Biomechanical characterization of the urethral musculature

【摘要】  Rigorous study of the associations between urethral structural anatomy and biomechanical function is necessary to advance the understanding of the development, progression, and treatment of urethral pathologies. An ex vivo model was utilized to define the relative biomechanical contributions of the active (muscle) elements of the female urethra relative to its passive (noncontractile) elements. Whole urethras from female, adult rats were tested under a range of applied intraluminal pressures (0 to 20 mmHg) as a laser micrometer simultaneously measured midurethral outer diameter. Active tissue characterization was performed during induced contraction of either smooth muscle alone ( N -nitro- L -arginine, phenylephrine), striated muscle alone (sodium nitroprusside, atropine, hexamethonium, acetylcholine), or during collective activation of both muscles ( N -nitro- L -arginine, phenylephrine, acetylcholine). The subsequent collection of paired passive biomechanical responses permitted the determination of parameters related to intrinsic muscle contractile function. Activation of each muscle layer significantly influenced the biomechanical responses of the tissue. Measures of muscle responsiveness over a wide range of sustained opposing pressures indicated that an activated striated muscle component was approximately one-third as effective as activated smooth muscle in resisting tissue deformation. The maximum circumferential stress generated by the striated muscle component under these conditions was also determined to be approximately one-third of that generated by the smooth muscle (748 ± 379 vs. 2,229 ± 409 N/m 2 ). The experiments quantitatively reveal the relative influence of the intrinsic urethral smooth and striated muscle layers with regard to their effect on the mechanical properties and maximum functional responses of the urethra to applied intralumenal stresses in the complete absence of extrinsic influences.

【关键词】  smooth striated characterization

THE URETHRA IS A DYNAMIC TISSUE comprised of both circumferential and longitudinally oriented smooth and striated muscle within a connective tissue matrix. It serves a dual function by maintaining continence during bladder filling and aiding in the release of urine from the bladder during micturition. Unlike the smooth muscle layer, which is relatively uniform and continuous throughout the entire axial length, the striated muscle layer becomes much more dense within the midurethral region which has been termed the rhabdosphincter ( 15, 21, 25 ). Continence is maintained through the generation of urethral closure pressures that exceed intravesical pressures.

Experimental studies involving pharmacological neuromuscular blockage demonstrate the fundamental importance of the urethral musculature in maintaining continence, both at rest and during strenuous activities (35-37). Clinical associations have been established between the emergence of certain forms of incontinence (e.g., stress urinary incontinence) and deficits in the urethral musculature and/or associated nerves ( 1, 18, 22, 26 - 28, 40 ). Furthermore, the anatomic establishment of minimum limits on urethral structure and associated musculature necessary to maintain continence has been serendipitously revealed through required resections (e.g., orthotopic lower urinary tract reconstruction) ( 30 ). Despite such clinical observations and associations, the female urinary continence mechanism as a whole remains somewhat controversial. Experimental studies have yet to clearly discern the quantitative contribution of the structures that contribute to the generation of closure forces under normal and stressful (e.g., coughing, sneezing, straining) conditions ( 35 ). The lack of agreement on fundamental issues regarding the maintenance of continence speaks to the necessity of clarifying basic urethra structure-function relationships. To begin to address such deficiencies, and gain an improved understanding of urethral function, the experiments described herein were designed to quantify the individual contributions of both the urethral smooth and striated muscle layers in the female rat urethral in response to a wide range of applied transmural pressure loads. To accomplish this, a recently developed ex vivo whole urethra experimental model was utilized ( 19 ).

There remains a fundamental need for clear identification of structure-function relationships encompassing both the urethra and its supporting tissues. We demonstrate here, through the validation of traditional biomechanics methodology and concepts, how the activation of each individual muscle component can affect the overall function of this complex tissue. Our results embody the first known detailed ex vivo description and comparison of the biomechanical behavior of the active and passive female urethra.

MATERIALS AND METHODS

Animals. Urethras were obtained from adult female Sprague-Dawley rats (260-320 g, 8-10 wk of age; Harlan) that were housed at the University of Pittsburgh under the supervision of the Department of Laboratory Animal Resources. The policies and procedures of the animal laboratory are in accordance with those detailed in the Guide for the Care and Use of Laboratory Animals published by the US Department of Health and Human Services. Procedural protocols were approved by the Institutional Animal Care and Use Committee.

Urethra isolation and experimental testing apparatus. Intact urethras were isolated under halothane anesthesia (4%), as previously described ( 19 ). To maintain in vivo length, a catheter (PE-50) was inserted into the urethral lumen before excision of the tissue and secured with sutures at the midbladder and at the urethral meatus. The entire bladder-urethra unit was removed and immediately placed into cold, oxygenated Medium 199 (Sigma) for transport to a previously described testing system ( 19 ).

