Oestrogen receptors and their relation to neural receptive tissue of the labia minora

Authors

  • Nieves Martin-Alguacil,

    1. Department of Neurobiology and Behaviour, The Rockefeller University, New York, NY,
    2. Department of Anatomy and Embryology, School of Veterinary Medicine, Universidad Complutense de Madrid, Spain
    Search for more papers by this author
  • Donald W. Pfaff,

    1. Department of Neurobiology and Behaviour, The Rockefeller University, New York, NY,
    Search for more papers by this author
  • Lee-Ming Kow,

    1. Department of Neurobiology and Behaviour, The Rockefeller University, New York, NY,
    Search for more papers by this author
  • Justine M. Schober

    Corresponding author
    1. Department of Neurobiology and Behaviour, The Rockefeller University, New York, NY,
    2. Hamot Medical Center, Erie, PA, USA, and
      Justine Schober, 333 State Street, Suite 201, Erie, PA 16507, USA.
      e-mail: Schobermd@aol.com
    Search for more papers by this author

Justine Schober, 333 State Street, Suite 201, Erie, PA 16507, USA.
e-mail: Schobermd@aol.com

Abstract

Associate Editor

Michael G. Wyllie

Editorial Board

Ian Eardley, UK

Jean Fourcroy, USA

Sidney Glina, Brazil

Julia Heiman, USA

Chris McMahon, Australia

Bob Millar, UK

Alvaro Morales, Canada

Michael Perelman, USA

Marcel Waldinger, Netherlands

OBJECTIVE

To assess the cellular distributions of oestrogen receptors α and β (ERα and ERβ) and neuronal nitric oxide synthase (nNOS) in the labia minora, as knowledge about ER type and function may clarify the role of oestrogens in vaginal scar formation and improve outcomes in female genital surgery.

SUBJECTS AND METHODS

Labial samples were taken from 10 girls (aged 2–9 years) who underwent surgery for labial fusion. The waste tissue strips obtained were used for immunohistochemical identification of ERα and ERβ, and nNOS in the labia minora.

RESULTS

There was ERα nuclear staining in the stroma of the labia minora close to the clitoris, and basal and suprabasal in the epidermal cells membrane restricted to superficial sections of the labia minora. ERβ was found in the stroma of the labia minora closer to the clitoris and in superficial sections, in the basal epidermal cells membrane and apocrine glandular epithelial cells membrane. There was also ERβ cell membrane staining in the basal and suprabasal epithelial cells and fibroblasts in the lamina propria.

CONCLUSIONS

Established ER presence allows the consideration of the introitus of the vagina as a target for oestrogen therapy in various clinical and surgical situations. Continuing elucidation of the immunohistochemistry of this external genital tissue might assist in the development of molecular tools to treat genital abnormalities. Details of this immunohistochemistry may also advance the understanding of the effects of sexual differentiation on the brain and other organ systems.

INTRODUCTION

Hormones influence the growth, development, differentiation and sensation of the genitalia. Studies of the actions of hormones and their receptors suggest that female phenotypic differentiation of the external genitalia is an active process, which is (maternal) oestrogen dependent [1]. Oestrogen deprivation results in atrophic changes in the vaginal mucosa characterized by thinning of the epithelium, loss of vaginal rugae, and reduction in vaginal lubrication. It is well-known that the sensory condition of the vulvar epithelium is highly influenced by oestrogen status with enlargement of the genito-sensory field in response to this hormone [2,3]. Fusion of the labia in girls suggests oestrogen dependence of the vaginal introitus because of its resolution with oestrogen therapy or the oestrogen effects of puberty. Oestrogen accelerates cutaneous wound healing processes associated with enhanced matrix deposition, rapid epithelialization and dampening of the inflammatory response [4–8].

Previous studies have shown the presence of oestrogen receptors (ERs) in the vaginal walls, as well as in the uterosacral ligaments [9,10]. Regarding the ER, variations in the expression of two isoforms (ERα and ERβ) in the same tissue or organ and different physiological responses have been indicators of a variable target tissue response to the same oestrogen ligand. Further, ERα and ERβ differ in their actions within different tissues [11].

Understanding the locations within genital structures, particularly in the area of the labial minora, of the different isoforms of the ER may elucidate mechanisms of genital growth and development. The labia minora are said to be the female homologue of male genital structures that undergo ventral folding and fusion to form the penile urethra and corpus spongiosum [12–14]. If this is indeed true, failure or interruption in the process of labial fusion in the normal female fetus would create the introital opening of the vagina. This may be due in part to the lack of fetal androgen production and low 5α-reductase activity, in part to maternal oestrogen stimulation of ER-positive urethral folds causing the labia minora to diverge laterally [1].

