0022-5347/01/1662-0651/0 THE JOURNAL OF UROLOGY® Copyright © 2001 by AMERICAN UROLOGICAL ASSOCIATION, INC.®
Vol. 166, 651– 657, August 2001 Printed in U.S.A.
THE DECOMPENSATED DETRUSOR V: MOLECULAR CORRELATES OF BLADDER FUNCTION AFTER REVERSAL OF EXPERIMENTAL OUTLET OBSTRUCTION RAIMUND STEIN, JOEL C. HUTCHESON, LEV KRASNOPOLSKY, DOUGLAS A. CANNING, MICHAEL C. CARR AND STEPHEN A. ZDERIC From the Division of Urology, Children’s Hospital of Philadelphia, Philadelphia, Pennsylvania, and Department of Urology, University of Mainz, Mainz, Germany
ABSTRACT
Purpose: Calcium ion homeostasis has a significant role in smooth muscle function. Its regulation requires complex storage and release mechanisms via ion pumps and channels located within intracellular storage sites (sarcoplasmic reticulum) and at the plasma membrane. We have previously reported a dramatic loss of the 2 major sarcoplasmic reticulum proteins sarcoplasmic endoplasmic reticulum calcium magnesium adenosine triphosphatase (SERCA2) and the ryanodine sensitive ion channel, also called the ryanodine receptor, after outlet obstruction. In our current study we investigated the correlation of the expression of these 2 major sarcoplasmic reticulum components with bladder function recovery after the reversal of outlet obstruction. Materials and Methods: Standard partial bladder outlet obstruction was created in adult New Zealand White rabbits. Voiding patterns were monitored 2 and 4 weeks postoperatively, and rabbits were selected for outlet obstruction reversal based on a voiding pattern consistent with a decompensated state, as indicated by a frequency of greater than 30 voids daily and an average voided volume of less than 4 cc. Bladder biopsy was done when outlet obstruction was reversed. Voiding performance was monitored postoperatively and the animals were sacrificed 2 weeks later. Voiding patterns and muscle strip studies enabled us to define 2 functional outcome categories after reversal, namely normal versus minimally improved. Microsomal membrane protein fractions were prepared from the same bladder tissues before and after reversal, and probed by Western blot analysis for SERCA2 and ryanodine receptor expression. Results: Western blot analysis revealed a major loss of SERCA2 and ryanodine receptor expression at the time of reversal and biopsy. In 65% of bladders obstruction reversal resulted in a normalized voiding pattern with a recovery of ryanodine receptor expression that was 15% to 65% of control values. In contrast, in the 35% of bladders with persistent voiding symptoms there was minimal recovery of ryanodine receptor expression. SERCA2 expression increased slightly in each group after reversal but did not differ in bladders with normalized versus improved function. Conclusions: Bladder decompensation is highly associated with a loss of sarcoplasmic reticulum function. Furthermore, the decompensated detrusor recovers function after obstruction reversal, which is associated with the recovery of these sarcoplasmic reticulum components. KEY WORDS: bladder, bladder outlet obstruction, rabbits, sarcoplasmic reticulum
Partial bladder outlet obstruction in humans and in experimental animal models results in numerous changes within the detrusor. In human studies it is known that the detrusor may accommodate to imposed obstruction and continue to empty, although at higher pressure and increased effort. This response may be considered compensatory. In older patients with benign prostate hyperplasia Sullivan and Yalla observed a point at which no further compensation occurred and decompensation began, as manifested by a loss of detrusor reserve, that is the difference between maximal isometric pressure and maximal voiding pressure.1 This loss of detrusor reserve clinically correlated with increased post-void residual urine as well as patient symptoms of urinary frequency and decreased voided volume. It is also known that after the reversal of outlet obstruction in humans voiding symptoms may persist in up to 30%.2 To our knowledge the
molecular mechanisms leading to bladder wall decompensation and any markers for irreversible damage after surgical relief remain unclear. While in humans the degree, duration and cause of obstruction varies among individuals, animal models have the advantage of creating the same degree of obstruction in each animal, enabling bladder assessment at defined points. Animal models also provide the opportunity to study carefully phenotypic changes after outlet obstruction using a number of physiological measures, such as bladder mass, muscle strip performance, voiding pattern analysis and video urodynamics.3 Using these tools to characterize the state of the bladder is critical before any molecular analysis, since despite standard protocols to create experimental partial outlet obstruction, some bladders adapt to the imposed work load and others do not.4 Noninvasive means for monitoring bladder performance become especially important when performing a study of the outcome after the reversal of outlet obstruction. In our previous series we have shown that approximately 30% of bladders have compensated performance after outlet
