Origin and evolutionary landscape of Nr2f transcription factors across Metazoa.

BACKGROUND: Nuclear Receptor Subfamily 2 Group F (Nr2f) orphan nuclear hormone transcription factors (TFs) are fundamental regulators of many developmental processes in invertebrates and vertebrates. Despite the importance of these TFs throughout metazoan development, previous work has not clearly outlined their evolutionary history.
RESULTS: We integrated molecular phylogeny with comparisons of intron/exon structure, domain architecture, and syntenic conservation to define critical evolutionary events that distinguish the Nr2f gene family in Metazoa. Our data indicate that a single ancestral eumetazoan Nr2f gene predated six main Bilateria subfamilies, which include single Nr2f homologs, here referred to as Nr2f1/2/5/6, that are present in invertebrate protostomes and deuterostomes, Nr2f1/2 homologs in agnathans, and Nr2f1, Nr2f2, Nr2f5, and Nr2f6 orthologs that are found in gnathostomes. Four cnidarian Nr2f1/2/5/6 and three agnathan Nr2f1/2 members are each due to independent expansions, while the vertebrate Nr2f1/Nr2f2 and Nr2f5/Nr2f6 members each form paralogous groups that arose from the established series of whole-genome duplications (WGDs). Nr2f6 members are the most divergent Nr2f subfamily in gnathostomes. Interestingly, in contrast to the other gnathostome Nr2f subfamilies, Nr2f5 has been independently lost in numerous vertebrate lineages. Furthermore, our analysis shows there are differential expansions and losses of Nr2f genes in teleosts following their additional rounds of WGDs.
CONCLUSION: Overall, our analysis of Nr2f gene evolution helps to reveal the origins and previously unrecognized relationships of this ancient TF family, which may allow for greater insights into the conservation of Nr2f functions that shape Metazoan body plans.

Introduction

Nuclear hormone receptors (NRs) form a large, ancient superfamily of transcription factors (TFs) found in all Metazoa [1]. While NR functions are often dictated by interactions with specific ligands, including steroids, thyroid hormones, and retinoids [2, 3], the ligands for many NRs, called orphan NRs, are still not known [4]. Nuclear Receptor Subfamily 2 Group F Members (Nr2fs), initially named Chicken ovalbumin upstream promoter-transcription factors (Coup-TFs) due to their ability to bind the COUP element of the ovalbumin gene [5-7], are some of the most highly studied orphan NRs. Despite an overall expansion of the NR superfamily [1, 2], invertebrate phyla appear to have predominantly retained a single Nr2f gene. Only one Nr2f member is present in the protostome Drosophila melanogaster (fly), early-branching deuterostome Strongylocentrotus purpuratus (sea urchin) [8, 9], and invertebrate chordates Branchiostoma floridae (amphioxus) and Ciona robusta (sea squirt) [10, 11]. However, the number of Nr2f genes in early-branching metazoans is presently less clear. In cnidarians, one Nr2f has been reported in Hydractinia echinata [12, 13], while multiple have been reported in Nematostella and Hydra vulgaris [14, 15]. In contrast to most invertebrates, vertebrates have exhibited a significant expansion of the Nr2f family, with gnathostomes having multiple Nr2f genes. Furthermore, teleosts possess additional Nr2f Ohnologs (duplicates originating from whole-genome duplication (WGD)) [16], most likely reflecting the additional WGDs that have occurred in the teleost lineage [17, 18].

Nr2f proteins are highly conserved at the sequence level throughout Metazoa [19]. From the N-terminus to the C-terminus, all Nr2f proteins have six domains (Fig 1): an A/B domain, which contains the activating function-1 (AF-1) domain; the C domain, which contains the DNA-binding domain (DBD); the D domain (a linker); the E domain, which is comprised of the ligand-binding domain (LBD) and an AF-2 domain; the F domain (C-terminal) [20]. While the A/B domains are the most divergent in sequence, strikingly, the DBDs and LBDs of Nr2f members even from distantly related species (e.g. fly, sea urchin, frog, zebrafish, mouse, and human) are ~94% identical [21]. The extremely high degree of conservation among several species implies the preservation of critical roles for Nr2f in development and differentiation [21, 22]. Moreover, requirements for Nr2f genes have been found in organs of all three germ layers during embryogenesis [14, 22]. For instance, the Drosophila Nr2f homolog, called seven up (svp), is required for retinal, dorsal vessel, and liver development [23, 24]. Furthermore, Nr2f TFs in vertebrates appear to both have acquired diverse and retained redundant functions. For instance, in mice, Nr2f1 is predominantly required for neural development with a role in regulation of premigratory and migratory neural crest cells in the developing hindbrain [25, 26]. However, the mouse Nr2f2 gene is required for differentiation of mesodermal derivatives, including atrial cardiomyocytes of the heart and venous endothelial cells [22, 27, 28]. An example of redundancy is zebrafish nr2f1a and nr2f2, which are both required for proper ventricular cardiomyocyte and cranial muscle specification [29].

Fig 1

Schematic of conserved domain architecture of Nr2f TFs.

A/B (N-terminal variable domain with transactivating AF-1 domain), C (DBD, which contains two Zinc finger (Znf) motifs), D (a linker domain), E (LBD plus transactivating AF-2 domain), and F (C-terminal).

While Nr2f proteins were initially identified as transcriptional activators of chicken ovalbumin gene [5], they have since been shown to function directly as both transcriptional activators and repressors in several developmental contexts [22, 30, 31]. Nr2fs can bind a range of different response elements [32-34] and in signaling reporter assays can compete with and inhibit retinoic acid receptors [35]. In vivo they bind numerous targets that reflect their various requirements in the specific tissues. For instance, Nr2f1 KO mice also have inner ear defects [36]. In the mouse inner ear, direct targets of Nr2f1 include fatty acid binding protein 7 (FABP7), cellular retinoic acid binding protein 1 (CRABP1) [37], microRNA-140 (mi-R140), and Krüppel-like 9 (Klf9) [37, 38]. In adipogenesis, Nr2f2 directly represses peroxisome proliferator-activated receptor γ (PPARγ) downstream of canonical Wnt/β-catenin signaling [39]. In the mammalian heart, Nr2f2 is thought to directly orchestrate a regulatory network that facilitates atrial cardiomyocyte identity through concurrently promoting Tbx5 and repressing Irx4 and Hey2, the latter of which promote ventricular cardiomyocyte identity [40]. Thus, Nr2fs can activate and repress a range of direct targets related to their functions in specific tissues.

Despite the conservation and clear importance of this gene family to numerous developmental processes in Metazoa, we still do not completely understand the evolution of Nr2f TFs. Here, we investigated Nr2f family evolution through a combination of phylogenetic, domain architecture, intron/exon structure, and genomic synteny analyses. Our data show that the single Nr2f gene found in placozoans, represents the ancestral Nr2f to those found in cnidarians, protostomes, and deuterostomes. Importantly, a single Nr2f homolog, which we have named Nr2f1/2/5/6, is present in the majority of invertebrates, while most vertebrate genomes contain Nr2f1, Nr2f2, Nr2f5, and Nr2f6 orthologs, which are derived from established rounds of WGDs [41, 42]. Interestingly, the invertebrate Nr2f1/2/5/6 and agnathan Nr2f1/2 homologs have retained the greatest similarity with vertebrate Nr2f1 and Nr2f2 paralogs. With respect to the vertebrate Nr2f5 and Nr2f6 paralogs, Nr2f5 genes have been independently lost in some cartilaginous fish and amniote lineages, while the Nr2f6 subfamily is the most divergent with respect to sequence and genomic structure. Overall, our data clarify the relationships among Nr2f genes within Metazoa and define the expansion, divergence, and independent loss of extant Nr2f genes in vertebrates, which will allow us to make meaningful inferences about the conserved developmental functions of this family that have helped mold animal body plans.

