Aggregation of dispersed human cytotrophoblastic cells: lessons relevant to the morphogenesis of the placenta.

The syncytial trophoblast of the human placenta forms by the fusion of mononuclear cytotrophoblast cells. Cytotrophoblast cells only fuse with other trophoblastic cells, indicating a specificity to this interaction. To explore the cellular aggregation which precedes fusion, we examined the association of cytotrophoblast cells isolated from term placentae and JEG-3 choriocarcinoma cells, a cytotrophoblast-like cell line, in suspension culture. Cytotrophoblast cells were isolated by dispersion of chorionic villi in trypsin-DNase in Ca2+/Mg2(+)-free medium. JEG-3 cells were released from culture flasks by trypsinization in Versene-EDTA buffer. In suspension culture, each cell type aggregated forming tissue-like masses over a 24-hr period. Transmission electron microscope analysis demonstrated the formation of numerous desmosomes between the aggregated cells. In outgrowth culture, the aggregates created in suspension were maintained as microvilli-covered multicellular structures with hollow cores. The extent of aggregation was dependent upon the concentration of cells in the incubations with greater aggregation occurring with higher cell densities. Aggregation of both cytotrophoblast cells and JEG-3 cells progressed rapidly during the initial 10 hr of incubation and then continued at a slower rate. Aggregation took place in serum-containing and serum-free medium, but was impeded in Ca2+/Mg2(+)-free medium. Incubation of JEG-3 and cytotrophoblast cells in the presence of the protein synthesis inhibitor, cycloheximide, prevented aggregation, whereas the inhibitor of N-linked glycosylation, tunicamycin, did not. The inhibitor of RNA synthesis, actinomycin D, had no effect on the aggregation of the cells during the initial 6 hr of aggregation. These findings suggest that trypsin treatment in Ca2+/Mg2(+)-poor medium removed a protein(s) from the trophoblast cell surface which must be resynthesized for cell-cell association to take place.

Involvement of cyclic AMP in the functions of granulosa and luteal cells: regulation of steroidogenesis.

The findings reviewed above demonstrate that cAMP can act at several distinct loci to enhance steroidogenesis. Analogs of cAMP stimulate the accumulation of the mRNAs which encode components of the steroidogenic machinery, such as the receptor for LDL and the system for the cleavage of the cholesterol sidechain. This apparently coordinated accumulation of specific mRNAs results from increased transcription of the relevant genes. Transacting factors modulated by cyclic AMP may influence common enhancer sequences (e.g., TGACGTCA), and such interactions would account for the simultaneous increase in expression of specific genes on several different chromosomes. These actions of cyclic AMP enable ovarian cells to support long-term increases in steroid synthesis by increasing the quantities of proteins involved in steroidogenesis. Such changes would obviously be important during luteinization, when the potential of granulosa cells to secrete progesterone is greatly increased. Although we speak of these effects as "long-term", it is evident that they occur within a relatively short time (hours) after exposure of cells to the tropic agent. Cyclic AMP also acts to stimulate steroidogenesis post-transcriptionally. It may influence events at the translational level via interactions in the formation of "labile" proteins. In addition, the regulation of cholesteryl ester hydrolase, as well as of other enzymes involved the metabolism of cholesterol, seems to involve post-translational modifications (e.g., phosphorylation). The effects of cAMP on the cytoskeleton may be another manifestation of a post-translational response. These actions of cAMP promote acute increases in steroidogenesis (i.e., within minutes). They encompass the transport of cholesterol to the mitochondria and the regulation of access of sterol to the inner mitochondrial membranes. Future research should be directed at elucidating the exact mechanisms which permit cAMP to exert coordinate effects on the genome (Figure 8), as well as its post-transcriptional effects on various proteins and enzymes which play a role in the synthesis of steroid hormones.

Gonadotropins and cyclic adenosine 3',5'-monophosphate (cAMP) alter the morphology of cultured human granulosa cells.

Morphological changes in human granulosa cells in culture were observed by phase, fluorescent, scanning electron and transmission electron microscopy following the addition of human chorionic gonadotropin (hCG), luteinizing hormone (LH), 8-bromocyclic adenosine 3',5'-monophosphate (cAMP) and cytochalasins B and D. In response to these agents, polygon-shaped granulosa cells with granular cytoplasm became rounded, leaving fingerlike processes attached to the substratum and adjacent cells. The changes in cell shape were accompanied by a centripetal movement of mitochondria and lysosomes to a perinuclear location. The morphological alterations appeared to be mediated by cyclic AMP and to be the result of a dismantling and reorganization of microfilament-containing stress fibers. Follicle-stimulating hormone (FSH), prolactin (PRL), growth hormone (GH), and human placental lactogen (hPL) did not provoke cell shape changes. We conclude that tropic hormones capable of stimulating progestin secretion by luteinized granulosa cells cause a change in cell structure in vitro which leads to a redistribution of organelles involved in steroid synthesis. The possible relationship of the cytoskeleton to steroidogenesis is considered.

