Improved vectors for retron-mediated CRISPR-Cas9 genome editing in Saccharomyces cerevisiae

Abstract

In vivo site-directed mutagenesis is a powerful genetic tool for testing the effects of specific alleles in their normal genomic context. While the budding yeast Saccharomyces cerevisiae possesses classical tools for site-directed mutagenesis, more efficient recent CRISPR-based approaches use Cas 'cutting' combined with homologous recombination of a 'repair' template that introduces the desired edit. However, current approaches are limited for fully prototrophic yeast strains, and rely on relatively low efficiency cloning of short gRNAs. We were thus motivated to simplify the process by combining the gRNA and its cognate repair template in cis on a single oligonucleotide. Moreover, we wished to take advantage of a new approach that uses an E. coli retron (EcRT) to amplify repair templates as multi-copy single-stranded (ms)DNA in vivo, which are more efficient templates for homologous recombination. To this end, we have created a set of plasmids that express Cas9-EcRT, allowing for co-transformation with the gRNA-repair template plasmid in a single step. Our suite of plasmids contains different antibiotic (Nat, Hyg, Kan) or auxotrophic (HIS3, URA3) selectable markers, allowing for editing of fully prototrophic wild yeast strains. In addition to classic galactose induction, we generated a β-estradiol-inducible version of each plasmid to facilitate editing in yeast strains that grow poorly on galactose. The plasmid-based system results in >95% editing efficiencies for point mutations and >50% efficiencies for markerless deletions, in a minimum number of steps and time. We provide a detailed step-by-step guide for how to use this system.

Figure 1.. Maps of new Cas9 and CRISPEY plasmids.

Two-plasmid systems for either galactose induction (A) or estradiol induction (B). The pCRISPEY plasmids contain single restriction enzyme sites for cloning (Gibson assembly), NotI for GAL (A) or XhoI for Z₃ (B). gRNAs and their repair templates for targeted editing are synthesized as oligonucleotides and cloned into the vectors. Editing is be performed by induction from 24 – 72 hours following co-transformation of the pCRISPEY and pCAS9-Ec86 plasmids into the strain of interest.

Figure 2.. Plasmid-encoded Cas9-RT is as effective for editing as genomically encoded.

Cells contained either genomically integrated (at the his3Δ1 locus) or plasmid-encoded galactose inducible Cas9 and Ec86 retron (pCas9-EcRT-GAL), along with pCRISPEY-GAL-ADE2 containing a gRNA and repair template that introduces a frameshift mutation into ADE2. Cells were passaged for the indicated amount of time on galactose-containing selective media and then plated onto YPD to score editing efficiency. Error bars denote the mean and standard deviation of biological triplicates.

Figure 3.. The choice selectable marker has a small effect on editing efficiency.

Cells were transformed with galactose-inducible pCas9-EcRT-GAL and pCRISPEY-GAL-ADE2 plasmids containing the denoted antibiotic resistance markers. Cells were passaged for the indicated amount of time on galactose-containing selective media and then plated onto YPD to score editing efficiency. Error bars denote the mean and standard deviation of biological triplicates.

Figure 4.. High editing efficiencies in wild diploid yeast.

Diploid or haploid derivatives of a lab strain (S288C-derived DBY8268), wild oak (YPS163), vineyard (M22), and clinical (YJM339) were transformed with galactose-inducible pCas9-EcRT-GAL-Hyg and pCRISPEY-GAL-Nat-ADE2 plasmids containing the denoted antibiotic resistance markers. Cells were passaged for the indicated amount of time on galactose-containing selective media and then plated onto YPD to score editing efficiency. Error bars denote the mean and standard deviation of biological triplicates.

Figure 5.. High editing efficiencies with an estradiol-inducible CRISPEY system.

Cells were transformed with β-estradiol-inducible pCas9-EcRT-Z3 and pCRISPEY-Z3-ADE2 plasmids containing the denoted antibiotic resistance markers. Cells were passaged for the indicated amount of time on β-estradiol-containing selective media and then plated onto YPD to score editing efficiency. Error bars denote the mean and standard deviation of biological triplicates.