Skip to main content
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2026 May 16.
Published in final edited form as: Pain. 2026 Apr 17;167(7):e258–e272. doi: 10.1097/j.pain.0000000000003978

Neuroimmune pathways linking sleep disturbance to persistent pain after injury are mitigated by exercise

David M Klyne 1, Megan Van Der Bas 2, Michele Y Harris 3, Brendan A Hilliard 2, Parth R Patel 2, Abby Kegg 2, Astha Rajpal 2, Khyleisha Ceasar 2, Betsy A Kalicharan 2, Paul W Hodges 1, Mary F Barbe 2
PMCID: PMC13177422  NIHMSID: NIHMS2168136  PMID: 42017897

Abstract

Sleep and physical activity are modifiable behaviours linked to pain. Sleep disturbance often co-occurs with persistent pain and may contribute to its development. Exercise is a first-line treatment for chronic pain. Previous work showed that sleep disturbance worsens and prolongs post-injury pain behaviours, exercise mitigates these effects, and brain-derived neurotrophic factor (BDNF) may play a mechanistic role. Deeper insight requires a broader assessment of pain behaviours and systemic biomarkers related to inflammation, tissue repair, and neuromodulation. This study addresses these gaps. Twenty-nine adult female Sprague-Dawley rats performed an intensive lever-pulling task for 4 weeks to induce overuse injury, then underwent one of three 4-week interventions: intermittent sleep disturbance, voluntary exercise (access to a running wheel), or both. Pain-related behaviours and 71 blood analytes were measured immediately pre-injury, post-injury, and post-intervention. Overuse injury decreased grip strength and increased mechanical sensitivity in the injured forepaws. Following cessation of the injury inducing task, these changes persisted with sleep disturbance but recovered to, or exceeded, pre-injury levels with exercise, even with concurrent sleep disturbance. Biomarker analyses revealed distinct neuroimmune responses to injury and sleep disturbance, particularly mediators of inflammation and neuroplasticity, that were offset by exercise. Correlations between biomarkers and behavioural outcomes support mechanistic links between injury, sleep, exercise and recovery. Findings demonstrate that post-injury sleep disturbance induces neuroimmune changes that increase persistent pain vulnerability whereas aerobic exercise counters these effects. This highlights the interaction between sleep and exercise in recovery and their potential as strategies to prevent and manage chronic pain.

Keywords: Pain, transition to chronicity, sleep disturbance, exercise, inflammation, cytokines, neuroplasticity

Summary

Post-injury sleep disturbance drives neuroimmune changes that sustain pain behaviours, whereas concurrent aerobic exercise counteracts these effects, highlighting sleep–exercise interactions in shaping pain outcomes.

1. Introduction

Sleep and physical activity are essential and modifiable health behaviours, and both are closely associated with pain. Poor-quality or insufficient sleep (i.e., sleep disturbance) commonly co-occurs with acute pain [57], painful flare-ups [23], and persistent, chronic pain [101]. Once viewed as a consequence of pain, growing evidence suggests that sleep disturbance may also drive pain [21; 24; 42; 61; 74], and perhaps more strongly and frequently than the reverse [43]. Likewise, physical inactivity is closely associated with more severe and persistent pain and often precedes its onset [53; 72; 83]. Unsurprisingly, exercise is among the most effective treatments for pain [35; 67; 106]. These observations raise two compelling questions: (1) Can sleep disturbance promote the transition from acute to chronic pain? and (2) Can exercise protect against these adverse effects? Answering these questions could have meaningful clinical implications.

We recently explored these questions in a short communication by exposing rats to 4 weeks of intermittent sleep disturbance, aerobic exercise (wheel running), or both, following 4 weeks of intensive repetitive reaching and pulling [61] that results in overuse injury with pain [9; 22; 61; 102]. As predicted, post-injury forepaw mechanical sensitivity—assessed using a limited range of von Frey monofilaments—worsened with sleep disturbance but improved with exercise. Supporting our hypothesis, sensitivity also improved with concurrent exposure to both interventions, suggesting that exercise mitigated the pro-nociceptive effects of sleep disturbance. Similar patterns emerged for grip strength in the injured limb; however, post-injury exercise did not significantly offset the effects of injury when combined with sleep disturbance. This finding might be explained by inter-individual variability, as grip strength was not normalised to each animal’s baseline in that study [61].

These behavioural changes were paralleled by systemic levels of brain-derived neurotrophic factor (BDNF), which increased with injury and sleep disturbance but returned to baseline with exercise, with or without sleep disturbance. This is notable given BDNF’s roles in sleep regulation [27; 99], and in synaptic plasticity within both the peripheral and central nervous systems, where it likely contributes to sensitisation processes underlying pain chronification [82; 107; 118].

A critical next step is to establish more granular pain behaviour profiles in response to sleep disturbance and/or exercise and to explore how they relate to biomarkers of inflammation, repair, and neuromodulation. Sleep disturbance is thought to exacerbate pain, in part, by enhancing pro-inflammatory signalling [49; 61], whereas exercise may relieve pain by dampening these same pathways [61; 102; 116]. Key neuromodulators like BDNF, which support tissue repair and neural plasticity, are tightly regulated by inflammatory cascades. Thus, the interplay between inflammation and neuromodulation may constitute a core mechanism through which sleep and exercise shape pain outcomes—either facilitating recovery or promoting chronicity.

This study builds on our earlier work [61] by further characterising longitudinal changes in mechanical sensitivity, grip strength and a broad spectrum of systemic biomarkers associated with inflammation, repair and neuromodulation in rats exposed for 4 weeks of an injury-inducing repetitive task, followed by intermittent sleep disturbance, aerobic exercise, or both after task cessation. We hypothesised that sleep disturbance would prolong or worsen post-injury pain behaviours, that exercise would counteract these effects, and that specific neuroimmune signatures would underlie these outcomes.

2. Materials and Methods

2.1. Animals

Experiments were pre-approved by the Institutional Animal Care and Use Committee and were compliant with US NIH guidelines for the humane care and use of laboratory animals [41]. Forty-four female Sprague-Dawley rats were procured from Charles River (Wilmington, MA, USA). Rats were housed in an AAALAC-accredited central animal facility where they were handled daily under 12-hour light/12-hour dark cycle conditions, with free access to water, at 22–23°C room temperature, in standard filter topped cages (ventilated and with hardwood chip rodent bedding) that were changed twice per week. Rats were group housed until experimental onset. Thereafter, all rats were housed in individually transparent plastic cages due to the need for food restriction (necessary for motivation to perform the task as described below).

Rats were numbered and randomly assigned to groups at the beginning of the study to ensure blinding. Study rats were then food restricted to 5% (based on animal weight) less than age-matched normal control rats with free-access-to-food; these latter (age-matched food “unrestricted” control) rats were used for weight comparison purposes only. All rats were handled at least twice per week and provided cage enrichment toys daily that included chew bones and tunnels for environmental enrichment. Rats were checked daily for health, feed and water by laboratory or institutional laboratory animal staff, under the supervision of the institutional veterinarian. To reduce illness-related confounders, all animals were weighed weekly, examined at least twice weekly for indices of illness (one rat was excluded from analyses as illness was detected at study end as described below), and examined weekly for tumors (none were detected).

The decision to use only female rats was based on the biomechanical constraints of our injury model, which is currently tailored to the pulling strength of female rats. Inclusion of males would have compromised data quality and interpretation. Further, sex is an important factor in the modulation of pain, and the inclusion of males would have introduced confound in the data that would have not been adequately controlled for due to our group sizes (n ≤10) and many study arms. This limitation is partially offset by epidemiological data indicating that women are more likely to develop overuse injuries [84; 85; 104] and chronic pain conditions [13; 17; 108], particularly in association with poor sleep quality [49; 74]. To account for hormonal variability across the estrous cycle (rodent equivalent of the menstrual cycle), we assessed systemic estradiol levels and found no group differences at pre-injury or post-injury (see Section 2.2).

2.2. Study design

A total of 44 animals were procured for both the main interventional study (n=29) and supplemental studies (n=15, which included both negative and positive controls). One was excluded early as it failed to learn the injury-inducing task and another was excluded from analyses as it displayed signs of illness (i.e., evidenced by lack of gain weight, abnormal blood cell counts, and refusal to run on the exercise wheel)—resulting in a total of 27 animals for analysis of the effects of secondary interventions.

Following the induction of overuse injury of the median nerve and forelimb musculoskeletal tissues (as a consequence of 4 weeks of repetitive and forceful lever bar pulling) [3; 9] (see Section 2.3), main study rats were randomly assigned to one of three 4-week interventions: (1) sleep disturbance (SD; n=11), (2) voluntary exercise via access to a running wheel (Ex; n=9 initially, n=8 by study end due to illness) or (3) both (SD+Ex; n=8). Longitudinal measures of behavioural outcomes and systemic biomarkers were collected at three time-points: just prior to injury induction (pre-injury), immediately following the 4-week injury induction phase (post-injury), and immediately after the 4-week intervention/exposure phase (post-intervention). This within-subject, longitudinal design permitted repeated monitoring of each animal over time without the need to remove or euthanize animals at intermediate time-points, thereby ensuring robust evaluation of intervention effects.

An additional nine animals served as negative controls and six as positive controls. Negative control animals were housed similarly but did not perform the injury-inducing high repetition high force (HRHF) task and were not exposed to any interventions; these were termed “Normal Controls”. Positive control animals performed the same 4-week injury inducing HRHF task as the main study animals but rested during the 4-week intervention period (i.e., were not exposed to SD, Ex or SD+Ex) and were termed “HRHF+Rest”. Behavioural measures and select systemic biomarkers were longitudinally collected at the same time-points as the main study: pre-injury, 4-weeks after onset of the experiment or injury induction phase, and 4-weeks later. Results for these groups are provided in Supplemental Figures 2 and 3.

2.3. Induction of overuse injury by voluntary performance of a high repetition high force task for 4 weeks

Sixteen custom-designed operant behavioural chambers were used in which rats performed an operant reaching and lever pulling task, as previously described [6]. Specific details regarding training and task procedures for the current study animals have been reported elsewhere [61]. Briefly, rats were acclimated to the facility for 3 days before handling (5 days/week) commenced. They were then familiarized to the operant chambers for 1 week before being trained in them for over 4 weeks (15 min/day, 5 days/week) to learn a reaching and lever-pulling task at high loads, gradually increasing from naïve loads with no specific reach rate, as previously described [6]. Following training, rats performed a high-repetition, high-force reaching and lever‐pulling task for 4 weeks in the custom apparatuses. Animals reached through a shoulder height portal and isometrically pulled a force handle attached to a force transducer with a load cell (Futek Advanced Sensor Technology, Irvine, CA, USA) located outside the chamber wall, as depicted previously [11]. The load cell was interfaced with custom Force-Lever software (version 1.03.02, Med Associates, St. Albans, VT, USA). Auditory and light indicators cued the reaching rate. If reach and force criteria (see below) were met within a 5-second cueing period, a 45 mg food pellet was dispensed into a food trough as a reward. The reach and force criteria involved pulling the lever bar at 53% ± 5% of the rats’ maximum pulling force at a target reach rate of 4 reaches/min, for 2 hours/day, over four 30-minute intervals (separated by a 1.5-hour break), for 3 days/week. We have previously shown that this task induces sensorimotor declines and both local and systemic inflammation [6; 8] that are consistent with changes observed in humans with acute-onset upper extremity repetitive strain injuries [19; 39].

