Construction of a Quadruple Auxotrophic Mutant of an Industrial Polyploid Saccharomyces cerevisiae Strain by Using RNA-Guided Cas9 Nuclease
a,bAssociated Data
Abstract
Industrial polyploid yeast strains harbor numerous beneficial traits but suffer from a lack of available auxotrophic markers for genetic manipulation. Here we demonstrated a quick and efficient strategy to generate auxotrophic markers in industrial polyploid yeast strains with the RNA-guided Cas9 nuclease. We successfully constructed a quadruple auxotrophic mutant of a popular industrial polyploid yeast strain, Saccharomyces cerevisiae ATCC 4124, with ura3, trp1, leu2, and his3 auxotrophies through RNA-guided Cas9 nuclease. Even though multiple alleles of auxotrophic marker genes had to be disrupted simultaneously, we observed knockouts in up to 60% of the positive colonies after targeted gene disruption. In addition, growth-based spotting assays and fermentation experiments showed that the auxotrophic mutants inherited the beneficial traits of the parental strain, such as tolerance of major fermentation inhibitors and high temperature. Moreover, the auxotrophic mutants could be transformed with plasmids containing selection marker genes. These results indicate that precise gene disruptions based on the RNA-guided Cas9 nuclease now enable metabolic engineering of polyploid S. cerevisiae strains that have been widely used in the wine, beer, and fermentation industries.
INTRODUCTION
Auxotrophic mutants of the yeast Saccharomyces cerevisiae have been widely used for both fundamental and applied biology research (1). Among the auxotrophic mutants, ura3, trp1, leu2, and his3 mutants have been frequently employed to introduce genetic perturbations during genetic engineering and molecular biology processes (2). This process includes plasmid-based gene overexpression, one-step gene disruption, and gene integration (3, 4). By and large, most of the yeast gene/genome manipulation tools developed have been based on the auxotrophic mutant and selection marker system. Unfortunately, these efficient and convenient systems are highly restricted to haploid yeast strains but not polyploid yeast strains because of the difficulties in obtaining auxotrophic mutants of these strains (5). Nevertheless, polyploid yeast strains are preferred for industrial applications, as these yeast strains exhibit beneficial traits such as rapid sugar fermentation and higher tolerance of stress conditions (6).
While S. cerevisiae has been intensively employed as a metabolic engineering host for the production of biofuels and value-added chemicals with emerging tools of system biology and synthetic biology (7, 8), polyploid yeast strains have rarely been used as hosts in metabolic engineering because of the difficulties described above. Therefore, a convenient, labor- and time-saving genetic tool for engineering of polyploid strains is highly desired. In order to enable gene manipulation in polyploid yeast strains, dominant drug resistance genes have been used as selection markers to circumvent the absence of auxotrophic mutants (9, 10). An accompanying Cre-loxP technology to recycle these antibiotic markers for multiple genetic perturbations has been developed as well (11). Still, the addition of toxic antibiotic compounds, which may affect cellular physiology, is required to maintain the introduced expression of a target gene. While there is no problem in using the drug resistance marker for the overexpression of a target gene, it is quite difficult to delete multiple alleles of a target gene by using the drug resistance markers with the increased ploidy (12). Moreover, the use of antibiotic markers might lead to the resulting yeast being labeled a genetically modified organism (GMO) for industrial applications (13, 14).
Recently, the bacterial type II CRISPR-Cas system has emerged as a powerful genome-editing tool, being used with great success in various eukaryotic and bacterial hosts (15,–19). Cas9 functions as an RNA-guided endonuclease in these hosts. A double-strand break (DSB) is generated precisely at the guide RNA (gRNA) targeting sequence, followed by the protospacer adjacent motif (PAM) (Fig. 1A). Cotransformation of a 90-mer donor DNA or a 1.4-kb KanMX cassette (conferring G418 resistance) with both a Cas9-expressing plasmid and a gRNA expression plasmid resulted in a nearly 100% positive rate of targeted genome editing in a haploid strain of S. cerevisiae (19). This impressive result motivated us to examine the feasibility of generating auxotrophic mutants of an industrial polyploid S. cerevisiae strain through RNA-guided Cas9 nuclease.
CRISPR-Cas9-mediated gene disruption in polyploid yeast. (A) Diagram of CRISPR-Cas-directed gene disruption. (B) Flow chart for disruption of the auxotrophic genes in the polyploid industrial yeast strain S. cerevisiae ATCC 4124.
