Tips, Tricks, and Potential Pitfalls of CRISPR Genome Editing in Saccharomyces cerevisiae

Abstract

The versatility of clustered regularly interspaced short palindromic repeat (CRISPR)-associated (Cas) genome editing makes it a popular tool for many research and biotechnology applications. Recent advancements in genome editing in eukaryotic organisms, like fungi, allow for precise manipulation of genetic information and fine-tuned control of gene expression. Here, we provide an overview of CRISPR genome editing technologies in yeast, with a particular focus on Saccharomyces cerevisiae. We describe the tools and methods that have been previously developed for genome editing in Saccharomyces cerevisiae and discuss tips and experimental tricks for promoting efficient, marker-free genome editing in this model organism. These include sgRNA design and expression, multiplexing genome editing, optimizing Cas9 expression, allele-specific editing in diploid cells, and understanding the impact of chromatin on genome editing. Finally, we summarize recent studies describing the potential pitfalls of using CRISPR genome targeting in yeast, including the induction of background mutations.

Keywords: Cas9; background mutagenesis; biotechnology; genome engineering; guide RNA; off-target mutagenesis; synthetic biology; yeast.

FIGURE 1

Overview of Cas9 (or dCas9) targeting mechanism. Cas9 (or dCas9) recognizes and binds DNA using sgRNA. Briefly, Cas9:sgRNA complexes allow the PAM-interacting domain in Cas9 (or dCas9) to sample PAM motifs in the genome (Step 1). The Cas9:sgRNA complex also aligns the sgRNA for invasion of the target DNA. Following PAM sampling, recognition, and subsequent binding, the DNA duplex is locally unwound and sgRNA begins pairing with complementary target DNA (Step 2). Base pairing in the seed region (i.e., the nucleobases immediately adjacent to the PAM motif) is followed by the unidirectional propagation of the DNA:sgRNA heteroduplex (i.e., R-loop) away from the PAM motif (Step 3). Propagation and the subsequent formation of a stable Cas9-induced R-loop is coupled with conformational changes in Cas9 which dictate DNA cleavage activity by the Cas9 endonuclease. Completely paired R-loops for active Cas9 (i.e., 20 bp) result in double strand breaks (DSB) 3–4bp from the PAM motif (i.e., Full heteroduplex). The subsequent repair of these Cas9-induced DNA breaks results in the typical mutations observed during genome editing. Alternatively, Cas9 proteins that are involved in base editing or transcriptional modulation will result in deamination activity or gene activation or repression, respectively (Step 4).

FIGURE 2

DSB repair in S. cerevisiae allows for marker-free genome editing. The high efficiency of homologous recombination in yeast allows homologous donor templates such as oligonucleotides or PCR products to be efficiently incorporated into yeast cells. However, if cells escape Cas9 editing or editing occurs outside the PAM motif or sgRNA targeting space, it is possible that Cas9 could continually target DNA and iteratively form DSBs in the yeast genome. Additionally, this could lead to a low frequency of non-homologous end joining repair of DSBs, which generates random indels at the Cas9 target site. Importantly, the toxicity of Cas9-induced DSBs allows for strong selection against unedited cells because unedited cells typically die, especially if many DSBs are being generated under continual and repeated Cas9 targeting. This highlights a critical parameter for marker-free genome editing in yeast in that desired genomic edits can be obtained by selecting for yeast cells that actively express Cas9 and sgRNA. Moreover, this demonstrates how PAM inactivating mutations, which abrogate CRISPR/Cas9 targeting can be used to minimize the potential detrimental effects associated with excessive and iterative generation of DSBs while also allowing for correct genomic edits to still be enriched for under selective pressure (see key references: Garst et al., 2017; Gorter de Vries et al., 2019; Laughery and Wyrick, 2019).

FIGURE 3

Overview of Cas9 specificity and activity in eukaryotic chromatin. (A) Schematic describing the impact of PAM accessibility in nucleosome on Cas9 cleavage dynamics. Nucleosomes occlude Cas9 access to PAM motifs within chromatin reducing Cas9 binding and cleavage activity (arrow 1). PAM motifs within linker DNA, which are located outside the nucleosomes making up the chromatin are accessible and thus get cleaved strongly (arrow 3). PAM motifs located at the entry/exit sites of nucleosomes (arrow 2) display variable Cas9 activity because Cas9 activity is influenced by PAM orientation and nucleosome breathing dynamics for these PAM positions. Importantly, PAM motifs oriented away from the nucleosome at the entry/exit sites are typically cleaved better than inward facing PAM motifs. (B) Schematic showing how Cas9 targeting (i.e., binding and cleavage). is more efficient in open and transcriptionally active euchromatic regions relative to silent, transcriptionally silent heterochromatin regions.

