Microglial depletion and repopulation differentially modulate sleep and inflammation in a mouse model of traumatic brain injury

Abstract

Traumatic brain injury (TBI) causes persistent sleep disturbances, leading to long-term neurological consequences and reduced quality of life. We hypothesized that microglial depletion via PLX5622 (PLX), a colony-stimulating factor 1 receptor (CSFR1R) inhibitor, would exacerbate sleep disturbances and alter inflammatory profiles after TBI, and that microglial repopulation would ameliorate these effects. Male mice received PLX or control diets (21 days) followed by a midline fluid percussion injury (mFPI) or sham surgery. Physiological parameters were recorded non-invasively to determine sleep for 7 days post-injury. Subsequently, PLX was withdrawn to allow microglial repopulation, and sleep was assessed during the 7-day repopulation period. In a subset of mice, repeated blood draws were taken to quantify sleep regulatory cytokine concentrations (interleukin [IL]-6, IL-1β, tumor necrosis factor [TNF]-α). TBI significantly reduced sleep in mice on a control diet during the light period (3, 5, and 7 days post-injury), but not the dark period. In PLX-treated mice, TBI did not alter sleep in the light period, however, sleep in the dark period was increased at 3 days post-injury. During the microglial repopulation period, PLX-treated TBI mice slept significantly more in the dark period compared to PLX sham mice and sleep was similar in control TBI vs PLX TBI mice. Analyses revealed that elimination of microglia did not alter baseline cytokine levels. IL-6 was elevated in PLX TBI mice at 1 and 7 days post-injury compared to TBI mice on control diet, while IL-1β and TNF-α remained unchanged. This study highlights the critical role of microglia in modulating post-TBI sleep and inflammation. Findings suggest differential effects of TBI on sleep depending on microglial depletion or repopulation status, with IL-6 serving as a marker of the inflammatory response in microglia-depleted conditions.

Highlights

• •Microglial depletion via PLX5622 alters post-TBI sleep and neuroinflammatory responses.
• •Microglial repopulation normalizes post-TBI sleep patterns in PLX-treated mice.
• •IL-6 is elevated post-TBI in PLX-treated mice, suggesting a key role in neuroinflammation.
• •Study reveals microglia's dual role in modulating post-TBI sleep and inflammation.

1. Introduction

Traumatic brain injury (TBI) affects an estimated 69 million people worldwide each year (Dewan et al., 2019). TBI causes lifelong disabilities, including cognitive impairments (Zamani et al., 2019; Ichkova et al., 2017; Keenan et al., 2018) and psychological disturbances (Acerini and Tasker, 2007; Ortiz et al., 2020; Emelifeonwu et al., 2019). In the United States alone, the economic cost of nonfatal TBI exceeds $40 billion annually (Miller et al., 2021). These significant impacts emphasize the urgent need to develop effective therapies to reduce the long-term consequences of brain injuries.

Sleep is essential for cognitive function, immune health, endocrine balance, emotional stability, and overall well-being. Crucially, it also plays a pivotal role in promoting brain health and recovery from illnesses, infections, and injuries. Following TBI, sleep disturbances are common and likely contribute to post-traumatic complications (Gardani et al., 2015; Griesbach et al., 2018; Wolfe et al., 2018). These sleep disturbances not only exacerbate neuroinflammation but also impair cognitive recovery and overall health (Mullington et al., 2010; Haack et al., 2007; Zawar et al., 2023). Given the heterogeneity of TBI, the characteristics of sleep-wake disturbances vary widely, making it critical to understand how these disturbances influence long-term recovery (Fulda and Schulz, 2001). Current treatments for post-TBI sleep disturbances are disorder-specific (Viola-Saltzman and Watson, 2012), but identifying the causal mechanisms could lead to more effective pharmacological and rehabilitative interventions.

Inflammatory triggers, such as TBI, activate microglia, the primary immune cells of the central nervous system (CNS). Microglia rapidly respond to injury or disease by upregulating inflammatory molecules, clearing cellular debris, and initiating restorative processes to protect CNS function and maintain homeostasis (Loane and Faden, 2010; Miwa et al., 1997; Nakajima and Kohsaka, 2004). However, prolonged microglial reactivity can lead to neurotoxicity and tissue damage, highlighting the dual role of inflammation as both protective and potentially harmful when unregulated (Loane and Faden, 2010; Myer et al., 2006; Cardona et al., 2006; Limatola and Ransohoff, 2014; Bilbo and Stevens, 2017; Catalin et al., 2013). Microglia contribute to the neuroimmune response through the release of cytokines, which influence immune–brain interactions, including sleep regulation. Pro-inflammatory cytokines like interleukin (IL)-6, IL-1β, and tumor necrosis factor (TNF)-α promote sleep (Krueger and Majde, 1995; Krueger, 2008), while anti-inflammatory cytokines, such as IL-4, IL-10, and IL-13, inhibit sleep (Imeri and Opp, 2009; Krueger et al., 2008). Furthermore, sleep disturbances, even in healthy individuals, trigger inflammation characterized by increased levels of IL-1β, IL-6, and TNF-α, a response likely mediated by microglial activation (Mullington et al., 2010; Haack et al., 2004, Haack et al., 2007; Dinges et al., 1994; Frey et al., 2007; Irwin et al., 2006, Irwin et al., 2010; Irwin, 2002; Lekander et al., 2013; Meier-Ewert et al., 2004; Simpson and Dinges, 2007). Thus, microglia are both sources and targets of sleep-altering cytokines and play a key role in the brain's response to disturbed sleep (Artiushin and Sehgal, 2020). Studies show that sleep restriction activates microglia, alters their morphology, and increases their density in sleep-regulating brain regions (Hsu et al., 2003; Bellesi et al., 2017; Hall et al., 2020; Deurveilher et al., 2021; Green et al., 2020). Additionally, suppressing microglial reactivity with compounds like minocycline has been shown to reduce sleep rebound after deprivation (Wisor et al., 2011). These findings highlight the critical role of microglia in the neuroimmune response to sleep disturbances and support their involvement in the interplay between sleep and inflammation.

