ROS-mediated synthetic growth defect caused by impaired metabolism of sphingolipids and phosphatidylserine in budding yeast

ABSTRACT

Previously, we found that yeast exhibits a strong growth defect with the combination of a lack of gene involved in structural modification of sphingolipids and repression of the phosphatidylserine synthase gene. Here we found that the double gene mutation causes reactive oxygen species-mediated cell growth defect, which is suppressed by deletion of LEM3 encoding the subunit of phospholipid flippase.

KEYWORDS:

• Sphingolipids
• phosphatidylserine
• Saccharomyces cerevisiae

Complex sphingolipids and glycerophospholipids are essential for construction of eukaryotic biomembranes having diverse physiological functions; however, it is largely unknown how different classes of the membrane lipids functionally interact to maintain biomembrane functions. In the budding yeast Saccharomyces cerevisiae, complex sphingolipids can be classified into 15 subtypes; that is, the ceramide moiety of complex sphingolipids can be classified into five types according to the hydroxylation state, and the complex sphingolipids are divided into three types due to the differences in the hydrophilic head group, inositol phosphorylceramide (IPC), mannosylinositol phosphorylceramide (MIPC), and mannosyldiinositol phosphorylceramide (M(IP)₂C) (Fig. S1(a)) [1,2]. The structural diversity is thought to be important for complex sphingolipids to exert their multiple biological functions. Therefore, disruption of the gene encoding a protein involved in creating the structural diversity of complex sphingolipids causes defects of various cellular functions [2]. Previously, we found that deletion of SCS7 or CSG2 involved in hydroxylation of the ceramide moiety and biosynthesis of MIPCs from IPCs, respectively, causes a strong synthetic growth defect when expression of the gene encoding phosphatidylserine (PS) synthase (CHO1/PSS1) is repressed, indicating a functional relationship between complex sphingolipids and/or ceramides with a specific structure and PS [3,4]. In previous [3,4] and current studies, a mutant strain that carries the CHO1 gene under the control of the tetracycline-regulatable (Tet) promoter (tet-CHO1) was used in order to repress CHO1 gene expression. tet-CHO1 scs7∆ and tet-CHO1 csg2∆ cells, but not tet-CHO1 ones, exhibited a strong growth defect in the presence of doxycycline (Dox), which represses expression of the gene under the Tet promoter (Figure 1(a)) [3,4]; however, it remains unclear how synthetic growth defects are induced by the impaired biosynthesis of sphingolipids and PS. In this study, we further characterized these genetic interactions.

Figure 1. Deletion of SCS7 or CSG2 causes a synthetic growth defect with repression of the PS synthase gene but not with genes involved in biosynthesis of PE and/or PC

(a) Cells were cultured overnight in YPD medium, and then spotted onto YPD plates with or without 10 µg/mL doxycycline (Dox) in 10-fold serial dilutions starting with a density of 0.7 OD₆₀₀ U/mL (The S. cerevisiae strains used in this study are listed in Table S1). (b and c) Effects of deletion of SCS7 or CSG2 on the growth defect caused by impaired biosynthesis of PE and/or PC. Cells were cultured overnight in YPD medium, and then spotted onto SC plates with or without the indicated concentrations of ethanolamine (Etn) (b) or choline (c) in 10-fold serial dilutions starting with a density of 0.7 OD₆₀₀ U/mL. (d) Effects of the deletion of SCS7 or CSG2 on the growth of cells lacking the Kennedy pathway. All plates were incubated at 30°C and photographed after 2 days.

In yeast, the majority of the glycerophospholipids are phosphatidylcholine (PC), phosphatidylinositol (PI), phosphatidylethanolamine (PE), and PS [3]. We investigated whether or not the deletion of SCS7 or CSG2 causes a synthetic growth defect when the biosynthesis of PE and/or PC is repressed (PI is a precursor of complex sphingolipid IPCs (Fig. S1(b)), and thus we did not examine the effect of repression of PI synthase in this study). Double deletion of PSD1 and PSD2 encoding PS decarboxylase results in a lethal phenotype when ethanolamine (Etn), which is converted to PE via the Kennedy pathway, is depleted from the medium [5] (Fig. S1(b)). Similarly, cells lacking CHO2 and OPI3 encoding phospholipid methyltransferase hardly grow in the absence of choline, which is converted to PC via the Kennedy pathway [6] (Fig. S1(b)). As shown in Figure 1(b) and (c), psd1∆ psd2∆ or cho2∆ opi3∆ cells exhibited delayed cell growth when the concentration of Etn or choline in the SC medium was gradually reduced. Although growth of psd1∆ psd2∆ scs7∆ cells in the presence of 1 mM Etn was slightly delayed as compared with that of psd1∆ psd2∆ cells, the deletion of SCS7 did not reduce the rate of cell growth in the presence of 150 or 100 µM Etn (Figure 1(b)). Deletion of CSG2 did not affect the growth of psd1Δ psd2Δ double mutant cells (Figure 1(b)). The deletion of SCS7 or CSG2 did not affect the delay in the cell growth of cho2Δ opi3Δ double mutant cells with a reducing concentration of choline in the medium (Figure 1(c)). In addition, the deletion of SCS7 or CSG2 did not affect the growth of cells lacking the Kennedy pathway (ept1Δ cpt1Δ double mutant cells) (Figure 1(d)). Collectively, these results suggested that cells lacking SCS7 or CSG2 exhibit a marked defect of cell growth when the PS synthase gene is repressed, but not when the biosynthesis of PE and/or PC is reduced.

