Dihydroxyacetone (DHA) is the smallest ketotriose, and it is utilized by many organisms as an energy source. However, at higher concentrations, DHA becomes toxic towards several organisms including the budding yeast Saccharomyces cerevisiae. In the present study, we show that DHA toxicity is due to its spontaneous conversion to methylglyoxal (MG) within yeast cells. A mutant defective in MG-metabolizing enzymes (glo1Δgre2Δgre3Δ) exhibited higher susceptibility to DHA. Intracellular MG levels increased following the treatment of glo1Δgre2Δgre3Δ cells with DHA. We previously reported that MG depolarized the actin cytoskeleton and changed vacuolar morphology. We herein demonstrated the depolarization of actin and morphological changes in vacuoles following a treatment with DHA. Furthermore, we found that both MG and DHA caused the morphological change in nucleus, and inhibited the nuclear division. Our results suggest that the conversion of DHA to MG is a dominant contributor to its cytotoxicity.

Introduction

Dihydroxyacetone (DHA, CH2OHCOCH2OH) is the smallest ketotriose and is regarded as an energy source for different types of cells from bacteria to mammals [1]. In bacteria, DHA is formed by the oxidation of glycerol or by the aldol cleavage of fructose 6-phosphate [2,3]. Methylotrophic yeasts produce DHA as a primary product of the assimilation of methanol [4,5]. In mammals, DHA is an important gluconeogenic precursor [6,7]; however, limited information is currently available on its metabolic origin.

The budding yeast Saccharomyces cerevisiae has the ability to grow in medium containing DHA as the sole carbon source [8]. Dihydroxyacetone kinases (DAKs) (Figure 1) play a principal role in the assimilation process of DHA in S. cerevisiae [8]. DHA is converted by DAKs to dihydroxyacetone phosphate (DHAP), a glycolytic intermediate. However, higher concentrations of DHA exert toxic effects in yeast [8]. DHA is presumptively thought to be converted to formaldehyde by the action of transketolase (Tkl1 and Tkl2). In this reaction, glyceraldehyde 3-phosphate is predicted to function as an acceptor of a ketol moiety of DHA, and subsequently, xylulose 5-phosphate and formaldehyde are produced [8]. In the detoxification pathway of formaldehyde, glutathione spontaneously forms a complex with formaldehyde to give S-hydroxymethylglutathione, and this compound is converted to S-formylglutathione in the presence of NAD(P)+ by glutathione-dependent formaldehyde dehydrogenase (Sfa1), which is then hydrolyzed to glutathione and formate by S-formylglutathione hydrolase (Yjl068c). Formate is further degraded to CO2 in the presence of NADH by the action of formate dehydrogenase (Fdh1) (Figure 1). The deletion of SFA1 impaired adaptation for growth and survival in medium containing DHA as a sole source of carbon [9]. Nevertheless, the overexpression of SFA1 exhibited normal sensitivity to DHA [9]. Furthermore, although the generation time of cell growth increased ∼2-fold, the toxic effects of DHA were efficiently suppressed by glucose [8], yet the mechanism remains to be solved.

Scheme for the metabolism of MG and DHA.

Figure 1.
Scheme for the metabolism of MG and DHA.

Abbreviations used are: DHA, dihydroxyacetone; DHAP, dihydroxyacetone phosphate; GA3P, glyceraldehyde 3-phosphate; MG, methylglyoxal; S-LG, S-d-lactoylglutathione; Dak1/Dak2, dihydroxyacetone kinase 1/2; Tpi1, triosephosphate isomerase; Glo1, glyoxalase I; Glo2, glyoxalase II; Gre2, putative methylglyoxal reductase and Gre3, aldose reductase; Tkl1/Tkl2, transketolase 1/2; Sfa1, formaldehyde dehydrogenase; Yjl068c, S-formylglutathione hydrolase.

Figure 1.
Scheme for the metabolism of MG and DHA.

Abbreviations used are: DHA, dihydroxyacetone; DHAP, dihydroxyacetone phosphate; GA3P, glyceraldehyde 3-phosphate; MG, methylglyoxal; S-LG, S-d-lactoylglutathione; Dak1/Dak2, dihydroxyacetone kinase 1/2; Tpi1, triosephosphate isomerase; Glo1, glyoxalase I; Glo2, glyoxalase II; Gre2, putative methylglyoxal reductase and Gre3, aldose reductase; Tkl1/Tkl2, transketolase 1/2; Sfa1, formaldehyde dehydrogenase; Yjl068c, S-formylglutathione hydrolase.

Previous studies reported that DHA exerted cytotoxic effects not only in yeast, but also in several organisms, such as Escherichia coli [10] and Trypanosoma brucei [11]. Although the mechanisms by which the toxic effects of DHA are exerted have not yet been elucidated in detail, the involvement of methylglyoxal (MG, CH3COCHO) in the toxicity of DHA in E. coli cells has been reported [10]. MG is a metabolite derived from glycolysis and has the ability to inhibit the growth of cells in all types of organisms [12,13]. This toxic metabolite is mainly metabolized to d-lactate through a ubiquitous glutathione-dependent glyoxalase system consisting of glyoxalase I (Glo1) and glyoxalase II (Glo2) [13] (Figure 1). DHA may be non-enzymatically converted to MG [14], and a gloA mutant, defective in glyoxalase I, of E. coli showed greater sensitivity to DHA [10]. Therefore, a model in which DHA toxicity is due to the formation of MG has been proposed in E. coli [10]; however, a glo1Δ mutant of S. cerevisiae exhibited similar sensitivity to DHA as wild-type cells in medium containing DHA as the sole source of carbon [8]. However, the cultivation of yeast cells in medium containing DHA yielded large amounts of N-ε-carboxyethyl-lysine, an MG-derived advanced glycation end-product (AGE) [15], in yeast cells [9,16]. Therefore, it currently remains controversial whether the conversion of DHA to MG is committed to DHA toxicity in S. cerevisiae.

To elucidate the relationship between DHA toxicity and MG in S. cerevisiae, we investigated some phenotypes that we previously found to be induced by MG, such as the depolarization of the actin cytoskeleton [17] and the defragmentation of vacuoles [18]. We showed that the addition of DHA to glucose-containing medium increased intracellular MG levels. Mutants defective in MG-detoxification enzymes exhibited increased sensitivity to DHA, which was suppressed by the overexpression of DAK1, whereas MG sensitivity was not suppressed. We also demonstrated that actin depolarization and changes in vacuolar morphology occurred following a treatment with DHA, and these phenotypes were suppressed by the overexpression of DAK1. Furthermore, we found that DHA changed nuclear morphology and inhibited nuclear division, both of which were suppressed by the overexpression of DAK1. Morphological change in the nucleus and inhibition of nuclear division were also observed in cells treated with MG. These results indicate that the formation of MG from DHA is attributed to DHA-induced toxicity in yeast.

Experimental

Media and reagents

The media used were YPD (2% glucose, 1% yeast extract, and 2% peptone) and SD (synthetic dextrose; 2% glucose, 0.67% yeast nitrogen base without amino acids) with appropriate amino acids and bases being added where necessary. MG and dihydroxyacetone were purchased from Sigma and Nacalai Tesque.

Plasmids

The plasmids used in the present study are listed in Table 1.

Table 1
Plasmids used in the present study.
Plasmid Description Source/reference 
YCp50-PKC1R398P YCp50 (CEN type, URA3 marker) harboring PKC1R398P [50
pRS426-DAK1 pRS426 (2µ type, URA3 marker) harboring DAK1 The present study 
pRS426-DAK2 pRS426 (2µ type, URA3 marker) harboring DAK2 The present study 
YIp-NUP116-GFP NUP116-GFP in an integrate-type, URA3 marker plasmid [19
pRS304-NUP116-GFP NUP116-GFP in an integrate-type, TRP1 marker plasmid The present study 
pRS306-ATG18-GFP ATG18-GFP in an integrate-type, URA3 marker plasmid [18
pUGΔLeu2 pUC19 backbone, for deletion of GLO1 The present study 
Plasmid Description Source/reference 
YCp50-PKC1R398P YCp50 (CEN type, URA3 marker) harboring PKC1R398P [50
pRS426-DAK1 pRS426 (2µ type, URA3 marker) harboring DAK1 The present study 
pRS426-DAK2 pRS426 (2µ type, URA3 marker) harboring DAK2 The present study 
YIp-NUP116-GFP NUP116-GFP in an integrate-type, URA3 marker plasmid [19
pRS304-NUP116-GFP NUP116-GFP in an integrate-type, TRP1 marker plasmid The present study 
pRS306-ATG18-GFP ATG18-GFP in an integrate-type, URA3 marker plasmid [18
pUGΔLeu2 pUC19 backbone, for deletion of GLO1 The present study 

To construct pRS426-DAK1 and pRS426-DAK2, the genomic fragment containing DAK1 or DAK2 was amplified by PCR using the following primer sets: DAK1-F-EcoRI, 5′-ACGTAATTATGGGAATTCGTTGATGATTAG-3′; and DAK1-R, 5′-ACCCAGAATCAGCAATACGGAATCTACCGC-3′; or DAK2-F-SalI, 5′-TAGCGTTGGTGGTCGACTACTCTCGGAAGA-3′, and DAK2-R-BamHI, 5′-CAGACAACACAGGGGATCCGAGAAGGAATC­-3′. The amplified DNA fragment was digested with EcoRI/HindIII or SalI/BamHI, and the resultant fragment was ligated into the corresponding sites in pRS426.

