Faster protein degradation in response to decreases steady state levels of amino acylation of tRNAHis in Chinese hamster ovary cells.

The rate of protein degradation in cultured Chinese hamster ovary cells increases in response to histidine starvation. Using cell lines with defective histidyl-tRNA synthetase, or histidinol (a competitive inhibitor of the enzyme), we have previously demonstrated a functional connection between the increase in degradation and the amino acylation of this tRNA (Scornik, O. A., Ledbetter, M. L. S., and Malter, J. S. (1980) J. Biol. Chem. 255, 6322-6329). A correlation is shown here between the steady state level of histidyl-tRNA and the regulatory response. Cells were incubated for 15 min in the presence of L-[3H]histidine, at a concentration at which greater than 90% of histidine for protein synthesis derives from the medium. The level of histidyl-tRNA was measured by its radioactivity after purification by phenol extraction, ethanol precipitation, and mild alkaline hydrolysis. Protein degradation in each condition was determined by the release of acid-soluble radioactivity from cells labeled for 24 h with L-[1-14C]leucine. The steady state level of histidyl-tRNA was altered by either histidinol (which slows down its production) or cycloheximide (which interferes with its utilization). Cycloheximide counteracts the effects of histidinol both on the level of histidyl-tRNA and on the rate of protein degradation. Both effects can be obtained, however, even in the presence of cycloheximide, if higher concentrations of histidinol are used. The results indicate that this regulatory mechanism does not recognize the rate of amino acylation per se but rather, the steady state level of its product, amino acyl-tRNA.

The rate of protein degradation in cultured Chinese hamster ovary cells increases in response to histidine starvation. Using cell lines with defective histidyl-tRNA synthetase, or histidinol (a competitive inhibitor of the enzyme), we have previously demonstrated a functional connection between the increase in degradation and the amino acylation of this tRNA (Scornik,  A correlation is shown here between the steady state level of histidyl-tRNA and the regulatory response. Cells were incubated for 15 min in the presence of ~- [~H]histidine, at a concentration at which >90% of histidine for protein synthesis derives from the medium. The level of histidyl-tRNA was measured by its radioactivity after purification by phenol extraction, ethanol precipitation, and mild alkaline hydrolysis. Protein degradation in each condition was determined by the release of acid-soluble radioactivity from cells labeled for 24 h with ~-[l-'~C]leucine. The steady state level of histidyl-tRNA was altered by either histidinol (which slows down its production) or cycloheximide (which interferes with its utilization). Cycloheximide counteracts the effects of histidinol both on the level of histidyl-tRNA and on the rate of protein degradation. Both effects can be obtained, however, even in the presence of cycloheximide, if higher concentrations of histidinol are used. The results indicate that this regulatory mechanism does not recognize the rate of amino acylation per se but rather, the steady state level of its product, amino acyl-tRNA.
Most prokaryote and eukaryote cells respond to deprivation of one or more amino acids with a variety of physiological changes, including slower synthesis of RNA and protein and faster protein breakdown (see Ref. 1 and 2). In bacteria, this response is caused by the accumulation of deacylated tRNA (1) and correlates with the appearance of the unusual guanosine nucleotides ppGpp' (1) and ppGp (3). These nucleotides are not found in animal cells and the regulatory role of tRNA in these cells has been questioned (see Ref. 2). Recent work from this laboratory has demonstrated a functional connection between the utilization of histidine for protein synthesis and the increase in protein degradation in response to histidine starvation in CHO cells. It was found that cell lines in which * This work was supported by Grants AM 13336 and AG 01420 from the National Institutes of Health. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "aduertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. 1 The abbreviations used are: ppGpp, guanosine 5'-diphosphate, 3" diphosphate; ppGp, guanosine 5'-diphosphate, 3'-phosphate; ApdA, diadenosine 5',5'"-P1,P4-tetraphosphate; CHO, Chinese hamster ovary. mutations of histidyl-tRNA synthetase increase the concentration of histidine required for protein synthesis behave in the presence of normal histidine concentration as if they were deprived of the amino acid. It was also found that the increase in protein breakdown produced in normal CHO cells by histidine starvation could be mimicked by histidinol (a competitive inhibitor of the synthetase) but not by cycloheximide (an inhibitor of ribosomal protein synthesis). It was concluded that the regulatory response is functionally connected with the amino acylation of tRNA, or a consequence of it other than the inhibition of protein synthesis (2). One possibility is that the regulatory mechanism senses the activity of the synthetase itself (4). This enzyme is capable of generating at least one regulatory signal, the dinucleotide Ap4A (5). Alternatively, the cell may recognize the steady state level of amino acylation of each tRNA. In this paper, I present evidence for the latter possibility.
