If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
The ultimate stem cell, the oocyte, is frequently very large. For example, Drosophila and Xenopus oocytes are ∼105 times larger than normal somatic cells. Importantly, once the large oocytes are fertilized, the resulting embryonic cells proliferate rapidly. Moreover, these divisions occur in the absence of cell growth and are not governed by normal cell cycle controls. Observations suggest that mitogens and cell growth signals modulate proliferation by upregulating G1-phase cyclins, which in turn promote cell division. Like embryonic cells, the proliferation of cancer cells is largely independent of mitogens and growth factors. This occurs, in part, because many proteins that are known to modulate G1-phase cyclin activity are frequently mutated in cancer cells. Interestingly, we have found that both the expression and the activity of G1-phase cyclins is modulated by growth rate and cell size in yeast. These and other data suggest that proliferative capacity correlates with cell size. Thus, a major goal of our laboratory is to use yeast to investigate the relationship between proliferation rate, G1-phase cyclins, growth rate, and cell size. The elucidation of this relationship will help clarify the role of cell size in promoting proliferation in both normal and cancer cells.
The most obvious differences between different animals are differences in cell size, but for some reason the zoologists have paid singularly little attention to them.
J. B. S. Haldane, On Being the Right Size 1927
On Being the Right Size
Size is a fundamental and useful descriptive quality of all organisms. Remarkably, organisms display an almost incomprehensible range of sizes. For example, the largest organism, the Blue Whale, is over 19 billion times larger than the smallest single cell plankton. Comparing the largest and smallest multicellular organisms still reveals an amazing spectrum of size. The smallest marine rotifer has less than 100 cells as compared with the nearly 100 quadrillion cells of a Blue Whale. Given the amazing diversity of organism size, it is striking that cells themselves are quite uniform in size. Most animal cells are 10–20 μm in diameter and rarely vary more than 2-fold outside of this size range (
). The relative constancy of cell size within diverse organisms suggests that the mechanism of cell size regulation is conserved. But despite these observations, very little is known about the biological mechanisms that control the size of cells or organisms.
The remarkable homogeneity of cell size observed in populations of cells is achieved by coordination of cell growth with division. This occurs because external stimuli, such as nutrients, growth factors, and mitogens, stimulate cell growth and division equivalently. Although often used inter-changeably, it is important to stress that cell growth is not synonymous with proliferation. Proliferation refers to increases in cell numbers whereas growth refers to increases in cell mass (discussed in
). Recently, the mechanisms that control the regulation of cell size and act to coordinate cell growth with proliferation have become an area of intense research (reviewed in
). Prior to the 1950s research on cell size control was virtually non-existent. But a Pub-Med literature search reveals that in the past 5 y there has been an average of nearly 300 “cell size” manuscripts per year as compared with an average of less than two per year 40 y ago Figure 1. This trend is likely to continue as more and more important insights are being made into the genetic, biochemical, and molecular mechanisms that ensure cell size homeostasis (reviewed in
Figure 1The rapid pace of cell size research. A Pub-Med literature search showed that in recent years, the number of publications on cell size continues to increase at a very rapid pace.
). By analyzing temperature-sensitive mutants in the yeast Saccharomyces cerevisiae, they found that the inactivation of some genes essential for proliferation resulted in cells that arrested in specific phases of the cell cycle. Because these mutants blocked progression through the cell division cycle (cdc mutants), the genes encoding these mutants were called CDC genes (
). Careful analysis of these mutants revealed three fundamental details of the basic architecture of the cell cycle.
First, it was discovered that the cell cycle is composed of a series of inter-dependent steps that are initiated at the transition point between G1 and S phase. Because of the relationship between this transition and cell cycle progression, this point was named Start in yeast Figure 2 (
). Like CDC28, CDC2 is required for progression past the “restriction point.”
Figure 2Start is equivalent to the restriction point. In Saccharomyces cerevisiae, cell division is clearly asymmetric resulting in a larger mother cell (m) and a smaller daughter cell (d). Observations indicate that smaller daughters require cell growth to a “critical cell size” in order to progress past Start. Start is a point just prior to G1/S-phase boundary equivalent to the restriction point in mammalian cells. Interestingly, whereas by default mother cells are larger than the “critical cell size,” they still have a cell growth requirement for Start. It is not known if this requirement is also governed by cell size.
) Figure 2. A subset of G1-phase cdc mutants blocks cell growth. In this manner, cells smaller than the required minimum cell size arrest before Start (
). To date, little is known about the biochemical mechanisms responsible for linking cell growth and cell size to proliferation.
