342
Biotechnol. Prog. 1995, 1 I, 342-347
Tracking of Individual Cell Cohorts in Asynchronous Saccharomyces cereuisiae Populations Danilo Porrot and Friedrich Srienc* Department of Chemical Engineering and Materials Science and Institute for Advanced Studies in Biological Process Technology, University of Minnesota, MinneapolidSt. Paul, Minnesota 55108
A novel flow cytometric procedure has been developed with the aim to obtain the growth properties of individual Saccharomyces cereuisiae cells in asynchronous culture. The method is based on labeling of the cell surface with FITC-conjugated concanavalin A and detection of the single-cell fluorescence with flow cytometry after cell exposure to growth conditions. Because the formation of new cell wall material in budded cells is restricted to the bud tip, exposure of the stained cells to growth conditions results in three cell types: (i) stained cells, (ii)partially stained cells, and (iii)unstained cells. Analysis of the staining pattern over time permits the determination of the specific growth rate of the cell population, the length of the budded cell cycle phase, and the growth pattern during the cell cycle of newly formed, partially stained daughter cells. The procedure has been tested with yeast cell populations growing at different rates. The data suggest a n exponential increase in the size of individual cells during the cell cycle, as reflected by the forward angle light scattering (FALS) signals. It has been found that the apparent single-cell specific cell size growth rates, determined by FALS intensity, are significantly lower than the specific growth rates of the overall population. This could indicate that the tracking of a cohort of cells is significantly perturbed by a distribution of staining levels of daughter cells a t cell division and that FALS may not be a good indicator of the cell size.
Introduction Saccharomyces cerevisiae is a microorganism of significant biotechnological interest. It is used in wellestablished bioprocesses for the synthesis of products such as single-cell proteins, ethanol, and vitamins (Fiechter et al., 1981; Postma et al., 1989). More recently, it has found applications in bioremediation processes and in the production of specific chemicals, pharmaceutical agents, and vaccines (Porro et al., 1992; Dequin and Barre, 1994; Ramonos et al., 1992; Buckholz, 1993). In addition, it is an important organism for the study of cell cycle control in eukaryotic cells (Sherlock and Rosamond, 1993; Lew and Reed, 1992; McKinney and Cross, 1992; Nasmyth, 1990). Optimization and control of a bioprocess require the monitoring and manipulation of environmentalvariables to maintain the culture at conditions most favorable for the synthesis of the product. The performance of a biotechnological process depends to a large extent on the physiological state of the growing biomass and on how this physiological state is distributed within the cell population. Little information is available on these distributions since conventional sensors typically measure only population average quantities and their direct determination is difficult. Flow cytometry allows one to measure physical and/ or chemical bioparameters in single cells at rapid rate, such that the distribution of these properties in the whole population also can be obtained in a very short period of time (Shapiro, 1988). This analytical technique has been
* Author to whom correspondence should be addressed: Telephone (612) 624-9776; Fax (612) 625-1700. ' Present address: Dipartimento di Fisiologia e Biochimica Generali, Sez. Biochimica Comparata, Univ. di Milano, Via Celoria 26, 20133 Milan, Italy.
applied to growing yeast cells in several different applications. These range from the characterization of the single-cell growth dynamics (Agar and Bailey, 1981; Srienc q d Dien, 1992; Alberghina and Porro, 19931, to the analysQ of the kinetics of heterologous protein production (Eitpman and Srienc, 1991), to the selection of cells with specfic properties (Skowronek et al., 1990), to studies on the metabolic state of cell populations (Porro et al., 1994), to give a few examples. On the basis of the work of May and Mitchison (19861, we have developed a novel flow cytometry procedure that should permit the determination of single-cell growth properties of S. cerevisiae. The procedure is based on labeling the cell surface with a lectin such as Concanavalid ( C o d ) conjugated to a fluorescent marker such as fluorescein isothiocyanate (FITC). We have established experimental staining conditions that did not perturb cell growth &r the resuspension of stained cells in growth medium. Because cell growth in budded cells is concentrated in the bud (Ballou, 1988; Tkacz et al., 1971; Chung et al., 19651, newly synthesized cell wall is unstained, while the older cell wall components retain their initial fluorescence (Tkacz et al., 1971). It has been anticipated, therefore, that partially stained cells should be indicative of a certain cell cycle age. The analysis of the staining patterns has been used to directly provide information on the dynamic properties of individual cells and of the cell population. These include the specific growth rate of the overall population, the timing of the main phases of the cell cycle, and the growth behavior of individual cohorts of cells during the cell cycle. This methodology has been applied to asynchronous cell populations growing at different specific growth rates t o test the developed procedure and to find out more information on the growth behavior of individual cells.
