Article Text


NOS inhibitors in colitis: a suitable case for treatment?
  1. T D WARNER,
  1. Vascular Inflammation,
  2. The William Harvey Research Institute,
  3. St Bartholomew’s and the Royal London
  4. School of Medicine and Dentistry,
  5. Charterhouse Square,
  6. London EC1M 6BQ, UK

Statistics from

See article on page 180

Nitric oxide (NO) is now well characterised as a physiologically important molecule, acting most notably as an intra- and intercellular messenger within the cardiovascular and nervous systems.1 To fulfil these functions NO is produced in a tightly controlled manner by calcium dependent, constitutively expressed nitric oxide synthases (cNOS) present within endothelial and neural cells. Under pathological conditions, however, NO acts very differently from a benign signalling molecule. In these circumstances NO is produced in large, apparently tissue damaging amounts by inducible, calcium unregulated nitric oxide synthases (iNOS). For example, in inflammatory bowel disease the presence of high local concentrations of pro-inflammatory cytokines, such as interleukin 8, is associated with the induction of iNOS and so the continuous local generation of NO,2 3 possibly in very great amounts.4 Notably, this difference in NO production between cNOS and iNOS enzymes is not due, as is often remarked, to the inducible isoform of NO synthase being capable of producing more NO than the constitutive forms. Indeed, cNOS and iNOS enzymes function with very similar kinetics, co-factor requirements and substrate affinities. The important characteristic of iNOS enzymes is that they are not regulated by intracellular calcium. So, once iNOS is expressed within a cell it immediately sets to work forming NO.

The characterisation of cNOS and iNOS permitted a neat theory to be put forward to explain the conflicting roles of NO. cNOS enzymes made NO in low concentrations as a signalling molecule, whereas iNOS enzymes made NO in high, cell damaging, concentrations. However, there was a problem with this theory. NO by itself is not especially reactive. So, how does the production of NO in ungoverned amounts contribute to the tissue damage seen in many inflammatory states? The answer to this question was provided by Beckman and colleagues5 who showed that the reaction of NO with superoxide anion was not a simple inactivation process6 but rather an important activation step, resulting in the formation of peroxynitrite (ONOO). Unlike NO, peroxynitrite is a reactive species, giving rise to further damaging molecules, most notably the extremely destructive hydroxyl anion (OH). The formation of reactive peroxynitrite by this combination between NO and superoxide anion explains why—for example, neutrophils are required to produce both species if they are to be effective killers of invading cells.5 In inflamed tissue we should also expect peroxynitrite to be formed, as iNOS is present (see earlier) along with considerable populations of invading cells, many of which produce superoxide anion. As long as the NO and superoxide anion are produced in close proximity we must conclude that peroxynitrite is formed, and importantly that this could be central to the inflammatory damage seen. Indeed, instillation of the highly reactive and short-lived peroxynitrite directly into the rat colon unsurprisingly causes notable local damage and inflammation.7 In this issue (see page 180), Kimura and colleagues, supporting the work of Singer et al,8 provide further evidence that NO produced by iNOS within the colonic mucosa of active ulcerative colitis is directly associated with the local production of peroxynitrite. In essence, these authors have found iNOS to be present throughout the colonic mucosa in amounts that are proportional to the degree of mucosal inflammation. This does not show that peroxynitrite is also formed, and unfortunately because of its highly reactive nature there is no way to measure peroxynitrite concentrations directly within living tissues. However, peroxynitrite reacts with proteins leaving behind tell-tale nitrotyrosine residue “footprints”. These “footprints” can be detected by selective antibodies, which Kimuraet al have used to show that, as for iNOS, there is a clear correlation between nitrotyrosine groups and tissue inflammation.

Clearly, our most important question must be whether or not the observations of Kimura et al and others will lead to the use of NOS inhibitors as new therapies for inflammatory bowel disease. Unfortunately, our state of knowledge suggests that the answer to this question is no. Currently, there are few genuinely selective iNOS inhibitors and no reports of their use in animal models of these disease states. Non-selective cNOS and iNOS inhibitors have been tested and sometimes found to be therapeutically effective in animal models.9 However, these non-selective agents also inhibit cNOS very well and can exacerbate intestinal damage10 by both reducing local blood flow and promoting neutrophil adhesion to the blood vessel wall.11 Although it is hoped these side effects could be avoided by the use of selective iNOS inhibitors, it must be borne in mind that cNOS isoforms could also be up-regulated in inflamed tissues, making the selective targeting of a disease specific NOS isoform impossible.12 Also supporting our cautious conclusion are studies demonstrating that in a disease model closer to the human, rhesus monkeys displaying spontaneous chronic colonic inflammation, administration of NO synthase inhibitors reduces neither histological inflammatory scores nor diarrhoeal symptoms.13

One final point of interest arising from Kimura et al’s study is, as remarked by the authors, that many of the subjects were taking steroids without any noticeable effect on iNOS expression or activity, or peroxynitrite formation. Steroids dramatically down-regulate iNOS expression in numerous cell and animal models. Does this reflect different dosing levels of steroids in animal models and human disease, different pathways of iNOS regulation in acute and chronic inflammatory states, or differences between the human and animal models? The data presented here do not permit us to draw a conclusion. However, as we investigate the actions of iNOS inhibitors in human disease states, it is clearly vital to understand why such a powerful inhibitor of iNOS expression was without effect.

See article on page 180


View Abstract

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.

Linked Articles