Vol. 159, No. 1JOURNAL OF BACTERIOLOGY, JUlY 1984, P. 57-620021-9193/84/070057-06$02.00/0Copyright C 1984, American Society for Microbiology

Nucleotide Sequence of Escherichia coli pabB Indicates a CommonEvolutionary Origin of p-Atninobenzoate Synthetase and

Anthranilate SynthetasePAUL GONCHAROFF AND BRIAN P. NICHOLS*

Department of Biological Sciences, University of Illinois at Chicago, Chicago, Illinois 60680

Received 16 January 1984/Accepted 13 April 1984

Biochemical and immnunological experiments have suggested that the Escherichia coli ,enzyme p,7amiiloben-zoate synthetase and anthranilate synthetase are structurally related. Both enzymes are composed of twononidentical subunits. Anthranliate synthetase is composed of proteins encoded by the genes trp(G)D and trpE,whereas p-aminobenzoate synthetase is composed of proteins encoded by pabA and pabB. These two enzymescatalyze similar reactions atd produce similar products. The nucleotide sequences of pabA and trp(G)D havebeen determined and indicate a common evolutionary origin of these two genes. Here we present the nucleotidesequence ofpabB and compare it with that of trpE. Similarities are 26% at the amino acid level and 40% at thenucleotide level. We propose that pabB and trpE arose from a common ancestor and hence that there is acommon ancestry of genes encoding p-aminobenzoate synthetase and anthranilate synthetase.

Chorismate is a common precursor in the biosynthesis ofp-aminobenzoate (PABA) and anthranilate. PABA is a com-ponent of the vitamin folic acid, whereas anthranilate is anintertnediate in the tryptophan biosynthetic pathway. Thesetwo specific reactions involving chorismate are similar andyield similar products. Both of these reactions utilize choris-mate and glutamine to produce pyruvate, glutamate, andeither PABA or anthranilate. The structure of PABA andanthranilate differ only by the position of the amino group onthe benzene ring. The enzymes that catalyze these tworeactions, p-aminobenzoate synthetase (PAI3S) and anthra-nilate synthetase (AS), both consist of nonidentical subunits,component I (CoI) and component II (CoIl). In PABAsynthesis, PABS CoI cannot function with AS Coll. Similar-ly, in anthranilate synthesis, AS CoI cannot function withPABS Coll. This is inferred by the inability of wild-type trpgenes to complement pabA and pabB mutants and theinability of pab genes to complement trpE and trp(G)Dmutants (27). Nevertheless, it has been suggested that Esch-erichia coli PABS and AS are structurally related sinceantibodies raised against AS cross-react with fractionatedextracts containing PABS (19).

In each synthetase complex, Col alone can bind choris-mate and ammonia to form the aromatic product (8; S.Doktor and B. P. Nichols, unpublished results). CoII con-tains a glutamine amidotransferase activity whose function isto transfer the amide group from glutamine to Col. E. coliPASS CoI and CoIl are encoded by two unlinked genes,pabB and pabA, respectively (5, 6). AS CoI and Coll of E.coli are encoded by two linked genes, trpE and trpD,respectively (26). trpD encodes a bifunctional protein thatcontains the glutamine amidotransferase and the anthranilatephosphoribosyl transferase of the tryptophan pathway (26).In Setratia marcescens, these two enzymes are encoded byseparate genes, trpG and trpD, respectively (15). To avoidconfusion, we will use trp(G) to refer to that portion of thetrpD gene that encodes the glutamine amidotransferaseactivity of E. coli AS CoIl (2).

Recently, common evolutionary origins of several genepairs have been reported (3, 11, 25). Gene duplications

* Corresponding author.

increase the genetic potential of an organism (9) since after agene duplication event, one of the genes is free to respond toselective pressures which may result in an altered function.It has been hypothesized that E. coli pabA and trp(G) arosefrom a common ancestor via a gene duplication event, andnucleotide sequence comparisons support this view (10).This work investigates the evolutionary relationship of theCoI subunits of PABS and AS encoded by E. coli pabB andtrpE.

MATERIALS AND METHODSBacterial strains. E. coli AB3303 (pabB3 thi-J his4 argE3

lacYJ galK2 xyl-5 mtl-l rpsL-704 tsx-29 or tsx-358 supE44)(6) was provided by B. Bachmann. E. coli JM103 (13) andphages M13mp8 and M13mp9 (14) were obtained from NewEngland Biolabs.Enzymes and DNA manipulations. Restriction endonucle-

ases were, purchased from New England Biolabs, BoehringerMannheim, or P-L Biochemicals, Inc., or were prepared inthis laboratory by published procedures. E. coli DNA poly-merase I (Klenow fragment) was purchased from BoehringerMannheim. Plasmid DNA was prepared by the Birnboim andDoly method (1). Transformations of bacterial cells withplasmid DNA were performed as described by Mandel andHiga (12).DNA sequence determination and analysis. DNA sequence

analysis was performed by the method of Sanger et al. (22).M13 manipulations were performed as described by Messinget al. (13). The polyacrylamide-urea gel electrophoresissystem described by Sanger and Coulson (21) was used.Sequence analysis was performed in part by a computer withthe program of J. B. Kaplan (unpublished).

