INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

OREXIN A AND B, ALSO KNOWN as hypocretin 1 and 2, are 33- and 28-amino acid peptides expressed in neurons of the rat lateral hypothalamus (7, 18). The initial observation that the peptide when injected into the ventricles promoted feeding behaviors, as assessed by food consumption in rats, has generated a considerable interest relative to its role in food intake and obesity (18). Recent studies have revealed that the peptide may also be an important signaling molecule in sleep/wakefulness. Disruption of the hypocretin receptor 2 gene results in a sleep disorder resembling narcolepsy in canines (16), and orexin knockout mice displayed signs and symptoms similar to those of human narcolepsy (2).

Although investigations relative to the role of orexins in feeding and narcolepsy have been the major focus of interest, there is evidence that the peptide may also be an important messenger molecule in the central regulation of autonomic activity including cardiorespiratory. For example, orexin-immunoreactive fibers are noted in areas of the medulla, including the nucleus of the solitary tract and ventral medulla, that are known to influence cardiorespiratory and other autonomic functions (6, 10, 17). Several functional studies have shown that orexins, with intracerebroventricular or intracisternal injection, increased mean arterial pressure (MAP) and heart rate (HR) in anesthetized or conscious rats (4, 19, 22). These studies suggest that orexins may act to increase sympathetic activity at the level of the medulla.

In addition to the medulla, orexin-immunoreactive fibers have been detected in the spinal cord, including the intermediolateral cell column (IML) where the majority of sympathetic preganglionic neurons (SPNs) is located (6, 24). The present study was conducted to evaluate the hypothesis that orexins by acting on SPNs may augment spinal sympathetic outflow, as assessed by a change in the MAP and HR in anesthetized rats and in the membrane potential of individual SPNs in thoracolumbar spinal cord slices.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

A breeding colony of Sprague-Dawley rats, purchased from Harlan (Indianapolis, IN), was established at the Division of Laboratory Animal Resources, East Tennessee State University. Animals were housed two per cage in a room maintained at 22 ± 1°C with an alternating 12:12-h light-dark cycle; food and water were available ad libitum. Male rats weighing 325-350 g were used in the in vivo experiments, and 12- to 16-day-old rats were used in the in vitro study. Animal protocols were reviewed and approved by the Institution Animal Care and Use Committee.

In vivo experiments. Procedures for intrathecal injection to anesthetized rats were similar to those described earlier (14, 15). Under urethane anesthesia (1.2 g/kg ip), the right femoral artery was cannulated with a polyethylene tubing and connected to a pressure transducer with its output to a Gould pen recorder. The blood pressure signal was used to trigger a Biotach amplifier (Gould ECG/BiTac) for HR recording. The right femoral vein was cannulated for intravenous injection of adrenergic receptor antagonists in one series of experiments. For intrathecal injection, a polyethylene tubing (Intramedic PE-10) was passed through a slit in the dura at the atlantooccipital junction to the T2-T3 segments; the position of the tubing was visually verified at the end of each experiment. Orexins were dissolved in deionized water and frozen in aliquots. A few minutes before administration, orexins were thawed and dissolved in physiological saline to desired concentrations. Intrathecal injections were made at a volume of 10 µl followed by 10 µl saline to wash in the peptide. Orexin A antiserum (1:500 dilution) or normal rabbit serum albumin (1:500 dilution) was injected intrathecally at a volume of 10 µl followed by 10 µl saline in a manner similar to that described for orexin injections. The protocol for the above series of experiments is as follows. First, saline was injected and followed by orexin A (1 nmol). After MAP and HR had returned to the basal level, orexin A antiserum and orexin A (1 nmol) were administered in succession, with a 10-min interval between injections. Lastly, orexin A was administered to determine whether or not the response to the peptide had recovered from the blocking effects of antiserum. MAP was calculated using the equation [(S  D)/3 + D], where S is systolic pressure and D is diastolic pressure, which were measured directly from the precalibrated chart recorder.

Whole cell patch-recording techniques. Procedures used in obtaining 400-µm transverse thoracolumbar spinal cord slices of young (12-16 days old) rats were similar to those described earlier (14, 15, 25). The spinal cord slice was held between two nylon meshes in a tissue chamber and continuously perfused with Krebs solution of the following composition (in mM): 127 NaCl, 1.9 KCl, 1.2 KH₂PO₄, 2.4 CaCl₂, 1.3 MgCl₂, 26 NaHCO₃, and 10 glucose. The solution was saturated with 95% O₂ and 5% CO₂. Whole cell recordings were made from antidromically identified SPNs under current-clamp mode with the use of an Axopatch 1D (Axon Instruments, Foster City, CA) as described previously (14, 15, 20, 25). Patch electrodes, filled with a solution containing (in mM) 130 K⁺ gluconate, 1 MgCl₂, 2 CaCl₂, 4 ATP, 0.3 guanosine 5'-triphosphate, 10 EGTA, and 10 HEPES, had a resistance of 3-5 M. A bipolar stimulating electrode was placed at the ventral root exit to antidromically identify the SPNs (20). Recordings were made at the room temperature (20 ± 1°C). Orexin A and TTX were dissolved in Krebs solution and applied to the spinal cord slices in known concentrations.

