Muscarinic receptors 2 and 5 regulate bitter response of urethral brush cells via negative feedback

Abstract

We have recently identified a cholinergic chemosensory cell in the urethral epithelium, urethral brush cell (UBC), that, upon stimulation with bitter or bacterial substances, initiates a reflex detrusor activation. Here, we elucidated cholinergic mechanisms that modulate UBC responsiveness. We analyzed muscarinic acetylcholine receptor (M1–5 mAChR) expression by using RT-PCR in UBCs, recorded [Ca²⁺]_i responses to a bitter stimulus in isolated UBCs of wild-type and mAChR-deficient mice, and performed cystometry in all involved strains. The bitter response of UBCs was enhanced by global cholinergic and selective M2 inhibition, diminished by positive allosteric modulation of M5, and unaffected by M1, M3, and M4 mAChR inhibitors. This effect was not observed in M2 and M5 mAChR-deficient mice. In cystometry, M5 mAChR-deficient mice demonstrated signs of detrusor overactivity. In conclusion, M2 and M5 mAChRs attenuate the bitter response of UBC via a cholinergic negative autocrine feedback mechanism. Cystometry suggests that dysfunction, particularly of the M5 receptor, may lead to such symptoms as bladder overactivity.—Deckmann, K., Rafiq, A., Erdmann, C., Illig, C., Durschnabel, M., Wess, J., Weidner, W., Bschleipfer, T., Kummer, W. Muscarinic receptors 2 and 5 regulate bitter response of urethral brush cells via negative feedback.

We have recently identified a novel cholinergic component in sensory control of bladder function, that is, a specialized cholinergic epithelial cell in the urethra, termed the urethral brush cell (UBC) (1, 2). UBCs are polymodal chemosensory cells that express both canonical bitter and umami receptors and their common downstream channel, transient receptor potential cation channel subfamily M member 5 (TRPM5). They respond to both bitter substances and glutamate with an increase in [Ca²⁺]_i. Both stimuli represent a potential danger signal in the urethra as many bacterial products have bitter quality, and glutamate (umami) facilitates bacterial growth in urine. Accordingly, these cells also respond to heat-inactivated uropathogenic Escherichia coli (2). UBCs are cholinergic, release acetylcholine upon stimulation, and are approached by sensory nerve fibers that carry nicotinic acetylcholine receptors. Bitter application into the urethral lumen reflexively triggers enhanced detrusor activity, which has been interpreted as a protective reflex, as potential hazardous content is expelled from the urethra via micturition (2, 3).

In the course of these studies, we obtained first evidence of an autocrine cholinergic feedback mechanism that regulates the sensitivity of this system. Using knockout mice for all 5 known muscarinic acetylcholine receptor (mAChR) subtypes (M1–5), we have clarified the underlying receptors and demonstrate in cystometry in awake mice that the targeted deletion of the M5 receptor is associated with signs of overactivity.

Overactive bladder syndrome (OAB) is a common disorder that affects approximately 12–16% of the population in Western countries and millions of people worldwide (4, 5). OAB symptoms are urgency, with or without urinary incontinence, usually with frequency and nocturia, and OAB is often associated with involuntary contractions of the detrusor muscle, that is, detrusor overactivity (6). Antagonists of mAChRs are drugs of choice in OAB, originally based on the concept of inhibiting detrusor contraction, which is primarily mediated via acetylcholine (7); however, as mAChR blockers are effective in the bladder-filling phase during which no contractions occur, and because they do not impair physiologic voiding, a mode of action on the sensory component is currently favored, yet still not fully elucidated.

MATERIALS AND METHODS

Animals

The generation of M1–5^−/− and M2/3^−/− mice has been described previously (8–13). Corresponding wild-type mice were used as controls. Choline acetyltransferase [ChAT(BAC)-eGFP] mice that expressed enhanced green fluorescent protein (eGFP) in UBCs (2) and C57BL/6 mice were obtained from The Jackson Laboratory (Bar Harbor, ME, USA). All urodynamic experiments were performed with adult female mice (>14 wk). Experiments were conducted in accordance with the European Communities Council Directive of November 24, 1986 (86/609/EEC), and approved by the local authorities (Administration of Giessen, Germany; No. 571_M, 572_M, 574_M, A44-48/2011, V44/2008).

