Modulation of inflammatory pain by myeloid FABP5

Abstract

The initiation and maintenance of pain involves complex neuroimmune interactions, with myeloid cell output finetuning the balance between pronociceptive and antinociceptive signaling. Fatty acid binding protein 5 (FABP5) is a lipid chaperone whose inhibition produces antinociceptive effects in models of inflammatory pain. While FABP5 is highly expressed in myeloid cells that are recruited to sites of inflammation, the contribution of myeloid FABP5 to nociception is unknown. Utilizing myeloid-specific FABP5 conditional knockout (cKO) mice, we demonstrate that FABP5⁺ myeloid cells infiltrate into the paw following the onset of carrageenan-induced inflammation. However, myeloid cell recruitment was unaffected by FABP5 deletion. FABP5 cKO were characterized by attenuated development of inflammatory thermal hyperalgesia, which was observed throughout the experimental time course and was evident in both sexes. Transcriptomic profiling of myeloid cells obtained from carrageenan injected paws identified numerous differentially expressed genes between FABP5 cKO mice and Flox controls. Notably, FABP5 cKO myeloid cells featured elevated expression of interleukin-10 (IL-10) and CD206, indicative of polarization towards an M2-like anti-inflammatory state. Collectively, this study ascribes a new role to myeloid FABP5 in pain modulation, thereby expanding the repertoire of cell-types wherein FABP5 regulates nociception to myeloid cells.

Supplementary Information

The online version contains supplementary material available at https://doi.org/10.1038/s41598-026-65946-y.

Introduction

Pain affects over 20% of the population and negatively impacts quality of life¹, necessitating the development of novel efficacious analgesics. Noxious stimuli are transmitted by nociceptors from sites of inflammation to the dorsal horn of the spinal cord². Inflammatory mediators released within sites of tissue injury enhance nociceptor sensitization and activation to exacerbate pain³. It is now well-established that immune cells recruited to inflamed tissue act in concert with resident immune cells to bidirectionally regulate the initiation and resolution of inflammatory pain^4,5. Notably, myeloid cells exert key roles in driving the initiation and maintenance of pain, with pro-inflammatory M1-like myeloid cells enhancing pain by releasing a variety of pronociceptive cytokines and chemokines^6,7. In contrast, M2-like polarized myeloid cells, characterized by the expression of interleukin-10 (IL-10), promote antinociceptive effects and pain resolution^8–11. Therefore, enhancement of myeloid cell polarization towards an M2-like anti-inflammatory phenotype may attenuate the development of pain¹².

Fatty acid binding protein 5 (FABP5) is an intracellular lipid chaperone that coordinates intracellular lipid transport, nuclear receptor signaling, and the expression of pro-inflammatory cytokines and chemokines^13–18. Pharmacological or genetic inhibition of FABP5 produces antinociceptive effects in models of inflammatory, neuropathic, and arthritic pain^13–20. Mechanistically, FABP5 deletion in sensory neurons augments endocannabinoid signaling, leading to activation of cannabinoid receptor 1 and peroxisome proliferator-activated receptor alpha to suppress transient receptor potential vanilloid 1 (TRPV1) sensitization and nociceptor activation¹⁶. FABP5 is additionally expressed in myeloid cells, including macrophages, wherein it governs cellular polarization toward an M1-like state^21–23. In contrast, FABP5 deletion reduces pro-inflammatory cytokine release and promotes polarization towards an anti-inflammatory M2-like phenotype^21–23. These converging findings support the possibility that FABP5 may modulate the transcriptomic landscape of myeloid cells and in turn regulate nociception, which was the goal of the current study. Our findings demonstrate that FABP5 deletion from myeloid cells confers antinociceptive effects that are accompanied by transcriptomic changes including the upregulation of IL-10.

