BIOLOGY OF REPRODUCTION 55, (1996)

Transcription

1 BIOLOGY OF REPRODUCTION 55, (1996) Steady-State Luteinizing Hormone Receptor Messenger Ribonucleic Acid Levels and Endothelial Cell Composition in Bovine Normal- and Short-Lived Corpora Lutea Gary D. Smith,1" 2 Heywood R. Sawyer, 4 Mark A. Mirando,2 Michael D. Griswold,3 Annama Sadhu,s and Jerry 1. Reeves 2 Department of Animal Sciences 2 and Department of Biochemistry and Biophysics, 3 Washington State University, Pullman, Washington Animal Reproduction and Biotechnology Laboratory 4 Colorado State University, Fort Collins, Colorado Department of Obstetrics and Gynecology, 5 The University of Chicago, Chicago, Illinois ABSTRACT The short-lived corpus luteum (CL) contributes to reproductive inefficiency during the postpartum period in beef cows. The cause for the early demise of the short-lived CL is not fully understood but is believed to involve a premature release of prostaglandin F 2,,. The objectives of this study were to evaluate norgestomet-hcg-induced normal-lived CL and hcg-induced shortlived CL in postpartum cows with respect to serum progesterone (P 4 ) and 13,14-dihydro-15-keto, prostaglandin F 2, (PGFM) concentrations and luteal LH receptor (LH-R) concentrations, LH-R mrna levels, and vascularity. Although serum P 4 profiles from the time of hcg administration (Day 0) until luteectomy (Day 6, 7, or 8) were similar between CL life span groups, PGFM concentrations were elevated (p < 0.05) on Day 8 in cows expected to have short-lived CL compared to normal-lived CL. The LH-R concentrations were similar between normal- and shortlived CL on all days measured. Irrespective of luteal life span and day of luteectomy, all CL possessed a 4.4-kb LH-R transcript. Actin-normalized LH-R mrna levels were similar between normal- and short-lived CL on Days 6 and 7; however, Day 8 shortlived CL contained less (p < 0.05) LH-R mrna than Day 8 normal-lived CL. Although the area of luteal tissue occupied by capillaries in normal- and short-lived CL was similar on Days 6 and 7, the area occupied by capillaries in short-lived CL was lower (p < 0.05) than that for normal-lived CL on Day 8. Collectively, these results indicate that there is a decrease in steady-state LH-R mrna and a reduction in luteal vascularity in CL expected to be short-lived. These changes occur concomitantly with a rise in serum PGFM, but prior to a decline in serum P 4. INTRODUCTION Subnormal corpora lutea (CL) have been reported in humans [1], nonhuman primates [2], sheep [3], and cattle [4]. Subnormal CL in the cow, which form during the postpartum period either spontaneously [5] or after administration of GnRH [6] or hcg [7], have a shortened life span (under 14 days, and averaging approximately 8 days [7, 8]). Since prostaglandin (PG) F 2. levels are significantly higher between Days 4 and 9 in cows exhibiting short cycles compared to cows with normal luteal phases [9], it appears that the early demise of the short-lived CL is due, at least in part, to the premature release of uterine PGF 2. [10-13]. In dairy cows, luteolysis in both short and normal first estrous cycles occurs concomitantly with a rise in 13,14-dihydro-15-keto-prostaglandin F 2. (PGFM) [14]. In addition, indomethacin treatment [15] or hysterectomy [16] Accepted June 14, Received August 28, 'Correspondence and current address: Dr. Gary D. Smith, Department of Obstetrics and Gynecology, The University of Chicago, MC South Maryland Ave., Chicago, IL FAX: (312) prolongs luteal function in cows expected to form shortlived CL. Many mechanisms have been postulated to explain the luteolytic effect of PGF 2,, including a decrease in LH receptor (LH-R) number and function [17, 18] and decreased luteal blood flow [19]. Luteolytic doses of PGF 2 reduce luteal blood flow within 2 h of administration [19], and, at 12 h, morphological changes in endothelial cells consistent with the process of apoptosis are evident [20]. Luteinizing hormone plays a significant role in regulating ovarian function and is necessary in order for the processes of folliculogenesis [21], ovulation [22], and luteogenesis [23] to proceed normally. The initial step in the mechanism of action of LH is the binding of LH to specific receptors located in the plasma membrane of target cells. This binding results in increased adenylate cyclase activity, mediated by intracellular membrane-associated G proteins [24], an increase in intracellular camp, and ultimately activation of protein kinase A [25]. The LH-R has a molecular mass of 94 kda, and its cdna has been characterized in several species [26-28]. The LH-R is a single polypeptide composed of 694 amino acids and, similar to other G protein-coupled receptors, contains 7 hydrophobic regions that represent transmembrane portions. Characterization of the transcript that encodes the LH-R in various species has been reported [26, 29] and ranges from 7.7 kb to 1.2 kb with the predominant transcript at 4.4 kb. Concentrations of LH-R have been reported to be similar between cows with induced normal and subnormal CL on Days 5 and 6.5 after hcg or GnRH injection [7, 30]. However, on Day 7, CL expected to be short-lived were less responsive to LH stimulation in vitro than CL expected to have a normal life span [6]. The objectives of this study were to evaluate the luteal steady-state LH-R mrna levels, LH-R concentrations, and degree of vascularity, as well as serum progesterone (P 4 ) and PGFM concentrations, on Days 6, 7, and 8 after hcg administration in cows expected to have normal or subnormal CL. MATERIALS AND METHODS Animals Fifty postpartum beef cows in fair to good body condition were induced to ovulate by administration of a single i.m. injection of 1000 IU hcg (Sigma Chemical Company, St. Louis, MO) between 33 and 40 days postpartum. This regimen has been shown to induce CL with a shortened life span unless preceded by P 4 treatment [31, 32]. Beginning 11 days prior to hcg administration, 25 cows received an s.c. implant of norgestomet (synthetic progestin; CEVA Laboratories, Inc., Overland Park, KS) for 9 days (cows