RESULTSPlasmid construction and sequence determination of E. coli

pabB. Two independent cloning experiments yielded plas-mids carrying E. coli pabB. E. coli genomic DNA waspartially digested with either EcoRI or HindlIl. Fragmentsresulting from these two digests were ligated into eitherEcoRI-digested or Hindill-digested plasmid pBR322. Theproducts of these constructions were used to transform E.coli AB3303 (pabB3) to PABA independence and atnpicillin

57

on March 16, 2021 by guest

http://jb.asm.org/

Dow

nloaded from

58 GONCHAROFF AND NICHOLS

I I T

Ic~

qz~~~I Szc \c0 -.._

pPG3 -I . - -

II-

-__

pobS

FIG. 1. (a) Linear restriction maps of pabB-containing plasmids. The heavy dark lines represent pBR322 vector DNA, and the light linesrepresent inserted DNA. The open arrow within the inserted DNA corresponds to the coding region of the pabB and indicates direction oftranscription. (b) DNA sequencing strategy. Arrows represent DNA sequence determined from specific Sau3A, TaqI, or HaeIII fragmentscloned into M13 bacteriophage vectors.

resistance. Plasmids were isolated from the resulting colo-nies. Two of these plasmids were named pBNpabB(R3) andpBNpabB(HJ), corresponding to the EcoRI and HindlllconstrUctions, respectively. Besides the 4.3 kilobases (kb) ofpBR322 vector DNA, pBNpabB(R3) contained 9 kb of E.coli DNA; pBNpabB(Hl) contained 8.7 kb. The restrictionmaps of these two plasmids (Fig. la) were found to beidentical within the 6.5-kb HindIII-EcoRI fragment. Thisindicated that the pabB gene was contained within thisfragment, Further subclones were constructed frompBNpabB(R3).The subcloning strategy of pBNpabB(R3) is shown in Fig.

la. Plasmids pBNpabB(R3) and pBR322 were digested withBamHI and EcoRI. This mixture of fragments was ligatedand used to transform E. coli AB3303 to PABA indepen-dence and ampicillin resistance. One of the plasmids isolatedin this construction, pPG1, was a pBR322 derivative thatcontained a 5.6-kb BamHI-EcoRI fragment frompBNpabB(R3). A SalI reduction of pPG1 did not yieldPABA-independent colonies, suggesting that at least one ofthe Sall sites lay within pabB. A 1.9-kb fragment wasisolated from pPG1 after a partial SalI digest. The fragmentcontained the 275-base pair (bp) SalI-BamHI portion ofpBR322 and a 1,622-bp portion of E. coli DNA containingpabB. pPG3 was constructed by ligation of the 1.9-kb SalIfragment into the Sall site of pBR322 and transformation ofE. coli AB3303 to ampicillin resistance and PABA indepen-dence. pPG3 contains the 1,622-bp BamHI-SalI fragmentflanked by 275-bp direct repeats.The 1,622-bp BamHI-SalI fragment of plasmid pPG3 was

the source of smaller DNA fragments used to determine thenucleotide sequence of pabB. Sau3A, TaqI, and HaeIIIrestriction fragments were isolated frotn 5% polyacrylamidegels and were ligated into M13mp8 or M13mp9 bacterio-

phage vectors restricted with BamHI, AccI, and HincIl,respectively. Single-stranded DNA was prepared from E.coli JM103 that had been transfected with the recombinantphages. The nucleotide sequence of the 1,622-bp BamHI-SalI fragment was determined completely on both strands ofthe DNA (Fig. 2).

Translational reading frame of pabB. Examination of allpossible translational frames of the 1,622-bp BamHI-SalIfragment identified one 1,359-bp continuous reading franme.No other reading frame exceeded a total length of 300 bp.This 1,359-bp open reading frame has the potential forencoding a protein containing 453 amino acid residues with acalculated molecular weight of 50,958. This figure is in closeagreement with two independently determined molecularweights for E. coli PABS CoT. Gel permeation chromatogra-phy yielded a molecular weight of 46,000 (5), and sodiumdodecyl sulfate-polyacrylamide gel electrophoresis yielded amolecular weight of 53,000 (Seibold and Nichols, unpub-lished results). These data and the fact that plasmid pPG3can complement the pabB mutation of strain AB3303 pro-vide strong evidence that the nucleotide and antino acidsequences shown in Fig. 2 are those of E. coli pabB.Amino acid and nucleotide sindilarities of pabB and thpE.