The intact urethra was secured at the maintained in vivo length inside of the experimental apparatus and was enclosed within a chamber that was perfused with bathing medium (Medium 199; Sigma) at 37°C and bubbled with a 95% O 2 -5% CO 2 gas mixture. A minimum of 30 min was allowed to equilibrate the urethra segments in the bathing chamber before testing. A range of urethra intraluminal pressures was generated via a height-adjustable fluid reservoir, and a laser micrometer was positioned to measure urethral outer-diameter (OD) at a single axial location near the mid-urethra (40-60% along axial length, starting from the apex of the bladder). Both pressure and OD measurements were recorded concurrently using LabView software.

Paired pressure-diameter (P-D) responses for both active and passive states of the excised urethra were evaluated. Here, an "active" tissue state refers to that occurring during a pharmacologically induced contraction of one or more urethral muscle components, as described in detail below. A passive tissue state refers to the subsequent elimination of all induced contractions, along with any intrinsic muscle tone that may have also been present, also described below. Tissue samples were used only for the collection of a single set of paired active and passive P-D data.

Active state P-D measurements. Protocols for the activation of smooth and striated muscle components were based on previous pharmacological responsiveness of female rat urethra segments ( 19 ). Three separate muscle activation protocols were employed to selectively stimulate the smooth and striated muscles, as described in Table 1. These muscle activation protocols also employed selective suppression of potential basal activity, where applicable. Suppression of basal smooth muscle tone in the striated muscle alone activation protocol was accomplished through addition of sodium nitroprusside. Previous experiments, showing the near complete elimination of baseline tone through administration of sodium nitroprusside, demonstrate the smooth muscle nature of this baseline tone in the tissue preparation used and confirm that baseline striated tone is likely negligible ( 19 ); therefore, no suppression of striated activity was used in the smooth muscle alone activation protocol. However, although no evidence to the contrary exists at this time, we cannot rule out the possibility that the suppression of smooth muscle activity has an inadvertent affect on striated muscle activation.

Table 1. Urethra muscle activation protocols

For smooth muscle activation alone ( n = 8), N -nitro- L -arginine (100 µM) was first added to the tissue bath to eliminate endogenous nitric oxide synthase activity and nitric oxide production, which can mediate a tonic relaxation of the smooth muscle in some isolated tissues ( 19 ). Following a 20-min incubation period, phenylephrine (PE; 40 µM) was added to specifically stimulate smooth muscle contraction; verified through stabilization to a constricted OD value. For tissues in which concurrent smooth and striated muscle activation was desired ( n = 6), the abovementioned protocol was followed by the addition of acetylcholine (ACh; 5 mM), immediately after verification of the PE-induced contraction. For striated muscle activation alone ( n = 6), sodium nitroprusside (10 µM) was first added to the tissue bath for a period of 30 min to completely inhibit smooth muscle activity as evidenced by dilation of the tissue to a new stable baseline OD. Both atropine (1 µM) and hexamethonium (100 µM) were also added along with the sodium nitroprusside, to further eliminate any possibility of cholinergic-induced smooth muscle activation, before stimulation with ACh (5 mM). Preliminary PE and ACh dose-response studies were performed to ensure maximal induced contractions and to ensure the effectiveness of the blocking agents at the stated concentrations.

For all active tissue experiments, immediately following confirmation of contraction, intralumenal pressure was increased from 0 to 20 mmHg in 2-mmHg increments. P-D data were collected at a 20-Hz sampling rate over a 10-s period for each 2-mmHg pressure step. The OD data were then averaged over the 200 data points obtained at each incremental pressure step to obtain discrete values associated with each applied pressure. The P-D collection period (100 s = 10 s x 10 pressure steps) was chosen to be within the duration of maximal striated muscle contraction based on preliminary measurements. The 20-mmHg pressure range was selected based on preliminary studies to identify the degree of tissue strain in the ex vivo system and to ensure that this degree of strain remains consistent with estimations of the normal physiological urethral strain (occurring during micturition). This pressure is consistent with normal physiological (non-stress) mean bladder pressure in the female rat (3.4 kPa 25 mmHg) ( 32 ).

Passive state P-D measurements. Immediately following collection of active state data, a passive state devoid of muscle tone was induced by addition of EDTA (3 mM final concentration) to the bath, allowing a 30-min equilibration period for elimination of muscle activity ( 19 ). To minimize viscoelastic effects, a mechanical preconditioning protocol, consisting of 10 pressure cycles from 0 to 4 mmHg at 5-s increments, was performed before collection of the passive P-D data. P-D responses became repeatable generally by the sixth to eighth inflation-deflation cycle, as observed through examination of resulting hysteresis loops. Note that preconditioning was not performed before the collection of active state data and is commonly avoided in studies of active mechanics due to concerns of compromising contractile function with repetitive stretching ( 12, 14, 17 ).

