Formalin-evoked pain triggers sex-specific behavior and spinal immune response

Abstract

Mounting evidence shows sex-related differences in the experience of pain with women suffering more from chronic pain than men. Yet, our understanding of the biological basis underlying those differences remains incomplete. Using an adapted model of formalin-induced chemical/inflammatory pain, we report here that in contrast to male mice, females distinctly display two types of nocifensive responses to formalin, distinguishable by the duration of the interphase. Females in proestrus and in metestrus exhibited respectively a short-lasting and a long-lasting interphase, underscoring the influence of the estrus cycle on the duration of the interphase, rather than the transcriptional content of the dorsal horn of the spinal cord (DHSC). Additionally, deep RNA-sequencing of DHSC showed that formalin-evoked pain was accompanied by a male-preponderant enrichment in genes associated with the immune modulation of pain, revealing an unanticipated contribution of neutrophils. Taking advantage of the male-enriched transcript encoding the neutrophil associated protein Lipocalin 2 (Lcn2) and using flow cytometry, we confirmed that formalin triggered the recruitment of LCN2-expressing neutrophils in the pia mater of spinal meninges, preferentially in males. Our data consolidate the contribution of female estrus cycle to pain perception and provide evidence supporting a sex-specific immune regulation of formalin-evoked pain.

Figure 1

Female mice display two types of formalin-induced pain behaviors, respectively marked by short- and long-lasting interphases. (a) Cumulative duration (left) and the duration per 5 min intervals (time-course, right) of formalin-evoked pain behavior (in seconds, s) in a small cohort of n = 8 male and n = 8 female mice. Data are presented as mean ± SEM. (*p < 0.05; **p < 0.01). (b) Individual data for the time-course of formalin-evoked pain in the females included the (a). (c) Cumulative duration (left) and time-course (right) of formalin-evoked nocifensive behavior (in seconds, s) in a large group of mice (n = 24 males and 53 females). Data are presented as mean ± SEM. (*p < 0.05). (d) K-means based subdivision of males and females; shows the percentage of subjects in each sub-group (males and outlier males and F1 and F2 females). (e) Cumulative duration (left) and time-course (right) of formalin-evoked nocifensive behavior (in seconds, s) in F1 (n = 29) and F2 (n = 24) females and in males (n = 24). Data are presented as mean ± SEM. (M-F1: ^$p < 0.05; ^$$$p < 0.001) (M-F2: ^øp < 0.05; ^øøp < 0.01) (F1-F2: *p < 0.05; ***p < 0.001). (f) Duration of the interphase in males (n = 24), F1 (n = 29) and F2 (n = 24) females (in minutes, min). Data are presented as mean ± SEM. (M-F1: ^$$$p < 0.001) (M-F2: ^øp < 0.05) (F1-F2: ***p < 0.001).

Figure 2

The duration of the interphase in females is influenced by the estrus cycle. (a) Vaginal cytology used for the identification of the different stages of the estrus cycle and the percentage of females in proestrus, estrus, metestrus and diestrus in F1 and F2 subgroups. (b) Cumulative duration (left) and time-course (right) of formalin-evoked nocifensive behavior (in seconds, s) in females in proestrus (n = 12) as compared to females in metestrus (n = 15). Data are presented as mean ± SEM. (*p < 0.05; ***p < 0.001). (c) Duration of the interphase (in minutes, min) in females in proestrus (n = 12) and females in metestrus (n = 15). Data are presented as mean ± SEM. (***p < 0.001). (d) Illustration of the experimental design for RNA-Seq preformed on the dorsal horn of the spinal cord (DHSC) of proestrus (n = 6) and metestrus (n = 6) females following formalin injection. (e) Volcano-plot representation of the RNA-Seq data, highlighting the transcripts enriched in proestrus females and those in metestrus females (FDR5).

Figure 3

Sex-related transcriptional differences in the ipsilateral DHSC following formalin administration. (a) Illustration of the experimental design for RNA-Seq preformed on the dorsal horn of the spinal cord (DHSC) of males (n = 6) and females (n = 12, including n = 6 F1/proestrus and n = 6 F2/metestrus) following formalin injection. (b) Volcano-plot representation of the RNA-Seq identified DEG between males and females (FDR5). For a matter of representation, the following DEG were not represented on the volcano plot: Xist, Tsix, Gm2223 for females and Kdm5d, Eif2s3y, Uty, Ddx3y for males. (c) Metascape functional enrichment of transcripts enriched in males and those enriched in females. (d) IPA disease and function prediction showing clusters of male-enriched transcripts associated with immune response.

Figure 4

Formalin triggers LCN2-positive neutrophils infiltration in the pia layer of spinal meninges, preferentially in males. (a) RT-qPCR analysis of Lcn2 expression levels in the ipsilateral DHSC of naïve males (n = 3) and free-cycling females (n = 3) and formalin-injected males (n = 8) and free-cycling females (n = 12). Data are presented as mean of relative expression normalized to β-actin and expressed in arbitrary units (a.u) ± SEM. (**p < 0.01). (b) Whole-mount immunostaining on L3-L5 segments of the DHSC with anti-CD31 (red) and anti-LCN2 (green) antibodies, of naïve and formalin-injected males and free-cycling females. The tissue is oriented to allow visualization of the dorsal pia mater. Representative of 2 to 3 independent experiments. Scale bars: 100 µm. (c) Fold change between formalin and naive conditions of the percentages of CD45⁺LCN2⁺ cells in the pia and dura maters and in the ipsilateral SC in males and in free-cycling females, naïve and 2H after formalin injection. Data are expressed as mean ± SEM calculated from 3 to 6 independent samples. (*p < 0.05). (d) Flow cytometry analysis of CD45⁺LCN2⁺ cells in the pia mater of naive and formalin-injected males and free-cycling females. From left to right are shown: the percentages of CD45⁺ LCN2⁺ cells within live CD45⁺ cells, the percentages of Ly6C⁺Ly6G⁺ (LCN2⁺ neutrophils) and Ly6C⁺Ly6G^- (LCN2⁺ monocytes) in the CD45⁺LCN2⁺CD11b⁺CD3^- gate. Data are expressed as mean ± SEM.