- Email: [email protected]
Botulinum toxin A inhibits ATP release from bladder urothelium after chronic spinal cord injury
Neurochemistry International 45 (2004) 987–993
Botulinum toxin A inhibits ATP release from bladder urothelium after chronic spinal cord injury Mohit Khera, George T. Somogyi∗ , Susanna Kiss, Timothy B. Boone, Christopher P. Smith Scott Department of Urology, Baylor College of Medicine, One Baylor Plaza, Alkek Building N720, Houston, TX 77030, USA Received 22 April 2004; accepted 1 June 2004 Available online 23 July 2004
Abstract The effects of mechanoreceptor stimulation and subsequent ATP release in spinal cord injured and normal bladders was examined to demonstrate if spinal cord injury (SCI) modulates the basal or evoked release of ATP from bladder urothelium and whether intravesical botulinum toxin A (BTX-A) administration inhibits urothelial ATP release, a measure of sensory nerve activation. A Ussing chamber was used to isolate and separately measure resting and mechanoreceptor evoked (e.g. hypoosmotic stimulation) ATP release from urothelial and serosal sides of the bladder. Following spinal cord injury, resting urothelial release of ATP was ninefold higher than that of normal rats. Botulinum toxin A instillation did not significantly affect the resting release of ATP after spinal cord injury. Evoked ATP release following hypoosmotic stimulation was significantly higher in chronic spinal cord injured compared to normal rat bladders. However, botulinum toxin A treatment markedly reduced ATP release in spinal cord injured bladders by 53% suggesting that ATP release by mechanoreceptor stimulation, as opposed to basal release, occurs by exocytotic mechanisms. In contrast, there was no significant difference in basal or evoked ATP release from bladder serosa following spinal cord injury. Moreover, intravesical instillation of botulinum toxin A did not affect ATP release from the serosal side after spinal cord injury, suggesting that its effects were confined to the urothelial side of the bladder preparation. In summary: (1) increased release of ATP from the urothelium of spinal cord injured bladders may contribute to the development of bladder hyperactivity and, (2) mechanoreceptor stimulated vesicular ATP release, as opposed to basal non-vesicular release of ATP, is significantly inhibited in spinal cord injured bladders by intravesical instillation of botulinum toxin A. These results may have important relevance in our understanding of the mechanisms underlying plasticity of bladder afferent pathways following SCI. © 2004 Elsevier Ltd. All rights reserved. Keywords: Botulinum toxin; Bladder; Spinal cord injury; ATP; Urothelium; Afferent nerve; Ussing chamber
1. Introduction Spinal cord injury (SCI) induces significant plasticity within neural pathways innervating the lower urinary tract that leads to functional changes such as increased bladder activity (e.g. detrusor hyperreflexia) and bladder outlet obstruction (e.g. detrusor sphincter dyssynergia). Underlying these changes in efferent outflow are significant increases in bladder afferent activity that are secondary, in large part, to the unmasking of normally quiescent C-fiber nociceptors (De Groat et al., 1990). Following pathologic injury such as SCI or bladder inflammation, C-fiber afferent nerves respond to physiologic (pressure and volume) changes in addition to noxious stimuli. However, increased afferent ∗
Corresponding author. Tel.: +1 713 798 3541; fax: +1 713 798 6454. E-mail address: [email protected] (G.T. Somogyi).
0197-0186/$ – see front matter © 2004 Elsevier Ltd. All rights reserved. doi:10.1016/j.neuint.2004.06.001
nerve activity results not only from the recruitment of normally silent C-fiber nerves, but also because responding nerves become more excitable (Yoshimura and de Groat, 1997). Maggi has presented evidence that bladder sensory nerves have dual afferent and efferent nerve functions (Maggi, 1993). Sensory efferent functions include the release of transmitters such as ATP, substance P and CGRP that can act on nearby tissues as well as on afferent nerve terminals in an autocrine fashion to increase afferent nerve activity. In addition to the role that afferent nerve transmitter release plays in sensory function, significant recent interest has been directed to the existence and function of non-neuronal sources of transmitters. For example, it has been shown that acetylcholine (Reinheimer et al., 1996; Klapproth et al., 1997) as well as ATP (Ferguson et al., 1997; van der Wijk et al., 2003) can be released from
988
M. Khera et al. / Neurochemistry International 45 (2004) 987–993
the epithelium of bronchi, intestine and urinary bladder. Investigators postulate that these non-neuronal sources of neurotransmitters have important paracrine functions on surrounding tissues to modulate transmitter release and afferent nerve activity. Neuro-epithelial interactions between epithelium and afferent nerves has gained widespread interest among investigators focused on the role that sensory input plays in the development of various models of bladder dysfunction (Wessler et al., 1998). In this regard, urothelium is thought to play an important role in sensory transduction mechanisms that modulate micturition, particularly in conditions of increased sensory nerve transmission such as following chronic inflammation and SCI. Studies have demonstrated that urothelial cells can release the neurotransmitter ATP when stimulated by stretch (Sun et al., 2001). Released ATP can activate suburothelially located P2X3 receptors to increase sensory nerve transmission, possibly in a non-synaptic manner, a phenomenon that has been previously described regarding the interaction between efferent nerve terminals (Vizi, 1979). The importance of P2X3 receptors in sensory nerve transduction mechanisms is underscored by the significant decrease in bladder activity found in P2X3 knockout mice (Vlaskovska et al., 2001). Botulinum toxin A (BTX-A) is well established as an inhibitor of vesicular acetylcholine release from motor nerve terminals that acts by cleaving the SNARE protein SNAP-25 (Schiavo et al., 1993). Our lab has demonstrated that ACh and norepinephrine (NE) release from the rat bladder and urethra, respectively, is inhibited by BTX-A (Smith et al., 2003b). Furthermore, contractile data suggests that BTX-A may impair ATP release in addition to ACh release from isolated bladder tissue (Smith et al., 2003a). Yet, while traditionally used to treat disorders of muscle spasticity, recent basic and clinical evidence suggests that BTX-A may have antinociceptive effects unrelated to its actions on efferent nerve terminals (Cui and Aoki, 2000; Vemulakonda et al., 2004a,b; Smith and Chancellor, 2004). By impairing urothelial or afferent nerve transmitter release, particularly under conditions of chronic inflammation or SCI, BTX-A could reduce peripheral sensitization mechanisms that are thought to play an important role in increasing afferent nerve activity. In fact, experiments have shown that hypoosmotic stimulus evoked release of ATP from intestinal epithelium is inhibited by botulinum toxin (van der Wijk et al., 2003). These results suggest not only that evoked ATP release from epithelium is via vesicular mechanisms, but also demonstrate that BTX-A can inhibit transmitter release from non-neuronal tissue. In the present experiments, we studied the effects of mechanoreceptor stimulation and subsequent ATP release in spinal cord injured and normal bladders to examine whether: (1) SCI modulates the basal or evoked release of ATP from bladder urothelium and, (2) BTX-A administration inhibits urothelial ATP release, a measure of sensory nerve activation.
2. Materials and methods 2.1. Surgery Female Sprague–Dawley rats (250–300 g) were used for these experiments. Spinal cord injury (SCI) was accomplished by performing a laminectomy and a complete spinal cord transection between T8-T10 under isoflurane anesthesia. In order to prevent infection, rats received daily subcutaneous ampicillin (100 mg/kg i.m.) for a period of 1 week. During the spinal shock phase (10–14 days post-SCI), urinary bladders were manually expressed twice per day until spontaneous voiding bladder contractions reappeared. Experiments began in rats 21 days after SCI. All animal experiments were carried out in accordance with the National Institutes of Health guide for the care and use of laboratory animals. 2.2. BTX-A instillation Under isoflurane anesthesia, 1 ml of protamine sulfate (10 mg/ml) was instilled into normal or SCI bladders 5 days before the experiments began in order to increase urothelial permeability and allow BTX-A access to urothelium and suburothelial nerves (Niku et al., 1994). Protamine sulfate was left in the bladder for a total of 40 min and the rats were flipped from ventral to dorsal position after 20 min. The bladder was then irrigated with normal saline after which 1 ml of BTX-A (20 units/ml) was instilled in the bladder of one group of animals (e.g. SCI + BTX-A) and 1 ml of saline was instilled into another group (e.g. SCI) using the same protocol as for protamine sulfate. In addition, a third group of animals (e.g. normal) did not undergo SCI, but received protamine sulfate followed by saline instillation and were used as control animals. 2.3. ATP release Rats were euthanized with CO2 inhalation and the bladders were surgically removed from the abdomen and placed on a silgard covered Petri dish containing oxygenated Krebs solution. The bladder was cut open longitudinally and a 10 mm diameter curved piece was cut and mounted on the pins of the male half of the Ussing chamber (WPI, Sarasota, FL). The female and male halves of the chamber were then attached and put in a special vise (WPI) to hermetically seal the two halves together. The bladder mounted in this way is separated into two chambers where one chamber communicates with the urothelium (urothelial side) and the other half communicates with the serosa (serosal side) (Fig. 1). Both halves of the chamber were perfused with oxygenated preheated 37 ◦ C Krebs solution (mm/l; NaCl 113, KCL 4.7, CaCl2 1.25, MgSO4 1.2, NaHCO3 25, KH2 PO4 1.2, glucose 11.5) at a rate of 0.5 mL/min. After 60 min of equilibration, 5 min effluents were collected using a fraction collector
M. Khera et al. / Neurochemistry International 45 (2004) 987–993
(Graphpad, San Diego, CA). A level of P < 0.05 was considered statistically significant. The data are expressed as means ± S.E.M.
bladder o O o
989
Perfusion out
2.5. Drugs
collection
BTX-A (Botox® ) was generously provided by Allergan Inc. (Irvine, CA). All constituents of the Krebs solution and the assay kit for the ATP measurement were purchased from Sigma Chemical Co. (St. Louis, MO).
