Somatostatin 2 Receptor Activation in the Rostral Ventrolateral Medulla Does Not Mediate the Decompensatory Phase of Haemorrhage

Abstract

Decompensation, a critical phase in the response to hemorrhage, is characterized by profound sympathoinhibition and the overriding of baroreflex mediated compensation. As sympathoexcitatory neurons of the rostral ventrolateral medulla (RVLM) maintain vasomotor tone and are essential for sympathetic baroreceptor reflex function, the RVLM is the likely mediator. However, how decompensation occurs is a mystery. Our previous work demonstrated that the inhibitory neuropeptide somatostatin (SST), evokes potent sympathoinhibition. Here we test the hypothesis that, in response to hypovolemia, SST in the RVLM evokes sympathoinhibition, driving decompensation and suppressing baroreflex compensation. We evaluated neuronal activation at sites that contain SST mRNA and project to the RVLM and, in SST2A expressing neurons in the RVLM. We determined the effects on cardiovascular and sympathetic responses to haemorrhage, of bilateral blockade of SST2 receptors in both the RVLM and A1 regions. Haemorrhage in conscious rats evoked c-Fos immunoreactivity in the amygdala, periaqueductal gray, and parabrachial nuclei, regions previously associated with hemorrhage, shown to contain SST and project to the RVLM. Although c-Fos labeling was found throughout the ventrolateral medulla, only a small subset of RVLM SST2A receptor expressing neurons were activated, consistent with the idea that these neurons are inhibited during hemorrhage. However, SST2 receptor antagonists bilaterally injected in the RVLM or the A1 region did not affect the decompensation response to hemorrhage. Thus somatostatin in the RVLM does not mediate decompensation. The physiological role associated with somatostatin-induced sympathoinhibition in the RVLM together with the central mechanisms responsible for decompensation remain elusive.

INTRODUCTION

Persistent rapid blood loss (hemorrhage) initiates a cascade of neurohumoral and hemodynamic events, resulting in three distinct phases of response: compensation, decompensation, and recompensation ^(1–3). Initial hypovolemia is compensated by baroreflex mediated increase in vasomotor tone and heart rate, in an effort to maintain blood pressure in defense of decreasing cardiac output ^{(2, 4)}. Following a continuing decline in blood volume, the second phase (decompensation) is initiated by withdrawal of sympathetic tone to the heart and blood vessels and increased cardiac vagal tone ⁽⁴⁾. Decompensation is likely triggered by activation of cardiopulmonary vagal ⁽⁵⁾ and cardiac spinal afferents ⁽⁶⁾, which provide cardioprotection against poor diastolic filling ⁽⁷⁾. If blood loss ceases, recompensation occurs via sympathetic reactivation, promoting return to normotensive levels of blood pressure ⁽⁸⁾.

Sympathetic tone to the heart and blood vessels is regulated by neural and humoral mechanisms. Sympathoexcitatory premotor neurons in the rostral ventrolateral medulla (RVLM) are highly baroreceptor sensitive and tonically active with this activity crucial for the maintenance and stability of vasomotor tone and blood pressure. Serotonin and opioid actions in the RVLM indicate that these neurons contribute to decompensation following hemorrhage ^(9–12). As profound sympathoinhibition and suppression of baroreceptor-mediated compensation are characteristic of decompensation, inhibition of RVLM neurons is a likely mechanism. During hemorrhage, RVLM sympathetic premotor neurons show reduced neuronal discharge ⁽⁹⁾, directly correlated with the low expression of the early oncogene c-Fos ^{(10, 13)}.

The somatostatin 2A receptor is abundantly expressed on RVLM sympathoexcitatory premotor neurons ^{(14, 15)}. RVLM microinjection of the inhibitory neuropeptide somatostatin (SST) elicits intense sympathoinhibition and bradycardia akin to that seen during decompensation, mediated by SST2 receptors ⁽¹⁴⁾ via activation of GIRK channels on premotor neurons ⁽¹⁶⁾. Few neurochemicals evoke sympathoinhibition when injected into the RVLM ⁽¹⁷⁾, and it is difficult to identify other physiological mechanisms that would underlie such a debilitating effect. Somatostatinergic projections to the RVLM ⁽¹⁶⁾ do include nuclei implicated in the autonomic response to hemorrhage such as the central nucleus of the amygdala, ventrolateral peraquaductal gray, and lateral parabrachial nucleus ^{(8, 18, 19)}. As SST is not tonically active in the RVLM ⁽¹⁴⁾, the possibility exists that it is released in the RVLM in response to hemorrhage. We test the hypothesis that in response to hypovolemia, SST in the RVLM evokes sympathoinhibition driving the decompensation phase of hemorrhage and SST arises from neurons in the central amygdala, ventrolateral periaqueductal gray, and/or lateral parabrachial nucleus.

