The effects of different carbon sources on optimizing the expression of the inducible EGFP gene under the control of the GAL1 promoter in S. cerevisiae
A unique platform for gene expression studies in S. cerevisiae was developed using CRISPR/Cas9 technology10. A stable EGFP gene expression system was created by precise genome editing to integrate the EGFP reporter gene under the control of the GAL1 promoter with simultaneous deletion of the GAL1 gene, as shown in Fig. 1. Under these conditions, galactose metabolism in yeast cells is blocked, and galactose in the culture medium acts solely as an inducer of expression9.
The resulting system allows stable and precise induction of gene expression in the presence of galactose, providing a useful platform for further gene regulation studies in a yeast model. Integration of the EGFP reporter gene directly into the native GAL1 locus, with all its regulatory elements, ensures precise EGFP expression in a natural genomic context. Unlike plasmid-based systems, this system allows the use of rich media, such as YP with a carbon source, thereby avoiding the limitations of synthetic media, which often result in a slow growth and metabolic stress.
As yeast metabolism is notably modified by the available carbon sources, the system was validated using a whole array of both fermentative and nonfermentative media. EGFP fluorescence was measured to assess GAL1 promoter activity and expression efficiency (Fig. 2A). Under control conditions (YPD medium without galactose), the fluorescence levels were insignificant, confirming the lack of promoter induction in the presence of glucose as the sole carbon source. In contrast, evident activation of the GAL1 promoter and an increase in the fluorescence signal were observed in all the tested media supplemented with galactose.
As shown in Fig. 2B, detailed analysis of the quantitative data allowed comparison of the levels of fluorescence in cells growing in media containing various carbon sources (i.e., sucrose, lactate, glycerol, ethanol or a combination of glycerol and ethanol) and galactose. To eliminate the effects of differences in culture growth rates among various media, the data were normalized to the optical density (OD600).
The results of the measurements revealed that the highest levels of fluorescence were obtained in YPG (glycerol; 10938 ± 1232) and YPS (sucrose; 10442 ± 763.7). In contrast, in the presence of glycerol and ethanol (YPEG medium), the fluorescence signal was moderate (7888 ± 2413), which may have been due to the simultaneous effects of the two carbon sources on yeast metabolism. The lowest induction of the GAL1 promoter among the galactose media tested was observed for YPL (lactate; 7267 ± 457.8).
Importantly, clear activation of the GAL1 promoter was observed even in a medium containing glucose (the strongest catabolic repressor), demonstrating the high flexibility of this expression system. Significant differences between YPD + Gal and YPS + Gal and YPG + Gal conditions were confirmed, highlighting the importance of careful carbon source selection in the design of efficient expression systems. In contrast, no significant differences were found between the YPS + Gal and YPG + Gal media, suggesting similar efficacies of these substrates in the induction of GAL1 promoter-controlled expression.
The developed GAL1 promoter-based expression system showed great flexibility in terms of the carbon source used, including glucose, a well-known catabolic repressor. These results clearly indicate that sucrose and glycerol may be good alternatives.
The SARS-CoV-2 Mpro as an S. cerevisiae growth inhibitor under fermentative and respiratory conditions
Our expression system was used to assess the influence of the SARS-CoV-2 Mpro on the yeast under different conditions. The gene encoding the EGFP protein was linked to Mpro through two different linkers, SAVLQ and DDDDK (D4K), and both constructs were introduced into yeast as previously described9. As shown in Fig. 3A, the two linkers differed in their effects on Mpro functionality. The SAVLQ linker is specifically recognized and cleaved by Mpro (Fig. 3B), which resulted in autocatalytic release of the protease with its native structure and full activity. Active Mpro was toxic to yeast, resulting in impaired cell growth (Fig. 3C). In contrast, with the D4K linker, EGFP remains attached to the N-terminus of Mpro (Fig. 3B), which is believed to limit protease activity. As a result, the growth of yeast cells that expressed EGFP-D4K-Mpro was comparable to that of control yeast cells that expressed EGFP alone (Fig. 3C). A control strain with a GAL1 deletion (to match the genetic background of the other strains) was also included and showed similar results.
To evaluate the effects of Mpro expression under fermentative and respiratory conditions, different carbon sources were used. These included glucose (YPD) and sucrose (YPS)11 which promote rapid fermentation and limit mitochondrial activity, as well as glycerol (YPG)12,13, a mixture of glycerol and ethanol (YPEG), or lactate (YPL)14 which force cells to activate respiratory pathways. This approach enabled us to assess whether the cytotoxic effects of Mpro are more pronounced under conditions that require high mitochondrial activity.
Figure 4 compares the growth of the three strains, allowing a direct assessment of the effect of Mpro expression on yeast cell viability. No toxicity was observed in YPD medium without galactose (Fig. 4A). There were no significant differences between the strains (EGFP, EGFP-SAVLQ-Mpro and EGFP-D4K-Mpro); thus, the introduction of the construct alone did not have a negative effect on the cells.
Under strictly fermentative conditions (YPD + Gal or YPS + Gal), the growth of the EGFP-SAVLQ-Mpro strain was significantly reduced compared with that of both the EGFP control strain and the EGFP-D4K-Mpro variant. This difference was pronounced after induction with galactose in both YPD and YPS, suggesting that active Mpro (formed by the cleavage of the SAVLQ linker) negatively affects cell metabolism under predominantly fermentative conditions. Strong catabolic repression in YPD leads to the discrepancy between the effect in YPD and YPS.
