Central Control Circuit for Context-Dependent Micturition

Abstract

Urine release (micturition) serves an essential physiological function as well as a critical role in social communication in many animals. Here, we show a combined effect of olfaction and social hierarchy on micturition patterns in adult male mice, confirming the existence of a micturition control center that integrates pro- and anti-micturition cues. Furthermore, we demonstrate that a cluster of neurons expressing corticotropin-releasing hormone (Crh) in the pontine micturition center (PMC) is electrophysiologically distinct from their Crh-negative neighbors and sends glutamatergic projections to the spinal cord. The activity of PMC Crh-expressing neurons correlates with and is sufficient to drive bladder contraction, and when silenced impairs micturition behavior. These neurons receive convergent input from widespread higher brain areas that are capable of carrying diverse pro- and anti-micturition signals, and whose activity modulates hierarchy-dependent micturition. Taken together, our results indicate that PMC Crh-expressing neurons are likely the integration center for context-dependent micturition behavior.

Keywords: Barrington's nucleus; bladder; corticotropin-releasing hormone; medial pre-optic area; micturition; pons.

Figure 1.. Context-dependent territorial marking correlates with PMC activity

A, Schematic (top) of behavior setup and examples of urine marks deposited (bottom) demonstrating the context-dependency of territorial marking. Pair-housed adult males are separated and tested for 2hr in response to olfactory stimuli: female urine or saline (solid or dotted circles). Social dominance (crown) is determined at the end of the 2hr period. B, Quantification of urine marks deposited in each context. Black bars, mean. *p<0.05, ns: not significant, two-tailed Mann-Whitney U test. C, Probability distribution of distances between urine-marked pixels and stimulus center for the four contexts with mean (solid lines) and SEM (shaded areas). The olfactory stimulus spot at cage center has a radius of 1.8 cm (dotted line). D, Example images (left) and quantification (right) of c-fos protein in the brainstem nucleus PMC in relation to micturition. left, PMC (dashed line) is identified by its location relative to locus coeruleus (LC), highlighted by immunostaining against tyrosine hydroxylase (TH, magenta). These are representative images from a subordinate male in response to saline (top, red border) and from a dominant male in response to female urine (bottom, blue border); right, Numbers of c-fos+ cells and urine marks positively correlate for each mouse (color scheme as in C). Scale bar, 100 μm.

Figure 2.. Crh+ neurons of the PMC are glutamatergic and project to the spinal cord

A, Schematic (left) and fluorescently labeled coronal section (right) of the pons illustrating the location of the PMC next to the LC. The PMC (arrowheads) are identified by tdTomato fluorescence (magenta) from a Crh^ires-Cre::tdTomato mouse, whereas immunostaining against TH (green) highlights the LC. Scale bar, 100 μm. B, Higher magnification image of the PMC from a section of a Crh^ires-Cre::tdTomato mouse immunostained for NeuN (green). Scale bar, 100 μm. C, In situ hybridization of Crh (magenta), Vglut (Slc17a7, Slc17a6, Slc17a8, green) and Gad (Gad1 and Gad2, red) mRNAs reveals that a vast majority of cells expressing Crh also express vesicular glutamate transporters but not GABA synthetic enzymes. Scale bar, 100 μm. The boxed region from the overlay is shown enlarged in inset (Scale bar, 10 μm). D, Quantification of detected mRNA copies per cell for Crh vs. Vglut (left) and Crh vs. Gad (right) showing a cell-by-cell correlation between the expression of Crh and Vglut but not with that of Gad. Inset: ROC analysis shows that Crh copy number is a good classifier of Vglut vs. Gad cell types (area under curve=0.84). The true positive rate (TPR) vs. false positive rate (FPR) of detecting a Vglut neuron based on a sliding threshold of Crh copy number is shown. E, Percentages of Crh+ (magenta) and Crh− neurons (black) co-labeled by Vglut or Gad. F, Schematic of AAV-DIO-PLAP under the CAG promoter in a Cre-dependent manner (left). This virus, packaged as serotype 2.9, was injected into the PMC of Crh^ires-Cre mice (middle, arrowhead), resulting in PLAP enzymatic activity (visualized by purple precipitate) in the PMC with axons found bilaterally in the spinal cord at the sacral parasympathetic nuclei (right, arrows). Scale bars, 100 μm.

