Inhibition of Brain Interleukin-1 Attenuates Sleep Rebound After Sleep Deprivation in Rabbits

Satoshi Takahashi, Jidong Fang, Levente Kapás, Ying Wang and James M. Krueger

Department of Physiology and Biophysics, University of Tennessee, Memphis, TN 38163 USA

 

KEY WORDS: IL-1 soluble receptor, brain temperature, EEG slow-wave activity, REM sleep, cytokine

Abstract

It is hypothesized that interleukin-1 (IL-1) is involved in physiological sleep. If this hypothesis is correct, inhibition of IL-1 should attenuate sleep responses after sleep deprivation. We tested the effect of intracerebroventricular (i.c.v.) or intravenous (i.v.) injection of an IL-1 inhibitor, an IL-1 receptor fragment (IL-1RF), on sleep rebound after sleep deprivation in rabbits. Six hours of total sleep deprivation significantly increased non-rapid eye movement sleep (NREMS) and enhanced electroencephalogram (EEG) slow-wave activity (SWA) during NREMS. I.c.v. treatment with the IL-1RF (50 mg) significantly attenuated the sleep responses after sleep deprivation. Furthermore, 1.0 mg/kg i.v. injection of the IL-1RF significantly suppressed spontaneous NREMS in rabbits that were not sleep deprived. However, i.v. administration of the IL-1RF (1.0 mg/kg) failed to attenuate the sleep responses following the 6-h sleep deprivation period. These results support the hypothesis that central pools of IL-1 are important for physiological sleep regulation.

Introduction

Sleep is posited to be regulated by homeostatic and circadian rhythm processes (reviewed 2). Both these processes are composed of neuronal and humoral mechanisms. Perhaps the best evidence for a sleep homeostatic process is the observation that the amount of time spent in sleep increases after sleep deprivation. Sleep deprivation experiments were also useful in demonstrating the existence of an endogenous humoral somnogenic agent. Many laboratories, beginning with Ishimori (11) at the turn of the century, have shown that during sleep deprivation a somnogenic substance(s) accumulates in cerebrospinal fluid (CSF). If CSF from sleep-deprived animals is transferred to normal animals, the recipients sleep more than normal (reviewed 12, 15, 16). Within the past few years several candidate substances have been proposed as being the active somnogenic agent in CSF obtained from sleep-deprived animals; the list includes interleukin-1b (IL-1b), tumor necrosis factor-a (TNFa), growth hormone releasing hormone (GHRH), uridine, prostaglandin D2 and adenosine (reviewed 12, 15, 16). All of these substances have also been implicated in physiological sleep regulation as well.

IL-1b is one of the better characterized sleep regulatory substances. Administration of exogenous IL-1b via intraventricular (i.c.v.), intravenous (i.v.) or intraperitoneal (i.p.) routes induces relatively large increases in non-rapid eye movement sleep (NREMS) in rats (20, 30), rabbits (14), mice (5, 6), cats (25), and monkeys (8) and sleepiness in humans (4). Inhibition of IL-1b actions using anti-IL-1b antibodies, the IL-1 soluble receptor, the IL-1 receptor antagonist, or a-melanocyte stimulating hormone inhibits spontaneous sleep (reviewed 15, 16). Similarly, substances which inhibit IL-1b production, such as prostaglandin E2 (13), corticotropin releasing hormone, and IL-10 (21), also inhibit spontaneous sleep (reviewed 15, 16). NREMS responses induced by mild increases in ambient temperatures (Takahashi, et al., unpublished) or by bacterial products (28) are attenuated if animals are pretreated with inhibitors of IL-1b, such as the IL-1 soluble receptor or a synthetic fragment of the soluble IL-1 receptor (IL-1RF). IL-1b and other members of the IL-1 family of molecules are constitutively expressed in brain (reviewed 16). There is a diurnal rhythm of IL-1b mRNA in the hypothalamus, hippocampus and cortex in rats (26). Further, hypothalamic and brain stem IL-1b mRNA levels increase during sleep deprivation (18). CSF levels of IL-1-like bioactivity also vary in phase with sleep-wake cycles (17). Plasma levels of IL-1b or the ability of circulating leukocytes to produce IL-1 vary with sleep-wake cycles (19, 31) and increase during sleep deprivation (19). Collectively, these data suggest IL-1b is involved in physiological sleep regulation and possibly in sleep responses induced by sleep deprivation. This report provides direct evidence that IL-1 is involved in the excess NREMS induced by sleep deprivation. We now show that by pretreating rabbits with an IL-1 inhibitor (the IL-1RF) prior to sleep deprivation, expected NREMS responses are blunted.

