Biochemical Journal

Research article

Acidic stress induces the formation of P-bodies, but not stress granules, with mild attenuation of bulk translation in Saccharomyces cerevisiae

Aya Iwaki , Shingo Izawa

Abstract

The stress response of eukaryotic cells often causes an attenuation of bulk translation activity and the accumulation of non-translating mRNAs into cytoplasmic mRNP (messenger ribonucleoprotein) granules termed cytoplasmic P-bodies (processing bodies) and SGs (stress granules). We examined effects of acidic stress on the formation of mRNP granules compared with other forms of stress such as glucose deprivation and a high Ca2+ level in Saccharomyces cerevisiae. Treatment with lactic acid clearly caused the formation of P-bodies, but not SGs, and also caused an attenuation of translation initiation, albeit to a lesser extent than glucose depletion. P-body formation was also induced by hydrochloric acid and sulfuric acid. However, lactic acid in SD (synthetic dextrose) medium with a pH greater than 3.0, propionic acid and acetic acid did not induce P-body formation. The results of the present study suggest that the assembly of yeast P-bodies can be induced by external conditions with a low pH and the threshold was around pH 2.5. The P-body formation upon acidic stress required Scd6 (suppressor of clathrin deficiency 6), a component of P-bodies, indicating that P-bodies induced by acidic stress have rules of assembly different from those induced by glucose deprivation or high Ca2+ levels.

  • acidic stress
  • lactic acid
  • processing body (P-body)
  • Saccharomyces cerevisiae
  • stress granule
  • suppressor of clathrin deficiency 6 (Scd6)

INTRODUCTION

P-bodies (processing bodies) and SGs (stress granules) in eukaryotes are cytoplasmic mRNP (messenger ribonucleoprotein) granules consisting of non-translating mRNAs and various RNA-associated proteins. They play a role in the regulation of translation and turnover of mRNA in response to different forms of stress [13]. P-bodies and SGs can interact with each other by dynamically overlapping and sharing specific mRNAs and proteins [46]. SGs typically consist of a subset of translation initiation factors, 40S ribosomal subunits and non-translating mRNAs [2,3,7]. On the other hand, P-bodies generally contain the mRNA decay machinery including exonuclease and decapping enzymes [8,9]. Therefore P-bodies were initially identified as sites where mRNAs can be degraded [10,11]. Previous studies, however, indicate that P-bodies, as well as SGs, serve as sites of non-translating mRNA storage under several stressed conditions and are responsible for translational repression [2,12,13].

Various forms of stress can block translational initiation and consequently cause the elevation of non-translating mRNA levels. Non-translating mRNAs and associated proteins can be recruited from polysomes and segregated into P-bodies or SGs under stressed conditions [13,8]. It has been reported that the mRNAs in SGs or P-bodies are translationally inactive, but can exit these mRNP granules and return to translation once the stress is eliminated [1416]. Recently, however, Arribere et al. [17] reported that such a translational resurrection is restricted to a subset of mRNAs in P-bodies. Whatever the case, these findings strongly suggest that P-bodies and SGs contribute to the regulation of efficient and selective translation under stressed conditions. Additionally, both granules might be critical for a rapid recovery from stress, since the segregated mRNAs in these granules can rapidly re-enter the translation initiation process when stress is eliminated [13,14,18].

Lactic acid is one of the most important weak acids associated with the fermentation industry. Lactic acid functions as a flavour enhancer in traditional sourdough breadmaking, Japanese sake brewing and malolactic fermentation of wine [1922]. Lactic acid has been also used as an antimicrobial preservative in the food industry because of its bacteriocidal and fungicidal powers [23,24]. Additionally, lactic acid is used as a chemical raw material in various industrial applications, and lactic acid production has been achieved using metabolically engineered Saccharomyces cerevisiae [25, 26]. However, lactic acid is a stressor for yeast cells and inhibits the growth and ethanol productivity of S. cerevisiae [23,27,28]. It has been reported that the minimum inhibitory concentration of lactic acid for yeast growth is 2.5% (w/v) (278 mM) at 30°C [23].

