* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
Download Globins in Nonvertebrate Species: Dispersal by Horizontal Gene
Transcriptional regulation wikipedia , lookup
Molecular ecology wikipedia , lookup
Gene nomenclature wikipedia , lookup
Gene regulatory network wikipedia , lookup
Biosynthesis wikipedia , lookup
Interactome wikipedia , lookup
Evolution of metal ions in biological systems wikipedia , lookup
Gene expression wikipedia , lookup
Amino acid synthesis wikipedia , lookup
Western blot wikipedia , lookup
Magnesium transporter wikipedia , lookup
Silencer (genetics) wikipedia , lookup
Expression vector wikipedia , lookup
Structural alignment wikipedia , lookup
Genetic code wikipedia , lookup
Protein–protein interaction wikipedia , lookup
Endogenous retrovirus wikipedia , lookup
Protein structure prediction wikipedia , lookup
Artificial gene synthesis wikipedia , lookup
Proteolysis wikipedia , lookup
Biochemistry wikipedia , lookup
Point mutation wikipedia , lookup
Molecular evolution wikipedia , lookup
Ancestral sequence reconstruction wikipedia , lookup
Globins in Nonvertebrate Species: Dispersal by Horizontal Gene Transfer and Evolution of the Structure-Function Relationships Luc Moens,* Jacques VanJEeteren,-f Yves Van de Peer,* Kris Peeters,* Oscar Kapp,$ John Czeluzniak, 5 Morris Goodman, 5 Mark Blaxter,ll and Serge Vinogradofl *Department of Biochemistry, University of Antwerp; TDepartment of Morphology, Systematics and Ecology, University of Ghent; *Department of Radiology and Enrico Fermi Institute, University of Chicago; SDepartment of Anatomy, Wayne State University School of Medicine; IlWellcome Centre for Parasitic Infections, Department of Biology, Imperial College of Science, Technology and Medicine; and (#Department of Biochemistry, Wayne State University School of Medicine Using a new template based on an alignment of 145 nonvertebrate globins we examined several recently determined sequences of putative globins and globin-like hemeproteins. We propose that all globins have evolved from a family of ancestral, approx. 17-kDa hemeproteins, which displayed the globin fold and functioned as redox proteins. Once atmospheric 0, became available the acquisition of oxygen-binding properties was initiated, culminating in the various highly specialized functions known at present. During this evolutionary process, we suggest that (1) high oxygen affinity may have been acquired repeatedly and (2) the formation of chimeric proteins containing both a globin and a flavin binding domain was an additional and distinct evolutionary trend. Furthermore, globin-like hemeproteins encompass hemeproteins produced through convergent evolution from nonglobin ancestral proteins to carry out O,-binding functions as well as hemeproteins whose sequences exhibit the loss of some or all of the structural determinants of the globin fold. We also propose that there occurred two cases of horizontal globin gene transfer, one from an ancestor common to the ciliates Paramecium and Tetruhymena and the green alga Chlumydomonus to a cyanobacterium ancestor and the other, from a eukaryote ancestor of the yeasts Succhuromyces and Cundidu to a bacterial ancestor of the proteobacterial genera Escherichiu, Alculigenes, and Vitreoscillu. Introduction In contrast to vertebrate (V) globins, nonvertebrate (NV) globins exhibit an extensive variability in all aspects of their structures (Vinogradov 1985; Vinogradov et al. 1993). They comprise single chain, one-domain globins, two-domain and multidomain globins, which consist of repeated complete globin units, and chimeric proteins containing a globin domain linked to a flavinbinding protein (Vinogradov et al. 1993). In addition, some of these putative globins particularly from bacteria (Wakabayashi, Matsubara, and Webster 1986; Vasudevan et al. 1991; Potts et al. 1992; Cramm, Siddiqui, and Friedrich 1994), unicellular algae (Couture et al. 1994), protists (Iwaasa, Takagi, and Shikama 1989; Yamauchi, Ochiai, and Usuki 1992; Takagi 1993; Takagi et al. 1993), and yeasts (Iwaasa, Takagi, and Shikama 1992; Zhu and Riggs 1992) may perform functions other than oxygen binding. We have used the alignment of 145 NV globins (Kapp et al. 1995) to construct a new NV template and employ it to evaluate several NV putative globins and globin-like sequences. We also attempt to reconstruct the evolution of the structure-function relaAbbreviations: Hmp, hemeprotein; V, vertebrate; NV, nonvertebrate; Hb, hemoglobin; Mb, myoglobin; UBP, unit evolutionary period. Key words: globins, structure, evolution, horizontal gene transfer. Address for correspondence and reprints: Prof. Dr. L. Moens, Department of Biochemistry, University of Antwerp, Universiteitsplein 1, B-2610 Wilrijk, Belgium. E-mail: [email protected]. Mol. Biol. Evol. 13(2):324-333. 1996 0 1996 by the Society for Molecular Biology and Evolution. 324 ISSN: 0737-4038 tionship of the globin domain, suggest that all extant globin proteins can be derived from a monotypic ancestral globin, and propose that two horizontal gene transfers may have occurred between eukaryote and prokaryote ancestors. Materials and Methods The accession numbers or references for the sequences studied are listed in table 1. The NV globin template (table 2) was constructed as follows. One hundred forty five NV globin amino acid sequences were aligned using the following guidelines: (1) Related globins with known secondary structure were used as references for matching the helical segments. When the appropriate crystal structures were not available, the sperm whale myoglobin (Mb) structure was used. (2) The number of positions assigned to interhelical regions was minimized (just enough to accommodate the longest sequences). (3) The interhelical regions were aligned with each other heuristically, accounting for the preferred conservative amino acid substitutions. The NV globin template was derived as outlined for construction of the Bashford, Chothia, and Lesk (1987) template II with the following modifications. Zero penalty scores were assigned to residues occupying any given position in more than 2% of the sequences. Tyr and Trp were classified as polar residues and given a score of 0.7 when occupying internal positions at frequences below the 2% cutoff value. Resi- Globin Evolution in Nonvertebrate Species 325 Table 1 Species List of Globin Sequences Used in This Study Alcaligenes eutrophus flavoHmp ............. Azorhizobium caulinodans Hmp .............. A. caulinodans MTRC ...................... Aplysia limacina Mb ....................... Bradyrhizobium japonicum Hmp ............. Caenorhabditis elegans Mb ................. Candida norvegensis flavoHmp .............. Casuarina glauca 1egHb .................... Chlamydomonas eugametos Hb L14 10 ........ C. eugametos Hb L1637 .................... Chironomus thumni Hb III .................. Escherichia coli colicin A ................... . coliHmp .............................. Glycera dibranchiata globin MI1 ............. Glycine maxima IegHb ..................... Hansenula anomala Hmp ................... Mastigocladus laminosus phycocyanin OL ...... Nostoc commune Mb ....................... Nostoc sp. Mb ............................ Paramecium caudatum Hb .................. Parasponia andersonii 1egHb ................ Phaseolus vulgaris 1egHb ................... Physeter catodon Mb ....................... Rhizobium meliloti Hmp .................... Saccharomyces cerevisiae flavoHmp .......... Scapharca inaequivalvis HbI ................ Sesbania rostrata 1egHb II .................. Sulculus diversicolor “Mb” ................. Tetrahymena pyriformis Hb ................. Tetrahymena thermophyla Hb ............... Urechis caupo Hb FI ....................... Vitreoscilla sp. Hb ......................... Proteobacteria Proteobacteria Proteobacteria Mollusca Proteobacteria Nematoda Fungi (Deuteromycota) Plantae (Spermatophyta) Plantae (Chlorophyta) Plantae (Chlorophyta) Insecta Proteobacteria Proteobacteria Annelida Plantae (Spermatophyta) Fungi (Ascomycota) Cyanobacteria Cyanobacteria Cyanobacteria Protozoa (Ciliophora) Plantae (Spermatophyta) Plantae (Spermatophyta) Mammalia Proteobacteria Fungi (Ascomycota) Mollusca Plantae (Spermatophyta) Mollusca Protozoa (Ciliophora) Protozoa (Ciliophora) Echiura Proteobacteria dues occurring at surface positions with charge restriction were assigned scores of 0.5 and 0.2 when having the opposite charge or no charge, respectively. Scores of 0.5 and 1.0 were assigned to Val and Trp or Tyr when they occupied these positions at frequencies below the cut-off value. At positions with volume restriction, penalty scores were assigned as to internal positions. Evolutionary trees based on the neighbor-joining method (Saitou and Nei 1987) were constructed using the software package TREECON version 3.0, which was adapted for the analysis of amino acid sequences, including bootstrapping (Van de Peer and De Wachter 1993, 1994). Maximum parsimony analysis was performed using the PHYLIP software package (version 3.5) of Felsenstein (1993). Results and Discussion Putative Globins and Globin-like Sequences The stable folding of a polypeptide strand around the heme group is mainly dictated by the hydrophobic x74334 ACFIXLJ ACNTRYXA A0253 1 BJFIXLJ 211115 X68849 SO0560 X72915 X72916 A0255 1 Ml5691 S15992 A02538 A02560 S55723 M75599 M92437 SO5230 A02563 PO2234 PO2185 RHMFIXLJG A45383 A02535 SO2206 S50084 D13920 D13919 A25537 A02564 Genbank Genbank EMBL PIR Genbank Genbank Genbank PIR Genbank Genbank Genbank Genbank PIR PIR PIR PIR Genbank Genbank (Mulligan PIR PIR PIR PIR Genbank PIR PIR PIR Genbank Genbank Genbank PIR PIR 1993) nature of the residues involved in protein-to-heme and protein-to-protein contacts. Nature evolved a number of protein folds (e.g., cytochrome c, cytochrome b, catalase, etc.) surrounding the heme group, thereby modulating its functional versatility. As based on a number of crystal structures, the globin fold consists of seven to eight helices (A-H) that are connected by short loops and lock up the heme moiety within a three-on-three helical sandwich (Pastore and Lesk 1990; Holm and Sander 1993). There exists considerable tolerance toward both globin size variation and amino acid substitution, however. Alignment of several hundreds of V and NV globins necessitates the use of 182 positions, of which less than half, 84 positions, are common to all the globins (Kapp et al. 1995). Amino acid residues that specify the globin fold are restricted by the hydrophobicity and volume of their side chains. Their restrictions can be described by templates, and the deviation of the standard pattern as penalty scores (Bashford, Chothia, and Lesk 1987). However, the Bashford templates I and 326 Moens et al. Table 2 Nonvertebrate Table 2 Continued Globin Template PENALTY 0.0 SCORE 0.2 PENALTY 0.5 RKDENQHGASCT RKDENQHGASCT ASCTVILM RKDENQHGASCT ILMFYW RKDENQHGASCTP ASCTVILMF PV PV F GYW PV V GASCT V GYW Pattern B-C .. s .. i .. i .. i .. i GASCTPDNVE ASCTVILMF ILMFYW VILMFY ILMF P .. i ASCTVILMFY .. s RKDENQHGASCTP . . req F . . s RKDENQHGASCTP Q V T V G ST HLIMK GYW GASCT GASCW GASCTYW AQ GW V V Pattern E El E2 E3 E4 E5 E6 E7 E8 El0 El1 El2 El5 El8 El9 E20 . . . . . . . . . . . . . . . . . . . . . . . . 0.2 0.5 0.7 Pattern H A4 . . . s A6 . . . s A8 . . . i A10 . . s Al2 . . i Al3 . . s Al5 . . i B6 . B9 . BlO B13 B14 C2...i C4 . C6 . CD1 CD2 0.0 0.7 Pattern A SCORE s s s i s s i i .. s .. i .. i .. i .. i .. i .. s RKDENQHGASCT RKDENQHGASCTP RKDENQHGASCTP GASCTVILMF RKDENQHGASCTP RKDENQHGASCTP PV V V V V VL HQ GASCTDP RKDENQHGASCT ILMF GASCTVILMFYW CTVILMF GASCTVILMFY ASCTVIL RKDENQHGASCT NV EQHILMKR PV V GASCTYW S GAYW W GFYW M PV Pattern F F2 . . . s RKDENQHGASCT F3 . . . s RKDENQHGASCTP F4 . . . i TVILMF F8 . . . req H FlO . . s RKDENQHGASCTP PV V C GASYW V Pattern G FG4 . . i G5 . . . i G8 . . . i GlO . . s G12 . . i G15 . . i G16 . . i G17 . . s G18 . . s TVILM ILMFYW ASCTVILMF RKDENQHGASCTP VILMF ASCTVILMFY VILMF RKDENQHGASCTP RKDENQHGASCT CF V GASYW GASCT GYW V T GASCYW GW GASCYW T V PV H6 H7 H8 H9 Hll H12 H13 H15 H17 H19 . . . . . . . . . . . . . . .. .. .. .. .. NOTE.