Ng313
Molec. gen. Genet. 170, 225-230 (1979)
© by Springer-Verlag 1979
Peptide Chain Elongation Rate and Ribosomal Activity in Saccharomyces cerevisiae as a Function of the Growth Rate Bjarne Bonven and Kay Gullov Department of Molecular Biology, University of Odense, Campusvej 55, DK-5230 Odense M, Denmark
Summary. The peptide-chain elongation rate of Saccharomyces cerevisiae at two different growth rates was estimated by the kinetics of radioactive labelling of nascent and finished polypeptides as described by Gausing, 1972, and Young and Bremer, 1976. The elongation rates of a diploid strain cultured in yeast nitrogen base supplemented with glucose or acetate were 9.3 amino acids/s and 5.5 amino acids/s at 30° C, respectively. These data together with published values on the "ribosomal efficiency" as a function of growth rate (Waldron and Lacroute, 1975) enable us to estimate the rate of synthesis of ribosomal proteins as a function of the rate of total protein synthesis, c~r, and the fraction of ribosomes that are active in protein synthesis. We conclude that in S. cerevisiae er is largely independent of the growth rate while the fraction of active ribosomes decreases with decreasing growth rate.
Introduction
During steady state growth of a unicellular organism, the amount of any cellular component will be doubled within one generation time. From measurements of the amount of protein and the number of ribosomes synthesized per cell generation, the rate of protein synthesis per ribosome also denoted the "average ribosome efficiency'; is calculated. Assuming that all ribosomes are active and the turnover of cellular protein is negligible the "average ribosome efficiency" equals the peptide-chain growth rate. In bacteria, approximately eighty percent of the ribosomes are actively engaged in protein synthesis. This figure is largely independent of the growth rate except for extreme low growth rates where the n u m For offprints contact: Kay Gullov
ber of active ribosomes seems to decrease (Forchhammer and Lindahl, 1971; Lacroute and Stent, 1968; Young and Bremer, 1976; Kjeldgaard and Gausing, 1974). In the budding yeast Saccharomyces cerevisiae the cellular protein and rRNA content was measured over a ten fold variation of the steady state growth rates by Waldron and Lacroute (1975). Contrary to the observation in bacteria, the "average ribosome efficiency" was found to decrease with decreasing growth rate in a linear fashion. A similar result was reported by Boehlke and Friesen (1975). Furthermore, these authors state that in yeast, this observation reflects a decreasing peptide-chain growth rate, since approximately ninety percent of the ribosomes are registered within the polysome region in metabolically active spheroplasts independent of the growth media. In principle a 3 to 5 fold variation of the peptide-chain growth rate could be accounted for by 1) limitation of endogenous amino acid supply during slow growth or limitation during a growth rate dependent interval in the cell cycle; 2) rate limiting, synthesis of tRNA at low growth rates; 3) disaggregation of polysomes during a growth rate dependent interval in the cell cycle as registered in mammalian cells during mitoses (Steward et al., 1968; Fan and Penman, 1970); 4) growth rate dependent variation in mRNA supply during cell cycle which might reflect the discontinuous enzyme synthesis during the cell cycle (Halvorson et al., 1971). Of these possibilities at least 2) seems to be ruled out since the tRNA/rRNA ratio tends to increase slightly with decreasing growth rates (Waldron and Lacroute, 1975). Measurements of the fraction of active ribosomes directly from the number of ribosomes in polysomes is a dubious method since the cells must either be converted to metabolically active spheroplasts before lysis or treated with cycloheximide before spheroplasting. Incubation with cycloheximide 0026-8925/79/0170/0225/$01.20
226 is k n o w n to cause a c c u m u l a t i o n of m o n o s o m e s o n m R N A ( F a n a n d P e n m a n , 1970). A metabolically active spheroplast is devoid of the cell wall which constitutes m o r e t h a n 20 ~ of the total cellular protein. It is likely that these p e r t u r b e d cells are in a state of derepression, synthesizing cell wall proteins at high rates a n d thus increasing their polysome fraction as c o m p a r e d with n o r m a l cells. Experiments reported here seem to confirm this n o t i o n . Therefore, we t u r n e d to estimate the peptide-chain e l o n g a t i o n rate of i n d i v i d u a l proteins by the kinetics of radioactive labelling of n a s c e n t a n d finished polypeptides originally described by G a u s i n g (1972).
Materials and Methods a) Strain and Growth Conditions. A diploid strain X951 (Str~mnaes, 1968) was used in all experiments. This strain has no growth requirement for leucine. Cells were cultivated in Wickerham's minimal medium supplemented with either 2 % w/v glucose or sodium acetate as carbon source. Metabolically active spheroplasts were prepared and cultivated as described earlier (Gullov and Friis, 1978). b) Kinetic of Labelling Nascent and Fin•hed Polypeptide Chains. The dual labelling method of Young and Bremer (1976) was used. 150 ml of cells, cultivated in balanced growth for several generations, were pulse labelled at a cell density of 8 x 10 6 cells/ml with 3Hleucine (3gC/ml; 53 Ci/mmole). Samples (3.5ml) were taken at approx. 5 s intervals into test tubes containing 8 ml ethanol plus 100gg/ml leucine, 200gg/ml cycloheximide, 15raM NAN3, and 2 % mercaptoethanol. The test tubes were stored in a cold bath at -20 ° C. A 50 ml reference culture was labelled for two generations with 14C-leucine(0.1 p.C/ml; 2.5 mCi/mmole) followed by a 30 rain chase with 100 gg/ml leucine. Each sample of the chilled cells was supplemented with an aliquot of reference cells. c) Protein Extraction. The fixed cells were harvested at 4°C and washed 3 times with 5 ml iced water supplemented with 100 pg/ml leucine, 200gg/ml cycloheximide and 15 mM NaN 3. Finally the cells were resuspended in 50gl extraction buffer (62.5mM TrisHC1, 0.1 M EDTA, 5 ~ SDS, 10 % glycerol, 5 % mercaptoethanol, pH6.8). Samples were heated to 100°C for 4min , chilled on ice and centrifuged at 18,000g to sediment non-solubilized material. More than 90 % of labelled material was extracted by this method. This supernatant was subdivided. One fraction was subjected to electrophoresis, the other was used to determine total incorporation of labelled leucine. d) Electrophoresis. The extracted yeast proteins were separated by SDS-gel electrophoresis. A Pharmacia GE-4 equipment for slabgel electrophoresis was used. The separation gel was composed of 10 % acrylamide, 0.3 % bis-acrylamide, and 0.1% SDS in 0.375M Tris-HC1 at pH 8.8. A stacking gel composed of 3 % acrylamide, and 0.1% SDS in 0.125M Tris-HC1 at pH6.8 was used. As electrode buffers was used 0.025M Tris, 0.194M glycine at pH 8.3. Voltage: 100V. Current: 15 to 20mA/slab. Gels were stained overnight with 0.25% coomassie blue R-250 dissolved in methanol/acetic acid/water: 4.5:1:4.5 and destained for two days with methanol/acetic acid/water: 1:1.5:5. Nine marker proteins with molecular weights ranging from 11,700 to 165,000 were coelectrophoresed with the samples. Approximately fifty of the most
B. Bonven and K. Gullov: Regulation of Protein Synthesis in Yeast prominent proteins from the supernatant were clearly visible following destaining. Of these, only twelve bands were selected, on the basis of molecular weights, staining intensity, and how well they were resolved from the neighbouring bands.
