Step one to understanding the evolution of CD-NTases, STINGs, and viperins was to accumulate sequences for these proteins from throughout the eukaryotic tree. To seek for various immune homologs, we employed a hidden Markov mannequin (HMM) technique, which has excessive sensitivity, a low variety of false positives, and the flexibility to individually analyze a number of (probably independently evolving) domains in the identical protein [22–24]. We used this HMM technique to go looking the EukProt database, which has been developed to mirror the true scope of eukaryotic variety by means of the genomes and transcriptomes of almost 1,000 species, particularly chosen to span the eukaryotic tree [25]. EukProt incorporates sequences from NCBI and Ensemble, plus many diverged eukaryotic species not present in some other database, making it a novel useful resource for eukaryotic variety [25]. Whereas it may be difficult to accumulate various eukaryotic sequences from conventional databases attributable to an overrepresentation of metazoan knowledge [26], EukProt ameliorates this bias by downsampling historically overrepresented taxa.

To broaden our searches from preliminary animal homologs to eukaryotic sequences extra usually, we used iterative HMM searches of the EukProt database, incorporating the hits from every search into the following HMM. After utilizing this strategy to create pan-eukaryotic HMMs for every protein household, we then added in bacterial homologs to generate common HMMs (Figs 1A and S1), persevering with our iterative searches till we both failed to seek out any new protein sequences or started discovering proteins exterior of the household of curiosity (S1 Fig). To outline the boundaries that separated our proteins of curiosity from neighboring gene households, we centered on together with homologs that shared protein domains that outlined that household (see Supplies and strategies for area designations) and had been nearer to in-group sequences than the outgroup sequences on a phylogenetic tree (outgroup sequences are famous within the Supplies and strategies).

Our searches for CD-NTases, STINGs, and viperins recovered tons of of eukaryotic proteins from every household, together with a very giant variety of metazoan sequences (pink bars, Fig 1B). It isn’t shocking that we discovered so many metazoan homologs, as every of those proteins was found and characterised in metazoans and these animal genomes are usually of upper high quality than different taxa (S2 Fig). We additionally recovered homologs from different species unfold throughout the eukaryotic tree, demonstrating that our strategy may efficiently determine deeply diverged homologs (Fig 1B). Nonetheless, exterior of Metazoa, these homologs had been sparsely distributed, such that for many species in our dataset (711/993), we didn’t recuperate proteins from any of the three immune households examined (white house, lack of coloured bars, Fig 1B). Whereas a few of these absences could also be attributable to technical errors or dataset incompleteness (S2 Fig), we interpret this sample as a mirrored image of ongoing, repeated gene losses throughout eukaryotes, as has been discovered for different innate immune proteins [27–29] and different forms of gene households surveyed throughout eukaryotes [28,30–32]. Certainly, most of the species that lacked any of the immune homologs had been represented by high-quality datasets (Ex: Metazoa, Chlorplastida, and Fungi). Thus, though it’s all the time attainable that our strategy has missed some homologs, we consider the ensuing knowledge represents a good evaluation of the variety throughout eukaryotes, at the least for these species at present included inside EukProt.

Fig 1. HMM searches to seek out homologs throughout the eukaryotic Tree of Life.

(A) A schematic of the HMM search course of. Ranging from preliminary, animal-dominated HMM profiles for every protein household, we used iterative HMM searches of the EukProt database to generate pan-eukaryotic HMMs. These had been mixed with bacterial sequences to allow discovery of bacteria-like homologs in eukaryotes. Every set of searches was repeated till few or no further eukaryotic sequences had been recovered which was between 3 and 5 instances in all circumstances. (B) Phylogenetic tree of eukaryotes, with main supergroups colour coded. The peak of the coloured rectangles for every group is proportional to its species illustration in EukProt. Horizontal, coloured bars mark every eukaryotic species by which we discovered homologs of STINGs, CD-NTases, or viperins. White house signifies species the place we searched however didn’t recuperate any homologs. The CD-NTase hits are divided into the three eukaryotic superfamilies, outlined in Fig 2. Particular person knowledge can be found in S1 File. CD-NTase, cGAS-DncV-like nucleotidyltransferase; HMM, hidden Markov mannequin; STING, Stimulator of Interferon Genes.

