Continuous labeling and pulse-chase techniques were employed to study the synthesis and secretion of multiple forms of immunoreactive beta- endorphin by cultured dispersed rat anterior lobe cells and intact neurointermediate pituitary lobe. Cell and medium extract immunoreactive beta-endorphin (specific immunoprecipitation and radioimmunoassay) exhibiting a Kav similar to authentic beta-endorphin upon gel filtration was characterized further by nonequilibrium isoelectric focusing, cation exchange chromatography, reverse phase high pressure liquid chromatography, and partial tryptic and chymotryptic mapping. Intact neurointermediate lobes incorporated radiolabeled amino acids into four to six forms of immunoreactive beta- endorphin. Four of these forms were physicochemically similar to authentic beta-endorphin, N-acetylated beta-endorphin, beta-endorphin- (1–27), and N-acetylated beta-endorphin-(1–27). Pulse-chase studies indicated that a beta-lipotropin-like molecule served as a metabolic intermediate for a beta-endorphin-like molecule. As beta-endorphin-like material accumulated in the cell, some of it was N-acetylated (approximately 18% at 2 hr chase and approximately 65% at 18 hr chase). At later chase times, beta-endorphin-(1–27)- and N-acetylated beta- endorphin-(1–27)-like peptides were the predominant molecular species detected. All endorphin forms were detected in unlabeled tissue maintained in culture or tissue continuously labeled for 72 hr and were released into the medium under basal, stimulatory (10(-8) M norepinephrine), or inhibitory (10(-7) M dopamine) incubation conditions. In all cases, beta-endorphin-(1–27)-like species were the predominant forms (more than 70% of total) present in the cells and released into the medium. In contrast, approximately 90% of radiolabeled immunoreactive beta-endorphin extracted from anterior lobe cells and medium similarly incubated appeared to represent the authentic beta-endorphin molecule. Continuous labeling (72 hr) revealed the beta-lipotropin/beta-endorphin molar ratio to be approximately 4. We conclude that, in anterior lobe, most of the beta-endorphin is not processed further and is released intact, while in neurointermediate lobe, it serves as a biosynthetic intermediate.