Molec. gen. Genet, 170, I29-135 (1979) © by Springer-Verlag 1979
Individual Messenger RNA Half Lives in Saccharomyces cerevisiae H. Koch Strahlenzentrum der Justus-Liebig-Universitfit,D-6300 GieBen, Federal Republic of Germany J.D. Friesen Department of Biology, York University, 4700 Keele Street, Downsview, Ontario, Canada M3J 1P3
Summary. We have measured the decay half-life of functional messenger R N A ( m R N A ) for some thirty different proteins in the yeast Saccharomyces cerevisiae. Production of newly synthesized m R N A was halted by raising the temperature of a culture of a temperature-sensitive mutant, ts 136. Aliquots of this culture were pulsed-labelled with [3sS]-methionine at various times after the temperature shift and the radioactive proteins separated on the two-dimensional gel electrophoresis system of O'Farrell. We find a range in the decay half lives of individual m R N A species which varies f r o m 3.5 min to greater than 70 min. We find three general classes of decay curves, (a) simple exponential (first order); some of these showed a shoulder before onset of exponential decay; (b) bi-component or multi-component concave upward; (c) initial stimulation of rate of m R N A synthesis, followed by virtually undetectable decay.
Introduction The decay constant for messenger R N A ( m R N A ) in the yeast Saccharomyces cerevisiae has been measured both for global m R N A as well as for certain individual m R N A species (Hartwell, 1967; Tonnesen and Friesen, 1973; Hereford and Rosbash, 1977; Cooper et al., 1978). These results point to the likelihood of considerable variation in individual decay rates of different m R N A species, allowing for fast regulatory response for some translation products and comparably slow response for others. The average half-life for overall m R N A decay appears to be approximately 17 __+3 min. The methods used previously to determine m R N A decay constants have suffered from certain shortcomings. F o r example, the use of inhibitors inFor offprints contact: H. Koch
troduces the uncertainty that the inhibitor itself might interfere with the initiation of translation or the stability of m R N A (Singer and Penman, 1972). Or, measurements of m R N A life time by the kinetics of precursor incorporation assume r a n d o m decay (Petersen et al., 1976), which is not necessarily the case and furthermore requires certain assumptions regarding equilibration of intracellular pools with exogenous radioactive precursors. Again, measurements in spheroplasts may not apply to physiological condition of normal cell growth. We have sought to circumvent some of these problems and to provide an independent means of measuring individual functional m R N A half-lives by combining the use of a temperature sensitive (ts) mutant with 2-dimensional gel electrophoresis. There is evidence that mutant ts 136 of Saccharomyces cerevisiae is defective in some function which leads to the appearance of inhibited transport of all classes of R N A from the nucleus into the cytoplasm at the non-permissive temperature (Hutchinson et al., 1969; Hartwell et al., 1970; Tonnesen and Friesen, 1973; Shiokawa and Pogo, 1974; Hereford and Rosbash, 1977; Cooper et al., 1978). Certainly it appears that all classes of R N A synthesis are inhibited in this mutant at 37 ° C. This strain has been used to measure the availability of functional m R N A for translation (Tonnesen and Friesen, 1973).
Material and Methods Strains and Media. Saccharomycescerevisiae strain ts 136 (Hutchinson et al., 1969), was given to us by L.H. Hartwell. The strain was grown in yeast nitrogen base (YNB) medium supplemented (per litre) with 5 g succinic acid, 3 g NaOH, 20 g glucose, 20 mg each of histidine, lysine and tyrosine, 125 nag each of adenine and uracil and 0.5 g yeast extract (Difco). The strain was grown at 23° C; the doubling time was approximately 190 rain. Growth was monitored by measuring the absorbance at 436 nm with a filter
0026- 892 5/79/0170/0129/$01.40
130
H. Koch and J.D. Friesen: Individual Messenger RNA Half Lives in Saeeharomyces cerevisiae
photometer. An optical density of 1 corresponds to approximately 2 x 107 cells/ml.
Radioactive Labelling. Exponentially growing cultures were shifted at A436= 1 to 37°C and pulse-labelled with 17.5 ~tCi/ml [35S]methionine (755Ci/mmol, Amersham) at various times after the shift. The label was diluted with non-radioactive methionine to give a final concentration of 1.3 x 10 -3 gmol/ml in the culture. The labelling period of 6 min was terminated by adding 250 Ixg/ml methionine; 9 min later the cultures were rapidly chilled to 0° C. Portions of 20 gl were withdrawn and precipitated in 3 ml of icecold 5% trichloroacetic acid. The precipitates were collected on nitrocellulose membrane filters and washed three times with 3 ml each of 5% trichloroacetic acid. 5 ml xylene based scintillation mix was added after drying and the radioactivity was determined in a liquid scintillation spectrometer. Reference cells were taken from the original culture and labelled with 10gCi/ml [methyl-3H]-methionine (8.8 Ci/mmol, 59 mCi/mg, Amersham) for 140 rain at 23° C. The labelling was terminated as described for pulseqabelled cells.
Extraction of Protein. 35S-methionine pulse-labelled cells and 3Hlabelled cells were washed. Equal amounts of 3H-labelled cells were added to each pulselabelled culture. The samples were transferred to Eppendorf vials and 0.3 g glass beads (0.45-0.5 mm diameter) were added. The vials were put into a pre-cooled adapter fitting into a Braun homogenizer and the cells were disrupted for 5 min. The homogenate was collected and the glass beads were washed twice. To the combined homogenate and washings a mixture was added to give a final concentration of 50 mM Tris-HC1, 0.5 mM MgClz at pH 7.4 and 20 gg/ml of bovine pancreas deoxyribonuclease and ribonuclease each (DN 25 and R 4875, Sigma). The samples were kept at 0°C for 30 min, then frozen in Eppendorf vials at - 6 0 ° C and freeze-dried overnight. The dried samples were dissolved in 50 gl lysis buffer (O'Farrell, 1975). Cell debris was sedimented in an Eppendorf centrifuge and from the supernatant protein extract 2 gl portions were precipitated in 3 ml 5% TCA for radioactivity determination and were treated as described above.