During the training and task period, rats were food restricted to weigh 5% less than age-matched normal control rats with free access to food (these latter rats were used for weight comparison purposes only). This food restriction was necessary to motivate rats to work for a food reward. All rats were weighed twice per week, provided regular rat chow daily in addition to food reward pellets (banana [F0024] and chocolate grain-based [F0165] dustless precision pellets, Bio-Serv, Flemington, NJ, USA), to allow for body weight gains over the course of the experiment [102].

2.4. Post-injury sleep disturbance for 4 weeks

Sleep disturbance involved frequent arousing of rats during the light-phase (12 h from 0600–1800, as rats are nocturnal) on four random days of the week to avoid circadian and sleep-pattern adaptations. This was achieved by replacing an old object with a new stimulatory object in their cage when the rat appeared to become drowsy. Stimulatory objects included plastic toys and tubes of different sizes, plastic boxes containing sucrose food reward pellets that could be sensed but not accessed by the animal, and nesting material (see Supplemental Fig 1). When necessary, additional arousal was achieved by disturbing the bedding/nest or adding new bedding material. Cages were also exchanged twice a day (1200 and 1400) for fresh ones with fresh bedding to provide additional stimulation. This method avoids the confounding impact of acute stress associated with most sleep restriction/disturbance methods and better reflects sleep disturbance in humans in the real-world, as previously described (see Appendix A Methods in [61]).

Time spent sleeping was assessed immediately following the 4-week intervention period (post-intervention) in all rats, regardless of whether they were exposed to sleep disturbance. Rats were left undisturbed in a quiet room between 15:40 and 17:50, corresponding to the end of the light (inactive) phase and prior to the onset of the dark (active) phase (18:00–06:00). Sleep was monitored in the home cage using side-mounted cameras across this 2.1-h period, during three 10-min sampling windows (0–10, 60–70, and 120–130 min). Rats were classified as asleep if they remained completely inactive for ≥10 s.

2.5. Post-injury voluntary exercise for 4 weeks

Voluntary, unrestricted exercise was provided by individually housing rats in cages equipped with a built-in running wheel (Panlab Harvard Apparatus, Holliston, MA, USA) for 12 hours during the dark-phase (1800–600), 5 days a week. This method of “voluntary” aerobic exercise, as previously described [61], avoids complications such as tissue damage [102] and acute stress [105] that can be induced by forced exercise protocols. Daily wheel revolutions were recorded digitally using an automated rotameter (LE3806 multicounter; Harvard Apparatus) and exported using SeDaCom software (Harvard Apparatus). No differences in total running distance were observed between groups with wheel access, as previously reported (mean ± SEM km: Ex = 3.0 ± 0.3; SD+Ex = 2.6 ± 0.3; p = 0.09) [61]. To prevent potential disruptions to sleep from the stimulatory effects of running, wheel access was not allowed during the light-phase (600–1800), during which rats were returned to their usual cage (i.e., without a wheel). “Non-exercise” rats remained in their usual cage without wheels. All rats were housed with tunnels and chew toys and were handled daily, allowing for some voluntary physical activity.

2.6. Longitudinal measures of pain-related behaviours

All animals were acclimated to the measurement apparatuses and experimenters for a minimum of 2 weeks prior to data collection, which included at least 30 minutes of exposure to each apparatus. Following acclimation, reflexive grip strength (lower values indicative of muscle myalgia [44]) and forepaw mechanical sensitivity were assessed at three time-points: (1) pre-injury, (2) post-injury, and (3) post-intervention. All assessments were performed by experienced experimenters who were blinded to group allocation.

Reflexive grip strength was assessed using a rat grip strength meter (1027SR-D58, Columbus Instruments, Columbus, OH, USA). Each forelimb was tested 3–5 times, with the maximum grip strength for each limb recorded. Data from the preferred reach limb was reported in two ways: (1) raw maximum grip strength (cN) and (2) percent change from baseline (i.e., pre-injury) to both follow-up time-points (i.e., post-injury and post-intervention).

Forepaw mechanical sensitivity was assessed bilaterally using von Frey monofilaments from a touch-test monofilament set (North Coast Medical Inc), as previously described [22]. Animals were placed—five at a time—in an elevated clear acrylic apparatus with five individual chambers (for each rat) and a wire-mesh floor (8 mm spacing) and allowed to habituate for at least 10 minutes. The volar surface of each forepaw was stimulated using a series of 5 calibrated monofilaments (ranging from 0.16, 0.4, 1, 4, 7.8, and 15 cN; corresponding to 0.16, 0.4, 1, 4, 8, and 15g, and evaluator sizes 3.22, 3.16, 4.08, 4.56, 4.93 and 5.18, respectively). Filaments were applied in ascending order of force, with each applied 10 times in a pseudorandom order to the midplantar surface of each forepaw, and held in place for 5 seconds or until a withdrawal occurred. A forelimb withdrawal was defined as: elevating the forepaw, elevating the forepaw and licking it, or elevating the forepaw and shaking it. Stimuli were only delivered when the paw was in contact with the mesh floor. For each filament, the number of withdrawals (maximum score = 10) was recorded per forepaw, with higher scores indicative of greater sensitivity. Data from the preferred reach limb was reported and used for analysis.

2.7. Longitudinal measures of systemic biomarkers

Tail vein blood (0.5 to 0.7 ml) was collected under brief, low-dose isoflurane anaesthesia (to minimise stress) at each respective time-point within a fixed time-window (0900–1000, to control for circadian variation in analyte levels [90; 98; 110]) for determination of serum concentrations of 71 analytes of interest. Anaesthesia was applied uniformly across all animals, minimizing its potential impact on group differences. For the final post-intervention time-point, blood was collected 16 and 28 hours after the last sleep disturbance and exercise session, respectively, to avoid the acute effects of these activities on cytokine fluctuations [20; 80]. Tail vein blood was collected into uncoated tubes, allowed to clot for ~45 min, and centrifuged at 12000 revolutions per minute for 20 minutes at 4 °C. The resulting serum (the supernatant) was aliquoted into 200 μl microcentrifuge tubes and stored at −80 °C until assayed.

Sixty-seven protein analytes were assayed using RayBiotech’s Quantibody Rat Cytokine Array 67 platform (QAR-CAA-67, RayBiotech, Norcross, GA, USA), performed as a manufacturer-provided service. This multiplex platform enables the simultaneous quantification of 67 biomarkers—including cytokines, inflammatory factors, growth factors, and receptors—using matched antibody pairs. Each sample (and each biomarker) was analysed four times, with results averaged and normalized to internal positive and negative controls. Absolute concentrations were calculated based on standard curves generated in parallel for each biomarker as part of the array. All procedures were performed in-house by RayBiotech (Norcross, GA, USA) according to their protocol, antibody verification steps, and software analysis on blinded samples. The manufacturer reports an inter-assay variability of ≤12% coefficient of variation (CV) for its standard sandwich ELISA kits.

In addition, select analytes were assessed in-house via individual ELISAs using commercially available kits: BDNF (DBNT00, R&D Systems, Minneapolis, MN, USA; sensitivity: 1.35 pg/ml), estradiol (ES180S-100, CalBiotech Inc., El Cajon, CA, USA; sensitivity: 3 pg/ml), corticosterone (55-CORMS-E01, Alpco, Salem, NH, USA; sensitivity: 6.1 ng/ml), and TNFα (EA100366, OriGene, Rockville, MD, USA; sensitivity: 1 pg/ml). Absorbance was measured using either a Paradigm (Beckman Coulter, Inc., CA, USA) or GloMax® (GM3030, Promega, Madison, WI, USA) microplate reader, following each manufacturer’s protocol. Values below the assay detection threshold were recorded as zero. ELISAs were performed by an experimenter blinded to group allocation.

All analytes are listed and defined in Supplemental Table 1. Select analytes were verified using slot blot methodology; results are provided in Supplemental Figures 4–10.

2.8. Statistical analysis

GraphPad PRISM version 10.3.1 was used for analyses (GraphPad Software, Boston, MA, USA). Sample sizes of 6 to 11 rats per group was based on our prior studies showing that a minimum of 5 rats per group was sufficient to detect differences across the measured outcomes [4; 7; 9; 16; 102]. Statistical significance was set at p <0.05. Data are presented as mean ± SEM, and post-hoc results are reported in the text, figures, and/or tables, with adjusted p-values provided where applicable.

Of the 71 biomarkers assessed by protein microarray or ELISA, 21 were excluded from analyses due to low detectability—defined as “0” or “negative” values in ≥65% of the entire dataset (i.e., all cases across all time-points) and/or in ≥85% of cases for a single time-point. In total, 46 biomarkers assayed using protein microarray and four biomarkers assayed by ELISA (total = 50) were included in the final analysis. Only biomarkers showing significant main effects and/or interactions are plotted; full results for all biomarkers are provided in Supplemental Tables 2 and 3.

Raw maximum reflexive grip strength and systemic biomarkers were compared between Groups (SD vs. Ex vs. SD+Ex) over Time (pre-injury vs. post-injury vs. post-intervention) using repeated-measures mixed-effects models (Restricted Maximum Likelihood, REML). Percent change in maximum reflexive grip strength was similarly analysed between Groups but across only two Time intervals (pre-injury to post-injury vs. pre-injury to post-intervention). Mechanical sensitivity was also assessed using repeated-measures mixed-effects models in two ways. First, by comparing the number of withdrawal responses over Time (pre-injury vs. post-injury vs. post-intervention) to different Filament Sizes, for each group, separately. Second, by comparing the number of withdrawal responses to different Filament Sizes between Groups (SD vs. Ex vs. SD+Ex) for each time-point (pre-injury, post-injury and post-intervention), separately. Post-hoc comparisons between groups and time-points were performed using Tukey’s or Fisher’s Least Significant Difference tests, with p-values adjusted for multiple comparisons and reported.

In Supplemental Figure 2, the same repeated-measures mixed-effects model and post-hoc tests were extended to include the HRHF+Rest (positive control) group to evaluate differences in time spent sleeping across three post-intervention sleep tracking time windows (0–10, 60–70, and 120–130 min; see Section 2.4). In Supplemental Figure 3, the same models were extended to include the Normal Control (negative control) and HRHF+Rest (positive control) groups, to evaluate differences between these and the intervention Groups (Normal Control vs. HRHF+Rest vs. SD vs. Ex vs. SD+Ex) over Time (pre-injury vs. post-injury vs. post-intervention) for selected pain-behavioural measures and biomarkers (see Section 2.2).

For biomarkers found to be significantly impacted by Group, Time and/or their interaction in the above analyses, Pearson’s or Spearman’s rank correlations were applied as appropriate (Pearson’s for continuous data, Spearman’s for ordinal data) to examine relationships with behavioural outcomes. These correlations do not account for repeated measures within animals and therefore reflect overall associations across time rather than within-animal changes. Each biomarker was correlated individually with each behavioural measure, using compiled values across all time-points. Correlation coefficients (r) are reported as follows: <0.3 (or >−0.3), weak; 0.3 to 0.59 (or −0.3 to −0.59), moderate; >0.59 (or <−0.59), strong.

