| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Female Reproductive Tract |
Department of Molecular and Integrative Physiology, and the R.L. Smith Mental Retardation Research Center, Kansas University Medical Center, Kansas City, Kansas 66160
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
estradiol, neurotransmitters, ovulatory cycle, steroid hormones, vagina
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Another mechanism by which altered estrogen status may contribute to postmenopausal dysfunction is by affecting vaginal innervation. The mammalian vagina is richly imbued with sensory, parasympathetic, and sympathetic nerve fibers. Sensory nociceptor and tactile nerves are present within dermis and connective tissue, and autonomic sympathetic and parasympathetic fibers abundantly innervate vaginal smooth muscle and vasculature [5, 6]. These nerves are believed to play important roles in mediating vaginal sensitivity and pain, and in regulating blood flow, secretions, and tonus of the vaginal wall [6–9].
Estrogen is a potent regulator of peripheral innervation, particularly with respect to organs of reproduction. Estrogen receptors have been localized to sympathetic, parasympathetic, and sensory neurons projecting to the female reproductive tract [10, 11], and elevated estrogen levels are associated with increased density of calcitonin gene-related immunoreactive sensory innervation to the mammary gland [12], and sympathetic nerve depletion of the uterus [13, 14]. It is therefore possible that estrogen directly affects vaginal innervation, thereby contributing to changes in vaginal function.
In the present study, we explored the possibility that estrogen promotes nerve remodeling in the rodent vagina by quantifying autonomic and sensory innervation density under conditions of high estrogen that occur during the estrus phase of the estrous cycle and after estrogen pellet implantation, and low estrogen "menopause" induced by ovariectomy.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Virgin female rats (Harlan Breeding Laboratories, Indianapolis, IN) obtained at 60 days of age were housed three per cage in a climate- and light- (12L:12D) controlled environment, and received food and water ad libitum. Within a week of arrival, 10 rats were anesthetized by i.p. injection of a mixture of ketamine hydrochloride (60 mg/kg; Phoenix Scientific, St. Joseph, MO), atropine sulfate (0.4 mg/kg; American Pharmaceutical Partners, Los Angeles, CA), and xylazine (8 mg/kg; Phoenix Scientific), and ovariectomy (OVX) was performed aseptically through bilateral incisions [13]. Seven days later, rats were anesthetized as described above, and a pellet was implanted s.c. between the scapulae. Five rats received a 21-day-release pellet containing 0.1 mg of 17ß-estradiol (E-121; Innovative Research of America, Sarasota, FL) and five rats received a matching placebo pellet (C-111; Innovative Research of America). A previous study using identical procedures showed that plasma estrogen levels in OVX rats receiving placebo pellets is below the limits of detection, and 17ß-estradiol pellets produce sustained plasma estrogen levels in the physiological range (220 ± 33 pg/ml) [13]. Five intact rats were allowed to cycle normally, and estrous cycle stages were determined using the criteria described by Long and Evans [15]. Only rats exhibiting at least two consecutive 4- to 5-day estrous cycles were used.
Seven days after pellet implantation or on the first day of estrus, rats were anesthetized as described above. A ventral midline incision was made in the lower abdomen, the pubic bone was removed, and the vagina was exposed. The vagina was transected at its junction with the cervix, separated from adjacent connective tissue, and removed as a single block that included the external vulva. All vaginal tissues were weighed and their lengths were measured. Animals were killed by asphyxiation. All experimental protocols followed the National Institutes of Health Guide for the Care and Use of Laboratory Animals, and were approved by the University of Kansas Medical Center Animal Care and Use Committee.
Tissue Processing
Vaginal tissues were immersed in Zamboni fixative [16] for 24 h at 4°C and washed daily for 7 days with 0.1 M PBS (pH 7.5). Tissue was cut transversely into four segments of equal length, snap-frozen in tissue-freezing medium (Triangle Biomedical Sciences, Durham, NC), and stored at –80°C.