The amino acid sequences of PABS Col and AS Col arealigned and presented in Fig. 3. Six gaps have been insertedbetween the two sequences at positions 124 to 125, 130 to131, 160 to 161, 228, 271 to 272, and 281 to 282 to align areasof similarity. Other alignments are possible if more gaps areinserted, but care was taken to use a minimal number of gapsin aligning the two sequences. Some gaps have been insertedarbitrarily within specific regions of the two sequences. Forexample, the gap present between positions 160 and 161could have been placed between positions 148 and 149without affecting the overall amount of similarity. No gaps

(a) F

J. BACTERIOL.

(b)r

_,

r _

-I i i ... m

1%. 4t -zrX NStt a

41 0,04 401 1

on March 16, 2021 by guest

http://jb.asm.org/

Dow

nloaded from

E. COLI pabB NUCLEOTIDE SEQUENCE 59

-130 -120GGATCCGCTCGACAG

-110 -100 -90 -80 -70 -60 -50 -40 -30 -20 -10CTTTTTGCTGCTGTTCATGGGTGAGTGAAGGTAAACCTGCAAACATTGTTAACTCCTGCTAAATTGTTGGCGCTAATTATTTCATGCTACCCGGCACATAGCCAGTAGAGTCAGGACTG

10 20 30Met Lys Thr Leu Ser Pro Ala Val Ile ThrATG AAG ACG TTA TCT CCC GCT GTG ATT ACT

100 110 120Ala Met Leu Leu His Ser Gly Tyr Ala AspGCG ATG CTT TTA CAC TCC GGC TAT GCC GAT

190 200 210Gly Lys Glu Thr Val Val Ser Glu Ser GluGGT AAA GAA ACC GTT GTT AGT GAA AGC GAA

280 290 300Asp Ile Arg Pro Thr His Asn Glu Asp LeuGAC ATT CGC CCA ACG CAT AAC GAA GAT TTG

370 380 390Leu Pro Glu Ile Ala Glu Gln Asp Il ValCTG CCA GAA ATT GCG GAA CAA GAT ATC GTT

460 470 480Thr Val Ser Leu Leu Ser His Asn Asp ValACA GTT TCT TTG CTG AGT CAT AAT GAT GTC

40 50Leu Leu Trp Arg Gin Asp Ala Ala GluTTA CTC TGG CGT CAG GAC GCC GCT GAA

60 70 80 90Phe Tyr phe Ser Arg Leu Ser His Leu Pro TrpTTT TAT TTC TCC CGC TTA AGC CAC CTG CCG TGG

130 140 150 160 170His Pro Tyr Ser Arg Phe Asp Ile Val Val Ala Glu Pro Ile Cys Thr LeuCAT CCG TAT AGC CGC TTT GAT ATT GTG GTC GCC GAG CCG ATT TGC ACT TTA

180Thr Thr PheACC ACT TTC

220 230 240 250 260 270Lys Arg Thr Thr Thr Thr Asp Asp Pro Leu Gin Val Leu Gin Gln Val Leu Asp Arg AlaAAA CGC ACA ACG ACC ACT GAT GAC CCG CTA CAG GTG CTC CAG CAG GTG CTG GAT CGC GCA

310 320Pro Phe Gln Gly Gly Ala LeuCCA TTT CAG GGC GGC GCA CTG

400 410Leu Pro Asp Met Ala Val GlyCTG CCG GAT ATG GCA GTG GGT

490 500Asn Ala Arg Arg Ala Trp LeuAAT GCC CGT CGG GCC TGG CTG

550 560 570 580 590Thr Ser Asp Trp Gin Ser Asn Met Thr Arg Glu Gln Tyr Gly Glu Lys Phe Arg GinACT TCC GAC TGG CAA TCC AAT ATG ACC CGC GAG CAG TAC GGC GAA AAA TTT CGC CAG

640 650 660Gln Val Asn Leu Ala Gln Arg Phe His AlaCAG GTG AAT CTC GCC CAA CGT TTT CAT GCG

670 680 690Thr Tyr Ser G1y Asp Glu Trp Gin Ala PheACC TAT TCT GGC GAT GAA TGG CAG GCA TTC

330 340 350Gly Leu Phe Gly Tyr Asp Leu Gly Arg ArgGGG TTG TTT GGC TAC GAT CTG GGC CGC CGT

