Sex-Specific Disruption of Distinct mPFC Inhibitory Neurons in Spared-Nerve Injury Model of Neuropathic Pain

Abstract

The medial prefrontal cortex (mPFC) modulates a range of behaviors, including responses to noxious stimuli. While various pain modalities alter mPFC function, our understanding of changes to specific cell types underlying pain-induced mPFC dysfunction remains incomplete. Proper activity of cortical GABAergic interneurons is essential for normal circuit function. We find that nerve injury increases excitability of layer 5 parvalbumin-expressing neurons in the prelimbic (PL) region of the mPFC from male, but not female, mice. Conversely, nerve injury dampens excitability in somatostatin-expressing neurons in layer 2/3 of the PL region; however, effects are differential between males and females. Nerve injury slightly increases the frequency of spontaneous excitatory post-synaptic currents (sEPSCs) in layer 5 parvalbumin-expressing neurons in males but reduces frequency of sEPSCs in layer 2/3 somatostatin-expressing neurons in females. Our findings provide key insight into how nerve injury drives maladaptive and sex-specific alterations to GABAergic circuits in cortical regions implicated in chronic pain.

Keywords: GABAergic neurons; PVIN; SOM; medial prefrontal cortex; nerve injury; pain; parvalbumin; sex-specific; slice electrophysiology; somatostatin.

Figure 1.. SNI Increases PV Excitability in PL Cortex of Male, but Not Female, Mice

(A) Mechanical allodynia at baseline and post-operative day 7 (POD7) measured in male and female sham and SNI mice. Two-way repeated-measures ANOVA: interaction, F(1, 59) = 26.87, p < 0.0001; injury, F(1, 59) = 10.11, p = 0.0024; time, F(1, 59) = 60.31, p < 0.0001. Sidak post hoc tests revealed no significant differences between males and females at baseline and POD7. *p < 0.05; **p < 0.0001. (B) Fluorescent confocal 10× image displaying distribution of PV-tdTomato neurons (PVINs) in the PL cortex (D, dorsal; V, ventral; M, medial; L, lateral). (C) Representative images of a PVIN recording in PL cortex at 4× (left) and 60× (right). (D and E) Example traces of action potential (AP) firing (D) and mean (±SEM) number of APs elicited in response to increasing step current from male and female sham mice (E) (two-way ANOVA: interaction, F(12, 408) = 5.00, p < 0.0001; sex, F(1, 34) = 1.34, p = 0.01; current, F(12, 408) = 108.9, p < 0.0001; *p < 0.05, Sidak post hoc test). (F) Raw data and boxplot representation for input resistance versus normalized soma location for total (left: Mann-Whitney U = 90, n₁ = 20, n₂ = 16, p = 0.03), L2/3 (middle: t(12) = 0.27, p = 0.79), and L5 (right: t(20) = 2.35, p = 0.03) PVINs from sham males (total, n = 20 neurons; 9 mice) and females (total, n = 16 neurons; 6 mice). *p < 0.05. (G and H) Example traces of AP firing (G) and mean (±SEM) number of APs elicited from male sham and SNI mice (H) (two-way ANOVA: interaction, F(12, 432) = 5.03, p < 0.0001; injury, F(1, 36) = 4.49, p = 0.04; current, F(12, 432) = 143.5, p < 0.0001; *p < 0.05, Sidak post hoc test). (I) Input resistance versus normalized soma location for total (left: Mann-Whitney U = 123, n₁ = 18, n₂ = 20, p = 0.10), L2/3 (middle: t(11) = 0.15, p = 0.89), and L5 (right: t(23) = 2.10, p = 0.05) PVINs from male sham (total, n = 20 neurons; 9 mice) and SNI (total, n = 18 neurons; 6 mice). *p = 0.05. (J and K) Example traces of AP firing (J) and mean (±SEM) number of APs elicited from female sham and SNI mice (K) (two-way ANOVA: interaction, F(12, 444) = 0.13, p = 0.99; injury, F(1, 37) = 0.04, p = 0.87; current, F(12, 444) = 128.1, p < 0.0001). (L) Input resistance versus normalized soma location for total (t(37) = 0.31, p = 0.76), L2/3 (t(16) = 0.71, p = 0.49), and L5 (Mann-Whitney U = 39, n₁ = 13, n₂ = 8, p = 0.37) PVINs from female sham (total, n = 16 neurons; 6 mice) and SNI (total, n = 22 neurons; 7 mice). Plus symbol indicates outlier in MATLAB. Data represent mean ± SEM.