Determination of muscle volume fraction and tissue thickness. Immediately following collection of P-D data, and before tissue removal for histological processing, the location at which the laser was positioned for OD measurements was identified by marking the outside of the tissue with a permanent ink pen. The tissue was then fixed at the maintained in vivo length using 2% paraformaldehyde. The local content of smooth and striated muscle at the location of measurement was determined through immunofluorescent labeling of fast and slow myosin heavy chain and -smooth muscle actin [protocols previously described ( 19 ); all antibodies from Sigma]. For each specimen, local muscle content and tissue thickness were determined by averaging values obtained from three cryosections taken over the marked segment, 1 mm in axial length. The actual urethral wall cross-sectional area (WCSA) for each section was determined through image processing (Northern Eclipse software, v6.0; Epix Imaging,) from the difference in the cross-sectional areas (CSA) enclosed by manually drawn borders of both the lumenal and ablumenal surfaces. Unloaded tissue thickness, h, was calculated as the difference between the estimated unpressurized outer radius ( R o ) and inner radius ( R i ) based on these CSA measurements and an approximation of the tissue to a cylindrical geometry, as described in equations 1 - 3.

( 1 )

( 2 )

( 3 )

Estimation of inner radius. The tissue inner radius, R i x, under a given pressure, P x, was estimated from application of the incompressibility assumption, which is commonly invoked in (and has been validated for) arterial mechanics studies ( 5 ). This assumption, that the cross-sectional area of the tissue wall will remain unchanged under the various levels of tissue strain, allows for the estimation of R ix at each pressure step (P x ) based on corresponding OD measurements (yielding outer radii, R o x ) and from known tissue dimensions ( R o, R i ) taken at a reference pressure (P o = 0 mmHg). Because the tissue was tethered at both ends, changes in axial length were not permitted. This constant tissue length, along with conservation of tissue volume, dictate that the tissue maintains a constant wall cross-sectional area (i.e., WCSA P x = WCSA Po ). Thus assuming an approximate cylindrical geometry yields the estimation of R i x at any pressure (P x ) based on the following equations 4 - 5 :

( 4 )

( 5 )

Previous estimations of R i x using these assumptions, specifically examined for various blood vessels over a wide range of pressure conditions, were found to be within a mean of less than 1% of actual measurements ( 13 ).

Muscle contraction parameters. Parameters descriptive of the relative contractile function of the active muscle element(s) over the entire applied pressure range were obtained from the P-D data. Each parameter is graphically depicted in Fig. 1.

Fig. 1. Two measures of contractile function obtained from urethral pressure-diameter (P-D) data. The total contraction capacity ( A ) is a single dimensionless value representing the ratio of the area between the passive and active P-D curves (A1) to the area under the passive P-D curve (A2) alone. The functional contraction ratio ( B ), computed separately for each applied pressure, defines the dimensionless ratio of the difference between the active and passive normalized outer diameter measurements ( D p - D a ) relative to the passive normalized outer diameter measurement alone ( D p ). Adapted from Ref. 14.

The total contraction capacity (TCC), defined as the ratio of the area between the active and passive curves to the area under the passive curve alone ( Fig. 1 A ), may be considered the relative ratio of the work performed by the activated muscle to the elastic energy stored by the passive elements of the tissue alone ( 14 ). To obtain this parameter, the OD data were normalized to the initial diameter value (i.e., D / D 0 = 1.0 at P = 0 mmHg) to attain a common starting point for each set of passive and active curves. To calculate the area under each curve, each set of P-D data was fit separately to a third-order polynomial function ( y = ax 3 + bx 2 + cx + d ) using nonlinear regression (SigmaStat, v2.0). Once the coefficients were determined, the function describing the P-D curve was integrated over the same normalized diameter range for each paired passive and active data set to estimate the representative areas under each respective curve. The lower limit was taken to be the initial normalized diameter starting point (i.e., 1.0), and the upper limit taken to be the maximum active normalized diameter value (i.e., at P = 20 mmHg). Because the active tissue deformed to a lesser degree than the passive tissue, the active maximum normalized diameter was chosen as the upper integration limit so that the active curve did not have to be extrapolated.

An additional muscle contraction parameter was calculated from the difference between the measured passive and active diameters for each applied pressure, which provides an indication of the relative degree of isobaric contraction for the activated muscle component(s) ( Fig. 1 B ). The functional contraction ratio (FCR) was previously defined as the difference between the passive ( D p ) and active ( D a ) diameters relative to the passive diameter (i.e., D p - D a / D p ) ( 14 ). This ratio was calculated for each pressure step from the normalized P-D data for each activation state.

