A Role for Benzamil-Sensitive Proteins of the Central Nervous System in the Pathogenesis of Salt-Dependent Hypertension

SUMMARY

1. While increasing evidence suggests that salt-sensitive hypertension is a disorder of the central nervous system, little is known about the critical proteins (e.g., ion channels or exchangers) that play a role in the pathogenesis of the disease.

2. Central pathways involved in the regulation of arterial pressure have been investigated. In addition, systems such as the renin-angiotensin-aldosterone axis, initially characterized in the periphery, are present in the central nervous system, and seem to play a role in the regulation of arterial pressure.

3. Central administration of amiloride, or its analogue benzamil hydrochloride, has been shown to attenuate several forms of salt-sensitive hypertension. In addition, intracerebroventricular (ICV) benzamil effectively blocks pressor responses to acute osmotic stimuli, such as ICV hypertonic saline. Amiloride or its analogues have been shown to interact with the brain renin-angiotensin-aldosterone system (RAAS) and to effect the expression of endogenous ouabain-like compounds; both could play a role in the interaction between amiloride compounds and arterial pressure. Peripheral treatments with benzamil, even at higher doses than those given centrally, have little or no effect on arterial pressure. These data provide strong evidence that benzamil-sensitive proteins (BSPs) of the central nervous system play a role in cardiovascular responsiveness to sodium.

4. Mineralocorticoids have been linked to human hypertension; many patients with essential hypertension respond well to pharmacological agents antagonizing the mineralocorticoid receptor (MR), and certain genetic forms of hypertension are due to chronically elevated levels of aldosterone. The deoxycorticosterone acetate (DOCA) -salt model of hypertension is a benzamil-sensitive model that incorporates several factors implicated in the etiology of human disease, including mineralocorticoid action and increased dietary sodium. The DOCA-salt model is ideal for investigating the role of BSPs in the pathogenesis of hypertension, because mineralocorticoid action has been shown to modulate the activity of at least one benzamil-sensitive protein, the epithelial sodium channel (ENaC).

5. Characterizing the BSPs involved in the pathogenesis of hypertension may provide a novel clinical target. Further studies are necessary to determine which BSPs are involved, and where in the nervous system they are located.

INTRODUCTION

Evidence suggests that in humans, hypertension is a disorder of the central nervous system; specifically, sympathetic nervous activity (SNA) is increased in the human disease state [1, 2]. Furthermore, increasing dietary sodium has been shown to affect both arterial pressure and SNA [3–9]. In addition, several animal models of salt-sensitive hypertension exhibit increased SNA, or increased levels of plasma norepinephrine (an indicator of increased sympathetic drive) [10–12]. While high sympathetic tone seems to be a hallmark of many salt-dependent models of hypertension, central administration of amiloride, or its analogue benzamil, attenuates both the level of hypertension reached and circulating levels of norepinephrine [13]. These data suggest an interaction between benzamil-sensitive proteins (BSPs) and SNA; however, it remains to be shown where within autonomic pathways the BSPs mediating this interaction are located. This review will address what is known about the neural pathways involved in the regulation of arterial pressure, as well as the characteristics of known benzamil-sensitive proteins. In addition, we will attempt to summarize possible interactions between the activity of BSPs and central neural control of cardiovascular function.

THE ETIOLOGY OF SALT-SENSITIVE HYPERTENSION

Central Pathways

Some of what is currently understood about the neural pathways mediating salt-dependent forms of hypertension has come from studies employing acute osmotic stressors, and extrapolating those results to long-term salt sensitivity of arterial pressure. Drawing on these results, Figure 1 summarizes the hypothetical central pathways involved in the pathogenesis of salt-dependent hypertension. The CVOs, with their fenestrated capillaries, exhibit a weakened blood brain barrier; because of this, CVOs are uniquely situated in the CNS-- able to sense changes in plasma osmolality or hormone levels and to drive subsequent neural activity (e.g., [14], figure 1). Discrete lesions of the area postrema (AP), a hindbrain CVO, prevent MAP increases in the DOCA-salt model of hypertension [15]. Lesions of the anteroventral third ventricle (AV3V region), which contains both the organum vasculosum of the lamina terminalis (OVLT), and projections from subfornical organ (SFO) to downstream targets, decrease MAP in the DOCA-salt model [16] and other salt-sensitive models of hypertension [17, 18]. OVLT and SFO are forebrain CVOs.