Microglial depletion has been widely studied to understand their role in neuropathology (Barnett et al., 2021; Boland and Kokiko-Cochran, 2024). Methods include pharmacological approaches, such as colony-stimulating factor 1 receptor (CSF1R) inhibitors (e.g., PLX5622), and genetic techniques inducing diphtheria toxin receptor expression in microglia (Barnett et al., 2021). After CSF1R inhibitor removal, microglia rapidly repopulate, resuming maintenance functions similar to resident microglia (Barnett et al., 2021). Importantly, repopulated microglia in the context of disease models exhibit distinct transcriptomes from pathological pre-depletion microglia (Henry et al., 2020). These findings suggest that microglial repopulation could be a promising strategy for studying and treating neuropathology. Targeting microglia may also mitigate sleep disturbances after a TBI. Our previous work demonstrates that microglia are essential for modulating sleep following peripheral immune challenges (Rowe et al., 2022). Using PLX5622 to deplete microglia in mice, we observed transient increases in sleep during the depletion period and exaggerated sleep responses to lipopolysaccharide (LPS), a potent immune stimulant (Rowe et al., 2022). During the microglial repopulation phase, these exaggerated responses normalized (Rowe et al., 2022). These findings highlight the involvement of microglia in regulating acute and adaptive neuroimmune responses to inflammation and their influence on sleep regulation.

In this present study, we investigated how microglial depletion and repopulation impact post-traumatic sleep disturbances and inflammation following diffuse TBI. We hypothesized that microglial depletion using PLX5622 (PLX) would result in dysregulated sleep and altered inflammatory cytokine profiles, while subsequent microglial repopulation would restore normal sleep patterns. To test this, we subjected male mice to a midline fluid percussion injury (mFPI) or sham surgery, with or without prior microglial depletion via PLX. Physiological parameters to determine sleep were assessed over a 7-day post-injury period, followed by an additional 7-day period to examine recovery during microglial repopulation after PLX withdrawal. We also quantified sleep-regulatory cytokines, including IL-6, IL-1β, and TNF-α, through repeated longitudinal sampling to evaluate the peripheral inflammatory response. This study aimed to uncover the role of microglia in modulating post-TBI sleep disturbances and neuroinflammation and to provide insights into potential therapeutic strategies for mitigating TBI-related complications.

2. Materials and methods

2.1. Rigor

All studies were conducted in accordance with the guidelines established by the Institutional Animal Care and Use Committee (IACUC) at the University of Arizona and the National Institutes of Health (NIH) guidelines for the care and use of laboratory animals. Studies are reported following the Animal Research: Reporting In Vivo Experiments (ARRIVE) guidelines (Kilkenny et al., 2010). Mice were randomly assigned to treatment groups before the initiation of the study to ensure equal distribution of experimental conditions (e.g., diet, injury) across groups. Data collection stopped at pre-determined endpoints based on days post-injury for each animal. Determination of sleep-wake behavior based on physiological parameters, and quantification of cytokines were done by investigators blinded to experimental conditions.

2.2. Animals

Adult (13-15 week-old) male C57BL/6 J mice (Jackson Laboratories, Bar Harbor, ME) were used (total n = 39). Mice were group-housed in a 14-h light 10-h dark cycle (200 lux, cool white, fluorescent light) at a fixed temperature (23 °C ± 2 °C) with food (AIN-76 A rodent chow, Research Diets, Inc., New Brunswick, NJ) and water available ad libitum according to the Association for Assessment and Accreditation of Laboratory Animal Care International. For sleep studies, mice were individually housed and acclimated to non-invasive piezoelectric sleep cages (Signal Solutions, Lexington, KY, USA) for a minimum of 5 days prior to initiation of data collection. Midline fluid percussion injuries were administered between zeitgeber (ZT) 2 and 4. Post-injury sleep measurements began at ZT 4 for all experimental cohorts.

2.3. Plexxikon (PLX) administration

After a 7-day acclimation period following shipping, mice were placed on Control AIN-76A rodent chow (Research Diets, Inc., New Brunswick, NJ) or diet with PLX5622 (1200 mg/kg, Plexxikon; PLX) formulated in AIN-76A rodent chow for 21 days prior to experimental procedures. For sleep studies, mice remained on the assigned diet for 7 days post-injury, after which all mice in the study were placed on control diet for 7 days to allow for microglial repopulation (Fig. 1). For the subset of mice used for cytokine analysis, mice were randomly assigned to control or PLX diet for 21 days and remained on the assigned diet until the end of the 7-day post-injury period (Fig. 1). Access to food and water remained ad libitum throughout all studies.