Results

Phylogenetic reconstruction of Nr2f evolution in animals

Although previous work has investigated the homology of some Nr2fs within metazoans, these analyses were primarily focused on their relationship to other NRs and were limited by the comparatively little genomic information at the time [1, 3, 14, 43, 44]. Therefore, the relatively few Nr2f family members examined in the previous analysis did not provide a specific and detailed understanding of Nr2f evolution. To garner a better understanding of how the Nr2f family has evolved in animals, we performed a phylogenetic analysis using 153 Nr2f proteins with representatives from placozoans to mammals (Fig 2; S1 File). Early-branching metazoan models Amphimedon queenslandica (sponge) and Mnemiopsis leidyi (ctenophore) were not included, as we did not find putative Nr2f orthologs based on current databases, consistent with published phylogenetic studies of the NR superfamily [13, 15]. The placozoan Trichoplax adhaerens Nr2f, which was previously shown to cluster with vertebrate Nr2fs in phylogenetic analyses [45], was used as the outgroup in a maximum-likelihood (ML) phylogenetic tree. Protein sequences from groups that caused long branch artifacts due to significant divergence were not included in the phylogenetic trees (S2 File). This phylogenetic analysis provided evidence for the existence of distinct Nr2f subfamilies (Fig 2). Moreover, the same relationships were also supported using a Bayesian model selection (S1 Fig). Present information allowed us to identify four Nr2fs in the cnidaria Nematostella vectensis and Acropora millepora, three in Hydra vulgaris, and one for Hydractinia echinata. However, while identifiable as Nr2fs, an A. millepora, the H. vulgaris, and the H. echinata Nr2fs caused long-branch artifacts and were consequently excluded (S2 File). Interestingly, the tree incorporating the N. vectensis and remaining A. millepora Nr2f members, which we now call Nr2f1/2/5/6a-d based on their relationship to Bilateria Nr2fs, were found at the base of the eumetazoan Nr2f proteins and are likely the result of gene duplications within cnidaria [15] (Fig 2). The protostome and deuterostome Nr2f sequences clustered into six subfamilies, which we have called Nr2f1/2/5/6, Nr2f1/2, Nr2f1, Nr2f2, Nr2f5, and Nr2f6. Single Nr2f1/2/5/6 subfamily genes, which are highly conserved, yet evolutionary divergent from the Nr2f1/2/5/6 genes present in early-branching eumetazoa, were found in invertebrate protostomes, invertebrate deuterostomes (hemichordates, echinoderms), and invertebrate chordates (amphioxus, tunicates). An older nomenclature proposal suggested that the Drosophila Nr2f (svp) should be designated Nr2f3 [19], implying the other invertebrate Nr2fs should follow this nomenclature. However, this designation seems to obfuscate the homology of these genes revealed here and imply a different evolutionary relationship, as there is no distinct Nr2f3 subfamily. Thus, we propose using Nr2f1/2/5/6 or in the future potentially just Nr2f for the early-branching eumetazoan and invertebrate Nr2f genes. We have used Nr2f1/2/5/6 in this manuscript to refer to the invertebrate Nr2fs to reinforce their evolutionary relationship within the Nr2f family. The invertebrate Nr2f1/2/5/6 group is more closely related to the branch that includes Nr2f1/2s from the agnathan (lamprey and hagfish) and vertebrate Nr2f1 and Nr2f2 proteins than the vertebrate Nr2f5 and Nr2f6 subfamilies (Fig 2). The clustering of the invertebrate Nr2f1/2/5/6 and agnathan Nr2f1/2 proteins with Nr2f1 and Nr2f2 of gnathostomes suggests that these paralogous gnathostome genes arose from distinct duplicative events during vertebrate evolution [41, 42]. In addition, the three agnathan Nr2f proteins found in Sea lamprey (Petromyzon marinus) and hagfish (Eptatretus burgeri) (Fig 2), which we have called Nr2f1/2A, Nr2f1/2B, and Nr2f1/2C, diverge and cluster together at the base of the vertebrate Nr2f1 and Nr2f2 proteins (Fig 2), supporting that the duplications leading to these proteins in agnathans were distinct from those that gave rise to the Nr2f paralogs in gnathostomes.

Fig 2

Evolutionary reconstruction of Nr2f TFs in metazoans.

Phylogenetic tree of Nr2f members demonstrate the existence of six protein classes: Nr2f1/2/5/6 (violet box), Nr2f1/2 (yellow box), Nr2f1 (pink box), Nr2f2 (green box), Nr2f5 (blue box), Nr2f6 (orange box). The placozoan Trichoplax adhaerens Nr2f was used as the outgroup. Values at the branches indicate replicates obtained using the aLRT method.

Our analysis also shows that Nr2f5 and Nr2f6 form a separate branch and are sisters groups, implying that they are paralogous and derived from the second of the vertebrate WGDs [41, 42]. Importantly, while all gnathostomes examined have retained Nr2f1, Nr2f2 and Nr2f6, current genomic data support that Nr2f5 has been independently lost by multiple vertebrate groups. Cartilaginous fish, including Whale shark (Rinchodon typus) [46] and the Great white shark (Carcharadon charcarias) [47], have retained Nr2f5 genes, while they are absent in chimaera [48] and skates (S3 File). In amniotes, Nr2f5 genes were found in reptiles, such as American alligator (Alligator mississippiensis), gecko (Gekko japonicus), and the Green sea turtle (Chelonia mydas) (Fig 2; S1 Fig), but absent from the Chinese sea turtle (Pelodiscus sinesis), as well as birds and mammals (S3 File). Although previous work had also designated a Xenopus laevis Nr2f4 [19], our data indicate there is no evidence for a separate Nr2f4 subfamily and that this gene should be called Nr2f5. Comparing the vertebrate Nr2f1/Nr2f2 and Nr2f5/Nr2f6 clusters, the branching and distances from our phylogenetic trees indicate that Nr2f1 and Nr2f2 are more highly conserved, while Nr2f6 TFs comprise the most divergent vertebrate Nr2f subfamily (Fig 2; S1 Fig).

To analyze the impact of additional WGDs on Nr2f genes, which took place in teleosts [17, 18], and specifically, in salmonids [49], we surveyed the Nr2f proteins of 12 teleost species (Fig 2; S1 Fig). Consistent with the WGDs in these species, there was a tremendous expansion of the Nr2f family in this clade, although it was accompanied by differential Nr2f paralog losses in some species (Fig 2; S1 Fig). To further interrogate the evolution of the Nr2f proteins, we examined alignments of the highly conserved zinc-fingers (Znf) within their DNA-binding domains (DBDs) using representatives from each subfamily (Fig 3). Although there is a high degree of conservation in all the examined Nr2fs, the amino acid changes in the DBDs parallels the phylogenetic results of the whole proteins. The Nr2f1/2/5/6 proteins of early-branching eumetazoans showed a high degree of variability and multiple differences with respect to Nr2f1/2/5/6 DBDs of protostome and deuterostome invertebrates and the Nr2f DBDs in vertebrates. There is high similarity between Nr2f1/2, Nr2f1 and Nr2f2 DBDs in agnathans and gnathostomes, whereas Nr2f5 and Nr2f6 DBDs of gnathostomes exhibited specific changes that are consistent with their positions in the phylogenetic trees (Figs 2 and 3). Interestingly, single amino acid changes found in most Nr2f5 and Nr2f6 proteins are also found in some early-branching eumetazoans and invertebrate Nr2fs. However, the functional significance of these changes, if any, is not clear. Thus, our phylogenetic reconstruction of Nr2f genes in metazoans overall shows the presence of single orthologs in invertebrates and a significant expansion of the family in vertebrates that is punctuated with independent losses of Nr2f5 in some cartilaginous fishes and amniotes.

Fig 3

Zinc finger (Znf) motifs within the DBD of the Nr2f family.

Alignments of first (I) and second (II) Znfs found in Nr2f TFs. Yellow represents highly conserved amino acids throughout all species. White indicates amino acids that are not conserved. Turquoise and blue indicate amino acid changes that are conserved within Znf I of Nr2f5 and -6, respectively. The valine change found in some Nr2f6 LBDs is also found in the placozoan and cnidaria Nr2fs. Magenta and red indicate amino acid changes at the same residue that are conserved within Znf II of Nr2f5 and Nr2f6, respectively. A glycine residue is also found at the same position in some cnidaria and invertebrate Nr2fs. Green indicates a conserved change found in most Nr2f5 and Nr2f6 Znf IIs.

Nr2f genes have conserved intron codes

To complement the phylogenetic analysis of Nr2f genes, we first analyzed the conservation of Nr2f intron/exon structure [50-52]. Intron/exon junctions from early-branching eumetazoans and vertebrates matching the transcripts and the translated proteins were mapped and given a score for the intron phases (S4 File), with 0, 1 and 2 introns falling before the first, second and third bases of a codon, respectively. The introns were then mapped on a protein alignment comprising the highly conserved Nr2f protein DBDs and LBDs (S4 File). We found that two “phase 1” introns (one within the 3’ end of the DBD and one within the LBD) are preserved in all the extant Nr2f subfamilies (Fig 4A; S4 File). However, Nr2f6 genes also have a “phase 2” intron inside the second zinc-finger domain belonging to the DBD (Fig 4A; S4 File). The conservation of intron/exon junctions in the examined Nr2f genes allows two groups to be distinguished: one constituted by Nr2f, Nr2f1/2/5/6, Nr2f1/2, Nr2f1, Nr2f2, Nr2f5, and one comprising only vertebrate Nr2f6 (Fig 4B), implying this unique intron/exon boundary originated after the duplication event that generated Nr2f5 and Nr2f6. Thus, our analysis of intron/exon boundaries demonstrates the existence of a highly conserved intron code throughout eumetazoan Nr2f family members and the divergence of Nr2f6 genes following the second WGD.

