Download Impact of Cell Cycle Progression on the G1 Period in Saccharomyces cerevisiae and more Study notes Cell Biology in PDF only on Docsity!
Proc. Nati. Acad. Sci. USA Vol. 78, No. 5, pp. 3030-3033, May 1981 Cell Biology
Nature of the G1 phase of the yeast Saccharomyces cerevisiae
(cell cycle/hydroxyurea)
R. A. SINGER AND G. C. JOHNSTON
Faculty of Medicine, Dalhousie University, Halifax, Nova Scotia B3H 4H7 Canada
Communicated by David Marshall Prescott, Frebruary 3, 1981
ABSTRACT Under conditions that protract the S phase for Saccharomyces cerevisiae without affecting steady-state rates of cell growth or proliferation, there were striking decreases in the length of the GI period. These decreases were localized in the period between mitosis and the start event that initiates a new cell cycle. We conclude that this major fraction of the GI period has no functional role in the DNA-division sequence of cell cycle events.
The cell cycle involves the periodic replication of DNA and seg-
regation of replicated DNA with cellular constituents to prog-
eny cells. Mitchison (1) has described^ the^ eukaryotic^ cell^ cycle
by a model composed of two independent cycles, the DNA-
division cycle and the growth cycle. The^ DNA-division^ cycle
consists of events concerned with the replication and segrega-
tion of DNA. The growth cycle, on the other^ hand, is a^ more
loosely defined concept, normally used to describe processes
that provide the bulk of the new^ cytoplasm. The^ performance
of these two cycles is independent (1), but cells have mecha-
nisms so that repeated DNA-division cycles do^ not^ outstrip the
process of growth. This coordination generally occurs in the GC
period ofthe cell cycle, after mitosis (M phase) is completed but
before DNA replication (S phase) is initiated (for review, see
ref. 2).
Much is known about the regulation of the cell cycle of the
yeast Saccharomyces cerevisiae (3). For example, one event in
the DNA-division cycle requires growth to a critical size for its
completion (2-4). This point of regulation has been referred to
as "start" and lies temporally near the end of the G1 period (2).
The G1 period of the cell cycle also is (for other eukaryotes as
well) the most sensitive to growth conditions because the com-
bined length of the cell cycle periods S +^ G2 +^ M generally
remains constant under varied growth conditions (5-9). As nu-
tritional conditions are changed so that growth rates are in-
creased, the^ time^ in^ GC is^ correspondingly decreased.
It is generally held that, throughout the eukaryotic G1 pe-
riod, a^ sequence ofcell^ cycle-specific^ events must occur^ in^ prep-
aration for DNA synthesis (1, 10, 11). Recently this view has
been questioned. Both Cooper (12) and Liskay et al. (13) have
theorized that^ G1 is^ present in^ eukaryotic cells,^ not^ because
events specific to G1 are taking place but because events specific
to growth have not yet occurred in sufficient quantity. In this
view, G1 is present in cells solely to allow accumulation of suf-
ficient "division potential" to initiate a new S phase.
This alternative view of the nature of the G1 period is based
upon two types of observations. First, Cooper has developed
his hypothesis by extending the model for bacterial cell division.
Bacteria have the ability to initiate a new round ofchromosome
replication (a new DNA-division cycle) when sufficient mass has
accumulated, even prior to the completion of previously initi-
ated rounds of replication. Thus, cell cyles in bacteria can over-
lap (14). In this view of the cell cycle, an interval separating the
initiation of a new DNA-division cycle from the preceding cell
division event is simply a reflection of an insufficient accumu-
lation of mass during the previous cell cyle. Second, Liskay and
coworkers have described Chinese hamster cells that have no
discernible GC period (13, 15, 16). This observation has led
them to suggest that Gl need not be an integral part of the eu-
karyotic cell cycle.
Whether the GC period is (i) necessary for the execution of
a sequence of cell cycle-specific events or (ii) simply part of a
larger period for growth, altering the growth rate would be ex-
pected to alter the length ofGC in much the same way. In con-
trast, the possibility of decreasing the rate of performance of
DNA-division-cycle events, without affecting overall growth
rate [a procedure used for cell cycle studies in bacteria (17)],
leads to quite different predictions for Gl. For example, if^ Gl
is an obligatory part ofthe cell cycle, reflecting the need for GI-
specific events, then slowing the rate of progression of some
other aspect ofthe DNA-division cycle without affecting overall
growth rate should not markedly alter the GC period. If, how-
ever, GC simply represents a^ period of^ ongoing growth, then
(because growth and cell division are independent processes)
slowing the rate of progression of some aspect of^ the^ DNA-di-
vision cycle should allow greater than normal growth during the
protracted performance of these events and should result^ in^ a
shorter than normal GC period. Because such experiments may
be conducted under identical growth (nutritional) conditions,
slowing the DNA-division cycle should clearly distinguish be-
tween the two views of the GC period previously described.
