L-NAME, a nitric oxide synthase inhibitor, diminishes oxidative damage in urinary bladder partial outlet obstruction
http://www.100md.com
《美国生理学杂志》
1Albany Medical College,2Albany College of Pharmacy
3Stratton Veterans Affairs Medical Center Albany, New York
ABSTRACT
Partial bladder outlet obstruction (PBOO) results in cellular damage due to ischemia and reperfusion injury. Our study seeks to establish how early this damage can occur and the role that nitric oxide may play in its pathophysiology. Surgical PBOO (1, 3, and 7 days) were performed on male New Zealand White rabbits. Half of the animals were premedicated for 3 days with NG-nitro-L-arginine methyl ester(L-NAME), an inhibitor of nitric oxide synthase before obstruction. Bladder weight increased with duration of PBOO but was significantly lower at 3 and 7 days in animals treated with L-NAME compared with their untreated counterparts. Contractile function decreased progressively with PBOO duration. At 1 day postobstruction, bladder contractility was significantly lower in the L-NAME rabbits than in the untreated rabbits. At 3 and 7 days, contractility of the L-NAME bladders was equal or higher than the untreated bladders. The level of hypoxia at 1 day after obstruction was significantly higher in the L-NAME-treated animals than in the untreated controls but equal at 3 and 7 days obstruction. Increased nitrotyrosine was seen by Western blot in all obstructed animals. However, the amount was significantly less in the L-NAME-treated animals at 3 and especially at 7 days. Nerve density decreased progressively after obstruction; however, it decreased to a significantly lesser degree in the L-NAME-treated bladders than in the untreated groups. These results suggest that L-NAME pretreatment enhanced ischemic damage at 1 day after obstruction but protected the bladder from nitric oxide-generated free radical damage at the later time periods by inhibiting the generation of nitrotyrosine.
ischemia-reperfusion; rabbit
PARTIAL OUTLET OBSTRUCTION of the urinary bladder is one of the major pathological effects of benign prostatic hyperplasia in men. It has been demonstrated that ischemia followed by reperfusion is an etiology for the progression of bladder dysfunction associated with partial bladder outlet obstruction (PBOO) in a variety of animal models (2, 8, 18). Cyclic episodes of ischemia-reperfusion (I/R) lead to both direct ischemic damage, and the generation of free radicals, both reactive nitrogen species (RNS) and reactive oxygen species (ROS), which result in membrane and subcellular organelle damage.
The initial structural response to partial outlet obstruction in the rabbit is a rapid remodeling of the bladder wall that includes urothelial and serosal hyperplasia and smooth muscle hypertrophy. It has been demonstrated that significant changes in blood flow occur within bladder tissue very early in this process. Lieb et al. (14) demonstrated that, as early as 1 day after obstruction, there is a significant increase in blood flow to the bladder, which is caused by nitric oxide (NO)-stimulated vascular relaxation. Schroeder et al. (18) demonstrated that a decrease in blood flow resulting from I/R is correlated with decreased contractile function. In this study, we hypothesize that ischemia and I/R are induced very early postobstruction and that both processes may be related to NO and the generation of RNS and subsequent free radical damage.
NO can be produced from exogenous sources, such as nitrovasodilators, or endogenously from L-arginine by one of several isoforms of NO synthase (NOS). NO has been implicated as a nonadrenergic, noncholinergic inhibitory neurotransmitter that mediates relaxation activity in the lower urinary tract (1, 4). It is also known to be a powerful and ubiquitous regulator of vascular tone and can lead to reperfusion-based injury. NO-related tissue injury is likely to be partially due to peroxynitrite, a relatively long-lived, strong oxidant that is generated by the near-diffusion-limited reaction between NO and superoxide (11). A major product from the reaction of peroxynitrite with proteins is nitrotyrosine (12). With the use of antibodies that specifically recognize nitrotyrosine, extensive protein nitration has been detected in human coronary artery atherosclerotic lesions (3) and in acute lung injury (10, 13). Nitrotyrosine has been demonstrated to be a marker of free radical damage due to RNS and will be used in our study as a marker of reperfusion-based injury.
To investigate further the role of I/R injury in bladder function in patients with benign prostatic hyperplasia (BPH), we employed an animal model of New Zealand White male adult rabbits with surgical PBOO. The location and severity of this damage as it evolves over time will be characterized in the immediate postobstructive period using several modalities. It is likely that NO performs a key role in the pathophysiology of this process, and it is hypothesized here that substances that decrease reperfusion following ischemia will mitigate damage in this setting. NG-nitro-L-arginine methyl ester (L-NAME), which acts as a competitive inhibitor of NOS due to its structural similarity to arginine, the substrate of NOS, may act as such an agent.
MATERIALS AND METHODS
Surgical procedure. Young New Zealand White male rabbits (3–5 kg; 15–20 wk old) were obtained from Milbrook Breeding Labs (Amherst, MA). Animals were divided into two groups. One group was premedicated with L-NAME by implanting an osmotic pump at an infusion rate of 6 mg·kg–1·day–1 in the subcutaneous tissue behind the rabbit's neck and allowing 1 wk for equilibrium to reach adequate serum levels of L-NAME; the infusion was continued during the entire experimental period.
The other group of animals received no premedication. For both groups, surgical bladder outlet obstruction was performed under general anesthesia with ketamine/xylazine. Animals were catheterized with an 8-Fr Coude tip catheter before the procedure. The bladder was identified, and a 3-0 silk suture was tied around the catheterized bladder neck, with care taken to avoid damaging the vesical vessels. The tension of the suture was standardized, with all obstructions performed by the same surgical team. After the procedure, the catheter was removed. At 1, 3, and 7 days postoperatively, four animals from each group were euthanized with pentobarbital sodium. Two hours before euthanasia, 2 ml hypoxyprobe-1 (20 mg/kg body wt) were administered subcutaneously to the rabbits. The bladder was harvested from each animal and placed in Tyrode solution (a physiological medium of dextrose, calcium chloride, sodium bicarbonate, potassium chloride, magnesium chloride, and sodium phosphate). The perivascular fat and connective tissues were removed, and the bladders were weighed. Three full-thickness bladder strips were obtained from the supratrigonal dorsal aspect of each rabbit bladder for contractile studies. Two full-thickness strips were isolated and fixed in formalin for histology and immunohistochemistry. The balance of the bladder body was separated into muscle and mucosa, flash-frozen in liquid nitrogen, and used for protein extraction and biochemical assays. Nonobstructed control animals, both with (n = 4) and without (n = 4) exposure to L-NAME, were euthanized and had tissue isolated for study in an analogous manner.
