Unique Characteristics of the Pyrrolysine System in the 7th Order of Methanogens: Implications for...

Research ArticleUnique Characteristics of the Pyrrolysine System inthe 7th Order of Methanogens Implications for the Evolutionof a Genetic Code Expansion Cassette

Guillaume Borrel12 Nadia Gaci1 Pierre Peyret1 Paul W OrsquoToole2

Simonetta Gribaldo3 and Jean-Franccedilois Brugegravere1

1 1EA-4678 CIDAM Clermont Universite Universite drsquoAuvergne Place Henri Dunant 63001 Clermont-Ferrand France2Department of Microbiology and Alimentary Pharmabiotic Centre University College Cork Western Road Cork Ireland3 Institut Pasteur Department of Microbiology Unite de Biologie Moleculaire du Gene chez les Extremophiles 28 rue du Dr Roux75015 Paris France

Correspondence should be addressed to Jean-Francois Brugere jfbrugereudamailfr

Received 30 August 2013 Accepted 19 October 2013 Published 27 January 2014

Academic Editor Kyung Mo Kim

Copyright copy 2014 Guillaume Borrel et al This is an open access article distributed under the Creative Commons AttributionLicense which permits unrestricted use distribution and reproduction in any medium provided the original work is properlycited

Pyrrolysine (Pyl) the 22nd proteogenic amino acid was restricted until recently to few organisms Its translational use necessitatesthe presence of enzymes for synthesizing it from lysine a dedicated amber stop codon suppressor tRNA and a specific amino-acyl tRNA synthetase The three genomes of the recently proposed Thermoplasmata-related 7th order of methanogens containthe complete genetic set for Pyl synthesis and its translational use Here we have analyzed the genomic features of the Pyl-coding system in these three genomes with those previously known from Bacteria and Archaea and analyzed the phylogeny ofeach component This shows unique peculiarities notably an amber tRNAPyl with an imperfect anticodon stem and a shortenedtRNAPyl synthetase Phylogenetic analysis indicates that a Pyl-coding system was present in the ancestor of the seventh orderof methanogens and appears more closely related to Bacteria than to Methanosarcinaceae suggesting the involvement of lateralgene transfer in the spreading of pyrrolysine between the two prokaryotic domains We propose that the Pyl-coding system likelyemerged once in Archaea in a hydrogenotrophic andmethanol-H

2-dependentmethylotrophic methanogenThe close relationship

between methanogenesis and the Pyl system provides a possible example of expansion of a still evolving genetic code shaped bymetabolic requirements

1 Introduction

Protein synthesis relies on 20 canonical amino acids encodedaccordingly to a genetic code each codon being recognizedby an aminoacyl-tRNA The molecular basis of the genotypeto phenotype correspondence relies on the conjunction oftRNAs and of aminoacyl-tRNA synthetases (aaRS) Post-translational modifications of amino acids extend the chemi-cal nature of proteins with functional implications in cellularprocesses [1 2] Another naturally occurring mechanismexpands the genetic code to 22 amino acids by addingSelenocysteine (Sec 2-selenoalanine) [3 4] and Pyrrolysine(Pyl 4-methyl-pyrroline-5-carboxylate linked to the 120576N of

lysine through an amide linkage) [5 6] Sec is present inmanyorganisms from the three domains of life [7] with potentiallyalmost a quarter of sequenced bacteria synthesizing it [8]It is at least present in two orders of Euryarchaeota themethanogens Methanococcales and Methanopyrales [9 10]Its synthesis and incorporation differ fromother amino acidsas it is synthesized after a serine has been branched intotRNASec

UCA (recognizing the opal codon UGA) that is nextmodified into a Sec-tRNASec

UCAIn contrast Pyl is restricted to a very small number of

organisms and proteins It necessitates a complex systemwithspecialized enzymes for biosynthesis of Pyl a dedicated tRNAand an associated unique aaRS [11 12] Pyl is first synthesized

Hindawi Publishing CorporationArchaeaVolume 2014 Article ID 374146 11 pageshttpdxdoiorg1011552014374146

2 Archaea

as a free amino acid in the cell by the products of the pylBCDgenes [13 14] from two lysines one being methylated into 3-methylornithine (catalyzed by PylB a lysine mutase-proline-2 methylase) and condensed to the second lysine to form3-methylornithinyl-N6-lysine (PylC Pyrrolysine synthetase)[15] The pyrrole ring is then formed by oxidation with anatypical dehydrogenase the Proline reductase (also calledPyl synthase PylD) [16] with concomitant release of anamino group during the cycle formation The gene productof pylT forms a dedicated tRNA used for the translationalincorporation of Pyl It is an amber decoding tRNACUA [6]showing some unusual features compared to other tRNAs[11 17ndash19] the anticodon stem is longer (6 nucleotides insteadof 5) while other parts are shorter D-loop with only 5 basesacceptor and D-stems separated by one base instead of 2and a variable loop of only 3 bases Moreover differing fromalmost all tRNAs the D-loop does not contain the G

18G19

bases nor the usual T54Ψ55C56

bases in the T-loop Finallythe aaRS encoded by pylS catalyzes the ligation of Pyl to itscognate tRNA species warranting the right correspondencebetween DNA and proteins The anticodon of the tRNAseems not to be recognized by the PylRS at least in Dhafniense [20] Whereas the archaeal PylRS is encoded bya single gene (pylS) the bacterial PylRS is encoded by twogenes pylSn for the firstN-terminal 140 amino acids and pylScfor the remaining sequenceThis pyrrolysyl-tRNA synthetasebelongs to the class II aaRS and its structure has beenresolved with and without its substratesanalogs [17 21ndash24]It has likely a homodimeric or homotetrameric quaternarystructure

To date Pyl-containing proteins have been found only inthe methanogens of the family Methanosarcinaceae and in afew bacteria belonging to Firmicutes and two mem-bers of deltaproteobacteria Bilophila wadsworthia and anendosymbiont of the worm Olavius algarvensis [25] Pyl ispresent almost specifically in the methyltransferases (MT)involved in the methanogenesis pathway frommonomethyl-dimethyl- and trimethylamine in methanogens respectivelyencoded by the genes mtmB mtbB and mttB [26 27] and intheir bacterial homologs whose function is unclearHowevermany bacteria harbor mttB homologs that lack the in-frameamber codon and the Methanosarcinales Methanococcoidesburtonii hasmttB genes with and without an in-frame ambercodon [11 25] In contrast themtbB gene is almost exclusivelypresent inMethanosarcinales and harbors an in-frame ambercodonThemtmB gene represents an intermediary case witha few Pyl-containing bacteria possessing it leading either to aMT with Pyl (Acetohalobium arabaticum Desulfotomaculumacetoxidans) or not (Bilophila wadsworthia) [25]

The existence of a 7th order of methanogens related toThermoplasmatales was proposed from molecular data of16S rRNA and mcrA (methanogenesis marker) sequencesretrieved from human stools [28 29]This is strengthened bygrowing molecular data from various environments [30 31]and by phylogenomics studies [32] established from the firstgenomes of members of this clade [33 34] Moreover a firstmember of this orderMethanomassiliicoccus luminyensis hasbeen isolated and grows onmethanol +H

2[35]We identified

the presence of genes formethanogenesis frommethanol and

also from dimethylsulfide andmethylamines in the two othersequenced genomes (ldquoCandidatus Methanomethylophilusalvusrdquo [33] ldquoCandidatusMethanomassiliicoccus intestinalisrdquo[36]) together with the Pyl-coding genes and in-frame ambercodon in the mtmB mtbB and mttB genes of these threespecies It has now been shown that the H

2-dependent

methanogenesis from trimethylamine is effective for Mluminyensis [37] Also ruminal methanogens of the samelineage as ldquoCa M alvusrdquo (Rumen Cluster C or RCC) whencultured in consortia with trimethylamine (TMA) show anincrease of methanogenesis and mRNAs for MTs with anamber in-frame codon are detected [38] Altogether thisgreatly suggests the presence of Pyl in these MTs

The recent uncovered lineage with all the featuresneeded for Pyl encoding and use representing a newarchaeal methanogenic Order provides an opportunity tobetter understand the origin distribution and diversity ofPyrrolysine-coding systems

2 Materials and Methods

Genomic sequence data were obtained through GenBankFor the 7th order Thermoplasmata-related methano-gens the accession numbers are CP0040491 (ldquoCa Methano-methylophilus alvusrdquo) CP0059341 (ldquoCa Methanomassilii-coccus intestinalisrdquo) and NZ CAJE0100001 to NZ CAJE010-0026 (Methanomassiliicoccus luminyensis) For the genomicorganization and comparison of pyl genes genomic sequen-ces were either treated with the RAST annotation server [39]or a local Artemis platform [40] Blast searches [41] wereeither performed directly on the RAST server on the nrdatabase atNCBI or locally Sequence alignment (DNARNAand proteins) was performed using ClustalW [42] T-coffee[43] and MUSCLE [44] For RNAs the dedicated R-Coffeeprogram was also used [45] accessible at httpwwwtcoffeeorg which generates a multiple sequence align-ment using structural information RNAfold [46] (availa-ble at httprnatbiunivieacatcgi-binRNAfoldcgi) andRNAstructure [47] (available at httprnaurmcroches-teredu) were used to determine the secondary structure oftRNAThe structure of the tRNAPyl was also manually drawnby comparison with the structure of the D hafniense andMacetivorans tRNAPyl [11 20] Phylogenies were both inferredfrom maximum likelihood and Bayesian proceduresDatasets of protein homologues were aligned by MUSCLE[44] with default parameters and unambiguously alignedpositions were automatically selected by using the BMGEsoftware for multiple-alignment trimming [48] with aBLOSUM30 substitution matrix Maximum likelihood treeswere calculated by PhyML [49] and the LG amino acidsubstitution model [50] with 4 rate categories as suggestedby the AIC criterion implemented in Treefinder [51] Treeswere also calculated by Bayesian analysis with PhyloBayes[52] with the LG model (for single gene trees) or the CATmodel (for the concatenated dataset) and 4 categories ofevolutionary rates In this case two MCMC chains were runin parallel until convergence and the consensus tree wascalculated by removing the first 25 of trees as burnin

