Chronic Pain Induced by Social Defeat Stress in Juvenile Mice Depends on TLR4

1. Introduction

Stress and pain are multifactorial constructs that are linked at several levels. Both activate similar pathways of the nervous system, such as the corticolimbic system, hippocampus, prefrontal cortex, and can be affected by genetic factors and psychosocial triggers (fears, emotional distress, beliefs, goals, threats) [1,2,3,4]. In acute conditions, stress may act as a protective response, inducing systemic cortisol release that provides energy to 'escape' from the stressor (a fight or flight state) [5]. However, stress becomes harmful when it occurs repeatedly for an extended period, leading to dysregulation in the neuroendocrine and immune systems [1]. Evidence shows that chronic stress is a major trigger for several mental disorders, such as anxiety and depression, and when it occurs in childhood or adolescent years, the individual becomes predisposed to the development of psychopathologies and pain sensitivity associated with increases in circulating inflammatory mediators released during stressful events [6]. In patients with depression, an increase of cytokines amounts in the peripheral blood was detected, which was also related to the participation of endogenous damage-associated molecular patterns (DAMPs) [7].

Preclinical models of physical and psychological stress have been used in rodents by applying unpleasant stressors such as repeated restraint, forced swims, sleep deprivation, maternal separation, and social defeat stress (SDS) [8]. The SDS model involves the resident-intruder paradigm in which a submissive mouse intruder is physically and sensorially exposed to an aggressor animal (resident) [9]. SDS was shown to promote behavioral alterations in adult mice, including anxiety, anhedonia, and social avoidance [10]. Interestingly, SDS also triggers widespread chronic pain phenotype, mimicking a condition similar to human fibromyalgia [10]. At present, there is an increasing interest in the study of pain phenotypes evoked by SDS in juvenile mice, given some clinical findings linking children and adolescent bullying with mental disorders and increased susceptibility to develop chronic pain in adulthood [11,12,13].

Our group has focused on studying the potential mechanisms and molecules for addressing chronic pain. In such an effort, toll-like receptor 4 (TLR4) has emerged as a key factor for neuronal sensitization [14,15,16]. These receptors are highly expressed in neural tissues, particularly in microglia and macrophages [17], and are demonstrated to participate in initiating and maintaining pain states, such as peripheral neuropathic pain [18] induced by chemotherapy and arthritic pain [19]. Despite this, there are many gaps to be filled regarding the mechanisms involved in the hyperexcitability of pain pathways as a result of chronic social stress, especially at early stages. Here, using social defeat stress (SDS), we investigated the chronic pain and anxiety-like behavior phenotype in juvenile male mice and the contribution of TLR4 activity in the dorsal root ganglia, spinal cord, and prefrontal cortex in the development of persistent pain caused by social stress.

2. Materials and Methods

2.1. Animals

Male C57Bl6/J mice at four to five weeks old were used as the experimental mice randomly divided into stressed (SDS) or no stressed (no-SDS) groups, and twelve-week-old male CD1 retired breeders as aggressors (purchased from Jackson laboratory^© or ENVIGO^©). They were maintained on isolated ventilated racks in a 12:12 hour light or dark cycle with controlled humidity and temperature (22°C). Food and water were available ad libitum. Tlr4^-/- mice (TLR4-KO) were a gift from Dr. S. Akira (Osaka University, Japan) and backcrossed onto the C57Bl/6 background for ten generations [20]. Animal handling and behavioral experiments were performed as per the International Association for the Study of Pain (IASP) guidelines for using animals in pain research. The study was conducted in accordance with the rules established by the Institutional Animal Care and Use Committees of the University of California, San Diego, USA. All animal procedures performed in this study were approved by the Committee (IACUC – #S00137M).

2.2. Social Defeat Stress (SDS)

Social defeat stress is an ethologically relevant and widely used stress model to mimic the symptomatology of human depression and chronic pain, based on a resident-intruder paradigm [9]. We first screened several CD1 mice to categorize aggressive behavior. For this, we used a separate group of C57BL6/J mice as screeners. The screening consisted of 3 consecutive daily sessions of 3 minutes of physical contact between a CD1 with a C57 who was placed in the aggressor cage. A new screener mouse was placed in the CD1 cage in each session. The latency of CD1's first attack and the number of assaults against the screener were monitored and recorded following previous protocol standards [21]. Intermediated CD1 aggressive behavior was selected to perform the SDS protocol.

