Pannexin 1 channels mediate the release of ATP into the lumen of the rat urinary bladder

Abstract

Key points: ATP is released through pannexin channels into the lumen of the rat urinary bladder in response to distension or stimulation with bacterial endotoxins. Luminal ATP plays a physiological role in the control of micturition because intravesical perfusion of apyrase or the ecto-ATPase inhibitor ARL67156 altered reflex bladder activity in the anaesthetized rat. The release of ATP from the apical and basolateral surfaces of the urothelium appears to be mediated by separate mechanisms because intravesical administration of the pannexin channel antagonist Brilliant Blue FCF increased bladder capacity, whereas i.v. administration did not. Intravesical instillation of small interfering RNA-containing liposomes decreased pannexin 1 expression in the rat urothelium in vivo and increased bladder capacity. These data indicate a role for pannexin-mediated luminal ATP release in both the physiological and pathophysiological control of micturition and suggest that urothelial pannexin may be a viable target for the treatment of overactive bladder disorders.

Abstract: ATP is released from the bladder epithelium, also termed the urothelium, in response to mechanical or chemical stimuli. Although numerous studies have described the contribution of this release to the development of various bladder disorders, little information exists regarding the mechanisms of release. In the present study, we examined the role of pannexin channels in mechanically-induced ATP release from the urothelium. PCR confirmed the presence of pannexin 1 and 2 mRNA in rat urothelial tissue, whereas immunofluorescence experiments localized pannexin 1 to all three layers of the urothelium. During continuous bladder cystometry in anaesthetized rats, inhibition of pannexin 1 channels using carbenoxolone (CBX) or Brilliant Blue FCF (BB-FCF) (1-100 μm, intravesically), or by using intravesical small interfering RNA, increased the interval between voiding contractions. Intravenous administration of BB-FCF (1-100 μg kg(-1) ) did not alter bladder activity. CBX or BB-FCF (100 μm intravesically) also decreased basal ATP concentrations in the perfusate from non-distended bladders and inhibited increases in ATP concentrations in response to bladder distension (15 and 30 cmH2 O pressure). Intravesical perfusion of the ATP diphosphohydrolase apyrase (2 U ml(-1) ), or the ATPase inhibitor ARL67156 (10 μm) increased or decreased reflex bladder activity, respectively. Intravesical instillation of bacterial lipopolysaccharides (LPS) (Escherichia coli 055:B5, 100 μg ml(-1) ) increased ATP concentrations in the bladder perfusate, and also increased voiding frequency; these effects were suppressed by BB-FCF. These data indicate that pannexin channels contribute to distension- or LPS-evoked ATP release into the lumen of the bladder and that luminal release can modulate voiding function.

Figure 1. The rat urothelium expresses pannexin 1 channels

A, products of PCR amplification of pannexin genes in urothelial tissue of Sprague–Dawley (SD, left) and Long–Evans (LE, right) rats. Note the positive PCR products for Panx1 and Panx2 for SD and Panx1 and Panx3 for LE. Positive products for the housekeeping gene β-actin are also shown as a positive control. Expected product sizes are listed in the Methods. Results shown are typical of those obtained from three separate rats. B, immunofluorescence localization of pannexin 1 in whole bladder tissue sections of a SD rat. Pannexin 1 (Panx1) staining is shown upper left, whereas the umbrella cell marker cytokeratin-20 (Cyto 20) is shown upper right. The nuclear stain 4',6-diamidino-2-phenylindole is shown middle left. A colourized, merged imaged is shown middle right. Scale bar = 50 μm. Bottom: a 4× digital zoom of the area denoted by a dashed rectangle (middle right). The white line denotes the boundary of the umbrella cells as determined by cytokeratin staining. Note that pannexin 1 is expressed in the umbrella cell layer (white arrow), as well as the underlying intermediate and basal cell layers (arrowheads). Results shown are typical of those obtained from three separate rats.

Figure 2. Intravesical perfusion of pannexin channel antagonists suppresses reflex bladder activity in the rat

A and B, representative tracings of bladder cystometrograms in the anaesthetized rat taken before and after BB-FCF. C, summary of the measured intercontraction intervals during cystometry experiments involving intravesical perfusion of BB-FCF (1–100 μm; n = 6 rats) 0 μm = Krebs solution without drug. D, summary of the measured intercontraction intervals during cystometry experiments involving intravesical perfusion of carbenoxolone (1–100 μm; n = 5 rats). E, summary of the measured micturition threshold pressures during cystometry experiments involving intravesical perfusion of BB-FCF (n = 6). F, summary of the measured peak micturition pressures during cystometry experiments involving intravesical perfusion of BB-FCF (n = 6). **P < 0.05 compared to control by one-way ANOVA with Tukey's post hoc test. ##P < 0.05 compared to 100 μm BB-FCF.

