Origin of the decline in pancreatic amylase secretion with time in vitro

The release of alpha-amylase (EC 3.2.1.1) into the medium bathing strips of rabbit pancreas decreased gradually with time in vitro so that after 3 h of incubation the secretory rate was only 20% of that seen initially. This decrease was not the result of an intrinsic loss of cellular responsiveness or viability, because a cholinergic agonist was still able to augment amylase secretion from the tissue after this incubation period to the same, or greater, degree than from "fresh" tissue. Rather, the inhibition of secretion was apparently due to the release of a nondialyzable substance from the tissue, which appears to be amylase itself. The observations that led to this conclusion are 1) replacing the bathing medium every 30 min eliminated the time-dependent decrease in enzyme secretion, 2) this replacement was ineffective if the tissue was separated from the medium by a dialysis membrane, and 3) adding an excess of the amylase substrate glycogen to the medium also prevented the falloff in amylase secretion, the effect of glycogen presumably being due to the formation of an enzyme-substrate complex in the medium. These observations are consistent with other observations that demonstrate that the rate of enzyme secretion by the pancreas depends on the rate of removal of product from the site of its secretion [Isenman and Rothman, Proc. Natl. Acad. Sci. USA 74: 4068-4072, 1977; Isenman and Rothman, Science 204: 1212-1215, 1979; Ho and Rothman, Am. J. Physiol. 242 (Gastrointest. Liver Physiol. 5): G32-G39, 1982].