Cardiovascular and Respiratory Effects of Endothelin in the Ventrolateral Medulla of the Normotensive Rat
Abstract We studied the relevance of the ventrolateral medulla for the cardiovascular and respiratory effects of endothelin-1 in urethane-anesthetized rats. Microinjection of endothelin-1 into the rostral ventrolateral medulla (RVLM) evoked pressor and bradycardic effects followed by sustained decreases in blood pressure, bradycardia, and respiratory depression. These effects were inhibited by endothelin-A receptor antagonists (BQ-123 and BQ-610) but not by endothelin-B antagonists. In the caudal ventrolateral medulla (CVLM) endothelin-1 decreased blood pressure, renal sympathetic nerve activity, respiratory frequency, and phrenic nerve activity, whereas heart rate increased. Pretreatment with BQ-123 in the CVLM increased respiratory frequency by 15±6 breaths per minute and prevented the effects of intra-CVLM administration of endothelin-1. In separate experiments, the intracisternal administration of endothelin-1 (20 pmol) to rats pretreated with saline in both the RVLM and CVLM resulted in a hypotensive and bradycardic phase that was followed by hypertension (50±15 mm Hg), bradycardia, and 100% mortality. In a separate group, pretreatment with BQ-123 in the RVLM and CVLM completely inhibited the hypotensive phase and reduced by 83% the subsequent rise in blood pressure evoked by endothelin-1. Cardiorespiratory arrest was prevented in all the rats in this group. Selective endothelin receptor blockade in the RVLM attenuated the hypertensive period of intracisternal administration of endothelin-1 and prevented mortality by 33%, whereas in the CVLM the endothelin receptor antagonist inhibited the initial hypotension and reduced mortality by 25%. Our results support the concept that in the ventral medulla, endothelin-1 can modulate cardiovascular and respiratory function.
Endothelin is a potent vasoconstrictive 21–amino acid peptide for which at least three different isoforms have been identified (ET-1, ET-2, and ET-3).1 These isoforms activate specific receptor subtypes that have been classified as ETA (selective for ET-1 and ET-2) and ETB (equally sensitive for all endothelin isoforms).2 3 There is also some evidence for a third receptor, ETC, which seems to have higher affinity for ET-3.4
A growing number of reports indicate that endothelin, in addition to its systemic vascular effects, may also function as a neuromodulator within the central nervous system.5 6 Autoradiographic studies have documented the presence of endothelin binding sites in brain nuclei that, for instance, modulate cardiovascular function. These sites include hypothalamic (ie, the supraoptic nucleus) and lower brain stem medullary centers (ie, the RVLM).7 Importantly, studies using immunohistochemistry,8 Northern blot,9 and in situ hybridization techniques8 9 demonstrated neural as opposed to vascular localization within the brain. In addition, pharmacological studies have indicated specific and selective cardiovascular changes after central endothelin administration.5 10 11 12 13 14 In particular, several reports documented that the RVLM is highly sensitive to the effects of ET-1 administration.5 14 Microinjection of ET-1 into the RVLM increases BP and RSNA. In a high proportion of animals, these cardiovascular events are followed by respiratory impairment, cardiovascular collapse, and death.5 14
The present report, in addition to further delineating the actions of ET-1 in the RVLM, characterizes the effect of microadministration of this peptide on cardiovascular and respiratory function in the CVLM. We also present results from studies demonstrating the relevance of the ventrolateral medulla to the hemodynamic effects of endothelin circulating in the cerebrospinal fluid.
Normotensive Sprague-Dawley rats (Sasco Sprague-Dawley) weighing 300 to 350 g were used. The rats were fed normal rat chow, were given tap water ad libitum, and were housed under a 12-hour controlled light/dark cycle.
All procedures were done in accordance with institutional guidelines and approved by the Vanderbilt University Animal Care Committee. Experiments were done in rats anesthetized with urethane (1.0 g/kg IP plus 300 mg/kg IV). For direct measurement of BP a polyethylene cannula (PE-50) was placed in the femoral artery and connected to a pressure transducer (Gould P23ID) and polygraph (Gould RS3800). HR was monitored continuously by a tachograph preamplifier (Gould 13-4615-65) driven by the BP signal. The rats were then placed in a stereotaxic frame (David Kopf Instruments) for controlled administration of drugs into the cisterna magna or different areas of the ventrolateral medulla.
