INTRODUCTION

Top Abstract Introduction Materials & Methods Results Discussion References

VARIOUS COMPONENTS of the growth hormone (GH)-releasing hormone (GHRH)-somatostatin (SRIH)-GH-insulin-like growth factor I somatotropic axis are involved in the regulation of sleep. GH was the first hormone for which a sleep-related secretory pattern was discovered. In humans, the major GH release is associated with deep non-rapid eye movement sleep (NREMS) onset, generally occurring soon after sleep onset (31, 37, reviewed in Ref. 39). Correlation between NREMS and GH secretion also occurs in other species including the rat (20). In addition, experiments involving sleep deprivation provide support for the coupling of NREMS and GH secretion. GH release tends to be suppressed during deprivation, whereas recovery sleep is often associated with larger than normal GH surges (31, 37), although the sleep deprivation-induced changes in GH secretion might depend on the age of the subject (21). The sleep-associated secretion of GH is attributed to hypothalamic GHRH simultaneously stimulating NREMS and pituitary GH release under normal conditions (13).

GHRH has now become one of the best documented sleep-promoting substances. Systemic injection of GHRH promotes NREMS in human subjects if administered in repeated pulses or several hours after sleep onset (12, 16, 34) and in rats (26). Central administration of GHRH promotes NREMS and behavioral signs of NREMS in rats and rabbits (7, 24, 25). In addition to increasing the duration of NREMS, GHRH also enhances slow-wave activity in the electroencephalograph (EEG) during NREMS (7, 24, 25). Similar increases in EEG slow-wave activity are observed after sleep deprivation (29), and it is thought that these supranormal EEG waves reflect the intensity of NREMS (3). Inhibition of endogenous GHRH using either a peptide antagonist (27) or anti-GHRH antibodies (28) suppresses spontaneous sleep. Furthermore, immunoneutralization of GHRH attenuates sleep rebound during recovery after sleep deprivation (28). NREMS is also suppressed in a transgenic mouse model with a deficiency in the somatotropic system (42). Reduced NREMS in these transgenic mice might be attributed to the reduction of GHRH observed in the hypothalami of these transgenic mice.

Hypothalamic GHRH mRNA displays a diurnal rhythm; highest levels occur at the onset of daylight hours in rats (4). Furthermore, with the use of in situ hybridization, Toppila et al. (38) observed that in rats 6 h of sleep deprivation, starting at light onset, induced increased GHRH mRNA expression in the paraventricular nucleus of the hypothalamus. However, they failed to find changes in GHRH mRNA expression after 12 h of sleep deprivation during the dark period nor did they determine whether GHRH mRNA levels returned to normal after sleep recovery. The aim of this study was to confirm, using different methodology, the findings that GHRH mRNA has a diurnal variation in the hypothalamus and that its levels are increased during sleep deprivation. Furthermore, we examined the effects of sleep recovery after sleep deprivation on GHRH mRNA.

The pulsatile GH secretion results from reciprocal stimulatory and inhibitory actions of GHRH and SRIH at the levels of both the pituitary and the hypothalamus (41). GHRH and SRIH mRNA expressions in the arcuate nucleus exhibit a 3-h ultradian rhythm with a 180° phase shift between them. It seems that GHRH and SRIH also exert opposite effects on sleep: GHRH enhances NREMS, whereas somatostatinergic stimulation elicits a prompt suppression of NREMS (1). To determine whether sleep deprivation alters SRIH transcription, mRNA levels of SRIH were measured in the same hypothalamic samples used to detect GHRH mRNA. Finally, SRIH mRNA was determined at the two time points of the diurnal cycle where GHRH mRNA expression exhibits the largest difference: at 1 h after light onset and at 1 h after dark onset.

	MATERIALS AND METHODS

Top Abstract Introduction Materials & Methods Results Discussion References

Animals. Adult male Sprague-Dawley rats (300-350 g) were housed individually in sound-attenuated environmental chambers at 25 ± 1°C. Animals were kept on a 12:12-h light-dark cycle with lights on at 0800. Food and water were available ad libitum. Animals were acclimated to these housing conditions for at least 2 wk before the experiment.

