Giant DNA Virus Mimivirus Encodes Pathway for Biosynthesis ofUnusual Sugar 4-Amino-4,6-dideoxy-D-glucose (Viosamine)*□S

Received for publication, October 25, 2011, and in revised form, December 7, 2011 Published, JBC Papers in Press, December 8, 2011, DOI 10.1074/jbc.M111.314559

Francesco Piacente‡, Margherita Marin‡, Antonio Molinaro§, Cristina De Castro§, Virginie Seltzer¶, Annalisa Salis‡,Gianluca Damonte‡, Cinzia Bernardi‡, Jean-Michel Claverie¶, Chantal Abergel¶1, and Michela Tonetti‡2

From the ‡Department of Experimental Medicine and Center of Excellence for Biomedical Research, University of Genova, VialeBenedetto XV/1, 16132 Genova, Italy, the §Department of Organic Chemistry and Biochemistry, University of Napoli “Federico II,”Napoli, Italy, and ¶Information Génomique et Structurale, CNRS-UPR 2589, Aix-Marseille University, Institut de Microbiologie de laMediterranee, Parc Scientifique de Luminy, Case 934, FR-13288 Marseille, France

Background:Mimivirus is highly glycosylated; however, nothing is known about its glycan composition and structure.Results:We identified aMimivirusUDP-viosamine synthetic pathway, andwe determined the sugar composition of viral fibers.Conclusion:Our data give further support to the presence of a Mimivirus-encoded glycosylation machinery.Significance: These results contribute to shed light on the origin of viral glycosylation systems.

Mimivirus is one the largest DNA virus identified so far,infecting several Acanthamoeba species. Analysis of its genomerevealed the presence of a nine-gene cluster containing genespotentially involved in glycan formation. All of these genes areco-expressed at late stages of infection, suggesting their role inthe formation of the long fibers covering the viral surface.Among them, we identified the L136 gene as a pyridoxal phos-phate-dependent sugar aminotransferase. This enzyme wasshown to catalyze the formation of UDP-4-amino-4,6-dideoxy-D-glucose (UDP-viosamine) from UDP-4-keto-6-deoxy-D-glu-cose, a key compound involved also in the biosynthesis of L-rh-amnose. This finding further supports the hypothesis thatMimivirus encodes a glycosylation system that is completelyindependent of the amoebal host. Viosamine, together withrhamnose, (N-acetyl)glucosamine, and glucose, was found as amajor component of the viral glycans. Most of the sugars wereassociated with the fibers, confirming a capsular-like nature ofthe viral surface. Phylogenetic analysis clearly indicated thatL136 was not a recent acquisition from bacteria through hori-zontal gene transfer, but it was acquired very early during evo-lution. Implications for the origin of the glycosylation machin-ery in giant DNA virus are also discussed.

Mimivirus is a member of the nucleocytoplasmic large DNAvirus group (1) and one the largest viruses ever described, with adiameter of 700 nm and a 1.2-Mbp genome encoding more than1000 genes (2, 3). The large size and theGram-positive staining ofthe viral particles led to its initialmisidentification as a small intra-cellularbacterium,namedBradfordcoccus. Itwas later recognizedas a giant virus with a typical icosahedral capsid surrounded bylong fibers�120nm in length, supposed to be highly glycosylated,

this explaining the Gram staining (4–7). The Mimivirus mode ofinfection is unique because it penetrates the phagocytic cells usinginternalization vacuoles. Later, its nucleocapsid is transferred intoits host cytoplasm through the opening of a specific structure, thestargate, and the fusion of one of the two virion internal mem-branes with the vacuole one (8). From the very beginning, it waspredicted that the glycosylated fibers could be of a capsular natureandused by the virion to promote the phagocytosis of host cells bymimickingbacterial polysaccharide structures. They can also con-tribute to the increased virion resistance to proteases and otherenvironmental stresses.Adetailed analysis of the composition andstructure of Mimivirus glycans is thus needed to understand itsmode of infection.On a more general level, it is now becoming increasingly

obvious that, in contrast to other viruses, somemembers of thenucleocytoplasmic large DNA virus encode at least part, if notall, of the enzymes required for the glycosylation of their struc-tural proteins. For instance, Chlorella viruses (Phycodnaviri-dae), which infect unicellular green algae, encode functionalenzymes involved in the production of modified nucleotidesugars, glycosyltransferases, and glycosidases (9–12). Indeed,much evidence has indicated that the glycosylation of theirmajor capsid occurs independently of the algal host endoplas-mic reticulum/Golgi machinery (13). Interestingly, Mimivirusgenome analysis revealedmany genes potentially encoding pro-teins involved in glycan formation, including glycosyltrans-ferases and sugar-modifying enzymes (2). This strongly sug-gests that this virus also encodes an independent glycosylationsystem. We have previously demonstrated that two Mimivirusgenes, one being part of this cluster, encode two functionalenzymes required for the biosynthesis of UDP-L-Rha3 (14).Putative enzymes involved in the synthesis of 3-deoxy-D-

* This work was supported by CNRS, Agence Nationale de la Recherche, GrantANRBLAN08-0089.

□S This article contains supplemental Tables S1 and S2 and Figs. S1–S6.1 To whom correspondence may be addressed. Tel.: 33-491-825422; Fax:

33-491-825421; E-mail: Chantal.Abergel@igs.cnrs-mrs.fr.2 To whom correspondence may be addressed. Tel.: 39-010-3538151; Fax:

39-010-354415; E-mail: tonetti@unige.it.

3 The abbreviations used are: Rha, rhamnose; Vio, viosamine; GlcN(Ac),(N-acetyl)glucosamine; PLP, pyridoxal phosphate; UGD, UDP-D-Glc4,6-dehydratase; COSY, correlated spectroscopy; DQF-COSY, doublequantum-filtered COSY; T4K, ((2R,3R,4S,5S,6R)-3,4-dihydroxy-5-((3-hy-droxy-2-methyl-5-(phosphonooxymethyl)pyridin-4-yl)methylideneamino)-6-methyl-oxan-2-yl) (hydroxy-(((2R,3S,5R)-3-hydroxy-5-(5-methyl-2,4-di-oxo-pyrimidin-1-yl)oxolan-2-yl)methoxy)phosphoryl) hydrogen phosphate;contig, group of overlapping clones.

THE JOURNAL OF BIOLOGICAL CHEMISTRY VOL. 287, NO. 5, pp. 3009 –3018, January 27, 2012© 2012 by The American Society for Biochemistry and Molecular Biology, Inc. Published in the U.S.A.

JANUARY 27, 2012 • VOLUME 287 • NUMBER 5 JOURNAL OF BIOLOGICAL CHEMISTRY 3009

by guest on January 25, 2020http://w

ww

.jbc.org/D

ownloaded from

manno-octulosonate were also found in the recently sequencedmarine giant virus CroV, which infects the microflagellategrazer Cafeteria roembergensis (15).In this study, we characterized the Mimivirus L136 gene as a

new enzyme responsible for the production of the unusualmonosaccharide 4-amino-4,6-dideoxy-D-glucopyranose (vios-amine (Vio) or 4-aminoquinovose). Vio is found in several bac-terial glycans, such as the LPSO-antigens of Shigella disenteriaetype 7 and of the Shiga toxin-producing Escherichia coli O121(16), the fiber-associated pentasaccharide of Bacillus anthracisexosporium (17), the O-polysaccharide of Francisella tularen-sis and of the emerging pathogen Photorabdus asymbyotica(18), the glycans of Pseudomonas syringae flagellin (19), and thecapsular polysaccharides of several marine bacteria belongingto the genus Pseudoalteromonas and Shewanella (20–21). Wealso analyzed the sugar composition of Mimivirus by GC-MSand were able to demonstrate that, together with Vio, majorcomponents of the viral glycans are Rha, Glc, and (N-acetyl)-glucosamine (GlcN(Ac)). With the exception of Glc, most ofthe sugars were found associated with the fibers covering theviral capsids, thus confirming the capsular-like nature ofMimi-virus fibers.

