Comparative analysis of phenotype and gene expression in Saccharomyces cerevisiae grown in solid versus liquid culture formats

Abstract

Background

Saccharomyces cerevisiae is a pivotal model organism in biological research. In practice, the choice between solid and liquid formats of the same culture medium is often based on convenience, with an implicit assumption that the physical format (solid vs. liquid) of the medium has minimal impact on core physiological outcomes. This study systematically tested this assumption by comparing the phenotypic and transcriptional profiles of yeast grown in identical YPD medium prepared as either agar plates or shaken liquid cultures.

Methods

Yeast S288C was pre-cultured in YPD and then inoculated onto solid YPD agar plates or into liquid YPD medium, both of which were incubated in the same incubator with two-tier system at 29°C for 7 days to reach stationary phase, with liquid cultures shaken at 150 rpm. We analyzed cell viability, size, morphology, and the distribution of quiescent (Q) and non-quiescent (NQ) sub-populations. Extracellular vesicles (EVs) were isolated and characterized. Transcriptomic differences were assessed via RNA sequencing and validated by qRT-PCR for key pathways.

Results

After 7 days, total biomass and overall cell viability were similar between the two culture formats. However, cells from liquid culture exhibited larger sizes. Solid culture yielded a higher proportion and viability of Q cells, although these cells were smaller. Contrary to expectations, solid cultures produced a higher concentration of larger EVs. Transcriptomic analysis, coupled with KEGG enrichment, revealed significant differential expression, with solid cultures showing marked upregulation of genes involved in ribosome biogenesis, secondary metabolite biosynthesis, and carbon metabolism. The upregulation of genes involved in key pathways was further validated by qRT-PCR.

Conclusions

The culture format (solid vs. liquid) profoundly influences the phenotypic and transcriptional state of S. cerevisiae, even when the chemical medium composition is identical. These differences, likely driven by variations in nutrient/oxygen gradients and mechanical environment, underscore the importance of carefully considering culture format in experimental design and interpretation, particularly in studies of stationary-phase physiology, metabolism, and extracellular vesicle biology.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12866-026-04855-6.

Introduction

The budding yeast Saccharomyces cerevisiae (S. cerevisiae) is a cornerstone model organism for eukaryotic biology [1]. Its utility extends from fundamental studies of cell cycle and metabolism to modeling human disease processes [2–4]. A critical, yet often underexplored, aspect of yeast experimental design is the choice of culture format - solid agar plates versus liquid broth. While the chemical composition of the medium is rightly scrutinized, the physical format is frequently selected based on technical convenience, with an implicit expectation that it does not fundamentally alter biological outcomes.

However, solid and liquid formats create vastly different microenvironments. Liquid cultures, typically aerated by shaking, offer relatively homogeneous access to nutrients and oxygen. In contrast, growth on solid surfaces leads to structured colonies characterized by internal gradients of nutrients, oxygen, pH, and metabolic waste products [5, 6]. These gradients are known to influence microbial physiology, stress responses, and population heterogeneity [7–9]. Despite this knowledge, a systematic comparison of how these inherent differences between solid and liquid formats of the same rich medium affect global yeast physiology is lacking.

During our study, we observed that experimental outcomes derived from medium with the same composition in solid and liquid cultures were sometimes inconsistent. We hypothesized that the distinct microenvironments of solid and liquid culture formats would lead to significant phenotypic and transcriptional divergence. We chose a 7-day incubation period to allow cultures to enter stationary phase, enabling the examination of long-term adaptive responses and population heterogeneity relevant to studies of aging, fermentation, and microbial survival [10–13]. Using the haploid laboratory strain S288C, we conducted an integrated analysis of cell morphology, sub-population dynamics, extracellular vesicle production, and genome-wide gene expression. Our findings reveal that culture format is a major determinant of yeast physiology, challenging the notion of format equivalence and providing a resource for more deliberate experimental design.