8756-7938/95/3011-0342$09.00/00 1995 American Chemical Society and American Institute of Chemical Engineers
Biotechnol. Prog., 1995,Vol. 11, No. 3
M a t e r i a l s and Methods Yeast Strain and Media. The homozygous diploid S . cerevisiae strain D 6034 (Mat a, ura3-52, lys 2-801, met, his 3, ade2-101, reg 1-501) (Eitzman and Srienc, 1991) was used in this study. Cells were grown in Erlenmeyer flasks by shaking at 30 "C in 0.67% (wh) Yeast Nitrogen Base (YNB, Difco, Detroit, MI) minimal medium with the appropriate supplements (50 pg/mL). Carbon sources were glucose (GLU), galactose (GAL),or raffinose (RAF); they were used at concentration of 2% (w/v). Staining Conditions. Cells growing exponentially under balanced growth conditions [balanced growth typically was observed in a cell concentration range between 2 x lo6 and (8-10) x lo6 cells/mLl were collected by centrifugation (5 min, 5000 rpm) and resuspended in precooled, fresh YNB-based medium, containing 120 mg/mL conjugated Cod-FITC (approximately 3.6 mol of FITC per mole of lectin; Sigma, St. Louis, MO) at a cell concentration of 2 x IO8 celldml. ARer 7 min of staining, cells were recovered by rapid centrifugation and resuspended in identical fresh medium. All of the operations were carried out at 4 "C in the dark. Cell Number, Cell Volume, Budding Index, and TB Determination. Cells were counted aRer sonication with an electronic Coulter Counter. Cell volume distributions were determined with a Channelyzer C-1000 interfaced with a microcomputer (IBM PC) for data acquisition. The percentage of budded cells or budding index (BI) was calculated after the microscopic examination of at least 600 cells. The specific growth rate was obtained by plotting cell number against time on a semilogarithmic scale, while the duration of the budded phase or time of budding (TB), comprising the S, G2, M, and G1* cell cycle phases, was determined from the following equation (Lord and Wheals, 1980):
where FBIis the fraction of budded cells in the population and Tis the duplication time of the whole cell population. Flow Cytometric Analysis. Forward angle light scatter (FALSI, indicating cell size, and FITC fluorescence (FLU, indicating the amount of surface fluorescence, signal intensities were acquired from a Cytofluorograf 50 H (Ortho Instruments, Westwood, MA) and a FACStarP'us(Becton & Dickinson, Palo Alto, CAI equipped with an argon ion laser (excitation wavelength 488 nm, laser power 200 mW). The sample flow rate during analysis was approximately500 cellds. Typically 50 000 cells were analyzed per sample. Deconvolution Analysis. A least-squares algorithm has been applied to estimate the weights of (i),completely unstained daughter cells, (ii)partially stained daughters cells, and (iii) the rest of the population consisting of completely stained cells. This algorithm assumes that the experimental distributions consist of the s u m of weighted distributions of populations i-iii and that the distributions of the pure populations i and iii can be obtained experimentally from separate measurements. The time course of the completely unstained cell fraction F,, has been described by the following equation:
where Fu,(t) is the fraction of completely unstained cells at time t.
