Vol. 127, No. 1 Printed in U.S.A.
JouRNAL oF BACTERIOLOGY, July 1976, P. 109-113 Copyright C 1976 American Society for Microbiology
Transport of Glucose and Glycine in Schizosaccharomyces pombe During the Cell Cycle H. E. KUBITSCHEK* AND R. V. CLAYMEN Division of Biological and Medical Research, Argonne National Laboratory, Argonne, Illinois 60439 Received for publication 26 February 1976
Cell growth and uptake of glucose and glycine during the cell cycle were studied in synchronous cultures of Schizosaccharomyces pombe. Rates of accumulation of glucose and glycine were constant during most of the cell cycle, implying a constant rate of cell mass increase. Rates of uptake of glycine appeared to double at an average cell age of 0.9 generations.
Mitchison (9) demonstrated that total dry weight increased at a constant rate in single cells of Schizosaccharomyces pombe growing in complex media throughout most of the cell cycle (linear cell growth). Although acid-extractable pools in the same strain fluctuated in a cyclic manner during the cell cycle, increasing to a maximum value near the middle of the cycle, the sum ofthe acid-soluble and -insoluble fractions again gave the observed linear increase in total cell mass (11). These and other studies on cell mass synthesis supported the suggestion (5) that linear growth is the general growth pattern in steady state cultures of prokaryotic and eukaryotic cells. On the other hand, in later measurements by Stebbing (15, 16), on synchronized cultures of S. pombe in minimal media, both total cell dry weight and pool sizes of amino acids and nucleotides increased exponentially. To distinguish between different cell growth patterns, however, measurements of cell mass and cell dry weight must be very accurate. For cells that double in mass, for example, the maximum deviation between cells growing at a constant rate and cells growing at an exponentially increasing rate is less than 6% (3). In principle, a far more sensitive approach is to examine rates of cell mass accumulation: such rates would be constant for linear growth and would increase steadily during the cycle for exponential mass increase. Unfortunately, only rough determinates of rates of mass increase are presently possible; but because cell mass increase is due to transport of materials into the cell, it is possible to distinguish between several growth patterns on the basis of transport measurements. We have therefore measured rates of uptake of ['4C]glucose and [14C]glycine during the cell cycle of synchronous cultures of S. pombe, and our results are reported below. We first discuss, however, the criteria that
must be met by cells in synchronous culture before they can be considered in natural synchrony, i.e., representative of cells in the "steady state" growth occurring in exponentially increasing cultures. The use of forcing procedures, such as metabolic shock, may give synchronized cultures that increase in a stepwise fashion but cannot give natural synchrony. The differing results of investigators in the past, using very similar techniques with S. pombe and other strains of yeast as well, can be explained by failure to use cultures in natural synchrony. Criteria for natural synchrony. The requirement that growth of cells in synchronized cultures should represent that in steady state cultures leads to the following criteria for synchronous growth and division. (i) The mean generation time of the synchronous culture should be indistinguishable from that for steady state cultures upder the same growth conditions. (ii) Generation time distributions also should be indistinguishable from those in steady state cultures. Since present evidence supports a (truncated) normal distribution of generation rates (6), cell numbers in synchronous cultures should increase in the manner specified by this distribution during the first period of division, with essentially unaltered values for the coefficients of variation of the generation time (CVT) or the generation rate (CVr) distributions. Whereas it is generally recognized that synchrony is poor when steps in cell number increase become too broad, it has been less apparent that unduly abrupt steps also indicate a deviation from natural synchrony and result from marked metabolic alteration in the synchronized culture, as shown earlier (8). (iii) The degree of synchrony must decay with each successive division. Cultures in natural synchrony cannot maintain the same degree of 109
KUBTCHEK AND CLAYMAN 110 division synchrony in successive divisions (although some synchronization techniques give cultures that do so) because there is only weak correlation between division times of successive generations in exponential cultures (2). The actual number of reasonably sharp steps depends upon the generation time distribution of the cells and upon the rate of decay of synchrony from generation to generation, but from published generation time distributions (2) very little residual synchrony may be expected after two or three divisions for the bacteria, and after about four divisions for higher cells which have smaller values of CVr. When division is essentially uncorrelated from generation to generation, the broadening of the steps due to desynchronization increases as n 12, where n is the number of gene generations (14). (iv) Average cell mass, size, and qther cell properties must remain nearly the same at corresponding points of successive cell cycles, reflecting the invariance of these averages in steady state cultures. (v) Average cell mass and other extensive properties must double during the division cycle. MATERIALS AND METHODS