Materials and Methods

1. Materials.

IL-1RF was synthesized by Dr. J. M. Seyer (Department of Biochemistry, University of Tennessee, Memphis). Its amino acid sequence is YCLRIKISAK; which corresponds to amino acid residues 86-95 of the human type I IL-1 receptor (29). It contains the IL-1-binding sites of the parent molecule and inhibits IL-1 actions in vitro and in vivo (29). Indeed, the IL-1RF blocks IL-1b-induced sleep and fever in rabbits (28). Substances were dissolved in pyrogen-free isotonic NaCl (PFS) (Abbott Laboratories, North Chicago, IL).

2. Animals.

Male New Zealand White Pasteurella-free rabbits (3.5-5.5 kg) were surgically implanted with electroencephalographic (EEG) electrodes, a calibrated 30 k? brain thermistor (Model #44008, Omega Engineering, Stanford, CT), and an i.c.v. guide cannula under ketamine-xylazine (35 and 5 mg/kg) anesthesia as previously described (28). In brief, the stainless steel EEG electrodes were placed over the frontal and parietal cortices. The thermistor was implanted on the dura mater over the parietal cortex to measure brain temperature (Tbr). The i.c.v. guide cannula was placed in the left lateral ventricle stereotaxically. The leads from the EEG electrodes and the thermistor were routed to a Teflon pedestal. The pedestal, leads, and the guide cannula were attached to the skull with dental acrylic (Duz-All, Coralite Dental Product, Skokie, IL). After a 2 week recovery period, the rabbits were placed in sleep-recording chambers (Hot Pack 352600, Philadelphia, PA) for at least one 24-h habituation. The animals were kept on a 12:12 h light/dark cycle (lights on at 06:00 h) at 21 ± 1°C ambient temperature. Water and food were available ad libitum throughout the experiment.

3. Apparatus and Recording.

The rabbits were allowed relatively unrestricted movement inside the recording chamber. A flexible tether connected the electrode and thermistor leads to an electronic swivel. Ultrasonic detectors (Biomedical Instrumentation, University of Tennessee, Memphis) were used to detect body movements. The leads from the electronic swivel and movement detectors were routed to Grass Model #7D polygraphs in an adjacent room. The amplified signals were digitized at a frequency of 128 Hz for EEG, and at 2 Hz for Tbr and motor activity. Tbr data were saved on computer at 10 s intervals. The hourly averages of Tbr values collected at 10 min intervals were used for statistical analysis. On-line Fourier analysis of the EEG was performed. The vigilance states: wakefulness (W), NREMS and rapid eye movement sleep (REMS) were visually identified off-line in 10-s epochs using criteria previously reported (28). Briefly, W was characterized by fast low-amplitude EEG waves, gradually increasing Tbr, and high incidence of gross body movements. NREMS was associated with slow high-amplitude EEG waves, slowly decreasing Tbr, and lack of body movements. In contrast, REMS was characterized by fast low-amplitude EEG waves, appearance of rhythmic theta EEG, rapidly increasing Tbr at REMS onset, and lack of body movement. The amount of time spent in each vigilance state was calculated for 1-h intervals and for the entire recording periods. The sleep time for 1-h intervals was used for statistical analyses, and the amount of sleep time for 3-h intervals was used for figures (the last period was a 2-h time block). Furthermore, the EEG power density values were summed in four frequency bands for each 10-s epoch. Hourly averages of the EEG delta wave (0.5-4.0 Hz) activity during NREMS, also called EEG slow-wave activity (SWA), was determined. The percent change in SWA from time-matched values of the baseline period was calculated.

Sleep deprivation was achieved by gentle handling on a table. During sleep deprivation, the EEG and Tbr were recorded from each animal to allow quantitation of effectiveness of the deprivation procedure.

