A possible oligosaccharide-conjugate vaccine candidate for clostridium difficile is antigenic and immunogenic

Article abstract of DOI:10.1016/j.chembiol.2011.03.009

Nosocomial infections with the Gram-positive pathogen Clostridium difficile pose a major risk for hospitalized patients and result in significant costs to health care systems. Here, we present the chemical synthesis of a PS-II hapten of a cell wall polysa

Full text of DOI:10.1016/j.chembiol.2011.03.009

Chemistry & Biology
Article
A Possible Oligosaccharide-Conjugate Vaccine
Candidate for Clostridium difficile
Is Antigenic and Immunogenic
Matthias A. Oberli,1,4,5 Marie-Lyn Hecht,1,4 Pascal Bindschadler,1,6 Alexander Adibekian,1,7 Thomas Adam,3
¨
and Peter H. Seeberger1,2,
*
1Department of Biomolecular Systems, Max-Planck Institute for Colloids and Interfaces, 14476 Potsdam, Germany
2
¨
Freie Universitat Berlin, Arnimallee 12, 14195 Berlin, Germany
3Institut fu¨ r Mikrobiologie und Hygiene, Universita¨ tsmedizin Charite´ , Hindenburgdamm 27, 12203 Berlin, Germany
4These authors contributed equally to this work
5Present address: Department of Chemistry, Massachusetts Institute of Technology, Cambridge, MA 02139, USA
6Present address: BASF SE, GVA/IO–B009, 67056 Ludwigshafen, Germany
7Present address: The Scripps Research Institute, 10550 North Torrey Pines Road, La Jolla, CA 92037, USA
*Correspondence: seeberger@mpikg.mpg.de
DOI 10.1016/j.chembiol.2011.03.009
SUMMARY
rates also increased significantly (Kuijper et al., 2007). The North
American and European outbreaks coincided with the emer-
Nosocomial infections with the Gram-positive path- gence of a hypervirulent strain of C. difficile, alternatively desig-
nated by the synonymous terms as PCR ribotype 027, toxin
type III, NAP1, and BI (McDonald et al., 2005; Loo et al., 2005).
Thehypervirulence of ribotype 027 hasbeen ascribed to itshigher
ogen Clostridium difficile pose a major risk for hospi-
talized patients and result in significant costs to
health care systems. Here, we present the chemical
synthesis of a PS-II hapten of a cell wall polysaccha-
ride of hypervirulent ribotype 027 of C. difficile. Mice
were immunized with a conjugate consisting of the
synthetic hexasaccharide and the diphtheria toxoid
variant CRM197. The immunogenicity of the glycan
repeating unit was demonstrated by the presence
of specific IgG antibodies in the serum of immunized
˚
toxin yields and an increased rate of sporulation (Akerlund et al.,
2008). Higher toxin content is due to an additional toxin referredto
as thebinarytoxinanda genetic mutationin atoxinregulator gene
(tcdC), encoding a negative regulator of the C. difficile pathoge-
nicity locus. The isolates obtained during the North American
and European epidemics were genetically closely related and,
in addition, resistant to fluoroquinolones (Clements et al., 2010).
The medical significance of C. difficile has prompted intense
mice. Murine monoclonal antibodies interact with the studies aiming to elucidate the structural composition of its cell
wall. Two capsular polysaccharides, PS-I and PS-II, were identi-
fied (Ganeshapillai et al., 2008). PS-I has a branched pentaglyco-
synthetic hexasaccharide, as determined by micro-
array analysis. Finally, we found that specific IgA
antibodies in the stool of hospital patients infected
with C. difficile recognize the synthetic PS-II hexa-
syl phosphate repeating unit, and PS-II a hexaglycosyl phos-
phate repeating unit. Both are found on the highly virulent
ribotype 027 (Figure 1). Currently, to our knowledge, no licensed
vaccine against C. difficile is available.
saccharide hapten.
Several marketed vaccines are based either on natural poly-
saccharides alone (Lucas and Reason, 1999) or on polysaccha-
rides linked to immunogenic protein carriers (Hecht et al., 2009;
INTRODUCTION
The Gram-positive bacteria of the genus Clostridium difficile have Roy, 2008). Polysaccharide-protein conjugate vaccines based
long been recognized as the cause of a range of gastrointestinal on isolated carbohydrates are an effective measure against at
diseases (Hookman and Barkin, 2009). Infection and the develop- least five different bacteria: Haemophilus influenzae type b,
ment of C. difficile-associated diseases (CDADs) are linked to the Streptococcus pneumoniae, Neisseria meningitidis, Salmonella
use of antibiotics that disrupt the normal intestinal flora and allow typhi, and Staphylococcus aureus infections (Ada and Isaacs,
for proliferation of C. difficile (Thomas et al., 2003). C. difficile 2003). A synthetic carbohydrate-protein conjugate vaccine
infection in its most severe form can cause toxic megacolon against Haemophilus influenzae type b was developed and mar-
with subsequent colonic perforation, peritonitis, shock, and keted in Cuba (Verez-Bencomo et al., 2004; Roy 2008).