Results

Recruitment of FABP5⁺ myeloid cells to inflamed tissue

We recently developed FABP5^Flox mice that possess a STOP-Flox tdTomato cassette²⁴, enabling tdTomato expression in cells following Cre-mediated FABP5 deletion. FABP5^Flox mice were subsequently crossed with myeloid-specific LysM-Cre mice to yield mice lacking FABP5 in cells of myeloid lineage (termed FABP5 cKO). Intraplanar injection of carrageenan resulted in the recruitment of tdTomato⁺ cells into the inflamed paw, which was not observed in FABP5^Flox controls, confirming that tdTomato labeling is specific to FABP5 cKO mice (Fig. 1). We employed flow cytometry to profile the recruitment of myeloid cells into the paws following carrageenan injection. CD45⁺/CD11b⁺/Ly6G⁻ myeloid cells were profiled in carrageenan injected paws and were further subdivided into Ly6C⁺ and Ly6C⁻ myeloid cells representing infiltrating and resident myeloid cells, respectively^7,25–27(Fig. 2A). TdTomato was expressed in > 50% of Ly6C⁺ and Ly6C⁻ myeloid cells obtained from FABP5 cKO mice (Fig. 2B).

Fig. 1.

Characterization of FABP5 and tdTomato expression in FABP5^Flox and FABP5 cKO paws. Immunofluorescence of FABP5 and tdTomato expression in paws 4 h after injection of vehicle (control) or carrageenan. Note the infiltration of FABP5⁺ cells and a corresponding lack of tdTomato expression in FABP5^Flox paws. In contrast, FABP5 cKO paws are characterized by infiltration of tdTomato⁺ cells and a corresponding loss of FABP5 expression. Scale bar: 50 μm.

Fig. 2.

Analysis of myeloid cells by flow cytometry. (A) Representative flow cytometry plots depicting the gating strategy to evaluate myeloid cell populations in paws. (B) Representative results of tdTomato expression in non-neutrophil myeloid (CD45⁺/CD11b⁺/Ly6G⁻) cells from FABP5^Flox and FABP5 cKO paws 4 h after carrageenan or saline administration.

We subsequently characterized immune cell infiltration patterns in FABP5^Flox and FABP5 cKO mice following carrageenan injection. Profiling of total myeloid cells, Ly6C⁺ myeloid cells, Ly6C⁻ myeloid cells, and Ly6G⁺ neutrophils 4 h after carrageenan administration revealed increased infiltration of myeloid cells, principally neutrophils, but no differences in cell populations between the genotypes (Fig. 3). Similar results were observed at the 24 h time point, with large increases in Ly6C⁺ myeloid cells that were comparable in magnitude between FABP5^Flox and FABP5 cKO mice (Fig. 4). These results indicate that FABP5 deletion in myeloid cells does not impact the recruitment of immune cells and/or proliferation of resident immune cells in the inflamed paw.

Fig. 3.

Impact of FABP5 deletion on paw myeloid cell composition 4 h after carrageenan injection. Myeloid cells were quantified from FABP5^Flox and FABP5 cKO paws 4 h after an injection of carrageenan or saline. (A) Total myeloid cells (CD45⁺/CD11b⁺), (B) neutrophils (CD45⁺/CD11b⁺/Ly6G⁺), (C) non-neutrophil myeloid cells (CD45⁺/CD11b⁺/Ly6G⁻), (D) Ly6C⁻ myeloid cells (CD45⁺/CD11b⁺/Ly6G⁻/Ly6C⁻), and (E) Ly6C⁺ myeloid cells (CD45⁺/CD11b⁺/Ly6G⁻/Ly6C⁺) were quantified and normalized to all live cells. *, p < 0.05 (n = 4).

Fig. 4.

Myeloid cell composition 24 h after carrageenan injection in FABP5^Flox and FABP5 cKO paws. (A) Total myeloid cells (CD45⁺/CD11b⁺), (B) neutrophils (CD45⁺/CD11b⁺/Ly6G⁺), (C) non-neutrophil myeloid cells (CD45⁺/CD11b⁺/Ly6G⁻), (D) Ly6C⁻ myeloid cells (CD45⁺/CD11b⁺/Ly6G⁻/Ly6C⁻), and (E) Ly6C⁺ myeloid cells (CD45⁺/CD11b⁺/Ly6G⁻/Ly6C⁺) were quantified and normalized to all live cells. *, p < 0.05; **, p < 0.01; ***, p < 0.001 (n = 3).

