Behavioral Perturbation and Sleep in Healthy and Virus-Infected Inbred Mice

Abstract

Murine gammaherpesvirus (MuGHV) is a natural pathogen of wild rodents that has been studied extensively in terms of host immune responses to herpesviruses during acute infection, latency, and reactivation from latency. Although herpesvirus infections in people can be associated with fatigue and excessive sleepiness during both acute and latent infection, MuGHV has not been assessed extensively as a model for studying the behavioral consequences of chronic latent herpesvirus infections. To assess MuGHV infection as a model for evaluating fatigue and assessing potential mechanisms that underlie the exacerbation of fatigue during chronic viral disease, we evaluated sleep, temperature, and activity after exposure of healthy and latently MuGHV-infected mice to sleep fragmentation and social interaction. Neither treatment nor infection significantly affected temperature. However, at some time points, latently infected mice that underwent sleep fragmentation had less locomotor activity and more slow-wave sleep than did mice exposed to social interaction. In addition, delta-wave amplitude during slow-wave sleep was lower in infected mice exposed to sleep fragmentation compared with uninfected mice exposed to the same treatment. Both reduced locomotor activity and increased time asleep could indicate fatigue in infected mice after sleep fragmentation; reduced delta-wave amplitude during slow-wave sleep indicates a light plane of sleep from which subjects would be aroused easily. Identifying the mechanisms that underlie sleep responses of mice with chronic latent MuGHV infection may increase our understanding of fatigue during infections and eventually contribute to improving the quality of life for people with chronic viral infections.

Epstein–Barr virus (EBV) is a ubiquitous human γ-herpesvirus that causes acute disease, establishes life-long latency and is associated with the common syndrome of infectious mononucleosis. Murine gammaherpesvirus (MuGHV) is a natural pathogen of wild rodents that provides an experimental animal model for studying the pathophysiology of an EBV-like gammaherpesvirus.^10,27 A comparison of MuGHV strain 68 (MuGHV68) infection in bank voles (a natural host) and laboratory mice (Mus musculus) using in vivo luciferase imaging and classic virologic methods showed that the 2 host species have quantitative differences in the magnitude of the infection yet exhibit comparable patterns of viral replication and sites of viral latency.¹² These findings support using MuGHV68 and Mus musculus for the study of gammaherpesvirus pathogenesis. Although MuGHV has been studied extensively with regard to host immune responses to herpesviruses during acute infection, latency, and reactivation from latency,^10,27 it has not been assessed extensively as a model for the study of the behavioral consequences of chronic latent herpesvirus infections.

EBV infection in humans is often associated with fatigue and excessive sleepiness during both acute and latent phases of infection.^{1,16,19,38,43,44} Many people similarly experience fatigue in association with other chronic medical conditions that have either specific viral etiologies (for example, hepatitis C, HIV) or symptoms suggestive of viral infections (for example, chronic fatigue syndrome (CFS), fibromyalgia syndrome).^3,21,23 Acute viral infections, persistent viral infections, and reactivation of latent virus may trigger or exacerbate fatigue.^15,29,35 Furthermore, exacerbation of fatigue in response to various physiologic and psychologic challenges is prominent in many disease conditions associated with chronic fatigue. For example, exercise is reported to precipitate or exacerbate fatigue or malaise in persons with CFS,^2,26 and CFS both reduces activity and blunts the circadian rhythm of activity in patients.⁴² In addition, stressful life events are reported to precipitate CFS and to exacerbate fatigue in persons with CFS,^6,37and stress can be a factor in reactivation of latent EBV and other herpesviruses.¹⁵ Fatigue, like other so-called ‘sickness behaviors’ (for example, anorexia, anhedonia, reduced social interaction), has been linked causally to various cytokines that are facets of the immune response.⁷ Therefore, the host immune response to chronic infection or inflammation is likely to create a powerful and unrelenting stimulus for fatigue.

