CCiittyy UUnniivveerrssiittyy ooff NNeeww YYoorrkk ((CCUUNNYY)) CCUUNNYY AAccaaddeemmiicc WWoorrkkss Publications and Research City College of New York 2002 SSiiggnniifificcaannccee ooff ttwwoo ddiissttiinncctt ttyyppeess ooff ttrryyppttoopphhaann ssyynntthhaassee bbeettaa cchhaaiinn iinn BBaacctteerriiaa,, AArrcchhaaeeaa aanndd hhiigghheerr ppllaannttss Gary Xie Los Alamos National Laboratory Christian Forst Los Alamos National Laboratory Carol Bonner University of Florida Roy A. Jensen CUNY City College How does access to this work benefit you? Let us know! More information about this work at: https://academicworks.cuny.edu/cc_pubs/153 Discover additional works at: https://academicworks.cuny.edu This work is made publicly available by the City University of New York (CUNY). Contact: [email protected] http://genomebiology.com/2001/3/1/research/0004.1 Research Significance of two distinct types of tryptophan synthase beta c chain in Bacteria, Archaea and higher plants o m m Gary Xie*†, Christian Forst*, Carol Bonner† and Roy A Jensen*‡ e n t Addresses: *BioScience Division, Los Alamos National Laboratory, Los Alamos, NM 87544, USA. †Department of Microbiology and Cell Science, University of Florida, Gainesville, FL 32611, USA. ‡Department of Chemistry, City College of New York, New York, NY 10031, USA. Correspondence: Roy A Jensen. E-mail: [email protected] r e v ie Published: 14 December 2001 Received: 24 September 2001 w s Revised: 30 October 2001 GenomeBiology2001, 3(1):research0004.1–0004.13 Accepted: 30 October 2001 The electronic version of this article is the complete one and can be found online at http://genomebiology.com/2001/3/1/research/0004 © 2001 Xie et al., licensee BioMed Central Ltd (Print ISSN 1465-6906; Online ISSN 1465-6914) r e p o r t Abstract s Background:Tryptophan synthase consists of two subunits, (cid:1)and (cid:2). Two distinct subgroups of (cid:2) chain exist. The major group (TrpEb_1) includes the well-studied (cid:2) chain of Salmonella typhimurium. The minor group of (cid:2) chain (TrpEb_2) is most frequently found in the Archaea. d e Most of the amino-acid residues important for catalysis are highly conserved between both TrpE po s subfamilies. it e d r e Results:Conserved amino-acid residues of TrpEb_1 that make allosteric contact with the TrpEa s e subunit (the (cid:1) chain) are absent in TrpEb_2. Representatives of Archaea, Bacteria and higher ar c h plants all exist that possess both TrpEb_1 and TrpEb_2. In those prokaryotes where two trpEb genes coexist, one is usually trpEb_1 and is adjacent to trpEa, whereas the second is trpEb_2 and is usually unlinked with other tryptophan-pathway genes. r e fe r e Conclusions: TrpEb_1 is nearly always partnered with TrpEa in the tryptophan synthase e d reaction. However, by default at least six lineages of the Archaea are likely to use TrpEb_2 as the r e functional (cid:2)chain, as TrpEb_1 is absent. The six lineages show a distinctive divergence within the se a overall TrpEa phylogenetic tree, consistent with the lack of selection for amino-acid residues in rc h TrpEa that are otherwise conserved for interfacing with TrpEb_1. We suggest that the stand- alone function of TrpEb_2 might be to catalyze the serine deaminase reaction, an established catalytic capability of tryptophan synthase (cid:2) chains. A coincident finding of interest is that the Archaea seem to use the citramalate pathway, rather than threonine deaminase (IlvA), to initiate in t the pathway of isoleucine biosynthesis. er a c t io n s Background system for many years and continues to receive attention. In Tryptophan biosynthesis is absent in mammals but is a the current genomics era, emerging surprises reveal there is general metabolic capability of prokaryotes, eukaryotic still much to learn. These surprises include the revelations microorganisms and higher plants. It has been a classic that operons are being organizationally reshuffled, invaded in fo system for elucidation of gene-enzyme relationships and reg- by insertion of apparently unrelated genes, disrupted by rm ulation, thanks largely to the lifelong efforts of Charles either partial or complete dispersal of genes to extra-operon at io Yanofsky [1]. Tryptophan is biochemically the most expen- locations, or complicated by the seemingly unnecessary n sive of the amino acids to synthesize [2]. The clustered orga- presence of additional operon-gene copies located outside of nization of the trp genes into an operon has been a model the operon. 