Survey
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
BIOLOGY OF REPRODUCTION 55, 996-1002 (1996) Effects of Growth Hormone and Pregnancy on Expression of Growth Hormone Receptor, Insulin-Like Growth Factor-I, and Insulin-Like Growth Factor Binding Protein-2 and -3 Genes in Bovine Uterus, Ovary, and Oviduct' Crystal J. Kirby,3 William W. Thatcher, 4 Robert . Collier,5 Frank A. Simmen, 4 and Matthew C. Lucy2 ,3 Department of Animal Sciences, 3 University of Missouri, Columbia, Missouri 65211 Department of Dairy and Poultry Sciences, 4 University of Florida, Gainesville, Florida 3261 1 Monsanto Company,s St. Louis, Missouri 63198 ABSTRACT The effects of growth hormone (GH) and pregnancy on insulin-like growth factor (IGF)-I, IGF binding protein (IGFBP)-2, and IGFBP-3 mRNA in reproductive tissues were studied in cattle. Lactating dairy cows were inseminated at estrus and treated with 25 mg/day GH (n = 8) or saline (n = 8) for 16 days. Corpus luteum (CL), ovary (CL removed), oviduct, endometrium, and myometrium were collected at the end of treatment. Messenger RNA for GH receptor, IGF-I, IGFBP-2, IGFBP-3, and actin were measured by nuclease protection assays. The CL contained more GH receptor mRNA than the other reproductive tissues examined. Expression of IGF-I mRNA was highest in myometrium, with lower amounts found in endometrium; the CL expressed the least amount of IGF-I mRNA. The IGFBP-2 mRNA was most abundant in endometrium and least abundant in CL. Expression of IGFBP-3 mRNA was detected in all reproductive tissues examined. However, endometrium, a tissue that expressed the most IGFBP-2 mRNA, had the lowest amount of IGFBP-3 mRNA. The GH receptor mRNA was decreased in cows treated with GH whereas the mRNA for IGF-I, IGFBP-2, or IGFBP-3 was not changed. In the reproductive tissues evaluated, cows that contained a conceptus at tissue collection (pregnant) had higher amounts of IGF-I mRNA than did nonpregnant cows. In summary, the level of mRNA encoding GH receptor, IGF-I, IGFBP-2, and IGFBP-3 varied within the tissues examined, suggesting that these genes may play a variety of roles in the bovine female reproductive tract. Supplemental GH failed to change the expression of IGF-I, IGFBP-2, and IGFBP-3 mRNA, possibly because of low GH receptor mRNA levels in tissues other than CL. A direct action of GH on IGF-I, IGFBP-2, or IGFBP-3 gene expression within cow reproductive tissues was not supported because the amount of IGF-I, IGFBP-2, or IGFBP-3 mRNA was not altered by GH. INTRODUCTION Components of the insulin-like growth factor (IGF) axis play an important role in follicular growth and embryonic development [1-5]. Most reproductive tissues are capable of IGF-I synthesis, and receptors for IGF-I are expressed in the ovary, uterus, and embryo [1-11]. Therefore, ovarian and uterine function, and embryonic growth and differentiation, may be dependent partially on local (autocrine) and systemic (endocrine) IGF-I. As an example, ovarian folli- cles synthesize IGF-I and IGF-II [1, 2, 5, 11], and the synergistic actions of IGF, LH, and FSH are involved in follicular growth [1]. In the uterus, elongating pig blastocysts secrete estradiol that increases IGF-I secretion into the uterine lumen [7, 12]. Unlike the pig, in the cow, endometrial IGF-I mRNA abundance is not increased during early pregnancy [8]. Additional control of reproductive function may occur through the actions of IGF binding proteins (IGFBP) [13]. These IGFBP are found within the ovary and uterus and can modulate IGF action [1, 2, 5, 9, 13]. Little is known about the endocrine mechanisms that control the local production of IGF-I, IGFBP-2, or IGFBP-3 within the female reproductive system. One possibility is that the local production of IGF-I, IGFBP-2, and IGFBP-3 in the ovary and uterus is controlled by growth hormone (GH). GH receptors are found in the bovine corpus luteum (CL) [14], and an alternate GH receptor (exon I variant) is found in CL and endometrium [15, 16]. In the rat, GH receptors occur within all parts of the reproductive tract, but receptor levels vary across different tissues [17]. When GH was administered to laboratory animals, IGF-I mRNA was increased in hepatic tissue (major site of GH receptor mRNA) [18, 19], and it may be increased similarly in reproductive tissues. Furthermore, production of IGFBP-2 or IGFBP-3 in reproductive tissues may be