343
Results and Discussion Cell Growth after Surface Staining. Yeast cell wall is constructed almost entirely of two classes of polysaccharides: polymers of mannose covalently linked to peptides (mannoproteins) and polymers of glucose (glucans). A third sugar polymer of N-acetylglucosamine (chitin) is present only in minor amounts (Ballou, 1982; Cabib et al., 1982). Concanavalin A, the lectin obtained from Jack beans (Canaualiaensiformis)(Goldstein et al., 19651, specifically binds the mannoproteins of the yeast cell wall (Tkacz et al., 1971). To obtain information on the growth kinetics of S . cereuisiue cells, we have developed a new procedure based on the labeling of the cell surface of growing cells with the lectin C o d conjugated to fluorescein (Cod-FITC). Yeast cells were cultured to the exponential phase, harvested, stained, and resuspended in the growth medium described in the Material and Methods section. We have also carried out initial staining experiments with cell wall-specific antibodies (Douglas and Ballou, 1980) using a secondary FITC-conjugated antibody for fluorescence detection. However, it was more convenient to work with direct fluorescence using the conjugated lectin. In order to use the surface staining as a permanent label for observation of the growth properties of a cohort of cells, the following two requirements must be met: (i) the labeling procedure should not perturb the growth behavior of cells and (ii) the surface label should be retained by the cells over the subsequent growth period. We have established staining conditions that did not detectably perturb cell growth, as suggested by the representative data shown in Figures 1 and 2. A cell culture was grown to exponential phase. At time 0 (see Figure lA), a fraction of the culture was stained, resuspended in fresh growth medium, and compared to the unstained control culture. The growth rate in cell number, as well as the fraction of budded cells (budding index), was very similar in both cases. The specific growth rates were 0.233 and 0.231 h-l for the control culture and the stained culture, respectively. Furthermore, the volume distributions of the two cell populations also appeared to be identical over the period of the experiment (Figure 1B). We therefore conclude that the staining procedure causes only minor perturbations in the growth behavior of the cells, if any at all. Figure 2 shows the growth curve, average fluorescence, and total fluorescence per unit of culture volume of a cell population cultured and stained as described in Figure 1. It can be seen that the increase in cell number concentration and decrease in the average fluorescence per cell follow the same first-order rate law, with very similar rate constants of 0.234 and 0.24 h-l, respectively. The coefficient of correlation in both cases was better than 0.99. The dye attached to the cells appears to be only diluted by the new growth. This can also be seen in the time course of the total fluorescence per unit of culture volume (average fluorescence multiplied by cell number concentration), which remains constant over several hours of cell growth, indicating that the attached stain is not lost into the growth medium. Length of Budded Phase and Specific Growth Rate. A critical point in the cell cycle regulation of cell wall synthesis is the initiation of a new bud (Sherlock and Rosamond, 1993). Localization of cell wall synthesis during growth of the bud has been demonstrated with fluorescent dyes and autoradiography (Tkacz et al., 1971; Chung et al., 1965). It is believed that the cell wall of the growing bud is not derived from the material of the mother cell wall, but is synthesized de novo (Ballou, 1982,
Biotechnol. Prog., 1995, Vol. 11, No. 3
344
A
A
I Budded
60
I
50
G1
IS
+
G2
+
Phase1 M
+
G1*
I
40 10’
30
STAINED 20
- 2 - 1 0
1
3
2
5
4
TIME (HOURS)
FLUORESCENCE
CELL V O L U M E Figure 1. 1. (A) Effect of staining on cell growth. D603-i yeast cells were grown on galactose medium until the exponential phase; one fraction was harvested, stained with ConA-FITC, and then allowed to grow in fresh galactose culture medium (see Materials and Methods): untreated cells (0,0);stained cells (U, 0).(B)Cell volume histograms of unstained (0) and stained ( 0 )cells. Cell volumes have been determined with a Coulter Counter Channalyzer on culture samples withdrawn at time T = 2 h of part A. n
B
28
0 W
A
21
c
0
J
Ih
I 31
14
v
-I LL
-e
n
LL
‘3; J J
W
0
23
3
7
1
-
1
0
1
2
3
4
5
TIME (HOURS) Figure 2. Cell number and single-cellfluorescence of a stained cell population. Exponentially growing cells have been stained and resuspended at time T = 0 into fresh galactose medium: cell number/mL (0);average single-cell fluorescence (AV. FL.) (U); total fluorescence per unit of culture volume (TOT. FL.) (0).
