A neural circuit for sex-dependent conditioned pain hypersensitivity in mice

Abstract

The neural mechanisms underlying sex-specific pain, in which males and females exhibit distinct responses to pain, remain poorly understood. Here we show that in a mouse model of male-specific pain hypersensitivity response to pain conditioning environments (contextual pain hypersensitivity model), elevated free-testosterone leads to hyperactivity of glutamatergic neurons in the medial preoptic area (Glu^mPOA) through activation of androgen receptor signaling, which in turn induces contextual pain hypersensitivity in male mice. Although not observed in naïve female mice, this pain phenotype could be induced in females via chronic administration of testosterone propionate. In addition, Glu^mPOA neurons send excitatory inputs to GABAergic neurons in the ventrolateral periaqueductal gray (GABA^vlPAG) that are required for contextual pain hypersensitivity. Our study thus demonstrates that testosterone/androgen receptor signaling enhances Glu^mPOA → GABA^vlPAG pathway activity, which drives a male-specific contextual pain hypersensitivity, providing insight into the basis of sexually dimorphic pain response.

Fig. 1. Contextual pain hypersensitivity in response to pain conditioning in male, but not female, mice.

a Schematic for the contextual pain hypersensitivity mouse model. b, c Example raster plots and statistic data for writhing and constriction reponses of saline- and AA-injected male and female mice (n = 5; Writhing, F_3,16 = 13.51, p = 0.0001; Constriction, F_3,16 = 58.05, p < 0.0001). d, e Paw withdrawal threshold (PWT) and latency (PWL) of the male^DC, female^DC, male^SC, and female^SC mice (PWT: n = 7 for male^DC, p = 0.4812; n = 6 for female^DC, p = 0.5208; PWL: n = 5 per group, p = 0.6061 for male^DC, p = 0.7439 for female^DC; PWT: n = 10 for male^SC, p < 0.0001; n = 7 for female^SC, p = 0.8053; PWL: n = 9 per group, p = 0.0045 for male^SC, p = 0.7832 for female^SC). f, g PWT and PWL for 7 days after AA injection in male and female mice (PWT: n = 6 per group, F_1,10 = 20.48, p = 0.0011; PWL: n = 5 for male^DC, n = 9 for male^SC, F_1,12 = 5.137, p = 0.0427; PWT: n = 5 for female^DC, n = 4 for female^SC, F_1,7 = 0.4099, p = 0.5424; PWL: n = 5 for female^DC, n = 9 for female^SC, F_1,12 = 4.676, p = 0.0515). h Schematic for the auditory-cue-induced contextual pain hypersensitivity paradigm. i PWT and PWL of the male^DC and male^SC mice (n = 5 per group; PWT, F_1,8 = 3.169, p = 0.0288; PWL, F_1,8 = 7.233, p = 0.0275). j As indicated in panel i, but for female mice (n = 5 per group; PWT, F_1,8 = 0.1501, p = 0.7086; PWL: F_1,8 = 0.2094, p = 0.6594). Significance was assessed by one-way ANOVA with post hoc Bonferroni’s test between groups in c, two-tailed paired Student’s t-test (d, e) and two-way repeated-measures ANOVA with post hoc comparison between groups (f–j). The data are presented as the mean ± SEMs. * P < 0.05, ** P < 0.01, *** P < 0.001; n.s. not significant. See also Supplementary Table 1. Source data are provided as a Source Data file.

Fig. 2. Conditioned context elicits pain hypersensitivity by activating Glu^mPOA neurons.

a, b Schematic and representative images for fiber photometry recording. c–e Heatmaps, representative traces, and average ΔF/F of Glu^mPOA neuronal activity aligned to the time from onset of contextual changes in mice (n = 6 for male; n = 5 for female; male, p = 0.0071; female, p = s. f, g Schematic and representative images for multi-tetrode recordings. h Example recording of spontaneous and light-evoked spikes from a Glu^mPOA neuron. i, j Raster plots with typical traces and quantitative data of spontaneous firing rates of Glu^mPOA neurons in male^DC, male^SC, female^DC, and female^SC mice (n = 31 units from 4 mice for male^DC and female group; n = 35 units from 4 mice for male^SC group; male, p < 0.0001; female, p = 0.2108). k, l Representative images of virus expression and traces of action potentials evoked by 473-nm light (blue bars) recorded from ChR2-expressing Glu^mPOA neurons. m, n Effects of photostimulation on PWT and PWL in male^DC (n = 5 per group; PWT, F_1,8 = 8.312, p = 0.0204; PWL, F_1,8 = 12.11, p = 0.0083) and female^DC mice (PWT: n = 6 per group, F_1,10 = 8.962, p = 0.0135; PWL: n = 5 per group, F_1,8 = 22.21, p = 0.0015). o, p As indicated in panels k, l, but for optogenetic inhibition. q, r Effects of photostimulation on PWT and PWL in male^SC (n = 5 per group; PWT, F_1,8 = 10.10, p = 0.0099; PWL, F_1,8 = 23.19, p = 0.0013) and female^SC mice (n = 5 per group; PWT, F_1,8 = 0.5575, p = 0.4766; PWL, F_1,8 = 2.125, p = 0.1830). Significance was assessed by two-tailed paired Student’s t-test in (e), two-tailed unpaired Student’s t-test (j) and two-way repeated-measures ANOVA with post hoc comparison between groups (m, n, q, r). Scale bars, 100 μm and 10 μm (zoom). The data are presented as the mean ± SEMs. * P < 0.05, ** P < 0.01, *** P < 0.001; n.s. not significant. See also Supplementary Table 1. Source data are provided as a Source Data file.

