Production of pyruvate from mannitol by mannitol-assimilating pyruvate decarboxylase-negative Saccharomyces cerevisiae

Abstract

Mannitol is contained in brown macroalgae up to 33% (w/w, dry weight), and thus is a promising carbon source for white biotechnology. However, Saccharomyces cerevisiae, a key cell factory, is generally regarded to be unable to assimilate mannitol for growth. We have recently succeeded in producing S. cerevisiae that can assimilate mannitol through spontaneous mutations of Tup1-Cyc8, each of which constitutes a general corepressor complex. In this study, we demonstrate production of pyruvate from mannitol using this mannitol-assimilating S. cerevisiae through deletions of all 3 pyruvate decarboxylase genes. The resultant mannitol-assimilating pyruvate decarboxylase-negative strain produced 0.86 g/L pyruvate without use of acetate after cultivation for 4 days, with an overall yield of 0.77 g of pyruvate per g of mannitol (the theoretical yield was 79%). Although acetate was not needed for growth of this strain in mannitol-containing medium, addition of acetate had a significant beneficial effect on production of pyruvate. This is the first report of production of a valuable compound (other than ethanol) from mannitol using S. cerevisiae, and is an initial platform from which the productivity of pyruvate from mannitol can be improved.

Introduction

Mannitol is a sugar alcohol derivative of mannose and a promising carbon source for white biotechnology, since brown macroalgae contains mannitol at up to 33% (w/w, dry weight).^1-3 The budding yeast Saccharomyces cerevisiae is a key cell factory that is used for production of a wide range of industrial products.⁴ However, S. cerevisiae including the S288C reference strain is generally thought to be unable to assimilate mannitol for growth.⁵ However, we⁶ and Enquist-Newman et al.⁷ have recently succeeded in producing S. cerevisiae that can utilize mannitol, thus opening a new way to produce valuable compounds from mannitol.

Enquist-Newman et al. overexpressed genes for mannitol dehydrogenase and mannitol transporter and produced a S. cerevisiae strain that assimilates mannitol.⁷ We also produced a S. cerevisiae strain that assimilates mannitol using spontaneous mutations of Tup1-Cyc8, each of which constitute a general corepressor complex that regulates many genes.⁶ We demonstrated production of ethanol from mannitol using this S. cerevisiae strain.⁶ These mannitol-assimilating S. cerevisiae strains may also have potential for production of other valuable compounds from mannitol.

Pyruvate is widely used for production of crop-protection agents, polymers, cosmetics, and food additives, and as a starting material in the biosynthesis of pharmaceuticals (e.g., L-DOPA, alanine, L-tryptophan, and L-tyrosine).^8,9 Pyruvate production from glucose has been achieved using a pyruvate decarboxylase (Pdc)-negative S. cerevisiae TAM strain.¹⁰ The goal of this study was to produce pyruvate from mannitol using our mannitol-assimilating S. cerevisiae strain.⁶

Materials and Methods

Strains and media

The S. cerevisiae strains used in the study are listed in Table 1. ADE2 of MK4416 was removed and TRP1 was replaced with trp1Δ63 using plasmid YRp14-trp1Δ63 (ATCC 77148),¹¹ resulting in the MK5316 strain. PDC1 and PDC6 of MK5316 were eliminated¹¹ and PDC5 was deleted,¹² resulting in the MK5376 strain. Detailed information is described in Supplementary Methods. Standard yeast media were used,¹³ including SG, SM, SGE (ethanol 0.15%), and SGE (ethanol 0.3%) media consisting of 0.67% w/v yeast nitrogen base w/o amino acids (BD), complete amino acids/nucleosides (Clontech), and carbon sources: 2% v/v glycerol (SG), 2% w/v mannitol (SM), 2% v/v glycerol plus 0.15% v/v ethanol [SGE (ethanol 0.15%)], and 2% v/v glycerol plus 0.3% v/v ethanol [SGE (ethanol 0.3%)]. Glycerol (30% v/v) and sodium acetate (27.1% w/v, pH 9.1; equal to 20% w/v acetate) were each autoclaved separately from the other components and ethanol was also separately sterilized by filtration (0.20-µm pore size). Liquid medium was solidified at 2% w/v agar. Amino acids/nucleosides are removed from the medium when necessary. 5-Fluoroorotic Acid (FOA) for the counter-selection of yeast was added to the solid medium at 1 mg/ml (26). The MK5316 and MK5376 strains were maintained on SGE (0.3% ethanol) solid medium at room temperature and stored in the presence of 17% v/v glycerol at -80°C. Cultivation was conducted with a Personal Lt-10F (Taitec, Tokyo, Japan).

Table 1.

