Wu et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Control of Transcription by Cell Size
1Department of Biology, Massachusetts Institute of Technology, Cambridge, Massachusetts, United States of America
2Whitehead Institute for Biomedical Research, Cambridge, Massachusetts, United States of America
3Computer Science and Artificial Intelligence Laboratory, Department of Electrical Engineering and Computer Science, Massachusetts Institute of Technology, Cambridge, Massachusetts, United States of America
4Broad Institute of MIT and Harvard, Cambridge, Massachusetts, United States of America
Stephen J. Elledge, Academic Editor
Harvard Medical School, United States of America
Received April 1, 2010; Accepted September 7, 2010.
Cell size increases significantly with increasing ploidy. Differences in cell size and ploidy are associated with alterations in gene expression, although no direct connection has been made between cell size and transcription. Here we show that ploidy-associated changes in gene expression reflect transcriptional adjustment to a larger cell size, implicating cellular geometry as a key parameter in gene regulation. Using RNA-seq, we identified genes whose expression was altered in a tetraploid as compared with the isogenic haploid. A significant fraction of these genes encode cell surface proteins, suggesting an effect of the enlarged cell size on the differential regulation of these genes. To test this hypothesis, we examined expression of these genes in haploid mutants that also produce enlarged size. Surprisingly, many genes differentially regulated in the tetraploid are identically regulated in the enlarged haploids, and the magnitude of change in gene expression correlates with the degree of size enlargement. These results indicate a causal relationship between cell size and transcription, with a size-sensing mechanism that alters transcription in response to size. The genes responding to cell size are enriched for those regulated by two mitogen-activated protein kinase pathways, and components in those pathways were found to mediate size-dependent gene regulation. Transcriptional adjustment to enlarged cell size could underlie other cellular changes associated with polyploidy. The causal relationship between cell size and transcription suggests that cell size homeostasis serves a regulatory role in transcriptome maintenance.
Cells of the same type, whether microbial, plant, or metazoan in origin, exhibit remarkable uniformity in size. This uniformity arises from control mechanisms that respond to internal cellular changes as well as external environmental factors. Although precise control of cell size is a universal phenomenon, its relationship to cellular physiology is underexplored. In this study using yeast we show a causal relationship between cell size and gene regulation: changes in cell size correlate with changes in the expression of a set of genes. Hence, the maintenance of uniformity in cell size could be a homeostatic mechanism for the maintenance of gene expression in a cell or in a population of cells within a tissue. The relationship between cell size and gene expression uncovered in this study may have fundamental implications in evolution, in the development of multicellular organisms, and in the formation of tumors, as these processes often involve genome duplication accompanied by enlarged cell size.
The size of cells can vary significantly within an organism, and cells of the same type display pronounced increase in size with increasing ploidy 
. During development, specific cell types in many diploid organisms perform endoreplication and differentiate into polyploid cells that are functionally distinct from their diploid progenitors 
. Polyploidy also occurs as an intermediate state in aneuploid tumor formation 
and as a mechanism to create substrates for evolution 
. From yeast to mammals, polyploidy is associated with enlarged cell size and altered cellular physiology 
. How polyploidy changes physiology is a long-standing question. Furthermore, a causal relationship between enlarged cell size and altered physiology has not been discovered.
Yeast offers a unique advantage in studying the physiological consequences of polyploidy, because it is possible to construct isogenic strains of increasing ploidy. There were two previous analyses that compared transcription between cells of different ploidy. The first analysis of transcription in a yeast ploidy series identified a few genes whose transcript abundance in the transcriptome was altered by ploidy 
. These included some genes that were strongly repressed and others that were strongly induced in polyploids. Although this study established a clear effect of ploidy on transcription, the limited set of identified genes did not reveal a functional relationship between ploidy and gene expression. The scope and sensitivity of this early investigation were hampered by technical limitations. Because the genome sequence of the studied yeast strain (Σ1278b) was not known at the time, microarrays designed for a related yeast strain (S288c) were employed. Recent genome analysis comparing these two yeast strain backgrounds has revealed many polymorphisms and changes in genomic organization 
that compromised the power of detection by hybridization in the earlier study.
A subsequent analysis of polyploid yeast detected no significant differences between the diploid and tetraploid transcriptomes by microarrays 
, raising the possibility that the differences found in the first study were strain specific. Alternatively, experimental differences between the two studies could account for the different conclusions. Expanding the second study to compare strains with a greater difference in ploidy (i.e., between haploids and tetraploids) might have uncovered significant transcriptional changes related to ploidy, as was observed in the first study. More importantly, this later study used a different laboratory strain (S288c). Unlike the strain used in the first study (Σ1278b), the S288c strain background does not express FLO11 
, the gene that was most affected by ploidy in the first study 
We address the issues raised by both of these studies by examining ploidy effects in different yeast strains with more sensitive assays. The recently acquired genome sequence of the Σ1278b strain 
, combined with advances in transcriptome profiling by RNA-seq 
, provides the resolution necessary for genome-wide determination of a functional connection among genes regulated by ploidy. The dynamic range of quantitative linearity in RNA-seq is at least 10-fold higher than that of microarrays, making RNA-seq superior at comparing transcript abundance 
. In this study, RNA-seq enabled identification of a much larger set of differentially expressed genes between Σ1278b haploid and tetraploid strains and the discovery of related genes by Gene Ontology (GO) analysis. The enriched GO terms suggested a causal relationship between cell size and gene expression, and this relationship was then confirmed by analyzing gene expression patterns in cells of varying sizes. The genes repressed in large cells of the Σ1278b background were also found to be repressed in tetraploids of S288c, suggesting that the casual relationship between cell size and gene expression is a general feature.