altered in response to GH or IGF-I. In the present study, tissues were identified that express GH receptor, IGF-I, IGFBP-2, and IGFBP-3 mRNA in the cow reproductive tract on Day 17 of the estrous cycle. At the same time, the expression of IGF-I, IGFBP-2, and IGFBP-3 in the ovary and uterus after the administration of either GH or saline (control) was evaluated. This was done to determine whether GH controlled transcription of IGF-I, IGFBP-2, or IGFBP-3 mRNA within several cow reproductive tissues. The GH receptor, IGF-I, IGFBP-2, and IGFBP-3 mRNA also was evaluated in pregnant compared to nonpregnant cattle. MATERIALS AND METHODS Animals and Treatments Sixteen lactating Holstein dairy cows (> 250 days of lactation) from the University of Florida Dairy Research Unit at Hague, Florida, were used. Cows were housed outdoors in concrete lots with access to dirt exercise yards and were fed diets balanced nutritionally according to National Research Council [20] guidelines. Estrous cycles were synchronized by using a combination of norgestomet ear implant (Synchromate-B implant; Sanofi Animal Health Inc., Overland Park, KS; inserted for 6 days) and prostaglandin F 2. injection (25 mg/cow; Lutalyse, Pharmacia and Upjohn Inc., Kalamazoo, MI; administered 5 days after norgestomet implant). Cows were inseminated artificially at 12 and Accepted June 20, 1996. Received November 16, 1995. 'Contribution from the Missouri Agricultural Experiment Station, Journal Series Number 12422. This project was partially supported by the USDA under the National Research Initiative Competition Grants Program. Grant number 92-37203-8341 awarded to M.C.L. 2Correspondence: Matthew C.Lucy, Department of Animal Sciences, 164 Animal Science Research Center, University of Missouri, Columbia, MO 65211. FAX: (573) 882-6827; e-mail: [email protected] 996 GROWTH HORMONE, IGF, AND REPRODUCTION 997 24 h after the first behavioral sign of estrus (cow remained motionless during a mount by a herd-mate) with semen from the same bull. Beginning on the day of behavioral estrus, cows were treated daily with either 25 mg methionyl bovine GH (sometribove; Monsanto Company, St. Louis, MO; n = 8 cows) or saline (SAL; n = 8 cows). Daily injections of either GH or SAL continued for the next 16 days. TABLE 1. Bovine cDNA and subcloning vectors used for the production of antisense ribonucleotide probes for analyses of bovine mRNA.* Ovary and Embryo Removal * Size of the protected fragment was shorter than the ribonucleotide probe because of RNA synthesis from the polymerase start site and additional polylinker sequences. On Day 17 after estrus and insemination, the ovaries and uterus were collected from each cow after cows were killed by captive bolt stunning and exsanguination. Time of slaughter was approximately 24 h after the last injection of GH or SAL. Tissues were immersed in ice and brought to a laboratory next to the slaughter facility. Ovaries were kept on ice, and the CL was dissected free from surrounding ovarian stroma. None of the CL had undergone luteolysis before slaughter. This was confirmed by plasma progesterone analyses that showed progesterone concentration to be at least 6.6 ng/ml [21]. Embryos were flushed from the tip of the uterine horn ipsilateral to the ovary bearing the CL by inserting 40 ml of a mixture (1:1) of F-12 Ham and Dulbecco's Modified Eagle's Medium (Sigma Chemical Company, St. Louis, MO) into the contralateral uterine horn at the uterotubal junction. Flushing medium was massaged gently through the uterine horn and the uterine body to the ipsilateral uterine horn to recover the embryo. A second flush (40 ml) was done if no embryo was recovered in the first flush. The cow was classified as nonpregnant if no embryo was found in the second flush. On Day 17 of pregnancy, six embryos were recovered from cows treated with GH (75% conception rate), and five embryos were collected from cows treated with SAL (63% conception rate). Processing of Reproductive Tissues After the embryos were collected, reproductive tissues (CL, ovary ipsilateral to the CL [luteal tissue removed], ovary contralateral to the CL, ipsilateral oviduct, contralateral oviduct, ipsilateral endometrium, contralateral endometrium, ipsilateral myometrium, and contralateral myometrium) were placed in a plastic bag, immersed in liquid nitrogen until frozen (approximately 5 min), and stored