Cell isolation

Cell isolation was performed as described previously (2). Isolated UBCs were identified by either eGFP-fluorescence or labeled with a TRPM5 Ab (1:125; Abcam, Cambridge, United Kingdom), followed by FITC-conjugated donkey anti-rabbit IgG (1:125; EMD Millipore, Billerica, MA, USA). TRPM5 labeling and eGFP expression matched nearly 1:1 (2). For RT-PCR analysis, UBCs were isolated with the TRPM5 Ab and magnetic beads (Thermo Fisher Scientific, Waltham, MA, USA) that were coated with goat anti-rabbit IgG (H + L; PI65-6100; Thermo Fisher Scientific), followed by harvesting by magnetic cell separation (2).

RT-PCR

Total RNA from pooled isolated cells (n = 3 samples) was extracted by using an RNeasy Kit (Qiagen, Hilden, Germany) according to the manufacturer’s protocol. RT-PCR was performed as described previously (primer sequences are given in Supplemental Table 1) (2).

Measurement of intracellular calcium concentration

Measurement of intracellular calcium concentration ([Ca²⁺]_i) was performed as described previously (2). In brief, isolated cells were loaded with the fluorescent calcium indicator, Calcium Orange AM (0.01 µg/μl; Thermo Fisher Scientific), and plated on coverslips. [Ca²⁺]_i was analyzed with a confocal laser scanning microscope (LSM 710; Carl Zeiss, Jena, Germany) during continuous superfusion (2.5 ml/min). Fluorescence intensities at the start of the recording period were set arbitrarily at 100%. Test stimuli and concentrations were denatonium benzoate (25 mM; Molekula, Munich, Germany), mecamylamine hydrochloride (0.02 mM; Sigma-Aldrich, St. Louis, MO, USA), atropine sulfate (0.002 mM; Sigma-Aldrich), tropicamide (0.001 mM; Sigma-Aldrich), methoctramine hydrate (0.001 mM; Sigma-Aldrich), pirenzepine dihydrochloride (0.001 mM; Sigma-Aldrich), 4-DAMP (1,1-dimethyl-4-diphenylacetoxypiperidinium iodide; 0.001 mM; Sigma-Aldrich), and VU 0238429 (0.005 mM; Abcam).

Cystometric analysis

Cystometric analyses were performed as described elsewhere (14). In brief, animals were anesthetized with an injection of atropine/ketamine/xylazine (0.05/100/15 mg/kg, i.p.) and meloxicam (1.5 mg/kg). Intramedic PE-10 polyethylene tubing (Becton Dickinson, Franklin Lakes, NJ, USA) was then inserted in the dome of the bladder and tunneled subcutaneously to the neck. Three days later, cystometry was performed in conscious, freely moving mice. Saline (room temperature) was infused into the bladder at a rate of 1.5 ml/h. After a stabilization phase of 15–30 min, intravesical pressure and micturition were recorded continuously. We analyzed basal pressure level, maximum detrusor pressure during micturition, threshold pressure, bladder capacity, time interval between 2 pressure peaks, micturition interval, micturition volume, and postmicturition residual volume. Residual volume was measured by discontinuing the tube between animal and infusion pump after micturition. In this way, we were able to collect and weigh the saline that had remained in the bladder.

Statistical analysis

Student’s paired/unpaired, 2-tailed Student’s t test or 1-way factorial ANOVA followed by Dunnett’s multiple comparison test was used to compare intracellular or δ maximum calcium concentration. Cytometric parameters between groups of mice was compared using 1-way factorial ANOVA, followed by Dunnett’s multiple comparison test. Throughout, a value of P < 0.05 was considered significant.

