Intravesical protamine sulfate and potassium chloride as a model for bladder hyperactivity

BASIC SCIENCE

INTRAVESICAL PROTAMINE SULFATE AND POTASSIUM CHLORIDE AS A MODEL FOR BLADDER HYPERACTIVITY YAO-CHI CHUANG, MICHAEL B. CHANCELLOR, SATOSHI SEKI, NAOKI YOSHIMURA, PRADEEP TYAGI, LEAF HUANG, JOHN P. LAVELLE, WILLIAM C. DE GROAT, AND MATTHEW O. FRASER

ABSTRACT Objectives. An acute animal model for hyperactive bladder in rats was developed using intravesical infusion of protamine sulfate (PS), an agent thought to break down urothelial barrier function, and physiologic concentrations of potassium chloride (KCl). Methods. Continuous cystometrograms (CMGs) were performed in urethane-anesthetized female rats by filling the bladder (0.04 mL/min) with normal saline followed by intravesical infusion of a test solution consisting of either KCl (100 or 500 mM) or PS (10 or 30 mg/mL) for 60 minutes. Subsequently, the 10 mg/mL PS-treated animals were infused intravesically with 100, 300, or 500 mM KCl. Some animals were pretreated with capsaicin (125 mg/kg subcutaneously) 4 days before the experiments. Results. Unlike KCl (100 or 500 mM) or a low concentration of PS (10 mg/mL) alone, the intravesical administration of a high concentration of PS (30 mg/mL) produced irritative effects with a decreased intercontraction interval (by 80.6%). After infusion of a low concentration of PS, infusion of 300 or 500 mM KCl produced irritative effects (intercontraction interval decreased by 76.9% or 82.9%, respectively). The onset of irritation occurred more rapidly after 500 mM KCl (10 to 15 minutes) than after 300 mM KCl (20 to 30 minutes). Capsaicin pretreatment delayed the onset (approximately 60 minutes) and reduced the magnitude (intercontraction interval decreased by 35.5%) of irritative effects. Conclusions. Intravesical administration of KCl after PS treatment activates capsaicin-sensitive afferents and detrusor muscle and presumably capsaicin-resistant afferents. Modest, noncytotoxic affronts to urothelial barrier function can result in dramatic irritative responses. This model may be useful in the study of bladder irritation and hyperactivity. UROLOGY 61: 664–670, 2003. © 2003, Elsevier Science Inc.

B

ladder hypersensitivity conditions, such as interstitial cystitis, are characterized by bladder pain, urinary frequency, urgency, and sterile urine. Although the pathogenesis of bladder hypersensi-

This study was supported by the Interstitial Cystitis Association and Fishbein Family Foundation, a grant from the Taiwan Government (NSC90-2314-B-182A-149), and grants NICHD HD39768 and NIDDK K12 DK02656. From the Division of Urology, Chang Gung Memorial Hospital, Kaohsiung; Department of Urology, National Yang Ming University School of Medicine, Taipei, Taiwan; Departments of Urology and Pharmacology, University of Pittsburgh School of Medicine; University of Pittsburgh School of Pharmacy, Pittsburgh, Pennsylvania; and Department of Urology, University of North Carolina School of Medicine, Chapel Hill, North Carolina. Reprint requests: Michael B. Chancellor, M.D., Department of Urology, University of Pittsburgh School of Medicine, 3471 Fifth Avenue, Kaufmann Building, Suite 700, Pittsburgh, PA 15213 Submitted: April 2, 2002, accepted (with revisions): September 30, 2002

664

© 2003, ELSEVIER SCIENCE INC. ALL RIGHTS RESERVED

tivity is uncertain, it has been proposed that a leaky urothelium plays a major role.1,2 This results in the transepithelial migration of solutes, such as potassium, that can depolarize subepithelial afferent nerves and provoke irritative urinary symptoms.3,4 An animal model of irritating cystitis that involves the movement of a neuroexcitatory substance through a leaky urothelium would be useful. Current animal models relevant to bladder hypersensitivity and hyperactivity use the application of noxious chemicals, such as acetic acid, xylene, mustard oil, turpentine, cyclophosphamide, and protamine sulfate (PS)/lipopolysaccharide, that induce inflammation or injury of the bladder wall.5–9 These methods damage the bladder but are not likely to produce the more subtle urothelial dysfunction associated with bladder hypersensitivity and hyperactivity. In the present study, we combined PS treatment, thought to break down 0090-4295/03/$30.00 doi:10.1016/S0090-4295(02)02280-X

