ebook img

Metabolism of halophilic archaea PDF

20 Pages·2008·0.52 MB·English·1 Downloads
Save to my drive
Quick download

Preview Metabolism of halophilic archaea

Extremophiles(2008)12:177–196 DOI10.1007/s00792-008-0138-x REVIEW Metabolism of halophilic archaea Michaela Falb Æ Kerstin Mu¨ller Æ Lisa Ko¨nigsmaier Æ Tanja Oberwinkler Æ Patrick Horn Æ Susanne von Gronau Æ Orland Gonzalez Æ Friedhelm Pfeiffer Æ Erich Bornberg-Bauer Æ Dieter Oesterhelt Received:27July2007/Accepted:6January2008/Publishedonline:16February2008 (cid:1)Springer2008 Abstract In spite of their common hypersaline environ- Abbreviations ment, halophilic archaea are surprisingly different in their DAP Diaminopimelate pathway nutritional demands and metabolic pathways. The meta- DHA(P) Dihydroxyacetone (phosphate) bolicdiversityofhalophilicarchaeawasinvestigatedatthe DKFP 6-Deoxy-5-ketofructose-1-phosphate genomiclevelthroughsystematicmetabolicreconstruction ED Entner-Douderoff pathway and comparative analysis of four completely sequenced EM Embden-Meyerhoff pathway species: Halobacterium salinarum, Haloarcula marismor- IPP Isopentenyl diphosphate (isopentenyl tui, Haloquadratum walsbyi, and the haloalkaliphile pyrophosphate) Natronomonas pharaonis. The comparative study reveals KD(P)G 2-Dehydro-3-deoxy-(6-phospho)gluconate differentsetsofenzymegenesamongsthalophilicarchaea, (2-keto-3-deoxy-(6-phospho)gluconate) e.g. in glycerol degradation, pentose metabolism, and PAB p-Aminobenzoate folate synthesis. The carefully assessed metabolic data PP Pentose-phosphate pathway represent a reliable resource for future system biology PRPP 5-Phospho-D-ribosyl-1-pyrophosphate approaches as it also links to current experimental data on RuBisCO Ribulose-bisphosphate carboxylase (halo)archaea from the literature. RuMP Ribulose monophosphate pathway TCA Tricarboxylic acid cycle Keywords Metabolism (cid:1) Archaea (cid:1) Haloarchaea (cid:1) THF 5,6,7,8-Tetrahydrofolate Halobacterium salinarum (cid:1) Pathway database (cid:1) Metabolic pathways (cid:1) Enzymes (cid:1) Comparative genomics Introduction CommunicatedbyD.A.Cowan. Electronicsupplementarymaterial Theonlineversionofthis Extremely halophilic archaea are a diverse group of eury- article(doi:10.1007/s00792-008-0138-x)containssupplementary archaeota that inhabit hypersaline environments (3–5 M) material,whichisavailabletoauthorizedusers. suchassaltlakes,saltponds,andmarinesalterns.Theyare oftenreferredtoas‘‘halobacteria,’’namedafterthemodel M.Falb(cid:1)K.Mu¨ller(cid:1)L.Ko¨nigsmaier(cid:1)T.Oberwinkler(cid:1) P.Horn(cid:1)S.vonGronau(cid:1)O.Gonzalez(cid:1)F.Pfeiffer(cid:1) organism Halobacterium salinarum, whose proton pump D.Oesterhelt(&) bacteriorhodopsin is one of the best-studied membrane DepartmentofMembraneBiochemistry, proteins. Although haloarchaea share certain features in Max-Planck-InstituteofBiochemistry, order to adapt to their extreme environment, i.e. acidic AmKlopferspitz18,82152Martinsried,Germany e-mail:[email protected] protein machineries, respiratory chains and rhodopsins, theirmetabolismisconsiderablydifferentfromeachother. E.Bornberg-Bauer Although there are carbohydrate-utilizing species such as EvolutionaryBioinformatics,InstituteforEvolution Haloferax mediterranei, Haloarcula marismortui, and andBiodiversity,Westfa¨lischeWilhelmsUniversity, Schlossplatz4,48149Mu¨nster,Germany Halococcus saccharolyticus, which catabolize hexoses 123 178 Extremophiles(2008)12:177–196 (glucose,fructose),pentoses(arabinose,xylulose),sucrose, Escalante-Semerena2004;Zayasetal.2006).Archaeaalso and lactose (Rawal et al. 1988; Altekar and Rangaswamy employ novel enzymes and precursors for pentose forma- 1992; Johnsen et al. 2001; Johnsen and Schonheit 2004), tion (Grochowski et al. 2005) and aromatic amino acid other haloarchaea like H. salinarum are not capable of synthesis (White 2004; Porat et al. 2006) circumventing sugar degradation (Rawal et al. 1988). Instead, non-car- absentenzymesoftheclassicalpentose-phosphatepathway. bohydrate-utilizing species thrive on amino acids and For this detailed review of haloarchaeal metabolism, typical compounds of hypersaline habitats. Haloferax vol- metabolic pathways of halophilic archaea were systemati- canii,forexample,isabletogrowonglycerolandorganic cally reconstructed. Currently, genome sequences of four acids (Kauri et al. 1990) excreted by primary halophilic diverse haloarchaeal species are available for comparative producers Dunaliella salina (Elevi Bardavid et al. 2006) analysis (Table 1), namely that of the model organism andMicrocoleuschthonoplastes(Zviagintsevaetal.1995), H. salinarum (Ng et al. 2000; Pfeiffer et al. 2008; respectively. Haloarchaea differ not only in their catabolic http://www.halolex.mpg.de), the metabolic-versatile pathways but also in their nutritional requirements. While H. marismortui (Baliga et al. 2004), the haloalkaliphile simple growth media were described for Haloferax volca- N. pharaonis (Falb et al. 2005), and the square-shaped nii (Kauri et al. 1990) and Natronomonas pharaonis (Falb Haloquadratum walsbyi (Bolhuis et al. 2006). The pre- et al. 2005), H. salinarum exhibits complex nutritional sentedmetabolicdatawillbeavaluableresourceforfuture demands.GrowthofHalobacteriumcellsisoftenlimitedin systembiologyapproaches,aseachmetabolicreactionhas synthetic media (Oesterhelt and Krippahl 1973; Grey and been carefully assessed and linked to experimental data Fitt 1976), in spite of rich amino acid (at least 10 amino from the literature. acids) and cofactor supplements (folate, biotin, thiamine). Themetabolicdiversityofhalophilicarchaeahasnotyet