Different immune cells mediate mechanical pain hypersensitivity in male and female mice

Journal name:
Nature Neuroscience
Volume:
18,
Pages:
1081–1083
Year published:
DOI:
doi:10.1038/nn.4053
Received
Accepted
Published online

A large and rapidly increasing body of evidence indicates that microglia-to-neuron signaling is essential for chronic pain hypersensitivity. Using multiple approaches, we found that microglia are not required for mechanical pain hypersensitivity in female mice; female mice achieved similar levels of pain hypersensitivity using adaptive immune cells, likely T lymphocytes. This sexual dimorphism suggests that male mice cannot be used as proxies for females in pain research.

At a glance

Figures

  1. Mechanical allodynia after nerve injury is reversed by microglial inhibition in male but not female mice.
    Figure 1: Mechanical allodynia after nerve injury is reversed by microglial inhibition in male but not female mice.

    (a) Reversal of established SNI-induced mechanical allodynia by intrathecal minocycline (MCL) in male, but not female, mice (see also Supplementary Fig. 1a). Data are presented as mean ± s.e.m. Shown is 50% withdrawal threshold from von Frey filaments before surgery (BL), 7 d after surgery (pre-injection; D7) and 10–120 min post-injection of minocycline (n = 4–5 mice per sex per dose). A different set of mice (n = 4 mice per sex) were tested similarly 28 d post-SNI (D28, right). (b) Male, but not female, mice treated with Mac-1–saporin (SAP) displayed significantly reduced SNI allodynia at 4 h post-treatment. Data are presented as mean ± s.e.m. Shown is 50% withdrawal threshold from von Frey filaments before surgery (BL), 7 d after surgery, pre-injection (D7), and 1, 2, 4 and 24 h post-SAP (n = 11 mice per sex per condition). (c) Intrathecal administration of the P2X inhibitor TNP-ATP, the p38 MAPK inhibitor SB203580, the NGF/BDNF inhibitor Y1036 or the BDNF-sequestering fusion protein TrkB-Fc all blocked SNI-induced allodynia in male, but not female, mice. The NMDA receptor antagonist AP5 blocked allodynia equally in both sexes. Data are presented as mean ± s.e.m. percentage of maximal anti-allodynia (Online Methods; n = 8 mice per sex per drug). (d) Development of SNI-induced mechanical allodynia in male and female mice lacking CNS microglial BDNF (that is, with tamoxifen-induced Cre-loxP–mediated deletion of the Bdnf gene in CX3CR1-positive cells). Mutant (Bdnf−/−) male mice failed to develop full allodynia displayed by female Bdnf−/− and wild-type (Bdnf+/+) mice. Data are presented as mean ± s.e.m. absolute withdrawal threshold from von Frey filaments before and 3, 7, 10 and 14 d after surgery (n = 4–8 mice per sex per genotype). *P < 0.05, **P < 0.01, ***P < 0.001 compared with corresponding female mice by t test.

  2. Mechanical allodynia after nerve injury is mediated by adaptive immune cells in female mice.
    Figure 2: Mechanical allodynia after nerve injury is mediated by adaptive immune cells in female mice.

    (a) Reversal of mechanical allodynia after SNI by intrathecally administered glial inhibitors minocycline (MCL), fluorocitrate (FC) and propentofylline (PPF) in male CD-1 mice and in immunocompromised nude mice of both sexes, but not in female CD-1 mice. Data are presented as mean ± s.e.m. percentage of maximal anti-allodynia (n = 4–7 mice per sex per drug per genotype). (b) Male-specific reversal of allodynia from the PPARα ligand, fenofibrate (FFB) was blocked by the PPARα antagonist GW6471 and by castration (TX). Data are presented as in a (n = 4–6 mice per sex per condition). (c) Female-specific reversal of allodynia from the PPARγ ligand pioglitazone (PIO) was blocked by the PPARγ antagonist GW9662 and by testosterone proprionate (TP). Data are presented as in a (n = 7–10 mice per sex per condition). *P < 0.05, **P < 0.01 compared with corresponding female mice by t test. #P < 0.05, ##P < 0.01, ###P < 0.001 compared with same-sex wild-type (CD-1) or vehicle group by t test. n.t. = not tested.

  3. Dose-dependent reversal of SNI-induced allodynia by intrathecal glial inhibitors in male but not female mice.
    Supplementary Fig. 1: Dose-dependent reversal of SNI-induced allodynia by intrathecal glial inhibitors in male but not female mice.

    Reversal of SNI-induced allodynia by the glial inhibitors minocycline (MCL; a), fluorocitrate (FC; b) and propentofylline (PPF; c). Symbols represent mean ± SEM percentage of maximum possible anti-allodynia (see Methods); n=4–6 mice/dose/sex/drug. ANOVAs revealed significant main effects of sex in each case (MCL: F1,23 = 15.3, p=0.001; FC: F1,19 = 19.3, p<0.001; PPF: F1,26 = 66.7, p<0.001). ***p<0.001 compared to female mice.