at -80°C. The total time from animal slaughter to liquid nitrogen freezing was 1 h. Total cellular RNA was extracted from all tissues. Ipsilateral and contralateral tissues were combined after preliminary analyses indicated no effect of location (ipsilateral versus contralateral). Frozen tissue was removed from -80°C storage, placed in a mortar, covered with liquid nitrogen, and crushed to a fine powder with a pestle. The powder was then transferred to a hand-held homogenizer and dissociated in 4 M guanidinium thiocyanate (Sigma). Total cellular RNA was isolated according to Chomczynski and Sacchi [22]. Cellular RNA that had 28S and 18S ribosomal bands readily visible under ultraviolet illumination after electrophoresis and ethidium bromide staining was classified as intact and used in subsequent analyses. Purified RNA was dissolved in water, and concentration was determined from an A260 measurement. Storage was at -80°C. Analysis of RNA by Nuclease Protection Assay Subcloning of bovine cDNA fragments into plasmid vectors for production of complementary ribonucleotide probes Bovine gene Actin GH Receptor IGF-I IGFBP-2 IGFBP-3 Vector pGEM pGEM pGEM pGEM pGEM 3Z Blue 3Z 3Z 3Z Riboprobe length 137 498 86 126 177 bp bp bp bp bp Protected length 80 460 63 108 136 bp bp bp bp bp Reference [23] [231 [231 [24] [25] has been described previously for GH receptor, IGF-I, and actin [23]. For the GH receptor, the ribonucleotide probe was specific for the extracellular domain. Therefore, the probe was specific for both the GH receptor and the GH binding protein mRNA. These are not processed as separate mRNA in the cow [23]. For IGF-I, the probe was specific for the "A" domain of IGF-I protein and should detect all of the IGF-I transcripts [19]. Additional plasmids were constructed from bovine cDNA for IGFBP-2 and IGFBP-3. One hundred-eight base pairs (bp) of the IGFBP-2 and 136 bp of the IGFBP-3 cDNA were amplified by polymerase chain reaction (PCR). Primers for the PCR were designed to amplify cDNA nucleotides 1084-1192 of IGFBP-2 [24] and 845-981 of IGFBP-3 [25]. Amplified cDNA was subcloned (Table 1), and orientation and integrity were verified by DNA sequencing. Before their use for ribonucleotide probe synthesis, plasmids were linearized by digestion with appropriate restriction enzymes, extracted with a 1:1 mixture of phenol and chloroform, precipitated with 0.1 volume 3 M sodium acetate (pH 5.2) and two volumes absolute ethanol, dissolved in water, and stored at -20°C. Antisense ribonucleotide probes were generated by using a Riboprobe Gemini II Core system (Promega Corporation, Madison, WI). Two hundred nanograms of linearized plasmid were incubated with either SP6 or T7 polymerase (Promega), [ 32 P]-rCTP (New England Nuclear, Boston, MA), and appropriate buffers to yield a ribonucleotide probe that was antisense to the specific mRNA. Before use in nuclease protection assays, ribonucleotide probes were extracted with phenol:chloroform (1:1) and then chloroform. Unincorporated [3 2 P]-rCTP was removed by centrifugation through a G50 Sephadex spin column (Boehringer Mannheim, Indianapolis, IN). Ribonuclease protection assays [26] were done on 20 ug of total cellular RNA by using the RPA II kit (Ambion Inc., Austin, TX). Protected mRNA fragments were identified by their electrophoretic mobility through 8% acrylamide, 8 M urea gels (Acryl-A-Mix 8, Promega). Gels were dried and autoradiography was done by using X-OMAT-AR film (Eastman Kodak, Rochester, NY) for 48 h (GH receptor, actin, IGFBP-2, and IGFBP-3) or six days (IGF-I) at - 80C with intensifying screens. Size of the protected fragment was shorter than the ribonucleotide probe because of RNA synthesis from the polymerase start site and additional polylinker sequences (Table 1). The difference in probe length and protected fragment length was verified by counting single bases on a nucleotide ladder from the full-length probe to the protected fragment. Relative amounts of [ 32 P]-labeled probe hybridized to mRNA were measured by using GP Tools, version 3.0 (BioPhotonics Corporation, Ann Arbor, MI). 