Fig 4

Intron code of the Nr2f family in metazoans.

(A) Protein alignment showing conservation of intron/exon structures within the DBDs (black) and LBDs (red) of Nr2f members. Znfs in the DBDs are underlined. Phase 0 introns—yellow, phase 1 introns—green, and phase 2 introns—turquoise. Asterisks indicate 100% amino acid conservation. Colons indicate high levels (>90%) amino acid conservation. Periods indicate moderate levels (50–89%) of amino acid conservation. (B) Schematization of intron/exon boundaries of Nr2f genes as they relate the Nr2f protein DBD and LBDs. Black box indicates DBD. Purple boxes represent the zinc-fingers motifs within the DBD. Red boxes indicate the LDB. Colored bars indicate the conserved Nr2f Phase 1 introns (green) and the Nr2f6-specific Phase 2 intron (turquoise).

Synteny analysis defines differential duplications and losses in the Nr2f family

In order to confirm the specific homologies indicated from the phylogenetic analysis, we next carried out an examination of synteny within the Nr2f genomic environments. With respect to representatives of the more ancient Nr2f genes, we did not find evidence of synteny between the single Nr2f in the placozoan T. adhaerens and the multiple Nr2fs in cnidarians. However, the location of the four Nr2f genes in N. vectensis and A. millepora genomes indicates they were likely derived from an initial duplication event followed by a tandem duplication event (Fig 5). Interestingly, Mef2 and Rbm8 homologs were associated with Nr2f1/2/5/6b in N. vectensis and A. millepora, while an Arrdc homolog is associated with Nr2f1/2/5/6a in N. vectensis. In vertebrates, Mef2 paralogs (Mef2c, Mef2b, Mef2b) are associated with Nr2f1, Nr2f2, and Nr2f6, Arrdc paralogs (Arrdc3, Arrdc4, Txnip/Arrdc6) are associated with Nr2f1, Nr2f2, and Nr2f5, and Rbm8a is associated with Nr2f5 (Figs 6–8), implying an ancient association of these genes within eumetazoan genomes.

Fig 5

Synteny analysis of Nr2f1/2/5/6 genes in cnidaria.

Schematic of the loci flanking Nr2f1/2/5/6 gene duplications in cnidaria N. vectensis, A. millepora, and H vulgaris. Only the one H. vulgaris gene presently has available genomic information. Arrows indicate transcription orientation.

Fig 6

Synteny analysis of vertebrate Nr2f1 and Nr2f2 genes.

Schematization of conserved genomic environments of gnathostome Nr2f1 and Nr2f2 genes (red rectangles) in selected species with relative chromosomes/scaffolds. Flanking orthologous genes are represented employing rectangles of the same color. Arrows indicate transcription orientation.

Fig 8

Synteny analysis of vertebrate Nr2f6 genes.

Schematization of conserved genomic environments of gnathostome Nr2f6 genes (red rectangles) in selected species with relative chromosomes/scaffolds. Flanking orthologous genes are represented using the same color code. Arrows indicate transcription orientation.

In invertebrates, despite the synteny suggested between Nr2f, Mef2, Arrdc, and Rbm8 genes in cnidaria and vertebrates, we only found limited preservation of the Nr2f1/2/5/6 loci between two slow-evolving deuterostomes: the amphioxus (B. belcheri) [53] and the hemichordate (S. kowalevskii) [54] (S2 Fig). However, the limited synteny still corroborates the existence of the invertebrate Nr2f1/2/5/6 cluster shown in the phylogenetic trees (Fig 2; S1 Fig). Furthermore, the only remaining synteny between Nr2f1/2/5/6 in invertebrates and vertebrate orthologs appears to be the linkage between UNC45A and NR2F2 of primates and Unc45a and Nr2f1/2/5/6 of the tunicate C. robusta (S3 Fig), which is considered the closest living relative of vertebrates [55]. Focusing on Nr2f1 and Nr2f2 in the genomes of gnathostomes, including Great white sharks, coelacanths, spotted gars, zebrafish, chickens, and humans, we found a high degree of synteny for Nr2f1 and Nr2f2 loci and conservation of the location of flanking genes among these taxa (Fig 6). Specifically, Nr2f1 and Nr2f2 genes exhibited remarkably conserved syntenic environments, clustering with putative orthologs belonging to other families. Lysmd3, Arrdc3, Mctp1 and Mef2a flank Nr2f1 orthologs, while Nr2f2 orthologs are flanked by Lysmd4, Arrdc4, Mctp2 and Mef2c paralogs. Furthermore, in teleosts like zebrafish, two Nr2f1 Ohnologs (nr2f1a and nr2f1b) also shared significant conservation of paralogous genes (Fig 6), which is consistent with an origin from the teleost-specific genome duplication (TSGD) [17, 18]. However, the nr2f1b gene has been lost by several teleost species (Fig 2; S1 Fig). Although the genomic information is somewhat fragmented, orthologs of flanking genes found in gnathostome Nr2f1 and Nr2f2, such as Arrdc2/3, Lysmd3, Fam172a, were also found near each of the three Sea lamprey Nr2f1/2 genes and the hagfish Nr2f1/2C gene (S4 Fig), which is consistent with these genes arising from genome duplication(s) within the agnathan lineage [56]. Together, these results suggest that Nr2f1 and Nr2f2 of gnathostomes have a common origin and are derived from a WGD event [41, 42].

Examining Nr2f5 loci in representative gnathostomes showed a high degree of conservation in both species that have retained and lost the gene. The adjacent genomic environments in the majority of examined Nr2f5 loci have retained an association with Rbm8a (Fig 7), whose homolog in cnidarians flanks Nr2f1/2/5/6b (Fig 5). The synteny is generally not shared with gnathostome Nr2f1 and Nr2f2 orthologs (Fig 6). However, the Nr2f5 loci in coelacanth and amphibians have retained Txnip (Fig 7), which is also named Arrdc6. As aforementioned, Arrdc family members flank the N. vectensis Nr2f1/2/5/6a (Fig 5) and both Nr2f1 and Nr2f2 genes (Fig 6). Interestingly, amniotes that have lost Nr2f5 (representatives including Chinese soft-shell turtles, chickens, and humans) (Fig 7; S3 File) have largely preserved the flanking genomic loci that are present in cartilaginous fish, zebrafish, coelacanth, frogs, and Green see turtles (Fig 7). In contrast, the absence of Nr2f5 in some cartilaginous fish, such as C. milii, correlates with the lack of the entire locus. Within the Actinopterygii (ray-finned fishes), the synteny of genes has been lost only on one side of the Nr2f5 loci (Fig 7). With respect to the lamprey, its Nr2f1/2C ortholog is flanked by a Bola1 ortholog, as well as orthologs of genes that flank gnathostome Nr2f1 and Nr2f2 (S4 Fig; Figs 6 and 7), which further suggests ancestral linkage with the single Nr2f1/2/5/6 genes (Fig 2; S1 Fig).

Fig 7

Synteny analysis of vertebrate Nr2f5 genes.

Schematization of conserved genomic environments of gnathostome Nr2f5 genes (red rectangles) in selected species with relative chromosomes/scaffolds. Flanking orthologous genes are represented using rectangles of the same color. Arrows indicate transcription orientation.

As might be expected given the divergence, the Nr2f6 subfamily did not share many common elements with the other Nr2f loci in gnathostomes (Fig 8). However, Mef2b was syntenic in Great white sharks, Spotted gar, chicken, and human genomes, similar to cnidarian Nr2f1/2/5/6b (Fig 5) and gnathostome Nr2f1 and Nr2f2 (Fig 6). Within gnathostomes, the Nr2f6 loci were highly conserved from cartilaginous fish to mammals, although there were significant gene losses surrounding nr2f6a and nr2f6b loci in zebrafish and one side of the Nr2f6 locus in humans. Furthermore, the presence of conserved orthologs (ano8a and ano8b, plvapa and plvapb) flanking nr2f6a and nr2f6b zebrafish genes suggested that they originated from the TSGD. Together, these findings show that despite the greater divergence of the Nr2f5 and Nr2f6 within vertebrates the genomic environments have retained some synteny and surrounding Nr2f5 and Nr2f6 loci are highly conserved within gnathostomes.