Myeloid FABP5 deletion attenuates inflammatory hyperalgesia

We next examined whether myeloid FABP5 modulates the development of thermal hyperalgesia. Our focus on thermal responses is based on previous findings demonstrating that FABP5 inhibition suppresses TRPV1 sensitization and attenuates thermal hyperalgesia^13–16. Intraplantar injection of carrageenan induced thermal hypersensitivity in male and female FABP5^Flox mice (Fig. 5A and B). In contrast, paw withdrawal latencies were elevated in FABP5 cKO mice at both 4 h and 24 h time points, indicative of antihyperalgesic effects (Fig. 5A and B). Comparable antinociceptive effects were observed in male and female FABP5 cKO mice (Fig. 5A and B), consistent with previous findings in global FABP5 KO mice¹⁶. These findings demonstrate that myeloid FABP5 deletion suppresses the development of inflammatory thermal hyperalgesia in male and female mice.

Fig. 5.

FABP5 deletion in myeloid cells suppresses inflammatory hyperalgesia. (A, B). Thermal paw withdrawal latencies in male and female FABP5^Flox and FABP5 cKO mice were assessed at baseline, 4 h, and 24 h after intraplantar injection of carrageenan. *, p < 0.05; ***, p < 0.001 (n = 6).

Carrageenan-induced hyperalgesia is accompanied by the upregulation of pro-inflammatory cytokines in WT mice, which is blunted in FABP5 KO mice¹⁵. Accordingly, we examined the expression of tumor necrosis factor alpha (TNFα) and interleukin-1β (IL-1β) in the paws of FABP5^Flox and FABP5 cKO mice following injection of carrageenan or vehicle. Comparable upregulation of TNFα and IL-1β was observed in FABP5^Flox and FABP5 cKO mice (Fig. 6), suggesting that FABP5 deletion may modulate myeloid cell transcriptomic programs that are not reflected in whole paws, which we subsequently examined.

Fig. 6.

Expression of pro-inflammatory cytokines in paws 4 h after carrageenan administration. Expression of (A) TNFα and (B) IL-1β in paws of FABP5^Flox and FABP5 cKO mice after intraplantar injection of carrageenan or vehicle. Transcript expression was normalized to vehicle injected FABP5^Flox paws. *, p < 0.05; ***, p < 0.001 (n = 3).

FABP5 deletion alters inflammatory gene expression in myeloid cells

Myeloid cells were isolated from paws of FABP5^Flox and FABP5 cKO mice by flow cytometry 4 h after injection of carrageenan or saline control. Transcriptomic profiling revealed significant alterations in gene expression between the genotypes (Fig. 7A and B). There were 415 downregulated transcripts in FABP5 cKO mice compared to FABP5^Flox post-carrageenan, with 90 of those overlapping the vehicle controls (Fig. 7C). Furthermore, 353 transcripts were upregulated in FABP5 cKO compared to FABP5^Flox, with 60 shared with the vehicle control. As expected, FABP5 expression was downregulated in FABP5 cKO myeloid cells (Fig. 7A and B). Anti-inflammatory mediators including IL-10 were elevated in FABP5 cKO compared to FABP5^Flox cells (Fig. 7B). Conversely, pro-inflammatory cytokines including CCL17, CCL22, and TGF-β1 were downregulated. We subsequently employed qPCR to validate findings from the RNAseq analysis. Indeed, carrageenan injection resulted in an upregulation of CCL17 in FABP5^Flox myeloid cells that was not observed in FABP5 cKO cells (Fig. 8A). Importantly, FABP5 cKO cells were characterized by the upregulation of IL-10 (Figs. 7B and 8B). No differences in CCL22 and TGFβ expression were noted between the conditions or genotypes (Fig. 8C and D). To further establish a shift toward an M2-like phenotype in FABP5 cKO mice, we examined the expression of the M2 marker CD206/Mrc1 in paw tissue after carrageenan injection. Indeed, our results revealed elevated expression of CD206 in paws from FABP5 cKO compared to FABP5^Flox littermates (Fig. 9), which was also reflected in the transcriptomic analysis (Fig. 7B).