Organisn. The fission yeast Schizosaccharomyces pombe 972h-, kindly supplied by H. Gutz of the University of Texas at Dallas, was used throughout. Initially, several other strains were investigated, including the N.C.Y.C. 132 strain (10, 16), but these were unsuitable for electronic cell counting and sizing because of clumping even when they were vigorously aerated- and shaken. Medium and growth. Cells were grown at 32 C in 400-ml cultures under vigorous aeration through a glass tube extending through the bottom of the culture vessel. The cultures were grown in glucose minimal medium EMM 1 (10) with the addition of 4% lactose to prevent adverse osmotic effects during later gradient centrifugation. Because earlier results with bacteria indicated that balanced growth was not always maintained when large amounts of labeled precursors were added (4), glycine also was added to give a concentration of 6.5 mg/ml when the cultures were to be used to examine uptake of labeled glycine. Uptake of labeled precursors was less than 10% of that being supplied. Rates of uptake appeared to be constant for periods longer than 10 min, the labeling period chosen for these experiments. During growth, cell numbers and size were measured with a Coulter Counter-multichannel analyzer system (7), using a sensing aperture approximately 50 sm in diameter. The counter was calibrated by hemocytometer and plate counts. Cultures were inoculated about 18 h prior to use from refrigerated suspensions that had grown from single-colony isolates. The steady state generation time (doubling time) of cells in these cultures during exponential growth phase was 195 min. When these
J. BACTziUOL. cultures reached cell counts of 2,000 to 4,000 (approximately 0.5 x 106 to 10 cells/ml), they were used to establish synchronous cultures. Synchronous cultures were prepared by the method of Mitchison and Vincent (12), except that gradients were made from a nonutilizable sugar, lactose. The use of lactose prevented possible metabolic shock that might occur in sucrose gradients and also prevented the production of carbon dioxide bubbles which could disturb layered cells. To prepare a synchronous culture, most of the exponential phase, starting culture was filtered through a Bac-T-Flex membrane (0.45-jem pore size, 25-mm diameter, Schleicher & Schuell Inc., Keene, N.H.). The cells were then resuspended in 0.2 to 0.4 ml of fresh medium minus glucose, and 0.2 ml was layered on a linear 6 to 16% (wt/vol) lactose gradient (8 ml) and centrifuged at 500 rpm (approximately 60 x g) in a Sorvall swinging bucket rotor (7) for 1 min. This centrifugation produced a band ofcells approximately 2 cm in width, with small cells near the top and larger cells near the bottom of the band. For synchronous cultures, about 0.1 ml of the cell suspension was removed from the top of the band and diluted into 100 to 200 ml ofthe original medium. All of these operations were carried out at 32 C. Labeling. Radioactive glucose and glycine were used in the following concentrations: [U-_4C]glucose, 200 gCi/ml and 31 mg/ml; [2-14C]glycine, 10 ,sCi/ml and 6.5 mg/ml. After filtration ofabout 20 to 40 ml of the starting culture these cells were suspended in 1.2 ml of the original experimental culture, giving a final density of about 107 to 2 x 107 cells per ml. Then, 0.1 to 0.3 ml of the radioactive tracer was added. This cell concentration procedure was required to obtain sufficient uptake of label in the presence of the much larger amounts of unlabeled glucose or glycine. These dense cultures were exposed to label for 10 min, then filtered again and thoroughly washed with about 100 ml of fresh medium to remove any excess radioactivity, resuspended in 0.2 ml of fresh medium, banded, and fractionated by size (and age) as described above. Samples (0.1 to 0.2 ml) of cells of increasing size were removed at equal intervals down the gradient, as described earlier (7), and each was diluted to 2 ml in ice cold medium to prevent further metabolism. Three 0.5-ml aliquots were obtained from each sample. One was diluted into 7 ml offresh medium, and the cells were counted and sized with the Coulter Counter-multichannel analyzer system. Each of the other two was mixed with 10 ml of XDC scintillation fluid (1) and counted in a Beckman liquid scintillation counter (CPM-100). The possibility that transport might have been affected by the dense concentrations of cells during application of the radioactive label was tested by measuring glucose uptake in samples from exponentially growing cultures concentrated to different densities. The results show that glucose uptake was proportional to cell density and, therefore, that the concentration procedure had no observable effect on uptake (Fig. 1). Determination of cell ages. In cell samples obtained from gradients, average cell ages were determined from average cell volumes, as described ear-
TRANSPORT IN S. POMBE DURING THE CELL CYCLE
VOL. 127, 1976
111
first division corresponds to a value of the coefficient of variation of generation rate, CVr, of 0.125. The curve describing the second step has a larger spread, reflecting the increasing loss of synchronization expected in successive steps in cell number increase. Sinclair and Ross (14) pointed out that the increase in the coeflicient of variation is proportional to n"2, where n is the number of divisions, in cultures in which division is uncorrelated from generation to generation. Correspondingly, we have fitted the second step with the increase expected for CVr = 0.125 (2) 12 = 0.18. It is apparent from the fit
a.)
