Introduction

Cells can sense a wide range of stimuli and stresses, enabling them to mount appropriate responses that ensure survival under adverse conditions. These stress responses are critical for cellular viability and involve a complex interplay between the regulation of gene expression and post-translational control mechanisms that modulate the activity of effector proteins. In eukaryotes, most of these responses are mediated through signal transduction pathways, which are activated by specific stress signals or by intrinsic factors such as DNA damage or disturbances in cellular homeostasis1,2,3.

Among the factors that play essential roles in post-transcriptional gene regulation are RNA-binding proteins (RBPs), which influence RNA stability, processing, transport, and translation. In yeast, numerous RBPs coordinate these processes, ensuring proper gene expression and cellular function. Two of them, Nab2 and Hrp1 are key regulators of mRNA metabolism4. Nab2 is an mRNA poly(A)-binding protein primarily involved in mRNA export and poly(A) tail length control, acting as a critical factor in the nuclear export pathway5. Hrp1, a component of the cleavage and polyadenylation machinery, participates in pre-mRNA 3’-end formation and mRNA stability6. In Saccharomyces cerevisiae, mRNA degradation occurs through two main pathways: one involves 3′–5′ decay via the exosome after poly(A) tail removal, while the other relies on 5′–3′ degradation by Xrn1 following decapping7. Several decay factors, including Xrn1, Dhh1, Lsm1–7, Pat1, Dcp1/2, and Edc3, that contribute to this process can act as transcriptional activators8,9,10,11.

In yeast, stress responses have been extensively studied at the genome level for various stresses, including oxidative, osmotic, alkaline pH, heat shock, and nutritional stress12,13,14,15,16,17with a focus on changes in transcription and mRNA amounts. However, the levels of a specific mRNA at any given moment are determined not only by the synthesis rate (SR) but also by the tight interplay between SR and the decay rate8,18,19, typically expressed as a half-life or as a first-order degradation constant, kD.

In recent years, various techniques have been developed to assess the relative contributions of transcription and degradation to mRNA abundance20,21. Methods such as Genomic Run-On (GRO)22 and comparative Dynamic Transcriptome Analysis (cDTA)23,24 have proven valuable for investigating these dynamics, particularly in yeast cells subjected to different types of cellular stress. Recent evidence has shown that changes in mRNA levels in yeast cells subjected to oxidative stress25,26, osmotic stress23,27, heat shock28, alkaline pH29, and iron deprivation30,31 are not only the result of alterations in transcriptional regulation (seen as SR variations) but also reflect changes in mRNA stability. Although each stress presents a distinct global response, resulting in different patterns of SR, mRNA levels, and degradation rates, most share a common feature: mRNA destabilization at early stages of stress exposure, with varying degrees of severity depending on the type of stress29. Moreover, mRNA destabilization seems to play an important role in environmental stress response down-regulated genes (ESRdown)11 across most of the stresses studied. The most notable functional categories in this group—ribosome biogenesis (RiBi) and ribosomal protein (RP) genes—are regulated at both the mRNA synthesis and stability level, albeit with stress-specific nuances. Furthermore, changes in both mRNA stability and synthesis also occur transiently in upregulated environmental stress response (ESRup) genes11, but in a manner nearly opposite to that observed for RiBi and RP genes29. Regarding the global SR, this varies depending on the type of stress: it decreases under alkaline pH and iron deprivation, while it increases at different levels and times during heat shock, osmotic stress, and oxidative stress29,30. Thus, it seems that transcriptional response varies in intensity and kinetics depending on the applied stress.

In contrast, no studies address how cell wall stress affects transcription and mRNA stability. The yeast cell wall is a vital structure that can undergo stress. It surrounds fungal cells, determining their shape and protecting them from osmotic bursting caused by internal turgor pressure or environmental stresses32,33. Since cell wall biogenesis is essential for fungal survival, exposure to cell wall-perturbing agents triggers robust rescue mechanisms. These compensatory mechanisms include transcriptional changes that contribute to the synthesis of cell wall components and modifications in the cross-linking of cell wall polymers, both of which are essential for maintaining cellular integrity32,34,35.

The Cell Wall Integrity (CWI) MAPK pathway serves as the primary regulatory mechanism for adaptive responses to cell wall stress32,34. However, additional pathways, such as the HOG and PKA signalling cascades36,37, contribute depending on the nature of the stress. In all cases, cell wall stress triggers a transcriptional adaptive response that is predominantly controlled by the MAPK Slt2/Mpk1 and the transcription factor Rlm138,40, with further input from SBF (Swi4/Swi6) transcription factors35,41. This process involves the recruitment of Rlm1, Slt2, and RNA Polymerase II (RNAPII) to the promoters of CWI-regulated genes, alongside the participation of chromatin remodeling complexes such as SWI/SNF and SAGA, as well as other regulatory factors like Pat111,40,42,43. These coordinated activities ensure precise and effective transcriptional regulation during cell wall stress adaptation.