Potential neuroimmune patterns in response to sleep disturbance and/or exercise were explored using unbiased hierarchical clustering. Clustering was performed on post-intervention concentrations of biomarkers identified as significantly affected by group, time, and/or their interaction (i.e., those included in correlation analyses). Analysis was conducted using the publicly available ClustVis web tool (https://biit.cs.ut.ee/clustvis/), applying the correlation distance metric and average linkage method. The optimal number of clusters was determined by inspecting the dendrogram/heatmap and limiting the minimum number of animals per cluster to six. After clustering, 1-way ANOVA or Pearson Chi-squared tests were used to evaluate differences between clusters in: (1) biomarker levels (i.e., to define their neuroimmune profiles), (2) behavioural outcomes (i.e., % change in grip strength from pre-injury, maximum grip strength, forepaw sensitivity), and (3) group representation (SD vs. Ex vs. SD+Ex).

3. Results

3.1. Time spent sleeping was greater in response to sleep disturbance than normal sleep, independent of exercise

Time spent sleeping was assessed at the end of the final post-intervention week (i.e., 4 weeks post-intervention) in all experimental groups, including the no-intervention control group (HRHF+Rest). Main effects for Group (F: 1, 38 = 7.56, p=0.004) and Time (F: 2, 26 = 12.87, p<0.001) were observed (Supplemental Figure 2). Groups exposed to sleep disturbance (SD and SD+Ex) spent more time sleeping than groups allowed to sleep normally (Ex and HRHF+Rest) at all camera-tracking time-points assessed during the 2.1-h observation period (0–10, 60–70, and 120–130 min; p<0.05). Across groups, most sleep occurred during the 60–70 min window.

3.2. Post-injury grip strength was restored by exercise, even with sleep disturbance, but did not recover with sleep disturbance alone

A significant Group × Time interaction was found for maximum reflexive grip strength (F: 4, 44 = 2.8, p=0.039). Following injury, maximum grip strength had declined from pre-injury levels (all p<0.01) before recovering to varying degrees after the intervention (all p<0.01), for each group (Fig. 1A). Specifically, Ex and SD+Ex rats returned to or exceeded their pre-injury levels, whereas SD rats did not (p=0.04). Post-intervention, maximum reflexive grip strength was lower in SD animals than Ex animals (p=0.003).

Figure 1.

Figure 1.

Reflexive grip strength. Effects of injury (assessed immediately after 4 weeks [wks] of a high repetition high force [HRHF] lever pulling task) and then of 4 weeks of sleep disturbance (SD), exercise (Ex) or both (SD+Ex) on: (A) maximum reflexive grip strength, and (B) percent change in reflexive grip strength from pre-injury (pre-inj) to post-injury (post-inj) and post-intervention (post-int). *p<0.05, **p<0.01.

Main effects for Group (F: 2, 24 = 4.9, p=0.016) and Time (F: 1, 22 = 47.54, p<0.001) were found for percent change in grip strength relative to pre-injury levels (Fig. 1B). Similar declines were observed post-injury for each group (ranging from a 21 to 30% reduction; Fig. 1B). Although grip strength improved in all groups post-intervention (i.e., relative to the percentage change from pre-injury to post-injury, p<0.006), the degree of recovery varied substantially. SD rats remained well below (11% reduction) pre-injury levels, SD+Ex rats recovered (1% reduction) to pre-injury levels, and Ex rats surpassed (15% increase) pre-injury levels. Overall recovery across timepoints (i.e., pre-injury to both post-injury and post-intervention) was lower in SD than Ex rats (p=0.014).

Supplemental Figure 3A includes longitudinal reflexive grip strength data for Normal Control and HRHF+Rest animals. A significant Group × Time interaction was observed (F: 8, 74 = 3.0, p=0.006). In Normal Control (uninjured [no exposure to repeated loading], no intervention) rats, grip strength did not differ from other groups (SD, Ex, SD+Ex, HRHF+Rest) at baseline (pre-injury), increased over the 8-week study period (p=0.009 vs. pre-injury), and remained higher than all groups at 4 weeks (post-injury time-point; p<0.004), and higher than SD (p<0.001) and SD+Ex (p=0.022) groups at 8 weeks (post-intervention time-point). In HRHF+Rest (no intervention) rats, grip strength did not differ from other groups at baseline, decreased after injury (p=0.006), then recovered to pre-injury levels following 4 weeks of rest (p=0.001), and was greater than SD rats at this final 8-week time-point (p=0.039).

3.3. Post-injury forepaw mechanical sensitivity was restored by exercise, even with sleep disturbance, but enhanced by sleep disturbance alone

A significant Filament Size × Time interaction was found for SD (F: 10, 136 = 3.1, p=0.002), Ex (F: 10, 95 = 3.3, p=0.001) and SD+Ex (F: 10, 105 = 2.9, p=0.003) rats. With respect to the SD group, greater sensitivity (as defined by a greater number of withdrawal responses) was observed in response to 4 cN and 8 cN monofilament probings at the post-injury time-point, and 1 cN, 4 cN and 8 cN probings at the post-intervention time-point, relative to pre-injury (p<0.05; Fig. 2A). With respect to the Ex and SD+Ex groups, greater sensitivity in response to 1 cN and/or 4 cN probings was observed only post-injury, relative to pre-injury (p<0.035; Fig. 2B, C). Between-group comparisons revealed no interactions or main effects at the pre-injury (Fig. 2D) and post-injury (Fig. 2E) time-points. However, a significant Filament Size × Group interaction emerged post-intervention (F: 10, 105 = 2.5, p=0.011), with SD rats showing greater sensitivity to 4 cN monofilament probings compared to Ex+SD rats (p=0.043; Fig. 2F).

Figure 2.

Figure 2.

Forepaw mechanical sensitivity. (A-C) Effects of injury and then of 4 weeks of sleep disturbance (SD; A), exercise (Ex; B), or both (SD+Ex; C) on forepaw mechanical sensitivity to 0.16, 0.4, 1, 4, 8 and 15 cN monofilament testing. Number of limb withdrawals to 10 probings per filament is reported. (D) Post-intervention results for each group (SD vs. Ex vs. SD+Ex) are compared. (A-C): #p<0.05, ##p<0.01 and ###p<0.001, pre-injury (Pre-inj) vs. post-injury (Post-inj); $p<0.05, pre-injury vs. post-intervention (Post-int); (D): *p=0.043, SD vs. SD+Ex.

Supplemental Figure 3B includes post-intervention (8-week time-point) data for Normal Control and HRHF+Rest animals. A significant Filament Size × Group interaction was observed (F: 20, 170 = 2.3, p=0.002), with lower sensitivity to 1 cN, 4 cN, 8cN, and/or 15 cN monofilament probings in Normal Control (i.e., uninjured) rats than all other groups (SD, Ex, SD+Ex. HRHF+Rest; p<0.038), and greater sensitivity to 1 cN and/or 4 cN probings in HRHF+Rest (i.e., injury only) rats than Normal Control, Ex and SD+Ex groups (p<0.05).

3.4. Systemic biomarkers show responses specific to injury and/or intervention, with several influenced by sleep disturbance alone but not when combined with exercise

Full summary statistics and mixed-effects model results for all biomarkers are presented in Supplemental Tables 2 and 3. Of these biomarkers, a significant main effect for Group and/or Time, and/or a Group × Time interaction, was found and plotted for 18 (Fig. 3A–R).

Figure 3.

Figure 3.

Systemic biomarkers. Effects of injury and then of 4 weeks of sleep disturbance (SD), exercise (Ex), or both (SD+Ex) on systemic concentrations of: (A) BDNF, (B) CINC-1, (C) corticosterone, (D) eotaxin, (E) estradiol, (F) GAS-1, (G) HGF, (H) IFNγ, (I) IL-1β, (J) LIX, (K) MCP-1 (L) Neuropilin-2, (M) NOTCH-2, (N) RANTES, (O) TCK-1, (P) TIM-1, (Q) TIMP-1, and (R) TNFα. Biomarker abbreviations and basic functions are defined in Supplemental Table 1. Post hoc differences for interaction effects are shown by the “✱” symbol: *p<0.05, **p<0.01, ***p<0.001. Post hoc differences for Time effects are shown by the “#” symbol: #p<0.05, ##p<0.01 and ###p<0.001. Post hoc differences for Group effects are shown by the “$” symbol: $ $p<0.01.

A significant Group × Time interaction was observed for six biomarkers: BDNF, corticosterone, estradiol, GAS-1, Neuropilin 2, and tissue inhibitor of metalloproteinase-1 (TIMP-1). As previously reported, BDNF increased post-injury and was further increased by sleep disturbance (i.e., SD rats), but returned toward pre-injury levels with exercise (i.e., Ex rats) or combined sleep disturbance and exercise (i.e., SD+Ex rats). Post-intervention levels were higher in SD than Ex rats (Fig. 3A). Corticosterone was not increased post-injury relative to pre-injury, was markedly increased after the exercise intervention, but decreased or remained unchanged with sleep disturbance or combined sleep disturbance and exercise, respectively. Post-intervention levels were higher in Ex rats than the other groups (Fig. 3C). Estradiol was not increased post-injury relative to pre-injury, was maintained with sleep disturbance, trended upward with exercise, and increased significantly with combined sleep disturbance and exercise. Post-intervention levels were lower in SD than SD+Ex rats (Fig 3E). GAS-1 was not increased post-injury relative to pre-injury, was decreased after exposure to sleep disturbance, but was preserved with exercise and combined sleep disturbance and exercise. No significant group differences were detected post-intervention (Fig 3F). Neuropilin 2 was not increased post-injury relative to pre-injury, was markedly increased with sleep disturbance, but remained unchanged with exercise or combined sleep disturbance and exercise. Post-intervention levels were higher in SD rats than the other groups (Fig 3L). TIMP-1 was not increased post-injury relative to pre-injury, was markedly decreased with sleep disturbance, but increased with exercise and combined sleep disturbance and exercise. Post-intervention levels were lower in SD rats than the other groups (Fig 3Q).

A main effect for Time, but no interaction, was found for all other biomarkers. Most generally increased in response to injury (CINC-1, eotaxin, IFNγ, IL-1β, LIX, MCP-1, RANTES, and TNFα; Fig. 3B, D, H, I, J, K, N, R; noting that BDNF also increased in response to injury, Fig. 3A). These biomarkers either decreased (CINC-1, eotaxin, IFNγ, IL-1β; Fig. 3B, D, H, I), persisted (LIX, MCP-1, TNFα; Fig. 3 J, K, R), or further increased (RANTES; Fig. 3N) post-intervention, regardless of group. Although no interaction or main effect for Group was detected for CINC-1, post-intervention decreases were largely attributable to Ex (Fig. 3B). Some biomarkers did not respond to injury yet increased (HGF and TIM-1; Fig. 3G, Q) or decreased (Notch-2 and TCK-1; Fig. 3M, O) following the intervention. A main effect for Group was also observed for Notch-2, which was elevated generally in Ex rats relative to other groups.