Third, although it was shown that proliferation was dependent upon cell growth, it was found that the converse is not true. Most cdc mutants that arrest cells in G1 phase continue to grow in mass at near normal rates (
). The manner in which these cdc mutants prevent proliferation despite normal cell growth is not well understood. Thus, a major aim of the cell cycle field is the dissection of the molecular mechanism that links cell growth with proliferation.
The genetic study of mutations that disrupt normal cell size control in yeast has been extremely useful in elucidating the mechanisms that coordinate cell growth with proliferation (reviewed in
). The cloning and analysis of this gene revealed that it encoded a truncated and stabilized G1-phase cyclin, Cln3, that upregulates the activity of the Cdc28 (
). These observations have demonstrated that Cln–Cdc28 is integral to cell size homeostasis.
Interestingly, much of what is known about cell size homeostasis in higher eukaryotes has come from studying cells where growth is not coordinated with proliferation Figure 3. Physiologically, this is a relatively rare event. For instance, oocytes, neurons, and adipocytes can grow without dividing leading to very large cells Figure 3b (discussed in
). For example, Drosophila and Xenopus oocytes are ∼105 times larger than normal somatic cells. Once the large oocytes are fertilized, the resulting embryonic cells proliferate rapidly. Moreover, these divisions occur in the absence of cell growth and are not governed by normal cell cycle controls Figure 3c. These observations have led to the theory that cell size may modulate the proliferative capacity of cells. Specifically, it has been suggested that commitment to proliferation is dependent upon the attainment of a minimum “critical cell size” (discussed in
). It is this phenomenon that allows extremely large oocytes to return to the normal size of somatic cells Figure 3c. Because it is proposed that cells are unable to commit to cell division until a minimum cell size is attained, this mechanism also prevents normal somatic cells from getting continually smaller after each division (
Figure 3Cell growth and cell division. (A) In nearly all somatic cells, cell division is dependent upon cell growth. Homeostasis is maintained because on the average, cells double their mass before dividing. (B) In contrast, in some cell types, cell growth and cell division can be uncoupled. For example, neurons, oocytes, and adipocytes can grow without dividing. (C) Cells that are unusually large (e.g., oocytes) can proliferate in absence of cell growth. In this case, normal G1-phase cell cycle controls are absent and cells divide rapidly. Homeostasis is maintained because once cells become smaller than normal, cell division again becomes dependent upon cell growth.
). But because size is a rather amorphous characteristic, it has proved very difficult to extend these observations from a correlative to causative relationship. Moreover, despite the fact that G1-phase Cdks are integral to both cell size homeostasis and proliferation, the relationship between cell size, G1-phase Cdk activity, and proliferative capacity is not well understood. In addition, whereas cell growth is required for proliferation, it is not known how cell size affects cell growth. Finally, it is unclear how a cell might sense its size or how cell size might trigger cell division. Thus, the two major goals in this field are: (1) To determine if cell size has a causative role in promoting cell division, and (2) To determine the molecular mechanism that links cell growth to proliferative potential.
Like embryonic cells, the proliferation of cancer cells is largely independent of mitogens and growth factors. This occurs, in part, because the pathways known to modulate G1-phase cyclin activity are mutated or disrupted in nearly every cancer cell (
). Because of the central role of G1-phase Cdks in coordinating cell growth with division, the elucidation of the genetic, molecular, and biochemical details of these mechanisms is likely to provide significant insight into the role of these processes in promoting proliferation in both normal and cancer cells. As the basic cell cycle machinery is highly conserved, data obtained from model systems like yeast will continue to be invaluable in dissecting the intricacies and importance of appropriate cell cycle controls.
Here we show, in yeast cultures, that proliferative capacity correlates with cell size and cell growth rates. We found that the time to Start is not constant for all cells but has an inverse relationship to cell size. Large cells have a higher proliferative potential. Supporting previous observations, we show that whereas G1-phase cyclin (Cln) abundance is modulated by cell cycle position and cell size, it correlates most closely with growth rate. For example, large rapidly growing cells express considerably more Cln than do small slowly growing cells. Investigation of new cell size mutants revealed that a number of highly conserved signal transduction pathways are involved in cell size homeostasis. Detailed analysis of specific cell size mutants indicates that cell size homeostasis is achieved in part through modulation of CLN transcription. But the discovery of novel classes of cell size mutants indicates that the molecular mechanism that coordinates cell growth with proliferation is highly complex. Further analysis of genetic data obtained in yeast should help to elucidate the complex relationship between cell size, G1-phase Cdk activity, and proliferative potential and shed light on the role of these processes in carcinogenesis.
