Interleukin 1β (IL-1) is a key mediator of the acute phase response in an infected host and acts centrally to coordinate responses to an immune challenge, such as fever and increased non-rapid eye movement (NREM) sleep. The preoptic area (POA) is a primary sleep regulatory center in the brain: the ventrolateral POA (VLPO) and median preoptic nucleus (MnPN) each contain high numbers of c-Fos protein immunoreactive (IR) neurons after sleep but not after waking. We hypothesized that IL-1 mediates increased NREM sleep through activation of these sleep-active sites. Rats injected intracerebroventricularly with IL-1 (10 ng) at dark onset spent significantly more time in NREM sleep 4–5 h after injection. This increase in NREM sleep was associated with increased numbers of Fos-IR neurons in the MnPN, but not in the VLPO. Fos IR in the rostral MnPN was significantly increased 2 h post IL-1 injection, although the percentage of NREM sleep in the preceding 2 h was the same as controls. Fos IR was also increased in the extended VLPO 2 h postinjection. Finally, Fos IR in the MnPN did not differ significantly between IL-1 and vehicle-treated rats that had been sleep deprived for 2 h postinjection, but it was increased in VLPO core. Taken together, these results suggest that Fos IR in the MnPN after IL-1 is not independent of behavioral state and may require some threshold amount of sleep for its expression. Our results support a hypothesis that IL-1 enhances NREM sleep, in part, through activation of neurons in the MnPN of the hypothalamus.
- preoptic area
- rapid eye movement sleep
interleukin 1β (il-1) is a polypeptide that mediates many of the host responses to infection and inflammation (5), including fever, appetite suppression, and alterations in sleep (3, 25). IL-1, administered peripherally or intracerebroventricularly (icv), promotes non-rapid eye movement (NREM) sleep and inhibits rapid eye movement (REM) sleep in several species (for a review, see Ref. 15). IL-1 is also involved in physiological sleep regulation. For example, targeting the IL-1 system with receptor antagonists or antibodies inhibits spontaneous NREM sleep and blocks the increase in NREM sleep that follows sleep deprivation (27, 28).
Although IL-1 may enhance NREM sleep, irrespective of timing of administration, there are circadian influences that determine the full impact of IL-1 on sleep-wake behavior (16, 29). When administered into rats at dark onset, effective doses of IL-1 induce biphasic effects on NREM sleep; there is an initial increase in NREM sleep during the first hour, and a more prominent increase 4–5 h postinjection (26, 29). IL-1 may enhance NREM sleep through several mechanisms, including activation of nuclear factor-κB (NFκB), inducing nitric oxide production, adenosine release, and prostaglandin synthesis, and increasing growth hormone releasing hormone (GHRH) release (reviewed in Ref. 25).
There are several neuroanatomic sites at which IL-1 administration may modulate NREM sleep. IL-1 infused into the subarachnoid space enhances NREM sleep, probably by promoting localized prostaglandin D2 synthesis (41). IL-1 microinjected into the dorsal raphe nucleus enhances NREM sleep, likely by inhibition of serotonergic neurons (19). Another area of the brain potentially involved in the NREM sleep-inducing actions of IL-1 is the preoptic area (POA); IL-1 perfused into the POA increases discharge rates of sleep-active neurons and reduces discharge rates of wake-active neurons (1). The POA plays an important role in normal sleep-wake regulation. Lesions of the POA produce arousal (13, 18, 21, 40), and microinfusion of sleep-promoting agents into the POA promotes sleep (22, 42, 44). In particular, the ventrolateral preoptic area (VLPO) and the median preoptic nucleus (MnPN) contain a substantial number of sleep-active neurons (38, 39). Also, neurons in both VLPO and MnPN of rats exhibit immunoreactivity (IR) for the c-Fos protein after sustained sleep but not after sustained wakefulness (9, 10, 37).