2 LH-R mrna IN SHORT-LIVED CL 903 expected to form CL with normal life span), whereas the remaining 25 cows received an s.c. placebo implant for 9 days (cows expected to form CL with shortened life span). A 5-ml blood sample was collected via median caudal venipuncture from all postpartum cows twice weekly beginning 14 days postpartum until the time of the hcg injection, and serum concentrations of P 4 were determined to identify ovulation prior to hcg treatment. Five cows from the short-lived CL group had elevated P 4 levels prior to hcg injection and were removed from the study. Blood samples were collected beginning at the time of hcg administration until the time of luteectomy (n = 29) to evaluate both serum P 4 and PGFM. In addition, blood samples were collected for 20 consecutive days from nonsurgically treated postpartum cows expected to form normal-lived (n = 10) or short-lived (n = 6) CL in order to observe P 4 values during these periods and to verify efficacy of treatment regimens. Cows were considered to have short-lived CL if P 4 concentrations fell below 1.0 ng/ml before Day 14. Serum was harvested from all blood samples and subsequently stored at C until P 4 and PGFM assays were performed. Radioimmunoassay Concentrations of P 4 were determined, as previously described, by RIA [33]. Intra- and interassay coefficients of variation (CV) were 8.8% and 12.5%, respectively. The 11lla-hemisuccinate-tyrosine methyl ester progesterone for iodination was provided by A. Belanger (Le Centre Hospitalier de L' Universite Laval, Quebec, PQ, Canada). Concentrations of PGFM in 200 Ril cow serum were quantified by RIA without extraction using [5,6,8, 9,11,12,14,- 3 H(N)]-PGFM (spec. act. 182 Ci/mmol; Amersham, Arlington Heights, IL), PGFM antiserum J57 (1: final tube dilution), and radioinert PGFM (Sigma) for standards as previously described [34, 35]. For preparation of PGFM standards, blood was collected from steers directly into chilled heparinized tubes containing 1 mg/ml aspirin and immediately centrifuged for 10 min at 1800 x g, 4 C; the plasma obtained was used to dilute the PGFM standards. Separation of free and antibody-bound PGFM was achieved by precipitation of the antibody-bound PGFM with addition of an equal volume of 40% polyethylene glycol (w:v) in 50 mm Tris-HCI (ph 7.5) and 0.01% NaN 3 (w:v), followed by centrifugation for 20 min at 1800 x g and 4 C. Supernatants were decanted, the pellets were resuspended, and the precipitation/centrifugation was repeated to further reduce nonspecific binding. Concentrations of PGFM in serum were calculated using a log-logit regression program [36]. Assay sensitivity was 2.5 pg/tube (p < 0.01). Intra- and interassay CV were 11.0% and 5.5%, respectively. The assay was validated by quantifying concentrations of PGFM in 6 replicates each of 100, 150, 200, 250, and 300 Rl of a pool made from cow sera. The slope of the generated displacement curve was parallel to that of the standards. The quantity of PGFM measured after addition of 10, 20, 40, 80, 120, or 160 pg PGFM to 200 Rl of pooled cow serum averaged 99.5% of the amount added, and regression of the amount of PGFM recovered on the amount added was Y = X (r 2 = 0.99, p < 0.001). The antisera cross-reacted < 1% with PGF 2,, PGF,,, PGE 2, and arachidonic acid [37]. Collection of CL Luteectomy was performed as previously described [38] by manual enucleation of the CL through a vaginal incision following caudal epidural analgesia. Corpora lutea were collected from cows expected to have normal-lived and short-lived CL on Day 6 (n = 5 for each CL life-span group), Day 7 (n = 5 for each CL life-span group), or Day 8 (normal-lived n = 5; short-lived n = 4). A small portion (1-2 mm 3 ) was removed from each CL and placed immediately in fixative containing 50% picric acid, 4% paraformaldehyde, and 0.25% glutaraldehyde in 0.1 M phosphate buffer, ph 7.4. Tissue was fixed for 2 h and stored in 0.1 M phosphate buffer at 4 C until processed as described below. The remainder of the luteal tissue was used for determining LH-R content and steady-state LH-R mrna levels. Luteal Vascularity Fixed luteal tissue was embedded in paraffin, and four sections approximately 5 im in thickness were cut from each block at 50-1pm intervals. Sections were placed on slides and processed for immunohistochemical identification of vascular endothelial cells using a rabbit anti-human von Willebrand factor antibody (A082, Dako Polyclonal; Dako, Santa Barbara, CA) [39]. Sections from a canine cutaneous hemangiosarcoma sample served as a positive control for the staining procedure. Negative controls were conducted for each luteal sample evaluated; these consisted of substitution of a normal rabbit