Physiological studies. Bladder strips were mounted in a 30-ml physiological medium of Tyrode solution with dextrose (1 mg/ml) and gassed with 95% O2-5% CO2 at 37°C. Two grams of tension were applied to each strip, and the strip was equilibrated for 15 min. The maximal response to electrical field stimulation (2, 8, and 32 Hz, 80 V, 1-ms duration, 15-s train) was determined. Strips were then exposed successively and independently to carbachol (20 μM) and potassium chloride (120 mM), and the contractile response was determined. Peak tension was recorded for each of the different stimuli. The concentrations in these experiments were based on the results of dose-response studies showing that they produced maximal responses in normal and obstructed rabbit bladder strips. Between additions of pharmacological agents, each strip was washed three times with fresh buffer at 10-min intervals.
Protein isolation and SDS-PAGE and Western blotting. Frozen bladder tissues (100 mg) were ground to a fine powder in a mortar cooled in liquid nitrogen and homogenized in buffer containing 20% glycerol, 50 mM Tris·HCl (pH 6.8), 0.5% (vol/vol) Tween 20, and protease inhibitors (0.5 mM PMSF, 2 μM pepstatin, 2 μM antipain, and 0.1 mg/ml trypsin inhibitor). After addition of SDS (final concentration, 1%), the sample was boiled for 4 min and centrifuged at 10,000 rpm for 15 min. Protein concentration in the supernatant was measured using the Pierce BCA protein assay kit. Equal amounts (20 μg) of total protein from control, obstructed, and L-NAME-treated rabbit bladders were loaded on 12% SDS-PAGE gels and transferred to Immobilon-P membranes with Towbin buffer [25 mM Tris, 192 mM glycine and 20% (vol/vol) methanol]. Membranes were blocked with 5% nonfat milk in 0.05% Tween 20 in PBS for 1 h at room temperature and then incubated with primary antibody, monoclonal antibody to nitrotyrosine (Alexis Biochemicals, San Diego, CA). After treatment with the primary antibody, the membranes were washed with 20 mM Tris, 500 mM NaCl, and 0.05% Tween 20 and incubated with secondary antibody (goat anti-mouse IgG at 1:10,000). Substrates were visualized by using echochemiluminescence (Amersham Pharmacia Biotech, Buckinghamshire, UK) and by exposing the membranes to autoradiographic films (Kodak X-OMAT). Band intensities were scanned and analyzed with a Kodak Image Station 440CF and a Kodak ID image analysis software (Scientific Image System, Rochester, NY). Standard curves were constructed to establish that the protein concentrations used for analysis fell within the linear portion of the curves.
Histologic studies. Full-thickness specimens were cut from the bladder, placed in OCT compound (TBS, Durham, NC), snap-frozen in liquid nitrogen, and stored at –70°C. Transverse sections 5-μm thick were then cut on a Microm HM505E cryostat at –25°C and mounted on Bond Rite-treated slides (Richard Allan Scientific, Kalamazoo, MI).
The frozen sections were air dried for 1 h, fixed in ice-cold acetone for 10 min, and then air dried again. Before sections were stained, they were soaked in two changes of PBS for 15 min each to remove the OCT. Another section of full-thickness bladder tissue from each specimen was fixed in 10% neutral buffered formalin for 4–8 h, processed routinely, and embedded in a paraffin block. Before embedding, the square section of tissue was sliced into two or three strips, and each was embedded on edge to provide a cross-sectional view of bladder after microtome sectioning. Five-micrometer-thick sections were cut from each block and mounted on positively charged slides, deparaffinized in xylene, and used for the various immunostaining studies.
Immunostaining. Tissue sections were incubated with the mouse monoclonal antibody to nitrotyrosine and another set with hypoxyprobe primary antibody on the Ventana ES automated immunostainer (Ventana Medical Systems, Tucson, AZ) using an avidin-biotin and diaminobenzidene chromagin detection followed by a light hematoxylin counterstain. The detection kit from Ventana was used with the dilutions, with times and temperatures set by the supplier except for the following: initial blocking done with serum-free protein block (DAKO, Carpinteria, CA) for 12 min. Nitrotyrosine primary antibody was applied for 1 h at room temperature at a dilution of 1:50 and hypoxyprobe primary antibody at a dilution of 1:100 in antibody diluent (DAKO). The secondary antibody in the kit was substituted with biotinylated goat anti-mouse predilute (Lab Vision, Fremont, CA). After the immunostaining procedure, all slides were dehydrated through graded alcohols and then mounted with Permount (Fisher Scientific, Pittsburgh, PA) mounting medium.
Neurofilament staining. Nerve damage is investigated through immunohistochemistry evaluation of the presence and density of neurofilaments within the axons of bladder nerves. Tissue sections were stained with the mouse monoclonal antibody anti-neurofilament clone 2f11 (Neomarkers, Fremont, CA) on the Ventana ES automated immunostainer (Ventana Medical Systems) using an avidin-biotin and diaminobenzidene chromagin detection followed by a light hematoxylin counterstain. Neurofilament primary antibody (100 μl) was applied for 32 min at a dilution of 1:20 in antibody diluent (DAKO). The secondary antibody in the kit was substituted with biotinylated goat anti-mouse predilute (Lab Vision). After the immunostaining, all slides were dehydrated through graded alcohols and two changes of xylene and then mounted with Permount (Fisher Scientific) mounting medium.
Nerve calculations. The nerve data including size and quantity were acquired from the stained cross sections using Image Pro Plus (Media Cybernetics, Silver Spring, MD) image analysis software. Images for analysis were acquired using a Spot charge-coupled device color digital camera linked to an Olympus BX-60 microscope. The transverse section of each specimen was captured in four frames at x100 magnification. The nerves in each image were highlighted using the Image Pro Plus software by manually selecting the pixel value within the stained nerves using the color cube-based tool in the count/size application of the program. The area measurement was then selected for all the nerves highlighted, and the data were saved to Excel and later compiled and compared for all the groups.
Statistical analysis. A Student's t-test was used to calculate difference between means in the contractile function and bladder weights obtained in this experiment. Statistical significance was judged by a P value of <0.05.
RESULTS
There was no significant difference in bladder weight between those control animals premedicated with L-NAME and those that were not. There was a progressive increase in bladder weight among the obstructed animals. However, in those animals premedicated with L-NAME, there was a significantly smaller increase in bladder weight at both 3 and 7 days of PBOO (Fig. 1).
