Microbiology (2010), 156, 3079–3084
DOI 10.1099/mic.0.039867-0
Campylobacter jejuni 81-176 forms distinct
microcolonies on in vitro-infected human small
intestinal tissue prior to biofilm formation
Graham Haddock,1 Margaret Mullin,2 Amanda MacCallum,3 Aileen Sherry,3
Laurence Tetley,2 Eleanor Watson,4 Mark Dagleish,4 David G. E. Smith3,4
and Paul Everest3
Correspondence
1
Paul Everest
2
Department of Paediatric Surgery, Royal Hospital for Sick Children, Glasgow, UK
Institute of Biological and Life Sciences, University of Glasgow, Glasgow, UK
[email protected]
3
Institute of Comparative Medicine, University of Glasgow Veterinary School, Glasgow, UK
4
Moredun Research Institute, Penicuik, Midlothian, UK
Received 18 March 2010
Revised
09 June 2010
Accepted 6 July 2010
Human small and large intestinal tissue was used to study the interaction of Campylobacter jejuni
with its target tissue. The strain used for the study was 81-176 (+pVir). Tissue was processed for
scanning and transmission electron microscopy, and by immunohistochemistry for light
microscopy. Organisms adhered to the apical surface of ileal tissues at all time points in large
numbers, in areas where mucus was present and in distinct groups. Microcolony formation was
evident at 1–2 h, with bacteria adhering to mucus on the tissue surface and to each other by
flagellar interaction. At later time points (3–4 h), biofilm formation on ileal tissue was evident.
Flagellar mutants did not form microcolonies or biofilms in tissue. Few organisms were observed
in colonic tissue, with organisms present but not as abundant as in the ileal tissue. This study
shows that C. jejuni 81-176 can form microcolonies and biofilms on human intestinal tissue and
that this may be an essential step in its ability to cause diarrhoea in man.
INTRODUCTION
Campylobacter jejuni is the commonest cause of bacterial
foodborne disease in the developed world and causes a large
amount of disease, particularly in the young, in developing
countries. Symptoms include bloody or watery diarrhoea,
abdominal pain and fever (Skirrow & Blaser, 2000). It is
evident that the organism must interact with host
gastrointestinal epithelium in some way that manifests as
the clinical presentation of diarrhoea (Konkel et al., 2001).
The interaction of C. jejuni with human tissue and how the
organism causes diarrhoea in an infected host are not well
understood. Experimental evidence using tissue culture
cells as models of gastrointestinal epithelia suggests that
bacteria interact with cells via their flagella. The flagellar
export system delivers effector molecules into the host cell,
which facilitates cell invasion (Konkel et al., 2001). CiaB
has been shown to be important for the clinical presentation of diarrhoeal disease in a pig model of infection
(Konkel et al., 2001). The organism may also bind to extracellular matrix molecules on the basolateral cell surface, which results in membrane ruffling and bacterial
internalization (Konkel et al., 2001; Krause-Gruszczynska
Abbreviation: SEM, scanning electron microscopy.
039867 G 2010 SGM
et al., 2007). Using C. jejuni-infected Caco-2 cells as a
model of host cell infection, it is clear that bacterial
adhesion to the apical cell surface is minimal, with most
adhesion observed at cellular junctions (Hu et al., 2008). It
is difficult to study the interaction of this organism with
human tissue because biopsies are not always obtained
from patients with C. jejuni enteritis due to the risk of
intestinal perforation, or if a mucosal biopsy is taken at
acute presentation the diagnosis of a specific Campylobacter aetiology may not be known. We have obtained
human intestinal tissue (both ileal and colonic) from children undergoing elective intestinal resection and infected it
in vitro with C. jejuni strain 81-176. This has allowed us to
study the interaction of C. jejuni with human gastrointestinal tissue by electron and light microscopy.
METHODS
Bacteria. Bacteria were grown in Mueller–Hinton broth and agar
(Oxoid) and incubated at 37 uC in a variable atmosphere incubator
(VAIN, Don Whitley) under microaerophilic conditions in an
atmosphere of 6 % hydrogen, 5 % carbon dioxide, 5 % oxygen and
84 % nitrogen. The strain of C. jejuni used in this study was 81-176
(+pVir). This strain has been shown to cause clinical disease in
human volunteers (Black et al., 1988). Flagella mutants in flaA and
flaB were also tested in both ileal and colonic tissue.