immunoglobulin fraction (X903, Dako Polyclonal) for the rabbit anti-human von Willebrand factor antibody. Microscopic assessment of vascularity was performed blind to treatment using a Nikon optiphot microscope (Nikon Inc., Melville, NY) with a 20x objective. For each luteal sample, single randomly selected fields ( iim 2 ) from each of the four sections were evaluated. Endothelial cells that stained positive were used to identify vascular components. The outer border of each capillary in each field was circumscribed, and area of each field occupied by vascular components was calculated using Bioquant System IV (R&M Biometrics Inc., Nashville, TN) image analysis software. Values reported are a mean percentage of the set area assessed that stained positive for von Willebrand factor (endothelial cells). Assay of LH-R To assess LH-R concentrations in normal and subnormal CL, tissue samples were homogenized in assay buffer (0.01 M Tris-HCI, 0.1% NaN 3, 5 mm MgCl 2 ; ph 7.4, 20 C) followed by centrifugation at x g for 15 min. Resulting pellets were resuspended in assay buffer to a final concentration of 50 mg/ml based on prehomogenization luteal sample weight. Quantification of LH-R was performed as previously described [40]; purified hcg (CR121; IU/mg) was used as the labeled ligand [41], and crude hcg (Sigma; 3100 IU/mg) was used as the unlabeled ligand. Specific binding was determined by subtracting nonspecific from total binding. The intraassay CV was 7.9%. Analysis of LH-R mrna Bovine luteal total RNA was isolated using the RNAzol method as previously described [42]. Approximately 1 g of tissue, frozen in liquid nitrogen within 10 min of excision, was homogenized in 10 ml of 4 M guanidinium isothiocyanate, 25 mm sodium citrate (ph 7.0), 0.5% sarkosyl, and 0.1 M 2-mercaptoethanol. After RNA extraction and

3 904 SMITH ET AL. precipitation, polyadenylated RNA was isolated by a double passage over an oligo(dt)-cellulose (type III; Collaborative Research, Waltham, MA) column [43]. The possibility of RNA degradation during or after both total and poly(a) + RNA preparation was monitored by electrophoresis on a 1.2% agarose gel under denaturing conditions with formaldehyde followed by visualization of nucleic acids with ethidium bromide. Specificity of a 180-bp cdna encoding a portion of the rat LH-R [44] for the bovine LH-R was assessed by Northern blot analysis. This rat cdna was labeled using a random primer DNA labeling system (Gibco BRL, Gaithersburg, MD). Northern blots were prepared by electrophoresing bovine CL, kidney, liver, and lung poly(a) + RNA (20- lg samples) through 1.2% agarose gels under denaturing conditions with formaldehyde; this was followed by transfer to a nitrocellulose membrane using standard wicking procedures with 20-strength SSC (single-strength SSC = 150 mm NaCl, 15 mm citric acid; ph 7.0) as the transfer buffer. Transfer was terminated after 24 h, and the nitrocellulose was washed and baked at 80C for 2 h under vacuum. Before hybridization, blots were incubated in prehybridization buffer containing 50% formamide, 5-strength SSPE (150 mm NaC1, 10 mm NaH 2 PO 4.H 2 0, 1 mm EDTA; ph 7.4), double-strength Denhardt's reagent, 0.1% SDS, and 100 plg denatured salmon sperm DNA for 1 h at 42 C. Hybridization of the membranes was performed for 16 h at 42 0 C in fresh buffer (as described above) with the addition of denatured [ 32 P]rat 180-bp cdna. After hybridization, filters were washed twice in single-strength SSC plus 0.1% SDS at 20 C for 15 min each; they were then washed twice more in 0.25-strength SSC plus 0.1% SDS at 20 C for 15 min each. Finally, filters were washed twice for 15 min/wash with 0.2-strength SSC plus 0.1% SDS at 50 C. Radiography was carried out with preflashed Kodak XAR-5 film (Eastman Kodak Co., Rochester, NY) at -80 C in the presence of an intensifying screen. Luteinizing hormone receptor mrna from normal and subnormal CL was quantified by Northern blot analysis of 20 lpg luteal poly(a) + RNA using [ 3 2 P]rat LH-R cdna and [ 3 2 P]human actin cdna [45] as described above. Four filters were required to analyze all samples. Influence of filter was removed by analyzing samples from all three days (both treatments), plus one extra day (both treatments) per filter. Radiography was conducted for 2 h and 18 h with the same filters for visualization of actin mrna and LH-R mrna bands, respectively. Densitometry was conducted using whole band analysis, and because LH-R and actin signals were determined from the same filters, the LH-R mrna values were normalized for actin signal intensity to control for loading and/or transfer variation. Statistical Analysis Differences in serum P 4 and PGFM were examined by analysis of variance for repeat measures followed by pairwise comparisons between CL life span groups on similar days. Log transformations were performed on PGFM values prior to analysis because of heterogeneity of variance within CL life span groups on individual days. Luteal endothelial cell composition, luteal LH-R concentration, and luteal LH-R and actin mrna levels were compared between CL life span groups on similar days and within CL life span groups on different days using unpaired Student's t-tests. Differences were considered significant at p < E 0 o a) E 2 60 E E2 CD Co n B -o- Normal-lived Short-lived i= t i--= s~~~~~~~~tzi ^~~ Post-hCG (days) FIG. 1. A) Serum progesterone profiles from cows