Accepted for publication March 9, 2001. Supported by National Institutes of Health Grants P50DK52620, K12DK12-02196 and DFG931/1-1, and the Leonard and Madlyn Abramson Chair in Pediatric Urology. 651
652
DECOMPENSATED DETRUSOR
obstruction, whereas we classified 70% as decompensated.4 We have previously shown that the analysis of voiding patterns may correlate highly with muscle strip physiology and whole bladder video urodynamic analysis.3, 5 In other words, by monitoring voiding patterns noninvasively we selected for reversal procedures animals with bladder decompensation. Our earlier study has shown that major changes develop with decompensation in the expression of 2 major components of the sarcoplasmic reticulum, namely the sarcoplasmic endoplasmic reticulum calcium magnesium adenosine triphosphatase (SERCA2)4, 6 and ryanodine sensitive ion channel, also known as the ryanodine receptor.5 Further study at our laboratory indicated that while there is a major loss of SERCA2 and ryanodine receptor expression with decompensation, minimal changes developed in the ␣-1 subunit expression of the voltage operated calcium channel.5 This finding led us to hypothesize that major alterations in sarcoplasmic reticulum composition and subsequent disruptions in cytosolic calcium handling are an important step in the pathways leading to decompensation. In addition to initiating muscle contractility, cytosolic calcium has a pivotal role in other cellular functions, such as exocytosis, endocytosis, protein synthesis and assembly. Calcium ion dysregulation is a central feature of the process of cellular injury and may result in fibrosis or apoptosis.7 If our hypothesis is correct that major alterations in sarcoplasmic reticulum are an integral step in the pathway to bladder decompensation, it should follow that recovery from the decompensated state should be associated with recovery of the sarcoplasmic reticulum. We tested this hypothesis by reversing partial outlet obstruction that was created experimentally in a well characterized rabbit model. To ensure that a more homogeneous population of bladders was selected for reversal, voiding patterns were monitored to choose only those meeting our criteria for decompensation. Bladder biopsy was done at laparotomy to reverse outlet obstruction and voiding patterns were monitored after the second surgical procedure (fig. 1). This scenario enabled comparison of sarcoplasmic reticulum protein expression in a single bladder at 2 time points. Our hypothesis was that the sarcoplasmic reticulum protein expression (SERCA2 and ryanodine receptor) would be substantially improved after reversal in bladders with recovered function and minimally affected in those with minimal recovery. In contrast, we anticipated that there
FIG. 1. Overview of experimental protocol. Of 27 rabbits initially obstructed 20 met criteria for decompensated detrusor based on voiding pattern analysis. BOO, bladder outlet obstruction.
would be little change in the ␣-1 subunit of the voltage operated calcium channel, which is expressed at the plasma membrane. MATERIALS AND METHODS
Operative procedure/obstruction. All studies were performed in 4-month-old male New Zealand white rabbits and approved by the Children’s Hospital of Philadelphia animal use committee. After sedation with 35 mg./kg. ketamine and 5 mg./kg. xylazine intravenous administration of sodium pentobarbital (thiopental) was used to induce deep anesthesia. After inserting an 8Fr catheter into the bladder via the urethra the urethra and bladder neck were exposed through a small vertical lower abdominal incision. Extraperitoneal dissection enabled for minimal urethral mobilization and mobilization of the periurethral fat was decreased to a minimum. A right angle clamp was passed around the urethra and a 2-zero silk suture was used to create obstruction. To maximize the standardization of partial outlet obstruction another 8Fr catheter was placed outside of the urethra and the silk suture was tied around each catheter. The catheters were then removed. In the sham operated group the silk suture was cut and removed after identical dissection. All operations were performed by one of us (R. S.). Voiding patterns. The rabbits were placed in metabolic cages with free access to food and water, and allowed a 24-hour period to adjust to the new environment. Voided urine was then monitored by collection on digital scales interfaced to a computer. Scales were monitored at 2-minute intervals. Collected data were transferred to a spreadsheet for graphing and analysis. Frequency data are expressed as voids per 24 hours and average voided volume is presented as cc per void. Noninvasive monitoring of the rabbit voiding pattern was initiated 10 days after obstruction. Rabbits selected for obstruction reversal demonstrated certain selection criteria characteristic of a decompensated bladder,3, 5 including greater than 30 voids daily and an average voided volume of less than 4 cc. After the reversal of outlet obstruction and biopsy voiding patterns were again monitored to assess the degree of functional recovery. Reversal surgery and tissue procurement. After 14 or 28 days deep anesthesia was induced and the bladder was exposed via the same midline incision. An intraperitoneal