Although IL-1 induces c-Fos expression in numerous brain regions, including the POA (14, 24), the potential involvement of the VLPO and the MnPN in mediating increases in NREM sleep induced by IL-1 is unknown. The aim of our study, therefore, was to determine whether the MnPN and VLPO are activated in association with IL-1-induced increases in NREM sleep. To test this hypothesis, we injected IL-1 intracerebroventricularly (icv) into rats at dark onset, killed them either 2 or 5 h later, and quantified the extent of Fos IR in the VLPO and the rostral and caudal portions of the MnPN. These times were selected to capture both the first phase of the biphasic sleep response to IL-1, which ends within 2 h, and the second phase, which is maximal at 5 h, postinjection (29). To determine the impact of IL-1 on c-Fos activation in the MnPN and VLPO independent of IL-1-induced changes in sleep, we included a third group of rats that were sleep deprived (SD) for 2 h post IL-1 injection and then killed. We now report that IL-1 administered intracerebroventricularly enhances NREM sleep and increases the number of Fos-IR neurons in the MnPN, but not in VLPO, in spontaneously sleeping rats in the dark phase, 4–5 h after injection.
MATERIALS AND METHODS
Thirty-six male Sprague-Dawley rats, weighing 280–350 g at the beginning of the experiment, were used. All experimental procedures were approved by the Animal Care and Use Committee at the Veterans Affairs Greater Los Angeles Healthcare System and were conducted according to the guidelines of the National Research Council. Rats were anesthetized with ketamine/xylazine (80/10 mg/kg ip) and surgically implanted with chronic cortical EEG and dorsal neck electromyogram (EMG) electrodes for determining arousal state. Briefly, stainless steel screw electrodes were implanted in the skull for EEG recordings and flexible, insulated stainless steel wires were threaded into neck muscles for EMG recording. A stainless steel guide tube with an outer diameter (OD) of 0.6 mm was stereotaxically placed deep in the cortex, according to coordinates of Paxinos and Watson (30); anterioposterior (AP), +3 mm from bregma; lateral, −3 mm from bregma; dorsoventral, 4 mm below the skull, through which a thermistor (bead diameter 0.41 mm; Thermometrics, Edison, NJ) was inserted for recording brain temperature. A stainless steel guide cannula (OD 0.5 mm) was chronically placed in the right lateral ventricle and sealed with a removable obdurator (OD 0.3 mm). Leads from the electrodes and the thermistor were soldered to a small Amphenol connector, and the complete assembly was anchored to the skull with dental acrylic.
Rats were allowed a recovery period of between 4 and 7 days after surgery, during which time they were housed in individual cages, in a temperature-controlled recording chamber (23 ± 1°C). They were maintained on a 12:12-h light-dark cycle (lights on at 0600) throughout the experiment. Rats had ad libitum access to food and water. After recovery from surgery, to assess cannula patency and localization (29), rats were injected with ANG II [500 ng in pyrogen-free saline (PFS) icv]; ANG II elicits a drinking response by stimulating preoptic structures (6). Three rats that did not show a drinking response after angiotensin treatment were still used in the experiment, as vehicle-treated controls. The remaining animals were randomly assigned to experimental and control groups. Placement of the cannula in the right lateral ventricle was histologically confirmed at the end of the experiment.
Rats were handled daily for at least 5 days before the experiment. For 3 consecutive days before the experiment, they were adapted to the experimental procedures. Rats were connected, for 3–5 h each day, to a recording cable that was lightly suspended above them by a counter-weighted beam. The cable connected the miniature connector on the animal’s head to AC-coupled amplifiers (A-M Systems, Carlsborg, WA). The outputs of the amplifiers were directed to a data acquisition board (Data Translation, DT-2821) installed in a Pentium III processor-based PC. The EEG, EMG, and brain temperature recordings were digitally displayed and stored continuously to disk using Pass Plus software (Delta Software, St. Louis, MO).
IL-1 (rat recombinant IL-1 expressed in Escherichia coli) was purchased from R&D Systems (Minneapolis, MN). Lyophilized IL-1 was dissolved in PFS containing 0.1% BSA. Aliquots were kept frozen at −70°C and were used within a 6-mo period. Just before injection, an aliquot was thawed and brought to the appropriate volume with a stock PFS (0.1% BSA) solution. Controls were injected with vehicle (PFS containing 0.1% BSA).