Results
Cln abundance is modulated by cell cycle position and cell size
The yeast G1-phase cyclin, Cln3, modulates the transcription of two other G1-phase cyclins, Cln1 and Cln2 (reviewed in
). Cln3 activates the transcription of CLN1 and CLN2 in part by phosphorylating the yeast orthologue, Whi5, of the mammalian RB tumor suppressor gene (
). Initial experiments were done by expressing CLN1 from a non-cell cycle regulated promoter, the GAL1 promoter, and measuring Cln1 protein abundance as a function of cell size. Here, we have expanded upon this observation by showing that Cln2 protein is similarly upregulated in a size-dependent manner in the absence of changes in transcription.
To conduct these experiments, we used a construct where the CLN2 gene was fused to the constitutive spADH promoter. Under these circumstances, CLN2 transcription is unaffected by cell cycle position or cell size. This construct was transformed into a conditional Cln strain (e.g., GAL–CLN1 cln2 cln3) (
). In this strain, the expression of CLN1 is dependent upon the presence of galactose in the medium. In the absence of galactose, the transformed spADH–CLN2 construct is the strain's sole CLN gene. This arrangement allowed us to investigate the role of the transformed spADH–CLN2 construct in promoting cell cycle progression independent of all endogenous G1-phase cyclins. Cultures of this strain were grown to mid-log phase in the absence of galactose, and centrifugal elutriation was used to obtain fractions of cells on the basis of size. In this manner, eight fractions were obtained ranging in size from 29 to 60 fL Figure 4. Protein was isolated from each fraction, and western analysis revealed that Cln2 abundance increased dramatically as the cells got larger Figure 4a. Quantitation of the western data and normalization to the B-tubulin loading controls revealed that Cln2 protein abundance was more than 30-fold higher in the large 60 fL cells as compared with the smaller 29 fL cells Figure 4b. These data confirm that both Cln1 and Cln2 protein abundance are modulated by cell size.
Figure 4Cln protein levels are is modulated by cell size. A conditional Cln yeast strain BS111 was transformed with a spADH–CLN2–HA construct. In the presence of glucose, this construct is the only source of Cln for this strain. In this case, in contrast to wild-type cells, CLN2 mRNA levels are constitutive and not cell size dependent. In contrast, western analysis of elutriated fractions demonstrates that Cln2 protein levels are size dependent. B-tubulin is used for a loading control. Lanes N and P are negative (e.g., a strain with an untagged CLN2 gene) and positive (e.g., CLN2–HA3 over-expressed from the GAL promoter) controls. (B) Quantitation of the western data and normalization to the B-tubulin loading controls revealed that Cln2 protein abundance was more than 30-fold higher in the large 60 fL cells as compared with the smaller 29 fL cells.
Cell cycle position correlates strongly with cell size. That is, G1-phase cells tend to be smaller than S/G2/M-phase cells. In addition, progression past Start is believed to be size dependent. But because the molecular details that link cell size to cell cycle progression are not well understood, it remains plausible that the relationship between cell size and proliferation is entirely correlative and not causative. Here, we have examined the relationship between cell size and proliferative capacity using growth-arrested wild-type yeast cells. The objective of these experiments was to answer three questions. First, in minimally perturbed cultures that are growth arrested in G1 phase, does the proliferative capacity of cells correlate closely with cell size or is the time to Start constant regardless of cell size? Second, how does cell size affect cell growth rates? Finally, given that Clns are linked to proliferative capacity, does the abundance of Cln protein correlate more closely with cell size or cell growth rate?
To conduct these experiments, we grew a yeast culture to saturation. After 4 d in culture, cells stopped growing and proliferating. Flow cytometry analysis revealed that >99% of the cells were in the G1 phase of the cell cycle. Subsequently, centrifugal elutriation was used to fractionate cells on the basis of size. In this manner, six fractions were obtained of cells that ranged from an average of 48 to 230 fL in volume. Each fraction was resuspended in fresh medium and returned to standard culture conditions. Cell cycle progression was monitored and the time to Start was determined as a measure of proliferative potential. Using this minimally perturbed culture, it was found that even though the cell division time of the asynchronous culture was constant and stable, the time to Start for individual populations of cells was extremely variable. Specifically, times to Start ranged from 1.8 h for the largest cells to 9.5 h for the smallest cells Figure 5a. The near linear relationship between cell size and proliferative capacity demonstrates that under these conditions cell size correlates strongly with the probability of cell division.