Archaea 3

T S B C D

T S B C D

T Sc B C D Sn

Sc B C DSnT

T Sc B C D Snkinase

T Sc B C D SnramA

T Sc B C DHydantoinaseoxoprolinasemttB mttC Sn

T S BC D

T S B C D

Contig 4T S B C D

Contig 23T S B C D TC

ldquoCa Methanomassiliicoccus intestinalisrdquo (CP005934)

ldquoCa Methanomethylophilus alvusrdquo (CP004049)

Thermincola sp JR (NC 014152)

Methanosarcina acetivorans (NC 003552)Methanococcoides burtonii (NC 007955)Methanohalophilus mahii (NC 014002)Methanomethylovorans hollandica (NC 019977)

Methanohalobium evestigatum (NC 014253)

Desulfitobacterium hafniense (NC 011830)

Bilophila wadsworthia (NZ ADCP00000000)Choline

ethanolamine

Desulfotomaculum acetoxidans (NC 013216)

TAG Acetohalobium arabaticum (NC 014378)

(NZ CAJE01000001 NZ CAJE01000026)Methanomassiliicoccus luminyensis

sim06 Mb

sim38 Mb

sim07 Mb

sim15 kb

Figure 1 Gene organization of the Pyl systemOn the left the gene organization of the pylBCD and pylS (pylSc and pylSn in bacteria) is shownfor Methanosarcinaceae and some representative bacteria (adapted and updated from [11 13 53]) On the right is shown the organization ofthese genes in the 7th order of methanogens

3 Results

31 Genomic Organization of the pyl Genes The pyl genesusually occur in close association in archaeal and bacte-rial genomes [11 53] In archaeal genomes they form acluster pylTSBCD that is not interrupted by other genesexcept for Methanohalobium evestigatum (Figure 1) In bac-terial genomes the pyl genes are generally organized aspylTScBCDSn or pylSnScBCD Moreover the cluster canbe interrupted by one or several genes Each of the threegenomes affiliated to 7th order of methanogen displays adistinct organization of the pyl genes ldquoCa M intestinalisrdquohas a single pyl cluster akin to the general organizationobserved in most of the Methanosarcinales Two copies of anidentically organized cluster are also found inM luminyensis(contigs 4 and 23) together with a third isolated copy of pylCand pylT present 15 kb away on the complementary strand(contig 23) In the ldquoCaM alvusrdquo genome the pyl genes occurin single copy and the pylB gene is sim07Mb distant from thepylTSCD cluster (Figure 1)

Usually these genes are closely associated with themethylamines MT genes (mtmB mtbB and mttB) and othergenes involved in these pathways (which do not encode Pyl-containing proteins) A similar gathering of the pyl genescluster with the genes involved in methylotrophic methano-genesis including mtmB mtbB and mttB is observed in thegenomes of the 7th order of methanogens (data not shown)

32 tRNAPyl The tRNAPyl homologues were retrieved fromthe three 7th order genomes by using the tRNAPyl ofMethanosarcina barkeri strain MS-DSM 800 (Accessionnumber AY0644011) as seed As mentioned above the ldquoCaM alvusrdquo and rdquoCa M intestinalisrdquo genomes harbor one pylTgene while three are present in the M luminyensis genome(Figure 1) The third tRNAPyl of M luminyensis shows notypical stem-loop structures of tRNAs using dedicated bioin-formatics tools [54] and is therefore likely a pseudogene Onthe contrary the remaining tRNAs from the three genomesall have a similar shape different from previously knowntRNAPyl and with stabilities comprised between minus285 andminus235 kJsdotmolminus1 (Figure 2) The D-loop already shortened to5 bases in other tRNAPyl is here even shorter with 4 bases inldquoCa M alvusrdquo and in one of the M luminyensis sequences(contig 4 tRNAPyl no 1) and even with 3 bases in ldquoCaM intestinalisrdquo and in one of the M luminyensis sequences(contig 23 tRNAPyl no 2) The acceptor- and D-stems areeither not separated (ldquoCa M alvusrdquo) or separated by one(ldquoCa M intestinalisrdquo tRNAPyl no 2 ofM luminyensis) or twobases (tRNAPyl no 2 of M luminyensis) The variable loopis conserved in all sequences and is equivalent to that of Dhafniense formed of the three bases CAG In the anticodonloop the adjacent base of the anticodon CUA is A in ldquoCaM alvusrdquo as observed in D hafniense and M barkeri and Cin the two other species However the most striking feature

4 Archaea

G

G

G

G

G

G

U

C U A GGG

AA

G A U CA

UAC

A

C

G

G

A

G

U

G

C

U

UC AU AC U A

GA

C

A

C

C

G

C

C

C

C

U

C

A

G C G G G

U G C C CU

U

C

G

A

AA

G

G

G

G

G

A

C

C U G G

G A C CAG

C

G

G

G

U

C

G

A

U

C

C

AC AU AC U A

GA

C

A

C

C

G

C

U

C

U

C

U

G

C G G G G U

G C C C C GC

U

A

CGG

G

C

C

G

G

A

G

G

U G GU

G A C C

C

C

G

G

C

G

G

C

A

C

C

GC AU CC U A

GA

C

C C G G G

G G C C C

C

G

C

U

U

C

C

U

U

C

U

A

CG

A

G

CG

(A)

U

G

G

G

A

A

C

C

C U A GUA

UG

G A U CA

UGA

A

U

G

G

A

U

U

G

C

C

UC AU AC U A

GA

U

A

C

C

G

C

C

U

U

U

C

G

C C G G G

G G C C CU

U

U

U

A

AG

G

G

A

G

U

G

U

C U G GU

CG

G A C CG AC

C

G

G

C

G

G

C

A

C

C

GC AU CC U A

GA

C

A

C

C

G

C

U

U

C

A

C

G

C C G G G

G G C C CU

U

C

U

A

CA

A

G

G

A

G

U

G

U

U U G G

UC

CG A C C

G

G AC

C

G

G

C

G

G

C

A

C

C

GC AU CC U A

GA

C

C C G G G

G G C C C

A

C

C

G

C

U

U

C

A

C

G

U

U

C

U

A

CG

A

G

39984003

9984003998400

3998400

3998400

3998400

5998400

5998400

5998400 5

998400 5998400

5998400

Desulfitobacterium hafniense Methanosarcina barkeri

ldquoCa Methanomethylophilus alvusrdquo ldquoCa Methanomassiliicoccus intestinalisrdquo Methanomassiliicoccus luminyensis

tRNAPyl no 1 tRNAPyl no 2

TΨC loop

TΨC stem

D-stem

D-loop

Anticodon stem

Anticodon loop

Variable loop

Amino acid acceptor stem

Amino acid attachment site

G

CA

G

C

U

C

Figure 2 Secondary structure of the tRNAPyl in the Thermoplasmata-related 7th order The stem-loop structure of the tRNAPyl in the 7thmethanogenic order is shown in comparison with the structure in bacteria (Desulfitobacterium hafniense left) and in Methanosarcinaceae(Methanosarcina barkeri right) The name of each region of the tRNA is indicated The anticodon CUA (corresponding to the amber codon)is outlined in blue For the two tRNAPyl in M luminyensis the modified bases in no 2 compared to no 1 are written in red Adapted andupdated from [11 13]

is the anticodon stem which is broken in all the tRNAPyl

observed in the 7th order of methanogensThis forms a smallloop with a different shape in ldquoCa M alvusrdquo and in ldquoCa MintestinalisrdquoM luminyensis (Figure 2)

33 pyl-tRNA Synthetase The pylS gene encoding the Pyl-tRNA synthetase PylRS is a class II-aaRS (subclass IIc) [21]The four homologues present in the three genomes areshortened in their 51015840 end compared to their Methanosarci-nales counterparts (usually around 420ndash460AA) and encoderespectively 275 amino acids in ldquoCa M alvusrdquo ldquoCa Mintestinalisrdquo and for one of the two PylRS ofM luminyensisand 271 amino acids (or possibly 308 due to uncertainty ofthe start codon) for the second This should lead to an N-tertruncated protein with sim140 residues less similar to the PylScproteins that are present in bacteriaWe could not identify inany of the three complete genomes a gene encoding a protein

similar to these sim140 N-terminal residues present in theMethanosarcinaceae PylS or the bacterial PylSn Moreoverwe found no homologue in these genomes of the particulardomain TIGR03912 that is present in archaeal PylS andbacterial PylSn proteins and that enhances the interactionof the tRNA synthetase to its specific tRNA [55] Theseparticular features of the PylRS may be linked to the peculiarcharacteristics of the tRNAPyl and suggest a different kind ofinteraction

