Unfolding protein response signaling is involved in development,maintenance, and regression of the corpus luteum during the bovine estrous cycle

The corpus luteum (CL) is a transient endocrine organ. Development, maintenance, and regression of CL are effectively controlled by dynamic changes in gene expression. However, it is unknown what types of gene are affected during the CL life span of the estrous cycle in bovine. Here, we determined whether unfolded protein response (UPR) signaling via eIF2α/ATF4/GADD34, p90ATF6/p50ATF6, and IRE1/XBP1, which is a cellular stress response associated with the endoplasmic reticulum (ER), is involved in the bovine CL life span. Our results indicated that expression of Grp78/Bip, the master UPR regulator, was increased during the maintenance stage and rapidly decreased at the regression stage. Additionally, UPR signaling pathways genes were found to be involved in luteal phase progression during the estrous cycle. Our findings suggested that Grp78/Bip, ATF6, and XBP1 act as ER chaperones for initiating CL development and maintaining the CL. In addition, we investigated whether ER stress-mediated apoptosis is occurred through three UPR signaling pathways in CL regression stage. Interestingly, pIRE1 and CHOP were found to be involved in both the adaptive response and ER stress-mediated apoptosis. During the CL regression stage, increased expression of pJNK and CHOP, two components of ER stress-mediated apoptotic cascades, occurred before increased level of cleaved caspase 3 were observed. The present investigation was performed to identify a functional link between UPR signaling and CL life span during the bovine estrous cycle. Taken together, results from this study demonstrated that UPR protein/gene expression levels were different at various stages of the bovine CL life span. Variations in the expression of these protein/genes may play important roles in luteal stage progression during the estrous cycle.     

Systemic and intraovarian effects of corpus luteum on follicular dynamics during estrous cycle in hair breed sheep

The present study aimed to determine systemic and local effects of corpora lutea (CL), on follicular dynamics throughout the estrous cycle. All follicles >or=2 mm and CL were assessed by daily transrectal ultrasonography in 12 West African ewes. Blood samples were collected to determine plasma concentration of progesterone. Fifteen estrous cycles were evaluated with a mean interovulatory interval of 16.8+/-0.2 days. Two (13.3%), 10 (66.7%) and 3 (20%) of the estrous cycles had 2, 3 and 4 waves of follicular development, respectively. In sheep with three waves of follicular development, both the length of growing phase and the growth rate of dominant follicles from midluteal wave II were diminished (3.4+/-0.3 days, P<0.0001, and 0.4+/-0.1 mm/day, P<0.01, respectively) when compared to follicles from early luteal phase (wave I, 4.1+/-0.2 days, and 0.7+/-0.1 mm/day) or late luteal phase (wave III, 6.3+/-0.4 mm and 0.6+/-0.1 mm/day). The diameter of the dominant follicle was smaller during the midluteal phase (3.9+/-0.1 mm, P<0.0001) than in the early and late luteal phase (5.0+/-0.2 and 5.7+/-0.2 mm; respectively). The effect of the dominant follicle was less during midluteal phase, because number of accompanying smaller follicles was fewer (P<0.01) in waves I and III (6.3+/-0.9 compared with 3.4+/-0.8 and 2.3+/-0.7). The number of follicles was also different between ovaries that had CL and those that did not. The total number of large follicles during the luteal phase was less in ovaries with CL (0.9+/-0.5 compared with 2.7+/-0.3; P<0.01), as was the mean daily number of both large (0.1+/-0.02 compared with 0.2+/-0.02; P<0.001) and total number of follicles >or=2 mm (2.5+/-0.1 compared with 3.3+/-0.1; P<0.01). Current results indicate that the presence of a functional CL may exert both systemic and local effects on the population of follicles, affecting the dominance exerted by large follicles.     

