Nucleotide Sequence Analysis of the purEK Operon Encoding 5'-Phosphoribosyl-5-Aminoimidazole Carboxylase of Escherichia coli K-12 - Journal of ...
←
→
Page content transcription
If your browser does not render page correctly, please read the page content below
JOURNAL OF BACTERIOLOGY, Jan. 1989, p. 205-212 Vol. 171, No. 1
0021-9193/89/010205-08$02.00/0
Copyright © 1989, American Society for Microbiology
Nucleotide Sequence Analysis of the purEK Operon Encoding
5'-Phosphoribosyl-5-Aminoimidazole Carboxylase
of Escherichia coli K-12
AMELIA A. TIEDEMAN,'t JACQUELINE KEYHANI,2 JOHN KAMHOLZ,2 HENRY A. DAUM III,1t
JOSEPH S. GOTS,2 AND JOHN M. SMITH'*
Seattle Biomedical Research Institute, 4 Nickerson Street, Seattle, Washington 98109,1 and Department of Microbiology,
School of Medicine, University of Pennsylvania, Philadelphia, Pennsylvania 19J042
Received 3 October 1988/Accepted 2 November 1988
5'-Phosphoribosyl-5-aminoimidgzole (AIR) carboxylase (EC 4.1.1.21) catalyzes step 6, the carboxylation of
Downloaded from http://jb.asm.org/ on April 28, 2021 by guest
AIR to 5'-phosphoribosyl-5-aminoimidazole-4-carboxylic acid, in the de novo biosynthesis of purine nucleo-
tides. As deduced from the DNA sequence of restriction fragments encoding AIR carboxylase and supported
by maxicell analyses, AIR carboxylase was found to be composed of two nonidentical subunits. In agreement
with established complementation data, the catalytic subunit (deduced Mr, 17,782) was encoded by the purE
gene, while the CO2-binding subunit (deduced Mr, 39,385) was encoded by the purK gene. These two genes
formed an operon in which the termination codon of the purE gene overlapped the initiation codon of the purK
gene. The 5' end of the purEK mnRNA was determined by mung bean nuclease mapping and was located 41
nucleotides upstream of the proposed initiation codon. The purEK operon is regulated by the purR gene
product, and a purR regulatory-protein-binding site related to the sequences found in other pur loci was
identified in the purEK operon control region.
5'-Phosphoribosyl-5-aminoimidazole (AIR) carboxylase typhimurium, the purE locus is unlinked to the other scat-
(EC 4.1.1.21) catalyzes the carboxylation of AIR to 5'-phos- tered loci of the purine de novo biosynthesis pathway (2, 36),
phoribosyl-5-aminoimidazole-4-carboxylic acid (CAIR): in contrast to Bacillus subtilis, in which the equivalent loci
AIR + CO2 ± CAIR. This reaction constitutes step 6 in the are organized as a single large operon (7).
de novo synthesis of IMP, the first complete nucleotide of In this paper, we report the nucleotide sequence of the
purine biosynthesis (20). While the enzyme has been only purE locus, the identification of its control region, and
partially purified from several sources (20, 32), the gene regulation by the purR gene product. The DNA sequencing
encoding AIR carboxylase has been the focus of several studies revealed that the purE locus is composed of two
studies (10, 19, 44) and has been cloned from several diverse separate genes that form the purEK operon. In consultation
organisms (7, 11, 12, 17, 31). with B. Bachmann of the E. coli Genetic Stock Center, the
Escherichia coli and Salmonella typhimuruim strains with purE1 locus has been designated purE, and the purE2 locus
mutations in the purE locus have been divided into two has been designated purK. Consistent with previous com-
classes, first by complementation analysis and then accord- plementation analyses (10), the first gene in the operon
ing to their growth response to high CO2 concentrations (4, (purE) encoded the catalytic subunit, while the second gene
10). Those mutants that have an absolute requirement for (purK) encoded the C02-binding subunit. The termination
exogenous purines have been designated purE1 mutants, codon of the purE gene overlapped the initiation codon of
while purE2 mutants are CO2 conditional in that they can the purK gene, and thus the expression of the two AIR
grow without exogenous purines in the presence of increased carboxylase subunits should be translationally coupled. The
concentrations of CO2 (10). This indicates that the purE2 purEK mRNA 5' end was identified by mung bean nuclease
product is required for optimal CO2 binding. mapping, and a purR regulatory-protein-binding site
As a historical note, the gene encoding AIR carboxylase (GCAAACGTTTGC) found in the control regions of the
has the distinction of being involved in the first lac fusion purF, purMN, and other sequenced pur loci (A. A. Tiede-
described by Jacob et al. (16). This and other lac fusions man, A. Shiau, S. A. Wolfe, C. G. Gaines, and J. M. Smith,
have been extensively utilized to investigate the regulation Fed. Proc. 46:2218, 1987) was also found in the purEK
of this gene and its potential coregulation with the remaining control region. Introduction of a purR mutation into a
loci of the de novo purine biosynthesis pathway (19, 44). purE-lac fusion strain led to loss of regulation of the purEK
These studies have established that the purE loci of E. coli operon.
and S. typhimuruim are regulated by a mechanism in com-
mon with the other pur loci (10, 19, 44), and regulatory MATERIALS AND METHODS
mutants with mutations unlinked and linked to the purE
locus have been characterized (10, 19, 44). In E. coli and S. Strains and media. E. coli K-12 TXS17 [ara A(gpt-pro-lac)
purE204 srlC300::TnJO recA56], in addition to the strains
previously described (17), was constructed and used as a
*
Corresponding author. recipient for the isolation of purEK plasmids by complemen-
t Present address: Department of Molecular Genetics, The Uni- tation. The recipient strain JM83 (25) was used to identify
versity of Texas System Cancer Center, M.D. Anderson H4ospital subclones containing restriction fragments from the purEK
and Tumor Institute, Houston, TX 77030. region, while strain JM101 (25) was employed for the prop-
t Present address: FMC Bioproducts, Rockland, ME 04841-9987. agation of M13 bacteriophages. E. coli CSR603 (35) was used
205206 TIEDEMAN ET AL. J. BACTERIOL.
for maxicell analyses (34), and strain TX302 (49) is our Enzymes and chemicals. [t-32P]dATP was obtained from
standard strain for genetic and regulatory studies. Strains New England Nuclear Corp. (Boston, Mass.), and [-y-
were made competent and transformed by the procedure of 32P]ATP was obtained from ICN (Irving, Calif.). T4 DNA
Dagert and Ehrlich (6). The minimal medium of Neidhardt et ligase, DNA polymerase I (Klenow fragment), T4 polynu-
al. (28) and the rich media described by Miller (27) were used cleotide kinase, calf intestinal alkaline phosphatase, and
for the growth of the E. coli K-12 strains. SmaI restriction enzyme were obtained from Boehringer
A purE-lacZ fusion [J?(purE'-lacZ' Y:: Kan)214] was cre- Mannheim (Indianapolis, Ind.). All other enzymes were
ated in vitro by the digestion of plasmid pJS131 with AsuII obtained from New England BioLabs (Beverly, Mass.).