In general, there was a progressive decrease in contractile response to all forms of stimulation with duration of obstruction. With regard to the contractile responses to field stimulation, for all frequencies of stimulation, after 1 day of obstruction, those animals treated with L-NAME had significantly poorer contractile function compared with those with PBOO alone. This trend reversed at 3 and 7 days of PBOO, with the L-NAME group contracting better than the PBOO alone group, which showed significant difference at 32 Hz (Fig. 2). Similarly, for both carbachol and KCl (Fig. 3) after 1 day of obstruction, those animals treated with L-NAME had significantly poorer contractile function compared with those with PBOO alone. This trend reversed at 3 and 7 days of PBOO, with the L-NAME group contracting better than the PBOO alone group.
There was a progressive and significant increase in the expression of nitrotyrosine over the 7-day period of obstruction. In contrast, at the 3- and 7-day time periods, the level of nitrotyrosine in the L-NAME-treated-obstructed rabbits decreased to control levels, providing strong evidence that NO-generated RNS is involved in the response to obstruction (Fig. 4).
Immunohistochemistry also demonstrated increased expression of nitrotyrosine in obstructed rabbits. Again, there was significantly less staining of nitrotyrosine in L-NAME-treated groups (Fig. 5). The presence of nitrotyrosine in the PBOO rabbit bladders indicates that oxidants derived from NO such as peroxynitrite are generated during bladder outlet obstruction and may be involved in its pathogenesis. Using hypoxyprobe-1, we found no evidence of hypoxia in the control groups. After PBOO, at day 1 there was minimal staining only in the urothelium. After 3 days, there was moderate staining in the urothelium as well as light staining in the vascular endothelium. At day 7, there was intense staining in the urothelium and vascular endothelium, as well as small amounts of hypoxia in the muscularis (Fig. 6, A–D). When examining the samples from animals treated with L-NAME, we found differences at day 1 of obstruction; in addition, there was significantly greater staining of the urothelium compared with those with PBOO alone and significant vascular endothelial uptake. No differences were seen at days 3 and 7 with L-NAME vs. PBOO alone (Fig. 6, E and F).
In the neurofilament studies, tissue from control animals demonstrated staining of nerve fibers of all sizes throughout the submucosa and muscularis. There were no changes seen after 1 day of obstruction. At 3 days, more than one-half of small nerve tracts were absent, but large nerve bundles were still present in similar number as controls. By day 7, there was no small nerve tract staining, and there was a small decrement in the number of large nerve tracts (Fig. 7). At each time point, the nerve density of the L-NAME-treated groups was significantly higher than their nontreated counterparts (Fig. 8). Thus treatment with L-NAME provided significant protection of the nerves, which are the most sensitive subcellular structures to ischemic and reperfusion damage.
DISCUSSION
In this study, we investigated the effects of L-NAME, a NOS inhibitor, on the response to PBOO-mediated injury in rabbits at 1, 3, and 7 days after obstruction. We found a striking time course of effect, with L-NAME reducing bladder function 1 day after obstruction but enhancing bladder function at 3 and 7 days.
We found that L-NAME negatively affected bladder contractile properties and histological appearance 1 day after surgical obstruction. This has never been studied previously. It is well recognized that blood flow generally speaking is reduced after partial bladder obstruction. However, in the first day after obstruction, there is an NO-stimulated increase in blood flow that is thought to mediate the maintenance of function and the bladder hypertrophy that begins at this time. This is confirmed by histological studies that demonstrate no significant hypoxia or neural injury. Our study uses L-NAME to block NO. Our findings of decreased function and increased ischemia in L-NAME-treated rabbits confirm the previous findings that NO serves to help maintain function during the first day after obstruction. An alternative cause of bladder dysfunction 1 day postobstruction in L-NAME-treated animals may be inducible NOS expression. By day 1 of PBOO, inducible NOS is not likely to be expressed, and L-NAME will inhibit constitutive endothelial NOS and neuronal NOS. This would decrease blood flow and exacerbate ischemia, decreasing contractility as observed. The implications of our study are significant in that they give further confirmation that blood flow is a key mediator of function after PBOO.
We also found that L-NAME blockade of NO helped to prevent deterioration of bladder function 3 and 7 days after obstruction. This paradoxical finding has never been reported previously. However, our findings are entirely consistent in this area. Not only field stimulation (which reflects nerve function) but also carbachol and KCl (which reflect muscarinic receptor stimulation and smooth muscle contraction) responses were all similar. In addition, our histological studies of nerve staining supports nerves being more intact when L-NAME is used during obstruction. However, we believe that the explanation is found by studying the response to I/R.
In the contractile studies, L-NAME significantly enhanced the contractile dysfunctions to all forms of stimulation at 1 day but showed improved contractile responses compared with untreated obstructed bladders at 3 and 7 days. Previous studies (14) demonstrated that PBOO resulted in a significant NO-stimulated increase in blood flow at 1 day, which returned to control levels at 3 and 7 days postobstruction. This increase in blood flow would be expected to compensate for the ischemia mediated by the obstruction. In the L-NAME-treated rabbits, this increase in blood flow would have been reduced or eliminated and can account for both the increased hypoxia observed at the 1-day period of obstruction and the enhanced contractile dysfunctions observed at this time. However, L-NAME would also be expected to reduce or eliminate the RNS generated by the increased NO, with the resulting reduction in RNS-stimulated generation of nitrotyrosine and reduction in membrane damage. This can account for the improved contractile function at 3 and 7 days and also the higher nerve densities at these time periods.
PBOO causes an increase in bladder mass primarily because of smooth muscle hypertrophy and hyperplasia of the urothelium and interstitial fibroblasts (6, 16). The increased wall thickness and wall tension result in cyclical I/R during and subsequent to each voiding contraction (8, 9). Cyclical I/R leads to an increase in free Ca2+ that activates Ca2+-dependent hydrolytic enzymes such as calpain, phospholipase A2, and endonucleases (20, 21). The ROS and RNS generated subsequently result in characteristic damage to proteins, phospholipids, and DNA that eventually disrupt cellular function (7, 17).