2 GenomeBiology Vol 3 No 1 Xie et al. Tryptophan synthase, catalyzing the final step of tryptophan The status of nomenclature for the tryptophan pathway is biosynthesis, is one of the most rigorously documented not so chaotic as the genes in almost all prokaryotes have examples of an enzyme complex [3-6]. It consists of an been named in line with E.coli, but even here distinct prob- (cid:1)subunit, which cleaves indoleglycerol phosphate to indole lems have arisen because gene fusions that exist in E.coliare and glyceraldehyde 3-phosphate, and a (cid:2) subunit, which often absent elsewhere. Much of the literature up to at least condenses indole and L-serine to yield L-tryptophan. The 1996 used five gene designations for the seven protein (cid:1)(cid:2)(cid:2)(cid:1)complex forms a tunnel into which enzymatically gen- domains [15]. In B.subtilisthese seven protein domains are erated indole is released. The (cid:1) monomers and (cid:2) dimers encoded by separate genes [16]. Thus, E. coli trpD encodes contact one another via highly sophisticated mechanisms of the equivalent of B.subtilisTrpG and TrpD. The E.coligene allostery [7], and it is little wonder that the genes encoding fusion has been re-designated with the convention of a bullet these two subunits are almost always closely linked, fre- to represent a fusion: trpG(cid:3)trpD [17]. Likewise, the E. coli quently being translationally coupled. trpC encodes the equivalent of B. subtilis TrpC and TrpF, and the E. coli gene fusion has been re-designated Against this background, it seems curious that a significant trpC(cid:3)trpF. The nomenclature we advocate is more easily number of organisms possess more than one gene encoding remembered because genes are named in the order of the (cid:2)chain of tryptophan synthase. Usually, but not always, pathway reaction. Subunits for a given reaction are named at the ‘extra’ gene is unlinked to the gene encoding the (cid:1)chain, the same hierarchical level: anthranilate synthase (first reac- and it also defines a distinct subgroup of the (cid:2) chain. This tion) consists of TrpAa and TrpAb, and tryptophan synthase has been recognized in the COGs database [8] as “alternative (fifth reaction) consists of TrpEa and TrpEb. The implemen- tryptophan synthase” (COG 1350). In this paper we present a tation of a logical and consistent nomenclature, as illustrated detailed analysis of the distribution of the two subgroups of in Figure1, should be most helpful in the long term. the (cid:2)chain in prokaryotes within the context of the surpris- ingly dynamic past and ongoing alterations of organization for genes responsible for tryptophan biosynthesis. Results and discussion Phylogenetic tree construction Nomenclature Initial amino-acid alignments were generated using In recent publications involving bioinformatic analysis of aro- ClustalW software, version 1.4 [18]. Manual adjustments matic amino-acid biosynthesis [9-14], we have implemented a were made through visual inspection to bring conserved nomenclature that attempts logical and consistent naming of motifs and residues into register. This was implemented by genes and their gene products for different organisms. The use of the BioEdit multiple alignment tool [19]. Inferences problem is exemplified by the contemporary naming in two about the evolutionary relationships within the TrpEa and model organisms of 3-deoxy-D-arabino-heptulosonate 7-phos- TrpEb protein families were made using the PHYLIP phate (DAHP) synthase, the initial enzyme step of aromatic package of programs [20]. The Protpars program was used amino-acid biosynthesis. In Bacillus subtilis the gene, appro- to generate a maximum parsimony tree, and the neighbor- priately enough, has been named aroA, but in Escherichia coli joining and Fitch programs were used to generate a dis- the equivalent function is