RESULTS

Calcium response of UBCs upon repetitive bitter stimulation

Repetitive stimulation of isolated UBCs with the bitter stimulus, denatonium, led to repetitive increases in [Ca²⁺]_i in both ChAT(BAC)-eGFP and C57BL/6 mice, with a small decrease in the second response (significant only in C57BL/6 mice), which was indicative of a slight desensitization; however, the response to the second stimulus was significantly increased in the presence of a muscarinic/nicotinic blocker cocktail (atropine/mecamylamine), which indicated a cholinergic negative feedback mechanism (Fig. 1A–E, H). This increase accounted for 42% in ChAT-eGFP and 44% in C57BL/6 mice compared with the immediately preceding stimulus, and, taking into account the slight loss of response that was observed at the second stimulus without receptor blockade, we observed a total increase of 47% in ChAT-eGFP and 60% in C57BL/6 mice (Supplemental Table 2). Because only 1–5 UBCs per coverslip were present in these preparations, lying approximately 100 µm apart from each other, this cholinergic negative feedback likely reflects autoinhibition rather than paracrine signaling. To identify whether this cholinergic negative feedback mechanism operates via muscarinic or nicotinic receptors, only one or the other of these inhibitors was used. The response, then, to the second stimulus was significantly increased only in the presence of atropine, but not in the presence of mecamylamine (Fig. 1F–H). This indicated a muscarinic feedback mechanism and led us to assess mAChR expression in UBCs.

Figure 1.

Negative cholinergic feedback in the calcium response of UBCs to bitter stimulation and expression of mAChRs. A–H) Recording of changes in Calcium Orange fluorescence of isolated UBCs during repetitive stimulation with denatonium (Den1 and Den2; 25 mM) with or without the addition of muscarinic (2 µM atropine) and nicotinic (20 µM mecamylamine) acetylcholine receptor blockers or a cocktail (A/M) of both between the 2 stimuli. All drugs were added under continuous flow in the chamber so that indicated concentrations were reached initially, then washed out. The y-axis depicts arbitrary units (AU) that correlate to [Ca²⁺]_i. UBCs were isolated and identified by means of eGFP fluorescence from ChAT reporter mice or by binding to a TRPM5 Ab in the case of wild-type (WT; C57BL/6) mice. A, B, D–G) Shown are recordings over time (means ± sem). C, H) Depicted are peak values after the first (Den1) and second stimulation (Den2) without or with blocker addition (A/M) between these 2 stimuli, analyzed with paired t test, and the difference between the peak responses under these conditions (Δ), analyzed with unpaired t test or 1-way factorial ANOVA followed by Dunnett’s multiple comparisons test. I) RT-PCR, agarose gel, M1 (Chrm1; 198 bp), M2 (Chrm2; 193 bp), M3 (Chrm3; 222), M4 (Chrm4; 156), and M5 (Chrm5; 180). UBCs were isolated by using magnetic beads coated with TRPM5 Ab. ^+/− RT, aliquots were processed with/without reverse transcription. *P < 0.05, **P < 0.01, ^§§P < 0.01, ^§§§P < 0.001.

Expression of mAChR in UBCs

RT-PCR revealed the expression of mRNAs that coded for all mAChRs in isolated UBCs (Fig. 1I). As previous immunohistochemical findings suggest heterogeneity within the UBC population (2), it seems likely that not all 5 mAChRs are simultaneously expressed within 1 cell.

Calcium response of UBCs to bitter in mAChR-deficient mice and selective pharmacologic intervention

Repetitive bitter stimulation in the absence and presence of cholinergic blockade was compared in mAChR-deficient mice. As in the corresponding wild-type (C57BL/6) mice, the second stimulus provoked a slightly smaller response without blockers, but an enhanced response in the presence of blockers with a significant difference between these responses in mice that lacked M1, M3, and M4 mAChRs (Fig. 2A, C, E). The enhancing effect of cholinergic receptor blockade was not observed in UBCs from mice that lacked the M2 mAChR, either alone or in combination with the M3 mAChR, nor in M5 mAChR-deficient UBCs (Fig. 2B, D, F). Taking these differences in second responses in the absence or presence of cholinergic blockade as parameters, M2 and M5 mAChR-deficient UBCs were statistically different from wild-type mice, whereas those from other mAChR-deficient strains were not (Fig. 2G).

Figure 2.

