ALTERED INDUCIBLE NITRIC OXIDE SYNTHASE EXPRESSION AND NITRIC OXIDE PRODUCTION IN THE BLADDER OF CATS WITH FELINE INTERSTITIAL CYSTITIS

0022-5347/05/1732-0625/0 THE JOURNAL OF UROLOGY® Copyright © 2005 by AMERICAN UROLOGICAL ASSOCIATION

Vol. 173, 625– 629, February 2005 Printed in U.S.A.

DOI: 10.1097/01.ju.0000145900.22849.1d

ALTERED INDUCIBLE NITRIC OXIDE SYNTHASE EXPRESSION AND NITRIC OXIDE PRODUCTION IN THE BLADDER OF CATS WITH FELINE INTERSTITIAL CYSTITIS L. A. BIRDER,*, † A. WOLF-JOHNSTON, C. A. BUFFINGTON, J. R. ROPPOLO, W. C. AND A. J. KANAI

DE

GROAT‡

From the Departments of Medicine (LAB, AW-J, AJK) and Pharmacology (LAB, JRR, WCdG, AJK), University of Pittsburgh School of Medicine, Pittsburgh, Pennsylvania, and Department of Clinical Veterinary Sciences (CAB), Ohio State University, Columbus, Ohio

ABSTRACT

Purpose: Alterations in nitric oxide (NO) levels have been demonstrated in some humans with interstitial cystitis (IC) as well as in chemically induced animal models of cystitis. Thus, in the current study we investigated whether inducible NO synthase (iNOS) mediated NO production is altered in the bladder of cats with a naturally occurring model of IC termed feline IC (FIC). Materials and Methods: We examined iNOS expression using Western immunoblotting and baseline NO production using an NO microsensor from smooth muscle and mucosal bladder strips in 9 healthy cats and 6 diagnosed with FIC. Results: There was a significant increase in baseline NO production in cats with FIC compared with that in healthy cats in smooth muscle and mucosal strips. This production was not ablated in the absence of extracellular Ca2⫹ (100 ␮M egtazic acid) or following incubation with the calmodulin antagonist trifluoroperazine (20 ␮M), indicating iNOS mediated Ca2⫹ independent NO production. Release was significantly decreased following incubation with the NOS antagonist L-NAME (N-nitro-L-arginine methyl ester) (100 ␮M). Furthermore, immunoblotting revealed a trend toward increased iNOS expression in smooth muscle and mucosal strips from FIC cats but not from healthy cats. Conclusions: In light of previous findings that the barrier property of the urothelial surface is disrupted in FIC and iNOS mediated increase in NO alters barrier function in other types of epithelium our findings suggest that iNOS dependent NO production may have a role in epithelial barrier dysfunction in FIC. KEY WORDS: bladder; cystitis, interstitial; nitric oxide; urothelium; cats

Interstitial cystitis (IC) is a chronic syndrome characterized by sensory symptoms such as urinary urgency, frequency and pain.1, 2 Although IC has been considered by some investigators to be an inflammatory disease of the bladder urothelium or mucosa, to our knowledge the pathological condition responsible for IC symptoms has not been identified. Alterations in the responsiveness of bladder sensory nerves due to decreased epithelial integrity or altered transmitter release may underlie part of the etiology of this syndrome. In cats there is a naturally occurring form of idiopathic cystitis termed feline IC (FIC) that demonstrates almost all of the characteristics of human IC, including most if not all of the symptoms.3 Studies have been hampered by the lack of an adequate animal model for this disease. Thus, this naturally occurring model of IC in cats may permit more relevant studies than acute models of inflammation/injury of the bladder in healthy animals.

While bladder urothelium normally functions as an efficient barrier to the passage of water and various solutes, urothelial ultrastructure and permeability are altered in FIC cats.4 A number of factors may have a role in these changes, such as the production of bioactive substances such as nitric oxide (NO) by various cell types, including bladder nerves and urothelium.5 The putative physiological functions of NO produced by constitutively expressed Ca2⫹ dependent NO synthase (NOS) isoforms in the lower urinary tract include the relaxation of urethral smooth muscle, modulation of transmitter release from efferent nerves, regulation of urothelial permeability and modulation of afferent nerve activity.6 In contrast, overproduction of NO by the Ca2⫹ independent inducible NO synthase (iNOS) isoform can increase epithelial permeability to hydrophilic macromolecules, which could contribute to the loss of barrier function in many disease states.7 These and other data raise the possibility that the transmitter function of NO is plastic and can be altered by chronic pathological conditions. Because it has been reported that NO levels are altered in patients with IC and also in chemically induced models of cystitis,8, 9 in the current studies we examined whether NO levels are also altered in FIC bladder. We used the functional measurements of NO release as well as immunoblot analysis from bladder muscle and mucosal strips. Our findings, which have been presented in a preliminary report,10 demonstrate increased baseline NO production due to iNOS in FIC bladder.