To this end, with S. cerevisiae we have used procedures that
affect the rate of progression through S phase. Under identical
nutritional conditions and without affecting the overall rate of
cell number increase or growth, we find that^ slowing the^ rate
of progress through the DNA-division sequence causes a strik-
ing decrease in the time spent in^ the GC period. Thus, in^ this
communication we provide evidence that in an organism that
normally exhibits a significant GC period, most of^ this^ GC is^ not
an obligatory part of the cell cycle.
MATERIALS AND METHODS
Strains and Media. The haploid strain GR2 (a ural his6)
(ATCC 42564) has been described (18). Strain 13052 that carries
the temperature-sensitive allele cdc8-3 (19) and strain 428 that
carries cdc13-1 (20) are both derivatives of strain A364A and
were provided by L. H. Hartwell. Strain GR150 is a cdc8 strain
constructed to be isogenic to strain GR2. Cells^ were^ grown at
room temperature in the complex medium YM1 (21). Where
appropriate, hydroxyurea was added to 1.5 mg/ml from a stock
solution of 20 mg/ml in YM 1. The mating pheromone a^ factor
was prepared by the method of Bficking-Throm et al. (22).
Analysis of^ Cellular^ Parameters.^ Cell^ number^ was^ deter-
mined with an electronic particle counter (Coulter Electronics,
The publication costs ofthis article were defrayed in part by page charge payment. This article must therefore be hereby marked "advertise- ment" in accordance with 18 U. S. C.^ §1734 solely to^ indicate^ this fact.
Proc. Natl. Acad. Sci. USA 78 (1981) 3031
Hialeah, FL). Prior^ to^ counting, cells were either sonicated
briefly (23) or^ treated^ with^ the^ enzyme glusulase to^ separate cells that (^) had (^) undergone cytokinesis (24). Execution (^) point was
determined by increase in cell number as described (8).
RESULTS
Three procedures were employed to slow the progress of cells
through the^ DNA-division^ cycle. Each^ of^ these^ procedures in-
volved affecting the rate of progress through S phase.
Effect of Hydroxyurea on Cell Cycle Progression. The first
procedure employed the^ DNA^ synthesis inhibitor hydroxyurea.
A high concentration of hydroxyurea causes arrest of cells in S
phase (25). At^ lower^ concentrations of hydroxyurea, we found
rather different effects on the progression of cells through the
cell cycle. Hydroxyurea-treated cells, after an initial lag period,
grew at^ the^ same^ 2.7-^ to^ 2.8-hr generation time^ as^ untreated
cells. The rate ofgrowth (cell-mass increase) must be unaffected
by these conditions because^ the^ normal^ rate^ of cell^ number^ in-
crease of these hydroxyurea-treated cells could continue in-
definitely (data not^ shown). We^ continually subcultured^ cells for several (^) days in (^) the (^) presence ofhydroxyurea before (^) perform- ing subsequent experiments.
A low concentration of hydroxyurea was used to retard but
not to block progress through S phase. Because of the difficulty
in measuring the length of S phase in yeast, no direct mea-
surements were made. (^) However, to (^) ensure that S (^) phase was
protracted by hydroxyurea treatment, we^ examined^ the^ effect
of (^) hydroxyurea on the (^) timing of (^) completion of cell (^) cycle steps.
Certain mutations, the cdc mutations, affect cell cycle progress,
and may be used to define events that function at specific points
in the cell (^) cycle. The last (^) point in the cell (^) cycle affected (^) by a
cdc mutation is referred to as the execution point of that mu-
tation (^) [execution point is (^) operationally defined as that (^) point in the cell (^) cycle past which a cell is no (^) longer sensitive to non- permissive conditions^ (26)]. We^ examined the^ execution^ point
of the cdc8 mutation, which affects DNA synthesis (27), and
found that (^) hydroxyurea treatment moved (^) completion of this
step from^ 0.37^ to^ 0.56^ of the^ yeast cell^ cycle. Another^ cdc^ mu-
tation, cdcM3, defines^ a^ step normally completed at^ the end of S (^) phase (^) (20). In similar (^) fashion, execution (^) point determination showed that (^) hydroxyurea treatment moved (^) completion of the
cdcM3 step from^ 0.35^ to^ 0.50^ of the cell cycle. Thus, without
affecting the^ rate^ of cell^ number increase, hydroxyurea treat- ment (^) lengthens the (^) period (^) encompassing S (^) phase. Upon hydroxyurea treatment, the^ proportion of^ unbudded
cells fell from 45% to 15% (Fig. 1). Because the proportion of
unbudded cells (^) usually approximates the (^) proportion of cells in GI period (^) (7, 9), these results show (^) that, without (^) affecting the
rate of cell number increase, hydroxyurea treatment shortens
the unbudded period of the cell cycle and, thus, may shorten
the GC (^) period. Effect of (^) Hydroxyurea on (^) Length ofGI. To demonstrate that
the decreased proportion of cells without buds also indicated
a shorter (^) length of time in the (^) GC (^) period prior to the initiation ofnew (^) cell (^) cycles, we examined the effect of (^) hydroxyurea treat-
ment on the timing of another DNA-division-sequence event,
start. Start is the earliest known gene-mediated step in the
DNA-division sequence (3). Before commitment to a new DNA-
division sequence by the completion of start, cells of the a mat-
ing type are^ still^ sensitive^ to^ the^ mating pheromone a^ factor
(28). We^ determined^ the^ point ofloss^ of^ a-factor^ sensitivity (the execution (^) point) for both (^) hydroxyurea-treated and untreated
populations ofcells^ of^ strain^ GR2.^ In^ the untreated^ population,
these cells executed the a-factor sensitive step (start) at ap-
proximately 0.30^ of the cell^ cycle, measured from^ cell^ separa-
tion. In contrast, cells growing in the presence of hydroxyurea
0 U 4 128
x (^) Time, hr -4 2.0-
~1.