gel. Care was taken to avoid trapping air between the gels. The gel set was clamped to an electrophoresis apparatus, the chambers were filled with O'Farrell's (1975) precooled electrophoresis buffer and the gels run at 4 ° C. The proteins were electrophoresed off of the first dimension gel applying 400 V for 2 min. Then the voltage was reduced to 110 V and thereafter gradually increased to 200 V during a 7 h separation, limiting the power to 3.5 watts/ gel. The separation was terminated when the bromophenolblue approached the bottom of the gel. The gels were fixed and prestained overnight in 1% TCA, 7.5% acetic acid, 50% methanol and 50 rag/1 methylene blue. The gels were stained for 90 min in 1% TCA, 7.5% acetic acid and 50 mg/1 methylene blue and destained in 7.5% acetic acid. The gels were dried and exposed to Du Pont Cronex 2 X-ray film. Protein spots were punched out, transferred into scintillation vials and covered with 75 gl H20 2 to which 2% NH4OH was added. The vials were tightly capped and heated to 70°C for 2 h. Immediately after the vials were put on a metal sheet that had been cooled to - 8 0 ° C and were than kept at - 8 0 ° C for 1 h. The cooling procedure trapped gaseous 3H20 efficiently at the bottom of the vials. After melting, the samples were treated with 100 gl beef liver catalase for 10 rain (Sigma C-30, 12,600 units/rag, 23 mg/ml, 0.5% (v/v) Catalase and 25 mM Tris-HC1, pH 8). 0.5 ml Lumasove (Lumac) and 5 ml scintillation mix were added and the 3sS/3H ratio was determined in a liquid scintillation spectrometer.
Results The rate of total protein synthesis decreases exponentially in t h e s t r a i n ts 136, as m e a s u r e d b y i n c o r p o r a t i o n o f six m i n p u l s e s o f 3 S S - m e t h i o n i n e a t v a r i o u s
1 ."2_ 0.8 0
Electrophoresis. Electrophoresis procedures and media were those of O'Farrell (1975) except where otherwise indicated.
\
E=,.,0.6 x
>
a) First Dimension Isoelectric focusing gels were prepared with 1.5% Ampholines pH 5-7 and 0.5% pH 3.5-10 (LKB) in glass tubing of 110 m m x 1.5 mm inside diameter. 1"he gels were pre-run (O'Farrell, 1975), then approximately 20 ~tl protein samples were loaded onto the gels. Isoelectric focusing was accomplished in approximately 12h at 300 V and 45 rain at 800 V. The gels were extruded from the glass tubing, and gently shaken for 10 rain in 5 ml SDS sample buffer (O'Farrell, 1975), to which 0.4% of a 1% bromophenol blue solution had been added.
b) Second Dimension The apparatus and procedure for SDS-polyacrylamide gel electrophoresis have been simplified extensively (H. Koch, manuscript in preparation), 10% separation gel was made as described by O'Farrell (1975). After 30 min the stacking gel (O'Farrell, 1975) was applied. Immediately after polymerization the first dimension gel was gently pushed in between the 1.1 mm spacing on top of the stacking
o 0.4
\
\
x
0.2
°lo
~2
lb 2b
,\
6b 70 rain
Fig.1. The relative rate of overall protein synthesis in S. cerevisiae ts 136 following temperature shift. Samples were putseAabelled with [35S]-methionine as indicated at various times after shift and mixed with [3H]-methionine-labelled reference cells. Approx. 1 gl of each of these mixtures was precipitated in TCA as described in Materials and Methods. The ratio of 35S/all of the first pulse was normalized to 1. Symbols refer to the average of four experiments
H. Koch and J.D. Yriesen: individual Messenger RNA Half Lives in Saccharom),ces cerel,isiae
131
Fig. 2A-C. Two-dimensional gel electrophoresis of soluble yeast proteins. Electrophoresis was performed by a technique of O'Farrell (4) modified as indicated in Materials and Methods. The pH gradient runs from the left to the right starting at pH 3.5 and going to pH 10. SDS-electrophoresis is from top to bottom. Yeast cells were pulse-labelled as in Fig. 1. Autoradiograms were made using X-ray film that is sensitive only to 3ss. Autoradiograms are: A 0-6 min, B i5-21 min, C 50 55 min. Proteins used for quantitative evaluation were numbered from high to low molecular weight
times after t e m p e r a t u r e shift f r o m 2 3 ° C to 3 7 ° C (Fig. 1). I n c o r p o r a t i o n of label into total soluble p r o t e i n is reduced to 50% within 16 min. This is comp a r a b l e with data for m R N A half-life derived by various a u t h o r s ( H u t c h i n s o n et al., 1969; Hartwell
a n d M c L a u g h l i n , 1969; T o n n e s e n a n d Friesen, 1973) a n d identical to the elaborate m e a s u r e m e n t m a d e at 23 ° a n d 36 ° by Petersen et al., 1976). O u r measurem e n t s were m a d e only o n soluble p r o t e i n extract since a c o n s t a n t p r o p o r t i o n of label was f o u n d in cellular
132
H. K o c h and J.D. Friesen: Individual Messenger R N A Half Lives in Saccharomyces cerevisiae
debris not soluble in 9.5 M urea and 5%/?-mercaptoethanol following cell disintegration. Differential measurements of the residual rate of protein synthesis made by pulse labelling (Fig. 1) in our hands yielded greater reproducibility compared to integral measurements of the total protein synthesis (i.e. continuous uptake of label) with the temperature sensitive strain. The slope of the decay curve, shown in Fig. 1, is slightly reduced at later times. Several mechanisms might account for this effect, among them leakiness of the mutant, mitochondrial protein synthesis or a superposition of many individual decay functions with different decay time constants. The data presented below bear out the last possibility. The decay constant of individual proteins was measured by separating the total protein extract of yeast into its components by two dimensional gel electrophoresis. The autoradiograms of several such electropherograms are shown in Fig. 2. Up to 1850 different proteins spots can be counted on a single electropherogram loaded with approximately 50 gg of protein. The mode of the protein distribution is approximately half a pH unit more acidic for S. cerevisiae than for E. coli, as has been found by an electrophoresis of combined extracts of the two organisms (data not shown). For many different haploid and diploid yeast strains very similar protein profiles were found (data not shown). No significant differences in the spot pattern were found in a comparison of extracts from the mutant ts 136 and its parent strain A364A when both were grown at the permissible temperature (23 ° C) (unpublished observation). Interesting protein spots were selected after visual comparison of the autoradiograms derived from samples labelled at various times following temperature shift. Spots appearing to be translated by extremely short- or long-lived mRNA were included. The spots selected were numbered from high to low molecular weight on all electropherograms, cut out and the ratio of 3 5 S / 3 H determined. The relative rate of synthesis of each protein so analyzed was plotted against time after temperatureshift (Fig. 3). A number of different decay patterns were found for the proteins investigated. The variations in decay half-life range from 3.5 to more than 70 rain (Table 1). Several of the decay curves show a sigmoidal shape with a shoulder width varying from 0 to 30 min (Fig. 3). Three of the proteins investigated show an initial stimulation in synthesis after temperature shift (Fig. 3) followed by a decay. Some proteins show a "tail" in their decay curve of up to 65% of the normal synthetic capacity (Fig. 3). Messenger RNA molecules for the major class
