Inhibition of macroautophagy and proteolysis in the isolated rat hepatocyte by a nontransportable derivative of the multiple antigen peptide Leu8-Lys4-Lys2-Lys-beta Ala.

The multiple antigen peptide derivative, Leu8-Lys4-Lys2-Lys-beta Ala (Leu8-MAP), was synthesized by attaching the carboxyl of leucine to the NH2 termini of a branched lysine core, termed MAP, creating a molecule of about 1900 Da with 8 leucine residues. On a molar basis (independent of the number of leucine substitutions), Leu8-MAP was as effective as leucine in suppressing macroautophagy and proteolysis; moreover, it exhibited the same apparent Km (about 0.1 mM). The effect was specific for leucine since Ile8-MAP was inactive. It is of interest, though, that Leu8-MAP did not elicit the multiphasic response typical of leucine but instead evoked the single site inhibition normally seen with leucine plus the co-regulator alanine. Some free leucine was produced from Leu8-MAP during hepatocyte incubations, but the amounts were insufficient to account for the inhibition. Although this degradation created species of Leu-MAP that had lost 1-3 residues of leucine, their inhibitory effectiveness was not diminished. Because the extracellular/intracellular distribution ratio of [3H]-Leu8-MAP was 100:1 or greater, the direct transport of Leu8-MAP across the plasma membrane into the cytosolic compartment can be excluded. Hence, cytosolic concentrations of Leu8-MAP will be at least 100-fold smaller than those of leucine under conditions of comparable proteolytic inhibition. For these and related reasons, effects attributable to the recognition of Leu8-MAP cannot be explained by signals generated within the cytosol. They could, however, be mediated from site(s) on the plasma membrane or within associated vesicles.

Leucine-specific binding of photoreactive Leu7-MAP to a high molecular weight protein on the plasma membrane of the isolated rat hepatocyte.

Leu8-MAP (Multiple Antigen Peptide) is an effective inhibitor of macroautophagy and proteolysis in the isolated rat hepatocyte, having an apparent Km (0.1 mM) equaling leucine. Since it is not transported into the cytosolic compartment, it very likely mediates its effect through a plasma membrane site. In an attempt to identify the site we photoreacted intact cells with a biologically active, iodinatable azide derivative of Leu7-MAP. A approximately 340,000 M(r) protein whose labeling was protected 83% with 20 mM Leu was found in plasma membrane fractions when electrophoresed in 7.5-20% gradient gels under nonreducing conditions; addition of 20 mM dithiothreitol generated smaller m.w. products, possibly subunits, of consistent size. No specific labeling was observed with photoreactive derivatives of Ile7-MAP or Val7-MAP.

De novo autophagic vacuole formation in hepatocytes permeabilized by Staphylococcus aureus alpha-toxin. Inhibition by nonhydrolyzable GTP analogs.

The role of GTP-binding proteins in autophagic vacuole formation was investigated in isolated rat hepatocytes permeabilized by alpha-toxin from Staphylococcus aureus, an agent which creates stable plasma membrane channels allowing exchange of small (< or = 1000 Da) molecules. Vacuole formation was monitored from the uptake of 125I-tyramine-cellobiitol (125ITC) into osmotically sensitive vacuoles isolated on colloidal silica density gradients. Separation was based on an established observation that autophagic vacuoles are retained in a heavy midgradient band when samples are layered, but are selectively shifted to dense fractions when they are previously dispersed in the gradient material. The vacuolar uptake of 125ITC was concentration-dependent and required exogenous ATP: 94% was directly mediated by sequestration; 6% was acquired by fluid-phase endocytosis as monitored by [carboxyl-14C]dextran-carboxyl. Although the amino acid control of proteolysis was lost, addition of the nonhydrolyzable GTP analog GTP gamma S (as well as GMP-PNP) decreased fractional rates of direct vacuolar 125ITC uptake and long-lived proteolysis by similar amounts (1.02-1.03% h-1), substantiating the notion that the effects were the direct result of autophagic inhibition. These and associated findings, supported by quantitative electron microscopy, indicate the presence of ongoing macro- and microautophagy in alpha-toxin-permeabilized cells and suggest that one or more GTP-binding proteins is required in macroautophagic vacuole formation.

Multiphasic control of proteolysis by leucine and alanine in the isolated rat hepatocyte.