Ten-micron cryosections were obtained perpendicular to the longitudinal axis of the vagina from the two central quarterns, corresponding to the region where smooth muscle and neural contributions are most fully developed [6]. The block containing the distal-most portion of the vagina with external vulva was bisected longitudinally and sectioned along the saggital plane, which provides optimal visualization of vulvar innervation. Sections were thaw-mounted onto Fisherbrand Colorfrost/Plus precleaned slides (Fisher Scientific, Pittsburgh, PA). Sections from each block were collected in seven serial sets and air-dried.
One set of sections was stained with hematoxylin and eosin and coverslips were placed over them with Permount (Fisher Scientific, Fair Lawn, NJ). Additional sets were rinsed in PBS containing 0.3% Triton X-100 (PBST; Sigma, St. Louis, MO; pH 7.5) and incubated overnight at room temperature with antiserum against protein gene product 9.5 (PGP 9.5, 1: 200; rabbit polyclonal immunoglobulin G [IgG]; Biogenesis, Brentwood, NH), tyrosine hydroxylase (TH, 1:400; rabbit polyclonal IgG; Chemicon, Temecula, CA), vesicular acetylcholine transporter (VAChT, 1:400; goat polyclonal IgG; Chemicon), or calcitonin gene-related peptide (CGRP, 1: 800; rabbit polyclonal IgG; Chemicon). An additional set was incubated for 48 h at room temperature with antiserum against neurofilament 200 (RT-97, 1:400; rabbit polyclonal IgG; Sigma, St. Louis, MO). PGP 9.5 is a panneuronal marker that labels all intact axons, whereas TH, VAChT, and CGRP are markers for sympathetic, parasympathetic, and sensory nociceptor nerve fibers, respectively. RT-97 antiserum selectively marks larger-diameter myelinated primary sensory axons. Primary antibody binding was visualized with Cy3 goat anti-rabbit IgG (H+L) (1:400; Jackson, West Grove, PA) for PGP 9.5, CGRP, TH, and RT97; or with Cy3, donkey anti-goat IgG (H+L) (1:400; Jackson) for VAChT. All sections stained for neural antigens were concurrently double-stained with an antibody directed against 
-smooth muscle actin (-SMA, 1:300; mouse monoclonal IgG; Sigma), visualized with Cy2 donkey anti-mouse IgG (H+L) (1:100; Jackson). All incubations were carried out in a humidified chamber and slides had coverslips placed on them with Fluoromount G (Southern Biotechnology Associates, Inc., Birmingham, AL). Antibody specificities were confirmed by primary antibody omissions and by preadsorptions with excess antigen.
Quantitative Analyses
To provide an index of vaginal innervation density, the apparent percentage area occupied by axons immunoreactive for PGP 9.5, TH, VAChT, and CGRP was assessed. Digital microscopic images (Nikon Eclipse TE300 with an Optronix MagnaFire camera) were captured from two randomly selected sections from each of the two blocks containing the middle quarterns of the vagina. In each section, six randomly selected fields containing submucosal muscular/connective tissue were captured (combined area = 0.894 mm2, or 
17% of the entire area of each section analyzed); the epithelial layer was largely devoid of innervation and was therefore excluded. In each field, images were obtained using illumination for neural (Cy3 channel) and smooth muscle (fluorescein channel) fluorescence. The total area of the submucosal tissue compartment was measured planimetrically (analySIS v. 3.2; Soft Imaging System, Lakewood, CO). The apparent percentage area occupied by immunoreactive nerves was determined by threshold discrimination, and divided by the total area of the sampled field to provide an index of innervation density.
To determine whether changes in innervation density arise secondary to alterations in the amount of vaginal tissue, nerve density was normalized to vaginal tissue volume within the sampled segment. Submucosal muscular/connective tissue area was measured planimetrically from hematoxylin and eosin-stained transverse sections. Tissue volume for each of the two sampled blocks was then calculated by multiplying cross sectional area by the length of the tissue block. Innervation volume was computed by multiplying innervation density (% area) by tissue volume for the block. Values are expressed as the average of the measurements from the two central quarterns of the vagina.
