Delineating the prime mover action of progesterone for endometrial receptivity in primates

Introduction

Progesterone is essential for endometrial receptivity and pregnancy maintenance in the human and non-human primates12345. As shown in Table I, there are a few reports on the molecular characteristics of the ‘window’ of implantation in which high density cDNA microarray screening was performed for global gene profiling to essentially identify changes in gene expression between early luteal (pre-receptive) and mid luteal (receptive) phases in primate and human endometrium to elucidate steroidal regulation of endometrial physiology towards receptivity67891011121314. There are two studies describing the differential display of genomic expression in human endometrium in response to high affinity progesterone antagonist (mifepristone) during the period of progesterone dominance in vitro and in vivo1014. However, it is now clearly evident that a dialogue between embryo and endometrium under progesterone dominance results in complex interaction of genomic expressions that collectively manifests endometrial receptivity to an implanting blastocyst151617. The potential effect of human chorionic gondotropin (hCG), a putative preimplantation stage embryo-derived signal on transcript profiles in implantation stage baboon endometrium has been demonstrated13. All of the above-mentioned studies, except two1013 adopted an inductive approach based on genome wide expression arrays. We, on the contrary, delineated the transcripts profile during endometrial differentiation for ‘receptivity’ under the dual control of steroidal regulation and embryo-derived signal using a deductive approach18. To this effect, we have compared transcript profiles for known 409 genes between ‘receptive’, and mifepristone-induced desynchronized ‘non-receptive’ stage monkey endometrial samples from mated, potential conception cycles, using cDNA arrays containing sequence-verified clones known to be important in endometrial function18. In the present review, we discuss a few issues related to luteal phase progesterone action on endometrial receptivity using mifepristone as a potent antiprogestin in terms of its transcriptomic expression in fecund cycles of rhesus monkeys as a non-human primate model. The significance of transforming growth factor β-tumour necrosis factor α (TGFβ-TNFα) pathways and signal transducer and activator of transcription (STAT)signaling system under the receptivity-hostility paradigm of endometrial preparation for blastocyst implantation has been specifically addressed.

Table I Summary of differential expression studies in endometrial receptivity

Transcriptomics of receptive endometrium of conception cycle

A large number of gene products (31 out of 54) showed changes within five fold, while relatively less (23) number of genes showed more than five fold changes in both days 4 (10 out of 26 gene products) and 6 (13 out of 28 gene products) post-ovulation endometrium18. Further analysis revealed a total of 26 genes (20 increased and 6 decreased) on day four, and 28 genes (25 increased and 3 decreased) on day 6 after fertilization displayed significant differential expression in endometrium as a result of mifepristone treatment on day 2 after ovulation in proven fertile cycles18. The relative abundance of transcripts of nine gene products (Jun B, KDR, Leptin receptor, matrix metalloprotease 9 (MMP 9), ribosomal protein s7 (Rsp7), ribosomal protein s9 (Rsp9), STAT3, Tastin, TGF beta1) was higher in endometrial samples on both days 4 and 6 after ovulation following mifepristone treatment as compared to control samples and these showed progressive increase between days 4 and 6 after ovulation in mifepristone treated group. A set of another twelve gene products showed changes following mifepristone treatment only in day 4 samples; ten of these (α1-anti-trypsin, Bystin, Follistatin-like1, 3β-HSD, Integrin-α, Keratan sulphotransferase1, matrix metalloprotease 3 (MMP3), STAT-PIASγ, Tissue factor/Thromplastin and Wnt inhibitor factor were increased and two of these [erythroblastic leukemia viral oncogene B (ErB3) and Prolactin receptor] showed decreased expression. Similarly, a different set of fourteen gene products showed changes only in day 6 samples, twelve of these (CD44, cyclooxygenase 1 (COX1), DLX4, E74-like factor 1 (ELF1), interlukin 6 (IL6), JAK1, JUN, MIF, MMP10, PGFR, Thioredoxin, TNFα,) were increased and two (17, 20-desmolase, MUC1) were decreased following mifepristone treatment18.

New leads in transcriptomics of receptive endometrium

Table II shows the common gene products appeared to be regulated by progesterone in previous reports including our study6789101112131418. Table II also shows a new group of 28 gene products that were affected in implantation stage endometrium following early luteal phase administration of anti-progestin mifepristone18 and those were not reported to be affected by progesterone in earlier array-based studies using the model of comparative study between pre-receptive and receptive stage endometrium67891011121314. We have validated our cDNA array based data for nine gene products using TaqMan validation; of the nine gene products, seven (CD 44, CSF-1, DLX4, Enigma, KDR, Leptin receptor, and MIF) appeared to be novel candidates, while two (MUC1 and Uteroglobin) have been reported earlier1112. While previous reports indicated that wnt-associated pathways were involved in progesterone action on endometrial differentiation12, changes in transcripts profiles of homeobox genes like distal less homeobox 4 (DLX4) and enigma, as well as, of transcription and translation related gene products like JunB, c-Jun, Elf1, and ribosomal proteins in implantation stage endometrium following mifepristone treatment18 have not been reported earlier. It is assumed that endometrial receptivity in the primate is associated with categorical reprogramming in endometrial transcriptomics as observed in human endometrial fibroblasts in vitro19, human placental villous cytotrophoblasts in vitro20, and mouse decidual cells21. Thus, the elucidation of the time course pattern of endometrial receptivity associated transcriptomic networks will be beneficial to clinical sciences for improvising strategies to assist the establishment of pregnancy.

