Search

Cells from the inner and outer granulosa cell layers of the ovarian follicles differ in function, probably because of their different origins from the surface epithelium and from the rete. This suggestion has not so far been thoroughly investigated in the human ovary. We examined fetal ovaries from the early, middle and late gestational periods, ovaries from fertile women, and preovulatory follicular cells obtained from patients under in vitro fertilization therapy (IVF). Indirect immunohistology and immunocytology were used to detect the presence of cytokeratin (CK)-positive epithelial cells. In fetal ovaries from the early gestational period, prominent rete tubules (sometimes with oocytes) appeared to be fused with the sex cords and primordial follicles. Both showed CK-positively, detected with the pan-CK antibody Lu-5. Cytokeratin 19 was clearly expressed in the fusion area. In the fetal and adult ovaries, CK-positive follicular or granulosa cells were noted in the primordial and primary follicles as well as the preovulatory follicles. Cytokeratin was not detected in the granulosa cells of growing follicles, CK-positive and -negative luteal cells were identified in the developing corpus luteum. We conclude for the human ovary: (1) the heterogeneous morphology of granulosa cells may be explained by their twofold origin from the surface epithelium and the rete, (2) the rete tubules appear to be involved in folliculogenesis, (3) the transient absence of CK expression in growing follicles compared to resting and mature follicles or to the developing corpus luteum indicates a particular role of CK-positive cells at the periovulatory period.

Serial sections of 62 rat ovaries and 66 ovaries from Syrian hamsters (Mesocricetus auratus) have been examined by light microscopy. The relative and absolute percentage of preantral intra-ovarian oocyte release (PIOR) was recorded. PIOR was incomplete when the rupture of the basal membrane was not followed by the release of the oocyte from its preantral follicle into the ovarian stroma; conversely, a PIOR was regarded to be complete if the oocyte was released from the preantral follicle into the adjacent stroma. PIOR occurred in 23% of rat ovaries and in 32% of hamster ovaries; meiosis was not resumed. In rat ovaries the frequency of complete PIOR was 71%; the oocyte disappeared quickly after its release and herniated granulosa cells could invade capillaries. By contrast, incomplete PIOR was prevalent in hamster ovaries (67%); oocytes survived in the interstitial cortical tissue after the disappearance of their granulosa and theca cells ("free oocytes"). It is concluded that PIOR reflects the onset of follicular atresia and that different local factors are probably responsible for species differences.

Follicular ruptures with intra-ovarian oocyte release (IOR) were studied in 17, 21 and 24-day-old rats by morphological methods. Using a light microscope, it was seen that IOR occurred at all times and the IOR frequency did not change. IOR developed in preantral follicles. Their oocytes were mostly found within the follicular compartment (incomplete IOR). Using an electron microscope, a circumscribed dissolution of the basal lamina was observed. IOR granulosa cells appeared activated. They rarely underwent typical necrosis after herniation into the extrafollicular area. Herniated granulosa cells tended either to stay intact or to shed cytoplasmic components into the extracellular space, whilst nuclei of active cell function were maintained. Tissue adjacent to an IOR seemed inactive with the exception of endothelial cells. Some endothelial cells underwent necrosis. Additionally, the endothelium was discontinuous. The morphological data support the hypothesis that the mechanism of follicular rupture represents an inside to outside process.