In order to enable the expression of Cas9 and gRNA in an industrial S. cerevisiae strain, ATCC 4124 (20, 21), we modified the existing Cas9 and gRNA plasmids by replacing auxotrophic markers (TRP1 and URA3) with drug resistance markers (NAT (nourseothricin resistance) and HyB (hygromycin B resistance). To introduce a designed mutation into a polyploid yeast strain, we expressed Cas9 and a gRNA containing 20-bp sequences targeted to defined sites to create a DSB. While the site-specific DSB generation by Cas9 and the gRNA could be lethal, cotransformation of a 90-bp repair DNA fragment with the designed mutation in the gRNA binding sequences allows the selection of transformants harboring the designed mutation in the genome. For disruption of a target gene, we introduced a TAA stop codon into the repair DNA.
By using this simple procedure, we were able to generate auxotrophic mutants of S. cerevisiae ATCC 4124, which has been reported to be a superior host for metabolic engineering (21).As a result, a quadruple auxotrophic (ura3 trp1 leu2 his3) mutant was obtained. This successful disruption of auxotrophic markers in a polyploid yeast strain not only allows us to exploit numerous preexisting auxotrophic marker-based plasmids in polyploid yeast but also provides a proof of concept of a precise, clean, and labor-saving method of genome editing with the CRISPR-Cas system in yeast.
MATERIALS AND METHODS
Strains and media.
Escherichia coli Top10 was used for the construction and routine propagation of plasmids. E. coli was grown in LB medium (5 g/liter yeast extract, 10 g/liter tryptone, 10 g/liter NaCl, pH 7.0) at 37°C, and ampicillin (100 μg/ml) was added when required. The S. cerevisiae strain used for the disruption of auxotrophic marker genes with the CRISPR-Cas system was polyploid industrial strain ATCC 4124 (20, 21), which was isolated from a molasses distillery. The polyploidy of this strain was confirmed by flow cytometry assay (see Fig. S1 in the supplemental material). Laboratory S. cerevisiae strains D452-2 (MATa leu2 his3 ura3 can1), CEN.PK2-1D (MATα ura3-52 leu2-3,112 trp1-289 his3Δ MAL2-8c SUC2), and BY4742 (MATα his3Δ1 leu2Δ0 lys2Δ0 ura3Δ0) were used as references for evaluation of tolerance of various inhibitor molecules and fermentation capability. Yeast strains were routinely grown on yeast extract-peptone (YP) medium (10 g/liter yeast extract, 20 g/liter peptone) containing glucose (YPD) for transformation or fermentation experiments. Yeast strains transformed with plasmids containing antibiotic markers were propagated on YPD plates supplemented with the corresponding antibiotics. The notations of selection media with antibiotics are as follows. YPDH represents a YPD plate containing 300 μg/ml of hygromycin B, YPDN represents a YPD plate containing 100 μg/ml of nourseothricin, and YPDHN represents a YPD plate containing both hygromycin B and nourseothricin). Synthetic complete (SC) medium minus the appropriate auxotrophic compounds was used for auxotrophic phenotype confirmation.
Plasmid construction.
Plasmid p414-TEF1p-Cas9-CYC1t (no. 43802) carrying the Cas9 cassette was obtained from Addgene Inc. A further modification of this plasmid was made to change a selection marker gene from TRP1 to NAT1 (confers nourseothricin resistance) as follows. First, the NAT1 gene was PCR amplified with primers Cas9-NAT1-U and Cas9-NAT1-D. Second, the PCR fragment obtained was double digested with SnaBI and MfeI and then ligated into plasmid p414-TEF1p-Cas9-CYC1t, which was treated with the same enzymes to generate plasmid Cas9-NAT (see Table 2). The TRP1 gene in the resulting plasmid was disrupted as the NAT1 gene was inserted into the middle of TRP1. Construction of a customized gRNA-expressing plasmid was performed according to the gRNA synthesis protocol of Addgene Inc., with modifications. First, four gRNA cassettes with different sequences for the recognition of TRP1, URA3, HIS3, and LEU2 were synthesized with flanking sequences for SacI and KpnI digestion from IDT Inc. (Table 1). Second, the gRNA DNA fragments synthesized were digested with SacI and KpnI and then ligated into plasmid pRS42H (EUROSCARF), which was digested with the same enzymes to construct plasmids gRNA-ura-HYB, gRNA-trp-HYB, gRNA-leu-HYB, and gRNA-his-HYB, respectively (Table 2).