420 430 440Ile Tyr Asp Trp Ala Leu Ile Val Asp HisATC TAC GAT TGG GCG CTC ATT GTC GAC CAC

510 520 530Glu Ser Gln Gln Phe Ser Pro Gln Glu AspGAA AGC CAG CAA TTC TCG CCG CAG GAA GAT

600 610 620Val Gln Glu Tyr Leu His Ser GlyGTA CAG GAA TAT CTG CAC AGC GGT

700 710Leu Gln Leu Asn Gln Ala AsnCTT CAG CTT AAT CAG GCC AAC

360Phe Glu SerTTT GAG TCA

450Gln Arg HisCAG CGT CAT

540Phe Thr LeuTTC ACG CTC

630Asp Cys TyrGAT TGC TAT

720Arg Ala ProCGC GCG CCA

730 740 750 760 770Phe Ser Ala Phe Leu Arg Leu Glu Gln Gly Ala lie Leu Ser Leu Ser ProTTT AGC GCT TTT TTA CGT CTT GAA CAG GGT GCA ATT TTA AGC CTT TCG CCA

780 790 800Glu Arg Phe Ile Leu Cys Asp Asn Ser Glu IleGAG CGG TTT ATT CTT TGT GAT AAT AGT GAA ATC

820 830 840Arg Pro Ile Lys Gly Thr Leu Pro Arg LouCGC CCG ATT AAA GGC ACG CTA CCA CGC CTG

850 860Pro Asp Pro Gln Glu Asp Ser Lys GlnCCC GAT CCT CAG GAA GAT AGC AAA CAA

870 880 890 900Ala Val Lys Leu Ala Asn Ser Ala Lys Asp ArgGCA GTA AAA CTG GCG AAC TCA GCG AAA GAT CGT

910 920 930 940 950Ala Glu Asn Leu Met Ile Val Asp Lou Met Arg Asn Asp Ile Gly Arg ValGCC GAA AAT CTG ATG ATT GTC GAT TTA ATG CGT AAT GAT ATC GGT CGT GTT

960 970 980 990Ala Val Ala Gly Ser Val Lys Val Pro Glu Lou Phe ValGCC GTA GCA GGT TCG GTA AAA GTA CCA GAG CTG TTC GTG

1000 1010 1020 1030 1040 1050 1060 1070 1080Val Glu Pro Pho Pro Ala Val Hig His Lou Val Ser Thr Ile Thr Ala Gln Lou Pro Glu Gln Leu His Ala Ser Asp Leu Leu Arg AlaGTG GAA CCC TTC CCT GCC GTG CAT CAT CTG GTC AGC ACC ATA ACG GCG CAA CTA CCA GAA CAG TTA CAC GCC AGC GAT CTG CTG CGC GCA

1090 1100 1110 1120 1130 1140 1150 1160 1170Ala Phe Pro Gly Gly Ser Ile Thr Gly Ala Pro Lys Val Arg Ala Met Glu Ile Ile Asp Glu Leu Glu Pro Gin Arg Arg Asn Ala TrpGCT TTT CCT GGT GGC TCA ATA ACC GGG GCT CCG AAA GTA CGG GCT ATG GAA ATT ATC GAC GAA CTG GAA CCG CAG CGA CGC AAT GCC TGG

1i80 1i90 1200 1210 1220 1230 1240 1250 1260Cys Gly Ser Ile Gly Tyr Leu Se& Phe Cys Gly Asn Met Asp Thr Ser Ile Thr Ile Arg Thr Leu Thr Ala Ile Asn Gly Gln Ile PheTGC GGC AGC ATT GGC TAT TTG AGC TTT TGC GGC AAC ATG GAT ACC AGT ATT ACT ATC CGC ACG CTG ACT GCC ATT AAC GGA CAA ATT TTC

1270 1280 1290 1300 1310 1320 1330 1340 1350Cys Ser Ala Gly Gly Gly Ile Val Ala Asp Ser Gln Glu Glu Ala Glu Tyr Gln Glu Thr Phe Asp Lys Val Asn Arg Ile Leu Lys GlnTGC TCT GCG GGC GGT GGA ATT GTC GCC GAT AGC CAG GAA GAA GCG GAA TAT CAG GAA ACT TTT GAT AAA GTT AAT CGT ATC CTG AAG CAA

1370 1380 1390 1400 1410 1420 1430 1440 1450 1460Leu Glu Lys EndCTG GAG AAG TAA GACGTGGAATACCGTAGCCTGACGCTTGATGATTTTTTATCGCGCTTTCAACTTTTGCGCCCGCAAATTAACCGGGAAACCCTAAATCATCGTCAGGCTGCTG