~,,.. --~o
"z
\
b
11
2
a°o Ib ~ 3b ~ 5b 6b 7'o
i
,
10 20
10
,
,
30
/.0
,
,
i
50 60 70 rnin
001
0.1
c
o.ol
\ 11 12
13
lb ~ 3b 4'o s'o do 7b o Ib ~ 3b ~b ~.b 7'o
).01
min
.= 1¸
0
O.i
e
o.o~
\ 21
16 2b ~ 4b ~ ~ 7b
\ 24
lb 2b 3~ ~b s; ~o ~o
3.01
min
Fig. 3a-f. Relative residual synthetic ratio of proteins in S. cerevisiae ts 136 as a function of time following temperature shift. N u m b e r s refer to the autoradiograms shown in Fig. 2. Each protein was punched out and the ratio of 3sS/3H was determined. For each individual protein the ratio of the first pulse was normalized to 1
of proteins, including those with the highest individual abundance, show simple first order (exponential) decay kinetics. This class is represented by spots No. 2, 4, 10, 11, 12, 13, 15, 16, 17, 18, 19, 24, 26. Some of those showed a shoulder before the onset of ex-
H. Koch and J.D. Friesen: Individual Messenger R N A Half Lives in Saccharomyees cerevisiae Table 1. Individual m R N A half-lives. Figures were obtained by measuring the slopes of the inactivation curves not regarding the shoulders. For composite curves the shortest half life obtained is indicated in the first column, the longer half life in the second column. Values in brackets indicate that the relative synthesis increases before decrease. - - - means no significant decay observable Spot No.
1
2 3 4 5 6 7 8 9 10 11 12 13 14 "
T 1/2 in rain
T 1[2 in min
Spot No.
T 1/2 in min
G)
-
15
14
16 17 18 19 20 21 22 23 24 25 26 27
12 6,5 10 11.5 3.5 5 11 16 8.5 12 14.5
11.5 6.5 16.5 30.5 " 18 13.5 15 12.5 11.5 10 14.5 (17)
60
38.5 37
T 1/2 in rain
133
altered by the experimental treatment. An alteration in protein processing during the t-shift should show up in the gels by dislocation of spots either in position or quantity increasing with time after t-shift. Neither autoradiograms nor staining of the gels show such an effect. Therefore we conclude that if the t-shift should change protein processing, the influence on our results should be negligible.
Discussion
34_5 15 35
half-life values cannot be derived from the curve
ponential decay (No. 4, 16, 18). There is one high molecular weight protein (No. 5) with a very long shoulder before exponential decay. A second class of m R N A shows concave decay curves: (spots No. 3, 7, 8, 9, 20, 21, 22, 23, 25, 27). Their possible significance will be discussed later. Two of the proteins falling into this class show an initial decrease in synthesis followed by an apparently stable component resulting in 65% of the initial synthesis (spot 25) or 35% (spot 23). A third class of translation products shows an initial increase in synthesis rate (spots 1, 6, 14) which is followed by a very slow, in some cases undetectable, decay. The high molecular weight protein represented by spot No. 1 is remarkable. Its synthesis rate does not increase for 10 min after temperature shift, then increases by at least a factor of two and finally 30 min after shift shows a very slow decay. Spots 7, 8, 10 and 11 have nearly the same molecular weights and identical decay curves. Spots 13, 16 and 18 represent those with the highest individual abundance as can be seen from Fig. 2. These three proteins show very similar isoelectric points, widely different molecular weights but rather similar decay curves. Possible effects of protein processing are largely circumvented by our technique. As both t-shifted and reference cells are methionine-labelled and are handled together the ratio of 35s/all cannot become
We believe that the method used to measure m R N A decay rate used in the present study has certain advantages over other approaches. Functional m R N A in general is defined by its capability for translation, and is not necessarily identical to its chemical half-life, as has been determined by others (Petersen et al., 1976). Moreover, the use of inhibitors to block m R N A synthesis has disadvantages, such as possible interference with the function of m R N A (Lee et al., 1971). Investigations employing polysome-bound m R N A to measure the half life involve no alteration of the cellular metabolism but they require the assumption that the decay of m R N A is random (Petersen et al., 1976), which is not the case as our data show; neither does this method readily allow the measurement of individual m R N A species. In the present study a temperature sensitive mutant (ts 136), which shows a marked inhibition of either m R N A export, transcription or processing at 36 ° C, was chosen to block appearance of new mRNA. Since the half life obtained for total m R N A in the present study is in good agreement with other estimates we feel we have a sound basis for the estimation of individual m R N A half lives. From our measurements we can differentiate decay curves into three discrete classes.
1. Simple Exponential This is the shape of m R N A decay curve one would expect if there were random withdrawal of m R N A from the translatable pool. Even if m R N A for a specific protein were longer than the actual coding sequence, as has been suggested for ribosomal proteins (Mager et al., 1977), one would expect first-order decay kinetics since a proportion of m R N A molecules would be unmasked in the act of being destroyed at the moment of temperature shift. In other words at the time of the temperature shift one expects that the system for production and destruction of m R N A would have been in equilibrium. A variation on this
134
H. Koch and J.D. Friesen: Individual Messenger RNA Half Lives in Saccharomyces cerevisiae
theme is the appearance of a slight shoulder on some of the decay curves. A n u m b e r o f possible explanations could account for this class. First, the temperature shift could effect a preferential increase in the translation efficiency of these classes of m R N A . F o r a short time this could establish a new steady state of protein p r o d u c t i o n for that gene product. Eventually, as the source of new m R N A has been cut off, n o r m a l decay is revealed. Second, one could imagine that the defect in ts 136 lies not in the transport o f R N A f r o m nucleus to cytoplasm (Branes and Pogo, 1975), but in the processing o f nongenic sequences f r o m the m R N A (A. Hopper, personal communication). Thus, only the class o f m R N A which has no insertion would continue to be p r o d u c e d and translated under nonpermissive conditions. F o r m R N A species with nontranslated sequences, their site within the m R N A molecule might have an influence on the shape of the decay curve. If m R N A processing and transcription were to occur paripassu, then those molecules which had been transcribed past the position of the nontranslated sequence at the time of temperature shift could proceed to translation. The effect would be a delay in the onset o f decay. The position of the non-translated sequence could influence the extent o f the delay: the closer is the p r o m o t e r to the nontranslated sequence, the longer is the delay.