The dorsal surface of the medulla was exposed by a limited craniotomy for endothelin microinjections into the RVLM or CVLM. Triple-barreled glass microcannulas (1.2-mm OD, 0.68-mm ID; Kwik-Fil, World Precision Instruments) were prepared with an external tip diameter of approximately 40 μm and connected to a nitrogen-pressured, multichannel pneumophoresis pump (PV800, World Precision Instruments). One barrel was prefilled with glutamate (74 pmol/60 nL); another was prefilled with ET-1 (0.25, 0.5, 1.0, 2.0, 4.0, or 6.0 pmol/60 nL) or saline (60 nL); and the third was filled with endothelin receptor antagonists (BQ-123, BQ-610, or BQ-788) or fast green ink. Micropipettes were individually calibrated before insertion into the brain tissue and after withdrawal to deliver a volume of 60 nL over 15 seconds. The diameter of the ejected droplet was measured with a micrometer eyepiece (Acts Instruments Inc). During the experiment microinjection of the substance into the appropriate brain tissue was monitored by observing the movement of the meniscus inside the glass micropipette. The microcannulas were then stereotaxically implanted into the RVLM with coordinates of +2.5 to 2.7 mm anteroposterior, 1.8 to 2.1 mm mediolateral, and 2.4 to 2.8 mm dorsoventral, with the obex used as reference. Within these coordinates the final injection sites within the RVLM were selected by identifying the most sensitive location that could be found for the pressor effects of glutamate as described previously.15 Similarly, for the CVLM the cannulas were first positioned with the use of the following coordinates from the obex: anteroposterior, 0.8 mm; mediolateral, 1.7 mm; and dorsoventral, 2.4 mm; then the most sensitive sites for the hypotensive effects of 74 pmol glutamate were located.
Effects of Endothelin in the CVLM
In a group of 38 rats we determined whether microinjection of different doses of ET-1 into the CVLM affects BP and HR in a dose-dependent manner. The rats were allowed to rest for at least 60 minutes after initial instrumentation. This period was followed by microinjection of 60 nL saline in all the rats (for assessment of volume effects) followed by a randomly selected ET-1 dose (0.25 pmol, n=6; 0.5 pmol, n=4; 1.0 pmol, n=8; 2.0 pmol, n=6; 4.0 pmol, n=4; and 6.0 pmol, n=6). The effects on BP and HR were recorded for the next hour. Only one dose per rat was used in each experiment.
Effects on RSNA
In a group of six rats we investigated the effects of ET-1 microinjection (1 pmol) into the CVLM on RSNA and its relationship with hemodynamic changes. RSNA recordings were obtained as described elsewhere.5 Briefly, the left renal nerve was dissected and placed in situ on bipolar hook electrodes (polytetrafluoroethylene-insulated, 0.003-inch stainless steel; Medwire Corp). The nerve was covered with low-viscosity polyvinylsiloxane dental impression material to electrically isolate the nerve-electrode junction. Multiunit recordings were amplified 100 000 times in two stages by an isolated preamplifier (Gould 11-5407-58) and a universal amplifier (Gould 13-4615-58). Nerve activity was full-wave–rectified, integrated, and expressed in arbitrary units proportional to volts per second by an integrator-amplifier (Gould 13-4615-70). Total sympathetic nerve activity (equal to the sum of efferent and afferent activities plus electrical noise) was recorded, and the signal remaining after administration of hexamethonium (20 mg/kg IV) was assumed to represent electrical noise and afferent activity. The magnitude of this signal was considered to be an estimate of zero efferent nerve activity, and estimates of this activity were obtained by subtracting zero activity from the total recorded activity. To confirm that these recordings represented sympathetic nerve activity, we gave bolus doses of phenylephrine (10 μg IV). In the rats in which the electrodes were placed properly on the renal nerve, phenylephrine inhibited nerve activity by 89±7%. This was assumed to indicate the sympathetic nature of the recorded activity. The rats were then placed in a stereotaxic frame and instrumented for ET-1 microinjections (1 pmol) into the CVLM as described above. The effects on BP, HR, and RSNA were followed during the next 60 minutes.
Effects on Respiratory Function
In different groups of rats we investigated the effects of microinjection of ET-1 into the CVLM on RF and its relationship to hemodynamic changes. After induction of anesthesia the rats were instrumented for BP and HR recordings as described above. The trachea was intubated, and a thermistor probe (YS1 series, model I-8456-00, Cole-Parmer Instrument Co) was placed in the cannulated tracheal tube. Breathing rate was monitored from the temperature fluctuation in the airway where the probe was located. With the use of a Gould tachograph, respiratory rate was derived. After instrumentation a preselected ET-1 dose (0.25 to 6.0 pmol/60 nL) was microinjected into the CVLM, and the effects on RF, BP, and HR were followed as described above.
In an additional group of rats we evaluated the effects of microinjection of ET-1 into the CVLM on arterial blood gases at the time of the maximal cardiovascular effects. This group was instrumented as above with the exception that we placed an additional arterial line in the other femoral artery for blood gas determination. After the rats were placed in the stereotaxic frame, the CVLM was identified and the rats were rested for 30 minutes; a blood sample was taken (0.25 mL) in heparinized capillary tubes, and the same volume was replaced with blood from a donor rat. We determined blood gases in a Corning 158 pH/blood gas analyzer. After BP and HR were stable, the rats received a microinjection of ET-1 (1 pmol/60 nL) into the CVLM. At the peak of the hemodynamic response another sample was obtained for blood gas determination.