Construction of the plasmid that contains the mutant rat GHRH mRNA as an internal standard. cDNA for PCR templates was synthesized from total RNA extracted from a normal male rat hypothalamus. Two separate PCRs were performed (RoboCycler gradient 40, Stratagene, La Jolla, CA) using two different sets of primers (Table 1) based on the cloned rat GHRH cDNA sequence (17). In the first PCR, the sense primer 1 had the same sequence as rat GHRH cDNA from position 201 to 220 and the antisense primer 1 was a composite oligonucleotide with the complementary sequence as rat GHRH cDNA from position 286 to 300 plus position 361 to 374. This PCR preferentially amplified the same sequence as rat GHRH cDNA from position 201 to 374 with a 60-bp deletion from position 301 to 360. Similarly, in the second PCR the sense primer 2 was also a composite oligonucleotide with the same sequence as rat GHRH cDNA from position 286 to 300 plus position 361 to 374. The preferential PCR product from this reaction had the same sequence as rat GHRH cDNA from position 286 to 600 with a 60-bp deletion from position 301-360. Therefore, the 3'-terminal end of the product from the first PCR was complementary to the 5'-terminal end of the product from the second PCR, and they would anneal together if mixed under the optimal condition. These PCR products were mixed and used as templates for the subsequent PCR using sense primer 1 and antisense primer 2 (Table 1). The resulting PCR product was a mutant rat GHRH cDNA with 60-bp deletion from 301 to 360. The PCR product was purified with GeneClean II Kit (Bio101, La Jolla, CA) following the manufacturer's instructions.

The purified PCR product was directly ligated into a linearized pCRII vector (Invitrogen TA Cloning Kit, Carlsbad, CA) using T4 DNA ligase. Colonies containing the plasmid were selected. To confirm the correct insert, the plasmid was used as a DNA template in a PCR using sense primer 1 and antisense primer 2 (Table 1). The colony that produced a 340-bp product was chosen to be further amplified in 25 ml liquid bacterial medium containing ampicillin (50 µg/ml) at 37°C overnight in a rotary shaker. The plasmids were purified using QIAGEN Plasmid Kit (QIAGEN, Santa Clarita, CA). Because the linearized vector has a single 3'-deoxythymidine (T) extruding at both ends, the inserted PCR product could be oriented in either 5'- to 3'- or 3'- to 5'-end direction. Restriction mapping using Hind III and Xba I digestion of the purified plasmid yielded one 160-bp and one 289-bp fragment as expected, indicating that the insert of mutant GHRH sequence was from the 5'- to 3'-end (+). The orientation and sequence of this insert were further verified by sequencing at Center for Biotechnology, St. Jude Children's Research Hospital (Memphis, TN).

In vitro transcription of the mutant rat GHRH and SRIF internal standards. In vitro transcription was done using an RNA transcription kit (Stratagene). Briefly, the purified plasmid containing mutant GHRH insert was linearized by Xho I digestion at the multiple cloning site downstream of the insert. The transcription reaction consisted of 20 units of SP6 polymerase, 1 µl of 10 mM rATP-rGTP-rCTP-rUTP, 1 µl of 0.75 M DTT, 1.4 µg of the Xho I-linearized plasmid as DNA template, and 5 µ1 of 5× transcription buffer in a final volume of 25 µl. The reaction was carried out at 37°C for 30 min. After the transcription, excess RNase-free DNase I was added to completely digest the plasmid DNA template. The mutant rat GHRH internal standard cRNA was purified by phenol-chloroform extraction, and isopropanol precipitation was followed by two washes with 75 and 100% ethanol. The pellet was dissolved in an appropriate amount of Tris-EDTA buffer. The amount of mutant GHRH cRNA internal standard was quantified with spectrophotometry in terms of optical density at 260 nm.

This mutant rat GHRH cRNA was used as an internal standard for subsequent RT-PCR analysis of RNA samples in this study. The advantage of using this internal control was that it controls for errors that may be induced by different efficiencies in either RT or PCR. The cRNA internal standard was added into the sample RNA before the RT reaction so both of them were reverse transcribed simultaneously. Only one pair of primers was used for the PCR step for both internal standard cDNA and sample cDNA from the same RT reaction in the same tube. The target GHRH mRNA and internal control were expected to amplify with equal efficiency because amplification efficiency is primarily determined by the primer sequences and their annealing conditions when the sizes of the amplified sequences are similar (33, 40).