We have shown that these I/R-initiated events begin in transient hypoxic foci within the bladder wall during the bladder's initial response to obstruction and that RNS generation would also result from this cyclical I/R. There may be several stimuli for NOS expression in response to bladder obstruction. Mechanical stretch is known to induce NOS expression (15). Hence, the increased pressure toward the bladder that accompanies bladder outlet obstruction may cause production of NOS in detrusor smooth muscle cells. Oxidatively modified proteins have been used as early markers of oxidative stress in a variety of physiological and pathophysiological processes. Damage to proteins by ROS and RNS affects the function of enzymes, receptors, and signal transduction mechanisms (5). Nitrotyrosine, generated on reaction of the RNS (ONOO–) with protein tyrosine residues, was associated with neuronal ischemic injury and numerous neurodegenerative diseases (19). Antibodies specific to nitrotyrosine can measure this biomarker of oxidative stress. We have found an overexpression of nitrotyrosine in obstructed rabbit bladders compared with control bladders.
In conclusion, there is clear evidence of early cellular damage after PBOO. Hypoxia, which may be extrapolated to ischemia, occurs in bladder detrusor smooth muscle tissues, causing severe damage to nerves and the related contractile response to field stimulation. It has been shown that NO exerts a positive influence at 1 day of obstruction by increasing the blood flow to the bladder tissue but subsequently causes reperfusion-based injury due to free radical damage. Pretreatment with L-NAME significantly enhanced the ischemic damage at 1 day of obstruction but significantly inhibited the generation of nitrotyrosine and resulted in a remarkable preservation of nerve density, which indicates that L-NAME reduces free radical damage associated with PBOO. Further studies will be directed toward proteomic methods, which can identify specific proteins nitrated in outlet obstruction and the complex effect of NO on biological systems.
GRANTS
This work was supported in part by Yamaneuchi USA Research Foundation, the Office of Research and Development Medical Research Service, Department of Veteran's Affairs, and by National Institute of Diabetes and Digestive and Kidney Diseases Grant RO-1 DK-067114.
FOOTNOTES
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
REFERENCES
Andersson KE and Persson K. The L-arginine/nitric oxide pathway and nonandergenic, non-cholinergic relaxation of the lower urinary tract. Gen Pharmacol 24: 833–899, 1993.
Azadzoi KM, Tarcan T, Kozlowski R, Krane RJ, and Siroky MB. Overactivity and structural changes in the chronically ischemic bladder. J Urol 162: 1768–1778, 1999.
Beckman JS, Ye YZ, Anderson P, Chen J, Accavitti MA, Tarpey MM, and White CR. Extensive nitration of protein tyrosines in human atherosclerosis detected by immunohistochemistry. Biol Chem Hoppe-Seyler 375: 81–88, 1994.
Bennett BC, Kruse MN, Roppolo JR, Flood HD, Fraser M, and de Groat. Neural control of urethral outlet activity in vivo: role of nitric oxide. J Urol 153: 2004–2009, 1995.
Davis KL, Martin E, Turko IV, and Murad F. Novel effects of nitric oxide. Annu Rev Pharmacol Toxicol 41: 203–236, 2001.
Gabella G and Uvelius B. Urinary bladder of rat: fine structure of normal and hypertrophic musculature. Cell Tissue Res 262: 67–79, 1990.
Gosling G, Kung LS, Dixon JS, Horan P, Whitbeck C, and Levin RM. Correlation between the structure and function of the urinary bladder following partial outlet obstruction. J Urol 163: 1349–1356, 2000.
Greenland JE, Hvistendahl JJ, Anderson H, Jorgensen TM, McMurray G, Cortina-Borja M, Brading AF, and Frokiaer J. The effect of bladder outlet obstruction on tissue oxygen tension and blood flow in the pig bladder. Br J Urol 85: 1109–1114, 2000.
Greenland JE and Brading AF. The effect of bladder outflow obstruction on detrusor blood flow changes during the voiding cycle in conscious pigs. J Urol 165: 245–248, 2001.
Haddad IY, Pataki G, Hu P, Beckman JS, and Matalon S. Quantification of nitrotyrosine levels in lung sections of patients and animals with acute lung injury. J Clin Invest 94: 2407–2413, 1994.
Huie RE and Padmaja S. The reaction of NO with superoxide. Free Radic Res Commun 18: 195–199, 1993.
Ischiropoulos HL, Zhu J, Chen M, Tsai JC, Martin CD, Smith, and Beckman JS. Peroxynitrite mediated tyrosine nitration catalyzed by superoxide dismutase. Arch Biochem Biophys 298: 431–437, 1992.
Kooy NW, Royall JA, Ye YZ, Kelly DR, and Beckman JS. Evidence for in vivo peroxynitrite production in human acute lung injury. Am J Respir Crit Care Med 151: 1250–1254, 1995.
Lieb JI, Chichester P, Kogan B, Das AK, Leggett RE, Schroder A, and Levin RM. Rabbit urinary bladder blood flow changes during the initial stage of partial outlet obstruction. J Urol 164: 1390–1397, 2000.
Miyajima A, Chen J, Kirman I, Poppas DP, Vaughan ED, and Felsen D. Interaction of nitric oxide and transforming growth factor-1 induced by angiotensin 11 and mechanical stretch in rat renal tubular epithelial cells. J Urol 164: 1729–1734, 2000.
Monson FC, Wein AJ, Eika B, Murphy M, and Levin RM. Stimulation of the proliferation of rabbit bladder urothelium by partial outlet obstruction and acute overdistension. Neurourol Urodyn 13: 51–62, 1994.
Nordling J, Artibani W, Hald TI, Keuppens F, Levin R, and Thuroff J. Pathophysiology of the urinary bladder in obstruction and ageing. In: Benign Prostatic Hyperplasia, edited by Chatelain C, Denis L, Foo KT, Khoury S, and McConnell J. Plymouth, UK: Plymbridge Distributors, 2001.
Schroder A, Chicester P, Kogan B, Longhurst P, Lieb JI, Das AK, and Levin RM. Effect of chronic bladder outlet obstruction on blood flow of the rabbit bladder. J Urol 165: 640–646, 2001.
Smith MA, Richey-Harris PL, Satre LM, Beckman JS, and Perry G. Widespread peroxynitrite damage in Alzheimer's disease. J Neurosci 17: 2653–2657, 1997.
Tominaga T, Kure S, Narisawa K, and Yoshimoto T. Endonuclease activation following focal ischemic injury in the rat brain. Brain Res 608: 21–26, 1992.