The adaptive response to lactic acid in S. cerevisiae has been widely investigated. It has been clarified that lactic acid activates the transcription of genes controlled by the transcriptional activators Haa1, Aft1 and Msn2/Msn4, and currently there is information available on the yeast transcriptional response to lactic acid [24,29,30]. However, the effects of lactic acid on the formation of cytoplasmic mRNP granules and translational initiation remain unclear. To gain a greater understanding of post-transcriptional events in the response to lactic acid in yeast cells, we investigated whether lactic acid causes the formation of P-bodies and/or SGs. We also investigated the effects of other acids such as propionic acid and acetic acid on the formation of cytoplasmic mRNP granules and bulk translation activity. We found that lactic acid and acidic stress with low pH (<2.5) conditions cause the assembly of P-bodies, but not SGs, and weak repression of translation in S. cerevisiae.

EXPERIMENTAL

Strains and medium

W303-1A (MATa his3-11, 15 leu2-3, 112 trp1-1 ade2-1 ura3-1 can1-100) was used as a wild-type strain. BY4741 (MATa his3Δ1 ura3Δ0 leu2Δ0 met15Δ0) and its isogenic knockout mutants (pat1Δ and scd6Δ) were purchased from Open Biosystems. Cells were cultured in 50 ml of SD (synthetic dextrose) medium (2% glucose and 0.67% yeast nitrogen base without amino acids) with appropriate supplements of amino acids and bases, at 28°C with reciprocal shaking (120 rev./min) in Erlenmeyer flasks (300 ml). Exponentially growing cells were harvested at D600=0.5. The pH of the medium was measured using a meter HORIBA D-21 (Horiba). All of the acids including lactic acid were obtained from Nakarai Tesque.

Plasmids

All of the GFP (green fluorescent protein)- or mRFP (monomeric red fluorescent protein)-tagged proteins were expressed using integrative plasmids except for pRP1662, and the transcription of each gene was under the control of its own promoter.

A 1.5-kb fragment encoding the promoter region and ORF (open reading frame) of DCP (mRNA decapping) 1 was amplified using 5′-AAAAAGCAGCTTGGGAATTTCTAGAGGAGA-3′ and 5′-CCTCTCTCGAGAAGCAAAAGAATCTTTTGGCTCATT-3′ as primers. The amplicon was digested with XbaI/XhoI and cloned into the XbaI/XhoI sites of pAUR-ZRC1-GFP [31] to construct pAUR-DCP1-GFP. The plasmid was linearized by StuI and introduced into yeast cells.

A 0.4-kb fragment encoding part of the ORF of LSM1 (like Sm 1) was amplified using the primers 5′-TTATATCTAGACCAGTATAACTTCACTACC-3′ and 5′-TTTTCCTCGAGAGTACATGTCAGATTTATG-3′. The amplicon was digested with XbaI/XhoI and cloned into the XbaI/XhoI sites of pPS1630 [32] to construct YIp-LSM1-GFP. The plasmid was linearized by ClaI and introduced into yeast cells.

A 2.1-kb fragment encoding part of the ORF of PAT1 (protein associated with topoisomerase 1) was amplified using the primers 5′-GGACCTCTAGACCTGAAGCCAATGGAATCT-3′ and 5′-GGAAACTCGAGACTTTAGTTCTGATATTTC-3′. The amplicon was digested with XbaI/XhoI and cloned into the XbaI/XhoI sites of pPS1630 to construct YIp-PAT1-GFP. The plasmid was linearized by partial digestion with HindIII and introduced into yeast cells.

A 0.8-kb fragment encoding part of the ORF of EDC3 (enhancer of mRNA decapping 3) was amplified using 5′-TCAAAGAGCTCTAAGCAGAACATTCCTATG-3′ and 5′-ATTCACTCGAGACAAATCTAATAGCAGGGA-3′. The amplicon was digested with SacI/XhoI and cloned into the SacI/XhoI sites of pPS1630 to construct YIp-EDC3-GFP. The plasmid was linearized by ClaI and introduced into yeast cells.

