PIEZO2 in sensory neurons and urothelial cells coordinates urination

Abstract

Henry Miller stated that "to relieve a full bladder is one of the great human joys". Urination is critically important in health and ailments of the lower urinary tract cause high pathological burden. Although there have been advances in understanding the central circuitry in the brain that facilitates urination^1-3, there is a lack of in-depth mechanistic insight into the process. In addition to central control, micturition reflexes that govern urination are all initiated by peripheral mechanical stimuli such as bladder stretch and urethral flow⁴. The mechanotransduction molecules and cell types that function as the primary stretch and pressure detectors in the urinary tract mostly remain unknown. Here we identify expression of the mechanosensitive ion channel PIEZO2 in lower urinary tract tissues, where it is required for low-threshold bladder-stretch sensing and urethral micturition reflexes. We show that PIEZO2 acts as a sensor in both the bladder urothelium and innervating sensory neurons. Humans and mice lacking functional PIEZO2 have impaired bladder control, and humans lacking functional PIEZO2 report deficient bladder-filling sensation. This study identifies PIEZO2 as a key mechanosensor in urinary function. These findings set the foundation for future work to identify the interactions between urothelial cells and sensory neurons that control urination.

Extended Data Figure 1.

a, FISH in bladder tissue with probes against Krt20 (green) and Piezo1 (white). DAPI in blue. b, FISH in bladder tissue with probes against Krt20 (green) and Tmem63b (white). DAPI in blue. c, z-projection of the standard deviation of responses from genital pinch in WT and d, Piezo2^cKO DRG. e, quantification of peak responses during pinch shown as percent of baseline (each data point is one cell). N=3 DRGs, 40 cells for WT, 4 DRGs and 69 cells for Piezo2^cKO DRGs.

Extended Data Figure 2.

a, HoxB8^Cre;Ai9 bladder tissue, fixed, frozen and mounted to show tdTomato (red) throughout the tissue, labeled with DAPI (blue). Scale is 100 μm. Expression was evaluated in two mice. b, Example pressure and urethra activity traces from three wildtype males and c, three HoxB8^Cre;Piezo2^f/f knockout male littermates. d, heatmap of individual bladder contraction events in wildtype and e, knockout male mice, with corresponding urethra activity below in f and g respectively. h, Bladder contraction intervals for males. i, bladder pressures five seconds before peak contraction for males. Note: 1200 s was the length of one recording. These dots represent recording periods where the animal had no successful urination events. j, Total bladder pressure for males and k, sum of urethra activity during bladder contractions. N = 6 males per group. P<0.0001 for graphs in h, i, j and k, two-sided Student’s t test with Welch’s correction. l, body weights from a subset of mice whose bladder weights are shown in Figure 2t, and m, bladder weights from animals in l, shown as a percentage of body weight. Red horizontal lines indicate means, vertical red bars indicate +/− standard deviation (shown where possible).

Extended Data Figure 3.

a, UPKII^Cre; Ai9 bladder tissue fixed, frozen and mounted to show tdTomato (red) throughout the urothelium, labeled with DAPI (blue). Expression was evaluated in two mice. b, SNS^Cre: Ai9 bladder tissue fixed, frozen and mounted to show tdTomato (red) is not present. Expression was evaluated in two mice. Thin cryosections made neuronal endings difficult to visualize. Scale: 200 μm, applies to a and b. c, SNS^Cre: Ai9 DRG tissue showing tdTomato (red) in the majority of neurons, and d, a cell backlabled with CTB-Alexa 488 injected into bladder. e, merge of c and d, DAPI in blue. 9/9 backlabeled bladder cells analyzed from two mice were tdTomato positive. f, Quantification of freshly excised bladder weights from four UPKII^Cre; Piezo2^f/f knockout and wildtype littermates. Age-matched littermates were 10–11 months old, which could account for greater variability. g, Bladder weights from age-matched SNS^Cre;Piezo2^f/f knockout mice and wildtype littermates, 7–8 months old. Red lines indicate mean values.

Figure 1.. Human PIEZO2-deficient subjects have urinary dysfunction.

Patient numbers correspond to those in Supplementary Table 1. Grey indicates a neutral or non-pathological answer. Urinary frequency information is scored differently than other questions, and is color coded by the pathological score assigned to the answer in the questionnaire. Asterisk indicates an unanswered question. Unless otherwise noted, answers follow the scale: grey: never (no pathology), blue: less than half of the time (pathology score of 1), yellow: half of the time (pathology score of 2), orange: more than half of the time (pathology score of 3), red: every day or every night if nighttime is indicated in question (pathology score of 4). *Question not answered. Ŝubject answered 2X per day during clinical interview.