e) Radioactivity Assay. Individual bands were cut out manually from the gel slabs and digested with 250gl of 30 % H20 2 supplemented with 5% v/v NH~OH in sealed scintillation vials overnight at 37°C. Total incorporation was assayed by applying 20gl of the protein extract on a gel strip cut from pre-run unloaded gels followed by HzO2 digestion. Digested samples were dissolved in 15 ml toluene (PPO-POPOP) scintillation liquid supplemented with 10 % v/v NCS (Searle).
Results a ) Estimation o f the Peptide-Chain Growth Rate. The p a r a m e t e r F*(m,t) described by G a u s i n g (1972) was d e t e r m i n e d according to Y o u n g a n d Bremer (1976). The peptide-chain e l o n g a t i o n rate, Cp ( a m i n o acids/s per active r i b o s o m e is equal to: m
Cp=-tm •a where m is the m o l e c u l a r weight of the polypeptide, t,, the time of synthesis of the polypeptide of M w m, a n d a the average molecular weight of a m i n o acid residues in protein. I n yeast a is a s s u m e d to be 107 ( W a l d r o n a n d Lacroute, 1975). As a prerequisite for sharp, well defined breakpoints at t,, a n d linearity of the initial section of the F* kinetics, total uptake of 3H-leucine should increase strictly linearly with time within the pulse interval. This r e q u i r e m e n t is fulfilled in yeast at leucine c o n c e n t r a t i o n s below 0.1 g M as seen in Fig. 1. I n
91 A ZX
U I
._o rr
3
o
~
50
0
0
100
150
Time (sec) Fig. 1. Total incorporation of ~H-leucine.Cells were grown exponentially at 30°C in a minimal medium supplemented with glucose or acetate. 3H-leucine was added to the cultures. Samples were taken into chilled supplemented ethanol, whereby incorporation was stopped instantaneously. 14C-leucine labelled reference cells were added and the samples were processed as described in Materials and Methods. (zx) Glucose culture. (o) Acetate culture
B. Bonven and K. Gullov: Regulation of Protein Synthesis in Yeast
227
F*
1.0
_
/
• • • O0. !
•
0.5
I 1.0
-~Z
I
I 9
I
•--•~u--O
r
49.000 120
I
i
• •
0.5
60
[
180
•-/r I
39.000 I 120
6O
240
P
] 180
T
l 240
Time (sec) F* 26,500
18.000
m ~ 2.0 oc.I (9
Fig. 2. a F* kinetics of selected protein bands from glucose grown cells. Samples taken as described in legend to Fig. 1 were prepared for electrophoresis as described in Materials and Methods. For a given band the ratio of 3H/14C-leucine in each sample was divided by the 3H/~4C-Ieucine ratio of total protein (Fig. 1). Individual molecular weights are indicated on the figures b Comparative F* kinetics of selected protein bands at two different growth rates. F* kinetics of two protein bands of molecular weight of 26,500 and 18,000 were determined in cells grown in glucose minimal medium and in acetate minimal medium, respectively
1.0
p
I
I
3.0
2.0
2 =_ _ O ~-o
1.0
t.... ~ oo ~
I
I
1
I
I
so
120
1so
60
12o
Time
• -
ee,~
[
180
(sec)
acetate grown cells a slight increase in the uptake rate is registered at 70 s; therefore, only polypeptides with t m values below this figure were included in Cp calculations. Figure 2a and 2b show a number of individual F* kinetics from both glucose and acetate grown cells. In most cases the t,, values were defined from the distinct breaks of the whole kinetics. In cases where the breakpoints were less distinct, with polypeptides of high molecular weight, the slope of the initial kinetic, equal to 0.5 x t~ 1 of the theoretical F*(m,t) function, was considered in order to determine tin. This case is examplified by the polypeptide of molecular weight 85,000. In Fig. 3 is recorded the relation between t,,-values and the molecular weights of the polypeptide bands as calculated from the individual F* kinetics. For glucose grown cells (/~ =0.50 doublings/h) Cp is calculated to be 9.3 amino acids/s, while Cp for acetate grown cells (#=0.17 doublings/h) is calculated to be 5.5 amino acids/s.