Experimental Protocol

1. Effects of I.C.V. IL-1RF on Sleep Rebound After Sleep Deprivation.

Sleep recordings were performed under three different conditions. In all conditions, injections were performed between 08:45 h and 09:15 h. For baseline recording, 10 rabbits were injected with 50 ml PFS i.c.v. EEG, Tbr and motor activity were recorded for 23 h after injections without sleep deprivation in the environmental chamber (non-sleep deprivation with PFS-i.c.v.). The effects of sleep deprivation alone were evaluated by 6 h sleep deprivation after i.c.v. injection of 50 ml PFS (sleep deprivation with PFS-i.c.v.). The effects of i.c.v. injection of the IL-1RF on Sleep deprivation-induced sleep responses were studied by injecting 50 mg IL-1RF in a volume of 50 ml PFS at the beginning of the 6 h sleep deprivation (sleep deprivation with IL-1RF-i.c.v.). Sleep deprivation started immediately after injection of the IL-1RF. After 6 h sleep deprivation, the rabbits were placed back into the sleep recording chambers and the EEG, Tbr and motor activity were recorded for an additional 17 h. A minimum of one week was allowed between two deprivation periods. Five rabbits received sleep deprivation with PFS-i.c.v. first, and the other five received sleep deprivation with IL-1RF-i.c.v. first.

2. Effects of I.V. Injection of the IL-1RF on Spontaneous Sleep in Rabbits.

IL-1RF (0.01 mg/kg, n = 7; 0.1 mg/kg, n = 8 or 1.0 mg/kg, n = 7) was injected i.v. via the left marginal ear vein in a volume of 0.5 ml/kg PFS between 08:45 and 09:15. On a separate control day, the same rabbits received i.v. injection of the same volume of PFS. After injection, EEG, Tbr and motor activity were recorded for 23 h.

3. Effects of I.V. IL-1RF on Sleep Rebound After Sleep Deprivation.

For baseline recording, 6 rabbits were injected with 0.5 ml/kg PFS i.v. EEG, Tbr and motor activity were recorded for 23 h after injections without sleep deprivation in the environmental chamber (non-sleep deprivation with PFS-i.v.). The effects of sleep deprivation alone were evaluated by 6 h sleep deprivation after i.v. injection of 0.5 ml/kg PFS (sleep deprivation with PFS-i.v.). The effects of i.v. injection of the IL-1RF on sleep deprivation-induced sleep responses were studied by injecting 1.0 mg/kg IL-1RF in a volume of 0.5 ml/kg PFS at the beginning of the 6 h sleep deprivation (sleep deprivation with IL-1RF-i.v.). Sleep deprivation started immediately after injection of the IL-1RF. After 6 h of sleep deprivation, the rabbits were placed back into the recording chambers and the EEG, Tbr and motor activity were recorded for an additional 17 h. A minimum of one week was allowed between two deprivation periods. Three rabbits received sleep deprivation with PFS-i.v. first, and the other three received sleep deprivation with IL-1RF-i.v. first.

Statistical Analysis

All data were analyzed using two-way analysis of variance for repeated measures (ANOVA). When treatment effects were significant, the post hoc Student-Newman-Keuls (SNK) test was performed. A significant level of p < 0.05 was accepted.

Results

Effects of I.C.V. IL-1RF on Sleep After Sleep Deprivation

NREMS was significantly increased after 6-h of sleep deprivation (Table 1, Fig. 1). The enhanced NREMS was evident during the first h after sleep deprivation and remained above corresponding control values for the remainder of the recording period. Concurrent with the enhanced NREMS was a significant increase in EEG SWA. In contrast, duration of REMS after sleep deprivation was not significantly affected by the sleep deprivation procedure but quickly returned to control values as soon as the animals were left undisturbed (Fig. 1).

If the rabbits were pretreated with the IL-1RF prior to sleep deprivation the expected excess NREMS response was significantly attenuated compared to the response induced by sleep deprivation alone (Table 1, Fig. 1). The effects of the IL-1RF were clearest during the last 11 hours of recording; during that period NREMS in the sleep deprived plus IL-1RF-treated rabbits was indistinguishable from that obtained from non-sleep deprived rabbits, whereas those rabbits subjected only to sleep deprivation, exhibited excess NREMS during this period (Fig. 1). The sleep deprivation-induced increases in EEG SWAs were also significantly attenuated if animals were pretreated with the IL-1RF (Table 1, Fig. 1). This IL-1RF-attenuated EEG SWA was most evident during the first 6 h after sleep deprivation (Fig. 1). Nevertheless, EEG SW values in IL-1RF-sleep deprived rabbits remained significantly above values obtained during control recordings.

