Articles in PresS. Am J Physiol Gastrointest Liver Physiol (December 31, 2014). doi:10.1152/ajpgi.00091.2014
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37
Human Clostridium difficile infection: Altered mucus production and composition
Melinda A. Engevik 1, Mary Beth Yacyshyn3, Kristen A. Engevik1 Jiang Wang4, Benjamin Darien5, Daniel J. Hassett2 Bruce R. Yacyshyn3,6, Roger T. Worrell 1,6 1
Department of Molecular and Cellular Physiology Department of Molecular Genetics, Biochemistry and Microbiology 3 Department of Medicine Division of Digestive Diseases 4 Department of Pathology and Lab Medicine University of Cincinnati College of Medicine, Cincinnati, OH 45267 5 University Wisconsin-Madison Animal Health and Biomedical Sciences, Madison, WI 53706 6 Digestive Health Center of Cincinnati Children’s Hospital, Cincinnati, OH 45229 2
*Corresponding Author: Roger T. Worrell, Ph.D. Department of Molecular and Cellular Physiology University of Cincinnati College of Medicine Cincinnati, OH 45267 Phone: 513-558-6489 Email:
[email protected]
1
Copyright © 2014 by the American Physiological Society.
38
ABSTRACT
39
The majority of antibiotic-induced diarrhea is caused by Clostridium difficile (C. difficile).
40
Hospitalizations for Clostridium difficile infection (CDI) have tripled in the last decade,
41
emphasizing the need to better understand how the organism colonizes the intestine and maintain
42
infection. The mucus provides an interface for bacterial-host interactions and changes in
43
intestinal mucus have been linked host health. To assess mucus production and composition in
44
healthy and CDI patients, the main mucins MUC1 and MUC2 and mucus oligosaccharides were
45
examined. In comparison to healthy subjects, CDI patients demonstrated decreased MUC2 with
46
no changes in surface MUC1. Although MUC1 did not change at the level of the epithelia,
47
MUC1 was the primary constituent of secreted mucus in CDI patients. CDI mucus also exhibited
48
decreased N-Acetylgalactosamine (GalNAc), increased N-Acetylglucosamine(GlcNAc) and
49
increased terminal galactose residues. Increased galactose in CDI specimens are of particular
50
interest since terminal galactose sugars are known as C. difficile toxin A receptor in animals. In
51
vitro, C. difficile is capable of metabolizing fucose, mannose, galactose, GlcNAc, and GalNAc
52
for growth under healthy stool conditions (low [Na+], pH 6.0). Injection of C. difficile into
53
human intestinal organoids (HIOs) demonstrated that C. difficile alone is sufficient to reduce
54
MUC2 production, but is not capable of altering host mucus oligosaccharide composition. We
55
also demonstrate that C. difficile binds preferentially to mucus extracted from CDI patients
56
compared with healthy subjects. Our results provide insight into a mechanism of C. difficile
57
colonization and may provide novel target(s) for the development of alternative therapeutic
58
agents.
59 60
2
61
INTRODUCTION
62
Clostridium difficile infection (CDI) represents an increasing public health problem as it
63
is a primary cause of antibiotic-induced diarrhea and colitis. Since the cause and current
64
treatment of C. difficile is antibiotics, CDI will likely persist as a major health concern. Multiple
65
studies have focused on the effects of C. difficile toxin production on the intestinal epithelium,
66
however the human C. difficile mucus interaction and the process of colonization remains
67
unknown. GI mucus consists of a firmly attached inner mucus layer, devoid of bacteria under
68
normal conditions, and a loose outer mucus layer, which is colonized by bacteria (54, 55). The
69
intestinal mucus is composed primarily of MUC2 and secreted MUC2 forms a gel-like structure
70
which provides a physical barrier to protect the intestinal epithelium from the gut microbiota (6,
71
8, 53-55, 59, 107). The colon harbors the densest bacterial population and contains the thickest
72
intestinal mucus as a result of MUC2 secretion (96). Secreted and adherent mucus are
73
continuously renewed and can be rapidly secreted by the host to limit host-bacterial interaction
74
(65, 71). In addition to host-modulated mucus changes, several bacterial species are also capable
75
of altering host mucus (27, 31, 63, 65, 70, 71, 96). Secretion of MUC2 mucus has been proposed
76
to be one mechanism used by commensal bacteria to improve barrier function and exclude
77
pathogens (67, 70, 77). If bacteria are able to penetrate the secreted MUC2 mucus layer they are
78
able to interact with the glycocalyx consisting primarily cell-surface mucins, such as MUC1
79
(96).