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Fri, 16 Jun 2017 12:02:41
Printed in Great Britain
3079
G. Haddock and others
Infection of human intestinal tissue. Ethical permission to obtain
and use human gastrointestinal tissue was sought and obtained from
the Royal Hospital for Sick Children, research ethics committee,
North Glasgow NHS Trust. Primary human small and large intestinal
tissue was obtained from healthy sections of intestine removed during
elective surgery. For this study at least 10 independent tissue samples
were used from different patients. The tissue samples were initially
transported to the laboratory in Dulbecco’s modified Eagle’s medium
(DMEM) with 10 % fetal calf serum (FCS) and penicillin/streptomycin (Gibco). On arrival in the laboratory, tissue samples were washed
in antibiotic-free DMEM with 10 % FCS, cut into explants (1–2 cm2)
and placed into the top well of a 12-well Transwell tissue culture plate
(Corning). Explants were inoculated with 50 ml bacterial suspension
(C. jejuni 81-176), containing ~16108 c.f.u., or left uninfected as
controls. Explants were incubated for 1–4 h at 37 uC in a 6 % CO2
humidified atmosphere to allow the bacteria to interact with the
tissue. Uninfected control tissue from each different tissue source was
used in the same experiments as infected tissue. Tissue was washed in
three rinses of PBS prior to fixation and processing for microscopy. Tissue morphology was checked after infection by normal
histological processing using haematoxylin and eosin staining. Only
morphologically normal tissue was used for the study. Five
independent experiments on ileal tissue were performed using flaA
and flaB mutants of C. jejuni 81-176.
Specimen processing for scanning electron microscopy (SEM).
Specimens were fixed for 1 h at room temperature in 2.5 %
glutaraldehyde in 0.1 M phosphate buffer. Following three 5 min
rinses with 0.1 M phosphate buffer, specimens were fixed for 1 h in
1 % osmium tetroxide. After three 10 min washes with distilled water,
specimens were dehydrated through an ascending series of acetone
solutions (30, 50, 70, 90 %) twice for 10 min each, ending in 100 %
acetone twice for 10 min and dried 100 % acetone twice for 10 min.
Specimens were then dried in a critical point dryer (Polaron E3000)
for 80 min and mounted on stubs using double-sided copper tape
and silver paint. A Polaron SC515 SEM coating system was used to
coat the specimens with gold–palladium (20 nm thickness) and they
were viewed on a JEOL 6400 scanning electron microscope.
Specimen processing for immunohistochemistry. Human ileal
sections of 4 mm thickness were mounted onto slides and allowed to
dry before placing in a 37 uC oven overnight. Slides were marked with
Fig. 1. (a, b). Low-power views of C. jejuni 81-176 on ileal tissue, 1–2 h after infection. Organisms form microcolonies where
mucus overlying intestinal villi is abundant. (c, d) Details of a microcolony. Bacteria adhere together in distinct clumps. Bacteria
adhere to mucus via flagella and adhere to each other with a distinct interaction between flagella. (e) Detail of different
microcolonies, showing distinct flagellar morphology.
3080
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Fri, 16 Jun 2017 12:02:41
Microbiology 156
C. jejuni infection, biofilms and human tissue
the antibody and date, dewaxed in xylene and rinsed in 100 % alcohol
then 95 % alcohol. Slides were then placed in 3 % hydrogen peroxide
in methanol for 20 min at room temperature. After washing in water,
antigen retrieval was performed by placing the slides in an autoclave
for 10 min in citrate buffer, pH 6, at 121 uC. Slides were left to cool
to 50 uC before the transfer of slides to a staining chamber. Slides
were washed in PBS/Tween-20 twice before applying 25 % normal
goat serum for 30 min at room temperature. Antibody against C.
jejuni was added to the slides overnight at 4 uC (100 ml, 1 : 600 mouse
monoclonal, BGN/2E10, 1 mg ml21, Abcam ab54125). A control
isotype IgM antibody was used at equivalent dilution (Sigma M5909,
0.2 mg ml21). Slides were washed in PBS/Tween-20 twice before
applying EnVision polymer (goat anti-mouse horseradish peroxidase
conjugate, DakoCytomation) for 30 min. Slides were washed with
PBS/Tween-20 twice before adding Vector NovaRED for 10 min, and
again washed with water. Slides were counterstained with haematoxylin for 1 min, washed in water, dehydrated through alcohols then
rinsed in xylene before mounting sections in DPX mountant.