  4. Reversal of complete Freund/'s adjuvant (CFA)‑induced mechanical allodynia by intrathecal glial inhibitors minocycline (MCL), fluorocitrate (FC) and propentofylline (PPF) in male but not female mice.
    Supplementary Fig. 2: Reversal of complete Freund’s adjuvant (CFA)‑induced mechanical allodynia by intrathecal glial inhibitors minocycline (MCL), fluorocitrate (FC) and propentofylline (PPF) in male but not female mice.

    Bars represent mean ± SEM percentage anti-allodynia; n=5–7 mice/sex/drug. All three glial inhibitors reversed allodynia in male but not female mice (MCL: t7 = 5.7, p<0.001; FC: t11 = 3.4, p=0.005; PPF: t8 = 3.8, p=0.005). **p<0.01, ***p<0.001 compared to female mice.

  5. Repeated systemic (i.p.) injections of minocycline (MCL) reverse SNI- and CFA‑induced mechanical allodynia in male but not female mice.
    Supplementary Fig. 3: Repeated systemic (i.p.) injections of minocycline (MCL) reverse SNI- and CFA‑induced mechanical allodynia in male but not female mice.

    Symbols represent mean ± SEM mechanical withdrawal thresholds (g) before (BL) and after SNI surgery (a) or CFA injection (b), and one day after three daily injections of MCL at 25 mg/kg/day (i.p.); n=6–8 mice/sex/drug in each experiment. In both experiments, a significant sex x drug x repeated measure (post-BL time points only) was observed (F1,20 = 7.6, p=0.01; F1,26 = 68.0, p<0.001, respectively). *p<0.05, ***p<0.001 compared to all other groups. Single injections of MCL produced no effects even at high doses (data not shown); repeated injections have previously shown to increase the anti-allodynic efficacy of MCL (Nazemi et al., Pharmacol. Biochem. Behav., 2012). Note that these observations are in contradiction to those of Bastos and colleagues, who observed partial reversal of chronic constriction injury (CCI)‑mediated mechanical allodynia by 100 mg/kg minocycline in female C57BL/6 mice (Bastos et al., Neurosci. Lett., 543:157-162, 2013) and partial reversal of late‑phase (15–30 min post-injection) formalin test responding by 50 and 100 mg/kg minocycline in male and female mice (Bastos et al., Neurosci. Lett., 553:110-114, 2013). In experiments by another group, 50 mg/kg minocycline was found to reverse thermal hyperalgesia induced by interleukin-1β in female heterozygous G-protein-coupled receptor kinase 2 (GRK2) mutant mice on a C57BL/6 genetic background (Willemen et al., Pain, 2010). It is unclear whether the differences are due to assay, dose, measures or test parameters.

  6. Testosterone‑dependence of the efficacy of minocycline in reversing CFA-induced allodynia.
    Supplementary Fig. 4: Testosterone‑dependence of the efficacy of minocycline in reversing CFA-induced allodynia.

    ANOVA revealed a significant sex x hormonal condition interaction (F3,24 = 3.9, p=0.02). Minocycline (50 μg) was ineffective in castrated male mice (male GDX) and young (4 week-old) mice of both sexes; its efficacy was reinstated in gonadectomized (GDX; castrated or ovariectomized) mice of both sexes given testosterone proprionate replacement (GDX + TP). Bars represent mean ± SEM percentage anti-allodynia (n=4 mice/sex/hormonal condition).

  7. Spinal microgliosis 7 days after SNI in male (left) and female (right) mice.
    Supplementary Fig. 5: Spinal microgliosis 7 days after SNI in male (left) and female (right) mice.

    Microglial proliferation is shown by Iba1 immunoreactivity (red). Insets are high power images of transverse sections of lumbar spinal cord. Dotted region shows region of high power image. Scale bar = 100 μm. No sex differences were seen in NeuN-positive or GFAP‑positive cells (data not shown; also see Supplementary Fig. 8c).

  8. Microglial depletion by intrathecal Mac-1-saporin treatment in male and female mice.
    Supplementary Fig. 6: Microglial depletion by intrathecal Mac-1-saporin treatment in male and female mice.

    Depletion of microglia ipsilateral (Ipsi.) to the SNI (7 days post‑surgery) in both male (M) and female (F) mice 4 h after treatment with Mac‑1‑saporin (Mac-1-SAP) toxin compared to mice treated with saporin (SAP) vehicle. Bars represent mean ± SEM Iba1‑positive (Iba1+) cells in the lumbar spinal cord dorsal horn (n=5 mice/sex/condition). Cont.=contralateral. Similar microglial depletion (≈25%) was observed in both sexes (male: t8 = 2.4, p=0.02; female: t8 =2.8, p=0.001). *p<0.05.

  9. SNI upregulates expression of the Itgam (CD11b), Emr1 (F4/80; a surface marker for microglia), Irf5, and Irf8 genes (see below) in the dorsal horn of the spinal cord equally in both sexes, but only upregulates the P2rx4 gene (P2X4R) in male mice.
    Supplementary Fig. 7: SNI upregulates expression of the Itgam (CD11b), Emr1 (F4/80; a surface marker for microglia), Irf5, and Irf8 genes (see below) in the dorsal horn of the spinal cord equally in both sexes, but only upregulates the P2rx4 gene (P2X4R) in male mice.