Fig. 1.

Study Design. For the Sleep Study, male mice were acclimated to non-invasive piezoelectric sleep cages for 7 days then randomly assigned to a diet (control, PLX) for 21 days. Following microglial depletion, mice received a midline fluid percussion injury (mFPI), or control sham surgery and physiological parameters were measured for 7 days to determine post-traumatic sleep. At 7 days post-injury, all mice were placed on control diet to allow for microglial repopulation. Sleep was assessed for the duration of the repopulation period. For the Cytokine Study, male mice were acclimated to the vivarium and a baseline blood draw was taken at the initiation of the study. Mice were randomly assigned to a diet (control, PLX) for 21 days. After 21 days of diet administration a blood draw was taken (prior to TBI). Blood was collected at 1 and 7 days post-injury. Pro-inflammatory cytokine levels of interleukin (IL)-6, IL-1β, and tumor necrosis factor (TNF)-α were assessed.

2.4. Midline fluid percussion injury

Mice were anesthetized with 5% isoflurane in 100% oxygen for 5 min, after which they were secured in a stereotaxic frame with a continuous flow of 2.5% isoflurane in 100% oxygen via a nosecone. Ophthalmic ointment was applied, and the surgical site was cleaned with betadine and ethanol. A midline excision was made to expose the skull. A craniectomy (3 mm diameter) was trephined midway between bregma and lambda on the sagittal suture. An injury cap was prepared from a Luer-Loc needle hub and fixed over the craniectomy using cyanoacrylate gel and methyl-methacrylate (Hygenic Corp., Akron, OH). The hub was filled with saline and closed with a cap made from a modified syringe tip to prevent debris and air exposure.

Mice were re-anesthetized 24 h after surgery with 5% isoflurane in 100% oxygen for 3 min. The cap was removed from the hub, and the dura inspected to ensure it was intact. The hub was then filled with saline and attached to an extension tube connected to the fluid percussion injury (FPI) device (Custom Design and Fabrication, Virginia Commonwealth University, Richmond, VA). When a pedal withdrawal response was detected, the pendulum was released, and a fluid impulse was delivered to the intact dura, as we routinely publish (Rowe et al., 2014, Rowe et al., 2018a, Rowe et al., 2018b, Saber et al., 20211). The righting reflex time was recorded and used as an indicator of injury severity. Sham (control) animals received a craniectomy and were attached to the FPI device, but the pendulum was not released. Once the mouse had returned to a supine position, it was briefly re-anesthetized, and the injury site was cleaned and closed with sutures. Antibiotic ointment (bacitracin) was applied, and the animal was returned to a heated holding cage.

2.5. Piezoelectric sleep recording parameters

To determine sleep-wake behavior, physiological parameters were recorded using a non-invasive piezoelectric cage system (Signal Solutions, Lexington, KY, USA), as previously described (Rowe et al., 2018b; Harrison et al., 2015; Saber et al., 2019). Briefly, each cage has an open bottom that allows the mouse to be placed directly on a cage floor with bedding over a Polyvinylidene difluoride (PVDF) sensor. These sensors are coupled to an input differential amplifier to generate pressure signals. These pressure signals are then converted to voltages, the amplitude of which is proportional to the pressure signal. Regular breathing movements characterize sleep [3 Hz, regular amplitude signals (Donohue et al., 2008)], whereas signals from awake mice are of higher amplitude with irregular spiking associated with volitional movements. In this study, the piezoelectric signals were analyzed over 10-s epochs at a 2-s interval. Data collected from the sleep cages were binned at each hour using a rolling average of the percentage of recording time spent in sleep and total minutes slept (Ouellette and Donohue, 2022).

2.6. Submandibular blood collection

At predetermined time points, submandibular blood samples were obtained from mice (Fig. 1). Mice were briefly restrained manually, and a lancet was used to make a small puncture at the submandibular vein. Approximately 250 μL of blood was collected into BD Microtainer® MAP microtubes coated in ethylene diamine tetra acetic acid (EDTA) (Becton, Dickinson, and Company, Franklin Lakes, NJ). Hemostasis was ensured by applying gentle pressure to the puncture site for 30 s with sterile gauze. Mice were monitored for signs of distress or complications. Blood collected from mice was centrifuged at 2000×g for 10 min at 4 °C to separate plasma from cellular components. Plasma was carefully aspirated, and samples were stored at −80 °C until cytokine quantification. All steps were performed on ice to preserve sample integrity. To account for circadian fluctuations in cytokines, all blood collections occurred at the same time of day (ZT 3–4).

