INTRODUCTION
Carotenoids are diverse class of C40 isoprenoids widely produced by plants, bacteria, fungi and microalgae (Berman et al., 2015; Henríquez, Escobar, Galarza, & Gimpel, 2016). Of all known carotenoids, β-carotene is believed to be the most important due to its nutritional role as pro-vitamin A (Dowling & Wald, 1960) and health-promoting potential as an antioxidant (Palozza & Krinsky, 1992) and an anti-tumor agent (Williams, Boileau, Zhou, Clinton, & Erdman, 2000). Its wide applications in nutraceutical, feed and cosmetic industries lead to a fast-growing world market (Irwandi Jaswir, 2011). Currently, chemical synthesis remains the major route of commercial β-carotene production. Considering the safety concerns of chemical synthesis, and consumer preferences for natural additives, microbial production of β-carotene via metabolic engineering gains increasing interests and becomes an attractive alternative (Yoon et al., 2007; Zhao et al., 2013). The biological pathway of all isoprenoids use isopentenyl diphosphate (IPP) as precursor, which is synthesized through either MVA pathway in eukaryotes, or the 2-C-methyl-D-erythritol-4-phosphate (MEP) pathway in prokaryotes. Among potential microbial hosts, Saccharomyces cerevisiae has superior traits in industrial production of isoprenoids such as the GRAS (generally recognized as safe) status, ease of genetic manipulation, industrial robustness (Auesukaree et al., 2009), and the native MVA pathway which is generally considered as an effective supplier of isoprenoid precursor from acetyl-CoA (Vickers, Williams, Peng, & Cherry, 2017).
Researchers expend great efforts in heterologous production of carotenoids using engineered S. cerevisiae . Those efforts have far involved the optimization of metabolic flux, and balancing necessary cofactors by manipulating the expression levels of targeted genes (Das et al., 2007; Peralta-Yahya et al., 2011; Verwaal et al., 2007; Yan, Wen, & Duan, 2012). Among all the reported manipulation targets, overexpression of a truncated, soluble form of 3-hydroxy-3-methylglutaryl-coenzyme A reductase (tHMG1 ), a major rate-limiting enzyme of the MVA pathway, has been consistently recognized as an essential strategy for high-level production of carotenoids (Verwaal et al., 2007; Xie, Lv, Ye, Zhou, & Yu, 2015; Zhou et al., 2017) and other isoprenoids, such as artemisinic acid (Ro et al., 2006), farnesene (Meadows et al., 2016), squalene and amorphadiene (Kwak et al., 2017) in S. cerevisiae . In addition to tHMG1overexpression, up-regulation of the MVA pathway related genes such asERG8 , ERG12 , ERG19 , IDI1 , ERG20 (Y. Sun, Sun, Shang, & Yan, 2016) and down-regulation of the ergosterol pathway related genes such as ERG9 (Yan et al., 2012) have been attempted to increase the production of carotenoids. However, extensive genetic manipulations usually increase metabolic burdens on the host and thus cause instable performance in industrial-scale fermentation (Hollinshead, He, & Tang, 2014).
More importantly, despite intensive genetic perturbations for driving metabolic fluxes towards carotenoids production, ethanol remains a major product due to the entirely fermentative metabolism of S. cerevisiae on glucose even in the presence of oxygen (Pfeiffer & Morley, 2014), which hindered the high-level production of carotenoids. This well-known metabolic regulation, termed the Crabtree effect, was not observed while using non-native sugar xylose as a carbon source (Y.-S. Jin, Laplaza, & Jeffries, 2004; Kwak et al., 2017; Matsushika, Goshima, & Hoshino, 2014). We, therefore, assumed that xylose fermentation by engineered S. cerevisiae might facilitate carotenoids production by alleviating glucose-dependent repression on respiratory metabolism. Additionally, xylose, comprising up to 30-40 % of lignocellulosic biomass, is the second most abundant sugar in nature that derived from non-edible sources (Kim, Ha, Wei, Oh, & Jin, 2012). Efficient production of value-added chemicals like carotenoids and vitamin A from xylose is an important step toward economically feasible and sustainable bioconversion processes of lignocellulosic biomass (Kwak, Jo, Yun, Jin, & Seo, 2019; L. Sun, Kwak, & Jin, 2019). However, carotenoids production from xylose in engineered S. cerevisiaehas yet been reported.
As such, in this study, we sought to overproduce β-carotene from xylose in engineered S. cerevisiae . High-level production of β-carotene was achieved using xylose as a carbon source without tHMG1overexpression and other genetic perturbations. In order to explore the advantageous traits of xylose utilization for β-carotene production in engineered yeast, we assessed the differences in β-carotene production patterns from glucose and xylose via fermentation profiling, metabolites analysis and comparative transcriptional studies. To the best of our knowledge, the titer of β-carotene achieved in this study is among the highest reported in engineered S. cerevisiae (López et al., 2019; Xie, Ye, Lv, Xu, & Yu, 2015). This study demonstrated that using xylose as a carbon source would be a promising strategy potentially bypassing extensive genetic perturbations for high-level and stable production of carotenoids and other isoprenoids in S.cerevisiae .