Effects of TSGD on the Nr2f gene repertoire

We next wanted to measure the impact of the series of additional WGDs that have occurred in teleosts on Nr2f gene number (Fig 2; S1 Fig). For this comparison, we examined all the Nr2f loci in zebrafish, the Asian arowana (S. formousus), which is documented to retain duplicates [57], and the Atlantic salmon (S. salar), which has a salmonid-specific genome duplication (SSGD) [49]. We found that each of these teleosts retained two Nr2f1 Ohnologs (Fig 9), suggesting they either were not duplicated or that one pair of Ohnologs was lost in salmonids. Zebrafish lost one nr2f2 Ohnolog, maintaining only the nr2f2a ortholog, while Asian arowana retained two Nr2f2 Ohnologs. Salmonids have 3 Nr2f2 genes (Nr2f2a1, Nr2f2a2, and Nr2f2b1), due to a loss of the one of the Nr2f2b Ohnologs following their additional WGD. With respect to Nr2f5, only Atlantic salmon showed two copies, implying these were generated during the SSGD event, as suggested by the presence of two Nr2f5 Ohnologs in other salmonids (Oncorhynchus spp., Coregonus clupeaformis) (S5 File). Finally, zebrafish and Asian arowana each possess two Nr2f6 genes, while Atlantic salmon has 3 similar to what is found in the Nr2f2 subfamily (Fig 9). Inspecting other teleost Nr2f gene family repertoires (Fig 2; S1 Fig), we found that the Channel catfish (Ictalurus punctatus), Red-bellied piranha (Pygocentrus nattereri), cavefish (Astyanax mexicanus) and Sheepshead minnow (Cyprinodon variegatus) all retained only Nr2f6b. The Sheepshead minnow and Princess cichlid (Neolamprologus brichardi) also lost Nr2f5. However, other cichlids like Nile tilapia (Oreochromis niloticus) and Zebra mbuna (Maylandia zebra) did not lose Nr2f5 (S5 File). Intriguingly, the Monterrey platifish (Xiphophorus couchianus) is the only gnathostome without any Nr2f1 paralogs, differing from its sibling species, the common platifish (X. maculatus), which possesses Nr2f1a. Therefore, teleosts show an expansion of Nr2f genes following TSGD and SSGD, which were followed by high variability in species-specific losses of Nr2f Ohnologs.

Fig 9

Synteny analysis of Nr2f genes in teleosts.

Comparison of Nr2f genome environments in selected teleosts (zebrafish, Asian arowana, Atlantic salmon) with relative chromosomes/scaffolds. Rectangles of the same color represent flanking orthologous genes. Arrows indicate transcription orientation.

Discussion

We have performed an examination of Nr2f gene evolution in metazoans. Our analysis corroborates previous work showing that Nr2f genes are present in some representative early-branching eumetazoans (placozoans and cnidarians) [15, 58], but that they are absent in early-branching metazoans, i.e. sponges and ctenophores [15, 58]. Importantly, our data support a model in which a single Nr2f gene, which is present in a representative placozoan, predated a Nr2f1/2/5/6 subfamily found in cnidaria and six Bilateria subfamilies that include Nr2f1/2/5/6 (found in invertebrate protostome and deuterostomes), Nr2f1/2 (found in agnathans), and Nr2f1, Nr2f2, Nr2f5, and Nr2f6 (found in vertebrates; Fig 10). Single, conserved Nr2f1/2/5/6 genes are predominantly found throughout invertebrate protostomes and deuterostomes and have even been retained in species traditionally considered gene losers, such as the tunicates [52, 59, 60]. There has been significant expansion and retention of Nr2fs in gnathostomes, particularly in teleosts. Although initial analysis in lampreys suggested they may possess only one Nr2f gene [61], our evolutionary assessment shows that extant agnathans have three Nr2f members, which appear to have originated in part from an agnathan WGD event [56]. Interestingly, the single Nr2f1/2/5/6 proteins in invertebrates are also highly conserved at the sequence level and cluster with the Nr2f1/2 proteins in agnathans and Nr2f1 and Nr2f2 proteins in gnathostomes. Furthermore, our data support a parsimonious view that Nr2f1 and Nr2f2 are paralogous and Nr2f5 and Nr2f6 are paralogous, consistent with each of the Nr2f1/2 and Nr2f5/6 branches being created from an initial WGD [41, 42]. Within gnathostomes, the genomic environments of each the Nr2f1, Nr2f2, Nr2f5, and Nr2f6 orthologs have retained significant synteny of their loci [16, 21, 22]. Remarkably, while limited synteny exists between the Nr2f1/2 and Nr2f5/6 branches and within the Nr2f5/6 branch, members of these families have retained association with Mef2, Arrdc, and Rbm8 homologs within their genomic environments, which is also found in cnidaria. However, this genomic association was not found in other examined invertebrate genomes. Our analysis also shows the Nr2f5 subfamily is the smallest in vertebrates, having been independently lost in multiple gnathostomes (some cartilaginous fishes, amniotes—some reptiles, absent in birds and mammals) (Fig 10). In contrast to Nr2f5, the Nr2f6 subfamily has been retained by all the evaluated gnathostomes, despite being the most divergent at the sequence level, with respect to synteny, and intron/exon structure.

Fig 10

Model summarizing the evolutionary events of the Nr2f family in Metazoa.

A single Nr2f of placozoans (white box) represents the ancestor of extant Nr2fs. There were duplicative events specific to cnidaria leading to the expansion of Nr2f1/2/5/6 (pink circle). Invertebrate protostomes and deuterostomes have predominantly retained a single Nr2f1/2/5/6 homolog. There were duplicative events specific to agnathans leading to an expansion of Nr2f1/2 genes (orange circle). WGDs within vertebrates (green circles) generated the four Nr2fs found in vertebrates, with Nr2f1/Nr2f2 being paralogous and Nr2f5/Nr2f6 being paralogous. Nr2f5 has been independently lost in multiple vertebrate groups (red circle). It is lost in some cartilaginous fish and turtles (reptilian amniotes), and is absent in avian and mammalian amniotes. Teleosts have additional Nr2f Ohnologs due to TSGDs (blue circles) and SSGDs (yellow circles).

Although overall there has been relatively limited comparative analysis of Nr2f gene expression beyond major model organisms, integrating our phylogenetic assessment with available expression and functional analyses of the Nr2f members in evolutionarily distant animals [12, 14] presently supports a hypothesis that Nr2f expression originated in neural tissue and regulation of neuronal differentiation may be the most ancient Nr2f function. Foremost, the two Nr2f members (both Nr2f1/2/5/6c) of the diploblastic cnidaria H. vulgaris and H. echinata thus far examined appear to be expressed in neurons and have requirements in neurogenesis [12, 14]. Clearly, the expression of the additional Nr2f cnidarian homologs that have been identified needs to be examined and if found to be expressed in endoderm would alter this hypothesis. Nevertheless, the function of Nr2f1/2/5/6 orthologs of protostome invertebrates nematodes and flies have been extensively studied in neural tissues and neural sensory cell differentiation [8, 62, 63]. In invertebrate deuterostomes, the single Nr2f1/2/5/6 orthologs are expressed in neural tissue of sea urchin (Strongylocentrotus purpuratus), amphioxus, and sea squirt embryos [11, 64–66]. Recent functional analysis of the Mediterranean sea urchin (Paracentrotus lividus) Nr2f1/2/5/6 shows that it is required for the development of neural and ectodermal derivatives [67]. A Nr2f1/2/5/6 ortholog from the agnathan River lamprey (Lampetra fluviatilis) is also expressed in the developing nervous system [61]. However, our identification of three Nr2f1/2 members in agnathans suggests that additional expression and potentially functional analysis should be performed in the Sea lamprey (P. marinus) and/or hagfish (E. burgeri) to understand the conservation of the different agnathan paralogs compared to Nr2fs in vertebrates. Both Nr2f1 and Nr2f2 orthologs share overlapping central nervous system (CNS) expression in mouse and zebrafish [16, 21, 68]. However, nr2f1a and nr2f2 are both expressed more extensively in neural tissue of zebrafish embryos, while Nr2f1 is predominantly expressed in neural tissues of mice [21, 22]. Nr2f5 is expressed in neural tissue and derivatives, including in the eyes of zebrafish and newts [68-70]. Nr2f6 genes have conserved expression within the central nervous system of mammals [16], as well as both zebrafish nr2f6 Ohnologs. Thus, all Nr2fs examined are expressed in neural tissue, with experiments in cnidaria and invertebrates presently supporting their ancestral requirements may be in neural cell differentiation.

