Role of Intestinal Microbiota in Ulcerative Colitis

1046 J. Holck et al. / Process Biochemistry 46 (2011) 1039–1049
Fig. 6. MS/MS high energy CID spectrum of galA2rha2gal2 (R1, Fig. 5) illustrating the fragmentation pattern and nomenclature, and the proposed structure.
ysis of galA 3rha 2gal 2 revealed the co-existence of two different
structures (Fig. 8A and B). One of the structures consisted of two
galacturonic acids, the non-reducing being unsaturated, followed
by a rha-galA-rha segment with both rhamnoses substituted with
a single galactose (Fig. 8A). This molecule corresponded well with
the previous findings of the connecting linkage between homo- and
rhamnogalacturonan I in apple pectin [44] but this is the first report
of this connecting linkage in sugar beet pectin. The second structure
showed an alternating galacturonic acid-rhamnose backbone
structure, but with a galacturonic acid in both the reducing and
the non-reducing end, the latter being unsaturated. Galactose was
distributed evenly (Fig. 8B). These two structures indicated that
rhamnogalacturonan I is covalently linked to homogalacturonan in
both the reducing and the non-reducing end in sugar beet pectin.
Particularly the finding of the two consecutive galacturonic acid
moieties in R6 supports the comprehension of pectin being built of
consecutive homogalacturonan and rhamnogalacturonan I structural
elements. In turn, this finding rejects the alternative model
structure proposed by Vincken [45] with homogalacturonan being
a side chain of the rhamnogalacturonan I backbone. While the finding
of the molecule with a galacturonic acid in both the reducing
and non-reducing end also supports the classic pectin model (Fig. 1)
it cannot rule out the alternative model.
HO
OH
O
OH
O
O
HO
A 1,3
C
H 3
Fig. 7. Type A cross-ring cleavage of rhamnose in a RGI subfragment (CRFx). As an
example of a potential cleavage pattern an A 1,3 cleavage is indicated by a dotted line.
Principle adapted from [43].
O
R 1
O
O
R 2
Table 5
Identified compounds, peak identity and yields from RGI permeate. Yield in mg/g
initial pectin. For peak identities, refer to (Fig. 5).
Peak ID Structure Yield mg/g pectin
R1 galA2rha2gal2 16.5
R2 galA2rha2gal 13.9
R3 galA2rha2 20.2
R4 galA3rha3gal3 5.1
R5 galA3rha3gal2 6.7
R6 galA3rha3
galA3rha2gal2
5.2
Total 67.6
The yield for each RGI like oligosaccharide ranged from 5 to
20 mg/g pectin with a total of 67.6 mg/g pectin (Table 5). The yield of
each type of compound per run ranged from 47 to 184 mg/g sample
load. This yield per sample load is larger than reported for homogalacturonides
due to a smaller population of different molecules
in the sample. With the employed chromatography conditions, this
yield was equivalent to an isolation efficiency of 0.1–0.4 mg/peak h,
which was much lower than for the homogalacturonides, but the
sample load was never increased in order to increase the efficiency.
The yield of R6 was based on an average molecular weight of the
two compounds described, since no quantification was possible.
According to the mass balance, the RGI permeate accounted for
110 mg/g pectin. Out of this, approximately 60% was thus recovered
as defined RGI like molecules. As seen for the homogalacturonides,
the chromatographic elution profile of RGI permeate (Fig. 5) indicated
the presence of small amounts of RGI like molecules with
higher DP than the one observed in R6, but the lack of MS/MS verification
prevented these peaks from further investigation and yield
calculations.
3.5. In vitro fermentation
Quantitative RealTime PCR from in vitro fermentations showed
differences in the relative ratio of Bacteroidetes and Firmicutes
dependent on the type of substrate. The density of Bacteroidetes
was significantly lower in both fermentations compared to the