To achieve this we first determined whether hemorrhage in conscious animals, induced c-Fos expression within regions known to both project to the RVLM and express SST mRNA. Secondly, we determined the relationship between hemorrhage induced c-Fos expression and SST2A receptor expression throughout the ventrolateral medulla (VLM). Next we examined, in anaesthetized rats, the effect of selective SST2 receptor antagonists bilaterally injected into the RVLM and a more caudal site on the sympathetic responses to hemorrhage.

MATERIALS AND METHODS

Animal welfare and ethical approval

All experiments conducted were in accordance with the Australian Code of Practice for the Care and Use of Animals for Scientific Purposes. Experiments were carried out with the approval of the Macquarie University Ethics Committee (2011/057, 2013/044). Animals were housed under constant 12 h light/dark cycles and allowed standard rat chow ad libitum.

Conscious hemorrhage: cannulation and blood withdrawal

Male Sprague–Dawley rats (n = 14) were deeply anaesthetized with a mixture of ketamine/medetomidine (0.75 and 0.5 mg/kg, respectively, i.p., Norbrook Pharmaceuticals, Melbourne, Australia) and then received cephazolin (200 mg/kg, i.m., Norbrook Pharmaceuticals, Australia) and carprofen (2.5 mg/kg, s.c., Norbrook Pharmaceuticals, Australia). The right femoral and carotid arteries were exposed using a ventral approach, and heparinized saline filled catheters (polyethylene, ID 0.58 mm, OD 0.96 mm, Sterihealth, Sydney, Australia) inserted, with the capped end tunneled under the skin emerging between the scapula, to enable measurement of arterial pressure via a flexible connection to a pressure transducer, and allow arterial blood withdrawal. Rats were allowed to recover in their home cages for 48 h prior to hemorrhage. Carprofen (2.5 mg/kg) was administered every 12–16 h after surgery.

Rats were randomly allocated as either hemorrhage or sham-operated controls. On the day of experimentation, carotid or femoral catheters were connected for recording arterial pressure while rats remained in their home cage undisturbed for 60 min. Blood withdrawal was begun at a rate of 0.6 mL/min for 10 min. Sham rats were connected for recording but had no blood withdrawn. All rats were then allowed to recover undisturbed for 120 min before being euthanized with an overdose of sodium pentobarbital (100 mg/kg, i.p.) and transcardially perfused with ice-cold saline followed by 4% paraformaldehyde (PFA; 4% PFA, 76.7 mM Na₂HPO₄, 26.6 mM NaH₂PO₄, pH 7.4) and the brain was removed and stored in PFA overnight at 4°C. The brain was then placed in cryoprotectant (Cryo; 500 μM polyvinylpyrrolidone, 76.7 mM Na₂HPO₄, 26.6 mM NaH₂PO₄, 876 mM sucrose, 5 mM ethylene glycol, Sigma-Aldrich, Sydney, Australia) and kept at −80°C until sectioning.

Immunohistochemistry

Coronal brain sections were cut using a vibrating microtome (Leica VT 1200S; Leica Biosystems, Sydney, Australia) into 40 μm serial sections collected in five series. Free-floating sections were permeabilized in 50% ethanol for 30 min at RT and washed in tris-phosphate-buffered saline (PBS). Sections were incubated in primary antibodies (diluted in 10% normal horse serum in tris-PBS) for SST2A (guinea pig anti-SST2A, 1:1,000, Biotrend, Cologne, Germany) and c-Fos (mouse anti-c-Fos, 1:500, Santa Cruz Biotechnology Inc., Dallas, Tex) for 48 h at 4°C. Sections were then incubated in fluorescent conjugated secondaries (Cy5-conjugated donkey anti-guinea pig IgG, 1:250 and 488-conjugated donkey anti-mouse IgG, 1:250 respectively, Jackson ImmunoResearch, West Grove, Pa) overnight at 4°C and then washed, mounted on glass slides using Dako mounting medium, and visualized using a light microscope (Zeiss, Axio Imager Z.2, Oberkochen, Germany). Images were acquired using ZEN 2012 imaging software (Zeiss, Germany). High-power images were taken using a confocal microscope (Leica TCS SP8, Nussloch, Germany) and acquired using Leica Application Software AF (LAS AF; Leica, Nussloch, Germany). All images were imported and analyzed using the ImageJ plugin Fiji.

