HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) Supplementary Reference List All Versions of this Article: 93/6/725    most recent expphysiol.2007.041814v1 Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Google Scholar Articles by Kasparov, S. Articles by Teschemacher, A. G. PubMed PubMed Citation Articles by Kasparov, S. Articles by Teschemacher, A. G. Related Collections Cardiovascular control Review Articles

¹ Department of Physiology and Pharmacology, University of Bristol, Bristol BS8 1TD, UK

	Abstract

Top Abstract Introduction Links between central NA... Central NA signalling and... New insights into regulation... How do we take... References

 
Numerous studies, some of which date back more than three decades, have established a link between disorders of the cardiovascular system and the catecholaminergic system of the brain. Central noradrenergic (and putative adrenergic) neurones are involved in numerous brain functions, and there appears to be more than one mechanism via which a dysfunction of central nor/adrenergic signalling may be detrimental to the cardiovascular system. Moreover, in some cases, such as essential hypertension, altered noradrenergic transmission could play a causative role. Numerous controversies are evident throughout the literature, which are very difficult to explain without much better understanding of the basic physiology of central noradrenergic transmission. Recently, using a combination of novel molecular, electrochemical and imaging techniques, we have started to unravel how noradrenergic neurones in the brain store and release their transmitter. Targeted long-term modulation of specific noradrenergic cell groups in defined brain areas using viral gene transfer is helping to clarify the links between central catecholamines and cardiovascular control in health and disease. These studies may reveal new therapeutic strategies for various cardiovascular diseases which are accompanied by heightened sympathetic nerve activity.

(Received 18 December 2007; accepted after revision 29 February 2008; first published online 7 March 2008)
Corresponding author A. G. Teschemacher: Department of Physiology and Pharmacology, University of Bristol, Bristol BS8 1TD, UK. Email: Anja.Teschemacher{at}bristol.ac.uk

	Introduction

Top Abstract Introduction Links between central NA... Central NA signalling and... New insights into regulation... How do we take... References

 
In the course of more than three decades, the question of how central noradrenaline (NA) and adrenaline (ADR) transmission impacts on cardiovascular function and dysfunction has been addressed by many research groups using a wide variety of experimental approaches. These studies reveal a highly complex organization and regulation of the central catecholaminergic system, which interacts with a multitude of the areas of the brain involved in autonomic control of the cardiovascular system (CVS). While drugs which interact with the NA and ADR receptors (adrenoceptors) are widely used in CVS medicine, their mechanisms of action are still subject to much debate. A number of recent studies suggest additional scope for pharmacotherapy of CVS disease via modulation of central catecholaminergic transmission in more selective ways.

This review examines several decades of research efforts which have shaped the current understanding of the role of catecholaminergic transmission in CVS control. The catecholaminergic neurones involved with central CVS control are predominantly noradrenergic (NAergic) but a subset of them are also equipped for downstream synthesis of ADR. Hence we will be addressing NAergic transmission and pointing out potential involvement of ADR where applicable. We also acknowledge the fact that ‘catecholaminergic’ neurones almost certainly release other signalling molecules in addition to NA or ADR, such as ATP, peptides or possibly even glutamate.

	Links between central NA transmission and cardiovascular diseases

Top Abstract Introduction Links between central NA... Central NA signalling and... New insights into regulation... How do we take... References

 
The NAergic system is recruited and activated as a part of the central nervous system (CNS) response to stress (S108; please see Supplementary material for additional references numbered as S). This could be one fundamental reason why it is affected by most cardiovascular pathologies, including myocardial infarction, cardiac failure and hypertension (both primary and secondary), and related conditions such as coronary heart disease. This primary central NAergic response, most probably, pursues physiological, homeostatic purposes initially, but when it becomes persistent its role is thought to be largely maladaptive and detrimental.

Numerous studies, some of which were performed more than three decades ago, attempted to link central NA transmission with various physiological and pathophysiological mechanisms which could be either directly damaging to the CVS or detrimental when the CVS is already under strain of a disease. At least three such mechanisms are evident from the literature, namely, regulation of sympathetic nervous system activity, interference with water and salt homeostasis, and modulation of cardiovascular reflexes, which also feed into the control of sympathetic outflow.

Central NA transmission, sympathetic nervous system and CVS disease

In humans, central sympathetic NA spillover from subcorcital but not cortical areas is clearly correlated with the activity of the peripheral muscle sympathetic nerves, indicating that sympathetic outflow may be directly influenced by the activity of the brainstem NAergic neurones (Lambert et al. 1997). This strategic position of central NA transmission has extensive pathophysiological implications.

The role of the sympathetic system in hypertension is indisputable. The link between central NAergic activity, sympathetic output and blood pressure has been documented by dozens of human and animal studies (for example: Isaac, 1980; Patel et al. 1981; Reis et al. 1984b; Elghozi et al. 1989; Head, 1991; Woo et al. 1993; Esler et al. 1995, 2006; Oparil et al. 1996; Zoccal et al. 2007; S75, S165). Some of these studies will be analysed in more detail below, but hypertension is not the only CVS pathology in which central NA transmission has been implicated.

Heightened central sympathetic drive has been demonstrated to play a causative role in triggering episodes of angina pectoris. For example, in angina patients, a reduction of the central autonomic outflow with clonidine, a centrally acting ₂-adrenoceptor agonist, dramatically decreased the incidence of anginal attacks (it is acknowledged that clonidine is also a ligand at the imidazoline binding sites but at present this aspect of its activity is not well understood). This was detectable in over 60% of borderline hypertensive and also in normotensive patients (S19). Similar observations were made by Thomas et al. (1986) and Cocco et al. (1979). Predictably, the effectiveness of clonidine in angina patients correlates with the initial level of sympathetic activity as measured by urinary excretion of NA (Zaleska et al. 1982). Repeated preoperative administration of clonidine in patients at risk of coronary artery disease can significantly reduce the incidence of perioperative myocardial ischaemia and postoperative death, underscoring the benefits of normalizing sympathetic outflow in angina (S170).

Cardiac arrhythmia is yet another manifestation of an elevated sympathetic drive in some patients (Scardi et al. 1993; Esler & Kaye, 2000; S147). This is particularly evident in patients with anxiety disorders which are associated with an increase in mortality from CVS diseases. These patients frequently show signs of a decreased cardiac vagal function with prevalent and increased sympathetic drive. Pharmacological evidence pointed to central NAergic hyperactivity as a possible link between increased sympathetic tone and heightened anxiety levels. Anxiety patients with panic disorder became more anxious after administration of the ₂-adrenoceptor blocker yohimbine, which also increased the signs of sympathetic drive to the heart. The opposite was observed with clonidine. These changes were not seen in healthy control subjects (S176). The anti-arrythmic β-adrenoceptor blocker propranolol is also used in anti-anxiety treatment. Some studies suggested that propranolol has a central component in its anti-arrhythmic effect (S121).