80
In order to reach the epithelium, GI pathogens must develop specific virulence strategies
81
to address the presence of mucus (71). The mucus barrier has been shown to provide partial
82
protection against several enteric bacterial pathogens, including Yersinia enterocolitica, Shigella
83
flexneri, Citrobacter rodentium, and Salmonella enterica Serovar Typhimurium (8, 69, 76, 107).
3
84
Although multiple GI pathogens have been shown to evade or alter the host mucus barrier (8, 69,
85
76, 107), no studies have examined how C. difficile infection affects host mucus production or
86
composition. It has been well documented that antibiotic disruption of the gut microbiota
87
provides an open niche which favors C. difficile colonization and toxin production (3, 5, 9-12,
88
14). C. difficile has been shown to bind to mucus in cell lines and animal models (16, 32, 45, 56,
89
60, 98, 103, 104), but data on C. difficile mucus binding and colonization in humans is scarce.
90
Adherence of GI pathogens to the intestinal mucus has been hypothesized to allow for optimal
91
delivery of toxins to the host. C. difficile toxin A has been shown to bind to α-Gal-(1,3)-β-Gal-
92
(1,4)-β-GlcNAc in animal models (19, 22, 28, 40, 41, 47, 60, 82, 83, 98), but the receptors in
93
humans have not been identified. Furthermore it is unknown whether C. difficile is able to
94
manipulate the mucus oligosaccharide composition thereby exposing or stimulating production
95
of binding and toxin receptors. A better understanding of C. difficile pathogenesis, especially
96
during the colonization phase, is critical for developing new therapeutics for treatment of
97
prevention of CDI.
98
In this study, we demonstrate that patients with CDI secrete acidic mucus that consists
99
primarily of MUC1. Patients with CDI have decreased MUC2 expression indicating a mucus
100
barrier defect. Injection of human intestinal organoids (HIOs) with C. difficile demonstrates that
101
C. difficile alone is sufficient to decrease MUC2 production. In addition, we will show that
102
patients with CDI exhibit an altered mucus oligosaccharide composition and in vitro C. difficile
103
is able to use free oligosaccharides in media for proliferation. C. difficile injection into HIOs
104
reveals that C. difficile alone is not sufficient to alter the mucus oligosaccharide composition.
105
However HIOs injected with CDI stool supernatant is sufficient to elicit the changes observed in
4
106
biopsy specimens and surgical resections from patients with CDI. Finally, we demonstrate that
107
C. difficile binds better to mucus from CDI patients compared with control patient mucus.
108 109
METHODS
110
Patient information: Colonic biopsy specimens or surgical sections were obtained from healthy
111
volunteers or patients with recurrent CDI at the University of Cincinnati Medical Center
112
Hospital, Cincinnati, OH. All subjects provided informed consent approved by the University of
113
Cincinnati IRB. Patients with recurrent CDI were defined as >1 C. difficile positive laboratory
114
test (as determined by ELISA or LAMP test) and completion of at least one round of antibiotic
115
treatment. Biopsy specimens were collected from five healthy volunteers, fixed in neutral
116
buffered formalin and paraffin-embedded. Healthy subjects were defined as presenting without
117
previous or current GI symptoms, history of chronic intestinal disease or cancer. Healthy subject
118
demographics are as follows: average age 52, age range 45-63, 3 females, 2 males. Formalin-
119
fixed paraffin sections were obtained from 5 de-identified patients with current CDI diagnosis
120
(C. difficile-positive toxin test). Demographics for the CDI patients are as follows: average
121
patient age: 44, age range: 28-65, 2 females, 3 males. Selected patients and volunteers did not
122
have history of cancer, diverticulosis, colostomy, Inflammatory Bowel Disease (IBD), small
123
bowel obstruction.
124
Stool mucus was prepared from feces collected from healthy and recurrent CDI patients.
125
For mucus extraction, stool was collected from 7 healthy subjects (average age: 41, age range:
126
28-64, 5 female, 3 male) and 7 recurrent CDI patients (average age: 47, age range: 34-71, 3
127
female, 4 male). Total stool DNA was prepared from feces collected from 9 healthy (average
128
age: 43, age range: 28-57, 5 females, 4 males) and 9 recurrent CDI patients (average age: 51, age
5
129
range 32-64, 6 females, 3 males). The same criteria for patient inclusion for biopsy specimen
130
collection were used for stool samples.
