Rhodosporidium toruloides: a new platform organism for conversion of lignocellulose into terpene biofuels and bioproducts

Abstract

Background: Economical conversion of lignocellulosic biomass into biofuels and bioproducts is central to the establishment of a robust bioeconomy. This requires a conversion host that is able to both efficiently assimilate the major lignocellulose-derived carbon sources and divert their metabolites toward specific bioproducts.

Results: In this study, the carotenogenic yeast Rhodosporidium toruloides was examined for its ability to convert lignocellulose into two non-native sesquiterpenes with biofuel (bisabolene) and pharmaceutical (amorphadiene) applications. We found that R. toruloides can efficiently convert a mixture of glucose and xylose from hydrolyzed lignocellulose into these bioproducts, and unlike many conventional production hosts, its growth and productivity were enhanced in lignocellulosic hydrolysates relative to purified substrates. This organism was demonstrated to have superior growth in corn stover hydrolysates prepared by two different pretreatment methods, one using a novel biocompatible ionic liquid (IL) choline α-ketoglutarate, which produced 261 mg/L of bisabolene at bench scale, and the other using an alkaline pretreatment, which produced 680 mg/L of bisabolene in a high-gravity fed-batch bioreactor. Interestingly, R. toruloides was also observed to assimilate p-coumaric acid liberated from acylated grass lignin in the IL hydrolysate, a finding we verified with purified substrates. R. toruloides was also able to consume several additional compounds with aromatic motifs similar to lignin monomers, suggesting that this organism may have the metabolic potential to convert depolymerized lignin streams alongside lignocellulosic sugars.

Conclusions: This study highlights the natural compatibility of R. toruloides with bioprocess conditions relevant to lignocellulosic biorefineries and demonstrates its ability to produce non-native terpenes.

Keywords: Amorphadiene; Bisabolene; Heterologous expression; Multiple carbon source utilization; Plant biomass-derived hydrolysate; Rhodosporidium toruloides; Terpenes.

Fig. 1

Schematic of R. toruloides as a new platform for the production of lignocellulosic biofuels and bioproducts

Fig. 2

Terpene titers of R. toruloides transformants. a Bisabolene and b amorphadiene titers in selected strains grown in SD medium with 2% glucose. 5 mL cultures in test tubes were set up at a starting OD of 0.1 with a 20% dodecane overlay. At day 7, the dodecane layer was sampled and analyzed for bisabolene or amorphadiene. (n = 3, data shown as average ± standard deviation, representative from two independent experiments)

Fig. 3

Stability of bisabolene production in serial cultures. Cultures in SD medium with 2% glucose were passaged consecutively every 6 days. (n = 3, data shown as average ± standard deviation, from a single experiment)

Fig. 4

The effect of pH on bisabolene titers. (n = 3, data shown as average ± standard deviation, from a single experiment in SD medium with 2% glucose)

Fig. 5

Conversion of glucose, xylose, and p-coumaric acid, both individually and mixed, into bisabolene by R. toruloides. Bisabolene titers, growth, and carbon utilization of strain BIS3 grown in SD medium supplemented with different carbon sources: a 0.5% glucose, b 0.5% xylose, c 0.5% p-coumaric acid, and d 0.5% glucose, 0.5% xylose, 0.5% p-coumaric acid. Left panels: lines represent ODs, bars represent bisabolene titers. Right panels: glucose (red), xylose (black), p-coumaric acid (blue). 5 mL cultures in test tubes were set up at a starting OD of 0.1 with a 20% dodecane overlay. At each time point, the dodecane layer was sampled and analyzed for bisabolene and the aqueous layer was sampled for OD measurement and carbon utilization analysis (n = 3, data shown as average ± standard deviation, representative from at least four independent experiments)

Fig. 6

Utilization of several lignin-related aromatic compounds by R. toruloides. Carbon source utilization of R. toruloides grown in SD medium supplemented with 2 g/L of either p-coumaric acid, p-hydroxybenzoic acid (4-HBA), ferulic acid, vanillic acid, or benzoic acid (n = 3, data shown as average ± standard deviation, from a single experiment)

Fig. 7

Conversion of biomass-derived glucose, xylose, and p-coumaric acid into bisabolene by R. toruloides. Bisabolene titers, growth, and carbon source utilization of strain BIS3 grown in (a) corn stover hydrolysate and (b) SD medium supplemented with individual components at the same concentration as those found in the corn stover hydrolysate: glucose (17.1 g/L), xylose (9.1 g/L), p-coumaric acid (383 mg/L), alpha-ketoglutarate (254 mM), and choline (586 mM). A low level of arabinose (0.98 g/L) was also detected in the hydrolysate and included in the control medium. Left panels: lines represent ODs, bars represent bisabolene titers. Right panels: glucose (red), xylose (black), p-coumaric acid (blue). 5 mL cultures in test tubes were set up at a starting OD of 0.1 with a 20% dodecane overlay. At each time point, the dodecane layer was sampled and analyzed for bisabolene and the aqueous layer was sampled for OD measurement and carbon utilization analysis (n = 3, data shown as average ± standard deviation, representative from at least two individual experiments)

Fig. 8

High-carbon fed-batch fermentation of R. toruloides. Bioreactor cultivation of strain BIS3 in (a) alkaline hydrolysate (b) SD medium with glucose. Left panels: lines represent dry cell weight (DCW), bars represent bisabolene titers. Bisabolene titers were measured three times per time point, average value is shown. Right panels: measured glucose (red) and xylose (black) concentration profiles. At each time point, 10 mL of the culture was sampled. After separation, the dodecane layer was used for bisabolene measurement. 5 mL of the aqueous layer was used for the measurement of DCW