RNA-Seq Reveals Novel Genes Differentially Regulated in Tetraploids as Compared with Haploids
To identify transcripts whose relative abundance in the transcriptome is changed by ploidy, poly(A) RNA transcripts isolated from isogenic haploids and tetraploids of the Σ1278b strain background were analyzed by RNA-seq (). Approximately 9 million sequence reads were obtained from each cDNA library, and the majority of reads mapped to annotated ORFs (). Pair-wise comparison of tetraploid samples with haploid samples revealed that ploidy affects the abundance of only a small proportion of the total transcripts (). By comparison with haploids, 35 transcripts were significantly and reproducibly repressed and 30 transcripts were induced in tetraploids (). The differentially expressed genes included several of the strongly regulated genes identified in the previous study on a Σ1278b ploidy series, and the majority of the remaining genes showed consistent regulatory trends in both studies (; Dataset S1
. The disproportional expression of these genes appeared unrelated to the cell cycle, since there was no systematic bias for genes expressed in specific cell cycle stages (Table S1
Identification of ploidy-regulated genes in yeast (Σ1278b).
Genes differentially regulated in the Σ1278b tetraploid and their compartmental bias for encoding cell surface components.
Gene Ontology Analysis Suggests That Cell Size but Not Ploidy Alters Gene Expression in Tetraploids
Interestingly, the genes differentially expressed in tetraploid cells are significantly enriched not for those associated with chromosomes but for those encoding proteins localized to the cell surface (cell wall, extracellular space, and plasma membrane) and for genes that encode regulators of cell surface components (; and S2
). This compartmental bias suggests that the differential gene expression in tetraploids is not directly caused by an increase in the genome content, but by a difference in cell size/geometry: for a spherical cell, a 4-fold increase in volume corresponds to only a ~2.5-fold increase in surface area. In other words, although tetraploid yeast cells are 4-fold larger in volume than haploid cells 
, the ratio of surface area to volume is smaller in tetraploids than in haploids. Reduction in surface area relative to volume is likely to trigger differential regulation of components associated with the cell surface, where signaling and transport processes take place dynamically. A reduction in relative cell surface area could alter interactions between surface and cytoplasmic signaling pathway components and affect the cell's ability to transport metabolites across the plasma membrane. Either type of perturbation caused by a reduced surface area relative to volume could alter gene expression in enlarged cells.
Cellular compartmental Gene Ontology terms for genes differentially regulated in the Σ1278b tetraploid.
Haploid Size Mutants Also Demonstrate Regulation of Gene Expression by Size
The relationship between cell size and transcription could be assessed by examining gene expression levels in haploid mutants with altered cell size relative to wild type (WT). Cell size mutants () previously identified in a genome-wide study 
were selected because they effectively enlarge cell size without significantly affecting fitness or cell shape. In addition, the underlying mutations have no reported functional relationship with the differentially expressed genes identified in tetraploid cells. We considered additional mutations to increase cell size in haploids 
but did not pursue them because of technical concerns that would preclude a clear interpretation of experimental results (Text S1
Enlarged cell size represses FLO11.
If differential expression of a gene in tetraploids is caused by enlarged cell size rather than directly by higher ploidy, the gene should be differentially expressed in haploid size mutants as well. Furthermore, the magnitude of change in the gene's expression level should, ideally, correspond to the magnitude of change in cell size. This cell size–transcription hypothesis was initially tested by investigating the effect of cell size on expression of FLO11
, a gene encoding a cell surface glycoprotein 
showed the highest degree of repression in the tetraploid (), providing a wide range of detection for changes in transcript levels.
Expression levels of FLO11
were measured in WT Σ1278b and isogenic size mutant haploids () treated with nocodazole for cell cycle arrest in M-phase, when the FLO11
transcript is most abundant (). Because the size mutants manifest an altered cell cycle 
, arresting the cell cycle was necessary to separate the transcriptional effect of cell size from that of cell cycle. We analyzed mutant alleles of the CLN3
gene that altered cell size significantly and arrested efficiently in the presence of nocodazole. The CLN3-2
mutant arrested as small cells (66% of WT volume), whereas the cln3
Δ mutant arrested as large cells (185% of WT volume) (). In this haploid size series, we found an inverse correlation between FLO11
expression and cell size: FLO11
transcript abundance is highest in the small CLN3-2
haploid and lowest in the large cln3
Δ haploid, the same relationship observed between FLO11
expression and cell size in haploid versus tetraploid cells. Expression of FLO11
was also significantly repressed in the bck2
Δ and eap1
Δ mutants that displayed enlarged cell size at 124% and 137% of WT volume, respectively (). The reduced expression of FLO11
in these large mutants mimics the down-regulation of FLO11
in tetraploids (). The results from the haploid size mutants demonstrate an inverse correlation between FLO11
expression and cell size, and this correlation is independent of ploidy.