To prepare internal standard for SRIH mRNA, a modified version of the procedure reported by Riedy et al. (30) was followed. A fragment of rat SRIH gene was amplified by PCR using cDNA that was reverse transcribed from rat brain total RNA. The 5'-oligonucleotide was 5'-ATGCTGTCCTGCCGTCTCCAGCTGCCACCGGGAAACAGGAACTGGC-3'. The first 20 bases starting from the 5'-end are the same sequence as the SRIH sense primers (see below); there is a 99-bp deletion from the SRIH sequence between the first 20 bases and the last 26 bases of this primer. The 3'-antisense oligonucleotide was 5'-CTAACAGGATGTGAATGTCTTCC-3', thus a 99-bp deletion was produced in the final PCR product. This PCR product was then cloned using the TA cloning kit (Invitrogen) according to the instructions provided. The plasmid construct was amplified in DH5 bacteria as described above. The SRIH internal standard cRNA was generated by in vitro transcription of the mutant gene using SP6 RNA polymerase followed by DNase I digestion.

Experimental protocol. Sleep deprivation was performed by gentle handling while animals were kept in their home cages. The rats were monitored continuously during sleep deprivation. They were aroused by light stimuli, e.g., knocking on the cage, touching the animal whenever they appeared to go to sleep, but the rats were never lifted out of the cages. The deprivation started at light onset. In rats, the duration and intensity of NREMS (3, 8) and hypothalamic GHRH mRNA levels (4) are highest during the first few hours after light onset. Groups of rats were killed by decapitation after 8 (n = 10) and 12 h (n = 4) of sleep deprivation and 1 and 2 h after 8 h of sleep deprivation (n = 10 for each group) and 2 h after 12 h of sleep deprivation (n = 4). Control time-matched undisturbed rats (n = 10 for each of the 3 groups corresponding to the 8 h of sleep deprivation groups and n = 4 corresponding to each of the 2 12 h of sleep deprivation groups) were killed at the corresponding time points (1600, 1700, 1800, 2000, and 2200). The brains were immediately removed, and the hypothalami were dissected by using the landmarks as follows: optic chiasma, lateral sulci, mammillary bodies, and a depth of 2 mm. The samples were snap-frozen in liquid nitrogen and stored in an ultra-low-temperature freezer at 80°C until RNA extraction. The total time between removal of a rat from its cage and when the dissected area of brain was frozen was <2 min.

To determine time-of-day variations in SRIH mRNA levels, groups of rats (n = 8 each) were killed at 1 h after light onset and at 1 h after dark onset. In addition to the hypothalamus, samples were also collected from other brain regions as follows: hippocampus (dorsal hippocampus from both hemispheres), brain stem (mesencephalon + pons), and cerebral cortex (parietal cortex from both hemispheres).

RNA extraction. Total RNA was extracted following the modified method of Chomczynski and Sacchi (5) using RNA STAT-60 according to the manufacturer's protocol (TelTest "B," Friendswood, TX). Briefly, tissues were homogenized in 1 ml RNA STAT-60 solution then extracted with 0.2 ml CHCl₃. After centrifugation at 4°C, the aqueous upper phase was transferred to a fresh tube and RNA was precipitated by adding 0.5 ml isopropanol. After centrifugation for another 10 min at 4°C, the RNA pellet was washed once with 1 ml 75% (vol/vol) ethanol and then washed again with absolute ethanol. After the last centrifugation, the RNA pellet was briefly air-dried then dissolved in 20 µl sterile, RNase-free water. The integrity of the RNA was checked by denaturation and formaldehyde-containing agarose gel electrophoresis. The amount of RNA was measured in terms of optical density at 260 nm.

RT-PCR. First-strand cDNA was synthesized using Superscript II RNase H RT according to the manufacturer's instruction (GIBCO-BRL, Gaithersburg, MD) with minor modification. Briefly, 1 µg total sample RNA and a proper amount of cRNA internal standard were incubated with 50 ng antisense primer 2 (Table 1) for 90 min at 42°C. The amount of cRNA internal standard added was determined by pilot PCRs of random samples from various control and experimental groups; in all cases equal amounts of cRNA internal standard were added to control and experimental RNA extracts. The RT reaction was terminated by heating to 95°C for 10 min followed by a brief centrifugation at room temperature. cDNA was cooled to room temperature and then stored at 20°C until further analysis.