Another group (n=6) was used for assessment of the endothelin effects on PNA. In these rats the trachea was intubated and connected to a small-animal ventilator (model 50-1908, Harvard Apparatus); the acromiotrapezius muscle then was divided longitudinally. The phrenic nerve was isolated from the ventral division of the fifth cervical plexus and cut for placement on a bipolar hook electrode. This preparation was covered with low-viscosity polyvinylsiloxane dental impression material to electrically isolate the nerve-electrode junction. These rats were then vagotomized by dissecting and cutting bilaterally the vagus nerve. After completion of the surgery the ventilatory rate was adjusted to a level similar to that of spontaneously breathing animals (80 to 100 breaths per minute) and in which significant activity of phrenic nerve discharge was present. All these rats were initially paralyzed with pancuronium bromide (1 mg/kg IV) and supplemented as needed for suppression of respiratory motor activity. Subsequently, the rats were placed in a stereotaxic frame for microinjections of ET-1 into the CVLM, and the effects on BP, HR, and PNA were followed as described above.
Specificity of the Endothelin Response
To assess the specificity of the endothelin response we studied whether pretreatment with specific receptor antagonists inhibited the effects of this peptide in the CVLM. After rats were instrumented for recording of BP, HR, and RF, they were divided into groups that received either saline (60 nL, n=4) or the ETA receptor antagonist BQ-123 (2 pmol/60 nL, n=4). Fifteen minutes after the antagonist microinjection, ET-1 (1 pmol/60 nL) was microinjected and the effects were followed during the next 60 minutes. A different group of rats (n=6) was pretreated with the tripeptide endothelin receptor antagonist BQ-610, which has been reported to be a potent ETA receptor antagonist.16 Rats were microinjected with BQ-610 (0.03 pmol/60 nL), and after they had recovered from the effects, ET-1 (1 pmol/60 nL) was similarly administered into the CVLM.
To elucidate the involvement of the ETB receptor subtype on the cardiovascular and respiratory effects of ET-1 in the CVLM, we studied a subgroup of rats that were pretreated with the selective ETB receptor antagonist BQ-788.17 Rats were instrumented for recording of hemodynamics and RF and pretreated with 2 pmol/60 nL BQ-788. Fifteen minutes later the rats received a microinjection of 1 pmol ET-1 (n=6), and changes in BP, HR, and RF were followed during the next 60 minutes.
Endothelin Administration Into the RVLM
We have previously reported the effects of endothelin on BP, HR, RSNA, RF, and blood gases in the RVLM.5 In the present studies we documented the effects of selective endothelin antagonists on the hemodynamic and respiratory actions evoked by ET-1 in the RVLM. After initial instrumentation and resting periods, different rat groups received microinjections of either saline (60 nL), BQ-123 (2, 200, or 2000 pmol/60 nL), or BQ-788 (same concentrations), and the effects on BP, HR, and RF were followed during the next 15 minutes. After this period rats received a microinjection of ET-1 (1 pmol), and the hemodynamic changes in BP and HR were followed for the next hour.
To investigate whether the effects of endothelin receptor antagonists were selective for endothelin, we studied the effects of BQ-123 on the cardiovascular effects of substances such as glutamate or nicotine. In a group of six rats glutamate (74 pmol/60 nL) was microinjected into the RVLM before and after similar administration of 2000 pmol/60 nL BQ-123. In the other group we studied the potential inhibitory effects of 2000 pmol/60 nL BQ-123 on the cardiovascular effects evoked by nicotine administration (369 pmol/60 nL, n=6).
Blockade of Endothelin Effects
A different subset of studies evaluated the relevance of the ventrolateral medulla for the cardiovascular effects evoked by intracisternal administration of endothelin. In these rats the cisterna magna was exposed, and the tip of a 10-cm-long polyethylene cannula (PE-10) was inserted through the leptomeninges. This cannula was kept in place by cyanoacrylate biomedical glue; adequate location of the cannula was verified by efflux of clear cerebrospinal fluid. Four trephine holes were made in the occipital area for implantation of cannulas in sites of the left and right RVLM and left and right CVLM. The rats were then randomly assigned to one of five different experimental groups. Group 1 rats (n=6) were instrumented with triple-barreled micropipettes for bilateral microinjections of saline (60 nL) in both the RVLM and CVLM. The final injection site was selected with the use of the cardiovascular responses to glutamate as described above. Group 2 rats (n=8) were similarly instrumented but received the endothelin receptor antagonist BQ-123 bilaterally (2 pmol/60 nL per site) in the RVLM and CVLM. Group 3 rats (n=8) received the same endothelin receptor antagonist only in the RVLM (2 pmol per site), whereas saline (60 nL) was microinjected bilaterally into the CVLM. Group 4 rats (n=8) received bilateral microinjections of saline into the RVLM in conjunction with bilateral administration of BQ-123 (same dose as in group 3) into the CVLM. Group 5 rats (n=4) were used as an additional control group and were microinjected with BQ-123 in sites surrounding or proximal to the RVLM and CVLM. Fifteen minutes after the administration of substances into the ventral medulla, ET-1 (20 pg) was administered through the intracisternal catheter, and hemodynamic effects were followed for at least the next 60 minutes.
After completion of the experiments, 120 nL fast green ink was injected through the cannula, and the rats were perfused with saline followed by a solution of 4% formaldehyde. Sections of the brain stem (40 μm) were stained with cresyl violet, and placement of the pipette tip in the CVLM or RVLM was verified by examination of the histological sections. Only results from those rats in which histology documented adequate placement within the appropriate brain nucleus were included for final evaluation.
