patterns of protein synthesis during the cell cycle of

5 downloads 0 Views 1MB Size Report
Mar 21, 1982 - thesis using pulse labels of amino acids in synchronous cultures prepared ... these steps generate periodicities in the rate cf protein synthesis, then the prediction ..... first-order kinetics with a time to half-saturation cf 7-5 min.
J. Cell Sci. 5», 263-285 (1982) Printed in Great Britain © Company of Biologists Limited 1982

PATTERNS OF PROTEIN SYNTHESIS DURING THE CELL CYCLE OF THE FISSION YEAST SCHIZOSACCHAROMYCES POMBE J. CREANOR AND J. M. MITCHISON* Department of Zoology, University of Edinburgh, West Mams Road, Edinburgh EHg 2JT, Scotland WITH A STATISTICAL APPENDIX BY D. A. WILLIAMS,

Department of Statistics, University of Edinburgh

SUMMARY The rate of protein synthesis through the cell cycle of Schizosaccharomyces pombe has been determined from the incorporation of pulses of [JH]tryptophan in synchronous cultures prepared by selection in an elutriating rotor. This selection procedure caused minimal perturbations as judged by asynchronous control cultures, which had also been put through the rotor. The rate of synthesis showed a periodic pattern rather than a smooth exponential increase. There was a sharp increase in the rate at an 'acceleration point' at about 0-9 of the cycle. Model-fitting by a novel procedure suggests that the average single cell has an increasing rate of protein synthesis for the first 60 % of the cycle and a constant rate for the remaining 40 %. The same pattern was shown in less extensive experiments with PHjleucine and [3H]phenylalanine. It was also shown in a series of size mutants, which indicates that the pattern is not size-related, in contrast to earlier work on the rates of synthesis of messenger RNA. However, one large mutant (ede 2.M35T2O) had a significantly earlier acceleration point. Care was taken to justify the assumption that the rate of incorporation of tryptophan was a valid measure of the rate of protein synthesis. A tryptophan auxotroph was used to eliminate the problem of endogenous supply and the size of the metabolic pool was measured through th; cycle. This pool did not show cell-cycle related fluctuations. An operational model of the pools is presented.

INTRODUCTION The cell cycle of the fission yeast Schizosaccharomyces pombe has been extensively studied over the last 25 years in terms both of division control and of growth during the cycle. Nevertheless, the pattern of total protein synthesis, a major component of growth, has not been defined clearly. An early study by Mitchison & Wilbur (1962) on autoradiographs using a method equivalent to ' age-fractionation' showed an increasing rate of incorporation of labelled amino acids through the cycle but the method, like most other age-fractionations, was not sufficiently sensitive to show the fine detail of rate changes. We therefore set out to measure the rate of synthesis using pulse labels of amino acids in synchronous cultures prepared by selection with an elutriating rotor (Creanor & Mitchison, 1979). This method causes less • Author for correspondence.

264

J- Creanor andj. M. Mitchison

perturbations than other methods of synchronization in S. pombe and also allows the production of asynchronous controls. There were two particular reasons for undertaking this study. The first was to see whether protein synthesis followed one or ether of two simple patterns - a smooth exponential without periodicities in rate or a 'linear' pattern with a doubling in rate once per cycle (Mitchison, 1969). Linear patterns with periodic rate doubling have been found in S. pombe for total dry mass (Mitchison, 1957), enzyme activity (Mitchison & Creanor, 1969), messenger and ribosomal RNA (Fraser & Moreno, 1976) and carbon dioxide production (Creanor, 1978). Periodic pattern are of interest because they pose questions about the nature of the rate controls and whether they are tightly linked to the DNA-division cycle. In the case of carbon dioxide production (Creanor, 1978) and enzyme potential (Benitez, Nurse & Mitchison, 1980), the controls do not appear to be closely linked since the rate changes continue for a time after the DNAdivision cycle has been delayed or blocked. The second reason for examining protein synthesis stemmed from the work of Fraser & Nurse (1978, 1979). Their evidence suggested that the cycle position of the rate-doubling step in the linear pattern for messenger and ribosomal RNA varied according to cell size. In cells small at division, for example wee mutants, the step was later in the cycle than in wild-type cells. If these steps generate periodicities in the rate cf protein synthesis, then the prediction would be that the timing of these periodicites would also be size-related. In the first half of this paper, we present evidence that the rate of incorporation of amino acids, primarily tryptophan, dees show cell cycle periodicity, though it does not follow the simple linear pattern. The pattern is also not size-related. In the second half, we show the rate of incorporation is a valid measure of the rate of protein synthesis and examine the tryptophan pools and the balance between endogenous and exogenous supply of the precursor. MATERIALS AND METHODS