Chemicals and statistical data analysis. Orexin A or B peptide and orexin A antiserum were from Phoenix Pharmaceuticals (Belmont, CA), normal rabbit serum albumin was from Vector Laboratories (Burlingame, CA), and TTX was from Research Biochemicals International (Natick, MA). All other chemicals were from Sigma (St. Louis, MO). Data were analyzed statistically using the one-factor ANOVA followed by the Student-Newman-Keuls test, with P < 0.05 considered statistically significant. Results are expressed as means ± SE.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Effects of intrathecal orexins on MAP and HR. The mean MAP and HR in anesthetized rats were 100.8 ± 10.7 mmHg and 335.7 ± 33.1 beats/min (n = 21). Orexin A and B consistently increased MAP and HR in all rats tested. The increase developed slowly, reached a peak in ~3-5 min, and lasted for 15 to over 60 min. Generally, MAP and HR recovered to the basal level within 60 min after orexin A (1 nmol) injection; a representative experiment is shown in Fig. 1. The mean increases of MAP and HR by 0.3 nmol (n = 4), 1 nmol (n = 10), and 10 nmol (n = 3) orexin A were 5.3 ± 1.6, 18.4 ± 1.8, and 30.2 ± 4.4 mmHg, and 10.3 ± 1.1, 42.3 ± 6.1, and 85.7 ± 16.3 beats/min, respectively, above basal values. Orexin B (3 nmol; n = 4) caused a significant increase of MAP and HR of 11.4 ± 1.7 mmHg and 40.2 ± 5.7 beats/min over basal values. Figure 2 illustrates the change of MAP and HR induced by orexin A (1 nmol) and orexin B (3 nmol) relative to time. Intrathecal injection of saline before or after orexin A or B injection caused no significant change of MAP and HR in any of the rats tested (Figs. 1 and 2).

View larger version (12K): [in this window] [in a new window]   Fig. 1.   Pressor response induced by intrathecal injection of orexin A but not saline in urethane-anesthetized rats. Intrathecal injection of orexin A (1 nmol/10 µl) followed by 10 µl saline caused a prolonged elevation of mean arterial pressure (MAP) and heart rate (HR), whereas saline (20 µl) injection had no significant effects.  Indicates time of intrathecal injection. Bpm, beats/min.

View larger version (20K): [in this window] [in a new window]   Fig. 2.   Change of MAP and HR with time following intrathecal injection of orexin A, orexin B, or saline to anesthetized rats. Top: change in MAP and HR in relation to time after intrathecal injection of 1 nmol orexin A or saline. Peak response occurred 3 to 5 min after orexin A injection. Bottom: change in MAP and HR with time after intrathecal injection of 3 nmol orexin B or saline. Vertical bars are means ± SE. *Statistically significant difference of P < 0.05.

Antagonism of orexin A by orexin A antibodies. As specific orexin A receptor antagonists are currently unavailable, the effects of orexin A antiserum were evaluated in relation to the pressor effects induced by the peptide. The specificity of orexin A antiserum has been verified in our immunohistochemical studies in which the immunostaining of orexin A antiserum was completely blocked by preabsorbing the antiserum with orexin A (3, 10).

View larger version (22K): [in this window] [in a new window]   Fig. 3.   Histograms showing the peak percent change of MAP and HR induced by intrathecal injection of orexin A (1 nmol) before and after administration of orexin A antiserum (1:500 dilution). Anesthetized rats had a mean MAP and HR of 103.8 ± 9.8 mmHg and 340.1 ± 35.1 beats/min (n = 5). Saline or orexin A antiserum had no appreciable effects on MAP and HR, whereas the pressor response to orexin A was significantly diminished by prior injection of orexin A antiserum. Vertical bars are means ± SE. *Statistically significant difference between orexin A-treated and saline injection groups, and ** indicates a statistically significant difference between orexin A-treated and orexin A antiserum + orexin A-treated groups.

