Urothelium muscarinic activation phosphorylates CBSSer227 via cGMP/PKG pathway causing human bladder relaxation through H2S production

The urothelium modulates detrusor activity through releasing factors whose nature has not been clearly defined. Here we have investigated the involvement of H2S as possible mediator released downstream following muscarinic (M) activation, by using human bladder and urothelial T24 cell line. Carbachol stimulation enhances H2S production and in turn cGMP in human urothelium or in T24 cells. This effect is reversed by cysthationine-β-synthase (CBS) inhibition. The blockade of M1 and M3 receptors reverses the increase in H2S production in human urothelium. In T24 cells, the blockade of M1 receptor significantly reduces carbachol-induced H2S production. In the functional studies, the urothelium removal from human bladder strips leads to an increase in carbachol-induced contraction that is mimicked by CBS inhibition. Instead, the CSE blockade does not significantly affect carbachol-induced contraction. The increase in H2S production and in turn of cGMP is driven by CBS-cGMP/PKG-dependent phosphorylation at Ser227 following carbachol stimulation. The finding of the presence of this crosstalk between the cGMP/PKG and H2S pathway downstream to the M1/M3 receptor in the human urothelium further implies a key role for H2S in bladder physiopathology. Thus, the modulation of the H2S pathway can represent a feasible therapeutic target to develop drugs for bladder disorders.


Carbachol-induced contraction is affected by H 2 S in human bladder strips. The urothelium
removal significantly potentiates carbachol-induced contraction (1 μ M) (* p < 0.05; Fig. 1A). This finding confirms that the urothelium contributes to bladder homeostasis by producing relaxing factors. In order to assess the possible involvement of the endogenous H 2 S in carbachol-induced contraction, human bladder strips with urothelium were contracted with carbachol (10 nM-30 μ M) in presence of aminooxyacetc acid (AOAA, 1 mM), a CBS inhibitor, DL-propargylglicine (PAG, 10 mM), a CSE inhibitor, or vehicle. AOAA but not PAG significantly increases carbachol-induced contraction (* p < 0.05; Fig. 1B). Of note, the effect of AOAA is higher compared with PAG treatment (°p < 0.05; Fig. 1B). The vehicle does not affect the concentration-dependent contraction induced by carbachol (Fig. 1B).

Carbachol increases H 2 S production in human urothelium and T24 cells.
The incubation with carbachol (0.1, 1 and 10 μ M) causes a significant increase in H 2 S generation compared with vehicle (* * p < 0.01, ***p < 0.001) in both human urothelium ( Fig. 2A) and T24 cells (Fig. 2B). To evaluate the involvement of CBS in H 2 S production induced by carbachol (1 μ M), we have incubated human urothelium as well as T24 cells with AOAA (1 mM). CBS inhibition significantly reduces the increase in carbachol-induced H 2 S production in human urothelium (°°°p < 0.001 vs carbachol; Fig. 2C) as well as in T24 cells (°°p < 0.01 vs carbachol; Fig. 2D). In order to confirm the CBS role in human urothelium, as previously demonstrated 21 , we have stably silenced CBS in T24 cells namely T24Δ CBS (Fig. 2E). T24Δ CBS produces a detectable amount of H 2 S in basal condition (Fig. 2F). When T24Δ CBS cells are stimulated with carbachol (0.1, 1 and 10 μ M) the increased H 2 S production is blunted (Fig. 2F) as compared to control cells (Fig. 2B).
Carbachol phosphorylates CBS through cGMP/PKG pathway. Firstly, we showed the presence of eNOS in T24 cells or in human urothelium (Insert in Fig 3A). Since human CBS is activated through cGMP/ PKG-dependent phosphorylation at Ser 227,21 we have evaluated whether carbachol increases cGMP in T24 cells. Carbachol stimulation (0.1, 1, 10 μ M) significantly increases cGMP levels (* p < 0.05 vs vehicle; Fig 3A). As, cGMP/PKG is downstream from the NO signal, we used L-NIO (1 μ M), as eNOS inhibitor, or KT5823 (10 μ M), as PKG inhibitor. Both significantly inhibit the increase in H 2 S production induced by carbachol in T24 cells (°°°p < 0.001 vs carbachol; Fig 3B).
Finally, western blot analysis shows that carbachol triggers a significant increase in p-CBS 227 compared to vehicle (* * p < 0.01 and * * * p < 0.001; Fig 3C,D). The involvement of cGMP/PKG pathway is confirmed by the finding that CBS phosphorylation is significantly reduced upon PKG inhibition (°°p < 0.01, °°°p < 0.001 vs carbachol; Fig 3C,D).

