Progesterone regulates FGF10, MET, IGFBP1, and IGFBP3 in the endometrium of the ovine uterus

Progesterone (P4) is unequivocally required to maintain a uterine environment conducive to pregnancy. This study investigated the effects of P4 treatment on expression of selected growth factors (fibroblast growth factor 7 [FGF7], FGF10, hepatocyte growth factor [HGF], and insulin-like growth factors [IGF1 and IGF2]), their receptors (MET, FGFR2(IIIB), and IGF1R), and IGF binding proteins (IGFBPs) in the ovine uterus. Ewes received daily injections of corn oil vehicle (CO) or 25 mg of P4 in vehicle from 36 h after mating (Day 0) to hysterectomy on Day 9 or Day 12. Another group received P4 to Day 8 and 75 mg of mifepristone (RU486, a P4 receptor antagonist) from Day 8 through Day 12. Endometrial FGF10 mRNA levels increased between Day 9 and Day 12 and in response to P4 on Day 9 in CO-treated ewes, which had larger blastocysts, and were substantially reduced in P4+RU486-treated ewes, which had no blastocysts on Day 12. Endometrial FGF7 or HGF mRNA levels were not affected by day or reduced by RU486 treatment, but MET mRNA levels were higher in P4-treated ewes on Day 9 and Day 12. Levels of IGF1, IGF2, and IGF1R mRNA in the endometria were not affected by early P4 treatment. Although stromal IGFBPs were unaffected by P4, levels of IGFBP1 and IGFBP3 mRNA in uterine luminal epithelia were increased substantially between Day 9 and Day 12 of pregnancy in CO-treated ewes and on Day 9 in early P4-treated ewes. Therefore, FGF10, MET, IGFBP1, and IGFBP3 are P4-regulated factors within the endometrium of the ovine uterus that have potential effects on endometrial function and peri-implantation blastocyst growth and development.     

Ovine osteopontin: I. Cloning and expression of messenger ribonucleic acid in the uterus during the periimplantation period.

Trophoblast-derived interferon tau (IFNtau) acts on the endometrium to increase secretion of several proteins during the pregnancy recognition period in ruminants. One of these is a 70-kDa acidic protein that has not been identified. Our hypothesis was that the 70-kDa acidic protein is osteopontin (OPN). OPN is an acidic glycoprotein that fragments upon freezing and thawing or treatment with proteases including thrombin. OPN contains a Gly-Arg-Gly-Asp-Ser (GRGDS) sequence that binds to cell surface integrins to promote cell-cell attachment and cell spreading. Using antisera to recombinant human OPN, both 70-kDa and 45-kDa proteins were identified in uterine flushings from pregnant ewes by Western blotting. A clone containing the entire ovine OPN cDNA coding sequence was isolated by screening a Day 15 pregnant ovine endometrial cDNA library with a partial ovine OPN cDNA. In pregnant ewes, steady-state levels of OPN endometrial mRNA increased (P < 0. 01) after Day 17. In both cyclic and pregnant ewes, in situ hybridization analysis showed that OPN mRNA was localized on unidentified immune cells within the stratum compactum of the endometrium. In pregnant ewes, OPN mRNA was also expressed by the glandular epithelium. Results suggest that progesterone and/or IFNtau induce expression and secretion of OPN by uterine glands during the periimplantation period and that OPN may induce adhesion between luminal epithelium and trophectoderm to facilitate superficial implantation.     

Select nutrients in the ovine uterine lumen. ii. glucose transporters in the uterus and peri-implantation conceptuses