Maximum stress generation. For the urethra preparation described here, the tissue thickness to radius ratio (h/r) was always greater than 0.05, and thus the tissue could not be treated biomechanically as a thin-walled vessel ( 8 ). Circumferential stress ( ) within a thick-walled, linearly elastic, isotropic cylinder as a function of any radial position (r) within the wall has been previously derived ( 8 ) and is given below in equation 6 :

( 6 )

where P x is the internal (transmural) applied pressure and R i x and R o x are the previously defined estimated and measured inner and outer radii, respectively.

Using equation 6, urethra midwall responses were calculated and were subsequently used to determine the maximum stress generated by each activated muscle component. To accomplish this, nonlinear regression was used to fit the experimental data to either an exponential ( y = a *exp bx; for active striated and all passive responses) or a third-order polynomial function ( y = ax 3 + bx 2 + cx; for active smooth and smooth-striated responses), based on maximizing R 2 values, using SigmatStat (v2.0; Jandel Scientific). Active and passive fitted values were then obtained over the entire normalized diameter (OD/OD o ) range at 0.02 increments. Maximum generated by the active muscle component alone for each tissue was determined as the maximum difference between the fitted active and passive values at each normalized diameter increment. Fitted vs. normalized diameter responses, and subsequent estimation of maximum, were performed within the limits of the experimentally observed diameter range (i.e., requiring no extrapolation).

Statistical analysis. Comparison of OD and ID responses between active and corresponding passive tissue states at each applied pressure was performed using paired t -test, or the equivalent nonparametric Wilcoxon signed rank test. Comparison of the contraction parameters TCC and FCR among active tissue states (smooth, striated, and smooth striated) was performed using one-way ANOVA with Student-Newman-Keuls post hoc pairwise comparisons, or the equivalent nonparametric ANOVA on ranks with Dunn's post hoc analysis. FCR comparisons were performed at each applied pressure, separately. P values <0.05 are reported as significant, and data are expressed as means ± SE. All statistical comparisons were performed using SigmaStat (v2.0; Jandel Scientific) statistical software.

RESULTS

Tissue histology. The striated muscle layer within the mid-urethral region contained a prevailing alignment of myofibers in the circumferential direction ( Fig. 2, left ). Inspection of the smooth muscle layer within the midregion also revealed a predominant circumferentially oriented muscle layer, along with some longitudinally oriented fibers both intermingled within the circumferential layer and located closer to the lumen of the tissue. For tissues utilized in these experiments, smooth and striated muscle content in the midregion as a percentage of total cross-sectional area was within the same range as previously reported (smooth, 32.8-37.8%; striated 17.2-19.7%) ( 19 ). The striated muscle layer was predominantly composed of type II (fast) fibers, also as previously reported ( 19 ).

Fig. 2. Immunofluorescent labeled cross-sections of the female rat mid-urthera, representative of the muscle orientations found within the mid-urethra. Predominant circumferential layers of both smooth and striated muscle are apparent, as well as a mixture of longitudinally oriented smooth muscle. Striated muscle was revealed with antibodies against both fast and slow myosin heavy chain and smooth muscle with an antibody against -smooth muscle actin.

P-D responses. Average P-D plots displaying responses collected within the mid-urethral region for all three active and passive muscle states are provided in Fig. 3. In general, all active tissue states displayed an initial resistance to deformation on increased applied pressure, resulting in an S-shaped P-D curve. This was more pronounced in the presence of an activated smooth muscle component. In contrast, the presence of active striated muscle alone was not sufficient to provide resistance to deformation distinguishable from that of the corresponding passive state at both the lowest and highest applied pressures ( Fig. 3, middle ).

Fig. 3. Applied pressure vs. measured normalized outer diameter (OD/OD o ) responses measured at the mid-urethra. Tissue responses measured in the absence (passive) and presence (active) of smooth muscle activity alone ( n = 8), striated muscle activity alone ( n = 6), and combined smooth and striated muscle activity ( n = 6). * P < 0.05, ** P < 0.01.

Muscle contraction parameters. Initial diameters, to which the OD diameter data were normalized, demonstrated consistency in both the tissue preparation and state of the tissue at the time of experimentation (1.651 ± 0.182). The nonlinear regression performed to obtain the TCC parameter provided an excellent fit for both the active and passive P-D data for all three experimental protocols ( r 2 : striated alone, 0.977 ± 0.020 and 0.996 ± 0.001; smooth alone, 0.954 ± 0.013 and 0.994 ± 0.002; smooth-striated, 0.941 ± 0.015 and 0.993 ± 0.002, for active and passive data fits, respectively). A significant difference in TCC was observed between tissues in which striated muscle alone was activated compared with those in which both smooth and striated muscle was activated ( Fig. 4 A ). No statistically significant difference was apparent for smooth muscle-activated tissues in the presence or absence of striated muscle activity.