incision was made and the bladder urine was aspirated via a 14 gauge needle. The bladder was gently cradled with moistened gauze sponges to minimize any urine leakage into the peritoneum. A 2 ⫻ 3 cm. section of the anterior bladder wall was excised with a scalpel. The muscle and mucosa from this detrusor biopsy were separated and stored separately in liquid nitrogen until analysis. The detrusor wall was over sewn in 2 layers with 5-zero chromic suture and the abdominal wall was closed in layers of 3-zero polyglactin suture. Two weeks after biopsy and outlet obstruction reversal deep anesthesia was again induced and the bladders were exposed via the same midline incision, enabling the whole bladder and urethra to be excised. The bladder was blotted dry with paper towels and weighed. The bladder trigone was removed and the mucosa was peeled away, leaving the detrusor muscle with the attached serosa. Three 0.2 ⫻ 0.2 ⫻ 1 cm. muscle strips were prepared and the remainder of the detrusor muscle was immediately transferred for storage in liquid nitrogen. Smooth muscle physiology. The muscle strips were attached by a 4-zero silk suture to a post in 5 ml. of tissue-organ bath (Radnoti Glass Technology, Inc., Monrovia, California) at 1 end and an isometric force transducer (Grass Instruments, Quincy, Massachusetts) at the other. The transducer was calibrated with known weights and the output directed to a Model 7H polygraph (Grass instruments). The tissue organ baths were maintained at 37C and contained Tyrode’s
653
DECOMPENSATED DETRUSOR
solution (125 mM. sodium chloride, 2.7 mM. potassium chloride, 1.8 mM. calcium chloride, 0.5 mM. magnesium chloride, 23.8 mM. sodium bicarbonate, 0.4 mM. sodium biphosphate and 5.6 mM. glucose) perfused by a bubbled mixture of 95% oxygen/5% carbon dioxide. After equilibration for 30 minutes hour at slack length the strips were gradually stretched to the length at which optimal force is generated. Peak tension was measured in response to 32 Hz., 80 V. field stimulation to a high potassium chloride containing solution (127 mM. potassium chloride, 1.8 mM. calcium chloride, 0.5 mM. magnesium chloride, 23.8 mM. sodium bicarbonate, 0.4 mM. sodium biphosphate and 5.6 mM. glucose), and direct cholinergic stimulation with 200 mol. bethanechol. After completion of the experiments the muscle between the 2 silk sutures was weighed. All results are expressed in gm. of tension per 100 mg. tissue. Histology. Full-thickness strips from the detrusor obtained at bladder biopsy and final sacrifice were fixed in 5% buffered formalin. The strips were embedded in paraffin and 6 m. tissue sections were cut. Sections were stained with conventional hematoxylin and eosin and Masson’s trichrome preparations. The muscle fraction was analyzed using a microscope equipped with a digital camera and image analysis software. Bladder classification after reversal. The rabbits were divided into a normalized and an improved group after reversal based on muscle strip performance in response to direct cholinergic stimulation with bethanechol, normalized that is a force of greater and less than 10 gr. tension per 100 mg. tissue, respectively. Because all data were available for all bladders, it was also possible to define the normalized and improved groups based on voiding parameters alone as a voiding pattern that returned to normal with less than 10 voids daily and an average voided volume of greater than 20 cc per void, and a voiding pattern that improved after reversal with greater than 10 voids daily and an average voided volume of less than 20 cc per void. All results are presented using these definitions. Protein preparation. Each detrusor smooth muscle sample was stored individually, enabling us to correlate molecular findings with individual bladder performance before and after the reversal of obstruction. The frozen detrusor muscle was minced by scissors on ice, followed by homogenization in 10 mM. sodium bicarbonate Trizma buffer (15 ml./1 gm. tissue) containing protease inhibitor cocktail (100 l./gm. tissue) (Sigma Chemical Co., St. Louis, Missouri). Homogenization was followed by low speed centrifugation at 25 minutes and 1,600 ⫻ gravity at 4C. The resulting supernatant was centrifuged for 35 minutes at 8,600 ⫻ gravity at 4C. This supernatant was then ultracentrifuged at 93,000 ⫻ gravity for 120 minutes at 4C. Membrane pellets were suspended in the same buffer, as described, and stored at ⫺80C until use. Total protein content was determined using a modified Lowry method.8, 9 Western blot analysis. The membrane fraction was dissolved in 6% sodium Dodecyl sulfate, 50 mM. dithiothreitol, 10 mM. ethylenediaminetetraacetic acid, 0.83 mM. benzamide, 0.23 mM. phenylmethyl sulfonylfluoride, 1 mM. lodoacetamide, 0.5 M. sucrose and 130 mM. Trizma base adjusted to pH 6.8 and heated at 100C for 5 minutes. Standard gel electrophoresis (4% polyacrylamide stacking gel, 7.5% polyacrylamide separating gel and 20 g. membrane protein per lane) was done at 110 V. The running buffer was 0.025 M. Trizma base, 0.192 M. glycine and 10% sodium dodecyl sulfate, adjusted to pH 8.3. Gels were removed and proteins were blotted onto nitrocellulose membranes using the Bio-Rad Mini transfer system (BioRad Laboratories, Hercules, California) at 60 V. overnight at 4C. Transfer buffer consisted 0.025 M. Trizma base, 0.192 M. glycine and 20% methanol. The membrane was then immersed in 10% nonfat milk in phosphate buffered saline (PBS) containing 1.85 mM. NaH2PO4, 8.39 mM. Na2HPO4 and 150 mM. sodium chloride