Total glucose in ovine uterine lumenal fluid increases 6-fold between Days 10 and 15 of gestation, but not the estrous cycle; however, mechanisms for glucose transport into the uterine lumen and uptake by conceptuses (embryo/fetus and associated membranes) are not established. This study determined the effects of the estrous cycle, pregnancy, progesterone (P4), and interferon tau (IFNT) on expression of both facilitative (SLC2A1, SLC2A3, and SLC2A4) and sodium-dependent (SLC5A1 and SLC5A11) glucose transporters in ovine uterine endometria from Days 10 to 16 of the estrous cycle and Days 10 to 20 of pregnancy, as well as in conceptuses from Days 10 to 20 of pregnancy. The SLC2A1 and SLC5A1 mRNAs and proteins were most abundant in uterine luminal epithelia and superficial glandular epithelia (LE/sGE), whereas SLC2A4 was present in stromal cells and glandular epithelia (GE). SLC5A11 mRNA was most abundant in endometrial GE, whereas SLC2A3 mRNA was not detectable in endometria. SLC2A1, SLC2A3, SLC2A4, SLC5A1, and SLC5A11 were expressed in the trophectoderm and endoderm of conceptuses. Steady-state levels of SLC2A1, SLC5A1, and SLC5A11 mRNAs, but not SLC2A4 mRNA, were greater in endometria from pregnant than from cyclic ewes. Progesterone increased SLC2A1, SLC5A11, and SLC2A4 mRNAs in the LE/sGE and SLC5A1 in the GE of ovariectomized ewes. Expression of SLC5A1 was inhibited by ZK136,317 (progesterone receptor antagonist), and the combination of ZK136,317 and IFNT further decreased expression in GE. In constrast, P4 induced and IFNT stimulated expression of SLC2A1 and SLC5A11, and these effects were blocked by ZK136,317. Results of this study indicate differential expression of facilitative and sodium-dependent glucose transporters in ovine uteri and conceptuses for transport and uptake of glucose, and that P4 or P4 and IFNT regulate their expression during the peri-implantation period of pregnancy.     

Progesterone and placentation increase secreted phosphoprotein one (SPP1 or osteopontin) in uterine glands and stroma for histotrophic and hematotrophic support of ovine pregnancy

Secreted phosphoprotein one (SPP1, osteopontin) may regulate conceptus implantation and placentation. We investigated effects of progesterone (P(4)) and the conceptus on expression and localization of SPP1 in the ovine uterus. Steady-state levels of SPP1 mRNA in the endometrium of unilaterally pregnant ewes did not differ significantly between nongravid and gravid horns within their respective days of pregnancy; however, levels did increase as pregnancy progressed. SPP1 mRNA was detectable in the glandular epithelium (GE) of both nongravid and gravid horns via in situ hybridization. SPP1 protein was localized to the apical surface of the luminal epithelium of both nongravid and gravid uterine horns. Gravid horns exhibited extensive stromal SPP1 on Days 40 through 120, whereas SPP1 was markedly lower in the stroma of nongravid uterine horns through Day 80 of pregnancy. By Day 120, stromal expression of SPP1 between nongravid and gravid horns was similar. Long-term P(4) treatment of ovariectomized ewes induced SPP1 in the uterine stroma and GE. A bioactive 45-kDa SPP1 fragment was purified from uterine secretions and promoted ovine trophectoderm cell attachment in vitro. Interestingly, increased stromal cell expression of SPP1 was positively associated with vascularization as assessed by von Willebrand factor staining. Finally, ovine uterine artery endothelial cells produced SPP1 during outgrowth into three-dimensional collagen matrices in an in vitro model system that recapitulates angiogenesis. Collectively, P(4) induces and the conceptus further stimulates SPP1 in uterine GE and stroma, where SPP1 likely influences histotrophic and hematotrophic support of conceptus development.     

Endometrial HSD11B1 and cortisol regeneration in the ovine uterus: effects of pregnancy, interferon tau, and prostaglandins

In ruminants, the elongating conceptus secretes interferon tau (IFNT), the pregnancy recognition signal, and prostaglandins (PGs). Progesterone from the ovary induces prostaglandin synthase two (PTGS2) and hydroxysteroid (11-beta) dehydrogenase 1 (HSD11B1) in the endometrial epithelia, and PTGS2-derived PGs regulate endometrial functions and conceptus elongation. The enzyme HSD11B1 interconverts inactive cortisone and active cortisol. These studies determined the effects of pregnancy, IFNT, and PGs on endometrial HSD11B1 expression and activity in the ovine uterus. Study one found that HSD11B1 activity was present in both the endometrium and conceptus during early pregnancy. In study two, ewes received intrauterine infusions of vehicle as a control (CX) or meloxicam (MEL), a PTGS2 inhibitor, from Days 8 to 14 of pregnancy. Endometrial HSD11B1 activity and cortisol in the uterine lumen were substantially lower in MEL-infused ewes. In study three, cyclic ewes received intrauterine infusions of vehicle as a CX, MEL, recombinant ovine IFNT, or IFNT and MEL. Infusion of IFNT increased endometrial HSD11B1 expression and activity and cortisol in the uterine lumen, and this effect was diminished by coinfusion of MEL. In study four, cyclic ewes were infused with vehicle as a CX, IFNT, PGE2, PGF2 alpha, or PGI2. Infusion of all the PGs and IFNT increased endometrial HSD11B1 expression and activity, and IFNT and PGI2 infusion increased cortisol in the uterine lumen. These studies support the idea that IFNT and PGs from the conceptus regulate endometrial HSD11B1 expression and activity that regenerates bioactive cortisol in the ovine uterus during early pregnancy to influence endometrial functions and conceptus elongation.     

