Article Text

Download PDFPDF

Expression of isoforms of nitric oxide synthase in collagenous colitis
  1. E A B Cameron1,
  2. S J Middleton1,
  3. P J Roberts2
  1. 1Department of Gastroenterology, Addenbrooke's NHS Trust, Cambridge CB2 2QQ, UK
  2. 2Department of Gastroenterology, Hinchingbrooke NHS Trust, Huntingdon, Cambridge PE18 8NT, UK
  1. Correspondence to:
    S J Middleton;
  1. A Perner3,
  2. L Andresen3,
  3. J Rask-Madsen3
  1. 3Department of Gastroenterology, Herlev Hospital, University of Copenhagen, Denmark

Statistics from

We read with interest the study by Perner et al (Gut 2001;49:387–94) investigating expression of various isoforms of nitric oxide synthase (iNOS, eNOS, and nNOS) in non-inflamed colon, collagenous colitis, and ulcerative colitis. Inducible NOS (iNOS) was identified by immunohistochemical analysis in the epithelium of patients with non-inflamed colon. The authors concluded that this might be a result of bowel preparation with bisacodyl.

Increased synthesis of nitric oxide has been detected by a number of different methods in patients with ulcerative colitis.1,2 We have previously found physiological expression of iNOS in histologically normal colon using reverse transcription-polymerase chain reaction (RT-PCR), immunohistochemistry, and immunoblotting.3 Tissue from three different sources was studied. Surgical specimens were obtained from patients undergoing colectomy for colorectal cancer who had undergone bowel preparation with sodium picosulphate, colonoscopic biopsies from patients also prepared with sodium picosulphate, and rectal biopsies at sigmoidoscopy from patients who had received no bowel preparation. This last group of patients also underwent colonoscopy with sodium picosulphate preparation confirming the absence of colonic pathology. iNOS mRNA was identified in all samples by RT-PCR. iNOS protein was detected by immunoblotting in 77% of samples, by immunostaining in 80% of surgical specimens, and in 90% of biopsy specimens.3 It is therefore possible that expression of iNOS in epithelial cells reported by Perner et al is, as we found, a result of physiological expression of iNOS rather than a secondary phenomenon as a result of bisacodyl bowel preparation.

The reason for iNOS expression in normal colonic epithelium is not currently clear. Nitric oxide production may aid maintenance of the epithelial barrier by preventing bacterial translocation or by inducing apoptosis. It is also possible that its presence represents a link between dietary or other luminal factors and the development of colorectal cancer.

Figure 1

Expression of inducible nitric oxide synthase in mucosal biopsies from the unprepared (Unprep) sigmoid colon and 12 hours following bowel preparation with bisacodyl (Bis) or polyethylene glycol 3000 (PEG), analysed by western blotting and quantified by densitometry relative to a reference. Values in individual subjects are represented by circles and connected by lines. OD, optical density.


Authors' reply

We thank Cameron et al for their comment on our recent publication and important observations that suggest physiological expression of inducible nitric oxide synthase (iNOS) in histologically normal human colon. As we observed subtle iNOS labelling in colonic mucosal biopsies from our group of controls with non-inflamed bowel, we have subsequently studied whether bowel preparation with bisacodyl or polyethylene glycol prior to sigmoidoscopy might induce iNOS expression.

Ten healthy non-smoking male subjects were investigated. Mucosal biopsies were taken from the sigmoid colon prior to bowel preparation and again 12 hours after rectal administration of an enema consisting of bisacodyl (100 mg) or polyethylene glycol 3000 (6.4 g in 100 ml of water) in randomised order. Expression of iNOS protein was quantified by western blot analysis and localised by immunohistochemistry.

iNOS was expressed in the colonic mucosal biopsies from all subjects and localised in epithelial cells, particularly at the luminal border of the epithelial cells and more pronounced in the crypt epithelium. Expression of iNOS was unaffected by bowel preparation with bisacodyl or polyethylene glycol (fig 1).

Hence we agree with Cameron et al that expression of iNOS in epithelial cells is possibly a result of physiological expression of iNOS rather than a secondary phenomenon as a result of the bowel preparation per se or the effect of the secretagogue laxative bisacodyl. For the reasons given above, we also agree that nitric oxide may be important in maintaining the epithelial barrier and may represent a link between dietary or other luminal factors and the development of colorectal cancer, as hypothesised by Cameron et al, although high iNOS expression in collagenous colitis is not associated with an increased risk of malignancy.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.