Urethane was from Aldrich Chemical Co; l-glutamic acid from NBC; ET-1, BQ-610, and BQ-788 from Peptides International Inc; phenylephrine, nicotine, and pancuronium bromide from Sigma Chemical Co; and the endothelin receptor antagonist BQ-123 was a generous gift from BANYU Pharmaceutical.
For evaluation of the results, Student’s paired t test or ANOVA followed by Dunnett’s test for significant differences was used as appropriate. Differences with a value of P<.05 were considered significant. All data are presented as mean±SEM.
Effects of Endothelin Into the CVLM
The first set of experiments evaluated the cardiovascular effects of different doses of ET-1 in the CVLM. In an area of the CVLM where glutamate maximally reduces BP (−22±4 mm Hg) and HR (−113±26 beats per minute [bpm]), microinjection of low ET-1 doses (0.25, 0.5, and 1 pmol) modestly decreased BP and increased HR (Fig 1⇓). The cardiovascular effects of these endothelin doses peaked 15±5 minutes after microinjection of the peptide (Fig 2⇓) and were maximal at the 1 pmol dose (−12±4 mm Hg and 38±7 bpm for BP and HR, respectively; P<.01). Doses higher than 2 pmol resulted in variable BP responses with more consistent initial tachycardia, followed by respiratory failure and subsequent cardiovascular collapse in up to 90% of the rats.
Effects of Microinjection of ET-1 into the CVLM on RSNA
To assess the relevance of sympathetic mechanisms, we studied the changes in RSNA evoked by microinjection of ET-1 into the CVLM. Basal BP and HR were 98±6 mm Hg and 342±15 bpm, respectively (n=6). Vehicle microinjection did not modify RSNA (3±9%), BP (99±8 mm Hg), or HR (339±21 bpm). However, administration of 1 pmol ET-1 into the CVLM resulted in a progressive decrease in RSNA and BP with concomitant tachycardia (Table 1⇓). Maximal decreases in RSNA (−52±15%) and BP (−18±8 mm Hg) were observed at 15±2 minutes after peptide administration and progressively recovered within the next 45 minutes.
Effects on Respiratory Function of ET-1 Microinjection Into the CVLM
In additional groups of rats we monitored the effects on RF of ET-1 microinjection into the CVLM. Basal respiratory rate in the different groups ranged between 88±4 and 99±8 breaths per minute. ET-1 administration into the CVLM evoked a dose-related decrease in RF that was maximal at the 1 pmol dose (−18±10 breaths per minute; n=4, P<.01, Fig 3⇓). Although the decrease in RF developed concomitantly with the hypotensive effect, it peaked earlier (10±3 minutes, Fig 2⇑). After administration of higher ET-1 doses, the moderate decrease in RF was followed by abrupt apnea and cardiovascular failure.
In one additional rat group (n=6) we investigated whether significant changes in arterial blood gases were associated with the hypotensive and respiratory effects of ET-1 in the CVLM. Basal BP, HR, and RF values were 88±5 mm Hg, 343±21 bpm, and 98±4 breaths per minute, respectively. Control values for blood gases included a pH of 7.347±0.018, Po2 of 103±16 mm Hg, and Pco2 of 46±3 mm Hg. At the time of the maximal cardiovascular effects (−16±5 mm Hg and 45±13 bpm for BP and HR, respectively; P<.05), RF decreased by 14±5 breaths per minute (P<.05), with no significant changes in pH (7.310±0.016), Po2 (100±6 mm Hg), or Pco2 (50±3 mm Hg).
To further examine the effects of ET-1 on respiratory function, we studied the effects produced by this peptide on phrenic nerve activity and hemodynamic function in a group of artificially ventilated rats (n=6). Basal mean BP and HR values were 81±4 mm Hg and 437±23 bpm, respectively. Microinjection of 1 pmol ET-1 into the CVLM of artificially ventilated rats resulted in a hypotensive (−12±5 mm Hg) and tachycardic (29±15 bpm) response; these BP and HR values were not significantly different from those obtained in spontaneously breathing rats. Phrenic nerve activity decreased from control values by 52±15% (P<.01) and remained below baseline values for up to 60 minutes. Cardiovascular failure was not observed in this group.
Specificity of the Endothelin Response in the CVLM
The specificity of the endothelin response was elucidated with the use of selective endothelin receptor antagonists. We first determined whether the cardiovascular and respiratory effects of ET-1 in the CVLM were affected by pretreatment with the ETA receptor antagonist BQ-123. The administration of 2 pmol BQ-123 did not affect basal BP or HR. However, RF increased (15±6 breaths per minute; n=5, P<.05, Table 2⇓) and remained elevated for up to 14 minutes. Subsequent microinjection of ET-1 (1 pmol) did not affect BP, HR, or RF. The involvement of the ETA receptor subtype was further confirmed with the use of the more potent ETA receptor antagonist BQ-610 (0.03 pmol/60 nL; n=6). Like BQ-123, microinjection of BQ-610 increased resting RF (25±11 breaths per minute, P<.05) and completely prevented the cardiorespiratory effects of 1 pmol ET-1 (Table 2⇓). In related experiments, we investigated the potential involvement of the ETB receptor subtype. Microinjection of 2 pmol BQ-788 did not affect basal BP, HR, or RF. Administration of 1 pmol ET-1 resulted in hemodynamic (hypotension and tachycardia) and respiratory (decrease in RF) effects that were not different from those in the group pretreated with saline (Table 2⇓).