represented by genes encoding three tance-based tree. The distance matrix used in the latter differentially regulated paralogs. These E.coligenes have been programs was produced using the program Protdist with a named aroF, aroGand aroH. In B.subtilisthese latter three Dayoff PAM matrix. The Seqboot and Consense programs gene designations refer to 5-enolpyruvylshikimate-3-phos- were then used to assess the statistical strength of the tree phate synthase (step 6 of chorismate synthesis), chorismate using bootstrap resampling. Neighbor-joining (PHYLIP), synthase (step 7) and chorismate mutase (initial step of pheny- Fitch and Margolash (Fitch in PHYLIP), and maximum par- lalanine and tyrosine biosynthesis). Even in B.subtilis, where simony methods [21] all produced trees consistent with one the naming was intended to follow an orderly progression in another. Despite low bootstrap values at many individual terms of order of reaction steps, there is the complication that internal nodes, the clusters formed and arrangement of taxa DAHP synthase is expressed as a fusion of two catalytic within them were largely identical. Ninety TrpEb and 63 domains, one being a class of chorismate mutase called AroQ. TrpEa sequences were analyzed. This requires naming at the level of domain and the desig- nation aroQ(cid:3)aroA was implemented to denote such a Distinctly different types of TrpEb fusion [9-12]. Thus, a single enzymatic function in one TrpEb proteins divide into two distinctly different groups, as organism is accommodated through the cumulative expres- illustrated by the unrooted tree shown in Figure2. The major sion of three paralog genes, but in another organism is only group, denoted TrpEb_1, includes the well-studied enzymes encoded by a portion of a single gene. A universal nomen- from such organisms as E. coli, Salmonella typhimurium, clature is needed that labels at the level of domain, that and B.subtilis. The minor group, denoted TrpEb_2, is repre- labels in synchrony with order of reaction steps as much as sented heavily, but not exclusively, by archaeal proteins. possible, and that labels isofunctional paralogs at the same Among the current inventory of completed archaeal genomes, hierarchical level but with discriminating identifiers. only Methanococcus jannaschii lacks TrpEb_2. Seven http://genomebiology.com/2001/3/1/research/0004.3 TrpAb [TrpG] Gln Glu Pyr c COO–CCH2 NH3 Mg2+ COOO–NHCC2HC2OO– NH2 COO– PRPP PPi P OH2C O–NOHOC omment O COO– 2-Amino-2-deoxy- TrpB OH isochorismate Anthranilate [TrpD] OHOH Chorismate N-(5′-Phosphoribosyl) -anthranilate TrpAa [TrpE] r e v TrpC ie [TrpF] w s CH2 COO– Ser G3P CH2O P CO2 + H2O COO OH OH CH PLP CH CH HO C CH CH r N NH3+ TrpEb N TrpEa N OH OH TrpD NH CH CH2O P epo H [TrpB] H [TrpA] H [TrpC] 1-(o-Carboxyphenylamino) rts L-Tryptophan Indole Indole 3-glycerol -1-deoxyribulose 5-phosphate phosphate Tryptophan synthase d e p Figure1 o s Biochemical pathway of tryptophan biosynthesis. Acronyms that are currently used for the seven monofunctional proteins of Bacillus subtilisare shown in it e brackets below the acronyms used in this paper. TrpAa, large aminase subunit of anthranilate synthase; TrpAb, small glutamine-binding subunit of d r anthranilate synthase; TrpB, anthranilate phosphoribosyl transferase; TrpC, phosphoribosyl-anthranilate isomerase; TrpD, indoleglycerol phosphate e s synthase; TrpEa, (cid:1)subunit of tryptophan synthase; TrpEb, (cid:2)subunit of tryptophan synthase. ea r c h archaeal genomes possess both TrpEb_1 and TrpEb_2 cases where two TrpEb_1 species coexist, where two r e (Methanosarcina barkeripossesses two paralogs of TrpEb_1 TrpEb_2 species coexist, or where TrpEb_1 and TrpEb_2 fe r e in addition to one species of TrpEb_2). Six archaeal genomes coexist in the same organism. A number of organisms e d possess TrpEb_2, but not TrpEb_1. (M. barkeri, Rhodobacter capsulatus, Chlamydia psittaci, r e s and Corynebacterium diphtheriae) possess two copies of e a r Bacterial TrpEb_2 proteins are thus far limited to Aquifex, TrpEb_1. In each case the trpEb_1 copy that is linked to c h Thermotoga, Mycobacterium, Geobacter, Chlorobium and trpEa is highly conserved, whereas