Tailorable Trimethyl Chitosans as Adjuvant for ... - TI Pharma
Tailorable Trimethyl Chitosans as Adjuvant for ... - TI Pharma
Tailorable Trimethyl Chitosans as Adjuvant for ... - TI Pharma
Create successful ePaper yourself
Turn your PDF publications into a flip-book with our unique Google optimized e-Paper software.
TAILORABLE TRIMETHYL CHITOSANS<br />
AS ADJUVANT FOR INTRANASAL<br />
IMMUNIZA<strong>TI</strong>ON
The printing of this thesis w<strong>as</strong> financially supported by:<br />
J.E. Jurriaanse Stichting<br />
Utrecht Institute <strong>for</strong> <strong>Pharma</strong>ceutical Sciences<br />
ISBN: 978-90-39354292<br />
© 2010 RJ Verheul, Utrecht<br />
Cover: Drawings taken from ‘Poissons, écrevisses et crabs’ by Louis Renard. First<br />
published in 1719, the book describes exotic sea creatures (including a mermaid) from<br />
the Netherlands West-Indies, now Indonesia. Many thanks to Marijn van Hoorn,<br />
curator of the Teylers Museum, Haarlem<br />
The research presented in this thesis w<strong>as</strong> per<strong>for</strong>med under the framework of <strong>TI</strong><br />
<strong>Pharma</strong> project D5-106-1; Vaccine delivery: alternatives <strong>for</strong> conventional multiple<br />
injection vaccines<br />
Printed by: Vandenberg, Maarn
TAILORABLE TRIMETHYL CHITOSANS AS<br />
ADJUVANT FOR INTRANASAL IMMUNIZA<strong>TI</strong>ON<br />
Varieerbare <strong>Trimethyl</strong> Chitosanen als Adjuvans voor Intran<strong>as</strong>ale Vaccinatie<br />
(met een samenvatting in het Nederlands)<br />
Proefschrift<br />
ter verkrijging van de graad van doctor aan de Universiteit Utrecht op<br />
gezag van de rector magnificus, prof. dr. J.C. Stoof, ingevolge het besluit<br />
van het college voor promoties in het openbaar te verdedigen op<br />
maandag 8 november 2010 des middags te 2.30 uur<br />
door<br />
Rudolf Johannus Verheul<br />
geboren op 8 maart 1980 te Oss
Promotoren: Prof. dr. ir. W.E. Hennink<br />
Prof. dr. W. Jiskoot
Voor mijn ouders
Table of Contents<br />
Chapter 1<br />
General introduction<br />
Chapter 2<br />
Synthesis, characterization and in vitro biological properties of O-methyl free<br />
N,N,N,-trimethylated chitosan<br />
Chapter 3<br />
Influence of the degree of acetylation on the enzymatic degradation and in vitro<br />
biological properties of trimethylated chitosans<br />
Chapter 4<br />
Relationship between structure and adjuvanticity of trimethyl chitosan (TMC)<br />
structural variants in a n<strong>as</strong>al influenza vaccine<br />
Chapter 5A<br />
A step-by-step approach to study the influence of N-acetylation on the<br />
adjuvanticity of N,N,N-trimethyl chitosan (TMC) in an intran<strong>as</strong>al whole<br />
inactivated influenza virus vaccine<br />
Chapter 5B<br />
Maturation of human monocyte derived dendritic cells by trimethyl chitosan is<br />
correlated with its N-acetyl glucosamine (GlcNAc) content<br />
Chapter 6<br />
<strong>Tailorable</strong> thiolated trimethyl chitosans <strong>for</strong> covalently stabilized nanoparticles<br />
Chapter 7<br />
Covalently stabilized trimethyl chitosan-hyaluronic acid nanoparticles <strong>for</strong> n<strong>as</strong>al<br />
and intradermal vaccination<br />
Chapter 8<br />
Summary and future perspectives<br />
Appendices<br />
Affiliations of collaborating authors<br />
List of abbreviations<br />
Curriculum vitae<br />
List of publications<br />
Nederlandse samenvatting<br />
Dankwoord<br />
Page<br />
9<br />
25<br />
47<br />
69<br />
91<br />
113<br />
125<br />
149<br />
169<br />
183
CHAPTER 1<br />
GENERAL INTRODUC<strong>TI</strong>ON
General Introduction<br />
N<strong>as</strong>al Vaccination<br />
Ever since Edward Jenner’s successful inoculations with cowpox to prevent a potentially<br />
lethal smallpox infection in the end of the 18 th century, active vaccination h<strong>as</strong> proven to be the<br />
most (cost) effective tool in the fight against infectious dise<strong>as</strong>es. Active vaccination, or<br />
immunization, involves activation of the immune system by controlled exposure to a<br />
milder/inactivated <strong>for</strong>m of the pathogen causing the dise<strong>as</strong>e, or to components derived from it,<br />
thereby inducing (immunological) memory and <strong>for</strong>tifying the host’s response towards the real<br />
pathogen. Vaccines are usually made of live attenuated or inactivated pathogens (e.g. viruses<br />
or bacteria) or purified immunogenic protein(conjugate)s derived from these pathogens. After<br />
two hundred years, still, most vaccines are administered via parenteral injection due to the<br />
limited absorption and enhanced degradation of these large molecular structures when using<br />
alternative administration routes [1]. There is however a need <strong>for</strong> vaccines that can be<br />
administered via non-inv<strong>as</strong>ive other routes <strong>for</strong> re<strong>as</strong>ons pointed out below. Table I gives an<br />
overview of the (dis)advantages of the different immunization routes.<br />
Table I. Advantages and disadvantages of different immunization routes (adopted from Slütter et al.<br />
[2]).<br />
Immunization route Advantages Disadvantages<br />
Parenteral<br />
N<strong>as</strong>al<br />
Oral<br />
Pulmonary<br />
Dermal<br />
Powerful systemic immune<br />
response<br />
Accurate dosing<br />
Non-inv<strong>as</strong>ive<br />
Mucosal and systemic immune<br />
responses<br />
E<strong>as</strong>ily accessible<br />
Little antigen degradation<br />
(compared to oral)<br />
Non-inv<strong>as</strong>ive<br />
Mucosal and systemic immune<br />
responses<br />
Large surface area<br />
Non-inv<strong>as</strong>ive<br />
Mucosal and systemic immune<br />
responses<br />
Little degradation (compared to<br />
oral)<br />
Non or minimally inv<strong>as</strong>ive<br />
Large, e<strong>as</strong>ily accessible application<br />
area<br />
High density of immune cells in skin<br />
Mucosal and systemic immune<br />
responses<br />
Inv<strong>as</strong>ive<br />
Limited mucosal immune response<br />
Risk of contaminated needles<br />
Need <strong>for</strong> trained personnel<br />
Mucociliary clearance<br />
Inefficient uptake of antigen<br />
Application device needed<br />
Vaccine/antigen digestion in<br />
stomach and gut<br />
Inefficient uptake of antigen<br />
Mucosal tolerance<br />
Highly variable antigen delivery<br />
Dry powder inhaler or nebulizer<br />
needed<br />
Clearance from lungs<br />
May require inv<strong>as</strong>ive technology<br />
(e.g. tattooing, microneedles)<br />
Patch or application device needed<br />
Less established technology<br />
11
Chapter 1<br />
Of these alternatives, n<strong>as</strong>al vaccination in particular h<strong>as</strong> some interesting advantages. The<br />
nose is e<strong>as</strong>ily accessible, commonly used and accepted <strong>for</strong> drug administration (e.g. nosesprays).<br />
A relatively simple, painless, device can be used without the aid of trained personnel<br />
and the n<strong>as</strong>al environment is less harsh <strong>for</strong> vaccine components than the oral route [3]. Also,<br />
high numbers of antigen presenting cells (APCs) are present in the n<strong>as</strong>al mucosal linings<br />
mediating both mucosal and systemic immune responses against <strong>for</strong>eign pathogens that try to<br />
invade the human body through the respiratory tract. Importantly, this mucosal immunity,<br />
hardly induced by parenteral immunization, may highly contribute to overall protection<br />
against a <strong>for</strong>eign pathogen. The excretion of secretory immunoglobulins is not only limited to<br />
the area of antigen-exposure but can occur on mucosal surfaces across the body (e.g. n<strong>as</strong>al<br />
immunization against sexually transmitted human papilloma virus may thus be an interesting<br />
option) [2, 4, 5].<br />
All these potential benefits taken into consideration, it should be mentioned that currently<br />
only one n<strong>as</strong>al vaccine is on the market: a live attenuated influenza vaccine administered via a<br />
n<strong>as</strong>al spray, thereby mimicking the natural route of inv<strong>as</strong>ion of the pathogen. Although<br />
effective, immunization with live attenuated viruses is under debate because of their potential<br />
to mutate, thereby escaping the immune system and regaining their pathogenicity. As a<br />
consequence, elderly, small infants and immuno-compromised people (e.g. by AIDS) are<br />
excluded from such vaccines. Subunit vaccines consisting of better defined and characterized<br />
antigenic proteins cannot mutate into pathogenic <strong>for</strong>ms but are generally less immunogenic<br />
and need potent adjuvant(system)s to elicit an adequate immune response [3, 6]. As vaccine<br />
safety is now top priority, a number of hurdles needs to be tackled to make intran<strong>as</strong>al (i.n.)<br />
vaccination a success-story.<br />
Successful n<strong>as</strong>al vaccine delivery<br />
After administration of the vaccine to the nose with an appropriate device, several successive<br />
steps can be identified that should lead to an adequate immune response (Figure 1). In short, a<br />
successful i.n. vaccine <strong>for</strong>mulation h<strong>as</strong> to adhere to the mucosal surfaces of the n<strong>as</strong>al cavity and<br />
provide protection against the proteolytic degradation of the antigen in the n<strong>as</strong>al environment.<br />
Prolonging n<strong>as</strong>al residence time may be an important mode of action <strong>for</strong> adjuvants; normally<br />
rapid mucociliarly clearance of the antigen will limit contact of the antigen with the epithelial<br />
barrier. Next, sufficient uptake and/or transport of antigen through the epithelial barrier<br />
should be achieved. Macromolecules up to a certain size (some suggest 22 kDa [1]) may use<br />
12
General Introduction<br />
paracellular pathways via cellular tight junctions (that can be opened by penetration<br />
enhancers) to overcome the epithelial barrier [1].<br />
Figure 1. Schematic overview of the consecutive steps towards successful n<strong>as</strong>al vaccine delivery: 1)<br />
muco-adhesion; 2) antigen uptake, by M-cell transport; 3) delivery to and subsequent<br />
activation/maturation of DC; 4) induction of B- and T-cell responses. DC= dendritic cell, M-cell =<br />
microfold cell, Th cell = helper T cell (adopted from Slütter et al. [2]).<br />
Microfold (M) cells play an important role especially <strong>for</strong> particulate systems, since they are<br />
capable of transporting the antigen by transcytosis to n<strong>as</strong>al <strong>as</strong>sociated lymphoid tissue (NALT)<br />
[2-4]. NALT consists of agglomerates of cells involved in the initiation and execution of an<br />
immune response, like dendritic cells (DCs), T- and B-cells. Alternatively, DCs may establish<br />
contact with the antigen through close interaction with epithelial cells [3]. After transport<br />
through the epithelium, the antigen h<strong>as</strong> to be taken up by antigen presenting cells (APC), most<br />
likely DCs or macrophages, which subsequently should mature, migrate and interact with<br />
helper T (Th) cells. Here, DCs will present a peptide (epitope) of the (degraded) antigen via the<br />
13
Chapter 1<br />
major histocompatibility complex cl<strong>as</strong>s II (MHC II) to the Th cells. Upon recognition of the MHC<br />
II-peptide complex and co-stimulation of the APC, naïve Th cells differentiate into effector Th<br />
cells, which can be divided in two major subtypes: Th1 and Th2 cells. Th1 cells are mainly<br />
involved in activation and proliferation of the cellular immune system, i.e. stimulate the<br />
cytotoxic T cells (CTL). Th2 cells are involved in stimulation of B-cells to differentiate into<br />
pl<strong>as</strong>ma cells and incre<strong>as</strong>e the humoral immune responses i.e. the production of<br />
immunoglobulins (IgG) also named antibodies. Interestingly, mucosal B-cells may differentiate<br />
into mucosal pl<strong>as</strong>ma cells that secrete prote<strong>as</strong>e resistant secretory dimeric IgA (sIgA) into the<br />
lumen to prevent mucosal inv<strong>as</strong>ion [7].<br />
Ideally, adaptive immune responses elicited by vaccination should comprise both cellular<br />
and humoral immune responses and should result in long-lived specific T- and B-memory cells,<br />
<strong>as</strong> well <strong>as</strong> some readily available circulating antibodies [3]. Importantly, DC or APC signaling<br />
determines the fate of the naïve Th cell and this can be modulated by the use of delivery<br />
systems and/or adjuvants. Thus, a vaccine <strong>for</strong>mulation may not only enhance the antigen<br />
availability or incre<strong>as</strong>e the immune response (i.e. “the danger signal”) but can also influence<br />
the type of immune response (i.e. immunomodulation) [2, 3, 8].<br />
Polymeric carrier systems<br />
Polymeric carriers have the advantage over other delivery systems (e.g. liposomes, ISCOMs)<br />
because of their endless potential of chemical modifications thereby allowing fine-tuning of the<br />
physico-chemical properties of the carrier system [6]. Although many polymers have been<br />
studied <strong>for</strong> n<strong>as</strong>al immunization [4], most research h<strong>as</strong> been done with the synthetic polymer<br />
poly(lactide-co-glycolide) (PLGA) and chitosan (derivatives) which are obtained from naturally<br />
abundant chitin. Interestingly, chitosan and its derivatives are much more effective in eliciting<br />
immune responses in micro- or nanoparticulate <strong>for</strong>m than <strong>as</strong> plain polymer solution [9-12],<br />
most likely because of the particle’s resemblance to the original pathogen, their multimeric<br />
antigen presentation and improved protection of the antigen against degradation [7].<br />
Furthermore, particles are better taken up by APCs and can co-deliver antigen and adjuvant to<br />
the same cell (13). While PLGA is FDA approved <strong>for</strong> several therapeutic applications and is<br />
considered biodegradable and biocompatible in humans, chitosan derivatives are gaining<br />
interest because of their superior efficacy in mucosal antigen delivery [2, 6, 14].<br />
14
General Introduction<br />
Chitosan<br />
Chitin (poly β1→4 N-acetyl D-glucosamine) is the second most abundant natural biopolymer<br />
and is derived from exoskeletons of crustaceans, insects and cell walls of several fungi [15].<br />
Chitosan (Scheme 1) is a polysaccharide consisting of β1→4-D-glucosamine and β1→4 N-<br />
acetyl D-glucosamine units and is obtained from chitin by partial deacetylation, which makes it<br />
water-soluble in an acidic aqueous solution [1]. It can vary in size (molecular weight) and/or<br />
degree of acetylation and is used in many are<strong>as</strong> of drug delivery and tissue engineering [16]. It<br />
is approved <strong>for</strong> dietary applications in several countries and the FDA h<strong>as</strong> approved chitosan<br />
<strong>for</strong> use in wound dressings [16]. Chitosan’s muco-adhesiveness, penetration-enhancing<br />
abilities and its properties allowing the preparation of particles without the use of organic<br />
solvents h<strong>as</strong> led to many investigations on chitosan <strong>for</strong>mulations <strong>for</strong> mucosal vaccination [1, 5,<br />
17]. Additionally, the functional groups (amines and hydroxyls) allow many modifications<br />
depending on the <strong>for</strong>eseen application [15]. However, despite many studies on toxicity and<br />
biocompatibility, chitosan is not (yet) considered a GRAS (generally regarded <strong>as</strong> safe) material<br />
and no approval <strong>for</strong> use in drug delivery h<strong>as</strong> been given so far. One re<strong>as</strong>on might be that<br />
chitosan’s high chemical variability confuses regulatory authorities [16].<br />
Scheme 1. Chemical structure of chitosan. The degree of acetylation (x) is variable.<br />
Chitosan’s primary amine-groups have a pK a of around 6, resulting in low aqueous solubility<br />
and loss of penetration-enhancing abilities at neutral pH [1]. Carboxylation of the amines<br />
and/or hydroxyl moieties and N-PEGylation have proven to be useful strategies <strong>for</strong> improving<br />
the aqueous solubility of chitosan at neutral pH [18-20]. Other modifications encomp<strong>as</strong>s the<br />
introduction of quaternary amine groups thereby giving chitosan a permanent, pH<br />
independent positive charge and thus aqueous solubility also at neutral pH. While sometimes<br />
15
Chapter 1<br />
quaternary ammonium groups are introduced directly e.g. by coupling 2-diethylaminoethyl<br />
chloride to chitosan [21, 22], mostly quaternization is achieved by partial trimethylation of the<br />
amine resulting in N,N,N-trimethylated chitosan (TMC) [23].<br />
N,N,N,-<strong>Trimethyl</strong>ated chitosan (TMC)<br />
TMC (Scheme 2) h<strong>as</strong> been shown to have muco-adhesive properties [27] and is able to open<br />
tight junctions above a degree of quaternization (DQ) of 20% [25-29]. In addition, TMC h<strong>as</strong><br />
been used to complex and condense DNA to yield polyplexes <strong>for</strong> gene delivery purposes [30-<br />
32]. TMC is synthesized b<strong>as</strong>ed on the method first published by Domard and coworkers [33]<br />
and later modified by Sieval et al. [23]. They showed that alkylation of primary amines of<br />
chitosan occurred by reaction of this polymer in strong alkaline conditions with an excess of<br />
iodomethane at elevated temperature (60°C) using N-methyl-2-pyrrolidone (NMP) <strong>as</strong> solvent.<br />
These relatively vigorous reaction conditions also lead to polymer chain scission [34] <strong>as</strong> well<br />
<strong>as</strong> to partial and uncontrolled methylation of the C-3 and C-6 hydroxyl groups of chitosan [35,<br />
36].<br />
Scheme 2. Chemical structure of TMC. TMC can vary in degree of acetylation (x), quaternization (y),<br />
dimethylation (z) and O-methylation (z). The various substitutions are randomly distributed throughout<br />
the polymer; O-methylation (z) may also occur on the quaternized and acetylated units.<br />
Several studies have been per<strong>for</strong>med to determine the optimal DQ <strong>for</strong> either trans-epithelial<br />
delivery of low molecular weight drug molecules and/or proteins, or to incre<strong>as</strong>e the<br />
transfection potential of complexes of TMC with pl<strong>as</strong>mid DNA. A DQ of about 40-50% w<strong>as</strong><br />
found to be the optimum <strong>for</strong> transepithelial delivery of both low molecular weight compounds<br />
and proteins [28, 37-40]. Furthermore, TMC h<strong>as</strong> been used successfully <strong>for</strong> n<strong>as</strong>al immunization<br />
in mice [41-44]. Amidi et al. showed high levels of serum IgG and HI titers after i.n.<br />
administration of influenza A subunit encapsulated in TMC-tripolyphosphate (TPP)<br />
16
General Introduction<br />
nanoparticles <strong>as</strong> compared to plain antigen and antigen with TMC in solution [9]. Additionally,<br />
our results and those of others suggest that optimizing the DQ of TMC may lead to further<br />
improvement of the vaccine <strong>for</strong>mulation: Boonyo et al. proposed an optimal DQ of 40% b<strong>as</strong>ed<br />
upon an i.n. immunization with ovalbumin [41] and we found that TMC with a DQ of 37% may<br />
be superior to TMC with a DQ of 15% in an i.n. vaccination with whole inactivated influenza<br />
virus [44]. Importantly, in these studies the TMCs used also had a variable extent of O-<br />
methylation (DOM), degree of acetylation (DAc) and differences in polymer molecular weights.<br />
While incre<strong>as</strong>ing TMC polymer molecular weight leads to an incre<strong>as</strong>e in toxicity [45], the<br />
effects of the other polymer compositional variables are currently unknown. In particular the<br />
role of the DAc can be anticipated to be quite substantial: in chitosan the DAc is <strong>as</strong>sociated with<br />
its enzymatic degradability [16], penetration-enhancing capability [46] and stimulating effect<br />
on APCs [47-50]. Tailorability of these variables (DQ, DOM and DAc) will allow better<br />
understanding of the contributions of each of those side groups to the physico-chemical and<br />
biological properties of TMC. Additionally, the introduction of novel substitutions such <strong>as</strong> thiolmoieties<br />
may further broaden the potential pharmaceutical applications of TMC by enhancing<br />
its muco-adhesive potential and allowing further chemical derivatization reactions via<br />
reducible disulfide bridges [51-54].<br />
TMC nanoparticle preparation<br />
As mentioned above, nanoparticles composed of (subunit) antigen and TMC are more<br />
effective in i.n. vaccination than soluble antigen and polymer. In some c<strong>as</strong>es TMC is used <strong>as</strong><br />
coating <strong>for</strong> particulate systems (e.g. PLGA nanoparticles or whole inactivated influenza virus<br />
particle) but more frequently TMC nanoparticles are prepared by ionic crosslinking methods.<br />
The complexation between the positively charged TMC and oppositely charged<br />
(macro)molecules added drop-wise under stirring in low ionic strength buffer results in<br />
spontaneous <strong>for</strong>mation of nanoparticles. This ionic gelation method is simple and mild <strong>for</strong><br />
proteins <strong>as</strong> no chemical crosslinkers, organic solvents or elevated temperatures are required<br />
[1]. It h<strong>as</strong> been reported that is some c<strong>as</strong>es the antigen or therapeutic protein alone <strong>as</strong> such act<br />
<strong>as</strong> crosslinker to yield nanoparticles [55], but often an additional crosslinker is needed. Both<br />
small molecules such <strong>as</strong> tripolyphosphate (TPP) [9, 56-58] and larger anionic macromolecules<br />
like polyglutamic acid [59] and hyaluronic acid [60] have been successfully used <strong>for</strong><br />
nanoparticle preparation. However, the physico-chemical stability of these complexes is<br />
dependent on the characteristics of the crosslinker used and they often have a limited stability<br />
17
Chapter 1<br />
in physiological saline [60, 61] or when the pH drops [59, 62]. So, although TMC nanoparticles<br />
are successfully used in n<strong>as</strong>al vaccination, many variables such <strong>as</strong> the type of crosslinker used<br />
and particle stability leave room <strong>for</strong> improvement.<br />
TMC mode of action<br />
Several studies have shown the superiority of positively charged particles above negatively<br />
charged or neutral particles in n<strong>as</strong>al immunization. E.g. chitosan coated PLGA particles<br />
improved immunogenicity of ‘naked’ PLGA particles [63], TMC particles were superior in i.n.<br />
vaccination with tetanus toxoid <strong>as</strong> compared to negatively charged particles made from<br />
carboxylated chitosan [19] and TMC coated whole inactivated influenza virus resulted in<br />
protection against a live virus challenge, in contr<strong>as</strong>t with the uncoated virus which w<strong>as</strong> not<br />
effective [44]. Often, muco-adhesion due to charge interactions between polymer and mucus is<br />
suggested <strong>as</strong> potential mechanism of action <strong>for</strong> these penetration-enhancing polymers. It h<strong>as</strong><br />
further been postulated that muco-adhesive polymers incre<strong>as</strong>e n<strong>as</strong>al residence time and<br />
improve interaction of the antigen with the epithelial cell barrier leading to an incre<strong>as</strong>ed<br />
antigen uptake [1, 24, 43]. Alternatively, the opening of cellular tight junctions (<strong>as</strong> determined<br />
in a transepithelial electrical resistance (TEER) <strong>as</strong>say) may result in enhanced antigen/protein<br />
uptake [59].<br />
Chitosan and TMC are known to induce a rearrangement of cytoskeletal F-actin and tight<br />
junction protein ZO-1, leading to enhanced permeability of the epithelium [59, 64]. However,<br />
fully quaternized diethyl aminoethyl dextran w<strong>as</strong> ineffective <strong>as</strong> absorption enhancer indicating<br />
that solely cationic charge is not sufficient <strong>for</strong> a polymer to have adjuvant activity [65].<br />
Additionally, nanoparticles are considered too big to p<strong>as</strong>s through these junctions [1] and also<br />
non-TEER reducing TMCs can improve immunogenicity of i.n. vaccine <strong>for</strong>mulations [9, 44].<br />
Also, the direct adjuvant effect of TMC(-particles) on APCs or DCs h<strong>as</strong> hardly been studied, but<br />
<strong>as</strong> mentioned be<strong>for</strong>e, the N-acetyl glucosamine units may interact with C-type lectin receptors<br />
present on these cells [47-50]. Taken together, clearly, the exact mode of action of TMC is not<br />
yet fully understood.<br />
18
General Introduction<br />
Aim and thesis outline<br />
As pointed out in the previous paragraphs, TMCs are promising polymers <strong>for</strong> use in n<strong>as</strong>al<br />
immunization. However, many opportunities <strong>for</strong> optimizing TMC structure and TMC- b<strong>as</strong>ed<br />
nanoparticles still remain and mechanistic insights in the mode of action of TMC is lacking. The<br />
aim of this thesis is to develop synthetic routes to make TMC structural variants in a<br />
controllable and tailorable manner and introduce substitutions such <strong>as</strong> thiol-moieties that may<br />
further improve TMC’s properties. In this way, structure-activity relationships can be<br />
investigated in in vitro <strong>as</strong>says and in in vivo n<strong>as</strong>al vaccination studies. Also, novel covalently<br />
stabilized TMC-b<strong>as</strong>ed particles are prepared and investigated in in vivo immunization studies.<br />
Chapter 2 challenges the current synthetic method that introduces several side-reactions<br />
such <strong>as</strong> O-methylation and chain scission. A novel synthesis route is proposed that allows<br />
tailorability of the DQ without introducing other modifications. These novel TMC polymers are<br />
studied <strong>for</strong> physico-chemical properties, evaluated <strong>for</strong> cytotoxicity and the ability to open tight<br />
junctions, and compared with TMCs synthesized via the traditional method.<br />
Chapter 3 describes the synthesis of TMC with a high degree of acetylation (DAc) and<br />
investigates the lysozyme-catalyzed degradability of TMCs with different DAcs and with or<br />
without O-methylated groups, and evaluates these polymers <strong>for</strong> their physico-chemical<br />
properties, cytotoxicity and ability to open tight junctions.<br />
In Chapter 4 the adjuvant effect of several TMC types is investigated <strong>for</strong> i.n. administered<br />
whole inactivate influenza virus (WIV). In particular, O-methyl free TMCs with varying DQs,<br />
reacetylated O-methyl free TMC (TMC-RA), and conventional O-methylated TMCs with similar<br />
DQ are compared. The TMC-WIV vaccines are physicochemically characterized and the<br />
immunogenicity and protectivity of the vaccines are <strong>as</strong>sessed in a murine challenge model.<br />
Additionally, the influence of TMC:WIV ratio on the type and extent of humoral immune<br />
responses is investigated.<br />
To understand the differences in adjuvanticity of the different TMCs in Chapter 5A TMC-<br />
WIV, TMC-RA-WIV and WIV <strong>for</strong>mulations are compared at six potentially critical steps in the<br />
induction of an immune response after i.n. administration. In particular, (1) the degradation of<br />
TMC and TMC-RA in n<strong>as</strong>al w<strong>as</strong>hes, (2) n<strong>as</strong>al residence time, (3) n<strong>as</strong>al distribution patterns, (4)<br />
cellular uptake and (5) transport through an epithelial (Calu-3) cell line of WIV <strong>for</strong>mulated<br />
19
Chapter 1<br />
without or with TMC(-RA), and (6) the effect of the different <strong>for</strong>mulations on maturation of<br />
murine bone marrow derived dendritic cells (DCs) are studied. In Chapter 5B the effects of<br />
TMC and TMC-RA in solution and in combination with WIV on human monocyte-derived<br />
human DCs are investigated.<br />
Chapter 6 introduces a novel synthetic method to yield thiolated TMCs with a high DQ. The<br />
different thiolated TMCs are physico-chemically characterized, evaluated in in vitro<br />
cytotoxicity <strong>as</strong>says and used <strong>for</strong> preparation of covalently stabilized nanoparticles with<br />
thiolated hyaluronic acid.<br />
Physico-chemically characterized covalently stabilized and/or PEGylated TMC-hyaluronic<br />
acid nanoparticles are used in an intran<strong>as</strong>al and intradermal vaccination study in mice in<br />
Chapter 7. For comparison, non-stabilized particles are used <strong>as</strong> control. Both the type and the<br />
extent of the immune responses are investigated.<br />
Chapter 8 summarizes the findings and conclusions of this thesis and addresses the future<br />
opportunities <strong>for</strong> these novel TMCs and TMC-b<strong>as</strong>ed delivery systems.<br />
20
General Introduction<br />
References<br />
1. M. Amidi, E. M<strong>as</strong>trobattista, W. Jiskoot, and W. E. Hennink. Chitosan-b<strong>as</strong>ed delivery systems<br />
<strong>for</strong> protein therapeutics and antigens. Adv Drug Deliv Rev 62: 59-82 (2010).<br />
2. B. Slütter, N. Hagenaars, and W. Jiskoot. Rational design of n<strong>as</strong>al vaccines. J Drug Target 16: 1-17<br />
(2008).<br />
3. N. Csaba, M. Garcia-Fuentes, and M. J. Alonso. Nanoparticles <strong>for</strong> n<strong>as</strong>al vaccination. Adv Drug<br />
Deliv Rev 61: 140-157 (2009).<br />
4. S. Chadwick, C. Kriegel, and M. Amiji. Nanotechnology solutions <strong>for</strong> mucosal immunization.<br />
Adv Drug Deliv Rev 62: 394-407 (2009).<br />
5. L. Illum, I. Jabbal-Gill, M. Hinchcliffe, A. N. Fisher, and S. S. Davis. Chitosan <strong>as</strong> a novel n<strong>as</strong>al<br />
delivery system <strong>for</strong> vaccines. Adv Drug Deliv Rev 51: 81-96 (2001).<br />
6. N. Mishra, A. K. Goyal, S. Tiwari, R. Paliwal, S. R. Paliwal, B. Vaidya, S. Mangal, M. Gupta, D.<br />
Dube, A. Mehta, and S. P. Vy<strong>as</strong>. Recent advances in mucosal delivery of vaccines: Role of<br />
mucoadhesive/biodegradable polymeric carriers. Exp Opin Ther Patents 20: 661-679 (2010).<br />
7. M. R. Neutra and P. A. Kozlowski. Mucosal vaccines: The promise and the challenge. Nat Rev<br />
Immunol 6: 148-158 (2006).<br />
8. V. E. Schijns. Immunological concepts of vaccine adjuvant activity. Curr Opin Immunol 12: 456-<br />
463 (2000).<br />
9. M. Amidi, S. G. Romeijn, J. C. Verhoef, H. E. Junginger, L. Bungener, A. Huckriede, D. J. A.<br />
Crommelin, and W. Jiskoot. N-<strong>Trimethyl</strong> chitosan (TMC) nanoparticles loaded with influenza<br />
subunit antigen <strong>for</strong> intran<strong>as</strong>al vaccination: Biological properties and immunogenicity in a mouse<br />
model. Vaccine 25: 144-153 (2007).<br />
10. L. Feng, X. J. Zhou, and X. R. Qi. Preparation, rele<strong>as</strong>e and immunogenicity evaluation of<br />
HBsAg-PLGA microspheres. Beijing da xue xue bao. Yi xue ban = Journal of Peking University. Health<br />
sciences 37: 527-531 (2005).<br />
11. D. T. O'Hagan, M. Singh, and J. B. Ulmer. Microparticle-b<strong>as</strong>ed technologies <strong>for</strong> vaccines.<br />
Methods 40: 10-19 (2006).<br />
12. M. J. Shephard, D. Todd, B. M. Adair, A. Li Wan Po, D. P. Mackie, and E. M. Scott.<br />
Immunogenicity of bovine parainfluenza type 3 virus proteins encapsulated in nanoparticle<br />
vaccines, following intran<strong>as</strong>al administration to mice. Res Vet Sci 74: 187-190 (2003).<br />
13. A. C. Rice-Ficht, A. M. Aren<strong>as</strong>-Gamboa, M. M. Kahl-McDonagh, and T. A. Ficht. Polymeric<br />
particles in vaccine delivery. Curr Opin Microbiology 13: 106-112 (2010).<br />
14. B. Slütter, L. Plapied, V. Fievez, M. Alonso Sande, A. des Rieux, Y. J. Schneider, E. Van Riet,<br />
W. Jiskoot, and V. Préat. Mechanistic study of the adjuvant effect of biodegradable<br />
nanoparticles in mucosal vaccination. J Control Rele<strong>as</strong>e 138: 113-121 (2009).<br />
15. V. K. Mourya and N. N. Inamdar. Chitosan-modifications and applications: Opportunities<br />
galore. React Funct Polym 68: 1013-1051 (2008).<br />
16. T. Kean and M. Thanou. Biodegradation, biodistribution and toxicity of chitosan. Adv Drug<br />
Deliv Rev 62: 3-11 (2010).<br />
17. H. C. Arca, M. Günbeyaz, and S. Şenel. Chitosan-b<strong>as</strong>ed systems <strong>for</strong> the delivery of vaccine<br />
antigens. Exp Rev Vaccines 8: 937-953 (2009).<br />
18. Y. Ohya, R. Cai, H. Nishizawa, K. Hara, and T. Ouchi. Preparation of PEG-grafted chitosan<br />
nanoparticles <strong>as</strong> peptide drug carriers. S.T.P. <strong>Pharma</strong> Sciences 10: 77-82 (2000).<br />
19. B. Sayin, S. Somavarapu, X. W. Li, M. Thanou, D. Sesardic, H. O. Alpar, and S. Şenel. Mono-<br />
N-carboxymethyl chitosan (MCC) and N-trimethyl chitosan (TMC) nanoparticles <strong>for</strong> noninv<strong>as</strong>ive<br />
vaccine delivery. Int J Pharm 363: 139-148 (2008).<br />
20. M. Thanou, M. T. Nihot, M. Jansen, J. C. Verhoef, and H. E. Junginger. Mono-Ncarboxymethyl<br />
chitosan (MCC), a polyampholytic chitosan derivative, enhances the intestinal<br />
absorption of low molecular weight heparin across intestinal epithelia in vitro and in vivo. J<br />
Pharm Sci 90: 38-46 (2001).<br />
21
Chapter 1<br />
21. Y. Xu, Y. Du, R. Huang, and L. Gao. Preparation and modification of N-(2-hydroxyl) propyl-3-<br />
trimethyl ammonium chitosan chloride nanoparticle <strong>as</strong> a protein carrier. Biomaterials 24: 5015-<br />
5022 (2003).<br />
22. Y. Zambito, C. Zaino, G. Uccello-Barretta, F. Balzano, and G. Di Colo. Improved synthesis of<br />
quaternary ammonium-chitosan conjugates (N+-Ch) <strong>for</strong> enhanced intestinal drug permeation.<br />
Eur J Pharm Sci 33: 343-350 (2008).<br />
23. A. B. Sieval, M. Thanou, A. F. Kotze, J. C. Verhoef, J. Brussee, and H. E. Junginger.<br />
Preparation and NMR characterization of highly substituted N-trimethyl chitosan chloride.<br />
Carbohydr Polym 36: 157-165 (1998).<br />
24. V. Grabovac, D. Guggi, and A. Bernkop-Schnürch. Comparison of the mucoadhesive<br />
properties of various polymers. Adv Drug Deliv Rev 57: 1713-23 (2005).<br />
25. A. F. Kotzé, H. L. Lueßen, B. J. De Leeuw, B. G. De Boer, J. C. Verhoef, and H. E. Junginger.<br />
N-<strong>Trimethyl</strong> chitosan chloride <strong>as</strong> a potential absorption enhancer across mucosal surfaces: In<br />
vitro evaluation in intestinal epithelial cells (Caco-2). Pharm Res 14: 1197-1202 (1997).<br />
26. G. Sandri, M. C. Bonferoni, S. Rossi, F. Ferrari, C. Boselli, and C. Caramella. Insulin-loaded<br />
nanoparticles b<strong>as</strong>ed on N-trimethyl chitosan: In vitro (caco-2 model) and ex vivo (excised rat<br />
jejunum, duodenum, and ileum) evaluation of penetration enhancement properties. AAPS<br />
PharmSciTech 11: 362-371 (2010).<br />
27. D. Snyman, J. H. Hamman, and A. F. Kotze. Evaluation of the mucoadhesive properties of N-<br />
trimethyl chitosan chloride. Drug Develop Ind Pharm 29: 61-69 (2003).<br />
28. M. M. Thanou, A. F. Kotze, T. Scharringhausen, H. L. Lueßen, A. G. De Boer, J. C. Verhoef,<br />
and H. E. Junginger. Effect of degree of quaternization of N-trimethyl chitosan chloride <strong>for</strong><br />
enhanced transport of hydrophilic compounds across intestinal Caco-2 cell monolayers. J<br />
Control Rele<strong>as</strong>e 64: 15-25 (2000).<br />
29. I. M. Van der Lubben, J. C. Verhoef, M. M. Fretz, O. Van, I. Mesu, G. Kersten, and H. E.<br />
Junginger. <strong>Trimethyl</strong> chitosan chloride (TMC) <strong>as</strong> a novel excipient <strong>for</strong> oral and n<strong>as</strong>al<br />
immunisation against diphtheria. S.T.P. <strong>Pharma</strong> Sciences 12: 235-242 (2002).<br />
30. T. Kean, S. Roth, and M. Thanou. <strong>Trimethyl</strong>ated chitosans <strong>as</strong> non-viral gene delivery vectors:<br />
Cytotoxicity and transfection efficiency. J Control Rele<strong>as</strong>e 103: 643-653 (2005).<br />
31. Z. Mao, M. Lie, Y. Jiang, M. Yan, C. Gao, and J. Shen. N,N,N-trimethylchitosan chloride <strong>as</strong> a<br />
gene vector: Synthesis and application. Macromol Biosci 7: 855-863 (2007).<br />
32. M. Thanou, B. I. Florea, M. Geldof, H. E. Junginger, and G. Borchard. Quaternized chitosan<br />
oligomers <strong>as</strong> novel gene delivery vectors in epithelial cell lines. Biomaterials 23: 153-159 (2002).<br />
33. A. Domard, M. Rinaudo, and C. Terr<strong>as</strong>sin. New method <strong>for</strong> the quaternization of chitosan. Int J<br />
Bio Macromol 8: 105-107 (1986).<br />
34. D. Snyman, J. H. Hamman, J. S. Kotze, J. E. Rollings, and A. F. Kotze. The relationship<br />
between the absolute molecular weight and the degree of quaternisation of N-trimethyl chitosan<br />
chloride. Carbohydr Polym 50: 145-150 (2002).<br />
35. E. Curti, D. De Britto, and S. P. Campana-Filho. Methylation of chitosan with iodomethane:<br />
Effect of reaction conditions on chemoselectivity and degree of substitution. Macromol Biosci 3:<br />
571-576 (2003).<br />
36. A. Polnok, G. Borchard, J. C. Verhoef, N. Sarisuta, and H. E. Junginger. Influence of<br />
methylation process on the degree of quaternization of N-trimethyl chitosan chloride. Eur J<br />
Pharm Biopharm 57: 77-83 (2004).<br />
37. J. H. Hamman, C. M. Schultz, and A. F. Kotzé. N-trimethyl chitosan chloride: Optimum degree<br />
of quaternization <strong>for</strong> drug absorption enhancement across epithelial cells. Drug Develop Ind<br />
Pharm 29: 161-172 (2003).<br />
38. A. F. Kotze, M. M. Thanou, H. L. Luessen, A. B. G. De Boer, J. C. Verhoef, and H. E.<br />
Junginger. Effect of the degree of quaternization of N-trimethyl chitosan chloride on the<br />
permeability of intestinal epithelial cells (Caco-2). Eur J Pharm Biopharm 47: 269-274 (1999).<br />
22
General Introduction<br />
39. G. Sandri, S. Rossi, M. C. Bonferoni, F. Ferrari, Y. Zambito, G. Di Colo, and C. Caramella.<br />
Buccal penetration enhancement properties of N-trimethyl chitosan: Influence of quaternization<br />
degree on absorption of a high molecular weight molecule. Int J Pharm 297: 146-155 (2005).<br />
40. S. M. Van Der Merwe, J. C. Verhoef, J. H. M. Verheijden, A. F. Kotzé, and H. E. Junginger.<br />
<strong>Trimethyl</strong>ated chitosan <strong>as</strong> polymeric absorption enhancer <strong>for</strong> improved peroral delivery of<br />
peptide drugs. Eur J Pharm Biopharm 58: 225-235 (2004).<br />
41. W. Boonyo, H. E. Junginger, N. Waranuch, A. Polnok, and T. Pitaksuteepong. Chitosan and<br />
trimethyl chitosan chloride (TMC) <strong>as</strong> adjuvants <strong>for</strong> inducing immune responses to ovalbumin in<br />
mice following n<strong>as</strong>al administration. J Control Rele<strong>as</strong>e 121: 168-175 (2007).<br />
42. M. Amidi, S. G. Romeijn, G. Borchard, H. E. Junginger, W. E. Hennink, and W. Jiskoot.<br />
Preparation and characterization of protein-loaded N-trimethyl chitosan nanoparticles <strong>as</strong> n<strong>as</strong>al<br />
delivery system. J Control Rele<strong>as</strong>e 111: 107-116 (2006).<br />
43. B. C. Baudner, J. C. Verhoef, M. M. Giuliani, S. Peppoloni, R. Rappuoli, G. Del Giudice, and H.<br />
E. Junginger. Protective immune responses to meningococcal C conjugate vaccine after<br />
intran<strong>as</strong>al immunization of mice with the LTK63 mutant plus chitosan or trimethyl chitosan<br />
chloride <strong>as</strong> novel delivery plat<strong>for</strong>m. J Drug Target 13: 489-498 (2005).<br />
44. N. Hagenaars, E. M<strong>as</strong>trobattista, R. J. Verheul, I. Mooren, H. L. Glansbeek, J. G. M. Heldens,<br />
H. Van Den Bosch, and W. Jiskoot. Physicochemical and immunological characterization of<br />
N,N,N-trimethyl chitosan-coated whole inactivated influenza virus vaccine <strong>for</strong> intran<strong>as</strong>al<br />
administration. Pharm Res 26: 1353-1364 (2009).<br />
45. S. Mao, X. Shuai, F. Unger, M. Wittmar, X. Xie, and T. Kissel. Synthesis, characterization and<br />
cytotoxicity of poly(ethylene glycol)-graft-trimethyl chitosan block copolymers. Biomaterials 26:<br />
6343-6356 (2005).<br />
46. N. G. M. Schipper, K. M. Vårum, and P. Artursson. <strong>Chitosans</strong> <strong>as</strong> absorption enhancers <strong>for</strong><br />
poorly absorbable drugs. 1: Influence of molecular weight and degree of acetylation on drug<br />
transport across human intestinal epithelial (Caco-2) cells. Pharm Res 13: 1686-1692 (1996).<br />
47. K. Nishimura, S. Nishimura, and H. Seo. Macrophage activation with multi-porous beads<br />
prepared from partially deacetylated chitin. J Biomed Mat Res 20: 1359-1372 (1986).<br />
48. M. J. Robinson, D. Sancho, E. C. Slack, S. LeibundGut-Landmann, and C. R. Sousa. Myeloid C-<br />
type lectins in innate immunity. Nat Immunology 7: 1258-1265 (2006).<br />
49. J. Nadesalingam, A. W. Dodds, K. B. M. Reid, and N. Palaniyar. Mannose-binding lectin<br />
recognizes peptidoglycan via the N-acetyl glucosamine moiety, and inhibits ligand-induced<br />
proinflammatory effect and promotes chemokine production by macrophages. J Immunol 175:<br />
1785-1794 (2005).<br />
50. P. Zhang, S. Snyder, P. Feng, P. Azadi, S. Zhang, S. Bulgheresi, K. E. Sanderson, J. He, J.<br />
Klena, and T. Chen. Role of N-acetylglucosamine within core lipopolysaccharide of several<br />
species of Gram-negative bacteria in targeting the DC-SIGN (CD209). J Immunol 177: 4002-<br />
4011 (2006).<br />
51. K. Albrecht and A. Bernkop-Schnürch. Thiomers: Forms, functions and applications to<br />
nanomedicine. Nanomedicine 2: 41-50 (2007).<br />
52. A. Bernkop-Schnürch and A. Greimel. Thiomers: The next generation of mucoadhesive<br />
polymers. Amer J Drug Deliv 3: 141-154 (2005).<br />
53. T. M<strong>as</strong>uko, A. Minami, N. Iw<strong>as</strong>aki, T. Majima, S. I. Nishimura, and Y. C. Lee. Thiolation of<br />
chitosan. Attachment of proteins via thioether <strong>for</strong>mation. Biomacromol 6: 880-884 (2005).<br />
54. L. Yin, J. Ding, C. He, L. Cui, C. Tang, and C. Yin. Drug permeability and mucoadhesion<br />
properties of thiolated trimethyl chitosan nanoparticles in oral insulin delivery. Biomaterials 30:<br />
5691-5700 (2009).<br />
55. S. Mao, U. Bakowsky, A. Jintapattanakit, and T. Kissel. Self-<strong>as</strong>sembled polyelectrolyte<br />
nanocomplexes between chitosan derivatives and insulin. J Pharm Sci 95: 1035-1048 (2006).<br />
56. F. Chen, Z. R. Zhang, and Y. Huang. Evaluation and modification of N-trimethyl chitosan<br />
chloride nanoparticles <strong>as</strong> protein carriers. Int J Pharm 336: 166-173 (2007).<br />
23
Chapter 1<br />
57. F. Chen, Z. R. Zhang, F. Yuan, X. Qin, M. Wang, and Y. Huang. In vitro and in vivo study of<br />
N-trimethyl chitosan nanoparticles <strong>for</strong> oral protein delivery. Int J Pharm 349: 226-233 (2008).<br />
58. G. Sandri, M. C. Bonferoni, S. Rossi, F. Ferrari, S. Gibin, Y. Zambito, G. Di Colo, and C.<br />
Caramella. Nanoparticles b<strong>as</strong>ed on N-trimethylchitosan: Evaluation of absorption properties<br />
using in vitro (Caco-2 cells) and ex vivo (excised rat jejunum) models. Eur JPharm Biopharm 65:<br />
68-77 (2007).<br />
59. F. L. Mi, Y. Y. Wu, Y. H. Lin, K. Sonaje, Y. C. Ho, C. T. Chen, J. H. Juang, and H. W. Sung.<br />
Oral delivery of peptide drugs using nanoparticles self-<strong>as</strong>sembled by poly(γ-glutamic acid) and a<br />
chitosan derivative functionalized by trimethylation. Bioconj Chem 19: 1248-1255 (2008).<br />
60. S. Boddohi, N. Moore, P. A. Johnson, and M. J. Kipper. Polysaccharide-b<strong>as</strong>ed polyelectrolyte<br />
complex nanoparticles from chitosan, heparin, and hyaluronan. Biomacromol 10: 1402-1409<br />
(2009).<br />
61. D. V. Pergushov, H. M. Buchhammer, and K. Lunkwitz. Effect of a low-molecular-weight salt<br />
on colloidal dispersions of interpolyelectrolyte complexes. Colloid Polym Sci 277: 101-107 (1999).<br />
62. A. Bernkop-Schnürch, A. Weithaler, K. Albrecht, and A. Greimel. Thiomers: Preparation and in<br />
vitro evaluation of a mucoadhesive nanoparticulate drug delivery system. Int J Pharm 317: 76-81<br />
(2006).<br />
63. K. S. Jaganathan and S. P. Vy<strong>as</strong>. Strong systemic and mucosal immune responses to surfacemodified<br />
PLGA microspheres containing recombinant Hepatitis B antigen administered<br />
intran<strong>as</strong>ally. Vaccine 24: 4201-4211 (2006).<br />
64. N. G. M. Schipper, S. Olsson, J. A. Hoogstraate, A. G. DeBoer, K. M. Vårum, and P.<br />
Artursson. <strong>Chitosans</strong> <strong>as</strong> absorption enhancers <strong>for</strong> poorly absorbable drugs 2: Mechanism of<br />
absorption enhancement. Pharm Res 14: 923-929 (1997).<br />
65. G. Di Colo, S. Burgal<strong>as</strong>si, Y. Zambito, D. Monti, and P. Chetoni. Effects of different N-<br />
trimethyl chitosans on in vitro/in vivo ofloxacin transcorneal permeation. J Pharm Sci 93: 2851-<br />
2862 (2004).<br />
24
CHAPTER 2<br />
SYNTHESIS, CHARACTERIZA<strong>TI</strong>ON AND IN<br />
VITRO BIOLOGICAL PROPER<strong>TI</strong>ES OF O-<br />
METHYL FREE N,N,N-TRIMETHYLATED<br />
CHITOSAN<br />
Rolf J. Verheul, Maryam Amidi, Steffen van der Wal,<br />
Elly van Riet, Wim Jiskoot, Wim E. Hennink.<br />
Biomaterials 2008, 29, 3642-3649
Chapter 2<br />
Abstract<br />
N,N,N-trimethylated chitosan (TMC) with varying degrees of quaternization (DQs) is<br />
currently being investigated in mucosal drug, vaccine and in gene delivery. However, besides<br />
N-methylation, also O-methylation and chain scission occur during the synthesis of this<br />
polymer. Since both side reactions may affect the polymer characteristics, there is a need <strong>for</strong><br />
TMCs without O-methylation and disparities in chain lengths while varying the DQ. In this<br />
study, O-methyl free TMC with varying DQs w<strong>as</strong> successfully synthesized by using a two-step<br />
method. First, chitosan w<strong>as</strong> quantitatively dimethylated using <strong>for</strong>mic acid and <strong>for</strong>maldehyde.<br />
Then, in presence of an excess amount of iodomethane, TMC w<strong>as</strong> obtained with different DQs<br />
by varying reaction time. TMC obtained by this two-step method showed no detectable O-<br />
methylation ( 1 H-NMR) and a slight incre<strong>as</strong>e in molecular weight with incre<strong>as</strong>ing DQ (GPC),<br />
implying that no chain scission occurred during synthesis. The solubility in aqueous solutions<br />
at pH 7 of O-methyl free TMC with DQ < 22% w<strong>as</strong> less <strong>as</strong> compared to O-methylated TMC with<br />
the same DQ. On the other hand, O-methyl free TMC with DQ > 30% had a good aqueous<br />
solubility. On Caco-2 cells O-methyl free TMCs demonstrated a larger decre<strong>as</strong>e in transepithelial<br />
electrical resistance (TEER) than O-methylated TMCs. Also, with incre<strong>as</strong>ing DQ, an<br />
incre<strong>as</strong>e in cytotoxicity (MTT) and membrane permeability (LDH) w<strong>as</strong> observed.<br />
26
Synthesis and Characterization of O-methyl Free TMC<br />
Introduction<br />
Chitosan is a polysaccharide consisting of β1→4-D-glucosamine and β1→4 N-acetyl D-<br />
glucosamine units and is obtained by partial deacetylation of the natural polymer chitin. In<br />
recent years, chitosan h<strong>as</strong> been under investigation <strong>for</strong> various biomedical and pharmaceutical<br />
applications [1-4]. However, its poor aqueous solubility and loss of penetration-enhancing<br />
activity above pH 6 is a major drawback <strong>for</strong> its use at physiological conditions [5]. Partial<br />
quaternization of chitosan’s primary amine groups h<strong>as</strong> been used to obtain a chitosan<br />
derivative that is soluble at physiological conditions. N,N,N-trimethylated chitosan (TMC) h<strong>as</strong><br />
been shown to have muco-adhesive properties and is able to open tight junctions above a<br />
certain degree of quaternization (DQ) [6-10]. In addition, TMC h<strong>as</strong> been used to complex and<br />
condense DNA to yield polyplexes <strong>for</strong> gene delivery purposes [11, 12].<br />
Without exceptions, TMC is synthesized b<strong>as</strong>ed on the method first published by Domard and<br />
coworkers [13] and later modified by Sieval et al [7]. They showed the alkylation of primary<br />
amines of chitosan by reaction of this polymer in strong alkaline conditions with an excess of<br />
iodomethane using N-methyl-2-pyrrolidone (NMP) <strong>as</strong> solvent. The relatively vigorous reaction<br />
conditions lead to polymer chain scission [14] and, importantly, partial and uncontrolled<br />
methylation of the C-3 and C-6 hydroxyl groups of chitosan [15, 16]. Furthermore, the DQ<br />
proved difficult to control and often multiple reaction steps are required to obtain the desired<br />
TMC [17].<br />
Several studies have been per<strong>for</strong>med to determine the optimal DQ <strong>for</strong> either transepithelial<br />
delivery of low molecular weight drug molecules and/or proteins, or to incre<strong>as</strong>e the<br />
transfection potential of complexes of TMC with pl<strong>as</strong>mid DNA. It h<strong>as</strong> been reported that a DQ<br />
of about 40-50% is the optimum <strong>for</strong> transepithelial delivery of both low molecular weight<br />
compounds [17, 18] and proteins [19]. In these studies the TMCs used also had a variable<br />
extent of O-methylation and disparities in polymer chain lengths. Since O-methylation and<br />
variations in polymer chain length may affect the physicochemical properties and, likely, also<br />
the biological properties of TMC, there is a need <strong>for</strong> a synthetic method that yields TMC<br />
without O-methylated groups and prevents chain scission.<br />
Muzzarelli and Tanfani reported on the synthesis of TMC using iodomethane and N-dimethyl<br />
chitosan (DMC) obtained by reaction of chitosan with <strong>for</strong>maldehyde and sodium borohydride<br />
[20]. They showed that up to 60% of the amine groups could be trimethylated by this method,<br />
but no investigations were done on the tailorability of the DQ. Interestingly, this two-step<br />
method likely prevents chain scission and deacetylation of remaining N-acetyl groups, and<br />
might result in TMC without O-methylation. Nevertheless, this w<strong>as</strong> not investigated by these<br />
27
Chapter 2<br />
researchers. Recently, several adjustments to this method were presented by Jia et al. [21] and<br />
Guo et al. [22] introducing sodium hydroxide in the second step to incre<strong>as</strong>e the degree of<br />
substitution. However, the use of a strong b<strong>as</strong>e when trimethylating DMC will very likely result<br />
in O-methylation. So far, TMC synthesized by these methods h<strong>as</strong> only been studied <strong>for</strong> its<br />
antibacterial activity [20-22].<br />
The aim of this study w<strong>as</strong> to investigate a two-step method to synthesize TMC with tailorable<br />
DQ avoiding O-methylation and chain scission <strong>as</strong> side reactions. The synthesized TMC with<br />
different DQs were studied <strong>for</strong> physico-chemical properties, evaluated <strong>for</strong> cytotoxicity and the<br />
ability to open tight junctions, and compared with TMC synthesized via the traditional method.<br />
Materials and Methods<br />
Materials. Chitosan with a residual degree of acetylation of 17% (determined by NMR) and<br />
M n of 25 kDa, M w of 42 kDa (determined by Viscotek triple detection system <strong>as</strong> described<br />
below) w<strong>as</strong> purch<strong>as</strong>ed from Primex (Siglufjordur, Iceland). N-methyl-2-pyrrolidone (NMP),<br />
<strong>for</strong>maldehyde (37% stabilized with methanol), <strong>for</strong>mic acid, thiazolyl blue tetrazolium bromide<br />
(MTT), sodium acetate, acetic acid (anhydrous), sodium hydroxide and hydrochloric acid were<br />
obtained from Sigma-Aldrich Chemical Co. Dulbecco’s Modified Eagle’s Medium (DMEM),<br />
Hank’s balanced salt solution (HBSS), Fetal calf serum (FCS) were obtained from Invitrogen<br />
(Breda, The Netherlands). Sodium dodecyl sulfate (SDS) w<strong>as</strong> ordered from Merck (Darmstadt,<br />
Germany). Iodomethane 99% stabilized with copper w<strong>as</strong> obtained from Acros Organics (Geel,<br />
Belgium). All other chemicals used were of analytical grade.<br />
Synthesis of dimethylated chitosan (DMC). The two-step reaction pathway to synthesize<br />
TMC avoiding O-methylation is depicted in scheme 1 and is b<strong>as</strong>ed on the method published by<br />
Muzzarelli and Tanfani [20], with some modifications. In detail, a <strong>for</strong>mic acid-<strong>for</strong>maldehyde<br />
methylation (Eschweiler-Clarke) w<strong>as</strong> used to synthesize N,N-dimethylated chitosan [23].<br />
Instead of sodium borohydride <strong>as</strong> the reducing agent [20-22], we used <strong>for</strong>mic acid that allows<br />
chitosan to dissolve in the aqueous solution without the use of an acetate buffer. Ten grams of<br />
chitosan w<strong>as</strong> transferred into a 500 ml roundbottom fl<strong>as</strong>k. Next, 30 ml of <strong>for</strong>mic acid w<strong>as</strong><br />
added followed by 40 ml of 37% <strong>for</strong>maldehyde and 180 ml of distilled water yielding a total<br />
volume of about 250 ml. A reflux condenser w<strong>as</strong> attached and the solution w<strong>as</strong> heated to 70 °C<br />
and stirred using a magnetic stirrer <strong>for</strong> 118 hours. Then, the slightly yellow, viscous solution<br />
w<strong>as</strong> evaporated under reduced pressure and 1 M NaOH solution w<strong>as</strong> used to incre<strong>as</strong>e the pH<br />
28
Synthesis and Characterization of O-methyl Free TMC<br />
to 12 at which gel <strong>for</strong>mation occurred. This gel w<strong>as</strong> w<strong>as</strong>hed with deionized water over a gl<strong>as</strong>s<br />
filter to remove impurities. Then, the DMC w<strong>as</strong> dissolved in deionized water at pH 4 (adjusted<br />
with 1 M HCl), filtered over a gl<strong>as</strong>s filter and dialyzed against deionized water <strong>for</strong> three days<br />
(changing buffer twice-daily). Finally, the product w<strong>as</strong> filtered through a 0.8 µm filter and<br />
freeze dried.<br />
Synthesis of trimethylated chitosan from DMC. DMC w<strong>as</strong> reacted with iodomethane to<br />
yield TMC following the method described by Muzzarelli, with some modifications (see Scheme<br />
1) [20]. To prevent O-methylation, the reaction of DMC with iodomethane w<strong>as</strong> done in NMP,<br />
without the addition of a b<strong>as</strong>e catalyst [21, 22]. In detail, 250 mg of DMC w<strong>as</strong> dissolved in 40<br />
ml deionized water and the pH w<strong>as</strong> adjusted to 11 with NaOH by which gel <strong>for</strong>mation<br />
occurred. This step is per<strong>for</strong>med to ensure deprotonation of the tertiary amino groups of the<br />
DMC. Then, the gel w<strong>as</strong> w<strong>as</strong>hed with water and finally three times with acetone. Next, DMC<br />
w<strong>as</strong> suspended in 50 ml NMP and 2 ml iodomethane w<strong>as</strong> added. The dispersion w<strong>as</strong> stirred at<br />
40°C <strong>for</strong> the desired time and subsequently dropped in 150 ml of an ethanol/diethyl ether<br />
mixture (50/50). The precipitate (TMC) w<strong>as</strong> isolated by centrifugation and w<strong>as</strong>hed extensively<br />
with diethyl ether. After drying overnight, the TMC w<strong>as</strong> dissolved in 100 ml of an aqueous 10%<br />
NaCl solution and put on a shaker <strong>for</strong> a minimum of 18 hours <strong>for</strong> ion-exchange. Finally, the<br />
TMC w<strong>as</strong> dialyzed against deionized water <strong>for</strong> 3 days changing buffer twice daily, filtered<br />
through a 0.8 µm filter and freeze dried. Analysis on NMR 500 MHz w<strong>as</strong> per<strong>for</strong>med to<br />
determine the degree of quaternization (DQ) and to confirm the absence of O-methylation.<br />
TMCs with different DQs were synthesized starting with one gram of DMC following the same<br />
procedure but using 3 ml of iodomethane, 100 ml of NMP and a 250 ml round bottom fl<strong>as</strong>k and<br />
varying reaction time.<br />
29
Chapter 2<br />