Impact of mAChR deficiency on the effect of cholinergic blockade on the calcium response of UBCs to bitter stimuli. A–F) UBCs were isolated and identified from M1–5 mAChR-deficient and M2/3 mAChR-deficient mice via TRPM5 Ab. UBCs were repetitively stimulated with denatonium (Den1 and Den2; 25 mM) with or without the addition of a cocktail (A/M) of muscarinic (2 µM atropine) and nicotinic (20 µM mecamylamine) acetylcholine receptor blockers between the 2 stimuli, and [Ca²⁺]_i was recorded as Calcium Orange fluorescence in arbitrary units (AU). Depicted are peak values after the first (Den1) and second stimulation (Den2) without or with blocker addition (A/M) between these 2 stimuli, analyzed with paired t test, and the difference between the peak responses under these conditions (Δ), analyzed with unpaired Student’s t test. All drugs were added under continuous flow in the chamber so that indicated concentrations were reached initially, then washed out. G) A 1-way factorial ANOVA, followed by Dunnett’s multiple comparison test was used to compare difference between peak responses under these conditions (Δ). *P < 0.05, **P < 0.01, ***P < 0.001, ^§§P < 0.01, ^§§§P < 0.001.

Acute pharmacologic inhibition of either M1 (pirenzepine), M3 (4-DAMP), or M4 receptors (tropicamide) had no impact on the response to subsequently applied denatonium, whereas M2 receptor inhibition (methoctramine) alone fully mimicked the enhancing effect observed with the complete blocker cocktail. Conversely, positive allosteric modulation of the M5 mAChR (VU 238429) suppressed the response to subsequently applied denatonium (Fig. 3).

Figure 3.

Impact of mAChR-selective pharmacologic intervention on the effect of cholinergic blockade on the calcium response of UBCs to bitter stimulus. We used the same technique and data presentation as in Fig. 2. All drugs were added under continuous flow in the chamber so that indicated concentrations were reached initially, then washed out. Initial concentrations of drugs were as follows: denatonium (Den1 and Den2; 25 mM), pirenzepine (0.001 mM; M1 receptor inhibitor), methoctramine (0.001 mM; M2 receptor inhibitor), 4-DAMP (0.001 mM; M3 receptor inhibitor), tropicamide (0.001 mM; M4 receptor inhibitor), and VU 0238429 (0.005 mM; positive allosteric modulator of the M5 receptor). A 1-way factorial ANOVA, followed by Dunnett’s multiple comparison test was used to compare differences between peak responses under these conditions (Δ). *P < 0.05, ***P < 0.001 (paired Student’s t test), ^§P < 0.05, ^§§§P < 0.001.

Urodynamic measurements in M1–5 mAChR-deficient mice

In all mouse strains, the basal pressure level recorded in transvesical cystometry was in the reported physiologic range between 10 and 25 cm H₂O (15, 16), with the lowest values observed in M2^−/−, M2/3^−/−, and M4^−/− mice (Fig. 4). We did not record a significant difference to wild-type mice in either parameter in M1^−/− and M4^−/− mice. M2^−/− mice demonstrated significantly lower threshold pressure, M3^−/− mice a significantly shorter peak-to-peak interval, and M2/3^−/− mice a significantly shorter peak-to-peak interval and lower peak and threshold pressure. Most parameters were altered in M5^−/− mice; significantly lowered micturition volume, bladder capacity, and shortened pressure peak-to-peak and micturition interval (Fig. 4).

Figure 4.

Urodynamic measurements in M1–5 mAChR-deficient mice. Cystometry was performed in conscious, freely moving mice. Saline (room temperature) was infused into the bladder at a rate of 1.5 ml/h. After a stabilization phase of 15–30 min, intravesical pressure and micturition were recorded continuously. We analyzed the basal pressure level (P_base), maximum detrusor pressure during micturition (P_max), threshold pressure (TP), bladder capacity (BC), time interval between 2 pressure peaks (PP), micturition interval (MI), micturition volume (MV), and residual volume (RV). A 1-way factorial ANOVA followed by Dunnett’s multiple comparison test was used to compare cystometric parameters between groups of mice. *P < 0.05, **P < 0.01, ***P < 0.001.