and BsmI (nucleotides 411 and 660, respectively), treatment Mung bean nuclease and deoxy- and dideoxynucleotide
with T4 DNA polymerase to create blunt ends, and ligation triphosphates were obtained from Pharmacia (Piscataway,
with a SmaI-digested lacZY::Kan cassette (46). After trans- N.J.). 5-Bromo-4-chloro-3-indoyl-,-D-galactoside and all
formation and verification by restriction digestion, the re- other chemicals were obtained from Sigma Chemical Co.
sulting purE-lacZ fusion was transferred to the E. coli (St. Louis, Mo.).
chromosome by the procedure of Winans et al. (48). P1 Maxicell analysis. The labeling of plasmid proteins in
transduction was then used to move the purE-lacZ fusion by UV-irradiated whole cells (maxicells) was carried out as
selection for Kanr into strain TX302 to create strain TX725 described by Sancar et al. (34), except that after UV irradi-
[A(lac)U169 ID(purE'-lacZ' Y::Kan)214]. ation, cells were aerated for 1 h at 37°C, cycloserine was
Downloaded from http://jb.asm.org/ on April 28, 2021 by guest
Plasmids. Plasmid pLC8-25 from the library of Clarke and added at a final concentration of 100 ,ug/ml, and aeration at
Carbon (5) has been described previously (17). A 3.0- 37°C was then resumed and continued for 16 h. The growth
kilobase-pair (kb) BgIII fragment from plasmid pLC8-25 medium was as described previously (34), except that it was
containing the purEK operon (17) was subcloned into plas- supplemented with thiamine (2 ,ug/ml) and, for the plasmid-
mid pUC19 (50) to create plasmid pJS131. This plasmid, in carrying derivatives of CSR603, ampicillin (100 ,ug/ml).
addition to plasmids pJK3 and pJK43 (17), was subsequently When indicated, adenine was also added to the medium at a
employed as a source of restriction fragments for DNA final concentration of 0.5 mg/ml). Samples were loaded on a
sequencing studies. Plasmids pJK42 and pJK43 (17) carrying 16% polyacrylamide-sodium dodecyl sulfate gel for electro-
portions of the purEK operon were utilized for maxicell phoresis.
analysis (34). Plasmids pTE405 and pTE407 and their TnS AIR carboxylase comparisons. The AIR carboxylase se-
derivatives containing the purE gene of Methanobrevibacter quences were aligned with the Genealign program of H.
smithii were obtained from P. Hamilton (11). Martinez (23) as implemented on BIONET, and local adjust-
DNA isolation and sequence analysis. The DNA isolation ments were made by visual inspection.
procedures previously described were employed (47). DNA
sequences were determined by the dideoxy-chain termina- RESULTS AND DISCUSSION
tion method of Sanger et al. (38). A 1.9-kb HpaI restriction Nucleotide sequence of the purEK operon. The sequence of
fragment containing the purEK operon was digested with the 1.9-kb HpaI restriction fragment encoding the purE locus
AlI, FnuDII, HaeIII, Sau3A, TaqI, HpaI, and DraI restric- (17) was determined for both strands from overlapping DNA
tion enzymes. Additional restriction fragments to extend the restriction fragments. An additional and overlapping DNA
sequence beyond the purK gene were obtained from plasmid sequence from the HpaI site to a BgIII site 480 bp down-
pJS131. The resulting restriction fragments were then ligated stream of the purK gene was also determined after subclon-
into the appropriate cloning sites in M13mpl8 and 19 (50) ing of additional restriction fragments from plasmid pJS131.
and transformed into JM101 (26). Colorless plaques were A detailed restriction map and the specific DNA fragments
individually picked and propagated for the preparation of sequenced are shown in Fig. 1. The DNA sequence and the
sequencing DNA (37). The 17-base primer used for sequenc- deduced amino acid sequence are shown in Fig. 2, and the
ing (5'-GTTTTCCCAGTCACGAC-3') was obtained from sequence is numbered from the upstream HpaI restriction
B. A. Roe, University of Oklahoma. The DNA sequences site to the BglII site.
were compiled and analyzed by computer (18) and in part by Analyses of the sequence in Fig. 2 for open reading frames
the facilities of the BIONET resource. The DNA sequence (ORFs) revealed two overlapping ORFs. The first, encoding
data presented in this paper have been submitted to the a polypeptide of Mr 17,782, initiated with an ATG codon and
EMBL Data Library with the accession number X12982. extended from nucleotide 322 to nucleotide 831, with the
RNA studies. RNA was extracted from strain TX517 translational stop codon, TGA, overlapping the ATG start
containing plasmid pJS131 by the sodium dodecyl sulfate hot codon of the second ORF. The latter also initiated with an
phenol method (33). Mung bean nuclease was employed to ATG codon, started at nucleotide 828, and extended to
determine the 5' end of the purEK mRNA as previously nucleotide 1895, where the termination codon (TAA) was
described (40). A 416-bp TaqI restriction fragment that part of the HpaI recognition sequence (GTTAAC) and would
spanned the end of the purEK locus was 5'-end labeled with encode a polypeptide of Mr 39,385. Previous studies of the
[y-32PIATP and polynucleotide kinase after the fragment had purE locus have indicated two complementation groups,
been dephosphorylated with calf intestinal alkaline phospha- purEl and purE2, where purEl contained the catalytic activ-
tase (22). Total RNA was hybridized to the labeled fragment ity while purE2 contained the C02-fixing activity (10). The
for 8 h at 49°C. Mung bean nuclease at 60 and 120 U was then first ORF also lay within the 866-bp HpaI-HaeIII restriction
added to duplicate hybridization reactions, and the mixtures fragment previously shown to encode the catalytic activity
were incubated for 15 min at 37°C. After termination of the of AIR carboxylase and to contain the control region (17).
reaction by phenol extraction, the mung bean nuclease Accordingly, the first ORF should be designated purE, the
digestion mix was precipitated with ethanol, suspended in catalytic subunit, while the second ORF should be desig-
loading buffer, and loaded onto a sequencing gel. The size of nated purK and contains the C02-binding activity of AIR
the protected fragment, corrected for the phosphate group carboxylase.
(41), was determined after autoradiography by comparison The ATG codon at nucleotide 322 is indicated as the start
with an accompanying dideoxy sequencing ladder. of the purE coding region as based on the subunit size andVOL. 171, 1989
A/uI
FnuDII
HoaeM
Sou3A
I
I I
I
I
~~
HpoI DroI Hg/AI AsvuI
.
Il
E
BsmI
It If
i
Bo/I AhoI
I . I
I If
_-
a
K
HglAI KpnI
FspI Bc/I Bc/I HpoI (H\gAI NruI
...!...... ............................................................--.........
--
a
I I
purEK OPERON
.....................................I . .
8g9II
I I
207
Downloaded from http://jb.asm.org/ on April 28, 2021 by guest
. 1z1L
Toq I
HpoaI Hpo I HqiA Nru I Bq/1I
Specifics , ,. It", Ft
I
i-0 --tf
FIG. 1. Restriction endonuclease sites and sequencing strategy for the purEK operon. The location of the major 6-bp restriction enzyme
sites in the purEK operon are shown at the top. The restriction enzymes employed to generate DNA fragments for sequencing are indicated.