Scheme 1. Two-step synthetic pathway <strong>for</strong> the preparation of TMC avoiding O-methylation.<br />
Synthesis of O-methylated TMC from chitosan. TMC w<strong>as</strong> synthesized by methylation of<br />
chitosan with iodomethane in presence of an aqueous solution of NaOH essentially <strong>as</strong><br />
described previously [7]. In detail, 1 g of chitosan and 2.4 g of sodium iodide were added to a<br />
mixture of 40 ml of NMP and 6 ml of 15% w/v aqueous NaOH solution. Subsequently, the<br />
mixture w<strong>as</strong> heated to 60°C and after stirring <strong>for</strong> 20 min, 6 ml of methyl iodide w<strong>as</strong> added and<br />
30
Synthesis and Characterization of O-methyl Free TMC<br />
the reaction mixture w<strong>as</strong> refluxed <strong>for</strong> 60 min. The reaction w<strong>as</strong> stopped by dropping the<br />
mixture in a 200 ml mixture of diethyl ether and ethanol (50/50). The obtained precipitate<br />
w<strong>as</strong> w<strong>as</strong>hed extensively with diethyl ether. In this way, TMC with a DQ of about 20% and<br />
comparable O-methylation w<strong>as</strong> obtained (step 1). To synthesize TMC with a DQ of around 40-<br />
50%, be<strong>for</strong>e precipitation, 3 ml of 15% NaOH solution and 3 ml of iodomethane were added<br />
and the solution w<strong>as</strong> stirred <strong>for</strong> another 60 minutes be<strong>for</strong>e stopping the reaction <strong>as</strong> described<br />
above (step 1.5). TMCs with higher DQs (60% to 90%) were synthesized by dissolving the<br />
dried TMC (DQ ~ 20) together with 2.4 g of sodium iodide in 40 ml of NMP at 60°C.<br />
Subsequently, 5.5 ml of an aqueous 15% w/v NaOH solution w<strong>as</strong> added and, after stirring <strong>for</strong><br />
20 min 3.5 ml CH 3I w<strong>as</strong> added and the reaction w<strong>as</strong> done <strong>for</strong> 45 minutes under refluxing to<br />
yield TMC with a DQ of about 60 to 70% (step 2). To obtain TMC with a DQ of about 80 to 90%,<br />
after 45 minutes, 0.6 g NaOH and 1 ml iodomethane were added and the reaction w<strong>as</strong><br />
continued <strong>for</strong> 60 min at 60°C (step 2.5). The reaction w<strong>as</strong> terminated by precipitation the<br />
reaction mixture in 200 ml of a mixture of diethyl ether and ethanol (50:50) and the<br />
precipitate w<strong>as</strong> w<strong>as</strong>hed extensively with diethyl ether.<br />
Finally, the products were dissolved in 50 ml aqueous 10% w/v NaCl solution and put on a<br />
shaker <strong>for</strong> a minimum of 18 hours <strong>for</strong> ion-exchange. The obtained solution w<strong>as</strong> dialyzed at<br />
room temperature against deionized water <strong>for</strong> 3 days changing buffer twice daily, filtered<br />
through a 0.8 µm filter and freeze dried. Analysis on NMR 500MHz w<strong>as</strong> per<strong>for</strong>med to<br />
determine the DQ and the degree of O-methylation.<br />
31
Chapter 2<br />
Scheme 2. Synthetic pathway <strong>for</strong> the preparation of TMC according to the method of Sieval et al [7].<br />
Determination of the degrees of dimethylation, quaternization and O-methylation. The<br />
1H-NMR spectra of the DMC and the various TMCs were recorded with a Varian INOVA<br />
500MHz NMR spectrometer (Varian Inc., Palo Alto, Ca, USA) at 80°C in D 2O. The degree of<br />
dimethylation of the DMC w<strong>as</strong> calculated <strong>as</strong> follows:<br />
DDM = [(CH 3) 2]/[H2-H6] x 100<br />
Here, [(CH 3) 2] is the integral of the hydrogens of the dimethyl amino groups at 2.9 ppm and<br />
[H2-H6] is the integral corresponding the H-2 to H-6 protons between 4.0 and 3.2 ppm.<br />
32
Synthesis and Characterization of O-methyl Free TMC<br />
The DQ, degree of dimethylation (DM) and degree of 3- and 6-O-methylation (DOM-3 and<br />
DOM-6, respectively) of the TMCs were calculated according to previous described methods[7,<br />
15, 24].<br />
DQ = [[(CH 3) 3]/[H] × 1/9] × 100<br />
DM = [[(CH 3) 2]/[H] × 1/6] × 100<br />
DOM = [[CH 3]/[H] × 1/3] × 100<br />
Here, [(CH 3) 3], [(CH 3) 2] and [CH 3] are the integrals of the hydrogens of the trimethylated amino<br />
groups at 3.3 ppm, the dimethylated amino groups at 2.9 ppm and the methylated hydroxyl<br />
groups at either 3.4 (DOM-6) or 3.5 (DOM-3) ppm, respectively. [H] is the integral of the H-1<br />
peaks between 4.7 and 5.7 ppm; the signal related to the hydrogen atoms bound to the C-1’s of<br />
the TMC molecule. For the DMC and TMC synthesized with the mild method and a DQ below<br />
25% addition of 0.05 ml of DCl w<strong>as</strong> needed to dissolve the polymers in D 2O.<br />
Determination of M n and M w of chitosan and TMC. M n and M w of chitosan and the various<br />
TMCs were determined by gel permeation chromatography (GPC) on a Viscotek-triple<br />
detection system using a Shodex OHPak SB-806 column (30 cm) and 0.3 M sodium acetate pH<br />
4.4 (adjusted with acetic acid) <strong>as</strong> running buffer [25]. To remove residual water, chitosan and<br />
the TMC samples were dried in a vacuum oven at 40°C overnight. Then, the samples were<br />
dissolved overnight in the running buffer at a concentration of 5 mg/ml, filtered through a 0.2<br />
µm filter and injected (50 µl); the flow rate w<strong>as</strong> 0.7 ml/min. Data from the l<strong>as</strong>er photometer (λ<br />
= 670 nm) (right (90 0 ) and low (7 0 ) angle light scattering), refracting index detector and<br />
viscosity detector were integrated using the provided Viscotek-software to calculate the M n,<br />
M w and dn/dc of the different samples. Pullulan (M n = 102 kDa, M w = 106 kDa) obtained from<br />
Viscotek Benelux (Oss, the Netherlands) w<strong>as</strong> used <strong>for</strong> calibration.<br />
Water solubility of chitosan and various TMCs. Aqueous solubility of the different<br />
polymers w<strong>as</strong> determined at pH 7 at room temperature. First the polymers were dissolved<br />
overnight in a 0.5% acetic acid solution at 2.5 mg/ml. Then the pH w<strong>as</strong> adjusted to 7 using 1 M<br />
NaOH and the transmittance of the solutions at 500 nm w<strong>as</strong> me<strong>as</strong>ured on an UV/VIS<br />
spectrophotometer (UV-2450, Shimadzu, Japan). The polymers were considered insoluble<br />
when the transmittance w<strong>as</strong> less than 90% compared to the transmittance of 0.5% acetic acid<br />
solution [26].<br />
33
Chapter 2<br />
Transepithelial electrical resistance (TEER) me<strong>as</strong>urements. Caco-2 cells were seeded at<br />
a density of 2x10 5 cells per well on 12-transwell plates with a microporous membrane. The<br />
cells were cultured in Dulbecco’s Modified Eagle’s Medium (DMEM) <strong>for</strong> 10 days until a<br />
confluent cell layer w<strong>as</strong> <strong>for</strong>med. The medium w<strong>as</strong> replaced by Hank’s Balanced Salt Solution<br />
(HBSS) at the b<strong>as</strong>olateral side 10 minutes be<strong>for</strong>e the start of the experiments. Then, 0.5 ml<br />
solution of TMC (with various DQs, with or without O-methylation, dissolved (2 mg/ml) in<br />
HBSS, pH adjusted to 7 with 0.1 M NaOH) w<strong>as</strong> applied at the apical site of the cell monolayers.<br />
SDS (10 mg/ml) w<strong>as</strong> used <strong>as</strong> positive control and HBSS <strong>as</strong> reference. The resistance me<strong>as</strong>ured<br />
of the membrane without cells w<strong>as</strong> used <strong>as</strong> blank. The TEER of the Caco-2 cells w<strong>as</strong> me<strong>as</strong>ured<br />
with a Millicell-ERS (Millipore, Billerica, USA) me<strong>as</strong>uring device at certain time points (0, 15,<br />
30, 45, 60 and 90 min.) after addition of the stimuli. After 90 minutes the cells were w<strong>as</strong>hed<br />
with HBSS and incubation of the cells w<strong>as</strong> continued in DMEM <strong>for</strong> 24 hours at 37°C, CO 2 5% to<br />
determine the recovery of the TEER [24, 27].<br />
MTT cell toxicity <strong>as</strong>say. Caco-2 cells were seeded in a 96-well plate at a density of 4x10 4<br />
cells per well and incubated <strong>for</strong> 2 days at 37°C, CO 2 5% in culture medium (DMEM, high<br />
glucose, 10% FCS, L-glutamine, pyruvate, non essential amino acids). The medium w<strong>as</strong><br />
removed and the cells were incubated <strong>for</strong> 2.5 hours with 100 µl TMC solutions in HBSS (TMC<br />
concentrations were 0.1, 1 and 10 mg/ml, pH set at 7 with 0.1 M NaOH). SDS (10 mg/ml) w<strong>as</strong><br />
used <strong>as</strong> positive control and HBSS <strong>as</strong> reference <strong>for</strong> 100% cell viability. Thereafter, the HBSS<br />
w<strong>as</strong> removed and the cells were w<strong>as</strong>hed with phosphate buffered saline. One hundred µl of a<br />
freshly prepared solution of 0.5 mg/ml MTT in DMEM, without any additions, w<strong>as</strong> added and<br />
the cells were incubated <strong>for</strong> 3 hours at 37°C and 5% CO 2. Subsequently, the wells were<br />
emptied, 100 µl of DMSO w<strong>as</strong> used to dissolve the <strong>for</strong>med <strong>for</strong>mazan crystals and the<br />
absorbance w<strong>as</strong> read at 595 nm [28].<br />
LDH <strong>as</strong>say. Caco-2 cells were seeded in a 96-well plate at a density of 4x10 4 cells per well<br />
and incubated <strong>for</strong> 2 days at 37°C, CO 2 5% in culture medium (see section on MTT <strong>as</strong>say <strong>for</strong><br />
composition). The medium w<strong>as</strong> removed and the cells were w<strong>as</strong>hed with HBSS and incubated<br />
<strong>for</strong> 2.5 hours with 100 µl TMC solutions in HBSS (TMC concentrations were 0.1, 1 and 10<br />
mg/ml, pH set at 7 with 0.1 M NaOH). After incubation, the concentration of LDH present in the<br />
supernatant of the samples w<strong>as</strong> determined with the Cytotoxicity Detection Kit-Plus (Roche<br />
Diagnostics, Mannheim, Germany) by me<strong>as</strong>uring absorbance at 490 nm with 650 nm <strong>as</strong> a<br />
reference wavelength. A calibration curve w<strong>as</strong> made with the lysis buffer provided by the<br />
34
Synthesis and Characterization of O-methyl Free TMC<br />
manufacturer, setting the LDH concentration me<strong>as</strong>ured with the undiluted lysis buffer at 100%<br />
LDH rele<strong>as</strong>e. HBSS w<strong>as</strong> used <strong>as</strong> a negative control.<br />
Results and discussion<br />
Synthesis and characterization. To synthesize O-methyl free TMC, chitosan w<strong>as</strong> first<br />
converted into DMC using the Eschweiler-Clarke reaction with <strong>for</strong>maldehyde and <strong>for</strong>mic acid<br />
(Figure 1). 1 H-NMR analysis showed that the obtained polymer had a degree of dimethylation<br />
of 83%. Since the chitosan used in this study had a degree of acetylation of around 17 % it can<br />
be concluded that the free amines were quantitatively dimethylated. In the next step, DMC<br />
dissolved in NMP w<strong>as</strong> converted into TMC using an excess amount of iodomethane. Figure 1<br />
shows that the degree of quaternization can be accurately tailored by varying the reaction time<br />
while keeping reaction temperature and DMC/CH 3I ratio constant. In other published<br />
procedures to synthesize TMC, sodium iodide w<strong>as</strong> added to the reaction mixture of chitosan<br />
and iodomethane [7, 21, 22]. However, the presence of sodium iodide (0.011 M) during the<br />
synthesis of TMC from DMC using CH 3I did not affect the obtained DQ in our studies (data not<br />
shown).<br />
degree of quaternization<br />
(%)<br />
70<br />
60<br />
50<br />
40<br />
30<br />
20<br />
10<br />
0<br />
0 10 20 30 40 50 60 70<br />
reaction time (h)<br />
Figure 1. Effect of reaction time on the DQ of TMC. The reaction temperature w<strong>as</strong> 40°C. Error bars<br />
represent standard deviation of three independent syntheses.<br />
35
Chapter 2<br />
Figure 2. 1 H-NMR-spectra of TMC with a DQ of about 60% synthesized with a) one-step method under<br />
alkaline conditions and b) the two-step method presented in this article. Red oval indicates signals<br />
correlated to O-methylation.<br />
Typical 1 H-NMR spectra of TMC synthesized with the new procedure described in this paper<br />
and with the procedure described by Sieval et al. are shown in Figure 2. From the spectra it can<br />
be concluded that both TMCs have a DQ of about 60%. The spectrum of TMC synthesized<br />
36
Synthesis and Characterization of O-methyl Free TMC<br />
according to the procedure of Sieval et al. (Figure 2a) clearly shows O-methylation of the<br />
hydroxyl groups at the C-3 and C-6 of the glucosamine units (peaks observed at 3.5 and 3.4<br />
ppm, respectively). It w<strong>as</strong> calculated that DOM-6 and DOM-3 are 56 and 44%, respectively. The<br />
higher degree of methylation observed on the C-6 hydroxyl groups is likely because this<br />
hydroxyl group is less sterically hindered than the hydroxyl group on the C-3. An overview of<br />
the 1 H-NMR results of the O-methylated TMCs is presented in Table 1. As also reported by<br />
others [15, 16], both the DQ and the DOM-3 and DOM-6 incre<strong>as</strong>ed with the number of reaction<br />
steps, but the DQ and DOM proved hard to control. In contr<strong>as</strong>t, 1 H-NMR analysis of TMC with a<br />
DQ of 56% synthesized with the new method described in this paper shows no peaks at 3.4<br />
and 3.5 ppm (Figure 2b), demonstrating the absence of O-methylation.<br />
Table 1. Degree of quaternization (DQ), dimethylation (DM), O-methylation on C-6 (DOM-6) and on<br />
C-3 (DOM-3) of TMCs synthesized according to Sieval et al. [7] <strong>as</strong> determined by 1 H-NMR analysis.<br />
DQ DM DOM-6 DOM-3<br />
Step 1 22% 61% 18% 12%<br />
Step 1.5 50% 35% 45% 40%<br />
Step 2 61% 25% 56% 44%<br />
Step 2.5 86% 1% 76% 72%<br />
In line with the results of Snyman et al. [14], GPC analysis showed a slightly decre<strong>as</strong>ed<br />
molecular weight of TMC synthesized in presence of aqueous NaOH with incre<strong>as</strong>ing DQ (Table<br />
2). Under the alkaline reaction conditions hydrolysis of some glycosidic bonds linking the<br />
glucosamine units likely occurs, thus with incre<strong>as</strong>ing reaction time, the molecular weight and<br />
the polymer chain length of the O-methylated TMCs decre<strong>as</strong>es. Although acid hydrolysis of<br />
chitosan h<strong>as</strong> been reported [29], we show that the molecular weight incre<strong>as</strong>es after converting<br />
chitosan into DMC (Table 2) which demonstrates that no or only limited chain scission occurs<br />
during reaction. The slight incre<strong>as</strong>e in molecular weight can be <strong>as</strong>cribed to the addition of<br />
methyl groups to the molecule. Further, GPC analysis also shows that TMC synthesized from<br />
DMC shows a slight incre<strong>as</strong>e in molecular weight with incre<strong>as</strong>ing DQ (Table 2). There<strong>for</strong>e, <strong>as</strong><br />
expected, the trimethylation of DMC with the synthetic method described in this thesis is not<br />
<strong>as</strong>sociated with chain scission, and all O-methyl free TMCs will have the same polymer chain<br />
length. This is in contr<strong>as</strong>t with the ‘standard’ method to synthesize TMC <strong>as</strong> described by Sieval<br />
et al [7] where TMCs with varying DQ will also have discrepancies in polymer chain lengths.<br />
37
Chapter 2<br />
Table 2. Molecular weights and water solubility of various derivatives of chitosan.<br />
M n M w dn/dc Solubility in water<br />
at pH 7<br />
Chitosan 25 kDa 42 kDa 0.18 -<br />
TMC-OM 22% 34 kDa 56 kDa 0.15 +<br />
TMC-OM 50% 32 kDa 49 kDa 0.15 +<br />
TMC-OM 61% 31 kDa 49 kDa 0.14 +<br />
TMC-OM 86% 29 kDa 44 kDa 0.14 +<br />
DMC 28 kDa 57 kDa 0.16 -<br />
TMC 22% 31 kDa 60 kDa 0.16 -<br />
TMC 30% 33 kDa 59 kDa 0.15 +<br />
TMC 43% 36 kDa 75 kDa 0.15 +<br />
TMC 56% 37 kDa 78 kDa 0.15 +<br />
TMC 68% 39 kDa 84 kDa 0.15 +<br />
All TMCs tested (with or without O-methylation) and a DQ >22% were readily soluble in<br />
aqueous solutions at pH 7. As expected, chitosan and DMC became insoluble when the pH w<strong>as</strong><br />
incre<strong>as</strong>ed to 7. Remarkably, O-methyl free TMC with a DQ 22% w<strong>as</strong> insoluble at pH 7 at a<br />
concentration of 2.5 mg/ml while TMC with around the same molecular weight and DQ (22%)<br />
and a DOM of 18 and 12% at C-6 and C-3, respectively, w<strong>as</strong> readily dissolved in the same<br />
solvent. This is rather unexpected since O-methylated glucosamine units can only act <strong>as</strong><br />
hydrogen-acceptor, where<strong>as</strong> non O-methylated units can both accept and donate hydrogen<br />
bridges in interaction with aqueous solvents. Possibly, TMC with low DQs will still partially<br />
resemble the behavior of chitosan, where intra- and intermolecular interactions decre<strong>as</strong>e the<br />
aqueous solubility at pH 7 [1]. Partial O-methylation might reduce these interactions and<br />
there<strong>for</strong>e result in better aqueous solubility of O-methylated TMCs with a DQ
Synthesis and Characterization of O-methyl Free TMC<br />
50% <strong>for</strong> penetration enhancing effects [10, 18, 19]. However, the possible contribution of O-<br />
methylation on these effects is unknown. The availability of TMCs with varying DQ but without<br />
O-methylated groups or discrepancies in polymer chain lengths allows a better evaluation of<br />
the effect of DQ on the biological properties of TMC. Figures 3a and 3b show the effect of TMCs<br />
with different DQ with and without (partial) O-methylation on the TEER. The results of the O-<br />
methylated TMCs are in line with those reported by other researchers [6, 10, 18, 24], namely<br />
TMC with a DQ of 22% and DOM-3 and DOM-6 of 12 and 18%, respectively, showed no effect<br />
on the TEER, indicating that the polymer is unable to open the tight junctions. The TMC with a<br />
DQ of 61% and similar O-methylation demonstrates the largest decre<strong>as</strong>e in the TEER and<br />
showed marginal toxicity using the MTT <strong>as</strong>say (Figure 4a) implying that this polymer induces<br />
opening of tight junctions without exerting major acute toxic effects. However, since the LDH<br />
levels (Figure 5a) are elevated, this TMC apparently induces some membrane damage to cells.<br />
Interestingly, <strong>as</strong> compared to O-methylated TMC with a DQ of 61%, TMC with the highest DQ<br />
(86%) had similar effect on the TEER while it showed substantial cellular toxicity (MTT <strong>as</strong>say)<br />
at a concentration of 10 mg/ml. The highly elevated LDH rele<strong>as</strong>e upon exposure of the Caco-2<br />
cells to this polymer supports the data obtained with the MTT <strong>as</strong>say. Other groups [10, 18]<br />
reported the highest tight-junction opening activity of O-methylated TMC with a DQ of about<br />
40-50%. However, the molecular weights of the TMCs used in those studies were higher which<br />
may account <strong>for</strong> the somewhat different optimal DQ found in these studies.<br />
39
Chapter 2<br />
TEER (% HBSS)<br />
A<br />
SDS 1%<br />
100 TMC-OM 22%<br />
TMC-OM 50%<br />
80<br />
TMC-OM 61%<br />
60<br />
TMC-OM 86%<br />
40<br />
20<br />
0<br />
0 15 30 45 60 75 90<br />
Time (min.)<br />
1140<br />
TEER (% HBSS)<br />
B<br />
100 SDS 1%<br />
80<br />
60<br />
40<br />
20<br />
TMC 30%<br />
TMC 43%<br />
TMC 56%<br />
TMC 68%<br />
0<br />
0 15 30 45 60 75 90<br />
Time (min.)<br />
1140<br />
Figure 3. Effect of (a) O-methylated TMC and (b) O-methyl free TMC with various DQs on the TEER of<br />
Caco-2 cells at a TMC concentration of 2 mg/ml. SDS 10 mg/ml w<strong>as</strong> used <strong>as</strong> a positive control. Error bars<br />
represent the standard deviation of six me<strong>as</strong>urements.<br />
Figure 3 shows that O-methyl free TMC with a DQ of 30% had a similar TEER effect <strong>as</strong> the O-<br />
methylated TMC with a DQ of 56%, where<strong>as</strong> the MTT and LDH <strong>as</strong>says showed some toxicity<br />
above a concentration of 1 mg/ml (Figures 4b and 5b). Figure 3b also shows that the effect on<br />
the TEER slightly incre<strong>as</strong>ed (reduction from 35% to 20% remaining resistance compared with<br />
HBSS) with incre<strong>as</strong>ing DQ (from 30 to 68%). This might indicate that with incre<strong>as</strong>ing DQ the<br />
polymers have greater capacity to open the tight junctions, but the decre<strong>as</strong>e in TEER can also<br />
be the result of acute toxic effects on the Caco-2 cells (see Figures 4b and 5b). The MTT <strong>as</strong>say<br />
clearly displays a DQ depending cytotoxicity <strong>for</strong> O-methyl free TMCs and these polymers show<br />
40
Synthesis and Characterization of O-methyl Free TMC<br />
a larger decre<strong>as</strong>e in cell viability than O-methylated TMC. The LDH <strong>as</strong>say shows a DQdependent<br />
LDH rele<strong>as</strong>e <strong>for</strong> both O-methylated and O-methyl free TMCs. Interestingly, although<br />
the O-methyl free TMCs demonstrate a higher LDH rele<strong>as</strong>e than the O-methylated TMCs, the<br />
difference is much less prominent <strong>as</strong> with the MTT <strong>as</strong>say; O-methylated TMC with a DQ of 56%<br />
induces a similar LDH rele<strong>as</strong>e <strong>as</strong> O-methyl free TMC with a DQ of 30% and the LDH rele<strong>as</strong>e of<br />
O-methylated TMC with a DQ of 86% is comparable to the amount of LDH rele<strong>as</strong>ed by<br />
application of O-methyl free TMC with a DQ of 68%. Partial recovery of the TEER w<strong>as</strong> observed<br />
(24 hours after removal of the polymer) <strong>for</strong> all TMCs (with or without O-methylation), which<br />
indicates that cells were still viable after the removal of the stimulus. Taken together, O-methyl<br />
free TMCs lead to a larger decre<strong>as</strong>e of the TEER than O-methylated TMCs but the penetrationenhancing<br />
effects of the TMCs without O-methylation remain to be established.<br />
Cell viability (%)<br />
A<br />
100<br />
80<br />
60<br />
40<br />
20<br />
TMC-OM 22%<br />
TMC-OM 50%<br />
TMC-OM 61%<br />
TMC-OM 86%<br />
0<br />
0.1 1 10<br />
Conc. (mg/ml)<br />
Cell viability (%)<br />
B<br />
100 TMC 30%<br />
TMC 43%<br />
80<br />
TMC 56%<br />
60<br />
TMC 68%<br />
40<br />
20<br />
0<br />
0.1 1 10<br />
Conc. (mg/ml)<br />
Figure 4. Effect of (a) O-methylated TMC and (b) O-methyl free TMC with various DQs on the viability of<br />
Caco-2 cells (MTT <strong>as</strong>say) at different concentrations. Error bars represent the standard deviation of six<br />
me<strong>as</strong>urements.<br />
41
Chapter 2<br />
The synthesis method described in this paper provides a better defined TMC compared to<br />
the traditionally synthesized TMC with unavoidable O-methylation [15]. Moreover, the lack of<br />
chain scission during the quaternization of TMC allows studies of TMCs with varying DQ<br />
without discrepancies in polymer chain lengths.<br />
Our studies demonstrate that O-methyl free TMCs have a stronger effect on the Caco-2 cells<br />
in the TEER, MTT and LDH <strong>as</strong>says than the O-methylated TMCs. Since the molecular weights of<br />
the polymers used in the present study are comparable (Table 2), the results imply a reduction<br />
in toxicity of TMC by the (partial) O-methylation of the polymer. Additionally, TMC without O-<br />
methylation resembles to a larger extent the original chitosan polymer. This may have<br />
beneficial effects on the enzymatic degradability of the polymer.<br />
LDH rele<strong>as</strong>e (%)<br />
A<br />
30<br />
20<br />
10<br />
HBSS<br />
TMC-OM 22%<br />
TMC-OM 50%<br />
TMC-OM 61%<br />
TMC-OM 86%<br />
0<br />
0.1 1.0 10.0<br />
Conc. (mg/ml)<br />
LDH rele<strong>as</strong>e (%)<br />
B<br />
30<br />
20<br />
10<br />
HBSS<br />
TMC 30%<br />
TMC 43%<br />
TMC 56%<br />
TMC 68%<br />
0<br />
0.1 1.0 10.0<br />
Conc. (mg/ml)<br />
Figure 5. Effect of (a) O-methylated TMC and (b) O-methyl free TMC with various DQs on the viability of<br />
Caco-2 cells (LDH rele<strong>as</strong>e) at different concentrations. Error bars represent the standard deviation of<br />
three me<strong>as</strong>urements.<br />
42
Synthesis and Characterization of O-methyl Free TMC<br />
TMCs without O-methylation may improve the penetration-enhancing effects of the<br />
‘traditional’ TMCs, demonstrating a strong (reversible) effect of the opening of tight junctions.<br />
Furthermore, it shows limited toxicity at concentrations used <strong>for</strong> cell transfection experiments<br />
[11, 12]. Since incorporation into nanoparticles reduces the toxicity of TMC [24], this non O-<br />
methylated TMC may be very suitable <strong>for</strong> incorporation into nanoparticles <strong>for</strong> mucosal<br />
vaccination, which is currently under investigation.<br />
Conclusion<br />
This paper shows a straight<strong>for</strong>ward method to synthesize TMC avoiding O-methylation while<br />
preventing chain scission and to tailor the DQ of the TMC by varying the reaction time. O-<br />
methyl free TMCs with a DQ > 30% are readily soluble in aqueous solutions at pH 7. In the<br />
TEER, MTT and LDH <strong>as</strong>says O-methyl free TMCs have a stronger effect on the Caco-2 cells than<br />
the O-methylated TMCs. In conclusion, these new well-characterized O-methyl free TMCs have<br />
potentially favorable biological characteristics over the traditional TMC and allow to effectively<br />
study the influence of the DQ of TMC in various delivery systems.<br />
Acknowledgement. This research w<strong>as</strong> per<strong>for</strong>med under the framework of <strong>TI</strong> <strong>Pharma</strong><br />
project number D5-106-1; Vaccine delivery: alternatives <strong>for</strong> conventional multiple injection<br />
vaccines.<br />
43
Chapter 2<br />
References<br />
1. Kumar MNVR, Muzzarelli RAA, Muzzarelli C, S<strong>as</strong>hiwa H, Domb AJ. Chitosan chemistry<br />
and pharmaceutical perspectives. Chem Rev 104:6017-84 (2004)<br />
2. Agnihotri SA, Mallikarjuna NN, Aminabhavi TM. Recent advances on chitosan-b<strong>as</strong>ed<br />
micro- and nanoparticles in drug delivery. J Control Rele<strong>as</strong>e 100:5-28 (2004).<br />
3. Aspden TJ, Illum L, Skaugrud Ø. Chitosan <strong>as</strong> a n<strong>as</strong>al delivery system: evaluation of insulin<br />
absorption enhancement and effect on n<strong>as</strong>al membrane integrity using rat models. Eur J<br />
Pharm Sci 4:23-31 (1996).<br />
4. Van Der Lubben IM, Kersten G, Fretz MM, Beuvery C, Verhoef JC, Junginger HE.<br />
Chitosan microparticles <strong>for</strong> mucosal vaccination against diphtheria: oral and n<strong>as</strong>al efficacy<br />
studies in mice. Vaccine 21:1400-8 (2003).<br />
5. Kotze AF, Luessen HL, de Boer AG, Verhoef JC, Junginger HE. Chitosan <strong>for</strong> enhanced<br />
intestinal permeability: prospects <strong>for</strong> derivatives soluble in neutral and b<strong>as</strong>ic environments.<br />
Eur J Pharm Sci 7:145-51 (1999).<br />
6. Kotze AF, Luessen HL, de Leeuw BJ, de Boer AG, Verhoef JC, Junginger HE. N-<strong>Trimethyl</strong><br />
chitosan chloride <strong>as</strong> a potential absorption enhancer across mucosal surfaces: in vitro evalutation<br />
in intestinal epithelial cells (Caco-2). Pharm Res 14:1197-202 (1997).<br />
7. Sieval AB, Thanou M, Kotze AF, Verhoef JC, Brussee J, Junginger HE. Preparation and<br />
NMR characterization of highly substituted N-trimethyl chitosan chloride. Carbohydrate<br />
Polym 36:157-65 (1998).<br />
8. Van der Lubben IM, Verhoef JC, Fretz MM, Van O, Mesu I, Kersten G, et al. <strong>Trimethyl</strong><br />
chitosan chloride (TMC) <strong>as</strong> a novel excipient <strong>for</strong> oral and n<strong>as</strong>al immunisation against<br />
diphtheria. STP Pharm Sci 12:235-42 (2002).<br />
9. Snyman D, Hamman JH, Kotze AF. Evaluation of the mucoadhesive properties of N-<br />
trimethyl chitosan chloride. Drug Develop Ind Pharm 29:61-9 (2003).<br />
10. Thanou MM, Kotze AF, Scharringhausen T, Lueßen HL, De Boer AG, Verhoef JC, et al.<br />
Effect of degree of quaternization of N-trimethyl chitosan chloride <strong>for</strong> enhanced transport of<br />
hydrophilic compounds across intestinal Caco-2 cell monolayers. J Control Rele<strong>as</strong>e 64:15-25<br />
(2000).<br />
11. Thanou M, Florea BI, Geldof M, Junginger HE, Borchard G. Quaternized chitosan oligomers<br />
<strong>as</strong> novel gene delivery vectors in epithelial cell lines. Biomaterials 23:153-9 (2002).<br />
12. Mao Z, Ma L, Jiang Y, Yan M, Gao C, Shen J. N,N,N-<strong>Trimethyl</strong>chitosan chloride <strong>as</strong> a<br />
gene vector: synthesis and application. Macromol Biosci 7:855-63 (2007).<br />
13. Domard A, Rinaudo M, Terr<strong>as</strong>sin C. New method <strong>for</strong> the quaternization of chitosan. Int J Biol<br />
Macromol 8:105-7 (1986).<br />
14. Snyman D, Hamman JH, Kotze JS, Rollings JE, Kotze AF. The relationship between the<br />
absolute molecular weight and the degree of quaternisation of N-trimethyl chitosan chloride.<br />
Carbohydrate Polym 50:145-50 (2002).<br />
15. Polnok A, Borchard G, Verhoef JC, Sarisuta N, Junginger HE. Influence of methylation<br />
process on the degree of quaternization of N-trimethyl chitosan chloride. Eur J Pharm<br />
Biopharm 57:77-83 (2004).<br />
16. Curti E, De Britto D, Campana-Filho SP. Methylation of chitosan with iodomethane: effect of<br />
reaction conditions on chemoselectivity and degree of substitution. Macromol Biosci 3:571-6<br />
(2003).<br />
17. Di Colo G, Burgal<strong>as</strong>si S, Zambito Y, Monti D, Chetoni P. Effects of different N-trimethyl<br />
chitosans on in vitro/in vivo ofloxacin transcorneal permeation. J Pharm Sci 93:2851-62 (2004).<br />
18. Hamman JH, Stander M, Kotze AF. Effect of the degree of quaternisation of N-trimethyl<br />
chitosan chloride on absorption enhancement: in vivo evaluation in rat n<strong>as</strong>al epithelia. Int J<br />
Pharm 232:235-42 (2002).<br />
44
Synthesis and Characterization of O-methyl Free TMC<br />
19. Boonyo W, Junginger HE, Waranuch N, Polnok A, Pitaksuteepong T. Chitosan and<br />
<strong>Trimethyl</strong> chitosan chloride (TMC) <strong>as</strong> adjuvants <strong>for</strong> inducing immune responses to ovalbumin<br />
in mice following n<strong>as</strong>al administration. J Control Rele<strong>as</strong>e 121:168-75 (2007).<br />
20. Muzzarelli RAA, Tanfani F. The N-permethylation of chitosan and the preparation of N-<br />
trimethyl chitosan iodide. Carbohydr Polym 5:297-307 (1985).<br />
21. Jia Z, shen D, Xu W. Synthesis and antibacterial activities of quaternary ammonium salt of<br />
chitosan. Carbohydr Res 333:1-6 (2001).<br />
22. Guo Z, Xing R, Liu S, Zhong Z, Ji X, Wang L, Li P. Antifungal properties of Schiff b<strong>as</strong>es of<br />
chitosan, N-substituted chitosan and quaternized chitosan. Carbohydr Res 342:1329-32 (2007).<br />
23. Pine SH, Sanchez BL. Formic acid-<strong>for</strong>maldehyde methylation of amines. J Org Chem 36:829-32<br />
(1971).<br />
24. Amidi M, Romeijn SG, Borchard G, Junginger HE, Hennink WE, Jiskoot W. Preparation<br />
and characterization of protein-loaded N-trimethyl chitosan nanoparticles <strong>as</strong> n<strong>as</strong>al delivery<br />
system. J Control Rele<strong>as</strong>e 111:107-16 (2006).<br />
25. Jiang X, van der Horst A, van Steenbergen M, Akeroyd N, van Nostrum C, Schoenmakers P,<br />
Hennink WE. Molar-m<strong>as</strong>s characterization of cationic polymers <strong>for</strong> gene delivery by<br />
aqueous size-exclusion chromatography. Pharm Res 23:595-603 (2006).<br />
26. Mao S, Shuai X, Unger F, Wittmar M, Xie X, Kissel T. Synthesis, characterization and<br />
cytotoxicity of poly(ethylene glycol)-graft-trimethyl chitosan block copolymers. Biomaterials<br />
26:6343-56 (2005).<br />
27. Nerurkar MM, Burton PS, Borchardt RT. The use of surfactants to enhance the permeability<br />
of peptides through Caco-2 cells by inhibition of an apically polarized efflux system. Pharm Res<br />
13:528-34 (1996).<br />
28. Mosmann TJ. Rapid colorimetric <strong>as</strong>say <strong>for</strong> cellular growth and survival: application to<br />
proliferation and cytotoxicity <strong>as</strong>says. J Immunol Methods 65:55-65 (1983).<br />
29. Varum KM, Ottoy MH, Smidsrod O. Acid hydrolysis of chitosans. Carbohydrate Polym 46:89-98<br />
(2001).<br />
45
CHAPTER 3<br />
INFLUENCE OF THE DEGREE OF<br />
ACETYLA<strong>TI</strong>ON ON THE ENZYMA<strong>TI</strong>C<br />
DEGRADA<strong>TI</strong>ON AND IN VITRO BIOLOGICAL<br />
PROPER<strong>TI</strong>ES OF TRIMETHYLATED<br />
CHITOSANS<br />
Rolf J. Verheul, Maryam Amidi, Mies van Steenbergen,<br />
Elly van Riet, Wim Jiskoot, Wim E. Hennink.<br />
Biomaterials 2009, 30, 3129-3135
Chapter 3<br />
Abstract<br />
Chitosan derivatives such <strong>as</strong> N,N,N-trimethylated chitosan (TMC) are currently being<br />
investigated <strong>for</strong> the delivery of drugs, vaccines and genes. However, the influence of the extent<br />
of N-acetylation of these polymers on their enzymatic degradability and biological properties<br />
is unknown. In this study, TMCs with a degree of acetylation (DAc) ranging from 11 to 55%<br />
were synthesized by using a three-step method. First, chitosan w<strong>as</strong> partially re-acetylated<br />
using acetic anhydride followed by quantitative dimethylation using <strong>for</strong>maldehyde and sodium<br />
borohydride. Then, in presence of an excess amount of iodomethane, TMC w<strong>as</strong> synthesized.<br />
The TMCs obtained by this method showed neither detectable O-methylation nor loss in acetyl<br />
groups ( 1 H-NMR) and a slight incre<strong>as</strong>e in molecular weight (GPC) with incre<strong>as</strong>ing degree of<br />
substitution, implying that no chain scission occurred during synthesis. The extent of<br />
lysozyme-catalyzed degradation of TMC, and that of its precursors chitosan and dimethyl<br />
chitosan, w<strong>as</strong> highly dependent on the DAc and polymers with the highest DAc showed the<br />
largest decre<strong>as</strong>e in molecular weight. On Caco-2 cells, TMCs with a high DAc (~50%), a DQ of<br />
around 44% and with or without O-methylated groups, were not able to open tight junctions in<br />
the trans-epithelial electrical resistance (TEER) <strong>as</strong>say, in contr<strong>as</strong>t with TMCs (both O-<br />
methylated and O-methyl free; concentration 2.5 mg/ml) with a similar DQ but a lower DAc<br />
which were able to reduce the TEER with 30 and 70%, respectively. Additionally, TMCs with a<br />
high DAc (~50%) demonstrated no cell toxicity (MTT, LDH rele<strong>as</strong>e) up to a concentration of 10<br />
mg/ml.<br />
48
Influence of Degree of Acetylation on Degradation and Properties of TMC<br />
Introduction<br />
Chitosan is a polysaccharide consisting of β1→4-D-glucosamine and β1→4 N-acetyl D-<br />
glucosamine units and is obtained from the natural polymer chitin (poly β1→4 N-acetyl D-<br />
glucosamine) by partial deacetylation. Chitosan h<strong>as</strong> been under investigation <strong>for</strong> various<br />
biomedical and pharmaceutical applications due to its biocompatibility, low toxicity and its<br />
muco-adhesive properties [1-4]. It h<strong>as</strong> been reported that chitosan can be degraded in human<br />
tissues by several enzymes (e.g. lysozyme, chitin<strong>as</strong>e), by acid hydrolysis and by oxidativereductive<br />
depolymerization (ORD) reactions [5, 6]. Nordtveit et al. demonstrated that the ORD<br />
of chitosan driven by hydrogen peroxide is independent of the amount of residual N-acetylated<br />
units [7]. However, in vitro studies on the degradation of chitosan by acid hydrolysis [8] and by<br />
enzymes [6, 7, 9-14] revealed a major dependency on the degree of acetylation (DAc).<br />
Completely de-acetylated chitosan shows very limited degradation by (enzyme catalyzed)<br />
hydrolysis while the extent of degradation and degradation kinetics incre<strong>as</strong>e with the extent of<br />
acetylation. Lysozyme is widely present in human body fluids (e.g. serum, saliva, tears) and is<br />
actively secreted by macrophages and neutrophils [15], and consequently the enzymatic<br />
degradation of chitosan by lysozyme h<strong>as</strong> been studied in depth [7, 9-13, 16, 17]. Temperature,<br />
pH and ionic strength have been found to influence degradation kinetics [10]. Additionally, in<br />
vivo experiments on the degradability of chitosan have shown that the DAc also plays a key<br />
role in the depolymerization of chitosan in living animals [13, 17]. Furthermore, uptake-,<br />
toxicity- and cell transfection studies with chitosan suggest that the degree of acetylation h<strong>as</strong><br />
an important influence on biological polymer characteristics <strong>as</strong> well [18-20]. Finally, partially<br />
acetylated chitosan h<strong>as</strong> been found to have better adjuvant properties <strong>for</strong> macrophage<br />
stimulation compared to chitosan with a low amount of residual acetylated groups or chitin<br />
[21-23], and recently, it w<strong>as</strong> reported that N-acetylated glucosamines can bind to C-type lectin<br />
receptors on denditric cells thereby working <strong>as</strong> an adjuvant [24].<br />
In contr<strong>as</strong>t to chitosan, N,N,N,-trimethylated chitosan (TMC), a partially quaternized<br />
derivative of chitosan, is water-soluble at neutral pH. TMC h<strong>as</strong> been widely studied in the<br />
biomedical field <strong>as</strong> drug, vaccine and gene delivery vehicle [25-31]. It h<strong>as</strong> been shown in<br />
several in vitro and in vivo models that TMC h<strong>as</strong> limited toxicity, possesses muco-adhesive<br />
properties and can incre<strong>as</strong>e the uptake of small drug molecules <strong>as</strong> well <strong>as</strong> proteins via various<br />
mucosal routes [25-27, 29, 30, 32-35]. Several investigators have studied the optimal degree of<br />
quaternization (DQ) of TMC <strong>for</strong> mucosal transport and gene delivery [26, 33, 36, 37]. However,<br />
molecular weight [38, 39] and, <strong>as</strong> demonstrated in chapter 2, O-methylation [40] also have a<br />
major impact on the physical and biological characteristics of TMC. Besides its solubility at pH<br />
49
Chapter 3<br />
7, another often suggested potentially beneficial characteristic of TMC is its biodegradability.<br />
These biodegradability claims, however, are b<strong>as</strong>ed on degradation studies per<strong>for</strong>med on<br />
chitosan, and no investigations on the enzymatic degradability of TMC have been carried out so<br />
far. The aim of this paper w<strong>as</strong> to investigate the lysozyme-catalyzed degradability of TMCs<br />
with different DAcs and with or without O-methylated groups, and to evaluate these polymers<br />
<strong>for</strong> their physico-chemical properties, cytotoxicity and ability to open tight junctions.<br />
Materials and Methods<br />
Materials. Chitosan with a residual degree of acetylation of 17% (determined with 1 H-NMR<br />
<strong>as</strong> described in below) and a number average molecular weight (M n) and weight average<br />
molecular weight (M w) of 28 and 43 kDa, (determined with GPC-TD <strong>as</strong> described below),<br />
respectively, w<strong>as</strong> purch<strong>as</strong>ed from Primex (Siglufjodur, Iceland). Hen-egg white lysozyme<br />
(41800 units per mg solid), acetic anhydride, sodium borohydride, <strong>for</strong>mic acid, <strong>for</strong>maldehyde<br />
37% (stabilized with methanol), DCl 35% w/w in D 2O, sodium azide, deuterium oxide, sodium<br />
acetate, acetic acid (anhydrous), sodium hydroxide and hydrochloric acid were obtained from<br />
Sigma-Aldrich Chemical Co. Dulbecco’s Modified Eagle’s Medium (DMEM), Hank’s balanced salt<br />
solution (HBSS) and fetal calf serum (FCS) were obtained from Invitrogen (Breda, The<br />
Netherlands). Sodium dodecyl sulfate (SDS) and Sicapent were ordered from Merck<br />
(Darmstadt, Germany). Iodomethane 99% stabilized with copper w<strong>as</strong> obtained from Acros<br />
Organics (Geel, Belgium). All other chemicals used were of analytical grade.<br />
Acetylation of chitosan. Chitosan (degree of acetylation 17%) w<strong>as</strong> used <strong>as</strong> obtained and<br />
acetylation w<strong>as</strong> carried out according to a previously described method [19]. Briefly, chitosan<br />
(10 g) w<strong>as</strong> dissolved in 1% acetic acid (500 ml). Then, 500 ml methanol and 2.7 ml acetic<br />
anhydride were added. The resulting mixture w<strong>as</strong> stirred overnight at room temperature in a<br />
roundbottom fl<strong>as</strong>k. Next, the reacetylated chitosan w<strong>as</strong> precipitated by dropping the reaction<br />
mixture into an aqueous solution of 1 M NaOH, filtrated using a gl<strong>as</strong>s filter and w<strong>as</strong>hed<br />
extensively with methanol. Then, the precipitate w<strong>as</strong> dissolved in an aqueous acidic solution<br />
(pH 4, adjusted with 1 M HCl) and dialyzed against deionized water <strong>for</strong> 3 days changing buffer<br />
twice-daily. Finally, the solution w<strong>as</strong> filtered through a 0.8 µm filter and the polymer w<strong>as</strong><br />
collected after freeze-drying.<br />
50
Influence of Degree of Acetylation on Degradation and Properties of TMC<br />
Synthesis of dimethylated chitosan. Dimethylated chitosan (DMC) with a degree of<br />
acetylation of 17% w<strong>as</strong> synthesized <strong>as</strong> described previously in chapter 2 [40]. In short,<br />
chitosan (5 g) w<strong>as</strong> dissolved in a mixture of 90 ml deionized water and 15 ml <strong>for</strong>mic acid.<br />
Subsequently, 20 ml of 37% <strong>for</strong>maldehyde w<strong>as</strong> added, and the solution w<strong>as</strong> heated to 70 °C<br />
and stirred under refluxing <strong>for</strong> 2 days. The slightly yellow, viscous solution w<strong>as</strong> evaporated<br />
under reduced pressure and the pH w<strong>as</strong> adjusted to 8 using 1 M NaOH resulting in the<br />
<strong>for</strong>mation of a gel, which w<strong>as</strong> extensively w<strong>as</strong>hed with deionized water to remove impurities.<br />
Then, the DMC w<strong>as</strong> dissolved in deionized water at pH 4 (adjusted with 1 M HCl), filtered over<br />
a gl<strong>as</strong>s filter and dialyzed against deionized water <strong>for</strong> three days (changing buffer twice-daily).<br />
Finally, the product w<strong>as</strong> filtered through a 0.8 µm filter and collected after freeze-drying.<br />
GPC analysis revealed that under these conditions, some chain scission occurred with<br />
chitosan with a DAc of 55% and there<strong>for</strong>e less vigorous reaction conditions were applied to<br />
obtain dimethylated chitosan with a high degree of acetylation. Here, sodium borohydride<br />
instead of <strong>for</strong>mic acid w<strong>as</strong> used <strong>as</strong> a reductor to allow the reaction to take place at room<br />
temperature [41, 42]. In detail, chitosan (1 g) with a degree of acetylation of 55% w<strong>as</strong><br />
dissolved in 50 ml of 2% acetic acid (v/v). Next, 10 ml of 37% <strong>for</strong>maldehyde solution w<strong>as</strong><br />
added, the pH w<strong>as</strong> adjusted to 5 with 1M NaOH and the mixture w<strong>as</strong> stirred <strong>for</strong> 2 hours at<br />
room temperature. Subsequently, sodium borohydride (2.5 g) w<strong>as</strong> added in portions of 250 mg<br />
over a period of 24 hours, adjusting the pH to 5 with 1M HCl after each addition. After<br />
precipitation with acetone, the product w<strong>as</strong> w<strong>as</strong>hed extensively with acetone and dried<br />
overnight.<br />
Synthesis of O-methyl free trimethylated chitosan. DMC with different degrees of<br />
acetylation synthesized from chitosan with a DAc of 55% or 17% <strong>as</strong> described above, were<br />
reacted with iodomethane to yield trimethylated chitosan (TMC). To prevent O-methylation,<br />
the reaction of DMC with iodomethane w<strong>as</strong> done in NMP, without the addition of a b<strong>as</strong>e<br />
catalyst. TMC with a degree of acetylation of 17% and a degree of quaternization of around<br />
45% w<strong>as</strong> obtained by the method described previously [40]. TMC with a degree of acetylation<br />
of 55% and a DQ of 45% w<strong>as</strong> obtained with a slightly modified method. In detail, DMC (500<br />
mg) with a degree of acetylation of 55% w<strong>as</strong> dissolved in 80 ml deionized water and the pH<br />
w<strong>as</strong> adjusted to 11 with a 1 M solution of NaOH, resulting in gel <strong>for</strong>mation. Then, the gel w<strong>as</strong><br />
w<strong>as</strong>hed with water followed by acetone. To remove residual solvents, the DMC w<strong>as</strong> dried<br />
under vacuum <strong>for</strong> 4 hours. Next, DMC w<strong>as</strong> suspended in 125 ml NMP followed by the addition<br />
of 8 ml iodomethane. The dispersion w<strong>as</strong> stirred at 40°C <strong>for</strong> 50 hours and subsequently<br />
51
Chapter 3<br />
dropped into 400 ml of an ethanol/diethyl ether mixture (50/50) to precipitate the <strong>for</strong>med<br />
TMC, which w<strong>as</strong> collected by centrifugation and subsequently extensively w<strong>as</strong>hed with diethyl<br />
ether. After drying overnight at room temperature, the obtained TMC w<strong>as</strong> dissolved in 100 ml<br />
of an aqueous 10% NaCl solution <strong>for</strong> ion-exchange. Finally, the TMC w<strong>as</strong> dialyzed against<br />
deionized water <strong>for</strong> 3 days changing buffer twice daily, filtered through a 0.8 µm filter and<br />
collected after freeze-drying.<br />
Synthesis of O-methylated TMC from chitosan. O-methylated TMC with a DQ of about 45%<br />
w<strong>as</strong> synthesized by methylation of chitosan with iodomethane in a mixture of an aqueous<br />
solution of NaOH and NMP essentially <strong>as</strong> described previously [43]. In detail, chitosan (500<br />
mg) with a degree of acetylation of 17 or 55% and sodium iodide (1.2 g) were dispersed in a<br />
mixture of 40 ml of NMP and 5 ml of 15% w/v aqueous NaOH solution. Subsequently, the<br />
mixture w<strong>as</strong> heated to 60°C and after stirring <strong>for</strong> 20 min, 4 ml of methyl iodide w<strong>as</strong> added and<br />
the reaction mixture w<strong>as</strong> refluxed <strong>for</strong> 60 min. Then, 2 ml of 15% NaOH solution and 1.5 ml of<br />
iodomethane were added and the solution w<strong>as</strong> stirred <strong>for</strong> 60 minutes. Next, the reaction<br />
mixture w<strong>as</strong> dropped into 200 ml of a mixture of diethyl ether and ethanol (50/50) to<br />
precipitate the O-methylated TMC, which w<strong>as</strong> subsequently w<strong>as</strong>hed extensively with diethyl<br />
ether. Finally, the product w<strong>as</strong> dissolved in 50 ml aqueous 10% w/v NaCl solution, put on a<br />
shaker <strong>for</strong> 18 hours <strong>for</strong> ion-exchange and the obtained solution w<strong>as</strong> dialyzed at room<br />
temperature against deionized water <strong>for</strong> 3 days changing buffer twice daily, filtered through a<br />
0.8 µm filter and freeze dried.<br />
Determination of the degrees of acetylation, dimethylation, and quaternization. The<br />
1H-NMR spectra of the various chitosans, DMC and TMCs were recorded with a Varian INOVA<br />
500MHz NMR spectrometer (Varian Inc., Palo Alto, Ca, USA) at 80°C in D 2O. For the DMC and<br />
the chitosans addition of DCl (35% w/w in D 2O) w<strong>as</strong> needed to dissolve the polymers.<br />
The degree of acetylation of the chitosans, DMC and TMCs w<strong>as</strong> calculated <strong>as</strong> described<br />
previously [44]:<br />
DAc = [[CH 3]/[H2-H6] x 1/2] x 100<br />
Here, [CH 3] is the integral of the three hydrogens of the acetyl groups at 2.0 ppm and [H2-H6]<br />
is the integral corresponding the six H-2 to H-6 protons between 3.9 and 3.0 ppm.<br />
52
Influence of Degree of Acetylation on Degradation and Properties of TMC<br />
The degree of dimethylation of the DMC w<strong>as</strong> calculated <strong>as</strong> follows:<br />
DDM = [(CH 3) 2]/[H2-H6] x 100<br />
Here, [(CH 3) 2] is the integral of the hydrogens of the dimethyl amino groups at 2.9 ppm and<br />
[H2-H6] is the integral corresponding the H-2 to H-6 protons between 3.9 and 3.0 ppm.<br />
The DQ and degree of dimethylation (DM) of the TMCs were calculated <strong>as</strong> previously described<br />
[25, 43, 45].<br />
DQ = [[(CH 3) 3]/[H] × 1/9] × 100<br />
DM = [[(CH 3) 2]/[H] × 1/6] × 100<br />
Here, [(CH 3) 3] and [(CH 3) 2] are the integrals of the hydrogens of the trimethylated amino<br />
groups at 3.3 ppm and the dimethylated amino groups at 2.9 ppm, respectively. [H] is the<br />
integral of the H-1 peaks between 4.7 and 5.7 ppm; the signal related to the hydrogen atoms<br />
bound to the C-1’s of TMC.<br />
Enzymatic degradation of various TMCs. Table 1 gives an overview of the different<br />
chitosans and TMCs used <strong>for</strong> evaluation of the lysozyme-catalyzed degradation. The different<br />
polymers were placed overnight in a vacuum oven at 40 o C in presence of Sicapent to remove<br />
residual water. Next, the samples were dissolved in 2% acetic acid solution <strong>for</strong> at le<strong>as</strong>t 18<br />
hours. Then, lysozyme dissolved in H 2O, and 1 M NaCl and 1 M NaOH were added to the<br />
polymer solutions to obtain a lysozyme concentration of 38 µg/ml, polymer concentrations of<br />
5 mg/ml, 150 mM NaCl, 1% acetic acid and a pH of 4.5.<br />
The degradation of the different TMC polymers w<strong>as</strong> also studied at physiological pH in a<br />
phosphate-buffered salt (PBS) buffer (8.2 g/l NaCl, 3.1 g/l Na 2HPO 4 12 H 2O, 0.3 g/l NaH 2PO 4 2<br />
H 2O, pH 7.4). In detail, the different polymers were placed overnight in a vacuum oven at 40 o C<br />
in presence of Sicapent to remove residual water. Next, the samples were dissolved in PBS with<br />
0.02% sodium azide <strong>for</strong> 18 hours. Then, lysozyme dissolved in PBS (also containing 0.02%<br />
sodium azide) w<strong>as</strong> added to the polymer solutions to obtain a lysozyme concentration of 38<br />
µg/ml, polymer concentrations of 5 mg/ml and a pH of 7.4. Additionally, the influence of<br />
lysozyme concentration on polymer degradation w<strong>as</strong> studied. To this end, TMC-RA w<strong>as</strong><br />
dissolved in PBS with 0.02% sodium azide <strong>as</strong> described above and incubated with varying<br />
concentrations of lysozyme (9 µg/ml to 38 µg/ml) in PBS with 0.02% sodium azide.<br />
53
Chapter 3<br />
Polymer solutions were incubated at 37 °C and samples were taken at various time points<br />
during 6 days. Polymer molecular weights were determined with a GPC-triple detection<br />
described below. Polymers not exposed to lysozyme were used <strong>as</strong> controls.<br />
Determination of M n and M w of the different polymers. The number average weight (M n)<br />
and weight average weight (M w) of chitosan and the various TMCs were determined by gel<br />
permeation chromatography (GPC) on a Viscotek-triple detection system using a Shodex<br />
OHPak SB-806 column (15 cm) and 0.3 M sodium acetate pH 4.4 (adjusted with acetic acid) <strong>as</strong><br />
running buffer [46]. Data from the l<strong>as</strong>er photometer (λ = 670 nm) (right (90 0 ) and low (7 0 )<br />
angle light scattering), refractive index detector and viscosity detector were integrated using<br />
the provided Omnisec-software to calculate the M n, M w, dn/dc and the intrinsic viscosity ([η])<br />
of the different samples. Pullulan (M n = 102 kDa, M w = 106 kDa) obtained from Viscotek<br />
Benelux (Oss, the Netherlands) w<strong>as</strong> used <strong>for</strong> calibration.<br />
Transepithelial electrical resistance (TEER) me<strong>as</strong>urements. Caco-2 cells were seeded at<br />
a density of 2x10 5 cells per well on 12-transwell plates with a microporous membrane. The<br />
cells were cultured in Dulbecco’s Modified Eagle’s Medium (DMEM) <strong>for</strong> 10 days until a<br />
confluent cell layer w<strong>as</strong> <strong>for</strong>med. The medium w<strong>as</strong> replaced by Hank’s Balanced Salt Solution<br />
(HBSS) at the b<strong>as</strong>olateral side 10 minutes be<strong>for</strong>e the start of the experiments. Then, 0.5 ml<br />
solution of TMC (with various DQs, DAcs and with or without O-methylation, dissolved (2.5<br />
mg/ml) in HBSS, pH adjusted to 7 with 0.1 M NaOH) w<strong>as</strong> applied at the apical site of the cell<br />
monolayers. SDS (10 mg/ml) w<strong>as</strong> used <strong>as</strong> positive control and HBSS <strong>as</strong> reference. The<br />
resistance me<strong>as</strong>ured of the membrane without cells w<strong>as</strong> used <strong>as</strong> blank. The TEER of the Caco-2<br />
cells at certain time points (0, 15, 30, 45, 60 and 90 min.) after addition of the stimuli w<strong>as</strong><br />
me<strong>as</strong>ured with a Millicell-ERS (Millipore, Billerica, USA) me<strong>as</strong>uring device. After 90 minutes<br />
the cells were w<strong>as</strong>hed with HBSS and incubation of the cells w<strong>as</strong> continued in DMEM <strong>for</strong> 24<br />
hours at 37°C, CO 2 5% to determine the recovery of the TEER [25, 47].<br />
MTT cell toxicity <strong>as</strong>say. Caco-2 cells were seeded in a 96-well plate at a density of 4x10 4<br />
cells per well and incubated <strong>for</strong> 2 days at 37°C, CO 2 5% in culture medium (DMEM, high<br />
glucose, 10% FCS, L-glutamine, pyruvate, non essential amino acids). The medium w<strong>as</strong><br />
removed and the cells were incubated <strong>for</strong> 2.5 hours with 100 µl TMC solutions in HBSS (TMC<br />
concentrations were 0.1, 1 and 10 mg/ml, pH set at 7 with 0.1 M NaOH). SDS (10 mg/ml) w<strong>as</strong><br />
used <strong>as</strong> positive control and HBSS <strong>as</strong> reference <strong>for</strong> 100% cell viability. Thereafter, the HBSS<br />
54
Influence of Degree of Acetylation on Degradation and Properties of TMC<br />
w<strong>as</strong> removed and the cells were w<strong>as</strong>hed with phosphate buffered saline. One hundred µl of a<br />
freshly prepared solution of 0.5 mg/ml MTT in DMEM (without any additions), w<strong>as</strong> added and<br />
the cells were incubated <strong>for</strong> 3 hours at 37°C and 5% CO 2. Subsequently, the wells were<br />
emptied, 100 µl of DMSO w<strong>as</strong> used to dissolve the <strong>for</strong>med <strong>for</strong>mazan crystals and the<br />
absorbance w<strong>as</strong> read at 595 nm [48].<br />
LDH <strong>as</strong>say. Caco-2 cells were seeded in a 96-well plate at a density of 4x10 4 cells per well<br />
and incubated <strong>for</strong> 2 days at 37°C, CO 2 5% in culture medium (see section above <strong>for</strong><br />
composition). The medium w<strong>as</strong> removed and the cells were w<strong>as</strong>hed with HBSS and incubated<br />
<strong>for</strong> 2.5 hours with 100 µl TMC solutions in HBSS (TMC concentrations were 0.1, 1 and 10<br />
mg/ml, pH set at 7 with 0.1 M NaOH). After incubation, the concentration of LDH present in the<br />
supernatant of the samples w<strong>as</strong> determined with the Cytotoxicity Detection Kit-Plus (Roche<br />
Diagnostics, Mannheim, Germany) by me<strong>as</strong>uring absorbance at 490 nm with 650 nm <strong>as</strong> a<br />
reference wavelength. A calibration curve w<strong>as</strong> made with the lysis buffer provided by the<br />
manufacturer, setting the LDH concentration me<strong>as</strong>ured with the undiluted lysis buffer at 100%<br />
LDH rele<strong>as</strong>e. HBSS w<strong>as</strong> used <strong>as</strong> a negative control.<br />
Results and discussion<br />
Synthesis and characterization of different polymers. To investigate the influence of the<br />
degree of acetylation on the biological properties and enzymatic degradability of trimethylated<br />
chitosan (TMC), chitosan (CS, DAc 17%) w<strong>as</strong> partially re-acetylated using acetic anhydride to a<br />
DA of 55% (CS-RA, DAc 55%) <strong>as</strong> determined by 1 H-NMR analysis. Both CS and CS-RA were<br />
subsequently trimethylated to a degree of about 45% according to the method described by<br />
Sieval [43]. 1 H-NMR analysis (Table 1) showed that this synthesis method also leads to<br />
considerable O-methylation and loss of N-acetylated units, likely due to the alkaline reaction<br />
conditions.<br />
Besides the frequently used method of Sieval et al, CS and CS-RA were also trimethylated to a<br />