While we propose that Nr2fs may have originated with requirements in neural differentiation, they are also required for the development of mesodermal and endodermal-derived tissues through Bilateria. Thus, it is interesting to consider some of these requirements in light of our phylogenetic analysis. In addition to neural differentiation, Nr2f homologs are necessary for copulation control in nematodes [71] and heart vessel specification in flies [23, 24]. Furthermore, the recent work with the Mediterranean sea urchin suggests that it is required for the development of mesendodermal derivatives [67]. The functions of Nr2f1 and Nr2f2 genes have been intensely investigated in vertebrate models and they are required for proper human development [22, 31, 72]. Both expression and functional analysis of Nr2f1 and Nr2f2 genes in vertebrates show that they have acquired distinct developmental roles during evolution. Following overlapping expression early in mouse embryos, murine Nr2f1 and Nr2f2 become predominantly expressed in neural and mesendodermal tissues, respectively [21, 22]. Analysis of these Nr2f genes in mice and zebrafish support the functional divergence of these proteins. Murine Nr2f1 KOs have glial differentiation defects [73], while Nr2f2 is required for proper development of many mesendodermal-derived tissues, including atrial chamber and arterial-venous differentiation [40, 74]. Intriguingly, mouse Nr2f2 and zebrafish nr2f1a are functional homologs with respect to heart development, as both are required for atrial differentiation [75], further supporting the common evolutionary origins of these paralogs. While zebrafish nr2f2 is not required for early atrial or vein development [29], NR2F1 and NR2F2 TFs do appear to have redundant requirements, for instance promoting atrial cardiomyocyte differentiation in human embryonic stem cells [76, 77]. It is interesting that the single Nr2f1/2/5/6 (svp/Nr2f3) homolog of flies is also required for dorsal vessel (heart) development [23]. However, if these similar roles in mesodermally-derived heart tissues reflect homologous requirements within Bilateria for cardiac differentiation requires functional studies from many additional model organisms [67]. With respect to analysis of the expansion of Nr2f1 and Nr2f2 Ohnologs in teleosts, Nr2f1b actually has been lost in the majority of surveyed teleosts. Nr2f1b zebrafish mutants are viable [78] and surprisingly do not exhibit redundancy with nr2f1a in atrial cardiomyocyte differentiation [29], but do exhibit some redundancy with multiple other Nr2f genes in neural crest cells that promote jaw development [78]. Virtually all the analyzed gnathostome genomes have a single Nr2f2 gene, excluding the teleosts S. formosus (2) and S. salar (3), implying there may be some dosage sensitivity that favors the retention of single orthologs in gnathostomes.

With respect to the function of Nr2f5 and Nr2f6 genes, zebrafish nr2f5 mutants are viable, yet like zebrafish nr2f1b mutants they function redundantly with other nr2f genes for proper upper-jaw development [78]. While expression and functional analysis from other organisms that have retained Nr2f5 (coelacanth, spotted gar, and frog) may provide insights into conservation of Nr2f5 orthologs, the independent loss of Nr2f5 genes in multiple vertebrate lineages, as well as the lack of overt requirements alone in zebrafish, suggests that Nr2f5 orthologs likely have retained minimal developmental requirements and its loss can be tolerated. Murine Nr2f6 KO mice have forebrain defects. Specifically, these mutants show a loss of neurons that regulate the circadian clock genes [79]. However, Nr2f6 also has a critical role in lymphocyte differentiation and T-cell mediated tumor surveillance, suggesting requirements in mesodermally-derived tissues and neofunctionalization in adaptive immunity [80, 81]. Altogether, minimally, expression and functional data support requirements for Nr2fs in all three germ layers of Bilateria. However, the conservation of these requirements and if they reflect homologous roles in the different germ layers throughout Bilateria is not yet as clear.

In examining the evolution of the Nr2f TFs, it is also worthwhile to note that in early-branching eumetazoans through invertebrate chordates and gnathostomes there is conserved responsiveness to retinoic acid (RA) signaling [82], a critical molecule involved early patterning of vertebrate embryos [83-85], implying this relationship may form the core of an ancient gene regulatory network. Nr2f genes from placozoans [58] and the invertebrate chordates Ciona and amphioxus are all RA-responsive [11, 64]. Furthermore, in vertebrates, where the earliest requirement for RA is posteriorization of the embryo [86], virtually all the Nr2f genes have been shown to be responsive to RA signaling in developmental contexts involving all three germ layers. Specifically, RA signaling has been shown to positively regulate all the Nr2fs in zebrafish in the developing zebrafish endoderm [87], the CNS [68], and anterior lateral plate mesoderm (ALPM) [29]. RA signaling also positively regulates Nr2f1, Nr2f2 and Nr2f6 in mice [88, 89], and NR2F1 and NR2F2 in humans [90, 91]. Nr2fs can inhibit RA signaling in some contexts, suggesting it may form a negative feedback loop. One role Nr2fs may play is through direct competition with retinoic acid receptors (RARs) in binding retinoic acid response elements (RAREs) [21]. Moreover, it has been shown that the cnidarian Nr2f1/2/5/6c possesses the ability to inhibit RA signaling in in vitro signaling assays [14]. Thus, the responsiveness of the Nr2f family to RA may have evolved very early and has been highly maintained through the diversification of multiple vertebrate Nr2f genes, implying there is high selection to maintain this relationship.

Conclusions

Overall, our evolutionary assessment sheds new light on the events that have shaped the extant Nr2f family in Metazoa. The phylogenetic analysis defines the individual Nr2f subfamilies and their relationships across metazoan phyla, which complements available expression and functional data presently supporting an origin of their requirements in the development of neural tissue. Interestingly, the functions of Nr2f proteins are found to regulate development of all germ layers of Bilateria. The detailed evolutionary understanding of the Nr2f gene family we now have will allow us to infer more meaningful conclusions about the origins and conserved requirements of Nr2f genes in normal metazoan development and their role in sculpting diverse body plans.

Methods

Ethics statement

Ethical approval is not required. No animals were used in this study.

Genome database searches and phylogenetic reconstruction

Homo sapiens NR2F protein sequences were employed as queries in BLASTp and tBLASTn in genome databases of selected species (NCBI, Ensembl, Ensembl Metazoa, SkateBase [92], ANISEED [93]). The entire dataset of protein sequences for domain architecture was analyzed by using the domain database provided by Expasy, named PROSITE [94] and then, manually annotated. All the surveyed sequences were verified to be Nr2f proteins through analysis of DBDs and LBDs (S6 File). The analysis was weighted with 30 species from agnathans to primates to take into account the impact of multiple WGDs in vertebrates [41, 42] and in teleosts [17, 18]. Orthology of the Nr2f members was initially assessed by using a reciprocal best blast hit (RBBH) approach employing default parameters and corroborated by phylogenetic analyses. Protein alignment for phylogeny was generated using L-INS-i (accurate; for alignment of <200 sequences) on MAFFT [95, 96] (S7 File). The phylogenetic reconstruction of Fig 2 was performed on the entire protein sequences and based on maximum-likelihood (ML) inferences calculated with PhyML 3.0 [97], employing automatic Akaike Information Criterion (AIC) by Smart Model Substitution (SMS) [98], which selected the JTT+G+F model employing discrete gamma distribution in categories. All parameters (gamma shape = 0.7; proportion of invariants (fixed) = 0.000) were established from the dataset. Branch support was provided by aLRT [99]. The phylogeny of S1 Fig was carried out employing Bayesian Information Criterion (BIC) by SMS, which sorted the JTT+G+F model using discrete gamma distribution in categories. All parameters (gamma shape = 0.7; proportion of invariants (fixed) = 0.000) were established from the dataset, with branch support calculated employing aBayes method [100]. Accession numbers and protein sequences used for phylogenetic tree reconstructions are provided in S1, S6, and S7 Files, while those excluded for their divergence are listed in S2 File. Common and Latin names for species used in this study are listed in S8 File.

Analysis of intron/exon structures and phases

Gene structures were deduced after merging the genomic sequences with ESTs when available, as previously described [50-52]. Introns were classified as phase 0, phase 1, and phase 2, according to their positions with respect to the protein-reading frame. The amino-acid residues with the conserved introns were manually mapped on a ClustalX alignment [101] of selected Nr2f proteins (S9 File).

Evaluation of synteny

We evaluated the presence/absence of synteny examining the chromosomes on public genome databases (NCBI, Ensembl, Ensembl Metazoa, ANISEED [93]). We verified the existence of duplicates using Genomicus [102] and Vertebrate Ohnologs [103]. The window considered for the locus analyses was twenty flanking genes. Genes that were not conserved were excluded from the analysis. All the genes were represented employing colored rectangles, using the same color for all Nr2f genes (red).

Phylogenetic tree of the Nr2f family, using Bayesian Information Criterion (BIC).

The same color code as Fig 2 is used. Values at the branches indicate replicates obtained employing the aBayes method.

(TIF)

Click here for additional data file.

Synteny analysis of Nr2f1/2/5/6 genes found invertebrates.

Schematic of limited conservation for Nr2f1/2/5/6 loci between the hemichordate S. kowalevskii and amphioxus B. belcheri. Black arrows indicate transcription orientation.

(TIF)

Click here for additional data file.

Unc45–Nr2f gene duplet preservation.

Schematic of Unc45-Nr2f duplet conservation in genomes of ascidians (Ciona) and primates. The duplet is absent in other vertebrate models, including zebrafish and mouse.

(TIF)

Click here for additional data file.

Synteny analysis of Nr2f1/2 genes in agnathans.