Figure 1. Central Pathways.

CVOs (AP, OVLT, SFO) have a weakened blood-brain barrier due to their fenestrated capillaries. These CNS sites may be important for sensing circulating hormones and/or plasma osmolality [14]. The CVOs send projections to integrative and regulatory sites, such as the MnPO (involved in salt appetite and regulating drinking behavior) and the PVN (critical for defense of homeostasis; e.g., [19]). These pathways continue onto pre-sympathetic areas, such as the RVLM and the IML of the spinal cord [35], which allows for modulation of sympathetic activity and arterial pressure [95, 96].

The CVOs feed unto midbrain sites, and then to bulbospinal pathways that are critical for control of sympathetic nervous activity [19]. Activity of the median preoptic area (MnPO), a downstream target of the OVLT and SFO, seems to play a role in many models of hypertension, [20–22], salt appetite [23], and drinking behavior (of interest because drinking is increased in some forms of salt-sensitive hypertension) [24, 25]. The MnPO, in turn, sends projections to the Zona Incerta (ZI) and to the paraventricular nucleus of the hypothalamus (PVN). The ZI plays a role in osmotically induced drinking. In single-cell electrophysiological studies, ZI has been shown to be responsive to changes in osmolality in the AV3V region and at MnPO [26]; firing rates of ZI were affected by both hyperosmotic stimuli (ICV hypertonic saline of varying concentrations) and a hypoosmotic stimulus (distilled water) injected at upstream connection sites. In addition, in electrophysiological recordings from ZI of conscious animals, firing rates increase to the sight of water following osmotic stress (ICV hypertonic saline administration) but not in control animals [27]. Finally, lesions of ZI or ZI antagonism (through local injection of pharmaceutical agents) disrupts drinking behavior in response to osmotic stressors [28–32].

The PVN, another downstream target of the CVOs and the MnPO [33], has been implicated in the control of MAP. The PVN is a hypothalamic nucleus, which plays a role in regulating neuroendocrine responses, such as modulating systemic vasopressin levels through activity of its magnocellular neurons. In addition, the PVN is a presympathetic site, sending projections from parvocellular neurons to the intramediolateral cell column of the spinal cord and to the rostroventrolateral medulla [34–36]. Lesions of the PVN attenuate several forms of hypertension [37, 38]. In addition, OVLT, MnPO, and PVN all show increased activity, as measured by c-Fos immunoreactivity, following acute DOCA-salt treatment [39].

Ultimately, integration of these upstream sites are thought to affect SNA by converging on sympathetic premotor neurons in the rostral ventrolateral medulla (RVLM). The RVLM is believed to be the key brainstem site for regulation of spinal sympathetic preganglionic neurons in the intermediolateral cell column (IML) of the spinal cord [40].

The Role of Central Mineralocorticoids in Salt-sensitive Hypertension

Aldosterone, the primary mineralocorticoid, is a steroid hormone secreted by the adrenal gland, which plays a major role in sodium balance and body fluid homeostasis. Unlike many other steroid hormones, aldosterone does not readily cross the blood-brain barrier [41– 43], and so it has been hypothesized that the CVOs may be critical for driving aldosterone-sensitive responses in the CNS [44]. Basic research implicates mineralocorticoids in the control of arterial pressure. Chronic intracerebroventricular (ICV) infusion of aldosterone results in increased arterial pressure [45, 46]. While systemic administration of aldosterone, or its precursor deoxycorticosterone acetate (DOCA), also leads to increased arterial pressure, ICV infusion of MR antagonists blocks the hypertensive effect of systemic mineralocorticoids [45, 47], indicating that central responses are necessary for the hypertensive effects seen in systemic aldosterone treatment. Furthermore, following ICV administration of MR antagonists, Dahl salt-sensitive rats remain normotensive, even on a high salt diet [48, 49].