1470 1480TGTTAATCCCCATCGTCCGTCGAC

FIG. 2. Nucleotide and amino acid sequence of E. coli pabB. The complete nucleotide sequence of the 1,622-bp BamHI-SalI fragment isshown along with the predicted amino acid sequence. Underlined portions of the sequence correspond to possible regulatory regionsdiscussed in the text.

had to be inserted from position 282 to the end of thecomparison. Out of 450 amino acids compared, 26% similar-ity exists between PABS CoI and AS Col. Using this samealignment, a similarity of 40% exists at the nucleotide level.The extent of similarity at the nucleotide level throughoutthe alignment of pabB and trpE is shown in Fig. 4.Codon usage in pabB. Codon usage is believed to be

primarily dependent on the relative amounts of isoacceptingtRNA molecules present in the cell (7). The codon usage ofpabB is similar to that of trpE (Table 1) and is consistent withthe codon usage of E. coli genes which are not highly

expressed (4). These data, including the amino acid align-ment of pabB with trpE, further support the proposedtranslational frame of E. coli pabB (Fig. 2).PABS CoI contains seven tryptophan residues, whereas

AS Col contains none. Furthermore, it has been reportedthat PABS CoIl contains three tryptophan residues, whereasits counterpart, AS CoII, contains none (10, 16). It has beenproposed that AS CoI and CoII, which together make up AS,are tryptophan free so that chorismate can be channeled intothe tryptophan biosynthetic pathway under severe trypto-phan starvation conditions (17). Thus, the presence of tryp-

VOL. 159, 1984

810Gln ThrCAG ACC

1360

on March 16, 2021 by guest

http://jb.asm.org/

Dow

nloaded from

60 GONCHAROFF AND NICHOLS

10PABS COI M K T Lm P A V I T L W ROAS COI M O T O K P T L E L L T C E G A Y R D N P T A L F H Q L c a D R P A T L L L E S A DID J K D L K S JL L V

20 30 40 50 60 70AEA E F Y F S R L S H L P W A MrL H SrGY D H P Y S R F D I V V A E P I C T L T T F G K E T V V S EnE K R T T TjS JL R I T A L G D T V T IO Ala S G NLUJEWL L A L L D N A L P A G V E S E Q S P N C R V L R F P P VL5JP L L D E D80 90 100 120 130T D D P L V L O O V L D R A D I R P T H N E D L P F IG G A L G rFGYD1L G R R[1E S E I - A E Q D I V L|P DIHA R L C S L S V F OA F R L L O N L L N V P K E E R E LM F F S FSI D LIV A G F E D LQ L SIA E N N C - - P F140 150 160 170 180A V G I Y D W A L I V R H T V S L L S H N D V - - - - - - - - - - - - - N A R RJWnA E S Q F S P Q E D F T LTC F Y L A E T L M V I |D OK K S T R I A 5 L F A P NE E E K O R L T A R LOJNE LJOOQQ T E A A P P L P V V S V P H

190 200 210 220 2 0 240S DW QS n M11T R QYIT E K F1FT VV[E Y L H S7nD C Y N LAQRF H AT Y S G D EW QA F LQf Nl N R A[ FMRCECW JO S DCE FLOG V V|E LWK A I R AGE I F10 VIV 5 RIt FI P C P 5 P L A Y Y V K 5lYP s Y

250 260 270 280 290S AIR L R L E 0 G A I L S Lr PER F I L C E1NS E I T R - -1K GTLPrVR|L P - - - - - - EP 0 ErD SK Q A V K LM F EM O D N D F T L F G AI P ES5 L K Y JA T 5 R I E I YI IAIG TIRIP RIG R R AD G SL R D LID SI I E L E M

300 310 320 330 340 350A N S Ar1D R A Nr MI rV-T mHR N DII GRV A V A G SIV K P E F VrE P F P ArH H L V S|T I T A QMP E Q HRTD HWJE L SH M|L|VDL A D|L AJ I C T PGJR Y A D T KVD R Y 5 YWM H L V 51RVVG EULR H D D360 370 380 390 400

T1S DL LIPF-I-A F P GmS I TGA WKV AME IID FL P QOR RIN A W C G YSIYLS F CnN N[D TIS TI R1T L TSL H A Y C M N M T L S|G A P K V R A M L A A G RR RG S Y GWA VIG YIF T A HUG D LID TI RIS A L

420 430 440 450AI [N G 1I F CSI1 G G I ArA6T Q E E AEY QOE TIF DrKjV N R IfK Q L E KV EIN GI I A T V OA GIA G V V L0D SIV P OSEA DIE TIRNS A R A V R A I A T A H H A Q E T F

FIG. 3. Amino acid alignment of PABS Col and AS Col. Reference numbers above the two sequences correspond to the PABS Colsequence.

tophan residues in PABS Col and CoIl would aid in thisprocess. Presumably, other proteins which use chorismateas a substrate, like those encoded by pheA and tyrA, will befound to contain tryptophan residues as well.

