Elsevier

Methods in Enzymology

Volume 546, 2014, Pages 473-489
Methods in Enzymology

Chapter Twenty-Two - Multiplex Engineering of Industrial Yeast Genomes Using CRISPRm

https://doi.org/10.1016/B978-0-12-801185-0.00023-4Get rights and content

Abstract

Global demand has driven the use of industrial strains of the yeast Saccharomyces cerevisiae for large-scale production of biofuels and renewable chemicals. However, the genetic basis of desired domestication traits is poorly understood because robust genetic tools do not exist for industrial hosts. We present an efficient, marker-free, high-throughput, and multiplexed genome editing platform for industrial strains of S. cerevisiae that uses plasmid-based expression of the CRISPR/Cas9 endonuclease and multiple ribozyme-protected single guide RNAs. With this multiplex CRISPR (CRISPRm) system, it is possible to integrate DNA libraries into the chromosome for evolution experiments, and to engineer multiple loci simultaneously. The CRISPRm tools should therefore find use in many higher-order synthetic biology applications to accelerate improvements in industrial microorganisms.

Introduction

For thousands of years, humans have domesticated Baker's yeast Saccharomyces cerevisiae for the production of alcohol and bread. More recently, global demand has driven the use of industrial strains of S. cerevisiae for large-scale production of biofuels and renewable chemicals (Farrell et al., 2006, Rubin, 2008). However, the genetic basis of desired domestication traits is poorly understood because robust genetic tools do not exist for industrial production hosts. Industrial S. cerevisiae strains are more stress tolerant and produce much higher yields of desired biofuel or renewable chemical end products than laboratory strains. However, linking genotypes of industrial yeasts to their phenotypes remains difficult because these strains are often polyploid with low-efficiency mating and sporulation. The standard genetic tool of integrating linear DNA into the genome by homologous recombination (HR) is too inefficient for the creation of loss-of-function phenotypes in these strains, and current technologies rely on dominant selectable markers for chromosomal integrations or plasmid maintenance. Since only a small number of markers exist, deciphering and improving important complex multigenic phenotypes in industrial strains remains a challenge.

Due to the lack of genetic tools, most industrially relevant phenotypes must be tested in haploid derivatives of the industrial strains or in lab strains. These haploid derivatives may not cosegregate the alleles required for the relevant phenotype, particularly if the phenotype is complex, so in many cases they are not ideal surrogates for industrial isolates. Phenotypes observed in one segregant may not be similar to the others. Therefore, phenotypes are best tested within the relevant industrial strain in the exact state to which it is found in the industrial process. Further, lab strains do not act as good proxies for strain-specific phenotypes, as even the most straightforward phenotypes such as essentiality in rich medium can differ substantially between two laboratory strains due to complex genetics with unpredictable allelic combinations (Dowell et al., 2010).

Present technologies for heterologous gene expression by integrating genes into yeast chromosomes require the recombination and cointegration of the gene to be expressed and a dominant selectable marker to identify cells with the integrated DNA. The efficiency of chromosomal integration is low and homozygous integrations need to be made by iteratively incorporating the gene with a different selectable marker. Further complicating matters, the gene integrated second may replace the first integration by recombination. Finally, at the end of the process, the selectable markers need to be removed before the engineered yeast can be used in an industrial setting (Solis-Escalante, Kuijpers, van der Linden, Pronk, & Daran-Lapujade, 2014). Therefore, it is difficult, time-consuming and labor-intensive to generate homozygous integrations in diploid (or higher ploidy) yeast strains. Ideally, an experimenter needs a targeting method that does not require an integrated marker and precisely cuts all chromosomes without the requirement of any premade genetic modifications to the cell, such as auxotrophic markers. A system such as this would be ready for use in any industrial, wild or unmodified isolate, including those with higher chromosome copy number.

Bacterial type II CRISPR/Cas9 genome editing has been used successfully in several eukaryotic organisms but has not been adapted for genome-wide studies or for heterologous protein engineering in industrially important eukaryotic microbes. CRISPR/Cas systems require a Cas9 endonuclease that is targeted to specific DNA sequences by a noncoding single guide RNA (sgRNA) (Jinek et al., 2012). The Cas9–sgRNA ribonucleoprotein complex precisely generates double-strand breaks (DSBs) in eukaryotic genomes at sites specified by a twenty-nucleotide guide sequence at the 5′ end of the sgRNA that base pairs with the protospacer DNA sequence preceding a genomic Protospacer adjacent motif (PAM) (Sternberg, Redding, Jinek, Greene, & Doudna, 2014). Repair by nonhomologous end joining results in small deletions or insertions in the genome 5′ of the PAM motif (Cong et al., 2013, Mali, Esvelt and Church, 2013, Mali, Yang, et al., 2013). Alternatively, the presence of the Cas9-produced DSB in genomic DNA can increase the rate of HR with linear DNA at the DSB locus by several thousand-fold (DiCarlo et al., 2013), potentially enabling high-throughput genetic studies.