998 KIRBY ET AL. FIG. 1. Expression of GH receptor mRNA in reproductive tissues (CL, ovary [Ov; minus luteal tissue], oviduct [Ovid]; endometrium [Endol, and myometrium [Myo]). Samples were collected on Day 17 after estrus from cows treated with either GH (n = 8 samples/tissue) or SAL (n = 8 samples/ tissue) from Days 0 to 16 after estrus. A) Means for cows treated with GH or SAL. Superscripts on x-axis label denote differences in mRNA expression among tissues (tested by Duncan's Multiple Range test; tissues with different letters are not the same with p < 0.05). B)Representative autoradiograph from nuclease protection assay of 2 of 16 cows analyzed. Probe, undigested probe; tRNA, negative control. Statistical Analysis Eighty samples of RNA (five tissues from each of sixteen cows) were assayed by nuclease protection. The tissue means were tested for homogeneity of variance by using a test initially described by Bartlett [27]. Variances of tissue means for actin mRNA were homogeneous. However, variances of tissue means for GH receptor, IGF-I, IGFBP-2, and IGFBP-3 were heterogeneous. Therefore, these data were logo1 -transformed. After transformation, the variances of tissue means were homogeneous for GH receptor, IGF-I, and IGFBP-2. However, variance for IGFBP-3 remained heterogeneous because of smaller myometrium tissue variance. Alternative transformations (natural log and square root) failed to reduce the heterogeneity of variance. Therefore, the IGFBP-3 data were analyzed without myometrium. Data were analyzed by least squares analysis of variance by using the Statistical Analysis System (SAS; [28]). The statistical model for mRNA (actin) or loglo mRNA (GH receptor, IGF-I, IGFBP-2, and IGFBP-3) analyses was Yijkl = I. + Ai + Bj + ABp + C(AB)ijk + Dl + ADil + BDjI + ABDijl + eijkl, where Yijkl = dependent observation, I. = overall mean, A i = effect of treatment i (GH or SAL), Bj = effect of pregnancy status j (pregnant or nonpregnant), ABij = treatment-by-pregnancy interaction, C(AB)ijk = effect of animal k nested within treatment i and pregnancy j, D = effect of tissue 1 (CL, ovary, oviduct, endometrium, myometrium), ADil = treatment-by-tissue interaction, BDjl = pregnancy-by-tissue interaction, ABDij = treatment-bypregnancy-tissue interaction, and eijkl = residual. A i, Bj, and ABij were tested by using C(AB)ijk. Remaining terms were tested with the residual. Tissue means were separated by Duncan's multiple comparison procedures of SAS. For clarity, actual means of the untransformed data are presented in figures. Standard errors are not presented for GH receptor, IGF-I, IGFBP-2, and IGFBP-3 because tests of significance were based on log 10-transformed data. Means for log 1o-transformed data for IGFBP-3 within myometrium were not heterogeneous. Therefore, a single analysis of loglo myometrial mRNA for IGFBP-3 was done by using a statistical model without tissue or tissue interactions. RESULTS Statistical Interactions of Treatment, Pregnancy, and Tissue Source The solutions to the statistical model for GH receptor, IGF-I, IGFBP-2, and IGFBP-3 each showed a main effect of tissue. For some genes, an effect of pregnancy or treatment was detected. The effects of treatment by pregnancy, treatment by tissue, pregnancy by tissue, and treatment by pregnancy by tissue were not significant (p > 0.10) for actin, GH receptor, IGF-I, IGFBP-2, and IGFBP-3. Therefore, only the main effects of treatment, pregnancy, and tissue are presented in the subsequent sections. Actin Actin was measured in all RNA samples. Tissue means for actin mRNA tended to be different (p < 0.07). Mean actin was (arbitrary units) 17.5 + 1.8, 18.6 ± 1.7, 15.3 + 1.6, 17.4 + 1.5, and 16.4 + 1.3 for CL, ovary, oviduct, endometrium, and myometrium, respectively. Treatment and pregnancy were not significant (p > 0.10). GH Receptor There was an effect of tissue on GH receptor mRNA (p < 0.001). The mRNA level for GH receptor was higher in the CL compared to the ovary (CL removed), oviduct, endometrium, or myometrium (Fig. 1). Expression of the GH receptor mRNA within endometrium and myometrium was similar, and these tissues contained more mRNA for the GH receptor than ovary or oviduct. Treatment with GH GROWTH HORMONE, IGF, AND REPRODUCTION 999 tended (p < 0.07; Fig. 1A) to decrease GH receptor mRNA in all tissues. Pregnancy had no effect on GH receptor mRNA expression (p > 0.10). IGF-I There was an effect of tissue on IGF-I mRNA (p < 0.001). Highest expression of IGF-I mRNA was found within the myometrium (Fig. 2). Endometrium contained less IGF-I mRNA than did myometrium but more IGF-I mRNA than did the oviduct, ovary, or CL. The lowest amounts of IGF-I