After habituation to handling and recording procedures, rats were injected with either IL-1 (10 ng in 3 μl PFS/0.1% BSA icv) or vehicle (3 μl PFS/0.1% BSA icv) using a stainless steel needle (OD 0.3 mm) connected via polyethylene tubing to a Hamilton microsyringe. Injections were made over a 1-min period, after which the injection cannula was left in place for an additional 30 s before removal. All injections were given 15 min before dark onset. Rats were then connected to the recording cable so that EEG, EMG, and brain temperature could be recorded.
Rats were divided into three experimental groups, with each group including both IL-1 and vehicle-treated rats. Recordings were made in two groups of rats that were left undisturbed for either 2 or 5 h after injection. A third group of rats was sleep deprived by gentle handling for the 2-h period immediately after injections. During sleep deprivation, if the EEG of one rat showed high-amplitude activity indicative of ensuing sleep, both rats were handled in the same way, by stroking their fur with a cotton tip applicator. At the end of the recording period, rats were removed from the recording chamber and immediately given a lethal dose of pentobarbital sodium (100 mg/kg), which was followed by cardiac perfusion.
Sleep and temperature analysis.
Sleep-wake states of the rats were determined by an experienced scorer on the basis of the predominant state within each 10-s epoch. The scorer was blind to experimental conditions and to all cell count data. Wakefulness was defined by low-voltage, high-frequency activity combined with elevated neck muscle tone. NREM sleep was defined by high-amplitude EEG with prominent activity in the 2- to 4-Hz range. Rapid eye movement (REM) sleep was defined by moderate-amplitude EEG with dominant theta frequency activity (6–8 Hz) combined with minimal neck EMG tonus, except for occasional brief twitches. The percentage of time spent in each state was calculated per hour and for the total recording period.
Thermistors were calibrated, by water immersion, against a precision thermometer (YSI 4600, Dayton, OH), to an accuracy of 0.01°C. Brain temperature values were extracted from the Pass Plus system for each 10-s epoch, and averaged across recording time. Thermistors malfunctioned in four animals: one control rat in the 2-h postinjection group; one control and one experimental rat in the 5-h postinjection group; one experimental rat in the 2-h SD postinjection group. Temperature data are therefore missing from these animals.
Animals were perfused transcardially with 0.1 M PBS, followed by 300 ml of fixative containing 3% paraformaldehyde and 15% picric acid, in 0.1 M PBS, followed by 10% and then 30% sucrose. The brains were then stored in 30% sucrose at 4°C until they sank. Coronal sections were cut at 40 μm on a freezing microtome. Sections were processed for Fos immunohistochemistry as previously described (9, 10). Briefly, sections were incubated with a rabbit anti-c-Fos primary antiserum (AB-5, Oncogene Science, Cambridge, MA; 1:10,000) for 24 h. Sections were subsequently incubated with a biotinylated goat anti-rabbit IgG (Vector Laboratories, Burlingame, CA; 1:800) for 2 h and then reacted with avidin-biotin complex (Vector Elite Kit, 1:400) and developed with nickel-diaminobenzidine tetrahydrochloride, which produced a black reaction product in cell nuclei. There was no nuclear staining in the absence of primary antiserum.