Human fatigue can be viewed as continued performance of essential activities (for example, employment, care-giving) with curtailment of nonessential, voluntary, or recreational activities.³² Similarly, fatigue in mice can be defined operationally as a reduction in voluntary activity (that is, wheel running) as compared with essential activity (that is, locomotion in the cage to obtain food and water).^28,30,33 Problems with sleep can either reflect or contribute to fatigue. For example, excessive sleep may reflect fatigue associated with other causes, whereas poor sleep or inadequate amounts of sleep may cause fatigue.

Mice inoculated with MuGHV develop changes in sleep and activity that may indicate fatigue. On days 7 through 11 after inoculation (that is, at times associated with the peak lytic infection and its resolution), infected mice show hypothermia, increased somnolence, and reduced running-wheel activity during the dark (active) phase of the diurnal cycle as well as reduced food intake and body weight.²⁸ These measures all return to normal during days 11 through 30 after inoculation, as the virus enters the latent phase.²⁸ After this time, these behavioral indicators of fatigue can again be elicited by challenging mice with LPS. In uninfected mice, LPS administration induces modest and relatively transient (that is, less than 18 h) hypothermia, reduced running-wheel activity, and alterations in sleep.²⁸ In contrast, mice with latent MuGHV infection develop prolonged hypothermia, hypoactivity, hypersomnolence, and fragmented sleep, all of which persist for as long as 5 d after LPS administration.²⁸ This duration is consistent with the time reported for viral reactivation from latency in MuGHV-infected mice treated with LPS.¹⁴ Therefore, as compared with uninfected mice, mice with latent infections appear to show a greater magnitude and duration of LPS-induced behavioral perturbations that may reflect fatigue, perhaps in association with virus reactivation.

To further assess MuGHV infection as a model system for evaluation of fatigue and assessment of potential mechanisms that underlie exacerbation of fatigue during chronic viral disease, we extended our previous work^28,40 by evaluating sleep, temperature, and activity after exposure of uninfected and latently infected mice to sleep fragmentation (SF) and social interaction (SI), both of which are likely to create behavioral stress.

Materials and Methods

Experimental animal procurement and care.

Male C57BL/6J mice were purchased from the Jackson Laboratory (Bar Harbor, ME) at 6 wk of age. Prior to experimental use, mice were housed in groups of 5 under a 12:12-h light:dark cycle at 25 ± 1 °C. Food and water were available ad libitum. Mice were housed individually in the same conditions after assignment to an experimental group. All mice were free of known infections with common rodent microbial and parasitic agents, as monitored by using monthly testing of sentinel mice housed in the same room. All procedures involving mice were approved by the Laboratory Animal Care and Use Committee of the Southern Illinois University School of Medicine (protocol number 168-07-007).

Virus inoculation.

Mice were randomly assigned to receive intranasal administration of either MuGHV or vehicle. Mice were inoculated under isoflurane anesthesia with 25 µL containing approximately 5000 pfu MuGHV, as described previously.^11,17 Control mice received the same volume of pyrogen-free saline. Mice were used experimentally no sooner than 42 d after inoculation, consistent with recovery from the acute lytic infection and establishment of latent infection.^9,28

Surgery.

At about 30 d after inoculation, mice were implanted with intraabdominal transmitters (Data Sciences International, St Paul, MN) to allow telemetric recording of locomotor activity and core body temperature. Transmitters were gas-sterilized prior to implantation. Mice were anesthetized by subcutaneous injection of a mixture of ketamine (50 mg/kg) and xylazine (50 mg/kg) and were supplemented with additional anesthetic during surgery when needed. All surgery was conducted by using standard aseptic techniques. The abdomen was shaved by using a no. 40 clipper blade, and the skin was disinfected with alternating scrubs of povidone–iodine and 70% ethanol. A midline incision of approximately 3 cm was made in the abdominal skin. The linea alba was identified and incised, exposing the abdominal cavity. The transmitter was implanted with the rounded end directed cranially. Sterile saline (1 mL) was added to the abdominal cavity to lubricate the transmitter and support hydration in the mouse. The abdominal muscles were closed in a simple continuous pattern with 4-0 absorbable suture material, and the skin was closed with no fewer than 3 simple interrupted sutures of the same material. Mice recovered from anesthesia in a cage placed under a heating lamp. Any remaining skin sutures were removed at 10 d after surgery.