TABLE I. Effects of 100 mM and 500 mM KCl on CMG parameters Control (n ⴝ 4) PT (cm H2O) Amplitude (cm H2O) Compliance (mL/cm H2O) ICI (min)

7.2 25.5 0.206 13.9

⫾ ⫾ ⫾ ⫾

0.5 1.4 0.013 2.4

After 100 mM KCl (n ⴝ 4) 5.9 28.2 0.205 13.8

⫾ ⫾ ⫾ ⫾

0.5 0.5 0.032 2.0

Control (n ⴝ 4) 7.3 26.5 0.268 15.6

⫾ ⫾ ⫾ ⫾

0.8 1.1 0.013 2.1

After 500 mM KCl (n ⴝ 4) 8.2 28.8 0.237 15.8

⫾ ⫾ ⫾ ⫾

0.9 1.2 0.023 2.4

KEY: KCl ⫽ potassium chloride; CMG ⫽ cystometrography; PT ⫽ pressure threshold; ICI ⫽ intercontraction interval. Data presented as the mean ⫾ SE. No statistically significant differences were observed between 100 and 500 mM KCl treatment.

urothelial barrier function,8 and physiologic concentrations of potassium chloride (KCl). To understand the effects of potassium on bladder sensory pathways and detrusor muscle, some animals were pretreated with capsaicin to desensitize the C fiber afferents and another group of animals were treated with either intravenous hexamethonium injection or transection of bilateral pelvic nerves to suppress the micturition reflex. MATERIAL AND METHODS ANIMAL PREPARATION The study was performed on 40 female Sprague-Dawley rats weighing 250 to 300 g. The animals were anesthetized with 1.2 g/kg urethane injected subcutaneously. The body temperature was maintained in the physiologic range using a heating lamp.

CYSTOMETROGRAPHY PE-50 tubing (Clay-Adams, Parsippany, NJ) was inserted into the bladder through the urethra and connected by a threeway stopcock to a pressure transducer and syringe pump to record the intravesical pressure and infuse solutions into the bladder. A control cystometrogram (CMG) was performed by slowly filling the bladder with saline (0.04 mL/min) to elicit repetitive voiding. The amplitude, pressure threshold, compliance, and intercontraction interval (ICI) of the reflex bladder contractions were recorded. The pressure threshold is the pressure that induces the first bladder contraction and has often been used as a parameter that corresponds to afferent nerve activity to induce reflex bladder contractions. The measurements in each animal represent the average of three to five bladder contractions.

INDUCTION OF HYPERACTIVE BLADDER After performing control CMGs with saline infusion, three types of experiments were performed by intravesical infusion of various solutions: (a) KCl (100 or 500 mM in saline) for 1 hour (n ⫽ 4 in each group), (b) PS (30 mg/mL in saline; Sigma Chemical) for 1 hour (n ⫽ 6), (c) PS (10 mg/mL) for 1 hour, followed by 100, 300, or 500 mM KCl for 1 hour (n ⫽ 6 in each subgroup). In 4 rats, capsaicin dissolved in a vehicle containing 10% ethanol, 10% Tween-80, and 80% physiologic saline, at a concentration of 20 mg/mL, was given subcutaneously in divided doses on 2 consecutive days: 25 and 50 mg/kg at a 12-hour interval on the first day and 50 mg/kg on the second day.10 All injections were performed under halothane anesthesia. Four days after the last dose of capsaicin, the animals were anesthetized and treated with intravesical administration of PS (10 mg/mL) for 1 hour followed by KCl (500 mM) infusion. UROLOGY 61 (3), 2003

SUPPRESSION OF MICTURITION REFLEX To evaluate the direct effects of potassium on the detrusor muscle, the micturition reflex was blocked in animals by either intravenous hexamethonium injection (25 mg/kg; n ⫽ 2) or transection of bilateral pelvic nerves (n ⫽ 2).