EXPERIMENTAL PROCEDURES

L136 Sequence and Phylogenetic Analyses—Themost similarhomologues of the Mimivirus L136 PLP-dependent transami-nase were first identified using the Blast-Explorer tool (22) onthe Phylogeny.fr server (23). The environmental sequenceswere specifically retrieved using the NCBI server (24) and the“env_nr” database (25). A subset of sequences was selectedbased on the alignment quality (preserving enough informativepositions). An optimal multiple alignment was then computedusing MAFFT version 6 (26) on the CBRC-AIST server. Theneighbor-joining tree was computed from 234 ungapped posi-tions using the JTT default substitution model.For structural comparison of PLP-dependent aminotrans-

ferases, reference structures were retrieved from the ProteinData Bank (27). The L136 model was downloaded from theModBase server (28). The L136 model and the reference struc-tures were superimposed using the Coot software (29), and thebiggest ligand coordinates were retrieved and inserted in allcoordinate files to identify in all structures the contacting resi-dues less than 3.5 Å from it.Mimivirus 9 Gene Cluster Expression Profiles—The Mimivi-

rus transcriptomewas performedbyRNA-seq as described pre-viously (30). Briefly, cells were infected by Mimivirus with amultiplicity of infection of 1000 to obtain synchronization.Total RNA was extracted from infected cells collected everyhour from the beginning of the infectious cycle until 7 h tosample efficiently the early and intermediate phase of the infec-tion (31). A last point was collected at 11 h. We thus generatednine barcoded transcriptome libraries from 1 �g of total RNA,each with the SOLiDTM whole transcriptome analysis kit,pooled at equimolar concentrations. After emulsion PCR, themonoclonal beads were loaded on one slide of a SOLiDTM 3Plus system and sequenced (50-base pair reads) with theSOLiDTM Opti Fragment Library Sequencing chemistry. Atotal of 202,436,309 sequence reads were generated and subse-

quently aligned to the Mimivirus genome using Bfast. Tran-scription data of theMimivirus genes are available at theMimi-virus Genome Browser Web site.Expression and Purification of Recombinant L136 Protein—

The recombinant protein was expressed in BL21 Gold E. colicells (Stratagene) as glutathione S-transferase (GST) fusionprotein using the pGEX-6-P1 vector (GE Healthcare), asdescribed previously (9). Viral DNA was purified by standardprocedures. The sequence corresponding to L136 ORF in theviral genome (RefSeq ID NC_014649) was amplified by PCRusing the following primers: AATTGGATCCATGGGTCTT-GAAAAACTTAC (forward) andAATTCTCGAGTTATTTT-TATCAGCAAATTC (reverse), containing the BamHI andXhoI restriction sites (underlined), respectively. The recombi-nant protein (40,346 Da) was concentrated to 2–3mg/ml usingthe Centricon YM10 system (Millipore) and stored at 4 °C inTris-buffered saline (50 mM Tris-HCl, pH 7.5, 150 mM NaCl)containing 1mMDTT (TBSD). The absorbance spectrumof thepurified protein was obtained using a Beckman Coulter DU800spectrophotometer, and concentration was estimated byabsorbance at 280 nm using a computed extinction coefficientof 36330 M�1 cm�1 (32). Protein purity, determined by SDS-PAGE, exceeded 95% in all preparations.Enzymatic Activity Assays—The L136 enzymatic activity was

assayed in TBSD at 25 °C using 0.025–1.5 mM UDP(dTDP)-4-keto-6-deoxy-Glc and 1–10mMglutamate as substrates. Unlessotherwise indicated, all chemicals were obtained from Sigma-Aldrich. UDP(dTDP)-4-keto-6-deoxy-D-Glc was producedfrom UDP-D-Glc and dTDP-D-Glc by ATCV-1 and MimivirusUDP-D-Glc 4,6-dehydratase (UGD) (14). The reaction wasstopped by heat inactivation for 3 min at 80 °C, and the dena-tured enzymewas removed by ultrafiltration (Microcon YM10,Millipore). L136 enzymatic activitywas determined in a discon-tinuous assay. Aliquots were removed at each time point, andthe reaction was immediately stopped by heating at 80 °C for 3min. After clarification, the conversion of UDP(dTDP)-4-keto-6-deoxy-D-Glc to the product was determined by ion exchangeHPLC, as described (9). The formation of �-ketoglutaratederived from the transamination of glutamatewas followed by acoupled assaywith glutamate dehydrogenase bymonitoring thedisappearance of NADH at 340 nm. To monitor the reversereaction, the purified UDP(dTDP)-4-amino-6-deoxyhexoseproduced by L136 was purified and reacted again with L136 inthe presence of �-ketoglutarate. Kinetic parameters weredetermined by theMichaelis-Menten equation using nonlinearregression (GraphPad Prism).Purification of L136 Product—L136 product was purified by

anion exchange HPLC using the described procedure (9) andthen subjected to solid phase extraction using Carbographultraclean columns (150 mg/4 ml; Alltech). The solid phaseextraction columns were pretreated with 3 ml of 60% acetoni-trile/H2O containing 0.3% ammonium formate (pH 9), fol-lowed by one wash with 2 ml of H2O and one with 3 ml of 250mM NH4HCO3. The HPLC-purified compound was thenapplied to the solid phase extraction column. After a wash with3 ml of H2O, L136 product was eluted using 1 ml of 60% aceto-nitrile/H2O containing 0.3% ammonium formate and driedunder vacuum. The amount of the UDP-sugar recovered after

Viosamine Synthesis in Mimivirus

3010 JOURNAL OF BIOLOGICAL CHEMISTRY VOLUME 287 • NUMBER 5 • JANUARY 27, 2012

by guest on January 25, 2020http://w

ww

.jbc.org/D

ownloaded from

purification was determined by UV absorbance using �262 �10,000 M�1 cm�1 for UTP.Structural Characterization of L136 Product—Electrospray

ionization MS analysis was performed as described (14). Thestructural assignment of the UDP-4-amino-6-deoxyhexo-samine was obtained by NMR spectroscopy. One- and two-dimensional NMR spectra were recorded on a Bruker 600 DRXequipped with a cryoprobe on a solution of 500 �l of D2O.Double quantum-filtered phase-sensitive COSY experimentswere performed using data sets of 2048 � 512 points (33, 34);the data matrix was zero-filled in both dimensions to give amatrix of 4K � 2K points and was resolution-enhanced in bothdimensions by a cosine-bell function before Fourier transfor-mation. Coupling constants were determined on a first orderbasis from high resolution one-dimensional spectra or by two-dimensional phase-sensitive DQF-COSY. Heteronuclear singlequantumcorrelation spectroscopy andheteronuclearmultiple-bond correlation spectroscopy were measured in the 1H-de-tected mode via single quantum coherence with proton decou-pling in the 13C domain, using data sets of 2048 � 512 points.Experiments were carried out in the phase-sensitive mode (35),and the data matrix was extended to 2048 � 1024 points usingforward linear prediction extrapolation.Analysis ofMimivirusMonosaccharideComposition—Mimi-

virus was purified from infected Acanthamoeba castellani cul-ture as described (31). Briefly, the virus was collected by 10minof centrifugation at 15,000 � g and 4 °C and resuspended intoPBS with CsCl to obtain a density of 1.15. 20 ml of the suspen-sion were layered onto a discontinuous gradient of CsCl (1.45/1.35/1.25) and centrifuged at 18,000 � g for 30 min. The virusbandwas collected andwashed/centrifuged 5 timeswith 50mM