M1 and M3 receptors are involved in carbachol-induced increase in H 2 S production in human
urothelium and T24 cells. M receptors (M1, M2, M3, M4 and M5) are expressed as mRNA either in human urothelium or in T24 cells (Fig. 4A). In order to address the involvement of a specific subtype of M receptor in H 2 S generation, we have used selective M antagonists. Notably, the blockade of M1 or M3 receptor, with telenzepine (10 nM) or 4-DAMP (100 nM) respectively, abrogates carbachol-induced increase in H 2 S production in human urothelium (°p < 0.05 and °°p < 0.01 vs carbachol; Fig. 4B). In T24 cells, only the blockade of M1 receptor with telenzepine, significantly reduces carbachol-induced increase in H 2 S production (°°°p < 0.001 vs carbachol; Fig. 4C).

Discussion
The urinary bladder expresses all M receptor subtypes that are widely and heterogeneously distributed, in the different area of the bladder 22 . Activation of M receptors on the detrusor muscle triggers the contractile response Scientific RepoRts | 6:31491 | DOI: 10.1038/srep31491 necessary for urinary bladder voiding. However, M receptors are also involved in relaxing bladder 9,10,23 . This Janus effect (doubled edged effect) implies that different downstream signaling as well as mediators must be involved following M receptors activation within the bladder structure. Here, we have addressed the role of H 2 S as possible downstream mediator following activation of M receptors within the urothelium. Indeed, endogenous H 2 S is a signaling molecule which participates in the regulation of various physiological processes in human urogenital-tract 11,16,17,21,[24][25][26] . The choice of the urothelium stems from the finding that CBS is markedly expressed in the urothelium as compared to CSE and that its activity is specifically enhanced upon phosphorylation 21 . Indeed, CBS is phosphorylated at Ser 227 in a cGMP/PKG-dependent mechanism 21 . We have shown that the human bladder, as also reported by others, contracts to carbachol in vitro and this contractile effect is enhanced upon urothelium removal 9,10,27,28 . In addition, the inhibition of CBS, significantly increased carbachol-induced contraction in human bladder strips with intact urothelium while, PAG, a CSE inhibitor, did not significantly modify carbachol effect. This data indicates that i) urothelium dynamically contributes to pacing the contractile cholinergic stimulation and ii) endogenous production of H 2 S is involved in bladder tone. This hypothesis is further sustained by the finding that carbachol causes a significant concentration-dependent increase in H 2 S production in human urothelium. As it happened in the functional study, the inhibition of CBS blocks carbachol-induced H 2 S production in human tissue. Therefore, H 2 S is involved as downstream signaling mediator following cholinergic stimulation in urothelium. In order to further address this issue, we used the T24 cell line that mimics the human urothelium biochemical profile for what concerns the H 2 S pathway, as previously demonstrated 21 . T24 cells express all M receptor subtypes and, as expected, carbachol causes a significant enhancement in H 2 S production in T24 cells that is reversed by AOAA. Since AOAA is not selective for CBS 29 , in order to confirm the role hypothesized for CBS-derived H 2 S, we silenced CBS in T24 cells. In the T24 CBS silenced cells challenge with carbachol failed to increase H 2 S production confirming a key role for CBS-derived H 2 S. Thus, following activation of the M receptors the downstream activation of CBS is necessary in order to trigger H 2 S production. Since PKA is almost absent in human urothelium as well as in T24 cells we have investigated the possible involvement of cGMP 21 . A role for the NO/cGMP in relaxing effect specially in bladder neck and urethra has been demonstrated 30 . Here, the activation of M receptors by carbachol causes an increase in cGMP levels in T24 cells. In parallel, the western blot analysis confirmed the presence of eNOS in the urothelium. Thus, our results suggest a role for NO/cGMP in detrusor muscle dome through urothelium M stimulation. In order to evaluate the involvement of the NO/cGMP/PKG pathway following M receptor activation in urothelium we modulated eNOS or PKG activity by using the selective inhibitors, L-NIO or KT5823 respectively. Both treatments inhibited carbachol-induced H 2 S production bringing back to control levels. As expected, the increased cGMP production translated into an increased phosphorylation of CBS at Ser 227 in T24 cells. Therefore, taken together all these findings imply that following M receptor stimulation eNOS/cGMP pathway is activated and cGMP produced in turn phosphorylates CBS in a PKG-dependent manner, thereby increasing H 2 S production. In order to investigate which M receptor (M1-M5) was involved in carbachol-induced H 2 S generation, we used different selective M1 to M4 receptor inhibitors since there are no available selective M5 receptor inhibitors. In T24 cells only telenzepine, an M1 antagonist, reduced H 2 S production while in human urothelium, the blockade of either M1 or M3 receptor reversed carbachol-induced increase in H 2 S production. Thus, although all M receptors are expressed in human bladder urothelium, only M1 and M3 appear