Perfusion in
Serosal side
Urothelial side
3. Results 3.1. Urothelial ATP release
Pump
Krebs
Fig. 1. Schematic diagram of the Ussing chamber used to measure ATP release. The bladder was cut open and used as a separation membrane between two hemichambers. Each chamber was perfused with oxygenated Krebs solution and the effluent was collected using a fraction collector. This apparatus allows one to separately collect released ATP from urothelial and serosal sides.
(Gilson, Middleton, WI) from both sides of the chamber. At the 10th min of the perfusion, a bolus of hypoosmotic Krebs solution was perfused for 6 min on either the urothelial or the serosal side of the chamber. The hypoosmosotic solution was created by reducing the concentration of NaCl by 40% from 113 to 67.8 mM. A 50 uL aliquot of each 5 min effluent was placed in a luminometer (Turner, Sunnyvale, CA) and the ATP content was measured by using the luciferin-luciferace assay. The luminescence of the collected samples was compared to that of standard ATP. Basal release of ATP was calculated by averaging the first two collected samples after a 60 min equilibration period. Evoked release of ATP by hypoosmotic stimulation was computed by adding the increased ATP values above baseline levels. Serosal release of ATP was standardized on the basis of tissue weight (e.g. fmol/g) because the number of nerve terminals and smooth muscle fibers, sources of serosal ATP, are proportional to tissue mass. In contrast, urothelial ATP release was standardized on the basis of the surface area of the Ussing chamber (e.g. fmol/cm2 ) considering that the predominant source of released ATP is the urothelium that should not vary with changes in mass of the tissue.
3.1.1. Resting release The first two samples after the 60 min equilibrium and before the hypoosmotic stimulation was considered as a measure of the resting release of ATP. The average of the two initial samples was calculated and given as the resting release value for the three experimental groups (e.g. normal rats, SCI rats, and SCI + BTX-A rats). As shown in Fig. 2, the SCI group had an almost ninefold higher resting release of ATP from the urothelial side than that of normal rats (1393 ± 158 versus 157 ± 76.4 fmol/cm2 , P < 0.05). In contrast, BTX-A instillation did not significantly affect the resting release of ATP following SCI (P > 0.05). 3.1.2. Hypoosmotic stimulation Hypoosmotic stimulation was applied as a bolus starting from the 10th min of the perfusion. Considering the lag time (e.g. 3 min) between the reservoir and the Ussing chamber as well as the dead space of one half of the chamber (e.g. 400 L), the effect of the hypoosmotic stimulation was initially delayed and persisted after switching back to normoosmotic Krebs solution. However, while the evoked release of
2.4. Statistical analysis Results obtained from different groups were analyzed using either non-parametric one-way ANOVA (e.g. Kruskal–Wallis) with Dunn’s post-test or by unpaired t-test where appropriate using the Prism statistical program
Fig. 2. Resting release of ATP into the urothelial compartment. Note there is a significant increase in the release of ATP under basal conditions in both SCI (n = 5) and SCI + BTX (n = 5) (e.g. SCI + BTX-A) groups compared to normal control animals (n = 4), ∗ denotes P < 0.05 using non-parametric ANOVA).
990
M. Khera et al. / Neurochemistry International 45 (2004) 987–993
Fig. 3. Hypoosmotic stimulation evoked ATP release into the urothelial compartment. Note that the released ATP was significantly higher in SCI bladders (n = 5) and SCI + BTX bladders (n = 5) compared to normal bladders (n = 4). BTX-A treatment (n = 5) significantly inhibited ATP release in SCI bladders. The symbol (**) denotes significant difference between normal and SCI bladders (P < 0.01), (*) denotes significant difference between normal and SCI + BTX bladders (P < 0.05), and (#) denotes significant different between SCI and SCI + BTX bladders (P < 0.05), using non-parametric ANOVA.
ATP was delayed, the area of the release curve was not affected. The hypoosmotic stimulation-evoked release of ATP was calculated by integration of the release curve above baseline values for the three experimental groups. While epithelial release of ATP in normal rats was minimal, ATP release was significantly higher in chronic SCI rat bladders (e.g. 208.8 ± 129.5 fmol/cm2 versus 5419 ± 964.5 fmol/cm2 , respectively, P < 0.01) (Fig. 3). BTX-A instillation significantly reduced ATP release in SCI bladders by 53% but ATP release still remained elevated above control levels (e.g. 2577 ± 774 fmol/cm2 in SCI + BTX-A versus 5419 ± 964.5 fmol/cm2 in SCI, P < 0.05). No measurable serosal release of ATP in any group occurred following urothelial hypoosmotic stimulation (data not shown). This data signifies that: (1) urothelial and serosal compartments are physically isolated (e.g. urothelial ATP does not leak around into serosal chamber), (2) ATP measured in the urothelial compartment is derived from superficial tissue sources. Otherwise, if deeper layers were responsible for a significant component of urothelial ATP levels, we would expect to also measure significant release of ATP within the serosal chamber following urothelial hypoosmotic stimulation and (3) Urothelial ATP does not diffuse through or stimulate deeper tissue layers to release ATP into the serosal compartment.
Fig. 4. Resting release of ATP into the serosal compartment. Compared to findings on the urothelial side of the bladder, there was no significant change in basal efflux of serosal ATP following SCI (n = between 4 and 5, P > 0.6 using non-parametric ANOVA).
ples was calculated and given as the resting release value for the three experimental groups (e.g. normal rats, SCI rats and SCI + BTX-A rats. In contrast to our findings on the urothelial side of the bladder, there was no significant change in basal efflux of serosal ATP following SCI (P > 0.6) (Fig. 4). 3.2.2. Hypoosmotic stimulation Using the same paradigm of hypoosmotic stimulation, the serosal side was exposed to reduced osmolarity Krebs (e.g. 60%) and the ATP content was measured in 5 min effluents. The ATP release was calculated by integrating the evoked ATP efflux curve. As shown in Fig. 5, serosal release of ATP was not significantly increased in SCI bladders when compared to normal bladders (P > 0.5). In addition, intravesical instillation of BTX-A was not able to significantly reduce the amount of ATP released from the SCI serosal surface after hypoosmotic stimulation suggesting that the mechanism of ATP release from the serosal side does not involve vesicular exocytosis. In addition to measuring serosal release of ATP to serosal hypoosmotic stimulation, in a subset of normal and SCI
3.2. Serosal ATP release 3.2.1. Resting release Similar to urothelial experiments, the first two samples after a 60 min equilibrium period and before hypoosmotic stimulation was considered as a measure of the serosal resting release of ATP. The average of the two initial sam-
Fig. 5. Hypoosmotic stimulation evoked release of ATP into the serosal compartment. Following SCI, no significant difference in evoked ATP release was demonstrated (n = between 4 and 5, P > 0.5 using non-parametric ANOVA).
M. Khera et al. / Neurochemistry International 45 (2004) 987–993
animals, we also investigated the effect of serosal stimulation on urothelial ATP release. We found no significant urothelial release of ATP in either group to serosal hypoosmotic stimulation (data not shown), demonstrating that: (1) serosal hypoosmotic stimulation does not stimulate urothelial sources of ATP, (2) urothelial and serosal chambers are physically separated (e.g. serosal ATP does not leak into urothelial chamber) and (3) serosal released ATP does not diffuse through tissue and induce urothelial release of ATP.