In heart failure, which often results from extensive myocardial infarctions, the level of sympathetic drive to the heart is a major negative prognostic factor with a clear impact on mortality (reviewed by Esler & Kaye, 2000). In patients with heart failure, increased NA turnover from subcortical areas of the brain (measured from the spilover of NA metabolites) has been demonstrated directly, clearly implicating central NA in the pathological sympathoexcitation characteristic of this condition (Aggarwal et al. 2002). Use of β-blockers is one of the currently preferred strategies for treatment of cardiac failure. Remarkably, not all β-blockers are equally effective for treatment of the failing heart (Neil-Dwyer et al. 1981; S93, S119, S125). Interestingly, lipophilic compounds which are able to cross the blood–brain barrier perform better (Neil-Dwyer et al. 1981). Recent clinical data indicate that in patients with severe heart failure, β-blockers greatly reduce the severity of sleep apnoeas, a clearly beneficial central therapeutic effect, which is likely to help to reduce sympathetic hyperactivity (Kohnlein & Welte, 2007).

In summary, high and detrimental sympathetic outflow from the brain has been linked to the central NA system and is a commonality in a number of the most prevalent disorders of the CVS.

Central NA and the regulation of water and salt balance

It has long been known that NA regulates vasopressin release at the level of the paraventricular nucleus (PVN) and that application of NA in the PVN increases blood pressure (Camacho et al. 1981; Lightman et al. 1984; Benetos et al. 1986; S28). In our studies, ‘silencing’ of A2 neurones, a significant cluster of noradrenergic cells located within the nucleus of the solitary tract (NTS) had an acute negative effect on water consumption in Wistar rats, but not in spontaneously hypertensive rats (SHR; Duale et al. 2007). There are also indications that NA may regulate salt appetite and salt consumption induced by angiotensin II (S34, S41). Microinjections of NA into the median preoptic nucleus increased urine outflow, sodium excretion and blood pressure in rats (S34). Interestingly, hypertonic saline (but not an equivalent osmotic stimulus) applied to the NTS, which harbours the NAergic A2 cell group, caused massive sympathetically mediated hypertension in conscious rats (Vlahakos & Gavras, 1988). In contrast, in anaesthetized rats a hypotensive effect was found after microinjections of hypertonic saline into the medial subnucleus of the NTS (S70). Thus, the effect of hypertonicity seems to be dependent on anaesthesia and/or region. More studies are needed to clarify the specific role of NAergic neurones in water and salt homeostasis. However, it is clear that in the case of central NA dysfunction, inadequate regulation of water and salt balance may have a negative impact on the CVS and contribute to the progression of disease.

Central NA and control of various reflexes involved in CVS regulation

Various central autonomic reflexes, such as the baroreceptor-mediated vagal and sympathetic reflex or the peripheral chemoreceptor reflex, which can lead to sympathetic activation or deactivation, play an important role in CVS homeostasis. Blunted cardiac baroreflexes in patients with essential hypertension were documented long ago (Bristow et al. 1969; Grassi et al. 1998). Interestingly, clonidine greatly enhanced baroreflex-mediated bradycardia, which points towards a role for central NA in regulation of this reflex in both rodent models and human patients (Tank et al. 2004; S159). The data regarding the influence of central NA on CVS reflexes are somewhat inconsistent between different NAergic areas, which is further discussed below. Some authors found that chemical disruption or ‘genetic silencing’ of the A2 neurones in the NTS increased blood pressure lability (Snyder et al. 1978; Duale et al. 2007), while others reported a decrease (S78). Interestingly, microinjection of NA in NTS also affected cardiovascular reflex changes triggered by chemoreceptor activation (Silva de Oliveira et al. 2007) and this might mean that some central NAergic neurones may affect sympathetic outflow via modulation of pathways other than those involved in the baroreceptor reflex. The NTS is not the only site where NA may affect cardiovascular reflexes, since the hypothalamus (Oparil et al. 1996; Hwang et al. 1998; S146) and the rostral ventrolateral medulla (RVLM; Reis et al. 1984b; S95) have also been documented as crucial locations. Thus, it is possible that the dysregulation of homeostatic reflexes characteristic of many CVS diseases is to some degree a result of changes in central NA/ADR transmission.

Other possible links between the central NAergic system and CVS disease

In addition to the above-mentioned mechanisms, there are a number of less direct but possibly not less important mechanisms which could underpin an association between changes in central NA signalling and CVS diseases.

One interesting and almost completely unexplored role of central NA which might have direct relevance to hypertension is the regulation of the cerebral microcirculation. As early as 1975, it was demonstrated that stimulation of the NA-containing cell bodies within the locus coeruleus leads to a prompt reduction in cerebral blood flow and an increase in brain vascular permeability to water (Raichle et al. 1975). Taking into account that brain microvessels have abundant NAergic innervation (S84), it might be hypothesized that if NAergic innervation is affected, the autoregulatory capacity of the brain microcirculation may be compromised. We have recently demonstrated that the lumen of main brain arteries in young, prehypertensive SHR is clearly reduced compared with normotensive Wistar–Kyoto (WKY) rats (S123). This could bring the circulatory system of the brain close to the edge of its autoregulatory capacity. If coupled with a dysfunction of NA-mediated control of brain microcirculation, it could lead to inadequate oxygenation of the brain structures controlling sympathetic outflow and result in Cushing-like sympathetic responses.

Another relevant function of the central NA system is the control of emotional reactivity and anxiety (S108). It is clear that disproportional levels of anxiety may lead to activation of the hypothalamo–pituitary–adrenal axis, followed by a plethora of effects on the CVS, most of which are bound to be detrimental for patients already suffering from a CVS disease. It has been shown directly that in middle-aged women, higher urinary NA excretion correlates with higher levels of anxiety (S75), presumably reflecting a high level of sympathetic system activity in anxious, stressed and depressed but otherwise healthy individuals. Panic disorder commonly associates with hypertension. Although the gross NA outflow into the jugular blood between panic episodes is apparently normal (Esler et al. 2006), NA transporter deficiencies in panic disorder patients have been documented (Alvarenga et al. 2006). There is also clear evidence for sympathetic hyperactivation in panic disorder patients suffering from angina (S47). Interestingly, NA microinjections into the amygdala, one of the structures involved in emotional responses, antagonize glutamate-induced decreases in heart rate and blood pressure (S136). Thus, at the level of amygdala, high release of NA might interfere with an as yet unknown antihypertensive glutamatergic mechanism.

Yet another potentially important link between central NA and CVS disease may be via the well-documented role of central NA in control of appetite (S3, S14, S139). Abnormality in central NA signalling might therefore ultimately underpin the known association between obesity and hypertension (S4, S11, S23, S53, S171). In other words, it is not impossible that an altered central NA transmission may be one of the factors contributing to the current epidemics of obesity and atherosclerosis and, via that route, indirectly promoting CVS pathology. In our experiments, ‘genetic silencing’ of the A2 neurones in rats led to a significant increase in weight gain over time (Duale et al. 2007). These findings are consistent with data showing that destruction of the local NA innervation in the NTS using a saporin-conjugated dopamine-β-hydroxylase (DBH) antibody (DBH–saporin) attenuates the appetite-suppressing effect of cholecystokinin (S134). Noradrenergic effects on food consumption are best documented at the level of the hypothalamus, which receives a large portion of its NAergic axons from the A2 group (Sawchenko & Swanson, 1982; S25, S26). Future studies should clarify whether a defect in this particular role of A2 neurones has a role in obesity and its numerous complications highly relevant to CVS disease.