The arrows denote sequencing direction, and the length of the arrow is proportional to the number of nucleotides determined. An HpaI site
is designated as the 5' end. The coding regions for purE and purK are indicated by solid and stippled regions, respectively.
alignment with AIR carboxylase sequences from other or- CSR603, whether it was grown in the absence or presence of
ganisms (discussed below). This proposed initiation codon adenine. These results are shown in Fig. 3.
was preceded by two potential Shine-Dalgarno (39) se- Comparison of E. coli AIR carboxylase with AIR carboxyl-
quences. The AGGAG at nucleotides 306 to 310 showed the ases from other organisms. The genes encoding AIR carbox-
greatest identity with the end of the 16S rRNA (39). The ylase have been cloned and sequenced from Bacillus subtilis
other potential ribosomal binding site was the sequence (7), Methanobrevibacter smithii (12), and Methanobacte-
TAAG at nucleotide 314 to 317, but as discussed by Gold and rium thermoautotrophicum (11). The AIR carboxylase activ-
Stormo (9), the AGGAG sequence would be more commonly ity of the two methanogens appears to reside on a single
used for translational initiation. polypeptide chain (11, 12). The purE gene of Methanobrevi-
The proposed ribosomal binding site for the purK gene bacter smithii encodes a single polypeptide with a molecular
was a more typical GAGG sequence (9) at nucleotides 818 to weight of about 37,000. Hamilton and Reeve (11) have
821, lay entirely within the purE coding region, and preceded introduced Tn5 insertions into the gene, and these deriva-
an ATG initiation codon. The overlapping of the purE tives produced truncated polypeptides with Mr ranging from
termination codon with the purK translational initiation 7,000 to 28,000 that are unable to complement the purE
codon suggests not only that the purEK genes are an operon mutants of E. coli. We found that the plasmid with the most
but also that they are translationally coupled (30). The distal insert (E3::TnS) which produces the truncated 28,000-
boundaries as well as the correctness of the sequence for the Mr protein did indeed complement purE mutants of both E.
purEK genes is also supported by maxicell analysis and coli and S. typhimurium, but only in the presence of high
comparisons with the sequences of AIR carboxylase from CO2 tension. This means that it has purE but not purK
other organisms. activity and that the parent polypeptide contains two do-
Maxicell analyses. Strain CSR603 carrying plasmid pJK42, mains analogous to the two subunits of E. coli AIR carbox-
which contains the entire purEK operon (17), showed two ylase. Analysis of the complementation patterns of the TnS
protein bands of molecular weights 15,000 and 41,000, re- derivatives placed the purE product between 12,000 and
spectively, while strain CSR603 carrying plasmid pJK43, 27,000 Mr and established that the terminal 75 amino acids
which contains only the control region of the purEK operon are required for purK activity.
and the purE gene, showed only the protein band of molec- It has also been noted that the AIR carboxylase polypep-
ular weight 15,000. Growth in the presence of adenine led to tide of Methanobrevibacter smithii contains a duplication (7)
much less labeling of these protein bands. Besides these where the carboxy-terminal half of the polypeptide can be
bands, two more protein bands of molecular weights 28,000 aligned with the front amino-terminal half. This duplication
and 31,000, respectively, were mostly noticeable. Labeling is also present in the sequence of the homologous enzyme
in these bands was not diminished when the cells were from Methanobacterium thermoautotrophicum such that the
grown in the presence of adenine. These bands are attributed AIR carboxylases from the two methanogens can be readily
to the proteins encoded by the gene responsible for ampicil- aligned along their length (11). This argues for a gene
lin resistance (34), which is present in both plasmids pJK42 duplication of ancient origin, since these two methanogens
and pJK43. None of the bands described were detected in are widely separated on an evolutionary scale (11). Another208 TIEDEMAN ET AL. J. BACTERIOL.
10 20 30 40 50 60 70 1208 1223 1238 1253
GTTAACCAAA ACGCGGTGGT CAGTGCGATG GAAAAACATC AGGTGCAATG GCTGATCCAC GGGCATACCC TAT GAC GGT C0C CGT CM TGG cGT TTA CGC GCA MT GAA ACC GM CAG TTA CCG GCA
Tyr Asp Gly Arg Gly Gln Trp Arg Leu Arg Ala Asn Glu Thr Glu Gln Leu Pro Ala
80 90 100 110 120 130 140
ATCGCCCGGC GGTGCATGAA CTTATCGCCA ATCACCAACC TGCTTTTCGC GTGGTACTGG GTGCCTGGCA 1268 1283 1298 1313
GAG TGT TAC GGC GM TGT ATT GTC GAG CAG GGC ATT MC TTC TCT GGT GM GTG TCG
Glu Cys Tyr Gly Glu Cys Ile Val Glu Gln Gly Ile Asn Phe Ser Gly Glu Val Ser
150 160 170 180 190 200 210
TACGGAAGGT TCAATGGTGA AAGTCACGGC GGATGACGTT GAGCTGATTC ATTTTCCGTT TTAAAAAACC
1328 1343 1358 1373
CTG GTT GGC GCG CGC GGC TTT OAT GGC AGC ACC GTG TTT TAT CCG CTG ACG CAT MC
220 230 240 250 260 270 280 Leu Val Gly Ala Arg Gly Phe Asp Gly Ser Thr Val Phe Tyr Pro Leu Thr His Asn
CGCAACTTTG CTGATTTCAC AGCCACGCAA CCGTnCT TGCTCTCTTT CCGTGCTATT CTCTGTGCCC
-35 -10 *
1388 1403 1418
CTC CAT CAG GAC GGT ATT TTG CGC ACC AGC GTC GCT TTT CCG CAG GCC MC GCA CAG
Leu His Gln Asp Gly Ile Leu Arg Thr Ser Val Ala Phe Pro Gln Ala Asn Ala Gln
290 300 310 320 336
TCTAAAGCCG AGAGTTGTGC ACCACAGGAG TTTTMGACG C ATG TCT TCC CGC AAT AAT CCG GCG
S/D S/D MET Ser Ser Arg Asn Asn Pro Ala 1433 1448 1463 1478
CAG CAG GC0 CM GCC GM GAG ATG CTG TCG GCG ATT ATG CAG GAG CTG 0GC TAT CTG
Gln Gln Ala Gln Ala Glu Glu MET Leu Ser Ala Ile MET Gln Glu Leu Gly Tyr Val