The Neurolucida computer-aided plotting system (Microbrightfield, Williston, VT) was used to identify and quantify neurons that were labeled for c-Fos immunoreactivity. Cell counts were made by an individual who was blind to the experimental conditions of the animals. Section outlines were drawn under ×20 magnification. Fos-IR neurons were mapped in the section outlines under ×400 magnification. All cell counts were calculated for constant rectangular grids corresponding to three areas of interest: 1) the rostral MnPN (rMnPN) grid was a 600 μm × 600 μm square, centered on the apex of the third ventricle, rostral to the decussation of the anterior commissure and to bregma (∼AP: 0.1 mm) (10); 2) the caudal MnPN (cMnPN) grid was placed immediately dorsal to the third ventricle at the level of the decussation of the anterior commissure, extending 150 μm laterally and 600 μm dorsally just caudal to bregma (∼AP: −0.26 mm) (10); 3) The VLPO counting grid was placed at the level 160 μm or more caudal to the organum vasculosum of the lamina terminalis (∼AP: from −0.3 mm to −0.7 mm relative to bregma) and was subdivided into core and extended sections (17). The VLPO core box was 300 μm wide by 300 μm high, placed along the base of the brain, with its far border 400 μm lateral to the lateral edge of the optic chiasm (Fig. 1). The medial extended VLPO box was medial to the VLPO core, 400 μm wide by 300 μm high. The dorsal extended VLPO box was 200 μm wide by 300 μm high, positioned above the VLPO core and medial extended VLPO boxes and centered over their border (Fig. 1).
For both the rostral and caudal MnPN, cell counts were made in three sections and averaged to yield a single value for each rat. For VLPO, cell counts were made bilaterally in three sections containing the largest part of the VLPO. Those six counts were then averaged to yield a single value for both the VLPO core and extended VLPO (medial and dorsal sections combined) for each rat. Because Fos IR is high within the ependymal cells lining the ventricular walls following an intracerebroventricular injection of IL-1, and to a lesser extent, saline (4, 33), we were careful to exclude c-Fos labeling of these cells.
Two rats were excluded from all analyses due to incomplete sleep-wake data. In the final analysis, the 2-h postinjection group comprised five experimental and five control animals; the 5-h postinjection group contained seven experimental and six control rats; the 2-h SD postinjection group contained six experimental and five control animals. For rats left undisturbed after injection, we analyzed average sleep-wake states over the total recording period (2 or 5 h) and, for the 5-h postinjection group, also over the last 2-h period before the animals were perfused (4- to 5-h time block). Independent t-tests were used to assess significant differences in percentage of NREM and REM sleep in the first hour postinjection, as well as in the 2-h period preceding death, between IL-1 and vehicle-treated rats in each group of sleeping rats. For the 5-h postinjection group, average percentage NREM and REM sleep, within the total 5-h recording period, was also compared between IL-1 and vehicle-treated rats. Independent t-tests were also used to identify significant differences in brain temperature between IL-1 and vehicle-treated rats in the 5-h postinjection group. A one-way ANOVA was used to assess significant differences in brain temperature between all 2-h postinjection groups of rats that were either treated with IL-1 or vehicle followed by SD or no further intervention. The Student-Newman-Keuls (SNK) post hoc test was used to investigate the source of any significant differences. To assess the influence of IL-1 on Fos IR in the MnPN and VLPO, in all groups of rats, one-way ANOVAs were used, followed by the SNK post hoc test where appropriate. A significance level of P < 0.05 was set for all statistical comparisons. All results are reported as means ± SE. Data from vehicle and IL-1-treated rats that were killed 5 h after injection were combined to investigate correlations between cell counts and percentage of NREM sleep within the last 2 h before death, using the Pearson correlation coefficient.
Sleep and Brain Temperature
Two hours postinjection.
Sleep-wake states and brain temperature were assessed in rats that were left undisturbed for 2 h after an injection of either IL-1 or vehicle. The percentage of recording time spent in NREM sleep was similar during each hour (Table 1) and there was no significant difference between IL-1-treated rats [t(8) = 0.5, P = 0.8] and control rats for this 2-h time block (30.4 ± 3.4% vs. 28.5 ± 2.3%). However, REM sleep was significantly suppressed over 2 h [t(1, 8) = 3.7, P = 0.006] in IL-1-treated rats compared with vehicle-treated rats (0.1 ± 0.1% vs. 5.0 ± 1.3%). Wakefulness did not differ significantly [t(8) = 0.3, P = 0.8] between the IL-1 and vehicle-treated rats (69.6 ± 3.4% vs. 68.2 ± 3.0%). As shown in Fig. 2, average brain temperature was significantly elevated [ANOVA F(3,15)=17.7, P < 0.0001; SNK P = 0.0008] in IL-1-treated rats compared with controls (38.60 ± 0.1°C vs. 38.07 ± 0.01°C). Brain temperatures also differed between undisturbed and SD rats.
Five hours postinjection.
Table 1 reports the distribution of NREM sleep and REM sleep across 5 h after dark onset in rats injected with either IL-1 or vehicle. During the first hour postinjection, IL-1-treated rats had significantly more NREM sleep [t(11) = 2.2, P = 0.01]. Also, during the postinjection 4- to 5-h time block, IL-1-treated rats spent significantly more time in NREM sleep than did control rats [46.2 ± 4% vs. 30.7 ± 5%; t(11)=2.5, P = 0.03]. There was no significant difference [t(11) = 1.5, P = 0.16] in NREM sleep between IL-1 and vehicle-treated rats across the entire 5-h recording period (38.6 ± 4% vs. 30.9 ± 3%). REM sleep was significantly suppressed [t(11) = 3.7, P = 0.003] across the entire 5-h recording period (0.5 ± 0.3% vs. 5.0 ± 1.2%) and during the postinjection 4- to 5-h time block [t(11) = 2.3, P = 0.04] in rats injected with IL-1 compared with controls (0.5 ± 0.3% vs. 3.7 ± 1.5%). Wakefulness across 5 h did not significantly differ [t(11) = 0.7, ns] between the IL-1 and vehicle-treated rats (60.4 ± 3.7% vs. 64 ± 3.3%). Wakefulness during the 4- to 5-h time block tended to be less [t(11) = 1.7, P = 0.1] in the IL-1 compared with the vehicle-treated rats (53.3 ± 3.7% vs. 64.2 ± 5.5%).
As shown in Fig. 3, brain temperature increased within the first 2 h after an IL-1 injection and remained elevated across the 5-h recording period, compared with controls. Brain temperature averaged over 5 h postinjection was significantly elevated in IL-1-treated rats compared with controls [38.62 ± 0.2°C vs. 37.95 ± 0.1°C; t(9) = 3.19, P = 0.01]. The maximum febrile response to IL-1 occurred in the last hour before death, when average brain temperature was approximately 1°C higher in IL-treated rats than in controls (38.89 ± 0.3°C vs. 37.75 ± 0.25°C; Fig. 3).
Two hours of sleep deprivation postinjection.
A third group of rats was sleep deprived for 2 h after either an injection of IL-1 or vehicle. All rats spent less than 4% of recording time in NREM sleep during this period. No REM sleep was observed in any rat during this time. Brain temperatures in SD IL-1-treated rats tended to be higher [ANOVA F(3,15) = 17.7, P < 0.0001; SNK P = 0.06] compared with SD vehicle-treated rats (38.92 ± 0.04°C vs. 38.69 ± 0.1°C). SD IL-1-treated rats had significantly higher brain temperatures than did undisturbed IL-1-treated rats (SNK P = 0.02), and both groups of SD rats had significantly higher brain temperatures compared with undisturbed vehicle-treated rats (SNK P < 0.001) (Fig. 2).
Figure 4 shows representative photomicrographs from the rMnPN of IL-1 and vehicle-treated rats. Fos protein immunoreactivity is stained black and confined to the nucleus. Average cell count data for all experimental conditions are shown in Fig. 5A. ANOVA showed a significant overall group effect for the number of Fos-IR neurons in the rMnPN [F(5,28) = 16.4, P < 0.0001]. IL-1-treated rats killed 2 h postinjection had significantly more (SNK P = 0.0002) Fos IR neurons in the rMnPN compared with controls. The number of Fos IR neurons in the rMnPN was also significantly greater (SNK P = 0.0002) 5 h postinjection in IL-1-treated rats compared with controls. IL-1-treated rats that were left undisturbed for either 2 or 5 h had similar c-Fos counts (SNK P = 0.3). The number of Fos IR neurons in IL-1-treated rats that had been SD for 2 h postinjection was not significantly different (SNK P = 0.3) from that of the SD controls. c-Fos counts in the IL-1-treated SD rats were significantly lower (SNK P < 0.001) compared with the IL-1-treated rats that were left undisturbed for either 2 or 5 h (Fig. 5A). All vehicle-treated rats had similar numbers of Fos-IR neurons (P > 0.1).
