Final Progress Report
Proposal No. IBD-0183
Principal Investigator: Hazel M. Mitchell, Ph.D.
Applicant Organization: The University of New South Wales (Australia)
Project Title: Investigation of the role of members of the Helicobacteriaceae in Crohn’s disease
Period of Award: September 7, 2006 – March 6, 2008
A summary of project aims:
The aim of our research project was to further investigate the role of bacteria belonging to the Helicobacteriaceae family, in the initiation and/or continuation of the inflammation associated with CD. We chose to conduct our studies in children as they are in the early stages of disease and relatively free of significant confounding factors that make such studies in adults more difficult. The specific aims of the study are outlined below.
Aim1: To collect intestinal biopsies and faecal samples from children undergoing colonoscopy and diagnosed with CD at hospitals in 3 different centres in Australia, Canada and USA.
Aim 2: To collect stool samples from well children with no bowel problems (of the same age & ethnicity as CD children) attending each of the 3 locations.
Aim 3: To detect using molecular techniques, the presence of members of the Helicobacteriaceae in colonic samples and/or stools from these children and where present identity the species involved.
Aim 4: To detect using molecular techniques, the presence of members of the Helicobacteriaceae in colonic samples and/or stools from these children and where present identity the species involved.
Aim 5: To assess, in children with CD, the location (if present) of these bacteria using special techniques.
Aim 6: To determine if an association exists between all members, or a subset of species, from the Helicobacteriaceae familyand CD.
Accomplishments towards meeting those aims:
Aim1:
The study aimed over a 1-year period to collect biopsies from a total of 30 children (aged 2-18 years), diagnosed with CD and 15 control children from each of three geographic locations (Australia, Canada and USA).
To date we have collected biopsy samples from 33 CD patients. The number of biopsy specimens from CD patients is well below our planned number. This was due to a number of reasons. Firstly our USA collaborator failed to supply any samples (Letter previously submitted to the Broad Foundation). To try to increase the numbers of samples collected last year we established a collaboration with a number of gastroenterologists working at the Children’s Hospital Westmead, Sydney. This collaboration has worked well and although it took some time to get ethics approval for this new site we are now regularly receiving specimens. To date the number of samples from Canada has been low due to the fact that it took nearly 6 months for this site to obtain ethics approval. Although late last year we shipped 6 specimens to Australia, given the large cost involved with the transport of specimens from Canada ($1800 dollars/shipment) we decided to wait until a large number of samples had been collected before shipping the next batch to Australia. Specimen collection is continuing in Canada and these specimens will be processed before the end of this year. In addition to the above biopsy specimens from the children with CD we have also collected faecal samples from a further cohort of 29 children with CD and thus over this time frame we have collected and examined 62 samples from CD patients.
Aim 2:
To date we have collected faecal samples 60 control children. These are made up of 32 normal healthy controls (faecal samples) and 28 non-IBD controls (biopsy controls).
Aim 3:
The presence of members of the Helicobacteraceae has been determined in all biopsy and 29 stool samples collected from children with CD as well as all 60 controls using a Helicobacteraceae specific PCR. For all biopsy samples positive for members of the Helicobacteraceae by PCR the PCR products were sequenced to putatively identify the specific Helicobacter species involved.
Due to the fact that in our culture studies (see below) we isolated 4 Campylobacter species from CD patients and controls we also decided to examine all of the DNA from the biopsy and faecal samples for the presence of Campylobacter species in both the CD patients (n=59) and 60 controls using a Campylobacter specific PCR. All positive PCR samples were sequenced and putative identification of the Campylobacter species identified. Our rational for doing this was the fact that the Campylobacter genus is closely related phylogenetically to the Helicobacter genus and alsothey are mucus-associated bacteria.
Aim 4:
All biopsy samples collected from the two Sydney hospitals were cultured for the isolation of Helicobacter and Campylobacter species. Three different culture media were used for the isolation of Campylobacter species. These were 1) Horse blood agar plate containing cefoperazone (20μg/mL) (Sigma Aldrich, Sydney, Australia), vancomycin (10μg/mL) (Sigma) and amphotericin B (2μg/mL) (Bristol-Myers Squibb, Australia); 2) Horse blood agar plates containing trimethoprim (10μg/ml) and vancomycin (10μg/ml) and 3) Horse blood agar plates containing vancomycin (10μg/ml). In addition a filtration methodology was employed. All plates were cultured under microaerophilic and anaerobic conditions. Mixed bacterial cultures of members of the Helicobacteraceae a of Helicobacter and other bacteria were identified in 4 children and these were identified by PCR. The identity of these Helicobacters as determined by sequencing were H. pylori In addition we isolated four different non-jejuni Campylobacter species from biopsies of children with CD and three different non-jejuni Campylobacter species from controls.
