|
|
||||||||
a Departments of OB/GYN and Physiology, University of Massachusetts Medical School, Worcester, Massachusetts 01655
| ABSTRACT |
|---|
|
|
|---|
| INTRODUCTION |
|---|
|
|
|---|
The molecular mechanisms, which act to achieve temporal and spatial control of differential growth in the nonhuman primate endometrium, are largely unknown. It is evident, however, that target tissues for steroid hormones contain high-affinity binding proteins that mediate the hormonal response. The evolving model of P4 action in normal tissues requires the binding of hormone to the P4 receptor (PR), which increases the affinity of the steroid-receptor complex to P4 response elements that function as DNA enhancer and repressor elements to control transcription of specific genes and gene networks [47]. Thus, it is envisaged that a cascade of gene expression initiated by PR, involving up- and down-regulation of ligands, growth factors, receptors, extracellular matrix proteins, and enzymes controlling cellular metabolism, is required for correct endometrial maturation. Endometria that cannot support implantation due to an inadequate P4 level are expected to show differences in gene expression of factors required for normal endometrial differentiation.
We have constructed cDNA populations from rhesus monkey endometrial tissue in artificial menstrual cycles (AMCs) under normal (adequate) and inadequate secretory phases and have used differential-display reverse transcription-polymerase chain reactions (DDRT-PCR) to identify genes regulated by P4 in vivo. These cDNA populations represent mRNA expression in 1) the E-dominant proliferative phase (EcDNA), 2) the P4-dominant secretory phase (PcDNA), and 3) inadequate secretory phases (IcDNA). In addition, we have analyzed by DDRT-PCR two P4-enriched cDNA populations created by subtractive hybridization. Using this approach we have identified several different levels of P4-dependent regulation of endometrial genes in the normal secretory phase. These studies are a further step toward the identification of irregularities of gene expression that may result in infertile, inadequate endometria.
| MATERIALS AND METHODS |
|---|
|
|
|---|
The animal model and the AMCs used in this study have been described in detail previously [8, 9]. Briefly, the approach uses ovariectomized female rhesus monkeys in which the menstrual cycle is simulated artificially by controlled administration of E- followed by P4-releasing silicone elastomer implants [8, 9]. Peak levels of E occur on Day 13 and mimic the natural proliferative phase [8]. Similarly, the secretory phase of the menstrual cycle is induced by P4 administration starting on Day 13, with peak P4 levels occurring between Days 20 and 23. Creating low levels of P4 (23 ng/ml) during the secretory phase that are within ranges determined to be inadequate in the rhesus monkey [3, 9] simulates the inadequate menstrual cycle. All procedures were approved by the Institutional Animal Care and Use Committee at UMass Medical Center.
Synthesis and Amplification of cDNA Populations
We and our colleagues have previously described the detailed procedure for the isolation of EcDNA, PcDNA, and IcDNA populations [9, 10]. To obtain cDNA populations representing proliferative, secretory, and inadequate phases, poly(A)+ RNA was prepared from surgically removed biopsies of rhesus monkey endometria pooled from 46 animals on Days 9, 11, and 13 of normal AMCs (E dominant), Days 21 and 23 of normal AMCs (P4 dominant), and Days 21 and 23 of AMCs. All procedures were as previously described [911]. The mRNA populations were then converted to double-stranded cDNA. The cDNA populations were end-ligated with EcoRI adaptors and amplified by the polymerase chain reaction (PCR) using adaptor-complementary primers [10]. Two libraries enriched for P4-dependent fragments were constructed by subtractive hybridization of PcDNA (target) from EcDNA (driver) to yield P4-enriched cDNA [10], and PcDNA from IcDNA to yield adequate-enriched cDNA.