been investigated at the genomic level by metabolic Materials and methods reconstruction and comparative analysis. The absence of enzymegenesfornumerousmetabolicreactionsinarchaeal Metabolic reconstruction procedure genomeshas limitedreconstructionofmetabolicpathways so far. However, many of these pathway gaps have been MetabolicpathwaysofH.salinarumwerereconstructedin elucidated recently with the discovery of novel non- a two-step procedure. At first, relevant reactions that take orthologous enzymes in archaea, e.g. for the de novo syn- part in a given metabolic pathway were chosen from the thesisofmevalonate(Barkleyetal.2004;Grochowskietal. complete set of reference reactions downloaded from the 2006b),purines(Graupneretal.2002;Ownbyetal.2005), KEGG ligand database (Kanehisa et al. 2004). In the sec- and cobamide (Woodson et al. 2003; Woodson and ond step, organism-specific metabolic pathways were Table1 Overviewofthecurrentlysequencedhaloarchaea H.salinarumstrains H.marismortui H.walsbyi N.pharaonis NRC-1/R1a Geneidentifier VNG/OE rrn,pNG HQ NP Saltoptimum[M] 4–5M 4.5M 3.3M 3.5M(pH8.5) Isolation Saltedfish DeadSea(Israel) Solarsaltern(Spain) Sodalake(Egypt) Mainresearchinterests Rhodopsins,signal Versatilenitrogen Square-shapedcells, Haloalkaliphilicity, transduction metabolism halomucin respiratorychain Genomesize[Mb] 2.61/2.72 4.37 3.24 2.80 #Plasmids 2/4 8(incl.CHRII) 1 2 %GCchromosome 68.0 62.4 47.9 63.4 rRNAoperons 1 3 2 1 Flagenes(motility) Yes Yes No Yes #Transducergenes 18 21(18) 0 19 #Rhodopsingenes(bop, 4(1,1,2) 6(3,1,2) 3(2,1,0) 2(0,1,1) hop,sop) a TwostrainsofH.salinarumhavebeensequenced,strainsNRC-1(Ngetal.2000)andR1(Pfeifferetal.2008;http://www.halolex.mpg.de). These are virtually identical and differ only in their distribution of insertion elements and their plasmid arrangements. Unless mentioned otherwise,thetwostrainsarenotdistinguishedinthisreview,becausetheyexhibitanalogoussetsofenzymeencodinggenes.Geneidentifiersof H.salinarumstrainR1(e.g.OE1001F)willbeusedthroughoutthetext 123 Extremophiles(2008)12:177–196 179 reconstructed by flagging each reaction existent or non- Enzymeannotationsfromanautomaticassignmentroutine existent in H. salinarum with a certain confidence value. andfromcuratedgenomedata(http://www.halolex.mpg.de) The enzyme gene predicted to catalyze this reaction in were carefully assessed in order to avoid misassignments H. salinarum was linked to the reaction. In case experi- due to domain rearrangements (e.g. purine biosynthesis, mentaldataisavailablefromtheliterature,acommentwas Fig. 2) or overannotation of the exact substrate specifity linked to the reaction or enzyme gene, which includes the (e.g. 2-oxoacid dehydrogenase complex). Novel enzyme PubMed identifier of the external reference. Each reaction variantsclosingformerpathwaygapsinarchaeaaswellas was manually assessed based on automatic enzyme previously published experimental data of H. salinarum assignments derived from various similarity searches (i.e. were also considered throughout the reconstruction pro- blast search with UniProt enzymes, COG assignments, cess. Comparative analysis of the currently sequenced Pfam search), genome data (http://www.halolex.mpg.de), haloarchaea,H.marismortui,H.walsbyi,N.pharaonis,and andexperimentalliteraturedataforH.salinarum.Incase,a H. salinarum, disclosed major metabolic differences, i.e. reaction or enzyme is confirmed by experiments but no alternative enzymes and metabolic pathways employed by genetic evidence was found in the genome (indicating haloarchaea (Supplementary Material S2). novel yet unknown enzyme variants), the reaction was In the following, reconstructed metabolic pathways of marked existent and the conflict was labeled. The meta- halophilicarchaeawillbereviewedindetailwithemphasis bolic data for H. salinarum given in Supplementary on different enzymegenes andmetabolic pathways, which MaterialS1wasinternallystoredinaMySQLdatabaseand lead to the diverse catabolic and anabolic capabilities of managed via a Web-interface. halophilicarchaea.Incaseanenzymegeneispresentinall four haloarchaeal species, only the N. pharaonis identifier (e.g. NP1002A) will be exemplarily mentioned. Comparative analysis of the complete haloarchaeal gene sets Glycolytic pathways For comparison of haloarchaeal gene sets, protein data- bases (fasta format) of the four completely sequenced Major differences in sugar degradation capabilities were haloarchaeal genomes, H. salinarum strain R1, N. phara- identified amongst halophilic archaea that reflect the pre- onis, H.walsbyi, and H. marismortui, were blasted against vious classification of halophilic archaea into eachotherandagainstthenon-redundant(nr)database.For carbohydrate- and non-carbohydrate utilizers (Fig. 1, example, each predicted protein sequence of N. pharaonis Supplementary Material S2). H. salinarum, H. marismor- (species1)wassearchedagainsttheH.salinarumdatabase tui, and H. walsbyi possess all required enzyme genes for (species 2) and the nr database (all species), then the dif- the semi-phosphorylated Entner-Douderoff (ED) pathway ference between the best blast score for the general previously described for carbohydrate-utilizing haloar- database (nr) and for the related species H. salinarum was chaea (Rawal et al. 1988; Danson and Hough 1992; calculated. Listing N. pharaonis sequences by these dif- Verheesetal.2003).However,onlyH.marismortuiandH. ferences in descending order detects genes present in walsbyi encode a classical KDPG aldolase, which is the N. pharaonis but