Supplemental Figure 3C–E includes longitudinal BDNF, corticosterone, and TNFα data for Normal Control and HRHF+Rest animals. Significant Group × Time interactions were observed for BDNF and TNFα (Supplemental Figure 3C, E). In Normal Control (i.e., uninjured) rats, these biomarkers did not differ from other groups (SD, Ex, SD+Ex, HRHF+Rest) at baseline (pre-injury), showed no change over time, and remained lower than all groups (SD, Ex, and SD+Ex significantly, p<0.019; HRHF+Rest non-significantly p=0.119) at 4 weeks (equivalent post-injury time-point). At 8 weeks (equivalent post-intervention time-point), BDNF was lower relative to SD rats (p=0.005) and TNFα was lower relative to all other groups (p<0.001). In HRHF+Rest (i.e., injury only) rats, BDNF and TNFα did not differ from other groups at baseline, increased in response to injury (BDNF non-significantly, p=0.156; TNFα significantly, p<0.001), and were maintained or trended toward pre-injury levels following 4 weeks of rest (i.e., post-intervention). At 8 weeks (i.e., after the injury and rest period; equivalent post-intervention time-point), BDNF was lower than in SD rats (p=0.049) whereas TNFα remained elevated compared to Normal Control rats (p=0.002). For corticosterone, no significant interaction or main effects were observed, with levels remaining stable across all timepoints.

3.5. Several systemic biomarkers correlated with pain-related behaviours

Correlation analyses between biomarkers and pain-related behaviours (aggregated across the three time-points) revealed several significant relationships (Table 1). Lower grip strength correlated moderately with higher levels of BDNF, CINC-1, eotaxin, IFNγ, IL-1β, Neuropilin-2, and TNFα. Percent improvements in grip strength correlated moderately and positively with corticosterone, RANTES and TIM-1, and lower levels of IFNγ and TCK-1; i.e., higher levels of the biomarkers related to greater improvement in grip strength. Greater forepaw mechanical sensitivity to 1cN and/or 4cN sized monofilaments correlated moderately with higher levels of BDNF, eotaxin, HGF, and TNFα. Additional weak correlations (r <0.3) are presented in italicized text in Table 1.

Table 1.

Correlations between systemic biomarkers and behavioural outcomes, all timepoints.

Biomarker Maximum grip strength (all timepoints) a (XY pairs=75) % Change in grip strength from pre-injury a (XY pairs=49) 1 cN monofilament (all timepoints) b (XY pairs=75) 4 cN monofilament (all timepoints) b (XY pairs=75)
BDNF (pg/ml) r= −0.30, p<0.01 r= −0.18, p=0.11 r=0.34, p<0.01 r=0.50, p<0.01
CINC-1 (pg/ml) r= −0.30, p<0.01 r= −0.08, p=0.29 r=0.18, p=0.05 r=0.23, p=0.02
Corticosterone (ng/ml) r=0.24, p=0.02 r=0.37, p<0.01 r=0.11, p=0.18 r=0.19, p=0.05
Eotaxin (pg/ml) r= −0.34, p<0.01 r= −0.24, p=0.04 r=0.21, p=0.03 r=0.30, p=0.01
Estradiol (pg/ml) r=0.14, p=0.12 r= 0.24, p=0.04 r=0.10, p=0.20 r= −0.04, p=0.36
GAS-1 r= −0.04, p=0.36 r=0.03, p=0.42 r=0.03, p=0.38 r= 0.11, p=0.16
HGF (pg/ml) r= −0.27, p<0.01 r= −0.05, p=0.39 r=0.35, p<0.01 r=0.33, p<0.01
IFNγ (pg/ml) r= −0.31, p<0.01 r= −0.45, p<0.01 r=0.10, p=0.19 r=0.23, p=0.02
IL-1β (pg/ml) r= −0.30, p<0.01 r= −0.16, p=0.13 r=0.04, p=0.35 r=0.06, p=0.28
LIX r= −0.16, p=0.08 r= −0.17, p=0.13 r=0.17, p=0.07 r= 0.23, p=0.02
MCP-1 (pg/ml) r= −0.16, p=0.08 r=0.07, p=0.31 r=0.21, p=0.03 r=0.15, p=0.10
Neuropilin-2 (pg/ml) r= −0.30, p<0.01 r= −0.25, p=0.04 r=0.00, p=0.21 r=0.11, p=0.17
NOTCH-2 (pg/ml) r= −0.24, p=0.02 r= −0.20, p=0.09 r= −0.03, p=0.39 r=0.03, p=0.41
RANTES (pg/ml) r= −0.03, p=0.39 r=0.45, p<0.01 r=0.25, p=0.01 r=0.07, p=0.28
TCK-1 (pg/ml) r= −0.14, p=0.11 r= −0.30, p=0.02 r= −0.04, p=0.48 r= −0.03, p=0.39
TIM-1 (pg/ml) r=0.28, p<0.01 r=0.41, p<0.01 r=0.15, p=0.09 r= −0.11, p=0.18
TIMP-1 r= −0.08, p=0.22 r= −0.03, p=0.41 r=0.15, p=0.10 r= 0.15, p=0.09
TNF-α (pg/ml) r= −0.34, p<0.01 r=0.19, p=0.10 r=0.32, p<0.03 r=0.32, p<0.03
a

Pearson’s correlation

b

Spearman’s correlation

Moderate to strong correlations are in bold; Weak correlations are in italics.

3.6. Intervention-specific neuroimmune response profiles captured by cluster analysis

Clustering of post-intervention systemic biomarker data revealed three clusters with distinct neuroimmune profiles (Fig. 4), termed: “low inflammatory” (Cluster 1), “high CINC-1, IFNγ & IL-1β” (Cluster 2), and “high inflammatory” (Cluster 3). ANOVAs identified biomarkers that differentiated the clusters (Table 2) and post-hoc analyses identified between-cluster differences. Cluster 3 was characterised by higher BDNF and neuropilin-2 than other clusters (all p<0.047), higher TNFα than Cluster 2 (p=0.057) and lower TIMP-1 than Cluster 1 (p=0.018). Cluster 2 was characterised by higher CINC-1, IFNγ and IL-1β than other clusters (all p<0.035). In contrast, Cluster 1 showed the opposite pattern to Clusters 2 and 3, with significantly lower BNDF (vs. Cluster 3: p=0.047), neuropilin-2 (vs. Cluster 3: p=0.026), CINC-1 (vs. Cluster 2: p<0.001), IFNγ (vs. Cluster 2: p=0.031) and IL-1β (vs. Cluster 2: p>0.001). Cluster 1 also had higher estradiol than Cluster 3 (p=0.025).

Figure 4.

Figure 4.

Neuroimmune response profiles. Heatmap of normalized (Z-scores) systemic biomarker concentrations at the final (post-intervention) time-point in animals exposed to 4 weeks of sleep disturbance, exercise, or their combination after injury. Each column corresponds to an individual animal and each row to a specific biomarker. Red indicates higher expression and blue indicates lower expression. Unbiased hierarchical clustering identified four distinct subgroups (clusters), represented by the dendrogram at the top: “low inflammatory” (Cluster 1), “high CINC-1, IFNγ & IL-1β” (Cluster 2), and “high inflammatory” (Cluster 3). The proportions (%) of animals from each intervention group (sleep disturbance, SD; exercise, Ex, combined sleep disturbance and exercise, SD+Ex) within the three clusters are shown in the pie charts at the bottom. Biomarker abbreviations and basic functions are defined in Supplemental Table 1.

Table 2.

Post-intervention systemic concentrations (Means, SEM) of each biomarker used in the cluster analysis and compared across the three clusters.

Biomarker Cluster 1 Cluster 2 Cluster 3 P-value
Mean SEM Mean SEM Mean SEM
BDNF (pg/ml) 606.67 70.51 527.76 113.15 1014.98 165.84 0.020
CINC-1 (pg/ml) 8.59 3.54 60.79 13.93 25.52 6.31 <0.001
Corticosterone (ng/ml) 75.90 25.87 44.51 22.60 22.92 3.95 0.190
Eotaxin (pg/ml) 395.21 22.02 374.00 20.29 409.23 102.62 0.927
Estradiol (pg/ml) 7.29 1.22 4.47 0.66 3.77 0.43 0.022
GAS-1 (pg/ml) 480.98 36.50 411.83 28.04 1088.28 702.10 0.451
HGF (pg/ml) 7172.96 338.23 5790.24 423.57 13140.31 6688.04 0.388
IFNγ (pg/ml) 0.05 0.05 5.09 2.58 0.00 0.00 0.019
IL-1β (pg/ml) 26.62 11.75 128.86 22.72 25.45 10.00 <0.001
LIX (pg/ml) 623.29 20.97 556.23 53.32 593.16 53.20 0.558
MCP-1 (pg/ml) 454.80 24.62 443.43 40.38 366.41 36.16 0.144
Neuropilin-2 (pg/ml) 43.33 36.62 36.06 33.46 288.91 97.08 0.015
Notch-2 (pg/ml) 2519.48 202.48 2235.12 341.13 1921.76 218.59 0.244
RANTES (pg/ml) 148.79 6.81 152.13 11.32 154.84 12.04 0.908
TCK-1 (pg/ml) 3388.46 442.69 2840.94 271.90 3424.04 290.71 0.481
TIM-1 (pg/ml) 71.10 6.39 73.37 25.32 53.56 10.10 0.587
TIMP-1 (pg/ml) 1732.03 62.34 1410.53 157.09 1303.70 93.29 0.018
TNFα (pg/ml) 174.17 2.58 171.44 2.26 183.51 4.83 0.054

Biomarkers were compared across all three clusters using 1-way ANOVAs. Refer to Supplementary Table 1 for biomarker abbreviation definitions. Significant values are in bold. Near significant values are in italics.

Clusters differed in the distribution of animals from each intervention groups (Pearson χ2, p=0.005; Fig. 4): Cluster 3 contained the highest proportion of SD animals (78%), with only one SD+Ex animal and no Ex animal. Conversely, Cluster 1 included no SD animals, but the highest proportion of SD+Ex (71%) and Ex (50%) animals. Cluster 2 comprised a mix of SD (22%), SD+Ex (29%), and Ex (38%) animals.

Although post-intervention pain-related behaviours did not significantly differ between clusters (Table 3), Cluster 3 showed the least improvement in grip strength (percent change from pre-injury), the lowest maximum grip strength, and greater mechanical sensitivity to 1 cN and 4 cN monofilament probing.

Table 3.

Post-intervention behavioral measures (Mean, SEM) compared between the three clusters.

Biomarker Cluster 1 Cluster 2 Cluster 3 P-value
Mean SEM Mean SEM Mean SEM
% Change in grip strength 3.70 7.49 6.69 6.91 −11.92 6.42 0.165
Maximum grip strength 753.44 43.01 811.86 41.10 667.63 29.71 0.057
Number of withdrawals (out of 10) to 1 cN monofilament probing 3.00 0.97 1.43 0.81 3.38 0.92 0.335
Number of withdrawals (out of 10) to 4 cN monofilament probing 4.22 1.00 2.71 1.36 5.50 1.51 0.352

Behavioural measures were compared across all three clusters using 1-way ANOVAs. Near significant values are in italics.