The construction of pRS306-ATG18-GFP was described recently [18]. pRS306-ATG18-GFP was digested with ClaI, and linearized DNA was introduced at the ATG18 locus.

To construct pUGΔLeu2, the fragment containing a part of the GLO1 ORF was amplified by PCR with the following primers: 5′-TATAGCATGCAACTGCCAAAATTTTCGGGTA-3′ and 5′-GGGGCATGCAAAATGGTGTAACACAAATCT-3′. The 740-bp fragment yielded by PCR was digested with HindIII and BamHI, and was cloned into the HindIII/BamHI sites of pUC19 (pUBHa). pUBHa was digested with EcoRV and HpaI in order to delete a part of the GLO1 ORF, and it was replaced with the LEU2 gene to form pUGΔLeu2.

YIp-NUP116-GFP [19] was digested by SacI and KpnI, and the resultant fragment was cloned into the SacI and KpnI sites of pRS304 to yield pRS304-NUP116-GFP.

Strains

The yeast strains used in the present study are listed in Table 2. The S. cerevisiae strains used had the YPH250 and BY4741 backgrounds. To disrupt GLO1, pUGΔLeu2 was digested with HindIII and BamHI, and the resultant HindIII–BamHI fragment was introduced into YPH250. Gene deletions of GRE2 and GRE3 were constructed by PCR-based methods with KanMX or his5+ selective markers [20]. The deletion allele was amplified by PCR from BY4741-based deletion mutants (Invitrogen), and the corresponding locus of YPH250 was disrupted using PCR products.

Table 2
Saccharomyces cerevisiae strains used in the present study.
Strain Relevant genotype/description Source/reference 
YPH250 MATa trp1-Δ1 his3-Δ200 leu2-Δ1 lys2-801 ade2-101 ura3-52 Laboratory stock 
BY4741 MATa his3Δ1 leu2Δ0 met15Δ0 ura3Δ0 Laboratory stock 
glo1Δ YPH250, glo1Δ::LEU2 The present study 
glo1Δ gre2Δ YPH250, glo1Δ::LEU2 gre2Δ::KanMX4 The present study 
glo1Δ gre2Δgre3Δ YPH250, glo1Δ::LEU2 gre2Δ::KanMX4 gre3Δ::his5+ The present study 
adh1Δ BY4741, adh1Δ:: KanMX4 Invitrogen 
adh2Δ BY4741, adh2Δ:: KanMX4 Invitrogen 
adh3Δ BY4741, adh3Δ:: KanMX4 Invitrogen 
adh4Δ BY4741, adh4Δ:: KanMX4 Invitrogen 
adh5Δ BY4741, adh5Δ:: KanMX4 Invitrogen 
adh6Δ BY4741, adh6Δ:: KanMX4 Invitrogen 
adh7Δ BY4741, adh7Δ:: KanMX4 Invitrogen 
bdh1Δ BY4741, bdh1Δ:: KanMX4 Invitrogen 
sfa1Δ BY4741, sfa1Δ:: KanMX4 Invitrogen 
Strain Relevant genotype/description Source/reference 
YPH250 MATa trp1-Δ1 his3-Δ200 leu2-Δ1 lys2-801 ade2-101 ura3-52 Laboratory stock 
BY4741 MATa his3Δ1 leu2Δ0 met15Δ0 ura3Δ0 Laboratory stock 
glo1Δ YPH250, glo1Δ::LEU2 The present study 
glo1Δ gre2Δ YPH250, glo1Δ::LEU2 gre2Δ::KanMX4 The present study 
glo1Δ gre2Δgre3Δ YPH250, glo1Δ::LEU2 gre2Δ::KanMX4 gre3Δ::his5+ The present study 
adh1Δ BY4741, adh1Δ:: KanMX4 Invitrogen 
adh2Δ BY4741, adh2Δ:: KanMX4 Invitrogen 
adh3Δ BY4741, adh3Δ:: KanMX4 Invitrogen 
adh4Δ BY4741, adh4Δ:: KanMX4 Invitrogen 
adh5Δ BY4741, adh5Δ:: KanMX4 Invitrogen 
adh6Δ BY4741, adh6Δ:: KanMX4 Invitrogen 
adh7Δ BY4741, adh7Δ:: KanMX4 Invitrogen 
bdh1Δ BY4741, bdh1Δ:: KanMX4 Invitrogen 
sfa1Δ BY4741, sfa1Δ:: KanMX4 Invitrogen 

To add a GFP tag at the C-terminus of Nup116, YIp-NUP116-GFP [19] or pRS304-NUP116-GFP was digested with EcoRI, and the linearized fragment was introduced into the locus of NUP116.

Spot assay

Cells were cultured in SD medium until the early log phase of growth, and then diluted to A610 = 0.1 with sterilized 0.85% NaCl solution. Cell suspensions were diluted serially (1 : 10) with sterilized 0.85% NaCl solution and spotted (4 µl) onto SD agar plates containing various concentrations of DHA or MG.

Measurement of intracellular MG content

Cells of the wild-type strain and glo1Δgre2Δgre3Δ mutant were cultured in SD minimal medium until A610 = 0.5–0.7 and exposed to MG or DHA. Cells (180 OD unit) were collected, and cellular MG levels were measured as described previously [21,22].

Actin staining

Cells were fixed with formaldehyde (final concentration, 4%) at room temperature for 1 h. After fixation, cells were harvested, washed twice with phosphate-buffered saline (PBS; pH 7.4), and suspended in 30 µl of PBS. Rhodamine–phalloidin (Molecular Probes) was added to the cell suspension to a final concentration of 33 units/ml (1.1 µM), and the cell suspension was then incubated at 4°C in the dark overnight. Cells were collected by centrifugation and washed twice with PBS, and the distribution of actin was subsequently observed using a fluorescence microscope.

Western blotting

Procedures for the detection of Mpk1 were described previously [17]. Anti-phospho-p44/42 MAP kinase (#9101; Cell Signaling) and anti-Mpk1 (sc6803; Santa Cruz Biotechnology) were used as the primary antibody. Immunoreactive bands were visualized with an Immobilon Western chemiluminescent horseradish peroxidase substrate (Millipore) using a LAS-4000 mini-imaging system (Fujifilm).

Vacuolar staining

Yeast vacuoles were visualized in vivo by labeling with FM4-64 (Molecular Probes) [23].

Measurement of nuclear division

Cells were cultured in SD medium until the A610 reached 0.3–0.5, and 1 µg/ml Hoechst 33342 (Hoechst AG) (Molecular Probes, Inc., Eugene, OR, U.S.A.) was added to stain the nuclei. To quantify the distribution of nuclei, cells with a large bud (bud diameter approximately two-thirds that of the mother cell) containing the nucleus were counted after taking pictures with fluorescence microscopy. Approximately 100–200 cells were counted for each experiment.

Fluorescence microscopy

The fluorescence microscopes BX51 and BX63 (OLYMPUS) equipped with the digital cameras DP70 (OLYMPUS) and ORCA-R2 (HAMAMATSU PHOTONICS) were used.

Results

Mutants defective in MG metabolism show DHA sensitivity

MG is metabolized to lactate through two distinct pathways, i.e. the glutathione-dependent glyoxalase pathway and the redox pathway (Figure 1). MG is converted to d-lactate through S-d-lactoylglutathione by the sequential reactions of Glo1 and Glo2. On the other hand, MG is reduced to l-lactaldehyde by methylglyoxal reductase [24] and is then oxidized to l-lactate by lactaldehyde dehydrogenase [25]. Alternatively, MG is reduced to acetol (hydroxyl-2-propane, hydroxyacetone) by aldose reductase. GRE2 and GRE3 have been reported to code for methylglyoxal reductase and aldose reductase, respectively [13,26,27]. Mutants defective in enzymes constituting these pathways exhibit increased susceptibility to MG [2729]. To clarify whether the toxicity of DHA is exerted through the formation of MG, we examined the susceptibility of glo1Δ, glo1Δgre2Δ, and glo1Δgre2Δgre3Δ mutants to DHA.

Firstly, we spotted these mutants on glucose media containing low concentrations of MG, which did not affect the growth of wild-type cells, in an attempt to verify the effectiveness of these mutants in evaluations of sensitivity to MG. As shown in Figure 2A, additive effects in terms of MG sensitivity were observed; i.e. the extent of MG sensitivity was in the order of glo1Δgre2Δgre3Δ > glo1Δgre2Δ >glo1Δ.