Hemorrhage under anesthesia

Surgery and electrophysiological recordings

Male Sprague–Dawley rats (n = 23) were anaesthetized with urethane (1.3 g/kg, i.p. 10% solution in saline, Sigma-Aldrich). Core temperature was maintained between 36.5 and 37°C using a homeothermic regulated heating blanket (Harvard Apparatus, Holliston, Mass). Both femoral arteries (left and right) and the right femoral vein were cannulated using heparinized saline-filled catheters (polyethylene, ID 0.58 mm, OD 0.96 mm; Sterihealth, Australia) for measuring arterial pressure, withdrawal of blood and administration of drugs and saline, respectively. The trachea was intubated for artificial oxygen-enriched ventilation with end tidal CO₂ maintained between 3.5% and 4.5% following neuromuscular blockade (pancuronium bromide, 0.8 mg/kg, i.v. administered every 60 min). In order to assess the level of anesthesia the pinch reflex was tested every 30 min by monitoring sympathetic, heart rate and blood pressure responses. When a response in any parameter was evoked additional anesthetic (0.13 g/kg i.v.) was administered. It should be noted that pancuronium bromide is vagolytic, blocking only efferent vagal traffic by muscarinic receptor blockade ⁽²⁰⁾. The protocols associated with the use of the paralyzing agent, pancuronium bromide were reviewed, approved, and monitored by the Macquarie University Animal Ethics Committee. Animals were then placed in a stereotaxic frame and the left greater splanchnic nerve isolated, cut, and prepared for recording, using bipolar silver wire electrodes as described previously ⁽¹⁴⁾. Nerve recordings were amplified (×10,000, CWE Inc., Ardmore, Pa), band-pass filtered (0.1–3 kHz), sampled at 5 kHz (CED Micro 1401, Cambridge Electronic Design; CED Ltd, Cambridge, UK) and recorded using Spike2 software (CED, Ltd).

RVLM and A1 region microinjection

Following removal of the occipital plate, the dorsal surface of the medulla was exposed and RVLM and A1 region were targeted for injection of the peptide antagonist. Pressor responses obtained following l-glutamate microinjections (50 nL, 100 mM, Sigma-Aldrich, dissolved in PBS [pH 7.4]) greater than 35 mmHg were selected as the RVLM target sites ⁽²¹⁾. The SST2 receptor antagonist BIM-23627 (1.0 mM, Bachem, Bubenderf, Switzerland) was dissolved in 5% dimethylsulphoxide (Sigma-Aldrich) with PBS and microinjected (100 nL) bilaterally into either the RVLM (n = 6) or A1 region (n = 5). Previous reports from our laboratory demonstrated only small transient pressor effects following microinjection of dimethylsulphoxide/PBS vehicle into the RVLM ⁽¹⁴⁾.

Fixed pressure hemorrhage and baroreceptor reflex stimulation

Arterial blood withdrawal began 10 min after drug microinjection initially at a rate of 0.6 mL/min for 10 min to a fixed level of mean arterial pressure (MAP) set at 30 mmHg, then blood was removed or replaced periodically (∼ 1 mL blood) to maintain this pressure for up to 60 min as described previously ⁽²²⁾. In a separate subset of experiments (n = 3), before and 10 min following injection of BIM-23627 the baroreceptor reflex was assessed using the vasoactive drugs; phenylephrine (10 μg/kg, i.v., Sigma-Aldrich) and sodium nitroprusside (10 μg/kg, i.v., Sigma-Aldrich).

At the end of recording, rats were euthanized with an overdose of sodium pentobarbital (80 mg/kg, i.v.). Brains were removed and drop fixed in 4% PFA for confirmation of microinjection sites, identified by the pipette track.

Data analysis

Neurograms were amplified, rectified, and smoothed (sSNA, 1 s time constant). Time course analysis was obtained from data averaged over 5 min intervals, from 10 min prior to blood withdrawal to 60 min following blood withdrawal. Peak changes during compensatory, decompensatory, and recompensatory phases of hemorrhage were measured as percentage change from baseline activity (period before any intervention) at 10, 20, and 60 min, respectively. Statistical analysis was performed using GraphPad Prism 6 (Graphpad Prism 6, San Diego, Calif) and data were analyzed using one- or two-way ANOVA with Bonferroni's correction. Statistical significance was considered at P < 0.05, data are expressed as mean ± SE.