For rats killed 5 h postinjection, data from IL-1 and vehicle-treated rats were combined for correlational analysis. Fos IR in the rMnPN was significantly and positively correlated (r = 0.68, P = 0.01) with the percentage of NREM sleep in the 2-h period before death for IL-1 and vehicle-treated rats (Fig. 6).
Average cell count data for the cMnPN, for all three experimental conditions, are shown in Fig. 5B. ANOVA showed a significant group effect [F(5,28) = 2.5, P = 0.05], but none of the post hoc comparisons was significant. However, rats injected with IL-1 and killed 5 h later tended to have more c-Fos (SNK, P = 0.1) than did vehicle-treated rats. There was no significant correlation between cell counts and percentage NREM sleep in the preceding 2 h (r = 0.43, P = 0.14). There was no significant difference in Fos-IR between SD IL-1 and vehicle-treated rats (SNK P = 0.6).
Average c-Fos counts in the VLPO core and extended VLPO, for rats in the three experimental conditions, are shown in Fig. 7. ANOVA revealed significant group effects for both VLPO core [ANOVA F(5,28) = 3.0, P = 0.03] and extended VLPO [ANOVA F(5,28) = 3.8, P = 0.009]. For the VLPO core, post hoc analysis showed that SD IL-1-treated rats had significantly more c-Fos compared with SD controls (SNK, P = 0.04). In the undisturbed 2-h and 5-h postinjection groups, however, there were no significant differences in the numbers of Fos-IR neurons in the VLPO core of IL-1-treated rats compared with controls (SNK P > 0.1). In the extended VLPO, IL-1-treated rats had significantly more Fos-IR neurons compared with controlsin the 2-h postinjection group (SNK P = 0.02). They also had significantly more Fos-IR neurons compared with IL-1-treated rats in the 5-h postinjection group (SNK P = 0.02). However, in the 5-h postinjection group, the number of c-Fos-IR neurons in the extended VLPO of IL-1-treated rats was not significantly different from that of controls (SNK, P = 0.3). Although the SD IL-1-treated rats appeared to have a greater number of c-Fos-IR neurons in the extended VLPO compared with controls (Fig. 7B), there was no significant difference (P = 0.19), probably due to the small sample size. There also were no significant correlations between cell counts in either the extended VLPO (r = −0.2, P = 0.5) or VLPO core (r = 0.2, P = 0.6) and percentage NREM sleep in the 2 h before death, for the 5-h postinjection group of rats.
We have found that IL-1, injected intracerebroventricularly into rats at dark onset, increases NREM sleep, suppresses REM sleep, and induces fever. These results are in agreement with previous reports and indicate that rat recombinant IL-1 alters sleep-wake behavior and brain temperature of rats in a manner generally similar to human recombinant IL-1 used previously (26, 29). These increases in NREM sleep and brain temperature are associated with increases in Fos-IR in the MnPN, an area containing sleep-related cells (10, 38). Our findings support the hypothesis that the MnPN may play a role in mediating increases in NREM sleep after IL-1 treatment.
The c-Fos protein is a useful and widely used marker of neuronal activation (23). However, we cannot state on the basis of this study whether IL-1-induced c-Fos IR in the MnPN is a direct effect of IL-1 or is a consequence of physiological changes, such as increases in NREM sleep or temperature. As an initial attempt to address this issue, we included in our study a group of rats that were kept awake after IL-1 injection. The number of c-Fos-IR cells in the rMnPN was increased 2 h after an IL-1 injection when rats were allowed to sleep spontaneously. In contrast, the number of c-Fos-IR cells in the rMnPN of rats injected with IL-1 and kept awake for 2 h was not significantly different from that of SD vehicle-treated controls and was significantly lower than that of IL-1-treated rats allowed to sleep for 2 h postinjection. These results suggest that neurons in this region, and under the conditions of this study, are active during sleep, and not in response to IL-1 administered into the ventricular system. However, the number of c-Fos-IR cells in the rMnPN was significantly greater 2 h postinjection in rats allowed to sleep after receiving IL-1 compared with vehicle, even though the amount of time spent in NREM sleep was identical. This observation suggests that neurons in the rMnPN of rats are activated in the presence of IL-1, but only in freely behaving animals allowed to sleep. Expression of c-Fos in the MnPN may depend on conditions that permit the occurrence of some threshold amount of sleep and may require the withdrawal of inhibitory inputs from wake-promoting signals mediated by noradrenergic inputs to the MnPN (2). Once the conditions for sleep are present, the expression of c-Fos may then represent the degree of sleep drive encoded by sleep-promoting cells in the MnPN, which could be increased following IL-1 treatment.
