The factors operating at the apical side of the endocrine cell releasing cholecystokinin (CCK) were investigated using the isolated vascularly perfused rat duodenojejunum. In the protease-free intestinal segment, a 30-min infusion of glucose (280 mM), oleic acid (100 mM), or triglycerides containing short- or long-chain fatty acids did not alter significantly the basal level of portal CCK-like immunoreactivity (CCK-LI), while octanoic acid (100 mM) produced a transient rise of plasma CCK-LI to approximately 250% of basal. Infusion of proteins (5% solutions of ovalbumin or casein) or of a mixture of all amino acids brought about a modest CCK secretion. In contrast, isocaloric amounts of an ovalbumin hydrolysate produced a sharp rise of portal CCK-LI to 530% of basal followed by a well-sustained plateau secretion (420% of basal) until the end of the infusion. An acid casein hydrolysate induced a slightly less pronounced CCK-LI release and was followed in decreasing order by meat, casein, and soybean peptones. Simultaneous infusion of trypsin with ovalbumin or casein hydrolysate reduced by approximately 60% the CCK release induced by peptone alone. This effect was reversed by coinfusion of soybean trypsin inhibitor (SBTI) with the trypsin-peptone mixture. Arterial infusion of tetrodotoxin (10(-6) M) or atropine (10(-5) M) had no significant effect on the trypsin-induced inhibition of peptone-mediated CCK-LI release. Administration of SBTI or camostate alone or in combination with trypsin did not alter basal CCK. Monitor peptide produced a dose-dependent transient rise of portal CCK-LI over the range from 2 to 12 micrograms.(ABSTRACT TRUNCATED AT 250 WORDS)