To see whether other genes differentially expressed in tetraploid cells were also influenced by cell size, we used quantitative PCR (qPCR) to compare expression of these genes in WT and cln3
Δ haploids. Among the mutants we examined, the cln3
Δ haploid displayed the most pronounced change in cell size and arrested in M-phase with efficiency most similar to WT (; Table S6
). The majority of differentially regulated genes in the tetraploid were regulated in the same direction in the cln3
Δ haploid (, S7
, and S8
), especially the top-ranking genes, i.e., those that displayed the strongest differential expression in tetraploids (). These top-ranking genes likely represent those that respond most robustly to changes in cell size, since the cln3
Δ haploid (185% of WT haploid volume) is still much smaller than the tetraploid (400% of WT haploid volume). Notably, the fold changes in expression levels of the top-ranking genes appeared to correlate with the increase in cell size: they were smaller in the cln3
Δ haploid and larger in the WT tetraploid (Tables S7
), a trend consistent with a functional relationship between cell size and gene expression.
A substantial number of differentially expressed genes in the tetraploid are also differentially expressed in the cln3Δ haploid.
To determine whether the magnitude of change in transcription correlates with the magnitude of change in cell size, we compared expression of the top-ranking size-responsive genes in enlarged haploid mutants and the WT tetraploid. Expression levels in each enlarged strain were measured by qPCR and normalized to those in an isogenic WT haploid. The juxtaposed datasets show a negative correlation (for repressed genes) or a positive correlation (for induced genes) between gene expression levels and cell size ( and ; Table S9
). The results indicate that the differential regulation of these genes is enhanced with increasing cell size. This observation strongly supports our cell size–transcription hypothesis and the idea that incremental changes in cell size can be sensed by the cell and lead to incremental transcriptional responses.
Differential regulation of the top-ranking size-repressed genes correlates with cell size.
Differential regulation of the top-ranking size-induced genes correlates with cell size.
Gene Ontology and Transcription Factor Motifs Reveal Pathways That Mediate Size-Dependent Gene Regulation
GO analysis indicated that many of the genes repressed by large cell size are regulated by the mating and the filamentation mitogen-activated protein kinase (MAPK) pathways (Table S3
; see complete GO analysis results in Dataset S1
). Analysis of transcription factor binding motifs also suggested that these MAPK pathways mediate differential gene regulation in response to cell size: the binding motifs of Dig1 and Ste12, transcription factors that function in both pathways, were significantly enriched (Table S10
; Dataset S2
). Ste12 is a transcriptional activator crucial for mating and filamentation 
. Dig1 mediates transcriptional repression in both cellular processes by inhibiting the activity of Ste12 
. When either of the MAPK pathways is active, Ste12 is phosphorylated by the MAPK and released from inhibition by Dig1.
We compared transcription in haploid and tetraploid cells of a different strain background (S288c) to see whether the mating pathway, which is conserved among different species of yeast 
, was affected by cell size in this background as it is in Σ1278b. The genes downstream of the mating pathway were differentially repressed in the S288c tetraploid (), suggesting that the effect of cell size on the mating pathway could be a general characteristic in yeast. We also constructed isogenic haploid and tetraploid S288c cells expressing FLO11 
and found that expression of FLO11
was significantly repressed in this tetraploid strain (), a result consistent with our finding in Σ1278b.
Differential down-regulation of the top-ranking size-repressed genes also occurs in the S288c strain background.
Although genes up-regulated in large cells were significantly enriched for those containing the binding motifs of Ace2, Swi5, Rfx1, and Yap7 in their promoters (Dataset S2
), this group gave no obvious clues concerning the molecular pathway causing their differential regulation (Text S2
). Moreover, when the transcription of this group of genes in haploids and tetraploids of the S288c background was compared, there was no difference in the levels of their expression (data not shown). This difference between the two yeast strains probably reflects the many regulatory differences between them 
The Mating and the Filamentation Pathways Contribute to Differential Gene Expression in Response to Changes in Cell Size
To understand the roles of the mating and the filamentation MAPK pathways in mediating size-dependent gene regulation, we disrupted signaling in these pathways in enlarged cells by making mutations in key pathway components. In the absence of the transcriptional repressor Dig1, several size-repressed genes were less repressed in large cells and became insensitive or much less responsive to size enlargement (). The reduced effect of cell size on gene expression in the absence of Dig1 shows that this transcription factor is involved in gene regulation by cell size. Assessment of the role of the MAPK pathways required a double mutant lacking both Fus3 and Kss1, MAPKs of the mating pathway and the filamentation pathway, respectively. The double mutant was necessary as these MAPKs have partially overlapping functions as transcriptional repressors 
. The kss1
Δ double mutation reduced the effect of enlarged cell size on the transcription of downstream genes (). Results from the dig1
Δ and kss1
Δ mutants suggest that reduced activities in the mating and the filamentation pathways contribute to differential gene regulation in large cells.