Results

Culture format influences cell size but not total biomass after long-term incubation

In order to figure out the impact of different culture format, yeast was cultivated for 7 days on solid YPD agar plates or in shaken liquid YPD medium. We strictly controlled the experimental conditions, including the composition of the YPD medium, equal culture volume of either liquid or solid formats, the initial inoculation amount, the choice of incubator etc. (see M&M section) The haploid S288C strain used carries a non-functional HO endonuclease gene [14], preventing mating-type switching and ensuring population homogeneity in terms of ploidy under these conditions. When yeast was cultured in liquid medium, it was easy to follow the growth of the cells by optical density (OD) changes, but it was difficult to follow it in solid medium [15]. Therefore, we detected the growth of the cells by measuring the biomass (wet weight) after 7 days of culture. As shown in Fig. 1A, the yeast biomass produced in liquid culture or solid agar plates is not significantly different. We also detected the cell viability of the cells by Trypan blue staining, no significant difference in the survival of the cells were observed (Fig. 1B). In addition, we measured the cell concentration and diameter, and found that cells grown in liquid medium had a lower concentration (Fig. 1C), but were significantly larger than those grown in solid medium (Fig. 1D and E). The results of transmission electron microscopy observations were consistent with the results of particle size measurements, cells cultured by liquid medium were larger (Fig. 1F). This inverse relationship between cell size and concentration explains the comparable total biomass.

Fig. 1.

Phenotypic differences of yeast cells cultured in different media. A Yeast Biomass. B Cell viability of yeast in different media. C Cell concentration analysis. D Cell diameter distribution. E Cell size analysis. F Cell morphology observed by electron microscopy. LM: liquid medium. SM: solid medium. Data are presented as mean ± SD of at least 3 biologically independent experiments, the blue dots in the bar charts represent biological replicates. Unpaired student t-test was used for the statistical analysis. ns: non-significant, **p ≤ 0.01

Solid culture format favors the accumulation of viable quiescent cells

After 7 days of cultivation, the yeast cultures had entered a stable phase characterized by population heterogeneity. As expected from previous studies [10, 12], we observed the accumulation of distinct sub-populations of quiescent (Q) and non-quiescent (NQ) cells. NQ cells are composed of older, asynchronous cells that rapidly lose their reproductive capacity. These cells are typically larger, have looser cytoplasm, thinner cell walls, and lower buoyancy density [10–13]. Q cells are mostly unbudded and uniformly sized daughter cells with thick cell walls and high buoyancy density [10–13]. We used Percoll density gradient centrifugation to separate Q cells and NQ cells [10, 12]. As shown in Fig. 2A, two distinct cell layers were successfully separated using the Percoll density separation system. We observed that the distribution patterns of Q cells and NQ cells differed between liquid and solid culture formats. Specifically, we collected cells from the different density layers and, upon microscopic examination, confirmed that the upper layer predominantly contained NQ cells, while the lower layer was primarily composed of Q cells (Fig. 2B). The concentration and particle size of the cells in each layer were measured. The results indicated no significant difference in the number of Q cells and NQ cells obtained after 7 days of liquid culture. However, after 7 days of solid culture, the number of Q cells was found to be 4.29 times greater than that of NQ cells (p < 0.0001) (Fig. 2C). In contrast, the number of Q cells and NQ cells derived from liquid culture was 0.57 times (p = 0.0001) and 1.8 times (p = 0.021) that of cells from solid agar plates, respectively. When the total numbers of Q and NQ cells from the same batch were summed, solid culture produced a higher total cell count (p = 0.024) (Fig. S1A), consistent with the findings in Fig. 1C. The average particle size of NQ cells generated under both culture formats was similar, but Q cells grown in liquid culture were slightly larger (p = 0.015) (Fig. 2D and E). Furthermore, there was no significant difference in the cell viability of Q cells and NQ cells cultured in liquid medium. However, Q cells grown on solid agar plates exhibited significantly higher survival rates compared to NQ cells (p = 0.001) (Fig. 2F). Additionally, NQ cells derived from liquid cultures demonstrated greater viability than those from solid culture (p = 0.018), while Q cells showed no significant difference between the two culture formats (Fig. 2F). It is worth noting that, consistent with the result presented in Fig. 1B, the combined viability of Q and NQ cells in the two culture formats is not significantly different (Fig. S1B).

Fig. 2.

