{"paper_id":"29d78de2-e62e-4deb-8c1c-d8698502bf67","body_text":"1 \nFunctional and phylogenomic approaches  reveal novel \ntypes of M42 peptidases with contrasted enzymatic \nproperties in Archaea \n \nEmilie Chagny1, Najwa Taib 2,3, Daphna Fenel 4, Eric Girard 4, Simonetta Gribaldo 2, Didier Flament 1*, \nBruno Franzetti4*  \n \n1 Univ. Brest, Ifremer, BEEP, F-29280 Plouzané, France \n2 Institut Pasteur, Université Paris Cité, Evolutionary Biology of the Microbial Cell Laboratory, Paris, \nFrance \n3 Institut Pasteur, Université Paris Cité, Bioinformatics and Biostatistics Hub, Paris, France \n4 Univ. Grenoble Alpes, CNRS, CEA, IBS, F-38000 Grenoble  \n*Corresponding authors: didier.flament@ifremer.fr, bruno.franzetti@ibs.fr  \n \nAbstract \n \nM42 peptidases are half-megadalton aminopeptidases characterized by a tetrahedral architecture (TET) \nubiquitous across all domains of life. Despite their widespread occurrence, their evolutionary history and \nfunctional diversity remain largely unexplored. Here we show  an unsuspected and largely untapped \nwealth of archaeal TET peptidases, exhibiting remarkable functional heterogeneity, as illustrated by the \ncharacterization of six novel enzymes. Using structural biology, phylogeny, and enzymatic studies, we \nestablish robust criteria for high -throughput identification of TET peptidases and perform the first \nsystematic study of their genomic distribution and functional diversity across the archaeal kingdom. We \npropose an 11-group classification for these enzymes and identify one group as the ancestral lineage \nthat first emerged in Archaea. By coupling taxonomic distribution patterns with functional insights , we \nhighlight the presence of multiple TET enzymes with selective activities in heterotrophic and mixotrophic \norganisms, suggesting a role for TET peptidases in the degradation of environmental peptides. Overall, \nthis work illuminates the underexplored diversity of TET enzymes, uncovering a complex evolutionary \nhistory linked to their potential biological function. \n  \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted March 2, 2025. ; https://doi.org/10.1101/2025.02.26.640355doi: bioRxiv preprint \n\n 2 \nIntroduction \n \nThe development of environmental surveys and culture -independent approaches, such as \nmetagenomics and high-throughput sequencing, have greatly enhanced our understanding of the global \nmicrobial inventory, particularly in marine ecosystems1–3. This surge in genomic data availability unveiled \nan unforeseen level of taxonomic and functional diversity among prokaryotes, leading to the discovery \nof novel metabolic pathways, insights into enzyme functions related to environmental adaptation, and \nthe identification of new biocatalysts for biotechnological applications4,5. However, functional annotation \nbased on sequence or structural similarity, even with deep learning -based bioinformatics tools, often \nfalls short of accurately classifying proteins 6 and requires direct experimental validation. While high-\nthroughput functional screening approaches have been used to identify enzymes of interest from \ngenomic data7, these methods  only allow exploration of a limited range of substrates and activation \nconditions. This is especially problematic for large complexes or extremophilic enzymes, which often \nrequire specific activation conditions. To address these challenges, we introduce an innovative hybrid \napproach integrating phylogenetic analysis with biochemical characterization, enabling the exploration \nof the functional diversity of M42 aminopeptidases, a unique type of giant self -compartmentalized \naminopeptidases forming distinctive tetrahedral structures, named TET. \n \nTET peptidases are known to sequentially degrade the N-terminal residues of peptides up to 40 \namino-acids in length 8. These enzymes belong to the  M18 and M42 families  of the  MEROPS \nclassification9. M18 members are predominantly found in eukaryotes and bacteria, while M42 peptidases \nare restricted to prokaryotes 10. Although their biological significance remains poorly understood, they \nhave been hypothesized to be involved in intracellular proteolysis  downstream of the proteasome 8,11, \nwith no clear supporting evidence to date. Archaeal TET peptidases, in particular, have been studied \nextensively. Following the discovery of the first TET complex in Haloarcula marismortui8, several high-\nresolution structures of archaeal enzymes  revealed a conserved architecture, characterized by the \nassembly of twelve subunits into a ~450 kDa hollow tetrahedral complex. Dimers, which are the building \nblock of the dodecamer, are positioned along the edges of the complex.  The faces of the tetrahedron \nare defined by three dimers forming a central opening, which is presumed to function as the substrate \nentry site, leading to a wide inner cavity. All catalytic sites, situated in the apexes of the complex, are \noriented toward this chamber11–14. The active site comprises seven catalytic residues, with five of these \nresidues coordinating two metallic ion cofactors12,15–17. \n \nDespite a high degree of structural conservation, characterization of several archaeal TETs \nrevealed significant functional disparities, along with variations in the copy number per organism . The \nsingle TET of H. marismortui was characterized as a broad -spectrum aminopeptidase, whereas four \nhomologous TETs exhibiting distinct and narrower substrate specificities were identified in Pyrococcus \nhorikoshii: PhTET1, PhTET2, PhTET3, and PhTET4 are glutamyl -, leucyl-, lysyl-, and glycyl -specific \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted March 2, 2025. ; https://doi.org/10.1101/2025.02.26.640355doi: bioRxiv preprint \n\n 3 \naminopeptidases, respectively13,18–21. Owing to this functional versatility, TET peptidases hold significant \npotential for biotechnologial applications in nutrition, health, and cosmetics. For instance, archaeal TET \npeptidases can increase the diversity of bioactive peptides in hydrolysates derived from natural biomass, \nwhich could be used in the agri-food sector to improve the nutritional quality of feeds for aquaculture22,23. \nHowever, with only a few archaeal enzymes thoroughly functionally characterized so far 8,13,15,18–21,24, \nmostly from closely related species within  the Thermococcales order, it is uncertain whether current \nknowledge fully accounts for the functional diversity and biological activities of these enzymatic \ncomplexes. \n \nIn this study, we carried out a high-throughput screening for M42 TET peptidases in 3,702 \narchaeal genomes  using structure-based identification  criteria. We uncover a previously unknown \ndiversity of TET spanning the whole tree of A rchaea. By combining phylogenetic and biochemical \nanalyses, we propose to  classify archaeal TETs in to eleven groups. Six new TETs from previously \nundescribed groups were characterized, revealing a large functional diversity of these enzymes. Finally, \nwe infer the evolutionary history of TET peptidases  and discuss new