Cell size alters gene expression through the mating and the filamentation pathways.
Cell size homeostasis has been intensively studied and shown to be controlled by a complex coordination of cell growth and cell division 
. However, the functional significance of this intricate and efficient maintenance of cell size has not been addressed. Moreover, when the genome is duplicated, as in polyploids, the accompanying increase in cell size is maintained upon cell division. Such whole genome duplication has been invoked to explain evolution, development, and diseases 
, with little attention to the transcriptional consequences of the enlarged cell size that accompanies polyploidy.
To explore the functional relationship between enlarged cell size and gene expression, we first profiled the transcriptomes of isogenic Σ1278b haploid and tetraploid strains by RNA-seq. This strategy revealed a more complete catalog of the genes influenced by ploidy than was possible to achieve in a previous study 
. Genes encoding cell-surface-related proteins were overrepresented, suggesting an effect of cell size on gene expression. The top-ranking genes whose transcription was down-regulated by an increase in genome size were also down-regulated in another laboratory strain, S288c, that has significant physiological differences from Σ1278b 
. The fact that these genes behaved similarly in two different strains supported the hypothesis that cell size was responsible for the transcriptional effects. This cell size–transcription hypothesis was plausible because the volume of yeast cells increases proportionally with ploidy 
. Moreover, as the cell surface area relative to cell volume decreases with increasing cell size, the reduction in cell surface area could alter gene expression by affecting cell-surface-related signaling molecules and impairing molecular transport across the plasma membrane.
The cell size–transcription hypothesis was supported by an independent assessment that measured gene expression in haploid mutants that make large cells. These large haploid mutants also showed a causal relationship between cell size and gene transcription, indicating the existence of a size-sensing mechanism that alters transcription independently of ploidy. An alternative model, in which polyploidy and size-altering mutations change transcription of size-regulated genes, whose altered expression then enlarges cell size, is unlikely. The set of size-regulated genes we identified do not themselves regulate cell size 
. Moreover, the haploid size mutations and tetraploidy enlarge cell size by different mechanisms 
and affect different cellular processes 
. The similar transcriptional effects observed in these physiologically different contexts of increased cell size and the enhancing of transcriptional changes with increasing cell size support our hypothesis that cell size sensing is involved in differential transcriptional regulation.
Our analysis of the transcriptional data suggests that the cell size signal may be transmitted by the mating and the filamentation MAPK pathways. Genes regulated by these pathways in Σ1278b were preferentially down-regulated in large cells and composed the most significant category in GO analysis. A previous study did not detect differential regulation of these genes, as the strains employed (MATa
/alpha S288c) were inactive for mating and filamentation 
. We showed that in S288c strains of a suitable mating type and with a greater difference in ploidy, genes downstream of the mating pathway were also repressed. Genetic disruptions in the mating and the filamentation MAPKs as well as their common downstream transcriptional repressor DIG1
reduced the effect of cell size on target gene expression. These results confirmed a decrease in activities of the mating and the filamentation pathways in large cells. The switch-like dual functions of the MAPKs 
and the positive feedback loops involving downstream transcription factors 
likely exacerbate differences in pathway activity between the active state (in small cells) and the inactive state (in large cells). These attributes of the pathways may account for a nonlinear transcriptional response to changes in cell size. Consequently, the magnitude of change in gene expression did not appear to correlate with cell size or cell surface area in a simple fashion (Table S9
Although the exact signal that initiates a size-dependent change in transcription is not known, these MAPK pathways have an architecture that is well suited to transmitting a signal from the cell surface to the nucleus. In both MAPK pathways, plasma-membrane-bound G proteins recruit and activate the MAPKKK upon stimulation by the mating pheromone or by nutrient starvation. Through a series of further protein–protein interactions, the MAPKKK in turn activates downstream kinases including the MAPK, which then translocates from the cellular periphery to the nucleus to induce gene expression 
. An enlarged cell size could affect one or more of the molecular events in the process of pathway activation. Because the nuclear size is proportional to overall cell size in yeast 
, both the nuclear surface and the cell surface experience a reduction in area relative to the enclosed volume in large cells. The reduced relative nuclear surface area could impair translocation of the MAPKs. The reduced relative cell surface area could affect the interactions between plasma-membrane-bound and cytoplasmic components in the pathway.
In summary, we showed that polyploidy-associated differential gene regulation is largely caused by an increase in cell size. The newly uncovered regulatory relationship between cell size and gene expression suggests that the uniformity of cell size in unicellular organisms and within tissues in multicellular organisms could be necessary to maintain the homeostasis of transcription. Our finding also suggests that cells monitor their geometric properties (i.e., size and shape) and adjust transcription accordingly. These physical features have not been typically considered a regulatory factor in cellular biology, especially in gene expression studies. In metazoans, the control of gene expression by cell size could contribute to the altered development of large or polyploid cells in normal tissues or to the aberrant physiology of tumor cells.