adjusted to pH 7.4 for 1 hour. The membrane was then washed 4 times for 10 minutes each with PBS with 0.3% Tween. The monoclonal antibodies tested (ryanodine receptor against isoforms 1 and 2, SERCA2 and the voltage operated calcium channel ␣-1 subunit) (RDI, Flanders, New Jersey), were diluted 1:1,000 in PBS containing 0.77 mM. 0.1% sodium azide, 0.1% NP-40 and 3% bovine albumin fraction, and incubated with the membrane overnight at 4C. The membrane was washed 5 times with PBS with 0.3% Tween and incubated for 1.5 hours at room temperature with horseradish peroxidase labeled goat anti-mouse IgG antibody diluted 1:10,000 in PBS containing 0.1% NP-40. After discarding the antibody solution the membrane were washed 5 times in PBS/Tween buffer and incubated for 1 minute in ECL Western blot detection reagents (Amersham Life Sciences, Arlington Heights, Illinois). Membranes were placed in a cassette and x-ray film was developed to show the bands. Correlation was done with known molecular weight standards. Densitometry of the Western blots was done on each blot using commercially available computer software. On each Western blot 20 g. of 2 control bladders were loaded as well as samples from the same bladder before and after reversal. The intensity of each band was normalized to the band of the control bladders on the same blot. Control bladders were considered to show 100% expression. The specificity of each band was confirmed by repeating the blot in the presence of all reagents and secondary antibody in the absence of primary antibody. For statistical analysis the 2-sided t test was performed with p ⬍0.05 considered significant. RESULTS
Physiology. The table lists the mean results plus or minus standard error of mean of urinary frequency and voided volume before and after obstruction. The table also lists bladder characteristics after the reversal of outlet obstruction at sacrifice. While bladder weight did not vary significantly, there were statistically significant differences in the responses to bethanechol and in the voiding patterns. Sham surgery induced no significant changes in the voiding pattern in terms of mean frequency or average voided volume versus nonoperated controls (6 ⫾ 3 versus 4 ⫾ 3 voids daily and 26 ⫾ 16 versus 31 ⫾ 17 cc, respectively). However, following outlet obstruction the decompensated group voided with much greater frequency than the control or sham group and with much lower voided volume (p ⬍0.05, see table). The preoperative differences in frequency and voided volume in the normalized and improved groups after reversal were not statistically significant. However, postoperatively after the reversal of outlet obstruction 2 distinct patterns were noted. In 13 rabbits voiding function recovered to almost normal, while in 7 there was minimal improvement in voiding parameters. The differences in voiding parameters in these 2 groups was highly significant (p ⬍0.05). There were also significant differences in muscle strip performance in these groups. Western blot analysis. SERCA2 and ryanodine receptor expression were significantly decreased in all bladder biopsies obtained at reversal, whereas there was a minimal effect
Normalized No. Obstruction (mean ⫾ SEM): No. voids/day (cc) Voided vol. (cc) Reversal (mean ⫾ SEM): No. voids/day (cc) Voided vol. (cc) Bladder mass Bethanechol
13 43 ⫾ 5 3⫾1 7⫾2 43 ⫾ 6 4.5 ⫾ 2 24.1 ⫾ 4.1
Improved 7 47 ⫾ 12 3⫾1 19 ⫾ 5 (p ⬍0.05) 14 ⫾ 4 (p ⬍0.05) 5.1 ⫾ 1 7.4 ⫾ 1.1 (p ⬍0.05)
654
DECOMPENSATED DETRUSOR
on the ␣-1 subunit of voltage operated calcium channel expression. SERCA2 and ryanodine receptor expression were increased after the removal of obstruction and a 2-week recovery period (fig. 2). A minimal increase in ryanodine receptor expression was observed in bladders with a minimal improvement in function. In contrast, major improvement in bladder function after reversal was associated with a major increase in ryanodine receptor expression (fig. 3). SERCA2 expression increased after the reversal of outlet obstruction but there was no difference in expression in the normalized and improved groups (fig. 4). The greatest increase in the recovery of SERCA2 expression was noted in the group with minimal improvement. However, the final expression of SERCA2 was still significantly depressed at 42% of the baseline control value. In contrast, there were minimal shifts in voltage operated calcium channel expression at reversal and after recovery. Expression of the ␣1 subunit of the voltage operated calcium channel remained at approximately 120% of control values at each time points in each outcome group. Histology. There was no significant difference in the percent of smooth muscle fraction in these 2 groups (fig. 5). However, in bladders with improvement there was a 10% decrease in the muscle fraction after the release of the outlet obstruction from 49% ⫾ 3% to 39% ⫾ 3%, which was statistically significant.