Figure 5Proliferative capacity correlates with cell size. (A) A wild-type yeast strain where the endogenous CLN1 gene is tagged with an HA epitope was grown for 4 d at room temperature to saturation. Cell fractions of different sizes were collected by centrifugal elutriation and then released into fresh medium and incubated at room temperature. Time and size to Start (50% budded) was determined for six different sized fractions. A near linear relationship between cell size and proliferative capacity was observed indicating that under these conditions cell size determines the probability of cell division. (B) The rate of cell volume increase (fL per h) was plotted for each fraction and found to be cell size dependent for fractions 1–5. (C) The expression of Cln1–HA protein at Start in each fraction was analyzed by the western blot. Initial size, cell size at Start, and time to Start are given. B-tubulin is used for a loading control. Lanes N and P are negative (e.g., a strain with an untagged CLN1 gene) and positive (e.g., CLN1–HA3 over-expressed from the GAL promoter) controls. (D) Quantitation of the western data and normalization to the B-tubulin loading controls revealed that Cln1 protein abundance correlated more closely to growth rate than to cell size.
In cycling cells, cell growth is required for progression past Start. But the relationship between cell growth rate and cell size is less clear. It is known that in asynchronous cultures, rapidly proliferating cells are considerably larger than slowly proliferating cells (
). In addition, it is not known if all cells have the same growth requirement in order to progress past Start. To examine these questions, we measured cell growth over time in each of the six different size fractions that were collected. Analysis of these data revealed that in the first five fractions of cells, cell growth was size dependent Figure 5b. That is, the largest cells grew more rapidly than the smaller cells. For instance, cells in the fifth fraction grew at an average of nearly 14 fL per h Figure 5b. In contrast, the smallest cells grew at an average of about 1 fL per h Figure 5b. Strikingly, the overall growth requirement varied considerably. For example, cells in the sixth fraction grew only 3 fL before Start whereas cells in the fourth fraction grew 51 fL before Start Figure 5b, c. This clearly demonstrates that whereas the probability that a given cell will divide correlates with cell size, this size is not the same for all cells even under identical culture conditions. Rather, these data suggest that proliferation potential is linked to both cell size and cell growth rate.
Cln abundance is growth rate dependent
We have previously shown that not only are Clns rate limiting for proliferation but that a minimum threshold of Cln is required for progression past Start (
). This mechanism helps link proliferation to cell size. In addition, we have shown that Cln abundance and the Cln threshold level are strongly modulated by growth rate and cell size (
). Because of this, we sought to determine the relationship between growth rate, cell size, and Cln abundance in this experiment. Western analysis of cell fractions at Start revealed that the abundance of Cln2 protein after normalization to B-tubulin was size dependent in all but the largest cell fractions Figure 5c, d. Comparison of the relationship between relative Cln abundance Figure 5d and relative growth rates Figure 5b reveals that the two curves are nearly super-imposable. This strongly supports our previous findings that Cln abundance and the apparent Cln requirement are strongly modulated by both cell size and growth rate.
But the very large cells in the sixth fraction illustrate an anomalous situation that runs counter to this generalization. These cells, the largest of all the cell fractions, progress past Start in the shortest period of time with the least amount of Cln and the least amount of cell growth. Thus, like embryonic cells, these yeast cells have the highest proliferative potential and appear to be capable of dividing in absence of normal growth and cell cycle controls. At this time, the mechanisms responsible for these observations are not known.
Cell size genetics and CLN transcription
The analysis of yeast cell size mutants has helped clarify the molecular mechanisms involved in cell size homeostasis. But because until recently very few cell size control genes were known, the genetic pathways responsible for cell size homeostasis have remained relatively obscure. Furthermore, elucidation of the mechanism of cell size homeostasis has been recalcitrant to genetic analysis primarily because of the difficulty in cloning cell size control genes. One great advantage to yeast is that nearly all of the 6200 ORFs have been systematically deleted in individual strains (
). Microscopic analysis of these deletions revealed that it was feasible to perform a brute-force systematic genome-wide genetic screen for cell size mutants Figure 6a. Therefore, to identify new size control genes, the effect of 5958 single gene deletions (4792 homozygous and 1166 heterozygous gene deletions) on cell size in yeast grown to saturation was systematically determined using a Coulter Counter Channelyzer Figure 6b. From these data, 49 genes were identified that dramatically altered cell size without any obvious growth defects (
). Interestingly, these genes clustered non-randomly into pathways. For example, 36 of these are involved in transcription, signal transduction, or cell cycle control; 89% of these genes have putative human homologues (
Figure 6Cell size mutants. (A) Photomicroscopy of a mixture of wild-type and yeast mutants revealed that it was possible to differentiate large (U) and small (W) cell mutants from wild-type cells (WT). Measurements indicate that the largest mutants are ∼10-fold larger than the smallest. Scale bar∼5 μM. (B) Coulter counter analysis of 5958 single gene deletion mutants. The number of deletion mutants was plotted as a function of their mean size.