REMS significantly increased in the i.c.v. IL-1RF-treated sleep deprived rabbits despite the fact that after sleep deprivation alone there was no increase in REMS in this group of rabbits (Table 1, Fig. 1). Tbr was not affected by i.c.v. IL-1RF treatment (Table 1, Fig. 1).

Effects of I.V. Injection of the IL-1RF

I.v. injection of the two lower doses of the IL-1RF failed to significantly affect any of the parameters measured in this study (Table 2, Fig. 2). In contrast, the highest dose tested, 1 mg/kg, inhibited NREMS (Table 2, Fig. 2). The loss of NREMS induced by i.v. IL-1RF was evident in hs 3-6 postinjection then persisted for the next 12 hs (Fig. 2). During this period of IL-1RF-induced NREMS inhibition, REMS remained near control values. Although Tbr was not significantly affected across the 23 h recording period, there was a transient increase in Tbr during hs 2-4 post-IL-1RF i.v. injection; this affect was present after the 0.1 mg/kg dose but was most obvious after the 1.0 mg/kg dose (Fig. 2).

Effects of I.V. IL-1RF on Sleep After Sleep Deprivation

In the group of animals used in this study, 6 h of sleep deprivation induced a significant increase in NREMS duration and EEG SWA (Table 3, Fig. 3). These effects began during the first h after sleep deprivation then persisted throughout most of the remaining recording period. In this group of rabbits, REMS also significantly increased after sleep deprivation. Tbr was elevated during the sleep deprivation period but returned to control levels as soon as animals were left undisturbed. None of the sleep deprivation-induced affects were significantly affected by pretreating the rabbits with the IL-1RF i.v. (Table 3, Fig. 3).

Discussion

Current results clearly suggest that central pools of IL-1 may be involved in NREMS and EEG SWA responses induced by sleep deprivation since both these responses were attentuated if the IL-1RF was given centrally to rabbits before sleep deprivation began. These results are consistent with previous results obtained in rats (22) and rabbits (23) in which it was shown that central administration of an anti-IL-1 antibody attenuated sleep deprivation-induced NREMS responses. The present study extends those studies by: a) using a longer period of sleep deprivation 6 hs vs. 4 hs (rabbits) or 3 hs (rats); b) using a different inhibitor of IL-1 (the IL-1RF) which has longer lasting affects on NREMS; and c) determining the role that central vs. systemic pools of IL-1 may have in sleep deprivation-induced NREMS rebound.

Previously we published the results of i.c.v. administration of a 50 mg dose of the IL-1RF on spontaneous sleep and Tbr (28). This dose of i.c.v. IL-1RF inhibited NREMS over a 23 h recording period. It is thus possible that in the current study that the effects of sleep deprivation, tending to increase NREMS, and the effects of the IL-1RF, tending to decrease NREMS, are simply additive. However, this seems unlikely for several reasons. First, i.c.v. administration of the IL-1RF does not affect EEG SWAs yet after sleep deprivation i.c.v. IL-1RF attenuated sleep deprivation-induced increases in EEG SWAs. Further, i.c.v. administration of IL-1RF to normal animals does not affect duration of REMS (28). In the group of rabbits used for the i.c.v. IL-1RF study, 6 hs of sleep deprivation did not result in an increase in REMS; this finding is consistent with two other previous studies in rabbits in which either 4 hs or 6 hs of sleep deprivation failed to affect duration of REMS. Yet, if the animals in the current study were pretreated with the IL-1RF there was a significant increase in REMS after sleep deprivation (Fig. 1). At the very least, this shows that the IL-1RF does not have a non-specific activation effect and could suggest that, if NREMS pressure is reduced by binding of the IL-1RF to IL-1, then REMS rebound can be manifested. Regardless of these possibilities results do suggest that the effects of sleep deprivation and i.c.v. IL-1RF are not simply additive. Similar conclusions were made in studies in which anti-IL-1 antibodies were used to attenuate sleep rebound after sleep deprivation (22, 23). In those studies the anti-IL-1 antibodies by themselves inhibited spontaneous NREMS but did not affect duration of REMS nor EEG SWAs yet in rabbits the antibodies attenuated sleep deprivation-induced NREMS and EEG SWA responses (23).