Human chorionic gonadotropin and 8-bromo-adenosine 3',5'-monophosphate stimulate [125I]low density lipoprotein uptake and metabolism by luteinized human granulosa cells in culture.

In nonsteroidogenic cells, cellular cholesterol requirements and sterol availability determine low density lipoprotein (LDL) receptor expression and LDL metabolism. We wished to learn if hCG and cAMP increase LDL metabolism by cultured luteinized human granulosa cells and whether this increase is dependent on enhanced metabolism of cellular cholesterol stores to steroid. Granulosa cells were cultured for 48 h in medium containing 20% human male serum and then for 48 h in serum- and hormone-free medium. The cells then received either fresh medium (no additions) or one of the following treatments: 500 mIU hCG/ml, 1.5 mM 8-bromo-cAMP, 100 micrograms aminoglutethimide (AG)/ml to inhibit cholesterol metabolism to steroid hormones, hCG plus AG, or 8-bromo-cAMP plus AG. After 6-48 h of exposure to tropic agents, specific metabolism of [125I]LDL was determined. hCG and 8-bromo-cAMP significantly increased (P less than 0.05) the amount of [125I]LDL bound (2.2-fold), internalized (2.3-fold), and degraded (2.9-fold) by the luteinized granulosa cells. The apparent Km values for LDL degradation in control and hCG-treated cells were similar (2.0 and 2.6 micrograms/ml, respectively). As little as 10 mIU hCG/ml stimulated LDL metabolism in a time-dependent fashion: a stimulatory effect was detected within 6 h of exposure to hCG and was greater after 24 h. AG attenuated but did not prevent the hCG- or 8-bromo-cAMP-stimulated increase in both LDL uptake and metabolism, although it completely inhibited the steroidogenic response. AG alone had no significant effect on [125I] LDL metabolism. We conclude that hCG and cAMP increase LDL metabolism by luteinized human granulosa cells. These effects are apparently not simply a consequence of enhanced cellular cholesterol metabolism to steroids.

Effects of prolactin on progestin secretion by human granulosa cells in culture.

Concentrations of human prolactin (hPrl) greater than or equal to 600 ng/ml produced inhibition of progestin production in cultures of granulosa cells pooled from follicles of women stimulated with clomiphene citrate-human chorionic gonadotropin (hCG). However, cells collected from follicles of human menopausal gonadotropin (HMG)-hCG-treated patients did not demonstrate a significant reduction in progestin secretion in response to hPrl. We conclude that high concentrations of hPrl can result in inhibition of steroidogenesis, but the expression of the inhibitory effects of Prl depends upon the hormonal treatments used to stimulate follicular growth.

Pathology of primary tumors and axillary lymph nodes in British and Japanese women with breast cancer.

Histopathologic features of the primary tumor and axillary lymph nodes from 97 consecutive patients with breast cancer from Japan were compared with those from 164 patients from England. Between the two groups, there were statistically significant differences in the morphology of the primary tumors regarding nuclear grade and patterns of tumor infiltration. In axillary lymph nodes, sinus histiocytosis was much more common in Japanese cases than in British cases, and was related to a diminished frequency of axillary node metastases. Germinal centers were also more common in the nodes of Japanese patients and were similarly associated with diminished frequency of metastases.

An experimental approach to the breast cancer, reserpine problem.

Reserpine was administered intraperitoneally 3 times weekly to inbred female Syrian BIO hamsters of the 15.16 strain previously shown to be susceptible to methylcholanthrene (MC) induction of mammary cancer. Other groups of hamsters received non-carcinogenic doses of MC along with the reserpine administrations, and an additional group received a carcinogenic dose of MC alone. This last group demonstrated that BIO 15.6 females were indeed susceptible to MC mammary cancer induction, since 4 mg of MC administered by stomach tube (a total dose of 200 mg) caused mammary cancer in 52% of the animals. Mammary cancer was not observed in any of the animals given reserpine or reserpine in combination with the non-carcinogenic dose of MC.

Bladder cancer as seen in giant histologic sections.

Giant sections from cystectomy specimens in 45 cases of bladder cancer were examined microscopically after en bloc fixation and processing. There were 35 transitional cell, seven squamous cell, and three mixed transitional cell and squamous carcinomas. Broad front invasion was associated with papillary and superficial tumors while tentacular invasion was associated with solid tumors and a generally poorer prognosis. Carcinoma in situ merged with the invasive cancer in 33 cases, and neoplasia in these cases tended to be multifocal. In 10 cases there was no carcinoma in situ next to the invasive lesion, and the cancer was unifocal. These findings support the concept that there may be two pathogenetic types of bladder cancer, one arising in an extensive field of abnormal epithelium and one arising in a focal area of abnormality. The findings also underscore the importance in clinical management of selected mucosal biopsies adjacent to the site of any visible bladder tumors.