The present study used tissue of the labia minora, obtained during surgical separation of complete fusion, to assess the differential cellular distributions of ERα, ERβ, and neuronal nitric oxide synthase (nNOS).

SUBJECTS AND METHODS

The labial samples were taken from 10 girls aged 2–9 years who underwent surgery of the vagina for labial fusion. All specimens were obtained by the senior investigator (J.M.S.), a paediatric urologist, after informed consent had been obtained from the parent or legal guardian. The specimens were waste tissue strips obtained at surgical separation of the labial fusion. Each specimen was marked with a suture on the posterior end of the strip. This study was approved by the Institutional Review Board of the hospital where the surgical procedures were performed.

The tissue was fixed in 4% paraformaldehyde, stored in 0.1 m phosphate buffer and 10% sucrose for 24 h, frozen and cut into 30 µm thick sections on a cryostat. The sections were washed twice for 5 min in PBS (pH 7.4), incubated in blocking solution (10% methanol, 3% H2O2 in PBS) for 10 min, rinsed twice in PBS followed by the gelatin-blocking step (0.75% gelatin in PBS) for 10 min and rinsed twice in PBS. Matching series of sections from each specimen were incubated for 48 h at 4 °C, with the following antibodies in a humidified chamber: rabbit anti-ERα 1 µg/mL (Upstate, cell signalling solutions, Charlottesville, VA, USA), rabbit anti-ERβ 1 µg/mL (Zymed Laboratories Inc., San Francisco, CA, USA), and rabbit anti-nNOS 2 µg/mL (Zymed Laboratories Inc.). After incubation in primary antibody, the sections were then processed according to a standard procedure for the rabbit IgG Vectastain ABC Kit (Vector Laboratories, Burlingame, CA, USA). Sections were then placed with agitation in two 5-min washes of PBS (pH 7.4) followed by a 5-min wash in 0.05 m Tris buffer, pH 7.6 (Sigma Chemicals Corp., St. Louis, MO, USA). The sections were then immersed in 0.05% 3,3′-diaminobenzidine tetrahydrochloride (Sigma) with 0.03% H2O2 in 0.05 m Tris buffer (pH 7.6) and reacted for 10 min, rinsed with 0.05 m Tris buffer followed by distilled water. The reaction product appeared as a dark brown stain. In the negative control primary antibody was absorbed by pre-incubation with the respective synthetic peptide.

One section every 10 sections was lightly counterstained with cresyl violet. One complete set of sections was stained with a slightly modified Bielschowsky’s silver stain [15]. All sections were dehydrated through graded alcohols, cleared in xylene and coverslips applied with Permount.

RESULTS

ERα nuclear staining was present in the stroma of the labia minora but was only detected in a restricted area closer to the clitoris (Fig. 1A). There was also basal and suprabasal ERα staining in the epidermal cells membrane but restricted to superficial sections of the labia minora (Fig. 2A).

Figure 1.

A, ERα nuclear staining in the stroma of the labia minora, ×20. B, ERβ cell membrane staining in the basal and suprabasal epithelial cells, and fibroblasts in the lamina propria, ×4.

Figure 2.

A, ERα staining was also present in the basal and suprabasal epidermal cells membrane but restricted to superficial sections of the labia minora, ×20. B, In superficial sections ERβ was located in the basal epidermal cells membrane and apocrine glandular epithelial cells membrane, ×20.

This contrasted with the distribution of ERβ; few cells were nuclear stained for ERβ in the stroma of the labia minora closer to the clitoris. There was cell membrane ERβ staining in the basal and suprabasal epithelial cells and fibroblasts in the lamina propria (Fig. 1B). In superficial sections, ERβ was located by stain in the basal epidermal cells membrane and apocrine glandular epithelial cells membrane (Fig. 2B).

Figure 3 shows nerve bundles and fibres in the stroma of the labia using a modified Bielschowsky’s silver staining technique (Fig. 3A) and nNOS immunostaining (Fig. 3B).

Figure 3.

A, Nerve bundles and fibres in the stroma of the labia shown with Bielschowsky’s modified silver staining technique. B, Nerve bundles and fibres detected with nNOS immunostaining, ×20.