Phenotypic differences between quiescent (Q) and non-quiescent (NQ) cells isolated after cultivation in different culture formats. A Percoll density gradient centrifugation separated two cell layers. B Morphological observation of two cell layers by microscopy (bar = 10 μm). C Comparison of cell concentration of the two cell layers. D Cell diameter distribution of the two cell layers. E Comparison of the mean cell diameter of the two cell layers. F Comparison of the cell viability of the two cell layers. Data are presented as mean ± SD of 3 biologically independent experiments, the blue dots in the bar charts represent biological replicates. Ordinary one-way ANOVA was used for the statistical analysis. *: p ≤ 0.05, ***: p ≤ 0.001

Enhanced production of extracellular vesicles in solid culture format

Extracellular vesicles (EVs) are functional structures that play a crucial role in intercellular communications [16]. To determine whether the extracellular vesicles produced in different culture format exhibit any differences, we collected and analyzed the vesicles using differential centrifugation [17]. Contrary to the notion that higher metabolic activity in liquid cultures drives EVs release [18], the solid medium culture yielded a significantly higher number of vesicles (Fig. 3A, p = 0.002). These EVs were notably larger in size (Fig. 3B, p = 0.045), while no difference in membrane potential was detected (Fig. 3C).

Fig. 3.

Yeast cultured in solid medium produced more and larger vesicles. A Comparison of extracellular vesicle concentration between groups. B Comparison of extracellular vesicle particle size between groups. C Comparison of membrane potential of Extracellular vesicles. Data are presented as mean ± SD of 3 biologically independent experiments, the blue dots in the bar charts represent biological replicates. Unpaired student t-test was used for the statistical analysis. ns: p > 0.05, *: p ≤ 0.05, **: p ≤ 0.01

Transcriptional reprogramming in solid culture format features upregulation of ribosomal and metabolic pathways

Based on the above observations, we further investigated the gene expression differences in S. cerevisiae under different culture formats using RNA sequencing. For each sample, a minimum of 45 × 10⁶ mRNA-Seq reads were mapped to the S. cerevisiae genome, which covered roughly 95% of the reference genome. Sample correlation analysis and principal component analysis (PCA) ensured the reliability of subsequent analyses (Fig. S2A and S2B). As shown in Fig. 4A, obvious differences in gene expression were observed between yeasts grown on the solid and liquid culture formats. Venn analysis revealed that 78 genes were exclusively expressed in the liquid culture group, of which 48 are tRNA genes and 16 uncharacterized protein genes with unknown functions. On the other hand, 24 genes were uniquely expressed in the solid culture group, including 8 tRNA genes and 8 transposon-related protein genes (Fig. S2C). After excluding genes exclusively expressed in one condition, using stringent criteria (adjusted p ≤ 0.05, |log2FC| ≥ 1), differential expression analysis revealed that 487 genes were upregulated, while 449 genes were downregulated in the solid culture versus liquid culture format (Fig. 4B, full list in Supplementary Table S1). We then subjected these genes to KEGG analysis. As shown in Fig. 4C, they were significantly enriched in genes for Ribosome (p = 1.07 × 10^− 24), biosynthesis of secondary metabolites (p = 0.0108), and carbon metabolism (p = 0.0083). GO enrichment analysis further revealed that the most significant differences of Biological Process (BP), Cellular Component (CC), or Molecular Function (MF), were enriched in “translation”, “ribosome”, and “structural constituent of ribosome”, respectively (Fig. S2D). Subsequently, we performed a separate analysis of 55 differentially expressed genes (DEGs) associated with the synthesis of the Ribosome 40S small subunit and 76 DEGs linked to the synthesis of the Ribosome 60S large subunit. Our findings indicated that the majority of these genes were upregulated in the solid culture (Fig. 4D). To gain deeper insights, KEGG enrichment analysis revealed that only the upregulated genes were significantly enriched in the “Ribosome” pathway (Fig. S2E and S2F). Finally, the increased expression of the top 5 differentially expressed ribosomal RPL (ribosomal large subunit) and RPS (ribosomal small subunit) genes in the solid culture was confirmed through qRT-PCR (Fig. 4E).

Fig. 4.

Differential gene expression of yeasts cultured in different media. A Differential gene clustering heatmap of all 936 differentially expressed genes (adjusted p ≤ 0.05, |log2FC| ≥ 1). B Differential gene volcano diagram. C Differential gene KEGG enrichment analysis. D Heatmap of differentially expressed genes (DEGs) of ribosome 40 S subunit proteins and ribosome 60 S subunit proteins. E Relative expression of ribosomal genes to β-tubulin (TUB2) detected by qRT-PCR. F Heatmap of DEGs associated with carbon metabolism. G Relative expression of metabolite transporter genes to β-tubulin (TUB2) detected by qRT-PCR. H Relative expression of fatty acid biosynthesis genes to β-tubulin (TUB2) detected by qRT-PCR. 5 biologically independent cultures were processed for each condition for RNA sequencing analysis. qRT-PCR are presented as mean ± SD of 5 biologically independent cultures, the blue and pink dots in the bar charts represent biological replicates. Multiple student t-tests was used for the statistical analysis of E, G and H. *: p ≤ 0.05, **: p ≤ 0.01; ***: p ≤ 0.001

Differential expression of genes involved in carbon metabolism, metabolite transport and fatty acid biosynthesis

Analysis of carbon metabolism genes revealed a complex expression pattern (Fig. 4F). Among the 28 differentially expressed genes related to carbon metabolism, 12 were downregulated and 16 were upregulated in solid culture format. The most pronounced changes were observed in ACS1, ICL1, CTT1, and HXK1 (Fig. 4F). ACS1 encodes acetyl-CoA synthetase, which catalyzes the conversion of acetate to acetyl-CoA, playing a crucial role in key metabolic pathways, including fatty acid and cholesterol synthesis, as well as gluconeogenesis [19, 20]. ICL1 encodes isocitrate lyase, a key enzyme in the glyoxylate cycle, which is involved in converting isocitrate to oxaloacetate. Key gluconeogenic genes (ACS1, ICL1) were higher in liquid culture, suggesting active utilization of non-glucose carbon sources [21]. CTT1 encodes catalase T1, an enzyme responsible for mediating the cellular response to oxidative stress [22, 23]. The upregulation of CTT1 is consistent with a heightened oxidative stress response, possibly due to oxygen gradients within colonies [24, 25]. The upregulation of HXK1 (hexokinase 1) but not HXK2 (hexokinase 2, data not shown) in solid cultures is notable, as HXK1 is typically associated with regulatory functions and stress responses, while HXK2 is the major glucokinase for glycolysis. This specific pattern suggests a metabolic rewiring in solid format that may prioritize signaling or stress adaptation over maximizing glycolytic flux [26].

Additionally, we performed qRT-PCR analysis to investigate the expression levels of metabolite transporter genes and fatty acid biosynthesis genes in both solid and liquid culture formats. Among the 4 metabolite transporter genes examined, 3 showed significantly higher expression in solid culture format (Fig. 4G), though the fold changes were relatively modest. All 5 fatty acid biosynthesis genes tested were significantly upregulated in solid culture format (Fig. 4H). The upregulation of fatty acid biosynthesis genes is consistent with the observed increase in EV size and yield (Fig. 3B), suggesting a potential link between lipid metabolism and EV biogenesis under these conditions.

Discussion

Our study provides a comprehensive demonstration that the format of yeast cultivation - solid agar plate versus shaken liquid culture - is a major determinant of cellular phenotype and global gene expression, even when the chemical composition (YPD) is identical. This challenges a common operational assumption in microbiology and underscores the need for explicit consideration of culture format in experimental design.

The solid format promoted a smaller average cell size but a higher proportion of viable, dense Q cells. This is likely a consequence of the spatially constrained, gradient-rich environment of a colony, where limited diffusion creates micro-niches [5, 24, 25]. Studies have shown that by adding solidifying agents to liquid culture media can induce a transition from planktonic to colony formation in microbial growth. Researchers suggest that this change in the medium state leads to a decrease in the growth rate when transitioning from liquid to solid media [27], as microorganisms face limited access to nutrients and oxygen. Cells in the interior may experience nutrient and oxygen limitation, favoring entry into a protected quiescent state, a well-documented survival strategy [11, 12, 28]. Conversely, the homogeneous and well-aerated liquid environment supported larger cell size and a more uniform distribution between Q and NQ states, reflecting a different trajectory into stationary phase.

A striking and seemingly paradoxical finding was the strong upregulation of ribosomal protein (RP) genes in solid cultures, juxtaposed with the “slow-growth” associated phenotypes of smaller size and higher Q-cell proportion. We propose that this RP gene induction does not necessarily indicate higher translation rates but may represent a stress-ready or maintenance response. In the suboptimal, gradient-driven environment of a colony, cells may invest in ribosome biogenesis to rapidly resume translation upon encountering favorable conditions or to repair stress-induced damage. Alternatively, this signature could originate from a metabolically active sub-population at the colony periphery, diluted in bulk analysis. This hypothesis resolves the apparent contradiction and aligns with known complex regulation of ribosome synthesis in response to environmental cues [29, 30].