Differential regulation of gonadotropin subunit messenger ribonucleic acids by gonadotropin-releasing hormone pulse frequency in ewes

Changes in the frequency of GnRH and LH pulses have been shown to occur between the luteal and preovulatory periods in the ovine estrous cycle. We examined the effect of these different frequencies of GnRH pulses on pituitary concentrations of LH and FSH subunit mRNAs. Eighteen ovariectomized ewes were implanted with progesterone to eliminate endogenous GnRH release during the nonbreeding season. These animals then received 3 ng/kg body weight GnRH in frequencies of once every 4, 1, or 0.5 h for 4 days. These frequencies represent those observed during the luteal and follicular phases, and the preovulatory LH and FSH surge of the ovine estrous cycle, respectively. On day 4, the ewes were killed and their anterior pituitary glands were removed for measurements of pituitary LH, FSH, and their subunit mRNAs. Pituitary content of LH and FSH, as assessed by RIA, did not change (P greater than 0.10) in response to the three different GnRH pulse frequencies. However, subunit mRNA concentrations, assessed by solution hybridization assays and expressed as femtomoles per mg total RNA, did change as a result of different GnRH frequencies. alpha mRNA concentrations were higher (P less than 0.05) when the GnRH pulse frequency was 1/0.5 h and 1 h, whereas LH beta and FSH beta mRNA concentrations were maximal (P less than 0.05) only at a pulse frequency of 1/h. Additionally, pituitary LH and FSH secretory response to GnRH on day 4 was maximal (P = 0.05) when the pulse infusion was 1/h.(ABSTRACT TRUNCATED AT 250 WORDS)     

Cellular localization of an embryonic interferon, ovine trophoblastin and its mRNA in sheep embryos during early pregnancy.

The ovine embryo produces an interferon named ovine Trophoblastin (oTP) which is involved in the maternal recognition of pregnancy and ensures the maintenance of progesterone secretion by the corpus luteum. We have used indirect immunohistofluorescence and in situ hybridization on histological sections to investigate the fate of this protein and its mRNA in ovine embryos from days 3 to 25 of pregnancy. The level of expression was measured by image analysis of the autoradiographs after in situ hybridization. Both techniques clearly demonstrated that oTP and its mRNA were specifically localized in the extra-embryonic trophoblast. Neither the embryonic cells, nor the yolk sac or the amniotic tissues produced the protein or its mRNA. The protein could be detected by d 11 of pregnancy in the elongated blastocyst. Maximum of expression is observed at d 14 and the level decreased by d 16 of pregnancy. The arrest of expression occurred in the regions of trophoblast which have established cellular contacts with the uterine epithelium during the implantation process.     

A single intrauterine infusion of sustained recombinant ovine interferon-tau extends corpus luteum lifespan in cyclic ewes

Delivery carriers were developed to permit sustained release of recombinant ovine tau-interferon (roIFN-tau) to increase corpus luteum (CL) lifespan in cyclic ewes following a single intrauterine administration on Day 10 post estrus. A single infusion with 1.7 mg roIFN-tau covalently bound to carboxymethyl biogel agarose (carbodiimide coupling) significantly increased the interestrus interval (P < 0.01) in treated (n = 4) versus control animals (n = 6), whereas liposomally encapsulated roIFN-tau administered to experimental ewes (n = 8) versus control ewes (n = 6) was less effective (P < 0.05). RoIFN-tau covalently bound to trisacryl (glutaraldehyde coupling) was also effective in cyclic ewes (n = 6), but covalent binding to Eupergit C through oxirane bonds yielded ineffective preparations. Ewes that were given 1.7 mg soluble roIFN-tau (n = 8) displayed slight extension of the CL lifespan compared with ewes that were given 1.7 mg soluble BSA (n = 6), but this extension lacked significance in the Mann-Whitney U-test (P > 0.05). These results are consistent with previous data from experiments performed with daily intrauterine infusion of soluble, native or recombinant oIFN-tau. In addition, because CL maintenance requires only a single administration, these methods are efficient and simple to use since they avoid animal catheterization and allow for reduced injection frequency. Moreover, they may permit the use of smaller amounts of IFN. It is concluded that the use of oIFN-tau sustained in some delivery systems may allow for the development of an experimental sheep pseudopregnancy model.     