FIG. 3. Western blot densitometry data on ryanodine receptor expression when obstructed bladder underwent bladder biopsy (OBS) and after 2 weeks of recovery following reversal (REV). Difference in expression in bladders with normalized function (Norm) after reversal was highly significant (p ⬍0.05). Difference in 2 groups during obstruction was not significant. Increase in expression after reversal in improved (Imp) group was not significant.
DISCUSSION
The fate of the bladder following partial outlet obstruction has been extensively studied and reviewed.10 It remains controversial as to how much the physical properties of the bladder change and to which bladder wall fraction these physiological alterations may be attributed. Others believe that smooth muscle is minimally affected in terms of its contractile parameters11 and the major changes in bladder performance may be attributable to the deposition of major amounts of extracellular matrix throughout the bladder wall.12 Cher et al observed no change in the percent of myosin light chain phosphorylation determined at peak tension and concluded that smooth muscle fiber contractile performance was unaffected by obstruction.13 In contrast, Xiaoling and Moreland reported minimal to no changes in peak force but a 10-fold decrease in the velocity of shortening after outlet obstruction and a major increase in baseline myosin light chain phosphorylation.14 These results support the hypothesis that smooth muscle cell performance is affected by outlet obstruction.
FIG. 2. Representative Western blot analysis. A, SERCA2. B, ryanodine sensitive ion channel. Eight bladders, including 2 controls (CH) and matched pairs of bladders with obstruction (o) and then reversed after 2-week recovery period (r). Also shown is whether bladder had normal function after reversal (Norm) or improvement (Imp).
FIG. 4. Western blot densitometry data on SERCA2 expression when obstructed bladder underwent biopsy (OBS) and after 2 weeks of recovery following reversal (REV). Differences in SERCA2 expression before and after reversal did not approach statistical significance. Norm, normalized function. Imp, improved function.
Our previous data have indicated a major loss in the expression of SERCA2 and ryanodine receptor with bladder decompensation as well as minimal shifts in the expression of the voltage operated calcium channel.3– 6 These results imply that significant changes develop within the sarcoplasmic reticulum after outlet obstruction that may correlate with a decrease in bladder wall contractile performance. This damage to the sarcoplasmic reticulum may result in slow and sustained increases in cytosolic calcium, which may in turn lead to cellular hypertrophy and decompensation via the calcineurin pathway.15, 16 If our hypothesis is true that sarcoplasmic reticulum perturbations are a major contributing pathway to bladder wall decompensation, there should also be recovered sarcoplasmic reticulum protein expression in bladders with recovered function after the removal of anatomical obstruction. We used a well characterized model of partial outlet obstruction and voiding pattern analysis to select bladders for reversal with an equivalent degree of severe bladder dysfunction (fig. 1). The selection criteria for severe decompensation were defined as a urinary frequency of greater than 30 voids daily and an average voided volume of less than 4 cc. These criteria were established in previous studies in which voiding patterns were correlated with muscle strip physiology3, 5 and with video urodynamic outcomes (unpublished data). In our standard model of partial outlet obstruction rabbits that met
DECOMPENSATED DETRUSOR
FIG. 5. Percent of smooth muscle demonstrated no major shift in smooth muscle content in 2 bladder populations during obstruction. There was no shift in muscle fraction in those with normalized function (Norm) after reversal of outlet obstruction (REV). In bladders with minimal improvement (Imp) there was 10% mean decrease in muscle fraction from 49% ⫾ 3% to 39% ⫾ 3% (p ⬍0.05). OBS, obstructed bladders.