Previous studies have extensively characterized stress-induced transcriptomic responses, shedding light on the modes and mechanisms of stress defense. Specifically, transcriptomic analyses of cell wall stress induced by agents that disrupt cell wall biogenesis—such as Congo Red11,39,40,42, Calcofluor White44, Zymolyase36, caspofungin37,45, and poacic acid46—have identified key regulatory elements of the cell wall transcriptional response. These studies also emphasize subtle yet significant differences in the regulatory mechanisms activated by each agent, reflecting their distinct modes of action. In all these works, the relative abundance was estimated, revealing that in most wall stresses, the mRNA levels of genes significantly increased by approximately 1.5–2%, while genes with decreased mRNA levels were quite rare.

To better understand the roles of SR and mRNA stability in stress adaptation, we investigated the yeast cell wall stress response and compared it with other stress responses using the GRO method. In this study, we analyzed genome-wide changes in mRNA concentrations ([mRNA]) to determine how synthesis rates and mRNA stability influence mRNA levels during treatment with Congo Red. Our findings indicate that alterations in SR primarily drive changes in [mRNA], whereas mRNA stability remains largely unaffected under our conditions, in contrast to other stress responses. However, some genes previously known to be induced by Congo Red show that mRNA stabilization contributes, in part, to their increased mRNA levels. Additionally, the mRNA levels of RP, RiBi, and ESRup genes are influenced by both mRNA stability and SR, albeit to a lesser extent than in other stress conditions29. Moreover, we identified: (a) a small subset of genes whose mRNA levels are co-regulated by both changes in SR and mRNA stability; (b) previously unidentified genes whose mRNA levels increase in the presence of Congo Red; and (c) RNA-binding proteins (RBPs) such as Nab2 and Hrp1 as potential regulators of genes induced in response to cell wall stress.

Results

Global kinetic analysis of mRNA levels, synthesis rates, and mRNA stabilities after cell wall stress

In recent years, our lab has characterized the global transcriptional response of S. cerevisiae to Congo Red, a dye that interferes with cell wall chitin11,39,40,42. These previous works primarily focused on analyzing mRNA levels, providing insights into changes in gene expression under stress conditions. However, mRNA levels can be influenced by various factors, including alterations in the SR of genes, mediated by transcription factors that bind to gene promoters, and variations in mRNA stability, modulated by RNA-binding proteins (RBP) that bind to mRNAs. In this study, we expand on this analysis using the GRO method21,22 to assess transcriptional regulation and mRNA stability in response to Congo Red stress. This approach provides a more comprehensive understanding of gene expression dynamics under stress conditions.

Before initiating the GRO experiments, it was necessary to assess whether Congo Red affects cell volume. This evaluation is important because an increase in cell volume has been associated with a decrease in the overall mRNA turnover rate, which helps maintain a relatively stable total [mRNA]47,48. Additionally, SR is calculated as changes in [mRNA] with time that, in turn, requires to know cell volume changes along the experimental conditions (see48 for a detailed explanation). To address this, we monitored cell volume variation over time in untreated (time 0) and Congo Red-treated cells (30, 60, 120, 180, and 240 min). As shown in Supplementary Figure S1, untreated cells maintained a relatively constant volume throughout the time course. In contrast, Congo Red-treated cells exhibited a progressive increase in volume starting after 1 h of treatment, and reaching approximately 1.2-fold at 2 h, 1.4-fold at 3 h, and up to 1.6-fold relative to control cells after 4 h of treatment. These factors were used to correct raw transcription rate (TR) and mRNA data to convert them into SR and [mRNA] data (see “Methods” for description). It is important to note that no cell volume changes were observed in previous stress-response studies (e.g29). Therefore, TR and mRNA amount data by relative cell volume were not corrected in those cases. Since GRO experiments use the same number of cells in each sample, the optical density measurements at each time point were complemented by cell counts performed using a flow cytometer (to ensure the same cell number).

After establishing the appropriate experimental parameters, we quantified the global [mRNA] and SR as global median value for all yeast genes over a time course of Congo Red treatment (30 µg/ml) using the GRO method21,22. Additionally, changes in global mRNA stability, represented as variation in the mRNA degradation constant (kD), were calculated using a previously developed algorithm (the higher the kD, the greater the destabilization of mRNAs)49,50. As shown in Fig. 1, mRNA level decreases after 60 min of exposure to Congo Red, stabilizing at approximately 70% of the initial level for the remainder of the time course. The SR exhibits a similar trend to that of [mRNA], though it begins to decline slightly earlier and ultimately stabilizes at around 60% of the initial value. As expected, since [mRNA] and SR follow similar kinetics, the global stability of the mRNAs fluctuates minimally, with variations not exceeding 10%.

Fig. 1
figure 1

Effect of cell wall stress on total mRNA concentration [mRNA], global synthesis rate (SR), and degradation constant (kD). Cells growing exponentially in YEPD were collected at time 0, then exposed to Congo red, and aliquots were collected at 30, 60, 120, 180, and 240 min to measure [mRNA], SR and kD. See “Methods” for details. The parameters are expressed relative to their respective time 0 values (defined as the unit). kD values above 1 indicate mRNA destabilization, while values below 1 indicate stabilization. The mean ± SD from three independent experiments are shown.