During the same surgery, mice were surgically implanted with electrodes for monitoring EEG and EMG. EEG electrodes consisted of 4 insulated stainless steel wires (Plastics One, Roanoke, VA) that were positioned visually parallel to and under the skull in bilateral frontal (1 mm anterior to bregma and 2 mm to the left and right of midline) and parietotemporal (3 to 4 mm posterior to bregma and 2 mm to the left and right of midline) positions. All electrodes were inserted into a pedestal that was secured to the skull with dental acrylic. One of the electrodes was made continuous with cable shielding and served as a ground; this electrode was not used for data acquisition. Two of the other 3 electrodes were referenced against each other in the combination that provided the best visual differentiation of 3 vigilance states (wakefulness, slow-wave sleep [SWS] and rapid-eye-movement sleep [REMS]). EMG electrodes (Plastics One) were placed subcutaneously overlying nuchal muscles of the left and right sides of the body and were referenced against each other.

After surgery, mice were housed in individual cages in a sound-shielded chamber under a 12:12-h light:dark cycle at 25 ± 1 °C. Ibuprofen (1 mg/mL) was provided in drinking water from 1 d before through 7 d after surgery, to provide analgesia.

Experimental design.

A minimum of 2 wk was permitted for recovery from surgery and acclimation to the running wheel, and a minimum of 6 wk was provided for clearance of the acute lytic MuGHV infection and establishment of latent infection (Figure 1). After that period, mice were exposed weekly during the first 6 h of the light phase to a single 6-h exposure of either sleep fragmentation (SF) or to a social interaction (SI) in block-random design (Figure 1). SI involved placing the experimental male mouse into the occupied home cage of another male mouse. For induction of SF, mice were placed on a disk that rotates slowly in a random direction for 8 s of every 30-s interval as described in detail elsewhere.⁴¹ Briefly, the disk diameter was 15 in., and the wall height was 10 in. A perforated fitted cover was placed over this cylinder to prevent the mice from escaping. The disk rotated at a speed of 20 to 25 s per full rotation. An 8-s period of disk rotation was chosen because at the speed used, 8 s allowed the disk to rotate 180°, thus assuring that the mouse would contact the wall during each rotational period. Mice that are asleep when rotation occurs awaken when they contact the wall that bisects the cylinder. Sleep, body temperature, and locomotor activity were measured for 24 h before (baseline) and 18 h after the 6-h period of perturbation (postperturbation). Data were not collected during SF or SI. At the end of the final data collection period, mice were euthanized by exsanguination under isoflurane anesthesia.

Figure 1.

Diagram of experimental design.

Physiologic measures.

Locomotor activity and core body temperatures were monitored via the intraperitoneal transmitters, which emitted frequencies that were received by (for temperature) and located on (for activity) a receiver (RPC1, Data Sciences International) that was positioned under each cage. Collected signals were processed through an analog converter. Locomotor activity was detected as movement within the cage based on the transmitter location above the receiver and was recorded as number of events per 10 min. Temperature readings were collected once every 10 min. Data were summarized for every 2-h interval across the entire recording period and analyzed by using DQ3 software (Transoma Medical, St Paul, MN).

For collection of EEG and EMG data, mice were tethered to a 6-channel electrical swivel with lightweight cables (Plastics One) that permitted unrestricted movement. Mice were acclimated to the tether for at least 2 d before data collection began. Throughout all recording sessions, the mice could move freely in their cages and had continuous access to food and water. Sleep data were scored by assigning a specific vigilance state (SWS, REMS, or waking) to each 10-s epoch of the recording period by using a computer-assisted scoring method with custom software (Quality Software, Springfield, IL). EEG tracings initially were examined visually to set a threshold value for δ-wave amplitude (DWA) that was associated with SWS for each mouse. Thresholds also were set for EMG amplitudes associated with periods of movement and for θ-to-δ ratios associated with REMS. The computer algorithm used these thresholds to score the animal's vigilance states over the entire recording period. A mouse was considered to be in SWS whenever DWA exceeded the SWS threshold in the context of with a low-amplitude EMG signal. REMS was identified by low-amplitude EEG and EMG signals that occurred simultaneously with a high θ-to-δ ratio. At all other times, mice were considered to be awake. All computer-scored data were verified visually prior to data analysis. At the end of the recording period, mice were euthanized by exsanguination under isoflurane anesthesia.