these voiding criteria had increased bladder mass, bladder wall decompensation on muscle strip analysis, elevated voiding pressures consistent with outlet obstruction and elevated post-void residual urine. Furthermore, this noninvasive technology enabled us to separate the groups that we defined as compensated versus decompensated, which is especially crucial when performing a systematic study of reversal outcomes. Our previous data show that approximately 30% of the rabbits obstructed by our technique had compensated bladder function, although they had obstruction according to elevated voiding pressure. Including these animals with compensated function in a study of the effect of reversal would have confused the results. Voiding pattern analysis enabled us to categorize the phenotype before reversal, so that the most uniform possible cohort of animals was selected. By selecting the worst cases for reversal we are also more confident that any increased function after reversal represented true increases in bladder performance. After releasing obstruction in 20 bladders deemed to be decompensated on voiding pattern analysis almost normal recovery of function occurred in 13 bladders (65%) (see table). In 7 other bladders there was some improvement in voiding parameters, which led us to define this group as minimally improved. There were residual symptoms in 35% of rabbits, similar to the rate of lower urinary tract symptoms reported in men after surgical procedures to relieve prostatic obstruction.2 It is plausible that these residual symptoms reflect persistent obstruction despite the procedure to remove the ligature. However, bladder weight did not differ significantly in the normal and minimally improved groups, while based on our previous data, any significant obstruction should result in increased bladder mass.3– 6 An additional study in this regard would be video urodynamic characterization of animals with residual voiding symptoms after the reversal of obstruction. In a rat model Chai et al identified elevated voiding pressure despite the surgical removal of outlet obstruction.17 Based on our simple muscle strip analysis 2 groups of bladders were defined after outlet obstruction release. Clearly there were differences based on the contractile response to bethanechol in these groups. Likewise animals with residual urinary frequency and decreased volume had poor contractile performance based on these measurements of bladder wall performance (see table). Is it possible that these dramatic differences that we report reflect differences in muscle content? While slight shifts occurred in smooth muscle content on morphometric analysis (fig. 5), these differences do not explain the physiology that we observed.
655
As in our previous study of the ␣-1 subunit of voltage operated calcium channel expression in bladder decompensation, there was no significant change in expression after outlet obstruction or after reversal irrespective of whether bladder function normalized. These data imply that the major shifts in calcium regulatory protein occur at the level of the sarcoplasmic reticulum and not at the plasma membrane. This finding further strengthens the association of sarcoplasmic reticulum function with the response to outlet obstruction. This Western blot analysis is done with a monoclonal antibody specific for the ␣-1 subunit of the voltage operated calcium channel. May shifts occur in other subunits of the voltage operated calcium channel? This possibility must be explored using another independent method of radioligand binding using H3PN-200, which would allow not only the determination of channel density, but also its dissociation constant. These experiments have been deferred in favor of other assays because of the large protein load required for a Scatchard analysis. In contrast to the relatively constant expression of the ␣1 subunit of voltage operated calcium channel, dramatic differences were observed in the expression of the ryanodine receptor (fig. 3). Relative to controls ryanodine receptor expression on Western blot analysis was significantly decreased after the initial period of outlet obstruction. There were small preoperative differences in ryanodine receptor expression in bladders with normal function versus those with minimal improvement after reversal, but they were not statistically significant. However, the differences in ryanodine receptor expression after reversal were highly significant (p ⬍0.05) with bladders with normalized function showing a dramatic recovery of function to a mean of 65% ⫾ 15% of the control value. In contrast, bladders with minimal improvement in ryanodine receptor expression after reversal had little functional improvement. As in our previous study of SERCA2 and ryanodine receptor expression with bladder outlet obstruction each proteins was down-regulated.5 However, the greatest shift developed in the ryanodine receptor, making it appear to be a more sensitive molecular marker of outlet obstruction. SERCA2 expression was markedly decreased after outlet obstruction, consistent with our previous series.4 – 6 There was a slight, statistically insignificant preoperative difference in SERCA2 expression in bladders with normal function versus those with minimal improvement after reversal. After the release of outlet obstruction SERCA2 expression increased in each group and the final expression was equal. Thus, while SERCA2 expression was clearly decreased after outlet obstruction and increased after reversal, its reappearance did not correlate as well with the normalization of function. Why should