CRISPR gene targeting lends itself to the efficient targeting of yeast genomes, for both loss-of-function analysis and heterologous gene expression. In the yeast system developed in our lab, the CRISPR/Cas9 endonuclease is coexpressed with multiple ribozyme-protected sgRNAs (Fig. 22.1). This system enables efficient, marker-free, single step, and multiplexed genome editing in industrial strains of S. cerevisiae. The power of multiplex CRISPR (CRISPRm) can be used to accelerate discoveries of the genetic and molecular determinants of improved industrial microorganisms. Further, CRISPRm can be used in any prototrophic yeast isolate, making the system plug-and-play ready. Unlike other systems, CRISPRm requires no previous genetic modifications (Wingler & Cornish, 2011) and the drug-resistant plasmid used to coexpress the Cas9 protein and sgRNA confers dominant drug resistance. Because there is no fitness advantage in nonselective conditions the Cas9 plasmid (pCAS) is readily lost in rich medium shortly after the drug has been removed from the medium. CRISPRm results in homozygous mutants of yeast cells with diploid (or higher) copy number; to date we have not recovered heterozygous mutants. This is likely the case because Cas9 protein will cut all of the targeted chromosomes. We suggest that the linear DNA is used as a homology directed repair template to correct one chromosome and then that repaired chromosome is then used as DNA template to repair the other chromosomes by HR. With respect to the genome, CRISPRm is marker-free and iterative, enabling much more complex genome editing practices in lab and industrial yeast.

Section snippets

Plasmid Design

The first step in creating a coexpression system is to build a plasmid that can be: (A) stably maintained in the expression host and (B) propagated in bacteria. A plasmid that can be maintained in both bacteria and yeast requires a species-specific origin of replication and a dual function selection marker. Coexpressing the two components of the CRISPRm system (Cas9, sgRNAs) from a single plasmid has the advantage of not needing to be cotransformed, coinherited, or coexpressed and requires only

Cas9 Expression

Cas9 genome editing requires the coexpression of the Cas9 endonuclease and the guide RNA (Jinek et al., 2012). Correctly expressed, the guide RNA binds to Cas9 and forms a functional ribonucleoprotein, equipped with a precise targeting sequence within the guide RNA (Jinek et al., 2014). In vivo, this means that both the protein and RNA components need to be expressed at physiologically tolerable (nontoxic) levels by the cell, colocalized and correctly folded. Overexpression of proteins can be

Guide RNA Expression

Ascomycete yeasts express all of their transfer RNAs (tRNAs), the U6 spliceosomal RNA SNR6, the snoRNA SNR52, the RNA component of RNase P RPR1 and the RNA component of the signal recognition particle SCR1 using RNA Polymerase III (RNA Pol III) promoters (Marck et al., 2006). These RNA Pol III transcripts have varying architectures but all contain the essential components for RNA Pol III transcription initiation. They contain A Box and B Box binding domains and a TATA Box binding domain. Only

Screening Method

For CRISPRm to become a common practice in nonspecialized laboratories it is important that the protocol and reagent sets are simple, cost efficient and rely on well-established protocols. The pCAS plasmid developed in our lab can be used in any S. cerevisiae strain, including prototrophic isolates. Only one modification to pCAS is required for genome targeting. The researcher only needs to clone the target (protospacer) 20-mer sequence into the sgRNA encoded in the pCAS plasmid.

Screening

Concluding Remarks

In this chapter, we describe methods for integrating linear DNA into yeast chromosomes for: (A) loss of function studies and (B) heterologous protein expression. Because CRISPRm does not require selectable markers to be integrated with the DNA, this method is fully scalable. We propose that by using iterations of CRISPRm, a synthetic chromosome (or genome) in theory could be reduced from a wild-type organism. This is particularly possible because of the multiplex function with minimal reduced

Acknowledgments

We thank J. Doudna for helpful discussions on the manuscript. This work was supported by funding from the Energy Biosciences Institute.

Competing Financial Interests: Some of the authors have filed a patent application related to the results presented here.

References (28)

  • A.E. Farrell et al.

    Ethanol can contribute to energy and environmental goals

    Science

    (2006)
  • J.M. Galazka et al.

    Cellodextrin transport in yeast for improved biofuel production

    Science

    (2010)
  • G. Giaever et al.

    Functional profiling of the Saccharomyces cerevisiae genome

    Nature

    (2002)
  • P.D. Hsu et al.

    DNA targeting specificity of RNA-guided Cas9 nucleases

    Nature Biotechnology

    (2013)
  • Cited by (69)

    • CRISPR-based tools for microbial cell factories

      2021, Microbial Cell Factories Engineering for Production of Biomolecules
    View all citing articles on Scopus
    View full text