Metabolic Engineering and Process Intensification for Muconic Acid Production Using Saccharomyces cerevisiae

Abstract

The efficient production of biobased organic acids is crucial to move to a more sustainable and eco-friendly economy, where muconic acid is gaining interest as a versatile platform chemical to produce industrial building blocks, including adipic acid and terephthalic acid. In this study, a Saccharomyces cerevisiae platform strain able to convert glucose and xylose into cis,cis-muconic acid was further engineered to eliminate C2 dependency, improve muconic acid tolerance, enhance production and growth performance, and substantially reduce the side production of the intermediate protocatechuic acid. This was achieved by reintroducing the PDC5 gene and overexpression of QDR3 genes. The improved strain was integrated in low-pH fed-batch fermentations at bioreactor scale with integrated in situ product recovery. By adding a biocompatible organic phase consisting of CYTOP 503 and canola oil to the process, a continuous extraction of muconic acid was achieved, resulting in significant alleviation of product inhibition. Through this, the muconic acid titer and peak productivity were improved by 300% and 185%, respectively, reaching 9.3 g/L and 0.100 g/L/h in the in situ product recovery process as compared to 3.1 g/L and 0.054 g/L/h in the control process without ISPR.

Keywords: Saccharomyces cerevisiae; in situ product recovery; metabolic engineering; muconic acid; reactive extraction; yeast cell factory.

Figure 1

Schematic representation of the genetic targets engineered to establish muconic acid (MA) production in S. cerevisiae strain TN22 (red), and the modifications performed in this study (blue) to enhance its performance. In TN22, MA production was established by the expression of a 3-dehydroshikimate (DHS) dehydratase from P. anserina (aroZ_pan), protocatechuic acid (PCA) decarboxylase (PCAD) from K. pneumoniae (aroY-C^iso), and oxygen-consuming catechol 1,2-dioxygenase (CDO) from C. albicans (Hqd2). Feedback inhibition by aromatic amino acids for entry of carbon into the shikimate pathway was alleviated by expressing a feedback-resistant DAHP-synthase (Aro4^K229L) and the production of prenylated flavine mononucleotide (prFMN), a co-factor of PCAD, was enhanced by overexpressing PAD1. Moreover, ethanol production was eliminated by the deletion of PDC1, PDC5 and PDC6. In this work, PDC5 was reintegrated to eliminate the C2 dependency for ccMA production and the multidrug transporter encoded by QDR3 was expressed to enhance MA export (Created with BioRender.com).

Figure 2

Relieving the C2 dependency of S. cerevisiae TN22. (A) ccMA production at shake flask level by S. cerevisiae TN22 in yeast extract-peptone 4% dextrose (YP4%D) supplemented with 1% ethanol (●) and yeast extract-peptone-xylose (YP4%X) supplemented with 1% ethanol (■), YP4%D (◯), and YP4%X (◻). (B) ccMA production by 129 evolved clones (black), compared to TN22 (red) after 72 hours in YP4%D. (C,D) Effect of PDC5 overexpression on MA production in S. cerevisiae TN22. Strains were fermented at shake flask level in a glucose-containing medium, YP4%D (C), or xylose-containing medium, YP4%X (D). TN22 (●) was compared to strains overexpressing PDC5, in order of increasing promoter strengths: TN22 REV1p_PDC5 (△), TN22 SAC6p_PDC5 (▲), TN22 RPL18Bp_PDC5 (◻), TN22 TEF1p_PDC5 (■). Results are the means of three biological replicates. Error bars show the standard deviation at each time point.

Figure 3

Overexpression of QDR3 in S. cerevisiae TN22 improves ccMA (A,B) and PCA (C,D) production. Strains were fermented in glucose-containing medium, YP4%D1%E (A,C), or xylose-containing medium, YP4%X1%E (B,D). S. cerevisiae TN22 (●) was compared to strains overexpressing QDR3, in order of increasing promoter strengths TN22 REV1p_QDR3 (△), TN22 SAC6p_QDR3 (▲), TN22 RPL18Bp_QDR3 (◻), TN22 TEF1p_QDR3 (■). Results are the means of three biological replicates. Error bars show the standard deviation at each time point.

Figure 4

Overexpression of QDR3 and PDC5 in S. cerevisiae TN22 results in improved ccMA (A) and PCA (B) production. Strains were fermented in YP4%D. S. cerevisiae TN22 (●) was compared to S. cerevisiae TN22 RPL18Bp_PDC5 (◻) and S. cerevisiae TN22 RPL18Bp_PDC5; TEF1p_QDR3 (▲). Results are the means of three biological replicates. Error bars show the standard deviation at each time point.

Figure 5

MA fermentations in 2L bioreactors using S. cerevisiae TN22 RPL18Bp_PDC5; TEF1p_QDR3. Solid lines represent in situ product recovery (ISPR) fermentations, where 20% of solvent (v/v) composed of CYTOP 503 (12.5 v%) in canola oil was added after 74 h of fermentation (vertical line). Dashed lines represent control fermentations. Cell dry weight (CDW) (A), MA titer (B), productivity (C), glucose consumption (D), pH (E), and PCA titer (F) are displayed as mean values of biological duplicates. Error bars show the deviations at each time point.