microcolony was evident. Using mutants in flaA and flaB
of C. jejuni 81-176 we were never able to show microcolony
or biofilm formation in infected tissue. The flagella also
exhibited a unique morphology (Fig. 1e), which was not
evident in organisms grown in broth or suspended in tissue
culture media alone (data not shown). These observations
suggest microcolony formation of bacteria as an initial
colonization step for the human mucosa. Microcolonies
were particularly evident in small intestinal material, where
the majority of these bacterial groupings occurred. There
was a distinct difference in how organisms colonized the
small and large intestinal tissue, with bacteria being more
abundant in small bowel compared with colonic tissue. At
later time points, organisms were observed as a biofilm
over the surface of the ileal mucosa (Fig. 2). Immunohistochemical staining of human ileal tissue and viewing by
light microscopy also showed a biofilm over the surface of
the ileal tissue (Fig. 3).
RESULTS
In colonic tissue, bacteria were spread over the surface
mucus, and distinct groups or microcolonies were not
observed at up to 2 h post-challenge (Fig. 4). Occasional
groups of organisms were observed in colonic crypts.
Organisms were much more difficult to locate in the
colonic tissue than in small intestinal tissue infected for the
same time. Fig. 5 shows normal uninfected ileum and
colon, demonstrating the normal morphology of these
tissues at the end of the experimental time period (4 h).
In small intestinal tissue, bacteria were observed to adhere
to mucus and cells via their flagella in distinct groups, as
observed in Fig. 1. Organisms were particularly abundant
on mucus overlying cells. Bacteria adhered to cells where
mucus was absent but in much lower numbers or as
solitary bacteria only. Separate groups formed on the
mucosa distinctly adjacent to, but separate from, nearby
groupings. Organisms initially, during early time points,
did not coalesce over the entire surface of the mucosa and
organisms were found to have formed distinct microcolonies on the tissue surface (Fig. 1–d). Bacteria in
microcolonies adhered to the tissue via flagella. Bacterial
aggregation in these groupings also occurred by flagellar
interaction between individual organisms. Organisms
adhered to each other and flagellar interaction within the
Transmission electron microscopy (TEM) showed organisms interacting with surface mucus in close contact with
cell microvilli (data not shown). Despite numerous
attempts to locate large numbers of organisms by TEM,
only small numbers were observed. It is not known
whether the TEM processing affects organisms in tissue or
removes most of the mucus overlying the epithelium.
Fig. 2. C. jejuni 81-176 forms distinct biofilms on human ileal tissue 3–4 h after infection in vitro. Bacteria are adherent to the
mucosa, as tissue was washed before fixation.
http://mic.sgmjournals.org
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Fri, 16 Jun 2017 12:02:41
3081
G. Haddock and others
tissue via flagella in discrete colonies or multicellular
groups. In this study we have used a different strain of C.
jejuni and larger tissue samples from human small and
large intestine, and found differences in the bacterial
distribution on tissues.
Fig. 3. Immunohistochemistry of human ileal tissue in vitro-infected
with C. jejuni 81-176 (3–4 h after infection). The red staining is
specific for C. jejuni. No staining was observed for uninfected
tissue (not shown). The organism forms a biofilm on human ileal
tissue. Bar, 500 mm.
DISCUSSION
In similar work by Grant et al. (2006), using smaller
endoscopic biopsies of duodenum and then infecting in
vitro, C. jejuni 11168 was shown to adhere to mucosal
Fig. 4. C. jejuni 81-176 attaches to human colonic mucosa but
without the obvious microcolony or biofilm formation observed in
ileal tissue (4 h of incubation). The organisms were occasionally
observed as small clumps in colonic crypts (not shown).
3082
C. jejuni adheres to human small intestinal mucus in
microcolonies via flagella and at later time points as a
biofilm, shown as diffuse sheets of bacteria attached to
human ileal tissue. Microcolony formation of this type is
not observed in cell culture models of infection, which may
reflect the lack of mucus in these model systems. Adhesion
to small intestinal villus tip cells was also mediated by
bacterial flagella (data not shown). The flagella have
distinct morphology and mediate the adhesion of bacteria
to each other within these microcolonies. The significance
of the change in flagellar morphology is not known.
