Abstract
Metrial glands are normal structures located in the mesometrial triangle of the pregnant rat uterus from gestational day (GD) 8 through termination of pregnancy. Metrial glands are composed of a dynamic mixed cell population of granulated metrial gland (GMG) cells, endometrial stromal cells, trophoblasts, blood vessels, and fibroblasts. Collections of similar cells may be seen in association with pseudopregnancy and other hormonal disturbances. Granulated metrial gland cells are the hallmark cell of the metrial gland. They are bone-marrow-derived, perforin-positive, natural killer cells that proliferate in the pregnant uterus. Understanding the normal histogenesis of the metrial gland and recognizing the possible existence of GMG cells and a reactive metrial gland in the nonpregnant state are important when examining any uterine lesion that contains granulated cells. This report demonstrates that the cellular composition, morphology, and immunohistochemical staining profile of normal metrial glands are similar to reported granular cell neoplasms in rats and mice. The possibility of a non-neoplastic lesion involving the metrial gland should be considered when proliferative lesions involving granulated cells are observed in the uterus of mice and rats from nonclinical toxicity studies. Positive immunohistochemical staining for perforin and S100 would assist in the classification of such lesions as a reactive metrial gland or decidual reaction.
Keywords
Introduction
Metrial glands are nodular aggregates of heterogeneous tissue that develops in the mesometrial triangle in the uterine wall of rats and mice during pregnancy (Peel 1989). These glands are important in fetal growth and successful placentation (Greenwood et al. 2000; Stewart 1998). The glands are composed of a dynamic mixed cell population of granulated metrial gland (GMG) cells, spindle to decidualized endometrial stromal cells, trophoblasts, blood vessels, and fibroblasts. Collections of similar cells, except for trophoblasts, may be seen in the nonpregnant state in association with pseudopregnancy, other hormonal disturbances, chemical stimulation, and physical trauma (Finn and Keen 1963; Peel 1989). Granulated metrial gland cells, the hallmark cell of the metrial gland, are bone-marrow-derived, perforin-positive, natural killer cells that proliferate within the pregnant uterus. Granulated metrial gland cells are approximately 40 μm in diameter and have abundant distinct (2 μm in diameter) eosinophilic glycoprotein granules that are periodic acid-Schiff (PAS) stain positive and diastase resistant (Croy and Kiso 1993; Head et al. 1994; Peel et al. 1989). Granulated metrial glands are frequently binucleated, exhibit some mitotic activity, and can have a high rate of apoptosis.
Although the metrial gland has received attention in human endocrinology, immunology, and reproductive physiology literature, the metrial gland has rarely been mentioned in the veterinary or toxicologic pathology literature (Karbe et al. 2000; Nagatani et al. 2001; Yoshida 1997). The four most recent reports and characterizations of granular cell tumors of the rat and mouse genital tract (1) do not mention the GMG cell or metrial gland; (2) do not discuss the GMG as a possible cell involved in the proliferative lesion; and (3) do not entertain the possibility that proliferative lesions involving granular cells could be non-neoplastic (Courtney et al. 1992; Markovits et al. 2000; Nyska et al. 1991; Veit et al. 2008).
The normal metrial gland, with its dynamic and heterogeneous cell populations, has gross pathologic, histopathologic, and immunohistochemical features similar to those described for granular cell neoplasms in non-pregnant rats and mice. We propose, therefore, that when evaluating lesions containing granulated cells in the uterus of rats and mice on nonclinical toxicity study, a diagnosis of a reactive metrial gland or decidual reaction (if no GMG cells are present) be considered as a diagnosis. Furthermore, as developmental and reproductive toxicology (DART) studies become an integral part of the safety assessment profile for food additives, chemicals, and pharmaceuticals, veterinary pathologists should be familiar with the normal histogenesis and morphology of this gland, which is critical to the maintenance of pregnancy in rats and mice (U.S. EPA 1998; U.S. FDA 2000).
This paper describes the histogenesis, morphology, and immunohistochemical staining characteristics of the normal metrial gland in the pregnant rat. By so doing, we highlight the dynamic nature of the cell populations of the metrial gland during gestation. We also present the histopathologic and immunohistochemical characteristics of metrial gland tissue in rats and suggest that immunohistochemistry can be used to identify such tissue in the nonpregnant rat as a reactive metrial gland.
Materials and Methods
Animals
Twelve adult female Cr1:CD Sprague Dawley rats used for the study were kept under standard conditions and were fed LLC Certified Rodent LabDiet 5002 and water ad libitum.
