Dynamics of oscillatory phenotypes in S. cerevisiae reveal a network of genome-wide transcriptional oscillators

Shwe L Chin; Ian M Marcus; Robert R Klevecz; Caroline M Li

doi:10.1111/j.1742-4658.2012.08508.x

. Author manuscript; available in PMC: 2013 Mar 1.

Published in final edited form as: FEBS J. 2012 Feb 27;279(6):1119–1130. doi: 10.1111/j.1742-4658.2012.08508.x

Dynamics of oscillatory phenotypes in S. cerevisiae reveal a network of genome-wide transcriptional oscillators

Shwe L Chin, Ian M Marcus, Robert R Klevecz, Caroline M Li ^*

PMCID: PMC3368069 NIHMSID: NIHMS354329 PMID: 22289124

Abstract

Genetic and environmental factors are well-studied influences on phenotype; however, time is a variable that is rarely considered when studying changes in cellular phenotype. Time-resolved microarray data revealed genome-wide transcriptional oscillation in a yeast continuous culture system with ~2 and ~4 h periods. We mapped the global patterns of transcriptional oscillations into a 3D map to represent different cellular phenotypes of redox cycles. This map shows the dynamic nature of gene expression in that transcripts are ordered and coupled to each other through time and concentration space. Although cells differed in oscillation periods, transcripts involved in certain processes were conserved in a deterministic way. When oscillation period lengthened, the peak to trough ratio of transcripts increased and the fraction of cells in the unbudded (G0/G1) phase of the cell division cycle increased. Decreasing the glucose level in the culture media was one way to increase the redox cycle, possibly from changes in metabolic flux. The period may be responding to lower glucose levels by increasing the fraction of cells in G1 and reducing S-phase gating so that cells can spend more time in catabolic processes. Our results support that gene transcripts are coordinated with metabolic functions and the cell division cycle.

Keywords: dynamics, microarray, oscillation, genome-wide, S. cerevisiae

Introduction

Oscillations are found throughout nature, from particle physics to stellar and planetary motion and in biological clocks. Day length on Earth, another oscillation, has increased at least three-fold since the first photoautotrophs appeared. To adapt to the change in day length, the cell’s chemistry has evolved to coordinate with daily changes in the external environment [1].

Saccharomyces cerevisiae grown continuously under aerobic growth at constant pH, aeration, temperature, agitation, and dilution rate is an excellent model system for studying biological clocks. The cell population spontaneously displays an ultradian (<24 h) redox cycle that is easily monitored with the rise and fall of dissolved oxygen (DO) in the media [2]. The expression of mRNA and metabolites peak at three distinct phases (respiratory, early reductive, and late reductive phase) of each redox cycle while S-phase is gated through the early reductive phase [3–7]. Even though growth parameters may be at steady state, the cells are not, due to oscillations in metabolites and transcripts. The cells grow densely to less than one cell diameter apart [1] and are thought to self synchronize through cell-to-cell signaling with small molecules such as hydrogen sulfide and acetaldehyde [8–9].

The period of the redox cycle varies with mutations, drug treatments or altered growth conditions [10–14]. We found that the same batch of culture changed in dilution rate or changed to media with 40% less glucose resulted in doubling in period length. The system may have stable limit cycles separated by unstable limit cycles where perturbations, such as changes in dilution rate or changes in glucose concentration, determine if the dynamics end up on a different stable cycle. In this study, the genome-wide organization of two redox oscillations, ~2 and ~4 h, was compared over time in the S. cerevisiae strain CEN.PK113-7D through microarray and cell cycle analysis to better understand the molecular details of phenotypic change.

Results

Previous work from our laboratory showed that transcripts peaked either at the respiratory or reductive phase through three redox cycles [3, 6]. In this work, more densely spaced samples from a single cycle of CEN.PK113-7D with ~2 and ~4 h oscillation periods were taken for a more detailed analysis. These results were compared with our earlier microarray data from strain IFO0233 oscillating with ~40 min period before adding phenelzine [4]. To maintain consistency for comparative analysis, the same growth conditions used for IFO0233 [4] were applied to CEN.PK113-7D. Only the dilution rate (medium flow rate/culture volume) was varied to maintain stable and different oscillations. During each cycle, 32 samples were taken for the ~2 h CEN.PK113-7D oscillation data set, and 49 were taken for the ~4 h oscillation data set (Fig. 1A, C). In contrast, previously published data had fewer than 12 samples for each cycle. To avoid sample bias, samples were randomized after mRNA isolation and then reordered during data analysis. This higher resolution data set provided more detailed information on when transcripts were made or degraded.

Fig. 1 — Expression of three transcripts that peak at the respiratory, early reductive, and late reductive phase. The black line represents DO levels. DO was set to 100% before cells were inoculated in the fermentor. The actual sample time was adjusted so that the first sample starts at 1 h. (A) Red circles represent each sample taken from CEN.PK113-7D oscillating with ~2 h period. A total of 32 samples were taken, one every 4 min. (B) mRNA expression levels of *AAH1* (blue circle), *RFA2* (orange circle), and *GND2* (yellow circle) from cells oscillating with ~2 h period. (C) Red circles represent each sample from CEN.PK113-7D oscillating with ~4 h period with one sample taken every 5 min (49 samples total). (D) *AAH1* (blue circle), *RFA2* (orange circle), and *GND2* (yellow circle) represent the expression levels for each sample in CEN.PK113-7D oscillating with ~4 h period. (A and C) The colored shading represents the respiratory (blue), early reductive (orange) and late reductive (yellow) phase.

Gene annotation analysis revealed that transcripts involved in certain processes (P<0.001) peaked at the three major phases (respiratory, early reductive, and late reductive) (Table 1) in cell populations with differing redox oscillation periods. Transcripts for genes involved in ribosome biogenesis, RNA metabolism, gene expression and sulfur amino acid metabolism peaked during the respiratory phase. Transcripts that peaked in the early reductive phase participate in DNA-dependent DNA replication and DNA metabolism. Finally, transcripts that peaked in the late reductive phase involved cellular catabolism, glycolysis, autophagy and response to stress and starvation. These findings are consistent with previous reports for IFO0233 with ~40 min period and CEN.PK122 with ~4–5 h period [3, 6]. The annotation analysis was also used to more clearly define the borders, relative to DO, among the three phases from densely sampled microarray data (Fig. 1A, C). Examples of specific transcripts that peaked at the respiratory, early reductive, and late reductive phases regardless of the oscillation period and are involved in core processes include AAH1, RFA2, and GND2, respectively (Fig. 1B, D). AAH1 is involved in ribosome biogenesis, RFA2 is involved in DNA-dependent DNA replication, and GND2 is involved in the pentose phosphate pathway for D-glucose catabolism.

Table 1.

Gene Ontology annotation for transcripts that were maximally expressed during the three phases (respiratory, early reductive, and late reductive) in S. cerevisiae oscillating with ~40 min (IFO0233), ~2 h (CEN.PK113-7D), and ~4 h (CEN.PK113-7D) redox periods (P<0.001).