degree of 45% via the two-step synthesis route described recently by Verheul et al. [40] (with<br />
some modifications <strong>for</strong> the synthesis of TMC-RA) which avoids O-methylation and loss of N-<br />
acetylated units (Table 1). Chitosan with a DAc of 17% w<strong>as</strong> quantitatively dimethylated with<br />
<strong>for</strong>maldehyde using <strong>for</strong>mic acid <strong>as</strong> reducing agent. However, using these reaction conditions to<br />
dimethylate chitosan with a DAc of 55%, which is more susceptible to acidic hydrolysis than<br />
55
Chapter 3<br />
chitosan with a DAc of 17% [8], resulted in the reduction of molecular weights (data not<br />
shown). When sodium borohydride w<strong>as</strong> used <strong>as</strong> reducing agent, relatively mild reaction<br />
conditions (pH 4, room temperature) could be applied to quantitively dimethylate chitosan.<br />
Table 1. Characteristics of the chitosans and TMCs. Degree of acetylation (DAc), quaternization (DQ),<br />
dimethylation (DM), O-methylation on C-6 (DOM-6) and on C-3 (DOM-3) of TMCs <strong>as</strong> determined by 1 H-<br />
NMR analysis. M n, M w and dn/dc were determined by GPC-triple detection.<br />
DAc DQ DM DOM-6 DOM-3 Mn Mw dn/dc<br />
(kDa) (kDa)<br />
CS 17% - - - - 28 43 0.18<br />
TMC 17% 43% 40% - - 39 64 0.16<br />
TMC-OM 11% 45% 44% 25% 16% 35 48 0.16<br />
CS-RA 55% - - - - 35 65 0.16<br />
DMC-RA 55% - 45% - - 42 71 0.16<br />
TMC-RA 55% 44% 1% - - 43 79 0.16<br />
TMC-RA-OM 49% 46% 5% 77% 57% 40 63 0.15<br />
Table 1 shows that particularly in c<strong>as</strong>e of the re-acetylated chitosan, these milder reaction<br />
conditions prevented polymer chain scission; the molecular weights of CS, CS-RA, DMC-RA and<br />
TMC-RA slightly incre<strong>as</strong>ed with incre<strong>as</strong>ing extent of substitutions (TMC-RA>DMC-RA>CS-<br />
RA>CS). As reported previously, TMC synthesized according to the method of Sieval et al.<br />
resulted in polymers with a lower molecular weight than the polymers obtained via the<br />
method without the use of a strong b<strong>as</strong>e [40]. Likely, the strong b<strong>as</strong>ic reaction conditions<br />
caused some chain scission during synthesis [40, 49]. Importantly, the mild two-step synthetic<br />
method described in this paper yields TMC with varying DQs while O-methylation, polymer<br />
chain scission and the hydrolysis of N-acetylated groups do not occur. Previous studies showed<br />
that changes in TMC molecular weight [38, 39] and the presence of O-methyl groups [40]<br />
influence polymer characteristics, such <strong>as</strong> cytotoxicity and the ability to open tight junctions.<br />
GPC-TD analysis showed that re-acetylated chitosan and TMCs contained some high molecular<br />
weight material (about 3% of the injected amount) which can be likely <strong>as</strong>cribed to aggregation<br />
[50-52]. However, this high-molecular-weight material had no impact on the degradation<br />
studies due to its relatively small contribution and it w<strong>as</strong> e<strong>as</strong>ily separated on the GPC column<br />
from the free polymer.<br />
Enzymatic degradation of various chitosans and TMCs. Figure 1 shows the decre<strong>as</strong>e of<br />
the M n of TMC-RA in presence of various concentrations of hen egg-white lysozyme at pH 7.4 at<br />
37 °C. Clearly, the degradation rate incre<strong>as</strong>es with lysozyme concentration while without<br />
lysozyme almost no decre<strong>as</strong>e in molecular weight w<strong>as</strong> observed. The enzymatic degradation<br />
essentially occurred in the first 24 hours and after this time period hardly any decre<strong>as</strong>e of the<br />
56
Influence of Degree of Acetylation on Degradation and Properties of TMC<br />
M n of TMC-RA w<strong>as</strong> detected. Addition of fresh lysozyme after 70 hours did not result in a<br />
further drop of molecular weight (results not shown), implying that arrest of the polymer<br />
degradation is not due to inactivation of lysozyme but that lysozyme is not able to catalyze the<br />
hydrolysis of the remaining glyosidic bonds. As expected, the molecular weight obtained after<br />
prolonged exposure to lysozyme w<strong>as</strong> independent of the concentration of the enzyme. Overall,<br />
in the presence of lysozyme, the M n of TMC-RA dropped from 43 kDa to about 11 kDa<br />
corresponding on average to approximately three cuts per molecule. Figure 2 shows the<br />
changes in M n, M w and intrinsic viscosity ([η]) of TMC-RA in presence of lysozyme (38 µg/ml)<br />
at pH 7.4 at 37 °C <strong>as</strong> obtained with the GPC-triple detection method. As expected, both the M w<br />
and the [η] followed a similar pattern <strong>as</strong> the M n of TMC-RA. This w<strong>as</strong> observed <strong>for</strong> all analyzed<br />
polymers (data not shown).<br />
4.0×10 4 TMC-RA 38 μg/ml lyso<br />
Mn (Da)<br />
3.0×10 4<br />
2.0×10 4<br />
TMC-RA 0μg/ml lyso<br />
TMC-RA 9 μg/ml lyso<br />
TMC-RA 19 μg/ml lyso<br />
1.0×10 4<br />
0<br />
0 5 10<br />
100 200<br />
time (h)<br />
Figure 1. Degradation of TMC-RA by lysozyme (various concentrations) at pH 7.4 and 37°C. Error bars<br />
represent the standard deviation of three me<strong>as</strong>urements.<br />
The lysozyme catalyzed enzymatic degradation of different chitosans and TMCs w<strong>as</strong><br />
compared <strong>as</strong> shown in Figure 3. Since CS, CS-RA and DMC-RA are not soluble at physiological<br />
pH, degradation w<strong>as</strong> studied at pH 4.5 and at 37 °C in presence of 38 µg/ml lysozyme. This<br />
figure clearly demonstrates that the extent of lysozyme-catalyzed degradation of the<br />
investigated polymers is highly dependent on the degree of acetylation. Polymers with a DAc ≤<br />
17% (CS, TMC and TMC-OM) are less susceptible to lysozyme-catalyzed degradation than the<br />
re-acetylated polymers with a DAc ≥49%. Although the changes in molecular weights are less<br />
57
Chapter 3<br />
prominent, it is obvious that TMC-OM, with the lowest DAc (DAc 11%), w<strong>as</strong> also to the lowest<br />
extent degraded by lysozyme after 6 days (decre<strong>as</strong>e in M n from 35 kDa to 30 kDa). In contr<strong>as</strong>t,<br />
the M n of TMC-RA with a DAc of 55% dropped from 43 kDa to 10 kDa under the same<br />
conditions. Re-acetylated CS-RA (DAc 55%), in agreement with others [6, 7, 10, 14, 17, 53], w<strong>as</strong><br />
much more susceptible to lysozyme-catalyzed degradation than CS (DAc 17%), and their M n’s<br />
decre<strong>as</strong>ed from 35 kDa to 7 kDa and from 28 kDa to 20 kDa, respectively.<br />
0.4<br />
Mn or Mw (Da)<br />
8.0×10 4 100 200<br />
6.0×10 4<br />
4.0×10 4<br />
2.0×10 4<br />
[η]<br />
Mn<br />
Mw<br />
0.3<br />
0.2<br />
0.1<br />
[η] (dl/g)<br />
0<br />
0 5 10<br />
time (h)<br />
0.0<br />
Figure 2. Changes in the M n, M w and intrinsic viscosity of TMC-RA by lysozyme (38 µg/ml) at pH 7.4 and<br />
37°C. Error bars represent the standard deviation of three me<strong>as</strong>urements.<br />
Nordtveit et al. demonstrated that at le<strong>as</strong>t 3 to 4 acetylated units are required in the<br />
hexameric binding pocket <strong>for</strong> lysozyme to allow enzymatic cleavage [7]. Since these domains<br />
containing the required number of acetylated units are destroyed through cleavage of the<br />
glycosidic bond [7, 16], degradation by lysozyme occurs till a minimum molecular weight<br />
depending on the DAc of the chitosan [11]. As DMC-RA w<strong>as</strong> degraded to practically the same<br />
molecular weight <strong>as</strong> CS-RA (Figure 3), it can be concluded that methylation of the free NH 2<br />
groups of chitosan did not affect its extent of lysozyme-catalyzed degradation. TMC-RA w<strong>as</strong><br />
cleaved up to a slightly higher M n than CS-RA (10 kDa vs. 7 kDa, respectively), however, this<br />
can be explained by its higher degree of substitution due to the quaternization of the free<br />
amines thereby incre<strong>as</strong>ing the weight per glucosamine-residue. There<strong>for</strong>e, it can be stated that<br />
quaternized chitosan synthesized according to the two-step method presented here is<br />
58
Influence of Degree of Acetylation on Degradation and Properties of TMC<br />
degraded by lysozyme to the same extent <strong>as</strong> the original chitosan. In contr<strong>as</strong>t, chitosan<br />
quaternized using iodomethane and a strong b<strong>as</strong>e (TMC-RA-OM) w<strong>as</strong> cleaved less than the<br />
original chitosan (CS-RA) or the O-methyl free TMC-RA, likely due to loss of N-acetylated units<br />
during synthesis. To explain, the M n of TMC-RA-OM with a DAc of 49% decre<strong>as</strong>ed from 40 kDa<br />
to 18 kDa. Although TMC-RA-OM h<strong>as</strong> a modestly higher weight per glucosamine unit than CS-<br />
RA, the difference in final M n of 7 kDa of CS-RA and 18 kDa of TMC-RA-OM h<strong>as</strong> to be attributed<br />
to less chain scission by lysozyme <strong>for</strong> TMC-RA-OM due to its lower DAc <strong>as</strong> compared to the<br />
other re-acetylated polymers. Interestingly, TMC-RA and, especially DMC-RA, were f<strong>as</strong>ter<br />
cleaved by lysozyme than CS-RA implying that enzyme-affinity and/or maximum substrate<br />
conversion rates may be enhanced by the methylation of the free amines.<br />
Mn (Da)<br />
4.0×10 4 CS<br />
CS-RA<br />
DMC-RA<br />
3.0×10 4<br />
TMC-RA<br />
TMC-RA-OM<br />
TMC<br />
2.0×10 4<br />
TMC-OM<br />
1.0×10 4<br />
0<br />
0 5 10<br />
100 200<br />
time (h)<br />
Figure 3. Degradation of various chitosans and TMCs by lysozyme (38 µg/ml) at pH 4.5 and 37°C. Error<br />
bars represent the standard deviation of three me<strong>as</strong>urements.<br />
To investigate the lysozyme-catalyzed degradation of quaternized chitosans under<br />
physiological conditions, changes in molecular weight of TMCs in the presence of lysozyme<br />
were studied in PBS at pH 7.4. The extent of lysozyme-catalyzed degradation of quaternized<br />
chitosans w<strong>as</strong> comparable to the degradation observed at pH 4.5 (Figure 4). In detail, at pH 7.4<br />
59
Chapter 3<br />
the polymer with the lowest DAc of 11% (TMC-OM) showed the smallest decre<strong>as</strong>e in molecular<br />
weight (M n from 34 to 29 kDa) while the TMC with a DAc of 55% (TMC-RA) w<strong>as</strong> readily<br />
degraded (M n from 43 to 11 kDa). At pH 4.5 the lysozyme-catalyzed degradation of the reacetylated<br />
TMC-RA and TMC-RA-OM w<strong>as</strong> slightly f<strong>as</strong>ter than at pH 7.4 likely due to higher<br />
enzyme activity at lower pH [9]. Overall, our data show that the degradation of quaternized<br />
chitosans by lysozyme both in terms of kinetics and in final molecular weight of the<br />
degradation products is hardly affected by the pH in the range studied (pH 4.5-7.4). This w<strong>as</strong><br />
not observed <strong>for</strong> the degradation of chitosan, where the degradation rate at pH 7.4 is<br />
substantially slower than that in acidic environment [14]. At a pH above 6.5 chitosans are<br />
insoluble and degradation by enzymes decre<strong>as</strong>es due to limited accessibility of the binding<br />
sites <strong>for</strong> lysozyme. Since TMCs are soluble at physiological pH, solubility does not limit the<br />
lysozyme-catalyzed degradation.<br />
4.0×10 4 TMC<br />
Mn (Da)<br />
3.0×10 4<br />
2.0×10 4<br />
TMC-RA<br />
TMC-OM<br />
TMC-RA-OM<br />
1.0×10 4<br />
0<br />
0 5 10<br />
100 200<br />
time (h)<br />
Figure 4. Degradation of various TMCs by lysozyme (38 µg/ml) at pH 7.4 and 37°C. Error bars represent<br />
the standard deviation of three me<strong>as</strong>urements.<br />
The degradability of TMCs will have important consequences <strong>for</strong> its pharmaceutical<br />
applications. Onishi and co-workers demonstrated in mice that, after intraperitoneally<br />
injection, chitosan with a DAc of 50% is readily degraded to a molecular weight of less than 10<br />
60
Influence of Degree of Acetylation on Degradation and Properties of TMC<br />
kDa and after 24 hours the full dose w<strong>as</strong> excreted by the kidneys [54]. Since glomerular<br />
filtration by kidneys is size-dependent, our results imply that the degradation products of the<br />
re-acetylated polymers, due to their lower final molecular weight, will be more e<strong>as</strong>ily removed<br />
from the body than the chitosans with a low DAc. Also, TMC-carrier systems may be tailored<br />
<strong>for</strong> its degradation characteristics; ideally, the carrier molecule will degrade after delivery at<br />
the target site thereby rele<strong>as</strong>ing its pharmaceutically active compounds.<br />
In conclusion, our data demonstrate that the lysozyme-catalyzed degradation of TMCs is, like<br />
that of chitosan, facilitated <strong>for</strong> polymers with a high DAc. However, the degradation of TMC is<br />
similar at pH 4.5 and 7.4.<br />
Evaluation of various TMCs on TEER, MTT and LDH <strong>as</strong>says. TMC is capable to open tight<br />
junctions of epithelial cells and the highest activity w<strong>as</strong> observed with DQs of around 40-50%<br />
[29, 35, 55]. Recently, it w<strong>as</strong> shown that O-methylation substantially reduces the activity of<br />
TMC on transepithelial electrical resistance (TEER) [40]. In this study, TMCs were tested with<br />
only minor variations in DQ (see Table 1), so the charge ratio and the cationic character of<br />
these polymers were considered similar. However, the polymers differed in DAc and in extent<br />
of O-methylation.<br />
TEER (% HBSS)<br />
120<br />
100<br />
80<br />
60<br />
40<br />
20<br />
SDS 1%<br />
TMC<br />
TMC-OM<br />
TMC-RA<br />
TMC-RA-OM<br />
0<br />
0 15 30 45 60 75 90<br />
time (min.)<br />
1140<br />
Figure 5. Effect of TMCs with various DAcs and with or without O-methylated units on the TEER of<br />
Caco-2 cells at a TMC concentration of 2.5 mg/ml. SDS 10 mg/ml w<strong>as</strong> used <strong>as</strong> a positive control. Error<br />
bars represent the standard deviation of six me<strong>as</strong>urements.<br />
Figure 5 shows the effect of various TMCs on the TEER of a Caco-2 monolayer. In line with<br />
previous data [40], O-methyl free TMC with a DQ of 43% and a DAc of 17% demonstrated the<br />
61
Chapter 3<br />
largest decre<strong>as</strong>e in TEER (about 70% reduction), while the partially O-methylated TMC with a<br />
DQ of 42% and a DAc of 11% reduced the TEER with about 30% compared to the TEER<br />
observed with HBSS. Importantly, the Caco-2 monolayer partially recovered after removal of<br />
these polymers implying that no permanent toxicity w<strong>as</strong> induced on the cells. F<strong>as</strong>cinatingly,<br />
TMC polymers with a similar DQ but with higher DAcs had no effect on the TEER, indicating<br />
that these polymers, although positively charged, were unable to open tight junctions. Also, the<br />
results of the MTT and LDH <strong>as</strong>says (Figures 6 and 7) obtained with TMC and TMC-OM were <strong>as</strong><br />
expected [40], showing a higher toxicity <strong>for</strong> O-methyl free TMC. Additionally, even at a<br />
concentration of 10 mg/ml both TMC-RA and TMC-RA-OM clearly show neither a decre<strong>as</strong>e in<br />
cell viability nor LDH rele<strong>as</strong>e. Studies with chitosan have shown that an incre<strong>as</strong>e in DAc lead to<br />
lower penetration through a Caco-2 monolayer [20], a decre<strong>as</strong>e in cytotoxicity [18, 20] and a<br />
reduction in DNA transfection [19]. However, these experiments were carried out in slightly<br />
acidic environment to solubilize the chitosan polymer. In acidic environment the primary<br />
amines will be (partially) protonated and thus become positively charged. A higher DAc will<br />
reduce the number of amines available <strong>for</strong> protonation thereby decre<strong>as</strong>ing the charge density<br />
of chitosan. There<strong>for</strong>e, with chitosan, it w<strong>as</strong> not possible to study the effect of DAc on polymer<br />
characteristics without altering the charge density of the polymer.<br />
cell viability (%)<br />
125<br />
100<br />
75<br />
50<br />
25<br />
TMC<br />
TMC-OM<br />
TMC-RA<br />
TMC-RA-OM<br />
0<br />
0.1 1 10<br />
conc. (mg/ml)<br />
Figure 6. Effect of TMCs with various DAcs and with or without O-methylated units on the viability of<br />
Caco-2 cells (MTT <strong>as</strong>say) at different concentrations. Error bars represent the standard deviation of six<br />
me<strong>as</strong>urements.<br />
62
Influence of Degree of Acetylation on Degradation and Properties of TMC<br />
LDH rele<strong>as</strong>e (%)<br />
70<br />
60<br />
50<br />
40<br />
30<br />
20<br />
10<br />
HBSS<br />
TMC-RA-OM<br />
TMC-RA<br />
TMC-OM<br />
TMC<br />
0<br />
0.1 1 10<br />
conc. (mg/ml)<br />
Figure 7. Effect of TMCs with various DAcs and with or without O-methylated units on the viability of<br />
Caco-2 cells (LDH rele<strong>as</strong>e) at different concentrations. Error bars represent the standard deviation of six<br />
me<strong>as</strong>urements.<br />
Our studies were done at physiological pH and with TMCs with a similar DQ (and thus charge<br />
density) allowing evaluation of the influence of the DAc without changing other polymer<br />
characteristics. Consequently the remarkable decre<strong>as</strong>e in toxicity and ability to open tight<br />
junctions when incre<strong>as</strong>ing the DAc h<strong>as</strong> to be contributed to intra- and/or intermolecular<br />
changes induced by the N-acetylated units. Possibly, the N-acetylated glucosamine parts of the<br />
TMC can <strong>for</strong>m hydrophobic domains, thereby shielding the positively charged quaternized<br />
parts of the macromolecule. However, further studies are needed to provide more insight into<br />
the macromolecular changes that occur when incre<strong>as</strong>ing the DAc.<br />
Conclusion<br />
This paper presents a method to synthesize TMC with a high DAc without introducing other<br />
alterations of the polymer such <strong>as</strong> O-methylation, chain scission or loss of N-acetylation. The<br />
enzymatic degradation of TMC by lysozyme w<strong>as</strong> essentially identical to the degradation of<br />
chitosan (thus highly dependent on the degree of acetylation) however, in contr<strong>as</strong>t to chitosan,<br />
this degradation w<strong>as</strong> pH independent. TMCs with a DAc of ~50%, a DQ of around 44% and<br />
with or without O-methylation, were not able to open tight junctions of Caco-2 cells, in contr<strong>as</strong>t<br />
to TMCs (O-methylated or O-methyl free) with a similar DQ but a lower DAc. Additionally,<br />
TMCs with a high DAc (~50%) showed no cell toxicity up to a concentration of 10 mg/ml. In<br />
conclusion, the degree of N-acetylation h<strong>as</strong> great impact on the biological characteristics of<br />
63
Chapter 3<br />
TMC and future research will be done to determine the optimal polymer structure <strong>for</strong> the<br />
various potential pharmaceutical applications of TMC.<br />
Acknowledgement. This research w<strong>as</strong> per<strong>for</strong>med under the framework of <strong>TI</strong> <strong>Pharma</strong><br />
project number D5-106-1; Vaccine delivery: alternatives <strong>for</strong> conventional multiple injection<br />
vaccines. The authors thank Sara Riera Riv<strong>as</strong> <strong>for</strong> her contributions to the project.<br />
64
Influence of Degree of Acetylation on Degradation and Properties of TMC<br />
References<br />
1. Mourya VK, Inamdar NN. Chitosan-modifications and applications: Opportunities galore. React<br />
Funct Polym 68:1013-51 (2008).<br />
2. Agnihotri SA, Mallikarjuna NN, Aminabhavi TM. Recent advances on chitosan-b<strong>as</strong>ed microand<br />
nanoparticles in drug delivery. J Control Rele<strong>as</strong>e 100:5-28 (2004).<br />
3. Kumar MNVR, Muzzarelli RAA, Muzzareilli C, S<strong>as</strong>hiwa H, Domb AJ. Chitosan chemistry and<br />
pharmaceutical perspectives. Chem Rev 104:6017-84 (2004).<br />
4. Van Der Lubben IM, Kersten G, Fretz MM, Beuvery C, Verhoef JC, Junginger HE. Chitosan<br />
microparticles <strong>for</strong> mucosal vaccination against diphtheria: Oral and n<strong>as</strong>al efficacy studies in<br />
mice. Vaccine 21:1400-8 (2003).<br />
5. Muzzarelli RAA. Human enzymatic activities related to the therapeutic administration of chitin<br />
derivatives. Cel Mol Life Sci 53:131-40 (1997).<br />
6. Vårum KM, Myhr MM, Hjerde RJN, Smidsrød O. In vitro degradation rates of partially N-<br />
acetylated chitosans in human serum. Carbohydr Res 299:99-101 (1997).<br />
7. Nordtveit RJ, Varum KM, Smidsrod O. Degradation of fully water-soluble, partially N-<br />
acetylated chitosans with lysozyme. Carbohydr Polym 23:253-60 (1994).<br />
8. Varum KM, Ottoy MH, Smidsrod O. Acid hydrolysis of chitosans. Carbohydr Polym 46:89-98<br />
(2001).<br />
9. Freier T, Koh HS, Kazazian K, Shoichet MS. Controlling cell adhesion and degradation of<br />
chitosan films by N-acetylation. Biomaterials 26:5872-78 (2005).<br />
10. Nordtveit RJ, Varum KM, Smidsrod O. Degradation of partially N-acetylated chitosans with<br />
hen egg white and human lysozyme. Carbohydr Polym 29:163-7 (1996).<br />
11. Ren D, Yi H, Wang W, Ma X. The enzymatic degradation and swelling properties of chitosan<br />
matrices with different degrees of N-acetylation. Carbohydr Res 340:2403-10 (2005).<br />
12. S<strong>as</strong>hiwa H, Saimoto H, Shigem<strong>as</strong>a Y, Ogawa R, Tokura S. Lysozyme susceptibility of partially<br />
deacetylated chitin. Int J Biol Macromol 12:295-6 (1990).<br />
13. Tomihata K, Ikada Y. In vitro and in vivo degradation of films of chitin and its deacetylated<br />
derivatives. Biomaterials 18:567-75 (1997).<br />
14. Shigem<strong>as</strong>a Y, Saito K, S<strong>as</strong>hiwa H, Saimoto H. Enzymatic degradation of chitins and partially<br />
deacetylated chitins. Int J Biol Macromol 16:43-9 (1994).<br />
15. Lewis CE, McCarthy SP, Lorenzen J, McGee JOD. Differential effects of LPS, IFN-γ and<br />
TNFα on the secretion of lysozyme by individual human mononuclear phagocytes: relationship<br />
to cell maturity. Immunology 9:402-8 (1990).<br />
16. Pangburn SH, Trescony PV, Heller J. Lysozyme degradation of partially deacetylated chitin, its<br />
films and hydrogels. Biomaterials 3:105-8 (1982).<br />
17. Yang Y, Hu W, Wang X, Gu X. The controlling biodegradation of chitosan fibers by N-<br />
acetylation in vitro and in vivo. J Mater Sci: Mater Med 18:2117-21 (2007).<br />
18. Huang M, Khor E, Lim LY. Uptake and cytotoxicity of chitosan molecules and nanoparticles:<br />
effects of molecular weight and degree of deacetylation. Pharm Res 21:344-53 (2004).<br />
19. Kiang T, Wen J, Lim HW, Leong KWKW. The effect of the degree of chitosan deacetylation<br />
on the efficiency of gene transfection. Biomaterials 25:5293-301 (2004).<br />
20. Schipper NGM, Vårum KM, Artursson P. <strong>Chitosans</strong> <strong>as</strong> absorption enhancers <strong>for</strong> poorly<br />
absorbable drugs. 1: Influence of molecular weight and degree of acetylation on drug transport<br />
across human intestinal epithelial (Caco-2) cells. Pharm Res 13:1686-92 (1996).<br />
21. Nishimura K, Ishihara C, Ukei S, Tokura S, Azuma I. Stimulation of cytokine production in<br />
mice using deacetylated chitin. Vaccine 4:151-56 (1986).<br />
22. Nishimura K, Nishimura S, Nishi N. Immunological activity of chitin and its derivatives. Vaccine<br />
2:93-98 (1984).<br />
23. Nishimura K, Nishimura S-i, Nishi N, Numata F, Tone Y, Tokura S et al. <strong>Adjuvant</strong> activity of<br />
chitin derivatives in mice and guinea-pigs. Vaccine 3:379-84 (1985).<br />
65
Chapter 3<br />
24. Robinson MJ, Sancho D, Slack EC, LeibundGut-Landmann S, Reis e Sousa C. Myeloid C-type<br />
lectins in innate immunity. Nat Immunol 7:1258-65 (2006).<br />
25. Amidi M, Romeijn SG, Borchard G, Junginger HE, Hennink WE, Jiskoot W. Preparation and<br />
characterization of protein-loaded N-trimethyl chitosan nanoparticles <strong>as</strong> n<strong>as</strong>al delivery system. J<br />
Control Rele<strong>as</strong>e 111:107-16 (2006).<br />
26. Boonyo W, Junginger HE, Waranuch N, Polnok A, Pitaksuteepong T. Chitosan and trimethyl<br />
chitosan chloride (TMC) <strong>as</strong> adjuvants <strong>for</strong> inducing immune responses to ovalbumin in mice<br />
following n<strong>as</strong>al administration. J Control Rele<strong>as</strong>e 121:168-75 (2007).<br />
27. Chen F, Zhang Z-R, Yuan F, Qin X, Wang M, Huang Y. In vitro and in vivo study of N-<br />
trimethyl chitosan nanoparticles <strong>for</strong> oral protein delivery. Int J Pharm 349:226-33 (2008).<br />
28. Kean T, Roth S, Thanou M. <strong>Trimethyl</strong>ated chitosans <strong>as</strong> non-viral gene delivery vectors:<br />
Cytotoxicity and transfection efficiency. J Control Rele<strong>as</strong>e 103:643-53 (2005).<br />
29. Van der Lubben IM, Verhoef JC, Fretz MM, Van O, Mesu I, Kersten G, Junginger HE.<br />
<strong>Trimethyl</strong> chitosan chloride (TMC) <strong>as</strong> a novel excipient <strong>for</strong> oral and n<strong>as</strong>al immunisation against<br />
diphtheria. STP Pharm Sci 12:235-42 (2002).<br />
30. Sayin B, Somavarapu S, Li XW, Thanou M, Sesardic D, Alpar HO et al. Mono-Ncarboxymethyl<br />
chitosan (MCC) and N-trimethyl chitosan (TMC) nanoparticles <strong>for</strong> non-inv<strong>as</strong>ive<br />
vaccine delivery. Int J Pharm 363:139-48 (2008).<br />
31. Thanou M, Florea BI, Geldof M, Junginger HE, Borchard G. Quaternized chitosan oligomers<br />
<strong>as</strong> novel gene delivery vectors in epithelial cell lines. Biomaterials 23:153-9 (2002).<br />
32. Chen F, Zhang ZR, Huang Y. Evaluation and modification of N-trimethyl chitosan chloride<br />
nanoparticles <strong>as</strong> protein carriers. Int J Pharm 336:166-73 (2007).<br />
33. Di Colo G, Burgal<strong>as</strong>si S, Zambito Y, Monti D, Chetoni P. Effects of different N-trimethyl<br />
chitosans on in vitro/in vivo ofloxacin transcorneal permeation. J Pharm Sci 93:2851-62 (2004).<br />
34. Snyman D, Hamman JH, Kotze AF. Evaluation of the mucoadhesive properties of N-trimethyl<br />
chitosan chloride. Drug Develop Ind Pharm 29:61-9 (2003).<br />
35. Thanou MM, Kotze AF, Scharringhausen T, Lueßen HL, De Boer AG, Verhoef JC et al. Effect<br />
of degree of quaternization of N-trimethyl chitosan chloride <strong>for</strong> enhanced transport of<br />
hydrophilic compounds across intestinal Caco-2 cell monolayers. J Control Rele<strong>as</strong>e 64:15-25<br />
(2000).<br />
36. Hamman JH, Stander M, Kotze AF. Effect of the degree of quaternisation of N-trimethyl<br />
chitosan chloride on absorption enhancement: In vivo evaluation in rat n<strong>as</strong>al epithelia. Int J<br />
Pharm 232:235-42 (2002).<br />
37. Sandri G, Bonferoni MC, Rossi S, Ferrari F, Gibin S, Zambito Y et al. Nanoparticles b<strong>as</strong>ed on<br />
N-trimethylchitosan: Evaluation of absorption properties using in vitro (Caco-2 cells) and ex<br />
vivo (excised rat jejunum) models. Eur J Pharm and Biopharm 65:68-77 (2007).<br />
38. Mao S, Shuai X, Unger F, Wittmar M, Xie X, Kissel T. Synthesis, characterization and<br />
cytotoxicity of poly(ethylene glycol)-graft-trimethyl chitosan block copolymers. Biomaterials<br />
26:6343-56 (2005).<br />
39. Guo Z, Xing R, Liu S, Zhong Z, Ji X, Wang L et al. The influence of molecular weight of<br />
quaternized chitosan on antifungal activity. Carbohydr Polym 71:694-7 (2008).<br />
40. Verheul RJ, Amidi M, van der Wal S, van Riet E, Jiskoot W, Hennink WE. Synthesis,<br />
characterization and in vitro biological properties of O-methyl free N,N,N-trimethylated<br />
chitosan. Biomaterials 29:3642-9 (2008).<br />
41. Guo Z, Xing R, Liu S, Zhong Z, Ji X, Wang L et al. Antifungal properties of Schiff b<strong>as</strong>es of<br />
chitosan, N-substituted chitosan and quaternized chitosan. Carbohydr Res 2007;342:1329-32.<br />
42. Jia Z, shen D, Xu W. Synthesis and antibacterial activities of quaternary ammonium salt of<br />
chitosan. Carbohydr Res 333:1-6 (2001).<br />
43. Sieval AB, Thanou M, Kotze AF, Verhoef JC, Brussee J, Junginger HE. Preparation and NMR<br />
characterization of highly substituted N-trimethyl chitosan chloride. Carbohydr Polym 36:157-65<br />
(1998).<br />
66
Influence of Degree of Acetylation on Degradation and Properties of TMC<br />
44. Lavertu M, Xia Z, Serreqi AN, Berrada M, Rodrigues A, Wang D et al. A validated 1H NMR<br />
method <strong>for</strong> the determination of the degree of deacetylation of chitosan. J Pharm Biomed Anal<br />
32:1149-58 (2003).<br />
45. Polnok A, Borchard G, Verhoef JC, Sarisuta N, Junginger HE. Influence of methylation<br />
process on the degree of quaternization of N-trimethyl chitosan chloride. Eur J Pharm Biopharm<br />
57:77-83 (2004).<br />
46. Jiang X, van der Horst A, van Steenbergen M, Akeroyd N, van Nostrum C, Schoenmakers P et<br />
al. Molar-m<strong>as</strong>s characterization of cationic polymers <strong>for</strong> gene delivery by aqueous size-exclusion<br />
chromatography. Pharm Res 23:595-603 (2006).<br />
47. Nerurkar MM, Burton PS, Borchardt RT. The use of surfactants to enhance the permeability of<br />
peptides through Caco-2 cells by inhibition of an apically polarized efflux system. Pharm Res<br />
13:528-34 (1996).<br />
48. Mosmann T. Rapid colorimetric <strong>as</strong>say <strong>for</strong> cellular growth and survival: Application to<br />
proliferation and cytotoxicity <strong>as</strong>says. J Immunol Methods 65:55-65 (1983).<br />
49. Snyman D, Hamman JH, Kotze JS, Rollings JE, Kotze AF. The relationship between the<br />
absolute molecular weight and the degree of quaternisation of N-trimethyl chitosan chloride.<br />
Carbohydr Polym 50:145-50 (2002).<br />
50. Hu Y, Du Y, Yang J, Tang Y, Li J, Wang X. Self-aggregation and antibacterial activity of N-<br />
acylated chitosan. Polymer 48:3098-106 (2007).<br />
51. Schatz C, Viton C, Delair T, Pichot C, Domard A. Typical physicochemical behaviors of<br />
chitosan in aqueous solution. Biomacromolecules 4:641-8 (2003).<br />
52. Yanagisawa M, Kato Y, Yoshida Y, Isogai A. SEC-MALS study on aggregates of chitosan<br />
molecules in aqueous solvents: Influence of residual N-acetyl groups. Carbohydr Polym 66:192-98<br />
(2006).<br />
53. Zhang H, Neau SH. In vitro degradation of chitosan by bacterial enzymes from rat cecal and<br />
colonic contents. Biomaterials 23:2761-66 (2002).<br />
54. Onishi H, Machida Y. Biodegradation and distribution of water-soluble chitosan in mice.<br />
Biomaterials 20:175-182 (1999).<br />
55. Kotze AF, Luessen HL, de Leeuw BJ, de Boer AG, Verhoef JC, Junginger HE. N-<strong>Trimethyl</strong><br />
chitosan chloride <strong>as</strong> a potential absorption enhancer across mucosal surfaces: in vitro evalutation<br />
in intestinal epithelial cells (Caco-2). Pharm Res 14:1197-202 (1997).<br />
67
CHAPTER 4<br />
RELA<strong>TI</strong>ONSHIP BETWEEN STRUCTURE AND<br />
ADJUVAN<strong>TI</strong>CITY OF N,N,N-TRIMETHYL<br />
CHITOSAN (TMC) STRUCTURAL VARIANTS<br />
IN A NASAL INFLUENZA VACCINE<br />
Rolf J. Verheul*, Niels Hagenaars*, Imke Mooren, P<strong>as</strong>cal H.J.L.F. de Jong,<br />
Enrico M<strong>as</strong>trobattista, Harrie L. Glansbeek, Jacco G.M. Heldens<br />
Han van den Bosch, Wim E. Hennink, Wim Jiskoot.<br />
*authors contributed equally<br />
Journal of Controlled Rele<strong>as</strong>e 2009, 140, 126-133
Chapter 4<br />
Abstract<br />
The aim of this study w<strong>as</strong> to <strong>as</strong>sess the influence of structural properties of N,N,N-trimethyl<br />
chitosan (TMC) on its adjuvanticity. There<strong>for</strong>e, TMCs with varying degrees of quaternization<br />
(DQ, 22-86%), O-methylation (DOM, 0-76%) and acetylation (DAc 9-54%) were <strong>for</strong>mulated<br />
with whole inactivated influenza virus (WIV). The <strong>for</strong>mulations were characterized<br />
physicochemically and evaluated <strong>for</strong> their immunogenicity in an intran<strong>as</strong>al (i.n.)<br />
vaccination/challenge study in mice.<br />
Simple mixing of the TMCs with WIV at a 1:1 (w/w) ratio resulted in comparable positively<br />
charged nanoparticles, indicating coating of WIV with TMC. The amount of free TMC in solution<br />
w<strong>as</strong> comparable <strong>for</strong> all TMC-WIV <strong>for</strong>mulations. After i.n. immunization of mice with WIV and<br />
TMC-WIV on day 0 and 21, all TMC-WIV <strong>for</strong>mulations induced stronger total IgG, IgG1 and<br />
IgG2a/c responses than WIV alone, except WIV <strong>for</strong>mulated with reacetylated TMC with a DAc<br />
of 54% and a DQ of 44% (TMC-RA44). No significant differences in antibody titers were<br />
observed <strong>for</strong> TMCs that varied in DQ or DOM, indicating that these structural characteristics<br />
play a minor role in their adjuvant properties. TMC with a DQ of 56% (TMC56) <strong>for</strong>mulated<br />
with WIV at a ratio of 5:1 (w/w) resulted in significantly lower IgG2a/c:IgG1 ratio’s compared<br />
to TMC56 mixed in ratios of 0.2:1 and 1:1, implying a shift towards a Th2 type immune<br />
response. Challenge of vaccinated mice with aerosolized virus demonstrated protection <strong>for</strong> all<br />
TMC-WIV <strong>for</strong>mulations with the exception of TMC-RA44-WIV.<br />
In conclusion, <strong>for</strong>mulating WIV with TMCs strongly enhances the immunogenicity and<br />
induced protection after i.n. vaccination with WIV. The adjuvant properties of TMCs <strong>as</strong> i.n.<br />
adjuvant are strongly decre<strong>as</strong>ed by reacetylation of TMC, where<strong>as</strong> the DQ and DOM hardly<br />
affect the adjuvanticity of TMC.<br />
70
Relationship between Structure and <strong>Adjuvant</strong>icity of TMC<br />
Introduction<br />
Intran<strong>as</strong>al (i.n.) vaccination offers several advantages over the intramuscular (i.m.) route,<br />
like simple, needle-free administration without the need <strong>for</strong> trained personnel, potentially less<br />
adverse effects and the induction of local mucosal immune responses [1]. On the other hand,<br />
vaccines administered via the i.n. route generally induce low systemic immune responses<br />
when compared to i.m. administration likely due to mucociliary clearance and low antigen<br />
uptake. Mucoadhesive polymers have been used to incre<strong>as</strong>e the immunogenicity of i.n.<br />
vaccines by incre<strong>as</strong>ing n<strong>as</strong>al residence time and enhancing antigen presentation [1].<br />
Chitosan, a polysaccharide that is obtained by deacetylation of the natural polymer chitin,<br />
h<strong>as</strong> mucoadhesive properties and showed promising results <strong>as</strong> an adjuvant in n<strong>as</strong>al vaccines<br />
[2-5]. However, the unfavorable pH-dependent solubility and charge density led to the<br />
synthesis of its quaternized derivative N,N,N-trimethyl chitosan (TMC) (Scheme 1), which is<br />
well soluble in aqueous solution at neutral pH.<br />
Scheme 1. General structure of TMC. Depending on the synthesis route TMCs can be varied in degree of<br />
acetylation (see block ‘x’), quaternization (see block ‘y’), and O-methylation (see block ‘z’). The various<br />
substitutions are randomly distributed throughout the polymer; O-methylation (block ‘z’) may also<br />
occur on the quaternized and acetylated units (blocks ‘x’ and ‘y’).<br />
TMC is traditionally synthesized by reaction of chitosan with excess iodomethane in strong<br />
alkaline conditions with N-methyl-2-pyrrolidone (NMP) <strong>as</strong> solvent and the degree of<br />
quaternization (DQ) can be varied by varying the number of reaction steps [6]. Besides N-<br />
methylation this synthesis method also introduces substantial O-methylation on the hydroxyl<br />
groups located at the C-3 and C-6 of the glucosamine unit. The degree of O-methylation (DOM)<br />
incre<strong>as</strong>es with incre<strong>as</strong>ing DQ (up to 80-90%) [7, 8]. Recently, O-methyl free TMC w<strong>as</strong><br />
synthesized using a novel two-step synthesis procedure, allowing good control of the DQ<br />
without altering other structural properties. Both DQ and DOM were found to influence<br />
toxicity and transepithelial electrical resistance (TEER), an indicator <strong>for</strong> opening of tight<br />
71
Chapter 4<br />
junctions, in a Caco-2 cell model (Chapter 2). A higher DQ leads to more toxicity and a stronger<br />
TEER effect, with a maximum effect on TEER at a DQ above 60%. Furthermore, O-methyl free<br />
TMC h<strong>as</strong> a much stronger effect on TEER than O-methylated TMC (TMC-OM) and shows more<br />
in vitro cell toxicity [8]. Another characteristic of TMCs is the degree of N-acetylation (DAc).<br />
Partial reacetylation of TMC (from 17 to 54%) decre<strong>as</strong>ed the in vitro cell toxicity and effect on<br />
TEER but incre<strong>as</strong>ed the enzymatic degradability of TMC by lysozyme (Chapter 3) [9].<br />
Little is known about the relationship between the structural characteristics and adjuvant<br />
properties of TMCs in vivo. For TMC-OM solutions in i.n. vaccination with ovalbumin an<br />
optimal DQ of 40% w<strong>as</strong> reported although differences were small [10]. Previously, whole<br />
inactivated influenza virus (WIV) vaccine w<strong>as</strong> <strong>for</strong>mulated with TMC-OM with a DQ of 15% or<br />
37%. This resulted in positively charged nanoparticles with partially bound TMC-OM. These<br />
particles had an intact viral ultr<strong>as</strong>tructure. Strong, protective immune responses were induced<br />
after i.n. vaccination [11]. No significant differences were observed between the two different<br />
TMC-OMs. Most likely, TMC exerts its adjuvant effect by an improved antigen delivery, through<br />
an incre<strong>as</strong>ed n<strong>as</strong>al residence time and/or enhanced uptake through the epithelium and by<br />
antigen presenting cells. Besides differences in DQ, the TMC-OMs used in these studies also<br />
differed in DOM and, likely, polymer molecular weight. So, the individual contributions of DQ,<br />
DAc and DOM on the adjuvant effect of TMC are unknown.<br />
In the present study we investigated <strong>for</strong> i.n. administered WIV the adjuvant properties of O-<br />
methyl free TMCs with varying DQs and reacetylated O-methyl free TMC in comparison to<br />
conventional TMC-OMs with similar DQ. The TMC-WIV vaccines were physicochemically<br />
characterized and the immunogenicity and protectivity of the vaccines were <strong>as</strong>sessed in a<br />
murine challenge model. Additionally, the influence of TMC:WIV ratio on the quality and<br />
quantity of humoral immune responses w<strong>as</strong> investigated.<br />
Materials and Methods<br />
Materials. Chitosan with a DAc of 17% (determined with 1 H-NMR <strong>as</strong> described in [9]) and a<br />
number average molecular weight (M n) and weight average molecular weight (M w) of 28 and<br />
43 kDa, <strong>as</strong> determined by gel permeation chromatography (GPC) <strong>as</strong> described in [8],<br />
respectively, w<strong>as</strong> purch<strong>as</strong>ed from Primex (Siglufjordur, Iceland). Acetic anhydride, sodium<br />
borohydrate, <strong>for</strong>mic acid, <strong>for</strong>maldehyde 37% (stabilized with methanol), deuterium oxide,<br />
sodium acetate, acetic acid (anhydrous), sodium hydroxide and hydrochloric acid were<br />
obtained from Sigma-Aldrich Chemical Co. Iodomethane 99% stabilized with copper w<strong>as</strong><br />
72
Relationship between Structure and <strong>Adjuvant</strong>icity of TMC<br />
obtained from Acros Organics (Geel, Belgium). Live, egg-grown, mouse adapted influenza<br />
A/Puerto Rico/8/34 virus (A/PR/8/34) and purified, cell culture-grown (Madin-Darby Canine<br />
Kidney (MDCK) cells), β-propiolacton (BPL)-inactivated A/PR/8/34, <strong>as</strong> well <strong>as</strong> polyclonal<br />
rabbit anti-A/PR/8/34 serum were from Nobilon International BV, Boxmeer, The Netherlands.<br />
PO-labeled goat anti mouse -IgG (H+L), -IgG1, -IgG2a/c and -IgA(Fc) were purch<strong>as</strong>ed from<br />
Nordic Immunological Laboratories (Tilburg, The Netherlands). All other chemicals used were<br />
of analytical grade.<br />
Synthesis and characterization of methylated chitosans. N,N,N-<strong>Trimethyl</strong>ated chitosans<br />
with varying DQ and DAc, and DOM were synthesized from chitosan <strong>as</strong> described previously<br />
[8, 9]. Briefly, O-methyl free TMCs were made with a two-step method: first quantitative<br />
dimethylation of the free amino-groups of chitosan with <strong>for</strong>maldehyde and <strong>for</strong>mic acid w<strong>as</strong><br />
carried out, followed by reaction of the dimethylated chitosan with an excess of iodomethane.<br />
By varying the reaction time the DQ of the TMCs could be tailored [8]. TMC-OMs, with<br />
substantial O-methylation of the hydroxyl groups on the C-3 and C-6 of the glucosamine units,<br />
were synthesized according to the method of Sieval et al. [6, 8]. Here, chitosan w<strong>as</strong><br />
trimethylated to various extents by reacting with iodomethane in the presence of a strong b<strong>as</strong>e<br />
(NaOH) <strong>for</strong> several times depending on the desired DQ. Finally, to obtain TMC with a high<br />
degree of acetylation, chitosan w<strong>as</strong> first re-acetylated using acetic anhydride [9, 12]. Then, this<br />
re-acetylated chitosan w<strong>as</strong> quantitatively dimethylated with <strong>for</strong>maldehyde and sodium<br />
borohydrate followed by complete trimethylation of the dimethylated amino groups with<br />
iodomethane [9]. All synthesized TMCs were dissolved in an aqueous 10% w/v NaCl solution,<br />
put on a shaker overnight <strong>for</strong> ion-exchange and the obtained solution w<strong>as</strong> dialyzed at room<br />
temperature against deionized water <strong>for</strong> 3 days changing water twice daily, filtered through a<br />
0.8 µm filter and freeze dried.<br />
The DQ, DAc and DOM of the hydroxyl groups on C-3 and C-6 (DOM-3 and DOM-6,<br />
respectively) of the TMCs were determined with 1 H-NMR on a Varian INOVA 500MHz NMR<br />
spectrometer (Varian Inc., Palo Alto, Ca, USA) at 80 °C in D 2O [9]. Furthermore, M n and M w of<br />
the various TMCs were determined, <strong>as</strong> described previously [9, 13], by GPC on a Viscotek<br />
system detecting refractive index, viscosity and light scattering. A Shodex OHPak SB-806<br />
column (30 cm) w<strong>as</strong> used with 0.3 M sodium acetate pH 4.4 (adjusted with acetic acid) <strong>as</strong><br />
running buffer. The structural characteristics of the synthesized TMCs are summarized in<br />
Table 1.<br />
73
Chapter 4<br />
Preparation of TMC-WIV <strong>for</strong>mulations. Purified, cell culture-derived, BPL-inactivated<br />
A/PR/8/34 suspended in a 10 mM phosphate buffered saline solution (150 mM NaCl, pH 7.4)<br />
(PBS) w<strong>as</strong> concentrated by centrifugation at 22,000 x g <strong>for</strong> 30 min at 4 °C and resuspended in 5<br />
mM HEPES buffer (pH 7.4). The WIV concentration is expressed <strong>as</strong> mg total protein/ml <strong>as</strong><br />
determined by DC protein <strong>as</strong>say (Bio-Rad, Hercules, CA, USA). The amount of hemagglutinin<br />
(HA) w<strong>as</strong> approximately 35 % of the total protein content, <strong>as</strong> determined previously [14]. The<br />
TMC-WIV vaccines were prepared by adding equal volumes of TMC solution (in 5 mM HEPES,<br />
pH 7.4) to a WIV dispersion (in 5 mM HEPES pH 7.4) at a 1:1 w/w ratio using a Gilson pipette<br />
while gently mixing <strong>for</strong> 5 seconds. TMC-WIV w<strong>as</strong> <strong>for</strong>mulated at a final WIV concentration of<br />
1.25 mg/ml, except <strong>for</strong> the samples used <strong>for</strong> dynamic light scattering (DLS) and zeta-potential<br />
me<strong>as</strong>urements, which were per<strong>for</strong>med at lower concentrations (62.5 µg/ml) <strong>for</strong> optimal testconditions.<br />
To study the immunogenicity of TMC-WIV at other ratios, TMC56 w<strong>as</strong> also<br />
<strong>for</strong>mulated with WIV at ratios of 0.2:1 (TMC56-WIV(0.2:1))and 5:1(TMC56-WIV(5:1)), by<br />
varying the TMC concentration added to WIV.<br />
Dynamic light scattering and zeta-potential me<strong>as</strong>urements. WIV and TMC-WIV<br />
<strong>for</strong>mulations were prepared at a final WIV concentration of 62.5 µg/ml in 5 mM HEPES buffer<br />
pH 7.4. Particle size w<strong>as</strong> me<strong>as</strong>ured by dynamic light scattering (DLS) using a Malvern ALV CGS-<br />
3 (Malvern Instruments, Malvern, UK). DLS results are given <strong>as</strong> a z-average particle size<br />
diameter and a polydispersity index (PDI). The PDI can range from 0 (indicating monodisperse<br />
particles) to 1 (a completely heterodisperse system). Zeta-potential w<strong>as</strong> me<strong>as</strong>ured using a<br />
Zet<strong>as</strong>izer Nano (Malvern Instruments, Malvern, UK).<br />
Quantification of unbound TMCs by GPC. The fraction of the various TMCs that w<strong>as</strong> not<br />
bound to WIV w<strong>as</strong> quantified in the supernatant of centrifuged TMC-WIV <strong>for</strong>mulations, using<br />
the GPC method described earlier [11]. TMC-WIV <strong>for</strong>mulations were centrifuged <strong>for</strong> 40 min at<br />
22,000 x g at 4 °C and the supernatant w<strong>as</strong> collected. Prior to injection, 20 µl GPC running<br />
buffer w<strong>as</strong> added to 100 µl supernatant to adjust the pH of the sample to pH 4.4. The sample<br />
concentration w<strong>as</strong> determined using refractive index detection.<br />
Immunization protocol. Animal experiments were conducted according to the guidelines<br />
provided by the Dutch Animal Protection Act and were approved by a Committee <strong>for</strong> Animal<br />
Experimentation. For all experiments 6-8 weeks old female C57-BL/6 mice (Charles River)<br />
were used. Mice were housed in groups of 7-11 mice and food and water were provided ad<br />
74
Relationship between Structure and <strong>Adjuvant</strong>icity of TMC<br />
libitum. Prime and boost immunizations at day 0 and 21, respectively, were per<strong>for</strong>med without<br />
anesthesia. Groups of 11 mice were vaccinated i.n. with the various TMC-WIV <strong>for</strong>mulations at a<br />
dose of 12.5 µg WIV (corresponding to approximately 4.3 µg HA). All TMC-WIV vaccines were<br />
freshly prepared by mixing WIV dispersion with solutions of the various TMCs, <strong>as</strong> described<br />
above. Additionally, a group of 11 mice were vaccinated with WIV i.n. without TMC. As<br />
negative control groups, one group of 11 mice w<strong>as</strong> treated i.n. with 5 mM HEPES and another<br />
with TMC56 solution (1.25 mg/ml in 5 mM HEPES pH 7.4) without WIV. For i.n. immunization,<br />
mice were held in supine position without anesthesia and the <strong>for</strong>mulations were administered<br />
to the left and right nostril in a total volume of 10 µl. As a reference, one group of mice w<strong>as</strong><br />
vaccinated i.m. with WIV at a dose of 12.5 µg protein in a volume of 100 μl in the left and right<br />
quadriceps <strong>for</strong> prime and boost vaccination, respectively.<br />
Blood sampling and n<strong>as</strong>al w<strong>as</strong>hes. Blood samples were collected by orbital puncture in<br />
MINICOLLECT® serum separator tubes coated with SiO 2 (Greiner Bio-One, Alphen a/d Rijn,<br />
the Netherlands) 20 days after prime vaccination and 17 days after boost vaccination.<br />
Coagulated blood samples were centrifuged at 6,500 x g <strong>for</strong> 8 min at room temperature to<br />
obtain serum samples. Individual serum samples were stored at -20 °C until further analysis.<br />
Seventeen days after boost vaccination, 4 mice from each group were sacrificed by a lethal<br />
intraperitoneal injection of 100 µl sodium pentobarbital (200 mg/ml). The trachea w<strong>as</strong> then<br />
cannulated towards the n<strong>as</strong>opharyngeal duct with a PVC tube (inner/outer diameter 0.5/1.0<br />
mm). PBS (500 μl) containing complete Mini, EDTA free prote<strong>as</strong>e inhibitor (Roche Diagnostics,<br />
Indianapolis, IN, USA) at a concentration of 1 tablet / 7 ml PBS w<strong>as</strong> flushed through the n<strong>as</strong>al<br />
cavity and collected from the nostrils 3 times. The combined n<strong>as</strong>al w<strong>as</strong>hes were stored at -70<br />
°C until further analysis.<br />
Challenge. Twenty one days after the boost vaccination, mice were challenged with 50 ml (2<br />
x 10 8 x the 50% egg infectious dose (EID50)/ml) aerosolized, egg-grown A/PR/8/34 using a<br />
DeVilbiss Ultra-Neb 2000 ultr<strong>as</strong>onic nebulizer (Direct Medical Ltd, Lecarrow, Ireland) <strong>for</strong> 25 min.<br />
After challenge, mice were put back in their cages and any observed signs of illness like<br />
lethargy, standing fur and curved back were recorded. Additionally, their body weight w<strong>as</strong><br />
monitored daily <strong>for</strong> 15 days. For comparison of loss in body weight, the average area under the<br />
curve (AUC) w<strong>as</strong> calculated <strong>for</strong> each group from relative body weight curves of individual mice.<br />
All i.n. groups were compared to the negative control group (PBS i.n.) and the positive control<br />
75
Chapter 4<br />
group (WIV i.m.) by the average (AUC) of individual mice using a one-way ANOVA and<br />
Bonferroni’s correction <strong>for</strong> multiple comparisons.<br />
Hemagglutination inhibition test. First, 25 μl serum w<strong>as</strong> incubated <strong>for</strong> 18 h at 37 °C with<br />
75 μl Receptor Destroying Enzyme (RDE) solution (Denka Seiken UK Ltd, Coventry, UK) to<br />
suppress nonspecific hemagglutination inhibition. RDE w<strong>as</strong> then inactivated by incubating the<br />
mixture <strong>for</strong> 30 min at 56 °C. Next, 150 μl PBS w<strong>as</strong> added to obtain a final 10-fold serum<br />
dilution. Fifty μl diluted serum w<strong>as</strong> transferred in duplicate to V-bottom 96-wells plates<br />
(Greiner, Alphen a/d Rijn, The Netherlands) and serially diluted twofold in PBS. Next, 4<br />
hemagglutination units (HAU) of A/PR/8/34 (in 25 μl PBS) were added and the mixture w<strong>as</strong><br />
incubated <strong>for</strong> 40 min at room temperature. Finally, 50 μl 0.5% (v/v) chicken erythrocytes in<br />
PBS w<strong>as</strong> added. Plates were incubated <strong>for</strong> 1 h at room temperature. The HI titer is expressed <strong>as</strong><br />
the reciprocal value of the highest serum dilution capable of completely inhibiting the virusinduced<br />
agglutination of chicken erythrocytes. If no complete inhibition could be detected in<br />
the first lane, serum w<strong>as</strong> arbitrarily scored 5. Comparison between different experimental<br />
groups from the same dose w<strong>as</strong> made by a one-way ANOVA test and the Tukey post test on the<br />
log trans<strong>for</strong>med HI titers.<br />
Antibody <strong>as</strong>says. Antigen specific serum antibody responses were determined by a<br />
sandwich ELISA. Maxisorp ELISA plates (Nunc, Roskilde, Denmark) were coated overnight<br />
with polyclonal rabbit anti-A/PR/8/34 serum (dilution 1:1620). Plates were w<strong>as</strong>hed in<br />
between all prescribed steps with w<strong>as</strong>h buffer (0.64 M NaCl, 3 mM KCl, 0.15% polysorbate 20<br />
in 10 mM phosphate buffer pH 7.2) using a Skanw<strong>as</strong>her 300 (Molecular Devices, Sunnyvale,<br />
CA, USA). Next, plates were incubated <strong>for</strong> 1 h at 37 °C with blocking buffer (0.2% (w/w) c<strong>as</strong>ein,<br />
4% (w/w) sucrose, 0.05% (w/w) Triton X-100 and 0.01% (w/w) sodium azide in 30 mM TRIS<br />
pH 7.4), followed by incubation <strong>for</strong> 1 h at 37 °C with egg-grown, BPL-inactivated A/PR/8/34<br />
(25.6 HAU/ml). Plates were then incubated with twofold serially diluted sera (100 μl/well) <strong>for</strong><br />
1 h at 37 °C. Next, plates were incubated with 100 μl of a 1:2500 dilution of horseradish<br />
peroxid<strong>as</strong>e linked goat anti mouse -IgG (H+L), -IgG1, -IgG2a/c or -IgA(Fc) (Nordic<br />
Laboratories, Tilburg, the Netherlands) <strong>for</strong> 30 min, and w<strong>as</strong>hed twice. Finally, 100 μl 3,3’,5,5’-<br />
Tetramethylbenzidine (TMB) substrate solution w<strong>as</strong> added and plates were incubated <strong>for</strong> 15<br />
min at room temperature be<strong>for</strong>e enzymatic conversion w<strong>as</strong> stopped by adding 50 μl 2 M<br />
sulfuric acid. Optical density (OD) w<strong>as</strong> then me<strong>as</strong>ured at 450 nm using a Tecan Sunrise plate<br />
reader (Tecan Trading AG, Zurich Switzerland). Titers are given <strong>as</strong> the reciprocal sample<br />
76
Relationship between Structure and <strong>Adjuvant</strong>icity of TMC<br />
dilution corresponding to 20% of the maximal ELISA signal above background. Seronegative<br />
sera were arbitrarily scored with a titer of 15. Comparison between different experimental<br />
groups w<strong>as</strong> made using the log trans<strong>for</strong>med titers by a one-way ANOVA test and the Tukey<br />
post test. Boost IgG2a/c:IgG1 ratios were calculated and compared using the log trans<strong>for</strong>med<br />
data by a one-way ANOVA test and Bonferroni’s correction <strong>for</strong> multiple comparison.<br />
Results and discussion<br />
Structural properties of TMCs. The structural properties of the TMCs used in this study are<br />
summarized in Table 1. The DQ of the O-methyl free TMCs ranged between 30 and 68%,<br />
allowing us to selectively study the influence of trimethylation on adjuvant properties of TMC.<br />
With the TMC-OM group (DQ varying from 22 to 86%, O-methylation from 12 to 76% along<br />
with incre<strong>as</strong>ing DQ) the combined effect of charge density and O-methylation on adjuvanticity<br />
can be studied. The effect of O-methylation can be evaluated by comparing the O-methylated<br />
TMCs with O-methyl free TMCs. Finally, the re-acetylated TMC with a DAc of 54% and a DQ of<br />
44% can be used to <strong>as</strong>sess the role of N-acetylated units in the adjuvant properties of TMCs.<br />
Especially the comparison of TMC43, TMC-OM45 and TMC-RA44 will provide insight into the<br />
optimal structural properties of TMC <strong>for</strong> its use in n<strong>as</strong>al vaccine delivery. Importantly, it is<br />
unlikely that the minor differences in molecular weights between the various TMCs (Table 1)<br />
caused by variation in the degrees of substitutions and/or synthesis methods will affect the<br />
biological properties of the polymers [15].<br />
Table 1. Structural properties of synthesized TMCs. 1)<br />
M n<br />
M w<br />
Abbreviation<br />
(kDa) (kDa)<br />
DQ<br />
(%)<br />
DAc<br />
(%)<br />
DOM-6<br />
(%)<br />
DOM-3<br />
(%)<br />
TMC30 33 59 30 17 - -<br />
TMC43 36 75 43 17 - -<br />
TMC56 37 78 56 17 - -<br />
TMC68 39 84 68 17 - -<br />
TMC-OM22 34 56 22 12 18 12<br />
TMC-OM45 32 49 45 11 25 16<br />
TMC-OM61 31 49 61 10 56 44<br />
TMC-OM86 29 44 86 9 76 72<br />
TMC-RA44 43 83 44 54 - -<br />
1) Degree of acetylation (DAc), quaternization (DQ), O-methylation on C-6 (DOM-6) and on<br />
C-3 (DOM-3) of TMCs were determined by 1 H-NMR analysis. M n, M w were determined by GPC.<br />
77
Chapter 4<br />
Characterization of TMC-WIV <strong>for</strong>mulations. Various TMC-WIV <strong>for</strong>mulations were<br />
prepared by mixing the two components at a 1:1 (w/w) ratio. For comparison, TMC56-WIV<br />
<strong>for</strong>mulations were also prepared at 0.2:1 and 5:1 (w/w) ratios. Particle size and size<br />
distribution were determined and compared with those of plain WIV. Figure 1 shows that plain<br />
WIV had a diameter of approximately 170 nm with a polydispersity index (PDI) of 0.2,<br />
indicating a fairly homogeneous particle size distribution. Formulating the WIV particles with<br />
the various TMCs led to a small incre<strong>as</strong>e in particle size (200-220 nm) and comparable size<br />
distributions, independent of the type of TMC used or the TMC/WIV ratio.<br />
Figure 1. Diameter and size distribution (PDI) of TMC-WIV <strong>for</strong>mulations <strong>as</strong> determined by dynamic<br />
light scattering. Error bars represent standard deviations of three me<strong>as</strong>urements.<br />
Furthermore, the zeta-potential of the TMC-WIV particles w<strong>as</strong> analyzed in 5 mM HEPES pH<br />
7.4. The addition of TMC to negatively charged WIV (-13 mV) in a 1:1 (w/w) ratio resulted in<br />
positively charged particles (+12 to +20 mV) <strong>as</strong> seen in Figure 2, suggesting that all TMCs<br />