Organisms Some experiments were done with strain N.C.Y.C. 132 (A.T.C.C. 24751) of S. pombe. Most experiments, however, were done on a wild-type strain 972h~ (originally obtained from Professor U. Leupold, Bern) and its mutants. The mutants wee 1.50 and wee 2.1 divide at about half the size of the wild-type strain (Thuriaux, Nurse & Carter, 1978). wee 1.302 is partially defective in the wee 1 gene and is intermediate in size between wee 1.50 and wild type (Fantes, 1981). ede 2.M35 is an allele of ede 2 (Nurse & Thuriaux, 1980), which divides at a larger size than wild type when grown at 25 °C but is blocked at nuclear division when transfered to 35 °C. ede 2.M35T2O is likely to be a revertant in this gene. It was obtained by selecting for growth and large size at 35 °C after ultraviolet irradiation of ede 2.M35. After backcrossing to wild type and tetrad dissection, there was no reappearance of the cdc~ phenotype in five tetrads. This was confirmed with free spore analysis where there was no cdc~ segregants in 565 progeny colonies. Of 100 colonies examined, 51 had cells of normal size and 49 had large cells, indicating that a single gene was involved, trp 4.47 is a tryptophan auxotroph (provided by P. Thuriaux) with a defective phosphoribosyl transferase (Schweingruber & Dietrich, 1973).

Protein synthesis in the yeast cell cycle

265

Media and growth conditions Strain 132 was grown in a minimal medium with acetate buffer, EMM2 (Mitchison, 1970). Strain 972 and its mutants were grown in EMM3, a modified version of EMM2 with a phthalate buffer to reduce clumping with this strain. Its composition was (per litre): glucose, 10 g; NH4C1, 5 g; Na,SO4, 0.1 g; MgCl,.6 H,O, 1 g; CaCl,.2 H,O, 15 mg; KH phthalate, 3 g; Na,HPO4, i-8 g; vitamins and trace elements as in EMM2. The 'EMM3-glutamate' used for the trp 4 mutant experiments had sodium glutamate at 5 g/1 in place of the NH4C1. Cultures were shaken or stirred and, except where specified, the growth temperature was 35 °C. Some of the earlier experiments, especially with 132, were done at 32 °C but the doubling time was the same (145 min). Synchronous cultures and asynchronous control cultures Synchronous cultures were prepared by selecting small cells in a Beckman JE-6 elutriator rotor. In summary, an exponential-phase culture was pumped through the rotor and the yeast cells accumulated in the rotor cell. When sufficient cells had accumulated, the pump speed was increased and the first fraction of the effluent contained small cells early in the cycle. These were diluted, if necessary, with the same medium and grown on as a synchronous culture. Asynchronous control cultures were made in the same way except that the whole contents of the rotor cell was used. This method has been described in detail by Creanor & Mitchison (1979). The only change in technique was that the rotor, after treatment with diethyl pyrocarbonate, was washed out with 1 1 sterile distilled water followed by 1 1 warm growth medium. Radioactive tracers, cell collection and cell counting The following radioactive amino acids were used: L-[G-'H]rryptophan, 93-326 GBq/mmol (Radiochemical Centre, Amersham and New England Nuclear); L-[4,s-3H]leucine, 1-962-15 TBq/mmol; and L-phenyl-[2,3-*H]alanine, 592 GBq/mmol (both from Radiochemical Centre, Amersham). It was necessary to filter off fine paniculate matter in the ['H]tryptophan solution, immediately before use, with a Millex-GS filter (Millipore, 0-22 /tm pore size). Cells were collected on glass fibre (Whatman GF/C). For total uptake samples, the filters were washed four times with the appropriate non-radioactive amino acid at 1 mg/ml, and four times with water. For acid-insoluble incorporation, the preceding washes were followed by treatment with 10 % trichloracetic acid for 10 s and then two further washes with water. The filters were then dried, immersed in 0.56 % butyl-PDB in toluene and counted in a Packard model 2425 liquid scintilation spectrometer. Cell numbers were determined on a Coulter Counter (Industrial model D) with a 100/tm aperture.