4. Discussion The main findings of these experiments are that, after spinal cord injury: (1) resting release of ATP from bladder urothelium is increased, (2) there is an increased release of ATP from the urothelial but not the serosal side of the bladder following hypoosmotic stimulation and (3) urothelial but not serosal hypoosmotic stimulation-evoked ATP release is inhibited by BTX-A instilled into the bladder lumen. Our results have significance in several respects. First, the increase in basal and hypoosmotic stimulation induced ATP release after SCI may evoke an increase in the frequency of bladder contractions by activating P2X3 receptors on afferent nerve terminals (Vlaskovska et al., 2001). Second, the fact that BTX-A pretreatment significantly reduced ATP release from the urothelium supports the concept that BTX-A treatment inhibits transmitter release not only from efferent nerve endings but also from the sensory nerve terminals and/or urothelium. ATP is well established as a co-transmitter that is released along with ACh and NE from peripheral tissues. Although ATP release has been widely demonstrated from efferent cholinergic and adrenergic terminals (Sperlagh and Vizi, 1992; Todorov et al., 1996; Tong et al., 1997), ATP release from bladder urothelial cells has only recently been confirmed within the scientific literature (Ferguson et al., 1997; Ferguson, 1999; Sun et al., 2001; Birder et al., 2002). Since discovering that ATP release can be generated by increasing the hydrostatic pressure on the urothelial side of the bladder (Ferguson et al., 1997; Birder et al., 2002) or in the lumen of the ureter (Knight et al., 2002), it has become increasingly evident that urothelium plays a significant role as a pressure sensing system, utilizing the neurotransmitter ATP in response to mechanical stimulation. This concept has been proven not only in intact ex vivo bladder tissue experiments but also in urothelial cell culture preparations in response to stretch (Sun et al., 2001) or hypoosmotic stress (Birder et al., 2002). In addition, increased stretch activated ATP release was reported from human urothelial cells cultured from the bladders of patients with IC (Sun et al., 2001), a chronic inflammatory and painful bladder disorder in which alterations in urothelium and afferent nerve function are thought to be prominent etiologic components (Steers and Tuttle, 1997).
991
SCI bladders also demonstrate increased afferent nerve activity that can be measured by cystometrogram (CMG) as an increase in the frequency of bladder contractions (Cheng et al., 1995; Khera et al., 2004) compared to that of normal bladders. In conjunction with these findings, the resting release of ATP from the urothelium of SCI bladders was higher in our experiments than ATP levels in normal bladders. However, the resting release of ATP from the serosal side did not increase after SCI despite the known presence of sensory nerve endings within the bladder smooth muscle layer (Gabella and Davis, 1998). We hypothesize that serosal basal efflux of ATP includes sources other than sensory nerves such as efferent nerve terminals or smooth muscle itself whose basal ATP release may not be directly affected by SCI. Not surprisingly, the resting release of ATP on the urothelial or serosal side did not respond to treatment with BTX-A. Since BTX-A inhibits SNAP-25 mediated vesicle docking and exocytotic release of transmitters, these findings imply that the SNARE protein SNAP-25 does not play an active role in the resting release of ATP. One might suggest that urothelial integrity is impaired after SCI such that ATP is simply leaking out from damaged urothelial cells. However, it has recently been shown that transepithelial resistance, a marker of the permeability of bladder urothelium, is unchanged after chronic spinal cord injury (Apodaca et al., 2003). Diffusion of ATP from the serosal side to the urothelial side is also improbable under resting conditions because urothelial ATP release was significantly higher in SCI bladders than in normal bladders, whereas there was no difference in resting ATP release between both groups on the serosal side (Figs. 2 and 4). The most relevant difference between normal and SCI bladders was the markedly elevated release of ATP from the urothelial side of SCI bladders after hypoosmotic stimulation. Epithelial cells have been shown to respond to hypoosmotic shock by releasing ATP in airway (Taylor et al., 1998) and intestine (van der Wijk et al., 2003) as well as in cultured urothelial cells (Birder et al., 2002). We were somewhat surprised to find the relatively small increase in ATP release after hypoosmotic shock in normal bladder urothelium as compared to its corresponding resting ATP release. It appears that urothelial cells integrated in our whole bladder preparation are more resistant to hypoosmostic stimulation evoked volume change than are solitary urothelial cells in primary culture. For this reason, hypoosmotic stimulation evokes only a small degree of ATP release. This phenomenon, however, does not apply to all changes in fluid osmolarity since a significant decrease in transmembrane resistance was observed in rabbit urothelium exposed to hyperosmotic solutions (Ferguson, 1999). However, changes in resistance do not necessarily correlate with changes in ATP release. Nevertheless, this finding can have important physiologic implications since bladder urothelium is constantly exposed to urine of differing osmolarities. Given our findings, it appears that normal bladders
992
M. Khera et al. / Neurochemistry International 45 (2004) 987–993
are less sensitive to hypoosmotic compared to hyperosmotic stimuli. In addition, under pathological conditions such as SCI, hypoosmotic shock evokes a significant release of ATP from bladder urothelium indicating that urothelium has become much more sensitive to, at least, decreases in urine osmolarity. Another significant finding of these experiments was that BTX-A instilled into the bladder lumen inhibited the release of ATP from the urothelial but not the serosal side of the bladder. Mechanical stress induces release of ATP from various epithelial tissues including intestine and bladder (Birder et al., 2002; Ferguson et al., 1997; van der Wijk et al., 2003). Moreover, hypoosmotic evoked ATP release can be inhibited by BTX or by depletion of intracellular calcium stores (van der Wijk et al., 2003). In addition, calcium-dependent ATP release has also been reported in ureteral epithelium (Knight et al., 2002) further suggesting that ATP release from epithelial tissues occurs via exocytotic mechanisms. Evoked release of ATP, at least from bladder urothelium, is not inhibited by tetrodotoxin implying that urothelial ATP release is not significantly affected by neuronal activity (Ferguson et al., 1997). Our results demonstrating that hypoosmotic stimulation of bladder urothelium evokes a significant release of ATP which is markedly inhibited by BTX-A suggests that: (1) ATP release from bladder urothelium is exocytotic in nature and (2) Impairment of urothelial ATP release is the one of the major mechanisms by which BTX-A reduces afferent activation following pathologic insult such as SCI. In contrast to urothelium, basal and evoked serosal ATP release was not significantly different between normal and SCI bladders. Since release of serosal ATP was similar in normal and SCI bladders, we speculate that smooth muscle may serve as a significant source of ATP, masking release from other areas (i.e. afferent or efferent nerves). In addition, neither basal nor evoked serosal ATP release was affected by BTX-A instilled into the bladder lumen after protamine sulfate treatment. Protamine sulfate is a caustic agent known to disrupt the integrity of epithelium and to dramatically decrease transmembrane resistance (Niku et al., 1994). However, normal resistance and urothelial integrity is restored within a few days after treatment (Apodaca et al., 2003). We applied protamine sulfate to allow BTX-A access to urothelium and suburothelial afferent nerves. The ineffectiveness of BTX-A to inhibit evoked serosal release of ATP may signify that BTX-A could not penetrate deeper layers of the bladder such that its effects were confined to superficial bladder tissues (e.g. urothelium and afferent nerve terminals). Although elevated amounts of ATP are released on the serosal side of the bladder, this release does not appear to significantly affect the afferent system adjacent to the urothelium. In our experiments, we could not detect any overflow of ATP from the serosal to urothelial sides or vice versa. Thus, ubiquitously released ATP appears to be localized within the bladder in such a way that ATP released from deeper lay-
ers (i.e. serosal hypoosmotic stimulation) does not significantly modulate afferent signals originating from urothelium in normal or pathologic (SCI) states. What is the significance of the increased release of ATP in SCI bladders? The prevailing notion is that urothelium serves as a pressure sensor whose function may be modulated under inflammatory (Vemulakonda et al., 2004) or neurological (spinal cord injury) conditions (Apodaca et al., 2003; Khera et al., 2004). While there is no evidence that urothelium is wired directly to afferent nerves by forming synapses with intraepithelial or subepithelial afferent fibers (Gabella and Davis, 1998), it can communicate with the sensory nerves via release of transmitters such as ATP (Birder et al., 2002). Thus, the connection between urothelium and the afferent nerve endings is non-synaptic, a form of interaction already described amongst efferent nerve terminals (e.g. adrenergic and cholinergic) within the peripheral (Vizi, 1979; Somogyi and de Groat, 1990) and the central nervous system (Vizi, 2000). Higher amounts of ATP release following SCI could activate P2X3 receptors in epithelial and subepithelial bladder layers to increase afferent nerve activity and, consequently, account for the higher frequency of bladder contractions seen in both human and animal models of SCI. The finding that BTX-A significantly impairs urothelial ATP release following SCI gives another possible explanation for its clinical efficacy in treating human patients with neurogenic bladders after SCI. In addition, by inhibiting mechanosensory mechanisms within the bladder, use of BTX-A may be extended to treat other disorders of increased bladder afferent activity such as interstitial cystitis, a chronic, debilitating inflammatory disorder characterized by symptoms of urinary frequency, urgency, and pelvic pain. In summary, there is an increased release of ATP from the urothelium of SCI bladders that may contribute to the development of bladder hyperactivity by activating P2X3 receptors on epithelial and subepithelial afferent nerve endings. Since afferent nerve endings are not forming synapses with urothelial cells, the interaction appears to be non-synaptic in nature. ATP release evoked by hypoosmotic stimulation appears to occur through exocytotic mechanisms that can be inhibited by BTX-A, a finding that helps explain the inhibitory effects of BTX-A on bladder sensory mechanisms. These results may have important relevance in our understanding of the mechanisms underlying plasticity of bladder afferent pathways following SCI.