BENZAMIL-SENSITIVE PROTEINS

The Deg/ENaC Superfamily

Benzamil and other amiloride-analogues act at two large classes of proteins. The first are members of the Degenerin/Epithelial Sodium Channel (Deg/ENaC) Superfamily of ion channels. All family members share a common topology, with subunits containing two membrane-spanning regions and a large, cysteine-rich extracellular loop. Mammalian family members include the ENaCs and the acid sensitive ion channels, or ASICs. In both groups, the proteins form multimers composed of four subunits. Two subunits, which are the main structural components of the channel, are repeated. The remaining subunits are modulatory or accessory components of the channel [50–55].

ENaCs are sodium channels, and are involved in salt homeostasis. ENaCs play a critical role in sodium reabsorption in the distal nephron, as well as at the lung and colon [55, 56]. ENaCs may also play a role in the arterial baroreceptor reflex [57]. A δ subunit, sharing much homology with the α subunit, has been identified. Although δ ENaC has a broad neural distribution, to date, it has only been found in primates, and so probably does not play a role in rodent sodium homeostasis [58]. At least one lab has shown the presence of the benzamil-blockable Epithelial Sodium Channel (ENaC) in many autonomic regulatory sites; however, these results remain to be replicated by other labs [59]. In attempts to replicate these findings, using primers published in this study, we were unable to amplify a clear, single RT product [60].

The ASICs are proton gated, non-selective cation channels, which are widely expressed in neural tissue. ASICs play a role in such diverse functions as nociception, response to ischemic events, and ability to taste salt. Gating properties and channel function are dramatically affected by the particular subunits present, and so the specific multimer formed may dictate the physiological function of the channel.

Ion Transport Systems

The second large class of benzamil-sensitive proteins are ion transport systems. These include the Na⁺/H⁺ exchanger, the Na⁺/Ca⁺⁺ exchanger, the Na⁺ pump, and the Ca⁺⁺ pump. The Na⁺/H⁺ ion exchanger, or antiport, is a membrane-localized protein found in a wide variety of cell types. It relies on secondary active transport to facilitate the movement of ions (i.e., ion flux generated by active transport at other proteins, such as the Na⁺/K⁺ ATPase generate the gradients needed to run these ion exchangers). Increased activity of the Na⁺/H⁺ exchanger has been linked to primary hypertension in humans [61].

The Na⁺/Ca⁺⁺ exchanger is a bidirectional transporter, whose activity depends on the electrochemical gradients present at the membrane. The Na⁺/Ca⁺⁺ exchanger’s activity is coupled to that of the Na⁺/H⁺ antiport, and may also play a role in hypertension; while benzamil treatment has been shown to reduce certain models of salt-sensitive hypertension, Keep and colleagues [62] found that amiloride analogues with greater affinity for the Na⁺/Ca⁺⁺ exchanger (and very low affinity for Na⁺ channels) are equally effective in blocking the hypertensive effect of DOCA-salt treatment.

While amiloride and benzamil are inhibitors of these proteins, the ion transport systems have higher affinity for other amiloride analogues, such as 3',4'-dichlorobenzamil; 2',4'-dimethylbenzamil; 5-(N-ethyl-N-isopropyl) amiloride; and 5-(N-methyl-N-isobutyl) amiloride.

DOCA-SALT HYPERTENSION

DOCA-salt: a neural model

Although peripheral effects of DOCA-salt hypertension (i.e., renal and vascular) are well documented [63], the model is considered a neurogenic form of hypertension. Lesions of the AV3V region [16], or of the AP [15], attenuate DOCA-salt hypertension. In addition, following treatment with a sympatholytic agent, such as hexamethonium, DOCA animals show greater decreases in MAP than control animals [64, 65]. Furthermore, SNA and plasma norepinephrine levels are increased upon DOCA-salt treatment [66]. DOCA [67] or other mineralocorticoid [68] treatment has been shown to increase levels of plasma vasopressin, as well as salt appetite [69], further indication of increased CNS activity following mineralocorticoid treatment. Thirst may also be increased in DOCA-salt hypertension; however, this is often conflated with salt appetite, as animals are frequently given saline in place of drinking water. Finally, ICV benzamil treatment attenuates DOCA-salt hypertension, while peripheral benzamil administration has no effect [13, 70]. Taken together, these data provide strong evidence that the DOCA-salt model is driven by neural activity.