Figure 2.. The lower urinary tract expresses Piezo2, and sensory neurons require PIEZO2 for detecting low-pressure bladder filling.

a, DRG neurons were retrogradely labeled using CTB-488. (cyan, left) fluorescent in situ hybridization (FISH) of DRGs with probes targeting Piezo2 (magenta, middle). Arrowheads point to Piezo2+ bladder neurons. Scale: 50 μm. Tracing experiment was repeated using three mice, N=22–36 cells analyzed per animal. b, FISH of bladder, probes against Krt20 (green) and Piezo2 (magenta). DAPI in blue. Arrowheads point to Piezo2+ umbrella cell examples. Scale: 50 μm. RNAscope was performed on three bladders, with two technical replications. Analysis was performed on 80–117 nuclei per bladder. c, Image z-stack from GCaMP6f^+/+ control mouse S1 DRG during bladder fill. d, Example image from Piezo2^cKO mouse S1 DRG during bladder fill. e, Count of cells responding to low- (black) and high-pressure (red) stimuli in WT (N=3 animals) and KO (N=4 animals) DRG. f, Example pressure trace from a wildtype DRG and g, knockout DRG. Stimuli were interleaved during recording, but are shown sorted low to high, hence the discontinuous line. Data below these graphs is sorted together with the respective pressure peaks. h, i, Average percent change in calcium fluorescence for all responding cells during the pressure peaks shown above in f and g, respectively. j, Calcium traces for individual wildtype cells that responded to pressure stimuli in f (N=17). Scale for cell number applies to j-m. k, Calcium traces from all knockout cells responding to pressure stimuli shown in g (N=6). Each cell’s responses are shown on the same horizontal line. Cells are sorted by cumulative response to the four lowest-pressure stimuli. l, m, Maximum calcium response for the cells shown above, 1 s after pressure peak.

Figure 3.. PIEZO2 is required for efficient micturition reflexes.

a, Example pressure traces from three female WT mice and b, three female HoxB8^Cre;Piezo2^f/f KO mice during continuous bladder filling. Scale in a also applies to b. c, Bladder contraction intervals (P<0.0001). d, bladder pressure five seconds before contraction peaks (P<0.0001). e, Heatmaps showing bladder contractions for six WT female mice and f, five HoxB8^Cre;Piezo2^f/f female KO mice. Color scale in e also applies to f. Each row represents bladder pressure during a single micturition event, with peaks aligned at 0. Arrows mark where data from one animal ends and another begins. g, Peak bladder pressures (P<0.0001) and h, area under the curve for bladder contractions (P<0.0001). i, Heatmap showing urethra activity, with each row from corresponding to bladder contraction events in e. j, Urethra activity from the HoxB8^Cre;Piezo2^f/f KO bladder contraction events shown in f. Scale in i also applies to j. k, Urethra activity during micturition (P<0.0001). l, Void volume measurements (P=0.03). P values from Student’s t-tests with Welch’s correction. N=6 WT and N=5 KO female mice for c through l. N=10–29 bladder contractions analyzed per mouse. m, Urination patterns of five WT mice and n, five HoxB8^Cre;Piezo2^f/f KO mice. o, Quantification of urine in the middle 50% of the cage (P=0.0001). N=11 female mice per group. p, Mice spend the same amount of time in cage center. q, H&E staining from WT and r, KO bladder sections, from littermates, 6–7 months, scale 100 μm. Muscle layer is marked by vertical lines. Scale for q and r is 200 μm. s, Bladder muscle wall thickness (N=5, P=0.016) and t, total bladder weights (N=9 and 8, P=0.0002). For o, s and t, Mann-Whitney test, all others are two-sided Student’s t-tests with Welch’s correction. Red lines indicate mean +/− standard deviation.

Figure 4.. PIEZO2 functions in both bladder urothelium and sensory neurons.

a-h, Cystometry data from UPKII^Cre;Piezo2^f/f animals, N=5 WT and 4 KO female mice, N=18–49 bladder contractions analyzed per mouse.and i-p, SNS^Cre;Piezo2^f/f animals, N=3 WT and 3 KO female mice, 11–24 contractions per mouse. Cartoons in the top right depict the lower urinary tract, with Piezo2 KO tissue in red. a, Intervals between bladder contraction voids (P<0.0001) and b, bladder pressures five seconds before peak contraction (P=0.001) in UPKII^Cre;Piezo2^f/f knockout mice and wildtype littermates. c, Bladder pressure events during continuous filling cystometry in wildtype and d, UPKII^Cre;Piezo2^f/f knockout mice. Color scale in d also applies to c. e, Urethra activity recorded during the bladder contraction events shown in c. f, Urothelial Piezo2 KO urethra activity during bladder contraction events shown in d. g, Bladder pressure during micturition events (P<0.0001) and h, urethral reflex responses during micturition (P=0.03). i, Intervals between bladder contractions in SNS^Cre;Piezo2^f/f and wildtype littermates (P=0.002). j, Bladder pressures five seconds before peak contraction. k, Bladder pressure events during continuous filling cystometry in wildtype and l, SNS^Cre;Piezo2^f/f knockout mice. Color scale in l also apply to k. m, Urethra activity recorded during the bladder contraction events shown in k. n, Urethra activity during bladder contraction events shown in l. o, Bladder pressure during micturition events (P=0.004) and p, urethral reflex responses during micturition (P<0.0001). Red horizontal lines indicate means, vertical red bars indicate +/− standard deviation. All P values are from two-sided Student’s t test with Welch’s correction. * is P≤0.05, ** is P≤0.01, *** is P≤0.001, **** is P≤0.0001.