DDRT-PCR
Twenty nanograms of cDNA was amplified in 100-µl reactions containing 2.5 µM dNTPs, 1.5 mM magnesium chloride, single-strength buffer, 12.5 µCi [
-35S]dATP, 2 U Tfl polymerase, and 0.5 µM arbitrary primers (10-mers, listed below). Reactions were carried out at 94°C/1 min, 40°C/2 min, and 72°C/30 sec for 40 cycles and analyzed by denaturing polyacrylamide gel electrophoresis, omitting the fixing stage. Experiments were duplicated to ensure reproducibility and to avoid isolating false-positive bands. The autoradiogram and gel were aligned by needle punctures, and individual bands were carefully excised from the gel with a razor blade. Gel slices attached to filter paper were boiled in 100 µl water and spun to remove debris, and the supernatant was precipitated with glycogen. DNA fragments thus isolated were reamplified as described above except with 250 µM dNTPs and no radiolabel. Products were blunt-ended using T4 DNA polymerase and cloned into the plasmid vector pCDNAII (Invitrogen, San Diego, CA) and then sequenced. Complementary DNA fragments isolated in this fashion are flanked by sequences homologous to the arbitrary primers. Arbitrary primers were ARB-1: CTGATCCATG; ARB-2: CTTGATTGCC; ARB-4: GTTGCGATCC; ARB-5: GACCGCTTGT; ARB-7: CTTTGGTCAG; ARB-8: CAAGCGAGGT; ARB-9: AACGCGCAAC.
Semiquantitative PCR Analysis
Twenty nanograms of template cDNA was amplified in 100-µl reactions containing 0.5 µM primers, 0.25 mM dNTPs, 1.5 mM magnesium chloride, single-strength PCR buffer (Sigma Chemical Co., St. Louis, MO), and 2 U of Tfl polymerase for 2050 cycles. Each experiment was repeated three times to ensure reproducibility. Products were analyzed by 0.51.0% ethidium bromide-stained agarose gel electrophoresis, and comparative quantitation was performed by densitometric analysis of photographed gels. Representative gels of PCR products determined to be within the linear range of amplification are shown. The primers used and the expected product sizes are as follows: I1 upper: 5'-AAAGTATGGGAATGCCAACG, I1 lower: 5'-CTATGGATGGAGAGAGGCCA, product: 117 bp; I2 upper: 5'-GTCCATGGCAAAAGAAAAGC, I2 lower: 5'-TGATGTTGGTTGCTGCAAAT, product: 137 bp; I4 upper: 5'-GCGATCCTGGTCTCTGAAAA, I4 lower: 5'-ACCGCTTGTTGAATTGATCC, product: 100 bp.
| RESULTS |
|---|
|
|
|---|
|
|
|
|
We first compared EcDNA, PcDNA, and P4-enriched cDNA by differential display to identify P4-dependent fragments. Our criteria for selection of these fragments required that the fragment be absent in EcDNA and strongly expressed in PcDNA. Fragments P1 and P2 met these criteria as shown in Figure 2a and were subsequently isolated. These two P4-dependent fragments were also more abundant in a subtracted cDNA population enriched for P4-dependent genes, reinforcing our assumption that these are genuine P4-induced sequences and not artifacts of the technique. This pattern of fragments in the cDNA populations was reproducible in several repeat experiments using the same set of arbitrary primers (data not shown).
Figure 2, bf, shows additional gels using different primer sets in which IcDNA and the other P4-enriched subtracted library, adequate-enriched cDNA, were analyzed. From these gels the P4-dependent fragments P3, P4, P5, P6, P7, and P8 (shown in Fig. 2) were isolated. It should also be noted that in panels bf of Figure 2 the expression of fragments P3P8 in IcDNA was absent or very low, providing additional confidence that these fragments required normal levels of serum P4 for proper expression. In addition to identification of genes that are up-regulated during an adequate secretory phase (PcDNA), Figure 2 also shows that there is most often a coincident reduction in mRNA abundance in IcDNA.
All the fragments (P1P8) were between approximately 50 and 400 base pairs (bp) in length. In some cases we were able to identify the genes that these fragments represent by DNA and predicted protein homology with the GenBank databases (Table 1). P1 is 103 bp, and its DNA sequence matched an as yet uncharacterized human P1 (bacteriophage) artificial chromosome clone (PAC) from chromosome 6. Fragments P2, P3, and P4 yielded no significant sequence homologies to GenBank entries. P5 is 91 bp and showed limited, gapped DNA homology to an uncharacterized human bacterial artificial chromosome (BAC) clone from chromosome 7. P6 is 261 bp, of which 71 bp exhibited limited homology to an uncharacterized human expressed sequence tag (EST) from fetal brain. P7 and P8 showed no homology to database entries. Since fragments P1P8 have, to date, yielded no significant gene matches, we searched for potential encoded protein domains using the Prosite (Swiss Institute for Bioinformatics, Geneva, Switzerland) database, but again no significant homologies were found.
|
In addition to identification of fragments up-regulated by P4 in the normal secretory phase, we also identified several fragments that were highly expressed in IcDNA but showed low or absent expression in PcDNA and EcDNA (Fig. 3). The rationale for the isolation of inadequate-specific fragments is based on the hypothesis that some genes are rapidly turned on by low levels of P4 and subsequently turned off during the midsecretory phase during elevated serum P4 levels (autologous down-regulation). Indeed, Figure 3 shows that there are numerous gene fragments that are up-regulated in inadequate endometrium (low P4) and repressed during an adequate secretory phase. These genes do not appear to be E-dependent, because their expression is similar in E- and P4-dominant endometria.