DISCUSSION

Here, we demonstrate that the initial step of a bladder activity–enhancing reflex, that is, bitter activation of UBCs assessed by an increase in [Ca²⁺]_i, is modulated by an autocrine cholinergic negative feedback loop as it is enhanced in the presence of a general cholinergic blocker cocktail. This situation differs from chemosensory lingual taste buds for which acetylcholine that is released from receptor cells enhances, rather than depresses, taste signaling via M3 mAChRs (17); however, in these rather complex sensory structures, the M3 receptor-mediated enhancement takes place at type III cells, which are considered to sense acidic substances, but not at type II cells that carry bitter and umami receptors (18), the receptors also expressed by UBCs (2).

In our approach, we unmasked the negative cholinergic feedback in UBCs by inhibiting all cholinergic receptors, thereby enhancing the response to the bitter stimulus, denatonium. Depending on the mode of calculation and mouse strain (ChAT-eGFP, C57BL/6), the average increase in response accounted for 42–60%. This is well in the range of increases in response to a blockade of autoinhibitory cholinergic feedback loops in peripheral cholinergic neurons that have been reported for rat myenteric plexus, rat heart, guinea pig trachea, mouse atrium, and mouse urinary bladder (19–22). This effect was lost in mice that lacked either M2 or M5 mAChRs. Accordingly, preferential M2 inhibition alone enhanced, and positive allosteric M5 modulation dampened, the bitter response. In contrast, this negative feedback loop was neither affected in mice with deficiency of any other muscarinic receptor (M1, M3, M4), nor could it be unmasked with M1-, M3-, and M4-preferring inhibitors. As UBCs are a rare cell type in the urethral epithelium, an entire murine urethra harbors approximately 380 (male) and 550 (female) UBCs (2), the paucity of available material precludes extensive dose-response curves in (ant)agonist experiments to determine K_d values, which limits the interpretation of data obtained with a single concentration. Still, the effects observed with the M2-preferring inhibitor and the M5-preferring allosteric modulator shall be ascribed to these and not to off-target effects on other mAChR subtypes, as inhibitors that preferentially address those did not demonstrate this effect. Hence, data from both inhibitor and knockout mice experiments revealed M2 and M5 receptors to be the crucial players (Fig. 5).

Figure 5.

Schematic drawing of the hypothesized mechanism. The activation of a canonical taste transduction cascade in UBCs by a bitter stimulus leads to [Ca²⁺]_i increase followed by acetylcholine release. M2 and M5 mAChRs attenuate the bitter response of UBCs via a cholinergic negative autocrine feedback mechanism.

Autocrine negative feedback loops that operate via M2 receptors are commonly observed in cholinergic neurons, including striatal neurons (23), α-motoneurons (24), and postganglionic parasympathetic neurons (25). In parasympathetic cholinergic neurons, the M4 receptor can also serve as an autoreceptor that inhibits acetylcholine release, and it is the dominant, if not sole, receptor that confers this function in the mouse urinary bladder (19). Both M2 and M4 receptors typically act via G_i/o proteins and the downstream inhibition of adenylate cyclase activity and voltage-activated Ca²⁺ channels (26), which fits to the inhibitory effect of the M2 receptor on the bitter stimulus increase in [Ca²⁺]_i that was observed in the present study.

The M5 receptor, however, is generally associated with an increase in [Ca²⁺]_i (26, 27), and, accordingly, enhances the activity of primary afferent terminals in the spinal cord, spinal glutamatergic interneurons, and somatodendrites of substantia nigra neurons (27, 28). This profile is in clear contrast to the dampening effect on stimulated [Ca²⁺]_i that was observed here in UBCs. Still, an M5 receptor–mediated inhibition of transmitter release that operates via a yet unresolved pathway has also been described for dopaminergic terminals in the striatum (27), so that the present finding of inhibitory M5 receptors is not unprecedented. The underlying signaling pathways still need to be identified.