Materials and Methods
Yeast Growth Conditions
Yeast strains used in this study are listed in . Strains L6437 (WT MATa haploid) and L6440 (WT MATa tetraploid) were grown in synthetic complete medium plus 2% glucose at 30°C until mid-log phase for transcriptome profiling by RNA-seq. Cells were handled with caution to minimize passaging in order to avoid aneuploidy in the tetraploid. For cell cycle arrest using nocodazole, cultures were inoculated at low density in yeast extract peptone dextrose (YPD) plus 1% DMSO from overnight precultures and incubated at 30°C. After a few hours, the cultures in exponential phase were diluted to ~0.15 O.D.600 in prewarmed fresh medium and incubated for another 30 min. Nocodazole was added to a final concentration of 15 µg/ml to arrest cell cycle for 3 h. Enrichment of cells in M-phase was monitored by SWI5 RNA transcript abundance and by counting the percentage of large budded cells with DAPI-stained nuclei at the mother–bud junction. To compare gene expression levels in isogenic strains at different ploidies, cells were cultured in YPD at 30°C until mid-log phase. Cells were collected by centrifugation for RNA extraction and microscopy.
Yeast strains used in this study.
Preparation of cDNA Libraries for Sequencing
Total RNA was extracted from yeast cultures in mid-log phase with acidic phenol. After enrichment of poly(A) RNA (Qiagen Oligotex mRNA kit), the resultant mRNA was processed for cDNA library construction and sequencing as previously described 
. The libraries were sequenced for 36 cycles on Illumina Genome Analyzer 2 using the standard protocol.
Mapping Algorithm for RNA-Seq Reads
Reads were mapped to the Σ1278b genome using the Bowtie alignment software (version 0.10.0; http://bowtie-bio.sourceforge.net/index.shtml
). Reads were either mapped uniquely (bowtie –solexa-quals -k 1 -m 1 –best –strata -p 2 –strandfix) or multiply (bowtie –solexa-quals -k 100 -m 100 –best –strata -p 2 –strandfix). We used unique mappings only to look at differential gene expression. Multiple mappings were used to assess how many reads aligned to TY elements, the rDNA cluster, and other repetitive sequences; a read that mapped to n
genomic locations was assigned a weight of 1/n
, and the “number of reads” mapping to a repetitive element was the sum of the weights of the hits in that element. The complete RNA-seq data are available at the Gene Expression Omnibus (http://www.ncbi.nlm.nih.gov/geo/
) repository with the accession number GSE19685.
Enriching Differentially Expressed Candidate Genes in the RNA-Seq Data
Mapped reads were stored in the David K. Gifford group's in-house ChIP/RNA-seq database and analyzed with the code provided in DifferentialExpression.java. This code performs the following procedure on each annotated ORF in the Σ1278b genome. (1) Determine the total number of uniquely mapped reads in the haploid and tetraploid experiments. (2) Determine the number of uniquely mapped reads on both strands in the ORF in the haploid and tetraploid experiments. (3) Compute frequency_haploid
(haploid count for gene)/(total haploid count) and frequency_tetraploid
(tetraploid count for gene)/(total tetraploid count). (4) Use frequency_haploid to compute a p
-value for the observed reads in the tetraploid experiment using a binomial model given the frequency in the haploid experiment. That is
The CMF is the cumulative mass function (the discrete equivalent of a cumulative distribution function) and is the sum of the probabilities for all counts less than or equal to the observed count. This is the p
-value for the haploid observation given the tetraploid observation. (5) Compute the p
-value for the tetraploid observation given the haploid observation. (6) Retain genes with p
The worksheets “sample pair A” and “sample pair B” in Dataset S1
show the results.
Identification of Genes Transcriptionally Affected by Ploidy in the RNA-Seq Data
Based on read counts of known silenced genes (hypoxia response and sporulation specific), a threshold of 15 was set as the minimal expressed level. In total, 5,613 genes were considered expressed and constituted the “background gene list” for subsequent GO analysis on Saccharomyces Genome Database (SGD). The list of differentially expressed genes with read count of 15 or greater provided the set of candidate genes.
The differentially regulated candidate genes were sorted by fold change after gene read counts had been normalized by the total number of reads in each dataset. Equal numbers of the top-ranking candidates from the two haploid–tetraploid replicates were compared, and the overlapping candidates were identified as differentially regulated. The numbers of top-ranking candidates from the replicates were selected to obtain a sufficient number of overlapping genes for GO analysis while ensuring that the overlap between replicates was highly statistically significant (p<e−10) by hypergeomtric test using MATLAB (MathWorks).
Binding Motif Analysis for Genes Disproportionally Expressed in the Tetraploid
Given the list of differentially expressed genes, we scanned the upstream promoter region for all motifs published 
. For each gene, we determined the promoter region as the region (1) extending 50 bp downstream of the annotated coding start site and (2) extending upstream of the annotated coding start site to the first annotated feature (ORF, Ty, tRNA, etc.) or 5 kb away. Features that overlap the transcription start site are not used as the first upstream feature.