4. Discussion

This study examined the interplay between sleep, exercise, systemic biomarkers of inflammation and neural plasticity, and pain-related behaviours before and after injury. Results revealed consistent yet contrasting findings. First, pain-related behaviours persisted post-injury with sleep disturbance alone but returned to or exceeded pre-injury levels with exercise, either alone or in combination with sleep disturbance. Interestingly, 4 weeks of rest post-injury, in place of a sleep and/or exercise intervention, also restored grip strength (though not above baseline as with exercise) but mechanical sensitivity remained impaired, albeit less severely than with sleep disturbance. Second, circulating biomarkers showed intervention-specific responses. Sleep disturbance increased BDNF and neuropilin-2, decreased TIMP-1 and GAS-1, and maintained estradiol. Conversely, exercise (alone or with sleep disturbance) elicited opposite responses, including increased estradiol. Notably, CINC-1 (decreased) and corticosterone (increased) responded only to exercise alone, suggesting that sleep disturbance may override and suppress their normal exercise-induced regulation. In contrast, rest alone post-injury failed to restore any of the three biomarkers (BDNF, TNFα, corticosterone) assessed in this group (HRHF-Rest). Third, cluster analysis of post-intervention biomarker data revealed two intervention-specific neuroimmune profiles: one linked to sleep disturbance alone (Cluster 3: high BDNF, neuropilin-2, TNFα; low TIMP-1, estradiol) and the other to exercise alone (Cluster 1: low BDNF, neuropilin-2, CINC-1; high TIMP-1, estradiol). Fourth, several of these biomarkers correlated with pain-related behaviours. Together, these findings support the hypothesis that post-injury sleep disturbance promotes persistent pain via neuroimmune mechanisms, which are largely counteracted by exercise.

That 4 weeks of irregular, low-stress sleep disturbance impaired post-injury recovery (forepaw mechanical sensitivity and grip strength) corroborates and extends on our prior work [61]. Rodents exposed solely to sleep disturbance continued to exhibit post-injury mechanical hypersensitive across a range of stimulus intensities (1 cN, 4 cN, 8 cN monofilament sizes) relative to baseline (pre-injury), and greater sensitivity (4 cN) relative to rodents exposed to exercise alone or combined with sleep disturbance. Comparable findings have been reported in pain-free humans. For example, next-day pain thresholds in response to a pressure stimulus are lower following one or two nights of partial sleep loss [28; 76; 86]. Moreover, sleep fragmentation and deprivation impairs endogenous pain modulatory pathways and enhances spontaneous pain, sensitivity to various stimuli (e.g., pressure and cold), and signs considered to imply peripheral and central sensitisation [21; 112]. Our data suggest these changes may be amplified and/or sustained post-injury, potentially facilitating the transition to chronic pain—a proposition supported by human cohort studies [45; 56; 58; 60; 62–65]. With respect to grip strength, failure to recover to pre-injury levels suggests sleep disturbance contributes to persistent and worsening functional deficits in the affected tissues (in this case, the neuromuscular, tendinous, and/or ligamentous components of the forelimb). Although few, if any, studies have examined the long-term effects of sleep on injured structures, poor sleep is known to weaken immunity [6; 22; 89] and impair healing [59]. Longitudinal human studies are required to validate our pre-clinical insights and explore the intriguing, yet untested, possibility that sleep-focused interventions could improve both pain and function.

The extent to which aerobic exercise countered the negative impact of sleep disturbance on sensorimotor function highlights its potential to prevent and manage persistent pain in those unable to improve their sleep, such as shift workers or individuals with treatment-resistant sleep disorders. Like “healthy” sleep, aerobic exercise provides wide-ranging biological, psychological, social, and behavioural benefits [91; 113; 114]. At the neuroimmune level, both are thought to reduce inflammation [37; 75; 102], promoting: (1) adaptive changes within the peripheral and central nervous systems that lower pain [30; 59; 61], and (2) improvements in musculoskeletal tissues that promote recovery and enhance functional capacity [51; 59]. Conversely, our data suggest that sleep disturbance, as opposed to normal/healthy sleep (i.e., rest only after injury), drives opposing neuroimmune effects that exercise may offset or even reverse via shared pathways. Clinically, exercise is a first-line treatment for persistent pain [36; 40], and when combined with sleep-focused interventions (for those with comorbid sleep impairments), may offer additive or synergistic benefits for pain prevention/management. Confirmation of these finding could have important clinical implications, particularly given that sleep disturbance co-occurs with chronic pain in up to 88% of cases [1; 30; 79]. Importantly, exercise more effectively restored sensorimotor and neuroimmune profiles than rest alone, which failed to normalise post-injury impairments, consistent with previous findings [12; 117].

Physiological insight into how sleep affects pain and musculoskeletal function—and how exercise might mitigate these effects—can be drawn from the examined biomarkers. The sustained increase in systemic BDNF in response to sleep disturbance (without exercise) likely reflects altered neuronal activity and a heightened inflammatory milieu. BDNF both influences and is influenced by sleep architecture and homeostatic sleep drive, and changes in sleep are consistently associated with changes in circulating BDNF [27; 34; 92]. Although ~75% of BDNF is produced within the central nervous system, systemic levels closely correlate due to BDNF’s capacity to cross the blood–brain barrier [71]. Importantly, BDNF is also upregulated in response to inflammation [82; 118], which is relevant given the pro-inflammatory consequences of overuse injury [5; 10; 31; 32; 109] and sleep disturbance [50; 59; 80]. This is consistent with the current findings showing increased levels of pro-inflammatory proteins (e.g., neuropilin-2) and decreased levels of anti-inflammatory proteins (e.g., TIMP-1) with sleep disturbance alone. In contrast, exercise—irrespective of sleep—tended to elicit the opposite profile. This divergence corresponded with sensorimotor outcomes, with higher BDNF and pro-inflammatory protein levels correlating with worse pain-related behaviours. Given BDNF’s role in synaptic plasticity, excessive and/or prolonged elevations, especially under persisting inflammatory conditions, may drive maladaptive neuroplasticity underlying central sensitisation and chronic pain development [61; 82].

Particularly interesting was the contrasting effects of sleep disturbance and exercise (with or without sleep disturbance) on TIMP-1. TIMP-1 is a multi-functional glycoprotein with emerging key roles in the suppression and resolution of inflammation, sensitisation and pain. Supporting our data, low TIMP-1 levels have been linked with excessive inflammation, hypersensitivity, and the onset of chronic pain after injury [66; 103]. Prior work also supports the potential for exercise to upregulate TIMP-1 expression [73]. Other inflammation-modulating biomarkers that correlated with sensorimotor function—such as neuropilin 2 (pro-inflammatory [97]) and estradiol (anti-inflammatory/analgesic [88; 115])—were also oppositely influenced by sleep disturbance (increased and supressed, respectively) and exercise (with/without sleep disturbance). These findings highlight anti-inflammatory pathways through which aerobic exercise may counteract sleep-induced pain. However, CINC-1 (decreased) and corticosterone (increased) responded exclusively to exercise alone, suggesting sleep disturbance may supress their normal exercise-induced response. CINC-1, analogous to human IL-8, supports early tissue repair via recruiting neutrophils [77; 119], but can promote hyper-inflammation and persistent pain when levels persist excessively [55; 69; 77; 96]. Corticosterone, analogous to human cortisol, dampens inflammation [100; 120], though excessive suppression early post-injury can hinder healing [26; 38; 87; 111]. Although both biomarkers are modulated by exercise [18; 25; 46; 48; 81], our data suggest sleep disturbance exerts dominant regulatory effects, potentially disrupting corticosterone’s role in inflammatory resolution and tissue repair.

Cluster analyses reinforced and extended these neuroimmune profiles associated with sleep disturbance and exercise. Cluster 3, characterised by high BDNF, neuropilin-2, and TNFα, and low TIMP-1, comprised only sleep-disturbed animals, except one that also exercised (SD+Ex). This profile mirrors individual biomarker findings for the sleep disturbance group, with the exception of TNFα, which did not differ at the group level across interventions. Its elevation in this cluster suggests that sleep disturbance enhances TNFα for many, but not all. TNF is a potent pro-inflammatory cytokine, and its dysregulation promotes chronic inflammation and impairs healing [33; 54; 78]. It also sensitises nociceptors and promotes central sensitisation through neuronal, glial, and immune mechanisms [2; 14; 15; 47; 68; 93; 95; 122]. These processes are implicated in the transition from acute to chronic pain [52; 70] and likely contribute to the poor recovery observed in Cluster 3.

In contrast, the most favourable recovery profile (Cluster 1) featured low BDNF, neuropilin-2, CINC-1, IFNγ, and IL-1β, and included only animals exposed to exercise (with or without sleep disturbance). This low-BDNF/anti-inflammatory profile aligns with group/intervention level findings and suggests exercise dampens adverse neuroimmune responses to injury and sleep disturbance. That IFNγ and IL-1β—pro-inflammatory cytokines that activate nociceptors and promote pain-sensitising mediators [29; 94; 117; 121]—were low in this cluster but did not differ between intervention groups likely reflects individual variability in inflammatory responsiveness to exercise. Cluster 2, characterised by elevated CINC-1, IFNγ and IL-1β, comprised animals from all three intervention groups, suggesting inherent neuroimmune reactivity independent of sleep or exercise, potentially driven by biological predisposition and/or variability in injury response.

This study has some limitations. First, only female rats were included due to biomechanical constraints of the injury model, as detailed and addressed in Section 2.2 and [61]. Second, the 4-week post-injury follow-up period is insufficient to fully capture the transition from acute to chronic pain. Extended observation periods are needed to characterise the long-term effects of sleep disturbance on pain. Third, nervous system tissues were not examined to confirm peripheral or central sensitization. These limitations should be addressed in future studies to validate the role of sleep in the development and maintenance of chronic pain.

5. Conclusion

Findings reveal that post-injury sleep disturbance triggers neuroimmune changes that increase persistent pain risk, whereas aerobic exercise mitigates these effects, highlighting the interactive role of sleep and exercise in shaping pain outcomes and their potential as lifestyle-based strategies for preventing and managing chronic pain.

Supplementary Material

Supplemental Tables
Supplemental FIgures 1-10 and Supplemental Table 4

Acknowledgements

This research was funded by the Assistant Secretary of Defense for Health Affairs endorsed by the U.S. Department of Defense through the FY19 Chronic Pain Management Research Program (DMK, Grant ID: W81XWH2010909), National Health and Medical Research Council (NHMRC) of Australia (DMK, Grant ID: 2027008; PWH, Grant ID: 1194937), and U.S. National Institutes of Health (NIH) National Institute of Arthritis and Musculoskeletal and Skin Diseases (NIAMS) (MFB, Grant ID: RO1AR056019).

Footnotes

Conflict of Interest Statement

The authors have no conflicts of interest to declare.

Data Availability

All study data included in the article are provided in Tables 1–3 or in Supplemental Tables 1–3. Study data that are not included in supplementary files will be provided upon request.

