Metabolic diversity within the globally abundant Marine Group II Euryarchaea offers insight into ecological patterns

Despite their discovery over 25 years ago, the Marine Group II Euryarchaea (MGII) remain a difficult group of organisms to study, lacking cultured isolates and genome references. The MGII have been identified in marine samples from around the world, and evidence supports a photoheterotrophic lifestyle combining phototrophy via proteorhodopsins with the remineralization of high molecular weight organic matter. Divided between two clades, the MGII have distinct ecological patterns that are not understood based on the limited number of available genomes. Here, I present a comparative genomic analysis of 250 MGII genomes, providing a comprehensive investigation of these mesophilic archaea. This analysis identifies 17 distinct subclades including nine subclades that previously lacked reference genomes. The metabolic potential and distribution of the MGII genera reveals distinct roles in the environment, identifying algal-saccharide-degrading coastal subclades, protein-degrading oligotrophic surface ocean subclades, and mesopelagic subclades lacking proteorhodopsins, common in all other subclades.

In the revised manuscript "Metabolic Diversity within the Globally Abundant Marine Group II Euryarchaea Offers Insight into Ecological Patterns" Benjamin Tully convincingly describes the diversity and potential ecological role of MG II Euryarchaea. Most of my initial comments have been addressed to my satisfaction. The naming scheme has been updated and has gone away from the Lord of the Rings-based names (which I appreciate a lot, particularly as it might have caused legal issues regarding the copyright of the names). As a side note, the Hobbit-naming scheme is still mentioned in the Acknowledgements. I am very much in favor of recommending this manuscript fir publication, as it has undergone substantial improvements compared to what I have seen in the previous version. More robust statistics have been applied as well (e.g. PERMANOVA) and methods have now been described in a satisfying manner. The only concern I see with the paper is, that in the meantime a similar story about MGII Archaea has a appeared in The ISME Journal (Rinke et al.). However, the story by Benjamin Tully has already been on BioRxiv for quite some time and Rinke et al. simply ignored Benjamin Tully's work. I think the current paper still deserves publication if a comparison to the Rinke et al paper is Editorial Note: This manuscript has been previously reviewed at another journal that is not operating a transparent peer review scheme. This document only contains reviewer comments and rebuttal letters for versions considered at Nature Communications .
included in the discussion section and the phylogeny and taxonomy are properly compared. A comparison of the potential metabolic traits would be necessary too. All these requests need, of course, to be in line with the journal policy about studies that get "scooped" while in peer review.

Reviewer #1
This manuscript provides a new analysis and proposes a classification for the important marine Archaea of group II. The manuscript has merit and provides interesting views about this important group. Unfortunately, this paper published recently has a similar proposal: https://www.nature.com/articles/s41396-018-0282-y To avoid confusion and also for scientific consistency, this manuscript needs to be rewritten considering the published proposal. Response: All three reviewers and the editor made this suggestion. Many of the core metabolic functions of the MGII are identified in both manuscript. Major differences in the manuscript stem from areas of interest unique to each author (Rinke et al. emphasized amino acid and membrane biosynthesis pathways, while I explored the role of Na + motive force in the ETC and carbohydrate degradation. Rinke et al. emphasized a HGT-based theory for proteorhodopsin prevalence, a theory that is not in contradiction to the PR data presented in this manuscript). The phylogenetic trees presented in both manuscripts are virtually identical with only small differences affecting the number of subclades and genera present.
The following text has been added at the end of the Discussion: It should be noted that during review a separate analysis of the MGII, derived predominantly from the UBA 22 dataset, was published by Rinke et al. 71 . The two analyses provide a complementary analysis of MGII phylogeny, metabolic functional prediction, and global ecology. Importantly, both manuscripts provide a framework for MGII phylogenetic assignment that are directly comparable (Supplemental Figure 9). Based on normalized ranks derived from relative evolutionary divergence 72 , Rinke et al. propose an Order-level classification for the MGII (Candidatus Poseidoniales) and subsequent classification of MGIIa and MGIIb in to the families Ca. Poseidoniaceae and Ca. Thalassarchaeaceae, respectively. Thus groups of MGII genomes designated as subclades in this manuscript would correspond to genera in Rinke et al. There is a discrepancy in the number of genera identified by Rinke et al. (n = 21) and the 17 subclades described here. Five of the Rinke et al. genera (J1-3 and Q1-2) consisted of two or less genomes, and while counterparts are distinguishable amongst this analyzed dataset, the deep-branching nature of these genomes did not meet the criteria necessary for designations as a subclade (Supplemental Figure 9). The single genome representing genus L4 in Rinke et al. did not meet the criteria for inclusion in the non-redundant dataset, but is present in the same phylogenetic position for the full redundant dataset (Supplemental Figure 1). Further, Rinke et al. genera L1 and O3 was split in to two subclades each. This is likely the result of the specific cutoffs used to split clades O and L using relative evolutionary divergence 71 (Supplemental Figure 9).
Both manuscripts highlight the potential role of the MGII in particle degradation, specifically through the breakdown of proteins and fatty acids, particle attachment via flotillins, the presence of an archaeal flagellum operon in many subclades, and a varied distribution of green-and blue-light adapted proteorhodopsins among subclades. The horizontal gene transfer system proposed by Rinke et al. may be a possible explanation to account for the finescale differences observed within subclades detailed in this study. In addition to metabolic features discussed here, Rinke et al. discuss the lack of amino acid biosynthesis genes and the potential for mixed membrane synthesis among the MGII, which was not addressed in this manuscript. Though the ecological approaches for the manuscripts are not directly comparable, both analyses identify obligate mesopelagic MGII genomes. and MGIIb.13/genus O3 could be resolved by slight modifications in the cutoff value for relative evolutionary distance applied to each genus. This change that would likely maintain the rank normalization groupings presented in Figure S5 in Rinke et al.

