Search

We determined whether intraportal caffeine infusion, at rates designed to create concentrations similar to that seen with normal dietary intake, would enhance net hepatic glucose uptake (NHGU) during a glucose load. Dogs (n = 15) were implanted with sampling and infusion catheters as well as flow probes >16 d before the studies. After a basal sampling period, dogs were administered a somatostatin infusion (0-150 min) as well as intraportal infusions of glucose [18 micromol/(kg . min)], basal glucagon [0.5 ng/(kg . min)], and insulin [8.3 pmol/(kg . min)] to establish mild hyperinsulinemia. Arterial glucose was clamped at 10 mmol/L with a peripheral glucose infusion. At 80 min, either saline (Control; n = 7) or caffeine [1.5 micromol/(kg . min); n = 8] was infused into the portal vein. Arterial insulin, glucagon, norepinephrine, and glucose did not differ between groups. In dogs infused with caffeine, NHGU was significantly higher than in controls [21.2 +/- 4.3 vs. 11.2 +/- 1.6 micromol/(kg . min)]. Caffeine increased net hepatic lactate output compared with controls [12.5 +/- 3.8 vs. 5.5 +/- 1.5 micromol/(kg . min)]. These findings indicate that physiologic circulating levels of caffeine can enhance NHGU during a glucose load, and the added glucose consumed by the liver is in part converted to lactate.

Consumption of large amounts of coffee has been shown to decrease the incidence of type 2 diabetes. However, the specific compounds and mechanisms responsible for this effect are not known. The aim of this study was to determine the effects of a decaffeinated coffee extract and a synthetic quinide, representative of those found in roasted coffee, 3,4-diferuloyl-1,5-quinolactone, on insulin-stimulated glucose disposal and muscle glucose uptake. Experiments were performed on conscious rats during hyperinsulinemic, euglycemic clamps receiving gastric infusions of saline, a decaffeinated coffee extract (DECAF) (220 mg/kg), or 3,4-diferuloyl-1,5-quinide (DIFEQ) (110 mg/kg). Following treatment, rats received an intravenous bolus of deoxy-[2-3H] glucose to assess muscle glucose uptake (Rg, micromol x 100 g(-1) x min(-1)). Glucose infusions [mg/(kg x min)] required to maintain euglycemia during the tracer period were higher with DIFEQ (14.6 +/- 0.7) than with saline (10.8 +/- 0.7) and DECAF (11.5 +/- 1.1). Despite increased glucose requirements, Rg in skeletal (soleus, gastrocnemius, superficial vastus lateralis) and cardiac muscle were unchanged. DECAF or DIFEQ did not affect heart rate, blood pressure, plasma nonesterified fatty acids or liver aminotransferase activity. These results demonstrate that DIFEQ increases whole-body glucose disposal independently of skeletal muscle Rg.

To determine the contributions of transporter-mediated and passive absorption during an intraduodenal glucose infusion in a large animal model, six mongrel dogs had sampling catheters (portal vein, femoral artery, duodenum), infusion catheters (vena cava, duodenum) and a portal vein flow probe implanted 17 d before an experiment. Protocols consisted of a basal (-30 to 0 min) and an experimental (0-90 min) period. An intraduodenal glucose infusion of 44 micromol/(kg. min) was initiated at t = 0 min. At t = 20 and 80 min, 3-O-[3H]methylglucose and L-[14C]glucose (L-Glc) were injected intraduodenally. Phloridzin, an inhibitor of the Na+/K+ ATP-dependent transporter (SGLT1), was infused from t = 60 to 90 min in the presence of a peripheral isoglycemic clamp. Net gut glucose output was 21.1 +/- 3.0 micromol/(kg. min) from t = 0 to 60 min. Transporter-mediated glucose absorption was calculated using three approaches, which involved either direct measurements or indirect estimates of duodenal glucose analog radioactivities, to account for the assumptions and difficulties inherent to duodenal sampling. Values were essentially the same regardless of calculations used because transporter-mediated absorption was 89 +/- 1%, 90 +/- 2% and 91 +/- 2% of net gut glucose output. Phloridzin-induced inhibition of transporter-mediated absorption completely abolished passive absorption of L-Glc. We conclude that in dogs, transporter-mediated glucose absorption constitutes the vast majority of glucose absorbed from the gut and is required for passive glucose absorption. The method described here is applicable to investigation of the mechanisms of gut glucose absorption under a variety of nutritional, physiologic and pathophysiologic conditions.