TABLE 1
Primers and gBlock products used in this study
| Primer/gBlock product | Direction | Sequencea |
|---|---|---|
| Cas9-NAT1-U | Sense | TCTAGAGCGGCCGCTACGTAGATCTGTTTAGCTTGCCTTG |
| Cas9-NAT1-D | Antisense | TCTAGAGCGGCCGCCAATTGCGTTTTCGACACTGGATGGC |
| gRNA-HyB-U | Sense | TCTAGAGCGGCCGCGATATCTCTGTTTAGCTTGCCTTGTC |
| gRNA-HyB-D | Antisense | TCTAGAGCGGCCGCAGGCCTGACACTGGATGGCGGCGTTA |
| gBlock-clone-U | Sense | TCTACAGCGGCCGCGAGCTCTCT |
| gBlock-clone-D | Antisense | TATAGAGCGGCCGCGGTACCAGA |
| URA3donor-U | Sense | TCCATGGAGGGCACAGTTAAGCCGCTAAAGGCATTATAAGCCAAGTACAATTTTTTACTC |
| URA3donor-D | Antisense | ACCAATGTCAGCAAATTTTCTGTCTTCGAAGAGTAAAAAATTGTACTTGGCTTATAATGC |
| TRP1donor-U | Sense | TCCGATGCTGACTTGCTGGGTATTATATGTGTGTAAAATAGAAAGAGAACAATTGACCCG |
| TRP1donor-D | Antisense | TACAAGACTTGAAATTTTCCTTGCAATAACCGGGTCAATTGTTCTCTTTCTATTTTACAC |
| LEU2donor-U | Sense | CCAGGTGACCACGTTGGTCAAGAAATCACAGCCGAAGCCATTAAGTAACTTAAAGCTATT |
| LEU2donor-D | Antisense | ATCGAACTTGACATTGGAACGAACATCAGAAATAGCTTTAAGTTACTTAATGGCTTCGGC |
| HIS3donor-U | Sense | GTAAAGCGTATTACAAATGAAACCAAGATTCAGATTGCGATCTCTTTAAAGGGTTAACCC |
| HIS3donor-D | Antisense | TTCTGGGAAGATCGAGTGCTCTATCGCTAGGGGTTAACCCTTTAAAGAGATCGCAATCTG |
| URA3-U | Sense | GAGCACAGACTTAGATTGGT |
| URA3-D | Antisense | ATAGAAATCATTACGACCGA |
| URA3-Sequencing-U | Sense | TGTGGTGTTGAAGAAACATG |
| URA3-Sequencing-D | Antisense | GATTCCCGGGTAATAACTGA |
| TRP1-U | Sense | GCCGATTAAGAATTCGGTCG |
| TRP1-D | Antisense | GTACAATCAATCAAAAAGCC |
| TRP1-Sequencing-U | Sense | CATTGGTGACTATTGAGCAC |
| TRP1-Sequencing-D | Antisense | CAAAAGGCCTGCAGGCAAGT |
| LEU2-U | Sense | TGACTAAATGCTTGCATCAC |
| LEU2-D | Antisense | GATTGCGTATATAGTTTCGT |
| LEU2-Sequencing-U | Sense | TAGGTGGTTAGCAATCGTCT |
| LEU2-Sequencing-D | Antisense | CTACCCTATGAACATATTCC |
| HIS3-U | Sense | TCCACCTAGCGGATGACTCT |
| HIS3-D | Antisense | TGCATTACCTTGTCATCTTC |
| HIS3-Sequencing-U | Sense | TTAGCGATTGGCATTATCAC |
| HIS3-Sequencing-D | Antisense | CGTATACATACTTACTGACA |
| gBlock-URA3 | TCTACAGCGGCCGCGAGCTCTCTTTGAAAAGATAATGTATGATTATGCTTTCACTCATATTTATACAGAAACTTGATGTTTTCTTTCGAGTATATACAAGGTGATTACATGTACGTTTGAAGTACAACTCTAGATTTTGTAGTGCCCTCTTGGGCTAGCGGTAAAGGTGCGCATTTTTTCACACCCTACAATGTTCTGTTCAAAAGATTTTGGTCAAACGCTGTAGAAGTGAAAGTTGGTGCGCATGTTTCGGCGTTCGAAACTTCTCCGCAGTGAAAGATAAATGATCGAGTAAAAAATTGTACTTGGGTTTTAGAGCTAGAAATAGCAAGTTAAAATAAGGCTAGTCCGTTATCAACTTGAAAAAGTGGCACCGAGTCGGTGGTGCTTTTTTTGTTTTTTATGTCTGGTACCGCGGCCGCTCTATA | |