Effects of adrenergic receptor antagonists. This series of experiments evaluated whether or not the orexin A-induced pressor response is mediated via the activation of the sympathetic nervous system. Propranolol (0.5 mg/kg) and prazosin (0.5 mg/kg) were selected as the -adrenergic and ₁-adrenergic receptor antagonists. Because intravenous administration of prazosin and propranolol caused a fall in MAP and HR, the second injection of orexin A was made after MAP and HR had stabilized to a new level, which was 65.5 ± 8.3 mmHg and 320.2 ± 7.4 beats/min in prazosin-treated rats and 97.3 ± 19.2 mmHg and 259.4 ± 17.3 beats/min in propranolol-treated rats. Propranolol and prazosin attenuated, respectively, the increase in HR and MAP caused by subsequent intrathecal administration of orexin A (Fig. 4).

View larger version (23K): [in this window] [in a new window]   Fig. 4.   Histograms showing the peak percent change in MAP and HR induced by orexin A (1 nmol) before and after intravenous administration of prazosin (0.5 mg/kg) and propranolol (0.5 mg/kg). Anesthetized rats in the prazosin-treated group had a mean basal MAP and HR of 113.8 ± 14.2 mmHg and 340 ± 22.4 beats/min (n = 5), and the mean basal MAP and HR of propranolol-treated rats were 100.3 ± 18.3 mmHg and 315.6 ± 28.5 beats/min. Top: saline injection had no appreciable effects on MAP and HR, whereas prazosin significantly attenuated the MAP increase induced by intrathecal injection of orexin A. The second injection of orexin A was made after MAP and HR had stabilized to a new level, which was 65.5 ± 8.3 mmHg and 320.2 ± 7.4 beats/min. Bottom: saline injection had no appreciable effects on MAP and HR, whereas propranolol significantly attenuated the HR increase induced by intrathecal injection of orexin A. The second injection of orexin A was made after MAP and HR had stabilized to a new level (97.3 ± 19.2 mmHg and 259.4 ± 17.3 beats/min). *Statistically significant difference between orexin A and saline-injected groups, and ** denotes a statistically significant difference between orexin A- and orexin A + prazosin or propranolol-treated groups. Vertical bars are means ± SE.

Orexin A-stimulated SPNs. Whole cell recordings were made from SPNs located in the IML of lower thoracic and upper lumbar spinal cord slices, as described previously (25). SPNs had a mean resting potential of 57 ± 2 mV and input resistance of 670 ± 52 M (n = 17), which were comparable to those reported earlier (14, 15, 25). Also, as reported earlier (21), some of the SPNs in spinal cord slices were found to discharge spontaneously (n = 7) (Fig. 5A). Five of seven (70%) spontaneous active cells responded to superfusion of orexin A or B (100 nM) with an increase in discharge frequency and/or a small depolarization (Fig. 5A). The peptide caused a slow membrane depolarization and/or increase of neuronal discharges in 7 of 10 (70%) silent SPNs tested (Fig. 5, B and D). Orexin A- or B-induced hyperpolarization was not observed in any of the SPNs studied. At the higher dose of 300 nM, orexin A or B caused a larger depolarization and intense neuronal discharges in four of five SPNs (Fig. 5, B and D); the mean depolarization was 6.2 ± 0.8 mV. Superfusion of the slices with a Krebs solution containing TTX (0.5 µM) blocked the neuronal discharge, but not the depolarization, induced by orexin A (Fig. 5C). The depolarization was associated with either a small increase (10-25%) or no apparent change in input resistance in responsive neurons.

View larger version (65K): [in this window] [in a new window]   Fig. 5.   Membrane depolarizations and neuronal discharges induced by orexin A or B from 3 sympathetic preganglionic neurons (SPNs) (A, B, and D) in spinal cord slices. The 4 traces are chart recordings. Orexin A or B was applied to the bath by superfusion for a period indicated by the solid bar. A: orexin A (100 nM) increased the frequency of neuronal discharges and caused a small depolarization in this spontaneously active SPN. The frequency of spontaneous discharges was low and became higher following superfusion of orexin. B: orexin A (300 nM) caused a depolarization and neuronal discharge in this silent SPN. C: orexin A caused a membrane depolarization only in the presence of TTX (0.5 µM). D: orexin B (300 nM) caused a membrane depolarization associated with neuronal discharges. Downward deflections superimposed on the membrane potential are hyperpolarizing electrotonic potentials induced by constant current pulses (not shown). The amplitude of the hyperpolarizing electrotonic potentials is slightly increased (<10%) at the peak of orexin-induced depolarization (C), indicating a small increase in membrane resistance. Note the peak amplitude of neuronal discharges shown in A, B, and D was truncated due to the limited frequency response of the pen recorder. The traces in A, B, and D are from 3 different SPNs. Recordings shown in B and C are from the same SPN.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

In vivo and in vitro experiments were conducted in the rats to evaluate the hypothesis that the peptide orexin modulates spinal sympathetic outflow by interacting with SPNs of the thoracolumbar spinal cord.