Submitted for publication August 27, 2004. Study received Institutional Animal Care and Use Committee, University of Pittsburgh School of Medicine approval. Supported by National Institutes of Health Grants DK57284 (LAB), DK64539 (LAB AND CAB), Roche Palo Alto (LAB) and DK64280 (AJK). * Correspondence: Department of Medicine, University of Pittsburgh School of Medicine, A1207 Scaife Hall, Pittsburgh, Pennsylvania 15261 (telephone: 412-383-7368; FAX: 412-648-7197; e-mail: [email protected]). † Financial interest and/or other relationship with Roche Palo Alto. ‡ Financial interest and/or other relationship with Abbott Pharmaceuticals, Roche Palo Alto, Eli Lilly Pharmaceuticals, Pfizer Pharmaceuticals, Dynagen, Novartis Pharmaceuticals, Urogene, Johnson and Johnson, Newron, ICOS and Neurogen. 625

626

NITRIC OXIDE SYNTHASE AND NITRIC OXIDE IN FELINE INTERSTITIAL CYSTITIS METHODS

Animals. All protocols involving use of animals in this study were approved by the Institutional Animal Care and Use Committee, University of Pittsburgh School of Medicine. Nine healthy and 6 FIC adult cats were used for this study. All cats with FIC were obtained as donations from clients due to a history of chronic recurrent strangury, hematuria, pollakiuria and/or urination in inappropriate locations and were evaluated at Ohio State University Veterinary Teaching Hospital. Evaluation consisted of a complete physical examination (including body weight), complete blood count, serum biochemical analysis, urinalysis, urine bacteriological culture and cystoscopy. Cystoscopy was performed using a 9Fr rigid pediatric cystoscope (Karl Storz, Endoscopy America, Culver City, California) in female cats and a 3Fr flexible fiber optic cystoscope (Five Star Medical, San Jose, California) in male cats. The diagnosis of FIC was based on compatible history and the consideration of standard National Institutes of Health inclusion and exclusion criteria after the results of the described laboratory tests were obtained, including the presence of submucosal petechial hemorrhages (glomerulations) at cystoscopy. Healthy, age matched cats obtained from commercial vendors and determined to be free of disease and signs referable to the lower urinary tract according to the same diagnostic criteria as cats with FIC served as controls. All cats were housed in stainless steel cages at Ohio State University animal facilities and allowed to acclimate to their environment for at least 3 months before the study. Tissue preparation. The bladders of deeply anesthetized (␣-chloralose, 60 to 70 mg/kg intravenously) FIC and normal cats were excised and the mucosal layer was stripped from the smooth muscle layer. Subsequently the mucosa and smooth muscle portions were diced and homogenized using a tissue tearer (BioSpec Products, Bartlesville, Oklahoma) in Hanks balanced salt solution composed of 5 mM KCl, 0.3 mM KH2PO4, 138 mM NaCl, 4 mM NaHCO3, 0.3 mM Na2HPO4, 5.6 mM glucose, 10 mM HEPES and complete protease inhibitor cocktail tablets, 1 tablet per 10 ml (Roche, Basel, Switzerland), pH 7.4 (Sigma Chemical Co., St. Louis, Missouri). The homogenate was centrifuged at 13,000 rpm for 15 minutes and protein concentrations of the supernatant were determined using the BCA protein assay (Pierce, Rockford, Illinois). Immunoblotting. After denaturation at 100C for 5 minutes 5 ␮g protein per sample were separated on an 8% to 16% gradient sodium dodecyl sulfate-polyacrylamide gel electrophoresis gel (Gradipore, New York, New York). The proteins were transferred to polyvinylidene fluoride membranes (BioRad Laboratories, Hercules, California) and the membranes were blocked with 5% Milk tris buffered saline-Tween (TBS-T) for 1 hour. After a brief rinse in TBS-T the membranes were incubated overnight at 4C with primary antibody, that is rabbit anti-iNOS, endothelial NOS (eNOS) or neuronal NOS (nNOS) antibodies (BD Transduction Laboratories, San Diego, California) at 1:2,000 in 5% Milk TBS-T. Following extensive washing the membranes were incubated in secondary antibody (goat antirabbit IgG horseradish peroxidase, Santa Cruz Biotechnology, Santa Cruz, California) at 1:10,000 for 1 hour in 5% Milk TBS-T, washed, and then developed with ECL Plus (Amersham International, Arlington Heights, Illinois) and exposed to film. The volume of each band was determined using a Personal Densitometer SI (Molecular Probes, Eugene, Oregon). The membranes were stripped using a membrane recycling kit (Alpha Diagnostic International, San Antonio, Texas). As a loading control, the membranes were reprobed overnight using similar conditions with rabbit anti ␤-actin (1:40,000 in 5% Milk TBS-T, Abcam, Ltd., Cambridge, United Kingdom). NO measurement. Urothelial tissue strips obtained from deeply anesthetized (␣-chloralose) healthy and FIC cats were