351 366 381 396
CGT GTC GCC ATC GTG ATG GGG TCC MA AGC GAC TGG GCT ACC ATG CAG TTC GCC 0CC
Arg Val Ala Ile Val MET Gly Ser Lys Ser Asp Trp Ala Thr MET Gln Phe Ala Ala 1493 1508 1523 1538
Downloaded from http://jb.asm.org/ on April 28, 2021 by guest
GGC GTG ATG GCG ATG GAG TGT TTT GTC ACC CCG CM GGT CTG TTG ATC MC GM CTG
Gly Val MET Ala MET Glu Cys Phe Val Thr Pro Gln Gly Leu Leu Ile Asn Glu Leu
411 426 441 456
GM ATC TTC GM ATC CTG MT GTC CCG CAC CAC GTT GM GTG GTT TCT GCT CAC CGC
Glu Ile Phe Glu Ile Leu Asn Val Pro His His Val Glu Val Val Ser Ala His Arg 1553 1568 1583 1598
CCA CCG CGT GTG CAT MC AGC GGT CAC TGG ACA CM MC GGT GCC AGC ATC AGC CAG
Ala Pro Arg Val His Asn Ser Gly His Trp Thr Gln Asn Gly Ala Ser Ile Ser Gln
471 486 501 516
ACC CCC GAT MA CTG TTC AGC TTC GCC GM AGC GCC GM GAG MC GGT TAT CAG GTG
Thr Pro Asp Lys Leu Phe Ser Phe Ala Glu Ser Ala Glu Glu Asn Gly Tyr Gln Val 1613 1628 1643 1658
TTT GAG CTG CAT CTG CGG GCG ATT ACC GAT CTG CCG TTA CCG CAA CCA GTG GTG MT
Phe Glu Leu His Leu Arg Ala Ile Thr Asp Leu Pro Leu Pro Gln Pro Val Val Asn
531 546 561
ATT ATT GCG GGC GCA GGC GGC GCA GCG CAT CTG CCA GGC ATG ATT GCC GCC MA ACG
Ile Ile Ala Gly Ala Gly Gly Ala Ala His Leu Pro Gly MET Ile Ala Ala Lys Thr 1673 1688 1703
MT CCG TCG GTG ATG ATC MT CTG ATT GGT AGC GAT GTG MT TAT GAC TGG CTG AAA
Asn Pro Ser Val MET Ile Asn Leu Ile Gly Ser Asp Val Asn Tyr Asp Trp Leu Lys
576 591 606 621
CTG GTG CCG GTG CTG GGC GTG CCA GTA CAG AGC GCC GCA CTG AGC GGT GTC GAT AGC
Leu Val Pro Val Leu Gly Val Pro Val Gln Ser Ala Ala Leu Ser Gly Val Asp Ser 1718 1733 1748 1763
CTG CCO CTG GTG CAT CTG CAC TGG TAC GAC MA GM GTC CGT CCG GGG CGT MA GTG
636 651 666 681 Leu Pro Leu Val His Leu His Trp Tyr Asp Lys Glu Val Arg Pro Gly Arg Lys Val
CTC TAC TCC ATC GTA CM ATG CCG C0C 0C0 ATT CCG GTG GGT ACG CTG GCG ATT GGT
Leu Tyr Ser Ile Val Gln MET Pro Arg Gly Ile Pro Val Gly Thr Leu Ala Ile Gly 1778 1793 1808 1823
G0G CAT CTG MT TTG ACC GAC AGC GAC ACA TCG CGT CTG ACT GCG ACG CTG GAA GCC
Gly His Leu Asn Leu Thr Asp Ser Asp Thr Ser Arg Leu Thr Ala Thr Leu Glu Ala
696 711 726 741
AM GCT GGC GCG GCA MC GCG GCG TTA CTG GCA GCA CM ATT CTT GCG ACT CAT GAT
Lys Ala Gly Ala Ala Asn Ala Ala Leu Leu Ala Ala Gln Ile Leu Ala Thr His Asp 1838 1853 1868 1883
TTA ATC CCG CTG CTG CCG CCG GM TAT GCC AGC GGC GTG ATT TGG GCG CAG AGT MG
Leu Ile Pro Leu Leu Pro Pro Glu Tyr Ala Ser Gly Val Ile Trp Ala Gln Ser Lys
756 771 786 801
MA GMA CTG GAO G COT CTG MT GAC TGG CGC MA GCC CAG ACC GAC GM GTG CTG
Lys Glu Leu His Gln Arg Leu Asn Asp Trp Arg Lys Ala Gln Thr Asp Glu Val Leu 1905 1915 1925, 1935 1945 1955
TTC GOT TM CTGGTGCTCT ATTCTTGCCG GATGCGGCGT AAACGCCTTA TCCGGCCTAC CGATCCCGTA
Phe Gly REP
816 831 848
GM MC CCG GAC CCG CGA GGT GCG GCA TG AAA CAG GTT TGC GTC CTC GGT MC GGG CAG
Glu Asn Pro Asp Pro Arg Gly Ala Ala 1965 1975 1985 1995 2005 2015 2025.
S/D Met Lys Gln Val Cys Val Leu Gly Asn Gly Gln CCCATTGTAG GCCTGATMG ATGCGTCMG CATCGCATCA GGCATTGTGC ACCMTTGCC GCATGCGGCA
REP
863 878 893 908
TTA GGC CGT ATG CTG CGT CAG CCA GGC GM CCG TTA GGC ATT GCT GTC TGG CCA GTC 2035 2045 2055 2065 2075 2085 2095
Leu Gly Arg MET Leu Arg Gln Ala Gly Glu Pro Leu Gly Ile Ala Val Trp Pro Val CCGGTTGTAG OCCTGATMG ACGCGTCMG CGTCGCATCA GGCACAMTG TCTMTGCCT ACGACTACAG
REP
923 938 953 968
GGG CTG GAC GCT GM CCG GCG 0C0 GTG CCT TTT CM CM AGC CTG ATT ACC GCT GAG 2105 2115 2125 2135 2145 2155 2165
Gly Leu Asp Ala Glu Pro Ala Ala Val Pro Phe Gln Gln Ser Val Ile Thr Ala Glu CGMATACAG GTCCCCGCTT CGCCCGCCAG CGTCTCTTCA ATTCGCGATA ACGCCCMTC CACGCGGGTT
983 998 1013 1028 2175 2185 2195 2205 2215 2225 2235
ATA GM CGC TGG CCG GM ACC GCA TTA ACC CGC GAG CTG GCG CCC GAT CCG GCC TTT TACCACGGCT TCTGACATM CCACTCACCG CCGTTACATT CGGCCCCATC GMCCATCGG CTTTGGCMA
Ile Glu Arg Trp Pro Glu Thr Ala Leu Thr Arg Glu Leu Ala Pro Asp Pro Ala Phe
2245 2255 2265 2275 2285 2295 2305
1043 1058 1073 1088
TGGCGCTMC TCATCCGGTG TGGCATGCCG MTGGCACGT TGCTGCGGCG TTCCCCAGTT TTCATATACC
GTG MC CGC GAT GTG TTC CCG ATT ATT GCT GAC CGT CTG ACT CAG MG CAG CTT TTC
Val Asn Arg Asp Val Phe Pro Ile Ile Ala Asp Arg Leu Thr Gln Lys Gln Leu Phe
2315 2325 2335 2345 2355 2365 2375
1103 1118 1133 CCATCAGCAT CGGTGAGGAT CACCAGTCCA TCTGCATTM TCTGCTCGGC GACCMCGCA GCGGCGAGAT
OAT MG CTC CAC CTG CCG ACT GCA CCG TGG CAG TTA CTT GCC GM CGC AGC GAG TGG
Asp Lys Leu His Leu Pro Thr Ala Pro Trp Gln Leu Leu Ala Glu Arg Ser Glu Trp
CT
1148 1163 1178 1193
CCT GC0 GTG TTT GAT CGT TTA GGT GAG CTG GCG ATT GTT MG CGT CGC ACT GOT GOT
Pro Ala Val Phe Asp Arg Leu Gly Glu Leu Ala Ile Val Lys Arg Arg Thr Gly Gly
FIG. 2. Nucleotide and deduced amino acid sequences of the purEK operon. The DNA sequence of the sense strand of the purEK operon
is shown. The numbering is from an HpaI site to a BgIII site at nucleotide 2377. The purE coding region is shown from nucleotide 322 to 831,
and the purK coding region is shown from nucleotide 828 to 1895. The proposed Shine-Dalgarno sequences (39) (S/D) for purE and purK are
underlined. The -10 and -35 regions are underlined and labeled. A common region of homology (purR binding site) found in other pur loci
(Tiedeman et al., Fed. Proc. 46:2218, 1987) is double underlined. An asterisk indicates the probable transcription start, and regions of dyad
symmetry distal to the purEK operon that form REP (8, 42) sequences are overlined.VOL. 171, 1989 purEK OPERON 209
1 2 3 methanogen AIR carboxylases, E. coli purE shared 38%
identity with Methanobrevibacter smithii and 40% identity
-MW with Methanobacterium thermoautotrophicum, which in-
creased to 59 and 62%, respectively, when conserved resi-
66,000 dues where included as aligned in Fig. 4.