Acknowledgements We express our sincere gratitude to K. Roger Aoki, Ph.D., for valuable discussions and providing the BTX-A (Botox® , Allergan, Irvine, CA) for our experiments. This work was supported by the NIH grant RO1 DK 45741; The Scott Department of Urology Neurourology Fund; and an Unrestricted Educational Grant from Allergan, Irvine, CA.
M. Khera et al. / Neurochemistry International 45 (2004) 987–993
References Apodaca, G., Kiss, S., Ruiz, W., Meyers, S., Zeidel, M., Birder, L., 2003. Disruption of bladder epithelium barrier function after spinal cord injury. Am. J. Physiol. 284, 966–976. Birder, L.A., Nakamura, Y., Kiss, S., Nealen, M.L., Barrick, S., Kanai, A.J., Wang, E., Ruiz, G., de Groat, W.C., Apodaca, G., Watkins, S., Caterina, M.J., 2002. Altered urinary bladder function in mice lacking the vanilloid receptor TRPV1. Nat. Neurosci. 5, 856–860. Cheng, C.L., Ma, C.P., de Groat, W.C., 1995. Effect of capsaicin on micturition and associated reflexes in chronic spinal rats. Brain Res. 678, 40–48. Cui, M., Aoki, K.R., 2000. Botulinum toxin type A (BTX-A) reduces inflammatory pain in the rat formalin model. Cephalagia 20, 414. De Groat, W.C., Kawatani, M., Hisamitsu, T., Cheng, C.L., Ma, C.P., Thor, K., Steers, W., Roppolo, J.R., 1990. Mechanisms underlying the recovery of urinary bladder function following spinal cord injury. J. Autonomic Nervous Sys. 30, 71–77. Ferguson, D.R., Kennedy, I., Burton, T.J., 1997. ATP is released from rabbit urinary bladder epithelial cells by hydrostatic pressure changes—a possible sensory mechanism? J. Physiol. 505, 503–511. Ferguson, D.R., 1999. Urothelial function. Br. J. Urol. Int. 84, 235–242. Gabella, G., Davis, C., 1998. Distribution of afferent axons in the bladder of rats. J. Neuro. Cytol. 27, 141–155. Khera, M., Somogyi, G.T., Salas, N., Vemulakonda, V.M., Evans, J.L., Kiss, S., Smith, C.P., Boone, T.B., 2004. Effects of botulinum toxin with protamine on bladder contractility after spinal cord injury. J. Urol. 171, 459. Klapproth, H., Reinheimer, T., Metzen, J., Munch, M., Bittinger, F., Kirkpatrick, C.J., Hohle, K.D., Schemann, M., Racke, K., Wessler, I., 1997. Non-neuronal acetylcholine, a signaling molecule synthesized by surface cells of rat and man. Naunyn Schmiedebergs Arch. Pharmacol. 355, 515–523. Knight, G.E., Bodin, P., de Groat, W.C., Burnstock, G., 2002. ATP is released from guinea pig ureter epithelium on distension. Am. J. Physiol., Renal Physiol. 282, 281–288. Maggi, C.A., 1993. The dual, sensory and efferent functions of the capsaicin-sensitive primary sensory nerves in the bladder and urethra. In: Maggi, C.A. (Ed.), The Autonomic Nervous System: Nervous Control of the Urogenital System, vol. 3. Harwood Academic Publishers, London, pp. 383 (Chapter 11). Niku, S.D., Stein, P.C., Scherz, H.C., Parsons, C.L., 1994. A new method for cytodestruction of bladder epithelium using protamine sulfate and urea. J. Urol. 152, 1025–1028. Reinheimer, T., Bernedo, P., Klapproth, H., Oelert, H., Zeiske, B., Racke, K., Wessler, I., 1996. Acetylcholine in isolated airways of rat, guinea pig, and human: species differences in role of airway mucosa. Am. J. Physiol. 270, 722–728. Schiavo, G., Santucci, A., DasGupta, B.R., et al., 1993. Botulinum neurotoxins serotypes A and E cleave Snap-25 at distinct COOH-terminal peptide bonds. FEBS Lett. 335, 99. Smith, C.P., Boone, T.B., de Groat, W.C., Chancellor, M.B., Somogyi, G.T., 2003a. Effect of stimulation intensity and botulinum toxin isoform on rat bladder strip contractions. Brain Res. Bull. 61, 165–171. Smith, C.P., Chancellor, M.B., 2004. Emerging role of botulinum toxin in the treatment of voiding dysfunction. J. Urol. 171, 2128–2137.
993
Smith, C.P., Franks, M.E., McNeil, B.K., Ghosh, R., de Groat, W.C., Chancellor, M.B., Somogyi, G.T., 2003b. Effect of botulinum toxin A on the autonomic nervous system of the rat lower urinary tract. J. Urol. 169, 1896–1900. Somogyi, G.T., de Groat, W.C., 1990. Modulation of the release of [3H] norepinephrine from the base and body of the rat urinary bladder by endogenous adrenergic and cholinergic mechanicsms. J. Pharmacol. Exp. Ther. 255, 204–210. Sperlagh, B., Vizi, E.S., 1992. Is the neuronal ATP release from guinea-pig vas deferens subject to alpha2-adrenoceptor-mediated modulation? Neuroscience 51, 203–209. Steers W.D., Tuttle J.B., 1997. Neurogenic inflammation and nerve growth factor: possible roles in interstitial cystitis. In: Sant, G.R. (Ed.), Interstitial Cystitis. Lippincott-Raven, Philadelphia, pp. 67– 75. Sun, Y., Keay, S., De Deyne, P.G., Chai, T.C., 2001. Augmented stretch activated adenosine triphosphate release from bladder uroepithelial cells in patients with interstitial cystitis. J. Urol. 166, 1951–1956. Taylor, A.L., Kudlow, B.A., Marrs, K.L., Gruenert, D.C., Guggino, W.B., Schwiebert, E.M., 1998. Bioluminescence detection of ATP release mechanisms in epithelia. Am. J. Physiol. 275, 1391–1406. Todorov, L.D., Mihaylova-Todorova, S., Craviso, G.L., Bjur, R.A., Westfall, D.P., 1996. Evidence for the differential release of the cotransmitters ATP and noradrenaline from sympathetic nerves of the guinea-pig vas deferens. J. Physiol. 496, 731–748. Tong, Y.C., Hung, Y.C., Shinozuka, K., Kunitomo, M., Cheng, J.T., 1997. Evidence of adenosine 5 -triphosphate release from nerve and P2X-purinoceptor mediated contraction during electrical stimulation of rat urinary bladder smooth muscle. J. Urol. 158, 1973–1977. van der Wijk, T., Tomassen, S.F., Houtsmuller, A.B., de Jonge, H.R., Tilly, B.C., 2003. Increased vesicle recycling in response to osmotic cell swelling. Cause and consequence of hypotonicity-provoked ATP release. J. Biol. Chem. 278, 40020–40025. Vemulakonda, V.M., Smith, C.P., Kiss, S., Khera, M., Evans, J.L., Boone, T.B., Somogyi, G.T., 2004a. Botulinum toxin A inhibits mechanoreceptor-mediated urothelial release of ATP in conditions of bladder inflammation. J. Urol. 171 (Suppl), 95. Vemulakonda, V.M., Somogyi, G.T., Kiss, S., Evans, J.L., Khera, M., Boone, T.B., Smith, C.P., 2004b. Intravesical botulinum toxin A inhibits the afferent response in the chronically inflamed bladder. J. Urol. 171 (Suppl), 458. Vizi, E.S., 1979. Presynaptic modulation of neurochemical transmission. Prog. Neurobiol. 12, 181–290. Vizi, E.S., 2000. Role of high-affinity receptors and membrane transporters in nonsynaptic communication and drug action in the central nervous system. Pharmacol. Rev. 52, 63–89. Vlaskovska, M., Kasakov, L., Rong, W., Bodin, P., Bardini, M., Cockayne, D.A., Ford, A.P., Burnstock, G., 2001. P2X3 knock-out mice reveal a major sensory role for urothelially released ATP. J. Neurosci. 21, 5670–5677. Wessler, I., Kirkpatrick, C.J., Racke, K., 1998. Non-neuronal acetylcholine, a locally acting molecule, widely distributed in biological systems: expression and function in humans. Pharmacol. Ther. 77, 59–79. Yoshimura, N., de Groat, W.C., 1997. Plasticity of Na+ channels in afferent neurons innervating rat urinary bladder following spinal cord injury. J. Physiol. 503, 269–276.
