Regulation of the metR Gene of Salmonella typhimurium MARK L. URBANOWSKI AND GEORGE V. STAUFFER * Department of Microbiology , University of Iowa , Iowa City , Iowa 52242 Received 11 May 1987/Accepted 14 September 1987 Regulation of the SalmoneUa typhimurium metR gene was studied by measuring I-galactosidase levels in Escherichia coli strains lysogenic for a lambda bacteriophage carrying a metR-lacZ fusion .
The results indicate that the metR gene is negatively regulated by its own gene product and that this autoregulation involves homocysteine as a corepressor .
In addition , the results indicate that the metR gene is negatively regulated by the metj gene product over a 70-to 80-fold range .
The methionine biosynthetic genes are scattered around the chromosomes of Escherichia coli and Salmonella typhi-murium and form a regulon ( 9 ; for a review , see reference 7 ) .
Expression of the regulon is controlled by noncoordinate repression of transcription by the metJ gene product and S-adenosylmethionine .
Recently , it has been shown that the metR gene product , a positive-acting protein , is required for expression of the metE and metH genes ( 11 ) .
These two genes encode the vitamin-B12-independent and vitamin-B12-dependent homocysteine transmethylases , respectively ( 7 ) .
Both of these enzymes catalyze the last step in methionine biosynthesis .
To understand the role of the metR gene product in the regulation of methionine biosynthesis , we have examined the regulation of the metR gene itself .
Since no assay for the metR gene product is available , a metR-lacZ gene fusion was constructed and used to study metR gene expression .
In this system , the synthesis of a functional chimeric , B-galactosi-dase enzyme is directed by the metR gene control region .
Construction of k lysogens carrying a metR-lacZ gene fusion .
The metR gene from S. typhimurium has been cloned ( 11 ) , and its nucleotide sequence has been determined ( 5 ) .
To construct a metR-lacZ gene fusion , plasmid pGS191 , which carries the S. typhimurium metR gene , was digested with restriction enzymes MluI and PstI ( Fig. 1 ) .
A 581-base-pair DNA fragment carrying the metR control region was isolated , the MluI site was filled in with the large fragment of DNA polymerase 1 ( 3 ) , and the DNA fragment was partially digested with the restriction enzyme Sau3AI .
A 458-base-pair DNA fragment carrying the metR promoter region and the first 39 codons of the metR gene was isolated and ligated into the lacZYA fusion vector pMC1403 ( 1 ) at the SmaI and BamHI restriction sites .
In the resulting plasmid , designated pRlac , the site of fusion of codon 39 of the metR gene to codon 8 of the lacZ gene was verified by DNA sequencing .
The metR-lacZ gene fusion and the lacY and lacA genes carried on plasmid pRlac were cloned into the single EcoRI site in bacteriophage Xgt2 by the method described previously ( 10 ) .
The resultant phage , designated XRlac , was plaque purified and was used to lysogenize the E. coli strains shown in Table 1 .
All lysogens were tested to verify that they were single lysogens ( 8 ) .
The metR gene is negatively regulated by the metj repressor .
The metR gene lies near the metE gene in both S. typhimu-rium and E. coli ( 11 ) .
In the former , it has been shown that the two genes are divergently transcribed and that the two promoter elements overlap substantially ( Fig. 1 [ 5 ] ) .
Since expression of the metE gene is subject both to negative control via the metJ repressor and to positive control via the metR gene product , it seems likely that expression of the overlapping metR promoter would also be influenced by these two regulatory gene products .
To test whether the metR gene was subject to metJ-mediated repression , ARlac lysogens were grown in glucose minimal-medium plus phe-nylalanine and vitamin-Bi ( GMM ) supplemented with either L-methionine ( repressing condition ) or D-methionine ( nonrepressing condition ) , and,-galactosidase levels were assayed ( 4 ) as an indication of metR gene expression .
In the Met ' parent lysogen 162XRlac , addition of L-methionine to the growth medium caused a two-to threefold repression of P-galactosidase activity ( Table 2 ) .
In contrast , the metJ lysogen 597XRlac showed elevated P-galactosidase activity that was not repressed by the addition of L-methionine to the growth medium .
These results indicate that the metR gene is subject to metJ-mediated repression over about a 70-to 80-fold range .
The meiR gene is negatively autoregulated .
Many positiveacting regulatory genes in E. coli employ autoregulation as a mechanism to maintain relatively constant levels of gene product in the cell ( 6 ) .
Thus , the metR-lacZ fusion phage was used to determine whether the metR gene product regulates its own synthesis .