-s s s i s i i s i s i RKDENQHGASCTP GAS ILMFYW RKDENQHGASCT GASCTVILMF GASCTVILMFYW RKDENQHGASCT TVILMF RKDENQHGASCT GASCTVILMF = Surface position; V CT V DPNVEQHLI GASCT PV PV C GASYW PV i = internal position; YW req = essential residue. II are considerably biased toward V sequences as the aligned set of 226 globin sequences comprised only about 10% NV sequences. Our template is based on 145 NV sequences exclusively (table 2). This template reflects the high tolerance to amino acid substitution seen in NV globin. As a result it is generally less restrictive than the V-based Bashford template II. Some additional stringency was introduced by giving more weight to volume and charge restriction, classifying Trp and Tyr as polar instead of hydrophobic residues, and restricting zero penalty score assignment to internal residues present in no less than 2% of the sequences. Figure 1 compares the scores obtained by the Bashford, Chothia, and Lesk (1987) V and our NV templates. Although the V globins match up well against both templates, the NV template yields superior results with the NV sequences. Several putative globins and globin-like heme proteins were evaluated by the above criteria. Figure 2 represents the block of aligned sequences comprising (A) two representative “true” globins, sperm whale Mb and lupin hemoglobin (Hb), (B) several representative putative globins and (C), several globin-like hemeproteins (Hmp). We also included the alignments of phycocyanin C and colicin A, which share the three-on-three helical sandwich with globins (Pastore and Lesk 1990; Holm and Sander 1993). It can be seen in table 3, that the scores obtained with the group B sequences, although higher than those obtained with the “true” globins, exemplified by the Mb and the plant Hb, are much smaller than those calculated for the group C globin-like sequences. This simply means that the three-dimensional structures of the proteins listed in group C definitely deviate from the globin fold. The penalty scores of the NV globin sequences (table 3, group B) obtained with the nonvertebrate template are generally lower than those obtained by applying the Bashford template II, suggesting a better match Globin Evolution Non-vertebrate a alignment . a I 2 3 4 5 6 7 8 9 10 11 12 13 14 Penalty score New template m Vertebrate Bashford iemplate alignment 120 t g 100 s ; 80 ' $ 60 2 1 40 20 0 0 1 2 3 4 5 New template 6 7 8 9 Penalty score m 10 11 12 13 14 Bashford template FIG. 1.-Comparison of penalty scores of V (Q) and NV (b) globin sequences against the new NV template and the original Bashford, Chothia, and Lesk (1987) template II. of their sequences against the NV template. The relatively greater range in penalty scores obtained with this template reflects the high range of amino acid substitutions at most positions in NV globin sequences. This variability tends to become masked when using the Bashford template due to its weaker performance with NV sequences. As a whole, both templates separate the B and C groups very well. We do expect, however, that the NV template will prove a superior tool for aligning and resolving novel NV globin and globin-like sequences to be detected in the future. It has been argued that the truncated Hbs of the protists Paramecium and Tetrahymena and the cyanabacterium Nostoc, though similar to each other, show no homology to other globins and might be evolved from in Nonvertebrate Species 327 a different ancestral gene (Takagi 1993; Takagi et al. 1993). However, these proteins fit all the essential determinants of the globin fold, and their V and NV templates scores are still comparable to the scores for other NV globins (table 3). Furthermore, the monomeric globin sequences from the bacterium Vitreoscilla, the green alga Chlamydomonas, and the N-terminal globin domains of the chimeric proteins from the yeasts Saccharomyces and Candida and the Proteobacteria Escherichia and Alcaligenes also appear to be “true” globins by the above criteria. In contrast, none of the group C sequences listed in table 3 meets these criteria. The globin-like domains of the chimeric proteins from the mollusc SuZcuZus, the yeast Hansenula, and the Rhizobium spp. can be aligned to have CD1 Phe and F8 His, only at the expense of a considerable increase in gap length. Furthermore, their penalty scores are considerably higher when compared to those of “true” globins (table 3 groups A and B). Phycocyanin C and colicin A share the three-on-three a-helical sandwich with the globins (Pastore and Lesk 1990; Holm and Sander 1993). However, they lack the CD1 and F8 His residues and have high scores with both templates (table 3), in agreement with the earlier suggestion (Holm and Sander 1993) that their three-dimensional structures converged toward a stable three-onthree a-helical sandwich similar to the globin fold. Such convergence is not surprising in view of the increased evidence for a limited number of supersecondary structures (Dorit, Schoenbach, and Gilbert 1990; Chothia 1992; Blundell and Johnson 1993; Orengo et al. 1993). The recent detection of a globin-like fold in diphtheria toxin (Orengo and Taylor 1993) is probably yet another example. Origin and Early Evolution of Globin We hypothesize that redox Hmps existed during the early Proterozoic, approx. 