Udea N, Walker PD, Hsu SM, and Shah SV. Activation of a 15-kDA endonuclease in hypoxia/reoxygenation injury without morphologic features of apoptosis. Proc Natl Acad Sci USA 92: 7202–7206, 1995.(William Conners, Catherine Whitebeck, Pa)
3Stratton Veterans Affairs Medical Center Albany, New York
ABSTRACT
Partial bladder outlet obstruction (PBOO) results in cellular damage due to ischemia and reperfusion injury. Our study seeks to establish how early this damage can occur and the role that nitric oxide may play in its pathophysiology. Surgical PBOO (1, 3, and 7 days) were performed on male New Zealand White rabbits. Half of the animals were premedicated for 3 days with NG-nitro-L-arginine methyl ester(L-NAME), an inhibitor of nitric oxide synthase before obstruction. Bladder weight increased with duration of PBOO but was significantly lower at 3 and 7 days in animals treated with L-NAME compared with their untreated counterparts. Contractile function decreased progressively with PBOO duration. At 1 day postobstruction, bladder contractility was significantly lower in the L-NAME rabbits than in the untreated rabbits. At 3 and 7 days, contractility of the L-NAME bladders was equal or higher than the untreated bladders. The level of hypoxia at 1 day after obstruction was significantly higher in the L-NAME-treated animals than in the untreated controls but equal at 3 and 7 days obstruction. Increased nitrotyrosine was seen by Western blot in all obstructed animals. However, the amount was significantly less in the L-NAME-treated animals at 3 and especially at 7 days. Nerve density decreased progressively after obstruction; however, it decreased to a significantly lesser degree in the L-NAME-treated bladders than in the untreated groups. These results suggest that L-NAME pretreatment enhanced ischemic damage at 1 day after obstruction but protected the bladder from nitric oxide-generated free radical damage at the later time periods by inhibiting the generation of nitrotyrosine.
ischemia-reperfusion; rabbit
PARTIAL OUTLET OBSTRUCTION of the urinary bladder is one of the major pathological effects of benign prostatic hyperplasia in men. It has been demonstrated that ischemia followed by reperfusion is an etiology for the progression of bladder dysfunction associated with partial bladder outlet obstruction (PBOO) in a variety of animal models (2, 8, 18). Cyclic episodes of ischemia-reperfusion (I/R) lead to both direct ischemic damage, and the generation of free radicals, both reactive nitrogen species (RNS) and reactive oxygen species (ROS), which result in membrane and subcellular organelle damage.
The initial structural response to partial outlet obstruction in the rabbit is a rapid remodeling of the bladder wall that includes urothelial and serosal hyperplasia and smooth muscle hypertrophy. It has been demonstrated that significant changes in blood flow occur within bladder tissue very early in this process. Lieb et al. (14) demonstrated that, as early as 1 day after obstruction, there is a significant increase in blood flow to the bladder, which is caused by nitric oxide (NO)-stimulated vascular relaxation. Schroeder et al. (18) demonstrated that a decrease in blood flow resulting from I/R is correlated with decreased contractile function. In this study, we hypothesize that ischemia and I/R are induced very early postobstruction and that both processes may be related to NO and the generation of RNS and subsequent free radical damage.
NO can be produced from exogenous sources, such as nitrovasodilators, or endogenously from L-arginine by one of several isoforms of NO synthase (NOS). NO has been implicated as a nonadrenergic, noncholinergic inhibitory neurotransmitter that mediates relaxation activity in the lower urinary tract (1, 4). It is also known to be a powerful and ubiquitous regulator of vascular tone and can lead to reperfusion-based injury. NO-related tissue injury is likely to be partially due to peroxynitrite, a relatively long-lived, strong oxidant that is generated by the near-diffusion-limited reaction between NO and superoxide (11). A major product from the reaction of peroxynitrite with proteins is nitrotyrosine (12). With the use of antibodies that specifically recognize nitrotyrosine, extensive protein nitration has been detected in human coronary artery atherosclerotic lesions (3) and in acute lung injury (10, 13). Nitrotyrosine has been demonstrated to be a marker of free radical damage due to RNS and will be used in our study as a marker of reperfusion-based injury.
To investigate further the role of I/R injury in bladder function in patients with benign prostatic hyperplasia (BPH), we employed an animal model of New Zealand White male adult rabbits with surgical PBOO. The location and severity of this damage as it evolves over time will be characterized in the immediate postobstructive period using several modalities. It is likely that NO performs a key role in the pathophysiology of this process, and it is hypothesized here that substances that decrease reperfusion following ischemia will mitigate damage in this setting. NG-nitro-L-arginine methyl ester (L-NAME), which acts as a competitive inhibitor of NOS due to its structural similarity to arginine, the substrate of NOS, may act as such an agent.
MATERIALS AND METHODS
Surgical procedure. Young New Zealand White male rabbits (3–5 kg; 15–20 wk old) were obtained from Milbrook Breeding Labs (Amherst, MA). Animals were divided into two groups. One group was premedicated with L-NAME by implanting an osmotic pump at an infusion rate of 6 mg·kg–1·day–1 in the subcutaneous tissue behind the rabbit's neck and allowing 1 wk for equilibrium to reach adequate serum levels of L-NAME; the infusion was continued during the entire experimental period.
The other group of animals received no premedication. For both groups, surgical bladder outlet obstruction was performed under general anesthesia with ketamine/xylazine. Animals were catheterized with an 8-Fr Coude tip catheter before the procedure. The bladder was identified, and a 3-0 silk suture was tied around the catheterized bladder neck, with care taken to avoid damaging the vesical vessels. The tension of the suture was standardized, with all obstructions performed by the same surgical team. After the procedure, the catheter was removed. At 1, 3, and 7 days postoperatively, four animals from each group were euthanized with pentobarbital sodium. Two hours before euthanasia, 2 ml hypoxyprobe-1 (20 mg/kg body wt) were administered subcutaneously to the rabbits. The bladder was harvested from each animal and placed in Tyrode solution (a physiological medium of dextrose, calcium chloride, sodium bicarbonate, potassium chloride, magnesium chloride, and sodium phosphate). The perivascular fat and connective tissues were removed, and the bladders were weighed. Three full-thickness bladder strips were obtained from the supratrigonal dorsal aspect of each rabbit bladder for contractile studies. Two full-thickness strips were isolated and fixed in formalin for histology and immunohistochemistry. The balance of the bladder body was separated into muscle and mucosa, flash-frozen in liquid nitrogen, and used for protein extraction and biochemical assays. Nonobstructed control animals, both with (n = 4) and without (n = 4) exposure to L-NAME, were euthanized and had tissue isolated for study in an analogous manner.
