Aging‐associated decrease of PGC‐1α promotes pain chronification

Abstract Aging is generally associated with declining somatosensory function, which seems at odds with the high prevalence of chronic pain in older people. This discrepancy is partly related to the high prevalence of degenerative diseases such as osteoarthritis in older people. However, whether aging alters pain processing in the primary somatosensory cortex (S1), and if so, whether it promotes pain chronification is largely unknown. Herein, we report that older mice displayed prolonged nociceptive behavior following nerve injury when compared with mature adult mice. The expression of peroxisome proliferator‐activated receptor‐gamma coactivator‐1α (PGC‐1α) in S1 was decreased in older mice, whereas PGC‐1α haploinsufficiency promoted prolonged nociceptive behavior after nerve injury. Both aging and PGC‐1α haploinsufficiency led to abnormal S1 neural dynamics, revealed by intravital two‐photon calcium imaging. Manipulating S1 neural dynamics affected nociceptive behavior after nerve injury: chemogenetic inhibition of S1 interneurons aggravated nociceptive behavior in naive mice; chemogenetic activation of S1 interneurons alleviated nociceptive behavior in older mice. More interestingly, adeno‐associated virus‐mediated expression of PGC‐1α in S1 interneurons ameliorated aging‐associated chronification of nociceptive behavior as well as aging‐related S1 neural dynamic changes. Taken together, our results showed that aging‐associated decrease of PGC‐1α promotes pain chronification, which might be harnessed to alleviate the burden of chronic pain in older individuals.


| INTRODUC TI ON
Aging is accompanied by gradual, yet significant, physiological changes in all organ systems, including the nervous system.For example, main sensory modalities including vision, hearing, smell, and taste are generally in decline with advanced age (Cavazzana et al., 2018;Schumm et al., 2009).Additionally, advanced age is associated with declining somatosensation, evidenced by increased mechanical, vibratory, cold, and warmth detection thresholds (Dunn et al., 2015;Johnson et al., 2021;Low Choy et al., 2007).Chronic pain, which has a key somatosensory domain, is highly prevalent in older individuals (Thornbury & Mistretta, 1981;Wang et al., 2006).Notably, age-related decline in somatosensation seems at odds with the high prevalence of chronic pain in older adults.This discrepancy can be explained, in part, by the high prevalence of degenerative diseases such as osteoarthritis in older people (Marks, 2018;Zhang et al., 2019).These age-related degenerative diseases provide peripheral inputs to the nociceptive system underlying chronic pain.However, it is unclear if central pain processing mechanisms also contribute to the high prevalence of chronic pain in older individuals.
The primary somatosensory cortex (S1) is a critical brain region for the processing of somatosensory information, which has been implicated in pain perception (Bushnell et al., 1999;Vierck et al., 2013).For example, in a mouse model of trigeminal neuralgia (Ding, Fischer, et al., 2023), S1 regional neural dynamics were pushed into a synchronized state, which was indispensable for spontaneous pain-like behavior (Ding, Yang, et al., 2023).
Neuropathic pain was found to be linked to S1 somatostatin interneuron dysfunction.Activation of these interneurons prevented the development of neuropathic pain (Cichon et al., 2017).In a mouse model of inflammatory pain, S1 was found to have increased neuronal activity and connectivity (Okada et al., 2021).In humans, postherpetic neuralgia was associated with S1 plasticity, shown as functional reorganization in magnetic resonance imaging, implicated in both evoked and spontaneous pain intensities (Li et al., 2022).Successful treatment of carpal tunnel syndrome using acupuncture was linked to the rewiring of the S1 (Maeda et al., 2017).
Despite the prominent role of S1 in pain processing, its potential role in age-related changes in pain processing has only begun to unravel (Yezierski, 2012).Using resting-state functional connectivity analysis of magnetic resonance imaging, a recent study found that older, as compared to younger adults, displayed higher pain thresholds, enhanced functional connectivity of somatosensory cortices, and reduced connectivity between regions involved in pain inhibition (Gonzalez-Roldan et al., 2020).Particularly, older participants showed reduced functional connectivity between key nodes of the descending pain inhibitory pathway (Gonzalez-Roldan et al., 2020).
These results highlight a plausible role for the central nervous system in the greater vulnerability to chronic pain in older adults.In line with this, older mice displayed exaggerated nociceptive behavior in response to tonic painful stimuli such as acetone and capsaicin (Millecamps et al., 2020).Using a monoiodoacetate-induced knee osteoarthritis model, acute pain processing was similar between young and older rats.Hyperalgesia-like nociceptive behavior lasted significantly longer in older rats than young adults (Ro et al., 2020).This model of aging-related pain chronification was related to changes in the pain modulatory network in older animals (Da Silva et al., 2021).
Peroxisome Proliferator-Activated Receptor Gamma Coactivator 1-alpha (PGC-1α) serves as a crucial transcriptional coactivator, exerting control over mitochondrial biogenesis, energy metabolism, and neuronal function (Baur et al., 2006;Liu et al., 2007;Sahin et al., 2011).These processes are integral to pain modulation and processing within the somatosensory cortex.Recent research indicates that PGC-1α may play a role in modulating pain sensitivity and the development of chronic pain conditions (Miao, Adkins-Threats, et al., 2020).It is implicated in regulating the expression of genes involved in nociception, neuroinflammation, and synaptic transmission within pain pathways.Furthermore, PGC-1α influences neuronal excitability and synaptic plasticity in regions like the dorsal horn of the spinal cord and the somatosensory cortex (Cheng et al., 2012;Vanaveski et al., 2021), implicated in pain processing.Changes in PGC-1α expression or activity have been associated with alterations in pain behavior in animal models of neuropathic and inflammatory pain.While upregulation of PGC-1α is linked to reduced pain sensitivity, downregulation or dysfunction may worsen pain symptoms (Miao, Adkins-Threats, et al., 2020).However, the specific influence of cortical PGC-1α on pain in older individuals and its role in promoting pain chronification remains largely unexplored.
In this study, we examined the role of S1 in aging-related pain chronification by studying nociceptive behavior and S1 neural dynamics following chronic constriction injury of the sciatic nerve (CCI) in mature adult and older mice.Our results suggest that nerve injury in older mice induced prolonged nociceptive behavior and sustained abnormal S1 neural dynamics, revealed by intravital two-photon calcium imaging.Manipulating S1 neural dynamics using chemogenetics influenced nociceptive behavior after nerve injury.Mechanistically, aging was accompanied by decreased expression of PGC-1α in S1.
Adeno-associated virus-mediated expression of PGC-1α in S1 interneurons ameliorated aging-related pain chronification.Taken together, our results reveal a cortical mechanism underlying agingrelated pain chronification that could be harnessed for effective pain treatment in older adults.