RESULTS Pools and perturbation The only satisfactory method of determining the fine detail of the rate of synthesis of macromolecules is to use pulse labels of an appropriate precursor. Given certain assumptions, the amount of percursor incorporated over a short period is a measure of the rate of synthesis of the macromolecule. An important assumption is that there is only a small pool of precursor, which is used directly for synthesis. This pool we call the ' metabolic pool'. If this pool is small, incorporation will rise linearly from the origin in a time-course experiment in which incorporation is followed after the addition of label. If the pool is large, incorporation will proceed at an increasing rate until the pool reaches a constant specific activity. A single pulse measurement is not a

266

]f. Creanor and J. M. Mitchison

20

40 0 Time (min)

INT

40

Fig. i. Kinetics of total uptake (A —A) and incorporation (O—O) of labelled amino acids added to asynchronous exponential-phase cultures, A. f'HJleucine in strain 132 growing in EMM2 at 32 °C; 0-5 ml samples of a culture treated at time o with 19 kBq/ml. Initial cell number 2-4x10' cells/ml. B. [sH]phenylalanine in strain 972 growing in EMM3 + 1 gh/ml phenylalanine at 35 °C; 0-25 ml samples of a culture treated at time o with 93 kBq/ml. Initial cell number, I-I x i o 1 cells/ml. C. [3H]tryptophan in strain 972 growing in EMM3 + 10 /tg/ml tryptophan at 35 CC; 0-25 ml samples of a culture treated at time o with 740 kBq/ml. Initial cell number, 2-0 x 10" cells/ml. D. [aH]tryptophan in Up 4.47 growing in EMM3-glutamate + I2'S j"g/ml tryptophan at 35 °C; 02 ml samples of a culture treated at time o with 463 kBq/ml. Initial cell number, 2-9 x 10' cells/ml. ( ) Backward extrapolation of the line through the points between 30 and 50 min. It intercepts the abscissa at INT (12 min).

satisfactory measurement of synthesis in a situation where the pool is large and the incorporation rate is still increasing, since its value will depend on pool size as well as on rate of synthesis. Fig. 1 A, B, c shows that the metabolic pool is small with leucine in strain 132 and with phenylalanine and tryptophan in strain 972. Incorporation rises linearly from the origin, or nearly so. These experiments were dene with asynchronous cultures and we consider later whether there may be changes in the size of the metabolic pool through the cycle. There is a marked contrast between these curves and that for the trp 4.47 mutant in Fig. 1 D. In the mutant, the pool is large since it takes 30 min before the incorporation achieves a constant rate. As with the other experiments,

267

Protein synthesis in the yeast cell cycle 100

50

1-

-

0

1

2

3 Time(h)

4

5

6

Fig. 2. Rate of leucine incorporation in strain 132 at 32 °C in EMM2. Each point is the mean of two samples. Curve A, synchronous culture. Samples (025 ml) labelled for 14 min with 93 kBq [*H]leucine. Arrows indicate acceleration points. 10 arbitrary units (a.u.) = 4420 c.p.m. Curve B, cell numbers in synchronous culture A. 10 a.u. = 1-50x10' cells/ml. Curve C, synchronous culture. Labelling as for A. 10 a.u. = 1150 c.p.m. Curve D, cell numbers in synchronous culture C. 10 a.u. = 4-06 x io 8 cells/ml. Curve E, asynchronous control culture. Samples (1 ml) labelled for 13 min with 148 kBq [3H]leucine. 1 a.u. = 4550 c.p.m. Curve F, cell numbers in control culture (E). 1 a.u. = 152 x io* cells/ml.