DHA-induced increases in intracellular MG levels.

Figure 2.
DHA-induced increases in intracellular MG levels.

(A) Wild-type (YPH250), glo1Δ, glo1Δgre2Δ, and glo1Δgre2Δgre3Δ cells were cultured in SD medium until A610 = 0.3, 4 µl of each cell suspension was spotted onto SD agar plates with or without DHA and MG, and cells were incubated at 28°C for 3–4 days. (B) Cells (YPH250) of the wild-type and glo1Δgre2Δgre3Δ mutant were cultured in SD medium until A610 = 0.5–0.7, treated with 10 mM MG for 60 min, or 200 or 400 mM DHA for 90 min, and intracellular MG levels were assessed. Similarly, intracellular MG levels in glo1Δgre2Δgre3Δ cells carrying an empty plasmid (pRS426) or pRS426-DAK1 were measured. Data are for three independent experiments (average ± SD). (C) Wild-type (YPH250) and glo1Δgre2Δgre3Δ cells carrying either an empty plasmid (vector, pRS426), pRS426-DAK1, or pRS426-DAK2 were cultured in SD medium until A610 = 0.3, 4 µl of each cell suspension was spotted onto SD agar plates with or without DHA (100 mM) and MG (1 mM), and cells were then incubated at 28°C for 3 days. (D) Cells of the wild-type (BY4741) and each mutant (adh17Δ, bdh1Δ, and sfa1Δ) were cultured in SD medium until A610 = 0.3, 4 µl of each cell suspension was spotted onto SD agar plates with or without MG (8 mM), and cells were incubated at 28°C for 3 days.

Figure 2.
DHA-induced increases in intracellular MG levels.

(A) Wild-type (YPH250), glo1Δ, glo1Δgre2Δ, and glo1Δgre2Δgre3Δ cells were cultured in SD medium until A610 = 0.3, 4 µl of each cell suspension was spotted onto SD agar plates with or without DHA and MG, and cells were incubated at 28°C for 3–4 days. (B) Cells (YPH250) of the wild-type and glo1Δgre2Δgre3Δ mutant were cultured in SD medium until A610 = 0.5–0.7, treated with 10 mM MG for 60 min, or 200 or 400 mM DHA for 90 min, and intracellular MG levels were assessed. Similarly, intracellular MG levels in glo1Δgre2Δgre3Δ cells carrying an empty plasmid (pRS426) or pRS426-DAK1 were measured. Data are for three independent experiments (average ± SD). (C) Wild-type (YPH250) and glo1Δgre2Δgre3Δ cells carrying either an empty plasmid (vector, pRS426), pRS426-DAK1, or pRS426-DAK2 were cultured in SD medium until A610 = 0.3, 4 µl of each cell suspension was spotted onto SD agar plates with or without DHA (100 mM) and MG (1 mM), and cells were then incubated at 28°C for 3 days. (D) Cells of the wild-type (BY4741) and each mutant (adh17Δ, bdh1Δ, and sfa1Δ) were cultured in SD medium until A610 = 0.3, 4 µl of each cell suspension was spotted onto SD agar plates with or without MG (8 mM), and cells were incubated at 28°C for 3 days.

Next, we investigated the sensitivity of these mutants to DHA. A previous study showed that DHA (200 mM) exhibited toxicity when it was used as the sole source of carbon in medium, and this toxicity was suppressed by the addition of glucose [8]. As shown in Figure 2A, wild-type cells grew in glucose medium containing 100 mM DHA; however, these mutants showed susceptibility to DHA even in the glucose-containing medium, and the extent of susceptibility to DHA coincided with that to MG, i.e. glo1Δgre2Δgre3Δ cells exhibited the greatest sensitivity to DHA.

A previous study reported that DHA passed through the plasma membrane by simple diffusion, and its uptake occurred independently of the carbon source [8]. Since DHA is non-enzymatically converted to MG [14], we investigated whether extracellularly added DHA is taken up by cells and affects MG levels within yeast cells. The steady-state level of MG in glo1Δgre2Δgre3Δ cells was ∼1.5-fold higher than that in wild-type cells because of its insufficient metabolism. After a treatment with 200 mM DHA for 90 min, intracellular MG levels in glo1Δgre2Δgre3Δ cells were similar to those in wild-type cells treated with 10 mM MG (Figure 2B). These results suggest that the susceptibility of the mutants defective in the MG metabolic pathways to DHA may be due to an increase in the intracellular MG levels formed from DHA, which is taken up from the medium by cells.

A previous study reported that a yeast mutant defective in DAKs (dak1Δdak2Δ) showed hypersensitivity to DHA; however, the overexpression of DAK1 or DAK2 restored growth in DHA medium [8], which is expected to convert DHA to DHAP in the assimilating pathway (Figure 1). We investigated whether the susceptibility of glo1Δgre2Δgre3Δ cells to DHA in glucose medium was also suppressed by the overexpression of DAK1 or DAK2. As shown in Figure 2C, the overexpression of DAK1 markedly restored the growth of glo1Δgre2Δgre3Δ cells in glucose medium containing DHA; however, the suppressive effects of DAK2 were not as strong as those of DAK1. Neither DAK1 nor DAK2 suppressed the susceptibility of the glo1Δgre2Δgre3Δ mutant to MG (Figure 2C). These results suggest that DAKs avert the formation of MG from DHA in yeast cells. To verify more directly whether DHA is converted to MG, and Dak1 averts the formation of MG, we measured the intracellular levels of MG in glo1Δgre2Δgre3Δ cells overexpressing DAK1 following the treatment with DHA. As shown in Figure 2B, an increase in the level of intracellular MG was suppressed by overexpression of DAK1. These results indicate that DHA is converted to MG in yeast cells, and Dak1 averts it by converting DHA to DHAP, a glycolytic intermediate.

A previous study reported that DHA is likely to be converted to formaldehyde, and this reactive aldehyde has been proposed as a cause of the toxicity of DHA [9]. Glutathione-dependent formaldehyde dehydrogenase (Sfa1) was shown to play a role in the detoxification of formaldehyde derived from DHA, and sfa1Δ cells showed impaired adaptation to DHA [9]. To clarify whether Sfa1 is involved in the metabolism of MG in yeast cells, the susceptibility of an sfa1Δ mutant to MG (8 mM) was examined. As shown in Figure 2D, sfa1Δ cells exhibited MG sensitivity. However, as shown in Supplementary Figure S1A, 1 mM MG, the concentration at which glo1Δ cells did not grow, gave no obvious effect on growth of sfa1Δ cells. Therefore, even though Sfa1 is active towards MG [30], Sfa1 seems less effective on detoxification of MG. Since Sfa1 is an enzyme involved in oxidation of formaldehyde to formic acid, an sfa1Δ mutant exhibited sensitivity to formaldehyde (Supplementary Figure S1B). Meanwhile, sfa1Δ cells did not grow on the SD agar plates containing 100 mM DHA, but glo1Δ cells showed a poor growth. Nonetheless, neither sfa1Δ nor glo1Δ cells were able to grow in 125 mM DHA-containing medium (Supplementary Figure S1A), suggesting that DHA is converted to formaldehyde, besides DHA being converted to MG, and both Sfa1 and Glo1 seem concertedly detoxify the toxicity of DHA.

Sfa1 is a bifunctional enzyme that exhibits alcohol dehydrogenase and formaldehyde dehydrogenase activities [3032]. Therefore, we examined the susceptibility of other mutants defective in alcohol dehydrogenase (adh17Δ, and bdh1Δ) to MG; however, none of the mutants, except for an adh3Δ mutant, exhibited MG sensitivity (Figure 2D). The adh3Δ mutant exhibited poor growth in glucose medium without MG and was slightly sensitive to MG, although the reason for it is yet to be determined.

DHA depolarizes actin patches

S. cerevisiae exhibits polarized growth, and the actin cytoskeleton has a strong influence on cellular polarity [33]. F-actin forms dynamic structures such as cortical actin patches and actin cables in yeast cells. Actin patches in logarithmically growing S. cerevisiae cells accumulate in the bud, which affects the polarity of the cell. We previously reported that MG induced the depolarization of actin patches, which is attributed, in part, to the inhibition of polarized growth in the presence of MG [17]. Therefore, to establish whether DHA causes the same phenomenon as MG in terms of its effects on the polarity of actin patches, we investigated the depolarization of actin patches following a treatment with DHA.

We initially verified that DHA did not decrease cell viability within the range of the concentrations used (∼400 mM) and during the exposure period (∼120 min) (Figure 3A), indicating that the phenotypes obtained following the treatment with DHA were not as a consequence of cell death.

DHA induces actin depolarization.

Figure 3.
DHA induces actin depolarization.