Phase	Period	Biological Processes
Respiratory	~40 min	ribosome biogenesis, RNA metabolic process, gene expression, RNA biosynthetic process, and sulfur amino acid metabolic process
	~2 h	ribosome biogenesis, RNA metabolic process, gene expression, RNA biosynthetic process, and sulfur amino acid metabolic process
	~4 h	ribosome assembly, RNA metabolic process, gene expression, and sulfur amino acid metabolic process

Early Reductive	~40 min	DNA-dependent DNA replication and nitrogen compound metabolic process
	~2 h	cellular biosynthetic process, cellular metabolic process, DNA metabolic process, histone modification, and mitosis
	~4 h	cellular biosynthetic process, DNA-dependent DNA replication, DNA metabolic process, histone modification, and mitosis

Late Reductive	~40 min	proteolysis, response to stress, and proteolysis involved in cellular protein catabolic process
	~2 h	autophagy, celluar response to stress and starvation, cellular catabolic process, and lipid biosynthetic process
	~4 h	autophagy, cellular response to stress and starvation, and cellular catabolic process

Open in a new tab

To identify DNA-binding transcriptional regulators that might be controlling gene expression during the three major phases of the redox cycle, groups of transcripts that peaked at each phase were analyzed for common DNA binding sites [15] of transcriptional regulators (Table 2). The identified DNA-binding transcriptional regulators and the genes to which they bind are listed in Table S1. Transcriptional regulators involved in the sulfur amino acid metabolic process (Met4p, Met32p, and Met31p) and regulation of ribosomal protein transcription (Fhl1p) were associated with binding sites related to transcripts that peaked in the respiratory phase. The transcriptional regulators associated with transcripts that peaked at the early reductive phase included Mbp1p, Gcn4p, Swi6p, and Swi4p. Mbp1p, Swi4p, and Swi6p are involved in regulating the G1 to S-phase of the mitotic cell division cycle [16], and Gcn4p responds to amino acid starvation by activating amino acid biosynthetic genes [17–18]. The transcriptional regulators Msn2p and Rpn4p bind to sequences associated with transcripts that peaked at the late reductive phase. Msn2p is involved in cellular response to glucose starvation and response to stress [19–20]. Rpn4p is involved in expression of proteosome genes and genes involved in catabolic processes in response to various stresses [21]. The presence of DNA-binding sites associated with specific transcriptional regulators suggests when the regulators are active during the redox cycle and gives insights into how the cell tightly controls the expression of transcripts.

Table 2.

Transcriptional regulators that bind significantly to DNA binding site sequences were deduced from the sequences of transcript groups peaking at the respiratory, early reductive, or late reductive phase.

DNA-binding Transcriptional Regulator	# Binding Site in ~40 min period (P value)	# Binding Site in ~2 h period (P value)	# Binding Site in ~4 h period (P value)	DNA-Binding Site Sequence
Respiratory
Met4p	7 (8.82x10⁻⁵)	6 (3.46x10⁻³)	6 (1.13x10⁻⁴)	RMMAWSTGKSGYGSC
Met32p	12 (1.36x10⁻⁵)	11 (8.77x10⁻³)	9 (1.65x10⁻⁴)	AAACTGTGG, ANTGTGGCGY, and KGTGGCK
Met31p	6 (2.84x10⁻⁴)	5 (4.51x10⁻⁴)	5 (4.78x10⁻⁴)	AAACTGTGG and DNNGTGGCK
Fhl1p	59 (1.85x10⁻⁹)	45 (1.32x10⁻²)	34 (7.19x10⁻⁴)	GACGCA, GACGCAV, and TGTAYGGRTG
Rap1p	44 (9.59x10⁻⁶)			CACCCRWACA, CAYCCRTRCA, KGGTGTACNG, and WRMACCCATACAYY

Early Reductive
Mbp1p		42 (1.56x10⁻¹¹)	73 (6.39x10⁻⁹)	ACGCGT, ACGCGTCA, and DCGCGH
Gcn4p		24 (1.42x10⁻³)		ARTGACTCW, RTGASTCAY, and TGASTCA
Swi4p		20 (2.11x10⁻⁶)		CACGAAAA, CGCSAAA, and DCGCGAAW
Swi6p		26 (7.20x10⁻⁷)		CACGAAAA and CGCGAAA

Late Reductive
Msn2p		19 (1.47x10⁻³)	19 (7.25x10⁻⁶)	AAGGGG, MAGGGGSGG, and RGGGG
Rpn4p		52 (4.73x10⁻⁵)	40 (6.65x10⁻⁴)	CGCCACCC, GGTGGCAAA, and RGTGGCG

Open in a new tab

The number of transcripts that peaked in each of the three phases was determined (Table 3) to further compare molecular details of phenotypic change. As the period lengthened, more genes peaked at the early reductive phase (from 252 to 1979), whereas fewer genes peaked at the late reductive phase (from 3335 to 2142). Not all genes were conserved among transcripts that peaked with various periods (Table 3). Up to ~90% transcripts were conserved between ~2 and ~4 h oscillations in all phases. For the three oscillation periods compared, up to ~60%, ~20%, and ~70% of transcripts occurred similarly in the respiratory, early reductive, and late reductive phase, respectively. Although transcripts involved in certain processes peaked at a specific phase (Table 1) and transcripts showed oscillatory behavior, a specific gene transcript may not necessarily peak at the same phase for cells oscillating with differing periods, supporting the importance of looking at global transcriptional trends.

Table 3.

Number of conserved genes and comparisons for ~40 min (IFO0233), ~2 h (CEN.PK113-7D), and ~4 h (CEN.PK113-7D) oscillation periods in each phase.

Phase	Genes in IFO0233 with ~40 min period	Genes in CEN.PK with ~2 h period	Genes in CEN.PK with ~4 h period	Conserved genes between ~2 h and ~4 h period	Conserved genes among ~40 min, ~2 h, and ~4 h
Respiratory	1624	1817	1090	985	654
Early reductive	252	494	1979	428	38
Late reductive	3335	2900	2142	1881	1481

Total	5211	5211	5211	3294	2173

Open in a new tab

Singular value decomposition (SVD) analysis of microarray data ([22]) provides a nonbiased method to break down a matrix of genome-wide transcriptional variations over time. It is a method to reduce dimensionality by taking 5,211 down to orthogonal vectors. To directly compare two cycles of differing periods, time was converted to phase angle based on the DO (Fig. 2A, B). SVD revealed three major eigengenes from ~40 min, ~2 h and ~4 h redox oscillations that display nonlinear genome-wide dynamic behavior of gene expression data having similarities to an attractor surface. The first eigengene is used to normalize the data. The next three eigengenes (Fig. 2A and 2B) show that more transcripts peak at the respiratory phase for ~2 h redox cycles than in ~4 h redox cycles. In contrast, ~4 h redox cycles display more early reductive phase peaked transcripts, as supported in Table 3, possibly from overactive gene regulation of cells spending more time in the early reductive phase. The SVD analysis shows a genome-wide analysis of the entire time-course microarray data.

Eigengenes 2, 3, and 4 was determined by SVD analysis and plotted into a 3D map (Fig. 2C) to compare cells undergoing ~40 min [4], ~2 h, and ~4 h oscillation periods. This map visually represents the temporal and global cyclical nature of transcripts in a synchronously oscillating cell population. The three eigengenes circle around the steady state, yet never reach the steady state due to oscillations that fluctuate through time out of phase with one another. In addition, the oscillations are non-linearly coupled to each other. K-means cluster analysis was used to determine that 3 clusters gave the most distinct patterns within the gene expression data. The centroids of these clusters peaked at the respiratory, early reductive, and late reductive phase for cells with ~2 and ~4 h oscillation periods (Fig. S1). These three transcript peaks were also reported for ~40 min transcriptional oscillations [3] under similar growth conditions and in 4–5 h transcriptional oscillations in diploid CEN.PK grown under different growth conditions [6].

To elucidate the molecular dynamics of phenotypic changes (~2 and ~4 h oscillations), transcriptional and cell division cycle parameters were compared. The average peak/trough ratio of transcripts increased as the oscillation period increased, even with low amplitude (peak/trough<2) and high amplitude (peak/trough>2) transcripts (Fig. 3). To determine where cells were in the cell cycle, the fraction of cells in G0/G1 was measured by Sytox green staining of DNA for CEN.PK cells oscillating with ~2 and ~4 h periods (Fig. 4A, B). Fig. 4C show that cell cycle oscillations for ~2 h oscillations are more difficult to see than in ~4 h redox oscillations. However, time course cell cycle analysis in Fig. 4A does reveal low amplitude oscillations in the cell cycle for ~2 h redox oscillations. The cell cycle oscillations correspond with redox cycles arguing that redox oscillations regulate cell cycle processes, more apparent in longer period redox oscillations.

