Neurotransmitter Switching Coupled to β-Adrenergic Signaling in Sympathetic Neurons in Prehypertensive States

Supplemental Digital Content is available in the text.

T he myocardial β-adrenergic receptor (βAR) signaling pathway plays a pivotal role in the pathogenesis of many cardiovascular diseases. Chronic cardiac adrenergic activation and impaired myocardial cyclic nucleotide (cN) signaling, resulting from enhanced catecholaminergic neurotransmission, are well-established contributors to ventricular hypertrophy, arrhythmia, and cardiomyocyte apoptosis. [1][2][3] Sympathetic overactivity and vagal impairment (dysautonomia) are recurrent features in normotensive subjects with a familial predisposition for hypertension 4,5 and in animal models of this disease. [6][7][8] Moreover, patients with familial dysautonomia experience catecholaminergic supersensitivity, episodic hypertension, and have a high propensity for fatal cardiac events. 9 β-Blockers are a mainstay treatment for many cardiovascular diseases and stress-related events. 10 Chronic β-blocker therapy affords patients a wide-range of beneficial effects, including β 1 AR resensitization and restoration of intracellular cN signaling pathways, improvements in cardiac myocyte contractility, and reversal of ventricular remodeling. 1,3 The precise mechanisms, however, that mediate and sustain the beneficial effects of β-blockers in disease remain unclear, 11,12 although the presence of potentiating Gα s -coupled presynaptic βARs on presynaptic sympathetic terminals suggests a role for β-blockers in regulating cardiacneuronal communication. [13][14][15][16][17][18][19][20][21][22] The Adrenaline Hypothesis of hypertension argues that small incremental increases in plasma adrenaline (epinephrine) enhance sympathetic activity through sustained activation of presynaptic sympathetic βARs, leading to the development of hypertension. 17,23,24 Whether epinephrine synthesis occurs before the onset of hypertension is not known, as there is limited cellular and molecular data within the sympathetic stellate ganglia to confirm this idea.
In this study, we investigated whether sympathetic βARs are present on human and rat sympathetic stellate ganglia (cervicothoracic ganglia, T1-T3) that preferentially innervate the heart. [25][26][27][28] We aimed to establish whether intracellular second messenger signaling coupled to presynaptic βARs is impaired in prehypertensive states and contributes to altered Ca 2+ and cN signaling before increases in arterial blood pressure. Finally,

Rat and Human Sympathetic Stellate Ganglia Express β 1 and β 2 Adrenoceptors
We sequenced the transcriptome of the sympathetic stellate ganglia from 16-week-old male Wistar rats (n=4) and SHR (n=4). At 16 weeks, it is well-established that SHR display hypertension and sympathetic hyperactivity. 6,[29][30][31]33,35,36,42 Using quasi-mapping RNAseq 43 and quantitative real-time (qRT)polymerase chain reaction (PCR), we identified the presence of β 1 AR (Adrb1) and β 2 AR (Adrb2) mRNA transcripts, in addition to α 2A AR (Adra2a) and tyrosine hydroxylase (Th) mRNA transcripts, markers of presynaptic sympathetic neurons, respectively ( Figure 1A; Figure S1). We selected the α 2A AR isoform as an indicator of presynaptic neuronal phenotype based on reports that the α 2A AR primarily regulates presynaptic sympathetic activity. 44 The α 2C AR isoform plays a secondary role in regulating presynaptic norepinephrine release, 44 whereas the α 2B AR isoform has a preferential role within the vasculature. 44 The mRNA expression for α 2A AR was also found to be significantly higher than α 2C AR expression identified by RNAseq (data not shown). Using RNAseq, we found that Adrb2 mRNA expression was significantly lower in SHR ganglia compared with Wistar ( Figure 1A; Figure S1C; P.adj=0.00945). Data points represent mean raw counts±SEM ( Figure 1A).