RESULTS

CNS nuclei activated following hemorrhage in conscious rats

Neurons in a number of suprabulbar regions are activated in response to hemorrhage ^{(8, 18, 19)}. Of these, we have previously shown that the central amygdaloidal nuclei (encompassing the central (CeC), lateral (CeL), and medial (CeM) subdivisions, Fig. 1A), ventrolateral periaqueductal gray (vlPAG, Fig. 1B), and lateral parabrachial nucleus (LPBN, Fig. 1C) contain SST mRNA and project directly to the RVLM ⁽¹⁶⁾. We therefore assessed the level of neuronal activation (using c-Fos immunoreactivity) within these nuclei following haemorrhage in conscious rats (Fig. 1).

Fig. 1: Brain regions that express SST mRNA and project to the RVLM are activated by haemorrhage in conscious animals.

Sham-operated controls (n = 4) showed little or no c-Fos immunoreactivity in the regions assessed (Fig. 1, Ai, Bi, and Ci). Following conscious hemorrhage, c-Fos immunoreactivity was consistently present within; the central amygdaloidal nuclei (predominately found in the CeC, with sparse labeling present within the CeL and CeM (Fig. 1Aii, n = 4), the basolateral amygdala (BLA, Fig. 1Aii, n = 4), the vlPAG (Fig. 1Bii, n = 4), together with some labeling in the dorsolateral PAG (dlPAG, Fig. 1Bii, n = 4) and the LPBN (Fig. 1Cii, n = 1).

Most SST2A expressing neurons in the VLM are not activated following hemorrhage

We next determined whether somatostatin 2A receptor (SST2A) receptor expressing neurons throughout the VLM, were activated following conscious hemorrhage. The region encompassed by dashed lines in Figure 2A depicts the area analyzed: lateral border, the spinal trigeminal nucleus (sp5), dorsal border, the dorsal boundary of nucleus ambiguus (AmbC) where possible, and the medial border the inferior olive (IO). SST2A immunoreactive neurons were present throughout the rostrocaudal extent of the VLM (Fig. 2D). Within the RVLM (Fig. 2A) hemorrhage evoked c-Fos expression (Fig. 2Bi, green and Fig. 2C) with SST2A immunoreactive neurons present in the same region (Fig. 2Bii, magenta and Fig. 2D); however, only some SST2A immunoreactive cells expressed c-Fos (Fig. 2D). More c-Fos immunoreactive neurons were present throughout the VLM following hemorrhage compared with control (Fig. 2C, P < 0.0001, two-way ANOVA, n = 6 and n = 5, respectively) with many seen caudally, particularly at the level of the noradrenergic A1 cell group (Fig. 2C, Bregma: −14 mm) and rostrally at the level of RVLM or rostral C1 region (Fig. 2C, Bregma: −12 mm). Despite many neurons expressing SST2A throughout the VLM (Fig. 2D, magenta circles, n = 6) only a small population expressed c-Fos following hemorrhage (Fig. 2D, open circles, n = 6) however this population was significantly greater than that seen in sham-operated controls (Fig. 2D, closed circles, P < 0.0001, two-way ANOVA, n = 6).

Fig. 2: Most sst2AR-ir neurons in the ventrolateral medulla are not activated following hemorrhage.

SST2 receptor blockade in the RVLM does not affect the sympathoinhibitory phase of hemorrhage

Figure 3A shows, in an anaesthetized rat, the heart rate and sympathetic response to fixed pressure hemorrhage during which blood was removed until MAP reached a level of 30 mmHg and was replaced or removed to maintain 30 mmHg pressure. Grouped data show the responses evoked in controls and following bilateral microinjection of the SST2 receptor antagonist BIM-23627 into the RVLM in Fig. 3B and C.

Fig. 3: Bilateral injections of the SST2 receptor antagonist BIM-23627 in the RVLM do not alter decompensation.