Autophagically mediated proteolysis in the perfused rat liver is under complex multiphasic control by a small group of amino acids dominated by leucine. Because there have been no prior reports of such regulation in the isolated hepatocyte, our goal was to determine whether it is a manifestation of interactions between diverse cells in the intact liver or, alternatively, the expression of a unique control mechanism within a single population of cells. Hepatocytes were isolated from livers of ad libitum-fed rats and incubated with cycloheximide at low density (approximately 10(6) cells/ml) for the determination of valine release. As in perfusion experiments with synchronously fed rats, proteolytic responses to leucine in cells from fed rats were mediated through two inhibitory mechanisms that alternated randomly on a day-to-day basis. The first (L) represented a typical multiphasic dose-response with low- and high-concentration inhibition separated by a sharp zonal loss of inhibition that could be abolished by alanine. The second (H) mediated inhibition only at high concentrations. It disappeared after 24 h of starvation, leaving L as the prevailing mode. The findings indicate that both macroautophagy and the multiphasic mechanism for regulating it coexist in a single population of hepatocytes, making the cells suitable for studies aimed at defining the putative plasma membrane site of leucine recognition.

Parallel control of hepatic proteolysis by phenylalanine and phenylpyruvate through independent inhibitory sites at the plasma membrane.

Intracellular protein degradation in the rat hepatocyte is regulated by 7 amino acids of which Leu, Gln, and Tyr play major roles. Although Phe is known to inhibit proteolysis as effectively as Tyr at high concentrations, it has not been regarded as an active regulator because of its rapid hydroxylation to Tyr. We now show that proteolytic responses to Phe during liver perfusion differ strikingly from effects of the multiphasic regulators Leu, Gln, and Tyr in eliciting mirror image responses at half-normal and normal plasma concentrations. Since response curves to phenylpyruvate and Phe were identical, we considered the possibility that phenylpyruvate mediated its anomalous inhibition intracellularly. However, when phenylpyruvate was produced from phenyllactate intracellularly at a rate providing the same rate of transamination (and intracellular concentration) as that derived from the uptake of phenylpyruvate, no response was obtained. Hence, the effect of phenylpyruvate was not initiated within the cell but more likely from the plasma membrane. Comparable evidence for Phe is less direct. Recent findings indicate that recognition sites for Leu and Gln are located at the plasma membrane. Since Phe augments the concerted inhibition by Leu and Gln at 4-fold normal levels, Phe is probably recognized in close proximity to them. However, the failure of phenylpyruvate to substitute for Phe in this interaction suggests that proteolytic inhibition by phenylpyruvate and Phe is mediated through similar, although independent, plasma membrane sites.

Control of hepatic proteolysis by leucine and isovaleryl-L-carnitine through a common locus. Evidence for a possible mechanism of recognition at the plasma membrane.

Deprivation-induced proteolysis in the perfused rat liver is controlled through the multiphasic action of 7 regulatory amino acids of which L-leucine plays the dominant role. Recently, isovaleryl-L-carnitine (IVC) was shown to mimic the leucine's effects, suggesting that the two molecules share structural features that are recognized at a common site(s). In this study we find that each evokes identical responses consisting of inhibitory effects at 0.08 and 0.8 mM, separated by a sharp zonal loss of inhibition at 0.15 mM. As monitored by density shifts of beta-hexosaminidase in colloidal silica gradients, macroautophagy is suppressed by both. Responses to Leu and IVC at 0.08 and 0.15 mM are stereospecific and require a reactive group at the alpha-carbon (or equivalent) and a high degree of branched chain specificity. In addition, 0.5 mM Ala coregulates with IVC and Leu by decreasing the zonal loss at 0.15 mM. The fact that the multiphasic responses can be duplicated with equimolar mixtures of Leu + IVC indicates that both react at the same site(s). IVC is readily taken up by a saturable process, but owing to its rapid hydrolysis in the cell, the ratio of internal to external IVC remains low over a 4-fold concentration range. These findings, together with a kinetic analysis of concerted responses to regulatory amino acids, suggest that the recognition sites are at a position in the cell, possibly at the plasma membrane, to react reversibly with plasma amino acids.

4-Amino-6-methylhept-2-enoic acid: a leucine analogue and potential probe for localizing sites of proteolytic control in the hepatocyte.

A recent analysis of leucine analogues has suggested that the carboxyl group is not required for mediating low concentration proteolytic inhibition in liver cells. In designing a probe to localize the regulatory site(s), we tested this hypothesis by synthesizing an analogue with a 2-carbon insert between the carboxyl and alpha-carbon. The Wittig product, a trans olefin, was fully active. Surprisingly, low concentration activity was lost when the double bond was eliminated by hydrogenation although some inhibitory effectiveness at high concentrations was evident. Since the double bond extends the carboxyl group away from the alpha-carbon, the results support the above hypothesis as well as the feasibility of adding functional groups to the carboxyl end of leucine.