Putative regulatory sequences in pabB. Unlike trpE, pabBcontains no attenuator-like sequences in its 5' flankingregion. In addition, little similarity exists between the 5'flanking regions of these two genes. Nevertheless, the 5'region of pabB contains a promoter-like sequence extendingfrom nucleotide positions -81 to -35. The region from posi-tions -73 to -67 has a possible RNA polymerase binding site(5'-TGTTAAC-3' [20]), and 21 bp downstream from this areathere is a Pribnow-like box (5'-TAAT-3' [181) at positions-46 to -43. The positions of these putative regulatorysequences suggest that transcription initiation occurs atposition -36 before the initiation codon (23). Within theputative mRNA leader sequence, there is a Shine-Delgarnoribosome binding site at positions -9 to -4 (5'-TCAGGA-3'[24]; these putative regulatory sequences are underlined in

6pbB I

trpE I

!050-

3 39

Fig. 2). At present, the translational start point of pabB isunknown. We believe, however, that the translation ofpabBinitiates at the methionine codon shown in Fig. 2 because theputative regulatory regions lie directly upstream from it.

Examination of the 3' region of pabB revealed no canoni-cal rho-independent termination sequence. Currently, ex-periments are in progress to determine whether the above-mentioned proposed regulatory regions affect pabBexpression.

DISCUSSION

We have established the complete nucleotide sequence ofE. coli pabB, including the 5' and 3' flanking regions. thecoding sequence of pabB can be aligned with that of E. colitrpE. The two genes have similarities of 40% at the nucleo-tide level and 26% at the amino acid level. These datasuggest that pabB and trpE arose from a common ancestorsince similarities exist throughout most of the alignment of

3

6 8

/NhJkA

06 w 26o 36o 460 500 60 7o0 8o00s6ooKo io00sb o 40o oobpFIG. 4. Graphic representation of nucleotide similarity between pabB and trpE coding and flanking regions. Each point represents the

amount of similarity present within a stretch of 40 nucleotides and is placed at the beginning of the corresponding nucleotide region. Abovethis graph are open arrows corresponding to the coding regions of pabB and trpE. Relative positions of the gaps introduced in the amino acidalignment of these two gene products are indicated by vertical lines drawn through the open arrows. Reference numbers corresponding tothese lines represent the numbers of nucleotides deleted.

J. BACTERIOL.

V^--TVVVVIVNVW-Nnv %11%( %Off%"-" V-.

on March 16, 2021 by guest

http://jb.asm.org/

Dow

nloaded from

E. COLI pabB NUCLEOTIDE SEQUENCE 61

TABLE 1. Codon usage for E. coli pabB and trpE

Amino Codons in gene: Amino Codons in genea:acid Codon pabB trpE acid Codon pabB trpE

Phe YTTT 13 (62) 8 (40) TAG 0 (0) 0 (0)

Leu

Ile

Met

Val

Ser

Pro

Thr

Ala

Tyr

End

TTC

TTATTGCTTCTCCTACTG

ATTATCATA

ATG

GTTGTCGTAGTG

TCTTCCTCATCG

CCTCCCCCACCG

ACTACCACAACG

GCTGCCGCAGCG

TATTAC

TAA

I-

8 (38)

9 (19)4 (9)6 (13)5 (11)3 (6)

20 (43)

16 (64)7 (28)2 (8)

8 (100)

6 (22)6 (22)6 (22)9 (33)

4 (13)4 (13)3 (10)3 (10)

3 (13)3 (13)8 (35)9 (39)

8 (31)9 (35)2 (8)7 (27)

6 (16)14 (37)8 (21)

10 (26)

8 (73)3 (27)

1 (100)

12 (60)

7 (11)5 (8)6 (9)10 (15)4 (6)

35 (52)

12 (71)5 (29)0 (0)

12 (100)

5 (15)6 (18)8 (24)

14 (42)

6 (16)5 (14)4 (11)5 (14)

3 (11)5 (18)5 (18)

15 (54)

4 (16)14 (56)4 (16)3 (12)

12 (23)19 (36)6 (11)16 (30)

9 (64)5 (36)

0 (0)