2,000 Mya, well before dioxygen accumulated in the atmosphere (Jenkins 1991). Oxygen-reactive globins likely evolved as an offshoot with the advent of atmospheric oxygen (Keilin 1966, p.95-104; Riggs 1991). Structural similarities with phycocyanin, colicin, and diphteria toxin, but also with hemocyanin, hemerythrin (Volbeda and Ho1 1982), and the cytochromes b2 and b5 (Runnegar 1984) may reflect very remote common ancestry as well as convergence. The origin of globin is rather hypothetical, but stronger evidence exists for the hypothesis that ancestral globin functioned as an oxygen-utilizing enzyme. The reactivity with O2 and its binding are determined by the nature of the amino acid residues surrounding the heme (Perutz 1989). Many V and NV Hbs and Mbs show monooxygenase activity and their ferric derivatives can be reduced by NADH (Ferraiolo and Mieyal 1982; Ferraiolo, 328 Moens et al. 6 7 5 3 4 8 9 1 2 xx12345678901234567-89012345678901234-567890--12345678901-2345678-------------9012345678901234-56789012345678901234567890 10 15 20 15 10 15 1 1 1 5 1 5 1 5 10 15 5 Max xx___-----____~_AAAAAAAAAAAA--________ BBBBB--BBBBBBBBBBB-CCCCCCC-------------CCCC-----DDDDDDD-EEEEEEEEEEEEEEEEEEEEEEEEEE D Mbfld------------NN~-~~a~~--------BBB~--BBBbbB_~BB-CCCcC~C-------------C~CCCC-CCDDDDDDD-EEEeE~e~e~eeE~ee~-----AA DDDDDD-DD (a) Phc GIY Ure Sea Apl Chi Gli LUP 1 0 ------------VLSEGEW-QLVLHVWAKVEA-------DVAGH--GQDILI~S-HPETLEK-------------FDRFKH-LKTEAEMKA-SEDLKKHGVTVLTALGH ------------GLSAAQR-QVIAATWKDIAGAD--------NGAGV--GKDCLI~SA-HP~V-------------FG-FSG--------AS-DP GVAALGAKVLAQIGVAVSHL--GD ------------GLTTAQI-KAIaDHWFLNIKGC-_------LQ~--ADSIFFK~TA-YPGDLAF-------------FHKFSS-V-PLYGLRS-NPAY~~L~~D~~G------PSVYDAAAQLTADVK-KDLRDSWKVIGS---------DKKGN--GV~T~AD-NQETIGY-------------FK~-GNV-S~-M-A-ND~RGHSIT~~QNFIDQL---D ------------SLSAAEA-DLAGKSWAPVFA---------NKN~--GLD~V~EK-FPDS~F-------------FADFKGK--SVADI~-SP~R~S~~E~---~ -------------LSADQI-STvQASFDKvKG-----------D--PVGILYAVFKA_DSKIIGEL--------------VAFTEKQD-ALVSSSFEAFKA---------NIPQY--S~TSILEK-~~DL-------------FSFL----~GVDP-T-NP~TG~~~~~QL~SG -----------GALTESQA-ALVKSSWEEFNA----------NIPKH--TH~IL~EI-APAAKDL-------------FSFLKG--TSEV-PQN-NPELQ~~~~~QLE~G (b) Cd Pha SeS Cas Pan Par TeP Tet CYP CYM Ch4 Ch6 Vit Hmp Ale Sac Can -----------SMNRQEIS-DLCVKSLEGRMVGTEA----QNIEN--GNA~Y~TN-FPDLRVY-------------~G-AEKYT-ADDVKK-SE~DK~~L~HL~-V--YT -----------GAFTEKQE-ALVNSBWEAE'K G-------NIPQY--SVVFYTSILEK-APAAKNL-------------FSFL----ANGVDP-T-NPKLTAS~a~~QLRANG ------------GFTDKQE-ALVNASYEK-------NLPGH--SVLFYSFILEK-EPAAKGL-------------FSFLKD--SDGV-PQN-NPSLQAHAEKVFC;LVHDAG CC) PW CoA Sul Han Rhm Acf Azo Bjf 1 1 1 1 1 1 1 1 1 6 7 3 4 5 8 1 2 0 12345678--------9-012345678901234-5--67890123456789012345---67890-12345678901234567890123456789012xxxxx 15 20 25 1 5 PlO 1 5 10 15 1 5 10 FFFFFFFF--------F-FFFFFFFFFFF--------GGGGGGGGGGGGGG~GGGGG--------HHHHHHHHH~HHHHHHHHHHHHHHHH-----xxxxx P ----------------F-FFffFFFFFFX--FF_F--FF-F--~~G~~~GgGGgg~GX------- ---HHHHHHHhHHhhHHhHHHhXHHHHHHHHHH--------G---GG-G GGGG (a) Phc Gly Ure Sea Apl Chi Gli LUP EAE-------------L-KPLAQSHATKH--K--I--PIKYLEFISEAIIHVLHSR----HPGDF-GADAQGAMNKALELFRKDIAAKYKELGYQG-------EGKMVAQ--------M-KAVCVRHKGYG-NKH-I--KAQISGLQS-------------GNAGAL--------M-KSHDA--MG---I--TP~~QLLKL~~QEEF-SA------DPTTV~~DMG~V~----------------NPDDLVCV--------V-EKHIT----RK-I-_SAA-----------------NAGKMSAM--------L-SQFAKEHVGFG-----V--GSAQFENVRSMFPGFVAS-VAAPP------AGALIIDALKAAGK--------------PNIEAD--------V-NTFVASHKPRG-----V-_THD-------------TW----A--------D-AALGSVHAQKA-------TDPQ---------------VWSDAT--------L-KNLGSVHVSKG-----V--AD~FP~EAILKTIKEWGA----KW-SEELNSA~IAYDE~IVIKKEMDD~---------- (b) Gel Pha SaS Cas Pan Par Tap Tet CYP CYM Ch4 Ch6 Vit MP Ale Sac Can NEEVFKGY--------V-RETINRIIRI--Y-K--M--DPALWMAFFTVFTGnESNLPHV-----AV----VA--------D-ASIHSQKG-----V--NDSQFL~~LK~K~VGD----~-TDELST~E~D~IK~YA------------WV---LA--------D-ASLSVBVQKG----V-_TDPHS------------HAWYDNNT--------L-KRLCSIHLKN---K--I--TDPHFEVMKGALLGTIKEIKE----NW-SDEMGQA~EAYNQLVATIKAEMKE------------KVTVKESD--------L-KRIGAIHFKTG-----V--~E~~RFALLPSST--------NA--WT----------G-RNLICEVHAN--MG---V--SNAQFPTVIGHLRSAGTGA-GV---AAA-LVEQTVAVAET-----------------EHT?mGK_--_____------NH--YK----------G-K~~KG--M-N--L--QNLHFDAIIE~TLKE-LGV-------TDAVIN~ NH__YK_-----_-_-G-Km KG--M-N--L--QNSHFDAIIENLAAl'LKE-LGV-------SDQIIGEAAKVI EHTRKDCLGK---_________--KQ--YG----------G-RPKTHAG--L-N--L--QQPILNK--------------KQ--YG----------G-RPKTHAG--L-N--L--QQPILNK--------------AE--WK----------G-KCMRTAHKD--LVPH-L--TDV~QAW~LSD~E-L---GVT---PGDIAD~~TKTE~~PR~G~SNR-----SE--WK----------G-KCMRTARKD--LVPH-L--SDVHFQA~SD~TE-L---GVP---PEDITD~~STRTE~~P~------------LPAILPA---------V-KKIAVKRCQ-----AG-V--~YPI~Q~L~KEVL---GD~--TDDILDA~~GVI~~IQVEADLYA~VE----LPALLPA---------V-EK~KHTS--F--Q-I--KPEQ~I~E~LA~DEMF----SP---GQEVLDA~~G~ INREAF,IYNENAS----PNSLMAV---------L-KNIANKIIAS--L--G-V--KPEQYPIVGEHLL~KEVL---GN~--TDDIISA~~G~D~GMESELYE~~----LSVLMDH---------V-KQIGHKHRA--L--L--Q-I--KPEHYPIVGE~L~~VL---GD~--TPEIINAWGEAYQAIADIFITVEKKMYEEAL------LTPISGF---------V-NQIVLKHCG--LG---I--KPDQYPWGESLVQAPKQTLIDAEASVYKTLA------ (c) Phy CoA SUl Han Rhm Acf Azo Bjf FIG. 2.-Alignment known crystal structure: of some NV globins and globin-like sequences with globins of known three-dimensional structure. (a) Globins with Physeter catodon (Phc), Glyceru dibrunchiutu MI1 (Gly), Urechis cuupo (Ure), Scuphurcu inequivulvis I (Sea), Aplysiu Globin Evolution Table 3 Penalty Scores Against the V Template of Bashford, Chothia, and Lesk (1987) and the New NV Template Group A ..... Organism/protein NV Template V Template Physeter Mb Lupinus Hb 0.2 2.2 0.5 2.0 8.8 4.4 5.9 7.3 8.3 0.9 3.9 1.2 0.7 3.1 12.8 8.4 10.9 13.1 12.3 6.5 9.8 7.8 7.7 7.8 14.1 18.2 25.0 25.0 20.4 27.0 22.2 22.9 18.7 22.9 29.1 27.6 20.7 31.2 21.9 22.0 B .. ... Paramecium Hb Tetrahymena pyriformis Hb Tetrahymena thermophyla Hb Chlamydomonas 4 Nostoc commune Saccharomyces flavoHb Candida flavoHb Vitreoscilla Hb Escherichia coli HMP Alcaligenes Hmp C ..... Sulculus Mb Hansenula Hmp Rhizobium Hmp Azorhizobium (Acf) Hmp Azorhizobium (MTRC) Hmp Bradirhizobium Hmp Phycocyanin C E. coli colicin A Onady, and Mieyal 1984; Bakan et al. 1991). Structurally engineered cytochrome c in which Met 80 is replaced by Ala, evinces spectral characteristics resembling those of Mb and is capable of reversible O2 binding to the Fe 2+ heme (Wallace and Clark-Levis 1992; Lu et al. 1993). The purple sulfur