Uptake and degradation of cytoplasmic RNA by hepatic lysosomes. Quantitative relationship to RNA turnover.

Past evidence has suggested that the lysosomal pathway is an important site of cytoplasmic RNA degradation in the hepatic parenchymal cell (Lardeux, B. R., Heydrick, S. J., and Mortimore, G. E. (1987) J. Biol. Chem. 262, 14507-14519). We now provide additional support for this notion by quantitating degradable RNA in lysosomes and correlating its pool size with hepatic RNA degradation. Rat livers, previously labeled with [6-14C]orotic acid, were perfused with graded levels of amino acids over the full range of induced autophagy; RNA degradation was determined from [14C]cytidine release. Close correspondence between the marker beta-acetylglucosaminidase and the breakdown of RNA to cytidine in subcellular fractions indicated that the lysosome was the main site of catabolism, a conclusion supported by the fact that degradation was enhanced when external pH was lowered from 7 to 6. Although [14C]cytidine was also released in homogenates by the action of natural ribonucleases on cytosolic RNA, this source was eliminated by unlabeled exogenous RNA. The size of the degradable RNA pool in lysosomes, determined from the total release of cytidine in homogenates, correlated directly with rates of hepatic RNA degradation over the full range of basal and induced degradation. A direct correlation was also seen between RNA degradation and cytidine pools within lysosomal particles. Because cytosolic cytidine was not taken up by lysosomes under these conditions, the pool could only have arisen from the breakdown of intralysosomal RNA. As determined by cytidine production, these findings support the view that the lysosomal-vacuolar system is the main, if not sole, site of induced and basal RNA degradation in liver.

Amino acid control of proteolysis in perfused livers of synchronously fed rats. Mechanism and specificity of alanine co-regulation.

The primary control of autophagically mediated proteolysis in perfused rat liver is carried out via two alternate mechanisms in response to specific regulatory amino acids. One (L) elicits direct inhibition at low and high plasma levels, but requires a co-regulatory amino acid to express inhibition at normal concentrations. The second (H) is ineffective at normal levels and below, but active at higher concentrations. Because regulation is subject to unpredictable variability with ad libitum feeding, we have utilized rats synchronously fed 4 h day-1 to stabilize responses. Proteolytic control is seen to evolve in stages: H appears 12 h after the start of feeding; by 18 h L emerges, alternating with H in a statistically predictable way; with omission of the 24-h feeding, H disappears and L remains constant through 42 h. In both 18- and 42-h rats, alanine, glutamate, and aspartate exhibit similar inhibitory activity when added singly to the regulatory group at normal plasma concentrations. However, since alanine, but not glutamate or aspartate, evokes proteolytic acceleration when it is deleted from a full plasma mixture, alanine appears to be the sole co-regulator. Alanine yields co-regulatory effects with normal plasma leucine (0.2 mM) in 18- and 42-h animals and interacts synergistically with 0.8 mM leucine in 42-h but not in 18-h rats where leucine alone inhibits strongly. Because the inactivation of alanine amino-transferase by aminooxyacetate (determined from the conversion of [14C]alanine to glucose) does not alter the co-regulatory and synergistic effects of alanine, regulation by alanine must be mediated from a site of recognition before transamination.

Modulation of kidney cell protein degradation by insulin.

Insulin is an established regulator of intracellular proteolysis in several mammalian tissues but little is known about its role in the kidney. The present study was undertaken to determine whether insulin influences protein degradation in isolated rat renal proximal tubules and to investigate its mechanism of action in cultured proximal-like tubular epithelial cells from the opossum kidney. Long-lived protein degradation was determined from the release of carbon 14-labeled valine from previously labeled cellular protein under conditions designed to minimize label reutilization. In isolated tubules, the mean control rate of proteolysis was 2.18% per hour, indicating an appreciable turnover of cellular protein. Insulin (10(-6) mol/L) decreased the rate by 23%. In cultured kidney cells, the rate of protein degradation averaged 1.25% per hour in the presence of serum and 1.68% per hour in its absence, an increase of 34%. High insulin concentrations suppressed this acceleration completely, and physiologic levels inhibited it partially. No evidence was obtained to indicate that insulin action is mediated through stimulation of Na(+)-H+ antiport or through increased amino acid utilization. Ammonium chloride, however, strongly attenuated the serum deprivation response and the inhibitory effect of insulin. The exact mechanisms whereby insulin inhibits proteolysis is not known, but these findings are consistent with an inhibitory action of insulin on the lysosomal pathway.