The presence of Adrb1, Adrb2, Adra2a, and Th mRNA transcripts was identified and quantified by quantitative real time PCR (qRT-PCR) using RNA extracted from 4-week pre-SHR and Wistar rats (n=3 rats/group, unpooled; Figure S1D) and 16-week SHR and Wistar rats (n=4 rats/group, unpooled; Figure S1D). qRT-PCR data were analyzed using the ∆∆C T method, where raw counts in both strains were first normalized to a control housekeeping gene B2m, and the difference in counts between SHR and Wistar was calculated. 45 Data points represent log 2 (fold change)±SEM. There was no significant difference in the levels of mRNA for Adrb1, Adrb2, Adra2a, or Th between strains or between age groups by qRT-PCR although the trend for a reduction in Adrb2 expression remained.
Sandwich ELISAs were used to quantify the relative protein expression of β 1 AR and β 2 ARs in postganglionic sympathetic neurons obtained from 3-to 5-week-old normotensive pre-SHR and Wistar rats ( Figure 1B; 32 stellates, 16 rats/ group, pooled) or 19-to 20-week-old SHR and age-matched Wistar rats ( Figure 1C; 20 stellates, 10 rats/group, pooled). The ELISA assays were biologically powered where 20 to 32 stellates were used per sample; however, the stellates tissue was pooled to obtain an adequate protein concentration for the ELISA assays, therefore no statistical comparisons were made. Data points indicate mean±SEM (of 3-4 technical replicates).

βAR-Evoked cAMP Generation, PKA Activity, and [Ca 2+ ] i Are Enhanced in Pre-SHR Neurons
To determine whether the presence of presynaptic βARs on sympathetic ganglia plays a functional role in modulating intracellular cN signaling pathways, we used FRET to quantify the relative levels of cAMP and PKA (protein kinase A) activity in response to a relatively nonselective βAR agonist, isoprenaline. To assess whether βAR-mediated cAMP generation facilitated signaling via the canonical cAMP-PKA-Ca 2+ pathway, we used the loss-of-FRET sensor Epacs1-H187 (EpacH187) 46 to measure changes in intracellular cAMP. Isoprenaline administration at 10 nmol/L led to significantly greater cAMP generation in pre-SHR (55.6%±16.8%) compared with that measured in Wistar neurons (7.1%±1.4%; 2-way ANOVA; P<0.0001) that was also observed at higher concentrations of isoprenaline (Figure 2A and 2B). PKA activity was measured using the gain-of-FRET sensor AKAR4. 47 Using the same concentration of isoprenaline (10 nmol/L), we found that PKA activity was significantly higher in pre-SHR (26.1%±5.4%) versus Wistar neurons (4.5%±1.1%; 2-way ANOVA; P<0.0001), which was also observed at 100 nmol/L isoprenaline ( Figure 2C and 2D). Raw YFP (yellow fluorescent protein) and CFP (cyan fluorescent protein) fluorescence traces as emitted from the cytosolic loss-of FRET sensor EpacH187 and the gain-of-FRET sensor AKAR4 in Using RNA sequencing, we identified the presence of β 1 AR (Adrb1), β 2 AR (Adrb2), and α 2A AR (Adra2a) mRNA transcripts (A) from 16-wk-old male Wistar rats (n=4) and SHR (n=4). Adrb2 expression was significantly lower in SHR ganglia compared with Wistar (P. adj=0.00945; Salmon-DESeq2 method). There was no significant difference in the levels of mRNA for Adrb1 or Adra2a between strains or between age groups. Data points represent raw counts±SEM for each transcript (A). ELISAs confirmed protein expression of β 1 AR and β 2 ARs in 3.5-to 5-wk-old rat neurons (32 stellates, 16 animals/group) and 20-wkold neurons (20 stellates, 10 animals/group); however, no statistical tests were conducted as stellates were pooled into a single sample to obtain adequate protein concentrations for the ELISA assays. In young rat stellates (B), the concentration of β 1 AR protein was calculated as 869.1±50.6 pg/mL (Wistar) and 114.2±23.7 pg/mL (pre-SHR). β 2 AR protein expression was calculated as 363.5±43.6 pg/ mL (Wistar) and 82.2±20.0 pg/mL (pre-SHR). In adult rat stellates (C), β 1 AR expression was calculated as 674.1±44.6 pg/mL (Wistar) and 489.4±26.3 pg/mL (SHR). β 2 AR protein expression quantified as 353.3±11.2 pg/mL (Wistar) and 147.4±20.7 pg/mL (SHR). Data points depict mean±SEM of 3 to 4 technical replicates. β 1 AR (516±99.17 pg/mL) and β 2 AR (340±104.3 pg/mL) expression was also detected in stellate ganglia from human donors. Data points represent mean±SEM (6 replicates), from 3 pooled stellates obtained from 2 patients (D). Immunocytochemistry depicts β 1 AR (E) and β 2 AR (F) expression on TH (tyrosine hydroxylase)-positive neurons from 4-wk control rats. White arrows demonstrate the localization of β 2 AR on synaptic terminals. response to isoprenaline (10-100 nmol/L) are presented in the online-only Data Supplement ( Figure S2A and S2B).