SDS protocol consisted of 6 physical sessions of 10 minutes between a C57BL6/J experimental mouse and a CD1 (resident) previously selected. They were maintained in the same cage for ten days, separated by a clear perforated plexiglass divider, allowing only sensorial contact (Supplementary Figure 1A). The physical sessions were split every two days (protocol days 1,2,5,6,9 and 10). The control group had no physical contact with the CD1 aggressor or other mice, being paired for ten days with another naive C57 separated by a perforated plexiglass divider (Supplementary Figure 1D). For acclimation, the divider was placed two days before the first stress session in the aggressor and control cages. After the end of the SDS protocol, on day 11, all experimental mice (SDS and no-SDS) were single-housed for further behavioral assessment.

2.3. Mechanical and Thermal Threshold Evaluation

The mechanical threshold was measured using Von Frey filaments (Bioseb) using the up-down method [22]. Different gauges or stiffness filaments (range from 2.44 to 4.31; 0.02-2.00 g) were applied to the plantar surface of both hind paws, and the withdrawal response was registered. Before the initiation of mechanical stimulation, mice were placed in clear, plastic, bottomless cages over a wire mesh surface and acclimated for 1 hour. Mechanical values for the left and right hind paws were measured and averaged to produce a single data point per measurement day.

Thermal hyperalgesia was evaluated by a Hargreaves-type testing device (UARDG, Department of Anesthesiology, University of California, San Diego)[23]. The device consists of a box with a glass surface on which the mice rest in plexiglass cubicles for 1 hour. A bulb is positioned below the glass surface, and when activated, it projects a focus heat light at 36-43 ^oC. The thermal stimulus is applied to both hind paws, and latency is defined as the time required for the mouse to produce a brisk paw withdrawal as detected by photodiode motion sensors that stop the timer and terminate the thermal stimuli. Von Frey and Hargreaves tests were performed before the stress session (day 0) and on days 7, 11, and 21 of the SDS protocol (see schema in Figure 1A). A blind investigator conducted both behavioral tests from 9:00 a.m. to 5:00 p.m.

2.4. Open-Field and Light-and-Dark Tests

Mice were subjected to the Open-Field test to evaluate locomotor activity and anxiety-like behavior. The test consisted of recording the mouse freely moving in an arena (40 × 40 × 40 cm3; 800 lx) for 10 minutes [24]. Specialized software (Ethovision 3.1, Noldus Information Technology BV, Wageningen, The Netherlands) was used to assess the real-time activity of each animal (nose, center of the body, and tail base). Each mouse was tested individually in a separate arena; 4 arenas were used per experiment. The parameters measured were: (1) number of entries to the center, (2) time spent in the center, (3) number of entries to the corners, (4) time spent in the corners, (5) total distance traveled, (6) body mobility, (7) velocity, and (8) maximum acceleration.

The light-and-dark test is performed in an apparatus with two interconnected compartments equally sized (90 × 90 × 165 cm) [25]. One compartment is illuminated with 7000 lux white light, and the other is painted dark and covered with an opaque lid. Mice started in the dark compartment and stayed for 10 minutes in the arena. The video tracker system registers the latency for the first entry into the light side and the time spent. Both behavioral tests were conducted on days 10 and 17 of the protocol between 9:00 a.m. and 5:00 p.m. Before each test, animals were placed in the room for 60 min for acclimation.