In contrast, the C02-binding subunit encoded by E. coli
.-45,000 purK showed only 32% identity when aligned with the
sequence from B. subtilis; the identity increased to 58%
when conserved residues were considered. This alignment
includes several gaps, and thus the C02-binding subunits of
these two bacteria have diverged more than the catalytic
-25,00 subunits. This may be due to a relatively nonspecific nature
of CO2 binding. When either the E. coli or B. subtilis purK
sequence was compared with the AIR carboxylase methano-
gen sequences, no significant identity could be detected.
purEK operon REP sequences. Downstream of the purK
Downloaded from http://jb.asm.org/ on April 28, 2021 by guest
gene, three complete and one half repetitive extragenic
..;..00
........ palindromic (REP) sequences (8, 42) could be identified from
nucleotide 1912 to nucleotide 2069 (overlined in Fig. 2). In
addition to other postulated roles in the cell, REP sequences
have been shown to be involved with mRNA stability (24,
29), and thus these REP sequences may play the same role in
the expression of the purEK operon. In the interval between
FIG. 3. Labeling of plasmid proteins in UV-irradiated E. coli the REP sequences and the BglII site, no sequences with the
CSR603 and in its plasmid-carrying derivatives. Twenty-microliter features of rho-independent terminators (15) could be iden-
samples from CSR603 (lane 1), CSR603(pJK42) (lane 2), and tified with confidence. Thus, the terminus of the purEK
CSR603(pJK43) (lane 3) grown in the absence of adenine were operon remains uncharacterized.
loaded onto a 16% polyacrylamide-sodium dodecyl sulfate gel. After Identification of the 5' end of purEK mRNA. The 5' end of
the electrophoresis run (22 h at 40 V), the gel was dried and the purEK mRNA was determined by mung bean nuclease
subjected to autoradiography. Bovine serum albumin (66,000), oval- mapping. A 416-bp TaqI fragment that spanned the end of
bumin (45,000), chymotrypsin A (25,000), and ribonuclease A the purE gene was labeled at the 5' end with polynucleotide
(13,700) were used as molecular weight (MW) standards. Arrow- kinase and [y-32P]ATP after dephosphorylation with calf
heads indicate the protein bands of molecular weights 41,000 and
15,000 found in CSR603(pJK42) and the protein band of molecular intestinal alkaline phosphatase according to the procedures
weight 15,000 found in CSR603(pJK43). of Maniatis et al. (22). Total RNA extracted from strain
TX517 containing plasmid pJS131 was hybridized to the
labeled fragment, and after mung bean nuclease treatment,
example of a gene duplication involving the purine genes has the protected fragment was sized on a DNA sequencing gel
been reported for the AIR synthetase portion of the trifunc- with a Sanger dideoxy sequencing ladder as a standard. A
tional protein of Drosphila melanogaster (14'. protected fragment of approximately 131 nucleotides was
The purEK genes of B. subtilis were identified by the detected after treatment with various concentrations of
homology of their deduced amino acid sequence with the mung bean nuclease (Fig. 5). The size of this protected
sequences reported here (7). Thus, the AIR carboxylase fragment would indicate that the probable transcription
enzyme of B. subtilis resembles that of E. coli, in which two initiation nucleotide is 280 or 281. Within reasonable spacing
separate subunits are required for activity. While AIR car- constraints for RNA polymerase, the TATTCT sequence at
boxylase in methanogens appears to be a single polypeptide nucleotides 267 to 272 is the best candidate for the purEK
enzyme with two domains, AIR carboxylase from the eu- -10 region. This assignment is also supported by the RNA
bacteria E. coli and B. subtilis are enzymes made up of two polymerase binding studies reported earlier which localized
subunits. In contrast, AIR carboxylase appears to reside on the RNA polymerase-binding site to this region (17). At 17
the same polypeptide as SAICAR synthetase in chicken liver bp upstream of this -10 region, the sequence TTTTCC is the
(32). In summary, with the results reported here, we have best fit to the consensus -35 region (TTGACA; 13). Similar
three broad patterns of evolutionary strategy for AIR car- to the other pur loci that have been characterized to date,
boxylase activity: separate subunits in eubacteria, a single including the purF (21), purMN (40), purL (F. J. Schendal,
duplicated subunit in archaebacteria, and a bifunctional E. Mueller, J. Stubbe, A. Shiau, and J. M. Smith, Biochem-
polypeptide in eucaryotes. istry, in press), purA (S. A. Wolfe and J. M. Smith, J. Biol.
In Fig. 4, the AIR carboxylase sequences from B. subtilis, Chem., in press), and guaBA (43, 45) loci, the purEK control
M. smithii, and M. thermoautotrophicum are aligned for region mRNA does not appear to have potential for second-
comparison with the deduced E. coli sequences. The cata- ary structure.
lytic subunit (purE) of E. coli AIR carboxylase showed very Regulation of the purEK operon. As other previous studies
strong identity when aligned with the sequences from B. have shown, the purEK operon is coregulated with other loci
subtilis, and in turn, the sequences of these two eubacteria of the de novo purine biosynthesis pathway (10, 44). This
could be aligned and shared considerable identity with the coregulation was ascribed to the action of a regulatory
amino-terminal portion of the methanogen AIR carboxyl- protein encoded by a "purR" locus, but these trans-acting
ases. Overall, the E. coli and B. subtilis catalytic subunits regulatory mutations have not been genetically character-
shared 57% identity, which increased to 72% if conserved ized or confirmed in their mode of action. Recently a purR
residues were included and no gaps were needed to align the locus encoding a regulatory protein controlling the expres-
sequences. When compared with the first 170 residues of the sion of the purF operon has been cloned and sequenced. The210 TIEDEMAN ET AL. J. BACTERIOL.
