The time-dependent release of molecular variants of cholecystokinin (CCK) into the circulation was studied before and 1, 2, and 4 h after a test meal in six healthy volunteers. At each time period, 100 ml of blood were drawn in a manner to inhibit CCK degradation. Plasma was formed and CCK concentrated by Sep-Pak C18 cartridge chromatography. Molecular variants of CCK and gastrin were well separated from each other by high-performance liquid chromatography (HPLC). Molecular forms of CCK and gastrin were measured by radioimmunoassay using an antibody that requires the presence of the carboxyl-terminal phenylalanine amide for full recognition, implying that biologically active forms were detected. HPLC elution positions of gastrin forms were determined using a gastrin-specific antibody. Chromatographic separation of CCK from gastrin forms was complete, allowing separate integration of gastrin and CCK forms. Therefore no subtraction of gastrin-like immunoreactivity from CCK-like immunoreactivity (CCK-LI) was necessary and CCK-LI could be directly determined. Peaks of CCK-LI were integrated in the column eluates and the plasma concentrations were calculated. Total plasma CCK-LI rose from a value of 2.4 +/- 0.6 pM before the test meal to 6.4 +/- 0.8, 6.6 +/- 0.9, and 5.8 +/- 1.2 pM 1, 2, and 4 h postprandially. The major molecular forms released into the circulation eluted on HPLC in the position of CCK-58 and CCK-39 (which coelutes with CCK-33). Minor amounts were detected in the position of CCK-8. There was no significant difference in the relative proportions of the molecular forms released at the different time periods.(ABSTRACT TRUNCATED AT 250 WORDS)