In the Met-metR lysogen 244ARlac , 3-galactosidase activity was more than fourfold higher than in the Met-metE control lysogen 243XRlac when the cells were grown in GMM plus D-methionine ( Table 2 ) , suggesting that the metR gene is involved in negatively regulating its own expression .
Although the control lysogen 243ARlac and the metR lysogen 244XRlac are both functionally metE mutants ( 11 ) , their metE alleles are different .
To rule out the possibility that the differences seen are due to heterogenous metE alleles , we tested lysogens-705XRlac ( metEJ63 : : TnlO ) and 750XRlac ( metEJ63 : : TnJO metR ) under the same conditions and found a similar four-to fivefold negative autoregulation ( Table 2 ) .
Interestingly , the addition of L-methionine to the growth medium caused repression of P-galactosidase synthesis in both the 244ARlac and 750ARlac lysogens , indicating that negative control by the metJ repressor can override derepression caused by the metR mutation .
Since the metJ repressor negatively regulates metR gene expression , the apparent negative regulation of metR by its own gene product could actually be an indirect result of positive regulation of the metJ gene by the metR gene product .
This hypothesis was tested in two ways .
First , we compared the P-galactosidase levels in the metJ lysogen 597XRlac and the metR lysogen 244XRlac with those in the metJ metR lysogen 748XRlac ( Table 2 ) .
In lysogen 748XRlac , P-galactosidase levels were two-to threefold higher than in lysogen 597XRlac , indicating that a metR mutation could cause derepression of the metR gene beyond that caused by inactivation of the metJ system .
Conversely , in lysogen 244XRlac , where no metR gene product is produced , there was enough metJ repressor present to repress,3-galactosi-dase synthesis when methionine was added to the growth medium ( Table 2 ) .
Second , we determined the direct effect of the metR mutation on metJ gene expression by comparing 584 35 -10 -10 -35 FIG. 1 .
Construction of the metR-lacZ gene fusion .
The fusion consists of a 458-base-pair MluI-Sau3AI DNA fragment containing the metE-metR control region of S. typhimurium from plasmid pGS191 ( 11 ) fused at codon 39 of the metR gene to codon 8 of the E. coli lacZ gene .
Synthesis of the chimeric , B-galactosidase product was dependent on transcription and translation initiation from the metR control region .
The-10 and -35 regions of the rightward-transcribed metR promoter and leftward-transcribed metE promoter overlap substantially .
Heavy lines indicate plasmid pBR322 vector DNA .
Open boxes indicate structural gene segments ; the primed gene symbol denotes a truncated gene .
Heavy arrows indicate the directions of transcription of the metE and metR genes .
Abbreviations : B , BamHI ; H , Hindlll ; M , MluI ; P , PstI ; S , Sau3AI ; Sa , Sail ; Sm , SmaI .
E. coli strain descriptions and origins Relevant markersa Source AlacUI69 G. Zurawski AmetE : : Mu AlacU169 This laboratory AmetR : : Mu AlacU169 This laboratory metJ97 AlacU169 This laboratory metEJ63 : : TnlO AlacU169 This laboratory metBI metJ97 AlacU169 This laboratory metBI metC162 : : TnlO metJ97 This laboratory AlacUl69 metBI metC162 : : TnlO metJ97 This laboratory AmetF : : Mu AlacUl69 metBI metE163 : : TnIO metJ97 This laboratory AlacU169 metJ97 AmetR : : Mu AlacU169 This laboratory metE163 : : TnJO AmetR : : Mu This laboratory metBI metC162 : : TnlO metJ97 This laboratory AmetF : : Mu metR AlacU169 TABLE 2 .
Effects of the metJ and metR gene products on expression of the S. typhimurium metR-lacZ gene fusion Strain 3-Galactosidase activitya GS162 GS243 LysL e o ~ - M h ogynsogenD-Methi .
on.ine ~ ~ ~ ~ BBD +1 etMionine GS244 GS597 GS705 GS719 GS720 162XRlac ( Met + ) 597XRlac ( met !
) 50 1,680 80 20 2,630 90 320 3,820 80 260 1,410 20 30 4,100 20 30 243XRlac ( metE ) 244ARlac ( metR ) 748XRlac ( metJ metR ) 705XRlac ( metE ) 750XRlac ( metE metR ) 243AJlac ( metE ) 244XJlac ( metR ) 100 440 3,850 70 340 GS723 GS747 170 160 90 90 NDb ND GS748 GS750 GS753 a In addition to the relevant markers , all strains carry the pheA905 , araD129 , rpsL , and thi mutations .
P-galactosidase levels in lysogens carrying the metJ-lacZ fusion phage XJlac ( 10 ) .
The metR lysogen 244AJ1ac showed no significant difference from the control metE lysogen 243AJlac when the cells were grown in GMM supplemented with either D-methionine or L-methionine ( Table 2 ) .
