Antagonistic coevolution of parasite infectivity and host resistance may alter the biological functionality of species, yet these dynamics in nature are still poorly understood. Here we show the molecular details of a long-term phage–bacterium arms race in the environment. Bacteria (Flavobacterium columnare) are generally resistant to phages from the past and susceptible to phages isolated in years after bacterial isolation. Bacterial resistance selects for increased phage infectivity and host range, which is also associated with expansion of phage genome size. We identified two CRISPR loci in the bacterial host: a type II-C locus and a type VI-B locus. While maintaining a core set of conserved spacers, phage-matching spacers appear in the variable ends of both loci over time. The spacers mostly target the terminal end of the phage genomes, which also exhibit the most variation across time, resulting in arms-race-like changes in the protospacers of the coevolving phage population.

The antagonistic coevolution of parasite infectivity and host resistance alters the biological functionality of species, with effects spanning to communities and ecosystems. Still, studies describing long-term host-parasite coevolutionary dynamics in nature are largely missing.Furthermore, the role of host resistance mechanisms for parasite evolution is poorly understood, necessitating for the molecular and phenotypic characterization of both coevolving parasites and their hosts. We combined long-term field sampling (2007)(2008)(2009)(2010)(2011)(2012)(2013)(2014), in vitro cross-infections and timeshift experiments with bacteriophage whole genome sequencing and bacterial (Flavobacterium columnare) CRISPR (Clustered Regularly Interspaced Short Palindromic Repeats) profiling to show the molecular details of the phage-bacterium arms race in the environment. Bacteria were generally resistant to phages from the past and susceptible to phages in the future. The bacterial resistance selected for increased phage infectivity and host range, correlating directly with the expansion of phage genome size by 2656 bp. In the bacterial host, two CRISPR loci were identified: a type II-C locus and an RNA-targeting type VI-B locus. While maintaining a core set of conserved spacers, phage-matching spacers appeared in the variable end of both CRISPR loci over time. The appearance of these CRISPR spacers in the bacterial host often corresponded with arms race -manner molecular changes in the protospacers of the coevolving phage population. However, the phenotypic data indicated that the relative role of constitutive defence may be more important in high phage pressure, highlighting the importance of our findings for understanding microbial community ecology and in the development of phage therapy applications.

CRISPR-Cas immune systems adapt to new threats by acquiring new spacers from invading nucleic acids such as phage genomes. However, some CRISPR-Cas loci lack genes necessary for spacer acquisition despite variation in spacer content between microbial strains. It has been suggested that such loci may use acquisition machinery from cooccurring CRISPR-Cas systems within the same strain. Here, following infection by a virulent phage with a double-stranded DNA (dsDNA) genome, we observed spacer acquisition in the native host Flavobacterium columnare that carries an acquisition-deficient CRISPR-Cas subtype VI-B system and a complete subtype II-C system. We show that the VI-B locus acquires spacers from both the bacterial and phage genomes, while the newly acquired II-C spacers mainly target the viral genome. Both loci preferably target the terminal end of the phage genome, with priming-like patterns around a preexisting II-C protospacer. Through gene deletion, we show that the RNA-cleaving VI-B system acquires spacers in trans using acquisition machinery from the DNA-cleaving II-C system. Our observations support the concept of cross talk between CRISPR-Cas systems and raise further questions regarding the plasticity of adaptation modules. IMPORTANCE CRISPR-Cas systems are immune systems that protect bacteria and archaea against their viruses, bacteriophages. Immunity is achieved through the acquisition of short DNA fragments from the viral invader’s genome. These fragments, called spacers, are integrated into a memory bank on the bacterial genome called the CRISPR array. The spacers allow for the recognition of the same invader upon subsequent infection. Most CRISPR-Cas systems target DNA, but recently, systems that exclusively target RNA have been discovered. RNA-targeting CRISPR-Cas systems often lack genes necessary for spacer acquisition, and it is thus unknown how new spacers are acquired and if they can be acquired from DNA phages. Here, we show that an RNA-targeting system “borrows” acquisition machinery from another CRISPR-Cas locus in the genome. Most new spacers in this locus are unable to target phage mRNA and are therefore likely redundant. Our results reveal collaboration between distinct CRISPR-Cas types and raise further questions on how other CRISPR-Cas loci may cooperate.

Gliding motility facilitates the movement of bacteria along surfaces in many Bacteroidetes species and results in spreading colonies. The adhesins required for the gliding are secreted through a gliding motility-associated protein secretion system, known as the type IX secretion system (T9SS). The fish pathogen Flavobacterium columnare produces spreading (rhizoid [Rz], soft [S]) and non-spreading (rough [R]) colony types, of which only the spreading Rz type is virulent. In this study, we explored the spreading behavior of these colony types by microscopic imaging and measured the expression of genes associated with gliding motility and T9SS (gldG, gldH, gldL, sprA, sprB, sprE, sprF, sprT, and porV) under high and low resource levels by using RT-qPCR (reverse transcription quantitative PCR). The spreading colony types responded to the low resource level with increased colony size. The non-spreading colony type, as well as the cells growing under high nutrient level expressed only moderate cell movements. Yet, a low nutrient level provoked more active gliding motility in individual cells and increased spreading by cooperative gliding. The gene expression survey demonstrated an increased expression level of sprA (a core component of T9SS) and sprF (needed for adhesin secretion) under low nutrient conditions. Surprisingly, the expression of gliding motility genes was not consistently associated with more active spreading behavior. Furthermore, no genetic differences were found between spreading and non-spreading colony types in the studied genes associated with gliding motility. Our study demonstrates that environmental nutrient level is an important regulator of both gliding motility and the expression of some of the associated genes. These results may help to understand the connections between nutrient concentration, gliding motility, and virulence of F. columnare.

Phage therapy is becoming a widely recognized alternative for fighting pathogenic bacteria due to increasing antibiotic resistance problems. However, one of the common concerns related to the use of phages is the evolution of bacterial resistance against the phages, putatively disabling the treatment. Experimental adaptation of the phage (phage training) to infect a resistant host has been used to combat this problem. Yet, there is very little information on the trade-offs of phage infectivity and host range. Here we co-cultured a myophage FCV-1 with its host, the fish pathogen Flavobacterium columnare, in lake water and monitored the interaction for a one-month period. Phage resistance was detected within one day of co-culture in the majority of the bacterial isolates (16 out of the 18 co-evolved clones). The primary phage resistance mechanism suggests defense via surface modifications, as the phage numbers rose in the first two days of the experiment and remained stable thereafter. However, one bacterial isolate had acquired a spacer in its CRISPR (Clustered Regularly Interspaced Short Palindromic Repeat)-Cas locus, indicating that also CRISPR-Cas defense was employed in the phage-host interactions. After a week of co-culture, a phage isolate was obtained that was able to infect 18 out of the 32 otherwise resistant clones isolated during the experiment. Phage genome sequencing revealed several mutations in two open reading frames (ORFs) likely to be involved in the regained infectivity of the evolved phage. Their location in the genome suggests that they encode tail genes. Characterization of this evolved phage, however, showed a direct cost for the ability to infect several otherwise resistant clones—adsorption was significantly lower than in the ancestral phage. This work describes a method for adapting the phage to overcome phage resistance in a fish pathogenic system.