References

  • [1].Andersen ML, Araujo P, Frange C, Tufik S. Sleep Disturbance and Pain: A Tale of Two Common Problems. Chest 2018;154(5):1249–1259. [DOI] [PubMed] [Google Scholar]
  • [2].Andrade P, Visser-Vandewalle V, Hoffmann C, Steinbusch HW, Daemen MA, Hoogland G. Role of TNF-alpha during central sensitization in preclinical studies. Neurol Sci 2011;32(5):757–771. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [3].Barbe MF, Amin M, Gingery A, Lambi AG, Popoff SN. Blocking CCN2 preferentially inhibits osteoclastogenesis induced by repetitive high force bone loading. Connect Tissue Res 2021;62(1):115–132. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [4].Barbe MF, Chen FL, Loomis RH, Harris MY, Kim BM, Xie K, Hilliard BA, McGonagle ER, Bailey TD, Gares RP, Van Der Bas M, Kalicharan BA, Holt-Bright L, Stone LS, Hodges PW, Klyne DM. Characterization of pain-related behaviors in a rat model of acute-to-chronic low back pain: single vs. multi-level disc injury. Front Pain Res (Lausanne) 2024;5:1394017. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [5].Barbe MF, Elliott MB, Abdelmagid SM, Amin M, Popoff SN, Safadi FF, Barr AE. Serum and tissue cytokines and chemokines increase with repetitive upper extremity tasks. J Orthop Res 2008;26(10):1320–1326. [DOI] [PubMed] [Google Scholar]
  • [6].Barbe MF, Gallagher S, Massicotte VS, Tytell M, Popoff SN, Barr-Gillespie AE. The interaction of force and repetition on musculoskeletal and neural tissue responses and sensorimotor behavior in a rat model of work-related musculoskeletal disorders. BMC Musculoskelet Disord 2013;14:303. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [7].Barbe MF, Hilliard BA, Amin M, Harris MY, Hobson LJ, Cruz GE, Dorotan JT, Paul RW, Klyne DM, Popoff SN. Blocking CTGF/CCN2 reverses neural fibrosis and sensorimotor declines in a rat model of overuse-induced median mononeuropathy. J Orthop Res 2020;38(11):2396–2408. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [8].Barbe MF, Hilliard BA, Amin M, Harris MY, Hobson LJ, Cruz GE, Popoff SN. Blocking CTGF/CCN2 reduces established skeletal muscle fibrosis in a rat model of overuse injury. FASEB J 2020;34(5):6554–6569. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [9].Barbe MF, Hilliard BA, Delany SP, Iannarone VJ, Harris MY, Amin M, Cruz GE, Barreto-Cruz Y, Tran N, Day EP, Hobson LJ, Assari S, Popoff SN. Blocking CCN2 Reduces Progression of Sensorimotor Declines and Fibrosis in a Rat Model of Chronic Repetitive Overuse. J Orthop Res 2019;37(9):2004–2018. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [10].Barbe MF, Lambi AG, Harris MY, Patel PR, Tamas IP, McGonagle ER, Van Der Bas M, Kalicharan BA, Bright-Rowe L, Popoff SN, Klyne DM. Overuse injury induces persistent behavioral declines that correlate with higher IL-6 expression in the affected musculoskeletal tissues, circulation, and brain. Front Physiol 2025;16:1500795. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [11].Barbe MF, Massicotte VS, Assari S, Monroy MA, Frara N, Harris MY, Amin M, King T, Cruz GE, Popoff SN. Prolonged high force high repetition pulling induces osteocyte apoptosis and trabecular bone loss in distal radius, while low force high repetition pulling induces bone anabolism. Bone 2018;110:267–283. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [12].Barbe MF, Panibatla ST, Harris MY, Amin M, Dorotan JT, Cruz GE, Bove GM. Manual Therapy With Rest as a Treatment for Established Inflammation and Fibrosis in a Rat Model of Repetitive Strain Injury. Front Physiol 2021;12:755923. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [13].Blyth FM, March LM, Brnabic AJ, Jorm LR, Williamson M, Cousins MJ. Chronic pain in Australia: a prevalence study. Pain 2001;89(2–3):127–134. [DOI] [PubMed] [Google Scholar]
  • [14].Boettger MK, Hensellek S, Richter F, Gajda M, Stockigt R, von Banchet GS, Brauer R, Schaible HG. Antinociceptive effects of tumor necrosis factor alpha neutralization in a rat model of antigen-induced arthritis: evidence of a neuronal target. Arthritis Rheum 2008;58(8):2368–2378. [DOI] [PubMed] [Google Scholar]
  • [15].Boettger MK, Weber K, Grossmann D, Gajda M, Bauer R, Bar KJ, Schulz S, Voss A, Geis C, Brauer R, Schaible HG. Spinal tumor necrosis factor alpha neutralization reduces peripheral inflammation and hyperalgesia and suppresses autonomic responses in experimental arthritis: a role for spinal tumor necrosis factor alpha during induction and maintenance of peripheral inflammation. Arthritis Rheum 2010;62(5):1308–1318. [DOI] [PubMed] [Google Scholar]
  • [16].Bove GM, Delany SP, Hobson L, Cruz GE, Harris MY, Amin M, Chapelle SL, Barbe MF. Manual therapy prevents onset of nociceptor activity, sensorimotor dysfunction, and neural fibrosis induced by a volitional repetitive task. Pain 2019;160(3):632–644. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [17].Breivik H, Collett B, Ventafridda V, Cohen R, Gallacher D. Survey of chronic pain in Europe: prevalence, impact on daily life, and treatment. Eur J Pain 2006;10(4):287–333. [DOI] [PubMed] [Google Scholar]
  • [18].Caplin A, Chen FS, Beauchamp MR, Puterman E. The effects of exercise intensity on the cortisol response to a subsequent acute psychosocial stressor. Psychoneuroendocrinology 2021;131:105336. [DOI] [PubMed] [Google Scholar]
  • [19].Carp SJ, Barbe MF, Winter KA, Amin M, Barr AE. Inflammatory biomarkers increase with severity of upper-extremity overuse disorders. Clin Sci (Lond) 2007;112(5):305–314. [DOI] [PubMed] [Google Scholar]
  • [20].Cerqueira E, Marinho DA, Neiva HP, Lourenco O. Inflammatory Effects of High and Moderate Intensity Exercise-A Systematic Review. Front Physiol 2019;10:1550. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [21].Chang JR, Fu SN, Li X, Li SX, Wang X, Zhou Z, Pinto SM, Samartzis D, Karppinen J, Wong AY. The differential effects of sleep deprivation on pain perception in individuals with or without chronic pain: A systematic review and meta-analysis. Sleep Med Rev 2022;66:101695. [DOI] [PubMed] [Google Scholar]
  • [22].Clark BD, Al-Shatti TA, Barr AE, Amin M, Barbe MF. Performance of a high-repetition, high-force task induces carpal tunnel syndrome in rats. J Orthop Sports Phys Ther 2004;34(5):244–253. [DOI] [PubMed] [Google Scholar]
  • [23].Costa N, Smits E, Kasza J, Salomoni S, Ferreira M, Sullivan M, Hodges PW. ISSLS PRIZE IN CLINICAL SCIENCE 2021: What are the risk factors for low back pain flares and does this depend on how flare is defined? Eur Spine J 2021;30(5):1089–1097. [DOI] [PubMed] [Google Scholar]
  • [24].Crooks E, Pavlenco C, Miller L, Wambold L, Schroeder S, Weeks DL. Non-cardiac Pain Prevalence, Pain Intensity, Pain Interference, Sleep, and Cardiopulmonary Function in Outpatient Cardiac Rehabilitation. Arch Phys Med Rehabil 2025. [DOI] [PubMed] [Google Scholar]
  • [25].Docherty S, Harley R, McAuley JJ, Crowe LAN, Pedret C, Kirwan PD, Siebert S, Millar NL. The effect of exercise on cytokines: implications for musculoskeletal health: a narrative review. BMC Sports Sci Med Rehabil 2022;14(1):5. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [26].Ebrecht M, Hextall J, Kirtley L-G, Taylor A, Dyson M, Weinman J. Perceived stress and cortisol levels predict speed of wound healing in healthy male adults. Psychoneuroendocrinology 2004;29(6):798–809. [DOI] [PubMed] [Google Scholar]
  • [27].ElGrawani W, Sun G, Kliem FP, Sennhauser S, Pierre-Ferrer S, Rosi-Andersen A, Boccalaro I, Bethge P, Heo WD, Helmchen F, Adamantidis AR, Forger DB, Robles MS, Brown SA. BDNF-TrkB signaling orchestrates the buildup process of local sleep. Cell Reports 2024;43(7):114500. [DOI] [PubMed] [Google Scholar]
  • [28].Faraut B, Léger D, Medkour T, Dubois A, Bayon V, Chennaoui M, Perrot S. Napping reverses increased pain sensitivity due to sleep restriction. PLoS One 2015;10(2):e0117425. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [29].Ferrara V, Toti A, Ghelardini C, Mannelli LDC. Interferon-gamma and neuropathy: balance between pain and neuroprotection. Neural Regeneration Research 2022;17(12). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [30].Finan PH, Goodin BR, Smith MT. The association of sleep and pain: an update and a path forward. J Pain 2013;14(12):1539–1552. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [31].Fisher PW, Zhao Y, Rico MC, Massicotte VS, Wade CK, Litvin J, Bove GM, Popoff SN, Barbe MF. Increased CCN2, substance P and tissue fibrosis are associated with sensorimotor declines in a rat model of repetitive overuse injury. J Cell Commun Signal 2015;9(1):37–54. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [32].Gao HG, Fisher PW, Lambi AG, Wade CK, Barr-Gillespie AE, Popoff SN, Barbe MF. Increased serum and musculotendinous fibrogenic proteins following persistent low-grade inflammation in a rat model of long-term upper extremity overuse. PLoS One 2013;8(8):e71875. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [33].Garcia-Dominguez M The Role of TNF-alpha in Neuropathic Pain: An Immunotherapeutic Perspective. Life (Basel) 2025;15(5). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [34].Garner JM, Chambers J, Barnes AK, Datta S. Changes in Brain-Derived Neurotrophic Factor Expression Influence Sleep-Wake Activity and Homeostatic Regulation of Rapid Eye Movement Sleep. Sleep 2018;41(2). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [35].Geneen LJ, Moore RA, Clarke C, Martin D, Colvin LA, Smith BH. Physical activity and exercise for chronic pain in adults: an overview of Cochrane Reviews. Cochrane Db Syst Rev 2017(4). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [36].Geneen LJ, Moore RA, Clarke C, Martin D, Colvin LA, Smith BH. Physical activity and exercise for chronic pain in adults: an overview of Cochrane Reviews. Cochrane Database Syst Rev 2017;4(4):CD011279. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [37].Gleeson M, Bishop NC, Stensel DJ, Lindley MR, Mastana SS, Nimmo MA. The anti-inflammatory effects of exercise: mechanisms and implications for the prevention and treatment of disease. Nat Rev Immunol 2011;11(9):607–615. [DOI] [PubMed] [Google Scholar]
  • [38].Godbout JP, Glaser R. Stress-Induced Immune Dysregulation: Implications for Wound Healing, Infectious Disease and Cancer. Journal of Neuroimmune Pharmacology 2006;1(4):421–427. [DOI] [PubMed] [Google Scholar]