| Age-related differences in nociceptive behavior following chronic constriction injury in mice
Mature adult (4-month-old adult) and older (18-months-old) C57BL/6 mice (N = 8 mice) underwent chronic constriction injury to the sciatic nerve (CCI) followed by testing for nociceptive behavior, including hindpaw mechanical and thermal withdrawal thresholds (Figure 1a).
Older mice displayed slightly higher, albeit not statistically significant, | 3 of 16 baseline withdrawal parameters compared with mature adults.After CCI, mature adults decreased their withdrawal thresholds, which returned to pre-injury baseline levels around 4 weeks post-injury.
However, older mice displayed significantly lower withdrawal parameters than adult mice starting 3 weeks post-injury, which then gradually returned to baseline levels at 7 weeks post-nerve injury (Figure 1b,c).Therefore, older mice displayed prolonged nociceptive behaviors, referred to as 'aging-associated pain chronification'.
Additionally, our data did not reveal significant disparities between the adult male and female mice (Figure S1a,b).

| Altered S1 neuronal dynamics in aging-associated pain chronification
To examine S1 neuronal dynamics in aging-associated pain chronification, two-color intravital two-photon calcium imaging was established to simultaneously investigate excitatory neurons and interneurons.A widely accepted viral strategy based on Dlx promoter was used for interneuron targeting (Dimidschstein et al., 2016).
Previously, we reported efficient pan-interneuron targeting using this strategy in both S1 and the hippocampus (Ding, Fischer, et al., 2023;Zhou et al., 2023).AAV8-Dlx-GCaMP6f (green color) and AAV8-CaMKII-jRGECO1a (red color) were used to image interneuron and excitatory neurons (Figure 3a), respectively, in mature adult and older mice at resting state without anesthesia (N = 4 mice).For excitatory neurons, at day 7 post-nerve injury, both mature adult and older mice displayed exaggerated calcium dynamics when compared with day 0 (Figure 3b).Excitatory neuron hyperactivities largely resolved by day 35 in adult but not older mice (Figure 3c).For interneurons, at day 7 post-nerve injury, both adult and older mice displayed muted calcium dynamics when compared with day 0 (Figure 3d).
Interneuron hypoactivities largely resolved day 35 in adult but not older mice (Figure 3e).Importantly, at day 35, mature adult but not older mice were able to resolve nociceptive behavior following nerve injury.As such, aging-related pain chronification was accompanied by sustained abnormal S1 neuron dynamics characterized by exaggerated excitatory neuron activities and muted interneuron activities.Therefore, it is plausible that altered S1 neural dynamics were responsible for aging-related pain chronification.