2.5. Immunostaining

Animals were euthanized eleven days after the last session of stress, day 22 of the protocol. The spinal cord (lumbar region), dorsal root ganglia (L3, L4, and L5), and prefrontal cortex were harvested and used for Flow Cytometry assay or Immunofluorescence. Tissues were dissociated using the Neural Tissue Dissociation Kit (Miltenyi Biotec, San Diego, CA, #130.092.628), following the manufacturer protocol. A debris removal solution (Kit Miltenyi Biotec, San Diego, CA, #130.109.398) was applied to the spinal cord and brain samples to eliminate debris. Single-cell suspensions were incubated for 45 min on ice with the following antibodies: APC-Cy7 live/dead Ghost dye (#18452, Cell Signaling, Danvers, MA) or 540 live/dead Ghost dye (#72086, Cell Signaling, Danvers, MA); PC7 -conjugated TLR4/MD2 (#MTS510; ThermoFisher); APC-conjugated total TLR4 (#SA15-21; Biolegend), 488-conjugated F4-80 (#123110; Biolegend), PercpCy 5.5-conjugated CD11b (#101228; Biolegend). After staining, cells were washed and analyzed using a CytoFLEX flow cytometer (Beckman Coulter, Brea, CA). To calculate TLR4 dimers, the ratio of the geometric mean fluorescence intensity of TLR4/MD2 monomer antibody over total TLR4 antibody was calculated, TLR4 dimers % = 100 − (TLR4 mono gMean × %)/(TLR4 total gMean × %) × (100%)/(highest TLR4 ratio) [14,26].

A second set of C57 mice was subjected to the SDS protocol, and the brain was harvested on day 22 for an immunofluorescence assay. The prefrontal cortex (PFC) was dissected post-fixed in a solution containing 4% formaldehyde in 0.01M PBS for 24 hours and cryoprotected using 30% sucrose until the tissue sank into the solution. Using cryocut in a Leica cryostat, 30 mm slices were performed and stored in an antifreeze solution at -20 ^oC. After all the PFCs were cut, 5-6 slices from each sample were processed as free-floating sections, rinsed in 0.01M PBS and blocked using 2% BSA for 60 min at room temperature. Next, the sections were incubated overnight with an anti-Iba1 antibody (1:500, Wako, 019-19741) at room temperature. The secondary antibody goat Anti-rabbit Alexa 488 (1:1000, Thermo Fisher #A11034) was applied for 1 hour. Sections were mounted in superfrost plus slides and coverslipped with ProLong™ Gold Antifade Mounting (Thermo Fisher, #P36935). Z-stacks imaging from three different ROIs of a section and three sections from each animal were executed using a Leica TCS SP5 confocal microscope with LAS AF version 2.7.3.9723 software. For quantification, the images were transformed into maximum projection, and a blind experimenter manually counted the number of positive cells for Iba1.

2.6. Statistical Analysis

Statistical analyses were performed in GraphPad Prism v10 (GraphPad Software Inc., La Jolla, CA, United States). Data are expressed as mean ± Standard Error of the Mean (SEM), and statistical differences were determined using one-way or two-way variance analysis (ANOVA) followed by Bonferroni’s multiple comparison test. Means were considered significantly different when p < 0.05. The Institutional Animal Care and Use Committees of the University of California, San Diego, approved all animal experimental procedures performed in this work (IACUC - protocol number #S00137M).

3. Results

3.1. SDS Induces Persistent Mechanical Allodynia and Thermal Hyperalgesia

To assess whether chronic SDS can induce pain in juvenile mice, von Frey and Hargraves tests were performed at baseline, after four physical sessions (day 7), and after the end of the SDS protocol (days 11 and 21) (see experimental timeline in Figure 1A). Compared to the control group (no-SDS), stressed mice (SDS) demonstrated a slight decrease in the hind paw mechanical threshold on day 7, although not significant. From day 11, SDS animals showed a significant decrease in the paw threshold, which persisted until day 21 of the protocol, ten days after the last defeat session (Figure 1B). Similar results were observed in the thermal stimulus test, where the SDS group demonstrated a significantly reduced hind paw withdrawal latency to the heat (~43 ^oC) compared to controls on days 11 and 21 (Figure 1C). These findings show that SDS in juvenile male mice induces chronic mechanical allodynia and thermal hyperalgesia.

Figure 1. Behavioral profile of WT mice. (A) Experimental timeline, SDS protocol, and behavior evaluation time points. (B) Paw mechanical threshold by von Frey filament of SDS and no-SDS groups, days 0, 7, 11, and 21. (C) Paw thermal threshold by Hargreaves test of SDS and no-SDS groups, days 0, 7, 11, and 21. (D) Open-field test: total distance traveled in the arena (cm). (E) Mean body velocity of each animal (cm/s). (F) Maximum body acceleration (cm/s²). (G) Body mean mobility (%). Data represent the mean ± SEM. Two-way ANOVA followed by Bonferroni's multiple comparisons test for (A-B). One-way ANOVA followed by Bonferroni's multiple comparisons test for (D-G). *p < 0.05, **p < 0.01, ***p < 0.001, ***p < 0.0001, ***p < 0.00001, n= 6 (SDS); n=7 (no-SDS).