To assess changes in vaginal smooth muscle content, images of -SMA immunostained sections were captured at low magnification and assembled as montages. The area of the muscularis layer was measured planimetrically from the two sample regions, multiplied by the block lengths, and the values were averaged. To assess whether smooth muscle cell composition of the muscularis was altered, eight fields from the muscular layer were selected randomly at high magnification, and the percentage area occupied by -SMA-ir cells was measured.
RT-97-ir fibers were assessed semiquantitatively in two saggital sections containing the external vulvar regions from each preparation. A blinded observer ranked nerve density using a 4-point scale of 0 (no innervation) to 3 (frequent innervation); this approach was used because the uneven distribution and relatively low abundance of these axons was not amenable to our other quantitative analytical methods.
All values are presented as the mean ± SEM. Statistical comparisons were made using one-way analysis of variance (SigmaStat for Windows, v 2.03; Jandel Scientific, San Rafael, CA) with posthoc comparisons by the Student-Neuman-Kuels test. Differences were considered significant when P 
0.05.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
In rats at the estrus stage of the estrous cycle, the vagina weighed 396 ± 18 mg and was 25.0 ± 0.00 mm in length (Table 1). Microscopy revealed a thickened luminal epithelium surrounded by submucosal connective tissue containing a distinct muscularis smooth muscle layer (Fig. 1a). The submucosal connective/smooth muscle tissue compartment volume of the two central quarterns averaged 33.2 ± 3.4 mm3, and the muscularis volume averaged 11.0 ± 1.1 mm3 (Table 1). -SMA-ir cells occupied 33.1% ± 3.8% of the area within sections through the muscularis.
|
|
PGP 9.5-ir fibers were moderately abundant throughout the connective tissue and muscularis (Fig. 2a), and were only rarely present within the epithelium. PGP 9.5-ir nerve fiber density was 0.57% ± 0.24% (Fig. 3a), corresponding to a total volume of 0.19 ± 0.01 mm3 (Fig. 3b).
|
|
TH-ir fibers were localized predominantly to blood vessels and muscularis smooth muscle (Fig. 2b). TH-ir fiber density was 0.19% ± 0.02% (Fig.3a), and nerve volume was 0.06 ± 0.01 mm3 (Fig. 3b).
VAChT-ir fibers were associated primarily with the vasculature and nonvascular smooth muscle (Fig. 2c). VAChT-ir nerve density was 0.19% ± 0.01% (Fig. 3a), corresponding to a volume of 0.06 ± 0.01 mm3 (Fig. 3b).
CGRP-ir fibers were distributed within connective tissue, blood vessels, and smooth muscle (Fig. 2d). Fibers were rarely present in the epithelium of the proximal vagina, but were sometimes encountered in the distal epithelium. CGRP-ir nerve fiber density was 0.27% ± 0.01% (Fig. 3a), and CGRP-ir nerve volume was 0.09 ± 0.01 mm3 (Fig. 3b).
RT-97-ir nerve fibers were observed almost exclusively in large bundles in the adventitia of the central quarterns of the vagina. Terminal axons were moderately abundant (Table 1) in the external vulval tissue in association with hair follicles and smooth muscle, and as free or specialized nerve endings immediately beneath and within the epithelium (Fig. 4).
|
Ovariectomized Rat Vaginal Morphology and Innervation
In rats 14 days after OVX, vaginal weight was approximately half that of rats in estrus (P < 0.001), and vaginal length was reduced by 25% (P < 0.001). The luminal epithelium was considerably atrophied relative to that of rats in estrus. However, the volume of the proximal and central submucosal tissue compartment was comparable to that of estrus rats (Table 1). The muscularis was somewhat thinner in OVX rats (Fig. 1b). Volume of the muscularis was decreased by 42% (P = 0.01), although the content of smooth muscle within the muscular layer was not significantly affected (Table 1).
The overall distribution pattern of PGP 9.5-ir fibers in OVX rats was comparable to that of rats in estrus (Figs. 3e). However, PGP 9.5-ir innervation density was increased by 59% (P < 0.001). Similarly, PGP 9.5-ir innervation volume was increased by 42% (P = 0.005).