Even more pronounced toxicity of Mpro was observed under conditions requiring intensive cellular respiration (YPG, YPEG and YPL media). The growth of the strain with EGFP-SAVLQ-Mpro often decreased by several orders of magnitude compared with that of the control, clearly indicating that Mpro most strongly interferes with the growth of cells that use mitochondrial metabolism as their main energy source (Fig. 4A). This relationship is perfectly illustrated in Fig. 4B, which shows that no colonies of the strain with SAVLQ were observed on plates supplemented with galactose, whereas the strains expressing EGFP alone or EGFP-D4K-Mpro formed numerous and well-formed colonies.
Interestingly, the expression of the EGFP-D₄K-Mpro construct also significantly affected yeast growth under respiratory conditions, although not as severely as the SAVLQ variant did. The data presented in Fig. 4A suggest that EGFP-D₄K-Mpro partially retained the activity of Mpro, which can disrupt cellular processes, particularly those related to mitochondrial function. As a result, although the toxic effect was not as drastic as that of the SAVLQ variant (in which the protease is fully activated), a significant reduction in colony formation could still be observed under conditions when cells depended on aerobic respiration as their primary source of ATP, ultimately leading to reduced viability and growth of yeast cells.
The results described herein indicate that the presence of SAVLQ in the construct, which is specifically recognized by Mpro, causes full protease activation and inhibition of yeast cell growth, especially when mitochondrial pathways play a dominant role in energy transduction.
Active Mpro causes mitochondrial dysfunction
To study the direct effects of active Mpro on mitochondria, a series of experiments were performed to analyze the intensity of cellular respiration and the morphology of the mitochondria. The results are shown in Figs. 5 and 6, which compare three yeast strains grown on YPG medium.
The cellular oxygen consumption rates (OCR) in the presence or absence of mitochondrial respiration modulating substances were assessed in yeast strains expressing either EGFP only (black), EGFP-D4K-Mpro (pink; less active protease) or EGFP-SAVLQ-Mpro (blue; fully active protease) at 4 and 24 h after induction (Fig. 5). The representative result of respirometry measurements after 24 h is presented (Fig. 5B). Compared with the EGFP-SAVLQ-Mpro strain, both EGFP and the EGFP-D4K-Mpro strains presented a significantly greater basal respiration rate, resulting in a more pronounced and rapid decrease in oxygen concentration over time (Fig. 5B). As shown in Fig. 5A basal respiration was significantly reduced in the EGFP-SAVLQ-Mpro strain compared to both EGFP and EGFP-D4K-Mpro strains after 24 h, demonstrating that the presence of the active form of Mpro limits aerobic metabolism in yeast cells.
Additionally, spare respiratory capacity decreased significantly in the EGFP-SAVLQ-Mpro strain (1.43 ± 0.43 nmolO2*(ml*min*OD600)−1) after 24 h compared to the EGFP (3.50 ± 0.84 nmolO2*(ml*min*OD600)−1) and EGFP-D4K-Mpro strains (3.35 ± 0.45 nmolO2*(ml*min*OD600)−1) (Fig. 5C), suggesting reduced metabolic flexibility. Bioenergetic parameter analysis (Fig. 5D) revealed that, after 4 h, the EGFP-D4K-Mpro strain showed elevated ATP-linked respiration (61.73 ± 12.34% of basal respiration) and decreased proton leak (38.27 ± 12.34% of basal respiration) compared to the Mpro-active strain (ATP synthesis 48.93 ± 8.36; proton leak 51.07 ± 8.36% of basal respiration). However, after 24 h proton leak dominated in the EGFP-SAVLQ-Mpro strain, which is consistent with respiratory impairment.
Confocal microscopy was subsequently used to assess mitochondrial morphology, the mitochondrial membrane potential (TMRM dye) and the distribution of the fusion proteins (EGFP signal) in yeast strains EGFP (control), EGFP-D4K-Mpro (less active protease), and EGFP-SAVLQ-Mpro (fully active protease), at 4 and 24 h post-induction (Fig. 6).
In the EGFP and EGFP-D4K-Mpro control strains after 4 h, EGFP expression was evenly distributed and the mitochondria showed intense, homogeneous TMRM staining, indicating intact mitochondrial potential and structure. However, after 24 h some cells in both strains showed a decrease in TMRM staining intensity, reflecting a loss of mitochondrial membrane potential. This is consistent with the observed changes in bioenergetic parameters and spare respiratory capacity measured between 4 and 24 h (Fig. 5C and D).
In contrast, the EGFP-SAVLQ-Mpro strain exhibited abnormal mitochondrial morphology and aberrantly intense accumulation of TMRM (Fig. 6; green arrows) in certain cells already at 4 h, indicative of hyperpolarization. These unusual mitochondrial abnormalities persisted at 24 h. Notably, despite this hyperpolarization, bioenergetic parameters measured did not improve and remained compromised (Fig. 5C and D). These morphological aberrations (Fig. 6; red arrows) observed at 24 h resemble previously described mitochondrial phenotypes associated with defects in essential cellular processes, such as mitochondrial protein import and maintenance of mitochondrial structure, as reported in earlier studies15.
Analyses of both the respiratory parameters (Fig. 5) and confocal images (Fig. 6) indicate that the highly active Mpro in yeast can lead to a decrease in respiratory activity and induce mitochondrial malfunction, which is manifested by changes in the membrane potential and alterations in the morphology of these organelles. In contrast, in the variant with an less active form of Mpro (D4K), oxygen metabolism and the mitochondrial structure are preserved, suggesting that when Mpro can self-cleave and be fully activated, it becomes a factor that induces mitochondrial dysfunction. These results provide evidence of the toxic effects of Mpro in yeast cells and further confirm that mitochondria are sensitive targets for the active enzyme.