Botulinum toxin A inhibits ATP release from bladder urothelium after chronic spinal cord injury Mohit Khera, George T. Somogyi∗ , Susanna Kiss, Timothy B. Boone, Christopher P. Smith Scott Department of Urology, Baylor College of Medicine, One Baylor Plaza, Alkek Building N720, Houston, TX 77030, USA Received 22 April 2004; accepted 1 June 2004 Available online 23 July 2004
Abstract The effects of mechanoreceptor stimulation and subsequent ATP release in spinal cord injured and normal bladders was examined to demonstrate if spinal cord injury (SCI) modulates the basal or evoked release of ATP from bladder urothelium and whether intravesical botulinum toxin A (BTX-A) administration inhibits urothelial ATP release, a measure of sensory nerve activation. A Ussing chamber was used to isolate and separately measure resting and mechanoreceptor evoked (e.g. hypoosmotic stimulation) ATP release from urothelial and serosal sides of the bladder. Following spinal cord injury, resting urothelial release of ATP was ninefold higher than that of normal rats. Botulinum toxin A instillation did not significantly affect the resting release of ATP after spinal cord injury. Evoked ATP release following hypoosmotic stimulation was significantly higher in chronic spinal cord injured compared to normal rat bladders. However, botulinum toxin A treatment markedly reduced ATP release in spinal cord injured bladders by 53% suggesting that ATP release by mechanoreceptor stimulation, as opposed to basal release, occurs by exocytotic mechanisms. In contrast, there was no significant difference in basal or evoked ATP release from bladder serosa following spinal cord injury. Moreover, intravesical instillation of botulinum toxin A did not affect ATP release from the serosal side after spinal cord injury, suggesting that its effects were confined to the urothelial side of the bladder preparation. In summary: (1) increased release of ATP from the urothelium of spinal cord injured bladders may contribute to the development of bladder hyperactivity and, (2) mechanoreceptor stimulated vesicular ATP release, as opposed to basal non-vesicular release of ATP, is significantly inhibited in spinal cord injured bladders by intravesical instillation of botulinum toxin A. These results may have important relevance in our understanding of the mechanisms underlying plasticity of bladder afferent pathways following SCI. © 2004 Elsevier Ltd. All rights reserved. Keywords: Botulinum toxin; Bladder; Spinal cord injury; ATP; Urothelium; Afferent nerve; Ussing chamber
1. Introduction Spinal cord injury (SCI) induces significant plasticity within neural pathways innervating the lower urinary tract that leads to functional changes such as increased bladder activity (e.g. detrusor hyperreflexia) and bladder outlet obstruction (e.g. detrusor sphincter dyssynergia). Underlying these changes in efferent outflow are significant increases in bladder afferent activity that are secondary, in large part, to the unmasking of normally quiescent C-fiber nociceptors (De Groat et al., 1990). Following pathologic injury such as SCI or bladder inflammation, C-fiber afferent nerves respond to physiologic (pressure and volume) changes in addition to noxious stimuli. However, increased afferent ∗
Corresponding author. Tel.: +1 713 798 3541; fax: +1 713 798 6454. E-mail address: [email protected] (G.T. Somogyi).
0197-0186/$ – see front matter © 2004 Elsevier Ltd. All rights reserved. doi:10.1016/j.neuint.2004.06.001
nerve activity results not only from the recruitment of normally silent C-fiber nerves, but also because responding nerves become more excitable (Yoshimura and de Groat, 1997). Maggi has presented evidence that bladder sensory nerves have dual afferent and efferent nerve functions (Maggi, 1993). Sensory efferent functions include the release of transmitters such as ATP, substance P and CGRP that can act on nearby tissues as well as on afferent nerve terminals in an autocrine fashion to increase afferent nerve activity. In addition to the role that afferent nerve transmitter release plays in sensory function, significant recent interest has been directed to the existence and function of non-neuronal sources of transmitters. For example, it has been shown that acetylcholine (Reinheimer et al., 1996; Klapproth et al., 1997) as well as ATP (Ferguson et al., 1997; van der Wijk et al., 2003) can be released from
988
M. Khera et al. / Neurochemistry International 45 (2004) 987–993
the epithelium of bronchi, intestine and urinary bladder. Investigators postulate that these non-neuronal sources of neurotransmitters have important paracrine functions on surrounding tissues to modulate transmitter release and afferent nerve activity. Neuro-epithelial interactions between epithelium and afferent nerves has gained widespread interest among investigators focused on the role that sensory input plays in the development of various models of bladder dysfunction (Wessler et al., 1998). In this regard, urothelium is thought to play an important role in sensory transduction mechanisms that modulate micturition, particularly in conditions of increased sensory nerve transmission such as following chronic inflammation and SCI. Studies have demonstrated that urothelial cells can release the neurotransmitter ATP when stimulated by stretch (Sun et al., 2001). Released ATP can activate suburothelially located P2X3 receptors to increase sensory nerve transmission, possibly in a non-synaptic manner, a phenomenon that has been previously described regarding the interaction between efferent nerve terminals (Vizi, 1979). The importance of P2X3 receptors in sensory nerve transduction mechanisms is underscored by the significant decrease in bladder activity found in P2X3 knockout mice (Vlaskovska et al., 2001). Botulinum toxin A (BTX-A) is well established as an inhibitor of vesicular acetylcholine release from motor nerve terminals that acts by cleaving the SNARE protein SNAP-25 (Schiavo et al., 1993). Our lab has demonstrated that ACh and norepinephrine (NE) release from the rat bladder and urethra, respectively, is inhibited by BTX-A (Smith et al., 2003b). Furthermore, contractile data suggests that BTX-A may impair ATP release in addition to ACh release from isolated bladder tissue (Smith et al., 2003a). Yet, while traditionally used to treat disorders of muscle spasticity, recent basic and clinical evidence suggests that BTX-A may have antinociceptive effects unrelated to its actions on efferent nerve terminals (Cui and Aoki, 2000; Vemulakonda et al., 2004a,b; Smith and Chancellor, 2004). By impairing urothelial or afferent nerve transmitter release, particularly under conditions of chronic inflammation or SCI, BTX-A could reduce peripheral sensitization mechanisms that are thought to play an important role in increasing afferent nerve activity. In fact, experiments have shown that hypoosmotic stimulus evoked release of ATP from intestinal epithelium is inhibited by botulinum toxin (van der Wijk et al., 2003). These results suggest not only that evoked ATP release from epithelium is via vesicular mechanisms, but also demonstrate that BTX-A can inhibit transmitter release from non-neuronal tissue. In the present experiments, we studied the effects of mechanoreceptor stimulation and subsequent ATP release in spinal cord injured and normal bladders to examine whether: (1) SCI modulates the basal or evoked release of ATP from bladder urothelium and, (2) BTX-A administration inhibits urothelial ATP release, a measure of sensory nerve activation.
2. Materials and methods 2.1. Surgery Female Sprague–Dawley rats (250–300 g) were used for these experiments. Spinal cord injury (SCI) was accomplished by performing a laminectomy and a complete spinal cord transection between T8-T10 under isoflurane anesthesia. In order to prevent infection, rats received daily subcutaneous ampicillin (100 mg/kg i.m.) for a period of 1 week. During the spinal shock phase (10–14 days post-SCI), urinary bladders were manually expressed twice per day until spontaneous voiding bladder contractions reappeared. Experiments began in rats 21 days after SCI. All animal experiments were carried out in accordance with the National Institutes of Health guide for the care and use of laboratory animals. 2.2. BTX-A instillation Under isoflurane anesthesia, 1 ml of protamine sulfate (10 mg/ml) was instilled into normal or SCI bladders 5 days before the experiments began in order to increase urothelial permeability and allow BTX-A access to urothelium and suburothelial nerves (Niku et al., 1994). Protamine sulfate was left in the bladder for a total of 40 min and the rats were flipped from ventral to dorsal position after 20 min. The bladder was then irrigated with normal saline after which 1 ml of BTX-A (20 units/ml) was instilled in the bladder of one group of animals (e.g. SCI + BTX-A) and 1 ml of saline was instilled into another group (e.g. SCI) using the same protocol as for protamine sulfate. In addition, a third group of animals (e.g. normal) did not undergo SCI, but received protamine sulfate followed by saline instillation and were used as control animals. 2.3. ATP release Rats were euthanized with CO2 inhalation and the bladders were surgically removed from the abdomen and placed on a silgard covered Petri dish containing oxygenated Krebs solution. The bladder was cut open longitudinally and a 10 mm diameter curved piece was cut and mounted on the pins of the male half of the Ussing chamber (WPI, Sarasota, FL). The female and male halves of the chamber were then attached and put in a special vise (WPI) to hermetically seal the two halves together. The bladder mounted in this way is separated into two chambers where one chamber communicates with the urothelium (urothelial side) and the other half communicates with the serosa (serosal side) (Fig. 1). Both halves of the chamber were perfused with oxygenated preheated 37 ◦ C Krebs solution (mm/l; NaCl 113, KCL 4.7, CaCl2 1.25, MgSO4 1.2, NaHCO3 25, KH2 PO4 1.2, glucose 11.5) at a rate of 0.5 mL/min. After 60 min of equilibration, 5 min effluents were collected using a fraction collector
M. Khera et al. / Neurochemistry International 45 (2004) 987–993
(Graphpad, San Diego, CA). A level of P < 0.05 was considered statistically significant. The data are expressed as means ± S.E.M.
bladder o O o
989
Perfusion out
2.5. Drugs
collection
BTX-A (Botox® ) was generously provided by Allergan Inc. (Irvine, CA). All constituents of the Krebs solution and the assay kit for the ATP measurement were purchased from Sigma Chemical Co. (St. Louis, MO).