Fig. 4. Muscle contraction parameters: P-D-derived measures of relative muscle contractile function within the mid-urethra over the entire applied pressure range [total contraction capacity (TCC); A ], or at each pressure step individually [functional contraction ratio (FCR); B ]. Active tissue responses, compared with passive responses, measured in the presence of smooth muscle activity alone ( n = 8), striated muscle activity alone ( n = 6), and combined smooth and striated muscle activity ( n = 6). A : * P < 0.05 striated vs. smooth-striated. B : * P < 0.05 striated vs. smooth-striated and ** P < 0.05 striated vs. smooth.

Comparison of contractile function between the activated muscle elements at each applied pressure step, as defined by the FCR parameter, is shown in Fig. 4 B. Both striated and smooth muscle exhibit their maximal influence within approximately the same pressure range (from 4 to 10 mmHg). However, smooth muscle activation, either with or without concomitant striated activity, resulted in a significantly increased contraction within the 4- to 14-mmHg range, and again at 20 mmHg, compared with striated muscle activation alone (nonsignificant multiple comparisons: P = 0.141, 0.051, and 0.055 at 2, 16, and 18 mmHg, respectively). No significant differences were observed however for smooth muscle-activated tissues in the presence or absence of striated muscle activity.

Maximum stress generation. A good correlation between actual and fitted data were observed for all of the midwall circumferential stress ( ) vs. normalized diameter responses, which were subsequently used to calculate maximum generated by each activated muscle component. The best fit for tissues with active striated muscle, as well as for all passive tissues, was obtained using an exponential function ( r 2 : striated active, 0.994 ± 0.003; striated passive, 0.996 ± 0.001; smooth passive, 0.993 ± 0.002; smooth-striated passive, 0.995 ± 0.001), whereas the best fit for smooth and smooth-striated active tissues was obtained using a polynomial function ( r 2 : 0.985 ± 0.007 and 0.987 ± 0.004, respectively). An example of these fitted passive and active plots for each of the activation protocols in relation to the measured data is shown in Fig. 5. Also shown in this figure are plots of the difference between the active and passive responses, which is representative of that which would be hypothetically observed from the activated muscle component alone (i.e., if the underlying passive component could be removed from the tissue).

Fig. 5. Experimental and fitted (model) passive and active midwall circumferential stress ( ) responses of the midurethra plotted against changes in normalized outer diameter (OD/OD o - 1). Examples for each of the 3 activated tissue states examined are shown: smooth muscle alone ( top ), striated muscle alone ( middle ), and both smooth striated simultaneous activation ( bottom ). Also shown are the hypothetical muscle-only responses, obtained as the difference between the active and passive fitted data.

From these responses, the average maximum was estimated to be 748 ± 379 N/m 2 for striated muscle alone, 2,229 ± 409 N/m 2 for smooth muscle alone, and 2,335 ± 239 N/m 2 for smooth and striated muscle activated tissues; all of these maximum stresses occurred at a similar degree of midwall deformation for each activated tissue state ( 0.1-0.25 change in normalized OD). However, from Fig. 5, it can be seen that the active smooth muscle component generated a more immediate and continuous amount of stress over a wider range of tissue deformation, in contrast to that observed for striated muscle activation alone.

DISCUSSION

Biomechanical characterization of the intact female rat urethra was performed through ex vivo experimental measures of changes in the outer diameter of active and passive tissues in response to controlled increases in intralumenal pressure. Previously, we demonstrated that such measurements performed during the application of a single fixed pressure can be used to delineate regional urethra contractile responses in relation to axial changes in muscle composition and content ( 19 ). In principal, we now demonstrate that the experimental methodology utilized herein, which has traditionally been employed to characterize vascular tissue responses ( 12, 14, 17 ), may also be applied for the investigation of urethra functional properties within a physiological pressure range. The results obtained utilizing this methodology indicate that the female rat intrinsic urethral smooth muscle, compared with striated muscle, is capable of imparting a greater influence on the long-term functional responses of the urethra; the smooth muscle generated higher levels of maximum circumferential stress and also sustained higher levels of stress over a broader range of applied opposing pressures.