  • [39].Gold JE, Hallman DM, Hellstrom F, Bjorklund M, Crenshaw AG, Djupsjobacka M, Heiden M, Mathiassen SE, Piligian G, Barbe MF. Systematic review of biochemical biomarkers for neck and upper-extremity musculoskeletal disorders. Scand J Work Environ Health 2016;42(2):103–124. [DOI] [PubMed] [Google Scholar]
  • [40].Gordon R, Bloxham S. A Systematic Review of the Effects of Exercise and Physical Activity on Non-Specific Chronic Low Back Pain. Healthcare (Basel) 2016;4(2). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [41].Grundy D Principles and standards for reporting animal experiments in The Journal of Physiology and Experimental Physiology. J Physiol 2015;593(12):2547–2549. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [42].Haack M, Sanchez E, Mullington JM. Elevated inflammatory markers in response to prolonged sleep restriction are associated with increased pain experience in healthy volunteers. Sleep 2007;30(9):1145–1152. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [43].Haack M, Simpson N, Sethna N, Kaur S, Mullington J. Sleep deficiency and chronic pain: potential underlying mechanisms and clinical implications. Neuropsychopharmacology 2020;45(1):205–216. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [44].Hadrevi J, Barbe MF, Ortenblad N, Frandsen U, Boyle E, Lazar S, Sjogaard G, Sogaard K. Calcium Fluxes in Work-Related Muscle Disorder: Implications from a Rat Model. Biomed Res Int 2019;2019:5040818. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [45].Hall M, Allison K, Knox G, McManus F, Hinman RS, Bennell KL, Spiers L, De Silva AP, Klyne DM, Plinsinga ML, Costa RJS, Murphy NJ, Wu W, Dobson FL. Addition of aerobic physical activity to resistance exercise for hip osteoarthritis (PHOENIX): a randomised comparative effectiveness trial. Lancet Rheumatol 2025;7(5):e343–e354. [DOI] [PubMed] [Google Scholar]
  • [46].Hare BD, Beierle JA, Toufexis DJ, Hammack SE, Falls WA. Exercise-associated changes in the corticosterone response to acute restraint stress: evidence for increased adrenal sensitivity and reduced corticosterone response duration. Neuropsychopharmacology 2014;39(5):1262–1269. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [47].Hess A, Axmann R, Rech J, Finzel S, Heindl C, Kreitz S, Sergeeva M, Saake M, Garcia M, Kollias G, Straub RH, Sporns O, Doerfler A, Brune K, Schett G. Blockade of TNF-alpha rapidly inhibits pain responses in the central nervous system. Proc Natl Acad Sci U S A 2011;108(9):3731–3736. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [48].Hill EE, Zack E, Battaglini C, Viru M, Viru A, Hackney AC. Exercise and circulating cortisol levels: the intensity threshold effect. J Endocrinol Invest 2008;31(7):587–591. [DOI] [PubMed] [Google Scholar]
  • [49].Irwin MR, Carrillo C, Olmstead R. Sleep loss activates cellular markers of inflammation: sex differences. Brain Behav Immun 2010;24(1):54–57. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [50].Irwin MR, Olmstead R, Carroll JE. Sleep Disturbance, Sleep Duration, and Inflammation: A Systematic Review and Meta-Analysis of Cohort Studies and Experimental Sleep Deprivation. Biol Psychiatry 2016;80(1):40–52. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [51].James G, Klyne DM, Millecamps M, Stone LS, Hodges PW. ISSLS Prize in Basic science 2019: Physical activity attenuates fibrotic alterations to the multifidus muscle associated with intervertebral disc degeneration. Eur Spine J 2019;28(5):893–904. [DOI] [PubMed] [Google Scholar]
  • [52].Ji RR, Nackley A, Huh Y, Terrando N, Maixner W. Neuroinflammation and Central Sensitization in Chronic and Widespread Pain. Anesthesiology 2018;129(2):343–366. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [53].Jiang X, Tang L, Zhang Y, Bai Y, Luo H, Wang R, Bi X, Chen R, Wang X. Does sedentary time and physical activity predict chronic back pain and morphological brain changes? A UK biobank cohort study in 33,402 participants. BMC Public Health 2024;24(1):2685. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [54].Kaye AD, Perilloux DM, Hawkins AM, Wester GC, Ragaland AR, Hebert SV, Kim J, Heisler M, Kelkar RA, Chami AA, Shekoohi S, Kaye AM. Tumor Necrosis Factor and Interleukin Modulators for Pathologic Pain States: A Narrative Review. Pain Ther 2024;13(3):481–493. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [55].Khan J, Hassun H, Zusman T, Korczeniewska O, Eliav E. Interleukin-8 levels in rat models of nerve damage and neuropathic pain. Neurosci Lett 2017;657:106–112. [DOI] [PubMed] [Google Scholar]
  • [56].Klyne D, Barbe M, Van Den Hoorn W, Hodges P. Longitudinal analysis of inflammatory, psychological, and sleep-related factors following an acute low back pain episode: A potential factor in tissue effects in low back pain. Journal of Bodywork and Movement Therapies 2018;22(4):866. [DOI] [PubMed] [Google Scholar]
  • [57].Klyne DM, Barbe MF, Hodges PW. Systemic inflammatory profiles and their relationships with demographic, behavioural and clinical features in acute low back pain. Brain Behav Immun 2017;60:84–92. [DOI] [PubMed] [Google Scholar]
  • [58].Klyne DM, Barbe MF, Hodges PW. Relationship between systemic inflammation and recovery over 12 months after an acute episode of low back pain. Spine J 2022;22(2):214–225. [DOI] [PubMed] [Google Scholar]
  • [59].Klyne DM, Barbe MF, James G, Hodges PW. Does the Interaction between Local and Systemic Inflammation Provide a Link from Psychology and Lifestyle to Tissue Health in Musculoskeletal Conditions? Int J Mol Sci 2021;22(14). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [60].Klyne DM, Barbe MF, van den Hoorn W, Hodges PW. ISSLS PRIZE IN CLINICAL SCIENCE 2018: longitudinal analysis of inflammatory, psychological, and sleep-related factors following an acute low back pain episode-the good, the bad, and the ugly. Eur Spine J 2018;27(4):763–777. [DOI] [PubMed] [Google Scholar]
  • [61].Klyne DM, Hilliard BA, Harris MY, Amin M, Hall M, Besomi M, Mustafa S, Farrell SF, Rawashdeh O, Han FY, Hodges PW, Frara N, Barbe MF. Poor sleep versus exercise: A duel to decide whether pain resolves or persists after injury. Brain Behav Immun Health 2024;35:100714. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [62].Klyne DM, Hodges PW. Circulating Adipokines in Predicting the Transition from Acute to Persistent Low Back Pain. Pain Med 2020;21(11):2975–2985. [DOI] [PubMed] [Google Scholar]
  • [63].Klyne DM, Moseley GL, Sterling M, Barbe MF, Hodges PW. Individual Variation in Pain Sensitivity and Conditioned Pain Modulation in Acute Low Back Pain: Effect of Stimulus Type, Sleep, and Psychological and Lifestyle Factors. J Pain 2018;19(8):942 e941–942 e918. [DOI] [PubMed] [Google Scholar]
  • [64].Klyne DM, Moseley GL, Sterling M, Barbe MF, Hodges PW. Are Signs of Central Sensitization in Acute Low Back Pain a Precursor to Poor Outcome? J Pain 2019;20(8):994–1009. [DOI] [PubMed] [Google Scholar]
  • [65].Klyne DM, van den Hoorn W, Barbe MF, Cholewicki J, L MH, Khan A, Meroni R, Moseley GL, Nicholas M, O’Sullivan L, Park R, Russell G, Sterling M, Hodges PW. Cohort profile: why do people keep hurting their back? BMC Res Notes 2020;13(1):538. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [66].Knight BE, Kozlowski N, Havelin J, King T, Crocker SJ, Young EE, Baumbauer KM. TIMP-1 Attenuates the Development of Inflammatory Pain Through MMP-Dependent and Receptor-Mediated Cell Signaling Mechanisms. Front Mol Neurosci 2019;12:220. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [67].Kocak M, Sevgin O. The effect of exercise on menstrual symptoms: a randomized controlled trial. BMC Womens Health 2025;25(1):406. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [68].Konig C, Zharsky M, Moller C, Schaible HG, Ebersberger A. Involvement of peripheral and spinal tumor necrosis factor alpha in spinal cord hyperexcitability during knee joint inflammation in rats. Arthritis Rheumatol 2014;66(3):599–609. [DOI] [PubMed] [Google Scholar]
  • [69].Krock E, Millecamps M, Anderson KM, Srivastava A, Reihsen TE, Hari P, Sun YR, Jang SH, Wilcox GL, Belani KG, Beebe DS, Ouellet J, Pinto MR, Kehl LJ, Haglund L, Stone LS. Interleukin-8 as a therapeutic target for chronic low back pain: Upregulation in human cerebrospinal fluid and pre-clinical validation with chronic reparixin in the SPARC-null mouse model. EBioMedicine 2019;43:487–500. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [70].Latremoliere A, Woolf CJ. Central Sensitization: A Generator of Pain Hypersensitivity by Central Neural Plasticity. Journal of Pain 2009;10(9):895–926. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [71].Li S, Weinstein G, Zare H, Teumer A, Volker U, Friedrich N, Knol MJ, Satizabal CL, Petyuk VA, Adams HHH, Launer LJ, Bennett DA, De Jager PL, Grabe HJ, Ikram MA, Gudnason V, Yang Q, Seshadri S. The genetics of circulating BDNF: towards understanding the role of BDNF in brain structure and function in middle and old ages. Brain Commun 2020;2(2). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [72].Lindell M, Grimby-Ekman A. Stress, non-restorative sleep, and physical inactivity as risk factors for chronic pain in young adults: A cohort study. PLoS One 2022;17(1):e0262601. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [73].Lo Presti R, Hopps E, Caimi G. Gelatinases and physical exercise: A systematic review of evidence from human studies. Medicine (Baltimore) 2017;96(37):e8072. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [74].Luo X, Li H, Hou X, Li L, Mao Y. The gender difference in the comorbidity of chronic pain and depression, a longitudinal study. J Affect Disord 2025:120109. [DOI] [PubMed] [Google Scholar]
  • [75].Mathur N, Pedersen BK. Exercise as a Mean to Control Low-Grade Systemic Inflammation. Mediat Inflamm 2008;2008. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [76].Matre D, Hu L, Viken LA, Hjelle IB, Wigemyr M, Knardahl S, Sand T, Nilsen KB. Experimental Sleep Restriction Facilitates Pain and Electrically Induced Cortical Responses. Sleep 2015;38(10):1607–1617. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [77].Matsushima K, Yang D, Oppenheim JJ. Interleukin-8: An evolving chemokine. Cytokine 2022;153:155828. [DOI] [PubMed] [Google Scholar]
  • [78].McDermott MF. TNF and TNFR biology in health and disease. Cell Mol Biol (Noisy-le-grand) 2001;47(4):619–635. [PubMed] [Google Scholar]
