FSH receptor (FSHR) expression in human extragonadal reproductive tissues and the developing placenta, and the impact of its deletion on pregnancy in mice

Abstract

Expression and function of the follicle-stimulating hormone receptor (FSHR) in females were long thought to be limited to the ovary. Here, however, we identify extragonadal FSHR in both the human female reproductive tract and the placenta, and test its physiological relevance in mice. We show that in nonpregnant women FSHR is present on: endothelial cells of blood vessels in the endometrium, myometrium, and cervix; endometrial glands of the proliferative and secretory endometrium; cervical glands and the cervical stroma; and (at low levels) stromal cells and muscle fibers of the myometrium. In pregnant women, placental FSHR was detected as early as 8-10 wk of gestation and continued through term. It was expressed on: endothelial cells in fetal portions of the placenta and the umbilical cord; epithelial cells of the amnion; decidualized cells surrounding the maternal arteries in the maternal decidua; and the stromal cells and muscle fibers of the myometrium, with particularly strong expression at term. These findings suggest that FSHR expression is upregulated during decidualization and upregulated in myometrium as a function of pregnancy. The presence of FSHR in the placental vasculature suggests a role in placental angiogenesis. Analysis of genetically modified mice in which Fshr is lacking in fetal portions of the placenta revealed adverse effects on fetoplacental development. Our data further demonstrate FSHB and CGA mRNAs in placenta and uterus, consistent with potential local sources of FSH. Collectively, our data suggest heretofore unappreciated roles of extragonadal FSHR in female reproductive physiology.

Keywords: FSH receptor; cervix; follicle-stimulating hormone; gonadotropin; placenta; pregnancy; uterus.

FIG. 1

FSHR expression in human placenta and associated tissues at 38–40 wk gestation. Tissues were stained with antibody FSHR-323 IgG2a (brown) and counterstained with hematoxylin (blue). A and B) Chorionic villi (magnifications ×200 [A] and ×600 [B]). Labeled are endothelial cells of the villi vessels (V), the chorionic stromal core (S), and trophoblasts (arrowhead labeled T). C) Amnion-chorion interface, including the amnion, chorion, and maternal decidua (magnification ×200; inset ×600). Labeled are the amniotic epithelium (AE), amniotic fibroblasts (AF), maternal decidua (D), and endothelial cells of the maternal vessels (MV). D) Maternal decidua (magnification ×200; inset ×600). Labeled are decidua strongly stained for FSHR (D₁), decidua moderately stained for FSHR (D₂), chorionic villi (C), and nonspecific staining of neutrophils (N). Corresponding negative controls stained with nonimmune IgG2a at the same concentration are shown in Supplemental Figure S2.

FIG. 2

FSHR expression in human umbilical cord at 38–40 wk gestation. Tissues were stained with antibody FSHR-323 IgG2a (brown) and counterstained with hematoxylin (blue). A) Umbilical artery (magnification ×100; inset ×200). Labeled are the endothelium (UAE), the inner layer of smooth muscle (M₁), and the outer layer of smooth muscle (M₂). B) Umbilical vein (magnification ×100; inset ×200). Labeled are the tunica intima (TI), tunica media (TM), and Wharton jelly (WJ). C) Cord amnion (magnification ×200; inset ×600). Labeled are the amniotic epithelium (AE) and Wharton jelly (WJ). Corresponding negative controls stained with nonimmune IgG2a at the same concentration are shown in Supplemental Figure S3.

FIG. 3

FSHR expression in human placenta and associated tissues at 28–30 wk gestation. Tissues were stained with antibody FSHR-323 IgG2a (brown) and counterstained with hematoxylin (blue). A) Chorionic villi (magnification ×100). Labeled are the endothelial cells of villi vessels (V), the chorionic stromal core (S), and trophoblasts (arrowhead labeled T). B) Amnion-chorion interface (magnification ×200). Labeled are the amniotic epithelium (AE), amniotic fibroblasts (AF), chorionic villi (C), and trophoblasts of the chorionic villi (arrowhead labeled T). C) Maternal decidua (magnification ×200). Labeled are decidua strongly stained for FSHR (D₁), decidua moderately stained for FSHR (D₂), endothelial cells of the maternal vessels (MV), and chorionic villi (C). Corresponding negative controls stained with nonimmune IgG2a at the same concentration are shown in Supplemental Figure S4.

FIG. 4

FSHR expression in human placenta and associated tissues at 18–20 wk gestation. Tissues were stained with antibody FSHR-323 IgG2a (brown) and counterstained with hematoxylin (blue). A) Chorionic villi (magnification ×200). Labeled are endothelial cells of the developing villi vessels (V), the chorionic stromal core (S), and trophoblasts (arrowhead labeled T). B) Amnion-chorion interface (magnification ×200). Labeled are the amniotic epithelium (AE) and amniotic fibroblasts (AF). C) Umbilical artery (magnification ×100; inset ×200). Labeled are the umbilical artery epithelium (UAE), smooth muscle layers (M), and Wharton jelly (WJ). D) Umbilical vein (magnification ×100; inset ×200). Labeled are the endothelial tunica intima (TI), the tunica media (TM), and Wharton jelly (WJ). Corresponding negative controls stained with nonimmune IgG2a at the same concentration are shown in Supplemental Figure S5.