expected to have normal- and short-lived CL from the day of hcg administration (Day 0) until the day of luteectomy. Each point represents the mean _+ SE. B) Serum PGFM profiles from cows expected to have normal- and short-lived CL from the day of hcg administration (Day 0) until the day of luteectomy. Each point represents the mean + SE. Days indicated with an asterisk (*) have different (p < 0.05) PGFM values between luteal life span groups. RESULTS A/ Five of the original cows expected to display short-lived CL (4 nonluteectomized and I Day 8 luteecetomized) were removed from the study because of high P 4 before hcg treatment. In the remaining 16 nonluteectomized cows, 1000 IU hcg induced ovulation and stimulated luteal development as expected. In nonluteectomized cows, normallived CL were induced in 9 of 10 cows (90%) that received norgestomet implants, with an average estrous cycle length of 18 days; in contrast, subnormal CL were induced in 5 of 6 cows (83%) that received placebo implants before hcg injection, while I cow had a luteal phase of normal duration yielding an average estrous cycle length of 11 days. In luteectomized cows, P 4 values were similar between cows expected to have normal-lived and short-lived CL from Day 0 until the day of CL removal (Fig. A,) indicating that CL were functionally similarly between normal and subnormal CL treatment groups during this sampling period. Serum PGFM values were not significantly different from Day 0 through Day 7 after hcg administration in luteectomized cows expected to have normal-lived or shortlived CL (Fig. B). However, on Day 8 after administration of hcg, cows expected to have short-lived CL had higher (p < 0.05) serum PGFM concentrations than cows expected to have normal-lived CL. The percentage of luteal tissue composed of vascular components, as visualized by immunohistochemical analysis with an anti-von Willebrand factor antibody, was de-

4 LH-R mrna IN SHORT-LIVED CL 905 TABLE 1. Concentrations of LH-R in CL removed on Day 6, 7, or 8 after hcg administration (Day 0) from cows expected to have normal- or shortlived CL.* Luteal LH-R concentrations (fmol/g CL tissue) CL Day 6 Day 7 Day 8 Normal-lived Short-lived _ * n = 5 for each luteal life span group on each day except n = 4 for short-lived on Day 8; values are expressed as mean SE concentrations based on luteal weight. FIG. 2. The degree of vascularity of CL removed on Day 6, 7, or 8 following hcg administration (Day 0) from cows expected to have normal- or short-lived CL. Corpora lutea (n = 5 for each luteal life span group on each day except for short-lived group on Day 8, for which n = 4) sections were assessed by immunohistochemical analysis with an antihuman von Willebrand factor antibody specific for endothelial cells. For each luteal sample, a single field ( m 2 ) per section with four sections per CL was analyzed. The degree of vascularity represents the mean percentage (%) SE of the set area assessed, by image analysis, that stained positive for von Willebrand factor. Columns with an asterisk (*) are different (p < 0.05). termined for each normal- and short-lived CL sample. The endothelial cell composition of CL expected to be either normal- or short-lived was not significantly different between life span groups on Day 6 and 7 after hcg administration (Fig. 2). However, CL collected on Day 8 from cows expected to exhibit subnormal luteal phases contained 33% less (p < 0.05) vascular components per luteal area analyzed than CL from cows expected to have normal luteal phases. The endothelial cell composition within a CL life span group did not significantly change between Days 6, 7, and 8. Concentrations of LH-R in normal- and short-lived CL were expressed based on wet luteal weight. The LH-R concentrations did not differ between CL expected to be normal-lived and short-lived on any day measured, and there was no significant difference in LH-R concentration within CL life span groups on different days sampled (Table 1). The 180-bp rat LH-R cdna used in these experiments hybridized to a transcript of approximately 4.4 kb, which was specifically expressed in bovine CL but not in bovine kidney, liver, or lung (Fig. 3). All cows, irrespective of CL life span and day of sampling, expressed an LH-R transcript at the 4.4-kb range (Fig. 4). When the LH-R transcripts were normalized with actin hybridization intensity, on an individual animal basis, it was found that there were no significant differences in LH-R mrna levels between cows expected to form normal-lived and short-lived CL on either Day 6 or 7 (Fig. 5). However, CL expected to have a shortened life span contained 48% less (p < 0.05) LH-R mrna on Day 8 than normal-lived CL. Comparison of actin mrna levels between normal-lived and short-lived CL groups, respectively, revealed no significant differences on Days 6 ( vs ; p = 0.90), 7 ( vs ; p = 0.5), or 8 ( vs ; p = 0.9). Over time, the expression of the luteal LH-R and actin genes did not significantly change within CL life span groups. FIG. 3. Northern blot analysis of LH-R mrna in various bovine tissues. Twenty micrograms of mrna from bovine CL (two different samples; lane A and B), kidney (lane C), liver (lane D), and lung (lane E) were submitted to electrophoresis, transferred to nitrocellulose membrane, and hybridized with a [32P]random-labeled rat LH-R cdna. Radiography was performed at -80 C for 14 h.