Fig. 3 — Comparison of average peak to trough ratio of transcripts from cells oscillating with different periods. (A) Average peak to trough ratio of mRNA transcripts for cells with ~40, ~120, and ~240 min dissolved oxygen period. (B) Comparison of gene groups from the ~40 min oscillation data set with peak to trough ratios less than 2 (black bars, low amplitude genes) and peak to trough ratios greater than 2 (gray bars, high amplitude genes).

At any time, there were always some cells in S-phase, even at the respiratory phase (Fig. 4). Throughout the cycle, the fraction of cells in G0/G1 varied from 59–78% for the haploid yeast strain CEN.PK113-7D with ~2 h oscillation period, whereas the fraction of cells in G0/G1 varied from 66–94% for the population with ~4 h oscillation period. The diploid yeast strain CEN.PK122 also had more cells in G0/G1 for populations with longer period oscillations (Fig. S2). In the diploid strain, cells in G0/G1 varied from 68–80% for ~2 h oscillation period and 54–94% for ~4 h oscillation period. The larger amplitude transcriptional oscillations and higher fraction of cells in G0/G1 for the longer period oscillation suggest better transcriptional and cell cycle synchrony.

We were also interested in controlling the causes of seemingly spontaneous fluctuations in redox period to better understand these phenotypic changes. Previously, the dilution rate was found to increase the period, as reported for the S. cerevisiae strain S288C [10], CEN.PK122 [23] and IFO0233 [24]. We found that the initial dilution rate can vary yet give rise to the same oscillation period once the cells settle to a stable phenotype. CEN.PK113-7D mostly oscillated with ~2 h redox period with a dilution rate of ~0.08 h⁻¹ and ~8 h doubling time. Cells that oscillated with ~4 h redox period were generally more dilute, preferred a dilution rate of ~0.06 h⁻¹ and had ~12 h doubling time. However, cells with an initial dilution rate of ~0.06 h⁻¹ and ~0.08 h⁻¹ were capable of oscillating with ~2 or ~4 hour periods, respectively. Once a stable oscillation was achieved, decreasing the initial dilution rate resulted in a longer period, whereas increasing the dilution rate resulted in a shorter oscillation period. After changing the dilution rates, the oscillation stabilized to the new period.

To determine which component of the media was limiting to period changes, we reduced the amount of each ingredient of the media by 40% (Fig. 5), starting from a ~2 h oscillation period. Within 24 h, the period did not change in the presence of 40% less ammonium sulfate, calcium chloride, magnesium sulfate, or mineral solution (FeSO₄, ZnSO₄, CuSO₄, and MnCl₂). The oscillation was gradually lost when the media was changed to 40% less yeast extract. When medium with 40% less phosphate was added to cells oscillating with ~2 h period, the results were not consistent with period changes (Fig. S3A). Three out of four experiments did not result in period lengthening, yet period increased in one experiment for an unknown reason. Substituting the media with 40% less glucose (from 1.75% to 1% glucose) in three independent experiments resulted in consistent period changes from ~2 to ~4 h (Fig. 5). As the media was changed back to 1.75% glucose, the period returned to ~2 h from ~4 (Fig. S3B). When the media was reduced from 1% to 0.8% glucose, the cell density decreased and oscillations went away, possibly due to lack of cell signaling. Perhaps intercellular signal strength reached below a threshold for maintaining population synchrony. Overall, the concentration of glucose appears to be an important component for consistently controlling doubling in period length under our growth conditions.

Discussion

Growing evidence has identified redox cycles in yeast synchronous cultures, asynchronous cultures, and single-cells. In the yeast strain IFO0233 oscillating with a ~40 min period, each redox cycle has three major clusters of mRNA transcripts that peak at the respiratory, early reductive, and late reductive phase [3]. A similar pattern of transcriptional oscillations was reported in the yeast strain CEN.PK122 oscillating with a 4–5 h period [6]. Principal component analysis not only revealed genome-wide oscillatory behavior in synchronous cultures of yeast and mammalian cells but also found oscillations in asynchronous yeast cultures [25]. Quantitative analysis of RNA by fluorescence in situ hybridization suggested that metabolic cycles occur at the single-cell level in unsynchronized cultures of yeast strain S288C growing in limited glucose or phosphate [26–27]. In addition to yeast, redox cycles are found in other organisms such as plants, mosquitos, mice and humans. In these organisms, oscillations in expression of cytochrome c oxidase (an oxidative gene) and NADH dehydrogenase (a reductive gene) are out of phase with each other [28]. Thus, redox cycles may be more ubiquitous than originally considered.

fWe seek to determine the molecular details of phenotypic change for S. cerevisiae growing in continuous cultures, as a way of understanding the basis of cellular oscillators. Cells in the continuous culture system resemble cells growing densely in a natural environment, such as on the surface of grape skins or oak leaves. In addition, these cells constantly receive oxygen and nutrients to maintain a constant density. Cells in the continuous culture system self-synchronize and undergo redox cycles. Fourier transform analysis of earlier microarray expression data showed that almost all transcripts oscillate in the yeast strain IFO0233 growing with a ~40 min period [4]. More than half of the transcripts in CEN.PK122 oscillating with a 4–5 h period were reported to be periodic with a periodicity algorithm [6], whereas almost all genes from CEN.PK122 were found to be periodic from an analysis by digital signal processing [29]. Since previous studies showed periodic oscillations, we monitored transcriptional level fluctuations through time at high resolution for one cycle in synchronous cultures of CEN.PK stably oscillating with ~2 and ~4 h periodsm (Fig. 1). Under both conditions, transcripts peaked at three major phases along the redox cycle: respiratory, early reductive, and late reductive phase. Common DNA-binding transcriptional regulators for different oscillation periods were identified in these three major phases (Table 2), giving information on the order in which transcriptional regulators are active during the cycle. Annotation analysis showed that cellular phenotype with a ~40 min, ~2 h, and ~4 h oscillation still maintained core regulatory elements or core biological processes in the three major groups of transcripts (Table 1), providing insights into ordered transcriptional events needed for maintaining stable redox oscillations.

The order of core processes gives insights into when certain cellular processes occur. Mitochondria are active at the respiratory phase where the electron transport chain generates large amounts of energy. The levels of ATP are low during this phase [30–31], supporting that the cell uses ATP during the respiratory phase. Large amounts of energy are likely produced and consumed in the respiratory phase to allow energy intense processes to occur, such as ribosome biogenesis [32–33]. Transcripts involved in RNA metabolic process, gene expression, and sulfur amino acid metabolic process peak at the respiratory phase (Table 1) possibly to accumulate materials for translation and cell growth needed during the budding of S-phase. Transcripts involved in the DNA metabolic processes and DNA replication peak during the early reductive phase (Table 1). Cell cycle studies support that the cell population moves towards DNA-replication at this phase (Fig. 4). In budding yeast, cells grow and increase in size until the daughter cell buds off. Cytokinesis occurs primarily during the late reductive phase, as determined from an increase in cell concentration from our lab and others [34]. The late reductive phase also allows the cell population to respond to stress/starvation. Storage carbohydrates such as glycogen and trehalose accumulate and break down at the late reductive phase in previously reported studies [10, 13, 31]. In addition, cells undergo glycolysis and other cellular catabolic processes such as autophagy during this phase. The transcript that peak at the late reductive phase are involved in response to stress/starvation and catabolic processes (Table 1). Future assessment of protein levels and activity would provide stronger evidence of when certain processes occur and identify the enzymes involved.