Thus, there is certainly more than one mechanism and more than one area of the brain where altered activity of the NA system may translate into a mechanism detrimental to the CVS. In some cardiovascular pathologies, such as myocardial infarction-induced heart failure, changes in central NA signalling are probably triggered in a reactive manner. In contrast, in essential hypertension central NA imbalance occurs at very early stages, when it should be rather seen as a possible causative factor. This can be exemplified by the SHR, in which changes in brain NA transmission are detectable in vitro, in early postnatal tissue, long before hypertension develops in this model (Veerasingham et al. 2005; S174). The causative role of central NA in SHR hypertension is supported by studies which show that depletion of central NA in young prehypertensive SHR prevents the development of hypertension (Haeusler et al. 1972; Erinoff et al. 1975).

	Central NA signalling and control of blood pressure

Top Abstract Introduction Links between central NA... Central NA signalling and... New insights into regulation... How do we take... References

 
Methodological considerations

The studies cited above link central NAergic transmission with various CVS diseases. However, the role of NA and ADR in CVS regulation and pathophysiology is most probably different, depending on the site of their release within the CNS. Numerous studies attempted to reveal these links using microinjections of NA or its agonists or antagonists into selected brain nuclei. However, interpretation of such experiments is not straightforward because such drugs almost invariably affect NAergic neurones themselves. Thus, it is usually very difficult to decide whether the effect of a microinjected adrenoceptor ligand is due to an action on the NAergic neurones (or their axons), or due to modulation of non-NAergic neurones present in the targeted structure. Another method commonly used to assess the role of NAergic neurones in vivo is the destruction of NAergic cells using 6-hydroxydopamine (6-OHDA) or saporin conjugates. This, however, triggers inevitable reorganization of local networks and reactive changes which are hard to foresee. Moreover, removal of NA (or ADR) from both the medullary nucleus, which harbours NAergic cell bodies, and their target locations might lead to mutually opposing effects on sympathetic outflow and blood pressure. Perhaps even worse, this approach inevitably leads to retrograde destruction of NAergic neurones in the nuclei which project into the injected area. Thus, injections of such toxins, for example in the NTS, may not discriminate between the effect on A2 neurones from the effects of retrograde destruction of cells in areas A1, C1 and other areas which send their axons to or through the NTS. These considerations should be taken into account when analysing the studies cited below.

General evidence for a role of central NA/ADR in hypertension

Given an enormous medical and social burden of hypertension, better understanding of its pathophysiological mechanisms is pertinent. Below we review some of the studies which attempted to reveal the role of central NA in this widespread CVS disease in more detail.

Early studies revealed that in patients with essential hypertension, NA concentrations in the cerebrospinal fluid were almost twice as high as in normotensive subjects (Cubeddu & Hoffman, 1987). Moreover, as well as decreasing blood pressure, clonidine reduced this elevated central NA outflow (Cubeddu et al. 1984), an effect which was ascribed to the inhibition of NA release by activation of ‘presynaptic’ ₂-adrenoceptors. Many studies used SHR as a model of essential hypertension. In the SHR, in accordance with the human data, prolonged treatment with clonidine decreased blood pressure and NA turnover as estimated from levels of its metabolites in renal cortex and medulla. Withdrawal of clonidine, conversely, increased NA turnover (S9). In contrast, long-term depletion of central NA using the catecholamine-selective neurotoxin 6-OHDA prevented the development of hypertension in SHR (Haeusler et al. 1972; Erinoff et al. 1975; S40). Interestingly, after widespread destruction of NAergic neurones by 6-OHDA in normotensive rabbits, resting mean arterial pressure slightly increased (by 14%), while baroreflex gain and heart rate did not change (Elghozi et al. 1989).

It may be asked whether the effects of clonidine (S9) are due to its inhibitory effect on NAergic neurones themselves or on the cells which are the recipients of NA signalling in the brain. It turned out that clonidine was still effective after destruction of the ascending NA projections, while -methyldopa (which depends on NAergic neurones for its conversion in -methyl-NA) was not (S164). Essentially the same was found for C1 neurones where, in anaesthetized rats, DBH–saporin pretreatment did not prevent the hypotensive and sympatholytic effect of systemically administered clonidine, suggesting that clonidine reduced sympathetic outflow by a combination of effects on multiple cell types both within and outside the RVLM (Schreihofer & Guyenet, 2000). This is consistent with the report that in 6-OHDA-lesioned rabbits the hypotension produced by intracerebroventricular clonidine was not affected (Elghozi et al. 1989). Thus, since hypotensive effects of ₂-agonists can still be demonstrated after NA depletion, it is logical to conclude that these drugs do not act by blocking NAergic neurone activity or NA release. Rather, they mimic the endogenous NA effect on some target neurones (S66). In has to be acknowledged that this conclusion is not shared by all authors (Madden & Sved, 2003). Experimental and clinical data also indicate that clonidine increases baroreflex control of blood pressure (Hayashi et al. 1993; Tank et al. 2004; S120, S159) although it still is not clear at what level of the CNS that effect takes place.

Which brain areas are involved?

The A2 cell group in the NTS.  Most of the available literature argues that the activity of A2 neurones is ‘hypotensive’, i.e. that the A2 neurones are part of an antihypertensive homeostatic mechanism. One of the first observations which linked A2 noradrenergic neurones with hypertension was published in 1978 by D. J. Reis's group (Snyder et al. 1978). It reported that injections of high doses of 6-OHDA into the NTS led to severe hypertension in rats, while low doses led to chronic lability of pressure. In follow-up studies, it was found that the A2 lesions attenuated the bradycardic responses to phenylephrine stimulation, implying the importance of A2 neurones in the baroreceptor reflex (Snyder et al. 1978; Talman et al. 1980c). It also became apparent that the hypertension triggered by large doses of 6-OHDA was probably due to unspecific cell damage. While it was later suggested that blood pressure lability due to A2 lesions may not translate into hypertension, the effects 6-OHDA on blood pressure were clearly proportional to the damage to A2 neurones (Talman et al. 1980a,b).

Release of NA in the NTS in response to a blood pressure increase or stimulation of the aortic nerve (which contains baroreceptor afferents) was demonstrated directly using microdialysis (Kobilansky et al. 1988; S146) This is consistent with the idea that the A2 cells are a component of a circuit involved in the reflex control of arterial pressure. Indeed, dose-dependent depressor and bradycardic responses in anaesthetized rats were registered in response to microinjections of L-threo-dihydroxy-phenylserine, a synthetic precursor of noradrenaline, into the medial NTS (S103). Also, a dose-dependent decrease of blood pressure and heart rate was produced in anaesthetized rats by NTS microinjections of a range of adrenergic agonists, with ADR being the most effective drug, followed by NA, dopamine, -methyl-NA and octopamine (Zandberg et al. 1979). Similarly, a vasodepressor response to NA microinjected into NTS was found (S117). Surprisingly, microinjections in conscious animals seemed to demonstrate the opposite effects. For example, microinjection of NA or clonidine into the NTS in conscious rats produced an immediate, sharp increase in blood pressure and decrease in heart rate. However, light ether, nembutal or urethane anaesthesia drastically reduced, abolished or even reversed the pressor effect of local microinjection of the same substance (Vlahakos et al. 1985). In a recent study by B. Machado's group, NA microinjections in conscious rats were also hypertensive. In that study, NA significantly attenuated a pressor response to chemoreceptor stimulation, which indicates that the effect of NA on blood pressure control does not necessarily need to be via modulation of the baroreceptor pathway (Silva de Oliveira et al. 2007). It is not clear at this point why the role of A2 neurones in control of blood pressure may be so critically dependent on anaesthesia. Interestingly, projections from the locus coeruleus into the NTS could contribute to NA-mediated inhibition of the baroreceptor reflex (Chan et al. 1992).