Different expression of PGE synthase,PGF receptor,TNF, Fas and oxytocin in the bovine corpus luteum of the estrous cycle and pregnancy

Functional differences between the corpus luteum (CL) of pregnancy and CL of the cycle in cows were examined. Messenger RNA and protein levels of prostaglandin (PG) E synthase (PGES), PGF2α receptor (PGFR), tumor necrosis factor-α (TNF) and Fas were found to be higher in the CL of pregnancy than in CL of the cycle. Oxytocin (OT) mRNA and protein levels were lower in the CL of pregnancy. Messenger RNA levels of progesterone receptor (PR), luteinizing hormone receptor (LHR), PGE2 receptor (PGER), PGF synthase (PGFS), TNF receptor type I (TNFRI) and TNF receptor type II (TNFRII) did not differ between the cycle and pregnancy. PGE2 and PGF2α production by cultured bovine endometrial tissues was decreased by a supernatant derived from the homogenized CL of pregnancy but not by that of the CL of the cycle, suggesting that specific substances in the CL of pregnancy affect endometrial PG production in cows. Collectively, PGES, PGFR, TNF, Fas or OT may contribute to differences between the CL of pregnancy and CL of the estrous cycle in cows.     

Corpus luteum function following single and double ovulation during estrous cycle in Sanjabi ewes

This study compared the effect of double and single ovulation on serum progesterone concentrations and luteal characteristics in Sanjabi ewes at different days of the estrous cycle. The estrous cycles of 197 Sanjabi ewes were synchronized by a 12-day treatment with intravaginal sponges (Chronogest^®). Estrus was detected in 144 ewes 27–39 h after sponge removal. Daily blood samples were taken every morning and analyzed for serum progesterone (P4). Ewes were then transported to a local abattoir, where nine ewes were slaughtered on each experimental day (days 1–16 after estrus) for ovary collection. The ovarian follicles were measured and categorized by size (very small <2 mm; small 2–3.5 mm; medium 3.5–5 mm; large >5 mm). On each slaughter day, the number of corpora lutea per ewe was classified as single and double ovulation. The results show that the effect of dominant follicles was less during the mid-luteal phase. Ovulation rate of right, left and both ovaries were (54.9%), (23.6%) and (21.5%), respectively. The incidence of double ovulations was 40.2%. In the case of ewes exhibiting double ovulation, 46.6% occurred unilateral (ewes exhibited both ovulations on the right ovary); whereas 53.4% occurred bilateral (ewes exhibited ovulations on the right and left ovaries). Unilateral double ovulation was not observed in the left ovary. The right ovary appeared to play a significantly greater role in ewes showing single and double ovulations than the left ovary (P < 0.05). Serum progesterone concentration showed minimum and maximum levels of 0.29 ± 0.15 and 5.51 ± 0.75 ng/ml on days 16 and 11 post-estrous, respectively (P < 0.001). The mean volume of individual corpus lutea in ewes with single ovulations was significantly higher than in ewes with double ovulations (P < 0.01). However, the total volume of corpus lutea in ewes with single ovulation was significantly lower than in ewes with double ovulations in some days of estrous cycle (P < 0.01). The serum progesterone concentration was significantly higher in double than single ovulating animals on days 1–16 of the estrous cycle (P < 0.001). These results indicated a relatively high incidence of double ovulation in ewes associated with increasing total luteal volume and high circulating concentrations of progesterone.     

Differences of microvascular endothelium in the bovine corpus luterum of pregnancy and the corpus luteum of the estrous cycle

Summary— The purpose of the present study was to investigate potential modulations of endothelial cells of the bovine corpus luteum (CL) during pregnancy. Luteal endothelia of pregnant and non-pregnant cows were isolated and purity of cultures was verified by flow cytometric quantification of three independent endothelial markers (von Willebrand factor, angiotensin converting enzyme, Bandeiraea simplicifolia agglutinin I ligands). Different cellular parameters including light and electron microscopical investigation of morphology and growth characteristics as well as quantification of cellular lectin binding sites were compared. Extensive heterogeneity between luteal endothelial cells in pregnant and non-pregnant animals could be demonstrated, reflected in functional attributes like angiogenic activity, ultrastructural characteristics and the quantitative expression of cellular carbohydrates. Two different morphological types of cells (‘cobblestone growth pattern’ and ‘arcuate growth pattern’) were isolated from the CL of pregnancy as well as from the cyclic CL. Spontaneous angiogenic activities, including cellular migration in band-like structures and formation of ring-like structures, were observed in endothelial cells isolated from the CL of pregnant cows exclusively. This strongly suggests that microvascular luteal endothelium of pregnant animals, in contrast to the one of non-pregnant animals, is able to produce quantitatively and/or qualitatively specific angiogenesis factor(s). Heterogeneity between luteal endothelial cells in the pregnant and non-pregnant animal could also be demonstrated by quantification of lectin (Bandeiraea simplicifolia agglutinin I, concanavalin A, Dolichos biflorus agglutinin, Ulex europaeus agglutinin I, wheat germ agglutinin) binding sites: quantitative expression of specific endothelial cell surface carbohydrates could be correlated to be status of pregnancy, thus emphasizing the actual need of quantification of lectin binding.     

Localization of thyrotropin receptor and thyroglobulin in the bovine corpus luteum

The receptor of the Thyroid Stimulating Hormone (TSHr) and thyroglobin (TGB), are two proteic factors necessary for the synthesis of hormones, in the thyrocite. In mammals, many immuno-histochemical reports indicate the presence of the TSHr in extra-thyroidal tissues, but not in the ovary. Triiodothyronine (T₃) and thyroxine (T₄) have been widely shown to affect ovarian functions and the synthesis of progesterone (P₄). The aim of this study was to determine if by immunohistochemistry techniques TSHr and TGB could be found in the bovine corpora haemorragica, lutea and albicantia.     

Expression and localization of progesterone receptor membrane component 1 and 2 and serpine mRNA binding protein 1 in the bovine corpus luteum during the estrous cycle and the first trimester of pregnancy

The aim of this study was to evaluate the mRNA and protein expression and the localization of progesterone receptor membrane component 1 (PGRMC1), PGRMC2, and the PGRMC1 partner serpine mRNA binding protein 1 (SERBP1) in the bovine CL on Days 2 to 5, 6 to 10, 11 to 16, and 17 to 20 of the estrous cycle as well as during Weeks 3 to 5, 6 to 8, and 9 to 12 of pregnancy (n = 5–6 per each period). The highest levels of PGRMC1 and PGRMC2 mRNA expression were found on Days 6 to 16 (P < 0.05) and 11 to 16, respectively, of the estrous cycle and during pregnancy (P < 0.001). The level of PGRMC1 protein was the highest (P < 0.05) on Days 11 to 16 of the estrous cycle compared with the other stages of the estrous cycle and pregnancy, whereas PGRMC2 protein expression (P < 0.001) was the highest on Days 17 to 20 and also during pregnancy. The mRNA expression of SERBP1 was increased (P < 0.05) on Days 11 to 16, whereas the level of its protein product was decreased (P < 0.05) on Days 6 to 10 of the estrous cycle and was at its lowest (P < 0.001) on Days 17 to 20. In pregnant cows, the patterns of SERBP1 mRNA and protein expression remained constant and were comparable with those observed during the estrous cycle. Progesterone receptor membrane component 1 and PGRMC2 localized to both large and small luteal cells, whereas SERBP1 was observed mainly in small luteal cells and much less frequently in large luteal cells. All proteins were also localized in the endothelial cells of blood vessels. The data obtained indicate the variable expression of PGRMC1, PGRMC2, and SERBP1 mRNA and protein in the bovine CL and suggest that progesterone may regulate CL function via its membrane receptors during both the estrous cycle and pregnancy.     

Pregnancy rates and corpus luteum-related factors affecting pregnancy establishment in bovine recipients synchronized for fixed-time embryo transfer

The objective was to investigate the influence of corpora lutea physical and functional characteristics on pregnancy rates in bovine recipients synchronized for fixed-time embryo transfer (FTET). Crossbred (Bos taurus taurus × Bos taurus indicus) nonlactating cows and heifers (n = 259) were treated with the following protocol: 2 mg estradiol benzoate (EB) plus an intravaginal progesterone device (CIDR 1.9 g progesterone; Day 0); 400 IU equine chorionic gonadotropin (eCG; Day 5); prostaglandin F_2α (PGF_2α) and CIDR withdrawal (Day 8); and 1 mg EB (Day 9). Ovarian ultrasonography and blood sample collections were performed on Day 17. Of the 259 cattle initially treated, 197 (76.1%) were suitable recipients; they received a single, fresh, quality grade 1 or 2 in vivo-derived (n = 90) or in vitro-produced (n = 87) embryo on Day 17. Pregnancy rates (23 d after embryo transfer) were higher for in vivo-derived embryos than for in vitro-produced embryos (58.8% vs. 31.0%, respectively; P < 0.001). Mean (±SD) plasma progesterone (P₄) concentration was higher in cattle that became pregnant than that in nonpregnant cattle (5.2 ± 5.0 vs. 3.8 ± 2.4 ng/mL; P = 0.02). Mean pixel values (71.8 ± 1.3 vs. 71.2 ± 1.1) and pixel heterogeneity (14.8 ± 0.3 vs. 14.5 ± 0.5) were similar between pregnant and nonpregnant recipients (P > 0.10). No significant relationship was detected between pregnancy outcome and plasma P₄, corpus luteum area, or corpus luteum echotexture. Embryo type, however, affected the odds of pregnancy. In conclusion, corpus luteum-related traits were poor predictors of pregnancy in recipients. The type of embryo, however, was a major factor affecting pregnancy outcome.     