the remaining copy has Rhodopseudomonas genera. In addition, one of the multiple diverged to the extent that it may be a pseudogene (Table 2). TrpEb proteins present in the higher plant, Arabidopsis The TrpEb_1 products of these probable pseudogenes have thaliana, belongs to the TrpEb_2 subfamily. In view of the elongated branches (highlighted yellow) on the protein tree in t distinct divergence of TrpEb_2 from TrpEb_1, one might shown in Figure2. er a expect that either TrpEb_2 has lost the ability to interact ct io allosterically with TrpEa, or perhaps that a divergent sub- It is noteworthy that cyanobacterial and higher plant amino- n s group of TrpEa has coevolved with TrpEb_2. Multiple copies acid sequences form a cohesive cluster for TrpEb_1, of TrpEb are often present in genomes. Examples include as shown in phylogram form in Figure 3 (left panel). An in Figure 2 (see figure on the next page) fo Unrooted phylogenetic tree (radial view) of the TrpEb protein family consisting of subfamily TrpEb_1 (top) and TrpEb_2 (bottom). Phylogenetic rm reconstruction of the inferred amino-acid sequence was accomplished by the neighbor-joining method using the PHYLIP program. Organismal acronyms a t are defined in Table 1. For economy of space, a single branch is used to represent proteins that have diverged very recently, for example, TrpEb_1 io n proteins from E.coli and S. typhimurium(Eco/Sty). Archaeal proteins are highlighted in magenta. The detailed order of branching for TrpEb_1 proteins in cyanobacteria and higher plants is shown in Figure3. Probable pseudogenes are shown in yellow. 4 GenomeBiology Vol 3 No 1 Xie et al. Cyanobacteria/ higher plants Cte-1/Det Sau Mlo/Ccr Rca-1/Rpa-1 Smu/Lla Ngo/Nme Spn Bpe/Neu Pae Dra Cps-2 Xfa Bha Bst/Bsu Lpn Mja Gsu-2 Aae-1 Tma-1 Mth-1 Ctr Mba-1/Hal Afu-1 Cps-1 Pab-1/Pfu-1 Cdip-1 Cac Msm-1 Mle Cdip-2 Mbo/Mtu Hin/Aac Pmu Tfu Yps/Ype Eco/Sty Sco Kpn Vch Hpy Mba-2 Bsp TrpEb_1 Cje Rca-2 Ape-1 TrpEb_2 Gsu-1 Paero-1 Tma-2 Mth-2 Tac/Tvo Cte-2 Sso-2 Mba-3 Afu-2 Sso-1 Aae-2 0.1 Fac Pho-2 Ape-2 Pfu-2/ Paero-2 Pab-2 Ath-3 Msm-2 Rpa-2 Figure2(see legend on the previous page) http://genomebiology.com/2001/3/1/research/0004.5 Table 1 Table 1 (continued) Key to sequence identifiers NCBI GI number c o NCBI GI number Species name Acronym trpEb-1 trpEb-2 m m e Species name Acronym trpEb-1 trpEb-2 Prochlorococcus marinus Pma N/A n t Pseudomonas aeruginosa Paeru 12230946 Actinobacillus Aac N/A Pyrobaculum aerophilum Paero-1 N/A actinomycetemcomitans Pyrobaculum aerophilum Paero-2 N/A Aeropyrum pernix Ape-1 7674393 Pyrococcus abyssi Pab-1 14520675 Aeropyrum pernix Ape-2 7674395 Pyrococcus abyssi Pab-2 14423981 Anabaena sp. Asp-1 N/A Pyrococcus furiosus Pfu-1 N/A Anabaena sp. Asp-2 N/A Pyrococcus furiosus Pfu-2 N/A r Aquifex aeolicus Aae-1 6226273 Pyrococcus horikoshii Pho 14591361 ev Aquifex aeolicus Aae-2 7674374 Rhodobacter capsulatus Rca-1 N/A ie w Arabidopsis thaliana Ath-1 136251 Rhodobacter capsulatus Rca-2 N/A s Arabidopsis thaliana Ath-2 1174779 Rhodopseudomonas palustris Rpa-1 N/A Arabidopsis thaliana Ath-3 10176821 Rhodopseudomonas palustris Rpa-2 N/A Archaeoglobus fulgidus Afu-1 3334387 Salmonella typhimurium Sty 136281 Archaeoglobus fulgidus Afu-2 7674372 Staphylococcus aureus Sau 13701169 Bacillus halodurans Bha 10174280 Streptococcus mutans Smu N/A Bacillus stearothermophilus Bst 226585 Streptococcus pneumoniae Spn N/A Bacillus subtilis Bsu 136270 Streptomyces coelicolor Sco 6226276 r Bordetella pertussis Bpe N/A Sulfolobus solfataricus Sso-1 14424473 ep o Buchnera sp. APS Bsp 11182450 Sulfolobus solfataricus Sso-2 13814334 r t Campylobacter jejuni Cje 11269304 Synechococcus sp. Syn N/A s Caulobacter crescentus Ccr 136272 Synechocystis sp. Ssp 2501413 Chlamydia trachomatis Ctr 6226274 Thermomonospora fusca Tfu N/A Chlamydia psittaci Cps-1 N/A Thermoplasma acidophilum Tac 13878841 Chlamydia psittaci Cps-2 N/A Thermoplasma volcanium Tvo 13541762 Chlorobium tepidum Cte-1 N/A Thermotoga maritima Tma-1 1717761 d e Chlorobium tepidum Cte-2 N/A Thermotoga maritima Tma-2 7674388 po Clostridium acetobutylicum Cac N/A Vibrio cholerae Vch 11269279 sit Corynebacterium diphtheriae Cdip-1 N/A Xylella fastidiosa