A 1.4-kb fragment encoding part of the ORF of XRN1 (exoribonuclease 1) was amplified using 5′-AAGAGTCTAGACAAACTTTCCCAGATTTTT-3′ and 5′-TCGTACTCGAGAAGTAGATTCGTCTTTTTT-3′. The amplicon was digested with XbaI/XhoI and cloned into the XbaI/XhoI sites of pPS1630 to construct YIp-XRN1-GFP. The plasmid was linearized by ClaI and introduced into yeast cells.

A 2.3-kb fragment encoding SCD6 (suppressor of clathrin deficiency 6) was amplified using the primers 5′-CAAAAGAATTCAACAAAATTCTGGACAAAT-3′ and 5′-CCGCCGAATTCCGCTAGCTTCGAAGTTATG-3′. The amplicon was digested with EcoRI and cloned into the EcoRI site of pRS313 [33] to construct pRS-SCD6.

Construction of YIp-NGR1-GFP, YIp-DCP2-GFP, YIp-DHH1-GFP, YIp-PBP1-GFP, YIp-RPG1-GFP, YIp-PRT1-GFP, YIp-NIP1-GFP, YIp-RPS30A-GFP, and YIp-DCP2-mRFP was as described previously [18,34]. pRP1662 [Pub1 (polyuridylate binding)–mCherry) was donated by Dr Roy Parker (University of Arizona, Tucson, AZ, U.S.A.) [6].

Microscopy

A Leica AF6500 fluorescence microscope system was used for the microscopic analysis. The cells treated with various types of stress were immediately observed without fixation. In order to count the number of P-bodies (Dcp2–GFP) in the entire cell cytoplasm, experimental images were captured as Z-stacks of six images.

Polysome profile analysis

The preparation of yeast extract and sucrose gradient separation for the polysome analysis was carried out using a gradient master and fractionator (107-201M and 152-002, BioComp Instruments) by the methods of Inada and Aiba [35].

Stress tolerance assay

Cells were cultivated in SD medium until D600=0.5 and were harvested and resuspended in fresh SD medium to obtain an initial D600 value of 0.1. To examine their susceptibility to acid stress, the cells were treated with various acids at 28°C for 60 min. They were then spotted (10 μl) on to plates of SD medium in 10-fold serial dilutions and incubated at 28°C for 2 days.

RESULTS

Lactic acid causes the formation of P-bodies, but not SGs

First, we investigated the effects of lactic acid (0.5–2.0% v/v) on the formation of P-bodies and SGs in yeast cells at a mid-exponential growth phase in SD medium. Dcp2–mRFP and Ngr1 (negative growth regulatory 1)–GFP were used as a P-body marker and a SG marker respectively [3,13]. As well as glucose depletion, lactic acid enhanced the formation of clear foci of Dcp2–mRFP in the cytoplasm (Figure 1A). As far as we examined, the minimum concentration of lactic acid able to generate the foci of Dcp2–mRFP was around 0.75% (v/v) (88.5 mM). The number of cytoplasmic granules of Dcp2 increased with the concentration of lactic acid from 0.75 to 1.5% (Figures 1A and 1C). These results strongly suggest that lactic acid (>0.75% v/v) can induce the formation of P-bodies.