(^0 4 8 ) Time, hr
FIG. (^) 1. Effect of hydroxyurea on cellular parameters. Cells of strain GR2 were grown in the complex medium YM1. At time 0, (^) hy- droxyurea was added to 1.5 mg/ml. Cell concentrations and the (^) per- centage unbudded were determined as described. *, With (^) hydroxyurea; o, without hydroxyurea.
executed start at 0.0 of the yeast cell cycle. Thus, not only is
the unbudded period shortened in the presence ofhydroxyurea,
but cells are also able to execute the first (^) step in the DNA-di-
vision sequence immediately after completion of cell separation.
Normally yeast cell number determinations are made after
brief sonication to separate cells that have completed both the
cytokinesis and cell separation steps but have failed to com-
pletely detach. Thus, the usual procedure with mild sonication
really only allows the measurement of execution point relative
to cell separation. Conceivably during hydroxyurea treatment,
the period between cytokinesis and cell separation may expand
in a manner compensatory to the decrease in the unbudded
period. To examine this possibility, we measured execution
points relative to the cytokinesis process. This can be accom-
plished by using the enzyme mixture glusulase to separate cells
by cell wall digestion (24). When glusulase treatment was em-
ployed instead of sonication, we found that here too the exe-
cution point of strain GR2 for a-factor sensitivity was moved by
hydroxyurea treatment to 0.05 of the cell cycle.
The GC period may be thought to begin not with cytokinesis
but after mitosis. Consequently, the shortened GC period be-
tween cytokinesis and start brought about by hydroxyurea treat-
ment may result simply from a large compensatory expansion
of the normally short interval between mitosis and cytokinesis.
To test this hypothesis, we determined the proportion of GR
cells that had two nuclei and, thus, had completed mitosis but
had not yet undergone cell separation, which follows cytoki-
nesis. After fixation and sonication to separate cells, staining
with Giemsa (29) showed that 6.6% of >1500 hydroxyurea-
treated cells and 5.7% of >1000 untreated cells (^) were in the in-
terval between mitosis and cell separation. This finding of a
constant interval between mitosis and cell separation, coupled
with the demonstration that cytokinesis immediately precedes
start, shows that hydroxyurea treatment does in fact shorten the
period between mitosis and start.
Effect of Hydroxyurea on Bud Initiation of Mother and
Daughter Cells.^ Because of the budding mode of division of S.
cerevisiae, a daughter cell is usually smaller than a mother cell
Proc. Natl. Acad. Sci. USA 78 (^) (1981) 3033
to determine the extent of cell cycle overlap.
The finding that start can be executed at 0.0 of the yeast cell
cycle is supported by our observation that for both mother and
daughter cells bud initiation occurs at roughly the same time
under these steady-state growth conditions. This result was an-
ticipated by earlier work of Yamata and Ito (34), who used sim-
ilar hydroxyurea concentrations in an attempt to synchronize
the budding mode of mother and daughter cell pairs. Although
they only looked at the first few divisions after hydroxyurea ad-
dition, they too found that, in the presence ofhydroxyurea, bud
initiation was simultaneous on both mother and daughter cells.
In the absence ofgenetic evidence for any prestart steps, we
suggest that a prestart GC period is merely a manifestation of
a block at the start event of the DNA-division sequence caused
by insufficient growth. This view suggests that the GI period
need not be considered an extended sequence ofcell cycle spe-
cffic events. There need be no interval at all between the end
of mitosis and cytokinesis and the execution of start. Although
temporally the start event may be completed near the end of
the Gl time interval, functionally (with respect to the cell cycle)
start may initiate the pre-S phase. We further agree with the
suggestion by Cooper (12) that use of the term cycle may not
be appropriate. The events that make up cell division are better
described as a sequence as shown in Fig. 2, with a beginning
(completion of start) and an end (completion of mitosis). We
would suggest that the bulk of the GI period we observe as a
temporal entity is in fact without functional significance with
respect to the DNA-division sequence.