Xfa 11269281 ed Corynebacterium diphtheriae Cdip-2 N/A Yersinia pestis Ype N/A re Dehalococcoides ethenogenes Det N/A Yersinia pseudotuberculosis Yps N/A se Deinococcus radiodurans Dra 7474051 Zea mays Zma-1 1174780 ar c Escherichia coli Eco 136273 Zea mays Zma-2 1174778 h Ferroplasma acidarmanus Fac N/A Geobacter sulfurreducens Gsu-1 N/A Geobacter sulfurreducens Gsu-2 N/A re Haemophilus influenzae Hin 1174785 analysis of TrpEa proteins from the same organisms yields a fe r Halobacterium sp. Hal 14423973 very similar phylogram output (Figure 3, right panel). This ee Helicobacter pylori Hpy 7674399 higher plant/cyanobacteria relationship is pleasingly consis- d r Klebsiella pneumoniae Kpn N/A es tent with the endosymbiotic hypothesis of organelle evolu- e Lactococcus lactis Lla 267168 a r Legionella pneumophila Lpn N/A tion. In each case Prochlorococcus marinus and c h Mesorhizobium loti Mlo 13474230 Synechococcusspecies are the outlying sequence group, with Methanobacterium Mth-1 3334383 the other cyanobacterial sequences (Nostoc punctiformeand thermoautotrophicum Anabaena species) being closer to the higher plant Methanobacterium Mth-2 7674371 thermoautotrophicum sequences from A. thalianaand corn (Zea mays). The order in t Methanococcus jannaschii Mja 2501412 of branching shown is supported by very high bootstrap er a Methanosarcina barkeri Mba-1 N/A values. Zma-3 is the TrpEa protein that has been proposed ct Methanosarcina barkeri Mba-2 N/A io [22] to function independently of a TrpEb partner, produc- n Methanosarcina barkeri Mba-3 N/A s Mycobacterium bovis Mbo N/A ing indole for entry into a pathway other than tryptophan. In Mycobacterium leprae Mle 13093205 this case indole serves as a precursor for a defense metabo- Mycobacterium smegmatis Msm-1 N/A lite that is active against insects, bacteria and fungi. Mycobacterium smegmatis Msm-2 N/A Mycobacterium tuberculosis Mtu 3024761 TrpEa in organisms lacking TrpEb_1 in Neisseria gonorrhoeae Ngo N/A fo Neisseria meningitides Nme 11269306 Six organisms (all Archaea) possess intact tryptophan path- rm Nitrosomonas europaea Neu N/A ways, but they lack TrpEb_1. TrpEb in Thermoplasma vol- at Nostoc punctiforme Npu-1 N/A io canii, T. acidophilum, and Ferroplasma acidarmanus is n Nostoc punctiforme Npu-2 N/A represented only by a single species of TrpEb_2. Although Pasteurella multocida Pmu 13432266 Sulfolobus solfataricus, Aeropyrum pernixand Pyrobaculum 6 GenomeBiology Vol 3 No 1 Xie et al. Table 2 Table 2 (continued) Exceptions to residue invariance Invariant Protein residue† Exceptions: organism (homologous residue) Invariant Protein residue* Exceptions: organism (homologous residue) S184 Sso-1 (T) S203 Fac (T) TrpEa 57P Sso (A) I206 Msm-2 (M) 60D Hal (E) A207 Fac (G) 61G Ctr (N) S209 Ape-1 (A) 64I Mth (V) E210 Sso-2 (D) 183T Ctr (R) G226 Fac (A) 184G Hin (S) N265 Paero-2 (S) 211G Ctr (R) P272 Mba-3 (E) 212F/L Ctr (R) G300 Cte-2 (A) 213G Ctr (D) Paero (A) Ape (S) F/Y325 Mba-3 (H) 234G Ctr (K) F371 Ape-1 (M) TrpEb_1 10G Ctr (H) Cps-2 (E) Cps-1 (Y) P379 Fac (A) 21L Cje (A) S410 Ath-3 (C) 41F Asp-1 (Y) 55R Mth-1 (K) *Residues are numbered according to the S. typhimuriumsequence. 78E Ccr (D) †Residues are numbered according to the P. furiosussequence. 79D Rca-1 (E) Paeru (E) 80L Mth-1 (M) Dra (Q) aerophilum all possess two species of TrpEb, both are the 82H Mba-2 (Q) Dra (F) 94Q Rca-2 (E) TrpEb_2 variety. Thus, in all six of these lineages TrpEa 96L Rpa-1 (M) Spn (W) Bsp (M) either might be unable to form a tight complex with TrpEb_2, 97L Cps-2 (I) or might have evolved different protein-protein contacts. In 102G Bsp (K) the latter case, distinct TrpEa subgroupings might be 114Q Rpa-1 (M) Paeru (M) 124A Zma-2 (R) Cps-1 (G) expected in parallel with the two TrpEb subgroupings. On the 133F/Y Rca-2 (H) contrary, all TrpEa sequences fall into a single group 141R Rca-2 (K) (Figure 4). However, in contrast to sequences present in 146V Mba-1 (A) those Archaea that do possess TrpEb_1 (for example, 149M Cps-2 (I) Archaeoglobus, species of Pyrococcus, Methanococcus, 153G Hal (D) 156V Cje (I) Methanobacterium, Methanosarcina and Halobacterium), 159V Bha (A) the six archaeal lineages that possess only TrpEb_2 have very 162G Cdip-2 (E) distinctive elongated branches on the TrpEa tree (Figure4). 