Fig. 7.

Transcriptomic profiles of Ly6G⁻ myeloid cells from FABP5^Flox and FABP5 cKO. Myeloid cells were flow sorted from mice 4 h after injection of carrageenan or vehicle. (A, B) Volcano plots depicting differential transcript expression between FABP5^Flox and FABP5 cKO cells, presented as log₂FC of cKO/Flox abundance ratios with the log₂FC threshold set at 0.58 and -log₁₀ p-value threshold set at 1.3. Differential gene expression profiles for (A) vehicle and (B) carrageenan sorted cells are shown. (C) Venn diagram depicting transcripts upregulated and downregulated following carrageenan injection that are shared with saline controls, as well as those unique to each condition.

Fig. 8.

qPCR validation of select transcripts in Ly6G⁻ myeloid cells. Relative expression of (A) CCL17, (B) IL-10, (C) CCL22, and (D) TGFβ1 in myeloid cells sorted from FABP5^Flox and FABP5 cKO paws 4 h after injection of carrageenan or vehicle. Transcript expression was normalized to vehicle injected FABP5^Flox cells. *, p < 0.05 (n = 5).

Fig. 9.

CD206 expression in paws after carrageenan administration. Immunofluorescence of CD206 and tdTomato expression in paws of FABP5^Flox and FABP5 cKO mice 4 h after injection of vehicle (control) or carrageenan. CD206 was expressed in a greater proportion of cells in FABP5 cKO paws compared to FABP5^Flox controls following carrageenan injection (FABP5^Flox = 12.0 ± 2.0% vs. FABP5 cKO = 17.4 ± 4.3%; p = 0.034, n = 5). Scale bar: 50 μm.

Pathway enrichment analysis of flow-sorted myeloid cells was subsequently performed to identify differentially expressed immune-related pathways between the genotypes following saline and carrageenan injection (Fig. 10). In the saline condition, gene sets induced by IL-1 family cytokines were suppressed in FABP5 cKO myeloid cells relative to FABP5^Flox controls (Fig. 10A), consistent with reduced IL-1–driven inflammatory signaling (Fig. 10A). The majority of the IL-1 associated signatures remained suppressed following carrageenan injection, which additionally featured a distinct transcriptional response characterized by the downregulation of IFN-associated signaling pathways (Fig. 10B and Supplementary Table 1).

Fig. 10.

Suppression of IL-1-related pathways in FABP5 cKO myeloid cells. Pathway enrichment analysis was performed on RNA-seq data from flow-sorted myeloid cells isolated from the paws of FABP5 cKO and FABP5^Flox mice 4 h after saline or carrageenan injections. IL-1 family signature pathways (highlighted in red) are suppressed in FABP5 cKO relative to FABP5^Flox for (A) saline and (B) carrageenan conditions based on adjusted p-values < 0.05. Differentially regulated non-IL-1 pathways (highlighted in gray) are shown for comparison. Pathways are separated by activation and suppression based on normalized enrichment scores.

Discussion

The current study investigated the contribution of myeloid FABP5 to nociception and demonstrates that deletion of FABP5 in myeloid cells reduces inflammatory hyperalgesia, which coincides with a transcriptional shift towards an anti-inflammatory phenotype. Specifically, FABP5 cKO mice were characterized by upregulation of IL-10 and CD206 as well as suppression of IL-1 family pathways, which is consistent with the noted shift toward an M2-like state in FABP5 KO macrophages^21–23,28. Previous studies have established that IL-10 along with IL-4 attenuate pain in inflammatory pain models and upregulate endogenous opioid peptides^29,30. IL-10 additionally activates STAT3 to suppress IFN-γ, providing a mechanistic basis for the downregulation of IFN-associated pathways observed in the pathway enrichment analysis of FABP5 cKO myeloid cells^31,32. FABP5 deletion also blunted the upregulation of the pronociceptive chemokine CCL17. Notably, CCL17 (and CCL22) induces hyperalgesia while inhibition of CCR4, the receptor for CCL17 and CCL22, attenuates pain^33–36. The upregulation of IL-10 and CD206 along with the concomitant suppression of pro-inflammatory cytokines supports a model wherein FABP5 deletion shifts myeloid cells away from a pro-inflammatory and towards an anti-inflammatory state, potentially through multiple downstream effector pathways.