Figure 3. Intravenous administration of pannexin channel antagonists does not suppress bladder reflexes

A, summary of the measured intercontraction intervals during cystometry experiments involving i.v. administration of BB-FCF (1–100 μg kg⁻¹; n = 5 rats). 0 μg kg⁻¹ = Krebs solution without drug. NS, not significant. B, summary of the measured intercontraction intervals during cystometry experiments involving i.v. administration of the purinergic antagonist PPADS (3 μg kg⁻¹; n = 4 rats). C, summary of the measured intercontraction intervals during cystometry experiments involving intravesical administration of PPADS (0.1–10 μm; n = 4 rats). 0 μm = Krebs solution without drug. **P < 0.05 compared to control by an unpaired Student's t test.

Figure 4. Luminal ATP release in response to stretch is suppressed by pannexin, but not by a connexin channel blocker

Effects of BB-FCF (100 μm; n = 5 rats), CBX (100 μm; n = 3 rats) and 18 α-GA (50 μm; n = 4 rats) on distension-evoked (0–30 cmH₂O) ATP concentrations measured in the urinary bladder perfusate. All concentrations are presented as the percentage change from non-distended (0 cmH₂O) Krebs-infused controls (n = 13). **P < 0.05 by one-way ANOVA with Tukey's post hoc test; ##P < 0.05 compared to pressure matched control (Krebs) by an unpaired Student's t test.

Figure 5. Modulators of ATP catabolism influence luminal ATP concentrations and bladder activity in response to distension

A, effects of apyrase (2 U ml⁻¹; n = 3 rats) or ARL67156 (10 μm; n = 3 rats) on distension-evoked ATP concentrations in the urinary bladder perfusate. **P < 0.05 as measured by one-way ANOVA with Tukey's post hoc test. ##P < 0.05 compared to pressure matched control (Krebs) by an unpaired Student's t test. B and C, effects of intravesical apyrase (B) (2 U ml⁻¹; n = 4 rats) or ARL67156 (C) (10 μm, n = 4 rats) on the intercontraction interval during cystometry in the anaesthetized rat. **P < 0.05 compared to pressure matched control (Krebs) by an unpaired Student's t test.

Figure 6. Release of ATP and excitation of reflex bladder activity in the LPS-treated anaesthetized rat is diminished by BB-FCF

A, inhibitory effect of BB-FCF (100 μm; n = 5 rats) on ATP concentrations in the urinary bladder perfusate following stimulation with 100 μg ml⁻¹ LPS at two different distension pressures (0 and 15 cmH₂O). **P < 0.05 compared by two-way ANOVA with Dunnett's multiple comparisons test. B, excitatory effect of LPS (100 μg ml⁻¹; n = 5 rats) on reflex bladder activity and the ability of BB-FCF (100 μm) to reverse that excitation. **P < 0.05 compared by two-way ANOVA with Dunnett's multiple comparisons tests.

Figure 7. Knockdown of pannexin 1 channels through siRNA suppresses reflex bladder activity and prevents the actions of BB-FF

A, measured intercontraction intervals during cystometry experiments involving BB-FCF (100 μm) in rats treated intravesically for 5 days with siRNA against pannexin 1 or scrambled siRNA as a control (n = 3 for each). **P < 0.05 compared to Krebs-infused, no siRNA control by two-way ANOVA with Dunnett's multiple comparisons test. ++P < 0.05 compared to Krebs-infused controls by two-way ANOVA with Dunnett's multiple comparisons tests. B, relative pannexin 1 channel expression measured by western blotting in whole rat bladders taken from the cystometrograms (A). The inset depicts sample blots. **P < 0.05 compared to no siRNA controls by an unpaired Student's t test. C, fluorescence (left) and differential interference contrast/fluorescence (right) micrographs of urinary bladder tissue from a rat instilled intravesically with liposomes containing fluorescent siRNA (red) as a transfection control. Scale bar = 100 μm.

Figure 8. Summary of hypothetical purinergic signalling in the urothelium

Stretch (or stimulation by LPS, not shown) activates pannexin channels, leading to ATP release into the lumen of the bladder by an as yet unknown mechanism. Luminal ATP can be broken down by endogenous ecto-nucleotidases (NTPDases) or, in the present study, exogenous apyrase perfused intravesically. These NTPDases can be inhibited by ARL67156, increasing ATP concentrations. Luminal ATP can act on P2X or P2Y receptors on the urothelium, resulting in the release of neurotransmitters (NT) from the basolateral surface of the urothelium, which act on afferent nerves to increase their excitability. It is possible that ATP is one of these transmitters because systemic administration of the purinergic antagonist PPADS decreases reflex bladder activity. The basolateral release of ATP appears to occur via a mechanism other than pannexin channels because systemic BB-FCF has no effect on bladder reflexes.