Volume 13 Number 3 1985 Nucleotide sequence and biochemical characterization of the metJ gene from SalmoneUa typhimurium LT2 Mark L.Urbanowski and George V.Stauffer Department of Microbiology , University of Iowa , Iowa City , IA 52242 , USA ABSTRACT The nucleotide sequence of the Salmonella typhimurium metW gene is presented along with the sequence of the promoter region for the closely linked metB gene .
The two genes are transcribed in opposite directions , with transcription initiating from a single promoter for metB , and from two apparent promoters for metJ .
RNA polymerase binding sites for metJ and metB , determined by in-vitro protection studies , lie adjacent to each other and may overlap .
The two metJ promoters , P and PJ2 , are separated by approximately 65 base pairs .
Binding of RNApolymerase in-vitro could only be observed for PJll even though transcripts are initiated from both promoters in-vivo .
The metJ gene codes for a polypeptide of 105 amino-acids with a calculated Mr of 12,110 .
The translation start site was determined by N-terminal amino-acid sequence analysis of a metJ-lacZ fusion protein .
INTRODUCTION The metJ gene in Salmonella typhimurium and Escherichia coli has been shown to code for a protein involved in the regulation of the methionine biosynthetic pathway , possibly acting as a classical repressor ( 1-3 , 17 ) .
We have recently reported the cloning of this gene and the closely linked metB gene of S. typhimurium ( 4 ) .
Here we present the complete nucleotide sequence of the metJ gene and a biochemical characterization of its control region and the control region of the metB gene .
MATERIALS AND METHODS Bacteria and plasmids The S. typhimurium wild type strain JL781 was used for preparing total cellular RNA .
Plasmid pGS107 carries the S. typhimurium metJ and metB genes and has been described ( 4 ) .
Plasmid isolation Plasmid DNA was prepared as described previously ( 4 ) .
DNA sequence analysis DNA sequencing methods were those of Maxam and Gilbert ( 5 ) .
Gel elec ¬ © I RL Press Limited , Oxford , England .
The SI nuclease mapping procedure of Weaver and Weissman ( 6 ) was used .
A 396 base pairs ( bp ) MluI fragment , which includes the metJ and metB trans-cription initiation regions , was labeled at the 5 ' ends with y-32P-ATP using T4 polynucleotide kinase ( 14 ) .
The labeled strands were separated ( 5 ) and aliquots of each strand precipitated with 50 pg of total cellular RNA isolated from either JL781 grown in Luria broth , or JL781 carrying pGS107 grown either in glucose minimal-medium or glucose minimal-medium plus methionine .
The mixtures were resuspended in 20 p1 of hybridization buffer ( 0.4 M NaCl , 0.04 M PIPES , pH 6.4 , 1 mM EDTA ) , heated to 90 °C for 2 minutes , and then allowed to hybridize at 55 °C for 1 hour .
Two hundred p1 of cold SI nuclease buffer ( 0.25 M NaCl , 0.06 M Na acetate , pH 4.6 , 1 mM ZnSO4 ' and 5 % glycerol ) containing 300 units of Si nuclease was added and digestion was carried out at 20 °C for 1 hour .
The Si nuclease-resistant products were then precipitated and run alongside a sequencing ladder of the same DNA strand .
3 ' SI nuclease mapping The procedure used to map the 3 ' terminus of the meWt transcript was similar to the 5 ' SI nuclease mapping given above , with the following modifications .
Suitable fragments containing the termination region of the metJ gene could not be separated into single strands .
Therefore , a double stranded 621 bp MluI-ClaI fragment labeled at the 3 ' MluI end using a-32P-dCTP and the large fragment of E. coli polymerase I ( 14 ) was used as a probe .
The labeled DNA was precipitated along with 50 pg of total cellular RNA isolated from JL781 carrying plasmid pGS107 and grown in glucose minimal-medium .
The mixture was resuspended in 20 pl of an 80 % formamide hybridization buffer , heated to 900C for 2 minutes , and allowed to hybridize 4 hours at 60 °C .
The sample was then treated as in the 5 ' mapping procedure above .