This study represents a continuance of previous work performed with a primary goal of assigning biomechanical, and subsequently clinical, significance to the inherent structures of the urethra. It is generally believed that both the smooth and striated muscle of the urethra play a substantial role in the production of closure force at rest ( 2, 4, 33 ). It is also generally believed that the urethral striated muscle, along with that of the supporting pelvic floor, is a chief source of adjunct closure pressure required to maintain continence during stress episodes ( 35 ). Indeed, deficits of innate striated muscle function and composition, not smooth muscle, are most associated with the clinical presentation of stress-related incontinence ( 18, 23, 26, 27 ). Despite such clinical associations, the urinary continence mechanism in women as a whole remains a considerably controversial issue. Contradictions in the perceived relative importance of each of the intraurethral muscle layers (smooth and striated), including the potential quantitative contribution of each component toward the overall generation of urethral closure force, continue to persist within the urologic community without resolution. For instance, it has been proposed that both the smooth and striated musculature, as well as the external pelvic floor musculature, may all coordinately contribute to increased urethral closure pressures during stress episodes in a bladder-to-urethra reflex mechanism ( 20 ). Yet others have proposed that external compression of the urethra by pelvic floor contraction, rather than inherent contraction of urethral musculature, contributes to such increased pressures ( 38 ). Even during resting conditions some have asserted that the striated muscle, rather than the smooth muscle, provides the bulk of resting urethral closure pressure ( 11, 31 ).

Attempts to add clarity to these issues as well as to definitively quantify the contributions of each intraurethral muscle component have met with further discrepancy. Studies performed in the canine have estimated striated contributions to the urethral pressure profile to be one-third ( 3 ) to over one-half ( 34 ). In striking contrast, other studies, also performed in the canine, have been unable to detect any significant influence by this muscle component ( 6 ). Additionally, studies performed in the pig have also shown little contribution of striated muscle to resting pressures ( 16 ). The results presented herein in the female rat mid-urethra estimate the maximal striated muscle contribution to force generation to be approximately one-third of that of the smooth muscle; furthermore, sustainability of these contractions suggest that the striated muscle layer is more likely to act as an adjunct, shorter-term closure mechanism to the continuous principal influence of the smooth muscle during normal (nonstressful) urinary storage. However, it should also be stated the temporal role of the striated muscle's contribution may ultimately be more relevant to the overall continence mechanism than absolute force generation capability. This is a topic requiring further research.

As has been astutely observed, animal studies cannot and should not simply be extrapolated directly to the human clinical situation, owing in the least to differences in: species anatomy, receptor distribution and composition, and possible confounding effects of anesthesia during measurements ( 35 ). Although anesthesia effects are plainly absent during ex vivo measurements, there remain several important experimental issues to consider with regard to the conclusions presented herein, in particular, and the ex vivo or in vitro study of urethral striated muscle function, in general. First, although the female rat is a widely accepted model with which to perform functional urethra studies, the predominantly fast (type II) myosin isotype of the striated muscle observed in the young rats utilized here is contradictory to the predominantly slow (type I) isotype of striated muscle histological findings in both the human female and in older rats ( 2, 15, 29 ). The temporal nature of striated muscle is fundamentally influenced by the predominance of fiber type present. Our findings of a predominantly type II urethral striated layer in the female rat, and a conclusion that this muscle is more suitable to responding to sudden and brief periods of applied stress, coincide with previous urethra strip studies in both the canine and sheep demonstrating fiber-type influence over contractile speed and sustainability ( 9 ). In addition, the use of pharmacological agents to induce striated muscle contraction can result in desensitization of the nicotinic cholinergic receptors, contributing to a more transient striated muscle response. Previous experience has demonstrated the length of maximal ACh-induced striated contraction to be of the same order of time required to complete the incremental pressure ramping ( 19 ). The priming volume of the system, together with chemically evoked receptor desensitization, precluded the possibility of challenging the contractile function of the tissue before initiation of the mechanical testing protocol. Thus no method of assessing uniform strength of contraction during application of the opposing intralumenal pressures was possible. Therefore, currently, we cannot rule out the possibility at this time of underestimation of striated maximum contractile responses or the duration of such responses. Modification of the system to allow electrical field stimulation (EFS)-induced muscle contractions may help to clarify and lessen the potential impact of such issues ( 39 ). Nonetheless, the overall experimental biomechanics approach described herein is adeptly suited toward delineating the quantitative contributions of the intrinsic muscle responses of the urethra and we have no reason to believe that this system is not perfectly suited at the present toward addressing smooth muscle functional responses in both an accurate and reliable manner.