Figure 1 Lactic acid stress causes the assembly of P-bodies, but not SGs

(A) S. cerevisiae cells in an exponential phase of growth in SD medium at 28°C were treated with lactic acid (0.5–2.0% v/v) for 30 min or glucose deprivation stress (−Glc) for 15 min. The assembly of P-bodies and SGs was examined with Dcp2–mRFP and Ngr1–GFP respectively. The pH in each condition is indicated. Scale bar, 5 μm. BF, bright field; w/o, without. (B) Time course of P-body/SG formation caused by lactic acid stress. Cells were treated with 1.5% (v/v) lactic acid for 0–120 min. The assembly of P-bodies was examined with Dcp2–GFP and Dhh1–GFP as P-body markers. The assembly of SGs was examined with Ngr1–GFP. (C) Quantification of P-bodies in cells treated with lactic acid or sulfuric acid. The number of P-bodies (Dcp2–GFP) per cell was counted using a minimum of 150 cells from at least two independent experiments for each condition. Results are means±S.E.M. N.S., no statistically significant difference. (D) Cells were treated with CHX (100 μg/ml) for 1 min prior to treatment with lactic acid (1.0 or 1.5% v/v) for 30 min or glucose deprivation (−Glc) for 15 min. The assembly of P-bodies was examined with Dcp2–GFP. (E) The assembly of P-bodies was confirmed using other components of P-bodies. Cells were treated with lactic acid (1.5% v/v) for 30 min or glucose deprivation (−Glc) for 15 min.

It took over 25 min for the assembly of P-bodies upon lactic acid stress (Figure 1B), whereas glucose depletion caused P-body formation within 15 min (Figure 1A). It has been reported that high Ca2+ levels and hyperosmotic stress also cause P-body formation within 15 min [36]. Additionally, ethanol stress (10% v/v) caused P-body formation within 10 min [18]. These results indicated that lactic acid elicits P-body formation more slowly than other stressors already reported.

It is well established that CHX (cycloheximide) can trap mRNAs in polysomes and prevent the assembly of P-bodies and SGs in glucose-deprived yeast cells [6,13]. We also examined the effects of CHX on the formation of P-bodies caused by lactic acid. Cells were treated with CHX (100 μg/ml) for 1 min prior to the treatment with lactic acid, then the assembly of P-bodies was investigated. Consistent with previous studies, pre-treatment with CHX prevented the formation of the foci of Dcp2–GFP upon glucose deprivation (Figure 1D) [13]. Likewise, CHX prevented the formation of the foci of Dcp2–GFP caused by lactic acid (Figure 1D), indicating that lactic acid causes P-body formation via the release of non-translating mRNAs from polysomes.

We verified that lactic acid caused the formation of P-bodies using other markers. Dcp1, Dhh1 (DEAD box helicase homologue 1), Edc3, Lsm1, Pat1 and Xrn1 are core components of P-bodies in S. cerevisiae [3]. As shown in Figure 1(E), these markers also revealed clear cytoplasmic foci in the presence of lactic acid. Therefore we concluded that lactic acid stress induces the formation of P-bodies.

Conversely, Ngr1–GFP did not accumulate and diffused in the cytoplasm in the presence of lactic acid (Figure 1A). Even prolonged treatment (for 120 min) with 1.5% lactic acid did not produce foci of Ngr1–GFP (Figure 1B). Cells treated with a higher concentration of lactic acid (>2.0%) were unanalysable for cellular localization of Ngr1–GFP because the intracellular structure was disrupted (Figure 1A). Additionally, treatment with 2.0% lactic acid caused a decrease in cell survival (Figure 2). We investigated the localization of other SG components such as Pbp1 (Pab1p-binding protein 1)–GFP, Pub1–mCherry, Rpg1–GFP, Prt1 (protein synthesis 1)–GFP, Nip1 (nuclear import 1)–GFP and Rps30A (ribosomal protein of the small subunit)–GFP [6,18]. We verified that all of these SG components also diffused in the cytoplasm of cells treated with lactic acid (Supplementary Figure S1 at http://www.BiochemJ.org/bj/446/bj4460225add.htm). These results indicated that the assembly of SGs was not induced by lactic acid.

Figure 2 Cellular susceptibility to various types of weak carboxylic acid stress

Cells cultured in SD medium until D600=0.5 were treated with weak carboxylic acids at 28°C for 60 min, then spotted (10 μl) on to plates of SD medium in 10-fold serial dilutions and incubated at 28°C for 2 days. w/o, without.