Figure 1. Behavioral profile of WT mice. (A) Experimental timeline, SDS protocol, and behavior evaluation time points. (B) Paw mechanical threshold by von Frey filament of SDS and no-SDS groups, days 0, 7, 11, and 21. (C) Paw thermal threshold by Hargreaves test of SDS and no-SDS groups, days 0, 7, 11, and 21. (D) Open-field test: total distance traveled in the arena (cm). (E) Mean body velocity of each animal (cm/s). (F) Maximum body acceleration (cm/s²). (G) Body mean mobility (%). Data represent the mean ± SEM. Two-way ANOVA followed by Bonferroni's multiple comparisons test for (A-B). One-way ANOVA followed by Bonferroni's multiple comparisons test for (D-G). *p < 0.05, **p < 0.01, ***p < 0.001, ***p < 0.0001, ***p < 0.00001, n= 6 (SDS); n=7 (no-SDS).

3.2. SDS Induces Long-Term Locomotor Changes and Short-Term Anxiety-Like Behavior

To assess anxiety-like behavior and locomotor activity, we performed the open-field and light-and-dark tests after the last stress session (day 10) and seven days after SDS (day 17). As observed in Figure 1D, stressed animals showed a significant decrease in spontaneous activity behavior (less distance traveled in the arena) during the open-field test compared to the no-SDS group on both days evaluated. Corroborating this result, the mean velocity and mobility were also decreased in SDS group (Figure 1E,F). Considering that mobility parameters indicate whether the mouse is highly mobile, moderately mobile, or immobile, our result suggests that chronic SDS resulted in a reduction of mobility, meaning less spontaneous motivation in the exploration of the arena, a profile that was not observed in the no-SDS group. Consequently, the decrease in mobility is reflected in the velocity measure since the speed is calculated by the distance traveled by the center, nose, or tail base point per unit of time. Interestingly, the max acceleration was significantly reduced on day 10 in SDS mice and reverted to a similar measurement of the control group on day 17 (Figure 1F). This behavioral profile uncovered that SDS in juvenile mice promoted persistent reduction in locomotor activity.

Furthermore, the open-field data showed that SDS triggers transient anxiety-like behavior. On day 10, the SDS group showed an anxiety-like phenotype, spending more time in the corners than in the center when compared to the non-SDS group (Figure 2A–H). This pattern was not continued after 7 days of the SDS; stressed animals show a recovery effect of the anxious state, presenting similar data to the no-SDS group at day 17 (Figure 2A–H). Despite these findings, the SDS protocol did not promote significant changes in the behavior of the stressed group in the light-and-dark test on day 10 or 17 compared to non-SDS (Figure 2I–K). Thus, an aversion to the light was not observed after SDS sessions of male juvenile mice.

3.3. TLR4 Activation in Spinal Cord, Dorsal Root Ganglia, and Prefrontal Cortex After SDS

TLR4 activation was analyzed by flow cytometry by calculating dimerization (ratio monometers to total expression) in immune cells of the DRG, spinal cord, and prefrontal cortex from the SDS and non-SDS groups. Significant changes were observed in the DRG and the spinal cord but not in the PFC. As seen in Figure 3B, TLR4 dimerization was significantly increased in microglia cells (CD11b⁺) of the lumbar spinal cord in the SDS group compared to the no-SDS group. Similarly, TLR4 dimerization was augmented on macrophage cells (CD11b⁺/F4-80⁺) of the DRG from stressed animals compared to non-stressed mice (Figure 3D). Besides this, analyses from Prefrontal cortex tissue in the flow cytometry or immunofluorescence assays did not show significant changes between the groups of animals regarding the number of Iba1⁺ cells (microglia activation marker) nor TLR4 dimerization in this population (Supplemental Figure 2A-D).