A distinct distribution of nNOS immunostaining was identified in the stroma of the labia. The immunoreactivity was diffusedly distributed in the labia corresponding with nerve bundles and fibres located in this region. These results correspond with the pattern seen with those sections prepared with the Bielschowsky’s modified silver staining technique. Nuclear immunostaining was detected in stromal cells of the labia (Fig. 3).

For each figure, negative control experiments in which the complete immunocytochemical procedure was carried out but omitting the primary antibody, failed to show staining, thus showing the specificity of the positive results (data not shown).

DISCUSSION

The cellular distribution of ERs in pubic skin of adult women have previously been studied by saturation analysis using isotopically labelled oestradiol. No significant differences between premenopausal and menopausal skin was detected [16]. ERs were seen in the basal and suprabasal cells of the vaginal epithelium and the epidermis of labia minora but were restricted to the basal keratinocytes in true skin. They were seen in stromal fibroblasts and vaginal smooth muscle, and dermal fibroblasts of the skin. ERs levels were highest in the vaginal epithelium and stroma, and lowest in suprapubic skin [17]. In this study, the transition from vagina to vulva was marked by a decrease in ERs. This distribution of ERs would indicate a limited role for oestrogen creams on the vulva [17].

Considering the present data, one might anticipate a greater oestrogen effect when administered vaginally, compared with extravaginal administration. The differential therapeutic effects for oestrogens when administered under different background physiological conditions of the oestrous or menstrual cycle have yet to be studied.

ERs were not seen in skin blood appendages except for occasional apocrine sweat glands in which luminal epithelial cells stained strongly. ERs were not seen in skin blood vessels or lymphatics [17].

Oestrogen regulates cell proliferation and differentiation in vaginal epithelium; this is reflected by the abundance of ERs in the basal and suprabasal compartments. The distribution of ERs in labial tissue in the present study is in agreement with that reported by Press et al. [18] for vaginal tissue. The abundance of ERs in vaginal stromal cells suggests that stromal–epithelial interactions could be involved in oestrogen regulation of the vaginal epithelium during adult life. Intense ER staining is also seen in the epidermis and dermis of the labia minora [17]. The study of Kalloo et al. [1] also noted intense ER staining in the primordia of the inner labia minora in female embryos, so that the potential exists for the direct action of oestrogen on these tissues from mid-gestation into adult life. Oestrogens may modulate the behaviour of epidermal keratinocytes, especially in genital skin. In this context it is of interest that oestrogens have recently been shown in other tissues to regulate synthesis of interleukin-6 [19], a key pro-inflammatory cytokine produced by epidermal keratinocytes [20,21].

Inactivation of the ER gene appears to have had no significant effect on the embryonic development of female external genitalia [22]. However, there is extensive circumstantial evidence that oestradiol-17β influences the postnatal physiology of extra-genital and especially, genital (vulval) skin in the human female [23].

The effects of oestrogen in the treatment of human skin thickness [24], collagen content [25] and blood flow [26] have been reported, but it was unclear whether these represent direct actions of oestrogens on the skin or systematic effects.

The results of using immunohistochemistry to assess the cell distribution and protein expression of ERα and ERβ are shown in Figs 1 and 2. ERα was distributed in both the epithelium and stroma, with the greatest density in the parabasal cells. ERβ had a cellular distribution similar to that of ERα but was not nearly as abundant or prominent. Although nuclear staining was present for ERβ, there was a great deal of cytoplasmic and interstitial staining, which yielded nonspecific results and prevented accurate interpretation [27].

In the human corpus cavernosum and corpus spongiosum smooth muscle was immunoreactive for the ERα and strongly reactive for ERβ. Endothelial cells were sporadically positive for ERα, and positive for ERβ. Urethral epithelium showed nuclear expression of ERα in the intermediate cells. ERβ was highly expressed in almost all urethral nuclei and, much more weakly in cytoplasm [28].

ER signals localized to the mesenchyme and subepithelia are in coincidence with the studies on the ER expression pattern in developing reproductive tissues in the female rat [29]. This pattern is also in good agreement with the proposed concept on mesenchyme and epithelia interaction [30]. Urogenital mesenchyme under the influence of various factors, most notably oestrogen and testosterone, is thought to promote epithelial morphogenesis, proliferation, and differentiation and evoke expression of tissue-specific secretory proteins.

In the human vagina, only ERα could be detected: there was a nuclear reaction in deep layers of the stratified epithelium as well as in stromal and muscle cells [31]. In canine vaginal and vulvar tissue, there was nuclear staining for ERα in surface epithelium, stromal and smooth muscle cells. Receptors were also expressed in vulvar skin. No cyclic changes in ER immunostaining were observed [32].