mRNA were in the CL. There was no effect of treatment on IGF-I mRNA within reproductive tissues (p > 0.10; Fig. 2A). Tissues from pregnant animals had higher amounts of IGF-I mRNA (p < 0.003; Fig. 2B) compared to nonpregnant animals on estrous cycle Day 17. IGFBP-2 There was an effect of tissue for IGFBP-2 mRNA (p < 0.001). Messenger RNA for IGFBP-2 was detected in highest amounts in endometrium (Fig. 3), with the least amounts in CL. The ovary, oviduct, and myometrium had intermediate amounts of mRNA for IGFBP-2. There was no effect of treatment (p > 0.10; Fig. 3A) or pregnancy (p > 0.10) on the amount of IGFBP-2 mRNA. IGFBP-3 The tissue means for IGFBP-3 mRNA were heterogeneous for variance. This was caused by the smaller variance in myometrium. Removal of myometrium from the analyses led to homogeneous variance. Therefore, myometrium was removed from statistical analyses of other tissues. There was an effect of tissue (p < 0.001) for expression of IGFBP-3 mRNA (Fig. 4). Ovary contained the highest amount of IGFBP-3 mRNA. Endometrium contained the least amount of IGFBP-3 mRNA. There was no effect of treatment (p > 0.10; Fig. 4A) or pregnancy (p > 0.10) on the amount of IGFBP-3 mRNA. The IGFBP-3 mRNA in myometrium was analyzed separately from that in other tissues. There was an effect of pregnancy (p < 0.035) on myometrial IGFBP-3 mRNA: The level of IGFBP-3 mRNA was higher in pregnant (18.9 ± 1.4) compared to nonpregnant (13.2 ± 2.2) cows. DISCUSSION The cow reproductive tract expressed GH receptor, IGF-I, IGFBP-2, and IGFBP-3 mRNA in a tissue-specific manner (effect of tissue, p < 0.001, for each mRNA). Although the expression of IGF-I and IGFBP-2 has been compared within the ruminant ovary and within the uterus [8, 10, 11], no studies have compared ovarian, oviductal, and uterine tissue expression of mRNA. The results of the present study demonstrated diverse expression of GH receptor, IGF-I, IGFBP-2, and IGFBP-3 mRNA in the tissues that we tested. Actin mRNA was measured as a marker for mRNA extraction, measurement, and loading. There was a tendency (p < 0.07) for actin mRNA amount to differ among tissues. However, the difference in actin mRNA between tissues (1.2-fold difference) was minor compared to the differences in GH receptor, IGF-I, IGFBP-2, or IGFBP-3 mRNA among tissues (5- to 10-fold difference). The differences in mRNA expression for GH receptor, IGF-I, IGFBP-2, and IGFBP-3 mRNA among individual tissues suggest roles of variable importance for GH, IGF-I, IGFBP-2, and IGFBP-3 in the ovary, oviduct, and uterus FIG. 2. Expression of IGF-I mRNA in reproductive tissues (CL, ovary [Ov; minus luteal tissue], oviduct [Ovid]; endometrium [Endo], and myometrium [Myol). Samples were collected on Day 17 after estrus from cows treated with either GH (n = 8 samples/tissue) or SAL (n = 8 samples/tissue) from Days 0 to 16 after estrus. Bars represent means for (A) cows treated with GH or SAL or (B)cows that were either pregnant (n = 11 samples/tissue) or nonpregnant (n = 5 samples/tissue) at the time of tissue collection. There was a significant effect of pregnancy (p < 0.003) and tissue (p < 0.001). Superscripts on x-axis label denote differences in mRNA expression among tissues (tested by Duncan's Multiple Range test; tissues with different letters are not the same with p < 0.05). C) Representative autoradiograph from nuclease protection assay of 2 pregnant cows of 16 cows analyzed. Probe, undigested probe; tRNA, negative control. 1000 KIRBY ET AL. FIG. 4. Expression of IGFBP-3 in reproductive tissues (CL, ovary [Ov; FIG. 3. Expression of IGFBP-2 in reproductive tissues (CL, ovary [Ov; minus luteal tissue], oviduct [Ovid]; endometrium [Endo], and myometrium [Myol). Samples were collected on Day 17 after estrus from cows treated with either GH (n = 8 samples/tissue) or SAL (n = 8 samples/ tissue) from Days 0 to 16 after estrus. A)Means for cows treated with GH or SAL. Superscripts on x-axis label denote differences in mRNA expression among tissues (tested by Duncan's Multiple Range test; tissues with different letters are not the same, with p < 0.05). B) Representative autoradiograph from nuclease protection assay of 2 of 16 cows analyzed. Probe, undigested probe; tRNA, negative control. of cattle. These data represent gene expression on a single day of the estrous cycle (Day 17; late luteal phase, before CL regression). In previous