131 132
Histology. Transverse colon biopsy specimens, which had been previously formalin fixed and
133
embedded in paraffin, were obtained for CDI patients. For consistency, transverse colon biopsy
134
specimens from healthy subjects were collected, fixed overnight at 4°C in neutral-buffered
135
formalin and embedded in paraffin. Serial 6–7 m thick sections were applied to glass slides.
136
Intestine architecture was examined by H&E stain. Mucus composition was examined by
137
periodic acid-Schiff (PAS)-Alcian Blue (AB) stain. Briefly, deparaffinized slides were exposed
138
to 3% acetic acid, followed by staining with Alcian blue, 1% Periodic Acid, Shiff Reagent and
139
sodium metabisulfite (Fisher Scientific, Walthman, MA). Expression of mucus oligosaccharides
140
was examined using FITC-conjugated lectins: Ulex europaeus agglutinin 1 (UEA-1, fucose),
141
Concanavalin A (ConA, mannose), Dolichos biflorus agglutinin (DBA, N-Acetylgalactosamine),
142
Peanut Agglutinin (PNA, galactose), and Wheat germ agglutinin (WGA, N-acteylglucosamine,
143
Vector Laboratories, Burlingame, CA) were used as previously described (29, 49, 68). Briefly,
144
sections were deparaffinized, blocked with PBS containing 10% BSA, and stained with FITC-
145
labeled lectin (10 g/ml) for 1 h at room temperature. Sections were then washed three times in
146
PBS, counterstained with Hoechst (Fisher Scientific), and analyzed by confocal laser scanning
147
microscopy (Zeiss LSM Confocal 710, Carl Zeiss, Germany).
148
Expression of MUC1 examined with an anti-human MUC1 antibody (dilution 1:100, RB-
149
9222, Thermo Fisher Scientific), MUC2 examined with an anti-human MUC2 antibody (dilution
150
1:100, MS-1037, Thermo Fisher Scientific) and C. difficile binding was examined with rabbit
151
anti-C. difficile cell surface protein antibody (dilution 1:100, ab93728, ABCAM, Cambridge,
6
152
MA). Briefly, sections were stripped of paraffin and incubated for 40 min at 97°C with Tris–
153
EDTA–SDS buffer as previously described (99). Sections were then blocked with PBS
154
containing 10% serum, and stained with primary antibody overnight at 4°C. Sections were then
155
washed three times in PBS, incubated with goat-anti-rabbit IgG Alexa Fluor® 488 secondary
156
antibody (1:100 dilution) (Life Technologies, Grand Island, NY) for 1 hr at room temperature
157
and counterstained with Hoechst (Fisher Scientific). For MUC1 and MUC2 stains, MUC1 was
158
stained with goat-anti-rabbit IgG Alexa Fluor® 488(1:100 dilution) and MUC2 was stained with
159
donkey anti-mouse IgG Alexa Fluor® 633(1:100 dilution) and counterstained with Hoechst
160
(Fisher Scientific). Sections were analyzed by confocal laser scanning microscopy (Zeiss LSM
161
Confocal 710). Digital images of slides were evaluated by tabulating mean pixel intensity of the
162
respective color channel on each image using Image J software (NIH) and reported as relative
163
fluorescence. Five regions of interests per image, four images per slide, and n=5 healthy and CDI
164
patients were used for semi-quantitation of stain intensity.