death. Furthermore, C. difficile is a major cause of diarrhea in
Here, we report the synthesis of a hapten of the C. difficile
hospital and long-term care facility patients due to the frequent PS-II, its conjugation to the diphtheria toxoid CRM197, and the
use of antibiotics, contamination of these facilities with resistant production of monoclonal antibodies that specifically recognize
spores, and because of the high density of susceptible persons. the glycan hapten. Polysaccharide-specific IgA antibodies were
A dramatic increase in C. difficile incidents was recorded in many detected in patients diagnosed with C. difficile infections. The
developed countries (Polk et al., 2006) starting with reports of PS-II structure was selected as synthetic target because it was
´
hospital outbreaks in Canada in 2003 (Pepin et al., 2004). With
reported to be found on several C. difficile strains, including ribo-
the increasing severity of the incidents, relapse and mortality type 027 (Ganeshapillai et al., 2008).
580 Chemistry & Biology 18, 580–588, May 27, 2011 ª2011 Elsevier Ltd All rights reserved

Chemistry & Biology
A Possible Vaccine Candidate for C. difficile
Figure 1. Retrosynthetic Analysis of Hexasaccharide Repeating Unit 2
(A) Structure of PS-II found on C. difficile.
(B) Retrosynthetic analysis of PS-II hapten analog 2.
RESULTS
synthesized from commercially available unprotected monosac-
charides following known methodology.
Preparation of the reducing terminus commenced with glyco-
Oligosaccharide Repeating Unit Synthesis
The PS-II repeating units are interconnected via a (1/6) phos- sylation of the protected spacer N-benzyl-N-benzyloxycarbonyl-
phate diester linkage in the natural polysaccharide (Figure 1). 5-aminopentanol 8 (Delcros et al., 2002) with mannose building
We aimed to synthesize oligosaccharide 2 (Figure 1), which block 9 (Figures 1 and 2). The 2-O-benzoyl and 3-O-levulinoyl
constitutes a nonphosphorylated PS-II hexasaccharide hapten. protection groups on building block 9 were crucial for the
The oligosaccharide was designed to carry a primary amine at success of this glycosylation because when the alternative
the reducing terminus via a spacer to facilitate conjugation to mannose-based building block bearing 2-O-acetate and
a protein carrier and attachment to microarrays. Based on our 3-O-fluorenylmethyloxycarbonyl (Fmoc) protecting groups was
retrosynthetic analysis, the hexasaccharide will be assembled used instead, it afforded mainly the orthoester product (Kong,
`
from the monosaccharide building blocks 3 and 4, and the disac- 2007; Ravida et al., 2006) and only small amounts of the desired
charide building block 5 that appears twice in the target struc- product. The levulinoyl ester in mannose glycoside 10 was selec-
ture. Disaccharide 5 will be derived in turn from monosaccharide tively cleaved using hydrazine monohydrate to reveal the C3
building blocks 6 and 7. The building blocks 3, 4, 6, and 7 were hydroxyl group to give glycosyl acceptor 11.
Chemistry & Biology 18, 580–588, May 27, 2011 ª2011 Elsevier Ltd All rights reserved 581

Chemistry & Biology
A Possible Vaccine Candidate for C. difficile
Figure 2. Synthesis of Building Blocks 11, 15, and 18
Reagents and conditions: (a) NIS, TMSOTf, CH2Cl2, 90%; (b) N2H4$H2O, AcOH, Py, quantitative; (c) TMSOTf, CH2Cl2, À30ꢀC, 78%; (d) Et3SiH, TfOH,
CH2Cl2, À78ꢀC, 68%; (e) LevOH, DMAP, DIPC, CH2Cl2, 94%; (f) NIS, aqueous HCl, THF, 96%; (g) CF3C(NPh)Cl, Cs2CO3, CH2Cl2, 78%; (h) FmocCl, Py, CH2Cl2,
72%; (i) NBS, aqueous HCl, THF, 70%; (j) CF3C(NPh)Cl, Cs2CO3, CH2Cl2, quantitative. Lev, Levulinoyl; Bn, benzyl; Bz, benzoyl; Cbz, benzyloxycarbonyl; TCA,
trichloroacetyl.
Disaccharide building block 5 resulted from the union of galac- were obtained when disaccharide 12 was treated with triethylsi-
tosamine 6 (Chen et al., 2005) with the previously characterized lane and triflic acid at À78ꢀC. Other methods that relied on using
`
glucosyl phosphate 7 (Ravida et al., 2006) via trimethylsilyl tri- trimethylsilyl trifluoromethanesulfonate as Lewis acid, or sodium
fluoromethanesulfonate-mediated activation of the anomeric cyanoborohydride as a reducing agent, furnished inseparable
dibutyl phosphate leaving group in 7 followed by nucleophilic mixtures of the 4-hydroxyl and 6-hydroxyl-regioisomers. The
substitution by 6 to afford disaccharide 12 (Figures 1 and 2). free C4 hydroxyl group in disaccharide 13 was subsequently
The selective opening of the benzylidene acetal in compound masked as levulinoyl ester to afford glycosylating agent 5.
