An expanded enzyme toolbox for production of cis, cis-muconic acid and other shikimate pathway derivatives in Saccharomyces cerevisiae

An expanded enzyme toolbox for production of cis, cis-muconic acid and other shikimate pathway... Abstract A wide range of commercially relevant aromatic chemicals can be synthesized via the shikimic acid pathway. Thus, this pathway has been the target of diverse metabolic engineering strategies. In the present work, an optimized yeast strain for production of the shikimic acid pathway intermediate 3-dehydroshikimate (3-DHS) was generated, which is a precursor for the production of the valuable compounds cis, cis-muconic acid (CCM) and gallic acid (GA). Production of CCM requires the overexpression of the heterologous enzymes 3-DHS dehydratase AroZ, protocatechuic acid (PCA) decarboxylase AroY and catechol dioxygenase CatA. The activity of AroY limits the yield of the pathway. This repertoire of enzymes was expanded by a novel fungal decarboxylase. Introducing this enzyme into the pathway in the optimized strain, a titer of 1244 mg L−1 CCM could be achieved, yielding 31 mg g−1 glucose. This represents the highest yield of this compound reported in Saccharomyces cerevisiae to date. To demonstrate the applicability of the optimized strain for production of other compounds from 3-DHS, we overexpressed AroZ together with a mutant of a para-hydroxybenzoic acid hydroxylase with improved substrate specificity for PCA, PobAY385F. Thereby, we could demonstrate the production of GA for the first time in S. cerevisiae. Saccharomyces cerevisiae, muconic acid, gallic acid, gallic acid decarboxylase, shikimic acid pathway, metabolic engineering, biotechnology INTRODUCTION Aromatic compounds and their derivatives have a wide range of applications, from pharmaceuticals to bulk and fine chemicals. Today, their production relies mainly on the use of fossil resources. As the supply of fossil resources will become sparse, an environmentally sustainable production from renewable feedstocks becomes more and more attractive. Thus, an increasing demand for biotechnological production processes by microbial fermentations has developed. Consequently, the shikimic acid pathway as a major source of aromatic compounds in microbial cells is an important target for metabolic engineering strategies (Suastegui and Sha 2016; Gottardi et al. 2017). cis, cis-muconic acid (CCM) is a precursor of the bulk chemical adipic acid, which is used for production of nylon, plastics, lubricants and softeners. Adipic acid is industrially synthesized from benzol by reduction to cyclohexane, followed by oxidation to cyclohexanone and cyclohexanol, in an energy-consuming process that leads to the release of greenhouse gasses, disadvantages which have stimulated the development of alternative microbial production processes. A route for production of CCM from shikimic acid pathway intermediate 3-dehydroshikimate (3-DHS) via protocatechuic acid (PCA) and catechol has been first established in Escherichia coli (Draths and Frost 1994). However, due to its acid tolerance and resistance to microbial contaminations, the yeast Saccharomyces cerevisiae would be particularly well suited as an industrial production host. The heterologous CCM production pathway first established in S. cerevisiae consists of a 3-DHS dehydratase (AroZ) from Podospora anserina; a PCA decarboxylase (AroY) consisting of subunits B, C and D from Klebsiella pneumoniae; and catechol-1, 2-dioxygenase (CatA) from Acinetobacter calcoaceticus (Weber et al. 2012) (Fig. 1). In this study, a drain of the intermediate 3-DHS was prevented by deletion of the E-domain in Aro1. Aro1 is a pentafunctional protein and catalyzes step 2 through 6 in the biosynthesis of chorismate, which is a precursor to aromatic amino acids. The E-domain encodes the dehydrogenase function required for conversion of 3-DHS into shikimate. The same effect could be achieved in a more recent study, introducing the point mutation D1409A in the E-domain, thus destroying its function (Suastegui et al. 2016). Introduction of a feedback resistant variant of 3-deoxy-D-arabino-heptulosonate-7-phosphate (DAHP) synthase Aro4 catalyzing the entrance reaction into the shikimic acid pathway, Aro4K229L, has allowed for the production of even higher CCM titers (Curran et al. 2013; Suastegui et al. 2016, 2017). Moreover, optimizing the availability of the precursor erythrose-4-phosphate (E4P) has been another strategy to enhance the flux into the shikimic acid pathway (Fig. 1). Deletion of glucose-6-phosphate dehydrogenase encoded by ZWF1 forces the entry of glycolytic intermediates into the non-oxidative pentose phosphate pathway, which in combination with overexpression of transketolase TKL1 diverts the flux from the glycolytic intermediates fructose-6-phosphate (F6P) and glyceraldehyde-3-phosphate (G3P) to E4P and xylulose-5-phosphate (Curran et al. 2013). These modifications have permitted the production of high titers of CCM (Curran et al. 2013; Suastegui et al. 2016, 2017; Leavitt et al. 2017). Despite these achievements, the decarboxylase reaction converting PCA to catechol remains a bottleneck of the established pathway. The activity of AroY could be improved by applying oxygen-limited conditions for culture growth (Suastegui et al. 2016). AroY isomeric subunit C (AroY-Ciso), which contains the decarboxylase function of the enzyme, requires the presence of the cofactor prenylated flavin mononucleotide (prFMN) for its activity (Payer et al. 2017; Weber et al. 2017). The biosynthesis of this cofactor from FMN is catalyzed by AroY subunit B or, more efficiently, by the endogenous yeast enzyme Pad1. Consequently, the activity of AroY-C could be enhanced by optimizing the cofactor supply (Leavitt et al. 2017; Weber et al. 2017). In this study, we have successfully utilized a novel class of fungal gallic acid (GA)/PCA decarboxylases for production of CCM in S. cerevisiae. We have developed a strain overexpressing feedback resistant variants of DAHP synthases, Aro3K222L and Aro4K229L, together with an Aro1 variant with a mutation in the E-domain, Aro1K1370A, leading to an enhanced flux into the shikimic acid pathway and allowing for the accumulation of the intermediate 3-DHS. Moreover, the ZWF1 gene was replaced by integration of an expression cassette containing TKL1 under control of the strong, constitutive TDH3 promoter. Expressing the novel decarboxylases in the optimized strain, we could achieve a titer of 1244 mg L−1 and a yield of 31 mg g−1 glucose in a batch fermentation, which represents the highest CCM yield in S. cerevisiae published to date (Table S1, Supporting Information). Figure 1. View largeDownload slide Overview of optimization strategies for the synthesis of aromatic compounds cis, cis-muconic acid (CCM) and gallic acid (GA) in S. cerevisiae. A deletion of the ZWF1 (Δzwf1) gene encoding glucose-6-phosphate dehydrogenase in combination with overexpression of transketolase Tkl1 abolishes entry of glucose-6-phosphate (G6P) into the oxidative part of the pentose phosphate pathway and forces entry of glycolytic intermediates fructose-6-phosphate (F6P) and glycerin-aldehyde-3-phosphate (GAP) into the non-oxidative part. Thereby, the formation of shikimic acid pathway precursor erythrose-4-phosphate (E4P) is improved. Overexpression of feedback resistant variants of 3-Deoxy-D-arabinoheptulosonate 7-phosphate (DAHP) synthases Aro3 and Aro4 with a lysine to leucine exchange at position 222 or 229, Aro3K222Land Aro4K229L, catalyzing the reaction of E4P and phosphoenolpyruvate (PEP) to DAHP, enhances the flux into the shikimic acid pathway. Deletion of the E-domain harboring the shikimate dehydrogenase function in the pentafunctional enzyme Aro1 (ΔaroE) and overexpression of mutants with a disrupted function of the E-domain (Aro1 K1370A, Aro1D1407A or Aro1ΔE) cause an accumulation of the shikimic acid intermediate 3-dehydroshikimate (3-DHS). 3-DHS is the precursor of heterologous muconic acid (CCM) and GA biosynthetic pathways. 3-DHS dehydratase AroZ catalyzes the first step of the heterologous pathways, the conversion of 3-DHS to protochatechuic acid (PCA). For production of CCM, a bacterial (AroY) or eukaryotic PCA decarboxylase is required to generate catechol, which is subsequently converted to CCM by catechol dioxygenase CatA. Hydroxylation of PCA to GA requires expression of PobAY385F, a para-hydroxy benzoic acid hydroxylase with a tyrosine to phenylalanine exchange causing a higher substrate affinity for PCA. Multiple enzymatic steps are shown as dashed lines. Figure 1. View largeDownload slide Overview of optimization strategies for the synthesis of aromatic compounds cis, cis-muconic acid (CCM) and gallic acid (GA) in S. cerevisiae. A deletion of the ZWF1 (Δzwf1) gene encoding glucose-6-phosphate dehydrogenase in combination with overexpression of transketolase Tkl1 abolishes entry of glucose-6-phosphate (G6P) into the oxidative part of the pentose phosphate pathway and forces entry of glycolytic intermediates fructose-6-phosphate (F6P) and glycerin-aldehyde-3-phosphate (GAP) into the non-oxidative part. Thereby, the formation of shikimic acid pathway precursor erythrose-4-phosphate (E4P) is improved. Overexpression of feedback resistant variants of 3-Deoxy-D-arabinoheptulosonate 7-phosphate (DAHP) synthases Aro3 and Aro4 with a lysine to leucine exchange at position 222 or 229, Aro3K222Land Aro4K229L, catalyzing the reaction of E4P and phosphoenolpyruvate (PEP) to DAHP, enhances the flux into the shikimic acid pathway. Deletion of the E-domain harboring the shikimate dehydrogenase function in the pentafunctional enzyme Aro1 (ΔaroE) and overexpression of mutants with a disrupted function of the E-domain (Aro1 K1370A, Aro1D1407A or Aro1ΔE) cause an accumulation of the shikimic acid intermediate 3-dehydroshikimate (3-DHS). 3-DHS is the precursor of heterologous muconic acid (CCM) and GA biosynthetic pathways. 3-DHS dehydratase AroZ catalyzes the first step of the heterologous pathways, the conversion of 3-DHS to protochatechuic acid (PCA). For production of CCM, a bacterial (AroY) or eukaryotic PCA decarboxylase is required to generate catechol, which is subsequently converted to CCM by catechol dioxygenase CatA. Hydroxylation of PCA to GA requires expression of PobAY385F, a para-hydroxy benzoic acid hydroxylase with a tyrosine to phenylalanine exchange causing a higher substrate affinity for PCA. Multiple enzymatic steps are shown as dashed lines. To expand the application of our newly developed strain, we decided to establish the production of GA, a phenolic compound naturally occurring in several plant species. GA has several applications in pharmaceutical and chemical industries and is mainly produced through acid hydrolysis of tannins or biotransformation procedures (Seth and Chand 2000; Bajpai and Patil 2008; Aguilar-Zárate et al. 2015). These procedures suffer from low yields and low purity and cause environmental pollution. A pathway for production of GA from 3-DHS has already been established in E. coli (Kambourakis, Draths and Frost 2000). Using P. anserina AroZ and a mutated version of para-hydroxybenzoic acid hydroxylase PobA from Corynebacterium glutamicum with a changed substrate specificity, we were able to demonstrate the production of GA for the first time in S. cerevisiae, reaching a titer of 682 mg L−1 from glucose. MATERIAL AND METHODS Strains and media Yeast strains used in this work are listed in Table 1. Saccharomyces cerevisiae strains were grown at 30°C with 180 rpm shaking in synthetic complete (SC) medium (6.7 g L−1 Difco yeast nitrogen base without amino acids), supplemented with amino acids as described previously (Bruder et al. 2016) and with 20 g L−1 D-glucose as a carbon source (SCD), adjusted to pH 6.3, or in YEPD medium (20 g L−1 peptone, 10 g L−1 yeast extract, 20 g L−1 glucose). For propagation of strains CEN.PK2–1C ΔaroE, JTY1 and JTY4 in SC medium, concentrations of aromatic amino acids tyrosine, phenylalanine and methionine were doubled. For maintenance of plasmids, selective SCD media lacked the auxotrophic markers or/and contained 200 mg L−1 G418 (geneticin) or 200 mg L−1 hygromycin, respectively. Table 1. Strains used in this study. Strain  Genotype  Reference  CEN.PK2–1C  MATa leu2–3112 ura3–52 trp1–289 his3-Δ1 MAL2–8c SUC2  EUROSCARF, Frankfurt  CEN.PK2–1CΔaroE  MATa leu2–3112 ura3–52 trp1–289 his3-Δ1 MAL2–8c SUC2 ARO1Δ1359–1588::loxP  Weber et al. (2012)  JTY1  CEN.PK2–1C ARO1Δ1359–1588::loxP pyk2Δ::TPI1p-aro4K229L;HXT7p-aro3K222L;FBA1p-aro1K1370A;kanMX  This work  JTY4  JTY1 zwf1Δ::TDH3p-TKL1  This work  Strain  Genotype  Reference  CEN.PK2–1C  MATa leu2–3112 ura3–52 trp1–289 his3-Δ1 MAL2–8c SUC2  EUROSCARF, Frankfurt  CEN.PK2–1CΔaroE  MATa leu2–3112 ura3–52 trp1–289 his3-Δ1 MAL2–8c SUC2 ARO1Δ1359–1588::loxP  Weber et al. (2012)  JTY1  CEN.PK2–1C ARO1Δ1359–1588::loxP pyk2Δ::TPI1p-aro4K229L;HXT7p-aro3K222L;FBA1p-aro1K1370A;kanMX  This work  JTY4  JTY1 zwf1Δ::TDH3p-TKL1  This work  View Large Metabolite analysis For metabolite analysis, yeast cells were removed from samples by centrifugation. Proteins were precipitated by addition of sulfosalicylic acid to a final concentration of 5%, and concentrations of metabolites were measured by high-performance liquid chromatography (HPLC). The metabolites were separated by HPLC chromatography (Dionex) using a Nucleogel Sugar 810 H column (Macherey-Nagel GmbH & Co, Germany). The column was eluted with 5 mM H2SO4 as mobile phase and a flow rate of 0.6 ml min−1 at the temperature of 65°C or 57°C. The detection of glucose was by means of a Shodex RI-101 refractive index detector. A UV detector was used for the detection of PCA (220 nm), catechol (220 nm), GA (220 nm) and CCM (250 nm). For data evaluation, the Chromeleon software (version 6.50) was used. Plasmid and strain construction Plasmids and primers used in this study are listed in Tables 2 and 3. Molecular techniques were performed according to published procedures (Wiedemann and Boles 2008). Codon-optimized gene versions for increased protein expression in S. cerevisiae were obtained from DNA2.0 (USA) or GeneArt (Germany). Yeast transformations and reisolation of plasmid DNA from yeast cells were carried out as described previously (Boles and Zimmermann 1993; Xiao, Gietz and Woods 2006). Genes were cloned by homologous recombination. The coding regions of the respective genes were amplified by PCR from genomic DNA or plasmids by using specific primer pairs with 5΄-extensions overlapping vector sequences. PCR fragments were co-transformed into yeast together with a linearized vector. Synthetic genes for construction of p426-MET25-TaGDC1 and p426-MET25-MmGDC1 were ordered with 30 base pairs overlap to the MET25 promoter and CYC1 terminator and directly cloned by homologous recombination. All vector encoded genes were controlled by strong promotors (Table 2). Plasmids were amplified in E. coli strain DH10b (Life Technologies). Escherichia coli transformations were performed via electroporation according to the methods of Dower, Miller and Ragsdale (1988). Escherichia coli was grown on LB (Luria-Bertani) medium with 40 μg ml−1 ampicillin for plasmid selection. Genomic integrations were constructed by using the cre-loxP-kanMX4-loxP system (Güldener et al. 1996) or using the CRISPR/Cas system (Generoso et al. 2016). The loxP-kanMX-loxP deletion/integration cassette was PCR amplified from the plasmid pHD8 by using primers binding in the region homologous to the integration site in the PYK2 locus present in the plasmid, allowing homologous recombination. The cassette was transformed into CEN.PK2–1C and selected for geneticin (G418) resistance. The deletion/integration was verified by PCR analysis. Table 2. Plasmids used in this study. Plasmid  Regulatory elements, markers, genes  Genbank accession no  Reference  p423-H7  2μ, HXT7p, CYC1t, HIS3  –  Hamacher et al. (2002)  p425-H7  2μ, HXT7p, CYC1t, LEU2  –  Hamacher et al. (2002)  p426-H7  2μ, HXT7p, CYC1t, URA3  –  Hamacher et al. (2002)  pRS42K  2μ, HXT7p, CYC1t, kanMX  –  Taxis and Knop (2006)  pRS42H  2μ, HXT7p, FBAt, HygromycinR  –  Taxis and Knop (2006)  pRS41H  CEN6_ARS4, TEFp, CYC1t, HygromycinR  –  Taxis and Knop (2006)  pRS62H  2μ, HXT7p, CYC1t, HygromycinR  –  Farwick et al. (2014)  pRS62N  2μ, HXT7p, CYC1t, ClonNatR  –  Farwick et al. (2014)  p425-MET25  2μ, MET25p, CYC1t, LEU2  –  Mumberg et al. (1994)  p426-MET25  2μ, MET25p, CYC1t, URA3  –  Mumberg et al. (1994)  p423-H7-PaAroZ  aroZ from Podospora anserina  CAD60599  Weber et al. (2012)  p426-MET25-AroY-B-C-D  MET25p; aroY subunit B; CYC1t; PGK1p; aroY subunit C isoform; PGK1t; TPI1p; aroY subunit D; TAL1t; all subunits from Klebsiella pneumoniae codon-optimized (DNA2.0 optimized)  AAY57854 AB479384 AAY57856  Weber et al. (2017)  p426 -AGDC1  PGK1p; agdc1 from Arxula adeninivorans, codon optimized (GeneArt); PGK1t  SJN60119.1  This work  p425-H7-ArCatA  catA from Acinetobacter radioresistens codon optimized (DNA2.0)  AF380158  Weber et al. (2012)  p425-H7-Aro3  ARO3 from Saccharomyces cerevisiae  YDR035W  This work  p425-H7-Aro3K222L  ARO3 from Saccharomyces cerevisiae with amino acid exchange K222L codon optimized (GeneArt)  YDR035W  This work  pRS42K-Aro4K229L  HXT7p, ARO4 from Saccharomyces cerevisiae with amino acid exchange K229L codon optimized (GeneArt), CYC1t  YBR249C  This work  pRS42H-Aro1ΔaroE  HXT7p; ARO1 from Saccharomyces cerevisiae lacking the aroE encoding domain (amino acids 1359–1588); FBA1t  YDR127W  Weber et al. (2012)  pRS42H-Aro1K1370A  2μ, HXT7p, ARO1 from Saccharomyces cerevisiae with amino acid exchange K1370A, FBA1t, HygromycinR  YDR127W  This work  pRS42H-Aro1D1407A  HXT7p, ARO1 from Saccharomyces cerevisiae with amino acid exchange D1407A, FBA1t  YDR127W  This work  p426-MET25-AGDC1  Arxula adeninivorans Gallat Decarboxylase codon optimized (GeneArt)  SJN60119.1  This work  p426-MET25-TaGDC1  Talaromyces atroroseus Gallat Decarboxylase codon optimized (JCat)  XP_020118381.1  This work  p426-MET25-MmGDC1  Madurella mycetomatis Gallat Decarboxylase codon optimized (JCat)  KXX81388.1  This work  pRCC-N  2μ, ROX3p-cas9-CYC1t; SNR52p-gRNA-SUP4t, natMX  –  Generoso et al. (2016)  pRCC-N zwf1  zwf1 gRNA  YNL241C  This work  pHD-Aro3K222L, Aro4K229L, Aro1K1370A  2μ, TPI1p‐ARO3K222L‐CYC1t, HXT7p‐ARO4K229L‐ARO4t, FBA1p-ARO1K1370A-RKI1t-KanMX  YDR035W YBR249C YDR127W  This work  p423-H7-ArCatA, TaGDC1, PaAroZ  HXT7p, catA from Acinetobacter radioresistens (codon optimized), RPE1t, ZEO1p, Talaromyces atroroseus Gallat Decarboxylase (codon optimized), TAL1t, TDH3p, aroZ from Podospora anserina, pPGK1t  AF380158 XP_020118381.1 CAD60599  This work  p425-H7-CgPobA  Corynebacterium glutamicum pobA  WP_011014104.1  This work  p427-H7-CgPobAY385F  Corynebacterium glutamicum pobA with amino acid exchange Y385F  WP_011014104.1  This work  Plasmid  Regulatory elements, markers, genes  Genbank accession no  Reference  p423-H7  2μ, HXT7p, CYC1t, HIS3  –  Hamacher et al. (2002)  p425-H7  2μ, HXT7p, CYC1t, LEU2  –  Hamacher et al. (2002)  p426-H7  2μ, HXT7p, CYC1t, URA3  –  Hamacher et al. (2002)  pRS42K  2μ, HXT7p, CYC1t, kanMX  –  Taxis and Knop (2006)  pRS42H  2μ, HXT7p, FBAt, HygromycinR  –  Taxis and Knop (2006)  pRS41H  CEN6_ARS4, TEFp, CYC1t, HygromycinR  –  Taxis and Knop (2006)  pRS62H  2μ, HXT7p, CYC1t, HygromycinR  –  Farwick et al. (2014)  pRS62N  2μ, HXT7p, CYC1t, ClonNatR  –  Farwick et al. (2014)  p425-MET25  2μ, MET25p, CYC1t, LEU2  –  Mumberg et al. (1994)  p426-MET25  2μ, MET25p, CYC1t, URA3  –  Mumberg et al. (1994)  p423-H7-PaAroZ  aroZ from Podospora anserina  CAD60599  Weber et al. (2012)  p426-MET25-AroY-B-C-D  MET25p; aroY subunit B; CYC1t; PGK1p; aroY subunit C isoform; PGK1t; TPI1p; aroY subunit D; TAL1t; all subunits from Klebsiella pneumoniae codon-optimized (DNA2.0 optimized)  AAY57854 AB479384 AAY57856  Weber et al. (2017)  p426 -AGDC1  PGK1p; agdc1 from Arxula adeninivorans, codon optimized (GeneArt); PGK1t  SJN60119.1  This work  p425-H7-ArCatA  catA from Acinetobacter radioresistens codon optimized (DNA2.0)  AF380158  Weber et al. (2012)  p425-H7-Aro3  ARO3 from Saccharomyces cerevisiae  YDR035W  This work  p425-H7-Aro3K222L  ARO3 from Saccharomyces cerevisiae with amino acid exchange K222L codon optimized (GeneArt)  YDR035W  This work  pRS42K-Aro4K229L  HXT7p, ARO4 from Saccharomyces cerevisiae with amino acid exchange K229L codon optimized (GeneArt), CYC1t  YBR249C  This work  pRS42H-Aro1ΔaroE  HXT7p; ARO1 from Saccharomyces cerevisiae lacking the aroE encoding domain (amino acids 1359–1588); FBA1t  YDR127W  Weber et al. (2012)  pRS42H-Aro1K1370A  2μ, HXT7p, ARO1 from Saccharomyces cerevisiae with amino acid exchange K1370A, FBA1t, HygromycinR  YDR127W  This work  pRS42H-Aro1D1407A  HXT7p, ARO1 from Saccharomyces cerevisiae with amino acid exchange D1407A, FBA1t  YDR127W  This work  p426-MET25-AGDC1  Arxula adeninivorans Gallat Decarboxylase codon optimized (GeneArt)  SJN60119.1  This work  p426-MET25-TaGDC1  Talaromyces atroroseus Gallat Decarboxylase codon optimized (JCat)  XP_020118381.1  This work  p426-MET25-MmGDC1  Madurella mycetomatis Gallat Decarboxylase codon optimized (JCat)  KXX81388.1  This work  pRCC-N  2μ, ROX3p-cas9-CYC1t; SNR52p-gRNA-SUP4t, natMX  –  Generoso et al. (2016)  pRCC-N zwf1  zwf1 gRNA  YNL241C  This work  pHD-Aro3K222L, Aro4K229L, Aro1K1370A  2μ, TPI1p‐ARO3K222L‐CYC1t, HXT7p‐ARO4K229L‐ARO4t, FBA1p-ARO1K1370A-RKI1t-KanMX  YDR035W YBR249C YDR127W  This work  p423-H7-ArCatA, TaGDC1, PaAroZ  HXT7p, catA from Acinetobacter radioresistens (codon optimized), RPE1t, ZEO1p, Talaromyces atroroseus Gallat Decarboxylase (codon optimized), TAL1t, TDH3p, aroZ from Podospora anserina, pPGK1t  AF380158 XP_020118381.1 CAD60599  This work  p425-H7-CgPobA  Corynebacterium glutamicum pobA  WP_011014104.1  This work  p427-H7-CgPobAY385F  Corynebacterium glutamicum pobA with amino acid exchange Y385F  WP_011014104.1  This work  If not stated otherwise, cloned genes are under control of promoters and terminators included in the basic vectors. Promoters were taken from 1–500 bps upstream and terminators 1–300 bps downstream of the respective open reading frames, with the exception of the HXT7 promoter (Hamacher et al. 2002), the MET25 promoter (Mumberg et al. 1994) and the ZEO1 promoter (Tochigi et al. 2010). p, promoter, t, terminator. View Large Table 3. Primers used in this study. Primer  5΄-3΄-Sequence  Target  JTP118-pPGK1-ovp426-fw  ATTAACCCTCACTAAAGGGAACAAAAGCTGGCTAGCCCCGGGGTCGACTGTTTGCAAAAA GAACAAAACTG  p426-MET25-AGDC1  CBp160 pPGK1_r  TGTTTTATATTTGTTGTAAAAAGTAGATAATTAC  See above  HWP12 GallatDCovpPGK1 for  TTATCTACTTTTTACAACAAATATAAAACAATGACCACCTCTTACGAACC  See above  HWP13 GallatDCovtPGK1rev  ATTGATCTATCGATTTCAATTCAATTCAATTTACCAGTGCAAGTCAATTTCCTTAC  See above  CBp95 tPGK1_f  ATTGAATTGAATTGAAATCGATAG  See above  CBp163 tPGK1_r  AAATAATATCCTTCTCGAAAGCTTTAACG  See above  CBp196 pHXT7_ARO3K222L_f  CAAAAAGTTTTTTTAATTTTAATCAAAAAGTTAACATGCATATGTTCATTAAGAACGAT CACGC  p425-H7-Aro3K222L  CBp197 tCYC1_ARO3K222L_r  GAGGGCGTGAATGTAAGCGTGACATAACTAATTACATGACTTACTTCTTCAAAGCCTTT CTTCTG  See above  CBp194 ARO3wt_f  CTGTCACCAAGCCAGGTGTCACTGC  p425-H7-Aro3  CBp195 ARO3wt_r  CACCTGGCTTGGTGACAGACAAGAAGTAG  See above  CBp175 pHXT7_pRS42K_f  GGCCCCCCCTCGAGGTCGACGGTATCGATAAGCTTGATATCGCTCGTAGGAACAATTTC GG  pRS42K-Aro4K229L  CBp176 pHXT7_ARO4K2229L_r  TTTTTGATTAAAATTAAAAAAACTTTTTG  See above  CBp149 pHXt7_Aro4K222L_neu_f  GAATAAACACAAAAACAAAAAGTTTTTTTAATTTTAATCAAAAAATGAGTGAATCTCCA ATGTTCG  See above  CBp116 Aro4_r_neu  CAACACCATGCAAAGTAACACCCATGAAATGGTGAGAATGAG  See above  CBp105 Aro4K229L_f_neu  CATGGGTGTTACTTTGCATGGTGTTG  See above  CBp107 Aro4K229L_tCYC1_neu_r  GTGGGGGGAGGGCGTGAATGTAAGCGTGACATAACTAATTCTATTTCTTGTTAACTTCT CTTCTTTG  See above  CBp177 tCYC1_ARO4K229L_f  AATTAGTTATGTCACGCTTACATTC  See above  CBp178 tCYC1_pRS42K_r  GCTGGAGCTCCACCGCGGTGGCGGCCGCTCTAGAACTAGTGGCGAATTGGGTACCG  See above  CBp184 pHXT7_ARO1_f  CACAAAAACAAAAAGTTTTTTTAATTTTAATCAAAAAATGGTGCAGTTAGCCAAAGTC  pRS42H-AroK1370A  CBp134 Aro1K1370A_r  GCATTATATCTAAAGCCAGAGGAATTGTG  See above  CBp135 Aro1K1370A_f  CAATTCCTCTGGCTTTAGATATAATGC  See above  CBp136 Aro1K1370A_tCYC1_r  GAGGGCGTGAATGTAAGCGTGACATAACTAATTACATGACCTACTCTTTCGTAACGGCA TC  See above  CBp179 pHXT7_Aro1_f  CTGGAGCTCCACCGCGGTGGCGGCCGCTCTAGAACTAGTGAGCTCGTAGGAACAATTTC GG  See above  CBp180 pHXT7_Aro1_r  GATAATATCATTTCCTAGAATTGGGACTTTGGCTAACTGCACCATTTTTTGATTAAAAT TAAAAAAACTTTTTGTTTTTGTG  See above  CBp181 tCYC1_Aro1_f  GCCCTTTCAAGGCCATTTTTGATGCCGTTACGAAAGAGTAGGTCATGTAATTAGTTATG TCACGC  See above  CBp182 tCYC1_Aro1_r  GGAAACCGACGCCCCAGCACTCGTCCGAGGGCAAAGGAATAAGGTACCGGCCGCAAATT AAAG  See above  CBp137 Aro1K1407A_r  CGGATACCTAACCAAGCGGTATTATCAC  pRS42H-Aro1D1407A  CBp138 Aro1K1407A_f  GGGTGATAATACCGCTTGGTTAGGTATCCG  See above  JTP31-GDC-ovMet25-fw  AATTCTATTACCCCCATCCATACTCTAGAAATGACCACCTCTTACGAAC  p426-MET25-AGDC1  JTP174-GDCovcyc1-rev  GGCGTGAATGTAAGCGTGACATAACTAATTACATGATTACCAGTGCAAGTCAATTTCC  See above  JTP35-TPI1p-ovMO1-fw  GGGCAGCTTGGTCTATACGTATACCCTTAGGCGGCCGCGGATCCATTTAAACTGTGAGG ACCTTAATAC  pHD-Aro3K222L, Aro4K229L, Aro1K1370A  MGP2-TPI1p-oAro4 revv  GTTGGCAGCGAACATTGGAGATTCACTCATTTTTAGTTTATGTATGTGTTTTTTGTAG  See above  MGP3-optAro4-oTPIp fw  TCTATAACTACAAAAAACACATACATAAACTAAAAATGAGTGAATCTCCAATGTT  See above  JTP36-Aro4t-ovhxt7p-rev  CGCTAGCGTGGCGGCCGCTCTAGAGGATCCAACTTATGTATGTTTCGATG  See above  JTP37-Hxt7p-ovAro4t-fw  GGATCCTCTAGAGCGGCCGCCACGCTAGCGGAGCTCGTAGGAACAATTTCGGG  See above  MGP109-pHXT7-rev  TTTTGATTAAAATTAAAAAAACTTTTTGT  See above  MGP25-ARO3opHXT7-rev  GATCGTTCTTAATGAACATTTTTTGATTAAAATTAAAAAAAC  See above  MGP24- ARO3opHXT7- fw  AAGTTTTTTTAATTTTAATCAAAAAATGTTCATTAAGAACGATC  See above  JTP38-Aro3-ovcyc1t-rev  TTACATGACTCGAGTTACTTCTTCAAAGCCTTTC  See above  JTP39-cyc1t-ovAro3-fw  GGCTTTGAAGAAGTAACTCGAGTCATGTAATTAGTTATGTC  See above  JTP40-cyc1p-ovFBA1p-rev  GTCGACGCTAGCGGTACCGGCCGCAAATTAAAGCCTTCGAG  See above  JTP41-FBA1p-ovcyc1t-fw  TTTAATTTGCGGCCGGTACCGCTAGCGTCGACTGGGTCATTACGTAAATAATG  See above  JTP42-FBA1p-ovAro1-rev  GACTTTGGCTAACTGCACCATTTTGAATATGTATTACTTGG  See above  JTP43-Aro1-ovFBA1p-fw  ATATTCAAAATGGTGCAGTTAGCCAAAGTC  See above  JTP44-Aro1-ovRKIt-rev  TGCCTTTGATCTGCCTACTCTTTCGTAACGGCATC  See above  JTP45-RKIt-ovAro1-fw  CGTTACGAAAGAGTAGGCAGATCAAAGGCAAAGACAG  See above  JTP46-RKI1t-ovMO1-rev  CATACATTATACGAAGTTATATTAAGGGTTGTCGACCTTGGTGTGTCATCGGTAGTAAC  See above  sbp324-Clon-pCrCas-K/N-Zwf1-for  AGATTCAGATCTGTGACTGTTTTAGAGCTAGAAATAGCAAGTTAAAATAAGG  pRCC-N zwf1  sbp310-Clon-pCrCas-K/N-Zwf1-rev  GTCACAGATCTGAATCTAAGATCATTTATCTTTCACTGCGGAG  See above  JTP221-pTDH3ovZWF1-Donor-fw  TTCAGTGACTTAGCCGATAAATGAATGTGCTTGCATTTTTACAGTTTATTCCTGGCAT CC  Integration of TDH3p-TKL1-TKL1t  JTP222-TKL1tovZWF1-Donor-rev  ACATAACACCTAAAGTGGCTTCCTCCTGCCCCTCTCTCCCATATTCTTTATTGCTTTA TACTTGAATGG  See above  JTP249-CatArovHxt7p-fw  ACAAAAAGTTTTTTTAATTTTAATCAAAAAATGACTGCAGCAAATGTTAAG  p423-H7-ArCatA, TaGDC1, PaAroZ  CBp76 catA radio-R  TCGTATAGTATAGAGAGTATAAATATAAGAAATGCCGCATATGTACAATTAGGCTTGC AATCTTGGTCTATCC  See above  CBp80 rRPE1-F  TTGTACATATGCGGCATTTCTTATA  See above  JTP250-tRPE1-ovZEO1-rev  ACAATTTGTGGTCTTCGTCGTGCTCTGTGGGGATCCAAATGGATATTGATCTAGATGG CG  See above  JTP156-ZEO1p-fw  CCACAGAGCACGACGAAG  See above  JTP158-pZEO1-rev  ATTGATATAAACGTAGTTTTGTATGTTTC  See above  JTP251-TaGDCovZEO1p-fw  AGAAACATACAAAACTACGTTTATATCAATATGGTTTACAACGTTACTGAAAAG  See above  JTP252-TaGDCovTAL1t-rev  GGCCTAAATTAATATTTCCGAGATACTTCCCTACTTTGGAACTTCCAACAAC  See above  CBp214 tTAL1_f  GGAAGTATCTCGGAAATATTAATTTAGGC  See above  CBp158 tTAL_r  GACGTTGATTTAAGGTGGTTCCG  See above  JTP253-pTDH3ovTAL1t-fw  AAACATCCGGAACCACCTTAAATCAACGTCGTCGACACAGTTTATTCCTGGCATCC  See above  CBp222 pTDH3_r  TTTGTTTGTTTATGTGTGTTTATTCG  See above  JTP254-AroZovTDH3-fw  GTTTCGAATAAACACACATAAACAAACAAAATGCCAAGTAAACTGGCTATTAC  See above  JTP255-tPGK1ovAroZ-rev  ATTGATCTATCGATTTCAATTCAATTCAATTTAAAGTGCGGCAGATAGAG  See above  CBp9tPGK_f  ATTGAATTGAATTGAAATCGATAGAT  See above  JTP256-tPGK1ovp423-r  CCAGTGAGCGCGCGTAATACGACTCACTATAGCGGCCGCAAATAATATCCTTCTCGAA AGCTTTAAC  See above  pobA_Coryne_F  AACACAAAAACAAAAAGTTTTTTTAATTTTAATCAAAAAATGAACCACGTACCAGTGG CAATT  P425-H7-CgPobA  pobA_Coryne_wt_R  GAATGTAAGCGTGACATAACTAATTACATGACTCGAGTTATACCTCGAAGCGTGGTAG  See above  pobA_Coryne_R  GAATGTAAGCGTGACATAACTAATTACATGACTCGAGTTATACCTCGAAGCGTGGTAG GTCGCGCCCAACGAACTGCTC  P425-H7-CgPobAY385F  Primer  5΄-3΄-Sequence  Target  JTP118-pPGK1-ovp426-fw  ATTAACCCTCACTAAAGGGAACAAAAGCTGGCTAGCCCCGGGGTCGACTGTTTGCAAAAA GAACAAAACTG  p426-MET25-AGDC1  CBp160 pPGK1_r  TGTTTTATATTTGTTGTAAAAAGTAGATAATTAC  See above  HWP12 GallatDCovpPGK1 for  TTATCTACTTTTTACAACAAATATAAAACAATGACCACCTCTTACGAACC  See above  HWP13 GallatDCovtPGK1rev  ATTGATCTATCGATTTCAATTCAATTCAATTTACCAGTGCAAGTCAATTTCCTTAC  See above  CBp95 tPGK1_f  ATTGAATTGAATTGAAATCGATAG  See above  CBp163 tPGK1_r  AAATAATATCCTTCTCGAAAGCTTTAACG  See above  CBp196 pHXT7_ARO3K222L_f  CAAAAAGTTTTTTTAATTTTAATCAAAAAGTTAACATGCATATGTTCATTAAGAACGAT CACGC  p425-H7-Aro3K222L  CBp197 tCYC1_ARO3K222L_r  GAGGGCGTGAATGTAAGCGTGACATAACTAATTACATGACTTACTTCTTCAAAGCCTTT CTTCTG  See above  CBp194 ARO3wt_f  CTGTCACCAAGCCAGGTGTCACTGC  p425-H7-Aro3  CBp195 ARO3wt_r  CACCTGGCTTGGTGACAGACAAGAAGTAG  See above  CBp175 pHXT7_pRS42K_f  GGCCCCCCCTCGAGGTCGACGGTATCGATAAGCTTGATATCGCTCGTAGGAACAATTTC GG  pRS42K-Aro4K229L  CBp176 pHXT7_ARO4K2229L_r  TTTTTGATTAAAATTAAAAAAACTTTTTG  See above  CBp149 pHXt7_Aro4K222L_neu_f  GAATAAACACAAAAACAAAAAGTTTTTTTAATTTTAATCAAAAAATGAGTGAATCTCCA ATGTTCG  See above  CBp116 Aro4_r_neu  CAACACCATGCAAAGTAACACCCATGAAATGGTGAGAATGAG  See above  CBp105 Aro4K229L_f_neu  CATGGGTGTTACTTTGCATGGTGTTG  See above  CBp107 Aro4K229L_tCYC1_neu_r  GTGGGGGGAGGGCGTGAATGTAAGCGTGACATAACTAATTCTATTTCTTGTTAACTTCT CTTCTTTG  See above  CBp177 tCYC1_ARO4K229L_f  AATTAGTTATGTCACGCTTACATTC  See above  CBp178 tCYC1_pRS42K_r  GCTGGAGCTCCACCGCGGTGGCGGCCGCTCTAGAACTAGTGGCGAATTGGGTACCG  See above  CBp184 pHXT7_ARO1_f  CACAAAAACAAAAAGTTTTTTTAATTTTAATCAAAAAATGGTGCAGTTAGCCAAAGTC  pRS42H-AroK1370A  CBp134 Aro1K1370A_r  GCATTATATCTAAAGCCAGAGGAATTGTG  See above  CBp135 Aro1K1370A_f  CAATTCCTCTGGCTTTAGATATAATGC  See above  CBp136 Aro1K1370A_tCYC1_r  GAGGGCGTGAATGTAAGCGTGACATAACTAATTACATGACCTACTCTTTCGTAACGGCA TC  See above  CBp179 pHXT7_Aro1_f  CTGGAGCTCCACCGCGGTGGCGGCCGCTCTAGAACTAGTGAGCTCGTAGGAACAATTTC GG  See above  CBp180 pHXT7_Aro1_r  GATAATATCATTTCCTAGAATTGGGACTTTGGCTAACTGCACCATTTTTTGATTAAAAT TAAAAAAACTTTTTGTTTTTGTG  See above  CBp181 tCYC1_Aro1_f  GCCCTTTCAAGGCCATTTTTGATGCCGTTACGAAAGAGTAGGTCATGTAATTAGTTATG TCACGC  See above  CBp182 tCYC1_Aro1_r  GGAAACCGACGCCCCAGCACTCGTCCGAGGGCAAAGGAATAAGGTACCGGCCGCAAATT AAAG  See above  CBp137 Aro1K1407A_r  CGGATACCTAACCAAGCGGTATTATCAC  pRS42H-Aro1D1407A  CBp138 Aro1K1407A_f  GGGTGATAATACCGCTTGGTTAGGTATCCG  See above  JTP31-GDC-ovMet25-fw  AATTCTATTACCCCCATCCATACTCTAGAAATGACCACCTCTTACGAAC  p426-MET25-AGDC1  JTP174-GDCovcyc1-rev  GGCGTGAATGTAAGCGTGACATAACTAATTACATGATTACCAGTGCAAGTCAATTTCC  See above  JTP35-TPI1p-ovMO1-fw  GGGCAGCTTGGTCTATACGTATACCCTTAGGCGGCCGCGGATCCATTTAAACTGTGAGG ACCTTAATAC  pHD-Aro3K222L, Aro4K229L, Aro1K1370A  MGP2-TPI1p-oAro4 revv  GTTGGCAGCGAACATTGGAGATTCACTCATTTTTAGTTTATGTATGTGTTTTTTGTAG  See above  MGP3-optAro4-oTPIp fw  TCTATAACTACAAAAAACACATACATAAACTAAAAATGAGTGAATCTCCAATGTT  See above  JTP36-Aro4t-ovhxt7p-rev  CGCTAGCGTGGCGGCCGCTCTAGAGGATCCAACTTATGTATGTTTCGATG  See above  JTP37-Hxt7p-ovAro4t-fw  GGATCCTCTAGAGCGGCCGCCACGCTAGCGGAGCTCGTAGGAACAATTTCGGG  See above  MGP109-pHXT7-rev  TTTTGATTAAAATTAAAAAAACTTTTTGT  See above  MGP25-ARO3opHXT7-rev  GATCGTTCTTAATGAACATTTTTTGATTAAAATTAAAAAAAC  See above  MGP24- ARO3opHXT7- fw  AAGTTTTTTTAATTTTAATCAAAAAATGTTCATTAAGAACGATC  See above  JTP38-Aro3-ovcyc1t-rev  TTACATGACTCGAGTTACTTCTTCAAAGCCTTTC  See above  JTP39-cyc1t-ovAro3-fw  GGCTTTGAAGAAGTAACTCGAGTCATGTAATTAGTTATGTC  See above  JTP40-cyc1p-ovFBA1p-rev  GTCGACGCTAGCGGTACCGGCCGCAAATTAAAGCCTTCGAG  See above  JTP41-FBA1p-ovcyc1t-fw  TTTAATTTGCGGCCGGTACCGCTAGCGTCGACTGGGTCATTACGTAAATAATG  See above  JTP42-FBA1p-ovAro1-rev  GACTTTGGCTAACTGCACCATTTTGAATATGTATTACTTGG  See above  JTP43-Aro1-ovFBA1p-fw  ATATTCAAAATGGTGCAGTTAGCCAAAGTC  See above  JTP44-Aro1-ovRKIt-rev  TGCCTTTGATCTGCCTACTCTTTCGTAACGGCATC  See above  JTP45-RKIt-ovAro1-fw  CGTTACGAAAGAGTAGGCAGATCAAAGGCAAAGACAG  See above  JTP46-RKI1t-ovMO1-rev  CATACATTATACGAAGTTATATTAAGGGTTGTCGACCTTGGTGTGTCATCGGTAGTAAC  See above  sbp324-Clon-pCrCas-K/N-Zwf1-for  AGATTCAGATCTGTGACTGTTTTAGAGCTAGAAATAGCAAGTTAAAATAAGG  pRCC-N zwf1  sbp310-Clon-pCrCas-K/N-Zwf1-rev  GTCACAGATCTGAATCTAAGATCATTTATCTTTCACTGCGGAG  See above  JTP221-pTDH3ovZWF1-Donor-fw  TTCAGTGACTTAGCCGATAAATGAATGTGCTTGCATTTTTACAGTTTATTCCTGGCAT CC  Integration of TDH3p-TKL1-TKL1t  JTP222-TKL1tovZWF1-Donor-rev  ACATAACACCTAAAGTGGCTTCCTCCTGCCCCTCTCTCCCATATTCTTTATTGCTTTA TACTTGAATGG  See above  JTP249-CatArovHxt7p-fw  ACAAAAAGTTTTTTTAATTTTAATCAAAAAATGACTGCAGCAAATGTTAAG  p423-H7-ArCatA, TaGDC1, PaAroZ  CBp76 catA radio-R  TCGTATAGTATAGAGAGTATAAATATAAGAAATGCCGCATATGTACAATTAGGCTTGC AATCTTGGTCTATCC  See above  CBp80 rRPE1-F  TTGTACATATGCGGCATTTCTTATA  See above  JTP250-tRPE1-ovZEO1-rev  ACAATTTGTGGTCTTCGTCGTGCTCTGTGGGGATCCAAATGGATATTGATCTAGATGG CG  See above  JTP156-ZEO1p-fw  CCACAGAGCACGACGAAG  See above  JTP158-pZEO1-rev  ATTGATATAAACGTAGTTTTGTATGTTTC  See above  JTP251-TaGDCovZEO1p-fw  AGAAACATACAAAACTACGTTTATATCAATATGGTTTACAACGTTACTGAAAAG  See above  JTP252-TaGDCovTAL1t-rev  GGCCTAAATTAATATTTCCGAGATACTTCCCTACTTTGGAACTTCCAACAAC  See above  CBp214 tTAL1_f  GGAAGTATCTCGGAAATATTAATTTAGGC  See above  CBp158 tTAL_r  GACGTTGATTTAAGGTGGTTCCG  See above  JTP253-pTDH3ovTAL1t-fw  AAACATCCGGAACCACCTTAAATCAACGTCGTCGACACAGTTTATTCCTGGCATCC  See above  CBp222 pTDH3_r  TTTGTTTGTTTATGTGTGTTTATTCG  See above  JTP254-AroZovTDH3-fw  GTTTCGAATAAACACACATAAACAAACAAAATGCCAAGTAAACTGGCTATTAC  See above  JTP255-tPGK1ovAroZ-rev  ATTGATCTATCGATTTCAATTCAATTCAATTTAAAGTGCGGCAGATAGAG  See above  CBp9tPGK_f  ATTGAATTGAATTGAAATCGATAGAT  See above  JTP256-tPGK1ovp423-r  CCAGTGAGCGCGCGTAATACGACTCACTATAGCGGCCGCAAATAATATCCTTCTCGAA AGCTTTAAC  See above  pobA_Coryne_F  AACACAAAAACAAAAAGTTTTTTTAATTTTAATCAAAAAATGAACCACGTACCAGTGG CAATT  P425-H7-CgPobA  pobA_Coryne_wt_R  GAATGTAAGCGTGACATAACTAATTACATGACTCGAGTTATACCTCGAAGCGTGGTAG  See above  pobA_Coryne_R  GAATGTAAGCGTGACATAACTAATTACATGACTCGAGTTATACCTCGAAGCGTGGTAG GTCGCGCCCAACGAACTGCTC  P425-H7-CgPobAY385F  View Large Fermentations and feeding experiments Cultures of yeast strains (50 ml) were grown in 300 ml shake flasks (Erlenmeyer flasks) at 30°C in a rotary shaker (180 rpm). Precultures were grown into the exponential phase in selective SCD media. Cells were washed with sterile water and inoculated to an optical density at 600 nm (OD600) of 0.8 in the same medium. For PCA feeding experiments, cells were inoculated to an OD600 of 0.8 in selective SCD medium containing 5 mM PCA. SCD medium with 5 mM PCA was prepared as follows. PCA was dissolved in distilled water to a concentration of 100 mM and sterile filtered. The stock solution was added to non-buffered, selective SCD medium to a concentration of 5 mM, resulting in a pH of 4.3. Growth/fermentation experiments were performed up to three times by using the same precultures, with the given standard deviations. Samples for the measurement of the OD600 and the metabolite analyses were taken at different time points. RESULTS Improving the flux into the shikimic acid pathway In order to enhance CCM production, it is important to enhance the substrate flux into the shikimic acid pathway. The two isomeric DAHP synthases Aro3 and Aro4, which catalyze the entrance reaction into the shikimic acid pathway, are known to be feedback inhibited by products of the shikimic acid pathway, tyrosine and phenylalanine, respectively (Braus 1991). Therefore, we chose to overexpress variants with abolished negative regulation in addition to the natural pathway (Fig. 2). For Aro4, the mutation K229L is known to lack feedback sensitivity (Hartmann et al. 2003) and has been successfully used in several studies (Curran et al. 2013; McKenna et al. 2014; Li et al. 2015; Rodriguez et al. 2015; Suastegui et al. 2016). As Aro3 was reported to have a higher affinity for E4P and phosphoenolpyruvate (PEP) compared to Aro4 (Hartmann et al. 2003), we also created a mutated Aro3 with an analogous K222L point mutation. For comparison, the mutated enzymes or the wild type Aro3 were overexpressed under control of the strong constitutive HXT7 promoter (Hamacher et al. 2002) together with Aro1 lacking the E-domain responsible for conversion of 3-DHS to shikimate (Aro1ΔE) in strain CEN.PK2–1C ΔaroE harboring a genomic deletion of the E-domain (Weber et al. 2012). This deletion is crucial for the accumulation of the precursor 3-DHS. Additionally, 3-DHS dehydratase AroZ was overexpressed and formation of PCA was measured (Fig. 2). In fermentations in selective SC medium, the overexpression of ARO4K229L resulted in 250 mg L−1, of wild-type ARO3 in a codon-optimized form in 304 mg L−1, and of the codon-optimized mutant ARO3K222L in 469 mg L−1 after 96 h. Combined overexpression of ARO3 K222L and ARO4K229L led to the highest PCA titer with 653 mg L−1, which is a 3.3-fold increase compared to the empty plasmid control with 197 mg L−1. Figure 2. View largeDownload slide Activity of feedback resistant DAHP synthase mutants. CEN.PK2–1C ΔaroE cells were transformed with expression plasmids p425-H7-Aro3 (Aro3 (wt)) or p425-H7-Aro3K222L (Aro3K222L) and pRS42K-Aro4K229L (Aro4K229L), or with a combination of these plasmids. Empty plasmids were transformed as controls (Plasmid). To analyze the activity of the enzymes, pRS42H-Aro1ΔE (Aro1ΔE) and p423-H7-PaAroZ (AroZ) were co-transformed, and the final protocatechuic acid (PCA) concentrations after cultivation for 96 h in shake flasks in selective synthetic complete (SCD) medium (2% glucose) at 30°C were analyzed. Error bars represent the standard deviation of two independent replicates. Figure 2. View largeDownload slide Activity of feedback resistant DAHP synthase mutants. CEN.PK2–1C ΔaroE cells were transformed with expression plasmids p425-H7-Aro3 (Aro3 (wt)) or p425-H7-Aro3K222L (Aro3K222L) and pRS42K-Aro4K229L (Aro4K229L), or with a combination of these plasmids. Empty plasmids were transformed as controls (Plasmid). To analyze the activity of the enzymes, pRS42H-Aro1ΔE (Aro1ΔE) and p423-H7-PaAroZ (AroZ) were co-transformed, and the final protocatechuic acid (PCA) concentrations after cultivation for 96 h in shake flasks in selective synthetic complete (SCD) medium (2% glucose) at 30°C were analyzed. Error bars represent the standard deviation of two independent replicates. Mutation and overexpression of Aro1 In our previous study, the whole E-domain of the pentafunctional enzyme Aro1 was deleted in order to prevent conversion of 3-DHS to shikimate, thereby improving the availability of the CCM pathway substrate 3-DHS (Weber et al. 2012). Concerns about possible negative effects on protein folding and enzyme activities prompted us to search for point mutations inside the E-domain, which should also lead to a specific loss of dehydrogenase function. Two invariant catalytic residues in the active site of the AroE homolog YdiB from E. coli, K71 and D107, had previously been identified as crucial for substrate binding (Lindner et al. 2005). In the E-domain of S. cerevisiae Aro1, these residues are located at position 1370 and 1407, respectively. The K1370A mutation and a D1409A mutation have already been described in a previous study (Suastegui et al. 2016). To test if the single amino acid exchanges cause a complete loss of the shikimate dehydrogenase function of Aro1, ARO1K1370A and ARO1D1407A expressed under control of the strong constitutive HXT7 promoter were tested for their ability to complement the aromatic amino acid auxotrophy of the CEN.PK2–1C ΔaroE strain. For this aim, a growth test was performed on media with or without the aromatic amino acids tyrosine and phenylalanine (Fig. 3A). CEN.PK2–1C ΔaroE and CEN.PK2–1C transformed with empty vector were used as controls. Both mutated ARO1 alleles were not able to complement the growth defect on media without tyrosine and phenylalanine, indicating a loss of shikimate dehydrogenase activity and making them suitable candidates for the CCM pathway. Figure 3. View largeDownload slide (A) Growth test with Aro1 point mutations. CEN.PK2–1C ΔaroE cells transformed with expression vectors pRS42H-Aro1K1370A (Aro1K1370A) or pRS42H-Aro1D1407A (Aro1D1307A) were cultivated in selective synthetic complete medium (SCD) with 2% glucose, harvested at OD600 = 1 and spotted onto selective SCD agar (2% glucose) with or without the aromatic amino acids tyrosine (Tyr) and phenylalanine (Phe) in several dilutions. CEN.PK2–1C ΔaroE and CEN.PK2–1C cells transformed with empty plasmid (Plasmid) were used as controls. The picture shows growth after 4 days at 30°C. (B) Activity of Aro1 variants. CEN.PK2–1C ΔaroE cells were transformed with pRS42H-Aro1ΔE (Aro1ΔE), pRS42H-Aro1K1370A (Aro1K1370A) or pRS42H-Aro1D1407A (Aro1D1407A) in combination with p425-H7-Aro3K222L (Aro3K222L) and p423-H7-PaAroZ (AroZ) and subsequently cultivated in selective SCD medium (2% glucose) in shake flasks. The production of PCA was determined after 96 h at 30°C. The error bars show the standard deviation of two independent replicates. Figure 3. View largeDownload slide (A) Growth test with Aro1 point mutations. CEN.PK2–1C ΔaroE cells transformed with expression vectors pRS42H-Aro1K1370A (Aro1K1370A) or pRS42H-Aro1D1407A (Aro1D1307A) were cultivated in selective synthetic complete medium (SCD) with 2% glucose, harvested at OD600 = 1 and spotted onto selective SCD agar (2% glucose) with or without the aromatic amino acids tyrosine (Tyr) and phenylalanine (Phe) in several dilutions. CEN.PK2–1C ΔaroE and CEN.PK2–1C cells transformed with empty plasmid (Plasmid) were used as controls. The picture shows growth after 4 days at 30°C. (B) Activity of Aro1 variants. CEN.PK2–1C ΔaroE cells were transformed with pRS42H-Aro1ΔE (Aro1ΔE), pRS42H-Aro1K1370A (Aro1K1370A) or pRS42H-Aro1D1407A (Aro1D1407A) in combination with p425-H7-Aro3K222L (Aro3K222L) and p423-H7-PaAroZ (AroZ) and subsequently cultivated in selective SCD medium (2% glucose) in shake flasks. The production of PCA was determined after 96 h at 30°C. The error bars show the standard deviation of two independent replicates. To test the activity of the Aro1 variants, plasmids overexpressing ARO1K1370A, ARO1D1407A and ARO1ΔaroE were transformed into CEN.PK2–1C ΔaroE. ARO3K222L was overexpressed for increased substrate flux, and AroZ for production of PCA, which was determined after 96 h of incubation in selective SCD medium in the culture supernatant (Fig. 3B). The point mutations ARO1K1370A (1014 mg L−1) and ARO1D1407A (957 mg L−1) resulted in a significant increase in PCA formation compared to the deleted version ARO1ΔaroE (597 mg L−1). Moreover, this experiment also shows the importance of additional ARO1 overexpression, as the strain carrying the empty vector instead of a vector overexpressing one of the ARO1 variants showed very low PCA production (18 mg L−1). The mutant ARO1K1370A was chosen for further experiments to prevent possible negative effects of the aroE deletion. Construction of an optimized production strain In the next step, we decided to integrate the information gained in the previous experiments into the construction of a new production strain. ARO4K229L, ARO3K222L and ARO1K1370A were combined on one plasmid under control of the strong constitutive TPI1, HXT7 and FBA1 promoters, respectively, using the backbone of 2μ plasmid pHD8 (Demeke et al. 2013) (Table 2), resulting in plasmid pHD-Aro3K222L, Aro4K229L and Aro1K1370A. In the newly generated plasmid, the integrative expression cassette is flanked by regions homologous to PYK2, a non-essential gene expressed only under growth on non-fermentable carbon sources (Boles et al. 1997). The expression cassette was subsequently stably integrated into the PYK2 locus in strain CEN.PK2–1C ΔaroE, thereby generating strain JTY1 (Table 1). The performance of the two strains was tested in comparison to strain CEN.PK2–1C and to strain CEN.PK2–1C ΔaroE transformed with the 2μ plasmid pHD-Aro3K222L, Aro4K229L and Aro1K1370A containing the expression cassette for the three genes (Fig. 4A). Upon growth in selective YEPD medium (2% glucose) containing G418, no 3-DHS was detectable in the culture supernatant of strain CEN.PK2–1C after 96 h of cultivation, while strain CEN.PK2–1C ΔaroE produced 19 mg L−1 3-DHS, confirming the importance of the deletion for accumulation of the precursor. Strain CEN.PK2–1C ΔaroE transformed with the high copy plasmid expressing the three enzymes Aro4K229L, Aro3K222L and Aro1K1370A produced 188 mg L−1 3-DHS. Finally, stable genomic integration of the expression cassette allowed for the production of up to 482 mg L−1 3-DHS with strain JTY1 after incubation for 96 h. Figure 4. View largeDownload slide Strain optimization for improved precursor supply. (A) Aro3K222L, Aro4K229L and Aro1K1370A were expressed from a multicopy plasmid (pHD-Aro3K222L, Aro4K229L, Aro1K1370A) in strain CEN.PK2–1C ΔaroE (2μ ARO3K222L, ARO4K229L, ARO1K1370A), or the expression cassette was integrated into the genome, thereby replacing the PYK2 gene (pyk2Δ::ARO3K222L, ARO4K229L, ARO1K1370A), generating strain JTY1. Cells were grown in shake flasks using YEPD medium (2% glucose) containing G418 for 96 h at 30°C, and the formation of 3-dehydroshikimate (3-DHS) in the culture supernatant was analyzed by HPLC. The error bars show the standard deviation of three independent replicates. (B) Strain JTY4 (zwf1Δ::TKL1) was generated by replacement of the ZWF1 gene by an expression cassette encoding Tkl1 in strain JTY1. Concentrations of 3-DHS in the culture supernatant were determined after 96 h at 30°C of cultivation in YEPD medium (2% glucose). The error bars show the standard deviation of two independent replicates. Figure 4. View largeDownload slide Strain optimization for improved precursor supply. (A) Aro3K222L, Aro4K229L and Aro1K1370A were expressed from a multicopy plasmid (pHD-Aro3K222L, Aro4K229L, Aro1K1370A) in strain CEN.PK2–1C ΔaroE (2μ ARO3K222L, ARO4K229L, ARO1K1370A), or the expression cassette was integrated into the genome, thereby replacing the PYK2 gene (pyk2Δ::ARO3K222L, ARO4K229L, ARO1K1370A), generating strain JTY1. Cells were grown in shake flasks using YEPD medium (2% glucose) containing G418 for 96 h at 30°C, and the formation of 3-dehydroshikimate (3-DHS) in the culture supernatant was analyzed by HPLC. The error bars show the standard deviation of three independent replicates. (B) Strain JTY4 (zwf1Δ::TKL1) was generated by replacement of the ZWF1 gene by an expression cassette encoding Tkl1 in strain JTY1. Concentrations of 3-DHS in the culture supernatant were determined after 96 h at 30°C of cultivation in YEPD medium (2% glucose). The error bars show the standard deviation of two independent replicates. To improve the performance of the strain even further, a copy of the gene encoding transketolase 1 (TKL1) was integrated into the ZWF1 locus under control of the strong TDH3 promoter, thereby disrupting the glucose-6-phosphate dehydrogenase function of the gene. These manipulations force the entry of glycolytic intermediates into the non-oxidative part of the pentose phosphate cycle via the Tkl1 reaction, leading to enhanced supply of the shikimic acid precursor E4P (Curran et al. 2013). Introducing these changes into strain JTY1 resulted in an improved titer of 3-DHS in the newly generated strain JTY4 (1552 mg L−1) compared to strain JTY1 (754 mg L−1) after 96 h (Fig. 4B). The use of non-selective YEPD medium in this experiment might be responsible for the higher titer of 3-DHS in strain JTY1 compared to the titers achieved with the same strain in the previous experiment shown in Fig. 4A, where cells were grown in YEPD medium containing G418. A novel PCA decarboxylase In previous studies in S. cerevisiae (Weber et al. 2012; Curran et al. 2013; Skjoedt et al. 2016; Suastegui et al. 2016, 2017; Leavitt et al. 2017) and E. coli (Draths and Frost 1994; Niu, Draths and Frost 2002; Sonoki et al. 2014; Zhang et al. 2015), the bacterial decarboxylase AroY has been used for the conversion of PCA to catechol, which represents the limiting step in the heterologous CCM pathway. AroY-C, the enzyme with the actual decarboxylase function, requires the prenylated flavin mononucleotide (prFMN) cofactor for its activity, which is synthesized with the help of AroY subunit B (AroY-B) (Payer et al. 2017; Weber et al. 2017). In bacterial genomes, AroY- C and -B are frequently encoded in the same operon together with subunit D (AroY-D), whose function is still unknown. Recently, a novel eukaryotic decarboxylase from the yeast Arxula adeninivorans, AGDC1, has been described (Meier et al. 2017), which shows no homology to previously known bacterial PCA decarboxylases. The enzyme has been shown to accept GA and PCA as substrates. For comparison of their activities, AGDC1 and AroY from K. pneumoniae were expressed as codon optimized versions in a PCA feeding experiment in strain CEN.PK2–1C, whereby AGDC1 and AroY-C are under control of the strong, constitutive PGK1 promoter (Fig. 5). All transformants were cultivated in selective SCD media containing 5 mM PCA for 144 hours. The pH of the medium supplemented with PCA is 4.3 at the start of the feeding experiment, corresponding to the pKa of 4.26 of the substance (Perrin and Sillen 1979), and is dropping to pH 2.8 in the course of the experiment (not shown). Thus, sufficient amounts of the undissociated form of the acid should be taken up passively by the cell. The results show that expression of AGDC1 from A. adeninivorans enhanced formation of catechol (140 mg L−1) compared to the expression of the bacterial enzyme AroY (109 mg L−1). To control that catechol is actually formed from PCA taken up by the cell and not from an intracellular source, we performed an experiment, where cells expressing AGDC1 were fed with distilled water or PCA (Fig. S1, Supporting Information). Despite the presence of a functional decarboxylase, cells fed with water did not produce measurable amounts of catechol, proving that catechol produced by AGDC1 expressing cells is solely generated by conversion of externally added PCA. Figure 5. View largeDownload slide Activities of two different decarboxylases. Expression vectors encoding bacterial KpAroY from Klebsiella pneumoniae (p426-AroY-BCD) and eukaryotic AGDC1 from Arxula adeninivorans (p426-AGDC1) were transformed in strain CEN.PK2–1C. Empty plasmid was transformed as control (Plasmid). All transformants were cultivated in shake flasks in selective synthetic complete (SCD) media (2% glucose) containing 5 mM protocatechuic acid (PCA) at 30°C. Final catechol concentrations were measured after 144 h. Error bars represent the standard deviation of three independent replicates. Figure 5. View largeDownload slide Activities of two different decarboxylases. Expression vectors encoding bacterial KpAroY from Klebsiella pneumoniae (p426-AroY-BCD) and eukaryotic AGDC1 from Arxula adeninivorans (p426-AGDC1) were transformed in strain CEN.PK2–1C. Empty plasmid was transformed as control (Plasmid). All transformants were cultivated in shake flasks in selective synthetic complete (SCD) media (2% glucose) containing 5 mM protocatechuic acid (PCA) at 30°C. Final catechol concentrations were measured after 144 h. Error bars represent the standard deviation of three independent replicates. To find other related enzymes with a potentially higher activity, we performed a BLAST search with non-redundant protein sequences (https://blast.ncbi.nlm.nih.gov/Blast.cgi). Our BLAST search revealed a plethora of sequences from diverse fungal species with identities of maximally 57% to AGDC1. For further analysis, we chose two proteins from Talaromyces atroroseus (Acc. No. XP_020118381.1; TaGDC1) and Madurella mycetomatis (Acc. No. KXX81388.1; MmGDC1) with 56% and 52% identity compared to AGDC1. Coding sequences for both proteins and AGDC1 were cloned in codon-optimized form under control of the strong, methionine-repressible MET25 promoter (Mumberg, Müller and Funk 1994). Subsequently, PCA decarboxylase activities of both proteins were analyzed in comparison to AGDC1 in a feeding experiment with 5 mM PCA in strain CEN.PK2–1C (Fig. 6). Compared to AGDC1 expressing cells reaching a catechol titer of 460 mg L−1, both other proteins turned out to have superior activity, with catechol titers of 619 mg L−1 (TaGDC1) and 584 mg L−1 (MmGDC1) after 120 h cultivation in selective SCD medium lacking methionine. TaGDC1 was chosen for further experiments. Figure 6. View largeDownload slide Activities of gallic acid decarboxylases from different species. Expression vectors encoding eukaryotic enzymes from Arxula adeninivorans (p426-MET25-AGDC1), Madurella mycetomatis (p426-MET25-MmGDC1) and Talaromyces atroroseus (p426-MET25-TaGDC1) were transformed in strain CEN.PK2–1C. The empty vector was transformed as a control (Plasmid). All transformants were cultivated in shake flasks in selective synthetic complete (SCD) medium (2% glucose) lacking methionine and containing 5 mM protochatechuic acid (PCA) for 96 h at 30°C. Final catechol concentrations were measured after 96 h. Error bars represent the standard deviation of two independent replicates. Figure 6. View largeDownload slide Activities of gallic acid decarboxylases from different species. Expression vectors encoding eukaryotic enzymes from Arxula adeninivorans (p426-MET25-AGDC1), Madurella mycetomatis (p426-MET25-MmGDC1) and Talaromyces atroroseus (p426-MET25-TaGDC1) were transformed in strain CEN.PK2–1C. The empty vector was transformed as a control (Plasmid). All transformants were cultivated in shake flasks in selective synthetic complete (SCD) medium (2% glucose) lacking methionine and containing 5 mM protochatechuic acid (PCA) for 96 h at 30°C. Final catechol concentrations were measured after 96 h. Error bars represent the standard deviation of two independent replicates. To test the performance of the new decarboxylase in optimized chassis strain JTY4 (Table 1), the three enzymes of the heterologous CCM production pathway were combined on one plasmid under control of strong promoters. The production of CCM and the pathway intermediates 3-DHS, catechol and PCA was monitored in a fermentation starting with an OD600 of 10 in selective SCD medium with 4% glucose over a period of 120 h (Fig. 7). While CCM titers reached 1244 mg L−1, intermediates 3-DHS and PCA still accumulated to levels of 2141 and 3063 mg L−1, respectively. With 23 mg L−1, catechol titers were negligible. To our knowledge, the yield of 31 mg g−1 glucose represents the so far highest yield of CCM obtained in S. cerevisiae. Figure 7. View largeDownload slide cis, cis-Muconic acid production at high cells density in a batch fermentation. An expression plasmid encoding AroZ from Podospora anserina, GDC1 from Talaromyces atroroseus and CatA from Acinetobacter radioresistens (p423-H7-ArCatA, TaGDC1, PaAroZ) was transformed into yeast strain JTY4. Cells were inoculated to an OD600 of 10 from a preculture and cultivated in shake flasks for 120 h in selective synthetic complete (SCD) medium (4% glucose) at 30°C. Samples from culture supernatants were taken at the indicated time points, and concentrations of 3-dehydroshikimate (3-DHS), protocatechuic acid (PCA) and cis, cis-muconic acid (CCM) were determined by HPLC. Error bars represent the standard deviation of three independent replicates, but are masked by the size of the symbols. Figure 7. View largeDownload slide cis, cis-Muconic acid production at high cells density in a batch fermentation. An expression plasmid encoding AroZ from Podospora anserina, GDC1 from Talaromyces atroroseus and CatA from Acinetobacter radioresistens (p423-H7-ArCatA, TaGDC1, PaAroZ) was transformed into yeast strain JTY4. Cells were inoculated to an OD600 of 10 from a preculture and cultivated in shake flasks for 120 h in selective synthetic complete (SCD) medium (4% glucose) at 30°C. Samples from culture supernatants were taken at the indicated time points, and concentrations of 3-dehydroshikimate (3-DHS), protocatechuic acid (PCA) and cis, cis-muconic acid (CCM) were determined by HPLC. Error bars represent the standard deviation of three independent replicates, but are masked by the size of the symbols. GA production with the optimized chassis yeast strain In the next step, we wanted to test the suitability of the JTY4 chassis strain for production of other aromatic compounds derived from 3-DHS. For this aim, we chose to introduce a pathway for production of GA, a compound with known antioxidant, anti-inflammative, antimutagenic and anticancer properties (Badhani, Sharma and Kakkar 2015). A pathway for production of GA from 3-DHS via PCA was already established before in E. coli, making use of a mutant form of p-hydroxybenzoate hydroxylase PobA from Pseudomonas aeruginosa (Kambourakis, Draths and Frost 2000). A conserved tyrosine 385 was replaced by phenylalanine (Y385F), changing substrate specificity and allowing for the enzyme to hydroxylate PCA to GA (Entsch et al. 1991; Eschrich et al. 1993). The reaction requires O2 and reduction equivalents. We created an analogous Y385F mutant of PobA from C. glutamicum (Huang et al. 2008) with the ability to hydroxylate PCA to GA. NADPH is the preferred cofactor for the enzyme from C. glutamicum (Huang et al. 2008). For production of GA in S. cerevisiae, wild-type pobA and the pobAY385F mutant were expressed in strain JTY4 together with aroZ, and the formation of GA was determined after 120 h (Fig. 8A). As anticipated, PobAY385F produced much higher GA titers (298 mg L−1) than the wild-type enzyme (33 mg L−1). To boost the titer even more, we applied the conditions optimized for CCM production, as shown in Fig. 7 (Fig. 8B). With a starting OD600 of 10, strain JTY4 expressing aroZ and pobAY385F from individual plasmids produced 682 mg L−1 of GA. Despite the relatively high titer of GA obtained under these conditions, activity of PobAY385F is still limiting, as PCA accumulated to 6599 mg L−1 in the culture supernatant, probably due to NADPH limitations in the zwf1 deletion strain JTY4 (see discussion). Remarkably, the total amount of aromatic compounds analyzed (3-DHS, PCA and GA together) was higher than observed for production of CCM (3-DHS, PCA and CCM together) in Fig. 7. Here, it has to be considered that different promoters were used in the two heterologous pathways (Table 2), which might influence the flux through the upstream pathway. Differences in toxicities of the intermediates and products and in their secretion might be also responsible for this observation. Figure 8. View largeDownload slide Gallic acid (GA) production in strain JTY4. (A) Activity of a mutant version of p-hydroxybenzoic acid hydroxylase PobA. Yeast strain JTY4 was transformed with expression plasmids p423-H7-PaAroZ and p425-H7-CgPobA (AroZ + PobA) or p425-H7-CgPobAY385F (AroZ + PobAY385F), respectively. Empty plasmids were transformed as control (Plasmid). For production of GA, cells were inoculated to an OD600 of 0.8 from precultures in selective synthetic complete (SCD) medium (2% glucose), and the formation of GA was determined after cultivation in shake flasks for 120 h at 30°C by HPLC analysis of the culture supernatant. (B) Gallic acid production at high cell density in a batch fermentation. Yeast strain JTY4 was transformed with expression plasmids p423-H7-PaAroZ and p425-H7-CgPobAY385F. Cells were inoculated to an OD600 of 10 from a preculture and cultivated for 120 h at 30°C in shake flasks in selective SCD medium (4% glucose). Culture supernatants were analyzed for production of 3-dehydroshikimate (3-DHS), protocatechuic acid (PCA) and GA using HPLC. Error bars represent the standard deviation of three independent replicates. Figure 8. View largeDownload slide Gallic acid (GA) production in strain JTY4. (A) Activity of a mutant version of p-hydroxybenzoic acid hydroxylase PobA. Yeast strain JTY4 was transformed with expression plasmids p423-H7-PaAroZ and p425-H7-CgPobA (AroZ + PobA) or p425-H7-CgPobAY385F (AroZ + PobAY385F), respectively. Empty plasmids were transformed as control (Plasmid). For production of GA, cells were inoculated to an OD600 of 0.8 from precultures in selective synthetic complete (SCD) medium (2% glucose), and the formation of GA was determined after cultivation in shake flasks for 120 h at 30°C by HPLC analysis of the culture supernatant. (B) Gallic acid production at high cell density in a batch fermentation. Yeast strain JTY4 was transformed with expression plasmids p423-H7-PaAroZ and p425-H7-CgPobAY385F. Cells were inoculated to an OD600 of 10 from a preculture and cultivated for 120 h at 30°C in shake flasks in selective SCD medium (4% glucose). Culture supernatants were analyzed for production of 3-dehydroshikimate (3-DHS), protocatechuic acid (PCA) and GA using HPLC. Error bars represent the standard deviation of three independent replicates. DISCUSSION In the past years, several attempts have been made to optimize the production of CCM with yeast, which have led to considerable improvements in titers and yields (Curran et al. 2013; Skjoedt et al. 2016; Suastegui et al. 2016, 2017; Leavitt et al. 2017). Genetic manipulations and experimental conditions that have permitted these improvements are summarized in Table S1 (Supporting Information). Nevertheless, the decarboxylation step from PCA to catechol still remains the main bottleneck of the pathway. In our study, we have created a strain with an optimized production of the shikimic acid pathway intermediate 3-DHS (Fig. 4). In this strain, we have successfully utilized the recently identified eukaryotic PCA/GA decarboxylase AGDC1 (Meier et al. 2017) (Fig. 5) and two of its homologs from other organisms for production of CCM (Fig. 6). Using the same strain, we could demonstrate the production of GA from 3-DHS via PCA for the first time in S. cerevisiae (Fig. 8). In S. cerevisiae, the entrance reaction into the shikimic acid pathway is catalyzed by DAHP synthases Aro3 and Aro4. These highly homologous proteins are feedback regulated by aromatic amino acids phenylalanine and tyrosine, respectively (Braus 1991). With lower Km values for E4P and PEP and a higher Ki for its specific inhibitor (Hartmann et al. 2003), Aro3 can be regarded as the more effective enzyme. Nonetheless, so far only feedback-resistant mutants of Aro4 have been utilized in metabolic engineering applications (Curran et al. 2013; McKenna et al. 2014; Li et al. 2015; Rodriguez et al. 2015; Suastegui et al. 2016). We have introduced an analogous mutation to the K229L exchange in the effector binding cavity of Aro4 into Aro3 (Hartmann et al. 2003), generating Aro3K222L. Overexpressed together with Aro4K229L, the flux into the shikimic acid pathway could be considerably improved compared to the individual mutants (Fig. 2). Nevertheless, we cannot exclude an additional positive effect due to the simultaneous codon optimization of Aro3K222L. The next steps in the shikimic acid pathway are catalyzed by the multifunctional Aro1 protein. To block further conversion of 3-DHS to aromatic amino acids, the dehydrogenase domain E had been completely deleted in a previous study (Weber et al. 2012). Here, we could show (i) that in general overexpression of ARO1 mutant versions without dehydrogenase activity is beneficial for 3-DHS and PCA production but (ii) that even higher product titers could be achieved by implementing an Aro1D1370A mutant with a point mutation in the E-domain, a mutation that is presumably less detrimental for the overall structure and activity of Aro1 than a deletion of the whole E-domain (Fig. 3). The deletion of the ZWF1 gene encoding glucose-6-phosphate dehydrogenase ZWF1 in combination with an overexpression of TKL1 has already been described to lead to an enhanced production of CCM (Curran et al. 2013), an effect that was not observed consistently in a more recent study in a different strain background (Suastegui et al. 2017). In our study, an integration of an overexpression construct for TKL1 combined with the deletion of ZWF1 in a ΔaroE strain already harboring integrations of overexpression constructs for ARO3K222L, ARO4K229L and ARO1K1370A led to a 2.1-fold enhancement in the formation of 3-DHS (Fig. 4B), and after integration of the complete CCM pathway allowed for a final titer of 1244 mg L−1 of CCM (Fig. 7). Glucose-6-phosphate dehydrogenase is the major source of cytosolic NADPH, and a deletion of the corresponding gene has been shown to cause defects in NADPH-requiring processes, such as assimiliation of sulfur, leading to a requirement for an organic sulfur source, such as methionine or cysteine (Thomas, Cherest and Surdin-Kerjan 1991). If such a deficiency in NADPH supply would lead to an impairment in growth, it would make a strain unattractive in industrial applications. Growth of the Δzwf1 strain JTY4 was not impaired in YEPD medium. Nevertheless, upon cultivation in SCD medium lacking methionine growth was slightly impaired and CCM titers were reduced, although growth tests revealed that the strain was not completely auxotrophic for methionine (not shown). Moreover, in a previous report about shikimic acid production, which involves the NADPH-dependent shikimic acid dehydrogenase, a deletion of ZWF1 led to an enhanced accumulation of 3-DHS and a reduced level of the desired product shikimic acid, indicating a reduced activity of the NADPH-dependent shikimate dehydrogenase function of Aro1 (Suastegui et al. 2017). Thus, if NADPH demands are high, especially due to the implementation of NADPH requiring reactions, limitations in product titers can be expected. Other cytosolic sources for NADPH exist, which could compensate for the deficiency. It has been shown that the cytosolic aldehyde dehydrogenase Ald6 can partially compensate for the reduced NADPH/NADP ratio in yeast cells with a deletion in ZWF1 (Grabowska and Chelstowska 2003). Isocitrate dehydrogenase (Idp2) is another source for cytosolic NADPH when glucose is exhausted (Minard and McAlister-Henn 2005). In our study about GA production, we also used the zwf1 deletion strain JTY4. The heterologous enzyme CgPobA prefers NADPH over NADH as a cofactor (Huang et al. 2008). Although we could achieve GA titers up to 682 mg L−1 (Fig. 8), the intermediate PCA still accumulated to very high amounts, which might be indicative for NADPH limitations. Other bacterial homologous with a preference for NADH exist, which could be used alternatively for the conversion of PCA to GA (Fujii and Kaneda 1985; Jadan et al. 2001; Iwaki et al. 2005). In addition to the replacement of NADPH with NADH consuming reactions (Brochado et al. 2010), implementing transhydrogenases converting NADH to NADPH (Fiaux et al. 2003) or establishing NADPH generating reactions (Verho et al. 2002) might be valuable strategies to broaden the applicability of a Δzwf1 strain. Moreover, as also the catalytic activity of PobA might be limiting, mutant forms could be used, such as the combined mutations of T294A and Y385F. The T294A mutation, which is located in the catalytic pocket of the enzyme, has recently been described to improve the hydrogen-bond mediated interaction between PobAY385F and PCA, thereby enhancing the catalytic activity of P. aeruginosa PobAY385F (Chen et al. 2017). The eukaryotic decarboxylase AGDC1 has been described to be specific not only for PCA, but also for GA (Meier et al. 2017). Engineering the specificity of AGDC1 could allow for the extension of our newly implemented pathway, enabling the production of pyrogallol, a compound with applications in the pharmaceutical and chemical industry (Kambourakis, Draths and Frost 2000). For the production of CCM, the eukaryotic enzyme AGDC1 exhibits several advantages over its prokaryotic AroY analog. The decarboxylase function of AroY is performed by subunit C and requires the cofactor prenylated flavin mononucleotide (prFMN), which is synthesized by AroY-B. prFMN can be synthesized with the help of the endogenous yeast enzyme Pad1, but this enzyme is absent in some yeast strains, such as the strain CEN.PK2–1C used in this study (Weber et al. 2017). Even though AroY-Ciso from K. pneumoniae showed reduced oxygen sensitivity compared to other C subunits tested (Weber et al. 2012), the enzyme showed only full activity upon cultivation of the host strain under oxygen-limited conditions (Suastegui et al. 2016). In contrast to AroY (Weber et al. 2017), AGDC1 and its homologs show high activity in strain CEN.PK2–1C (Figs 5 and 6) without the requirement to express an additional cofactor generating enzyme, reducing the metabolic burden imposed on the production organism. Moreover, a fine-tuning of expression of the decarboxylase and the cofactor generating enzyme can be omitted. Despite all these advantages and the titer of 1244 mg L−1 CCM achieved with the new enzyme, the decarboxylase substrate PCA still accumulates to 3063 mg L−1, still indicating a major limitation. However, the requirement for only one enzyme will certainly ease multiple integrations of the enzyme or enzyme engineering, which will be needed to overcome the bottleneck. In conclusion, although just applying a few rational design strategies in combination with a novel enzyme has led to production of a significant yield of CCM, for an industrial process these must be still improved. Novel high-throughput systems utilizing biosensors (Skjoedt et al. 2016; Leavitt et al. 2017) in combination with adaptive laboratory evolution strategies have a high potential to identify new targets for further manipulations, which might finally allow for an industrial application of yeast for CCM and GA production. SUPPLEMENTARY DATA Supplementary data are available at FEMSYR online. Acknowledgements The authors thank Dr Christian Weber for generating expression plasmids for CgPobA. The authors are grateful to Dr Jun-yong Choe (Rosalind Franklin University, North Chicago, USA) for his helpful suggestions. FUNDING This work was supported by the German Federal Ministry of Education and Research following a decision of the German Bundestag [grants 031A542 and 031B0218 to EB]. Conflict of interest. None declared. REFERENCES Aguilar-Zárate P, Cruz MA, Montañez J et al.   Gallic acid production under anaerobic submerged fermentation by two bacilli strains. Microb Cell Fact  2015; 14: 209. Google Scholar CrossRef Search ADS PubMed  Badhani B, Sharma N, Kakkar R. Gallic acid: a versatile antioxidant with promising therapeutic and industrial applications. RSC Adv  2015; 5: 27540– 57. Google Scholar CrossRef Search ADS   Bajpai B, Patil S. A new approach to microbial production of gallic acid. Braz J Microbiol  2008; 39: 708– 11. Google Scholar CrossRef Search ADS PubMed  Boles E, Zimmermann FK. Saccharomyces cerevisiae phosphoglucose isomerase and fructose bisphosphate aldolase can be replaced functionally by the corresponding enzymes of Escherichia coli and Drosophila melanogaster. Curr Genet  1993; 23: 187– 91. Google Scholar CrossRef Search ADS PubMed  Boles E, Schulte F, Miosga T et al.   Characterization of a glucose-repressed pyruvate kinase (Pyk2p) in Saccharomyces cerevisiae that is catalytically insensitive to fructose-1,6-bisphosphate. J Bacteriol  1997; 179: 2987– 93. Google Scholar CrossRef Search ADS PubMed  Braus GH. Aromatic amino acid biosynthesis in the yeast Saccharomyces cerevisiae: a model system for the regulation of a eukaryotic biosynthetic pathway. Microbiol Rev  1991; 55: 349– 70. Google Scholar PubMed  Brochado AR, Matos C, Moller BL et al.   Improved vanillin production in baker's yeast through in silico design. Microb Cell Fact  2010; 9: 84. Google Scholar CrossRef Search ADS PubMed  Bruder S, Reifenrath M, Thomik T et al.   Parallelised online biomass monitoring in shake flasks enables efficient strain and carbon source dependent growth characterisation of Saccharomyces cerevisiae. Microb Cell Fact  2016; 15: 127. Google Scholar CrossRef Search ADS PubMed  Chen Z, Shen X, Wang J et al.   Rational engineering of p-hydroxybenzoate hydroxylase to enable efficient gallic acid synthesis via a novel artificial biosynthetic pathway. Biotechnol Bioeng  2017; 114: 2571– 80. Google Scholar CrossRef Search ADS PubMed  Curran KA, Leavitt JM, Karim AS et al.   Metabolic engineering of muconic acid production in Saccharomyces cerevisiae. Metab Eng  2013; 15: 55– 66. Google Scholar CrossRef Search ADS PubMed  Demeke MM, Dietz H, Li Y et al.   Development of a D-xylose fermenting and inhibitor tolerant industrial Saccharomyces cerevisiae strain with high performance in lignocellulose hydrolysates using metabolic and evolutionary engineering. Biotechnol Biofuels  2013; 6: 89. Google Scholar CrossRef Search ADS PubMed  Dower WJ, Miller JF, Ragsdale CW. High efficiency transformation of E. coli by high voltage electroporation. Nucleic Acids Res  1988; 16: 6127– 45. Google Scholar CrossRef Search ADS PubMed  Draths KM, Frost JW. Environmentally compatible synthesis of adipic acid from D-glucose. J Am Chem Soc  1994; 116: 399– 400. Google Scholar CrossRef Search ADS   Entsch B, Palfey BA, Ballou DP et al.   Catalytic function of tyrosine residues in para-hydroxybenzoate hydroxylase as determined by the study of site-directed mutants. J Biol Chem  1991; 266: 17341– 9. Google Scholar PubMed  Eschrich K, van der Bolt FJ, de Kok A et al.   Role of Tyr201 and Tyr385 in substrate activation by p-hydroxybenzoate hydroxylase from Pseudomonas fluorescens. Eur J Biochem  1993; 216: 137– 46. Google Scholar CrossRef Search ADS PubMed  Farwick A, Bruder S, Schadeweg V et al.   Engineering of yeast hexose transporters to transport D-xylose without inhibition by D-glucose. P Natl Acad Sci USA  2014; 111: 5159– 64. Google Scholar CrossRef Search ADS   Fiaux J, Cakar ZP, Sonderegger M et al.   Metabolic-flux profiling of the yeasts Saccharomyces cerevisiae and Pichia stipitis. Eukaryot Cell  2003; 2: 170– 80. Google Scholar CrossRef Search ADS PubMed  Fujii T, Kaneda T. Purification and properties of NADH/NADPH-dependent p-hydroxybenzoate hydroxylase from Corynebacterium cyclohexanicum. Eur J Biochem  1985; 147: 97– 104. Google Scholar CrossRef Search ADS PubMed  Generoso WC, Gottardi M, Oreb M et al.   Simplified CRISPR-Cas genome editing for Saccharomyces cerevisiae. J Microbiol Methods  2016; 127: 203– 5. Google Scholar CrossRef Search ADS PubMed  Gottardi M, Reifenrath M, Boles E et al.   Pathway engineering for the production of heterologous aromatic chemicals and their derivatives in Saccharomyces cerevisiae: bioconversion from glucose. FEMS Yeast Res  2017; 17: fox035. Grabowska D, Chelstowska A. The ALD6 gene product is indispensable for providing NADPH in yeast cells lacking glucose-6-phosphate dehydrogenase activity. J Biol Chem  2003; 278: 13984– 8. Güldener U, Heck S, Fielder T et al.   A new efficient gene disruption cassette for repeated use in budding yeast. Nucleic Acids Res  1996; 24: 2519– 24. Google Scholar CrossRef Search ADS PubMed  Hamacher T, Becker J, Gardonyi M et al.   Characterization of the xylose-transporting properties of yeast hexose transporters and their influence on xylose utilization. Microbiology  2002; 148: 2783– 8. Google Scholar CrossRef Search ADS PubMed  Hartmann M, Schneider TR, Pfeil A et al.   Evolution of feedback-inhibited beta /alpha barrel isoenzymes by gene duplication and a single mutation. P Natl Acad Sci USA  2003; 100: 862– 7. Google Scholar CrossRef Search ADS   Huang Y, Zhao KX, Shen XH et al.   Genetic and biochemical characterization of a 4-hydroxybenzoate hydroxylase from Corynebacterium glutamicum. Appl Microbiol Biot  2008; 78: 75– 83. Google Scholar CrossRef Search ADS   Iwaki H, Saji H, Abe K et al.   Cloning and sequence analysis of the 4-hydroxybenzoate 3-hydroxylase gene from a cyclohexanecarboxylate-degrading Gram-positive bacterium, “Corynebacterium cyclohexanicum” strain ATCC 51369. Microb Environ  2005; 20: 144– 50. Google Scholar CrossRef Search ADS   Jadan AP, van Berkel WJ, Golovleva LA et al.   Purification and properties of p-hydroxybenzoate hydroxylases from Rhodococcus strains. Biochemistry (Mosc)  2001; 66: 898– 903. Google Scholar CrossRef Search ADS PubMed  Kambourakis S, Draths KM, Frost JW. Synthesis of gallic acid and pyrogallol from glucose:  replacing natural product isolation with microbial catalysis. J Am Chem Soc  2000; 122: 9042– 3. Google Scholar CrossRef Search ADS   Leavitt JM, Wagner JM, Tu CC et al.   Biosensor-enabled directed evolution to improve muconic acid production in Saccharomyces cerevisiae. Biotechnol J  2017; 12: 1600687. Google Scholar CrossRef Search ADS   Li M, Kildegaard KR, Chen Y et al.   De novo production of resveratrol from glucose or ethanol by engineered Saccharomyces cerevisiae. Metab Eng  2015; 32: 1– 11. Google Scholar CrossRef Search ADS PubMed  Lindner HA, Nadeau G, Matte A et al.   Site-directed mutagenesis of the active site region in the quinate/shikimate 5-dehydrogenase YdiB of Escherichia coli. J Biol Chem  2005; 280: 7162– 9. Google Scholar CrossRef Search ADS PubMed  McKenna R, Thompson B, Pugh S et al.   Rational and combinatorial approaches to engineering styrene production by Saccharomyces cerevisiae. Microb Cell Fact  2014; 13: 123. Google Scholar CrossRef Search ADS PubMed  Meier AK, Worch S, Boer E et al.   Agdc1p - a gallic acid decarboxylase involved in the degradation of tannic acid in the yeast blastobotrys (Arxula) adeninivorans. Front Microbiol  2017; 8: 1777. Google Scholar CrossRef Search ADS PubMed  Minard KI, McAlister-Henn L. Sources of NADPH in yeast vary with carbon source. J Biol Chem  2005; 280: 39890– 6. Google Scholar CrossRef Search ADS PubMed  Mumberg D, Müller R, Funk M. Regulatable promoters of Saccharomyces cerevisiae: comparison of transcriptional activity and their use for heterologous expression. Nucleic Acids Res  1994; 22: 5767– 8. Google Scholar CrossRef Search ADS PubMed  Niu W, Draths KM, Frost JW. Benzene-free synthesis of adipic acid. Biotechnol Prog  2002; 18: 201– 11. Google Scholar CrossRef Search ADS PubMed  Payer SE, Marshall SA, Barland N et al.   Regioselective para-carboxylation of catechols with a prenylated flavin dependent decarboxylase. Angew Chem Int Ed  2017; 56: 13893– 7. Google Scholar CrossRef Search ADS   Perrin DD, Sillen LG. Stability Constants of Metal-Ion Complexes, Part B: Organic Ligands . Oxford; New York: Pergamon Press, 1979. Rodriguez A, Kildegaard KR, Li M et al.   Establishment of a yeast platform strain for production of p- coumaric acid through metabolic engineering of aromatic amino acid biosynthesis. Metab Eng  2015; 31: 181– 8. Google Scholar CrossRef Search ADS PubMed  Seth M, Chand S. Biosynthesis of tannase and hydrolysis of tannins to gallic acid by Aspergillus awamori — optimisation of process parameters - ScienceDirect. Process Biochem  2000; 36: 39– 44. Google Scholar CrossRef Search ADS   Skjoedt ML, Snoek T, Kildegaard KR et al.   Engineering prokaryotic transcriptional activators as metabolite biosensors in yeast. Nat Chem Biol  2016; 12: 951– 8. Google Scholar CrossRef Search ADS PubMed  Sonoki T, Morooka M, Sakamoto K et al.   Enhancement of protocatechuate decarboxylase activity for the effective production of muconate from lignin-related aromatic compounds. J Biotechnol  2014; 192: 71– 7. Google Scholar CrossRef Search ADS PubMed  Suastegui M, Shao Z. Yeast factories for the production of aromatic compounds: from building blocks to plant secondary metabolites. J Ind Microbiol Biot  2016; 43: 1611– 24. Google Scholar CrossRef Search ADS   Suastegui M, Yu Ng C, Chowdhury A et al.   Multilevel engineering of the upstream module of aromatic amino acid biosynthesis in Saccharomyces cerevisiae for high production of polymer and drug precursors. Metab Eng  2017; 42: 134– 44. Google Scholar CrossRef Search ADS PubMed  Suastegui M, Matthiesen JE, Carraher JM et al.   Combining metabolic engineering and electrocatalysis: application to the production of polyamides from sugar. Angew Chem Int Ed  2016; 55: 2368– 73. Google Scholar CrossRef Search ADS   Taxis C, Knop M. System of centromeric, episomal, and integrative vectors based on drug resistance markers for Saccharomyces cerevisiae. Biotechiques  2006; 40: 73– 78. Google Scholar CrossRef Search ADS   Thomas D, Cherest H, Surdin-Kerjan Y. Identification of the structural gene for glucose-6-phosphate dehydrogenase in yeast. Inactivation leads to a nutritional requirement for organic sulfur. EMBO J  1991; 10: 547– 53. Google Scholar PubMed  Tochigi Y, Sato N, Sahara T et al.   Sensitive and convenient yeast reporter assay for high-throughput analysis by using a secretory luciferase from Cypridina noctiluca. Anal Chem  2010; 82: 5768– 76. Google Scholar CrossRef Search ADS PubMed  Verho R, Richard P, Jonson PH et al.   Identification of the first fungal NADP-GAPDH from Kluyveromyces lactis. Biochemistry  2002; 41: 13833– 8. Google Scholar CrossRef Search ADS PubMed  Weber C, Brückner C, Weinreb S et al.   Biosynthesis of cis,cis-muconic acid and its aromatic precursors, catechol and protocatechuic acid, from renewable feedstocks by Saccharomyces cerevisiae. Appl Environ Microb  2012; 78: 8421– 30. Google Scholar CrossRef Search ADS   Weber HE, Gottardi M, Bruckner C et al.   Requirement of a functional flavin mononucleotide prenyltransferase for the activity of a bacterial decarboxylase in a heterologous muconic acid pathway in Saccharomyces cerevisiae. Appl Environ Microb  2017; 83: 03472– 16. Google Scholar CrossRef Search ADS   Wiedemann B, Boles E. Codon-optimized bacterial genes improve l-arabinose fermentation in recombinant Saccharomyces cerevisiae. Appl Environ Microb  2008; 74: 2043– 50. Google Scholar CrossRef Search ADS   Xiao W, Gietz RD, Woods RA. Yeast transformation by the LiAc/SS carrier DNA/PEG method. In: Walker JM (ed). Yeast Protocol, Vol. 313 , Totowa, NJ: Humana Press, 2006, 107– 20. Zhang H, Pereira B, Li Z et al.   Engineering Escherichia coli coculture systems for the production of biochemical products. P Natl Acad Sci USA  2015; 112: 8266– 71. Google Scholar CrossRef Search ADS   © FEMS 2018. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png FEMS Yeast Research Oxford University Press

An expanded enzyme toolbox for production of cis, cis-muconic acid and other shikimate pathway derivatives in Saccharomyces cerevisiae

Loading next page...
 
/lp/ou_press/an-expanded-enzyme-toolbox-for-production-of-cis-cis-muconic-acid-and-UJTjcRrfnd
Publisher
Blackwell
Copyright
© FEMS 2018. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com
ISSN
1567-1356
eISSN
1567-1364
D.O.I.
10.1093/femsyr/foy017
Publisher site
See Article on Publisher Site

Abstract

Abstract A wide range of commercially relevant aromatic chemicals can be synthesized via the shikimic acid pathway. Thus, this pathway has been the target of diverse metabolic engineering strategies. In the present work, an optimized yeast strain for production of the shikimic acid pathway intermediate 3-dehydroshikimate (3-DHS) was generated, which is a precursor for the production of the valuable compounds cis, cis-muconic acid (CCM) and gallic acid (GA). Production of CCM requires the overexpression of the heterologous enzymes 3-DHS dehydratase AroZ, protocatechuic acid (PCA) decarboxylase AroY and catechol dioxygenase CatA. The activity of AroY limits the yield of the pathway. This repertoire of enzymes was expanded by a novel fungal decarboxylase. Introducing this enzyme into the pathway in the optimized strain, a titer of 1244 mg L−1 CCM could be achieved, yielding 31 mg g−1 glucose. This represents the highest yield of this compound reported in Saccharomyces cerevisiae to date. To demonstrate the applicability of the optimized strain for production of other compounds from 3-DHS, we overexpressed AroZ together with a mutant of a para-hydroxybenzoic acid hydroxylase with improved substrate specificity for PCA, PobAY385F. Thereby, we could demonstrate the production of GA for the first time in S. cerevisiae. Saccharomyces cerevisiae, muconic acid, gallic acid, gallic acid decarboxylase, shikimic acid pathway, metabolic engineering, biotechnology INTRODUCTION Aromatic compounds and their derivatives have a wide range of applications, from pharmaceuticals to bulk and fine chemicals. Today, their production relies mainly on the use of fossil resources. As the supply of fossil resources will become sparse, an environmentally sustainable production from renewable feedstocks becomes more and more attractive. Thus, an increasing demand for biotechnological production processes by microbial fermentations has developed. Consequently, the shikimic acid pathway as a major source of aromatic compounds in microbial cells is an important target for metabolic engineering strategies (Suastegui and Sha 2016; Gottardi et al. 2017). cis, cis-muconic acid (CCM) is a precursor of the bulk chemical adipic acid, which is used for production of nylon, plastics, lubricants and softeners. Adipic acid is industrially synthesized from benzol by reduction to cyclohexane, followed by oxidation to cyclohexanone and cyclohexanol, in an energy-consuming process that leads to the release of greenhouse gasses, disadvantages which have stimulated the development of alternative microbial production processes. A route for production of CCM from shikimic acid pathway intermediate 3-dehydroshikimate (3-DHS) via protocatechuic acid (PCA) and catechol has been first established in Escherichia coli (Draths and Frost 1994). However, due to its acid tolerance and resistance to microbial contaminations, the yeast Saccharomyces cerevisiae would be particularly well suited as an industrial production host. The heterologous CCM production pathway first established in S. cerevisiae consists of a 3-DHS dehydratase (AroZ) from Podospora anserina; a PCA decarboxylase (AroY) consisting of subunits B, C and D from Klebsiella pneumoniae; and catechol-1, 2-dioxygenase (CatA) from Acinetobacter calcoaceticus (Weber et al. 2012) (Fig. 1). In this study, a drain of the intermediate 3-DHS was prevented by deletion of the E-domain in Aro1. Aro1 is a pentafunctional protein and catalyzes step 2 through 6 in the biosynthesis of chorismate, which is a precursor to aromatic amino acids. The E-domain encodes the dehydrogenase function required for conversion of 3-DHS into shikimate. The same effect could be achieved in a more recent study, introducing the point mutation D1409A in the E-domain, thus destroying its function (Suastegui et al. 2016). Introduction of a feedback resistant variant of 3-deoxy-D-arabino-heptulosonate-7-phosphate (DAHP) synthase Aro4 catalyzing the entrance reaction into the shikimic acid pathway, Aro4K229L, has allowed for the production of even higher CCM titers (Curran et al. 2013; Suastegui et al. 2016, 2017). Moreover, optimizing the availability of the precursor erythrose-4-phosphate (E4P) has been another strategy to enhance the flux into the shikimic acid pathway (Fig. 1). Deletion of glucose-6-phosphate dehydrogenase encoded by ZWF1 forces the entry of glycolytic intermediates into the non-oxidative pentose phosphate pathway, which in combination with overexpression of transketolase TKL1 diverts the flux from the glycolytic intermediates fructose-6-phosphate (F6P) and glyceraldehyde-3-phosphate (G3P) to E4P and xylulose-5-phosphate (Curran et al. 2013). These modifications have permitted the production of high titers of CCM (Curran et al. 2013; Suastegui et al. 2016, 2017; Leavitt et al. 2017). Despite these achievements, the decarboxylase reaction converting PCA to catechol remains a bottleneck of the established pathway. The activity of AroY could be improved by applying oxygen-limited conditions for culture growth (Suastegui et al. 2016). AroY isomeric subunit C (AroY-Ciso), which contains the decarboxylase function of the enzyme, requires the presence of the cofactor prenylated flavin mononucleotide (prFMN) for its activity (Payer et al. 2017; Weber et al. 2017). The biosynthesis of this cofactor from FMN is catalyzed by AroY subunit B or, more efficiently, by the endogenous yeast enzyme Pad1. Consequently, the activity of AroY-C could be enhanced by optimizing the cofactor supply (Leavitt et al. 2017; Weber et al. 2017). In this study, we have successfully utilized a novel class of fungal gallic acid (GA)/PCA decarboxylases for production of CCM in S. cerevisiae. We have developed a strain overexpressing feedback resistant variants of DAHP synthases, Aro3K222L and Aro4K229L, together with an Aro1 variant with a mutation in the E-domain, Aro1K1370A, leading to an enhanced flux into the shikimic acid pathway and allowing for the accumulation of the intermediate 3-DHS. Moreover, the ZWF1 gene was replaced by integration of an expression cassette containing TKL1 under control of the strong, constitutive TDH3 promoter. Expressing the novel decarboxylases in the optimized strain, we could achieve a titer of 1244 mg L−1 and a yield of 31 mg g−1 glucose in a batch fermentation, which represents the highest CCM yield in S. cerevisiae published to date (Table S1, Supporting Information). Figure 1. View largeDownload slide Overview of optimization strategies for the synthesis of aromatic compounds cis, cis-muconic acid (CCM) and gallic acid (GA) in S. cerevisiae. A deletion of the ZWF1 (Δzwf1) gene encoding glucose-6-phosphate dehydrogenase in combination with overexpression of transketolase Tkl1 abolishes entry of glucose-6-phosphate (G6P) into the oxidative part of the pentose phosphate pathway and forces entry of glycolytic intermediates fructose-6-phosphate (F6P) and glycerin-aldehyde-3-phosphate (GAP) into the non-oxidative part. Thereby, the formation of shikimic acid pathway precursor erythrose-4-phosphate (E4P) is improved. Overexpression of feedback resistant variants of 3-Deoxy-D-arabinoheptulosonate 7-phosphate (DAHP) synthases Aro3 and Aro4 with a lysine to leucine exchange at position 222 or 229, Aro3K222Land Aro4K229L, catalyzing the reaction of E4P and phosphoenolpyruvate (PEP) to DAHP, enhances the flux into the shikimic acid pathway. Deletion of the E-domain harboring the shikimate dehydrogenase function in the pentafunctional enzyme Aro1 (ΔaroE) and overexpression of mutants with a disrupted function of the E-domain (Aro1 K1370A, Aro1D1407A or Aro1ΔE) cause an accumulation of the shikimic acid intermediate 3-dehydroshikimate (3-DHS). 3-DHS is the precursor of heterologous muconic acid (CCM) and GA biosynthetic pathways. 3-DHS dehydratase AroZ catalyzes the first step of the heterologous pathways, the conversion of 3-DHS to protochatechuic acid (PCA). For production of CCM, a bacterial (AroY) or eukaryotic PCA decarboxylase is required to generate catechol, which is subsequently converted to CCM by catechol dioxygenase CatA. Hydroxylation of PCA to GA requires expression of PobAY385F, a para-hydroxy benzoic acid hydroxylase with a tyrosine to phenylalanine exchange causing a higher substrate affinity for PCA. Multiple enzymatic steps are shown as dashed lines. Figure 1. View largeDownload slide Overview of optimization strategies for the synthesis of aromatic compounds cis, cis-muconic acid (CCM) and gallic acid (GA) in S. cerevisiae. A deletion of the ZWF1 (Δzwf1) gene encoding glucose-6-phosphate dehydrogenase in combination with overexpression of transketolase Tkl1 abolishes entry of glucose-6-phosphate (G6P) into the oxidative part of the pentose phosphate pathway and forces entry of glycolytic intermediates fructose-6-phosphate (F6P) and glycerin-aldehyde-3-phosphate (GAP) into the non-oxidative part. Thereby, the formation of shikimic acid pathway precursor erythrose-4-phosphate (E4P) is improved. Overexpression of feedback resistant variants of 3-Deoxy-D-arabinoheptulosonate 7-phosphate (DAHP) synthases Aro3 and Aro4 with a lysine to leucine exchange at position 222 or 229, Aro3K222Land Aro4K229L, catalyzing the reaction of E4P and phosphoenolpyruvate (PEP) to DAHP, enhances the flux into the shikimic acid pathway. Deletion of the E-domain harboring the shikimate dehydrogenase function in the pentafunctional enzyme Aro1 (ΔaroE) and overexpression of mutants with a disrupted function of the E-domain (Aro1 K1370A, Aro1D1407A or Aro1ΔE) cause an accumulation of the shikimic acid intermediate 3-dehydroshikimate (3-DHS). 3-DHS is the precursor of heterologous muconic acid (CCM) and GA biosynthetic pathways. 3-DHS dehydratase AroZ catalyzes the first step of the heterologous pathways, the conversion of 3-DHS to protochatechuic acid (PCA). For production of CCM, a bacterial (AroY) or eukaryotic PCA decarboxylase is required to generate catechol, which is subsequently converted to CCM by catechol dioxygenase CatA. Hydroxylation of PCA to GA requires expression of PobAY385F, a para-hydroxy benzoic acid hydroxylase with a tyrosine to phenylalanine exchange causing a higher substrate affinity for PCA. Multiple enzymatic steps are shown as dashed lines. To expand the application of our newly developed strain, we decided to establish the production of GA, a phenolic compound naturally occurring in several plant species. GA has several applications in pharmaceutical and chemical industries and is mainly produced through acid hydrolysis of tannins or biotransformation procedures (Seth and Chand 2000; Bajpai and Patil 2008; Aguilar-Zárate et al. 2015). These procedures suffer from low yields and low purity and cause environmental pollution. A pathway for production of GA from 3-DHS has already been established in E. coli (Kambourakis, Draths and Frost 2000). Using P. anserina AroZ and a mutated version of para-hydroxybenzoic acid hydroxylase PobA from Corynebacterium glutamicum with a changed substrate specificity, we were able to demonstrate the production of GA for the first time in S. cerevisiae, reaching a titer of 682 mg L−1 from glucose. MATERIAL AND METHODS Strains and media Yeast strains used in this work are listed in Table 1. Saccharomyces cerevisiae strains were grown at 30°C with 180 rpm shaking in synthetic complete (SC) medium (6.7 g L−1 Difco yeast nitrogen base without amino acids), supplemented with amino acids as described previously (Bruder et al. 2016) and with 20 g L−1 D-glucose as a carbon source (SCD), adjusted to pH 6.3, or in YEPD medium (20 g L−1 peptone, 10 g L−1 yeast extract, 20 g L−1 glucose). For propagation of strains CEN.PK2–1C ΔaroE, JTY1 and JTY4 in SC medium, concentrations of aromatic amino acids tyrosine, phenylalanine and methionine were doubled. For maintenance of plasmids, selective SCD media lacked the auxotrophic markers or/and contained 200 mg L−1 G418 (geneticin) or 200 mg L−1 hygromycin, respectively. Table 1. Strains used in this study. Strain  Genotype  Reference  CEN.PK2–1C  MATa leu2–3112 ura3–52 trp1–289 his3-Δ1 MAL2–8c SUC2  EUROSCARF, Frankfurt  CEN.PK2–1CΔaroE  MATa leu2–3112 ura3–52 trp1–289 his3-Δ1 MAL2–8c SUC2 ARO1Δ1359–1588::loxP  Weber et al. (2012)  JTY1  CEN.PK2–1C ARO1Δ1359–1588::loxP pyk2Δ::TPI1p-aro4K229L;HXT7p-aro3K222L;FBA1p-aro1K1370A;kanMX  This work  JTY4  JTY1 zwf1Δ::TDH3p-TKL1  This work  Strain  Genotype  Reference  CEN.PK2–1C  MATa leu2–3112 ura3–52 trp1–289 his3-Δ1 MAL2–8c SUC2  EUROSCARF, Frankfurt  CEN.PK2–1CΔaroE  MATa leu2–3112 ura3–52 trp1–289 his3-Δ1 MAL2–8c SUC2 ARO1Δ1359–1588::loxP  Weber et al. (2012)  JTY1  CEN.PK2–1C ARO1Δ1359–1588::loxP pyk2Δ::TPI1p-aro4K229L;HXT7p-aro3K222L;FBA1p-aro1K1370A;kanMX  This work  JTY4  JTY1 zwf1Δ::TDH3p-TKL1  This work  View Large Metabolite analysis For metabolite analysis, yeast cells were removed from samples by centrifugation. Proteins were precipitated by addition of sulfosalicylic acid to a final concentration of 5%, and concentrations of metabolites were measured by high-performance liquid chromatography (HPLC). The metabolites were separated by HPLC chromatography (Dionex) using a Nucleogel Sugar 810 H column (Macherey-Nagel GmbH & Co, Germany). The column was eluted with 5 mM H2SO4 as mobile phase and a flow rate of 0.6 ml min−1 at the temperature of 65°C or 57°C. The detection of glucose was by means of a Shodex RI-101 refractive index detector. A UV detector was used for the detection of PCA (220 nm), catechol (220 nm), GA (220 nm) and CCM (250 nm). For data evaluation, the Chromeleon software (version 6.50) was used. Plasmid and strain construction Plasmids and primers used in this study are listed in Tables 2 and 3. Molecular techniques were performed according to published procedures (Wiedemann and Boles 2008). Codon-optimized gene versions for increased protein expression in S. cerevisiae were obtained from DNA2.0 (USA) or GeneArt (Germany). Yeast transformations and reisolation of plasmid DNA from yeast cells were carried out as described previously (Boles and Zimmermann 1993; Xiao, Gietz and Woods 2006). Genes were cloned by homologous recombination. The coding regions of the respective genes were amplified by PCR from genomic DNA or plasmids by using specific primer pairs with 5΄-extensions overlapping vector sequences. PCR fragments were co-transformed into yeast together with a linearized vector. Synthetic genes for construction of p426-MET25-TaGDC1 and p426-MET25-MmGDC1 were ordered with 30 base pairs overlap to the MET25 promoter and CYC1 terminator and directly cloned by homologous recombination. All vector encoded genes were controlled by strong promotors (Table 2). Plasmids were amplified in E. coli strain DH10b (Life Technologies). Escherichia coli transformations were performed via electroporation according to the methods of Dower, Miller and Ragsdale (1988). Escherichia coli was grown on LB (Luria-Bertani) medium with 40 μg ml−1 ampicillin for plasmid selection. Genomic integrations were constructed by using the cre-loxP-kanMX4-loxP system (Güldener et al. 1996) or using the CRISPR/Cas system (Generoso et al. 2016). The loxP-kanMX-loxP deletion/integration cassette was PCR amplified from the plasmid pHD8 by using primers binding in the region homologous to the integration site in the PYK2 locus present in the plasmid, allowing homologous recombination. The cassette was transformed into CEN.PK2–1C and selected for geneticin (G418) resistance. The deletion/integration was verified by PCR analysis. Table 2. Plasmids used in this study. Plasmid  Regulatory elements, markers, genes  Genbank accession no  Reference  p423-H7  2μ, HXT7p, CYC1t, HIS3  –  Hamacher et al. (2002)  p425-H7  2μ, HXT7p, CYC1t, LEU2  –  Hamacher et al. (2002)  p426-H7  2μ, HXT7p, CYC1t, URA3  –  Hamacher et al. (2002)  pRS42K  2μ, HXT7p, CYC1t, kanMX  –  Taxis and Knop (2006)  pRS42H  2μ, HXT7p, FBAt, HygromycinR  –  Taxis and Knop (2006)  pRS41H  CEN6_ARS4, TEFp, CYC1t, HygromycinR  –  Taxis and Knop (2006)  pRS62H  2μ, HXT7p, CYC1t, HygromycinR  –  Farwick et al. (2014)  pRS62N  2μ, HXT7p, CYC1t, ClonNatR  –  Farwick et al. (2014)  p425-MET25  2μ, MET25p, CYC1t, LEU2  –  Mumberg et al. (1994)  p426-MET25  2μ, MET25p, CYC1t, URA3  –  Mumberg et al. (1994)  p423-H7-PaAroZ  aroZ from Podospora anserina  CAD60599  Weber et al. (2012)  p426-MET25-AroY-B-C-D  MET25p; aroY subunit B; CYC1t; PGK1p; aroY subunit C isoform; PGK1t; TPI1p; aroY subunit D; TAL1t; all subunits from Klebsiella pneumoniae codon-optimized (DNA2.0 optimized)  AAY57854 AB479384 AAY57856  Weber et al. (2017)  p426 -AGDC1  PGK1p; agdc1 from Arxula adeninivorans, codon optimized (GeneArt); PGK1t  SJN60119.1  This work  p425-H7-ArCatA  catA from Acinetobacter radioresistens codon optimized (DNA2.0)  AF380158  Weber et al. (2012)  p425-H7-Aro3  ARO3 from Saccharomyces cerevisiae  YDR035W  This work  p425-H7-Aro3K222L  ARO3 from Saccharomyces cerevisiae with amino acid exchange K222L codon optimized (GeneArt)  YDR035W  This work  pRS42K-Aro4K229L  HXT7p, ARO4 from Saccharomyces cerevisiae with amino acid exchange K229L codon optimized (GeneArt), CYC1t  YBR249C  This work  pRS42H-Aro1ΔaroE  HXT7p; ARO1 from Saccharomyces cerevisiae lacking the aroE encoding domain (amino acids 1359–1588); FBA1t  YDR127W  Weber et al. (2012)  pRS42H-Aro1K1370A  2μ, HXT7p, ARO1 from Saccharomyces cerevisiae with amino acid exchange K1370A, FBA1t, HygromycinR  YDR127W  This work  pRS42H-Aro1D1407A  HXT7p, ARO1 from Saccharomyces cerevisiae with amino acid exchange D1407A, FBA1t  YDR127W  This work  p426-MET25-AGDC1  Arxula adeninivorans Gallat Decarboxylase codon optimized (GeneArt)  SJN60119.1  This work  p426-MET25-TaGDC1  Talaromyces atroroseus Gallat Decarboxylase codon optimized (JCat)  XP_020118381.1  This work  p426-MET25-MmGDC1  Madurella mycetomatis Gallat Decarboxylase codon optimized (JCat)  KXX81388.1  This work  pRCC-N  2μ, ROX3p-cas9-CYC1t; SNR52p-gRNA-SUP4t, natMX  –  Generoso et al. (2016)  pRCC-N zwf1  zwf1 gRNA  YNL241C  This work  pHD-Aro3K222L, Aro4K229L, Aro1K1370A  2μ, TPI1p‐ARO3K222L‐CYC1t, HXT7p‐ARO4K229L‐ARO4t, FBA1p-ARO1K1370A-RKI1t-KanMX  YDR035W YBR249C YDR127W  This work  p423-H7-ArCatA, TaGDC1, PaAroZ  HXT7p, catA from Acinetobacter radioresistens (codon optimized), RPE1t, ZEO1p, Talaromyces atroroseus Gallat Decarboxylase (codon optimized), TAL1t, TDH3p, aroZ from Podospora anserina, pPGK1t  AF380158 XP_020118381.1 CAD60599  This work  p425-H7-CgPobA  Corynebacterium glutamicum pobA  WP_011014104.1  This work  p427-H7-CgPobAY385F  Corynebacterium glutamicum pobA with amino acid exchange Y385F  WP_011014104.1  This work  Plasmid  Regulatory elements, markers, genes  Genbank accession no  Reference  p423-H7  2μ, HXT7p, CYC1t, HIS3  –  Hamacher et al. (2002)  p425-H7  2μ, HXT7p, CYC1t, LEU2  –  Hamacher et al. (2002)  p426-H7  2μ, HXT7p, CYC1t, URA3  –  Hamacher et al. (2002)  pRS42K  2μ, HXT7p, CYC1t, kanMX  –  Taxis and Knop (2006)  pRS42H  2μ, HXT7p, FBAt, HygromycinR  –  Taxis and Knop (2006)  pRS41H  CEN6_ARS4, TEFp, CYC1t, HygromycinR  –  Taxis and Knop (2006)  pRS62H  2μ, HXT7p, CYC1t, HygromycinR  –  Farwick et al. (2014)  pRS62N  2μ, HXT7p, CYC1t, ClonNatR  –  Farwick et al. (2014)  p425-MET25  2μ, MET25p, CYC1t, LEU2  –  Mumberg et al. (1994)  p426-MET25  2μ, MET25p, CYC1t, URA3  –  Mumberg et al. (1994)  p423-H7-PaAroZ  aroZ from Podospora anserina  CAD60599  Weber et al. (2012)  p426-MET25-AroY-B-C-D  MET25p; aroY subunit B; CYC1t; PGK1p; aroY subunit C isoform; PGK1t; TPI1p; aroY subunit D; TAL1t; all subunits from Klebsiella pneumoniae codon-optimized (DNA2.0 optimized)  AAY57854 AB479384 AAY57856  Weber et al. (2017)  p426 -AGDC1  PGK1p; agdc1 from Arxula adeninivorans, codon optimized (GeneArt); PGK1t  SJN60119.1  This work  p425-H7-ArCatA  catA from Acinetobacter radioresistens codon optimized (DNA2.0)  AF380158  Weber et al. (2012)  p425-H7-Aro3  ARO3 from Saccharomyces cerevisiae  YDR035W  This work  p425-H7-Aro3K222L  ARO3 from Saccharomyces cerevisiae with amino acid exchange K222L codon optimized (GeneArt)  YDR035W  This work  pRS42K-Aro4K229L  HXT7p, ARO4 from Saccharomyces cerevisiae with amino acid exchange K229L codon optimized (GeneArt), CYC1t  YBR249C  This work  pRS42H-Aro1ΔaroE  HXT7p; ARO1 from Saccharomyces cerevisiae lacking the aroE encoding domain (amino acids 1359–1588); FBA1t  YDR127W  Weber et al. (2012)  pRS42H-Aro1K1370A  2μ, HXT7p, ARO1 from Saccharomyces cerevisiae with amino acid exchange K1370A, FBA1t, HygromycinR  YDR127W  This work  pRS42H-Aro1D1407A  HXT7p, ARO1 from Saccharomyces cerevisiae with amino acid exchange D1407A, FBA1t  YDR127W  This work  p426-MET25-AGDC1  Arxula adeninivorans Gallat Decarboxylase codon optimized (GeneArt)  SJN60119.1  This work  p426-MET25-TaGDC1  Talaromyces atroroseus Gallat Decarboxylase codon optimized (JCat)  XP_020118381.1  This work  p426-MET25-MmGDC1  Madurella mycetomatis Gallat Decarboxylase codon optimized (JCat)  KXX81388.1  This work  pRCC-N  2μ, ROX3p-cas9-CYC1t; SNR52p-gRNA-SUP4t, natMX  –  Generoso et al. (2016)  pRCC-N zwf1  zwf1 gRNA  YNL241C  This work  pHD-Aro3K222L, Aro4K229L, Aro1K1370A  2μ, TPI1p‐ARO3K222L‐CYC1t, HXT7p‐ARO4K229L‐ARO4t, FBA1p-ARO1K1370A-RKI1t-KanMX  YDR035W YBR249C YDR127W  This work  p423-H7-ArCatA, TaGDC1, PaAroZ  HXT7p, catA from Acinetobacter radioresistens (codon optimized), RPE1t, ZEO1p, Talaromyces atroroseus Gallat Decarboxylase (codon optimized), TAL1t, TDH3p, aroZ from Podospora anserina, pPGK1t  AF380158 XP_020118381.1 CAD60599  This work  p425-H7-CgPobA  Corynebacterium glutamicum pobA  WP_011014104.1  This work  p427-H7-CgPobAY385F  Corynebacterium glutamicum pobA with amino acid exchange Y385F  WP_011014104.1  This work  If not stated otherwise, cloned genes are under control of promoters and terminators included in the basic vectors. Promoters were taken from 1–500 bps upstream and terminators 1–300 bps downstream of the respective open reading frames, with the exception of the HXT7 promoter (Hamacher et al. 2002), the MET25 promoter (Mumberg et al. 1994) and the ZEO1 promoter (Tochigi et al. 2010). p, promoter, t, terminator. View Large Table 3. Primers used in this study. Primer  5΄-3΄-Sequence  Target  JTP118-pPGK1-ovp426-fw  ATTAACCCTCACTAAAGGGAACAAAAGCTGGCTAGCCCCGGGGTCGACTGTTTGCAAAAA GAACAAAACTG  p426-MET25-AGDC1  CBp160 pPGK1_r  TGTTTTATATTTGTTGTAAAAAGTAGATAATTAC  See above  HWP12 GallatDCovpPGK1 for  TTATCTACTTTTTACAACAAATATAAAACAATGACCACCTCTTACGAACC  See above  HWP13 GallatDCovtPGK1rev  ATTGATCTATCGATTTCAATTCAATTCAATTTACCAGTGCAAGTCAATTTCCTTAC  See above  CBp95 tPGK1_f  ATTGAATTGAATTGAAATCGATAG  See above  CBp163 tPGK1_r  AAATAATATCCTTCTCGAAAGCTTTAACG  See above  CBp196 pHXT7_ARO3K222L_f  CAAAAAGTTTTTTTAATTTTAATCAAAAAGTTAACATGCATATGTTCATTAAGAACGAT CACGC  p425-H7-Aro3K222L  CBp197 tCYC1_ARO3K222L_r  GAGGGCGTGAATGTAAGCGTGACATAACTAATTACATGACTTACTTCTTCAAAGCCTTT CTTCTG  See above  CBp194 ARO3wt_f  CTGTCACCAAGCCAGGTGTCACTGC  p425-H7-Aro3  CBp195 ARO3wt_r  CACCTGGCTTGGTGACAGACAAGAAGTAG  See above  CBp175 pHXT7_pRS42K_f  GGCCCCCCCTCGAGGTCGACGGTATCGATAAGCTTGATATCGCTCGTAGGAACAATTTC GG  pRS42K-Aro4K229L  CBp176 pHXT7_ARO4K2229L_r  TTTTTGATTAAAATTAAAAAAACTTTTTG  See above  CBp149 pHXt7_Aro4K222L_neu_f  GAATAAACACAAAAACAAAAAGTTTTTTTAATTTTAATCAAAAAATGAGTGAATCTCCA ATGTTCG  See above  CBp116 Aro4_r_neu  CAACACCATGCAAAGTAACACCCATGAAATGGTGAGAATGAG  See above  CBp105 Aro4K229L_f_neu  CATGGGTGTTACTTTGCATGGTGTTG  See above  CBp107 Aro4K229L_tCYC1_neu_r  GTGGGGGGAGGGCGTGAATGTAAGCGTGACATAACTAATTCTATTTCTTGTTAACTTCT CTTCTTTG  See above  CBp177 tCYC1_ARO4K229L_f  AATTAGTTATGTCACGCTTACATTC  See above  CBp178 tCYC1_pRS42K_r  GCTGGAGCTCCACCGCGGTGGCGGCCGCTCTAGAACTAGTGGCGAATTGGGTACCG  See above  CBp184 pHXT7_ARO1_f  CACAAAAACAAAAAGTTTTTTTAATTTTAATCAAAAAATGGTGCAGTTAGCCAAAGTC  pRS42H-AroK1370A  CBp134 Aro1K1370A_r  GCATTATATCTAAAGCCAGAGGAATTGTG  See above  CBp135 Aro1K1370A_f  CAATTCCTCTGGCTTTAGATATAATGC  See above  CBp136 Aro1K1370A_tCYC1_r  GAGGGCGTGAATGTAAGCGTGACATAACTAATTACATGACCTACTCTTTCGTAACGGCA TC  See above  CBp179 pHXT7_Aro1_f  CTGGAGCTCCACCGCGGTGGCGGCCGCTCTAGAACTAGTGAGCTCGTAGGAACAATTTC GG  See above  CBp180 pHXT7_Aro1_r  GATAATATCATTTCCTAGAATTGGGACTTTGGCTAACTGCACCATTTTTTGATTAAAAT TAAAAAAACTTTTTGTTTTTGTG  See above  CBp181 tCYC1_Aro1_f  GCCCTTTCAAGGCCATTTTTGATGCCGTTACGAAAGAGTAGGTCATGTAATTAGTTATG TCACGC  See above  CBp182 tCYC1_Aro1_r  GGAAACCGACGCCCCAGCACTCGTCCGAGGGCAAAGGAATAAGGTACCGGCCGCAAATT AAAG  See above  CBp137 Aro1K1407A_r  CGGATACCTAACCAAGCGGTATTATCAC  pRS42H-Aro1D1407A  CBp138 Aro1K1407A_f  GGGTGATAATACCGCTTGGTTAGGTATCCG  See above  JTP31-GDC-ovMet25-fw  AATTCTATTACCCCCATCCATACTCTAGAAATGACCACCTCTTACGAAC  p426-MET25-AGDC1  JTP174-GDCovcyc1-rev  GGCGTGAATGTAAGCGTGACATAACTAATTACATGATTACCAGTGCAAGTCAATTTCC  See above  JTP35-TPI1p-ovMO1-fw  GGGCAGCTTGGTCTATACGTATACCCTTAGGCGGCCGCGGATCCATTTAAACTGTGAGG ACCTTAATAC  pHD-Aro3K222L, Aro4K229L, Aro1K1370A  MGP2-TPI1p-oAro4 revv  GTTGGCAGCGAACATTGGAGATTCACTCATTTTTAGTTTATGTATGTGTTTTTTGTAG  See above  MGP3-optAro4-oTPIp fw  TCTATAACTACAAAAAACACATACATAAACTAAAAATGAGTGAATCTCCAATGTT  See above  JTP36-Aro4t-ovhxt7p-rev  CGCTAGCGTGGCGGCCGCTCTAGAGGATCCAACTTATGTATGTTTCGATG  See above  JTP37-Hxt7p-ovAro4t-fw  GGATCCTCTAGAGCGGCCGCCACGCTAGCGGAGCTCGTAGGAACAATTTCGGG  See above  MGP109-pHXT7-rev  TTTTGATTAAAATTAAAAAAACTTTTTGT  See above  MGP25-ARO3opHXT7-rev  GATCGTTCTTAATGAACATTTTTTGATTAAAATTAAAAAAAC  See above  MGP24- ARO3opHXT7- fw  AAGTTTTTTTAATTTTAATCAAAAAATGTTCATTAAGAACGATC  See above  JTP38-Aro3-ovcyc1t-rev  TTACATGACTCGAGTTACTTCTTCAAAGCCTTTC  See above  JTP39-cyc1t-ovAro3-fw  GGCTTTGAAGAAGTAACTCGAGTCATGTAATTAGTTATGTC  See above  JTP40-cyc1p-ovFBA1p-rev  GTCGACGCTAGCGGTACCGGCCGCAAATTAAAGCCTTCGAG  See above  JTP41-FBA1p-ovcyc1t-fw  TTTAATTTGCGGCCGGTACCGCTAGCGTCGACTGGGTCATTACGTAAATAATG  See above  JTP42-FBA1p-ovAro1-rev  GACTTTGGCTAACTGCACCATTTTGAATATGTATTACTTGG  See above  JTP43-Aro1-ovFBA1p-fw  ATATTCAAAATGGTGCAGTTAGCCAAAGTC  See above  JTP44-Aro1-ovRKIt-rev  TGCCTTTGATCTGCCTACTCTTTCGTAACGGCATC  See above  JTP45-RKIt-ovAro1-fw  CGTTACGAAAGAGTAGGCAGATCAAAGGCAAAGACAG  See above  JTP46-RKI1t-ovMO1-rev  CATACATTATACGAAGTTATATTAAGGGTTGTCGACCTTGGTGTGTCATCGGTAGTAAC  See above  sbp324-Clon-pCrCas-K/N-Zwf1-for  AGATTCAGATCTGTGACTGTTTTAGAGCTAGAAATAGCAAGTTAAAATAAGG  pRCC-N zwf1  sbp310-Clon-pCrCas-K/N-Zwf1-rev  GTCACAGATCTGAATCTAAGATCATTTATCTTTCACTGCGGAG  See above  JTP221-pTDH3ovZWF1-Donor-fw  TTCAGTGACTTAGCCGATAAATGAATGTGCTTGCATTTTTACAGTTTATTCCTGGCAT CC  Integration of TDH3p-TKL1-TKL1t  JTP222-TKL1tovZWF1-Donor-rev  ACATAACACCTAAAGTGGCTTCCTCCTGCCCCTCTCTCCCATATTCTTTATTGCTTTA TACTTGAATGG  See above  JTP249-CatArovHxt7p-fw  ACAAAAAGTTTTTTTAATTTTAATCAAAAAATGACTGCAGCAAATGTTAAG  p423-H7-ArCatA, TaGDC1, PaAroZ  CBp76 catA radio-R  TCGTATAGTATAGAGAGTATAAATATAAGAAATGCCGCATATGTACAATTAGGCTTGC AATCTTGGTCTATCC  See above  CBp80 rRPE1-F  TTGTACATATGCGGCATTTCTTATA  See above  JTP250-tRPE1-ovZEO1-rev  ACAATTTGTGGTCTTCGTCGTGCTCTGTGGGGATCCAAATGGATATTGATCTAGATGG CG  See above  JTP156-ZEO1p-fw  CCACAGAGCACGACGAAG  See above  JTP158-pZEO1-rev  ATTGATATAAACGTAGTTTTGTATGTTTC  See above  JTP251-TaGDCovZEO1p-fw  AGAAACATACAAAACTACGTTTATATCAATATGGTTTACAACGTTACTGAAAAG  See above  JTP252-TaGDCovTAL1t-rev  GGCCTAAATTAATATTTCCGAGATACTTCCCTACTTTGGAACTTCCAACAAC  See above  CBp214 tTAL1_f  GGAAGTATCTCGGAAATATTAATTTAGGC  See above  CBp158 tTAL_r  GACGTTGATTTAAGGTGGTTCCG  See above  JTP253-pTDH3ovTAL1t-fw  AAACATCCGGAACCACCTTAAATCAACGTCGTCGACACAGTTTATTCCTGGCATCC  See above  CBp222 pTDH3_r  TTTGTTTGTTTATGTGTGTTTATTCG  See above  JTP254-AroZovTDH3-fw  GTTTCGAATAAACACACATAAACAAACAAAATGCCAAGTAAACTGGCTATTAC  See above  JTP255-tPGK1ovAroZ-rev  ATTGATCTATCGATTTCAATTCAATTCAATTTAAAGTGCGGCAGATAGAG  See above  CBp9tPGK_f  ATTGAATTGAATTGAAATCGATAGAT  See above  JTP256-tPGK1ovp423-r  CCAGTGAGCGCGCGTAATACGACTCACTATAGCGGCCGCAAATAATATCCTTCTCGAA AGCTTTAAC  See above  pobA_Coryne_F  AACACAAAAACAAAAAGTTTTTTTAATTTTAATCAAAAAATGAACCACGTACCAGTGG CAATT  P425-H7-CgPobA  pobA_Coryne_wt_R  GAATGTAAGCGTGACATAACTAATTACATGACTCGAGTTATACCTCGAAGCGTGGTAG  See above  pobA_Coryne_R  GAATGTAAGCGTGACATAACTAATTACATGACTCGAGTTATACCTCGAAGCGTGGTAG GTCGCGCCCAACGAACTGCTC  P425-H7-CgPobAY385F  Primer  5΄-3΄-Sequence  Target  JTP118-pPGK1-ovp426-fw  ATTAACCCTCACTAAAGGGAACAAAAGCTGGCTAGCCCCGGGGTCGACTGTTTGCAAAAA GAACAAAACTG  p426-MET25-AGDC1  CBp160 pPGK1_r  TGTTTTATATTTGTTGTAAAAAGTAGATAATTAC  See above  HWP12 GallatDCovpPGK1 for  TTATCTACTTTTTACAACAAATATAAAACAATGACCACCTCTTACGAACC  See above  HWP13 GallatDCovtPGK1rev  ATTGATCTATCGATTTCAATTCAATTCAATTTACCAGTGCAAGTCAATTTCCTTAC  See above  CBp95 tPGK1_f  ATTGAATTGAATTGAAATCGATAG  See above  CBp163 tPGK1_r  AAATAATATCCTTCTCGAAAGCTTTAACG  See above  CBp196 pHXT7_ARO3K222L_f  CAAAAAGTTTTTTTAATTTTAATCAAAAAGTTAACATGCATATGTTCATTAAGAACGAT CACGC  p425-H7-Aro3K222L  CBp197 tCYC1_ARO3K222L_r  GAGGGCGTGAATGTAAGCGTGACATAACTAATTACATGACTTACTTCTTCAAAGCCTTT CTTCTG  See above  CBp194 ARO3wt_f  CTGTCACCAAGCCAGGTGTCACTGC  p425-H7-Aro3  CBp195 ARO3wt_r  CACCTGGCTTGGTGACAGACAAGAAGTAG  See above  CBp175 pHXT7_pRS42K_f  GGCCCCCCCTCGAGGTCGACGGTATCGATAAGCTTGATATCGCTCGTAGGAACAATTTC GG  pRS42K-Aro4K229L  CBp176 pHXT7_ARO4K2229L_r  TTTTTGATTAAAATTAAAAAAACTTTTTG  See above  CBp149 pHXt7_Aro4K222L_neu_f  GAATAAACACAAAAACAAAAAGTTTTTTTAATTTTAATCAAAAAATGAGTGAATCTCCA ATGTTCG  See above  CBp116 Aro4_r_neu  CAACACCATGCAAAGTAACACCCATGAAATGGTGAGAATGAG  See above  CBp105 Aro4K229L_f_neu  CATGGGTGTTACTTTGCATGGTGTTG  See above  CBp107 Aro4K229L_tCYC1_neu_r  GTGGGGGGAGGGCGTGAATGTAAGCGTGACATAACTAATTCTATTTCTTGTTAACTTCT CTTCTTTG  See above  CBp177 tCYC1_ARO4K229L_f  AATTAGTTATGTCACGCTTACATTC  See above  CBp178 tCYC1_pRS42K_r  GCTGGAGCTCCACCGCGGTGGCGGCCGCTCTAGAACTAGTGGCGAATTGGGTACCG  See above  CBp184 pHXT7_ARO1_f  CACAAAAACAAAAAGTTTTTTTAATTTTAATCAAAAAATGGTGCAGTTAGCCAAAGTC  pRS42H-AroK1370A  CBp134 Aro1K1370A_r  GCATTATATCTAAAGCCAGAGGAATTGTG  See above  CBp135 Aro1K1370A_f  CAATTCCTCTGGCTTTAGATATAATGC  See above  CBp136 Aro1K1370A_tCYC1_r  GAGGGCGTGAATGTAAGCGTGACATAACTAATTACATGACCTACTCTTTCGTAACGGCA TC  See above  CBp179 pHXT7_Aro1_f  CTGGAGCTCCACCGCGGTGGCGGCCGCTCTAGAACTAGTGAGCTCGTAGGAACAATTTC GG  See above  CBp180 pHXT7_Aro1_r  GATAATATCATTTCCTAGAATTGGGACTTTGGCTAACTGCACCATTTTTTGATTAAAAT TAAAAAAACTTTTTGTTTTTGTG  See above  CBp181 tCYC1_Aro1_f  GCCCTTTCAAGGCCATTTTTGATGCCGTTACGAAAGAGTAGGTCATGTAATTAGTTATG TCACGC  See above  CBp182 tCYC1_Aro1_r  GGAAACCGACGCCCCAGCACTCGTCCGAGGGCAAAGGAATAAGGTACCGGCCGCAAATT AAAG  See above  CBp137 Aro1K1407A_r  CGGATACCTAACCAAGCGGTATTATCAC  pRS42H-Aro1D1407A  CBp138 Aro1K1407A_f  GGGTGATAATACCGCTTGGTTAGGTATCCG  See above  JTP31-GDC-ovMet25-fw  AATTCTATTACCCCCATCCATACTCTAGAAATGACCACCTCTTACGAAC  p426-MET25-AGDC1  JTP174-GDCovcyc1-rev  GGCGTGAATGTAAGCGTGACATAACTAATTACATGATTACCAGTGCAAGTCAATTTCC  See above  JTP35-TPI1p-ovMO1-fw  GGGCAGCTTGGTCTATACGTATACCCTTAGGCGGCCGCGGATCCATTTAAACTGTGAGG ACCTTAATAC  pHD-Aro3K222L, Aro4K229L, Aro1K1370A  MGP2-TPI1p-oAro4 revv  GTTGGCAGCGAACATTGGAGATTCACTCATTTTTAGTTTATGTATGTGTTTTTTGTAG  See above  MGP3-optAro4-oTPIp fw  TCTATAACTACAAAAAACACATACATAAACTAAAAATGAGTGAATCTCCAATGTT  See above  JTP36-Aro4t-ovhxt7p-rev  CGCTAGCGTGGCGGCCGCTCTAGAGGATCCAACTTATGTATGTTTCGATG  See above  JTP37-Hxt7p-ovAro4t-fw  GGATCCTCTAGAGCGGCCGCCACGCTAGCGGAGCTCGTAGGAACAATTTCGGG  See above  MGP109-pHXT7-rev  TTTTGATTAAAATTAAAAAAACTTTTTGT  See above  MGP25-ARO3opHXT7-rev  GATCGTTCTTAATGAACATTTTTTGATTAAAATTAAAAAAAC  See above  MGP24- ARO3opHXT7- fw  AAGTTTTTTTAATTTTAATCAAAAAATGTTCATTAAGAACGATC  See above  JTP38-Aro3-ovcyc1t-rev  TTACATGACTCGAGTTACTTCTTCAAAGCCTTTC  See above  JTP39-cyc1t-ovAro3-fw  GGCTTTGAAGAAGTAACTCGAGTCATGTAATTAGTTATGTC  See above  JTP40-cyc1p-ovFBA1p-rev  GTCGACGCTAGCGGTACCGGCCGCAAATTAAAGCCTTCGAG  See above  JTP41-FBA1p-ovcyc1t-fw  TTTAATTTGCGGCCGGTACCGCTAGCGTCGACTGGGTCATTACGTAAATAATG  See above  JTP42-FBA1p-ovAro1-rev  GACTTTGGCTAACTGCACCATTTTGAATATGTATTACTTGG  See above  JTP43-Aro1-ovFBA1p-fw  ATATTCAAAATGGTGCAGTTAGCCAAAGTC  See above  JTP44-Aro1-ovRKIt-rev  TGCCTTTGATCTGCCTACTCTTTCGTAACGGCATC  See above  JTP45-RKIt-ovAro1-fw  CGTTACGAAAGAGTAGGCAGATCAAAGGCAAAGACAG  See above  JTP46-RKI1t-ovMO1-rev  CATACATTATACGAAGTTATATTAAGGGTTGTCGACCTTGGTGTGTCATCGGTAGTAAC  See above  sbp324-Clon-pCrCas-K/N-Zwf1-for  AGATTCAGATCTGTGACTGTTTTAGAGCTAGAAATAGCAAGTTAAAATAAGG  pRCC-N zwf1  sbp310-Clon-pCrCas-K/N-Zwf1-rev  GTCACAGATCTGAATCTAAGATCATTTATCTTTCACTGCGGAG  See above  JTP221-pTDH3ovZWF1-Donor-fw  TTCAGTGACTTAGCCGATAAATGAATGTGCTTGCATTTTTACAGTTTATTCCTGGCAT CC  Integration of TDH3p-TKL1-TKL1t  JTP222-TKL1tovZWF1-Donor-rev  ACATAACACCTAAAGTGGCTTCCTCCTGCCCCTCTCTCCCATATTCTTTATTGCTTTA TACTTGAATGG  See above  JTP249-CatArovHxt7p-fw  ACAAAAAGTTTTTTTAATTTTAATCAAAAAATGACTGCAGCAAATGTTAAG  p423-H7-ArCatA, TaGDC1, PaAroZ  CBp76 catA radio-R  TCGTATAGTATAGAGAGTATAAATATAAGAAATGCCGCATATGTACAATTAGGCTTGC AATCTTGGTCTATCC  See above  CBp80 rRPE1-F  TTGTACATATGCGGCATTTCTTATA  See above  JTP250-tRPE1-ovZEO1-rev  ACAATTTGTGGTCTTCGTCGTGCTCTGTGGGGATCCAAATGGATATTGATCTAGATGG CG  See above  JTP156-ZEO1p-fw  CCACAGAGCACGACGAAG  See above  JTP158-pZEO1-rev  ATTGATATAAACGTAGTTTTGTATGTTTC  See above  JTP251-TaGDCovZEO1p-fw  AGAAACATACAAAACTACGTTTATATCAATATGGTTTACAACGTTACTGAAAAG  See above  JTP252-TaGDCovTAL1t-rev  GGCCTAAATTAATATTTCCGAGATACTTCCCTACTTTGGAACTTCCAACAAC  See above  CBp214 tTAL1_f  GGAAGTATCTCGGAAATATTAATTTAGGC  See above  CBp158 tTAL_r  GACGTTGATTTAAGGTGGTTCCG  See above  JTP253-pTDH3ovTAL1t-fw  AAACATCCGGAACCACCTTAAATCAACGTCGTCGACACAGTTTATTCCTGGCATCC  See above  CBp222 pTDH3_r  TTTGTTTGTTTATGTGTGTTTATTCG  See above  JTP254-AroZovTDH3-fw  GTTTCGAATAAACACACATAAACAAACAAAATGCCAAGTAAACTGGCTATTAC  See above  JTP255-tPGK1ovAroZ-rev  ATTGATCTATCGATTTCAATTCAATTCAATTTAAAGTGCGGCAGATAGAG  See above  CBp9tPGK_f  ATTGAATTGAATTGAAATCGATAGAT  See above  JTP256-tPGK1ovp423-r  CCAGTGAGCGCGCGTAATACGACTCACTATAGCGGCCGCAAATAATATCCTTCTCGAA AGCTTTAAC  See above  pobA_Coryne_F  AACACAAAAACAAAAAGTTTTTTTAATTTTAATCAAAAAATGAACCACGTACCAGTGG CAATT  P425-H7-CgPobA  pobA_Coryne_wt_R  GAATGTAAGCGTGACATAACTAATTACATGACTCGAGTTATACCTCGAAGCGTGGTAG  See above  pobA_Coryne_R  GAATGTAAGCGTGACATAACTAATTACATGACTCGAGTTATACCTCGAAGCGTGGTAG GTCGCGCCCAACGAACTGCTC  P425-H7-CgPobAY385F  View Large Fermentations and feeding experiments Cultures of yeast strains (50 ml) were grown in 300 ml shake flasks (Erlenmeyer flasks) at 30°C in a rotary shaker (180 rpm). Precultures were grown into the exponential phase in selective SCD media. Cells were washed with sterile water and inoculated to an optical density at 600 nm (OD600) of 0.8 in the same medium. For PCA feeding experiments, cells were inoculated to an OD600 of 0.8 in selective SCD medium containing 5 mM PCA. SCD medium with 5 mM PCA was prepared as follows. PCA was dissolved in distilled water to a concentration of 100 mM and sterile filtered. The stock solution was added to non-buffered, selective SCD medium to a concentration of 5 mM, resulting in a pH of 4.3. Growth/fermentation experiments were performed up to three times by using the same precultures, with the given standard deviations. Samples for the measurement of the OD600 and the metabolite analyses were taken at different time points. RESULTS Improving the flux into the shikimic acid pathway In order to enhance CCM production, it is important to enhance the substrate flux into the shikimic acid pathway. The two isomeric DAHP synthases Aro3 and Aro4, which catalyze the entrance reaction into the shikimic acid pathway, are known to be feedback inhibited by products of the shikimic acid pathway, tyrosine and phenylalanine, respectively (Braus 1991). Therefore, we chose to overexpress variants with abolished negative regulation in addition to the natural pathway (Fig. 2). For Aro4, the mutation K229L is known to lack feedback sensitivity (Hartmann et al. 2003) and has been successfully used in several studies (Curran et al. 2013; McKenna et al. 2014; Li et al. 2015; Rodriguez et al. 2015; Suastegui et al. 2016). As Aro3 was reported to have a higher affinity for E4P and phosphoenolpyruvate (PEP) compared to Aro4 (Hartmann et al. 2003), we also created a mutated Aro3 with an analogous K222L point mutation. For comparison, the mutated enzymes or the wild type Aro3 were overexpressed under control of the strong constitutive HXT7 promoter (Hamacher et al. 2002) together with Aro1 lacking the E-domain responsible for conversion of 3-DHS to shikimate (Aro1ΔE) in strain CEN.PK2–1C ΔaroE harboring a genomic deletion of the E-domain (Weber et al. 2012). This deletion is crucial for the accumulation of the precursor 3-DHS. Additionally, 3-DHS dehydratase AroZ was overexpressed and formation of PCA was measured (Fig. 2). In fermentations in selective SC medium, the overexpression of ARO4K229L resulted in 250 mg L−1, of wild-type ARO3 in a codon-optimized form in 304 mg L−1, and of the codon-optimized mutant ARO3K222L in 469 mg L−1 after 96 h. Combined overexpression of ARO3 K222L and ARO4K229L led to the highest PCA titer with 653 mg L−1, which is a 3.3-fold increase compared to the empty plasmid control with 197 mg L−1. Figure 2. View largeDownload slide Activity of feedback resistant DAHP synthase mutants. CEN.PK2–1C ΔaroE cells were transformed with expression plasmids p425-H7-Aro3 (Aro3 (wt)) or p425-H7-Aro3K222L (Aro3K222L) and pRS42K-Aro4K229L (Aro4K229L), or with a combination of these plasmids. Empty plasmids were transformed as controls (Plasmid). To analyze the activity of the enzymes, pRS42H-Aro1ΔE (Aro1ΔE) and p423-H7-PaAroZ (AroZ) were co-transformed, and the final protocatechuic acid (PCA) concentrations after cultivation for 96 h in shake flasks in selective synthetic complete (SCD) medium (2% glucose) at 30°C were analyzed. Error bars represent the standard deviation of two independent replicates. Figure 2. View largeDownload slide Activity of feedback resistant DAHP synthase mutants. CEN.PK2–1C ΔaroE cells were transformed with expression plasmids p425-H7-Aro3 (Aro3 (wt)) or p425-H7-Aro3K222L (Aro3K222L) and pRS42K-Aro4K229L (Aro4K229L), or with a combination of these plasmids. Empty plasmids were transformed as controls (Plasmid). To analyze the activity of the enzymes, pRS42H-Aro1ΔE (Aro1ΔE) and p423-H7-PaAroZ (AroZ) were co-transformed, and the final protocatechuic acid (PCA) concentrations after cultivation for 96 h in shake flasks in selective synthetic complete (SCD) medium (2% glucose) at 30°C were analyzed. Error bars represent the standard deviation of two independent replicates. Mutation and overexpression of Aro1 In our previous study, the whole E-domain of the pentafunctional enzyme Aro1 was deleted in order to prevent conversion of 3-DHS to shikimate, thereby improving the availability of the CCM pathway substrate 3-DHS (Weber et al. 2012). Concerns about possible negative effects on protein folding and enzyme activities prompted us to search for point mutations inside the E-domain, which should also lead to a specific loss of dehydrogenase function. Two invariant catalytic residues in the active site of the AroE homolog YdiB from E. coli, K71 and D107, had previously been identified as crucial for substrate binding (Lindner et al. 2005). In the E-domain of S. cerevisiae Aro1, these residues are located at position 1370 and 1407, respectively. The K1370A mutation and a D1409A mutation have already been described in a previous study (Suastegui et al. 2016). To test if the single amino acid exchanges cause a complete loss of the shikimate dehydrogenase function of Aro1, ARO1K1370A and ARO1D1407A expressed under control of the strong constitutive HXT7 promoter were tested for their ability to complement the aromatic amino acid auxotrophy of the CEN.PK2–1C ΔaroE strain. For this aim, a growth test was performed on media with or without the aromatic amino acids tyrosine and phenylalanine (Fig. 3A). CEN.PK2–1C ΔaroE and CEN.PK2–1C transformed with empty vector were used as controls. Both mutated ARO1 alleles were not able to complement the growth defect on media without tyrosine and phenylalanine, indicating a loss of shikimate dehydrogenase activity and making them suitable candidates for the CCM pathway. Figure 3. View largeDownload slide (A) Growth test with Aro1 point mutations. CEN.PK2–1C ΔaroE cells transformed with expression vectors pRS42H-Aro1K1370A (Aro1K1370A) or pRS42H-Aro1D1407A (Aro1D1307A) were cultivated in selective synthetic complete medium (SCD) with 2% glucose, harvested at OD600 = 1 and spotted onto selective SCD agar (2% glucose) with or without the aromatic amino acids tyrosine (Tyr) and phenylalanine (Phe) in several dilutions. CEN.PK2–1C ΔaroE and CEN.PK2–1C cells transformed with empty plasmid (Plasmid) were used as controls. The picture shows growth after 4 days at 30°C. (B) Activity of Aro1 variants. CEN.PK2–1C ΔaroE cells were transformed with pRS42H-Aro1ΔE (Aro1ΔE), pRS42H-Aro1K1370A (Aro1K1370A) or pRS42H-Aro1D1407A (Aro1D1407A) in combination with p425-H7-Aro3K222L (Aro3K222L) and p423-H7-PaAroZ (AroZ) and subsequently cultivated in selective SCD medium (2% glucose) in shake flasks. The production of PCA was determined after 96 h at 30°C. The error bars show the standard deviation of two independent replicates. Figure 3. View largeDownload slide (A) Growth test with Aro1 point mutations. CEN.PK2–1C ΔaroE cells transformed with expression vectors pRS42H-Aro1K1370A (Aro1K1370A) or pRS42H-Aro1D1407A (Aro1D1307A) were cultivated in selective synthetic complete medium (SCD) with 2% glucose, harvested at OD600 = 1 and spotted onto selective SCD agar (2% glucose) with or without the aromatic amino acids tyrosine (Tyr) and phenylalanine (Phe) in several dilutions. CEN.PK2–1C ΔaroE and CEN.PK2–1C cells transformed with empty plasmid (Plasmid) were used as controls. The picture shows growth after 4 days at 30°C. (B) Activity of Aro1 variants. CEN.PK2–1C ΔaroE cells were transformed with pRS42H-Aro1ΔE (Aro1ΔE), pRS42H-Aro1K1370A (Aro1K1370A) or pRS42H-Aro1D1407A (Aro1D1407A) in combination with p425-H7-Aro3K222L (Aro3K222L) and p423-H7-PaAroZ (AroZ) and subsequently cultivated in selective SCD medium (2% glucose) in shake flasks. The production of PCA was determined after 96 h at 30°C. The error bars show the standard deviation of two independent replicates. To test the activity of the Aro1 variants, plasmids overexpressing ARO1K1370A, ARO1D1407A and ARO1ΔaroE were transformed into CEN.PK2–1C ΔaroE. ARO3K222L was overexpressed for increased substrate flux, and AroZ for production of PCA, which was determined after 96 h of incubation in selective SCD medium in the culture supernatant (Fig. 3B). The point mutations ARO1K1370A (1014 mg L−1) and ARO1D1407A (957 mg L−1) resulted in a significant increase in PCA formation compared to the deleted version ARO1ΔaroE (597 mg L−1). Moreover, this experiment also shows the importance of additional ARO1 overexpression, as the strain carrying the empty vector instead of a vector overexpressing one of the ARO1 variants showed very low PCA production (18 mg L−1). The mutant ARO1K1370A was chosen for further experiments to prevent possible negative effects of the aroE deletion. Construction of an optimized production strain In the next step, we decided to integrate the information gained in the previous experiments into the construction of a new production strain. ARO4K229L, ARO3K222L and ARO1K1370A were combined on one plasmid under control of the strong constitutive TPI1, HXT7 and FBA1 promoters, respectively, using the backbone of 2μ plasmid pHD8 (Demeke et al. 2013) (Table 2), resulting in plasmid pHD-Aro3K222L, Aro4K229L and Aro1K1370A. In the newly generated plasmid, the integrative expression cassette is flanked by regions homologous to PYK2, a non-essential gene expressed only under growth on non-fermentable carbon sources (Boles et al. 1997). The expression cassette was subsequently stably integrated into the PYK2 locus in strain CEN.PK2–1C ΔaroE, thereby generating strain JTY1 (Table 1). The performance of the two strains was tested in comparison to strain CEN.PK2–1C and to strain CEN.PK2–1C ΔaroE transformed with the 2μ plasmid pHD-Aro3K222L, Aro4K229L and Aro1K1370A containing the expression cassette for the three genes (Fig. 4A). Upon growth in selective YEPD medium (2% glucose) containing G418, no 3-DHS was detectable in the culture supernatant of strain CEN.PK2–1C after 96 h of cultivation, while strain CEN.PK2–1C ΔaroE produced 19 mg L−1 3-DHS, confirming the importance of the deletion for accumulation of the precursor. Strain CEN.PK2–1C ΔaroE transformed with the high copy plasmid expressing the three enzymes Aro4K229L, Aro3K222L and Aro1K1370A produced 188 mg L−1 3-DHS. Finally, stable genomic integration of the expression cassette allowed for the production of up to 482 mg L−1 3-DHS with strain JTY1 after incubation for 96 h. Figure 4. View largeDownload slide Strain optimization for improved precursor supply. (A) Aro3K222L, Aro4K229L and Aro1K1370A were expressed from a multicopy plasmid (pHD-Aro3K222L, Aro4K229L, Aro1K1370A) in strain CEN.PK2–1C ΔaroE (2μ ARO3K222L, ARO4K229L, ARO1K1370A), or the expression cassette was integrated into the genome, thereby replacing the PYK2 gene (pyk2Δ::ARO3K222L, ARO4K229L, ARO1K1370A), generating strain JTY1. Cells were grown in shake flasks using YEPD medium (2% glucose) containing G418 for 96 h at 30°C, and the formation of 3-dehydroshikimate (3-DHS) in the culture supernatant was analyzed by HPLC. The error bars show the standard deviation of three independent replicates. (B) Strain JTY4 (zwf1Δ::TKL1) was generated by replacement of the ZWF1 gene by an expression cassette encoding Tkl1 in strain JTY1. Concentrations of 3-DHS in the culture supernatant were determined after 96 h at 30°C of cultivation in YEPD medium (2% glucose). The error bars show the standard deviation of two independent replicates. Figure 4. View largeDownload slide Strain optimization for improved precursor supply. (A) Aro3K222L, Aro4K229L and Aro1K1370A were expressed from a multicopy plasmid (pHD-Aro3K222L, Aro4K229L, Aro1K1370A) in strain CEN.PK2–1C ΔaroE (2μ ARO3K222L, ARO4K229L, ARO1K1370A), or the expression cassette was integrated into the genome, thereby replacing the PYK2 gene (pyk2Δ::ARO3K222L, ARO4K229L, ARO1K1370A), generating strain JTY1. Cells were grown in shake flasks using YEPD medium (2% glucose) containing G418 for 96 h at 30°C, and the formation of 3-dehydroshikimate (3-DHS) in the culture supernatant was analyzed by HPLC. The error bars show the standard deviation of three independent replicates. (B) Strain JTY4 (zwf1Δ::TKL1) was generated by replacement of the ZWF1 gene by an expression cassette encoding Tkl1 in strain JTY1. Concentrations of 3-DHS in the culture supernatant were determined after 96 h at 30°C of cultivation in YEPD medium (2% glucose). The error bars show the standard deviation of two independent replicates. To improve the performance of the strain even further, a copy of the gene encoding transketolase 1 (TKL1) was integrated into the ZWF1 locus under control of the strong TDH3 promoter, thereby disrupting the glucose-6-phosphate dehydrogenase function of the gene. These manipulations force the entry of glycolytic intermediates into the non-oxidative part of the pentose phosphate cycle via the Tkl1 reaction, leading to enhanced supply of the shikimic acid precursor E4P (Curran et al. 2013). Introducing these changes into strain JTY1 resulted in an improved titer of 3-DHS in the newly generated strain JTY4 (1552 mg L−1) compared to strain JTY1 (754 mg L−1) after 96 h (Fig. 4B). The use of non-selective YEPD medium in this experiment might be responsible for the higher titer of 3-DHS in strain JTY1 compared to the titers achieved with the same strain in the previous experiment shown in Fig. 4A, where cells were grown in YEPD medium containing G418. A novel PCA decarboxylase In previous studies in S. cerevisiae (Weber et al. 2012; Curran et al. 2013; Skjoedt et al. 2016; Suastegui et al. 2016, 2017; Leavitt et al. 2017) and E. coli (Draths and Frost 1994; Niu, Draths and Frost 2002; Sonoki et al. 2014; Zhang et al. 2015), the bacterial decarboxylase AroY has been used for the conversion of PCA to catechol, which represents the limiting step in the heterologous CCM pathway. AroY-C, the enzyme with the actual decarboxylase function, requires the prenylated flavin mononucleotide (prFMN) cofactor for its activity, which is synthesized with the help of AroY subunit B (AroY-B) (Payer et al. 2017; Weber et al. 2017). In bacterial genomes, AroY- C and -B are frequently encoded in the same operon together with subunit D (AroY-D), whose function is still unknown. Recently, a novel eukaryotic decarboxylase from the yeast Arxula adeninivorans, AGDC1, has been described (Meier et al. 2017), which shows no homology to previously known bacterial PCA decarboxylases. The enzyme has been shown to accept GA and PCA as substrates. For comparison of their activities, AGDC1 and AroY from K. pneumoniae were expressed as codon optimized versions in a PCA feeding experiment in strain CEN.PK2–1C, whereby AGDC1 and AroY-C are under control of the strong, constitutive PGK1 promoter (Fig. 5). All transformants were cultivated in selective SCD media containing 5 mM PCA for 144 hours. The pH of the medium supplemented with PCA is 4.3 at the start of the feeding experiment, corresponding to the pKa of 4.26 of the substance (Perrin and Sillen 1979), and is dropping to pH 2.8 in the course of the experiment (not shown). Thus, sufficient amounts of the undissociated form of the acid should be taken up passively by the cell. The results show that expression of AGDC1 from A. adeninivorans enhanced formation of catechol (140 mg L−1) compared to the expression of the bacterial enzyme AroY (109 mg L−1). To control that catechol is actually formed from PCA taken up by the cell and not from an intracellular source, we performed an experiment, where cells expressing AGDC1 were fed with distilled water or PCA (Fig. S1, Supporting Information). Despite the presence of a functional decarboxylase, cells fed with water did not produce measurable amounts of catechol, proving that catechol produced by AGDC1 expressing cells is solely generated by conversion of externally added PCA. Figure 5. View largeDownload slide Activities of two different decarboxylases. Expression vectors encoding bacterial KpAroY from Klebsiella pneumoniae (p426-AroY-BCD) and eukaryotic AGDC1 from Arxula adeninivorans (p426-AGDC1) were transformed in strain CEN.PK2–1C. Empty plasmid was transformed as control (Plasmid). All transformants were cultivated in shake flasks in selective synthetic complete (SCD) media (2% glucose) containing 5 mM protocatechuic acid (PCA) at 30°C. Final catechol concentrations were measured after 144 h. Error bars represent the standard deviation of three independent replicates. Figure 5. View largeDownload slide Activities of two different decarboxylases. Expression vectors encoding bacterial KpAroY from Klebsiella pneumoniae (p426-AroY-BCD) and eukaryotic AGDC1 from Arxula adeninivorans (p426-AGDC1) were transformed in strain CEN.PK2–1C. Empty plasmid was transformed as control (Plasmid). All transformants were cultivated in shake flasks in selective synthetic complete (SCD) media (2% glucose) containing 5 mM protocatechuic acid (PCA) at 30°C. Final catechol concentrations were measured after 144 h. Error bars represent the standard deviation of three independent replicates. To find other related enzymes with a potentially higher activity, we performed a BLAST search with non-redundant protein sequences (https://blast.ncbi.nlm.nih.gov/Blast.cgi). Our BLAST search revealed a plethora of sequences from diverse fungal species with identities of maximally 57% to AGDC1. For further analysis, we chose two proteins from Talaromyces atroroseus (Acc. No. XP_020118381.1; TaGDC1) and Madurella mycetomatis (Acc. No. KXX81388.1; MmGDC1) with 56% and 52% identity compared to AGDC1. Coding sequences for both proteins and AGDC1 were cloned in codon-optimized form under control of the strong, methionine-repressible MET25 promoter (Mumberg, Müller and Funk 1994). Subsequently, PCA decarboxylase activities of both proteins were analyzed in comparison to AGDC1 in a feeding experiment with 5 mM PCA in strain CEN.PK2–1C (Fig. 6). Compared to AGDC1 expressing cells reaching a catechol titer of 460 mg L−1, both other proteins turned out to have superior activity, with catechol titers of 619 mg L−1 (TaGDC1) and 584 mg L−1 (MmGDC1) after 120 h cultivation in selective SCD medium lacking methionine. TaGDC1 was chosen for further experiments. Figure 6. View largeDownload slide Activities of gallic acid decarboxylases from different species. Expression vectors encoding eukaryotic enzymes from Arxula adeninivorans (p426-MET25-AGDC1), Madurella mycetomatis (p426-MET25-MmGDC1) and Talaromyces atroroseus (p426-MET25-TaGDC1) were transformed in strain CEN.PK2–1C. The empty vector was transformed as a control (Plasmid). All transformants were cultivated in shake flasks in selective synthetic complete (SCD) medium (2% glucose) lacking methionine and containing 5 mM protochatechuic acid (PCA) for 96 h at 30°C. Final catechol concentrations were measured after 96 h. Error bars represent the standard deviation of two independent replicates. Figure 6. View largeDownload slide Activities of gallic acid decarboxylases from different species. Expression vectors encoding eukaryotic enzymes from Arxula adeninivorans (p426-MET25-AGDC1), Madurella mycetomatis (p426-MET25-MmGDC1) and Talaromyces atroroseus (p426-MET25-TaGDC1) were transformed in strain CEN.PK2–1C. The empty vector was transformed as a control (Plasmid). All transformants were cultivated in shake flasks in selective synthetic complete (SCD) medium (2% glucose) lacking methionine and containing 5 mM protochatechuic acid (PCA) for 96 h at 30°C. Final catechol concentrations were measured after 96 h. Error bars represent the standard deviation of two independent replicates. To test the performance of the new decarboxylase in optimized chassis strain JTY4 (Table 1), the three enzymes of the heterologous CCM production pathway were combined on one plasmid under control of strong promoters. The production of CCM and the pathway intermediates 3-DHS, catechol and PCA was monitored in a fermentation starting with an OD600 of 10 in selective SCD medium with 4% glucose over a period of 120 h (Fig. 7). While CCM titers reached 1244 mg L−1, intermediates 3-DHS and PCA still accumulated to levels of 2141 and 3063 mg L−1, respectively. With 23 mg L−1, catechol titers were negligible. To our knowledge, the yield of 31 mg g−1 glucose represents the so far highest yield of CCM obtained in S. cerevisiae. Figure 7. View largeDownload slide cis, cis-Muconic acid production at high cells density in a batch fermentation. An expression plasmid encoding AroZ from Podospora anserina, GDC1 from Talaromyces atroroseus and CatA from Acinetobacter radioresistens (p423-H7-ArCatA, TaGDC1, PaAroZ) was transformed into yeast strain JTY4. Cells were inoculated to an OD600 of 10 from a preculture and cultivated in shake flasks for 120 h in selective synthetic complete (SCD) medium (4% glucose) at 30°C. Samples from culture supernatants were taken at the indicated time points, and concentrations of 3-dehydroshikimate (3-DHS), protocatechuic acid (PCA) and cis, cis-muconic acid (CCM) were determined by HPLC. Error bars represent the standard deviation of three independent replicates, but are masked by the size of the symbols. Figure 7. View largeDownload slide cis, cis-Muconic acid production at high cells density in a batch fermentation. An expression plasmid encoding AroZ from Podospora anserina, GDC1 from Talaromyces atroroseus and CatA from Acinetobacter radioresistens (p423-H7-ArCatA, TaGDC1, PaAroZ) was transformed into yeast strain JTY4. Cells were inoculated to an OD600 of 10 from a preculture and cultivated in shake flasks for 120 h in selective synthetic complete (SCD) medium (4% glucose) at 30°C. Samples from culture supernatants were taken at the indicated time points, and concentrations of 3-dehydroshikimate (3-DHS), protocatechuic acid (PCA) and cis, cis-muconic acid (CCM) were determined by HPLC. Error bars represent the standard deviation of three independent replicates, but are masked by the size of the symbols. GA production with the optimized chassis yeast strain In the next step, we wanted to test the suitability of the JTY4 chassis strain for production of other aromatic compounds derived from 3-DHS. For this aim, we chose to introduce a pathway for production of GA, a compound with known antioxidant, anti-inflammative, antimutagenic and anticancer properties (Badhani, Sharma and Kakkar 2015). A pathway for production of GA from 3-DHS via PCA was already established before in E. coli, making use of a mutant form of p-hydroxybenzoate hydroxylase PobA from Pseudomonas aeruginosa (Kambourakis, Draths and Frost 2000). A conserved tyrosine 385 was replaced by phenylalanine (Y385F), changing substrate specificity and allowing for the enzyme to hydroxylate PCA to GA (Entsch et al. 1991; Eschrich et al. 1993). The reaction requires O2 and reduction equivalents. We created an analogous Y385F mutant of PobA from C. glutamicum (Huang et al. 2008) with the ability to hydroxylate PCA to GA. NADPH is the preferred cofactor for the enzyme from C. glutamicum (Huang et al. 2008). For production of GA in S. cerevisiae, wild-type pobA and the pobAY385F mutant were expressed in strain JTY4 together with aroZ, and the formation of GA was determined after 120 h (Fig. 8A). As anticipated, PobAY385F produced much higher GA titers (298 mg L−1) than the wild-type enzyme (33 mg L−1). To boost the titer even more, we applied the conditions optimized for CCM production, as shown in Fig. 7 (Fig. 8B). With a starting OD600 of 10, strain JTY4 expressing aroZ and pobAY385F from individual plasmids produced 682 mg L−1 of GA. Despite the relatively high titer of GA obtained under these conditions, activity of PobAY385F is still limiting, as PCA accumulated to 6599 mg L−1 in the culture supernatant, probably due to NADPH limitations in the zwf1 deletion strain JTY4 (see discussion). Remarkably, the total amount of aromatic compounds analyzed (3-DHS, PCA and GA together) was higher than observed for production of CCM (3-DHS, PCA and CCM together) in Fig. 7. Here, it has to be considered that different promoters were used in the two heterologous pathways (Table 2), which might influence the flux through the upstream pathway. Differences in toxicities of the intermediates and products and in their secretion might be also responsible for this observation. Figure 8. View largeDownload slide Gallic acid (GA) production in strain JTY4. (A) Activity of a mutant version of p-hydroxybenzoic acid hydroxylase PobA. Yeast strain JTY4 was transformed with expression plasmids p423-H7-PaAroZ and p425-H7-CgPobA (AroZ + PobA) or p425-H7-CgPobAY385F (AroZ + PobAY385F), respectively. Empty plasmids were transformed as control (Plasmid). For production of GA, cells were inoculated to an OD600 of 0.8 from precultures in selective synthetic complete (SCD) medium (2% glucose), and the formation of GA was determined after cultivation in shake flasks for 120 h at 30°C by HPLC analysis of the culture supernatant. (B) Gallic acid production at high cell density in a batch fermentation. Yeast strain JTY4 was transformed with expression plasmids p423-H7-PaAroZ and p425-H7-CgPobAY385F. Cells were inoculated to an OD600 of 10 from a preculture and cultivated for 120 h at 30°C in shake flasks in selective SCD medium (4% glucose). Culture supernatants were analyzed for production of 3-dehydroshikimate (3-DHS), protocatechuic acid (PCA) and GA using HPLC. Error bars represent the standard deviation of three independent replicates. Figure 8. View largeDownload slide Gallic acid (GA) production in strain JTY4. (A) Activity of a mutant version of p-hydroxybenzoic acid hydroxylase PobA. Yeast strain JTY4 was transformed with expression plasmids p423-H7-PaAroZ and p425-H7-CgPobA (AroZ + PobA) or p425-H7-CgPobAY385F (AroZ + PobAY385F), respectively. Empty plasmids were transformed as control (Plasmid). For production of GA, cells were inoculated to an OD600 of 0.8 from precultures in selective synthetic complete (SCD) medium (2% glucose), and the formation of GA was determined after cultivation in shake flasks for 120 h at 30°C by HPLC analysis of the culture supernatant. (B) Gallic acid production at high cell density in a batch fermentation. Yeast strain JTY4 was transformed with expression plasmids p423-H7-PaAroZ and p425-H7-CgPobAY385F. Cells were inoculated to an OD600 of 10 from a preculture and cultivated for 120 h at 30°C in shake flasks in selective SCD medium (4% glucose). Culture supernatants were analyzed for production of 3-dehydroshikimate (3-DHS), protocatechuic acid (PCA) and GA using HPLC. Error bars represent the standard deviation of three independent replicates. DISCUSSION In the past years, several attempts have been made to optimize the production of CCM with yeast, which have led to considerable improvements in titers and yields (Curran et al. 2013; Skjoedt et al. 2016; Suastegui et al. 2016, 2017; Leavitt et al. 2017). Genetic manipulations and experimental conditions that have permitted these improvements are summarized in Table S1 (Supporting Information). Nevertheless, the decarboxylation step from PCA to catechol still remains the main bottleneck of the pathway. In our study, we have created a strain with an optimized production of the shikimic acid pathway intermediate 3-DHS (Fig. 4). In this strain, we have successfully utilized the recently identified eukaryotic PCA/GA decarboxylase AGDC1 (Meier et al. 2017) (Fig. 5) and two of its homologs from other organisms for production of CCM (Fig. 6). Using the same strain, we could demonstrate the production of GA from 3-DHS via PCA for the first time in S. cerevisiae (Fig. 8). In S. cerevisiae, the entrance reaction into the shikimic acid pathway is catalyzed by DAHP synthases Aro3 and Aro4. These highly homologous proteins are feedback regulated by aromatic amino acids phenylalanine and tyrosine, respectively (Braus 1991). With lower Km values for E4P and PEP and a higher Ki for its specific inhibitor (Hartmann et al. 2003), Aro3 can be regarded as the more effective enzyme. Nonetheless, so far only feedback-resistant mutants of Aro4 have been utilized in metabolic engineering applications (Curran et al. 2013; McKenna et al. 2014; Li et al. 2015; Rodriguez et al. 2015; Suastegui et al. 2016). We have introduced an analogous mutation to the K229L exchange in the effector binding cavity of Aro4 into Aro3 (Hartmann et al. 2003), generating Aro3K222L. Overexpressed together with Aro4K229L, the flux into the shikimic acid pathway could be considerably improved compared to the individual mutants (Fig. 2). Nevertheless, we cannot exclude an additional positive effect due to the simultaneous codon optimization of Aro3K222L. The next steps in the shikimic acid pathway are catalyzed by the multifunctional Aro1 protein. To block further conversion of 3-DHS to aromatic amino acids, the dehydrogenase domain E had been completely deleted in a previous study (Weber et al. 2012). Here, we could show (i) that in general overexpression of ARO1 mutant versions without dehydrogenase activity is beneficial for 3-DHS and PCA production but (ii) that even higher product titers could be achieved by implementing an Aro1D1370A mutant with a point mutation in the E-domain, a mutation that is presumably less detrimental for the overall structure and activity of Aro1 than a deletion of the whole E-domain (Fig. 3). The deletion of the ZWF1 gene encoding glucose-6-phosphate dehydrogenase ZWF1 in combination with an overexpression of TKL1 has already been described to lead to an enhanced production of CCM (Curran et al. 2013), an effect that was not observed consistently in a more recent study in a different strain background (Suastegui et al. 2017). In our study, an integration of an overexpression construct for TKL1 combined with the deletion of ZWF1 in a ΔaroE strain already harboring integrations of overexpression constructs for ARO3K222L, ARO4K229L and ARO1K1370A led to a 2.1-fold enhancement in the formation of 3-DHS (Fig. 4B), and after integration of the complete CCM pathway allowed for a final titer of 1244 mg L−1 of CCM (Fig. 7). Glucose-6-phosphate dehydrogenase is the major source of cytosolic NADPH, and a deletion of the corresponding gene has been shown to cause defects in NADPH-requiring processes, such as assimiliation of sulfur, leading to a requirement for an organic sulfur source, such as methionine or cysteine (Thomas, Cherest and Surdin-Kerjan 1991). If such a deficiency in NADPH supply would lead to an impairment in growth, it would make a strain unattractive in industrial applications. Growth of the Δzwf1 strain JTY4 was not impaired in YEPD medium. Nevertheless, upon cultivation in SCD medium lacking methionine growth was slightly impaired and CCM titers were reduced, although growth tests revealed that the strain was not completely auxotrophic for methionine (not shown). Moreover, in a previous report about shikimic acid production, which involves the NADPH-dependent shikimic acid dehydrogenase, a deletion of ZWF1 led to an enhanced accumulation of 3-DHS and a reduced level of the desired product shikimic acid, indicating a reduced activity of the NADPH-dependent shikimate dehydrogenase function of Aro1 (Suastegui et al. 2017). Thus, if NADPH demands are high, especially due to the implementation of NADPH requiring reactions, limitations in product titers can be expected. Other cytosolic sources for NADPH exist, which could compensate for the deficiency. It has been shown that the cytosolic aldehyde dehydrogenase Ald6 can partially compensate for the reduced NADPH/NADP ratio in yeast cells with a deletion in ZWF1 (Grabowska and Chelstowska 2003). Isocitrate dehydrogenase (Idp2) is another source for cytosolic NADPH when glucose is exhausted (Minard and McAlister-Henn 2005). In our study about GA production, we also used the zwf1 deletion strain JTY4. The heterologous enzyme CgPobA prefers NADPH over NADH as a cofactor (Huang et al. 2008). Although we could achieve GA titers up to 682 mg L−1 (Fig. 8), the intermediate PCA still accumulated to very high amounts, which might be indicative for NADPH limitations. Other bacterial homologous with a preference for NADH exist, which could be used alternatively for the conversion of PCA to GA (Fujii and Kaneda 1985; Jadan et al. 2001; Iwaki et al. 2005). In addition to the replacement of NADPH with NADH consuming reactions (Brochado et al. 2010), implementing transhydrogenases converting NADH to NADPH (Fiaux et al. 2003) or establishing NADPH generating reactions (Verho et al. 2002) might be valuable strategies to broaden the applicability of a Δzwf1 strain. Moreover, as also the catalytic activity of PobA might be limiting, mutant forms could be used, such as the combined mutations of T294A and Y385F. The T294A mutation, which is located in the catalytic pocket of the enzyme, has recently been described to improve the hydrogen-bond mediated interaction between PobAY385F and PCA, thereby enhancing the catalytic activity of P. aeruginosa PobAY385F (Chen et al. 2017). The eukaryotic decarboxylase AGDC1 has been described to be specific not only for PCA, but also for GA (Meier et al. 2017). Engineering the specificity of AGDC1 could allow for the extension of our newly implemented pathway, enabling the production of pyrogallol, a compound with applications in the pharmaceutical and chemical industry (Kambourakis, Draths and Frost 2000). For the production of CCM, the eukaryotic enzyme AGDC1 exhibits several advantages over its prokaryotic AroY analog. The decarboxylase function of AroY is performed by subunit C and requires the cofactor prenylated flavin mononucleotide (prFMN), which is synthesized by AroY-B. prFMN can be synthesized with the help of the endogenous yeast enzyme Pad1, but this enzyme is absent in some yeast strains, such as the strain CEN.PK2–1C used in this study (Weber et al. 2017). Even though AroY-Ciso from K. pneumoniae showed reduced oxygen sensitivity compared to other C subunits tested (Weber et al. 2012), the enzyme showed only full activity upon cultivation of the host strain under oxygen-limited conditions (Suastegui et al. 2016). In contrast to AroY (Weber et al. 2017), AGDC1 and its homologs show high activity in strain CEN.PK2–1C (Figs 5 and 6) without the requirement to express an additional cofactor generating enzyme, reducing the metabolic burden imposed on the production organism. Moreover, a fine-tuning of expression of the decarboxylase and the cofactor generating enzyme can be omitted. Despite all these advantages and the titer of 1244 mg L−1 CCM achieved with the new enzyme, the decarboxylase substrate PCA still accumulates to 3063 mg L−1, still indicating a major limitation. However, the requirement for only one enzyme will certainly ease multiple integrations of the enzyme or enzyme engineering, which will be needed to overcome the bottleneck. In conclusion, although just applying a few rational design strategies in combination with a novel enzyme has led to production of a significant yield of CCM, for an industrial process these must be still improved. Novel high-throughput systems utilizing biosensors (Skjoedt et al. 2016; Leavitt et al. 2017) in combination with adaptive laboratory evolution strategies have a high potential to identify new targets for further manipulations, which might finally allow for an industrial application of yeast for CCM and GA production. SUPPLEMENTARY DATA Supplementary data are available at FEMSYR online. Acknowledgements The authors thank Dr Christian Weber for generating expression plasmids for CgPobA. The authors are grateful to Dr Jun-yong Choe (Rosalind Franklin University, North Chicago, USA) for his helpful suggestions. FUNDING This work was supported by the German Federal Ministry of Education and Research following a decision of the German Bundestag [grants 031A542 and 031B0218 to EB]. Conflict of interest. None declared. REFERENCES Aguilar-Zárate P, Cruz MA, Montañez J et al.   Gallic acid production under anaerobic submerged fermentation by two bacilli strains. Microb Cell Fact  2015; 14: 209. Google Scholar CrossRef Search ADS PubMed  Badhani B, Sharma N, Kakkar R. Gallic acid: a versatile antioxidant with promising therapeutic and industrial applications. RSC Adv  2015; 5: 27540– 57. Google Scholar CrossRef Search ADS   Bajpai B, Patil S. A new approach to microbial production of gallic acid. Braz J Microbiol  2008; 39: 708– 11. Google Scholar CrossRef Search ADS PubMed  Boles E, Zimmermann FK. Saccharomyces cerevisiae phosphoglucose isomerase and fructose bisphosphate aldolase can be replaced functionally by the corresponding enzymes of Escherichia coli and Drosophila melanogaster. Curr Genet  1993; 23: 187– 91. Google Scholar CrossRef Search ADS PubMed  Boles E, Schulte F, Miosga T et al.   Characterization of a glucose-repressed pyruvate kinase (Pyk2p) in Saccharomyces cerevisiae that is catalytically insensitive to fructose-1,6-bisphosphate. J Bacteriol  1997; 179: 2987– 93. Google Scholar CrossRef Search ADS PubMed  Braus GH. Aromatic amino acid biosynthesis in the yeast Saccharomyces cerevisiae: a model system for the regulation of a eukaryotic biosynthetic pathway. Microbiol Rev  1991; 55: 349– 70. Google Scholar PubMed  Brochado AR, Matos C, Moller BL et al.   Improved vanillin production in baker's yeast through in silico design. Microb Cell Fact  2010; 9: 84. Google Scholar CrossRef Search ADS PubMed  Bruder S, Reifenrath M, Thomik T et al.   Parallelised online biomass monitoring in shake flasks enables efficient strain and carbon source dependent growth characterisation of Saccharomyces cerevisiae. Microb Cell Fact  2016; 15: 127. Google Scholar CrossRef Search ADS PubMed  Chen Z, Shen X, Wang J et al.   Rational engineering of p-hydroxybenzoate hydroxylase to enable efficient gallic acid synthesis via a novel artificial biosynthetic pathway. Biotechnol Bioeng  2017; 114: 2571– 80. Google Scholar CrossRef Search ADS PubMed  Curran KA, Leavitt JM, Karim AS et al.   Metabolic engineering of muconic acid production in Saccharomyces cerevisiae. Metab Eng  2013; 15: 55– 66. Google Scholar CrossRef Search ADS PubMed  Demeke MM, Dietz H, Li Y et al.   Development of a D-xylose fermenting and inhibitor tolerant industrial Saccharomyces cerevisiae strain with high performance in lignocellulose hydrolysates using metabolic and evolutionary engineering. Biotechnol Biofuels  2013; 6: 89. Google Scholar CrossRef Search ADS PubMed  Dower WJ, Miller JF, Ragsdale CW. High efficiency transformation of E. coli by high voltage electroporation. Nucleic Acids Res  1988; 16: 6127– 45. Google Scholar CrossRef Search ADS PubMed  Draths KM, Frost JW. Environmentally compatible synthesis of adipic acid from D-glucose. J Am Chem Soc  1994; 116: 399– 400. Google Scholar CrossRef Search ADS   Entsch B, Palfey BA, Ballou DP et al.   Catalytic function of tyrosine residues in para-hydroxybenzoate hydroxylase as determined by the study of site-directed mutants. J Biol Chem  1991; 266: 17341– 9. Google Scholar PubMed  Eschrich K, van der Bolt FJ, de Kok A et al.   Role of Tyr201 and Tyr385 in substrate activation by p-hydroxybenzoate hydroxylase from Pseudomonas fluorescens. Eur J Biochem  1993; 216: 137– 46. Google Scholar CrossRef Search ADS PubMed  Farwick A, Bruder S, Schadeweg V et al.   Engineering of yeast hexose transporters to transport D-xylose without inhibition by D-glucose. P Natl Acad Sci USA  2014; 111: 5159– 64. Google Scholar CrossRef Search ADS   Fiaux J, Cakar ZP, Sonderegger M et al.   Metabolic-flux profiling of the yeasts Saccharomyces cerevisiae and Pichia stipitis. Eukaryot Cell  2003; 2: 170– 80. Google Scholar CrossRef Search ADS PubMed  Fujii T, Kaneda T. Purification and properties of NADH/NADPH-dependent p-hydroxybenzoate hydroxylase from Corynebacterium cyclohexanicum. Eur J Biochem  1985; 147: 97– 104. Google Scholar CrossRef Search ADS PubMed  Generoso WC, Gottardi M, Oreb M et al.   Simplified CRISPR-Cas genome editing for Saccharomyces cerevisiae. J Microbiol Methods  2016; 127: 203– 5. Google Scholar CrossRef Search ADS PubMed  Gottardi M, Reifenrath M, Boles E et al.   Pathway engineering for the production of heterologous aromatic chemicals and their derivatives in Saccharomyces cerevisiae: bioconversion from glucose. FEMS Yeast Res  2017; 17: fox035. Grabowska D, Chelstowska A. The ALD6 gene product is indispensable for providing NADPH in yeast cells lacking glucose-6-phosphate dehydrogenase activity. J Biol Chem  2003; 278: 13984– 8. Güldener U, Heck S, Fielder T et al.   A new efficient gene disruption cassette for repeated use in budding yeast. Nucleic Acids Res  1996; 24: 2519– 24. Google Scholar CrossRef Search ADS PubMed  Hamacher T, Becker J, Gardonyi M et al.   Characterization of the xylose-transporting properties of yeast hexose transporters and their influence on xylose utilization. Microbiology  2002; 148: 2783– 8. Google Scholar CrossRef Search ADS PubMed  Hartmann M, Schneider TR, Pfeil A et al.   Evolution of feedback-inhibited beta /alpha barrel isoenzymes by gene duplication and a single mutation. P Natl Acad Sci USA  2003; 100: 862– 7. Google Scholar CrossRef Search ADS   Huang Y, Zhao KX, Shen XH et al.   Genetic and biochemical characterization of a 4-hydroxybenzoate hydroxylase from Corynebacterium glutamicum. Appl Microbiol Biot  2008; 78: 75– 83. Google Scholar CrossRef Search ADS   Iwaki H, Saji H, Abe K et al.   Cloning and sequence analysis of the 4-hydroxybenzoate 3-hydroxylase gene from a cyclohexanecarboxylate-degrading Gram-positive bacterium, “Corynebacterium cyclohexanicum” strain ATCC 51369. Microb Environ  2005; 20: 144– 50. Google Scholar CrossRef Search ADS   Jadan AP, van Berkel WJ, Golovleva LA et al.   Purification and properties of p-hydroxybenzoate hydroxylases from Rhodococcus strains. Biochemistry (Mosc)  2001; 66: 898– 903. Google Scholar CrossRef Search ADS PubMed  Kambourakis S, Draths KM, Frost JW. Synthesis of gallic acid and pyrogallol from glucose:  replacing natural product isolation with microbial catalysis. J Am Chem Soc  2000; 122: 9042– 3. Google Scholar CrossRef Search ADS   Leavitt JM, Wagner JM, Tu CC et al.   Biosensor-enabled directed evolution to improve muconic acid production in Saccharomyces cerevisiae. Biotechnol J  2017; 12: 1600687. Google Scholar CrossRef Search ADS   Li M, Kildegaard KR, Chen Y et al.   De novo production of resveratrol from glucose or ethanol by engineered Saccharomyces cerevisiae. Metab Eng  2015; 32: 1– 11. Google Scholar CrossRef Search ADS PubMed  Lindner HA, Nadeau G, Matte A et al.   Site-directed mutagenesis of the active site region in the quinate/shikimate 5-dehydrogenase YdiB of Escherichia coli. J Biol Chem  2005; 280: 7162– 9. Google Scholar CrossRef Search ADS PubMed  McKenna R, Thompson B, Pugh S et al.   Rational and combinatorial approaches to engineering styrene production by Saccharomyces cerevisiae. Microb Cell Fact  2014; 13: 123. Google Scholar CrossRef Search ADS PubMed  Meier AK, Worch S, Boer E et al.   Agdc1p - a gallic acid decarboxylase involved in the degradation of tannic acid in the yeast blastobotrys (Arxula) adeninivorans. Front Microbiol  2017; 8: 1777. Google Scholar CrossRef Search ADS PubMed  Minard KI, McAlister-Henn L. Sources of NADPH in yeast vary with carbon source. J Biol Chem  2005; 280: 39890– 6. Google Scholar CrossRef Search ADS PubMed  Mumberg D, Müller R, Funk M. Regulatable promoters of Saccharomyces cerevisiae: comparison of transcriptional activity and their use for heterologous expression. Nucleic Acids Res  1994; 22: 5767– 8. Google Scholar CrossRef Search ADS PubMed  Niu W, Draths KM, Frost JW. Benzene-free synthesis of adipic acid. Biotechnol Prog  2002; 18: 201– 11. Google Scholar CrossRef Search ADS PubMed  Payer SE, Marshall SA, Barland N et al.   Regioselective para-carboxylation of catechols with a prenylated flavin dependent decarboxylase. Angew Chem Int Ed  2017; 56: 13893– 7. Google Scholar CrossRef Search ADS   Perrin DD, Sillen LG. Stability Constants of Metal-Ion Complexes, Part B: Organic Ligands . Oxford; New York: Pergamon Press, 1979. Rodriguez A, Kildegaard KR, Li M et al.   Establishment of a yeast platform strain for production of p- coumaric acid through metabolic engineering of aromatic amino acid biosynthesis. Metab Eng  2015; 31: 181– 8. Google Scholar CrossRef Search ADS PubMed  Seth M, Chand S. Biosynthesis of tannase and hydrolysis of tannins to gallic acid by Aspergillus awamori — optimisation of process parameters - ScienceDirect. Process Biochem  2000; 36: 39– 44. Google Scholar CrossRef Search ADS   Skjoedt ML, Snoek T, Kildegaard KR et al.   Engineering prokaryotic transcriptional activators as metabolite biosensors in yeast. Nat Chem Biol  2016; 12: 951– 8. Google Scholar CrossRef Search ADS PubMed  Sonoki T, Morooka M, Sakamoto K et al.   Enhancement of protocatechuate decarboxylase activity for the effective production of muconate from lignin-related aromatic compounds. J Biotechnol  2014; 192: 71– 7. Google Scholar CrossRef Search ADS PubMed  Suastegui M, Shao Z. Yeast factories for the production of aromatic compounds: from building blocks to plant secondary metabolites. J Ind Microbiol Biot  2016; 43: 1611– 24. Google Scholar CrossRef Search ADS   Suastegui M, Yu Ng C, Chowdhury A et al.   Multilevel engineering of the upstream module of aromatic amino acid biosynthesis in Saccharomyces cerevisiae for high production of polymer and drug precursors. Metab Eng  2017; 42: 134– 44. Google Scholar CrossRef Search ADS PubMed  Suastegui M, Matthiesen JE, Carraher JM et al.   Combining metabolic engineering and electrocatalysis: application to the production of polyamides from sugar. Angew Chem Int Ed  2016; 55: 2368– 73. Google Scholar CrossRef Search ADS   Taxis C, Knop M. System of centromeric, episomal, and integrative vectors based on drug resistance markers for Saccharomyces cerevisiae. Biotechiques  2006; 40: 73– 78. Google Scholar CrossRef Search ADS   Thomas D, Cherest H, Surdin-Kerjan Y. Identification of the structural gene for glucose-6-phosphate dehydrogenase in yeast. Inactivation leads to a nutritional requirement for organic sulfur. EMBO J  1991; 10: 547– 53. Google Scholar PubMed  Tochigi Y, Sato N, Sahara T et al.   Sensitive and convenient yeast reporter assay for high-throughput analysis by using a secretory luciferase from Cypridina noctiluca. Anal Chem  2010; 82: 5768– 76. Google Scholar CrossRef Search ADS PubMed  Verho R, Richard P, Jonson PH et al.   Identification of the first fungal NADP-GAPDH from Kluyveromyces lactis. Biochemistry  2002; 41: 13833– 8. Google Scholar CrossRef Search ADS PubMed  Weber C, Brückner C, Weinreb S et al.   Biosynthesis of cis,cis-muconic acid and its aromatic precursors, catechol and protocatechuic acid, from renewable feedstocks by Saccharomyces cerevisiae. Appl Environ Microb  2012; 78: 8421– 30. Google Scholar CrossRef Search ADS   Weber HE, Gottardi M, Bruckner C et al.   Requirement of a functional flavin mononucleotide prenyltransferase for the activity of a bacterial decarboxylase in a heterologous muconic acid pathway in Saccharomyces cerevisiae. Appl Environ Microb  2017; 83: 03472– 16. Google Scholar CrossRef Search ADS   Wiedemann B, Boles E. Codon-optimized bacterial genes improve l-arabinose fermentation in recombinant Saccharomyces cerevisiae. Appl Environ Microb  2008; 74: 2043– 50. Google Scholar CrossRef Search ADS   Xiao W, Gietz RD, Woods RA. Yeast transformation by the LiAc/SS carrier DNA/PEG method. In: Walker JM (ed). Yeast Protocol, Vol. 313 , Totowa, NJ: Humana Press, 2006, 107– 20. Zhang H, Pereira B, Li Z et al.   Engineering Escherichia coli coculture systems for the production of biochemical products. P Natl Acad Sci USA  2015; 112: 8266– 71. Google Scholar CrossRef Search ADS   © FEMS 2018. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com

Journal

FEMS Yeast ResearchOxford University Press

Published: Mar 1, 2018

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off