80c o
.;
";
60-
O 40
~Z 20-
i
20
40
60
80
t m (see) Fig. 3. Time of synthesis of S. cerevisiae proteins of different molecular weights and different growth rates. The t,, values from the individual F*(m, t) kinetics of Fig. 2a and b are plotted against the corresponding molecular weights. (e) Glucose minimal culture. (o) Acetate minimal culture
228
B. Bonven and K. Gullov: Regulation of Protein Synthesis in Yeast
Table 1. Estimation of ~r and fl, at two growth rates
o
Growth rate Cp e~ generations/h aminoacids/s aminoacids/s
ctr ~
~, b
0.5 0.17
0.34 0.32
0.59 0.36
9.3 5.5
5.5 2.0
a Data from Waldron and Lacroute (1975). Calculated from the relation er = #c/etr at balanced growth (Schleif, 1967), where e is the number of amino acids in a monosome, c is assumed to be equal to 2.0 x 104 according to Wool and StSffler (1974) b Calculatedfrom the relation er=fl,. Cp
b) Ribosome Activity at Different Growth Rates. The average rate of protein synthesis per unit volume of cells (dP/dt) is, dP/dt = er. R, where e~ is the "average ribosome efficiency" and R the number of ribosomes. dP(ribosomal) /dP Introducing the parameter e r = dt / dt during balanced growth, e~=l~.c/er, (Schleif, 1967), where/~ is the specific growth rate and c the number of amino acids in a ribosome, c~ (the rate of synthesis of ribosomal proteins as a fraction of the rate of total protein synthesis) might thus be calculated from the knowledge of e~ at different growth rates. According to Bakalkin et al. (1976) the rate of intracellular protein turnover during early exponential, aerobic growth of S. cerevisiae is only 3.3 ~ of total cell protein per hour. Therefore, the least reliable figure is presumably c, since no published data for yeast are available. The value from rat liver cells used in this estimation is supposed to be representative, since both the number of ribosomal proteins in yeast and liver cells and the sedimentation coefficients are largely the same. The results are given in Table 1 together with the estimation of fir (the fraction of ribosomes that are active in protein synthesis).
c) Protein Synthesis During Nutritional Shift-Up. To see if the more than two fold increase of the rate of protein synthesis per ribosome which has been observed in glucose grown cells as compared with acetate grown cells might be due to a severe limitation of the amino acid supply at lower growth rates, the following experiments were carried out. Acetate grown cells were supplemented with glucose and/or amino acids, and the uptake rate of labelled leucine was registered during the shifts. Since this strain has no growth requirement for leucine, labelled leucine was added in high concentration to ensure that the increment of leucine incorporation is proportional to the increment in protein. Labelled leucine was added to a final concentration of 10 gg/ml which results in at least a ten fold swelling of the leucine pool (Watson, 1976) but at the expense of accuracy at the early
20-
X
.c_ l O .......
U
5-
15
30
45
Time (rain) Fig. 4. Effect of nutritional shift-up on the synthesis of protein. Cells were cultivated in acetate minimal medium for several generations to a cell density of 3 x 106 cells/ml. At time zero the culture was labelled with 3H-leucine (2gC/ml; 10gg/ml). l ml samples were taken into l ml 10 ~ TCA supplemented with 200gg/ml leucine at the indicated time. 15rain later the culture was sub~ divided into 4 cultures receiving acetate (e); 2 ~ glucose (o); all common L-amino acids (each of 20 gg/ml) except for leucine (A) and glucose+L-amino acids (zx). Sampling was continued for 30 rain. Samples were filtered and washed three times on Whatman GF/C glass-paper filters with 5 ~ TCA
time points. Consequently a 15min labelling period before the shifts was necessary to measure the preshift uptake rate. As seen in Fig. 4, the pre-shift rate is maintained initially after the shift, when the medium is enriched with all the common amino acids. The gradually reduced rate observed following glucose and amino acid addition is likely to be due to a multirepression effect imposed on an appreciable number of structural genes. A similar response has been described earlier (Wehr and Parks, 1969). The absence of an abrupt increase in incorporation following amino acid addition seems to rule out the possibility that severe amino acid limitation causes the decrease of the rate of protein synthesis per ribosome at lower growth rates.
d) Protein Synthesis in Metabolically Active Spheroplasts. The obvious discrepancy between our estimations of fir and measurements of the size of the polysome fraction in metabolically active spheroplasts (Boehlke and Friesen, 1975) was revealed by the following experiment. A glucose culture was supplemented with 14C-uracil for several generations in order to label stable RNA. One half of the culture was converted to spheroplasts, the other half was treated in an analogous manner, but snail gut enzyme was omitted. Both cultures were resuspended in 3H-leucine labelled prewarmed medium supplemented with
B. Bonven and K. Gullov: Regulation of Protein Synthesis in Yeast
5 o U
~-
o
m 3 O
c~ nr"
1
20
60 Time
100 (min)
Fig. 5. Rate of protein synthesis in metabolically active spheroplasts and normal cells in sorbitol supported medium. Cells were labelled with 1*C-uracil (0.5ktC/ml; 5 #g/ml) for two generations in glucose minimal medium. At a density of 2 x 106 cells/ml incorporation was stopped by addition of a forty fold excess of unlabelled uracil. Cells were harvested and washed with 0.9 % KCI. One half was converted to spheroplast. Cells and spheroplasts were resuspended a a celt density of 8 x 106 cells/ml in prewarmed glucose medium supplemented with 1 M sorbitol and labelled with 3H-leucine (0.3 ~g/ml; 10 gg/mI). 1 ml samples were taken into 1 ml l0 % TCA supplemented with 100gg/ml leucine at the indicated times. TCA insoluble material was registered. (o) Normal cells. (o) Metabolically active spheroplasts
1 M sorbitol, and the incorpartion of label into protein was followed for two hours. This procedure was used to exclude from the calculations variable cell recoveries due to cell lysis during spheroplasting. As seen in the graph in Fig. 5 the rate of increment of newly synthesized protein per equivalent of pre-synthesized RNA is marked higher in the spheroplasts as compared with normal cells. Since both cultures were subjected to the same starvation conditions, etc. we suppose that this difference reflects a higher net rate of protein synthesis in spheroplasts, rather than a preferential degredation of pre-synthesized RNA in spheroplasts. Discussion
Our results unambiguously reveal that the peptidechain elongation rate in yeast is independent of the molecular weights of the finished polypeptides. The time needed to synthesize a finished chain is thus proportional to the number of amino acid residues of the finished chain (Fig. 3.). This observation disagrees with the report of Waldron et al. (1974) stating a chain length dependent variation of the peptide-chain elongation rates. Although nearly identical techniques have been employed by these authors and us, the discrepancy is likely due to different experimental conditions. Their labelling kinetics were carried out at rather low specific activities causing a non linear