Early studies suggested that the cell cycle is somehow regulated in continuous cultures of S. cerevisiae because a constant fraction of budded cells is present in each cycle [10]. The population is heterogeneous with mother and daughter cells with different generation times, dependent on the cell’s size and age [10, 23]. The phase when most cells are in G1 for CEN.PK113-7D with ~4 h redox oscillation agrees with earlier reported autonomous oscillations of the same yeast strain [34]. Consistent with IFO0233 at ~40 min period [3] and CEN.PK122 with a 4–5 h oscillation [6], a distinct population of cells gate through the S-phase at the early reductive phase. The fraction of unbudded cells (G1) starts to decrease when S-phase starts to increase near the end of the respiratory phase and into the early reductive phase (Fig. 4). Cells were hypothesized to prefer performing DNA synthesis at the reductive phase as an evolutionary mechanism to avoid oxidative damage [3, 6]. In contrast, the S-phase for the yeast strain DBY12007 was found to peak at the respiratory phase when grown with a dilution rate of 0.1 h⁻¹ but peak at the reductive phase with dilution rate of 0.133 h⁻¹ [35]. Cells that undergo DNA replication during the respiratory phase may be more prone to oxidative damage and greater spontaneous mutation rates as observed in strains with cell cycle mutants [14]. Since S-phase in our cell population increase toward the early reductive phase and not the respiratory phase, the cells may be less prone to genetic mutations. Our yeast strain and growth conditions may be ideal for maintaining genetically stable populations. Although the period of the cell cycle is consistent with the redox period, it is not clear what signals progression of the cell cycle in the continuous culture system.

Understanding the causes of redox cycle changes can give further insights into the basis of phenotypic changes. Drug treatments, mutations, and altered growth conditions can all influence the oscillation period [10–14]. No one clear mechanism explains what causes period changes. It is also not clear why oscillation periods for wild-type yeast strains can vary under identical growth conditions with a preferred period. In the event of a perturbation, the phenotype becomes unstable and can change. For example, the dilution rate was reported to alter the DO oscillation period in S. cerevisiae [10, 23–24]. We also observed that dilution rate changes affect the oscillation period in CEN.PK113-7D. Thus, adjusting the media rate entering the fermentor is one way to adjust the oscillation period, supporting evidence of a tuneable attractor [36].

Changes of the incoming media from 1.75% to 1% glucose consistently lead to a doubling of period length (Fig. 5). With less carbon source in the media, period lengthening may allow the cell population to spend more time conducting catabolic processes and less time in cell division and population growth, thus giving the population functional advantage for survival. Cell cycle analyses support that a longer oscillation period results in a higher fraction of cells that accumulate in G0/G1, thus blocking the G1 to S-phase progression. Not surprisingly, we found active transcriptional regulators that are involved in regulating the progression from G1 to S-phase (Mbp1p, Swi4p, and Swi6p) present during the early reductive phase for CEN.PK113-7D. Signals that prevent the S-phase may allow cells time to recover from energy deficiency, thus resulting in synchronization of the population. In addition, growth rate differences could contribute to the timing of redox cycles. Thus, our results suggest that the cell’s response to glucose levels and energetics contribute to doubling in period length in redox cycles.

We are also interested in modeling the temporal organization of different phenotypes of redox cycles to help understand the mechanics of period changes. Changes in ultradian cycles also give clues to how cells have differentiated and evolved to adapt to environmental changes [1]. In our previous work, genome-wide transcriptional oscillations were used to map a 3D attractor surface circling around the steady-state through time after perturbation of the yeast strain IFO0233 with a drug that increased the redox period. The maps of transcriptional changes through phase space resembled a modified Rossler attractor [4] that supports a dynamic system of gene transcripts that is sensitive to initial conditions, such as dilution rate, and leads to phenotypic changes, such as period changes. The chaotic characteristics of continuous self-synchronized cultures of yeast with regard to respiratory dynamics have also been previously demonstrated [37].

In the current studies, a 3D attractor surface was plotted using SVD analysis from CEN.PK stably oscillating with a ~2 or ~4 h period (Fig. 2C). The maps can also be drawn from transcripts peaking at the respiratory, early reductive, and late reductive phase. Changes in the map surface provide a simplified representation of the temporal organization of different phenotypes. As with IFO0233 with ~40 min period, cellular phenotype can be viewed as an attractor consisting of an organized network of mRNA transcripts coupled to each other in a nonlinear way to maintain temporal organization and stability. The model provides a dynamic perspective that can be applied to predict the paths of phenotypic changes, such as those that occur during cell differentiation or disease progression. Recognizing that networks of genome-wide oscillators could manifest in other systems encourages one to design experiments that consider changes over time, rather than changes before and after an effect. Without consideration of changes over time, experimental replicates may artificially appear stochastic. Thus, the model demonstrates the importance of taking time into consideration when applying statistical analysis to dynamic systems.

Materials and Methods

Fermentor Conditions

The prototrophic yeast strain CEN.PK113-7D (MATa URA3 HIS3 LEU2 TRP1 MAL2-8 SUC2) and CEN.PK122 (Mata/MATαURA3/URA3 HIS3/HIS3 LEU2/LEU2 TRP1/TRP1 MAL2-8/MAL2-8 SUC2/SUC2) used in this study were kindly provided by Dr. Peter Kotter (Institute for Microbiology, Johann Wolfgang Goethe-Universitat Frankfurt, Germany). The fermentor equipment and conditions were previously described [4, 38] with minor changes. CEN.PK grown overnight in 20 ml YPD at 275 rpm and 30°C, was inoculated into the fermentor. In the fermentor with automated control systems, the pH, agitation speed, aeration, and temperature were maintained at 4.0, 750 rpm, 150 ml/min, and 30 °C, respectively. Cells were grown in either a B. Braun Biotech Biolab CP (Aylesbury, Buckinghamshire, UK) in 650 ml or a New Brunswick Scientific Bioflo 110 Fermentor/Bioreactor (Edison, NJ, USA) in 850 ml volume. Continuous flow was started after 24–30 h by pumping in fermentor media (19.25 g/l D(+)glucose monohydrate, 5 g/l (NH₄)₂SO₄, 2 g/l KH₂PO₄, 0.5 g/l MgSO₄, 1 g/l yeast extract, 0.2 ml/l antifoam A, 0.1 g/l CaCl₂, 0.02 g/l FeSO₄, 0.01 g/l ZnSO₄, 0.005 g/l CuSO₄, 0.001 g/l MnCl₂, and 1 ml/l 75% H₂SO₄) at a dilution rate of 0.086 h⁻¹. Media with excess cells was pumped out of the fermentor to maintain a constant fermentor volume. The media was maintained at pH 4.0 with 2.5 N NaOH. The dilution rate was adjusted to 0.06–0.086 h⁻¹ to maintain the period.

Samples for Microarray Data Analysis

The pellet from 0.5 ml cells was immediately frozen in liquid nitrogen and stored at -80 °C. Samples from one cycle with a small overlap into the next cycle were taken every 4 min from cells with ~2 h DO period (32 samples total) and every 5 min from cells with ~4 h period (49 samples total). RNA was purified from the cells and DNA was digested using the RiboPure^™ Yeast kit (Ambion, Austin, TX, USA) according to the manufacturer’s instructions, except that cells were lysed in a minibead beater (BioSpec Products, Bartlesville, OK, USA) rather than by vortexing. RNA samples were randomized before they were further prepared to avoid artifacts in sample handling. RNA samples for the Affymetrix GeneChip Yeast Genome 2.0 Array (Santa Clara, CA, USA) were processed as described previously [4].

Data Analysis

Data was processed and analyzed as described in [4], and is available in Dataset S1, Dataset S2, and at the National Center for Biotechnology Information/Gene Expression Omnibus web site under accession number GSE30053 (http://www.ncbi.nlm.nih.gov/geo/). K-means cluster analysis was done in Matlab (MathWorks, Natick, MA, USA). Gene Ontology (GO) [39] was used to annotate genes that were maximally expressed in each sample of IFO0233 oscillating with a ~40 min period before phenelzine was added [4], CEN.PK113-7D oscillating with a ~2 h period, and CEN.PK113-7D oscillating with a ~4 h period. The GO database was accessed by Amigo, a term enrichment tool [40]. To identify DNA-binding transcriptional regulators for each phase, DNA-binding sites and associated transcriptional regulators were extracted from Harbison et al. [15]. Identification was processed using MatLab. The most statistically significant transcription regulators were found by determining the probability that the regulators were counted by chance using a hypergeometric probability distribution function that returns the p-value associated for each transcriptional regulator.