In addition to what may be subtle effects of IL-1 on NREM sleep-related c-Fos activation within the first 2 h postinjection, there are substantial effects of this manipulation 5 h after injection. Rats injected with IL-1 spend, on average, 52% more time in NREM sleep 4–5 h postinjection compared with controls, and the number of c-Fos-IR cells in the MnPN is increased. Indeed, the amount of NREM sleep of rats in the 2 h before death correlates with the number of Fos-IR neurons in the rMnPN. A previous study from our group (10) demonstrates a positive correlation between the number of c-Fos-IR cells in the MnPN and total sleep time, which includes NREM and REM sleep. Because REM sleep is suppressed by IL-1 (this study; 8, 11, 26, 29), our results suggest the intriguing possibility that sleep-associated Fos IR in the rMnPN may be most strongly related to NREM sleep.
Data in this study also suggest that IL-1-induced activation of the MnPN precedes by 2 h an increase in NREM sleep; the number of Fos IR cells increases within the first 2 h postinjection, but a robust increase in NREM sleep occurs only 4–5 h after injection. Most studies demonstrate IL-1-induced increases in NREM sleep in the first hour postinjection [e.g., (8, 11, 26)], and, indeed, our rats injected with IL-1 and allowed to freely behave for 5 h spent more time in NREM sleep during the first postinjection hour than did control animals. The lack of increase in NREM sleep during postinjection hour 1 of rats allowed only 2 h of spontaneous behavior before death is due to the increased amount of NREM sleep of the corresponding control rats. Nevertheless, although there were no differences in this study in the amount of time spent in NREM sleep during the first 2 h postinjection, there were more Fos-IR neurons in the rMnPN of animals injected with IL-1 than in animals injected with vehicle. Interestingly, the number of Fos-IR neurons in the rMnPN (∼50) of IL-1-treated rats allowed to sleep after injection at dark onset, in our study, is similar to that previously observed in rats spending large amounts of time (∼70%) in spontaneous NREM sleep during the 2 h preceding death in the light period (10). Although expression of c-Fos in the rMnPN normally shows diurnal variation, being higher during the light period (32), MnPN neurons may be activated to a similar extent when near-maximal NREM sleep is attained for a particular circadian time point, whether during the light or dark phase. An alternative possibility is that IL-1 activates both sleep-related and non-sleep-related MnPN neurons, such that Fos-IR cell counts are higher at a given amount of sleep after IL-1 administration.
We cannot state with certainty from this present study that Fos-IR neurons in the MnPN activated by IL-1 are sleep-regulatory neurons, although several lines of evidence suggest they may be (9, 10, 20, 38). Activation of neurons in the MnPN after IL-1 treatment might reflect other responses to IL-1. For example, increased Fos IR in the MnPN is associated with PGE2-induced fever (45), or heat stress (35), and 20–30% of neurons in the MnPN are warm-sensitive (43). However, it is unlikely that elevated brain temperature after IL-1 is a major determinant of Fos IR in the rMnPN, at least under the conditions of this study. The progression of fever was unrestricted in IL-1-treated rats that were sleep deprived, yet the number of Fos IR neurons in the rMnPN was not significantly different from that of vehicle-treated controls that were left undisturbed and that had significantly lower brain temperatures. Also, SD vehicle-treated rats had raised brain temperatures, probably due to continuous wakefulness (7), but no difference in c-Fos counts in the rMnPN compared with undisturbed controls. It is still plausible, however, that elevated brain temperature may activate the rMnPN only if accompanied by some sleep; mild ambient warming is associated with increased Fos IR in the rMnPN in rats allowed to sleep but not in waking rats (10). Additional studies are needed to elucidate further the interactions between IL-1-induced alterations in sleep and the role of neurons of the MnPN activated by IL-1.