34 Phylogenetic Analysis Phylogenetic analysis was carriedout with each of the Pyl gene products by recoveringall homologues available in current sequence databases Itappears that bacterial members belong to related Firmi-cutes species and a single deltaproteobacterium Bilophilawadsworthia a common resident of the gut For PylCPylD and PylS no evident closely related homologues

Archaea 5

Therm

incola potens

Methanolobus psychrophilus

87099

1001

1001

02

Acetohalobium arabaticum

Thermoacetogenium phaeum

Bilophila wadsworthia

Desulfotomaculum

gibsoniae

Desulfotom

aculum acetoxidans

Des

ulfo

spor

osin

us or

ientis

Desu

lfosp

oros

inus

mer

idiei

Desulfo

sporos

inus y

oung

iae

Desulfitobacter

ium dehalogenans

Desulfitobacterium hafniense

Methanosarcina acetivorans

Methanosalsum zhilinae

Methanosarcina mazei

Meth

anosa

rcina

barke

ri

Meth

anoh

aloph

ilus m

ahii

Meth

anoc

occo

ides

burto

nii

Methanom

ethylovorans hollandica

Methanohalobium evestigatum

Methanomassiliicoccus luminyensis

ldquoCa Methanomassiliicoccus intestinalisrdquo

ldquoCa Methanomethylophilus alvusrdquo

Figure 3 Phylogenetic trees of the pylSCD geneproducts The phylogenetical analysis (PhyML) was performed from a concatenation (672positions) See Sections 2 and 3 for more details Individual trees are also reported in the additional file (supplementary Figures S1 to S3 withindication of the corresponding accession numbers) The 7th methanogenic order members are indicated in red the Methanosarcinaceae inblue and the bacteria in black

could be found outside of the known Pyl-containingorganisms whereas PylB had a few distant homologuesin various bacteria (not shown) Pyl proteins are verywell conserved at the sequence level among Archaea andBacteria (alignments are available on request from theauthors) The PylC PylD and PylS datasets gave consistentresults (additional files Figures S1 to S3 available online athttpdxdoiorg1011552014374146) andwere therefore con-catenated to provide increased phylogenetic signal (Figure 3)ldquoCa M alvusrdquo ldquoCa M intestinalisrdquo and ldquoM luminyensisrdquoform a robustly supported monophyletic cluster indicating acommon origin of the Pyl system in the 7th methanogenicorder The products of the two pyl gene clusters of

M luminyensis (including the third extra PylC) are moreclosely related among them than to the other sequencesindicating that they originated from a specific duplicationMethanosarcinaceae and bacterial sequences also form twodistinct and robustly supportedmonophyletic clustersMore-over the sequences from the 7th methanogenic order appearto be more closely related to their bacterial counterparts thanto Methanosarcinaceae (Figure 3)

Concerning PylB phylogenetic analyses gave moreambiguous results with a poorly supported branching ofthe 7th methanogenic order within Bacteria and differentinferred evolutionary relationships according to exclusion orinclusion of the distant bacterial outgroup (data not shown)

6 Archaea

An unrooted PhyML phylogenetic tree inferred exclusivelyfrom PylB (without the distant bacterial outgroup) is givenin the additional file Figure S4

4 Discussion

We have shown here that all the genetic components for Pylsynthesis and use are present in a new order of Archaeaperforming methylotrophic methanogenesis from methanoland methylamines therefore enlarging the taxonomic dis-tribution of this genetic code expansion cassette Moreoverthis Pyl system has unique features and appears moreclosely related to the bacterial one than to those presentin Methanosarcinaceae Whether or not the Pyl system isactive in these archaeons is beyond the scope of this paperHowever there are many arguments supporting a functionalsystem These include for the three genomes available andanalyzed to date at least the global presence of all the genesfor Pyl synthesis charging and cotranslational incorpora-tion in cooccurrence with all the genes for methylotrophicmethanogenesis frommethylamines that is those coding formethylamines-corrinoid protein methyltransferases (MT)genes (mtmB mtbB andmttB) all bearing an in-frame ambercodon It is interesting to note that as a nonsense (stop)codon the amber codon is used with a low frequency thisvalue being maximal in the M luminyensis genome withonly 11 of stop codon being an amber one It is likely anadaptive strategy to the presence of the Pyl system in thisorder similarly to what is observed in Methanosarcinaceaebut different from Bacteria these lasts deal with the presenceof Pyl through regulation rather than codon avoidance [25]Ability to metabolize TMA into methane has been shownfor M luminyensis [37] Moreover it has also been shownthat methylotrophic methanogenesis from methylamines isactive in the rumen very likely carried out by Rumen ClusterC members of the 7th order of methanogens which areclose neighbors or similar to ldquoCa M alvusrdquo [38] This istherefore conceivable that methylotrophic methanogenesis isconstitutive of this whole order by dedicated MTs containingPylThe presence of unique features in the Pyl system in thesearchaeons raises the exciting question of how it functionsPylRS harbors a shorter form more closely related to theproduct of the bacterial pylSc gene than to the C-terminusof the PylS present in Methanosarcinaceae and lacks thesim140 amino acids equivalent to the bacterial pylSn or to theN-terminal of the archaeal PylS Bacterial PylSn and theNter part of Methanosarcinaceae PylRS contain the proteindomain TIGR03912 [55] In the three genomes of the 7thorder of methanogens no CDS containing this peculiardomain was found by an InterProScan analysis [56] Thelack of these elements in the tRNAPyl synthetase (PylSn-TIGR03912 domain) indicate that PylRS must solely relyon a homomeric structure and might be able to fulfill thisrole without the domain corresponding to PylSn Indeedit has been shown that the D hafniense PylSc is alonesufficient to charge a cognate tRNAPyl with an analogue ofpyrrolysine in a recombinant system [20] Alternatively itmay rely on the presence of a yet unknown enhancing

component Also exciting are the sequence and inferredstructure of the tRNAPyl its extremely condensed naturewith a peculiar broken anticodon stem is of concern forfuture functional and evolutionary studies together with themolecular mechanisms which sustain this orthogonal codebetween the tRNAPyl and the PylRSThe peculiar structure ofthe tRNAPyl may strengthen the interaction with its cognatePylRS and be an adaptation following the loss of the PylSndomain Alternatively mutations in the sequence of thetRNAPyl may have led to the loss of the PylSn However itis also possible that this domain was never present in thePyl system of these archaeons In any case active or notthe well-supported monophyly of the Pyl encoding genes inthe three genomes argue for their presence in their commonancestor Because these represent at least two distinct clades[32] this indicates the likely presence of a Pyl system earlyin the 7th order of methanogens or even in the ancestor ofthe whole order The same conclusion stands also for theMethanosarcinaceae

How the 22nd proteogenic amino acid Pyrrolysine wasadded to the genetic code is an exciting question Based onexperimental data it has been proposed that the universalcodon catalog was anterior to the aminoacylation systems[57] 3D structure-based phylogeny of PylRS has led to thesuggestion that it was already present in the last universalcommon ancestor (LUCA) of all present-day life beingsand has a common origin with the phenylalanine tRNAsynthetase PheRS [21] If true this would mean that Pylwas inherited from the LUCA and retained only in the fewpresent-day methylamines-utilizing bacterial and archaealspecies Similarly Fournier and colleagues proposed thatPyl originated in a pre-LUCA lineage representing a yetunidentified or more probably extinct 4th domain of life [58ndash60] In this hypothesis Pyl would have been acquired ina few Bacteria and Archaea via several independent lateralgene transfers (LGT) from members of this hypothetical 4thdomain of life However its highly specific function andrestricted distribution make it more likely that the additionof Pyl is as a more recent event

The strong link of the Pyl system with methanogenesisespecially the methylotrophic one argues for a commonevolutionary history Pyl is now essential for methylotrophicmethanogenesis from methylamines while methylotrophicmethanogenesis frommethanol is independent of Pyl Pyl hasbeen found inArchaea almost exclusively in the methyltrans-ferases involved in methanogenesis from mono- di- andtri-methylamines (MtmB MtbB and MttB) Moreover pylgenes and mtm-mtb-mtt genes are clustered in the genomesof methylamines-utilizing archaeal species therefore a stim-ulating hypothesis is that genes necessary for Pyl synthe-sisincorporation and for methylotrophic methanogenesisfrom methylamines have coevolved whereas this was notthe case for methylotrophic methanogenesis from methanolwhich is independent of pyrrolysine-containing enzymes

Methanogenesis is believed to have arisen early inthe evolution of Euryarchaeota likely after the divergenceof Thermococcales and to have been lost several timesindependently in various lineages (eg Halobacteriales

Archaea 7

Archaeoglobales and Thermoplasmatales) [32 61 62] Thefirst form of methanogenesis might have been the hydro-genotrophic one (present in the class I methanogens com-posed of Methanopyrales Methanobacteriales and Metha-nococcales) while other types of methanogenesis (methy-lotrophic non-H