RNA polymerase binding sites The RNA polymerase binding sites for the meWt and metB genes were determined using the DNAse I `` footprinting '' procedure of Schmitz and Galas ( 8 ) .
The 396 bp MluI fragment carrying the metJ and metB control regions was labeled at a single 3 ' end using a-32P-dCTP and the large fragment of E. coli polymerase I .
The fragment was incubated 15 minutes at 37 °C in 100 pl of binding buffer ( 10 mM Tris-HCl , pH 7.9 , 0.1 mM ATP and GTP , 125 mM KCl , 10 mM MgCl2 , 10 mM CaCl2 , 0.1 mM dithiothreitol , and 25 % glycerol ) , and then split into two samples .
To one sample 7 pg of RNA polymerase holoenzyme was added , and the incubation continued for 10 minutes at 37 °C .
trophoresis was carried out according to Sanger and Coulson ( 13 ) .
5 ' SI nuclease mapping DNAse I was then added to both tubes to give a final concentration of 0.13 pg/ml .
The DNAse I reaction was stopped after 30 seconds by the addition of 25 pl of a 3 M ammonium acetate and 0.25 M EDTA solution containing 0.15 mg/ml sonicated calf thymus DNA .
The DNAse I-generated fragments were ethanol precipitated and run alongside a sequencing ladder of the same fragment .
Purification of the metJ-lacZ fusion protein The construction of the metJ-lacZ fusion will be described elsewhere .
It consists of the first 41 amino-acid codons of the metJ gene fused to the eighth codon of the f-galactosidase gene carried on plasmid pMC1403 , constructed according to the method of Casadaban et al. ( 15 ) .
Purification of the hybrid protein was by affinity chromatography according to Steers ( 16 ) .
Approximately 3 mg of the purified fusion protein was subjected to automated N-terminal amino-acid sequencing on a Beckman 890-C sequencer .
Enzymes and chemicals Restriction enzymes and other DNA modifying enzymes were obtained from BRL ( Gaithersburg , MD ) or NEB ( Beverly , MA ) with the exception of T4 poly-nucleotide kinase which was obtained from P-L Biochemicals ( Milwaukee , WI ) .
The p-aminophenyl-p-D-thio-galactopyranoside-agarose resin used in the affinity chromatography purification of the metJ-lacZ fusion protein was purchased from Sigma ( St. Louis , MO ) .
All other chemicals were reagent grade and commercially available .
RESULTS The metJ gene was originally isolated on a 19 kb EcoRI DNA fragment and was subcloned on a 3100 bp ClaI fragment into plasmid pBR322 ( designated pGS107 ) ( 4 ) .
Further analysis localized the gene to a 1000 bp MluI-ClaI fragment within pGS107 .
A physical map of this 1000 bp MluI-ClaI fragment is presented in Fig. 1 along with the DNA sequencing strategy employed .
The DNA sequence was determined for both strands by the Maxam-Gilbert technique ( 5 ) , and all restriction enzyme sites were overlapped .
The nucleotide sequence for the S. typhimurium metJ gene and the deduced amino-acid sequence are shown in Fig. 2 .
Also shown is the nucleotide sequence for the control region of the metB gene .
Location of the 5 ' ends of metJ and metB mRNA Our previous work using Tn5 insertion mutagenesis indicated that the metJ promoter region would most likely be situated on a 396 bp MluI fragment ( 4 ) .
The Sl-nuclease mapping procedure of Weaver and Weissman ( 6 ) was used to map the 5 ' ends of the metW and metB mRNA transcripts using the NluI fragment as a probe .
As shown in Figure 3 , two apparent 5 ' termini exist for the metJ transcripts , separated by approximately 65 bp .
Although the S1-nuclease-mapping procedure is not strictly quantitative , a comparison of the band intensities of the protected fragments within a particular lane is valid .
Note in Fig. 3A an increase in the fraction of transcripts initiating from PJ2 relative to PJ1 when the RNA for hybridization was isolated from cells grown with methionine ( lane c ) compared to those grown without methionine ( lane b ) .