Administration of Endothelin Into the RVLM
In agreement with earlier findings,5 administration of 2 pmol ET-1 into the RVLM of anesthetized, spontaneously breathing rats caused initial pressor and bradycardic effects followed by a sustained decrease in BP and more pronounced bradycardia. The hypotensive phase lasted 24±3 minutes, and in 38% of the rats cardiorespiratory failure developed. In a separate rat group the effects of the same dose of ET-1 were studied after pretreatment with either 2, 200, or 2000 pmol/60 nL of the ETA receptor antagonist BQ-123. With the lowest dose of BQ-123 (n=4) no significant changes in resting BP (1±6 mm Hg, from a basal level of 91±4 mm Hg) or HR (−5±6 bpm, from a basal level of 351±19 bpm) were observed. This dose of the antagonist slightly attenuated the early pressor and bradycardic effects (P<.05) and the subsequent hypotensive phase (Fig 4⇓). The intermediate dose (200 pmol) increased BP by 5±3 mm Hg (from 93±8 mm Hg, P<.05), without affecting HR (4±14 bpm). In addition, this BQ-123 dose inhibited the effects of endothelin by 56% (n=6, P<.05; Figs 4⇓ and 5⇓). In a different group (n=4) microinjection of 2000 pmol BQ-123 increased basal BP by 11±1 mm Hg and HR by 15±3 bpm (P<.05) and completely antagonized the cardiovascular effects of ET-1 administration into the RVLM (89±4 mm Hg and 326±14 bpm versus 91±4 mm Hg and 321±16 bpm before and after ET-1 microinjection, respectively). In any of the groups treated with the receptor antagonist, apnea or cardiovascular failure was observed (data not shown).
To test further the specificity of BQ-123 actions, we investigated whether the cardiovascular effects of glutamate were affected by this ETA receptor antagonist. Glutamate administration into the RVLM resulted in pressor (32±4 mm Hg) and tachycardic (35±21 bpm, P<.01) effects that were not significantly affected by pretreatment with 2000 pmol BQ-123 (15 minutes after the antagonist: 29±6 mm Hg and 30±18 bpm for BP and HR, respectively). In related experiments we observed that the cardiovascular effects of nicotine in the RVLM were not affected by BQ-123 administration. During the control period, microinjection of 369 pmol nicotine resulted in an increase in BP (15±4 mm Hg) and HR (11±13 bpm, P<.05). Similar pressor (18±4 mm Hg) and tachycardic (13±16 bpm) effects in response to nicotine administration were observed 15 minutes after microinjection of 2000 pmol BQ-123.
In contrast to the inhibitory effects of BQ-123, the prior administration of the selective ETB receptor antagonist BQ-788 did not affect resting BP or HR or the subsequent effects of ET-1 administration into the RVLM (Table 3⇓).
Effects of Microinjection of BQ-123 Into the Ventrolateral Medulla on the Cardiovascular Actions Evoked by Intracisternal ET-1 Administration
Intracisternal administration of ET-1 (20 pmol) in rats bilaterally pretreated with saline (group 1) in the RVLM and CVLM resulted in a decrease in arterial BP from 89±8 to 71±5 mm Hg and HR from 373±14 to 335±23 bpm within 5 minutes of the peptide administration (Fig 6⇓). Twenty-five percent of the rats died immediately after this particular period as a result of abrupt apnea. In the remaining rats this phase was followed by pronounced hypertension (+50±15 mm Hg) and bradycardia. Subsequently, the rats experienced severe hypotension, bradycardia, and respiratory impairment that resulted in 100% mortality.
In group 2, bilateral administration of BQ-123 in both the RVLM and CVLM resulted in an increment in mean BP (from 95±2 to 100±4 mm Hg, P<.05) and no significant change in heart rate (308±14 to 313±14 bpm). These short-term hemodynamic effects of the endothelin receptor antagonist resolved by 15 minutes of administration. In addition, the prior microinjection of the endothelin receptor antagonist affected the cardiovascular actions of subsequent intracisternal application of ET-1 (20 pmol). In the initial phase, the decrease in BP was abolished, whereas the bradycardic effects were similar to those of controls. The subsequent pressor period, present after the initial 5 minutes in controls, was reduced in the rats pretreated with the antagonist and lasted up to 15 minutes. In clear contrast with the results in the control group, the previous administration of BQ-123 inhibited the bradycardia and completely prevented the cardiorespiratory depression evoked by intracisternal administration of endothelin (Fig 6⇑).