Physiological studies. Bladder strips were mounted in a 30-ml physiological medium of Tyrode solution with dextrose (1 mg/ml) and gassed with 95% O2-5% CO2 at 37°C. Two grams of tension were applied to each strip, and the strip was equilibrated for 15 min. The maximal response to electrical field stimulation (2, 8, and 32 Hz, 80 V, 1-ms duration, 15-s train) was determined. Strips were then exposed successively and independently to carbachol (20 μM) and potassium chloride (120 mM), and the contractile response was determined. Peak tension was recorded for each of the different stimuli. The concentrations in these experiments were based on the results of dose-response studies showing that they produced maximal responses in normal and obstructed rabbit bladder strips. Between additions of pharmacological agents, each strip was washed three times with fresh buffer at 10-min intervals.
Protein isolation and SDS-PAGE and Western blotting. Frozen bladder tissues (100 mg) were ground to a fine powder in a mortar cooled in liquid nitrogen and homogenized in buffer containing 20% glycerol, 50 mM Tris·HCl (pH 6.8), 0.5% (vol/vol) Tween 20, and protease inhibitors (0.5 mM PMSF, 2 μM pepstatin, 2 μM antipain, and 0.1 mg/ml trypsin inhibitor). After addition of SDS (final concentration, 1%), the sample was boiled for 4 min and centrifuged at 10,000 rpm for 15 min. Protein concentration in the supernatant was measured using the Pierce BCA protein assay kit. Equal amounts (20 μg) of total protein from control, obstructed, and L-NAME-treated rabbit bladders were loaded on 12% SDS-PAGE gels and transferred to Immobilon-P membranes with Towbin buffer [25 mM Tris, 192 mM glycine and 20% (vol/vol) methanol]. Membranes were blocked with 5% nonfat milk in 0.05% Tween 20 in PBS for 1 h at room temperature and then incubated with primary antibody, monoclonal antibody to nitrotyrosine (Alexis Biochemicals, San Diego, CA). After treatment with the primary antibody, the membranes were washed with 20 mM Tris, 500 mM NaCl, and 0.05% Tween 20 and incubated with secondary antibody (goat anti-mouse IgG at 1:10,000). Substrates were visualized by using echochemiluminescence (Amersham Pharmacia Biotech, Buckinghamshire, UK) and by exposing the membranes to autoradiographic films (Kodak X-OMAT). Band intensities were scanned and analyzed with a Kodak Image Station 440CF and a Kodak ID image analysis software (Scientific Image System, Rochester, NY). Standard curves were constructed to establish that the protein concentrations used for analysis fell within the linear portion of the curves.
Histologic studies. Full-thickness specimens were cut from the bladder, placed in OCT compound (TBS, Durham, NC), snap-frozen in liquid nitrogen, and stored at –70°C. Transverse sections 5-μm thick were then cut on a Microm HM505E cryostat at –25°C and mounted on Bond Rite-treated slides (Richard Allan Scientific, Kalamazoo, MI).
The frozen sections were air dried for 1 h, fixed in ice-cold acetone for 10 min, and then air dried again. Before sections were stained, they were soaked in two changes of PBS for 15 min each to remove the OCT. Another section of full-thickness bladder tissue from each specimen was fixed in 10% neutral buffered formalin for 4–8 h, processed routinely, and embedded in a paraffin block. Before embedding, the square section of tissue was sliced into two or three strips, and each was embedded on edge to provide a cross-sectional view of bladder after microtome sectioning. Five-micrometer-thick sections were cut from each block and mounted on positively charged slides, deparaffinized in xylene, and used for the various immunostaining studies.
Immunostaining. Tissue sections were incubated with the mouse monoclonal antibody to nitrotyrosine and another set with hypoxyprobe primary antibody on the Ventana ES automated immunostainer (Ventana Medical Systems, Tucson, AZ) using an avidin-biotin and diaminobenzidene chromagin detection followed by a light hematoxylin counterstain. The detection kit from Ventana was used with the dilutions, with times and temperatures set by the supplier except for the following: initial blocking done with serum-free protein block (DAKO, Carpinteria, CA) for 12 min. Nitrotyrosine primary antibody was applied for 1 h at room temperature at a dilution of 1:50 and hypoxyprobe primary antibody at a dilution of 1:100 in antibody diluent (DAKO). The secondary antibody in the kit was substituted with biotinylated goat anti-mouse predilute (Lab Vision, Fremont, CA). After the immunostaining procedure, all slides were dehydrated through graded alcohols and then mounted with Permount (Fisher Scientific, Pittsburgh, PA) mounting medium.
Neurofilament staining. Nerve damage is investigated through immunohistochemistry evaluation of the presence and density of neurofilaments within the axons of bladder nerves. Tissue sections were stained with the mouse monoclonal antibody anti-neurofilament clone 2f11 (Neomarkers, Fremont, CA) on the Ventana ES automated immunostainer (Ventana Medical Systems) using an avidin-biotin and diaminobenzidene chromagin detection followed by a light hematoxylin counterstain. Neurofilament primary antibody (100 μl) was applied for 32 min at a dilution of 1:20 in antibody diluent (DAKO). The secondary antibody in the kit was substituted with biotinylated goat anti-mouse predilute (Lab Vision). After the immunostaining, all slides were dehydrated through graded alcohols and two changes of xylene and then mounted with Permount (Fisher Scientific) mounting medium.
Nerve calculations. The nerve data including size and quantity were acquired from the stained cross sections using Image Pro Plus (Media Cybernetics, Silver Spring, MD) image analysis software. Images for analysis were acquired using a Spot charge-coupled device color digital camera linked to an Olympus BX-60 microscope. The transverse section of each specimen was captured in four frames at x100 magnification. The nerves in each image were highlighted using the Image Pro Plus software by manually selecting the pixel value within the stained nerves using the color cube-based tool in the count/size application of the program. The area measurement was then selected for all the nerves highlighted, and the data were saved to Excel and later compiled and compared for all the groups.
Statistical analysis. A Student's t-test was used to calculate difference between means in the contractile function and bladder weights obtained in this experiment. Statistical significance was judged by a P value of <0.05.
RESULTS
There was no significant difference in bladder weight between those control animals premedicated with L-NAME and those that were not. There was a progressive increase in bladder weight among the obstructed animals. However, in those animals premedicated with L-NAME, there was a significantly smaller increase in bladder weight at both 3 and 7 days of PBOO (Fig. 1).
In general, there was a progressive decrease in contractile response to all forms of stimulation with duration of obstruction. With regard to the contractile responses to field stimulation, for all frequencies of stimulation, after 1 day of obstruction, those animals treated with L-NAME had significantly poorer contractile function compared with those with PBOO alone. This trend reversed at 3 and 7 days of PBOO, with the L-NAME group contracting better than the PBOO alone group, which showed significant difference at 32 Hz (Fig. 2). Similarly, for both carbachol and KCl (Fig. 3) after 1 day of obstruction, those animals treated with L-NAME had significantly poorer contractile function compared with those with PBOO alone. This trend reversed at 3 and 7 days of PBOO, with the L-NAME group contracting better than the PBOO alone group.
There was a progressive and significant increase in the expression of nitrotyrosine over the 7-day period of obstruction. In contrast, at the 3- and 7-day time periods, the level of nitrotyrosine in the L-NAME-treated-obstructed rabbits decreased to control levels, providing strong evidence that NO-generated RNS is involved in the response to obstruction (Fig. 4).
Immunohistochemistry also demonstrated increased expression of nitrotyrosine in obstructed rabbits. Again, there was significantly less staining of nitrotyrosine in L-NAME-treated groups (Fig. 5). The presence of nitrotyrosine in the PBOO rabbit bladders indicates that oxidants derived from NO such as peroxynitrite are generated during bladder outlet obstruction and may be involved in its pathogenesis. Using hypoxyprobe-1, we found no evidence of hypoxia in the control groups. After PBOO, at day 1 there was minimal staining only in the urothelium. After 3 days, there was moderate staining in the urothelium as well as light staining in the vascular endothelium. At day 7, there was intense staining in the urothelium and vascular endothelium, as well as small amounts of hypoxia in the muscularis (Fig. 6, A–D). When examining the samples from animals treated with L-NAME, we found differences at day 1 of obstruction; in addition, there was significantly greater staining of the urothelium compared with those with PBOO alone and significant vascular endothelial uptake. No differences were seen at days 3 and 7 with L-NAME vs. PBOO alone (Fig. 6, E and F).
In the neurofilament studies, tissue from control animals demonstrated staining of nerve fibers of all sizes throughout the submucosa and muscularis. There were no changes seen after 1 day of obstruction. At 3 days, more than one-half of small nerve tracts were absent, but large nerve bundles were still present in similar number as controls. By day 7, there was no small nerve tract staining, and there was a small decrement in the number of large nerve tracts (Fig. 7). At each time point, the nerve density of the L-NAME-treated groups was significantly higher than their nontreated counterparts (Fig. 8). Thus treatment with L-NAME provided significant protection of the nerves, which are the most sensitive subcellular structures to ischemic and reperfusion damage.
DISCUSSION
In this study, we investigated the effects of L-NAME, a NOS inhibitor, on the response to PBOO-mediated injury in rabbits at 1, 3, and 7 days after obstruction. We found a striking time course of effect, with L-NAME reducing bladder function 1 day after obstruction but enhancing bladder function at 3 and 7 days.
We found that L-NAME negatively affected bladder contractile properties and histological appearance 1 day after surgical obstruction. This has never been studied previously. It is well recognized that blood flow generally speaking is reduced after partial bladder obstruction. However, in the first day after obstruction, there is an NO-stimulated increase in blood flow that is thought to mediate the maintenance of function and the bladder hypertrophy that begins at this time. This is confirmed by histological studies that demonstrate no significant hypoxia or neural injury. Our study uses L-NAME to block NO. Our findings of decreased function and increased ischemia in L-NAME-treated rabbits confirm the previous findings that NO serves to help maintain function during the first day after obstruction. An alternative cause of bladder dysfunction 1 day postobstruction in L-NAME-treated animals may be inducible NOS expression. By day 1 of PBOO, inducible NOS is not likely to be expressed, and L-NAME will inhibit constitutive endothelial NOS and neuronal NOS. This would decrease blood flow and exacerbate ischemia, decreasing contractility as observed. The implications of our study are significant in that they give further confirmation that blood flow is a key mediator of function after PBOO.
We also found that L-NAME blockade of NO helped to prevent deterioration of bladder function 3 and 7 days after obstruction. This paradoxical finding has never been reported previously. However, our findings are entirely consistent in this area. Not only field stimulation (which reflects nerve function) but also carbachol and KCl (which reflect muscarinic receptor stimulation and smooth muscle contraction) responses were all similar. In addition, our histological studies of nerve staining supports nerves being more intact when L-NAME is used during obstruction. However, we believe that the explanation is found by studying the response to I/R.
In the contractile studies, L-NAME significantly enhanced the contractile dysfunctions to all forms of stimulation at 1 day but showed improved contractile responses compared with untreated obstructed bladders at 3 and 7 days. Previous studies (14) demonstrated that PBOO resulted in a significant NO-stimulated increase in blood flow at 1 day, which returned to control levels at 3 and 7 days postobstruction. This increase in blood flow would be expected to compensate for the ischemia mediated by the obstruction. In the L-NAME-treated rabbits, this increase in blood flow would have been reduced or eliminated and can account for both the increased hypoxia observed at the 1-day period of obstruction and the enhanced contractile dysfunctions observed at this time. However, L-NAME would also be expected to reduce or eliminate the RNS generated by the increased NO, with the resulting reduction in RNS-stimulated generation of nitrotyrosine and reduction in membrane damage. This can account for the improved contractile function at 3 and 7 days and also the higher nerve densities at these time periods.
PBOO causes an increase in bladder mass primarily because of smooth muscle hypertrophy and hyperplasia of the urothelium and interstitial fibroblasts (6, 16). The increased wall thickness and wall tension result in cyclical I/R during and subsequent to each voiding contraction (8, 9). Cyclical I/R leads to an increase in free Ca2+ that activates Ca2+-dependent hydrolytic enzymes such as calpain, phospholipase A2, and endonucleases (20, 21). The ROS and RNS generated subsequently result in characteristic damage to proteins, phospholipids, and DNA that eventually disrupt cellular function (7, 17).