Schematic of lamprey (P. marinus) Nr2f1/2 loci with relative chromosomes and available genomic data from the hagfish (E. burgeri). Genomic data could only be obtained for the hagfish Nr2f1/2C gene. Same color code of Figs 6–8 is used. Flanking genes are in common with gnathostomes, with Arrdc2 and Arrdc3 (green) that form a conserved duplet with Nr2f1/2B and Nr2f1/2C. Nr2f1/2C is adjacent to Fam172a in both lamprey and hagfish. Arrows indicate transcription orientation.

(TIF)

Click here for additional data file.

List of all protein sequences employed in Nr2f phylogenetic tree with accession numbers.

(TXT)

Click here for additional data file.

List of protein sequences excluded from Nr2f phylogenetic tree due to their high degree of divergence.

(TXT)

Click here for additional data file.

List of examined species whose genomes lacked Nr2f5 with their common names, Latin names, and phyla.

(XLSX)

Click here for additional data file.

Intron/Exon structure of Nr2f genes in Metazoa.

Alignment of specific and conserved intron/exon boundaries within the Zinc finger motifs of DBD (underlined) and LBDs (red). The intron phases have been depicted using color code: Phase 0 (yellow), Phase 1 (green), Phase 2 (turquoise).

(DOCX)

Click here for additional data file.

Sequences of additional salmonid and cichlid Nr2f5 proteins.

(TXT)

Click here for additional data file.

Nr2f domain architectures during metazoan evolution.

Sequences used in analysis with DBDs (yellow) and LBDs (green) domains in metazoan Nr2f proteins indicated. the Zinc-finger motifs within the DBDs are underlined.

(DOCX)

Click here for additional data file.

MAFFT alignment of protein sequences used for phylogenetic analysis of Fig 2 and S1 Fig.

(TXT)

Click here for additional data file.

List of species used for our evolutionary analyses with their common names, Latin names, and phyla.

(XLSX)

Click here for additional data file.

Selected Nr2f transcripts and translations used for analysis with the positions and phases of intron/exon boundaries indicated.

(DOCX)

Click here for additional data file.

16 Jul 2021

PONE-D-21-18033

Origin and evolutionary landscape of Nr2f transcription factors across Metazoa

PLOS ONE

Dear Dr. Waxman,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that comprehensively addresses the points raised during the review process.

Please submit your revised manuscript by Aug 30 2021 11:59PM. If you will need more time than this to complete your revisions, please reply to this message or contact the journal office at plosone@plos.org. When you're ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

Please include the following items when submitting your revised manuscript:

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: http://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols. Additionally, PLOS ONE offers an option for publishing peer-reviewed Lab Protocol articles, which describe protocols hosted on protocols.io. Read more information on sharing protocols at https://plos.org/protocols?utm_medium=editorial-email&utm_source=authorletters&utm_campaign=protocols.

We look forward to receiving your revised manuscript.

Kind regards,

Michael Schubert

Academic Editor

PLOS ONE

Journal Requirements:

When submitting your revision, we need you to address these additional requirements.

1. Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at and https://journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

2. We note that you have included the phrase “data not shown” in your manuscript. Unfortunately, this does not meet our data sharing requirements. PLOS does not permit references to inaccessible data. We require that authors provide all relevant data within the paper, Supporting Information files, or in an acceptable, public repository. Please add a citation to support this phrase or upload the data that corresponds with these findings to a stable repository (such as Figshare or Dryad) and provide and URLs, DOIs, or accession numbers that may be used to access these data. Or, if the data are not a core part of the research being presented in your study, we ask that you remove the phrase that refers to these data.

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Partly

Reviewer #2: No

**********

2. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

Reviewer #2: Yes

**********

3. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #2: Yes

**********

4. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: Yes

**********

5. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: This papers provides a very useful in depth actualisation of the NR2F repertoire, especially in vertebrates. However, at the eumetazoan scale, it it less extensive, and thus in the present state leads to less robust conclusions.

The weakest point is the data substantiating the author's hypothesis that NR2F6 was lost in most bilaterians but retained only in gnathostomes. Actually, this may be true but should be backed by more extensive sampling of NR2F sequences in cnidarians. The minimum would be to incorporate all the sequences previousely discussed as cnidarian NR2Fs, so four, not only three N. vectensis paralogs, and the sequences from H. vulgaris (mentionned in the text) and from A. millepora. Those were already incorporated in the Bridgham et al Plos Biology paper form 2010. See Supp Fig 2. More recently, an additional sequence from the hydrozoan Hydractinia echinata has been incorporated into a global NR analysis (Beinsteiner et al., Plos Genetics, 2021, Supp Fig 3), suggesting that this sequence belongs also to the NR2F group. I am pretty sure the authors could significantly improve the cnidarian dataset just by doing targeted blast queries on cnidarian sequences available in Genbank. With that they would probably get the data necessarily to back their hypothesis more solidely.

Another problem is that some sequences discussed in the text (like the nematode UNC-55) and the three hagfish sequences are missing. The hagfish ones should be present in the tree to demonstrate their orthology to the lamprey ones. Regarding the UNC-55, I can guess that the author were embarrassed by its unstable placing. According to Bertrand et al., MBE 2004, figure 1, this sequence did not branch clearly with the drosophila SVP/NR2F3. This is probably a long-branch artefact but could be easily resolved incorporating other nematode sequences like the one from Brugia malayi (EDP32461.1). Perhaps the authors could even found a Priapulid sequence that may help group together nematodes and arthropods.

Regarding nomenclature, it would be preferable to keep the name NR2F3 for the group the author call here NR2F1/2/5/6 should be named NR2F3, regarding the fact that the SVP Drosophila sequence was used to define that group in the official nomenclature paper of 1999 (doi:10.1016/s0092-8674(00)80726-6). At least for the ecdysozoan (or even protostome sequences) that are orthologous. Also, it should be mentionned somewhere in the paper that the xenopus sequence that the authors put in the NR2F5 group was initially named NR2F4. Actually, the coelacanth sequence could be also named NR2F4, being closer to the xenopus one than to the teleost one. And in that case, a name like NR2F4/5 would be appropriate for the entire gnathostome group. In any case, for all readers that are not familiar with the nomenclature, it would be useful to add trivial names, for sequences that have already been functionally characterized.

Regarding the rooting of the tree, the T. adhaerens sequence is not strickly speaking a NR2F. Its weak branching at the basis of the NR2F in the global trees of Bridgham and Beinsteiner indicates that it could be homologous to both NR2F and NR2E families in eumetazoans. Therefore it has been called NR2I in the Beinsteiner paper (see Fig 7). Based on this interpretation, I recommend rooting the tree using this sequence instead of the full cnidarian/placozoan group during the next analysis run.

Fig3: the comparison is unsufficiently exploited here. The authors should clearly highlight the colored position in the zinc fingers that are also present in Nematostella and Trichoplax. For example, the turquoise L residue from X. tropicalis is also present in T. adhaerens. The blue V is present in all thre N. vectensis and in the T. adharens sequences. The magenta G is present in two N. vectensis sequences. The green D is present in one N. vectensis sequence. An then, are those changes linked of functional significance? Did the author look at structural data related to the DBD?

Regarding the discussion of ancestral functions, it is hard to follow as currently written. I would recommend that the author implement a caracter mapping approach using the phytools package in R, mesquite or whatever program they would be comfortable with. This would once again give more solidity to their claims regarding the ancestrality of nervous system expression of NR2F genes.

Typos:

L64: which two contains two Zinc finger motifs -> which contains

L76: Kruppel-like 9 -> Krüppel-like factor 9

Reviewer #2: In this manuscript Coppola and Waxman address the evolution of NR2F transcription factors in this manuscript. This is an important group of nuclear receptors, which display a complex evolutionary history. The authors have made an effort to elucidate some outstanding questions regarding the evolution of these genes. I raise below a few questions that deserve attention prior to acceptance.

1. The authors refer NR2F5 absent from “aminotes” – confirm, as am I sure that there are orthologues in bird and reptile genomes. This needs clarification.

2. The authors consider NR2F originating “pre-metazoan” (line 8)– this cannot be. NRs emerged in the ancestor of Metazoa (check and cite https://doi.org/10.1371/journal.pbio.1000497). Clarify conclusions. NR2F emerged in the ancestor of Bilateria. Not including Porifera and others groups does not allow the sort of conclusions described in the manuscript.

3. NR2F6: unclear why the authors suggest this have been lost independently in so many lineages (figure 9; a more parsimonious scenario would be that NR2f6 is a highly divergent paralogue from 2R genome duplications? It would be relevant to investigate the paralogy relationships between these regions).

4. NR2F5 in cartilaginous fish: maybe consider sampling more species using a synteny based approach. Use Bola1 as an anchor gene? Sample more genomes available and readdress this conclusion. Please have a look and cite: DOI: 10.1016/j.ygcen.2020.113527

5. “Pre-metazoan models Amphimedon queenslandica (sponge) and Mnemiopsis leidyi (ctenophore)”: you mean pre Bilateria? Correct throughout the manuscript.