In group 3 (the rats that only received pretreatment with the endothelin receptor antagonist into the RVLM) the initial hypotensive and bradycardic phase evoked by the intracisternal administration of ET-1 was unchanged compared with the control group (Fig 7⇓). The cardiovascular effects of the subsequent period were reduced; the increment in BP (15±7 mm Hg) and reduction in HR (−82±26 bpm) were less pronounced than in controls (see control values above, P<.05). In addition, mortality was significantly reduced, with 33% of the rats experiencing cardiorespiratory failure at this time. Overall mortality in this group was 67%.
In the rats that received BQ-123 only in the CVLM, the initial hypotension but not the bradycardia was significantly reduced. On the other hand, the first minutes of the pressor/bradycardic phase evoked by intracisternal administration of ET-1 did not differ from the control group (Fig 7⇑). In this group, treatment with BQ-123 in the CVLM reduced overall mortality by 25%.
To test the anatomic specificity for the protective actions of BQ-123, we added an additional control group in which the receptor antagonist was microinjected in four sites outside of the RVLM and CVLM. All the hemodynamic phases previously characterized in the absence of the receptor antagonist were present, and cardiorespiratory failure developed in 100% of this rat group.
Different neuronal cell groups located in the ventrolateral medulla, an area that corresponds to the ventrolateral quadrant of the medullary reticular formation, affect autonomic function. At rostral levels of the nucleus ambiguus, the ventrolateral medulla contains preganglionic parasympathetic neurons (which project to the heart and to different structures of the respiratory system) and bulbospinal respiratory premotoneurons (expiratory and inspiratory).18 The RVLM (which lies rostral to the lateral reticular nucleus, caudal to the facial nucleus, and extends to the ventral medullary surface) contains cells (adrenergic and nonaminergic) that project to the intermediolateral column of the spinal cord and regulate vasomotor function.19 Since autoradiographic studies7 documented endothelin receptor binding sites in the RVLM, it is not surprising that microinjection of endothelin in this area affects both cardiovascular and respiratory functions.
We previously reported5 that microinjection of cumulative doses of endothelin (0.5 to 2 pmol/60 nL) into the RVLM results in a biphasic response. During the initial phase BP and RSNA increase, with a concomitant decrease in HR. This phase is followed by a period characterized by hypotension and bradycardia that in turn precedes cardiorespiratory failure in close to 40% of the animals.5 The present studies using a single administration of 1 pmol endothelin confirm these observations and indicate that the potent cardiorespiratory effects of this peptide do not result from recruitment of actions caused by cumulative doses. In addition, the results also indicate that the cardiovascular and respiratory actions of endothelin are mediated by the ETA receptor subtype. Although two different ETA receptor antagonists inhibited the cardiovascular effects of ET-1, the ETB antagonist was ineffective in modifying endothelin actions. A potential tonic regulatory effect of endothelin in the RVLM is suggested by the pressor effect resulting after microinjections of selective ETA receptor antagonists. It can be argued, however, that BQ-123 effects can be related to nonspecific actions of this antagonist (particularly at higher doses). This has to be considered because the BQ-123 concentrations used in the present study (between 33 μmol/L and 33 mmol/L) were higher than the ones used in binding experiments (22 nmol/L to 6 μmol/L20 21 ). However, nonspecific effects of BQ-123 are unlikely because (1) administration of equimolar concentrations of the ETB receptor antagonist did not affect ET-1 actions on resting BP, HR, or RF; (2) microinjection of lower doses of a more potent ETA receptor antagonist (BQ-610, 0.5 μmol/L) also exerted similar inhibitory effects on endothelin actions; and (3) pretreatment with the highest dose of BQ-123 did not affect the cardiovascular effects of glutamate or nicotine in the RVLM. Furthermore, the BQ-123 concentrations are similar to or less than those used in in vivo studies implicating ETA receptor activation in neuroprotection22 or fluid homeostasis.23
The CVLM also regulates cardiovascular function.24 25 This area contains, among others, a population of noradrenergic neurons of group A1.24 Chemical or electrical stimulation of the CVLM results in hypotension and variable bradycardia.25 In the present study microinjection of endothelin into the CVLM resulted in tachycardia, with a modest decrease in BP, RF, and phrenic nerve activity. Although the hypotension seems to result from an inhibition of sympathetic tone (decrease in RSNA), the changes in HR are probably baroreflex mediated. In a high proportion of the rats, cardiorespiratory collapse followed these effects. As in the RVLM, prior microinjection of the endothelin receptor antagonist inhibited both the cardiovascular and respiratory effects of endothelin. In contrast with the results obtained in the RVLM, the endothelin receptor antagonist did not affect basal BP or HR. However, RF significantly increased after 2 pmol BQ-123, suggesting that endothelin has a tonic modulatory effect in respiratory function. The effects of ET-1 on respiratory function are further supported by the experiments evaluating phrenic nerve activity. In these rats ET-1 inhibited phrenic nerve activity for a rather prolonged period. Overall, the effects of ET-1 and BQ-123 in the ventrolateral medulla suggest that endogenous endothelin can be involved in the regulation of cardiorespiratory function in this area and potentially in the pathophysiological changes observed during increased intracranial pressure (Cushing’s reflex).26