To assess whether isoprenaline-dependent βAR activation enhances intracellular Ca 2+ ([Ca 2+ ] i ), we measured responses to KCl in the absence or presence of isoprenaline. Ca 2+ recordings were obtained using Indo-1AM labeled sympathetic neurons from 4-week pre-SHR and Wistar rats. In pre-SHR stellate neurons, KCl stimulation in the presence of isoprenaline led to significantly higher [Ca 2+ ] i than KCl stimulations alone ( Figure 2E and 2F; P=0.0272). There was significantly higher KCl-evoked [Ca 2+ ] i in the presence of isoprenaline in pre-SHR neurons compared with that recorded in control neurons ( Figure 2E and 2F; P=0.0027). A time-controlled example trace is shown in the online-only Data Supplement ( Figure S2C).

Epinephrine-Synthesizing Enzyme PNMT Is Present in Rat and Human Stellate Ganglia
RNAseq was performed to obtain an overview of the transcriptome in stellate ganglia obtained from 16-week-old male SHR (n=4) and Wistar rats (n=4). We identified the presence of mRNA transcripts-encoding enzymes required for norepinephrine synthesis ( Figure 4A): phenylalanine hydroxylase (Pah), Th, L-DOPA decarboxylase (Ddc), dopamine β-hydroxylase (Dbh). Furthermore, RNAseq identified the presence of the mRNA transcript encoding phenylethanolamine-N-methyltransferase (Pnmt), the enzyme required for the conversion of norepinephrine to epinephrine in both Wistar and SHR stellate ganglia. In the RNAseq data set, Pah and Ddc mRNA transcript expression were also shown to be significantly lower in SHR neurons ( Figure 4B; Figure S3A; P.adj=0.0719; 6.64×10 -15 , Pah, Ddc, respectively). These findings were validated in 4-week Wistar and pre-SHR by qRT-PCR, and a significant ≈4-fold reduction in Pah expression was observed in pre-SHR ganglia ( Figure 4C; P=0.0098). Data were normalized to a control housekeeping gene (B2m), and SHR gene counts were subsequently normalized to Wistar using the ∆∆C T method. Data are presented as Log 2 (fold change). 45 Using the same method, we also confirmed the presence of Pnmt and Th by qRT-PCR in neurons from 16-week Wistar and SHR ( Figure  S3B). There was no significant difference in mRNA expression of either Th or Pnmt between age groups or between phenotypes. ELISA assays confirmed protein expression of TH ( Figure 4D) and PNMT ( Figure 4E) in stellate ganglia from 4-week pre-SHR, 20-week-old SHR, and agematched Wistar rats. The ELISA assays were high-powered biologically, where 20 to 32 stellates were used per sample; however, stellates were pooled to obtain adequate protein concentrations for the ELISA assays, therefore no statistical comparisons were made. Data points indicate mean±SEM (of 2-3 technical replicates).
To assess whether the presence of PNMT in sympathetic stellate ganglia is conserved in higher species, we obtained stellate ganglia from male human donors. qRT-PCR demonstrated the presence of both Th and Pnmt mRNA transcripts in human sympathetic stellate ganglia ( Figure S3C). Data were normalized to a control housekeeping gene B2m using the ∆C T method 45 and expressed as normalized count values (3 patients, 4 stellates). We also used ELISAs to confirm protein expression of both TH and PNMT in human stellate samples ( Figure 4F). Data points represent mean±SEM (2-3 replicates) from 3 pooled stellates obtained from 2 patients.