Previous work has demonstrated that FABP5 inhibition modulates pain through multiple overlapping mechanisms. Genetic deletion of FABP5 in nociceptors promotes the emergence of antinociceptive effects that are mediated by cannabinoid receptor 1 and peroxisome proliferator-activated receptor alpha¹⁶. Specifically, FABP5 deletion attenuates inflammatory TRPV1 sensitization driven by nerve growth factor and nociceptor output^15,16. The observed upregulation of IL-10 in FABP5 cKO mice may likewise dampen nociceptor hyperexcitability as previously reported¹⁰. FABP5 inhibition within the human and rat osteoarthritic synovium reduces the expression and release of pronociceptive cytokines and chemokines and ameliorates osteoarthritis-induced pain¹⁷. Our current findings establish that FABP5 deletion induces a shift toward an M2-like state and position myeloid FABP5 as an additional locus of FABP5-mediated inflammatory and nociceptive signaling. Taken together, these results provide additional support for a model wherein FABP5 inhibition leads to the engagement of functionally distinct mechanisms that culminate in the suppression of pain.

While our study provides evidence that myeloid FABP5 modulates nociception, future studies will be needed to identify the precise molecular mechanisms underlying these effects. For example, FABP5 inhibition reduces the expression of pro-inflammatory cytokines in the carrageenan-injected paw¹⁵. In contrast, the expression of such cytokines was unaffected in paws from FABP5 cKO mice (Fig. 6). This suggests that myeloid FABP5 may exert more subtle effects on the inflammatory tissue environment that are not reflected in the whole paw. It is also worth noting that FABP5 deletion was evident in > 50% of myeloid cells (Fig. 2). This level of deletion was still sufficient to produce robust antihyperalgesic effects across both sexes and experimental time points. Although our study focused on myeloid cells within the inflamed tissue, myeloid cells additionally infiltrate dorsal root ganglia and modulate the initiation and resolution of pain^4,5,37. Indeed, IL-10 released by myeloid cells within dorsal root ganglia has an established role in dampening pain, and it is likely that enhanced IL-10 signaling within dorsal root ganglia may contribute at least in part to the antinociceptive effects observed in FABP5 cKO mice^38,39. Future studies will be required to establish a potential role for FABP5 as a modulator of immune-neuronal crosstalk between myeloid cells and nociceptors in dorsal root ganglia. The recent development of orally active FABP5 inhibitors^18,20 coupled with the capacity of FABP5 to modulate nociception via distinct mechanisms highlights the unique therapeutic potential of FABP5 inhibition to treat diverse pain conditions.

Materials and methods

Animals

All experiments were conducted with approval from Stony Brook University Institutional Animal Care and Use Committee (#277150) and in accordance with the ARRIVE guidelines. All methods were carried out in accordance with relevant guidelines and regulations. Adult FABP5^Flox24 and LysM-Cre mice (Jackson Labs #4781) were maintained on a 12:12-hour light: dark cycle with ad libitum access to food and water in a temperature-controlled environment. FABP5^Flox mice were crossed with LysM-Cre mice to generate the corresponding FABP5 cKO mice, which were subsequently backcrossed to FABP5^Flox mice for at least six generations. Ten- to fourteen-week-old FABP5 cKO and FABP5^Flox (Cre⁻) littermates were used for all experiments.