bacterium Chromatium vinosum has a covalently bound heme able to bind O2 and CO, exhibiting spectroscopic properties similar to ferric Mb (Gaul, Gray, and Knaff 1983). A number of bacterial oxidases and mixed function oxidases, collectively called cytochromes o, are Hmps able to bind CO, CN, and O2 (Jurtshuk and Yang 1980; Poole 1984). The creation of a reversibly oxygen-binding cytochrome c mentioned previously illustrates that a functional shift can arise from minor structural changes in the vicinity of the heme group without any alteration in the protein fold. Thus functional shifts are more readily in Nonvertebrate Species 329 achieved than fundamental alterations of the protein fold. Hence, we assume that globin sequences that exhibit all the determinants of the globin fold (A and B, but not C groups of fig. 1 and table 3) are derived from a direct common monomeric globin ancestor. Evolution of Oxygen Affinity As the concentration of atmospheric O2 increased after the radiation of the photosynthetic bacteria, about 2,000 Mya (Jenkins 1991), specialized proteins evolved to protect cells against oxidative damage, e.g., superoxide dismutase and the various peroxidases, some of which are Hmps (Hassan and Schiavone 1991). Conceivably, ancestral globin was also recruited for oxygen binding and scavenging. This would imply a selective pressure toward enhanced O2 affinity. In the plant 1egHbs and globin from the yeast Candida (Oshino et al. 1973) and the trematode flatworm Dicrocoelium (Di Iorio et al. 1985; Smit et al. 1986) this is achieved by a very high “on” rate. In contrast, globin from the insect Gastrophilus (Phelps et al. 1972) and, particularly, the pseudocoelic globin from the nematode Ascaris (Gibson et al. 1993) have high oxygen avidity resulting from a very low “off” rate. The clam Lucina has two globin species: type I being characterized by a high “on” rate, type II having a slow “off” rate (Kraus and Wittenberg 1990). The simplest conclusion is that high oxygen affinity may have evolved repeatedly in the past. The molecular basis of high oxygen avidity is generally unknown, but in Ascaris pseudocoelic globin it appears to be due to the formation of a second hydrogen bond between O2 and the hydroxyl group of B 10 Tyr, in addition to the usual one formed with E7 Gln (His) (De Baere et al. 1994; Kloek et al. 1994). Because these residues occur in Ascaris muscle globin (Blaxter et al. 1994) and several other NV globins, such as plant Hbs and the chimeric Hbs of the yeasts Saccharomyces and Candida (Oshino et al. 1973; Gibson et al. 1989; Iwaasa, Takagi, and Shikama 1992; Zhu and Riggs 1992), which have much higher “off” rates, other residues are probably needed to position the B helix properly with respect to the heme. c limacina (Apl), Chironomus tumni III (Chi), Glycine maxima (Gli), Lupinus luteus II (Lup). (b) Monomeric globin from Caenorhabditis elegans (Cel), Phaseolus vulgaris (Pha), Sesbania rostrata II (Ses), Casuarina glauca I (Cas), and Parasponia andersonii (Pan), Nostoc commune (CyP) and Nostoc sp. (CyM), and the globin domains of the chimeric flavoHmps from Saccharomyces cerevisiae (Sac), Candida norvegensis (Can), Escherichia coli (HMP), and Alcaligenes eutorophus (Ale). (c) Phycocyanin (Ychain from Mastigocladus laminosus (Phy), the carboxy-terminal fragment of colicine A (CoA) from Escherichia coli, and the globin-like domain in Hmps from Sulculus diversicolor (Sul), Hansenula anomala (Han), Rhizobium meliloti (Rhm), Azorhizobium caulonidans (Acf and Azo, the MTRC gene), and Bradyrhizobium japonicum (Bjf). The characteristics of the globin fold are indicated on top of the block of aligned sequences. The maximum length (182 positions) of the helical segments is indicated as Max. The myoglobin fold (Mb fold) displays the template I of Bashford, Chothia, and Lesk (1987), which consists of 71 positions including 34 interior positions with low, medium, and severe restrictions in size (indicated by lowercase, lowercase boldface, and capital boldface letters, respectively), 32 positions occupied by surface residues (underlined), and 5 positions where any residue is acceptable (boldface X). The distal and proximal residues are indicated by D and I? 330 Moens et al. Evolution of Chime& A Globins A different evolutionary trend is highlighted by the chimeric flavoHbs of bacteria and yeasts (Vasudevan et al. 1991; Iwaasa, Takagi, and Shikama 1992; Zhu and Riggs 1992; Cramm, Siddiqui, and Friedrich 1994), which are part of group B (table 3). The globin and the C-terminal FAD-binding domains show considerable homology to corresponding domains in other members of the group. The flavin-binding domain appears to be homologous to other flavoprotein reductases: NADPH sulfite reductase (Ostrowski et al. 1989), toluate 1,2 dioxygenase (Neidle et al. 1991), cytochrome P450 reductase (Karplus, Daniels, and Herriott 1991), and nitric oxide synthase (Bredt et al. 1991). The HMP protein of E. coli acts as a dihydropteridine and ferrisiderophore reductase (Vasudevan et al. 1991; Andrews et al. 1992). Because O2 binding by the heme moiety limits FAD reduction, HMP could serve as an oxygen sensor, which might regulate the activity of transcriptional regulator proteins (Poole, Ioannidis, and Or-ii 1994), such as the Fnr protein (Spiro and Guest 1991). Vitreoscilla Hb lacks the second domain but interacts with a separate NADPH reductase (Dikshit et al. 1989). Reduced AZcaligenes Hmp binds O2 reversibly and mediates reduction of various dyes and cytochrome c (Cramm, Siddiqui, and Friedrich 1994), which is suggestive of superoxide anion production. Possibly, this Hmp utilizes O2 as a catalyst in the reduction of nitrate or nitrite to nitrous oxide and nitrogen. Vitreoscilla Hb supports aerobic growth in E. coli lacking functional terminal oxidases (Dikshit et al. 1991), and expression of the globin from Saccharomyces is considerably enhanced after disruption of the electron transport system. Thus these