| gBlock-TRP1 | TCTACAGCGGCCGCGAGCTCTCTTTGAAAAGATAATGTATGATTATGCTTTCACTCATATTTATACAGAAACTTGATGTTTTCTTTCGAGTATATACAAGGTGATTACATGTACGTTTGAAGTACAACTCTAGATTTTGTAGTGCCCTCTTGGGCTAGCGGTAAAGGTGCGCATTTTTTCACACCCTACAATGTTCTGTTCAAAAGATTTTGGTCAAACGCTGTAGAAGTGAAAGTTGGTGCGCATGTTTCGGCGTTCGAAACTTCTCCGCAGTGAAAGATAAATGATCGTCAATTGTTCTCTTTCTATGTTTTAGAGCTAGAAATAGCAAGTTAAAATAAGGCTAGTCCGTTATCAACTTGAAAAAGTGGCACCGAGTCGGTGGTGCTTTTTTTGTTTTTTATGTCTGGTACCGCGGCCGCTCTATA | |
| gBlock-LEU2 | TCTACAGCGGCCGCGAGCTCTCTTTGAAAAGATAATGTATGATTATGCTTTCACTCATATTTATACAGAAACTTGATGTTTTCTTTCGAGTATATACAAGGTGATTACATGTACGTTTGAAGTACAACTCTAGATTTTGTAGTGCCCTCTTGGGCTAGCGGTAAAGGTGCGCATTTTTTCACACCCTACAATGTTCTGTTCAAAAGATTTTGGTCAAACGCTGTAGAAGTGAAAGTTGGTGCGCATGTTTCGGCGTTCGAAACTTCTCCGCAGTGAAAGATAAATGATCAATCACAGCCGAAGCCATTAGTTTTAGAGCTAGAAATAGCAAGTTAAAATAAGGCTAGTCCGTTATCAACTTGAAAAAGTGGCACCGAGTCGGTGGTGCTTTTTTTGTTTTTTATGTCTGGTACCGCGGCCGCTCTATA | |
| gBlock-HIS3 | TCTACAGCGGCCGCGAGCTCTCTTTGAAAAGATAATGTATGATTATGCTTTCACTCATATTTATACAGAAACTTGATGTTTTCTTTCGAGTATATACAAGGTGATTACATGTACGTTTGAAGTACAACTCTAGATTTTGTAGTGCCCTCTTGGGCTAGCGGTAAAGGTGCGCATTTTTTCACACCCTACAATGTTCTGTTCAAAAGATTTTGGTCAAACGCTGTAGAAGTGAAAGTTGGTGCGCATGTTTCGGCGTTCGAAACTTCTCCGCAGTGAAAGATAAATGATCATTGCGATCTCTTTAAAGGGGTTTTAGAGCTAGAAATAGCAAGTTAAAATAAGGCTAGTCCGTTATCAACTTGAAAAAGTGGCACCGAGTCGGTGGTGCTTTTTTTGTTTTTTATGTCTGGTACCGCGGCCGCTCTATA |
TABLE 2
Plasmids used in this study
| Plasmid | Description | Marker | Source |
|---|---|---|---|
| p414-TEF1p-Cas9-CYC1t | Cas9 expression plasmid | TRP1 | Addgene Inc. |
| pRS42H | HyB | EUROSCARF | |
| Cas9-NAT | Cas9 expression plasmid | NAT1 | This study |
| gRNA-ura-HYB | URA3 disruption gRNA cassette | HyB | This study |
| gRNA-trp-HYB | TRP1 disruption gRNA cassette | HyB | This study |
| gRNA-leu-HYB | LEU2 disruption gRNA cassette | HyB | This study |
| gRNA-his-HYB | HIS3 disruption gRNA cassette | HyB | This study |
| pAG25 | PCR template for NAT1 cassette | NAT1 | EUROSCARF |
Procedures for auxotrophic marker disruption in a polyploid yeast.