The fragments I1I4 were thus identified (Fig. 3), cut from the gel, reamplified, and cloned into a prokaryotic vector. Sequence homology analysis against GenBank entries was then performed (Table 1). Fragment I1 contained an open reading frame (ORF) that matched (100%) human serine/threonine protein phosphatase 2A. Fragment I2 contained an ORF that showed significant homology (83%) to human oxobutanoate dehydrogenase. The DNA sequence of fragment I3 was homologous to an uncharacterized human EST clone, and fragment I4 contained an ORF that was 88% identical to primate line-1 reverse transcriptase homologue.
Using differential display we were also able to identify E-dependent genes that are inhibited by P4 (Fig. 4). We isolated, cloned, and sequenced four of these fragments (E1E4) as noted in Figure 4 and Table 1. Fragments E2 and E9 contained no ORFs and showed no significant homology to GenBank entries. These fragments represent undefined noncoding regions. E3 was significantly homologous (probability = e-36) to an as yet uncharacterized human sequence site tag (STS) clone (see Table 1). Fragment E4 was also homologous (e-17) to uncharacterized human genomic DNA near the Down syndrome region on chromosome 21.
For the three fragments that correspond to previously characterized mRNAs (I1, I2, and I4), we designed specific primers for semiquantitative PCR in order to confirm their regulation by P4. Figure 5 shows that I1 and I2 were up-regulated (5- and 3-fold, respectively) in IcDNA compared to EcDNA and PcDNA, as expected. However, I4 appears to be a false positive from DDRT-PCR, since its expression was greatest in PcDNA (i.e., up-regulated by P4 in the normal secretory phase). This observation shows the necessity for confirming the regulation of differential-display fragments of known genes and avoiding false positives.
|
| DISCUSSION |
|---|
|
|
|---|
We used three primary endometrial cDNA populations and two P4-enriched populations, created by subtractive hybridization, as templates for analysis by differential gene display. The primary cDNAs represent mRNA populations from the proliferative (E dominant), secretory (P4 dominant), and inadequate secretory (low P4) phases of the menstrual cycle (EcDNA, PcDNA, and IcDNA, respectively). The subtracted cDNAs were made by subtracting E/P-common fragments from PcDNA to yield P4-enriched cDNA, and I/P-common fragments from adequate (P4) cDNA to yield adequate-enriched cDNA, respectively [10]. In this study we focused on the different levels of P4-dependent regulation identified by differential display and isolated 16 gene fragments for cloning and sequencing.
Seven fragments (P2, P3, P4, P7, P8, E2, and E9) appeared to represent previously undescribed transcripts since they showed no match to GenBank entries. Six fragments (P1, P5, P6, I3, E3, and E4) exhibited homology to uncharacterized sequences on cosmid and ESTs, for which there is no further genetic information. Three fragments (I1, I2, and I4) corresponded to previously characterized mRNA sequence entries.
We have identified two gene fragments (I1,I2) that are overexpressed in IcDNA (inadequate) compared to PcDNA (adequate). These fragments are also underrepresented or absent in EcDNA, suggesting that an inadequate level of P4 can induce P4-dependent gene expression but that an adequate level of P4 appears to exert an autologous down-regulation of these early secretory phase response genes. Homology searches for fragments I1 and I2, which are induced by a low P4 level, resulted in virtually identical, extensive matches (95100%) to previously known genes. I1 encodes protein phosphatase 2A (PP2A), a heterotrimeric serine/threonine phosphatase present in most tissue and cell types that regulates such diverse cellular processes as DNA replication, cell cycle progression, signal transduction, and intermediary metabolism [14]. PP2A can both positively and negatively regulate cellular processes in conjunction with different catalytic subunits and protein kinases by reversible phosphorylation of factors that function at different steps of signaling cascades [15].