The significant increase in EV yield and size from solid cultures was unexpected and contrasts with the typical association of EV release with high metabolic activity [18, 31]. The concurrent upregulation of fatty acid biosynthesis genes suggests a potential link between lipid metabolism and EV biogenesis under these conditions. We speculate that EVs may play an enhanced role in colony communities, perhaps in waste removal, stress signaling, or structural support within the biofilm-like matrix.

Our transcriptomic analysis reveals that growth on solid culture format is not merely a slower version of liquid growth but triggers a distinct adaptive program. The key features of this program - enhanced stress response (e.g., CTT1), metabolic flexibility (e.g., HXK1/ICL1 pattern), and increased biosynthetic capacity for lipids and ribosomal components - can be logically traced to the physical constraints of the colony lifestyle. Nutrient and oxygen gradients create internal regions of starvation and stress [5, 7], selecting for cells that can resist oxidation, utilize alternative carbon sources, and invest in community-level structures like extracellular vesicles. The upregulation of ribosomal genes, in this context, may reflect a strategy to maintain protein synthesis capacity for rapid response to changing local conditions, rather than for prolific growth. Collectively, the transcriptomic data paint a picture of yeast on solid medium adopting a “stress-preparedness and community maintenance” phenotype.

Several limitations of our study warrant mention. First, the experiment only considered two types of culture formats, liquid and solid, while semi-solid, another commonly used microbial growth condition, was not included. Future research could explore its role in yeast growth. Second, although we observed the impact of liquid culture on yeast cell size and EVs production, the underlying molecular mechanisms remain unclear. Further investigations, incorporating genomics and metabolomics, are needed to systematically analyze the physiological changes in yeast under different culture conditions and to better understand the comprehensive effects of culture conditions on yeast growth and metabolism. Third, our study presents a snapshot at a single time point (7 days). A time-course analysis would be invaluable to delineate the dynamics of the transition from exponential growth to the steady-state differences we observed, revealing how and when these phenotypic and transcriptional divergences initiate.

Conclusion

In conclusion, we show that the choice between solid and liquid culture formats is not methodologically neutral but actively shapes the physiological state of S. cerevisiae. Researchers should be cognizant of this “format effect”, particularly in studies focusing on stationary-phase physiology, aging-related phenotypes, metabolism, and extracellular vesicle biology. Our work provides a foundational dataset and highlights the importance of the physical and spatial context in microbial cell biology.

Materials and methods

Yeast strains and culture medium culture conditions

The Saccharomyces cerevisiae strain S288C (MATα, ura3-52, his3Δ200, leu2-3,112, trp1-1, lys2-801) was used in this study. The haploid S288C strain used here carries a non-functional HO endonuclease gene [14], preventing mating-type switching and diploid formation under the conditions tested, thereby ensuring population homogeneity in terms of ploidy. Yeasts were first inoculated onto YPD agar plates for single colony isolation. The isolated colonies were then transferred into 20 mL of YPD liquid medium, which contains 1% (w/v) yeast extract, 2% (w/v) peptone, and 2% (w/v) glucose. The culture was grown at 29°C with shaking at 150 rpm for 24 h to prepare the seed culture. The YPD agar plates were supplemented with 2% (w/v) agar. To ensure consistent culture volume and initial inoculum across both liquid and solid culture conditions, the cell concentration of the seed culture was measured. A 100 µL aliquot of the seed culture, with a concentration of 2 × 10³ cells/mL, was either inoculated into 20 mL of liquid medium or spread onto 20 mL YPD agar plates. Both groups were incubated at 29°C for 7 days under the same conditions. The liquid culture group was shaken at 150 rpm, while the solid culture group remained static in the same incubator.

Biomass measurement

The yield (biomass) of yeast culture after 7 days of incubation was determined as wet weight. Cells from liquid cultures or colonies scraped from agar plates were harvested, washed three times with distilled water and pelleted. The wet weight per total culture volume of 20 mL is reported.

Cell viability assay

For the solid culture group, the culture plates were washed with PBS, and the cell cultures were collected. For the liquid culture group, the cell cultures were collected directly. Both groups of cell cultures were centrifuged at 1000 × g for 3 min to precipitate the cells, followed by 3 washes with PBS under the same centrifugation conditions. To assess cell viability, the cells were resuspended in PBS and adjusted to an OD₆₀₀= 4. Subsequently, 10 µL of the cell suspension was mixed with an equal volume of 0.4% (w/v) Trypan blue solution, incubated at room temperature for 3 min, and analyzed using an automatic cell counter (Denovix, CellDropBF, USA). The assay was performed with three biological replicates for each condition. A minimum of 300 cells were counted per replicate to determine the percentage of viable cells.

Cell number and size measurement

To detect cell number and size distribution in both liquid and solid culture conditions, we measured the cell concentration and size using a Beckman Coulter Multisizer particle counter (Beckman Coulter, Multisizer 4eC, USA). Both liquid and solid cultures were harvested by centrifugation, followed by 3 washes with 20 mL PBS. The cells were then resuspended in 1 mL PBS. For particle analysis, 10 µL of the cell suspension was diluted into 10 mL of dilution solution and thoroughly mixed before measurement. All procedures were performed according to the manufacturer’s instructions. The assay was performed with three biological replicates for each condition.

Extracellular vesicles measurement

Yeast cells from liquid culture were first removed by centrifuging the culture at 1,000 × g for 3 min. In order to eliminate errors caused by liquid evaporation, the precise volume of the resulting supernatant was measured, and an equivalent volume of PBS was added to wash the colonies from the solid culture. The suspension of the solid culture was also centrifuged at 1,000 × g for 3 min to remove yeast cells, and the supernatant was retained.

Subsequently, the supernatant underwent centrifugation at 2,000 × g for 10 min to remove large debris, followed by a 10-minute centrifugation at 10,000 × g to further eliminate impurities. The final step involved centrifuging at 16,000 × g for 30 min to pellet the vesicles. The resulting vesicle pellet was resuspended in 10 µL of PBS for subsequent analysis. Vesicle concentration and size distribution were analyzed using Nanoparticle Tracking Analysis (NTA, Particle Metrix, ZetaView^®, Germany). The assay was performed with three biological replicates for each condition.

Separation of quiescent and non-quiescent yeast cell sub-populations

Quiescent (Q) and non-quiescent (NQ) yeast cells were separated by density gradient centrifugation using Percoll (Sigma). Briefly, a 90% (v/v) Percoll gradient was prepared in Tris-HCl buffer. Yeast cells harvested from both liquid and solid cultures were washed, resuspended to an OD₆₀₀ of 200, and layered over the gradient. After centrifugation at 400 × g for 1 h, cells banding at different densities were collected, washed, and processed for subsequent analysis.

Electron microscopy

Yeast cells were collected by centrifugation and washed 3 times with PBS. The cells were then embedded in agarose and fixed with 1% osmium tetroxide for 2 h at room temperature, protected from light. Following fixation, the slides were rinsed with PBS 3 times for 15 min each. The cells were dehydrated through a graded alcohol series, with each step lasting 20 min, followed by dehydration in 100% acetone for 15 min. After dehydration, the resin block was sectioned into 60–80 nm slices using an ultrathin slicer. The sections were transferred onto 150-mesh copper grids with an aromatic membrane. The copper grids were stained with 2% uranyl acetate in saturated alcohol solution for 8 min in the dark, washed 3 times with 70% alcohol, and then rinsed 3 times with ultrapure water. Subsequently, the sections were stained with 2.6% lead citrate solution for 8 min, avoiding exposure to carbon dioxide. The sections were washed again with ultrapure water, placed into a copper grid box, and dried at room temperature overnight. The sections were observed under a transmission electron microscope (Hitachi, HT7800, Japan).

Bulk RNA sequencing

RNA was extracted from the yeast cells using the RNAprep Pure Total RNA Extraction Kit (Tiangen, DP430). RNA library preparation and quality control were performed by Novogene (Beijing). Briefly, the preparation involves using total RNA as the starting material. mRNA with polyA tails is enriched using Oligo(dT) magnetic beads. The enriched mRNA is then fragmented into smaller pieces using divalent cations in a fragmentation buffer. The fragmented mRNA serves as a template for the first strand of cDNA synthesis, which is performed using random oligonucleotide primers and the M-MuLV reverse transcriptase system. After the RNA strand is degraded by RNase H, the second strand of cDNA is synthesized in a DNA polymerase I system using dNTPs. The resulting double-stranded cDNA is subjected to end repair, A-tail addition, and adapter ligation for sequencing. cDNA fragments of approximately 370–420 bp are selected using AMPure XP beads, then amplified by PCR and purified again with AMPure XP beads to generate the final library. Following library quality control, different libraries are pooled according to their effective concentration and target sequencing requirements. The pooled libraries are then sequenced using Illumina. The image data generated by the high-throughput sequencers are converted into sequence data (reads) via CASAVA base calling, producing files in fastq format. Raw data is then filtered for further analysis.

