Article Text


A key cytokine unlocks the door
  1. H SHIRIN,
  2. S F MOSS
  1. Department of Medicine
  2. Division of Gastroenterology
  3. St Luke’s-Roosevelt Hospital Centre
  4. College of Physicians and Surgeons
  5. Columbia University, New York, NY 10025, USA
  1. Professor Moss (email: smoss1{at}

Statistics from

See article on page 24

The synthesis and secretion of a variety of pro-inflammatory cytokines are increased in the chronic inflammatory reaction associated with Helicobacter pyloriinfection.1 In this issue (see page 24), Shibata and colleagues confirm increased secretion of tumour necrosis factor (TNF) by the inflamed mucosa, and provide evidence that duringH pylori infection there is both secretion of TNF and of the plasma membrane receptors for TNF, which are cleaved and secreted by gastric epithelial cells. Further experiments suggest that this secretion of soluble TNF receptors may be important in allowing the epithelial cell to protect itself from TNF induced death. Although there are relatively few normal controls in the human biopsy study, these results do suggest that the secretion of soluble TNF receptors may be an important defensive reaction, and may have implications for other inflammatory conditions of the gastrointestinal tract.

Since its discovery over 20 years ago, TNF has been recognised as an important mediator in the pathogenesis of a wide variety of diseases. TNF is one of family of related peptides, including Fas ligand and nerve growth factor, that acts through specific related cell surface associated receptors.2 The multiple biological effects of TNF are mediated through two specific, high affinity cell surface receptors (a 55 kDa TNF-R1 and a 75 kDa TNF-R2) which are found on most mammalian cells. Both TNF receptors may exist in soluble forms by proteolytic cleavage of the extracellular domain of the TNF receptor or by the translation of specific transcripts formed by alternative splicing mechanisms. Enzymes responsible for the shedding of TNF and for its receptors are currently being investigated, as are the roles of “decoy”, non-functioning receptors. As in all highly regulated systems, mechanisms exist to regulate the effects of important mediators and to this effect, soluble TNF receptors seem to adjust TNF bioactivity in a dose dependent manner. At high concentrations they compete for TNF binding and reduce TNF bioactivity3 and at lower concentrations they stabilise the trimeric structure of TNF, thereby providing a reservoir for the slow release of TNF bound to its receptor.4 Measuring increased plasma concentrations of soluble TNF receptors has been proposed for the diagnosis of several neoplastic and inflammatory diseases.5

Although only one of many cytokines in a complex soup, TNF has a variety of biological effects which may be relevant to the understanding of the pathogenesis of diseases associated withH pylori. These include chronic incitement of the inflammatory cascade, immunomodulation, cytotoxicity (including apoptosis), and even mitogenic stimulation and/or tumour promotion.2 6 By examining the secretion of soluble TNF receptors in H pylori associated, TNF induced apoptosis, Shibata and colleagues’ paper adds to our understanding of the complex regulatory systems that must be in place to limit TNF induced damage. However, many other issues related to the interaction between TNF and H pylori have still to be resolved. For example, it is not clear how TNF-R1 is somehow capable of transmitting several, sometimes antagonistic, signals. The intracellular portion of TNF-R1 contains a cytoplasmic death domain that is required for the signalling of certain activities such as apoptosis and activation of transcription factors nuclear factor κB (NF-κB) and AP-1. Binding of TNF to the extracellular domain of TNF-R1 induces receptor trimerisation, allowing the death domain of TNF-R1 to recruit the adapter proteins TRADD (TNF receptor associated death domain), FADD (Fas associated death domain), TRAF2 (TNF receptor associated factor), and RIP (receptor interacting protein).7 This TNF-R1 complex then activates signalling cascades leading to apoptosis, activation of JNK/SAPK (jun kinase/stress activated protein kinase) pathways and NF-κB activation. Although FADD interacts directly with the apoptotic proteases, thus triggering cell death, stimulation of cytoprotective gene expression mainly occurs through specific TRAFs. Experiments using TRAF2−/− mice have suggested that transcription of cytoprotective genes by TRAF2 is the first and the most important role of TNF gene stimulation as these mice were unable to upregulate JNK/SAPK activity after TNF treatment, yet were highly sensitive to TNF dependent apoptosis.8 TNF-R1 mediated cell survival also seems to involve RIP, dominant negative forms of which can inhibit NF-κB and JNK activation, and overexpression of which induces both JNK and NF-κB. During infection withH pylori, does TNF act predominantly to damage epithelial cells, or to protect them from injury? In cell culture experiments, there is evidence that TNF can augmentH pylori induced apoptosis,9 as well as data showing that H pylori can induce NF-κB.10 Cell lines derived from gastric cancer may not be the best models for examining the effects of TNF or ofH pylori however. This is because despite TNFs original designation as an agent causing tumour cell death, most cancer cells are usually resistant to apoptosis induced by TNF unless sensitised by translational inhibitors or blockers of the TRAF2/NF-κB pathway. Infecting mice lacking the 55 kDa TNF receptor withH pylori could clarify the role of TNF inH pylori induced gastritis and apoptosis. These mice, which are moderately resistant to the lethal effects of lipopolysaccharide but highly susceptible to infection by Listeria monocytogenes,11 may teach us whether TNF is more important for inducing the inflammatory process or in protecting the mucosa from damage.

The pharmacological modulation of TNF in chronic inflammatory diseases such as rheumatoid arthritis and Crohn’s disease is already showing great promise, but our knowledge of how the gut responds to and regulates TNF is still at a relatively primitive stage. The observation that TNF may provide mitogenic signals to cancer cells may be surprising,6 especially as TNF was first described as a substance capable of killing cancer cells, but it serves to illustrate how difficult it can be to predict the biological effects of the same substance in different systems. In this context, we should not be surprised if further paradoxical effects of TNF are observed as the mechanisms of its regulation are uncovered.

See article on page 24


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