2.7. Cytokine quantification

Cytokine assays (Bio-Plex Pro™ Mouse Cytokine 3-Plex, Bio-Rad Laboratories, Hercules, CA) were performed according to manufacturer's instructions to quantify levels of IL-6, IL-1β, TNF-α in plasma. Briefly, samples were diluted with sample diluent at 1:4. Samples and supplied standards were transferred to a 96-well plate containing antibodies coupled to magnetic beads for each analyte and incubated on a shaker for 30 min at room temperature. The samples were washed three times with supplied wash buffer and then incubated with detection antibodies on a shaker for 30 min at room temperature. The samples were washed three times with supplied wash buffer and then incubated with Streptavidin-PE on a shaker for 10 min at room temperature. Following another three washes, beads were resuspended for 5 min and quantified on a Bio-Plex® 200 system (Bio-Rad Laboratories, Hercules, CA). All samples were run in duplicate.

2.8. Statistical analyses

To examine sleep percentages over time following injury, we performed a repeated measures two-way analysis of variance (ANOVA) with a Geisser-Greenhouse correction. Fixed effects included injury, time, and their interaction (injury × time). When a significant injury effect was detected, Sidak's multiple comparisons post hoc analysis was used to identify specific group differences. To investigate the effect of the PLX diet, TBI mice on PLX diet were compared to TBI mice on control diet using a two-way ANOVA with fixed effects for diet, time, and their interaction (diet × time). Total minutes of sleep and cytokine levels were analyzed using unpaired t-tests. For both percent sleep and total minutes slept, data were analyzed separately for the light and dark periods. Statistical analyses and data visualization were conducted using GraphPad Prism 9.4.1 (GraphPad Software, LLC).

Exclusion criteria were predetermined: mice that lost >20% of their body weight or had visible signs of pain or distress were euthanized and excluded from the study (n = 0). Predetermined inclusion criteria included a righting reflex time of >315 s for moderate TBI (mean = 431.8 ± 101.1 s). To ensure injury severity was equivalent in mice on control diet compared to mice on PLX diet, an unpaired t-test was used to compare righting reflex times. There was no significant difference in righting reflex times of TBI control mice compared to TBI PLX mice (t₉ = 0.5, p = 0.6287). A total of 4 mice were excluded for low righting reflex times. The group sizes for the sleep experiments used in statistical analyses were: control sham n = 5, control TBI n = 5, PLX sham n = 7, PLX TBI n = 6. The group sizes for the cytokine quantification used in statistical analyses were: control TBI n = 6, PLX TBI n = 6.

3. Results

3.1. TBI altered sleep in mice on control and PLX diets

At 3 days post-injury, TBI altered sleep in both control and PLX diet mice compared to sham mice (Fig. 2A). In mice on control diet, TBI significantly reduced percent sleep in the light period (injury: F_1,8 = 30.11, p = 0.0006; time: F_4.7,37.7 = 3.04, p = 0.023) and a significant time effect was observed (F_3.3,26.3 = 4.65, p = 0.008) in the dark period. This reduction in percent sleep was reflected in total minutes slept, with TBI significantly reducing sleep during the light period (t₈ = 5.00, p = 0.001) but not altering dark period sleep (t₈ = 1.22, p = 0.257). In PLX diet mice, TBI caused a significant time effect in the light period (F_5.2,57.6 = 3.22, p = 0.011) and significant injury (F_1,11 = 8.19, p = 0.016) and time effects (F_4.1,45.3 = 12.99, p < 0.0001) in the dark period. For total minutes slept, TBI did not alter light period sleep (t₁₁ = 1.11, p = 0.292) but increased sleep in the dark period (t₁₁ = 2.86, p = 0.016). Comparing control and PLX diet TBI groups, significant time effects were observed in both the light (F_3.7,33.7 = 2.73, p = 0.048) and dark periods (F_3.6,32.5 = 10.16, p < 0.0001), but no differences were found in total minutes slept (light: t₉ = 1.79, p = 0.107; dark: t₉ = 1.83, p = 0.101).

Fig. 2.

TBI altered sleep in mice on control and PLX diets. (A) At 3 days post-injury, TBI significantly decreased sleep during the light period in mice on control diet, and increased sleep in the dark period in the PLX diet group. (B) At 5 days post-injury, disrupted sleep patterns persisted in TBI mice on control diet in the light period. For the PLX diet group, a significant injury × time interaction was observed in the dark period. (C) At 7 days post-injury, TBI-induced sleep disturbances were sustained, with significant effects of injury in both the light and dark periods for mice on control diet, and significant injury × time interaction in the light and dark periods of mice on PLX diet. Hourly percent sleep and total minutes slept are presented as mean ± SEM. Black bars indicate the dark period (zeitgeber time (ZT) 14–23). Significant differences are denoted as follows: ∗ indicates a significant effect of injury, # indicates a significant injury × time interaction, (p < 0.05).