In OVX rats, TH-ir fiber distribution was similar to that of rats in estrus (Fig. 2f), but density was increased by 70% (P < 0.001, Fig. 3a), and volume by 55% (P < 0.001, Fig. 3b). VAChT-ir fiber distribution in OVX rats was also unchanged (Fig. 2g), whereas innervation density increased by 93% (P < 0.001, Fig. 3a) and volume increased by 106% (P < 0.001, Fig. 3b).
In sections of OVX vaginas immunostained for CGRP-ir (Fig. 2h), nerve fiber density was increased by 84% (P = 0.008, Fig. 3a) and volume was 66% greater (P < 0.001, Fig. 3b) than in rats at estrus. The distribution and the number of RT-97-ir nerve fibers in vaginal and vulvar tissues appeared comparable to that of rats in estrus (Table 1).
Effects of Estrogen on Vaginal Morphology and Innervation
Following 1 wk of sustained estrogen administration, vaginal weight was increased by 52% relative to that of OVX rats (P = 0.006) but remained below that of rats in estrus (P = 0.004, Table 1). Vaginal length was greater than in OVX rats (P < 0.001) and comparable to that of rats in estrus (Table 1). Vaginal epithelium was hypertrophied after estrogen treatment, but to a lesser extent than rats in estrus. Submucosal tissue volumes were comparable to those of OVX rats and rats in estrus (Table 1). The muscularis thickness of estrogen-treated rats was similar to that of OVX rats (Fig 1c), and volume was comparable to that of OVX rats and less than in estrogen-treated rats (P = 0.004, Table 1). Muscularis smooth muscle cell content was not affected (Table 1).
Sustained estrogen administration decreased PGP 9.5-ir innervation (Fig. 2i) relative to that of OVX rats by 35% (P = 0.001, Fig. 3a), and reduced PGP 9.5-ir volume by 32% (P = 0.001, Fig. 3b), to levels that were comparable to those of rats in estrus. TH-ir innervation (Fig. 2j) was reduced with respect to density and total volume (both by 45%, P < 0.001) to levels that were not significantly different from those of rats in estrus. VAChT-ir innervation (Fig. 2k) was reduced by 43% with regard to density (Fig. 3a, P < 0.001) and 43% by volume (Fig. 3b, P < 0.001), to levels similar to those of rats at estrus.
CGRP-ir innervation after estrogen treatment (Fig. 2l) was reduced relative to that of OVX rats by 45% as assessed by density (Fig. 3a, P < 0.001) and by 44% as determined by volume (Fig. 3b, P < 0.001). Innervation densities and volumes in estrogen-treated rats were comparable to those of rats in estrus. Estrogen administration did not affect RT-97-ir nerve fiber distribution or density in the distal vaginal and vulvar tissues (Table 1).
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Sustained decreases in estrogen that follow natural or surgically induced menopause in humans are accompanied by a variety of symptoms that may reflect direct effects on vaginal tissue, or indirect effects mediated by alterations in innervation. A prominent feature of menopause in humans is atrophy of vaginal tissues, and our findings confirmed that vaginas of OVX rats weighed less and were shorter than those of rats in estrus. Low estrogen levels result in profound atrophy of the vaginal mucosa [4, 17], and while we did not directly quantify these changes, diminished mucosal content is likely to account for the bulk of the reduction in vaginal mass in OVX rats. In fact, overall submucosal tissue volume did not differ significantly in OVX, estrus, or estrogen-treated rats, leading to the conclusion that changes in the largely uninnervated mucosal epithelium account predominantly for differences in vaginal mass.
Vaginal tone and smooth muscle content both decline substantially in menopause [18], and our measurements reveal significant reductions in muscularis smooth muscle volume in the OVX rat. This is concordant with previous findings that uterine smooth muscle atrophies following ovariectomy [19], suggesting that reproductive hormones may govern smooth muscle mass throughout the female reproductive tract. However, there appear to be important differences. Exogenous estrogen reverses the gonadectomy-induced decrease in uterine smooth muscle mass [13]. In contrast, sustained estrogen administration did not increase vaginal smooth muscle mass, implying that reproductive hormones other than estrogen are responsible for regulating the size of this target. Therefore, estrogen apparently is capable of exerting direct trophic effects on some types of reproductive tract smooth muscle but not on others. If a similar relationship exists in human females, then estrogen alone may offer only limited effectiveness as a therapeutic agent in reversing postmenopausal vaginal smooth muscle atrophy and atony.