For each motif, we determined the numerical score of the log-likelihood matrix match against all genes as the best match of the motif to the upstream promoter. Each log-likelihood matrix has a positive maximum score representing the highest possible value that the matrix can assign to a DNA sequence seen while scanning; as these are log-likelihood matrices, a score of zero indicates that the sequence in question matches the background model as well as it matches the matrix. The motif-scanning code then tested cutoff scores between 0.3 and 1.0 (in increments of 0.05) times the maximum possible score for the motif to determine the score to use to call a motif “match”; the score cutoff used was the one that produced the most significant result when comparing the motif frequency in genes up-regulated in tetraploid against motif frequency in all genes using a binomial test. We performed a similar process on the down-regulated genes.
We filtered the results to include only motifs that showed a p-value of 0.005 or less, were found in at least 25% of up-/down-regulated genes, and for which the fold change in frequency was at least 1.5. We used these filters to retain motifs mostly likely to be relevant to the lists of identified genes.
shows the results of the motif scanning.
Quantifying Expression Levels of Specific Genes by PCR
Total RNA extracted using acidic phenol was processed for cDNA synthesis using the QuantiTect Reverse Transcription Kit (Qiagen). Expression levels were measured on an Applied Biosystems 7500 Real-Time PCR System with SYBR Green in Absolute Quantification mode following manufacture's procedure. Unless specified, we used the abundance of ACT1 transcript to normalize expression levels of genes of interest. The representation of ACT1 transcript in total RNA is constant in all strains used in this study. Statistical treatment (unpaired t test for two-tailed p-values) of qPCR data was performed using GraphPad Prism (GraphPad Software).
Measurement of Cell Size
Cells were fixed in 3.7% formaldehyde at 4°C overnight and digested with a mixture of zymolyase and glusulase in the presence of 1.2 M sorbitol citrate to relieve aggregation. Microscopy images of more than 50 cells per strain were analyzed using ImageJ (United States National Institutes of Health). In experiments involving cell cycle arrest with nocodazole, cell size was calculated from the measured width and length of both mother and bud of large budded cells, assuming rotational symmetry about the long axis. For actively cycling cultures, cell size was calculated from the measured width and length of the mother of budded cells. Statistical comparison of cell size was performed using Student's t test.
Construction of Isogenic MATaa Diploids
Haploid strains with suitable genotypes were transformed with a plasmid encoding the HO endonuclease inducible by galatose to enable mating-type switching. Transformed strains were pre-grown overnight in synthetic complete drop-out medium supplemented with 0.1% glucose. After sufficient washing with water, cells were resuspended in synthetic complete drop-out medium plus 2% galactose and grown for approximately one doubling time to enable mating-type switching before plating on YPD agar. Mating-type-switched candidates were passaged multiple times on YPD to ensure loss of the HO-encoding plasmid prior to mating-type assessment by auxotrophic marker complementation or a pheromone-dependent growth inhibition assay.
Results for the RNA-seq comparison between the transcriptomes of haploid and tetraploid cells. In the worksheets “sample pair A” and “sample pair B,” read counts of annotated ORFs in the two pairs of RNA-seq datasets are listed. Also included are p-values for calling differentially expressed genes and the sample total read counts. The worksheet “gene lists” shows genes in the “background list” for GO analysis and the lists of genes differentially regulated in tetraploids sorted by average fold change in the RNA-seq datasets. The remaining worksheets show the complete GO term search results from SGD for genes differentially expressed in tetraploids. GO terms for biological process, molecular function, and cellular compartment are listed separately.
(2.33 MB XLS)
Overrepresented transcription factor motifs in the promoters of differentially regulated genes. Worksheet “Dig1, Ste12, Mcm1” shows the likelihood of these transcription factors binding to the promoters of interest. The remaining worksheets show motifs found to be significantly enriched in specific gene sets.
(0.05 MB XLS)
Regulation of disproportionally expressed genes in the tetraploid is not correlated with stages in the mitotic cell cycle.
(0.07 MB DOC)
Cellular compartment GO terms for genes repressed in the tetraploid.
(0.03 MB DOC)
Biological process GO terms for genes repressed in the tetraploid.
(0.03 MB DOC)
Molecular function GO terms for genes repressed in the tetraploid.
(0.03 MB DOC)
GO terms for genes induced in the tetraploid.
(0.03 MB DOC)
Mitotic arrest efficiency, measured as percentages of arrested cells, in experiments shown in .
(0.03 MB DOC)
Expression levels of genes down-regulated in the WT tetraploid in the cln3Δ haploid.
(0.06 MB DOC)
Expression levels of genes up-regulated in the WT tetraploid in the cln3Δ haploid.
(0.06 MB DOC)
Relationship between cell size/surface area and gene expression analyzed with linear regression.
(0.05 MB DOC)
The binding motifs of Dig1 and Ste12 are overrepresented in the promoters of genes repressed in the Σ1278b tetraploid.
(0.05 MB DOC)
Cell size mutants incompatible with cell size–transcription analysis.
(0.03 MB DOC)
Discussion on genes up-regulated in large cells.