167K Cdip-2 (S) Sau (S) This suggests an elevated rate of evolutionary divergence, due 173A Hal (T) Rca-2 (C) 177W Cps-2 (F) Rca-2 (Y) either to selection for new productive contacts of TrpEa with 186Y Cps-2 (F) Rca-2 (F) TrpEb_2 or to lack of constraint to maintain TrpEa residues 208I Xfa (V) previously important for contacts with TrpEb_1. 213R/K Bpe (L) 224P Cje (V) 225D Bha (T) Rca-2 (A) The long branch of the TrpEa sequence of Chlamydia tra- 267H Rca-2 (N) chomatis reflects its likely status as a pseudogene. This is 269A Sau (L) Mba-1 (S) consistent with the observation that C. trachomatisTrpEb_1 274G Aac (A) (Figure 2) also seems to be a pseudogene. One does not 299S Rca-2 (T) expect positive selection for maintenance of function in 310G Mba-2 (S) 345I Rca-2 (V) C. trachomatis as it lacks an intact tryptophan pathway. 383D Yps (E) Indeed, the alteration in C. trachomatis of many otherwise invariant amino-acid residues is evident from the informa- Invariant tion given in Table 2. Protein residue† Exceptions: organism (homologous residue) TrpEb_2 P10 Paero-1 (gap) Overview comparison of TrpEb_1 and TrpEb_2 Y14 Aae-2 (L) Figure5 shows an alignment of the amino-acid sequence of E54 Sso-1 (Q) TrpEb_1 from S. typhimuriumwith TrpEb_2 from P. furio- L86 Msm-2 (F) sus. Each sequence is shown as a template for its own sub- E87 Paero-2 (D) E101 Paero-2 (Q) family, as extracted from a refined multiple alignment. S108 Paero-2 (N) Conserved residues deduced from a full multiple alignment A114 Mba-3 (S) (available from the author on request) are indicated, as are W138 Sso-1 (R) the gap positions present in the full alignment. Functional G176 Fac (N) roles in catalysis and allosteric regulation are indicated for http://genomebiology.com/2001/3/1/research/0004.7 Pma Pma 963 Syn 989 Syn Ssp c 1000 994 NpSus-2p 895 999 964 ANsppu-2-2 omm 998 898 e Asp-2 Asp-1 n 953 1000 t Npu-1 Npu-1 926 983 978 Asp-1 Ath-1 999 Ath-1 Ath-2 1000 999 Ath-2 Zma-1 1000 786 Zma-2 Zma-2 1000 1000 Zma-3 Zma-1 r e v ie TrpEb_1 TrpEa w s 0.1 Figure3 Unrooted phylogenetic tree (phylogram view) of cyanobacterial and higher plant TrpEb_1 (zoom-in expansion from Figure2) and TrpEa (zoom-in expansion from Figure4) protein sequences. The higher-plant lineage is shown in green. Bootstrap values (from 1,000 replications) supporting the order of branching shown are given at the nodes. r e p o r t the S. typhimuriumTrpEb_1 sequence in order to compare G181 of TrpEa. Competing allosteric conformations are s similarities and differences between TrpEb_1 and TrpEb_2 mediated by alternative salt bridges between K167 of proteins. Residues that are ligands of pyridoxal phosphate or TrpEb_1 and D305 of TrpEb_1 on the one hand, or between that interact with pyridoxal phosphate are scattered K167 of TrpEb_1 and D56 of TrpEa, on the other. When throughout the sequences, including the catalytic K87, and D305 of TrpEb_1 is not occupied with K167, it forms an d e are highly conserved. Residue E109 has been shown to alternative salt bridge with R141, as shown in Figure5. po s render indole more nucleophilic via proton abstraction from it e d N1 [23]. The serine substrate-binding region is highly con- As intersubunit signaling between TrpEb_2 and TrpEa is r e served, as is a monovalent cation (MVC) binding region [7] either lacking or involves different contacts, one might se a coordinating with G232, F/Y306, and G/A/S308. A number expect the important catalytic residues, but not the allosteric r c h of indels (insertions/deletions) distinguish TrpEb_1 and residues, to be conserved in comparison of TrpEb_2 with TrpEb_2, and TrpEb_2 is about 50 residues longer overall TrpEb_1. This comparison is shown in Figure5. Likewise, in than TrpEb_1. In addition to other residues conserved those TrpEa proteins that lack a TrpEb_1 partner, instead r e between both TrpEb_1 and TrpEb_2, each subgroup has its being forced to function in concert with TrpEb_2, one might fe r e own repertoire of uniquely conserved residues. The COMM expect retention of catalytic residues but loss of allosteric e d domain [24], a rigid but mobile domain as originally defined residues. This comparison is shown in Figure6. Of the inter- r e s with S. typhimurium TrpEb_1 [25], differs from the corre- subunit signaling pair TrpEb_1 S178/TrpEa G181, S178 is e a r sponding region of the TrpEb_2 subfamily by the presence not conserved within its own subfamily and equivalent c h of an indel. Key TrpEb_1 regulatory residues (R141, K167) residues seem to exist [26]. It is, however, striking that G181 within this region as well as one residue near the MVC site is invariant in all TrpEa proteins, except for those belonging (D305) are not conserved in the TrpEb_2 subgroup. to the six archaeal lineages lacking TrpEb_1. It is even more striking that K167, which participates in the alternative (cid:2)-(cid:2) in t Loss of intersubunit contacts in TrpEa-TrpEb_2 or (cid:2)-(cid:1) salt bridges is invariant in TrpEb_1, but absent in er a systems TrpEb_2. Likewise, the salt-bridge partner D305 is invariant ct io The tryptophan synthase of S. typhimurium is a rigorously in TrpEb_1, but absent in TrpEb_2. The salt-bridge partner n s documented example of substrate channeling in which D56 in TrpEa is invariant except for the six archaeal lineages indole generated as an intermediate is passed through an that lack TrpEb_1. Figure5 shows that TrpEb_2 sequences internalized tunnel ([7] and references therein). Ligand carry an insertion of 16 residues in the general region corre- binding at the (cid:1)-site and covalent transformations at the sponding to the COMM domain of TrpEb_1 proteins. (cid:2)-site accomplish mutually reinforcing overall allostery. The in fo movable COMM domain is comprised of residues G102 to Tryptophan gene organization in TrpEb_2-containing rm G189 in S. typhimurium TrpEb_2. COMM interacts with genomes at both the (cid:2)-active site and with (cid:1)-subunit loops 2 and 6 in The organization of tryptophan-pathway genes in the ten ion response to allosteric signals. Within the COMM domain of archaeal genomes and six bacterial genomes that are thus far TrpEb_1, S178 participates in intersubunit signaling with known to possess TrpEb_2 are displayed on a 16S rRNA tree 8 GenomeBiology Vol 3 No 1 Xie et al. Tvo Ape Paero Ctr Tac Cte Cps Dra MsmMle Mtu/Mbo Fac Det Sco Tfu Xfa Gsu Neu Sso Nme/Ngo Paeru Tma Rpa Cac Rca Aae Smu Mja Spn Mba Sau Hal Mth Cje Bsu Bha 0.1 Hpy Cyanobacteria and higher plants Cdip Vch Pfu Ype Bsp Eco/Sty Pab Afu Aac Hin Pmu Figure4 Unrooted phylogenetic tree (radial view) of the TrpEa protein family. Organismal acronyms are defined in Table 1. Phylogenetic reconstruction of the inferred amino-acid sequence was accomplished by the neighbor-joining method using the PHYLIP program. The detailed order of branching for cyanobacterial and higher-plant sequences is shown in Figure3. Archaeal proteins are visualized in magenta, and the yellow line indicates the TrpEa protein from C. trachomatis, which is probably encoded by a pseudogene. in Figure 7. Pyrococcus horikoshii has lost the entire species of Thermoplasma possess an otherwise intact trp pathway, and only trpEb_2 is present. In Aquifex and operon that lacks trpEb_1 (also not present elsewhere in the Chlorobium all trp genes are dispersed, and trpEa is not genome). Hence, by default it appears that the unlinked linked to either trpEb_1 or trpEb_2. In Bacteria and trpEb_2 must function for tryptophan biosynthesis in these Archaea the gene order trpEb_1(cid:4)trpEais one of the most organisms. In A. pernix and S. solfataricus all of the trp highly conserved of genomic gene arrangements. Often genes are adjacent, but trpEb_2flanks trpEa, and trpEb_1is translational coupling exists, as seen in Figure7 for P. furio- absent from these genomes. In each case, a second unlinked sus, P. abyssi, M. thermoautotrophicum, T. volcanium, copy of trpEb_2 is present. In Geobacter sulfurreducens, Archaeoglobus fulgidus and T. maritima. In each of the both trpEaand trpEb_1(presumably partnered based on the latter genomes trpEb_2 is outside the trp operon. Both results shown in Figure4) are unlinked to one another, and http://genomebiology.com/2001/3/1/research/0004.9 F/Y Y/F Sty 1 