Epinephrine Is Released From Pre-SHR but Not Wistar Whole-Stellate Ganglia
After the identification of PNMT, we investigated whether epinephrine is released from the whole rat stellate ganglia under basal conditions or with electric field stimulation. We measured significantly greater total norepinephrine content in homogenized Wistar stellates (43.3±2.173 pg; n=8) compared with pre-SHR stellate ganglia (29.82±6.366 pg; n=4; P=0.0294). In the same homogenate, we measured a greater content of epinephrine in pre-SHR ganglia (14.14±5.399 pg) compared with that measured in Wistar stellates (3.937±0.820 pg; P=0.0019), suggesting that a significant amount of norepinephrine is converted to epinephrine in prehypertensive states ( Figure 5A).
We also investigated whether epinephrine is released from rat stellate ganglia with electric stimulation ( Figure 5B). Electrically evoked concentrations of norepinephrine were significantly higher in samples obtained from pre-SHR (4.32±1.523 pg; n=4) versus Wistar ganglia (1.477±0.316 pg; n=8; P=0.0396). Moreover, in the same samples, the concentrations of electrically evoked epinephrine were significantly

Discussion
In this study, we have obtained evidence for β 1 AR and β 2 AR mRNA and protein expression on presynaptic postganglionic sympathetic neurons from human and rat ganglia. We have further demonstrated that in isolated sympathetic neurons, βAR agonists elevate cAMP and activate PKA. The effects were more pronounced in neurons from pre-SHR rats. We also observed that βAR agonists enhanced [Ca 2+ ] i in response to depolarization by high K + in pre-SHR neurons only. In addition, we demonstrate the presence of mRNA and protein expression of PNMT, the enzyme involved in the synthesis of epinephrine in human and rat sympathetic stellate neurons. Moreover, we observed that epinephrine is present in diseased states and is actively released from prehypertensive, but not healthy rat neurons, suggesting preferential switching of neurotransmitter synthesis in disease.
Single or combinatorial administration of β-blockers is a mainstay treatment strategy for diseases caused by sympathetic overactivity, although the precise mechanisms that underpin the long-term beneficial effects are not entirely clear. 12 Current dogma suggests that the observed antihypertensive and cardioprotective effects of β-blockers are mediated through inhibition of cardiac and vascular βARs, reducing myocardial work and total peripheral resistance. 52 Our findings suggest that the efficacy of clinical β-blockers may be attributed, at least in part, to a reduction in sympathetic hyperactivity and neurotransmission at the end-organ.
What is the cause for increased sympathetic neurotransmission before the onset of neurogenic hypertension? Emerging evidence suggests that impaired nitric oxide synthesis and reductions in cGMP-PKG (protein kinase G) signaling lead to pathological increases in [Ca 2+ ] i and norepinephrine release at the end-organ. 31,53 Recently, we demonstrated that decreased cGMP signaling leads to enhanced N-type Ca 2+ channel (Ca v 2.2) currents and that this effect may be ameliorated by artificially increasing cytosolic cGMP. 54,55 cN signaling is acutely regulated by phosphodiesterase enzymes, and in early prehypertensive states, phosphodiesterase signaling is impaired, resulting in an imbalance between cAMP and cGMP signaling. 54 We, therefore, sought to ask the question, could high levels of neurotransmitter release act in an autocrine or paracrine fashion to increase neuronal cAMP and potentiate neurotransmission in a feedforward manner?
Although it has been previously reported that presynaptic βARs are present 56 and may be capable of facilitating norepinephrine release in several peripheral autonomic ganglia in rat, guinea pig, cat, rabbit, dog, and human [13][14][15][16][17]19,[57][58][59][60][61][62] , the role of adrenergic signaling within the sympathetic stellate ganglia remains unclear, particularly in disease. In the present study, we confirmed the presence of both β 1 AR and β 2 AR isoforms in stellate ganglia from human and rat and found that activation of βARs on rat sympathetic neuron led to a significantly greater increase in intracellular cAMP generation, PKA activity in pre-SHR compared with control neurons (Figure 2).