adsorb onto WIV in a similar f<strong>as</strong>hion. The zeta-potential incre<strong>as</strong>ed slightly with incre<strong>as</strong>ing DQ<br />
and the lowest surface charge w<strong>as</strong> observed with TMC-OM22-WIV. As expected, compared to<br />
the 1:1 ratio a lower TMC56:WIV ratio (0.2:1 (w/w)) resulted in a lower surface charge (+12<br />
mV vs. +18 mV), but a higher ratio (5:1 (w/w)) did not result in a higher zeta-potential.<br />
78
Relationship between Structure and <strong>Adjuvant</strong>icity of TMC<br />
Figure 2. Zeta-potential of TMC-WIV <strong>for</strong>mulations determined in 5 mM HEPES pH 7.4 Error bars<br />
indicate standard deviations of three samples.<br />
The amount of free TMC present in the TMC-WIV <strong>for</strong>mulations w<strong>as</strong> quantified using GPC. The<br />
concentration and relative amount of free TMC are depicted in Table 2. TMC-WIV <strong>for</strong>mulations<br />
at a ratio of 1:1 (w/w) had an average free TMC content between 0.95 and 1.15 mg/ml<br />
(corresponding to 76 and 91% of total TMC in the <strong>for</strong>mulation), independent of the DQ of the<br />
polymers. TMC-RA44 had the highest amount of TMC bound to the WIV particles. In the<br />
TMC56-WIV(0.2:1) mixture about 50% of the TMC56 remained free in solution, where<strong>as</strong> in the<br />
TMC56-WIV(5:1) almost 98% of the total TMC56 in the <strong>for</strong>mulation remained unbound. These<br />
results, together with the results of the zeta-potential me<strong>as</strong>urements, indicate that WIV is<br />
likely saturated with TMCs at TMC:WIV ratios 1:1 and 5:1 (w/w).<br />
79
Chapter 4<br />
Table 2. Specifications of TMC in various TMC-WIV <strong>for</strong>mulations.<br />
Total TMC in<br />
WIV<br />
TMC:WIV<br />
Formulation<br />
<strong>for</strong>mulation<br />
(µg)<br />
(w/w) ratio<br />
(µg)<br />
Unbound TMC<br />
(mg/ml) ab<br />
Unbound TMC<br />
(% of total) b<br />
TMC30-WIV 12.5 12.5 1:1 1.13± 0.04 90.0±3.3<br />
TMC43-WIV 12.5 12.5 1:1 1.14± 0.03 91.2±2.6<br />
TMC56-WIV 12.5 12.5 1:1 1.09± 0.01 87.1±0.8<br />
TMC68-WIV 12.5 12.5 1:1 1.08± 0.08 86.1±6.2<br />
TMC-OM22-WIV 12.5 12.5 1:1 1.02± 0.04 81.8±3.4<br />
TMC-OM45-WIV 12.5 12.5 1:1 1.05± 0.06 83.9±4.4<br />
TMC-OM61-WIV 12.5 12.5 1:1 1.01± 0.04 80.7±3.2<br />
TMC-OM86-WIV 12.5 12.5 1:1 1.06± 0.01 85.1±0.9<br />
TMC-RA44-WIV 12.5 12.5 1:1 0.95± 0.00 76.0±0.4<br />
TMC56-WIV (0.2:1) 12.5 2.5 0.2:1 0.10± 0.05 49.3±1.5<br />
TMC56-WIV (5:1) 12.5 62.5 5:1 6.11± 0.07 97.7±1.1<br />
WIV 12.5 - - - -<br />
a Amount of free TMC in the TMC-WIV <strong>for</strong>mulations determined by GPC.<br />
b Values are presented <strong>as</strong> the average of three samples ± standard deviation.<br />
Summarizing, mixing WIV with various TMCs in a ratio of 1:1 resulted in particles with<br />
similar size, size distribution, surface charge and amount of free TMC, thus allowing a fair<br />
evaluation of the influence of the structural characteristics of the TMCs on their adjuvant<br />
properties. Altering the TMC:WIV ratio mostly changed the amount of free TMC present in the<br />
<strong>for</strong>mulation and here the contribution of free TMC can be <strong>as</strong>sessed.<br />
Serum antigen specific total IgG after prime and boost vaccination. The TMC-WIV<br />
<strong>for</strong>mulations were compared to WIV alone in an intran<strong>as</strong>al vaccination study. Serum samples<br />
were analyzed <strong>for</strong> antigen-specific total IgG responses three weeks after prime and boost<br />
vaccination (Figure 3). IgG responses were induced by all <strong>for</strong>mulations containing WIV after<br />
prime vaccination and had incre<strong>as</strong>ed after boost vaccination. TMCs with a DAc ≤ 17% had a<br />
strong adjuvant effect that is not critically affected by DOM and DQ. All TMC-WIV <strong>for</strong>mulations,<br />
except TMC-RA44-WIV, showed a higher number of responding mice and elicited significantly<br />
higher IgG titers than WIV alone. Interestingly, TMC-RA44-WIV induced significantly lower<br />
total IgG responses than all other TMC-WIV <strong>for</strong>mulations with a TMC:WIV ratio of 1:1.<br />
Additionally, no significant differences were observed between the various TMC56-WIV ratio<br />
<strong>for</strong>mulations, indicating that free TMC does not strongly influence total IgG titers and that the<br />
lowest dose of TMC that w<strong>as</strong> tested already significantly incre<strong>as</strong>ed immune responses after i.n.<br />
vaccination with WIV.<br />
80
Relationship between Structure and <strong>Adjuvant</strong>icity of TMC<br />
Figure 3. Geometric mean antigen specific total IgG titers three weeks after prime and boost<br />
vaccination. Error bars indicate 95% confidence intervals (n=11). Non-responding mice were arbitrarily<br />
given a total IgG serum titer of 15. Below the x-axis the number of seropositive mice per group after<br />
boost vaccination are depicted. All titers after boost vaccination were compared using a one-way ANOVA<br />
test and Tukey’s post test. *** p
Chapter 4<br />
the IgG2a/c:IgG1 ratio per individual mouse illustrates that the Th1/Th2 balance of humoral<br />
immune responses shifts towards Th2 with incre<strong>as</strong>ing TMC:WIV ratio (Figure 4B).<br />
A<br />
B<br />
Figure 4. A) IgG1 and IgG2a/c titers elicited by WIV <strong>for</strong>mulations with varying TMC56:WIV ratios<br />
after prime and boost vaccination. Error bars indicate 95% confidence intervals (n=11). Indicated<br />
above the bars is the number of mice that developed detectable IgG1 or IgG2a/c titers (e.g. 1/11<br />
indicates one out of eleven mice). ** p
Relationship between Structure and <strong>Adjuvant</strong>icity of TMC<br />
HI titers. After boost vaccination, HI titers were hardly detectable in any of the i.n.<br />
vaccinated groups (data not shown). Only WIV i.m. induced substantial HI titers (average HI<br />
titer of 160). This is in line with previous results [11, 14]<br />
Antigen-specific secretory IgA (sIgA) in n<strong>as</strong>al w<strong>as</strong>hes. N<strong>as</strong>al w<strong>as</strong>hes were per<strong>for</strong>med 17<br />
days after boost vaccination to determine whether antigen-specific sIgA were elicited by any of<br />
the various <strong>for</strong>mulations. As shown in Figure 5 only in the TMC-WIV <strong>for</strong>mulations some mice<br />
developed detectable sIgA responses in the n<strong>as</strong>al cavity. Interestingly, although WIV i.m.<br />
immunization showed the highest serum IgG titers, no sIgA w<strong>as</strong> found in the n<strong>as</strong>al w<strong>as</strong>hes of<br />
any of these mice. Altogether, the determination of sIgA in the n<strong>as</strong>al mucus showed a relatively<br />
high variation within the <strong>for</strong>mulation-groups (<strong>as</strong> observed by others [10]) likely due to the<br />
collection and detection methods.<br />
OD 450<br />
1.4<br />
1.2<br />
1.0<br />
0.8<br />
0.6<br />
0.4<br />
0.2<br />
0.0<br />
TMC30-WIV<br />
TMC43-WIV<br />
TM56-WIV<br />
TMC68-WIV<br />
TMC-OM20-WIV<br />
TMC-OM45-WIV<br />
TMC-OM61-WIV<br />
TMC-OM86-WIV<br />
TMC-RA44-WIV<br />
TMC56-WIV (0.2:1)<br />
TMC56-WIV (5:1)<br />
WIV<br />
TMC56 sol<br />
WIV i.m.<br />
HEPES<br />
Figure 5. Antigen-specific secretory IgA levels in n<strong>as</strong>al w<strong>as</strong>hes of four individual mice per <strong>for</strong>mulation<br />
17 days after boost vaccination. Only in groups represented by black dots sIgA positive w<strong>as</strong>hes were<br />
found.<br />
Challenge with live, aerosolized virus. To <strong>as</strong>sess the protective effect of the immunization,<br />
seven mice per group were challenged with potentially lethal, homologous, egg-grown<br />
influenza virus and loss of body weight <strong>as</strong> a me<strong>as</strong>ure of illness w<strong>as</strong> monitored. All mice<br />
vaccinated i.n. with O-methyl free or O-methylated TMC-WIV <strong>for</strong>mulations were protected<br />
against the live virus. As representative examples the average body weight over 15 days after<br />
83
Chapter 4<br />
the challenge of the mice immunized with TMC43-WIV and TMC-OM45-WIV <strong>for</strong>mulations are<br />
shown in Figure 6A and B.<br />
A<br />
B<br />
C<br />
D<br />
Figure 6. Average relative body weight curves after challenge of mice (n=7) vaccinated with a) TMC43-<br />
WIV; b) TMC-OM45-WIV; c) TMC-RA44-WIV and d) WIV i.n.. For comparison each panel contains the<br />
curves of WIV i.m. and HEPES i.n.. Error bars indicate the 95% confidence intervals. *** p< 0.001 and **<br />
p< 0.01 indicate that average AUCs are significantly higher than the HEPES i.n. group. ### p< 0.001 and #<br />
p
Relationship between Structure and <strong>Adjuvant</strong>icity of TMC<br />
Additionally, there were no signs of illness observed in these groups and the AUC of the<br />
average body weight w<strong>as</strong> significantly higher (p
Chapter 4<br />
Furthermore, O-methylated TMCs had similar effects on both antibody titers and protection <strong>as</strong><br />
O-methyl free TMCs, indicating that the DOM does not affect the adjuvant properties of TMC.<br />
Interestingly, these findings do not correlate with the large influence of DQ and O-methylation<br />
in vitro on cell toxicity and TEER. This teaches us that DQ and DOM have a much more<br />
pronounced effect on in vitro cell toxicity and TEER <strong>as</strong>says than on mucosal adjuvant<br />
properties. It should be noted that the DOM and DQ may have an influence on the activity of<br />
TMC <strong>as</strong> a penetration enhancer in mucosal drug delivery, <strong>as</strong> suggested by reports on the<br />
influence of DQ [18-20]. As HA likely is too big to p<strong>as</strong>s even fully opened tight junctions, it is<br />
unlikely that opening of tight junctions will directly improve antigen uptake from the n<strong>as</strong>al<br />
cavity.<br />
The reacetylation of TMC on the other hand induced a strong decre<strong>as</strong>e in adjuvant effect<br />
illustrated by lower antibody titers and poor protection against challenge with live influenza<br />
virus. Since the mucoadhesion of particulate systems can be attributed to positive charge and<br />
hydrophobic effect [3], it is likely that TMC-RA44-WIV h<strong>as</strong> similar or even better<br />
mucoadhesive properties <strong>as</strong> the other TMC-WIV vaccines. Also, the zeta-potential of the<br />
particles w<strong>as</strong> similar to the other TMC-coated particles, implying that the introduction of a<br />
positive surface charge alone is not sufficient to improve adjuvanticity. There<strong>for</strong>e other factors<br />
must be responsible <strong>for</strong> the decre<strong>as</strong>ed adjuvant effect of reacetylated TMC.<br />
The first explanation <strong>for</strong> the decre<strong>as</strong>ed adjuvant effect of TMC-RA44 is a difference in<br />
interaction with cells, illustrated by a much lower in vitro toxicity and TEER effect than most<br />
other TMCs on Caco-2 cells [9]. TMC-OM22, however, hardly induced a TEER effect or cell<br />
toxicity either but showed to be a good mucosal adjuvant. This indicates that TEER and toxicity<br />
studies, <strong>as</strong> carried out <strong>for</strong> these TMCs [8, 9] cannot fully explain the loss of adjuvant effect by<br />
reacetylation.<br />
A second explanation <strong>for</strong> the poor adjuvant properties of TMC-RA44 is its enhanced<br />
enzymatic degradation by lysozyme compared to other TMCs. Previous research showed that<br />
the extent of lysozyme-catalyzed degradation of TMC is highly dependent on the DAc; TMC-<br />
RA44 showed a large decre<strong>as</strong>e in molecular weight while TMCs with DAc of ≤17% were only<br />
slightly degraded in presence of lysozyme [9]. Lysozyme is a strong antibacterial cationic<br />
protein that is excreted in high concentrations in the n<strong>as</strong>al cavity [21]. This may result in rapid<br />
degradation of the TMC-RA44 after i.n. administration, thereby strongly limiting its adjuvant<br />
effect on the WIV vaccination. Finally, chitin, an insoluble polysaccharide of N-<br />
acetylglucosamine units, which are also present in TMC-RA44, can be recognized by specific<br />
receptors of the innate immune system [22]. It h<strong>as</strong> recently been suggested that chitin induces<br />
86
Relationship between Structure and <strong>Adjuvant</strong>icity of TMC<br />
pro-inflammatory but also anti-inflammatory signals depending on its size [23]. Further<br />
studies should be done to investigate whether TMC-RA44 also induces anti-inflammatory<br />
signals through interaction with the innate immune system.<br />
In addition to changes in the type of TMC, the influence of free TMC w<strong>as</strong> studied by varying<br />
the TMC concentration in the <strong>for</strong>mulation. An excess of free TMC induced a shift towards<br />
higher IgG1 titers after boost vaccination. Several studies investigated the influence of TMC on<br />
the quality of immune responses [24-26] while other studies determined the effect of adjuvant<br />
dose on the type of responses elicited, using different adjuvants [27, 28]. However, since the<br />
type of adjuvant, animal model used, route of administration and animal model can affect the<br />
quality of immune responses, it is difficult to compare these studies to our data. This is the first<br />
time that the effect of adjuvant dose w<strong>as</strong> studied <strong>for</strong> TMC and additional studies should be<br />
per<strong>for</strong>med to provide mechanistical insight.<br />
Conclusion<br />
All TMC-WIV <strong>for</strong>mulations had comparable physicochemical properties, and there<strong>for</strong>e<br />
observed differences in immunogenicity are related to the various chemical structures of the<br />
TMCs. Formulating WIV with TMCs strongly enhances the immunogenicity and protection of<br />
i.n. vaccination with WIV. The adjuvant properties of TMCs <strong>as</strong> i.n. adjuvant are strongly<br />
decre<strong>as</strong>ed by reacetylation of TMC, where<strong>as</strong> the DQ and DOM did not significantly affect the<br />
adjuvant effect of TMC.<br />
Acknowledgement. This research w<strong>as</strong> partially per<strong>for</strong>med under the framework of <strong>TI</strong><br />
<strong>Pharma</strong> project number D5-106-1; Vaccine delivery: alternatives <strong>for</strong> conventional multiple<br />
injection vaccines. The authors acknowledge Frouke Kuijer <strong>for</strong> making the graphical abstract.<br />
87
Chapter 4<br />
References<br />
1. B. Slütter, N. Hagenaars, W. Jiskoot. Rational design of n<strong>as</strong>al vaccines. J Drug Target 16:1-17<br />
2008).<br />
2. V. Grabovac, D. Guggi, A. Bernkop-Schnürch. Comparison of the mucoadhesive properties of<br />
various polymers. Adv Drug Deliv Rev 57:1713-23 (2005).<br />
3. S.E. Harding. Mucoadhesive interactions. Biochem Soc Trans 31:1036-41 (2003).<br />
4. R.C. Read, S.C. Naylor, C.W. Potter, J. Bond, I. Jabbal-Gill, A. Fisher, L. Illum, R. Jennings.<br />
Effective n<strong>as</strong>al influenza vaccine delivery using chitosan. Vaccine 23:4367-74 (2005).<br />
5. A. Vila, A. Sanchez, K. Janes, I. Behrens, T. Kissel, J.L. Vila Jato, M.J. Alonso. Low molecular<br />
weight chitosan nanoparticles <strong>as</strong> new carriers <strong>for</strong> n<strong>as</strong>al vaccine delivery in mice. Eur J Pharm<br />
Biopharm 57:123-31 (2004).<br />
6. A.B. Sieval, M. Thanou, A.F. Kotze, J.E. Verhoef, J. Brussee, H.E. Junginger. Preparation and<br />
NMR characterization of highly substituted N-trimethyl chitosan chloride. Carbohydr<br />
Polym 36:157-65 (1998).<br />
7. A. Polnok, G. Borchard, J.C. Verhoef, N. Sarisuta, H.E. Junginger. Influence of methylation<br />
process on the degree of quaternization of N-trimethyl chitosan chloride. Eur J Pharm<br />
Biopharm 57:77-83 (2004).<br />
8. R.J. Verheul, M. Amidi, S. van der Wal, E. van Riet, W. Jiskoot, W.E. Hennink. Synthesis,<br />
characterization and in vitro biological properties of O-methyl free N,N,N-trimethylated<br />
chitosan. Biomaterials 29:3642-9 (2008).<br />
9. R.J. Verheul, M. Amidi, M.J. van Steenbergen, E. van Riet, W. Jiskoot, W.E. Hennink. Influence<br />
of the degree of acetylation on the enzymatic degradation and in vitro biological properties of<br />
trimethylated chitosans. Biomaterials 30:3129-35 (2009).<br />
10. W. Boonyo, H.E. Junginger, N. Waranuch, A. Polnok, T. Pitaksuteepong. Chitosan and<br />
trimethyl chitosan chloride (TMC) <strong>as</strong> adjuvants <strong>for</strong> inducing immune responses to ovalbumin in<br />
mice following n<strong>as</strong>al administration. J Control Rele<strong>as</strong>e 121:168-75 (2007).<br />
11. N. Hagenaars, E. M<strong>as</strong>trobattista, R.J. Verheul, I. Mooren, H.L. Glansbeek, J.G. Heldens, H. van<br />
den Bosch, W. Jiskoot. Physicochemical and Immunological Characterization of N,N,N-<br />
<strong>Trimethyl</strong> Chitosan-Coated Whole Inactivated Influenza Virus Vaccine <strong>for</strong> Intran<strong>as</strong>al<br />
Administration. Pharm Res 26:1353-64 (2009)<br />
12. T. Kiang, J. Wen, H.W. Lim, K.W. Leong. The effect of the degree of chitosan deacetylation on<br />
the efficiency of gene transfection. Biomaterials 25:5293-301 (2004).<br />
13. X. Jiang, A. van der Horst, M.J. van Steenbergen, N. Akeroyd, C.F. van Nostrum, P.J.<br />
Schoenmakers, W.E. Hennink. Molar-m<strong>as</strong>s characterization of cationic polymers <strong>for</strong> gene<br />
delivery by aqueous size-exclusion chromatography. Pharm Res 23:595-603 (2006).<br />
14. N. Hagenaars, E. M<strong>as</strong>trobattista, H. Glansbeek, J. Heldens, H. van den Bosch, V. Schijns, D.<br />
Betbeder, H. Vromans, W. Jiskoot. Head-to-head comparison of four nonadjuvanted<br />
inactivated cell culture-derived influenza vaccines: Effect of composition, spatial organization<br />
and immunization route on the immunogenicity in a murine challenge model. Vaccine 26<br />
:6555-63 (2008).<br />
15. S. Mao, X. Shuai, F. Unger, M. Wittmar, X. Xie, T. Kissel. Synthesis, characterization and<br />
cytotoxicity of poly(ethylene glycol)-graft-trimethyl chitosan block copolymers. Biomaterials 26<br />
:6343-56 (2005).<br />
16. D. Mei, S. Mao, W. Sun, Y. Wang, T. Kissel. Effect of chitosan structure properties and<br />
molecular weight on the intran<strong>as</strong>al absorption of tetramethylpyrazine phosphate in rats. Eur J<br />
Pharm Biopharm 70:874-881 (2008).<br />
17. M. Huang, E. Khor, L.Y. Lim. Uptake and cytotoxicity of chitosan molecules and nanoparticles:<br />
effects of molecular weight and degree of deacetylation. Pharm Res 21:344-53 (2004).<br />
18. M.M. Thanou, A.F. Kotze, T. Scharringhausen, H.L. Luessen, A.G. de Boer, J.C. Verhoef, H.E.<br />
Junginger. Effect of degree of quaternization of N-trimethyl chitosan chloride <strong>for</strong> enhanced<br />
88
Relationship between Structure and <strong>Adjuvant</strong>icity of TMC<br />
transport of hydrophilic compounds across intestinal caco-2 cell monolayers. J Control Rele<strong>as</strong>e<br />
64:15-25 (2000).<br />
19. J.H. Hamman, M. Stander, A.F. Kotze. Effect of the degree of quaternisation of N-trimethyl<br />
chitosan chloride on absorption enhancement: in vivo evaluation in rat n<strong>as</strong>al epithelia. Int J<br />
Pharm 232:235-42 (2002).<br />
20. G. Sandri, S. Rossi, M.C. Bonferoni, F. Ferrari, Y. Zambito, G. Di Colo, C. Caramella. Buccal<br />
penetration enhancement properties of N-trimethyl chitosan: Influence of quaternization degree<br />
on absorption of a high molecular weight molecule. Int J Pharm 297:146-55 (2005).<br />
21. A.M. Cole, H.I. Liao, O. Stuchlik, J. Tilan, J. Pohl, T. Ganz. Cationic polypeptides are required<br />
<strong>for</strong> antibacterial activity of human airway fluid. J Immunol 169:6985-91 (2002).<br />
22. C.G. Lee, C.A. Da Silva, J.Y. Lee, D. Hartl, J.A. Eli<strong>as</strong>. Chitin regulation of immune responses:<br />
an old molecule with new roles. Curr Opin Immunol 20:684-9 (2008).<br />
23. C.A. Da Silva, C. Chalouni, A. Williams, D. Hartl, C.G. Lee, J.A. Eli<strong>as</strong>. Chitin is a sizedependent<br />
regulator of macrophage TNF and IL-10 production. J Immunol 182:3573- 82 (2009).<br />
24. M. Amidi, H.C. Pellikaan, H. Hirschberg, A.H. de Boer, D.J. Crommelin, W.E. Hennink, G.<br />
Kersten, W. Jiskoot. Diphtheria toxoid-containing microparticulate powder <strong>for</strong>mulations <strong>for</strong><br />
pulmonary vaccination: preparation, characterization and evaluation in guinea pigs. Vaccine 25<br />
:6818-29 (2007).<br />
25. M. Amidi, S.G. Romeijn, J.C. Verhoef, H.E. Junginger, L. Bungener, A. Huckriede, D.J.<br />
Crommelin, W. Jiskoot. N-trimethyl chitosan (TMC) nanoparticles loaded with influenza<br />
subunit antigen <strong>for</strong> intran<strong>as</strong>al vaccination: biological properties and immunogenicity in a mouse<br />
model. Vaccine 25:144-53 (2007).<br />
26. B. Sayin, S. Somavarapu, X.W. Li, M. Thanou, D. Sesardic, H.O. Alpar, S. Senel. Mono-Ncarboxymethyl<br />
chitosan (MCC) and N-trimethyl chitosan (TMC) nanoparticles <strong>for</strong> non-inv<strong>as</strong>ive<br />
vaccine delivery. Int J Pharm 363:139-48 (2008).<br />
27. M.J. McCluskie, R.D. Weeratna, J.D. Clements, H.L. Davis. Mucosal immunization of mice<br />
using CpG DNA and/or mutants of the heat-labile enterotoxin of Escherichia coli <strong>as</strong> adjuvants.<br />
Vaccine 19:3759-68 (2001).<br />
28. K. Riedl, R. Riedl, A. von Gabain, E. Nagy, K. Lingnau. The novel adjuvant IC31 strongly<br />
improves influenza vaccine-specific cellular and humoral immune responses in young adult and<br />
aged mice. Vaccine 26:3461-8 (2008).<br />
89
CHAPTER 5A<br />
A STEP-BY-STEP APPROACH<br />
TO STUDY THE INFLUENCE OF N-<br />
ACETYLA<strong>TI</strong>ON ON THE ADJUVAN<strong>TI</strong>CITY OF<br />
N,N,N-TRIMETHYL CHITOSAN (TMC) IN AN<br />
INTRANASAL WHOLE INAC<strong>TI</strong>VATED<br />
INFLUENZA VIRUS VACCINE<br />
Rolf J. Verheul*, Niels Hagenaars*, Thom<strong>as</strong> van Es, Ethlinn van Gaal,<br />
P<strong>as</strong>cal H.J.L.F. de Jong, Sven Bruins, Ivo Que, Bram Slütter,<br />
Enrico M<strong>as</strong>trobattista, Harrie L. Glansbeek, Jacco G.M. Heldens,<br />
Han van den Bosch, Wim E. Hennink, Wim Jiskoot.<br />
*authors contributed equally<br />
Manuscript submitted
Chapter 5A<br />
Abstract<br />
In a previous study, we observed that reacetylation of N,N,N-trimethyl chitosan (TMC)<br />
reduced the adjuvant effect of TMC in mice after intran<strong>as</strong>al (i.n.) administration of whole<br />
inactivated influenza virus (WIV) vaccine. The aim of the present study w<strong>as</strong> to elucidate the<br />
re<strong>as</strong>on <strong>for</strong> the lack of adjuvanticity of reacetylated TMC (TMC-RA) by comparing TMC-RA<br />
(degree of acetylation 54%) with TMC (degree of acetylation 17%) at six potentially critical<br />
steps in the induction of an immune response after i.n. administration in mice: chemical<br />
stability of the polymer in a n<strong>as</strong>al w<strong>as</strong>h, local i.n. distribution of WIV, n<strong>as</strong>al residence time of<br />
WIV, cellular uptake of WIV by epithelial cells, transport of WIV by epithelial cells, and capacity<br />
of the <strong>for</strong>mulation to induce maturation of murine bone marrow derived dendritic cells (DCs).<br />
GPC analysis showed that TMC-RA w<strong>as</strong> degraded in a n<strong>as</strong>al w<strong>as</strong>h to a slightly larger extent<br />
than TMC. The local i.n. distribution and n<strong>as</strong>al clearance were similar <strong>for</strong> both TMC types.<br />
Fluorescently labeled WIV w<strong>as</strong> taken up more efficiently by Calu-3 cells when <strong>for</strong>mulated with<br />
TMC-RA compared to TMC and both TMCs significantly reduced transport of WIV over a Calu-3<br />
monolayer. Murine bone-marrow derived dendritic cell activation w<strong>as</strong> similar <strong>for</strong> plain WIV,<br />
and WIV <strong>for</strong>mulated with TMC-RA or TMC.<br />
In conclusion, the inferior adjuvant effect of TMC-RA over that of TMC might be caused by a<br />
slightly lower stability of TMC-RA-WIV in the n<strong>as</strong>al cavity, rather than by any of the other<br />
factors studied in this paper.<br />
92
Influence of N-Acetylation on the <strong>Adjuvant</strong>icity of TMC<br />
Introduction<br />
Intran<strong>as</strong>al (i.n.) immunization offers several advantages over traditional parenteral routes,<br />
such <strong>as</strong> painless, needle-free administration without the need <strong>for</strong> trained personnel and the<br />
potential induction of both systemic and local immune responses [1-3]. Nonetheless, its<br />
success h<strong>as</strong> been limited due to delivery issues <strong>as</strong>sociated with the physiology of the n<strong>as</strong>al<br />
epithelium and the tolerogenic nature of immune cells in the n<strong>as</strong>al <strong>as</strong>sociated lymphoid tissue.<br />
This is illustrated by the strong immune responses that the same vaccines elicit after<br />
intramuscular administration compared to n<strong>as</strong>al administration [4]. A successful i.n. vaccine<br />
<strong>for</strong>mulation should there<strong>for</strong>e adhere to the mucosal surfaces of the n<strong>as</strong>al cavity and provide<br />
protection against the degradation of the antigen in the n<strong>as</strong>al environment. Next, the<br />
<strong>for</strong>mulation should induce the uptake and/or transport of antigen through the epithelium, to<br />
ensure that a sufficient amount of antigen will reach the antigen presenting cells in the<br />
subepithelial space. Finally, these APCs (most noticeably dendritic cells (DCs)), have to be<br />
activated (matured) either by the antigen or by addition of an adjuvant, in order <strong>for</strong> them to<br />
migrate to the nearby lymph nodes and elicit the desired type of immune response [5-8].<br />
N,N,N-trimethyl chitosan (TMC), a quaternized, water-soluble derivative of chitosan, h<strong>as</strong><br />
previously been shown to possess mucoadhesive [9] and absorption enhancing [10-12]<br />
characteristics and when <strong>for</strong>mulated with whole inactivated influenza virus vaccine (WIV)<br />
greatly enhanced the efficacy of the i.n. administered vaccine in mice [13]. To further improve<br />
its adjuvant effect, the structural properties of TMC, like the degree of quaternization (DQ), the<br />
degree of O-methylation (DOM), and the degree of N-acetylation (DAc), can be varied during<br />
synthesis [14, 15]. In a previous study on the influence of these structural properties on the<br />
adjuvant effect of TMC in i.n. WIV vaccines in mice [16], we showed that both the DQ and the<br />
DOM of TMC did not have a significant effect on the immunogenicity of the i.n. WIV vaccines,<br />
where<strong>as</strong> the DAc did. A striking loss in adjuvanticity of TMC w<strong>as</strong> observed when WIV w<strong>as</strong><br />
<strong>for</strong>mulated with re-acetylated TMC (TMC-RA) with a DAc of 54% and a DQ of 44% (compared<br />
to a DAc of 17% <strong>for</strong> other tested TMCs). Importantly, the physical characteristics of WIV<br />
<strong>for</strong>mulated with TMC-RA or with TMC (both polymers having a DQ of about 45%) were similar<br />
<strong>for</strong> size, zeta-potential and amount of unbound TMC, indicating that these physical properties<br />
can not be held accountable <strong>for</strong> the difference in adjuvant effect. In contr<strong>as</strong>t to TMC, TMC-RA is<br />
a substrate <strong>for</strong> lysozyme, h<strong>as</strong> a lower toxicity profile than TMC and h<strong>as</strong> no effect on the<br />
transepithelial electrical resistance (TEER) of Caco-2 cells [15].<br />
The aim of this study, w<strong>as</strong> to understand why TMC-RA, in contr<strong>as</strong>t to TMC, does not act <strong>as</strong> an<br />
adjuvant <strong>for</strong> an i.n. WIV vaccine in mice. This will provide more insight into the mechanism of<br />
93
Chapter 5A<br />
adjuvanticity of TMC and may aid in the rational, structural design of TMC <strong>as</strong> an adjuvant. TMC-<br />
WIV, TMC-RA-WIV and WIV <strong>for</strong>mulations were compared at six potentially critical steps in the<br />
induction of an immune response after i.n. administration. In particular, we studied (i) the<br />
stability of TMC and TMC-RA in n<strong>as</strong>al w<strong>as</strong>hes, (ii) the effect of both polymers on the n<strong>as</strong>al<br />
residence time of WIV, (iii) n<strong>as</strong>al distribution patterns, (iv) cellular uptake and (v) transport of<br />
WIV through an epithelial (Calu-3) cell line and (vi) the effect of the different <strong>for</strong>mulations on<br />
maturation of murine bone-marrow derived dendritic cells (DCs).<br />
Materials and Methods<br />
Materials. Chitosan w<strong>as</strong> purch<strong>as</strong>ed from Primex (Siglufjordur, Iceland) and had a DAc of 17%<br />
and a number average molecular weight (M n) and weight average molecular weight (M w) of 28<br />
and 43 kDa, respectively. Acetic anhydride, sodium borohydrate, <strong>for</strong>mic acid, <strong>for</strong>maldehyde<br />
37% (stabilized with methanol), deuterium oxide, sodium acetate, acetic acid (anhydrous),<br />
sodium hydroxide and hydrochloric acid were obtained from Sigma-Aldrich Chemical Co.<br />
Iodomethane 99% stabilized with copper w<strong>as</strong> obtained from Acros Organics (Geel, Belgium).<br />
Purified, cell culture-grown (Madin-Darby Canine Kidney (MDCK) cells), β-propiolacton (BPL)-<br />
inactivated A/PR/8/34, <strong>as</strong> well <strong>as</strong> polyclonal rabbit anti-A/PR/8/34 serum were from Nobilon<br />
International BV (Boxmeer, The Netherlands). Alexa Fluor® 488-labeled goat anti-rabbit IgG,<br />
Alexa Fluor® 488 labeling kit and 4,6-diamidino-2-phenylindole dihydrochloride (DAPI) were<br />
purch<strong>as</strong>ed from Invitrogen (Breda, the Netherlands). Osteosoft decalcifier (10% EDTA) w<strong>as</strong><br />
from Merck Serono (Schiphol, the Netherlands). Citric acid, Triton-X, trypsin from bovine<br />
pancre<strong>as</strong>, normal goat serum and normal rabbit serum were obtained from Sigma<br />
(Zwijndrecht, the Netherlands). Phosphate buffered saline (10 mM phosphate buffer, pH 7.4,<br />
150 mM NaCl; PBS) w<strong>as</strong> purch<strong>as</strong>ed from Braun (Melsungen, Germany). IRDye800CW®-labeled<br />
epidermal growth factor (EGF) and the IRDye800CW® protein labeling kit were from LI-COR<br />
Biosciences (Lincoln, NE, USA). Fixative (4% <strong>for</strong>malin in phosphate buffer pH 7.0) w<strong>as</strong><br />
obtained from Klinipath (Duiven, the Netherlands). Calu-3 cells were obtained from ATCC<br />
(Man<strong>as</strong>s<strong>as</strong>, VA, USA). Virally trans<strong>for</strong>med human bronchial epithelial cell line 16HBE14o- w<strong>as</strong><br />
a kind gift from Dr. D. Gruenert (University of Cali<strong>for</strong>nia at San Francisco). Trypsin/EDTA 10x,<br />
Plain DMEM (Dulbecco’s modified Eagle’s medium, with 3.7 g/l sodium bicarbonate, 1 g/l L-<br />
glucose, L-glutamine) and antibiotics/antimycotics (penicillin, streptomycin sulfate,<br />
amphotericin B) were from PAA Laboratories GmbH (P<strong>as</strong>ching, Austria) and propidium iodide<br />
(PI) and MEM with 0.292g/L L-glutamine, 1 g/L Glucose, 2.2 g/L NaHCO3, w<strong>as</strong> from Invitrogen<br />
94
Influence of N-Acetylation on the <strong>Adjuvant</strong>icity of TMC<br />
(Breda, The Netherlands). Matrigel ® w<strong>as</strong> acquired from Becton Dickinson (Franklin Lakes, NJ<br />
USA). All other chemicals used were of analytical grade. Animal experiments were conducted<br />
according to the guidelines provided by the Dutch Animal Protection Act and were approved<br />
by a Committee <strong>for</strong> Animal Experimentation.<br />
Synthesis and characterization of trimethylated chitosans. O-methyl free TMCs with a<br />
DQ of 45% and varying DAc were synthesized from chitosan <strong>as</strong> described previously [15]. The<br />
DQ and DAc of the TMCs were determined with 1H-NMR on a Varian INOVA 500 MHz NMR<br />
spectrometer (Varian Inc., Palo Alto, Ca, USA) at 80 °C in D 2O. Furthermore, number average<br />
weight (M n) and weight average weight (M w) of the TMCs were determined <strong>as</strong> described be<strong>for</strong>e<br />
[15] by GPC on a Viscotek system detecting refractive index, viscosity and light scattering. A<br />
Shodex OHPak SB-806 column (30 cm) w<strong>as</strong> used with 0.3 M sodium acetate pH 4.4 (adjusted<br />
with acetic acid) <strong>as</strong> running buffer. The structural characteristics of the synthesized TMCs are<br />
summarized in Table 1.<br />
Preparation and characterization of TMC-WIV <strong>for</strong>mulations. TMC-WIV <strong>for</strong>mulations<br />
were prepared <strong>as</strong> described previously (16). Briefly, purified, cell culture-grown (Madin-Darby<br />
Canine Kidney (MDCK) cells), β-propiolacton (BPL)-inactivated mouse adapted influenza<br />
A/Puerto Rico/8/34 virus (A/PR/8/34) suspended in a 10 mM phosphate buffered saline (150<br />
mM NaCl, pH 7.4) w<strong>as</strong> concentrated by centrifugation at 22,000xg <strong>for</strong> 30 min at 4 °C and<br />
resuspended in 5 mM HEPES buffer (pH 7.4). WIV suspension w<strong>as</strong> mixed with equal volumes<br />
of TMC or TMC-RA in 1:1 or 5:1 weight/weight ratio. For dynamic light scattering (DLS)<br />
(Malvern ALV CGS-3, Malvern Instruments, Malvern, UK) and zeta-potential me<strong>as</strong>urements<br />
(Zet<strong>as</strong>izer Nano, Malvern Instruments, Malvern, UK) a final WIV concentration of 62.5 μg/ml<br />
(expressed <strong>as</strong> total protein concentration) in 5 mM HEPES pH 7.4 w<strong>as</strong> used. DLS results are<br />
given <strong>as</strong> z-average particle size diameter and a polydispersity index (PDI). The PDI can vary<br />
from 0 (indicating monodisperse particles) to 1 (a completely heterodisperse particle size<br />
distribution)).<br />
Degradation of TMCs in n<strong>as</strong>al w<strong>as</strong>h. Six female 6-8 weeks old Balb/c nu/nu mice (Charles<br />
River, L’Arbresle, France) were sacrificed by a lethal intraperitoneal injection of 100 μl sodium<br />
pentobarbital (200 mg/ml). The trachea of each mouse w<strong>as</strong> then cannulated towards the<br />
n<strong>as</strong>opharyngeal duct with a PVC tube (inner/outer diameter 0.5/1.0 mm). PBS (600 μl) w<strong>as</strong><br />
flushed through the n<strong>as</strong>al cavity and collected from the nostrils 3 times; the collected samples<br />
95
Chapter 5A<br />
were pooled. TMC and TMC-RA were dissolved in 5 mM HEPES at 2.5 mg/ml, mixed 1:1 (v/v)<br />
with n<strong>as</strong>al w<strong>as</strong>h and the mixtures were incubated at 37 °C. Samples were taken after 4 and 24<br />
hours and 5 and 9 days. M n and M w were determined by GPC on a Viscotek system detecting<br />
refractive index, viscosity and light scattering. A Shodex OHPak SB806 column (15 cm) w<strong>as</strong><br />
used with 0.3 M sodium acetate, pH 4.4 (adjusted with acetic acid), <strong>as</strong> running buffer [17].<br />
Pullulan (M n 102 kDa, M n 106 kDa) obtained from Viscotek Benelux (Oss, The Netherlands)<br />
w<strong>as</strong> used <strong>for</strong> calibration and polymer solutions mixed with PBS 1:1 (v/v) were used <strong>as</strong><br />
controls. Prior to injection, 30 μl GPC running buffer w<strong>as</strong> added to 120 μl sample to adjust the<br />
pH of the sample to pH 4.4.<br />
In vivo imaging of n<strong>as</strong>al residence time of TMC-WIV <strong>for</strong>mulations. In vivo imaging of<br />
TMC-WIV <strong>for</strong>mulations w<strong>as</strong> done according to a previous study [18]. Female nude (Balb/c<br />
nu/nu) mice were obtained from Charles River (L’Arbresle, France) and received plain<br />
IRDye800CW®-labeled WIV (n=9) or <strong>for</strong>mulated with TMC (n=9) or TMC-RA (n=3) in a ratio<br />
of 1:1 (w/w) under light anesthesia with isoflurane inhalation. Next, mice were scanned with<br />
an IVIS Spectrum imaging system from Caliper Life Sciences (Hopkinton, MA, USA). Scans were<br />
per<strong>for</strong>med regularly over at le<strong>as</strong>t 2 h. Scanned images were analyzed using Living Image 3.1<br />
software from Caliper Life Sciences (Hopkinton, MA, USA). The threshold w<strong>as</strong> set using the<br />
background scan made from each mouse be<strong>for</strong>e administration of the IRDye800CW®-labeled<br />
<strong>for</strong>mulations. The excitation wavelength w<strong>as</strong> set at 710 nm and emitted light w<strong>as</strong> me<strong>as</strong>ured at<br />
760; 780; 800; 820 and 840 nm. Spectral unmixing w<strong>as</strong> per<strong>for</strong>med to decompose the emitted<br />
light into auto fluorescence and label-specific fluorescence. To compare the fluorescence in<br />
different mice and different groups, the absolute fluorescence w<strong>as</strong> converted to relative<br />
fluorescence (% of the maximal fluorescence in the n<strong>as</strong>al cavity). The are<strong>as</strong> under the curve<br />
(AUC) of the relative fluorescence of individual mice were used to compare the fluorescence<br />
over time <strong>for</strong> the different groups.<br />
Immunostaining of TMC-WIV <strong>for</strong>mulations in n<strong>as</strong>al cross-sections. Immunostaining of<br />
TMC-WIV <strong>for</strong>mulations w<strong>as</strong> carried out essentially <strong>as</strong> described be<strong>for</strong>e [18]. Six to eight weeks<br />
old female C57-BL/6 mice from Charles River (L’Arbresle, France) were i.n. vaccinated with<br />
TMC-RA-WIV or TMC-WIV in a (w/w) ratio of 1:1 or plain WIV <strong>for</strong>mulations. As negative<br />
controls, mice received PBS or solutions of TMCs and 4 mice were left untreated. After 20<br />
minutes or 1 hour, animals were sacrificed by cervical dislocation and the n<strong>as</strong>al cavity w<strong>as</strong><br />
isolated by removing the brains, lower jaw, skin and muscle tissue. The n<strong>as</strong>al cavities were<br />
96
Influence of N-Acetylation on the <strong>Adjuvant</strong>icity of TMC<br />
then fixed in 10% <strong>for</strong>malin, decalcified, embedded in paraffin and 3.5 µm thick cross-sections<br />
were made at different depths of the n<strong>as</strong>al cavity using a Microm HM 355 S microtome from<br />
Thermo Fisher Scientific (Walldorf, Germany). Cross-sections <strong>for</strong> immunohistochemistry were<br />
mounted on superfrost plus gl<strong>as</strong>s slides from Menzel-Gläser (Braunschweig, Germany). After<br />
deparaffinization, hydration and a heat-induced epitope retrieval (HIER) step with citrate<br />
buffer, cross-sections were w<strong>as</strong>hed 3 x 5 min with 0.2% Triton-X in PBS. Subsequently, crosssections<br />
were incubated first with normal goat serum <strong>for</strong> 15 minutes and then with a 1:500<br />
dilution of polyclonal WIV-specific rabbit antiserum overnight at 4°C. As a negative control,<br />
one of the two paraffin sections w<strong>as</strong> incubated with nonspecific rabbit serum <strong>as</strong> primary<br />
antibody. After w<strong>as</strong>hing the slides 3x <strong>for</strong> 5 min with 0.2% Triton-X in PBS, a 1:200 dilution of<br />
Alexa Fluor® 488-labeled goat anti-rabbit IgG w<strong>as</strong> applied. The samples were incubated <strong>for</strong> 45<br />
minutes at room temperature. Next, the slides were w<strong>as</strong>hed 3x with PBS and stained with a<br />
1:25000 dilution of DAPI <strong>for</strong> 6 min at room temperature and w<strong>as</strong>hed 3x with PBS again.<br />
Finally, the slides were mounted with Fluorosave from Calbiochem (San Diego, CA, USA). All<br />
slides were examined using a fluorescence microscope (Nikon Eclipse TE-2000 Nikon,<br />
Amstelveen, the Netherlands) equipped with a Digital Sight DS-2Mv camera. Slides were<br />
blindly scored <strong>for</strong> the presence, location, pattern and intensity of antigen staining and pictures<br />
were taken with fixed camera settings <strong>for</strong> a fair comparison.<br />
Uptake by Calu-3 epithelial cells of fluorescently labeled TMC-WIV <strong>for</strong>mulations. Calu-<br />
3 were grown in MEM with 0.292g/L L Glutamine, 1g /L Glucose, 2.2g/L NaHCO3 (Invitrogen,<br />
Breda, The Netherlands) supplemented with antibiotics/antimycotics and 10% heatinactivated<br />
fetal calf serum (FCS; Integro, Zaandam, The Netherlands). Cells were cultured in<br />
fibronectin-coated fl<strong>as</strong>ks and maintained at 37°C in a 5% CO2 humidified air atmosphere and<br />
split once a week.<br />
Cells were transferred into a flat bottom 24-well plate (500,000 cells/well) and after<br />
maturation <strong>for</strong> 2 days cells were incubated with 500 μl of plain AlexaFluor-488®-labeled WIV<br />
or <strong>for</strong>mulated with TMC or TMC-RA in PBS at a TMC:WIV (w/w) ratio of 5:1 and diluted <strong>as</strong><br />
indicated (50µg WIV/ml diluted 10-160x) <strong>for</strong> 1 h at 37 ºC or 4 ºC. After incubation, cells were<br />
w<strong>as</strong>hed with PBS (400 μl) and detached with 100 μl trypsin/EDTA at 37 ºC. Then, 150 μl MEM<br />
+ 10% FCS w<strong>as</strong> added and the cells were transferred into a 96 well plate. Cells were<br />
centrifuged (250 x g <strong>for</strong> 5 minutes at 4 ºC) and w<strong>as</strong>hed with three times with 200 μl phosphate<br />
buffered albumin (PBA, 1 g albumin per 100 ml PBS) and finally resuspended in 200 μl PBA.<br />
Immediately prior to me<strong>as</strong>urement, 20 μl propidium iodide solution (PI; 1 μg/ml in water) w<strong>as</strong><br />
97
Chapter 5A<br />
added <strong>for</strong> live/dead cell discrimination. Flow cytometric analysis w<strong>as</strong> per<strong>for</strong>med on a FACS<br />
Canto (Becton and Dickinson, Mountain View, CA, USA) using a 15 mW 488 nm, air-cooled<br />
argon-ion l<strong>as</strong>er and data were analyzed using FACS Diva software (Becton and Dickinson,<br />
Mountain View, CA, USA). 10,000 cells were recorded per sample to determine bead-uptake<br />
(FL1-channel) and PI-staining (FL3-channel).<br />
Transport of fluorescently labeled TMC-WIV <strong>for</strong>mulations over Calu-3 monolayer.<br />
Calu-3 cells were seeded at a density of 5x10 5 cells per well on 12-transwell plates with<br />
Matrigel ® coated 2 µm microporous membranes. The cells were cultured in Dulbecco’s<br />
Modified Eagle’s Medium (DMEM) <strong>for</strong> 14 days until a confluent cell layer w<strong>as</strong> <strong>for</strong>med. The<br />
medium w<strong>as</strong> replaced by Hank’s Balanced Salt Solution (HBSS) at the apical (300 μl) and<br />
b<strong>as</strong>olateral (1.2 ml) sides and after 30 minutes equilibration the transepithelial electrical<br />
resistance (TEER) w<strong>as</strong> me<strong>as</strong>ured using a home made dipstick electrode. Then, HBSS w<strong>as</strong><br />
replaced by 300 μl of plain AlexaFluor-488®-labeled WIV or <strong>for</strong>mulated with TMC or TMC-RA<br />
at a TMC:WIV (w/w) ratio of 5:1 (25µg WIV/ml in HBSS). After incubation <strong>for</strong> 60 minutes at 37<br />
ºC the TEER w<strong>as</strong> me<strong>as</strong>ured again and the amount of fluorescently labeled WIV particles in the<br />
b<strong>as</strong>olateral acceptor compartment w<strong>as</strong> determined by flow cytometry.<br />
Maturation of murine bone marrow derived dendritic cells. The femur and tibia of C57-<br />
BL/6 mice were removed, both ends were cut and bone marrow w<strong>as</strong> flushed with Iscove’s<br />
Modified Dulbecco’s Medium (IMDM; Gibco, CA, USA) using a syringe with a 0.45 mm diameter<br />
needle. The bone marrow suspension w<strong>as</strong> vigorously resuspended and p<strong>as</strong>sed over a 100 μm<br />
gauge to obtain a single cell suspension. After w<strong>as</strong>hing, cells were seeded 2x10 6 cells per 100<br />
mm petridish (Greiner Bio-One, Alphen aan den Rijn, The Netherlands) in 10 ml IMDM,<br />
supplemented with 2 mM L-glutamine, 50 U/ml penicillin, 50 ug/ml streptomycin and 50 μM<br />
β-mercaptoethanol (Merck, Darmstadt, Germany) and 30 ng/ml recombinant murine GM-CSF<br />
(rmGM-CSF). At day 2, 10 ml medium containing 30 ng/ml rmGM-CSF w<strong>as</strong> added. At day 5<br />
another 30 ng/ml rmGM-CSF w<strong>as</strong> added to each plate. From day 6 onwards, the non-adherent<br />
DCs were harvested and used <strong>for</strong> subsequent experiments.<br />
These immature DCs were seeded (50,000 cells/well) in a 96-well plate in a total volume of<br />
100 μl RPMI 10% FCS in the presence of TMC (0.001-25 μg/ml), TMC-WIV <strong>for</strong>mulations at a<br />
1:1 or 5:1 (w/w) ratios or lipopolysaccharide (LPS) (0.01-100 ng/ml) <strong>as</strong> a positive control. DCs<br />
were incubated with the samples <strong>for</strong> 16 h at 37 °C. DC maturation w<strong>as</strong> determined by<br />
98
Influence of N-Acetylation on the <strong>Adjuvant</strong>icity of TMC<br />
analyzing cell-surface expression of co-stimulatory molecules (CD40) on CD11 and MHC II<br />
positive cells by flow cytometry [19].<br />
Results<br />
Properties of TMC-WIV <strong>for</strong>mulations. The structural characteristics of the synthesized<br />
TMCs are summarized in Table 1.<br />
Table 1. Structural characteristics of TMC and TMC-RA used in this study.<br />
Polymer M n (kDa) M w (kDa) DQ (%) DAc (%)<br />
TMC 36 75 43 17<br />
TMC-RA 43 83 44 54<br />
As in a previous study [16], WIV w<strong>as</strong> mixed in a 1:1 or 5:1 w/w ratio with TMC or TMC-RA<br />
and analyzed <strong>for</strong> size, PDI and zeta potential. Figure 1 shows that mixing of WIV with TMC led<br />
to a minor incre<strong>as</strong>e in size and PDI, and the zeta potential reversed from -17 mV (plain WIV) to<br />
about +18 mV, independent of type or amount of TMC used <strong>for</strong> coating. It can there<strong>for</strong>e be<br />
stated that all <strong>for</strong>mulations were physically similar, except <strong>for</strong> the amount of free polymer;<br />
<strong>for</strong>mulations with a 5:1 w/w polymer to WIV ratio will very likely contain more free polymer<br />
[16].<br />
Diameter (nm)<br />
A<br />
400<br />
300<br />
200<br />
100<br />
PDI<br />
Diameter<br />
1.0<br />
0.8<br />
0.6<br />
0.4<br />
0.2<br />
PDI<br />
Zeta potential (mV)<br />
B<br />
25<br />
15<br />
5<br />
-5<br />
-15<br />
0<br />
WIV<br />
TMC-WIV 1:1<br />
TMC-WIV 5:1<br />
TMC-RA-WIV 1:1<br />
TMC-RA-WIV 5:1<br />
0.0<br />
-25<br />
WIV<br />
TMC-WIV 1:1<br />
TMC-WIV 5:1<br />
TMC-RA-WIV 1:1<br />
TMC-RA-WIV 5:1<br />
Figure 1. Z-average diameter and PDI (A) and zeta potential (B) of the WIV <strong>for</strong>mulations. Error bars<br />
represent the standard deviation (n=5).<br />
99
Chapter 5A<br />
Biodegradation. An important difference between TMC and TMC-RA is their enzymatic<br />
biodegradability by lysozyme [15], an enzyme which is present in the n<strong>as</strong>al cavity [20].<br />
There<strong>for</strong>e, after i.n. immunization TMC-RA-WIV may be more rapidly degraded in the n<strong>as</strong>al<br />
cavity. To study whether TMC and TMC-RA are also degraded upon contact with enzymes<br />
present in the n<strong>as</strong>al cavity, solutions of these TMCs were incubated at 37 °C with murine n<strong>as</strong>al<br />
w<strong>as</strong>hes. At different time points, the solutions were analyzed by GPC to determine the<br />
molecular weight (M n and M w) of the TMCs. As expected, TMCs in PBS that were taken <strong>as</strong><br />
controls, did not show decre<strong>as</strong>e in M n/M w. As shown in Figure 2, TMC-RA degraded to a greater<br />
extent than TMC in a n<strong>as</strong>al w<strong>as</strong>h. However, the degradation of both TMC-RA and TMC w<strong>as</strong><br />
relatively slow in the n<strong>as</strong>al w<strong>as</strong>h (compared to the in vitro findings) and up to 24 h of<br />
incubation, no significant differences were observed between the two TMCs.<br />
Figure 2. Biodegradation of TMC (squares) and TMC-RA (dots) in pooled n<strong>as</strong>al w<strong>as</strong>h of mice. Error bars<br />
indicate the standard deviation (n=3).<br />
In vivo fluorescence imaging. In vivo fluorescence imaging w<strong>as</strong> used to compare the effect<br />
of TMC and TMC-RA on the n<strong>as</strong>al clearance of WIV. Fluorescently labeled WIV w<strong>as</strong> <strong>for</strong>mulated<br />
with TMC and TMC-RA and the <strong>for</strong>mulations were administered i.n. to mice. Imaging of the<br />
fluorescence in the n<strong>as</strong>al cavity <strong>as</strong> a function of time did not reveal a significant difference<br />
between the <strong>for</strong>mulations (p>0.05), <strong>as</strong> calculated from the AUC (see Figure 3). After an initial<br />
incre<strong>as</strong>e in fluorescence, likely due to fluorescence dequenching [18], all <strong>for</strong>mulations showed<br />
a comparable decre<strong>as</strong>e in fluorescence over time. This suggests that the n<strong>as</strong>al clearance of<br />
plain WIV, TMC-WIV and TMC-RA-WIV is comparable.<br />
100
Influence of N-Acetylation on the <strong>Adjuvant</strong>icity of TMC<br />
Relative fluorescence<br />
(%)<br />
110<br />
100<br />
90<br />
80<br />
70<br />
60<br />
50<br />
40<br />
30<br />
20<br />
10<br />
0<br />
0 25 50 75 100 125<br />
Time (min)<br />
WIV<br />
TMC-WIV<br />
TMC-RA-WIV<br />
Figure 3. Average relative fluorescence in the n<strong>as</strong>al cavity over time after intran<strong>as</strong>al administration of<br />
fluorescently labeled WIV <strong>for</strong>mulations. Error bars indicate standard deviation (n=3 mice).<br />
Local distribution in the n<strong>as</strong>al cavity of intran<strong>as</strong>ally applied vaccine <strong>for</strong>mulations. In a<br />
previous study it w<strong>as</strong> shown that in the murine n<strong>as</strong>al cavity, TMC-WIV <strong>for</strong>mulations exhibit a<br />
completely different distribution pattern <strong>as</strong> plain WIV: TMC coated WIV w<strong>as</strong> located tightly<br />
along the epithelial cells while plain WIV w<strong>as</strong> mainly observed in mucosal blobs [18]. To<br />
investigate whether the distribution pattern is different <strong>for</strong> TMC-RA-WIV, mice were<br />
administered TMC-RA-WIV and TMC-WIV and n<strong>as</strong>al sections were stained to visualize the<br />
location of WIV. After 20 min, WIV-specific staining w<strong>as</strong> found on the epithelial surfaces of the<br />
n<strong>as</strong>o- and maxilloturbinates in all mice, <strong>for</strong> both TMC-WIV and TMC-RA-WIV (see Figure 4 <strong>for</strong><br />
representative examples, n=5). This indicates that the degree of acetylation is unlikely to affect<br />
the contact between WIV with the mucosal surfaces. After 1 hour, WIV w<strong>as</strong> still present on the<br />
n<strong>as</strong>o- and maxilloturbinates but the staining w<strong>as</strong> less intense (results not shown). TMC<br />
solutions showed no fluorescence upon excitation nor did staining with non-specific rabbit<br />
serum [18].<br />
101
Chapter 5A<br />
A<br />
B<br />
Figure 4. Representative picture of fluorescently labeled WIV (depicted in green) on the mucosal<br />
surfaces of the n<strong>as</strong>oturbinates when <strong>for</strong>mulated with TMC-RA (A) and TMC (B) 20 minutes after<br />
administration.<br />
102
Influence of N-Acetylation on the <strong>Adjuvant</strong>icity of TMC<br />
A<br />
Fluorescence Geo Mean<br />
(arb. units)<br />
2000<br />
1500<br />
1000<br />
500<br />
0<br />
5.00<br />
*<br />
2.50<br />
*<br />
1.25<br />
*<br />
0.62<br />
WIV<br />
TMC-WIV<br />
TMC-RA-WIV<br />
PBS<br />
** **<br />
0.31<br />
B<br />
Fluorescence Geo Mean<br />
(arb. units)<br />
1250<br />
1000<br />
750<br />
500<br />
250<br />
0<br />
**<br />
**<br />
5.00<br />
** ***<br />
2.50<br />
** ** ***<br />
***<br />
1.25<br />
WIV<br />
TMC-WIV<br />
TMC-RA-WIV<br />
PBS<br />
0.62<br />
0.31<br />
Conc. WIV (μg/ml)<br />
Conc. WIV (μg/ml)<br />
Figure 5. Uptake/<strong>as</strong>sociation of fluorescently labeled WIV, uncoated or coated with TMC or TMC-RA by<br />
Calu-3 cells when incubated with solution of different concentrations <strong>for</strong> 1 hour at 37°C (A) or 4°C (B). *<br />
p
Chapter 5A<br />
Change in TEER:<br />
% of WIV transported<br />
1.0×10 -2 -1% (±9) -26% (±7) -7% (±4)<br />
7.5×10 -3<br />
5.0×10 -3<br />
2.5×10 -3<br />
0<br />
oo<br />
***<br />
WIV<br />
TMC-WIV<br />
***<br />
TMC-RA-WIV<br />
Figure 6. Transport of fluorescently labeled WIV, uncoated or coated with TMC or TMC-RA, by Calu-3<br />
human bronchial epithelial cells after incubation at 37°C <strong>for</strong> 1 h. Changes in TEER values (± standard<br />
deviation) after 1 h incubation are depicted above the bars. Error bars indicate standard deviations<br />
(n=3). *** value significantly (p
Influence of N-Acetylation on the <strong>Adjuvant</strong>icity of TMC<br />
CD40 exp. (MFI)<br />
500<br />
400<br />
300<br />
200<br />
100<br />
medium<br />
0<br />
WIV<br />
TMC-WIV 5:1<br />
TMC-RA-WIV 5:1<br />
LPS 100 ng/ml<br />
TMC 25 μg/ml<br />
TMC-RA 25 μg/ml<br />
0.63μg/ml WIV<br />
0.31<br />
0.16<br />
0.08<br />
0.04<br />
Controls<br />
Figure 7. A representative example of DC maturation (n=2) (monitored by CD40 expression) by plain<br />
WIV, TMC-WIV and TMC-RA-WIV <strong>for</strong>mulations at 5:1 TMC:WIV (w/w) ratio in different concentrations.<br />
Discussion<br />
As pointed out be<strong>for</strong>e, adjuvants <strong>for</strong> n<strong>as</strong>al immunization can act by enhancing the<br />
immunoavailability of the antigen, or <strong>as</strong> immune potentiator by providing or incre<strong>as</strong>ing a<br />
“danger” signal [1, 2, 5, 6]. In general, TMC is <strong>as</strong>sociated with improving antigen delivery<br />
through its muco-adhesiveness [6, 13, 21, 22], thereby incre<strong>as</strong>ing n<strong>as</strong>al residence time and/or<br />
improving antigen-epithelial barrier interactions [18]. Although of importance <strong>for</strong> mucosal<br />
peptide and protein delivery [6, 12], TMC’s capability to open tight junctions seems not to be of<br />
major relevance <strong>for</strong> n<strong>as</strong>al vaccination <strong>as</strong> several studies showed successful n<strong>as</strong>al immunization<br />
with TMCs with low DQ (
Chapter 5A<br />
In the present study, we found that TMC-RA is degraded to a larger extent in n<strong>as</strong>al w<strong>as</strong>h than<br />
TMC, although this degradation appeared to be rather slow. However, note that the<br />
biodegradability of TMC in a n<strong>as</strong>al w<strong>as</strong>h may not accurately reflect the degradation in the n<strong>as</strong>al<br />
cavity after i.n. administration. It is likely that the concentration of lysozyme and other<br />
prote<strong>as</strong>es in a n<strong>as</strong>al w<strong>as</strong>h is much lower than the concentration in the n<strong>as</strong>al cavity, firstly<br />
because of a strong dilution effect in the w<strong>as</strong>h and secondly because most of the lysozyme is<br />
stored and excreted from submucosal glands [20] and may not be efficiently extracted by a<br />
n<strong>as</strong>al w<strong>as</strong>h. Thus, although in vitro degradation occurred quite slowly, the intran<strong>as</strong>al<br />
degradation of TMC-RA during its n<strong>as</strong>al residence may be significant in vivo. Our results<br />
consequently suggest that TMC-RA is enzymatically degraded more extensively than TMC.<br />
However, both the extent of degradation and the degradation rate in the n<strong>as</strong>al cavity are<br />
difficult to establish quantitatively.<br />
This reduction of polymer molecular weight may lead to a decre<strong>as</strong>e of cytotoxicity [24],<br />
however, TMC-RA h<strong>as</strong> already a very low toxicity profile showing no reduction of cell viability<br />
or incre<strong>as</strong>ed lactate dehydrogen<strong>as</strong>e (LDH) rele<strong>as</strong>e in Caco-2 cells even up to a concentration of<br />
10 mg/ml [15]. Furthermore, both in vivo WIV n<strong>as</strong>al clearance and local n<strong>as</strong>al distribution<br />
patterns are similar <strong>for</strong> TMC and TMC-RA coated WIV, indicating that possible n<strong>as</strong>al<br />
degradation of TMC-RA does not affect antigen exposure to the epithelial barrier. There<strong>for</strong>e,<br />
although n<strong>as</strong>al degradation of TMC-RA in vivo may occur, its effects on the exposure of the<br />