700
600[
Cell Density
FIG. 1. [14C]glucose uptake in S. pombe as a function of cell density. Uptake is given in counts per minute for 50-pJ samples of cell culture after exposure to label for 10 min at 30 C. Cell densities are given in cells per milliliter. The different symbols refer to data from two independent experiments. Abbreviations: CPM, counts per minute; OD.i, optical density at 600 nm.
lier (7). Briefly, mean cell volumes at birth, Vb, and at termination, 2Vb, were first determined from the average volume V of cells in the parent culture in its exponential growth phase, Vh = V ln 2. The corresponding ages at birth and at termination are labeled 0 and 1, and intermediate ages have a linear correspondence with intermediate sample volumes.
RESULTS Division synchrony. Figures 2 and 3 are representative of growth and division in our synchronous cultures. The cell counts in Fig. 2 indicate two division steps, in which cell numbers increased almost by a factor of two. The 10% deviation from a twofold increase may be due to the presence of nongrowing cells, as described by Mitchison (10). The data show that the cells in the growing population divided twice. The vertical arrows indicate the midpoint positions of logarithmic increase for each of these steps. The interval between the two arrows is the generation time of the synchronous culture, 3.2 h, in excellent agreement with the generation time of the parent culture (3.25 h). The data are fitted with theoretical curves for numbers increase. The first step is that expected on the basis of a normal distribution of generation rates, the so-called reciprocal normal distribution of generation times (6). The spread in increase in cell numbers during the
500s
z 400 0
300[
200
2
3
4
7
8
TIME, hours
FIG. 2. Increase in cell numbers for a synchronous culture of S. pombe. Values are those obtained with the Coulter counter. The vertical arrows locate the midpoint of the logarithmic increase in cell numbers during the first and second divisions. Vertical bar (#) indicates displacement of one standard error above and below the observed mean.
/
10 w
9-
8-~~~~~~~~
-j
-j W
7
w0 w 40
TIME, hours
FIG. 3. Increase in average cell volume (relative units) with time. Same culture as that for Fig. 1. The data are averages computed from cell volume distributions obtained with the counter analyzer system.
112
KUBITSCHEK AND CLAYMAN
J BACTERIOL .