maintained in vitro in a temperature regulated (37C) and oxygenated bath. They were perfused (1 ml per minute) with a solution containing 4.8 mmol/l KCl, 120 mmol/l NaCl, 1.2 mmol/l NaH2PO4, 1.1 mmol/l MgSO4, 15.5 mmol/l NaHCO3, 2 mmol/l CaCl2 and 11 mmol/l glucose, pH 7.4. The tip of an NO selective porphyrinic microsensor (NO detection limit 1 nM, response time 1 millisecond, tip diameter 10 to 20 ␮m) prepared according to previously described methods5 was placed directly onto the luminal surface of isolated bladder strips. The microsensors were characterized by differential pulse voltammetry to determine the redox potential of the oxidation of NO to NO⫹. Chronoamperometry performed at a constant potential of 50 mV more positive than redox potential was used to determine the NO concentration. High purity (greater than 99.99%) NO standards were prepared daily to calibrate accurately the electrodes. Currents generated by the oxidation of NO to NO⫹ at the porphyrinic interface were amplified, converted to voltages and digitized for analysis. Controls were evaluated without external stimuli and after the direct application of perfusate to ensure that flow did not cause cellular disruption. In general bladder strips were bubbled in oxygenated solution and used within 1 to 4 hours after tissue removal without any appreciable decrease in the ability to produce NO.5 All compounds were applied locally at a distance 100 ␮m from the cell using a microperfusion system (ALA Instruments, Westbury, New York). Unless specified, all chemicals were reagent grade or better. Data were analyzed using the Student unpaired t test with significance at p ⬍0.05. RESULTS

iNOS expression. Western blot analysis of iNOS protein showed a trend toward increased iNOS levels in bladder mucosa and smooth muscle from FIC bladders as compared with healthy cats (fig. 1, A to C). In addition, analysis of nNOS or eNOS isoforms showed a trend toward decreased expression in bladder mucosa and to a lesser extent in smooth muscle. However, when these results were quantified as a percent of loading control (␤ actin), the changes were not statistically significant (p ⬎0.05, fig. 1, A to C). NO production. Tissue strips were studied within 1 to 4 hours after isolation from bladder smooth muscle or mucosa. In the absence of chemical stimulation baseline NO production was not detectable (sensitivity 1 to 5 nM NO) from smooth muscle strips or mucosa obtained from healthy cats. In addition, the application of perfusate, which provided mild mechanical stimulation of the tissues, did not elicit NO release. As a control to test for tissue strip viability, release was elicited by the application of isoproterenol, a nonselective ␤-adrenoceptor agonist, which in our previous reports also elicited NO release from rat bladder strips (fig. 2).5 In normal bladder tissue peak NO levels (500 to 750 nM) evoked by isoproterenol in mucosal strips from different cats was similar, whereas release could not be elicited in smooth muscle strips from any cats following the application of agonist. Agonist (isoproterenol) evoked NO production was also detected in FIC mucosal strips, although production was less (45%) compared with that in strips from healthy cats. In contrast, continuous baseline NO was recorded from mucosal as well as smooth muscle strips from FIC bladders (figs. 2 and 3). The peak level of NO varied among preparations (baseline NO release 140 to 365 nM), although in each case it was significantly increased compared with that in normal bladder strips, in which we did not detect baseline NO (p ⬍0.05). NO production in FIC bladder strips did not decrease with time and it continued for the duration of the recordings (30 minutes). Involvement of calcium. As noted in our previous studies, ␤-adrenoceptor evoked NO depended on extracellular Ca2⫹ since isoproterenol evoked NO release from bladder mucosal

NITRIC OXIDE SYNTHASE AND NITRIC OXIDE IN FELINE INTERSTITIAL CYSTITIS

627

FIG. 1. Mean NOS, nNOS and eNOS expression ⫾ SEM in FIC and normal mucosa and smooth muscle. A and B, representative Western blots. C and D, graphs show expression expressed as percent of loading control (␤-actin). Black bars indicate FIC mucosa or smooth muscle strips. Open bars indicate normal mucosa or smooth muscle strips. Bars represent data on at least 3 independent experiments.

FIG. 2. A, 10– 6 M isoproterenol (Iso) evoked NO release from isolated mucosal strip. B, NO release evoked by 10-6 M isoproterenol was decreased in FIC mucosal strips (black bars) compared with normal mucosal strips (open bars). In addition, there was constant baseline (Basal) release of NO in FIC mucosal strips (black bar). Baseline NO release was not detected in normal mucosal strips. Values represent mean ⫾ SEM. Asterisk indicates significantly different from control. Bar represents data on least 3 independent experiments and minimum of 3 bladder strips per bladder from 6 to 9 animals.

strips obtained from healthy cats was significantly decreased (mean ⫾ SEM 93% ⫾ 4%) after removal and chelation (100 ␮M egtazic acid) of extracellular Ca2⫹. On the other hand, the baseline production of NO in FIC bladder strips (mucosa and smooth muscle) was not altered in the presence of egtazic acid and zero Ca2⫹ (fig. 3). In addition, incubation with the calmodulin antagonist trifluoroperazine (20 ␮M), which blocks isoproterenol induced NO release,11 did not significantly decrease baseline NO. Baseline NO in a total of 25 FIC bladder tissue strips from 6 cats was significantly decreased by the NOS antagonist L-NAME (N-nitro-L-arginine methyl ester) (100 ␮M). Similar findings (increased NO production) were measured in smooth muscle and mucosal strips. DISCUSSION

This study revealed that FIC bladder mucosa and smooth muscle produce significantly higher levels of NO compared with strips from healthy cat bladders and these augmented levels were significantly decreased by incubation with a NOS antagonist. Alterations in NO production have been reported

in some humans with IC9 and, thus, regulation of NOS activity may have an important role in IC. These and other data demonstrating altered levels of adenosine triphosphate release from FIC urothelium, similar to that reported in patients with IC, suggests there may be similarities between FIC cats and some humans with IC. Although the mechanism as well as the specific type of cell that leads to altered NO release is unclear, our findings demonstrate that increased baseline NO production in FIC is due to changes in iNOS activity. Three isoforms of nitric oxide synthase have been characterized. nNOS and eNOS isoforms are constitutively expressed and Ca2⫹ dependent, while iNOS is inducible by injury or inflammation and Ca2⫹ independent. We have previously reported that eNOS and iNOS isoforms are expressed in bladder urothelium5 and the current findings show a trend toward decreased expression of the 2 isoforms in mucosal strips in FIC samples compared with controls. In the current studies agonist evoked Ca2⫹ dependent NO production was significantly decreased in FIC mucosal strips compared with

628

NITRIC OXIDE SYNTHASE AND NITRIC OXIDE IN FELINE INTERSTITIAL CYSTITIS

FIG. 3. Mean effect ⫾ SEM of antagonists on baseline NO release in FIC bladder mucosa. Baseline (Basal) NO release was not altered by pre-incubation with 20 ␮M of calmodulin antagonist trifluoroperazine (TFP) or after removal of extracellular Ca2⫹ (Zero Ca2⫹) but it was significantly decreased after incubation with 100 ␮M of NOS antagonist L-NAME. Asterisk indicates significantly different from untreated FIC mucosa (baseline release).

production in strips from normal, unaffected cats. Similarly it was reported that NO levels are decreased in some patients with IC.12 Because alterations in NO levels can affect the excitability of sensory nerves in the bladder, it is possible that a decrease in NO levels (decrease in phasic release of NO due to constitutive NOS isoforms) could sensitize afferents, leading to bladder symptoms in patients with IC. Although the current findings show that iNOS is expressed in normal cat smooth muscle and mucosal strips, we did not detect agonist independent NO production in normal cat bladder tissue. The reason for the discrepancy between iNOS expression and the apparent absence of baseline NO release from normal tissue is unclear. The possible reasons are NO levels that are below the detection limits of our microsensor (1 to 10 nM) and/or the presence of factors that inhibit iNOS activity. Baseline NO production has been reported in normal, non inflamed epithelium of the colon, in which it has been speculated that NO produced by iNOS may have a role in host defense mechanisms.13 Thus, additional studies are necessary to evaluate the role of iNOS in normal bladder urothelium. Nevertheless, iNOS expression in the normal cat bladder suggests that Ca2⫹ independent NO production may be rapidly induced by pathological stimuli, such as that which occurs after injury/inflammation. Although we did not detect a significant increase in iNOS protein levels in FIC bladder tissue, the significant increase in baseline NO release recorded in all FIC cat mucosal strips suggests that relatively small changes in iNOS synthesis may result in a substantial increase in NO production. Our results in the FIC bladder are in agreement with those in studies showing a rapid increase in iNOS mediated NO production following inflammation or injury.14 Although this increased NO production may be part of the defense mechanism of the body, excessive NO release for prolonged periods may elicit cellular damage. In the bladder the latter could compromise the urothelial barrier and lead to increased sensitivity/excitability of nearby bladder afferent nerves. In this context studies show that the intravesical administration of NO donors alters the voiding reflex by suppressing bladder nerve excitability.15 There is also evidence that increased iNOS activity, leading to increased NO in a number of inflammatory models, may be a key factor in altering the expression, localization and function of epithelial tight junction proteins.14 In support of this idea we have reported that intravesical administration of a high concentration (0.5 mM) of an NO donor (eg sodium nitroprusside) administered to the bladder in vivo or appli-

cation to urothelial cultures that typically exhibit high transepithelial resistance leads to a significant decrease in transepithelial resistance and increase in urothelial permeability.16 Increased NO levels due to iNOS are also thought to be involved in urothelial cell shedding by interfering with urothelial cell differentiation or by promoting the sloughing of damaged cells.17 Whether alterations in NO levels have a role in the changes in mucosal permeability and in urothelial ultrastructure seen in FIC bladders (disruption of tight junctions and detachment of umbrella cells from each other)4 remains to be further explored. It has also been shown that iNOS expression can be altered in bladder smooth muscle cells following exposure to noxious stimuli such as lipopolysaccharide or cyclophosphamide.18 In addition, following bladder outlet obstruction alterations in NO due to iNOS were found to have a role in the growth of smooth muscle cells, thus, influencing the progress of bladder hypertrophy.19 Additional evidence has raised the possibility that augmented iNOS in the bladder wall may be involved in the smooth muscle fibrotic response to chronic urinary tract infections.20 Whether up-regulation of iNOS in FIC produces similar alterations in bladder smooth muscle remains to be determined. CONCLUSIONS

Our findings demonstrate that FIC in cats results in increased baseline production of NO due to iNOS. These changes in transmitter release may have a role in the alterations in barrier properties that occur in FIC. Understanding the mechanism contributing to and maintaining this type of barrier dysfunction may provide important insight into targets for the clinical management of IC. All chemicals were obtained from Sigma Chemical Co. REFERENCES

1. Nickel, J. C.: Editorial: interstitial cystitis—an elusive target? J Urol, 170: 816, 2003 2. Bouchelouche, K. and Nordling, J.: Recent developments in the management of interstitial cystitis. Curr Opin Urol, 13: 309, 2003 3. Buffington, C. A., Chew, D. J. and Woodworth, B. E.: Feline interstitial cystitis. J Am Vet Med Assoc, 215: 682, 1999 4. Lavelle, J. P., Meyers, S. A., Ruiz, W. G., Buffington, C. A., Zeidel, M. L. and Apodaca, G.: Urothelial pathophysiological changes in feline interstitial cystitis: a human model. Am J Physiol, 278: F540, 2000 5. Birder, L. A., Nealen, M. L., Kiss, S., de Groat, W. C., Caterina, M. J., Wang, E. et al: Beta-adrenergic agonists stimulate endothelial nitric oxide synthase in rat urinary bladder urothelial cells. J Neurosci, 22: 8063, 2002 6. Mumtaz, F. H., Khan, M. A., Thompson, C. S., Morgan, K. J. and Mikhailidis, D. B.: Nitric oxide in the lower urinary tract: physiological and pathological implications. BJU Int, 85: 567, 2000 7. Xu, D. Z., Lu, Q. and Deitch, E. A.: Nitric oxide directly impairs intestinal barrier function. Shock, 17: 139, 2002 8. Erickson, D. R.: Urine markers of interstitial cystitis. Urology, 57: 15, 2001 9. Logadottir, Y., Here´n, I., Fall, M., Wiklund, N. P. and Peeker, R.: Intravesical nitric oxide production discriminates between classic and nonulcer interstitial cystitis. J Urol, 171: 1148, 2004 10. Buffington, C. A. T., Kiss, S., Kanai, A. J., Dineley, K., Reynolds, I., de Groat, W.C. et al: Alterations in bladder afferent neurons and urothelium in feline interstitial cystitis. Soc Neurosci Abstr, 26: 930, 2000 11. Bredt, D. S. and Snyder, S. H.: Isolation of nitric oxide synthase, a calmodulin-requiring enzyme. Proc Natl Acad Sci USA, 87: 682, 1990 12. Smith, S. D., Wheeler, M. A., Foster, H. E., Jr. and Weiss, R. M.: Improvement in interstitial cystitis symptoms scores during treatment with oral L-arginine. J Urol, 158: 703, 1997 13. Roberts, P. J., Riley, G. P., Morgan, K., Miller, R., Hunter, J. O.

NITRIC OXIDE SYNTHASE AND NITRIC OXIDE IN FELINE INTERSTITIAL CYSTITIS

14. 15. 16.

17.

and Middleton, S. J.: The physiological expression of inducible nitric oxide synthase (iNOS) in the human colon. J Clin Pathol, 54: 293, 2001 Han, X., Fink, M. P., Yang, R. and Delude, R. L.: Increased iNOS activity is essential for intestinal epithelial tight junction dysfunction in endotoxemic mice. Shock, 21: 261, 2004 Ozawa, H., Chancellor, M. B., Jung, S.-Y., Yokoyama, T., Fraser, M. O., Yu, Y. et al: Effect of intravesical nitric oxide therapy on cyclophosphamide-induced cystitis. J Urol, 162: 2211, 1999 Truschel, S., Ruiz, W. G., Kanai, A. J., Kiss, S., Meyers, S., Apodaca, G. et al.: Involvement of nitric oxide (NO) in bladder afferent and urothelial abnormalities following chronic spinal cord injury. Soc Neurosci Abstr, 27: abstract 842.9, 2001 Romih, R., Korosec, P., Jezernik, K., Sedmak, B., Trisinar, B., Deng, F. M. et al: Inverse expression of uroplakins and induc-

629

ible nitric oxide synthase in the urothelium of patients with bladder outlet obstruction. BJU Int, 91: 507, 2003 18. Xu, X., Cubeddu, L. X. and Malave, A.: Expression of inducible nitric oxide synthase in primary culture of rat bladder smooth muscle cells by plasma from cyclophosphamide-treated rats. Eur J Pharmacol, 416: 1, 2001 19. Johansson, R., Pandita, R. K., Poljakovic, M., Garcia-Pascual, A., de Vente, J. and Persson, K.: Activity and expression of nitric oxide synthase in the hypertrophied rat bladder and the effect of nitric oxide on bladder smooth muscle growth. J Urol, 168: 2689, 2002 20. Austin, P. F., Casale, A. J., Cain, M. P., Rink, R. C. and Weintraub, S. J.: Lipopolysaccharide and inflammatory cytokines cause an inducible nitric oxide synthase dependent bladder smooth muscle fibrotic response. J Urol, 170: 645, 2003