His

Gln

Asn

Lys

Asp

Glu

Cys

End

Trp

Arg

Ser

Arg

Gly

CATCAC

CAACAG

AATAAC

AAAAAG

GATGAC

GAAGAG

TGTTGC

TGA

TGG

CGTCGCCGACGG

AGTAGC

AGAAGG

GGTGGCGGAGGG

,-

7 (58)5 (42)

8 (27)22 (73)

10 (67)5 (33)

10 (77)3 (23)

23 (79)6 (21)

25 (81)6 (19)

1 (17)5 (83)

0 (0)

7 (100)

10 (36)14 (50)1 (4)3 (11)

4 (13)13 (42)

0 (0)0 (0)

8 (32)13 (52)2 (8)2 (8)

,-

7 (64)4 (36)

8 (38)13 (62)

10 (59)7 (41)

12 (80)3 (20)

21 (60)14 (40)

30 (86)5 (14)

5 (42)7 (58)

1 (100)

0 (0)

17 (43)21 (53)1 (3)0 (0)

4 (11)13 (35)

1 (3)0 (0)

11 (39)12 (43)3 (11)2 (7)

a Numbers in parentheses show the percentage of residues of the indicated amino acid coded by the indicated codon.

the two genes (Fig. 3). If these genes arose through conver-gent evolution, we would expect to see little similaritybetween them. The data show that this is not the case.Stretches of no similarity between the two sequences do,however, exist. This is an indication of the great amount ofdivergence which has occurred between the two genes.

In addition to our results indicating that PABS Col and ASCol have a common evolutionary origin, it has been pro-posed that PABS CoIl and AS CoIl have also evolved from acommon ancestor (10). Both sets of data are in agreementwith immunological studies suggesting that E. coli PABS andAS share common antigenic determinants (19). The amountof similarity between PABS Coll and AS CoIl, 44% at theamino acid level, is greater than the 26% amino acid similar-ity present between PABS Col and AS CoL. This is to beexpected since PABS CoIl and AS CoIl have identical rolesof transferring the amino group from glutamine to the Colsubunit of each respective enzyme complex.

On the other hand, the lower amount of similarity presentbetween PABS Col and AS Col can be ascribed to severalfactors. (i) AS Col responds to feedback inhibition bytryptophan, whereas PABS Col does not. Therefore, trypto-phan binding areas of AS Col will represent an area (orareas) with no similarity to PABS CoI. This region couldpossibly exist in the amino-terminal end of AS Col since thisarea shows the least amount of similarity to PABS CoI. (ii)PABS Col and AS Col have slightly different catalyticfunctions. (iii) Subunit interaction areas are different inPABS CoI and AS Col.

Since the complete nucleotide sequences of all the genescoding for PABS and AS are known, we can now considerthe evolutionary origins of these two enzymes. It has beenproposed that the development of new enzyme functions ismost easily achieved by recruiting proteins which alreadyexist and catalyze similar reactions (9). We believe that theexistence of PABS and AS is an actual case of acquisitive

_a

VOL. 159, 1984

on March 16, 2021 by guest

http://jb.asm.org/

Dow

nloaded from

62 GONCHAROFF AND NICHOLS

evolution which occurred after duplication of the genesencoding the initial synthetase complex. The fact that thisduplication event actually occurred is supported by datapresented here and elsewhere (10).

ACKNOWLEDGMENTSWe thank Jeffrey B. Kaplan for advice and criticism and for help

with the figures. We also thank Charles Yanofsky, in whoselaboratory pBNpabB(R3) and pBNpabB(H1) were constructed.

This work was supported by a grant from the University of Illinoisat Chicago Circle Research Board and by Public Health Servicegrant Al 18639 from the National Institutes of Health.

LITERATURE CITED1. Birnboim, H. C., and J. Doly. 1979. A rapid alkaline extraction

procedure for screening recombinant plasmid DNA. NucleicAcids Res. 7:1513-1523.

2. Crawford, I. P. 1975. Gene rearrangements in the evolution ofthe tryptophan synthetic pathway. Bacteriol. Rev. 39:87-120.

3. Gough, J. A., and N. E. Murray. 1983. Sequence diversityamong related genes for recognition of specific targets in DNAmolecules. J. Mol. Biol. 166:1-19.

4. Grantham, R., C. Gautier, M. Gouy, M. Jacobzone, and R.Mercier. 1981. Codon catalog usage is a genome strategymodulated for gene expressivity. Nucleic Acids Res. 9:r43-r74.

5. Huang, M., and F. Gibson. 1970. Biosynthesis of 4-aminoben-zoate in Escherichia coli. J. Bacteriol. 102:767-773.

6. Huang, M., and J. Pittard. 1967. Genetic analysis of mutantstrains of Escherichia coli requiring p-aminobenzoic acid forgrowth. J. Bacteriol. 93:1938-1942.

