Glenn Mortimore

The continuous turnover of intracellular protein and other macromolecules is a basic cellular process that serves, among other functions, to regulate cytoplasmic content and provide amino acids for ongoing oxidative and biosynthetic reactions during nutrient deprivation. The intensity of breakdown and pattern of regulation, though, vary widely among cells. Rat hepatocytes, for example, exhibit high absolute rates of proteolysis and regulatory effects that diminish during starvation, while corresponding responses in skeletal and cardiac muscle move in the opposite direction. It is also becoming apparent that effects of insulin and other acute regulatory agents on muscle breakdown are limited to nonmyofibrillar components. The latter may be sequestered and degraded within autophagic vacuoles, whereas myofibrillar proteins require an initial attack by calcium-dependent proteases in the cytosol. By contrast, most if not all of the breakdown of resident (long-lived) proteins as well as RNA in the hepatocyte can be explained by lysosomal mechanisms. The uptake of cytoplasmic components by lysosomes can be divided into two major categories, macroautophagy and micro- or basal autophagy. The first is induced by amino acid or insulin/serum deprivation. In the hepatocyte, amino acids alone can regulate this process almost instantaneously over two thirds of the full range of proteolysis, 4.5% to 1.5% per hour. Glucagon, cyclic AMP, and beta-agonists also stimulate macroautophagy in hepatocytes but have opposite effects in skeletal and cardiac myocytes. Basal autophagy differs from the macro type in that the cytoplasmic "bite" is smaller and sequestration is not acutely regulated. It is, however, adaptively decreased during starvation in parallel with absolute rates of basal turnover. Since endoplasmic reticulum comprises an appreciable fraction of the vacuolar content, volume sequestration would be compatible with the known heterogeneity of individual protein turnover if some proteins (or altered proteins) selectively bind to membranes. The amino acid control of macroautophagy in the hepatocyte is accomplished by a small group of direct inhibitors (Leu, Tyr/Phe, Gln, Pro, Met, Trp, and His) and the permissive effect of alanine whereas only leucine is involved in myocytes and adipocytes. Of unusual interest is the fact that the inhibitory amino acid group alone evokes responses in perfused livers that are identical to those of a complete plasma mixture at 0.5 and 4 times normal plasma levels but loses effectiveness almost completely at normal concentrations.(ABSTRACT TRUNCATED AT 400 WORDS)

Abstract Proteolysis was assessed during perfusion of isolated livers of nonfasted rats from the turnover of free valine, measured after single pulse additions of l-valine-1-14C to the perfusate, and from the release of label from livers previously labeled with l-valine-1-14C in vivo and then perfused with 15 mm unlabeled valine to minimize reincorporation. Since valine is not synthesized in the rat and was shown not to be metabolized significantly in these experiments, it could be assumed that any net gain or loss of free valine would reflect net changes in the protein content of the system. In control experiments, free valine accumulated slowly for 15 to 30 min; the rate then spontaneously increased rather abruptly and remained elevated. Since the utilization of extracellular valine by protein synthesis did not fall off with time, the spontaneous increase in valine accumulation was ascribed to an enhancement of proteolysis. Insulin in vitro completely inhibited this spontaneous net increase and also inhibited the release of label from previously labeled livers. No specific hormonal effects were noted on the incorporation of free valine into protein or its distribution in intra- and extracellular water. The inhibitory effect of insulin, which was sustained for periods up to 180 min of perfusion, was estimated to represent 1.0% of total liver protein per hour.

Amino acid deprivation and glucagon are both potent inducers of autography and proteolysis in liver. Because glucagon enhanced the metabolic utilization of some amino acids, the catabolic response to both of these stimuli could be achieved by a lowering of intracellular amino acid pools. Alternatively, glucagon could act independently of amino acids. To clarify the mode of hormonal action and also the relationship between the two cellular responses, livers from fed rats were perfused, with and without glucagon, with plasma amino acids over a concentration range of 0 to 10 times normal. Individual amino acids constancy at each level was ensured by perfusion in the single-pass mode. Amino acids alone strongly regulated autophagy and proteolysis in a coordinated fashion; maximal suppression was achieved at twice normal concentration; both effects increased rapidly to maximum at less than normal concentration. Corresponding effects of glucagon, however, could be elicited only at intermediate amino acid levels. None was noted at 4 and 10 times normal; at 0, hormonal stimulation was minimal. The amino acid inhibition was selective because it did not block cyclic AMP production or glycogenolysis. Intracellular pool measurements and systematic alteration of perfusate amino acid composition indicated that the autophagic and proteolytic effects of glucagon are mediated by a hormonally induced depletion of glycine, alanine, glutamate, and glutamine; of these, glutamine alone is the most effective. We conclude that the stimulation of intracellular protein degradation in liver is a manifestation of deprivation-induced autophagy which results from a decrease in certain intracellular glucogenic amino acids, notably glutamine.

Abstract Livers from nonfasted rats were perfused in situ with concentrations of valine (plus [1-14C]valine) in the medium ranging from 0.3 to 15 mm. The relationship of total free intracellular valine (y) to external valine (x) was of the form y = a + bx, where the slope was about 1.0 and the intercept nearly 0.4 mm. At low valine concentrations the ratio of intracellular to extracellular valine, after isotopic equilibrium was attained, was less than 0.5; as external valine was increased, the ratio increased and approached unity. These findings suggested that the intracellular valine pool comprised two components, one which readily equilibrated with extracellular valine and another which was independent of external valine. Rates of valine incorporation into liver protein, calculated from the average specific activity of total intracellular valine, were not constant over the concentration range of external valine, as expected, but were 2- to 3-fold greater at the lowest concentrations than at the highest levels. This lack of constancy was explained by assuming that the two valine components represented two separate pools and that precursor sites of protein synthesis were in continuity with the expandable or equilibrating pool. Thus if label transported inward from the medium were restricted to the expandable pool, its specific activity and that of precursor amino acids would be higher than the determined specific activity of total intracellular valine. Recalculation of valine incorporation, based on the presumed specific activity of the expandable pool, gave uniform rates at all external valine concentrations. The second or nonexpandable pool, which apparently failed to equilibrate with label in the medium, was thought to arise from intracellular proteolysis since total intracellular and extracellular specific activities were the same when the only source of free label was the breakdown of protein in perfused livers that had been previously labeled with [1-14C]valine in vivo.