Endothelin is a highly potent vasoconstrictor agent. Studies in isolated arteries indicate that it is more potent than neuropeptide Y and has a more prolonged vasoconstrictive effect than angiotensin II.27 Because of these characteristics it has been difficult to establish in the central nervous system whether the cardiovascular effects of endothelin result from actions on neuronal function or from vascular ischemia produced by intense vasoconstriction. Furthermore, Robinson et al28 have reported that ET-1 applied to the middle cerebral artery of the rat reduced local cerebral blood flow to pathological levels. These authors suggested that the hypertension observed after intracerebroventricular or intracisternal endothelin administration is a consequence of medullary ischemia. However, previous studies by our group5 and others13 in combination with our present results do not support this hypothesis. For instance, in the NTS microinjection of ET-1 decreases BP and HR.5 13 29 If the cardiovascular effects of ET-1 in the NTS were the result of ischemia, we should have observed instead an increase in BP similar to the one reported after impairment of NTS function.30 These results are at variance with a report of Lee et al.31 In that study, ET-1 administration into the NTS resulted in inconsistent changes in BP (after unilateral injection of 1 pmol/100 nL) or pressor effects, with a 45% reduction in cerebral blood flow after bilateral administration of ET-1. Although the reduction in cerebral blood flow may indicate severe vasoconstriction, it is important to note that chemical stimulation of the NTS with agents such as glutamate (a putative neurotransmitter of the baroreflex synapse in the NTS30 ) also decreases cerebral blood flow, with increases in cerebrovascular resistance.32 Since these results suggest that cell bodies within the NTS play a role in the control of the cerebral circulation, the reduction in cerebral blood flow observed after ET-1 administration does not necessarily imply a direct vasoconstrictive effect. It is also relevant to mention that although we did not measure blood flow after ET-1 administration into the RVLM, we have demonstrated that similar microinjections of much more potent cerebral vasoconstrictor agents (ie, vasopressin) do not result in significant cardiovascular effects.5
Also arguing against deleterious effects of endothelin on blood flow are our present results demonstrating that bilateral microinjections of BQ-123 into the ventrolateral medulla inhibit the cardiovascular actions of intracisternal administration of ET-1. If the effects of intracisternal administration of endothelin were mediated by intense general vasoconstriction of the cerebral vasculature, selective blockade of endothelin receptors in specific brain areas should be ineffective in antagonizing endothelin actions. In our studies we clearly demonstrated that combined pretreatment of the RVLM and CVLM with the ETA receptor antagonist abolished the initial hypotensive phase of intracisternal administration of endothelin, inhibited the subsequent pressor effect, and prevented the respiratory depression. In contrast, pretreatment with saline in the RVLM and CVLM did not modify the intracisternal effects of endothelin administration. The studies in rats pretreated with BQ-123 in areas proximal to the ventrolateral medulla (group 5) discounted the possibility that these results are explained by generalized vascular endothelin receptor blockade from leakage of the endothelin receptor antagonist.
Study of the results obtained after selective pretreatment of the RVLM or CVLM with BQ-123 provides additional information on the relevance of those areas in the overall response to intracisternal administration of endothelin. For example, selective blockade of endothelin receptors in the RVLM reduced the hypertensive/bradycardic phase. This seems to indicate that at least part of the pressor effect of intracisternal administration of endothelin results from activation of cardiovascular neurons in the RVLM. The inability of BQ-123 administration to completely prevent the pressor effect of endothelin could be the result of endothelin receptor activation in areas of the RVLM that were not reached and therefore not blocked by the antagonist. This is likely because BQ-123 was delivered in a rather limited volume (60 nL per each side). Alternatively, the remaining pressor effect could result from the participation of additional medullary centers outside the ventrolateral medulla or from the activation of ET-1 receptors that are not blocked by BQ-123.
The initial hypotensive phase evoked by intracisternal administration of endothelin could result from activation of cardiovascular neurons in the CVLM. This is supported by the findings of a reduction in BP after microinjection of endothelin into the CVLM and by the inhibition of the hypotensive phase of intracisternal administration of endothelin by blockade of endothelin receptors in the CVLM.
The respiratory effects of endothelin apparently result from actions on neuronal cell groups in both the RVLM and CVLM. Combined treatment of these two areas with the endothelin receptor antagonist prevented the abrupt apnea observed in the control rats. Since impairment of neuronal function at this level may result in respiratory failure of the animal, it is possible to indicate that the hemodynamic and respiratory actions observed after intracisternal administration of endothelin are the result of selective actions in the ventrolateral medulla and not of medullary ischemia.
In summary, the present observations further support the notion that endothelin within the central nervous system acts as a cardiovascular neuropeptide. They also indicate that endothelin can modulate respiratory function in the rostral medulla and that this anatomic region plays a pivotal role in the effects of endothelin circulating in the cerebrospinal fluid.