Thus , transcription initiation at these two sites appears to be regulated in the opposite manner by the presence of methionine in the growth medium .
A single 5 ' terminus was observed for the metB transcript .
The most likely transcription initiation sites deduced from these results are shown in Figure 2 .
Sequences that resemble the consensus sequences for the -10 Pribnow box and -35 regions ( 7 ) are underlined for both metW transcripts ( PJ1 and PJ2 ) and the metB transcript ( PB ) .
In vitro RNA Polnerase protection of the NluI fragment Interactions of RNA polymerase with the proposed promoter regions were examined by binding RNA polymerase to the 396 bp NluI fragment and observing protection from subsequent DNAse I digestion according to the methods of Schmitz and Galas ( 8 ) .
The protected `` footprints '' of RNA polymerase bound to the metW and metB promoter regions are shown in Figure 4 .
The corres-ponding protected regions of the sequence are shown in Figure 2 .
~ I -4 ~ ~ ~ ~ ~ ~ ~ ~ ¬ MIu I Mulu Sph I ClalI 200 400 00 1000 met J Hpa 11 Rsa I Figure 1 .
Restriction endonuclease recognition sites and nucleotide sequence determinations used to establish the metJ sequence .
Arrows indicate the extent of each sequence determination .
Location of the 3 ' terminus of metJ mRNA The 3 ' end of metJ mRNA was determined by the Sl mapping procedure using a 621 bp MluI-ClaI fragment 32P-labeled at the MluI 31 terminus as a probe ( Materials and Methods ) .
The results of the Sl mapping gel and the proposed termination region are shown in Fig. 5 .
This proposed termination region is located about 40 bases distal to the two translation termination codons ending the metJ coding sequence ( Fig. 2 ) .
Amino acid sequence and composition The deduced amino-acid sequence of the metJ gene product shown in Fig. 2 specifies a polypeptide containing 105 amino-acid-residues .
A likely Shine-Dalgarno ribosome binding sequence ( 9 ) precedes the AUG initiator codon located at bases 126-128 .
The choice of this codon as the correct initiation codon is based on N-terminal amino-acid sequence studies of a fusion protein containing the amino-terminal 41 amino-acids of the metJ protein fused to the 8th amino-acid of 0-galactosidase .
Construction of this fusion will be presented elsewhere .
These studies show a perfect match of the first 12 amino-acids of the fusion protein with-amino-acids 2 through 13 of the deduced amino-acid sequence shown in Fig. 2 .
The formyl-methionine apparently 360-360-400 Figure 2 .
The nucleotide and deduced amino-acid sequence of the S. typhimurium metJ gene .
Heavy brackets above the sequence indicate the regions protected by RNA polymerase .
Arrows indicate the direction and most likely transcription initiation sites for the two metJ promoters ( P1i and PJ2 ) and the metB promoter ( PB ) .
The most likely Pribnow box and -35 regions for PJ , PJ2 ' and PB are underlined , as are the possible ribosome binding sites Shine-Dalgarno sequences ) for both genes .
A possible operator sequence ( 10 ) is boxed .
The two translation termination codons for metJ are indicated by asterisks , and the proposed transcription termination region is indicated by the vertical arrows .
B A A T T A G G C c a b c ) r f r r tI , l 0.4 G ATA TC ~ J1 a-A rr .
\ J2 o > a -- C > * TG GCC C C T A A T A C C .
The amino-acid composition of the metJ gene product , as deduced from the DNA sequence , is shown in Table 1 .
The calculated molecular weight of Mr 12,110 is in good agreement with the Mr 12,000 value estimated from SDS-polyacrylamide gel electrophoresis of the metJ gene product ( 4 ) .
The codon usage frequency is given in Table 2 .
DISCUSSION The promoter regions of the metJ and metB genes of Salmonella typhimur-ium appear to be complex .
S1-nuclease-mapping experiments show that the metJ gene has two promoters active in-vivo ( PJ1 and PJ2 ) , separated by approximately 65 bp .
However , when these regions are examined by the `` footprinting '' technique to determine the binding sites for E. coli RNA polymerase , only one of the metJ promoters ( PJ1 ) shows significant binding of polymerase Figure 3 .