2-dependent and acetoclastic involving

cytochromes) would have emerged later Based on phyloge-nomic studies we have recently shown that the pathwayfor methylotrophic methanogenesis at least from methanolwas already present in the ancestor of Methanobacterialesin the ancestor of the 7th order and in the ancestor ofMethanosarcinales which is the common ancestor of mosteuryarchaeal lineages excludingThermococcales [32]There-fore H

2-dependent methylotrophic methanogenesis from

methanolmay also be an ancient type ofmethanogenesisThefact that it does not involve cytochromes and needs fewergenes could favor such hypothesis The question is can thesame conclusions be made for the methylotrophic methano-genesis frommethylaminesThe fact that this methanogene-sis is presently restricted to Methanosarcinaceae and the 7thorder of methanogens which emerged after class I metha-nogens argues for a more recent event It has also to bestressed that no mta gene (involved in methylotrophicmethanogenesis from methanol) harbors an amber-Pylcodon Moreover the last MT enzyme of the methanogenesispathway (encoded bymtaA gene for methanol) that precedesthe steps corresponding to core methanogenesis (methyl-coM reduction to generate methane) is involved in methano-genesis from both methanol and some methylamines inMethanosarcinaceae and likely also in the 7th order (data notshown) Therefore it is tempting to speculate that the Pyl-independent H

2-dependent methanogenesis frommethanol

preceded the Pyl-dependent H2-dependent methylotrophic

methanogenesis from methylaminesTaking together these pieces of evidence it is likely that in

the archaeal domain the Pyl system arose in a methanogeniceuryarchaeote in correspondence to the emergence of thegenes coding for methylamines-corrinoid protein methyl-transferases (MT) (mtmB mtbB and mttB) that all bear anin-frame amber codon allowing the use of new substrates(mono- di- and trimethylamine resp) We will refer to thearchaeon where the system arose as the archaeal Pyrrolysineancestor (APA) The APA may correspond to the ancestorof Methanosarcinaceae the ancestor of the 7th order ofmethanogens or the common ancestor of both (depicted onFigure 4) Now it can be asked how the Pyl system arosein this hypothetical euryarchaeote Two possibilities can beenvisaged (i) it was acquired via a lateral gene transfer (LGT)or (ii) it emerged autogenously

Concerning the hypothesis of an LGT from a 4th domainof life by Fournier and colleagues [58 59] in the light ofour results it may be reformulated by positing a single LGTfrom this extinct lineage into Archaea after the divergenceof Thermococcales to give birth to the APA (Figure 4 dottedgreen arrow)The systemwould have then been retained onlyin present-day pyl-containing organisms for example the7th order and Methanosarcinaceae whereas it would havebeen lost multiple times independently in most euryarchaeallineages including methanogenic ones (Figure 4 red dots)

However the hypothetical existence of a pre-LUCA lineagethat gave rise to the Pyl system remains questionable becauseof the necessary coexistence of the donor and the APA andbecause the LGT would have taken place in a euryarchaeonrelatively recent and distant from LUCA the 4th domainwould have been composed at that time ofmany different lin-eages that would have subsequently all disappeared (brokendotted lines in the hypothetical 4th Domain Figure 4)

Our data also make the hypothesis that the Pyl systemarose in Bacteria and was introduced in the APA via LGTless likely (Figure 4) In fact its presence in the 7th order nowmakes its taxonomic distributionmuch larger inArchaeathanthe few pyl-containing bacteria representing a sublineageof Firmicutes and two deltaproteobacteria and its link tomethylotrophic methanogenesis stronger

Therefore our preferred hypothesis is that the Pyl sys-tem is an archaeal invention (green box Figure 4) PylRSmight have emerged by gene duplication followed by fastevolutionary rates from another class II RS gene whilea mutation in the anticodon of a duplicated tRNA couldhave led to an amber decoding tRNA This would havehad less detrimental effects for the cell than the recodingof another codon affecting potentially all the proteins ofthe cell Interestingly PylRS does not need to recognize theanticodon of the tRNAPyl [17 55] and this may be a remnantof the birth of the orthogonal pair tRNAPylPylRS Moreoverthe recent discovery of the whole synthesis pathway of Pylwith PylB being a lysine mutase [13] shows that Pyl isentirely a derivative of a proteogenic amino acid (two lysines)and this could make sense in the light of the coevolutiontheory such are the cases of AspAsn and GluGln inArchaea[63] APA may have been an early diverging eur-yarchaeote performing H

2-dependent methanogenesis from

methanol perhaps the ancestor ofMethanosarcinales and the7th order of methanogens and therefore after the emergenceof Thermococcales and methanogens class I (Methanococ-cales Methanobacteriales and Methanopyrales) Verticalinheritance of the Pyl system would have paralleled that ofH2-dependent methanogenesis from methylamines whereas

multiple independent losses would have occurred over timein most euryarchaeal lineages including methanogenic ones(Figure 4 red dots) but retained only in the Methanosarci-naceae and the 7th order Alternatively the system wouldhave emerged later in either the ancestor of Methanosarci-naceae or in the ancestor of the 7th order and then spreadamong them via LGT (not indicated in the Figure 4 forclarity) Because their Pyl systems are different (notably theabsence of the N-terminus of PylS and the unique peculiarityof the tRNAPyl) it may be asked which on the two is theancestral one However it is likely that the unique system ofthe 7th order is derived from a more ldquoclassicalrdquo one such asthose present in bacteria and Methanosarcinaceae

Under the hypothesis of an archaeal invention of thePyl system the bacterial one may have arisen via LGTfrom the 7th order of methanogens considering their closerevolutionary relations before the loss of the N-terminusof PylS (Figure 4 PylS structure indicated above the Pyl-containing groups) It is unclear if one or several LGTs

8 Archaea

Methanosarcinaceae

Thermoplasmatales

HypotheticalPre-LUCA 4th domain

Pyl system invention

LGT from

BacteriaHypothetical 4th domain

APA

Arc

haea

LUCA

Methanocellales

Halobacteriales

Methanomicrobiales

Thermococcales

Methanosaetaceae

Archaeoglobales

Methanococcales

Methanobacteriales Methanopyrales

Interdomain LGT(s)

PylS (c)

Some firmicutes 120575 proteobacteria (2)PylScPylSnPylScPylSn

Intradomain (Bacteria) LGT(s)

PylSc

PylSn2

33

1

7th order

Bacteria

Eukarya

PylS (n + c) fused

Figure 4 Proposed scenarios for the origin and evolution of the pyrrolysine system One of the various possible models leading to the genesisof APA the ancestral Pyl-coding archaeon is schematized This model supports less LGTs than previously proposed ones APA (blue star)emerged in the Euryarchaeota (black circle no 1) after the birth of the methanogenesis and was an ancestor of the Methanosarcinalesandorof the Thermoplasmata-related 7th order of methanogens Only the hypothesis of a common ancestor of both MethanosarcinalesandThermoplasmata-related 7th order of methanogens is depicted here (see text for an alternative model with APA as a more recent ancestor ofthe Methanosarcinalesor of theThermoplasmata-related 7th order of methanogens but not both)The putative genesis of APA (green boxes)relies either from an archaeal invention (Box 1) from a bacterial contribution or from awhole functional system acquired from a hypotheticalpre-LUCA4th domain of life APAwas therefore likely amethanogenic archaeon performing both amethanol-H

2-dependentmethylotrophic

methanogenesis and a hydrogenotrophic oneThe Pyl system was vertically inherited (symbolized by large blue lines) toMethanosarcinaceaeand to the 7th methanogenic order and was lost several times independently on various branches including methanogenic ones (red dots)Due to the closest relationship of the Pyl system (orange square) between the 7th order and bacteria at least one LGT is supposed to haveoccurred from a 7th order ancestor (before the loss of the pylSc gene in this lineage) likely to a Firmicute(no 2 orange arrow) Other LGTsfrom the 7th order to Bacteria (ie a deltaproteobacterium) are also conceivable In a third period intradomain LGT(s) could have originatednext putatively among Firmicute andor into a deltaproteobacterium (no 3 red arrows) Pyl-coding organisms are symbolized in blue boxesThe nature of the PylRS is depicted with each Pyl-containing group in grey boxes and is either formed by a unique PylS (Methanosarcinaceae)a split complete version (PylScPylSn in bacteria) or a unique PylSc form (7th methanogenic order)

occurred from the 7th order onto bacteria (Firmicutes anddeltaproteobacteria) The group of Firmicutes is likely therecipient of this first transfer and would have then giventhe system to a deltaproteobacterium Alternatively due totheir close environmental relationship (the gut) a distinctLGT could have arisen from a 7th order member of thegut into Bilophila sp However it cannot be excluded thatbacteria took it from Methanosarcinaceae and split the PylSgene into PylSn and PylSc and then they later passed onlyPylSc to the 7th order The tree of PylB may suggest suchhypothesis albeit statistical support is not very strong and

no similar pattern is observed for the other componentsof the system Unfortunately phylogenetic analysis preventsconcluding the direction of these transfers because of theabsence of outgroup sequences In every case that theseinter- andor intradomain LGTs were not rejected is afascinating event Introducing a stop codon suppressor isin fact likely deleterious and has to be compensated by astrong selective advantage for the organism such as the use ofnew metabolic substrates (eg methylamines) Introducinga Pyl-coding cassette into a heterologous genome has beensuccessfully experimentally realized [14 64] It is possible that