insights into their potential \nbiological roles. \nResults \nStructure based analysis identifies a large diversity of TET peptidases in Archaea \nDespite strong structural homology, M42 peptidase primary sequences exhibit high divergence14. This \nvariation, coupled with frequent misannotations as cellulases or endoglucanases 24,25, make s it \nchallenging to identify M42 peptidases in genomic or proteomic databases  by sequence homology \nsearches. Using the structural elements involved in the formation of their unique tetrahedral architecture \nas key determinants, we identified several residues to better delineate M42 aminopeptidases (Fig. 1). \nFirst, the presence of the seven conserved catalytic residues (His62, Asp64, Asp173, Glu205, Glu206, \nAsp/Glu228, and His307 in PhTET1, PDB code 2WYR) coordinating two metallic cofactors is essential \nfor both the catalytic activity and structural integrity of the complex15–17. These residues being shared by \nall peptidases of the MEROPS MH clan (i.e., M18, M20, M28, and M42 families), additional criteria are \nneeded for accurate M42 peptidase segregation 26. Sequence and structure comparison of MH clan \nenzymes highlighted the unique presence of five glycine residues (Gly44, 77, 85, 86, 211 in PhTET1) in \nM42 peptidases that confer the required flexibility for proper protein folding in the periphery of the active \nsite. Taken together, these two criteria provide a robust method for screening M42 aminopeptidases in \ngenomic and proteomic databases. To ensure clarity and consistency  with previous studies, the term \n'TET peptidases' will be used exclusively for the remainder of this article. \n \nAn HMM profile was built using 210 archaeal TET sequences gathered from the MEROPS and \nNCBI nr databases. This profile was used to conduct an exhaustive homology search of TET peptidases \nagainst a large database  containing 3,702 genomes of Archaea and covering all currently available \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted March 2, 2025. ; https://doi.org/10.1101/2025.02.26.640355doi: bioRxiv preprint \n\n 4 \ndiversity. All positive hits were retrieved and filtered based on the presence of the conserved catalytic \nsite and glycine residues mentioned above, resulting in  1,826 archaeal TET homologues \n(Supplementary Table 1). TET peptidases were found in 39. 0% of archaeal genomes (1, 443) with \nvarying numbers of TET copies per species, ranging from one to four. While TET homologues are widely \npresent in Euryarchaeota and Asgard, their distribution is uneven to sporadic in the TACK and the \nDPANN superphyla , respectively (Supplementary Table 1). Intriguingly, some TET peptidase \nsequences from Asgard display a specific pattern consisting in an  insertion of approximately 20 \nresidues, which has never been detected in any TET peptidase described to date . T his peculiar \ninsertion, located just after the dimerization interface21, was found in 74 Asgard species and exhibits a \nstrong charge contrast. A similar but shorter insertion (11-12 residues) was also observed in 23 species \nof Thermoplasmata (Supplementary Fig. 1).  \n \nWe investigated the presence of possible functional analogs in genomes lacking TET peptidase. \nAs previous studies suggested complementarity between M42 and M18 or TRI peptidases 11,25,27, we \nused the PFAM domains PF02127 and PF14684 to search for M18 and TRI homologues in our local \ndatabase of archaeal genomes (Supplementary Table 1). Our results challenge these hypotheses; \nM18 and TRI homologues were only sparsely detected, primarily in species possessing M42 peptidases. \nFurthermore, several lineages ( e.g. Theionarchaea, Pontarchaeia, Thalassoarchaeia, Methanocellia, \nThaumarchaeota) lack all three peptidase families. This points to a more complex relationship and \nsuggests the existence of other functional analogs yet to be identified. \n \nTo understand how archaeal TET peptidases are related to each other, we inferred a maximum \nlikelihood tree using the 1,826 archaeal M42 sequences (Fig. 2). Members of the TET1, TET2, TET3, \nand TET4 groups were already identified in Thermococcales species prior to this study, with several \ncharacterized representatives from Pyrococcus horikoshii , Pyrococcus furiosus  and Thermococcus \nonnurineus13,15,18–21,28,29. In the TET phylogeny, t hese groups form  four distinct and supported \nmonophyletic clades (with UFB values of 100%). While TET2, TET3, and TET4 contain exclusively \nsequences from Thermococcales, TET1 also includes three sequences from Geoglobus, likely arising \nfrom a horizontal gene transfer (HGT) from Thermococcales to Archaeoglobales (Fig. 2). \n \nArchaeal TET peptidases exhibit contrasting substrate specificities  \nPrior studies on the four TET of P. horikoshii reavealed distinct substrate specificities, already unveiling \nthe functional versatility of these enzymes : PhTET2 functions as a broad -spectrum leucyl -\naminopeptidase18, while PhTET1, PhTET3, and PhTET4 specifically  target acidic, basic, and glycine  \nresidues, respectively13,19,20. However, a striking observation from the present analysis is that previous \ncharacterization of these few archaeal TET peptidases remains marginal with regard to the real  \ntaxonomic distribution and overall diversity  of TET peptidases in archaeal genomes 8,13,15,18–20. To fully \nexplore their functional diversity in Archaea, we selected thirteen sequences, phylogenetically distant \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted March 2, 2025. ; https://doi.org/10.1101/2025.02.26.640355doi: bioRxiv preprint \n\n 5 \nand from nine different species (1 TACK, 1 DPANN, 3 Asgardarchaeota, 2 Methanotecta, and 2 \nMethanomada) for recombinant protein expression and purification (Table 1, sequences provided in \nSupplementary Table 2). Notably, three Asgard sequences featuring the newly identified insertion were \nselected.  \n \n \nOf the thirteen initially targeted proteins, six were successfully produced and purified to near \nhomogeneity from  E. coli  extracts (i.e. HoTETb, PsTETa, PsTETc, ThTET, TaTET, and MtTET) . \nTogether with previously characterized TET peptidases, these new enzymes capture  the core \nphylogenetic and taxonomic diversity of archaeal TETs. During the final gel filtration chromatography \nstep, all proteins eluted as well-separated high molecular mass complexes corresponding to particles of \na molecular mass of c. 450 kDa , indicating the formation of homo -dodecameric complexes \n(Supplementary Fig. 2). These findings were further validated by negative -stain electron microscopy \nobservations, which revealed homogeneous populations of tetrahedral particles for most enzymes \n(Supplementary Fig. 2a). Lower purity of PsTETc and TaTET samples precluding satisfactory negative-\nstaining imaging, structure predictions were generated using AlphaFold3 (Supplementary Fig. 2b). The \nresulting models displayed the expected hollow tetrahedral edifices with high confidence scores (ipTM \n0.88 and 0.91), further supporting PsTETc and TaTET ability to form high molecular weight assemblies.  \n \n To investigate whether the diversity observed through phylogenetic analysis reflects functional \ndiversity, cleavage specificities were studied using chromogenic [para-nitroaniline (pNA) conjugated] \nand fluorogenic [7 -amino-4-methylcoumarin (AMC) conjugated] aminoacyl substrates  (Fig. 3 ). The \nobserved activity spectra are heterogenous and can be divided into two main clusters. PsTETa, ThTET, \nMtTET, and T aTET exhibit broad-spectrum activities and can be classified as generalist enzymes.  \nThese enzymes were found to preferentially cleave hydrophobic residues, with PsTET a and ThTET \ndisplaying broader specificities. PsTETa, ThTET, MtTET, and TaTET optimal amidolytic activities were \nobserved with Ile-pNA, Leu-pNA, Met-pNA, and Met-pNA, respectively (Fig. 3). Similarities with the \ncleavage profile of PhTET2 18 can be outlined, but PsTETa, MtTET, and TaTET represent the first  \ndescription of methionyl and isoleucyl aminopeptidases in the TET family. Conversely, PsTETc and \nHoTETb exhibit more selective activities and can be classified as specialized enzymes.  Analogous to \nthe previously described PhTET1 peptidase, they specifically target acidic amino acids19. Interestingly, \nPsTETc maximum activity was measured on Glu-pNA, whereas no hydrolysis could be detected on Asp-\npNA despite the similarity of these substrates (Fig. 3). The same substrate specificity has already been \nreported for the MHJ_0125 glutamyl-aminopeptidase of Mycoplasma hyopneumiae 30. In addition to \nthese varied substrate specificities, the characterized TET peptidases exhibit distinct activation profiles, \nwith variations in optimal temperature, pH, and metal cofactor requirements (Supplementary Fig. 3). \n \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted March 2, 2025. ; https://doi.org/10.1101/2025.02.26.640355doi: bioRxiv preprint \n\n 6 \nArchaeal TET peptidases can be delineated into 11 groups \nConsidering that the pre-existing TET1 to TET4 groups cover only a small fraction of TET diversity, we \nused the distinct biochemical properties of the characterized enzymes and their taxonomic distribution \nto delineate new groups (Fig. 3). Specifically, contrasted substrate specificities led us to separate the \nTET6 and TET9 groups. Conversely, the broad TET11 group was maintained as a single group\tdue to \nthe similarities between the substrate specificities of the enzymes of Methanocaldococcus jannaschii \n(Atalah et al.,  in preparation)  and Methanoculleus thermophilus, along with a robust phylogenetic \nsupport. Similarly, TaTET has been identified as a generalist enzyme predominantly targeting \nhydrophobic residues, which is consistent with the substrate specificity of the previously characterized \nAPDkam589 peptidase belonging to the same group24. Consequently, seven new groups were \ndescribed across the tree, which we re named TET5 to TET11 in accordance with the existing \nnomenclature (Supplementary Fig. 4), and each characterized enzyme was renamed according to its \nclassification within the eleven defined groups. Notably, all sequences featuring the novel insertion \ndescribed above were found in the TET7 group. Collectively, these eleven groups account for 1, 500 \nsequences, the remaining 326 sequences were not affiliated to any group due to poorly supported \nbranching or lack of characterized representatives. \n \nInterestingly, TET11 emerges as the most prevalent group, distributed across the entire tree of \nArchaea, and is consistently present in methanogenic species, suggesting that this group may have \nbeen the first to appear in Archaea  (Fig. 4 ). In contrast, TET1, TET2, TET3, and TET4 groups —\npreviously the only known TET groups—appear to be restricted to Thermocci species (to the exception \nof a few TET1 sequences found in Archaeoglobales). TET2 and TET3 are sister clades, indicating that \nthey arose from a gene duplication event. Interestingly, prior studies showed the in vitro and in vivo \nformation of PhTET2 -PhTET3 heterocomplexes in  P. horikoshii 16,31. TET5 and TET6 members are \nexclusively detected in Halobacteria . TET7 and TET8 groups span two superphyla and are found in \nspecies from the Asgard and TACK groups. These sister clades also possess unique TET groups, with \nTET9 and TET10 being found exclusively in Heimdallarchaeia and Crenarchaeota, respectively (Fig. 4).  \n \nFinally, to investigate the origin of the TET peptidases, we searched for homologues in 401 \nrepresentative bacterial genomes.  We retrieved 187 bacterial TET homologues from 30% of the \nanalyzed genomes  displaying a patchy distribution across diverse phyla such as Thermotogae, \nProteobacteria, Firmicutes, Chloroflexi, Deinococcota, Atribacteria, Bipolaricaulota, and Verrumicrobia \n(Supplementary Table 1). We inferred a maximum likelihood tree ( Supplementary Fig. 5), which \nrevealed a complex evolutionary history, shaped by multiple HGT intra and inter domains and several \nduplications. Two distinct groups can be delineated: the first group primarily contains archaeal \nsequences, spanning the full diversity of Archaea,  with representatives from the TET1, TET4, and \nTET11 groups. Interestingly, sequences belonging to Bacteria, mainly Elusimicrobia, Thermotogae, \nFirmicutes, and Proteobacteria branch within this group indicating several independent HGT between \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted March 2, 2025. ; https://doi.org/10.1101/2025.02.26.640355doi: bioRxiv preprint \n\n 7 \narchaea and these bacteria. The second group consists of a mixture of archaeal, mainly from the TACK \nand Asgard superphyla, and bacterial sequences, encompassing the remaining TET groups. \nDiscussion \nIn this study, we used  structure-based identification criteria for high -throughput screening of M42 \npeptidases to demonstrate a vast prevalence and diversity of these enzymes in archaea. Our approach \nnotably revealed a ~20-residue insertion, located close to the dimerization domain essential for TET \nparticle assembly, in some Asgard TET peptidases. This insertion does not hinder oligomerization, since \nboth PsTET7 and ThTET7 still form typical hollow tetrahedral particles. AlphaFold3 predictions suggest \nthat these insertions may form protruding two-stranded b-sheets, potentially obstructing the substrate \nentry pores of the tetrahedral particle 21 (Supplementary Fig. 6). Given that PsTET 7 and ThTET 7 \ndemonstrate a broader substrate specificity relative to PhTET2, MtTET11, and TaTET10, this insertion \nmay play a role in substrate recognition. Alternatively, this insertion might facilitate interactions with \npartner proteins. To fully elucidate the role of this novel insertion, further functional and structural studies \non representative members of this group should be conducted.  \n \nPhylogenetic analysis and functional characterization of archaeal TET peptidases allowed us to \npropose a classification into eleven groups, seven of which ha d never been investigated before. An \nimportant finding from this study is the discovery of the TET11 group. These enzymes are  the most \nprevalent and have nevertheless been completely overlooked until now. Indeed, while most of the TET \ngroups are restricted to few tax a (e.g., TET1-4 in Thermococci, TET5 -6 in Halobacteria), the TET11 \ngroup is widespread in Archaea and is found in all superphyla, suggesting that it was present in the last \narchaeal common ancestor. It is also notable that species with a single TET enzyme tend to possess a \nmember of the TET11 group. According to the characterization of MtTET 11, TET11 members would \ndisplay broad-spectrum activities. The ancestral origin of TET11, coupled with its enzymatic activity, \nsuggest that this group was the first to appear in Archaea, followed by several duplications and \nhorizontal transfers giving rise to multiple groups with different activities. Genetic studies on members \nof the proposed ancestral group TET11, such as the single TET peptidase of the genetically tractable \nspecies Methanocaldococcus jannaschii32, should be prioritized to establish the basic physiological role \nof the TET enzymes.  \n \nThe study of TET co -occurrence within archaeal genomes showed that narrow substrate \nspecificity occurs only in species harboring multiple TETs. This is the case for  P. horikoshii, which \npossesses four TET peptidases  of TET1, TET2, TET3, and TET4 groups. The four enzymes exhibit \ncomplementary activity spectra , suggesting that they function in concert to better achieve complete \npeptide hydrolysis13,18–20. Synergic specific activities of multiple TET peptidases have also been reported \nfor the bacteria Geobacillus stearothermophilus33 and Symbiobacterium thermophilum34. Similarly, we \nidentified specialized enzymes belonging to the TET8 and TET9 groups in the genomes of Ca. \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted March 2, 2025. ; https://doi.org/10.1101/2025.02.26.640355doi: bioRxiv preprint \n\n 8 \nHodarchaeales archaeon LC_ 3 and Ca. P. syntrophicum  containing two and three putative TET \npeptidase genes, respectively . Multiplicity is also observed in other types of archaea such as \nHalobacteriales, which typically harbor both a TET5 and a TET6. HmTET6, the only characterized \nenzyme of the TET6 group, displays broad-spectrum activity. Although no enzyme from the TET5 group \nhas been characterized to date, it could be  hypothesized that this group exhibits narrower activity \nprofiles. Accordingly, since TET-other group members are found in species possessing either a single \nTET or both a TET-other and an additional specialized TET, it can be hypothesized that enzymes display \nbroad-spectrum activities.  \n \nPhylogenetic analysis indicates that TET multiplicity arose independently multiple times. This \nphenomenon is not exclusively attributable to HGTs, as evidenced by the emergence of the TET1 and \nTET4 groups by duplication within archaea. Contrasted substrate specificity and multiplicity of TET \nenzymes within archaeal proteomes does not appear to stem from environmental adaptation , as no \ncorrelation was identified between TET distribution and specific biotopes. For example, although \nArchaeoglobales, Thermococc ales, Methanococcales, and Desulfurococcales all share the same \necological niche as  primary colonizers of deep -sea hydrothermal vents 35–37, these organisms exhibit \nmarkedly different TET distribution patterns. Conversely, the number and degree of specificity of TET \npeptidases present in an organism may correlate with its metabolic capabilities. Indeed, multiple TETs \nare found in heterotrophic and mixotrophic Hadarchaea38, Thermococcales 39–41, Halobacteriales 42, \nCrenarchaeota43 (Vulcanisaeta genus), Heimdallarchaeota 44–46, Korarchaeota 47,48, and \nBathyarchaeota49–51 species. This suggests that TETs may play a metabolic role in the degradation of \nenvironmental peptides used as carbon sources, enabling efficient organic matter utilization. In contrast, \nTET peptidases are typically found in single copy in autotrophic species such as methanogens, which \ndo not depend on the degradation of exogenous peptides, hinting at an alternative physiological role for \nthese enzymes. As initially proposed, TETs may participate in protein homeostasis and amino acid \nrecycling by processi ng peptides downstream of the proteasome and other related proteolytic \ncomplexes8,11. \n \nTo conclude , the hybrid approach adopted in this study —integrating structural biology, \nphylogeny, and biochemistry—revealed an unsuspected diversity of TET peptidases. This strategy shed \nlight on a complex evolutionary history, uncovering an ancient subgroup of archaeal enzymes that had \nso far gone unnoticed. Moreover, we could classify archaeal TET peptidases into eleven distinct groups. \nCharacterization of  representatives from these groups revealed contrasting biochemical properties, \nunderscoring the value of this approach  to facilitate the discovery of proteins with discriminating \ncharacteristics within the same enzymatic family. \n \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted March 2, 2025. ; https://doi.org/10.1101/2025.02.26.640355doi: bioRxiv preprint \n\n 9 \nMaterial and methods \nM42 peptidases identification in bacterial and archaeal genomes \nTo study the taxonomic distribution and the evolution of the M42 peptidase family in Archaea, we \nassembled a large  database containing 3,702 archaeal genomes  and 401 bacterial genomes  \nrepresentatives of all major phyla available in public databases as of January 2022 (Supplementary \nTable 1). \n \nFor homology searches, we built a specific HMM profile for the M42 archaeal peptidase family. \nFor this, we used the MEROPS database (v12.0)9 and retrieved all archaeal M42 peptidase sequences \nlonger than 260 amino acids  (195 sequences). We estimated TET4 homologues to be under -\nrepresented in this dataset so in parallel, the LD[AE][EL]EKKED pattern, canonical for the TET4 group, \nwas used to search the National Center for Biotechnology Information (NCBI) nr database restricted to \nArchaea using PHI-BLAST (default parameters)52. We retrieved 43 hits and used the T-Coffee trim tool \n(v11.0.8)53 to identify the 15 more divergent sequences, which were added to the initial set. Finally, the \n210 resulting sequences were aligned with T-Coffee using default parameters, and the alignment was \nused to build an HMM profile using the HMMBUILD tool from the HMMER suite (v3.3.2)54. \n \nThis profile was used to carry out homology-based searches against our local Archaea database \nusing HMMSEARCH. All hits were retrieved, aligned using MAFFT (v7.481, with the option -auto)55 and \nfiltered upon the presence of the conserved motifs characterizing M42 peptidases (i.e., residues Gly44, \nHis62, Asp64, Gly77, Gly85, Gly86, Asp173, Gl205, Glu206, Gly211, Asp/Glu228 and His307 according \nto PhTET1 numbering, PDB code 2WYR). This resulted in 1,826 archaeal M42 peptidase homologues \n(Supplementary Table 1). These sequences were aligned using MAFFT (with the option -linsi), and the \nresulting alignment was trimmed using trimAl (v1.5.0, with the option -gappy-out)56. Finally, a maximum \nlikelihood phylogeny was inferred using IQ -TREE (v2.0.6)57 with the model LG+F+R10  selected by \nModelFinder according to BIC criteria58.  \n \nTo investigate the origin and evolution of archaeal M42 peptidase, we extracted a reduced and \ntaxonomically balanced database covering both archaea and bacteria from our local database (593 \nArchaea and 401 Bacteria, see Supplementary Table 1). Homology searches, alignment, and filtering \nsteps were conducted as described above, yielding in 339 archaeal and 187 bacterial M42 peptidase \nhomologues (Supplementary Table 1). The 526 sequences were aligned using MAFFT (with the option \n-linsi), and trimmed using trimAl (v1.5.0, with the option  -gappy-out)56. A maximum likelihood \nphylogenetic tree was inferred using IQ-TREE and the model LG+R10  selected by ModelFinder \naccording to BIC criteria (Supplementary Fig. 5). All phylogenies were annotated using IToL59. \n \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted March 2, 2025. ; https://doi.org/10.1101/2025.02.26.640355doi: bioRxiv preprint \n\n 10 \nFinally, we used HMMSEARCH (with the option -cut_nc) and the PFAM domains PF02127 and \nPF14684 to search for peptidases M18, and TRI, respectively, in our local database of Archaea \n(Supplementary Table 1). \n \nBacterial strains and general information  \nEscherichia coli DH5a and Rosetta 2(DE3)pLysS chemically competent cells were used for cloning and \nrecombinant expression, respectively. Cells were grown in lysogeny broth (LB) media in a rotary shaker \nat 37°C (or 20°C when specified), 140 rpm. When used, final concentrations of kanamycin and \nchloramphenicol were 30 μg/mL and 34 μg/mL, respectively. \n \nFor SDS-PAGE analysis, protein samples were mixed with loading buffer (50 mM Tris-HCl, 8 M \nurea, 2 M thiourea, 75 mM DTT, 3% SDS, 0.05% bromophenol blue, pH 6.8) in a 1:3 ratio, heated to \n100°C for 4 min, and loaded on 12% CriterionTM XT Bis-Tris Protein gels (BioRad). Protein bands were \nvisualized by staining with InstantBlue (Expedon). Molecular weights were estimated relative to \nPrecision Plus Protein All Blue Prestained Standards (Biorad). \n \nExpression and purification  \nThe open reading frames of the selected genes were optimized for E. coli codon usage and synthesized \nby Twist Bioscience. For tagged protein expression, synthetic genes were digested with NdeI and \nBamHI restriction enzymes and inserted into the pET28a(+) vector, in frame with a thrombin-cleavable \nN-terminal His6 -tag. For untagged protein expression, genes were digested with NdeI and XhoI \nrestriction enzymes and cloned into the pET41c(+) vector. Cloning accuracy was assessed by Sanger \nsequencing (Eurofins).  \n \nThe resulting recombinant plasmids were used for transformation of E. coli Rosetta \n2(DE3)pLysS cells according to standard procedures60. Overnight cultures were diluted 1:100 and grown \nat 37°C, 140 rpm until OD600 reached 0.6. Protein overexpression was induced with 1 mM of isopropyl-\nb-D-thiogalactopyranoside (IPTG) for 16 h at 20°C. Cells were harvested by centrifugation at 8,000 ´ g \nfor 45 min at 4°C, and pellets were stored at -80°C. Cells were resuspended in lysis buffer (50 mM Tris-\nHCl, 150 mM NaCl, 0.1% Triton ´100, pH 8.0) supplemented with 0.05 mg/mL lysozyme, 0.01 mL/mL \nMgSO4 2M, 1 mg/mL Pefabloc SC, 0.05 mg/mL DNase, 0.2 mg/mL RNase, and were disrupted on ice \nin a Vibra-Cell sonifier (35% amplitude with five on/off cycles of  30 s each). For thermostable proteins, \nthe lysate was heated at 70°C for 15 min. Insoluble particles were pelleted by centrifugation (16,000 ´ \ng for 30 min at 4°C) and the cleared extract was filtered at 0.45 μm and 0.22 μm. The recombinant \nproteins were purified from the soluble fractions to near homogeneity using various combinations of \naffinity, anion exchange and gel filtration chromatography. \n \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted March 2, 2025. ; https://doi.org/10.1101/2025.02.26.640355doi: bioRxiv preprint \n\n 11 \nFor ThTET purific ation, after cell lysis, incubation at 70 °C for 15 min and clarification, the \nresulting supernatant was supplemented with imidazole (final concentration 10 mM) and loaded on a \nHiTrap Chelating HP 5 mL column (Cityva) equilibrated with 50 mM Tris -HCl, 150 mM NaCl,  10 mM \nimidazole, pH 8.0. Bound proteins were eluted with a linear gradient of imidazole (10 to 500 mM). \nFractions corresponding to the elution peak at 400 mM imidazole were pooled, dialysed against 50 mM \nTris-HCl, 20 mM NaCl, pH 8.0 and loaded on a ResourceQ column (Cytiva) equilibrated with the same \nbuffer. Elution was achieved by a linear NaCl gradient (20 to 500 mM) and fractions containing protein \nof similar mass (37-39 kDa) according to SDS-PAGE were combined and concentrated using an Amicon \nUltra-15 ultrafiltration unit (Millipore) with a 30 kDa cutoff. The protein was utlimately loaded on a \nSuperose 6 Increase 10/300 GL column (Cytiva) in 50 mM Tris, 150 mM NaCl, pH 8.0. Fractions from \nthe elution peak corresponding to a molecular mass around 450 kDa  were pooled and subsequently \nconcentrated using an an Amicon Ultra-15 ultrafiltration unit (Millipore) with a 30 kDa cutoff. \n \nFor HoTETb purification, a fter cell lysis, incubation at 70 °C for 15 min and clarification, the \nresulting supernatant was diluted to a final NaCl concentration of 75 mM and loaded on a ResourceQ \ncolumn (Cytiva) equilibrated with 50 mM Tris-HCl, 75 mM NaCl, pH 8.0. Elution was achieved by a linear \nNaCl gradient (75 to 300 mM) and fractions containing protein of similar mass (37-39 kDa) according to \nSDS-PAGE were combined and concentrated using an Amicon Ultra-15 ultrafiltration unit (Millipore) with \na 30 kDa cutoff. The protein was then loaded on a Superose 6 Increase 10/300 GL column (Cytiva) in \n50 mM Tris-HCl, 150 mM NaCl, pH 8.0. Fractions from the elution peak corresponding to a molecular \nmass around 450 kDa were pooled and subsequently concentrated using an Amicon Ultra -15 \nultrafiltration unit (Millipore) with a 30 kDa cutoff.  \n \nFor T aTET purification, a fter cell lysis, incubation at 70 °C for 15 min and clarification, the \nresulting supernatant was dialysed against 50 mM Tris -HCl, 50 mM NaCl, pH 8.0 and loaded on a \nResourceQ column (Cytiva) equilibrated with the same buffer. Elution was achieved by a linear NaCl \ngradient (50 mM to 1 M) and fractions containing protein of similar mass (37-39 kDa) according to SDS-\nPAGE were combined and concentrated using an Amicon Ultra -15 ultrafiltration unit (Millipore) with a \n30 kDa cutoff. The protein was then loaded on  a Superdex 200 10/300 GL column (Cytiva) in 50 mM \nTris-HCl, 150 mM NaCl, pH 8.0. Fractions from the elution peak corresponding to a molecular mass \naround 450 kDa were pooled and subsequently concentrated using an Amicon Ultra -15 ultrafiltration \nunit with a 30 kDa cutoff.  \n \nFor PsTETa, PsTETc, and MtTET purification, a fter cell lysis and clarification, the resulting \nsupernatant was dialysed against 50 mM Tris -HCl, 20 mM NaCl, pH 8.0 and loaded on a DEAE \nsepharose CL -6B resin (Cytiva, XK16/20 column) equilibrated with the same buffer. Elution was \nachieved by a linear NaCl gradient (20 to 600 mM) and fractions containing protein of similar mass (37-\n39 kDa) according to SDS-PAGE were combined, dialysed against 50 mM Tris-HCl, 50 mM NaCl, pH \n8.0 and loaded on a ResourceQ column (Cytiva) equilibrated with the same buffer. Elution was achieved \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted March 2, 2025. ; https://doi.org/10.1101/2025.02.26.640355doi: bioRxiv preprint \n\n 12 \nby a linear NaCl gradient (50 to 500 mM) and fractions containing the protein of interest were pooled \nand concentrated using an Amicon Ultra-15 ultrafiltration unit (Millipore) with a 30 kDa cutoff. The protein \nwas then loaded on a Superdex 200 10/300 GL column (Cytiva) in 50 mM Tris-HCl, 150 mM NaCl, pH \n8.0. Fractions from the elution peak corresponding to a molecular mass around 450 kDa were combined \nand subsequently concentrated using an Amicon Ultra-15 ultrafiltration unit with a 30 kDa cutoff.  \n \nNegative-stain electron microscopy  \n4 µL of purified protein samples (0.1 mg/mL) were absorbed onto the clean side of a carbon film on \nmica, stained, and transferred to a 400-mesh copper grid. Images were taken under low dose conditions \n(<10 e-/Å2) with defocus values between 1.2 and 2.5 μm on a Tecnai 12 LaB6 electron microscope at \n120 kV accelerating voltage using CCD Camera Gatan Orius 1000. \n \nAlphaFold model predictions \nAlphaFold model predictions were calculated using the AlphaFold3 server (https://alphafoldserver.com/ \naccessed on May 17th, 2024). \n \nEnzymatic characterization protocol  \nM42 peptidase hydrolytic activities on synthetic chromogenic and fluorogenic substrates were assayed \nusing aminoacyl-para-nitroaniline (pNA) and aminoacyl -7-amino-4-methylcoumarin (AMC) conjugates \nordered from Bachem. Substrates were solubilized in 100% di methylsulfoxide (DMSO) to a final \nconcentration of 20 mM. All assays described below were carried out according to the following standard \nprocedure18. Reactions were initiated by addition of 2 to 10 µg/mL of enzyme to a pre-warmed mixture \ncontaining 2.5 mM of the synthetic substrate in 50 mM buffer (pH 5,5 – 11), 150 mM KCl, and 1 mM \nCCl2 (X = Ca, Co, Fe, Mg, Mn, Ni or Zn) in a total volume of 60 μL. To avoid water evaporation, the total \nvolume was covered by 25 μL of mineral oil. Incubations were performed for 3 min to 1 h, reactions were \nstopped by the addition of 60 μL of 0.1 M acetic a cid, and samples were placed on ice. After \ncentrifugation at 6,00 0 ´ g for 3 min, liberated pNA or AMC quantities were quantified by OD 405 or \nfluorescence (excitation and emission wavelengths 360 nm and 460 nm, respectively) measurement in \na Synergy HT microplate reader (BioTek). Three replicates and two enzyme blanks were assayed for \neach experimental point. Enzyme concentrations and incubation durations were adjusted for each \npeptidase to produce a robust signal for accurate measurement. \n \nFor each enzyme, optimal temperature, pH and metallic cofactor were determined using the \nsubstrate on which maximum activity was measured. The effect of temperature on M42 peptidase \nactivities was evaluated between 20 and 100°C. Assays were conducted as previously described in \npresence of 50 mM HEPES, 150 mM KCl, and 1 mM CoCl 2, pH 7.5. To prevent enzyme denaturation \nand to ensure stable enzymatic activity, optimal pH and metallic cofactor were established 10°C below \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted March 2, 2025. ; https://doi.org/10.1101/2025.02.26.640355doi: bioRxiv preprint \n\n 13 \nthe determined optimal temperature. The effect of  metal cations on M42 peptidase activities was \nassessed using 1 mM (0.1 mM for TaTET) of XCl2 metal (X = Ca, Co, Fe, Mg, Mn, Ni or Zn) with 50 mM \nHEPES, 150 mM KCl, pH 7.5. The influence of pH was studied in presence of 1 mM CoCl2 (0.1 mM for \nTaTET) using the following buffers: MES (pH 5.5 to 6.5), HEPES (pH 7.0 to 8.0), CHES (pH 8.5 to 9.5), \nand CAPS (pH 10.0 to 11.0).  For each peptidase, substrate specificity was determined using optimal \nmetal cofactor and pH. Incubation was performed 10°C the established optimal temperature. \n \nTET Enzyme concentration Incubation duration \nHoTETb 2 µg/mL 3 min \nMtTET 8 µg/mL 10 min \nPsTETa 10 µg/mL 60 min \nPsTETc 10 µg/mL 20 min \nTaTET 10 µg/mL 15 min \nThTET 10 µg/mL 3 min \n \nData availability  \nData used to produce our results are provided as supporting data and can be found here :  \nhttps://data.mendeley.com/preview/rc9nntwy6b?a=d335eeda-c9b3-429e-ae60-c07f432bf7cf. \n \nAcknowledgments  \nWe thank Audrey Bossé for her technical support throughout this project. This work used the platforms \nof the Grenoble Instruct-ERIC center (ISBG ; UAR 3518 CNRS-CEA-UGA-EMBL) within the Grenoble \nPartnership for Structural Biology (PSB), supported by FRISBI (ANR -10-INBS-0005-02) and GRAL, \nfinanced within the University Grenoble Alpes graduate school (Ecoles Universitaires de Recherche) \nCBH-EUR-GS (ANR-17-EURE-0003). IBS acknowledges integration into the Interdisciplinary Research \nInstitute of Grenoble (IRIG,  CEA). This work was financially supported by the Région Bretagne , the \nFrench Research Institute for Exploitation of the Sea, and the Grenoble Alliance for Integrated Structural \nCell Biology (GRAL) . This work was supported by the Bettencourt -Schueller Foundation programme \nImpulscience (ENVOL) to S.G., Laboratoire d’Excellence ‘Integrative Biology of Emerging Infectious \nDiseases’ (grant no. ANR-10-LABX-62-IBEID), and the Fondation pour la Recherche Médicale (FRM). \nThis work used the computational and storage services (TARS cluster) provided by the IT department \nat Institut Pasteur, Paris. \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted March 2, 2025. ; https://doi.org/10.1101/2025.02.26.640355doi: bioRxiv preprint \n\n 14 \nReferences \n1. Sunagawa, S. et al.  Tara Oceans: towards global ocean ecosystems biology. Nat. Rev. \nMicrobiol. 18, 428–445 (2020). \n2. Moran, M. A. The global ocean microbiome. Science 350, aac8455 (2015). \n3. Kopf, A. et al. The ocean sampling day consortium. GigaSci. 4, 27 (2015). \n4. Madhavan, A., Sindhu, R., Parameswaran, B., Sukumaran, R. K. & Pandey, A. Metagenome \nAnalysis: a Powerful Tool for Enzyme Bioprospecting. Appl. Biochem. Biotechnol. 183, 636–651 (2017). \n5. Lema, N. K., Gemeda, M. T. & Woldesemayat, A. A. Recent Advances in Metagenomic \nApproaches, Applications, and Challenges. Curr. Microbiol. 80, 347 (2023). \n6. Ariaeenejad, S. et al. Precision enzyme discovery through targeted mining of metagenomic \ndata. Nat. Prod. Bioprospect. 14, 7 (2024). \n7. Ngara, T. R. & Zhang, H. Recent Advances in Function -Based Metagenomic Screening. \nGenom. Proteom. Bioinform. 