165 166
Stool mucus extraction. Crude mucus was extracted from human feces as previously described
167
(57, 79, 88). Briefly, stool was diluted in 4 volumes of ice-cold PBS (PBS; 10 mM phosphate,
168
pH 7.2) containing 0.5 µg/µl sodium azide (prevents bacterial growth), 1 mM
169
phenylmethylsulfonyl fluoride, iodoacetamide and 10 mM EDTA to inhibit proteases. The
170
suspension was shaken for 1 h at 43°C and centrifuged for 20 min at 4°C at 14,000 x g to pellet
171
stool solids. From the clear supernatant, the mucus was precipitated twice with ice cold 60%
172
ethanol and resuspended in ultrapure water. Mucus was then lyophilized and reconstituted based
173
on weight. Mucus was stored at -80°C until used. Crude mucus protein concentration was
174
determined by Bio-Rad protein assay performed according to the instructions of the
7
175
manufacturer (Bio-Rad Laboratories, Richmond, CA). The original non-lyophilized crude mucus
176
contained different amounts of proteins, with healthy mucus containing 0.4 µg/µl and CDI
177
mucus 2.2 µg/µl. Pooled lyophilized crude mucin preparation was reconstituted to 7 µg/µl
178
protein. C. difficile mucus binding was examined in two ways: microtiter plate and slide. For the
179
microtiter plate assay, 100 µl mucus was immobilized in polystyrene microtiter plate wells
180
(Maxisorp, Nunc) by overnight incubation at 4°C as previously described (79), which covers the
181
well with sufficient mucus. The wells were washed twice with PBS to remove excess mucus.
182
The microtiter plate was then placed in an anaerobic hood for 2 hr to create an anaerobic
183
environment. Once anaerobic, 100 µl of C. difficile ATTC-1870 culture (106 CFU) was added to
184
each well and incubated for 3 hr at 37°C within the Coy anaerobic chamber. After incubation, the
185
wells were washed twice with PBS to remove unattached bacteria. Mucus was then scrapped off
186
the microtiter plate and grown on C. difficile selective agar plates (Fisher Scientific). For the
187
slide binding assay, 50 µl mucus (385 µg protein) was added to glass slides and spread by an
188
inoculating loop. Slides were allowed to air-dry and then heat fixed using a Bunsen burner.
189
Slides were blocked with PBS containing 10% serum. An ImmEdge pen (Fisher Scientific) was
190
used to select areas for C. difficile binding. Slides were placed in an anaerobic chamber for 2 hr
191
and then 100 µl of C. difficile (106 CFU) was added to each well and incubated for 3 hr at 37°C
192
within the anaerobic chamber. Slides were then washed 3x in PBS to remove unattached bacteria
193
and slides were incubated with an anti-C. difficile antibody (dilution 1:100, ab93728, ABCAM)
194
for 1 hr at room temperature. Slides were then washed three times in PBS, incubated with goat-
195
anti-rabbit IgG Alexa Fluor® 488 secondary antibody (1:100 dilution) (Life Technologies),
196
cover slipped, and analyzed by confocal laser scanning microscopy (Zeiss LSM Confocal 710)
197
and Image J software (NIH).
8
198 199
Human Intestinal organoids (HIOs) and microinjection. Human intestinal organoids (HIOs),
200
hereafter referred to as HIOs, were generated by the Cincinnati Children’s Hospital Medical
201
Center (CCHMC) Pluripotent Stem Cell Facility. HIOs were generated by directed
202
differentiation of human pluripotent stem cells (hPSCs) by the facility and obtained in matrigel.
203
hPSCs were differentiated by culturing with ActivinA, fibroblast growth factor 4 (FGF4) and
204
Wnt3a. Addition of epidermal growth factor (EGF), R-spondin and Noggin to HIOs in matrigel
205
generated three-dimensional growth (105). HIOs resemble the proximal colon and contain
206
differentiated cell types (105). HIOs were injected with broth (control), C. difficile culture
207
(grown in Tryptone yeast TY broth (33)) and healthy or CDI patient stool supernatant using a
208
Nanoject microinjector (Drummon Scientific Company, Broomall, PA). They were then
209
incubated overnight as previously described (29). To generate stool supernatant, in an anaerobic
210
hood 0.5 g of healthy or CDI stool was added to 4.5 ml Tryptic Soy Broth (TSB) (Fisher
211
Scientific) in an anaerobic hood. Samples were vortexed and centrifuged at 150 x g for 10 min to
212
pellet solid materials and liquid supernatant containing bacteria were used for injection. After
213
overnight incubation, injected HIOs were processed for immuno-staining or RNA. RNA was
214
extracting from HIOs using TRIzol Reagent according to the manufacturer’s instructions
215