Study Procedure
Females were placed with a male rat in a cage overnight. When females had a vaginal plug, the day was designated as gestational day (GD) 1 of pregnancy. One pregnant female was euthanized by CO2 inhalation on GD8 and each subsequent day continuing through GD20.
Histology
The uterine horns were fixed in 10% neutral buffered formalin for forty-eight hours and transferred to 70% ethanol at 4 °C. Sections of uterus were trimmed from each fetal implantation site and placed in a separate cassette. After dehydration and clearing, tissues were embedded in paraffin, sectioned at 5 μm, and stained using hematoxylin and eosin (H&E) or periodic acid-Schiff (PAS) reaction with diastase to detect GMG cells (Stewart and Peel 1980).
Immunohistochemistry
Immunohistochemical staining of selected sections of metrial gland were performed throughout gestation to characterize the staining profile of various cell populations of the gland. In particular, perforin was used to identify GMG cells and to differentiate the GMG cells from the decidualized endometrial stromal cells. Neuron-specific enolase (NSE) and S100 were selected since granular cell tumors of rats were reported to be variably positive for S100 and NSE (Markovits et al. 2000), and decidual cells of the human uterus are reported to be positive for S100 (Lee et al. 1995; Nakamura et al. 1989).
Staining for perforin was performed using the avidin-biotin complex (ABC) immunohistochemical technique. Tissue sections (5 μm) were deparaffinized in xylene and rehydrated in decreasing ethanol concentrations before washing in phosphate-buffered saline (PBS). Endogenous peroxidase was blocked by 3% hydrogen peroxide in methanol. Epitope retrieval was performed using citrate buffer (pH = 6.0) in a decloaker for three minutes at 120 °C at 18 psi. Tissues were incubated in rabbit anti-rat perforin at a concentration of 200 μg/mL for thirty minutes at room temperature (Torrey Pines Biolabs, East Orange, NJ). The ABC Rabbit Elite Vector kit (Vector Labs, Burlingame, CA) was used to add the secondary antibody. Specific binding for perforin was visualized using diaminobenzidine (DAB) as a chromogen, and sections were counterstained with Mayer’s hematoxylin. For negative controls, nonspecific rabbit IgG (catalog number A0423, Dako Corporation, Carpinteria, CA, USA) was used in place of the primary antibody at the same protein concentration. Positive control tissue consisted of normal rat spleen.
Immunohistochemical staining for S100 and NSE was performed using a streptavidin biotin system (LSAB™ System-HRP kit, catalog number K0679, Dako Corporation). The procedures were used as above for perforin, with the exception that epitope retrieval was performed using high-pH citrate buffer (pH = 9.0). For NSE, the primary antibody was mouse monoclonal anti-NSE at 1:100 dilution (catalog number M0873, lot 00033597, clone BBS/NC/Vl-H14, Dako Corporation). For S100, the primary antibody was polyclonal rabbit anti-S100 at 1:400 dilution (catalog number Z0311, lot 00036387, Dako Corporation). Positive control tissue for S100 and NSE was normal rat colon.
Ethics
All procedures were approved and carried out under all applicable sections of the final Rules of the Animal Welfare Act regulations (9CFR), and all animals were maintained in accordance with the NIH Guide for the Care and Use of Laboratory Animals and were approved by the Institutional Animal Care Committee of WIL Research Laboratories, LLC.
Results
Histology
The histogenesis of the metrial gland is summarized as follows. At GD 8, the GMG cells are distributed as isolated, individualized cells located in the mesometrial triangle. By GD 11, the GMG cells become clearly organized into angiocentric aggregates. From GD 11 through GD 15, the GMG cells increase in number, and there is vascular remodeling and increased vascularity within the gland. At GD 13 through 15, the metrial gland reaches maximum size. Thereafter, GMG cells undergo apoptosis, retreat to a peripheral location in the gland, and allow for interstitial and endovascular trophoblasts to invade into the central part of the metrial gland. By GD 20, the gland is still approximately 1 cm in diameter and composed of centralized trophoblasts with GMG cells located more prominently in a peripheral portion of the gland.
Gestational Days 8 through 10:
The metrial gland begins to form at GD 8 and consists primarily of a collection of loosely arranged fibroblasts, elongated endometrial stromal cells, and normal capillaries lined by flattened endothelium (Figure 1). Several of the stromal cells are slightly plump in early stages of decidualization. These early decidual cells have occasional mitotic figures and some degree of pleomorphism (Figure 2). Individualized GMG cells are widely and randomly scattered in the metrial gland. These GMGs in the metrial gland have small eosinophilic, intracytoplasmic granules (Figure 2).
