By use of agarose gel electrophoresis, the sieving of spherical particles in agarose gels has been quantitated and modeled for spheres with a radius (R) between 13.3 and 149 nm. For quantitation, the electrophoretic mobility has been determined as a function of agarose percentage (A). Because a previously used model of sieving [D. Rodbard and A. Chrambach (1970) Proc. Natl. Acad. Sci. USA 65, 970-977] was found incompatible with some of these data, alternative models have been tested. By use of an underivatized agarose, two models, both based on the assumption of a single effective pore radius (PE) for each A, were found to yield PE values that were independent of R and that were in agreement with values of PE obtained independently (PE = 118 nm X A-0.74): sieving by altered hydrodynamics in a cylindrical tube of radius, PE, and sieving by steric exclusion from a circular hole of radius, PE. The same analysis applied to a 6.5% hydroxyethylated commercial agarose yielded a steeper PE vs A plot and also agreement of the above two models with the data. The PE vs A plot was significantly altered by both further hydroxyethylation and factors that cause variation in the electro-osmosis found in commercial agarose.

CaM kinase II, a multifunctional Ca2+/calmodulin-dependent protein kinase, is expressed in the pancreatic beta-cell and is activated by glucose and other secretagogues in a manner correlating with insulin secretion. It is proposed that the activation of CaM kinase II mediates some of the actions of Ca2+ on the exocytosis of insulin secretory granules. This suggestion is supported by the localization of CaM kinase II to the insulin secretory granule and by the identification of two secretory-relevant proteins, MAP-2 and synapsin I, as endogenous substrates in the beta-cell. Mechanistically, CaM kinase II appears to be involved in secretory steps proximal to granule fusion at the plasmalemma, and may facilitate protracted secretion through control of the interaction of granules with the cell cytoskeleton and their mobilization from intracellular synthesis sites. Through its unique regulatory properties, however, CaM kinase II is predicted to serve in more specialized aspects of the secretory process. In particular, the ability of CaM kinase II to remain active after cell stimulation is suggested to represent a mechanism by which releasable pools of granules are replenished between stimuli.

The present experiments were undertaken to examine the hypothesis that glucose-induced increased de novo synthesis of 1,2-diacyl-sn-glycerol (which has been observed in a number of different tissues, including retinal capillary endothelial cells exposed to elevated glucose levels in vitro) and associated activation of protein kinase C may play a role in mediating glucoseinduced vascular functional changes. We report here that twice daily instillation of 30 mM glucose over 10 -d in a rat skin chamber granulation tissue model induces approximately a 2.7-fold increase in diacylglycerol (DAG) levels (versus tissues exposed to 5 mM glucose) in association with marked increases in vascular clearance of albumin and blood flow. The glucose-induced increase in DAG levels as well as the vascular functional changes are prevented by addition of 3 mM pyruvate. Pharmacological activation of protein kinase C with the phorbol ester TPA in the presence of 5 mM glucose increases microvascular albumin clearance and blood flow, and similar effects are observed with 1-monoolein (MOG), a pharmacological inhibitor of the catabolism of endogenous DAG. A pharmacological inhibitor of protein kinase C (staurosporine) greatly attenuates the rise in microvascular albumin clearance (but not the rise in blood flow) induced by glucose or by MOG. These findings are compatible with the hypothesis that elevated concentrations of glucose increase tissue DAG content via de novo synthesis, resuIting in protein kinase C activation, and that these biochemical events are among the factors that generate the increased microvascular albumin clearance. (J. Clin. Invest. 1991.

-aminophenoxy) ethane-N, N, N,N-tetraacetate). Glucose and GLP-1 also synergistically activated NFAT (nuclear factor of activated T-cells)-mediated transcription from a minimal promoter construct containing tandem NFAT consensus sequences. Furthermore, twopoint base pair mutations in any of the three identified NFAT sites within the rat insulin I promoter resulted in a significant reduction in the combined effect of glucose and GLP-1. These data suggest that the synergistic action of glucose and GLP-1 to promote insulin gene transcription is mediated through NFAT via PKA-and calcineurin-dependent pathways in pancreatic ␤-cells.

'Initial' and 'total' activities of 3-hydroxy-3-methylglutaryl-CoA reductase (HMG-CoA reductase) were measured in cold-clamped samples of liver from rats at 2h intervals throughout the 24h light/dark cycle. Initial activities were obtained in microsomes (microsomal fractions) isolated and assayed in the presence of 100mM-KF, whereas 'total' activities were measured in microsomes prepared from the same homogenates but washed free of KF and incubated with exogenous partially purified rat liver protein phosphatase. The initial/total-activity ratio for HMG-CoA reductase underwent a diurnal cycle, which had a nadir 4h into the light phase (when initial activity was 28% of total activity) and a peak 12h later, i.e. 4h into the dark phase (when initial activity was 80% of total activity). These low and high points of the cycle were separated by gradual steady changes in the ratio. The characteristics of this diurnal cycle were different from those of the cycle observed for total activity, which had a plateau of high activity between 2 and 10h into the dark cycle preceded and succeeded by a very rapid increase and decrease, respectively, in the total activity of HMG-CoA reductase. The combination of the two cycles resulted in the dampening of the resultant cycle for the initial or effective activity of HMG-CoA reductase, such that the changes in initial activity around the beginning and and end of the dark phase were more gradual than would otherwise have been the case if the initial/total-activity ratio for HMG-CoA reductase were constant throughout the diurnal cycle. The physiological implications of the observed diurnal variation in the fraction of hepatic HMG-CoA reductase in the active form are discussed.