(A) glo1Δgre2Δgre3Δ cells were cultured in SD medium until A610 = 0.3 and treated with DHA. At the prescribed time, aliquots were plated onto YPD agar plates. After an incubation at 28°C for 2 days, colonies were counted. Viability at 0 min was taken as 100%. Data are for three independent experiments (average ± standard deviation, n ≥ 3). (B) Wild-type (YPH250) and glo1Δgre2Δgre3Δ cells were cultured in SD medium until A610 = 0.3 and were then treated with 200 mM DHA for 90 min or 10 mM MG for 60 min. Cells were stained for actin with rhodamine–phalloidin, and observed using a fluorescence microscope. Bar, 5 µm. (C) Wild-type (YPH250) and glo1Δgre2Δgre3Δ cells were cultured in SD medium until A610 = 0.3 and treated with 200 mM DHA for the prescribed time as indicated in the figure. Cells were stained for actin with rhodamine–phalloidin. The proportion of cells with depolarized actin was assessed by counting cells in which actin had not accumulated in the bud. More than ∼200 cells were counted in each experiment. Data are for three independent experiments (average ± standard deviation, n ≥ 3). (D) glo1Δgre2Δgre3Δ cells were cultured in SD medium until A610 = 0.3 and treated with DHA for the prescribed concentrations as indicated in the figure. The proportion of cells with depolarized actin was measured as described in (C). (E) glo1Δgre2Δgre3Δ cells carrying either an empty plasmid (vector, pRS426) or pRS426-DAK1 were cultured in SD medium until A610 = 0.3 and treated with 200 mM DHA for the prescribed time as indicated in the figure. The proportion of cells with depolarized actin was assessed as described in (C). (F) Viabilities of glo1Δgre2Δgre3Δ cells following the treatment with formaldehyde (0.1 and 0.6 mM) were measured as described in (A). glo1Δgre2Δgre3Δ cells were cultured in SD medium until A610 = 0.3 and were then treated with 0.1 mM HCHO for 120 min. Cells were stained for actin with rhodamine–phalloidin, and observed using a fluorescence microscope. Bar, 5 µm. (G) glo1Δgre2Δgre3Δ cells carrying either an empty plasmid (vector, YCp50) or YCp50-PKC1R398P were cultured in SD medium until A610 = 0.3, and treated with 200 mM DHA for the prescribed time as indicated in the figure. The proportion of cells with depolarized actin was assessed as described in (C).

Figure 3.
DHA induces actin depolarization.

(A) glo1Δgre2Δgre3Δ cells were cultured in SD medium until A610 = 0.3 and treated with DHA. At the prescribed time, aliquots were plated onto YPD agar plates. After an incubation at 28°C for 2 days, colonies were counted. Viability at 0 min was taken as 100%. Data are for three independent experiments (average ± standard deviation, n ≥ 3). (B) Wild-type (YPH250) and glo1Δgre2Δgre3Δ cells were cultured in SD medium until A610 = 0.3 and were then treated with 200 mM DHA for 90 min or 10 mM MG for 60 min. Cells were stained for actin with rhodamine–phalloidin, and observed using a fluorescence microscope. Bar, 5 µm. (C) Wild-type (YPH250) and glo1Δgre2Δgre3Δ cells were cultured in SD medium until A610 = 0.3 and treated with 200 mM DHA for the prescribed time as indicated in the figure. Cells were stained for actin with rhodamine–phalloidin. The proportion of cells with depolarized actin was assessed by counting cells in which actin had not accumulated in the bud. More than ∼200 cells were counted in each experiment. Data are for three independent experiments (average ± standard deviation, n ≥ 3). (D) glo1Δgre2Δgre3Δ cells were cultured in SD medium until A610 = 0.3 and treated with DHA for the prescribed concentrations as indicated in the figure. The proportion of cells with depolarized actin was measured as described in (C). (E) glo1Δgre2Δgre3Δ cells carrying either an empty plasmid (vector, pRS426) or pRS426-DAK1 were cultured in SD medium until A610 = 0.3 and treated with 200 mM DHA for the prescribed time as indicated in the figure. The proportion of cells with depolarized actin was assessed as described in (C). (F) Viabilities of glo1Δgre2Δgre3Δ cells following the treatment with formaldehyde (0.1 and 0.6 mM) were measured as described in (A). glo1Δgre2Δgre3Δ cells were cultured in SD medium until A610 = 0.3 and were then treated with 0.1 mM HCHO for 120 min. Cells were stained for actin with rhodamine–phalloidin, and observed using a fluorescence microscope. Bar, 5 µm. (G) glo1Δgre2Δgre3Δ cells carrying either an empty plasmid (vector, YCp50) or YCp50-PKC1R398P were cultured in SD medium until A610 = 0.3, and treated with 200 mM DHA for the prescribed time as indicated in the figure. The proportion of cells with depolarized actin was assessed as described in (C).

Actin patches in glo1Δgre2Δgre3Δ cells were depolarized following the treatment with 200 mM DHA (Figure 3B), and depolarization peaked after a 90-min treatment (Figure 3C). In contrast, actin polarity in wild-type cells was not affected by DHA (Figure 3B,C). The proportion of cells with depolarized actin patches in glo1Δgre2Δgre3Δ cells increased in a manner that was dependent on the dose of DHA added (Figure 3D). Since the overexpression of DAK1 suppressed the susceptibility of glo1Δgre2Δgre3Δ cells to DHA (Figure 2C), we investigated whether DAK1 has the ability to suppress DHA-induced actin depolarization. As shown in Figure 3E, the overexpression of DAK1 moderately suppressed the depolarization of actin patches. We verified that overexpression of DAK1 in wild-type cells did not affect the polarity of actin patches (Supplementary Figure S2A). A previous study reported that formaldehyde was the cause of DHA toxicity [9]; therefore, we examined whether formaldehyde has the ability to depolarize actin patches. Before experiments, we determined the conditions for formaldehyde treatment (concentration and exposure period) that did not decrease cell viability. As shown in Figure 3F, treatment of glo1Δgre2Δgre3Δ cells with 0.1 mM formaldehyde for 120 min reduced the growth rate ∼30%. Thirty percent slow-down of the growth was almost the same extent as that being caused by 200 mM DHA in this mutant (Figure 3A). Under these conditions, formaldehyde did not depolarize actin patches in glo1Δgre2Δgre3Δ cells. These results suggest that the depolarization of actin patches following the treatment with DHA was due to MG formed from DHA.

Protein kinase C (Pkc1) is involved in the actin organization [3436], and we previously reported that a constitutively active mutant of Pkc1 (Pkc1R398P) suppressed the MG-induced depolarization of actin patches [17]. Therefore, we examined whether Pkc1R398P also has the ability to suppress the DHA-induced depolarization of actin patches in glo1Δgre2Δgre3Δ cells. As expected, DHA-induced actin depolarization was suppressed by the expression of PKC1R398P (Figure 3G). These results suggest that DHA causes actin depolarization, and the underlying mechanism appears to be the same as that of MG.

Effects of DHA on vacuolar morphology

The vacuole is a dynamic organelle with a morphology that is influenced by extracellular and intracellular conditions [37]. We recently demonstrated that MG affected vacuolar morphology [18]. Logarithmically growing yeast cells predominantly have two to four fragmented vacuoles; however, vacuolar morphology changed from a fragmented to a large single form following the treatment with MG, suggesting that MG induced the fusion of vacuoles [18]. We then investigated whether DHA changes vacuolar morphology. As shown in Figure 4A, DHA also enhanced vacuolar fusion and caused a morphological change in vacuoles to a single swollen form in glo1Δgre2Δgre3Δ cells. This phenomenon was not observed following the treatment with formaldehyde (Figure 4A). Additionally, the DHA-induced defragmentation of vacuoles was suppressed by the overexpression of DAK1 in glo1Δgre2Δgre3Δ cells (Figure 4B). We verified that overexpression of DAK1 in wild-type cells did not affect the morphology of vacuole (Supplementary Figure S2B). These results suggest that morphological changes in vacuoles are caused by MG derived from DHA.

DHA induces vacuolar defragmentation.

Figure 4.
DHA induces vacuolar defragmentation.

(A) Wild-type (YPH250) and glo1Δgre2Δgre3Δ cells were cultured in SD medium until A610 = 0.3 and treated with 10 mM MG for 90 min, 400 mM DHA for 120 min, or 0.1 mM HCHO for 120 min. The vacuolar membrane was stained with FM4-64. Bar, 5 µm. (B) glo1Δgre2Δgre3Δ cells carrying either an empty plasmid (vector, pRS426) or pRS426-DAK1 were cultured in SD medium until A610 = 0.3 and treated with 400 mM DHA for 120 min. The vacuolar membrane was stained with FM4-64. Bar, 5 µm. (C) Wild-type (YPH250) and glo1Δgre2Δgre3Δ cells carrying ATG18-GFP were cultured in SD medium until A610 = 0.3 and treated with 10 mM MG for 90 min or 400 mM DHA for 120 min. Atg18-GFP and the vacuole (FM4-64) were observed using a fluorescence microscope. Bar, 5 µm.