https://doi.org/10.1371/journal.pbio.3002436.g001

We subsequent studied the evolution of the innate immune proteins, starting with cGAS and its broader household of CD-NTase enzymes. Following infections or mobile harm, cGAS binds cytosolic DNA and generates cyclic GMP-AMP (cGAMP) [33–36], which then prompts downstream immune responses through STING [35,37–39]. One other eukaryotic CD-NTase, 2′5′-Oligoadenylate Synthetase 1 (OAS1), synthesizes 2′,5′-oligoadenylates that bind and activate Ribonuclease L (RNase L) [40]. Activated RNase L is a potent endoribonuclease that degrades each host and viral RNA species, decreasing viral replication (reviewed right here [41,42]). Some bacterial CD-NTases resembling DncV behave just like animal cGAS; they’re activated by phage an infection and produce cGAMP [8,21,43]. These CD-NTases are generally discovered inside cyclic oligonucleotide-based antiphage signaling techniques (CBASS) throughout many bacterial phyla and archaea [8,21,44].

Along with the well-studied cGAS, a lot of different eukaryotic CD-NTases have been beforehand described: the OAS1 paralogs (OAS2/3), Male irregular 21-Like 1/2/3/4 (MAB21L1/2/3/4), Mab-21 area containing protein 2 (MB21D2), Mitochondrial dynamics protein 49/51(MID49/51), and Inositol 1,4,5 triphosphate receptor-interacting protein 1/2 (ITPRILP/1/2) [44]. Of those, cGAS and OAS1 are the perfect characterised and each play roles in immune signaling. Current work has proven that cGAS and associated animal proteins, the cGAS-like Receptors (cGLRs), are current in almost all metazoan taxa and generate various cyclic dinucleotide alerts [45]. Nonetheless, the immune features of Mab21L1 and MB21D2 stay unclear, though Mab21L1 has been proven to be vital for growth [46–48].

To research the evolutionary historical past of the eukaryotic CD-NTases, we searched EukProt v3 for homologs after which generated phylogenetic timber. We aligned the homologs with MAFFT and MUSCLE after which generated phylogenetic timber with IQtree, FastTree, and RaxML (see Supplies and strategies). We thought-about our outcomes to be sturdy in the event that they had been concordant throughout the vast majority of 6 timber generated per gene.

To start our sequence searches for eukaryotic CD-NTases, we used the Pfam area PF03281, representing the primary catalytic area of cGAS, as a place to begin. As consultant bacterial CD-NTases, we used 6,132 bacterial sequences, representing a large swath of CD-NTase variety [21]. Following our iterative HMM searches, we recovered 313 sequences from 109 eukaryotes, of which 34 had been metazoans (S30, S31 and S32 Information and Fig 1B). Inside the phylogenetic timber, most eukaryotic sequences clustered into certainly one of 2 distinct superfamilies: the cGLR superfamily (outlined by clade and containing a Mab21 PFAM area: PF03281) or the OAS superfamily (OAS1-C: PF10421) (Fig 2A). Bacterial CD-NTases sometimes had sequences matching the HMM for the Second Messenger Oligonucleotide or Dinucleotide Synthetase area (SMODS: PF18144).

Fig 2. Impartial HGT occasions gave rise to a number of CD-NTase superfamilies.

(A) Most probability phylogenetic tree generated by IQtree of CD-NTases spanning eukaryotic and bacterial variety. The cGLR superfamily (pink, high left) is essentially an animal-specific innovation, with many paralogs together with cGAS. In distinction, most different eukaryotic lineages encode CD-NTases from the OAS superfamily (multicolor, high proper). The comparatively small eSMODS superfamily (pink, backside left) probably arose from a current HGT between clade D micro organism and eukaryotes. Bacterial CD-NTase sequences proven in grey. Eukaryotic sequences are coloured based on eukaryotic supergroup as in Fig 1B. Tree is arbitrarily rooted on a department separating bacterial clades A, B, G, and H from the remainder of the bacterial CD-NTases. (B) Venn diagrams displaying the variety of species the place we detected at the least 1 STING, cGLR, and/or OAS homolog, both inside Metazoa (left) or in non-metazoan eukaryotes (proper). (C) Magnification of the CD-NTase phylogenetic tree in (A), displaying the area the place the OAS superfamily branches inside clade C bacterial CD-NTases (grey branches). (D) Magnification displaying clade D CD-NTases (grey branches), which have been horizontally transferred into eukaryotes a number of instances, giving rise to each the cGLR and the eSMODS superfamilies. Ultrafast bootstraps decided by IQtree proven at key nodes. See S4 Fig for full CD-NTase phylogenetic tree. Underlying Newick file is included in S2 File. Extra info on which species are CD-NTases of a given homolog (Fig 2B) may be present in S1 File. CD-NTase, cGAS-DncV-like nucleotidyltransferase; cGAS, cyclic GMP-AMP synthase; HGT, horizontal gene switch; STING, Stimulator of Interferon Genes.