Tris, pH 8.0, storage buffer. Removal of fibers was achievedusing the published procedure (5), and the treated sample wascontrolled by electron microscopy. Approximately 0.5–1 �1011 viral particles were used for each analysis. Briefly, 100 mM

DTT was added to 200 �l of virus suspensions; the tubes werethen heated at 99 °C for 15 min. After cooling, 5 units of Ben-zonasewere added and incubated for 60min at 37 °C to degradeDNA. Proteins were then precipitated using 5 volumes of ace-tone at�20 °Covernight. After removal of acetone, the sampleswere resuspended in H2O and briefly sonicated in order to dis-rupt protein aggregates. Hydrolysis was performed with 2 N

trifluoroacetic acid (TFA) at 121 °C for 2 h under N2 with con-tinuous stirring (36). Parallel experiments also with 4 N TFAgave comparable sugar recoveries. TFA was then removedunder vacuum, and samples were subjected to three washeswithmethanol. Alditols were obtained by the addition of 300�lof 0.25 M NaBH4 dissolved in 1 M NH4OH, followed by incuba-tion at 40 °C for 2 h. The reaction was stopped with 30% aceticacid in methanol, and methyl borates were removed by fourwashes with methanol/acetic acid (200:1). Acetylation was car-ried on with 200 �l of acetic anhydride/pyridine (1:1) for 1 h at80 °C, followed by drying and postderivatization cleaning fourtimes, using H2O/ethyl acetate (1:0.5). Samples resuspended inethyl acetate were injected in a HP5890 series II gas chromato-graph coupled to aHP5889Amass spectrometer equippedwithan electron impact ionization source (Hewlett-Packard). Sepa-ration was performed on an SE54 capillary column (0.32mm�

30 m; Alltech); the helium gas flow was 2 ml/min. The oventemperature gradient was as follows: initial temperature of170 °C, isothermal for 4 min, 170–280 °C (rate, 5 °C/min), andisothermal to 36 min. The MS analysis was performed in full-scan mode.Sugar quantificationwas done by comparing the area of sam-

ple peaks with that obtained for standard sugars, using selectedions. The most abundant ions were used for each monosaccha-ride;m/z 115was used for 6-deoxyhexoses, pentoses, and hexo-ses and m/z 84 for the amino sugars. Response proved to belinear in the range of the sugar concentrations used for theanalysis.

RESULTS

Analysis of L136 Sequence—We previously characterized theMimivirus R141 as a dehydratase and L780 as an epimerase/reductase, both involved in the synthesis of UDP-L-Rha (14).We then searched the viral genome for other putative enzymesable to catalyze the formation of modified monosaccharidesand identified a nine-gene cluster, including R141 (Fig. 1A).L136, the second in this cluster, has been annotated as a PLP-dependent aminotransferase and exhibited a weak homologywith bacterial sugar aminotransferases belonging to the DegT/DnrJ/EryC1/StrS family. The best BLAST hits correspond to aputative sugar aminotransferase from F. tularensis SCHU S4(29% identity over 361 amino acids) and from Shewanella sedi-minis HAW-EB3 (27% identity over 359 amino acids). Identitywith B. anthracis AntC, which produces dTDP-4-amino-4,6-dideoxy-Glc in the anthrose biosynthetic pathway (17), was26% over 250 amino acids. E. coli and Salmonella entericaWecE, which are involved in the synthesis of D-fucosaminefound in the enterobacterial common antigen (37), were 25%identical with L136 over 354 and 227 amino acids, respectively.Finally, the BLAST search against the environmental sequencedatabase revealed closer homologues (up to 46% identity over354 amino acids) with hypothetical proteins from marine met-agenomic samples (25). To obtain some information on theL136 specificity, we retrieved itsModBasemodel (28), based onthe Caulobacter crescentus perosamine synthase structure(Protein Data Bank codes 3DR4 or 3DR7) (38) with whom itshares 26% identity over 236 amino acids. We then comparedthe model with the reference structure of PLP-dependent ami-notransferase from Campylobacter jejuni (Protein Data Bankcode 1O61) (39) and theDesI PLP-dependent aminotransferasefrom Streptomyces venezuelae (Protein Data Bank code 2PO3)(40). The multiple alignment of L136 with the bacterialenzymes and the reference structures is reported in supplemen-tal Fig. S1. All residues that have been identified in transami-nases and are involved in PLP binding and in the catalyticmechanism (40) are all well conserved in the L136 sequence.The ligand (T4K) of S. venezuelae aminotransferase 2PO3 (40)was used to measure contacting residues less than 3.5 Å fromT4K in all structures. They are highlighted in supplemental Fig.S1.The phylogenetic tree built using the L136 sequence, refer-

ence bacterial sequences, and selected environmentalsequences exhibited two main clusters (supplemental Fig. S2).One includes bacterial sequences fromdifferent classes (cluster

Viosamine Synthesis in Mimivirus

JANUARY 27, 2012 • VOLUME 287 • NUMBER 5 JOURNAL OF BIOLOGICAL CHEMISTRY 3011

by guest on January 25, 2020http://w

ww

.jbc.org/D

ownloaded from

2), whereas cluster 1 includes the L136 sequence, environmen-tal sequences, and one of the two S. sediminis paralogoussequences of PLP-dependent transaminases, the other onebelonging to cluster 2. We selected the environmentalsequences from contigs encompassing more than one gene inorder to obtain some hints about their putative origins using adirect BLAST search against the Refseq database. With thisapproach, we were able to discriminate between possible viralor prokaryotic sequences. Details are reported in supplementalFig. S2. Results from the phylogenetic analysis strongly suggestthat the Mimivirus L136 sequence is ancestral and, if originat-ing from bacteria, was acquired early in evolution.L136 is encoded in a nine-gene cluster including several

putative enzymes involved in glycan formation (Fig. 1A). In par-ticular, R135 has been proposed to be a component of Mimivi-rus fibers and proved to be glycosylated and to be involved inantigenic response (5, 41). L137, L138, R139 each contain puta-tive domains belonging to the GT2 glycosyltransferase family(42). L140 does not show a clear GT domain; however, it dis-plays a 45% identity with the C-terminal domain of L138. R141encodes a functional UGD (14). L142 is predicted to have anN-terminal NeuD sugarO-acetyltransferase domain (43) and aC-terminal domain similar to the GT2 glycosyltransferase fam-ily. The last gene of this region, L143, contains a polysaccharidepyruvyltransferase domain (44). Interestingly, all of these genesstart to be expressed during the intermediate phase of viralcycle, and they are transcribed all along the late phase of theinfection with slightly different transcription profiles (Fig. 1B),suggesting that they are co-expressed and involved in virionmaturation and fibril formation.Recently, a Mimivirus isolate namedM4 has been character-

ized (7). Themature virions present slightly different morphol-ogies, with bald icosahedral capsids lacking the Mimivirus sur-face fibers. The corresponding genome encodes 155 fewergenes than Mimivirus, and the nine-gene cluster in M4 is