Estrogen Regulates Vaginal Autonomic Innervation
The present study provides evidence that, in addition to inducing vaginal tissue remodeling, reproductive hormones also regulate vaginal innervation. PGP 9.5 is constitutively expressed in all neural tissues and is therefore an optimal marker for visualizing intact axons within the reproductive tract [20]. Immunostaining for PGP 9.5 confirmed that the vagina receives substantial innervation [6]. However, the extent of this innervation varied as a function of hormonal status, with PGP 9.5-ir axon density increasing significantly in OVX rats relative to those in estrus. This increase cannot be ascribed simply to changes in target volume. Because smaller target volumes could give the impression of increased innervation despite constant nerve numbers, we corrected for differences in the submucosal tissue compartment size; the increase persisted even after this normalization. These findings are reminiscent of the decline in PGP 9.5-ir innervation of uterine smooth muscle during pregnancy and proestrus/estrus, which occurs through degeneration of terminal axons [14, 19, 21]. Hormonal regulation of smooth muscle innervation therefore appears to be a common theme within the rodent reproductive system.
There is compelling evidence that estrogen is the primary hormone mediating changes in vaginal innervation. Innervation density is low at estrus, which occurs immediately following the estrous cycle surge in estrogen titer. OVX renders plasma estrogen levels undetectable [13], leading to a marked increase in PGP 9.5-ir nerve density, and estrogen administration reduces vaginal innervation to a level quantitatively identical to that of estrus without affecting smooth muscle size. These findings reveal a reciprocal relationship between estrogen level and reproductive tract innervation, supporting the idea that sustained reductions in estrogen in menopause may lead to increased vaginal innervation.
The functional implications of increased vaginal innervation have not been investigated, but should reflect the functional roles of the axon populations affected. TH-ir innervation was relatively sparse in rats at estrus, but increased substantially when estrogen levels were reduced following OVX, reminiscent of changes seen previously in the uterus [13]. Sympathetic axons release norepinephrine, which elicits smooth muscle contraction, leading to vasoconstriction and increased vaginal tone [6, 9]. Hence, increased sympathetic nerve density under low-estrogen conditions may lead to decreased vaginal blood flow and therefore decreased lubrication. Consistent with increased vasoconstrictor innervation, vaginal blood flow and lubrication are reduced during menopause and after surgical OVX in humans and animal models [22, 23]. However, vaginal tone is generally believed to be decreased in menopause. While this would seem to contrast with the increased sympathetic innervation observed in this study, it is important to bear in mind that vaginal smooth muscle mass is also substantially decreased after OVX, which may offset any increase in tone gained by increased excitatory innervation. If human vaginal smooth muscle is comparably affected in menopause, then estrogen therapy may not only be ineffective in increasing smooth muscle size, but actually may be detrimental to vaginal tone by promoting degeneration of excitatory sympathetic innervation.
Changes in parasympathetic innervation may also influence vaginal blood flow and smooth muscle tone. VAChT-ir parasympathetic axons project from the paracervical ganglion [24], and represent cholinergic nerves that coexpress nitric oxide synthase and release nitric oxide, inducing potent relaxation of vaginal vascular and nonvascular smooth muscle [6]. Relative to estrus, OVX rats showed substantial increases in parasympathetic axon density. The proliferation of parasympathetic axons after OVX suggests a further mechanism by which vaginal smooth muscle tone may be affected in menopause. Increased nitrergic parasympathetic innervation may provide greater inhibitory input to vaginal smooth muscle, thus further reducing tone. However, increased nitrergic innervation to vaginal blood vessels may also antagonize the increased sympathetic innervation, thereby acting to reverse vasoconstriction elicited by sympathetic hyperinnervation. It is important to note, however, that estrogen suppresses vaginal nitrergic neurotransmission [25], which could offset increases in parasympathetic nerve density.