Table II Functional clustering of progesterone regulated genes in implantation stage endometrium

Functional correlates

General: The results of our study in the rhesus monkey18 have substantiated earlier reports indicating that endometrial cyclooxygenase, prostaglandin receptor, matrix metalloproteinases, tissue necrosis factor, prolactin receptor and mucin are regulated in mid-luteal phase endometrium by progesterone222324252627282930313233. On the other hand, it is notable that relative abundance of transcripts for colony stimulating factor 1 (CSF1), interleukin 6 (IL-6), leptin receptor, and kinase insert domain receptor (KDR) were actually up-regulated in response to mifepristone treatment, contrary to the expected changes based on earlier reports34353637383940. Collectively, the Enrichment Analysis by Process Maps and Process Networks using Metacore platform (GeneGo, St. Joseph, MI, USA) revealed that the group of 40 genes showing differential expression in endometrium regulated by progesterone action computed upon cell cycle, differentiation and development, immune response and inflammation. Of the all the 40 genes that showed differential expression on inhibition of progesterone action in implantation stage endometrium, two gene products, namely TGFβ1 and STAT3, appear intriguing for disparate reasons as discussed below.

TGFβ: Transforming growth factor beta (TGFβ) has reportedly been observed in a good number of genome-wide expression array studies6781112 to be involved in the process of luteal phase endometrial differentiation under progesterone dominance. TGFβ denotes a family of structurally related, dimeric protein that controls proliferation, apoptosis, cellular differentiation, angiogenesis, tissue remodelling and repair, immune responses and other functions depending on cell types, physiological states and its concentration and availability of its receptors4142. Our observation of increased TGFβ1 expression along with higher expression MMPs and TNF in mifepristone treated endometrium corroborates well with the earlier reports that (i) progesterone mediates a balance of MMPs and TNF in endometrium involving paracrine action of TGFβ434445 and thus it plays an important role in the process of endometrial receptivity and blastocyst implantation4647, and (ii) a very high level of TGFβ may suppress progestational maturation of endometrium484950. Furthermore, we have observed that the protein expression for these mediators was higher in anti-progestin (mifepristone) treated monkey endometrium as compared to control endometrium. Thus, it appears that higher expression of TNFα and TGFβ may mediate endometrial hostility in mifepristone treated fecund cycle through a putative model function as shown in the Figure.

Close

Fig

A putative model based on enriched pathways analysis of transcriptomic profiles highlighting the basic seed genes operative in the progesterone-dominant implantation stage endometrium regulating endometrial receptivity (development) and hostility (death cycle). ErbB, erythroblastic leukemia viral oncogene B; Follistat, follistatin; IL6, interleukin 6; NF-kB, nuclear factor kappa-light chain enhancer of activated B cells; Smads, homologs of both the Drosophilla protein, MAD and the C. elegans protein SMA; stat, signal transducer and activator of transcription; TGFβ, transforming growth factor beta; TNFá, tumour necrosis factor alpha; VEFG, vascular endothelial growth factor.

Export to PPT

STAT3 signaling: STAT3 activation in endometrium was reportedly associated with endometrial receptivity to implantation in rodents515253. Furthermore, Janus Kinase 1 (JAK1) at the transcript level in human luteal phase endometrial explant cultures following mifepristone administration was reduced and immunopositve protein levels of JAK1 was higher in stroma and luminal epithelium of luteal phase endometrium as compared to other stages of menstrual cycle10. Thus, the observed relative higher abundance of transcripts for JAK1 and STAT3 in anti-progestin (mifepristone) treated endometrium in our study18 appears counter-intuitive. To further examine this issue, we investigated immunohistochemically the protein products expression of activated STAT3 in endometrial samples collected on days 4 and 6 after fertilization with and without mifepristone treatment. The immunopositive signals for JAK1 and phosphorylated STAT 3 (pSTAT3) were consistently and markedly higher in epithelial compartment and were depressed in stromal and vascular compartments on days 4 and 6 after ovulation, while mifepristone treatment resulted in higher pSTAT3 in the vascular compartment18. Although the mechanism and function of increased expression of pSTAT3 in vascular compartment following mifepristone treatment are unclear, it is likely that inflammatory changes are triggered in luteal phase endometrium by mifepristone induced vasocentric JAK1-STAT3 pathway5455, thereby rendering endometrial hostility to blastocyst implantation. This observation corroborated well with earlier observation that early luteal phase administration of mifepristone resulted in significant changes in mid-lueal phase endometrial vascular physiology associated with phenotypic hostility of endometrium towards blastocyst implantation56.

In conclusion, we propose that TGFβ-TNFα pathways and STAT signaling system are important effectors in mediating the combinatorial regulation of luteal phase progesterone action on endometrial receptivity in fecund cycle.