restricted to the R135 gene, L143 gene, and a split L142 gene(referred to asL142a andL142b in Fig. 1A). Another giant virus,Megavirus chilensis, amarine virus recently identified as amoredistant relative of Mimivirus, also encodes putative proteinsinvolved in glycoconjugate synthesis and, in particular,enzymes involved in amino sugar formation (45). The icosahe-dral capsid of Megavirus is surrounded by shorter hairs com-pared withMimivirus, and the nine-gene cluster is restricted totwo genes: mg878 (orthologous to Mimivirus R135) and tomg538 (orthologous to L143) (Fig. 1A).Biochemical Characterization of L136 Protein—We hypoth-

esized that the L136 protein could catalyze the transfer of anamino group from an amino acid donor to the C4 of UDP-4-keto-6-deoxy-D-Glc (Fig. 2). This latter compound is formed byUGD as an intermediate in the pathway leading to the produc-tion of UDP-L-Rha (14). Approximately 2 mg of pure L136 pro-tein per liter of bacterial culture were obtained after GST cleav-age. UV spectral analysis of the protein revealed an additionalpeak of light absorbance with a maximum at 330 nm, which isconsistent with the presence of the cofactor mainly as pyridox-amine phosphate (supplemental Fig. S3) (46).To verify the activity of the L136 protein, we incubated the

enzyme with UDP-4-keto-6-deoxy-D-Glc in the presence ofglutamate and observed the formation of a new compound byanion exchangeHPLC (Fig. 3, peak B). Formation of peak Bwasnot observed when L136 was heat-inactivated before incuba-tion (not shown). The aminotransferase activity was also con-firmedbymonitoring�-ketoglutarate formation in the reactionmixture, using a glutamate dehydrogenase-coupled assay. Theenzyme exhibited a kcat of 0.9 � 0.06 s�1 and aKm 146.5 � 44.3�M for UDP-4-keto-6-deoxy-D-Glc. dTDP-4-keto-6-deoxy-D-Glc could also be used as a substrate, but the resulting activitywas clearly reduced (about one-third compared with the UDP-bound substrate). The higher affinity of the L136 protein for theUDP-bound substrate is consistent with the preference of

FIGURE 1. The Mimivirus nine-gene cluster. A, conservation of the Mimivirus nine-gene cluster in Megavirus and in the Mimivirus isolate M4. The arrowsindicate the coding strand. The only two conserved genes are at the two edges of the cluster. B, Mimivirus gene expression profiles over the infection cycle. Thethree expression classes are color-coded (yellow for early, blue for intermediate, and gray for late). The early class of expression mostly corresponds to unknownfunction genes, the intermediate class corresponds to genes involved in DNA replication, and the late class of expression corresponds to genes involved incapsid production (structural proteins and proteins involved in fibril formation) and to genes encoding proteins that are associated to the capsid, such as theviral transcription machinery. L140 and L143 genes were co-expressed all along the intermediate and late infection cycle; L138, R139, L142, and L780 wereexpressed 1 h later until the end of the infection; L137, R141, and L136 were co-expressed at 4 h until the end; and R135 was the latest to be expressed, until theend of the infectious cycle (see the Mimivirus Genome Browser Web site). The putative functions of these genes are indicated under “Results.” L780 is not partof the nine-gene cluster; however, it has been reported because it catalyzes the last step of UDP-L-rhamnose synthesis.

Viosamine Synthesis in Mimivirus

3012 JOURNAL OF BIOLOGICAL CHEMISTRY VOLUME 287 • NUMBER 5 • JANUARY 27, 2012

by guest on January 25, 2020http://w

ww

.jbc.org/D

ownloaded from

Mimivirus UGD for UDP-D-Glc (14). At variance with bacterialsugar aminotransferases (16), when glutamate was substitutedwith glutamine as the amino group donor, the activity was neg-ligible. No effects were observed after the addition of differentconcentrations of PLP, confirming that the co-factor remainedtightly bound to the enzyme during the purification procedure,as already suggested by the UV spectral analysis of the protein.We addressed the possibility that L136 could also perform thereverse reaction, converting the UDP-amino-6-deoxysugarinto the UDP-4-keto-6-deoxy compound.We thus purified theUDP-6-deoxyhexosamine product and assayed the L136 enzy-matic activity in the presence of �-ketoglutarate. No activitywas detected as the production of either UDP-4-keto-6-deoxy-D-Glc or glutamate, demonstrating that the sugar transamina-tion is irreversible.Structural Characterization of UDP-4-keto-6-deoxyhexos-

amine—HPLC analysis indicated that L136 was able to catalyzethe transamination of UDP-4-keto-6-deoxy-D-Glc with theexpected formation of a UDP-4-amino-4,6-dideoxyhexose.Actually, electrospray ionization MS indicated an m/z (mass/charge) of 548 for the L136 product, consistent with the pres-ence of the proposed compound (supplemental Fig. S4).In order to elucidate the structure of the UDP-sugar and in

particular to identify the stereochemistry of the sugar bound toUDP, a complete NMR analysis was carried out. The 1H NMRspectrum (Fig. 4A) showed the presence of several signals, and acombination of homo- and heteronuclear two-dimensional

NMR experiments (DQF-COSY and 1H-13C heteronuclear sin-gle quantum correlation spectroscopy (Fig. 4, B and C) andheteronuclear multiple-bond correlation spectroscopy) wasexecuted to assign the three spin systems of themolecule: uracil(U), ribose (R), and the sugar residue (V). In fact, 1H and 13Cchemical shifts for uracil and ribose were in good agreementwith the literature (34), whereas sugar residue V was identifiedas Vio on the basis of its 1H and 13C chemical shifts (supple-mental Table S1) and JH,H vicinal coupling constants (47). Inagreement, with the exception of the anomeric proton signal,which is on a �-configured carbon atom (signal V1 in Fig. 4A),each ring proton displayed a large coupling constant (about 10Hz) with the neighboring protons, a clear indication that theseprotons occupy the axial position of the pyranose sugar ring.Therefore, the V residue monosaccharide possesses the glucostereochemistry (i.e. all substituents on the ring are oriented inthe equatorial position as is the case of Vio). Additionally, C-4chemical shift (58.2 ppm) of the V residue indicated that thiswas a nitrogen-bearing carbon atom; this information com-pleted the identification of this monosaccharide as Vio.SugarComposition ofMimivirusGlycans—Neutral sugar and

amino sugar presence in viral particleswas analyzed byGC-MS.Following acid hydrolysis, the free monosaccharides were con-verted to alditol acetates. Clear identification ofMimivirus sug-ars was obtained by comparing the retention times and thefragmentation spectra with standard monosaccharides. Reten-tion times of standard sugars and of the viral samples arereported in supplemental Table S2. The main components ofMimivirus glycans were Rha, Glc, and GlcN(Ac) (Fig. 5). Lowamounts of ribose, arabinose, xylose, mannose, and galactosewere also detected. Two peaks corresponding to unknown sug-ars were also contributing to theMimivirus glycan composition(indicated as peaks 1 and 2 in Fig. 5). Given the enzymatic activ-ity of the L136 protein and the identification of its product as aVio, we ran a control using this compound.As already reported,Vio was not substantially decomposed during acid hydrolysis(48). TheVio peak corresponded to a retention time of 20.7min(Fig. 6A), whichwas also observed in the virus sample chromat-ogram (Fig. 6C, peak 2); in agreement, the ion peak in the virussample has a fragmentation pattern identical to the Vio stand-ard (Fig. 6, B and D), definitely suggesting that Vio is a compo-nent of Mimivirus glycans. Another peak with a retention timeof 18.2 min could not be assigned to any available sugar stand-ards. However, the fragmentation spectrum indicated the pres-ence of an amino sugar, and, in particular, ion fragments werehighly suggestive of the presence of Vio modified with a 3-O-methyl group (supplemental Fig. S5). This hypothesis needs to

FIGURE 2. Proposed pathway for UDP-Vio synthesis in Mimivirus. The pathway utilizes UDP-4-keto-6-deoxy-D-Glc, the intermediate in UDP-L-Rha synthesis,and glutamate to produce UDP-D-Vio. UGD, encoded by Mimivirus R141, catalyzes the first step in this pathway.