5 906 SMITH ET AL. FIG. 4. Northern blot composite of LH-R and actin mrna hybridizations in normal (NL) and short-lived (SL) luteal tissues collected on Days 6, 7, and 8 after hcg administration. Twenty micrograms of mrna per lane was submitted to electrophoresis, transferred to nitrocellulose membrane, and hybridized to a [32Plrandom-labeled rat LH-R cdna and a [32P]random-labeled human actin cdna. Radiography was performed at C for 2 h and 18 h to visualize actin and LH-R mrna bands, respectively. DISCUSSION In this report we have described endocrine, transcriptional, and histological events that occur in close approximation to the early demise of the bovine short-lived CL. The short-lived CL in the postpartum beef cow has been defined as having a life span of less than 14 days [8, 31], and when subnormal luteal formation is induced with hcg FIG. 5. Levels of LH-R mrna in CL removed on Day 6, 7, or 8 following hcg administration (Day 0) from cows expected to have normal- or shortlived CL. Corpora lutea (n = 5 for each luteal life span group on each day except for the short-lived group on Day 8, for which n = 4) were assessed for LH-R and actin mrna content by Northern blot analysis with 2 respective P-labeled cdnas followed by radiography and whole band densitometry. Values are reported as mean + SE actin-normalized LH-R mrna levels (arbitrary units). Columns with an asterisk (*) are different (p < 0.05). injection, the mean estrous cycle length was reported to be 8.1 days [7]. In the present study, 90% of nonluteectomized cows expected to form normal-lived CL had luteal phases longer than 14 days, whereas 83% of the nonluteectomized cows expected to form short-lived CL had luteal phases less than 14 days long. Although the interestrous intervals of luteectomized cows are not known, we expect the above values to be representative. Serum concentrations of P 4 were similar between luteectomized cows expected to have normal-lived and short-lived CL through Day 8, indicating that CL were functionally comparable throughout the sampling period. It has been speculated that the shortened life span of the bovine postpartum CL could be caused by 1) inadequate circulating concentrations of gonadotropins prior to ovulation, 2) inadequate luteotropic stimulus and/or responsiveness, and/or 3) a premature luteolysin release and/or increased responsiveness of luteal tissue to a luteolysin [46]. Considerable data are available to support the premature release of uterine PGF 2 as the factor responsible, at least in part, for the early demise of the short-lived CL [9-16]. Hysterectomy prevents regression of CL anticipated to have a shortened life span [16]; and it is believed that this uterine influence is mediated by PGF 2,, which has been found to be elevated from Days 4 through 9 following the first induced postpartum estrus in cows that exhibited short luteal phases [9]. Results from the present experiment are in agreement with the other reports mentioned in that serum concentrations of PGFM, the major stable PGF 2 metabolite, were significantly elevated on Day 8 in cows expected to form short-lived CL in comparison to cows expected to have normal-lived CL. However, serum P 4 concentrations were similar between the two CL life span groups at this sampling period. Therefore, the CL that were sampled in

6 LH-R mrna IN SHORT-LIVED CL 907 this experiment on Day 8 had been subjected to elevated levels of PGF 2. but did not show functional signs of luteal regression such as diminished serum P 4 concentrations. One of the first consequences of elevated PGF 2. on the CL is a reduction in blood supply [47]. Elevated PGF 2. between Days 4 through 9 in cows expected to have short luteal phases [9] could cause a decrease in the degree of luteal vascularity during this time period. On Days 6 and 7 after administration of hcg, when serum PGFM concentrations were similar between CL life span groups in the present study, the luteal endothelial cell composition was similar between cows expected to form normal- and shortlived CL. However, on Day 8 after hcg, when serum PGFM concentrations were elevated in cows expected to form subnormal CL, the mean luteal endothelial cell composition of short-lived CL was reduced by 33% compared to that of normal-lived CL. This difference in vascularity on Day 8 appears to be due to a lack of vascular increase in short-lived CL, an increase that appeared to occur in normal-lived CL. This reduced degree of vascularity in the short-lived CL represents one of the first detectable changes in luteal tissue after elevation of circulatory prostaglandins but prior to the serum P 4 decline. The responsiveness of bovine subnormal CL to LH, as determined by quantification of luteal LH-R concentrations, was found to be similar between normal- and short-lived CL on Day 5 after hcg administration and on Day 6.5 after GnRH administration [7, 30]. However, on Day 7 after GnRH, CL from cows expected to form subnormal CL were found to be less responsive to LH stimulation in vitro than normal CL [6]. With the average