Intracellular protein degradation in Neurospora crassa.

In exponentially growing cultures of Neurospora crassa, the basal rate of protein degradation increases as the constant of the rate of growth decreases, so that in slow growing cells (mu = 0.13) the rate of protein degradation is about 25% of the rate of protein accumulation. During glucose starvation and shift-down transition of growth, the rate of protein degradation is greatly enhanced, and a moderate reduction (about 30%) of the ATP level is observed. Treatment of glucose-starved cells with 2-deoxyglucose reduces the ATP content by 70% and blocks protein degradation. The addition of cycloheximide, given at the onset of glucose starvation, prevents the enhancement of protein degradation; instead cycloheximide is without effect if added when proteolysis has already started. At a supraoptimal temperature (42 degrees C) the basal rate of protein degradation is not stimulated, contrary to the behavior observed in bacteria. Guanosine nucleotides, which appear to have a regulatory role for protein degradation in bacteria, are not found in N. crassa.

Protein degradation is one of the processes that contributes to the regulation of cell growth (1,2). In bacteria it is regulated coordinately with RNA accumulation and other growth-related processes, such as synthesis of ribosomal protein, lipids, and nucleotides (3)(4)(5). Enhanced proteolysis and reduced RNA synthesis are observed in poor growth conditions and correlate with a large increase in the level of guanosine tetraphosphate (1,5,6). Changes in the overall rate of protein degradation contribute also in mammalian cells to the growth kinetics under a variety of normal and pathologic conditions (1, 7-ll), whereas the mechanisms that control protein degradation in higher cells remain largely to be elucidated. Since most of the studies on this subject have been done either in bacterial or in mammalian cells, it seemed opportune t', extend them to a lower eukaryote, such as Neurospora crassa, for which several investigations have already been carried out on the regulation of growth and macromolecular synthe,+c,s (12)(13)(14)(15).
The studies presented in this paper aimed first to ascert:Gn the relevance of protein degradation for the dynamics of growth in Neurospora cells in different conditions of expoil!hntial growth. Then the following aspects of the regulation of the enhanced protein degradation upon  the culture was filtered, and nucleotides were extracted and chromatographed as described for the other "P-labeled cultures, but the autoradiography was exposed for at least 14 days.

Protein Degradation in Different Conditions
of Exponential Growth-Protein degradation in N. crassa in four different conditions of exponential growth is shown in Fig. 1. The rates of protein degradation are very low in glucose minimal medium (IJ = 0.51; )J = duplications/h) and in acetate medium (p = 0.41). Instead, proteolysis is substantial in media that support lower rates of exponential growth, such as glycerol (p = 0.26) and ethanol (p = 0.13). The kinetics of decay of labeled protein observed in glycerol indicated the presence of at least two components, a fast and a slow decaying one, whereas such biphasic kinetics data are not detectable with the other carbon sources.
The half-lives of protein calculated from the slopes of the best fitting straight lines on the semilogarithmic plot of Fig to recall at this point that glycerol kinase, a key enzyme for glycerol metabolism, has been shown in Neurospora to have a very high turnover rate (24). As for the cells growing on the other media it remains possible that the experimental approach used is not sensitive enough to detect the eventual presence of a very small percentage of rapidly turning over protein.

Protein
Turnover during Nutritional Shifts-During a shift-down transition from glucose to glycerol ( Fig. 2A), about 25% of the radioactive protein made during the exponential growth in glucose is degraded. The onset of proteolysis occurs immediately when glucose is exhausted and lasts for approximately 2.5 h (Fig. 2B), exactly in parallel with the block of rRNA accumulation ( Fig. 2A)  A, culture growing on limiting glucose (60 pg/ml) and an excess of glycerol.
At zero time glucose is exhausted and the diauxic lag begins (12). Cl, Growth as absorbance at 450 nm; 0, protein accumulation; a, stable RNA accumulation.

Intracellular
Protein Degradation in Neurospora crassa findings suggest that in some way the degradation process is a selective one and, furthermore, that it may facilitate the rearrangement of cellular structures that are quite different in the two nutritional conditions (25). On the contrary, during a shift-up transition from acetate to glucose, rRNA synthesis is enhanced (14), and no protein degradation is detectable (results not shown).