Expression and localization of locally produced growth factors regulating lymphangiogenesis during different stages of the estrous cycle in corpus luteum of buffalo (Bubalus bubalis)

Recent experiments using expression, immunolocalization, and cell culture approaches have provided leading insights into regulation of luteal angiogenesis by different growth factor systems and its role in the function of corpus luteum (CL) in buffalo. On the contrary, lymphangiogenesis and its regulation in the CL are still poorly understood. The aim of this study was to evaluate the expression and localization of lymphangiogenic factors (vascular endothelial growth factor [VEGF]-C and VEGFD), their receptor (VEGFR3), and lymphatic endothelial marker (LYVE1) in bubaline CL during different stages of the estrous cycle and to investigate functional role of VEGFC and VEGFD in luteal lymphangeogenesis. The mRNA and protein expression of VEGFC, VEGFD, and VEGFR3 was significantly greater in mid and late luteal phases, which correlated well with the expression of LYVE1. The lymphangiogenic factors were localized in luteal cells, exclusively in the cytoplasm. Immunoreactivity of VEGFC was greater during midluteal phase and that of VEGFD was greater during the mid and late luteal phases. Luteal cells were cultured in vitro and treated for different time duration (24, 48, and 72 hours) with VEGFC and VEGFD each at 50, 100, and 150 ng/mL concentration and VEGFC with VEGFD at 100 ng/mL concentration. The temporal increase in LYVE1 mRNA expression was significant (P < 0.05) in VEGFC and VEGFC with VEGFD treatment and no significant change was seen in VEGFD treatment. Thus, it seems likely that VEGFD itself has little role in lymphangiogenesis but along with VEGFC it might have a synergistic effect on VEGFR3 receptors for inducing lymphangiogenesis. In summary, the present study provided evidence that VEGFC and VEGFD, and their receptor VEGFR3, are expressed in bubaline CL and are localized exclusively in the cell cytoplasm, suggesting that these factors have a functional role in lymphangiogenesis of CL in buffalo.     

Morphological characteristics of the bovine corpus luteum during the estrous cycle and pregnancy

The corpus luteum, one of the biological clocks of the estrous cycle and pregnancy, is known foremost for its production of progesterone that blocks the pituitary release of gonadotropins and prepares the uterus for a pregnancy. The cellular sources of this progesterone are the steroidogenic small and large luteal cells. Other luteal cells that are not steroidogenic, but are believed to have an important role in the function of this gland are the fibroblast, macrophages and endothelial cells. The most prominent luteal cell is the large steroidogenic cell characterized by an abundance of smooth endoplasmic reticulum and densely packed spherical mitochondria that are indicative of its contribution to most of the circulating progesterone believed to be constitutively secreted and not under the control of LH. Other distinguishing features of the large luteal cell are the presence of rough endoplasmic reticulum, prominent Golgi, and secretory granules that are indicative of endocrine cells. This cell undergoes dynamic changes across the estrous cycle and pregnancy, believed to reflect a change in progesterone and protein secretion that will eventually influence a successful pregnancy or another ovulation if pregnancy fails. The morphological characteristics of the bovine luteal cells are the focus of this review.     

Immunocytochemical evidence for the presence of oxytocin and neurophysin in the large cells of the bovine corpus luteum

Summary The presence of neurophysin, oxytocin and vasopressin in the bovine corpus luteum was examined immunocytochemically. Tissue blocks of corpora lutea from pregnant and non-pregnant animals were fixed with glutaraldehyde/paraformaldehyde fixative and immunostained by the peroxidase-antiperoxidase (PAP) method. The simultaneous presence of immunoreactive oxytocin and immunoreactive oxytocin-neurophysin was demonstrated in large luteal cells of non-pregnant animals, while no staining for vasopressin or vasopressin-neurophysin was observed. None of the peptides were detected in the corpus luteum of pregnant animals. The small luteal cells were not found to be stainable at any time.     