Organisms within the same SEM field showed normal
flagellar morphology, so only a subset of bacteria within
the microcolony express this changed flagellar morphology.
Microcolony or biofilm formation was never observed on
ileal tissue when mutants in flaA or flaB were used. It has
been shown that flagella are important for biofilm formation (Kalmokoff et al., 2006; Guerry, 2007; Svensson
et al., 2008). At later time points, human tissue infection
resulted in a distinct biofilm over the surface of the ileal
tissue, a phenomenon observed in vitro with glass and
plastic supports but as yet not observed in the organism’s
interaction with tissue (Svensson et al., 2008). The fact that
the organisms appeared to cover the entire tissue surface
raises questions as to how this observation may be linked to
the organism’s ability to cause diarrhoeal disease in man.
For many diarrhoeal pathogens colonization and persistence are important in the initiation of clinical symptoms.
How C. jejuni causes diarrhoea in man is unknown in
terms of the bacterial molecules that are required for the
initiation of symptoms. Extensive colonization of the
human gut mucosa would seem a reasonable pre-requisite
for disease to occur. It is tempting to speculate that such a
large concentration of organisms on an absorptive surface
prevents the normal absorptive transport functions of
the ileal mucosa and evokes other pathophysiological
alterations.
It is clear from the published literature that C. jejuni can
form biofilms (Svensson et al., 2008; Hanning et al., 2009;
Asakura et al., 2007). Published work relates to the formation of biofilms in the environment as a mechanism of
persistence and possible transmission and survival within
the poultry house (Howard et al., 2009; Svensson et al.,
2009). Water in poultry houses has been shown to be a
possible source of infection for birds and it has also been
shown that C. jejuni is present in biofilms in drinking
apparatus (Hanning et al., 2008). The biofilm consists of
aggregates and planktonic bacteria that live in a community consisting of either a single species or, when present in
the environment, multiple species. Presumably this way
of life confers some survival benefit on the members of
the community. The organisms within the biofilm show
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Fri, 16 Jun 2017 12:02:41
Microbiology 156
C. jejuni infection, biofilms and human tissue
Fig. 5. Low-power SEM images of uninfected specimens of the types of tissue used in this study. (a) Ileum, (b) colon.
Bars, 20 mm.
changes in metabolic activity and cell shape, and increased
tolerance to various stresses including enhanced resistance
to antimicrobial agents (reviewed by Svensson et al., 2008).
Although microcolony formation has been shown to be
important for biofilm formation in vitro (Guerry, 2007;
Golden & Acheson, 2002), it has not been shown before
that biofilm formation occurs on tissue in vivo or using an
in vitro tissue infection model.
could be used to help identify disease-causing mechanisms
of C. jejuni.
Bacteria were more readily visualized in ileal tissue than in
colonic tissue, and whether this reflects the true tissue
distribution in human infection is unknown and merits
further study. This is perhaps surprising in view of the
dogma that C. jejuni infection is predominantly a colitis,
with major colonic involvement in disease (Skirrow &
Blaser, 2000). Our observations raise the possibility that
the infection starts in the ileum, with more severe disease
extending to the colon, as observed in salmonellosis
(Skirrow & Blaser, 2000). Such a scenario is supported by
earlier observations of ileal involvement in occasional
patients who had undergone laparotomy or were examined
post-mortem (Butzler, 1981).
REFERENCES
Human intestinal tissue obtained at operation is a valuable
resource for studying the interaction of C. jejuni with its
target tissues in an infected host, and has shown the
formation of multicellular communities on tissues to be a
consistent feature of different strains of C. jejuni (Grant
et al., 2006). These models should facilitate the study of the
interaction of the bacterium with its target tissue at a level
of complexity not possible except in animal models, and
http://mic.sgmjournals.org
ACKNOWLEDGEMENTS
Financial support was from a grant by the Chief Scientist Office of the
Scottish Health Department (grant number CZB/4/394).
Asakura, H., Yamasaki, M., Yamamoto, S. & Igimi, S. (2007).
Deletion of peb4 gene impairs cell adhesion and biofilm formation in
Campylobacter jejuni. FEMS Microbiol Lett 275, 278–285.