2, Concave Upward A trivial explanation for this class might involve the near coincidence of two different gene products of nearly identical isoelectric points and very similar mo= bility. We c a n n o t rule out this possibility, although the frequency of appearance of this class of decay curves seems rather too high to account for such a r a n d o m effect. One could also imagine that there could be more than one cistron coding for a given protein; the non-translated position of the two different m R N A s could result in different half-lives because of varieties in m R N A secondary structure. A n o t h e r possibility is that m R N A molecules transcribed f r o m one given gene might become segregated to different classes o f ribosomes, for example cytoplasmic and m e m b r a n e - b o u n d ones. Possible differences in translation step times or accessibility to ribonuclease between the two classes of ribosomes might account for discrete classes at m R N A decay rates.
3. Initial Stimulation, Slow Decay It is possible that ts 136 is leaky for specific m R N A species. If the m u t a n t has a defect in m R N A process-
ing, those messages showing no decay might have no non-translated sequences. A n o t h e r possibility might be that these messages are transcribed f r o m non-nuclear D N A . We do k n o w that several mitochondrial proteins are coded for by mitochondrial D N A (Perlman et al., 1977). Some of these translation products m a y appear in our gels. It can be seen f r o m our results that the bulk of m R N A is decreased after the t-shift. This m a y mean that those messages which are either of mitochondrial origin or are not subject to the m u t a t i o n induced decline find more ribosomes for their translation giving rise to the initial stimulation.
Acknowledgements. We thank Heidi Lotz for technical assistance, Gordon Temple and Udo Leins for photography and Joan Sheehy and Eva Maria Peter for typing. Very helpful suggestions by Jfirgen Kiefer are gratefully appreciated. This research was supported by the National Cancer Institute of Canada and the National Research Council of Canda (Grant No. A5734) and by a grant from the Bundesministerium ffir Forschung und Technologic, Germany, H.K. was supported during the initial phase of these studies by a NATO Fellowship.
References Branes, L., Pogo, A.O.: Biogenesis of polysomes and transport of messenger RNA in yeast. Eur. J. Biochem. 54, 317-328 (1975) Cooper, T.G., Marcelli, G., Sumrada, R.: Factors influencing the observed half-lives of specific synthetic capacities in Saccharomyees cerevisiae. Biochim. Biophys. Acta 517, 464 472 (1978) Hartwell, L.H. : Macromolecule synthesis in temperature-sensitive mutants of yeast. J. Bacteriol. 93, 1662-1670 (1967) Hartwe11, L.H., Hutchinson, H.T., Holland, T.M., McLaughlin, C.S. : The effect of cycloheximide upon polyribosome stability in two yeast mutants defective respectively in the initiation of polypeptide chains and messenger RNA synthesis. Mol. Gen. Genet. 106, 347-361 (1970) Hartwell, L.H., McLaughlin, C.S.: Temperature-sensitive mutants of yeast exhibiting a rapid inhibition of protein synthesis. J. Bacteriol. 96, 1664-1671 (1968) Hartwell, L.H., McLaughlin, C.S.: A mutant of yeast apparently defective in the initiation of protein synthesis. Proe. Natl. Acad. Sci. U.S.A. 62, 468-474 (1969) Hereford, L.M., Rosbash, M.: Regulation of a set of abundant mRNA sequences. Cell 10, 463-467 (1977) Hutchinson, H. Terry, Hartwell, L.H., McLaughlin, C.S, : Temperature-sensitive yeast mutant defective in ribonucleic acid production. J. Bacteriol. 93, 807-814 (1969) Lee, S.Y., Krsmanovic, V., Brawerman, G. : Initiation of polysome formation in mouse sarcoma 180 ascites cells. Utilization of cytoplasmic messenger ribonucleic acid. Biochemistry 10, 895-900 (1971) Mager, W.H., Retel, J., Planta, R.J., Bollen, G.H.P.M., de Regt, V.C.H.F., Hoving, H.: Transcriptional units for ribosomal proteins in yeast. Eur. J. Biochem. 78, 575-583 (1977) O'Farrell, P.H.: High resolution two-dimensional electrophoresis of proteins. J. Biol. Chem. 250, 4007-4021 (1975)
H. Koch and J.D. Friesen: Individual Messenger RNA Half Lives in Saccharomyces cerevisiae Perlman, P.S., Douglas, M.G., Strausberg, R.L., Butow, R.A.: Localization of genes for variant forms of mitochondrial proteins on mitochondrial DNA of Saccharomyces cerevisiae. J. Mol. Biol. 115, 675 694 (1977) Petersen, N.S., McLaughlin, C.S., Nierlich, D.P. : Half life of yeast messenger RNA. Nature 260, 70 72 (1976) Schlessinger, D., Jacobs, K.A., Gupta, R.S., Kano, Y., Imanoto, F.: Decay of individual Escherichia coli trp messenger RNA molecules is sequentially ordered. J. Mol. Biol. 110, 421 439 (1977) Shiokawa, K., Pogo, A.O.: The role of cytoplasmic membranes in controlling the transport of nuclear messenger RNA and initiation of protein synthesis. Proc. Natl. Acad. Sci. U.S.A. 71, 2658-2662 (1974) Singer, R.H., Penman, S. : Stability of HeLa ceil mRNA in actinomycin. Nature 240, 100 102 (1972)
135
Talkad, V., Schneider, E., Kennell, D. : Evidence for variable rates of ribosome movement in Escherichia coli. J. Mol. Biol. 104, 299-303 (1976) Tonnesen, T., Friesen, J.D. : Inhibitors of ribonucleic acid synthesis in Saccharomyces eerevisiae: decay rate of messenger ribonucleic acid. J. Bacteriol. 115, 889-896 (1973) Waldron, C., Jund, R., Lacroute, F.: Evidence for a high proportion of inactive ribosomes in slow-growing yeast cells. Biochem. J. 168, 409 415 (1977)
Communicated
by H.G. Wittmann
Received August 4 / November 1, 1978
Individual Messenger RNA Half Lives in Saccharomyces cerevisiae H. Koch Strahlenzentrum der Justus-Liebig-Universitfit,D-6300 GieBen, Federal Republic of Germany J.D. Friesen Department of Biology, York University, 4700 Keele Street, Downsview, Ontario, Canada M3J 1P3
Summary. We have measured the decay half-life of functional messenger R N A ( m R N A ) for some thirty different proteins in the yeast Saccharomyces cerevisiae. Production of newly synthesized m R N A was halted by raising the temperature of a culture of a temperature-sensitive mutant, ts 136. Aliquots of this culture were pulsed-labelled with [3sS]-methionine at various times after the temperature shift and the radioactive proteins separated on the two-dimensional gel electrophoresis system of O'Farrell. We find a range in the decay half lives of individual m R N A species which varies f r o m 3.5 min to greater than 70 min. We find three general classes of decay curves, (a) simple exponential (first order); some of these showed a shoulder before onset of exponential decay; (b) bi-component or multi-component concave upward; (c) initial stimulation of rate of m R N A synthesis, followed by virtually undetectable decay.