MTTLLNPYFG---EFGGMYVPQILMPA------LNQLEEAFVRAQKDPE--------FQAQFADLLKNYAGRRP TALTKCQNITAGTR-------- 70 c Pfu-2 1 MKVVLPDGRIPRRWYNILPDLPEPLAPPLDPET-NEPVDPKKLERIFAKELVKQEMSTKRYIKIPEEVRKMYSKIG-RRP TPLFRATNLEKYLNTP------ 93 o m m e A / S 8 6 8 7 Q / N G / A 1 0 9 1 1 0 1 1 4 Y/FC omm d1o4m1a in nt Sty 71 ---TTLYLKK REE DLLHGGG AHHKKT NN QVLGQALLAKRMGKSEIIAEETTGGAAGGQ HGGV ASALASALLGLKCRIYMGAKDVERR QSPNVFRMMR LMGAEVIPVHS-----GS 163 Pfu-2 94 ---ARIYFKK FEE GATVTGG SHHKKI NNT ALAQAYYAKKEGIERLVTEETTGGAAGGQ WGGT ALSLAGALMGIKVRVYMARASYEQKPYRKVLMMR IYGG AEVFPSPSENTEIGK 191 Salt bridge D-56 G-181 S/T-190 167 178 232-236 F/Y re v Sty 164 A-----------TLL KKD AA CNEALRDWSGSYETAHYYM LGG TAAGPHPYPTIVREFQRMIGEEE TKAQQI LDKEG--RLPPD AVIACVGGGGGGSSNN AIGMFADFINDTS-- 249 ie w s Pfu-2 192 R-FLSENPNHPGSLL GIAA ISEAIEDVLKDE-KARYYS LGG SVLN---H---VLMHQTVIGLEE AKQQQM EEFEE----PP DVIIGCVGGGGGGSS NN FAGLA--YPFVKEVL 278 Salt bridge M+ G / A 30330 5 F/Y-306 G/A/S-308 Sty 250 --------------VGLIGVEPGGHGIETGE------HGAPLKHGR--VGIYFGMKAPMMQTADGQIEESYSISAGGLL DDFFP SS VGPP QHAYLNSIGRADYVSIT 328 Pfu-2 279 DG---------DNEYEFIAVEPKAAPSMTRG------VYT-YDFGDS-GELTPKLKMHTLGHR---YHVPPIHAGGGLL RYYH GG VAPP TLSVLVNNGIVKPIAYH 359 r e p or ts E-350 377 378 Sty 329 DDEALEAFKTLCRHEEGG IIPAA LEES SHHAA LAHALKMMREQPEK--EQLLVVNLSGG RGG DKDDI FTVHDILKARGEI 397 Pfu-2 360 QTEVFEAAALFAKLEEEG IVPAA PEES AHHAAIKATIDKAIEAKREGKEIVILFNLSGG HGGL LDDL HGYEEYLEGRLQDYEPKDLPISNPLNPKP 446 E / Q Indole PLP Serine Metal dep o s it e d Figure5 r e Alignment of TrpEb_1 from S. typhimurium(Sty) and of TrpEb_2 from P. furiosus(Pfu-2). The sequences are shown as they appear in a comprehensive s e alignment, including gaps. Invariant residues in each subfamily are highlighted and near-invariant residues (delineated in Table 2) are shaded. Invariant or ar c near-invariant residues common to both subfamilies are boxed. Residues shown in Sty to be relevant to metal coordination or to the binding of indole, h serine or PLP are color-coded as indicated at the bottom. r e fe r e both are outside an otherwise intact trpoperon. In this case it because of limitation of precursors needed for L-phenylala- e d seems curious indeed that the operon contains trpEb_2. nine and L-tyrosine biosynthesis. This is because no other re s genes encoding DAHP synthase appear to be present in the e a r A snapshot of the incredibly dynamic alteration of trypto- genome. It would be interesting to know whether the Ferro- c h phan gene organization in prokaryotes is apparent from plasma DAHP synthase is sensitive to allosteric control or Figure7. Of the organisms shown in Figure7, the most con- not. The phenomenon of growth inhibition triggered by sistent linkage is that of trpAa and trpAb. In A. fulgidus exogenous amino acids is exemplified by the effect of exoge- trpD and trpB are fused, whereas in Rhodopseudomonas nous L-phenylalanine upon DAHP synthase in Neisseria in t palustris trpAaand trpAbare fused. Operons are sometimes gonorrhoeae [27] and in other organisms ([27] and refer- er a incomplete as with P. aerophilumand G. sulfurreducens, or ences therein). ct io fragmented, as with R. palustris. In A. pernix an inverted n s arrangement yields a divergent transcription of trpEa (cid:4) What is the function of TrpEb_2? As previously discussed, it trpEb_2(cid:4)trpC (cid:4)trpDand trpB(cid:4)trpAa(cid:4)trpAb. This is seems clear that TrpEb_2 has sometimes been pressed into also one of the very few cases where the order is trpEa (cid:4) service with TrpEa to function as tryptophan synthase. What trpEbinstead of the usual trpEb (cid:4)trpEa. might be its function in those situations where trpEb_2 is isolated away from a typical tryptophan operon, which pos- in fo The Ferroplasma genome illustrates a case where a non- sesses closely linked or translationally coupled genes speci- rm tryptophan pathway gene, aroAencoding DAHP synthase, is a fying trpEaand trpEb_1? at io member of the operon (translationally coupled). The implied n transcriptional control of DAHP synthase by L-tryptophan TrpEb_1 from S. typhimurium is the prototype member potentially could produce growth inhibition by L-tryptophan of a superfamily of pyridoxal phosphate (PLP)-dependent