Nucleic Acid Stain

Cells were fixed in 70% ethanol and stored at 4 °C. For the ~4 h oscillation, samples were taken every 10 min for a total of 80 samples. For the ~2 h oscillation, samples were taken every 5 min for a total of 50 samples. Cells were pelleted by centrifugation (1 min at 16,000 RCF), and ethanol was removed. The cells were washed and then diluted to 1x10⁷ cells/ml in sheath fluid (NERL Diagnostics, East Providence, RI, USA). Cells were incubated at 50 °C for 1–2 h in 0.25 mg/ml RNase A (Qiagen, Valencia, CA, USA) and then incubated in 1 μM SYTOX green (Invitrogen, Carlsbad, CA, USA) at room temperature. Fluorescence from 100,000 cells was measured on a CyAn ADP 9 color flow cytometer (Dako, Carpinteria, CA, USA) with an excitation wavelength of 488 nm and emission filter of 530/40 nm. FCS Express version 3 (De Novo Software, Los Angeles, CA, USA) was used for cell cycle analysis. The fluorescence vs. side scatter plot was gated so that the cell population with less fluorescence and lower side scatter represent cells in G0/G1.

Supplementary Material

Supp Data S1. Dataset S1.

Microarray data for CEN.PK113-7D with a 2 h redox period. The actual time was adjusted so that sample 1 starts at 1 h.

NIHMS354329-supplement-Supp_Data_S1.xlsx^{(2.7MB, xlsx)}

Supp Data S2. Dataset S2.

Microarray data for CEN.PK113-7D with a 4 h redox period. The actual time was adjusted so that sample 1 starts at 1 h.

NIHMS354329-supplement-Supp_Data_S2.xlsx^{(3.9MB, xlsx)}

Supp Table S1 & Fig S1-S3

Fig. S1. K-means cluster results from gene expression data from CEN.PK113-7D oscillating with ~2 and ~4 h periods. The black line indicates the DO in the media against phase angle. (A) ~2 h and (B) ~4 h oscillation periods show centroids of the three gene clusters that peak at the respiratory (blue line), early reductive (orange line), and late reductive (black line with yellow highlight) phases.

Fig. S2. Cell cycle analysis of CEN.PK122 oscillating with ~2 and ~4 h periods. Cells were stained with SYTOX green. The solid line represents the DO oscillation. (A) Fraction of cells (quantified on right axis) in G0 and G1 (black circles) and fraction of cells in S and G2 (white squares) from 100,000 cells oscillating with ~2 h periods. A total of 75 samples were taken. (B) The DO oscillation was converted to phase angle and the number of cells as compared to DNA content are indicated by histograms. (C) Fraction of cells in G0 and G1 (black circles) and in S and G2 (white squares) for 100,000 cells with ~4 h period. Thirty-eight samples were taken. (D) Time was converted to phase angle and the histogram (number of cells vs. DNA content) for points along the DO are shown for cells with a ~4 h oscillation period.

Fig. S3. Influence of DO oscillations from decreasing phosphate and glucose levels in the media. (A) Arrow indicates when CEN.PK113-7D with ~2 h DO oscillations (line) was changed with media containing of 40% less potassium phosphate. (B) At time 0, CEN.PK113-7D with ~2 h DO was growing in the fermentor medium with 1.75% glucose. The left arrow shows when the fermentor media with 1% glucose was changed. The right array shows when the fermentor media with 1.75% was returned.

Table S1. Genes associated with transcriptional regulators for ~40 min, ~2 h, and ~4 h redox periods.

NIHMS354329-supplement-Supp_Table_S1___Fig_S1-S3.pdf^{(257KB, pdf)}

Acknowledgments

We thank Maricela Covarrubias of the City of Hope Functional Genomics core facility for help with processing samples for the Affymetrix arrays; the City of Hope Analytical Cytometry core facility for assistance with flow cytometry; Matteo Pellegrini and David Casero from University of California, Los Angeles, for aiding in the DNA-binding transcriptional regulator analysis; Douglas Murray from Keio University, Tsuruoka City, Yamagata, Japan for informative discussions; and Keely Walker and Laura De Francesco for help with editing. This research was supported by grant 5R01GM81757 from the National Institute of General Medical Sciences. The authors are responsible for the content which does not necessarily represent the official views of the National Institute of General Medical Sciences or the National Institutes of Health.