The physiological roles of NA released within the NTS are not yet fully explained. On the one hand, studies such as those discussed above argue that it could be part of a mechanism which helps to offset blood pressure increases. On the other hand, microinjections of NA into the NTS also led to blood pressure increases in some (Vlahakos et al. 1985; Silva de Oliveira et al. 2007) but not other studies (S88). As already mentioned, interpretation of such microinjection studies is controversial because it is almost certain that they lead to silencing of the A2 neurones via powerful ₂-adrenoreceptor-mediated negative feedback and, at the same time, activate targets of NA transmission in the NTS. We used viral vectors to express an inwardly rectifying potassium channel in A2 neurones and monitored the cardiovascular system by radiotelemetry. This additional potassium conductance hyperpolarizes the membranes of neurones with the aim of ‘silencing’ the A2 neurones. We observed a gradual rise and increased lability of blood pressure, presumably because the ‘restraining’ effect of these neurones on homeostatic reflexes controlling blood pressure was reduced (Duale et al. 2007). In contrast to pharmacological treatment, in these experiments the targets of intervention were known, although we cannot at this point discriminate whether the changes we observed were due to the alterations in NA release within the NTS or in some frontal structures which receive NA projections from the A2 cell group.

The C1 cell group in the RVLM.  The C1 neurones in the RVLM contain phenyletholamine-N-methyl-transferase and are, therefore, thought to contain ADR. More than 25 years ago, it was postulated that the RVLM provides tonic stimulatory input to the spinal sympathetic preganglionic neurones within the intermediolateral column of the spinal cord (Dampney, 1981; Reis et al. 1984b; S37). Indeed, many of the C1 neurones project to the spinal cord (Goodchild et al. 1984). Interestingly, while NA/ADRergic terminals are associated with most preganglionic sympathetic neurones in the intermediolateral column, the varicosities tend not to form any tight oppositions with them (S7). This arrangement is consistent with the ‘volume’ mode of transmission operated by the C1 (and other catecholaminergic) neurones, whereby the signal is not delivered to any well-specified cellular target but modulates the activity of the target area diffusely and possibly via some intermediates, such as local glia. Using a juxtacellular labelling technique in anaesthetized rats, Schreihofer & Guyenet (1997) demonstrated that the majority (about two/thirds) of barosensitive neurones in RVLM with axons projecting towards the spinal cord were positive for phenyletholamine-N-methyl-transferase and thus, by definition, were of the C1 cells. Focal electrical or chemical stimulation (by glutamate or bicuculline injection) in the RVLM of the rat and rabbit evoked large increases in blood pressure when the stimulus sites were in the region containing a high density of C1 neurones, but much smaller or no responses were obtained when the sites were outside this region (Chalmers et al. 1981; Reis et al. 1984a,b; Smith & Barron, 1990; S37). Consistent with these results, electrolytic lesions in the RVLM or local chemical ‘silencing’ using GABA or tetrodotoxin injection led to a collapse of arterial pressure to a level comparable to that following spinal cord transection (Reis et al. 1984a,b). At the spinal level, ADR injected into the intermediolateral column caused an increase in heart rate mediated by ₁-adrenoceptors, presumably via sympatho-activation, when small doses were used but bradycardia via ₂-adrenoceptors with larger doses (Malhotra et al. 1993b). In summary, the case for the critical role of C1 cells in driving the sympathetic responses via their descending (and also ascending) projection is very strong.

However, not all available evidence agrees with the idea that C1 neurones are paramount to activity of spinal sympathetic preganglionic neurones. In normotensive rats, electrical stimulation in the C1-containing area of the RVLM increased arterial pressure, but the pressor responses were unaffected by intrathecal injections of the -adrenoceptor blocker phentolamine at doses which could block locally applied adrenergic agonists. This observation put into question the importance of spinal adrenergic receptors in mediating the effects of RVLM stimulation (Mills et al. 1988). In another study, pressor responses elicited by unilateral glutamate stimulation of the medullary pressor area (the location of the C1 cell group) were not blocked by bilateral spinal (T8–T10) injections of the ₁-blocker prazosin at a dose that blocked effects of spinally applied ADR. However, the pressor responses could be blocked by spinal microinjections of 2-amino-7-phosphonoheptanoic acid (AP-7; an NMDA receptor blocker), but, surprisingly, not 6,7-dinitro-quinoxaline-2,3-dione (DNQX; a non-NMDA receptor blocker; Malhotra et al. 1993a). Taken together, these observations rather argue for a role of descending glutamatergic but not ADRergic projections, although the lack of effect of DNQX is extremely hard to explain. One possible reason for these controversial results is that glutamate is likely to trigger a surge of NA release within the RVLM and a rapid shutdown of C1 activity via the negative ₂-receptor-mediated feedback. At the same time, more persistent action potential activity might be needed to trigger ADR release from varicosities located far away from the C1 somata. It is also possible that ADR/NA release in the spinal cord will only become evident when an increased frequency of action potentials is coupled with some permissive conditions at the site of release, which sensitize the local transmitter release machinery. Finally, it needs to be remembered that the above-mentioned experiments were performed under anaesthesia, which is notorious for its effects on NA transmission.

Other observations added additional doubts. Using DBH–saporin as a presumably more selective lesioning approach, it was reported that destruction of C1 neurones in anaesthetized normotensive rats had little effect on the maintenance of blood pressure but interfered with the depressor effect of clonidine microinjected in that area (Madden & Sved, 2003). Recent studies confirmed that a widespread (>80%) destruction of RVLM C1 neurones with DBH–saporin led to only a modest hypotension in normotensive Sprague–Dawley rats (S95). After DBH–saporin treatment, intravenous clonidine caused a normal degree of sympatho-inhibition, and RVLM inhibition with the GABA_A receptor agonist muscimol caused hypotension and sympatho-inhibition similar to that in intact rats (Guyenet et al. 2001). Furthermore, only few C1 neurones exhibited c-Fos expression in response to baroreceptor activation, which may be interpreted as an argument against their role in this reflex pathway (Li & Dampney, 1992). In contrast, many tyrosine hydroxylase-positive cells in that area expressed c-Fos after a prolonged hypotension, possibly suggesting that they are specifically activated by hypotension (S89), which would be consistent the data from Madden & Sved (2003). Some studies even questioned the neurochemical phenotype of C1 cells. Thus, the plasma membrane NA transporter was found in over 95% of A1, A2, A5, A6 and A7 noradrenergic groups. However, according to some sources, only 10% of C1 RVLM cells expressed this transporter (Comer et al. 1998). It was later proposed that C1 neurones utilize a transporter that is distinct from the other bioactive amines, or may not require rapid reuptake mechanisms to function (Lorang et al. 1994). Until the expression of a vesicular monoamine transporter in C1 neurones has been investigated, the question of their neurochemical phenotype will remain open. Furthermore, some studies suggest that C1 neurones could release glutamate (for additional discussion see Dampney, 1994).

Thus, the nature of the contribution of C1 neurones to blood pressure control and cardiovascular reflexes is still unclear, not only in the context of hypertension, but also in cardiac failure, anxiety-related angina and arrhythmia episodes. However, it is impossible to dismiss the extensive evidence which argues for an important role of these neurones in control of sympathetic outflow. Further studies in awake animals using novel molecular approaches will hopefully resolve the above-mentioned controversies.