Thermal hyperalgesia

Ten- to fourteen-week-old mice were acclimated to the experimental room and the Hargreaves plantar apparatus (IITC Life Science) for 2 days prior to testing. Baseline paw withdrawal latencies were obtained and mice subsequently received a unilateral intraplantar injection of carrageenan (1% in 20 µL in sterile saline) while the contralateral paw received saline control. Thermal paw withdrawal latencies were assessed 4 and 24 h later.

FACS and flow cytometry

Paws pad tissues were dissected and incubated in a digestion cocktail containing 300 µg/mL Liberase TM and 50 U/mL DNAseI in RPMI + 5% FBS. The samples were incubated for 90 min in 37 °C and strained in a 70 μm strainer into Falcon tubes using RPMI + 5% FBS. Samples were washed with 1% BSA in 1X PBS and stained with TruStain FCX Plus and antibody mix containing CD45 (FITC), CD11b (Brilliant Violet 510), Ly6G (Brilliant Violet 650), and Ly6C (Brilliant Violet 765 for BD Fortessa or APC/Cyanine7 for BD FACSAria). Zombie NIR dye was used to assess cell viability. Cells were subsequently quantified with the Kaluza software package. Following cell quantification, CD45+/CD11b+/Ly6G- cells were collected by FACS or quantified by flow cytometry at the Stony Brook University’s Flow Cytometry Core Facility.

transcriptomic Analysis

Paw tissues from 3 to 4 FABP5^Flox and FABP5 cKO mice were pooled and CD45⁺/CD11b⁺/Ly6G⁻ cells were collected. RNA was extracted using the PureLink RNA Micro kit (Invitrogen). RNA quality was assessed with an Agilent Bioanalyzer and processed by Lexogen NGS services (Vienna, Austria) for library preparation and paired-end RNA sequencing using the LUTHOR platform. Sequencing data was processed using the Lexogen in-house pipeline services, including Unique Molecular Identifiers (UMI)-based PCR duplicate removal, alignment with STAR aligner, and quantification. Differential gene expression analysis was performed using the DESeq2 package. Genes were considered significantly regulated using the criteria of log₂ fold change (log₂FC) > 0.58 and < −0.58 with a threshold of FDR adjusted p-value < 0.05. These differentially expressed genes were visualized in a volcano plot using EnhancedVolcano R package⁴⁰. For pathway enrichment, gene set enrichment analysis (GSEA) was conducted using the clusterProfiler R package with genes ranked by log2FC as described⁴¹. Pathway enrichment analysis was assessed for immune-related transcriptional programs while retaining broader pathway context using gene sets from Molecular Signatures Database (MSigDB), which are derived from and named for specific immune cell contexts but contain genes broadly expressed across immune cell populations. Specifically, Immunological Signatures and Canonical Pathways from BioCarta, and WikiPathways to encompass immune and inflammatory processes. Statistical significance was determined using permutation-based enrichment testing with FDR correction in which pathways with an adjusted p-value < 0.05 were considered significantly enriched.

qPCR

RNA from flow sorted CD45⁺/CD11b⁺/Ly6G⁻ cells was processed for cDNA synthesis using SuperScript III First Strand Synthesis (ThermoFisher) and qPCR was performed using PowerUp SYBR green (ThermoFisher) on a StepOnePlus instrument (Applied Biosystems). The following primers were used: GAPDH forward 5’- CTCCCACTCTTCCACCTTCG − 3’ and reverse 5’-GCCTCTCTTGCTCAGTGTCC − 3’; TGFβ1 forward 5’-TGATACGCCTGAGTGGCTGTCT-3’ and reverse 5’-CACAAGAGCAGTGAGCGCTGAA-3’; CCL22 forward 5’-GTGGAAGACAGTATCTGCTGCC-3’ and reverse 5’-AGGCTTGCGGCAGGATTTTGAG-3’; CCL17 forward 5’-CGAGAGTGCTGCCTGGATTACT-3’ and reverse 5’-GGTCTGCACAGATGAGCTTGCC-3’; IL-10 forward 5’-TGAATTCCCTGGGTGAGAAG-3’ and reverse 5’-TGGCCTTGTAGACACCTTGG-3’; IL-1β forward 5’-GCTTCAGGCAGGCAGTATC-3’ and reverse 5’-AGGATGGGCTCTTCTTCAAAG-3’; TNFα forward 5’-GAACTGGCAGAAGAGGCACT-3’ and reverse 5’-AGGGTCTGGGCCATAGAACT-3’. Quantification was performed with the 2^−ΔΔCt approach, with GAPDH as the housekeeping gene, and samples normalized based on FABP5^Flox vehicle.