Together , these results indicate that the metR protein is not involved in the regulation of metJ gene expression and that the metR gene is negatively regulated directly by its own gene product and by the metJ repressor .
Involvement of homocysteine in metR autoregulation .
The metR-mediated activation of the divergently transcribed metE gene involves the methionine intermediate homocys-a Units of specific activity are nanomoles of O-nitrophenol produced per minute per milligram of protein at 28 ` C .
The growth medium was GMM , supplemented as indicated with either D-methionine ( 150 , ug/ml ) , L-methionine ( 50 pLg/ml ) , or D-methionine plus vitamin-B1 ,2 ( 1 g/ml ) .
b ND , Not done teine as a coactivator ; O-succinylhomoserine , cystathionine , 50 1,680 80 20 2,630 90 320 3,820 80 260 1,410 20 30 4,100 20 30 243XRlac ( metE ) 244ARlac ( metR ) 748XRlac ( metJ metR ) 705XRlac ( metE ) 750XRlac ( metE metR ) 243AJlac ( metE ) 244XJlac ( metR ) 100 440 3,850 70 340 GS723 170 160 90 90 NDb ND 5-methyltetrahydrofolate ( 5-mTHF ) , and methionine have no effect ( M. L. Urbanowski and G. V. Stauffer , manuscript in preparation ) .
We therefore tested the effect of homocysteine on the expression of the metR-1acZ gene fusion .
The metJ metB metC lysogen 720XRlac and the metJ metB metC metF lysogen 723XRlac are defective in synthesis of homocysteine , although both strains , when supplemented with limiting amounts of methionine ( D-methionine ) , produce small amounts of homocysteine from end , ogenous S-adenosylmethionine via a regenerative pathway ( 2 ) .
Lysogen 720XRlac had high P-galactosidase activity when grown in GMM supplemented only with D-methionine ( Table 3 ) .
However , when homocysteine was added to the growth medium , P-galactosidase activity was reduced nearly 10-fold .
A similar reduiction was seen in the metJ lysogen 597ARlac , although the effect was less dramatic because of endogenous homocysteine production .
This reduction was not seen in the metJ metR lysogen 748XRlac , suggesting that homocysteine acts as a corepressor in metR autoregulation ; In contrast , the metF mutation ( which blocks 5-mTHF synthesis ) in lysogen 723 .
Rlac prevented the high ' - galactosidase expression seen in 720XRlac , which suggests two possibilities : ( i ) ` either the metF gene product or 5-mTHIF is directly required for metR gene expression , or ( ii ) since homocysteine facilitates repression of metR , 5-mTHF increases metR expression indirectly by allowing utilization of homocysteine via the homocysteine transmethylase reaction , thus preventing accumulation of homocysteine formed through the regenerative pathway discussed above .
To distinguish between these two possibilities , we tested metRlacZ ` expression in the metJ metB lysogen 719XRlac and in the metJ metB metE lysogen 747ARlac .
In 747XRlac the metE mutation prevents utilization of homocysteine and 5-mTHF and thus leads to an accumulation of both intermediates .
If 5-mTHF is required directly for expression of the metR-lacZ fusion , then P-galactosidase levels should be intermediate to high in 747XRlac , similar to the levels in 720XRlac .
Conversely , if expression responds only to homocysteine , then,-galactosidase levels should be low , similar to the levels in 723ARlac .
As shown in Table 3 , lysogen 747XRlac had low P-galactosidase levels , suggesting that the low levels seen in 723XRlac were a result of homocysteine repression due to an accumulation of this TABLE 3 .
Effects of homocysteine and the metF gene product on expression of the metR-lacZ gene fusion P-Galactosidase activitya D-MethiOnine + D-Methionine hhoommooccyysstt , neiinnee 4,770 500-290-210 1,750 250 4,350 3,850 3,070 430-330-260 4,500 3,890 Lysogen Relevant genotype 720XRlac metJ metB , metC 723XRlac metJ metB metC metF 597XRlac metJ 748ARlac metJ metR 719ARlac metJ metB 747XRlac metJ metB metE 753XRlac metJ metB metC metF metR 753ARlac metJ metB metC metF ( pGSmetR ) metRlmetR + a Units of specific activity are nanomoles of O-nitrophenol produced per minute per milligram of protein at 28 °C .
The growth medium was GMM , supplemented where indicated with D-methionine ( 150 , ug/ml ) or D-methionine plus DL-homocysteine ( 100 , ug/ml ) .
intermediate and that 5-mTHF acted indirectly by affecting the homocysteine pools .