proteins can act as alternative oxidases, a function also performed in plants by an unrelated flavoHmp (Kumar and Sol1 1992). The Hmps from the abalone molluscs SuZcuZus diversicolor (Suzuki and Takagi 1992) and Nordotis madaka (Suzuki 1994) function as a Mb but have sequences more similar to human indoleamine 2,3-dioxygenase. Alternatively, linkage of a globin domain to a different protein may lead to progressive alteration of the structural determinants of the globin fold during evolution, as in the case of Hansenula and rhizobial Hmps. Horizontal Gene Transfers in Globin Evolution The evolutionary tree depicted in fig. 3a clearly resolves three separate clusters, one comprising metazoan species and plants, a second comprising ciliates, Chlamydomonas, and Nostoc, and a third combining the two yeasts along with the three Proteobacteria. Essentially the same picture is obtained using a maximum parsimony (PROTPARS) method (fig. 3b). On the contrary, a phylogenetic tree based on SSU t-RNA sequenc- Distance 0.1 H ekgans Candida norvegensis Escherichia coli Aicaligenes eutrophus Saccharomyces cerevestae Vitreoscllla zs4588~ -i-_ Gly-zera dibranchiata I - Distance I Urechis caupo Chmxwnus thumml Scaphsrca insquwalvis Caenorhatditis elegans Aplysla limacma Physeter catodon Pararponla andersonii Casuanna glaucs Lupinw luteus Sesbania rosbats PhaseOlUSwgans Giycine max Tetrahymena tennoph~ia TefIahymeM pyllfon?llS Nostoccommure M Noetoccmmune P Pararnecitm~caudaium Chlamydcmonas rhetnatdii 6 Chlamydornonas rheinardii4 saccharomycss csrs”Ieiae Viiscdla Eschenchta coli Alcaligeros eutmphus Candida “ON~S”SS 0.1 I FIG. 3.-(u) Unrooted neighbor-joining tree constructed from globin sequences including those aligned in fig. 2. The evolutionary distance separating any pair of species is given by summing the lengths of the connecting branches along the horizontal axis, using the scale on top, which represents 10 amino acid replacements per 100 amino acids, corrected for superimposed events. Deletions/insertions were ignored. Bootstrap values higher than 50% are indicated at each node. (b) Consensus tree (PROTPARS) with bootstrap values using the same sequences as in a. One most parsimonious tree (unrooted) needed a total of 2,795 substitutions. The trees were constructed using the Philip 3.5 software package of Felsenstein (1993). (c) Neighbor-joining tree constructed from 16/18S rRNA sequences from the species (or closely related species) also listed in a, as well as a few informative additional data. Interpretation of the tree as in a. Globin Evolution in Nonvertebrate Species 331 A FAD:pyrophos <--_--_-_-_-_-_-_-_> Ale Hmp Sac Can Wos FAD:isoallx <----_-_-_-_-_--_> lQPGGWKGWRTFVIRKKRPESDVITSFILKPAXGPY ~PQ_PpcYVS_-----_ ----VNFEPGQYTSVAIDV--PALGLQQIRQYSLSDKPNGRTYRI *KAGGWKGTRDFRIVAKTPRSALITSFE LEPVDGGAV----AEYRPGQY LGVWIXP--SGFPHQKIRQYSLTRKPDGKGYRIAVKREEGGQ------VS------VKHEMRKNFPAGLVS------l----WPGWKP PEITAKKYVASDIVKFTVKPKFGSGIKLESLPITPGQYITVNTHPIRQKNQYDALRHYSLCSASTKNGLRFA *----WEEPKDERVTKLVKED~~L~~--FKLK--PIIPGEYISFRWDIIHNPDITDIQPREYSISQDVKENEYRISVR--------DIGIVS------lQ~FASLATNEKEKQ~~KG~EYEEWRWCIMPTIVS YHTRDGKGPVHHGVCS NADPIi:rilmse <--_-_-_----_--_---> Ale Hmp Sac Can No8 VHAmRDRLRmAKTYKNLDLFVFY NLLRDRVRVGDQVKLAAP YGSFHIDVD----AKTPIVLISGGVGLTPMVSMLK-VALQAPPRQWFVHGARNSA NWLHNHAMlCDVVKLVAPAGDFFUAVA----DDTPVTLISAGVGQTPMLANLDTUKAGHTAQWWFHAA=GDVHAFAD KVKKLGQSLPRFTAHTWY KYLHKDAKVGDKIKLSAPAGDFAINKELIIiQNKVPLVLLSSGVGVTPLLAKLEKQVKCNPNRPIYWIQSSYD~ ELLAKCANVDKII-EHALKDGKDVKLYYSNRSYQ--SKPFRKFFSNLKKKNNGKFKLN-DYINKKLQVGDI VPVHAPVG--THKYDSISKKG-KVAVLAGGIGITPMIPII SwLNR-IQ~DwPCpV-RGApS--FHLpRNpQvpcILVG~~I~~~~RQ~IQ~~PC~~SKIDHI~~G~ NADPH:adenine <----_-_--_---_-_> Ale _P Sac Can Nos B ----WPLPEDVQGRDYDYP-CLMVRQIEKSILLPDADYYICGPIP~QHD~~IH~I~~PD~~----------------RQPSKADRAKGQFDSE-GLKDLS KLEGAFSDPTHQFYLCGPVGFMQFTAKQLVDLGVKQKNIHYE---------------------------VHTDTE-PLID-AAPLKEKSPA1UDW------------------------NYISAE-~INPDEYD~~~A~KF~YLVGKGVSD--~EF~~DP--------------KL~AYSREPDRPKR~QD~~~~~~HI~~DV-~D~IQRI~~SK~A~IS~D~Y~IFG Distance 0.1 H HMP ALC S NOS FOG. 4.-(u) Alignment of the flavin-binding domain of the flavoHmps from Succharomyces cerevisiue, Cundidu norvegensis, Escherichiu coli, and Alculigenes eutrophus with human nitric oxide synthase (Bredt et al. 1991). The putative NADPH- and FAD-binding sites are indicated. (b) Neighbor-joining tree constructed from these sequences, with the nitric oxide synthase sequence taken as outgroup. Interpretation of the tree as in fig. 3. es of the same or closely related species clearly assigns all bacteria to a single cluster, well separated from the eukaryotes (fig. 3~). This agrees well with the generally accepted idea that eubacteria and eukaryotes have evolved separately for at least 1,800 million years (Doolittle et al. 1986) and possibly much longer. These conflicting lines of evidence can be reconciled by assuming that the Nostoc commune ancestor acquired the globin gene from a common ancestor to Paramecium, Tetruhymena, and Chlamydomonas. If the transferred globin has evolved at the same rate as observed in animals (UEP approx. 5 Myr) the postulated gene transfer occurred approx. 600 Mya. The evidence for globin gene transfer in the yeastProteobacteria cluster (fig. 3~) is even more compelling. A common ancestor to E. coli, A. eutrophus, and Vitreoscilla acquired a copy of the chimeric flavoHmp gene from an ancestor of S. cerevisiae not long after it separated from its earlier common ancestor with C. norvegensis, less than 400 Mya. The flavoprotein domain was subsequently removed in the line leading to the Vitreoscilla. This scenario is compatible with the evolutionary tree based on the flavoprotein domains (fig. 4). The branching order of E. coli, A. eutrophus, Vitreoscilia, S. cerevisiae, and C. norvegensis is not informa- tive as the branches are supported by bootstrapping values that are too low. Lateral gene transfers between an E. coli and a eukaryote ancestor were also reported for cytosolic glyceraldehyde-3-phosphate dehydrogenase (from eukaryote to prokaryote) and aldolase type II (from prokaryote to eukaryote) (Doolittle et al. 1990; Smith, Feng, and Doolittle 1992). In both instances the E. coli sequence clusters with the yeast sequences. Acknowledgments This work was supported by grant no 2.0023.94 from the Belgian Fund for Joint Basic Research. J.R.V. is a Research Director with the National Fund for Scientific Research. S.N.V. is supported in part by grants from the National Institutes of Health (DK 38674 and DK 30382). O.H.K. is supported by a grant from the Department of Energy no. DE-FG02-86-ER60437. LITERATURE CITED ANDREWS, S. C., D. SHIPLEY, J. N. KEEN, J. B. C. FINDLAY, I? M. HARRISON, and J. R. GUEST. 