abbreviations

DO: dissolved oxygen
GO: gene ontology
SVD: singular value decomposition

Footnotes

Designed research: Robert R. Klevecz and Caroline M. Li

Performed research: Shwe L. Chin and Caroline M. Li

Analyzed data: Shwe L. Chin, Ian M. Marcus, and Caroline Li

Wrote the paper: Caroline M. Li

References

1.Klevecz RR, Li CM. Evolution of the clock from yeast to man by period-doubling folds in the cellular oscillator. Cold Spring Harb Symp Quant Biol. 2007;72:421–9. doi: 10.1101/sqb.2007.72.040. [DOI] [PMC free article] [PubMed] [Google Scholar]
2.Lloyd D, Murray DB, Klevecz RR, Wolf J, Kuriyama H. The Ultradian Clock (~40 min) in Yeast (Saccharomyces cerevisiae) In: Lloyd D, Rossi EL, editors. Ultradian Rhythms from Molecules to Mind: A New Vision of Life. Springer Berlin-Heidelberg; Berlin, Germany: 2008. pp. 11–42. [Google Scholar]
3.Klevecz RR, Bolen J, Forrest G, Murray DB. A genomewide oscillation in transcription gates DNA replication and cell cycle. Proc Natl Acad Sci USA. 2004;101:1200–5. doi: 10.1073/pnas.0306490101. [DOI] [PMC free article] [PubMed] [Google Scholar]
4.Li CM, Klevecz RR. A rapid genome-scale response of the transcriptional oscillator to perturbation reveals a period-doubling path to phenotypic change. Proc Natl Acad Sci USA. 2006;103:16254–9. doi: 10.1073/pnas.0604860103. [DOI] [PMC free article] [PubMed] [Google Scholar]
5.Murray DB, Beckmann M, Kitano H. Regulation of yeast oscillatory dynamics. Proc Natl Acad Sci USA. 2007;104:2241–6. doi: 10.1073/pnas.0606677104. [DOI] [PMC free article] [PubMed] [Google Scholar]
6.Tu BP, Kudlicki A, Rowicka M, McKnight SL. Logic of the yeast metabolic cycle: temporal compartmentalization of cellular processes. Science. 2005;310:1152–8. doi: 10.1126/science.1120499. [DOI] [PubMed] [Google Scholar]
7.Tu BP, Mohler RE, Liu JC, Dombek KM, Young ET, Synovec RE, McKnight SL. Cyclic changes in metabolic state during the life of a yeast cell. Proc Natl Acad Sci USA. 2007;104:16886–91. doi: 10.1073/pnas.0708365104. [DOI] [PMC free article] [PubMed] [Google Scholar]
8.Murray DB, Klevecz RR, Lloyd D. Generation and maintenance of synchrony in Saccharomyces cerevisiae continuous culture. Exp Cell Res. 2003;287:10–5. doi: 10.1016/s0014-4827(03)00068-5. [DOI] [PubMed] [Google Scholar]
9.Sohn HY, Murray DB, Kuriyama H. Ultradian oscillation of Saccharomyces cerevisiae during aerobic continuous culture: hydrogen sulphide mediates population synchrony. Yeast. 2000;16:1185–90. doi: 10.1002/1097-0061(20000930)16:13<1185::AID-YEA619>3.0.CO;2-W. [DOI] [PubMed] [Google Scholar]
10.Porro D, Martegani E, Ranzi BM, Alberghina L. Oscillations in continuous cultures of budding yeast: a segregated parameter analysis. Biotechnol Bioeng. 1988;32:411–7. doi: 10.1002/bit.260320402. [DOI] [PubMed] [Google Scholar]
11.Adams CA, Kuriyama H, Lloyd D, Murray DB. The Gts1 protein stabilizes the autonomous oscillator in yeast. Yeast. 2003;20:463–70. doi: 10.1002/yea.976. [DOI] [PubMed] [Google Scholar]
12.Salgado E, Murray DB, Lloyd D. Some Antidepressant Agents (Li, Monoamine Oxidase Type A Inhibitors) Perturb the Ultradian Clock in Saccharomyces cerevisiae. Biological Rhythm Research. 2002;33:351–361. [Google Scholar]
13.Xu Z, Tsurugi K. Destabilization of energy-metabolism oscillation in the absence of trehalose synthesis in the chemostat culture of yeast. Arch Biochem Biophys. 2007;464:350–8. doi: 10.1016/j.abb.2007.04.029. [DOI] [PubMed] [Google Scholar]
14.Chen Z, Odstrcil EA, Tu BP, McKnight SL. Restriction of DNA replication to the reductive phase of the metabolic cycle protects genome integrity. Science. 2007;316:1916–9. doi: 10.1126/science.1140958. [DOI] [PubMed] [Google Scholar]
15.Harbison CT, Gordon DB, Lee TI, Rinaldi NJ, Macisaac KD, Danford TW, Hannett NM, Tagne JB, Reynolds DB, Yoo J, Jennings EG, Zeitlinger J, Pokholok DK, Kellis M, Rolfe PA, Takusagawa KT, Lander ES, Gifford DK, Fraenkel E, Young RA. Transcriptional regulatory code of a eukaryotic genome. Nature. 2004;431:99–104. doi: 10.1038/nature02800. [DOI] [PMC free article] [PubMed] [Google Scholar]
16.Koch C, Moll T, Neuberg M, Ahorn H, Nasmyth K. A role for the transcription factors Mbp1 and Swi4 in progression from G1 to S phase. Science. 1993;261:1551–7. doi: 10.1126/science.8372350. [DOI] [PubMed] [Google Scholar]
17.Natarajan K, Meyer MR, Jackson BM, Slade D, Roberts C, Hinnebusch AG, Marton MJ. Transcriptional profiling shows that Gcn4p is a master regulator of gene expression during amino acid starvation in yeast. Mol Cell Biol. 2001;21:4347–68. doi: 10.1128/MCB.21.13.4347-4368.2001. [DOI] [PMC free article] [PubMed] [Google Scholar]
18.Thireos G, Penn MD, Greer H. 5′ untranslated sequences are required for the translational control of a yeast regulatory gene. Proc Natl Acad Sci U S A. 1984;81:5096–100. doi: 10.1073/pnas.81.16.5096. [DOI] [PMC free article] [PubMed] [Google Scholar]
19.Martinez-Pastor MT, Marchler G, Schuller C, Marchler-Bauer A, Ruis H, Estruch F. The Saccharomyces cerevisiae zinc finger proteins Msn2p and Msn4p are required for transcriptional induction through the stress response element (STRE) Embo J. 1996;15:2227–35. [PMC free article] [PubMed] [Google Scholar]
20.Gorner W, Durchschlag E, Wolf J, Brown EL, Ammerer G, Ruis H, Schuller C. Acute glucose starvation activates the nuclear localization signal of a stress-specific yeast transcription factor. Embo J. 2002;21:135–44. doi: 10.1093/emboj/21.1.135. [DOI] [PMC free article] [PubMed] [Google Scholar]
21.Mannhaupt G, Schnall R, Karpov V, Vetter I, Feldmann H. Rpn4p acts as a transcription factor by binding to PACE, a nonamer box found upstream of 26S proteasomal and other genes in yeast. FEBS Lett. 1999;450:27–34. doi: 10.1016/s0014-5793(99)00467-6. [DOI] [PubMed] [Google Scholar]
22.Alter O, Brown PO, Botstein D. Singular value decomposition for genome-wide expression data processing and modeling. Proc Natl Acad Sci U S A. 2000;97:10101–6. doi: 10.1073/pnas.97.18.10101. [DOI] [PMC free article] [PubMed] [Google Scholar]
23.Beuse M, Bartling R, Kopmann A, Diekmann H, Thoma M. Effect of the dilution rate on the mode of oscillation in continuous cultures of Saccharomyces cerevisiae. J Biotechnol. 1998;61:15–31. doi: 10.1016/s0168-1656(98)00016-9. [DOI] [PubMed] [Google Scholar]
24.Murray DB, Roller S, Kuriyama H, Lloyd D. Clock control of ultradian respiratory oscillation found during yeast continuous culture. J Bacteriol. 2001;183:7253–9. doi: 10.1128/JB.183.24.7253-7259.2001. [DOI] [PMC free article] [PubMed] [Google Scholar]
25.Tsuchiya M, Wong ST, Yeo ZX, Colosimo A, Palumbo MC, Farina L, Crescenzi M, Mazzola A, Negri R, Bianchi MM, Selvarajoo K, Tomita M, Giuliani A. Gene expression waves. Cell cycle independent collective dynamics in cultured cells. Febs J. 2007;274:2878–86. doi: 10.1111/j.1742-4658.2007.05822.x. [DOI] [PubMed] [Google Scholar]
26.Wyart M, Botstein D, Wingreen NS. Evaluating gene expression dynamics using pairwise RNA FISH data. PLoS Comput Biol. 2010;6:e1000979. doi: 10.1371/journal.pcbi.1000979. [DOI] [PMC free article] [PubMed] [Google Scholar]
27.Silverman SJ, Petti AA, Slavov N, Parsons L, Briehof R, Thiberge SY, Zenklusen D, Gandhi SJ, Larson DR, Singer RH, Botstein D. Metabolic cycling in single yeast cells from unsynchronized steady-state populations limited on glucose or phosphate. Proc Natl Acad Sci U S A. 2010;107:6946–51. doi: 10.1073/pnas.1002422107. [DOI] [PMC free article] [PubMed] [Google Scholar]
28.Ptitsyn AA, Gimble JM. True or false: all genes are rhythmic. Ann Med. 2011;43:1–12. doi: 10.3109/07853890.2010.538078. [DOI] [PubMed] [Google Scholar]
29.Ptitsyn AA, Zvonic S, Gimble JM. Digital signal processing reveals circadian baseline oscillation in majority of mammalian genes. PLOS Comput Biol. 2007;3:e120. doi: 10.1371/journal.pcbi.0030120. [DOI] [PMC free article] [PubMed] [Google Scholar]
30.Xu Z, Tsurugi K. A potential mechanism of energy-metabolism oscillation in an aerobic chemostat culture of the yeast Saccharomyces cerevisiae. Febs J. 2006;273:1696–709. doi: 10.1111/j.1742-4658.2006.05201.x. [DOI] [PubMed] [Google Scholar]
31.Satroutdinov AD, Kuriyama H, Kobayashi H. Oscillatory metabolism of Saccharomyces cerevisiae in continuous culture. FEMS Microbiol Lett. 1992;77:261–7. doi: 10.1016/0378-1097(92)90167-m. [DOI] [PubMed] [Google Scholar]
32.Kressler D, Hurt E, Bassler J. Driving ribosome assembly. Biochimica Et Biophysica Acta-Molecular Cell Research. 2010;1803:673–683. doi: 10.1016/j.bbamcr.2009.10.009. [DOI] [PubMed] [Google Scholar]
33.Strunk BS, Karbstein K. Powering through ribosome assembly. Rna. 2009;15:2083–104. doi: 10.1261/rna.1792109. [DOI] [PMC free article] [PubMed] [Google Scholar]
34.Stowers CC, Robertson JB, Ban H, Tanner RD, Boczko EM. Periodic fermentor yield and enhanced product enrichment from autonomous oscillations. Appl Biochem Biotechnol. 2009;156:59–75. doi: 10.1007/s12010-008-8486-7. [DOI] [PubMed] [Google Scholar]
35.Slavov N, Botstein D. Coupling among growth rate response, metabolic cycle, and cell division cycle in yeast. Mol Biol Cell. 2011;22:1997–2009. doi: 10.1091/mbc.E11-02-0132. [DOI] [PMC free article] [PubMed] [Google Scholar]
36.Murray DB, Lloyd D. A tuneable attractor underlies yeast respiratory dynamics. Biosystems. 2007;90:287–94. doi: 10.1016/j.biosystems.2006.09.032. [DOI] [PubMed] [Google Scholar]
37.Roussel MR, Lloyd D. Observation of a chaotic multioscillatory metabolic attractor by real-time monitoring of a yeast continuous culture. FEBS J. 2007;274:1011–8. doi: 10.1111/j.1742-4658.2007.05651.x. [DOI] [PubMed] [Google Scholar]
38.Klevecz RR, Li CM, Bolen JL. Signal processing and the design of microarray time-series experiments. Methods Mol Biol. 2007;377:75–94. doi: 10.1007/978-1-59745-390-5_4. [DOI] [PubMed] [Google Scholar]
39.Ashburner M, Ball CA, Blake JA, Botstein D, Butler H, Cherry JM, Davis AP, Dolinski K, Dwight SS, Eppig JT, Harris MA, Hill DP, Issel-Tarver L, Kasarskis A, Lewis S, Matese JC, Richardson JE, Ringwald M, Rubin GM, Sherlock G. Gene ontology: tool for the unification of biology. The Gene Ontology Consortium. Nat Genet. 2000;25:25–9. doi: 10.1038/75556. [DOI] [PMC free article] [PubMed] [Google Scholar]
40.Boyle EI, Weng S, Gollub J, Jin H, Botstein D, Cherry JM, Sherlock G. GO::TermFinder--open source software for accessing Gene Ontology information and finding significantly enriched Gene Ontology terms associated with a list of genes. Bioinformatics. 2004;20:3710–5. doi: 10.1093/bioinformatics/bth456. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supp Data S1. Dataset S1.