FIG. 5

FSHR expression in human decidua at 18–20 wk gestation. Tissues were stained with antibody FSHR-323 IgG2a (brown) and counterstained with hematoxylin (blue). Shown in A–D is the maternal decidua. A) A tiling of adjacent images taken at magnification ×100. B) Magnified image (magnification ×200) of the area enclosed by the rectangle in A. C) A magnified image (magnification ×600) of the area enclosed by the lower rectangle in B. D) A magnified image (magnification ×600) of the area enclosed by the upper rectangle in B. Labeled are decidual cells with relatively high FSHR expression (D₁), decidual cells with lower FSHR expression (D₂), chorionic villi (C), extravillous trophoblasts (EVT), and maternal vessels (MV). Arrows indicate maternal vessels that are positive for FSHR expression, and arrowheads indicate MVs that are negative for FSHR expression. Negative controls stained with nonimmune IgG2a at the same concentration are shown in Supplemental Figure S6.

FIG. 6

FSHR expression in human placenta and associated tissues at 8–10 wk gestation. Tissues were stained with antibody FSHR-323 IgG2a (brown) and counterstained with hematoxylin (blue). A) Chorionic villi (magnification ×200). Labeled are endothelial cells of developing chorionic villi vessels (V), the chorionic stromal core (S), and trophoblasts (arrowheads labeled T). B and C) Maternal decidua (magnifications ×100 [B] and ×200 [C]). Labeled are decidua (D), maternal vessels (MV), maternal lacunae (ML), and chorionic villi (C). Extravillous trophoblasts are labeled EVT in C. Corresponding negative controls stained with nonimmune IgG2a at the same concentration are shown in Supplemental Figure S7.

FIG. 7

FSHR expression in the labyrinth layer of mouse placenta. The labyrinth layer of placentae (15 dpc) of littermate wild-type (A) and Fshr^−/− (B) mice were stained with anti-rat FSHR antibody. Images are shown at magnification ×600. Labeled are endothelial cells of mouse placental blood vessels (arrow labeled V) and mouse placental giant cells (arrowhead labeled GC). Corresponding negative controls stained with preimmune serum at the same concentration are shown in Supplemental Figure S8.

FIG. 8

Fshr genotype-specific differences in fetoplacental development among littermates. Fshr^+/− females were mated with Fshr^−/− males, and at 14–16 dpc, fetoplacental units were dissected and analyzed as described in Materials and Methods. A) Quantification of placental weight as a function of fetal Fshr genotype. Data are expressed as a percentage of the mean weight of littermate Fshr^+/− placentae. B) Quantification of fetal weight as a function of fetal Fshr genotype. Data are expressed as a percentage of the mean weight of littermate Fshr^+/− fetuses. Data shown in A and B are the mean ± SEM for 54 pups from 9 litters. Asterisks denote a difference of P < 0.05. C) Percentage of fetuses of each Fshr genotype.

FIG. 9

FSHR expression in human nonpregnant reproductive tissues. Tissues were stained with antibody FSHR-323 IgG2a (brown) and counterstained with hematoxylin (blue). A) Proliferative endometrium (magnification ×100). Labeled are glandular epithelium (G), endometrial stroma (S), and endothelial cells of an endometrial blood vessel (V). B) Secretory endometrium (magnification ×100). Labeled are glandular epithelium (G), endometrial stroma (S), and endothelial cells of the endometrial blood vessels (V). C and D) Cervix (magnification ×100). Labeled are cervical glands (G), cervical muscle (M), endothelial cells of cervical blood vessels (V), and arterial smooth muscle (AM). Corresponding negative controls stained with nonimmune IgG2a at the same concentration are shown in Supplemental Figure S9.

FIG. 10

Differential expression of FSHR in human myometrial stroma of nonpregnant versus pregnant (term) uterus. Tissues in A (nonpregnant myometrium) and B (pregnant myometrium at 38–40 wk gestation) were stained with antibody FSHR-323 IgG2a (brown) and counterstained with hematoxylin (blue) and are shown at magnification ×200. Labeled are myometrial muscle fibers (M with bracket), myometrial blood vessels (V), and arterial smooth muscle (AM). Images in A and B are representative of samples from three nonpregnant women and three pregnant (term) women, respectively. Corresponding negative controls stained with nonimmune IgG2a at the same concentration are shown in Supplemental Figure S10. C) Quantitation of FSHR expression in muscle (excluding that in blood vessels) from all of the nonpregnant and pregnant (term) samples, based on immunohistochemical staining of all slides in the same experiment and quantification as described in Materials and Methods. Data shown are the mean ± SEM. Asterisk denotes a difference of P < 0.05.

FIG. 11

FSHR mRNA in human term myometrium. RNAs from human myometrium from one patient at term pregnancy and from pooled human ovaries were examined for FSHR mRNA transcripts as described in Materials and Methods. FSHR mRNA from exons 2–3 (A), exons 1–4 (B), exons 4–7 (C), and exons 8–10 (D) indicated the presence of full-length FSHR mRNA and the absence of FSHR mRNA splice variants lacking exons 2, 6, or 9, respectively, in myometrium. Each panel represents data from one gel. In a given panel, spaces between images represent where one or more lanes containing irrelevant data were excised from the figure.

FIG. 12

Messenger RNAs encoding FSH subunits are present in extragonadal reproductive tissues. Polymerase chain reaction was used to amplify CGA mRNA encoding the glycoprotein α subunit (A) and FSHB mRNA encoding the β subunit of FSH (B) in human tissues, as described in Materials and Methods. Human tissues from term pregnancy included uteroplacental tissue (U-P), placenta (P; maternal decidua and amnion removed), uterine decidua (Dec), and uterine myometrium (Myo). The HUVECs (H) were also analyzed. Human pituitary (Pit) was included as a positive control. Each panel represents data from one gel. In a given panel, spaces between images represent where one or more lanes containing irrelevant data were excised from the figure.