Protein Degradation during Glucose Starvation-When IV. crassa mycelia are grown in low glucose (60 pg/ml of initial concentration), growth stops as soon as glucose is exhausted in the medium, with an immediate block of protein and RNA accumulations (results not shown), whereas at the same time protein degradation is largely enhanced (Fig. 3). In this condition we have determined protein degradation by measuring either the radioactivity retained in the hot trichloroacetic acid-precipitable fraction or the radioactivity released in the trichloroacetic acid-soluble fraction (Fig. 3, inset). The agreement between the two methods is good for the fist 2 to 3 h and then the radioactivity found in the trichloroacetic acidsoluble fraction is less than that expected by its disappearance from hot trichloroacetic acid-precipitable material, probably due to a metabolic utilization of the released amino acids. From the data of Fig. 3, a degradation rate of about 12% (h-l) can be calculated.
The addition of 3-0-methylglucose (100 pg/ml) or of amethylglucoside (100 pg/ml) to glucose-starved cells does not significantly modify the rate of proteolysis, whereas the addition of 2-deoxyglucose (2-DG)' (100 pg/ml) almost completely blocks it (Fig. 4). In N. crassa cY-methylglucoside and 3-0-methylglucose are substrates for the transport systems (26), but they are not utilized; 2-DG instead is taken up and phosphorylated by hexokinase (27). In our experimental conditions 2-DG neither supports growth or protein accumulation in glucose-starved cells for at least 3 h nor inhibits the growth rate of cells growing exponentially on glucose (200 pg/ml, initial concentration) (results not shown). Since several reports indicate that the ATP level and the adenylate energy charge are important factors in regulating protein degradation (1,5,28), we have measured the adenylate pools during glucose starvation with or without the addition of 2-DG (100 pg/ml). As shown in Fig. 5, the ATP level (expressed as nanomoles/A.,50 unit of culture) slightly decreases during glucose starvation, whereas in the presence of 100 pg/ml of 2-DG it is much more reduced. However, also in this case there is not a complete depletion of the ATP pool, whose level stabilizes around 30% of the initial value. The energy charge decreases from a value of 0.80 to 0.84 to a value of 0.70 to 0.75 in glucose starvation, and the addition of 2-DG lowers the energy charge to values of 0.65, causing at the same time a partial depletion of the total adenylate pool (Table I).
Proteolysis in glucose-starved cells is immediately blocked, as shown in Fig 6A, by the readdition of glucose, which allows growth to start again. 2-DG also blocks protein degradation (Fig. 6B), of course without resumption of growth. Although the two responses are very similar, glucose and 2-DG appear to act with different mechanisms, as indicated by experiments done with cycloheximide (see below). Effects of Cycloheximide-The addition of cycloheximide (CHI) (1 pg/ml) to N. crassa cultures severely inhibits the enhancement of protein degradation during glucose starvation. The effect of cycloheximide, added at the time of the glucose exhaustion, is shown in Fig. 7   Inset, the chase was done at the moment of glucose exhaustion (zero time). The trichloroacetic acid soluble radioactivity was determined at different times after the chase, and to each value the zero time value was subtracted. The data are expressed as percentage of radioactivity present in the trichloroacetic acid-precipitable material at zero time. CHI is added when proteolysis has already started, there is no effect (Table II and Fig. 8).
CHI (1 pg/ml) completely stops within 5 to 10 min the incorporation of r,-['?]leucine into proteins (Fig. 8, inset) both in exponentially growing and in starved cultures; this lack of incorporation is not due to an inhibition of leucine uptake and, thus, it reflects a real inhibition of protein synthesis (data not shown). Besides, we have observed that the same lack of inhibition of protein degradation is observed at higher concentrations of CHI (10 pg/ml) (results not shown). It has been previously shown (Fig. 6, A and B) that the readdition of glucose or 2-DG to glucose-starved cells immediately blocks protein degradation.