To assess whether βAR signaling facilitates cardiac-sympathetic neurotransmission, [Ca 2+ ] i was measured in response to KCl in the absence or presence of isoprenaline. Consistent with the observed increases in βAR-mediated cAMP-PKA signaling in pre-SHR neurons, isoprenaline also increased KCl-evoked [Ca 2+ ] i in prehypertensive states; whereas there was no effect of isoprenaline in control neurons ( Figure 2E and 2F). These data demonstrate that enhanced βAR-mediated signaling in sympathetic neurons contributes to the Ca 2+ phenotype and increases sympathetic transmission. Previous work has demonstrated that the N-type calcium channel is the primary voltage-gated channel responsible for Ca 2+ influx in sympathetic neurons and carries a significantly larger Ca 2+ current in pre-SHR and SHR neurons compared with controls. 55,63 N-type calcium channel activity is differentially regulated by PKA and PKG. 54,55 Therefore, we suggest that the isoprenaline-potentiated increases in [Ca 2+ ] i in pre-SHR neurons primarily occurs as a result of βAR-cAMP activation that increases PKA-dependent phosphorylation of N-type calcium channel (Ca v 2.2).
To establish whether the observed increases in cAMP-PKA activity occur downstream of β 1 AR or β 2 AR signaling, cells were perfused with selective agonists for either β 1 AR or β 2 AR subtypes in the presence of either alternate βAR antagonist. We found that selective activation of β 1 AR or β 2 AR led to significantly greater increases in cAMP in pre-SHR neurons compared with Wistar. Indeed, there was no measureable effect of β 1 AR activation on [cAMP] in normotensive controls (Figure 3). Furthermore, stimulation of pre-SHR neurons with the β 2 AR agonist salbutamol led to cAMP generation that was almost twice as high as β 1 AR-evoked cAMP within pre-SHR neurons, suggesting a dominant role for β 2 AR compared with β 1 AR signaling. To establish whether increased βAR signaling in pre-SHR results from increases in βAR expression, we measured levels of β 1 AR and β 2 AR mRNA via qRT-PCR and RNAseq and quantified protein levels using sandwich ELISAs. Surprisingly, we observed that βAR transcripts and protein expression are reduced in pre-SHR stellates, as well as in aged SHR with established hypertension, 6,[29][30][31]33,35,36,42 compared with agematched Wistar neurons, in a similar manner to that reported in the myocardium. 64, 65 We also report that in healthy ganglia, β 1 AR expression decreases with age, much like in the heart (Figure 1). Together, these data suggest that in diseased states, the potentiating effects of βAR agonists may be mediated through impaired second messengers coupled to cAMP and its effector PKA, probably via impairment of phosphodiesterases to hydrolyze cAMP 54,55 rather than the G-protein coupled receptors themselves.
Which neurotransmitter preferentially activates presynaptic βARs? Several studies suggest limited involvement of norepinephrine in potentiating presynaptic neurotransmission but argue for a critical role for epinephrine in enhancing release, particularly in patients with essential hypertension 66 or stress disorders. 67,68 The role of epinephrine in the pathogenesis of essential hypertension has been termed the Adrenaline Hypothesis 23 ; however, the origins of local concentration of epinephrine remain unclear. Most reports suggest that high circulating plasma epinephrine concentrations arise from the adrenal medulla with active reuptake into sympathetic nerve terminals. 19,22,23,[69][70][71] Others have identified heightened epinephrine synthesis within the central nervous system, specifically the nucleus tractus solitariuus 72 and hypothalamus 72,73 and suggest that this source of epinephrine may underpin the high plasma levels of epinephrine. Alternatively, some reports have identified in situ epinephrine synthesis within various sympathetic ganglia in rat and human, via a stress-inducible mechanism. 19,24,66,72,[74][75][76][77] Our identification of PNMT mRNA and protein expression in human and rat cardiac-sympathetic ganglia (Figure 4) supports the findings of these earlier studies that epinephrine is synthesized in sympathetic stellate ganglia in disease.
