Impaired hepatocyte glucose transport protein (GLUT2) internalization in chronic pancreatitis.

Chronic pancreatitis (CP) is associated with impaired glucose tolerance and with reduced hepatic sensitivity to insulin. We have previously shown that in normal and sham-operated rats, insulin suppresses hepatic glucose production, and this suppression is associated with a decrease in the hepatocyte plasma membrane-bound quantity of the facilitative glucose transport protein GLUT2. The insulin-mediated reduction in membrane-bound GLUT2 is impaired in CP, and may play a role in the glucose intolerance associated with CP. To determine whether GLUT2 is actively internalized and whether this mechanism is disordered in CP, livers from fed and fasting rats in whom CP had been induced 2-3 months earlier by pancreatic duct oleic acid infusion, and in sham-operated (sham) rats, were fractionated to yield endosome (E)- and plasma membrane (PM)-enriched fractions. Forty-five minutes after duodenal intubation alone (fasting) or intubation plus duodenal feeding, livers were removed, homogenized and ultracentrifuged, and microsomal pellets were separated by sucrose density gradient ultracentrifugation. GLUT2 content of fractions was determined by Western blotting and scanning densitometry. The E:PM ratio of GLUT2 increased from 0.68 +/- 0.11 (mean +/- SEM) in fasting sham livers (n = 8) to 1.04 +/- 0.09 in fed sham livers (n = 8; p < 0.05). However, there was no change in the E:PM ratio of GLUT2 in CP livers after duodenal feeding (0.90 +/- 0.12 vs. 0.86 +/- 0.10; n = 8,8; p = NS). To test our findings using confocal laser scanning microscopy, liver specimens from fed and fasting CP and sham rats were minced, fixed in 4% paraformaldehyde, sectioned, and stained with rabbit antirat GLUT2 antibody followed by rhodamine-labeled secondary antibody. GLUT2 was quantified by mean pixel intensity in an 8 x 16-pixel area of PM and a 16 x 16-pixel area of cytosol (CYT) in each of 30 random cells/field (400x) in each of three rats per group. As in the fractionation study, duodenal feeding increased the CYT:PM ratio of GLUT2 from 0.75 +/- 0.01 in fasting sham liver to 0.86 +/- 0.01 in fed sham liver (p < 0.0001), while the CYT:PM ratio in CP remained unchanged. We conclude that feeding induces a shift in GLUT2 from the plasma membrane to the endosomal pool. The feeding-induced internalization of GLUT2 is absent in livers from rats with CP and may play a role in the glucose intolerance associated with CP.