Stretch of the bladder mucosa, direct imposition of physical stresses on urothelial cells or local addition of muscarinic receptor agonists such as carbachol releases several modulators, including ATP.  This phenomenon has been proposed as a sensory transduction system to monitor bladder filling, as released ATP may then activate nearby afferent nerves [1].  In addition, ATP release is greater from tissues associated with pathologies such as overactive bladder (OAB) or bladder pain syndrome (BPS) where it may underlie symptoms of increased urgency.  Several features of urothelial ATP release that remain unclear would be useful to evaluate the pathophysiological basis of abnormal sensory responses arising from the bladder wall.  These include:  i) is ATP release dependent on the extent of bladder wall stretch or the change of wall tension as a result of stretch?  Bladders with increased fibrosis are stiffer (i.e. a given stretch will generate greater wall tension) and often associated with increased urgency. Thus, does a given stretch of a fibrotic bladder wall during filling lead to greater ATP release.  ii) what is the relative quantity of ATP released by physical stretch or carbachol?  This is relevant as antimuscarinic agents could significantly modulate stretch-induced ATP release.  We tested two hypotheses: i) wall tension rather than the extent of stretch determines ATP release; ii) carbachol- and stretch-induced ATP release are comparable.  These were tested in bladder wall preparations from young and aged mice; the latter used because fibrosis is greater in aged mice.

Experiments were performed on detrusor strips from young (9-12 weeks) or aged (24 months) C57/BL6 mice. Animals were killed by a Schedule 1 procedure, the bladder immediately removed, and a strip of detrusor (≤1 mm diameter, 5 mm length), with an intact mucosa, was tied to a fixed hook and an isometric force transducer.  The force transducer was mounted on a micromanipulator to allow variation of length according to pre-determined protocols.  With some preparations the mucosa was removed.  Preparations were superfused with Tyrode’s solution (24 mM HCO3-/ 5% CO2, pH 7.4, 36°C) and allowed to equilibrate for 30 min with a resting tension of 2 mN.  The stiffness, E, of the tissue was estimated from the increase of steady-state tension after a stretch of 1 mm (20% of resting length): values of E are expressed as mN.mm-2 tissue cross-section area.  ATP was measured in 100 µl superfusate samples taken 1 mm lateral to the preparation before and at 0.1, 2, 4, 6, 8, 10 and 30 mins after commencement of the 60-s stretch.  After the 30-min sample, 10 µM carbachol was introduced and a further sample taken after 2 min.
ATP was measured with a luciferin-luciferase assay (FLAAM, Sigma-Aldrich, Dorset, UK) and a luminometer (Glomax 20/20, Promega).  The system was calibrated between 10 fM and 1 µM ATP before each experiment.  Data are median values (25,75% interquartiles), n=number of preparations, as several data sets were significantly skewed).  Differences between pairs of data sets were tested by Wilcoxon tests.  The null hypothesis was rejected at *p<0.05.  For comparison of non-parametric ATP release data from young and aged mice, a median absolute deviation (MAD) for data from young mice was calculated [2], where 1.483*MAD (MAD(E)) is comparable to a standard deviation.  Aged mice data more than 3 MAD(E) values from those of young mice were classified as significantly (p<0.01) different.

Experiments with young mice.  ATP release measured over a 30 min time-course showed a peak at two minutes followed by a return to control after 30 minutes, followed by release two minutes after addition of carbachol (figure 1A).Paired comparison of carbachol and stretch-induced ATP release values, each after two minutes, showed that ATP release was greater with carbachol than rapid stretch (97 [59, 172] vs 44 [14, 68] fmol.ml-1.mg-1, p<0.01, n=10, figure 1B).  The relationship between ATP release and passive tension, after a constant extent of stretch, was constructed.  For this evaluation, the integral of ATP release over 10 minutes (∫ATP10; shaded area in figure 1A) was plotted as a function of passive tension where a significant positive association was obtained (r=0.78, p<0.005, n=12; figure 1C).  Control experiments showed that stretch-induced ATP release was very much less if the preparation was denuded of mucosa compared to intact samples (∫ATP10 values: 10 [8, 13] vs 128 [81,173] fmol.ml-1.mg-1, n=4, 12, p<0.05); no release was measured with 10 µM carbachol.
Experiments with aged mice.  A similar time course of ATP release after a rapid stretch was recorded (figure 2A). Carbachol-induced ATP release in aged mice was significantly less than in young mice (35 [20, 82] vs 97 [59, 172] fmol.ml-1.mg-1, p<0.01, n=19,10), but ATP release 2-min after stretch was similar to that in young mice.  Most noticeable was the wide range of stretch-induced ATP release that made analysis of such data on the basis of a homogenous set difficult.  Values of ∫ATP10 in aged mice preparations were divided into three groups (figure 2B): values comparable to those from young mice; much greater values (defined as more than three MAD(E) - see Methods - above those of young mice); much smaller values (more than three MAD(E) below those of young mice).  In view of the wide range of ∫ATP10 there were no significant relationships between stretch-induced ∫ATP10 values and either stretch-induced passive tension or carbachol-induced ATP release, as in tissue from young animals.

For young mice, data from these experiments are consistent with the original hypotheses.  Firstly, stretch-induced ATP release is a function of tension-generated in the preparation (i.e. tissue stiffness) and not the extent of stretch per se.  Secondly, carbachol induced ATP release is comparable, in fact greater, than stretch-induced release.  Stretch of the bladder wall, or shear stress applied to urothelial cells, releases more ACh than ATP and at lower stimulus intensities [3].  Thus, ACh release represents a positive autocrine pathway to augment stretch-activated ATP release.  Identification of ACh release channels from urothelial cells, and the cell pathways that regulate ATP release are important targets to downregulate excessive ATP release and the sensory pathways that ATP controls.  With bladder wall tissue from aged animals the relationship between tissue stiffness and ATP release is absent, due to the large variability of the amount of ATP released with stretch.  Enhanced ATP release is a feature of tissues from bladder pathologies and to this may be added a subset of aged bladders. Thus, data from aged animals does not represent a homogeneous group and that co-morbid conditions may overwhelm any effect of age per se, and if translated to elderly humans would underlie the variable incidence of urinary urgency.

An increase of bladder wall passive tension, as well as muscarinic receptor activation are equally potent stimuli for ATP release from the mucosa.  Stretch-activated release is augmented by increased bladder wall tension, say due to fibrosis and could contribute to increased urgency.  This suggests that antimuscarinic agents or that reduce fibrosis as targets to reduce this symptom. However, with aged animals other as yet unidentified factors greatly increase stretch-induced ATP release that is relatively independent of a muscarinic-dependent pathway.

Birder LA, Andersson KE.  Physiol Rev 2013; 93: 653-680.
Rousseeuw PJ, Croux C. J Am Stat Assoc 1993; 88: 1273–1283.
McLatchie LM, Young JS, Fry CH.  Br J Pharmacol 2014; 171: 3394-3403