12, to afford disaccharide 13 bearing a 6-O-benzyl protection
Hexasaccharide assembly commenced with the glycosylation
group and no protection group on the C4 hydroxyl group, of monosaccharide 11 with phenyl selenide 5 to furnish trisac-
´
strongly depended on the reaction conditions. Best results charide 19 with a yield of up to 61% (Depre et al., 1999). To
582 Chemistry & Biology 18, 580–588, May 27, 2011 ª2011 Elsevier Ltd All rights reserved

Chemistry & Biology
A Possible Vaccine Candidate for C. difficile
improve the coupling yields, we changed the synthetic strategy was selected from the multitude of methods for conjugation of
by replacing the anomeric leaving group in 5. The phenyl sele- carbohydrates to proteins (Kuberan and Linhardt, 2000; Stall-
nide in 5 was replaced by a different anomeric leaving group, forth et. al, 2009). First, the amine group of the spacer moiety
by a N-phenyl trifluoroacetimidate group (Yu and Tao, 2001). in hexasaccharide 2 was reacted with one of the vinylogous
Disaccharide 5 was converted, via lactol 14, to the correspond- ester groups of 3,4-di-ethoxy-3-cyclobutene-1,2-dione in
ing glycosyl N-phenyl trifluoroacetimidate 15. As envisioned, pH 7.2 phosphate buffer to form the corresponding monoamine
glycosylation of nucleophile 11 with disaccharide N-phenyl tri- 24 that was purified by reverse-phase HPLC (Figure 4A). The
fluoroacetimidate 15 yielded 82% of the desired product, which remaining ester group of monoamide 24 was subsequently
compared favorably to the 61% obtained when phenyl selenide coupled with the 3-amino groups of lysine on the diphtheria
5 was used as disaccharide-glycosylating agent. The C2 partici- toxoid CRM197 in bicarbonate buffer at pH 9.0 to afford the
pating trichloroacetamido group of galactosamine in 15 ensured neoglycoconjugate. Successful conjugation was confirmed by
the exclusive formation of the b linkage. Treatment of trisaccha- SDS-PAGE (Figure 4B), and the oligosaccharide/CRM197 ratio
ride 19 with hydrazine monohydrate resulted in cleavage of the was determined by matrix-assisted laser desorption/ionization
levulinoyl ester and furnished 20 bearing a free hydroxyl group time-of-flight mass spectrometry (MALDI-TOF MS) (Figure 4C).
ready for the next glycosylation. Glycosylation of trisaccharide The mass analysis of CRM197 yielded a m/z ion at 58.6 kDa.
20 with thioglycoside 4 afforded a 55% yield of tetrasaccharide The mass spectrum of the neoglycoconjugate revealed mass
21. The yield of this glycosylation was again improved when peaks between 59.9 and 67.3 kDa corresponding to mono- to
N-phenyl trifluoroacetimidate glycoside 18 was employed heptavalent glycoconjugates. On average four haptens 2 were
instead. This glycosylating agent in a mixture of methylene chlo- loaded on the diphtheria toxoid.
ride and diethyl ether at À45ꢀC afforded 83% yield of tetrasac-
charide 21 containing the a-linked glucose. Glucose building Immunization and Monoclonal Antibodies
block 18 was prepared from known alcohol 16 (van Steijn To test the immunogenicity of the PS-II hapten, two female
et al., 1992) via the procedure that was used for the conversion C57BL/6 mice were immunized with the neoglycoconjugate.
of 5 to 15. Treatment of tetrasaccharide 21 with triethylamine Mice were injected three times subcutaneously (s.c.) with the
resulted in cleavage of the Fmoc group and liberation of the glycoconjugate at 2-week intervals. For each injection 15 mg
hydroxyl group, ready for the next glycosylation. Hexasacchar- protein, as determined by Bradford analysis, was used. Consid-
ide 23 was obtained by the trimethylsilyl trifluoromethanesulfo- ering an average loading ratio of four haptens on each protein,
nate-catalyzed glycosylation of tetrasaccharide 22 with disac- the corresponding carbohydrate content was calculated as
charide building block 15.
being 1.3 mg. The anti-hapten 2 antibody titers were monitored
Hexasaccharide 23 was freed from all protecting groups via by glycan microarray analysis. Microarrays were designed for
a three-step procedure. First, the N-trichloroacetyl groups high-throughput analysis, such that 64 samples could be
were transformed into N-acetyl groups by treatment with tributyl analyzed on one array with each well displaying hapten 2 and
stannane and azobisisobutyronitrile (AIBN) in toluene at 90ꢀC seven control sugars in quadruplicates (Figure 5A). The two
´
(Figures 1 and 3) (Belot and Jacquinet, 2000; Rawat et al., immunized mice produced IgG antibodies that bound specifi-
2008). Subsequent saponification using potassium hydroxide cally to hapten 2 (Figure 5B), demonstrating that hapten 2 is
in tetrahydrofurane and methanol was followed by hydrogena- immunoreactive. An increase in the level of anti-hapten 2 specific
tion using hydrogen gas and palladium on charcoal. Thereby, IgG antibodies over time was observed with mouse 2805.
hexasaccharide hapten 2 was obtained (Figures 1 and 3).
To generate monoclonal antibodies, spleenocytes of the
The anomeric region of the 13C and 1H-NMR spectra of hexa- immunized mice were fused to myeloma cells by the traditional
¨
saccharide 2 are compared to that of the natural polysaccharide hybridoma technique (Kohler and Milstein, 1975). The individual
(see Supplemental Experimental Procedures), confirming the hybridoma clones were screened to identify clones that produce
structure of hexasaccharide 2 despite slight differences. During anti-hapten 2 antibodies. Three hybridoma clones that secrete
the preparation of this manuscript, an alternative synthesis of specific antibodies were obtained (Figure 5C). All three
a closely related hexasaccharide was published (Danieli et. al, hybridoma clones were derived from mouse 2805. Although
2011) where a similar comparison of NMR spectra of synthetic the monoclonal antibodies C2805.7 and C2805.21 bound exclu-
hexasaccharide and natural polysaccharide was studied.
sively to hapten 2, antibody C2805.25 also interacted with
glucose on the array.
Preparation and Characterization
of Oligosaccharide-Protein Conjugate
Specific IgA Antibodies in Infected Hospital Patients
Polysaccharide vaccines provoke a T cell-independent immune Given the immunogenicity of the hexasaccharide hapten 2 in
response and do not induce an immunoglobulin class switch. mice, we wanted to know whether patients with CDAD produce
Therefore, polysaccharides are conjugated to immunogenic antibodies against the native glycopolymer. To this end, stool
carrier proteins that, unlike polysaccharides, induce a T cell- supernatants of ten hospitalized patients with and without
dependent immune response. The synthetic hapten 2 of the C. difficile infection, as confirmed by the VIDAS immunoassay
´
C. difficile glycopolymer PS-II was conjugated to the protein (bioMerieux) that detects toxins A and B, were analyzed. Stool
carrier CRM197. The diphtheria toxoid CRM197 was chosen as supernatant rather than serum was chosen because the contact
a carrier because it is a constituent of licensed vaccines site of the immune system with the cell surface glycopolymer is
(Barocchi et al., 2007). A method based on the selective reaction the intestinal mucosa. Accordingly, IgA rather than IgG anti-
of the primary amine with squaric acid diester (Tietze et al., 1991) bodies are chiefly responsible for clearance of the pathogen
Chemistry & Biology 18, 580–588, May 27, 2011 ª2011 Elsevier Ltd All rights reserved 583

Chemistry & Biology
A Possible Vaccine Candidate for C. difficile
Figure 3. Synthesis of Hexasaccharide 2
Reagents and conditions: (a) TMSOTf, CH2Cl2, À30ꢀC, 82%; (b) N2H4$H2O, Py, AcOH, CH2Cl2, 91%; (c) 18, TMSOTf, Et2O, CH2Cl2, À45ꢀC, 83%; (d) Et3N,
CH2Cl2, 85%; (e) 15, TMSOTf, CH2Cl2, À30ꢀC, 63%; (f) 1. Bu3SnH, AIBN, toluene, 68%; 2. KOH, MeOH, THF, 86%; 3. H2, Pd/C, AcOH, THF, MeOH, H2O, 95%.
from the gut. In order to monitor the immune response in the 2093, 2118, and 2121, which had not been diagnosed with
epithelia against hexasaccharide 2, the glycan arrays described C. difficile toxin-positive disease. A possible explanation is colo-
above were incubated with the stool supernatants, and bound nization with a nontoxigenic C. difficile strain or previous contact
IgA antibodies were visualized. Three persons had high titers with the bacterium.
of anti-hexasaccharide 2 IgA antibodies in their stool (Figure 6).
Of these three patients, two had been diagnosed with C. difficile DISCUSSION
toxin A/B-positive disease, whereas the third patient had
a borderline VIDAS test. Low amounts of anti-hexasaccharide Wereportanelegantsynthesis ofC. difficile PS-IIhexasaccharide
2 recognizing IgA antibodies were also detected in patients hapten that correlates with the structural assignment based on
584 Chemistry & Biology 18, 580–588, May 27, 2011 ª2011 Elsevier Ltd All rights reserved

Chemistry & Biology
A Possible Vaccine Candidate for C. difficile
Figure 4. Conjugation and Analysis of the Hexasaccharide 2-CRM197 Glycoconjugate
(A) Hexasaccharide 2 was reacted with the carrier protein CRM197 via squaric acid route to yield a polyvalent neoglycoconjugate.
(B) SDS-PAGE analysis of the conjugation. Samples were electrophoresed on 12.5% SDS-PAGE gels and stained with Coomassie blue.
(C) MALDI-TOF MS analysis of the neoglycoconjugate. CRM197 with a m/z peak at 58.5 kDa; hexasaccharide 2-CRM197 conjugates with m/z peaks between 59.9
and 67.3 kDa.
isolated material. The hexasaccharide was assembled from four hospital patients. Two patients with significantly increased
monosaccharide building blocks using an efficient and conver- C. difficile toxin A and B levels and one patient with a borderline
gent approach. A neoglycoconjugate comprising the hexasac- test displayed high amounts of highly specific anti-hexasacchar-
charide hapten and the immunogenic carrier protein CRM197 ide 2 IgA antibodies in their excrement. These observations
was obtained. The outcome of the conjugation process was suggest that native glycopolymer PS-II exposes antigenic deter-
monitored by MALDI-TOF MS and SDS-PAGE. Mice were immu- minants that are the targets of the immune response induced by
nized with the neoglycoconjugate, and IgG antibody production some patients infected with C. difficile. Antibodies in stool are
against hexasaccharide 2 was monitored by glycan microarray subject to different dilutions depending on the amount of daily
analysis. The two animals produced antibodies specific for the elimination; therefore, small variations in the concentrations of
carbohydrate hapten, one of which showed a gradual increase the individual samples are likely. The three false-negative results
of the antibody’s affinity/concentration over the immunization may be explained by the fact that these individuals were infected
period.
with C. difficile strains that do not express PS-II. For the strains
High-throughput carbohydrate microarray analysis served as prevalent in European hospitals (Zaiss et al., 2010), the expres-
a fast method to detect antibodies in murine sera, hybridoma sion of PS-II is only confirmed for ribotype 027. Because ribotyp-
supernatant, and human excrement. Active ester conjugation ing is not performed routinely in European hospitals, the genetic
chemistry allowed for facile immobilization of the amine-termi- background of the pathogens responsible for the infections
nated synthetic hexasaccharide antigen to glass slides. In addi- analyzed in this study remains elusive. The low binding signal
tion to hexasaccharide 2, seven control carbohydrates were recorded for three samples of patients without diagnosed
printed onto the microarray slides, and the entire array was C. difficile infection can be accounted for by latent or previous
stable for more than 1 year. Carbohydrate microarray analysis infections with bacteria of the clostridium type carrying PS-II.
gave a detailed picture of the presence of antibodies, antibody
affinity and concentration, as well as cross-reactivity.
In conclusion we show here that PS-II is an antigenic determi-
nant upon infections of humans with C. difficile. Using a synthetic
Using the microarrays, we detected specific anti-hexasac- fragment of PS-II as hapten, we further demonstrate immunoge-
charide hapten 2 IgA antibodies in the stool supernatants of nicity in mice and provide monoclonal antibodies specifically
Chemistry & Biology 18, 580–588, May 27, 2011 ª2011 Elsevier Ltd All rights reserved 585

Chemistry & Biology
A Possible Vaccine Candidate for C. difficile
Figure 5. IgG in the Serum of Immunized Mice and Three Murine Monoclonal Antibodies Bind Hexasaccharide 2 on Glycan Microarrays
(A) Printing scheme of the glycan microarrays.
(B) IgG titers of immunized animals were analyzed before immunization (day 0), after the immunization (day 16), and after the first boost (day 30). Both mice
produced IgG antibodies to hexasaccharide 2. The microarray printing pattern is shown in the gray box.
(C) Monoclonal antibodies 2805.7, 2805.21, and 2805.25 raised against the neoglycoconjugate.
interacting with the glycan hapten as determined by microarray pathogenesis of C. difficile-associated diseases (CDADs).
analysis.
Thus, both the natural polysaccharide and the synthetic
substructure are further studied as potential carbohydrate
conjugate vaccine candidates against C. difficile.
SIGNIFICANCE
EXPERIMENTAL PROCEDURES
The synthesis of a hexasaccharide fragment of a C. difficile
cell surface polysaccharide gave access to chemically
defined and structurally homogeneous material equipped
with a primary amine handle. This handle allowed for conju-
gation of the synthetic oligosaccharide to the immunogenic
carrier protein CRM197 and to glass surfaces to produce
microarrays. The neoglycoconjugate was immunogenic in
mice and produced murine monoclonal antibodies that
specifically interact with the glycan hapten. The antibody-
binding specificities were determined by microarray anal-
ysis. Furthermore, microarrays were used to detect IgA anti-
bodies in the stool supernatant of infected hospital patients.
The presence of anti-hexasaccharide 2 IgA antibodies in
infected patients suggests a pivotal role of the PS-II in the
Chemical Synthesis
Detailed experimental procedures and characterization data for new
compounds are available online; see Supplemental Experimental Procedures.
Conjugation
Diethyl squarate (7.3 ml, 51 mmol) was added to a solution of hexasaccharide
C1 (2 mg, 1.7 mmol) in EtOH (0.2 ml) and phosphate buffer (0.2 ml, 50 mM
[pH 7.2]) and stirred for 18 hr at room temperature. Most ethanol was removed
by a stream of N2. The mixture was purified using a HPLC Superdex size exclu-
sion column (95:5 H2O/EtOH) to afford a colorless solid. A solution of the squa-
rate adduct (0.7 mg, 546 nmol) and the diphtheria toxoid CRM197 (Calbiochem;
0.7 mg, 11.1 nmol) in NaHCO3 buffer solution (0.4 ml, 0.1 M [pH 9]) was shaken
for 48 hr at room temperature. The resulting mixture was purified by ultrafiltra-
tion (30 K, Amicon; Millipore) with PBS. The protein concentration was deter-
mined by Bradford analysis (Bio-Rad).
586 Chemistry & Biology 18, 580–588, May 27, 2011 ª2011 Elsevier Ltd All rights reserved

Chemistry & Biology
A Possible Vaccine Candidate for C. difficile
Figure 6. IgA Analysis of Stool Supernatant of Hospitalized Persons
The presence of IgA antibodies specific for hexasaccharide 2 was analyzed in stool supernatants of ten hospital patients.
(A) Glycan microarray experiment. Stool supernatant of each patient was analyzed in three adjacent microarray wells.
(B) Quantification of the fluorescence signals detected in the glycan microarray experiment. High titers of anti-hexasaccharide 2 IgA antibodies were detected in
patients 2095 (positive), 2122 (borderline), and 2131 (positive). Low-intensity signals were also detected in patients 2093, 2118, and 2121 (all diagnosed negative).
SDS-PAGE
were incubated in a humid chamber to complete reaction for 24 hr and stored
in a desiccator until usage.
Hexasaccharide 2-CRM197 glycoconjugate and unconjugated CRM197 were
dissolved in Laemmli buffer (0.125 M Tris, 20% [v/v] glycerol, 4% [w/v] SDS,
5% [v/v] b-mercaptoethanol, bromophenol [pH 6.8]) and boiled for 5 min.
Samples were run in 12.5% polyacrylamide gel and stained with 0.025%
(w/v) Coomassie brilliant blue R-250 in an aqueous solution containing 40%
(v/v) methanol and 7% (v/v) acetic acid.
Microarray Binding Assays
A FlexWell 64 (Grace Bio-Labs, Bend, OR, USA) grid was applied to the slides.
The resulting 64 wells were used for 64 individual experiments. The slide was
blocked with 2.5% (w/v) BSA and 0.05% (v/v) Tween 20 in PBS for 1 hr at room
temperature. Blocked slides were washed with PBS and incubated with 5%
(v/v) serum in PBS or hybridoma culture supernatant for 1 hr at room temper-
ature. Slides were washed with PBS and incubated with 10 mg/ml Alexa Fluor
594 goat anti-mouse IgG and Alexa Flour 594 goat anti-mouse IgM (both
Invitrogen) secondary antibody solutions in PBS with 1% (w/v) BSA. Slides
were washed with PBS and centrifuged to dryness. Slides were scanned using
a GenePix 4300A scanner (Bucher Biotec, Basel, Switzerland) and evaluated
using the GenePix Pro 7 software (Bucher Biotec).
MALDI-TOF MS
Conjugation was confirmed by MALDI-TOF MS using an Ultraflex-II TOF/TOF
instrument (Bruker Daltonics, Bremen, Germany) equipped with a 200 Hz
solid-state Smart beamꢀ laser. The mass spectrometer was operated in the
positive linear mode. MS spectra were acquired over an m/z range of 4,000–
80,000, and data were analyzed using FlexAnalysisꢁ software provided with
the instrument. The samples were lyophilized from 25 mM NH4HCO3
(pH 7.8). Sinapinic acid was used as the matrix, and samples were spotted
using the dried droplet technique.
Monoclonal Antibody Purification
Supernatant of the hybridoma clones was filtered through a 0.2 mm filter. The
supernatant was mixed 1:1 with binding buffer (0.1 M NaP, 0.15 M NaCl
[pH 7.4]) and loaded onto a Midi Protein G spin column (Proteus, Oxford,
UK). The spin column was washed twice with binding buffer. Subsequently,
the IgG was eluted with elution buffer (0.2 M glycan/HCl [pH 2.5]) and immedi-
ately neutralized with 1 M Tris/HCl (pH 9). The eluted antibody solution was
purified by ultrafiltration (100 K, Amicon; Millipore) with PBS containing
0.01% (w/v) sodium azide. Protein-stabilizing cocktail (Pierce, Rockford, IL,
USA) was added to the concentrated antibody solution, and the protein
concentration was determined by Bradford analysis (Bio-Rad).
Immunizations
Two female C57BL/6 mice were immunized s.c. with 15 mg hexasaccharide
2-CRM197 conjugate in complete Freund’s adjuvants where 15 mg refers to
the protein content. The mice were boosted twice with 15 mg hexasaccharide
2-CRM197 conjugate in incomplete Freund’s adjuvants in 2-week intervals.