Microarray data for CEN.PK113-7D with a 2 h redox period. The actual time was adjusted so that sample 1 starts at 1 h.

NIHMS354329-supplement-Supp_Data_S1.xlsx^{(2.7MB, xlsx)}

Supp Data S2. Dataset S2.

Microarray data for CEN.PK113-7D with a 4 h redox period. The actual time was adjusted so that sample 1 starts at 1 h.

NIHMS354329-supplement-Supp_Data_S2.xlsx^{(3.9MB, xlsx)}

Supp Table S1 & Fig S1-S3

Table S1. Genes associated with transcriptional regulators for ~40 min, ~2 h, and ~4 h redox periods.

NIHMS354329-supplement-Supp_Table_S1___Fig_S1-S3.pdf^{(257KB, pdf)}

[R1] 1.Klevecz RR, Li CM. Evolution of the clock from yeast to man by period-doubling folds in the cellular oscillator. Cold Spring Harb Symp Quant Biol. 2007;72:421–9. doi: 10.1101/sqb.2007.72.040. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R2] 2.Lloyd D, Murray DB, Klevecz RR, Wolf J, Kuriyama H. The Ultradian Clock (~40 min) in Yeast (Saccharomyces cerevisiae) In: Lloyd D, Rossi EL, editors. Ultradian Rhythms from Molecules to Mind: A New Vision of Life. Springer Berlin-Heidelberg; Berlin, Germany: 2008. pp. 11–42. [Google Scholar]

[R3] 3.Klevecz RR, Bolen J, Forrest G, Murray DB. A genomewide oscillation in transcription gates DNA replication and cell cycle. Proc Natl Acad Sci USA. 2004;101:1200–5. doi: 10.1073/pnas.0306490101. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R4] 4.Li CM, Klevecz RR. A rapid genome-scale response of the transcriptional oscillator to perturbation reveals a period-doubling path to phenotypic change. Proc Natl Acad Sci USA. 2006;103:16254–9. doi: 10.1073/pnas.0604860103. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R5] 5.Murray DB, Beckmann M, Kitano H. Regulation of yeast oscillatory dynamics. Proc Natl Acad Sci USA. 2007;104:2241–6. doi: 10.1073/pnas.0606677104. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R6] 6.Tu BP, Kudlicki A, Rowicka M, McKnight SL. Logic of the yeast metabolic cycle: temporal compartmentalization of cellular processes. Science. 2005;310:1152–8. doi: 10.1126/science.1120499. [DOI] [PubMed] [Google Scholar]

[R7] 7.Tu BP, Mohler RE, Liu JC, Dombek KM, Young ET, Synovec RE, McKnight SL. Cyclic changes in metabolic state during the life of a yeast cell. Proc Natl Acad Sci USA. 2007;104:16886–91. doi: 10.1073/pnas.0708365104. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R8] 8.Murray DB, Klevecz RR, Lloyd D. Generation and maintenance of synchrony in Saccharomyces cerevisiae continuous culture. Exp Cell Res. 2003;287:10–5. doi: 10.1016/s0014-4827(03)00068-5. [DOI] [PubMed] [Google Scholar]

[R9] 9.Sohn HY, Murray DB, Kuriyama H. Ultradian oscillation of Saccharomyces cerevisiae during aerobic continuous culture: hydrogen sulphide mediates population synchrony. Yeast. 2000;16:1185–90. doi: 10.1002/1097-0061(20000930)16:13<1185::AID-YEA619>3.0.CO;2-W. [DOI] [PubMed] [Google Scholar]

[R10] 10.Porro D, Martegani E, Ranzi BM, Alberghina L. Oscillations in continuous cultures of budding yeast: a segregated parameter analysis. Biotechnol Bioeng. 1988;32:411–7. doi: 10.1002/bit.260320402. [DOI] [PubMed] [Google Scholar]

[R11] 11.Adams CA, Kuriyama H, Lloyd D, Murray DB. The Gts1 protein stabilizes the autonomous oscillator in yeast. Yeast. 2003;20:463–70. doi: 10.1002/yea.976. [DOI] [PubMed] [Google Scholar]

[R12] 12.Salgado E, Murray DB, Lloyd D. Some Antidepressant Agents (Li, Monoamine Oxidase Type A Inhibitors) Perturb the Ultradian Clock in Saccharomyces cerevisiae. Biological Rhythm Research. 2002;33:351–361. [Google Scholar]

[R13] 13.Xu Z, Tsurugi K. Destabilization of energy-metabolism oscillation in the absence of trehalose synthesis in the chemostat culture of yeast. Arch Biochem Biophys. 2007;464:350–8. doi: 10.1016/j.abb.2007.04.029. [DOI] [PubMed] [Google Scholar]

[R14] 14.Chen Z, Odstrcil EA, Tu BP, McKnight SL. Restriction of DNA replication to the reductive phase of the metabolic cycle protects genome integrity. Science. 2007;316:1916–9. doi: 10.1126/science.1140958. [DOI] [PubMed] [Google Scholar]

[R15] 15.Harbison CT, Gordon DB, Lee TI, Rinaldi NJ, Macisaac KD, Danford TW, Hannett NM, Tagne JB, Reynolds DB, Yoo J, Jennings EG, Zeitlinger J, Pokholok DK, Kellis M, Rolfe PA, Takusagawa KT, Lander ES, Gifford DK, Fraenkel E, Young RA. Transcriptional regulatory code of a eukaryotic genome. Nature. 2004;431:99–104. doi: 10.1038/nature02800. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R16] 16.Koch C, Moll T, Neuberg M, Ahorn H, Nasmyth K. A role for the transcription factors Mbp1 and Swi4 in progression from G1 to S phase. Science. 1993;261:1551–7. doi: 10.1126/science.8372350. [DOI] [PubMed] [Google Scholar]