Figure 7The Ccr4–Not transcriptional complex. Analysis of cell size mutants revealed that the Ccr4–Not transcriptional complex appears to be involved in cell size regulation. Deletion of the genes for Hpr1, Paf1, Flr1, Caf1, and Ccr4 (shaded gray) resulted in abnormally large cells. Interestingly, all five of these gene products interact physically in Ccr4–Not transcriptional complexes. Ccr4 is the core component of complexes involved in transcription and poly (A) deadenylation. In addition, four other gene products (shown in italics and underlined) are known to effect cell size, cell cycle progression, or CLN expression.
). Based on this evidence, we examined the expression patterns of CLN1 mRNA in a panel of the new cell size mutants. Mutant cultures were grown to mid-log phase and centrifugal elutriation was used to isolate synchronized small G1-phase cells. Subsequently, these cells were resuspended in fresh medium and cultured. Regular time points were taken, and the CLN1 mRNA expression patterns were examined by northern blotting Figure 8a. Analysis of these data revealed that in large cell size mutants (e.g., rpt2) CLN1 mRNA expression was delayed and peaked at larger than normal cell sizes Figure 8. In contrast, in small cell size mutants (e.g., sac1) CLN1 mRNA expression was advanced and peaked at smaller than normal cell sizes Figure 8. This general pattern has held true for all cell size mutants examined to date.
Figure 8CLN1 expression in cell size mutants. (A) Deletion of the gene for Rpt2 results in abnormally large cells. Asynchronous cultures were elutriated and synchronized G1-phase cells were followed through a time course. Northern analysis revealed that CLN1 mRNA peaked at 111 fL. Asynchronous cultures (AS) are shown for reference. Blots were probed with ACT1 to control for loading. (B) Deletion of the gene for Sac1 results in abnormally small cells. Asynchronous cultures were elutriated and synchronized G1-phase cells were followed through a time course. Northern analysis revealed that CLN1 mRNA peaked at 50 fL. AS are shown for reference. Blots were probed with ACT1 to control for loading. (C) Expression of relative levels CLN1 normalized to ACT1 in sac1Δ and rpt2Δ mutants is plotted as a function of cell size. Under these conditions, in wild-type cells CLN1 mRNA expression peaks around 65 fL.
But rigorous analysis of this mutant set has revealed that mechanisms that control cell size homeostasis are likely to be highly complex. All previous small cell size mutants, promoted premature progression past Start and shortened the G1 phase of the cell cycle (
). In contrast, G1 phase was not shortened in the majority of our new small cell size mutants. In addition, we have found that many of these mutants failed to correctly exit the cell cycle. Moreover, we have identified two new classes of cell size mutants. Both of these classes exhibit important growth defects. The first class contains 181 mutants, all of which have mitochondrial or respiratory defects. Interestingly, nearly all of these mutants are abnormally small. In addition, we identified 216 mutants that proliferated abnormally slowly. Of these, 58 were small cell size mutants, and 18 were large cell size mutants. To date, we have no molecular explanation for these results. But the fact that not all of the slowly proliferating mutants were cell size mutants implies that the proliferation defect alone is not responsible for the alteration in cell size.
Discussion
Weighing in on the “critical cell size” theory
The first objective of the work presented here was to examine the relationship between cell size and proliferative capacity. Specifically, we sought to determine how closely the proliferative capacity of cells correlates with cell size. The “critical cell size” theory postulates that the attainment of a minimum cell size is required for commitment to cell division. But opponents of this theory suggest that it is equally possible that commitment to cell division is time dependent rather than size dependent (
). That is, if under a given condition, cells grew at a constant rate, then the observed correlation between cell size and proliferative capacity could be time dependent (
). In this case, it would be expected that all cells would divide at a constant rate regardless of cell size. Historically, a number of experiments and observations suggest that this view is incorrect (discussed in
). For example, cell cycle and G1-phase lengths are highly variable suggesting that a “simple timer” mechanism does not promote cell division (discussed in
Kinetic evidence for a critical rate of protein synthesis in the Saccharomyces cerevisiae yeast cell cycle [published erratum appears in J Biol Chem 1988 Dec 5;263(34):18582].
Relation between cell growth and cell division. III. Changes in nuclear volume and growth rate and prevention of cell division in Amoeba proteus resulting from cytoplasmic amputations.
Relation between cell growth and cell division. III. Changes in nuclear volume and growth rate and prevention of cell division in Amoeba proteus resulting from cytoplasmic amputations.