life span of the induced short-lived CL being approximately 8 days [7, 30], it is possible that sampling at Day 5 and Day 6.5 after hcg is too early, with respect to the beginning of the P 4 decline in the subnormal CL, for detection of a difference in LH-R concentration. In the present experiment, LH-R concentrations were similar between normal and subnormal CL on all days measured. These results are in agreement with reports on Day 5 post-hcg [7] and Day 6.5 post-gnrh [30] and indicate that an inability of the CL to respond to LH, as measured by luteal LH-R concentrations, does not exist on Day 6, 7, or 8 in CL anticipated to be subnormal. The possibility exists that the functionality of the LH-R may be different between normal- and short-lived CL. One mechanism by which this functionality could be altered is through impairment of the LH-R second messenger system. In proximity to the time of luteal regression in cattle, adenylate cyclase activity is reduced and phosphodiesterase activity increased in comparison to what occurs in the midluteal stage [48]. In sheep, the administration of PGF 2 has been shown to result in a rapid decrease in adenylate cyclase activity and an increase in phosphodiesterase activity that potentially could reduce the intracellular camp levels and impede the second messenger system [49]. However, the activity of these enzymes, adenylate cyclase and phosphodiesterase, has been reported to be similar between CL life span groups on Day 5 [7]. Whether this is also true on Days 6, 7, and 8 remains to be determined. This paper provides the first description of the LH-R transcript and steady-state LH-R mrna levels in cattle. The bovine LH-R transcript recognized by a rat LH-R [ 32 P]cDNA was visualized by Northern blot analysis at approximately 4.4 kb. The size of this transcript is in agreement with that described in the rat [26, 50] and is similar to that of a transcript reported in the sheep [51, 52]. In these previous studies, various LH-R signals have been reported that range from 1.8 to 7.0 kb. The inability of the polymerase chain reaction-generated [ 3 2 P]cDNA utilized in the present study to recognize more than one transcript may be due to 1) the portion of the transcript that is recognized by the [ 32 P]cDNA, 2) tissue- or estrous stage-specific expression of the LH-R gene, 3) differences in tissue processing and RNA isolation, or 4) a species difference. The [ 32 P]cDNA utilized in this experiment was generated against a portion of the transmembrane region of the LH-R, whereas that used in the rat [26, 50] was made against the extracellular portion of the receptor cdna and recognized multiple transcripts. Although the most abundant message was always visualized at 4.4 kb, one nonstudy animal sample had a faint signal present at approximately 6.3 kb. Recently, the influences of luteolytic doses of PGF2 on ovine [52, 53] and rat [54] luteal LH-R mrna content have been characterized. Luteal LH-R mrna content was shown to be significantly reduced within 4 h [53], 6 h [52], and 7 h [54] following PGF 2 administration. In the present study, LH-R mrna was detectible in all samples, irrespective of anticipated life span, and this transcript was present at 4.4 kb. When this 4.4-kb transcript was quantified using densitometry, and values were normalized with an actin hybridization signal from the same filters, the LH-R mrna values were not significantly different between normal and subfunctional CL on either Day 6 or 7 after hcg administration. However, steady-state levels of LH-R mrna in short-lived CL on Day 8 were reduced by 48% compared to those in normal-lived CL on the same day. Interesting are the findings that while LH-R mrna levels were reduced in animals with short-lived CL compared to normal-lived CL on Day 8, the steady-state levels of actin mrna remained constant. This indicates that preferential reduction in LH-R mrna occurs that is not due to a general decrease in all luteal transcripts. Since this decline in luteal LH-R mrna in short-lived CL occurred when serum PGFM concentrations were elevated, it appears that one mechanism by which PGF 2. elicits its luteolytic response in both sheep and cattle is through regulating the transcription of the LH-R gene and/or the degradation of its message. In conclusion, we have confirmed that a premature release of PGF 2,, as determined by measurement of its major stable metabolite PGFM, is associated with subnormal CL. At the time of elevated PGFM, but prior to functional luteolysis, cows expected to have short-lived CL had decreased luteal endothelial cell composition and LH-R mrna content. In addition, we have shown that luteal LH-R mrna levels decrease prior to luteal LH-R concentrations following elevation in PGFM. It therefore appears that both luteal LH-R mrna and the degree of luteal vascularity may be regulated acutely by PGF 2 in cattle. ACKNOWLEDGMENTS The authors would like to thank Dan Conrad of the WS.U. beef unit for making animals available for use; David deavila, Dr. Tom Geary, Dr. Tim Rozell, and Yalcin Yalvas for assistance in sample collection; Dr. Carol Linder for assistance with the molecular biology techniques; and Dr. Fuller W Bazer for supplying PGFM antiserum J57. We further thank Dr. Carrie Cosola-Smith for critical review of the manuscript. REFERENCES 1. dizerega GS, Hodgen GD. Luteal phase dysfunction infertility: A sequel to aberrant folliculogenesis. Fertil Steril 1981; 35: Wilks JW, Hodgen GD, Ross GT. Luteal-phase defects in the rhesus monkeys: the significance of serum FSH:LH ratio. J Clin Endocrinol & Metab 1976; 43:

7 908 SMITH ET AL. 3. Coleman DA, Dailey RA. Effect of repeated removal of large ovarian follicles and treatment with progestin on ovarian function in the ewe. Biol Reprod 1983; 29: Odde JD, Ward HS, Kiracofe GH, McKee RM, Kittok RJ. Short estrous cycles and associated serum progesterone levels in beef cows. Theriogenology 1980; 14: LaVoie V, Han AW, Foster DB, Moody EL. Suckling effect on estrus and blood plasma progesterone in postpartum beef cows. J Anim Sci 1981; 52: Kesler DJ, Weston PG, Pimentel CA, Troxel TR, Vincent DL, Hixon JE. Diminution of the in vitro response to luteinizing hormone by corpora lutea induced by gonadotropin releasing hormone treatment of postpartum suckled beef cows. J Anim Sci 1981; 53: Smith ME, Garverick HA, Youngquist RS, Zahler WL. Luteinizing hormone receptor concentrations, adenylate cyclase activity and phosphodiesterase activity of bovine corpora lutea: comparison of short and normal estrous cycles. Domest Anim Endocrinol 1986; 3: Pratt BR, Berardinelli JG, Stevens LP, Inskeep EK. Induced corpora lutea in the postpartum cow. I. Comparison of gonadotropin releasing hormone and human chorionic gonadotropin and effects of progesterone and estrogen. J Anim Sci 1982; 54: Cooper DA, Carver DA, Villeneuve P Silvia WJ, Inskeep EK. Effect of progestagen treatment on concentration of prostaglandins and oxytocin in plasma from the posterior vena cava of post-partum beef cows. J Reprod Fertil 1991; 91: Copelin JP, Smith ME Kiesler DH, Garverick HA. Effect of active immunization of pre-partum and post-partum cows against prostaglandin F-2a on lifespan and progesterone secretion of short-lived corpora lutea. J Reprod Fertil 1989; 87: Zollers WG Jr, Garverick HA, Smith ME Oxytocin-induced release of prostaglandin F 2. in postpartum beef cows: comparison of short versus normal luteal phases. Biol Reprod 1989; 41: Zollers WG Jr, Garverick HA, Youngquist RS, Ottobre JS, Silcox RW, Copelin JP, Smith ME In vitro secretion of prostaglandins from endometrium of postpartum beef cows expected to have short or normal luteal phases. Biol Reprod 1991; 44: Zollers WG Jr, Garverick HA, Smith ME Moffatt RJ, Salfen BE, Youngquist RS. Concentrations of progesterone and oxytocin receptors in endometrium of postpartum cows expected to have a short or normal oestrous cycle. J Reprod Fertil 1993; 97: Peters AT, Bosu WTK, Liptrap RM, Cummings E. Temporal changes in serum prostaglandin F-2a and oxytocin in dairy cows with short luteal phases after the first postpartum ovulation. Theriogenology 1989; 32: Troxel TR, Kesler DJ. Ability of indomethacin to alter prostaglandin metabolite concentrations and to enhance the function of corpora lutea induced in postpartum suckled beef cows. J Anim Sci 1984; 59: Copelin JP, Smith ME Garverick HA, Youngquist RS. Effect of the uterus on subnormal luteal function in anestrous beef cows. J Anim Sci 1987; 64: Behrman H, Grinwich DH, Hichens M, MacDonald GJ. Effect of hypophysectomy, prolactin and prostaglandin F2. on gonadotropin binding in vivo in the corpus luteum. Endocrinology 1978; 103: Fletcher PW, Niswender GD. Effect of PGF 2. on progesterone secretion and adenylate cyclase activity in ovine luteal tissue. Prostaglandins 1982; 20: Nett TM, McClellan MC, Niswender GD. Effects of prostaglandins on the ovine corpus luteum: blood flow, secretion of progesterone and morphology. Biol Reprod 1976; 15: Sawyer HR, Niswender KD, Braden TD, Niswender GD. Nuclear changes in ovine luteal cells in response to PGF 2,. Domest Anim Endocrinol 1991; 7: Hisaw FL. Development of the graafian follicle and ovulation. Physiol Rev 1947; 27: Espey LL. Ovulation as an inflammatory reaction-a hypothesis. Biol Reprod 1980; 22: Niswender GD, Nett TM. The corpus luteum and its control. In: Knobil E, Neill JD, Ewing LL, Greenwald GS, Markert CL, Pfaff DW (eds.), The Physiology of Reproduction, First Edition. New York: Raven Press, Ltd.; 1988: Gilman AG. G proteins: transducers of receptor-generated signals. Annu Rev Biochem 1987; 56: Birnbaumer L, Codina J, Mattera R, Cerione RA, Hildebrandt JD, Sunyer T, Rojas FJ, Caron MG, Lefkowitz RJ, Iyengar R. Regulation of hormone receptors and adenylyl cyclases by guanine nucleotide binding N proteins. Recent Prog Horm Res 1985; 41: MacFarland KC, Sprengel R, Phillips HS, Kohler M, Rosenblit N, Nikolics K, Segaloff DL, Seeburg PH. Lutropin-choriogonadotropin receptor: an unusual member of the G protein-coupled receptor family. Science 1989; 245: Loosefelt H, Misrahi M, Atger M, Salesse R, Vu Hai-Luu Thi MT, Jolivet A, Guiochon-Mantel A, Sar S, Jallal B, Garnier J, Milgrom E. Cloning and sequencing of porcine LH-hCG receptor cdna: variants lacking transmembrane domain. Science 1989; 245: Minegishi T, Nakamura K, Takakura Y, Miyamoto K, Hasegawa Y, Ibuki Y, Igarashi M. Cloning and sequencing of human LH/hCG receptor cdna. Biochem Biophys Res Commun 1990; 172: Segaloff DL, Ascoli M. The lutropin/choriogonadotropin receptor... 