https://doi.org/10.1371/journal.pbio.3002436.g002

The cGLR superfamily consists nearly completely of metazoan sequences, with just a few homologs from Amoebozoa, choanoflagellates, and different eukaryotes (Fig 2A). Certainly, the vast majority of animal CD-NTases (cGAS, Mid51, Mab21, Mab21L1/2/3/4, Mb21d2, ITPRI) are paralogs inside the cGLR superfamily, which arose from repeated animal-specific duplications [49] (S4 Fig). In distinction, in contrast to the animal-dominated cGLR superfamily, the OAS superfamily spans a broad group of eukaryotic taxa, with OAS-like homologs current in 8/12 eukaryotic supergroups. This distribution makes OAS proteins the commonest CD-NTases discovered throughout eukaryotes and implies that they arose very early in eukaryotic historical past, probably earlier than the LECA.

Given the connections between cGAS and STING in each animals and a few micro organism [3,21,50], we requested whether or not species that encode STING even have cGLR and/or OAS proteins. As a result of the cGLR superfamily is essentially animal particular, we carried out this evaluation individually in both Metazoa or with all non-metazoan eukaryotes (Fig 2B). In animal species the place we discovered a STING homolog, we additionally sometimes discovered a cGLR superfamily sequence (32/34), and particularly a cGAS homolog in (26/34) species (Fig 2B), according to the consensus that these proteins are functionally linked. We additionally noticed 19 metazoan species that had a cGLR-like sequence with no detectable STING homolog. Nearly half of those species (10/19) had been arthropods, aligning with prior findings of STING sparseness amongst arthropods [50]. We did discover STING homologs in 8/19 arthropod species in EukProt v3, together with the beforehand recognized STINGs of Drosophila melanogaster, Apis mellifera, and Tribolium castaneum [50,51]. Exterior of animals, we discovered that species with a STING homolog sometimes didn’t have a detectable CD-NTase protein from both superfamily (22/34). Whereas it stays attainable that these STING proteins perform along with a to-be-discovered CD-NTase that was absent from our dataset, we subsequently hypothesize that many eukaryotes exterior of metazoans and their shut kinfolk [52] use STING and CD-NTase homologs independently of one another.

What was the evolutionary origin of eukaryotic CD-NTases? Apparently, the cGLR and OAS superfamilies are solely distantly associated to 1 one other. Every lies nested inside a special, beforehand outlined, bacterial CD-NTase clade (Fig 2C and 2D). The OAS superfamily falls inside bacterial Clade C (with the closest associated bacterial CD-NTases being these of subclade C02-C03, Fig 2C), whereas the metazoan cGLR superfamily lies inside bacterial Clade D (subclade D12) (Fig 2D). We notice that on this tree (Fig 2D), Clade D doesn’t kind a single coherent clade, as was additionally true within the phylogeny that initially outlined the bacterial CD-NTase clades [11].

We additionally noticed a lot of eukaryotic sequences scattered throughout totally different bacterial CD-NTase clades (Fig 2A, coloured branches inside grey clades). Whereas a few of these might mirror further HGT occasions, others probably come from technical artifacts resembling bacterial contamination of eukaryotic sequences. To reduce such false optimistic HGT calls, we took a conservative strategy in our analyses, contemplating potential micro organism–eukaryote HGT occasions to be reliable provided that: (1) eukaryotic and bacterial sequences branched adjoining to 1 one other with robust assist (bootstrap values >70); (2) the eukaryotic sequences fashioned a definite subclade, represented by at the least 2 species from the identical eukaryotic supergroup; (3) the eukaryotic sequences had been produced by at the least 2 totally different research; and (4) the place of the horizontally transferred sequences was sturdy throughout all alignment and phylogenetic reconstruction strategies used (S3A Fig). For species represented solely by transcriptomes, these standards should still have issue distinguishing eukaryote–micro organism HGT from sure particular situations such because the long-term presence of devoted, eukaryote-associated, bacterial symbionts. Nonetheless, as a result of these standards enable us to deal with comparatively outdated HGT occasions, they offer us larger confidence these occasions are prone to be actual.