229
uptake rate into labelled proteins and a low degree of labelling of individual protein bands at early times, notably in the case of F*-kinetics of polypeptides with low molecular weights. Furthermore, the labelling kinetics were performed at rather high cell densities (4 x 107cells/ml). Under these conditions a distinct decrease of the rate of protein synthesis is observed with our strain. The yeast peptide-chain elongation rate of 9.3 amino acids/s per ribosome in glucose minimal medium at 30 ° C is only slightly below the value of 12 amino acids/s in E. coli grown at similar conditions with respect to carbon source and temperature (Gausing, 1972). Although differing in structure the yeast ribosome seems to be almost as efficient as the bacterial ribosome with respect to the elongation rates when the growth conditions are nearly identical. The chain growth rate of acetate grown cells of 5.5 amino acids/s per ribosome represents a 40 % reduction of the chain growth rate corresponding to a 2.5 fold decrease in the growth rate. In E. coli an equivalent reduction of the growth rate, as compared with glucose growth, results in a 30 to 40 % reduction of the chain growth rate (Forchhammer and Lindahl, 1971; Young and Bremer, 1976). Thus the peptide-chain growth rates seem to be affected by the nutritional conditions in nearly the same manner in E. coli and S. cerevisiae. In E. coli the number of ribosomes per cell is
controlled in such a manner that the fraction of active ribosomes is constant within a broad range of growth rates. In S. cerevisiae the number of ribosomes per cell is largely proportional to the protein content per cell, and the rate of protein synthesis per ribosome decreases with decreasing growth rate (Waldron and Lacroute, 1975; Boehlke and Friesen, 1975). ~r seems to be constant and independent of the growth rate. When our Cp estimations are taken into account (Table 1) this pattern can be interpreted to mean that the number of active ribosomes decreases drastically with decreasing growth rate, resulting in an overproduction of ribosomes at low growth rates. The vast amount of idled ribosomes observed migh be seen throughout the cell cycle or could be restricted to discrete steps within the cell cycle. The discontinuous enzyme synthesis observed by Halvorson et al. (1971) favours the last assumption. Furthermore, they observed that most of the enzymes included in the analysis doubled their enzyme activities rather late in the cell division cycle. We have unpublished data which indicate that the net rate of protein synthesis varies approximately two fold during the division cycle of glucose grown cells. The lowest rate is observed during the interval between cell separation and emergence of the bud.
230 Acknowledgments. The authors thank S. Andersen and I. Holmblad for their skillful technical assistance and Dr. J. Friis for helpful discussions and advice.
References Balkalkin, G.Y., Kalhov, S.L., Zubatov, A.S., Luzikov, V.N.: Degradation of total cell protein at different stages of Saccaromyces cerevisiae yeast growth. FEBS Letters 63, 218-221 (1976) Boehlke, K.W., Friesen, J.D.: Cellular content of ribonucleic acid and protein in Saecharomyces cerevisiae as a function of exponential growth rate: Calculation of the apparent peptide chain elongation rate. J. Bacteriol. 121, 429-433 (1975) Fan, H., Penman, S.: Regulation of protein synthesis in mammalian cells. J. Molec. Biol. 50, 655-670(1970) Forchhammer, J., Lindahl, L.: Growth rate of polypeptide chains as a function of the cell growth rate in a mutant of Escherichia coB. J. Molec. Biol. 55, 563-568 (1971) Gausing, K.: Efficiency of protein and messenger RNA synthesis in bacteriophage T4-infected cells of Escherichia coli. J. Molec. Biol. 71, 529-545 (1972) Gullov, K.B., Friis, J.: Isolation of chromatin from Saccharomyces cerevisiae: The metabolic state in relation to the protein content of chromatin and its susceptibility to micrococcal nuclease. Exp. Mycology 2, in press (1978) Halvorson, H.O., Carter, B.L.A., Tauro, P.: Synthesis of enzymes during the cell cycle. Advan. Microbial Physiol. 6, 47-106 (1971) Kjeldgaard, N.O., Gausing, K.: Regulation of biosynthesis of ribosomes. In: Ribosomes (M. Nomura, A. Tissi~res, and P. Lengyel, eds.), pp. 369-392. New York: Cold Spring Harbor Laboratory 1974
B. Bonven and K. Gullov: Regulation of Protein Synthesis in Yeast Lacroute, F., Stent, G.S.: Peptide chain growth of/3-galactosidase in Escherichia coll. J. Molec. Biol. 35, 165-173 (1968) Schleif, R. : Control of production of ribosomal protein. J. Molec. Biol. 27, 41-55 (1967) Steward, D.L., Schaeffer, J.R., and Humphrey, R.M.: Breakdown and assembly of polyribosomes in synchronized chinese hamster cells. Science 161, 791-793 (1968) Stromnaes, ~.: Genetic changes in Saccharomyces cerevisiae grown on media containing DL-para-fluoro-phenylalanine. Hereditas. 59, 197-220 (1968) Watdron, C., Jund, R., and Lacroute, F.: The elongation rate of proteins of different molecular weight classes in yeast. FEBS letters 46, 11-16 (1974) Waldron, C., Lacroute, F.: Effect of growth rate on the amounts of ribosomal and transfer ribonucleic acids in yeast. J. Bacteriol. 122, 855-865 (1975) Watson, T.G.: Amino-acid pool composition of Saccharomyces cerevisiae as a function of growth rate and amino-acid nitrogen source2 J. Gen. Microbiol. 96, 263-268 (1976) Wehr, C.T., Parks, L.W.: Macromolecular synthesis in Saccharomyces cerevisiae in different growth media. J. Bacteriol. 98, 458-466 (1969) Wool, I.G., Sti3ffler, G.: Structure and function of eukaryotic ribosomes. In: Ribosomes (M. Nomura, A. Tissi6res, and P. Lengyel, eds.), pp. 417-460. New York: Cold Spring Harbor Laboratory 1974 Young, R., Bremer, H.: Polypeptide-chain elongation rate in Escherichia coli B/r as a function of growth rate. Biochem. J. 160, 185-194 (1976)
Communicated
by H.G. Wittmann
Received August 21 / October 23, 1978
Molec. gen. Genet. 170, 225-230 (1979)
© by Springer-Verlag 1979
Peptide Chain Elongation Rate and Ribosomal Activity in Saccharomyces cerevisiae as a Function of the Growth Rate Bjarne Bonven and Kay Gullov Department of Molecular Biology, University of Odense, Campusvej 55, DK-5230 Odense M, Denmark
Summary. The peptide-chain elongation rate of Saccharomyces cerevisiae at two different growth rates was estimated by the kinetics of radioactive labelling of nascent and finished polypeptides as described by Gausing, 1972, and Young and Bremer, 1976. The elongation rates of a diploid strain cultured in yeast nitrogen base supplemented with glucose or acetate were 9.3 amino acids/s and 5.5 amino acids/s at 30° C, respectively. These data together with published values on the "ribosomal efficiency" as a function of growth rate (Waldron and Lacroute, 1975) enable us to estimate the rate of synthesis of ribosomal proteins as a function of the rate of total protein synthesis, c~r, and the fraction of ribosomes that are active in protein synthesis. We conclude that in S. cerevisiae er is largely independent of the growth rate while the fraction of active ribosomes decreases with decreasing growth rate.