Statistics.

Sleep data for individual mice were compiled in 6-h intervals across the entire recording period including the 24 h before (baseline) and 18 h after perturbation. For analysis of baseline data, a 2-factor repeated-measures ANOVA was used to determine the effect of infection status on the repeated measures of temperature, locomotor activity, and percentage time in SWS and REMS. All baseline variables showed significant effects of time but no between-subjects effects for infection and no significant interactions (Figure 2). Therefore, postperturbation repeated measures were compared by using differences from baseline data at each time point (thus, each mouse served as its own control).

Figure 2.

Baseline patterns of temperature, locomotor activity and sleep in MuGHV-infected and uninfected mice. The black bar along the x axis indicates the dark portion of the light:dark cycle. For temperature and activity, groups comprised 9 uninfected and 9 infected mice; for SWS and REMS, groups comprised 8 uninfected and 7 infected mice. There was a significant (P < 0.001)effect of time for all 4 measures, but no significant effect of infection status.

For analysis of the postperturbation behavioral data, a 3-factor repeated-measures ANOVA design with infection status and perturbation (SF or SI) as between-subjects factors was used to test the overall model for each of the repeated measures of temperature, locomotor activity, percentage time in SWS and REMS, SWS and REMS bout length, and DWA during SWS. Step-down 2-factor repeated-measures ANOVA then were used to determine the effect of infection status for each perturbation type (SF or SI). When significant effects were noted, one-way ANOVAs with Tukey follow-up were performed to identify specific significant differences between groups (infected–SI, uninfected–SI, infected–SF, uninfected–SF) at individual time points. Descriptive statistics are expressed throughout as mean ± SEM, and P values of 0.05 or less were considered to indicate statistically significant differences between groups. SPSS version 20 was used for all data analysis.

Results

Baseline measures in infected and uninfected mice (Figure 2).

Two-factor repeated-measures ANOVA revealed significant (P < 0.001 for all) effects of time on temperature, locomotor activity, and percentage time in SWS and REMS. However, as previously reported,^28,40 infection status had no significant effect on these measures.

Effects of exposure to SI and SF on core temperatures and locomotor activity (Figure 3).

Figure 3.

Temperature and locomotor activity after exposure to 6 h of social interaction (SI) or sleep fragmentation (SF) in MuGHV-infected and uninfected mice. Data are expressed as differences from values obtained for each mouse during the comparable baseline period. Data were not collected during the period of exposure to SI or SF (hours 0 through 6). The black bar along the x axis indicates the dark portion of the light:dark cycle. Groups comprised 9 uninfected and 9 infected mice. With regard to temperature, there were no significant effects of treatment or infection status and no significant interactions. For locomotor activity, there were no significant effect of infection status and no significant interactions, but there was a significant (P = 0.003) effect of treatment. Infected-SF mice had significantly (*; P < 0.05, Tukey) less activity than did mice of either infection status after exposure to SI at the 18-h time point.

Temperature showed no significant between-subjects effects for treatment or infection status and no significant interactions. Within-subjects analysis showed a significant (P < 0.001) effect of time, with no significant interactions.

For locomotor activity, between-subjects analysis showed a significant (P = 0.003) effect of treatment, with no significant effect of infection status and no significant interactions. Within-subjects analysis showed a significant effect of time (P < 0.001) and a significant time × treatment interaction (P = 0.008). A one-way ANOVA with Tukey follow-up comparing all 4 groups at each time point showed that infected–SF mice had significantly (P < 0.05) less activity at the 18-h time point than did mice of either infection status after exposure to SI.

Effects of exposure to SI and SF on SWS (Figure 4).

Figure 4.