antigen to the epithelial barrier are likely rather small.<br />
Importantly, our data indicate that enhancing the interaction between the antigen and the<br />
epithelial barrier alone is not sufficient <strong>for</strong> an adequate immune response and consequently<br />
other factors have to play a role <strong>as</strong> well. One possibility is that a co-stimulatory ‘danger’ signal<br />
due to toxicity of the TMC is crucial in provoking an adequate immune response [25]. As TMC-<br />
RA showed the lowest in vitro cyto-toxicity of all TMC polymers [15] this polymer may not be<br />
capable of inducing such a danger signal. Interestingly, this difference in toxicity between TMC<br />
and TMC-RA may become even more pronounced due to the higher extent of n<strong>as</strong>al degradation<br />
of TMC-RA.<br />
The uptake and binding of TMC coated WIV by Calu-3 cells decre<strong>as</strong>ed compared to TMC-RA<br />
coated WIV and uncoated WIV. In contr<strong>as</strong>t, in other cell-types (e.g. 16-HBE14o and HeLa cells),<br />
TMC coated WIV w<strong>as</strong> taken up to a higher extent than plain WIV and TMC-RA coated WIV<br />
(results not shown). This implies that the extent of uptake of TMC coated WIV is highly<br />
dependent on the model cell line used. Calu-3 cells are, although derived from human<br />
bronchial epithelium, used extensively <strong>for</strong> ‘n<strong>as</strong>al’ transepithelial electrical resistance (TEER)<br />
106
Influence of N-Acetylation on the <strong>Adjuvant</strong>icity of TMC<br />
studies [26, 27] <strong>as</strong> they <strong>for</strong>m tight monolayers and also secrete mucus [28]. Several<br />
investigation have shown incre<strong>as</strong>ed transport of proteins or peptides through Caco-2 and Calu-<br />
3 monolayers by chitosan and TMC [29, 30], however the transport and uptake by Calu-3 cells<br />
of virus particles h<strong>as</strong> not been studied. Strong interactions with mucus (which is w<strong>as</strong>hed away<br />
after incubation) may be a re<strong>as</strong>on <strong>for</strong> the observed reduced uptake and binding of TMC-WIV by<br />
these cells <strong>as</strong> compared to WIV [31]. Interestingly, this w<strong>as</strong> not observed with the TMC-RA<br />
coated WIV, despite its similar zeta potential compared to TMC-WIV. Perhaps due to steric<br />
hindrance, the electrostatic interactions between TMC-RA and WIV are weaker than those<br />
between TMC and WIV, resulting in better <strong>as</strong>sociation properties <strong>for</strong> WIV in these mucusproducing<br />
cells due to loss of the TMC-RA coating. This suggests that the intran<strong>as</strong>al stability in<br />
the presence of mucus of the TMC-RA-WIV <strong>for</strong>mulation may be inferior compared to the TMC-<br />
WIV particle. This inferior stability <strong>for</strong> TMC-RA-WIV particles w<strong>as</strong> also observed in saline<br />
conditions [32] and may result in decre<strong>as</strong>ed protection against proteolytic enzymes.<br />
Additionally, the results obtained by studying the transport of WIV over the Calu-3<br />
monolayer (WIV>>TMC-RA-WIV>TMC-WIV) support the <strong>as</strong>sumption that WIV particles are<br />
too large to be transported through epithelial tight junctions, since the only <strong>for</strong>mulation that<br />
significantly decre<strong>as</strong>ed the TEER values (TMC-WIV) showed the lowest transport rate (Figure<br />
6). As the negatively charged uncoated WIV w<strong>as</strong> transported to the largest extent, interactions<br />
between the coated particles and mucus may play a key role: the positively charged TMC- and<br />
TMC-RA-WIV particles likely adhere to the negatively charged mucus and are there<strong>for</strong>e unable<br />
to p<strong>as</strong>s through the cell monolayer. Importantly, <strong>as</strong> in vivo i.n. immunization h<strong>as</strong> shown that<br />
TMC greatly enhances antigen immunogenicity, most likely by improved delivery [6, 22], our<br />
results imply that the uptake by and transport through n<strong>as</strong>al epithelial cells is not the main<br />
route <strong>for</strong> improved antigen delivery by TMC. This suggests that the uptake and transport of<br />
TMC b<strong>as</strong>ed particulate systems is mediated through M-cells [1, 3, 6] or direct DC sampling [2].<br />
Interestingly, influenza viruses and influenza derivated virosomes are known to specifically<br />
interact with M-cells [33] most likely due to recognition of sialic acid and galactose residues by<br />
lectin receptors [34]. Additionally, N-acetyl glucosamine moieties, which are more abundant in<br />
TMC-RA, are known to interact with several lectin receptors present on DCs and/or<br />
macrophages [35-38]. However, the effect of coating of WIV with TMC(-RA) on these<br />
interactions remains to be clarified. More advanced in vitro or ex vivo models [39] should be<br />
developed to further study the transport and uptake of WIV particulate systems over mucosal<br />
epithelial surfaces [40]. Possibly co-culturing Calu-3 cells with microfold (M)-cells may<br />
107
Chapter 5A<br />
improve the predictability of these models [21] <strong>for</strong> in vivo applications, <strong>as</strong> w<strong>as</strong> observed <strong>for</strong> a<br />
Caco-2/M-cell co-culture model [41].<br />
Recently it w<strong>as</strong> shown that TMC, both in particulate <strong>for</strong>m and in solution, can exert<br />
significant immunostimulatory effects on human monocyte derived DCs [41-43]. The<br />
mechanism, however, remains unclear although suggestions were made about residual N-<br />
acetylated glucosamine units capable of interacting with C-type lectin receptors present on<br />
DCs [36, 43]. However, since the differences in adjuvanticity between TMC and TMC-RA coated<br />
WIV were observed in mice, it is more appropriate to investigate the effect of TMC and TMC-RA<br />
coated WIV on the maturation of murine DCs. Since TMC-RA contains a higher number of N-<br />
acetyl glucosamine units, an altered interaction with DCs could be expected [36-38, 44]. Our<br />
data, however, suggest that WIV is the main inducer of DC maturation and that there is no<br />
additive effect of TMC or TMC-RA, or any effect of plain TMC or TMC-RA (Figure 7).<br />
Importantly, murine DCs express different C-type lectin receptors on their surface than human<br />
DCs and there<strong>for</strong>e adjuvants may have a different effect on murine or human DC-subsets [19].<br />
Further investigations on human DCs with these <strong>for</strong>mulations are there<strong>for</strong>e appropriate to<br />
clarify these findings.<br />
In summary, we found that TMC-RA w<strong>as</strong> degraded to a higher extent in n<strong>as</strong>al w<strong>as</strong>hes than<br />
TMC. In vivo fluorescence imaging did not reveal a significant difference in n<strong>as</strong>al clearance of<br />
TMC-WIV and TMC-RA-WIV. In n<strong>as</strong>al cross-sections both TMC-WIV <strong>for</strong>mulations were<br />
similarly distributed in the n<strong>as</strong>al cavity, indicating that enhancing the interaction between the<br />
antigen and the epithelial barrier alone is not sufficient <strong>for</strong> improving the immune response.<br />
The uptake and binding of fluorescent WIV in Calu-3 cells w<strong>as</strong> less pronounced when coated<br />
with TMC than with TMC-RA or uncoated and both TMCs significantly decre<strong>as</strong>e the transport<br />
of WIV over a Calu-3 monolayer. Furthermore, experiments with murine BM-DCs indicated<br />
that WIV is the main inducer of DC maturation, me<strong>as</strong>ured <strong>as</strong> CD40 expression, and that TMC<br />
and TMC-RA do not have an additive effect. Our data suggest that the loss of adjuvanticity by<br />
reacetylation of TMC might be due to the lower stability of TMC-RA-WIV in the n<strong>as</strong>al cavity,<br />
rather than by any of the other factors studied in this paper.<br />
Acknowledgement. This research w<strong>as</strong> partially per<strong>for</strong>med under the framework of <strong>TI</strong><br />
<strong>Pharma</strong> project number D5-106-1; Vaccine delivery: alternatives <strong>for</strong> conventional multiple<br />
injection vaccines. E. Kaijzel and C. Löwik are acknowledged <strong>for</strong> their help with the n<strong>as</strong>al<br />
residence time experiment.<br />
108
Influence of N-Acetylation on the <strong>Adjuvant</strong>icity of TMC<br />
References<br />
1. S. Chadwick, C. Kriegel, and M. Amiji. Nanotechnology solutions <strong>for</strong> mucosal immunization.<br />
Adv Drug Deliv Rev 62: 394-407 (2010).<br />
2. N. Csaba, M. Garcia-Fuentes, and M. J. Alonso. Nanoparticles <strong>for</strong> n<strong>as</strong>al vaccination. Adv Drug<br />
Deliv Rev 61: 140-157 (2009).<br />
3. B. Slütter, N. Hagenaars, and W. Jiskoot. Rational design of n<strong>as</strong>al vaccines. J Drug Target 16: 1-17<br />
(2008).<br />
4. N. Hagenaars, E. M<strong>as</strong>trobattista, H. Glansbeek, J. Heldens, H. van den Bosch, V. Schijns, D.<br />
Betbeder, H. Vromans, and W. Jiskoot. Head-to-head comparison of four nonadjuvanted<br />
inactivated cell culture-derived influenza vaccines: Effect of composition, spatial organization<br />
and immunization route on the immunogenicity in a murine challenge model. Vaccine 26: 6555-<br />
6563 (2008).<br />
5. N. Mishra, A. K. Goyal, S. Tiwari, R. Paliwal, S. R. Paliwal, B. Vaidya, S. Mangal, M. Gupta, D.<br />
Dube, A. Mehta, and S. P. Vy<strong>as</strong>. Recent advances in mucosal delivery of vaccines: Role of<br />
mucoadhesive/biodegradable polymeric carriers. Exp Opin Ther Patents 20: 661-679 (2010).<br />
6. M. Amidi, E. M<strong>as</strong>trobattista, W. Jiskoot, and W. E. Hennink. Chitosan-b<strong>as</strong>ed delivery systems<br />
<strong>for</strong> protein therapeutics and antigens. Adv Drug Deliv Rev 62: 59-82 (2010).<br />
7. H. C. Arca, M. Günbeyaz, and S. Şenel. Chitosan-b<strong>as</strong>ed systems <strong>for</strong> the delivery of vaccine<br />
antigens. Exp Rev Vaccines 8: 937-953 (2009).<br />
8. L. Illum, I. Jabbal-Gill, M. Hinchcliffe, A. N. Fisher, and S. S. Davis. Chitosan <strong>as</strong> a novel n<strong>as</strong>al<br />
delivery system <strong>for</strong> vaccines. Adv Drug Deliv Rev 51: 81-96 (2001).<br />
9. D. Snyman, J. H. Hamman, and A. F. Kotze. Evaluation of the mucoadhesive properties of N-<br />
trimethyl chitosan chloride. Drug Develop Ind Pharm 29: 61-69 (2003).<br />
10. G. Di Colo, S. Burgal<strong>as</strong>si, Y. Zambito, D. Monti, and P. Chetoni. Effects of different N-<br />
trimethyl chitosans on in vitro/in vivo ofloxacin transcorneal permeation. J Pharm Sci 93: 2851-<br />
2862 (2004).<br />
11. A. F. Kotze, H. L. Luessen, B. J. De Leeuw, B. G. De Boer, J. C. Verhoef, and H. E. Junginger.<br />
N-<strong>Trimethyl</strong> chitosan chloride <strong>as</strong> a potential absorption enhancer across mucosal surfaces: In<br />
vitro evaluation in intestinal epithelial cells (Caco-2). Pharm Res 14: 1197-1202 (1997).<br />
12. G. Sandri, S. Rossi, M. C. Bonferoni, F. Ferrari, Y. Zambito, G. Di Colo, and C. Caramella.<br />
Buccal penetration enhancement properties of N-trimethyl chitosan: Influence of quaternization<br />
degree on absorption of a high molecular weight molecule. Int J Pharm 297: 146-155 (2005).<br />
13. N. Hagenaars, E. M<strong>as</strong>trobattista, R. J. Verheul, I. Mooren, H. L. Glansbeek, J. G. M. Heldens,<br />
H. Van Den Bosch, and W. Jiskoot. Physicochemical and immunological characterization of<br />
N,N,N-trimethyl chitosan-coated whole inactivated influenza virus vaccine <strong>for</strong> intran<strong>as</strong>al<br />
administration. Pharm Res 26: 1353-1364 (2009).<br />
14. R. J. Verheul, M. Amidi, S. van der Wal, E. van Riet, W. Jiskoot, and W. E. Hennink. Synthesis,<br />
characterization and in vitro biological properties of O-methyl free N,N,N-trimethylated<br />
chitosan. Biomaterials 29: 3642-3649 (2008).<br />
15. R. J. Verheul, M. Amidi, M. J. van Steenbergen, E. van Riet, W. Jiskoot, and W. E. Hennink.<br />
Influence of the degree of acetylation on the enzymatic degradation and in vitro biological<br />
properties of trimethylated chitosans. Biomaterials 30: 3129-3135 (2009).<br />
16. N. Hagenaars, R. J. Verheul, I. Mooren, P. H. J. L. F. de Jong, E. M<strong>as</strong>trobattista, H. L.<br />
Glansbeek, J. G. M. Heldens, H. van den Bosch, W. E. Hennink, and W. Jiskoot. Relationship<br />
between structure and adjuvanticity of N,N,N-trimethyl chitosan (TMC) structural variants in a<br />
n<strong>as</strong>al influenza vaccine. J Control Rele<strong>as</strong>e 140: 126-133 (2009).<br />
17. X. Jiang, A. Van Der Horst, M. J. Van Steenbergen, N. Akeroyd, C. F. Van Nostrum, P. J.<br />
Schoenmakers, and W. E. Hennink. Molar-m<strong>as</strong>s characterization of cationic polymers <strong>for</strong> gene<br />
delivery by aqueous size-exclusion chromatography. Pharm Res 23: 595-603 (2006).<br />
18. N. Hagenaars, M. Mania, P. de Jong, I. Que, R. Nieuwland, B. Slütter, H. Glansbeek, J. Heldens,<br />
H. van den Bosch, C. Löwik, E. Kaijzel, E. M<strong>as</strong>trobattista, and W. Jiskoot. Role of<br />
109
Chapter 5A<br />
trimethylated chitosan (TMC) in n<strong>as</strong>al residence time, local distribution and toxicity of an<br />
intran<strong>as</strong>al influenza vaccine. J Control Rele<strong>as</strong>e 144: 17-24 (2010).<br />
19. S. K. Singh, J. Stephani, M. Schaefer, H. Kalay, J. J. García-Vallejo, J. den Haan, E. Saeland, T.<br />
Sparw<strong>as</strong>ser, and Y. van Kooyk. Targeting glycan modified OVA to murine DC-SIGN<br />
transgenic dendritic cells enhances MHC cl<strong>as</strong>s I and II presentation. Mol Immunol 47: 164-174<br />
(2009).<br />
20. A. M. Cole, H. I. Liao, O. Stuchlik, J. Tilan, J. Pohl, and T. Ganz. Cationic polypeptides are<br />
required <strong>for</strong> antibacterial activity of human airway fluid. J Immunol 169: 6985-6991 (2002).<br />
21. M. Amidi, S. G. Romeijn, J. C. Verhoef, H. E. Junginger, L. Bungener, A. Huckriede, D. J. A.<br />
Crommelin, and W. Jiskoot. N-<strong>Trimethyl</strong> chitosan (TMC) nanoparticles loaded with influenza<br />
subunit antigen <strong>for</strong> intran<strong>as</strong>al vaccination: Biological properties and immunogenicity in a mouse<br />
model. Vaccine 25: 144-153 (2007).<br />
22. S. M. Van Der Merwe, J. C. Verhoef, J. H. M. Verheijden, A. F. Kotze, and H. E. Junginger.<br />
<strong>Trimethyl</strong>ated chitosan <strong>as</strong> polymeric absorption enhancer <strong>for</strong> improved peroral delivery of<br />
peptide drugs. Eur J Pharm Biopharm 58: 225-235 (2004).<br />
23. I. M. Van der Lubben, J. C. Verhoef, M. M. Fretz, O. Van, I. Mesu, G. Kersten, and H. E.<br />
Junginger. <strong>Trimethyl</strong> chitosan chloride (TMC) <strong>as</strong> a novel excipient <strong>for</strong> oral and n<strong>as</strong>al<br />
immunisation against diphtheria. S.T.P. <strong>Pharma</strong> Sciences 12: 235-242 (2002).<br />
24. S. Mao, X. Shuai, F. Unger, M. Wittmar, X. Xie, and T. Kissel. Synthesis, characterization and<br />
cytotoxicity of poly(ethylene glycol)-graft-trimethyl chitosan block copolymers. Biomaterials 26:<br />
6343-6356 (2005).<br />
25. V. E. Schijns. Immunological concepts of vaccine adjuvant activity. Curr Opin Immunol 12: 456-<br />
463 (2000).<br />
26. M. Amidi, S. G. Romeijn, G. Borchard, H. E. Junginger, W. E. Hennink, and W. Jiskoot.<br />
Preparation and characterization of protein-loaded N-trimethyl chitosan nanoparticles <strong>as</strong> n<strong>as</strong>al<br />
delivery system. J Control Rele<strong>as</strong>e 111: 107-116 (2006).<br />
27. D. Teijeiro-Osorio, C. Remuñán-López, and M. J. Alonso. New generation of hybrid<br />
poly/oligosaccharide nanoparticles <strong>as</strong> carriers <strong>for</strong> the n<strong>as</strong>al delivery of macromolecules.<br />
Biomacromolecules 10: 243-249 (2009).<br />
28. K. A. Foster, M. L. Avery, M. Yazdanian, and K. L. Audus. Characterization of the Calu-3 cell<br />
line <strong>as</strong> a tool to screen pulmonary drug delivery. Int J Pharm 208: 1-11 (2000).<br />
29. B. I. Florea, M. Thanou, H. E. Junginger, and G. Borchard. Enhancement of bronchial<br />
octreotide absorption by chitosan and N-trimethyl chitosan shows linear in vitro/in vivo<br />
correlation. J Control Rele<strong>as</strong>e 110: 353-361 (2006).<br />
30. C. Witschi and R. J. Mrsny. In vitro evaluation of microparticles and polymer gels <strong>for</strong> use <strong>as</strong><br />
n<strong>as</strong>al plat<strong>for</strong>ms <strong>for</strong> protein delivery. Pharm Res 16: 382-390 (1999).<br />
31. S. K. Lai, Y.-Y. Wang, and J. Hanes. Mucus-penetrating nanoparticles <strong>for</strong> drug and gene<br />
delivery to mucosal tissues. Adv Drug Deliv Rev 61: 158-171 (2009).<br />
32. N. Hagenaars. Towards an intran<strong>as</strong>al influenza vaccine - B<strong>as</strong>ed on whole inactivated influenza<br />
virus with N,N,N-trimethyl chito<strong>as</strong>n <strong>as</strong> adjuvant. Utrecht University, Utrecht (2010).<br />
33. Y. Fujimura, M. Takeda, H. Ikai, K. Haruma, T. Akisada, T. Harada, T. Sakai, and M. Ohuchi.<br />
The role of M cells of human n<strong>as</strong>opharyngeal lymphoid tissue in influenza virus sampling.<br />
Virchows Archiv 444: 36-42 (2004).<br />
34. G. N. Rogers and J. C. Paulson. Receptor determinants of human and animal influenza virus<br />
isolates: Differences in receptor specificity of the H3 hemagglutinin b<strong>as</strong>ed on species of origin.<br />
Virology 127: 361-373 (1983).<br />
35. K. Nishimura, S. Nishimura, and H. Seo. Macrophage activation with multi-porous beads<br />
prepared from partially deacetylated chitin. J Biomed Mat Res 20: 1359-1372 (1986).<br />
36. M. J. Robinson, D. Sancho, E. C. Slack, S. LeibundGut-Landmann, and C. R. Sousa. Myeloid C-<br />
type lectins in innate immunity. Nat Immunol 7: 1258-1265 (2006).<br />
37. J. Nadesalingam, A. W. Dodds, K. B. M. Reid, and N. Palaniyar. Mannose-binding lectin<br />
recognizes peptidoglycan via the N-acetyl glucosamine moiety, and inhibits ligand-induced<br />
110
Influence of N-Acetylation on the <strong>Adjuvant</strong>icity of TMC<br />
proinflammatory effect and promotes chemokine production by macrophages. J Immunol 175:<br />
1785-1794 (2005).<br />
38. P. Zhang, S. Snyder, P. Feng, P. Azadi, S. Zhang, S. Bulgheresi, K. E. Sanderson, J. He, J.<br />
Klena, and T. Chen. Role of N-acetylglucosamine within core lipopolysaccharide of several<br />
species of Gram-negative bacteria in targeting the DC-SIGN (CD209). J Immunol 177: 4002-<br />
4011 (2006).<br />
39. G. Sandri, P. Poggi, M. C. Bonferoni, S. Rossi, F. Ferrari, and C. Caramella. Histological<br />
evaluation of buccal penetration enhancement properties of chitosan and trimethyl chitosan. J<br />
Pharm <strong>Pharma</strong>col 58: 1327 (2006).<br />
40. B. Forbes and C. Ehrhardt. Human respiratory epithelial cell culture <strong>for</strong> drug delivery<br />
applications. Eur J Pharm Biopharm 60: 193-205 (2005).<br />
41. B. Slütter, L. Plapied, V. Fievez, M. Alonso Sande, A. des Rieux, Y. J. Schneider, E. Van Riet,<br />
W. Jiskoot, and V. Préat. Mechanistic study of the adjuvant effect of biodegradable<br />
nanoparticles in mucosal vaccination. J Control Rele<strong>as</strong>e 138: 113-121 (2009).<br />
42. S. M. Bal, B. Slütter, E. van Riet, A. C. Kruithof, Z. Ding, G. F. A. Kersten, W. Jiskoot, and J.<br />
A. Bouwstra. Efficient induction of immune responses through intradermal vaccination with N-<br />
trimethyl chitosan containing antigen <strong>for</strong>mulations. J Control Rele<strong>as</strong>e 142: 374-383 (2010).<br />
43. B. Slütter, P. C. Soema, Z. Ding, R. Verheul, W. Hennink, and W. Jiskoot. Conjugation of<br />
ovalbumin to trimethyl chitosan improves immunogenicity of the antigen. J Control Rele<strong>as</strong>e 143:<br />
207-214 (2010).<br />
44. K. Nishimura, C. Ishihara, and S. Ukei. Stimulation of cytokine production in mice using<br />
deacetylated chitin. Vaccine 4: 151-156 (1986).<br />
111
CHAPTER 5B<br />
MATURA<strong>TI</strong>ON OF HUMAN MONOCYTE<br />
DERIVED DENDRI<strong>TI</strong>C CELLS BY TRIMETHYL<br />
CHITOSAN IS CORRELATED WITH ITS N-<br />
ACETYL GLUCOSAMINE (GLCNAC) CONTENT<br />
Rolf J. Verheul, Niels Hagenaars, Thom<strong>as</strong> van Es, Sven Bruins,<br />
Bram Slütter, Wim E. Hennink, Wim Jiskoot.<br />
Manuscript submitted
Chapter 5B<br />
Abstract<br />
N-acetylated glucosamine units or GlcNAcs present in trimethyl chitosan (TMC) have been<br />
described to interact with several pathogen <strong>as</strong>sociated molecular pattern receptors involved in<br />
the human innate immune response. The aim of this study w<strong>as</strong> to compare the immunomodulatory<br />
effects of TMC with a low (17%) and high (54%, TMC-RA) degree of acetylation or<br />
GlcNAc content on the uptake and maturation of human monocyte derived dendritic cells (MO-<br />
DCs) in vitro, using whole inactivated influenza virus (WIV) <strong>as</strong> antigen. The GlcNAc content of<br />
TMC had no effect on the uptake of TMC(-RA) coated WIV by MO-DCs. Interestingly, TMC-RA,<br />
either <strong>as</strong> a solution or when <strong>for</strong>mulated with WIV, induced a much stronger DC maturation<br />
than any of the other <strong>for</strong>mulations, <strong>as</strong> judged from CD86 expression and cytokine rele<strong>as</strong>e (IL-<br />
10, TNF-α and IL-12p40 and IL-12p70). Since both IL-10 and IL-12p70 levels were elevated, no<br />
polarization towards a Th1 or Th2 type immune response could be established. In conclusion,<br />
<strong>as</strong> compared to TMC, TMC-RA h<strong>as</strong> strong immuno-stimulatory effects in vitro on human MO-<br />
DCs. This implies that the degree of N-acetylation or GlcNAc content may be critical <strong>for</strong> the<br />
adjuvant effect of TMC in humans.<br />
114
Maturation of Human DCs is Correlated with the GlcNAc Content<br />
Introduction<br />
Subunit or inactivated vaccines are generally safer but less effective than live attenuated<br />
vaccines and there<strong>for</strong>e require potent adjuvants to elicit adequate immune responses [1, 2]. As<br />
few adjuvants are approved <strong>for</strong> use in humans [3], there is a need <strong>for</strong> novel compounds. N,N,Ntrimethyl<br />
chitosan (TMC), a quaternized chitosan derivative, h<strong>as</strong> successfully been used <strong>for</strong><br />
mucosal and intradermal vaccination [4-7] and the structural properties of TMC like the<br />
degree of quaternization (DQ), the degree of O-methylation (DOM), and the degree of N-<br />
acetylation (DAc) can be varied during synthesis [8, 9]. In vitro evaluation showed that<br />
incre<strong>as</strong>ing the DAc or N-acetyl glucosamine (GlcNAc) content of TMC resulted in a decre<strong>as</strong>e in<br />
cell-toxicity and better enzymatic degradation by lysozyme [9]. GlcNAc moieties are, <strong>as</strong><br />
pathogen-<strong>as</strong>sociated molecular patterns (PAMPs), known to interact with C-type lectins [10],<br />
like DC-SIGN [11, 12], mannose receptors [13, 14], and toll-like receptor type 2 (TLR2) [15]<br />
present on the surface of several antigen presenting cells (APCs), such <strong>as</strong> DCs and<br />
macrophages [10]. However, after intran<strong>as</strong>al (i.n.) immunization with whole inactivated<br />
influenza virus (WIV) in mice, a striking loss in adjuvanticity of TMC w<strong>as</strong> observed when WIV<br />
w<strong>as</strong> <strong>for</strong>mulated with reacetylated TMC (TMC-RA; DAc of 54%), compared to all other tested<br />
TMCs with a low DAc (17%) [16]. Since no differences in the physico-chemical properties of<br />
the <strong>for</strong>mulations were observed, this lack of adjuvanticity had to be attributed to the higher<br />
GlcNAc content of TMC-RA, however the precise mechanism still remains to be elucidated.<br />
A recent study by our group showed no significant difference in activation of murine bone<br />
marrow derived DCs after exposure to regular TMC or TMC-RA (Chapter 5A). Interestingly,<br />
human DCs may be different from mouse DCs <strong>as</strong> murine and human DCs express different<br />
receptors on their surface [11, 17]. There<strong>for</strong>e, in this study we further explored the differences<br />
in immuno-stimulatory effect between reacetylated TMC-RA (GlcNAc content 54%) and<br />
conventional TMC (GlcNAc content 17%) using human monocyte derived DCs (MO-DCs).<br />
Materials and Methods<br />
Materials. Chitosan with a DAc of 17% (determined with 1 H-NMR <strong>as</strong> described in [9] and a<br />
number average molecular weight (M n) and weight average molecular weight (M w) of 28 and<br />
43 kDa, respectively, <strong>as</strong> determined by gel permeation chromatography (GPC) <strong>as</strong> described in<br />
[8], w<strong>as</strong> purch<strong>as</strong>ed from Primex (Siglufjordur, Iceland). Acetic anhydride, sodium borohydrate,<br />
<strong>for</strong>mic acid, <strong>for</strong>maldehyde 37% (stabilized with methanol), deuterium oxide, sodium acetate,<br />
115
Chapter 5B<br />
acetic acid (anhydrous), sodium hydroxide and hydrochloric acid were obtained from Sigma-<br />
Aldrich Chemical Co (Zwijndrecht, The Netherlands). Iodomethane 99% stabilized with copper<br />
w<strong>as</strong> obtained from Acros Organics (Geel, Belgium). Purified, cell culture-grown (Madin-Darby<br />
Canine Kidney (MDCK) cells, β-propiolacton (BPL)-inactivated influenza virus, strain<br />
A/PR/8/34 (WIV) w<strong>as</strong> from Nobilon International BV (Boxmeer, The Netherlands). Alexa<br />
Fluor® 488 labeling kit w<strong>as</strong> purch<strong>as</strong>ed from Invitrogen (Breda, the Netherlands).<br />
Trypsin/EDTA 10x, Plain DMEM (Dulbecco’s modification of Eagle’s medium, with 3.7 g/l<br />
sodium bicarbonate, 1 g/l L-glucose, L-glutamine) and antibiotics/antimycotics (penicillin,<br />
streptomycin sulphate, amphotericin B) were from PAA Laboratories GmbH (P<strong>as</strong>ching,<br />
Austria) and MEM with 0.292 g/L L-glutamine, 1 g/L glucose, 2.2 g/L NaHCO 3 w<strong>as</strong> from<br />
(Invitrogen, Breda, The Netherlands). All other chemicals used were of analytical grade.<br />
Synthesis and characterization of methylated chitosans. O-methyl free N,N,Ntrimethylated<br />
chitosans (TMCs) with a DQ of about 45% and varying DAc were synthesized<br />
from chitosan <strong>as</strong> described previously [9]. The DQ and DAc of the TMCs were determined with<br />
1H NMR on a Varian INOVA 500 MHz NMR spectrometer (Varian Inc., Palo Alto, Ca, USA) at 80<br />
°C in D 2O [9]. Furthermore, M n and M w of the TMCs were determined, <strong>as</strong> described previously<br />
[8], by GPC on a Viscotek system detecting refractive index, viscosity and light scattering. A<br />
Shodex OHPak SB-806 column (30 cm) w<strong>as</strong> used with 0.3 M sodium acetate pH 4.4 (adjusted<br />
with acetic acid) <strong>as</strong> running buffer. The structural characteristics of the synthesized TMCs are<br />
summarized in Table 1.<br />
Table 1. Polymer characteristics of TMC and TMC-RA.<br />
Abbreviation M n (kDa) M w (kDa) DQ (%) DAc (%)<br />
TMC 36 75 43 17<br />
TMC-RA 43 83 44 54<br />
In vitro studies on human MO-DCs. Immature DCs were cultured <strong>as</strong> described be<strong>for</strong>e [18].<br />
In short, human blood monocytes were isolated from buffy coats by use of a Ficoll gradient and<br />
subsequently a Percoll gradient. Purified monocytes were differentiated into immature DCs in<br />
the presence of interleukin-4 (IL-4, 500 U/ml) and granulocyte-macrophage colonystimulating<br />
factor (GM-CSF, 800 U/ml).<br />
Immature DCs (day 6) were seeded (50,000 cells/well) in a 96-well plate in a total volume of<br />
100 μl RPMI 10% FCS in the presence of TMC (0.001-30 μg/ml), TMC-WIV <strong>for</strong>mulations at 1:1<br />
and 5:1 (w/w) ratios (0.001-6 μg/ml WIV) or lipopolysaccharide (LPS) (0.01-1000 ng/ml) <strong>as</strong> a<br />
116
Maturation of Human DCs is Correlated with the GlcNAc Content<br />
positive control. As described elsewhere, TMC-WIV <strong>for</strong>mulations had similar physico-chemical<br />
properties (size and zeta potential) independent of the type or amount of TMC used (Chapter<br />
5A). There<strong>for</strong>e, it can be stated that all <strong>for</strong>mulations were physically similar, except <strong>for</strong> the<br />
amount of free polymer.<br />
Uptake and binding of TMC-coated WIV <strong>for</strong>mulations by human MO-DCs. The uptake<br />
and binding of (TMC coated) WIV by human MO-DCs were studied with AlexaFluor-488®<br />
labeled WIV. WIV w<strong>as</strong> labeled with the AlexaFluor-488® labeling kit using the procedure<br />
provided by Invitrogen. DCs were incubated <strong>for</strong> 1 h at 37 or 4 °C with the labeled WIV<br />
<strong>for</strong>mulations and flow cytometry w<strong>as</strong> used to determine cell fluorescence (n=2).<br />
Maturation and cytokine production of human MO-DCs. Human MO-DCs were incubated<br />
<strong>for</strong> 16 h at 37 °C with the various polymers with or without WIV. The cells were w<strong>as</strong>hed<br />
extensively with PBS and incubated <strong>for</strong> 30 min with anti-CD86-APC (BD, Breda, The<br />
Netherlands). DC maturation w<strong>as</strong> determined by analyzing cell-surface expression of costimulatory<br />
molecules (CD86) by flow cytometry. After 16 h culture supernatants were<br />
harvested and frozen at −80 °C until analysis. The supernatants were analyzed <strong>for</strong> the presence<br />
of the cytokines IL-10 and IL-12p40, IL12p70 and TNFα by ELISA (Biosource International,<br />
CA). For the DC maturation studies, 4 donors were used. For the cytokine experiments, 2<br />
donors were used.<br />
Results<br />
Uptake and binding of (TMC coated) WIV by human MO-DCs. The influence of TMC and<br />
TMC-RA on the uptake and binding of fluorescently labeled WIV by human MO-DCs w<strong>as</strong><br />
studied. After incubation <strong>for</strong> 1 h, all <strong>for</strong>mulations showed a concentration dependent uptake<br />
and binding (Figure 1). At 37 °C (Figure 1A) much higher intensities were observed than at 4<br />
°C (Figure 1B), indicating active uptake of WIV. Surprisingly, uncoated WIV showed a slightly<br />
higher binding and uptake than TMC or TMC-RA coated WIV.<br />
117
Chapter 5B<br />
WIV binding/uptake<br />
(MFI)<br />
A<br />
200<br />
150<br />
100<br />
50<br />
WIV<br />
TMC-WIV<br />
TMC-RA-WIV<br />
WIV binding/uptake<br />
(MFI)<br />
B<br />
20<br />
15<br />
10<br />
5<br />
WIV<br />
TMC-WIV<br />
TMC-RA-WIV<br />
0<br />
3.0 1.5 0.8 0.4 0.2<br />
Conc. WIV (μg/ml)<br />
0<br />
3.0 1.5 0.8 0.4 0.2<br />
Conc. WIV (μg/ml)<br />
Figure 1. WIV uptake and/or binding by human MO-DCs after incubation <strong>for</strong> 1 hour with fluorescently<br />
labeled (TMC or TMC-RA coated at w/w ratio 5:1) WIV at 37 °C (A) and 4°C (B). Results from one<br />
representative donor are shown (n=2). TMC(-RA)-WIV <strong>for</strong>mulations at w/w ratio 1:1 showed a similar<br />
outcome (results not shown).<br />
Maturation and cytokine production by human MO-DCs. The effect of TMC and TMC-RA<br />
on MO-DC maturation w<strong>as</strong> studied by quantifying maturation marker CD86 and production of<br />
cytokines IL-10, TNF-α and IL-12. Incubation of human MO-DCs with TMC solutions and TMC-<br />
WIV <strong>for</strong>mulations ((w/w) ratio 5:1) revealed a striking difference in CD86 expression between<br />
TMC-RA and TMC (Figure 2).<br />
CD 86 expression<br />
(MFI)<br />
A<br />
300<br />
200<br />
100<br />
TMC<br />
TMC-RA<br />
CD 86 expression<br />
(MFI)<br />
B<br />
500<br />
400<br />
300<br />
200<br />
100<br />
WIV<br />
TMC-WIV<br />
TMC-RA-WIV<br />
0<br />
15.0<br />
7.5<br />
3.8<br />
1.9<br />
0.9<br />
Medium<br />
1 μg/ml LPS<br />
0<br />
3.0<br />
1.5<br />
0.8<br />
0.4<br />
0.2<br />
Medium<br />
1 μg/ml LPS<br />
Conc. TMC (μg/ml)<br />
Conc. WIV (μg/ml)<br />
Figure 2. CD 86 expression by human MO-DCs after incubation with TMC and TMC-RA solutions (A) and<br />
by TMC-WIV and TMC-RA-WIV (5:1 w/w ratio) <strong>for</strong>mulations (B). Results from one representative donor<br />
are shown (n=4).<br />
118
Maturation of Human DCs is Correlated with the GlcNAc Content<br />
TMC-RA induced a much higher concentration dependent CD86 expression than TMC, <strong>for</strong><br />
both soluble TMC-RA (Figure 2A) and TMC-RA co-<strong>for</strong>mulated with WIV (Figure 2B).<br />
Furthermore, TMC-RA-WIV also stimulated, in a concentration dependent manner, the<br />
secretion of pro-inflammatory cytokines IL-10, TNF-α, IL12p40 and IL-12p70 more strongly<br />
than the other TMC-WIV <strong>for</strong>mulations or naked WIV did (Figure 3). As compared to TMC-RA-<br />
WIV, TMC-RA in solution induced similar cytokine levels (results not shown). Importantly, an<br />
endotoxin dose-effect calibration on DCs verified that the observed effects on DCs cannot be<br />
attributed to the relatively low endotoxin levels in the TMCs (≤0.16 EU/µg TMC), <strong>as</strong> me<strong>as</strong>ured<br />
with a LAL test. As TMC-RA showed very low in vitro cytotoxicity compared to TMC [9], it is<br />
highly unlikely that the maturation of DCs by TMC-RA is attributable to its toxicity. Since both<br />
IL-10 (<strong>as</strong> marker <strong>for</strong> a Th2 type response) and IL-12p70 (<strong>as</strong> marker <strong>for</strong> a Th1 type response)<br />
were elevated by TMC-RA (Figure 3), no polarization towards a Th1 or Th2 type of immune<br />
response could be established. These results suggest that TMC-RA h<strong>as</strong> much stronger intrinsic<br />
adjuvant effect than conventional TMC due to its higher GlcNAc content and that this enhanced<br />
activation does not seem to be the result of incre<strong>as</strong>ed uptake of antigen.<br />
Discussion<br />
In general, cationic nanoparticles are better taken up by APCs than their anionic<br />
counterparts [19, 20], likely due to improved interaction with the negatively charged cell<br />
membrane. Despite being anionic, uncoated WIV demonstrated slightly better uptake<br />
compared to the cationic, TMC-coated WIV, indicating that besides electrostatic interactions<br />
other interactions may play a role. It is known that sialic acid and galactose residues on the<br />
surface of influenza viruses can interact with C-type lectin receptors of APCs and enhance<br />
uptake of these particles [21, 22]. The coating of WIV with TMC or TMC-RA may hinder these<br />
interactions. Additionally, the GlcNAc units present in TMC(-RA) may have altered the uptake<br />
mechanism [10, 13]. This obviously resulted in a slightly decre<strong>as</strong>ed uptake of TMC(-RA)-coated<br />
WIV compared to plain WIV. This effect w<strong>as</strong> similar <strong>for</strong> TMC (DAc 17%) and TMC-RA (DAc<br />
54%), indicating that the GlcNAc content of TMC does not critically influence antigen uptake.<br />
In agreement with our present results, in previous studies the extent of antigen uptake did<br />
not appear to be a major determinant <strong>for</strong> DC maturation [5, 23, 24]. However, previous studies<br />
with TMC with low DAc (
6.0<br />
0.2<br />
3.0<br />
6.0<br />
3.0<br />
6.0<br />
3.0<br />
1.5<br />
0.2<br />
Chapter 5B<br />
(with a low DAc) h<strong>as</strong> only a limited intrinsic adjuvant effect on human MO-DCs compared to<br />
TMC-RA (high DAc). TMC, <strong>as</strong> coating of WIV, w<strong>as</strong> not capable to induce additional MO-DC<br />
maturation compared to WIV alone. The different effects of TMC on the maturation of MO-DCs<br />
may be explained by the nature of the antigens used <strong>for</strong> these studies: WIV h<strong>as</strong> a<br />
nanoparticulate structure (appr. 180 nm) and already induces (minor) DC activation without<br />
TMC. Ovalbumin, <strong>as</strong> a soluble antigen, h<strong>as</strong> no effect on DC maturation and mixing with TMC<br />
will lead to (minor) self-<strong>as</strong>sembly into larger structures due to charge interactions [5, 25]. In<br />
the latter c<strong>as</strong>e, DC activation by TMC, either due to changes in the particulate nature of a<br />
<strong>for</strong>mulation [1, 26] or due to a direct effect of the TMC [5, 24], will be more noticeable.<br />
IL-10 (pg/ml)<br />
A<br />
120<br />
100<br />
80<br />
60<br />
40<br />
20<br />
WIV<br />
TMC-WIV<br />
TMC-RA-WIV<br />
1 μg/ml LPS<br />
TNFα (pg/ml)<br />
B<br />
10000<br />
7500<br />
5000<br />
2500<br />
WIV<br />
TMC-WIV<br />
TMC−RA-WIV<br />
1 μg/ml LPS<br />
0<br />
0<br />
0.8<br />
0.4<br />
6.0<br />
1.5<br />
0.8<br />
0.4<br />
0.2<br />
Conc. WIV (μg/ml)<br />
Conc. WIV (μg/ml)<br />
IL-12p40 (pg/ml)<br />
C<br />
10000<br />
7500<br />
5000<br />
2500<br />
WIV<br />
TMC-WIV<br />
TMC-RA-WIV<br />
1 μg/ml LPS<br />
IL-12p70 (pg/ml)<br />
D<br />
50<br />
40<br />
30<br />
20<br />
10<br />
WIV<br />
TMC-WIV<br />
TMC-RA-WIV<br />
1 μg/ml LPS<br />
0<br />
3.0<br />
1.5<br />
0.8<br />
0.4<br />
0<br />
1.5<br />
0.8<br />
0.4<br />
0.2<br />
Conc. WIV (μg/ml)<br />
Conc. WIV (μg/ml)<br />
Figure 3. Expression of IL-10 (A); TNFα (B); IL-12p40 (C) and IL12p70 (D) by human MO-DCs after<br />
incubation with TMC-WIV <strong>for</strong>mulations (w/w ratio 5:1). The figure shows results from one<br />
representative donor (n=2).<br />
TMC-RA, <strong>as</strong> coating of WIV or solution, lacked immuno-stimmulatory effects in murine bonemarrow<br />
derived DCs (Chapter 5A). In contr<strong>as</strong>t, our present data show that TMC-RA induces<br />
maturation of human monocyte derived DCs. This suggest that TMC-RA may activate human<br />
120
Maturation of Human DCs is Correlated with the GlcNAc Content<br />
DCs by binding to a receptor that is not present on murine DCs. GlcNAc moieties are known to<br />
interact with species-independent C-type lectins, like mannose receptors [10, 13, 14], and tolllike<br />
receptor type 2 (TLR2) [15] present on the surface of APCs. DC-SIGN is a human C-type<br />
lectin not present on murine DCs [11, 17], but experiments with DCs derived from DC-SIGN<br />
transgenic mice showed no immuno-stimulatory effect of TMC-RA (unpublished results). This<br />
suggests that other GlcNAc, human specific, receptors may play a role. Further studies should<br />
be done to investigate the mechanism of human MO-DC activation by TMC-RA in more detail.<br />
Conclusion<br />
As opposed to TMC with a low GlcNAc content, TMC-RA with a high GlcNAc content (54%),<br />
both <strong>as</strong> solution and <strong>as</strong> coating of WIV, w<strong>as</strong> shown to be capable of inducing maturation of<br />
human MO-DCs. This maturation w<strong>as</strong> not a result of incre<strong>as</strong>ed antigen uptake. Both IL-10 and<br />
IL-12p70 levels were elevated after stimulation with TMC-RA, indicating a mixed Th1/Th2<br />
type immune response. These immuno-stimulatory characteristics, together with its low<br />
toxicity profile, make TMC-RA a promising adjuvant <strong>for</strong> future vaccinations in humans.<br />
Acknowledgement. This research w<strong>as</strong> partially per<strong>for</strong>med under the framework of <strong>TI</strong><br />
<strong>Pharma</strong> project number D5-106-1; Vaccine delivery: alternatives <strong>for</strong> conventional multiple<br />
injection vaccines. Nobilon International BV (Boxmeer, The Netherlands) is acknowledged <strong>for</strong><br />
supplying the WIV.<br />
121
Chapter 5B<br />
References<br />
1. A.C. Rice-Ficht, A.M. Aren<strong>as</strong>-Gamboa, M.M. Kahl-McDonagh, T.A. Ficht. Polymeric particles<br />
in vaccine delivery. Curr Opin Microbiol 13: 106-112 (2010).<br />
2. V.E. Schijns. Immunological concepts of vaccine adjuvant activity. Curr Opin Immunol 12: 456-<br />
463 (2000).<br />
3. P. Nordly, H.B. Madsen, H.M. Nielsen, C. Foged. Status and future prospects of lipid-b<strong>as</strong>ed<br />
particulate delivery systems <strong>as</strong> vaccine adjuvants and their combination with<br />
immunostimulators. Exp Opin Drug Deliv 6:657-672 (2009).<br />
4. M. Amidi, S.G. Romeijn, J.C. Verhoef, H.E. Junginger, L. Bungener, A. Huckriede, D.J.A.<br />
Crommelin, W. Jiskoot. N-<strong>Trimethyl</strong> chitosan (TMC) nanoparticles loaded with influenza<br />
subunit antigen <strong>for</strong> intran<strong>as</strong>al vaccination: Biological properties and immunogenicity in a mouse<br />
model. Vaccine 25:144-153 (2007).<br />
5. S.M. Bal, B. Slütter, E. van Riet, A.C. Kruithof, Z. Ding, G.F.A. Kersten, W. Jiskoot, J.A.<br />
Bouwstra. Efficient induction of immune responses through intradermal vaccination with N-<br />
trimethyl chitosan containing antigen <strong>for</strong>mulations. J Control Rele<strong>as</strong>e 142:374-383 (2010).<br />
6. W. Boonyo, H.E. Junginger, N. Waranuch, A. Polnok, T. Pitaksuteepong. Chitosan and<br />
trimethyl chitosan chloride (TMC) <strong>as</strong> adjuvants <strong>for</strong> inducing immune responses to ovalbumin in<br />
mice following n<strong>as</strong>al administration. J Control Rele<strong>as</strong>e 121:168-175 (2007).<br />
7. N. Hagenaars, E. M<strong>as</strong>trobattista, R.J. Verheul, I. Mooren, H.L. Glansbeek, J.G.M. Heldens, H.<br />
Van Den Bosch, W. Jiskoot. Physicochemical and immunological characterization of N,N,Ntrimethyl<br />
chitosan-coated whole inactivated influenza virus vaccine <strong>for</strong> intran<strong>as</strong>al<br />
administration. Pharm Res 26:1353-1364 (2009).<br />
8. R.J. Verheul, M. Amidi, S. van der Wal, E. van Riet, W. Jiskoot, W.E. Hennink. Synthesis,<br />
characterization and in vitro biological properties of O-methyl free N,N,N-trimethylated<br />
chitosan. Biomaterials 29:3642-3649 (2008).<br />
9. R.J. Verheul, M. Amidi, M.J. van Steenbergen, E. van Riet, W. Jiskoot, W.E. Hennink. Influence<br />
of the degree of acetylation on the enzymatic degradation and in vitro biological properties of<br />
trimethylated chitosans. Biomaterials 30:3129-3135 (2009).<br />
10. M.J. Robinson, D. Sancho, E.C. Slack, S. LeibundGut-Landmann, C.R. Sousa. Myeloid C-type<br />
lectins in innate immunity. Nat Immunol 7:1258-1265 (2006).<br />
11. S.K. Singh, J. Stephani, M. Schaefer, H. Kalay, J.J. Garcia-Vallejo, J. den Haan, E. Saeland, T.<br />
Sparw<strong>as</strong>ser, Y. van Kooyk. Targeting glycan modified OVA to murine DC-SIGN transgenic<br />
dendritic cells enhances MHC cl<strong>as</strong>s I and II presentation. Mol Immunol 47:164-174 (2009).<br />
12. P. Zhang, S. Snyder, P. Feng, P. Azadi, S. Zhang, S. Bulgheresi, K.E. Sanderson, J. He, J. Klena,<br />
T. Chen. Role of N-acetylglucosamine within core lipopolysaccharide of several species of<br />
Gram-negative bacteria in targeting the DC-SIGN (CD209). J Immunol 177:4002-4011 (2006).<br />
13. J. Nadesalingam, A.W. Dodds, K.B.M. Reid, N. Palaniyar. Mannose-binding lectin recognizes<br />
peptidoglycan via the N-acetyl glucosamine moiety, and inhibits ligand-induced<br />
proinflammatory effect and promotes chemokine production by macrophages. J Immunol<br />
175:1785-1794 (2005).<br />
14. Y. Shibata, W. James Metzger, Q.N. Myrvik. Chitin particle-induced cell-mediated immunity is<br />
inhibited by soluble mannan: Mannose receptor-mediated phagocytosis initiates IL-12<br />
production. J Immunol 159:2462-2467 (1997).<br />
15. C.A. Da Silva, D. Hartl, W. Liu, C.G. Lee, J.A. Eli<strong>as</strong>. TLR-2 and IL-17A in chitin-induced<br />
macrophage activation and acute inflammation. J Immunol 181:4279-4286 (2008).<br />
16. N. Hagenaars, R.J. Verheul, I. Mooren, P.H.J.L.F. de Jong, E. M<strong>as</strong>trobattista, H.L. Glansbeek,<br />
J.G.M. Heldens, H. van den Bosch, W.E. Hennink, W. Jiskoot. Relationship between structure<br />
and adjuvanticity of N,N,N-trimethyl chitosan (TMC) structural variants in a n<strong>as</strong>al influenza<br />
vaccine. J Control Rele<strong>as</strong>e 140:126-133 (2009).<br />
17. K. Wethmar, Y. Helmus, K. Lühn, C. Jones, A. L<strong>as</strong>kowska, G. Varga, S. Grabbe, R. Lyck, B.<br />
Engelhardt, M.G. Bixel, S. Butz, K. Loser, S. Beissert, U. Ipe, D. Vestweber, M.K. Wild.<br />
122
Maturation of Human DCs is Correlated with the GlcNAc Content<br />
Migration of immature mouse DC across resting endothelium id mediated by ICAM-2 but<br />
independent ß2-integrins and murine DC-SIGN homologues. Eur J Immunol 36:2781-2794<br />
(2006).<br />
18. F. Sallusto A. Lanzavecchia. Efficient presentation of soluble antigen by cultured human<br />
dendritic cells is maintained by granulocyte/macrophage colony-stimulating factor plus<br />
interleukin 4 and downregulated by tumor necrosis factor α. J Exp Med 179:1109-1118 (1994).<br />
19. K.S. Jaganathan S.P. Vy<strong>as</strong>. Strong systemic and mucosal immune responses to surface-modified<br />
PLGA microspheres containing recombinant Hepatitis B antigen administered intran<strong>as</strong>ally.<br />
Vaccine 24:4201-4211 (2006).<br />
20. B. Sayin, S. Somavarapu, X.W. Li, M. Thanou, D. Sesardic, H.O. Alpar, S. Şenel. Mono-Ncarboxymethyl<br />
chitosan (MCC) and N-trimethyl chitosan (TMC) nanoparticles <strong>for</strong> non-inv<strong>as</strong>ive<br />
vaccine delivery. Int J Pharm 363:139-148 (2008).<br />
21. Y. Fujimura, M. Takeda, H. Ikai, K. Haruma, T. Akisada, T. Harada, T. Sakai, M. Ohuchi. The<br />
role of M cells of human n<strong>as</strong>opharyngeal lymphoid tissue in influenza virus sampling. Virch<br />
Archiv 444:36-42 (2004).<br />
22. G.N. Rogers, J.C. Paulson. Receptor determinants of human and animal influenza virus isolates:<br />
Differences in receptor specificity of the H3 hemagglutinin b<strong>as</strong>ed on species of origin. Virology<br />
127:361-373 (1983).<br />
23. B. Slütter, P.C. Soema, Z. Ding, R. Verheul, W. Hennink, and W. Jiskoot. Conjugation of<br />
ovalbumin to trimethyl chitosan improves immunogenicity of the antigen. J Control Rele<strong>as</strong>e 143<br />
:207-214 (2010).<br />
24. B. Slütter, S. Bal, C. Keijzer, R. Mallants, N. Hagenaars, I. Que, E. Kaijzel, W. van Eden, P.<br />
Augustijns, C. Löwik, J. Bouwstra, F. Broere, W. Jiskoot. N<strong>as</strong>al vaccination with N-trimethyl<br />
chitosan and PLGA b<strong>as</strong>ed nanoparticles: Nanoparticle characteristics determine quality and<br />
strength of the antibody response in mice against the encapsulated antigen. Vaccine 28:6282-<br />
6291 (2010).<br />
25. S. Bakowsky, A. Jintapattanakit, T. Kissel. Self-<strong>as</strong>sembled polyelectrolyte nanocomplexes<br />
between chitosan derivatives and insulin. J Pharm Sci 95:1035-1043 (2006).<br />
26. S.D. Xiang, A. Scholzen, G. Minigo, C. David, V. Apostolopoulos, P.L. Mottram, M. Plebanski.<br />
Pathogen recognition and development of particulate vaccines: Does size matter?. Methods 40:1-<br />
9 (2006).<br />
123
CHAPTER 6<br />
TAILORABLE THIOLATED TRIMETHYL<br />
CHITOSANS FOR COVALENTLY STABILIZED<br />
NANOPAR<strong>TI</strong>CLES<br />
Rolf J. Verheul, Steffen van der Wal, Wim E. Hennink.<br />
Biomacromolecules 2010, 11, 1965-1971
Chapter 6<br />
Abstract<br />
A novel four-step method is presented to synthesize partially thiolated trimethylated<br />
chitosan (TMC) with tailorable degree of quaternization and thiolation. First, chitosan w<strong>as</strong><br />
partially N-carboxylated with glyoxilic acid and sodium borohydride. Next, the remaining<br />
amines were quantitatively dimethylated with <strong>for</strong>maldehyde and sodium borohydride and<br />
then quaternized with iodomethane in NMP. Subsequently, these partially carboxylated TMCs<br />
dissolved in water were reacted with cystamine at pH 5.5 using EDC <strong>as</strong> coupling agent. After<br />
addition of DTT and dialysis, thiolated TMCs were obtained varying in degree of quaternization<br />
(25-54%) and degree of thiolation (5-7%) <strong>as</strong> determined with 1 H-NMR and Ellman’s <strong>as</strong>say. Gel<br />
permeation chromatography with light scattering detection indicated limited intermolecular<br />
crosslinking. All thiolated TMCs showed rapid oxidation to yield disulfide crosslinked TMC at<br />
pH 7.4 while the thiolated polymers were rather stable at pH 4.0. Using Calu-3 cells, XTT and<br />
LDH cell viability tests showed a slight reduction in cytotoxicity <strong>for</strong> thiolated TMCs <strong>as</strong><br />
compared to the non-thiolated polymers with similar DQs. Positively charged nanoparticles<br />
loaded with fluorescently labeled ovalbumin were made from thiolated TMCs and thiolated<br />
hyaluronic acid. The stability of these particles w<strong>as</strong> confirmed in 0.8 M NaCl, in contr<strong>as</strong>t to<br />
particles made from non-thiolated polymers which dissociated under these conditions<br />
demonstrating that the particles were held together by intermolecular disulfide bonds.<br />
Chitosan<br />
+<br />
Thiolated TMC<br />
N(CH 3 ) 3<br />
HS<br />
Thiolated<br />
Hyaluronic<br />
acid<br />
+<br />
―SH<br />
+<br />
+<br />
―SH<br />
+<br />
-<br />
-<br />
--<br />
+<br />
―SH<br />
―SH<br />
+<br />
+<br />
―SH<br />
Covalently stabilized<br />
nanoparticles<br />
126
Thiolated TMC <strong>for</strong> Covalently Stabilized Nanoparticles<br />
Introduction<br />
Chitosan, a polysaccharide consisting of β1→4-D-glucosamine and β1→4 N-acetyl D-<br />
glucosamine units, is under investigation <strong>for</strong> various biomedical and pharmaceutical<br />
applications [1-3]. However, its poor aqueous solubility and loss of penetration enhancing<br />
activity above pH 6 is a major drawback <strong>for</strong> its use at physiological conditions. N,N,N,-<br />
trimethylated chitosan (TMC), a partially quaternized derivative of chitosan, is water-soluble<br />
at physiological pH. TMC h<strong>as</strong> been widely studied in the biomedical field <strong>as</strong> drug, antigen and<br />
gene delivery vehicle. It h<strong>as</strong> been shown in several in vitro and in vivo studies that TMC h<strong>as</strong> low<br />
toxicity, possesses muco-adhesive properties and can facilitate the uptake of small drug<br />
molecules <strong>as</strong> well <strong>as</strong> proteins via various mucosal routes [4-11] Several studies have suggested<br />
an optimal degree of quaternization (DQ) of 40-50% <strong>for</strong> mucosal transport and gene delivery<br />
[12-16]. In addition, molecular weight [17, 18] and, <strong>as</strong> recently demonstrated, O-methylation<br />
[19] and degree of N-acetylation [20] also have a major impact on the physical and biological<br />
characteristics of TMC.<br />
Introduction of thiol-moieties will further broaden the potential pharmaceutical applications<br />
of TMC by enhancing its muco-adhesive potential and allowing further chemical derivatization<br />
reactions via reducible disulfide-bridges [11, 21-24]. Yin et al. [25] synthesized thiolated TMC<br />
by coupling cysteine to the remaining free –NH 2 units of the polymer. However, this method is<br />
dependent on available free amines remaining after quaternization, and there<strong>for</strong>e this<br />
synthetic route is only applicable <strong>for</strong> TMC with a low DQ (up to 30 %).<br />
Nanoparticulate systems have superior penetration enhancing characteristics in mucosal<br />
protein and/or antigen delivery over polymer solutions [1, 26, 27]. Although often<br />
tripolyphosphate (TPP) is used <strong>as</strong> a crosslinker to <strong>for</strong>m TMC nanoparticles via ionic gelation,<br />
recent results by Sayın et al. [28] showed that n<strong>as</strong>al immunization with tetanus toxoid loaded<br />
TMC:mono-N-carboxymethyl chitosan nanoparicles resulted in superior antibody titers<br />
compared to the TMC/TPP particles. The physico-chemical stability of these complexes is<br />
dependent on the characteristics of the polyelectrolytes used and they often have a limited<br />
stability in physiological salt conditions [29, 30] or at low pH [21]. Thiolated chitosans, and<br />
more recently, thiolated TMCs complexed with insulin or using TPP <strong>as</strong> crosslinker resulted in<br />
nanoparticles stabilized via intracellularly degradable disulfide bridges [21, 25, 31]. Combining<br />
the cationic, thiolated TMC polymer with the muco-adhesive [22], anionic, thiolated hyaluronic<br />
acid to <strong>for</strong>m stabilized nanoparticles may further improve these mucosal delivery vehicles.<br />
Further, it can be anticipated that the charge of the particles can be tuned by the<br />
127
Chapter 6<br />
TMC:hyaluronic acid ratio and that both anionically and cationically charged proteins can be<br />
loaded.<br />
In this paper, a novel synthetic method is described to yield thiolated TMCs with a high DQ.<br />
The different thiolated TMCs were physico-chemically characterized, evaluated in in vitro<br />
cytotoxicity <strong>as</strong>says and used <strong>for</strong> preparation of covalently stabilized nanoparticles with<br />
thiolated hyaluronic acid.<br />
Materials and Methods<br />
Materials. Chitosan with a residual degree of acetylation of 17% (determined with 1 H-NMR<br />
<strong>as</strong> described below) and a number average (M n) and weight average molecular weight (M w) of<br />
28 and 43 kDa, (determined with GPC-TD <strong>as</strong> below), respectively, w<strong>as</strong> purch<strong>as</strong>ed from Primex<br />
(Siglufjodur, Iceland). Sodium borohydride, <strong>for</strong>maldehyde 37% (stabilized with methanol),<br />
glyoxilic acid monohydrate, cystamine dihydrochloride, dithiotreitol (DTT), 1-ethyl-3-(3-<br />
dimethylaminopropyl) carbodiimide HCl (EDC), L-cysteine HCl monohydrate, deuterium oxide,<br />
sodium 3’-[1-(phenylaminocarbonyl)-3,4-tetrazolium]-bis(4-methoxy-6-nitro) benzene<br />
sulphonic acid hydrate (XTT), N-methyl dibenzopyrazine methylsulphate (PMS), sodium<br />
acetate, acetic acid (anhydrous), sodium hydroxide and hydrochloric acid were obtained from<br />
Sigma-Aldrich Chemical Co. Fluorescently labeled ovalbumin (OVA-FITC), minimal essential<br />
medium (MEM) and fetal calf serum (FCS) were obtained from Invitrogen (Breda, The<br />
Netherlands). Sicapent w<strong>as</strong> ordered from Merck (Darmstadt, Germany). Iodomethane 99%<br />
stabilized with copper w<strong>as</strong> obtained from Acros Organics (Geel, Belgium). 5,5-dithio-bis-(2-<br />
nitrobenzoic acid) (Ellman’s reagent) w<strong>as</strong> purch<strong>as</strong>ed from Pierce (Rock<strong>for</strong>d, IL, USA). Linear<br />
polyethylenimine (PEI) (22kDa) w<strong>as</strong> synthesized according to Thom<strong>as</strong> et al [32]. TMCs with<br />
DQs of 30% (M n 33 kDa, M w 59 kDa) and 56% (M n 37 kDa, M w 78 kDa) were synthesized and<br />
characterized <strong>as</strong> described previously [19]. HA-SH with a M n of 39.8 kDa, PDI of 1.26 and a<br />
degree of thiolation of 52% w<strong>as</strong> synthesized <strong>as</strong> described elsewhere [33]. All other chemicals<br />
used were of analytical grade.<br />
Synthesis of Partially N-Carboxylated Chitosan. Chitosan (degree of acetylation 17%) w<strong>as</strong><br />
used <strong>as</strong> obtained and selective, partial N-carboxylation w<strong>as</strong> carried out according to a<br />
previously described method with some adjustments [10, 13]. Briefly, chitosan (10 g) w<strong>as</strong><br />
dissolved in 1% (v/v) acetic acid (300 ml). Then, 2.3 g glyoxylic acid w<strong>as</strong> added and pH w<strong>as</strong><br />
raised to 4.5 with 1 M NaOH. Subsequently, 3 g of sodium borohydride dissolved in H 2O (5%<br />
128
Thiolated TMC <strong>for</strong> Covalently Stabilized Nanoparticles<br />
w/v) w<strong>as</strong> added drop wise over a period of 3 hours in portions of 500 mg re-adjusting the pH<br />
with 1 M HCl to 4.5 after each addition. Finally, the partially carboxylated chitosan (CS-COOH)<br />
w<strong>as</strong> precipitated by dropping the reaction mixture into an ethanol/diethyl ether mixture<br />
(50/50 v/v) and the product w<strong>as</strong> extensively w<strong>as</strong>hed with diethyl ether and dried overnight.<br />
Synthesis of Dimethylated, Partially N-Carboxylated Chitosan. N,N,-Dimethylated,<br />
partially N-carboxylated chitosan (DMC-COOH) w<strong>as</strong> synthesized from CS-COOH <strong>as</strong> described<br />
previously <strong>for</strong> the synthesis of N,N,-dimethylated chitosan [20, 34]. In short, CS-COOH (10 g)<br />
w<strong>as</strong> dissolved in 500 ml of 1% acetic acid (v/v). Next, 20 ml of 37% <strong>for</strong>maldehyde solution w<strong>as</strong><br />
added, the pH w<strong>as</strong> adjusted to 4.5 with 1M NaOH and the mixture w<strong>as</strong> stirred <strong>for</strong> 30 minutes at<br />