Full size image

Gene-specific responses in mRNA abundance, synthesis rates, and mRNA stability following cell wall stress

Additionally, the GRO method provided individual SR and [mRNA] values for the 4,468 yeast genes with valid data (Supplementary Table S1) in response to cell wall stress. We performed a clustering analysis of gene profiles integrating both SR and [mRNA] datasets to distinguish the different behaviors of yeast genes during the stress response (Fig. 2). For this analysis, we represented the [mRNA] and SR values on a logarithmic scale relative to their respective time 0. Subsequently, kD values were calculated for each [mRNA]/SR cluster and incorporated into the figure to facilitate the interpretation of the various clusters.

Fig. 2
figure 2

Clustering of SR and [mRNA] data from cell wall stress. Time-course profiles for both parameters were used for clustering, generated by the 4.9 MeV (Multi Experiment Viewer) software. Both datasets are normalized to the value at time 0 on a logarithmic scale to allow for comparison between SR and [mRNA] data. For each cluster, the number of genes and data profiles are indicated. kD values, inferred from [mRNA] and SR data, are included in each plot. The dendrogram at the top reflects the distances between the cluster profiles. Data are presented as the mean ± standard error of [mRNA], SR, and kD for all genes within each cluster.

Full size image

We adjusted the clustering protocol to yield a tree with nine clusters (Fig. 2), aiming to control cluster number and thereby improve result interpretability without compromising data integrity. The genes included in each cluster are listed in Supplementary Table S2. Notably, 87% of the genes belong to two clusters (8 and 9 with 1544 and 2338 genes, respectively), which show minimal change across all three parameters with a continuous decrease in [mRNA]. Consequently, these two clusters, especially cluster 9, account for the global profile depicted in Fig. 1. No statistically significant enrichment in GO categories was found in both clusters despite containing most of the RP and RiBi genes. Clusters 1–5 (559 genes) show an initial mRNA stabilization (kD decrease) caused by Congo Red treatment. In cluster 1 (435 genes), a slight increase in mRNA concentration is observed at 30–60 min after treatment with Congo Red, returning to baseline levels after two hours. The SR does not change and even slightly decreases after two hours of treatment. In contrast, there is a slight stabilization of the mRNAs from the beginning of the time course, which remains relatively constant over time and may explain the initial increase in [mRNA].

Clusters 2–4 (comprising 6, 11, and 39 genes, respectively) correspond to genes with increasing [mRNA] over the time course. This increase is more pronounced in Cluster 2 compared to the other two clusters and is primarily driven by an increase in the synthesis rate. The common behavior of all these clusters is the greater increase in [mRNA] than in SR caused by a decrease in kD. This stabilization of the mRNAs is most prominent in the genes of Cluster 2.

The genes in clusters 5 and 6 (68, and 10 genes, respectively) are characterized by an increase in their [mRNA], although this increase is less pronounced than in the genes of the previous clusters. This rise remains constant over time, with a tendency to decrease in the case of cluster 5. Thus, in cluster 5, the [mRNA] increase is due to a rise in the SR, whereas in cluster 6, the [mRNA] increase is also driven by SR but partially counteracted by an augment in kD.

The Gene Ontology (GO) analysis of the genes included in these clusters (Table 1) revealed significant enrichment of specific functional categories in cluster 1 (“carbohydrate metabolic process”, “autophagy”, and “cell budding”), clusters 2 and 3 (“cell wall organization or biogenesis”), cluster 4 (“hexose transport and metabolic process”), cluster 5 (“organic acid metabolic process”) and cluster 6 (“generation of precursor metabolites and energy”).

Cluster 7 (17 genes) comprises genes whose SR increases between 30 and 180 min and whose mRNA becomes destabilized after 30 min, resulting in minimal variation in [mRNA] over time. This cluster lacks significant GO categories; however, it contains several stress-related genes, such as ECM38, SIP18, and MTL1, the latter of which is a sensor in the CWI pathway in response to oxidative stress51.

Table 1 Functional classification of genes from various clusters according to the gene ontology term finder tool (p-value < 0.05) of Saccharomyces Genome Database. The table shows the frequency of genes annotated to each term within the cluster and genome-wide, the corrected p-value and the false discovery rate (FDR). Annotated genes corresponding to each term are listed.
Full size table

In addition, the GRO method allowed us to identify 58 genes (Supplementary Table S3), mainly distributed across clusters 1, 3, 4, 5, and 6, whose [mRNA] increased twofold or more at least at two time points and had not been previously identified in DNA microarray studies analyzing the response to Congo Red11,39,40,42. No significant GO categories were found for these genes.

Study of the enrichment of transcription factors and RNA-binding protein targets in clusters

Since several transcription factors are associated with stress responses, we aimed to identify potential activators of genes exhibiting an induced SR, specifically those in clusters 2 through 7, where the most significant changes in SR and/or mRNA stability occur.