absent in H. salinarum. In few cases, the key enzyme of the pathway, in addition to the novel ar- N. pharaonis gene had a close homolog in the general chaeal KD(P)G aldolase recently described for databasebutonlyamoredistanthomologinH.salinarum. thermophilicarchaea(Ahmedetal.2005).Thus,operation SuchageneisconsideredpresentinH.salinarumwhenthe of different ED pathway variants between H. salinarum blast hit is highly significant (E-value better than e-20) and the other two species is indicated. It should further be wereconsidered.Thedifferentialblastanalysisroutinewas noted that neither the ED pathway nor growth on glucose applied to all combinations of the set of four haloarchaeal as the sole carbon source has been established for Halo- genomes. A list of genes present in only a subset of gen- bacterium so far (Gochnauer and Kushner 1969; Rawal omes was obtained and filtered for metabolic function et al. 1988). N. pharaonis is likely to be incapable of (Supplementary Material S2). glucose degradation, because it completely lacks all enzyme genes of the ED pathway. The carbohydrate-utilizing H. marismortui has acquired Results and discussion a wide range of enzyme genes involved in the uptake and degradation for various sugars. Fructose and sucrose are The metabolism of H. salinarum was reconstructed by likely degraded via the modified Embden-Meyerhof (EM) evaluating metabolic reactions of biochemical reference pathway, which has been described for haloarchaea pathways from KEGG (Supplementary Material S1). (Altekar and Rangaswamy 1992). The pathway involves 123 180 Extremophiles(2008)12:177–196 Entner-Douderoff Embden-Meyerhoff Pentosephosphate pathway pathway pathway KDPG Gluc *Glc AMP Rib1,5P 2 AMPrecycling GAP Pyr *Glc6P Gluc6P Ribul5P Ribul1,5PCO2 PGA 2 ModifiedEMpathway *His,purines, Frc Frc1P *Frc6P Xyl5P Rib5P PRPP pyrimidines Glycerolmetabolism *Frc1,6P2 GAP Sed7P AraHex Glyn Ribulosemono- Glyc GlynP GAP Ery4P Frc6P phosphatepathway Glyc3P Glyc1P PGA Membrane Serfamily lipids (*Ser,Gly,Cys) Pyruvate Ala*,lactate*, metabolism PEP acetate* Aromaticaa Pyr EBranched-chain family 90% aafamily (*Trp,*Phe,Tyr) 10% CO 2 AcCoA Isoprenoids Aspfamily (*phytanylchains) *Asp OA (*Asn,*Thr, Met,ELys) Mal Glyox Icit AcCoA Suc Glufamily 2OG *Glu (*Gln,*Pro,EArg) Respiratory chain Suc Fig.1 The central intermediary metabolism in halophilic archaea. genesarepresentintheH.salinarumgenome(redarrowswithgreen The reaction arrows depict the reconstructed metabolism of the border). Compounds that have been identified through labeling reference species H. salinarum (green reaction exists, red reaction studiesaremarkedbyasterisks.ProposedessentialaminoacidsforH. absent). The four geometric symbols illustrate differences in salinarum are indicated (E). Compounds: AraHex—D-arabino-3- enzyme gene sets between the four sequenced haloarchaea (square: hexulose-6P, Ery4P—erythrose-4P, Frc—fructose, GAP—glyceral- H.marismortui,circle: H.walsbyi,diamond:N.pharaonis,triangle: dehyde-3P, Glc—glucose, Gluc—gluconate, Glyn—glycerone, H.salinarum,greengeneexists,redgeneabsent).Reactionsthathave Glyc—glycerol, Glyox—glyoxylate, Icit—isocitrate, KDPG—2-de- been investigated experimentally through NMR studies or enzyme hydro-3-deoxy-6-phosphogluconate,Mal—malate,OA—oxalacetate, activitytests inH.salinarum arehighlightedbybold arrows(green 2-OG—2-oxoglutarate, PEP—phosphoenolpyruvate, PGA—3-phos- reactionexists,redreactionabsent).Forsomeoftheexperimentally phoglycerate, Pyr—pyruvate, Rib5P—ribose-5P, Ribul5P—ribulose- verified reactions, there are currently no genetic evidences in the 5P, Suc—succinate, Xyl5P—xylulose-5P, Sed7P—sedoheptulose 7- H.salinarumgenome(greenarrowswithredborder).Viceversa,some phosphate,AcCoA—acetyl-CoA reactions have been experimentally excluded, but probable enzyme the key enzyme 1-phosphofructokinase (EC, Embden-Meyerhof pathway and gluconeogenesis rrnAC0342) and sucrose 6-phosphate hydrolase (EC, rrnAC1479). Haloarcula also contains Consistent with the previous biochemical findings (Rawal probable gene clusters for maltose uptake (maltose ABC et al. 1988; Altekar and Rangaswamy 1992), genes for 6- transporter, rrnAC2346-rrnAC2349) and metabolism (sev- phosphofructokinase (EC, the key enzyme of the eral a-glucosidases, e.g. rrnAC0224) as well as a gene classicalEmbden-Meyerhofpathway,areabsentinallfour cluster for D-xylulose oxidation (rrnAC3032–rrnAC3039). haloarchaeal genomes. Alternative archaeal types of 6- ThelatterencodesthepreviouslycharacterizedD-xylulose phosphofructokinasesthatdependonADP(Thermococcus, dehydrogenase (rrnAC3034) (Johnsen and Schonheit Pyrococcus)orPP(i)(Thermoproteus)(Kengenetal.1994; 2004). Selig et al. 1997) as co-substrates have not been found 123 Extremophiles(2008)12:177–196 181 either.However,a1-phosphofructokinasegeneforfructose Sulfolobus ED pathway cluster (Ahmed et al. 2005; Kim degradation via the haloarchaeal variant of the EM path- andLee2005)werefoundinH.salinarum,H.marismortui way (Altekar and Rangaswamy 1992; Rangaswamy and and H. walsbyi, namely genes encoding D-gluconate Altekar 1994a, b) is present in H. marismortui dehydratase (EC, gnaD, COG4948, e.g. (rrnAC0342). Although the upper, hexose part of the OE1664R), KDG kinase (EC, kdgK, COG0524, classicalEMpathwayismissinginhaloarchaea,thelower, e.g. OE1266R), and KD(P)G aldolase (kdgA, COG0329, triosepartofthepathwaythatleadstopyruvateislikelyto e.g. OE1665R). The ED pathway genes, except kdgK, are be functional in all species, as required enzyme genes are located in one gene cluster in the H. salinarum