Figure 4.
DHA induces vacuolar defragmentation.

(A) Wild-type (YPH250) and glo1Δgre2Δgre3Δ cells were cultured in SD medium until A610 = 0.3 and treated with 10 mM MG for 90 min, 400 mM DHA for 120 min, or 0.1 mM HCHO for 120 min. The vacuolar membrane was stained with FM4-64. Bar, 5 µm. (B) glo1Δgre2Δgre3Δ cells carrying either an empty plasmid (vector, pRS426) or pRS426-DAK1 were cultured in SD medium until A610 = 0.3 and treated with 400 mM DHA for 120 min. The vacuolar membrane was stained with FM4-64. Bar, 5 µm. (C) Wild-type (YPH250) and glo1Δgre2Δgre3Δ cells carrying ATG18-GFP were cultured in SD medium until A610 = 0.3 and treated with 10 mM MG for 90 min or 400 mM DHA for 120 min. Atg18-GFP and the vacuole (FM4-64) were observed using a fluorescence microscope. Bar, 5 µm.

A previous study reported that phosphatidylinositol 3,5-bisphosphate (PtdIns(3,5)P2) has an influence on vacuolar morphology [38]. Although the total cellular abundance of PtdIns(3,5)P2 is low, it is predominantly distributed in the vacuolar membrane [38]. We previously reported that MG increased PtdIns(3,5)P2 levels in the vacuolar membrane, which induced the accumulation of Atg18, a PtdIns(3,5)P2-binding protein [39], on the vacuolar membrane [18]. We observed the localization of GFP-tagged Atg18 in glo1Δgre2Δgre3Δ cells. As a result, the localization of Atg18-GFP on the vacuolar membrane was enhanced following the treatment with DHA (Figure 4C). These results suggest that extracellularly added DHA induces the defragmentation of vacuoles, which is accompanied by an increase in PtdIns(3,5)P2 levels, similar to MG.

DHA does not activate the Mpk1 MAP kinase cascade

We recently reported that MG activated TOR (target of rapamycin) complex 2 (TORC2)–Pkc1 signaling [17]. The Mpk1 mitogen-activated protein (MAP) kinase cascade lies downstream of Pkc1 [34,35,40]; Mpk1 is phosphorylated following a treatment with MG [17,18]. Therefore, we investigated whether DHA activates the Mpk1 MAP kinase cascade in glo1Δgre2Δgre3Δ cells. As shown in Figure 5, DHA did not enhance the phosphorylation level of Mpk1 not only in wild-type cells, but also in glo1Δgre2Δgre3Δ cells, suggesting that DHA does not contribute to the activation of the Mpk1 MAP kinase cascade.

Effects of DHA on Mpk1 phosphorylation.

Figure 5.
Effects of DHA on Mpk1 phosphorylation.

Wild-type (YPH250) and glo1Δgre2Δgre3Δ cells were cultured in SD medium until A610 = 0.3, and 400 mM DHA or 10 mM MG was added. The levels of phosphorylation of Mpk1 (p-Mpk1) and the Mpk1 protein (Mpk1) were measured after an incubation for 120 min (DHA) or 60 min (MG) by Western blotting.

Figure 5.
Effects of DHA on Mpk1 phosphorylation.

Wild-type (YPH250) and glo1Δgre2Δgre3Δ cells were cultured in SD medium until A610 = 0.3, and 400 mM DHA or 10 mM MG was added. The levels of phosphorylation of Mpk1 (p-Mpk1) and the Mpk1 protein (Mpk1) were measured after an incubation for 120 min (DHA) or 60 min (MG) by Western blotting.

Effect of DHA and MG on nuclear morphology

Besides vacuolar morphology, we noted that nuclear morphology observed using nuclear membrane-located Nup116-GFP [20,41] also appeared to be changed by the treatment with DHA in glo1Δgre2Δgre3Δ cells (Figure 6A). The nuclear morphology had changed from a globate shape to one with a central depression aligned with the mother–bud axis, which we referred to as a ‘jellybean-like shape’ nucleus, following the treatment with DHA (Figure 6A). Additionally, the DHA-induced morphological change in the nucleus was suppressed by the overexpression of DAK1 in glo1Δgre2Δgre3Δ cells (Figure 6B), suggesting that the change in nuclear morphology is caused by MG derived from DHA. To verify this, we treated wild-type cells with MG and observed the morphology of the nucleus. As shown in Figure 6C, MG also caused a morphological change in nucleus to jellybean-like shape in wild-type cells.

Effects of DHA on nuclear morphology.

Figure 6.
Effects of DHA on nuclear morphology.

(A) glo1Δgre2Δgre3Δ cells carrying NUP116-GFP were cultured in SD medium until A610 = 0.3 and treated with 400 mM DHA for 120 min. Nup116-GFP and the vacuole (FM4-64) were observed using a fluorescence microscope. Bar, 5 µm. (B) glo1Δgre2Δgre3Δ cells carrying NUP116-GFP and either an empty plasmid (vector, pRS426) or pRS426-DAK1 were cultured in SD medium until A610 = 0.3 and treated with 400 mM DHA for 120 min. The morphologies of the nucleus and vacuole were determined as described in (A). Bar, 5 µm. (C) Wild-type (YPH250) cells carrying NUP116-GFP were cultured in SD medium until A610 = 0.3 and treated with 10 mM MG for 90 min. The morphologies of the nucleus and vacuole were determined as described in (A). Bar, 5 µm. (D) Wild-type (YPH250) cells carrying NUP116-GFP were cultured in SD medium at 25°C until A610 = 0.3 and were then shifted to 39°C. After a 30-min exposure to heat shock stress (HS), the morphologies of the nucleus and vacuole were determined as described in (A). Bar, 5 µm. (E) Wild-type (YPH250) cells carrying NUP116-GFP were cultured in SD medium containing 1 M sorbitol until A610 = 0.3, harvested by centrifugation, and suspended in SD medium without 1 M sorbitol. After a 15-min exposure to hypotonic stress (Hypo), the morphologies of the nucleus and vacuole were determined as described in (A). Bar, 5 µm. (F) Wild-type (YPH250), vps41Δ, and vam3Δ cells carrying NUP116-GFP were cultured in SD medium until A610 = 0.3 and treated with 10 mM MG for 90 min. The morphologies of the nucleus and vacuole were determined as described in (A). Bar, 5 µm.

Figure 6.
Effects of DHA on nuclear morphology.

(A) glo1Δgre2Δgre3Δ cells carrying NUP116-GFP were cultured in SD medium until A610 = 0.3 and treated with 400 mM DHA for 120 min. Nup116-GFP and the vacuole (FM4-64) were observed using a fluorescence microscope. Bar, 5 µm. (B) glo1Δgre2Δgre3Δ cells carrying NUP116-GFP and either an empty plasmid (vector, pRS426) or pRS426-DAK1 were cultured in SD medium until A610 = 0.3 and treated with 400 mM DHA for 120 min. The morphologies of the nucleus and vacuole were determined as described in (A). Bar, 5 µm. (C) Wild-type (YPH250) cells carrying NUP116-GFP were cultured in SD medium until A610 = 0.3 and treated with 10 mM MG for 90 min. The morphologies of the nucleus and vacuole were determined as described in (A). Bar, 5 µm. (D) Wild-type (YPH250) cells carrying NUP116-GFP were cultured in SD medium at 25°C until A610 = 0.3 and were then shifted to 39°C. After a 30-min exposure to heat shock stress (HS), the morphologies of the nucleus and vacuole were determined as described in (A). Bar, 5 µm. (E) Wild-type (YPH250) cells carrying NUP116-GFP were cultured in SD medium containing 1 M sorbitol until A610 = 0.3, harvested by centrifugation, and suspended in SD medium without 1 M sorbitol. After a 15-min exposure to hypotonic stress (Hypo), the morphologies of the nucleus and vacuole were determined as described in (A). Bar, 5 µm. (F) Wild-type (YPH250), vps41Δ, and vam3Δ cells carrying NUP116-GFP were cultured in SD medium until A610 = 0.3 and treated with 10 mM MG for 90 min. The morphologies of the nucleus and vacuole were determined as described in (A). Bar, 5 µm.

This morphological change in the nucleus appeared to have been caused by a single swollen vacuole shoving the nucleus towards the bud; however, the change in nuclear morphology did not occurred by heat shock or hypoosmotic stress causing the vacuolar swelling (Figure 6D,E). To determine whether the change in vacuolar morphology was involved in the MG-induced morphological change in the nucleus, nuclear shapes were observed in vps41Δ and vam3Δ mutants, in which the vacuolar fusion is defective [42]. We previously reported that MG-induced vacuolar swelling did not occur in these mutants [18]. The results obtained showed that the morphology of the nucleus in vps41Δ and vam3Δ cells did not have a jellybean-like shape (Figure 6F). These results suggest that the occurrence of a jellybean-like shape nucleus is caused by MG-induced vacuolar swelling.