    Symbols represent normalized 2−ΔΔCt values (n=8 biological replicates/sex/surgical condition) compared to the average of four reference genes. All male vs. female t-test values p>0.40, except for P2rx4: t14 = 3.5, p=0.003. **p<0.01 compared to other sex. In male mice, P2X4R gene expression is under the transcriptional control of interferon regulatory factors 5 and 8 (IRF5 and IRF8). Following peripheral nerve injury, IRF8 is upregulated and directs gene expression changes associated with microglial reactivity, including motility, chemotaxis, Iba1 expression and increases in IRF5 (Masuda et al., Nat. Commun. 2014; Masuda et al., Purinergic Signal., 2014). IRF5 binds directly to the promoter region of P2rx4 and has direct transcriptional control over it, resulting in de novo expression of microglial P2X4R after peripheral nerve injury. Following nerve injury, Irf8 is upregulated (and, as shown above, equally in males and females), with a consequent spinal microglial proliferation and upregulation of Itgam (and Aif1, which codes for Iba1). Irf5 is in turn upregulated and, in males but not females, leads to increased P2rx4 gene expression. However, in females, despite the increased IRF5, P2rx4 expression is unaffected. This pattern of gene expression changes after nerve injury shows the point of divergence in the cellular and molecular pathways underlying neuropathic pain in male and female mice lies at the induction of P2X4R.

  10. Nerve injury-induced microglia responses are present in microglial-specific Bdnf mutant mice of both sexes.
    Supplementary Fig. 8: Nerve injury-induced microglia responses are present in microglial-specific Bdnf mutant mice of both sexes.

    a) Spinal cord dorsal horn microgliosis following SNI in Cx3cr1CreER x loxP‑Bdnf mice. Insets are high power images taken from the ipsilateral dorsal horn of the spinal cords from female (left) and male (right) mice. Microglia were labelled with Iba1 (red). Both males and females show the characteristic microglial proliferation around the central terminals of peripherally axotomised sciatic afferents. b) Microglial-specific Bdnf mutant express a YFP IRES element. YFP (green) colocalizes directly with microglia (Iba1; red) in the spinal cord. c) Other cellular populations are identical in male and female microglial-specific Bdnf mutant mice after SNI. Astrocytes and neuronal nuclei are shown by GFAP and NeuN immunohistochemistry (both blue), respectively.

  11. Reversal of developed SNI‑induced mechanical allodynia in male Bdnf-/- mice.
    Supplementary Fig. 9: Reversal of developed SNI‑induced mechanical allodynia in male Bdnf−/− mice.

    Reversal of SNI in male mutant mice (Bdnf−/−) in which central nervous system microglial BDNF is deleted following tamoxifen (TMX) treatment, but not male Bdnf+/+ or female mice of both genotypes. Repeated measures ANOVA revealed a significant sex x genotype x repeated measures interaction: F5,65 = 3.1, p=0.01. Symbols represent mean ± SEM absolute withdrawal thresholds from von Frey filaments before surgery, 1 week post‑surgery, and 4–8 weeks post‑TMX treatment (n=4 mice/sex/genotype). **p<0.01, ***p<0.001 compared to all other groups.

  12. Quantitative sex differences in baseline mechanical sensitivity (a) and SNI- and CFA-induced mechanical allodynia (b-e) in various experiments.
    Supplementary Fig. 10: Quantitative sex differences in baseline mechanical sensitivity (a) and SNI- and CFA-induced mechanical allodynia (b–e) in various experiments.

    Bars in a represent mean ± SEM baseline von Frey thresholds of various mouse populations by sex. Bars in b,c represent mean ± SEM mechanical allodynia in CD-1 mice measured at 7 days post-SNI surgery (b) or 3 days post-CFA injection (c) in various drug conditions or mouse populations by sex. Bars in d,e represent mean ± SEM mechanical allodynia measured at 7 days post-SNI surgery or 3 days post-CFA injection in CD-1 (wildtype) and nude mice (d) and C57BL/6 (wildtype) and Rag1−/− mice (e). *p<0.05, **p<0.01, ***p<0.001 compared to other sex within-genotype or condition by (uncorrected) t-test. The slightly but significantly increased neuropathic allodynia of female CD-1 mice (see graphs b,d) has not been observed previously, but is well-documented (Coyle et al., Neurosci. Lett. 1995; Dominguez et al., Eur. J. Pain, 2012; Dina et al., Neuroscience, 2007; LaCroix-Fralish et al., Neuroscience 2006; LaCroix-Fralish et al., Pain, 2005; Tall et al., Pharmacol. Biochem. Behav., 2001), although strain-dependent (DeLeo & Rutkowski, Neurosci. Lett., 2000), in the rat.