genome, encodedintheirgenomes.Infact,severalenzymeactivities which additionally contains a glucose 1-dehydrogenase have already been proven in H. salinarum (Rawal et al. gene (EC, e.g. OE1669F) (Bonete et al. 1996) as 1988).Thefinalglycolysisstepfromphosphoenolpyruvate well as two conserved genes (e.g. OE1668R, OE1672F), to pyruvate might be catalyzed by pyruvate kinase (EC which are candidates for the yet missing gluconolactonase, e.g. NP1746A) and pyruvate, water dikinase (EC (EC The non-carbohydrate-utilizing strain, e.g. NP1196A) in haloarchaea. In a thermophilic N. pharaonis lacks all enzyme genes for the ED pathway. archaeon,bothenzymesparticipatedinglycolysisbutonly In addition to their archaeal KD(P)G aldolase deletion of the pyruvate, water dikinase gene completely (rrnAC0960, HQ2365A), H. marismortui (rrnAC3121), omitted growth on sugars (Imanaka et al. 2006). H. walsbyi (HQ1495A), and Haloferax alicantei CompletegenesetsforthereverseEMpathway(gluco- (AAB40121, co-located with kdgK, AAB40122) also neogenesis)frompyruvatetophosphorylatedglucosewere encode the canonical bacterial-type KDPG aldolase (EC identified in all four haloarchaeal genomes. Consistently,, kdgA, COG0800), which is absent in Halobacte- labeling experiments (Sonawat et al. 1990; Ghosh and So- rium,Sulfolobus,andThermoplasmaspecies.Thissuggests nawat1998)andenzymeactivitystudies(Rawaletal.1988) operation of different ED pathway variants in the two have confirmed gluconeogenesisin H.salinarum, which is (halo)archaeal groups, and, probably, a ‘reduced’ semi- required to synthesize hexoses for membrane constituents. phosphorylated ED pathway in H. salinarum, which needs For example, glucose was found to be incorporated into to be investigated in future. So far, experiments with glu- different sugar moieties (glucose, mannose, galactose) of cose-grown cells of H. salinarum have shown the halobacterial glycolipids (Weik et al. 1998). Furthermore, conversionfromglucosetogluconate(Sonawatetal.1990; saccharide units are attached to surface proteins of H. sa- Bhaumik and Sonawat 1999), but have not detected sub- linarum such as the S-layer protein and flagellins (Sumper sequent reactions from gluconate to pyruvate and 1987). Sugar moieties of lipids and proteins are likely glyceraldehyde 3-phosphate when applying indirect synthesized via nucleotide sugars. In accordance with this, enzyme activity tests (Rawal et al. 1988). several nucleotide sugar enzymes such as UDP-glucose 4- epimerase(EC5.1.3.2,e.g.NP4662A)andUDP-glucose6- dehydrogenase(EC1.1.1.22,e.g.NP2322A,NP4668A)are Oxidative pentose phosphate pathway presentinhaloarchaealgenomes. Through the oxidative pentose phosphate pathway, phos- phorylatedglucoseisoxidizedtogluconate6-phosphate(C ) 6 Entner-Douderoff pathway andthenconvertedtoribulose5-phosphate(C )byoxidative 5 decarboxylation.Inthearchaealdomainoflife,theoxidative Variants of the classical Entner-Douderoff pathway are PPpathwayseemsnottoexist,becauseenzymegenesforthis commoninthearchaealdomainoflife(DansonandHough pathwayareabsentinarchaealgenomes.However,haloar- 1992). Halophilic archaea such as H. vallismortis (Altekar chaea show glucose 6-phosphate dehydrogenase activity and Rangaswamy 1992) operate the semi-phosphorylated (EC1.1.1.49)inspiteoflackingtherespectiveenzymegene pathway, where glucose is converted to 2-dehydro-3-de- (Aitken andBrown1969)andencodeorthologsof6-phos- oxygluconate (KDG). This intermediate is then phogluconate dehydrogenase (EC, COG1023, e.g. phosphorylated to KDPG and subsequently split into NP0286A).Thus,anoperative,albeitmodifiedoxidativePP pyruvate and glyceraldehyde 3-phosphate. In Sulfolobus pathwayisindicatedforhaloarchaea. andThermoplasma,anon-phosphorylatedEDpathwayhas been described (in addition to the semi-phophorylated pathway variant), where KDG is cleaved without prior Pentose metabolism phosphorylation by a novel bifunctional KD(P)G aldolase (same COG0329 as dihydrodipicolinate synthase) (Ahmed In bacteria, pentoses are synthesized via the non-oxidative et al. 2005). Orthologs for all genes of the characterized part of the PP pathway, where fructose 6-phosphate (C ) 6 123 182 Extremophiles(2008)12:177–196 and glyceraldehyde 3-phosphate (C ) are converted to while RuBisCO (NP2770A) is only encoded in the N. 3 ribulose 5-phosphate (C ) in a complex five-step pathway. pharaonisgenome.Recently,ithasbeenshownthatpurine 5 The enzyme gene set for this pathway, consisting of and pentose metabolism are connected in archaea, as ar- transaldolase 1 and 2 (EC, transketolase chaeal type III RuBisCO is involved in an AMP recycling (EC, and ribulose-phosphate 3-epimerase (EC pathway that is present in N. pharaonis (see below) (Sato, is absent in most archaea except Methanococcus et al. 2007). This novel pathway might be part of a cyclic jannaschii and Thermoplasma spp. (Soderberg 2005). CO fixation pathway in archaea consisting of (i) pentose 2 Individual genes encoding ribose 5-phosphate isomerase formation (PRPP) from fructose 6-phosphate (e.g. via the (EC,e.g.NP0786A)fortheconversionofribose5- RuMP pathway), (ii) conversion of PRPP and adenine to phosphate to ribulose 5-phosphate are present in archaeal AMP by adenine phosphoribosyltransferase (EC, genomes. e.g. NP1254A, NP1426A), (iii) AMP recycling releasing Although Methanocaldococcus jannaschii encodes all adenine and 3-phosphoglycerate (and involving the CO 2 enzyme genes for an operative non-oxidative PP pathway, fixation step), and (iv) gluconeogenesis for the conversion relevant intermediates (i.e. erythrose 4-phosphate, xylose of 