Effect of DHA and MG on nuclear division

Since S. cerevisiae undergoes polarized cell growth, organelles are transported from the mother cell into the bud [43]. The nuclear membrane in yeast cells is not degraded during mitosis (closed mitosis), the nucleus is also transported into the bud during the mitotic process, where microtubules and actin cables play important roles. A deficiency in organelle inheritance may affect cell growth, and an impairment in nuclear inheritance (i.e. nuclear division) will cause crucial defects in cell viability. To explore the mechanism by which DHA inhibits cell growth or cell division, we focused on nuclear division because DHA affected the nuclear morphology (Figure 6A). To evaluate the effects of DHA on nuclear division, we monitored the distribution of nuclei in cells with buds that were more than two-thirds the size of its mother cell. As a result, we found that the proportion of cells whose nuclei were not transported into the bud was increased following the treatment with DHA in glo1Δgre2Δgre3Δ cells (Figure 7A). The proportion of cells contained the nucleus in mother cells increased in a manner that was dependent of the dose of DHA added (Figure 7B), and the DHA-induced inhibition of nuclear division was suppressed by the overexpression of DAK1 (Figure 7C).

Effects of DHA on nuclear division.

Figure 7.
Effects of DHA on nuclear division.

(A) glo1Δgre2Δgre3Δ cells were cultured in SD medium until the log phase of growth and treated with 400 mM DHA. After 120 min, cells were stained with Hoechst 33342. Bar, 5 µm. (B) glo1Δgre2Δgre3Δ cells were cultured in SD medium until the log phase of growth and treated with DHA in the prescribed concentrations as indicated in the figure for 120 min. The inhibition rate of nuclear division was determined by evaluating the status of the nucleus (stained with Hoechst 33342) in cells having a large bud (bud diameter approximately two-thirds that of the mother cell). Data are for three independent experiments (average ± standard deviation), and more than 100 cells were counted for each experiment. (C) glo1Δgre2Δgre3Δ cells carrying either an empty plasmid (vector, pRS426) or pRS426-DAK1 were cultured in SD medium until the log phase of growth and treated with 400 mM DHA. After 120 min, the inhibition rate of nuclear division was determined as described in (B). (D) Wild-type cells were cultured in SD medium until the log phase of growth and treated with 10 mM MG. After 90 min, the inhibition rate of nuclear division was determined as described in (B). (E) Wild-type cells were cultured in SD medium until A610 = 0.3–0.5 and treated with 10 mM MG. After 90 min, cells were washed with a 0.85% NaCl solution and suspended in fresh SD medium with or without 10 mM MG. After 180 min, the inhibition rate of nuclear division was determined (left panel). Growth (A610) of cells after release to the fresh medium was also monitored (right panel).

Figure 7.
Effects of DHA on nuclear division.

(A) glo1Δgre2Δgre3Δ cells were cultured in SD medium until the log phase of growth and treated with 400 mM DHA. After 120 min, cells were stained with Hoechst 33342. Bar, 5 µm. (B) glo1Δgre2Δgre3Δ cells were cultured in SD medium until the log phase of growth and treated with DHA in the prescribed concentrations as indicated in the figure for 120 min. The inhibition rate of nuclear division was determined by evaluating the status of the nucleus (stained with Hoechst 33342) in cells having a large bud (bud diameter approximately two-thirds that of the mother cell). Data are for three independent experiments (average ± standard deviation), and more than 100 cells were counted for each experiment. (C) glo1Δgre2Δgre3Δ cells carrying either an empty plasmid (vector, pRS426) or pRS426-DAK1 were cultured in SD medium until the log phase of growth and treated with 400 mM DHA. After 120 min, the inhibition rate of nuclear division was determined as described in (B). (D) Wild-type cells were cultured in SD medium until the log phase of growth and treated with 10 mM MG. After 90 min, the inhibition rate of nuclear division was determined as described in (B). (E) Wild-type cells were cultured in SD medium until A610 = 0.3–0.5 and treated with 10 mM MG. After 90 min, cells were washed with a 0.85% NaCl solution and suspended in fresh SD medium with or without 10 mM MG. After 180 min, the inhibition rate of nuclear division was determined (left panel). Growth (A610) of cells after release to the fresh medium was also monitored (right panel).

Next, we examined whether MG caused the same phenomenon as DHA did in terms of the effect on nuclear division. As a result, undivided nuclei were accumulated in wild-type cells following the treatment with MG (Figure 7D), and the inhibitory effect of MG on nuclear division was reversible, i.e. the nucleus divided, and was transported into the bud when cells were washed to remove MG from the medium in accordance with the resumption of cell growth (Figure 7E). These results suggest that the accumulation of undivided nuclei by the formation of MG from DHA contributes to the DHA toxicity.

Discussion

DHA exerts cytotoxic effects on several organisms including the budding yeast [8,10,11]. Regarding the evaluation of the cytotoxic effects of DHA in S. cerevisiae, a previous study used medium containing DHA as a sole source of carbon to produce energy. In the present study, we demonstrated that extracellularly added DHA to medium containing glucose increased intracellular MG levels (Figure 2B). Previous studies reported that DHA toxicity was involved in the formation of AGEs [9,16]. The production of AGEs is initiated by the formation of Schiff bases between proteinaceous amines, such as lysine and arginine, and carbonyl groups (intracellular Maillard reaction), i.e. an oxygen atom in a carbonyl group withdraws an electron, a carbonyl carbon subsequently becomes positively charged, and the primary amine attacks the carbocation as a nucleophile to form an imine (Schiff base). Since Schiff base formation between glucose and proteinaceous amines frequently occurs in the intracellular Maillard reaction, and DHA is the precursor of gluconeogenesis in mammalian cells [6,7], glucose may form in yeast cells cultured in medium containing DHA as the sole source of carbon. However, the reactivity of MG with amine is >20 000-fold higher than that of glucose [44]. In addition, a previous study reported that N-ε-carboxyethyl-lysine, a major MG-derived AGE [15], abundantly formed in yeast cells cultured with DHA as a carbon source [9]. Therefore, the conversion of DHA to MG appears to contribute to the DHA-induced formation of AGEs, which may be one of the causes of DHA toxicity.

On the other hand, the production of formaldehyde is also predicted to be involved in DHA toxicity [9]. For example, the deletion of SFA1, glutathione-dependent formaldehyde dehydrogenase, resulted in a decreased survival rate at high concentrations of DHA [9]. Formaldehyde is a toxic metabolite that induces the growth arrest of yeast cells at low concentrations [45]; however, formaldehyde induced neither actin depolarization nor vacuolar morphological changes as far as the concentration we examined (Figures 3F and 4C). In the present study, we showed that the sfa1Δ mutant exhibited enhanced susceptibility to MG (Figure 2D) [46]. In addition, MG has been identified as a substrate for Sfa1 [30]. Therefore, a deficiency in the adaptation of the sfa1Δ mutant to high concentrations of DHA may be due to impairments in the capability of MG detoxification. Collectively, the commitment of formaldehyde to the exertion of DHA toxicity appears to be less than that of MG.

More evidence that supports DHA being converted to MG in yeast cells is the suppression of the DHA sensitivity of glo1Δgre2Δgre3Δ cells by the overexpression of DAK1 (Figure 2C). DAK catalyzes the phosphorylation of DHA to DHAP, an intermediate of glycolysis; therefore, the overexpression of DAK1 is expected to remove DHA from the toxic cycle producing MG. Since DHA is non-enzymatically converted to MG, spontaneous conversion may occur in the medium. However, the overexpression of DAK1 did not suppress the susceptibility of the glo1Δgre2Δgre3Δ mutant to MG, which indicates that DHA escaped being converted to MG within yeast cells by DAK. This observation provides us with another important insight into the phenotypes caused by MG, i.e. the target molecules of MG that lead to actin depolarization and vacuolar defragmentation may exist within the cell. In contrast, DHA did not activate the Mpk1 MAP kinase cascade (Figure 5). This result implies that extracellular MG induces the activation of the Mpk1 MAP kinase cascade, and some machinery exists that senses extracellular MG on the plasma membrane. Further studies are needed to elucidate the modes of action of intracellular and extracellular MG.

To date, we have reported that MG is involved in the activation of signal transduction [17,21,47], regulation of transcription factors [22], induction of the endocytosis of hexose transporters [48], depolarization of the actin cytoskeleton [17], and changes in vacuolar morphology [18]. The actin cytoskeleton is crucial for the establishment of cell polarity. S. cerevisiae exhibits polarized growth, and polarized actin is known to play an important role during growth [33]. Actin polarization is perturbed by moderate concentrations of MG, which arrest cell growth [17,21]. Therefore, MG-induced actin depolarization appears to contribute to the inhibition of cell growth by MG, which is one of the modes of toxic action of MG. In the present study, we demonstrated that DHA was converted to MG, and some phenotypes such as growth arrest (Figure 2), actin depolarization (Figure 3), and vacuolar morphological changes (Figure 4) caused by DHA were due to MG. In addition to these phenotypes, we found that MG inhibited the nuclear division (Figure 7). Although further investigations are needed to determine whether a functional relationship exists between MG-induced actin depolarization and inhibition of nuclear division, the inhibitory effect on nuclear division seems one of the mechanisms by which DHA as well as MG arrest cell growth in S. cerevisiae. Since the nuclear division is the most critical event in cell cycle in all kind of cells, an inhibition of nuclear division is thought to directly link to the toxicity of MG.