there be such a difference in the association of ryanodine receptor and SERCA2 expression with normalized function? It is a simplistic to view the sarcoplasmic reticulum as a homogenous intracellular component composed of SERCA, ryanodine receptor, the inositol triphosphate receptor, a calcium binding protein that is not yet defined in bladder and other components. Cellular calcium control is required for many other cellular functions in addition to contractility. The SERCA family comprises 3 major isoforms. Type I is located in fast twitch striated muscle systems and type III is restricted to nonmuscle tissue. Type II SERCA is present in slow twitch fibers of striated muscle and in smooth muscle. SERCA2 has 2 isoforms that were shown by MacLennan et al to be the result of alternative splicing.18 The 2a isoform predominates in the heart and is also evident in smooth muscle. The 2b isoform is present in smooth muscle and nonmuscle systems. It is believed to function more as a general housekeeping ion pump that maintains proper intracellular calcium gradients for a number of critical cellular functions in addition to contractility. Using an SN1 nuclease
656
DECOMPENSATED DETRUSOR
protection assay Lytton et al identified SERCA2a and 2b isoforms in rabbit bladder smooth muscle and estimated that the 2b isoform accounted for 80% of total SERCA2 messenger (m)RNA.19 It is critical to remember that our Western blot analysis of SERCA2 expression was based on the original 2D8 clone developed by Jorgensen et al, which does not differentiate the 2a and 2b isoforms.20 Perhaps if we probe specifically for SERCA2a isoform expression, a better correlation of SERCA expression and normalization of function would become evident. We strongly suspect that it is the case and we are in the process of performing polymerase chain reaction to pinpoint specifically the differences in SERCA 2a and 2b mRNA expression. This finding would then enable the generation of monospecific antibodies to recognize the SERCA 2a and 2b isoforms. Is the loss of sarcoplasmic reticulum protein expression in the form of ryanodine receptor or SERCA2 merely an association with poor bladder wall performance or a contributing factor in its pathogenesis? What evidence supports the notion that sarcoplasmic reticulum function is important for normal bladder smooth muscle function? The early description of Mostwin of receptor mediated intracellular calcium stores implied that intracellular calcium release is crucial for normal force development.21 We demonstrated that in vitro administration of ryanodine suppressed peak force generation, implying that intracellular calcium release is a factor contributing to force development.22 In the whole bladder model Damaser et al noted that applying ryanodine affected peak pressure and bladder emptying.23 These studies also have the distinct advantage of comparing the same tissue to itself before and after adding the drug, eliminating the confounding presence of simultaneous shifts in other bladder wall components, such as the extracellular matrix. Alterations in the sarcoplasmic reticulum associated with loss of function are not unique to the bladder. Human and animal models of aortic stenosis have resulted in end stage cardiomyopathy that has also been shown to be associated with the loss of sarcoplasmic reticulum function.24 –26 Much debate has centered on the best means of studying the pathogenesis of outlet obstruction (murine models, cultured cells or fetal models). The number of model systems may imply that each model has advantages and disadvantages. If a perfect model existed, everyone would be using it. Our current study in the rabbit is advantageous because the same bladder may be studied at a molecular level at 2 time points. Furthermore, the rabbit bladder size enables a biopsy adequate enough to determine protein and mRNA expression at each time point. While it is possible to reverse the aortic ligation and understand the effect that it may have on cardiac function and sarcoplasmic reticulum expression, to our knowledge obtaining myocardial biopsy at reversal to study protein expression and comparing it with the final outcome has not been reported for obvious reasons. Reversal studies in a functionally well defined population accompanied by biopsy, as we have done, cannot be done as readily in murine27 or fetal28 models. Ultimately dissecting the contributions of each bladder wall component to overall bladder function requires close phenotypic study of knockout and transgenic mice in which no surgical interventions are done. We have developed the ability to study murine bladder physiology, which may enable us to analyze the effects of individual gene deletion or over expression on bladder performance in vitro29 or in vivo. Lemack et al reported that successful bladder outlet obstruction may be created in mice.27, 30 We agree that this technology holds great promise for future investigation into the pathophysiology of outlet obstruction. A detailed understanding of the molecular pathophysiology of bladder smooth muscle cell dysfunction is essential for designing novel therapeutic strategies to prevent the onset of end stage bladder disease.
CONCLUSIONS