). From this experiment, it was concluded that these cells were continually monitoring their cell mass and not the amount that they had grown nor the time that had elapsed since their last division. Subsequently, researchers using mouse cells showed that the probability that a given cell would divide was proportional to its cell size (
). These studies have been greatly expanded upon in both budding and fission yeast where numerous experiments have revealed that cell size correlates closely with the probability that a cell will divide (discussed in
). In addition, recent experiments in mammalian cells support the conclusion that cell size correlates closely with the proliferative potential of cells (
The data presented here using minimally perturbed and growth arrested G1-phase yeast cultures also support the “critical cell size” theory. We found that the proliferative potential of cells correlated closely to the size of cells. That is the largest cells began to proliferate five times faster than the smallest cells, and in general this trend continued until the population reached a “normal” cell size. In this case, G1 phases and cell division times were not constant, but instead they inversely correlated to cell size.
The second objective of this work was to examine how cell size affects cell growth rates. Somatic cells require both cell growth and high rates of protein synthesis to proliferate. In contrast, embryonic cells can proliferate rapidly in the absence of cell growth. Blocking the cell cycle with inhibitors generates abnormally large cells (
Kinetic evidence for a critical rate of protein synthesis in the Saccharomyces cerevisiae yeast cell cycle [published erratum appears in J Biol Chem 1988 Dec 5;263(34):18582].
Kinetic evidence for a critical rate of protein synthesis in the Saccharomyces cerevisiae yeast cell cycle [published erratum appears in J Biol Chem 1988 Dec 5;263(34):18582].
). But the use of cell cycle inhibitors may create a non-physiological scenario. Therefore, we wished to examine the relationship between cell size, cell growth, and proliferation under normal physiological conditions.
Again, using minimally perturbed and growth arrested G1-phase yeast cultures, we measured the growth rate of cells released to fresh medium. From these data, we found that, in general, cell growth prior to Start was proportional to cell size. That is, large cells grew more rapidly than small cells. Interestingly, for the largest fraction of cells, this correlation did not hold true. These cells had the shortest G1 phase and the smallest growth prior to Start. In addition, we found that the growth requirement for Start was not the same for all cells despite the fact that all of the cultures were propagated under identical conditions. The largest cells grew only 3 fL or 1% whereas the smallest cells grew 12 fL or 25%. Interestingly, cell growth was highest in cells of intermediate cell size where some cells grew as much as 51 fL or 40%. In summary, these data indicate that asynchronous cultures are composed of discrete populations of cells, each of which has unique cell cycle requirements. For this reason, the relationship between cell size, cell growth, and the proliferative potential of cells will probably be best clarified by studying the kinetics of proliferation of single cells rather than populations. The results presented here suggest that there is apparently no absolute cell size, cell growth or time requirement that ensures a given cell will divide. These results demonstrate that the relationship between cell growth, cell size, and proliferative potential is perhaps more complex than has been previously suspected.
Cln substrates and Cln thresholds
Clns are linked to proliferative capacity, and we have shown that cells require a minimal threshold level of Cln in order to proliferate (
). Larger cells have a higher abundance of Cln. Finally, we and others have shown that over-expression of Cln leads to premature proliferation and dramatically reduces the size of cells (
). But to date, the molecular explanation for these results is still largely unknown.
The final objective of this work was to determine if the abundance of Cln protein correlates more closely with cell size or cell growth rate. Here, we have shown that whereas Cln protein abundance is modulated by cell size, the absolute levels of Cln protein correlate more closely with growth rate than with cell size. Because Clns are constitutively unstable proteins, this type of mechanism may allow cells to use Cln levels as a gauge of the synthetic capacity of a cell (
). It may be this mechanism that links cell growth to proliferation.
Data indicate that rapidly growing cells express and require more Cln for cell division than do slowly growing cells. To date, the molecular mechanisms responsible for these observations are not known. One possibility is that the levels of a G1-phase Cdk substrate may determine the Cln requirement for proliferation. There are several known substrates of Cln–Cdc28 kinase complexes. Most of these are proteins involved in proper cell cycle regulation. Of these, one of the most intriguing is Whi5. Deletion of WHI5 results in a small cell size phenotype (
). Thus, Whi5 makes an excellent potential candidate for setting the required Cln threshold for Start. The Cdk inhibitor Sic1 is also a G1-phase Cdk substrate and several observations suggest that is a putative candidate for setting the Cln threshold (
During gametogenesis, cells destined to become gametes arrest cell cycle progression, but cells continue to grow. Because of this, oocytes can achieve sizes that are ∼105 times larger than normal somatic cells. Importantly, once the large oocytes are fertilized, the resulting embryonic cells proliferate rapidly. In addition, these cell divisions occur in the absence of cell growth. We have recently shown that cell size also affects the developmental options available to yeast where entry into meiosis is dependent upon the attainment of a minimum cell size (
). Here, we have shown that “supersized” yeast cells can be selected from wild-type cultures grown under physiological conditions. This result illustrates that despite the fact that the average cell size in a population is remarkably stable, cells grown under physiological conditions display a wide range (∼5-fold) of cell sizes. These extra large cells display many of the characteristics of fertilized oocytes. Like embryonic cells, they proliferate rapidly in the absence of cell growth and normal cell cycle controls. This rapid proliferation in the absence of growth leads to a reduction in average cell size over time. In addition, we have shown that these “supersized” yeast cells express and appear to require very low levels of G1-phase cyclins. In an attempt to elucidate the molecular mechanisms responsible for these observations, we have identified and analyzed a panel of new cell size mutants. Data from this analysis has revealed that many of the pathways involved in cell size homeostasis are strongly conserved from yeast to man (
). Moreover, most of these pathways modulate the expression and activity of G1-phase Cdks. Components of these pathways are known to be mutated in nearly every cancer cell suggesting that these signal transduction cascades are vital to the maintenance of normal cellular reproduction (reviewed in
). The high degree of conservation of these basic cell cycle pathways suggests that genetic data obtained from model organisms like yeast will continue to be invaluable in dissecting the mechanism whereby cells couple cell growth with division to achieve balanced, stable, and developmentally appropriate proliferation.
