RR:C19 Evidence Scale rating by reviewer:
Reliable. The main study claims are generally justified by its methods and data. The results and conclusions are likely to be similar to the hypothetical ideal study. There are some minor caveats or limitations, but they would/do not change the major claims of the study. The study provides sufficient strength of evidence on its own that its main claims should be considered actionable, with some room for future revision.
***************************************
Review: The paper of Backman et al present the finding of a conserved viral cluster in Pseudomonas spp. that is identified as a tailocin. This tailocin seems important for the competition between bacteria, both in planta and in vivo.
Overall, the manuscript is well written, and the results are clearly presented. The logic of the manuscript is also clear, starting from the observation of the tailocins encoded in a particular clade of bacterial plant pathogens, evidence for its conservation in several strains, following with the phenotypic effect of the most critical genes, and finally by assessing its historical presence in the population.
Although not completely novel (the existence and effect of tailocins is fairly well known), this manuscript does contribute to our understanding of these molecules by providing an environmental (i.e., the particular species that produce and are target by tailocins) and temporal (i.e., by the analysis of ancient Pseudomonas genomes) context. This context is important to understand the applicability and evolutionary resilience of tailocins.
In our view, there are no major issues with the analysis. Nevertheless, some points should be further expanded or clarified to strengthen the message of the paper:
The authors define in their manuscript a tailocin as “a phage-derived element that bacteria use to kill competitors for interbacterial warfare”. However, we would prefer the authors to use a more precise terminology, e.g the one of Patz et al. in [REF : https://doi.org/10.1016/j.jare.2019.04.003 ]. Patz et al. classify tailocin, and more precisely R-type tailocin as extracellular particles that kill bacteria by destroying the proton motive force. In the Backman et al. manuscript, no clear demonstration of this particular mode of action is available. In this case, we think that the phage tail-like particles found by the authors are possibly Phage-Like Protein Translocation Structures (PLTSs), meaning that this extracellular particles could kill bacteria by delivering toxins. Moreover, the loci of this extracellular particles contain Colicin-M & Putative chitinase (Fig. 2B), i.e putative toxins. Thus, mentioning how precisely authors define a tailocin and precising it’s mode of action, or using another terminology, would add clarity to the manuscript. Also, the title “A weaponized phage …” could be clarified using the chosen terminology.
Without further information, the analysis of the HTF variants as haplotypes defined by their length is not very informative, and sometimes even a bit confusing. We understand the need to refer to these haplotypes by something that can identify them. But it would be relevant to expand on what is similar (or not) between them. For instance, are there domains or particular sequences that are present/absent between phenotypes (and what do they correspond to e.g., structurally, if known). This seems relevant to understand the context of the TnSeq experiment as well, which identified mutants resistant to the tailocins. If the length of a HTF is associated with a given O-antigen genomic profile (from the presence/absence of genes), is this associated with particular missing domains, or HTF haplotype-specific domains?
We found it interesting that the most frequent HTF haplotypes (1803bp and 1245bp) are also the ones where O-antigen genes targeted by TnSeq are absent. Could the authors think of any theories to explain such an association?
Although the authors mention that the TFA haplotypes are correlated with the haplotypes of HTF, it would be informative to have independent data of each gene in some of the analysis. For instance, does the congruence between a TFA haplotype tree and the results of TnSeq (parallel to Fig 4B) show similar results to the HTF haplotypes? The reason we mention this is that, in some cases (e.g., Fig5B and FigS11A) they do seem to differ.
Although Fig 5B gives an idea of the homology between ancient and more modern HTF genes, it would also be interesting for the readers to have an idea of the percentage of identity between the ancient HTF and the reference HTF used in Fig 5A (and the same for the TFA gene).
Some of the plots that state "Frequency" (e.g., Fig 1B and 3A) should "Abundance" or "Counts".
Finally, some of the phylogenetic trees do not seem to have a scale (e.g., Fig 3B).
Creative Commons Attribution 4.0 International License (CC-BY 4.0)