  • [79].Morin CM, LeBlanc M, Daley M, Gregoire JP, Merette C. Epidemiology of insomnia: prevalence, self-help treatments, consultations, and determinants of help-seeking behaviors. Sleep Med 2006;7(2):123–130. [DOI] [PubMed] [Google Scholar]
  • [80].Mullington JM, Simpson NS, Meier-Ewert HK, Haack M. Sleep loss and inflammation. Best Pract Res Clin Endocrinol Metab 2010;24(5):775–784. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [81].Nielsen AR, Pedersen BK. The biological roles of exercise-induced cytokines: IL-6, IL-8, and IL-15. Appl Physiol Nutr Metab 2007;32(5):833–839. [DOI] [PubMed] [Google Scholar]
  • [82].Nijs J, Meeus M, Versijpt J, Moens M, Bos I, Knaepen K, Meeusen R. Brain-derived neurotrophic factor as a driving force behind neuroplasticity in neuropathic and central sensitization pain: a new therapeutic target? Expert Opin Ther Tar 2015;19(4):565–576. [DOI] [PubMed] [Google Scholar]
  • [83].Núñez-Cortés R, Cruz-Montecinos C, López-Bueno R, Andersen LL, Calatayud J. Physical inactivity is the most important unhealthy lifestyle factor for pain severity in older adults with pain: A SHARE-based analysis of 27,528 cases from 28 countries. Musculoskeletal Science and Practice 2025;76:103270. [DOI] [PubMed] [Google Scholar]
  • [84].Occupational Safety and Health Administration. Ergonomics Program, 1999. [Google Scholar]
  • [85].Okezue OC, Anamezie TH, Nene JJ, Okwudili JD. Work-Related Musculoskeletal Disorders among Office Workers in Higher Education Institutions: A Cross-Sectional Study. Ethiop J Health Sci 2020;30(5):715–724. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [86].Onen SH, Alloui A, Gross A, Eschallier A, Dubray C. The effects of total sleep deprivation, selective sleep interruption and sleep recovery on pain tolerance thresholds in healthy subjects. J Sleep Res 2001;10(1):35–42. [DOI] [PubMed] [Google Scholar]
  • [87].Oxlund H, Fogdestam I, Viidik A. The influence of cortisol on wound healing of the skin and distant connective tissue response. Surg Gynecol Obstet 1979;148(6):876–880. [PubMed] [Google Scholar]
  • [88].Pang H, Chen S, Klyne DM, Harrich D, Ding W, Yang S, Han FY. Low back pain and osteoarthritis pain: a perspective of estrogen. Bone Res 2023;11(1):42. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [89].Patel PR, Tamas IP, Van Der Bas M, Kegg A, Hilliard BA, Lambi AG, Popoff SN, Barbe MF. Repetitive Overuse Injury Causes Entheseal Damage and Palmar Muscle Fibrosis in Older Rats. Int J Mol Sci 2024;25(24). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [90].Petrovsky N, McNair P, Harrison LC. Diurnal rhythms of pro-inflammatory cytokines: regulation by plasma cortisol and therapeutic implications. Cytokine 1998;10(4):307–312. [DOI] [PubMed] [Google Scholar]
  • [91].Posadzki P, Pieper D, Bajpai R, Makaruk H, Konsgen N, Neuhaus AL, Semwal M. Exercise/physical activity and health outcomes: an overview of Cochrane systematic reviews. BMC Public Health 2020;20(1):1724. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [92].Rahmani M, Rahmani F, Rezaei N. The Brain-Derived Neurotrophic Factor: Missing Link Between Sleep Deprivation, Insomnia, and Depression. Neurochem Res 2020;45(2):221–231. [DOI] [PubMed] [Google Scholar]
  • [93].Rani S, Barbe MF, Barr AE, Litivn J. Role of TNF alpha and PLF in bone remodeling in a rat model of repetitive reaching and grasping. J Cell Physiol 2010;225(1):152–167. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [94].Ren K, Torres R. Role of interleukin-1beta during pain and inflammation. Brain Res Rev 2009;60(1):57–64. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [95].Richter F, Natura G, Loser S, Schmidt K, Viisanen H, Schaible HG. Tumor necrosis factor causes persistent sensitization of joint nociceptors to mechanical stimuli in rats. Arthritis Rheum 2010;62(12):3806–3814. [DOI] [PubMed] [Google Scholar]
  • [96].Rodriguez-Pintó I, Agmon-Levin N, Howard A, Shoenfeld Y. Fibromyalgia and cytokines. Immunology Letters 2014;161(2):200–203. [DOI] [PubMed] [Google Scholar]
  • [97].Roy S, Bag AK, Singh RK, Talmadge JE, Batra SK, Datta K. Multifaceted Role of Neuropilins in the Immune System: Potential Targets for Immunotherapy. Front Immunol 2017;8:1228. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [98].Scheff JD, Calvano SE, Lowry SF, Androulakis IP. Modeling the influence of circadian rhythms on the acute inflammatory response. J Theor Biol 2010;264(3):1068–1076. [DOI] [PubMed] [Google Scholar]
  • [99].Schmitt K, Holsboer-Trachsler E, Eckert A. BDNF in sleep, insomnia, and sleep deprivation. Ann Med 2016;48(1–2):42–51. [DOI] [PubMed] [Google Scholar]
  • [100].Slavich GM, Irwin MR. From stress to inflammation and major depressive disorder: a social signal transduction theory of depression. Psychol Bull 2014;140(3):774–815. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [101].Smith MT, Haythornthwaite JA. How do sleep disturbance and chronic pain inter-relate? Insights from the longitudinal and cognitive-behavioral clinical trials literature. Sleep Med Rev 2004;8(2):119–132. [DOI] [PubMed] [Google Scholar]
  • [102].Smith TTG, Barr-Gillespie AE, Klyne DM, Harris MY, Amin M, Paul RW, Cruz GE, Zhao H, Gallagher S, Barbe MF. Forced treadmill running reduces systemic inflammation yet worsens upper limb discomfort in a rat model of work-related musculoskeletal disorders. BMC Musculoskelet Disord 2020;21(1):57. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [103].Stair R, Eller O, Young E, Baumbauer K. TIMP-1 As An Effector Of Ion Channel Expression And Inflammatory Pain. The Journal of Pain 2023;24(4, Supplement):11. [Google Scholar]
  • [104].Strazdins L, Bammer G. Women, work and musculoskeletal health. Social Science & Medicine 2004;58(6):997–1005. [DOI] [PubMed] [Google Scholar]
  • [105].Svensson M, Rosvall P, Boza-Serrano A, Andersson E, Lexell J, Deierborg T. Forced treadmill exercise can induce stress and increase neuronal damage in a mouse model of global cerebral ischemia. Neurobiol Stress 2016;5:8–18. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [106].Tan L, Cicuttini FM, Fairley J, Romero L, Estee M, Hussain SM, Urquhart DM. Does aerobic exercise effect pain sensitisation in individuals with musculoskeletal pain? A systematic review. BMC Musculoskelet Disord 2022;23(1):113. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [107].Thakkar B, Acevedo EO. BDNF as a biomarker for neuropathic pain: Consideration of mechanisms of action and associated measurement challenges. Brain Behav 2023;13(3):e2903. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [108].Tsang A, Von Korff M, Lee S, Alonso J, Karam E, Angermeyer MC, Borges GL, Bromet EJ, Demytteneare K, de Girolamo G, de Graaf R, Gureje O, Lepine JP, Haro JM, Levinson D, Oakley Browne MA, Posada-Villa J, Seedat S, Watanabe M. Common chronic pain conditions in developed and developing countries: gender and age differences and comorbidity with depression-anxiety disorders. J Pain 2008;9(10):883–891. [DOI] [PubMed] [Google Scholar]
  • [109].van Tulder M, Malmivaara A, Koes B. Repetitive strain injury. Lancet 2007;369(9575):1815–1822. [DOI] [PubMed] [Google Scholar]
  • [110].Vgontzas AN, Bixler EO, Lin HM, Prolo P, Trakada G, Chrousos GP. IL-6 and its circadian secretion in humans. Neuroimmunomodulation 2005;12(3):131–140. [DOI] [PubMed] [Google Scholar]
  • [111].Vileikyte L Stress and wound healing. Clinics in Dermatology 2007;25(1):49–55. [DOI] [PubMed] [Google Scholar]
  • [112].Vincenot M, Leonard G, Cloutier-Langevin C, Bordeleau M, Gendron L, Camirand Lemyre F, Marchand S. Exploring the Spectrum of Temporal Summation and Conditioned Pain Modulation Responses in Pain-Free Individuals Using a Tonic Heat Pain and Cold Pressor Test Paradigms. Eur J Pain 2025;29(5):e70019. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [113].Warburton DE, Nicol CW, Bredin SS. Health benefits of physical activity: the evidence. CMAJ 2006;174(6):801–809. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [114].Warburton DER, Bredin SSD. Health benefits of physical activity: a systematic review of current systematic reviews. Curr Opin Cardiol 2017;32(5):541–556. [DOI] [PubMed] [Google Scholar]
  • [115].Wise PM, Suzuki S, Brown CM. Estradiol: a hormone with diverse and contradictory neuroprotective actions. Dialogues Clin Neurosci 2009;11(3):297–303. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [116].Wu M, Song W, Zhang M, Teng L, Tang Q, Zhu L. Potential mechanisms of exercise for relieving inflammatory pain: a literature review of animal studies. Front Aging Neurosci 2024;16:1359455. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [117].Xin DL, Hadrévi J, Elliott ME, Amin M, Harris MY, Barr-Gillespie AE, Barbe MF. Effectiveness of conservative interventions for sickness and pain behaviors induced by a high repetition high force upper extremity task. BMC Neurosci 2017;18(1):36. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [118].Xiong HY, Hendrix J, Schabrun S, Wyns A, Campenhout JV, Nijs J, Polli A. The Role of the Brain-Derived Neurotrophic Factor in Chronic Pain: Links to Central Sensitization and Neuroinflammation. Biomolecules 2024;14(1). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [119].Yang L, Shi F, Cao F, Wang L, She J, He B, Xu X, Kong L, Cai B. Neutrophils in Tissue Injury and Repair: Molecular Mechanisms and Therapeutic Targets. MedComm (2020) 2025;6(5):e70184. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [120].Yeager MP, Pioli PA, Guyre PM. Cortisol exerts bi-phasic regulation of inflammation in humans. Dose Response 2011;9(3):332–347. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [121].Zhang J Yin and yang interplay of IFN-gamma in inflammation and autoimmune disease. J Clin Invest 2007;117(4):871–873. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [122].Zhang L, Berta T, Xu ZZ, Liu T, Park JY, Ji RR. TNF-alpha contributes to spinal cord synaptic plasticity and inflammatory pain: distinct role of TNF receptor subtypes 1 and 2. Pain 2011;152(2):419–427. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplemental Tables
Supplemental FIgures 1-10 and Supplemental Table 4

Data Availability Statement

All study data included in the article are provided in Tables 1–3 or in Supplemental Tables 1–3. Study data that are not included in supplementary files will be provided upon request.

RESOURCES