In the previous work (2), we sought to decrease the rate of amino acylation of tRNAHis (by low concentration of the amino acid, histidinol, or mutation of the synthetase). In the present paper, I focus on the steady state level of histidyl-tRNA, which can be altered not only by inhibiting the rate of its production with histidinol (a competitive inhibitor of histidyl-tRNA synthetase (6)) but also by decreasing its rate of utilization with cycloheximide: It is well known that inhibitors of protein synthesis, such as cycloheximide, interfere with the stimulation of protein degradation by amino acid starvation. One explanation for this phenomenon is that the slower protein synthesis permits the accumulation of amino acyl-tRNAs (or prevents the accumulation of the deacylated species) ( 2 ) . But other explanations are possible: the generation of a regulatory signal may require an active (although idle) polyribosome (7); a short lived protein may be necessary for the response (8); or different pathways may be responsible for the stimulated and the unstimulated degradation (9, 10) (the former being dependent on protein synthesis). In this paper I measure the steady state levels of histidyl-tRNA in CHO cells in the presence of different concentrations of histidinol and cycloheximide. Once a critical range of concentrations of these inhibitors has been established, the effects of one on the level of histidyl-tRNA can be altered significantly by the presence of the other. I then show that within this range, in the presence of cycloheximide protein degradation is no longer stimulated by the lower concentrations of histidinol, but the stimulation can still be obtained if higher concentrations of histidinol are used. These results are consistent with the hypothesis that this regulatory mechanism recognizes not the rate of amino acylation per se, but the steady state level of its product, histidyl-tRNA. EXPERIMENTAL PROCEDURES'

RESULTS AND DISCUSSION
Stimulation by Histidinol of Protein Degradation-In our previous paper, we mentioned that CHO cells were occasionally unresponsive both to histidinol and to histidine starvation (2). This erratic behavior has disappeared since we introduced the stricter temperature control during the incubation described under "Experimental Procedures." I can now obtain a reproducible 20 to 30% stimulation of protein degradation by concentrations of histidinol of 1 m~ or higher, in the presence of a usual concentration of histidine in Dulbecco's minimal essential medium, 0.2 m~ (Fig. 1). That this effect is due to competition of the inhibitor with the amino acid is demonstrated by the fact that if the concentration of histidine in the medium is increased 10-fold, the concentration of histidinol required to affect degradation increases accordingly (Fig. 1). The stimulation of protein breakdown by deprivation of all amino acids (2) or by addition of histidinol (not shown) can be obtained either in the presence or absence of serum. In all experiments shown in this paper, serum was present during the growth of the cells, but absent during the observation period. I preferred to study the effects of the inhibitors in the absence of serum because the incubation conditions were thus more precisely defined, and also because the cells became quiescent and I avoided complications arising from effects of the inhibitors on growth (2).
Combined Effects of the Inhibitors on Histidyl-tRNA-It was next necessary to establish at which concentrations histidinol and cycloheximide could best be shown to counteract each other's effects on the pool of histidyl-tRNA. The effects of different concentrations on protein synthesis under these conditions are shown in Fig. 3 (left). Increasing inhibition of ribosomal protein synthesis by 0.5 and 1.5 pg/ml (which decreases the rate at which amino acyl-tRNAs are utilized) had the desired effect: progressively higher concentrations of histidinol were necessary to bring down the steady state level of histidyl-tRNA, but the depletion by histidinol was still obtainable within a reasonable concentration range (Fig. 3,  right).
Combined Effects of the Inhibitors on Protein Degradation-Having thus established a range of concentrations within which cycloheximide would antagonize but not abolish the depletion of histidyl-tRNA, I investigated the combined effects of those inhibitors on protein degradation. If cycloheximide abolishes the stimulation by histidinol because it prevents the depletion of histidyl-tRNA (rather than the alternative explanations offered in the Introduction), then, within these critical concentrations, the effect should be re-established, even in the presence of cycloheximide, by using higher concentrations of histidinol. This was indeed the result. Fig.  4 shows two experiments in which the effects on protein degradation by 0, 1, 3, 6, 10, and 30 m~ histidinol were measured in the absence or presence of 0.15, 0.5, and 1.5 pg/ ml of cycloheximide. In both experiments, histidinol produced a full stimulation of protein breakdown at the lower concentration (   of any concentration of cycloheximide. The stimulation was apparent, however, even in the presence of the increasing concentrations of cycloheximide at progressively higher histidinol concentrations. These results are consistent with the hypothesis that the regulatory mechanism recognizes the steady state level of amino acylation of this (and other) tRNAs.