of the data to the calculated increases that 008 these values of CV, are also in good agreement with the data. 0.07 The relative values of average cell volumes for the same culture are shown in Fig. 3, as 006 Hdetermined from graphs of the cell volume disz tributions obtained with the counter-analyzer 0 system. Cell volumes increased uniformly for 0UJ 05F-;i EL.approximately 3 h for each of the first two w 004 FJ,-I cycles, and each division reduced cell volumes C) // by about a factor of two. Cell volumes increased 0 03Fagain at about the same rate during the second generation and approached this value during 002kthe third generation, indicative of cyclic reproducibility. 0.01 These results are consistent with the criteria for natural size change presented above and 10 1.5 20 b5 provide evidence that our techniques did not V, /Vb significantly disturb cell growth and division. FIG. 5. Relative rates off uptake of ['4C]glycine Uptake of radioisotopes. For uptake meas- during the cell cycle in exponentially growing culurements, exponentially growing cultures were tures of S. pombe. Same units and symbols as in exposed either to ['4C]glucose or to ['4C]glycine Fig. 4. for 10 min. Figures 4 and 5 show radioisotope uptake (in counts per minute divided by the (Materials and Methods). The horizontal cell observed counter values) in cell samples of dif- line represents the average value of all data ferent mean volumes (Vs) taken from gradients points for glucose (Fig. 4) and of all but the last formed after exposure. The average age of the four points for glycine (Fig. 5). Within the ercells in each sample was determined from the rors shown, the data are in agreement with ratio of its mean cell volume to that at birth, Vb constant rates of uptake of glucose or of glycine over most of the cell cycle. With glycine, values clearly increased near the end of the cycle. The ,/ increase in the final value for glucose is not statistically significant. The dashed lines in Fig. 4 and 5 show the 0 Io5kexpected radioactivities of the cell samples if uptake were proportional to cell age or volume and if the data were fitted at the birth volume, Vb. A similar proportionality would be expected 0.01 0if cell volume or mass increased exponentially during the cycle, as has occasionally been suggested for bacterial growth and was apparently L) observed in Stebbing's (15) experiments with S. pombe. Our data for S. pombe do not support 0005[that uptake pattern. The rate of uptake of glycine increased slightly at about three-quarters of the way through the cycle. If, however, this increase was due to a doubling of uptake when cells I05 10 1.5 20 reached a transition point within the cycle, as 7V, / Vb suggested earlier (4, 5), then we can provide a FIG. 4. Relative rates of uptake of ['C]glucose better estimate of the time of this doubling. This during the cell cycle in exponentially growing cul- estimate was made by constructing graphs of the tures ofS. pombe. Average radioactivities per cell are increase in the fraction of the population older plotted in observed units of counts per minute (CPM) than some specific age (i.e., 0.8 generations) at divided by the Coulter counter value as a function of sample mean cell volume, Vs. Vb is the estimated various times during the cell cycle, assuming a normal distribution of cell generation rates with mean cell volume at birth. The dashed line represents the increase expected if uptake rates were propor- CVr = 0.125. The observed increases in radiotional to cell volume. The vertical bar (2) indicates activity for the points at the right side of Fig. 5 displacement of one standard error above and below were then compared with predictions from these the observed mean. graphs at the same age, sample mean cell vol-
.
JouRNAL oF BACTERIOLOGY, July 1976, P. 109-113 Copyright C 1976 American Society for Microbiology
Transport of Glucose and Glycine in Schizosaccharomyces pombe During the Cell Cycle H. E. KUBITSCHEK* AND R. V. CLAYMEN Division of Biological and Medical Research, Argonne National Laboratory, Argonne, Illinois 60439 Received for publication 26 February 1976
Cell growth and uptake of glucose and glycine during the cell cycle were studied in synchronous cultures of Schizosaccharomyces pombe. Rates of accumulation of glucose and glycine were constant during most of the cell cycle, implying a constant rate of cell mass increase. Rates of uptake of glycine appeared to double at an average cell age of 0.9 generations.