Estrogen Diminution May Contribute to Vaginal Pain Syndromes
Pain, burning, and itching are prominent symptoms in many menopausal women. In particular, decreased estrogen levels have been associated with dysesthetic vulvodynia, which is seen typically at the onset of menopause [26]. The present findings reveal that decreased reproductive hormone levels following OVX result in increases in a subset of vaginal sensory nerves. CGRP-ir fibers represent projections of sensory nociceptor neurons located in the lumbar dorsal root ganglia [6, 27], and the density of these fibers was increased in OVX rats. In contrast, fibers labeled by RT-97 represent mechanoreceptive afferents [28], and these did not appear to be affected. Nociceptor neurons are the primary pathway for sensations of pain, heat, and itching and mediate erythemic vasodilation (flare response), which is often evident in vulvodynia. Increased numbers of these nerves could therefore play an etiological role in dysesthetic vulvodynia. Moreover, sympathetic nerves are known to be important in the establishment of many pain syndromes, including vulvodynia [29]. Thus sympathetic hyperinnervation may further augment sensations of pain associated with vaginal sensory nociceptor hyperinnervation following reduced ovarian function.
There is strong evidence that this increase in sensory innervation may have behavioral consequences. Ovariectomized rats have been shown to display vaginal hypersensitivity, and this is reduced by systemic estrogen administration [30]. Because estrogen reduces the vaginal content of both sensory and sympathetic nerves, this may explain in part why estrogen is reported to be therapeutically effective in reducing the painful symptoms in dysesthetic vulvodynia [31].
The significance of increased sensory nociceptor innervation under low estrogen conditions is less clear with respect to another common form of vulvodynia, vulvar vestibulitis, which occurs in premenopausal women. However, vulvar vestibulitis has been linked to alterations in hormonal status, particularly with regard to oral contraceptive use [32], and analysis of vestibular tissue obtained from patients indicates that this is also hyperinnervated [33, 34], although there is disagreement as to whether this is by CGRP-ir fibers [35–37]. Together, observations from rodent models and human patients appear to lend strong weight to the idea that hormones can play important roles in the establishment of vaginal hypersensitivity, and that altered sensitivity involves the proliferation of sensory nerves within vaginal tissues.
Mechanisms of Estrogen-Mediated Vaginal Nerve Plasticity
Changes in vaginal sympathetic innervation appear to be consistent with prior findings that uterine myometrial sympathetic innervation undergoes estrogen-dependent cyclical remodeling, with sustained hyperinnervation occurring after OVX [13, 19, 21]. Explant studies show that this is due primarily to effects of estrogen on the myometrium rather than on the neuron [38], and appears to be dependent on target synthesis and release of brain-derived neurotrophic factor under high-estrogen conditions [39]. It is therefore possible that similar changes in neurotrophin production may underlie vaginal sympathetic nerve depletion. However, there appear to be important differences between vaginal and uterine neuroplasticity. In the uterus, neither sensory nociceptor nor parasympathetic nerve densities vary with estrous cycle stage or following estrogen administration [21, 40, 41], implying that different mechanisms may be operating in these two closely related tissues. A better understanding of factors regulating plasticity of vaginal innervation may provide a basis for more effective therapeutic strategies for addressing vaginal dysfunction.
|
FOOTNOTES |
|---|
2 Correspondence: Peter G. Smith, Mailstop 3051, Mental Retardation Research Center, University of Kansas Medical Center, 3901 Rainbow Blvd., Kansas City, KS 66160-7401. FAX: 913 588 5677; psmith{at}kumc.edu
Received: 29 April 2004.
First decision: 20 May 2004.
Accepted: 27 May 2004.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  R. H. Straub The Complex Role of Estrogens in Inflammation Endocr. Rev., August 1, 2007; 28(5): 521 - 574. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. A. Pessina, R. F. Hoyt Jr., I. Goldstein, and A. M. Traish Differential Effects of Estradiol, Progesterone, and Testosterone on Vaginal Structural Integrity Endocrinology, January 1, 2006; 147(1): 61 - 69. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