Materials and Methods
Strains and media
Strains used in this work are derived from W303. Yeast cultures were grown in YEP-based media (20.0 g Difco Bacto peptone and 10.0 g Difco Bacto yeast extract were dissolved in 900 mL of water and autoclaved) or YNB-based media (6.7 g Difco-Bacto yeast nitrogen base lacking amino acids and ammonium sulfate was added to 900 mL of water and autoclaved). Required amino acids were supplemented at 50 mg per liter except for tryptophan (80 mg per liter), adenine sulfate (32 mg per liter), and p-aminobenzoic acid (5 mg per liter). After autoclaving, sterile filter carbon sources were added to a final concentration of 2% (glucose or galactose).
Preparation of RNA and northern analysis
Yeast cultures were grown to a mid-log phase (1–3 × 107 cells per mL). Cultures were chilled rapidly by adding an equal volume of ice to media. The cells were pelleted by centrifugation at 4°C and washed in ice-cold water. Cell pellets were frozen at -80°C. Pellets were resuspended in 250 μL of LETS buffer (100 mM LiCl, 10 mM EDTA, 10 mM Tris-HCl, pH 7.4, and 0.2% SDS). Subsequently, 300 μL of LETS-equilibrated phenol and an equal volume of 450 nm acid-washed glass beads were added. The cell suspensions were vortexed at maximum speed for 30 s, and then an additional 200 μL of LETS was added. The cell suspensions were vortexed briefly and then centrifuged for 5 min at 16,000 ×g. The upper aqueous phase was removed and extracted twice with phenol/chloroform. RNA was precipitated by adding 1/10 volume of 5 M LiCl and 2.5 volumes of ice cold ethanol, then incubated 1–12 h at -20°C. After precipitation, RNA was recovered by centrifuging 15 min at 14,000 r.p.m. followed by a wash with 70%–80% ethanol. RNA pellets were air-dried at room temperature and resuspended in 50–100 μL DEPC-treated water. Size separation of RNA was performed using 1.0% denaturing agarose gels containing 6.6% formaldehyde and 1 × MOPS. Typically, 10 μg of RNA was lyophilized in a Savant speed vac and resuspended in 5 μL of DEPC-treated water. Subsequently, 17.5 μL of RNA loading buffer was added (12.5 mM MOPS, pH 7.1, 2.5 mM NaOAC, 0.25 mM EDTA, 3.1% formaldehyde, 25% formamide, 2% glycerol dye, 4 mg per mL bromphenol blue, 4 mg per mL xylene blue, and 50 μg per mL ethidium bromide), heated at 65°C for 15 min and loaded onto a gel that was pre-run at 90 V for 20 min. Gels were run for 30 min at 45V and then at 90 V for 3–5 h. After electrophoresis, gels were soaked in DEPC-treated water with gentle shaking for 45 min and then transferred to Nytran-Plus nylon membranes (Schleicher and Schuell Florham Park, New Jersey) as recommended by the manufacturer. After transfer, nucleic acids were cross-linked to membranes using UV light (UV Stratalinker 1800, Stratagene La Jolla, California) as recommended by the manufacturer. Hybridization of membranes was performed as previously described using Church hybridization buffer (7% w/v SDS, 0.1% w/v BSA (fraction V (Sigma St. Louis, Missouri)), 0.1 mM EDTA, and 0.25 M Na2HPO4 pH 7.2) (
). Filters were pre-incubated in hybridization buffer for 30 min at 65°C. Radioactive probes were made using α32P-ATP and a random prime labeling kit from Roche Applied Science, Indianapolis, Indiana. Probes were purified on G-50 sephadex spin columns, denatured by boiling for 5 min, and then added to pre-hybridization buffer and incubated 12–16 h at 65°C. Subsequently, blots were washed once with 2 × SSC for 5 min, twice with 2 × SSC+0.1% SDS (pre-heated to 65°C) at 65°C for 15 min, twice with, and finally in 2 × SSC for 15 min. Filters were wrapped in plastic wrap and exposed to Kodak XAR film or a Molecular Dynamics (Rochester, New York) phosphoimager screen for further analysis.