(1.08 MB DOC)
We are grateful to members of the Fink lab, Andreas Hochwagen, Hannah Blitzblau, Angelika Amon, Alexi Gornanov, Tony Lee, David Weinberg, Whitehead Genome Technology Core, Stephen Bell, and Frank Solomon for reagents, technical assistance, and helpful discussion. We thank the Boone lab, Pellman lab, and Brian Chin for providing yeast strains. Also thanks to Claire Tseng, Stacie Bumgarner, Hannah Blitzblau, and Gerben Vader for critiquing the manuscript.
|MAPK||mitogen-activated protein kinase|
|SGD||Saccharomyces Genome Database|
|YPD||yeast extract peptone dextrose|
1. Cook M, Tyers M. Size control goes global. Curr Opin Biotechnol. 2007;18:341–350. [PubMed] 2. Lee H. O, Davidson J. M, Duronio R. J. Endoreplication: polyploidy with purpose. Genes Dev. 2009;23:2461–2477. [PubMed] 3. Ganem N. J, Storchova Z, Pellman D. Tetraploidy, aneuploidy and cancer. Curr Opin Genet Dev. 2007;17:157–162. [PubMed] 4. Kasahara M. The 2R hypothesis: an update. Curr Opin Immunol. 2007;19:547–552. [PubMed] 5. Semon M, Wolfe K. H. Consequences of genome duplication. Curr Opin Genet Dev. 2007;17:505–512. [PubMed] 6. Galitski T, Saldanha A. J, Styles C. A, Lander E. S, Fink G. R. Ploidy regulation of gene expression. Science. 1999;285:251–254. [PubMed] 7. Andalis A. A, Storchova Z, Styles C, Galitski T, Pellman D, et al. Defects arising from whole-genome duplications in Saccharomyces cerevisiae. Genetics. 2004;167:1109–1121. [PubMed] 8. Storchova Z, Breneman A, Cande J, Dunn J, Burbank K, et al. Genome-wide genetic analysis of polyploidy in yeast. Nature. 2006;443:541–547. [PubMed] 9. Dowell R. D, Ryan O, Jansen A, Cheung D, Agarwala S, et al. Genotype to phenotype: a complex problem. Science. 2010;328:469. [PubMed] 10. Liu H, Styles C. A, Fink G. R. Saccharomyces cerevisiae S288c has a mutation in FLO8, a gene required for filamentous growth. Genetics. 1996;144:967–978. [PubMed] 11. Mortazavi A, Williams B. A, McCue K, Schaeffer L, Wold B. Mapping and quantifying mammalian transcriptomes by RNA-Seq. Nat Methods. 2008;5:621–628. [PubMed] 12. Wang Z, Gerstein M, Snyder M. RNA-Seq: a revolutionary tool for transcriptomics. Nat Rev Genet. 2009;10:57–63. [PMC free article] [PubMed] 13. Di Talia S, Skotheim J. M, Bean J. M, Siggia E. D, Cross F. R. The effects of molecular noise and size control on variability in the budding yeast cell cycle. Nature. 2007;448:947–951. [PubMed] 14. Jorgensen P, Nishikawa J. L, Breitkreutz B. J, Tyers M. Systematic identification of pathways that couple cell growth and division in yeast. Science. 2002;297:395–400. [PubMed] 15. Bishop A. C, Ubersax J. A, Petsch D. T, Matheos D. P, Gray N. S, et al. A chemical switch for inhibitor-sensitive alleles of any protein kinase. Nature. 2000;407:395–401. [PubMed] 16. Goranov A. I, Cook M, Ricicova M, Ben-Ari G, Gonzalez C, et al. The rate of cell growth is governed by cell cycle stage. Genes Dev. 2009;23:1408–1422. [PubMed] 17. Lo W. S, Dranginis A. M. FLO11, a yeast gene related to the STA genes, encodes a novel cell surface flocculin. J Bacteriol. 1996;178:7144–7151. [PMC free article] [PubMed] 18. Lambrechts M. G, Bauer F. F, Marmur J, Pretorius I. S. Muc1, a mucin-like protein that is regulated by Mss10, is critical for pseudohyphal differentiation in yeast. Proc Natl Acad Sci U S A. 1996;93:8419–8424. [PubMed] 19. Roberts R. L, Fink G. R. Elements of a single MAP kinase cascade in Saccharomyces cerevisiae mediate two developmental programs in the same cell type: mating and invasive growth. Genes Dev. 1994;8:2974–2985. [PubMed] 20. Cook J. G, Bardwell L, Kron S. J, Thorner J. Two novel targets of the MAP kinase Kss1 are negative regulators of invasive growth in the yeast Saccharomyces cerevisiae. Genes Dev. 1996;10:2831–2848. [PubMed] 21. Tedford K, Kim S, Sa D, Stevens K, Tyers M. Regulation of the mating pheromone and invasive growth responses in yeast by two MAP kinase substrates. Curr Biol. 1997;7:228–238. [PubMed] 22. Bennett R. J, Johnson A. D. Mating in Candida albicans and the search for a sexual cycle. Annu Rev Microbiol. 