studies, the mRNAs for IGF-I and IGFBP-2 in reproductive tissues were similar during the luteal phase [8, 10, 11], but IGF-I mRNA changed at estrus [10, 29]. Therefore, extrapolation of these data to nonluteal phases of the estrous cycle may not be appropriate. An increase in IGF-I gene expression in response to GH supplementation was not observed in the present study. The change failed to occur despite a more than 2-fold increase in IGF-I protein in follicular fluid of these cows [21]. In a separate study of mice, hepatic IGF-I mRNA was increased in response to GH whereas IGF-I mRNA in nonhepatic tissues was unaffected [18]; GH also failed to increase minus luteal tissue], oviduct [Ovid]; endometrium [Endo], and myometrium [Myol). Samples were collected on Day 17 after estrus from cows treated with either GH (n = 8 samples/tissue) or SAL (n = 8 samples/ tissue) from Days 0 to 16 after estrus. A)Means for cows treated with GH or SAL. Superscripts on x-axis label denote differences in mRNA expression among tissues (tested by Duncan's Multiple Range test; tissues with different letter are not the same, with p < 0.05). B)Representative autoradiograph from nuclease protection assay for both IGFBP-3 and actin from 2 of 16 cows analyzed. Probe, undigested probe; tRNA, negative control. IGF-I in the ovine ovary [30]. Perhaps the low level of GH receptor mRNA expression (2-22% of liver; Lucy, unpublished data) in ovary, oviduct, and uterus is insufficient to yield measurable increases in IGF-I mRNA in response to GH. Alternatively, the GH signal transduction system, which stimulates IGF-I release in liver [18, 19], may not be present in other tissues. The low expression of IGF-I mRNA in CL compared to endometrium or myometrium shows that the CL is not a major source of IGF-I in the cow. A similar observation was made in sheep, in which luteal expression of IGF-I mRNA was low and did not change during the estrous cycle [11]. The GH receptor mRNA was found in all reproductive tissues that we tested; CL expressed highest levels. We previously observed that the CL expressed the highest level of GH receptor protein [14]. This supports the hypothesis that GH (or possibly a related molecule, placental lactogen [31 ]) is involved in CL development or function. Endometrium and myometrium had equivalent GH receptor expression, GROWTH HORMONE, IGF, AND REPRODUCTION whereas oviduct and ovary had lower levels of this mRNA. Treatment of cattle with GH tended to decrease the amounts of GH receptor mRNA. The decrease suggests that high concentrations of GH, found in GH-treated cattle, may cause a decrease in GH receptor mRNA expression. Similar down-regulation of receptor mRNA was observed in other systems in which hormones exceeded normal physiological concentrations [32-34]. Although we had a limited number of samples, we found that pregnant cows had higher IGF-I mRNA levels in reproductive tissues. Geisert et al. [8] examined endometrial IGF-I mRNA expression on different days of pregnancy in cattle and concluded that IGF-I mRNA expression was not increased by the conceptus. An alternative hypothesis is that cows with more IGF-I in either reproductive tissue or blood may be most likely to support early embryonic growth and establish pregnancy. Although direct evidence for this idea is not available, it is supported conceptually by the general link between higher concentrations of blood IGF-I and increased fertility in cows and humans [2, 4, 5, 35]. Neither IGFBP-2 nor IGFBP-3 mRNA was affected by either treatment or pregnancy. However, mRNA expression in the various tissues was different for IGFBP-2 and IGFBP-3. For example, CL, a tissue that expressed the least IGFBP-2 mRNA, was intermediate among tissues for IGFBP-3 mRNA, whereas IGFBP-2 mRNA predominated in endometrium and IGFBP-3 mRNA predominated in myometrium. This is consistent with pig data, in which the IGFBP-2 mRNA level was higher in endometrium than in myometrium [9]. These tissue differences may reflect different roles played by IGFBP-2 and IGFBP-3 in reproductive function. There were clear differences among tissues for the genes analyzed. However, the technique used (whole tissue homogenization and nuclease protection analyses) did not address potential cellular changes that might have occurred for these genes. This is particularly true for ovary because differences among cows for