3.4. Genetic Ablation of TLR4 Expression Prevented Pain Phenotype Induced by SDS

To confirm the participation of TLR4 receptors in the development of chronic pain or anxiety-like behavior phenotype in SDS, Tlr4^-/- mice were subjected to the same protocol as the WT: six sessions of SDS, pain behavioral evaluation, and open-field test (see experimental timeline in Figure 4A). As reflected by the Von Frey readouts, SDS in Tlr4^-/- mice did not show mechanical allodynia at any time point evaluated compared to no-SDS-KO (Figure 4B). A similar pattern was observed in thermal response after SDS in Tlr4^-/- animals (Figure 4C). In the open-field test, the SDS-KO group decreased the time spent in the center of the arena in both times evaluated, days 10 and 17. However, there were no significant changes regarding the frequency to the center, the corner, and the time spent in the corners in SDS-KO compared to the no-SDS-KO group (Figure 4D,FG). With respect to locomotor activity parameters, total distance, velocity, acceleration, and mean mobility, no significant differences were detected between groups (Figure 4H–K). Together, these results suggest that TLR4 receptor depletion prevented locomotor activity reduction and chronic pain development induced by SDS.

4. Discussion

The current work reveals for the first time that SDS induces pain sensitivity in juvenile mice by involving TLR4 dimerization in the dorsal root ganglia macrophages and spinal cord microglia. In addition, SDS promotes persistent locomotor activity reduction in stressed animals compared to non-stressed animals. The lack of TLR4 expression in knock-out mice prevented the development of chronic pain phenotypes and locomotor changes induced by SDS.

A physiological stress response can be triggered by physical or psychological stressors. In these circumstances, the amygdala, in communication with the brain stem, promotes sympathetic adrenergic catecholamines, norepinephrine, and epinephrine release, resulting in typical stress symptoms: palpitation, high blood pressure, sweating, and pupil dilatation [4]. Long-lasting stress evokes hyperactivity in this system, lasting for an extended period even after the stress is discontinued [27]. In humans, chronic stressful experiences during childhood and adolescence (periods of intense social interaction) are shown to significantly impact mental health, resulting in anxiety and depression disorders, which can persist in adulthood [28]. Data from 2019 demonstrated that 970 million adults were found to be suffering from mental health disorders, and around 301 million suffer from anxiety, out of which 58 million were children and adolescents [29]. In parallel, studies proved that mental disorders caused by chronic stress exposure significantly affect the nociception pathways [27]. Animals subjected to stressful events presented an enhanced c-fos expression in the spinal cord, an increase in glutamate, and a reduction of GABA, leading to enhanced central excitability [30,31,32]. In humans, chronic stress alongside psychological disorders proved to be the primary trigger for widespread musculoskeletal pain development, known as fibromyalgia, one of the most challenging syndromes in clinics regarding pain management [33]. Collectively, these findings highlight the strong linkage between stress and pain processing. Here, we demonstrate that six sessions of SDS promote chronic pain phenotype in juvenile male mice subjugated and intimated (physically and sensorially) by the retired aggressor CD1 mice. The SDS group presented a reduction of thermal and mechanical thresholds persisting up to 21 days, 10 days after the aggressor had been removed from the cage. Similarly, previous work showed that ten consecutive days of social stress in adult mice leads to persistent hyperalgesia, long-lasting depressive-like behavior, and social avoidance [10]. In our open field data, however, SDS animals transiently exhibited depressive-like behavior; significant differences between the groups were only detected on day 10. Stressed mice presented lower distance traveled in the arena and less visited to the center (a spot considered "unsafe" and favorable to attacks) compared to non-stressed. Thus, the data suggests that SDS reduces the spontaneous interest in exploring the arena, tending to spend more time in the corners. However, this phenotype was abolished after 7 days of social stress sessions, contrary to what was observed in the behavioral pain tests where allodynia persisted until day 21. These findings indicate that juvenile mice might exhibit different anxiety behavior patterns compared to adults, recovering their interest in exploring the arena after a few days of SDS. Interestingly, the SDS protocol did not promote the natural aversion of stressed mice to brightly illuminated areas on our light-and-dark test. Nevertheless, in another study, stressed juvenile mice did not show a significant difference in exploring the arena compared to the control group in the open-field test, although they exhibited an increase in immobility behavior at tail-suspension and forced swimming tests [11]. Similar changes in locomotor activity due to stress were found in our results; the SDS group had reduced velocity and acceleration on both time points analyzed, which was reflected in lower body mobility than the non-stressed group. Accordingly, it has been reported that neonatal maternal separation-induced stress reduces exploratory activity in adult rats [34]. Unlike, the chronic restraint stress model performed for 5 days in adult rats promotes an increase in exploratory behavior [35]. Even though conflicted results, this evidence suggests that breed, age, and type or duration of the stress protocol can lead to distinct anxiety-like behavior parameters in rodents.

