Endothelium-derived endothelin-1 (ET-1) and nitric oxide (NO) are pivotal regulators of corpus luteum (CL) function. To have a better insight into their synthesis and action, members of the ET system (ET-1, ET converting enzyme (ECE-1) isoforms a -d, ET A and ET B receptors) along with NO synthase (NOS) isoforms -endothelial (e)NOS and inducible (i)NOS -were quantified in CL-derived endothelial cells (CLEC). The expression of these genes in microvascular CLEC, obtained by lectin-coated magnetic beads, was compared with cells removed from the luteal microenvironment and maintained in culture for different durations, and with endothelial cells (EC) derived from a large blood vessel (i.e. bovine aortic endothelial cells, BAEC). The profile of gene expression in the different EC types was determined by quantitative real-time PCR. Freshly isolated EC from mid-cycle CL exhibited high ET-1 receptor expression (both ET A and ET B ), low ET-1 synthesizing ability (both prepro (pp) ET-1 and ECE-1), but elevated iNOS -the high throughput NOS isoform. The distinct phenotype of CLEC was lost soon after an overnight culture. ET A and ET B receptor levels declined, ppET-1 levels increased while iNOS was reduced. These changes were extenuated during long-term culture of CLEC. The general pattern of gene expression in BAEC and long-term cultured CLEC was similar yet some differences, reminiscent of freshly isolated CLEC, remained: ECE-1c, ET B receptor and NOS isoforms were expressed differently in BAEC as compared with lines of CLEC. This study suggests that the luteal microenvironment is necessary to sustain the selective phenotype of its resident endothelial cells. The inverse relationship between ppET-1 and iNOS observed in freshly isolated CLEC and in cultured cells is physiologically significant and suggests that ET-1 and NO may modulate the production of each other.

Endothelin-converting enzyme-1 (ECE-1) cleaves big endothelins, as well as bradykinin and ␤-amyloid peptide. Several isoforms of ECE-1 (a-d) have been identified to date; they differ only in their NH 2 terminus but share the catalytic domain located in the COOHterminal end. Using quantitative PCR, we found ECE-1d to be the most abundant type in several endothelial cells (EC) types. In addition to full-length ECE-1 forms we have identified novel, alternatively spliced mRNAs of ECE-1 b-d. These splice variants (SVs) lack exon 3, which codes for the transmembrane region and is present in full-length forms. SVs mRNA were highly expressed in EC derived from macro and microvascular beds but much less so in other, nonendothelial cells expressing ECE-1, which suggests that the splicing mechanism is cell-specific. Analyses of ECE-1d and its SV form in stably transfected HEK-293 cells revealed that both proteins were recognized by anti COOH-terminal ECE-1 antibodies, but anti NH 2 -terminal antibodies only bound ECE-1d. The novel protein, designated ECE-1 sv, has an apparent molecular mass of 75 kDa; by using site-directed mutagenesis its start site was identified in a region common to all ECE-1 forms suggesting that ECE-1 b-d SV mRNAs are translated into the same protein. In agreement with the findings demonstrating common COOH terminus for ECE-1sv and ECE-1d, both exhibited a similar catalytic activity. However, immunofluorescence staining and differential centrifugation revealed a distinct intracellular localization for these two proteins. The presence of ECE-1sv in different cellular compartments than full-length forms of the enzyme may suggest a distinct physiological role for these proteins.