numbers and sizes of follicles may have increased the variability in mRNA expression. Therefore, the sensitivity of our analyses may have been increased if in situ hybridizations had been used instead of nuclease protection analyses. A second concern is the accuracy of mRNA analyses for the prediction of protein expression within these tissues. Some variation probably exists in the actual relationship between mRNA levels and protein synthesis within specific tissues. One additional approach would be to examine the protein expression for these genes by immunocytochemistry. This would localize the protein to specific cells. It may not be appropriate, however, to examine protein in either follicular or uterine fluids since proteins may arise at least in part from hepatic synthesis. Therefore, measurement of mRNA may be the most reliable method for detecting potential differences in local tissue protein production. Finally, it is possible that collection of mRNA on Day 17 may have compromised the treatment effects because of GH receptor down-regulation. Perhaps collection of tissues at shorter intervals after GH treatment would have demonstrated a biphasic response in which genes were initially induced by GH. This should be investigated in subsequent studies. In summary, mRNA amount for GH receptor, IGF-I, IGFBP-2, and IGFBP-3 was affected in a tissue-specific fashion within the cow reproductive tract. Treatment with GH tended to decrease GH receptor mRNA in reproductive tissues but failed to up-regulate IGF-I, IGFBP-2, or 1001 IGFBP-3 mRNA. Pregnant cows appear to express elevated levels of IGF-I mRNA within reproductive tissues. These data suggest that GH does not directly stimulate the IGF system in the reproductive tissues examined. ACKNOWLEDGMENTS The authors would like to express their appreciation to J. Savio, G. Danet-Desnoyers, M.T. Moser, M.D. Meyer, and R.L. De La Sota of the Dairy and Poultry Sciences Department, University of Florida, Gainesville, for assisting during embryo collection and processing of tissues. REFERENCES 1. Hammond JM, Mondschein JS, Samaras SE, Canning SE The ovarian insulin-like growth factors, a local amplification mechanism for steroidogenesis and hormone action. J Steroid Biochem Mol Biol 1991; 40:411-416. 2. Giudice LC. Insulin-like growth factors and ovarian follicular development. Endocr Rev 1992; 13:641-669. 3. Simmen RCM, Ko Y, Simmen FA. Insulin-like growth factors and blastocyst development. Theriogenology 1993; 39:163-175. 4. Hammond JM, Samaras S, Grimes R, Hagen D, Guthrie D. Ovarian IGF system: interface with the gonadotropic and somatotrophic axes. In: Adashi EY, Thorner MO (eds.), The Somatotrophic Axis and the Reproductive Process in Health and Disease. New York: SpringerVerlag; 1995: 183-201. 5. Spicer LJ, Echternkamp SE. The ovarian insulin and insulin-like growth factor system with an emphasis on domestic animals. Domest Anim Endocrinol 1995; 12:223-245. 6. Watson AJ, Hogan A, Hahnel A, Wiemer KE, Shultz GA. Expression of growth factor ligand and receptor genes in the preimplantation bovine embryo. Mol Reprod Dev 1992; 31:87-95. 7. Green ML, Simmen RCM, Simmen E Developmental regulation of steroidogenic enzyme gene expression in the periimplantation porcine conceptus: a paracrine role for insulin-like growth factor-I. Endocrinology 1995; 136:3961-3970. 8. Geisert RD, Lee CY, Simmen FA, Zavy MT, Fliss AE, Bazer FW, Simmen RCM. Expression of messenger RNAs encoding insulin-like growth factor -I, -II, and insulin-like growth factor binding protein-2 in bovine endometrium during the estrous cycle and early pregnancy. Biol Reprod 1991; 45:975-983. 9. Simmen FA, Simmen RCM, Geisert RD, Martinat-Botte F, Bazer FW, Terqui M. Differential expression, during the estrous cycle and preand postimplantation conceptus development, of messenger ribonucleic acids encoding components of the pig uterine insulin-like growth factor system. Endocrinology 1992; 130:1547-1556. 10. Stevenson KR, Gilmour RS, Wathes DC. Localization of insulin-like growth factor-I (IGF-I) and -II messenger ribonucleic acid and type 1 IGF receptors in the ovine uterus during the estrous cycle and early pregnancy. Endocrinology 1994; 134:1655-1664. 11. Perks CM, Denning-Kendall PA, Gilmour RS, Wathes DC. Localization of messenger ribonucleic acids for insulin-like growth factor I (IGF-I), IGF-II, and the type 1 IGF receptor in the ovine ovary throughout the estrous cycle. Endocrinology 1995; 136:5266-5273. 