After each injection, blood was collected, and serum titers (IgG) were analyzed
using microarrays. Prior to being sacrificed, mice received additional 10 mg
hexasaccharide 2-CRM197 in PBS i.p. on 3 consecutive days.
Preparation of Clostridium Microarrays
Analysis of Stool Supernatant
Eight oligosaccharides bearing an amine linker were immobilized on NHS-
activated slides. Besides hexasaccharide 2, mannose, glucose, galactose,
fucose, acetylglucosamine, lactose, and a b-galactoside337 were printed in
0.5 mM concentration onto the slides. Each spot was printed in quadruplicate
using a piezoelectric spotting device (S11; Scienion, Berlin, Germany). Slides
A FlexWell 64 (Grace Bio-Labs) grid was applied to the slides. The wells were
blocked with 2.5% (w/v) BSA and 0.05% (v/v) Tween 20 in PBS for 1 hr at room
temperature. Blocked slides were washed with PBS and incubated with 20 ml
stool supernatant of ten hospitalized persons (Charite´ , Berlin) for 1 hr at room
temperature. Slides were washed with PBS and incubated with 10 mg/ml goat
Chemistry & Biology 18, 580–588, May 27, 2011 ª2011 Elsevier Ltd All rights reserved 587

Chemistry & Biology
A Possible Vaccine Candidate for C. difficile
anti-human IgA FITC Conjugate (Invitrogen) secondary antibody solutions in
PBS with 1% (w/v) BSA. Slides were washed with PBS and centrifuged to
dryness. Slides were scanned using a GenePix 4300A scanner and evaluated
using the GenePix Pro 7 software.
Ko¨ hler, G., and Milstein, C. (1975). Continuous cultures of fused cells secreting
antibody of predefined specificity. Nature 256, 495–497.
Kong, F. (2007). Recent studies on reaction pathways and applications of
sugar orthoesters in synthesis of oligosaccharides. Carbohydr. Res. 342,
345–373.
SUPPLEMENTAL INFORMATION
Kuberan, B., and Linhardt, R.J. (2000). Carbohydrate based vaccines. Curr.
Org. Chem. 4, 653–677.
Supplemental Information includes Supplemental Experimental Procedures
and can be found withthis articleonline at doi:10.1016/j.chembiol.2011.03.009.
Kuijper, E.J., van Dissel, J.T., and Wilcox, M.H. (2007). Clostridium difficile:
changing epidemiology and new treatment options. Curr. Opin. Infect. Dis.
20, 376–383.
ACKNOWLEDGMENTS
Loo, V.G., Poirier, L., Miller, M.A., Oughton, M., Libman, M.D., Michaud, S.,
Bourgault, A.M., Nguyen, T., Fernette, C., Kelly, M., et al. (2005). A predomi-
nantly clonal multi-institutional outbreak of Clostridium difficile-associated
diarrhea with high morbidity and mortality. N. Engl. J. Med. 353, 2442–2449.
¨
¨
We thank the Max-Planck Society and the Korber Foundation (Korber Prize for
P.H.S.) for generous financial support. We are grateful to Dr. V. Mountain for
critical review of the manuscript.
Lucas, A.H., and Reason, D.C. (1999). Polysaccharide vaccines as probes of
Received: January 10, 2011
Revised: February 23, 2011
Accepted: March 7, 2011
Published: May 26, 2011
antibody repertoires in man. Immunol. Rev. 171, 89–104.
McDonald, L.C., Killgore, G.E., Thompson, A., Owens, R.C., Kazakova, S.V.,
Sambol, S.P., Johnson, S., and Gerding, D.N. (2005). An epidemic, toxin
gene-variant strain of Clostridium difficile. N. Engl. J. Med. 353, 2433–2441.
REFERENCES
Pe´ pin, J., Valiquette, L., Alary, M.-E., Villemure, P., Pelletier, A., Forget, K.,
´
Pepin, K., and Chouinard, D. (2004). Clostridium difficile-associated diarrhea
Ada, G., and Isaacs, D. (2003). Carbohydrate-protein conjugate vaccines. Clin.
in a region of Quebec from 1991 to 2003: a changing pattern of disease
Microbiol. Infect. 9, 79–85.
severity. CMAJ 171, 466–472.
˚
Akerlund, T., Persson, I., Unemo, M., Nore´ n, T., Svenungsson, B., Wullt, M.,
Polk, R.E., Oinonen, M., and Pakyz, A. (2006). Epidemic Clostridium difficile.
N. Engl. J. Med. 354, 1199–1203.
and Burman, L.G. (2008). Increased sporulation rate of epidemic Clostridium
difficile type 027/NAP1. J. Clin. Microbiol. 46, 1530–1533.
Ravida` , A., Liu, X., Kovacs, L., and Seeberger, P.H. (2006). Synthesis of
glycosyl phosphates from 1,2-orthoesters and application to in situ glycosyla-
tion reactions. Org. Lett. 8, 1815–1818.