Luteal blood flow is a more appropriate indicator for luteal function during the bovine estrous cycle than luteal size

The objective of this study was to assess the reliability of luteal blood flow (LBF) as recorded by color Doppler sonography to monitor luteal function during the estrous cycle of dairy cows and to compare the results with that for the established criterion luteal size (LS) as determined by B-mode sonography. In total, 14 consecutive sonographic examinations were carried out in 10 synchronized lactating Holstein-Friesian cows (Bos taurus) on Days 4, 5, 6, 7, 8, 10, 12, 14, 16, -5, -4, -3, -2, -1 of the estrous cycle (Day 1 = ovulation). Plasma progesterone concentrations in venous blood (P₄) were quantified by enzyme immunoassay. Luteal size was determined by sonographic measurement of the maximal cross-sectional area of the corpus luteum (CL). Luteal blood supply was estimated by calculating the maximum colored area of the CL from power Doppler sonographic images. Luteal size doubled during the luteal growth phase (until Day 7) and remained at this level during the luteal static phase (Day 8 to 16) before decreasing rather slowly during luteal regression (Days -5 to -1). Luteal blood flow doubled during the growth phase, doubled furthermore during the static phase, and decreased rapidly during luteal regression. Thus, LBF values represented highly reliable predictors of luteal status. Luteal blood flow predicted reliably a P₄ > 1.0 ng/mL by reaching only 35% of the maximal values, whereas LS had to exceed 60% of the maximal values to indicate reliably a functional CL. It is concluded that LBF reflected luteal function better than LS specifically during luteal regression.     

Lipofuscin in the corpus luteum of macaque ovaries

The present study examined the histochemistry of pigments in the corpus luteum of the ovaries of cynomolgus monkeys (Macaca fascicularis), Japanese monkeys (Macaca fuscata), and rhesus monkeys (Macaca mulatta). Yellowish brown pigments were found in the regressing corpus luteum cells. Histochemical studies revealed that these pigments consisted of lipofuscin, the so-called age pigment. The findings obtained suggest that accumulation of lipofuscin might be related to cellular aging of the corpus luteum.     

Hormonal regulation and function of an RNA helicase,Ddx5 in corpus luteum of adult Wistar rats

Corpus luteum (CL) is an endocrine tissue involved in regulation of reproductive cycle and early pregnancy establishment. In the present study DEAD-box helicase-5 (Ddx5), a member of the DEAD box family of RNA helicases was investigated for its expression, regulation and function in CL of Wistar rats. Ddx5 was expressed in adult rat CL. Primary cell culture from supra-ovulated ovaries were established for in vitro studies. Addition of luteinizing hormone (LH; 100 ng/ml), a luteotrophic factor in primary cell culture, decreased Ddx5 RNA expression (foldchange:0.6 ± 0.075) while prostaglandin alpha (PGF_2α; 1μM), a luteolytic factor caused an increase (foldchange:2.4 ± 0.4) compared to control group. Under in vivo conditions, the administration of PGF_2α or gonadotropin-releasing hormone antagonist; cetrorelix (CET) caused luteolysis as well as an increase in the protein level of Ddx5 (foldchange:1.9 ± 0.27 and 1.4 ± 0.09 viz.; p < 0.05) in CL of adult rats. LH was administered post CET treatment which suppressed Ddx5 protein expression (foldchange:0.8 ± 0.16; p < 0.05) compared to CET treated group. Further, it was observed that the expression of Ddx5 was upregulated (foldchange:1.5 ± 0.23; p < 0.05) in CL during late pregnancy compared to mid pregnancy concomitant to luteolysis in adult rats. Overall, the results suggest for the first time that Ddx5 is expressed in rat CL and regulated by luteolytic and luteotrophic factors in an inverse fashion. Further, the data significantly correlates ddx5 expression to CL regression suggesting involvement of ddx5 in luteolysis. These results suggest a significant role of Ddx5 in female reproduction biology and warrant in depth examination of the function of Ddx5 in CL.     