At 5 days post-injury, TBI disrupted sleep in both control and PLX diet mice compared to respective shams (Fig. 2B). In mice on control diet, TBI significantly reduced percent sleep in the light period (injury: F_1,8 = 18.88, p = 0.0025; time: F_4.1,32.5 = 3.4, p = 0.0194). In the dark period, a significant time effect was observed (F_2.5,20.0 = 7.65, p = 0.0021). These reductions were consistent with findings for total minutes slept, where TBI significantly reduced sleep during the light period (t₈ = 3.55, p = 0.0075) but had no effect on sleep in the dark period (t₈ = 0.47, p = 0.649). In PLX diet mice, TBI caused a significant time effect in the light period (F_4.7,51.2 = 6.23, p = 0.0002) and in the dark period (F_3.7,40.9 = 5.93, p = 0.0009), with a time × injury interaction in the dark period (F_9,99 = 2.03, p = 0.0436). However, TBI did not alter total minutes slept in the light period (t₁₁ = 1.38, p = 0.196) or dark period (t₁₁ = 1.13, p = 0.283). Comparing control and PLX diet TBI groups, significant time effects were noted in both the light (F_4.7,42.3 = 4.22, p = 0.0039) and dark periods (F_2.6,23.1 = 10.3, p = 0.0003). However, no differences were found in total minutes slept between groups in either the light (t₉ = 0.91, p = 0.385) or dark periods (t₉ = 1.30, p = 0.227: Fig. 2B).

At 7 days post-injury, TBI-induced sleep disturbances persisted (Fig. 2C). In control diet mice, TBI reduced sleep in both the light (F_1,8 = 9.73, p = 0.0143) and dark periods (F_2.8,22.1 = 3.43, p = 0.0376). For total minutes slept, TBI significantly reduced sleep during the light period (t₈ = 3.12, p = 0.014) but did not alter sleep in the dark period (t₈ = 0.85, p = 0.479). In PLX diet mice, TBI caused a significant time effect (F_4.7,51.6 = 4.85, p = 0.0013) and a time × injury interaction (F_13,143 = 2.04, p = 0.0215) in the light period. In the dark period, significant time effects (F_4.6,50.9 = 10.85, p < 0.0001) and time × injury interactions (F_9,99 = 2.95, p = 0.0037) were observed. Despite these findings, TBI did not alter total minutes slept in either the light (t₁₁ = 1.50, p = 0.163) or dark periods (t₁₁ = 1.40, p = 0.189; Fig. 2C). Comparing control and PLX diet TBI groups, significant time effects were found in both the light (F_4.4,39.4 = 4.71, p = 0.0026) and dark periods (F_3.0,27.2 = 9.77, p = 0.0001). For total minutes slept, PLX-TBI mice slept more during the light period (t₉ = 2.25, p = 0.050), but no differences were found in the dark period (t₉ = 1.17, p = 0.273; Fig. 2C).

3.2. Hourly percent sleep was altered in mice with repopulating microglia

Hourly percent sleep was investigated during the repopulation period following PLX withdrawal to allow microglial repopulation. After 3 days of microglial repopulation, TBI effects were observed in both control and PLX-treated mice (Fig. 3A). In mice on control diet, percent sleep showed no significant injury or time effects in the light period, but a significant time effect was observed in the dark period (F_3.2,25.2 = 7.757, p = 0.0007; Fig. 3A). Total minutes slept showed no TBI-induced differences in either the light (t₈ = 1.756, p = 0.1172) or dark period (t₈ = 1.167, p = 0.2767). In PLX-treated mice with repopulating microglia, the percent sleep analysis revealed significant time effects in both the light (F_5.9,65.6 = 7.363, p < 0.0001) and dark periods (F_3.2,34.7 = 10.22, p < 0.0001), along with a significant injury effect (F_1,11 = 5.178, p = 0.0439) and injury × time interaction in the dark period (F_9,99 = 2.329, p = 0.0201; Fig. 3A). For total minutes slept, TBI did not alter sleep in the light period (t₁₁ = 0.9184, p = 0.3781) but increased sleep in the dark period (t₁₁ = 2.276, p = 0.0439; Fig. 3A). Comparisons between TBI groups indicated significant time effects in the light (F_4.8,43.4 = 6.244, p = 0.0004) and dark periods (F_2.9,23.2 = 11.62, p = 0.0001; Fig. 3A), but no differences in total minutes slept in either the light (t₉ = 0.599, p = 0.5639) or dark period (t₉ = 1.668, p = 0.1297; Fig. 3A).

Fig. 3.

Sleep was altered in mice with repopulating microglia. PLX diet was withdrawn, and all mice were placed on control diet to allow for repopulation of microglia. (A–B) At 3- and 5 days of microglial repopulation, mice previously on PLX diet that were subjected to TBI slept more than respective shams. (C) At 7 days of repopulation (14-days post-injury), TBI-induced sleep disturbances were noted in the light period of mice on control diet. Hourly percent sleep and total minutes slept are presented as mean ± SEM. Black bars indicate the dark period (zeitgeber time (ZT) 14–23). Significant differences are denoted as follows: ∗ indicates a significant effect of injury, # indicates a significant injury × time interaction, (p < 0.05).