cDNA samples were amplified by PCR using Taq polymerase (Promega, Madison, WI) in a total volume of 50 µl. To be in the linear range during amplification (Fig. 1), reactions for hypothalamic samples of GHRH mRNA were carried out for 33 cycles (94°C, 1 min; 60°C, 1 min; 72°C, 1 min, 7 min for the last extension). The primers for PCR were sense primer 3 and antisense primer 3 (Table 1). The resulting PCR products were 294 bp for wild-type GHRH mRNA and 234 bp for the internal standard. cDNA for SRIH was amplified for 26 cycles: annealing was at 59°C for 1 min. The primers for the SRIH were 5'-ATGCTGTCCTGCCGTCTCCA-3' (sense) and 5'-CTAACAGGATGTGAATGTCTTCC-3' (antisense), which amplify a 361-bp product corresponding to wild-type somatostatin and a 262-bp product corresponding to mutant SRIH.

View larger version (11K): [in this window] [in a new window]   Fig. 1.   Linear regions for PCR amplification. PCR was performed on hypothalamic samples, which included total RNA and an appropriate amount of growth hormone-releasing hormone (GHRH) internal standard cRNA. Amplification courses for GHRH mRNA wild-type () and for mutant GHRH internal standard () were calculated from densitometric measurements of the ethidium bromide-stained 2% agarose gel and plotted in a logarithmic scale against the cycle number. Arrow indicates number of cycles used in this study.

Gel electrophoresis. PCR products (10 µl) were loaded on a 2% agarose gel containing ethidium bromide (0.5 µg/ml). Gels were run at 60 V for 100 min and photographed under ultraviolet light using a charge-coupled device camera (The Gel Doc 1000 fluorescent gel documentation system, Bio-Rad Laboratories, Hercules, CA). The images were captured and stored in a computer for densitometric analysis using PC-image, an IBM-PC version of public domain software NIH Image, for one-dimensional gels according to the manual provided. The relative amounts of GHRH and SRIH mRNAs in hypothalamus samples were expressed as the ratio of densitometric measurements derived from PCR products of the target mRNA and internal control cRNA, respectively. Results from two or three repeated PCRs from each cDNA sample were averaged and then used for statistical analyses. All data are expressed as means ± SE and were subjected to one-way ANOVA or Student's t-test. When appropriate, post hoc analysis was done using Student-Newman-Keuls test. The results of SRIH mRNA levels were analyzed by means of two-way ANOVA. For each time point, the brain region (repeated measure) and the treatment (independent samples) were the two factors. The Student-Newman-Keuls test was used as a post hoc test to identify the brain area where significant differences occurred. In all tests, an  level of P < 0.05 was considered statistically significant.

	RESULTS

Top Abstract Introduction Materials & Methods Results Discussion References

Examples of the RT-PCR-amplified hypothalamic GHRH mRNA and its internal standard cRNA are shown in Fig. 2. After agarose gel electrophoresis, two expected bands were visible, representing the 294-bp fragment of wild-type GHRH mRNA and the 234-bp mutant cRNA internal standard, respectively. In control animals, the expression of hypothalamic GHRH mRNA displayed a time-dependent variation [F(4,36) = 19.1, P < 0.0001]. The GHRH mRNA values in samples obtained during the light phase (1600 and 1700) were significantly higher than values in samples obtained during the dark period (2000 and 2200) (Student-Newman-Keuls test, P < 0.05). Similar time-dependent variations were evident in samples taken after sleep deprivation (Fig. 3). Hypothalamic GHRH mRNA levels were significantly increased after 8 or 12 h of sleep deprivation (t-test, P < 0.05) (Fig. 3). Although the magnitude of sleep deprivation-induced increase in GHRH mRNA was greater after 8 h of sleep deprivation, the relative magnitude of the GHRH mRNA increase was greater after 12 h of sleep deprivation (91.4 vs. 29.7% increase, respectively). The elevated expression of GHRH mRNA was not significant after either 1 or 2 h of recovery after 8 h of sleep deprivation or after 2 h of recovery after 12 h of sleep deprivation (Fig. 3).