Location of the 5 ' termini of the metJ and metB gene transcripts .
A 396 bp MluI fragment carrying the control regions for both the metJ and metB genes was labeled at the 5 ' termini with 2p , the strands separated , and each strand hybridized to total cellular RNA isolated from JL781 grown in Luria broth ( lane a ) , JL781 transformed with plasmid pGS107 and grown in glucose minimal-medium ( lane b ) , or JL781 carrying pGS107 and grown in glucose minimal-medium plus methionine ( lane c ) .
Hybridization mixtures were then treated with Sl nuclease and the Sl nuclease-resistant DNA products electrophoresed alongside a sequencing ladder of the original DNA strand .
Base numbering of nucleotides is in reference to Fig. 2 .
A , location of the 3 ' ends of the protected metJ DNA probe corresponding to the 5 ' termini of metJ mRNA at nucleotides +1 ( PJ1 ) and +61 to +65 ( PJ2 ) ; B , location of the 3 ' end of the protected metB DNA probe corresponding to the 5 ' terminus of metB mRNA at nucleotide -95 ( PB ) .
In vitro binding of RNA polymerase to metJ and metB control regions .
A 32P-labeled double strand DNA fragment containing the metJ and metB control regions was incubated either without ( lane a ) or with ( lane b ) E. coli RNA polymerase in the presence of 0.1 mM GTP and ATP , and subsequently partially digested with DNAse I .
The resulting mixture of digestion products was denatured and run alongside of a sequencing ladder of the original fragment .
The brackets show the region of the fragment protected from DNAse I digestion when RNA polymerase was present in the initial incubation reaction .
Asterisks indicate positions which appear to have enhanced susceptibility to DNAse I digestion .
The corresponding protected sequence is given in Figure 2 .
4t 68 G C C A C A-T G-C A-T C-G G-C G-C C-G Am X CATG-CGCTTTTTTAACG .
Location of the 3 ' terminus of metJ mRNA .
A 621 bp M1uI-Clal fragment carrying the distal part of the metJ gene and 32p-labelled at the 3 ' NluI terminus was hybridized to total cellularr RNA isolated from JL781 bearing plasmid pGS107 .
The hybridization mixture was then treated with 51 nuclease and the 51 nuclease-resistant products run alongside a sequencing ladder of the original DNA fragment .
The 3 ' end of metJ mRNA ( lane a ) is indicated by brackets and occurs at nucleotides 482-484 as numbered in Fig. 2 .
It can not be determined from this experiment whether transcription actually terminated at this region or whether meLt mRNA was processed to this site from a longer transcript .
The proposed metJ termination region is also shown .
The major 3 ' termini determined by the SI mapping are indicated by the arrows .
`` v ¬ ] ~ ~ T G C C A C Ws Cys 1 Ile 7 Pro 6 Val 4 Total number of residues = 105 ; calculated MW = 12,110 .
under the in-vitro conditions used .
It is possible that the failure to see RNA polymerase protection of PJ2 is due to improper conditions in the in-vitro binding reaction , e.g. , the absence of some binding factor .
Alternatively , RNA polymerase may have a much higher affinity for P1i and , when bound at Pill excludes further RNA polymerase binding at PJ2 ' However , we have not ruled out the possibility that the transcript from PJ2 is a processed form of the transcript initiated from PJi .
The region of DNA between P and the metB promoter ( PB ) includes a short palindromic sequence ( bases -36 to -59 , Fig. 2 ) which in E. coli has been proposed as a possible operator sequence ( 10 ) .
This sequence is fully conserved in S. typhimurium .
Interestingly , in S. typhimurium it binding at PJ2 ' with PJ2 acting as a low level maintenance promoter .
Consistent with this hypothesis is the observation that the relative amount of mRNA initiated by PJ1 decreases when the RNA for hybridization was isolated from cells grown in the presence of methionine compared to when cells are grown without methionine ( Fig. 3A , lanes b and c ) .
The opposite is observed for mRNA initiated at PJ2 ( Fig. 3A , lanes b and c ) .