What is the relevance of epinephrine synthesis in prehypertensive sympathetic stellate ganglia? The Adrenaline Hypothesis of hypertension proposes that stress and subsequent small incremental increases in epinephrine plays a major role in the pathogenesis of hypertension, not via epinephrine directly, but as a result of increased sympathetic activity and enhanced norepinephrine release. 23 This sustained increase in sympathetic activity caused by epinephrine leads to the development of hypertension. We have shown that epinephrine is synthesized in pre-SHR to a greater extent than in healthy sympathetic stellate ganglia and is only released from pre-SHR ganglia. Importantly, epinephrine has a 10-fold higher affinity for β 2 AR than norepinephrine (EC 50 5.2, 53.7 nmol/L, respectively) and is capable of generating 3× more cAMP than norepinephrine via β 2 AR activation, a feature that may be mimicked by isoprenaline because of similarities in efficacy. 78 Subsequently, the high efficacy of epinephrine (and the relatively low efficacy of norepinephrine) at β 2 AR has been shown to result in epinephrine-dependent norepinephrine transmission. Indeed, low concentrations of epinephrine (0.1-10 nmol/L) have been shown to be 100× to 500× more potent than norepinephrine in enhancing activitydependent norepinephrine release. 19,[79][80][81] Moreover, epinephrine may have a more sustained effect on norepinephrine release because of the extended tissue half-life of epinephrine. 79 Epinephrine-induced norepinephrine release has been identified in a wide variety of peripheral tissues in rat, rabbit, and human. 18,19,22 We sought to investigate whether the presence of presynaptic PNMT plays a functional role in converting norepinephrine to epinephrine in rat stellate ganglia, by measuring total catecholamine content and electrically evoked catecholamines by high-pressure liquid chromatography coupled to electrochemical detection ( Figure 5). We identified a significant decrease in total norepinephrine content in pre-SHR ganglia (74.8% of total catecholamine content) compared with norepinephrine calculated in Wistar ganglia (91.5% of total catecholamine content). We have also observed that the total content of epinephrine was significantly higher in pre-SHR ganglia (25.2% of total catecholamine content) compared with epinephrine levels quantified in Wistar ganglia (8.5% of total catecholamines measured). Furthermore, we found that on electric stimulation, the percentage ratio of norepinephrine:epinephrine released from Wistar ganglia was calculated as 91%:9%; whereas in pre-SHR ganglia, the ratio of catecholamines released (norepinephrine:epinephrine) was 44%:56% ( Figure 5), although the total amounts of catecholamines released during electric stimulation remained fairly similar between the strains (≈11-12 pg).
One recurrent feature in human and animal models of hypertension is the reduction in norepinephrine reuptake transport (NET), leading to larger and more sustained extracellular catecholamine concentrations. 24,[82][83][84] Recently, it has been proposed that PNMT may also act as a DNA methylase, silencing NET transcription that may underpin the observed NET phenotype. 24 We have previously identified reductions in NET activity in the pre-SHR cardiacstellate ganglia. 82

Limitations
In this study, we investigated the role and mechanisms involved in feed-forward presynaptic signaling in the cardiac-sympathetic ganglia. We performed a hypothesis neutral, nonbiased approach to sequencing the transcriptome of sympathetic stellate in adult rats that revealed the presence of RNA transcripts involved βAR receptor expression and epinephrine synthesis. We assessed the functional relevance of these findings by probing the adrenergic intracellular signaling pathways coupled to Ca 2+ -mediated exocytosis. There were several limitations to these approaches. First, the stellate ganglion comprises a heterogeneous population of cell types. Indeed, we identified markers of fibroblasts and astrocytes, including vimentin and glial fibrillary acidic protein, respectively; however, we identified that a high number of transcripts were neuronal in phenotype. We also found that the subunit profile of nicotinic acetylcholine receptors matches those described for sympathetic neurons. 85 Moreover, immunocytochemistry highlighted the localization of βand αARs on the soma and dendrites of TH-positive neurons. In support of these data, our collaborators have also identified the presence of transcripts encoding presynaptic βARs in sorted sympathetic mouse neurons (Ana Domingos, personal communication, 2018). Second, in the absence of cardiac tracing experiments, we rely on anatomic literature, [25][26][27] and our own previous observations 32,[86][87][88] that the results presented here are relevant to cardiac-sympathetic communication because significant myocardial sympathetic innervation has been shown to arise from the cervicothoracic ganglia. Third, we used stellates obtained from male rats. Although sex differences in hypertension and cardiovascular disease incidence have been widely reported, 89 in this study, we focused on investigating the transcriptome of the male rat stellate ganglia given that the prevalence for cardiovascular diseases is significantly higher in males than premenopausal women. 89 Fourth, cNs and phosphodiesterases reside in distinct subcellular compartments, and their localization with βARs receptors is acutely regulated. [90][91][92] Similarly, the regulation of Ca 2+ channels by PKA/PKG occurs in distinct signalosomes, conferring site-specific regulation of Ca 2+ entry coupled to neurotransmission. Furthermore, the rate of phosphodiesterase hydrolysis is critically dependent on the concentration of both cAMP and cGMP that is reported to be different between cell types. 93 In this study, we measured global cytosolic cAMP, PKA, and Ca 2+ concentrations, therefore we cannot ascertain precisely where the key pathways converge. Site-specific FRET and Ca 2+ sensors will be required to resolve the question of microdomain impairments in cN and effector signaling.