Mitchison (9) demonstrated that total dry weight increased at a constant rate in single cells of Schizosaccharomyces pombe growing in complex media throughout most of the cell cycle (linear cell growth). Although acid-extractable pools in the same strain fluctuated in a cyclic manner during the cell cycle, increasing to a maximum value near the middle of the cycle, the sum ofthe acid-soluble and -insoluble fractions again gave the observed linear increase in total cell mass (11). These and other studies on cell mass synthesis supported the suggestion (5) that linear growth is the general growth pattern in steady state cultures of prokaryotic and eukaryotic cells. On the other hand, in later measurements by Stebbing (15, 16), on synchronized cultures of S. pombe in minimal media, both total cell dry weight and pool sizes of amino acids and nucleotides increased exponentially. To distinguish between different cell growth patterns, however, measurements of cell mass and cell dry weight must be very accurate. For cells that double in mass, for example, the maximum deviation between cells growing at a constant rate and cells growing at an exponentially increasing rate is less than 6% (3). In principle, a far more sensitive approach is to examine rates of cell mass accumulation: such rates would be constant for linear growth and would increase steadily during the cycle for exponential mass increase. Unfortunately, only rough determinates of rates of mass increase are presently possible; but because cell mass increase is due to transport of materials into the cell, it is possible to distinguish between several growth patterns on the basis of transport measurements. We have therefore measured rates of uptake of ['4C]glucose and [14C]glycine during the cell cycle of synchronous cultures of S. pombe, and our results are reported below. We first discuss, however, the criteria that
must be met by cells in synchronous culture before they can be considered in natural synchrony, i.e., representative of cells in the "steady state" growth occurring in exponentially increasing cultures. The use of forcing procedures, such as metabolic shock, may give synchronized cultures that increase in a stepwise fashion but cannot give natural synchrony. The differing results of investigators in the past, using very similar techniques with S. pombe and other strains of yeast as well, can be explained by failure to use cultures in natural synchrony. Criteria for natural synchrony. The requirement that growth of cells in synchronized cultures should represent that in steady state cultures leads to the following criteria for synchronous growth and division. (i) The mean generation time of the synchronous culture should be indistinguishable from that for steady state cultures upder the same growth conditions. (ii) Generation time distributions also should be indistinguishable from those in steady state cultures. Since present evidence supports a (truncated) normal distribution of generation rates (6), cell numbers in synchronous cultures should increase in the manner specified by this distribution during the first period of division, with essentially unaltered values for the coefficients of variation of the generation time (CVT) or the generation rate (CVr) distributions. Whereas it is generally recognized that synchrony is poor when steps in cell number increase become too broad, it has been less apparent that unduly abrupt steps also indicate a deviation from natural synchrony and result from marked metabolic alteration in the synchronized culture, as shown earlier (8). (iii) The degree of synchrony must decay with each successive division. Cultures in natural synchrony cannot maintain the same degree of 109
KUBTCHEK AND CLAYMAN 110 division synchrony in successive divisions (although some synchronization techniques give cultures that do so) because there is only weak correlation between division times of successive generations in exponential cultures (2). The actual number of reasonably sharp steps depends upon the generation time distribution of the cells and upon the rate of decay of synchrony from generation to generation, but from published generation time distributions (2) very little residual synchrony may be expected after two or three divisions for the bacteria, and after about four divisions for higher cells which have smaller values of CVr. When division is essentially uncorrelated from generation to generation, the broadening of the steps due to desynchronization increases as n 12, where n is the number of gene generations (14). (iv) Average cell mass, size, and qther cell properties must remain nearly the same at corresponding points of successive cell cycles, reflecting the invariance of these averages in steady state cultures. (v) Average cell mass and other extensive properties must double during the division cycle. MATERIALS AND METHODS
Organisn. The fission yeast Schizosaccharomyces pombe 972h-, kindly supplied by H. Gutz of the University of Texas at Dallas, was used throughout. Initially, several other strains were investigated, including the N.C.Y.C. 132 strain (10, 16), but these were unsuitable for electronic cell counting and sizing because of clumping even when they were vigorously aerated- and shaken. Medium and growth. Cells were grown at 32 C in 400-ml cultures under vigorous aeration through a glass tube extending through the bottom of the culture vessel. The cultures were grown in glucose minimal medium EMM 1 (10) with the addition of 4% lactose to prevent adverse osmotic effects during later gradient centrifugation. Because earlier results with bacteria indicated that balanced growth was not always maintained when large amounts of labeled precursors were added (4), glycine also was added to give a concentration of 6.5 mg/ml when the cultures were to be used to examine uptake of labeled glycine. Uptake of labeled precursors was less than 10% of that being supplied. Rates of uptake appeared to be constant for periods longer than 10 min, the labeling period chosen for these experiments. During growth, cell numbers and size were measured with a Coulter Counter-multichannel analyzer system (7), using a sensing aperture approximately 50 sm in diameter. The counter was calibrated by hemocytometer and plate counts. Cultures were inoculated about 18 h prior to use from refrigerated suspensions that had grown from single-colony isolates. The steady state generation time (doubling time) of cells in these cultures during exponential growth phase was 195 min. When these