to be involved in H 2 S production most likely for the specific anatomical localization. Indeed, in human urothelium, M1 receptor is localized on basal cells, M2 on umbrella   cells, M3 and M4 homogenously and M5 with a decreasing gradient from luminal to basal cells 31 . The M1 receptor localization on basal cells that are anatomically strictly attached to detrusor muscle, suggests that H 2 S, once generated, can easily diffuse to the detrusor muscle contributing to its relaxation. Besides, our results suggest H 2 S as a possible candidate for UDRF. Indeed, previous studies demonstrated that M1 and M3 receptors stimulation induce the release of UDRF in rat bladder 23 . The identification of H 2 S as an UDRF in human bladder implies a role for this pathway in bladder tone and possibly in micturition. The urothelium impermeability is necessary for the normal bladder functioning, but at the same time represents limitations regarding the uptake of drugs. Therefore, the development of releasing-H 2 S drugs may represent a feasible therapeutic approach to bladder disorders. However our data strongly imply that the cholinergic stimulation triggers the detrusor muscle contraction but at the same time triggers the release of H 2 S by the urothelium. Therefore, it is feasible to hypothesize that in healthy condition a balance between these two signals e.g. contracting and relaxing effect contributes to bladder homeostasis. Conversely, in pathological condition, such as in overactive bladder, the detrusor contracting effect may predominate while in hyporeactive bladder the urothelium relaxing effect may prevail. To date, the knowledge of human bladder physiology is still unclear but our findings may open new frontiers in this field and may contextually offer the basis for novel pharmacological approaches in the management of bladder disorders. Human bladder strips. The bladder sections (detrusor muscle plus urothelium), obtained from patients were excised from each donor sample, and longitudinal strips were isolated. Strips were suspended in organ chambers filled with Krebs buffer (37 °C, 95% oxygen and 5% carbon dioxide) with the following composition: sodium chloride,115.3 mM; potassium chloride, 4.9 mM; calcium chloride, 1.46 mM; magnesium sulfate, 1.2 mM; potassium dihydrogen phosphate, 1.2 mM; sodium bicarbonate, 25.0 mM; and glucose, 11.1 mM (Carlo Erba, Milan, Italy). The strips were connected to isometric force transducers (model 7002, Ugo Basile, Comerio, Italy) and a tension of 0.5 g was applied. The tension changes were registered and monitored by a computerized system (DataCapsule, Ugo Basile, Comerio, Italy). The tissues were equilibrated (60 min) and standardized by repeated contractions with carbachol (1 μ M; Sigma, Milan, Italy). The strips with or without urothelium were challenged with carbachol (1 μ M). The contraction was calculated as dyne/mg tissue and expressed as mean ± SE. The results were analyzed using Student's t-test. In another setting of experiments, a concentration-response curve of carbachol (10 nM up to 30 μ M) was performed in presence of urothelium and thereafter the curve was repeated in presence of vehicle, PAG, a CSE inhibitor (10 mM, Sigma, Milan, Italy), or AOAA, a CBS inhibitor (1 mM; Sigma, Milan, Italy) after 20 min of incubation. Data were calculated as delta percentage of increase in contraction between the two repeated concentration-response curve of carbachol. Data were expressed as mean ± SE and analyzed by using one way ANOVA. p < 0.05 was considered significant.

Methods
Cell culture. The urothelial T24 cell line (American Type Culture Collection, ATCC, HTB-4) was cultured as previously described 32 .
CBS silencing. To stably silence CBS, T24 cells were seeded the day before transfection in a 60 mm plate.
Cells were transfected 24 hours later, with 2 μ g of a CBS short hairpin (sh)RNA expressing vector purchased from Santa Cruz Biotechnology (Santa Cruz Biotechnology, Santa Cruz, CA, USA). CBS silenced cells (T24Δ CBS) were selected as previously described 33 .
The CBS depletion was evaluated by western blot using a rabbit polyclonal anti-CBS (Santa Cruz Biotechnology, DBA, Milan, Italy). The cell clone with lowest expression level was selected and used for further experiments. A short hairpin non-silencing construct was used as control.
RT-PCR analyses. The mRNA level of M receptors (M1-M5) in urothelial tissue and in T24 cells was determined by using RT-semi-quantitative PCR (RT-PCR) method 34 . Total RNA was extracted from samples by using an ultrapure TRizol reagent (GibcoBRL) as directed by the manufacturer. RNA (2 μ g) was reverse-transcribed to cDNA through SuperScript III First-Strand Synthesis system (iScript TM cDNA Syntesis Kit, Biorad, Italy) according to the manufacturer's protocol in the presence of random hexamers (5 μ M), 20 U of RNasin (Promega), dNTP (1mM) for 1 h at 42 °C. RT-PCR was performed on 1 ml of the reverse transcription reaction mixture in a final volume of 50 ml with 2 U of Taq DNA Polymerase (VWR, Italy) and 1 mM of appropriate primers (Table 1). To obtain a linear amplification curves, the cDNA mixtures were subjected to denaturation at 95 °C for 1 min, annealing at 58 °C for 1 min for M2, M3 and β -actin and at 60 °C for M1, M4 and M5 and extension at 72 °C for 15 min.