1988; Cabib et aE.,1982; Tkacz et al., 1971; Chung et al., 1965). The staining procedure, in combination with flow cytometry analysis, enables rapid detection and quantification of the newborn daughter cells as they evolve in
0
1
2
3
4
5
6
TIME (HOURS) Figure 3. (A) Single-cell fluorescence frequency distribution functions of the stained cell population growing on galactose medium. The evolution of partially stained and unstained cells with time is clearly visible: T = 0 h (1); T = 1 h (2);T = 2 h (3); T = 3 h (4).The inset shows the cell cycle phases of a growing yeast cell. (B) Fractions of unstained newborn daughter cells growing o n glucose (O), galactose (U), and raffinose (+) after staining and resuapension in growth medium. The TB values, as determined with eq 1, were 1.36, 1.81,and 3.39 h, respectively. The solid lines represent the predictions by eq 2 using the experimentally determined specific growth rates of the populations.
the growing culture (Figure 3A). At time of staining (2’ = 0), all of the cells are completely stained. Since they are at different cell cycle positions, they generate, with progressing time, daughter cells with a gradually decreasing degree of surface stain. The partially stained cells represent newborn daughter cells originating from cells stained in the budded phase, which includes the S, G2,M, and G1* cell cycle phases. Completely unstained cells represent the newborn daughters originating from
Biotechnol. Prog., 1995, Vol. 11, No. 3
cells stained while they were in the unbudded phase (Gl) (see inset, Figure 3A). The time of appearance of this last subpopulation of daughter cells depends on the length of the S+G2+M+G1* cell cycle phase. The experiment allows, therefore, direct determination of the length of the budded phase of the cell cycle. A deconvolution algorithm has been applied to estimate the weight of the subpopulation of unstained cells in the growing population. If it is assumed that the length of the budded cell cycle phase is the same for all cells (Lord and Wheals, 1981; Thompson and Wheals, 19801, the time course of the appearance of unstained cells can be described by eq 2. Figure 3B shows that the experimental data are closely predicted by this relationship. Therefore, it is possible to use this relationship together with the experimental data to estimate the specific growth rate of the entire cell population. Growth of Daughter Cells. To determine the growth properties of individual cells, it is normally necessary to observe and record individual cells in the microscope. Alternatively, one can synchronize the cell population and observe the synchronized cell population over time. We have used the described staining procedure to directly determine the cell growth pattern of individual cells in asynchronous cell populations. Cell size has been estimated on the basis of the forward angle light scattering (FALS) signal that has been measured for cells of the same fluorescence intensity, which presumably indicates the same cell cycle age. Since the newborn, partially stained daughter cells retain the initial fluorescence (see Figure 2), the selection of a cohort of cells of the same fluorescence (Le., the same age) in the growing population permits observation of this cell population during growth. In particular, evaluation of the FALS signal over time directly reflects the growth in cell size during the cell cycle. An example for the selection of such a cohort is shown in Figure 4 A,B. The time course of the mean FALS signal is shown in Figure 5 for cohorts of cells selected from populations growing at different specific growth rates. If it is assumed that the area of the measured light scattering signals at low angles (FALS)reasonably reflects the size of the cells, one can directly estimate from these data the growth pattern during the cell cycle. In every case, the cell cultures were sampled and analyzed for a time as long as 80% of the overall duplication time of the culture. One can see that the data suggest exponential growth for individual cells during the cell cycle since an exponential rate law fits the data reasonably well. However, the single-cell specific growth rates obtained were significantly lower than the specific growth rates found for the overall populations (Table 1). It is interesting to note that at slow growth rates the differences are much less pronounced than at high growth rates. Such information is generally inaccessible except by inference from other data using mathematical models (Alberghina and Porro, 1993; Srienc et al., 1992) or after cell synchronization procedures (Woldringh et al., 1993). Additional analysis was carried out to further verify the validity of the interpretation of the data obtained. In particular, if the identity of a cell subpopulation indeed remains the same during the experiment, one should anticipate that the time course of the cell number in a given gate is represented by a decreasing function of time as the overall cell population grows in cell number. However, this is not the case as shown in Figure 6. After an initial transient, the fraction of cells in the gate used for the time course analysis appears to remain constant. This indicates that the identity of cells selected in the gate is changingin time. In fact, if the gate were to select cells of the same age (i.e., born at the same time), then
345
A
J
U
1b1
s
FLUORESCENCE
B
Figure 4. (A) Cytograms of C o d - F I T C fluorescence (log scale) versus forward angle light scattering (FALS; linear scale) taken at different times after the resuspension of stained yeast cells in fresh galactose medium: T = 0.5 h (1);T = 1.5 h (2); T = 2.5 h (3); T = 3.5 h (4). The gate R1 shows the region of newborn partially stained daughter cell populations of the same age (time elapsed between staining procedure and cell division). (B) Forward angle light scattering (FALSIhistograms obtained from a partially stained subpopulation of cells falling into a gated region (see part A) at different times after resuspension in fresh medium: 0.5 h (1);1.0 h (2); 1.5 h (3);2.0 h ( 4 ) ;2.5 h (5); 3.0 h (6).