| DISCUSS ION
Chronic pain has high prevalence in older age group, posing significant societal and medical burdens.Whether aging-associated changes in the central nervous system predispose older individuals to chronic pain is unknown.In this report, using a mouse model of nerve injury-induced nociceptive pain, older mice displayed significantly prolonged nociceptive behavior than mature adults, consistent with aging-associated pain chronification.Using two-color intravital twophoton calcium imaging to simultaneously investigate excitatory neuron and interneuron activities in the S1, aging-associated pain chronification was accompanied by excitatory neuron hyperactivity and interneuron hypoactivity.Mechanistically, S1 in older mice had lower levels of PGC-1α than mature adults.PGC-1α has been shown to be critical for interneuron functional integrity (Lucas et al., 2010(Lucas et al., , 2014;;Vanaveski et al., 2021).Chemogenetic inhibition of S1 interneurons precipitated nociceptive behavior in naïve mice, whereas chemogenetic activation of S1 interneurons alleviated nociceptive behavior in older mice following nerve injury.More importantly, overexpression of PGC-1α in S1 interneurons of older mice alleviated aging-associated pain chronification and dampened S1 excitatory neuron hyperactivity.As such, these results support a critical role for PGC-1α in aging-associated pain chronification.to prevent neuropathic pain development in animal models (Cichon et al., 2017).Additionally, the importance of DRG and spinal cord alterations in pain processing has been elegantly examined in previous studies, including in Petitjean et al. (2015).Moreover, recent research has highlighted the intricate interactions between cortical and spinal contributions to pain processing.Corticospinal projections have been implicated in modulating touch and tactile neuropathic pain sensitivity (Liu et al., 2018), providing a framework to connect DRG/spinal cord and cortical modulations.In our recent study, we demonstrated that while cortical interneuron modulation influenced nociceptive behavior in a mouse model of trigeminal neuralgia, peripheral decompression led to dampened pain-related cortical neural dynamics and alleviation of nociceptive behavior (Ding, Yang, et al., 2023).These findings underscore the complex interplay between cortical and spinal mechanisms in pain modulation.Another potential limitation is that we employed reflex-based nociceptive behavior, while affective components of pain were not included in the study.Recent studies have highlighted the influence of S1 manipulations on both sensory and affective pain responses (Ziegler et al., 2023).Alterations in neural activity within the S1 cortex have been associated not only with changes in sensory thresholds but also with affective aspects of pain processing.Future studies incorporating affective behavioral readouts alongside sensory assessments will provide a more comprehensive understanding of our results.PGC-1α is known to be ubiquitously expressed, with particularly high levels observed in tissues with high metabolic demands, such as skeletal muscle, heart, and brain regions like the hippocampus and cortex (Sahin et al., 2011;Wang et al., 2021).This widespread expression underscores its importance in coordinating metabolic homeostasis and cellular energy production throughout the body.Previous research has shown that alterations in PGC-1α levels can impact various physiological processes, including mitochondrial biogenesis, oxidative metabolism, glucose homeostasis, and cellular responses to stress (Sahin et al., 2011;Zhao et al., 2008).As a mitochondrial biogenesis master regulator, PGC-1α has been found to be critical for neuronal function.
Interneurons provide fundamental mechanisms for information processing during sensory perception and memory formation through orchestrating neuronal network activities.From a neuroenergetic standpoint, inhibitory interneurons, including parvalbumin neurons, have high metabolomic demand and are highly dependent on mitochondrial function (Kann et al., 2014).As a result, they are susceptible to oxidative stress, neuroinflammation, and other insults (Crapser et al., 2020;Powell et al., 2012;Roque et al., 2023).
As such, we speculate that aging-related decrease of PGC-1α might predispose interneurons to dysfunction.Notably, mitochondrial dysfunction has been recognized as one of the hallmarks of aging Intravital two-photon calcium imaging has greatly accelerated the exploration of neuronal activities in physiology and pathological conditions (Svoboda & Yasuda, 2006).Using a red variant of genetically encoded calcium indicator (Dana et al., 2016) in conjunction with a green color calcium sensor, we could simultaneously image excitatory neuron and interneuron activities in the setting of sciatic nerve injury.Our results support that aging-related changes in the central nervous system predispose older individuals to chronic pain, which could be targeted to treat aging-related pain chronification and promote successful aging.