Lactic acid stress leads to a gradual attenuation of translation

The formation of P-bodies and SGs is often correlated with a reduction in translation initiation [8,11,37]. Additionally, a model regarding the dynamic equilibrium between translation attenuation and P-body formation in the response to glucose deprivation was proposed [8,12]. Since we found that lactic acid stress can induce P-body formation, we examined how lactic acid stress affects bulk translation activity by polysome profile analysis. As reported by Ashe et al. [38], glucose depletion caused a pronounced reduction in the polysome fraction (Figure 3A), indicating the reduction of translation initiation by glucose depletion. Treatment with 1.0% or 1.25% lactic acid for 30 min slightly reduced translation, but clearly induced P-body formation (Figures 1A and 3A). We also observed that treatment of cells with 1.5% lactic acid for 30–120 min led to a gradual reduction in the polysome fraction and a concomitant increase in the monosome fraction (80S), indicating that 1.5% lactic acid caused a bulk attenuation in translation initiation (Figure 3A). However, the repressive effect of lactic acid on the bulk translation was apparently less and slower than that of glucose depletion. Intriguingly, longer treatment with 1.5% lactic acid had little effect on the size and number of P-bodies (Figure 1B). Additionally, the Z-test indicated no statistically significant difference in the number of P-bodies between cells treated with 1.5% lactic acid for 30 min and cells treated with 1.5% lactic acid for 120 min (Figure 1C). This result indicates that the assembly of P-bodies caused by lactic acid does not correlate tightly with the strength of translation attenuation.

Figure 3 Lactic acid stress gradually causes translation inhibition

(A) Polysome profiles of cells treated with lactic acid were determined. Cells in an exponential phase were treated with 1.0–1.5% lactic acid or glucose depletion (−Glc). The polysome, 40S (small ribosomal subunit), 60S (large ribosomal subunit) and 80S (monosome) peaks are labelled. w/o, without. (B and C) After treatment with 1.5% lactic acid for 120 min, cells were collected and transferred into fresh SD medium without lactic acid and incubated for 0–30 min at 28°C. (B) Disassembly of P-bodies was examined using Dcp2–GFP and Dhh1–GFP as P-body markers. (C) Polysome profiles were determined from cells after shifting to fresh medium without lactic acid from 1.5% lactic acid stress.

Elimination of lactic acid causes quick disassembly of P-bodies and gradual restoration of polysomes

We further examined whether elimination of lactic acid causes the disassembly of P-bodies. After induction of P-body formation by the treatment with 1.5% lactic acid for 120 min, cells were collected and transferred into fresh SD medium without lactic acid and incubated at 28°C. Elimination of lactic acid quickly caused the disassembly of components of P-bodies in 5–10 min (Figure 3B), indicating that these components assemble and disassemble in response to lactic acid stress in a reversible manner. Intriguingly, the disassembly of P-bodies preceded the complete restoration of polysomes upon the elimination of lactic acid. P-bodies were completely disassembled within 10 min after the elimination of lactic acid stress (Figure 3B), whereas the polysome fraction was not completely restored and the 80S fraction was large at this point (Figure 3C). Although the polysome fraction was mostly restored at the 30 min time point after elimination of lactic acid, the 80S fraction was still slightly larger than that under non-stressed conditions even at this point (Figure 3C). The restoration of polysomes after the elimination of lactic acid stress was relatively gradual, and took 30 min or more (Figure 3C).

Effects of other weak acids on P-body formation and bulk translation

We also investigated whether other weak carboxylic acids, propionic acid and acetic acid, cause the formation of P-bodies (Figure 4A). Treatment with propionic acid or acetic acid (0.1–0.5% for 30 min) did not lead to the formation of P-bodies and SGs (Figure 4A). Treatment with over 0.5% of acetic acid caused the disruption of the intracellular structure and it was impossible to see whether Dcp2–mRFP and Ngr1–GFP make foci or not (results not shown). Additionally, treatment with acetic acid or propionic acid at over 0.5% caused a decrease in cell survival (Figure 2). Neither propionic acid nor acetic acid induced pronounced repression of translation (Figure 4B). On the other hand, treatment with hydrochloric acid or sulfuric acid induced P-body formation (Figure 4A). The polysome fraction was reduced, but did not completely disappear in cells treated with hydrochloric acid, and the 80S monosome fraction was increased by the treatment with hydrochloric acid or sulfuric acid (Figure 4B). Intriguingly, sulfuric acid caused little reduction of the polysome fraction, but distinctly induced P-body formation (Figures 4A and 4B). The number of P-bodies in cells treated with sulfuric acid was greater than that in cells treated with lactic acid (Figure 1C).

