Human ovarian cancer, cell lines, and primary ascites cells express the human Mullerian inhibiting substance (MIS) type II receptor, bind, and are responsive to MIS.

Six human ovarian cancer cell lines and samples of ascites cells isolated from 27 patients with stage III or IV ovarian papillary serous cystadenocarcinoma were studied individually to test whether recombinant human Mullerian inhibiting substance (rhMIS) acts via its receptor. To do these experiments, we scaled up production of rhMIS and labeled it successfully with biotin for binding studies, cloned the human MIS type II receptor for mRNA detection, and raised antibodies to an extracellular domain peptide for protein detection. These probes were first tested on the human ovarian cancer cell lines and then applied to primary ovarian ascites cells. rhMIS inhibited colony growth of five of six cell lines that expressed the human MIS type II receptor mRNA by Northern analysis while not inhibiting receptor-negative COS cells. Flow cytometry performed on MIS-sensitive ovarian cancer cell lines demonstrated specific and saturable binding of rhMIS (Kd = 10.2 nM). Ascites cells from 15 of 27 or 56% of patients tested bound biotinylated MIS (MIS-biotin) and, of the 11 that grew in soft agarose, 9 of 11 or 82% showed statistically significant inhibition of colony formation. Of the 15 patients who bound biotinylated MIS, mRNA was available for analysis from 9, and 8 of 9 expressed MIS type II receptor mRNA by reverse transcription-PCR, showing a statistically significant correlation, compared with binding, by chi2 analysis (P = 0.025). Solid ovarian cancers were positive for the MIS type II receptor protein by immunohistochemical staining, which colocalized with staining for antibody to CA-125 (OC-125). Thus, the detection of the MIS type I receptor by flow cytometry may be a useful predictor of therapeutic response to MIS and may be a modality to rapidly choose patients with late-stage ovarian cancer for treatment with MIS.

Mullerian duct regression and antiproliferative bioactivities of mullerian inhibiting substance reside in its carboxy-terminal domain.

A 25-kilodalton dimeric carboxy-terminal fragment of the recombinant human Mullerian inhibiting substance protein (rhMIS) was produced by proteolytic cleavage with plasmin and purified by size-exclusion chromatography. The identity of the isolated dimer as the carboxy-terminal fragment was confirmed by gel electrophoresis and Western analysis. As was true of every sample of the holo molecule, all preparations of the carboxy-terminal domain of rhMIS (n = 10), when added in the 0.5-5.0 micrograms/ml range, exhibited a dose-dependent partial to complete regression of the 14.5-day fetal rat Mullerian duct in an organ culture assay. The carboxy-terminal dimer also inhibited, in a dose-dependent manner, the growth of A431 cells in monolayer cultures. Daily addition of 5, 10, or 20 micrograms carboxy-terminus for 3 days resulted in 0%, 25%, and 100% inhibition of cell proliferation, respectively. Similar and higher doses of holo rhMIS had no or inconsistent antiproliferative activity (0-34% inhibition), even though the preparations caused Mullerian duct regression. All amino-terminal fragments prepared using this separation protocol were found to be inactive in these assays. These findings suggest that the bioactivity of rhMIS as a regressor of fetal Mullerian ducts and an inhibitor of A431 cell growth resides in its carboxy-terminal domain. These results indicate that the urogenital ridge tissue, but not A431 cells in culture, may be capable of cleaving intact MIS to a biologically active conformation.

Human müllerian inhibiting substance: enhanced purification imparts biochemical stability and restores antiproliferative effects.

Separation of copurifying protease activity from recombinant human Müllerian inhibiting substance (rhMIS) bound to a monoclonal antibody immunoaffinity column by a high-salt wash results in cleaner preparations of rhMIS resistant to cleavage upon storage. In addition, an inhibitor of rhMIS antiproliferative activity is removed. Proteolytic cleavages produced by either a copurifying protease or exogenous plasmin occur at residues 229 and 427 but do not abolish rhMIS biological activity. This report details the modified immunoaffinity column isolation protocol suitable for proteins such as rhMIS and describes the biochemical and antiproliferative properties of this protein.

ATP synthase complex from bovine heart mitochondria. Passive H+ conduction through F0 does not require oligomycin sensitivity-conferring protein.

Oligomycin sensitivity-conferring protein (OSCP) is a water-soluble subunit of bovine heart mitochondrial H(+)-ATPase (F1-F0). In order to investigate the requirement of OSCP for passive proton conductance through mitochondrial F0, OSCP-depleted membrane preparations were obtained by extracting purified F1-F0 complexes with 4.0 M urea. The residual complexes, referred to as UF0, were found to be deficient with respect to OSCP, as well as alpha, beta, and gamma subunits of F1-ATPase, but had a full complement of coupling factor 6 as judged by sodium dodecyl sulfate-polyacrylamide gel electrophoresis and Western blotting techniques. These UF0 complexes had no intrinsic ATPase activity and were able to bind nearly the same amount of F1-ATPase in the presence of either OSCP or NH4+ ions alone, or a combination of the two. However, the preparations exhibited an absolute dependency on OSCP for conferral of oligomycin sensitivity to membrane-bound ATPase. The passive proton conductance in UF0 proteoliposomes was measured by time-resolved quenching of 9-amino-6-chloro-2-methoxyacridine or 9-aminoacridine fluorescence following a valinomycin-induced K(+)-diffusion potential. The data clearly establish that OSCP is not a necessary component of the F0 proton channel nor is its presence required for conductance blockage by the inhibitors oligomycin or dicyclohexylcarbodiimide. Furthermore, OSCP does not prevent or block passive H+ leakage. Comparisons of OSCP with the F1-F0 subunits from Escherichia coli and chloroplast lead us to suggest that mitochondrial OSCP is, both structurally and functionally, a hybrid between the beta and delta subunits of the prokaryotic systems.