Like the MnPN, the VLPO also contains a large number of sleep-active neurons (39). The number of Fos-IR cells in the VLPO correlates with the amount of prior sleep (37) and, more precisely, the number of Fos-IR neurons in the VLPO core correlates with NREM sleep, whereas the number of Fos-IR neurons in the extended VLPO correlates with REM sleep (17). We found effects of IL-1 on Fos immunoreactivity in the VLPO only in the 2-h treatment groups. SD IL-1-treated rats had significantly more c-Fos in the VLPO core, whereas IL-1-treated rats that were undisturbed for 2 h postinjection had increased c-Fos in the extended VLPO, but not in the VLPO core, compared with controls. We did not find a significant increase in the number of Fos-IR neurons in the VLPO core in our 5-h postinjection group of IL-1-treated rats in association with their increased NREM sleep. Our data therefore indicate that IL-1 injected into the lateral ventricle at dark onset increases NREM sleep without activating VLPO neurons. Also, despite REM suppression in the 2-h, undisturbed IL-1-treated group, c-Fos immunoreactivity in the extended VLPO was increased compared with controls. Other sleep-promoting substances are associated with much greater increases in Fos IR in the VLPO; continuous infusion of prostaglandin D2 (34) or an adenosine A2a receptor agonist (36) beginning at dark onset increases NREM sleep and induces a fourfold increase in c-Fos expression in VLPO. These substances were infused into the subarachnoid space. Collectively, these results suggest the possibility that VLPO neurons, which are situated at the base of the brain, may respond to signals in the subarachnoid space, whereas MnPN neurons, which are proximal to the ventricles, may respond more to intracerebroventricular signals.
Specific mechanisms of action for IL-1-induced alterations in NREM sleep remain poorly understood. IL-1 injected intracerebroventricularly might act on the epithelial cells of the ventricles, in which IL-1 receptor mRNA is abundant (46), to induce the release of secondary mediators that activate neurons in the MnPN. IL-1 also induces nuclear translocation of the transcription factor NFκB, which leads to synthesis of more IL-1; IL-1 increases PGs, NO, and adenosine, all of which can influence sleep [reviewed (25)]. Further, the effects of IL-1 on NREM sleep are mediated in part by GHRH in the hypothalamus (31) and serotonin in the dorsal raphe nucleus (12, 19). We recently demonstrated that IL-1 perfused directly into the POA/basal forebrain suppresses the discharge rates of wake-active cells and increases discharge rates of a subpopulation of sleep-active cells (1). The IL-1 receptor antagonist attenuates some of these effects, suggesting specific, receptor-mediated actions on these neurons. Therefore, IL-1 may enhance NREM sleep, in part, by suppressing wake-active neurons in the POA/basal forebrain.
In conclusion, our findings indicate that increased NREM sleep in rats after an intracerebroventricular injection of IL-1 at dark onset is associated with increased Fos-IR, particularly in the rostral MnPN. Although these data suggest a role for the MnPN in mediating effects of IL-1 on NREM sleep, the contribution of other hypothalamic nuclei and brain regions cannot be ruled out. Additional studies are necessary to determine the extent to which nuclei in the hypothalamus, and other regions in the brain, mediate effects of IL-1 on sleep-wake behavior.
This research was supported by U.S. National Institutes of Health Grants MH-64843, MH-66323, MH-47480, and the U.S. Department of Veterans Affairs Medical Research Service.
We thank Feng Xu and Keng-Tee Chew for excellent technical assistance, and Natalia Suntsova for comments on the manuscript.
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
- Copyright © 2005 the American Physiological Society