Cellular mechanisms by which oxytocin mediates ovine endometrial prostaglandin F2alpha synthesis: role of G(i) proteins and mitogen-activated protein kinases

Oxytocin stimulates a rapid increase in ovine endometrial prostaglandin (PG) F2alpha synthesis. The overall objective of these experiments was to investigate the cellular mechanisms by which oxytocin induces endometrial PGF2alpha synthesis. The objective of experiment 1 was to determine whether G(i) proteins mediate oxytocin-induced PGF2alpha synthesis. Uteri were collected from four ovary-intact ewes on Day 14 postestrus. Caruncular endometrial explants were dissected and subjected to in vitro incubation. Pertussis toxin, an inhibitor of G(i) proteins, had no effect on the ability of oxytocin to induce PGF2alpha synthesis (P > 0.10). The objective of experiment 2 was to determine whether any of the three mitogen-activated protein kinases (MAPKs), extracellular signal regulated protein kinase (ERK1/2), c-Jun N-terminal/stress-activated protein kinase (JNK/SAPK), or p38 MAPK, mediate oxytocin-induced PGF(2alpha) synthesis. Eleven ovary-intact ewes were given an injection of oxytocin (10 IU; i.v.; n = 5) or physiological saline (i.v.; n = 6) on Day 15 postestrus. Uteri were collected 15 min after injection and caruncular endometrium was dissected. Endometrial homogenates were prepared and subjected to Western blotting. Membranes were probed for both total and phosphorylated forms of all three classes of MAPK. All classes of MAPK were detected in ovine endometrium, but oxytocin treatment had no effect on the expression of these proteins (P > 0.10). ERK1/2 was the only phosphorylated MAPK detected and its concentrations were higher in oxytocin-treated ewes (P < 0.01). The objective of experiment 3 was to further investigate the role of ERK1/2 during oxytocin-induced PGF2alpha synthesis. Uteri were collected from four ovary-intact ewes on Day 14 postestrus. Caruncular endometrial explants were dissected and subjected to in vitro incubation. PD98059, a specific inhibitor of ERK1/2 activity, blocked the ability of oxytocin to stimulate PGF(2alpha synthesis in a dose-dependent manner (P < 0.05). These results indicate that the ovine oxytocin receptor is not coupled to G(i) proteins. These results indicate that oxytocin induces phosphorylation of ERK1/2 and that this MAPK appears to mediate oxytocin-induced PGF2alpha synthesis in ovine endometrium.     

Gastrin-releasing peptide (GRP) in the ovine uterus: regulation by interferon tau and progesterone

Gastrin-releasing peptide (GRP) is abundantly expressed by endometrial glands of the ovine uterus and processed into different bioactive peptides, including GRP1-27, GRP18-27, and a C-terminus, that affect cell proliferation and migration. However, little information is available concerning the hormonal regulation of endometrial GRP and expression of GRP receptors in the ovine endometrium and conceptus. These studies determined the effects of pregnancy, progesterone (P4), interferon tau (IFNT), placental lactogen (CSH1), and growth hormone (GH) on expression of GRP in the endometrium and GRP receptors (GRPR, NMBR, BRS3) in the endometrium, conceptus, and placenta. In pregnant ewes, GRP mRNA and protein were first detected predominantly in endometrial glands after Day 10 and were abundant from Days 18 through 120 of gestation. Treatment with IFNT and progesterone but not CSH1 or GH stimulated GRP expression in the endometrial glands. Western blot analyses identified proGRP in uterine luminal fluid and allantoic fluid from Day 80 unilateral pregnant ewes but not in uterine luminal fluid of either cyclic or early pregnant ewes. GRPR mRNA was very low in the Day 18 conceptus and undetectable in the endometrium and placenta; NMBR and BRS3 mRNAs were undetectable in ovine uteroplacental tissues. Collectively, the present studies validate GRP as a novel IFNT-stimulated gene in the glands of the ovine uterus, revealed that IFNT induction of GRP is dependent on P4, and found that exposure of the ovine uterus to P4 for 20 days induces GRP expression in endometrial glands.     