  13. Female nude mice [ldquo]switch[rdquo] to a microglial-dependent system in the mediation of CFA allodynia.
    Supplementary Fig. 11: Female nude mice “switch” to a microglial-dependent system in the mediation of CFA allodynia.

    Graph shows reversal of mechanical allodynia 3 days after CFA by i.t. administered glial inhibitors minocycline (MCL), fluorocitrate (FC) and propentofylline (PPF) in male but not female CD-1 mice, but in immunocompromised nude mice of both sexes. Bars represent mean ± SEM percentage of maximum reversal of allodynia (% anti-allodynia) (n=4–7 mice/sex/drug/genotype). ANOVAs showed significant genotype x sex interactions in all cases (MCL: F1,19 = 7.1, p=0.01; FC: F1,16 = 7.7, p=0.01; PPF: F1,22 = 4.2, p=0.05). **p<0.01, ***p<0.001 compared to same‑strain female mice by t‑test. ••p<0.01, •••p<0.001 compared to same-sex CD-1 mice by t‑test.

  14. Female Rag1-/- mice [ldquo]switch[rdquo] to a microglial-dependent system in the mediation of SNI and CFA allodynia.
    Supplementary Fig. 12: Female Rag1−/− mice “switch” to a microglial-dependent system in the mediation of SNI and CFA allodynia.

    Graph shows reversal of mechanical allodynia after SNI or CFA by i.t. minocycline (MCL; 50 μg) in male but not female wildtype (+/+; C57BL/6) mice, but in immunocompromised Rag1−/− (-/-) mice of both sexes. Bars represent mean ± SEM percentage of maximum reversal of allodynia (% anti-allodynia) (n=6–8 mice/sex/genotype). ANOVAs revealed significant genotype x sex interactions in both cases (SNI: F1,20 = 9.5, p=0.006; CFA: F1,24 = 4.7, p=0.04). ***p<0.001 compared to same‑genotype female mice by t‑test. ••p<0.01, •••p<0.001 compared to same‑sex +/+ mice by t-test.

  15. Adoptive transfer of splenocytes into female Rag1-/- mice reinstates their use of the female glial-independent pathway.
    Supplementary Fig. 13: Adoptive transfer of splenocytes into female Rag1−/− mice reinstates their use of the female glial-independent pathway.

    a) Successful repopulation by adoptive transfer. Spleen sections taken from a female Rag1−/− mouse (left) and a female Rag1−/− mouse following adoptive splenocyte transfer (right). Spleen macrophages are labelled in green (Iba1) and repopulated T-cells in red (CD3). Scale bar = 200 μm. b) Adoptive transfer of splenocytes (Splen.) from immunocompetent female (F) Rag1+/+ into immunocompromised female Rag1−/− mice (i.e., Rag1−/− F + Splen. condition) restores the male (M)-like ability of i.t. MCL (50 μg; 3 days after injection; D3) to reverse CFA allodynia. Rag1−/− F + Veh. indicates mutant females which received the adoptive transfer vehicle without splenocytes. Symbols represent mean ± SEM 50% withdrawal threshold from von Frey filaments before CFA (BL), 3 days after CFA, pre-MCL injection (D3), and 10–120 min post‑injection of MCL (n=4–6 mice/sex/condition except for adoptive transfer group, n=10). Repeated measures ANOVA revealed a significant group x repeated measures interaction: F15,100 = 5.8, p<0.001. ***p<0.001 compared to Rag1+/+ F and Rag1−/− F + Splen. groups by Tukey’s posthoc test.

  16. Female mice have a larger pool of T-cells in the blood than do male mice.
    Supplementary Fig. 14: Female mice have a larger pool of T-cells in the blood than do male mice.

    The number of CD4+ and CD8+ T cells in the blood of male and female naïve mice was quantified using FACS analysis. a) Representative examples of flow cytometric analysis of peripheral CD4+ and CD8+ T-cells in the blood of male (top) and female (bottom) mice. (b,c) histograms represent the FACS analysis (bars represent mean ± SEM counts) of CD4+ (b) and CD8+ (c) T-cells in the blood obtained from 3 naïve mice per sex. Female mice have more CD4+ (t4 = 2.9, p=0.04) and CD8+ (t4 = 3.8, p=0.02) T-cells than males. d) The number of lymphocytes in the blood of male and female mice was further confirmed using standard complete blood counting (CBC) (n=6 mice/sex; bars represent mean ± SEM counts); female mice exhibited higher numbers of lymphocytes (t10 = 2.5, p=0.03). *p<0.05 compared to males.

  17. Higher expression of T-cell markers in the lumbar spinal cord dorsal horns of female versus male CD-1 mice 7 days after SNI.
    Supplementary Fig. 15: Higher expression of T-cell markers in the lumbar spinal cord dorsal horns of female versus male CD-1 mice 7 days after SNI.

    Bars represent Ct values ± SEM (n=5–6 mice/sex), expressed relative to the housekeeping gene, Gapdh. CD3e: t8 = 2.8, p=0.02; CD4: t6 = 3.3, p=0.02; CD8a: t9 = 2.2, p=0.05. *p<0.05 compared to males.