Mechanism and regulation of protein degradation in liver.

The degradation of intracellular protein and other cytoplasmic macromolecules in liver is an ongoing process that regulates cytoplasmic mass and provides amino acids for energy and other metabolic uses early in starvation. Cellular proteins are conveniently divided into two general classes according to readily discernable differences in average rates of turnover. A short-lived class, having a half-life of approximately 10 min, comprises about 0.6% of total protein. Its degradation is not physiologically controlled, and the mechanism is probably nonlysosomal in nature. The second or long-lived group, with an average half-life 250 times greater, constitutes more than 99% of the cell's protein. By contrast, its breakdown is strongly regulated, and the site of catabolism is believed to be the vacuolar-lysosomal system. Cytoplasmic sequestration by lysosomes can be divided into two categories; macro- and microautophagy. The first is induced by amino acid and/or insulin deprivation. Amino acids are considered to be primary regulators, since they can control this process over the full range of induced proteolysis in the absence of hormones. Glucagon, cyclic AMP, and beta-agonists also stimulate macroautophagy in hepatocytes but have opposite effects in myocytes. Micrautophagy differs from the former in that the cytoplasmic "bite" is smaller and the uptake process is not acutely regulated. However, the latter does decrease during starvation in parallel with basal proteolysis, effects that might be linked to the loss of endoplasmic reticulum. The primary control of macroautophagy is accomplished through a small group of direct regulators (Leu, Tyr/Phe, Gln, Pro, Met, His, and Trp) and a specific coregulatory action of alanine. As a group, regulatory amino acids produce direct inhibitory responses in the perfused rat liver that are identical to those of the complete amino acid mixture at 0.5x and 4x (times) normal plasma concentrations. However, they lose effectiveness almost completely within a narrow zone centered at normal levels, a loss that can be abolished by the addition of alanine at its normal plasma concentration (0.5 mM). At this level, alanine does not inhibit directly. Interestingly, this zonal loss is also eliminated by insulin. Glucagon, though, specifically blocks the initial inhibition evoked by 0.5x amino acid mixtures and thus induces maximal rates of protein degradation at normal amino acid concentrations.(ABSTRACT TRUNCATED AT 400 WORDS)

Mechanism and control of protein and RNA degradation in the rat hepatocyte: two modes of autophagic sequestration.

The control of protein and RNA degradation by amino acids, insulin, and glucagon was investigated in perfused livers from normal fed rats. Rates of breakdown were determined from the release of valine and cytidine after isotopic labelling in vivo. Stringent amino acid deprivation induced comparable increases (approximately 3.2% h-1) in the degradation of both macromolecular classes, and insulin inhibited them equally. By contrast, glucagon evoked the same proteolytic response at normal plasma concentrations but failed to stimulate RNA breakdown significantly. These and associated electron microscopic findings indicate the existence of two concentration-dependent modes of macroautophagy, one which sequesters both RNA and protein at low amino acid levels and a second which selectively takes up protein at normal concentrations. Control of macroautophagy is accomplished by seven regulatory amino acids and the permissive action of alanine. Alanine is required for effective inhibition by the regulatory group at normal concentrations; in its absence protein degradation accelerates sharply. This response, like that following the administration of glucagon, is mediated by the second mode.

Modulation of the amino acid control of hepatic protein degradation by caloric deprivation. Two modes of alanine co-regulation.

Intracellular protein degradation in perfused livers of fed rats has been shown to be directly regulated by 7 amino acids (Leu, Tyr, Gln, Pro, Met, His, and Trp) and co-regulated by alanine. Responses to graded increases of regulatory amino acids (individually or combined) are multiphasic and include (a) an initial inhibition at 0.5 times normal plasma concentrations, (b) a localized, zonal loss of inhibition at normal levels, and (c) suppression to basal rates at 4 times normal concentrations or greater; the zonal loss of inhibition is prevented by 0.5 mM (normal) alanine. In further perfusion studies carried out at the usual time (1100 h), we have occasionally observed a sharp decrease in proteolytic responsiveness at normal amino acid concentrations. The decrease, which occurred spontaneously in normal fed rats, was attributed to a nearly 90% loss in the sensitivity of alanine co-regulation. In all instances, alanine sensitivity was restored after 4 to 24 h of starvation. The cause of the insensitivity and the mechanism of its reversal by caloric deprivation are not presently known. Starvation for 24 h also appeared to alter the individual inhibitory effectiveness of Leu, Tyr, and Gln. On the other hand, inhibition by the full regulatory group at 4 times normal plasma levels was unchanged when compared with the complete plasma mixture except for a concentration shift in the peak zonal loss of proteolytic inhibition from 1.25 to 0.6 times plasma levels. Since the shift paralleled known changes in portal vein regulatory amino acids, it may have been adaptive in nature. As with fed animals, the zonal loss in starvation was abolished by 0.5 mM alanine, but not with high levels of lactate and pyruvate (10 mM), a finding consistent with the view that co-regulation is mediated by the recognition of alanine per se rather than its metabolism.