Several lines of evidence have suggested that aberrant metabolism of sphingolipids causes an increase in reactive oxygen species (ROS) levels, which has a detrimental effect on cell growth [6–9]. Thus, we next examined the ROS levels in CHO1-repressed scs7∆ or csg2∆ cells. To observe ROS, 2′,7′-dichlorodihydrofluorescein diacetate (H₂DCFDA) was used as an indicator of ROS in cells [9]. As shown in Figure 2(a), Dox-treated tet-CHO1 scs7∆ and tet-CHO1 csg2∆ cells, but not untreated cells, exhibited increases in intracellular ROS levels; however, the accumulation of ROS was not observed in Dox-treated tet-CHO1 cells. The increases in the ROS levels were suppressed by the addition of the membrane-permeable antioxidant N-acetylcysteine (NAC) (Figure 2(a)). Notably, the cell growth defect in Dox-treated tet-CHO1 scs7∆ and tet-CHO1 csg2∆ cells was also suppressed by the addition of NAC (Figure 2(b)). These results suggested that the synthetic growth defect caused by abnormal metabolism of sphingolipids and PS is at least partly mediated by ROS accumulation.

Figure 2. Increases in ROS levels in CHO1-repressed scs7∆ or csg2∆ cells

(a) Cells were cultured overnight in YPD medium at 30°C, diluted (0.1 OD₆₀₀ units/mL) in fresh YPD medium with or without 10 µg/mL Dox and 20 mM N-acetylcysteine (NAC), and then incubated for 7 h at 30°C. The cells were stained with 10 µM H₂DCFDA for 1 h at 30°C. The cells were collected by centrifugation and viewed under a fluorescence microscope. The graphs indicate the frequency distribution of the fluorescence intensity in individual cells. The fluorescence intensity was quantified with ImageJ software (National Institutes of Health, Bethesda, MD, USA). Data represent the values for 100 cells for individual strains. The results are typical of three independent experiments. (b) Cells were cultured overnight in YPD medium, spotted onto YPD plates with or without 10 µg/mL Dox and 20 mM NAC, and then photographed as described in Figure 1.

Recently, it was reported that deletion of CHO1 causes hypersensitivity to plasma membrane stresses, which include hypotonic shock and reduction in total levels of sphingolipids on treatment with myriocin, an inhibitor of serine palmitoyltransferase; however, the hypersensitivity is partly canceled by LEM3 encoding the essential subunit of phospholipid flippases (Dnf1 and Dnf2), suggesting that phospholipid regulation via Dnf1/2 antagonistically acts on the functions of PS under plasma membrane stresses [10]. We found that the growth defect caused by CHO1 repression and the deletion of SCS7 or CSG2 is also suppressed by the deletion of LEM3 (Figure 3(a)). In addition, the increases in ROS levels caused by the repression of CHO1 and the deletion of SCS7 or CSG2 were suppressed by the deletion of LEM3 (Figure 3(b)). Thus, it was indicated that the defect of phospholipid flippase can rescue the synthetic growth defect caused by the repression of PS synthase and the deletion of SCS7 or CSG2.

Figure 3. Deletion of LEM3 suppresses the growth defect and ROS production caused by the repression of CHO1 and the deletion of SCS7 or CSG2.

(a) Cells were cultured overnight in YPD medium, spotted onto YPD plates with or without 10 µg/mL Dox, and then photographed as described in Fig. 1. (b) Effect of deletion of LEM3 on the increases in ROS levels in Dox-treated tet-CHO1 scs7∆ and tet-CHO1 csg2∆ cells. The staining with H₂DCFDA was performed as described in Fig. 2(a). The results are typical of three independent experiments.

It remains unclear how increases in ROS levels are induced in CHO1-repressed scs7∆ or csg2∆ cells. It was reported that reduced activity of the TORC2/Ypk1 signaling pathway involved in the response to plasma membrane stresses and sphingolipid homeostasis causes ROS accumulation, and ROS upregulates the activity of TORC2 through a feedback loop [8]. Furthermore, it was also reported that, under plasma membrane stresses, PS at the inner leaflet of plasma membrane plays an important role in actin reorganization, which is mediated by activation of TORC2/Ypk1 followed by Rho1 GTPase [10]. Thus, investigation on the relationship of TORC2/Ypk1- and Rho1-mediated signaling in ROS accumulation in CHO1-repressed scs7∆ or csg2∆ cells is required in the future. It should be noted that loss of hydroxylation of sphingolipids on the deletion of SCS7 or the defect of MIPC biosynthesis affects the physical properties of plasma membranes [11,12], which may induce some stressful conditions on plasma membranes.

The present study indicates that double abnormal metabolism of sphingolipids and PS causes a ROS-mediated cell growth defect, which is suppressed by the deletion of LEM3. This information will provide a novel insight into the functional relationship between sphingolipids and glycerophospholipids.

Author contributions

M.T. and T.T. conceived the study, designed the experiments, and wrote the manuscript. T.T., A.U., A.K., C.T., and M.T. performed the experiments and analyzed the data.

Acknowledgments

We wish to thank Drs. O. Kuge, T. Ogishima, and N. Miyata (Kyushu University) for discussions. This study was funded by a KAKENHI (18H02139) from the Ministry of Education, Culture, Sports, Science, and Technology, the Institute for Fermentation, Osaka, and the Noda Institute for Scientific Research, Japan.

Disclosure statement

No potential conflict of interest was reported by the authors.

Supplementary material

Supplemental data for this article can be accessed here.