FIGURE 4

Tips and tricks for expressing sgRNA in S. cerevisiae. (A) Schematic with overview of important experimental considerations for how sgRNA is expressed in yeast. sgRNA is typically expressed from vectors with low-copy number origins of replication like CEN/ARS or from high copy number origins of replication like 2 µ. Promoters for sgRNA expression typically involve either constitutive or inducible expression of individual or multiple sgRNAs from either a RNA Pol III promoter, RNA Pol II promoter, or a synthetic promoter. Lastly, sgRNAs that get expressed in yeast cells are typically transcribed from plasmid-based vectors or from elements that were directly integrated into the yeast genome. (B) Examples of sgRNA expression vector architecture. URA3 is a commonly used selection marker for sgRNA expression vectors and can be used to select for transformants as well as to remove sgRNA expression machinery from yeast cells. LEU2 markers have also been used. BclI (3’ of promoter) and SwaI (in the sgRNA) restriction sites are used for simplifying construction of new targeting sgRNAs in sgRNA vectors. This is done by digesting the plasmid, then hybridizing a user-defined 20 nt DNA segment with a 5’ GATC overhang to facilitate the subsequent ligation of this cassette into the plasmid. (C) Schematic describing general approach for multiplexing CRISPR/Cas9 genome editing experiments in yeast. One key consideration is whether a plasmid-based or integration approach is going to be utilized. Plasmid-based approaches will typically involve pooled systems in which many different sgRNAs are expressed from separate promoters within the same construct or the different sgRNAs are generated upon independent processing by ribozymes or other RNA processing enzymes. Strategies employing gRNAs that are flanked by cleavage RNA sequences make use of self-cleavable ribozyme sequences (e.g., Hammerhead ribozyme and HDV ribozyme), exogenous cleavage factor recognition sequences (e.g., Cys4), and endogenous RNA processing sequences (e.g., tRNA sequences and introns). For more details see Stovicek et al., 2017; Lian et al., 2018; Raschmanová et al., 2018; Ding et al., 2020; Malcı et al., 2020; Utomo et al., 2021; Zhang et al., 2021.

FIGURE 5

Tips and tricks for expressing Cas9 in S. cerevisiae (A) Schematic with overview of important experimental considerations for how Cas9 is expressed in yeast. Cas9 is typically expressed from vectors with low-copy number origins of replication like CEN/ARS or from high copy number origins of replication like 2 µ. Promoters for Cas9 expression typically involve either strong or weak promoters. The choice of constitutive or inducible expression and selection of promoter strength for Cas9 are important considerations when looking to avoid potential Cas9-induced toxicity throughout the genome. A nuclear localization signal (NLS) is required to localize Cas9 in the nucleus of eukaryotic cells like S. cerevisiae. Importantly, Cas9 expression vectors have been used to introduce both Cas9 enzymes as well as Cas12a/Cpf1 enzymes in S. cerevisiae. Another consideration for optimizing Cas9 function in cells is codon optimization. Lastly, Cas9 expression vectors require terminators to stop transcription of Cas9. (B) Examples of Cas9 expression vector architecture. LEU2 and TRP1 are a commonly used selection markers for Cas9 expression vectors and can be useful to select for transformants as well as removal of Cas9 machinery from yeast cells. For more details see Stovicek et al., 2017; Raschmanová et al., 2018; Ding et al., 2020.

FIGURE 6

Model describing the molecular mechanisms associated with Cas9 mutagenesis in S. cerevisiae. R-loop formation and DNA binding by Cas9 endonuclease can inhibit endogenous cellular processes such as BER, NHEJ, DNA replication and RNA transcription resulting in distinct on- and off-target mutational profiles for both active Cas9 and the dCas9/nCas9 enzymes that are used in either base editing or transcriptional regulation applications. Each of the factors influencing mutagenesis at the top of figure can be generated upon R-loop formation by dCas9. Base substitutions (ex. C to T or C to G) can be explained by cytosine deamination and the subsequent activity of BER or TLS enzymes. Complex mutations such as indels and structural variations can be explained, in part, by targeting homopolymer sequences with dCas9 as well as replication stress from bound dCas9 (bottom of figure). Additionally, inhibition of other cellular processes, such as BER, NHEJ, and RNA polymerases are likely to impact both on- and off-target mutational outcomes with Cas9 proteins. For example, the extent to which inhibition of RNA polymerase by Cas9 binding impacts mutagenesis is likely reflected in how readily RNA polymerase can dislodge bound Cas9 from a genomic target site. Notably, RNA polymerase is more likely to dislodge bound Cas9 when it is targeted to the transcribed strand. This likely impacts Cas9 activity in cells by allowing Cas9 to better function as a multi-turnover enzyme in vivo. The PAM motif is indicated by the green box. For more details see Gilbert et al., 2013; Clarke et al., 2018; Laughery et al., 2019; Doi et al., 2021; Antony et al., 2022.