Perfusion in
Serosal side
Urothelial side
3. Results 3.1. Urothelial ATP release
Pump
Krebs
Fig. 1. Schematic diagram of the Ussing chamber used to measure ATP release. The bladder was cut open and used as a separation membrane between two hemichambers. Each chamber was perfused with oxygenated Krebs solution and the effluent was collected using a fraction collector. This apparatus allows one to separately collect released ATP from urothelial and serosal sides.
(Gilson, Middleton, WI) from both sides of the chamber. At the 10th min of the perfusion, a bolus of hypoosmotic Krebs solution was perfused for 6 min on either the urothelial or the serosal side of the chamber. The hypoosmosotic solution was created by reducing the concentration of NaCl by 40% from 113 to 67.8 mM. A 50 uL aliquot of each 5 min effluent was placed in a luminometer (Turner, Sunnyvale, CA) and the ATP content was measured by using the luciferin-luciferace assay. The luminescence of the collected samples was compared to that of standard ATP. Basal release of ATP was calculated by averaging the first two collected samples after a 60 min equilibration period. Evoked release of ATP by hypoosmotic stimulation was computed by adding the increased ATP values above baseline levels. Serosal release of ATP was standardized on the basis of tissue weight (e.g. fmol/g) because the number of nerve terminals and smooth muscle fibers, sources of serosal ATP, are proportional to tissue mass. In contrast, urothelial ATP release was standardized on the basis of the surface area of the Ussing chamber (e.g. fmol/cm2 ) considering that the predominant source of released ATP is the urothelium that should not vary with changes in mass of the tissue.
3.1.1. Resting release The first two samples after the 60 min equilibrium and before the hypoosmotic stimulation was considered as a measure of the resting release of ATP. The average of the two initial samples was calculated and given as the resting release value for the three experimental groups (e.g. normal rats, SCI rats, and SCI + BTX-A rats). As shown in Fig. 2, the SCI group had an almost ninefold higher resting release of ATP from the urothelial side than that of normal rats (1393 ± 158 versus 157 ± 76.4 fmol/cm2 , P < 0.05). In contrast, BTX-A instillation did not significantly affect the resting release of ATP following SCI (P > 0.05). 3.1.2. Hypoosmotic stimulation Hypoosmotic stimulation was applied as a bolus starting from the 10th min of the perfusion. Considering the lag time (e.g. 3 min) between the reservoir and the Ussing chamber as well as the dead space of one half of the chamber (e.g. 400 L), the effect of the hypoosmotic stimulation was initially delayed and persisted after switching back to normoosmotic Krebs solution. However, while the evoked release of
2.4. Statistical analysis Results obtained from different groups were analyzed using either non-parametric one-way ANOVA (e.g. Kruskal–Wallis) with Dunn’s post-test or by unpaired t-test where appropriate using the Prism statistical program
Fig. 2. Resting release of ATP into the urothelial compartment. Note there is a significant increase in the release of ATP under basal conditions in both SCI (n = 5) and SCI + BTX (n = 5) (e.g. SCI + BTX-A) groups compared to normal control animals (n = 4), ∗ denotes P < 0.05 using non-parametric ANOVA).
990
M. Khera et al. / Neurochemistry International 45 (2004) 987–993
Fig. 3. Hypoosmotic stimulation evoked ATP release into the urothelial compartment. Note that the released ATP was significantly higher in SCI bladders (n = 5) and SCI + BTX bladders (n = 5) compared to normal bladders (n = 4). BTX-A treatment (n = 5) significantly inhibited ATP release in SCI bladders. The symbol (**) denotes significant difference between normal and SCI bladders (P < 0.01), (*) denotes significant difference between normal and SCI + BTX bladders (P < 0.05), and (#) denotes significant different between SCI and SCI + BTX bladders (P < 0.05), using non-parametric ANOVA.
ATP was delayed, the area of the release curve was not affected. The hypoosmotic stimulation-evoked release of ATP was calculated by integration of the release curve above baseline values for the three experimental groups. While epithelial release of ATP in normal rats was minimal, ATP release was significantly higher in chronic SCI rat bladders (e.g. 208.8 ± 129.5 fmol/cm2 versus 5419 ± 964.5 fmol/cm2 , respectively, P < 0.01) (Fig. 3). BTX-A instillation significantly reduced ATP release in SCI bladders by 53% but ATP release still remained elevated above control levels (e.g. 2577 ± 774 fmol/cm2 in SCI + BTX-A versus 5419 ± 964.5 fmol/cm2 in SCI, P < 0.05). No measurable serosal release of ATP in any group occurred following urothelial hypoosmotic stimulation (data not shown). This data signifies that: (1) urothelial and serosal compartments are physically isolated (e.g. urothelial ATP does not leak around into serosal chamber), (2) ATP measured in the urothelial compartment is derived from superficial tissue sources. Otherwise, if deeper layers were responsible for a significant component of urothelial ATP levels, we would expect to also measure significant release of ATP within the serosal chamber following urothelial hypoosmotic stimulation and (3) Urothelial ATP does not diffuse through or stimulate deeper tissue layers to release ATP into the serosal compartment.
Fig. 4. Resting release of ATP into the serosal compartment. Compared to findings on the urothelial side of the bladder, there was no significant change in basal efflux of serosal ATP following SCI (n = between 4 and 5, P > 0.6 using non-parametric ANOVA).
ples was calculated and given as the resting release value for the three experimental groups (e.g. normal rats, SCI rats and SCI + BTX-A rats. In contrast to our findings on the urothelial side of the bladder, there was no significant change in basal efflux of serosal ATP following SCI (P > 0.6) (Fig. 4). 3.2.2. Hypoosmotic stimulation Using the same paradigm of hypoosmotic stimulation, the serosal side was exposed to reduced osmolarity Krebs (e.g. 60%) and the ATP content was measured in 5 min effluents. The ATP release was calculated by integrating the evoked ATP efflux curve. As shown in Fig. 5, serosal release of ATP was not significantly increased in SCI bladders when compared to normal bladders (P > 0.5). In addition, intravesical instillation of BTX-A was not able to significantly reduce the amount of ATP released from the SCI serosal surface after hypoosmotic stimulation suggesting that the mechanism of ATP release from the serosal side does not involve vesicular exocytosis. In addition to measuring serosal release of ATP to serosal hypoosmotic stimulation, in a subset of normal and SCI
3.2. Serosal ATP release 3.2.1. Resting release Similar to urothelial experiments, the first two samples after a 60 min equilibrium period and before hypoosmotic stimulation was considered as a measure of the serosal resting release of ATP. The average of the two initial sam-
Fig. 5. Hypoosmotic stimulation evoked release of ATP into the serosal compartment. Following SCI, no significant difference in evoked ATP release was demonstrated (n = between 4 and 5, P > 0.5 using non-parametric ANOVA).
M. Khera et al. / Neurochemistry International 45 (2004) 987–993
animals, we also investigated the effect of serosal stimulation on urothelial ATP release. We found no significant urothelial release of ATP in either group to serosal hypoosmotic stimulation (data not shown), demonstrating that: (1) serosal hypoosmotic stimulation does not stimulate urothelial sources of ATP, (2) urothelial and serosal chambers are physically separated (e.g. serosal ATP does not leak into urothelial chamber) and (3) serosal released ATP does not diffuse through tissue and induce urothelial release of ATP.