In the presence of CHI the responses to glucose and to 2-DG are different, whereas glucose (Fig. 9) is also able in this condition to inhibit proteolysis; the readdition of 2-DG is apparently without effect.   (Fig. 9A) is not due to residual growth since protein synthesis is completely blocked, but probably to a swelling and vacuolization of the hyphae, that occurs under these conditions.2 Effects ofAmino Acids-Since amino acids are the terminal products of proteolysis, it might be interesting to ascertain whether or not the addition of amino acids to glucose-starved cells blocks protein degradation, in a kind of negative feedback control. The addition of 1 mM casamino acids to glucosestarved cells gradually reduces protein degradation, and at the same time growth resumes (Fig. lo), whereas lower concentrations (100 PM) are without effect. Thus, the effect observed in the fist instance is probably related to the recovery of growth and not to a direct regulatory effect of the exogenous amino acids on protein degradation.
A process in which proteolysis has a significant physiological role is spore formation, both in fungi (29) and bacteria (30). The following experiments were performed to see whether or not there was an impairment of protein degradation in a number of aconidial mutants of Neurospora, which are able to produce conidia only in the TIME   presence of exogenous amino acids (31). Since growth starvation is required to start conidiation (31), and in the aconidial mutants exogenous amino acids have to be provided for the production of conidia, it might be thought that protein degradation may not be stimulated during glucose starvation in some aconidial mutants. Contrary to this prevision, all of the aconidial mutants tested (45 fl A, 1838 fl A, 291 ~01-3 A, 102 fr A, 88 sh A, 68 sp A, and 277 bis A) have been shown to be able to degrade intracellular protein during glucose starvation (results not shown).
Effect of Temperature-Temperatures higher than 38°C and lower than 20°C severely restrict growth of N. crassa mycelia growing exponentially in liquid medium in glucose (15). The experiments presented so far have been conducted at 30°C. We have previously reported that during exponential growth in glucose at 8°C protein degradation is undetectable (15). Experiments conducted during this investigation indicate that there is not an enhancement of the basal protein degradation even at supraoptimal temperatures, contrary to the behavior reported for E. coli (32). In fact, during exponential growth in glucose at 42°C (p = 0.37), the basal rate of protein degradation is almost undetectable and even lower than that observed at 30°C. These results suggest that either Neurospora protein is fairly resistant to the processes of thermal denaturation or that the same proteases are thermolabile at 42°C. The latter suggestion does not hold, at least for the proteases that are active in glucose starvation. In fact, under this condition at 42°C the rate of protein degradation is largely enhanced over the rate measured at 30°C; it is, in fact, about 20% (h-l).
A Search for Regulatory Signals--In E. coli a strict correlation between protein degradation and the net synthesis of ribosomal RNA (i-RNA) is apparent since all of the conditions that block rRNA accumulation enhance protein degradation (1,5,6). Thus, it has been suggested that the same regulatory signals, for instance guanosine 5'-diphosphate, 3'-diphosphate (ppGpp) as a negative signal or a guanosine nucleotide, called the "phantom spot" (22) as a positive signal, control both processes (1,5,6).
In N. crassa, ppGpp is not detectable under conditions that inhibit rRNA accumulation (21); thus, it seems interesting to extend the analysis to the phantom spot to find out whether or not it is present. For this purpose N. crassa cultures in exponential growth in glucose were labeled with [32P]orthophosphate, and the acid-soluble nucleotides were separated by two-dimensional thin layer chromatography on polyethyleneimine cellulose sheets (22). After autoradiography a small spot (indicated as X) was observed near GTP, in a position close to that occupied by the phantom spot (Fig. 1IA). This X spot is present also during shift-down; its level is quite constant with respect to that of the GTP (about 5% of the GTP counts are found in the X spot), and it is not absorbed by activated charcoal (Norit).
Furthermore, the X spot is not detectable after labeling the cultures with [U-'4C]guanosine (Fig. llB), thus indicating that it is not a guanine derivative. The X spot observed after [32P]orthophosphate labeling might, therefore, be a small polyphosphate (33)   with commercial GTP as ultraviolet marker.
The autoradiography was exposed for 72 h. B, a 20.ml culture was labeled for 30 min with [U-%]guanosine (final concentration IO-" M, specific activity 100 Ci/mol).