A
E.c. 1 mssrnnparVaIvMGSkSDWaTMqfAaeIfeiLNVPhhveVVSAHRTPDklFsfAEsAeEnGyqVIIAGAGG
1 1111 1I.. 1111
11 11 111111111 .1 .11.1 11111111
B.s. 1 MqPlVgIIMGStSDWeTMkhAcdILdeLNVPYekKVVSAHRTPDfmFEyAEtArErGIKVIIAGAGG
11 11 11 . II- 1--11 1--1111111-- 1-- . 11111 I11
M.s. 1 MtPkVMIILGSgSDiaIAEKsMkILEkLeIPYsLKiASAHRTPDlVrElVvqgtnAGIKVFIGIAGL
1.111.111 Il. 1iii 1 .1 III I..IIIIII 1- .1 Il. I111
M.t. 1 MkPrVMIlLGSaSDfrIAEKaMeIfEeLrIPYdLrvASAHRThekVkaiVseavkAGveVFIGIAGL
E.c. 73 AAHLPGMiAAKTlvPVlGVPVQSaALsGvDSLySIVQMPrGiPVgTlaIGKAGAaNAaLLAAQILathDke
1111111 1111 1111
11.III 11 1I111I11111 1.11 111111.11 1111111 .1.
B.s. 68 AAHLPGMtAAKTtlPVIGVPVQSkALnGmDSLlSIVQMPgGvPVATtsIGKAGAvNAGLLAAQILsafDed
111111. 111 1111111 .I1 1I11I1111.I1111 .. ....
M.s. 68 AAHLPGaIAAyTHkPVIGVPV-DVKvsGLDALySsVQMPyPsPVATVGiDRGdNGAILAArIlGLyDeeiR
111111 11 1111111 III. 11111. 111.1 IIIIII III I 11111 1.1 .1
M.t. 68 sAHLPGmIsAnTHrPVIGVPV-DVKlgGLDALfacsQMPfPaPVATVGvDRGeNaAILAAqIiGigDpgvR
Downloaded from http://jb.asm.org/ on April 28, 2021 by guest
E.c. 144 LhqrLndwRkaqtdeVLEnpDprgaa
1. .1. III
B.s. 139 LarkLderRentkqtVLEssDqlv
*. ...
M.s. 138 -kkVLeskeGyrqkViknnEeiVqkidnphitndflriknlElnetteefngsyinknaevVIivGrhtD1
M.t. 138 -erVadlrrGfyerVrrd EcqVlnsiegsy ---- yaplevEmppigdkvpsdsqddpmvsVI-pGsysDm
M.s. 208 itgKKvsvtLdRlkIphDmqVIcPIRsgkkFraYvNTMkNaKiFIgInsnSsqVsGglVgLtekPVIGVPC
11 .. 1.1. I. II III .1 I. 1.1.11 1.1 .1 111111
M.t. 202 kiaKKttmfLeRmgIsyDlnVIsPIRyperFerYlekMeNvKlFIaIsglSahVtGavVaLsdrPVIGVPC
M.S. 279 -enelGnnyLLStvNMPPGVPVaTVGVnNGrNAA vLsgEiLsInnpvllelleklKnkkini
111 *11111111 1111 11 111-1 1-1 1.I.*--
M.t. 273 plkmnGwdsLLSmiNMPPGVPVgTVGVgNGgNAAiLaaEmLgIydekiesrikriKsrSvkf
B
E.c. 1 MkqVcvlgnGqlgrmlrqage-PlGiaVwpVgldA-----Epaavpfqq
.. . . 1-1 1.
B.s. 1 lskqiiypgavigiigggqlgkMmaVsakqmGykvavvdpvkdsPcGq-VadVeitAhyndrEairklaei
E.c. 44-S-vITaEiErwpetALtrelapdpafvnrdvfpIiadRlTqKqlfdklhlptAPwqllaersEw-pAVfd
.11 11 ... *1 Ill ....- 11- 1- 11 .
B.s. 71 SdiITyEfEnidydALhwlkdhaylpqgselllItqnReTeKkaiqsagcevAPysivktknElkqAVqel
E.c. 112 RLgelAivKrrtGGYDGrGQwrlran-etEQlpAec-yGeCIvEqginFsgEvSlvgaRgfdGstvfyPlt
II. 11 11111 11. II.I.I .1 .I. .1 .1 .1
B.s. 142 RLp--AvlKtcrGGYDGkGQfvikeeaqmEQaaAllehGtCIlEswvsFkmElSvivvRsvnGeistfPta
E.c. 181 hNlHqdgILrtS-VafpqanaQQqaqAeemlsaimqELgyVGvmAmEcFvTpqG-LLiNELAPRvHNSGHw
1I
. 11
1 111 1.I11 11 .1.1.1.1 .1 11.111111 11111.
B.s. 211 eNiHhnnILfqSiVpavekgiQQ--kAadlavkladELnlVGplAvEmFlTedGeLLvNELAPRpHNSGHy
E.c. 250 TqngasiSQFElHlRAitdLPLpqpvvnnPsvMiNLiGsd--- VnyDwlkLplvhLhwYdK-EvrpGRKvG
. . 1111 11 III1 .1 11 . 1- I. 111 .1
B.s 280 TldlcetSQFEqHiRAvcgLPLgktdllkPgmMvNLlGdevklVeeDpelLkeakLyiYgKhEikkGRKmG
E.c. 317 HlnlTdsdtsrltatlEalIpllppeyasGviwaqskfg
1. .
B.s. 351 Hi--TfmkqpedewiqE--ItnkwmnrdgGqae
FIG. 4. Amino acid alignment of E. coli, B. subtilis, Methanobrevibacter smithii, and Methanobacterium thermoautotrophicum AIR
carboxylases (A) and alignment of purK subunits from E. coli and B. subtilis (B). The E. coli (Ec.) sequence was taken from Fig. 2 and aligned
by computer analysis (23) and visual inspection to the AIR carboxylase sequences of B. subtilis (B.s.; 7), M. smithii (M.s.; 12), and M.
thermoautotrophicum (M.t.; 11). All sequences start with residue 1 and the numbering scheme does not include gaps. Vertical lines indicate
identity; dots indicate conserved residues according to the system of Amuro et al. (1).VOL. 171, 1989 purEK OPERON 211
purR
purF CTACGCAAACGTTTTCTTTTTCTGTTAGAATGCGCCCCGAA
111111 1111
lilt 11111111i111
IlI
purMN TCTCGCAAACGTTTGCTTTCCCTGTTAGAAT-TGCGCCGAA
I 11111 1111111111 11 I 11 III
purEK CCACGCAACCGTTTTCTTTCCGTGCTATTCTCTGTGCCCTC
-10
CTTGCTCTC
FIG. 6. Comparison of the purEK, purMN, and purF control
131-'j regions. The alignment of the purMN and purF control regions are
taken from Smith and Daum (40), and the purEK sequence is from
Fig. 2. The -10 regions and purR binding site are labeled. The
alignment of the purEK control region sequence to the purMN and
purF alignment (3) was by visual examination with respect to the
respective -10 regions.