Slow-wave sleep after exposure to 6 h of social interaction (SI) or sleep fragmentation (SF) in MuGHV-infected and uninfected mice. Data are expressed as differences from values obtained for each mouse during the comparable baseline period. Data were not collected during the period of exposure to SI or SF (hours 0 through 6). The black bar along the x axis indicates the dark portion of the light:dark cycle. Groups comprise 7 uninfected and 6 infected mice. Treatment significantly affected percentage time in SWS (P < 0.01) and SWS bout length (P = 0.002). Time significantly affected both percentage time in SWS (P = 0.008) and SWS bout length (P < 0.001), with no significant interactions. DWA during SWS was also significantly affected by time (P < 0.001), with a significant time × infection interaction (P = 0.002). In particular, 1) at the 18-h time point, infected-SF mice spent significantly more time in SWS than did infected or uninfected mice exposed to SI P < 0.05, Tukey); and 2) at the 24-h time point, infected-SF mice had significantly (*; P < 0.05, Tukey) lower DWA than did infected uninfected mice exposed to SF.

Repeated-measures ANOVA was used to evaluate the effects of treatment (SI or SF) and infection status on SWS parameters over time, with data analyzed as differences from baseline values for each mouse at each time point. In the 3-factor repeated-measures ANOVA, between-subjects analysis showed a significant effect of treatment on percentage time in SWS (P < 0.01) and SWS bout length (P = 0.002), with no significant effect of infection and no significant interactions and with no significant effects on DWA during SWS. Within-subjects analysis showed significant effects of time on both percentage time in SWS (P = 0.008) and SWS bout length (P < 0.001), with no significant interactions. In addition, DWA during SWS was significantly affected by time (P < 0.001), with a significant time × infection interaction (P = 0.002).

The step-down 2-factor repeated-measures ANOVA within each treatment group showed no significant effect of infection on percentage time in SWS, SWS bout length, or DWA during SWS for either the SF or SI groups. SI groups showed a significant within-subjects effect of time for both SWS bout length (P = 0.012) and DWA during SWS (P < 0.001), with no significant interactions. The SF groups showed significant effects of time for percentage time in SWS (P = 0.011), SWS bout length (P < 0.001), and DWA during SWS (P < 0.001), with a significant time × infection interaction for DWA during SWS (P = 0.004).

A one-way ANOVA with Tukey follow-up comparing all 4 groups (infected–SI, uninfected–SI, infected–SF, uninfected–SF) at each time point showed that infected–SF mice spent significantly (P < 0.05) more time in SWS at the 18-h time point than did infected or uninfected mice exposed to SI and had significantly (P < 0.05) lower DWA at the 24-h time point than did uninfected mice exposed to SF.

Effects of exposure to SI and SF on REMS (Figure 5).

Figure 5.

Rapid-eye-movement sleep after exposure to 6 h of social interaction (SI) or sleep fragmentation (SF) in MuGHV-infected and uninfected mice. Data are expressed as differences from values obtained for each mouse during the comparable baseline period. Data were not collected during exposure to SI or SF (hours 0 through 6). The black bar along the x axis indicates the dark portion of the light:dark cycle. Groups comprise 7 uninfected and 6 infected mice. No significant effects of time, treatment, or infection status were identified for percentage time in REMS or REMS bout length.

Repeated-measures ANOVA was used to evaluate the effects of treatment (SI or SF) and infection status on REMS parameters over time, with data analyzed as differences from baseline values for each mouse. Whereas 3-factor repeated-measures ANOVA showed a significant between-subjects effect of treatment on REMS bout length (P = 0.035) and within-subjects analysis showed a significant effect of time on percentage time in REMS (P = 0.041), no significant effects were detected in subsequent step-down analyses and follow-up tests.

Recovery of sleep debt (Table 1).

Table 1.