DOCA-salt Hypertension, Mineralocorticoid Receptors, and Benzamil-Sensitive Proteins

The MR is a cytoplasmic receptor with broad neural distribution, which is typically found in association with heat shock and other chaperone proteins. When MR binds ligand, the protein undergoes a conformational change, which frees it from its chaperone proteins and exposes a nuclear translocation signal. In the nucleus, MR acts as a transcription factor [71]. Because of these properties of the MR, nuclear localization of the protein can be taken as an index of MR activity. Indeed, recent studies have shown that MR adjacent to the AP in neurons of the nucleus of the tractus solitarius (NTS) translocate to the cellular nucleus following aldosterone treatment [72]. This unique population of neurons expresses the enzyme 11 β-hydroxysteroid dehydrogenase type 2, or HSD2. HSD2 catalyzes glucocorticoids; because MR have a nearly equal affinity for aldosterone and the more prevalent glucocorticoids, under normal conditions, MR are thought to be nearly completely occupied by glucocorticoid. However, HSD2 plays a permissive role in neuronal responsiveness to aldosterone through MR. Cells expressing HSD2 are able to respond to changes in aldosterone level, despite higher systemic levels of glucocorticoid. In fact, the HSD2 neurons of the NTS may play a role in modulating responses to DOCA treatment; following one week of DOCA, activity in these neurons is increased [73].

The DOCA-salt model is an ideal model for investigating the role of BSPs in the pathogenesis of hypertension. Figure 2 summarizes the putative relationship between mineralocorticoid action, high-salt diet, and benzamil-blockable channels. Interactions between the mineralocorticoid receptor (MR) and ENaCs have been well established [e.g., 68, [74]]. MR is a transcription factor for the α subunit of the ENaC, a BSP known to play a role in sodium transport at the kidneys [75]. Not only has MR activity been linked to the expression of BSPs, but also to the enzymes that regulate the plasma membrane localization of these channels [74]. In addition to direct effects on ENaC transcription, MR is a regulator of Sgk1, a kinase that plays a role in trafficking ENaCs to the cell membrane and in increasing sodium transport across ENaCs. An additional outcome of MR activation and subsequent Sgk1 activity is the inhibitory phosphorylation of Nedd-4. Nedd-4 plays a role in ubiquitin-mediated degradation of ENaC channels. Because of these effects, ICV infusion of MR antagonist will not only block direct aldosterone effects, but may also cause long-term changes in sodium permeability of neurons. While the relationship between MR and ENaCs has been shown, this interaction remains to be established for ASICs or other BSPs.

Figure 2. Model of the Role of BSPs in the Pathogenesis of DOCA-Salt Hypertension.

We propose that benzamil-sensitive ion channels play a role in DOCA-salt hypertension. Following DOCA treatment and subsequent increased MR activity, there would be an increased localization of BSPs to the membrane of HSD2 expressing cells. We believe that this increase in membrane-localized BSPs is likely to occur at sodium-sensitive sites, such as the OVLT, because BSPs seem to play a role both in chronic models such as DOCA-salt hypertension, as well as in acute sensitivity to CSF sodium levels. Membrane localization of BSPs would increase neural activity in the presence of high sodium levels; this in turn, could modulate activity at downstream sites, and thereby regulate SNA and increases in arterial pressure.

Our model is based on the assumption that the BSPs involved in the model are ion channels. Because ENaCs have an established role in sodium homeostasis in the periphery, it has been hypothesized that they subserve this role in the central nervous system. In addition, benzamil has high affinity for the ENaC, and even low doses of benzamil can block acute pressor responses to osmotic stimuli. However, as noted earlier, amiloride analogues can also interact with ion transport systems, and these systems have been shown to play a role in the regulation of arterial pressure.

Benzamil and DOCA-salt Hypertension

DOCA-salt hypertension is attenuated by central treatment with amiloride, or its analogue benzamil [13, 70]. Benzamil is a fairly specific blocker, targeting the ENaC and ASIC proteins, and only at higher doses, certain ion transport systems. While ICV benzamil treatment attenuates DOCA-salt hypertension and decreases vasopressin and norepinephrine levels, systemic benzamil treatment has no significant effect (even when given at much higher doses), indicating a critical role for central BSPs.

Early studies of the effect of benzamil on DOCA-salt hypertension employed tail-cuff measurements in animals with established hypertension. We have recently used radio telemetry transmitters to record MAP in awake, behaving animals, 24 hours a day. These studies suggests that DOCA-salt hypertension develops in several distinct phases; generally, there is an initial rapid rise in MAP, followed by a slower, more prolonged period of increasing MAP, and that benzamil plays a role in attenuating the later, more slowly-developing phase of hypertension [76].

OTHER INTERACTIONS BETWEEN BENZAMIL AND SALT-SENSITIVE CARDIOVASCULAR RESPONSES

Sodium Sensitivity and BSPs

Administration of benzamil will block not only DOCA-salt hypertension, but genetic models of hypertension that are driven by high salt diet [77, 78]. Therefore, it is likely that the BSPs play a role in sodium sensitivity; specifically, we hypothesize that when MR binds ligand, there is a subsequent upregulation of BSPs, which would increase membrane permeability to sodium. This response, in the face of transient increases in plasma sodium levels, would lead to membrane depolarization and an increase in neuronal activity, driving sympathetic outflow and increases in MAP (see figure 2). In addition to a role in salt-sensitive forms of hypertension, BSPs seem to play a role in pressor responses to acute stimuli. The rapid rise in arterial pressure, typically seen following ICV hypertonic saline, is blocked by ICV benzamil (figure 3).

Figure 3. Acute Pressor Response to ICV Saline is Blocked by Benzamil Hydrochloride.

Pentobarbital anesthetized rats (n=6) were instrumented for direct measurement of mean arterial pressure (MAP) via a femoral arterial catheter. MAP was measured by connecting the catheter to a volume displacement pressure transducer coupled to a Grass polygraph. Under stereotaxic guidance, the tip of a 23-gauge stainless steel cannula was placed in the lateral cerebral ventricle. After obtaining baseline MAP for 5 minutes, 50 µL of 1.5% saline was injected over 30 seconds. MAP was recorded for an additional 20 minutes post injection. 20µL of 20nM benzamil hydrochloride (e.g., 0.4 picomoles) was then delivered ICV into the lateral ventricle. A second injection of saline was administered 5 minutes after the benzamil injection, and MAP recorded for 20 minutes. Arrow indicates time of ICV hypertonic saline injection.

In this experiment, male Sprague-Dawley rats (n=6, 250–310 grams) were anesthetized with sodium pentobarbital and were instrumented for direct measurement of mean arterial pressure (MAP) via a femoral arterial catheter. MAP was measured by connecting the catheter to a volume displacement pressure transducer coupled to a Grass polygraph.

Under stereotaxic guidance, the tip of a 23-gauge stainless steel cannula was placed in the lateral cerebral ventricle. After obtaining baseline MAP for 5 minutes, 50 µL of 1.5% saline was injected over 30 seconds. MAP was recorded for an additional 20 minutes post injection. 20µL of 20 nM benzamil hydrochloride (e.g., 0.4 picomoles benzamil) was then delivered ICV into the lateral ventricle. A second injection of saline was administered 5 minutes after the benzamil injection, and MAP recorded for 20 minutes.

Although not examined in our study, Brooks and colleagues have shown that acute pressor responses contain both a vasopressinergic and a non-vasopressinergic component: following V1 antagonism, pressor responses to intravenous hypertonic saline are attenuated but not blocked [79]. Nishimura and colleagues have demonstrated that ICV benzamil treatment not only eliminated the pressor response to an acute injection of 1.5M NaCl, but also attenuated levels of sympathetic discharge and plasma vasopressin as compared to control animals [13]. Central benzamil treatment blocks nearly all pressor response to acute hypertonic saline (figure 3), suggesting that both vasopressinergic and other mechanisms are sensitive to benzamil treatment. However, pressor responses to more chronic stimuli, such as DOCA-salt treatment, are merely attenuated. The molecular substrate for this acute/chronic dichotomy remains to be elucidated; however, it is likely that even in chronic protocols, activity of vasopressinergic neurons is altered following ICV benzamil [76].

