All patients fulfilled Rome-II criteria for IBS . A total of 65 patients (61 females and 4 males) with a median age of 48 (range 22-67) years and a median duration of IBS symptoms of 6.5 years (range 0.6-33.2 years) were investigated. Diarrhoea predominant IBS was present in 21 patients (32%), 22 patients (34%) had constipation predominant IBS and 22 patients (34%) had IBS with alternating bowel habits. All patients had severe symptoms of IBS  and 26 patients also exhibited abnormalities on small bowel manometry, thus qualifying for a pathophysiological diagnosis of enteric dysmotility . Full thickness jejunum biopsies had been taken in 60 of our patients. Previous histopathological analysis had revealed neuropathic changes in 58 patients [5, 17, 18]. Neuropathy was associated with low-grade inflammation (LG = lymphocytic ganglionitis) in 46 patients and 20 of these also exhibited increased numbers of intraepithelial lymphocytes (LEG = lymphocytic epithelio-ganglionitis), whereas 12 patients had degenerative neuropathy (DN) without inflammation. Deficient staining for alpha-actin without neuronal damage was observed in 2 patients.
The control group comprised 42 persons (29 females) in whom IBS and all other functional bowel disorders had been excluded by medical interview and a validated questionnaire for the Rome-II symptom criteria. Ten controls (7 females) were obese but otherwise healthy (BMI mean = 42.8, SD = 4.3). The rest of the control group (32 persons, 22 females) consisted of healthy volunteers. The median age of the controls was 36 (range 19-60) years.
Full thickness jejunum biopsy
Previously obtained full-thickness biopsy specimens were available in 60/65 patients. The biopsies had been taken from the proximal jejunum using a laparoscopy-assisted procedure described by Tornblom et al Ten obese controls underwent full thickness biopsy of the jejunum at the time of gastric by-pass surgery.
Mucosa specimens from the proximal jejunum were taken with a Watson capsule in 32 controls and 6 patients. The capsule was swallowed by the subject and brought by peristalsis to a position distal to the ligament of Treitz as determined by fluoroscopy. We analyzed archived endoscopic mucosa biopsies from the duodenum of 20 patients and in 15 of these full thickness biopsies were also available. In 2 patients we analyzed mucosa biopsies from both the jejunum and the duodenum.
Immunofluorescence was performed using a genus-specific mouse monoclonal antibody to Chlamydia lipopolysaccharide (LPS)-FITC conjugated with Evans blue (RDI-PROAC1FT, Fitzgerald Industries International, Concord, MA, USA) and C. trachomatis major outer membrane protein (MOMP) as primary antibody (GeneTex, San Antonio, TX, USA) with a polyclonal rabbit anti-mouse antibody-FITC conjugated (Dako, Glostrup, Danmark) as the secondary antibody. We used cultured HeLa-cells infected with C. trachomatis as positive control and uninfected cells as negative control. We used a species-specific mouse monoclonal antibody for C. pneumoniae as primary antibody (GeneTex, San Antonio, TX, USA) with a polyclonal rabbit anti-mouse antibody-FITC conjugated (Dako, Glostrup, Denmark) as the secondary antibody. The presence of Chlamydia LPS was also evaluated using immunohistochemistry with a polyclonal rabbit antibody to C. trachomatis LPS (Fitzgerald Industries International, Concord, MA) and an immunoenzymatic assay with Streptavidin-biotin complex (Dako, Glostrup, Danmark).
Enteroendocrine cells were identified using rabbit polyclonal antibodies to Chromogranin A (Abcam, Cambridge, UK), and we used rabbit polyclonal antibodies to CD117 (Dako, Glostrup, Denmark) for mast cells, rabbit polyclonal antibodies to CD68 (Santa Cruz Biotechnology, Santa Cruz, CA, USA) for macrophages and rabbit monoclonal antibodies to CD11c (Abcam, Cambridge, UK) for dendritic cells. Goat anti-rabbit antibodies conjugated with Alexa Fluor 568 or Alexia Fluor 350 (Invitrogen, Carlsbad, CA, USA) were used as secondary antibody for all of the rabbit antibodies.
Stained sections were examined using a Universal Laser Scanning Confocal Microscope System Leica TCS and a Fluorescent Microscope System Leica DMRXA (Leica Microsystems, Wetzlar, Germany). Two independent investigators (AD and GM), who were unaware of clinical data, made the final assessment of slides. The slides were reviewed by a third investigator (BV). The immunofluorescence stainings were considered positive if more than one cell showed fluorescence. If only one positive cell was found, the staining-procedure was repeated. The case was regarded positive if the colour signal was again present, otherwise the case was recorded as negative. Double immunofluorescence stainings were also performed for the identification of LPS-positive cells (LPS with chromogranin, CD68, CD117, or CD11c, respectively).
Snap-frozen biopsies from 4 patients that were positive for Chlamydia LPS staining, were examined by Western blotting. HeLa cells infected with C. trachomatis served as positive control and we used non-infected HeLa cells as negative control. Equivalent amounts of protein from each specimen were loaded onto a sodium dodecyl sulphate-polyacrylamide gel. After electrophoresis, samples were transferred to nitrocellulose membranes. We used a mouse monoclonal antibody to C. trachomatis LPS (AbD Serotec, Oxford, UK) as primary antibody and goat anti-mouse antibody conjugated to horseradish peroxidase (BioRad, Herculaes, CA, USA) as secondary antibody. For MOMP we used the same primary antibody as for immunofluorescence.
We used Laser Microdissection Pressure Catapulting (LMPC) for laser based non-contact extraction of tissue areas in paraffin embedded biopsies from 6 patients. Regions of interest were manually delineated using fluorescence microscopy and the LMPC software. Tissue collection was achieved by laser cutting along the delineation lines to separate tissue of interest from surrounding regions, and secondly the laser catapulted the tissue of interest up into the lid of an Eppendorf cap containing sterile water. DNA was extracted using the QIAamp DNA mini kit, according to the instructions of the manufacturer, and analyzed by PCR.
In the present study, we used the real-time PCR assay developed by Everett et al. , which amplifies 23S ribosomal DNA, and detects all members of the family Chlamydiaceae. An internal amplification control was included to monitor possible inhibition of the PCR. DNA from frozen biopsies, taken from 4 patients previously positive for Chlamydia LPS staining, was extracted using a Qiagen minikit according to the tissue protocol (Qiagen, Solna, Sweden). Extracted DNA was quantified and quality controlled using a Nanodrop spectrophotometer (Nanodrop Technologies, Wilmington, DE, USA) before being subjected to PCR.
Transmission electron microscopy
Biopsies from the distal duodenum of 4 patients that were positive for Chlamydia LPS staining were fixed according to a procedure described before . Semi-thin sections were cut and stained with toluidin blue and used for light microscopic analysis. Ultra-thin sections were contrasted with uranyl acetate followed by lead citrate and examined in a Tecnai 10 transmission electron microscope at 80 kV. Digital images were taken by using a MegaView III digital camera (Soft Imaging System, Münster, Germany).
We used logistic regression with age and gender as covariates for calculation of odds ratios and p-values for comparisons of proportions. The size of the study group was determined from the assumption that LPS positivity would be no greater than 20% among controls. In order to detect a risk factor with an odds ratio of at least 6 at p < 0.01 with power >90% we needed to include at least 37 patients and 37 controls (two-sided test).
All parts of the study were approved by the Regional Ethical Review Board in Stockholm. Informed consent was obtained from all patients and controls at the time of biopsy taking.