The two sets of multiple bands seen for PJ2 are possibly the result of `` nibbling '' of the DNA fragments by S1 nuclease ( 11 ) , or may in fact be two separate initiation sites for PJ2 ' A single promoter was observed for the metB gene , and appears to be responsive to methionine addition to the medium during cell growth ( Fig. 3B , lanes b and c ) .
The location of this promoter , determined by S1-nuclease-mapping , is also consistent with the RNA polymerase protection experiment shown in Fig. 4 and summarized in Fig. 2 .
This location is not consistent , however , with the promoter location reported for the metB gene in E. coli Table 1 .
Amino acid composition of the metJ gene product .
Ala 8 Gln 3 Leu 10 Ser 6 Arg 7 Glu 14 Lys 8 Thr 6 2 Asn 3 Gly 4 Met Trp 2 Asp 6 His 3 Phe 2 Tyr 3 Table 2 .
Codon usage frequency in meWt .
Ser UCU 0 Tyr UAU 1 Ser UCC 1 Tyr UAC 2 Ser UCA 0 END UAA 0 Ser UCG 0 END UAG 1 Phe UUU Phe UUC Leu UUA Leu UUG 2 0 1 3 Cys UGU Cys UGC END UGA Trp UGG 1 0 1 2 Leu CUU Leu CUC Leu CUA Leu CUG 0 1 Pro CCU Pro CCC Pro CCA Pro CCG 1 1 1 3 His CAU His CAC Gln CAA Gln CAG 0 3 Arg CGU Arg CGC Arg CGA Arg CGG 4 2 1 0 0 5 2 1 Ile AUU Ile AUC Ile AUA Met AUG 3 4 0 2 Thr ACU Thr ACC Thr ACA Thr ACG 0 3 5 3 0 3 0 3 Asn AAU Asn AAC Lys AAA Lys AAG Ser AGU Ser AGC Arg AGA Arg AGG 1 4 0 0 Val GUU Val GUC Val GUA Val GUG 1 1 0 2 Ala GCU 2 3 2 1 Asp GAU Asp GAC Glu GAA Glu GAG 6 0 8 6 Gly GGU Gly GGC Gly GGA Gly GGG 1 1 0 2 Ala GCC Ala GCA Ala GCG ( 10 ) , even though the sequences of the two organisms are very homologous in these control regions .
A major difference resulting from this discrepancy is that the sequence proposed as the possible operator site discussed above would be contained within the RNA polymerase binding site for the metB gene of S. typhimurium , whereas in E. coli this sequence would be contained within the metB mRNA transcript itself , which begins 97 bases further upstream at base -8 in Fig. 2 .
It is interesting that no S1 nuclease resistant DNA was observed for the PB promoter when the mRNA for hybridization was isolated from S. typhimurium not carrying plasmid pGS107 ( Fig. 3B , lane a ) .
Similar results were reported for the E. coli metB promoter ( 10 ) .
Using the same mRNA preparation , we did observe Sl nuclease resistant DNA for both P and PJ2 ( Fig. 3A , lane a ) .
These results suggest that the metJ gene may be more efficiently transcribed than the metB gene or that metJ mRNA may have a longer half life .
Since in-vivo studies using metB-lacZ and metJ-lacZ-fusions indicate higher levels of the metB gene product are produced than the metJ gene product ( Urbanowski and Stauffer , unpublished ) , translation of metJ mRNA could be an important point of control .
However , the amino-acid codon usage given in Table 2 shows little preference for rare codon usage , although some similarities of codon usage by other weakly expressed proteins ( 18 ) do exist , e.g. , the preferential use of the glycine codon GGG and the pheynl-alanine codon UUU .
The translation initiation site and the assignment of a ribosome binding site for the metB gene was based on a comparison of the deduced amino-acid sequence shown in Fig. 2 with the published amino-acid sequence of cystathionine y-synthase purified from E. coli ( 12 ) .
The translation initiation site and the assignment of a ribosome binding site for the metJ gene was based on a comparison of the deduced amino-acid sequence shown in Fig. 2 with the first 12 N-terminal amino-acids determined for the metJ protein fused to 0-galacto-sidase .