Selected Abbreviations and Acronyms
|CVLM||=||caudal ventrolateral medulla|
|NTS||=||nucleus tractus solitarius|
|PNA||=||phrenic nerve activity|
|RSNA||=||renal sympathetic nerve activity|
|RVLM||=||rostral ventrolateral medulla<\/.>|
Part of this work was supported by a grant from the Smokeless Tobacco Research Council and from a Grant-in-Aid from the American Heart Association. The editorial assistance of Winky Boemer is greatly appreciated.
Reprint requests to Rogelio Mosqueda-Garcia, MD, PhD, Medical Research Center, Vanderbilt University Hospital, Nashville, TN 37232.
- Received December 5, 1994.
- Revision received January 13, 1995.
- Accepted April 17, 1995.
Mosqueda-Garcia R, Inagami T, Appalsamy M, Sugiura M, Robertson RM. Endothelin as a neuropeptide: cardiovascular effects in the brainstem of normotensive rats. Circ Res. 1993;72:20-35.
Yoshizawa T, Shinmi O, Giaid A, Yanagisawa M, Gibson SJ, Kimura S, Uchiyama Y, Polak JM, Masaki T, Kanazawa I. Endothelin: a novel peptide in the posterior pituitary system. Science. 1990;247:462-464.
Koseki C, Imai M, Hirata Y, Yanagisawa M, Masaki T. Autoradiographic distribution in rat tissues of binding sites for endothelin: a neuropeptide? Am J Physiol. 1989;256:R858-R866.
Lee M, de la Monte SM, Ng S, Bloch KD, Quertermous T. Expression of the potent vasoconstrictor endothelin in the human central nervous system. J Clin Invest. 1990;86:141-147.
MacCumber MW, Ross CA, Glaser BM, Snyder SH. Endothelin: visualization of mRNAs by in situ hybridization provides evidence for local action. Proc Natl Acad Sci U S A. 1989;86:7285-7289.
Ouchi Y, Kim S, Souza AC, Iijima S, Hattori A, Orimo H, Yoshizumi M, Kurihara H, Yazaki Y. Central effect of endothelin on blood pressure in conscious rats. Am J Physiol. 1989;256:H1747-H1751.
Kawano Y, Yoshida K, Yoshimi H, Kuramochi M, Omae T. The cardiovascular effect of intracerebroventricular endothelin in rats. J Hypertens. 1989;7(suppl 6):S22-S23.
Hashim MA, Tadepalli AS. Hemodynamic responses evoked by endothelin from central cardiovascular neural substrates. Am J Physiol. 1992;262:H1-H9.
Mosqueda-Garcia R, Tseng C, Appalsamy M, Robertson D. Cardiovascular effects of microinjection of angiotensin II in the brainstem of renal hypertensive rats. J Pharmacol Exp Ther. 1990;255:374-381.
Ishikawa K, Fukami T, Nagase T, Mase T, Hayama K, Niiyama K, Fujita Y, Urakawa U, Kumagai T, Fukuroda T, Ihara M, Yano M. Endothelin antagonistic peptide derivatives with high selectivity for ETA receptors. In: Schneider CH, Eberle AN, eds. Peptides 1992. Leiden, Netherlands: ESCOM Science Publishers BV; 1993:685-686.
Guyenet PG. Role of the ventral medulla oblongata in blood pressure regulation. In: Loewy AD, Spyer KM, eds. Central Regulation of Autonomic Functions. New York, NY: Oxford University Press; 1990:145-167.
Ihara M, Noguchi K, Saeki T, Fukuroda T, Tsuchida S, Kimura S, Fukami T, Ishikawa K, Nishikibe M, Yano M. Biological profiles of highly potent novel endothelin antagonists selective for the ETA receptor. Life Sci. 1991;50:247-255.
Samson WK, Murphy TC. Antidipsogenic actions of endothelins are exerted via the endothelin-A receptor in the brain. Am J Physiol. 1993;265:R1212-R1215.
Blessing WW, West MJ, Chalmers J. Hypertension, bradycardia and pulmonary edema after brainstem lesion coinciding with the A1 group of catecholamine neurons. Circ Res. 1981;49:949-958.
Willette R, Krieger AJ, Barcas PP, Sapru HN. Medullary-aminobutyric acid (GABA) receptors and the regulation of blood pressure in the rat. J Pharmacol Exp Ther. 1983;266:893-899.
Oglivy CS, DuBois AB. Effect of increased intracranial pressure on blood pressure, heart rate, respiration and catecholamine levels in neonatal and adult rabbits. Biology Neonate. 1987;52:327-336.
Yanagisawa M, Inoue A, Ishikawa T, Kasuya Y, Kimura S, Kumagaye S, Nagajima K, Watanabe TX, Sakakibara S. Primary structure, synthesis, and biological activity of rat endothelin, an endothelium-derived vasoconstrictor peptide. Proc Natl Acad Sci U S A. 1988;85:6964-6967.
Talman WT, Perrone MH, Reis DJ. Acute hypertension after the local injection of kainic acid into the nucleus tractus solitarii of rats. Circ Res. 1981;48:292-298.
Lee CC, Ohta H, Talman WT. Endothelin-1 microinjected into nucleus tractus solitarii of rat blocks arterial baroreflexes. Genet Hypertens. 1992;218:109-111.