Biotechnol. Prog., 1995, Vol. 11, No. 3
346
A
GAL v)
RAF
v,
-I
a
L J
A
20'. 0
6
. ' " .
1
TIME (HOURS) Figure 5. Average forward angle light scattering (FALS)signal of a selected population of newborn, partially stained daughter cells as a function of time after resuspension. Cells were grown on different carbon sources: glucose (A);galactose (0);raffinose
. ' . ' . ' . "
. . . ' .
2
3
4
5
TIME (HOURS)
B k
(MI. Table 1. Specific Growth Rates for Cell Populations and for Single Cohorts of Cells on DifPerent Carbon Sources specific growth rate (h-l) carbon source overall population" cohort of cellsb
0.315 0.230
glucose galactose rafinose
0.183 (r = 0.99) 0.140 ( r = 0.99) 0.111 ( r = 0.99)
0.125
Specific growth rates of the overall population have been determined from the relative cell nuinber increase. Specific growth rates of single-cell size as determined by FALS have been determined on cohorts of cells of the same age; data in parentheses represent the coefficient of correlation for the estimation of the growth rate. h
v
2
0
7
I
I
W
J W
0
R
0
o
1
2
3
4
TIME (HOURS) Figure 6. Fraction of cells selected in a gate (0)and coefficient of variation (CV; M) of the relative FALS signal intensity for cells grown on galactose medium after resuspension at T = 0 h (see also Figure 5 ) . the fraction of cells should decrease over time with a rate constant given by the specific growth rate of the population. This evidently is not the case, indicating that during the experiment cells of different ages fall into the selected gate. There could be two possible ways to explain this observation: (i) cells that divide have a relatively wide distribution and/or (ii) cells lose some fluorescence and fall into the gate of lower fluorescence. The fact that the overall fluorescence of the population is constant (see Figure 2) excludes the last argument. However, the first argument cannot explain the data obtained either. If cells fall at different times into the selected gate due to their variability in the time of the budded phase, one should expect, over the time period of the experiment, a pronounced increase in the size variation of the selected cohort of cells. The coefficient
U
J 111
I*,
I
FLUORESCENCE
Figure 7. (A) Average forward angle light scattering (FALS) intensity of gated subpopulations of cells as a function of time after staining and resuspension in fresh medium. The cells are grown on galactose as carbon source. The numbers refer to the gates delineating cells with decreasing staining,levels shown in part B. (B)Cytogram of the cell population obtained at time T = 3 h. The gates show the selection of regions to identify subpopulations of cells with decreasing staining levels.
of variation of cell size for a cohort of cells growing on GAL-YNB medium remains, however, nearly constant for more than 3 h (Figure 6). Figure 7 demonstrates the cell size increase for different cohorts of newborn daughter cells selected from the same cell population. Single cohorts of cells have been selected according to the gates shown in Figure 7B. In all cases the size appears to increase exponentially with time; however, the specific growth rates for cohorts 4 and 5 are clearly lower than those for the other three cell populations. This is not surprising since the gates delineating cohorts 4 and 5 fall into the region of completely unstained cells. Thus, the cell size values are derived from a mixture of daughter cells of different ages that are generated during the time. In contrast, cohorts 1 , 2 , and 3 appear to have identical specificgrowth rates, indicating that these subpopulations of cells would represent true cohorts of cells of the same age. However, in this experiment the specific growth rates for individual cohorts of cells also were significantly lower than the specific growth rates determined for the entire cell population (Table 2). The discrepancy in specific growth rates of individual cells and the overall population appears to be partially due to the fact that FALS is not a good indicator of cell size. In fact, recent doublelabeling experiments, where cell size is determined from the cellular protein content, indicate that the determined specific growth rate of the single-cell size appears to be much closer to the growth rate of the overall cell population (D.Porro, manuscript in preparation).