At 5 days of microglial repopulation, percent sleep analysis showed mice on control diet had significant time effects in the light (F_4.6,36.6 = 2.67, p = 0.0412) and dark periods (F_2.9,23.5 = 4.745, p = 0.0103; Fig. 3B). Total minutes slept remained unaffected by TBI in both the light (t₈ = 1.794, p = 0.1106) and dark periods (t₈ = 1.017, p = 0.3388). In PLX-treated mice with repopulating microglia, percent sleep showed significant time effects in both the light (F_5.0,54.7 = 3.895, p = 0.0044) and dark periods (F_3.5,38.0 = 7.003, p = 0.0004), and a significant time × injury interaction in the dark period (F_9,99 = 2.006, p = 0.0463; Fig. 3B). There were no TBI-induced changes to total minutes slept in the light (t₁₁ = 0.6494, p = 0.5294) or dark periods (t₁₁ = 2.046, p = 0.0654). Comparisons between TBI groups showed significant time effects in the light (F_5.1,46.3 = 3.676, p = 0.0065) and dark periods (F_3.5,31.8 = 8.046, p = 0.0002; Fig. 3B), but no differences in total minutes slept during either the light (t₉ = 0.2986, p = 0.7720) or dark periods (t₉ = 1.481, p = 0.1728; Fig. 3B).

At 7 days of repopulation (14 days post-injury), control diet mice exhibited a significant injury effect in the light period (F_1,8 = 2.67, p = 0.0174) and time effects in the dark period (F_2.7,21.9 = 5.793, p = 0.0054; Fig. 3C). Total minutes slept showed a significant reduction due to TBI in the light period (t₈ = 2.986, p = 0.0174) but no change in the dark period (t₈ = 0.7528, p = 0.4731; Fig. 3C). For percent sleep, PLX-treated mice with repopulating microglia had significant time effects in both the light (F_4.7,52.2 = 3.474, p = 0.0098) and dark periods (F_3.0,33.5 = 6.037, p = 0.0020; Fig. 3C). Total minutes slept showed no TBI-induced changes in either the light (t₁₁ = 0.045, p = 0.9649) or dark periods (t₁₁ = 0.1774, p = 0.8624). For percent sleep, comparisons between TBI groups revealed significant time effects in the light (F_4.9,43.9 = 5.056, p = 0.0010) and dark periods (F_2.4,22.0 = 7.692, p = 0.0018; Fig. 3C), but no differences in total minutes slept during either the light (t₉ = 1.558, p = 0.1536) or dark periods (t₉ = 0.1013, p = 0.9215; Fig. 3C).

3.3. IL-6 concentrations were elevated after TBI in mice on PLX diet

IL-6 levels were not altered at the initiation of the study (t₁₀ = 0.606, p = 0.5507), or after 21 days of administration of PLX or control diet (t₁₀ = 1.684, p = 0.1077). Following TBI, IL-6 levels were significantly elevated in PLX-treated mice compared to control mice at both 1 day (t₁₀ = 2.185, p = 0.0399; Fig. 4A) and 7 days post-injury (t₈ = 2.118, p = 0.0483; Fig. 4A). There were no differences in IL-1β levels at any time point (Study Initiation: t₁₀ = 0.5915, p = 0.5673; 21 Days on Diet: t₁₀ = 0.1768, p = 0.8632; 1 Day Post-Injury: t₁₀ = 0.5006, p = 0.6275; 7 Days Post-Injury: t₈ = 0.6187, p = 0.5533; Fig. 4B) or TNF-α levels (Study Initiation: t₁₀ = 0.6843, p = 0.5094; 21 Days on Diet: t₁₀ = 1.223, p = 0.2495; 1 Da y Post-Injury: t₁₀ = 0.5700, p = 0.5812; 7 Days Post-Injury: t₈ = 0.1440, p = 0.8891; Fig. 4C).

Fig. 4.

IL-6 was elevated after TBI in mice on PLX diet. Pro-inflammatory cytokine levels were evaluated at the initiation of the study, following 21 days of administration of control or PLX diet, and at 1- and 7 days post-injury. (A) IL-6 was elevated after TBI in mice on PLX diet compared to mice subjected to TBI on control diet. (B–C) IL-1β and TNF-α levels were similar among groups across all time points. Cytokine levels are presented as mean ± SEM. Significant differences are denoted as follows: ∗ indicates a significant effect, (p < 0.05).

4. Discussion

This study reveals the critical role of microglial dynamics in modulating post-traumatic sleep disturbances and neuroinflammation following a diffuse TBI. Our findings indicate that under the conditions of this study, microglial depletion via PLX5622 alters sleep patterns during the acute post-TBI period. Specifically, microglial depletion increases sleep in the dark period at 3 days post-injury, whereas sleep in the light period remains unaffected. During the microglial repopulation phase, sleep patterns in PLX-treated TBI mice normalize, matching those of control diet TBI mice. Additionally, TBI significantly elevates IL-6 concentrations in PLX-treated mice at 1 and 7 days post-injury, suggesting an amplified neuroinflammatory response in the absence of microglia. Interestingly, IL-1β and TNF-α did not differ between groups, suggesting a specific role for IL-6 in the post-TBI inflammatory profile. These results emphasize the dual role of microglia in both contributing to and resolving neuroinflammatory processes and their critical impact on sleep regulation during recovery from TBI.

