PIEZO2 and perineal mechanosensation are essential for sexual function

Abstract

Despite the potential importance of genital mechanosensation for sexual reproduction, little is known about how perineal touch influences mating. We explored how mechanosensation affords exquisite awareness of the genitals and controls reproduction in mice and humans. Using genetic strategies and in vivo functional imaging, we demonstrated that the mechanosensitive ion channel PIEZO2 (piezo-type mechanosensitive ion channel component 2) is necessary for behavioral sensitivity to perineal touch. PIEZO2 function is needed for triggering a touch-evoked erection reflex and successful mating in both male and female mice. Humans with complete loss of PIEZO2 function have genital hyposensitivity and experience no direct pleasure from gentle touch or vibration. Together, our results help explain how perineal mechanoreceptors detect the gentlest of stimuli and trigger physiologically important sexual responses, thus providing a platform for exploring the sensory basis of sexual pleasure and its relationship to affective touch.

Fig. 1.. Behavioral sensitivity of mice to perineal touch and underlying anatomy.

(A to C) Reaction of mice to punctate touch (A) wild-type hind-paw, (B) wild-type perineum, and (C) Piezo2^Hoxb8 perineum. (Left) Example responses for individual mice (points and thin lines; four males and four females) and mean (solid lines) to a series of calibrated von Frey filaments (grams, each tested 10 times per mouse). (Middle) Quantitation of von Frey threshold (≥5/10; n = 12 males and 12 females). Thresholds are different between all three groups [one-way analysis of variance (ANOVA) on ranks P < 0.001]. Wild-type females exhibited a lower perineal touch threshold than that of males (Mann-Whitney t test; P < 0.0001); there were no significant differences in other responses (supplementary materials, statistical reporting). (D and E) Triple-color retrograde CTB tracing from the perineum (cyan), prepuce (yellow), and glans (magenta) showing (D) cell bodies of lumbar-sacral sensory neurons in the DRG and (E) termini in the dorsal spinal cord. The dotted line indicates approximate extent of dorsal horn. In (E) and (F), n = 4 mice. Scale bars, 100 μm. (F) Anatomy of sensory ending of Piezo2-expressing sensory neurons in the perineum. (Inset) A magnified view of a single hair (boxed) highlighting prominent lanceolate and circumferential endings (n = 2 males and 1 female). Scale bar, 50 μm.

Fig. 2.. Functional characterization of perineal mechanoreceptors and role of PIEZO2.

(A to C) Heatmaps representing calcium (GCaMP6f) responses to (left) repetitive application of naturalistic stimuli and (right) graded von Frey stimulation. LTMRs and HTMRs are separated, and relative fluorescence changes (DF/F) are colored as indicated. Scale bar, 10 s. (A) Wild-type hind-paw, n = 4 mice. (B) Wild-type perineum, n = 4 mice. (C) Piezo2^cKO perineum, n = 6 mice. Additional analysis is provided in figs. S3 and S4. (D) Spatial activity maps of control and Piezo2^cKO neurons to von Frey filaments. Scale indicates response intensity. Scale bar, 100 μm. (E) Quantitation of von Frey responsive neurons in control mice (gray), Piezo2^cKO mice (red), and response profile of control HTMRs (black) (mean ± SEM, n = 8 control mice, n = 6 Piezo2^cKO mice). Piezo2^cKO mice had fewer von Frey responsive neurons at all filament strengths (Mann Whitney U test; P < 0.0087).

Fig. 3.. A role for PIEZO2-dependent perineal mechanosensation in mating.

(A) Physiological responses of male mice to perineal stimulation with transparent soft tubing. Penile protrusion was scored for two sets of 10 trials. Bars indicate mean ± SEM, and points indicate individual responses. Control versus Piezo2^Hoxb8 mice and Piezo2^Pvalb versus Piezo2^Scn10a mice were different (Mann-Whitney U test; P < 0.0001; n = 18 control mice; n = 10 Piezo2-deleted mice). (B) Representative whole-mount ISH of sacral ganglion showing faithful recombination [tdTomato (TdT); magenta] of Scn10a-Cre mouse in Scn10a (green) neurons; >90% (670 of 738) TdT cells expressed Scn10a (n = 3 ganglia). Scale bar, 50 μm. (C) Example ISH of sacral ganglion section probed for Th (magenta), Scn10a (green), and Piezo2 (cyan), illustrating expression of Scn10a and Piezo2 in cLTMRs identified with Th (n = 6 ganglia). Scale bar, 50 μm. (D) Anatomy of sensory ending of Scn10a-expressing sensory neurons in the perineum (maximum projection, full-stack). (Right) Magnified and focal views of single hairs (boxed at left), highlighting lanceolate endings of Scn10a-Cre–labeled neurons (n = 2 male mice). (E) Successful mating scored by vaginal plugs (n = 10 mice). Differences are significant for C57Bl/6 versus Piezo2^Hoxb8 mice (P = 0.0031) and Piezo2^Pvalb versus Piezo2^Scn10a mice (P = 0.0325) (Fisher’s exact test, two-tailed). (Right) Mating success for female and male Piezo2^Hoxb8 mice with C57Bl/6 partners. (F) Representative ethogram plots showing sexual motivation of three isogenic pairings of C57Bl/6 and 3 Piezo2^Hoxb8 mice: social interaction (gray), anogenital investigation (pale blue), and mounting attempts (red).