Apparent alpha-inhibin subunit immunoactivity in porcine and ovine luteal extracts is due to interference by cytosolic proteases in the assay.

Immunoreactive alpha-inhibin (ir-inhibin) was measured in luteal homogenates and subcellular fractions of ovine and porcine corpora lutea (CL) and in pig granulosa cells (GCs), using a sensitive radioimmunoassay specific for the 1-26 amino acid sequence of the N-terminus of the alpha chain of porcine inhibin (p1-26 alpha-inhibin). Inclusion of N-ethylmaleimide (N-EM) and/or EDTA in the immunoassay had no effect on the measurement of p1-26 alpha-inhibin peptide standards, on ir-inhibin levels in ovine follicular fluid and serum, or on ir-inhibin in subcellular fractions of pig GC. Fractionation of porcine GC homogenates on sucrose gradients demonstrated a major particular peak of ir-inhibin (buoyant density, 1.15-1.21 g/cm3) with variable activity in the cytosol. The particulate ir-inhibin peak was released into the cytosol by pretreatment of GC homogenates with the saponin, digitonin, prior to fractionation. Porcine GC extracts contained a protein (M(r) 45,000) which immunoblotted against p1-26 alpha-inhibin antibody. In the absence of inhibitors of proteolysis, apparent ir-inhibin activity was very high in extracts of sheep and pig CL. However, inclusion of N-EM or EDTA in the radioimmunoassay significantly reduced ir-inhibin levels in porcine and ovine CL extracts in a dose-dependent manner. Measurements of peptide tracer integrity indicated that porcine luteal cytosol degraded 125I-labelled p1-26 alpha-inhibin peptide. Subcellular fractionation studies demonstrated high levels of apparent ir-inhibin in luteal cytosol fractions, with only minor activity peaks associated with particulate fractions; however, this material was not releasable by digitonin. Immunoblotting of detergent extracts of porcine luteal particulate fractions failed to demonstrate alpha-inhibin material, and immunocytochemical localization studies of alpha-inhibin in porcine and ovine luteal sections were negative. Our results are consistent with the intracellular packaging/storage of a form of alpha-inhibin (M(r) similar to that of alpha-inhibin subunit precursor) in the porcine granulosa cell. However, luteinization of the porcine follicle was associated with a dramatic fall in ir-inhibin content, and the loss of immunostaining for alpha-inhibin peptides. We conclude that porcine and ovine CL contain little, if any, authentic inhibin. These studies emphasize the importance of excluding proteolytic artefacts when measuring biological peptides in luteal tissue extracts by radioimmunoassay.