The Renin-Angiotensin-Aldosterone System (RAAS)

The RAAS is a well-studied system, which may serve as a “sensor” to the CNS; plasma hormone concentrations of this system change in response to dietary sodium levels. The RAAS may be involved in the DOCA-salt model; changes in the RAAS have been measured, both in the brain and the kidney, following DOCA-salt treatment. Renin and angiotensin-converting-enzyme mRNAs are downregulated in the kidney, following DOCA-salt treatment; however, mRNA levels for these transcripts do not change in the brain. Conversely, following ICV benzamil treatment, mRNAs for critical members of the RAAS are downregulated in the brain, but not at the kidney. Peripheral benzamil treatment has no effect on the members of the RAAS, either in the brain or the kidney. These data indicate that benzamil-sensitive proteins of the central nervous system have the ability to modulate activity of the central RAAS, and thereby modulate arterial pressure. [70]

Endogenous Ouabain-like Compounds and Hypertension

Endogenous Digitalis-like compounds (also termed ouabain-like compounds or endobains) were first found in human-derived tissues in 1991 [80], and these compounds have been implicated in human essential hypertension [81–83]. Endogenous ouabain-like compounds have since been characterized in bovine-derived and murine-derived tissues [84]. Ouabain is a well-known inhibitor of the Na+/K+ ATPase, as well as the Na+/Ca++ exchanger and sodium pump.

Using immunoassay, Gottlieb and colleagues found endobains increased in the plasma of heart failure patients [85]. In addition, plasma endobains have been shown to increase with sodium intake in humans [86, 87]; high sodium intake also leads to endobain increase in a variety of tissue-types in rats [88]. Furthermore, chronic infusions of ouabain have been shown to produce hypertension in the rat [89–91]. Increasing levels of ouabain-like compounds affect the central RAAS and could explain increased SNA [92].

Administration of benzamil has been shown to block the effect of increasing ouabain levels [93, 94]. In addition, benzamil can act directly at several ouabain-sensitive proteins (e.g., Na+/Ca++ exchanger, Na+ pump). These data suggest a potential interaction between endobains and BSPs of the central nervous system.

CONCLUSION

Evidence from multiple animal models suggests a critical role for BSPs in the pathogenesis of salt-sensitive hypertension; however, the specific proteins involved have not been characterized. While it has long been hypothesized that the ENaC may play a role in the central control of cardiovascular responses, finding ENaC in the CNS has proven difficult. While at least one lab has shown the presence of CNS ENaCs [59], other labs have been unable to replicate this result, or have only found evidence for certain ENaC subunits [60]. Furthermore, in a study testing the efficacy of several amiloride analogues in blocking DOCA-salt hypertension, Keep and colleagues found that even amiloride analogues with low affinity for sodium channels were able to attenuate DOCA-salt hypertension. They concluded that the antihypertensive effects of amiloride analogues may be due to inhibiting Na⁺/Ca⁺ exchange in the brain [62]. Investigations of the RAAS or of endobains provide further evidence that other proteins, either in the place of, or in addition to the ion channels, may be the BSPs involved in salt-sensitive hypertension. Proper characterization of the BSPs involved in salt-sensitive hypertension is critical, both for our understanding of the disease, and as an avenue for the development of novel clinical treatments.

Abbreviations

AP

area postrema

ASIC

Acid Sensitive Ion Channel

AV3V

anteroventral third ventrical region

BSP

Benzamil-sensitive protein

CNS

Central Nervous System

CVO

circumventricular organ

Deg

degenerin

DOCA

deoxycorticosterone acetate

ENaC

Epithelial Sodium Channel

HSD2

11-β-hydroxysteroid dehydrogenase type 2

ICV

intracerebroventricular

IML

intermediolateral cell column

MAP

mean arterial pressure

MnPO

median preoptic area

MR

mineralocorticoid receptor

NTS

nucleus of the tractus solitarious

OVLT

organum vasculosum of the lamina terminalis

PVN

paraventricular nucleus

RAA

Renin-angiotensin-aldosterone

SFO

subfornical organ

SNA

sympathetic nervous activity

ZI

zona incerta