View larger version (19K): [in this window] [in a new window]   Fig. 2.   Gel electrophoresis of RT-PCR amplified hypothalamic GHRH mRNA (Wt) and its corresponding internal standard cRNA (Mu) in response to 8 h of sleep deprivation (8h SD), 1 h recovery after 8 h of sleep deprivation (8h SD 1h Re), 2 h recovery after 8 h of sleep deprivation (8h SD 2h Re), 12 h of sleep deprivation (12h SD), 2 h recovery after 12 h of sleep deprivaton (12h SD 2h Re), and their corresponding time-matched control samples (C). PCR products for GHRH mRNA and its internal standard are 294 and 234 bp, respectively.

View larger version (17K): [in this window] [in a new window]   Fig. 3.   Effects of 8 and 12 h of sleep deprivation and recovery sleep on GHRH mRNA expression in rat hypothalamus. Amounts of GHRH mRNA in experimental (filled bars) and time-matched control (open bars) samples are expressed as ratios of densitometric measurements of samples to that of cRNA internal standard. Data are presented as means ± SE of values obtained after averaging 2 or 3 PCRs of the appropriate cDNA (* P < 0.05; t-test).

Sleep deprivation also altered SRIH mRNA levels (Fig. 4, Table 2). ANOVA was performed on the mRNA values measured at the termination of the 8 h of sleep deprivation. There were significant differences among the four regions studied [F(3,27) = 39.43, P < 0.05] and between the control and the sleep-deprived rats [F(1,9) = 3.44, P < 0.05] with the hypothalamus the only area where sleep deprivation had an effect on SRIH mRNA. SRIH mRNA decreased significantly in response to sleep deprivation (Fig. 4). Sleep deprivation-induced changes in SRIH mRNA levels were not found after 1 or 2 h of recovery sleep in any region.

View larger version (20K): [in this window] [in a new window]   Fig. 4.   Top: changes in hypothalamic mRNA levels at the end of an 8-h sleep deprivation (filled bars) and after 1- and 2-h recovery after sleep deprivation (filled bars). Control, undisturbed rats were killed at the corresponding time points of the light period (1600, 1700, and 1800, open bars). n = 10 rats for each group. Bottom: hypothalamic SRIH mRNA levels at 1 h after light onset (open bars) and at 1 h after dark onset (filled bars) in the hypothalamus (HT), hippocampus (HC), brain stem (BS), and cerebral cortex (CT). n = 8 rats for each group. mRNA levels (mean ± SE) are expressed as ratios of densitometric measurements with respect to internal standards. *Significant differences (ANOVA followed by the Stedent-Newman-Keuls test, P < 0.05).

Comparisons of SRIH mRNA levels in samples obtained 1 h after light and 1 h after dark onset indicated that SRIH mRNA expression varied significantly as a function of both the brain area [F(3,21) = 10.7, P < 0.05] and the time of the day [F(1,7) = 11.9, P < 0.05]. The significant diurnal variations resulted from a significantly higher SRIH mRNA expression at 1 h after dark onset than in the morning in the hypothalamus (Student-Newman-Keuls test) (Fig. 4). In contrast, significant differences were not found in the SRIH mRNA levels between the morning and the night in the hippocampus, brain stem, or the cerebral cortex.

	DISCUSSION

Top Abstract Introduction Materials & Methods Results Discussion References

Current findings are consistent with previous reports describing the diurnal variation of GHRH mRNA expression in the rat hypothalamus using RT-PCR (4) or in the periventromedial hypothalamic area using in situ hybridization (38). Furthermore, they are also consistent with the study showing increased GHRH mRNA levels in the paraventricular nucleus after 6 h of sleep deprivation (38). Our results extend those observations by demonstrating that the sleep deprivation-induced changes are superimposed on the diurnal changes in GHRH mRNA and that the hypothalamic GHRH mRNA levels begin to recover toward control values after sleep recovery after sleep deprivation.

Unlike many other neuropeptides, the distribution of GHRHergic neurons in the rat brain is mostly restricted to a small area in the hypothalamus (6, 19, 32). GHRHergic neurons are found in the arcuate nucleus and in a few extra-arcuate locations. GHRHergic neurons in the arcuate nucleus project to the median eminence and are the major source of GHRH involved in GH release (reviewed in Ref. 2). GHRH mRNA expression in the arcuate nucleus exhibits an ultradian rhythm (41), but it does not seem to be affected by sleep deprivation (38). In contrast, extra-acuate GHRHergic neurons are suggested to mediate sleep effects. Toppila et al. (38) reported that GHRH mRNA expression in the periventromedial area exhibits diurnal variations, and GHRH mRNA levels in the paraventricular nucleus display sleep deprivation-induced changes. Currently it is not possible to determine whether the GHRHergic neurons in the periventromedial area or in the paraventricular nucleus, or both, are involved in NREMS regulation. The diurnal changes observed in the periventromedial area may be secondary to diurnal rhythms other than sleep. Similarly the sleep deprivation-induced changes in paraventricular GHRH mRNA may be due to sleep deprivation-induced changes in behavior and/or stress, rise in metabolic rate and body temperature (9), etc.