3-phosphoglycerate back to fructose 6-phosphate (Sato 5-phosphate, sedoheptulose 7-phosphate) are absent in its et al. 2007). cells (Grochowski et al. 2005). Instead, ribulose 5-phos- phate is exclusively produced through the ribulose monophosphate (RuMP) pathway in M. jannaschii. This Glycerol metabolism pathwayiscommonlyemployedforformaldehydefixation anddetoxification inbacteria butoperatesinreverse mode Inmanyhypersalinehabitats,glycerolisahighlyabundant in archaea and, thus, substitutes for the classical non-oxi- carbon source that is produced by the halotolerant green dative PP pathway (Orita et al. 2006; Grochowski et al. algae Dunaliella to protect itself from osmotic pressure 2005). The RuMP pathway converts fructose 6-phosphate (Borowitzka et al. 1977; Phadwal and Singh 2003). Halo- to D-arabino-3-hexulose 6-phosphate and then to formal- archaea are able to catabolize the abundant glycerol dehyde and ribulose 5-phosphate involving 6-phospho-3- through phosphorylation to glycerol 3-phosphate and sub- hexuloisomerase (COG0794) and 3-hexulose-6-phosphate sequentformationofdihydroxyacetonephosphate(DHAP) synthase (COG0269). These enzymes are encoded in all (Rawal et al. 1988; Nishihara et al. 1999). Consistently, archaea except Thermoplasma and haloarchaea, which haloarchaeal genomes encode glycerol kinase (EC presumably operate the non-oxidative and a modified oxi-, e.g. OE3762R), the multi-subunit glycerol 3- dative PP pathway for pentose formation, respectively phosphate dehydrogenase (EC, e.g. OE3763F- (Soderberg 2005). OE3765F) and a potential sn-glycerol-3-phosphate ABC Absence ofthenon-oxidative branchofthePPpathway transport system (e.g. OE5166F-OE5170F). The haloalka- in most archaea also results in the absence of the pathway liphile N. pharaonis lacks glycerol degrading enzymes, intermediate erythrose 4-phosphate (C ), which is the which might be due to its soda lake habitat, where 4 precursor of aromatic amino acids in bacteria. While a Dunaliella is not a main primary producer. It should be group of archaea (e.g. Pyrococcus abyssi) encodes trans- noted that glycerol degradation and lipid formation occur ketolase required for erythrose 4-phosphate synthesis, through separate pathways in archaea, meaning that the many archaea, amongst them haloarchaea, lack transke- intermediate glycerol 3-phosphate derived from glycerol tolase. These archaea operate an alternative pathway for catabolism is not used for the synthesis of the glycero- aromatic amino acid synthesis starting from different pre- phosphate backbone of archaeal lipids (see below) cursors, 6-deoxy-5-ketofructose-1-phosphate (DKFP) and (Nishihara et al. 1999). L-aspartate semialdehyde (see below) (White 2004). InH.salinarum,glycerolcanalsobeconvertedtoDHA The key enzyme of the reductive branch of the PP by glycerol dehydrogenase (Rawal et al. 1988). For this pathway for CO fixation is ribulose-bisphosphate car- reaction, a plasmid-encoded glycerol dehydrogenase 2 boxylase (EC, RuBisCO). RuBisCO activity was (EC,OE5160F)is employed,which has been char- previouslydetectedinhaloarchaeasuchasH.mediterranei acterized and structurally elucidated (Offermann 2003; but not in H. salinarum (Rawal et al. 1988; Rajagopalan Horn 2006). The produced DHA might be phosphorylated and Altekar 1994). The CO acceptor and substrate of by DHA kinase (EC and fed into the lower 2 RuBisCO, ribulose 1,5-bisphosphate is not directly syn- EM pathway. However, potential DHA kinases genes thesizedfromribulose5-phosphateinarchaea,butfromthe (HQ2672A, HQ2673A) are only encoded by H. walsbyi, purine precursor 5-phospho-D-ribosyl-1-pyrophosphate where they probably depend on a cytosolic phosphoenol- (PRPP)(Finn andTabita 2004). Therequiredgene for this pyruvate-dependentphosphotransferasesystem(HQ2709A) conversion is present in all haloarchaea (e.g. NP5174A), (Bolhuisetal.2006). 123 Extremophiles(2008)12:177–196 183 Pyruvate metabolism and tricarboxylic acid cycle Tricarboxylic acid cycle Pyruvate metabolism Although TCA cycles are highly variable and often incomplete within the archaeal domain of life (Huynen The central metabolite pyruvate is converted to acetyl- etal.1999),haloarchaealgenomesencodethecompleteset CoA by pyruvate-ferredoxin oxidoreductase (EC, ofenzymes.ForH.salinarum,activityofalltheseenzymes porAB, OE2623R/OE2622R) (Kerscher and Oesterhelt has been proven (Aitken and Brown 1969; Hubbard and 1981a, b) and subsequently fed into the tricarboxylic Miller 1972; Kerscher and Oesterhelt 1981a, b; Gradin acid (TCA) cycle. NMR spectroscopy experiments for H. etal.1985)andanoperativeTCAcyclehasbeenshownby salinarum have shown that 90% of the flux is channelled NMR spectroscopy (Ghosh and Sonawat 1998). into the TCA cycle via pyruvate-ferredoxin oxidoreduc- Inhaloarchaea,pyruvateand2-oxoglutaratearenotcon- tase (Ghosh and Sonawat 1998; Bhaumik and Sonawat vertedbyclassical2-oxoaciddehydrogenasecomplexesbut 1994). The remaining 10% of the pyruvate is converted by2-oxoacid-ferredoxinoxidoreductasesencodedbyporAB to the TCA intermediate oxaloacetate by pyruvate car- and korAB genes, respectively (Kerscher and Oesterhelt boxylase (EC, in order to fill up the oxaloacetate 1981a, b). However, haloarchaeal and other archaeal gen- pool of the TCA cycle, when its intermediates are drawn omesfurthercontaingeneclustersencodingallcomponents off for biosynthetic purposes (Ghosh and Sonawat 1998). of a 2-oxoacid dehydrogenase complex. In Thermoplasma However, haloarchaea do not encode archaeal-type acidophilum, it has recently been shown that the E1 com- pyruvate