Gestational Days 11 and 12:
The metrial gland continues to enlarge and may be grossly detectable at this stage (Figure 3). Many endometrial stromal cells are epithelioid, with moderate mitotic activity and multinucleated cells. Vascular remodeling is starting and characterized by capillaries lined by plump endothelial cells having expanded cytoplasm, enlarged nuclei, and occasional mitotic figures. There is an expansion of the GMG population in the metrial gland with an angiocentric distribution pattern (Figure 4). Binucleated GMG cells are occasionally seen, and the individual intracytoplasmic granules enlarge.
Gestational Days 13 through 15:
The metrial gland reaches maximum size at approximately 1 cm in diameter and is grossly visible at this stage (Figure 5). There is high concentration and number of GMG cells in an angiocentric distribution admixed with ovoid decidualized stromal cells (Figure 6). The GMG cells have large, eosinophilic glycoprotein granules and abundant clear cytoplasm. Endovascular trophoblasts are present lining vessels within the metrial glands adjacent to the decidua basalis (Figure 7). Other vessels in the metrial gland continue to be lined by plump, mitotically active endothelium. The decidualized endometrial stromal cells are ovoid with abundant basophilic, coarsely stippled cytoplasm, and exhibit mitotic activity and a moderate degree of cellular pleomorphism.
Gestational Days 16 through 20:
The metrial gland is grossly detectable as a 1-cm nodule in the mesometrial triangle (Figure 8). Granulated metrial gland cells, gradually diminished in number, are less concentrated around blood vessels and retreat to a lateral and peripheral position in the gland. Apoptotic GMG cells are not uncommon. As the GMG cells retreat, interstitial trophoblasts invade through the decidua into the metrial gland (Figure 8). These interstitial trophoblasts have abundant clear cytoplasm and are referred to as glycogen cells (Figure 9). The endometrial stromal cell population at the margins of the metrial gland are loosely arranged, and their epithelioid, densely stippled basophilic cytoplasm is apparent. By day 20, the gland is composed of a nodular cluster of interstitial trophoblasts surrounded by a well-vascularized stroma composed of epithelioid decidual cells and angiocentric GMG cells (Figure 10).
Immunohistochemistry
Perforin:
Within the metrial gland, GMG cells are strongly positive for perforin at all stages of gestation, compatible with their natural killer (NK) cell lineage (Figure 11). All other cell types in the metrial gland, including decidualized stromal cells, trophoblasts, fibroblasts, and blood vessels, are negative for perforin.
Neuron-specific Enolase:
The decidualized endometrial stromal cells and endovascular trophoblasts stain moderately positive for NSE. Diffuse positive staining is present within the cytoplasm of both cell types. Granulated metrial gland cells and interstitial trophoblasts are negative for NSE.
S100:
The decidualized endometrial stromal cells are strongly positive for S100 (Figure 12). Granulated metrial gland cells and all trophoblasts are negative for S100.
Discussion
The metrial gland in the rat and mouse is a normal anatomic structure visible histologically from GD 8 through the end of pregnancy. Development of the metrial gland is part of decidualization, which is a sequential process of growth and differentiation of uterine stromal cells, remodeling of the extracellular matrix and maternal vasculature, and differentiation of GMG cells (Greenwood et al. 2000). Decidual cells are ovoid to polygonal epithelioid endometrial stromal cells. The process of decidualization is under the control of estrogen. These decidual stromal cells, in concert with prolactin, stimulate the differentiation and proliferation of GMG cells (Ain, Tash, and Soares 2003; Stewart 1983; Stewart 1998). The GMG cells in turn stimulate placental angiogenesis and regulate (i.e., restrict) trophoblast invasion from the placenta (Ain, Canham, Soares 2003; Kaloglu and Bulut 2007). The dynamic changes in the metrial gland in this study based on histologic interpretation are consistent with those reported by others using a variety of investigational tools (Delgado et al. 1996; Fukazawa et al. 1998; Kaloglu and Bulut 2007; Kokubu et al. 2005; Wang et al. 2003).
Granulated metrial gland cells are bone-marrow-derived NK cells characterized by their large glycoprotein-containing cytoplasmic granules. Mouse GMG cells are 30–35 μm with 2- to 4-μm granules. Rat GMG cells are slightly smaller, at 25 to 30 μm with 1- to 3-μm granules (Stewart 1998) Like all NK cells, they produce perforin, granzyme, and serine esterase (Stewart, Peel, and Wester 1996; Parr, et al. 1990). They encircle newly formed blood vessels in the mesometrial triangle, where they produce proteases that destroy basement membranes and vascular endothelial growth factor to stimulate endothelial cell proliferation (Kaloglu and Bulut 2007; Wang et al. 2003). The end effect is to promote angiogenesis and remodel the uterine vasculature at this uterine entry point of blood vessels (Charnock-Jones, Kaufman, and Mayhew 2004).
