The oxytocin receptor (OTR) is part of an ancient hormone system expressed in diverse phyla in relation to acute reproductive smooth muscle responses, such as egg-laying, birth, or milk letdown. The regulation of the OTR gene, while correlating with steroid levels in vivo, remains elusive. There appear to be both inhibitory and stimulatory influences acting upon a constitutive pattern of basal expression. We have found no evidence, however, for an effect of the sex steroids either directly on gene transcription, or on the receptor itself at the protein level. In the prostatic carcinoma cell line Du145, we have shown that up-regulation of the OTR gene transcription can be effected by cAMP. In an attempt to characterize the expression of the OTR protein in vivo, we have shown, using ligand-blotting, that the OTR can be expressed at different sizes in transfected cells and in myometrium. Also, in the myometrium at term, immunohistochemistry suggests that there is both an increase in OTR protein per cell, as well as in the number of smooth muscle cells expressing OTR, emphasizing that perinatal changes are the results of both individual gene activation events and gross cellular differentiation. The OTR is a valuable model system reflecting molecular changes in the perinatal period. When we understand how this important molecule is regulated, we will also be a long way towards understanding the mechanisms controlling myometrial contractility at birth. Experimental Physiology (2001) 86.2, 289-296.

The bovine gene for the steroidogenic acute regulatory protein (StAR) was cloned and se-

No abstract

The oxytocin receptor (OTR) is expressed in the cow uterus at high levels at estrus and at term of pregnancy. This expression appears to be controlled mostly at the transcriptional level and correlates with increasing estrogen concentration and progesterone withdrawal. Approximately 3200 base pairs of the upstream region of the bovine OTR gene were cloned and analyzed using a combination of bioinformatic, electrophoretic mobility shift (EMSA), and transfection analyses. Using nuclear proteins from high- and low-expressing tissues, EMSA indicated no significant quantitative or qualitative changes in specific DNA-protein binding, suggesting that transcription is probably controlled by signalling systems targeting constitutive factors. Using various cell types, including primary and immortalized ruminant endometrial epithelial cells, as hosts for transfection of promoter-reporter constructs showed that endogenous activity resided only in the longest, i.e., 3.2-kb, construct but not in those shorter than 1.0 kb. While estrogen appears to be important in vivo, no effect of estradiol was found on any construct directly; only when the longest 3.2-kb construct was used in combination with some cotransfected steroid receptor cofactors, e.g., SRC1e, was an estradiol-dependent effect observed. A putative interferon-responsive element (IRE) was found at approximately -2,400 from the transcription start site. This element was shown to bind mouse IRF1 and IRF2 as well as similar proteins from bovine endometrial and myometrial nuclear extracts. This element also responded to these factors when cotransfected into various cell types. The bovine equivalents to IRF1 and IRF2 were molecularly cloned from endometrial tissue and shown to be expressed in a temporal fashion, supporting the role of interferon-tau in maternal recognition of pregnancy. Of many factors tested or analyzed, these components of the IFN system are the only ones found to significantly influence the transcription of the bovine OTR gene.

In vivo there appears to be a marked association between oestrogen levels and the expression of the oxytocin (OT) gene in most tissues. Transfection and DNA-protein binding experiments using high levels of either oestrogen receptor (ER)alpha or ERbeta imply a direct interaction of these transcription factors with the multiple hormone response element (HRE) at approximately -160 from the transcription start site of the OT gene in most species. In an extensive set of experiments, we show, using both transfection and protein-DNA binding, that low to moderate amounts of either oestrogen receptor, while being able to interact directly with a classic oestrogen response element (ERE) fail to interact with the human OT -160 HRE. Instead, this element, similar to its bovine counterpart, has a high affinity for the orphan receptors steroidogenic factor 1 and chicken ovalbumin upstream promoter transcription factor. Second, the human and bovine OT promoter can be made artificially responsive towards oestrogen in a cotransfection system over-expressing ERalpha or ERbeta, but not in cells expressing natural levels of these steroid receptors. Interestingly, nuclear extracts from both ERalpha-positive MCF7 cells and ERalpha-negative MDA-MB231 cells both contain a transcription factor which binds specifically to both the hOT-HRE element and to a classic ERE, and which has orphan receptor-like binding properties rather than those of an oestrogen receptor. Together, these and other results suggest that oestrogen action in vivo on the OT gene in all species is more likely to involve a DNA-independent mechanism than classic direct interactions with dimeric oestrogen receptors.