Using the Yeastract + web tool52,53, we found that the transcription factor Rlm1 was significantly represented across clusters 2 to 7 (Table 2). This was expected, given that Rlm1 is the transcription factor of the CWI pathway54 and regulates most of the genes whose [mRNA] increase following exposure to Congo Red39,40. This agrees with the fact that 32.5% of the genes included in clusters 2 to 7 were previously identified using DNA microarrays11,39,40,42, whereas only 7.4%, 1.9%, and 1.0% of the genes in clusters 1, 8, and 9, respectively, had been previously reported. Genes identified previously by DNA microarrays are indicated in Supplementary Table S2. Notably, the genes belonging to cluster 7, which exhibit an upregulated SR but no change in [mRNA] and are therefore undetectable using techniques that only assess mRNA quantity, such as microarrays, also appear to be, at least 50%, controlled by Rlm1, although 4 of these genes had been previously detected in these earlier microarray studies.

Table 2 Transcription factor regulation enrichment for genes included in clusters 2 to 7, as determined by analysis using YEASTRACT. The table shows the percentage of transcription factor target genes in each cluster and their proportion relative to the entire genome, along with the corresponding p-value (< 0.001).
Full size table

Additionally, Met32 (zinc-finger DNA-binding transcription factor; involved in the transcriptional regulation of methionine biosynthetic genes) and Rpn4 (a transcription factor that stimulates the expression of proteasome genes) appear to be potential positive regulators of genes in clusters 2–3, and cluster 6, respectively.

On the other hand, 3’ UTR regions often contain specific sequence motifs that are recognized by RBPs involved in the regulation of mRNA stability. Since our analysis suggested that there are gene groups whose mRNA stability appears to be coordinately regulated, we investigated the possible existence of common motifs in their 3’ UTR regions that could help us identify RBPs responsible for this regulation. By analyzing the genes from clusters 2 to 7, using the XSTREME tool from the MEME suite55, we identified putative targets for two RBPs, Hrp1 (which plays a role in transcription termination and mRNA processing) and Nab2 (which is involved in mRNA export from the nucleus to the cytoplasm) that may influence mRNA processing4,5,6. These two proteins are putatively involved in regulating mRNAs from all the clusters (Supplementary Figure S2), but in the clusters where greater stability differences are observed (2, 3, 4, 6, and 7), there are more potential binding sites for these proteins (between 70% and 100% of the mRNAs) compared to cluster 5 (where only a slight stabilization of mRNAs is observed at the beginning of the Congo Red treatment) or when analyzing a random selection of genes where potential binding sites for Hrp1 and Nab2 are found in only 20–30% of the mRNAs. Additionally, in clusters 2, 3, 6, and 7, new potential regulatory motifs have been identified that are not assigned to known RBPs (Supplementary Figure S2).

Determination of mRNA stability and decay rates for representative genes

To validate the kD data obtained through GRO, we measured the mRNA half-lives of four genes whose individual kD kinetics are shown in Fig. 3a: KDX1 (protein kinase paralog of Slt256), the most highly induced gene in cluster 2, whose mRNA remains stabilized throughout the entire Congo Red treatment; CRH1 (a chitin transglycosylase whose function is to transfer chitin to β(1–6) and β(1–3) glucans in the cell wall57) from cluster 4, whose mRNA is also stabilized; APE1 (vacuolar aminopeptidase often used as a marker in studies of autophagy58), whose mRNA undergoes the greatest destabilization among the cluster 7 genes after prolonged cell wall stress; and SRL3 (a cell cycle protein related to the SBF complex (Swi4/Swi6)59), included in cluster 3, which, after initial stabilization, reaches an equilibrium kD after 45 min of Congo Red exposure. For this, we grew the wild-type strain in the absence and presence of Congo Red for two hours and measured mRNA decay rates (mRNA half-lives) at different time points after adding thiolutin, a compound that acts as a transcriptional shut-off.

As shown in Fig. 3b, Congo Red promotes the stabilization of KDX1 and CRH1 mRNAs, increasing their half-life from 14 min under non-stress conditions to 19 min in the presence of the stressor. In contrast, APE1 mRNA undergoes significant destabilization upon Congo Red treatment, decreasing its half-life from 279 to 57 min. In the case of SRL3, mRNA stability remains largely unchanged, with half-lives of 25 and 26 min in the absence and presence of Congo Red, respectively. These results are largely consistent with those obtained through GRO (see Fig. 3, left panels).

Fig. 3
figure 3

mRNA half-lives of the KDX1, APE1, CRH1, and SRL3 genes align with GRO experiment predictions. (a) Representation of kD kinetics for each gene from GRO experiment. (b) mRNA half-lives (t1/2) of selected transcripts were determined using the thiolutin (3 µg/mL) transcriptional shut-off method (see “Methods” for details) in the wild-type strain after 2 h of growth in the presence or absence of Congo Red (CR, 30 µg/mL). The mean ± SD from three independent experiments is shown. The arrow in left panels marks 120 min time when the thiolutin experiments on right panels were done.