Aim 5:
Biopsy samples have been collected from all children (CD and controls) for examination by Fluorescent in situ hybridization (FISH). If the biopsy is positive by Helicobacteraceae PCR then these samples are examined by FISH. Currently we have conducted FISH analysis on 10 biopsies collected from children with CD and from 6 biopsies collected from the control children. Helicobacteraceae specific FISH positive signals were detected in 4 biopsies collected from children with CD and 2 biopsies collected from controls. Processing of the remaining samples will be conduced this year.
Aim 6:
Statistical analysis was used to compare the presence of all members Helicobacteriaceae family as well as enterohepatic Helicobacter species and Helicobacter pylori in children with CD and controls.
A list of significant results (positive or negative):
Members of Helicobacteraceae:
1. A significantly higher prevalence of Helicobacteraceae DNA was detected by PCR in biopsy samples from children with CD (59%) as compared with controls (21%) (p<0.05).
2. Sequencing of the PCR products showed these to be similar to a range of non-pylori Helicobacter species and Helicobacter pylori.
3. FISH identified Helicobacteraceae specific positive signals in the mucus layer of 4 CD children and 2 controls, a finding that not only confirms the presence and location of these bacteria but also that they are viable.
4. Culture of biopsy specimens for Helicobacter species resulted in the growth of Helicobacter species from 4 children however unfortunately on all occasions these Helicobacter species could only be identified by PCR from mixed bacterial cultures of Helicobacter and other bacteria.
5. A significantly higher detection rate of members of the Helicobacteraceae was also observed by PCR in stool samples collected from children with CD (59%) as compared with controls (3%) (p<0.001).
Campylobacter species:
1. A significantly higher detection rate of Campylobacter species was observed using PCR on biopsy samples collected from CD children (82%) as compared with controls (19%) (p<0.001).
2. Sequencing of the PCR products showed these to be similar to a range of non-jejuni Campylobacter species with in children with CD 59% being Campylobacter concisus.
3. Four different non-jejuni Campylobacter species were isolated from biopsies of children with CD and three different non-jejuni Campylobacter species from controls.
Lay summary of the progress report:
Studies examining the cause of Inflammatory Bowel Disease (IBD), which includes both Crohn’s disease (CD) and ulcerative colitis (UC), have shown that bacteria play an important role. However, to date there is no clear evidence to support a role for any particular organism, or group of organisms, in IBD.
We were interested in studying the possible role of intestinal mucus associated bacteria particularly lower bowel Helicobacter species and Campylobacter species in the initiation of IBD. Studies in rodents have shown that as compared with other intestinal mucus-associated bacteria, lower bowel Helicobacter species and Campylobacter species colonize areas in close proximity to the intestinal epithelium. Given that the inflammatory response in IBD occurs at mucosal surfaces, we hypothesized that lower bowel Helicobacter species and Campylobacter species may be the most likely candidates to be involved in the initiation of IBD.
In this project, we investigated the presence of members of Helicobacteraceae family and Campylobacter species in both biopsy samples and stool samples collected from children with CD and controls using PCR and sequencing of PCR products. In addition we used a Helicobacteraceae specific fluorescent rRNA in situ hybridization (FISH) technique to assess the location of members of Helicobaceteraceae family in the intestinal tissues. Furthermore, we cultured biopsy samples to try to isolate Helicobacter and Campylobacter species.
Over the period of the grant, biopsies from 33 children with CD and 52 controls were collected. In addition, we have also collected stool samples from 29 children with CD and 37 controls.
Using PCR, we have detected significantly higher Helicobacteraceae DNA in biopsies collected from children with CD (58%) as compared with controls (21%) (p<0.05). Sequencing of the PCR products showed these to be similar to non-pylori Helicobacter species and Helicobacter pylori. FISH identified Helicobacteraceae specific positive signals in the mucus layer of 4 CD children and 2 controls. Culture of biopsy specimens for Helicobacter species resulted in Helicobacter species DNA being detected in mixed cultures from 4 CD children. Significantly higher presence of members of the Helicobacteraceae were detected using PCR in stool samples collected from children with CD (59%) to compared with controls (3%) (p<0.001).
We have also detected significantly higher presence of Campylobacter species using PCR in biopsies from CD children (82%) as compared with controls (19%) (p<0.001). Sequencing of the PCR products showed these to be similar to a range of non-jejuni Campylobacter species including C. concisus. The presence of C. concisus was significantly higher in children with CD than that in controls (p<0.05). In addition, we have successfully for the first time isolated four different non-jejuni Campylobacter species from biopsies of children with CD and three different non-jejuni Campylobacter species from controls.
Results produced from this project has led to a success in attracting three years funding from National Health and Medical Research Council of Australia for further investigation of the possible role of members of Helicobacteraceae in the development of human IBD. In addition, this project will result in three journal publications.