room temperature. Subsequently, sodium borohydride (5 g) w<strong>as</strong> added in portions of 500 mg<br />
over a period of 5 hours, adjusting the pH to 4.5 with 1M HCl after each addition. After<br />
precipitation with acetone, the product w<strong>as</strong> w<strong>as</strong>hed extensively with acetone and dried<br />
overnight. To obtain quantitative dimethylation a second methylation step w<strong>as</strong> per<strong>for</strong>med<br />
using the procedure described above.<br />
Synthesis of O-Methyl free, <strong>Trimethyl</strong>ated, Partially N-Carboxylated Chitosan. DMC-<br />
COOH w<strong>as</strong> reacted with iodomethane to yield trimethylated, partially N-carboxylated chitosan<br />
(TMC-COOH). To prevent O-methylation, the reaction of DMC with iodomethane w<strong>as</strong> done in<br />
NMP, without the addition of a b<strong>as</strong>e catalyst. TMC-COOH with a degree of quaternization of<br />
around 25% or 54% were obtained by the method described previously [19]. In detail, DMC-<br />
COOH (2.5 g) w<strong>as</strong> dissolved in 80 ml deionized water and the pH w<strong>as</strong> adjusted to 11 with a 1 M<br />
solution of NaOH, resulting in gel <strong>for</strong>mation. Then, the gel w<strong>as</strong> w<strong>as</strong>hed with water and<br />
subsequently with acetone. To remove residual solvents, the DMC w<strong>as</strong> dried under vacuum <strong>for</strong><br />
2 hours. Next, DMC-COOH w<strong>as</strong> suspended in 100 ml NMP followed by the addition of 5 or 15<br />
ml iodomethane. The dispersion w<strong>as</strong> stirred at 40°C <strong>for</strong> 40 hours and subsequently dropped<br />
into 400 ml of an ethanol/diethyl ether mixture (50/50 v/v) to precipitate the <strong>for</strong>med TMC-<br />
COOH, which w<strong>as</strong> collected by centrifugation and subsequently extensively w<strong>as</strong>hed with<br />
diethyl ether. After drying overnight at room temperature, the obtained TMC-COOH w<strong>as</strong><br />
dissolved in 100 ml of an aqueous 10% NaCl solution <strong>for</strong> ion-exchange. Finally, the TMC-COOH<br />
w<strong>as</strong> dialyzed against 1% NaCl in diluted HCl (pH 4) <strong>for</strong> 2 days followed by dialysis against<br />
diluted HCl (pH 4) <strong>for</strong> another 2 days (changing buffer twice daily). The polymer solution w<strong>as</strong><br />
filtered through a 0.8 µm filter and collected after freeze-drying.<br />
129
Chapter 6<br />
Synthesis of O-Methyl Free, <strong>Trimethyl</strong>ated, Partially Thiolated Chitosan. Thiolated TMC<br />
w<strong>as</strong> synthesized by first coupling cystamine to the carboxylic acid moieties of TMC-COOH<br />
using EDC <strong>as</strong> coupling agent followed by reduction of the S-S bonds of the coupled cystamine<br />
groups by dithiothreitol. In detail, TMC-COOH (1 g) w<strong>as</strong> dissolved in deionized water at a<br />
concentration of 100 mg/ml. Then, five hundred milligram of cystamine dihydrochloride w<strong>as</strong><br />
added (molar ratio COOH:cystamine approx. 1:4). After complete dissolution, EDC w<strong>as</strong> added<br />
to the reaction mixture (final concentration of 200 mM) to activate the carboxylic acid groups<br />
of the polymer and the pH w<strong>as</strong> adjusted to 5.5 with 1 M NaOH. After reacting <strong>for</strong> 6 hours at<br />
room temperature on a roller bank, pH w<strong>as</strong> raised to 8 with 1 M NaOH and DTT w<strong>as</strong> added in<br />
final concentration of 100 mM. After approximately 2 hours, NaCl w<strong>as</strong> added to the reaction<br />
mixture (final concentration 1% (g/v)), and the pH w<strong>as</strong> adjusted to 4 with 1M HCl. The<br />
resulting solution w<strong>as</strong> dialyzed against 1% NaCl in diluted HCl (pH 4) <strong>for</strong> 2 days changing<br />
buffer twice daily followed by dialysis against diluted HCl (pH 4) <strong>for</strong> another 2 days (changing<br />
buffer twice daily). Dialysis w<strong>as</strong> per<strong>for</strong>med at 4°C. Finally, the solution w<strong>as</strong> filtered through a<br />
0.8 µm filter and the polymers were collected after freeze-drying. To improve the reduction of<br />
remaining disulfide bonds, TMC-SH obtained after this first cycle w<strong>as</strong> subjected to a second<br />
reduction cycle with DTT <strong>as</strong> described above.<br />
Determination of the Degrees of Carboxylation, Dimethylation, Acetylation and<br />
Quaternization. The 1 H-NMR spectra of the various chitosan derivatives were recorded with<br />
a Varian INOVA 300MHz NMR spectrometer (Varian Inc., Palo Alto, Ca, USA) at 25°C in D 2O.<br />
The degree of carboxylation of CS-COOH w<strong>as</strong> calculation <strong>as</strong> follows:<br />
D Carb = [[CH 2]/[H2-H6] x 6/2] x 100%<br />
Here, [CH 2] is the integral of the two hydrogens of the N-carboxymethyl groups at 3.2 ppm [35,<br />
36] and [H2-H6] is the integral corresponding the six protons bound to the C-2 to C-6 ring<br />
carbons between 3.9 and 3.0 ppm (without the signal observed at 3.2 ppm).<br />
The degree of dimethylation (DDM) of the DMC-COOH w<strong>as</strong> calculated <strong>as</strong> follows [19]:<br />
DDM = [(CH 3) 2]/[H2-H6] x 100%<br />
130
Thiolated TMC <strong>for</strong> Covalently Stabilized Nanoparticles<br />
Here, [(CH 3) 2] is the integral of the hydrogens of the dimethyl amino groups at 2.9 ppm and<br />
[H2-H6] is the integral corresponding to the six protons bound to the C-2 to C-6 ring carbons<br />
between 3.9 and 3.0 ppm (without the signal observed at 3.2 ppm).<br />
The 1 H-NMR spectra of the carboxylated and thiolated TMCs were recorded with a Varian<br />
INOVA 500MHz NMR spectrometer (Varian Inc., Palo Alto, Ca, USA) at 80°C in D 2O. The DQ and<br />
degree of dimethylation (DDM) of the TMC-COOHs and TMC-SHs were calculated <strong>as</strong> previously<br />
described [4, 37, 38]:<br />
DQ = [[(CH 3) 3]/[H] × 1/9] × 100%<br />
DDM = [[(CH 3) 2]/[H] × 1/6] × 100%<br />
Here, [(CH 3) 3] and [(CH 3) 2] are the integrals of the hydrogens of the trimethylated amino<br />
groups at 3.3 ppm and the dimethylated amino groups at 2.9 ppm, respectively. [H] is the<br />
integral of the H-1 peaks between 4.7 and 5.7 ppm; the signal related to the hydrogen atoms<br />
bound to the C-1 ring carbon of TMC.<br />
The degree of acetylation (DAc) of the chitosan derivatives w<strong>as</strong> calculated <strong>as</strong> described<br />
previously [39]:<br />
DAc = [[CH 3]/[H] x 1/3] x 100<br />
Here, [CH 3] is the integral of the three hydrogens of the acetyl groups at 2.0 ppm and [H] is the<br />
integral of the H-1 peaks between 4.7 and 5.7 ppm; the signal related to the hydrogen atoms<br />
bound to the C-1 ring carbon of TMC.<br />
Determination of Free Thiol Group Content. The degree of thiolation w<strong>as</strong> quantified with<br />
5,5-dithio-bis-(2-nitrobenzoic acid) (DTNB, Ellman’s reagent). DTNB w<strong>as</strong> dissolved in 0.1 M<br />
sodium phosphate, pH 8.0 containing 1 mM EDTA in a final concentration of 100 μg/ml. TMC-<br />
SHs were dissolved in 0.1 M sodium phosphate, pH 8.0 containing 1 mM EDTA in a<br />
concentration of 2.5 mg/ml. Then, 20 μl of this solution w<strong>as</strong> mixed with 180 μl of DTNB<br />
reagent solution in a microplate and after incubating <strong>for</strong> 15 minutes at room temperature the<br />
absorbance w<strong>as</strong> me<strong>as</strong>ured with a microplate reader at a wavelength of 405 nm (Bio-Rad<br />
Novapath, Hercules, Ca, USA). Cysteine standards were used to calculate the amount of free<br />
thiol moieties in the polymer. The degree of thiolation (D thiol) w<strong>as</strong> calculated using the<br />
estimated average molecular weight per glucosamine unit.<br />
131
Chapter 6<br />
Determination of Total Thiol Content. The total thiol content (free thiol moieties and<br />
disulfides) on the polymers w<strong>as</strong> determined according to a slightly modified method described<br />
by Hombach et al. [40]. In detail, TMC-SHs were dissolved in 50 mM phosphate buffer pH 7.0 (1<br />
mg/ml). Subsequently, 400 μl of a freshly prepared 10% (w/v) sodium borohydride solution in<br />
deionized water w<strong>as</strong> added to 500 μl of polymer solution and the mixture w<strong>as</strong> slightly shaken<br />
at 37°C <strong>for</strong> 60 minutes. Thereafter, 1 ml of 1 M HCl w<strong>as</strong> added to decompose the remaining<br />
sodium borohydride and after 10 minutes the pH w<strong>as</strong> raised to 8 by adding 1 ml of 0.5 M<br />
phosphate buffer pH 8.0. Then, 100 μl of DTNB reagent solution (4 mg/ml in 0.5 M sodium<br />
phosphate, pH 8.0) w<strong>as</strong> added and after incubating <strong>for</strong> 15 minutes at room temperature, a 200<br />
μl sample w<strong>as</strong> transferred into a microplate. The absorbance w<strong>as</strong> me<strong>as</strong>ured with a microplate<br />
reader at a wavelength of 405 nm (Bio-Rad Novapath, Hercules, Ca, USA). Cysteine standards<br />
prepared in the same way <strong>as</strong> the samples were used to calculate the total amount of thiol<br />
moieties (<strong>as</strong> free thiols or disulfides) on the polymer.<br />
Thiol Oxidation. TMC-SH w<strong>as</strong> dissolved in 100 mM acetate buffer pH 4.0, 100 mM acetate<br />
buffer pH 5.0, 100 mM phosphate buffer pH 6.2, or 100 mM phosphate buffer pH 7.4, in a final<br />
concentration of 5 mg/ml. Samples were incubated at 37°C under permanent shaking and<br />
aliquots were taken at different time points. Immediately after collecting the aliquots, the<br />
amount of remaining thiol moieties w<strong>as</strong> determined with Ellman’s reagent <strong>as</strong> described be<strong>for</strong>e.<br />
Determination of M n and M w of the Different Polymers. Polymers were placed overnight<br />
in a vacuum oven at 40 o C in the presence of Sicapent to remove residual water and<br />
subsequently dissolved in 0.3 M sodium acetate pH 4.4. The number average weight (M n) and<br />
weight average weight (M w) of chitosan and the various TMCs were determined by gel<br />
permeation chromatography (GPC) on a Viscotek-triple detection system using a Shodex<br />
OHPak SB-806 column (30 cm) and 0.3 M sodium acetate pH 4.4 (adjusted with acetic acid) <strong>as</strong><br />
running buffer [41]. Data from the l<strong>as</strong>er photometer (λ = 670 nm) (right (90 0 ) and low (7 0 )<br />
angle light scattering), refractive index detector and viscosity detector were integrated using<br />
the provided Omnisec-software to calculate the M n, M w of the different samples. Pullulan (M n =<br />
102 kDa, M w = 106 kDa) obtained from Viscotek Benelux (Oss, the Netherlands) w<strong>as</strong> used <strong>for</strong><br />
calibration.<br />
XTT Cytotoxicity Assay. Calu-3 cells (ATCC, Teddington, UK) were seeded into a 96-well<br />
plate at a density of 2x10 4 cells per well and incubated <strong>for</strong> 2 days at 37°C, CO 2 5% in culture<br />
132
Thiolated TMC <strong>for</strong> Covalently Stabilized Nanoparticles<br />
medium (MEM with L-glutamine and sodium pyruvate supplemented with<br />
antibiotics/antimycotics and 10% FCS). The medium w<strong>as</strong> removed and the cells were<br />
incubated <strong>for</strong> 2.5 hours with 100 µl TMC solutions in MEM (TMC concentrations were 0.01, 0.1,<br />
1 and 10 mg/ml, pH set at 7 with 0.1 M NaOH). Linear PEI (0.1 and 0.01 mg/ml) w<strong>as</strong> used <strong>as</strong><br />
positive control and cells incubated with MEM were used <strong>as</strong> reference. Thereafter, the<br />
solutions were removed and the cells w<strong>as</strong>hed MEM. Then, 100 µl of MEM containing 10% FCS<br />
w<strong>as</strong> added to the cells together with 50 µl of a freshly prepared solution of 1 mg/ml XTT in<br />
MEM containing 25 µmol PMS <strong>as</strong> electron-coupling reagent. Cells were incubated <strong>for</strong> 1 hour at<br />
37°C in 5% CO 2 and the absorbance w<strong>as</strong> read at 490 nm with 655 nm <strong>as</strong> reference wavelength.<br />
Obtained values of the samples were related to the mitochondrial activity of Calu-3 cells<br />
incubated with MEM only [42].<br />
LDH Cytotoxicity Assay. Calu-3 cells were seeded into a 96-well plate at a density of 2x10 4<br />
cells per well and incubated <strong>for</strong> 2 days at 37°C, CO 2 5% in culture medium (MEM with L-<br />
glutamine and sodium pyruvate supplemented with antibiotics/antimycotics and 10% FCS).<br />
The medium w<strong>as</strong> removed and the cells were w<strong>as</strong>hed with MEM and incubated <strong>for</strong> 2.5 hours<br />
with 100 µl TMC solutions in MEM (TMC concentrations were 0.01, 0.1, 1 and 10 mg/ml, pH set<br />
at 7 with 0.1 M NaOH). After incubation, the concentration of lactate dehydrogen<strong>as</strong>e (LDH)<br />
present in the supernatant of the samples w<strong>as</strong> determined with the Cytotoxicity Detection Kit-<br />
Plus (Roche Diagnostics, Mannheim, Germany) by me<strong>as</strong>uring absorbance at 490 nm with 655<br />
nm <strong>as</strong> a reference wavelength. A calibration curve w<strong>as</strong> made with the lysis buffer provided by<br />
the manufacturer, setting the LDH concentration me<strong>as</strong>ured with the undiluted lysis buffer at<br />
100% LDH rele<strong>as</strong>e. Cells incubated with linear PEI (0.1 and 0.01 mg /ml) were used <strong>as</strong> control.<br />
One-way ANOVA with Tukey’s post-test w<strong>as</strong> used <strong>for</strong> comparison.<br />
Preparation and Characterization of Covalently Stabilized Nanoparticles.<br />
Covalently stabilized nanoparticles loaded with fluorescently labeled ovalbumin were<br />
prepared with TMC-SH and thiolated hyaluronic acid (HA-SH). TMC-SH with a DQ 25% and<br />
D thiol 5% or a DQ of 54% and D thiol 6% were dissolved in 10 mM HEPES pH 7.4 at 1 mg/ml and<br />
subsequently mixed with 2.5 mg/ml fluorescently labeled ovalbumin (OVA-FITC) in 10 mM<br />
HEPES pH 7.4 at a weight ratio of 10:1 (TMC-SH:OVA-FITC) under magnetic stirring. Then, HA-<br />
SH (0.5 mg/ml in 10 mM HEPES pH 7.4) w<strong>as</strong> added drop-wise yielding an opalescent<br />
nanoparticle dispersion which w<strong>as</strong> incubated at 37°C <strong>for</strong> 3 hours to allow disulfide <strong>for</strong>mation.<br />
As a control nanoparticles with non-thiolated TMC (DQ 30% or 56%), OVA-FITC and HA were<br />
133
Chapter 6<br />
prepared in a similar way. After 3 hours of incubation, the nanoparticle dispersions were<br />
mixed with either 10 mM HEPES pH 7.4 or 0.8 M NaCl in 10 mM HEPES pH 7.4 and<br />
subsequently incubated at 37°C <strong>for</strong> 1 hour.<br />
Particle size w<strong>as</strong> me<strong>as</strong>ured by dynamic light scattering (DLS) using a Malvern ALV CGS-3<br />
(Malvern Instruments, Malvern, UK). DLS results are given <strong>as</strong> a z-average particle size<br />
diameter and a polydispersity index (PDI). The zeta-potential of the nanoparticles w<strong>as</strong><br />
me<strong>as</strong>ured in 10 mM HEPES pH 7.4 using a Zet<strong>as</strong>izer Nano (Malvern Instruments, Malvern, UK).<br />
The protein loading w<strong>as</strong> determined by <strong>as</strong>saying the concentration of OVA-FITC in the<br />
supernatant obtained after centrifugation (10 min at 13000 rpm; Biofuge Pico, PP1/97 #3324;<br />
Heraeus Instruments, Osterode, Germany) of the nanoparticle dispersion using a Fluostar<br />
Optima with 488 nm <strong>as</strong> excitation wavelength and emission w<strong>as</strong> me<strong>as</strong>ured at 520 nm (BMG<br />
Labtech, Offenburg, Germany). The OVA-FITC <strong>as</strong>sociation efficiency (AE%) w<strong>as</strong> calculated<br />
using the following equation:<br />
AE% = (total OVA-FITC ― OVA-FITC in supernatant)/ total OVA-FITC x 100%<br />
134
Thiolated TMC <strong>for</strong> Covalently Stabilized Nanoparticles<br />
Scheme 1. Synthetic route <strong>for</strong> thiolated TMCs.<br />
135
Chapter 6<br />
Results and discussion<br />
Synthesis and Characterization of Thiolated TMC. Thiolated TMCs with tailorable degree<br />
of quaternization (DQ) and thiolation (D thiol) were obtained <strong>as</strong> depicted in Scheme 1. First,<br />
chitosan w<strong>as</strong> selectively N-carboxylated with glyoxylic acid and sodium borohydride to a<br />
degree of carboxylation of 12% <strong>as</strong> determined with 1 H-NMR. Higher degrees of carboxylation<br />
could be achieved by using larger quantities of glyoxylic acid (results not shown), however, a<br />
further incre<strong>as</strong>e in carboxylic acid moieties that ultimately leads to a higher degree of<br />
thiolation w<strong>as</strong> not considered useful <strong>for</strong> the <strong>for</strong>eseen applications; about 10-20 thiol groups<br />
per polymer chain is sufficient <strong>for</strong> intermolecular disulfide-bridge <strong>for</strong>mation. Partially<br />
carboxylated chitosan w<strong>as</strong> subsequently quantitatively dimethylated using <strong>for</strong>maldehyde and<br />
sodium borohydride, <strong>as</strong> confirmed with 1H-NMR. Reaction of dimethylated, partially<br />
carboxylated chitosan (DMC-COOH) with different amounts of iodomethane (5 or 15 ml) in<br />
NMP resulted in TMC-COOH with DQs of 25% and 54%, respectively, indicating tailorability of<br />
the DQ (Table 1). 1 H-NMR analysis also demonstrated that neither O-methylation nor loss of<br />
residual N-acetylated groups w<strong>as</strong> observed after quaternization, in contr<strong>as</strong>t to the synthesis<br />
method applied by others using a strong alkaline component to induce quaternization [38].<br />
TMC-COOH w<strong>as</strong> not soluble at pH 7.4 confirming the zwitterionic character of this polymer.<br />
Table 1. Characteristics of chitosan and the synthesized thiolated TMCs.<br />
DQ DAc<br />
Total thiol<br />
amount<br />
Amount of<br />
free –SH D thiol M n (kDa) M w (kDa)<br />
(μmol/g) (μmol/g)<br />
Chitosan - 17% - - - 28 43<br />
TMC-SH<br />
High DQ<br />
54% 17% 529 (±2) 283 (±21) 6% 66 174<br />
TMC-SH<br />
Low DQ<br />
25% 17% 478 (±10) 236 (±13) 5% 62 215<br />
TMC-SH<br />
Low DQ<br />
2 nd cycle<br />
25% 17% 478 (±10) 342 (±4) 7% 60 144<br />
The carboxylic acid groups of TMC-COOH were reacted with cystamine using EDC <strong>as</strong><br />
coupling agent. In agreement with previous studies [43] we found that this reaction proceeded<br />
most efficiently at pH 5.5. An excess of cystamine w<strong>as</strong> used to avoid the <strong>for</strong>mation of<br />
crosslinked products. Incubation of cystamine with TMC-COOH without EDC did not result in<br />
polymers with detectable thiol groups. Ellman’s <strong>as</strong>say showed that significant thiolation of<br />
TMC w<strong>as</strong> achieved with 200 mM EDC (Table 1). This table shows that about 529 μmol/g thiol-<br />
136
Thiolated TMC <strong>for</strong> Covalently Stabilized Nanoparticles<br />
moieties (<strong>as</strong> free thiols or disulfides) were introduced into TMC-COOH DQ 54% corresponding<br />
with approx. 12% of glucosamine-units and indicates quantitative substitution of the<br />
carboxylic acids with cystamine. Treatment with DTT resulted in 283 μmol/g free –SH groups<br />
in TMC-SH DQ 54% corresponding to a degree of thiolation (D thiol) of ~6%. For TMC-COOH<br />
with a DQ of 25% the conversion w<strong>as</strong> slightly less efficient (478 μmol/g thiol moieties <strong>as</strong> free<br />
thiols or disulfide groups, approx. 10% of glucosamine-units) and reduction with DTT resulted<br />
in 236 μmol/g free –SH groups (approx. D thiol of 5%). D thiol incre<strong>as</strong>ed up to 341 μmol/g polymer<br />
(about 7% degree of thiolation) after a second reduction cycle with DTT. GPC analysis<br />
demonstrated an incre<strong>as</strong>e in both M n and M w with incre<strong>as</strong>ing the substitution degree of the<br />
TMC-SHs. This suggests that some intermolecular crosslinking had occurred. A second<br />
reduction cycle with DTT resulted in a decre<strong>as</strong>e of M w (215 to 144 kDa) compared to TMC-SH<br />
DQ 25% with only one DTT treatment, indicating that the intermolecular crosslinks were<br />
partly broken.<br />
Remaining thiol content (%)<br />
A<br />
100<br />
80<br />
60<br />
40<br />
20<br />
0<br />
0 1 2 3 4 5 6<br />
time (h)<br />
pH 4.0<br />
pH 5.0<br />
pH 6.2<br />
pH 7.4<br />
Remaining thiol content (%)<br />
B<br />
100<br />
80<br />
60<br />
40<br />
20<br />
0<br />
0 1 2 3 4 5 6<br />
time (h)<br />
TMC-SH DQ 25% D thiol 5%<br />
TMC-SH DQ 54% D thiol 6%<br />
TMC-SH DQ 25% D thiol 7%<br />
Figure 1.Thiol group content of TMC-SH DQ 25%, D thiol 5% in a concentration of 5 mg/ml, at pH 4.0, 5.0,<br />
6.2 and 7.4 at 37°C (A). Thiol group content of TMC-SH DQ 25%, D thiol 5%, TMC-SH DQ 54%, D thiol 6% and<br />
TMC-SH DQ 25%, D thiol 7% in a concentration of 5 mg/ml, at pH 7.4 at 37°C (B). Error bars indicate<br />
standard deviation of three independent samples.<br />
The pH-dependent oxidation of thiol groups of TMC-SH DQ 25%, D thiol 5%, w<strong>as</strong> studied<br />
(Figure 1A) at 37°C. This figure shows that at pH 4.0 no reduction of free thiols w<strong>as</strong> observed<br />
while at pH 7.4 almost quantitative oxidation w<strong>as</strong> achieved after 6 hours. For oxidation the<br />
thiol anion is needed and thus oxidation occurred much f<strong>as</strong>ter at higher pH. It w<strong>as</strong> further<br />
observed that an incre<strong>as</strong>e in DQ lead to a slight reduction in oxidation rate likely due to<br />
137
Chapter 6<br />
incre<strong>as</strong>ed charge repulsion [25, 43] but this had no effect on the extent of oxidation: after 6<br />
hours TMC-SH DQ 54%, D thiol 6% w<strong>as</strong> quantitatively oxidized (Figure 1B).<br />
Evaluation of TMC-SH on Cell Viability. Figure 2A shows the effect of TMC-SH on the<br />
mitochondrial activity (XTT) of Calu-3 cells. In agreement with previous data [19], an incre<strong>as</strong>e<br />
in DQ of the TMCs resulted in a decre<strong>as</strong>e in mitochondrial activity, especially at 10 mg/ml. The<br />
thiolated TMCs with a DQ of 25% were less cytotoxic at 1 mg/ml (p < 0.001, one-way ANOVA),<br />
<strong>as</strong> me<strong>as</strong>ured with the XTT <strong>as</strong>say, than the non-thiolated TMC with DQ of 30%. This might be<br />
due to the lower charge density of the TMC-SHs. Importantly, all TMCs were relatively nontoxic<br />
compared to linear PEI at concentrations < 1 mg/ml (p < 0.001). The influence of the<br />
various TMCs on membrane permeability determined with the LDH <strong>as</strong>say resulted in similar<br />
trends <strong>as</strong> obtained with the XTT <strong>as</strong>say (Figure 2B). Here, TMC-SH with a DQ of 54% showed<br />
less LDH rele<strong>as</strong>e than TMC DQ 56% at 1 mg/ml (p < 0.001). The LDH <strong>as</strong>say also indicated that<br />
linear PEI w<strong>as</strong> much more cytotoxic than the different TMCs (p < 0.001).<br />
# cells (x 10 3 )<br />
A<br />
25<br />
20<br />
15<br />
10<br />
5<br />
0<br />
#<br />
#<br />
* *<br />
TMC-SH DQ 25% D thiol 5%<br />
TMC-SH DQ 25% D thiol 7%<br />
TMC DQ 30%<br />
TMC-SH DQ 54% D thiol<br />
6%<br />
TMC DQ 56%<br />
linear PEI<br />
MEM<br />
10 mg/ml<br />
1.0 mg/ml<br />
0.1 mg/ml<br />
0.01 mg/ml<br />
LDH rele<strong>as</strong>e (%)<br />
B<br />
70<br />
60<br />
50<br />
40<br />
30<br />
20<br />
10<br />
0<br />
ο<br />
TMC-SH DQ 25% D thiol<br />
5%<br />
ο<br />
TMC-SH DQ 25% D thiol<br />
10 mg/ml<br />
1.0 mg/ml<br />
0.1 mg/ml<br />
0.01 mg/ml<br />
7%<br />
TMC DQ 30%<br />
TMC-SH DQ 54% D thiol<br />
#<br />
6%<br />
TMC DQ 56%<br />
linear PEI<br />
Figure 2. Effect of TMCs with various DQs and with or without thiolated units on the viability of Calu-3<br />
cells (XTT <strong>as</strong>say) at different concentrations (A). Asterics (*) indicate that no reliable XTT values were<br />
obtained due to gel-<strong>for</strong>mation of the TMC-SHs with a DQ of 25% at 10 mg/ml and there<strong>for</strong>e no bars are<br />
depicted. # indicate that significantly (p
Thiolated TMC <strong>for</strong> Covalently Stabilized Nanoparticles<br />
Recently, Hagenaars et al. demonstrated that n<strong>as</strong>al administration of TMC with a DQ of 68%<br />
in a concentration of 1.25 mg/ml induced only minimal local toxicity in mice while a solution of<br />
PEI resulted in moderate local toxicity [44].<br />
Covalently Stabilized Nanoparticles. Thiolated and non-thiolated TMCs were mixed with<br />
OVA-FITC and thiolated or non-thiolated hyaluronic acid in low ionic strength buffer (10 mM<br />
HEPES pH 7.4) to yield positively charged nanoparticles. Since the thiolation of hyaluronic acid<br />
leads to a reduction of negatively charged carboxylic acid moieties [33], more HA-SH than HA<br />
w<strong>as</strong> needed <strong>for</strong> ionic cross-linking to obtain particles with a size around 200 nm. Similarly,<br />
when TMC with a higher DQ w<strong>as</strong> used, larger amounts of negatively charged HA-SH or HA<br />
were needed to obtain nanoparticles (Table 2). The polydispersity index (PDI) w<strong>as</strong> below 0.2<br />
<strong>for</strong> all <strong>for</strong>mulations indicating fairly narrow size distributions. After <strong>for</strong>mulation, particles<br />
were incubated at 37°C <strong>for</strong> 3 hours while shaking to allow disulfide-bridge <strong>for</strong>mation. Table 2<br />
shows that the zeta-potential of the particles w<strong>as</strong> dependent on the DQ of the TMC: TMCs with<br />
a DQ of ~55% gave zeta-potentials of around +20 mV while TMCs with DQs of 25-30% resulted<br />
in particles with zeta-potentials of +16 mV. Thiolation of the polymers had no effect on the<br />
zeta-potential.<br />
Interestingly, when non-thiolated particles were dispersed in high ionic strength buffer (0.8<br />
M NaCl), DLS me<strong>as</strong>urements showed a substantial drop in counts indicating that nanoparticles<br />
disintegrated in this high ionic strength buffer. Similar destabilization in high ionic strength<br />
buffer w<strong>as</strong> observed <strong>for</strong> particles prepared with one thiolated polymer (e.g. TMC-SH or HA-SH)<br />
and one non-thiolated polymer (e.g. TMC or HA). Immediately after preparation, particles<br />
composed of TMC-SH and HA-SH also destabilized after dispersion in 0.8 M NaCl (results not<br />
shown). Importantly, nanoparticles made with TMC-SH and HA-SH and stabilized <strong>for</strong> 3 hours<br />
at 37°C showed almost no loss of DLS counts when incubated in this high ionic strength buffer<br />
indicating that the particles are held together by the <strong>for</strong>med covalent disulfide bonds between<br />
TMC-SH and HA-SH. All TMC:HA particles showed considerable <strong>as</strong>sociation of OVA-FITC in 10<br />
mM HEPES pH 7.4 (AE% up to 30%). The non-thiolated particles lost all their OVA-FITC in 0.8<br />
M NaCl where<strong>as</strong> the TMC-SH:HA-SH particles retained significant amounts of OVA-FITC<br />
(approx. 30-40% of originally <strong>as</strong>sociated protein) in high ionic strength buffer likely losing<br />
only surface bound protein.<br />
139
Chapter 6<br />
Table 2. Characteristics of thiolated and non-thiolated TMC-HA nanoparticles after incubation <strong>for</strong> 3<br />
hours at 37°C and dispersing in 10 mM HEPES pH 7.4 or in 0.8 M NaCl 10mM HEPES pH 7.4. Values are<br />
presented <strong>as</strong> the average of three samples ± standard deviation.<br />
TMC-SH 25%:<br />
HA-SH<br />
TMC:HA<br />
ratio<br />
(w/w)<br />
TMC 30%:HA 10:2<br />
TMC-SH 54%:<br />
HA-SH<br />
Zetapotential<br />
(mV)<br />
HEPES<br />
HEPES<br />
Size<br />
(nm)<br />
+<br />
0.8 M NaCl HEPES<br />
OVA-FITC<br />
Association Efficiency<br />
(%)<br />
+<br />
0.8 M NaCl<br />
10:2.4 + 13 (±0.8) 188 (±7) 192 (±4) 19 (±7) 8 (±2)<br />
+ 16.5<br />
(±0.3)<br />
224 (±11)<br />
Low<br />
scattering<br />
intensity<br />
22 (±9) -0.3 (±1)<br />
10:5.5 + 21 (±0.5) 191 (±3) 178 (±2) 25 (±7) 7 (±3)<br />
TMC 56%:HA 10:4 + 19 (±0.2) 197 (±3)<br />
Conclusion<br />
Low<br />
scattering<br />
intensity<br />
30 (±2) 0 (±3)<br />
In this paper a synthetic method <strong>for</strong> partial thiolation of TMC is presented resulting in D thiol<br />
up to 7% and DQs varying from 25-54% allowing investigation of thiolated TMCs with a<br />
potentially optimal DQ <strong>for</strong> mucosal vaccination and/or protein delivery or other applications<br />
such <strong>as</strong> DNA or siRNA delivery is presented. These thiolated TMCs readily <strong>for</strong>med disulfides at<br />
pH 7.4 and 37°C. Cell viability <strong>as</strong>says indicated a minor reduction in cytotoxicity on Calu-3 cells<br />
of TMC-SHs compared to non-thiolated TMCs with a similar DQ. The functionality of the thiol<br />
moieties w<strong>as</strong> confirmed by preparing covalently stabilized nanoparticles with thiolated<br />
hyaluronic acid. The opportunity to modify the remaining free thiols on the surface of the<br />
particles with e.g. PEG and targeting ligands <strong>as</strong> well <strong>as</strong> the option to prepare both negatively<br />
and positively charged particles by varying the TMC:HA ratio in the <strong>for</strong>mulation opens up wide<br />
possibilities <strong>for</strong> future pharmaceutical applications.<br />
Acknowledgement. This research w<strong>as</strong> partially per<strong>for</strong>med under the framework of <strong>TI</strong><br />
<strong>Pharma</strong> project number D5-106-1; Vaccine delivery: alternatives <strong>for</strong> conventional multiple<br />
injection vaccines. Roberta Censi is acknowledged <strong>for</strong> the synthesis and characterization of the<br />
thiolated hyaluronic acid.<br />
Supporting In<strong>for</strong>mation Available. Copies of 1 H-NMR spectra of TMC-COOH and TMC-SH<br />
<strong>for</strong> both DQs are depicted in the Supporting In<strong>for</strong>mation.<br />
140
Thiolated TMC <strong>for</strong> Covalently Stabilized Nanoparticles<br />
References<br />
1. M. Amidi, E. M<strong>as</strong>trobattista, W. Jiskoot, and W. E. Hennink. Chitosan-b<strong>as</strong>ed delivery systems<br />
<strong>for</strong> protein therapeutics and antigens. Adv Drug Deliv Rev 62: 59-82 (2010).<br />
2. B. A. Kumar, M. C. Varadaraj, and R. N. Tharanathan. Low Molecular Weight Chitosan-<br />
Preparation with the Aid of Pepsin, Characterization, and Its Bactericidal Activity.<br />
Biomacromolecules 8: 566-572 (2007).<br />
3. S. Mao, W. Sun, and T. Kissel. Chitosan-b<strong>as</strong>ed <strong>for</strong>mulations <strong>for</strong> delivery of DNA and siRNA.<br />
Adv Drug Deliv Rev 62: 12-27 (2010).<br />
4. M. Amidi, S. G. Romeijn, G. Borchard, H. E. Junginger, W. E. Hennink, and W. Jiskoot.<br />
Preparation and characterization of protein-loaded N-trimethyl chitosan nanoparticles <strong>as</strong> n<strong>as</strong>al<br />
delivery system. J Control Rele<strong>as</strong>e 111: 107-116 (2006).<br />
5. F. Chen, Z. R. Zhang, F. Yuan, X. Qin, M. Wang, and Y. Huang. In vitro and in vivo study of<br />
N-trimethyl chitosan nanoparticles <strong>for</strong> oral protein delivery. Int J Pharm 349: 226-233 (2008).<br />
6. F. Chen, Z. R. Zhang, and Y. Huang. Evaluation and modification of N-trimethyl chitosan<br />
chloride nanoparticles <strong>as</strong> protein carriers. Int J Pharm 336: 166-173 (2007).<br />
7. D. Snyman, J. H. Hamman, and A. F. Kotze. Evaluation of the mucoadhesive properties of N-<br />
trimethyl chitosan chloride. Drug Develop Ind Pharm 29: 61-69 (2003).<br />
8. M. M. Thanou, A. F. Kotze, T. Scharringhausen, H. L. Lueßen, A. G. De Boer, J. C. Verhoef,<br />
and H. E. Junginger. Effect of degree of quaternization of N-trimethyl chitosan chloride <strong>for</strong><br />
enhanced transport of hydrophilic compounds across intestinal Caco-2 cell monolayers. J<br />
Control Rele<strong>as</strong>e 64: 15-25 (2000).<br />
9. I. M. Van der Lubben, J. C. Verhoef, M. M. Fretz, O. Van, I. Mesu, G. Kersten, and H. E.<br />
Junginger. <strong>Trimethyl</strong> chitosan chloride (TMC) <strong>as</strong> a novel excipient <strong>for</strong> oral and n<strong>as</strong>al<br />
immunisation against diphtheria. S.T.P. <strong>Pharma</strong> Sciences 12: 235-242 (2002).<br />
10. B. Sayin, S. Somavarapu, X. W. Li, M. Thanou, D. Sesardic, H. O. Alpar, and S. Şenel. Mono-<br />
N-carboxymethyl chitosan (MCC) and N-trimethyl chitosan (TMC) nanoparticles <strong>for</strong> noninv<strong>as</strong>ive<br />
vaccine delivery. Int J Pharm 363: 139-148 (2008).<br />
11. B. Slütter, L. Plapied, V. Fievez, M. Alonso Sande, A. des Rieux, Y. J. Schneider, E. Van Riet,<br />
W. Jiskoot, and V. Préat. Mechanistic study of the adjuvant effect of biodegradable<br />
nanoparticles in mucosal vaccination. J Control Rele<strong>as</strong>e 138: 113-121 (2009).<br />
12. W. Boonyo, H. E. Junginger, N. Waranuch, A. Polnok, and T. Pitaksuteepong. Chitosan and<br />
trimethyl chitosan chloride (TMC) <strong>as</strong> adjuvants <strong>for</strong> inducing immune responses to ovalbumin in<br />
mice following n<strong>as</strong>al administration. J Control Rele<strong>as</strong>e 121: 168-175 (2007).<br />
13. G. Di Colo, S. Burgal<strong>as</strong>si, Y. Zambito, D. Monti, and P. Chetoni. Effects of different N-<br />
trimethyl chitosans on in vitro/in vivo ofloxacin transcorneal permeation. J Pharm Sci 93: 2851-<br />
2862 (2004).<br />
14. J. H. Hamman, M. Stander, and A. F. Kotze. Effect of the degree of quaternisation of N-<br />
trimethyl chitosan chloride on absorption enhancement: In vivo evaluation in rat n<strong>as</strong>al epithelia.<br />
Int J Pharm 232: 235-242 (2002).<br />
15. G. Sandri, M. C. Bonferoni, S. Rossi, F. Ferrari, S. Gibin, Y. Zambito, G. Di Colo, and C.<br />
Caramella. Nanoparticles b<strong>as</strong>ed on N-trimethylchitosan: Evaluation of absorption properties<br />
using in vitro (Caco-2 cells) and ex vivo (excised rat jejunum) models. Eur J Pharm Biopharm 65:<br />
68-77 (2007).<br />
16. T. Kean, S. Roth, and M. Thanou. <strong>Trimethyl</strong>ated chitosans <strong>as</strong> non-viral gene delivery vectors:<br />
Cytotoxicity and transfection efficiency. J Control Rele<strong>as</strong>e 103: 643 (2005).<br />
17. S. Mao, X. Shuai, F. Unger, M. Wittmar, X. Xie, and T. Kissel. Synthesis, characterization and<br />
cytotoxicity of poly(ethylene glycol)-graft-trimethyl chitosan block copolymers. Biomaterials 26:<br />
6343-6356 (2005).<br />
18. Z. Guo, R. Xing, S. Liu, Z. Zhong, X. Ji, L. Wang, and P. Li. The influence of molecular weight<br />
of quaternized chitosan on antifungal activity. Carbohydr Polym 71: 694-697 (2008).<br />
141
Chapter 6<br />
19. R. J. Verheul, M. Amidi, S. van der Wal, E. van Riet, W. Jiskoot, and W. E. Hennink. Synthesis,<br />
characterization and in vitro biological properties of O-methyl free N,N,N-trimethylated<br />
chitosan. Biomaterials 29: 3642-3649 (2008).<br />
20. R. J. Verheul, M. Amidi, M. J. van Steenbergen, E. van Riet, W. Jiskoot, and W. E. Hennink.<br />
Influence of the degree of acetylation on the enzymatic degradation and in vitro biological<br />
properties of trimethylated chitosans. Biomaterials 30: 3129-3135 (2009).<br />
21. A. Bernkop-Schnüch, A. Weithaler, K. Albrecht, and A. Greimel. Thiomers: Preparation and in<br />
vitro evaluation of a mucoadhesive nanoparticulate drug delivery system. Int J Pharm 317: 76-81<br />
(2006).<br />
22. V. Grabovac, D. Guggi, and A. Bernkop-Schnürch. Comparison of the mucoadhesive<br />
properties of various polymers. Adv Drug Deliv Rev 57: 1713-1723 (2005).<br />
23. T. M<strong>as</strong>uko, A. Minami, N. Iw<strong>as</strong>aki, T. Majima, S. I. Nishimura, and Y. C. Lee. Thiolation of<br />
chitosan. Attachment of proteins via thioether <strong>for</strong>mation. Biomacromolecules 6: 880-884 (2005).<br />
24. F. Meng, W. E. Hennink, and Z. Zhong. Reduction-sensitive polymers and bioconjugates <strong>for</strong><br />
biomedical applications. Biomaterials 30: 2180-2198 (2009).<br />
25. L. Yin, J. Ding, C. He, L. Cui, C. Tang, and C. Yin. Drug permeability and mucoadhesion<br />
properties of thiolated trimethyl chitosan nanoparticles in oral insulin delivery. Biomaterials 30:<br />
5691-5700 (2009).<br />
26. M. Amidi, S. G. Romeijn, J. C. Verhoef, H. E. Junginger, L. Bungener, A. Huckriede, D. J. A.<br />
Crommelin, and W. Jiskoot. N-<strong>Trimethyl</strong> chitosan (TMC) nanoparticles loaded with influenza<br />
subunit antigen <strong>for</strong> intran<strong>as</strong>al vaccination: Biological properties and immunogenicity in a mouse<br />
model. Vaccine 25: 144-153 (2007).<br />
27. S. Chadwick, C. Kriegel, and M. Amiji. Nanotechnology solutions <strong>for</strong> mucosal immunization.<br />
Adv Drug Deliv Rev 62: 394-407 (2010).<br />
28. B. Sayin, S. Somavarapu, X. W. Li, D. Sesardic, S. Şenel, and O. H. Alpar. TMC-MCC (Ntrimethyl<br />
chitosan-mono-N-carboxymethyl chitosan) nanocomplexes <strong>for</strong> mucosal delivery of<br />
vaccines. Eur J Pharm Sci 38: 362-369 (2009).<br />
29. D. V. Pergushov, H. M. Buchhammer, and K. Lunkwitz. Effect of a low-molecular-weight salt<br />
on colloidal dispersions of interpolyelectrolyte complexes. Colloid Polym Sci 277: 101-107 (1999).<br />
30. S. Boddohi, N. Moore, P. A. Johnson, and M. J. Kipper. Polysaccharide-b<strong>as</strong>ed polyelectrolyte<br />
complex nanoparticles from chitosan, heparin, and hyaluronan. Biomacromolecules 10: 1402-1409<br />
(2009).<br />
31. S. Shu, X. Wang, X. Zhang, X. Zhang, Z. Wang, and C. Li. Disulfide cross-linked biodegradable<br />
polyelectrolyte nanoparticles <strong>for</strong> the oral delivery of protein drugs. New J Chem 33: 1882-1887<br />
(2009).<br />
32. M. Thom<strong>as</strong>, J. J. Lu, Q. Ge, C. Zhang, J. Chen, and A. M. Klibanov. Full deacylation of<br />
polyethylenimine dramatically boosts its gene delivery efficiency and specificity to mouse lung.<br />
PNAS 102: 5679-5684 (2005).<br />
33. X. Z. Shu, Y. Liu, Y. Luo, M. C. Roberts, and G. D. Prestwich. Disulfide cross-linked<br />
hyaluronan hydrogels. Biomacromolecules 3: 1304-1311 (2002).<br />
34. Z. Guo, R. Xing, S. Liu, Z. Zhong, X. Ji, L. Wang, and P. Li. Antifungal properties of Schiff<br />
b<strong>as</strong>es of chitosan, N-substituted chitosan and quaternized chitosan. Carbohydr Res 342: 1329-<br />
1332 (2007).<br />
35. X. G. Chen and H. J. Park. Chemical characteristics of O-carboxymethyl chitosans related to<br />
the preparation conditions. Carbohydr Polym 53: 355-359 (2003).<br />
36. G. Lu, L. Kong, B. Sheng, G. Wang, Y. Gong, and X. Zhang. Degradation of covalently crosslinked<br />
carboxymethyl chitosan and its potential application <strong>for</strong> peripheral nerve regeneration.<br />
Eur Polym J 43: 3807-3818 (2007).<br />
37. A. Polnok, G. Borchard, J. C. Verhoef, N. Sarisuta, and H. E. Junginger. Influence of<br />
methylation process on the degree of quaternization of N-trimethyl chitosan chloride. Eur J<br />
Pharm Biopharms 57: 77-83 (2004).<br />
142
Thiolated TMC <strong>for</strong> Covalently Stabilized Nanoparticles<br />
38. A. B. Sieval, M. Thanou, A. F. Kotze, J. C. Verhoef, J. Brussee, and H. E. Junginger.<br />
Preparation and NMR characterization of highly substituted N-trimethyl chitosan chloride.<br />
Carbohydr Polym 36: 157-165 (1998).<br />
39. M. Lavertu, Z. Xia, A. N. Serreqi, M. Berrada, A. Rodrigues, D. Wang, M. D. Buschmann, and<br />
A. Gupta. A validated 1H NMR method <strong>for</strong> the determination of the degree of deacetylation of<br />
chitosan. J Pharm Biomed Anal 32: 1149-1158 (2003).<br />
40. J. Hombach, H. Hoyer, and A. Bernkop-Schnürch. Thiolated chitosans: Development and in<br />
vitro evaluation of an oral tobramycin sulphate delivery system. Eur J Pharm Sci 33: 1-8 (2008).<br />
41. X. Jiang, A. Van Der Horst, M. J. Van Steenbergen, N. Akeroyd, C. F. Van Nostrum, P. J.<br />
Schoenmakers, and W. E. Hennink. Molar-m<strong>as</strong>s characterization of cationic polymers <strong>for</strong> gene<br />
delivery by aqueous size-exclusion chromatography. Pharm Res 23: 595-603 (2006).<br />
42. D. A. Scudiero, R. H. Shoemaker, K. D. Paull, A. Monks, S. Tierney, T. H. Nofziger, M. J.<br />
Currens, D. Seniff, and M. R. Boyd. Evaluation of a soluble tetrazolium/<strong>for</strong>mazan <strong>as</strong>say <strong>for</strong> cell<br />
growth and drug sensitivity in culture using human and other tumor cell lines. Cancer Res 48:<br />
4827-4833 (1988).<br />
43. C. E. K<strong>as</strong>t and A. Bernkop-Schnürch. Thiolated polymers - thiomers: Development and in vitro<br />
evaluation of chitosan-thioglycolic acid conjugates. Biomaterials 22: 2345-2352 (2001).<br />
44. N. Hagenaars, M. Mania, P. de Jong, I. Que, R. Nieuwland, B. Slütter, H. Glansbeek, J. Heldens,<br />
H. van den Bosch, C. Löwik, E. Kaijzel, E. M<strong>as</strong>trobattista, and W. Jiskoot. Role of<br />
trimethylated chitosan (TMC) in n<strong>as</strong>al residence time, local distribution and toxicity of an<br />
intran<strong>as</strong>al influenza vaccine. J Control Rele<strong>as</strong>e 144: 17-24 (2010).<br />
143
Chapter 6<br />
144
Thiolated TMC <strong>for</strong> Covalently Stabilized Nanoparticles<br />
Supporting In<strong>for</strong>mation:<br />
TAILORABLE THIOLATED TRIMETHYL CHITOSANS FOR<br />
COVALENTLY STABILIZED NANOPAR<strong>TI</strong>CLES<br />
Rolf J. Verheul, Steffen van der Wal, Wim E. Hennink<br />
Contents:<br />
1H-NMR (D 2O, 80°C, 500MHz), TMC-COOH DQ 25%<br />
1H-NMR (D 2O, 80°C, 500MHz), TMC-COOH DQ 54%<br />
1H-NMR (D 2O, 80°C, 500MHz), TMC-SH DQ 25%, D thiol 7%<br />
1H-NMR (D 2O, 80°C, 500MHz), TMC-SH DQ 54%, D thiol 6%<br />
S1<br />
S2<br />
S3<br />
S4<br />
145
Chapter 6<br />
ppm (f1)<br />
5.50<br />
5.00<br />
4.50<br />
S1. 1 H-NMR of TMC-COOH DQ 25%.<br />
4.00<br />
3.50<br />
3.00<br />
2.50<br />
2.00<br />
ppm (f1)<br />
5.50<br />
5.00<br />
4.50<br />
4.00<br />
3.50<br />
3.00<br />
2.50<br />
2.00<br />
S2. 1 H-NMR of TMC-COOH DQ 54%.<br />
146
Thiolated TMC <strong>for</strong> Covalently Stabilized Nanoparticles<br />
ppm (f1)<br />
5.50<br />
5.00<br />
4.50<br />
4.00<br />
S3. 1 H-NMR of TMC-SH DQ 25%, D thiol 7%.<br />
3.50<br />
3.00<br />
2.50<br />
2.00<br />
ppm (f1)<br />
5.50<br />
5.00<br />
4.50<br />
4.00<br />
S4. 1 H-NMR of TMC-SH DQ 54%, D thiol 6%.<br />
3.50<br />
3.00<br />
2.50<br />
2.00<br />
147
CHAPTER 7<br />
COVALENTLY STABILIZED<br />
TRIMETHYL CHITOSAN-HYALURONIC ACID<br />
NANOPAR<strong>TI</strong>CLES FOR NASAL AND<br />
INTRADERMAL VACCINA<strong>TI</strong>ON<br />
Rolf J. Verheul, Bram Slütter, Suzanne M. Bal,<br />
Joke A. Bouwstra, Wim Jiskoot, Wim E. Hennink.<br />
Manuscript submitted
Chapter 7<br />
Abstract<br />
The physical stability of polyelectrolyte nanocomplexes composed of trimethyl chitosan<br />
(TMC) and hyaluronic acid (HA) is limited in physiological conditions. This may minimize the<br />
favorable adjuvant effects <strong>as</strong>sociated with particulate systems <strong>for</strong> n<strong>as</strong>al and intradermal<br />
immunization. There<strong>for</strong>e, covalently stabilized nanoparticles loaded with ovalbumin (OVA)<br />
were prepared with thiolated TMC and thiolated HA via ionic gelation followed by<br />
spontaneous disulfide <strong>for</strong>mation after incubation at pH 7.4 and 37 °C. Also, maleimide PEG w<strong>as</strong><br />
coupled to the remaining thiol-moieties on the particles to shield their surface charge.<br />
OVA-loaded TMC/HA nanoparticles had a size of around 250-350 nm, a positive zeta<br />
potential and OVA loading efficiencies up to 60%. Reacting the thiolated particles with<br />
maleimide PEG resulted in a slight reduction of zeta potential (from +7 to +4 mV) and a minor<br />
incre<strong>as</strong>e in particle size. Stabilized TMC-S-S-HA particles (PEGylated or not) showed superior<br />
stability in saline solutions compared to non-stabilized particles (composed of nonthiolated<br />
polymers) but readily disintegrated upon incubation in a saline buffer containing 10 mM<br />
dithiothreitol. In both the n<strong>as</strong>al and intradermal immunization study, OVA loaded stabilized<br />
TMC-S-S-HA particles demonstrated superior immunogenicity compared to non-stabilized<br />
particles (indicated by higher IgG titers). Intran<strong>as</strong>al, PEGylation completely abolished the<br />
beneficial effects of stabilization and it induced no enhanced immune responses against OVA<br />
after intradermal administration. In conclusion, stabilization of the TMC/HA particulate<br />
system greatly enhances the immunogenicity of OVA in n<strong>as</strong>al and intradermal vaccination.<br />
+<br />
TMC/HA<br />
+<br />
+<br />
-<br />
+<br />
+<br />
-<br />
+<br />
TMC-S-S-HA<br />
+<br />
TMC-S-S-HA PEG<br />
+<br />
- +<br />
intran<strong>as</strong>al<br />
- -<br />
intradermal<br />
vaccination<br />
log IgG titres<br />
5<br />
4<br />
3<br />
2<br />
1<br />
0<br />
intran<strong>as</strong>al<br />
prime<br />
boost<br />
6/8<br />
1/8<br />
1/8<br />
TMC/HA<br />
TMC-S-S-HA<br />
TMC-S-S-HA PEG<br />
log IgG titres<br />
intradermal<br />
5 prime boost<br />
4<br />
3<br />
2<br />
1<br />
0<br />
OVA<br />
TMC/HA<br />
TMC-S-S-HA<br />
TMC-S-S-HA PEG<br />
OVA i.m.<br />
150
Covalently Stabilized Nanoparticles <strong>for</strong> N<strong>as</strong>al and Intradermal Vaccination<br />
Introduction<br />
Most vaccines under development today are subunit vaccines b<strong>as</strong>ed on highly purified and<br />
well-characterized antigens derived from the respective pathogens against which one wants to<br />
protect. Although favorable because of their safety profile, these purified proteins generally<br />
show reduced immunogenicity compared to inactivated or attenuated pathogens [1-3].<br />
There<strong>for</strong>e, subunit vaccines have to be <strong>for</strong>mulated with adjuvants, i.e. delivery systems and/or<br />
immune potentiators that improve the immunogenicity of the antigens to elicit adequate,<br />
protective immune responses [4]. For instance, when co-<strong>for</strong>mulated in micro- or nanoparticles,<br />
<strong>for</strong>eign proteins are much more effective in eliciting immune responses than <strong>as</strong> plain protein<br />
solution [5-9], most likely because of the particle’s resemblance to the original pathogen, their<br />
multimeric antigen presentation and improved protection of the antigen against degradation<br />
[10]. Furthermore, particles are better taken up by antigen presenting cells (APCs), they may<br />
prolong the residence time of the antigen at the site of action and can co-deliver antigen and<br />
adjuvant to the same cell [8, 11, 12]. As a consequence, several types of particulate systems<br />
have been studied <strong>for</strong> vaccine delivery, including liposomes, oil-in-water emulsions, virus like<br />
particles, ISCOMs and polymeric carriers [1, 8, 9, 13].<br />
Alternative vaccine administration routes to conventional intramuscular immunization have<br />
been widely studied [2, 3, 5, 10, 14]. Mucosal vaccination offers several advantages over<br />
inv<strong>as</strong>ive (intramuscular or subcutaneous) immunization routes, like needle-free<br />
administration, potentially less adverse effects and the induction of local mucosal immune<br />
responses [15]. However, adequate antigen delivery via the n<strong>as</strong>al route is challenging, because<br />
of intran<strong>as</strong>al degradation and poor antigen uptake through the n<strong>as</strong>al epithelium. As trimethyl<br />
chitosan (TMC, a quaternized, water-soluble derivative of chitosan) and hyaluronic acid (HA)<br />
have muco-adhesive properties [16, 17], both polymers have been investigated in particulate<br />
<strong>for</strong>m in mucosal vaccine delivery [5, 18-23]. In n<strong>as</strong>al vaccine delivery, TMC nanoparticles have<br />
proven to have excellent adjuvant properties, most likely due to improved antigen delivery [5,<br />
24], but also immunostimulatory effects of TMC on monocyte derived dendritic cells (DCs)<br />
were observed [12, 23]. In these studies tripolyphosphate (TPP) w<strong>as</strong> used <strong>as</strong> a physical<br />
crosslinker to <strong>for</strong>m TMC nanoparticles via ionic gelation. Interestingly, results by Sayın et al.<br />
[25] showed that n<strong>as</strong>al immunization with tetanus toxoid loaded TMC:mono-N-carboxymethyl<br />
chitosan nanoparticles resulted in superior antibody titers compared to the TMC/TPP<br />
particles, indicating that combining TMC with an anionic polymer may further improve the<br />
adjuvant activity. However, the physical stability of such polyelectrolyte complexes may be<br />
limited in physiological conditions [26, 27] or at low pH [28].<br />
151
Chapter 7<br />
Intradermal (ID) immunization is another interesting immunization route. While muscular<br />
and subcutaneous tissues contain only limited numbers of APCs [29], skin tissue is abundant in<br />
DCs that play a central role in eliciting an immune response [12]. Recently, TMC nanoparticles<br />
were studied <strong>as</strong> intradermal vaccine carrier system, showing superior antibody titers against<br />
ovalbumin and diphtheria toxoid (DT) compared to the plain antigens. For DT encapsulated in<br />
TMC nanoparticles comparable immunopotency <strong>as</strong> subcutaneously administered DT-alum w<strong>as</strong><br />
observed [12].<br />
Recently, we developed covalently stabilized nanoparticles made from two oppositely<br />
charged, partially thiolated polymers, namely, thiolated trimethyl chitosan (TMC-SH) and<br />
thiolated hyaluronic acid (HA-SH) [30]. Polyelectrolyte complexes prepared with these<br />
polymers had a size of about 200-300 nm, a positive zeta potential and showed antigen<br />
encapsulation capacity up to 30%. The intermolecular disulfide bonds resulted in incre<strong>as</strong>ed<br />
stability of the TMC-S-S-HA particles in saline <strong>as</strong> compared to particles made with their<br />
nonthiolated counterparts (TMC/HA particles), which were kept together only by electrostatic<br />
interactions. Importantly, these covalently stabilized particles still allowed simple, aqueous<br />
and low stress preparation conditions <strong>as</strong> used with the preparation of conventional<br />
polyelectrolyte complexes [20, 27].<br />
It can be expected that both n<strong>as</strong>al and ID immunization antigen-loaded covalently stabilized<br />
TMC-S-S-HA particles may show enhanced immunogenicity compared to non-stabilized<br />
particles because superior particle integrity in the external environment may result in<br />
improved antigen delivery to, and activation of, APCs.<br />
Furthermore, remaining thiol groups present on the surface of TMC-S-S-HA particles allow<br />
post-particle modifications [31, 32]. Selective PEGylation of the free thiol moieties with PEGmaleimide<br />
may prove an interesting strategy <strong>as</strong> PEGylation of chitosan led to improved<br />
antibody titers in a n<strong>as</strong>al vaccination study with diphtheria toxoid possibly due to improved<br />
stability [1]. Furthermore, shielding of cationic charges may be beneficial in ID vaccination by<br />
limiting interactions with the negatively charged extracellular matrix. There<strong>for</strong>e, PEGylation<br />
may result in incre<strong>as</strong>ed mobility and uptake by APCs [33].<br />
In the present study we investigated covalently stabilized TMC-S-S-HA nanoparticles, with<br />
and without PEG coating, <strong>as</strong> potential n<strong>as</strong>al and intradermal vaccine delivery systems and<br />
compared them with non-stabilized TMC/HA nanoparticles. The particles contained ovalbumin<br />
<strong>as</strong> a model antigen. The <strong>for</strong>mulations were physico-chemically characterized and their stability<br />
in buffered saline and in the presence or absence of a reducing agent w<strong>as</strong> studied.<br />
152
Covalently Stabilized Nanoparticles <strong>for</strong> N<strong>as</strong>al and Intradermal Vaccination<br />
Furthermore, the extent and type of immune response elicited after n<strong>as</strong>al and intradermal<br />
administration of the <strong>for</strong>mulations in mice w<strong>as</strong> determined.<br />
Materials and Methods<br />
Materials. Chitosan with a residual degree of acetylation of 17% (determined with 1 H-NMR)<br />
and a number average (M n) and weight average molecular weight (M w) of 28 and 43 kDa,<br />
(determined with GPC-TD <strong>as</strong> described below), respectively, w<strong>as</strong> purch<strong>as</strong>ed from Primex<br />
(Siglufjodur, Iceland). Sodium borohydride, <strong>for</strong>maldehyde 37% (stabilized with methanol),<br />
glyoxilic acid monohydrate, cystamine dihydrochloride, dithiotreitol (DTT), 1-ethyl-3-(3-<br />
dimethylaminopropyl) carbodiimide HCl (EDC), L-cysteine HCl monohydrate, hen egg-white<br />
ovalbumin (OVA, grade V), deuterium oxide, sodium acetate, acetic acid (anhydrous), sodium<br />
hydroxide and hydrochloric acid were obtained from Sigma-Aldrich Chemical Co.<br />
Fluorescently labeled ovalbumin (OVA-FITC) w<strong>as</strong> obtained from Invitrogen (Breda, The<br />
Netherlands). Horseradish peroxid<strong>as</strong>e (HRP) conjugated goat anti-mouse IgG (γ chain specific),<br />
IgG1 (γ1 chain specific) and IgG2a (γ2a chain specific) were purch<strong>as</strong>ed from Southern Biotech<br />
Birmingham, USA). Chromogen 3, 3′, 5, 5′-tetramethylbenzidine (TMB) and the substrate buffer<br />
were purch<strong>as</strong>ed from Invitrogen. Iodomethane 99% stabilized with copper w<strong>as</strong> obtained from<br />
Acros Organics (Geel, Belgium). 5,5-dithio-bis-(2-nitrobenzoic acid) (Ellman’s reagent) w<strong>as</strong><br />
purch<strong>as</strong>ed from Pierce (Rock<strong>for</strong>d, IL, USA). Hyaluronic acid (HA, molecular weight 17 kDa <strong>as</strong><br />
determined by manufacturer) w<strong>as</strong> obtained from Lifecore (Ch<strong>as</strong>ka, USA). Methoxy<br />
polyethylene glycol (mPEG) maleimide (M w 2000) w<strong>as</strong> purch<strong>as</strong>ed from JenKem Technology<br />
(Beijing, China). TMCs with DQs of 30% (M n 33 kDa, M w 59 kDa) and 56% (M n 37 kDa, M w 78<br />
kDa) were synthesized and characterized <strong>as</strong> described previously [34]. Thiolated hyaluronic<br />
acid (HA-SH) with a M n of 15 kDa, PDI of 2.6 and a degree of thiolation of 21% w<strong>as</strong> synthesized<br />
and characterized according to the procedure by Shu et al.[35, 36]. All other chemicals used<br />
were of analytical grade.<br />
Synthesis and characterization of O-methyl free, trimethylated, partially thiolated<br />
chitosan (TMC-SH). Thiolated TMCs with different degrees of quaternization (DQ) were<br />
synthesized <strong>as</strong> described be<strong>for</strong>e [30]. The degree of thiolation and the total thiol content (free<br />
thiol moieties and disulfides) of the TMC-SHs were quantified with 5,5-dithio-bis-(2-<br />
nitrobenzoic acid) (DTNB, Ellman’s reagent) <strong>as</strong> described elsewhere [30, 37]. The M n and M w of<br />
the TMC-SHs were determined, <strong>as</strong> described previously [38], by GPC on a Viscotek system<br />
153
Chapter 7<br />
detecting refractive index, viscosity and light scattering. A Shodex OHPak SB-806 column (30<br />
cm) w<strong>as</strong> used with 0.3 M sodium acetate pH 4.4 (adjusted with acetic acid) <strong>as</strong> running buffer.<br />
Preparation of covalently stabilized nanoparticles with or without post-PEGylation.<br />
Covalently stabilized nanoparticles loaded with ovalbumin (OVA) were prepared with TMC-SH<br />
and HA-SH essentially <strong>as</strong> described be<strong>for</strong>e [30]. TMC-SH with a DQ 25% and D thiol 5% or a DQ<br />
of 54% and D thiol 6% were dissolved in 10 mM HEPES pH 7.4 at 1 mg/ml and subsequently<br />
mixed with 2.5 mg/ml ovalbumin in 10 mM HEPES pH 7.4 at a weight ratio of 10:1 (TMC-<br />
SH:OVA) under magnetic stirring. Then, HA-SH (0.5 mg/ml in 10 mM HEPES pH 7.4) w<strong>as</strong> added<br />
drop-wise yielding an opalescent nanoparticle dispersion which w<strong>as</strong> incubated at 37 °C <strong>for</strong> 3<br />
hours to allow disulfide <strong>for</strong>mation. In c<strong>as</strong>e of PEGylated particles, after 30 minutes of<br />
incubation at 37 °C mPEG maleimide (M w 2000 Da) w<strong>as</strong> added to the TMC-S-S-HA particles in a<br />
TMC:PEG w/w ratio of 2/1 and particles were incubated <strong>for</strong> an additional 2.5 hours at 37 °C.<br />
Remaining thiol-moieties on the surface of the particles were used to react with the maleimide<br />
group on the PEG at pH 7.4 (post PEGylation of the TMC-S-S-HA particles). Also, nanoparticles<br />
with non-thiolated TMC (DQ 30% or 56%), OVA and hyaluronic acid (HA) were prepared in a<br />