Full size image

Effect on mRNA dynamics of ribosomal protein and ribosome biogenesis genes

Congo Red acts as a cell wall stress agent, and genes related to ribosomal protein (RP) and ribosome biogenesis (RiBi) are highly sensitive to cellular stress. Stress conditions often lead to the downregulation of these genes, resulting in a decrease in ribosome biogenesis as a protective mechanism that conserves resources and redirects cellular energy toward stress response pathways12,15,29,60. Following Congo Red treatment, the [mRNA] of both RiBi and RP genes declines progressively, reaching approximately 40% of their initial levels (Fig. 4a, b). This reduction in [mRNA] is attributed to a similarly patterned decrease in SRs but enhanced by mRNA destabilization, especially for RP genes (Fig. 4a). Nevertheless, the reductions in [mRNA] and SR and the increase in kD are not as pronounced as in other types of stress, such as oxidative, osmotic, heat shock, or alkaline pH stress where the kD values of RP genes fluctuate between 4-fold (oxidative stress) and 8.5-fold (osmotic stress) relative to time 0, and those of RiBi genes vary between 2.5-fold (alkaline pH) and 6-fold (osmotic stress)29. Under conditions of iron deprivation, destabilization also occurs at earlier stages of deprivation31. However, in contrast to other stresses, it is followed by prolonged stabilization.

Fig. 4
figure 4

[mRNA], SR, and kD values for ribosomal proteins (RP) and RiBi genes under cell wall stress. Genes encoding the ribosomal proteins (a), RiBi (b), and ESRup genes (c) were selected from the complete set from GRO experiment. The parameters are expressed relative to their respective time 0 values (defined as the unit). The mean ± standard error of [mRNA], SR, and kD corresponding to all genes in each group is represented.

Full size image

Specific kinetic analysis of upregulated environmentally responsive genes

The RP and RiBi genes are typically downregulated under stress conditions (ESRdown). In contrast, other environmentally stress response genes12 are upregulated (ESRup). As shown in Fig. 4c, upon exposure to Congo Red, the [mRNA] of these genes increases by 1.4-fold within short timeframes (30–60 min), primarily driven by mRNA stabilization. After two hours, [mRNA] returns to time 0 level as well, despite a 30% reduction in the synthesis rate, due to a slight re-stabilization of the transcripts. Under other stress conditions -oxidative stress, osmotic stress, heat shock, or alkaline pH-, the regulation of ESRup genes is predominantly influenced by changes in SR and complex transient bidirectional alterations in mRNA stability29. Most of the RP and RiBi genes are found in clusters 8 and 9 (as previously indicated), while the ESRup genes are distributed across all clusters and are marked in Supplementary Table S2. We can conclude that the ESRup response is not important to the Congo Red stress response.

Global comparison of transcriptional responses under various stress conditions

This study integrates previous GRO data on the transcriptional response of yeast cells subjected to various stress conditions—such as osmotic27, heat28, oxidative25, and alkaline stress29, and iron deprivation30,31—in combination with cell wall stress, aiming to provide a broader perspective on how yeast cells respond to different types of stress. This approach expands upon prior analyses conducted without considering cell wall stress29. Figure 5 shows the global mRNA levels, the transcriptional response profiles, and mRNA stability variations compared across all stress conditions. For cell wall stress, we use the SR values instead of the TR and [mRNA] instead of mRNA amounts as in previous works due to the progressive increase in cell volume observed upon Congo Red exposure, as previously noted. In the other stresses, cell volume did not change; therefore, the relative values of TR are equivalent to those of SR (SR = TR/1) and similarly for mRNA amounts and mRNA concentrations (we use here mRNA levels to refer for both). While heat shock resulted in an initial increase in mRNA levels, a decrease was observed under oxidative, osmotic, alkaline and cell wall stresses. However, this decrease is not immediate in the case of cell wall stress. TR/SR profiles revealed a common pattern for heat and oxidative stress, with an immediate increase in SR, followed by a gradual return to baseline. In contrast, alkaline and cell wall stresses and iron deprivation resulted in a sustained decrease in SR, while osmotic stress exhibited an initial drop followed by an increase. Additionally, kD increases and subsequently decreases at varying rates in the previously studied stress conditions. In all cases, the kD tends to return to baseline values, although to a lesser extent under oxidative stress, where initial levels are not fully reached, or even increase mRNA stability, as occurs with heat shock and especially under iron deprivation conditions. In contrast, cell wall stress is the only stress condition where kD remains largely unchanged, with only minimal fluctuations observed. These findings underscore the complexity of the stress response, in which transcriptional regulation, mRNA abundance, and stability collectively shape adaptive strategies tailored to specific environmental challenges.

Fig. 5
figure 5

Comparison of mRNA dynamics between different stresses. The overall [mRNA], synthesis rates (SR), and degradation constants (kD) for osmotic stress27, heat shock28, alkaline pH29, oxidative stress25, iron deprivation30,31, and cell wall stress are shown. The parameters are expressed relative to their respective time 0 values (defined as the unit). The data for all course values were extracted from the original reports based on GRO experiments.

Full size image

Discussion

In this work, we conducted a comprehensive analysis of the effects of cell wall stress on mRNA dynamics at the whole-genome level. Using the GRO method, we examined global changes in synthesis rates, mRNA quantities, and mRNA stabilities over time, allowing us to assess the relative contributions of transcriptional and post-transcriptional regulation in response to stress. These data complement previous studies that analyzed only the effect of Congo red on mRNA abundance11,39,40,42, providing a more complete understanding of gene expression regulation under cell wall stress conditions. Our study builds upon earlier investigations of oxidative stress, osmotic stress, moderate heat shock, and sudden alkalization, integrating new data on transcription and mRNA stability. In contrast to these stresses, Congo Red leads to a progressive increase in cell volume throughout exposure. Cell volume plays an important role in interpreting transcription and synthesis rates, as larger cells or those with greater volume may contain more genetic material and available transcription machinery: however, they may also require more time to process and export RNA, potentially influencing the total amount of RNA produced47,48.

