Harnessing short DNA repeats for genome engineering of phages
Alessandra G. de Melo 1*, Stéphanie Loignon 1,2, Jeffrey K. Cornuault 3, Sylvain Moineau 1,2*
- Département de biochimie, de microbiologie et de bio-informatique, Faculté des sciences et de génie, Université Laval, Québec, QC, Canada.
- Félix d’Hérelle Reference Center for Bacterial Viruses, Faculté de médecine dentaire, Université Laval, Québec, QC, Canada.
- Meinig School of Biomedical Engineering, Cornell University, Ithaca, NY, USA.
Sylvain Moineau: sylvain.moineau@bcm.ulaval.ca
Phages are ubiquitous, highly diverse, and can infect a variety of bacterial hosts in several ecological niches. To successfully infect their hosts, phages must develop strategies to escape bacterial defense mechanisms or barriers for infection. Interestingly, we recently noticed when studying Pseudomonas phage Ab09 that its genome contains several short DNA repeats, and these repeats could mediate recombination when the viral genome is cut by a CRISPR-Cas system. Additional bioinformatics analyses revealed that these short DNA repeats are in fact highly abundant in phage genomes. Here, we investigated whether we could exploit these short DNA fragments for genome engineering purposes using CRISPR-Cas9.
We aimed to modify four virulent phage genomes infecting various bacterial species, such as E. coli, Salmonella, Pseudomonas, and Streptococcus. For E. coli phage T7 and Salmonella phage Felix O1, we used pCas9 that contained the heterogenous CRISPR-Cas9 system from Streptococcus pyogenes. For the genome editing of phage 2972, we used the endogenous CRISPR-Cas system of its host Streptococcus thermophilus. To modify the genome of virulent Pseudomonas phage Ab09, we first adapted a plasmid containing an heterogenous CRISPR-Cas9 system from Neisseria meningitidis (NmeCas9) by adding the pRO1600 origin of replication and the gentamicin resistance gene, generating pNmeCas9P.
By using these heterogenous or endogenous CRISPR-Cas systems, we spacer-targeted specific phage genes that were naturally flanked by short DNA repeats to mediate recombination in the above four phage genomes. Using mostly 10-11 bp repeats, we successfully generated knockouts of three genes (orf09, orf10 and orf14) or two genomic regions (orf06-orf10 and orf80-orf83) of Pseudomonas phage Ab09. For Streptococcus phage 2972, we targeted the gene coding for receptor binding protein as well as orf41. In T7, we targeted genes 0.3-0.4, 1.1-1.2, 1.4-1.5, and 1.5 alone and generated deletions in all those genes. Interestingly, in coliphage T7 genome, the deletion of the genes 1.4 and 1.5 was mediated by repeats of 7 bp. For Salmonella Felix O1, which had repeats up to 57 bp, we were able to generate deletions in orf21, orf22, and orf19-orf20.
In conclusion, targeting specific phage genes with the CRISPR-Cas9 technology led to viral genome recombination mediated by short DNA repeats, and the generation of several knockout phage mutants. Compared to the traditional strategy using CRISPR-Cas9 for phage genome engineering, which includes cloning of both spacer and repair template, this approach described above reduces considerably the time required to generate knockouts. In terms of phage biology, our findings also raise questions on the importance of short DNA repeats in phage evolution.