Bilateral microinjection of BIM-23627 into the RVLM (sites depicted in Fig. 3D, n = 6) using a dose we previously have shown to block SST induced sympathoinhibition in the same region (1.0 mM, see Burke et al. ⁽¹⁴⁾) caused small transient sympathoexcitatory responses that returned to baseline within 5 min (Fig. 3Biii). Hemorrhage was initiated 10 min after injection of BIM-23627. As the rate of blood withdrawal remained constant and MAP was maintained at 30 mmHg, MAP levels were similar in BIM-23627 treated and control animals (Fig. 3Bi, P > 0.05, two-way ANOVA, n = 6). The HR response to hemorrhage was unaffected by BIM-23627 microinjection into the RVLM (Fig. 3Bii, P > 0.05, two-way ANOVA, n = 6). Surprisingly, the sympathoinhibition, characteristic of decompensation (10–20 min), was unaffected following BIM-23627 in the RVLM (Fig. 3C, control, −37.2 ± 4.9%, vs. BIM-23627, −33.4 ± 7.6% P > 0.05, one-way ANOVA, n = 6), although both sympathoexcitatory phases, compensation (0–10 min) (Fig. 3C, control, 19.3 ± 2.3% vs. BIM-23627, −3.7 ± 3.7%, P < 0.05, one-way ANOVA, n = 6), and recompensation (20–60 min) (Fig. 3C, control, 52.1 ± 5.7%, BIM-23627, 21.5 ± 10.5% P < 0.0001, one-way ANOVA, n = 6) were attenuated. Over all the inhibitory phase was unchanged by BIM-23627, but both excitatory phases of the SNA response to hemorrhage were attenuated (Fig. 3Biii, P < 0.0001, two-way ANOVA, n = 6).

We examined the effects of BIM-23627 bilaterally in the RVLM on baroreceptor reflex gain as the compensatory phases of hemorrhage are dependent on baroreflex mechanisms and the RVLM is integral to the sympathetic baroreceptor reflex pathway ^(23–25). Sympathetic baroreflex gain was not significantly altered by BIM-23627 (control −1.3 ± 0.2 %sSNA/mmHg vs. BIM-23627, −1.4 ± −0.1 %sSNA/mmHg, P > 0.05, paired-t-test, n = 3).

SST in the A1 region does not contribute to sympathetic responses evoked during haemorrhage

Within the region containing the A1 cell group (Fig. 4A), some SST2A immunoreactive neurons (magenta) were activated (c-Fos-ir, green) following conscious haemorrhage (Figs. 2D and 4Ai) so we next investigated whether SST receptors here influence the sympathetic responses to hemorrhage. Microinjection of BIM-23627 into the A1 region in the anaesthetized rat (sites depicted in Fig. 4D, n = 5) did not significantly alter the responses to hemorrhage in MAP (Fig. 4Bi, P > 0.05, two-way ANOVA, n = 5), HR (Fig. 4Bii, P > 0.05, two-way ANOVA, n = 5) or sSNA (Fig. 4Biii, P > 0.05, two-way ANOVA, n = 5) and therefore did not affect compensation (Fig. 4C, control, 19.3 ± 2.3% vs. BIM-23627, 9.0 ± 6.3%, P > 0.05, one-way ANOVA, n = 5), decompensation (Fig. 3.4C, control, −37.2 ± 4.9% vs. BIM-23627, −29.4 ± 12.5%, P > 0.05, one-way ANOVA, n = 5), or recompensation (Fig. 4C, control, 52.1 ± 5.7% vs. BIM-23627, 55.5 ± 12.4%, P > 0.05, one-way ANOVA, n = 5).

Fig. 4: Bilateral injections of BIM-23627 in the A1 region do not alter sympathetic or cardiovascular responses to hemorrhage.

DISCUSSION

We report that the amygdala, periaqueductal gray, and parabrachial nucleus, regions which express SST and project to the RVLM ⁽¹⁶⁾ were activated during hemorrhage, and only a minority of SST2A expressing neurons in the RVLM expressed c-Fos immunoreactivity. Previously we demonstrated that SST2A activation in RVLM leads to profound sympathoinhibition ⁽¹⁴⁾. Together, these observations supported the idea that SST release in the RVLM may inhibit sympathoexcitatory SST2A expressing RVLM neurons during decompensation. However, contrary to our hypothesis, we find that SST2A receptor activation in the RVLM does not play a role in the decompensation phase of hemorrhage, although it may facilitate compensation and recompensation. Furthermore, SST2A receptor blockade in the A1 region where SST2A expressing neurons were activated by hemorrhage also did not alter the sympathetic responses to hemorrhage.