Globally, our results demonstrate a moderate decrease in total [mRNA] after one hour of exposure to cell wall stress, due to parallel changes in SR. Therefore, overall mRNA stability remains largely unaffected over time. The clustering analysis enabled us to identify gene groups that deviate from the global trend. Interestingly, about 15% of the transcriptome (clusters 1–5) shows small but significant changes in mRNA stability. These changes were previously unnoticed, as this is the first study capable of capturing dynamic mRNA stability changes during the Congo Red stress response. Interestingly, a large proportion of genes involved in cell wall organization fall within this group. In fact, approximately 50% of the genes previously identified by microarray analysis as showing increased mRNA levels are located within these clusters. Notably, in these genes, mRNA accumulation results from a coordinated increase in SR and mRNA stabilization. In one gene group, associated with carbohydrate metabolic processes, autophagy, and cell budding, this increase in [mRNA] is modest. In contrast, genes with higher mRNA levels are enriched in functional groups related to cell wall organization, hexose metabolism, and organic acid metabolism. Previous studies in S. cerevisiae suggest that heat shock promotes the stabilization of heat-shock-inducible mRNAs, which may indicate that mRNA stabilization is a key factor in regulating eukaryotic gene expression alongside transcriptional activation, allowing cells to increase the levels of specific transcripts in response to environmental stress conditions61. In another small gene group, corresponding to cluster 6, the increase in [mRNA] does not correlate with the rise in synthesis rate (SR), due to a slight destabilization of mRNA at intermediate time points during cell wall stress treatment. This suggests that the regulation of mRNA stability in this gene group acts to compensate for the increase in synthesis. The reason for this counterintuitive mechanism is that, for these genes, the response may need to be transient, as previously proposed49. This cluster is abundant in genes related to the generation of precursor metabolites and energy. Overall, the enrichment of these functional groups can be explained by the fact that cell wall synthesis and repair require energy and metabolic precursors. Consequently, genes involved in carbohydrate and organic acid metabolism are activated, as these compounds are essential for the synthesis of glucans, chitin, and other structural components. Additionally, cells may activate autophagy to recycle cellular components and ensure the availability of essential nutrients for cell wall repair. Moreover, cell wall stress could compromise the viability of the daughter cell, leading to the activation of genes related to cell budding to reinforce its structure.

Notably, we have identified a cluster of 17 (cluster 7) genes that shows compensatory behavior of SR and kD although, in this case, [mRNA] remains mostly unaltered. Thus, the GRO method has enabled us to identify transcriptionally regulated genes, most of which remain undetected in standard DNA microarray analyses due to the absence of an increase in mRNA levels. These genes are not enriched in any specific functional group; however, they include stress-related genes such as ECM38, SIP18, and MTL1. MTL1 functions as a sensor in the CWI pathway, responsible for responding to oxidative stress51. The finding of MTL1 suggest that Mtl1 may play a secondary role in responding to cell wall stress induced by Congo Red, a process primarily signaled by Mid245. Additionally, we have identified 58 other genes whose mRNA levels increase in the presence of Congo Red, which had not been identified in previous DNA microarray studies11,39,40,42. These genes are distributed across different clusters and, although they are not enriched in any GO category, they include genes related to stress response, such as ASC1 and SNG1, to RNA decay, like NGL3, and to other signaling pathways, such as TPK1 from the RAS-cAMP/PKA pathway. Previous studies on yeast have also reported interactions between the CWI and cAMP/PKA pathways. For example, the Wsc1 sensor has been proposed to act downstream of Ras proteins under heat shock conditions62 and upstream of RAS pathway in response to cell wall stress induced by caspofungin37. Furthermore, the involvement of the Rho1 GTP/GDP exchange factor Rom2 in the down-regulation of the Ras-cAMP pathway in response to oxidative stress51,63, as well as the participation of the CWI pathway in TPK1 expression under heat stress, has also been reported64.

As previously mentioned, the GRO results show gene clusters with varying kD kinetics. Several methods can validate these results by determining mRNA decay rates after transcriptional arrest, each with its own advantages and limitations65. Previous GRO studies have utilized the doxycycline-regulable tetO2 promoter method25,28 and the thiolutin-based method31, both of which confirmed the obtained data. In this study, we used the latter approach and observed that four mRNAs, corresponding to genes from four different clusters—KDX1, CRH1, APE1, and SRL3—were stabilized (KDX1 and CRH1), destabilized (APE1), or remained mostly unchanged (SRL3), in agreement with the GRO predictions. In a previous study11, we employed the doxycycline-regulable tetO2 promoter method66 to determine the half-life of KDX1 mRNA in the presence and absence of Congo Red. Although to a lesser extent, these mRNAs also exhibited stabilization under stress conditions.