Introduction
During steady state growth of a unicellular organism, the amount of any cellular component will be doubled within one generation time. From measurements of the amount of protein and the number of ribosomes synthesized per cell generation, the rate of protein synthesis per ribosome also denoted the "average ribosome efficiency'; is calculated. Assuming that all ribosomes are active and the turnover of cellular protein is negligible the "average ribosome efficiency" equals the peptide-chain growth rate. In bacteria, approximately eighty percent of the ribosomes are actively engaged in protein synthesis. This figure is largely independent of the growth rate except for extreme low growth rates where the n u m For offprints contact: Kay Gullov
ber of active ribosomes seems to decrease (Forchhammer and Lindahl, 1971; Lacroute and Stent, 1968; Young and Bremer, 1976; Kjeldgaard and Gausing, 1974). In the budding yeast Saccharomyces cerevisiae the cellular protein and rRNA content was measured over a ten fold variation of the steady state growth rates by Waldron and Lacroute (1975). Contrary to the observation in bacteria, the "average ribosome efficiency" was found to decrease with decreasing growth rate in a linear fashion. A similar result was reported by Boehlke and Friesen (1975). Furthermore, these authors state that in yeast, this observation reflects a decreasing peptide-chain growth rate, since approximately ninety percent of the ribosomes are registered within the polysome region in metabolically active spheroplasts independent of the growth media. In principle a 3 to 5 fold variation of the peptide-chain growth rate could be accounted for by 1) limitation of endogenous amino acid supply during slow growth or limitation during a growth rate dependent interval in the cell cycle; 2) rate limiting, synthesis of tRNA at low growth rates; 3) disaggregation of polysomes during a growth rate dependent interval in the cell cycle as registered in mammalian cells during mitoses (Steward et al., 1968; Fan and Penman, 1970); 4) growth rate dependent variation in mRNA supply during cell cycle which might reflect the discontinuous enzyme synthesis during the cell cycle (Halvorson et al., 1971). Of these possibilities at least 2) seems to be ruled out since the tRNA/rRNA ratio tends to increase slightly with decreasing growth rates (Waldron and Lacroute, 1975). Measurements of the fraction of active ribosomes directly from the number of ribosomes in polysomes is a dubious method since the cells must either be converted to metabolically active spheroplasts before lysis or treated with cycloheximide before spheroplasting. Incubation with cycloheximide 0026-8925/79/0170/0225/$01.20
226 is k n o w n to cause a c c u m u l a t i o n of m o n o s o m e s o n m R N A ( F a n a n d P e n m a n , 1970). A metabolically active spheroplast is devoid of the cell wall which constitutes m o r e t h a n 20 ~ of the total cellular protein. It is likely that these p e r t u r b e d cells are in a state of derepression, synthesizing cell wall proteins at high rates a n d thus increasing their polysome fraction as c o m p a r e d with n o r m a l cells. Experiments reported here seem to confirm this n o t i o n . Therefore, we t u r n e d to estimate the peptide-chain e l o n g a t i o n rate of i n d i v i d u a l proteins by the kinetics of radioactive labelling of n a s c e n t a n d finished polypeptides originally described by G a u s i n g (1972).
Materials and Methods a) Strain and Growth Conditions. A diploid strain X951 (Str~mnaes, 1968) was used in all experiments. This strain has no growth requirement for leucine. Cells were cultivated in Wickerham's minimal medium supplemented with either 2 % w/v glucose or sodium acetate as carbon source. Metabolically active spheroplasts were prepared and cultivated as described earlier (Gullov and Friis, 1978). b) Kinetic of Labelling Nascent and Fin•hed Polypeptide Chains. The dual labelling method of Young and Bremer (1976) was used. 150 ml of cells, cultivated in balanced growth for several generations, were pulse labelled at a cell density of 8 x 10 6 cells/ml with 3Hleucine (3gC/ml; 53 Ci/mmole). Samples (3.5ml) were taken at approx. 5 s intervals into test tubes containing 8 ml ethanol plus 100gg/ml leucine, 200gg/ml cycloheximide, 15raM NAN3, and 2 % mercaptoethanol. The test tubes were stored in a cold bath at -20 ° C. A 50 ml reference culture was labelled for two generations with 14C-leucine(0.1 p.C/ml; 2.5 mCi/mmole) followed by a 30 rain chase with 100 gg/ml leucine. Each sample of the chilled cells was supplemented with an aliquot of reference cells. c) Protein Extraction. The fixed cells were harvested at 4°C and washed 3 times with 5 ml iced water supplemented with 100 pg/ml leucine, 200gg/ml cycloheximide and 15 mM NaN 3. Finally the cells were resuspended in 50gl extraction buffer (62.5mM TrisHC1, 0.1 M EDTA, 5 ~ SDS, 10 % glycerol, 5 % mercaptoethanol, pH6.8). Samples were heated to 100°C for 4min , chilled on ice and centrifuged at 18,000g to sediment non-solubilized material. More than 90 % of labelled material was extracted by this method. This supernatant was subdivided. One fraction was subjected to electrophoresis, the other was used to determine total incorporation of labelled leucine. d) Electrophoresis. The extracted yeast proteins were separated by SDS-gel electrophoresis. A Pharmacia GE-4 equipment for slabgel electrophoresis was used. The separation gel was composed of 10 % acrylamide, 0.3 % bis-acrylamide, and 0.1% SDS in 0.375M Tris-HC1 at pH 8.8. A stacking gel composed of 3 % acrylamide, and 0.1% SDS in 0.125M Tris-HC1 at pH6.8 was used. As electrode buffers was used 0.025M Tris, 0.194M glycine at pH 8.3. Voltage: 100V. Current: 15 to 20mA/slab. Gels were stained overnight with 0.25% coomassie blue R-250 dissolved in methanol/acetic acid/water: 4.5:1:4.5 and destained for two days with methanol/acetic acid/water: 1:1.5:5. Nine marker proteins with molecular weights ranging from 11,700 to 165,000 were coelectrophoresed with the samples. Approximately fifty of the most
B. Bonven and K. Gullov: Regulation of Protein Synthesis in Yeast prominent proteins from the supernatant were clearly visible following destaining. Of these, only twelve bands were selected, on the basis of molecular weights, staining intensity, and how well they were resolved from the neighbouring bands.