Yeast transformation was carried out by a standard lithium acetate transformation method (22). A detailed auxotrophic marker disruption procedure is described below, and the schematic map is shown in Fig. 1B. In step 1, plasmid Cas9-NAT (∼200 ng per transformation) was transformed into the prototrophic S. cerevisiae ATCC 4124 strain. After transformation of plasmid Cas9-NAT, cells were plated on selective medium (YPDN plate) and allowed to grow for 2 to 3 days until colonies were ready to pick. The resulting strain was named ATCC 4124 (Cas9). In step 2, a double-stranded 90-mer oligonucleotide repair DNA for URA3 disruption was PCR amplified with primers URA3donor-U and URA3donor-D (Table 1). These primers have a 30-mer overlapping region encompassing the PAM sequence and a 30-mer overhang. A TAA stop codon was incorporated into the PAM sequence to disrupt the URA3 gene and, at the same time, prevent repetitive Cas9 cleavage of the target site. The gRNA-ura-HYB plasmid (∼500 ng) was then transformed together with the 90-mer donor DNA (∼2 μg) into Cas9-expressing ATCC 4124 (Cas9). Cells were plated on a YPDHN plate and allowed to grow for 2 to 3 days until transformants were ready to pick. In step 3, all of the colonies were replica plated onto an SC-ura plate to screen for ura3 auxotrophic mutants. The sequence of the URA3 gene was confirmed by sequencing of the PCR product with primers URA3-U and URA3-D (Table 1). The confirmed ura3 auxotrophic mutant was named ATCC 4124 Δura3 (Cas9/gRNA-ud). In step 4, in order to drop out the gRNA-ura-HYB plasmid for disruption of the next auxotrophic marker, ATCC 4124 Δura3 (Cas9/gRNA-ud) was cultured in liquid YPDN medium for 24 h and then streaked onto a YPDN plate for single colonies, which were then replica plated onto a YPDH plate to confirm the release of the plasmids. In step 5, steps 2 to 4 were repeated to achieve sequential disruption of the TRP1, LEU2, and HIS3 genes with their corresponding gRNA plasmids and donor DNA. The sequences of the primers used for disruption are provided in Table 1.
Spotting assay for evaluation of fermentation inhibitor tolerance.
Yeast cells were grown in 5 ml of YPD containing glucose at 20 g/liter overnight. After the optical density at 600 nm was adjusted to 1, 10-fold serial dilutions in water were performed. For controls, 5-μl volumes of 10−1-, 10−2-, 10−3-, 10−4-, 10−5-, and 10−6-diluted cell suspensions were spotted onto YPD agar plates containing glucose at 20 g/liter. Plates were incubated at 30°C for 1 day. A heat tolerance test was performed by the same method, but the plate was incubated at 40°C for 2 days. To evaluate tolerance of the fermentation inhibitors (furfural and 5-hydroxymethylfurfural [5-HMF]) present in most cellulosic hydrolysates, YPD plates containing glucose at 20 g/liter with furfural at 1 g/liter or 5-HMF at 3 g/liter were used. The inhibitor-containing plates were incubated at 30°C for 2 days.
Fermentation and metabolite analysis.
Batch fermentation experiments were performed by inoculating yeast cells grown on YPD or SCD into 20 ml of fermentation medium (YP medium containing glucose at 120 g/liter or SC medium containing glucose at 40 g/liter) in a 100-ml Erlenmeyer flask at an initial cell optical density of ∼1 (corresponding to approximately 0.24 g/liter [dry cell mass]). All of the flasks were sealed with aluminum foil, incubated at 30°C and 100 rpm under oxygen-limited conditions. Samples were taken at appropriate intervals and then centrifuged at 15,000 rpm for 5 min. The supernatants were diluted appropriately and then used for chromatographic analysis. Metabolites were analyzed with a high-performance liquid chromatograph (Agilent Technologies 1200 Series) equipped with a refractive-index detector. Metabolites such as glucose, glycerol, acetate, and ethanol were measured with a Rezex ROA-Organic Acid H+ (8%) column (Phenomenex Inc., Torrance, CA). Columns were eluted with 0.005 N H2SO4 at 50°C, and the flow rate was set at 0.6 ml/min.
RESULTS
Sequential disruption of auxotrophic markers through CRISPR-Cas9 genome editing.
CRISPR-Cas9-based genome editing has been demonstrated in haploid S. cerevisiae strain S288C, in which CAN1 and ADE2 were disrupted (19). The considerably high deletion efficiency of the CRISPR-Cas system in a haploid strain motivated us to examine the feasibility of generating auxotrophic mutants through the deletion of multiple alleles of genes in a polyploid industrial yeast strain. Therefore, we attempted to disrupt four popular auxotrophic marker genes (LEU2, TRP1, URA3, and HIS3) in S. cerevisiae ATCC 4124, which is a polyploid industrial strain. In order to use the existing constructs of the CRISPR-Cas9 system (19) in a prototrophic industrial strain, we inserted the NAT1 antibiotic marker into the TRP1 auxotrophic marker in the Cas9 plasmid and then constructed gRNA plasmids with vector pRS42H harboring the HyB marker. These two modifications enabled the transformation of Cas9 and gRNA expression cassettes into prototrophic industrial strains.
After the sequential deletion procedures listed in Materials and Methods were performed, four auxotrophic mutant strains, ATCC 4124 Δura, ATCC 4124 Δura Δtrp, ATCC 4124 Δura Δtrp Δleu, and ATCC 4124 Δura Δtrp Δleu Δhis, were obtained. Positive colonies expected to harbor the designed mutations after transformation of the gRNA-expressing cassettes and corresponding repair DNA fragments were 15 to 60% of the transformants when randomly picked colonies were tested with SC plates lacking appropriate auxotrophic nutrients. Representative colonies were picked at each stage of the sequential gene disruption procedure to confirm their auxotrophy on various dropout media (Fig. 2). Additionally, the mutated sequences in each strain were determined by sequencing the PCR products of the mutated regions from three randomly picked colonies. As shown in Fig. 3, all of the selected colonies harbored the expected TAA stop codon conferred by the double-strained donor DNA at the previous PAM sequence location, which disrupted the function of the marker genes and, at the same time, prevented further nonspecific cleavage by Cas9 at those sites.
Auxotrophy of the resulting ATCC 4124 strains. After each round of disruption, auxotrophy was confirmed by plating the strains onto SCD plates minus the appropriate auxotrophic compounds.
Alignment of the sequenced marker gene disruption regions. The underlined sequence is the wild-type marker gene region. The PAM sequence, the incorporated TAA stop codon, and the gRNA guiding sequence are highlighted. The letter r in parentheses stands for reverse complement.
After obtaining four auxotrophic mutants, we tested whether these mutants could be converted into prototrophic strains by transforming them with pRS plasmids containing the corresponding auxotrophic marker genes (23). The growth of all of the mutants on minimal medium lacking uracil, tryptophan, histidine, and leucine could be restored (data not shown), suggesting that four auxotrophic mutants are transformable with complementing plasmids.
Beneficial phenotypes of an industrial S. cerevisiae strain are retained in the quadruple auxotrophic mutant.
Industrial S. cerevisiae strains are known to exhibit desirable phenotypes for industrial applications. Tolerance of the fermentation inhibitors prevalent in cellulosic hydrolysates is an important trait for producing cellulosic biofuels. Of these fermentation inhibitors, furfural and 5-HMF are known to inhibit the metabolism, as well as the growth, of yeast cells (24). Therefore, higher tolerance of furfural and 5-HMF is a prerequisite for a cellulosic biofuel producer. To compare the tolerance of the quadruple mutant with that of its parental strain and other laboratory S. cerevisiae strains, a growth-based assay was employed to evaluate stress tolerances in the presence of toxic levels of furfural and 5-HMF (Fig. 4). Both parental strain S. cerevisiae ATCC 4124 and the quadruple auxotrophic mutant exhibited furfural and 5-HMF tolerance higher than that of laboratory S. cerevisiae strains (D452-2, CEN-PK2-1D, and BY4742).
Spotting assays. The ATCC 4124 prototroph, the ATCC 4124 auxotroph, and three laboratory strains were characterized for tolerance of fermentation inhibitors and heat stress.
We also examined the thermotolerance of the industrial and laboratory strains. Again, the quadruple auxotrophic mutant grew as well as parental S. cerevisiae ATCC 4124, even at 40°C. While CEN-PK2-1D was able to grow at 40°C, two laboratory strains (D452-2 and BY4742) did not.