The cGLR superfamily handed all 4 of the HGT thresholds, as did one other eukaryotic clade of CD-NTases that had been all beforehand undescribed. We identify this clade the eukaryotic SMODS (eSMODS) superfamily, as a result of the highest scoring area from hmmscan for every sequence on this superfamily was the SMODS area (PF18144), which is often discovered solely in bacterial CD-NTases (S25 File). This sequence similarity means that eSMODS arose following a current HGT from micro organism and/or that these CD-NTases have diverged from their bacterial predecessors lower than the eukaryotic OAS and cGLR households have. Moreover, the entire eSMODS sequences had been predicted to have a Nucleotidyltransferase area (PF01909), and (8/12) had a Polymerase Beta area (PF18765), that are options shared with many bacterial CD-NTases in Clades D, E, and F (S25 File). The eSMODS superfamily is made up of sequences from Amoebozoa, choanoflagellates, Ancryomonadida, and 1 animal (the sponge Oscarella pearsei), which clustered collectively robustly and with excessive assist (ultrafast bootstrap worth of 99) inside bacterial Clade D (e.g., subclade D04, CD-NTase 22 from Myxococcus xanthus) (S4 Fig). The eSMODS placement on the tree was sturdy to all alignment and phylogenetic algorithms used (S3A Fig), suggesting that eSMODS signify a further, impartial acquisition of CD-NTases from micro organism.

CD-NTases from bacterial Clade C and Clade D are the one CD-NTases to supply cyclic trinucleotides, producing cyclic tri-Adenylate and cAAG, respectively [11,14,53,54]. Apparently, OAS produces linear adenylates, which is one step away from the cAAA product made by beforehand characterised Class C CD-NTases, and equally cGAMP (made by cGAS) is one adenylate away from the Clade D product cAAG. As of this writing, the Clade D CD-NTases closest to the eSMODS and cGLR superfamilies (D04 and D12, respectively) haven’t been properly characterised. Due to this fact, we argue that these CD-NTases ought to be a spotlight of future research, as they could trace on the evolutionary stepping stones that enable eukaryotes to accumulate bacterial immune proteins.

We subsequent turned to research STING proteins. In animals, STING is a essential cyclic dinucleotide sensor, vital throughout viral, bacterial, and parasitic infections (reviewed right here [55]). Structurally, most metazoan STINGs include an N-terminal transmembrane area (TM), made from 4 alpha helices fused to a C-terminal STING area [56]. Canonical animal STINGs present distant homology with STING effectors from the bacterial CBASS, with main variations in protein construction and pathway perform between these animal and bacterial defenses. For instance, in micro organism, the vast majority of STING proteins are fusions of a STING area to a TIR (Toll/interleukin-1 receptor) area (Fig 3A). Bacterial STING proteins acknowledge cyclic di-GMP and oligomerize upon activation, which promotes TIR enzymatic exercise [3,57,58]. Some micro organism, resembling Flavobacteriaceae, encode proteins that fuse a STING area to a transmembrane area, though it’s unclear how these bacterial TM-STINGs perform [3]. Different micro organism have STING area fusions with deoxyribohydrolase, α/β- hydrolase, or trypsin peptidase domains [19]. Along with eukaryotic TM-STINGs, a number of eukaryotes such because the oyster Crassostrea gigas have TIR-STING fusion proteins, though the precise position of their TIR area stays unclear [3,51,59].

Fig 3. Numerous eukaryotic STING proteins bridge the hole between metazoans and micro organism.

(A) Graphical depiction of widespread area architectures of STING proteins. (B) Most probability unrooted phylogenetic tree of STING domains from Metazoa and micro organism, that are separated by 1 lengthy department. Black dot (•) signifies proteins which have been beforehand experimentally characterised. Bacterial sequences are in grey and animal sequences are in pink. (C) Most probability unrooted phylogenetic tree of hits from iterative HMM searches for various eukaryotic STING domains. The STING domains from blSTINGs from various eukaryotes break up the lengthy department between bacterial and animal STINGs. Buildings of the indicated STING proteins are proven above, with these predicted by AlphaFold indicated by an asterisk. Homologs with X-ray crystal constructions are from [3,87]. Two area architectures exist in micro organism and eukaryotes (STING linked to a TIR area and STING linked to a transmembrane area), every of which have developed convergently by means of area shuffling. Ultrafast bootstraps decided by IQtree proven at key nodes. Eukaryotic sequences are coloured based on the eukaryotic group as in Fig 1B. See S5 Fig for full STING phylogenetic tree. Underlying Newick recordsdata are included in Supporting info (S3 File and S4 File). AlphaFold predicted constructions are additionally included within the Supporting info (S6 File, S7 File, and S8 File). blSTING, bacteria-like STING; HMM, hidden Markov mannequin; STING, Stimulator of Interferon Genes.

