Infection by defective bacterial viruses that cannot replicate has now been found to be the key feature enabling bacteria to rapidly develop adaptive immunity against functional viruses.
Surviving viral infections is a necessary ability for most life forms. Adaptive immunity, in which invasive elements are captured by the cell, allowing subsequent recognition and destruction of related viruses, is crucial for overcoming such infections. As such, adaptive immunity drives evolution, selection and fitness. Although the antibody–antigen basis of mammalian adaptive immunity has been extensively characterized, its counterpart in archaea and bacteria — CRISPR–Cas immune systems — remains largely mysterious. Writing in Nature Communications, Hynes et al.1 describe how defective virus particles trigger immunization events by CRISPR–Cas systems, conferring adaptive immunity in the bacteria against related functional viruses.
CRISPR–Cas systems have two components: DNA sequences comprised of clustered regularly interspaced short palindromic repeats (CRISPR), and CRISPR-associated sequence (Cas) endonuclease enzymes. Typically, immunity arises when invasive genetic elements (for example, DNA injected into the cell by bacterial viruses called bacteriophages, or phages) are incorporated into the genome as 'spacers' between CRISPR sequences2. Subsequent transcription of the CRISPR array containing the incorporated spacers leads to the production of small interfering CRISPR RNAs3, which guide Cas enzymes to target and cleave DNA sequences that are complementary to the spacers4,5,6. This adaptive immune system has proven to be effective against phage predation in dairy starter cultures, which are widely used in yogurt and cheese manufacturing2.
Although it has been established that the uptake of viral DNA into the host genome drives CRISPR-based immunity5, little is known about how bacteria sample the genomes of phages for spacer acquisition, or the dynamics of the immunization process, which can be thought of as a bacterial 'vaccination' against phages. Phages typically take over the molecular machinery of their host within minutes, and the ability of bacteria to mount a quick adaptive immune response has remained enigmatic.
Hynes et al. first exposed bacterial cells to defective phages. These were produced either by exposing phages to ultraviolet (UV) radiation or by using virulent phages that are susceptible to a restriction–modification (RM) system that uses restriction enzymes to cleave phage DNA after injection into the host. In both cases, the defective phages can inject DNA into the host, but cannot replicate. DNA injections by cleavage-sensitive phages result in a montage of phage DNA fragments in infected cells. Irradiation-weakened phages inject and present non-replicative DNA, which can potentially be sampled and acquired by CRISPR arrays.
The authors searched for surviving host bacteria, and found that survivors had acquired additional spacers in CRISPR sequences, an indication that the phage DNA was accessible to the CRISPR adaptation machinery (Fig. 1). Although most of the cells died, a fraction of the infected population captured phage-genome pieces in CRISPR sequences. Specifically, the presence of UV-inactivated and RM-susceptible viruses increased the generation of vaccinated bacteria by three- to fourfold and tenfold, respectively, when compared with the presence of functional phages. This implies that replication-deficient viruses drive immunization.
Next, Hynes and colleagues used a 'double viral challenge' scheme, in which bacterial hosts were concurrently infected with defective and functional phages, to demonstrate that non-replicating viruses can be used to trigger vaccination against a distinct but similar family of functional phages. This test showed that most of the vaccination events that protect the cells from the functional phages arise from the defective phage populations. The authors report a direct correlation between the proportion of replication-deficient phages used in the challenge and the number of vaccination events, compared with a challenge using only functional phages. This is reminiscent of the use of attenuated viruses and bacteria for human vaccination against pathogens.
This study provides crucial proof of concept that defective viruses can be used to trigger immunization events through CRISPR–Cas systems. Although the use of attenuated viruses for vaccination is not new, the finding that inactivated viruses can trigger CRISPR-dependent adaptive immunity in bacteria has practical implications. Thus far, analysis of CRISPR–Cas systems in general, and adaptive spacer acquisition in particular, has been hampered by the limited set of available CRISPR model systems able to acquire spacers (as opposed to just targeting and cleaving nucleic acids). We anticipate that the use of attenuated viruses will allow researchers to expand the effectiveness of CRISPR immunization, and to use CRISPR–Cas in bacteria in which the system was previously deemed to be inactive. Future studies should also establish whether sampling of chromosomal DNA and plasmid DNA (small, non-chromosomal circular DNA molecules found in bacteria and archaea) follows the same molecular rules as viral DNA.
With an increased pool of active CRISPR–Cas systems, more Cas-based molecular machines could be repurposed for biotechnological applications, such as engineering bacterial resistance to phages or plasmids, or using CRISPR–Cas technology to edit genomes and to regulate transcription in various life forms, from bacteria to animals7,8. Hynes and colleagues' findings will help us begin to understand the role of CRISPR in the arms race between microbial communities and their viral predators in natural habitats, and will set the stage for further applications of CRISPR–Cas systems.
About this article
Genome Biology (2015)