References

  1. Grace, P.M., Hutchinson, M.R., Maier, S.F. & Watkins, L.R. Nat. Rev. Immunol. 14, 217231 (2014).
  2. Watkins, L.R. & Maier, S.F. Nat. Rev. Drug Discov. 2, 973985 (2003).
  3. Mogil, J.S. & Chanda, M.L. Pain 117, 15 (2005).
  4. Sorge, R.E. et al. J. Neurosci. 31, 1545015454 (2011).
  5. Tsuda, M. et al. Nature 424, 778783 (2003).
  6. Beggs, S., Trang, T. & Salter, M.W. Nat. Neurosci. 15, 10681073 (2012).
  7. Jin, S.X., Zhuang, Z., Woolf, C.J. & Ji, R.R. J. Neurosci. 23, 40174022 (2003).
  8. Coull, J.A.M. et al. Nature 438, 10171021 (2005).
  9. Parkhurst, C.N. et al. Cell 155, 15961609 (2013).
  10. Costigan, M. et al. J. Neurosci. 29, 1441514422 (2009).
  11. Cao, L. & DeLeo, J.A. Eur. J. Immunol. 38, 448458 (2008).
  12. Hu, P., Bembrick, A.L., Keay, K.A. & McLachlan, E.M. Brain Behav. Immun. 21, 599616 (2007).
  13. Mogil, J.S., Sternberg, W.F., Kest, B., Marek, P. & Liebeskind, J.C. Pain 53, 1725 (1993).
  14. Juni, A. et al. Anesthesiology 112, 181188 (2010).
  15. Mogil, J.S. et al. Proc. Natl. Acad. Sci. USA 100, 48674872 (2003).
  16. Scotland, R.S., Stables, M.J., Madalli, S., Watson, P. & Gilroy, D.W. Blood 118, 59185927 (2011).
  17. Zhang, M.A. et al. Proc. Natl. Acad. Sci. USA 109, 95059510 (2012).
  18. Kim, C.F. & Moalem-Taylor, G. J. Pain 12, 370383 (2011).
  19. Draleau, K.S. et al. J. Pain Relief 3, 003 (2014).
  20. Sorge, R.E. et al. Nat. Methods 11, 629632 (2014).
  21. Chaplan, S.R., Bach, F.W., Pogrel, J.W., Chung, J.M. & Yaksh, T.L. J. Neurosci. Methods 53, 5563 (1994).
  22. Bourquin, A.-F. et al. Pain 122, 14.e1114.e14 (2006).
  23. Shields, S.D., Eckert, W.A. III & Basbaum, A.I. J. Pain 4, 465470 (2003).
  24. Bennett, G.J. & Xie, Y.-K. Pain 33, 87107 (1988).
  25. Hylden, J.L.K. & Wilcox, G.L. Eur. J. Pharmacol. 67, 313316 (1980).
  26. Yang, M., Rainone, A., Shi, X.Q., Fournier, S. & Zhang, J. Acta Neuropathol. Commun. 2, 5 (2014).
  27. Mogil, J.S. et al. Pain 126, 2434 (2006).

Download references

Author information

  1. These authors contributed equally to this work.

    • Robert E Sorge &
    • Josiane C S Mapplebeck

Affiliations

  1. Department of Psychology, McGill University, Montreal, Quebec, Canada.

    • Robert E Sorge,
    • Josiane C S Mapplebeck,
    • Sarah Rosen,
    • Loren J Martin,
    • Jean-Sebastien Austin,
    • Susana G Sotocinal,
    • Di Chen &
    • Jeffrey S Mogil
  2. Department of Psychology, University of Alabama at Birmingham, Birmingham, Alabama, USA.

    • Robert E Sorge
  3. Program in Neuroscience and Mental Health, Hospital for Sick Children, Toronto, Ontario, Canada.

    • Josiane C S Mapplebeck,
    • Simon Beggs,
    • Jessica K Alexander,
    • YuShan Tu &
    • Michael W Salter
  4. Department of Physiology, University of Toronto, Toronto, Ontario, Canada.

    • Josiane C S Mapplebeck,
    • Simon Beggs,
    • Jessica K Alexander &
    • Michael W Salter
  5. University of Toronto Centre for the Study of Pain, Toronto, Ontario, Canada.

    • Josiane C S Mapplebeck,
    • Simon Beggs,
    • Jessica K Alexander &
    • Michael W Salter
  6. Department of Anesthesiology, Duke University Medical Center, Durham, North Carolina, USA.

    • Sarah Taves &
    • Ru-Rong Ji
  7. Faculty of Dentistry, McGill University, Montreal, Quebec, Canada.

    • Mu Yang,
    • Xiang Qun Shi,
    • Hao Huang &
    • Ji Zhang
  8. Program in Cell Biology, Hospital for Sick Children, Toronto, Ontario, Canada.

    • Nicolas J Pillon,
    • Philip J Bilan &
    • Amira Klip
  9. Alan Edwards Centre for Research on Pain, McGill University, Montreal, Quebec, Canada.