S. cerevisiae strains used in this study

Strains	Descriptions	Sources
S. cerevisiae
BY4742	MATα his3Δ1 his3Δ1 leuΔ0 lys2Δ0 ura3Δ0	Euroscarf
MK4416a	BY4742 cyc8 (Δ1139–1164/ Q380ASCKTGRKX)	(14)
MK5286	MK4416 ade2Δ0	This study
MK5316	MK4416 ade2Δ0 trp1Δ63	This study
MK5327	MK4416 ade2Δ0 trp1Δ63 pdc6Δ0	This study
MK5336	MK4416 ade2Δ0 trp1Δ63 pdc6Δ0 pdc1Δ0	This study
MK5376	MK4416 ade2Δ0 trp1Δ63 pdc6Δ0 pdc1Δ0 pdc5Δ::URA3	This study

^aMK4416 strain spontaneously lacks the central small region (1139–1164 nt) of CYC8 (total, 2,901 nt) resulting in a nonsense mutation in which a stop codon was created after a short new peptide (³⁸⁰ASCKTGRK³⁸⁷).⁶ MK4416 strain acquired the ability to assimilate mannitol.

Pyruvate production from mannitol

Pyruvate production was conducted as follows, unless otherwise stated. The MK5376 strain was precultured on SM solid medium, transferred to SM liquid medium (5 mL in a 100-mL Erlenmeyer flask), and cultured at 30°C at 145 strokes per min (spm) for approximately 24 h. Cells precultured in each medium were transferred to SM liquid medium (10 mL in a 100-mL Erlenmeyer flask) and further cultured at 0, 95, or 145 spm at 30°C.

Analytical methods

Cultures were centrifuged at 20,000 g for 5 min at 4°C and each component in the supernatant was analyzed. Ethanol was assayed using an Ethanol Assay F-kit (Roche). The concentration of mannitol was determined using an HPLC equipped with an Aminex HPX-87H (300 × 7.8 mm) (Bio-Rad) column (65°C, elution with 5 mM H₂SO₄ at 0.65 ml/min) and a RID-10A detector (Shimadzu).⁶ Pdc activity was assayed as described elsewhere.¹⁴ The protein concentration was determined using a Bradford reagent assay (Sigma)¹⁵ with bovine serum albumin as a standard.

Results and Discussion

Growth phenotype of mannitol-assimilating Pdc-negative S. cerevisiae

We previously found that S. cerevisiae BY4742 cells capable of assimilating mannitol arise spontaneously from wild-type BY4742 cells during prolonged culture in mannitol-containing medium due to spontaneous mutations in genes encoding Tup1 or Cyc8, which constitute a general corepressor complex.⁶ Of the strains that acquired the ability to assimilate mannitol, the MK4416 strain had a spontaneous partial deletion in CYC8 (Table 1) and showed salt tolerance, as well as high ethanol productivity,⁶ and was chosen as the mannitol-assimilating strain in this study. Auxotrophy for Ade and Trp was introduced into the MK4416 strain to give the MK5316 strain, which was used as the parental mannitol-assimilating strain. The three genes for Pdc in S. cerevisiae (PDC1, PDC5, and PDC6)¹⁶ were deleted in the MK5316 strain to give the Pdc-negative and mannitol-assimilating MK5376 strain (Table 1), which was confirmed to have no Pdc activity (Supplementary Results).

The S. cerevisiae Pdc-negative strain in a T2–3D or CEN.PK 113–7D background cannot grow in the presence of glucose in a defined liquid or solid medium, but can grow in a glucose-limited chemostat culture in the presence of C2 compounds (ethanol or acetate).^10,14 This requirement for C2 compounds was attributed to a deficiency of this Pdc-negative strain to synthesize cytosolic acetyl-CoA.^17,18 On the defined medium, the Pdc-negative MK5376 strain exhibited no growth in the presence of glucose as reported,^10,14 but showed growth in the presence of glycerol, glycerol plus ethanol, pyruvate, and, in particular, with mannitol alone (Fig. S1). Cytosolic acetyl-CoA would be supplied from functional mitochondria when the MK5376 strain assimilates mannitol of which assimilation requires functional mitochondrial respiration.⁶ The Pdc-negative MK5376 strain in the BY4742 background showed poorer growth in a complex medium than in a defined medium (Fig. S1), in contrast to previous findings for the Pdc-negative strain in a T2–3D background.¹⁷

Pyruvate production from mannitol

In a previous attempt to produce pyruvate from glucose using S. cerevisiae, a Pdc-negative strain in a CEN.PK113–7D background was evolved to a TAM strain.¹⁰ The TAM strain was independent of C2 compounds and tolerant to glucose, and produced 135 g/L pyruvate with an overall yield of 0.54 g of pyruvate per g of glucose.¹⁰ To produce pyruvate from mannitol using our MK5376 strain, the conditions for preculture of MK5376 strain were first examined. SG and SM media gave better pyruvate production than SGE (0.15 or 0.3% ethanol) media, although growth was not affected by these 4 media (Fig. 1A). Thus, we chose SM for preculture. Among the tested shaking speeds of 0, 95, and 145 spm, a speed of 0 spm (i.e., a static batch culture) resulted in the best production of pyruvate from mannitol (Fig. 1B). Using these conditions (preculture in SM medium and pyruvate production using a static batch culture), the pyruvate and ethanol productivity of the mannitol-assimilating Pdc-negative MK5376 strain was compared with that of the parental Pdc-positive MK5316 strain (Fig. 2). The parental strain produced no pyruvate, but the Pdc-negative strain produced 0.86 g/L pyruvate through consumption of 1.12 g/L mannitol after cultivation for 4 days, with an overall yield of 0.77 g of pyruvate per g of mannitol (the theoretical yield was 79%). This was a higher yield, but lower productivity, compared to the TAM strain (yield of 0.54 g of pyruvate per g of glucose, 135 g/L pyruvate production, consumption of 250 g/L glucose for 4 days).¹⁰ The difference in pyruvate productivity between the Pdc-negative strain MK5376 and the TAM strain could be attributed to fact that MK5376 metabolized mannitol less efficiently than TAM metabolized glucose, as indicated by the amount of sugar consumption (1.12 g/L mannitol vs 250 g/L glucose) and biomass formation (A₆₀₀ of 1.7 [MK5376] vs. A₆₀₀ of 50 [TAM strain]) after 4 d of cultivation (Fig. 2).⁴ The Pdc-negative strain MK5316 must acquire the enhanced ability to metabolize mannitol, e.g., through adaptive evolution, as in the case of TAM.

Figure 1.

Production of pyruvate from mannitol by the mannitol-assimilating Pdc-negative MK5376 strain. Growth (left) and pyruvate concentration (right) for the MK5376 strain (A) precultured in several media, (B) cultured at several shaking speeds in SM medium, and (C) cultured in SM medium plus 0.3% w/v acetate. Pyruvate production was conducted as described in the Materials and Methods, except that (A) the MK5376 strain was precultured on SGE (0.3% ethanol) (closed circles), SGE (0.15%EtOH) (asterisks), SG (closed triangles), and SM (closed squares) solid media, further precultured in the same liquid media, and statically cultured at 0 spm at 30°C; (B) cultivation was conducted at 0 (closed squares), 95 (asterisks), and 145 (closed triangles) spm at 30°C; and (C) cultivation was conducted as in (B), but in the presence of 0.3% w/v acetate. (A-C) Averages and maximum and minimum values are presented (n=2). (A) For precultivation in SM medium, averages ± standard deviations (SD) are presented (n=6).

Figure 2.

Comparison of growth profiles between the Pdc-negative MK5376 strain (closed squares) and the Pdc-positive parental MK5316 strain (asterisks). Pyruvate production was conducted as described in the Materials and Methods. Averages and maximum and minimum values (n=2) are presented, except that the growth and pyruvate concentration for the MK5376 strain are averages ± SD (n=6), and concentrations of ethanol and mannitol for the MK5376 strain are averages ± SD (n=3).

Effect of acetate on pyruvate production

The mannitol-assimilating Pdc-negative MK5376 strain showed no requirement for C2 compounds, but there is a possibility that C2 compounds helped the supply of cytosolic acetyl-CoA and enhanced pyruvate production. As expected, addition of acetate up to 0.3% w/v had a significant effect on pyruvate production. First, the strain produced pyruvate at shaking speeds of 95 and 145 spm (Fig. 1C), whereas no pyruvate was produced at 95 and 145 spm in the absence of C2 compounds (Fig. 1B). Second, the strain produced a larger amount of pyruvate in the presence of acetate compared to that in the absence of acetate (Fig. 1B and C). After long-term cultivation (20 days) at 0 spm, 2.71 g/L pyruvate was produced through consumption of 2.2 g/L mannitol and 0.88 g/L acetate (Fig. S2). The mechanism underlying this beneficial effect of acetate is unclear and further challenges are needed to improve productivity of pyruvate. However, this is the first demonstration of production of a valuable compound, other than ethanol, from mannitol using S. cerevisiae.

Disclosure of Potential Conflicts of Interest

No potential conflicts of interest were disclosed

Funding

This work was supported by a JSBBA Innovative Research Program Award (to S. K.), the 41th Iwatani Naoji Foundation, Iwatani Scientific Technology, Research Grant-in-Aid (to S.K.), JST A-STEP FS-stage (AS262Z00582N) (to S.K.), Shorai Foundation for Science and Technology, Research Grant-in-Aid in 2015 (to S.K.), and Asahi Group Foundation, Research Grant-in-Aid in 2015 (to S.K.).

Supplemental Material

Supplemental data for this article can be accessed on the publisher's website.