Figure 3.. Activation of Crh+ neurons in the PMC induces bladder contraction

A, left, Expression of ChR2-tdTomato or tdTomato alone (control) is induced virally in Crh+ neurons of PMC and a fiber optic is implanted to deliver 473 nm light to the PMC. Bladder pressure is monitored via an acutely implanted catheter. right, Example bladder pressure traces from a control mouse expressing tdTomato in PMC Crh+ neurons during delivery of light pulses (blue bars indicate 20 Hz stimulation with 15 ms light pulses for 5 s, repeated every min). B, Example bladder pressure traces from a mouse expressing ChR2-tdTomato in PMC Crh+ neurons with the same light stimulation protocol. C, top, Overlay of 30 s of bladder pressure traces surrounding light delivery (blue bar, repeated every min). bottom, same traces as in top panel shown in a heatmap (with warmer color indicating higher bladder pressure) sorted based on the amplitude prior to light delivery. D, Bladder pressure of mouse shown in C at laser onset and 5 s later (end of laser pulses, blue circles) compared to randomly chosen time points in the same recording (grey dots). E, Pressure ratios (defined as the ratio of bladder pressure at the end of laser pulses over the pressure at laser onset, averaged across trails for each mouse) are higher in AAV-DIO-ChR2-tdTomato injected mice compared to AAV-DIO-tdTomato. *p<0.05, unpaired t test.

Figure 4.. Ca²⁺ influx into PMC Crh+ neurons is correlated with bladder contraction and with micturition

A, Expression of GCaMP6s is induced virally in Crh+ neurons of PMC and a fiber optic coupled to a photomultiplier tube (PMT) is implanted to record fluorescence arising from Ca²⁺ influx into PMC Crh+ neurons. Bladder pressure is monitored via an acutely implanted catheter. B, Example bladder pressure (top, black) and time-locked fluorescence (bottom, red) traces. Micturition events are denoted with *. C, Overlay of 20 s segments of bladder pressure traces surrounding onset of bladder contraction (time = 0 s) with the averaged transient in red. D, left, Cross-correlation between bladder pressure and fluorescence transients from the mouse as shown in B and C (red) compared to shuffled data and, right, summary of cross-correlation coefficients. *p < 0.05, two-tailed Mann-Whitney U test. E, Behavioral arena for real-time tracking of mouse position, micturition, and fiber photometry in awake behaving male of Crh^ires-Cre mice injected with AAV-DIO-GCaMP6s. F, top, Example fiber photometry trace from a male mouse expressing GCaMP6s in PMC Crh+ neurons, overlaid with black dotted bars indicating initiation of 3 micturition bouts. middle, Corresponding frames from mouse (middle) and urine (bottom) tracking videos at micturition initiation. G, Averaged fiber photometry signals aligned to the initiation of each micturition event at t=0 (n=4 mice, 75 micturition events, red), or to shuffled micturition events (black). Shaded area represents SEM. H, Heatmap of individual events from C aligned to micturition, with warmer color indicating higher Z-score of ΔF/F photometry values.

Figure 5.. Silencing Crh+ PMC neurons impairs micturition and urine marking behavior

A, Expression of hM4Di-mCherry is induced virally in Crh+ neurons of PMC. After 3 weeks of recovery and viral expression (the last week of which mice are habituated to unrestrained mock IP injection), each mouse was randomly assigned to receive CNO or saline IP injection and then returned to its home cage. 25 min after injection, mice were transferred to a test cage and the micturition pattern images after 2 hr, as in Figure 1. In the subsequent session, animals were tested again, switching delivery of CNO or saline. The testing was repeated again, with each animal exposed once more to CNO and once more to saline on different days. B, Example micturition pattern from a single mouse with 2 trials each of CNO and saline IP injection. C, Number of urine marks deposited (left) and total area on the filter paper marked by urine (right) were both reduced in CNO trials compared to saline trials. *p < 0.05, **p < 0.01, Wilcoxon matched-pairs signed rank test. D, Example image (left) and quantification (right) of hM4Di-mCherry expression in relation to changes in micturition behavior. left, representative image of hM4Di-mCherry expression in Crh+ PMC neurons (green, Crh^ires-Cre::ZsGreen mouse); right, urine marks ratio, defined as the difference in number of urine marks in saline (S) and CNO conditions divided by their sum (C-S)/(C+S), positively correlates with the number of bilateral Crh+ PMC cells expressing with hM4Di-mCherry. Scale bar, 50 μm.