Human tissues and T24 cells treatments. In order to measure H 2 S production, human urothelial samples, T24 cells or T24Δ CBS were incubated with carbachol (Sigma-Aldrich, Milan, Italy) for 5 min at different concentrations (0.1, 1 and 10 μ M). On the basis of these results, 1 μ M was chosen as the optimal concentration. Scientific RepoRts | 6:31491 | DOI: 10.1038/srep31491 Thus, in another setting of experiments, human urothelial samples or T24 cells were pre-treated with AOAA (1 mM, for 20 min), a CBS inhibitor, and then challenged with carbachol (1 μ M, for 5 min). The carbachol-induced effect was also evaluated in presence of telenzepine (10 nM, M1 antagonist), AFDX (100 nM, M2 antagonist), 4DAMP (100 nM, M3 antagonist) or PD102809 (100 nM, M4 antagonist). Finally, in order to investigate the underlying mechanism, T24 cells were pretreated with L-NIO, an endothelial nitric oxide synthase (eNOS) inhibitor (1 μ M, Tocris UK) or KT5823, a PKG inhibitor (10 μ M, Tocris UK) and thereafter stimulated with carbachol (1 μ M, for 5 min). H 2 S determination. H 2 S production was evaluated as reported by d'Emmanuele di Villa Bianca and coworkers 35 . Human urothelial samples or T24 cells were homogenized with lysis buffer containing potassium phosphate buffer (100 mM, pH 7.4), sodium orthovanadate (10 mM) and proteases inhibitors. Homogenates were added for 40 min in a reaction mixture (total volume 500 μ l) containing piridoxal-5′ -phosphate (2 mM, 20 μ l) and saline (20 μ l) or L-cysteine (10 mM, 20 μ l, Sigma-Aldrich, Milan, Italy) to measure enzyme activity i.e. H 2 S generation. The reaction was performed in parafilmed Eppendorf tubes and initiated by transferring tubes from ice to a water bath at 37 °C and carried out for 40 min. After incubation, zinc acetate (1%, 250 μ l, Sigma-Aldrich) and trichloroacetic acid (10%, 250 μ l) were added. Subsequently, N,N-dimethyl-p-phenylendiamine sulfate (DPD, 20 mM, Sigma-Aldrich) in HCl (7.2 M) and iron chloride (FeCl3, 30 mM, Sigma-Aldrich) in HCl (1.2 M) were added and optical absorbance of the solutions was measured after 20 min at a wavelength of 667 nm. All samples were assayed in duplicate and H 2 S concentration was calculated against a calibration curve of NaHS (3.9-250 μM, Sigma-Aldrich). Results were calculated as nmol/mg protein/min and expressed as mean ± SE. Data were analyzed by using one way ANOVA following by Bonferroni as post test. p < 0.05 was considered significant. cGMP measurement in T24 cells. In order to measure the cGMP content, T24 cells were treated with carbachol (0.1, 1 and10 μ M) for 5 min and then immediately frozen. T24 cells were treated with 15 μ l of 3.3 M HCl and then centrifuged (600× g for 10 min); the cGMP levels were measured in supernatants as previously reported 36,37 by using an immune assay kit (Cayman; Vinci Biochem, Vinci, Italy). Results were calculated as pmol/ml and expressed as mean ± SE. Data were analyzed by using one way ANOVA following by Bonferroni as post test. p < 0.05 was considered significant.
Western blot analysis in T24 cells. T24 cells treated with carbachol (0.1, 1 and 10 μ M, for 5 min) were analyzed by western blotting in presence or in absence of KT5823 (10 μ M), a PKG inhibitor (Tocris, UK). Protein samples (30 μ g) were resolved by 12% SDS-PAGE and analyzed as previously described 38,39 .The membranes were incubated with rabbit polyclonal anti-CBS, (1:1000 Santa Cruz Biotechnology, Heidelberg, Germany), and anti-pCBS Ser227 (1:400, PRIMM srl, Milano, Italy) 21 . In another setting of experiments T24 cells or human urothelium were blotted for anti-eNOS (1:1000, BD Transduction, CA, USA). The target protein band intensity was normalized over the intensity of the housekeeping β -actin (1:5000, Sigma-Aldrich, Milan, Italy). Data were expressed as mean ± SE. Results were analyzed by ANOVA following by Bonferroni as post test. p < 0.05 was considered significant.