The morphological change observed in the nucleus (a jellybean-like shape) in the presence of MG may have some influence on nuclear division. A jellybean-like shape of the nucleus seems a consequence of enlargement of the vacuole (Figure 6C); however, it did not appear to be sufficient to change the morphology of the nucleus. For example, vacuoles were shown to be swollen to a single enlarged morphology by hypoosmotic stress or heat shock stress [37,49]; nevertheless, morphological changes in the nucleus did not occur under these conditions (Figure 6D,E), and subsequently, inhibition of nuclear division did not occur either. Therefore, another factor, e.g. changes in the fluidity of nuclear membrane, may be involved in the MG/DHA-induced changes in the morphology of the nucleus. Further studies are needed to elucidate the modes of action of MG that cause the morphological change in nucleus and inhibition of nuclear division.

Abbreviations

     
  • AGE

    advanced glycation end-product

  •  
  • DAKs

    dihydroxyacetone kinases

  •  
  • DHA

    dihydroxyacetone

  •  
  • DHAP

    dihydroxyacetone phosphate

  •  
  • Glo1

    glyoxalase I

  •  
  • Glo2

    glyoxalase II

  •  
  • MAP

    mitogen-activated protein

  •  
  • MG

    methylglyoxal

  •  
  • PBS

    phosphate-buffered saline

  •  
  • Pkc1

    protein kinase C

  •  
  • PtdIns(3,5)P2

    phosphatidylinositol 3,5-bisphosphate

  •  
  • SD

    synthetic dextrose

  •  
  • Tkl

    transketolase

  •  
  • TOR

    target of rapamycin

Author Contribution

W.N. and Y.I. designed experiments, and W.N. and M.A. performed experiments. W.N. and Y.I. analyzed the data and wrote the manuscript.

Funding

This work was partly supported by JSPS KAKENHI Grant Number JP17K15414 (to W.N.), JSPS KAKENHI Grant Number 18H02168 (to Y.I.), and Lotte Shigemitsu Prize, Japan (to W.N.).

Acknowledgments

We thank M. Hashimoto, K. Ikeda, and Dr S. Izawa for their technical support and helpful discussion, and Dr T. Ohdate for plasmid construction. We are grateful to Dr M.N. Hall for providing the plasmids.

Competing Interests

The Authors declare that there are no competing interests associated with the manuscript.