I2 encodes branched-chain alpha-keto acid dehydrogenase (BCAD), a multienzyme complex occupying a key position in intermediary metabolism. BCAD catalyses an irreversible step in the catabolism of several essential amino acids, including the branched-chain amino acids. BCAD is located in mitochondria and is under stringent control by hormones and dietary factors. The promoter region of BCAD contains a progesterone/glucocorticoid response element [16] that is consistent with regulation by P4 [7]. Interestingly, BCAD is active only when dephosphorylated [17].
I4 was shown to be a false positive in that it was up-regulated by an adequate rather than inadequate level of P4. I4 has significant homology to a human reverse transcriptase homologue similar to RNA-dependent DNA polymerases found in transposable elements and retroviruses [18]. Primates and rodents have been shown to contain L1 transposable element-like repeats thought to be derived from a reverse transcriptase-related protein [18]. Although the fragments I1, I2, and I4 are considered identified, their roles in the endometrium remain unknown.
The homology search results presented above are all statistically significant at the 99% confidence level, but biological knowledge should temper the conclusions for these findings. For instance, the fragments P5 and P6 exhibit only limited, gapped DNA homology (7585%) to their respective genes/gene products listed in Table 1. No direct biological conclusions can be drawn from these homologies at this time, and we are currently obtaining longer sequences of each fragment from cDNA libraries in order to help resolve the significance of the results.
We envisage that coordinated, steroid-induced activation and repression of many genes during the changeover from E to P4 dominance is important for correct endometrial maturation. On the basis of the P4-dependent endometrial regulation identified in the present study and other studies [9, 11], we propose that waves of P4-dependent gene regulation, a cascade, are central to proper maturation of the primate endometrium. During a normal secretory phase, the early phase of P4 regulation likely involves both inhibition of E-dependent genes and induction of P4-dependent genes that are sensitive to the initial rise in serum P4 level. As the secretory phase proceeds, rising titers of serum P4 lead to further inhibition of E-dependent genes and continued induction of P4-dependent genes. Superimposed on these two regulatory actions of P4 is an autologous down-regulation of some but not all P4-dependent genes. In addition, several laboratories have shown that some genes are activated only during a sharply restricted window, just prior to and during potential implantation in the human and other animal models [1925]. Subsequently, P4-dependent endometrial gene regulation necessary for proper maturation subsides with falling titers of P4 in the late secretory phase, signaling a decline in receptivity of the endometrium.
The use of differential display with our hormonally distinct cDNA populations has allowed us to identify different P4-dependent mRNA regulatory patterns in the normal secretory phase of the rhesus monkey. Although these data provide evidence for the relative changes in mRNA abundance, other potential effects of steroid hormones, such as alterations in degradation of mRNAs and proteins, need to be addressed by other approaches. In addition, approaches such as immunohistochemistry and in situ hybridization will be necessary to address important issues of cell type-specific expression in hormonally distinct microenvironments within the primate endometrium [26, 27].
| NOTE ADDED IN PROOF |
|---|
|
|
|---|
| ACKNOWLEDGMENTS |
|---|
| FOOTNOTES |
|---|
2 Correspondence: William C. Okulicz, Dept. of OB/GYN, Umass Medical School, 55 Lake Ave. N., Worcester, MA 01655. FAX: 508 856 5016; william.okulicz{at}banyan.ummed.edu ![]()
Accepted: November 30, 1998.
Received: July 8, 1998.
| REFERENCES |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
V S Patil, G Sachdeva, D N Modi, R R Katkam, D D Manjramkar, I Hinduja, and C P Puri Rab coupling protein (RCP): a novel target of progesterone action in primate endometrium J. Mol. Endocrinol., October 1, 2005; 35(2): 357 - 372. [Abstract] [Full Text] [PDF] |
||||
![]() |
G. R. Bertani, C. D. Gladney, R. K. Johnson, and D. Pomp Evaluation of gene expression in pigs selected for enhanced reproduction using differential display PCR: II. Anterior pituitary J Anim Sci, January 1, 2004; 82(1): 32 - 40. [Abstract] [Full Text] [PDF] |
||||
![]() |
W. C. Okulicz and C. I. Ace Temporal Regulation of Gene Expression During the Expected Window of Receptivity in the Rhesus Monkey Endometrium Biol Reprod, November 1, 2003; 69(5): 1593 - 1599. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. S.T. Torres, C. I. Ace, and W. C. Okulicz Assessment and Application of Laser Microdissection for Analysis of Gene Expression in the Rhesus Monkey Endometrium Biol Reprod, October 1, 2002; 67(4): 1067 - 1072. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |