Sepsis is a complex, exaggerated and chaotic version of the usually well-organized inflammatory arm of our immune defences, and kills over 175,000 people each year in the United States alone1. Although a great deal of time and effort has been spent researching septic shock, it remains difficult to understand and treat. One promising lead was provided two years ago, when it was discovered that there is a connection between inflammation and the involuntary nervous system. The details of this link have, however, been unclear — until now. Writing on page 384 of this issue, Kevin Tracey and colleagues2 describe how they identified a receptor protein that is stimulated by the nervous system and which in turn inhibits a key molecular mediator of inflammation and septic shock. This receptor might make a good target for future drugs to treat sepsis.
Inflammation has several roles in the body, one of which is to contribute to the immune system’s ability to fight off intruding microorganisms. For instance, molecules that are produced during the inflammatory response increase blood flow to infected areas, or help to recruit immune cells. One way in which inflammation is triggered is in response to lipopolysaccharides — components of the cell walls of many bacteria — which activate the immune system’s macrophages. These cells in turn release ‘alarm’ molecules, namely cytokines, some of which have powerful pro-inflammatory properties. Tumour-necrosis factor (TNF) is one such molecule. This protein can affect nearly all cell types, and has a range of biological activities. For instance, it induces the expression of a large number of genes that encode essential inflammatory molecules (such as other cytokines; enzymes that help to break down the barriers between cells, allowing the migration of immune cells; and adhesion molecules that again enhance immune-cell migration)3, 4.
As long as TNF production remains confined to the site of infection, the inflammatory response is clearly beneficial. But once bacteria, and consequently TNF, invade the systemic blood circulation, blood ‘poisoning’ and sepsis can develop quickly. Furthermore, TNF has been found to be a central mediator of chronic inflammatory disorders such as rheumatoid arthritis and Crohn’s disease. So there is much interest in learning how to control the production, release and activity of TNF. Several means of doing so have been developed (Fig. 1), and have seen some success in treating certain inflammatory disorders5. For instance, there are drugs that inhibit the transcription of the TNF-encoding gene into messenger RNA, the translation of the mRNA into protein, or the release of the TNF protein. There are also antibodies and soluble receptors that bind to and block TNF once it has been released. But, although the value of these approaches is beyond doubt, they all take time to work — and time is usually short when treating patients with sepsis.
Tracey’s research team has been studying TNF since this protein was discovered (see, for instance, ref. 6). Recently, Tracey’s group described another level of control of TNF synthesis — namely by means of the vagus nerve7 — thereby providing a new and exciting link between the involuntary nervous system and inflammation. This ‘parasympathetic’ nerve emanates from the cranium and innervates all major organs in a subconscious way. It is finely branched and is composed of both sensory (input) and motor (output) fibres. This is of relevance because it means that the vagus nerve can on the one hand sense continuing inflammation (presumably by detecting cytokines through receptors on the nerve surface), and on the other hand suppress it. This suppression is efficient and, above all, a good deal faster than the mechanisms mentioned above. Tracey’s group found7 that, after injecting lipopolysaccharides into rats, electrically stimulating the vagus nerve prevented both the release of TNF from macrophages, and death. Conversely, surgically severing the nerve not only removed this protection but also sensitized the animals to lipopolysaccharide.
But how does the vagus nerve have this effect on macrophages? It was already known that, after this nerve is stimulated, its endings release the neurotransmitter molecule acetylcholine with lightning speed. Macrophages express acetylcholine receptors known as nicotinic receptors, and respond to the released acetylcholine (or the acetylcholine-mimicking nicotine) by suppressing TNF release. But the precise identity of the nicotinic receptors on macrophages was not known. From a therapeutic point of view, this is clearly important to know. It’s also very difficult to find out, as the receptors are pentamers containing different combinations of a possible 16 monomers.
In their latest paper, Tracey and colleagues2 pin down the relevant nicotinic acetylcholine receptor: it is one comprising five copies of the monomer 7. They started by using -bungarotoxin, a molecule that binds to just a subset of receptor monomers, to show that macrophages express the 7 subunit. When the authors blocked the expression of this protein, acetylcholine and nicotine were no longer able to prevent the release of TNF — data that the authors confirmed by studying 7-deficient mice. In fact, such mutant mice displayed an exaggerated response to lipopolysaccharide in terms of their production of the cytokines TNF, interleukin-1 and interleukin-6. Finally, in a technical tour de force, Tracey and colleagues showed that electrically stimulating the vagus nerve of 7-deficient mice no longer afforded protection against lipopolysaccharide (in contrast to the situation in wild-type mice).
These findings2 could have therapeutic implications. The discovery of the connection between the involuntary nervous system and inflammation had already yielded new ideas about treating inflammatory disorders such as sepsis: for instance, a small compound has been developed that can trigger the vagus nerve in rats, thereby reducing inflammation8. Looking to the future, it would be interesting to stimulate the vagus nerve electrically in people — as is currently done in thousands of epilepsy patients, showing that the procedure is safe and feasible — and to study the effect on inflammation. More specifically, the new findings suggest that molecules that stimulate the 7 subunit would also be worth developing.
On a different note, nicotine has been found to have powerful immunosuppressive and inflammation-suppressing effects. Of course, the health risks associated with smoking are immense. Yet epidemiological studies indicate that nicotine protects against several inflammatory diseases, such as ulcerative colitis, Parkinson’s disease and even Alzheimer’s disease. It can also reduce fever and protect against otherwise lethal infection with the influenza virus9. The demonstration2 that nicotine binds to the 7 subunit on macrophages fleshes out the details of how nicotine produces such effects.
The data also make me reconsider the possibilities and molecular biology of ‘alternative’ medicine. Pavlovian-type conditioning, hypnosis and meditation are well known (since the beginning of the twentieth century in some cases) to reduce inflammation10. It might be worth finding out whether these effects, as well as the reported beneficial effects of prayer and acupuncture on inflammation (the last of which is known to depend on acetylcholine)11, 12, are mediated by the vagus nerve and the 7 subunit.
|1.||Stone, R. Science 64, 365-367 (1994).|
|2.||Wang, H. et al. Nature421, 384-388 (2003); advance online publication, 22 December 2002 (doi: 10.1038/nature01339).|
|3.||Vassalli, P. Annu. Rev. Immunol. 10, 411-452 (1992). | Article ||
|4.||Wielockx, B. et al. Nature Med. 7, 1202-1208 (2001). | Article ||
|5.||Feldmann, M. Nature Rev. Immunol. 2, 364-371 (2002). | Article ||
|6.||Tracey, K. J. et al. Science 234, 470-474 (1986).|
|7.||Borovikova, L. V. Nature 405, 458-462 (2000). | Article ||
|8.||Bernik, T. R. et al. J. Exp. Med. 195, 781-788 (2002). | Article ||
|9.||Sopori, M. Nature Rev. Immunol. 2, 372-377 (2002). | Article ||
|10.||Talley, N. J. & Spiller, R. Lancet 360, 555-564 (2002). | Article ||
|11.||Son, Y. S. et al. Neurosci. Lett. 319, 45-48 (2002). | Article ||
|12.||King, D. E., Mainous, A. G.III, Steyer, T. E. & Pearson, W. Int. J. Psychiatry Med. 31, 415-425 (2001).|