Downloaded from http://jb.asm.org/ on April 28, 2021 by guest
pressed under excess-purine growth conditions. This defin-
ww itively establishes the coregulation of the purEK and purF
operons by a common regulatory protein encoded by the
purR locus.
When the purEK control region was compared with the
purMN and purF control regions, a 9-bp displacement was
necessary to align the purEK control region with the con-
served sequence (36 of 39 nucleotides) present in their
control regions. It is of interest that this displacement
corresponds to the approXimately one turn of the DNA
FIG. 5.Identification of the 5' end and purEK mRNA. RNA
double helix. The purF, purMN, and purEK control regions
from strain TX517 containing plasmid pJS131 was hybridized to a
are aligned for comparison in Fig. 6, and with the allowance
5'-end-labeled 416-bp TaqI fragment that spanned the 5' end of the
for the 9-bp displacement, the control regions of the three
purE gene. Lanes: 2, 4, 5, and 7, dideoxy sequencing reaction's used
loci are very similiar with regard to the location of the purR
as size standards; 1, control r'eaction lacking RNA; 3 and 6, binding site, -10 regions, and conserved nucleotides.
protected fragment obtained after mung bean nuclease digestion at
60 and 120 U, respectively. After adjustment for the phosphate
ACKNOWLEDGMENTS
group (41), a major protected band of 131 tiucleotides was obtained.
This work was supported by Public Health Service grants AI
20068 to J.M.S. and GM 29500 to I.S.G. from the National Institutes
binding site of this regulator'y 'p-rotein in the purF control
of Health. Computer resources used to carry out our studies were
region has been identified by mutational and DNase I foot-
provided in part by the BIONET National Computer Resource for
Molecular Biology funded by Public Health Service grant 1 U41
printing studies (R. J. Rolfes and H. Zalkin, J. Biol. Chem.,
RR-01685-02 from the National Institutes of Health.
in press). The purR-binding site is part of a highly conserved
We thank Koral Schlotman and Karen Kinch for excellent typing
sequence (36 of 39 nucleotides identical) found during an and manuscript preparation.
earlier comparison of the purMN and purF control regions
(40) and should be present in other loci regulated by the purR LITERATURE CITED
protein. We have found common sequences corresponding 1. Amuro, N., J. L. Paluh, and H. Zalkin. 1985. Replacement by
to the purR-binding site (consenisus sequence GCAAACGT site-directed mutagenesis indicates a role for histidine 170 in the
TTGC) in the cont'rol regions of the other pur loci charac- glutamine amide transfer function of anthranilate synthase. J.
terized to date, with the exception of the purA locus (Tie- Biol. Chem. 260:14844-14849.
demhan et al., Fed. Proc. 46:2218, 1987). 2. Bachmann, B. J. 1983. Linkage map of Escherichia coli K-12,
Analysis of the purEK control -region showed a potential
edition 7. Microbiol. Rev. 47:180-230.
purR-binding site at nucleotides 237 to 249 (underlined in
3. Becker, M. A., K. 0. Raivio, and J. E. Seegmilier. 1979.
Fig. 2), iti agreement with earlier studies showing coregula-
Synthesis of phosphoribosylpyrophosphate in mammalian cells.
Adv. Enzymol. 49:281-306.
tion with the purF operon. The interaction of the purR 4. Charles, H. P., and G. A. Roberts. 1968. Carbon dioxide as a
regulatory protein with the purEK operon control region was
growth factor for mutants of Escherichia coli. J. Gen. Micro-
established by the following experiments. By using a lacZY- biol. 51:211-224.
Kan fusion cassette (46) inserted into plasmid pJS131 at the 5. Clarke, L., and J. Carbon. 1976. A colony bank containing
AsuIl and BsmI restrictiofn sites (nucleotides 411 to 660, Fig. synthetic ColEl hybrid plasmids representative of the entire E.
2), the resulting purE-1acZYnd:Kan fusion was movred on'to coli genome. Cell 9:91-99.
the E. coli by the procedure of Witiatis et al.
chromosome 6. Dagert, M., and S. D. Ehrlich. 1979. Prolonged incubation in
(48). After transfer
wild-type pur genetic background
to a
calcium chloride improves the competence of Escherichia coli
cells. Gene 6:23-28.
(strain TX302) to form strain TX725, a purR: :TnJO mutation 7. Ebbole, D. J., apd H. Zalkin. 1987. Cloning and characterization
allelic to the purR mutations of Rolfes and Zalkin (in press) of a 12-gene cluster from Bacillus subtilis encoding nine en-
was introduced by P1 transduction to form strain TX729. In
zymes for de novo purine nucleotide synthesis. J. Biol. Chem.
strain TX729 the purE-1acZY:i:Kan fusion was no longer 262:8274-8287..
subject to normal regulation and was constitutively dere- 8. Gilson, E., J. M. Clement, D. Brutlag, and M. Hofnung. 1984. A212 TIEDEMAN ET AL. J. BACTIERIOL.
family of dispersed repetitive extragenic palindromic DNA 31. O'Reilly, C:., P. D. Tuer, P. F. Snith-Keary, and D. J.
sequences in E. coli. EMBO J. 3:1417-1421. McConneil. 1984. Molecular cloning of genes involved in purine
9. Gold, L., and G. Stormo. 1987. Translational initiation, p. 1302- biosynthetic and salvag pathways of Salmonella typhimurium.
1307. In F. C. Neidhardt, J. L. Ingraham, K. B. Low, B. Mol. Gen. Genet. 1%:152-157.
Magasanik, M. Schaechter, and H. E. Umbarger (ed.), Esche- 32. Patey, C. A., and G. Slew. 1973. Purification and properties of
richia coli and Salmonella typhimurium: cellular and ffiolecular an enzyme duet, phosphoribosylaninoimidazole carboxylase
biology. American Society for Microbiology, Washington, D.C. and phosphoribosylaminoinidazolesuccinocarboxamide syn-
10. Gots, J. S., C. E. Benson, B. Jochimapa, and K. R. Koduri. 1977. thetase, involved in the biosynthesis of purine nucleotides de
Microbial models and regulatory elements in the control of novo. Biochem. J. 135:543-545.
purine metabolism. CIBA Found. Symp. 48:23-41. 33. Salser, W., R. F. Gestelad, and A. Dole. 1967. In vitro
11. Hamiton, P. T., and J. N. Reeve. 1985. Sequence divergence of synthesis of bacteriophage lysozyme. Nature (London) 215:
an archaebacterial gene cloned from a mesophilic and a thermo- 588-591.
philic methanogen. J. Mol. Evol. 22:351-360. 34. Sancar, A., A. M. Hak, and W. D. Rup. 1979. Simple method
12. Hailton, P. T., and J. N. Reeve. 1985. Structure of genes and for identification of plasmid-coded proteins. J. Bacteriol. 137:
an insertion element in the methane producing archaebacterium 692493.