FIGURE 3. Anion exchange HPLC analysis of nucleotide sugars. UDP-4-keto-6-deoxy-D-Glc (peak A) was incubated with glutamate in the presence ofL136 protein at 25 °C. The progressive formation of a new compound (peak B)was observed. The chromatogram was obtained after a 15-min incubation.

Viosamine Synthesis in Mimivirus

JANUARY 27, 2012 • VOLUME 287 • NUMBER 5 JOURNAL OF BIOLOGICAL CHEMISTRY 3013

by guest on January 25, 2020http://w

ww

.jbc.org/D

ownloaded from

be confirmed by the production of a standard sugar and NMRanalysis. The modified bacterial amino sugars muramic acidand bacillosamine were not found, excluding the presence oftypical bacterial peptidoglycan or of bacteria-likeN-linked gly-cans (49).Because it has been postulated that the Mimivirus fibers are

highly glycosylated, we have also analyzed the sugar content ofthe defibered viral particles. The electron microscopy analysisofMimivirus particles confirmed that the treatment (5) led to amajority of bald particles (�80%). As a consequence, we meas-ured a significant reduction (about 90%) of Rha, Vio, GlcN(Ac),and the unidentified sugar, whereas the decrease in Glc and inthe other neutral sugarswas less pronounced (Fig. 7). This dem-

onstrates that most of theMimivirus glycans are indeed associ-ated with the fibers.As indicated previously in Fig. 1A, the nine-gene cluster

found inMimivirus is not present inMegavirus.Megavirus alsolacks a homologue of the Mimivirus L780 gene, which encodesthe epimerase/reductase involved in the last step of UDP-L-Rhasynthesis. These differences might be at the origin of the dis-tinct appearances of the fiber layers covering theMegavirus andMimivirus particles. We thus performed a comparison of thesugar composition of Megavirus with Mimivirus using GC-MS(supplemental Fig. S6). As expected from the comparison oftheir gene contents, their glycan composition appeared com-pletely different. The Megavirus lacks Rha, Vio, and the puta-

FIGURE 4. NMR analysis of the product of L136 enzymatic activity at 600 MHz, 300 K, D2O. A, the assigned proton spectrum of UDP-Vio together with itsmolecular structure where the labeling of the three spin systems (ribose (R), uracil (U), and Vio (V)) is indicated. B, expansion of the DQF-COSY spectrum ofUDP-Vio and attribution of the cross-peaks relating to Vio ring protons. C, expansion of heteronuclear single quantum correlation spectroscopy spectrumshowing the more pertinent cross-peaks of UDP-Vio and their attribution.

Viosamine Synthesis in Mimivirus

3014 JOURNAL OF BIOLOGICAL CHEMISTRY VOLUME 287 • NUMBER 5 • JANUARY 27, 2012

by guest on January 25, 2020http://w

ww

.jbc.org/D

ownloaded from

tive 3-O-methyl-Vio. On the other hand, three new sugarsabsent in the Mimivirus fibers appear in the Megavirus chro-matogram. Further studies are needed to characterize themanddetermine the Megavirus fiber glycan composition. BecauseMimivirus andMegavirus were grown in the same host (A. cas-tellani), our results link the composition of the proteoglycanicfiber layers with the viral gene contents and strongly supportthe notion that the capsid maturation proceeds independentlyfrom the cellular host.

DISCUSSION

In the present work, we provided evidence that Mimivirusencodes a functional sugar transaminase able to transfer anamino group from glutamate to UDP-4-keto-6-deoxy-D-Glc,leading to the formation of a UDP-D-Vio. The gluco-configura-tion was univocally demonstrated by NMR analysis, whichexcluded the presence of the galacto-configured fucosamine(Fuc4N), a constituent of the enterobacterial common antigen(37). This finding further supports the hypothesis that Mimivi-rus encodes a host-independent glycosylation machinery. Wealso demonstrated that Vio is one of the main components ofMimivirus fiber-associated oligo(poly)saccharides, and indeed,we provided the first characterization of neutral and amino

sugars composing Mimivirus glycans. Previous reports haveindicated that Mimivirus structural proteins are densely glyco-sylated, and much indirect evidence suggested that the glycansare mainly linked to the long fibers covering the viral particles(3–5). Our results proved that major components ofMimivirussaccharides are Rha, Glc, GlcN(Ac), and Vio. Rha and theamino sugars were found mainly associated with the fibers,whereas Glc was only partially depleted after fiber removal.Wehave no clues at the moment about these differences, but it ispossible that Glc might be involved in the modification of lessaccessible structural proteins. Indeed, Luther et al. (50) veryrecently reported the identification of Mimivirus L230 as abifunctional enzyme, able to catalyze hydroxylation of lysine inMimivirus collagen-like proteins and transfer of a Glc moietyon it. Thus, glucosylation might occur independently fromother types of glycans.Vio often occurs in bacterial polysaccharides, generally car-

ryingN-acylation. In theO-antigens of S. disenteriae type 7 andof the E. coliO121 and in theO-antigen of the emerging humanpathogen Photorhabdus asymbiotica, the amino group of Viocan be acylated by an acetyl or an acetyl-glycyl group (16, 18),whereas F. tularensis LPS O-antigen contains a formyl group(18). In B. anthracis spores, Vio is further modified to obtain2-O-methy-L-4-(3-hydroxy-3-methyl butanamido)-4,6-dide-oxy-D-Glc, which is named anthrose (17). This sugar, togetherwith N-acylation, also contains a 2-O-methyl group. The func-tion of this substituent might serve to block further chainextension.Anthrose is a component of the pentasaccharide thatmodifies the collagen-like BclA protein of the exosporiumexternal hairlike nap (51). Serological cross-reactivity betweenB. anthracis and other Vio-containing oligosaccharides wasobserved, in particular with P. syringae flagellin glycans andShewanella spp. MR4 capsular polysaccharide (51). Becausecross-reactivity between sera of patients affected by tularemiaand Mimivirus has been observed (41), it would be interestingto verify if this cross-reactivity could be also ascribed to thepresence of Vio.We cannot presently confirm that theMimivirus glycans also

contain modified forms of Vio (i.e. presenting N-acylationand/or O-methylation). The L136 gene is located in a Mimivi-rus genomic region containing several other genes possiblyinvolved in glycan formation (Fig. 1A). These include theN-ter-minal region of L142, which could be involved in the acylationof the sugar backbone. Putative O- and N-acyl(acetyl)trans-ferases are often strictly associated with the aminotransferasein bacterial Vio gene clusters (i.e.VioB is localized immediatelydownstream of the aminotransferase VioA in E. coli O7, and ithas been shown to catalyze the formation of an acetamidogroup on C-4 of the 6-deoxy sugar) (16). In addition, L143encodes a putative exonuclease V-like ketal pyruvate transfer-ase, which has been involved in the formation of acidic exopo-lysaccharides in symbiotic and environmental bacteria and inhuman pathogens (44, 52, 53). Moreover, putative O-methyl-transferases are also contained in the Mimivirus genome (2).Cloning and expression of these proteins, which could beinvolved in Viomodification, are currently under way, and theywill help us to identify the presence of derivatives of this sugarin Mimivirus glycans. All of the genes included in this genomic

FIGURE 5. GC-MS analysis of Mimivirus glycans. Alditol acetates obtainedafter acid release of Mimivirus monosaccharides were analyzed by electronimpact GC-MS in full scan mode. Selected ions (m/z 115, m/z 126, and m/z144), which are representative fragments derived from neutral sugars (6-de-oxyhexoses, pentoses, and hexoses), Vio, and hexosamines, respectively,were extracted to obtain the presented chromatograms. Peaks 1 and 2 corre-spond to unknown sugars.