1992. The haemogobinlike protein (HMP) of Escherichia cob has ferrisiderophore reductase activity and its C-terminal domain shares homology with ferredoxin NADP+ reductases. FEBS Lett. 302: 247-252. 332 Moens et al. BAKAN, A., I? SALTMAN, Y. THERIAULT, and l? WRIGHT. 1991. Kinetics and mechanisms of reduction of Cu(I1) and Fe(II1) complexes by soybean leghemoglobin. Biochim. Biophys. Acta 1079:182-196. BASHFORD, D., C. CHOTHIA, and A. A. LESK. 1987. Determinants of a protein fold: unique features of the globin amino acid sequences. J. Mol. Biol. 196:199-216. BLAXTER, M. L., J. R. VANFLETEREN,J. XIA, and L. MOENS. 1994. Structural characterization of an Ascaris myoglobin. J. Biol. Chem. 269:30181-30186. BLUNDELL, T. L., and M. S. JOHNSON. 1993. Catching a common fold. Protein Sci. 2:877-883. BREDT, D. S., I? M, HWANG, C. E. GLATT, C. LOWENSTEIN,R. R. REED, and S. H. SNYDER. 1991. Cloned and expressed nitric oxide synthase structurally resembles cytochrome P450 reductase. Nature (Lond.) 351:714-718. CHOTHIA, C. 1992. One thousand families for the molecular biologist. Nature (Lond.) 357543-544. COUTURE, M., H. CHAMBERLAND, B. ST.-PIERRE, J. LAFONTAINE, and M. GUERTIN. 1994. Nuclear genes encoding chloroplast hemoglobins in the unicellular green alga Chlamydomonas eugametos. Mol. Gen. Genet. 243: 185197. CRAMM, R., R. A. SIDDIQUI, and B. FRIEDRICH. 1994. Primary sequence and evidence for a physiological function of the flavohemoprotein of Alcaligenes eutrophus. J. Biol. Chem. 269:7349-7354. DE BAERE, I., M. E PERUTZ, L. KIGER, M. C. MARDEN, and C. POYART. 1994. Formation of two hydrogen bonds from the globin to the heme-linked oxygen molecule in Ascaris hemoglobin. Proc. Natl. Acad. Sci. USA 91:1594-1597. DIKSHIT, K. L., D. SPAULDING,A. BRAUN, and D. A. WEBSTER. 1989. Oxygen inhibition of globin gene transcription and bacterial haemoglobin synthesis in Vitreoscilla. J. Gen. Microbiol. 1352601-2609. DIKSHIT, R. l?, K. L. DIKSHIT, Y. LIU, and D. A. WEBSTER. 1991. The bacterial hemoglobin from Vitreoscilla can support aerobic growth of E. coli lacking terminal oxidases. Arch. Biochem. Biophys. 293:241-245. DI IORIO, E. E., U. T. MEIER, J. D. G. SMIT, and K. H. WINTERHALTER.1985. Kinetics of oxygen and carbon monoxide binding to liver fluke Dicrocoelium dendriticum hemoglobin, an extreme case? J. Biol. Chem. 260:2160-2164. DOOLITTLE, R. I?, D. E FENG, K. L. ANDERSON, and M. R. ALBERRO. 1990. A naturally occurring horizontal gene transfer from a eukaryote to a prokaryote. J. Mol. Evol. 31: 383-388. DOOLIT~LE, R. E, D. E FENG, M. S. JOHNSON, and M. A. McCLURE. 1986. Relationships of human protein sequences to those of other organisms. Cold Spring Harbor Symp. Quant. Biol. 51:447-455. DORIT, R. L., L. SCHOENBACH, and W. GILBERT. 1990. How big is the universe of exons? Science 250:1877-1882. FELSENSTEIN,J. (1993) PHYLIP: phylogeny inference package, version 3.5. Distributed by the author, Department of Genetics, University of Washington, Seattle. FERRAIOLO, B. L., and J. J. MIEYAL. 1982. Subunit selectivity in the monooxygenase-like activity of tetrameric hemoglobin. Mol. Pharmacol. 21:1-4. FERRAIOLO, B. L., G. M. ONADY, and J. J. MIEYAL. 1984. Monooxygenase activity of human hemoglobin: role of quaternary structure in the preponderant activity of the El subunits within the tetramer. Biochemistry 23:5528-5534. GAUL, D. E, 0. GRAY, and D. KNAFF. 1983. Isolation and characterization of two soluble heme c-containing proteins from Chromatium vinosum. Biochim. Biophys. Acta 723: 333-339. GIBSON, Q. H., R. REGAN, J. S. OLSON, T. E. CARVER, B. DU(ON, B. POHAJDAK, I? K. SHARMA, and S. N. VINOGRADOV. 1993. Kinetics of ligand binding to Pseudoterranova decipiens and Ascaris suum hemoglobins and to Leu-29 + Tyr sperm whale myoglobin mutant. J. Biol. Chem. 268: 16993-16998. GIBSON, Q. H., J. B. WI-I-~ENBERG,B. A. WITTENBERG,D. BoGUSZ, and C. A. APPLEBY. 1989. The kinetics of ligand binding to plant hemoglobins. J. Biol. Chem. 264:100-107. HASSAN, H. M., and J. J. SCHIAVONE.1991. The role of oxygen free radicals in biology and evolution. Pp. 19-37 in C. BRYANT, ed. Metazoan life without oxygen. Chapman and Hall, London. HOLM, L., and C. SANDER. 1993. Structural alignment of globins, phycocyanins and colicin A. FEBS Lett. 315:301-305. IWAASA, H., T TAKAGI, and K. SHIKAMA. 1989. Protozoan myoglobin from Paramecium caudatum. J. Mol. Biol. 208: 355-358. -. 1992. Amino acid sequence of yeast hemoglobin: a two-domain structure. J. Mol. Biol. 227:948-954. JENKINS, R. J. E 1991. The early environment. Pp. 38-64 in C. BRYANT, ed. Metazoan life without oxygen. Chapman and Hall, London. JURTSHUK, I?, and T. Y. YANG. 1980. Pp. 137-186 in C. J. KNOWLES, ed. Diversity of bacterial respiratory systems. CRC Press, Boca Raton, Fla. KAPP, 0. H., L. MOENS, J. R. VANFLETEREN,C. N. A. TROTMAN, T. SUZUKI, and S. N. VINOGRADOV. 1995. Alignment of 700 globin sequences: extent of amino acid substitution and its correlation with variation in volume. Protein Science 4:2179-2190. KARPLUS, I? A., M. J. DANIELS, and J. R. HERRIOTT. 199 1. Atomic structure of ferredoxin-NADP+ reductase: prototype for a structurally novel flavoenzyme family. Science 251: 60-66. KEILIN, D. 1966. The history of cell respiration and cytochromes. Cambridge Univ. Press, Cambridge. KLOEK, A. I?, J. YANG, I? S. MATHEWS, C. FRIEDEN, and D. E. GOLDBERG. 