Introduction The decay constant for messenger R N A ( m R N A ) in the yeast Saccharomyces cerevisiae has been measured both for global m R N A as well as for certain individual m R N A species (Hartwell, 1967; Tonnesen and Friesen, 1973; Hereford and Rosbash, 1977; Cooper et al., 1978). These results point to the likelihood of considerable variation in individual decay rates of different m R N A species, allowing for fast regulatory response for some translation products and comparably slow response for others. The average half-life for overall m R N A decay appears to be approximately 17 __+3 min. The methods used previously to determine m R N A decay constants have suffered from certain shortcomings. F o r example, the use of inhibitors inFor offprints contact: H. Koch
troduces the uncertainty that the inhibitor itself might interfere with the initiation of translation or the stability of m R N A (Singer and Penman, 1972). Or, measurements of m R N A life time by the kinetics of precursor incorporation assume r a n d o m decay (Petersen et al., 1976), which is not necessarily the case and furthermore requires certain assumptions regarding equilibration of intracellular pools with exogenous radioactive precursors. Again, measurements in spheroplasts may not apply to physiological condition of normal cell growth. We have sought to circumvent some of these problems and to provide an independent means of measuring individual functional m R N A half-lives by combining the use of a temperature sensitive (ts) mutant with 2-dimensional gel electrophoresis. There is evidence that mutant ts 136 of Saccharomyces cerevisiae is defective in some function which leads to the appearance of inhibited transport of all classes of R N A from the nucleus into the cytoplasm at the non-permissive temperature (Hutchinson et al., 1969; Hartwell et al., 1970; Tonnesen and Friesen, 1973; Shiokawa and Pogo, 1974; Hereford and Rosbash, 1977; Cooper et al., 1978). Certainly it appears that all classes of R N A synthesis are inhibited in this mutant at 37 ° C. This strain has been used to measure the availability of functional m R N A for translation (Tonnesen and Friesen, 1973).
Material and Methods Strains and Media. Saccharomycescerevisiae strain ts 136 (Hutchinson et al., 1969), was given to us by L.H. Hartwell. The strain was grown in yeast nitrogen base (YNB) medium supplemented (per litre) with 5 g succinic acid, 3 g NaOH, 20 g glucose, 20 mg each of histidine, lysine and tyrosine, 125 nag each of adenine and uracil and 0.5 g yeast extract (Difco). The strain was grown at 23° C; the doubling time was approximately 190 rain. Growth was monitored by measuring the absorbance at 436 nm with a filter
0026- 892 5/79/0170/0129/$01.40
130
H. Koch and J.D. Friesen: Individual Messenger RNA Half Lives in Saeeharomyces cerevisiae
photometer. An optical density of 1 corresponds to approximately 2 x 107 cells/ml.
Radioactive Labelling. Exponentially growing cultures were shifted at A436= 1 to 37°C and pulse-labelled with 17.5 ~tCi/ml [35S]methionine (755Ci/mmol, Amersham) at various times after the shift. The label was diluted with non-radioactive methionine to give a final concentration of 1.3 x 10 -3 gmol/ml in the culture. The labelling period of 6 min was terminated by adding 250 Ixg/ml methionine; 9 min later the cultures were rapidly chilled to 0° C. Portions of 20 gl were withdrawn and precipitated in 3 ml of icecold 5% trichloroacetic acid. The precipitates were collected on nitrocellulose membrane filters and washed three times with 3 ml each of 5% trichloroacetic acid. 5 ml xylene based scintillation mix was added after drying and the radioactivity was determined in a liquid scintillation spectrometer. Reference cells were taken from the original culture and labelled with 10gCi/ml [methyl-3H]-methionine (8.8 Ci/mmol, 59 mCi/mg, Amersham) for 140 rain at 23° C. The labelling was terminated as described for pulseqabelled cells.
Extraction of Protein. 35S-methionine pulse-labelled cells and 3Hlabelled cells were washed. Equal amounts of 3H-labelled cells were added to each pulselabelled culture. The samples were transferred to Eppendorf vials and 0.3 g glass beads (0.45-0.5 mm diameter) were added. The vials were put into a pre-cooled adapter fitting into a Braun homogenizer and the cells were disrupted for 5 min. The homogenate was collected and the glass beads were washed twice. To the combined homogenate and washings a mixture was added to give a final concentration of 50 mM Tris-HC1, 0.5 mM MgClz at pH 7.4 and 20 gg/ml of bovine pancreas deoxyribonuclease and ribonuclease each (DN 25 and R 4875, Sigma). The samples were kept at 0°C for 30 min, then frozen in Eppendorf vials at - 6 0 ° C and freeze-dried overnight. The dried samples were dissolved in 50 gl lysis buffer (O'Farrell, 1975). Cell debris was sedimented in an Eppendorf centrifuge and from the supernatant protein extract 2 gl portions were precipitated in 3 ml 5% TCA for radioactivity determination and were treated as described above.