Dr. James (Jim) Krueger's pioneering work has significantly advanced our understanding of the interplay between inflammation and sleep regulation (Krueger et al., 2001, Krueger et al., 2007, Krueger et al., 2011). His research established that sleep is modulated by pro-inflammatory cytokines such as IL-1β and TNF-α, which act as sleep regulatory substances (Krueger, 1993, Krueger, 2008, Krueger et al., 2007, Krueger et al., 2011, Krueger and Majde, 1995). Our study builds upon these foundational insights by examining how microglia, the primary immune cells of the CNS, influence post-traumatic sleep disturbances and neuroinflammation. Dr. Krueger's work emphasized that cytokine-driven sleep responses are essential for recovery from illness and injury. By depleting and repopulating microglia in a TBI model, we extend these principles to demonstrate that microglia play a critical role in shaping the inflammatory milieu that governs sleep patterns after brain injury. Specifically, our findings suggest that IL-6 may serve as a key inflammatory marker in the context of microglial depletion, contributing to sleep regulation following TBI.

Compared to our previous study using lipopolysaccharide (LPS) to induce a peripheral inflammatory challenge (Rowe et al., 2022), the current findings highlight the context-dependent role of microglia in sleep regulation. In our previously published LPS study, microglial depletion via PLX5622 exaggerated sleep responses, with increased sleep primarily during the first 4 days after the immune challenge. By contrast, in the current TBI study, we focused on later time points—3, 5, and 7 days post-injury—to examine the subacute phase of recovery. We found that microglial depletion in this TBI model increased sleep during the dark period, when nocturnal mice are typically active, while TBI in control mice reduced sleep during the light period. When comparing TBI mice on the PLX diet to those on the control diet, PLX-treated mice slept more in both the light and dark periods. Although these increases did not reach statistical significance, the combined light and dark period effect size at day 3 post-injury was 0.336, indicating a large and biologically relevant effect. Given this, we conclude that PLX does not produce a phase-dependent effect on sleep in the post-injury period. Together, these findings demonstrate the distinct effects of microglial depletion on TBI-induced sleep disturbances and emphasize the importance of therapeutic timing in modifying sleep responses.

Our previous work shows that TBI induces acute increases in sleep during the first 3 days post-injury, driven by the early inflammatory response (Saber et al., 2019; Rowe et al., 2013). In the current study, we focus on subacute time points to investigate how sleep is modulated during the later phases of recovery. We found that TBI disrupts sleep in control mice throughout the 7-day post-injury period. Notably, in control diet mice, TBI decreases sleep during the light period, when healthy mice typically sleep the most. This reduction suggests that TBI leads to sleep fragmentation and impairs sleep consolidation during the recovery period. These sustained TBI-induced sleep disturbances align with our other translational studies that show focal TBI disrupts sleep for up to two weeks post-injury in male mice (Thomasy and Opp, 2019).

We observed no significant differences in cytokine concentrations between PLX-treated and control mice either at the initiation of the study or after the 21-day PLX treatment period prior to brain injury. This indicates that microglial depletion does not alter baseline cytokine levels in the absence of an inflammatory challenge, demonstrating that cytokine homeostasis can be maintained without microglia under non-inflammatory conditions. Thus, the effects of PLX on cytokines are context-dependent and emerge only in response to an inflammatory challenge. TBI increases IL-6 levels in PLX-treated mice at both 1 and 7 days post-injury compared to TBI-control diet mice, aligning with our previously published findings (Giordano et al., 2023). The consistent elevation of IL-6 across studies highlights its role as a key marker of neuroinflammation following TBI, which is influenced by microglial reactivity. In contrast, we observed no changes in IL-1β or TNF-α, replicating our previous findings (Giordano et al., 2023). The absence of significant changes in IL-1β and TNF-α in the current study suggests differences in cytokine responses are selective and depend on the inflammatory stimulus. These results emphasize the critical role of microglia in shaping cytokine profiles during neuroinflammation and highlight the dynamic nature of their contributions, which are modulated by the timing and type of injury or inflammation.

IL-6 serves as a promising prognostic biomarker, and targeting IL-6 pathways therapeutically may mitigate sleep disturbances and neuroinflammation during TBI recovery. Clinical studies link exaggerated IL-6 levels to worse outcomes following TBI (Ooi et al., 2022; Kumar et al., 2015). In pediatric TBI patients, IL-6 levels reliably indicate TBI severity, with elevated levels predicting significant findings on computer tomography (CT) scans (Chiollaz et al., 2024). Uniquely, IL-6 also acts as a potential biomarker for sleep quality. Higher IL-6 levels correlate with lower-quality sleep, such as fragmented sleep, in healthy individuals (Hong et al., 2005). Recent findings reveal that short durations of slow-wave sleep are associated with elevated IL-6 levels and greater daytime sleepiness (Koreki et al., 2024). These findings suggest a potential feed-forward mechanism in which TBI elevates IL-6, and TBI-induced sleep disturbances may further amplify the IL-6 inflammatory response. Our research shows that in the absence of microglia, TBI leads to high IL-6 levels and exacerbates sleep disturbances, which could further increase IL-6 production. Therefore, eliminating microglia is not a viable therapeutic strategy. A more targeted approach reducing IL-6 production holds therapeutic potential and warrants further exploration. Additionally, we clarify that IL-6 appears to be a marker of neuroinflammation in PLX-treated TBI mice rather than a direct mediator of sleep changes. While IL-6 has been implicated in sleep regulation, its sustained elevation following microglial depletion suggests that its role in this context may be secondary to other inflammatory processes. Future studies assessing direct CNS IL-6 levels, rather than peripheral blood measurements, may provide greater insight into its mechanistic role in post-TBI sleep disturbances.