(Invitrogen Life Technologies, Grand Island, NY). To process HIOs for immune-staining, HIOs
216
were fixed in Carnoy’s fixative (60% ethanol, 30% chloroform and 10% glacial acetic acid) for
217
30 min at room temperature, washed with PBS, transferred to sucrose (30% in PBS) and
218
incubated overnight at 4°C. HIOs were placed in OCT (Optimal Cutting Temperature) medium,
219
frozen at -80°C and cut to 7 µm sections on a cryostat. Slides were stained with anti-human
220
MUC1 antibody (dilution 1:100, RB-9222, Thermo Fisher Scientific), or anti-human MUC2
9
221
antibody (dilution 1:100, MS-1037, Thermo Fisher Scientific) overnight at 4°C. Sections were
222
countered stained with Alexa Fluor® 488 or 633 secondary antibodies for 1 hr at room
223
temperature and also counterstained with Hoechst (0.1 µg/ml) (Fisher Scientific) for 10 min at
224
room temperature. Expression of mucus oligosaccharides in HIOs was examined with 10 µg/ml
225
FITC-conjugated lectins: Concanavalin A (ConA, mannose), Dolichos biflorus agglutinin (DBA,
226
N-Acetylgalactosamine), Peanut Agglutinin (PNA, galactose), and Wheat germ agglutinin
227
(WGA, N-acteylglucosamine) (Vector Laboratories, Burlingame, CA). Sections were stained for
228
1 hr at room temperature followed by Hoechst staining (Fisher Scientific) for 10 min at room
229
temperature. All slides were analyzed by confocal laser scanning microscopy (Zeiss LSM
230
Confocal 710, Carl Zeiss) and the relative fluorescence intensity was quantified using Image J
231
software (NIH) and reported as relative fluorescence. Five regions of interests per image, four
232
images per slide, n=3, were used for semi-quantitation of stain intensity.
233 234
Quantitative real time-PCR amplification of 16S sequences. qRT-PCR standard curves were
235
generated from pure cultures of Micrococcus luteus, Staphylococcus aureus, Escherichia coli,
236
Burkholderia cepacia, Bacteroidetes thetaiotaomicron ATCC 29741 (Thermo Fisher Scientific,
237
Waltham, MA) and Rhizobium leguminosarum (Carolina Biological Supply Company,
238
Burlington, NC) as previously described (29). Total DNA was isolated from stool using a
239
QIAamp DNA Stool kit (Qiagen, Valencia, CA, USA) with the addition of a 95°C incubation
240
and lysozyme (10 mg/ml, 37°C for 30 min) incubation as previously described (20, 29, 36, 75,
241
92, 93). Qauntitative PCR (qPCR) was performed using a Step One Real Time PCR machine
242
(Applied Biosystems (ABI) Life Technologies) with SYBR Green PCR master mix (ABI) and
243
bacteria-specific primers (Table 1). Standard curves were used to correlate Cycle of threshold
10
244
values (CT) with calculated bacteria number as previously described to access the abundance of
245
specific intestinal bacterial phyla (4, 29, 78, 92). Percent bacterial phyla were determined with
246
the calculated CFU values for each bacterial phylum represented as a percentage of the total
247
bacteria.
248 249
qRT-PCR of mRNA. To examine MUC1 and MUC2 message level, total RNA was extracted
250
from HIOs with TRIzol Reagent (Invitrogen) according to the manufacturer’s instructions and as
251
previously described. qRT-PCR was performed using SYBR Green PCR master mix (ABI) on a
252
Step One Real Time PCR Machine (ABI) with the following gene specific qRT-PCR primers
253
were derived from previous literature: Muc1 Forward 5'- CCAGACCCCTGCACTCTGAT-3',
254
Muc1
255
TGCCCACCTCCTCAAAGAC-3'; Muc2 Reverse 5'- TAGTTTCCGTTGGAACAGTGAA-3'
256
(38); human GAPDH Forward 5'- TGCACCACCAACTGCTTAGC -3' and GAPDH Reverse 5'-
257
GGCATGGACTGTGGTCATGAG -3' (86). Data were reported as the delta delta CT using
258
GAPDH as the standard. Differences in mRNA expression were determined by qRT-PCR and
259
expressed as the CT relative fold difference.
Reverse
5'-
CGCTTGACAAAGGGCATGA-3';
Muc2
Forward5'-
260 261
Bacterial strains and culture conditions. C. difficile ATTC BAA-1870 was grown in media
262
with 8 mM Na+, pH 6.0 mimicking healthy conditions 75 mM Na+, pH 7.0 mimicking CDI
263
conditions (see manuscript #1, Figure 2). To determine the ability of mucus oligosaccharides to
264
promote C. difficile growth, fucose, N-acetylglucosamine (GlcNAc), N-acetylgalactosamine
265
(GalNAc) (Thermo Fisher Scientific), mannose or galactose (Sigma-Aldrich) were added to
266
TYG media at a final concentration of 1 mM and their presence was then analyzed for an 11
267
influence on bacterial growth. Bacteria were grown under anaerobic conditions at 37°C to early
268
stationary phase in normal TYG media (12 hrs, O.D.