mM h i s t i d i n e (0). o r medium t o brhich h i s t i d i n e was added t o a f i n
It is interesting that cycloheximide slightly depressed protein degradation even in the absence of histidinol and the presence of large concentrations of all 20 amino acids. It is possible that this result indicates that the steady state levels of one or more amino acyl-tRNAs are less than 100% even when the supply of amino acids i-unlimited. In fact, this is the case for histidyl-tRNA, as evidenced by the increase in the plateau radioactivity in the presence of cycloheximide (Miniprint, Fig. 2). This effect of cycloheximide suggests that the mechanism regulating the rate of protein breakdown may be operating even in the presence of a full complement of amino acids.
Effects of Cycloheximide on the Intracellular Histidine Pool-In the preceding paragraphs, we have proposed that cycloheximide counteracts the effects of histidinol because it slows down the utilization of histidyl-tRNA. An alternative, although unlikely, possibility was that unutilized histidine did not leak out of cells fast enough and accumulated the resulting expansion of the histidine pools could have decreased the effectiveness of the competitive inhibitor. Only a very large expansion of the histidine pool could have caused this result. Effects on protein degradation, such as those exhibited by 0.5 pg/ml of cycloheximide in the experiments shown in Fig. 4, would have required at least a 10-fold expansion of the histidine pool (Fig. 1). No such expansion was observed. As shown in Table I, 0.5 pg/ml of cycloheximide expanded the pool by only 10 to 15%. Even a much larger concentration than those used in our experiments, 20 pg/ml, produced only a moderate expansion, 58 to 73%.
General Considerations-We have previously concluded that the regulation of protein degradation (and probably other cellular functions) is functionally connected with the amino acylation of tRNA or some consequence of it other than protein synthesis (2). I have now taken this conclusion one step further and provided evidence that the regulatory mechanism recognizes the steady state level of amino acyl-tRNAs, rather than the rate of amino acylation p e r se. The fact that cycloheximide is known to decrease or abolish the effects of amino acid starvation on protein degradation in a variety of animal cells (10, 16) makes it likely that this conclusion has general validity. Other effects of amino acids on protein synthesis (17) or degradation (18) have been obtained even in the presence of cycloheximide; some of them may be related to amino acid catabolism (18,19). Whether these effects have physiological significance or not remains to be established. There are also reports of regulatory effects of amino acids which are not accompanied by parallel changes in the levels of the corresponding amino acyl-tRNAs (see Ref. 20). Interpretation of these results is uncertain because if the regulation is successful, the changes may be subtle, and because, as shown in the Miniprint, the pools of amino acyl-tRNAs are very small and turn over quickly. The measured value may depend on how rapidly this process is stopped. In attached or suspended cells, the preferred procedure (used here) is to add acid to them (21,22). With organs, even close contact with metal previously cooled in liquid N B (20) may not be fast enough. In spite of these reservations, it is possible that amino acids may affect protein metabolism in more ways than one. The immediate effect of the regulatory mechanism discussed in this paper is to minimize the effects of starvation on protein synthesis. Other regulatory effects of amino acids, if they actually occur in intact animals, may respond to the need to store amino acids as additional tissue protein (23) or avoid unnecessary amino acid catabolism when excessive amounts of amino acids are supplied by the diet. It is not necessary to propose different mechanisms for these different situations, but they are conceivable.
The results provide no indication of whether the regulatory mechanism discussed here senses changes in the acylated or the deacylated tRNAs. On theoretical grounds, the second possibility seems more likely for the following reasons. (a) Amino acyl-tRNAs are normally largely charged (although as we have seen, not necessarily in f u l l ) ; subtle variations in this level will produce proportionally larger changes in the pool of deacylated tRNA. For instance, a decrease in the level of an amino acyl-tRNA from 90 to 80% may be more difficult to detect than the corresponding doubling of the deacylated species from 10 to 20%. ( b ) When only one amino acid is missing, a mechanism that recognizes the appearance of a signal (a deacylated tRNA) where none existed before, would be more sensitive than a mechanism that senses the disappearance of 1 out of 20 different amino acyl-tRNAs.
We also ignore the influence that the concentration and nature of each tRNA (including the isoacceptor species) may have in their recognition. These factors may explain why different cells respond to starvation of some amino acids but not others. The liver, for instance, responds in particular to methionine, phenylalanine, and tryptophan (16), whereas muscle is more sensitive to the presence of branched chain amino acids (18,(24)(25)(26).
Resolution of these issues may have to wait until these regulatory processes can be reproduced in cell-free systems or in permeabilized cells, or until a collection of cell lines carrying mutations in different tRNAs becomes available.