4 years later. Endocr Rev 1993; 14: Rutter LM, Carruthers TD, Manns JG. The postpartum corpus luteum: functional differences from that of cycling cows and the effect of progesterone pretreatment. Biol Reprod 1985; 33: Ramirez-Godinez JA, Kiracofe GH, McKee RM, Schalles RR, Kittock RJ. Reducing the incidence of short estrous cycles in beef cows with norgestomet. Theriogenology 1981; 15: Hu Y, Sanders JDH, Kurz SG, Ottobre JS, Day ML. In vitro prostaglandin production by corpora lutea destined to be normal or shortlived. Biol Reprod 1990; 42: Grieger DM, Scarborough R, deavila DM, Johnson HE, Reeves JJ. Active immunization of beef heifers against luteinizing hormone: III. Evaluation of dose and longevity. J Anim Sci 1990; 68: Fincher KB, Bazer FW, Hansen PJ, Thatcher WW, Roberts RM. Proteins secreted by the sheep conceptus suppress induction of uterine prostaglandin F-2a release by oestradiol and oxytocin. J Reprod Fertil 1986; 76: Mirando MA, Ott TL, Harney JP, Bazer FW. Ovine trophoblast protein-one inhibits development of endometrial responsiveness to oxytocin in ewes. Biol Reprod 1990; 43: Rodbard D. Simple RIA: programs for logit-log and four parameter logistic methods, with interactive re-weighting and for quality control analysis RIA 001, RIA 004 and QCRIA. In: Laboratory of Theoretical and Physical Biology, National Institute of Child Health and Human Development, National Institutes of Health; 1983; Bethesda, MD. 37. Meyer MD, Hansen PJ, Thatcher WW, Drost M, Badinga L. Extension of corpus luteum lifespan and reduction of uterine secretion of prostaglandin F 2, ~ of cows in response to recombinant interferon-r. J Dairy Sci 1995; 78: Smith GD, Appell LH, Reeves JJ. Transvaginal luteectomy. In: Dzuik P, Wheeler M (eds.), Handbook of Methods for Study of Reproductive Physiology in Domestic Animals. University of Illinois; 1992: Luteectomy section. 39. Weidner N, Semple JP, Welch WR, Folkman J. Tumor angiogenesis and metastasis-correlation in invasive breast carcinomas. New Engl J Med 1991; 324: Brown JL, Schoenemann HM, Reeves JJ. Effect of FSH treatment on LH and FSH receptors in chronic cystic-ovarian-disease dairy cows. J Anim Sci 1986; 62: Cusan L, Auclair L, Belanger A, Ferland L, Kelly PA, Seguin C, Labrie E Inhibitory effects of long term treatment with a luteinizing hormone-releasing hormone agonist on the pituitary-gonadal axis in male and female rats. Endocrinology 1979; 104: Chomczynski P, Sacchi N. Single-step method of RNA isolation by acid guanidinium thiocyanate-phenol-chloroform extraction. Anal Biochem 1987; 162: Jacobson A. Purification and fractionation of poly(a) + RNA. In: Berger SL, Kimmel AR (eds.), Methods in Enzymology. New York: Academic Press, Inc.; 1987: 152: Heckert LL, Griswold MD. Expression of follicle-stimulating hormone mrna in rat testes and Sertoli cells. Mol Endocrinol 1991; 5: Hanukoglu I, Tanese N, Fuchs E. Complementary DNA sequence of a human cytoplasmic actin. Interspecies divergence of 3' non-coding regions. J Mol Biol 1983; 163: Garverick HA, Zollers WG Jr, Smith ME Mechanisms associated with corpus luteum lifespan in animals having normal or subnormal luteal function. Anim Reprod Sci 1992; 28: Niswender GD, Reimers TJ, Diekman MA, Nett TM. Blood flow: a mediator of ovarian function. Biol Reprod 1976; 14: Garverick HA, Smith ME Elmore RG, Morehouse GL, Agudo LSp. Zahler WL. Changes and interrelationships among luteal LH recep-

8 LH-R mrna IN SHORT-LIVED CL 909 tors, adenylate cyclase activity and phosphodiesterase activity during the bovine estrous cycle. J Anim Sci 1985; 61: Agudo LSp, Zahler WL, Smith ME Effect of prostaglandin F 2 on the adenylate cyclase and phosphodiesterase activity of ovine corpora lutea. J Anim Sci 1984; 58: LaPolt PS, Oikawa M, Jia X, Dargan C, Hsueh AJW. Gonadotropininduced up- and down-regulation of rat ovarian LH receptor message levels during follicular growth, ovulation and luteinization. Endocrinology 1990; 126: Bacich DJ, Rohan RM, Norman RJ, Rodgers RJ. Characterization and relative abundance of alternatively spliced luteinizing hormone receptor messenger ribonucleic acid in the ovine ovary. Endocrinology 1994; 135: Guy MK, Juengel JL, Tandeski TR, Niswender GD. Steady-state concentrations of mrna encoding the receptor for luteinizing hormone during the estrous cycle and following prostaglandin F 20 treatment of ewes. Endocrine 1995; 3: Smith GW, Gentry PC, Smith MF, Roberts RM. Ontogeny of LH receptor mrna expression within ovine corpora lutea. Biol Reprod 1994; 49(suppl 1): Bjurulf E, Gunnar S. Rat luteinizing hormone receptor messenger ribonucleic acid expression and luteolysis: inhibition by prostaglandin F 2. Biol Reprod 1996; 54: