Genetic Engineering Approaches for the Microbial Production of Vanillin

Abstract

Vanilla flavour is widely used in various industries and is the most broadly used flavouring agent in the food industry. The demand for this flavour is, therefore, extremely high, yet vanilla bean extracts can only meet about 1% of the overall demand. Vanillin, the main constituent of vanilla flavour, can easily be obtained through chemical synthesis. Nonetheless, consumer demands for natural products and environmentally friendly industrial processes drive the development of biotechnological approaches for its production. Some microorganisms can naturally produce vanillin when fed with various substrates, including eugenol, isoeugenol, and ferulic acid. The characterisation of the genes and enzymes involved in these bioconversion pathways, as well as progress in the understanding of vanillin biosynthesis in Vanilla orchids, allowed the development of genetic engineering and synthetic biology approaches to increase vanillin production in naturally vanillin-producing microorganisms, or to implement novel vanillin biosynthetic pathways in microbial chassis. This review summarises and discusses these genetic engineering and synthetic biology approaches for the microbial production of vanillin.

Keywords: accessible raw material; bioconversion; metabolic engineering; microbial production; pathway design; synthetic biology; vanillin.

Figure 4

Optimisation of vanillin biosynthesis from ferulic acid. (A) Ferulic acid catabolic pathway and schematic representation of the constructions used to improve vanillin production; HMPHP-CoA: 4-hydroxy-3-methoxyphenyl-β-hydroxyprpionyl CoA. (B) List of genes which have been inactivated in different microorganisms to increase vanillin production [60,62,65,68,69,70,73,74].

Figure 1

Proposed vanillin biosynthetic pathways through 4-hydroxybenzandehyde and ferulic acid in V. planifolia. Based on Podstolski et al. [16], Gallage et al. [20], and Negishi and Negishi [21]. PAL: phenylalanine ammonia-lyase; C4H: cinnamate-4-hydroxylase; TAL: tyrosine ammonia-lyase; 4HBS: 4-hydroxybenzaldehyde synthetase; PPDO: phenylpropanoid 2,3-dioxygenase; CPR: cytochrome P450 reductase; OMT: O-methyltransferase; 4CL: 4-hydroxycinnamoyl-CoA ligase; HCT: hydroxycinnamoyl transferase; C3H: 4-coumaric acid 3-hydroxylase; VpVAN: vanillin synthase; UGT: UDP-glycosyltransferase; β-Gluc: β-glucosidase.

Figure 2

Vanillin biosynthesis and degradation pathways in vanillin-producing microorganisms. (A) Enzymes involved in vanillin production from different precursors. (B) Enzymes involved in vanillin degradation. IEM: isoeugenol monooxygenase; VAO: vanillyl alcohol oxidase; EUGH: eugenol hydroxylase; CADH/CALDH: coniferyl alcohol/aldehyde dehydrogenase; FCS: feruloyl-CoA synthetase; ECH/HCHL: enoyl-CoA hydratase/4-hydroxycinnamate CoA-hydratase/lyase; Lac: Laccase; DyP: dye-decolourising peroxidase; LiP: lignin peroxidase; MnP: manganese-dependent peroxidase; ALDR/VR: aldehyde reductase/vanillin reductase; VDH: vanillin dehydrogenase; VDC: vanillate decarboxylase; VDM: vanillate demethylase.

Figure 3

Vanillin production using different chassis: Advantages; Drawbacks; and Genetic engineering approaches.

Figure 5

Microbial vanillin biosynthesis from lignin. (A) Lignin degradation pathway with vanillin as an intermediate in Rhodococcus jostii RHA1 and schematic representation of the method used for deletion of vdh, in order to increase vanillin production. (B) Lignin degradation pathway with vanillin as an intermediate in Arthrobacter sp. C2 and schematic representation of the recombinant strain constructed to improve vanillin production. (C) Table summarising the function of the genes/enzymes used.

Figure 6

Genetic engineering approaches allowing the use of eugenol as a precursor of ferulic acid for vanillin bioproduction. (A) Schematic representation of the biosynthetic pathway and the applied constructions. (B) Table summarising the function and origin of the genes/enzymes used.

Figure 7

Vanillin production from eugenol or isoeugenol in E. coli. (A) Pathway from isoeugenol and schematic representation of the genetic construction used for its implementation. (B) Pathways from eugenol and schematic representation of the genetic constructions used for their implementation. (C) Table summarising the function and origin of the genes/enzymes used.

Figure 8

Vanillin production from ferulic acid in E. coli. (A) CoA-dependent non-β-oxidative deacetylation pathway and schematic representation of the construction used for its implementation. (B) Table summarising the function of genes/enzymes used. The origin of genes is indicated under the schematic representation of plasmids.

Figure 9

Vanillin production from L-tyrosine by a ferulate biosynthetic pathway. (A) Schematic representation of the biosynthetic pathway and of the constructions used to implement it in E. coli or S. cerevisiae. Different transcriptional terminators are represented by different colors. (B) Table summarising the function and origin of the genes/enzymes used.

Figure 10

Vanillin production from L-phenylalanine by a benzoate biosynthetic pathway. (A) Schematic representation of the biosynthetic pathway and of the constructions used to implement it. Different transcriptional terminators are represented by different colors. (B) Table summarising the function and origin of the genes/enzymes used.

Figure 11

Vanillin production from ferulic acid by a de novo CoA-independent decarboxylation pathway. (A) Schematic representation of the pathway and of the constructions used. (B) Table summarising the function and origin of genes/enzymes used. ADO* represents a modified ADO able to convert ferulic acid directly into vanillin.

Figure 12

Vanillin production from glucose by de novo biosynthetic pathways. (A) Vanillin biosynthetic pathways. The arrows representing the reactions are coloured as follows: grey—primary metabolism (e.g., shikimate pathway); blue—conversion of dehydroshikimic acid into vanillin; orange—conversion of chorismate into protocatechuic acid; yellow—conversion of hydroxyphenylpyruvate into protocatechualdehyde; and magenta—conversion of vanillin in vanillin glucoside. Green arrows and enzyme names indicate that the enzymes catalysing the reactions are overexpressed in optimised chassis. Red dotted arrows marked with an X and red enzyme names indicates that the enzyme catalysing the reactions are absent in optimised chassis, following the inactivation of the endogenous genes encoding the corresponding enzymes. (B) Strategies to improve the SAM supply for methylation step. (C) Modified steps of glycolysis or nitrogen metabolism to improve the NADPH supply. (D) Strategies to improve the E4P precursor supply. For panels B, C, and D, the green and red arrows and protein names are used, as in panel A, to indicate overexpression or absence, respectively. Abbreviations of metabolites: Ac-CoA—Acetyl-Coenzyme A; Ace—Acetate; AcP—acetyl phosphate; DAHP—3-deoxy-D-arabino-heptulosonate-7-phosphate; DHGA—3,4-dihydroxyphenylglyoxylate; DHMA—3,4-dihydroxymandelate; DHQ—3-dehydroquinic acid; DHS—3-dehydroshikimic acid; 1,3DPG—3-phospho-D-glyceroyl-phosphate; E4P—erythrose-4-phosphate; F6P—fructose 6-phosphate; GAP—D-glyceraldehyde 3-phosphate; G6P—glucose 6-phosphate; HCys—homocysteine; HMA—hydroxymandelate; HPP—hydroxyphenylpyruvate; Met—Methionine; PEP—phosphoenolpyruvate; Pyr—pyruvate; SAH—S-adenosyl-L-homocysteine; SAM—S-adenosylmethionine; SRH—S-ribosyl-L-homocysteine; Trp—Tryptophan. Abbreviations of proteins: ACAR—Aromatic carboxylic acid reductase; AroB—DHQ synthase; AroD—DHQ dehydratase; AroE—Shikimate dehydrogenase; AroF, AroG, AroH—DAHP synthase feedback-regulated isoenzymes; AroZ, 3DSD—3-dehydroshikimate dehydratase; BFD—Benzoylformate decarboxylase; CPL—Chorismate-pyruvate lyase; GapC—GAP dehydrogenase; GDH1/GDH2—NADP⁺/NAD⁺-dependent glutamate dehydrogenases; GPP1—glycerol-1-phosphate; HBH—Hydroxybenzoate hydroxylase; HmaS—Hydroxymandelate synthase; HMO—Hydroxymandelate oxidase; HpaBC—Two-component flavin-dependent monooxygenase; LuxS—S-ribosylhomocysteine lyase; MetE, MetH—Methionine synthases; MetJ—transcription regulator of methionine biosynthetic gene cluster; MetK—SAM synthetase; Mtn—5′-methylthioadenosine/S-adenosylhomocysteine nucleosidase; OMT—O-methyltransferase; PDC1—pyruvate decarboxylase; Pos5c—cytosol-relocalised NADH kinase; PpsA—phosphoenolpyruvate synthase; PPTase—Phosphopantetheine transferase; Pta—phosphotransacetylase; TktA—Transketolase; UGT—UDP-glycosyltransferase; Xfpk—phosphoketolase.

Figure 13

Implementation of de novo biosynthetic pathways to produce vanillin from glucose in different model chassis. (A1–A6) Schematic representation of the genetic constructions used to implement these pathways. Different transcriptional terminators are represented by different colors. Coloured arrows represent genes encoding enzymes involved in different metabolic steps. Blue—conversion of dehydroshikimic acid into vanillin; Orange—conversion of chorismate into protocatechuic acid; Yellow—conversion of HPP into protocatechualdehyde; Magenta—conversion of vanillin in vanillin glucoside; Grey—genes involved in primary metabolism which are overexpressed for chassis optimisation; Red rectangle within an arrow or ∆—endogenous genes deleted for chassis optimisation; (B) Table summarising the origin of the heterologous genes. List of genes deleted in model microorganisms to improve vanillin production: adh6—alcohol dehydrogenase; aroE—shikimate dehydrogenase; gdh1—glutamate dehydrogenase I; gpp1—glycerol-1-phosphatese; MARE—multiple-aldehyde reductases; metJ—transcription regulator of methionine biosynthetic cluster; NCgl0324—aromatic aldehyde reductase; pcaHG—protocatechuate dioxygenase; pdc1—pyruvate decarboxylase; serA—phosphoglycerate dehydrogenase; RARE—multiple-aldehyde reductases; trpE—anthranilate synthase; csm—chorismate mutase; vdh—vanillin dehydrogenase; vanAB—vanillate demethylase. List of endogenous genes overexpressed (in bold) for chassis optimisation: aroB—DHQ synthase; aroZ—3-dehydroshikimate dehydratase; gdh2—glutamate dehydrogenase II; luxS—S-ribosylhomocysteine lyase; mtn –5′-methylthioadenosine/S-adenosylhomocysteine nucleosidase; ppsA—phosphoenolpyruvate synthase; serA—phosphoglycerate dehydrogenase; tktA—transketolase.

Figure 14

Implementation of a de novo biosynthetic pathway to produce vanillin from a PET-derived monomer (terephthalate acid). (A) Schematic representation of the biosynthetic pathway and the used constructions to produce vanillin in E. coli. (B) Table summarising the function and origin of the genes/enzymes used.

Figure 15

Different routes for vanillin production.