Barocchi, M.A., Censini, S., and Rappuoli, R. (2007). Vaccines in the era of
genomics: the pneumococcal challenge. Vaccine 25, 2963–2973.
Be´ lot, F., and Jacquinet, J.C. (2000). Syntheses of chondroitin 4- and 6-sulfate
pentasaccharide derivatives having a methyl b-D-glucopyranosiduronic acid
at the reducing end. Carbohydr. Res. 326, 88–97.
Rawat, M., Gama, C.I., Matson, J.B., and Hsieh-Wilson, L.C. (2008).
Neuroactive chondroitin sulfate glycomimetics. J. Am. Chem. Soc. 130,
2959–2961.
Chen, C.T., Weng, S.S., Kao, J.Q., Lin, C.C., and Jan, M.D. (2005). Stripping off
water at ambient temperature: direct atom-efficient acetal formation between
aldehydes and diols catalyzed by water-tolerant and recoverable vanadyl
triflate. Org. Lett. 7, 3343–3346.
Roy, R. (2008). Carbohydrate-Based Vaccines (New York: Oxford University
Press).
Stallforth, P., Lepenies, B., Adibekian, A., and Seeberger, P.H. (2009).
Carbohydrates: a frontier in medicinal chemistry. J. Med. Chem. 52, 5561–
5577.
Clements, A.C.A., Soares Magalha˜ es, R.J., Tatem, A.J., Paterson, D.L., and
Riley, T.V. (2010). Clostridium difficile PCR ribotype 027: assessing the risks
of further worldwide spread. Lancet Infect. Dis. 10, 395–404.
Thomas, C., Stevenson, M., and Riley, T.V. (2003). Antibiotics and hospital-
acquired Clostridium difficile-associated diarrhoea:
J. Antimicrob. Chemother. 51, 1339–1350.
a systematic review.
Danieli, E., Lay, L., Proietti, D., Berti, F., Constantino, P., and Adamo, R. (2011).
First synthesis of C. difficile PS-II cell wall polysaccharide repeating unit. Org.
Lett. 13, 378–381.
¨
Tietze, L.F., Schroter, C., Gabius, S., Brinck, U., Goerlach-Graw, A., and
Gabius, H.J. (1991). Conjugation of p-aminophenyl glycosides with squaric
acid diester to a carrier protein and the use of the neoglycoprotein in the histo-
chemical detection of lectins. Bioconjug. Chem. 2, 148–153.
Delcros, J.G., Tomasi, S., Carrington, S., Martin, B., Renault, J., Blagbrough,
I.S., and Uriac, P. (2002). Effect of spermine conjugation on the cytotoxicity
and cellular transport of acridine. J. Med. Chem. 45, 5098–5111.
van Steijn, A.M.P., Kamerling, J.P., and Vliegenthart, J.F.G. (1992). Synthesis
of trisaccharide methyl glycosides related to fragments of the capsular poly-
saccharide of Streptococcus pneumoniae type 18C. Carbohydr. Res. 225,
229–245.
Depre´ , D., Du¨ ffels, A., Green, L.G., Lenz, R., Ley, S.V., and Wong, C.-H. (1999).
Synthesis of glycans from the glycodelins: two undeca-, two deca-, three
nona-, an octa- and a heptasaccharide. Chem. Eur. J. 5, 3326–3340.
Ganeshapillai, J., Vinogradov, E., Rousseau, J., Weese, J.S., and Monteiro,
M.A. (2008). Clostridium difficile cell-surface polysaccharides composed of
pentaglycosyl and hexaglycosyl phosphate repeating units. Carbohydr. Res.
343, 703–710.
´
Verez-Bencomo, V., Fernandez-Santana, V., Hardy, E., Toledo, M.E.,
Rodrı´guez, M.C., Heynngnezz, L., Rodriguez, A., Baly, A., Herrera, L.,
Izquierdo, M., et al. (2004). A synthetic conjugate polysaccharide vaccine
against Haemophilus influenzae type b. Science 305, 522–525.
Hecht, M.L., Stallforth, P., Varo´ n Silva, D., Adibekian, A., and Seeberger, P.H.
(2009). Recent advances in carbohydrate-based vaccines. Curr. Opin. Chem.
Biol. 13, 354–359.
Yu, B., and Tao, H. (2001). Glycosyl trifluoroacetimidates. Part 1: preparation
and application as new glycosyl donors. Tetrahedron Lett. 42, 2405–2407.
Hookman, P., and Barkin, J.S. (2009). Clostridium difficile associated infection,
diarrhea and colitis. World J. Gastroenterol. 15, 1554–1580.
Zaiss, N.H., Witte, W., and Nu¨ bel, U. (2010). Fluoroquinolone resistance and
Clostridium difficile, Germany. Emerg. Infect. Dis. 16, 675–677.
588 Chemistry & Biology 18, 580–588, May 27, 2011 ª2011 Elsevier Ltd All rights reserved