carboxylase (Mukhopadhyay et al. 2000), but ponent of the encoded 2-oxoacid dehydrogenase complex biotin carboxylases that are more likely to be involved in acceptsbranched-chain2-oxoacids(Heathetal.2004).Most fatty acid degradation (e.g. NP4250A/NP4252A located likely,thehaloarchaeal2-oxoaciddehydrogenasecomplexis within a fatty acid degradation cluster NP4230A– alsoinvolvedinbranched-chainaminoaciddegradation. NP4258A). H. marismortui and H. walsbyi encode a When grown on acetate, Haloferax volcanii operates a phosphoenolpyruvate carboxylase (EC, glyoxylate bypass operon involving isocitrate lyase (EC rrnAC0562, HQ3197A), which has been proposed to be and a new type of malate synthase (EC involved with a novel cytosolic phosphotransferase sys- (SerranoandBonete2001).Homologsofbothenzymesare tem (pNG7387-pNG7391, HQ1667A, HQ2709A) present in the N. pharaonis (NP4432A, NP4430A) and (Bolhuis et al. 2006). During gluconeogenesis, phos- H. walsbyi (HQ1720A, HQ3094A) genomes, while only a phoenolpyruvate is synthesized from oxaloacetate probable malate synthase is encoded in H. marismortui through malic enzyme (EC, e.g. (rrnAC1965). Halobacterium lacks both glyoxylate cycle NP0132A, NP1772A) and pyruvate, water dikinase (EC genes, although activity of both glyoxylate cycle enzymes, e.g. NP1196A) in haloarchaea. The former ana- has been demonstrated previously in this haloarchaeon plerotic enzyme has been shown to be active in (Aitken and Brown 1969). H. salinarum, while the anaplerotic reactions catalyzed by phosphoenolpyruvate carboxykinase (EC 49) and oxaloacetate decarboxylase (EC are Nucleotide metabolism missing in haloarchaea (Bhaumik and Sonawat 1994; Ghosh and Sonawat 1998). De novo synthesis of nucleotides Under anaerobic conditions, it has been shown for H. salinarum that pyruvate is primarly converted to ala- The complete gene sets that are required for de novo syn- nine, presumably by an aspartate transaminase, and to a thesis of purines from ribose 5-phosphate and for de novo larger extent to lactate and acetate (Bhaumik and Sonawat synthesis of pyrimidines from carbamoylphosphate and 1994;GhoshandSonawat1998).Inspiteofprovenlactate ribose 5-phosphate are present in haloarchaeal genomes. dehydrogenase activity in H. salinarum cell extracts Haloarchaea reveal an unusual domain fusion pattern of (Bhaumik and Sonawat 1994), no clear lactate dehydro- purine synthesis enzymes (Fig. 2), because they contain a genase homolog has been identified in haloarchaeal unique fusion of GAR and AICAR transformylases (EC genomes.InM.jannaschii,ithasbeenshownthatlactateis, purN/purH, e.g. NP1662A, OE1620R). produced from lactaldehyde, which might be derived from Haloarchaea do not encode the novel AICAR transformy- methylglyoxal (Grochowski et al. 2006a). The M. janna- lase(purP)(Ownbyetal.2005)presentinmostarchaea,but schii lactaldehyde dehydrogenase is similar to several encode the novel archaeal variant of IMP cyclohydrolase probablealdehydedehydrogenasesencodedinhaloarchaea (purO, e.g. NP0732A, OE4329F) (Graupner et al. 2002). (e.g. NP1686A, NP3020A). Thefourhaloarchaealstrainsshowonlyfewdifferencesin 123 184 Extremophiles(2008)12:177–196 Fig.2 Domainrearrangement step2 step5 step3 step9 step10 ofenzymesinvolvedinthede PURD(COG0151) PURM(COG0150) PURN(COG0299) PURH(COG0138) PURH(COG0138) novosynthesisofpurines.The PURT(COG0027) PURP(COG1759) PURO(COG3363) pathwaycomprises10steps human fromPRPPtoIMP.Fusionsof PUR2_HUMAN enzymegenesareindicatedby PUR9_HUMAN linkedboxes.Non-orthologous yeast 66..33..44..1133 enzymesareknownforsteps3, PUR2_YEAST PUR91_YEAST PUR92_YEAST 9,and10(unfilledboxes)but PUR3_YEAST furtherarchaealenzymesfor Escherichia 66..33..44..1133 purinesynthesisareunknown coliK12 PUR2_ECOLI PUR3_ECOLI (questionmarks) PUR5_ECOLI PUR9_ECOLI PURT_ECOLI Methanosarcina 66..33..44..1133 mazei PUR2_METMA Q8PZP9 PUR5_METMA Q8PYG4 Halobacterium 66..33..44..1133 salinarum OE2864F OE1620R OE2292F OE4329F Methanococcus 66..33..44..1133 jannaschii PUR2_METJA PURT_METJA PURO_METJA PUR5_METJA PURP_METJA Archaeoglobus 66..33..44..1133 3.5.?4.10 ? fulgidus PUR2_ARCFU O28464 PUR5_ARCFU O29983 Pyrococcus 66..33..44..1133 ? furiosus PUR2_PYRFU Q8U3M7 PUR5_PYRFU Q8U0R7 their nucleotide metabolism, namely in the occurrence of esters. Specifically, membranes of H. salinarum contain pyrimidine kinases (EC OE3159R, HQ1795, EC core lipids with two phytanyl side chains (C ), and to 20 OE2749F)(Supplementary Material S2). lesser extents also other isoprenoid constituents such as squalenes(C ),phytoenes(C ),menaquinones(C ),and 30 40 40 dolichol (C ) (Oesterhelt 1976; Lechner et al. 1985; Ku- Archaeal type III RuBisCO functions in AMP metabolism 60 shwaha et al. 1976) (Fig. 3). Furthermore, H. salinarum synthesizes several carotenoids from prenyl precursors, Ribulose-bisphosphate carboxylase (EC, RuBis preferentially bacterioruberins (C ) and photoactive reti- CO) is a key enzyme for CO fixation via the Calvin- 50 2 nal (C ) (Oesterhelt 1976; Oesterhelt and Stoeckenius Benson-Bassham cycle in plants. However, RuBisCO was 20 1973). Retinal is incorporated into bacteriorhodopsin and found to be involved in bacterial methionine cleavage and other retinal proteins, which are unique to haloarchaea in the AMP metabolism of archaea (Sato et al. 2007). The withinthearchaealdomainoflife.Althoughfattyacidsare latter pathway recycles the intracellular pool of AMP not part of archaeal membrane lipids, small amounts of produced by ADP-dependent (AMP-forming) sugar kina- fattyacids(C C C )havebeendetectedinmembrane ses, which are