Fig. 3. Free-testosterone contributes to male-specific contextual pain hypersensitivity.

a Summarized data of free-testosterone levels in the serum of male naïve mice (control), male^DC mice, and male^SC mice on D2, D4, and D8 after acetic acid injection (n = 5 per group; D2, p = 0.0078; D4, p = 0.0157; D8, p = 0.0716). b Schematic for castration surgery (TX) and experimental procedure in male mice. c Summarized data of free-testosterone levels in the serum of Sham, TX-, vehicle-treated TX-, or testosterone propionate (TP)-treated TX-male mice (n = 7 per group, F_3,24 = 22.21, p < 0.0001). d, e PWT and PWL of male^SC mice (PWT: n = 8 for sham group, n = 7 for TX group, n = 5 for TX + Vehicle and TX + TP group, F_3,21 = 9.776, p = 0.0003; PWL: n = 5 per group, F_3,16 = 5.492, p = 0.0087). f, g Representative traces and quantitative data of action potentials recorded from Glu^mPOA neurons in sham- and castration male^SC mice (n = 19 cells from 4 mice per group; F_1,36 = 4.141, p = 0.0493). h, i PWT and PWL of female^SC mice (PWT: n = 6 per group, F_1,10 = 7.450, p = 0.0212; PWL: n = 5 per group, F_1,8 = 4.550, p = 0.0655). j Summarized data of serum free-testosterone levels in female mice (n = 7 per group, p < 0.0001). k Schematic for ovariectomy (OVX) in female mice. l PWT and PWL of OVX female mice (PWT: n = 6 per group, p = 0.1422 for Vehicle, p < 0.0001 for TP; PWL: n = 5 per group, p = 0.1095 for Vehicle, p = 0.0155 for TP). Significance was assessed by two-tailed unpaired Student’s t-test (a, j, l), one-way ANOVA with post hoc Bonferroni’s test between groups (c) and two-way repeated-measures ANOVA with post hoc comparison between groups (d, e, g–i). The data are presented as the mean ± SEMs. * P < 0.05, ** P < 0.01, *** P < 0.001; n.s. not significant. See also Supplementary Table 1. Source data are provided as a Source Data file.

Fig. 4. Stress and memory system regulate male-specific contextual allodynia.

a Schematic for examing the effects of stress on contextual pain hypersensitivity. b PWT and PWL of male^SC mice (n = 5; PWT: F_1,8 = 6.027, p = 0.0396; PWL: F_1,8 = 2.817, p = 0.1318). c, d Raster plots with typical traces and quantitative data for spontaneous firing rates of Glu^mPOA neurons in male^SC mice (n = 24 units for D0-saline group and n = 22 units for other groups from 4 mice; F_1,44 = 14.85, p = 0.0004). e Summarized data for corticosterone and free-testosterone levels (n = 5; corticosterone, p = 0.0329; free-testosterone, p = 0.0042). f As indicated in panel e, but for male mice after TX surgery (n = 5; corticosterone, p = 0.0096; free-testosterone, p = 0.2800). g–i Schematic and pain thresholds of ACSF- and ZIP-treated male^SC mice (n = 5; PWT: F_1,8 = 1.847, p = 0.2112; PWL: F_1,8 = 10.83, p = 0.0110). j As indicated in panel e, but for ACSF- and ZIP-treated male^SC mice (n = 5; corticosterone, p = 0.0204; free-testosterone, p = 0.0039). k Schematic for examing the effects of dCA1 inactivation on contextual pain hypersensitivity in male mice. l, m PWT and PWL of EGFP- and TetTox-infected male^SC mice (n = 5; PWT: F_1,8 = 24.10, p = 0.0012; PWL: F_1,8 = 39.87, p = 0.0002). n, o As indicated in panel c, d, but for EGFP- and TetTox-infected male^SC mice (n = 21, 23, 20, 21 units from 4 mice for D0-EGFP, D0-TetTox, D2-EGFP, and D2-TetTox group, respectively; F_1,81 = 7.354, p = 0.0082). p, q As indicated in panel e, but for EGFP- and TetTox-infected male^SC mice (n = 5; corticosterone, p = 0.0458; free-testosterone, p = 0.0027). Significance was assessed by two-tailed unpaired Student’s t-test (b, e, f, j, p, q), two-way mixed-effects analysis with Bonferroni’s multiple comparisons test (d, o), and two-way repeated-measures ANOVA with post hoc comparison between groups (h, i, l, m). The data are presented as the mean ± SEMs. * P < 0.05, ** P < 0.01, *** P < 0.001; n.s. not significant. See also Supplementary Table 1. Source data are provided as a Source Data file.