12. Simmen RCM, Simmen FA, Hofig A, Farmer SJ, Bazer FW. Hormonal regulation of insulin-like growth factor gene expression in pig uterus. Endocrinology 1990; 127:2166-2174. 13. Jones JI, Clemmons DR. Insulin-like growth factors and their binding proteins: biological actions. Endocr Rev 1995; 16:3-34. 14. Lucy MC, Collier RJ, Kitchell ML, Dibner JJ, Hauser SD, Krivi GG. Immunohistochemical and nucleic acid analysis of growth hormone receptor populations in the bovine ovary. Biol Reprod 1993; 48:12191227. 15. Lucy MC, Heap D, Collier RJ, Boyd CK. Expression of alternate growth hormone receptor messenger RNA in endometrium and corpus luteum of cattle. J Anim Sci 1995; 73(suppl 1):219 (abstract 437). 16. Pratt SL, Anthony RV. The growth hormone receptor messenger ribonucleic acid present in ovine fetal tissue is a variant form. Endocrinology 1995; 136:2150-2155. 17. Lobie PE, Breipohl W, Aragon JG, Waters MJ. Cellular localization of the growth hormone receptor/binding protein in the male and female reproductive systems. Endocrinology 1990; 126:2214-2221. 18. Mathews LS, Norstedt G, Palmiter RD. Regulation of insulin-like growth factor I gene expression by growth hormone. Proc Natl Acad Sci USA 1989; 83:9343-9347. 1002 KIRBY ET AL. 19. Bichell DP, Kikuchi K, Rotwein P. Growth hormone rapidly activates insulin-like growth factor I gene transcription in vivo. Mol Endocrinol 1992; 6:1899-1908. 20. National Research Council. Nutrient Requirements of Dairy Cattle, 6th rev. ed. Washington, DC: Natl. Acad. Sci.; 1989. 21. Lucy MC, Thatcher WW, Collier RJ, Simmen FA, Ko Y, Savio JD, Badinga L. Effects of somatotropin on the conceptus, uterus and ovary during maternal recognition of pregnancy in cattle. Domest Anim Endocrinol 1995; 12:73-82. 22. Chomczynski P, Sacchi N. Single-step method of RNA isolation by acid guanidinium thiocyanate-phenol-chloroform extraction. Anal Biochem 1987; 162:156-159. 23. Hauser SD, McGrath ME Collier RJ, Krivi GG. Cloning and in vivo expression of bovine growth hormone receptor mRNA. Mol Cell Endocrinol 1990; 72:187-200. 24. Bourner MJ, Busby WH, Siegel NR, Krivi GG, McCusker RH, Clemmons DR. Cloning and sequence determination of bovine insulin-like growth factor binding protein-2 (IGFBP-2): comparison of its structural and functional properties with IGFBP-1. J Cell Biochem 1992; 48:215-226. 25. Spratt SK, Tatsuno GP, Sommer A. Cloning and characterization of bovine insulin-like growth factor binding protein-3 (bIGFBP-3). Biochem Biophys Res Commun 1991; 177:1025-1032. 26. Lee JJ, Costlow NA. A molecular titration assay to measure transcript prevalence levels. Methods Enzymol 1987; 152:633-648. 27. Snedecor GW, Cochran WG. One-way classifications: analysis of variance. In: Statistical Methods, 8th ed. Ames, IA: Iowa State University Press; 1989: 217-253. 28. SAS. SAS User's Guide: Statistics, Version 6 Edition. Cary, NC: Statistical Analysis System Institute, Inc.; 1987. 29. Eisnspanier R, Miyamoto A, Schams D, Mfiller M, Brem G. Tissue concentration, mRNA expression and stimulation of IGF-I in luteal tissue during the oestrous cycle and pregnancy of cows. J Reprod Fertil 1990; 90:439-445. 30. Wathes DC, Perks CM, Davis AJ, Denning-Kendall PA. Regulation of insulin-like growth factor-I and progesterone synthesis by insulin and growth hormone in the ovine ovary. Biol Reprod 1995; 53:882889. 31. Lucy MC, Byatt JC, Curran TL, Curran DE Collier RJ. Placental lactogen and somatotropin: Hormone binding to the corpus luteum and effects on the growth and functions of the ovary in heifers. Biol Reprod 1994; 50:1136-1144. 32. Hu Z, Tsai-Morris C, Buczko E, Dufau ML. Hormonal regulation of LH receptor mRNA and expression in the rat ovary. FEBS Lett 1990; 274:181-184. 33. LaPolt PS, Jia X, Sincich C, Hsueh AJW. Ligand-induced down-regulation of testicular and ovarian luteinizing hormone (LH) receptors is preceded by tissue-specific inhibition of alternatively processed LH receptor transcripts. Mol Endocrinol 1991; 5:397-403. 34. Tsutsumi M, Laws SC, Rodic C, Sealfon SC. Translational regulation of the gonadotropin-releasing hormone receptor in alpha T3-1 cells. Endocrinology 1995; 136:1128-1136. 35. Webb R, Gong JG, Bramley TA. Role of growth hormone and intrafollicular peptides in follicle development in cattle. Theriogenology 1994; 41:25-30.