We have shown that these I/R-initiated events begin in transient hypoxic foci within the bladder wall during the bladder's initial response to obstruction and that RNS generation would also result from this cyclical I/R. There may be several stimuli for NOS expression in response to bladder obstruction. Mechanical stretch is known to induce NOS expression (15). Hence, the increased pressure toward the bladder that accompanies bladder outlet obstruction may cause production of NOS in detrusor smooth muscle cells. Oxidatively modified proteins have been used as early markers of oxidative stress in a variety of physiological and pathophysiological processes. Damage to proteins by ROS and RNS affects the function of enzymes, receptors, and signal transduction mechanisms (5). Nitrotyrosine, generated on reaction of the RNS (ONOO–) with protein tyrosine residues, was associated with neuronal ischemic injury and numerous neurodegenerative diseases (19). Antibodies specific to nitrotyrosine can measure this biomarker of oxidative stress. We have found an overexpression of nitrotyrosine in obstructed rabbit bladders compared with control bladders.
In conclusion, there is clear evidence of early cellular damage after PBOO. Hypoxia, which may be extrapolated to ischemia, occurs in bladder detrusor smooth muscle tissues, causing severe damage to nerves and the related contractile response to field stimulation. It has been shown that NO exerts a positive influence at 1 day of obstruction by increasing the blood flow to the bladder tissue but subsequently causes reperfusion-based injury due to free radical damage. Pretreatment with L-NAME significantly enhanced the ischemic damage at 1 day of obstruction but significantly inhibited the generation of nitrotyrosine and resulted in a remarkable preservation of nerve density, which indicates that L-NAME reduces free radical damage associated with PBOO. Further studies will be directed toward proteomic methods, which can identify specific proteins nitrated in outlet obstruction and the complex effect of NO on biological systems.
GRANTS
This work was supported in part by Yamaneuchi USA Research Foundation, the Office of Research and Development Medical Research Service, Department of Veteran's Affairs, and by National Institute of Diabetes and Digestive and Kidney Diseases Grant RO-1 DK-067114.
FOOTNOTES
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
REFERENCES
Andersson KE and Persson K. The L-arginine/nitric oxide pathway and nonandergenic, non-cholinergic relaxation of the lower urinary tract. Gen Pharmacol 24: 833–899, 1993.
Azadzoi KM, Tarcan T, Kozlowski R, Krane RJ, and Siroky MB. Overactivity and structural changes in the chronically ischemic bladder. J Urol 162: 1768–1778, 1999.
Beckman JS, Ye YZ, Anderson P, Chen J, Accavitti MA, Tarpey MM, and White CR. Extensive nitration of protein tyrosines in human atherosclerosis detected by immunohistochemistry. Biol Chem Hoppe-Seyler 375: 81–88, 1994.
Bennett BC, Kruse MN, Roppolo JR, Flood HD, Fraser M, and de Groat. Neural control of urethral outlet activity in vivo: role of nitric oxide. J Urol 153: 2004–2009, 1995.
Davis KL, Martin E, Turko IV, and Murad F. Novel effects of nitric oxide. Annu Rev Pharmacol Toxicol 41: 203–236, 2001.
Gabella G and Uvelius B. Urinary bladder of rat: fine structure of normal and hypertrophic musculature. Cell Tissue Res 262: 67–79, 1990.
Gosling G, Kung LS, Dixon JS, Horan P, Whitbeck C, and Levin RM. Correlation between the structure and function of the urinary bladder following partial outlet obstruction. J Urol 163: 1349–1356, 2000.
Greenland JE, Hvistendahl JJ, Anderson H, Jorgensen TM, McMurray G, Cortina-Borja M, Brading AF, and Frokiaer J. The effect of bladder outlet obstruction on tissue oxygen tension and blood flow in the pig bladder. Br J Urol 85: 1109–1114, 2000.
Greenland JE and Brading AF. The effect of bladder outflow obstruction on detrusor blood flow changes during the voiding cycle in conscious pigs. J Urol 165: 245–248, 2001.
Haddad IY, Pataki G, Hu P, Beckman JS, and Matalon S. Quantification of nitrotyrosine levels in lung sections of patients and animals with acute lung injury. J Clin Invest 94: 2407–2413, 1994.
Huie RE and Padmaja S. The reaction of NO with superoxide. Free Radic Res Commun 18: 195–199, 1993.
Ischiropoulos HL, Zhu J, Chen M, Tsai JC, Martin CD, Smith, and Beckman JS. Peroxynitrite mediated tyrosine nitration catalyzed by superoxide dismutase. Arch Biochem Biophys 298: 431–437, 1992.
Kooy NW, Royall JA, Ye YZ, Kelly DR, and Beckman JS. Evidence for in vivo peroxynitrite production in human acute lung injury. Am J Respir Crit Care Med 151: 1250–1254, 1995.
Lieb JI, Chichester P, Kogan B, Das AK, Leggett RE, Schroder A, and Levin RM. Rabbit urinary bladder blood flow changes during the initial stage of partial outlet obstruction. J Urol 164: 1390–1397, 2000.
Miyajima A, Chen J, Kirman I, Poppas DP, Vaughan ED, and Felsen D. Interaction of nitric oxide and transforming growth factor-1 induced by angiotensin 11 and mechanical stretch in rat renal tubular epithelial cells. J Urol 164: 1729–1734, 2000.
Monson FC, Wein AJ, Eika B, Murphy M, and Levin RM. Stimulation of the proliferation of rabbit bladder urothelium by partial outlet obstruction and acute overdistension. Neurourol Urodyn 13: 51–62, 1994.
Nordling J, Artibani W, Hald TI, Keuppens F, Levin R, and Thuroff J. Pathophysiology of the urinary bladder in obstruction and ageing. In: Benign Prostatic Hyperplasia, edited by Chatelain C, Denis L, Foo KT, Khoury S, and McConnell J. Plymouth, UK: Plymbridge Distributors, 2001.
Schroder A, Chicester P, Kogan B, Longhurst P, Lieb JI, Das AK, and Levin RM. Effect of chronic bladder outlet obstruction on blood flow of the rabbit bladder. J Urol 165: 640–646, 2001.
Smith MA, Richey-Harris PL, Satre LM, Beckman JS, and Perry G. Widespread peroxynitrite damage in Alzheimer's disease. J Neurosci 17: 2653–2657, 1997.
Tominaga T, Kure S, Narisawa K, and Yoshimoto T. Endonuclease activation following focal ischemic injury in the rat brain. Brain Res 608: 21–26, 1992.
Udea N, Walker PD, Hsu SM, and Shah SV. Activation of a 15-kDA endonuclease in hypoxia/reoxygenation injury without morphologic features of apoptosis. Proc Natl Acad Sci USA 92: 7202–7206, 1995.(William Conners, Catherine Whitebeck, Pa)