The A1 cell group in the caudal rostral ventrolateral medulla (CVLM).  The role of A1 neurones in CVS homeostasis has not been extensively investigated. The activity of A1 neurones seems to play, overall, an ‘antihypertensive’ role because destruction of these neurones causes hypertension mediated by sympathetic activation and increased vasopressin release. In the rabbit, lesions of the CVLM coinciding with the location of A1 cell bodies produced prominent increases in the plasma levels of vasopressin and adrenaline, a twofold increase in plasma NA and a substantial increase in plasma renin activity (S44). Such lesions led to hypertension and bradycardia (Elliott et al. 1985). In rats, kainic acid-induced lesions in the CVLM also trigger a prohypertensive response, an effect which could be linked to CVLM projections into the RVLM. Initially, it was suggested that the A1 neurones have a tonic inhibitory influence on the RVLM (Granata et al. 1986) but later it became apparent that their main target is the hypothalamus (Cunningham & Sawchenko, 1988).

The locus coeruleus.  The A6 group, also known as locus coeruleus (LC), is the most prominent pontine cluster of NA-producing neurones. The LC has been implicated in a vast range of central functions, many of which seem to be related to the regulation of state of vigilance, anxiety, cognition and stress response (reviewed by Fillenz, 1990). This makes it rather difficult to dissect LC contributions to CVS control specifically because there are too many indirect ways in which it may be affected. Electrical stimulation of LC in normotensive rats (S82) and rabbits (S22) produced a pressor response. However, unilateral glutamate injection in LC decreased arterial pressure in anaesthetized rats (S104, S154), leading to the suggestion that the effects evoked by electrical stimulation are not due to activation of A6 neurones but rather of en passant fibres. It has to be noted, however, that injection of glutamate leads to a very short upstroke of neuronal activity because of the rapid uptake of glutamate and its conversion into GABA. At the same time, it is possible that only sustained high-frequency activity is able to deliver sufficient excitatory signal to remote varicosities of NA neurones (Fig. 1). If that is the case, then the results obtained using these two modes of stimulation effectively cannot be compared. Hypotension increased outflow of NA from the LC in conscious rats, but so did stress induced by noise or air jet (Singewald et al. 1999), supporting the idea that NA release in LC may play a more general role than only control of sympathetic outflow and blood pressure. In line with this is the proposed role of LC (and other NAergic groups, including A2) in responses to fearful stimuli (such as inhalation of CO₂) in humans, which lead to an increase in blood pressure and activation of the pituitary–adrenocortical axis (Bailey et al. 2003).

View larger version (27K): [in this window] [in a new window]   Figure 1.  Microamperometry reveals different vesicular populations involved in transmitter release from soma and axonal varicosities of NAergic neurones Fine carbon fibre electrodes (CFE) can be used to detect quantal release of NA (amperometric current; Iamp) from NAergic neurones genetically targeted to express enhanced green fluorescent protein (EGFP). These experiments have revealed that release occurs in both frequent but fairly small quanta and relatively rare but very large quanta. They have confirmed that NA is released not only from axonal varicosities, which are traditionally believed to be the sites of release, but also from neuronal soma and dendrites (Chiti & Teschemacher, 2007). Since somata of NA neurones are concentrated in brainstem nuclei while their axons are spread throughout the whole neuraxis, the physiological roles of somatic and axonal release may be different. Parts of this figure reproduced with permission from Chiti & Teschemacher (2007).

Differences in brain NA concentration and turnover in normo- and hypertensive animal models

Brainstem.  A considerable number of studies have looked at NA transmission in SHR as a model of essential hypertension. Several approaches have been taken to characterize the activity of the system: measurements of whole tissue concentrations of NA; of its metabolites as an indicator for degradation; and of extracellular NA concentration as an indicator for release.

Takami and co-workers found that in young, mature (10 weeks) severely hypertensive stroke-prone SHR, the tissue level of NA in the medulla oblongata was significantly lower than in the control strain WKY rat (Takami et al. 1993). Similar observations were reported by others and, interestingly, the medulla oblongata was the only brain region where a decrease in NA in the SHR was significant (S143). Consistent with these reports, between 6 and 40 weeks, NA content in brainstem and thoracic spinal cord fell progressively in SHR but not in WKY (S72). Interestingly, in a later publication, these authors found elevated NA in the striatum of SHR, which might indicate that the NA synthesis and/or release can be regulated differentially between somata and distant axons in NAergic neurones (Howes et al. 1984). Most studies used tissue extracted from the whole brainstem and not from any identified NA/ADR cell-containing nucleus. Some studies, however, focused on the NTS more specifically. For example, NA content was decreased in the NTS of SHR compared with WKY rats at 4, 8 and 16 weeks of age, whereas in 8- and 16-week-old SHR, the concentration of NA was increased in the ventral part of the brainstem, which harbours the A1, A5 and C1 cell groups (Yao et al. 1989). Thus, overall, most studies suggest a reduced NA content of the SHR medulla oblongata, and some are pointing towards the A2 cell group specifically. However, given that the lower total NA content of the tissue may be a reflection of lower production as well as higher release combined with fast degradation, these data might equally suggest a ‘hypo-’ or a ‘hyperfunction’ of NAergic transmission in the brainstem of SHR, including NTS.

The turnover of NA estimated from the level of its metabolites in the NTS of 4- and 16-week-old SHR and WKY was found not to be different (Yao et al. 1989). Another group even reported a decrease in -methyl-p-tyrosine (-MPT)-induced NA disappearance in the NTS of young SHR, possibly suggesting a slower release rate (Fujino, 1984). In contrast, no significant differences in basal and electrically evoked [³H]NA release was found in medulla oblongata slices from mature SHR and WKY rats (10 weeks old; Tsuda et al. 1991). Thus, the question of whether NA signalling within the brainstem nuclei in SHR is enhanced or decreased remains open.

Hypothalamus.  The majority of studies suggest that in SHR, at the level of the hypothalamus, contrary to the NTS, NA turnover and signalling is increased compared with normotensive control rats. For example, NA levels in the PVN of SHR were significantly higher than in WKY rats (Woo et al. 1993). In line with this, NA synthesis was reported to be increased in the PVN in SHR (S110), and electrically evoked release of [³H]NA from hypothalamic slices in adult SHR was increased (Tsuda et al. 1991), as was the extracellular concentration of NA and its metabolites. Using microdialysis, the NA concentration was found to be increased by 35% in the posterolateral hypothalamus of adult SHR compared with WKY rats. This led to the conclusion that NA synthesis, release and breakdown in the adult SHR hypothalamus are enhanced (Pacak et al. 1993). More specifically, (Kawasaki et al. 1991) have demonstrated that NA release in the posterior hypothalamus (the area associated with hypertension; Robinson et al. 1983) was increased after stimulation of the LC with glutamate and that this increase was greater in SHR, suggesting that this pathway is facilitated in these rats.

Some studies supported the idea that the increased hypothalamic NA turnover is a secondary response rather than a causative factor for hypertension. For example, while in 9-week-old SHR the NA turnover in hypothalamus and brainstem had increased significantly, this was not evident in younger, prehypertensive 5-week-old SHR (S122). Furthermore, a similar rise in hypothalmic NA levels was observed in SHR and in rats with secondary hypertension triggered by aortic stenosis (Woo et al. 1993). However, in vivo microdialysis in freely moving rats showed significant increases in basal and K⁺-stimulated NA release in the PVN of younger (7- to 10-week-old) and older (12- to 14-week-old) SHR compared with WKY and Sprague–Dawley rats, although their blood pressure is still rising at that age (Qualy & Westfall, 1988). The authors suggested that enhanced central NAergic drive to the PVN may be involved in the development and maintenance of hypertension in the SHR (Qualy & Westfall, 1988), and later re-enforced their conclusions by demonstrating that deoxycorticosterone acetate (DOCA)-salt hypertensive rats reached equivalent blood pressure levels without comparable increase in NA ‘overflow’ (Qualy & Westfall, 1995).

Thus, the bulk of the literature indicates that NA signalling at the level of at least some hypothalamic nuclei is enhanced in SHR. However, conflicting evidence includes: (i) a lower tissue concentration of NA in hypothalamus of SHR, although no distinction was made between its different regions (S58); (ii) a lower DBH activity in the hypothalamus of SHR (Saavedra et al. 1978); and (iii) an opposite age-related trend in electrically stimulated NA release from hypothalamic synaptosomes, in that more NA was released from younger SHR than WKY rat synaptosomes (Hano et al. 1989). When considering the evidence in favour of enhanced NA signalling in the hypothalamus of SHR, it must be kept in mind that hypothalamic NA is released from the axons of brainstem NAergic neurones, many of which originate from the A2 group. In light of this, the previously discussed data demonstrating decreased NA concentration in the medulla oblongata of the SHR are remarkable. This yet again highlights the need for better understanding of the mechanisms which control synthesis, storage and release of NA at the level of somatic and axonal compartments of NAergic neurones.

Changes in cellular properties of brain NAergic neurones in normo- and hypertensive animal models

Regulation of NA release is a complex and multifactorial process. Evidence from neurones and peripheral NAergic model cells suggests that central NA release is not only dependent on neuronal anatomy and projection patterns but also on neuronal excitability, intracellular signalling, the mechanisms which control Ca²⁺ concentration and the organization of vesicular pools for NA storage and release.

Electrophysiology.  A number of studies have investigated neuronal activity via extracellular recordings in NA-neurone-containing brain nuclei in SHR and control rats. For example, a marginally higher firing rate of retrogradely identified RVLM neurones was recorded in a brainstem–spinal cord preparation of SHR compared with WKY control animals (Matsuura et al. 2002). However, data specifically relating to NAergic neurone activity are scarce. Thus, we were unable to find a comparative study on differences between A2, A1, A5 or C1 neurones. In the locus coeruleus, which consists predominantly of NAergic neurones projecting into hippocampus and other front brain areas, the mean spontaneous firing rate is slightly decreased in SHR compared with the normotensive control rats (Olpe et al. 1985; Miyawaki et al. 1992; S30). Interestingly, while electrical stimulation of LC produced hypertension in normotensive rats, stronger stimulation was needed to achieve an increase in blood pressure in SHR. Moreover, lower stimulation intensities resulted in slight hypotension only in SHR (S82). For C1 and A2 neurones in vitro, we found area-specific differences in spontaneous activity between Wistar rats and SHR (Teschemacher et al. 2007). It appears likely, however, that the frequency of somatic action potentials is a poor predictor for the quantity of NA release in axonal target areas, and other locally acting factors need to be taken into account (Fig. 2 ; also see subsection ‘New insights into regulation of NA release from central neurones’ below).

View larger version (32K): [in this window] [in a new window]   Figure 2.  Hypothetical balance between electrical activity and NA/ADR release in NAergic cell bodies and axonal varicosities in different physiological settings Central NA/ADR neurones may be anticipated to have a high rate of action potential propagation failure along their axons. This suggestion is based on the fact that these axons are unmyelinated, branch extensively and have numerous varicosities, all of which represent obstacles for effective action potential propagation (S38). Ai, in unstimulated organotypic slice cultures where the frequency of action potentials in NA neurones is relatively low and often is completely absent (Z. Chiti and A.G.Teschemacher, unpublished observation from this laboratory), we found a trend to higher release activity at varicosities compared with somata (Chiti & Teschemacher, 2007). This suggests that release from varicosities depends more on factors other than the action potentials. Aii, sustained action potential activity is likely to increase the rate of action potential propagation towards remote varicosities. It can also be expected to release more transmitter from the soma. B, while somatic action potentials may be generated at a given frequency, local factors acting on the varicosity directly could sensitize the release apparatus and potentiate NA release within a specific target area.

Vesicle properties.  Interestingly, NA storage vesicles may be structurally and/or biochemically altered in SHR. It was reported that storage vesicles of SHR took up 37% more [³H]NA than those of WKY rats and the speed of uptake was also significantly faster (Rho et al. 1983). At the electron microscopic level, WKY rats displayed intact, largely well-delineated dense-cored vesicles, while in SHR these vesicles appeared swollen and distorted. This raises the possibility that in SHR NA is stored and released from an altered pool of vesicles with, possibly, a different transmitter load and cotransmitter composition (Rho et al. 1983).

Autoregulatory feedback.  Some of the differences in NA release noted by various studies may result from changes in the negative ₂-receptor-mediated feedback onto neuronal activity or transmitter release. For example, B_max (the maximum specific binding) of ₂-binding sites was reduced in medulla oblongata of SHR compared with WKY animals, as early as 4 weeks after birth (S113). Consistent with this, ₂-mediated inhibition of NA release in cerebral cortex was decreased in SHR, suggesting an upregulation of NA release at axonal varicosities located distant from the neuronal cell bodies in the brainstem (Russell et al. 2000). Pacak et al. (1993) demonstrated an augmented ₂-receptor-mediated restraint of NA release in juvenile SHR as estimated from the effect of the ₂-blocker yohimbine on the NA concentration in dialysate.

Angiotensin II (Ang II), central NA transmission and hypertension

A strong connection between the vasoconstrictor hormone Ang II and NA signalling is well established in peripheral blood pressure regulation. Angiotensin II may influence CNS processes via several routes, namely: (i) from the periphery by passing the permissive blood–brain barrier at structures such as the area postrema; (ii) via transvascular signalling through receptors on the luminal side of brain blood vessels (S124); or (iii) by independent production and release within the CNS (reviewed by Moffett et al. 1987).

The literature suggests that Ang II in the brain influences blood pressure and that this effect is, in part, due to activation of central NA transmission. When Ang II was injected intracisternally, the neurones which were activated, as shown by increased c-Fos expression, were located in the NTS, CVLM and RVLM, and included both NAergic and non-NAergic neurones (S69). Angiotensin II facilitated the field-stimulation-induced release of [³H]NA from the NTS but, surprisingly, not from the anterior hypothalamus of Sprague–Dawley and Wistar rats (Meldrum et al. 1984). Microinjections of Ang II in the locus coeruleus and other NAergic brainstem areas and also in the hypothalamus increased NA release and breakdown (Sumners & Phillips, 1983). Analogous to the ADR/NAergic activity of these brain areas (see above), microinjection of Ang II into the RVLM induced sympathetically mediated pressor responses (S5), whereas in the CVLM they caused hypotension (S140). Owing to the interaction of Ang II with multiple cell types in the NTS, the direct effect on A2 neurones is currently not clear.