e) Radioactivity Assay. Individual bands were cut out manually from the gel slabs and digested with 250gl of 30 % H20 2 supplemented with 5% v/v NH~OH in sealed scintillation vials overnight at 37°C. Total incorporation was assayed by applying 20gl of the protein extract on a gel strip cut from pre-run unloaded gels followed by HzO2 digestion. Digested samples were dissolved in 15 ml toluene (PPO-POPOP) scintillation liquid supplemented with 10 % v/v NCS (Searle).
Results a ) Estimation o f the Peptide-Chain Growth Rate. The p a r a m e t e r F*(m,t) described by G a u s i n g (1972) was d e t e r m i n e d according to Y o u n g a n d Bremer (1976). The peptide-chain e l o n g a t i o n rate, Cp ( a m i n o acids/s per active r i b o s o m e is equal to: m
Cp=-tm •a where m is the m o l e c u l a r weight of the polypeptide, t,, the time of synthesis of the polypeptide of M w m, a n d a the average molecular weight of a m i n o acid residues in protein. I n yeast a is a s s u m e d to be 107 ( W a l d r o n a n d Lacroute, 1975). As a prerequisite for sharp, well defined breakpoints at t,, a n d linearity of the initial section of the F* kinetics, total uptake of 3H-leucine should increase strictly linearly with time within the pulse interval. This r e q u i r e m e n t is fulfilled in yeast at leucine c o n c e n t r a t i o n s below 0.1 g M as seen in Fig. 1. I n
91 A ZX
U I
._o rr
3
o
~
50
0
0
100
150
Time (sec) Fig. 1. Total incorporation of ~H-leucine.Cells were grown exponentially at 30°C in a minimal medium supplemented with glucose or acetate. 3H-leucine was added to the cultures. Samples were taken into chilled supplemented ethanol, whereby incorporation was stopped instantaneously. 14C-leucine labelled reference cells were added and the samples were processed as described in Materials and Methods. (zx) Glucose culture. (o) Acetate culture
B. Bonven and K. Gullov: Regulation of Protein Synthesis in Yeast
227
F*
1.0
_
/
• • • O0. !
•
0.5
I 1.0
-~Z
I
I 9
I
•--•~u--O
r
49.000 120
I
i
• •
0.5
60
[
180
•-/r I
39.000 I 120
6O
240
P
] 180
T
l 240
Time (sec) F* 26,500
18.000
m ~ 2.0 oc.I (9
Fig. 2. a F* kinetics of selected protein bands from glucose grown cells. Samples taken as described in legend to Fig. 1 were prepared for electrophoresis as described in Materials and Methods. For a given band the ratio of 3H/14C-leucine in each sample was divided by the 3H/~4C-Ieucine ratio of total protein (Fig. 1). Individual molecular weights are indicated on the figures b Comparative F* kinetics of selected protein bands at two different growth rates. F* kinetics of two protein bands of molecular weight of 26,500 and 18,000 were determined in cells grown in glucose minimal medium and in acetate minimal medium, respectively
1.0
p
I
I
3.0
2.0
2 =_ _ O ~-o
1.0
t.... ~ oo ~
I
I
1
I
I
so
120
1so
60
12o
Time
• -
ee,~
[
180
(sec)
acetate grown cells a slight increase in the uptake rate is registered at 70 s; therefore, only polypeptides with t m values below this figure were included in Cp calculations. Figure 2a and 2b show a number of individual F* kinetics from both glucose and acetate grown cells. In most cases the t,, values were defined from the distinct breaks of the whole kinetics. In cases where the breakpoints were less distinct, with polypeptides of high molecular weight, the slope of the initial kinetic, equal to 0.5 x t~ 1 of the theoretical F*(m,t) function, was considered in order to determine tin. This case is examplified by the polypeptide of molecular weight 85,000. In Fig. 3 is recorded the relation between t,,-values and the molecular weights of the polypeptide bands as calculated from the individual F* kinetics. For glucose grown cells (/~ =0.50 doublings/h) Cp is calculated to be 9.3 amino acids/s, while Cp for acetate grown cells (#=0.17 doublings/h) is calculated to be 5.5 amino acids/s.