Methanobrevibacter smithii. Mol. Gen. Genet. 200:47-59. 35. Sancar, A., and C. S. Rupert. 1978. Determination of plasmid
13. Harley, C. B., and R. P. Reynolds. 1987. Analysis of E. coli molecular weights from ultraviolet sensitivities. Natulre
promoter sequences. Nucleic Acids Res. 15:2343-2361. (London) 272:471-472.
Downloaded from http://jb.asm.org/ on April 28, 2021 by guest
14. Henikoff, S. 1986. The Saccharomyces cerevisiae ADE5,7 pro- 36. Sanderson, K. E., and J. R. Roth. 1983. Linkage map of
tein is homologous to overlapping Drosophila melanogaster Salmonella typhimurium, edition VI. Microbiol. Rev. 47:410-
Gart polypeptides. J. Mol. Biol. 190:519-528. 453.
15. Holmes, W. M., T. Platt, and M. Rosenberg. 1983. Termination 37. Sanger, F., A. R. Couio, B. G. BDrrl, A. J. Smith, and B. A.
of transcription in E. coli. Cell 32:1029-1032. Roe. 1980. Cloning in single-stranded bacteriophage as an aid to
16. Jacob, F., A. Ulhnann, and J. Monod. 1965. Deletions fusion- rapid DNA sequencing. J. Mol. Biol. 143:161-178.
nant l'operon lactose et un operon purine chez Escherichia coli 38. Sanger, F., S. Nickn, and A. R. Co_s|. 1977. DNA sequenc-
J. Mol. Biol. 13:704-719. ing with chain-terminating inhibitors. Proc. Natl. Acad. Sci.
17. Kamholz, J., J. Keyhani, and J. S. Gots. 1986. Molecular cloning USA 74:5463-5467.
and characterization of the purE operon of Escherichia coli. 39. Shine, J., and L. D)ao. 1974. The 3'-terminal sequence of
Gene 44:55462. Escherichia coli 16S ribosomal RNA: conptementarity to non-
18. Larson, R., and J. Messing. 1983. Apple II computer software sense triplets and ribosome binding sites. Proc. Natl. Acad. Sci.
for DNA and protein sequence data. DNA 2:31-35. USA 71:1342-1346.
19. Levine, R. A., and M. W. Taylor. 1982. Regulation of purE 40. Smith, J. M., and H. A. Daum mH. 1986. Nucleotide sequence of
transcription in a purE::lac fusion strain of Escherichia coli. J. the purM gene encoding 5'-phosphoribosyl-5-aminoimidazole
Bacteriol. 149:1041-1049. synthetase of Escherichia coli K-12. J. Biol. Chem. 261:10632-
20. Lukens, L. N., and J. M. Buchanan. 1959. Biosynthesis of the 10636.
purines. XXIV. The enzymatic synthesis of 5-amino-1-ribosyl- 41. Soner-Webb, B., and R. H. Reeder. 1979. The nucleotide
4-imidazolecarboxylic acid 5'-phosphate from 5-amino-1-ribosy- sequence of the initiation and termination sites for ribosomal
limidazole 5'-phosphate and carbon dioxide. J. Biol. Chem. 234: RNA transcription in X. laevis. Ceil 18:485-499.
1799-1805. 42. Stern, M. J., G. F. Ames, N. H. Smidh, E. C. Robi , and
21. Makaroff, C. A., and H. Zalkin. 1985. Regulation of Escherichia C. F. HIggin. 1984. Repetitive extragenic palindromic se-
colt purF. Analysis of the control region of a pur regulon gene. quences: a major component of the bacterial genome. Cell 37:
J. Biol. Chem. 260:10378-10387. 1015-1026.
22. Manintls, T., E. F. Fritsch, and J. Sambrook. 1982. Molecular 43. Thoms, M. S., and W. T. Drabbe. 1985. Nucleotide sequence
cloning: a laboratory manual. Cold Spring Harbor Laboratory, and organisation of the gua promoter region of Eschertichia coli.
Cold Spring Harbor, N.Y. Gene 36:45-53.
23. Martinez, H. M. 1988. A flexible multiple sequence alignment 44. Thomulka, K. W., and J. S. Gos. 1982. Isolation and charac-
program. Nucleic Acids Res. 16:1683-1691. terization of purine regulatory mutants of Salmonella typhimu-
24. Merino, E., B. Becerril, F. Vaile, and F. Boivar. 1987. Deletion rium with an episomal purE-lac fusion. J. Bacteriol. 151:153-
of a repetitive extragenic palindromic (REP) sequence down- 161.
stream from the structural gene of Escherichia coli glutamate 45. Tiedeman, A. A., and J. M. Smith. 1985. Nucleotide sequence of
dehydrogenase affects the stability of its mRNA. Gene 58:305- the guaB locus encoding IMP dehydrogenase of Escherichia
309. coli K12. Nucleic Acids Res. 13:1303-1316.
25. Messing, J. 1979. A multi-purpose cloning system based on the 46. Tiedeman, A. A., and J. M. SmIth. 1988. lacZY gene fusion
single-stranded DNA bacteriophage M13. Recomb. DNA Tech. cassettes with KanR resistance. Nucleic Acids Res. 16:3587.
Bull. 2:43-48. 47. Tiedeman, A. A., J. M. , and H. ZaUin. 1985. Nucleotide
26. Messing, J. 1983. New M13 vectors for cloning. Methods sequence of the guaA gene encoding GMP synthetase of Esch-
Enzymol. 101:20-78. erichia coli K12. J. Biol. Chem. 2608676-8679.
27. Mller, J. H. 1972. Experiments in molecular genetics. Cold 48. Winans, S. C., S. J. Elledge, J. H. Krueger, and G. C. Waker.
Spring Harbor Laboratory, Cold Spring Harbor, N.Y. 1985. Site-directed insertion and deletion mutagenesis with
28. Neidhardt, F. C., P. L. Bloch, and D. F. Smith. 1974. Culture cloned fragments in Escherichia coli. J. Bacteriol. 161:1219-
medium for enterobacteria. J. Bacteriol. 119:736-747. 1221.
29. Newbury, S. F9., N. H. Smith, and C. F. Higgins. 1987. Differ- 49. Wolfe, S. A., anid J. M. Smith. 1985. Separate regulation of purA
ential mRNA stability controls relative gene expression within a and purB loci of Escherichia coli K-12. J. Bacteriol. 162:822-
polycistronic operon. Cell 51:1131-1143. 825.
30. Normark, S., S. Bergstrom, T. Edund, T. Grundstrom, B. 50. Yanisch-Perron, C., J. Vieirt, and J. Messig. 1985. Improved
Jaurin, F. P. Lindberg, and 0. (Mson. 1983. Overlapping genes. M13 phage cloning vectors and host strains: nucleotide se-
Annu. Rev. Genet. 17:499-525. quences of the M13mpl8 and pUC19 vectors. Gene 33:103-119.You can also read