Figure 6.. Crh+ PMC neurons receive converging inputs from multiple brain regions

A, Schematic of rabies-based monosynaptic retrograde transsynaptic tracing from Crh+ PMC neurons. PMC of Crh^ires-Cre mice was transduced unilaterally with 2 Cre-dependent AAVs encoding TVA-mCherry fusion protein (TVA-mCh) and rabies virus glycoprotein (RG), followed 3 weeks later by EnvA-pseudotyped, glycoprotein-deleted (ΔRG) EGFP expressing rabies virus (RbV). After the rabies injection, the front part of the brains were embedded in agarose, imaged under serial 2-photon tomography (STP) system, reconstructed in 3D, and registered to a reference atlas for analysis. B, Brainstem sections containing the PMC were manually sliced and examined to identify starter cells in the injection site and to detect potential viral leak into nearby areas. Example starter cells in PMC infected with both AAVs and RbV (yellow, enlarged in the inset). Scale bar: 1 mm, 125 μm in the inset. C, Representative coronal sections of RbV labeled cells displayed from anterior to posterior. RbV-EGFP expressing cells reveal candidate areas projecting directly to Crh+ neurons in PMC. Approximate distances from the bregma are shown above the top panels. Regions inside the white dotted boxes are magnified and displayed in the bottom panels. MO: Motor Cortex, ORB: Orbitofrontal Cortex, ACA: Anterior Cingulate Cortex, PL: Prelimbic Cortex, LS: Lateral Septum, SS: Somatosensory Cortex, BST: Bed Nuclei of Stria Terminalis, MPOA: Medial Preoptic Areas, PVH: Paraventricular Hypothalamic Nucleus, LHA: Lateral Hypothalamus, ZI: Zona Incerta, CEA: Central Amygdalar Nucleus, PH: Posterior Hypothalamus, PAG: Periaqueductal Gray, SCm: Superior Colliculus (motor related), MRN: Midbrain Reticular Nucleus. Scale bar: 500μm in the upper panels, 250μm in the lower panels. The images in the top row are shown on the same gray scale. The contrast of each image in the bottom row has been adjusted to highlight cell bodies. D, The distribution of candidate neurons presynaptic to Crh+ PMC neurons, shown as percentage of the total labeled cells with each group (Cortex, Cerebral Nuclei, Hypothalamus, and Midbrain). Mean ± SEM.

Figure 7.. GABAergic MPOA neurons modulate rank-dependent micturition patterns

A, left, Expression of hM4Di-mCherry is induced virally in MPOA GABAergic cells (defined by expression of vesicular GABA transporter, VGAT, encoded by Slc32a1); right, Representative expression pattern of hM4Di-mCherry in MPOA. Scale bar: 1 mm. B, Example of urine marks deposited from a pair of co-housed adult males (top: dominant; bottom: subordinate) demonstrates modified micturition pattern in CNO trials (right) compared to saline trials (left). C, Summary data show that in CNO trials the number of urine marks deposited decreases (left), total area on the filter paper marked by urine is unchanged (center), and the average urine mark size increases (right). *p<0.05, **p<0.01, ns, not significant, Wilcoxon matched-pairs signed rank test. D, Inhibiting MPOA VGAT+ cells (CNO trials) eliminates social-rank dependent differences in number of urine marks (left) and average urine mark size (right). **p<0.01, ns: not significant, two-tailed Mann-Whitney U test.