Current results suggest that circulating IL-1 has little role in sleep deprivation-induced NREMS rebound since i.v. administration of the IL-1RF failed to significantly affect sleep deprivation-induced NREMS rebound. This was surprising since there is evidence that systemic IL-1 can affect sleep or is altered by sleep. Thus, the highest i.v. dose of the IL-1RF used in this study inhibited NREMS (Fig. 2). Further, i.v. or i.p. injection of IL-1 enhances NREMS (5, 14). Sleep deprivation of rabbits (23) or humans (10, 19) induces increases in circulating IL-1. Further, the somnogenic actions of low doses of i.p. IL-1 (but not higher somnogenic doses) are blocked by vagotomy (9). Nevertheless, current results are consistent with preliminary results from our laboratory showing that sleep deprivation has similar effects on NREMS in both vagotomized and control rats (Hansen, unpublished). Thus, although moderate sleep deprivation may affect both central and systemic pools of IL-1, central IL-1 appears to play the dominant role in sleep deprivation-induced NREMS responses.

Perspectives

In a previous study we showed that sleep deprivation-induced sleep responses could also be attenuated if rabbits were pretreated with a blocker of TNF (27). In those studies, as in the current study, NREMS rebound after sleep deprivation was not completely suppressed thereby suggesting that other substances are also involved. We suggested that IL-1 may be one of those substances and indeed those results prompted the current study. There is an independent set of data suggesting that TNFa, like IL-1b, is involved in NREMS regulation (reviewed 15, 16). Briefly, administration of TNF to rabbits, rats or mice enhances NREMS while inhibition of TNF using either antibodies to TNF or TNF soluble receptors inhibits spontaneous sleep (reviewed 15, 16). Further, TNF mRNA (3) and TNF (7) expression in brain is highest in rats during daylight hours. Mice lacking the 55-kD TNF receptor sleep less than control strains (6). Since IL-1 and TNF can induce each other’s production (1, 24) it is reasonable to propose that both are part of a biochemical network involved in sleep regulation. In fact, preliminary data from our laboratory indicate that if rabbits are pretreated with the IL-1RF, TNF-induced NREMS responses are attenuated. Conversely, if TNF is inhibited using a TNF soluble receptor, IL-1-induced NREMS responses are attenuated. Perhaps cytokine regulation of NREMS is similar to cytokine involvement in immunocyte regulation in that it is characterized by redundant regulatory pathways sharing multiple pleiotropic substances.

In conclusion, current results are consistent with the hypothesis that IL-1b is responsible, in part, for the NREMS rebound induced by sleep deprivation.

Acknowledgments

The secretarial assistance of Ms. Maria Swayze is gratefully acknowledged. This work was supported, in part, by the National Institutes of Health grants (NS-27250, NS-25378, and NS-31453) and the Naito Foundation. The permanent address of Dr. Satoshi Takahashi is: Department of Anesthesiology, University of Hirosaki School of Medicine, Hirosaki 036, Japan and of Dr. Levente Kapás is: Department of Biological Sciences, Fordham University, Bronx, New York 10458-5170.

References

1. Bachwich, P. R., S. W. Chensue, J. W. Larrick, and S. L. Kunkel. Tumor necrosis factor stimulates interleukin-1 and prostaglandin E2 production in resting macrophages. Biochem. Biophys. Res. Commun. 136: 94-101, 1986.

2. Borbély, A. A. and I. Tobler. Endogenous sleep-promoting substances and sleep regulation. Physiol. Rev. 69: 605-670, 1989.

3. Bredow, S., P. Taishi, N. Guha-Thakurta, F. Obál, Jr., and J. M. Krueger. Diurnal variations of tumor necrosis factor-a mRNA and a-tubulin mRNA in rat brain. (submitted).

4. Dinarello, C. A. Interleukin-1 and interleukin-1 antagonism. Blood 77: 1627-1652, 1991.

5. Fang, J., L. Kapás, Y. Wang, and J. M. Krueger. The TNF 55kD receptor and the IL-1 type I receptor are involved in physiological sleep regulation. Soc. for Neurosci. Absts. 22: 147, 1996.

6. Fang, J., Y. Wang, and J. M. Krueger. The TNF 55-kD receptor and the IL-1 type I receptor are involved in physiological sleep regulation. Sleep Res. 26 (in press), 1997.

7. Floyd, R. A. and J. M. Krueger. Diurnal variations of TNFa in the rat brain. NeuroReport (in press).

8. Friedman, E. M., S. Boinski, and C. L. Coe. Interleukin-1 induces sleep-like behavior and alters call structure in juvenile rhesus macaques. Am. J. Primatol. 35: 145-153, 1995.