Effects of indomethacin, luteinizing hormone (LH), prostaglandin E2 (PGE2), trilostane, mifepristone, ethamoxytriphetol (MER-25) on secretion of prostaglandin E (PGE), prostaglandin F2alpha (PGF2alpha) and progesterone by ovine corpora lutea of pregnancy or the estrous cycle

Two experiments were conducted to determine the luteotropin of pregnancy in sheep and to examine autocrine and paracrine roles of progesterone and estradiol-17 beta on progesterone secretion by the ovine corpus luteum (CL). Secretion of progesterone per unit mass by day-8 or day-11 CL of the estrous cycle was similar to day-90 CL of pregnancy (P > or = 0.05). In experiment 1, secretion of progesterone in vitro by slices of CL from ewes on day-8 of the estrous cycle was increased (P < or = 0.05) by LH or PGE2. Secretion of progesterone in vitro by CL slices from day-90 pregnant ewes was not affected by LH (P > or = 0.05) while PGE2 increased (P < or = 0.05) secretion of progesterone. Day 8 ovine CL of the estrous cycle did not secrete (P > or = 0.05) detectable quantities of PGF2alpha or PGE while day-90 ovine CL of pregnancy secreted PGE (P < or = 0.05) but not PGF2alpha. Secretion of progesterone and PGE in vitro by day-90 CL of pregnancy was decreased (P < or = 0.05) by indomethacin. The addition of PGE2, but not LH, in combination with indomethacin overcame the decreases in progesterone by indomethacin (P < or = 0.05). In experiment 2, secretion of progesterone in vitro by day-11 CL of the estrous cycle was increased at 4-h (P < or = 0.05) in the absence of treatments. Both day-11 CL of the estrous cycle and day-90 CL of pregnancy secreted detectable quantities of PGE and PGF2alpha (P < or = 0.05). In experiment 1, PGF2alpha secretion by day-8 CL of the estrous cycle and day-90 ovine CL of pregnancy was undetectable, but was detectable in experiment 2 by day-90 CL. Day 90 ovine CL of pregnancy also secreted more PGE than day-11 CL of the estrous cycle (P < or = 0.05), whereas day-8 CL of the estrous cycle did not secrete detectable quantities of PGE (P > or = 0.05). Trilostane, mifepristone, or MER-25 did not affect secretion of progesterone, PGE, or PGF2alpha by day- 11 CL of the estrous cycle or day-90 CL of pregnancy (P > or = 0.05). It is concluded that PGE2, not LH, is the luteotropin at day-90 of pregnancy in sheep and that progesterone does not modify the response to luteotropins. Thus, we found no evidence for an autocrine or paracrine role for progesterone or estradiol-17 36 on luteal secretion of progesterone, PGE or PGF2alpha.     

Effects of indomethacin, luteinizing hormone (LH), prostaglandin E(2) (PGE(2)), trilostane, mifepristone, ethamoxytriphetol (MER-25) on secretion of prostaglandin E (PGE), prostaglandin F(2alpha) (PGF(2alpha)) and progesterone by ovine corpora lutea of pregnancy or the estrous cycle

Two experiments were conducted to determine the luteotropin of pregnancy in sheep and to examine autocrine and paracrine roles of progesterone and estradiol-17 beta on progesterone secretion by the ovine corpus luteum (CL). Secretion of progesterone per unit mass by day-8 or day-11 CL of the estrous cycle was similar to day-90 CL of pregnancy (P >/= 0.05). In experiment 1, secretion of progesterone in vitro by slices of CL from ewes on day-8 of the estrous cycle was increased (P /= 0.05) while PGE(2) increased (P /= 0.05) detectable quantities of PGF(2alpha) or PGE while day-90 ovine CL of pregnancy secreted PGE (P /= 0.05). Trilostane, mifepristone, or MER-25 did not affect secretion of progesterone, PGE, or PGF(2alpha) by day-11 CL of the estrous cycle or day-90 CL of pregnancy (P >/= 0.05). It is concluded that PGE(2), not LH, is the luteotropin at day-90 of pregnancy in sheep and that progesterone does not modify the response to luteotropins. Thus, we found no evidence for an autocrine or paracrine role for progesterone or estradiol-17 36 on luteal secretion of progesterone, PGE or PGF(2alpha).     

