TRENDS in Neurosciences Vol.25 No.12 December 2002 626 Review http://tins.trends.com 0166-2236/02/$ – see front matter © 2002 Elsevier Science Ltd. All rights reserved. PII: S0166-2236(02)02261-0 Review Julian F.R. Paton* Sergey Kasparov Dept of Physiology, School of Medical Sciences, University of Bristol, Bristol, UK BS8 1TD. *e-mail: julian.f.r.paton@ bris.ac.uk David J. Paterson University Laboratory of Physiology, University of Oxford, Parks Road, Oxford, UK OX1 3PT. Historically, angina has been managed clinically by administering vasodilating drugs (e.g. nitroglycerin) that mediate their effects by releasing nitric oxide (NO). The action of NO in this case has traditionally been explained by its effect upon vascular smooth muscle, into which it can freely diffuse. In addition, NO causes an increase in heart rate, which was believed to originate primarily from the unloading of baroreceptors in response to the decreased arterial pressure. However, it has recently been proposed that some of the effects of nitroglycerin might be, in part, due to actions within the brainstem [1] and directly on the heart itself [2–8]. As NO is well recognized as an endogenous neurotransmitter, neuromodulator and intercellular messenger, it has become clear that endogenous NO might be implicated in control of heart rate, acting at multiple sites (including visceral afferents [9], brainstem neurons that mediate cardiovascular reflexes [10–12] and cardiac autonomic ganglia [13]). Because NO spreads very rapidly and freely from its source in nervous tissue, the question arises of how any sense can be made from this diffusible signal, which penetrates through any membrane and affects cells indiscriminately within a circumscribed area. To achieve some site-targeted action within the bloodstream, NO becomes trapped by erythrocytes and is released only in regions of low oxygen tension, to aid blood flow [14]. However, it is not completely clear whether NO has comparable site-targeted effects within nervous tissue, such as the brainstem or cardiac ganglia, and, if so, what mechanisms might underlie specificity of its action. To analyse this issue, the role of NO as a signalling molecule in the autonomic control of cardiac rate must be examined. Reflex control of cardiac rate depends on afferent input from various types of visceral afferent, with baroreceptors playing the most important role. It has been demonstrated that NO can affect peripheral afferent excitability [9,15] but there are two other sites for NO modulation: the nucleus tractus solitarii (NTS; the brainstem termination site for baroreceptor afferents) and cardiac autonomic efferents at the level of the heart. At both these sites, there is extensive anatomical evidence to support sophisticated NO-mediated interactions [16–18]. Multiple actions of NO: questioning specificity Many studies relate the effects of NO to its ability to modulate (usually to increase) release of transmitters, such as glutamate, GABA and ACh [19–21]. If NO plays a physiological role in regulation of a neuronal circuit, a generalized action of increasing release of multiple transmitters (e.g. glutamate and GABA) simultaneously would make little sense. But does this really happen? There might be several ways to achieve a specific action. First, close spatial proximity of the source of NO and its target (e.g. Ref. [22]) could allow discrete actions. Although recent data indicate that, in a brain slice, NO affects targets >150 μm from its source [23], this distance might be much less in the brain in vivo and could depend on the ratio between efficacy of NO production and elimination. This raises the second possibility: that the NO synthase (NOS) isoform involved might matter. In normal circumstances, NO in nervous tissue is manufactured by two enzymes – neuronal NOS (nNOS or NOS-I) and endothelial NOS (eNOS or NOS-III) – and presumably these enzymes have very different spatial distributions. In nervous tissue, nNOS is probably confined to nerve cells, but it is unclear whether eNOS is expressed in both neurons and vasculature [24,25] or confined exclusively to blood vessels [26]. Moreover, these enzymes have very different potency in terms of NO production. In peripheral tissues, eNOS produces relatively small amounts of NO, whereas nNOS can produce NO in large quantities (reviewed in Ref. [27]). Hence, activation of nNOS might lead to more diffuse signalling than activation of eNOS. A third factor is the availability and sensitivity of the cellular biochemical targets for NO (reviewed in Ref. [28]) within its diffusion range. For example, the biochemical machinery that regulates release of GABA could be more sensitive to physiological levels of NO than that regulating release of glutamate. It has been demonstrated that the best understood Nitric oxide and autonomic control of heart rate:a question of specificity Julian F.R. Paton, Sergey Kasparov and David J. Paterson Despite its highly diffusible nature, the gaseous signalling molecule nitric oxide (NO) can exert specific effects within the CNS and PNS. To date, the specificity of the actions of NO remains an unsolved puzzle. There are several plausible mechanisms that might account for this specificity in the context of autonomic regulation of heart rate. NO acts at distinct levels w ithin the autonomic nervous system to control cardiac rate, with opposing effects at different sites. We discuss factors that might contribute to this diversity of action, and conclude that the isoform of enzyme involved in producing NO, the spatial proximity of the NO source to the target, and differences in the intracellular coupling within the target cell are all crucial for encoding the functional action of NO.