Barriers to genome editing with CRISPR in bacteria

Abstract

Genome editing is essential for probing genotype-phenotype relationships and for enhancing chemical production and phenotypic robustness in industrial bacteria. Currently, the most popular tools for genome editing couple recombineering with DNA cleavage by the CRISPR nuclease Cas9 from Streptococcus pyogenes. Although successful in some model strains, CRISPR-based genome editing has been slow to extend to the multitude of industrially relevant bacteria. In this review, we analyze existing barriers to implementing CRISPR-based editing across diverse bacterial species. We first compare the efficacy of current CRISPR-based editing strategies. Next, we discuss alternatives when the S. pyogenes Cas9 does not yield colonies. Finally, we describe different ways bacteria can evade editing and how elucidating these failure modes can improve CRISPR-based genome editing across strains. Together, this review highlights existing obstacles to CRISPR-based editing in bacteria and offers guidelines to help achieve and enhance editing in a wider range of bacterial species, including non-model strains.

Keywords: Bacteria; CRISPR; Genome editing; Nuclease; Recombineering.

Fig. 1. Different strategies for CRISPR-based genome editing in bacteria.

Column 1: Recombineering using a linear DNA template followed by counterselection with CRISPR nucleases. A plasmid encoding a heterologous recombinase (Rec) is introduced into the cell and induced before co-transforming the linear DNA template and CRISPR-nuclease plasmid. Column 2: Recombineering using a plasmid-encoded recombineering template (RT) with or without a heterologous recombinase. The recombineering template can be placed on the plasmid harboring the CRISPR machinery for an all-in-one plasmid system, or it can be placed on a separate plasmid before transforming the CRISPR nuclease/gRNA plasmid. While one-plasmid systems are more streamlined, the larger plasmid could prove harder to transform and co-encoding the nuclease and gRNA could interfere with cloning if the gRNA can target the genome of the cloning strain. If no exogenous recombinase is used, this method relies on the cell’s native recombineering machinery. Column 3: Editing via the non-homologous end-joining (NHEJ) pathway. Depending on the strain, ku and/or ligD can be encoded on the plasmid harboring the CRISPR machinery and transformed into the strain. All strategies require plasmid curing after recombineering and nuclease targeting to isolate the mutant strain before pursuing downstream applications.

Fig. 2. Circumventing lack of colonies when using SpCas9 in bacteria.

a) Expressing SpCas9 is cytotoxic in some bacteria (left), while other bacteria do not yield any colonies when attempting editing (right). b) Several alternative strategies can be explored to circumvent these issues. Upper left: Utilizing inducible systems to express SpCas9 following transformation and culturing. Via an inducible promoter, SpCas9 expression is strongly repressed without inducer present and only induced after culturing the cells to ensure a large number of cells possess all components necessary for editing. Upper right: Using less toxic nucleases to achieve editing. Cas9n, which only cleaves one strand of DNA, and Cas12a can be less toxic than SpCas9. Lower left: SpCas9-derived base editors eliminate the need to create a double-stranded break to achieve editing. A translational fusion of dCas9 or Cas9n, a cytidine deaminase domain, and a uracil DNA glycosylase inhibitor (UGI) is introduced on a plasmid into the cell. Upon nuclease binding and R-loop formation, cytidines on the non-target strand within a defined window adjacent to the PAM are rapidly converted to uracils. Lower right: Harnessing endogenous CRISPR nucleases for genome editing. For strains harboring native CRISPR nucleases, gRNAs can be introduced along with a recombineering template to achieve editing without expressing a heterologous CRISPR nuclease. One drawback is that native cRNAs can compete with the introduced genome-targeting gRNAs, although preventing crRNA biogenesis can eliminate this barrier.

Fig. 3. Reported failure modes of CRISPR-based genome editing in bacteria.

a) Escaper colonies can form that contain either the wild-type sequence or an unintended edit, and make screening for correctly edited cells more difficult. b) Deactivated CRISPR machinery. Bacteria sometimes avoid CRISPR cleavage by mutating the gRNA on the transformed plasmid or mutating the cas genes. c) Unintended genomic excision events and mutations result in loss of target sequence. Bacteria can remove the protospacer sequence targeted by the CRISPR nuclease via genomic excision events driven by homologous recombination with the genome or via mutations to the protospacer. d) Cell repair via homologous recombination with the genome. Upon CRISPR cleavage at the target site, the cell can use another copy of the chromosome to repair itself. e) Reversion of the recombineering template. The recombineering template containing the desired mutation can revert back to the wild-type sequence in the host cell, preventing editing.