**********

6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: No

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email PLOS at figures@plos.org. Please note that Supporting Information files do not need this step.

28 Sep 2021

Response to Reviewers: PONE-D-21-18033

We thank both reviewers for their thoughtful and insightful critiques of our manuscript. Both reviewers recognized the importance of understanding the evolution of Nr2f transcription factors. While supportive of the work, they both provided helpful suggestions for additional analysis that would improve our study. In particular, these questions centered around the relationship of cnidarian Nr2fs and the Nr2f6 family within the phylogenetic analyses, and the presence of Nr2f5 orthologues within vertebrates. In the revised manuscript, we have tried to directly address all the reviewers’ comments. The new analyses performed in response to the reviewers’ comments have helped to resolve discrepancies within the original manuscript and dramatically improved our study. Responses to the reviewers’ specific critiques are indicated below in blue. Changes made within the revised manuscript in response to the reviewers’ comments are indicated in the manuscript version with track changes.

--------------------

Reviewer #1: This papers provides a very useful in depth actualisation of the NR2F repertoire, especially in vertebrates. However, at the eumetazoan scale, it it less extensive, and thus in the present state leads to less robust conclusions.

1. The weakest point is the data substantiating the author's hypothesis that NR2F6 was lost in most bilaterians but retained only in gnathostomes. Actually, this may be true but should be backed by more extensive sampling of NR2F sequences in cnidarians. The minimum would be to incorporate all the sequences previousely discussed as cnidarian NR2Fs, so four, not only three N. vectensis paralogs, and the sequences from H. vulgaris (mentionned in the text) and from A. millepora. Those were already incorporated in the Bridgham et al Plos Biology paper form 2010. See Supp Fig 2. More recently, an additional sequence from the hydrozoan Hydractinia echinata has been incorporated into a global NR analysis (Beinsteiner et al., Plos Genetics, 2021, Supp Fig 3), suggesting that this sequence belongs also to the NR2F group. I am pretty sure the authors could significantly improve the cnidarian dataset just by doing targeted blast queries on cnidarian sequences available in Genbank. With that they would probably get the data necessarily to back their hypothesis more solidely.

As Reviewer #1 suggested, we have included additional cnidarian Nr2f sequences. We found that there are 4 cnidarian Nr2fs in N. vectensis and A. millepora, and 3 in H. vulgaris. We could not identify additional H. echinate Nr2fs based on available data. We found that the H. vulgaris, H. echinata, and an A. millepora Nr2f were significantly divergent, which caused long branch artifacts in the phylogenetic tress. Therefore, the analysis shown has excluded those. Nevertheless, the incorporation of the additional cnidarian sequences has improved our analysis of Nr2fs within cnidaria and the overall phylogenetic trees, and support that the additional Nr2fs in cnidaria originated from cnidaria-specific duplication events. The additional data is reported in the revised Fig 2, S1 Fig, Fig 5, on lines 120-127 of the revised manuscript, and included in the revised Abstract.

Regarding the Nr2f6 origin hypothesis reported in the original manuscript, we agree that it may have been confusing and needed greater support. However, it was based on our data at the time. In addition to the incorporation of additional cnidaria sequences, we included more vertebrate Nr2f5 sequences, and rooted the trees with the placozoan Nr2f, as was suggested. While we have not determined the individual effects of each these changes, the new analysis supports a Nr2f5/Nr2f6 branch, suggesting that Nr2f5 and Nr2f6 are paralogous, which is a more parsimonious evolutionary hypothesis and consistent with whole genome duplications in vertebrates. Furthermore, additional analysis of synteny revealed the presence of shared genes (Mef2 TFs) between Nr2f1/2 and Nr2f6 loci of vertebrates (see other reviewer’s comments) and that cnidarian Nr2f genes are flanked by Mef2, Arrdc, and Rbm8 homologs, which are found flanking the vertebrate Nr2f genes. Thus, the incorporation of new sequence data and rooting supports the orthology between cnidarian Nr2f1/2/5/6 and vertebrates’ Nr2fs, and that Nr2f6 is derived from genome duplication events in vertebrates (see revised Fig 2, S1 Fig, Fig 5, and Fig 8, lines 157-171 and 309-312, and included in the revised Abstract).

2. Another problem is that some sequences discussed in the text (like the nematode UNC-55) and the three hagfish sequences are missing. The hagfish ones should be present in the tree to demonstrate their orthology to the lamprey ones. Regarding the UNC-55, I can guess that the author were embarrassed by its unstable placing. According to Bertrand et al., MBE 2004, figure 1, this sequence did not branch clearly with the drosophila SVP/NR2F3. This is probably a long-branch artefact but could be easily resolved incorporating other nematode sequences like the one from Brugia malayi (EDP32461.1). Perhaps the authors could even found a Priapulid sequence that may help group together nematodes and arthropods.

As Reviewer suggested, we have incorporated the hagfish Nr2fs in the revised tree (see revised Fig 2). However, there is still limited genomic information to perform syntenic analysis with hagfish. We have incorporated synteny analysis for the one hagfish Nr2f where were able to find genomic information.

Yes, there were a few cases where we observed long branch artifacts created by significantly divergent sequences, with examples including some cnidaria (as mentioned above) and nematodes. We tried to be transparent about this in the original text and stated this in the original Methods. Sequences that were not included in the phylogenetic trees due to their high divergence and that caused long branch artifacts were presented in the original S5 File (revised S2 File). We did take Reviewer #1’s suggestion regarding the incorporation of additional nematode sequences (B. malayi, O. vulvusls), as well as used C. elegans, to try to alleviate these issues. However, the incorporation of these sequences always caused long branch artifacts, and we consequently excluded. We were able to include a Priapulid sequence. Thus, these new results are present in the revised phylogenetic trees (Fig 2 and S1 Fig).

3. Regarding nomenclature, it would be preferable to keep the name NR2F3 for the group the author call here NR2F1/2/5/6 should be named NR2F3, regarding the fact that the SVP Drosophila sequence was used to define that group in the official nomenclature paper of 1999 (doi:10.1016/s0092-8674(00)80726-6). At least for the ecdysozoan (or even protostome sequences) that are orthologous. Also, it should be mentionned somewhere in the paper that the xenopus sequence that the authors put in the NR2F5 group was initially named NR2F4. Actually, the coelacanth sequence could be also named NR2F4, being closer to the xenopus one than to the teleost one. And in that case, a name like NR2F4/5 would be appropriate for the entire gnathostome group. In any case, for all readers that are not familiar with the nomenclature, it would be useful to add trivial names, for sequences that have already been functionally characterized.

We agree with Reviewer #1 that within the manuscript and as a community we should incorporate proper nomenclature for Nr2fs and nuclear receptors in general. However, it is not clear to us that Nr2f3 and Nr2f4 that were suggested from the nomenclature committee >20 years ago are suitable names based in current analysis. These names were based on significantly fewer sequences from Nr2f members and we cannot even access their data based on the outdated links in that paper. Moreover, those suggested names have not really been adopted within the literature. We propose the names should reflect the proper homology of the genes, which is why we have incorporated Nr2f1/2/5/6 and suggest potentially Nr2f within the revised text. In the revised manuscript, we have tried to explain the rationale for our nomenclature proposal more thoroughly (lines 132-139 of revised manuscript). In addition, we have mentioned the previous names svp and Nr2f3 to alleviate any confusion. We have also mentioned that previous groups referred to Nr2f4 in Xenopus, as was suggested (lines 166-168 of revised manuscript). Again, we do not think there is rationale for this name based on the evolutionary analysis. It would seem to be unnecessarily confusing to incorporate this name for reptiles and coelacanth moving forward and think the names based on the relationships shown here should be corrected.

4. Regarding the rooting of the tree, the T. adhaerens sequence is not strickly speaking a NR2F. Its weak branching at the basis of the NR2F in the global trees of Bridgham and Beinsteiner indicates that it could be homologous to both NR2F and NR2E families in eumetazoans. Therefore it has been called NR2I in the Beinsteiner paper (see Fig 7). Based on this interpretation, I recommend rooting the tree using this sequence instead of the full cnidarian/placozoan group during the next analysis run.