Fig. 5. Effects of androgen receptor on contextual pain hypersensitivity.

a Schematic for in vivo microdialysis. b Summarized data of free-testosterone levels (n = 6 mice; F_2,15 = 16.60, p = 0.0002). c, d Representative traces and quantitative data of action potentials recorded from Glu^mPOA neurons (n = 20 cells from 6 mice; 0.05 μM: F_1,38 = 1.413, p = 0.2420; 0.3 μM: F_1,38 = 10.28, p = 0.0027; 1 μM: F_1,38 = 13.11, p = 0.0009). e, f Raster plots with typical traces and summarized data for spontaneous spikes of Glu^mPOA neurons (n = 28, 29, 29, 24 units from 4 mice on D0, n = 36, 27, 28, 25 units from 4 mice on D2; F_3,115 = 7.870, p < 0.0001). g–i Heatmaps, representative traces and average ΔF/F of Glu^mPOA neuronal activity (n = 5 for Sham and TX + Vehicle, n = 4 for TX and TX + TP; DC: Sham::TX, p = 0.6747; TX + Vehicle::TX + TP, p = 0.0916; SC: Sham::TX, p = 0.0108; TX+Vehicle::TX + TP, p = 0.0372). j, k Representative images of virus expression and western blot analysis of AR expression (n = 3 mice per group; p = 0.0041). The samples derive from the same experiment and that gels were processed in parallel. Scale bars, 100 μm and 10 μm (zoom). l PWT, PWL of EGFP- and shAR-infected male^SC mice (n = 5 for EGFP, n = 8 for shAR; PWT: F_1,11 = 24.79, p = 0.0004; PWL: F_1,11 = 7.061, p = 0.0223). m, n As indicated in panels e, f, but for EGFP- and shAR-infected male mice (n = 30, 26, 32, 29 units from 4 mice; DC, p = 0.8221; SC, p < 0.0001). o, p Schematic and pain thresholds for TP injection in EGFP and shAR groups (n = 5; PWT: F_1,8 = 91.12, p < 0.0001; PWL: F_1,8 = 94.86, p < 0.0001). Significance was assessed by two-tailed t-test (i, k, n), two-way mixed-effects (Bonferroni; f), two-way repeated-measures ANOVA (post hoc; d, l, p), and one-way ANOVA (post hoc Bonferroni; b). The data are presented as the means ± SEMs. * P < 0.05, ** P < 0.01, *** P < 0.001; n.s. not significant. See also Supplementary Table 1. Source data are provided as a Source Data file.

Fig. 6. Effects of Glu^mPOA → GABA^vlPAG pathway on contextual pain hypersensitivity.

a, b Schematic for AAV-DIO-ChR2-mCherry virus injection in the mPOA of VgluT2-Cre mice and representative images showing the mPOA injection site and mCherry⁺ fibers in the ventrolateral periaqueductal gray (vlPAG). Scale bars, 100 μm. c Schematic for the Cre-dependent anterograde trans-monosynaptic tracing strategy. d Representative images showing the mPOA injected with AAV-hSyn-Cre-EGFP virus and the vlPAG injected with AAV-DIO-EGFP virus. Scale bars, 200 μm. e, f Representative images and summarized data showing EGFP⁺ neurons within the vlPAG co-localized with glutamatergic neurons and GABAergic interneurons. Scale bars, 10 μm. g, h Schematic and representative image for optical activation of Glu^mPOA neuronal terminals and the simultaneous whole-cell patch clamp recordings of EGFP-labeled GABA^vlPAG neurons. Scale bars, 100 μm. i Representative traces and summarized data of excitatory postsynaptic currents (EPSCs) recorded from GABA^vlPAG neurons (n = 8 cells from 4 mice; p = 0.0003). j Schematic and representative images for chemogenetic inhibition of vlPAG-projecting Glu^mPOA neurons in male^SC mice. Scale bars, 100 μm and 10 μm (zoom). k, l PWT and PWL of male^SC mice treated with CNO in the mCherry and hM4Di groups (n = 5; PWT: F_1,8 = 41.80, p = 0.0002; PWL: F_1,8 = 50.73, p < 0.0001). m, n Schematic and pain thresholds for optogenetic activation of the Glu^mPOA→GABA^vlPAG pathway in male^DC mice (n = 5; PWT: F_1,8 = 16.79, p = 0.0035; PWL: F_1,8 = 18.60, p = 0.0026). o, p As indicated in panels m, n, but for optogenetic inhibition (n = 5; PWT: F_1,8 = 9.699, p = 0.0035; PWL: F_1,8 = 13.45, p = 0.0063). Significance was assessed by two-tailed unpaired Student’s t-test (i) and two-way repeated-measures ANOVA with post hoc comparison between groups (k, l, n, p). The data are presented as the mean ± SEMs. * P < 0.05, ** P < 0.01, *** P < 0.001; n.s. not significant. See also Supplementary Table 1. Source data are provided as a Source Data file.