Lycopene overproduction in Saccharomyces cerevisiae through combining pathway engineering with host engineering

Abstract

Background: Microbial production of lycopene, a commercially and medically important compound, has received increasing concern in recent years. Saccharomyces cerevisiae is regarded as a safer host for lycopene production than Escherichia coli. However, to date, the lycopene yield (mg/g DCW) in S. cerevisiae was lower than that in E. coli and did not facilitate downstream extraction process, which might be attributed to the incompatibility between host cell and heterologous pathway. Therefore, to achieve lycopene overproduction in S. cerevisiae, both host cell and heterologous pathway should be delicately engineered.

Results: In this study, lycopene biosynthesis pathway was constructed by integration of CrtE, CrtB and CrtI in S. cerevisiae CEN.PK2. When YPL062W, a distant genetic locus, was deleted, little acetate was accumulated and approximately 100 % increase in cytosolic acetyl-CoA pool was achieved relative to that in parental strain. Through screening CrtE, CrtB and CrtI from diverse species, an optimal carotenogenic enzyme combination was obtained, and CrtI from Blakeslea trispora (BtCrtI) was found to have excellent performance on lycopene production as well as lycopene proportion in carotenoid. Then, the expression level of BtCrtI was fine-tuned and the effect of cell mating types was also evaluated. Finally, potential distant genetic targets (YJL064W, ROX1, and DOS2) were deleted and a stress-responsive transcription factor INO2 was also up-regulated. Through the above modifications between host cell and carotenogenic pathway, lycopene yield was increased by approximately 22-fold (from 2.43 to 54.63 mg/g DCW). Eventually, in fed-batch fermentation, lycopene production reached 55.56 mg/g DCW, which is the highest reported yield in yeasts.

Conclusions: Saccharomyces cerevisiae was engineered to produce lycopene in this study. Through combining host engineering (distant genetic loci and cell mating types) with pathway engineering (enzyme screening and gene fine-tuning), lycopene yield was stepwise improved by 22-fold as compared to the starting strain. The highest lycopene yield (55.56 mg/g DCW) in yeasts was achieved in 5-L bioreactors. This study provides a good reference of combinatorial engineering of host cell and heterologous pathway for microbial overproduction of pharmaceutical and chemical products.

Keywords: Heterologous pathway; Lycopene; Metabolic engineering; Saccharomyces cerevisiae; Synthetic biology.

Fig. 1

Schematic representation of the engineering strategies for enhanced lycopene production in S. cerevisiae. a The engineered lycopene biosynthetic pathway in S. cerevisiae. Genetic modifications are noted by thick arrows. Red arrows indicate the heterologous lycopene biosynthetic pathway consisting of CrtE, CrtB and CrtI, which starts either from FPP (solid arrow) or directly from IPP and DMAPP (dashed arrow). The enzyme overexpressed in the described lycopene-producing strain is highlighted in blue. b Construction of plasmids and integration modules. c The rationale of experiment design: combinatorial engineering of host cell and heterologous pathway

Fig. 2

The effect of Δypl062w on lycopene production. S. cerevisiae SyBE_Sc14C07 and SyBE_Sc14C23 were cultivated in YPDG media containing different concentrations of glucose (2 %, left side; 4 %, right side), respectively, in shake-flasks for analysis of lycopene production (a, b), acetate accumulation (c, d) and cytosolic acetyl-CoA level (e, f). The error bars represent standard deviation calculated from triplicate experiments

Fig. 3

Combinatorial optimization of CrtE, CrtB and CrtI from diverse species. Thirty lycopene-producing strains were constructed by screening enzymes from various sources and tested for lycopene production. Pa, Pantoea agglomerans; Sa, Sulfolobus acidocaldarius; Af, Archaeoglobus fulgidus; Bt, Blakeslea trispora; Tm, Taxus x media; Aa, Paracoccus sp. (formerly Agrobacterium aurantiacum). The error bars represent standard deviation calculated from triplicate experiments

Fig. 4

Carotenoid production by fine-tuning of BtCrtI and selecting homologous haploid strains. BtCrtI expression was fine-tuned by adjusting copy number and promoter strength, and haploid cells with different mating types (a, α) were compared as well for carotenoid production. The error bars represent standard deviation calculated from triplicate experiments

Fig. 5

The effects of distant genetic loci on lycopene production. Three gene-deletion targets (ROX1, DOS2, and YJL064W) and one overexpression target (INO2) were investigated in S. cerevisiae SyBE_Sc14D07 for lycopene production. The error bars represent standard deviation calculated from triplicate experiments

Fig. 6

Lycopene production in fed-batch fermentation. a Profile of glucose, ethanol, cell density and lycopene accumulation of strain SyBE_Sc14D14 during fed-batch fermentation. b Percentage ratios of the produced carotenoid composition at 120 h. The red liquid in the bottles was the lycopene fermentation broth. The error bars represent standard deviation calculated from duplicate experiments