Viosamine Synthesis in Mimivirus

JANUARY 27, 2012 • VOLUME 287 • NUMBER 5 JOURNAL OF BIOLOGICAL CHEMISTRY 3015

by guest on January 25, 2020http://w

ww

.jbc.org/D

ownloaded from

region are expressed late during the viral infection, as alreadyobserved for the enzymes involved in the UDP-L-Rha biosyn-thetic pathway (14). Because fibrils are added at the last stage ofviral particle formation, this finding is in agreement with a rolefor all these proteins in fibrillogenesis.L136 belongs to the aspartate aminotransferase fold type I

(AAT_I) superfamily of PLP-dependent enzymes and in partic-ular to the DegT/DnrJ/EryC1/StrS family (pfam01041). Mem-bers of this family are involved in the transfer of amino groupsto several different substrates to obtain modified amino sugars,which are frequently found in bacterial polysaccharides or inantibiotics, such as erythromycin and other macrolides. L136

displays relatively low sequence homology with the character-ized enzymes of this family, in particular with VioA from E. coliand DesI from S. venezuelae, which catalyze a similar reaction,from dTDP-4-keto-6-deoxy-D-Glc to dTDP-Vio (16, 40). How-ever, all of the residues previously recognized as essential forPLP binding are conserved in L136. This is also confirmed bycomparison of the L136model with the published structures ofsugar transaminases.The phylogenetic analysis indicates with strong support that

the L136 gene was not acquired from a bacteria by a recenthorizontal transfer. However, the L136 sequence is even moredistant from eukaryotic homologues. Thus, its origin, as well asthe one of its closest homologues found in the environmentaldata set, remains unknown, as for the many “virus-only” genesfound inMimivirus.On the other hand, the same genomic clus-ter encodes R141, which catalyzes the first step of Vio biosyn-thesis and, together with L780, also leads to L-Rha. Both R141and L780 exhibit a clear affinity with protist homologues, sug-gesting a lateral gene transfer from a cellular host toMimivirus(14). Amixture of enzymes of unknown, eukaryotic, or bacterialorigins appears to constitute a Mimivirus-encoded UDP-N-acetylglucosamine biosynthetic pathway4 as well as the GDP-L-fucose biosynthetic pathways in Chlorella viruses (9, 10). Thiscould be explained by the co-existence of genes from a veryancestral glycosylation machinery predating the emergence ofthe eukarya, more recently complemented or replaced by genesof extant protozoans and bacteria. The recently discovered

4 M. Tonetti, F. Piacente, and C. Abergel, unpublished results.

FIGURE 6. Identification of viosamine in Mimivirus glycans by GC-MS. Chromatograms were obtained by extraction of m/z 126 ion current. A, standard Vio;B, fragmentation spectrum of standard Vio; C, virus sample (zoom view of the chromatogram reported in Fig. 5). Peaks 1 and 2 correspond to unknown sugars.D, fragmentation spectrum of peak 2.

FIGURE 7. Effects of fiber removal on Mimivirus sugar composition. Mon-osaccharide concentration was determined by GC-MS analysis. NT, untreatedwild-type virus; DF, defibered virus. The inset displays the relative amounts ofthe compound eluting at 18.1 min as area of the peaks.

Viosamine Synthesis in Mimivirus

3016 JOURNAL OF BIOLOGICAL CHEMISTRY VOLUME 287 • NUMBER 5 • JANUARY 27, 2012

by guest on January 25, 2020http://w

ww

.jbc.org/D

ownloaded from

Megavirus (45) supports this hypothesis by suggesting thatgiant viruses, most probably derived from an ancestral cellulargenome, might have conserved genes from an ancestral glyco-sylation machinery. The subsequent process of reductive evo-lution and lineage-specific losses, combined with episodic lat-eral gene transfer (including non-orthologuous replacements),would then have resulted in the complex picture we are seeingtoday. Thus, the study of the giant DNA virus-encoded glyco-enzymes as well as the characterization of their glycan struc-tures might turn out to be of general interest in probing theorigin and ancient evolution of today’s cellular glycosylationpathways.

Acknowledgments—We thankDr.Martin Young (Institute for Biolog-ical Sciences, National Research Council of Canada) for providing thebacillosamine standard, Prof. Antonio De Flora and Prof. UmbertoBenatti (University of Genova) for helpful discussion, and Dr. Mat-thieu Legendre for the transcriptomic data.We acknowledge the use ofthe PACA-Bioinfo (IBISA) bioinformatic platform.

REFERENCES1. Iyer, L. M., Balaji, S., Koonin, E. V., and Aravind, L. (2006) Evolutionary

genomics of nucleo-cytoplasmic large DNA viruses. Virus Res. 117,156–184

2. Raoult, D., Audic, S., Robert, C., Abergel, C., Renesto, P., Ogata, H., LaScola, B., Suzan, M., and Claverie, J. M. (2004) The 1.2-megabase genomesequence of Mimivirus. Science 306, 1344–1350

3. Claverie, J. M., and Abergel, C. (2010) Mimivirus. The emerging paradoxof quasi-autonomous viruses. Trends Genet. 26, 431–437

4. Renesto, P., Abergel, C., Decloquement, P., Moinier, D., Azza, S., Ogata,H., Fourquet, P., Gorvel, J. P., and Claverie, J. M. (2006) Mimivirus giantparticles incorporate a large fraction of anonymous and unique gene prod-ucts. J. Virol. 80, 11678–11685

5. Xiao, C., Kuznetsov, Y. G., Sun, S., Hafenstein, S. L., Kostyuchenko, V. A.,Chipman, P. R., Suzan-Monti, M., Raoult, D., McPherson, A., and Ross-mann,M. G. (2009) Structural studies of the giantmimivirus. PLoS Biol. 7,e92

6. Kuznetsov, Y. G., Xiao, C., Sun, S., Raoult, D., Rossmann,M., andMcPher-son, A. (2010) Atomic force microscopy investigation of the giant mimi-virus. Virology 404, 127–137

7. Boyer, M., Azza, S., Barrassi, L., Klose, T., Campocasso, A., Pagnier, I.,Fournous, G., Borg, A., Robert, C., Zhang, X., Desnues, C., Henrissat, B.,Rossmann, M. G., La Scola, B., and Raoult, D. (2011) Mimivirus showsdramatic genome reduction after intraamoebal culture. Proc. Natl. Acad.Sci. U.S.A. 108, 10296–102301

8. Mutsafi, Y., Zauberman, N., Sabanay, I., and Minsky, A. (2010) Vaccinia-like cytoplasmic replication of the giant Mimivirus. Proc. Natl. Acad. Sci.U.S.A. 107, 5978–5982

9. Tonetti, M., Zanardi, D., Gurnon, J. R., Fruscione, F., Armirotti, A., Da-monte, G., Sturla, L., De Flora, A., and Van Etten, J. L. (2003) Parameciumbursaria Chlorella virus 1 encodes two enzymes involved in the biosyn-thesis of GDP-L-fucose and GDP-D-rhamnose. J. Biol. Chem. 278,21559–21565