Recent works indicated chronic stress regulates nociception transmission through neuroinflammatory signaling involving microglia-mediated neuronal remodeling [36]. TLR4 receptors are abundantly expressed in brain and spinal microglia and typically, when activated, trigger an inflammatory cascade, leading to pro-nociceptive mediators release (e.g., TNF-α, COX-2, and interferon-β) [17]. According to our previous findings and in line with other works, TLR4 is a critical microglial receptor responsible for developing and maintaining pain [14,15,16,18,19,37,38]. In vivo TLR4 blockade attenuates hyperalgesia triggered by intrathecal LPS and neuropathic pain induced by nerve constriction injury or by chemotherapy drugs [15,16,18]. Although TLR4 is reported as an innate immune receptor, recent evidence shows that the interaction of TLR4 with N-Metil-D-Aspartate (NMDA) and Transient receptor potential vanilloid 1(TRPV1) can affect the nociceptive transmission via downstream phosphorylation of second messengers and posterior changes in primary afferent neurons transcriptome leading to changes in neuroplasticity [39] [14].

The dimerization process implies the recruitment of the TLR4 receptor at lipid rafts platforms via ligands, promoting cholesterol efflux and downstream signaling activation, which means the functionality of the receptor in the cell. In our data, TLR4 dimerization was associated with pain induced by social stress in juvenile mice since spinal microglia and dorsal root ganglia macrophages from stressed animals presented a higher level of TLR4 dimerization than the control group 10 days after the end of SDS protocol. Previously, using the social defeat stress model, a group found an increase in microglial activation and overexpression of inflammatory genes such as TLR4 in the spinal cord, which was associated with stress-induced hyperalgesia [36]. Accordingly, in this work, the global genetic depletion of TLR4 prevented the development of mechanical allodynia or thermal hyperalgesia, and no changes were observed in locomotor activity parameters, such as velocity, acceleration, or mean mobility in SDS-KO mice compared to no-SDS KO. However, a partial anxiety-like was observed in KO mice subjected to SDS since they spent less time in the center of the arena on days 10 and 17 of the protocol. Thus, the data indicated that TLR4 receptors seem to be more related to the pain phenotype than the anxiety disorder promoted by social stress in juvenile mice.

The HMGB1 (chromatin protein high-mobility group box1) is shown to be extracellularly released by immune cells such as macrophages and microglia under a stress state [40]. Into the extracellular space, through TLRs (including TLR4) signals, induce NF-κB signaling and proinflammatory cytokine production [41]. Considering the DRG is located outside the blood-brain barrier [42], circulating products released by stress from small amines (noradrenalin) to larger proteins (hormones and cytokines) can easily reach the cells of DRG, which can reflect changes in TLR4 signaling. Moreover, bulbospinal pathways are recognized to regulate dorsal horn excitability. Of note, serotonergic pathways facilitate pain signaling, activating dorsal horn neurons and microglia [43,44,45]. Serotonin receptor activation increases afferent terminal excitability [46] and activates DRG neurons [47]. Consistent with these observations, stress, as with models of conditioned fear mediated by supraspinal integration and learning, produces a profound change in spinal processing through the activation of bulbospinal serotonin projections [48]. These considerations jointly point to the possibility that higher order processes can initiate changes in downstream neuraxial signaling.

5. Conclusions

SDS in juvenile male mice promotes chronic pain phenotype and persistent locomotor activity reduction. These outcomes are related to the dimerization of TLR4 receptors in spinal microglia and dorsal root ganglia macrophages. The pain phenotype caused by SDS was prevented by genetic ablation of TLR4 receptors in stressed mice.