Thank you for the suggestions. As mentioned above, we did root the revised phylogenetic trees with placozoan Nr2f. However, we respectfully disagree with presently calling the placozoan “Nr2f” sequence used “Nr2I” based on the available data and analysis. In Beinsteiner et al, there is really no discussion of the rationale for the use of this name in the manuscript. However, in the phylogenetic trees presented in both Bridgham et al and Beinsteiner et al, the placozoan Nr2f gene segregates with other Nr2f genes and is not found at the base of Nr2f and Nr2e genes, which would provide greater evidence for it being “Nr2I.” Moreover, sequence alignments shows that the placozoan Nr2f protein is more highly conserved, sharing ~10% greater identity and ~30% similarity at the sequence level, with Nr2fs than Nr2fe in vertebrates, including key residues found in all Nr2f proteins. Additionally, as shown in this manuscript, the intron/exon structure of the placozoan Nr2f is conserved with invertebrate and vertebrate Nr2fs (Fig 4 and lines 211-212). This intron/exon structure is not conserved with Nr2e genes. Therefore, we think the present data support that the placozoan Nr2f sequence used should be named called “Nr2f” and not “Nr2I.”

5. Fig3: the comparison is unsufficiently exploited here. The authors should clearly highlight the colored position in the zinc fingers that are also present in Nematostella and Trichoplax. For example, the turquoise L residue from X. tropicalis is also present in T. adhaerens. The blue V is present in all thre N. vectensis and in the T. adharens sequences. The magenta G is present in two N. vectensis sequences. The green D is present in one N. vectensis sequence. An then, are those changes linked of functional significance? Did the author look at structural data related to the DBD?

We have revised Fig 3 to include more sequences and highlight the appropriate residues found throughout all the protein sequences shown (see revised Fig 3 and Fig 3 legend). Although there is quite a lot of structure information about LBDs, there is significantly less about Nr2f structures in DBD. Thus, we are not aware of structural information about the DBDs that we could use to make meaningful predictions about function. We have stated this in the revised text (lines 186-188).

6. Regarding the discussion of ancestral functions, it is hard to follow as currently written. I would recommend that the author implement a caracter mapping approach using the phytools package in R, mesquite or whatever program they would be comfortable with. This would once again give more solidity to their claims regarding the ancestrality of nervous system expression of NR2F genes.

Thank you for the suggestion. We apologize our discussion was confusing. While character mapping is certainly one approach that can be used to decipher hypothesis about ancestral function, it was not clear this analysis really added much given the limited data in a variety of models. Therefore, in the revised manuscript we tried to address this issue by clarifying and simplifying the Discussion regarding the ancestral functions of Nr2f genes within neural tissues based on current data (page lines 402-430).

7. Typos:

L64: which two contains two Zinc finger motifs -> which contains

L76: Kruppel-like 9 -> Krüppel-like factor 9

Thank you. We corrected these typos.

Reviewer #2: In this manuscript Coppola and Waxman address the evolution of NR2F transcription factors in this manuscript. This is an important group of nuclear receptors, which display a complex evolutionary history. The authors have made an effort to elucidate some outstanding questions regarding the evolution of these genes. I raise below a few questions that deserve attention prior to acceptance.

1. The authors refer NR2F5 absent from “aminotes” – confirm, as am I sure that there are orthologues in bird and reptile genomes. This needs clarification.

In the revised manuscript, we have tried to clarify the losses of Nr2f5 found in amniotes. Examining the genomes of additional vertebrate species, we have found Nr2f5 in some reptiles (alligators, lizards, and turtles). However, it also appears to be lost in some turtles (i.e. the soft-shell turtle) and we could not find it in any bird and mammalian genomes examined. Thus, the most parsimonious hypothesis is that Nr2f5 has had independent losses in some amniote lineages. In the revised manuscript, we have incorporated the additional analysis (revised Fig 2) and tried to clarify this point (lines 159-166 and lines 384-386). We have also provided a list of reptile, bird, and mammalian genomes in which we were not able to find Nr2f5 genes in the Supporting Information (revised S3 File).

2. The authors consider NR2F originating “pre-metazoan” (line 8)– this cannot be. NRs emerged in the ancestor of Metazoa (check and cite https://doi.org/10.1371/journal.pbio.1000497). Clarify conclusions. NR2F emerged in the ancestor of Bilateria. Not including Porifera and others groups does not allow the sort of conclusions described in the manuscript.

Thank you for pointing this out. We agree and apologize that our statements were confusing. Throughout the revised manuscript we have eliminated the term “pre-metazoan” and have tried to clarify the statements about the origins of the Nr2fs in metazoa and their evolution in eumetazoa.

3. NR2F6: unclear why the authors suggest this have been lost independently in so many lineages (figure 9; a more parsimonious scenario would be that NR2f6 is a highly divergent paralogue from 2R genome duplications? It would be relevant to investigate the paralogy relationships between these regions).

We agree that our previous hypothesis was not parsimonious based on the 2R genome duplications. We have now included additional sequences in our analysis, including more cnidaria Nr2fs and vertebrate Nr2f5 sequences, and rooted the phylogenetic trees with the placozoan Nr2f (see response to Reviewer #1, comment 1). This seems to have resolved this issue and our phylogenetic trees now supports the parsimonious scenario of a 2R-origin for Nr2f6 and that it is paralogous to Nr2f5. Furthermore, additional analysis of synteny shows that several Nr2f6 loci are flanked by Mef2b, while paralogues Mef2c and –a flank Nr2f1 and Nr2f2, respectively. These new analyses are incorporated in the text (lines 157-171 and 309-320), the revised phylogenetic trees (Fig 2 and S1 Fig), the revised synteny schematic (Fig 8), and mentioned in the revised Abstract.

4. NR2F5 in cartilaginous fish: maybe consider sampling more species using a synteny based approach. Use Bola1 as an anchor gene? Sample more genomes available and readdress this conclusion. Please have a look and cite: DOI: 10.1016/j.ygcen.2020.113527

Thank you. We have sampled additional cartilaginous species. We found that Nr2f5 is present in the Great white shark (C. carcharias) and incorporated the Nr2f5 sequence for the Whale shark reported in Fonseca et al. However, we were not able to identify Nr2f5 orthologues in ghost shark, catshark, and skates (S3 Fig), suggesting it had been independently lost in some cartilaginous fish. Importantly, the Great white shark genome allowed us to also perform syntenic analysis of the loci using Bola1 as an anchor as was suggested. These data are incorporated in the revised manuscript in revised Fig 2, S1 Fig, and Fig 7, and lines 159-166.

5. “Pre-metazoan models Amphimedon queenslandica (sponge) and Mnemiopsis leidyi (ctenophore)”: you mean pre Bilateria? Correct throughout the manuscript.

Thank you. Yes, we referred to those animals incorrectly. We have corrected this mistake throughout the manuscript.

Submitted filename: Response to Reviewers_PONE-D-21-18033.docx

Click here for additional data file.

8 Nov 2021

Origin and evolutionary landscape of Nr2f transcription factors across Metazoa

PONE-D-21-18033R1

Dear Dr. Waxman,

We’re pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it meets all outstanding technical requirements.

Within one week, you’ll receive an e-mail detailing the required amendments. When these have been addressed, you’ll receive a formal acceptance letter and your manuscript will be scheduled for publication.

An invoice for payment will follow shortly after the formal acceptance. To ensure an efficient process, please log into Editorial Manager at http://www.editorialmanager.com/pone/, click the 'Update My Information' link at the top of the page, and double check that your user information is up-to-date. If you have any billing related questions, please contact our Author Billing department directly at authorbilling@plos.org.

If your institution or institutions have a press office, please notify them about your upcoming paper to help maximize its impact. If they’ll be preparing press materials, please inform our press team as soon as possible -- no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

Kind regards,

Michael Schubert

Academic Editor

PLOS ONE

Additional Editor Comments (optional):

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. If the authors have adequately addressed your comments raised in a previous round of review and you feel that this manuscript is now acceptable for publication, you may indicate that here to bypass the “Comments to the Author” section, enter your conflict of interest statement in the “Confidential to Editor” section, and submit your "Accept" recommendation.

Reviewer #1: All comments have been addressed

Reviewer #2: All comments have been addressed

**********

2. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Yes

Reviewer #2: Yes

**********

3. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

Reviewer #2: Yes

**********

4. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #2: Yes

**********

5. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: Yes

**********

6. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: The authors addressed all requests in a satisfactory manner. The manuscript is therefore suitable for publication now.

Reviewer #2: The authors have addressed my comments in the first round of revision. Thus, I support the publication of this manuscript.

**********

7. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: No

12 Nov 2021

PONE-D-21-18033R1

Origin and evolutionary landscape of Nr2f transcription factors across Metazoa

Dear Dr. Waxman:

I'm pleased to inform you that your manuscript has been deemed suitable for publication in PLOS ONE. Congratulations! Your manuscript is now with our production department.

If your institution or institutions have a press office, please let them know about your upcoming paper now to help maximize its impact. If they'll be preparing press materials, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information please contact onepress@plos.org.

If we can help with anything else, please email us at plosone@plos.org.

Thank you for submitting your work to PLOS ONE and supporting open access.

Kind regards,

PLOS ONE Editorial Office Staff

on behalf of

Dr. Michael Schubert

Academic Editor

PLOS ONE