similar way to obtain ‘conventional’ particles only kept together by charge interactions.<br />
After incubation, the nanoparticle dispersions were centrifuged <strong>for</strong> 10 min at 10000 rpm<br />
(Biofuge Pico, PP1/97 #3324; Heraeus Instruments, Osterode, Germany) to remove the free<br />
polymers and unbound OVA. The obtained nanoparticle pellets were resuspended in 10 mM<br />
HEPES pH 7.4 and diluted to obtain a final OVA concentration of 0.5 mg/ml.<br />
Physical characterization of prepared nanoparticles. Particles were diluted in 10 mM<br />
HEPES until a slightly opalescent dispersion w<strong>as</strong> obtained. Particle size w<strong>as</strong> me<strong>as</strong>ured by<br />
dynamic light scattering (DLS) using a Malvern ALV CGS-3 (Malvern Instruments, Malvern,<br />
UK). DLS results are given <strong>as</strong> a z-average particle size diameter and a polydispersity index<br />
(PDI). The PDI can vary from 0 (indicating monodisperse particles) to 1 (indicating a<br />
completely heterodisperse system). The zeta potential of the nanoparticles w<strong>as</strong> me<strong>as</strong>ured in<br />
10 mM HEPES pH 7.4 using a Zet<strong>as</strong>izer Nano (Malvern Instruments, Malvern, UK).<br />
Determination of remaining thiol moieties on surface of particles. Remaining thiol<br />
groups on the surface of the particles after preparation were <strong>as</strong>sessed with 5,5-dithio-bis-(2-<br />
nitrobenzoic acid) (DTNB, Ellman’s reagent). DTNB w<strong>as</strong> dissolved in 0.1 M sodium phosphate,<br />
pH 8.0 containing 1 mM EDTA in a final concentration of 100 μg/ml. Then, 10 μl of particle<br />
154
Covalently Stabilized Nanoparticles <strong>for</strong> N<strong>as</strong>al and Intradermal Vaccination<br />
solution (corresponding to 0.5 mg/ml OVA) w<strong>as</strong> mixed with 400 μl of DTNB reagent solution.<br />
After incubation at room temperature <strong>for</strong> 10 min, the mixtures were centrifuged <strong>for</strong> 10 min at<br />
13000 rpm (Biofuge Pico, PP1/97 #3324; Heraeus Instruments, Osterode, Germany). The<br />
supernatant w<strong>as</strong> transferred in a microplate and the absorbance w<strong>as</strong> me<strong>as</strong>ured with a<br />
microplate reader at a wavelength of 405 nm (Bio-Rad Novapath, Hercules, Ca, USA). Cysteine<br />
standards were used to calculate the amount of remaining thiol moieties on the particles and<br />
non-incubated particles were used <strong>as</strong> a control.<br />
Loading efficiency of prepared nanoparticles. The protein loading w<strong>as</strong> determined by<br />
<strong>as</strong>saying the concentration of fluorescently labeled OVA (OVA-FITC) in the supernatant<br />
obtained after centrifugation of non-purified nanoparticles prepared in exactly the same way<br />
<strong>as</strong> OVA loaded particles (10 min at 10000 rpm; Biofuge Pico, PP1/97 #3324; Heraeus<br />
Instruments, Osterode, Germany). After resuspension of the particle pellet in 10 mM HEPES pH<br />
7.4, weakly <strong>as</strong>sociated OVA-FITC w<strong>as</strong> determined by centrifuging the purified particles <strong>for</strong> 10<br />
min 13000 rpm and then me<strong>as</strong>uring the OVA-FITC content in the supernatant. Fluorescence<br />
intensity w<strong>as</strong> determined using a Fluostar Optima with 488 nm <strong>as</strong> excitation wavelength and<br />
emission w<strong>as</strong> me<strong>as</strong>ured at 520 nm (BMG Labtech, Offenburg, Germany). The OVA loading<br />
efficiency (LE%) w<strong>as</strong> calculated using the following equation:<br />
LE% = (total OVA-FITC ― OVA-FITC in supernatant)/ total OVA-FITC x 100%<br />
Stability of nanoparticles To <strong>as</strong>sess the stability of the prepared nanoparticles, changes in<br />
size, PDI and number of nanoparticles (by optical density) were determined by dispersing the<br />
nanoparticles in physiological and high saline solutions buffered with 10 mM HEPES pH 7.4,<br />
with or without a disulfide reducing agent present. In detail, non-stabilized, stabilized and<br />
PEGylated nanoparticles prepared with TMC with high or low DQ <strong>as</strong> described be<strong>for</strong>e were<br />
diluted 50 fold in 10 mM HEPES pH 7.4 and dispersed 1:1 (v/v) in buffer or buffered saline<br />
solutions to final concentrations of 150 or 800 mM NaCl and with or without 10 mM DTT. After<br />
incubation <strong>for</strong> 30 minutes at room temperature, size and PDI were determined using DLS and<br />
samples were transferred into a 96-well plate to me<strong>as</strong>ure the optical density at 450 nm<br />
(microplate reader, Bio-Rad Novapath, Hercules, Ca, USA) to quantify the amount (turbidity) of<br />
nanoparticles in the various (saline) buffers [27]. The optical density of the particles me<strong>as</strong>ured<br />
in 10 mM HEPES w<strong>as</strong> set at 100%.<br />
155
Chapter 7<br />
N<strong>as</strong>al immunization. Groups of eight female Balb/c mice (Charles River, Boxmeer, The<br />
Netherlands), 6-8 weeks old, received two n<strong>as</strong>al doses of 10 μg OVA in the different<br />
<strong>for</strong>mulations with an interval of 3 weeks. Injections of 20 μg plain OVA were administered<br />
intramuscularly (i.m.) <strong>as</strong> control (n=5). For n<strong>as</strong>al administration, <strong>for</strong>mulations were applied in<br />
a volume of 10 μl 10 mM HEPES pH 7.4, 5 μl per nostril in two sessions. Blood samples were<br />
taken 3 weeks after the prime and booster dose. After sacrificing the animals, spleens were<br />
harvested and n<strong>as</strong>al w<strong>as</strong>hes collected.<br />
Intradermal immunization. The immunogenicity of intradermally (ID) administered OVA<br />
<strong>for</strong>mulations w<strong>as</strong> <strong>as</strong>sessed in female Balb/c mice (Charles River, Boxmeer, The Netherlands),<br />
6-8 weeks old. The mice were vaccinated twice with 3 weeks intervals. Groups of five mice<br />
were injected ID with a Hamilton syringe equipped with a 30-Gauge needle. A total volume of<br />
30 μl containing 2 μg OVA dissolved in PBS or co-<strong>for</strong>mulated with TMC/HA nanoparticles w<strong>as</strong><br />
injected into the abdominal skin under anaesthesia (by intraperitoneal injection of 150 mg/kg<br />
ketamine and 10 mg/kg xylazine). As a control 20 μg plain OVA w<strong>as</strong> injected i.m.. Blood<br />
samples were collected from the tail vein 3 weeks after the prime dose and three weeks after<br />
boost vaccination the mice were sacrificed. Just be<strong>for</strong>e euthan<strong>as</strong>ia total blood w<strong>as</strong> collected<br />
from the femoral artery. Blood samples were collected in MiniCollect® tubes (Greiner Bio-one,<br />
Alphen a/d Rijn, The Netherlands) till clot <strong>for</strong>mation and centrifuged 10 minutes at 10,000 g to<br />
obtain cell-free sera.<br />
Determination of serum IgG, IgG1, IgG2a and secretory IgA. Micro titer plates (Nunc,<br />
Roskilde, Denmark) were coated with OVA, by incubation of 1 μg/ml OVA in 40 mM sodium<br />
carbonate buffer pH 9.6 <strong>for</strong> 24 hours at 4°C. To reduce <strong>as</strong>pecific binding, wells were blocked<br />
with 1% (w/v) BSA in PBS <strong>for</strong> 1 hour at room temperature. After extensive w<strong>as</strong>hing with PBS<br />
serial dilutions of serum ranging from 10 to 2*10 6 were applied, where<strong>as</strong> n<strong>as</strong>al w<strong>as</strong>hes were<br />
added undiluted. After incubation <strong>for</strong> 1.5 hours at room temperature and extensive w<strong>as</strong>hing,<br />
OVA specific antibodies were detected using HRP conjugated goat anti mouse IgG, IgG1, IgG2a<br />
or IgA (1 hour room temperature) and by incubating with 0.1 mg/ml TMB and 30 μg/ml H 2O 2<br />
in 110 mM sodium acetate buffer pH 5.5 <strong>for</strong> 15 min at room temperature. Reaction w<strong>as</strong><br />
stopped with 2 M H 2SO 2 and absorbance w<strong>as</strong> determined at 450 nm with an EL808 microplate<br />
reader (Bio-Tek Instruments, Bad Friedrichshall, Germany).<br />
156
Covalently Stabilized Nanoparticles <strong>for</strong> N<strong>as</strong>al and Intradermal Vaccination<br />
Statistical analysis. Statistical analysis w<strong>as</strong> per<strong>for</strong>med with Prism 5 <strong>for</strong> Windows<br />
(Graphpad, San Diego, USA). Data are presented <strong>as</strong> mean ± standard deviation. Statistical<br />
significance w<strong>as</strong> determined by a one way analysis of variance (ANOVA) with a Bonferroni<br />
post-test.<br />
Results and discussion<br />
Characteristics of nanoparticle <strong>for</strong>mulations. The structural characteristics of the<br />
synthesized TMC-SHs are summarized in Table 1.<br />
D thiol M n (kDa) M w (kDa)<br />
Table 1. Characteristics of the synthesized thiolated TMCs.<br />
DQ DA<br />
Total thiol<br />
amount<br />
Amount of<br />
free –SH<br />
(μmol/g) a (μmol/g) a<br />
TMC-SH<br />
High DQ<br />
TMC-SH<br />
Low DQ<br />
54% 17% 529 (±2) 283 (±21) 6% 66 174<br />
25% 17% 478 (±10) 236 (±13) 5% 62 215<br />
a values given <strong>as</strong> mean (± standard deviation) (n=3).<br />
Thiolated and nonthiolated TMCs with high and low DQ were mixed with OVA and thiolated<br />
or nonthiolated hyaluronic acid in low ionic strength buffer (10 mM HEPES pH 7.4) to yield<br />
positively charged nanoparticles. As w<strong>as</strong> observed be<strong>for</strong>e [30], more HA-SH than HA w<strong>as</strong><br />
needed <strong>for</strong> ionic cross-linking to obtain particles and, similarly, when TMC with a higher DQ<br />
w<strong>as</strong> used, larger amounts of negatively charged HA-SH or HA were needed to obtain<br />
nanoparticles (Table 2). Additionally, mPEG maleimide w<strong>as</strong> added to thiolated particles after<br />
30 min of incubation at 37 °C to obtain PEGylation via selective maleimide coupling to the<br />
remaining free thiol moieties on the surface of the particle. Importantly, when mPEG<br />
maleimide w<strong>as</strong> added to the thiolated particles without this prior incubation step, complete<br />
disintegration of the particles w<strong>as</strong> observed; likely this is due to shielding of the chargeinteractions<br />
between TMC-SH and HA-SH by PEGylation. Apparently, some disulfide crosslinking<br />
connecting TMC-SH and HA-SH is required to allow PEGylation while maintaining<br />
particle integrity. The particle <strong>for</strong>mulations were incubated at 37 °C <strong>for</strong> 3 hours while shaking<br />
to allow disulfide-bridge <strong>for</strong>mation. Previously we demonstrated that after 3 hours incubation<br />
at 37 °C, pH 7.4 more than 80% of the thiol groups of TMC-SH polymers were converted into<br />
157
Chapter 7<br />
disulfides [30]. Nanoparticle <strong>for</strong>mulations prepared with TMC with a low DQ resulted in higher<br />
OVA loading (50-60%) than <strong>for</strong>mulations with a high DQ (ca. 30%). However, loading<br />
efficiencies were highly dependent on TMC:crosslinker ratio and the type of TMC used [39], so<br />
this may explain the differences between the <strong>for</strong>mulations. Post PEGylation of the particles had<br />
no effect on the <strong>as</strong>sociation efficiency.<br />
Table 2. Characteristics of nonthiolated, thiolated and post-PEGylated nanoparticles in 10 mM HEPES<br />
pH 7.4. Values are presented <strong>as</strong> the average of three samples ± standard deviation.<br />
OVA-FITC<br />
TMC:OVA TMC:HA TMC:mPEG<br />
Zeta<br />
Loading<br />
Size<br />
ratio ratio ratio<br />
potential<br />
Efficiency<br />
(nm)<br />
(w/w) (w/w) (w/w)<br />
(mV)<br />
(%)<br />
TMC-S-S-HA<br />
+ 7.3<br />
10:1 10:3.7 n.a. 58 (±1)<br />
338 (±73)<br />
low DQ<br />
TMC-S-S-HA<br />
low DQ PEG<br />
TMC/HA<br />
low DQ<br />
TMC-S-S-HA<br />
high DQ<br />
TMC-S-S-HA<br />
high DQ PEG<br />
TMC/HA<br />
high DQ<br />
n.a. = not applicable<br />
10:1 10:3.7 10:5 59 (±1)<br />
(±1.2)<br />
+ 4.4<br />
(±1.2)<br />
352 (±27)<br />
10:1 10:2.8 n.a. 52 (±1) + 13 (±1.6) 321 (±39)<br />
10:1 10:9.3 n.a. 30 (±3) + 16 (±0.7) 226 (±16)<br />
10:1 10:9.3 10:5 30 (±2) + 10 (±0.6) 251 (±16)<br />
10:1 10:6 n.a. 29 (±1) + 18 (±0.9) 290 (±26)<br />
Free polymers and un<strong>as</strong>sociated OVA were removed by centrifugation and resuspension of<br />
the obtained particle-pellet in 10 mM HEPES pH 7.4. All particle <strong>for</strong>mulations showed limited<br />
burst-rele<strong>as</strong>e of <strong>as</strong>sociated OVA (< 10%) after resuspension in 10 mM HEPES pH 7.4. The zeta<br />
potential of the non PEGylated particles varied from +18 mV to +7 mV (Table 2) most likely<br />
due to differences in the DQ of TMC and amount of cross-linker used [39]. Importantly,<br />
PEGylation of the thiolated particles reduced the zeta potential <strong>for</strong> both PEGylated<br />
<strong>for</strong>mulations compared to the non PEGylated TMC-S-S-HA particles (from +7 mV to +4 mV and<br />
+16 mV to +10 mV <strong>for</strong> TMC-S-S-HA with low and high DQ, respectively. When mPEG maleimide<br />
w<strong>as</strong> added to nonthiolated particles, no reduction in zeta potential w<strong>as</strong> observed, indicating<br />
that available thiol groups are required <strong>for</strong> PEGylation. Similarly, when mPEG maleimide w<strong>as</strong><br />
added to TMC-S-S-HA particles after three hours of incubation at 37 °C, no significant<br />
reduction in zeta potential w<strong>as</strong> observed, indicating that almost all thiols were converted into<br />
disulfides. This w<strong>as</strong> confirmed with Ellman’s <strong>as</strong>say <strong>as</strong> neglectable amounts of remaining thiol<br />
moieties on the surface of TMC-S-S-HA particles were found (< 4 % of amount of thiol groups<br />
on the surface of non-incubated particles). DLS analysis showed that the size of the <strong>for</strong>med<br />
158
Covalently Stabilized Nanoparticles <strong>for</strong> N<strong>as</strong>al and Intradermal Vaccination<br />
nanoparticle varied from 226 to 352 nm. TMCs with a low DQ yielded particles with a bigger<br />
size and higher PDI compared those prepared with TMC with a high DQ (Figure 1A). Also,<br />
PEGylation resulted in a slight incre<strong>as</strong>e in particle diameter by 15-25 nm.<br />
Particle stability in saline and in presence of a disulfide-reducing agent. In general, the<br />
stability of polyelectrolyte complexes is dependent on the type of polymers used and may be<br />
limited solutions of physiological salt concentrations [26, 27]. We there<strong>for</strong>e studied the<br />
physical stability of nonthiolated, thiolated and PEGylated nanoparticles in buffers with<br />
different sodium chloride concentrations. Figure 1A shows that non-stabilized TMC/HA high<br />
DQ particles demonstrated a large incre<strong>as</strong>e in size in 150 mM NaCl implying aggregation while<br />
an incre<strong>as</strong>e in polydispersity is seen in 800 mM NaCl. On the other hand, the particles made of<br />
thiolated polymers, with and without PEG-coating showed that particle size and PDI w<strong>as</strong><br />
hardly affected by salt in the buffer. The turbidity of particle solutions w<strong>as</strong> me<strong>as</strong>ured at 450<br />
nm and w<strong>as</strong> used to quantify the amount of particles remaining in physiological or high saline<br />
conditions and with or without DTT (Figure 1B). After dispersion of the non-stabilized<br />
TMC/HA particles in 150 mM NaCl, a reduction of more than 75% in optical density w<strong>as</strong><br />
observed which became even more dramatic in 800 mM NaCl. In contr<strong>as</strong>t, stabilized TMC-S-S-<br />
HA particles showed a much less reduction in OD 450 in 800 mM NaCl indicating superior<br />
stability in high ionic strength buffers. However, the TMC-S-S-HA low DQ particles showed less<br />
stability in 800 mM NaCl than w<strong>as</strong> observed in the previous study where almost no reduction<br />
in DLS counts w<strong>as</strong> observed [30]. This may be attributed to the lower degree of thiolation and<br />
lower molecular weight of the HA-SH used in this study (21 vs 52% and M n of 14 vs. 39 kDa).<br />
Both altered characteristics will likely decre<strong>as</strong>e the probability of the <strong>for</strong>mation of disulfideb<strong>as</strong>ed<br />
networks between TMC-SH and HA-SH polymers in a nanoparticle. Interestingly, TMC-S-<br />
S-HA high DQ showed less reduction of OD 450 nm in the saline solutions compared to TMC-S-<br />
S-HA low DQ. This higher stability of TMC-S-S-HA high DQ particles may be due to the higher<br />
amount of HA-SH incorporated in the particles (Table 2); more HA-SH present in a particle<br />
incre<strong>as</strong>es the chances of TMC-S-S-HA <strong>for</strong>mation which is crucial <strong>for</strong> the particle stability [30].<br />
When a disulfide-reducing agent (10 mM DTT) w<strong>as</strong> present in a physiological saline solution,<br />
the turbidity of the stabilized particles dramatically decre<strong>as</strong>ed and this shows that the<br />
stabilization of the particles is due to disulfide <strong>for</strong>mation. PEGylation of the thiolated<br />
nanoparticles had no effect on the stability in presence of salt and particles were still sensitive<br />
to DTT.<br />
159
Chapter 7<br />
OD 450 nm (% in HEPES)<br />
B<br />
100<br />
75<br />
50<br />
25<br />
0<br />
TMC/HA low DQ<br />
TMC-S-S-HA low DQ<br />
TMC-S-S-HA low DQ PEG<br />
TMC/HA high DQ<br />
TMC-S-S-HA high DQ<br />
TMC-S-S-HA high DQ PEG<br />
HEPES<br />
150 mM NaCl<br />
800 mM NaCl<br />
150 mM NaCl<br />
+10 mM DTT<br />
Figure 1. Physical characteristics and stability of the nanoparticle <strong>for</strong>mulations in presence of<br />
physiological or high saline concentrations after 30 minutes incubation at room temperature; Z-average<br />
diameter (bars) and PDI (dots) (A). Optical density at 450 nm (OD 450 nm) of the nanoparticle<br />
<strong>for</strong>mulations in presence of physiological or high saline concentrations and with or without DTT (B).<br />
Error bars represent the standard deviation (n=3).<br />
160
Covalently Stabilized Nanoparticles <strong>for</strong> N<strong>as</strong>al and Intradermal Vaccination<br />
Immunization studies. N<strong>as</strong>al and ID immunization studies were carried out with<br />
nanoparticles prepared with TMC with a low DQ because we previously demonstrated that the<br />
DQ of TMC plays only a minor role in its (n<strong>as</strong>al) adjuvanticity [22].<br />
Serum antigen-specific antibodies after n<strong>as</strong>al vaccination. The OVA specific total IgG<br />
antibody titers elicited after n<strong>as</strong>al administration of the antigen and the various TMC-HA<br />
<strong>for</strong>mulations were determined three weeks after prime and boost vaccination (Figure 2). After<br />
prime vaccination hardly any IgG titers were detected <strong>for</strong> all n<strong>as</strong>al <strong>for</strong>mulations. Interestingly,<br />
after the boost vaccination the stabilized particles showed significantly higher total IgG titers<br />
compared to the non-stabilized or the PEGylated system and six out of eight mice had a<br />
detectable IgG titer after intran<strong>as</strong>al vaccination with the stabilized particles. These results<br />
suggest that stabilization does indeed <strong>as</strong> anticipated improve the adjuvanticity of the TMC-HA<br />
nanoparticulate system and that PEGylation of these particles abolishes this effect. The<br />
administered dose w<strong>as</strong> apparently too low to (significantly) discriminate between plain OVA<br />
and the TMC-S-S-HA particles and this also led to hardly detectable IgG1 and IgG2a after boost<br />
vaccination and secretory IgA (sIgA) levels in the n<strong>as</strong>al w<strong>as</strong>h (results not shown).<br />
5<br />
prime<br />
boost<br />
5/5<br />
10 log IgG titres<br />
4<br />
3<br />
2<br />
1<br />
3/8<br />
1/8<br />
6/8<br />
* *<br />
1/8<br />
0<br />
OVA<br />
TMC/HA low DQ<br />
TMC-S-S-HA low DQ<br />
TMC-S-S-HA low DQ PEG<br />
OVA i.m.<br />
Figure 2. Antigen-specific total IgG antibody titers after n<strong>as</strong>al vaccination with OVA <strong>for</strong>mulations. Error<br />
bars indicate standard deviation (n=8 (n<strong>as</strong>al) or 5 (i.m.)). * p
Chapter 7<br />
Serum antigen-specific total IgG after intradermal vaccination. The antigen specific total<br />
IgG antibody titers elicited after ID administration of the various OVA <strong>for</strong>mulations are shown<br />
in Figure 3. After prime vaccination the stabilized particles (with or without PEGylation)<br />
showed significantly higher antibody titers compared to plain OVA (up to 75-fold) and the nonstabilized<br />
TMC/HA particles (up to 20-fold). Interestingly, the non-stabilized nanoparticles did<br />
not significantly improve immune responses compared to plain OVA. Also, after boost<br />
vaccination the stabilized systems showed incre<strong>as</strong>ed total IgG antibody titers compared to<br />
OVA, in contr<strong>as</strong>t to the non-stabilized <strong>for</strong>mulations. The immunogenicity of ID administered<br />
PEGylated stabilized particles w<strong>as</strong> similar to that of the non-PEGylated ones. These results<br />
indicate that stabilization, but not PEGylation of TMC/HA particles is essential <strong>for</strong> their ID<br />
adjuvanticity.<br />
10 log IgG titres<br />
5<br />
4<br />
3<br />
2<br />
1<br />
prime<br />
boost<br />
** **<br />
OOO<br />
OO<br />
***<br />
***<br />
***<br />
0<br />
OVA<br />
TMC/HA low DQ<br />
TMC-S-S-HA low DQ<br />
TMC-S-S-HA low DQ PEG<br />
OVA i.m.<br />
Figure 3. Antigen-specific total IgG antibody titers after intradermal vaccination with OVA <strong>for</strong>mulations.<br />
Error bars indicate standard deviation (n=5). *** p
Covalently Stabilized Nanoparticles <strong>for</strong> N<strong>as</strong>al and Intradermal Vaccination<br />
immune response. Log IgG1/IgG2a ratio <strong>for</strong> each individual mouse did not reveal a significant<br />
shifting of the type of immune response compared to plain OVA (results not shown). This is in<br />
line with previous results observed in an ID vaccination study with TMC/OVA/TPP particles<br />
[12].<br />
10 log IgG1/IgG2a titres<br />
4<br />
3<br />
2<br />
1<br />
0<br />
OVA<br />
IgG1 IgG2a<br />
*<br />
*<br />
TMC/HA low DQ<br />
TMC-S-S-HA low DQ<br />
TMC-S-S-HA low DQ PEG<br />
Figure 4. Antigen-specific IgG1 and IgG2a antibody titers after boost intradermal vaccination with OVA<br />
and nanoparticle <strong>for</strong>mulations. Error bars indicate standard deviation (n=5). * p
Chapter 7<br />
2). However, previously we found that <strong>for</strong> n<strong>as</strong>al vaccination with TMC-coated whole<br />
inactivated influenza virus, similar differences in zeta potential did not result in altered<br />
adjuvant effects [22]. This indicates that it is unlikely that the lower zeta potential of the<br />
stabilized particles is responsible <strong>for</strong> the observed beneficial effects in the present n<strong>as</strong>al<br />
vaccination study. Thus, it can be concluded that the stabilization of TMC-S-S-HA nanoparticles<br />
resulted in improved immunogenicity, however, the exact mechanism(s) need to be<br />
determined.<br />
PEGylation of the stabilized particles inhibited the beneficial effects of stabilization (n<strong>as</strong>al<br />
administration), while <strong>for</strong> ID administration the PEGylated particles had similar effects <strong>as</strong> the<br />
non-PEGylated ones. This indicates that PEGylation of particles decre<strong>as</strong>ed n<strong>as</strong>al antigen<br />
delivery. Although it h<strong>as</strong> been suggested that PEGylation of TMC improves muco-adhesion<br />
[41], possibly, PEGylation will also hinder electrostatic interactions between the cationic<br />
particle and the epithelial barrier thereby reducing particle transport over epithelial cells [42]<br />
or particle uptake by microfold (M)-cells. Van den Berg et al. showed that PEGylation of<br />
cationic polyplexes improves DNA transfection efficiencies after ID tattooing [33]. Although<br />
PEGylation may improve mobility in the intracellular matrix, reduced electrostatic interactions<br />
with DCs may lead to lower uptake [43]. Better understanding of these conflicting effects may<br />
result in optimal use of PEGylation <strong>for</strong> ID vaccination. Importantly, we showed that post<br />
particle modifications of the stabilized TMC-S-S-HA polyelectrolyte system are possible and<br />
this opens up a variety of options such <strong>as</strong> specific targeting towards DCs or M-cells.<br />
Conclusion<br />
In this paper we showed that stabilized nanoparticles can be prepared using TMC-SH with<br />
high and low DQ together with HA-SH and that these particles can be post PEGylated. These<br />
particles showed adequate OVA <strong>as</strong>sociation efficiency, preserved their particle integrity under<br />
saline conditions but readily disintegrated when a disulfide reducing agent w<strong>as</strong> present.<br />
Stabilized particles showed enhanced adjuvanticity in n<strong>as</strong>al and ID vaccination compared to<br />
non-stabilized particles. PEGylation abolished the beneficial effects of stabilization in n<strong>as</strong>al<br />
vaccine administration and showed similar immunogenicity <strong>as</strong> stabilized particles in ID<br />
immunization. Our results imply that the stabilized TMC-S-S-HA nanoparticles <strong>for</strong>m a highly<br />
versatile and promising vaccine carrier system while also offering options <strong>for</strong> post particle<br />
modifications. Further studies should be per<strong>for</strong>med to elaborate the exact mechanism by<br />
which stabilization results in improved immunogenicity.<br />
164
Covalently Stabilized Nanoparticles <strong>for</strong> N<strong>as</strong>al and Intradermal Vaccination<br />
Acknowledgement. This research w<strong>as</strong> partially per<strong>for</strong>med under the framework of <strong>TI</strong><br />
<strong>Pharma</strong> project number D5-106-1; Vaccine delivery: alternatives <strong>for</strong> conventional multiple<br />
injection vaccines.<br />
165
Chapter 7<br />
References<br />
1. N. Csaba, M. Garcia-Fuentes, and M. J. Alonso. Nanoparticles <strong>for</strong> n<strong>as</strong>al vaccination. Adv Drug<br />
Deliv Rev 61: 140-157 (2009).<br />
2. N. Mishra, A. K. Goyal, S. Tiwari, R. Paliwal, S. R. Paliwal, B. Vaidya, S. Mangal, M. Gupta, D.<br />
Dube, A. Mehta, and S. P. Vy<strong>as</strong>. Recent advances in mucosal delivery of vaccines: Role of<br />
mucoadhesive/biodegradable polymeric carriers. Exp Opin Ther Patents 20: 661-679 (2010).<br />
3. D. T. O'Hagan and R. Rappuoli. Novel approaches to vaccine delivery. Pharm Res 21: 1519-1530<br />
(2004).<br />
4. V. E. Schijns. Immunological concepts of vaccine adjuvant activity. Curr Opin Immunol 12: 456-<br />
463 (2000).<br />
5. M. Amidi, E. M<strong>as</strong>trobattista, W. Jiskoot, and W. E. Hennink. Chitosan-b<strong>as</strong>ed delivery systems<br />
<strong>for</strong> protein therapeutics and antigens. Adv Drug Deliv Rev 62: 59-82 (2010).<br />
6. L. Feng, X. J. Zhou, and X. R. Qi. Preparation, rele<strong>as</strong>e and immunogenicity evaluation of<br />
HBsAg-PLGA microspheres. Beijing da xue xue bao. Yi xue ban = J Peking University. Health sci 37:<br />
527-531 (2005).<br />
7. D. T. O'Hagan, M. Singh, and J. B. Ulmer. Microparticle-b<strong>as</strong>ed technologies <strong>for</strong> vaccines.<br />
Methods 40: 10 (2006).<br />
8. A. C. Rice-Ficht, A. M. Aren<strong>as</strong>-Gamboa, M. M. Kahl-McDonagh, and T. A. Ficht. Polymeric<br />
particles in vaccine delivery. Curr Opin Microbiol 13: 106-112 (2010).<br />
9. S. D. Xiang, A. Scholzen, G. Minigo, C. David, V. Apostolopoulos, P. L. Mottram, and M.<br />
Plebanski. Pathogen recognition and development of particulate vaccines: Does size matter?<br />
Methods 40: 1-9 (2006).<br />
10. M. R. Neutra and P. A. Kozlowski. Mucosal vaccines: The promise and the challenge. Nat Rev<br />
Immunol 6: 148-158 (2006).<br />
11. T. Akagi, X. Wang, T. Uto, M. Baba, and M. Ak<strong>as</strong>hi. Protein direct delivery to dendritic cells<br />
using nanoparticles b<strong>as</strong>ed on amphiphilic poly(amino acid) derivatives. Biomaterials 28: 3427-<br />
3436 (2007).<br />
12. S. M. Bal, B. Slütter, E. van Riet, A. C. Kruithof, Z. Ding, G. F. A. Kersten, W. Jiskoot, and J.<br />
A. Bouwstra. Efficient induction of immune responses through intradermal vaccination with N-<br />
trimethyl chitosan containing antigen <strong>for</strong>mulations. J Control Rele<strong>as</strong>e 142: 374-383 (2010).<br />
13. A. Pattani, V. B. Patravale, L. Panicker, and P. D. Potdar. Immunological Effects and<br />
Membrane Interactions of Chitosan Nanoparticles. Mol Pharm 6: 345-352 (2009).<br />
14. P. Kuo-Haller, Y. Cu, J. Blum, J. A. Appleton, and W. M. Saltzman. Vaccine Delivery by<br />
Polymeric Vehicles in the Mouse Reproductive Tract Induces Sustained Local and Systemic<br />
Immunity. Mol Pharm (2010).<br />
15. B. Slütter, N. Hagenaars, and W. Jiskoot. Rational design of n<strong>as</strong>al vaccines. J Drug Target 16: 1-17<br />
(2008).<br />
16. V. Grabovac, D. Guggi, and A. Bernkop-Schnürch. Comparison of the mucoadhesive<br />
properties of various polymers. Adv Drug Deliv Rev 57: 1713-1723 (2005).<br />
17. D. Snyman, J. H. Hamman, and A. F. Kotze. Evaluation of the mucoadhesive properties of N-<br />
trimethyl chitosan chloride. Drug Develop Ind Pharm 29: 61-69 (2003).<br />
18. P. A. O' Hagan D. Use of hyaluronic acid polymers <strong>for</strong> mucosal delivery of vaccine and<br />
adjuvants., 2004.<br />
19. M. Amidi, H. C. Pellikaan, H. Hirschberg, A. H. de Boer, D. J. A. Crommelin, W. E. Hennink,<br />
G. Kersten, and W. Jiskoot. Diphtheria toxoid-containing microparticulate powder <strong>for</strong>mulations<br />
<strong>for</strong> pulmonary vaccination: Preparation, characterization and evaluation in guinea pigs. Vaccine<br />
25: 6818-6829 (2007).<br />
20. M. Amidi, S. G. Romeijn, J. C. Verhoef, H. E. Junginger, L. Bungener, A. Huckriede, D. J. A.<br />
Crommelin, and W. Jiskoot. N-<strong>Trimethyl</strong> chitosan (TMC) nanoparticles loaded with influenza<br />
subunit antigen <strong>for</strong> intran<strong>as</strong>al vaccination: Biological properties and immunogenicity in a mouse<br />
model. Vaccine 25: 144-153 (2007).<br />
166
Covalently Stabilized Nanoparticles <strong>for</strong> N<strong>as</strong>al and Intradermal Vaccination<br />
21. W. Boonyo, H. E. Junginger, N. Waranuch, A. Polnok, and T. Pitaksuteepong. Chitosan and<br />
trimethyl chitosan chloride (TMC) <strong>as</strong> adjuvants <strong>for</strong> inducing immune responses to ovalbumin in<br />
mice following n<strong>as</strong>al administration. J Control Rele<strong>as</strong>e 121: 168-175 (2007).<br />
22. N. Hagenaars, R. J. Verheul, I. Mooren, P. H. J. L. F. de Jong, E. M<strong>as</strong>trobattista, H. L.<br />
Glansbeek, J. G. M. Heldens, H. van den Bosch, W. E. Hennink, and W. Jiskoot. Relationship<br />
between structure and adjuvanticity of N,N,N-trimethyl chitosan (TMC) structural variants in a<br />
n<strong>as</strong>al influenza vaccine. J Control Rele<strong>as</strong>e 140: 126-133 (2009).<br />
23. B. Slütter, L. Plapied, V. Fievez, M. Alonso Sande, A. des Rieux, Y. J. Schneider, E. Van Riet,<br />
W. Jiskoot, and V. Préat. Mechanistic study of the adjuvant effect of biodegradable<br />
nanoparticles in mucosal vaccination. J Control Rele<strong>as</strong>e 138: 113-121 (2009).<br />
24. N. Hagenaars, M. Mania, P. de Jong, I. Que, R. Nieuwland, B. Slütter, H. Glansbeek, J. Heldens,<br />
H. van den Bosch, C. Löwik, E. Kaijzel, E. M<strong>as</strong>trobattista, and W. Jiskoot. Role of<br />
trimethylated chitosan (TMC) in n<strong>as</strong>al residence time, local distribution and toxicity of an<br />
intran<strong>as</strong>al influenza vaccine. J Control Rele<strong>as</strong>e 144: 17-24 (2010).<br />
25. B. Sayin, S. Somavarapu, X. W. Li, D. Sesardic, S. Şenel, and O. H. Alpar. TMC-MCC (Ntrimethyl<br />
chitosan-mono-N-carboxymethyl chitosan) nanocomplexes <strong>for</strong> mucosal delivery of<br />
vaccines. Eur J Pharm Sci 38: 362-369 (2009).<br />
26. S. Boddohi, N. Moore, P. A. Johnson, and M. J. Kipper. Polysaccharide-b<strong>as</strong>ed polyelectrolyte<br />
complex nanoparticles from chitosan, heparin, and hyaluronan. Biomacromol 10: 1402-1409<br />
(2009).<br />
27. D. V. Pergushov, H. M. Buchhammer, and K. Lunkwitz. Effect of a low-molecular-weight salt<br />
on colloidal dispersions of interpolyelectrolyte complexes. Coll Polym Sci 277: 101-107 (1999).<br />
28. A. Bernkop-Schnürch, A. Weithaler, K. Albrecht, and A. Greimel. Thiomers: Preparation and in<br />
vitro evaluation of a mucoadhesive nanoparticulate drug delivery system. Int J Pharm 317: 76-81<br />
(2006).<br />
29. E. Raz, D. A. Carson, S. E. Parker, T. B. Parr, A. M. Abai, G. Aichinger, S. H. Gromkowski, M.<br />
Singh, D. Lew, M. A. Yankauck<strong>as</strong>, S. M. Baird, and G. H. Rhodes. Intradermal gene<br />
immunization: The possible role of DNA uptake in the induction of cellular immunity to<br />
viruses. PNAS 91: 9519-9523 (1994).<br />
30. R. J. Verheul, S. van der Wal, and W. E. Hennink. <strong>Tailorable</strong> Thiolated <strong>Trimethyl</strong> <strong>Chitosans</strong> <strong>for</strong><br />
Covalently Stabilized Nanoparticles. Biomacromol (2010).<br />
31. T. M<strong>as</strong>uko, A. Minami, N. Iw<strong>as</strong>aki, T. Majima, S. I. Nishimura, and Y. C. Lee. Thiolation of<br />
chitosan. Attachment of proteins via thioether <strong>for</strong>mation. Biomacromol 6: 880-884 (2005).<br />
32. B. Slütter, P. C. Soema, Z. Ding, R. Verheul, W. Hennink, and W. Jiskoot. Conjugation of<br />
ovalbumin to trimethyl chitosan improves immunogenicity of the antigen. J Control Rele<strong>as</strong>e 143:<br />
207-214 (2010).<br />
33. J. H. van den Berg, K. Oosterhuis, W. E. Hennink, G. Storm, L. J. van der Aa, J. F. J.<br />
Engbersen, J. B. A. G. Haanen, J. H. Beijnen, T. N. Schumacher, and B. Nuijen. Shielding the<br />
cationic charge of nanoparticle-<strong>for</strong>mulated dermal DNA vaccines is essential <strong>for</strong> antigen<br />
expression and immunogenicity. J Control Rele<strong>as</strong>e 141: 234-240 (2010).<br />
34. R. J. Verheul, M. Amidi, S. van der Wal, E. van Riet, W. Jiskoot, and W. E. Hennink. Synthesis,<br />
characterization and in vitro biological properties of O-methyl free N,N,N-trimethylated<br />
chitosan. Biomaterials 29: 3642-3649 (2008).<br />
35. R. Censi, P. J. Fieten, P. di Martino, W. E. Hennink, and T. Vermonden. In Situ Forming<br />
Hydrogels by Tandem Thermal Gelling and Michael Addition Reaction between<br />
Thermosensitive Triblock Copolymers and Thiolated Hyaluronan. Macromol 43: 5771-5778<br />
(2010).<br />
36. X. Z. Shu, Y. Liu, Y. Luo, M. C. Roberts, and G. D. Prestwich. Disulfide cross-linked<br />
hyaluronan hydrogels. Biomacromol 3: 1304-1311 (2002).<br />
37. J. Hombach, H. Hoyer, and A. Bernkop-Schnürch. Thiolated chitosans: Development and in<br />
vitro evaluation of an oral tobramycin sulphate delivery system. Eur J Pharm Sci 33: 1-8 (2008).<br />
167
Chapter 7<br />
38. X. Jiang, A. Van Der Horst, M. J. Van Steenbergen, N. Akeroyd, C. F. Van Nostrum, P. J.<br />
Schoenmakers, and W. E. Hennink. Molar-m<strong>as</strong>s characterization of cationic polymers <strong>for</strong> gene<br />
delivery by aqueous size-exclusion chromatography. Pharm Res 23: 595-603 (2006).<br />
39. A. Jintapattanakit, V. B. Junyapr<strong>as</strong>ert, S. Mao, J. Sitterberg, U. Bakowsky, and T. Kissel. Peroral<br />
delivery of insulin using chitosan derivatives: A comparative study of polyelectrolyte<br />
nanocomplexes and nanoparticles. Int J Pharm 342: 240-249 (2007).<br />
40. L. Yin, J. Ding, C. He, L. Cui, C. Tang, and C. Yin. Drug permeability and mucoadhesion<br />
properties of thiolated trimethyl chitosan nanoparticles in oral insulin delivery. Biomaterials 30:<br />
5691-5700 (2009).<br />
41. A. Jintapattanakit, V. B. Junyapr<strong>as</strong>ert, and T. Kissel. The role of mucoadhesion of trimethyl<br />
chitosan and PEGylated trimethyl chitosan nanocomplexes in insulin uptake. J Pharm Sci 98:<br />
4818-4830 (2009).<br />
42. S. Mao, O. Germershaus, D. Fischer, T. Linn, R. Schnepf, and T. Kissel. Uptake and transport<br />
of PEG-graft-trimethyl-chitosan copolymer-insulin nanocomplexes by epithelial cells. Pharm Res<br />
22: 2058 (2005).<br />
43. Y. Sheng, Y. Yuan, C. Liu, X. Tao, X. Shan, and F. Xu. In vitro macrophage uptake and in vivo<br />
biodistribution of PLA-PEG nanoparticles loaded with hemoglobin <strong>as</strong> blood substitutes: Effect<br />
of PEG content. J Mat Sci: Mat Med 20: 1881-1891 (2009).<br />
168
CHAPTER 8<br />
SUMMARY AND FUTURE PERSPEC<strong>TI</strong>VES
Summary and Future Perspectives<br />
Summary<br />
Active vaccination h<strong>as</strong> proven to be the most (cost) effective tool in the fight against<br />
infectious dise<strong>as</strong>es. Vaccines are usually made of live attenuated or inactivated pathogens (e.g.<br />
viruses or bacteria) or purified immunogenic protein(conjugate)s derived from these<br />
pathogens. Nowadays, most vaccines are administered via parenteral injection however, the<br />
risk <strong>for</strong> contaminated needles and need <strong>for</strong> trained personnel have risen interest <strong>for</strong><br />
alternative immunization routes.<br />
Chapter 1 discusses the (dis)advantages of intramuscular and alternative administration<br />
routes, in particular intran<strong>as</strong>al immunization that allows relatively simple, needle-free<br />
administration, reduces the need <strong>for</strong> trained professionals and may, importantly, elicit both<br />
mucosal and systemic immune responses. On the other hand, antigen degradation and poor<br />
delivery to antigen presenting cells (APCs) are major drawbacks of n<strong>as</strong>al vaccination. As live<br />
attenuated vaccines raise considerable safety issues, (n<strong>as</strong>al) vaccine development now mainly<br />
focuses on purified, well-characterized antigenic proteins. However, these subunit vaccines are<br />
generally less immunogenic and need potent adjuvant(system)s to elicit an adequate immune<br />
response. The use of muco-adhesive polymers like N,N,N-trimethylchitosan (TMC), a partially<br />
quaternized, water-soluble chitosan derivative, can enhance the immune responses against<br />
antigens, presumably by incre<strong>as</strong>ing the n<strong>as</strong>al residence time and/or improving the contact<br />
area between the antigen and the mucosal surface. TMC’s chemical structure can vary in the<br />
degree of quaternization (DQ), giving the polymer its cationic charge, but also in extent of O-<br />
methylation (DOM), degree of acetylation (DAc) and polymer molecular weight. Tailorability of<br />
these structural elements will allow better understanding of the contribution of these moieties<br />
to the physico-chemical and biological properties of TMC. Additionally, the introduction of side<br />
groups such <strong>as</strong> thiol-moieties may further be applied to improve and optimize TMC’s<br />
properties.<br />
Chitosan and its derivatives are much more effective in eliciting immune responses in microor<br />
nanoparticulate <strong>for</strong>m than <strong>as</strong> plain polymer solution. Ionic gelation b<strong>as</strong>ed upon electrostatic<br />
interactions is a simple, commonly used method to yield nanoparticulate systems. The<br />
complexation between the positively charged TMC and oppositely charged (macro)molecules<br />
added drop-wise under stirring in low ionic strength buffer results in the spontaneous<br />
<strong>for</strong>mation of nanoparticles. However, the physico-chemical stability and the immunogenicity of<br />
these antigen-loaded complexes are dependent on the characteristics of the crosslinker used,<br />
and the currently applied carrier-systems may be sub-optimal. Furthermore, although TMC’s<br />
171
Chapter 8<br />
mode of action is usually attributed to its muco-adhesive and penetration enhancing<br />
properties, detailed knowledge of the mechanism behind the favorable adjuvant<br />
characteristics is lacking (e.g., the effect of TMC on APCs is currently unknown). Concluding,<br />
many opportunities <strong>for</strong> optimizing TMC structure and TMC-b<strong>as</strong>ed nanoparticles remain and<br />
mechanistic insights into the mode of action of TMC have to be obtained.<br />
The aim of this thesis is to develop synthetic routes to synthesize TMC structural variants in<br />
a controllable and tailorable manner and introduce substitutions such <strong>as</strong> thiol-moieties that<br />
may further improve TMC’s properties. In this way, structure-activity relationships can be<br />
investigated exploiting in vitro <strong>as</strong>says and in in vivo (n<strong>as</strong>al) vaccination studies.<br />
Chapter 2 challenges the current synthetic method to synthesize TMC that is <strong>as</strong>sociated with<br />
several side-reactions such <strong>as</strong> O-methylation and chain scission. Since these side reactions may<br />
affect the polymer characteristics, there is a need <strong>for</strong> TMCs without O-methylation and<br />
disparities in chain lengths while varying the DQ. In this chapter, O-methyl free TMCs with<br />
varying DQs were successfully synthesized by using a two-step method. First, chitosan w<strong>as</strong><br />
quantitatively dimethylated using <strong>for</strong>mic acid and <strong>for</strong>maldehyde. Then, in presence of an<br />
excess amount of iodomethane, TMC w<strong>as</strong> obtained with different DQs (22-68%) by varying the<br />
reaction time. TMC obtained by this two-step method showed no detectable O-methylation<br />
( 1 H-NMR) and a slight incre<strong>as</strong>e in molecular weight with incre<strong>as</strong>ing DQ (GPC), implying that no<br />
chain scission occurred during synthesis. The solubility in aqueous solutions at pH 7 of O-<br />
methyl free TMC with DQ < 22% w<strong>as</strong> less <strong>as</strong> compared to O-methylated TMC with the same DQ.<br />
On the other hand, O-methyl free TMC with DQ > 30% had an excellent aqueous solubility. On<br />
Caco-2, cells O-methyl free TMCs demonstrated a larger decre<strong>as</strong>e in trans-epithelial electrical<br />
resistance (TEER) than O-methylated TMCs. Also, with incre<strong>as</strong>ing DQ, an incre<strong>as</strong>e in<br />
cytotoxicity (MTT) and membrane permeability (LDH) w<strong>as</strong> observed. This chapter clearly<br />
demonstrates that the DQ <strong>as</strong> well <strong>as</strong> O-methylation substantially influence the physicochemical<br />
and in vitro biological properties of TMC.<br />
The influence of another structural variant of TMC, the degree of acetylation (DAc), on in<br />
vitro degradation and biological properties w<strong>as</strong> evaluated in Chapter 3. TMCs with a DAc<br />
ranging from 11 to 55% were synthesized by using a three-step method. First, chitosan w<strong>as</strong><br />
partially re-acetylated using acetic anhydride followed by quantitative dimethylation using<br />
<strong>for</strong>maldehyde and sodium borohydride. Then, in presence of an excess amount of<br />
iodomethane, TMC w<strong>as</strong> synthesized. The TMCs obtained by this method showed neither<br />
172
Summary and Future Perspectives<br />
detectable O-methylation nor loss in acetyl groups ( 1 H-NMR) and a slight incre<strong>as</strong>e in molecular<br />
weight (GPC) with incre<strong>as</strong>ing degree of substitution, implying that no chain scission occurred<br />
during synthesis. The extent of lysozyme-catalyzed degradation of TMC, and that of its<br />
precursors chitosan and dimethyl chitosan, w<strong>as</strong> highly dependent on the DAc; polymers with<br />
the highest DAc showed the largest decre<strong>as</strong>e in molecular weight. On Caco-2 cells, TMCs with a<br />
high DAc (~50%), a DQ of around 44% and with or without O-methylated groups, were not<br />
able to open tight junctions in the trans-epithelial electrical resistance (TEER) <strong>as</strong>say. This in<br />
contr<strong>as</strong>t to TMCs (both O-methylated and O-methyl free; concentration 2.5 mg/ml) with a<br />
similar DQ but a lower DAc which were able to reduce the TEER with 30 and 70%,<br />
respectively. Additionally, TMCs with a high DAc (~50%) demonstrated no cell toxicity (MTT,<br />
LDH rele<strong>as</strong>e) up to a concentration of 10 mg/ml. This chapter shows that the degree of N-<br />
acetylation dramatically influences the enzymatic degradation and in vitro biological<br />
properties of TMC.<br />
The results of Chapter 2 and 3 demonstrate that the DQ, DOM and DAc all influence the in<br />
vitro biological properties of TMC. There<strong>for</strong>e we investigated the influence of the structural<br />
properties of TMC on its adjuvanticity in an in vivo intran<strong>as</strong>al (i.n.) immunization study. In<br />
Chapter 4, TMCs with varying degrees of quaternization (DQ, 22-86%), O-methylation (DOM,<br />
0-76%) and acetylation (DAc 9-54%) were <strong>for</strong>mulated with whole inactivated influenza virus<br />
(WIV). Simple mixing of the TMCs with WIV at a 1:1 (w/w) ratio resulted in comparable<br />
positively charged nanoparticles, indicating coating of the negatively charged WIV with TMC.<br />
The amount of free TMC in solution w<strong>as</strong> comparable <strong>for</strong> all TMC-WIV <strong>for</strong>mulations. After i.n.<br />
immunization of mice with WIV and TMC-WIV on day 0 and 21, all TMC-WIV <strong>for</strong>mulations<br />
induced stronger total IgG, IgG1 and IgG2a/c responses than WIV alone, except WIV<br />
<strong>for</strong>mulated with re-acetylated TMC with a DAc of 54% and a DQ of 44% (TMC-RA44). No<br />
significant differences in antibody titers were observed <strong>for</strong> TMCs that varied in DQ or DOM,<br />
indicating that these structural characteristics play a minor role in their adjuvant properties.<br />
TMC with a DQ of 56% (TMC56) <strong>for</strong>mulated with WIV at a ratio of 5:1 (w/w) resulted in<br />
significantly lower IgG2a/c:IgG1 ratio’s compared to TMC56 mixed in ratios of 0.2:1 and 1:1,<br />
implying a shift towards a Th2 type immune response. Challenge of vaccinated mice with<br />
aerosolized virus demonstrated protection <strong>for</strong> all TMC-WIV <strong>for</strong>mulations with the exception of<br />
TMC-RA44-WIV. This chapter demonstrates that coating of WIV with TMCs strongly enhances<br />
the immunogenicity and induced protection after i.n. vaccination with WIV. The adjuvant<br />
173
Chapter 8<br />
properties of TMCs <strong>as</strong> i.n. adjuvant are strongly decre<strong>as</strong>ed by re-acetylation of TMC, where<strong>as</strong><br />
the DQ and DOM hardly affect the adjuvanticity of TMC.<br />
The aim of Chapter 5A w<strong>as</strong> to elucidate the re<strong>as</strong>on <strong>for</strong> the lack of adjuvanticity of reacetylated<br />
TMC (TMC-RA) by comparing TMC-RA (degree of acetylation 54%) with TMC<br />
(degree of acetylation 17%) at six potentially critical steps in the induction of an immune<br />
response after intran<strong>as</strong>al (i.n.) administration in mice: chemical stability of the polymer in<br />
murine n<strong>as</strong>al w<strong>as</strong>hings, local i.n. distribution of WIV, n<strong>as</strong>al residence time of WIV, cellular<br />
uptake of WIV by epithelial cells, transport of WIV by epithelial cells, and capacity of the<br />
<strong>for</strong>mulation to induce maturation of murine bone marrow derived dendritic cells (DCs). TMC-<br />
RA w<strong>as</strong> degraded in a n<strong>as</strong>al w<strong>as</strong>h to a slightly larger extent than TMC. The local i.n. distribution<br />
and n<strong>as</strong>al clearance were similar <strong>for</strong> both TMC types. Fluorescently labeled WIV w<strong>as</strong> taken up<br />
more efficiently by Calu-3 cells when <strong>for</strong>mulated with TMC-RA compared to TMC and both<br />
TMCs significantly reduced transport of WIV over a Calu-3 monolayer. Murine bone-marrow<br />
derived dendritic cell activation w<strong>as</strong> similar <strong>for</strong> plain WIV, <strong>as</strong> well <strong>as</strong> <strong>for</strong> WIV <strong>for</strong>mulated with<br />
TMC-RA or TMC. The inferior adjuvant effect of TMC-RA over that of TMC might be caused by a<br />
slightly lower stability of TMC-RA in the n<strong>as</strong>al cavity, rather than by any of the other factors<br />
studied in this paper.<br />
Interestingly, N-acetylated glucosamine units or GlcNAcs present in TMC have been<br />
described to bind several human C-type lectins, a family of lectins involved in the human<br />
innate immune response. There<strong>for</strong>e the effect of TMC and re-acetylated TMC with a degree of<br />
acetylation of 54% (TMC-RA) on the uptake and maturation of human dendritic cells (DCs) w<strong>as</strong><br />
<strong>as</strong>sessed in Chapter 5B using whole inactivated influenza virus (WIV) <strong>as</strong> antigen. Studies on<br />
monocyte-derived human DCs indicated that the uptake of TMC(-RA) coated WIV w<strong>as</strong> slightly<br />
lower than plain WIV. TMC-RA, <strong>as</strong> a solution or when <strong>for</strong>mulated with WIV, induced much<br />
stronger DC maturation, <strong>as</strong> me<strong>as</strong>ured with CD86 expression, than any of the other<br />
<strong>for</strong>mulations. Also, only TMC-RA(-WIV) induced high IL-10, TNF-α and IL-12p40 and IL-12p70<br />
rele<strong>as</strong>e by DCs. Since both IL-10 and IL-12p70 levels were elevated, no polarization towards a<br />
Th1 or Th2 type immune response could be established. Concluding, TMC-RA h<strong>as</strong> strong<br />
immuno-stimulatory effects in vitro on human monocyte derived dendritic cells which<br />
indicates that the degree of N-acetylation is critical <strong>for</strong> the adjuvant effect of TMC on human<br />
DCs, but not on mice DCs.<br />
174
Summary and Future Perspectives<br />
Introduction of thiol-moieties will further broaden the potential pharmaceutical applications<br />
of TMC by enhancing its muco-adhesive properties, allowing further chemical derivatization<br />
reactions via reducible disulfide-bridges and opening up possibilities <strong>for</strong> covalently linked<br />
nanoparticles together with a thiolated anionic polymer. In Chapter 6, a novel four-step<br />
method is presented to synthesize partially thiolated TMC with tailorable degrees of<br />
quaternization and thiolation. First, chitosan w<strong>as</strong> partially N-carboxylated with glyoxylic acid<br />
and sodium borohydride. Next, the remaining amines were quantitatively dimethylated with<br />
<strong>for</strong>maldehyde and sodium borohydride and then quaternized with iodomethane in N-<br />
methylpyrrolidone. Subsequently, these partially carboxylated TMCs, dissolved in water, were<br />
reacted with cystamine at pH 5.5 using EDC <strong>as</strong> coupling agent. After addition of dithiothreitol<br />
<strong>for</strong> disulfide reduction and dialysis, thiolated TMCs were obtained varying in degree of<br />
quaternization (25-54%) and degree of thiolation (5-7%) <strong>as</strong> determined with 1 H-NMR and<br />
Ellman’s <strong>as</strong>say. Gel permeation chromatography with light scattering detection indicated<br />
limited intermolecular crosslinking. All thiolated TMCs showed rapid oxidation to yield<br />
disulfide crosslinked TMC at pH 7.4 while the thiolated polymers were rather stable at pH 4.0.<br />
Using Calu-3 cells, XTT and LDH cell viability tests showed a slight reduction in cytotoxicity <strong>for</strong><br />
thiolated TMCs <strong>as</strong> compared to the non-thiolated polymers with similar DQs. Positively<br />
charged nanoparticles loaded with fluorescently labeled ovalbumin were made from thiolated<br />
TMCs and thiolated hyaluronic acid. The stability of these particles w<strong>as</strong> confirmed in 0.8 M<br />
NaCl, in contr<strong>as</strong>t to particles made from non-thiolated polymers which dissociated under these<br />
conditions demonstrating that the particles were held together by intermolecular disulfide<br />
bonds.<br />
As shown in Chapter 6, the physical stability of polyelectrolyte nanoparticles composed of<br />
trimethyl chitosan (TMC) and hyaluronic acid (HA) is limited in physiological conditions. This<br />
may adversely affect the favorable adjuvant effects of particulate systems <strong>for</strong> n<strong>as</strong>al and<br />
intradermal immunization. There<strong>for</strong>e, in Chapter 7 the effect of covalent stabilization of<br />
antigen-loaded TMC/HA nanoparticles by disulfides on their immunogenicity is evaluated.<br />
Furthermore, remaining thiols on the surface of these particles allow PEGylation which may<br />
result in additional beneficial effects by hindering interactions with the extracellular matrix<br />
(intradermal) and/or enhanced muco-adhesion (intran<strong>as</strong>al). Ovalbumin (OVA) loaded,<br />
covalently stabilized nanoparticles were prepared with thiolated TMC and thiolated HA via<br />
ionic gelation followed by spontaneous disulfide <strong>for</strong>mation after incubation at pH 7.4 and 37°C.<br />
175
Chapter 8<br />
Also, PEG maleimide w<strong>as</strong> coupled to the remaining thiol-moieties on the particles to shield the<br />
surface charge.<br />
TMC/HA nanoparticles had a size of around 250-350 nm, a positive zeta potential and OVA<br />
<strong>as</strong>sociation efficiencies up to 60%. PEGylation resulted in a slight reduction of zeta potential<br />
and a minor incre<strong>as</strong>e in particle size. Stabilized TMC-S-S-HA particles (PEGylated or not)<br />
showed superior stability in saline solutions compared to non-stabilized TMC/HA particles<br />
(composed of nonthiolated polymers), but readily disintegrated when a disulfide reducing<br />
agent w<strong>as</strong> introduced. In both the n<strong>as</strong>al and intradermal immunization study, OVA-loaded<br />
stabilized TMC-S-S-HA particles demonstrated superior immunogenicity compared to nonstabilized<br />
particles. For intran<strong>as</strong>al immunization, PEGylation completely abolished the positive<br />
effects of stabilization and it had no additional effect when used <strong>for</strong> intradermal vaccination. In<br />
conclusion, stabilization of the TMC/HA particulate system enhances its immunogenicity in<br />
n<strong>as</strong>al and intradermal vaccination, however, PEGylation of these stabilized particles had no<br />
beneficial effects.<br />
The novel synthetic methods to obtain structural variants of TMC presented in this thesis<br />
allowed tailorability of the degrees of quaternization and acetylation and excluded the<br />
introduction of other alterations such <strong>as</strong> O-methylation and polymer chain scission.<br />
Additionally, thiol-moieties were introduced in a controllable manner. This tailorability of TMC<br />
provided the possibility to properly establish structure-activity relationships in in vitro<br />
biological <strong>as</strong>says and in in vivo (n<strong>as</strong>al) vaccination studies. Also, mechanistic insight into the<br />
mode of action of TMC w<strong>as</strong> acquired by investigating several potentially crucial steps in n<strong>as</strong>al<br />
vaccination. Finally, a promising new, covalently stabilized, polymeric carrier system w<strong>as</strong><br />
introduced b<strong>as</strong>ed upon the <strong>for</strong>mation of intracellularly degradable disulfides.<br />
Discussion and Future perspectives<br />
TMC-b<strong>as</strong>ed particulate systems <strong>for</strong> n<strong>as</strong>al vaccine delivery. Although already synthesized<br />
from chitosan in the mid-80s by Muzzarelli [1] and later by Domard [2], trimethyl chitosan<br />
(TMC) h<strong>as</strong> not been used <strong>for</strong> mucosal vaccination until the beginning of this millennium [3].<br />
Even more recently, in 2007, Amidi et al. demonstrated <strong>for</strong> the first time the potency of TMCtripolyphosphate<br />
ionically crosslinked nanoparticulate systems in an intran<strong>as</strong>al vaccination<br />
study with influenza subunit antigen [4]. Although this system showed promising results <strong>as</strong><br />
176
Summary and Future Perspectives<br />
n<strong>as</strong>al vaccine delivery system, it w<strong>as</strong> anticipated that much could be gained by optimizing, in a<br />
controllable way, the chemical structure of TMC. In particular the degree of quaternization<br />