Basic research should continue toward a full understanding of the coordinated efforts and biomechanical contributions of the urethral musculature and related structural components that comprise the continence securing mechanism. It is of fundamental importance to determine the magnitude of the effect caused by alterations of these components with regard to the overall mechanical properties of the tissue. This is true in both defining pathophysiological changes in tissue composition and function and subsequently devising therapeutic strategies to address such changes. The recent interest in cellular uromyoplasty therapy to regenerate functional urethral striated muscle illustrates this position ( 7, 10, 24, 31 ). Currently, estimation of the clinical impact imparted by increases in striated muscle cannot be gauged simply from evaluation of the currently available literature, which may confound the expectations of such a therapy in relation to the amount of regeneration which can be realistically attained through cellular transplantation techniques. Thus clarification of such fundamental issues will be indispensable to further the practical understanding of urethra pathophysiology, but also will be critical in the continued development of strategies for the reconstruction, restoration, or functional replacement of the dysfunctional urethra.

GRANTS

This work was partially supported by National Institutes of Health Grants RO1-HL-65745, RO1-DK-55387, RO1-AR-049398, K12-DK-02656, and RO1-DK-067226 to D.A. Vorp, J. Huard, M.B. Chancellor, and R.J. Jankowski.

ACKNOWLEDGMENTS

The authors thank J. Zhou for technical assistance.

【参考文献】
  Allen RE, Hosker GL, Smith AR, and Warrell DW. Pelvic floor damage and childbirth: a neurophysiological study. Br J Obstet Gynaecol 97: 770-779, 1990.

Ashton-Miller JA, Howard D, and DeLancey JO. The functional anatomy of the female pelvic floor and stress continence control system. Scand J Urol Nephrol Suppl 00: 1-7; discussion 106-125, 2001.

Awad SA and Downie JW. Relative contributions of smooth and striated muscles to the canine urethral pressure profile. Br J Urol 48: 347-354, 1976.

Brading AF. The physiology of the mammalian urinary outflow tract. Exp Physiol 84: 215-221, 1999.

Carew TE, Vaishnav RN, and Patel DJ. Compressibility of the arterial wall. Circ Res 23: 61-68, 1968.

Cass AS and Hinman F Jr. Constant urethral flow in female dog. I. Normal vesical and urethral pressures and effect of muscle relaxant. J Urol 99: 442-446, 1968.

Chancellor MB, Yokoyama T, Tirney S, Mattes CE, Ozawa H, Yoshimura N, de Groat WC, and Huard J. Preliminary results of myoblast injection into the urethra and bladder wall: a possible method for the treatment of stress urinary incontinence and impaired detrusor contractility. Neurourol Urodyn 19: 279-287, 2000. <a href="/cgi/external_ref?access_num=10.1002/(SICI)1520-6777(2000)19:3

Chandran KB. Cardiovascular Biomechanics. New York: New York Univ. Press, 1992.

Chen X and Creed KE. Histochemical and contractile properties of striated muscles of urethra and levator ani of dogs and sheep. Neurourol Urodyn 23: 702-708, 2004.

Chermansky CJ, Tarin T, Kwon DD, Jankowski RJ, Cannon TW, de Groat WC, Huard J, and Chancellor MB. Intraurethral muscle-derived cell injections increase leak point pressure in a rat model of intrinsic sphincter deficiency. Urology 63: 780-785, 2004.

Colleselli K, Stenzl A, Eder R, Strasser H, Poisel S, and Bartsch G. The female urethral sphincter: a morphological and topographical study. J Urol 160: 49-54, 1998.

Cox RH. Comparison of arterial wall mechanics in normotensive and spontaneously hypertensive rats. Am J Physiol Heart Circ Physiol 237: H159-H167, 1979.

Faury G, Maher GM, Li DY, Keating MT, Mecham RP, and Boyle WA. Relation between outer and luminal diameter in cannulated arteries. Am J Physiol Heart Circ Physiol 277: H1745-H1753, 1999.

Fridez P, Makino A, Miyazaki H, Meister JJ, Hayashi K, and Stergiopulos N. Short-term biomechanical adaptation of the rat carotid to acute hypertension: contribution of smooth muscle. Ann Biomed Eng 29: 26-34, 2001.

Gosling JA. The structure of the female lower urinary tract and pelvic floor. Urol Clin North Am 12: 207-214, 1985.

Greenland JE, Dass N, and Brading AF. Intrinsic urethral closure mechanisms in the female pig. Scand J Urol Nephrol Suppl 179: 75-80, 1996.

Greenwald SE, Newman DL, and Denyer HT. Effect of smooth muscle activity on the static and dynamic elastic properties of the rabbit carotid artery. Cardiovasc Res 16: 86-94, 1982.

Hale DS, Benson JT, Brubaker L, Heidkamp MC, and Russell B. Histologic analysis of needle biopsy of urethral sphincter from women with normal and stress incontinence with comparison of electromyographic findings. Am J Obstet Gynecol 180: 342-348, 1999.