The simplest model for homocysteine repression of metR is one in which this intermediate acts as a corepressor with the metR protein rather than acting by a metR-independent mechanism .
If this model is correct , then it should be possible to isolate metR mutants of 723XRlac on lactose minimal-medium showing high derepressed P-galactosidase levels .
To test this hypothesis , portions ( 0.05 ml at 2 x 109 cells per ml ) of four independent overnight cultures of lysogen 723ARlac were plated onto lactose minimal plates supplemented with phenylalanine , vitamin-B1 , D-methi-onine , and 1 mM phenylethyl -,3-D-thiogalactoside ( a lactose analog that reduces background growth of lysogen 723XRlac ) .
Lysogen 723XRlac grew very slowly on this medium .
From each selection plate , one Lac ' colony which arose after 48 h was purified and the P-galactosidase levels were measured .
All four independently isolated mutants showed similar derepressed,-galactosidase activity .
' The enzyme levels of one representative mutant , 753ARlac , are shown in Table 3 .
The new mutation in 753ARlac , and in each of the other three mutants , was shown tp lie in the meR gene by the following three criteria .
( i ) All four mutants , when transformed with a single-copy-number plasmid carrying only the metR gene ( plasmid pGSmetR ) , showed low,-galactosidase levels , indicating complementation of the mutations in trans by the mettR plasmid [ only 753XRlac - ( pGSmetR ) is shown in Table 3 ] .
( ii ) All four mutations were linked to the ilv locus by phage P1 transduction ( data not shown ) , consistent with the map position of metR ( 11 ) .
( iii ) New ilv + metR transductants isolated in the above P1 experiments , when lysogenized with a XElac phage carrying a metE-lacZ fusion , show greatly reduced levels of metE-lacZ expression ( data not shown ) , typical of a metR mutant ( 11 ) .
Involvement of vitarj B12 in metR regulation .
We tested whether vitamin ` B12 in enced expression of the metR gene .
In both the wild-type lysogen 162XRlac and the metJ lysogen 597XRlac , a small increase in P-galactosidase levels was seen when vitamin-B12 was added to the growth medium ( Table 2 ) .
It is not clear whether metR gene expression is directly enhanced by the metH holoenzyme or whether this enhanced expression is an indirect result of a turning off of the overlapping metE promoter by the metH holoenzyme .
This work was supported by Public Health Service grant GM-26878 from the National Institute of General Medical Sciences .
Casadaban , M. J. , J. Chou , and S. N. Cohen .
In vitro gene fusions that join an enzymatically active 13-galactosidase segment to amino-terminal fragments of exogenous proteins : Esch-erichia coli plasmid vectors for the detection and cloning of translational initiation signals .
Duerre , J. A. , and R. D. Walker .
Metabolism of adenosylhomocysteine , p. 43-57 .
In F. Salvatore , E. Borek , V. Zappia , H. G. Williams-Ashman , and F. Schlenk ( ed .
) , The biochemistry of adenosylmethionine .
Columbia University Press , New York .
Maniatis , T. , E. F. Fritsch , and J. Sambrook .
Molecular cloning : a laboratory manual .
Cold Spring Harbor Laboratory , Cold Spring Harbor , N.Y. 4 .
Experiments in molecular genetics , p. 352-355 .
Cold Spring Harbor Laboratory , Cold Spring Harbor , N.Y. 5 .
Plamann , L. S. , and G. V. Stauffer .
Nucleotide sequence of the Salmonella typhimurium metR gene and the metR-metE control region .
Raibaud , O. , and M. Schwartz .
Positive control of transcription initiation in bacteria .
Methionine biosynthesis and its regulation , p. 191-211 .
In K. M. fierrmann and R. L. Somerville ( ed .
) , Amino acids : biosynthesis and genetic regulation .
AddisonWesley Publishing Co. , Reading , Mass. 8 .
Shimadaj K. , R. A. Weisberg , and M. E. Gottesman .
Prophage X at unusual chromosomal locations .
I. Location of the secondary attachment sites and properties of the lysogens .
Shoeman , R. , B. Redfield , T. Coleman , N. Brot , H. Weissbach , R. C. Greene , A. A. Smith , I. Sainit-Girons , M. M. Zakin , and G. N. Cohen .
Regulation of the methionine regulon in Escherichia coli .
Urbanowski , M. L. , and G. V. Stauffer .
Autoregulation by tandem promoters of the Salmonella typhimurium LT2 metJ gene .
Urbanowski , M. L. , L. T. Stauffer , L. S. Plamann , and G. V. Stauffer .
A new methionine locus , metR , that encodes a trans-acting protein required for activation of metE and metH in Escherichia coli and Salmonella typhimurium .