References

References
1
Erni
,
B.
,
Siebold
,
C.
,
Christen
,
S.
,
Srinivas
,
A.
,
Oberholzer
,
A.
and
Baumann
,
U.
(
2006
)
Small substrate, big surprise: fold, function and phylogeny of dihydroxyacetone kinases
.
Cell. Mol. Life Sci.
63
,
890
900
2
Daniel
,
R.
,
Stuertz
,
K.
and
Gottschalk
,
G.
(
1995
)
Biochemical and molecular characterization of the oxidative branch of glycerol utilization by Citrobacter freundii
.
J. Bacteriol.
177
,
4392
4401
3
Schürmann
,
M.
and
Sprenger
,
G.A.
(
2001
)
Fructose-6-phosphate aldolase is a novel class I aldolase from Escherichia coli and is related to a novel group of bacterial transaldolases
.
J. Biol. Chem.
276
,
11055
11061
4
Lüers
,
G.H.
,
Advani
,
R.
,
Wenzel
,
T.
and
Subramani
,
S.
(
1998
)
The Pichia pastoris dihydroxyacetone kinase is a PTS1-containing, but cytosolic, protein that is essential for growth on methanol
.
Yeast
14
,
759
771
5
Sakai
,
Y.
,
Nakagawa
,
T.
,
Shimase
,
M.
and
Kato
,
N.
(
1998
)
Regulation and physiological role of the DAS1 gene, encoding dihydroxyacetone synthase, in the methylotrophic yeast Candida boidinii
.
J. Bacteriol.
180
,
5885
5890
PMID:
[PubMed]
6
Sumida
,
K.D.
,
Crandall
,
S.C.
,
Chadha
,
P.L.
and
Qureshi
,
T.
(
2002
)
Hepatic gluconeogenic capacity from various precursors in young versus old rats
.
Metabolism
51
,
876
880
7
Taguchi
,
T.
,
Murase
,
S.
and
Miwa
,
I.
(
2002
)
Glyceraldehyde metabolism in human erythrocytes in comparison with that of glucose and dihydroxyacetone
.
Cell Biochem. Funct.
20
,
223
226
8
Molin
,
M.
,
Norbeck
,
J.
and
Blomberg
,
A.
(
2003
)
Dihydroxyacetone kinases in Saccharomyces cerevisiae are involved in detoxification of dihydroxyacetone
.
J. Biol. Chem.
278
,
1415
1423
9
Molin
,
M.
and
Blomberg
,
A.
(
2006
)
Dihydroxyacetone detoxification in Saccharomyces cerevisiae involves formaldehyde dissimilation
.
Mol. Microbiol.
60
,
925
938
10
Subedi
,
K.P.
,
Kim
,
I.
,
Kim
,
J.
,
Min
,
B.
and
Park
,
C.
(
2008
)
Role of GldA in dihydroxyacetone and methylglyoxal metabolism of Escherichia coli K12
.
FEMS Microbiol. Lett.
279
,
180
187
11
Uzcátegui
,
N.L.
,
Carmona-Gutiérrez
,
D.
,
Denninger
,
V.
,
Schoenfeld
,
C.
,
Lang
,
F.
,
Figarella
,
K.
et al. 
(
2007
)
Antiproliferative effect of dihydroxyacetone on Trypanosoma brucei bloodstream forms: cell cycle progression, subcellular alterations, and cell death
.
Antimicrob. Agents Chemother.
51
,
3960
3968
12
Kalapos
,
M.P.
(
1999
)
Methylglyoxal in living organisms: chemistry, biochemistry, toxicology and biological implications
.
Toxicol. Lett.
110
,
145
175
13
Inoue
,
Y.
,
Maeta
,
K.
and
Nomura
,
W.
(
2011
)
Glyoxalase system in yeasts: structure, function, and physiology
.
Semin. Cell Dev. Biol.
22
,
278
284
14
Needham
,
J.
and
Lehmann
,
H.
(
1937
)
Intermediary carbohydrate metabolism in embryonic life: glyceraldehyde and glucolysis
.
Biochem. J.
31
,
1913
1925
15
Rabbani
,
N.
and
Thornalley
,
P.J.
(
2015
)
Dicarbonyl stress in cell and tissue dysfunction contributing to ageing and disease
.
Biochem. Biophys. Res. Commun.
458
,
221
226
16
Molin
,
M.
,
Pilon
,
M.
and
Blomberg
,
A.
(
2007
)
Dihydroxyacetone-induced death is accompanied by advanced glycation endproduct formation in selected proteins of Saccharomyces cerevisiae and Caenorhabditis elegans
.
Proteomics
7
,
3764
3774
17
Nomura
,
W.
and
Inoue
,
Y.
(
2015
)
Methylglyoxal activates the target of rapamycin complex 2-protein kinase C signaling pathway in Saccharomyces cerevisiae
.
Mol. Cell. Biol.
35
,
1269
1280
18
Nomura
,
W.
,
Maeta
,
K.
and
Inoue
,
Y.
(
2017
)
Phosphatidylinositol 3,5-bisphosphate is involved in methylglyoxal-induced activation of the Mpk1 mitogesn-activated protein kinase cascade in Saccharomyces cerevisiae
.
J. Biol. Chem.
292
,
15039
15048
19
Izawa
,
S.
,
Takemura
,
R.
and
Inoue
,
Y.
(
2004
)
Gle2p is essential to induce adaptation of the export of bulk poly(A)+ mRNA to heat shock in Saccharomyces cerevisiae
.
J. Biol. Chem.
279
,
35469
35478
20
Longtine
,
M.S.
,
McKenzie
, III,
A.
,
Demarini
,
D.J.
,
Shah
,
N.G.
,
Wach
,
A.
,
Brachat
,
A.
et al. 
(
1998
)
Additional modules for versatile and economical PCR-based gene deletion and modification in Saccharomyces cerevisiae
.
Yeast
14
,
953
961
21
Maeta
,
K.
,
Izawa
,
S.
and
Inoue
,
Y.
(
2005
)
Methylglyoxal, a metabolite derived from glycolysis, functions as a signal initiator of the high osmolarity glycerol-mitogen-activated protein kinase cascade and calcineurin/Crz1-mediated pathway in Saccharomyces cerevisiae
.
J. Biol. Chem.
280
,
253
260
22
Maeta
,
K.
,
Izawa
,
S.
,
Okazaki
,
S.
,
Kuge
,
S.
and
Inoue
,
Y.
(
2004
)
Activity of the Yap1 transcription factor in Saccharomyces cerevisiae is modulated by methylglyoxal, a metabolite derived from glycolysis
.
Mol. Cell. Biol.
24
,
8753
8764
23
Vida
,
T.A.
and
Emr
,
S.D.
(
1995
)
A new vital stain for visualizing vacuolar membrane dynamics and endocytosis in yeast
.
J. Cell Biol.
128
,
779
792
24
Murata
,
K.
,
Fukuda
,
Y.
,
Simosaka
,
M.
,
Watanabe
,
K.
,
Saikusa
,
T.
and
Kimura
,
A.
(
1985
)
Metabolism of 2-oxoaldehyde in yeasts. Purification and characterization of NADPH-dependent methylglyoxal-reducing enzyme from Saccharomyces cerevisiae
.
Eur. J. Biochem.
151
,
631
636
25
Inoue
,
Y.
,
Watanabe
,
K.
,
Shimosaka
,
M.
,
Saikusa
,
T.
,
Fukuda
,
Y.
,
Murata
,
K.
et al. 
(
1985
)
Metabolism of 2-oxoaldehydes in yeasts. Purification and characterization of lactaldehyde dehydrogenase from Saccharomyces cerevisiae
.
Eur. J. Biochem.
153
,
243
247
26
Chen
,
C.-N.
,
Porubleva
,
L.
,
Shearer
,
G.
,
Svrakic
,
M.
,
Holden
,
L.G.
,
Dover
,
J.L.
et al. 
(
2003
)
Associating protein activities with their genes: rapid identification of a gene encoding a methylglyoxal reductase in the yeast Saccharomyces cerevisiae
.
Yeast
20
,
545
554
27
Aguilera
,
J.
and
Prieto
,
J.
(
2001
)
The Saccharomyces cerevisiae aldose reductase is implied in the metabolism of methylglyoxal in response to stress conditions
.
Curr. Genet.
39
,
273
283
28
Inoue
,
Y.
and
Kimura
,
A.
(
1996
)
Identification of the structural gene for glyoxalase I from Saccharomyces cerevisiae
.
J. Biol. Chem.
271
,
25958
25965
29
Bito
,
A.
,
Haider
,
M.
,
Hadler
,
I.
and
Breitenbach
,
M.
(
1997
)
Identification and phenotypic analysis of two glyoxalase II encoding genes from Saccharomyces cerevisiae, GLO2 and GLO4, and intracellular localization of the corresponding proteins
.
J. Biol. Chem
272
,
21509
21519
30
Wehner
,
E.P.
,
Rao
,
E.
and
Brendel
,
M.
(
1993
)
Molecular structure and genetic regulation of SFA, a gene responsible for resistance to formaldehyde in Saccharomyces cerevisiae, and characterization of its protein product
.
Mol. Gen. Genet.
237
,
351
358
PMID:
[PubMed]
31
Dickinson
,
J.R.
,
Salgado
,
L.E.
and
Hewlins
,
M.J.E.
(
2003
)
The catabolism of amino acids to long chain and complex alcohols in Saccharomyces cerevisiae
.
J. Biol. Chem.
278
,
8028
8034
32
Fernández
,
M.R.
,
Biosca
,
J.A.
,
Norin
,
A.
,
Jörnvall
,
H.
and
Parés
,
X.
(
1995
)
Class III alcohol dehydrogenase from Saccharomyces cerevisiae: structural and enzymatic features differ toward the human/mammalian forms in a manner consistent with functional needs in formaldehyde detoxication
.
FEBS Lett.
370
,
23
26
33
Pruyne
,
D.
and
Bretscher
,
A.
(
2000
)
Polarization of cell growth in yeast
.
J. Cell Sci.
113
(
Pt 3
),
571
585
PMID:
[PubMed]
34
Levin
,
D.E.
(
2011
)
Regulation of cell wall biogenesis in Saccharomyces cerevisiae: the cell wall integrity signaling pathway
.
Genetics
189
,
1145
1175
35
Heinisch
,
J.J.
and
Rodicio
,
R.
(
2018
)
Protein kinase C in fungi—more than just cell wall integrity
.
FEMS Microbiol. Rev.
42
,
22
39
36
Nomura
,
W.
and
Inoue
,
Y.
(
2017
)
Contribution of phosphatidylserine to Rho1- and Pkc1-related repolarization of the actin cytoskeleton under stressed conditions in Saccharomyces cerevisiae
.
Small GTPases
113
,
1
7
37
Li
,
S.C.
and
Kane
,
P.M.
(
2009
)
The yeast lysosome-like vacuole: endpoint and crossroads
.
Biochim. Biophys. Acta, Mol. Cell Res.
1793
,
650
663
38
McCartney
,
A.J.
,
Zhang
,
Y.
and
Weisman
,
L.S.
(
2014
)
Phosphatidylinositol 3,5-bisphosphate: low abundance, high significance
.
BioEssays
36
,
52
64
39
Dove
,
S.K.
,
Piper
,
R.C.
,
McEwen
,
R.K.
,
Yu
,
J.W.
,
King
,
M.C.
,
Hughes
,
D.C.
et al. 
(
2004
)
Svp1p defines a family of phosphatidylinositol 3,5-bisphosphate effectors
.
EMBO J.
23
,
1922
1933
40
Kamada
,
Y.
,
Jung
,
U.S.
,
Piotrowski
,
J.
and
Levin
,
D.E.
(
1995
)
The protein kinase C-activated MAP kinase pathway of Saccharomyces cerevisiae mediates a novel aspect of the heat shock response
.
Genes Dev.
9
,
1559
1571
41
Fabre
,
E.
and
Hurt
,
E.
(
1997
)
Yeast genetics to dissect the nuclear pore complex and nucleocytoplasmic trafficking
.
Annu. Rev. Genet.
31
,
277
313
42
Balderhaar
,
H.J.k.
and
Ungermann
,
C.
(
2013
)
CORVET and HOPS tethering complexes—coordinators of endosome and lysosome fusion
.
J. Cell Sci.
126
,
1307
1316
43
Bretscher
,
A.
(
2003
)
Polarized growth and organelle segregation in yeast: the tracks, motors, and receptors
.
J. Cell Biol.
160
,
811
816
44
Thornalley
,
P.J.
(
2005
)
Dicarbonyl intermediates in the Maillard reaction
.
Ann. N. Y. Acad. Sci.
1043
,
111
117
45
Magaña-Schwencke
,
N.
and
Moustacchi
,
E.
(
1980
)
Biochemical analysis of damage induced in yeast by formaldehyde III. Repair of induced cross-links between DNA and proteins in the wild-type and in excision-deficient strains
.
Mutat. Res., Fundam. Mol. Mech. Mutagen.
70
,
29
35
46
Hoon
,
S.
,
Gebbia
,
M.
,
Costanzo
,
M.
,
Davis
,
R.W.
,
Giaever
,
G.
and
Nislow
,
C.
(
2011
)
A global perspective of the genetic basis for carbonyl stress resistance
.
G3
1
,
219
231
47
Nomura
,
W.
,
Ito
,
Y.
and
Inoue
,
Y.
(
2017
)
Role of phosphatidylserine in the activation of Rho1-related Pkc1 signaling in Saccharomyces cerevisiae
.
Cell. Signal.
31
,
146
153
48
Yoshida
,
A.
,
Wei
,
D.
,
Nomura
,
W.
,
Izawa
,
S.
and
Inoue
,
Y.
(
2012
)
Reduction of glucose uptake through inhibition of hexose transporters and enhancement of their endocytosis by methylglyoxal in Saccharomyces cerevisiae
.
J. Biol. Chem.
287
,
701
711
49
Meaden
,
P.G.
,
Arneborg
,
N.
,
Guldfeldt
,
L.U.
,
Siegumfeldt
,
H.
and
Jakobsen
,
M.
(
1999
)
Endocytosis and vacuolar morphology in Saccharomyces cerevisiae are altered in response to ethanol stress or heat shock
.
Yeast
15
,
1211
1222
50
Helliwell
,
S.B.
,
Schmidt
,
A.
,
Ohya
,
Y.
and
Hall
,
M.N.
(
1998
)
The Rho1 effector Pkc1, but not Bni1, mediates signalling from Tor2 to the actin cytoskeleton
.
Curr. Biol.
8
,
1211
1214

Author notes

*

Present address: Laboratory of Molecular Function of Food, Division of Food Science and Biotechnology, Graduate School of Agriculture, Kyoto University, Uji, Kyoto 611-0011, Japan.

Supplementary data