9. Hansen, M. K. and J. M. Krueger. Subdiaphragmatic vagotomy attenuates interleukin-1-induced sleep. Sleep Res. 26 (in press), 1997.

10. Hohagen, F., J. Timmer, A. Weyerbrock, R. Fritsch-Montero, U. Ganter, S. Krieger, M. Berger, and J. Bauer. Cytokine production during sleep and wakefulness and its relationship to cortisol in healthy humans. Neuropsychobiology 28: 9-16, 1993.

11. Ishimori, K. True cause of sleep - A hypnogenic substance as evidenced in the brain of sleep-deprived animals. Tokyo Igakkai Zasshi 33: 429-457, 1909.

12. Inoué, S. Biology of Sleep Substances. CRC Press, Inc., Boca Raton, Florida, 1989.

13. Krueger, J. M., L. Kapás, M. R. Opp, and F. Obál, Jr. Prostaglandins E2 and D2 have little effect on rabbit sleep. Physiol. and Behav. 51: 481-485, 1992.

14. Krueger, J. M., J. Walter, C. A. Dinarello, S. M. Wolff, and L. Chedid. Sleep-promoting effects of endogenous pyrogen (interleukin-1). Am. J. Physiol. 246: R994-R999, 1984.

15. Krueger, J. M. and F. Obál, Jr. Sleep Factors. In: Sleep and Breathing, N. A. Saunders and C. E. Sullivan (Eds.), Marcel Dekker, Inc., New York, pp. 79-112, 1994.

16. Krueger, J. M. and J. A. Majde. Microbial products and cytokines in sleep and fever regulation. Crit. Rev. Immunol. 14: 355-379, 1994.

17. Lue, F. A., M. Bail, J. Jephthah-Ocholo, K. Carayanniotis, R. Gorczynski, and H. Moldofsky. Sleep and cerebrospinal fluid interleukin-1 like activity in the cat. Intern. J. Neurosci. 42: 179-183, 1988.

18. Mackiewicz, M., P. J. Sollars, M. D. Ogilvie, and A. I. Pack. Modulation of IL-1b gene expression in the rat CNS during sleep deprivation. NeuroReport 7: 529-533, 1996.

19. Moldofsky, H., F. A. Lue, J. Eisen, E. Keystone, and R. M. Gorczynski. The relationship of interleukin-1 and immune functions to sleep in humans. Psychosom. Med. 48: 309-318, 1986.

20. Opp, M. R., F. Obál, Jr., and J. M. Krueger. Interleukin-1 alters rat sleep: Temporal and dose-related effects. Am. J. Physiol. 260: R52-R58, 1991.

21. Opp, M. R., E. M. Smith, and T. K. Hughes. Interleukin-10 acts in the central nervous system of rats to reduce sleep. J. Neuroimmunol. 60: 165-168, 1995.

22. Opp, M. R. and J. M. Krueger. Anti-interleukin-1b reduces sleep and sleep rebound after sleep deprivation in rats. Am. J. Physiol. 266: R688-R695, 1994.

23. Opp, M. R. and J. M. Krueger. Interleukin-1 is involved in responses to sleep deprivation in the rabbit. Brain Res. 639: 57-65, 1994.

24. Phillip, R., and L. B. Epstein. Tumor necrosis factor as an immunomodulator and mediator of monocyte cytotoxicity induced by itself, gamma interferon and interleukin-1. Nature 323: 86-89, 1986.

25. Susic, V. and S. Totic. "Recovery" function of sleep: Effects of purified human interleukin-1 on the sleep and febrile response of cats. Met. Brain Dis. 4: 73-80, 1989.

26. Taishi, P., S. Bredow, N. Guha-Thakurta, F. Obál, Jr., and J. M. Krueger. Diurnal variations of interleukin-1b mRNA and b-actin mRNA in rat brain. J. Neuroimmunol. 1996 (in press).

27. Takahashi, S., L. Kapás, J. M. Seyer, Y. Wang, and J. M. Krueger. Inhibition of tumor necrosis factor attenuates physiological sleep in rabbits. NeuroReport 7: 642-646, 1996.

28. Takahashi, S., L. Kapás, J. Fang, Y. Wang, J. M. Seyer, and J. M. Krueger. An interleukin-1 receptor fragment inhibits spontaneous sleep and muramyl dipeptide-induced sleep in rabbits. Am. J. Physiol. 271: R101-R108, 1996.

