Coordinated induction of multi-gene pathways in Saccharomyces cerevisiae

Abstract

Bacterial operons are nature's tool for regulating and coordinating multi-gene expression in prokaryotes. They are also a gene architecture commonly used in the biosynthesis of many pharmaceutically important compounds and industrially useful chemicals. Despite being an important eukaryotic production host, Saccharomyces cerevisiae has never had such gene architecture. Here, we report the development of a system to assemble and regulate a multi-gene pathway in S. cerevisiae. Full pathways can be constructed using pre-made parts from a plasmid toolbox. Subsequently, through the use of a yeast strain containing a stably integrated gene switch, the assembled pathway can be regulated using a readily available and inexpensive compound-estradiol-with extremely high sensitivity (10 nM). To demonstrate the use of the system, we assembled the five-gene zeaxanthin biosynthetic pathway in a single step and showed the ligand-dependent coordinated expression of all five genes as well as the tightly regulated production of zeaxanthin. Compared with a previously reported constitutive zeaxanthin pathway, our inducible pathway was shown to have 50-fold higher production level.

Figure 1.

The behaviour of GAL inducible promoters, monitored via GFP expression and induced using Gal-P65. The GFP channel histograms obtained from flow cytometry are shown. Black indicates uninduced sample and red indicates induced sample. The sample identity and the percentage of positives are shown within each histogram. (A) The behaviour of natural GAL inducible promoters cloned from S. cerevisiae. (B) The behaviour of synthetic GAL inducible promoters, which were generated by fusing the synthetic UAS sequences with the TATA box of the GAL1 promoter. (C) Similar to (B), but fused with the TATA box of the GAL2 promoter.

Figure 2.

(A) Gene switch activity monitored via GFP expression and measured using flow cytometry. GFP expression levels are represented by the arithmetic means of GFP fluorescence (arbitrary unit). The mean of negative control has been subtracted from that of the samples. Blue columns represent uninduced samples, and red columns represent induced samples. Values are the average of two independent samples, and error bar indicates the standard error. Labels: CTR is negative control with empty plasmids, and Y2H is a yeast two-hybrid system positive control from (26). GE, EG, AGE, AEG, 6GE and 6EG are different combinations and permutations of Gal4 DBD(G), ER LBD(E), P65(6) and Gal4 AD(A) ordered from the N-terminus of the resulting gene switch protein. (B) Inducibility of the YZE strains monitored via GFP expression. The GFP channel histograms are shown. Black indicates negative control without gene switch, blue indicates uninduced sample and red indicates induced sample.

Figure 3.

(A) The inducible pathway assembly scheme. The T–P plasmids have BsaI-excisable terminator and promoter fragments, and the receiver plasmid has BsaI-excisable LacZα marker for blue–white screening. The toolbox plasmids together with the PCR products of the pathway genes can be used to assemble an inducible pathway in a Golden Gate one-pot (GG 1-Pot) reaction. BsaI sites are in light blue, Kan is the kanamycin resistance gene, Spe is the spectinomycin resistance gene. (B) A schematic of the assembled inducible zeaxanthin pathway. All 4 intergenic T–P fragments are identical except for the 4 bp overhang region.

Figure 4.

(A) HPLC traces showing the inducible production of zeaxanthin. (B) Zeaxanthin production in different YZE hosts. Blue columns represent data from the first 24-h samples. At the end of the 24 h, cells were re-inoculated into fresh media for another 24-h. Red columns represent data from the second 24-h samples. The values represent the average of four independent samples, and the error bar represents the standard error.

Figure 5.

(A) Protein-level induction at the different positions using EGFP as the reporter. The genes along the pathway are arranged from 5′ to 3′ in the order Crt E-B-I-Y-Z. Zea is the negative control with the five-gene inducible zeaxanthin pathway without any EGFP gene in it. E, B, I, Y, Z:EGFP have their respective zeaxanthin pathway gene replaced by EGFP. The uninduced EGFP-replaced pathways have identical basal expression levels, and they have been collapsed under crt:EGFP. The values represent the average of four independent samples, and the error bar represents the standard error. (B) mRNA level induction at the different positions as measured by quantitative reverse transcriptase-PCR. The value represents the average of two independent samples, and the error bar represents the standard error. (C) Production time course that benchmarks the inducible pathway in two different YZE strains to a constitutive pathway in YM954. The values represent the average of three independent samples, and the error bar represents standard error. (D) Ligand-titration curve of the zeaxanthin production against the estradiol concentration. The values represent the average of two independent samples, and the error bar represents standard error.