there is a divergence between uptake and incorporation, though it is larger here. We consider the problem of this radioactive pool later. Synchronous cultures have certain advantages over age-fractionation techniques in the analysis of cell-cycle events, but they do suffer from the problems of perturbations caused by the selection procedure (Mitchison, 1977). It is essential therefore to examine control cultures that have been put through all the synchronizing procedure apart from the final step of selecting small cells. These cultures should show a smooth exponential rise in cell number and also in cellular parameters such as rate of incor-

J. Creanor and y. M. Mitchison

268 100

50

3

g

10

I

0

1

2

3 Time (h)

4

5

6

Fig. 3. Rate of tryptophan incorporation in strain 972 at 35 °C in EMM3 + 10 /ig/ml tryptophan. Each point in curves A and C is the mean of two samples. Curve A, synchronous culture. Samples (0-25 ml) labelled for 12 min with 307 kBq [JH]tryptophan. Arrows indicate acceleration points. 10 arbitrary units (a.u.) = 1560 c.p.m. Curve B, cell numbers in synchronous cultured. 10 a.u. = 0-92 x 10* cells/ml. Curve C, synchronous culture. Labelling as for A. 1 a.u. = 682 c.p.m. Curve D, cell numbers in synchronous culture (C). 1 a.u. = 0-35 x 10' cells/ml. Curve E, asynchronous control cultures. Samples labelled for 14 min with 370 kBq [JH]tryptophan. 1 a.u. = 4170 c.p.m. Curve F, cell numbers in control culture E. 1 a.u. = 1-85 x 10' cells/ml.

poration. As will be shown, such cultures can be obtained, but even so there may be some initial perturbation lasting for the first 45 min. We started this work some four years ago using selection from sucrose gradients (Mitchison & Vincent, 1965) but found marked perturbations in the controls with carrier-free [3H]leucine incorporation in strain 132. We then developed a method of using the elutriator rotor to select small growing cells and also to produce control cultures. This method avoided collecting, and perhaps starving, the cells on a filter and also centrifuging through a sucrose gradient. It very largely eliminated the perturbations in strain 132 but it did not do so with strain 972 in EMM3 (Creanor &

Protein synthesis in the yeast cell cycle

269

100 r-

50

3

g

10

!5

0

1

2

3 Time (h)

4

5

6

Fig. 4. Rate of tryptophan incorporation in mutant toee 2.1 at 35 °C in EMM3 + 10 /*g/ml tryptophan. Each point is the mean of two samples. Curve A, synchronous culture. Samples (0-25 ml) labelled for 12 min with 307 kBq [3H]tryptophan. Arrows indicate acceleration points. 10 arbitrary units a.u. = 1520 c.p.m. Curve B, cell numbers in synchronous culture A. 10 a.u. = 2-0 x io* cells/ml. Curve C, synchronous culture. Labelling as for A, except 266 kBq were used. 1 a.u. = 600 c.p.m. Curve D, cell numbers in synchronous culture C. 1 a.u. = 0-77 x io' cells/ml. Curve E, asynchronous control culture. Labelling as for A, except 370 kBq were used. 1 a.u. = 1870 c.p.m. Curve F, cell numbers in control culture E, 1 a.u. = 1-82 x 10* cells/ml.

Mitchison, 1979). Further experiments showed that the perrubations in 972 could be eliminated by adding leucine to the growth medium, but this expanded the metabolic pool and produced incorporation kinetics similar to Fig. 1 D. Carrier-free pHJphenylalanine gave smaller perturbations than leucine and a few experiments were done with this tracer, but the controls were usually more perturbed than the example shown later (Fig. 3E). Eventually we settled en pHJtryptophan as the best of the tracers we examined in 972. To avoid the perturbations found with carrier-free tracer, the medium contained 10/jg/ml tryptophan. At this level of exogenous tryptophan, the

270

1

J. Creanor and J. M. Mitchison 972

T

wee 1-50

35 °C T .T

A

AA

f

T

1

W

35 °C

T

I

A

A

A

T



I

T

A

1

1-

972

T

32 °C



A

972

T

i

972 PA

H

T A

L

wee 2-1

AA

ND G i

G2 lr 0

1

0-5 Cycle

i

cdc 2 M35 v 2 0 0

S

r 1

•'

T

.

l

T| A

A

TT A

. T

.

|T

35 °C 1

A A A

T .