10. Fruscione, F., Sturla, L., Duncan, G., Van Etten, J. L., Valbuzzi, P., De Flora,A., Di Zanni, E., and Tonetti, M. (2008) Differential role of NADP� andNADPH in the activity and structure of GDP-D-mannose 4,6-dehydratasefrom two Chlorella viruses. J. Biol. Chem. 283, 184–193

11. Zhang, Y., Xiang, Y., Van Etten, J. L., and Rossmann, M. G. (2007) Struc-ture and function of a Chlorella virus-encoded glycosyltransferase. Struc-ture 15, 1031–1039

12. Sun, L., Adams, B., Gurnon, J. R., Ye, Y., and Van Etten, J. L. (1999) Char-acterization of two chitinase genes and one chitosanase gene encoded byChlorella virus PBCV-1. Virology 263, 376–387

13. Van Etten, J. L., Gurnon, J. R., Yanai-Balser, G. M., Dunigan, D. D., and

Graves, M. V. (2010) Chlorella viruses encode most, if not all, of the ma-chinery to glycosylate their glycoproteins independent of the endoplasmicreticulum and Golgi. Biochim. Biophys. Acta 1800, 152–159

14. Parakkottil Chothi,M., Duncan, G. A., Armirotti, A., Abergel, C., Gurnon,J. R., Van Etten, J. L., Bernardi, C., Damonte, G., and Tonetti, M. (2010)Identification of an L-rhamnose synthetic pathway in two nucleocytoplas-mic large DNA viruses. J. Virol. 84, 8829–8838

15. Fischer, M. G., Allen, M. J., Wilson, W. H., and Suttle, C. A. (2010) Giantvirus with a remarkable complement of genes infectsmarine zooplankton.Proc. Natl. Acad. Sci. U.S.A. 107, 19508–19513

16. Wang, Y., Xu, Y., Perepelov, A. V., Qi, Y., Knirel, Y. A,Wang, L., and Feng,L. (2007) Biochemical characterization of dTDP-D-Qui4N and dTDP-D-Qui4NAc biosynthetic pathways in Shigella dysenteriae type 7 and Esch-erichia coli O7. J. Bacteriol. 189, 8626–8635

17. Dong, S., McPherson, S. A., Wang, Y., Li, M., Wang, P., Turnbough, C. L.Jr., and Pritchard, D. G. (2010) Characterization of the enzymes encodedby the anthrose biosynthetic operon of Bacillus anthracis. J. Bacteriol.192, 5053–5062

18. Kondakova, A. N., Kirsheva, N. A., Shashkov, A. S., Shaikhutdinova, R. Z.,Arabtsky, N.P., Ivanov, S. A., Anisimov, A. P., and Knirel, Y. A. (2011) Lowstructural diversity of the O-polysaccharides of Photorhabdus asymbi-otica subspp. asymbiotica and australis and their similarity to the O-po-lysaccharides of taxonomically remote bacteria including Francisella tu-larensis. Carbohydr. Res. 346, 1951–1955

19. Takeuchi, K., Ono, H., Yoshida, M., Ishii, T., Katoh, E., Taguchi, F., Miki,R.,Murata, K., Kaku, H., and Ichinose, Y. (2007) Flagellin glycans from twopathovars of Pseudomonas syringae contain rhamnose in D and L config-urations in different ratios and modified 4-amino-4,6-dideoxyglucose. J.Bacteriol. 189, 6945–6956

20. Nazarenko, E. L., Komandrova, N. A., Gorshkova, R. P., Tomshich, S. V.,Zubkov, V. A., Kilcoyne, M., and Savage, A. V. (2003) Structures of poly-saccharides and oligosaccharides of some Gram-negative marine proteo-bacteria. Carbohydr. Res. 338, 2449–2457

21. Leone, S., Silipo, A., Nazarenko, E. L., Lanzetta, R., Parrilli, M., and Mo-linaro, A. (2007) Molecular structure of endotoxins from Gram-negativemarine bacteria. An update.Marine Drugs 5, 85–112

22. Dereeper, A., Audic, S., Claverie, J. M., and Blanc, G. (2010) BLAST-EX-PLORER helps you building data sets for phylogenetic analysis. BMCEvol.Biol. 10, 8

23. Dereeper, A., Guignon, V., Blanc, G., Audic, S., Buffet, S., Chevenet, F.,Dufayard, J. F., Guindon, S., Lefort, V., Lescot, M., Claverie, J. M., andGascuel, O. (2008) Phylogeny.fr. Robust phylogenetic analysis for the non-specialist. Nucleic Acids Res. 36,W465–W469

24. Altschul, S. F., Madden, T. L., Schäffer, A. A., Zhang, J., Zhang, Z., Miller,W., and Lipman, D. J. (1997) Gapped BLAST and PSI-BLAST. A newgeneration of protein database search programs. Nucleic Acids Res. 25,3389–3402

25. Venter, J. C., Remington, K., Heidelberg, J. F., Halpern, A. L., Rusch, D.,Eisen, J. A.,Wu,D., Paulsen, I., Nelson, K. E., Nelson,W., Fouts, D. E., Levy,S., Knap, A. H., Lomas, M. W., Nealson, K., White, O., Peterson, J., Hoff-man, J., Parsons, R., Baden-Tillson, H., Pfannkoch, C., Rogers, Y. H., andSmith, H. O. (2004) Environmental genome shotgun sequencing of theSargasso Sea. Science 304, 66–74

26. Katoh, K., and Toh, H. (2008) Recent developments in the MAFFT mul-tiple sequence alignment program. Brief. Bioinform. 9, 286–298

27. Berman, H. M., Westbrook, J., Feng, Z., Gilliland, G., Bhat, T. N., Weissig,H., Shindyalov, I. N., and Bourne, P. E. (2000) The Protein Data Bank.Nucleic Acids Res. 28, 235–242

28. Pieper, U., Webb, B. M., Barkan, D. T., Schneidman-Duhovny, D., Sch-lessinger, A., Braberg, H., Yang, Z., Meng, E. C., Pettersen, E. F., Huang,C. C., Datta, R. S., Sampathkumar, P., Madhusudhan, M. S., Sjölander, K.,Ferrin, T. E., Burley, S. K., and Sali, A. (2011) ModBase. A database ofannotated comparative protein structure models and associated re-sources. Nucleic Acids Res. 39, D465–D474

29. Emsley, P., and Cowtan, K. (2004) Coot. Model-building tools for molec-ular graphics. Acta Crystallogr. D Biol. Crystallogr. 60, 2126–2132

30. Legendre, M., Santini, S., Rico, A., Abergel, C., and Claverie, J. M. (2011)Breaking the 1000-gene barrier for Mimivirus using ultra-deep genome

Viosamine Synthesis in Mimivirus

JANUARY 27, 2012 • VOLUME 287 • NUMBER 5 JOURNAL OF BIOLOGICAL CHEMISTRY 3017

by guest on January 25, 2020http://w

ww

.jbc.org/D

ownloaded from

and transcriptome sequencing. Virol. J. 8, 9931. Byrne, D., Grzela, R., Lartigue, A., Audic, S., Chenivesse, S., Encinas, S.,

Claverie, J. M., and Abergel, C. (2009) The polyadenylation site of Mimi-virus transcripts obeys a stringent “hairpin rule”. Genome Res. 19,1233–1242

32. Gasteiger, E., Hoogland, C., Gattiker, A., Duvaud, S.,Wilkins,M. R., Appel,R. D., and Bairoch, A. (2005) Protein Identification and Analysis Tools onthe ExPASy Server. in The Proteomics Protocols Handbook (Walker, J. M.,ed) pp. 571–607, Humana Press, Totowa, NJ