Stanniocalcin (STC) in the endometrial glands of the ovine uterus: regulation by progesterone and placental hormones

Stanniocalcin (STC) is a hormone in fish that regulates calcium levels. Mammals have two orthologs of STC with roles in calcium and phosphate metabolism and perhaps cell differentiation. In the kidney and gut, STC regulates calcium and phosphate homeostasis. In the mouse uterus, Stc1 increases in the mesometrial decidua during implantation. These studies determined the effects of pregnancy and related hormones on STC expression in the ovine uterus. In Days 10-16 cyclic and pregnant ewes, STC1 mRNA was not detected in the uterus. Intriguingly, STC1 mRNA appeared on Day 18 of pregnancy, specifically in the endometrial glands, increased from Day 18 to Day 80, and remained abundant to Day 120 of gestation. STC1 mRNA was not detected in the placenta, whereas STC2 mRNA was detected at low abundance in conceptus trophectoderm and endometrial glands during later pregnancy. Immunoreactive STC1 protein was detected predominantly in the endometrial glands after Day 16 of pregnancy and in areolae that transport uterine gland secretions across the placenta. In ovariectomized ewes, long-term progesterone therapy induced STC1 mRNA. Although interferon tau had no effect on endometrial STC1, intrauterine infusions of ovine placental lactogen (PL) increased endometrial gland STC1 mRNA abundance in progestinized ewes. These studies demonstrate that STC1 is induced by progesterone and increased by a placental hormone (PL) in endometrial glands of the ovine uterus during conceptus (embryo/fetus and extraembryonic membranes) implantation and placentation. Western blot analyses revealed the presence of a 25-kDa STC1 protein in the endometrium, uterine luminal fluid, and allantoic fluid. The data suggest that STC1 secreted by the endometrial glands is transported into the fetal circulation and allantoic fluid, where it is hypothesized to regulate growth and differentiation of the fetus and placenta, by placental areolae.     

Select nutrients in the ovine uterine lumen. IX. Differential effects of arginine, leucine, glutamine, and glucose on interferon tau, ornithine decarboxylase, and nitric oxide synthase in the ovine conceptus

Nutrients are primary requirements for development of conceptuses (embryo and extraembryonic membranes), including protein synthesis. We have shown that arginine (Arg), leucine (Leu), and glucose stimulate protein synthesis through phosphorylation of MTOR signaling molecules, thereby increasing proliferation of ovine trophectoderm cells. This study determined whether Arg, Leu, glutamine (Gln), and glucose influence gene expression and protein synthesis in explant cultures of ovine conceptuses recovered from ewes on Day 16 of pregnancy. Conceptuses were deprived of select nutrients and then cultured with either Arg, Leu, Gln, or glucose for 18 h, after which they were analyzed for abundance of MTOR, RPS6K, RPS6, EIF4EBP1 (also known as 4EBP1), IFNT, NOS2, NOS3, GCH1, and ODC1 mRNAs and proteins. Levels of MTOR, RPS6K, RPS6, and EIF4EBP1 mRNAs were not affected by treatment with any of the select nutrients. Similarly, expression of IFNT, NOS2, NOS3, and ODC1 mRNAs were not different. Interestingly, GCH1 mRNA levels increased in response to Arg treatment. Importantly, Arg, Leu, Gln, and glucose increased the abundance of phosphorylated MTOR, RPS6K, RPS6, and EIF4EBP1 proteins as well as NOS and ODC1 proteins, but only Arg increased the abundance of IFNT protein. These findings indicate that Arg, Leu, Gln, and glucose stimulate translation of mRNAs to increase synthesis of proteins through phosphorylation and activation of components of the MTOR signaling pathway. Increases in abundance of IFNT protein (the pregnancy recognition signal), NOS2, NOS3 and GCH1 for conversion of Arg to nitric oxide, and ODC1 for synthesis of polyamines are all important for growth and development of the ovine conceptus during pregnancy.