Haematologica 2003 - Supplements

Scientific sessions
1. Development of normal and malignant
plasma cell
001
The Essential Role of Oncogenic FGFR3 in
Maintenance of t(4:14) Myeloma.
Suzanne Trudel*, Scott Ely#, Yildiz Farooqui*, Davide F.
Robbiani*, Maurizio Affer*, Marta Chesi*, Peter L.
Bergsagel*.
Medicine, Division of Hematology/Oncology*, Cornell Medical
College, New York, NY, USA; Pathology#, Cornell Medical
College, New York, NY, USA.
Chromosomal translocations to the immunoglobulin heavy-chain
locus on chromosome 14q32 are present in the majority of
multiple myeloma (MM) patients and may represent the first and
defining genetic event that leads to the development of MM. The
t(4;14) translocation which occurs in approximately 15% of
patients results in the dysregulated expression of fibroblast
growth factor receptor 3 (FGFR3) and MMSET. Wild-type
FGFR3 appears to be weakly transforming in a hematopoietic
murine model. The subsequent acquisition of FGFR3 activating
mutations is associated with disease progression and is strongly
transforming in experimental models. These findings suggest a
pathogenic correlation between FGFR3 expression and myeloma
however it remains to be proven how dysregulation of FGFR3
mediates an early oncogenic process in MM and whether FGFR3
is required for tumor maintenance. We have used
pharmacological inactivation of FGFR3 to address this question
directly in human MM.
We have developed 3 screening assays for identification of
FGFR3 inhibitors and have used these to establish PD173074 as a
selective inhibitor of FGFR3. Using this inhibitor we confirmed
that inactivation of FGFR3 blocks its oncogenic potential. We
have previously shown that activated forms of FGFR3 induce
transformation of NIH 3T3. Using this same assay we tested the
ability of PD173074 to inhibit the Y373C-FGFR3 induced
transformation of NIH 3T3 cells. Although it had no effect on
Ras-induced transformation, it completely inhibited foci
formation induced by activated FGFR3. Similarly, PD173074
prevented in vivo growth of Y373C-FGFR3 transfected NIH 3T3
cells in nude mice but had no inhibitory effect on growth of Ras
V12 expressing cells.
To establish that FGFR3 activation provides a critical and nonredundant
pro-proliferative and anti-apoptotic signal in MM we
exposed FGFR3 expressing myeloma cell lines to PD173074.
PD173074 inhibited cell proliferation of FGFR3 expressing
KMS11 and KMS18 cells with an IC50 of 12.5 nM and 20 nM,
respectively. 8226 cells, which lack FGFR3 expression, displayed
no growth inhibition demonstrating that PD173074 exhibits
minimal nonspecific cytotoxicity. Further characterization of this
finding demonstrated that inhibition of cell growth is related to
G0/G1 cell cycle arrest. PD173074 also induced delayed, doseresponsive
apoptosis of these cells. Immunohistochemical
analysis demonstrated an increase in cleaved caspase 3 positivity,
suggesting that FGFR3 activation protects MM cells from
caspase-dependent cell death. To explain the marked delay in
apoptosis we speculated that inhibition of FGFR3 in these cells
induces cell cycle arrest and differentiation. Inhibition of FGFR3
resulted in the differentiation of KMS11 and KM18 cells from a
plasmablast-like phenotype to a more mature plasma cell
characterized morphologically and by the induction of CD31
expression and increase in light chain secretion. In addition,
FGFR3 inactivation had similar affects in vivo inducing growth
arrest, apoptosis and differentiation of KMS11 tumors in a
xenograph mouse model. Most importantly the reversion of the
malignant phenotype was associated with delayed tumor
progression and enhanced overall survival of PD173074 treated
mice. These results provide evidence that FGFR3 is important
for genesis and maintenance of myeloma. Further, they validate
FGFR3 as a therapeutic target for a subset of MM patients.
002
The role of MMSET in t(4;14) myeloma
Marta Chesi, Davide F. Robbiani, Maurizio Affer, Suzanne
Trudel, W. Michael Kuehl and P. Leif Bergsagel
Weill Medical College of Cornell University and Genetics
Department, NCI
The t(4;14)(p16;q32) translocation, that occurs in about 15-20%
of multiple myeloma (MM), causes the concomitant
dysregulation of two genes by their juxtaposition to the two
immunoglobulin enhancers. FGFR3, is brought on the der(14)
under the control of the IgH 3’ enhancer and the intronic
enhancer, Emu, is translocated on der(4) where it dysregulates
MMSET expression. The breakpoints are clustered in two groups:
one falls in the 5’ UTR of MMSET, outside the coding sequence;
the other falls into the 5’ coding exons, resulting into a N-term
truncation of MMSET protein. Although the acquisition by the
tumor cells of FGFR3 activating mutations indicates a role for
FGFR3 in tumor progression in the few informative cases, we
and other investigators reported the loss of der(14) or FGFR3
expression in about 20% of t(4;14) myelomas. On the contrary,
there is only one example of a t(4;14) MM that has lost der(4)
and does not contain Ig/MMSET hybrid transcripts. Therefore
MMSET seems to be the crucial gene in t(4;14) myeloma.
MMSET belongs to the trithorax family of nuclear proteins
characterized by the presence of a SET domain and several PHDtype
zinc fingers and involved in chromatin remodeling. One of
them, MLL, is located on 11q23 and translocated in acute
leukemia. MMSET, also known as Nsd2, Trx5 and WHSC1, is
the gene deleted in Wolf Hirsch Syndrome. Highly related genes,
Nsd1 and Nsd3 have also been implicated in neoplastic
transformation and found translocated in AML. There are two
classes of MMSET mRNA transcripts, based on alternative
splicing: MMSET type I encodes for a 674 aa protein, MMSET
type II extends at the 3’ end of the type I and encodes for a 1365
aa protein. Although we have found that the type I protein can
block transformation of NIH3T3 fibroblasts by a variety of
oncogenes, there is no direct evidence that MMSET can function
as an oncogene. To study the oncogenic contribution of MMSET
dysregulation in myeloma we followed two approaches 1) We
generated retroviral vectors carrying EGFP, MMSETI and
MMSETII fused to an IRES-neomycin cassette, and infected
human myeloma cell lines (HMCL) that do not have a t(4;14)
translocation and do not express MMSET; 2) we generated
transgenic mice in which MMSETI and –II are under the control
of the lck minimal promoter and Emu – a strategy that had
previously enabled us and others to generate mice expressing
transgenes in B and T cell lineages. Although we obtained
HMCL neomycin resistant clones expressing MMSET mRNA,
we were unable to detect MMSET protein expression. Under the
same condition, however, MMSET exogenous protein was
S88