gel. Care was taken to avoid trapping air between the gels. The gel set was clamped to an electrophoresis apparatus, the chambers were filled with O'Farrell's (1975) precooled electrophoresis buffer and the gels run at 4 ° C. The proteins were electrophoresed off of the first dimension gel applying 400 V for 2 min. Then the voltage was reduced to 110 V and thereafter gradually increased to 200 V during a 7 h separation, limiting the power to 3.5 watts/ gel. The separation was terminated when the bromophenolblue approached the bottom of the gel. The gels were fixed and prestained overnight in 1% TCA, 7.5% acetic acid, 50% methanol and 50 rag/1 methylene blue. The gels were stained for 90 min in 1% TCA, 7.5% acetic acid and 50 mg/1 methylene blue and destained in 7.5% acetic acid. The gels were dried and exposed to Du Pont Cronex 2 X-ray film. Protein spots were punched out, transferred into scintillation vials and covered with 75 gl H20 2 to which 2% NH4OH was added. The vials were tightly capped and heated to 70°C for 2 h. Immediately after the vials were put on a metal sheet that had been cooled to - 8 0 ° C and were than kept at - 8 0 ° C for 1 h. The cooling procedure trapped gaseous 3H20 efficiently at the bottom of the vials. After melting, the samples were treated with 100 gl beef liver catalase for 10 rain (Sigma C-30, 12,600 units/rag, 23 mg/ml, 0.5% (v/v) Catalase and 25 mM Tris-HC1, pH 8). 0.5 ml Lumasove (Lumac) and 5 ml scintillation mix were added and the 3sS/3H ratio was determined in a liquid scintillation spectrometer.
Results The rate of total protein synthesis decreases exponentially in t h e s t r a i n ts 136, as m e a s u r e d b y i n c o r p o r a t i o n o f six m i n p u l s e s o f 3 S S - m e t h i o n i n e a t v a r i o u s
1 ."2_ 0.8 0
Electrophoresis. Electrophoresis procedures and media were those of O'Farrell (1975) except where otherwise indicated.
\
E=,.,0.6 x
>
a) First Dimension Isoelectric focusing gels were prepared with 1.5% Ampholines pH 5-7 and 0.5% pH 3.5-10 (LKB) in glass tubing of 110 m m x 1.5 mm inside diameter. 1"he gels were pre-run (O'Farrell, 1975), then approximately 20 ~tl protein samples were loaded onto the gels. Isoelectric focusing was accomplished in approximately 12h at 300 V and 45 rain at 800 V. The gels were extruded from the glass tubing, and gently shaken for 10 rain in 5 ml SDS sample buffer (O'Farrell, 1975), to which 0.4% of a 1% bromophenol blue solution had been added.
b) Second Dimension The apparatus and procedure for SDS-polyacrylamide gel electrophoresis have been simplified extensively (H. Koch, manuscript in preparation), 10% separation gel was made as described by O'Farrell (1975). After 30 min the stacking gel (O'Farrell, 1975) was applied. Immediately after polymerization the first dimension gel was gently pushed in between the 1.1 mm spacing on top of the stacking
o 0.4
\
\
x
0.2
°lo
~2
lb 2b
,\
6b 70 rain
Fig.1. The relative rate of overall protein synthesis in S. cerevisiae ts 136 following temperature shift. Samples were putseAabelled with [35S]-methionine as indicated at various times after shift and mixed with [3H]-methionine-labelled reference cells. Approx. 1 gl of each of these mixtures was precipitated in TCA as described in Materials and Methods. The ratio of 35S/all of the first pulse was normalized to 1. Symbols refer to the average of four experiments
H. Koch and J.D. Yriesen: individual Messenger RNA Half Lives in Saccharom),ces cerel,isiae
131
Fig. 2A-C. Two-dimensional gel electrophoresis of soluble yeast proteins. Electrophoresis was performed by a technique of O'Farrell (4) modified as indicated in Materials and Methods. The pH gradient runs from the left to the right starting at pH 3.5 and going to pH 10. SDS-electrophoresis is from top to bottom. Yeast cells were pulse-labelled as in Fig. 1. Autoradiograms were made using X-ray film that is sensitive only to 3ss. Autoradiograms are: A 0-6 min, B i5-21 min, C 50 55 min. Proteins used for quantitative evaluation were numbered from high to low molecular weight
times after t e m p e r a t u r e shift f r o m 2 3 ° C to 3 7 ° C (Fig. 1). I n c o r p o r a t i o n of label into total soluble p r o t e i n is reduced to 50% within 16 min. This is comp a r a b l e with data for m R N A half-life derived by various a u t h o r s ( H u t c h i n s o n et al., 1969; Hartwell
a n d M c L a u g h l i n , 1969; T o n n e s e n a n d Friesen, 1973) a n d identical to the elaborate m e a s u r e m e n t m a d e at 23 ° a n d 36 ° by Petersen et al., 1976). O u r measurem e n t s were m a d e only o n soluble p r o t e i n extract since a c o n s t a n t p r o p o r t i o n of label was f o u n d in cellular
132
H. K o c h and J.D. Friesen: Individual Messenger R N A Half Lives in Saccharomyces cerevisiae
debris not soluble in 9.5 M urea and 5%/?-mercaptoethanol following cell disintegration. Differential measurements of the residual rate of protein synthesis made by pulse labelling (Fig. 1) in our hands yielded greater reproducibility compared to integral measurements of the total protein synthesis (i.e. continuous uptake of label) with the temperature sensitive strain. The slope of the decay curve, shown in Fig. 1, is slightly reduced at later times. Several mechanisms might account for this effect, among them leakiness of the mutant, mitochondrial protein synthesis or a superposition of many individual decay functions with different decay time constants. The data presented below bear out the last possibility. The decay constant of individual proteins was measured by separating the total protein extract of yeast into its components by two dimensional gel electrophoresis. The autoradiograms of several such electropherograms are shown in Fig. 2. Up to 1850 different proteins spots can be counted on a single electropherogram loaded with approximately 50 gg of protein. The mode of the protein distribution is approximately half a pH unit more acidic for S. cerevisiae than for E. coli, as has been found by an electrophoresis of combined extracts of the two organisms (data not shown). For many different haploid and diploid yeast strains very similar protein profiles were found (data not shown). No significant differences in the spot pattern were found in a comparison of extracts from the mutant ts 136 and its parent strain A364A when both were grown at the permissible temperature (23 ° C) (unpublished observation). Interesting protein spots were selected after visual comparison of the autoradiograms derived from samples labelled at various times following temperature shift. Spots appearing to be translated by extremely short- or long-lived mRNA were included. The spots selected were numbered from high to low molecular weight on all electropherograms, cut out and the ratio of 3 5 S / 3 H determined. The relative rate of synthesis of each protein so analyzed was plotted against time after temperatureshift (Fig. 3). A number of different decay patterns were found for the proteins investigated. The variations in decay half-life range from 3.5 to more than 70 rain (Table 1). Several of the decay curves show a sigmoidal shape with a shoulder width varying from 0 to 30 min (Fig. 3). Three of the proteins investigated show an initial stimulation in synthesis after temperature shift (Fig. 3) followed by a decay. Some proteins show a "tail" in their decay curve of up to 65% of the normal synthetic capacity (Fig. 3). Messenger RNA molecules for the major class