STATISTICAL ANALYSIS

Quantitative data are expressed throughout as the mean ⫾ standard error. Statistical analyses were performed using Student’s t test for paired or unpaired data as applicable, with P ⬍0.05 considered statistically significant.

RESULTS CMGS IN VARIOUS INFUSION GROUP As shown in Table I , the CMG parameters during infusion of 100 or 500 mM KCl alone were not significantly different from those during saline administration. These results indicate that the bladder barrier function was not affected by the control condition. The 1-hour period of instillation with 10 mg/mL PS did not show a significant change compared with saline administration (Fig. 1 and Table II). These results revealed that low-dose PS itself was not an irritant to the bladder. However, instillation of 30 mg/mL PS into the bladder resulted in an irritative effect (ICI decrease of 80.6%, compliance decrease of 63.6%, and pressure threshold increase of 36.8%) after a delay of 40 to 45 minutes (Fig. 1 and Table II). EFFECT OF KCl AFTER LOW PS CONCENTRATION INFUSION As shown in Figure 2 and Table III, CMGs performed with a 300 or 500 mM KCl infusion significantly changed the ICI (76.9% or 82.9% decrease, respectively), compliance (60% or 63.4% decrease, respectively), and contraction amplitude (23.7% or 21.4% increase, respectively). However, the pressure threshold was not significantly changed. The effect of 300 mM KCl occurred after a delay of 20 to 30 minutes, and the effect of 500 mM KCl occurred after about a 10 to 15-minute delay. This implies that a higher concentration of KCl has a faster penetration rate. A 1-hour infusion of 100 mM KCl did not result in a significant change in the CMG parameters. 665

FIGURE 1. CMG tracing during control and different concentrations of PS treatment. High concentrations of PS induced bladder hyperactivity (decreased ICI), and low concentrations of PS had no effect. (A) Control CMG before PS treatment. (B) CMG after low concentration of PS. (C) Control CMG before PS treatment. (D) CMG after high concentration of PS.

TABLE II. Effects of 10 and 30 mg/mL PS on CMG parameters Control (n ⴝ 18) PT (cm H2O) Amplitude (cm H2O) Compliance (mL/cm H2O) ICI (min)

4.2 24.1 0.169 16.8

⫾ ⫾ ⫾ ⫾

0.8 2.4 0.012 0.9

After 10 mg/mL PS (n ⴝ 18) 3.4 25.3 0.176 17.8

⫾ ⫾ ⫾ ⫾

0.9 2.0 0.013 0.7

Control (n ⴝ 6) 5.7 27.7 0.228 13.9

⫾ ⫾ ⫾ ⫾

0.6 0.8 0.024 2.1

After 30 mg/mL PS (n ⴝ 6) 7.8 28.3 0.083 2.7

⫾ ⫾ ⫾ ⫾

1.1* 3.1 0.022* 0.7*

KEY: PS ⫽ protamine sulfate; other abbreviations as in Table I. Data presented as the mean ⫾ SE. *P ⬍ 0.05, significantly different from control.

CMGS IN RATS PRETREATED WITH CAPSAICIN In the capsaicin-pretreated animals, the hyperactive bladder evoked by sequential infusion of PS (10 mg/mL) and KCl (500 mM) was delayed by about 1 hour and the changes in ICI and compliance were reduced by 36% and 55%, respectively, after 2 hours of infusion (ICI decreased from 20.3 666

⫾ 1.2 to 13.1 ⫾ 2.8 minutes; compliance decreased from 0.224 to 0.100 mL/cm H2O; Table IV). In addition, the ICI (13.1 ⫾ 2.8 minutes) after sequential infusion of PS (10 mg/mL) and KCl (500 mM) in these capsaicin-pretreated rats (Table IV) was significantly (P ⬍0.05) longer than the ICI (3.2 ⫾ 1.3 minutes) reduced by intravesical appliUROLOGY 61 (3), 2003

FIGURE 2. CMG tracing during control and different concentrations of KCl treatment after 1 hour of PS (10 mg/mL) administration. High concentrations of KCl induced bladder hyperactivity (decreased ICI), and low concentrations had no effect. (A) Control CMG before KCl treatment. (B) CMG during 100 mM KCl treatment. (C) Control CMG before KCl treatment. (D) CMG during 300 mM KCl treatment. (E) Control CMG before KCl treatment. (F) CMG during 500 mM KCl treatment.