J. BACTziUOL. cultures reached cell counts of 2,000 to 4,000 (approximately 0.5 x 106 to 10 cells/ml), they were used to establish synchronous cultures. Synchronous cultures were prepared by the method of Mitchison and Vincent (12), except that gradients were made from a nonutilizable sugar, lactose. The use of lactose prevented possible metabolic shock that might occur in sucrose gradients and also prevented the production of carbon dioxide bubbles which could disturb layered cells. To prepare a synchronous culture, most of the exponential phase, starting culture was filtered through a Bac-T-Flex membrane (0.45-jem pore size, 25-mm diameter, Schleicher & Schuell Inc., Keene, N.H.). The cells were then resuspended in 0.2 to 0.4 ml of fresh medium minus glucose, and 0.2 ml was layered on a linear 6 to 16% (wt/vol) lactose gradient (8 ml) and centrifuged at 500 rpm (approximately 60 x g) in a Sorvall swinging bucket rotor (7) for 1 min. This centrifugation produced a band ofcells approximately 2 cm in width, with small cells near the top and larger cells near the bottom of the band. For synchronous cultures, about 0.1 ml of the cell suspension was removed from the top of the band and diluted into 100 to 200 ml ofthe original medium. All of these operations were carried out at 32 C. Labeling. Radioactive glucose and glycine were used in the following concentrations: [U-_4C]glucose, 200 gCi/ml and 31 mg/ml; [2-14C]glycine, 10 ,sCi/ml and 6.5 mg/ml. After filtration ofabout 20 to 40 ml of the starting culture these cells were suspended in 1.2 ml of the original experimental culture, giving a final density of about 107 to 2 x 107 cells per ml. Then, 0.1 to 0.3 ml of the radioactive tracer was added. This cell concentration procedure was required to obtain sufficient uptake of label in the presence of the much larger amounts of unlabeled glucose or glycine. These dense cultures were exposed to label for 10 min, then filtered again and thoroughly washed with about 100 ml of fresh medium to remove any excess radioactivity, resuspended in 0.2 ml of fresh medium, banded, and fractionated by size (and age) as described above. Samples (0.1 to 0.2 ml) of cells of increasing size were removed at equal intervals down the gradient, as described earlier (7), and each was diluted to 2 ml in ice cold medium to prevent further metabolism. Three 0.5-ml aliquots were obtained from each sample. One was diluted into 7 ml offresh medium, and the cells were counted and sized with the Coulter Counter-multichannel analyzer system. Each of the other two was mixed with 10 ml of XDC scintillation fluid (1) and counted in a Beckman liquid scintillation counter (CPM-100). The possibility that transport might have been affected by the dense concentrations of cells during application of the radioactive label was tested by measuring glucose uptake in samples from exponentially growing cultures concentrated to different densities. The results show that glucose uptake was proportional to cell density and, therefore, that the concentration procedure had no observable effect on uptake (Fig. 1). Determination of cell ages. In cell samples obtained from gradients, average cell ages were determined from average cell volumes, as described ear-
TRANSPORT IN S. POMBE DURING THE CELL CYCLE
VOL. 127, 1976
111
first division corresponds to a value of the coefficient of variation of generation rate, CVr, of 0.125. The curve describing the second step has a larger spread, reflecting the increasing loss of synchronization expected in successive steps in cell number increase. Sinclair and Ross (14) pointed out that the increase in the coeflicient of variation is proportional to n"2, where n is the number of divisions, in cultures in which division is uncorrelated from generation to generation. Correspondingly, we have fitted the second step with the increase expected for CVr = 0.125 (2) 12 = 0.18. It is apparent from the fit
a.)
700
600[
Cell Density
FIG. 1. [14C]glucose uptake in S. pombe as a function of cell density. Uptake is given in counts per minute for 50-pJ samples of cell culture after exposure to label for 10 min at 30 C. Cell densities are given in cells per milliliter. The different symbols refer to data from two independent experiments. Abbreviations: CPM, counts per minute; OD.i, optical density at 600 nm.
lier (7). Briefly, mean cell volumes at birth, Vb, and at termination, 2Vb, were first determined from the average volume V of cells in the parent culture in its exponential growth phase, Vh = V ln 2. The corresponding ages at birth and at termination are labeled 0 and 1, and intermediate ages have a linear correspondence with intermediate sample volumes.