35 °C

T

1

1

.

LA

+

32 °c

1

I

trp 4-47

W

35 °C

T

1

H

32 °C

|

.

TTT

A

,J 1

• A

1

132

35 °C

T

1

1 A

*—

25 °C

1

wee 1-302

T

0-5 Cycle

.

T

10

n

10

Fig. 5. Cell-cycle maps of the acceleration points. Each solid triangle marks the cell cycle position of an acceleration point in a synchronous culture, those above the line being in the first cycle and those below in the second cycle. The cell cycle is from cell division to division. The arrow and cross-bar show the mean and standard error. T, with tryptophan; PA, with phenylalanine; L, with leucine. The bottom left map gives the approximate position of nuclear division (ND) and the stages of the DNA cycle (G,, S and G,).

metabolic pool was not significantly expanded (Fig. 1 c). We have no cogent explanation for the differences between the strains or of the cause of the perturbations. It is possible that they may be connected with variations in the endogenous supply, since they are least marked when this supply is cut off or diminished by the presence of exogenous amino acid. It will be shown later that the endogenous supply of tryptophan is probably shut off when the exogenous concentration is io/tg/ml. Synchronous cultures and asynchronous control cultures

We have measured the rate of amino acid incorporation in some 45 synchronous cultures using different strains and size mutants, and different amino acids. All the experiments were of the same form. A sample (sometimes in duplicate) was taken from the culture, incubated at the growth temperature for 12-14 min with the labelled amino acid, and a measurement was then made of the incorporated counts. This is proportional to the rate of incorporation during that period. This procedure was repeated every 15 min for 6-6-5 h at 32-35 °C (rather more than two cell cycles for most strains). Cell numbers were also counted. Asynchronous control cultures that had been through the elutriator rotor were treated in the same way, except that they were usually followed for a shorter time. Some 70 of these controls had to be examined because of the problems of reducing perturbations. The results from a limited number of cultures are shown in Figs. 2-4, each of which has two synchronous and one control culture. Fig. 2 shows leucine pulses in

Protein synthesis in the yeast cell cycle

271

Table 1. Protein content of size mutants at 350 C in EMM-$

Strain

Mean protein (pg)/cell at division

wee 1.50

io-o

wee 2.1 wee 1.302

10-0 I2'4

972 h- (WT) Cdc 2.M3ST2O

178 26-6

Measurements by S. G. Elliot (unpublished) using the folin reaction (Lowry, Rosebrough, Farr & Randall, 1951).

strain 132 growing in EMM.2 at 32 °C. These are the strain and growth conditions used in much of the earlier work on S. pombe in this laboratory. It is clear that there are periodic changes in the rate of incorporation. At, or a little before, the mid-point of rise in cell number in the synchronous cultures there is a sharp rise in the rate of incorporation. We call this the 'acceleration point'. The rate of rise tends to fall off after the acceleration point and may reach zero before the next acceleration point. These periodic changes appear to be cell-cycle events, since their period is nearly equal to the cell cycle and they are absent in the asynchronous control. These patterns of incorporation rate are clearly not in accord with a smooth exponential increase in rate through the cycle. Nor are they in accord with the linear pattern mentioned earlier. If there was a linear pattern in single cells, incorporation rate should show the same symmetrical sigmoid curve as number increase. The rate curves are not symmetrical on either side of their point of half rise and, on average, they rise less sharply than the number curves. The interpretation of these rate curves will be discussed later. Fig. 3. shows the pattern of incorporation of tryptophan for strain 972 at 35 °C in a slightly different minimal medium, EMM3. The rate patterns are basically the same as those in Fig. 2, though it is worth pointing out that one of the synchronous cultures has a more sharply defined pattern than the other even though there is little difference in the degree of synchrony shown in the number curves. We have found this happening a number of times in synchronous cultures and cannot explain it. The incorporation of another amino acid, phenylalanine, was examined in strain 972 at 32 °C and the patterns were similar to those in Fig. 3. A conclusion from these results is that the same pattern of incorporation occurs with three different amino acids. This makes it more likely that this is the pattern of total protein synthesis and that it is not distorted by periods of slow or rapid incorporation of any one amino acid. Fig. 4 shows the pattern of incorporation of tryptophan for the small mutant wee 2.1. Here again, there are the same pattern and approximately the same position of the acceleration points.