2005;59:233–255. [PubMed] 23. Madhani H. D, Styles C. A, Fink G. R. MAP kinases with distinct inhibitory functions impart signaling specificity during yeast differentiation. Cell. 1997;91:673–684. [PubMed] 24. Dolznig H, Grebien F, Sauer T, Beug H, Mullner E. W. Evidence for a size-sensing mechanism in animal cells. Nat Cell Biol. 2004;6:899–905. [PubMed] 25. Tzur A, Kafri R, LeBleu V. S, Lahav G, Kirschner M. W. Cell growth and size homeostasis in proliferating animal cells. Science. 2009;325:167–171. [PMC free article] [PubMed] 26. Nurse P. Genetic control of cell size at cell division in yeast. Nature. 1975;256:547–551. [PubMed] 27. Johnston G. C, Pringle J. R, Hartwell L. H. Coordination of growth with cell division in the yeast Saccharomyces cerevisiae. Exp Cell Res. 1977;105:79–98. [PubMed] 28. Moseley J. B, Mayeux A, Paoletti A, Nurse P. A spatial gradient coordinates cell size and mitotic entry in fission yeast. Nature. 2009;459:857–860. [PubMed] 29. Seeley E. S, Kato M, Margolis N, Wickner W, Eitzen G. Genomic analysis of homotypic vacuole fusion. Mol Biol Cell. 2002;13:782–794. [PMC free article] [PubMed] 30. Han B. K, Aramayo R, Polymenis M. The G1 cyclin Cln3p controls vacuolar biogenesis in Saccharomyces cerevisiae. Genetics. 2003;165:467–476. [PubMed] 31. Cosentino G. P, Schmelzle T, Haghighat A, Helliwell S. B, Hall M. N, et al. Eap1p, a novel eukaryotic translation initiation factor 4E-associated protein in Saccharomyces cerevisiae. Mol Cell Biol. 2000;20:4604–4613. [PMC free article] [PubMed] 32. Chial H. J, Stemm-Wolf A. J, McBratney S, Winey M. Yeast Eap1p, an eIF4E-associated protein, has a separate function involving genetic stability. Curr Biol. 2000;10:1519–1522. [PubMed] 33. Cook J. G, Bardwell L, Thorner J. Inhibitory and activating functions for MAPK Kss1 in the S. cerevisiae filamentous-growth signalling pathway. Nature. 1997;390:85–88. [PubMed] 34. Kohler T, Wesche S, Taheri N, Braus G. H, Mosch H. U. Dual role of the Saccharomyces cerevisiae TEA/ATTS family transcription factor Tec1p in regulation of gene expression and cellular development. Eukaryot Cell. 2002;1:673–686. [PMC free article] [PubMed] 35. Borneman A. R, Leigh-Bell J. A, Yu H, Bertone P, Gerstein M, et al. Target hub proteins serve as master regulators of development in yeast. Genes Dev. 2006;20:435–448. [PubMed] 36. Harbison C. T, Gordon D. B, Lee T. I, Rinaldi N. J, Macisaac K. D, et al. Transcriptional regulatory code of a eukaryotic genome. Nature. 2004;431:99–104. [PMC free article] [PubMed] 37. Zeitlinger J, Simon I, Harbison C. T, Hannett N. M, Volkert T. L, et al. Program-specific distribution of a transcription factor dependent on partner transcription factor and MAPK signaling. Cell. 2003;113:395–404. [PubMed] 38. Chen R. E, Thorner J. Function and regulation in MAPK signaling pathways: lessons learned from the yeast Saccharomyces cerevisiae. Biochim Biophys Acta. 2007;1773:1311–1340. [PMC free article] [PubMed] 39. Jorgensen P, Edgington N. P, Schneider B. L, Rupes I, Tyers M, et al. The size of the nucleus increases as yeast cells grow. Mol Biol Cell. 2007;18:3523–3532. [PMC free article] [PubMed] 40. Neumann F. R, Nurse P. Nuclear size control in fission yeast. J Cell Biol. 2007;179:593–600. [PMC free article] [PubMed] 41. MacIsaac K. D, Wang T, Gordon D. B, Gifford D. K, Stormo G. D, et al. An improved map of conserved regulatory sites for Saccharomyces cerevisiae. BMC Bioinformatics. 2006;7:113. [PMC free article] [PubMed] 42. Torres E. M, Sokolsky T, Tucker C. M, Chan L. Y, Boselli M, et al. Effects of aneuploidy on cellular physiology and cell division in haploid yeast. Science. 2007;317:916–924. [PubMed] 43. Spellman P. T, Sherlock G, Zhang M. Q, Iyer V. R, Anders K, et al. Comprehensive identification of cell cycle-regulated genes of the yeast Saccharomyces cerevisiae by microarray hybridization. Mol Biol Cell. 1998;9:3273–3297. [PMC free article] [PubMed] 44. Epstein C. B, Cross F. R. Genes that can bypass the CLN requirement for Saccharomyces cerevisiae cell cycle START. Mol Cell Biol. 1994;14:2041–2047. [PMC free article] [PubMed]