involved in glycolytic pathways of archaea 14, 16, 18 proteins of H. salinarum (Pugh and Kates, 1994). Other (Kengenetal.1994;Seligetal.1997).N.pharaonisisthe haloarchaeal species likely possess similar membrane only haloarchaeal species encoding all enzyme genes for constituents as H. salinarum because they have the same this novel AMP recycling pathway, i.e. AMP phosphory- genesetforthedenovosynthesisofisoprenoids.However, lase (NP3958A), ribose-1,5-biphosphate isomerase specific prenyl-based compounds might vary from species (NP3202A),andRuBisCO(NP2770A),whileH.salinarum to species, as in the case of diether core lipids found in exhibits only the isomerase gene (OE3610R) (Sato et al. haloalkaliphiles, e.g. N. pharaonis (C –C and C –C 2007).N.pharaonisandsomeotherarchaeadonotpossess 20 20 20 25 prenyl chains) (Tindall et al. 1984). ADP-dependent sugar kinases that would produce AMP Prenyl side chains of membrane lipids and other iso- though. Instead, AMP recycling might be part of a cyclic prenoids are derived via the mevalonate pathway in pathway for CO fixation as described earlier. 2 haloarchaea(Ekieletal.1986),whiletheglycerophosphate backbone of membrane lipids is formed from glycerol 1- Lipid metabolism phosphate. This membrane precursor is derived from the glycolytic intermediate DHAP via glycerol-1-phosphate Membrane lipids of archaea consist of glycerol diether dehydrogenase (EC, e.g. NP4492A), which is lipids with prenyl side chains instead of diacylglycerol ubiquitously found in archaea (Nishihara et al. 1999). 123 Extremophiles(2008)12:177–196 185 Mevalonatepathway acetyl-CoA acetoacetyl-CoA HMG-CoA mevalonate mevalonate-P C2-unita acetyl-CoA mevalonate-PP BB AA-CO2 isopentenyl-P A -CO2 dimethylallyl-PP(C) isopentenyl-PP(C) 5 5 +IPP HT B tetrahydro- squalene(C30) GPP(C10) +IPP HT +2[2H] HH squalene(C30) presqualene-PP FPP(C15) +IPP HT +IPP HT phytoene(C40) prephytoene-PP HH aGlGltPraPn(sC20) +3[2H] p(Ch2y0t)yl-PP dGiG-trPaPns(C,2c0i)s +4[2H] +IPP HT +[2H] +nIPP HT lycopene(C40) HH GFPP(C25) dihydro- dihydro- phytyl-PP phytyl-PP +2IPP +nIPP HT (C ) (C ) beta-carotene(C ) 20 20 40 retinal(C ),e.g.in bacterioruberines prenylsidechains phytanylchains(C ) dolichol(C ) 20 20 60 bacteriorhodopsin (C )ofthe (C -C )of ofmembranelipids aslipidcarrierin 50 30 45 purplemembrane menaquinones glycoproteinsynthesis Fig.3 Biosynthesis of isoprenoids in halophilic archaea. The between halorarchaea (square: H.marismortui, circle: H.walsbyi, isoprenoid precursor IPPis synthesized via themevalonate pathway diamond:N.pharaonis,triangle:H.salinarum,greengeneexists,red as shown by labeling studies (green reaction exists, red reaction geneabsent).Bacterial-(B)orarchaeal-type(A)enzymevariantsare absent,boldexperimentalverification).Variousisoprenoidsdetected indicated. Superscript ‘‘a’’ indicates C -prenyl units are synthesized 5 inmembranesofH.salinarum(listedinboxes)aresynthesizedbya viathemevalonatepathwaystartingfromtwoacetyl-CoAmolecules seriesofcondensationreactionswithIPP,whichisaddedinhead–tail and a still unknown C -unit arising from amino acid degradation 2 (HT) or head–head (HH) fashion, and through desaturase reactions (Ekieletal.1986) ([2H]).Enzymegenesetsforisoprenoidsynthesisdifferonlyslightly De novo synthesis of isoprenoids (mevalonate pathway) haloarchaea lack the proposed archaeal phosphomevalo- nate decarboxylase gene, and, instead, encode a bacterial- Like other archaea, halophiles synthesize activated C - type diphospho-mevalonate decarboxylase (e.g. 5 units[dimethylallylandisopentenyldiphosphate(IPP)]for NP1580A). Thus, neither the classical bacterial nor the polycondensation of prenyl chains via the mevalonate proposed archaeal mevalonate pathway is complete. For pathway (Fig. 3). A previous comparative analysis of the the last mevalonate pathway step, only N. pharaonis and mevalonatepathway(SmitandMushegian2000)identified H. salinarum encode an archaeal-type II IPP isomerase gapsforthreepathwaystepsinarchaea,namelythelackof (e.g. NP0360A), while all four haloarchaea possess a bacterial-type phosphomevalonate kinase (EC, bacterial-type IPP isomerase (e.g. NP4826A). Future COG3890), diphosphomevalonate decarboxylase investigations are needed in order to clarify whether ar- (EC, COG3407), and IPP isomerase (EC, chaealandbacterialenzymesareemployedsimultaneously COG1443). Recently, an alternative type II IPP isomerase in haloarchaea. The acquired bacterial enzymes might has been identified in archaea, which belongs to the same possibly have a higher substrate specificity or turnover for COG1304 as lactate dehydrogenase (Barkley et al. 2004). covering increased isoprenoid demands for retinal and Furthermore, it has been suggested that mevalonate phos- bacterioruberin biosynthesis. phate is first decarboxylated and then phosphorylated to A functional mevalonate pathway has been verified for synthesize IPP, while phosphorylation of mevalonate H.salinarumbylabelingstudies(Ekieletal.1986),which phosphate precedes the decarboxylation step in bacteria lead tothe proposal ofan unusualfirst step. Lipid labeling (Grochowski et al. 2006b). The archaeal-specific reactions patterns indicated that mevalonate is not synthesized from involve a predicted phosphomevalonate decarboxylase three activated acetate precursors but rather from two (COG1355) and a characterized isopentenyl phosphate acetyl-CoAmoleculesandanunknownC -unit.Thelatteris 2 kinase (COG1608). not derived from acetate but from degraded amino acids Interestingly, haloarchaea may operate a chimeric such as lysine. C -isoprenoid precursors