Protein extraction and Western analysis
Yeast extracts for Western analysis were prepared as previously described (
). Briefly, yeast pellets were lysed in a mini-beadbeater cell disrupter (Biospecs Bartlesvilte, Oklahoma) using 0.5 mm diameter acid-washed baked zirconia beads in the presence of buffer 3 (0.1% NP40, 250 mM NaCI, 50 mM NaF, 5 mM EDTA, and 50 mM Tris-HCl pH 7.5) and proteinase inhibitors (1 mM PMSF, l μg per mL leupeptin, 1 μg per mL pepstatin, 0.6 mM dimethylaminopurine, 10 μg per mL soybean trypsin inhibitor, and 1 μg per mL TPCK). Cell debris was pelleted by centrifuging at 16,000 ×g. for 15 min. Protein concentrations were determined with the Bio-Rad (Hercules, California) dye-binding assay according to the manufacturer's specifications. For western analysis, 50 μg of protein lysates were mixed with an equal volume of 2 × protein sample and samples were boiled for 2 min. Samples were loaded onto small 10% SDS-PAGE gels and run at 75–100 V. Protein gels were transferred to nitrocellulose using a semi-dry transfer apparatus (Millipore Billerica, Massachusetts) and probed consecutively with primary anti-HA antibody 12CA5 (diluted 1:5000) and secondary HRP-conjugated Sheep anti-mouse (1:20,000 Amersham Piscataway, New Jersey). Proteins were visualized using the Amersham ECL system or the Pierce Supersignal system according to the manufacturer's specifications.
Quantification of cell size, percent of budded cells, and cell cycle distributions
Cell cycle synchronizations were performed using centrifugal elutriation as previously described (
). Cell cycle synchrony was confirmed using microscopic analysis and flow cytometry. The percent of budded cells was determined by coding samples and then counting the number of cells with visible buds in a minimum of 200 cells. The percent of budded cells was verified in at least two independent experiments. For flow cytometry, yeast cells were harvested, washed, sonicated, and fixed overnight in 70% ethanol at 4°C. Cells were resuspended in 50 mM sodium citrate, washed in the same buffer, sonicated, treated with RNAse A (final concentration 0.25 mg per mL) for 1 h at 50°C, and treated with Proteinase K (final concentration 1 mg per mL) for an additional hour at 50°C. Before analysis the yeast cells were stained with propidium iodide at a final concentration of 16 mg per mL. Flow cytometry was performed on yeast cells stained with propidium iodide with an Epics XL (Beckman-Coulter Fullerton, California) flow cytometer. Analysis of the cell size distribution of yeast strains was done using cultures in mid-log phase. Samples of the cultures were resuspended in 10 mL of Isoton buffer, briefly sonicated, and immediately analyzed using a Coulter Counter Channelyzer Model Z2 (Beckman-Coulter) (100 micron aperture).
ACKNOWLEDGMENTS
We thank M. Tyers and B. Futcher for useful reagents and J. Hutson and S. Williams for helpful discussions and comments. This research was supported by grants from the American Heart Association, The CH Foundation, the Wendy Will Cancer Fund, the Houston Endowment Incorporation, the South Plains Foundation, and Texas Tech University Health Sciences Center to B. L. S.
References
Alberts B.
Bray D.
Lewis J.
Raff M.
Watson J.D.
Molecular Biology of the Cell. New York,
Garland1994
Kinetic evidence for a critical rate of protein synthesis in the Saccharomyces cerevisiae yeast cell cycle [published erratum appears in J Biol Chem 1988 Dec 5;263(34):18582].
Relation between cell growth and cell division. III. Changes in nuclear volume and growth rate and prevention of cell division in Amoeba proteus resulting from cytoplasmic amputations.
52575-X&id=gr1.gif){kind=link}
52575-X&id=gr2.jpg){kind=link}
52575-X&id=gr3.jpg){kind=link}
52575-X&id=gr4.jpg){kind=link}
52575-X&id=gr5.jpg){kind=link}
52575-X&id=gr6.jpg){kind=link}
52575-X&id=gr7.jpg){kind=link}
52575-X&id=gr8.jpg){kind=link}