80c o
.;
";
60-
O 40
~Z 20-
i
20
40
60
80
t m (see) Fig. 3. Time of synthesis of S. cerevisiae proteins of different molecular weights and different growth rates. The t,, values from the individual F*(m, t) kinetics of Fig. 2a and b are plotted against the corresponding molecular weights. (e) Glucose minimal culture. (o) Acetate minimal culture
228
B. Bonven and K. Gullov: Regulation of Protein Synthesis in Yeast
Table 1. Estimation of ~r and fl, at two growth rates
o
Growth rate Cp e~ generations/h aminoacids/s aminoacids/s
ctr ~
~, b
0.5 0.17
0.34 0.32
0.59 0.36
9.3 5.5
5.5 2.0
a Data from Waldron and Lacroute (1975). Calculated from the relation er = #c/etr at balanced growth (Schleif, 1967), where e is the number of amino acids in a monosome, c is assumed to be equal to 2.0 x 104 according to Wool and StSffler (1974) b Calculatedfrom the relation er=fl,. Cp
b) Ribosome Activity at Different Growth Rates. The average rate of protein synthesis per unit volume of cells (dP/dt) is, dP/dt = er. R, where e~ is the "average ribosome efficiency" and R the number of ribosomes. dP(ribosomal) /dP Introducing the parameter e r = dt / dt during balanced growth, e~=l~.c/er, (Schleif, 1967), where/~ is the specific growth rate and c the number of amino acids in a ribosome, c~ (the rate of synthesis of ribosomal proteins as a fraction of the rate of total protein synthesis) might thus be calculated from the knowledge of e~ at different growth rates. According to Bakalkin et al. (1976) the rate of intracellular protein turnover during early exponential, aerobic growth of S. cerevisiae is only 3.3 ~ of total cell protein per hour. Therefore, the least reliable figure is presumably c, since no published data for yeast are available. The value from rat liver cells used in this estimation is supposed to be representative, since both the number of ribosomal proteins in yeast and liver cells and the sedimentation coefficients are largely the same. The results are given in Table 1 together with the estimation of fir (the fraction of ribosomes that are active in protein synthesis).
c) Protein Synthesis During Nutritional Shift-Up. To see if the more than two fold increase of the rate of protein synthesis per ribosome which has been observed in glucose grown cells as compared with acetate grown cells might be due to a severe limitation of the amino acid supply at lower growth rates, the following experiments were carried out. Acetate grown cells were supplemented with glucose and/or amino acids, and the uptake rate of labelled leucine was registered during the shifts. Since this strain has no growth requirement for leucine, labelled leucine was added in high concentration to ensure that the increment of leucine incorporation is proportional to the increment in protein. Labelled leucine was added to a final concentration of 10 gg/ml which results in at least a ten fold swelling of the leucine pool (Watson, 1976) but at the expense of accuracy at the early
20-
X
.c_ l O .......
U
5-
15
30
45
Time (rain) Fig. 4. Effect of nutritional shift-up on the synthesis of protein. Cells were cultivated in acetate minimal medium for several generations to a cell density of 3 x 106 cells/ml. At time zero the culture was labelled with 3H-leucine (2gC/ml; 10gg/ml). l ml samples were taken into l ml 10 ~ TCA supplemented with 200gg/ml leucine at the indicated time. 15rain later the culture was sub~ divided into 4 cultures receiving acetate (e); 2 ~ glucose (o); all common L-amino acids (each of 20 gg/ml) except for leucine (A) and glucose+L-amino acids (zx). Sampling was continued for 30 rain. Samples were filtered and washed three times on Whatman GF/C glass-paper filters with 5 ~ TCA
time points. Consequently a 15min labelling period before the shifts was necessary to measure the preshift uptake rate. As seen in Fig. 4, the pre-shift rate is maintained initially after the shift, when the medium is enriched with all the common amino acids. The gradually reduced rate observed following glucose and amino acid addition is likely to be due to a multirepression effect imposed on an appreciable number of structural genes. A similar response has been described earlier (Wehr and Parks, 1969). The absence of an abrupt increase in incorporation following amino acid addition seems to rule out the possibility that severe amino acid limitation causes the decrease of the rate of protein synthesis per ribosome at lower growth rates.
d) Protein Synthesis in Metabolically Active Spheroplasts. The obvious discrepancy between our estimations of fir and measurements of the size of the polysome fraction in metabolically active spheroplasts (Boehlke and Friesen, 1975) was revealed by the following experiment. A glucose culture was supplemented with 14C-uracil for several generations in order to label stable RNA. One half of the culture was converted to spheroplasts, the other half was treated in an analogous manner, but snail gut enzyme was omitted. Both cultures were resuspended in 3H-leucine labelled prewarmed medium supplemented with
B. Bonven and K. Gullov: Regulation of Protein Synthesis in Yeast
5 o U
~-
o
m 3 O
c~ nr"
1
20
60 Time
100 (min)
Fig. 5. Rate of protein synthesis in metabolically active spheroplasts and normal cells in sorbitol supported medium. Cells were labelled with 1*C-uracil (0.5ktC/ml; 5 #g/ml) for two generations in glucose minimal medium. At a density of 2 x 106 cells/ml incorporation was stopped by addition of a forty fold excess of unlabelled uracil. Cells were harvested and washed with 0.9 % KCI. One half was converted to spheroplast. Cells and spheroplasts were resuspended a a celt density of 8 x 106 cells/ml in prewarmed glucose medium supplemented with 1 M sorbitol and labelled with 3H-leucine (0.3 ~g/ml; 10 gg/mI). 1 ml samples were taken into 1 ml l0 % TCA supplemented with 100gg/ml leucine at the indicated times. TCA insoluble material was registered. (o) Normal cells. (o) Metabolically active spheroplasts
1 M sorbitol, and the incorpartion of label into protein was followed for two hours. This procedure was used to exclude from the calculations variable cell recoveries due to cell lysis during spheroplasting. As seen in the graph in Fig. 5 the rate of increment of newly synthesized protein per equivalent of pre-synthesized RNA is marked higher in the spheroplasts as compared with normal cells. Since both cultures were subjected to the same starvation conditions, etc. we suppose that this difference reflects a higher net rate of protein synthesis in spheroplasts, rather than a preferential degredation of pre-synthesized RNA in spheroplasts. Discussion