| Animals
All experiments in this study are in accordance with the National Institutes of Health guidelines/regulations and approved by Massachusetts General Hospital Institutional Animal Care and Use Committee.Adult male/female C57BL/6 mice between 12 and 16 weeks of age and older male/female C57BL/6 mice between 17 and 18 months of age at the start of the experiment were used.
Efforts were made to minimize the number of animals used in this study.Mice were purchased from the Jackson Laboratory (USA) and housed in a temperature-controlled colony room on a 24-h light cycle (lights on at 7 am; lights off at 7 pm), such that behavioral testing was performed during the light phase of the cycle.Mice were well maintained on ad libitum food and water.Additionally, the animal care and monitoring procedures followed the ARRIVE guidelines for transparent reporting.Procedures to mitigate pain, suffering, and distress were detailed in the experimental protocols.Any anticipated or unforeseen adverse events were documented.Humane endpoints, including monitored signs and their frequency, were established for the study.C21 (Tocris, catalog 5548) was intraperitoneally administered at 1 mg/kg.

| Chronic constriction injury of sciatic nerve surgery (CCI)
Mice were anesthetized with oxygenated isoflurane (2.5% for induction; 1.5%-2% for maintenance) using a customized nose cone.After the right side of lower extremity was antiseptically prepared with 10% povidone-iodine solution (Medline Industries Inc., Northfield, IL), the right side sciatic nerve was dissected following 1% lidocaine infiltration.CCI was produced by loosely ligating the exposed sciatic nerve referring to the method of Bennett and Xie (Bennett & Xie, 1988;Zhou et al., 2023) using four 6-0 chromic gut sutures with 2 mm apart.Sham mice were made following the same surgical process except for never-ligation.The skin was opposed with three 6-0 vicryl sutures (Ethicon, Somerville, NJ).

| Mechanical withdrawal threshold
All behavioral tests were conducted by an investigator who was blinded to the animal groups and surgeries.von Frey test was used to assess mechanical withdrawal threshold following the method as previously described (Shen et al., 2017).Briefly, mice were habituated for 30 min for 3 consecutive days before baseline behavior assessment.On the testing day, mice were placed on an elevated mesh platform and covered with a transparent Plexiglas chamber (10 × 10 × 12 cm).After 10-20 min of acclimation, withdrawal responses to punctate mechanical stimuli were tested using the von Frey filaments (Figure 1b).Each fiber was applied five times to the plantar aspect of the left hind paw for 1 s with a 10-s interval between each stimulation.The test started with the force of 0.04 grams and continued ascending order up to 2 grams (cutoff).The mechanical withdrawal threshold was defined as the force at which withdrawal occurred at least three out of five applications, and two grams was recorded as the threshold if less than three positive responses to all filaments.

| Hargreaves test
Thermal hyperalgesia to heat stimulation was evaluated following the procedure as previously described (Yang et al., 2022).Briefly, mice were habituated 30 min for three consecutive days before baseline behavior assessment.On the testing day, mice were placed on a preheated glass platform (~28°C) and covered with transparent Plexiglas cubicles for ~10 min.Remove mice drops when necessary.Using a radiant heat source underneath the glass and focus the light beam to the middle of the hindpaw underwent CCI surgery (Figure 1b).Paw withdrawal latency was defined as the time duration (seconds) from the initiation of heat exposure to the hindpaw withdrawal.To avoid tissues burning, 20 s was set as the cut-off time.

| Virus injection and cranial window implantation
The procedures were performed as previously described (Ding, Fischer, et al., 2023;Ding, Yang, et al., 2023).Briefly, mice were anesthetized with oxygenated isoflurane (3% for induction and 1.5% for maintenance) while the respiratory rate was closely monitored.To avoid corneal dehydration, eye lubricant was applied to moisten the eyes of mice.To minimize perioperative pain, Ketorolac tromethamine (Althenex, Schaumburg, IL USA) was