Figure 4 Effects of various acids on the formation of P-bodies and SGs

Cells in an exponential phase of growth were treated with propionic acid, acetic acid or hydrochloric acid for 30 min, or with sulfuric acid for 10 min. (A) The assembly of P-bodies and SGs was examined with Dcp2–mRFP and Ngr1–GFP respectively. The pH in each condition is indicated. BF, bright field. (B) Polysome profiles of cells treated with various acids were determined. The concentration of each acid is indicated (% v/v). w/o, without. (C and D) Cells were incubated in SD medium containing 1.5% lactic acid at pH 3.0–4.0 for 30 min. The pH of each medium was adjusted with KOH. (C) Assembly of P-bodies was examined using Dcp2–GFP. BF, bright field. (D) Polysome profiles of cells were determined.

Since pH values of SD media containing lactic acid, hydrochloric acid or sulfuric acid were significantly lower than those with propionic acid and acetic acid (Figures 1A and 4A), we further examined the effects of lactic acid on P-body formation under conditions with higher pH. The pH levels of the SD medium containing lactic acid were adjusted to 3.0–4.0 with KOH. As shown in Figure 4(C), lactic acid (1.5% v/v) did not cause the assembly of P-bodies and translational attenuation in the SD medium with a pH greater than 3.0, indicating that a reduction in the pH of the surrounding environment by lactic acid is necessary to induce P-body formation in yeast cells.

Deletion of Scd6 causes a defect of P-body formation upon lactic acid stress

A previous study suggested that the importance of P-body assembly factors can vary according to the type of stress [36]. The Lsm-associated protein Pat1 and the Lsm homologue Scd6 are components of P-bodies, and are required for P-body formation at high Ca2+ levels, i.e. P-body assembly induced at high Ca2+ levels is strongly blocked in the null mutants scd6Δ and pat1Δ [36]. On the other hand, these null mutants show slightly reduced P-body assembly, but can still form P-bodies upon glucose deprivation and 0.5 M NaCl stress [36,39,40]. As shown in Figure 5(A), we verified that the pat1Δ and scd6Δ mutants could hardly produce P-bodies at 200 mM CaCl2. We examined whether these components are required for P-body formation caused by lactic acid. The pat1Δ mutant still formed P-bodies upon lactic acid stress as well as glucose deprivation stress (Figure 5A), indicating that P-bodies induced by lactic acid have distinct rules of assembly from those induced by high Ca2+ levels. On the other hand, the scd6Δ mutant showed a significant reduction in the formation of P-bodies upon lactic acid stress (Figure 5A), similar to prior observation at high Ca2+ levels [36]. P-body formation was not induced by hydrochloric acid or sulfuric acid in the scd6Δ mutant, either (Figure 5B). Defects in the P-body formation upon acidic stress of the scd6Δ mutant were rescued when this strain was transformed with a plasmid bearing the SCD6 gene (pRS-SCD6) (Figure 5B). These findings indicate that P-bodies induced by acidic stress have rules of assembly different from those induced by glucose deprivation or high Ca2+ levels.

Figure 5 Scd6, but not Pat1, is required for efficient P-body formation caused by lactic acid

(A) The wild-type (WT) (BY4741) and isogenic knockout mutants (scd6Δ and pat1Δ) were treated with 1.5% lactic acid for 30 min, 200 mM CaCl2 (Ca2+) for 15 min or glucose depletion (−Glc) for 15 min. The assembly of P-bodies was examined with Dcp2–GFP and Dhh1–GFP. (B) The scd6Δ mutant carrying pRS-SCD6 or pRS313 (empty vector) and wild-type cells were treated with 200 mM CaCl2 (Ca2+) for 15 min, 1.5% lactic acid for 30 min, 0.1% sulfuric acid for 10 min or 0.4% hydrochloric acid for 30 min. The assembly of P-bodies was examined with Dcp2–GFP. w/o, without.