TABLE III. Effects of 100, 300, and 500 mM KCl on CMG parameters after 10 mg/mL PS treatment Control (n ⴝ 6)

After 100 mM KCl (n ⴝ 6)

PT (cm H2O) 3.1 ⫾ 1.2 26.8 ⫾ 3.7 Amplitude (cm H2O) Compliance 0.161 ⫾ 0.022 (mL/cm H2O) ICI (min) 17.2 ⫾ 1.3

3.0 ⫾ 1.3 28.2 ⫾ 3.7

Control (n ⴝ 6) 2.9 ⫾ 0.6 25.3 ⫾ 0.9

0.157 ⫾ 0.022 18.8 ⫾ 2.1

After 300 mM KCl (n ⴝ 6) 3.3 ⫾ 0.9 31.3 ⫾ 0.7*

Control (n ⴝ 6) 4.0 ⫾ 0.9 24.3 ⫾ 2.1

After 500 mM KCl (n ⴝ 6) 5.4 ⫾ 1.0 29.5 ⫾ 3.4*

0.165 ⫾ 0.027 0.066 ⫾ 0.012* 0.205 ⫾ 0.018 0.075 ⫾ 0.009* 17.4 ⫾ 2.0

5.2 ⫾ 1.7*

18.8 ⫾ 1.7

3.2 ⫾ 1.3*

Abbreviations as in Tables I and II. Data presented as the mean ⫾ SE. *P ⬍ 0.05, significantly different from control; statistically significant differences observed between control and 300/500 mM KCl treatment.

TABLE IV. Effects of KCl (500 mM) after 1-hour infusion of PS (10 mg/mL) on CMG parameters in capsaicin-pretreated animals KCl Control (n ⴝ 4) PT (cm H2O) Amplitude (cm H2O) Compliance (mL/cm H2O) ICI (min)

8.4 18.3 0.244 20.3

⫾ ⫾ ⫾ ⫾

0.9 1.5 0.007 1.2

1 hr (n ⴝ 4) 5.8 20.3 0.218 21.3

⫾ ⫾ ⫾ ⫾

0.5 0.3 0.013 2.8

2 hr (n ⴝ 4) 7.6 16.0 0.100 13.1

⫾ ⫾ ⫾ ⫾

1.7 1.7 0.016* 2.8*

Abbreviations as in Tables I and II. Data presented as the mean ⫾ SE. *P ⬍ 0.05, significantly different from control.

cation of PS (10 mg/mL) and KCl (500 mM) in the untreated rats (Table III), indicating that C fiber desensitization by pretreatment of capsaicin suppressed PS/KCl-induced bladder hyperactivity. CMGS IN MICTURITION REFLEX-SUPPRESSED ANIMALS As shown in Figure 3, infusion of normal saline did not induce the micturition reflex in the rats that had undergone hexamethonium injection or transection of the pelvic nerves, indicating that these procedures blocked the micturition reflex. Infusion of KCl (500 mM) after PS (10 mg/mL) instillation resulted in a compliance decrease of 55.9% (from 0.093 ⫾ 0.026 to 0.041 ⫾ 0.011 mL/cm H2O). This revealed that potassium has effects on the direct stimulation of the detrusor muscle and results in decreased compliance. COMMENT The results of our study revealed that low-dose PS (10 mg/mL) is not a bladder irritant, but a noncytotoxic affront to urothelial barrier function and that the use of “physiologic” normal saline versus more appropriately physiologic 300 or 500 mM KCl,11,12 for cystometry affects our understanding of the function of the lower urinary tract in animal models of the hyperactive bladder. Although the exact nature of the pathogenesis of bladder hypersensitivity and interstitial cystitis 668