The analysis of potential transcription factors regulating the genes in the clusters with high SR revealed that Rlm1 is their main activator. This was expected, as this protein is the primary transcription factor in the CWI pathway in response to cell wall damage such as Congo red39,40. Met32 and Rpn4 may also be possible transcription factors for some genes, 16 and 10, respectively, under these conditions. Met32 is involved in the transcriptional regulation of methionine biosynthetic genes, while Rpn4 stimulates the expression of proteasome genes. Cell wall damage can lead to the accumulation of damaged proteins. The proteasomal system (regulated by Rpn4) may be activated to degrade these proteins and facilitate cellular adaptation. In fact, a recent study identified a mechanism by which the Slt2 and TORC1 complex regulated the proteasome subunit supply under challenging conditions, thereby sustaining proteasomal degradation and ensuring cell viability67. Although none of the potential targets of Rpn4 in our study directly participate in the processes regulated by these proteins, this result could suggest that some of these genes may have an undescribed function.

In addition to factors that control transcription, we investigated RNA-binding proteins (RBPs) that influence the stability of their mRNA targets. Thus, we searched for overrepresented RBP motifs in clusters of genes characterized by a shared mRNA stability profile during a time course in the presence of Congo Red. Our analysis identified putative new binding motifs in other potential post-transcriptional regulators, as well as some gene clusters with statistically overrepresented possible targets of Hrp1 and Nab2.

Hrp1 and Nab2 are key guard proteins involved in pre-mRNA maturation surveillance, functioning as both nuclear and cytoplasmic quality-control factors4. As such, they may indirectly influence mRNA stability. Hrp1 plays a crucial role in 3′ end cleavage and the release of pre-mRNA from RNAPII68, and it has been implicated in nonsense-mediated decay (NMD)69. Nab2 acts as the nuclear poly(A) tail-binding protein, facilitating mRNA transport through the nuclear pore complex (NPC)5,70 and contributing to the proper folding of messenger ribonucleoproteins (mRNPs)71. Furthermore, a study in Schizosaccharomyces pombe revealed that Nab2 helps to stabilize specific pre-mRNAs72. A study has described the role of the MAPK Slt2 in phosphorylating the adapter Nab2. Under heat stress, Slt2-dependent phosphorylation of Nab2 leads to the nuclear retention of mRNAs not associated with the stress response73. This finding raises the possibility that Slt2 may regulate Nab2 under cell wall stress conditions. These results suggest that Hrp1 and Nab2 may influence the stability and export of mRNAs from genes involved in cell wall synthesis and maintenance. In this context, a recent study reported that Nab6, another RBP, acts parallel to the CWI pathway to modulate the expression of cell wall-related genes during stress by binding to their 3′ UTRs74.

The analysis of RiBi, RP, and ESRup genes shows distinct regulatory responses to different stress conditions. Heat shock, alkaline pH, osmotic, oxidative and cell wall stresses cause rapid destabilization of RP and RiBi mRNAs—reflecting protein synthesis inhibition—although the effect of cell wall stress is milder. Notably, iron deprivation leads to mRNA stabilization despite reduced transcription, causing minimal changes in mRNA levels. Regarding ESRup genes, their regulation also differs between cell wall stress and other stress conditions. Under Congo Red exposure, the main regulatory change in ESRup genes is a slight increase in [mRNA] due to mRNA stabilization. In other stress conditions, the regulation of these genes is more complex, primarily influenced by changes in SR and transient bidirectional alterations in mRNA stability.

Global comparisons of responses across different stress conditions also reveal significant variations, likely because cellular responses to different types of stress in yeast vary considerably over time. Heat shock28 and osmotic stress27 typically induce rapid responses, and under these conditions, mRNAs undergo greater initial destabilization than in other stress types. Moreover, SR changes are also more pronounced under these stresses, although each stress exhibits its particular pattern. Alkaline pH29 and oxidative stress25 lead to relatively delayed responses compared to osmotic and oxidative stress. These conditions also increase mRNA destabilization; however, the kD and SR display fewer alterations. Iron deprivation30,31 and cell wall stress show more prolonged response patterns. While SR levels are similar to those observed in alkaline pH and oxidative stress, KD behaves differently: in cell wall stress, it remains largely unchanged, whereas in iron deprivation, kD progressively decreases over time. This diversity in responses, both at the global level and in the RiBi, RP, and ESRup genes, suggests the existence of specialized transcriptional and post-transcriptional regulatory mechanisms tailored to each type of stress.

In summary, this work broadens the list of genes regulated by the CWI pathway in response to cell wall stress and provides new insights into how mRNA stability influences this response. We confirmed that alterations in mRNA decay rates are a crucial component of the stress response for certain gene groups, further enhancing our understanding of mRNA dynamics under various stress conditions. Moreover, we conclude that Congo Red stress response has distinctive features that differentiate it, and although mRNA stability does play a role, its contribution is relatively minor compared to its more prominent function in other stress conditions. Thus, this study improves our knowledge of how external signals regulate gene expression at multiple levels, including transcriptional regulation and mRNA stability, to promote cell survival under stress. Undoubtedly, proteomic analyses could enhance studies on mRNA dynamics and correlate with alterations in protein levels. We also expect this knowledge will be valuable in characterizing mRNA dynamics in other clinically relevant pathogenic fungi.

