This thesis investigated the putative roles of A, adenosine receptor (AR) and caffeine (CAF, nonselective AR antagonist) in insulin regulation of skeletal muscle glucose transport. In Study 1, reduced skeletal muscle glucose uptake (~50%) was the major determinant of CAF impairment of insulin sensitivity during physiological hyperinsulinemia (100 [mu]U/ml) in humans. CAF did not affect insulin stimulation of muscle IRTK, IRS-1 PI3K and PBK activities, but increased plasma epinephrine and muscle cAMP concentrations, possibly indicating that this might contribute to its insulin antagonistic effects. In Study 2, CAF administered in conjunction with propranolol ([beta]-receptor blocker) abolished the CAF-induced impairment of glucose tolerance in humans. In contrast to previous reports in rat muscle, the insulin antagonistic effects of CAF are engendered by epinephrine and not AR antagonism in human skeletal muscle. In Study 3, the effects of adenosine on 3-O-methylglucose transport were characterized in isolated rat soleus muscle. Following adenosine removal (ADA and AOPCP), N6-cyclopentyladenosine (CPA, A1 agonist) caused a dose-dependent increase (~15%) in submaximal insulin (0.1 mU/ml), but not basal (no insulin) or maximal (10 mU/ml) insulin stimulated glucose transport. 8-cyclopentyl-1,3-dipropylxanthine (CPX, A1 antagonist) dose-dependently inhibited the maximal stimulatory effects of CPA on insulin (0.1 mU/ml)-mediated glucose transport. These effects were likely achieved by altering translocation and not intrinsic activity of GLUT4 because addition of CPA and CPX to muscles preincubated with 10 mU/ml insulin alone (30 min) did not alter glucose transport. These findings suggest A 1AR mediates adenosine-insulin interaction. Since the presence of A 2A and A2B, but not A1AR has been conclusively demonstrated in skeletal muscle, the possibility exists that A2AR might mediate adenosine-insulin interaction. The relevance of these data to humans is unknown since species difference in adenosine-insulin interaction and a lack of effect of adenosine on insulin sensitivity and muscle glucose transport in humans have previously been reported. CPA and CPX have greater affinity for binding AR than interstitial adenosine and despite their use at pharmacological concentrations, only a minor change in glucose transport (~15%) was observed. Collectively, these findings raise uncertainty as to whether adenosine is an integral component of insulin regulation of skeletal muscle glucose transport.