Black, R. E., Levine, M. M., Clements, M. L., Hughes, T. P. & Blaser,
M. J. (1988). Experimental Campylobacter jejuni infection in humans.
J Infect Dis 157, 472–479.
Butzler, J. P. (1981). New prospects for treatment and prevention. In
Acute Enteric Infections in Children, p. 63. Edited by T. Holme,
J. Holmgren, M. Merson & R. Mollby. Amsterdam: Elsevier.
Golden, N. J. & Acheson, D. W. K. (2002). Identification of motility
and autoagglutination of Campylobacter jejuni mutants by random
transposon mutagenesis. Infect Immun 70, 1761–1771.
Grant, A. J., Woodward, J. & Maskell, D. J. (2006). Development of an
ex vivo organ culture model using human gastro-intestinal tissue and
Campylobacter jejuni. FEMS Microbiol Lett 263, 240–243.
Guerry, P. (2007). Campylobacter flagella: not just for motility. Trends
Microbiol 15, 456–461.
Hanning, I., Jarquin, R. & Slavik, M. (2008). Campylobacter jejuni as a
secondary colonizer of poultry biofilms. J Appl Microbiol 105, 1199–
1208.
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Fri, 16 Jun 2017 12:02:41
3083
G. Haddock and others
Hanning, I., Donoghue, D. J., Jarquin, R., Kumar, G. S., Aguiar, V. F.,
Metcalf, J. H., Reyes-Herrera, I. & Slavik, M. (2009). Campylobacter
Krause-Gruszczynska, M., Rohde, M., Hartig, R., Genth, H., Schmidt, G.,
Keo, T., König, W., Miller, W. G., Konkel, M. E. & Backert, S. (2007). Role
biofilm phenotype exhibits reduced colonization potential in young
chickens and altered in vitro virulence. Poult Sci 88, 1102–1107.
of the small Rho GTPases Rac1 and Cdc42 in host cell invasion of
Campylobacter jejuni. Cell Microbiol 9, 2431–2444.
Howard, S. L., Jagannathan, A., Soo, E. C., Hui, J. P., Aubry, A. J.,
Ahmed, I., Karlyshev, A., Kelly, J. F., Jones, M. A. & other authors
(2009). Campylobacter jejuni glycosylation island important in cell
Skirrow, M. & Blaser, M. (2000). Clinical aspects of Campylo-
charge, legionaminic acid biosynthesis, and colonization of chickens.
Infect Immun 77, 2544–2556.
Hu, L., Tall, B. D., Curtis, S. K. & Kopecko, D. J. (2008). Enhanced
bacter infection. In Campylobacter, pp. 68–88. Edited by
I. Nachamkin & M. J. Blaser. Washington, DC: American Society
for Microbiology.
Svensson, S. L., Frirdich, E. & Gaynor, E. (2008). Survival strategies of
microscopic definition of Campylobacter jejuni 81-176 adherence to,
invasion of, translocation across, and exocytosis from polarized
human intestinal Caco-2 cells. Infect Immun 76, 5294–5304.
Campylobacter jejuni: stress responses, the viable but nonculturable
state and biofilms. In Campylobacter, 3rd edn, pp. 571–590. Edited by
I. Nachamkin, C. Szymanski & M. J. Blaser. Washington, DC:
American Society for Microbiology.
Kalmokoff, M., Lanthier, P., Tremblay, T. L., Foss, M., Lau, P. C.,
Sanders, G., Austin, J., Kelly, J. & Szymanski, C. M. (2006). Proteomic
Svensson, S. L., Davis, L. M., MacKichan, J. K., Allan, B. J.,
Pajaniappan, M., Thompson, S. A. & Gaynor, E. C. (2009). The
analysis of Campylobacter jejuni 11168 biofilms reveals a role for the
motility complex in biofilm formation. J Bacteriol 188, 4312–4320.
Konkel, M. E., Monteville, M. R., Rivera-Amill, V. & Joens, L. A. (2001).
CprS sensor kinase of the zoonotic pathogen Campylobacter jejuni
influences biofilm formation and is required for optimal chick
colonization. Mol Microbiol 71, 253–272.
The pathogenesis of Campylobacter jejuni-mediated enteritis. Curr
Issues Intest Microbiol 2, 55–71.
Edited by: P. C. F. Oyston
3084
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Fri, 16 Jun 2017 12:02:41
Microbiology 156