Immunofluorescence staining

Mice were anesthetized with isoflurane and transcardially perfused with saline, followed by 4% paraformaldehyde (PFA) in 0.1 M phosphate-buffered saline (PBS, pH 7.4). Paw tissues were dissected, left in the same fixative at 4℃ overnight, and transferred to a 30% sucrose solution in 0.1 M PBS for cryoprotection. The tissues were embedded in a gelatin-albumin mixture (3% gelatin, 30% egg albumin in deionized water) and frozen using liquid nitrogen-chilled isopentane. Frozen tissue blocks were sectioned using a cryostat set to − 16℃. Tissue sections were cut to a thickness of 18 μm, placed on gelatin/chromium-coated slides, air-dried, and stored in a − 20℃ freezer for future use.

Sections were thawed at room temperature for 20 min, fixed on the slides with 4% PFA for 5 min, washed three times for 10 min each, and treated with 0.2% Triton X-100 in PBS for 10 min to permeabilize the tissues. Blocking was carried out for 30 min using a solution of 10% normal donkey serum (NDS), 2% glycine, and 0.1% Triton X-100 in PBS. The tissues were incubated overnight at 4℃ with primary antibodies diluted in PBS containing 0.3% Triton X-100 and 5% NDS. The following primary antibodies were used: TdTomato (MyBioSource, MBS448092, 1:750), FABP5 (BioVendor R&D, RD181060100, 1:350), and CD206 (Cell Signaling Technology, 24595, 1:500). The sections were washed three times with PBS for 10 min each and incubated at room temperature for 1 h with secondary antibodies diluted in PBS containing 0.3% Triton X-100 and 5% NDS. Secondary antibodies included: donkey anti-rabbit and donkey anti-goat that were conjugated to either Alexa 488 or 594 diluted in PBS with 0.3% Triton X-100 and 5% NDS. Afterward, the sections were washed three times in PBS (10 min each) and mounted with ProLong Gold antifade mounting medium with DAPI (Vector Labs Burlingame, CA, Cat# H-2000). Slides were stored at 4℃ in the dark until imaging. Fluorescent signals were visualized using a Zeiss Axioplan 2 epifluorescence microscope. Images were captured using a Zeiss AxioCam HRm digital camera and processed with AxioVision Rel. 4.6 software. Only adjustments for brightness and contrast were applied. For CD206 quantification, CD206⁺ cells in paws were quantified using the automated cell detection feature in QuPath software (v 0.6.0) and normalized to the number of DAPI⁺ cells.

Statistical analysis

Data are presented as means ± S.D. Thermal hyperalgesia was analyzed by Two-Way ANOVA with Tukey post-hoc test. Infiltrating cell quantification was analyzed using Two-Way ANOVA followed by Tukey post-hoc test. CD206⁺ cells were analyzed by unpaired t-test. qPCR results were assessed by comparing ΔCT values using Two-Way ANOVA followed by Tukey post-hoc test with p < 0.05 set as the significance criterion.

Supplementary Information

Below is the link to the electronic supplementary material.

Author contributions

FD and MK designed the study. FD and LS performed all experiments. FD, LS, and MK analyzed the data. FD and MK wrote the manuscript. All authors have read and approved the final version of the manuscript.

Funding

This work was supported by a National Institutes of Health grants NS137595 and DA035949.

Data availability

The transcriptomic data were submitted to the GEO database as GSE298388.

Declarations

Competing interests

The authors declare no competing interests.