https://doi.org/10.1371/journal.pbio.3002436.g003

Given these main variations in area architectures, ligands, and downstream immune responses, how have animals and micro organism developed their STING-based defenses, and what are the relationships between them? Previous to this work, the phylogenetic relationship between animal and bacterial STINGs has been troublesome to characterize with excessive assist [19]. Certainly, once we made a tree of beforehand recognized animal and bacterial STING domains, we discovered that the metazoan sequences had been separated from the bacterial sequences by one very lengthy department, alongside which many adjustments had occurred (Fig 3B).

To enhance the phylogeny by means of the inclusion of a larger variety of eukaryotic STING sequences, we started by fastidiously figuring out the area of STING that was homologous between bacterial and animal STINGs, as we anticipated this area to be greatest conserved throughout various eukaryotes. Though Pfam area PF15009 (TMEM173) is usually used to outline animal STING domains, this HMM features a portion of STING’s transmembrane area which isn’t shared by bacterial STINGs. Due to this fact, we in contrast the crystal constructions of HsSTING (6NT5), Flavobacteriaceae sp. STING (6WT4), and Crassostrea gigas STING (6WT7) to outline a core “STING” area. We used the area equivalent to residues 145–353 of 6NT5 as an preliminary HMM seed alignment of 15 STING sequences from PF1500915 (“Reviewed” sequences on InterPro). Our searches yielded 146 eukaryotic sequences from 64 species, which included STING homologs from 34 metazoans (S31 File and Fig 1). Utilizing most probability phylogenetic reconstruction on the STING area alone, we recognized STING-like sequences from 26 various microeukaryotes whose STING domains clustered in between bacterial and metazoan sequences, breaking apart the lengthy department. We identify these sequences the bacteria-like STINGs (blSTINGs) as a result of they had been the one eukaryotic group of STINGs with a bacteria-like Prok_STING area (PF20300) and due to the quick department size (0.86 versus 1.8) separating them from bacterial STINGs on the tree (Fig 3C). Whereas a earlier research reported STING domains in 2 eukaryotic species (1 in Stramenopiles and 1 in Haptista) [19], we had been capable of develop this set to further species and in addition recuperate blSTINGs from Amoebozoa, Rhizaria, and choanoflagellates. This variety allowed us to position the sequences on the tree with excessive confidence (bootstrap worth >70), recovering a considerably totally different tree than earlier work [19]. As for CD-NTases, the tree topology we recovered was sturdy throughout a number of totally different alignment and phylogenetic tree building algorithms (S3A Fig).

Given the similarities between the STING domains of the blSTINGs and bacterial STINGs, we subsequent requested whether or not the area architectures of those proteins had been comparable utilizing Hmmscan and AlphaFold. Nearly all of the brand new eukaryotic blSTINGs had been predicted to have 4 N-terminal alpha helices (Fig 3C and S5 File and S6 File), just like human STING. Whereas bacterial TM-STINGs had been superficially just like N-terminal transmembrane domains, these proteins had been predicted to have solely 2 alpha helices and in all phylogenetic timber the STING domains from bacterial TM-STINGs had been extra just like different bacterial STINGs than to eukaryotic homologs (S3A Fig). These outcomes counsel that eukaryotes and micro organism independently converged on a standard TM-STING area structure by means of area shuffling.

Apparently, the same sample of convergent area shuffling seems to have occurred a second time with the TIR-STING proteins. Some eukaryotes, such because the oyster C. gigas, have a TIR-STING fusion protein [3,51,59]. The STING area of those TIR-STINGs clustered intently to different metazoan STINGs, suggesting an animal origin (Fig 3B). We additionally investigated the likelihood that C. gigas acquired the TIR-domain of its TIR-STING protein through HGT from micro organism; nonetheless, this evaluation additionally instructed an animal origin for the TIR area (S7 Fig), because the C. gigas TIR area clustered with different metazoan TIR domains resembling Homo sapiens TICAM1 and a couple of (ultrafast bootstrap worth of 75). Eukaryotic TIR-STINGs are additionally uncommon, additional supporting the speculation that this protein resulted from current convergence, the place animals independently fused STING and TIR domains to make a protein resembling bacterial TIR-STINGs, according to earlier reviews [19]. General, the phylogenetic tree we constructed (Fig 3C) suggests that there’s domain-level homology between bacterial and eukaryotic STINGs, however attributable to sparseness and lack of an acceptable outgroup, this tree doesn’t definitively clarify the eukaryotic origin of the STING area. Nonetheless, the information does clearly assist a mannequin by which convergent area shuffling in eukaryotes and micro organism generated comparable TM-STING and TIR-STING proteins independently. Apparently, the non-metazoan, blSTINGs (Fig 3C) which can be discovered within the Stramenopiles, Haptista, Rhizaria, Choanoflagellates, and Amoebozoa have a TM-STING area structure just like animal STINGs however a STING area extra just like bacterial STINGs.