DISCUSSION

In the present study we have demonstrated that lactic acid causes P-body formation, but not SG formation, in S. cerevisiae (Figure 1A). Although lactic acid also caused a translational attenuation (Figure 3A), the effects of lactic acid stress on the repression/restoration of translation were slower than those of glucose deprivation and NaN3 treatment, both of which induce SG formation [6,15,17,41]. Lactic acid induced a gradual repression of translation initiation like hyperosmotic shock (0.5 M NaCl) and a high Ca2+ level (200 mM CaCl2) [38]. In spite of the partial repression of translation, lactic acid stress as well as hyperosmotic shock and a high Ca2+ level were capable of inducing P-body formation (Figures 1 and 3A) [36]. These findings clearly indicate that severe repression of bulk translation is not necessarily required for P-body assembly.

It has been reported that hyperosmotic shock and high Ca2+ levels have little or no effect on the formation of yeast SGs [2,6,36], whereas glucose deprivation and NaN3 treatment cause SG formation and pronounced repression of translation [6,41]. Robust heat shock also causes SG formation with almost a complete disappearance of the polysome fractions [42]. Additionally, Buchan et al. [6,41] have reported that yeast SGs are assembled following P-body formation during glucose deprivation or NaN3 treatment. We have also reported that the formation of yeast SGs upon ethanol stress requires a longer period of time and more severe stress than that of P-bodies [18]. In the case of lactic acid stress, prolonged treatment (120 min) with 1.5% lactic acid caused neither the complete disappearance of the polysome fractions nor SG formation (Figures 1A, 1B, 3A and Supplementary Figure S1). These results suggest severe repression of bulk translation to be essential for SG formation. It may be feasible that cells show distinct responses to stress that causes severe repression of translation and stress that causes mild attenuation of translation. Cells might have the ability to cope without SGs in the response to stress that causes mild translational attenuation.

Kilchert et al. [36] demonstrated an uncoupling of the strength of translational repression and P-body formation in the response to the combined stress of high Ca2+ levels and glucose deprivation. We also demonstrated in the present study that P-body formation caused by lactic acid was not necessarily coupled with the strength of translational repression. As time progressed during the treatment with lactic acid, the number and size of P-bodies increased little, although the repression of translation was strengthened (Figures 1B, 1C and 3A). Furthermore, disassembly of P-bodies preceded the complete restoration of polysomes upon the elimination of lactic acid (Figures 3B and 3C). Additionally, sulfuric acid rapidly induced the formation of multiple P-bodies without a distinct reduction in the polysome fraction (Figures 1C, 4A and 4B). These results clearly indicate that the number and size of P-bodies do not correlate tightly with the strength of the translational attenuation upon acidic stress. Kilchert et al. [36] reported that mRNA release from polysomes via translational attenuation might not be the driving force for P-body formation in secretory mutants. The results of the present study may also imply that mRNA release from polysomes is not the only driving force for P-body formation in the response to acidic stress.

In contrast to lactic acid, acetic acid and propionic acid had little effect on the translation and P-body formation in yeast cells (Figure 4). S. cerevisiae is relatively resistant to these weak carboxylic acids [43,44]. The toxicity of weak carboxylic acids is dependent on pH and causes changes in the intracellular pH. Weak acids in the undissociated form can easily diffuse into cells across the plasma membrane and dissociate due to higher intracellular pH. Conversely, the plasma membrane is impermeable to weak acids in the dissociated form [23,2728,44]. The pH levels of the SD media were reduced to 2.66, 3.10 and 3.25 by addition of lactic acid, acetic acid and propionic acid (0.5% v/v each) respectively (Figures 1A and 4A). These results indicate that lactic acid has a higher capacity to acidulate SD medium than acetic acid and propionic acid. Indeed, the acid dissociation constant (pKa), a quantitative measure of acid strength in solution, of lactic acid is 3.86; lower than those of acetic acid (4.76) and propionic acid (4.86) [24]. Furthermore, hydrochloric acid (pKa=−8.0) and sulfuric acid (pKa=−3.0) caused P-body formation with a lowering of the pH of the SD media (Figure 4A). Since lactic acid could not induce P-body formation in the SD media with pH 3.0–4.0 (Figure 4C), lowering the pH of SD medium seems to contribute to the generation of P-bodies by lactic acid. Lowering the pH of the extracellular environment might trigger the formation of P-bodies.