[R17] 17.Natarajan K, Meyer MR, Jackson BM, Slade D, Roberts C, Hinnebusch AG, Marton MJ. Transcriptional profiling shows that Gcn4p is a master regulator of gene expression during amino acid starvation in yeast. Mol Cell Biol. 2001;21:4347–68. doi: 10.1128/MCB.21.13.4347-4368.2001. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R18] 18.Thireos G, Penn MD, Greer H. 5′ untranslated sequences are required for the translational control of a yeast regulatory gene. Proc Natl Acad Sci U S A. 1984;81:5096–100. doi: 10.1073/pnas.81.16.5096. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R19] 19.Martinez-Pastor MT, Marchler G, Schuller C, Marchler-Bauer A, Ruis H, Estruch F. The Saccharomyces cerevisiae zinc finger proteins Msn2p and Msn4p are required for transcriptional induction through the stress response element (STRE) Embo J. 1996;15:2227–35. [PMC free article] [PubMed] [Google Scholar]

[R20] 20.Gorner W, Durchschlag E, Wolf J, Brown EL, Ammerer G, Ruis H, Schuller C. Acute glucose starvation activates the nuclear localization signal of a stress-specific yeast transcription factor. Embo J. 2002;21:135–44. doi: 10.1093/emboj/21.1.135. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R21] 21.Mannhaupt G, Schnall R, Karpov V, Vetter I, Feldmann H. Rpn4p acts as a transcription factor by binding to PACE, a nonamer box found upstream of 26S proteasomal and other genes in yeast. FEBS Lett. 1999;450:27–34. doi: 10.1016/s0014-5793(99)00467-6. [DOI] [PubMed] [Google Scholar]

[R22] 22.Alter O, Brown PO, Botstein D. Singular value decomposition for genome-wide expression data processing and modeling. Proc Natl Acad Sci U S A. 2000;97:10101–6. doi: 10.1073/pnas.97.18.10101. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R23] 23.Beuse M, Bartling R, Kopmann A, Diekmann H, Thoma M. Effect of the dilution rate on the mode of oscillation in continuous cultures of Saccharomyces cerevisiae. J Biotechnol. 1998;61:15–31. doi: 10.1016/s0168-1656(98)00016-9. [DOI] [PubMed] [Google Scholar]

[R24] 24.Murray DB, Roller S, Kuriyama H, Lloyd D. Clock control of ultradian respiratory oscillation found during yeast continuous culture. J Bacteriol. 2001;183:7253–9. doi: 10.1128/JB.183.24.7253-7259.2001. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R25] 25.Tsuchiya M, Wong ST, Yeo ZX, Colosimo A, Palumbo MC, Farina L, Crescenzi M, Mazzola A, Negri R, Bianchi MM, Selvarajoo K, Tomita M, Giuliani A. Gene expression waves. Cell cycle independent collective dynamics in cultured cells. Febs J. 2007;274:2878–86. doi: 10.1111/j.1742-4658.2007.05822.x. [DOI] [PubMed] [Google Scholar]

[R26] 26.Wyart M, Botstein D, Wingreen NS. Evaluating gene expression dynamics using pairwise RNA FISH data. PLoS Comput Biol. 2010;6:e1000979. doi: 10.1371/journal.pcbi.1000979. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R27] 27.Silverman SJ, Petti AA, Slavov N, Parsons L, Briehof R, Thiberge SY, Zenklusen D, Gandhi SJ, Larson DR, Singer RH, Botstein D. Metabolic cycling in single yeast cells from unsynchronized steady-state populations limited on glucose or phosphate. Proc Natl Acad Sci U S A. 2010;107:6946–51. doi: 10.1073/pnas.1002422107. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R28] 28.Ptitsyn AA, Gimble JM. True or false: all genes are rhythmic. Ann Med. 2011;43:1–12. doi: 10.3109/07853890.2010.538078. [DOI] [PubMed] [Google Scholar]

[R29] 29.Ptitsyn AA, Zvonic S, Gimble JM. Digital signal processing reveals circadian baseline oscillation in majority of mammalian genes. PLOS Comput Biol. 2007;3:e120. doi: 10.1371/journal.pcbi.0030120. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R30] 30.Xu Z, Tsurugi K. A potential mechanism of energy-metabolism oscillation in an aerobic chemostat culture of the yeast Saccharomyces cerevisiae. Febs J. 2006;273:1696–709. doi: 10.1111/j.1742-4658.2006.05201.x. [DOI] [PubMed] [Google Scholar]

[R31] 31.Satroutdinov AD, Kuriyama H, Kobayashi H. Oscillatory metabolism of Saccharomyces cerevisiae in continuous culture. FEMS Microbiol Lett. 1992;77:261–7. doi: 10.1016/0378-1097(92)90167-m. [DOI] [PubMed] [Google Scholar]

[R32] 32.Kressler D, Hurt E, Bassler J. Driving ribosome assembly. Biochimica Et Biophysica Acta-Molecular Cell Research. 2010;1803:673–683. doi: 10.1016/j.bbamcr.2009.10.009. [DOI] [PubMed] [Google Scholar]

[R33] 33.Strunk BS, Karbstein K. Powering through ribosome assembly. Rna. 2009;15:2083–104. doi: 10.1261/rna.1792109. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R34] 34.Stowers CC, Robertson JB, Ban H, Tanner RD, Boczko EM. Periodic fermentor yield and enhanced product enrichment from autonomous oscillations. Appl Biochem Biotechnol. 2009;156:59–75. doi: 10.1007/s12010-008-8486-7. [DOI] [PubMed] [Google Scholar]

[R35] 35.Slavov N, Botstein D. Coupling among growth rate response, metabolic cycle, and cell division cycle in yeast. Mol Biol Cell. 2011;22:1997–2009. doi: 10.1091/mbc.E11-02-0132. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R36] 36.Murray DB, Lloyd D. A tuneable attractor underlies yeast respiratory dynamics. Biosystems. 2007;90:287–94. doi: 10.1016/j.biosystems.2006.09.032. [DOI] [PubMed] [Google Scholar]

[R37] 37.Roussel MR, Lloyd D. Observation of a chaotic multioscillatory metabolic attractor by real-time monitoring of a yeast continuous culture. FEBS J. 2007;274:1011–8. doi: 10.1111/j.1742-4658.2007.05651.x. [DOI] [PubMed] [Google Scholar]

[R38] 38.Klevecz RR, Li CM, Bolen JL. Signal processing and the design of microarray time-series experiments. Methods Mol Biol. 2007;377:75–94. doi: 10.1007/978-1-59745-390-5_4. [DOI] [PubMed] [Google Scholar]

[R39] 39.Ashburner M, Ball CA, Blake JA, Botstein D, Butler H, Cherry JM, Davis AP, Dolinski K, Dwight SS, Eppig JT, Harris MA, Hill DP, Issel-Tarver L, Kasarskis A, Lewis S, Matese JC, Richardson JE, Ringwald M, Rubin GM, Sherlock G. Gene ontology: tool for the unification of biology. The Gene Ontology Consortium. Nat Genet. 2000;25:25–9. doi: 10.1038/75556. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R40] 40.Boyle EI, Weng S, Gollub J, Jin H, Botstein D, Cherry JM, Sherlock G. GO::TermFinder--open source software for accessing Gene Ontology information and finding significantly enriched Gene Ontology terms associated with a list of genes. Bioinformatics. 2004;20:3710–5. doi: 10.1093/bioinformatics/bth456. [DOI] [PMC free article] [PubMed] [Google Scholar]

PERMALINK

Dynamics of oscillatory phenotypes in S. cerevisiae reveal a network of genome-wide transcriptional oscillators

Shwe L Chin

Ian M Marcus

Robert R Klevecz

Caroline M Li

Abstract

Introduction

Results