Orexin occurs in two forms: A and B (or hypocretin 1 and 2). In our earlier study, orexin A or B by intracisternal injection or microinjection to the rostral ventrolateral medulla elicited a qualitatively similar increase in MAP and HR in urethane-anesthetized rats (4). Here, our results indicate that intrathecal injections of orexin A or B produced qualitatively similar increases of MAP and HR. Orexin A, however, appears to be more potent than orexin B in eliciting a pressor response. For these reasons, orexin A was selected as the representative of the two peptides whose effects were more closely scrutinized in the present study.

Results from the in vivo experiments showed that orexin A upon intrathecal injection consistently caused a significant and prolonged increase of MAP and HR, whereas saline injection had no appreciable effects. The pressor response was markedly attenuated by prior injection of orexin A antiserum, which presumably neutralized the orexin A action by binding to the peptide. The specificity of orexin A antiserum used in our study has been verified in our previous studies in which preabsorption of the antiserum with the peptide orexin A rendered the former ineffective (3, 10). Intrathecal injection of normal serum albumin resulted in no appreciable change of the pressor response induced by orexin A, indicating that the antagonizing effect of orexin A antiserum was specific. Furthermore, the observation that the ₁-adrenoceptor and -adrenoceptor antagonists prazosin and propranolol differentially attenuated the increase of MAP and HR supports the contention that the pressor response induced by intrathecal administration of orexin A is mediated by the activation of the sympathetic nervous system.

Within the sympathetic nervous system, SPNs in the thoracolumbar spinal cord provide the known output to the sympathetic ganglia, which in turn, innervate the heart and blood vessels. The increase in HR and blood pressure after intrathecal orexin A should reflect an excitation of SPNs innervating the respective organs. The possibility that the peptide following intrathecal injection may enter the circulation and produce a pressor response via a peripheral action appears to be unlikely, because intravenous administration of orexin A or orexin B at higher doses has been found to be ineffective in eliciting a pressor response in urethane-anesthetized rats (4).

More importantly, results from in vitro experiments directly demonstrated an excitatory action of orexin on SPNs. Superfusion of orexin A or orexin B caused a depolarization and/or neuronal discharge in the majority of SPNs tested. The membrane depolarization, but not the neuronal discharge, persisted in a TTX-containing solution, indicating that the peptide acted directly on SPNs. The depolarization and associated increase of neuronal discharges of SPNs observed in vitro could explain the sympathoexcitatory effect of orexin A in vivo.

At the concentrations tested here, orexin A or B had no apparent membrane effects on a number of SPNs. A simple explanation is that SPNs that bear orexin receptors are target specific. Alternatively, the concentrations used here may not be optimal to some of the SPNs tested. Two subtypes of orexin receptors, OX₁R and OX₂R, have been characterized (18). The OX₁R shows high affinity for orexin A, and orexin A and B bind with about equal affinity to OX₂R (18). Both types of orexin receptors are G protein coupled (18). Activation of orexin receptors by orexin A or B on SPNs produced a membrane depolarization of relatively slow time course, which is characteristic to activation of other known G protein-coupled receptors.

The ionic mechanism underlying orexin A-induced depolarizations in SPNs has not been elucidated. The depolarization was accompanied by either an increase of input resistance in some neurons or no apparent change in others. Orexin-induced depolarizations in the rat dorsal motor nucleus of vagus neurons were associated with an increase of input resistance or no apparent change, which is attributed to an increase of nonselective cation conductance and a decrease of potassium conductance (11). In view of the similarity of membrane resistance change induced by orexin A in SPNs on one hand and dorsal motor nucleus of vagus neurons on the other, the ionic mechanism underlying the depolarization in these two types of neurons is expected to be similar. In the case of rat locus coeruleus neurons, the orexin-induced depolarization appears to be caused by a decrease of potassium conductance (12).

The question of whether endogenously released orexin A in the spinal cord may play a role in regulating sympathetic outflow remains to be addressed. Orexin A- and orexin B-immunoreactive fibers are present in the IML area of the rat thoracolumbar spinal cord (5, 24). The physiological and/or pathological conditions under which the peptide may be released within the IML area are not known. Another question that needs to be addressed in future studies is whether or not activation of orexin-containing neurons in the lateral hypothalamus produces a pressor response similar to that observed here. An earlier study showed that electrical stimulation applied to various sites in the forebrain and midbrain including areas of the lateral hypothalamus, where orexin-containing neurons are located, elicited vasoconstriction, as assessed by lumen diameter changes of mesenteric vascular beds (9). Both pressor and depressor responses were observed in more recent studies where L-glutamate was microinjected to the lateral hypothalamus (1, 23). The development of specific orexin receptor antagonists would be crucial in defining a physiological role of orexins in central regulation of sympathetic outflow.

Perspectives