Mice with genetic disruptions of ERα and ERβ display different phenotypes, showing that each receptor has a distinct action [11,33]. Moreover, microarray analysis in mice with deletions of ERα or ERβ showed that ERβ inhibited the transcription of 240 oestrogen-responsive genes by 46%[34]. Thus, the relative levels of expression of these two ER isoforms will affect the cellular response to oestrogens [35,36].

In vulvar skin there was ERα staining in the nuclei of apocrine sweat glands. Hair follicles sometimes expressed very weak nuclear staining for the progesterone receptor. Striated muscle cells that were present in some vulvar sections always stained negative for ERα. In the vaginal wall, there was ERα staining in the nuclei of epithelium cells, stromal cells of the mucosa and smooth muscle cells of the muscular layer [32]. There was no staining for ERs in blood vessels in the vulvar or vaginal tissue. Nuclear staining for ERα was also noticed in small bundles of smooth muscle cells that were sparsely present in the vulvar tissue. When different cell groups of the vulva were compared, it was observed that the stromal cells always stained with the highest intensity and the mucosal epithelium cells with the lowest intensity. In the vagina, there was higher staining for ERα in the epithelial and smooth muscle cells compared with the vulvar tissue.

Immunostaining for ERα in human vaginal and vulvar tissues was exclusively nuclear [17] and is also predominantly nuclear in the dog [32]. ERα in the canine vagina was located in the epithelium, fibroblasts and smooth muscle cells [32], as in the human [17] and mouse [33]. In the epithelium of the vulvar skin of the dog there was a weak staining of ERα occasionally. ERα was detected in some apocrine sweat glands of canine vulvar skin [32].

Considering the differences between ERα and ERβ; studies on female ER-knockout mice showed that ERα seems to be the critical receptor for mediating vaginal responses to oestradiol-17β[37]. In ERα-knockout mice, histological analysis consistently indicated a complete lack of vaginal oestrogenization of exogenous oestradiol [33,38], despite the immunostaining of ERβ mRNA [37]. On the other hand, the vaginal mucosa of ERβ-knockout mice appears to undergo the normal cyclic changes associated with ovarian steroidogenesis, indicating that this is an ERα-mediated process [33].

Generally, there was higher staining intensity for ERα in stromal cells than in epithelial cells in canine vaginal and vulvar tissue [32]. In the vagina, oestradiol-17β plays a critical role in tissue growth and is obligatory for normal epithelia differentiation [17,37,38]. In the human, the abundance of ERs in vaginal stromal cells suggests that stromal–epithelial interactions could be involved in oestrogen regulation of the vaginal epithelium [17]. Differentiation of this epithelium is most probably elicited by paracrine mechanisms via stromal ERα[38], as indicated by several tissue-recombinant studies using tissues of wild-type and ERα-knockout mice [38,39]. That stromal cells stain with a higher intensity than epithelial cells for ERα has also been reported in the canine uterus [40]. These results suggest that stromal–epithelial interactions may also be of importance in the female genital tract of the dog, but further studies are required to support this idea.

Oestrogen status and oestrogen binding are critical factors in the thickness of the epithelium and smooth muscle walls, pH and composition of the flora, surface lubrication, sensitivity and local inflammatory responses in vulvar tissue. Because vulvar skin homeostasis as well as proliferation and differentiation are affected by oestrogen presence, this area is a target for consideration of drug intervention. Understanding the differential profiles of the various receptor ligands may help develop specific receptor modulators directing site-specific desired effects.

In conclusion, these findings are of clinical importance in the pathophysiology of age-associated and hormonally associated, female genital disorders that include both functional and structural changes. In addition, the understanding of ER status of vulvar tissue is critical to medical and surgical therapy considerations affecting the surface integrity of the vulvar epithelium. A more complete understanding of the molecular mechanisms and individual actions associated with each ER may allow discovery of tissue-selective drugs that maximize beneficial effects while limiting adverse effects of oestrogen therapy. Continuing elucidation of the immunohistochemistry of this external genital tissue may be a step toward the development of molecular tools to treat genital abnormalities. Details of this immunohistolochemistry may also advance the understanding of effects of sexual differentiation on the brain and other organ systems.

CONFLICT OF INTEREST

None declared. Source of funding: Hess Roth Foundation, Rockefeller University.

Ancillary