16, 405–415 (2018). \n8. Franzetti, B. et al. Tetrahedral aminopeptidase: a novel large protease complex from archaea. \nEMBO J. 21, 2132–2138 (2002). \n9. Rawlings, N. D. et al. The MEROPS database of proteolytic enzymes, their substrates and \ninhibitors in 2017 and a comparison with peptidases in the PANTHER database. Nucl. Acids Res. 46, \n624–632 (2018). \n10. Paysan-Lafosse, T. et al. InterPro in 2022. Nucl. Acids Res. 51, D418–D427 (2023). \n11. Borissenko, L. & Groll, M. Crystal structure of TET protease reveals complementary protein \ndegradation pathways in prokaryotes. J. Mol. Biol. 346, 1207–1219 (2005). \n12. Russo, S. & Baumann, U. Crystal Structure of a Dodecameric tetrahedral -shaped \naminopeptidase. J. Biol. Chem. 279, 51275–51281 (2004). \n13. Durá, M. A. et al. The structural and biochemical characterizations of a novel TET peptidase \ncomplex from Pyrococcus horikoshii reveal an integrated peptide degradation system in \nhyperthermophilic Archaea: Characterization of P. horikoshii TET3 peptidase. Mol. Microbiol. 72, 26–40 \n(2009). \n14. Appolaire, A. et al. TET peptidases: a family of tetrahedral complexes conserved in prokaryotes. \nBiochimie 122, 188–196 (2016). \n15. Colombo, M., Girard, E. & Franzetti, B. Tuned by metals: the TET peptidase activity is controlled \nby 3 metal binding sites. Sci. Rep. 6, 20876–20888 (2016). \n16. Appolaire, A. et al. Small-angle neutron scattering reveals the assembly mode and oligomeric \narchitecture of TET, a large, dodecameric aminopeptidase. Acta Cryst. D 70, 2983–2993 (2014). \n17. Rosenbaum, E., Ferruit, M., Durá, M. A. & Franzetti, B. Studies on the parameters controlling \nthe stability of the TET peptidase superstructure from Pyrococcus horikoshii revealed a crucial role of \npH and catalytic metals in the oligomerization process. Biochim. Biophys. Acta Proteins Proteom. 1814, \n1289–1294 (2011). \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted March 2, 2025. ; https://doi.org/10.1101/2025.02.26.640355doi: bioRxiv preprint \n\n 15 \n18. Durá, M. A. et al.  Characterization of a TET -like aminopeptidase complex from the \nhyperthermophilic archaeon Pyrococcus horikoshii. Biochemistry 44, 3477–3486 (2005). \n19. Durá, M. A. & Franzetti, B. PhTET1 aminopeptidase. in Handbook of Proteolytic Enzymes vol. \n2 1638–1645 (Elsevier, 2013). \n20. Basbous, H., Appolaire, A., Girard, E. & Franzetti, B. Characterization of a glycyl -specific TET \naminopeptidase complex from Pyrococcus horikoshii. J. Bacteriol. 200, 1–11 (2018). \n21. Schoehn, G. et al. An archaeal peptidase assembles into two different quaternary structures. J. \nBiol. Chem. 281, 36327–36337 (2006). \n22. Franzetti, B., Girard, E., Appolaire, A. & Basbous, H. Peptidase and Its Uses. (2019). \n23. Franzetti, B., Girard, E. & Appolaire, A. Use of a Combination of Tet Exoproteases Obtained \nfrom Extremophilic Microorganisms for Hydrolyzing Polypeptides. (2019). \n24. Slutskaya, E. S. et al.  Characterization of a novel M42 aminopeptidase from crenarchaeon \nDesulfurococcus kamchatkensis. Dokl. Biochem. Biophys. 442, 30–32 (2012). \n25. Dutoit, R., Brandt, N., Legrain, C. & Bauvois, C. Functional characterization of two M42 \naminopeptidases erroneously annotated as cellulases. PLoS ONE 7, 1–8 (2012). \n26. Rawlings, N. D. & Barrett, A. J. Introduction: Metallopeptidases and Their Clans. in Handbook \nof Proteolytic Enzymes 325–370 (Elsevier, 2013). doi:10.1016/B978-0-12-382219-2.00077-6. \n27. Dutoit, R. The paradigm of self -compartmentalized M42 aminopeptidases: insight into their \noligomerization, substrate specificities, and physiological function. (Université libre de Bruxelles, Faculté \ndes Sciences – Biologie Moléculaire, 2021). \n28. Lee, H. S. et al.  Biochemical characterization of deblocking aminopeptidase from \nhyperthermophilic archaeon Thermococcus onnurineus NA1. J. Biosci. Bioeng. 104, 188–194 (2007). \n29. Lee, Y. G., Leem, S.-H., Chung, Y.-H. & Kim, S. I. Characterization of thermostable deblocking \naminopeptidases of archaeon Thermococcus onnurineus NA1 by proteomic and biochemical \napproaches. J. Microbiol. 50, 792–797 (2012). \n30. Robinson, M. W. et al. MHJ_0125 is an M42 glutamyl aminopeptidase that moonlights as a \nmultifunctional adhesin on the surface of Mycoplasma hyopneumoniae. Open Biol. 3, 1–15 (2013). \n31. Appolaire, A. et al. The TET2 and TET3 aminopeptidases from Pyrococcus horikoshii form a \nhetero-subunit peptidasome with enhanced peptide destruction properties. Mol. Microbiol. 94, 803–814 \n(2014). \n32. Susanti, D., Frazier, M. C. & Mukhopadhyay, B. A genetic system for Methanocaldococcus \njannaschii: an evolutionary deeply rooted hyperthermophilic methanarchaeon. Front. Microbiol.  10, \n(2019). \n33. Roncari, G. & Zuber, H. THERMOPHILIC AMINOPEPTIDASES FROM BACILLUS \nSTEAROTHERMOPHILUS. I. ISOLATION, SPECIFICITY, AND GENERAL PROPERTIES OF THE \nTHERMOSTABLE AMINOPEPTIDASE I *. Int. J. Protein Res. 1, 45–61 (1969). \n34. Kumaki, Y. et al. Family M42 aminopeptidase from the syntrophic bacterium Symbiobacterium \nthermophilum: characterization using recombinant protein. J. Biosci. Bioeng. 111, 134–139 (2011). \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted March 2, 2025. ; https://doi.org/10.1101/2025.02.26.640355doi: bioRxiv preprint \n\n 16 \n35. Huber, J. A., Butterfield, D. A. & Baross, J. A. Temporal Changes in Archaeal Diversity and \nChemistry in a Mid-Ocean Ridge Subseafloor Habitat. Appl. Environ. Microbiol. 68, 1585–1594 (2002). \n36. Nercessian, O., Reysenbach, A., Prieur, D. & Jeanthon, C. Archaeal diversity associated with \nin situ samplers deployed on hydrothermal vents on the East Pacific Rise (13°N). Environ. Microbiol. 5, \n492–502 (2003). \n37. Takai, K. et al. Spatial Distribution of Marine Crenarchaeota Group I in the Vicinity of Deep-Sea \nHydrothermal Systems. Appl Environ Microbiol 70, 2404–2413 (2004). \n38. Baker, B. J. et al. Genomic inference of the metabolism of cosmopolitan subsurface Archaea, \nHadesarchaea. Nat. Microbiol. 1, 16002–16008 (2016). \n39. Lee, H. S. et al. The Complete Genome Sequence of Thermococcus onnurineus NA1 Reveals \na Mixed Heterotrophic and Carboxydotrophic Metabolism. J. Bacteriol. 190, 7491–7499 (2008). \n40. Pillot, G. et al. Identification of enriched hyperthermophilic microbial communities from a deep-\nsea hydrothermal vent chimney under electrolithoautotrophic culture conditions. Sci. Rep. 11, 14782–\n14793 (2021). \n41. Bertoldo, C. & Antranikian, G. The Order Thermococcales. in The Prokaryotes (eds. Dworkin, \nM., Falkow, S., Rosenberg, E., Schleifer, K. -H. & Stackebrandt, E.) 69 –81 (Springer New York, New \nYork, NY, 2006). doi:10.1007/0-387-30743-5_5. \n42. Oren, A. The Order Halobacteriales. in The Prokaryotes  (eds. Dworkin, M., Falkow, S., \nRosenberg, E., Schleifer, K.-H. & Stackebrandt, E.) 113–164 (Springer New York, New York, NY, 2006). \ndoi:10.1007/0-387-30743-5_8. \n43. Itoh, T. Vulcanisaeta distributa gen. nov., sp. nov., and Vulcanisaeta souniana sp. nov., novel \nhyperthermophilic, rod-shaped crenarchaeotes isolated from hot springs in Japan. Int. J. Syst. Evol. \nMicrobiol. 52, 1097–1104 (2002). \n44. Cai, M. et al. Diverse Asgard archaea including the novel phylum Gerdarchaeota participate in \norganic matter degradation. Sci. China Life Sci. 63, 886–897 (2020). \n45. Liu, Y. et al. Expanded diversity of Asgard archaea and their relationships with eukaryotes. \nNature 593, 553–557 (2021). \n46. Bulzu, P.