It is possible that the protein was initiated at one of the two other in-phase methionines upstream ( positions 69-71 and 102-104 ) , and subsequently processed to our amino-acid number 2 .
However , neither of these other two methionine codons are preceded by a good ribosome binding lies right at the junction between the RNA polymerase binding sites for Pii and PB .
A repressor molecule bound at this site might prevent RNA polymerase binding at both Pii and PB .
The absence of RNA polymerase at Pii could allow RNA polymerase binding at the lower affinity site PJ2 * Indeed , bound repressor might even be necessary to facilitate polymerase 68 site .
The experimental results obtained in this study are consistent with the models previously proposed for regulation of the methionine genes ( 17 ) , wherein the product of the metJ gene functions as a trans-acting repressor .
One possible version of this model , consistent with the results presented here , is that transcription of the metB gene occurs from one promoter , PB ' and that transcription of the meWt gene occurs from two distinct promoters Pil and PJ2 ' When cells are grown in methionine-free medium , RNA polymerase can bind to and transcribe from both PB and Pil efficiently , while PJ2 is transcribed somewhat less efficiently .
In cells grown in medium containing repressing levels of methionine , however , repressor competes with RNA polymerase for binding sites in the DNA region between PB and Pil , reducing the efficiency of transcription initiation at these promoters .
In addition , the less frequent binding of RNA polymerase at P would allow more frequent binding at PJ2 ' thus stimulating increased transcription from PJ2 and maintaining repressor mRNA at relatively constant levels .
In support of this model , we have constructed a hybrid metJ-lacZ fusion protein whose synthesis is under the control of the metJ control region and have measured 0-galacto-sidase activity under various growth-conditions .
Although P-galactosidase levels in this system respond to methionine in the growth medium , the difference between repressed and derepressed states is only 1.5 to 3 fold ( manuscript in preparation ) .
We intend to quantitate more carefully the amounts of mRNA produced by P and PJ2 under various growth-conditions in order to show , first , if the regulation of enzyme activity levels seen in the ACKNOWLEDGEMENTS 1 .
Lawrence , D. A. , Smith , D. A. and Rowbury , R. J. ( 1968 ) Genetics 58 , 473-492 .
and Greene , R. C. ( 1971 ) Proc .
Urbanowski , M. L. and Stauffer , G. V 7submitted ) Gene .
Maxam , A. M. and Gilbert , W. ( 1980 ) Methods in Enzymol .
Weaver , R. F. and Weissman , C. ( 1979 ) Nucleic Acids Res .
Rosenberg , M. and Court , D. ( 1979 ) Ann .
Schmitz , A. and Galas , D. J. ( 1979 ) Nucleic Acids Res .
Shine , J. and Dalgarno , L. ( 1975 ) Nature 254 , 34-38 .
Duchange , N. , Zakin , M. M. , Ferrara , P. , Saint-Girons , T. , Park , I. , Tran , S. V. , Py , M.-C .
and Cohen , G. N. ( 1983 ) J. Biol .
Plamann , M. D. and Stauffer , G. V. ( 1983 ) Gene 22 , 9-18 .
Tran , S. V. , Schaeffer , E. , Bertrand , O. , Mariuzza , R. and Ferrara , P. ( 1983 ) J. Biol .
Sanger , F. and Coulson , A. R. ( 1978 ) FEBS Letters 87 , 107-110 .
Maniatis , T. , Fritsch , E. F. and Sambrook , J. ( 1982Y Molecular Cloning .
Cold Spring Harbor Laboratory , Cold Spring Harbor , N.Y. 15 .
Casadaban , M. J. , Chou , J. and Cohen , S. N. ( 1980 ) J. Bacteriol .
Steers , E. , Jr. , Cuatrecasas , P. and Pollard , H. B. ( 1971 ) J. Biol .
Rowbury , R. J. ( 1983 ) In : Amino Acids : Biosynthesis and Genetic Regulation , Herrman , K. M. and Somerville , R. L. ( eds .
Grosjean , H. and Fiers , W. ( 1982 ) .