F
I G U R E 3 Aging-associated pain chronification is accompanied by excitatory neuron hyperactivity and interneuron hypoactivity.(a) Diagram and sketch depict experiment design.Mixed AAV8-CaMKII-jRGECO1a and AAV8-Dlx-GCaMP6f was injected into the S1HL cortex of adult and older mice.a representative image of brain slice showed the viruses transfection in S1HL cortex.Intravital two-photon imaging was performed before CCI surgery and day 7, 35 after CCI surgery.A representative view of calcium signal under microscope.(b) Three representative neural activity traces of excitatory neuron in each group at indicated time points.(c) Integrated calcium signal ΔF/F.N = 4, box represents 25th, 50th, and 75th percentile and whisker represent minimum/maximum values.Two-way ANOVA followed by Bonferroni post hoc test was performed for the comparison between groups.*p < 0.05; NS: not significant (p > 0.05).(d) Three representative neural traces of inhibitory neuron in each group at indicated time points.(e) Integrated calcium signal ΔF/F.N = 4, box represents 25th, 50th and 75th percentile and whisker represent minimum/maximum values.Two-way ANOVA followed by Bonferroni post hoc test was performed for the comparison between groups.*p < 0.05; NS: not significant (p > 0.05).F I G U R E 4 Inhibition of interneuron activity in S1HL cortex results in hyperalgesia in older mice.(a) Representative immunostaining images.Brain slices throughout S1HL cortex were selected and stained with anti-GABA receptor.Immunofluorescence imaging showed 80% dTomato+ cells were GABA+ cells in S1HL cortex.N = 4, data are represented as mean ± SD.(b) Diagram and sketch depict experiment design.AAV9-Dlx-Gi DREADD-dTomato was injected into S1HL of adult mice.A representative image of brain slice with box region zoomed verified viral transfection in S1HL cortex.(c) Mechanical withdrawal threshold was significantly decreased after C21 administrated.(d) Hindpaw withdrawal latency in response to S1HL inhibition.Comparing with the mice subject to vector virus, mice injected Gi GREADDs exhibited significant decreasing in hindpaw withdrawal latency after C21 administration.N = 8, data are presented as mean ± SD, two-tailed paired t test, **p < 0.01, NS: not significant (p > 0.05).

F
Activation of interneurons in S1HL alleviates aging-associated pain chronification.(a) Diagram and sketch depict experimental design.AAV9-Dlx-Gq DREADD-dTomato was injected into S1HL of adult mice.Following baseline behavioral test, mice received CCI surgery and C21 was administrated at day 14 after surgery.Pain behaviors were assessed 30 min after C21.(b) Mechanical withdrawal threshold.N = 8 mice, in mice received vector virus, before versus after surgery ***p < 0.001, before versus after C21 ***p < 0.001; in mice received Gq DREADD, before versus after surgery ***p < 0.001, before versus after C21 **p < 0.01.(c) Hindpaw withdrawal latency.Comparing with mice subjected to vector virus, mice received Gq DREADD injection in S1HL and C21 administration exhibited significant increasing in mechanical withdrawal threshold and hindpaw withdrawal latency.N = 8, data are presented as mean ± SD in panel B. One-way ANOVA followed by Tukey post hoc test was performed for the comparison.**p < 0.01, ***p < 0.001.NS: not significant (p > 0.05).(d) Representative immunofluorescent images.Brain slices throughout S1HL cortex were selected and stained with anti-GABA.Immunofluorescence imaging showed 80% dTomato+ cells were GABA + cells in S1HL cortex.N = 4, data are represented as mean ± SD.One potential limitation of our study is that we did not examine spinal and dorsal root ganglia (DRG) mechanisms.Studies have demonstrated the significance of spinal cord and DRG in pain modulation, as well as cortical mechanisms in pain perception.For example, activation of cortical somatostatin interneurons has been shown

F
Overexpression of PGC-1α in S1HL interneurons promotes the resolution of excitatory neuron hyperactivity after CCI in older mice.(a) Diagram and sketch depict experiment design.Mixed viruses of AAV-mDlx-PGC-1a-2A-mCherry and AAV-CaMKII-GCaMP6f were microinjected into S1HL cortex of older mice.A representative image of brain slice showed viruses expression in S1HL cortex.Intravital two-photon imaging was performed before CCI surgery at day 0 and following imaging was performed at day 7 and 35 after surgery.(b) Three sample traces of excitatory neuron activity in each group at indicated time points.(c) Integrated calcium signals ΔF/F.N = 6, data are presented as mean ± SD, two-tailed paired t test, **p < 0.01.NS: not significant (p > 0.05).(Lopez-Otin et al., 2023).Our results showed that in mature adult mice, interneuron hypoactivity was restored to baseline by day 35 post sciatic nerve injury, along with resolution of nociceptive behavior.In contrast, in older mice, interneuron hypoactivity failed to restore during the same time frame, which led to prolonged nociceptive behavior.Thus, improving mitochondrial health might promote healthy aging and alleviate aging-associated comorbidities.