Uptake and intracellular fate of [14C]sucrose-insulin in perfused rat livers.

Insulin was covalently linked to [14C]sucrose by means of cyanuric chloride to provide a label that would remain entrapped within the vacuolar system. The uptake of the conjugate by the perfused rat liver was rapid (half-life = 2.9 min), competitively inhibited by native insulin, and abolished by alkali denaturation. As assessed by its distribution on self-generating gradients of colloidal silica-povidone, label in lysosome-enriched samples of liver taken at different times after the addition of the conjugate moved progressively during 15 min from the plasma membrane into an intermediate peak and then to dense lysosomal fractions. After 30-60 min, the label had equilibrated throughout the lysosomal-vacuolar system. The initial movement from the plasma membrane to the intermediate peak occurred between 2 and 5 min. Because label in the peak could be physically separated from the lysosomal marker, beta-acetylglucosaminidase, by dispersing the sample through the gradient mixture before centrifugation rather than layering it, we concluded that the intermediate particles in question were not lysosomal in nature. On gel-filtration chromatography, label extracted from the intermediate peak did not move with insulin but rather as a broad band of lower molecular weight products, suggesting that insulin is subject to early proteolytic attack within a nonlysosomal compartment.

Rates of rat liver RNA degradation in vivo as determined from cytidine release during brief cyclic perfusion in situ.

The breakdown of RNA and of long-lived proteins in rat liver is believed to occur largely within the lysosomal-vacuolar system. Both processes are induced by amino acid lack and suppressed by insulin, and in all circumstances a consistent lag of 15-20 min was observed between the introduction of a physiological regulator and onset of the degradative response. This lag has allowed us to determine rates of liver RNA degradation in vivo during brief cyclic perfusions, as was done previously for long-lived-protein breakdown [Hutson & Mortimore (1982) J. Biol. Chem. 257, 9548-9554]. Degradation was measured from the release of [14C]cytidine in livers of rats previously labelled in vivo with [6-14C]orotic acid. Release was linear and unaffected by physiological regulators between 2 and 12 min of perfusion. In contrast with protein breakdown, no short-lived component was observed. In animals trained to feed between 16:00 and 20:00 h, the content of liver RNA fell at an average rate of 0.26 mg/h per 100 g initial body wt. between 07:00 and 16:00 h, a loss that was within 9% of that predicted from the net release (total release minus reutilization) of cytidine in vivo. In addition, the total rate of RNA degradation determined at the end of the meal was only 12% of that at the start of the post-absorptive period 14 h later (2.1 versus 17.1%/day). This finding is fully consistent with a lysosomal mechanism for RNA degradation, since autophagy is strongly suppressed by food intake. This approach provides a comparatively simple means of approximating moment-to-moment rates of RNA degradation in the rat liver in vivo.

Regulation of microautophagy and basal protein turnover in rat liver. Effects of short-term starvation.

Basal rates of long-lived (resident) protein degradation in rat liver, measured during perfusion after amino acid suppression of macroautophagy, were shown to be strongly regulated by caloric deprivation, decreasing 70% over 48 h in animals fed a high protein diet and 50% in normal controls. Intralysosomal pools of degradable protein correlated directly with basal turnover over this range, yielding a slope (0.09 min-1) that was virtually identical with previous estimates of macroautophagic turnover. The specific radioactivity of valine released from lysosomes in previously labeled livers was the same as that in plasma in both basal and deprivation-induced states. Quantitative electron microscopy revealed a significant decrease with starvation in the absolute volume of a class of secondary lysosome (type A) previously associated with basal or microautophagy. By contrast, the volumes of other microautophagic forms, which comprised roughly 10% of the total, did not change. Taking 0.087 min-1 as the turnover constant of degradable intralysosomal protein and assuming that the concentration of sequestered protein was the same in all vacuoles as that in cytoplasm, we obtained close agreement between predicted and observed rates of basal protein turnover over the range of regulation. The results support the view that the lysosomal system is the final step in the basal degradation of long-lived proteins in the hepatocyte and that a specific class of secondary lysosome (type A) plays a direct role in its regulation during caloric starvation.