A host of studies support the notion that central Ang II is involved in neurogenic hypertension. In the brainstem of SHR, levels of Ang II as well as expression of Ang II type 1 receptors (AT₁Rs) are increased (S61, S109). In the RVLM, AT₁Rs were found on NAergic and also on glutamatergic and GABAergic neurones (S73). The numbers of glutamatergic and GABAergic cells in RVLM double-stained for AT₁Rs were similar between SHR and WKY rats, but the mean optic density of AT₁R staining per cell was higher in SHR than that in Wistar rats. Interestingly, the authors did not comment on any such differences in C1 neurones specifically (S73). Functional data indicated that the effects of Ang II on central ADR/NA signalling could be altered in some brainstem nuclei of SHR. For example, retrogradely identified, spinally projecting RVLM neurones (phenotype not further characterized) were depolarized by high concentrations of Ang II (6 µM) in SHR but not in WKY rats (Matsuura et al. 2002). Another study showed excitatory effects of Ang II microinjected into the RVLM which were similar between SHR and WKY rats, whereas injections into the CVLM produced a stronger hypotension in SHR than in WKY rats (Muratani et al. 1991). Interestingly, compared with glutamate, responses to Ang II were much slower in onset (Muratani et al. 1991).

The mechanisms underlying observations in whole animals were investigated in greater detail by numerous in vitro studies, many of which have suggested that Ang II signalling to central NAergic neurones is enhanced and/or altered in SHR. In primary brainstem–hypothalamic cultures of neonatal SHR and WKY rats, fourfold higher AT₁R levels were found in SHR compared with WKY rats (S129). Furthermore, Ang II had profound effects on NA synthesis, uptake and vesicular trafficking, and these effects were enhanced in SHR (S55; reviewed in S133).

Short-term effects of Ang II on NA transmission were reported to involve AT₁R-mediated inhibition of K⁺ channels and stimulation of Ca²⁺ channels; effects which should lead to faster firing rates of NAergic neurones and rapid NA release (reviewed in S166). Indeed, an excitatory effect of Ang II on a significant proportion of unidentified brainstem neurones was found in culture; Ang II elicited an approximately 90% increase in firing rate in WKY neurones while this action was significantly greater (>200%) in SHR neurones (Sun et al. 2007). Longer term effects of Ang II may recruit the Ras–Raf–mitogen-activated protein kinase cascade and ultimately lead to an increase in NA transporter, tyrosine hydroxylase and DBH expression (S56). It has been proposed that stronger effects of Ang II involve activation of protein kinase B (PKB) via an additional phosphoinositide 3-kinase (PI3K)-dependent pathway operating selectively in SHR (S174). This pathway has been reported to underlie both the effects of Ang II on gene expression and the enhanced chronotropic (e.g. the ability to increase neuronal firing rate) effect of Ang II in SHR brainstem cultures (Sun et al. 2003). Obviously, a detailed comparative study of Ang II effects on identified NA neurones of SHR and normotensive rats is pertinent to clarify the situation.

	New insights into regulation of NA release from central neurones

Top Abstract Introduction Links between central NA... Central NA signalling and... New insights into regulation... How do we take... References

 
Our recent findings based on micro-amperometric measurements in organotypic brainstem slice cultures have shed new light on the organization of NA release from central neurones. As expected, we found exocytotic release of NA or ADR at axonal varicosities of A2, A1 and C1 neurones (Fig. 1; Chiti & Teschemacher, 2007). The majority of release events had transmitter loads consistent with exocytosis of medium-sized (80–120 nm diameter) dense-core vesicles previously described in the brain and in sympathetic neurones, assuming the same intravesicular NA concentration as that postulated for adrenal chromaffin cells (Fillenz, 1990; S51). In addition, rare but very large release events were observed, each one of which delivers as much transmitter as 100 average medium vesicles. Their quantal size is well in the range of large chromaffin granules, as found in the adrenal medulla, but their morphological substrate within the brain at electron microscopic level so far has not been convincingly identified. Since catecholamine reuptake in brain tissue is saturable, it is tempting to speculate that NA released from these putative granule-like large vesicles may diffuse further to reach more distant receptors than NA released in small quanta (Fig. 3). In addition, large and small vesicles may have a different complement of cotransmitters.

View larger version (31K): [in this window] [in a new window]   Figure 3.  Biological roles of small and large quanta might be different Although the anatomical substrate of large release events is not yet clear, their transmitter load makes them comparable to chromaffin granules. It is likely that an instant release of a large quantity of NA will enable it to travel further in the extracellular space and reach targets which might not be accessible when NA arrives in small quanta. It is also possible that the small and large release events have different cotransmitter complements.

	How do we take this field forward?

Top Abstract Introduction Links between central NA... Central NA signalling and... New insights into regulation... How do we take... References

 
In general, both clinical and experimental data provide a compelling case for the important role of central NA(ADR) transmission in a range of cardiovascular diseases. We hope that better understanding of the fundamentals of central NA transmission and implementation of new research strategies will help to resolve many of the controversies and fill the gaps in knowledge discussed in this review.

As mentioned earlier, studies on the role of NA transmission in CVS regulation, and specifically hypertension, have heavily employed methods of local microinjection of adrenergic agonists and antagonists and destruction of NA neurones using toxins. Both approaches have severe limitations, briefly mentioned above. Moreover, in some cases the literature reveals controversies which are very hard to overcome without changing the way we currently think about NA neurones. Some of the problems inherent to previous experimentation can be illustrated by the following example. Noradrenaline release upon stimulation of baroreceptors or aortic nerve has been detected in the NTS by microdialysis (S146). The most logical explanation is that this NA comes from the local A2 neurones. However, it has been reported that the discharge rate of A2 neurones changes only minimally when baroreceptors are being stimulated (Moore & Guyenet, 1985). There are two ways out of this conundrum. Could it be that the NA released upon baroreceptor stimulation in NTS (S146) arrives from the local varicosities of other NAergic neurones which project into the NTS? This is possible but seems peculiar because the NTS receives the bulk of the input from baroreceptors. Alternatively, A2 neurones can release NA without a major change of the frequency of their action potentials. This possibility needs to be addressed directly, and the responsible alternative triggers need to be identified. Having found the answer to this question, and in order to establish the role of A2 neurones for blood pressure control long term, ideally one would want to selectively increase or decrease NA release from A2 neurones, without directly affecting other NAergic groups, and monitor the outcome in chronic unanaesthetized animals.

Molecular techniques offer valuable new opportunities (Kasparov et al. 2004, 2005). For example, lentiviral vectors do not transduce NA neurones retrogradely and can be used to alter functions of selected clusters of NA cells without the risk of off-site effects. In our recent studies, we attempted to electrically ‘silence’ A2 neurones by expression of an additional potassium conductance and observed an effect on blood pressure which was exaggerated in the SHR (Duale et al. 2007). In this case, we could be certain that NAergic cells projecting into the NTS were not affected. Was expression of potassium channels (and reduction of neuronal firing rates) the most effective way to reduce NA release from these cells? We do not know yet, but possibly other transgenes directly interfering with the NA release machinery will help us to distinguish between action potential-dependent and -independent NA release.