There are numerous experimental models of hemorrhage (for review see ⁽²⁶⁾) with fixed pressure, fixed volume, and uncontrolled hemorrhage models the most common. To test the hypothesis here a highly reproducible and reliable model was required so a fixed pressure model was selected ⁽²²⁾ as fixed volume (dependent upon body weight, which does not match blood volume) and uncontrolled models provide variable time frames, intensities of hemorrhagic shock and exclude the possibility of exploring resuscitation. The model used here reflects clinically realistic durations of severe shock ⁽²²⁾.

Hemorrhage activates neurons (characterized by c-Fos immunoreactivity) in multiple forebrain, midbrain, and pontine nuclei ^{(8, 18, 19)}. Of these, the central nucleus of the amygdala (CeA), vlPAG, and LPBN contain RVLM projecting neurons, which harbor the potential for SST production (contain SST mRNA ⁽¹⁶⁾). We confirm activation of neurons in these regions following hemorrhage; however, precisely when these neurons are activated or whether their activation results in release of SST in the RVLM remains unknown. These regions appear to participate during different phases of hemorrhage, with the amygdala implicated in compensation ⁽²⁷⁾, the vlPAG in decompensation ^{(28, 29)} and LPBN in recompensation ⁽⁸⁾.

We confirm that neuronal activation within the ventral medulla was present with most c-Fos expression found in the rostral (RVLM) and most caudal (A1 region) parts of the VLM ^{(10, 19, 30)} and, as demonstrated previously, these neurons are commonly catecholaminergic ⁽³¹⁾. There are limitations for the use of c-Fos in some cell groups (e.g., motor neurons) and under some stimulus conditions. A range of other immediate early genes can be used to identify activated neuronal populations, including ARC together with a range of phosphorylated proteins such as p-CREB and p-ERK. However, it is unlikely that alternate markers would change any data presented here as expressions commonly overlap (e.g., ARC and c-Fos ⁽³²⁾). What would be useful is the development of a marker of neurons, which have been inhibited. No such marker is currently available. Although many presympathetic RVLM neurons are inhibited during hemorrhage ⁽⁹⁾, it was not surprising to see some activated SST2A receptor expressing neurons (mostly catecholaminergic) as sympathetic targets such as the adrenal medulla are activated during decompensation ⁽³³⁾ and neurons are also likely activated during compensation ^{(9, 10)}. A1 neurons contribute to vasopressin secretion in response to hemorrhage participating in the compensatory phases ⁽³⁴⁾. Our data indicate that SST in the A1 region does not contribute to the sympathetic response to hemorrhage; however, any effect on vasopressin secretion or other humoral agents remains to be determined.

We confirm the expression of SST2A containing neurons throughout the ventral medulla ⁽¹⁵⁾, and demonstrate that most of these neurons were not activated following haemorrhage in line with our hypothesis that SST released in the RVLM inhibits neurons contributing to sympathetically mediated decompensation. However, our hypothesis is disproved, as BIM-23627 injected into the RVLM failed to alter decompensation diminishing only compensation and recompensation. Precisely in what state or under what circumstance SST evoked sympathoinhibition occurs in the RVLM remains to be determined. The results suggest that haemorrhage induced SST release in the RVLM may directly or indirectly attenuate the compensatory baroreceptor reflex pathway within the RVLM ^(23–25). Direct effects are unlikely as SST2 receptor antagonists alone in the RVLM did not alter sympathetic baroreceptor reflex gain, and exogenous application of either SST or lanreotide (SST2 selective analogue) to the RVLM do not alter aortic-depressor nerve (ADN) stimulation evoked baroreflex-mediated reductions in sSNA ⁽¹⁴⁾. As SST2A receptors are not found on GAD expressing cell bodies in the RVLM ⁽¹⁴⁾ a presynaptic effect of SST during hemorrhage is possible although effects at SST2B receptor expressing cells also present in the region ⁽¹⁴⁾ cannot be discounted.

Perspectives

The sympathoinhibitory peptide SST in the RVLM evokes sympathoinhibition; however, it does not contribute to decompensation following hemorrhage. The physiological event(s) eliciting release of such a potent agent are unknown. Inhibition of a subset of neurons in the RVLM that maintain sympathetic tone is the likely mediator of decompensation; however, the identity of such neurons remains elusive. Although serotonin and opioids participate in decompensation and the midline medulla is involved ⁽³⁾ the central mechanisms underlying decompensation continue to remain a mystery. Identification of neural circuitry driving decompensation is a crucial goal as is a continuing quest to identify the location of central opioid receptors critical to the commencement of decompensation ⁽³⁵⁾.