Perspectives
Our data here demonstrate that in prehypertensive and hypertensive states, epinephrine is synthesized within presynaptic sympathetic nerve terminals and released on activation at the end-organ. In prehypertensive states, evoked release of epinephrine (that is exacerbated by decreased NET activity) may act preferentially on presynaptic β 2 ARs to increase cAMP generation and PKA activity, thereby enhancing Ca 2+ levels and neurotransmission in disease, in a manner akin to positive feedback. We suggest that epinephrine release at the end-organ may play a role in the pathogenesis of hypertension. Figure 6 depicts the model signaling pathways in healthy and prehypertensive states and highlights potential sites for neural phenotypic targeting in disease. We suggest that in a model of early hypertension, activation of presynaptic βARs enhances cAMP generation, PKA activity, and [Ca 2+ ] i to greater levels than that measured in healthy neurons, facilitating both norepinephrine and epinephrine release. Presynaptic activation of β 2 ARs (and β 1 ARs to a lesser extent) further enhances neurotransmission in a potentiating feed-forward manner, with activation of α 2 ARs playing a role in negative feedback. Additional studies aimed at investigating the relative roles of epinephrine and norepinephrine in positive and negative feedbacks signaling may be of therapeutic relevance. Indeed, these findings may have implications beyond neurogenic hypertension and may offer benefit in other diseases of sympathetic overactivity, such as modulation of Figure 6. Model figure. The sympathetic stellate ganglia (cervicothoracic ganglia) are located alongside vertebrate T1 to T3. They are the primary sympathetic ganglia that innervate the heart and have been shown to exert the greatest control over increases in heart rate and contractility. [25][26][27][28] In healthy postganglionic sympathetic neurons (A), Ca 2+ -dependent exocytosis facilitates the release of norepinephrine (NE) onto cardiac myocytes, where postsynaptic β 1 -and β 2 -adrenergic receptors (ARs) are activated. Increases in extracellular NE acts on presynaptic α 2 ARs, reducing adenylyl cyclase (AC) activity through activation of inhibitory Gαi G-proteins. Acute regulation of cAMP is maintained by phosphodiesterases (PDEs). 90,91 cAMP-dependent PKA (protein kinase A) activity increases intracellular Ca 2+ ([Ca 2+ ] i ) via phosphorylation of the N-type Ca 2+ Channel (ICaN; CaV2.2) 55 ; regulation of endoplasmic reticulum stores and mitochondrial Ca 2+ release. 31 In neurons obtained from the prehypertensive SHR, a young genetic model of hypertension (B), Ca 2+ -dependent exocytosis facilitates the release of NE and epinephrine (Epi). Activation of presynaptic βARs in prehypertensive states [29][30][31][32][33][34][35][36] enhances cAMP generation, PKA activity, and [Ca 2+ ] i to greater levels than in healthy neurons, facilitating neurotransmission in a potentiating feed-forward manner. This occurs preferentially via β 2 AR activation. Catecholamines may also be supplied from the circulation. We propose that β-blockers may have efficacy at βARs expressed on peripheral neurons, by reducing cardiac-sympathetic communication in hypertension and dysautonomias. ACh indicates acetylcholine; and nAChRs, nicotinic acetylcholine receptors.