Jankowski RJ, Prantil RL, Fraser MO, Chancellor MB, De Groat WC, Huard J, and Vorp DA. Development of an experimental system for the study of urethral biomechanical function. Am J Physiol Renal Physiol 286: F225-F232, 2004.

Kamo I, Cannon TW, Conway DA, Torimoto K, Chancellor MB, de Groat WC, and Yoshimura N. The role of bladder-to-urethral reflexes in urinary continence mechanisms in rats. Am J Physiol Renal Physiol 287: F434-F441, 2004.

Keane DP and O'Sullivan S. Urinary incontinence: anatomy, physiology and pathophysiology. Baillieres Best Pract Res Clin Obstet Gynaecol 14: 207-226, 2000.

Koelbl H, Mostwin J, Boiteaux JP, Macarak E, Petri E, Schafer W, and Yamaguchi O. Pathophysiology. In: Incontinence (2nd ed.), edited by Abrams P, Cardozo LD, Khoury S, and Wein A. Plymouth, OK: Health Publication, 2002, p. 205-241.

Kuo HC. Transrectal sonographic investigation of urethral and paraurethral structures in women with stress urinary incontinence. J Ultrasound Med 17: 311-320, 1998.

Lee JY, Cannon TW, Pruchnic R, Fraser MO, Huard J, and Chancellor MB. The effects of periurethral muscle derived stem cell injection on leak point pressure in a rat model of stress urinary incontinence. International J Urogynecol In press.

Oelrich TM. The striated urogenital sphincter muscle in the female. Anat Rec 205: 223-232, 1983.

Perucchini D, DeLancey JO, Ashton-Miller JA, Galecki A, and Schaer GN. Age effects on urethral striated muscle. II. Anatomic location of muscle loss. Am J Obstet Gynecol 186: 356-360, 2002.

Perucchini D, DeLancey JO, Ashton-Miller JA, Peschers U, and Kataria T. Age effects on urethral striated muscle. I. Changes in number and diameter of striated muscle fibers in the ventral urethra. Am J Obstet Gynecol 186: 351-355, 2002.

Rud T. Urethral pressure profile in continent women from childhood to old age. Acta Obstet Gynecol Scand 59: 331-335, 1980.

Russell B, Baumann M, Heidkamp MC, and Svanborg A. Morphometry of the aging female rat urethra. Int Urogynecol J Pelvic Floor Dysfunct 7: 30-36, 1996.

Stenzl A, Colleselli K, and Bartsch G. Update of urethra-sparing approaches in cystectomy in women. World J Urol 15: 134-138, 1997.

Strasser H, Berjukow S, Marksteiner R, Margreiter E, Hering S, and Bartsch G. Stem cell therapy for urinary stress incontinence. Exp Gerontol 39: 1259-1265, 2004.

Streng T, Santti R, and Talo A. Similarities and differences in female and male rat voiding. Neurourol Urodyn 21: 136-141, 2002.

Sullivan MP and Yalla SV. Physiology of female micturition. Urol Clin North Am 29: 499-514, vii, 2002.

Tanagho EA, Meyers FH, and Smith DR. Urethral resistance: its components and implications. II. Striated muscle component. Investig Urol (Berl) 7: 195-205, 1969.

Thind P. The significance of smooth and striated muscles in the sphincter function of the urethra in healthy women. Neurourol Urodyn 14: 585-618, 1995.

Thind P and Lose G. The effect of bilateral pudendal blockade on the adjunctive urethral closure forces in healthy females. Scand J Urol Nephrol 28: 249-255, 1994.

Thind P and Lose G. The effect of bilateral pudendal blockade on the static urethral closure function in healthy females. Obstet Gynecol 80: 906-911, 1992.

Umek WH, Laml T, Stutterecker D, Obermair A, Leodolter S, and Hanzal E. The urethra during pelvic floor contraction: observations on three-dimensional ultrasound. Obstet Gynecol 100: 796-800, 2002.

Van der Werf BA, Hidaka T, and Creed KE. Continence and some properties of the urethral striated muscle of male greyhounds. BJU Int 85: 341-349, 2000.

Van Geelen JM, Lemmens WA, Eskes TK, and Martin CB Jr. The urethral pressure profile in pregnancy and after delivery in healthy nulliparous women. Am J Obstet Gynecol 144: 636-649, 1982.

作者单位：1 Bioengineering Department, University of Pittsburgh; 2 Growth and Development Laboratory, Children‘s Hospital of Pittsburgh; and 3 Department of Pharmacology, 4 Division of Urologic Surgery, 5 Department of Orthopaedic Surgery, and 6 Division of Vascular Surgery, University of Pittsburgh Sch