Biotechnol. Prog., 1995, Vol. 11, No. 3
347
Table 2. Specific Rate of Forward Angle Light Scattering Increase in Daughters Cells Growing on Galactose Medium with an Overall Specific Growth Rate of u = 0.230 h-’
gate no. (see Figure 7A)
specific growth rate (h-l) 0.141 0.143 0.137 0.082 0.059
coefficient of correlation 0.99 0.99 0.98 0.98 0.97
Conclusions We have shown in this work that multiparameter flow cytometry in combination with a surface labeling technique can be used to obtain growth data on individual cells. It was anticipated that the surface label identifies the age of newborn daughter cells. Cohorts of cells with the same surface label or age can then be tracked as cell growth proceeds, and additional single-cell properties can be determined as a function of time. Such information normally is not directly accessible in an asynchronously growing cell population. It requires either application of an entirely different approach to determine single-cell growth rates (Bugeja et al., 1985; Srienc and Dien, 1992) or cell synchrony. The establishment of synchronous cell populations usually requires a much more elaborate experimental effort. Furthermore, such synchronization techniques may perturb the state of the population much more severely. Although the surface label appears to be correlated with the cell age, its utility tracking individual cohorts of daughter cells in an asynchronous cell population appears to be somewhat limited, presumably because of a distribution of staining levels of daughter cells at cell division. The contributions of these effects can most likely be accounted for with an appropriate mathematical model that needs to be developed for this purpose. Because the present procedure is very rapid, it could be useful for cell cycle research, for the isolation of growth mutants, and the study of the physiological response of S. cerevisiae to nutritional shifts.
Acknowledgment The authors thank D. Block and K. Gordon for carrying out preliminary experiments, C. E. Ballou for donating antibodies, and C. Hatzis for offering his expert assistance in analyzing the data. This work has been partially supported by the National Science Foundation (BCS-9100385). Additional support was provided by the Italian Research Council (CNR), F’rogetto Finalizzato Biotecnologie e Biostrumentazioni, subproject no. 3, to Bianca Maria Ranzi and by CIB (Consorzio Italiano Biotecnologie). We are also grateful to the “Gruppo Italiano Citometria” (GIC), which awarded this study with the 1994 GIC Prize.
Literature Cited Agar, D. W.; Bailey, J. E. Continuous cultivation of fission yeast: analysis of single-cell protein synthesis kinetics. Biotechnol. Bioeng. 1981,23,2315-2331. Alberghina, L.; Porro, D. Quantitative flow cytometry: analysis of protein distributions in budding yeast. A mini review. Yeast 1993,9,815-823. Ballou, C. E. Yeast cell wall and cell surface. In The molecular biology of the yeast Saccharomyces cerevisiae. Metabolism and gene expression; Strathern, J. N., Jones, E. W., Broach, J. R., Eds.; Cold Spring Harbor Laboratory Press: Cold Spring Harbor, NY, 1982;pp 335-360. Ballou, C. E. Organization of the Saccharomyces cerevisiae cell wall. In Self-Assembly Architecture;Varner, J. E., Ed.; Alan R. Liss, Inc.: New York, 1988;pp 105-117.