33. Piantini, U., Sørensen, O. W., and Ernst, R. R. (1982) Multiple quantumfilters for elucidating NMR coupling networks. J. Am. Chem. Soc. 104,6800–6801

34. Rance, M., Sørensen, O. W., Bodenhausen, G., Wagner, G., Ernst, R. R.,andWüthrich, K. (1983) Improved spectral resolution in COSY 1H NMRspectra of proteins via double quantum filtering. Biochem. Biophys. Res.Commun. 117, 479–485

35. States, D. J., Haberkorn, R. A., and Ruben, D. J. (1982) A two-dimensionalnuclear Overhauser experiment with pure absorption phase in four quad-rants. J. Magn. Reson. 48, 286–292

36. Manzi, A. (2001) Acid hydrolysis for release of monosaccharides. Curr.Protoc. Mol. Biol. 17, 17.16.1–17.16.11

37. Hwang, B. Y., Lee, H. J., Yang, Y. H., Joo, H. S., and Kim, B. G. (2004)Characterization and investigation of substrate specificity of the sugaraminotransferase WecE from E. coli K12. Chem. Biol. 11, 915–925

38. Cook, P. D., Carney, A. E., and Holden, H. M. (2008) Accommodation ofGDP-linked sugars in the active site of GDP-perosamine synthase. Bio-chemistry 47, 10685–10693

39. Badger, J., Sauder, J. M., Adams, J. M., Antonysamy, S., Bain, K., Bergseid,M. G., Buchanan, S. G., Buchanan,M. D., Batiyenko, Y., Christopher, J. A.,Emtage, S., Eroshkina, A., Feil, I., Furlong, E. B., Gajiwala, K. S., Gao,X.,He,D., Hendle, J., Huber, A., Hoda, K., Kearins, P., Kissinger, C., Laubert, B.,Lewis, H. A., Lin, J., Loomis, K., Lorimer, D., Louie, G.,Maletic,M.,Marsh,C. D., Miller, I., Molinari, J., Muller-Dieckmann, H. J., Newman, J. M.,Noland, B.W., Pagarigan, B., Park, F., Peat, T. S., Post, K.W., Radojicic, S.,Ramos, A., Romero, R., Rutter, M. E., Sanderson, W. E., Schwinn, K. D.,Tresser, J., Winhoven, J., Wright, T. A., Wu, L., Xu, J., and Harris, T. J.(2005) Structural analysis of a set of proteins resulting from a bacterialgenomics project. Proteins 60, 787–796

40. Burgie, E. S., and Holden, H. M. (2007) Molecular architecture of DesI. Akey enzyme in the biosynthesis of desosamine. Biochemistry 46,8999–9006

41. Pelletier, N., Raoult, D., and La Scola, B. (2009) Specific recognition of themajor capsid protein of Acanthamoeba polyphaga mimivirus by sera ofpatients infected by Francisella tularensis. FEMS Microbiol. Lett. 297,

117–12342. Coutinho, P. M., Deleury, E., Davies, G. J., and Henrissat, B. (2003) An

evolving hierarchical family classification for glycosyltransferases. J. Mol.Biol. 328, 307–317

43. Lewis, A. L., Hensler, M. E., Varki, A., and Nizet, V. (2006) The group Bstreptococcal sialic acid O-acetyltransferase is encoded by neuD, a con-served component of bacterial sialic acid biosynthetic gene clusters. J. Biol.Chem. 281, 11186–11192

44. Ivashina, T. V., Fedorova, E. E., Ashina, N. P., Kalinchuk, N. A., Dru-zhinina, T.N., Shashkov, A. S., Shibaev, V.N., andKsenzenko, V.N. (2010)Mutation in the pssM gene encoding ketal pyruvate transferase leads todisruption of Rhizobium leguminosarum bv. viciae-Pisum sativum symbi-osis. J. Appl. Microbiol. 109, 731–742

45. Arslan, D., Legendre, M., Seltzer, V., Abergel, C., and Claverie J. M. (2011)DistantMimivirus relative with a larger genome highlights the fundamen-tal features of Megaviridae. Proc. Natl. Acad. Sci. U.S.A. 108,17486–17491

46. Drsata, J., Bousová, I., and Malon, P. (2005) Determination of quality ofpyridoxal-5�-phosphate enzyme preparations by spectroscopic methods.J. Pharm. Biomed. Anal. 37, 1173–1177

47. Knirel, Y. A., Dashunin, V. V., Shashkov, A. S., Kochetkov, N. K., Dmitriev,B. A., and Hofman, I. L. (1988) Somatic antigens of Shigella. Structure ofthe O-specific polysaccharide chain of the Shigella dysenteriae type 7 li-popolysaccharide. Carbohydr. Res. 179, 51–60

48. Jann, B., and Jann, K. (1967) 4-Amino-4,6-dideoxyhexoses isolated fromlipopolysaccharides of Escherichia coli. Eur. J. Biochem. 2, 26–31

49. Sharon, N. (2007) Celebrating the golden anniversary of the discovery ofbacillosamine, the diamino sugar of a Bacillus. Glycobiology 17,1150–1155

50. Luther, K. B., Hülsmeier, A. J., Schegg, B., Deuber, S. A., Raoult, D., andHennet, T. (2011) Mimivirus collagen is modified by bifunctional lysylhydroxylase and glycosyltransferase enzyme. J. Biol. Chem. 286,43701–43709

51. Kubler-Kielb, J., Vinogradov, E., Hu, H., Leppla, S. H., Robbins, J. B., andSchneerson, R. (2008) Saccharides cross-reactive with Bacillus anthracisspore glycoprotein as an anthrax vaccine component. Proc. Natl. Acad.Sci. U.S.A. 105, 8709–8712

52. Poli, A., Anzelmo, G., and Nicolaus, B. (2010) Bacterial exopolysaccha-rides from extreme marine habitats. Production, characterization, andbiological activities.Marine Drugs 8, 1779–1802

53. Herasimenka, Y., Cescutti, P., Impallomeni, G., and Rizzo, R. (2007) Ex-opolysaccharides produced by Inquilinus limosus, a new pathogen of cys-tic fibrosis patients. Novel structures with usual components. Carbohydr.Res. 342, 2404–2415

Viosamine Synthesis in Mimivirus

3018 JOURNAL OF BIOLOGICAL CHEMISTRY VOLUME 287 • NUMBER 5 • JANUARY 27, 2012

by guest on January 25, 2020http://w

ww

.jbc.org/D

ownloaded from

Chantal Abergel and Michela TonettiSeltzer, Annalisa Salis, Gianluca Damonte, Cinzia Bernardi, Jean-Michel Claverie,

Francesco Piacente, Margherita Marin, Antonio Molinaro, Cristina De Castro, Virginie4-Amino-4,6-dideoxy-d-glucose (Viosamine)

Giant DNA Virus Mimivirus Encodes Pathway for Biosynthesis of Unusual Sugar

doi: 10.1074/jbc.M111.314559 originally published online December 8, 20112012, 287:3009-3018.J. Biol. Chem. 

  10.1074/jbc.M111.314559Access the most updated version of this article at doi:

 Alerts:

  When a correction for this article is posted• 

When this article is cited• 

to choose from all of JBC's e-mail alertsClick here

Supplemental material:

  http://www.jbc.org/content/suppl/2011/12/08/M111.314559.DC1

  http://www.jbc.org/content/287/5/3009.full.html#ref-list-1

This article cites 52 references, 17 of which can be accessed free at

by guest on January 25, 2020http://w

ww

.jbc.org/D

ownloaded from