Methods

Strain and growth conditions

Experiments were performed with the Saccharomyces cerevisiae BY4741 strain (MATa; his3Δ1; leu2Δ0; met15Δ0; ura3Δ0) provided by Euroscarf (Frankfurt, Germany). Yeast cells were cultured overnight at 24 °C in YEPD medium of 1% yeast extract (Condalab, Madrid, Spain), 2% peptone (Condalab, Madrid, Spain), and 2% glucose (Thermo Fisher Scientific, Waltham, MA, USA) until an optical density of 0.8–1 at 600 nm was reached. The culture was then refreshed in YEPD to an optical density of 0.1 at 600 nm and grown for 2.5 h. After this period, a sublethal concentration of Congo Red (30 µg/ml; Merck KGaA, Darmstadt, Germany) was added, with a sample taken at time 0 for reference. Cells were collected at 30, 60, 120, 180, and 240 min following Congo Red treatment. The volume collected at each time point was adjusted based on cell volume. Flow cytometry (Guava easyCyte, Merck KGaA, Darmstadt, Germany) was then used to determine the optimal optical density based on volume variation, ensuring that the same number of cells was collected at each time point. Median cell volumes (in femtoliters) were measured using a Coulter Counter Z Series device (Beckman Coulter, Brea, CA, USA). Relative values are presented in Supplementary Figure S1.

Genomic Run-on

Genomic run-on (GRO) was performed as described in22 and as modified in75 at the following sampling times: 0 min (before Congo Red treatment), and 30, 60, 120, 180, and 240 min after CR treatment. Briefly by macroarray hybridization, GRO detects genome-wide, active elongating RNAPII, whose density per gene is taken as a measurement of its nascent transcription rate that is converted to SR by dividing it by the cell volume factor (Supplementary Figure S1). Global SR was calculated by taking the media of all individual gene SRs. At the same time, the protocol allows for the estimation of mRNA amounts for all the genes using the hybridization of labelled cDNA onto the same nylon filters as the SR samples. To obtain individual [mRNA] values, mRNA amounts were divided by the cell volume factor. Total mRNA concentration per cell was determined by quantifying polyA + RNA in total RNA samples by oligo-dT hybridization of a dot blot containing total RNA and dividing by average cell volume at each sampling time75.

The decay rate constant (kD) was calculated from the SR and [mRNA] data for all three independent GRO experiments using the mathematical method described in25 and assuming steady-state conditions before the Congo Red treatment (t0) and non-steady state conditions after the cell wall stress (t30-t240). Global kD was calculated as the median of all individual kD.

Clustering procedures

The clusters used to evaluate changes in the normalized average values of [mRNA] and SR were generated using MeV MultiExperiment Viewer software, version 4.9 (The Institute for Genomic Research, Rockville, MD, USA) with the SOTA (Self-Organizing Tree Algorithm) tool and employing Euclidean distance as the similarity metric. All parameters were left at their default settings except for the maximum number of iterations (max cycles), which was set to 8. This adjustment led to the formation of 9 clusters (max cycles + 1). To assess potential enrichment in Gene Ontology (GO) categories among the clusters obtained in this study, we used the GO Term Finder tool, with the GO category “Process”, included in the Saccharomyces Genome Database (SGD), considering GO categories with an adjusted p-value below 0.05 as significant.

Transcription factors and RNA-binding protein analyses

Rank Genes by TF tool (with the option “TF acting as activator”) from The YEASTRACT database (http://www.yeastract.com)52,53 was used to search for putative transcription factors in the target genes. The Fungi server from the Regulatory Sequence Analysis Tools (RSAT) website (https://rsat.france-bioinformatique.fr/fungi/)76 was employed to search the 200 bp downstream of the STOP codon for each analyzed gene. In the obtained sequences, thymine (T) was replaced by uracil (U), and the modified sequences were then used to search for potential RBP binding sites using the XTREME tool from the MEME suite (https://meme-suite.org/meme/)55.

Total RNA isolation and quantification of mRNA stability using RT-qPCR

After two hours of yeast growth, whether in the absence or presence of Congo red (30 µg/ml), both cultures were supplemented with 3 µg/ml of the transcriptional inhibitor thiolutin (Merck KGaA, Darmstadt, Germany). Cells were collected every 10 min following the addition of thiolutin, for over 50 min. Samples were then stored at −80 °C until RNA extraction. Total RNA isolation and RT-qPCR assays were performed as previously described11. For RT-qPCR quantification, the SCR1 ncRNA was used for input cDNA normalization. Relative mRNA abundance for each target gene (KDX1, CRH1, APE1, and SRL3) was calculated using the 2−ΔΔCt method77comparing mRNA levels at each time point after thiolutin addition to the baseline levels at time 0 (immediately before adding Congo red). mRNA half-lives were determined using the equation t1/2 = −0.693/k, as described by Hu and Coller78, where k is the slope of the best-fit line from a semi-logarithmic plot of time (in minutes) versus the remaining mRNA (as a percentage on a log scale). Primers used for RT-qPCR are available upon request.