Buckholz, R. G. Yeast systems for expression of heterologous gene products. Curr. Opin. Biotechnol. 1993,4,538-542. Bugeja, V. C.; Saunders, P. T.; Bazin, M. J. Estimating the mode of growth of individual microbial cells from cell volume distributions. Biosystems 1985,18,47-63. Cabib, E.; Roberts, R.; Bowers, B. Synthesis of the yeast cell wall and its regulation. Annu. Rev. Biochem. 1982,51,763793. Chung, K.L.; Hawirko, R. Z.; Isaac, P. K. Cell wall replication in Saccharomyces cerevisiae. Can. J. Microbiol. 1965, 1 1 , 953-957. Dequin, S.;Barre, P. Mixed lactic acid-alchoholicfermentation by Saccharomyces cerevisiae expressing the Lactobacillus casei L(+)-LDH. BiolTechnology 1994,12,173-177. Douglas, R. H.; Ballou, C. E. Isolation of Kluyveromyces lactis mannoprotein mutants by fluorescence-activated cell sorting. J. Biol. Chem. 1980,255,5979-5985. Eitzman, P.; Srienc, F. Dynamics of activation of a galactoseinducible promoter in Saccharomyces cerevisiae. J. Biotechnol. 1991,21,63-82. Fiechter, A.; Fuhrmann, G. F.; Kappeli, 0. Regulation of glucose metabolism in growing yeast cells.Adv. Microb. Physwl. 1981, 22,2031-2043. Goldstein, I. J.; So, L. L. Protein-carbohydrate interaction 111. Agar gel-diffusion studies on the interaction of concanavalin A, a lectin isolated from jack bean, with polysaccharides. Arch. Bwchem. Biophys. 1965,111,407-414. Lew, D. J.; Reed, S. I. A proliferation of cyclins. Trends Cell Biol. 1992,2, 77-80. Lord, P. G.; Wheals, A. E. Asymmetrical division of Saccharomyces cerevisiae. J. Bacteriol. 1980,142,808-818. Lord, P. G.; Wheals, A. E. Variability in individual cell cycles of Saccharomyces cerevisiae. J. Cell Sci. 1981,50,361-376. May, J. W.; Mitchison, J. M. Length growth in fission yeast cell measured by two novel techniques. Nature 1986,322,752754. McKinney, J.; Cross, F. A switch-hitter at the Start of the cell cycle. Curr. Biol. 1992,2,421-423. Nasmyth, K.FAR-reaching discoveries about the regulation of start. Cell 1990,62,1117-1120. Porro, D.; Martegani, E.; Ranzi, B. M.; Alberghina, L. Lactose/ whey utilization and ethanol production by transformed Saccharomyces cerevisiae cells. Biotechnol. Bioeng. 1992,39, 799-805. Porro, D.; Smeraldi, C.; Martegani, E.; Ranzi, B. M.; Alberghina, L. Flow-cytometric determination of the respiratory activity in growing Saccharomyces cerevisiae populations. Biotechnol. Prog. 1994,10,193-197. Postma, E.; Verduyn, C.; Alexander, W.; Van Dijken, J. P. Enzymic analysis of the Crabtree effect in glucose-limited chemostat cultures of Saccharomyces cerevisiae. Appl. Environ. Microbiol. 1989,55,468-477. Ramonos, M. A.; Scorer, C. A.; Clare, J. J. Foreign gene expression in yeast: a review. Yeast 1992,8, 423-488. Shapiro, H. M.Practical flow cytometry;Alan R. Liss, Inc.: New York, 1988. Sherlock, G.; Rosamond, J. Starting to cycle: G1 controls regulating cell division in budding yeast. J . Gen. Microbiol. 1993,139,2531-2541. Skowronek, P.; Krummeck, G.; Haferkamp, 0.;Rodel, G. Flow cytometry as a tool to discriminate respiratory-deficient yeast cells. Curr. Genet. 1990,18,265-267. Srienc, F.; Dien, S. Kinetics of the cell cycle of Saccharomyces cerevisiae. Ann. N.Y. Acad. Sci. 1992,665,59-71. Thompson, P. W.; Wheals, A. E. Asymmetrical division of Saccharomyces cerevisiae in glucose-limited chemostat culture. J. Gen. Microbiol. 1980,121,401-409. Tkacz, J. S.;Cybulska, E. B.; Lampen, J. 0. Specific staining of wall mannan in yeast cells with fluorescein-conjugated concanavalin A. J. Bacteriol. 1971,105, 1-5. Woldringh, C. L.; Huls, P. G.; Vischer, N. 0. E. Volume growth of daughter and parent cells during the cell cycle of Saccharomyces cerevisiu dalpha as determined by image cytometry. J. Bacteriol. 1993,175,3174-3181. Accepted January 27,1995.* BP950007Y @
Abstract published in Advance ACS Abstracts, April 1, 1995.