Reviewer #3
The author has responded to all my comments, and has modified analyses (e.g., naming of lineages; phylogenomic analysis) and the text accordingly to all reviewers' requests. I believe the manuscript has significantly improved.
In light of the recent publication by Rinke et al. (https://www.nature.com/articles/s41396-018-0282-y), I consider that the study by Tully still represents an important contribution to the fields of microbial ecology and marine microbiology. In my view, the manuscript only requires few adjustments in order to take into account the recent publication. Especially, the linkage between 17 subclades by Tully and 21 genera by Rinke et al. needs to be made, maybe as a supplemental figure or table. This effort will create a bridge between the two studies, and in my view would benefit to the study by Tully, and by extension to our scientific community. Response: Please see the response to Reviewer #1.

Reviewer #4
In the revised manuscript "Metabolic Diversity within the Globally Abundant Marine Group II Euryarchaea Offers Insight into Ecological Patterns" Benjamin Tully convincingly describes the diversity and potential ecological role of MG II Euryarchaea. Most of my initial comments have been addressed to my satisfaction. The naming scheme has been updated and has gone away from the Lord of the Rings-based names (which I appreciate a lot, particularly as it might have caused legal issues regarding the copyright of the names). As a side note, the Hobbit-naming scheme is still mentioned in the Acknowledgements. I am very much in favor of recommending this manuscript fir publication, as it has undergone substantial improvements compared to what I have seen in the previous version. More robust statistics have been applied as well (e.g. PERMANOVA) and methods have now been described in a satisfying manner. The only concern I see with the paper is, that in the meantime a similar story about MGII Archaea has a appeared in The ISME Journal (Rinke et al.). However, the story by Benjamin Tully has already been on BioRxiv for quite some time and Rinke et al. simply ignored Benjamin Tully's work. I think the current paper still deserves publication if a comparison to the Rinke et al paper is included in the discussion section and the phylogeny and taxonomy are properly compared. A comparison of the potential metabolic traits would be necessary too. All these requests need, of course, to be in line with the journal policy about studies that get "scooped" while in peer review. Response: Please see the response to Reviewer #1.
Minor comments: -Abbreviations like RPKM should be introduced to the reader the first time they are being mentioned (e.g. Line 121) Response: The abbreviation of RPKM in the Figure 2 legend has been explained. -Legends and axis labelings of figures are sometimes too small to be seen without zooming in I have made some adjustments to legends and axes for Figure 2, 3, and 5. I will work with the editorial team to make sure final figures are readable in the publication.