Development of the metrial gland is not limited to the pregnant uterus, or to the female animal, for that matter. Decidual-type reactions have been reported in the uterus during pseudopregnancy or exogenous estrogen administration (diethylstilbestrol or 17βestradiol) in the mouse (Dunn et al. 1963; Highman et al. 1980; Murphy 1961), and with physical or chemical stimulation (Finn and Keen 1963). These reactions have also been reported in the seminal vesicle and urinary bladder wall of aging male mice (Karbe 1987; Karbe et al. 2000). When it develops at times other than normal pregnancy, it is referred to as a decidual reaction. We suggest that a decidual reaction be termed a reactive metrial gland, when the reaction includes both decidualized stromal cells and GMG cells. When GMG cells are not present, decidual reaction would be a more appropriate term. Endocrine disrupters could conceivably lead to a reactive metrial gland in nonpregnant animals, since uterine NK cells are scattered throughout the nonpregnant endometrium and are particularly prominent in the progesterone-treated endometrium (Bulmer and Lash 2005).
The morphologic, histologic, and immunohistochemical characteristics of the normal metrial gland are similar to those reported for benign and malignant granular cell tumors of the mouse and rat uterus, and to decidual reactions of the male mouse accessory sex glands (Courtney et al. 1992; Halliwell 1998; Kaspareit and Deerberg 1987; Markovits et al. 2000; Veit et al. 2008). Granular cell tumors (GCT) of aging nonpregnant mice and rats are described as small nodules up to 1 cm in diameter, are commonly found within the myometrium extending to the adventitia, vary histologically from a collection of granular cells to a mixture of granular cells and spindle shaped cells, have multinucleated pleomorphic cells, and contain individual cell necrosis. The granular cell population is diastase resistant PAS positive, and areas of the reported granular cell tumors are variably positive for Leu 7, NSE, S100, desmin, vimentin, F4/80, cytokeratin, and smooth muscle actin (Courtney et al. 1992; Markovits et al. 2000; Veit et al. 2008).
Likewise, the normal metrial gland is composed of multiple cell types with a similar immunohistochemical staining profile. The metrial gland is a collection of pleomorphic granulated cells (which could be GMG cells and/or decidualized stromal cells) and spindle-shaped cells (which could be smooth muscle cells, fibroblasts, or endometrial stromal cells) with multinucleated cells and individual cell necrosis. These various cellular components result in a staining profile similar to that reported for granular cell tumors, as follows: decidualized endometrial stromal cells (positive for NSE, S100), smooth-muscle fibers (positive for desmin and actin), fibroblasts (positive for vimentin), NK cells (positive for Leu 7), macrophages (positive for F4/80), and lymphocytes (positive for Leu 7) (Karbe et al. 2000). Since the histomorphologic and immunohistochemical staining profiles are similar between the granular cell tumors of aging rats and mice and the metrial glands of pregnant rats, a pathologist should consider a reactive metrial gland when confronted with a lesion of the nonpreganant mouse or rat uterus containing granulated cells. If the lesion contains only decidualized stromal cells, then a decidual reaction may be more appropriate.
To aid in the diagnosis of either a decidual reaction or reactive metrial gland, we recommend that perforin stain be used to definitively identify the hallmark GMG cell population. Staining with S100 should also be done to identify the decidualized stromal cell population involved in this reactive metrial gland. Desmin, actin, and Leu 7 could additionally be used to identify additional cell types included in the proliferative lesions. Once the lesion is shown to represent a proliferation of a heterogeneous cell population, neoplasm would be an unlikely diagnosis.
Prolactin-secreting pituitary tumors or estrogen-producing ovarian cysts in aging rats may be a coincidental finding or play a causative role in decidual reactions of the nonpregnant aging rat uterus. Estrogen has been implicated as the cause of prolactin secreting pituitary adenomas in the rat (Prysor-Jones Siverlight, and Jenkins 1983). Estrogen in aging rats could be (1) stimulating the endometrial stromal cells to decidualize; and/or (2) stimulating prolactin secretion from pituitary tumors, which then causes proliferation of the GMG cells. Clearly, the association, if any, between prolactin-secreting pituitary adenomas or estrogen-producing ovarian cysts and a reactive metrial gland or decidual reaction requires further study.