Figure 2.. SNI Reduces sEPSCs in L5 PL PVINs from Male Mice Only

(A–D) Sample sEPSC traces (A); sEPSC amplitude (B), t(20) = 1.19, p = 0.23; sESPC frequency (C), t(20) = 4.73, **p = 0.0001; and mean cumulative distribution curves (D) (top, amplitude; bottom, inter-EPSC interval) for male (n = 12 neurons; 5 mice) and female (n = 13 neurons; 5 mice) sham mice. (E–H) Sample sEPSC traces (E); sEPSC amplitude (F), t(21) = 1.17, p = 0.26; sESPC frequency (G), t(22) = 1.58, p = 0.06); and mean cumulative distribution curves (H) for male sham (n = 12 neurons; 5 mice) and SNI (n = 12 neurons; 5 mice). (I–L) Sample sEPSC traces (I); sEPSC amplitude (J), t(22) = 0.98, p = 0.34; sESPC frequency (K), t(22) = 0.83, p = 0.41; and mean cumulative distribution curves (L) for female sham (n = 13 neurons; 5 mice) and SNI (n = 11 neurons; 4 mice). Data represent mean ± SEM.

Figure 3.. SNI Decreases Excitability of SOM Neurons in PL Cortex

(A) Mechanical allodynia at baseline and post-operative day 7 (POD7) measured in male and female sham and SNI mice. Two-way repeated-measures ANOVA: interaction, F(1, 63) = 20.39, p < 0.0001; treatment, F(1, 63) = 14.25, p = 0.0004; time, F(1, 63) = 60.29, p < 0.0001. Sidak post hoc tests revealed no significant differences between males and females at baseline and POD7. *p < 0.05; **p < 0.0001. (B) Example fluorescent confocal 10× image of a brain slice displaying distribution of SOM-tdTomato (SOM) neurons in the PL cortex. (C) Representative images of a SOM neuron recording in PL cortex at 4× (left) and 60× (right). (D and E) Example traces of action potential (AP) firing (D) and mean (±SEM) number of APs elicited in response to increasing step current from sham male and female mice (E) (two-way ANOVA: interaction, F(12, 420) = 0.52, p = 0.90; sex, F(1, 35) = 1.76, p = 0.19; current, F(12, 420) = 92.62, p < 0.0001). (F) Raw data and boxplots representation for input resistance versus normalized soma location for total (Mann-Whitney U = 144, n₁ = 19, n₂ = 17, p = 0.59), L2/3 (t(15) = 0.11, p = 0.91), and L5 t(15) = 0.62, p = 0.54) SOM neurons from sham females (total, n = 17 neurons; 8 mice) and males (total, n = 18 neurons; 7 mice). (G and H) Example traces of AP firing (G) and mean (±SEM) number of APs elicited from male sham and SNI mice (H) (two-way ANOVA: interaction, F(12, 408) = 1.03, p = 0.42; injury, F(1, 34) = 2.46, p = 0.13; current injection, F(12, 408) = 78.76, p < 0.0001). (I) Input resistance versus normalized soma location for total (t(34) = 1.83, p = 0.08), L2/3 (t(15) = 1.64, p = 0.12), and L5 (Mann-Whitney U = 31, n₁ = 8, n₂ = 9, p = 0.67) SOM neurons from male sham (total, n = 20 neurons; 7 mice) and SNI (total, n = 18 neurons; 8 mice). (J and K) Example traces of AP firing (J) and mean (±SEM) number of APs elicited from female sham and SNI mice (K) (two-way ANOVA: interaction, F(12, 384) = 0.67, p = 0.78; injury, F(1, 32) = 0.16, p = 0.69; current, F(12, 384) = 95.05, p < 0.0001). (L) Input resistance versus normalized soma location for total (Mann-Whitney U = 90, n₁ = 16, n₂ = 17, p = 0.10), L2/3 (t(16) = 2.39, p = 0.03), and L5 (Mann-Whitney U = 27, n₁ = 7, n₂ = 8, p = 0.96) SOM neurons from female sham (total, n = 17 neurons; 8 mice) and SNI (total, n = 16 neurons; 7 mice). Plus symbol indicates outlier in MATLAB. *p < 0.05. Data represent mean ± SEM.

Figure 4.. SNI Significantly Reduces Frequency of sEPSCs in L2/3 SOM Neurons of Female Mice

(A–D) Sample sEPSC traces (A), sEPSC amplitude (B) (t(13) = 1.20, p = 0.25), (C) sESPC frequency (t(12) = 0.87, p = 0.40), and mean cumulative distribution curves (D) (top, amplitude; bottom, inter-EPSC interval) for L2/3 SOM neurons from sham males (n = 7 neurons; 3 mice) and females (n = 7 neurons; 4 mice). (E–H) Sample sEPSC traces (E), sEPSC amplitude (F) (t(12) = 1.16, p = 0.27), sEPSC frequency (G) (t(13) = 0.81, p = 0.43), and mean cumulative distribution curves (H) for L2/3 SOM neurons from male sham (n = 7 neurons; 3 mice) and SNI (n = 8 neurons; 3 mice). (I–L) Sample sEPSC traces (I), sEPSC amplitude (J) (t(14) = 0.28, p = 0.78), sEPSC frequency (K) (t(12) = 2.59, p = 0.02), and mean cumulative distribution curves (L) for L2/3 SOM neurons from female sham (n = 7 neurons; 4 mice) and SNI (7 neurons; 5 mice). *p < 0.05. Data represent mean ± SEM.