269
containing varying [Na+]. To assess restoration of growth, C. difficile was grown first in high
270
[Na+], pH 7.0 overnight and then used to inoculate low [Na+] pH 6.0 cultures with or without
271
oligosaccharides. Growth from all experiments was measured as the optical density (O.D. 560 nm)
272
with an Amersham Biosciences Ultospec 3100 Spectrophotometer (GE Healthcare Life Sciences,
273
Pittsburgh, PA) after 24 hrs. C. difficile titers were determined by bacterial cell counts using a
274
Petroff-Hauser chamber (Hausser Scientific; Horsham, PA) and also by colony forming units
275
(CFU) (17, 18).
560 nm
~1) and used to inoculate media
276 277
Statistics. Data are presented as mean ± SEM. Comparisons between groups were made with
278
either One or Two Way Analysis of Variance (ANOVA), and the Holm-Sidak post-hoc test used
279
to determine significance between pair wise comparisons using SigmaPlot (Systat Software, Inc.,
280
San Jose, CA).
281
experiments performed.
A P < 0.05 value was considered significant while n is the number of
282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 12
298 299 300
RESULTS
301
Studies have demonstrated that some bacteria are capable of adhering to intestinal mucus and/or
302
epithelium (84) and changing the rate of mucus secretion (26, 62, 77). Although it has been
303
demonstrated that several GI pathogens alter mucus production, little data exists on C. difficile-
304
mucus alteration in the patient setting. H&E stains of healthy and CDI colon reveal the presence
305
of mucus and the formation of pseudomembranes in patients with CDI compared to healthy (Fig.
306
1A black arrows). Periodic acid-Schiff (PAS)-Alcian Blue (AB) stains of CDI specimens depict
307
the secretion of acidic mucins (dark blue) (Fig. 1A). To determine the type of mucin present in
308
CDI mucus, intestinal mucins MUC1 (cell-surface) and MUC2 (secreted) were examined in
309
human specimens by immunofluorescence (Fig. 1B and C). Although MUC1 secretion was
310
observed, no changes were observed in MUC1 levels at the epithelium. Interestingly, MUC2
311
levels were significantly decreased in CDI patients compared to healthy subjects (P < 0.001). To
312
assess whether MUC1 was the predominant mucus in CDI stool, mucus was extracted from
313
healthy and CDI patient stool and MUC1 and MUC2 was analyzed by immunofluorescence.
314
Healthy subject stool was composed primarily of secreted MUC2. In contrast, CDI stool was
315
found to be composed almost entirely of MUC1 (Fig. 1D and E). This data indicates that CDI
316
patients have decreased secreted MUC2 and the secretion of the normally adherent MUC1. To
317
determine if C. difficile is capable of altering host mucus production, C. difficile, CDI and
318
healthy stool supernatants were injected into HIOs (Fig. 2A-C). Injection of C. difficile and CDI
319
stool supernatant both resulted in decreased MUC2 expression at the level of mRNA (Fig 2A)
320
and protein (Fig 2B and C). This indicates that C. difficile alone is sufficient to inhibit MUC2
321
expression.
Mucus provides both an interface and barrier between luminal bacteria and the host (85).
13
322
In addition to serving as a barrier, the mucus serves as an interface between the bacteria
323
and the host: the mucus provides binding sites for both commensal and pathogenic bacteria (64,
324
87) and bacterial energy/food sources (2, 6, 26, 107). Gut mucus oligosaccharides N-
325
acetylglucosamine (GlcNAc), galactose (Gal), N-acetylgalactosamine (GalNAc), fucose (Fuc),
326
N-acetylneuraminic acid (NeuNAc), and mannose (53-55) are attached to mucin glycoproteins
327
and can be used by bacteria to maintain their relative niches. To address whether or not CDI
328
patients exhibit an altered stool microbiota composition compared with healthy subjects, stool
329
DNA was examined. Bacterial analysis of stool DNA revealed that patients with CDI have an
330
altered gut microbiota with increased Bacteroidetes (P