Recovery of SWS and REMS after SI or SF

	Uninfected		Infected		Uninfected		Infected
SWS time (min)	Base	SIa	Base	SIa	Base	SFa	Base	SFa
0–6	210 ± 9	—	215 ± 5	—	212 ± 10	—	215 ± 6	—
6–24	485 ± 17	516 ± 27	505 ± 30	518 ± 40	450 ± 27	532 ± 37	408 ± 12	563 ± 48
Total	695 ± 21	516 ± 27	720 ± 33	518 ± 40	662 ± 35	532 ± 37	623 ± 12	563 ± 48
Pb		<0.001		0.002		<0.001		0.306
% recoveryc		74% ± 3%		72% ± 5%		80% ± 2%		91% ± 9%
REMS time (min)
0–6	13 ± 2	—	16 ± 3	—	10 ± 2	—	16 ± 4	—
6–24	29 ± 7	35 ± 9	35 ± 5	47 ± 8	20 ± 4	36 ± 8	26 ± 6	47 ± 10
Total	42 ± 10	35 ± 9	51 ± 6	47 ± 8	30 ± 6	36 ± 8	42 ± 10	47 ± 10
Pb		0.077		0.430		0.286		0.036
% recoveryc		80% ± 10%		87% ± 13%		125% ± 23%		119% ± 9%

^aFor purposes of calculation of recovery of sleep debt, SWS and REMS were assumed to be 0 during treatment.

^bt test (infected total compared with uninfected total)

^cCalculated as total minutes in sleep on treatment day / total minutes on baseline day

To assess of recovery of lost sleep after SF or SI in infected and uninfected mice, we made the assumption that no sleep had occurred during the perturbations (we did not measure sleep during perturbations). The durations (in minutes) of SWS and REMS were calculated for baseline and postperturbation periods. In response to SI, both infected and uninfected mice recovered lost REMS but maintained a significant (P ≤ 0.002) SWS debt. In response to SF, uninfected mice recovered lost REMS but maintained a SWS deficit (P < 0.001), as occurred in our previous study.⁴¹ However, infected mice recovered lost SWS and showed significantly (P = 0.036) more REMS than occurred during the baseline period, suggesting a heightened homeostatic response to sleep loss in infected mice.

Discussion

As in our previous publications studying MuGHV infection in mice, latent infection did not significantly alter patterns of temperature, activity, or sleep under basal conditions.^28,41 In the current study, we also tested the effects of 2 perturbations, SI and SF, on subsequent patterns of core temperature, activity, and sleep. With regard to temperature, we found no significant effects of treatment (SF, SI, or baseline) or infection status. However, treatment had a significant effect with regard to locomotor activity: during the 12 to 18 h interval after treatment, infected mice exposed to SF showed less locomotor activity than did mice exposed to SI.

As might be expected, SF produced a greater effect on sleep than did SI, with significant effects of treatment on percentage time in SWS, SWS bout length, and DWA during SWS. In particular, during the interval between hours 12 and 18, infected–SF mice spent significantly more time in SWS than did infected or uninfected mice exposed to SI, and during hours 18 to 24, infected–SF mice had significantly lower DWA than did uninfected mice exposed to SF.

These data, taken together, indicate that latently infected mice exposed to SF respond with significant changes in locomotor activity, SWS time, and SWS bout length, as compared with mice in other experimental groups. In particular, reduced locomotor activity in infected mice after SF is consistent with increased time asleep, both of which behaviors could indicate fatigue. Reduced DWA during SWS is indicative of a light plane of sleep from which subjects would be aroused easily. Our previous study found that mice with latent MuGHV infection showed greater and more prolonged effects of LPS administration than did uninfected mice.²⁸ The robust response of mice with latent infections to subsequent challenges (that is, LPS or SF) suggests that they maintain an altered and more responsive physiologic steady-state that perhaps results from or enforces viral latency, as has been suggested by others.^4,14,25,45 Indeed, the heightened sensitivity of mice with latent MuGHV infection to LPS challenge is reminiscent of the profound effects of LPS that occur in mice whose immune responses are disrupted due to genetic deficiencies in IL6, IL10, or the p50 subunit of NFκB.^18,24,39 In contrast to LPS challenge, SI and SF have comparatively modest effects on the inflammatory steady state.⁴¹

The duration of exposure to a physiologic challenge and the interval between the challenge and the assessment of dependent markers could influence the development or detection of changes in those markers. Therefore, in mice with latent MuGHV infections, administration of LPS caused behavioral changes that persisted for at least 5 d.²⁸ MuGHV reactivation occurs 3 to 7 d after LPS administration.¹⁴ In the current study, mice were exposed to SI and SF for only 6 h and were monitored for only 18 h thereafter. Increased exposure to SI or SF or prolonged follow-up could reveal additional alterations in behavioral measures related to infection status.