RNA sequencing analysis

To ensure the quality and reliability of the data analysis, we filtered the raw data by removing reads that contained adapters, reads with ambiguous bases (represented by N), and low-quality reads where more than 50% of the bases had a Qphred score of 5 or less. After filtering the raw data and checking for sequencing error rates and GC content distribution, we obtained clean reads for subsequent analysis. We then used Hisat2 software to rapidly and accurately align these clean reads to the reference genome (ncbi_saccharomyces_cerevisiae_gcf_000146045_2_r64), allowing us to determine the positional information of the reads on the reference genome.

Differential expression analysis for two conditions, each with five biological replicates, was conducted using the DESeq2 R package (version 1.20.0). DESeq2 provides statistical methods to assess differential expression in digital gene expression data through a model based on the negative binomial distribution. The resulting p-values were adjusted using the Benjamini-Hochberg procedure to control the false discovery rate. Genes identified by DESeq2 with an adjusted p-value ≤ 0.05 and |log2FC| ≥ 1 were classified as differentially expressed.

Differential gene clustering groups together genes with similar expression patterns. Hierarchical clustering was performed on the FPKM values of the genes, with normalization applied to the rows using Z-scores. GO enrichment of differentially expressed genes was achieved using clusterProfiler (3.8.1) software, and statistical enrichment of differentially expressed genes in KEGG pathways was analyzed using clusterProfiler (3.8.1) software.

RT-qPCR

RNA was extracted from yeast using RNeasy Mini kit (Qiagen), following the manufacturer’s instructions. The RNA concentration and purity were determined using a NanoDrop spectrophotometer (Thermo Fisher Scientific). For complementary DNA (cDNA) synthesis, 1 µg of total RNA was reverse transcribed using the QuantiTect Rev. Transcription Kit (Qiagen), according to the manufacturer’s protocol. The resulting cDNA was then subjected to quantitative PCR using PowerUp SYBR Green Master Mix on a real-time qPCR systems (CFX96, Bio-Rad). The amplification was carried out using specific primers for genes of interest (a full list of primer sequences in supplementary Table S2), and the expression levels were normalized to Tubulin-β gene (TUB2) for each sample. The relative gene expression was calculated using the 2^−ΔΔCt method.

Statistical analysis

Data analysis in this study was performed using GraphPad Prism software. Quantitative data are expressed as the mean ± standard deviation (X ± SD). Intergroup differences were assessed using independent samples t-tests (for two-sample comparisons) or two-way analysis of variance (ANOVA) for multiple group comparisons. The significance threshold for all statistical tests was set at p < 0.05. For groups with significant differences in the ANOVA, post-hoc Tukey’s multiple comparison test was performed to identify specific group differences.

Abbreviations

BP

Biological Process

CC

Cellular Component

cDNA

Complementary DNA

DEGs

Differentially Expressed Genes

EV

Extracellular vesicle

GO

Gene Ontology

KEGG

Kyoto Encyclopedia of Genes and Genomes

LM

Liquid medium

MF

Molecular Function

NQ

Non-quiescent cells

OD

Optical density

PCA

Principal component analysis

Q

Quiescent cells

qRT-PCR

Quantitative Reverse Transcription Polymerase Chain Reaction

RPL

Ribosomal large subunit

RPS

Ribosomal small subunit

SM

Solid medium

Authors’ contributions

K.W. and J.D. conceived and designed the study. J.Q.-L., J.L., J.P.-L. and Z.Y.-X. performed the experiments. Q.W. and J.Q.-L. analyzed the data and wrote the manuscript draft. K.W. and J.D. revised the manuscript. The final manuscript has been read and approved by all authors.

Funding

This work was supported by the High-level Talents Program of Hainan Provincial Natural Science Foundation (823RC496 to KW), the Graduate Student Innovative Research of Hainan Medical University, China (Qhys2022-286 to J.Q.-L.).

Data availability

The raw RNA sequencing data supporting the findings of this study have been deposited in the Gene Sequence Archive (GSA) and are accessible under the accession number CRA031767. These data are publicly available and can be accessed at https://bigd.big.ac.cn/gsa.

Declarations

Ethics approval and consent to participate

Not applicable.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.