7. Ikemura, T. 1981. Correlation between the abundance of Esche-richia coli transfer RNAs and the occurrence of the respectivecodons in its protein genes: a proposal for a synonymous codonchoice that is optimal for the E. coli translational system. J.Mol. Biol. 151:389-409.

8. Ito, J., E. C. Cox, and C. Yanofsky. 1968. Anthranilate synthe-tase, an enzyme specified by the tryptophan operon of Esche-richia coli: purification and characterization of component I. J.Bacteriol. 97:725-733.

9. Jensen, R. A. 1976. Enzyme recruitment in evolution of newfunction. Annu. Rev. Microbiol. 30:409-425.

10. Kaplan, J. B., and B. P. Nichols. 1983. Nucleotide sequence ofEscherichia coli pabA and its evolutionary relationship totrp(G)D. J. Mol. Biol. 168:451-468.

11. Krikos, A., N. Mutoh, A. Boyd, and M. I. Simon. 1983. Sensorytransducers of E. coli are composed of discrete structural andfunctional domains. Cell 33:615-622.

12. Mandel, M., and A. Higa. 1970. Calcium-dependent bacterio-phage DNA infection. J. Mol. Biol. 53:159-162.

13. Messing, J., R. Crea, and P. H. Seeburg. 1981. A system forshotgun DNA sequencing. Nucleic Acids Res. 9:309-321.

14. Messing, J., and J. Vieira. 1982. A new pair of M13 vectors forselecting either DNA strand of double-digest restriction frag-ments. Gene 19:269-276.

15. Miozzari, G. F., and C. Yanofsky. 1979. The regulatory region ofthe trp operon of Serratia marcescens. Nature (London)276:684-689.

16. Nichols, B. P., G. F. Miozzari, M. vanCleemput, G. N. Bennet,and C. Yanofsky. 1980. Nucleotide sequences of the trpGregions of Escherichia coli, Shigella dysenteriae, Salmonellatyphimurium and Serratia marcescens. J. Mol. Biol. 142:503-517.

17. Nichols, B. P., M. vanCleemput, and C. Yanofsky. 1981. Nucleo-tide sequence of Escherichia coli trpE. Anthranilate synthetasecomponent I contains no tryptophan residues. J. Mol. Biol.146:45-54.

18. Pribnow, D. 1975. Bacteriophage T7 early promoters: nucleo-tide sequences of two RNA polymerase binding sites. J. Mol.Biol. 99:419-443.

19. Reiners, J. J., L. J. Messenger, and H. Zalkin. 1978. Immunolog-ical cross-reactivity of Escherichia coli anthranilate synthetase,glutamate synthase and other proteins. J. Biol. Chem. 253:1226-1233.

20. Rosenberg, M., and D. Court. 1979. Regulatory sequencesinvolved in the promotion and termination of RNA transcrip-tion. Annu. Rev. Genet. 13:319-353.

21. Sanger, F., and A. R. Coulson. 1978. Use of thin acrylamide gelsfor DNA sequencing. FEBS Lett. 87:107-110.

22. Sanger, F., S. Nicklen, and A. R. Coulson. 1977. DNA sequenc-ing with chain-terminating inhibitors. Proc. Natl. Acad. Sci.U.S.A. 74:5463-5467.

23. Scherer, G. E. F., M. D. Walkinshaw, and S. Arnott. 1978. Acomputer aided oligonucleotide analysis provides a model se-quence for RNA polymerase-promoter recognition in E. coli.Nucleic Acids Res. 5:3759-3773.

24. Shine, J., and L. Delgarno. 1974. The 3'-terminal sequence ofEscherichia coli 16S ribosomal RNA: complementarity to non-sense triplets and ribosome binding sites. Proc. Natl. Acad. Sci.U.S.A. 71:1342-1346.

25. Squires, C. H., M. DeFelice, J. Devereux, and J. M. Calvo. 1983.Molecular structure of ilvIH and its evolutionary relationship toilvG in Escherichia coli K12. Nucleic Acids Res. 11:5299-5313.

26. Yanofsky, C., T. Platt, I. P. Crawford, B. P. Nichols, G. E.Christie, H. Horowitz, M. vanCleemput, and A. M. Wu. 1981.The complete nucleotide sequence of the tryptophan operon ofEscherichia coli. Nucleic Acids Res. 9:6647-6668.

27. Zalkin, H., and T. Murphy. 1975. Utilization of ammonia fortryptophan synthesis. Biochem. Biophys. Res. Commun.67:1370-1377.

J. BACTERIOL.

on March 16, 2021 by guest

http://jb.asm.org/

Dow

nloaded from