J. Creanor and J. M. Mitchison

272

2 13 x

I 12

o

5

30 60 Time (min)

90

-

X

E

a. 6

Time(h)

Fig. 6. Kinetics of total uptake (A —A) and incorporation (O—O) with a 4-h pulse of labelled tryptophan. [3H]tryptophan (93 kBq/ml) was added at o h to trp 4.47 growing at 35 °C in EMM3-glutamate + 15 /ig/tryptophan. Initial cell number 0-9 x 10' cells/ml. Sample size, 02 ml. At 4 h, the cells were washed free of tracer and resuspended in the original medium. Inset: incorporation only after resuspension, in an identical experiment except that the initial cell number was 2 - n x io e cells/ml and the [8H]tryptophan level was 111 kBq/ml.

Figs. 2-4 illustrate patterns that were found in all the experiments on synchronous cultures. The cell-cycle maps in Fig. 5 show the timing of the acceleration points in a wider range of experiments including 972 at 25 °C, 972 at 32 °C, three other size mutants, wee 1.50, wee 1.302 and cdc 2.M35r2O, and also trp 4.47. The method of determining the rate of incorporation in trp 4.47 was somewhat different and will be described later. Except for the large size mutant cdc 2.M3sr2O, which has an early timing (074 of the cycle), all other strains have mean values for the acceleration point which lie in G1 or early S phase between o-8i and 093 of the cycle. These means do not differ significantly from the mean of 0-85 of the cycle for strain 972 with [3H]tryptophan at 35 °C (f-test, P > 0-05). There is no consistent difference between the timing of the acceleration points in the first and the second cycle. The main purpose of using the size mutants was to see whether the incorporation patterns were related to cell size. One model would be to have the acceleration point triggered by attaining a critical size. However, comparison of Fig. 5 and Table 1 shows that there is no simple relationship between cell size and the acceleration point. The timing of the acceleration points is the same in wild-type cells and in the three wee mutants that vary between 56 % and 70 % of the protein content of wild type. The large mutant cdc 2.M35r2O does have an acceleration point that is 01 of the cycle earlier than wild-type cells. But this point should have been 06 of the cycle earlier, if

Protein synthesis in the yeast cell cycle

273

Endogenous supply

I Metabolic pool

• Protein

Exchangeable pool

Outside cell

Inside cell

Fig. 7. Diagram of tryptophan pools.

it occurred when the mutant cells were the same size as wild-type cells at division (assuming exponential increase of protein). Tryptophan mutant, pools model and pools through the cycle

There are several reasons why the amounts of label incorporated after pulses might not reflect the rate of protein synthesis during the cycle. First, the metabolic pool might vary in size through the cycle. Secondly, the relative proportions of the exogenous and endogenous contributions to the pool might vary. Thirdly, there might be variable contributions to the metabolic pool from other storage pools. These could all alter the specific activity of the metabolic pool and distort the relation between measured incorporation and actual synthesis. In principle, the best way of avoiding these problems is to measure directly the specific activity of the metabolic pool, but in practice it is difficult to find a way of extracting the metabolic pool by itself. We were unable to find such a way with the tryptophan pool in S. pombe, and we therefore followed an alternative course using a tryptophan auxotroph trp 4.47 and kinetic experiments. The use of an auxotroph eliminates the endogenous supply except from a storage pool.

J. Creanor and J. M. Mitchison

274 100

r

50

10