29. Tanihara, M., Y. Suzaki, C. Fujiwara, C. Abe, T. Suda, and Y. Mizushima. A synthetic peptide corresponding to 86-93 of the human type I IL-1 receptor binds human recombinant IL-1 (a and b) and inhibits IL-1 actions in vivo and in vitro. Biochem. Biophys. Res. Commun. 188: 912-920, 1992.

30. Tobler, I., A. A. Borbély, M. Schwyzer, and A. Fontana. Interleukin-1 derived from astrocytes enhances slow-wave activity in sleep EEG of the rat. Eur. J. Pharmacol. 104: 191-192, 1984.

31. Uthgenannt, D., D. Schoolman, R. Pietrowsky, H. L. Fehm, and J. Born. Effects of sleep on the production of cytokines in humans. Psychosom. Med. 57: 97-104, 1995.

Figure Legends

Fig. 1. Intracerebroventricular (i.c.v.) injection of an interleukin-1 receptor fragment (IL-1RF) attenuates sleep deprivation-induced increases in non-rapid eye movement sleep (NREMS) and electroencephalographic (EEG) slow-waves in rabbits. Rabbits (n = 10) were recorded from after i.c.v. injection of pyrogen free saline (PFS) (open triangle; control), 6 hs of sleep deprivation plus i.c.v. PFS (open circles) or after 6 hs of sleep deprivation plus 50 mg i.c.v. of the IL-1RF (closed circles). The EEG, brain temperature (Tbr) and EEG slow-wave activity (SWA) were recorded during the 6 hs of sleep deprivation and for 17 hs thereafter. Upper left bar indicates 6 h of sleep deprivation.

Fig. 2. Intravenous (i.v.) injection of the interleukin-1 receptor fragment (IL-1RF) corresponding to amino acid residues 86-95 of the IL-1 type I receptor inhibits non-rapid eye movement sleep (NREMS) in rabbits. Three doses (0.01 mg/kg, n = 7; 0.1 mg/kg, n = 8; and 1.0 mg/kg, n = 7) of the IL-1RF were injected i.v. (closed circles) and the electroencephalographic (EEG) and brain temperature (Tbr) were recorded for the next 23 hrs. For control, the same rabbits were recorded from after i.v. injection of pyrogen free saline (open circles). Lights were off during hs 10-21.

Fig. 3. Intravenous (i.v.) injection of an interleukin-1 receptor fragment (IL-1RF) failed to affect sleep deprivation-induced changes in sleep and brain temperature (Tbr) in rabbits. Rabbits (n = 6) were recorded from after i.v. injection of pyrogen free saline (PFS) (control; open triangles), 6 hs of sleep deprivation plus i.v. injection of PFS (open circles), or 6 hs of sleep deprivation plus 1 mg/kg of the IL-1RF (closed circles).

Table 1. Effects of I.C.V. Injection of 50 mg IL-1RF on Sleep and Brain Temperature After Sleep Deprivation in Rabbits1

1Data were collected during a 17-h period after 6-h of sleep deprivation (SD); they are expressed as mean ± SE.
2Data were lost from two animals due to mechanical failure.

aSignificantly different from control (p < 0.001, two-way ANOVA; p < 0.01, SNK test).
bSignificantly different from sleep deprivation treatment alone (p < 0.001; p < 0.01, SNK test).
cSignificantly different from control (p < 0.01) and sleep deprivation treatment alone (p < 0.01).
dSignificantly different from control (p < 0.01, SNK test).
eSignificantly different from sleep deprivation treatment alone (p < 0.05, SNK test) and from control (p < 0.05, SNK test).

Table 2. Effects of I.V. Injection of IL-1RF on Spontaneous Sleep and Brain Temperature in Rabbits1

1Data were collected during a 23-h period after either saline or IL-1RF injection i.v.

2n = 6 (data were lost in 1 animal due to mechanical failure).

ap < 0.01 compared to saline injection.

NA = not available due to computer error.

Table 3. Effects of I.V. Injection of IL-1RF on Sleep and Brain Temperature After Sleep Deprivation in Rabbits1

1Data were collected during a 17-h period after 6-h of sleep deprivation (SD); they are expressed as mean ± SE.

ap < 0.01 compared to control (SNK test).

bp < 0.01 compared to control (SNK test).

cp < 0.05 compared to control (SNK test).

dp < 0.01 compared to control (SNK test).