RESULTS Division synchrony. Figures 2 and 3 are representative of growth and division in our synchronous cultures. The cell counts in Fig. 2 indicate two division steps, in which cell numbers increased almost by a factor of two. The 10% deviation from a twofold increase may be due to the presence of nongrowing cells, as described by Mitchison (10). The data show that the cells in the growing population divided twice. The vertical arrows indicate the midpoint positions of logarithmic increase for each of these steps. The interval between the two arrows is the generation time of the synchronous culture, 3.2 h, in excellent agreement with the generation time of the parent culture (3.25 h). The data are fitted with theoretical curves for numbers increase. The first step is that expected on the basis of a normal distribution of generation rates, the so-called reciprocal normal distribution of generation times (6). The spread in increase in cell numbers during the
500s
z 400 0
300[
200
2
3
4
7
8
TIME, hours
FIG. 2. Increase in cell numbers for a synchronous culture of S. pombe. Values are those obtained with the Coulter counter. The vertical arrows locate the midpoint of the logarithmic increase in cell numbers during the first and second divisions. Vertical bar (#) indicates displacement of one standard error above and below the observed mean.
/
10 w
9-
8-~~~~~~~~
-j
-j W
7
w0 w 40
TIME, hours
FIG. 3. Increase in average cell volume (relative units) with time. Same culture as that for Fig. 1. The data are averages computed from cell volume distributions obtained with the counter analyzer system.
112
KUBITSCHEK AND CLAYMAN
J BACTERIOL .
of the data to the calculated increases that 008 these values of CV, are also in good agreement with the data. 0.07 The relative values of average cell volumes for the same culture are shown in Fig. 3, as 006 Hdetermined from graphs of the cell volume disz tributions obtained with the counter-analyzer 0 system. Cell volumes increased uniformly for 0UJ 05F-;i EL.approximately 3 h for each of the first two w 004 FJ,-I cycles, and each division reduced cell volumes C) // by about a factor of two. Cell volumes increased 0 03Fagain at about the same rate during the second generation and approached this value during 002kthe third generation, indicative of cyclic reproducibility. 0.01 These results are consistent with the criteria for natural size change presented above and 10 1.5 20 b5 provide evidence that our techniques did not V, /Vb significantly disturb cell growth and division. FIG. 5. Relative rates off uptake of ['4C]glycine Uptake of radioisotopes. For uptake meas- during the cell cycle in exponentially growing culurements, exponentially growing cultures were tures of S. pombe. Same units and symbols as in exposed either to ['4C]glucose or to ['4C]glycine Fig. 4. for 10 min. Figures 4 and 5 show radioisotope uptake (in counts per minute divided by the (Materials and Methods). The horizontal cell observed counter values) in cell samples of dif- line represents the average value of all data ferent mean volumes (Vs) taken from gradients points for glucose (Fig. 4) and of all but the last formed after exposure. The average age of the four points for glycine (Fig. 5). Within the ercells in each sample was determined from the rors shown, the data are in agreement with ratio of its mean cell volume to that at birth, Vb constant rates of uptake of glucose or of glycine over most of the cell cycle. With glycine, values clearly increased near the end of the cycle. The ,/ increase in the final value for glucose is not statistically significant. The dashed lines in Fig. 4 and 5 show the 0 Io5kexpected radioactivities of the cell samples if uptake were proportional to cell age or volume and if the data were fitted at the birth volume, Vb. A similar proportionality would be expected 0.01 0if cell volume or mass increased exponentially during the cycle, as has occasionally been suggested for bacterial growth and was apparently L) observed in Stebbing's (15) experiments with S. pombe. Our data for S. pombe do not support 0005[that uptake pattern. The rate of uptake of glycine increased slightly at about three-quarters of the way through the cycle. If, however, this increase was due to a doubling of uptake when cells I05 10 1.5 20 reached a transition point within the cycle, as 7V, / Vb suggested earlier (4, 5), then we can provide a FIG. 4. Relative rates of uptake of ['C]glucose better estimate of the time of this doubling. This during the cell cycle in exponentially growing cul- estimate was made by constructing graphs of the tures ofS. pombe. Average radioactivities per cell are increase in the fraction of the population older plotted in observed units of counts per minute (CPM) than some specific age (i.e., 0.8 generations) at divided by the Coulter counter value as a function of sample mean cell volume, Vs. Vb is the estimated various times during the cell cycle, assuming a normal distribution of cell generation rates with mean cell volume at birth. The dashed line represents the increase expected if uptake rates were propor- CVr = 0.125. The observed increases in radiotional to cell volume. The vertical bar (2) indicates activity for the points at the right side of Fig. 5 displacement of one standard error above and below were then compared with predictions from these the observed mean. graphs at the same age, sample mean cell vol-
.