(DQ) of TMC seemed important since this is the polymer’s main determinant <strong>for</strong> charge<br />
density. Also, in vitro and in vivo studies with <strong>for</strong>mulation of TMC and both protein and low<br />
molecular weight compounds suggested an optimal DQ of 40-50% <strong>for</strong> transepithelial delivery<br />
[5-10].<br />
In this thesis we show that the DQ, but also the exclusion of O-methylation, h<strong>as</strong> no effect on<br />
the immunogenicity of TMC when used <strong>as</strong> coating on whole inactivated influenza virus (WIV).<br />
Incre<strong>as</strong>ing another variable, the degree of acetylation (DAc), resulted even in a reduction of the<br />
beneficial effects of coating WIV with TMC in n<strong>as</strong>al immunization in mice. So, looking from an<br />
immunological perspective, despite all ef<strong>for</strong>ts put in the development of novel synthetic routes<br />
to obtain structural variants of TMC, no improvements in immunological efficacy were<br />
achieved. However, looking from a pharmaceutical perspective, the availability of<br />
reproducible, well-characterized TMCs that can be structurally tailored in a controllable<br />
manner without introducing side reactions, may be considered <strong>as</strong> a major step <strong>for</strong>ward.<br />
Additionally, these structural variants (DQ, DOM, DAc) also influence others physicochemical<br />
characteristics and superior stability of O-methyl free TMC-WIV <strong>for</strong>mulations w<strong>as</strong> found [11].<br />
The covalently stabilized polyelectrolyte nanocomplexes of TMC-SH and HA-SH offer a novel<br />
vaccine delivery plat<strong>for</strong>m with high versatility. To mention, positively charged antigens may be<br />
entrapped using HA-SH <strong>for</strong> initial complexation and TMC-SH <strong>as</strong> crosslinker and also covalent<br />
coupling of antigens, adjuvants or targeting ligands is e<strong>as</strong>ily realizable on remaining free thiol<br />
moieties (<strong>as</strong> is already demonstrated <strong>for</strong> PEG maleimide). Active targeting to lectin receptors<br />
or integrins present on microfold (M)-cells can improve the immunogenicity of a <strong>for</strong>mulation<br />
[12] and thus coupling of sialic acid, galactose residues or specific antibodies (like IgA) on the<br />
particles’ surface may be an attractive next step to enhance binding and uptake of the particles<br />
by antigen presenting cells. Interestingly, these and similar sugar-moieties like mannose and<br />
N-acetyl glucosamine (GlucNAc) are also <strong>as</strong>sociated with targeting to dendritic cells (DCs).<br />
Indeed, the activation of DCs via GlucNAc is described in this thesis with TMC-RA (containing<br />
high GlucNAc) showing promising results on human DCs (Chapter 5B). Furthermore the type<br />
of immune response may be steered by incorporation of CpG motifs (usually towards Th1) and<br />
lipopolysaccharides (LPS, usually towards Th2) or other immuno-modulators. Finally, <strong>as</strong><br />
thiolation of TMC enhances its muco-adhesiveness [13], the use of TMC-SHs in ‘conventional’<br />
TMC-TPP particles may further incre<strong>as</strong>e n<strong>as</strong>al residence time of the particles resulting in<br />
better antigen delivery compared to nonthiolated TMCs.<br />
177
Chapter 8<br />
This thesis further shows the limitations of in vitro predictability and/or correlation <strong>for</strong> in<br />
vivo outcome <strong>as</strong> large differences were observed in vitro <strong>for</strong> toxicity and capacity to open<br />
cellular tight junctions <strong>for</strong> polymers that showed no differences in vivo n<strong>as</strong>al vaccination<br />
studies. Additionally, extensive investigation in various potentially crucial steps <strong>for</strong> n<strong>as</strong>al<br />
vaccination with both in vitro and in vivo models did not reveal major differences between an<br />
effective (TMC-WIV) and ineffective (TMC-RA-WIV) <strong>for</strong>mulation. Apparently, displaying<br />
cationic charges, muco-adhesiveness and improving interaction between the <strong>for</strong>mulation and<br />
the epithelial barrier are not sufficient <strong>for</strong> effective n<strong>as</strong>al vaccination. This emph<strong>as</strong>izes the<br />
need <strong>for</strong> even more detailed knowledge of TMC’s (and that of other adjuvants) mode of action.<br />
Only then, in a rational way, in vitro models can be developed that adequately predict in vivo<br />
outcome and ultimately minimize the use of animal models.<br />
Another issue that is raised by this thesis is the discrepancy in results between human and<br />
murine in vitro models (<strong>as</strong> w<strong>as</strong> observed <strong>for</strong> differences in the immuno-stimulatory effects of<br />
TMC-RA in human and murine DCs). Again, in depth knowledge on the mode of action of TMC<br />
is crucial <strong>for</strong> the predictability of murine in vivo/in vitro results <strong>for</strong> human outcome. E.g. if<br />
TMC’s adjuvant effect is achieved through improved muco-adhesion, minor differences may be<br />
anticipated between mice and men. In contr<strong>as</strong>t, M-cells are more abundant in murine n<strong>as</strong>al<br />
cavity and murine and human M-cells may express different receptors <strong>as</strong> may antigen<br />
presenting cells [14]; if TMC-b<strong>as</strong>ed systems accomplish their effect mainly via these<br />
mechanisms, limited predictability can be expected from studies with mice models.<br />
As can be concluded from the above, to choose just one type of TMC(-system) <strong>for</strong> further<br />
research <strong>for</strong> n<strong>as</strong>al vaccine development is not recommended: in particular, the effects of<br />
incre<strong>as</strong>ing the GlucNAc content are not fully understood yet and the full potential of the<br />
stabilized TMC-S-S-HA particles needs further investigation. Animal models (such <strong>as</strong> ferrets or<br />
transgenic mice) that mimic the human immune responses in a better extent may elaborate<br />
whether a future application of TMC <strong>for</strong> n<strong>as</strong>al vaccination in man is fe<strong>as</strong>ible.<br />
Other potential applications <strong>for</strong> novel TMCs. Next to mucosal vaccination, TMCs are<br />
frequently studied <strong>for</strong> mucosal delivery of pharmaceutically active proteins and low molecular<br />
weight compounds. Interestingly, <strong>for</strong> these applications, good correlations are found between<br />
the capacity of a polymer to open tight junctions (<strong>as</strong> me<strong>as</strong>ured in a transepithelial electrical<br />
resistance (TEER) <strong>as</strong>say) and in vivo uptake. This implies that, in particular, O-methyl free<br />
TMCs can improve the mucosal uptake compared to ‘conventional’ TMCs <strong>as</strong> they showed a<br />
better, reversible reduction of the TEER. Further studies should be carried out to confirm this.<br />
178
Summary and Future Perspectives<br />
TMCs have been used <strong>for</strong> DNA delivery with some encouraging results. However,<br />
introductory studies with O-methyl free TMCs <strong>for</strong> DNA transfection showed low transfection<br />
efficiencies probably due to poor endosomal escape. Interestingly, thiolation of TMC may not<br />
only enhance the transfection potential of DNA [15] but also gene silencing via delivery of<br />
siRNA into the cytosol (Varhouchi et al, manuscript submitted). Moreover, the stabilized TMC-<br />
S-S-HA carrier system shows excellent properties <strong>for</strong> DNA/siRNA delivery demonstrating<br />
enhanced stability under physiological conditions but they readily dissociate in a reductive<br />
environment (like in the cytosol). Depending on the target cell or the application route<br />
(systemic or local) PEG or targeting ligands (such <strong>as</strong> nano- or antibodies) can be attached. This<br />
makes TMC-SH and in particular the TMC-S-S-HA carrier systems highly interesting <strong>for</strong><br />
siRNA/DNA delivery and they are currently being evaluated <strong>for</strong> this purpose in our<br />
Department.<br />
PEGylated nanoparticles (e.g. PEG-liposomes) are widely studied <strong>for</strong> systemic delivery of<br />
proteins, cytokines and anti-cancer and immunomodulatory drugs to area’s of tumor growth<br />
or inflammation <strong>as</strong> a result of the enhanced permeability and retention effect in those tissues.<br />
The PEGylated, stabilized TMC-S-S-HA system may be a valuable alternative to currently used<br />
systems because of its simple preparation method and its potentially high loading capacity. As<br />
TMCs with a high DAc are readily degraded by lysozyme, an interesting option may be the<br />
incorporation of this enzymatically-degradable TMC into particles. In environments with high<br />
lysozyme concentrations, <strong>for</strong> example in inflammated or infectioned tissues (<strong>as</strong> lysozyme is<br />
excreted by macrophages), these particles should disintegrate and rele<strong>as</strong>e their contents.<br />
Finally, the thiolated polymers can be used <strong>for</strong> covalently linked layer-by-layer technologies<br />
[16, 17] and/or the preparation of covalently stabilized hydrogels with acrylated or<br />
methacrylated polymers (b<strong>as</strong>ed upon Michael addition) [18].<br />
TMC h<strong>as</strong> only been used <strong>for</strong> a decade or so <strong>for</strong> mucosal vaccination and other applications<br />
and the l<strong>as</strong>t couple of years interest h<strong>as</strong> risen rapidly due to its promising results. However, to<br />
look into the future of TMC much can be learned from its older precursor, chitosan. This<br />
polymer h<strong>as</strong> been investigated <strong>for</strong> over <strong>for</strong>ty years in various biomedical, nutritional and<br />
technological fields and widely studied <strong>for</strong> toxicity, biocompatibility and biodegradation. In<br />
contr<strong>as</strong>t to TMC, it h<strong>as</strong> only two (major) variables, the molecular weight and the DAc and their<br />
effects on the physicochemical and biological properties of chitosan have been thoroughly<br />
studied. Despite all this scientific ef<strong>for</strong>t, the FDA h<strong>as</strong> not yet approved chitosan <strong>as</strong> GRAS<br />
(generally regarded <strong>as</strong> safe) material which hinders its application in (biomedical) products<br />
179
Chapter 8<br />
[19]. Likely, in the p<strong>as</strong>t manufacturers disliked the difficulty of patentability <strong>for</strong> chitosanproducts<br />
and there<strong>for</strong>e little stress on regulatory authorities w<strong>as</strong> applied. Recently, however,<br />
pressure is build by researchers and clinicians to revise this regulatory perspective <strong>as</strong><br />
chitosan-like substances are highly awaited in the various biomedical fields. For instance, a<br />
chitosan-b<strong>as</strong>ed delivery plat<strong>for</strong>m is exploited by Archimedes (UK-b<strong>as</strong>ed pharmaceutical<br />
company) and this system is currently evaluated in ph<strong>as</strong>e I clinical trails <strong>for</strong> the n<strong>as</strong>al delivery<br />
of granisetron <strong>as</strong> anti-emetic (stated on the Archimedes website by December 2009). If<br />
encouraging results are obtained, it can be anticipated that approval by regulatory authorities<br />
will follow. As TMC h<strong>as</strong> superior characteristics to chitosan, TMC may shortly go behind.<br />
Further convincing studies showing TMC’s potential may speed up this process <strong>as</strong> may the<br />
availability of well-characterized and defined polymers described in this thesis.<br />
180
Summary and Future Perspectives<br />
References<br />
1. R. A. A. Muzzarelli and F. Tanfani. The N-permethylation of chitosan and the preparation of<br />
N-trimethyl chitosan iodide. Carbohydr Polym 5: 297-307 (1985).<br />
2. A. Domard, M. Rinaudo, and C. Terr<strong>as</strong>sin. New method <strong>for</strong> the quaternization of chitosan. Int J<br />
Bio Macromol 8: 105-107 (1986).<br />
3. I. M. Van der Lubben, J. C. Verhoef, M. M. Fretz, O. Van, I. Mesu, G. Kersten, and H. E.<br />
Junginger. <strong>Trimethyl</strong> chitosan chloride (TMC) <strong>as</strong> a novel excipient <strong>for</strong> oral and n<strong>as</strong>al<br />
immunisation against diphtheria. S.T.P. <strong>Pharma</strong> Sciences 12: 235-242 (2002).<br />
4. M. Amidi, S. G. Romeijn, J. C. Verhoef, H. E. Junginger, L. Bungener, A. Huckriede, D. J. A.<br />
Crommelin, and W. Jiskoot. N-<strong>Trimethyl</strong> chitosan (TMC) nanoparticles loaded with influenza<br />
subunit antigen <strong>for</strong> intran<strong>as</strong>al vaccination: Biological properties and immunogenicity in a mouse<br />
model. Vaccine 25: 144-153 (2007).<br />
5. F. Chen, Z. R. Zhang, F. Yuan, X. Qin, M. Wang, and Y. Huang. In vitro and in vivo study of<br />
N-trimethyl chitosan nanoparticles <strong>for</strong> oral protein delivery. Int J Pharm 349: 226-233 (2008).<br />
6. G. Di Colo, S. Burgal<strong>as</strong>si, Y. Zambito, D. Monti, and P. Chetoni. Effects of different N-<br />
trimethyl chitosans on in vitro/in vivo ofloxacin transcorneal permeation. J Pharm Sci 93: 2851-<br />
2862 (2004).<br />
7. J. H. Hamman, C. M. Schultz, and A. F. Kotze. N-trimethyl chitosan chloride: Optimum degree<br />
of quaternization <strong>for</strong> drug absorption enhancement across epithelial cells. Drug Develop Ind<br />
Pharm 29: 161-172 (2003).<br />
8. J. H. Hamman, M. Stander, and A. F. Kotze. Effect of the degree of quaternisation of N-<br />
trimethyl chitosan chloride on absorption enhancement: In vivo evaluation in rat n<strong>as</strong>al epithelia.<br />
Inter J Pharm 232: 235-242 (2002).<br />
9. A. F. Kotze, M. M. Thanou, H. L. Luessen, A. B. G. De Boer, J. C. Verhoef, and H. E.<br />
Junginger. Effect of the degree of quaternization of N-trimethyl chitosan chloride on the<br />
permeability of intestinal epithelial cells (Caco-2). Eur J Pharm Biopharm 47: 269-274 (1999).<br />
10. M. M. Thanou, A. F. Kotze, T. Scharringhausen, H. L. Lueßen, A. G. De Boer, J. C. Verhoef,<br />
and H. E. Junginger. Effect of degree of quaternization of N-trimethyl chitosan chloride <strong>for</strong><br />
enhanced transport of hydrophilic compounds across intestinal Caco-2 cell monolayers. J<br />
Control Rele<strong>as</strong>e 64: 15-25 (2000).<br />
11. N. Hagenaars. Towards an intran<strong>as</strong>al influenza vaccine - B<strong>as</strong>ed on whole inactivated influenza<br />
virus with N,N,N-trimethylchitosan <strong>as</strong> adjuvant. Utrecht University, Utrecht (2010).<br />
12. B. Slütter, N. Hagenaars, and W. Jiskoot. Rational design of n<strong>as</strong>al vaccines. J Drug Target 16: 1-17<br />
(2008).<br />
13. L. Yin, J. Ding, C. He, L. Cui, C. Tang, and C. Yin. Drug permeability and mucoadhesion<br />
properties of thiolated trimethyl chitosan nanoparticles in oral insulin delivery. Biomaterials 30:<br />
5691-5700 (2009).<br />
14. S. K. Singh, J. Stephani, M. Schaefer, H. Kalay, J. J. García-Vallejo, J. den Haan, E. Saeland, T.<br />
Sparw<strong>as</strong>ser, and Y. van Kooyk. Targeting glycan modified OVA to murine DC-SIGN<br />
transgenic dendritic cells enhances MHC cl<strong>as</strong>s I and II presentation. Mol Immunol 47: 164-174<br />
(2009).<br />
15. X. Zhao, L. Yin, J. Ding, C. Tang, S. Gu, C. Yin, and Y. Mao. Thiolated trimethyl chitosan<br />
nanocomplexes <strong>as</strong> gene carriers with high in vitro and in vivo transfection efficiency. J Control<br />
Rele<strong>as</strong>e 144: 46-54 (2010).<br />
16. B. G. De Geest, G. B. Sukhorukov, and H. Möhwald. The pros and cons of polyelectrolyte<br />
capsules in drug delivery. Exp Opin Drug Deliv 6: 613-624 (2009).<br />
17. A. Szarpak, D. Cui, F. Dubreuil, B. G. De Geest, L. J. De Cock, C. Picart, and R. Auzély-Velty.<br />
Designing hyaluronic acid-b<strong>as</strong>ed layer-by-layer capsules <strong>as</strong> a carrier <strong>for</strong> intracellular drug<br />
delivery. Biomacromolecules 11: 713-720 (2010).<br />
18. R. Censi, P. J. Fieten, P. di Martino, W. E. Hennink, and T. Vermonden. In Situ Forming<br />
Hydrogels by Tandem Thermal Gelling and Michael Addition Reaction between<br />
181
Chapter 8<br />
Thermosensitive Triblock Copolymers and Thiolated Hyaluronan. Macromolecules 43: 5771-5778<br />
(2010).<br />
19. T. Kean and M. Thanou. Biodegradation, biodistribution and toxicity of chitosan. Adv Drug<br />
Deliv Rev 62: 3-11 (2010).<br />
182
APPENDICES
Affiliations of Collaborating Authors<br />
Affiliations of collaborating authors:<br />
Maryam Amidi<br />
Department of <strong>Pharma</strong>ceutics, Utrecht Institute <strong>for</strong> <strong>Pharma</strong>ceutical Sciences, Utrecht<br />
University, Utrecht, The Netherlands<br />
Suzanne Bal<br />
Division of Drug Delivery Technology, Leiden/Amsterdam Center <strong>for</strong> Drug Research, Leiden<br />
University, Leiden, The Netherlands<br />
Han van den Bosch<br />
Nobilon, part of Merck, Boxmeer, The Netherlands<br />
Joke Bouwstra<br />
Division of Drug Delivery Technology, Leiden/Amsterdam Center <strong>for</strong> Drug Research, Leiden<br />
University, Leiden, The Netherlands<br />
Sven Bruins<br />
Department of Molecular Cell Biology and Immunology, Vrije University Medical Center,<br />
Amsterdam, The Netherlands<br />
Thom<strong>as</strong> van Es<br />
Department of Molecular Cell Biology and Immunology, Vrije University Medical Center,<br />
Amsterdam, The Netherlands<br />
Ethlinn van Gaal<br />
Department of <strong>Pharma</strong>ceutics, Utrecht Institute <strong>for</strong> <strong>Pharma</strong>ceutical Sciences, Utrecht<br />
University, Utrecht, The Netherlands<br />
Harrie Glansbeek<br />
Nobilon, part of Merck, Boxmeer, The Netherlands<br />
Niels Hagenaars<br />
Department of <strong>Pharma</strong>ceutics, Utrecht Institute <strong>for</strong> <strong>Pharma</strong>ceutical Sciences, Utrecht<br />
University, Utrecht, The Netherlands<br />
Jacco Heldens<br />
Nobilon, part of Merck, Boxmeer, The Netherlands<br />
Wim Hennink<br />
Department of <strong>Pharma</strong>ceutics, Utrecht Institute <strong>for</strong> <strong>Pharma</strong>ceutical Sciences, Utrecht<br />
University, Utrecht, The Netherlands<br />
Wim Jiskoot<br />
Division of Drug Delivery Technology, Leiden/Amsterdam Center <strong>for</strong> Drug Research, Leiden<br />
University, Leiden, The Netherlands<br />
P<strong>as</strong>cal de Jong<br />
Department of <strong>Pharma</strong>ceutics, Utrecht Institute <strong>for</strong> <strong>Pharma</strong>ceutical Sciences, Utrecht<br />
University, Utrecht, The Netherlands<br />
185
Affiliations of Collaborating Authors<br />
Enrico M<strong>as</strong>trobattista<br />
Department of <strong>Pharma</strong>ceutics, Utrecht Institute <strong>for</strong> <strong>Pharma</strong>ceutical Sciences, Utrecht<br />
University, Utrecht, The Netherlands<br />
Imke Mooren<br />
Animal Service Department, Intervet Animal Health part of Merck, Boxmeer, The Netherlands<br />
Ivo Que<br />
Department of Endocrinology and Metabolic Dise<strong>as</strong>es, Leiden University Medical Center,<br />
Leiden, The Netherlands<br />
Elly van Riet<br />
Division of Drug Delivery Technology, Leiden/Amsterdam Center <strong>for</strong> Drug Research, Leiden<br />
University, Leiden, The Netherlands<br />
Bram Slütter<br />
Division of Drug Delivery Technology, Leiden/Amsterdam Center <strong>for</strong> Drug Research, Leiden<br />
University, Leiden, The Netherlands<br />
Mies van Steenbergen<br />
Department of <strong>Pharma</strong>ceutics, Utrecht Institute <strong>for</strong> <strong>Pharma</strong>ceutical Sciences, Utrecht<br />
University, Utrecht, The Netherlands<br />
Steffen van der Wal<br />
Department of Medicinal Chemistry and Chemical Biology, Utrecht Institute <strong>for</strong> <strong>Pharma</strong>ceutical<br />
Sciences, Utrecht University, Utrecht, The Netherlands<br />
186
List of Abbreviations<br />
List of abbreviations:<br />
APC<br />
CpG<br />
CS<br />
CS-RA<br />
DAc<br />
DC<br />
D carb<br />
DDM<br />
DLS<br />
DMC<br />
DOM<br />
D thiol<br />
DTT<br />
DQ<br />
EDC<br />
ELISA<br />
FACS<br />
FCS<br />
FDA<br />
FITC<br />
GPC<br />
HA<br />
HA-SH<br />
HBSS<br />
HEPES<br />
ID<br />
IgG<br />
IL<br />
i.m.<br />
i.n.<br />
kDa<br />
LDH<br />
antigen presenting cell<br />
cytosine guanine dinucleotide<br />
chitosan<br />
re-acetylated chitosan<br />
degree of acetylation<br />
dendritic cell<br />
degree of carboxylation<br />
degree of dimethylation<br />
dynamic light scattering<br />
dimethyl chitosan<br />
degree of O-methylation<br />
degree of thiolation<br />
dithiothreitol<br />
degree of quaternization<br />
1-ethyl-3-(3-dimethylaminopropyl) carbodiimide)<br />
enzyme-linked immunosorbent <strong>as</strong>say<br />
fluorescence activated cell sorter<br />
fetal calf serum<br />
food and drug administration<br />
fluorescein isothiocyanate<br />
gel permeation chromatography<br />
hyaluronic acid<br />
thiolated hyaluronic acid<br />
Hank’s balanced salt solution<br />
N-2-hydroxyethylpiperazine-N’-2-ethanesulfonic acid<br />
intradermal<br />
immunoglobulin G<br />
interleukin<br />
intramuscular<br />
intran<strong>as</strong>al<br />
kilodalton<br />
lactate dehydrogen<strong>as</strong>e<br />
187
List of Abbreviations<br />
LPS<br />
lipopolysaccharide<br />
MALT<br />
mucosal <strong>as</strong>sociated lymphoid tissue<br />
MTT<br />
3-(4,5-Dimethylthiazol-2-yl)-2,5-diphenyltetrazolium<br />
bromide<br />
M n<br />
number average molecular weight<br />
M w<br />
weight average molecular weight<br />
NALT<br />
n<strong>as</strong>al <strong>as</strong>sociated lymphoid tissue<br />
NMP<br />
N-methyl-2-pyrrolidone<br />
NMR<br />
nuclear magnetic resonance<br />
OD<br />
optical density<br />
OVA<br />
ovalbumin<br />
PBS<br />
phosphate buffer saline<br />
PDI<br />
polydispersity index<br />
PEG<br />
poly(ethylene glycol)<br />
PEI<br />
polyethylenimine<br />
SDS<br />
sodium dodecyl sulphate<br />
sIgA<br />
secretory immunoglobulin A<br />
siRNA<br />
small interfering ribonucleic acid<br />
TEER<br />
trans epithelial electrical resistance<br />
Th1/Th2 T helper cell type 1/2<br />
TMC<br />
trimethyl chitosan<br />
TMC-COOH<br />
carboxylated trimethyl chitosan<br />
TMC-OM<br />
O-methylated trimethyl chitosan<br />
TMC-RA<br />
re-acetylated trimethyl chitosan<br />
TMC-SH<br />
thiolated trimethyl chitosan<br />
TNF-α<br />
tumor necrosis factor alfa<br />
TPP<br />
tripolyphosphate<br />
WIV<br />
whole inactivated influenza virus<br />
XTT<br />
sodium 3´-[1-(phenylaminocarbonyl)- 3,4-tetrazolium]-<br />
bis (4-methoxy-6-nitro) benzene sulfonicacid hydrate<br />
188
Curriculum Vitae<br />
Curriculum Vitae<br />
Rolf Verheul w<strong>as</strong> born on the 8 th of March 1980 in Oss, The<br />
Netherlands. After finishing pre-university education<br />
(Gymn<strong>as</strong>ium) at the Titus Brandsma Lyceum in Oss in 1998,<br />
he started studying pharmacy at Utrecht University. In 2003<br />
he obtained his M<strong>as</strong>ter’s degree in pharmaceutical sciences<br />
cum laude, followed by his pharmacist degree (PharmD) in<br />
2005. During his study, he completed a research traineeship<br />
at the University of British Columbia, Vancouver, Canada under the supervision of prof. dr.<br />
Cullis <strong>as</strong> well <strong>as</strong> internships at the National Institute <strong>for</strong> Public Health and the Environment<br />
(RIVM) and various public and hospital pharmacies. Also, under the framework of the honors<br />
program in pharmacy, he followed a minor in medical anthropology and sociology at the<br />
University of Amsterdam. Thereafter, he worked <strong>as</strong> a laboratory pharmacist in the hospital<br />
pharmacy of the Academic Medical Center, Amsterdam, <strong>for</strong> one year. In November 2006 he<br />
started his PhD research program at the department of <strong>Pharma</strong>ceutics, Utrecht University<br />
under the supervision of prof. dr. ir. W.E. Hennink and prof. dr. W. Jiskoot. The results of his<br />
work are presented in this thesis.<br />
189
List of Publications<br />
List of publications:<br />
RJ Verheul, M Amidi, S van der Wal, E van Riet, W Jiskoot, WE Hennink. Synthesis,<br />
characterization and in vitro biological properties of O-methyl free N,N,N,-trimethylated<br />
chitosan. Biomaterials 29, 3642-3649 (2008).<br />
RJ Verheul, M Amidi, M van Steenbergen, E van Riet, W Jiskoot, WE Hennink. Influence of the<br />
degree of acetylation on the enzymatic degradation and in vitro biological properties of<br />
trimethylated chitosans. Biomaterials 30, 3129-3135 (2009).<br />
N Hagenaars, E M<strong>as</strong>trobattista, RJ Verheul, I Mooren, HL Glansbeek, JGM Heldens, H van den<br />
Bosch, W Jiskoot. Physicochemical and immunological characterization of N,N,N-trimethyl<br />
chitosan-coated whole inactivated influenza virus vaccine <strong>for</strong> intran<strong>as</strong>al administration.<br />
<strong>Pharma</strong>ceutical Research 26,1353-1364 (2009).<br />
RJ Verheul, N Hagenaars, I Mooren, PH de Jong, E M<strong>as</strong>trobattista, HL Glansbeek, JGM Heldens,<br />
H van den Bosch, WE Hennink, W Jiskoot. Relationship between structure and adjuvanticity<br />
of N,N,N-trimethyl chitosan (TMC) structural variants in a n<strong>as</strong>al influenza vaccine. Journal of<br />
Controlled Rele<strong>as</strong>e 140, 126-133 (2009).<br />
B Slütter, PC Soema, Z Ding, R Verheul, W Hennink, W Jiskoot. Conjugation of ovalbumin to<br />
trimethyl chitosan improves immunogenicity of the antigen. Journal of Controlled Rele<strong>as</strong>e<br />
143, 207-214 (2010).<br />
RJ Verheul, S van der Wal, WE Hennink. <strong>Tailorable</strong> thiolated trimethyl chitosans <strong>for</strong><br />
covalently stabilized nanoparticles. Biomacromolecules 11, 1965-1971 (2010).<br />
RJ Verheul, N Hagenaars, T van Es, E van Gaal, P de Jong, S Bruins, I Que, B Slütter, E<br />
M<strong>as</strong>trobattista, HL Glansbeek, JGM Heldens, H van den Bosch, WE Hennink, W Jiskoot. A stepby-step<br />
approach to study the influence of N-acetylation on the adjuvanticity of N,N,Ntrimethyl<br />
chitosan (TMC) in an intran<strong>as</strong>al whole inactivated influenza virus vaccine.<br />
Submitted.<br />
RJ Verheul¸ N Hagenaars, T van Es, S Bruins, B Slütter, WE Hennink, W Jiskoot. Maturation of<br />
human monocyte derived dendritic cells by trimethyl chitosan is correlated with its N-acetyl<br />
glucosamine (GlcNAc) content. Submitted <strong>as</strong> short communication.<br />
RJ Verheul, B Slütter, SM Bal, JA Bouwstra, W Jiskoot, WE Hennink. Covalently stabilized<br />
trimethyl chitosan-hyaluronic acid nanoparticles <strong>for</strong> n<strong>as</strong>al and intradermal vaccination.<br />
Submitted.<br />
SM Bal, B Slütter, RJ Verheul, JA Bouwstra, W Jiskoot. <strong>Adjuvant</strong>ed, antigen loaded N-trimethyl<br />
chitosan nanoparticles <strong>for</strong> n<strong>as</strong>al and intradermal vaccination: adjuvant- and site-dependent<br />
immunogenicity in mice. Submitted.<br />
AK Varkouhi, RJ Verheul, RM Schiffelers, T Lammers, G Storm, WE Hennink. Gene silencing<br />
activity of siRNA polyplexes b<strong>as</strong>ed on thiolated N,N,N-trimethylated chitosan. Submitted.<br />
190
NEDERLANDSE SAMENVAT<strong>TI</strong>NG<br />
EN<br />
DANKWOORD
Nederlandse Samenvatting<br />
Nederlandse Samenvatting<br />
Vaccinatie is het actief opwekken van een afweerreactie tegen een bepaalde<br />
lichaamsvreemde stof of eiwit (antigeen) van een ziekteverwekker. Hierdoor zal het<br />
immuunsysteem in het vervolg deze indringer snel herkenen en het kunnen verwijderen<br />
voordat het daadwerkelijk schade kan veroorzaken. Over de jaren heeft vaccinatie bewezen<br />
om het meest (kosten)effectieve instrument te zijn in de strijd tegen besmettelijke ziekten.<br />
Vaccins worden meestal gemaakt van levende, verzwakte of geïnactiveerde ziekteverwekkers<br />
(bijv. virussen of bacteriën) of gezuiverde immunogene eiwitten die afkomstig zijn van deze<br />
pathogenen. Hierdoor zullen vaccins zelf geen ziekte induceren, maar kunnen ze wel het<br />
immuunsysteem activeren. Tegenwoordig worden de meeste vaccins toegediend via een<br />
injectienaald. Echter, vanwege het risico van besmette naalden, met name in derde wereld<br />
landen, en de noodzaak van geschoold personeel voor een dergelijke toediening, is de<br />
belangstelling voor alternatieve vaccinatieroutes toegenomen.<br />
Hoofdstuk 1 bediscussieert de voor- en nadelen van de intramusculaire en alternatieve<br />
toedieningsroutes voor vaccins. In het bijzonder wordt immunisatie via de neus (intran<strong>as</strong>ale<br />
vaccinatie) besproken, waarmee relatief eenvoudig met bijvoorbeeld een neusspray of<br />
neusdruppels, naald-vrij en zonder getraind personeel een vaccin toegediend kan worden.<br />
Bovendien kan intran<strong>as</strong>ale vaccinatie zowel mucosale (op de slijmvliezen) als systemische (in<br />
het lichaam) immuunresponsen veroorzaken wat een verbeterde bescherming zou kunnen<br />
betekenen t.o.v. intramusculaire toediening. Aan de andere kant zijn, vanwege de natuurlijke<br />
barrières in de neus, antigeenafbraak en het bereiken van de antigeen presenterende cellen<br />
(APC's) over de slijm- en cellagen heen, belangrijke nadelen van n<strong>as</strong>ale vaccinatie (Figuur 1).<br />
Aangezien verzwakte, levende vaccins een aanzienlijk veiligheidsrisico met zich meebrengen<br />
voor mensen met een suboptimaal immuun systeem (b.v. kinderen, ouderen, patiënten met<br />
bepaalde ziektes als HIV), richt de ontwikkeling van vaccins zich nu vooral op het gebruik van<br />
gezuiverde, goed gekarakteriseerde antigene eiwitten. Echter, deze subunit vaccins zijn over<br />
het algemeen minder immunogeen en hebben een krachtig adjuvans (systeem) nodig om een<br />
adequate afweerreactie op te roepen. Door een muco-adhesief polymeer (dat ‘plakt’ aan de<br />
slijmlaag), zoals N,N,N-trimethylchitosan (TMC), te gebruiken bij intran<strong>as</strong>ale vaccinatie kan de<br />
immuunrespons tegen een antigeen worden verhoogd.<br />
193
Nederlandse Samenvatting<br />
Figuur 1. Schematische weergave van de cruciale stappen bij intran<strong>as</strong>ale vaccinatie. Na toediening zal<br />
het antigeen in contact komen met de oppervlaktes van het neusslijmvlies (1) waarna het vervolgens<br />
getransporteerd wordt over het n<strong>as</strong>ale epitheel (2) en opgenomen wordt door de antigeen<br />
presenterende cellen (APCs) zoals dendritische cellen (DCs) (3). Deze APCs zullen na activering o.a. tot<br />
de aanmaak van antilichamen overgaan (4).<br />
Vermoedelijk werkt TMC, een in water-oplosbaar chitosan derivaat, door de verblijftijd in de<br />
neus te verlegen en/of het contact tussen het antigeen en het mucosale oppervlak te<br />
verbeteren. TMC (Figuur 2) is eigenlijk een verzamelnaam voor een groep van polymeren met<br />
veel overeenkomsten in de chemische structuur maar met specifieke verschillen (zoals de<br />
ketenlengte, ladingsdichtheid en mate van methylering en acetylering). Het gecontroleerd<br />
kunnen variëren van deze structurele elementen maakt het mogelijk beter inzicht te krijgen in<br />
de afzonderlijke bijdrage van deze groepen op de fysisch-chemische en biologische<br />
eigenschappen van TMC. Ook kunnen nieuwe funtionaliteiten worden geïntroduceerd, zoals<br />
194
Nederlandse Samenvatting<br />
thiolgroepen die kunnen bijdragen aan het verbeteren en optimaliseren van de eigenschappen<br />
van TMC.<br />
Chitosan, en zijn derivaten als TMC, zijn in de vorm van nano- of microdeeltjes effectiever in<br />
het induceren van een immuunrespons dan een ‘gewone’ polymeeroplossing. Nano-gelering,<br />
gebruikmakend van electrostatische interacties, is een eenvoudige, veel gebruikte methode om<br />
nanodeeltjes van TMC te maken. Door negatief geladen (macro)moleculen druppelsgewijs<br />
onder roeren aan de TMC oplossing toe te voegen, vormen zich spontaan nanodeeltjes in een<br />
buffer met lage ionsterkte. Echter, de fysisch-chemische stabiliteit en de immunogeniciteit van<br />
deze met antigeen beladen complexen zijn afhankelijk van de eigenschappen van de gebruikte<br />
componenten. Mogelijk kunnen de momenteel toegep<strong>as</strong>te systemen nog worden verbeterd.<br />
Bovendien, hoewel het werkingsmechanisme van TMC doorgaans wordt toegeschreven aan de<br />
muco-adhesieve en penetratie verbeterende eigenschappen, ontbreekt er gedetailleerde<br />
kennis over het mechanisme (zo is bijvoorbeeld het directe effect van TMC op APCs niet<br />
bekend). Concluderend kan gesteld worden dat er zijn vele mogelijkheden voor het<br />
optimaliseren van de structuur van TMC en de op TMC geb<strong>as</strong>eerde nanodeeltjes. Tegelijkertijd<br />
zal een beter inzicht in het werkingsmechanisme van TMC tot een rationeel ontwerp van<br />
dergelijke adjuvantia kunnen leiden.<br />
Figuur 2. Schematische weergave van de structurele variaties van TMC. TMC kan variëren in de mate<br />
van acetylering (blok x), quaternisering (blok y) en O-methylering (blok z). Deze variaties kunnen<br />
willekeurig over het polymeer verdeeld zijn.<br />
Het doel van dit proefschrift is om synthese-routes te ontwikkelen waarmee op een<br />
controleerbare wijze de chemische structuur van TMC kan worden gevarieerd. Verder worden<br />
er nieuwe groepen geïntroduceerd, zoals thiolen, die mogelijk de eigenschappen van TMC nog<br />
195
Nederlandse Samenvatting<br />
verder kunnen verbeteren. Op deze manier kunnen structuur-activiteit-relaties worden<br />
onderzocht in zowel in vitro studies als in in vivo (n<strong>as</strong>ale) vaccinatie studies.<br />
De huidige methode om TMC te synthetiseren veroorzaakt, na<strong>as</strong>t de gewenste trimethylering<br />
van de vrije amines, ook O-methylering op de hydroxylgroepen en ketenbreuk van het<br />
polymer. Omdat deze nevenreacties van invloed kunnen zijn op de eigenschappen van het<br />
polymeer, is er een behoefte aan een syntheseroute waarbij deze ongewenste nevenreacties<br />
niet optreden en waarbij de mate van quaternisering (oftewel het percentage amine groepen<br />
dat getrimethyleerd is (DQ)) vrijelijk gevarieerd en gecontroleerd kan worden.<br />
In Hoofdstuk 2 werden met behulp van een twee-staps methode O-methyl vrije TMCs met<br />
variërende DQs gesynthetiseerd. Eerst werd chitosan kwantitatief gedimethyleerd met behulp<br />
van mierenzuur en <strong>for</strong>maldehyde. Vervolgens kon TMC worden verkregen door dit<br />
dimethylchitosan te laten reageren met een overmaat aan iodomethaan. Door de reactietijd te<br />
variëren, konden TMCs met verschillende DQs (22-68%) worden gesynthetiseerd. Deze TMCs<br />
bleken geen waarneembare O-methylering (bepaald met 1 H-NMR) te hebben en uit een geringe<br />
verhoging van het molecuulgewicht met toenemende DQ (bepaald met gel permeatie<br />
chromatografie of GPC) kon geconcludeerd worden dat zich geen noemenswaardige<br />
ketenbreuk had voorgedaan tijdens de synthese. De oplosbaarheid in water met een pH van 7<br />
w<strong>as</strong> voor O-methyl vrij TMC met een DQ 30% losten uitstekend op in water met<br />
een pH van 7. O-methyl vrije TMCs gaven een grotere daling in de trans-epitheliale elektrische<br />
weerstand (TEER) van een caco-2 monolaag dan O-gemethyleerde TMCs. Door het meten van<br />
de TEER wordt een idee verkregen in hoeverre een polymeer in staat is de intercellulaire<br />
ruimtes tussen epitheelcellen te openen waardoor het transport van stoffen over deze cellaag<br />
gemakkelijker wordt. Een daling in de TEER geeft aan dat de weerstand afneemt en dus dat de<br />
cellaag permeabeler wordt. Ook werd met de verhoging van de DQ, een stijging van de<br />
cytotoxiciteit (m.b.v. een MTT test) en celmembraan-permeabiliteit (m.b.v. een LDH test)<br />
waargenomen. Dit hoofdstuk toont duidelijk aan dat zowel de DQ als de aan/afwezigheid van<br />
O-methylering een belangrijke invloed hebben op de fysische en in vitro biologische<br />
eigenschappen van TMC.<br />
De invloed van een andere variatie in de moleculaire structuur, de mate van acetylering<br />
(DAc), op de in vitro enzymatische afbraak en biologische eigenschappen van TMC is<br />
onderzocht in hoofdstuk 3. TMCs met een DAc variërend van 11 tot 55% werden<br />
196
Nederlandse Samenvatting<br />
gesynthetiseerd met behulp van een drie-stappen-methode. Eerst werden de amines van<br />
chitosan gedeeltelijk geacetyleerd met azijnzuuranhydride en natrium borohydride, gevolgd<br />
door de kwantitatieve dimethylering van de resterende amines met <strong>for</strong>maldehyde en natrium<br />
borohydride. Vervolgens werd TMC gesynthetiseerd in aanwezigheid van een overmaat aan<br />
iodomethaan. De TMCs verkregen via deze methode bleken geen O-methylering, noch een<br />
verlies in acetylgroepen ( 1 H-NMR) of ketenbreuk (GPC) te vertonen. De mate van afbraak van<br />
TMC, en dat van haar voorgangers chitosan en dimethyl chitosan, door lysozyme (een<br />
lichaamseigen enzym) w<strong>as</strong> sterk afhankelijk van de DAc; polymeren met de hoogste DAc<br />
vertoonde de grootste daling in het molecuulgewicht. TMCs met een hoge DAc (~ 50%), een<br />
DQ van ongeveer 44% en met of zonder O-gemethyleerde groepen, waren niet in staat om de<br />
intercellulaire ruimtes van een caco-2 monolaag te openen zoals onderzocht in de transepitheliale<br />
elektrische weerstand (TEER) test. Dit in tegenstelling tot TMCs (zowel O-<br />
gemethyleerde en O-methyl vrij; concentratie van 2.5 mg/ml) met een soortgelijke DQ maar<br />
een lagere DAc; deze polymeren waren in staat de TEER tot 70 en 30% te verminderen,<br />
respectievelijk voor de O-methyl vrij en de O-gemethyleerde TMCs. Daarna<strong>as</strong>t vertoonden<br />
TMCs met een hoge DAc (~ 50%) geen noemenswaardige celtoxiciteit (MTT, LDH afgifte) tot<br />
een concentratie van 10 mg/ml. Dit hoofdstuk toont aan dat de mate van N-acetylering van<br />
TMC een grote invloed heeft op de in vitro enzymatische afbraak en biologische eigenschappen<br />
van TMC.<br />
De resultaten uit hoofdstukken 2 en 3 tonen aan dat de DQ, DAc en mate van O-methylering<br />
van invloed zijn op de in vitro biologische eigenschappen van TMC. Daarom onderzochten we<br />
de effecten van deze structurele variaties van TMC op de werkzaamheid als adjuvans in een<br />
intran<strong>as</strong>ale (i.n.) immunisatie studie in muizen.<br />
In hoofdstuk 4 werden TMCs met variërende mate van quaternisering (DQ, 22-86%), O-<br />
methylering (DOM, 0-76%) en N-acetylering (DAc 9-54%) ge<strong>for</strong>muleerd met geïnactiveerd<br />
influenza virus (WIV). Eenvoudig mengen van het WIV met de TMCs in een 1:1<br />
gewichtsverhouding resulteerde in vergelijkbare, positief geladen nanodeeltjes door coating<br />
van het negatief geladen WIV met het positief geladen TMC. De hoeveelheid vrij TMC in<br />
oplossing w<strong>as</strong> vergelijkbaar voor alle TMC-WIV <strong>for</strong>muleringen en er is dus een overmaat aan<br />
TMC aanwezig in de <strong>for</strong>muleringen.<br />
Na intran<strong>as</strong>ale immunisatie van muizen met WIV en TMC-WIV op dag 0 en 21, vertoonden de<br />
TMC-WIV <strong>for</strong>muleringen een sterkere respons (gemeten als hogere totaal IgG, IgG1 en IgG2a/c<br />
titers) dan WIV zonder TMC, met uitzondering van WIV gecoat met gereacetyleerd TMC (DAc<br />
197
Nederlandse Samenvatting<br />
54 %, DQ 44%). Er werden geen significante verschillen in de antistof-titers waargenomen<br />
voor TMCs die varieerden in DQ of DOM, wat aangeeft dat deze structurele kenmerken een<br />
ondergeschikte rol spelen in de i.n. adjuvans eigenschappen. TMC met een DQ van 56%<br />
(TMC56) ge<strong>for</strong>muleerd met WIV in een gewichtsverhouding van 5:1 resulteerde in significant<br />
lagere IgG2a/c:IgG1 ratio's ten opzichte van TMC56 gemengd in de verhouding van 0.2:1 en<br />
1:1. De IgG2a/c:IgG1 ratio geeft een indicatie voor het type immuunrespons dat opgewekt<br />
wordt: een hoge IgG1 spiegel wijst op een type 2 (of humorale, met antilichamen,<br />
afweerreactie) terwijl een hoge IgG2a/c titer een type 1 (of cellulaire, met cytotoxische T-<br />
cellen, afweerreactie) indiceert. Deze resultaten impliceren dus een verschuiving naar een<br />
Th2-type immuunrespons bij een hogere TMC dosis. Blootstelling van de gevaccineerde<br />
muizen aan een actief, ziekmakend, virus bewezen dat alle TMC-WIV <strong>for</strong>muleringen<br />
bescherming boden met uitzondering van WIV gecoat met gereacetyleerd TMC. Dit hoofdstuk<br />
toont aan dat de immunogeniciteit van WIV sterk verbetert door het gebruik van TMC en dat<br />
bescherming tegen een actief virus na i.n. vaccinatie met TMC-WIV kan worden bewerkstelligd.<br />
De werkzaamheid van TMC als adjuvans wordt sterk verminderd door het acetyleren van TMC,<br />
terwijl de DQ en DOM nauwelijks van invloed zijn.<br />
Het doel van hoofdstuk 5A w<strong>as</strong> om de oorzaak te achterhalen voor het gebrek aan<br />
werkzaamheid van gereacetyleerd TMC (TMC-RA) als adjuvans. TMC-RA (mate van acetylering<br />
54%) werd vergeleken met TMC (mate van acetylering 17%) op zes potentieel kritieke<br />
stappen in de inductie van een immuunrespons na intran<strong>as</strong>ale (i.n.) toediening bij muizen:<br />
chemische stabiliteit van het polymeer in een neusw<strong>as</strong>sing, de locatie van WIV in de neus, de<br />
n<strong>as</strong>ale verblijftijd van WIV, de cellulaire opname van WIV door calu-3 epitheelcellen, het<br />
transport van WIV over een monolaag van epitheel cellen, en de capaciteit van de verschillende<br />
<strong>for</strong>muleringen om activering te induceren van muizen dendritische cellen (DCs).<br />
TMC-RA werd in een grotere mate afgebroken in de neusw<strong>as</strong>sing dan TMC. De locatie van het<br />
antigeen in de neus en de n<strong>as</strong>ale verblijftijd waren hetzelfde voor beide types TMC.<br />
Fluorescent gelabeld WIV <strong>as</strong>socieerde beter met calu-3 cellen indien gecoat met TMC-RA dan<br />
met TMC, en beide TMCs verminderden aanzienlijk het transport van WIV over een calu-3<br />
monolaag. Muizen DCs werden door alle <strong>for</strong>muleringen in gelijke mate geactiveerd.<br />
Samenvattend, de inferieure werkzaamheid van TMC-RA als adjuvans zou kunnen worden<br />
verklaard door een lagere stabiliteit van de TMC-RA-WIV in de neusholte, en het is in ieder<br />
geval onwaarschijnlijk dat een van de andere cruciale stappen die zijn onderzocht in deze<br />
studie, een rol spelen.<br />
198
Nederlandse Samenvatting<br />
Het is bekend dat de N-geacetyleerde glucosamine blokken (of GlcNAcs) aanwezig in TMC<br />
kunnen binden aan verschillende humane C-type lectines, een familie van cel-receptoren die<br />
betrokken zijn bij de immuunrespons. Mogelijk is hierdoor het effect van TMC en<br />
gereacetyleerd TMC (met DAc van 54%, TMC-RA) op opname en activering van menselijke<br />
dendritische cellen (DCs) verschillend. Dit werd onderzocht in hoofdstuk 5B met behulp van<br />
het geïnactiveerde influenza virus (WIV) als antigeen. Studies met monocyt-afgeleide humane<br />
DCs lieten zien dat de opname van TMC(-RA) gecoat WIV iets lager w<strong>as</strong> dan ‘gewoon’ WIV.<br />
TMC-RA, als oplossing of als coating van WIV, veroorzaakte een veel sterkere activering van<br />
DCs, zoals gemeten met CD86 expressie, dan de andere <strong>for</strong>muleringen. Verder induceerde<br />
alleen TMC-RA(-WIV) een hoge afgifte van de cytokines IL-10, TNF- α en IL-12p40 en IL-12p70<br />
door de DCs. Omdat zowel de afgifte van IL-10 en IL-12p70 w<strong>as</strong> verhoogd, konden er geen<br />
uitspaken worden gedaan over het mogelijk induceren van een Th1 of Th2 type<br />
immuunrespons. Concluderend, TMC-RA heeft sterke immuno-stimulerende effecten in vitro<br />
op humane dendritische cellen. Dit impliceert dat de mate van N-acetylering mogelijk van<br />
cruciaal belang is voor het adjuvans effect van TMC in mensen.<br />
Door thiol-groepen te introduceren in de structuur zullen de (farmaceutische) toep<strong>as</strong>singen<br />
van TMC nog verder verbreed kunnen worden. Thiolen kunnen de muco-adhesieve<br />
eigenschappen verbeteren, vergemakkelijken verdere derivatiseringsreacties via o.m.<br />
intracellulair afbreekbare disulfide-bruggen, en maken het mogelijk om covalent vernette<br />
nanodeeltjes te produceren met een gethioleerd anionisch polymeer als crosslinker. In<br />
Hoofdstuk 6 werd een nieuwe vier-stappen methode gepresenteerd om gedeeltelijk<br />
gethioleerd TMC te synthetiseren waarmee de mate van quaternisering en thiolering vrijelijk<br />
kunnen worden gevarieerd. Eerst werden de amines van chitosan gedeeltelijk gecarboxyleerd<br />
met glyoxylzuur en natrium borohydride. Daarna werden de resterende amines kwantitatief<br />
gedimethyleerd met <strong>for</strong>maldehyde en natrium borohydride en vervolgens konden<br />
gecarboxyleerde TMCs verkregen worden door dit polymeer te laten reageren met een<br />
overmaat aan iodomethaan in N-methylpyrrolidon. Hierna werden de gedeeltelijk<br />
gecarboxyleerde TMCs opgelost in water en werd cystamine aan de carboxylzuren gekoppeld<br />
bij een pH van 5.5 met behulp van EDC. Na toevoeging van dithiotreïtol disulfide om de<br />
disulfide-bruggen te breken, werden gethioleerde TMCs verkregen die varieerden in mate van<br />
quaternisering (25-54%) en thiolering (5-7%) zoals bepaald met 1 H-NMR en Ellman's test. Gel<br />
permeatie chromatografie toonde aan dat er een beperkt aantal intermoleculaire bindingen<br />
w<strong>as</strong> ontstaan. De thiol-groepen in de TMCs bleken bij pH 7.4 snel te oxideren tot disulfides,<br />
199
Nederlandse Samenvatting<br />
terwijl ze bij een pH van 4.0 redelijk intact bleven na 8 uur op 37 graden. Studies met calu-3<br />
cellen lieten zien dat er een lichte daling in cytotoxiciteit is voor de gethioleerde TMCs in<br />
vergelijking met de niet-gethioleerde polymeren met eenzelfde DQ.<br />
Positief geladen nanodeeltjes werden gemaakt van de gethioleerde TMCs en gethioleerd<br />
hyaluronzuur (HA-SH) en hierin werd fluorescent gelabeld ovalbumine als model-eiwit<br />
ingesloten. Deze covalent gestabiliseerde nanodeeltjes bleven stabiel in 0.8 M NaCl, in<br />
tegenstelling tot de deeltjes gemaakt van niet-gethioleerde polymeren die in deze sterk<br />
ionische buffer uit elkaar vielen. Dit toonde aan dat de gestabiliseerde deeltjes bij elkaar<br />
gehouden werden door intermoleculaire zwavelbruggen tussen TMC-SH en HA-SH.<br />
In hoofdstuk 6 werd aangetoond dat de fysische stabiliteit van polyelectroliet nanodeeltjes<br />
van trimethyl chitosan (TMC) en hyaluronzuur (HA) beperkt is onder fysiologische<br />
omstandigheden. Vooral in aanwezigheid van zout kan de stabiliteit verminderen aangezien zij<br />
de ladingsinteracties tussen het positief geladen TMC en het negatief geladen HA geringer<br />
worden. Dit kan de werkzaamheid van het adjuvans effect van de nanodeeltjes voor de<br />
intran<strong>as</strong>ale en mogelijk ook intradermale immunisatie verminderen. Zodoende werd in<br />
hoofdstuk 7 het effect van de stabilisatie van TMC/HA nanodeeltjes door disulfide-bruggen op<br />
hun immunogeniciteit geëvalueerd. Ook werden, als alternatief, de resterende thiol-groepen op<br />
het oppervlak van deze deeltjes gepegyleerd waardoor de positieve lading van deze deeltjes<br />
(enigszins) werd afgeschermd. Mogelijk heeft dit een additioneel gunstig effect aangezien PEG<br />
(poly(ethyleen glycol)) de interacties met de extracellulaire matrix zou kunnen verminderen<br />
(intradermaal) en/of de muco-adhesie zou kunnen verbeteren (intran<strong>as</strong>aal). Covalent<br />
gestabiliseerde nanodeeltjes beladen met ovalbumine (OVA) werden gemaakt met gethioleerd<br />
TMC en gethioleerd HA via gelering op b<strong>as</strong>is van ladings-interactie gevolgd door spontane<br />
vorming van disulfide-bruggen na een incubatie van 3 uur bij pH 7.4 en 37 °C. Tevens werd<br />
maleïmide PEG gekoppeld aan de resterende thiol-groepen aan het oppervlak van de deeltjes<br />
om de positieve lading af te schermen.<br />
De nanodeeltjes hadden een grootte van circa 250-350 nm, een positieve zeta potentiaal en<br />
OVA ladings-efficiënties tot 60%. Pegylering resulteerde in een kleine verlaging van de zeta<br />
potentiaal en een geringe toename in deeltjesgrootte. Gestabiliseerde TMC-S-S-HA deeltjes<br />
(gepegyleerd of niet) hadden een superieure stabiliteit in zoutoplossingen in vergelijking met<br />
niet-gestabiliseerde TMC/HA deeltjes (bestaande uit niet-gethioleerde polymeren). Echter, de<br />
TMC-S-S-HA deeltjes vielen momentaan uiteen wanneer een reagens werd toegevoegd dat<br />
disulfide-bruggen breekt. In zowel de intran<strong>as</strong>ale als intradermale vaccinatie-studie,<br />
200
Nederlandse Samenvatting<br />
vertoonden de gestabiliseerde TMC-S-S-HA deeltjes beladen met OVA superieure<br />
immunogeniciteit in vergelijking met niet-gestabiliseerde deeltjes. Pegylering resulteerde<br />
intran<strong>as</strong>aal in een volledige tenietdoening van de positieve effecten van stabilisatie en het had<br />
geen additioneel effect wanneer deze gepegyleerde deeltjes werden gebruikt voor<br />
intradermale vaccinatie. Concluderend, de stabilisatie van de TMC/HA deeltjes verbeterde de<br />
immunogeniciteit na intran<strong>as</strong>ale en intradermale vaccinatie, echter pegylering van deze<br />
gestabiliseerde deeltjes had geen gunstig effect.<br />
De nieuwe synthese-routes die in dit proefschrift worden beschreven, maken het mogelijk<br />
voor elke specifieke toep<strong>as</strong>sing de gewenste structurele variant van TMC te produceren. Zo<br />
kunnen de mate van quaternisering en acetylering worden gevarieerd zonder dat er andere<br />
wijzigingen, zoals O-methylering of ketenbreuk, in het polymeer optreden. Daarna<strong>as</strong>t wordt<br />
een methode beschreven om op een controleerbare manier thiol-groepen in TMC te<br />
introduceren. Het gecontroleerd kunnen variëren van de structurele eigenschappen van TMC<br />
geeft de mogelijkheid om op de juiste manier structuur-activiteit-relaties te onderzoek in in<br />
vitro biologische <strong>as</strong>says en in vivo (n<strong>as</strong>ale) vaccinatie-studies. Verder werd het inzicht in het<br />
werkingsmechanisme van TMC verdiept door verschillende, potentieel cruciale, stappen in<br />
intran<strong>as</strong>ale vaccinatie op een systematische manier te onderzoeken. Ten slotte werd een<br />
veelbelovend nieuw, covalent gestabiliseerd, drager systeem onderzocht. De bemoedigende<br />
resultaten met deze nieuwe TMC-varianten geven meer dan voldoende aanleiding tot verdere<br />
pre-klinische en eventueel klinische ontwikkeling.<br />
201
Dankwoord<br />
Dankwoord<br />
Beste mensen, mijn mooie periode als AIO zit erop! Natuurlijk zijn er wel eens wat<br />
tegenslagen geweest en heb ik af en toe hard moeten werken, maar al met al is het project<br />
soepel verlopen en heb ik me erg vermaakt op het Went. Dit is dan ook naar mijn mening het<br />
grootste compliment voor iedereen met wie ik, op welke manier dan ook, te maken mee heb<br />
gehad de afgelopen jaren. Zonder jullie hulp, enthousi<strong>as</strong>me en kritische noot zou dit boekje<br />
nooit op deze prettige wijze zijn voltooid. Heel erg bedankt! Een aantal mensen wil ik graag in<br />
het bijzonder bedanken:<br />
In den beginne is er Wim H, mijn promotor: ‘Wat een biologische meuk dat TMC (Thai<br />
M<strong>as</strong>sage Center)! Kun je er geen hydrogel van maken?’ Waarna de voetbaluitslagen van het<br />
weekend werden besproken, vaak met enig leedvermaak aangezien Ajax de afgelopen 4 jaar<br />
niet de beste periode uit haar geschiedenis kende. Beste Wim, bedankt dat je deze PSV-er<br />
zoveel vrijheid en vertrouwen heb gegeven maar ook voor me klaar stond als er even spijkers<br />
met koppen geslagen moesten worden.<br />
Beste Wim J, mijn andere promotor. Hoewel officieel p<strong>as</strong> op het einde aan boord geklommen,<br />
w<strong>as</strong> je eigenlijk vanaf het begin al de grote coördinator van het TMC (Toxiciteit Misschien<br />
Cruciaal) project. Mooie discussies en een gezonde rivaliteit maakten de tripjes naar Leiden<br />
een leuk vooruitzicht. Dank voor je wetenschappelijke hulp en tekstuele creativiteit!<br />
Dan zijn er natuurlijk nog mijn alter-AIOs uit Leiden, Suzanne en Bram. Toch mooi dat we na<br />
3 jaar vergaderen onze projecten hebben kunnen afsluiten met een gezamenlijk in vivo<br />
experiment! Heel veel succes in de toekomst en bedankt voor de leuke samenwerking. Ook Elly<br />
mag hier niet ontbreken voor alle hulp bij de cel-experimenten en Joke voor de bijdrage op het<br />
gebied van de intradermale vaccinaties.<br />
Mijn studenten, Sara de Madrid, much<strong>as</strong> graci<strong>as</strong> por todo en Francesco di Camarino, grazie<br />
mille! Steffen, jouw vrijdagmiddag-experimentje heeft mij een pikstart bezorgd en ook daarna<br />
heb je me regelmatig de weg gewezen uit het organisch-chemische moer<strong>as</strong>. Bijna w<strong>as</strong> ik in een<br />
chemicus veranderd maar gelukkig heeft het niet zover hoeven komen, dank. Natuurlijk mag<br />
ook de man die in elk dankwoord staat vermeld ook hier niet ontbreken: Mies, zonder jou ziet<br />
het leven er toch een stuk minder rooskleurig uit, crying at the Viscotek!<br />
203
Dankwoord<br />
Thom<strong>as</strong> en Sven van het VUmc, geweldig dat jullie als immunologen zo enthousi<strong>as</strong>t met onze<br />
polymeren aan de gang wilden gaan. Uiteindelijk heeft het nog mooie resultaten opgeleverd.<br />
De mensen van Nobilon en Intervet, de twee uur durende treinreis w<strong>as</strong> snel vergeten door de<br />
stimulerende en prettige sfeer die er in Boxmeer heerste. Heel veel succes in deze tijd.<br />
Dr. ‘so you think you can dance’ Ethlinn, dat je hebt tijdens je eigen lange weg nog tijd hebt<br />
gevonden om me te helpen met experimenten en tips en trucs te geven, heb ik zeer kunnen<br />
waarderen. Ook anderen die me op de een of andere manier met de experimenten hebben<br />
geholpen mogen natuurlijk niet ontbreken: Maryam, Enrico, Roel, P<strong>as</strong>cal en Roberta. Dank<br />
voor jullie hulp!<br />
Het sterrenensemble van Z605, Marina (sorry if I te<strong>as</strong>ed you too much), Hajar (sorry if I<br />
te<strong>as</strong>ed you too much) and Amir V (sorry if I te<strong>as</strong>ed you too much), thanks <strong>for</strong> the good times on<br />
and off the lab. De oude garde Sophie, Marjan, Marion, Frankie, Sabrina, Marcel, Jozef ‘je<br />
gedraagt je wel hè? Ik hoef maar dìt te zien...’ en de rijpe garde Wouter, Joris en Ellen bedankt<br />
voor de gezelligheid en de wijsheden. De jonge hindes van biofarmacie: Inge, Roy, DaMarkus,<br />
Bart, speedy Maria, Alex, Pieter, Emmy, Albert, Melody, Kimberley, Audrie, Luis, Kim, Barbara,<br />
Lydia, Peter en de anderen, heel veel plezier en succes met het afronden!<br />
Gelukkig waren er nog voldoende mensen die me na<strong>as</strong>t het werk bezig hielden. Mijn<br />
teamgenoten van de Ody-zondag, we gaan er weer een frank en vrij seizoen van maken! De<br />
oud-huisgenoten van de ple<strong>as</strong>uredome, Sindre & Torkel, Japus, de Farmacie-gang, dank voor<br />
jullie vriendschap. Gert, zonder je voedzame maaltijden en de waardevolle gesprekken zou er<br />
slechts een skelet resteren, got to get behind the mule in the morning and plow!<br />
Amir-idjan, alle (oude) wijze mannen komen uit het Oosten. Vaak hebben we Utrecht onveilig<br />
gemaakt en nog vaker heb je me uitgelegd hoe de dingen werkelijk in elkaar zitten. Over vijf<br />
jaar zitten we bij je op de veranda ergens in Toronto met een lekkere whisky maar eerst heb ik<br />
je nog hier even nodig. Merci!<br />
Nelis, met jou iets ondernemen staat garant voor een succesvol avontuur. Met -5 ⁰C bananen<br />
roosteren aan de Sunshine Co<strong>as</strong>t, raven op surrealistisch Roskilde of afreizen naar Boxmeer<br />
om muizen te pesten, het w<strong>as</strong> allemaal enerverend. Je bent een unieke gozer en ik weet zeker<br />
dat wij elkaar blijven zien.<br />
204
Dankwoord<br />
Lieve Femke, het avontuur met jou is eigenlijk p<strong>as</strong> net begonnen maar hoe meer ik van je<br />
ontdek hoe meer ik er van zeker van ben dat we een mooie toekomst voor ons hebben.<br />
En als laatste wil ik nog mijn familie bedanken. Oma, die afspraak van vier jaar geleden<br />
komen we allebei na, geweldig dat je hier bij kan zijn. Lieve Marc en Nienke, nog even wachten<br />
en dan is het jullie dag. Lieve ‘spap en ‘smam, dank voor de liefde, interesse en vertrouwen.<br />
Zonder jullie w<strong>as</strong> dit allemaal niet mogelijk geweest dus geniet ervan.<br />
Tijd voor een nieuw hoofdstuk!<br />
Rolf<br />
205