Bladder performance after outlet obstruction is influenced by the smooth muscle cell ability to maintain calcium homeostasis. The ryanodine sensitive ion channel is the gate that allows the sarcoplasmic reticulum to discharge calcium stores into the cytosol and its expression is markedly affected by outlet obstruction. After obstruction reversal ryanodine receptor expression returned only in bladders with normalized function. While SERCA2 expression decreased after outlet obstruction and recovered after reversal, its expression did not correlate as well with the normalization of function. We speculate that it may be due to our current inability to distinguish the 2a and 2b isoforms. In contrast, voltage operated calcium channel expression is only minimally affected, implying that the major changes in calcium handling proteins occur at the level of the sarcoplasmic reticulum. These data imply that the expression of sarcoplasmic reticulum components in smooth muscle is highly associated with bladder function after outlet obstruction. REFERENCES
1. Sullivan, M. P. and Yalla, S. V.: Detrusor contractility and compliance characteristics in adult male patients with obstructive and non-obstructive voiding dysfunction. J Urol, 155: 1995, 1996 2. McGuire, E. J.: Detrusor Response to Outlet Obstruction. National Institutes of Health Publication 87–2881. Bethesda, Maryland: Department of Health and Human Services, p. 227, 1987 3. Stein, R., Hutcheson, J. C., Gong, C. et al: The molecular response to outlet obstruction: a role for the sarcoplasmic reticulum. Unpublished data 4. Zderic, S. A., Rohrmann, D., Gong, C. et al: The decompensated detrusor II: evidence for loss of sarcoplasmic reticulum function after bladder outlet obstruction in the rabbit. J Urol, part 2, 156: 587, 1996 5. Stein, R., Hutcheson, J. C., Gong, C. et al: Functional outcome after experimental bladder outlet obstruction and its correlation with ryanodine and voltage operated calcium channel expression. BJU Int, 85: 31, 2000 6. Stein, R., Gong, C., Hutcheson, J. C. et al: The decompensated detrusor III: impact of bladder outlet obstruction on sarcoplasmic endoplasmic reticulum protein and gene expression. J Urol, part 2, 164: 1026, 2000 7. Trump, B. F. and Berezesky, I. K.: Calcium-mediated cell injury and death. FASEB J, 9: 219, 1995 8. Lowry, O. H., Rosenbaugh, N. J., Farr, A. L. et al: Protein measurement with the folin phenol reagent. J Biol Chem, 193: 265, 1951 9. Peterson, G. L.: Review of the folin phenol protein quantitation method of Lowry, Rosenbaugh, Farr, and Randall. Anal Biochem, 100: 201, 1979 10. Baskin, L. S. and Hayward, S. W.: Advances in Bladder Research. New York: Kluwer Academic/Plenum, 1999 11. Peters, C. A.: Congenital bladder obstruction: research strategies and directions. In: Muscle Matrix and Bladder Function. Edited by S. A. Zderic. New York: Plenum Press, p. 117, 1995 12. Ewalt, D. H., Howard, P. S., Blyth, B. et al: Is lamina propria matrix responsible for normal bladder compliance? J Urol, part 2, 148: 544, 1992 13. Cher, M. L., Kamm, K. E., D., M. J.: Stress generation and myosin phosphorylation in the obstructed bladder. J Urol, suppl., 143: 355A, 1990 14. Xiaoling, S. and Moreland, R. S.: Obstruction induced changes in smooth muscle of the rabbit urinary bladder. Biophysical J, 78: 111A, 2000 15. Molkentin, J. D., Lu, J. R., Antos, C. L. et al: A calcineurindependent transcriptional pathway for cardiac hypertrophy. Cell, 93: 215, 1998 16. Olsson, E. N. and Williams, R. S.: Calcineurin signaling and muscle remodeling. Cell, 101: 689, 2000 17. Chai, T. C., Gemalmaz, H., Andersson, K. E. et al: Persistently increased voiding frequency despite relief of bladder outlet obstruction. J Urol, 161: 1689, 1999 18. MacLennan, D. H., Brandl, C. J., Korczak, B. et al: Amino acid
DECOMPENSATED DETRUSOR
19.
20.
21. 22. 23.
sequence of a Ca2⫹ ⫹ Mg2⫹ dependent ATPase from rabbit muscle sarcoplasmic reticulum deduced from its complementary DNA sequence. Nature, 316: 696, 1985 Lytton, J., Zarian-Herber, A. A., Periasamy, M. et al: Molecular cloning of the mammalian smooth muscle sarco(endo)plasmic reticulum Ca2⫹ ATPase reticulum. J Biol Chem, 264: 7059, 1989 Jorgensen, A. O., Arnold, W., Pepper, D. R. et al: A monoclonal antibody to the Ca2⫹-ATPase of cardiac sarcoplasmic reticulum cross-reacts with slow type I but not with fast type II canine skeletal muscle fibers: an immunocytochemical and immunohistochemical study. Cell Motil Cytoskeleton, 9: 164, 1988 Mostwin, J. L.: Receptor operated intracellular calcium stores in the smooth muscle of the guinea pig bladder. J Urol, 133: 900, 1985 Zderic, S. A., Sillen, U., Liu, G. H. et al: Developmental aspects of the bladder contractile function, evidence for intracellular calcium pool. J Urol, part 2, 150: 623, 1993 Damaser, M. S., Kim, B., Longhurst, P. A. et al: Calcium regulation of urinary bladder function. J Urol, 157: 732, 1997
657
24. Hassenfuss, G., Reinecke, H., Studer, R. et al: Relation between myocardial function and expression of sarcoplasmic reticulum Ca2⫹ATPase in failing and nonfailing human myocardium. Circ Res, 75: 434, 1994 25. Chien, K. R.: Stress pathways and heart failure. Cell, 98: 555, 1999 26. Chien, K. R.: Meeting Koch’s postulates for calcium signalling in cardiac hypertrophy. J Clin Invest, 105: 1339, 2000 27. Lemack, G. E., Szabo, Z., Urban, C. D. et al: Altered bladder function in transgenic mice expressing rat elastin. Neurourol Urodyn, 18: 55, 1999 28. Quinn, T. M., Kirsch, A. J., Zderic, S. A. et al: Renal effects of partial bladder outlet obstruction in the fetal lamb. Surg Forum, 16: 808, 1997 29. Hutcheson, J. C., Stein, R., Chacko, S. et al: Murine in vitro whole bladder physiology. In: Advances in Bladder Research. Edited by A. Atala. New York: Kluwer, in press, 2001 30. Lemack, G. E., Burkhard, F., Zimmern, P. E. et al: Physiologic sequelae of partial infravesical obstructionin the mouse: role of inducible nitric oxide synthase. J Urol, 161: 1015, 1999