Circadian regulation of inflammation plays a crucial role in sleep disturbances after TBI. The immune system follows a rhythmic pattern of cytokine release, which influences sleep architecture (Curtis et al., 2014). To reduce the impact of these fluctuations, we administered brain injuries and collected all blood samples at the same time of day. However, we did not directly assess how microglial depletion affects circadian fluctuations in cytokine levels. Proinflammatory cytokine induction is significantly higher when mice receive a peripheral immune challenge at dark onset compared to light onset, including a greater increase in IL-6 (Curtis et al., 2014; Gibbs et al., 2012). Future studies should incorporate circadian or diurnal analyses of cytokine expression and sleep patterns to determine whether microglial depletion affects the rhythmicity of inflammatory responses and whether these changes underlie the observed sleep disturbances.

As with any study, there are limitations to this one. This study focused on time points within the first two weeks following TBI, a period often used to identify therapeutic targets for improving long-term outcomes (Rowe et al., 2018a; Lim et al., 2013; Apostol et al., 2022). However, chronic sleep disturbances are commonly observed in clinical populations, and recovery trajectories are highly heterogeneous. Future studies need to explore sleep patterns and recovery beyond this subacute phase to better understand the long-term impact of TBI on sleep. Additionally, the present study was conducted exclusively in male mice. While this approach allowed for consistency, our previous work and ongoing studies suggest that sex is a significant biological variable that influences sleep at baseline and post-TBI (Saber et al., 2019; Mannino et al., 2024). In the clinical setting, recovery differences between males and females have been attributed, in part, to the role of sex hormone signaling (Ma et al., 2019). Future research should address how sex and sex hormones shape post-traumatic sleep architecture and neuroinflammation following diffuse TBI.

While PLX5622 is a widely used tool for microglial depletion, we have shown that it eliminates a small subset of peripheral phagocytes (Giordano et al., 2023), somewhat limiting the full ability to isolate microglia-specific effects. This global elimination approach, though effective, may confound interpretations of central versus peripheral immune contributions to post-TBI outcomes. Peripheral infiltration of immune cells is also possible after a TBI, as there is disruption to the blood-brain barrier. We have previously shown that blood brain barrier permeability after experimental TBI influences microglia reactivity, likely through immune signaling from peripheral macrophages (Green et al., 2024). While prior studies suggest that microglia regulate IL-1β, IL-6, and TNF-α in the context of sleep disturbances, our findings show a selective elevation of IL-6 in PLX-treated TBI mice, with no significant changes in IL-1β or TNF-α. This specificity may reflect differences in cytokine responses depending on the presence or absence of microglia. One possibility is that compensatory mechanisms in the peripheral immune system may sustain IL-1β and TNF-α levels despite microglial depletion. Further research is needed to determine the precise role of microglia in shaping cytokine responses across different injury models and inflammatory conditions. Future studies should aim to utilize more refined techniques, such as targeted depletion or conditional knockout models and single-cell RNA sequencing, to differentiate central from peripheral sources of inflammation, and to delineate the precise role of microglia in TBI recovery and associated sleep disturbances.

In conclusion, this study highlights the critical importance of investigating sleep disturbances following TBI and the bidirectional relationship between sleep and inflammation. Sleep is a fundamental process for brain recovery, and disruptions in sleep architecture can exacerbate neuroinflammation, delay healing, and impair long-term functional outcomes. The findings of this study suggest that microglia, as key regulators of neuroinflammation, play a pivotal role in post-TBI sleep disturbances. Understanding how inflammation influences sleep after TBI may provide crucial insights into the mechanisms underlying chronic sleep disorders observed in clinical populations. Therapies aimed at modulating microglial reactivity, reducing pro-inflammatory cytokines like IL-6, or enhancing sleep consolidation could significantly improve recovery trajectories and quality of life for TBI patients.

CRediT authorship contribution statement

Katherine R. Giordano: Writing – original draft, Visualization, Methodology, Data curation. Tabitha R.F. Green: Writing – review & editing, Methodology, Data curation. Mark R. Opp: Writing – review & editing, Visualization, Supervision, Project administration, Methodology, Funding acquisition. Rachel K. Rowe: Writing – review & editing, Visualization, Supervision, Resources, Project administration, Methodology, Funding acquisition, Formal analysis, Data curation, Conceptualization.

5. Disclosure statement

The authors have no disclosures to report.

6. Funding

This work was supported in part by grants from the Department of Defense awards W81XWH-21-1-083 (RKR) and W81XWH-14-1-0384 (MO). KRG was supported by National Institute of Neurological Disorders and Stroke of the National Institutes of Health award number F31NS113408. RKR is supported, in part, by funds from the Department of Integrative Physiology at the University of Colorado Boulder.

Declaration of competing interest

The authors declare the following financial interests/personal relationships which may be considered as potential competing interests: Rachel Rowe reports financial support was provided by US Department of Defense. Katherine Giordano reports financial support was provided by National Institutes of Health. Mark Opp reports financial support was provided by US Department of Defense. If there are other authors, they declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Handling Editor: Christopher Colwell

Data availability

Data will be made available on request.