    • Ji Zhang &
    • Jeffrey S Mogil

Contributions

R.E.S., S.B., R.-R.J., M.W.S. and J.S.M. conceived the study. R.E.S. designed most of the experiments, and J.C.S.M. and J.Z. designed certain experiments. R.E.S., J.C.S.M., S.R., S.T., S.B., J.K.A., L.J.M., J.-S.A., S.G.S., D.C., M.Y., X.Q.S., H.H., N.J.P., P.J.B., Y.T. and A.K. collected and analyzed data. R.E.S., J.C.S.M., S.B., M.W.S. and J.S.M. wrote the paper.

Competing financial interests

The authors declare no competing financial interests.

Corresponding authors

Correspondence to:

Author details

Supplementary information

Supplementary Figures

  1. Supplementary Figure 1: Dose-dependent reversal of SNI-induced allodynia by intrathecal glial inhibitors in male but not female mice. (179 KB)

    Reversal of SNI-induced allodynia by the glial inhibitors minocycline (MCL; a), fluorocitrate (FC; b) and propentofylline (PPF; c). Symbols represent mean ± SEM percentage of maximum possible anti-allodynia (see Methods); n=4–6 mice/dose/sex/drug. ANOVAs revealed significant main effects of sex in each case (MCL: F1,23 = 15.3, p=0.001; FC: F1,19 = 19.3, p<0.001; PPF: F1,26 = 66.7, p<0.001). ***p<0.001 compared to female mice.

  2. Supplementary Figure 2: Reversal of complete Freund’s adjuvant (CFA)‑induced mechanical allodynia by intrathecal glial inhibitors minocycline (MCL), fluorocitrate (FC) and propentofylline (PPF) in male but not female mice. (116 KB)

    Bars represent mean ± SEM percentage anti-allodynia; n=5–7 mice/sex/drug. All three glial inhibitors reversed allodynia in male but not female mice (MCL: t7 = 5.7, p<0.001; FC: t11 = 3.4, p=0.005; PPF: t8 = 3.8, p=0.005). **p<0.01, ***p<0.001 compared to female mice.

  3. Supplementary Figure 3: Repeated systemic (i.p.) injections of minocycline (MCL) reverse SNI- and CFA‑induced mechanical allodynia in male but not female mice. (124 KB)

    Symbols represent mean ± SEM mechanical withdrawal thresholds (g) before (BL) and after SNI surgery (a) or CFA injection (b), and one day after three daily injections of MCL at 25 mg/kg/day (i.p.); n=6–8 mice/sex/drug in each experiment. In both experiments, a significant sex x drug x repeated measure (post-BL time points only) was observed (F1,20 = 7.6, p=0.01; F1,26 = 68.0, p<0.001, respectively). *p<0.05, ***p<0.001 compared to all other groups. Single injections of MCL produced no effects even at high doses (data not shown); repeated injections have previously shown to increase the anti-allodynic efficacy of MCL (Nazemi et al., Pharmacol. Biochem. Behav., 2012). Note that these observations are in contradiction to those of Bastos and colleagues, who observed partial reversal of chronic constriction injury (CCI)‑mediated mechanical allodynia by 100 mg/kg minocycline in female C57BL/6 mice (Bastos et al., Neurosci. Lett., 543:157-162, 2013) and partial reversal of late‑phase (15–30 min post-injection) formalin test responding by 50 and 100 mg/kg minocycline in male and female mice (Bastos et al., Neurosci. Lett., 553:110-114, 2013). In experiments by another group, 50 mg/kg minocycline was found to reverse thermal hyperalgesia induced by interleukin-1β in female heterozygous G-protein-coupled receptor kinase 2 (GRK2) mutant mice on a C57BL/6 genetic background (Willemen et al., Pain, 2010). It is unclear whether the differences are due to assay, dose, measures or test parameters.

  4. Supplementary Figure 4: Testosterone‑dependence of the efficacy of minocycline in reversing CFA-induced allodynia. (126 KB)

    ANOVA revealed a significant sex x hormonal condition interaction (F3,24 = 3.9, p=0.02). Minocycline (50 μg) was ineffective in castrated male mice (male GDX) and young (4 week-old) mice of both sexes; its efficacy was reinstated in gonadectomized (GDX; castrated or ovariectomized) mice of both sexes given testosterone proprionate replacement (GDX + TP). Bars represent mean ± SEM percentage anti-allodynia (n=4 mice/sex/hormonal condition).

  5. Supplementary Figure 5: Spinal microgliosis 7 days after SNI in male (left) and female (right) mice. (334 KB)

    Microglial proliferation is shown by Iba1 immunoreactivity (red). Insets are high power images of transverse sections of lumbar spinal cord. Dotted region shows region of high power image. Scale bar = 100 μm. No sex differences were seen in NeuN-positive or GFAP‑positive cells (data not shown; also see Supplementary Fig. 8c).