Twenty microliters of the 1 N formic acid extract were chromatographed with commercial GTP. The autoradiography was exposed for 14 days. i TIME the equilibration of the specific radioactivity of the intracellular precursor pool and the external medium during the chase (34,35).
In Neurospora amino acids are compartmentalized in two pools (36,37). One, relatively small, is used for protein synthesis and other metabolic activities and quickly equilibrates with the external amino acids. A second one, localized in vesicles, contains for many amino acids the bulk of their cytoplasmic content, is expandable, and exchanges slowly with the other cytoplasmic pool (38,39).
The intracellular leucine content during exponential growth in glucose is quite small and it is largely increased by the addition of even fairly low concentrations (10m5) of exogenous leucine (15), that appear to expand the vesicular pool. Under the experimental conditions used in this study, the chase with 10m4 M leucine blocks completely and immediately the incorporation of [ %]leucine into protein (Fig. l&Q, as expected if the specific activity of the nonorganellar cytoplasmic pool equilibrates very quickly with the external leucine. At the same time after the chase, the intracellular trichloroacetic acid-soluble radioactivity does not vary appreciably (Fig. 12 under the chase conditions, the radioactive amino acids released by protein degradation are found outside the cell, as measured by the difference between the radioactivity of the total trichloroacetic acid-soluble material of the culture (cells plus medium) and that of the cells (Fig. 12B). These findings suggest that protein degradation, as other metabolic processes, utilizes the nonorganellar cytoplasmic pool, which is in active exchange with the external leucine, as indicated also by the results of Fig. 12A. The extent of isotope reutilization should, therefore, be minimized. DISCUSSION The regulation of intracellular protein degradation in Neurospora, as in bacteria and in mammalian cells, appears to be part of a general control system that may be relevant to determine the dynamics of growth (2, 9) and allows the cells to adapt to a poor nutritional environment.
The studies reported in this paper indicate the presence of similarities and differences in the control of protein degradation in Neurospora as compared to other well known systems.
In different conditions of exponential growth, the average rate of protein degradation in Neurospora increases by decreasing the rate of growth (Fig. l), thus confirming the observation that slow growing cells have higher rates of pro-Intracellular Protein Degradation in Neurospora crassa 7053 tein degradation (1,4). The rate of protein degradation becomes an increasing fraction of the rate of protein synthesis at lower growth rates. It can be calculated that the rate of protein degradation in cells growing in glucose, acetate, glycerol, and ethanol media corresponds to 3%, 4%, 14%, and 25% of the respective rate of protein synthesis. Thus, in slow growing cells the considerable rate of protein degradation puts a heavy burden on growth metabolism since up to one-fourth of the protein synthetic activity is counteracted by protein degradation.
This may be the reason for the presence of the "extra" rRNA (13,40) observed in cells growing in ethanol.
The basal overall rate of protein degradation in cells growing exponentially on glucose is largely enhanced by carbon source starvation (Fig. 3). A slight decrease (about 30%) of the ATP level is observed during starvation (Fig. 5). A similar correlation is noted also during a shift-down transition of growth. The ATP level drops by 30% at the beginning of the diauxic lag (la), whereas protein degradation is largely stimulated (Fig. 2). A second distinct effect of reduced ATP availability on protein degradation is evidenced by the treatment with 2-deoxyglucose, which lowers the ATP level by 70% (Fig. 5) and greatly reduces the rate of protein degradation in glucose-starved cells (Fig. 4).
In Neurospora we would have, therefore, a situation similar to that described for E. coli, in which a moderate decrease of cellular ATP enhances the rate of protein degradation, whereas if the ATP level is almost depleted, protein degradation is blocked (1,5). The involvement of ATP in protein breakdown, which recalls to mind similar energy requirements of other catabolic processes such as glycolysis or fatty acids oxidation, has also been evidenced in in vitro systems of protein degradation (28). The energy requirement might perhaps explain the puzzling observation that the treatment with 2-deoxyglucose in the presence of cycloheximide does not block protein degradation (Fig. 9D), contrary to the effect of the same treatment without cycloheximide (Fig. 6B). It might be tentatively suggested that the treatment with cycloheximide spares ATP utilization, thus, keeping the ATP level, even in the presence of 2-deoxyglucose, at a value that allows protein degradation to occur. As regards the regulatory function that a moderate decrease in the availability of cellular high energy phosphates may have to enhance the rate of protein degradation and the fall in RNA synthesis, it has been proposed that it may be mediated in E. coli by a large accumulation of guanosine ppGpp through a reduction of its rate of degradation (5).