-A. et al. Casting light on Asgardarchaeota metabolism in a sunlit microoxic niche. Nat. \nMicrobiol. 4, 1129–1137 (2019). \n47. McKay, L. J. et al. Co-occurring genomic capacity for anaerobic methane and dissimilatory sulfur \nmetabolisms discovered in the Korarchaeota. Nat. Microbiol. 4, 614–622 (2019). \n48. Elkins, J. G. et al. A korarchaeal genome reveals insights into the evolution of the Archaea. \nProc. Natl. Acad. Sci. U.S.A. 105, 8102–8107 (2008). \n49. Feng, X., Wang, Y., Zubin, R. & Wang, F. Core Metabolic Features and Hot Origin of \nBathyarchaeota. Engineering 5, 498–504 (2019). \n50. Khomyakova, M. A., Merkel, A. Y., Mamiy, D. D., Klyukina, A. A. & Slobodkin, A. I. Phenotypic \nand genomic characterization of Bathyarchaeum tardum gen. nov., sp. nov., a cultivated representative \nof the archaeal class Bathyarchaeia. Front. Microbiol. 14, 1214631–1214645 (2023). \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted March 2, 2025. ; https://doi.org/10.1101/2025.02.26.640355doi: bioRxiv preprint \n\n 17 \n51. Hou, J. et al.  Taxonomic and carbon metabolic diversification of Bathyarchaeia during its \ncoevolution history with early Earth surface environment. Sci. Adv. 9, 1–17 (2023). \n52. Zhang, Z. et al. Protein sequence similarity searches using patterns as seeds. Nucl. Acids Res. \n26, 3986–3990 (1998). \n53. Notredame, C., Higgins, D. G. & Heringa, J. T -coffee: a novel method for fast and accurate \nmultiple sequence alignment. J. Mol. Biol. 302, 205–217 (2000). \n54. Johnson, L. S., Eddy, S. R. & Portugaly, E. Hidden Markov model speed heuristic and iterative \nHMM search procedure. BMC Bioinformatics 11, 431–438 (2010). \n55. Katoh, K. & Standley, D. M. MAFFT Multiple Sequence Alignment Software Version 7: \nImprovements in Performance and Usability. Mol. Biol. Evol. 30, 772–780 (2013). \n56. Capella-Gutiérrez, S., Silla-Martínez, J. M. & Gabaldón, T. trimAl: a tool for automated alignment \ntrimming in large-scale phylogenetic analyses. Bioinformatics 25, 1972–1973 (2009). \n57. Minh, B. Q. et al. IQ-TREE 2: New Models and Efficient Methods for Phylogenetic Inference in \nthe Genomic Era. Mol. Biol. Evol. 37, 1530–1534 (2020). \n58. Kalyaanamoorthy, S., Minh, B. Q., Wong, T. K. F., Von Haeseler, A. & Jermiin, L. S. \nModelFinder: fast model selection for accurate phylogenetic estimates. Nat. Methods 14, 587 –589 \n(2017). \n59. Letunic, I. & Bork, P. Interactive Tree Of Life (iTOL) v4: recent updates and new developments. \nNucl. Acids Res. 47, 256–259 (2019). \n60. Sambrook, J. & Russell, D. W. Molecular Cloning: A Laboratory Manual. (Cold Spring Harbor \nLaboratory Press, Cold Spring Harbor, N.Y, 2001). \n \n  \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted March 2, 2025. ; https://doi.org/10.1101/2025.02.26.640355doi: bioRxiv preprint \n\n 18 \n \nFig. 1: Structure-based determinants for identifying M42 peptidases.  Multiple sequence alignment of \nstructurally and functionally characterized M42 aminopeptidases. The represented secondary structure \ncorresponds to PhTET1 (PDB code: 2WYR). Two criteria were retained for the identification of TET \naminopeptidases: conserved catalytic residues H62, D64, D173, E205, E206, [DE]228, H307 (red stars), and \nconserved glycine residues 44, 77, 85, 86, and 211 (blue triangles). Residue numbering according to PhTET1 \nsequence. \n \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted March 2, 2025. ; https://doi.org/10.1101/2025.02.26.640355doi: bioRxiv preprint \n\n 19 \n  \nFig. 2: Phylogeny of archaeal TET peptidase homologues. Maximum-likelihood phylogeny obtained from \nan alignment of 1,826 sequences and 337 amino acid positions. The scale bar represents the average number \nof substitutions per site. Circles at the branches indicate ultra-fast bootstrap values >= 90 %. TET1 to TET4 \nfamilies were delineated based on the taxonomic distribution and the topology of the tree. Gray and red bars \non the inner circle indicate enzymes characterized prior to  or during  this study , respectively . Archaeal \ntaxonomic groups are represented on the outer circle. \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted March 2, 2025. ; https://doi.org/10.1101/2025.02.26.640355doi: bioRxiv preprint \n\n 20 \n \n  \nFig. 3: Characterized TET aminopeptidases exhibit diverse s ubstrate specificities.  Cleavage \nspecificities were assayed using synthetic chromogenic and fluorogenic substrates. For each enzyme, \nactivities are expressed as percentage of the maximum activity observed, which was attributed a value of \n100%. Enzymes characterized prior to this study are indicated by dashed lines. Error bars indicate ±s.d. with \nn=3. NA: not assessed. \n \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted March 2, 2025. ; https://doi.org/10.1101/2025.02.26.640355doi: bioRxiv preprint \n\n 21 \n \n  \nFig. 4: Phylogenetic distribution of the TET families in archaea. Distribution of the different TET families \nhomologs on a schematic reference phylogeny of Archaea based on Garcia et al 49. The sizes of the circles \nvary between 0% and 100% and indicate the percentage of genomes where a family is found. Circles are \ncolored according to the activity spectrum of the characterized representatives of each family: pink for generic \nactivities, green for specific activities, and yellow for undetermined activities \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted March 2, 2025. ; https://doi.org/10.1101/2025.02.26.640355doi: bioRxiv preprint \n\n 22 \nTable 1: Candidate proteins for further characterization. Thirteen proteins spanning the phylogenetic tree \nand representing a broad range of taxonomically diverse species were selected. \nTET Species Taxonomy \nHvTETa \nHaloferax volcanii \nMethanotecta; Halobacteria; Halobacteriales; Haloferacaceae; \nHaloferax; HvTETb \nHoTETa Ca. Hodarchaeales \narchaeon LC_3 \nAsgardarchaeota; Heimdallarchaeia; Hodarchaeales; LC-3; \nLC-3;  HoTETb \nPsTETa Ca. \nPrometheoarchaeum \nsyntrophicum \nAsgardarchaeota; Lokiarchaeia; CR-4; AMARA-1; \nPrometheoarchaeum;  \nPsTETb \nPsTETc \nThTET \nCa. Thorarchaeota \narchaeon MP8T-1 \nAsgardarchaeota; Thorarchaeia; Thorarchaeales; \nThorarchaeaceae; MP8T-1;  \nTaTET \nThermosphaera \naggregans \nTACK; Thermoproteia/Crenarchaeota; Sulfolobales; \nDesulfurococcaceae; Thermosphaera; \nMfTET \nMethanothermus \nfervidus \nMethanomada; Methanobacteria; Methanobacteriales; \nMethanothermaceae; Methanothermus; \nMtTET \nMethanoculleus \nthermophilus \nMethanotecta; Methanomicrobia; Methanomicrobiales; \nMethanoculleaceae; Methanoculleus; \nMkTET \nMethanopyrus \nkandleri \nMethanomada; Methanopyri; Methanopyrales; \nMethanopyraceae; Methanopyrus; \nAlTET \nAltarchaeia archaeon \nex4484_2 \nDPANN; Altarchaeia; IMC4; QMZM01; EX4484-2;  \n \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted March 2, 2025. ; https://doi.org/10.1101/2025.02.26.640355doi: bioRxiv preprint","source_license":"CC-BY-4.0","license_restricted":false}