DWA during SWS typically is reported to increase during recovery after short-term sleep loss in both people and animals and generally is considered to reflect the depth of sleep: if DWA is relatively high, arousal is less likely to occur spontaneously and requires a more intense stimulus to induce.^5,13 In the current study, exposure of mice to SI and SF occurred during the light phase, which is the normal diurnal period of somnolence for mice, and the mice therefore likely experienced sleep loss during that time.⁴⁰ However, although both infected and uninfected mice experienced SF, only the infected cohort responded with increases in DWA followed by a prolonged decrease, suggesting a greater physiologic response to the same challenge. These findings suggest that infected mice are more likely than are uninfected mice to respond with altered sleep depth after experiencing sleep loss.

Our assessment of recuperative sleep after sleep loss (Table 1) carries the important caveat that SWS and REMS were not measured during SI or SF, and therefore the actual amount of sleep loss experienced by these mice could not be evaluated. To perform the analysis of recovery of lost sleep after perturbation, we assumed that sleep did not occur during the 6-h period in which the perturbation was applied. According to our previous report⁴¹ and that of others using the same SF system,³⁶ we are confident that this assumption is reasonable with regard to SF, given that mice obtained little SWS and essentially no REMS during SF in those reports. In the current study, uninfected mice recovered lost REMS after exposure to SF but maintained a SWS deficit, as in our previous study.⁴¹ However, infected mice recovered lost SWS and showed significantly (P = 0.036) more REMS than they had exhibited during the baseline period, suggesting a heightened homeostatic response to sleep loss in infected mice. In response to SI, our analysis showed that both infected and uninfected mice recovered lost REMS but maintained a significant SWS debt. However, if our assumption was incorrect and the mice had experienced with relatively little sleep loss during SI, then mice exposed to SI would appear to maintain normal amounts of SWS time during the 24-h period that included SI yet show excess time in REMS. Although social contact and isolation are reported to differentially influence both ad libitum sleep and recovery from sleep loss in mice,^8,20 we are unaware of previous studies in which a novel social interaction was used as the stimulus for sleep disruption.

As mentioned earlier, EBV infection in humans is often associated with fatigue and excessive sleepiness during both acute and latent phases of infection.^{1,16,19,38,43,44} However, a polysomnographic study found no significant differences in objective measures of sleep in patients with CFS and controls,²² consistent with our failure of infected mice to show significant differences in baseline sleep patterns as compared with uninfected mice. However, compared with nonfatigued control subjects, CFS patients report significantly more problems with sleep and more accurately estimate their sleep latency,^22,31,34 suggesting that they are more aware of their sleep behavior than are healthy persons. Furthermore, stress-related exacerbations of fatigue are common in many disease conditions associated with chronic fatigue,^2,6,26,37 as perhaps is indicated by the changes in sleep and activity that occur in MuGHV-infected mice after SI or SF.

In summary, the current study has shown that 2 forms of behavioral perturbation, SI and SF, produce some similar and some differing effects on core temperature, activity, and sleep, with infection status also influencing responses to SF. The significant differences in some responses of infected and uninfected mice to SF is consistent with different basal immune or physiologic states in the 2 groups, as also was suggested by our previous study of sleep responses in infected mice challenged with LPS.²⁸ In contrast, infection status did not modify the effects of SI on sleep, suggesting that sleep perturbation interacts with infectious or immune status to further perturb sleep, whereas the same duration of social disruption has less effect on sleep, immune status, and their interactions. However, a prolonged or stressful SI intervention potentially could influence sleep measures. If virus-related immune perturbations underlie the immunologic, behavioral, and physiologic changes that occur in association with chronic viral infection and disease states such as CFS, then studying how various challenges affect mice with chronic latent MuGHV infection may further our understanding of how symptoms relate to the initial and chronic course of an infection. Such information eventually could support the development of strategies for improving the quality of life for people with chronic viral infections.