4. Discussion The main findings of these experiments are that, after spinal cord injury: (1) resting release of ATP from bladder urothelium is increased, (2) there is an increased release of ATP from the urothelial but not the serosal side of the bladder following hypoosmotic stimulation and (3) urothelial but not serosal hypoosmotic stimulation-evoked ATP release is inhibited by BTX-A instilled into the bladder lumen. Our results have significance in several respects. First, the increase in basal and hypoosmotic stimulation induced ATP release after SCI may evoke an increase in the frequency of bladder contractions by activating P2X3 receptors on afferent nerve terminals (Vlaskovska et al., 2001). Second, the fact that BTX-A pretreatment significantly reduced ATP release from the urothelium supports the concept that BTX-A treatment inhibits transmitter release not only from efferent nerve endings but also from the sensory nerve terminals and/or urothelium. ATP is well established as a co-transmitter that is released along with ACh and NE from peripheral tissues. Although ATP release has been widely demonstrated from efferent cholinergic and adrenergic terminals (Sperlagh and Vizi, 1992; Todorov et al., 1996; Tong et al., 1997), ATP release from bladder urothelial cells has only recently been confirmed within the scientific literature (Ferguson et al., 1997; Ferguson, 1999; Sun et al., 2001; Birder et al., 2002). Since discovering that ATP release can be generated by increasing the hydrostatic pressure on the urothelial side of the bladder (Ferguson et al., 1997; Birder et al., 2002) or in the lumen of the ureter (Knight et al., 2002), it has become increasingly evident that urothelium plays a significant role as a pressure sensing system, utilizing the neurotransmitter ATP in response to mechanical stimulation. This concept has been proven not only in intact ex vivo bladder tissue experiments but also in urothelial cell culture preparations in response to stretch (Sun et al., 2001) or hypoosmotic stress (Birder et al., 2002). In addition, increased stretch activated ATP release was reported from human urothelial cells cultured from the bladders of patients with IC (Sun et al., 2001), a chronic inflammatory and painful bladder disorder in which alterations in urothelium and afferent nerve function are thought to be prominent etiologic components (Steers and Tuttle, 1997).
991
SCI bladders also demonstrate increased afferent nerve activity that can be measured by cystometrogram (CMG) as an increase in the frequency of bladder contractions (Cheng et al., 1995; Khera et al., 2004) compared to that of normal bladders. In conjunction with these findings, the resting release of ATP from the urothelium of SCI bladders was higher in our experiments than ATP levels in normal bladders. However, the resting release of ATP from the serosal side did not increase after SCI despite the known presence of sensory nerve endings within the bladder smooth muscle layer (Gabella and Davis, 1998). We hypothesize that serosal basal efflux of ATP includes sources other than sensory nerves such as efferent nerve terminals or smooth muscle itself whose basal ATP release may not be directly affected by SCI. Not surprisingly, the resting release of ATP on the urothelial or serosal side did not respond to treatment with BTX-A. Since BTX-A inhibits SNAP-25 mediated vesicle docking and exocytotic release of transmitters, these findings imply that the SNARE protein SNAP-25 does not play an active role in the resting release of ATP. One might suggest that urothelial integrity is impaired after SCI such that ATP is simply leaking out from damaged urothelial cells. However, it has recently been shown that transepithelial resistance, a marker of the permeability of bladder urothelium, is unchanged after chronic spinal cord injury (Apodaca et al., 2003). Diffusion of ATP from the serosal side to the urothelial side is also improbable under resting conditions because urothelial ATP release was significantly higher in SCI bladders than in normal bladders, whereas there was no difference in resting ATP release between both groups on the serosal side (Figs. 2 and 4). The most relevant difference between normal and SCI bladders was the markedly elevated release of ATP from the urothelial side of SCI bladders after hypoosmotic stimulation. Epithelial cells have been shown to respond to hypoosmotic shock by releasing ATP in airway (Taylor et al., 1998) and intestine (van der Wijk et al., 2003) as well as in cultured urothelial cells (Birder et al., 2002). We were somewhat surprised to find the relatively small increase in ATP release after hypoosmotic shock in normal bladder urothelium as compared to its corresponding resting ATP release. It appears that urothelial cells integrated in our whole bladder preparation are more resistant to hypoosmostic stimulation evoked volume change than are solitary urothelial cells in primary culture. For this reason, hypoosmotic stimulation evokes only a small degree of ATP release. This phenomenon, however, does not apply to all changes in fluid osmolarity since a significant decrease in transmembrane resistance was observed in rabbit urothelium exposed to hyperosmotic solutions (Ferguson, 1999). However, changes in resistance do not necessarily correlate with changes in ATP release. Nevertheless, this finding can have important physiologic implications since bladder urothelium is constantly exposed to urine of differing osmolarities. Given our findings, it appears that normal bladders
992
M. Khera et al. / Neurochemistry International 45 (2004) 987–993
are less sensitive to hypoosmotic compared to hyperosmotic stimuli. In addition, under pathological conditions such as SCI, hypoosmotic shock evokes a significant release of ATP from bladder urothelium indicating that urothelium has become much more sensitive to, at least, decreases in urine osmolarity. Another significant finding of these experiments was that BTX-A instilled into the bladder lumen inhibited the release of ATP from the urothelial but not the serosal side of the bladder. Mechanical stress induces release of ATP from various epithelial tissues including intestine and bladder (Birder et al., 2002; Ferguson et al., 1997; van der Wijk et al., 2003). Moreover, hypoosmotic evoked ATP release can be inhibited by BTX or by depletion of intracellular calcium stores (van der Wijk et al., 2003). In addition, calcium-dependent ATP release has also been reported in ureteral epithelium (Knight et al., 2002) further suggesting that ATP release from epithelial tissues occurs via exocytotic mechanisms. Evoked release of ATP, at least from bladder urothelium, is not inhibited by tetrodotoxin implying that urothelial ATP release is not significantly affected by neuronal activity (Ferguson et al., 1997). Our results demonstrating that hypoosmotic stimulation of bladder urothelium evokes a significant release of ATP which is markedly inhibited by BTX-A suggests that: (1) ATP release from bladder urothelium is exocytotic in nature and (2) Impairment of urothelial ATP release is the one of the major mechanisms by which BTX-A reduces afferent activation following pathologic insult such as SCI. In contrast to urothelium, basal and evoked serosal ATP release was not significantly different between normal and SCI bladders. Since release of serosal ATP was similar in normal and SCI bladders, we speculate that smooth muscle may serve as a significant source of ATP, masking release from other areas (i.e. afferent or efferent nerves). In addition, neither basal nor evoked serosal ATP release was affected by BTX-A instilled into the bladder lumen after protamine sulfate treatment. Protamine sulfate is a caustic agent known to disrupt the integrity of epithelium and to dramatically decrease transmembrane resistance (Niku et al., 1994). However, normal resistance and urothelial integrity is restored within a few days after treatment (Apodaca et al., 2003). We applied protamine sulfate to allow BTX-A access to urothelium and suburothelial afferent nerves. The ineffectiveness of BTX-A to inhibit evoked serosal release of ATP may signify that BTX-A could not penetrate deeper layers of the bladder such that its effects were confined to superficial bladder tissues (e.g. urothelium and afferent nerve terminals). Although elevated amounts of ATP are released on the serosal side of the bladder, this release does not appear to significantly affect the afferent system adjacent to the urothelium. In our experiments, we could not detect any overflow of ATP from the serosal to urothelial sides or vice versa. Thus, ubiquitously released ATP appears to be localized within the bladder in such a way that ATP released from deeper lay-
ers (i.e. serosal hypoosmotic stimulation) does not significantly modulate afferent signals originating from urothelium in normal or pathologic (SCI) states. What is the significance of the increased release of ATP in SCI bladders? The prevailing notion is that urothelium serves as a pressure sensor whose function may be modulated under inflammatory (Vemulakonda et al., 2004) or neurological (spinal cord injury) conditions (Apodaca et al., 2003; Khera et al., 2004). While there is no evidence that urothelium is wired directly to afferent nerves by forming synapses with intraepithelial or subepithelial afferent fibers (Gabella and Davis, 1998), it can communicate with the sensory nerves via release of transmitters such as ATP (Birder et al., 2002). Thus, the connection between urothelium and the afferent nerve endings is non-synaptic, a form of interaction already described amongst efferent nerve terminals (e.g. adrenergic and cholinergic) within the peripheral (Vizi, 1979; Somogyi and de Groat, 1990) and the central nervous system (Vizi, 2000). Higher amounts of ATP release following SCI could activate P2X3 receptors in epithelial and subepithelial bladder layers to increase afferent nerve activity and, consequently, account for the higher frequency of bladder contractions seen in both human and animal models of SCI. The finding that BTX-A significantly impairs urothelial ATP release following SCI gives another possible explanation for its clinical efficacy in treating human patients with neurogenic bladders after SCI. In addition, by inhibiting mechanosensory mechanisms within the bladder, use of BTX-A may be extended to treat other disorders of increased bladder afferent activity such as interstitial cystitis, a chronic, debilitating inflammatory disorder characterized by symptoms of urinary frequency, urgency, and pelvic pain. In summary, there is an increased release of ATP from the urothelium of SCI bladders that may contribute to the development of bladder hyperactivity by activating P2X3 receptors on epithelial and subepithelial afferent nerve endings. Since afferent nerve endings are not forming synapses with urothelial cells, the interaction appears to be non-synaptic in nature. ATP release evoked by hypoosmotic stimulation appears to occur through exocytotic mechanisms that can be inhibited by BTX-A, a finding that helps explain the inhibitory effects of BTX-A on bladder sensory mechanisms. These results may have important relevance in our understanding of the mechanisms underlying plasticity of bladder afferent pathways following SCI.