Viperins are innate immune proteins that prohibit the replication of a various array of viruses by conversion of nucleotides into 3′-deoxy-3′,4′didehdro- (ddh) nucleotides [4,15–17]. Incorporation of those ddh nucleotides right into a nascent RNA molecule results in chain termination, blocking RNA synthesis and inhibiting viral replication [15,18]. Whereas metazoan viperin particularly catalyzes CTP to ddhCTP [15], homologs from archaea and micro organism can generate ddhCTP, ddhGTP, and ddhUTP [4,60]. Earlier structural and phylogenetic evaluation confirmed that eukaryotic viperins are extremely conserved at each the sequence and structural stage and that, phylogenetically, animal and fungal viperins kind a definite monophyletic clade in comparison with bacterial viperins [4,16,60].

As viperin proteins include a single Radical SAM protein area, we iteratively searched EukProt starting with area PF04055 (Radical_SAM). The 194 viperin-like proteins we recovered got here from 158 species spanning the total vary of eukaryotic variety, together with organisms from the entire main eukaryotic supergroups, in addition to some orphan taxa whose taxonomy stays open to debate (Fig 1, Ancyromonadida, Hemimastigophora, Malawimonadida). After we constructed phylogenetic timber from these sequences, we discovered that the big majority of the eukaryotic viperins cluster collectively in a single, monophyletic clade, separate from bacterial or archaeal viperins (Fig 4). Inside the eukaryotic viperin clade, sequences from extra intently associated eukaryotes typically clustered collectively (Fig 4, coloured blocks), as can be anticipated if viperins had been current and vertically inherited inside eukaryotes for an prolonged time frame. The huge species variety and tree topology each strongly assist the inference that viperins are a very historical immune module and have been current inside the eukaryotic lineage probably courting again to the LECA.

Fig 4. Viperin is a deeply conserved innate immune module.

Most probability phylogenetic tree generated by IQtree of viperins from eukaryotes, micro organism, and archaea. All main eukaryotic supergroups have at the least 2 species that encode a viperin homolog (coloured supergroups). Eukaryotic sequences are coloured based on eukaryotic group as in Fig 1B. Bacterial viperin sequences proven in grey and archaeal sequences in darkish grey. There are 2 clades of Chloroplastida (a gaggle inside Archaeplastida) sequences that department robustly (>80 ultrafast bootstrap worth) inside the micro organism clade. Ultrafast bootstraps decided by IQtree proven at key nodes. Tree is arbitrarily rooted between the foremost eukaryotic and bacterial clades. See S6 Fig for totally annotated viperin phylogenetic tree. Underlying Newick file is included in S8 File below Supporting info.

https://doi.org/10.1371/journal.pbio.3002436.g004

Along with this deep eukaryotic ancestry, we additionally uncovered 2 examples of micro organism–eukaryote HGT which have occurred rather more not too long ago, each in Chloroplastida, a gaggle inside Archaeplastida. The primary of those consists of a small clade of Archaeplastida (Clade A) consisting of marine algae resembling Chloroclados australicus and Nemeris dumetosa. These algal viperins cluster intently with the marine cyanobacteria Anabaena cylindrica and Plankthriodies (Figs 4 and S6). The second clade (Clade B) consists of 4 different Archaeplastida inexperienced algal species, largely Chlamydomonas spp. In a few of our timber, the Clade B viperins branched close to to eukaryotic sequences from different eukaryotic supergroups; nonetheless, the location of the neighboring eukaryotic sequences various relying on the algorithms we used; solely the Archaeplastida placement was constant (Figs 4 and S3A and S6). Taken collectively, we conclude that viperins signify a category of historical immune proteins which have probably been current in eukaryotes because the LECA. But, we additionally discover ongoing evolutionary innovation in viperins through HGT, each amongst eukaryotes and between eukaryotes and micro organism.