Alternatively, how can we selectively increase the activity of NAergic neurones? It has recently become possible to drive the activity of neurones by expressing a cation channel (ChR2) which can be switched on by light of certain wavelengths (S8, S39). Given that the expression of transgenes in NAergic neurones of various groups using viral vectors has been well established (Kasparov et al. 2004, 2005; Chiti & Teschemacher, 2007; Duale et al. 2007; S92), this opens a way for selective manipulation of NAergic circuits in specific nuclei in vivo and in vitro.

Hopefully, better understanding of the basic physiology of central NAergic neurones and successful implementation of novel molecular tools will help finally to clarify the roles of various NAergic nuclei in the central control of the CVS in health and disease.

	References

Top Abstract Introduction Links between central NA... Central NA signalling and... New insights into regulation... How do we take... References

 
(Selected references only. Please see the supplementary material for the full list of sources, including supplemental references.)

Aggarwal A, Esler MD, Lambert GW, Hastings J, Johnston L & Kaye DM (2002). Norepinephrine turnover is increased in suprabulbar subcortical brain regions and is related to whole-body sympathetic activity in human heart failure. Circulation 105, 1031–1033.[Abstract/Free Full Text]

Alvarenga ME, Richards JC, Lambert G & Esler MD (2006). Psychophysiological mechanisms in panic disorder: a correlative analysis of noradrenaline spillover, neuronal noradrenaline reuptake, power spectral analysis of heart rate variability, and psychological variables. Psychosom Med 68, 8–16.[Abstract/Free Full Text]

Bailey JE, Argyropoulos SV, Lightman SL & Nutt DJ (2003). Does the brain noradrenaline network mediate the effects of the CO₂ challenge? J Psychopharmacol 17, 252–259.[Abstract]

Benetos A, Gavras I & Gavras H (1986). Norepinephrine applied in the paraventricular hypothalamic nucleus stimulates vasopressin release. Brain Res 381, 322–326.[CrossRef][Medline]

Berridge CW & Abercrombie ED (1999). Relationship between locus coeruleus discharge rates and rates of norepinephrine release within neocortex as assessed by in vivo microdialysis. Neuroscience 93, 1263–1270.[CrossRef][Medline]

Bristow JD, Honour AJ, Pickering GW, Sleight P & Smyth HS (1969). Diminished baroreflex sensitivity in high blood pressure. Circulation 39, 48–54.[Abstract/Free Full Text]

Camacho A, Cornell A & Phillips MI (1981). Norepinephrine mediates the release of vasopressin and pressor effect to central angiotensin II. Fed Proc 40.

Chalmers JP, Blessing WW, West MJ, Howe PR, Costa M & Furness JB (1981). Importance of new catecholamine pathways in control of blood pressure. Clin Exp Hypertens 3, 393–416.[CrossRef][Medline]

Chan JYH, Jang S-F & Chan SHH (1992). Inhibition by locus coeruleus on the baroreceptor reflex response in the rat. Neurosci Lett 144, 225–228.[CrossRef][Medline]

Chiti Z & Teschemacher AG (2007). Exocytosis of norepinephrine at axon varicosities and neuronal cell bodies in the rat brain. FASEB J 21, 2540–2550.[Abstract/Free Full Text]

Cocco G, Strozzi C, Haeusler G, Chu D, Amrein R & Padovan GC (1979). The therapeutic value of clonidine in patients with coronary heart disease. Eur J Cardiol 10, 221–228.[Medline]

Comer AM, Qi J, Christie DL, Gibbons HM & Lipski J (1998). Noradrenaline transporter expression in the pons and medulla oblongata of the rat: localisation to noradrenergic and some C1 adrenergic neurones. Brain Res Mol Brain Res 62, 65–76.[Medline]

Cubeddu LX & Hoffman IS (1987). Cerebrospinal fluid norepinephrine levels in essential hypertension: effects of drug treatment and withdrawal. J Cardiovasc Pharmacol 10 (Suppl. 12), S205–S210.[CrossRef][Medline]

Cubeddu LX, Hoffmann IS, Davila J, Barbella YR & Ordaz P (1984). Clonidine reduces elevated cerebrospinal fluid catecholamine levels in patients with essential hypertension. Life Sci 35, 1365–1371.[CrossRef][Medline]

Cunningham ET Jr & Sawchenko PE (1988). Anatomical specificity of noradrenergic inputs to the paraventricular and supraoptic nuclei of the rat hypothalamus. J Comp Neurol 274, 60–76.[CrossRef][Medline]

Dampney RA (1981). Brain stem mechanisms in the control of arterial pressure. Clin Exp Hypertens 3, 379–391.[CrossRef][Medline]

Dampney RA (1994). Functional organization of central pathways regulating the cardiovascular system. Physiol Rev 74, 323–364.[Free Full Text]

Duale H, Waki H, Howorth P, Kasparov S, Teschemacher AG & Paton JFR (2007). Restraining infulence of A2 neurones in chronic control of blood pressure in SHR. Cardiovasc Res 76, 184–193.[Abstract/Free Full Text]

Elghozi JL, Head GA, Wolf WA, Anderson CR & Korner PI (1989). Importance of spinal noradrenergic pathways in cardiovascular reflexes and central actions of clonidine and -methyldopa in the rabbit. Brain Res 499, 39–52.[CrossRef][Medline]

Elliott JM, Stead BH, West MJ & Chalmers J (1985). Cardiovascular effects of intracisternal 6-hydroxydopamine and of subsequent lesions of the ventrolateral medulla coinciding with the Al group of noradrenaline cells in the rabbit. J Auton Nerv Syst 12, 117–130.[CrossRef][Medline]

Erinoff L, Heller A & Oparil S (1975). Prevention of hypertension in the SH rat: effects of differential central catecholamine depletion. Proc Soc Exp Biol Med 150, 748–754.[Abstract]

Esler M, Alvarenga M, Pier C, Richards J, El-Osta A, Barton D, Haikerwal D, Kaye D, Schlaich M, Guo L, Jennings G, Socratous F & Lambert G (2006). The neuronal noradrenaline transporter, anxiety and cardiovascular disease. J Psychopharmacol 20, 60–66.[Abstract/Free Full Text]

Esler M & Kaye D (2000). Measurement of sympathetic nervous system activity in heart failure: the role of norepinephrine kinetics. Heart Fail Rev 5, 17–25.[CrossRef][Medline]

Esler MD, Lambert GW, Ferrier C, Kaye DM, Wallin BG, Kalff V, Kelly MJ & Jennings GL (1995). Central nervous system noradrenergic control of sympathetic outflow in normotensive and hypertensive humans. Clin Exp Hypertens 17, 409–423.[Medline]

Fillenz M (1990). Noradrenergic Neurons. Cambridge University Press, Cambridge, UK.

Fujino K (1984). Brain catecholamines in spontaneously hypertensive and DOCA-salt hypertensive rats. Acta Medica Okayama 38, 325–340.[Medline]

Goodchild AK, Moon EA, Dampney RAL & Howe PRC (1984). Evidence that adrenaline neurons in the rostral ventrolateral medulla have a vasopressor function. Neurosci Lett 45, 267–272.[CrossRef][Medline]

Granata AR, Numao Y, Kumada M & Reis DJ (1986). A1 noradrenergic neurons tonically inhibit sympathoexcitatory neurons of C1 area in rat brainstem. Brain Res 377, 127–146.[CrossRef][Medline]

Grassi G, Cattaneo BM, Seravalle G, Lanfranchi A & Mancia G (1998). Baroreflex control of sympathetic nerv