~,,.. --~o
"z
\
b
11
2
a°o Ib ~ 3b ~ 5b 6b 7'o
i
,
10 20
10
,
,
30
/.0
,
,
i
50 60 70 rnin
001
0.1
c
o.ol
\ 11 12
13
lb ~ 3b 4'o s'o do 7b o Ib ~ 3b ~b ~.b 7'o
).01
min
.= 1¸
0
O.i
e
o.o~
\ 21
16 2b ~ 4b ~ ~ 7b
\ 24
lb 2b 3~ ~b s; ~o ~o
3.01
min
Fig. 3a-f. Relative residual synthetic ratio of proteins in S. cerevisiae ts 136 as a function of time following temperature shift. N u m b e r s refer to the autoradiograms shown in Fig. 2. Each protein was punched out and the ratio of 3sS/3H was determined. For each individual protein the ratio of the first pulse was normalized to 1
of proteins, including those with the highest individual abundance, show simple first order (exponential) decay kinetics. This class is represented by spots No. 2, 4, 10, 11, 12, 13, 15, 16, 17, 18, 19, 24, 26. Some of those showed a shoulder before the onset of ex-
H. Koch and J.D. Friesen: Individual Messenger R N A Half Lives in Saccharomyees cerevisiae Table 1. Individual m R N A half-lives. Figures were obtained by measuring the slopes of the inactivation curves not regarding the shoulders. For composite curves the shortest half life obtained is indicated in the first column, the longer half life in the second column. Values in brackets indicate that the relative synthesis increases before decrease. - - - means no significant decay observable Spot No.
1
2 3 4 5 6 7 8 9 10 11 12 13 14 "
T 1/2 in rain
T 1[2 in min
Spot No.
T 1/2 in min
G)
-
15
14
16 17 18 19 20 21 22 23 24 25 26 27
12 6,5 10 11.5 3.5 5 11 16 8.5 12 14.5
11.5 6.5 16.5 30.5 " 18 13.5 15 12.5 11.5 10 14.5 (17)
60
38.5 37
T 1/2 in rain
133
altered by the experimental treatment. An alteration in protein processing during the t-shift should show up in the gels by dislocation of spots either in position or quantity increasing with time after t-shift. Neither autoradiograms nor staining of the gels show such an effect. Therefore we conclude that if the t-shift should change protein processing, the influence on our results should be negligible.
Discussion
34_5 15 35
half-life values cannot be derived from the curve
ponential decay (No. 4, 16, 18). There is one high molecular weight protein (No. 5) with a very long shoulder before exponential decay. A second class of m R N A shows concave decay curves: (spots No. 3, 7, 8, 9, 20, 21, 22, 23, 25, 27). Their possible significance will be discussed later. Two of the proteins falling into this class show an initial decrease in synthesis followed by an apparently stable component resulting in 65% of the initial synthesis (spot 25) or 35% (spot 23). A third class of translation products shows an initial increase in synthesis rate (spots 1, 6, 14) which is followed by a very slow, in some cases undetectable, decay. The high molecular weight protein represented by spot No. 1 is remarkable. Its synthesis rate does not increase for 10 min after temperature shift, then increases by at least a factor of two and finally 30 min after shift shows a very slow decay. Spots 7, 8, 10 and 11 have nearly the same molecular weights and identical decay curves. Spots 13, 16 and 18 represent those with the highest individual abundance as can be seen from Fig. 2. These three proteins show very similar isoelectric points, widely different molecular weights but rather similar decay curves. Possible effects of protein processing are largely circumvented by our technique. As both t-shifted and reference cells are methionine-labelled and are handled together the ratio of 35s/all cannot become
We believe that the method used to measure m R N A decay rate used in the present study has certain advantages over other approaches. Functional m R N A in general is defined by its capability for translation, and is not necessarily identical to its chemical half-life, as has been determined by others (Petersen et al., 1976). Moreover, the use of inhibitors to block m R N A synthesis has disadvantages, such as possible interference with the function of m R N A (Lee et al., 1971). Investigations employing polysome-bound m R N A to measure the half life involve no alteration of the cellular metabolism but they require the assumption that the decay of m R N A is random (Petersen et al., 1976), which is not the case as our data show; neither does this method readily allow the measurement of individual m R N A species. In the present study a temperature sensitive mutant (ts 136), which shows a marked inhibition of either m R N A export, transcription or processing at 36 ° C, was chosen to block appearance of new mRNA. Since the half life obtained for total m R N A in the present study is in good agreement with other estimates we feel we have a sound basis for the estimation of individual m R N A half lives. From our measurements we can differentiate decay curves into three discrete classes.
1. Simple Exponential This is the shape of m R N A decay curve one would expect if there were random withdrawal of m R N A from the translatable pool. Even if m R N A for a specific protein were longer than the actual coding sequence, as has been suggested for ribosomal proteins (Mager et al., 1977), one would expect first-order decay kinetics since a proportion of m R N A molecules would be unmasked in the act of being destroyed at the moment of temperature shift. In other words at the time of the temperature shift one expects that the system for production and destruction of m R N A would have been in equilibrium. A variation on this
134
H. Koch and J.D. Friesen: Individual Messenger RNA Half Lives in Saccharomyces cerevisiae
theme is the appearance of a slight shoulder on some of the decay curves. A n u m b e r o f possible explanations could account for this class. First, the temperature shift could effect a preferential increase in the translation efficiency of these classes of m R N A . F o r a short time this could establish a new steady state of protein p r o d u c t i o n for that gene product. Eventually, as the source of new m R N A has been cut off, n o r m a l decay is revealed. Second, one could imagine that the defect in ts 136 lies not in the transport o f R N A f r o m nucleus to cytoplasm (Branes and Pogo, 1975), but in the processing o f nongenic sequences f r o m the m R N A (A. Hopper, personal communication). Thus, only the class o f m R N A which has no insertion would continue to be p r o d u c e d and translated under nonpermissive conditions. F o r m R N A species with nontranslated sequences, their site within the m R N A molecule might have an influence on the shape of the decay curve. If m R N A processing and transcription were to occur paripassu, then those molecules which had been transcribed past the position of the nontranslated sequence at the time of temperature shift could proceed to translation. The effect would be a delay in the onset o f decay. The position of the non-translated sequence could influence the extent o f the delay: the closer is the p r o m o t e r to the nontranslated sequence, the longer is the delay.