These results suggest that the prototrophic and auxotrophic ATCC 4124 strains showed similar tolerance of the fermentation inhibitors and high temperature, indicating that disruption of auxotrophic markers by the CRISPR-Cas system did not affect the inherent superior tolerance phenotypes of S. cerevisiae ATCC 4124.
Comparison of rates of glucose fermentation by ATCC 4124 prototroph, ATCC 4124 auxotroph, and laboratory strains.
During the production of biofuel or other value-added chemicals by yeast strains, rapid and efficient sugar fermentation is critical for the implementation of economic processes. This is the major reason why industrial polyploid yeasts are preferred as host strains for metabolic engineering. In order to investigate whether or not deletions of auxotrophic marker genes impact the rapid sugar-fermenting phenotype of S. cerevisiae ATCC 4124, a fermentation assay with YP medium containing glucose at 120 g/liter was carried out to evaluate the fermentation capability of industrial and laboratory strains. As shown in Fig. 5, in terms of glucose consumption rates, the ATCC 4124 prototroph and the lab strains could deplete glucose within 9 and 15 h, respectively, while the ATCC 4124 auxotroph depleted glucose within 12 h. A similar trend in ethanol productivity was observed, and a decent difference in growth rates was shown. The growth rate of the ATCC 4124 auxotroph was similar to that of the laboratory strains. These results suggest that the disruption of auxotrophic markers in the ATCC 4124 prototroph slowed down its sugar metabolism slightly. However, the resulting ATCC 4124 auxotroph still significantly outperformed all of the laboratory strains tested here.
Glucose consumption (A), ethanol production (B), and cell growth (C) rates of the ATCC 4124 prototroph, the ATCC 4124 auxotroph, and three laboratory strains were evaluated in YP medium containing glucose at 120 g/liter. Filled circles, ATCC 4124 prototroph; open circles, ATCC 4124 auxotroph; squares, BY4742; triangles, CEN.PK; diamonds, D452-2. Data are presented as mean values and standard deviations of three independent biological replicates.
Restoration of four auxotrophic marker genes can rescue the growth defect phenotypes of the ATCC 4124 auxotroph.
As our goal was to exploit numerous preexisting plasmids based on auxotrophic markers for metabolic engineering after we obtained the auxotrophic industrial strain, we investigated the fermentation performance of the ATCC 4124 prototroph and quadruple auxotroph strains before and after the restoration of four marker genes by transformation with four auxotrophic marker-based plasmids (pRS42X series). Fermentation assays of the strain before restoration of the marker genes were carried out with SC medium containing glucose at 40 g/liter and those of the strain after marker gene restoration were carried out with SC-4 (uracil, tryptophan, leucine, and histidine dropout) medium containing glucose at 40 g/liter. As shown in Fig. 6A to toC,C, the fermentation profiles of the ATCC 4124 prototroph and auxotroph in SC medium showed trends similar to those seen in YP medium. The auxotrophic strain was overall somewhat inferior to the prototrophic strain. However, according to Fig. 6E and andF,F, this defective-growth phenotype of the auxotrophic strain can be rescued to a great extent by marker gene restoration. These results suggest that the inferior phenotype of the ATCC 4124 auxotroph during the fermentation process was due almost exclusively to the disruption of the marker genes. On the other hand, these results indicate that the CRISPR-Cas technology is precise and clean.
Glucose consumption (A, D), ethanol production (B, E), and cell growth (C, F) rates of the ATCC 4124 prototroph and its quadruple auxotroph were evaluated in SC medium (A to C) or SC-ura-trp-leu-his medium (D to F) containing glucose at 40 g/liter. Filled circles, ATCC 4124 prototroph; open circles, ATCC 4124 quadruple auxotroph; triangles, marker-restored ATCC 4124 quadruple auxotroph. Data are presented as mean values and standard deviations of three independent biological replicates.