remains unknown, it has been proposed that a critical component of interstitial cystitis is a leaky urothelium.1,2,13 This condition of a compromised urothelial barrier results in an influx of highly concentrated noxious substances that normally pass through the entirety of the urinary tract without reabsorption.3,4 However, when the urothelial barrier function is broken down, they cross back into the bladder, where they stimulate activity of resident C fiber afferents—those that carry pain sensation—and cause sensory symptoms.4 This has been hypothesized to be due primarily to the influx of concentrated potassium from the urine to the submucosa, which depolarizes bladder wall sensory afferents and initiates hyperactive bladder.3,4 Although the site and mechanism of action of KCl to induce bladder hyperactivity is still uncertain, it is clear that bladder afferents must play a key role. The stimulation of silent C fibers has been implicated in the etiology of some hyperactive bladders; A␦ afferents are usually thought to be involved in normal voiding function.10 However, in rats pretreated with capsaicin to desensitize the C fibers, we found that the effect of intravesical KCl was delayed and reduced but not completely eliminated. These results presumably provide indirect evidence that sensitization of myelinated A␦ afferents may also play a role in the bladder hyperactivity induced by KCl. However, high concentraUROLOGY 61 (3), 2003

FIGURE 3. CMG tracing during control and KCl (500 mM) infusion after PS (10 mg/mL) infusion in micturition reflex-suppressed animals. KCl stimulated the detrusor muscle and decreased bladder compliance. (A) Control CMG before KCl treatment. (B) CMG tracing during KCl treatment.

tions of potassium penetrating through a leaky urothelium could also directly stimulate detrusor muscle and contribute to the decrease in bladder compliance.3,14 Our methods, which blocked the micturition reflex by an autonomic ganglion blockade (hexamethonium) or after pelvic nerve transection, but still demonstrated a decrease in compliance after intravesical infusion of KCl after PS treatment, have provided evidence of the stimulation of the detrusor by high concentrations of potassium. Thus, the ICI was decreased, but the pressure threshold did not change. High concentrations of potassium can still irritate the bladder neck and cause high outlet resistance.3 Therefore, the bladder contraction amplitude was elevated in some of our experiments. The use of PS is a well-established model for bladder injury that was developed previously in UROLOGY 61 (3), 2003

another laboratory by distension of the bladder for 45 minutes with 1 mL of PS (10 mg/mL).8 Clearly, prolonged overdistension of the bladder alters the properties of the bladder wall13,14 and might enhance the cytodestructive effects of PS and result in immediate urothelial sloughing.8 However, in our open CMG method, 1-hour exposure of urothelium to the same concentration of PS did not cause obvious CMG changes but compromised the urothelial barrier function, which resulted in the influx of high concentrations of potassium and stimulated the bladder. Before PS treatment, the same concentration of potassium did not induce a hyperactive bladder. These data support the concept that abnormal epithelial permeability with the addition of a high concentration of potassium in the urine is a key component to induce the symptoms of bladder hypersensitivity. We hypothesize that without the mechanical destruction of urothelium, prolonged exposure of PS might still lead to a subtle change but break down the barrier function. This model might elucidate the mechanism involved in the potassium test for the diagnosis of interstitial cystitis.4 Previously, other investigators have reported that intravesical instillation of 150 mM KCl through a pair of bladder dome catheters at a rate of 0.250 mL/min to a maximal pressure of 30 mm Hg can excite the afferent activity in hypogastric nerves, but afferent activity was rarely detected in the pelvic nerves.15 The average bladder volume used was 1.5 mL, about two to threefold the normal bladder capacity in rats and could result in overdistension of the bladder.15–17 Others have also shown the reduction of bladder capacity in a closed CMG method with intravesical isotonic KCl treatment after a period of delay, the effects of which were enhanced by pretreatment with 50% dimethyl sulfoxide.18 However, studies in our laboratory have revealed that the continuous infusion of KCl (500 mM) for 1 hour without pretreatment with PS did not induce significant bladder irritation. It is possible that altering the urothelium properties by either overdistension or dimethyl sulfoxide instillation could increase bladder permeability and induce afferent firing and hyperactive bladder by KCl administration. CONCLUSIONS The use of “physiologic,” normal saline versus more appropriately physiologic 300 or 500 mM KCl for cystometry greatly biases our understanding of the function of the lower urinary tract in animal models of hyperactive bladder. Modest, noncytotoxic affronts to urothelial barrier function can result in dramatic irritative responses, given the proper physiologic concentrations of KCl. 669