2, Concave Upward A trivial explanation for this class might involve the near coincidence of two different gene products of nearly identical isoelectric points and very similar mo= bility. We c a n n o t rule out this possibility, although the frequency of appearance of this class of decay curves seems rather too high to account for such a r a n d o m effect. One could also imagine that there could be more than one cistron coding for a given protein; the non-translated position of the two different m R N A s could result in different half-lives because of varieties in m R N A secondary structure. A n o t h e r possibility is that m R N A molecules transcribed f r o m one given gene might become segregated to different classes o f ribosomes, for example cytoplasmic and m e m b r a n e - b o u n d ones. Possible differences in translation step times or accessibility to ribonuclease between the two classes of ribosomes might account for discrete classes at m R N A decay rates.
3. Initial Stimulation, Slow Decay It is possible that ts 136 is leaky for specific m R N A species. If the m u t a n t has a defect in m R N A process-
ing, those messages showing no decay might have no non-translated sequences. A n o t h e r possibility might be that these messages are transcribed f r o m non-nuclear D N A . We do k n o w that several mitochondrial proteins are coded for by mitochondrial D N A (Perlman et al., 1977). Some of these translation products m a y appear in our gels. It can be seen f r o m our results that the bulk of m R N A is decreased after the t-shift. This m a y mean that those messages which are either of mitochondrial origin or are not subject to the m u t a t i o n induced decline find more ribosomes for their translation giving rise to the initial stimulation.
Acknowledgements. We thank Heidi Lotz for technical assistance, Gordon Temple and Udo Leins for photography and Joan Sheehy and Eva Maria Peter for typing. Very helpful suggestions by Jfirgen Kiefer are gratefully appreciated. This research was supported by the National Cancer Institute of Canada and the National Research Council of Canda (Grant No. A5734) and by a grant from the Bundesministerium ffir Forschung und Technologic, Germany, H.K. was supported during the initial phase of these studies by a NATO Fellowship.
References Branes, L., Pogo, A.O.: Biogenesis of polysomes and transport of messenger RNA in yeast. Eur. J. Biochem. 54, 317-328 (1975) Cooper, T.G., Marcelli, G., Sumrada, R.: Factors influencing the observed half-lives of specific synthetic capacities in Saccharomyees cerevisiae. Biochim. Biophys. Acta 517, 464 472 (1978) Hartwell, L.H. : Macromolecule synthesis in temperature-sensitive mutants of yeast. J. Bacteriol. 93, 1662-1670 (1967) Hartwe11, L.H., Hutchinson, H.T., Holland, T.M., McLaughlin, C.S. : The effect of cycloheximide upon polyribosome stability in two yeast mutants defective respectively in the initiation of polypeptide chains and messenger RNA synthesis. Mol. Gen. Genet. 106, 347-361 (1970) Hartwell, L.H., McLaughlin, C.S.: Temperature-sensitive mutants of yeast exhibiting a rapid inhibition of protein synthesis. J. Bacteriol. 96, 1664-1671 (1968) Hartwell, L.H., McLaughlin, C.S.: A mutant of yeast apparently defective in the initiation of protein synthesis. Proe. Natl. Acad. Sci. U.S.A. 62, 468-474 (1969) Hereford, L.M., Rosbash, M.: Regulation of a set of abundant mRNA sequences. Cell 10, 463-467 (1977) Hutchinson, H. Terry, Hartwell, L.H., McLaughlin, C.S, : Temperature-sensitive yeast mutant defective in ribonucleic acid production. J. Bacteriol. 93, 807-814 (1969) Lee, S.Y., Krsmanovic, V., Brawerman, G. : Initiation of polysome formation in mouse sarcoma 180 ascites cells. Utilization of cytoplasmic messenger ribonucleic acid. Biochemistry 10, 895-900 (1971) Mager, W.H., Retel, J., Planta, R.J., Bollen, G.H.P.M., de Regt, V.C.H.F., Hoving, H.: Transcriptional units for ribosomal proteins in yeast. Eur. J. Biochem. 78, 575-583 (1977) O'Farrell, P.H.: High resolution two-dimensional electrophoresis of proteins. J. Biol. Chem. 250, 4007-4021 (1975)
H. Koch and J.D. Friesen: Individual Messenger RNA Half Lives in Saccharomyces cerevisiae Perlman, P.S., Douglas, M.G., Strausberg, R.L., Butow, R.A.: Localization of genes for variant forms of mitochondrial proteins on mitochondrial DNA of Saccharomyces cerevisiae. J. Mol. Biol. 115, 675 694 (1977) Petersen, N.S., McLaughlin, C.S., Nierlich, D.P. : Half life of yeast messenger RNA. Nature 260, 70 72 (1976) Schlessinger, D., Jacobs, K.A., Gupta, R.S., Kano, Y., Imanoto, F.: Decay of individual Escherichia coli trp messenger RNA molecules is sequentially ordered. J. Mol. Biol. 110, 421 439 (1977) Shiokawa, K., Pogo, A.O.: The role of cytoplasmic membranes in controlling the transport of nuclear messenger RNA and initiation of protein synthesis. Proc. Natl. Acad. Sci. U.S.A. 71, 2658-2662 (1974) Singer, R.H., Penman, S. : Stability of HeLa ceil mRNA in actinomycin. Nature 240, 100 102 (1972)
135
Talkad, V., Schneider, E., Kennell, D. : Evidence for variable rates of ribosome movement in Escherichia coli. J. Mol. Biol. 104, 299-303 (1976) Tonnesen, T., Friesen, J.D. : Inhibitors of ribonucleic acid synthesis in Saccharomyces eerevisiae: decay rate of messenger ribonucleic acid. J. Bacteriol. 115, 889-896 (1973) Waldron, C., Jund, R., Lacroute, F.: Evidence for a high proportion of inactive ribosomes in slow-growing yeast cells. Biochem. J. 168, 409 415 (1977)
Communicated
by H.G. Wittmann
Received August 4 / November 1, 1978