  6. Supplementary Figure 6: Microglial depletion by intrathecal Mac-1-saporin treatment in male and female mice. (101 KB)

    Depletion of microglia ipsilateral (Ipsi.) to the SNI (7 days post‑surgery) in both male (M) and female (F) mice 4 h after treatment with Mac‑1‑saporin (Mac-1-SAP) toxin compared to mice treated with saporin (SAP) vehicle. Bars represent mean ± SEM Iba1‑positive (Iba1+) cells in the lumbar spinal cord dorsal horn (n=5 mice/sex/condition). Cont.=contralateral. Similar microglial depletion (≈25%) was observed in both sexes (male: t8 = 2.4, p=0.02; female: t8 =2.8, p=0.001). *p<0.05.

  7. Supplementary Figure 7: SNI upregulates expression of the Itgam (CD11b), Emr1 (F4/80; a surface marker for microglia), Irf5, and Irf8 genes (see below) in the dorsal horn of the spinal cord equally in both sexes, but only upregulates the P2rx4 gene (P2X4R) in male mice. (124 KB)

    Symbols represent normalized 2−ΔΔCt values (n=8 biological replicates/sex/surgical condition) compared to the average of four reference genes. All male vs. female t-test values p>0.40, except for P2rx4: t14 = 3.5, p=0.003. **p<0.01 compared to other sex. In male mice, P2X4R gene expression is under the transcriptional control of interferon regulatory factors 5 and 8 (IRF5 and IRF8). Following peripheral nerve injury, IRF8 is upregulated and directs gene expression changes associated with microglial reactivity, including motility, chemotaxis, Iba1 expression and increases in IRF5 (Masuda et al., Nat. Commun. 2014; Masuda et al., Purinergic Signal., 2014). IRF5 binds directly to the promoter region of P2rx4 and has direct transcriptional control over it, resulting in de novo expression of microglial P2X4R after peripheral nerve injury. Following nerve injury, Irf8 is upregulated (and, as shown above, equally in males and females), with a consequent spinal microglial proliferation and upregulation of Itgam (and Aif1, which codes for Iba1). Irf5 is in turn upregulated and, in males but not females, leads to increased P2rx4 gene expression. However, in females, despite the increased IRF5, P2rx4 expression is unaffected. This pattern of gene expression changes after nerve injury shows the point of divergence in the cellular and molecular pathways underlying neuropathic pain in male and female mice lies at the induction of P2X4R.

  8. Supplementary Figure 8: Nerve injury-induced microglia responses are present in microglial-specific Bdnf mutant mice of both sexes. (662 KB)

    a) Spinal cord dorsal horn microgliosis following SNI in Cx3cr1CreER x loxP‑Bdnf mice. Insets are high power images taken from the ipsilateral dorsal horn of the spinal cords from female (left) and male (right) mice. Microglia were labelled with Iba1 (red). Both males and females show the characteristic microglial proliferation around the central terminals of peripherally axotomised sciatic afferents. b) Microglial-specific Bdnf mutant express a YFP IRES element. YFP (green) colocalizes directly with microglia (Iba1; red) in the spinal cord. c) Other cellular populations are identical in male and female microglial-specific Bdnf mutant mice after SNI. Astrocytes and neuronal nuclei are shown by GFAP and NeuN immunohistochemistry (both blue), respectively.

  9. Supplementary Figure 9: Reversal of developed SNI‑induced mechanical allodynia in male Bdnf−/− mice. (225 KB)

    Reversal of SNI in male mutant mice (Bdnf−/−) in which central nervous system microglial BDNF is deleted following tamoxifen (TMX) treatment, but not male Bdnf+/+ or female mice of both genotypes. Repeated measures ANOVA revealed a significant sex x genotype x repeated measures interaction: F5,65 = 3.1, p=0.01. Symbols represent mean ± SEM absolute withdrawal thresholds from von Frey filaments before surgery, 1 week post‑surgery, and 4–8 weeks post‑TMX treatment (n=4 mice/sex/genotype). **p<0.01, ***p<0.001 compared to all other groups.

  10. Supplementary Figure 10: Quantitative sex differences in baseline mechanical sensitivity (a) and SNI- and CFA-induced mechanical allodynia (b–e) in various experiments. (262 KB)

    Bars in a represent mean ± SEM baseline von Frey thresholds of various mouse populations by sex. Bars in b,c represent mean ± SEM mechanical allodynia in CD-1 mice measured at 7 days post-SNI surgery (b) or 3 days post-CFA injection (c) in various drug conditions or mouse populations by sex. Bars in d,e represent mean ± SEM mechanical allodynia measured at 7 days post-SNI surgery or 3 days post-CFA injection in CD-1 (wildtype) and nude mice (d) and C57BL/6 (wildtype) and Rag1−/− mice (e). *p<0.05, **p<0.01, ***p<0.001 compared to other sex within-genotype or condition by (uncorrected) t-test. The slightly but significantly increased neuropathic allodynia of female CD-1 mice (see graphs b,d) has not been observed previously, but is well-documented (Coyle et al., Neurosci. Lett. 1995; Dominguez et al., Eur. J. Pain, 2012; Dina et al., Neuroscience, 2007; LaCroix-Fralish et al., Neuroscience 2006; LaCroix-Fralish et al., Pain, 2005; Tall et al., Pharmacol. Biochem. Behav., 2001), although strain-dependent (DeLeo & Rutkowski, Neurosci. Lett., 2000), in the rat.

