Corpora lutea were obtained from mares at days 3, 10 and 14 after ovulation, and examined histologically. The morphology of isolated luteal cells obtained by either mechanical or collagenase dissociation of the tissue was examined and the cells stained to detect the steroidogenic enzyme Δ⁵, β-hydroxysteroid dehydrogenase. The ratio of large:small cells was significantly higher for cells obtained from mechanically dissociated luteal tissue than for cells obtained by collagenase dissociation (P < 0.01). Cells obtained by both mechanical and collagenase dissociation secreted progesterone, although neither cell population responded to exogenous gonadotrophin with an increase in progesterone secretion. Homogenates of equine luteal tissue bound ¹²⁵I-labelled human LH with high affinity and specificity, and the specific activity and binding affinity of luteal LH receptors did not change significantly from day 3, to days 10 and 14 after ovulation. However, mechanically dissociated cells on days 10 and 14 bound significantly more LH than did collagenase-dissociated cells on these days (P < 0.05). These results indicate that (i) collagenase dissociation of mare luteal tissue yields a population of cells that is unrepresentative of the corpus luteum, and (ii) the mare corpus luteum is not responsive to LH in vitro at the stages examined.

Seasonally anoestrous Welsh Mountain ewes received 250 ng gonadotrophin-releasing hormone (GnRH) every 2 h, with (Group 1; n = 13) or without (Group 2; n = 14) progesterone priming for 48 h. Fourteen control ewes (Group 3) were studied during the luteal phase in the breeding season. Animals in Group 4 (n = 12) received progesterone priming followed by 250 ng GnRH at increasing frequency for 72 h, while ewes in Group 5 (n = 13) were given three bolus injections of 30 μg GnRH at 90-min intervals. All treatment regimens induced ovulation. However, only corpora lutea (CL) from ewes in Group 3 (breeding season) or Group 4 exhibited normal luteal function. Luteal luteinizing hormone (LH) receptor levels were significantly higher on day 12 than day 4, and CL from groups with adequate CL (3 and 4) had significantly higher ¹²⁵I-human chorionic gonadotrophin (hCG)-binding levels than the three groups with inadequate CL on day 12. LH-binding affinity was unchanged. Exogenous ovine LH (10 μg) in vivo on days 3 or 11 after ovulation induced a pulse of progesterone in ewes with adequate CL: however, ewes in Groups 1, 2 and 5 showed no significant response. Basal progesterone secretion in vitro was significantly greater on day 4 than on day 12. Maximal steroidogenic responses of adequate and inadequate CL to hCG and to dibutyryl cyclic-3′,5′-AMP were similar at both stages of the luteal phase. However, the EC₅₀ for hCG on days 4 and 12 was 10-fold lower for groups with an adequate CL (0.1 IU hCG/ml) than for inadequate-CL groups (1 IU hCG/ml; P <0.05). Thus, in addition to the well-characterized premature sensitivity of GnRH-induced inadequate CL to endometrial luteolysin, we have shown (1) a marked decrease in total number of cells in the CL, a profound reduction in vascular surface area, and a decrease in mean large luteal cell volume (with no change in large luteal cell numbers), (2) decreased luteal LH receptor and progesterone content compared with adequate CL and (3) that CL that were becoming, or were destined to become, inadequate failed to respond to ovine LH in vivo and were 10-fold less sensitive to hCG in terms of luteal progesterone secretion in vitro.