derived via the 5 mevalonate pathway (Fig. 3). Like other archaea, they mevalonate pathway are elongated to trans- and cis-poly- encode isopentenyl phosphate kinase (e.g. NP2852A) prenyl chains in head-to-tail fashion by (E)- and (Z)- insteadofabacterialphosphomevalonatekinase.However, prenyltransferases, respectively (E: NP3696A, NP4556A, 123 186 Extremophiles(2008)12:177–196 NP0604A,Z:NP4550A,NP4544A).Exactchain-specificity A series of genes were assigned for Natronomonas (37 of prenyltransferase orthologs needs to be determined genes) and Haloarcula (29 genes) for the repeated four- experimentally, but the biosynthesis of dolichol and men- step reaction sequence of the ß-oxidation pathway indi- aquinonesfoundinH.salinarumrequiresprenyltransferases cating a versatile fatty-acid metabolism in these two withlong-chainspecifities.Potentialenzymes forsqualene species. In contrast, few ß-oxidation genes are present in andphytoenesynthesisthroughhead-to-headcondensation H. salinarum (12 genes) and H. walsbyi (6 genes). In arealsoencodedinhaloarchaealgenomes. accordance to these findings, growth experiments showed that N. pharaonis is able to grow on fatty acids of various lengthsascarbonsource(especiallyC ),whilefattyacids 14 Synthesis of carotenoids and retinal seem not to be utilized by H. salinarum, as growth is reduced or diminished by long-chain (C –C ) and med- 14 18 Forcarotenebiosynthesis,phytoeneisreducedtolycopene ium-chain fatty acids (\C ), respectively (Konigsmaier 14 by phytoene desaturase that is encoded in haloarchaeal 2006). genomes (e.g. NP4764A, NP0204A). Lycopene is the Haloarchaea likely degrade odd-numbered fatty acids, branching point for the synthesis of b-carotene (C ) and because they possess probable propionyl-CoA carboxylase 40 bacterioruberins(C )(Oesterhelt1976).Thereactionsthat (EC, e.g. NP4250A/NP4252A), methylmalonyl- 50 lead to bacterioruberins have not yet been elucidated in CoA epimerase (EC, e.g. NP1228A) as well as detail, but lycopene cyclase activity (e.g. NP0652A) con- methylmalonyl-CoA mutase (EC, e.g. NP1226A, verting lycopene to b-carotene has been shown for NP2710A). Natronomonas further encodes enzymes that H. salinarum (Peck et al. 2002). The derived b-carotene is areprobablyinvolvedinpropionatecatabolism(NP6212A, cleaved by b-carotene mono-oxygenase into two retinal NP4432A, NP4820A). molecules (C ), which are incorporated in haloarchaeal 20 rhodopsins. H. walsbyi encodes two cyanobacterial-like (HQ2381A, HQ2020A) and one plant-like b-carotene Amino acid synthesis mono-oxygenase homologs (HQ3007A), and H. maris- mortui acquired a cyanobacterial-like b-carotene mono- While certain haloarchaea such as Haloarcula hispanica oxygenasegene(pNG7272)ononeofitsplasmids(Bolhuis (Hochuli et al. 1999) do not require any amino acids for etal.2006).H.salinarumandN.pharaonislackb-carotene growth, H. salinarum is grown in synthetic media with 10 mono-oxygenase homologs and must therefore possess a to15aminoacids(OesterheltandKrippahl1973;Greyand still unknown non-orthologous enzyme for the oxidative Fitt 1976). Through metabolic pathway reconstruction and cleavage of b-carotene. Previously, brp and blh have been comparison, it can be proposed that H. salinarum has shown to play a role in regulation or synthesis of retinal indeed reduced capabilities to synthesize amino acids. (Peck et al. 2001). Specifically, H. salinarum lacks gene clusters for valine, leucine, isoleucine, lysine, and arginine (Supplementary Material S2). Furthermore, because of reduced folate bio- Fatty acid metabolism synthesis, methionine biosynthesis via folate-dependent methionine synthase (EC might be affected in Even-numbered fatty acids like palmitic and stearic acid H. salinarum. The predicted set of essential amino acids are part of membrane proteins but not of membrane lipids fits well to the set of amino acids that can be sensed by in archaea (Pugh and Kates 1994). For example, palmitic H. salinarum (Oren 2002) except for lysine, which is not acidisassociatedwithhalorhodopsinasafreefattyacidin an attractant signal, and for cysteine, which is sensed by H. salinarum (Kolbe et al. 2000). The origin of these fatty BasT (Kokoeva et al. 2002) in spite of the fact that all acidsisunclearbecausearchaeadonotencodecomponents enzyme genes for cysteine biosynthesis are present in its ofafattyacidsynthasecomplex.Inmethanogenicarchaea, genome. The synthesis of several amino acids (glutamate, biosynthesis of fatty-acid-like compounds from 2-oxoglu- glutamine, proline, aspartate, asparagine, alanine, serine, tarate by repeated 2-oxoacid chain elongation has been phenylalanine, tryptophan, histidine) by H. salinarum has reported (White 1989). been verified by NMR labeling studies (Ekiel et al. 1986; For the degradation of activated fatty acids via the ß- Bhaumik and Sonawat 1994; Ghosh and Sonawat 1998; oxidation pathway, gene candidates are present in haloar- Engelhard et al. 1989) (Fig. 1). chaeaandmostotherarchaea.However,sincechain-length For H. marismortui, N. pharaonis, and H. walsbyi, specifityoftheseenzymesiscurrentlyunknown,fattyacid complete independence from supplemented amino acids degradation might still be limited to short chain lengths can be concluded from comparative analysis (Supplemen- (e.g. for derivatives of branched-chain amino acids). taryMaterialS2).Denovosynthesisofallaminoacidshas 123

Abstract. In spite of their common hypersaline environment, halophilic archaea are surprisingly different in their nutritional demands and metabolic
See more

Similar Metabolism of halophilic archaea