Extra-arcuate GHRHergic neurons project to the preoptic area (32). It is well established that the preoptic area of the basal forebrain plays a fundamental role in the regulation of NREMS (reviewed in Ref. 36). For example, low- or high-frequency electrical stimulation of the preoptic area elicits sleep (11, 35). Lesions of the preoptic area cause severe acute sleep loss (18) and increases in motor activity (22). Inactivation of preoptic neuronal activity with the gamma aminobutyric acid receptor agonist muscimol also inhibits spontaneous sleep (14). Enhancements in NREMS could thus result from GHRH modulating the function of the somnogenic structures in the preoptic area. Microinjection of GHRH directly into the preoptic area elicits enhanced NREMS (43), and microinjection of a GHRH antagonist into the same area suppresses spontaneous NREMS and attenuates the sleep rebound after sleep deprivation (J. Zhang, F. Obál, Jr., and J. M. Krueger, unpublished data). These observations suggest a direct effect of GHRH on NREMS-promoting mechanisms in this region.

Sleep deprivation causes depletion of hypothalamic content of GHRH. Furthermore, hypothalamic GHRH content increases gradually during sleep recovery after sleep deprivation, although GHRH content is still below the baseline value after 2 h of recovery (10). This gradual increase of peptide content is likely to result from an enhanced translation of the increased mRNA observed at the termination of sleep deprivation. Stimulation of the transcription of GHRH mRNA may represent a regulatory response to the depletion of intracellular GHRH stores due to an enhanced GHRH release during sleep deprivation. The sleep deprivation-induced increases in hypothalamic GHRH mRNA are likely to manifest themselves eventually as increases in GHRH production, which probably contributes to sleep propensity during sleep deprivation.

SRIH and GHRH mRNA levels varied reciprocally: SRIH mRNA decreased at the termination of sleep deprivation when GHRH mRNA was high. Also, with respect to SRIH mRNA levels in the morning, SRIH mRNA increased after dark onset, whereas GHRH mRNA is high in the morning and drops to very low levels at night (4). These changes in SRIH mRNA support the hypothesis that GHRH and SRIH play opposite roles in sleep regulation as they do in the control of GH secretion. SRIH may inhibit sleep via suppressing GHRHergic neurons or interacting with GHRH effects in the basal forebrain. Unlike GHRH, SRIH is a pleiotropic neurotransmitter in the brain including in the hypothalamus. The diurnal variation of SRIH in our samples is phase shifted by 180° from the rhythm of SRIH in the suprachiasmatic nucleus, where SRIH peaks around the onset of the light period (23). In situ hybridization failed to show diurnal variations in SRIH mRNA in the arcuate nucleus or the periventricular nucleus (38). Also, in that study, there were no signs of sleep deprivation-induced alterations in SRIH mRNA in the periventricular nucleus, i.e., in the hypophysiotropic somatostatinergic neurons. Somatostatinergic neurons inside the arcuate nucleus are involved in the regulation of GH secretion; these neurons control the intra-arcuate GHRHergic neurons projecting to the median eminence (15). Instead of reducing, sleep deprivation stimulated SRIH mRNA in the arcuate nucleus in the in situ hybridization experiment. Currently, therefore, it is not clear where the somatostatinergic neurons reside that are responsible for the changes in mRNA found in our studies but these neurons may provide direct or indirect inputs for those extra-arcuate GHRHergic neurons that are important for sleep regulation.

In conclusion, hypothalamic GHRH mRNA expression increases in response to sleep deprivation. These results provide further evidence implicating GHRH in sleep regulation and suggest that a release of GHRH to the preoptic area of the basal forebrain might contribute to the increases in sleepiness characteristic of sleep deprivation. Our findings also suggest that there is a reciprocal interaction between GHRHergic and somatostatinergic activities in the modulation of sleep-wake activity.