We have shown that also in Neurospora an increased protein degradation (see "Results") is coordinated with a strong reduction of rRNA synthesis (12); however, the response is not mediated by ppGpp, which is not found in Neurospora (21). Nor did we detect in Neurospora another guanosine nucleotide, the so-called "phantom spot," which might be implicated in the regulation of stringent response in E. coli (22).
Moreover, when considering the control role that energy metabolism may have on protein degradation and rRNA synthesis in Neurospora, a preliminary question may be considered: whether it is the change of the ATP level or that of the rate of ATP production that triggers the regulatory response. In fact, these two aspects of energy metabolism are not always correlated in Neurospora since there appears to be feedback processes that depress ATP utilization very quickly after ATP synthesis is inhibited (41). The existence of a careful balance between the rate of synthesis and that of utilization of ATP in Neurospora is supported by the observation that the intracellular ATP pool is very small as compared to the rate of ATP utilization for protein and RNA synthesis.' In conclusion, the available findings suggest that in Neurospora changes in nutritional conditions that bring about a decrease of energy production also provoke inhibition of rRNA synthesis and enhancement of protein degrrdation, presumably through the mediation of a metabolic signal(s) different from the one(s) that mediates similar responses in bacteria. The inverse correlation between rRNA synthesis and protein degradation is not absolute since there is at least one condition, the treatment with caffeine, that inhibits rRNA synthesis without stimulating protein degradation in Neurospora (44). It is conceivable that the different response is related to a different mechanism of inhibition of rRNA synthesis.
The enhancement of the rate of protein degradation upon carbon starvation is prevented by cycloheximide ( Fig. 7 and Table II). Inhibitors of protein synthesis have been reported to reduce the rate of protein degradation in starving bacteria (1,6,45) and in resting mammalian cells (46,47). Since in these cells protein degradation falls shortly after the addition of the protein synthesis inhibitor, it has been suggested by Hershko and co-workers (46,48) that protein synthesis is necessary for the continuous synthesis of the short-lived protein required for the stimulation of protein breakdown during starvation. The experiments of Fig. 8 rule out the presence of such rapidly turning over protein in Neurospora since protein degradation continues at the same rate for at least 3 h after CHI is added after the onset of starvation.
Similar results have been reported also in yeast (49).
In E. coli the inhibitors of protein synthesis cause a reduction of the ppGpp level and in this way appear to lower the rate of protein degradation (5,6). The effects of CHI on protein degradation in Neurospora can not be explained by a similar mechanism, with only the change of the metabolic signal involved. In fact, CHI prevents the onset of protein degradation with a clear time dependence (Table II), but does not affect protein breakdown once it is established (Fig. 8) or prevent the block of protein degradation by the readdition of glucose (Fig. 9). The simplest explanation seems, therefore, that the onset of protein degradation in starving Neurospora requires the synthesis of new, relatively stable protein, either proteolytic enzymes or regulatory polypeptides, in a different pattern of response than that of E. coli, in which the increased proteolysis seems to involve pre-existing proteases (6), but in a similar way as proposed for the regulation of protein degradation in sporulating yeast (49). On the other hand, the readdition of glucose (Figs. 6 and 9) and in a less pronounced way that of amino acids (Fig. 10) is able to quickly inactivate 'The ATP level in Neurospora cells growing exponentially in glucose at 30°C is 2.4 nmol/Arso unit of culture (IS), i.e. 1.45 X 10" molecules. The protein level per A4r,a unit of culture is about 140 pg, and a rate of protein synthesis (since protein degradation is negligible) of 4.1 x 10'" amino acids, (minutesK'/A450 unit) can be calculated (15), which corresponds to 16.4 x 10'" high energy phosphate bonds required for protein synthesis (minutesK'/Aral, unit). The rKNA content