The threshold of the external pH to induce P-body assembly seems to be around pH 2.5 (Figures 1A and 4A). Previous studies suggested that pH 2.5 is critical for yeast physiology. Most Saccharomyces strains can grow in medium with pH values of 3.0–8.5, and the kinetics of growth and fermentation are not affected at pH 3.5–6.0 [45]. Additionally, the intracellular pH level of S. cerevisiae is well maintained using the plasma membrane ATPase during a shift of external pH between 3.0 and 7.5 [46]. On the other hand, many S. cerevisiae strains cannot grow at pH 2.5 [27,47,48], and ATPase activity is reduced to approximately 30% of the maximum value and yeast growth is drastically inhibited at pH 2.5 [45]. These reports clearly indicate that conditions with a pH lower than 2.5 are quite harsh for yeast cells and may support the interpretation that pH 2.5 is the threshold to induce P-body formation.

Weak carboxylic acids are widely used as bacteriostats/fungicides and preservatives, and their effects on yeast cells have been studied from various angles (for reviews see [24,49]). However, almost no information is available regarding their effects on yeast translation and mRNA flux. This is the first description about the formation of cytoplasmic mRNP granules by lactic acid and low pH, bringing new information about the physiological effects of acidic stress on yeast cells. Although the mechanism of P-body assembly upon acidic stress still remains elusive, P-body formation in the null mutants pat1Δ and scd6Δ (Figure 5) indicates a difference in the P-body assembly rules according to the types of stress. Scd6 seems more important for P-body assembly caused by acidic stress and high Ca2+ levels than that caused by glucose depletion. On the other hand, Pat1 was required for P-body assembly caused by high Ca2+ levels, but not for that caused by lactic acid. The results of the present study confirm that different stresses elicit P-body formation in different ways [36]. Further detailed investigation regarding the process of P-body assembly in response to lactic acid or low pH is now underway.

AUTHOR CONTRIBUTION

Aya Iwaki performed experiments and analysed the data. Shingo Izawa designed the study, performed experiments and wrote the paper.

FUNDING

This study was supported by the Skylark Food Science Institute and the Japanese Ministry of Education, Culture, Sports, Science and Technology [grant number 23580113].

Acknowledgments

We are grateful to Dr Toshifumi Inada (Tohoku University, Miyagi, Japan) and Mr Toshiyuki Sekine (SK BIO International) for technical advice on the polysome profile analysis, Dr Pamela Silver (Harvard University, Cambridge, MA, U.S.A.) and Dr Roy Parker (The University of Arizona, Tucson, AZ, U.S.A.) for plasmids, and Dr M. Sugiyama (Osaka University, Osaka, Japan) for valuable discussions.

Abbreviations: CHX, cycloheximide; Dcp, mRNA decapping; Edc3, enhancer of mRNA decapping 3; GFP, green fluorescent protein; Lsm1, like Sm 1; mRFP, monomeric red fluorescent protein; mRNP, messenger ribonucleoprotein; Ngr1, negative growth regulatory 1; ORF, open reading frame; Pat1, protein associated with topoisomerase 1; P-body, processing body; Pub1, polyuridylate binding; Scd6, suppressor of clathrin deficiency 6; SD, synthetic dextrose; SG, stress granule; Xrn1, exoribonuclease 1

References

View Abstract