Our results unambiguously reveal that the peptidechain elongation rate in yeast is independent of the molecular weights of the finished polypeptides. The time needed to synthesize a finished chain is thus proportional to the number of amino acid residues of the finished chain (Fig. 3.). This observation disagrees with the report of Waldron et al. (1974) stating a chain length dependent variation of the peptide-chain elongation rates. Although nearly identical techniques have been employed by these authors and us, the discrepancy is likely due to different experimental conditions. Their labelling kinetics were carried out at rather low specific activities causing a non linear
229
uptake rate into labelled proteins and a low degree of labelling of individual protein bands at early times, notably in the case of F*-kinetics of polypeptides with low molecular weights. Furthermore, the labelling kinetics were performed at rather high cell densities (4 x 107cells/ml). Under these conditions a distinct decrease of the rate of protein synthesis is observed with our strain. The yeast peptide-chain elongation rate of 9.3 amino acids/s per ribosome in glucose minimal medium at 30 ° C is only slightly below the value of 12 amino acids/s in E. coli grown at similar conditions with respect to carbon source and temperature (Gausing, 1972). Although differing in structure the yeast ribosome seems to be almost as efficient as the bacterial ribosome with respect to the elongation rates when the growth conditions are nearly identical. The chain growth rate of acetate grown cells of 5.5 amino acids/s per ribosome represents a 40 % reduction of the chain growth rate corresponding to a 2.5 fold decrease in the growth rate. In E. coli an equivalent reduction of the growth rate, as compared with glucose growth, results in a 30 to 40 % reduction of the chain growth rate (Forchhammer and Lindahl, 1971; Young and Bremer, 1976). Thus the peptide-chain growth rates seem to be affected by the nutritional conditions in nearly the same manner in E. coli and S. cerevisiae. In E. coli the number of ribosomes per cell is
controlled in such a manner that the fraction of active ribosomes is constant within a broad range of growth rates. In S. cerevisiae the number of ribosomes per cell is largely proportional to the protein content per cell, and the rate of protein synthesis per ribosome decreases with decreasing growth rate (Waldron and Lacroute, 1975; Boehlke and Friesen, 1975). ~r seems to be constant and independent of the growth rate. When our Cp estimations are taken into account (Table 1) this pattern can be interpreted to mean that the number of active ribosomes decreases drastically with decreasing growth rate, resulting in an overproduction of ribosomes at low growth rates. The vast amount of idled ribosomes observed migh be seen throughout the cell cycle or could be restricted to discrete steps within the cell cycle. The discontinuous enzyme synthesis observed by Halvorson et al. (1971) favours the last assumption. Furthermore, they observed that most of the enzymes included in the analysis doubled their enzyme activities rather late in the cell division cycle. We have unpublished data which indicate that the net rate of protein synthesis varies approximately two fold during the division cycle of glucose grown cells. The lowest rate is observed during the interval between cell separation and emergence of the bud.
230 Acknowledgments. The authors thank S. Andersen and I. Holmblad for their skillful technical assistance and Dr. J. Friis for helpful discussions and advice.
References Balkalkin, G.Y., Kalhov, S.L., Zubatov, A.S., Luzikov, V.N.: Degradation of total cell protein at different stages of Saccaromyces cerevisiae yeast growth. FEBS Letters 63, 218-221 (1976) Boehlke, K.W., Friesen, J.D.: Cellular content of ribonucleic acid and protein in Saecharomyces cerevisiae as a function of exponential growth rate: Calculation of the apparent peptide chain elongation rate. J. Bacteriol. 121, 429-433 (1975) Fan, H., Penman, S.: Regulation of protein synthesis in mammalian cells. J. Molec. Biol. 50, 655-670(1970) Forchhammer, J., Lindahl, L.: Growth rate of polypeptide chains as a function of the cell growth rate in a mutant of Escherichia coB. J. Molec. Biol. 55, 563-568 (1971) Gausing, K.: Efficiency of protein and messenger RNA synthesis in bacteriophage T4-infected cells of Escherichia coli. J. Molec. Biol. 71, 529-545 (1972) Gullov, K.B., Friis, J.: Isolation of chromatin from Saccharomyces cerevisiae: The metabolic state in relation to the protein content of chromatin and its susceptibility to micrococcal nuclease. Exp. Mycology 2, in press (1978) Halvorson, H.O., Carter, B.L.A., Tauro, P.: Synthesis of enzymes during the cell cycle. Advan. Microbial Physiol. 6, 47-106 (1971) Kjeldgaard, N.O., Gausing, K.: Regulation of biosynthesis of ribosomes. In: Ribosomes (M. Nomura, A. Tissi~res, and P. Lengyel, eds.), pp. 369-392. New York: Cold Spring Harbor Laboratory 1974
B. Bonven and K. Gullov: Regulation of Protein Synthesis in Yeast Lacroute, F., Stent, G.S.: Peptide chain growth of/3-galactosidase in Escherichia coll. J. Molec. Biol. 35, 165-173 (1968) Schleif, R. : Control of production of ribosomal protein. J. Molec. Biol. 27, 41-55 (1967) Steward, D.L., Schaeffer, J.R., and Humphrey, R.M.: Breakdown and assembly of polyribosomes in synchronized chinese hamster cells. Science 161, 791-793 (1968) Stromnaes, ~.: Genetic changes in Saccharomyces cerevisiae grown on media containing DL-para-fluoro-phenylalanine. Hereditas. 59, 197-220 (1968) Watdron, C., Jund, R., and Lacroute, F.: The elongation rate of proteins of different molecular weight classes in yeast. FEBS letters 46, 11-16 (1974) Waldron, C., Lacroute, F.: Effect of growth rate on the amounts of ribosomal and transfer ribonucleic acids in yeast. J. Bacteriol. 122, 855-865 (1975) Watson, T.G.: Amino-acid pool composition of Saccharomyces cerevisiae as a function of growth rate and amino-acid nitrogen source2 J. Gen. Microbiol. 96, 263-268 (1976) Wehr, C.T., Parks, L.W.: Macromolecular synthesis in Saccharomyces cerevisiae in different growth media. J. Bacteriol. 98, 458-466 (1969) Wool, I.G., Sti3ffler, G.: Structure and function of eukaryotic ribosomes. In: Ribosomes (M. Nomura, A. Tissi6res, and P. Lengyel, eds.), pp. 417-460. New York: Cold Spring Harbor Laboratory 1974 Young, R., Bremer, H.: Polypeptide-chain elongation rate in Escherichia coli B/r as a function of growth rate. Biochem. J. 160, 185-194 (1976)
Communicated
by H.G. Wittmann
Received August 21 / October 23, 1978