DISCUSSION
In order to enable the use of numerous existing plasmid systems with auxotrophic markers in industrial yeast strains, a quick, labor-saving, and universal method for auxotrophic marker generation in polyploid S. cerevisiae strains is highly desired. However, because of the presumed polyploidy of most industrial yeast strains (25, 26), regular gene deletion methods, which have been routinely used with laboratory S. cerevisiae strains to generate auxotrophic markers, are not applicable to polyploid S. cerevisiae strains. The target gene deletion efficiencies of the regular method are not high enough to achieve the deletion of multiple alleles and overcome high frequencies of reversion of the mutated allele through homologous recombination with the wild-type allele from extra copies of the genome (27). Recently, transcriptional-activator-like effector nuclease (TALENs) has been widely adopted as a genome-editing tool that outperforms the traditional methods (28,–30). The application of TALENs in haploid yeast was shown to be feasible by Ting Li et al. (31). We also performed URA3 gene disruption with TALENs in a haploid laboratory strain and a polyploid industrial strain. In the case of the laboratory strain, a consistent result was observed as previously reported (31). The TALENs targeting URA3 can be used for targeted gene deletion, but the frequency at which positive colonies were obtained was 10%, even with a haploid strain. In the case of a polyploid industrial strain, we failed to obtain a correct deletion mutant with the TALENs. Recently, the CRISPR-Cas system has emerged as a powerful tool for genome engineering in bacterial, S. cerevisiae yeast, and even human cells (17, 19, 32). Especially for a haploid S. cerevisiae strain, the frequency at which positive colonies were obtained reached almost ∼100% for CAN1 deletion when the gRNA cassette was carried by a multicopy plasmid and was under the control of a constitutive yeast promoter (19). Thus, we presumed that the high efficiency of the CRISPR-Cas system would be able to help us achieve gene disruption in polyploid yeast strains.
After each round of targeted gene disruption in polyploid S. cerevisiae strain ATCC 4124, positive colonies harboring the intended gene disruption were roughly in the range of 15 to 60%. These frequencies are lower than the previously reported ∼100% for gene disruption in a haploid strain. While the targeted gene disruption efficiency of the CRISPR-Cas system was indeed affected by the increased ploidy number, 15 to 60% is still a fairly impressive percentage that could significantly reduce the laborious screening process of other marker gene disruption strategies, which require either large-scale screening (frequencies ranged from 10−4 to 10−3) or time-consuming procedures (27, 33). In addition, CRISPR-Cas directed gene disruption does not leave a scar or undesirable selection marker in the yeast genome. After target gene disruption, only a TAA stop codon was inserted. Additionally, both the Cas9 and gRNA plasmids could be easily dropped out of the host strain by a 24-h culture under nonselective culture conditions (e.g., YPD medium), a favorable strategy considering GMO issues in using engineered industrial yeast strains for food and agricultural applications.
An important task for producing fuels and chemicals from cellulosic biomass is to secure a robust metabolic engineering host strain that already exhibits desirable traits. As polyploid S. cerevisiae strains display numerous beneficial phenotypes for industrial applications, direct genetic manipulation of the polyploid yeast could be a promising approach. Our spotting assays clearly proved that industrial strain ATCC 4124 possesses much higher resistance to fermentation inhibitors and high temperature than three laboratory strains, regardless of their auxotrophy. This result indicated that the CRISPR-Cas system is reliable and not highly mutagenic. However, it is noteworthy that all of the auxotrophic mutants of strain ATCC 4124 were slightly inferior to their isogenic parental strain in terms of glucose fermentation. However, this was expected because there had been many reports regarding the growth defect of auxotrophic strains. Even standard YPD could not compensate for the growth defects of S. cerevisiae auxotrophic strains (34, 35). Nevertheless, the ATCC 4124 auxotroph still performed significantly better than three laboratory strains. Furthermore, the restoration of four marker genes was able to rescue the inferior phenotype of the ATCC 4124 auxotroph in our SC medium fermentation assay, which consolidated the practical application of this strain and, on the other hand, proved that the CRISPR-Cas system is not highly mutagenic. In conclusion, we have demonstrated the feasibility of developing a precise, high efficiency, scarless, and labor-saving method to generate auxotrophic markers in polyploid industrial strains by using the state-of-the-art CRISPR-Cas system.
ACKNOWLEDGMENTS
We are grateful to Timothy L. Turner (University of Illinois at Urbana-Champaign) for his proofreading and suggestions on the writing.
This work was supported by funding from the Energy Biosciences Institute.
Footnotes
Published ahead of print 3 October 2014
Supplemental material for this article may be found at http://dx.doi.org/10.1128/AEM.02310-14.