We wish to thank D. P. (^) Bedard, R. (^) M. Liskay, and J. R. Pringle for helpful discussions. This work was supported by grants from the Med- ical Research Council (^) of Canada and National Cancer (^) Institute of Canada.
- Mitchison, J. M. (1971) The (^) Biology ofthe Cell Cycle (Cambridge Univ. Press, Cambridge, England).
- Johnston, G. C., Pringle, J. R. & (^) Hartwell, L. H. (1977) Exp. Cell Res. 105, 79-98.
- Hartwell, L. H. (1974) Bacteriol. (^) Rev. 38, 164-198.
- Johnston, G. C., (^) Ehrhardt, C. (^) W., Lorincz, A. & Carter, B. L. A. (1979)J. Bacteriol. 237, 1-5.
- Unger, M. W. & (^) Hartwell, L. H. (1976) Proc. Natl. Acad. Sci. USA 73, 1664-1668.
6. Hartwell, L. H. & Unger, M. W. (1977)J. Cell Biol. 75, 422-435.
7. Slater, M. L., Sharrow, S. 0. & Gart, J. J. (1974) Proc. Natl.
Acad. Sci. USA 74, 3850-3854.
8. Jagadish, M. N. & Carter, B. L. A. (1977) Nature (London) 269,
9. Johnston, G. C., Singer, R. A., Sharrow, S. 0. & Slater, M. L.
(1980)J. Gen.^ Microbiol.^ 118, 479-484.
10. Prescott, D. M. (1976) Reproduction of Eukaryotic Cells (Aca-
demic, New^ York).
11. Pardee, A. B., Dubrow, R., Hamlin, J. L. & Kletzien, R. F.
(1978) Annu. Rev.^ Biochem.^ 47, 715-750.
12. Cooper, S. (1979) Nature (London) 280, 17-19.
13. Liskay, R. M., Leonard, K. E. & Prescott, D. M. (1979) Somatic
Cell Genet. 5, 615-623.
14. Cooper, S. & Helmstetter, C. E. (1968) J. Mol. Biol. 31,
519-540.
15. Liskay, R. M. (1978) Exp. Cell Res. 114, 69-77.
16. Liskay, R. M. & Prescott, D. M. (1978) Proc. Natl. Acad. Sci.
USA 75, 2873-2877.
17. Pritchard, R. H. & Zaritsky, A. (1970) Nature (London) 226,
126-131.
18. Johnston, G. C. & Singer, R. A. (1978) Cell 14, 951-958.
19. Hartwell, L. H. (1973)J. Bacteriol. 115, 966-979.
- (^) Culotti, J. & (^) Hartwell, L. H. (^) (1971) Exp. Cell (^) Res. 67, 389401.
- (^) Hartwell, L. H. (^) (1967)J. Bacteriol. (^) 93, 1662-1670.
22. Bficking-Throm, E., Duntze, W., Hartwell, L. H. & Manney, T.
R. (1973) Exp. Cell Res. 76, 99-110.
23. Hartwell, L. H. (1970)J. Bacteriol. 104, 1280-1285.
24. Pringle, J. R. & Mor, J. R. (1975) Methods Cell Biol. 11, 131-168.
25. Slater, M. L. (1973)J. Bacteriol. 113, 263-270.
- (^) Hartwell, L. (^) H., Culotti, J. & Reid, B. (1970) Proc. Natl. Acad.
Sci. USA 66, 352-359.
27. Hartwell, L. H. (1971)J. Mol. Biol. 59, 183-194.
- (^) Hereford, L. M. & Hartwell, L. H. (^) (1974) J. Mol. Biol. (^) 84,
- (^) Robinow, C. F. (^) (1975) Methods Cell Biol. (^) 11, 1-22.
- (^) Jaeniche, L., Scholz, K. & (^) Donike, M. (^) (1970) Eur. (^) J. Biochem.
31. Hartwell, L. H., Culotti, J., Pringle, J. R. & Reid, B. J. (1974)
Science 183, 46-51.
32. Liskay, R. M., Kornfeld, B., Fullerton, P. & Evans, R. (1980)J.
Cell. Physiol. 104, 461467.
- (^) Pringle, J. R. (^) (1978) J. Cell. (^) Physiol. 95, (^) 393-406.
- (^) Yamada, K. & (^) Ito, M. (^) (1979) Plant Cell Physiol. (^) 20, 1471-1479.