Expression of vascular endothelial growth factor and its receptors in the porcine corpus luteum during the estrous cycle and early pregnancy

The present study was undertaken to determine the expression of vascular endothelial growth factor (VEGF) and its receptors, fms-like tyrosine kinase (Flt-1) and fetal liver kinase-1/kinase insert domain-containing receptor (Flk-1/KDR), in the porcine corpus luteum (CL) during the estrous cycle and early pregnancy. Immunohistochemical studies localized proteins of VEGF ligand-receptor system in the cytoplasm of luteal cells and in some blood vessels. Western blot analysis revealed significantly higher levels of VEGF protein during early and mid-luteal phase (vs. late luteal phase; P<0.001 and P<0.01, respectively). Quantification of VEGF mRNA in the CL showed increased mRNA levels during entire luteal phase (vs. Days 16-17; P<0.05). Expression of Flt-1 protein remained high during luteal phase (P<0.001), but the mRNA levels tended to increase from the early to the late luteal phase. Elevated protein expression of Flk-1/KDR was found in the mid-luteal phase (vs. Days 16-17; P<0.05). However, induction of Flk-1/KDR mRNA expression occurred earlier, in early luteal phase. The lowest VEGF, Flt-1 and Flk-1/KDR mRNA and protein levels were observed in regressed CL (P<0.001). During pregnancy, VEGF, Flt-1 and Flk-1/KDR mRNA and protein expression was comparable to the mid-luteal phase. In conclusion, the present study has demonstrated dynamic expression of VEGF and its receptors in the porcine CL during the estrous cycle and early pregnancy. These data suggest that the VEGF ligand-receptor system may play an important role in the development and maintenance of the CL in pigs.     

Ultrastructural localisation of oxytocin and neurophysin in the ovine corpus luteum

Summary Polyclonal rabbit antisera raised against oxytocin, bovine neurophysin I and vasopressin were used, together with an immunogold complex, to localise the peptides in ultrathin sections of ovine corpus luteum. The only organelle which consistently showed gold labelling was the secretory granule of the large luteal cell. In non-consecutive sections of the same large luteal cell all the granules showed a similar level of labelling after oxytocin or neurophysin I antisera: however no immunolabelling was detected for vasopressin. Oxytocin and neurophysin seem to be rapidly lost after secretion since exocytosed granule cores showed no labelling above background levels.     

Different phenotyes of cultured microvessel endothelial cells obtained from bovine corpus luteum

Summary Morphological heterogeneity has not been documented for cultured endothelial cells isolated from the microvascular bed of any organ. As the corpus luteum depends on a rich microvascularization, endothelial cells were dislodged from developing corpora lutea by mechanical dissection followed either by collagenase digestion or by no digestion. Cell separation was carried out by Percoll density centrifugation. Although the yield of intact cells was higher with collagenase treatment than without, successful endothelial cell cultures were only established when cells remained untreated. Viewed by light microscopy after an average lag phase of 10 days, five different phenotypes of endothelial cells were found under similar simple culture conditions: isomorphic epithelioid, polymorphic epithelioid, spindle-shaped, round, and phase-dense phenotypes. Monolayers appeared within 2–4 weeks. After an additional period of 2–4 weeks, tubular forms with a specific pattern were noted for types 1–3, the so-called pseudotubular forms for type 4, and none for type 5. Cell types differed in their cytochemical and immunocytochemical responses. Examined by SEM, type 1 displayed a more conspicuous surface anatomy than type 2. Types 3–5 demonstrated striking cell processes that were characteristic of each type. Tubular forms of types 1 and 2 showed cell borders and a marked increase in surface specializations, whereas tubular forms of type 3 lacked detectable cell borders in the absence of a striking surface anatomy. Pseudotubular forms of type 4 developed no particular spatial organization. Thus, for the first time, morphological evidence is provided that different endothelial cell types are obtained from diverse segments of the microvascular bed.     

Localization of relaxin in human gestational corpus luteum

Summary Corpora lutea from 12 pregnant women were prepared for immunohistochemical localization of relaxin using a highly specific antiserum. A positive response is given by luteal cells that are diffusely distributed throughout the corpus luteum. These cells do not form a distinctive group in any particular area. A negative response is seen in the adjacent ovarian tissue, and also in nongestational corpora lutea in an early luteal phase.