We have previously demonstrated that urethral bitter stimulation markedly enhances detrusor activity during cystometric recordings with continuous bladder filling with saline through a reflex arc initiated by cholinergic (nicotinic) stimulation of sensory endings (2). Thus, factors that modulate the activation level of the initial sensor cell, UBCs, are expected to result in changes in the threshold setting of this reflex with an impact on bladder function. Then, in a simplistic model, the lack of the presently observed negative feedback, that is, deficiency of either M2 or M5 receptors, shall sensitize for enhanced detrusor activity. Of course, the overall outcome is determined not only by the threshold setting at the sensor cell, but also by the occurrence of muscarinic receptor subtypes along the entire pathway, including sensory neurons, spinal cord, parasympathetic neurons, and the bladder. This is particularly relevant for the M2 receptor, which, in addition to its role at UBCs, is prominently involved in sensory processing in the dorsal horn (28) and is abundantly expressed by the detrusor smooth muscle (29). The observed marked reduction in maximum detrusor pressure during micturition in the M2/M3 receptor double knockouts most likely reflects a direct effect at the detrusor, which is consistent with previous reports on single and double M2 and M3 receptor–deficient mice using cystometry and isolated bladders (15, 30, 31). Interference with the muscarinic regulation of ACh release from nerve terminals in the detrusor had no effect on cystometrically recorded parameters as mice with genetic deficiency of the M4 receptor, which mediates this function in the mouse bladder (19, 32), were indistinguishable from controls. Likewise, we did not observe functional alterations in deficiency of the M1 receptor, which facilitates the excitation of pelvic cholinergic neurons (33).

Of note, M5 receptor knockout animals exhibited the most extensive alterations in cystometry among the tested strains. Reduced bladder capacity and micturition volume, along with shortened micturition and pressure peak-to-peak intervals, are signs of overactivity that are compatible with the disinhibition of an activating input that originates from UBCs. Still, although its distribution is generally much more restricted than that of any other mAChR, it is not strictly limited to UBCs. In the mouse lumbar spinal cord, M5 receptor activation enhances glutamate release from afferent terminals (28, 34), which is also compatible with the observed phenotype in cystometry; however, with respect to bladder afferents, retrograde tracing experiments and RT-PCR analysis failed to detect M5 receptor expression in this specific population, despite readily detecting the M2, M3, and M4 subtypes (35), and the overall effect of mAChR activation on mouse bladder afferent neurons is inhibitory rather than excitatory (36). Lastly, M5 receptor mRNA is also expressed in the bladder wall, and the total binding of muscarinic agonists is reduced in the bladder of M5 knockout mice (37, 38). The function of M5 receptors in the bladder is unknown; they neither play a role in regulating stimulus-induced ACh release from nerve terminals, nor do they have a role in cholinergic detrusor contraction (30, 32). Collectively, disinhibited UBCs seem to represent likely candidates to initiate signs of bladder overactivity in the general absence of M5 receptors, but cell type–specific deletion will be required to finally determine the crucial cellular element(s). In any case, these data are the first unraveling of a role for the M5 receptor in bladder physiology, and its activation may be considered and explored as a new line of intervention. The present study identifies a cholinergic autocrine negative feedback mechanism in the bitter response of UBCs, driven by the M2 and M5 muscarinic receptor subtypes. Cystometric recordings from awake mice, in which this mechanism is disturbed as a result of M5 receptor deficiency, revealed lowered micturition volume and enhanced frequency, which suggests that dysfunction of this system may lead to such symptoms as bladder overactivity.

Supplementary Material

This article includes supplemental data. Please visit http://www.fasebj.org to obtain this information.

Glossary

ChAT

choline acetyltransferase

eGFP

enhanced green fluorescent protein

mAChR

muscarinic acetylcholine receptor

met

methoctramine hydrate

OAB

overactive bladder syndrome

TRPM5

transient receptor potential cation channel subfamily M member 5

UBC

urethral brush cell

AUTHOR CONTRIBUTIONS

K. Deckmann, T. Bschleipfer, and W. Kummer designed the research, analyzed data, performed statistical analysis, obtained funding, and drafted the manuscript; K. Deckmann, A. Rafiq, C. Erdmann, C. Illig, and M. Durschnabel performed research; J. Wess and W. Weidner contributed administrative, technical, and material support; and W. Weidner and W. Kummer supervised work.