  11. Supplementary Figure 11: Female nude mice “switch” to a microglial-dependent system in the mediation of CFA allodynia. (114 KB)

    Graph shows reversal of mechanical allodynia 3 days after CFA by i.t. administered glial inhibitors minocycline (MCL), fluorocitrate (FC) and propentofylline (PPF) in male but not female CD-1 mice, but in immunocompromised nude mice of both sexes. Bars represent mean ± SEM percentage of maximum reversal of allodynia (% anti-allodynia) (n=4–7 mice/sex/drug/genotype). ANOVAs showed significant genotype x sex interactions in all cases (MCL: F1,19 = 7.1, p=0.01; FC: F1,16 = 7.7, p=0.01; PPF: F1,22 = 4.2, p=0.05). **p<0.01, ***p<0.001 compared to same‑strain female mice by t‑test. ••p<0.01, •••p<0.001 compared to same-sex CD-1 mice by t‑test.

  12. Supplementary Figure 12: Female Rag1−/− mice “switch” to a microglial-dependent system in the mediation of SNI and CFA allodynia. (98 KB)

    Graph shows reversal of mechanical allodynia after SNI or CFA by i.t. minocycline (MCL; 50 μg) in male but not female wildtype (+/+; C57BL/6) mice, but in immunocompromised Rag1−/− (-/-) mice of both sexes. Bars represent mean ± SEM percentage of maximum reversal of allodynia (% anti-allodynia) (n=6–8 mice/sex/genotype). ANOVAs revealed significant genotype x sex interactions in both cases (SNI: F1,20 = 9.5, p=0.006; CFA: F1,24 = 4.7, p=0.04). ***p<0.001 compared to same‑genotype female mice by t‑test. ••p<0.01, •••p<0.001 compared to same‑sex +/+ mice by t-test.

  13. Supplementary Figure 13: Adoptive transfer of splenocytes into female Rag1−/− mice reinstates their use of the female glial-independent pathway. (462 KB)

    a) Successful repopulation by adoptive transfer. Spleen sections taken from a female Rag1−/− mouse (left) and a female Rag1−/− mouse following adoptive splenocyte transfer (right). Spleen macrophages are labelled in green (Iba1) and repopulated T-cells in red (CD3). Scale bar = 200 μm. b) Adoptive transfer of splenocytes (Splen.) from immunocompetent female (F) Rag1+/+ into immunocompromised female Rag1−/− mice (i.e., Rag1−/− F + Splen. condition) restores the male (M)-like ability of i.t. MCL (50 μg; 3 days after injection; D3) to reverse CFA allodynia. Rag1−/− F + Veh. indicates mutant females which received the adoptive transfer vehicle without splenocytes. Symbols represent mean ± SEM 50% withdrawal threshold from von Frey filaments before CFA (BL), 3 days after CFA, pre-MCL injection (D3), and 10–120 min post‑injection of MCL (n=4–6 mice/sex/condition except for adoptive transfer group, n=10). Repeated measures ANOVA revealed a significant group x repeated measures interaction: F15,100 = 5.8, p<0.001. ***p<0.001 compared to Rag1+/+ F and Rag1−/− F + Splen. groups by Tukey’s posthoc test.

  14. Supplementary Figure 14: Female mice have a larger pool of T-cells in the blood than do male mice. (194 KB)

    The number of CD4+ and CD8+ T cells in the blood of male and female naïve mice was quantified using FACS analysis. a) Representative examples of flow cytometric analysis of peripheral CD4+ and CD8+ T-cells in the blood of male (top) and female (bottom) mice. (b,c) histograms represent the FACS analysis (bars represent mean ± SEM counts) of CD4+ (b) and CD8+ (c) T-cells in the blood obtained from 3 naïve mice per sex. Female mice have more CD4+ (t4 = 2.9, p=0.04) and CD8+ (t4 = 3.8, p=0.02) T-cells than males. d) The number of lymphocytes in the blood of male and female mice was further confirmed using standard complete blood counting (CBC) (n=6 mice/sex; bars represent mean ± SEM counts); female mice exhibited higher numbers of lymphocytes (t10 = 2.5, p=0.03). *p<0.05 compared to males.

  15. Supplementary Figure 15: Higher expression of T-cell markers in the lumbar spinal cord dorsal horns of female versus male CD-1 mice 7 days after SNI. (100 KB)

    Bars represent Ct values ± SEM (n=5–6 mice/sex), expressed relative to the housekeeping gene, Gapdh. CD3e: t8 = 2.8, p=0.02; CD4: t6 = 3.3, p=0.02; CD8a: t9 = 2.2, p=0.05. *p<0.05 compared to males.

PDF files

  1. Supplementary Text and Figures (8,882 KB)

    Supplementary Figures 1–15

  2. Supplementary Methods Checklist (128 KB)

Additional data