Archaea 9

the acquired genes remained silent or with a low expressionlevel and would have been activated progressively leadingconcomitantly to the negative selection of UAG nonsensecodons A more likely hypothesis is an LGT leading toan inducible expression of the Pyl components such asdependence of the presence of substrates like methylaminesThe natural expansion of the genetic code in the FirmicuteAcetohalobium arabaticum able to genetically encode the 20usual amino acids when grown on pyruvate and to expand itsrepertoire to 21 by adding pyrrolysine when grown on TMA[25] provides the paradigm

In conclusion when considering at least the archaealdomain there has been a Pyl-coding ancestor It appearedlikely relatively recently as an ancestor of the Methan-osarcinales an ancestor of the 7th order of methanogens orthe common ancestor of both It was likely a methanogenperforming a methanolH

2-dependent methanogenesis

and considering the probable coevolution history betweenmethylotrophic methanogenesis and Pyl with their stronginterdependence it was concomitant or rapidly followedby the emergence of methanogenesis from methylaminescompounds This has led nowadays in the archaeal lineageto conservation of a Pyl system only in methylamines-utilizingPyl-dependent methanogens Therefore thisprovides an example that the genetic code may be still underevolution with a conceivable expansion shaped by metabolicrequirements

Conflict of Interests

The authors declare that there is no conflict of interestsregarding the publication of this paper

Acknowledgments

This work was funded by PhD Scholarship supports fromthe French ldquoMinistere de lrsquoEnseignement Superieur et dela Rechercherdquo to Nadia Gaci Simonetta Gribaldo was sup-ported by the French Agence Nationale de la Recherche(ANR) through Grant ANR-10-BINF-01-01 ldquoAncestromerdquoPaul W OrsquoToole was supported by Science FoundationIreland through a Principal Investigator award and by anFHRI award to the ELDERMET project by the Departmentof Agriculture Fisheries and Marine of the Government ofIreland

References

[1] A P Lothrop M P Torres and S M Fuchs ldquoDecipheringpost-translational modification codesrdquo FEBS Letters vol 587pp 1247ndash1257 2013

[2] F Wold ldquoIn vivo chemical modification of proteins (post-translational modification)rdquo Annual Review of Biochemistryvol 50 pp 783ndash814 1981

[3] A Bock and T C Stadtman ldquoSelenocysteine a highly specificcomponent of certain enzymes is incorporated by a UGA-directed co-translational mechanismrdquo BioFactors vol 1 no 3pp 245ndash250 1988

[4] T C Stadtman ldquoSelenocysteinerdquo Annual Review of Biochem-istry vol 65 pp 83ndash100 1996

[5] M Ibba and D Soll ldquoGenetic code introducing pyrrolysinerdquoCurrent Biology vol 12 no 13 pp R464ndashR466 2002

[6] G Srinivasan C M James and J A Krzycki ldquoPyrrolysineencoded by UAG in archaea charging of a UAG-decodingspecialized tRNArdquo Science vol 296 no 5572 pp 1459ndash14622002

[7] A Bock K Forchhammer J Heider et al ldquoSelenocysteine the21st amino acidrdquo Molecular Microbiology vol 5 no 3 pp 515ndash520 1991

[8] Y Zhang H Romero G Salinas and V N GladyshevldquoDynamic evolution of selenocysteine utilization in bacteria abalance between selenoprotein loss and evolution of selenocys-teine from redox active cysteine residuesrdquo Genome Biology vol7 no 10 article R94 2006

[9] M Rother A Resch R Wilting and A Bock ldquoSelenoproteinsynthesis in archaeardquoBioFactors vol 14 no 1ndash4 pp 75ndash83 2001

[10] T Stock and M Rother ldquoSelenoproteins in Archaea and Gram-positive bacteriardquo Biochimica et Biophysica Acta vol 1790 no11 pp 1520ndash1532 2009

[11] M A Gaston R Jiang and J A Krzycki ldquoFunctional contextbiosynthesis and genetic encoding of pyrrolysinerdquo CurrentOpinion in Microbiology vol 14 no 3 pp 342ndash349 2011

[12] M Rother and J A Krzycki ldquoSelenocysteine pyrrolysineand the unique energy metabolism of methanogenic archaeardquoArchaea vol 2010 Article ID 453642 14 pages 2010

[13] M A Gaston L Zhang K B Green-Church and J A KrzyckildquoThe complete biosynthesis of the genetically encoded aminoacid pyrrolysine from lysinerdquoNature vol 471 no 7340 pp 647ndash650 2011

[14] D G Longstaff R C Larue J E Faust et al ldquoA natural geneticcode expansion cassette enables transmissible biosynthesis andgenetic encoding of pyrrolysinerdquo Proceedings of the NationalAcademy of Sciences of the United States of America vol 104 no3 pp 1021ndash1026 2007

[15] J A Krzycki ldquoThe path of lysine to pyrrolysinerdquo CurrentOpinion in Chemical Biology vol 17 no 4 pp 619ndash625 2013

[16] F Quitterer P Beck A Bacher and M Groll ldquoStructure andreaction mechanism of pyrrolysine synthase (PylD)rdquo Ange-wandte Chemie International Edition vol 52 no 27 pp 7033ndash7037 2013

[17] K Nozawa P OrsquoDonoghue S Gundllapalli et al ldquoPyrrolysyl-tRNA synthetase-tRNAPyl structure reveals themolecular basisof orthogonalityrdquoNature vol 457 no 7233 pp 1163ndash1167 2009

[18] C Polycarpo A Ambrogelly A Berube et al ldquoAn aminoacyl-tRNA synthetase that specifically activates pyrrolysinerdquo Pro-ceedings of the National Academy of Sciences of the United Statesof America vol 101 no 34 pp 12450ndash12454 2004

[19] A Theobald-Dietrich M Frugier R Giege and J Rudinger-Thirion ldquoAtypical archaeal tRNApyrrolysine transcript behavestowards EF-Tu as a typical elongator tRNArdquo Nucleic AcidsResearch vol 32 no 3 pp 1091ndash1096 2004

[20] S Herring A Ambrogelly C R Polycarpo and D SollldquoRecognition of pyrrolysine tRNA by the Desulfitobacteriumhafniense pyrrolysyl-tRNA synthetaserdquo Nucleic Acids Researchvol 35 no 4 pp 1270ndash1278 2007

[21] J M Kavran S Gundllapalli P OrsquoDonoghue M Englert DSoll and T A Steitz ldquoStructure of pyrrolysyl-tRNA synthetasean archaeal enzyme for genetic code innovationrdquo Proceedings ofthe National Academy of Sciences of the United States of Americavol 104 no 27 pp 11268ndash11273 2007

10 Archaea

[22] M M Lee R Jiang R Jain R C Larue J Krzycki and MK Chan ldquoStructure of Desulfitobacterium hafniense PylSca pyrrolysyl-tRNA synthetaserdquo Biochemical and BiophysicalResearch Communications vol 374 no 3 pp 470ndash474 2008

[23] T Yanagisawa R Ishii R Fukunaga T Kobayashi K Sakamotoand S Yokoyama ldquoCrystallographic studies on multiple con-formational states of active-site loops in pyrrolysyl-tRNA syn-thetaserdquo Journal of Molecular Biology vol 378 no 3 pp 634ndash652 2008

[24] T Yanagisawa T Sumida R Ishii and S Yokoyama ldquoA novelcrystal form of pyrrolysyl-tRNA synthetase reveals the pre- andpost-aminoacyl-tRNA synthesis conformational states of theadenylate and aminoacyl moieties and an asparagine residuein the catalytic siterdquo Acta Crystallographica D vol 69 pp 5ndash152013

[25] L Prat I U Heinemann H R Aerni J Rinehart POrsquoDonoghue and D Soll ldquoCarbon source-dependent expan-sion of the genetic code in bacteriardquo Proceedings of the NationalAcademy of Sciences vol 109 pp 21070ndash21075 2012

[26] L Paul D J Ferguson Jr and J A Krzycki ldquoThe trimethy-lamine methyltransferase gene and multiple dimethylaminemethyltransferase genes of Methanosarcina barkeri contain in-frame and read- through amber codonsrdquo Journal of Bacteriol-ogy vol 182 no 9 pp 2520ndash2529 2000

[27] J A Soares L Zhang R L Pitsch et al ldquoThe residue mass of L-pyrrolysine in three distinct methylamine methyltransferasesrdquoThe Journal of Biological Chemistry vol 280 no 44 pp 36962ndash36969 2005

[28] A Mihajlovski M Alric and J-F Brugere ldquoA putative neworder of methanogenic Archaea inhabiting the human gut asrevealed by molecular analyses of the mcrA generdquo Research inMicrobiology vol 159 no 7-8 pp 516ndash521 2008

[29] A Mihajlovski J Dore F Levenez M Alric and J-F BrugereldquoMolecular evaluation of the human gutmethanogenic archaealmicrobiota reveals an age-associated increase of the diversityrdquoEnvironmental Microbiology Reports vol 2 no 2 pp 272ndash2802010

[30] T Iino H Tamaki S Tamazawa et al ldquoCandidatus Metha-nogranum caenicola a novel methanogen from the anaer-obic digested sludge and proposal of Methanomassiliicoc-caceae fam nov and Methanomassiliicoccales ord nov for amethanogenic lineage of the class Thermoplasmatardquo Microbesand Environments vol 28 pp 244ndash250 2013

[31] K Paul J O Nonoh L Mikulski and A Brune ldquolsquoMethanoplas-matalesrsquo Thermoplasmatales-related archaea in termite gutsand other environments are the seventh order ofmethanogensrdquoApplied andEnvironmentalMicrobiology vol 78 pp 8245ndash82532012

[32] G Borrel P W OrsquoToole P Peyret J F Brugere and SGribaldo ldquoPhylogenomic data support a seventh order ofmethylotrophic methanogens and provide insights into theevolution of methanogenesisrdquo Genome Biology and Evolutionvol 5 pp 1769ndash1780 2013

[33] G Borrel H M Harris W Tottey et al ldquoGenome sequenceof ldquoCandidatus Methanomethylophilus alvusrdquo Mx1201 amethanogenic archaeon from the human gut belonging to aseventh order of methanogensrdquo Journal of Bacteriology vol194 pp 6944ndash6945 2012

[34] A Gorlas C Robert G Gimenez M Drancourt and DRaoult ldquoComplete genome sequence ofMethanomassiliicoccusluminyensis the largest genome of a human-associatedArchaeaspeciesrdquo Journal of Bacteriology vol 194 no 17 p 4745 2012

[35] B Dridi M L Fardeau B Ollivier D Raoult and M Dran-court ldquoMethanomassiliicoccus luminyensis gen nov sp novamethanogenic archaeon isolated from human faecesrdquo Interna-tional Journal of Systematic and Evolutionary Microbiology vol62 pp 1902ndash1907 2012

[36] G Borrel H M Harris N Parisot et al ldquoGenome sequence ofldquoCandidatusmethanomassiliicoccus intestinalisrdquo Issoire-Mx1 athird Thermoplasmatales-related methanogenic archaeon fromhuman fecesrdquo Genome Announcements vol 1 no 4 Article IDe00453 13 pages 2013

[37] J F Brugere G Borrel N Gaci W Tottey P W O rsquoToole andCMalpuech-Brugere ldquoArchaebiotics proposed therapeutic useof archaea to prevent trimethylaminuria and cardiovasculardiseaserdquo Gut Microbes vol 5 no 1 2013

[38] M Poulsen C Schwab B B Jensen et al ldquoMethylotrophicmethanogenicThermoplasmata implicated in reducedmethaneemissions from bovine rumenrdquoNature Communications vol 4article 1428 2013

[39] R K Aziz D Bartels A Best et al ldquoThe RAST server rapidannotations using subsystems technologyrdquo BMCGenomics vol9 article 75 2008

[40] K Rutherford J Parkhill J Crook et al ldquoArtemis sequencevisualization and annotationrdquo Bioinformatics vol 16 no 10 pp944ndash945 2000

[41] S F AltschulW GishWMiller EWMyers and D J LipmanldquoBasic local alignment search toolrdquo Journal ofMolecular Biologyvol 215 no 3 pp 403ndash410 1990

[42] M A Larkin G Blackshields N P Brown et al ldquoClustalW andclustal X version 20rdquo Bioinformatics vol 23 no 21 pp 2947ndash2948 2007

[43] C Notredame D G Higgins and J Heringa ldquoT-coffee a novelmethod for fast and accurate multiple sequence alignmentrdquoJournal of Molecular Biology vol 302 no 1 pp 205ndash217 2000

[44] R C Edgar ldquoMUSCLE multiple sequence alignment with highaccuracy and high throughputrdquo Nucleic Acids Research vol 32no 5 pp 1792ndash1797 2004

[45] A Wilm D G Higgins and C Notredame ldquoR-Coffee amethod for multiple alignment of non-coding RNArdquo NucleicAcids Research vol 36 no 9 article e52 2008

[46] A R Gruber R Lorenz S H Bernhart R Neubock and I LHofacker ldquoThe Vienna RNA websuiterdquo Nucleic Acids Researchvol 36 pp W70ndashW74 2008

[47] J S Reuter and D H Mathews ldquoRNAstructure softwarefor RNA secondary structure prediction and analysisrdquo BMCBioinformatics vol 11 article 129 2010

[48] A Criscuolo and S Gribaldo ldquoBMGE (Block Mapping andGathering with Entropy) a new software for selection ofphylogenetic informative regions frommultiple sequence align-mentsrdquo BMC Evolutionary Biology vol 10 no 1 article 2102010

[49] SGuindon J-FDufayardV LefortMAnisimovaWHordijkand O Gascuel ldquoNew algorithms and methods to estimatemaximum-likelihood phylogenies assessing the performanceof PhyML 30rdquo Systematic Biology vol 59 no 3 pp 307ndash3212010

[50] S Q Le and O Gascuel ldquoAn improved general amino acidreplacement matrixrdquo Molecular Biology and Evolution vol 25no 7 pp 1307ndash1320 2008

[51] G Jobb A von Haeseler and K Strimmer ldquoTREEFINDER apowerful graphical analysis environment for molecular phylo-geneticsrdquo BMC Evolutionary Biology vol 4 article 18 2004

Archaea 11

[52] N Lartillot T Lepage and S Blanquart ldquoPhyloBayes 3 aBayesian software package for phylogenetic reconstruction andmolecular datingrdquo Bioinformatics vol 25 no 17 pp 2286ndash22882009

[53] J Yuan P OrsquoDonoghue A Ambrogelly et al ldquoDistinct geneticcode expansion strategies for selenocysteine and pyrrolysineare reflected in different aminoacyl-tRNA formation systemsrdquoFEBS Letters vol 584 no 2 pp 342ndash349 2010

[54] I L Hofacker ldquoRNA secondary structure analysis using theVienna RNA packagerdquo Current Protocols in Bioinformatics2004 Chapter 12 Unit 122

[55] R Jiang and J A Krzycki ldquoPylSn and the homologous N-terminal domain of pyrrolysyl-tRNA synthetase bind the tRNAthat is essential for the genetic encoding of pyrrolysinerdquo TheJournal of Biological Chemistry vol 287 pp 32738ndash32746 2012

[56] E M Zdobnov and R Apweiler ldquoInterProScanmdashan integrationplatform for the signature-recognition methods in InterPrordquoBioinformatics vol 17 no 9 pp 847ndash848 2001

[57] M J HohnH-S Park P OrsquoDonoghueM Schnitzbauer andDSoll ldquoEmergence of the universal genetic code imprinted in anRNA recordrdquo Proceedings of the National Academy of Sciences ofthe United States of America vol 103 no 48 pp 18095ndash181002006

[58] G Fournier ldquoHorizontal gene transfer and the evolution ofmethanogenic pathwaysrdquo Methods in Molecular Biology vol532 pp 163ndash179 2009

[59] G P Fournier J Huang and J PeterGogarten ldquoHorizontal genetransfer from extinct and extant lineages biological innovationand the coral of liferdquo Philosophical Transactions of the RoyalSociety B vol 364 no 1527 pp 2229ndash2239 2009

[60] A V Lobanov A A Turanov D L Hatfield and V NGladyshev ldquoDual functions of codons in the genetic coderdquoCritical Reviews in Biochemistry and Molecular Biology vol 45no 4 pp 257ndash265 2010

[61] E Bapteste C Brochier and Y Boucher ldquoHigher-level clas-sification of the Archaea evolution of methanogenesis andmethanogensrdquo Archaea vol 1 no 5 pp 353ndash363 2005

[62] S Gribaldo and C Brochier-Armanet ldquoThe origin and evolu-tion of Archaea a state of the artrdquo Philosophical Transactions ofthe Royal Society B vol 361 no 1470 pp 1007ndash1022 2006

[63] J T-FWong ldquoCoevolutlon theory of genetic code at age thirtyrdquoBioEssays vol 27 no 4 pp 416ndash425 2005

[64] D G Longstaff S K Blight L Zhang K B Green-Churchand J A Krzycki ldquoIn vivo contextual requirements for UAGtranslation as pyrrolysinerdquoMolecular Microbiology vol 63 no1 pp 229ndash241 2007

Impact Factor 173028 Days Fast Track Peer ReviewAll Subject Areas of ScienceSubmit at httpwwwtswjcom

Hindawi Publishing Corporation httpwwwhindawicom Volume 2013Hindawi Publishing Corporation httpwwwhindawicom Volume 2013

The Scientific World Journal

2 Archaea

as a free amino acid in the cell by the products of the pylBCDgenes [13 14] from two lysines one being methylated into 3-methylornithine (catalyzed by PylB a lysine mutase-proline-2 methylase) and condensed to the second lysine to form3-methylornithinyl-N6-lysine (PylC Pyrrolysine synthetase)[15] The pyrrole ring is then formed by oxidation with anatypical dehydrogenase the Proline reductase (also calledPyl synthase PylD) [16] with concomitant release of anamino group during the cycle formation The gene productof pylT forms a dedicated tRNA used for the translationalincorporation of Pyl It is an amber decoding tRNACUA [6]showing some unusual features compared to other tRNAs[11 17ndash19] the anticodon stem is longer (6 nucleotides insteadof 5) while other parts are shorter D-loop with only 5 basesacceptor and D-stems separated by one base instead of 2and a variable loop of only 3 bases Moreover differing fromalmost all tRNAs the D-loop does not contain the G

18G19

bases nor the usual T54Ψ55C56

bases in the T-loop Finallythe aaRS encoded by pylS catalyzes the ligation of Pyl to itscognate tRNA species warranting the right correspondencebetween DNA and proteins The anticodon of the tRNAseems not to be recognized by the PylRS at least in Dhafniense [20] Whereas the archaeal PylRS is encoded bya single gene (pylS) the bacterial PylRS is encoded by twogenes pylSn for the firstN-terminal 140 amino acids and pylScfor the remaining sequenceThis pyrrolysyl-tRNA synthetasebelongs to the class II aaRS and its structure has beenresolved with and without its substratesanalogs [17 21ndash24]It has likely a homodimeric or homotetrameric quaternarystructure

To date Pyl-containing proteins have been found only inthe methanogens of the family Methanosarcinaceae and in afew bacteria belonging to Firmicutes and two mem-bers of deltaproteobacteria Bilophila wadsworthia and anendosymbiont of the worm Olavius algarvensis [25] Pyl ispresent almost specifically in the methyltransferases (MT)involved in the methanogenesis pathway frommonomethyl-dimethyl- and trimethylamine in methanogens respectivelyencoded by the genes mtmB mtbB and mttB [26 27] and intheir bacterial homologs whose function is unclearHowevermany bacteria harbor mttB homologs that lack the in-frameamber codon and the Methanosarcinales Methanococcoidesburtonii hasmttB genes with and without an in-frame ambercodon [11 25] In contrast themtbB gene is almost exclusivelypresent inMethanosarcinales and harbors an in-frame ambercodonThemtmB gene represents an intermediary case witha few Pyl-containing bacteria possessing it leading either to aMT with Pyl (Acetohalobium arabaticum Desulfotomaculumacetoxidans) or not (Bilophila wadsworthia) [25]

The existence of a 7th order of methanogens related toThermoplasmatales was proposed from molecular data of16S rRNA and mcrA (methanogenesis marker) sequencesretrieved from human stools [28 29]This is strengthened bygrowing molecular data from various environments [30 31]and by phylogenomics studies [32] established from the firstgenomes of members of this clade [33 34] Moreover a firstmember of this orderMethanomassiliicoccus luminyensis hasbeen isolated and grows onmethanol +H

2[35]We identified

the presence of genes formethanogenesis frommethanol and

also from dimethylsulfide andmethylamines in the two othersequenced genomes (ldquoCandidatus Methanomethylophilusalvusrdquo [33] ldquoCandidatusMethanomassiliicoccus intestinalisrdquo[36]) together with the Pyl-coding genes and in-frame ambercodon in the mtmB mtbB and mttB genes of these threespecies It has now been shown that the H

2-dependent

methanogenesis from trimethylamine is effective for Mluminyensis [37] Also ruminal methanogens of the samelineage as ldquoCa M alvusrdquo (Rumen Cluster C or RCC) whencultured in consortia with trimethylamine (TMA) show anincrease of methanogenesis and mRNAs for MTs with anamber in-frame codon are detected [38] Altogether thisgreatly suggests the presence of Pyl in these MTs

The recent uncovered lineage with all the featuresneeded for Pyl encoding and use representing a newarchaeal methanogenic Order provides an opportunity tobetter understand the origin distribution and diversity ofPyrrolysine-coding systems

2 Materials and Methods

Genomic sequence data were obtained through GenBankFor the 7th order Thermoplasmata-related methano-gens the accession numbers are CP0040491 (ldquoCa Methano-methylophilus alvusrdquo) CP0059341 (ldquoCa Methanomassilii-coccus intestinalisrdquo) and NZ CAJE0100001 to NZ CAJE010-0026 (Methanomassiliicoccus luminyensis) For the genomicorganization and comparison of pyl genes genomic sequen-ces were either treated with the RAST annotation server [39]or a local Artemis platform [40] Blast searches [41] wereeither performed directly on the RAST server on the nrdatabase atNCBI or locally Sequence alignment (DNARNAand proteins) was performed using ClustalW [42] T-coffee[43] and MUSCLE [44] For RNAs the dedicated R-Coffeeprogram was also used [45] accessible at httpwwwtcoffeeorg which generates a multiple sequence align-ment using structural information RNAfold [46] (availa-ble at httprnatbiunivieacatcgi-binRNAfoldcgi) andRNAstructure [47] (available at httprnaurmcroches-teredu) were used to determine the secondary structure oftRNAThe structure of the tRNAPyl was also manually drawnby comparison with the structure of the D hafniense andMacetivorans tRNAPyl [11 20] Phylogenies were both inferredfrom maximum likelihood and Bayesian proceduresDatasets of protein homologues were aligned by MUSCLE[44] with default parameters and unambiguously alignedpositions were automatically selected by using the BMGEsoftware for multiple-alignment trimming [48] with aBLOSUM30 substitution matrix Maximum likelihood treeswere calculated by PhyML [49] and the LG amino acidsubstitution model [50] with 4 rate categories as suggestedby the AIC criterion implemented in Treefinder [51] Treeswere also calculated by Bayesian analysis with PhyloBayes[52] with the LG model (for single gene trees) or the CATmodel (for the concatenated dataset) and 4 categories ofevolutionary rates In this case two MCMC chains were runin parallel until convergence and the consensus tree wascalculated by removing the first 25 of trees as burnin

Archaea 3

T S B C D

T S B C D

T Sc B C D Sn

Sc B C DSnT

T Sc B C D Snkinase

T Sc B C D SnramA


T S BC D

T S B C D

Contig 4T S B C D









ethanolamine




sim06 Mb

sim38 Mb

sim07 Mb

sim15 kb


3 Results




4 Archaea

G

G

G

G

G

G

U

C U A GGG

AA

G A U CA

UAC

A

C

G

G

A

G

U

G

C

U

UC AU AC U A

GA

C

A

C

C

G

C

C

C

C

U

C

A

G C G G G

U G C C CU

U

C

G

A

AA

G

G

G

G

G

A

C

C U G G

G A C CAG

C

G

G

G

U

C

G

A

U

C

C

AC AU AC U A

GA

C

A

C

C

G

C

U

C

U

C

U

G

C G G G G U

G C C C C GC

U

A

CGG

G

C

C

G

G

A

G

G

U G GU

G A C C

C

C

G

G

C

G

G

C

A

C

C

GC AU CC U A

GA

C

C C G G G

G G C C C

C

G

C

U

U

C

C

U

U

C

U

A

CG

A

G

CG

(A)

U

G

G

G

A

A

C

C

C U A GUA

UG

G A U CA

UGA

A

U

G

G

A

U

U

G

C

C

UC AU AC U A

GA

U

A

C

C

G

C

C

U

U

U

C

G

C C G G G

G G C C CU

U

U

U

A

AG

G

G

A

G

U

G

U

C U G GU

CG

G A C CG AC

C

G

G

C

G

G

C

A

C

C

GC AU CC U A

GA

C

A

C

C

G

C

U

U

C

A

C

G

C C G G G

G G C C CU

U

C

U

A

CA

A

G

G

A

G

U

G

U

U U G G

UC

CG A C C

G

G AC

C

G

G

C

G

G

C

A

C

C

GC AU CC U A

GA

C

C C G G G

G G C C C

A

C

C

G

C

U

U

C

A

C

G

U

U

C

U

A

CG

A

G

39984003

9984003998400

3998400

3998400

3998400

5998400

5998400

5998400 5

998400 5998400

5998400




TΨC loop

TΨC stem

D-stem

D-loop

Anticodon stem

Anticodon loop

Variable loop



G

CA

G

C

U

C







Archaea 5

Therm

incola potens


87099

1001

1001

02




Desulfotomaculum

gibsoniae

Desulfotom

aculum acetoxidans

Des

ulfo

spor

osin

us or

ientis

Desu

lfosp

oros

inus

mer

idiei

Desulfo

sporos

inus y

oung

iae

Desulfitobacter

ium dehalogenans





Meth

anosa

rcina

barke

ri

Meth

anoh

aloph

ilus m

ahii

Meth

anoc

occo

ides

burto

nii

Methanom










6 Archaea


4 Discussion






Archaea 7


















8 Archaea

Methanosarcinaceae

Thermoplasmatales



LGT from


APA

Arc

haea

LUCA

Methanocellales

Halobacteriales

Methanomicrobiales

Thermococcales

Methanosaetaceae

Archaeoglobales

Methanococcales


Interdomain LGT(s)

PylS (c)



PylSc

PylSn2

33

1

7th order

Bacteria

Eukarya

PylS (n + c) fused






Archaea 9







Acknowledgments


References






















10 Archaea































Archaea 11

















Archaea 3

T S B C D

T S B C D

T Sc B C D Sn

Sc B C DSnT

T Sc B C D Snkinase

T Sc B C D SnramA


T S BC D

T S B C D

Contig 4T S B C D









ethanolamine




sim06 Mb

sim38 Mb

sim07 Mb

sim15 kb


3 Results




4 Archaea

G

G

G

G

G

G

U

C U A GGG

AA

G A U CA

UAC

A

C

G

G

A

G

U

G

C

U

UC AU AC U A

GA

C

A

C

C

G

C

C

C

C

U

C

A

G C G G G

U G C C CU

U

C

G

A

AA

G

G

G

G

G

A

C

C U G G

G A C CAG

C

G

G

G

U

C

G

A

U

C

C

AC AU AC U A

GA

C

A

C

C

G

C

U

C

U

C

U

G

C G G G G U

G C C C C GC

U

A

CGG

G

C

C

G

G

A

G

G

U G GU

G A C C

C

C

G

G

C

G

G

C

A

C

C

GC AU CC U A

GA

C

C C G G G

G G C C C

C

G

C

U

U

C

C

U

U

C

U

A

CG

A

G

CG

(A)

U

G

G

G

A

A

C

C

C U A GUA

UG

G A U CA

UGA

A

U

G

G

A

U

U

G

C

C

UC AU AC U A

GA

U

A

C

C

G

C

C

U

U

U

C

G

C C G G G

G G C C CU

U

U

U

A

AG

G

G

A

G

U

G

U

C U G GU

CG

G A C CG AC

C

G

G

C

G

G

C

A

C

C

GC AU CC U A

GA

C

A

C

C

G

C

U

U

C

A

C

G

C C G G G

G G C C CU

U

C

U

A

CA

A

G

G

A

G

U

G

U

U U G G

UC

CG A C C

G

G AC

C

G

G

C

G

G

C

A

C

C

GC AU CC U A

GA

C

C C G G G

G G C C C

A

C

C

G

C

U

U

C

A

C

G

U

U

C

U

A

CG

A

G

39984003

9984003998400

3998400

3998400

3998400

5998400

5998400

5998400 5

998400 5998400

5998400




TΨC loop

TΨC stem

D-stem

D-loop

Anticodon stem

Anticodon loop

Variable loop



G

CA

G

C

U

C







Archaea 5

Therm

incola potens


87099

1001

1001

02




Desulfotomaculum

gibsoniae

Desulfotom

aculum acetoxidans

Des

ulfo

spor

osin

us or

ientis

Desu

lfosp

oros

inus

mer

idiei

Desulfo

sporos

inus y

oung

iae

Desulfitobacter

ium dehalogenans





Meth

anosa

rcina

barke

ri

Meth

anoh

aloph

ilus m

ahii

Meth

anoc

occo

ides

burto

nii

Methanom










6 Archaea


4 Discussion






Archaea 7


















8 Archaea

Methanosarcinaceae

Thermoplasmatales



LGT from


APA

Arc

haea

LUCA

Methanocellales

Halobacteriales

Methanomicrobiales

Thermococcales

Methanosaetaceae

Archaeoglobales

Methanococcales


Interdomain LGT(s)

PylS (c)



PylSc

PylSn2

33

1

7th order

Bacteria

Eukarya

PylS (n + c) fused






Archaea 9







Acknowledgments


References






















10 Archaea































Archaea 11

















4 Archaea

G

G

G

G

G

G

U

C U A GGG

AA

G A U CA

UAC

A

C

G

G

A

G

U

G

C

U

UC AU AC U A

GA

C

A

C

C

G

C

C

C

C

U

C

A

G C G G G

U G C C CU

U

C

G

A

AA

G

G

G

G

G

A

C

C U G G

G A C CAG

C

G

G

G

U

C

G

A

U

C

C

AC AU AC U A

GA

C

A

C

C

G

C

U

C

U

C

U

G

C G G G G U

G C C C C GC

U

A

CGG

G

C

C

G

G

A

G

G

U G GU

G A C C

C

C

G

G

C

G

G

C

A

C

C

GC AU CC U A

GA

C

C C G G G

G G C C C

C

G

C

U

U

C

C

U

U

C

U

A

CG

A

G

CG

(A)

U

G

G

G

A

A

C

C

C U A GUA

UG

G A U CA

UGA

A

U

G

G

A

U

U

G

C

C

UC AU AC U A

GA

U

A

C

C

G

C

C

U

U

U

C

G

C C G G G

G G C C CU

U

U

U

A

AG

G

G

A

G

U

G

U

C U G GU

CG

G A C CG AC

C

G

G

C

G

G

C

A

C

C

GC AU CC U A

GA

C

A

C

C

G

C

U

U

C

A

C

G

C C G G G

G G C C CU

U

C

U

A

CA

A

G

G

A

G

U

G

U

U U G G

UC

CG A C C

G

G AC

C

G

G

C

G

G

C

A

C

C

GC AU CC U A

GA

C

C C G G G

G G C C C

A

C

C

G

C

U

U

C

A

C

G

U

U

C

U

A

CG

A

G

39984003

9984003998400

3998400

3998400

3998400

5998400

5998400

5998400 5

998400 5998400

5998400




TΨC loop

TΨC stem

D-stem

D-loop

Anticodon stem

Anticodon loop

Variable loop



G

CA

G

C

U

C







Archaea 5

Therm

incola potens


87099

1001

1001

02




Desulfotomaculum

gibsoniae

Desulfotom

aculum acetoxidans

Des

ulfo

spor

osin

us or

ientis

Desu

lfosp

oros

inus

mer

idiei

Desulfo

sporos

inus y

oung

iae

Desulfitobacter

ium dehalogenans





Meth

anosa

rcina

barke

ri

Meth

anoh

aloph

ilus m

ahii

Meth

anoc

occo

ides

burto

nii

Methanom










6 Archaea


4 Discussion






Archaea 7


















8 Archaea

Methanosarcinaceae

Thermoplasmatales



LGT from


APA

Arc

haea

LUCA

Methanocellales

Halobacteriales

Methanomicrobiales

Thermococcales

Methanosaetaceae

Archaeoglobales

Methanococcales


Interdomain LGT(s)

PylS (c)



PylSc

PylSn2

33

1

7th order

Bacteria

Eukarya

PylS (n + c) fused






Archaea 9







Acknowledgments


References






















10 Archaea































Archaea 11

















Archaea 5

Therm

incola potens


87099

1001

1001

02




Desulfotomaculum

gibsoniae

Desulfotom

aculum acetoxidans

Des

ulfo

spor

osin

us or

ientis

Desu

lfosp

oros

inus

mer

idiei

Desulfo

sporos

inus y

oung

iae

Desulfitobacter

ium dehalogenans





Meth

anosa

rcina

barke

ri

Meth

anoh

aloph

ilus m

ahii

Meth

anoc

occo

ides

burto

nii

Methanom










6 Archaea


4 Discussion






Archaea 7


















8 Archaea

Methanosarcinaceae

Thermoplasmatales



LGT from


APA

Arc

haea

LUCA

Methanocellales

Halobacteriales

Methanomicrobiales

Thermococcales

Methanosaetaceae

Archaeoglobales

Methanococcales


Interdomain LGT(s)

PylS (c)



PylSc

PylSn2

33

1

7th order

Bacteria

Eukarya

PylS (n + c) fused






Archaea 9







Acknowledgments


References






















10 Archaea































Archaea 11

















6 Archaea


4 Discussion






Archaea 7


















8 Archaea

Methanosarcinaceae

Thermoplasmatales



LGT from


APA

Arc

haea

LUCA

Methanocellales

Halobacteriales

Methanomicrobiales

Thermococcales

Methanosaetaceae

Archaeoglobales

Methanococcales


Interdomain LGT(s)

PylS (c)



PylSc

PylSn2

33

1

7th order

Bacteria

Eukarya

PylS (n + c) fused






Archaea 9







Acknowledgments


References






















10 Archaea































Archaea 11

















Archaea 7


















8 Archaea

Methanosarcinaceae

Thermoplasmatales



LGT from


APA

Arc

haea

LUCA

Methanocellales

Halobacteriales

Methanomicrobiales

Thermococcales

Methanosaetaceae

Archaeoglobales

Methanococcales


Interdomain LGT(s)

PylS (c)



PylSc

PylSn2

33

1

7th order

Bacteria

Eukarya

PylS (n + c) fused






Archaea 9







Acknowledgments


References






















10 Archaea































Archaea 11

















8 Archaea

Methanosarcinaceae

Thermoplasmatales



LGT from


APA

Arc

haea

LUCA

Methanocellales

Halobacteriales

Methanomicrobiales

Thermococcales

Methanosaetaceae

Archaeoglobales

Methanococcales


Interdomain LGT(s)

PylS (c)



PylSc

PylSn2

33

1

7th order

Bacteria

Eukarya

PylS (n + c) fused






Archaea 9







Acknowledgments


References






















10 Archaea































Archaea 11

















Archaea 9







Acknowledgments


References






















10 Archaea































Archaea 11

















10 Archaea































Archaea 11

















Archaea 11

















Unique Characteristics of the Pyrrolysine System in the 7th Order of Methanogens: Implications for...

Documents

Transcript of Unique Characteristics of the Pyrrolysine System in the 7th Order of Methanogens: Implications for...