Acknowledgements We express our sincere gratitude to K. Roger Aoki, Ph.D., for valuable discussions and providing the BTX-A (Botox® , Allergan, Irvine, CA) for our experiments. This work was supported by the NIH grant RO1 DK 45741; The Scott Department of Urology Neurourology Fund; and an Unrestricted Educational Grant from Allergan, Irvine, CA.
M. Khera et al. / Neurochemistry International 45 (2004) 987–993
References Apodaca, G., Kiss, S., Ruiz, W., Meyers, S., Zeidel, M., Birder, L., 2003. Disruption of bladder epithelium barrier function after spinal cord injury. Am. J. Physiol. 284, 966–976. Birder, L.A., Nakamura, Y., Kiss, S., Nealen, M.L., Barrick, S., Kanai, A.J., Wang, E., Ruiz, G., de Groat, W.C., Apodaca, G., Watkins, S., Caterina, M.J., 2002. Altered urinary bladder function in mice lacking the vanilloid receptor TRPV1. Nat. Neurosci. 5, 856–860. Cheng, C.L., Ma, C.P., de Groat, W.C., 1995. Effect of capsaicin on micturition and associated reflexes in chronic spinal rats. Brain Res. 678, 40–48. Cui, M., Aoki, K.R., 2000. Botulinum toxin type A (BTX-A) reduces inflammatory pain in the rat formalin model. Cephalagia 20, 414. De Groat, W.C., Kawatani, M., Hisamitsu, T., Cheng, C.L., Ma, C.P., Thor, K., Steers, W., Roppolo, J.R., 1990. Mechanisms underlying the recovery of urinary bladder function following spinal cord injury. J. Autonomic Nervous Sys. 30, 71–77. Ferguson, D.R., Kennedy, I., Burton, T.J., 1997. ATP is released from rabbit urinary bladder epithelial cells by hydrostatic pressure changes—a possible sensory mechanism? J. Physiol. 505, 503–511. Ferguson, D.R., 1999. Urothelial function. Br. J. Urol. Int. 84, 235–242. Gabella, G., Davis, C., 1998. Distribution of afferent axons in the bladder of rats. J. Neuro. Cytol. 27, 141–155. Khera, M., Somogyi, G.T., Salas, N., Vemulakonda, V.M., Evans, J.L., Kiss, S., Smith, C.P., Boone, T.B., 2004. Effects of botulinum toxin with protamine on bladder contractility after spinal cord injury. J. Urol. 171, 459. Klapproth, H., Reinheimer, T., Metzen, J., Munch, M., Bittinger, F., Kirkpatrick, C.J., Hohle, K.D., Schemann, M., Racke, K., Wessler, I., 1997. Non-neuronal acetylcholine, a signaling molecule synthesized by surface cells of rat and man. Naunyn Schmiedebergs Arch. Pharmacol. 355, 515–523. Knight, G.E., Bodin, P., de Groat, W.C., Burnstock, G., 2002. ATP is released from guinea pig ureter epithelium on distension. Am. J. Physiol., Renal Physiol. 282, 281–288. Maggi, C.A., 1993. The dual, sensory and efferent functions of the capsaicin-sensitive primary sensory nerves in the bladder and urethra. In: Maggi, C.A. (Ed.), The Autonomic Nervous System: Nervous Control of the Urogenital System, vol. 3. Harwood Academic Publishers, London, pp. 383 (Chapter 11). Niku, S.D., Stein, P.C., Scherz, H.C., Parsons, C.L., 1994. A new method for cytodestruction of bladder epithelium using protamine sulfate and urea. J. Urol. 152, 1025–1028. Reinheimer, T., Bernedo, P., Klapproth, H., Oelert, H., Zeiske, B., Racke, K., Wessler, I., 1996. Acetylcholine in isolated airways of rat, guinea pig, and human: species differences in role of airway mucosa. Am. J. Physiol. 270, 722–728. Schiavo, G., Santucci, A., DasGupta, B.R., et al., 1993. Botulinum neurotoxins serotypes A and E cleave Snap-25 at distinct COOH-terminal peptide bonds. FEBS Lett. 335, 99. Smith, C.P., Boone, T.B., de Groat, W.C., Chancellor, M.B., Somogyi, G.T., 2003a. Effect of stimulation intensity and botulinum toxin isoform on rat bladder strip contractions. Brain Res. Bull. 61, 165–171. Smith, C.P., Chancellor, M.B., 2004. Emerging role of botulinum toxin in the treatment of voiding dysfunction. J. Urol. 171, 2128–2137.
993
Smith, C.P., Franks, M.E., McNeil, B.K., Ghosh, R., de Groat, W.C., Chancellor, M.B., Somogyi, G.T., 2003b. Effect of botulinum toxin A on the autonomic nervous system of the rat lower urinary tract. J. Urol. 169, 1896–1900. Somogyi, G.T., de Groat, W.C., 1990. Modulation of the release of [3H] norepinephrine from the base and body of the rat urinary bladder by endogenous adrenergic and cholinergic mechanicsms. J. Pharmacol. Exp. Ther. 255, 204–210. Sperlagh, B., Vizi, E.S., 1992. Is the neuronal ATP release from guinea-pig vas deferens subject to alpha2-adrenoceptor-mediated modulation? Neuroscience 51, 203–209. Steers W.D., Tuttle J.B., 1997. Neurogenic inflammation and nerve growth factor: possible roles in interstitial cystitis. In: Sant, G.R. (Ed.), Interstitial Cystitis. Lippincott-Raven, Philadelphia, pp. 67– 75. Sun, Y., Keay, S., De Deyne, P.G., Chai, T.C., 2001. Augmented stretch activated adenosine triphosphate release from bladder uroepithelial cells in patients with interstitial cystitis. J. Urol. 166, 1951–1956. Taylor, A.L., Kudlow, B.A., Marrs, K.L., Gruenert, D.C., Guggino, W.B., Schwiebert, E.M., 1998. Bioluminescence detection of ATP release mechanisms in epithelia. Am. J. Physiol. 275, 1391–1406. Todorov, L.D., Mihaylova-Todorova, S., Craviso, G.L., Bjur, R.A., Westfall, D.P., 1996. Evidence for the differential release of the cotransmitters ATP and noradrenaline from sympathetic nerves of the guinea-pig vas deferens. J. Physiol. 496, 731–748. Tong, Y.C., Hung, Y.C., Shinozuka, K., Kunitomo, M., Cheng, J.T., 1997. Evidence of adenosine 5 -triphosphate release from nerve and P2X-purinoceptor mediated contraction during electrical stimulation of rat urinary bladder smooth muscle. J. Urol. 158, 1973–1977. van der Wijk, T., Tomassen, S.F., Houtsmuller, A.B., de Jonge, H.R., Tilly, B.C., 2003. Increased vesicle recycling in response to osmotic cell swelling. Cause and consequence of hypotonicity-provoked ATP release. J. Biol. Chem. 278, 40020–40025. Vemulakonda, V.M., Smith, C.P., Kiss, S., Khera, M., Evans, J.L., Boone, T.B., Somogyi, G.T., 2004a. Botulinum toxin A inhibits mechanoreceptor-mediated urothelial release of ATP in conditions of bladder inflammation. J. Urol. 171 (Suppl), 95. Vemulakonda, V.M., Somogyi, G.T., Kiss, S., Evans, J.L., Khera, M., Boone, T.B., Smith, C.P., 2004b. Intravesical botulinum toxin A inhibits the afferent response in the chronically inflamed bladder. J. Urol. 171 (Suppl), 458. Vizi, E.S., 1979. Presynaptic modulation of neurochemical transmission. Prog. Neurobiol. 12, 181–290. Vizi, E.S., 2000. Role of high-affinity receptors and membrane transporters in nonsynaptic communication and drug action in the central nervous system. Pharmacol. Rev. 52, 63–89. Vlaskovska, M., Kasakov, L., Rong, W., Bodin, P., Bardini, M., Cockayne, D.A., Ford, A.P., Burnstock, G., 2001. P2X3 knock-out mice reveal a major sensory role for urothelially released ATP. J. Neurosci. 21, 5670–5677. Wessler, I., Kirkpatrick, C.J., Racke, K., 1998. Non-neuronal acetylcholine, a locally acting molecule, widely distributed in biological systems: expression and function in humans. Pharmacol. Ther. 77, 59–79. Yoshimura, N., de Groat, W.C., 1997. Plasticity of Na+ channels in afferent neurons innervating rat urinary bladder following spinal cord injury. J. Physiol. 503, 269–276.