Whatever the mechanism underlying the excitatory effect of PS/KCl on the bladder, methods directed toward decreasing the action of PS/KCl may be effective in decreasing bladder irritation. REFERENCES 1. Parsons CL, Lilly JD, and Stein P: Epithelial dysfunction in nonbacterial cystitis (interstitial cystitis). J Urol 145: 732–735, 1991. 2. Parsons CL, Housley T, Schmidt JD, et al: Treatment of interstitial cystitis with intravesical heparin. Br J Urol 73: 504 –507, 1994. 3. Hohlbrugger G: Urinary potassium and the overactive bladder. Br J Urol 83: 22–28, 1999. 4. Parsons CL, Greenberger M, Gabal L, et al: The role of urinary potassium in the pathogenesis and diagnosis of interstitial cystitis. J Urol 159: 1862–1867, 1998. 5. Yu Y, and de Groat WC: Effects of ZD6169, a KATP channel opener, on bladder hyperactivity and spinal c-fos expression evoked by bladder irritation in rats. Brain Res 807: 11–18, 1998. 6. Birder LA, and de Groat WC: Increased c-fos expression in spinal neurons after irritation of the lower urinary tract in the rat. J Neurosci 12: 4878 –4889, 1992. 7. Ozawa H, Chancellor MB, Jung SY, et al: Effect of intravesical nitric oxide therapy on cyclophosphamide-induced cystitis. J Urol 162: 2211–2216, 1999. 8. Stein PC, Pham H, Ito T, et al: Bladder injury model induced in rats by exposure to protamine sulfate followed by bacterial endotoxin. J Urol 155: 1133–1138, 1996.

670

9. McMahon SB, Dmitrieva N, and Koltzenburg M: Visceral pain. Br J Anaesth 75: 132–144, 1995. 10. Cheng CL, Ma CP, and de Groat WC: Effects of capsaicin on micturition and associated reflexes in rats. Am J Physiol 265: R132–R138, 1993. 11. Ohnishi M, Yajima H, Takeuchi T, et al: Mechanism of urinary tract crystal formation following biphenyl treatment. Toxicol Appl Pharmacol 174: 122–129, 2001. 12. Morrison J, Wen J, and Kibble A: Activation of pelvic afferent nerves from the rat bladder during filling. Scand J Urol Nephrol Suppl 201: 73–75, 1999. 13. Keay S, Warren JW, Zhang CO, et al: Antiproliferative activity is present in bladder but not renal pelvic urine from interstitial cystitis patients. J Urol 162: 1487–1489, 1999. 14. Hohlbrugger G: The vesical blood-urine barrier: a relevant and dynamic interface between renal function and nervous bladder control. J Urol 154: 6 –15, 1995. 15. Moss NG, Harrington WW, and Tucker MS: Pressure, volume, and chemosensitivity in afferent innervation of urinary bladder in rats. Am J Physiol 272: R695–R703, 1997. 16. Leppilahti M, and Kallioninen M: Duration of increased mucosal permeability of the urinary bladder after acute overdistension: an experimental study in rats. Urol Res 27: 272– 276, 1999. 17. Chuang YC, Fraser MO, Yu Y, et al: The role of bladder afferent pathways in bladder hyperactivity induced by the intravesical administration of nerve growth factor. J Urol 165: 975–979, 2001. 18. Hohlbrugger G, and Lentsch P: Intravesical ions, osmolality and pH influence the volume pressure response in the normal rat bladder, and this is more pronounced after DMSO exposure. Eur Urol 11: 127–130, 1985.

UROLOGY 61 (3), 2003