1994. The tyrosine BlO hydroxyl is crucial for oxygen avidity of Ascaris hemoglobin. J. Biol. Chem. 269:2377-2379. KRAUS, D. W., and J. B. WIT~ENBERG. 1990. Hemoglobins of the Lucinia pectinatalbacteria symbiosis. I. Molecular properties. Kinetics and equilibria of reactions with ligands. J. Biol. Chem. 265: 16043-16053. KUMAR, A. M., and D. SOLL. 1992. Arabidopsis alternative oxidase sustains Escherichia coli respiration. Proc. Natl. Acad. Sci. USA 89: 10842-10846. Lu, Y., D. R. CASIMIRO, K. B. BREN, J. H. RICHARDS, and H. GRAY. 1993. Structurally engineered cytochromes with un- Globin Evolution usual ligand-binding properties: expression of Sczccharoc. Proc. myces cerevisiae Met-80 + Ala iso-1-cytochrome Natl. Acad. Sci. USA 90: 11456-l 1459. MULLIGAN, M. 1993. Genes encoding amino acid transporter, cyanoglobin found in NIF region. Cyanonews, 15.02.1993. NEIDLE, E. L., C. HARTNETT,L. N. ORNSTON, A. BAIROCH, M. REKIK, and S. HARAYAMA. 1991. Nucleotide sequences of the Acinetobacter calcoaceticus benABC genes for benzoate 1,2-dioxygenase reveal evolutionary relationships among multicomponent oxygenases. J. Bacterial. 1735385-5395. ORENGO, C. A., T. I? FLORES,W. R. TAYLOR, and J. M. THORNand classification of protein fold TON. 1993. Identification families. Protein Eng. 6:485-500. ORENGO, C. A., and W. R. TAYLOR. 1993. A local alignment method for protein structure motifs. J. Mol. Biol. 233:488497. OSHINO, R., T. ASAKURA, K. TAKIO, N. OSHINO, and B. CHANCE. 1973. Purification and molecular properties of yeast hemoglobin. Eur. J. Biochem. 39:581-590. OSTROWSKI,J., M. J. BARBER, D. C. RUEGER, B. E. MILLER, L. M. SIEGEL, and N. M. KREDICH. 1989. Characterization of the flavoprotein moieties of NADPH-sulfite reductase from Salmonella typhimurium and Escherichia coli. J. Biol. Chem. 264: 15796-15808. PASTORE, A., and A. LESK. 1990. Comparison of the structures of globins and phycocyanins: evidence for evolutionary relationship. Proteins Struct. Funct. Genet. 8: 133-154. PERUTZ, M. E 1989. Mechanisms of cooperativity and allosteric regulation in proteins. Q. Rev. Biophys. 22:139-236. PHELPS, C. E, E. ANTONINI, M. BRUNORI, and G. KELLETT. 1972. The kinetics of binding of oxygen and carbon monoxide to Gastrophilus hemoglobin. Biochem. J. 129:892896. POOLE, R. K. 1984. Bacterial cytochrome oxidases. A structurally and functionally diverse group of electron-transfer proteins. Biochim. Biophys. Acta 726:205-243. POOLE, R. K., N. IOANNIDIS, and Y. 0~11. 1994. Reactions of the Escherichia coli flavohaemoglobin (Hmp) with oxygen and reduced nicotinamide adenine dinucleotide: evidence for oxygen switching of flavin oxidoreduction and a mechanism for oxygen sensing. Proc. R. Sot. Lond. B 255:251258. POTTS, M., S. V. ANGELONI, R. E. EBEL, and D. BASSAM. 1992. Myoglobin in a cyanobacterium. Science 256: 1690-l 692. RIGGS, A. 1991. Aspects of the origin and evolution of nonvertebrate hemoglobins. Am. Zool. 31:535-545. RUNNEGAR, B. 1984. Derivation of the globins from type b cytochromes. J. Mol. Evol. 21:33-41. SAITOU, M., and M. NEI. 1987. The neighbor-joining method: a new method for reconstructing phylogenetic trees. Mol. Biol. Evol. 4:406425. SMIT, J. D. G., H. SICK, A. PETERHANS, and K. GERSONDE. 1986. Acid Bohr effect of a monomeric hemoglobin from in Nonvertebrate Species 333 Dicrocoelium dendriticum. Mechanism of the allosteric conformation transition. Eur. J. Biochem. 155:23 l-237. SMITH, M. W., D. E FENG, and R. E DOOLITTLE. 1992. Evolution by acquisition: the case for horizontal gene transfers. Trends Biochem. Sci. 17:489493. SPIRO, S., and J. R. GUEST. 1991. Adaptive responses to oxygen limitation in Escherichia coli. Trends Biochem. Sci. 16: 310-314. SUZUKI, T. 1994. Abalone myoglobins derived from indoleamine dioxygenase: the cDNA-derived amino acid sequence of myoglobin from Nordotis mad&u. J. Protein Chem. 14: 9-13. SUZUKI, T., and T. TAKAGI. 1992. A myoglobin evolved from indoleamine-2,3-dioxygenase. J. Mol. Biol. 228:698-700. TAKAGI, T. 1993. Hemoglobins from single-celled organisms. Cm-r. Biol. 3:413-418. TAKAGI, T., H. IWAASA, H. YUASA, K. SHIKAMA,T. TAKEMASA, and Y. WATANABE. 1993. Primary structure of Tetruhymena hemoglobin. Biochim. Biophys. Acta 1173:75-78. VAN DE PEER, Y., and R. DE WACHTER. 1993. TREECON: a software package for the construction and drawing of evolutionary trees. Comput. Appl. Biosci. 9: 177-182. . 1994. TREECON for Windows: a software package for the construction and drawing of evolutionary trees for the Microsoft Windows environment. Comput. Appl. Biosci. 10:569-570. VASUDEVAN, S. G., W. L. E ARMAREGO, D. C. SHAW, F? E. LILLEY, N. E. DIXON, and R. K. POOLE. 1991. Isolation and nucleotide sequence of the hmp gene that encodes a haemoglobin-like protein from Escherichia coli K-12. Mol. Gen. Genet. 226:49-58. VINOGRADOV,S. N. 1985. The structure of invertebrate extracellular hemoglobins (erythrocruorins and chlorocruorins). Comp. Biochem. Physiol. 828: 1-15. VINOGRADOV, S., D. A. WALZ, B. POHAJDAK,L. MOENS, 0. H. KAPP, T. SUZUKI, and C. N. A. TROTMAN. 1993. Adventitious variability? The amino acid sequences of nonvertebrate globins. Comp. Biochem. Physiol. 106: l-26. WAKABAYASHI,S., H. MATSUBARA, and D. A. WEBSTER. 1986. Primary sequence of a dimeric bacterial haemoglobin from Vitreoscilla. Nature (Lond.) 322:481-483. WALLACE, C. J. A., and I. CLARK-LEWIS. 1992. Functional role of heme ligation in cytochrome c. J. Biol. Chem. 267:38523861. YAMAUCHI, K., T. OCHIAI, and I. USUKI. 1992. The unique structure of the Paramecium caudutum hemoglobin gene: the presence of one intron in the middle of the coding region. Biochim. Biophys. Acta 1171:81-87. ZHU, H., and A. E RIGGS. 1992. Yeast flavohemoglobin is an ancient protein related to globins and a reductase family. Proc. Natl. Acad. Sci. USA 89:5015-5019. MANOLO GOUY, reviewing Accepted October 2, 1995 editor