Regulation of Salmonella typhimurium ilvYC Genes D. L. BLAZEYt * AND R. 0 .
BURNSt Department of Microbiology and Immunology , Duke University School of Medicine , Durham , North Carolina 27710 Received 15 March 1984/Accepted 5 June 1984 The Salmonella typhimurium LT2 ilvYC genes were studied by fusion of each gene to the Escherichia coli K-12 galK gene .
The expression of ilvY and ilvC could then be determined by measurement of the galK-encoded galactokinase enzyme .
The promoter for ilvC , Pc , was located by this technique to a 0.42-kilobase BglII-EcoRI fragment of the S. typhimurium ilvGEDAYC gene cluster .
This sequence was completely sufficient for aL-acetohydroxyacid-inducible galK expression .
The ilvY gene was located within a 1.0-kilobase XhoI-SalI fragment .
ilvY gene expression was constitutive with respect to ilv-specific control signals .
The ilvY gene was transcribed in the same direction as the other two transcriptional units in the ilvGEDAYC gene cluster , ilvGEDA and ilvC .
Transcription of the ilvC gene was completely dependent upon the activity of its own promoter , Pc , and independent from transcription of the ilvY gene .
The role of the intervening region between ilvY and ilvC in regulation of ilvC expression was explored .
The Salmonella typhimurium LT2 ilvGEDA YC gene cluster specifies the proteins necessary for the biosynthesis of isoleucine and valine ( 6 , 21 ) .
Four of the biosynthetic enzymes are encoded by the ilvGEDA genes , which constitute a single operon ( 4 ) .
Expression of ilvGEDA is negatively regulated by isoleucine , valine , and leucine ( 8 ) .
The control mechanism involves translational control of transcription termination , i.e. , attenuation ( 20 ) .
The ilvC gene encodes the remaining enzyme of the isoleucine-valine biosynthetic pathway , a-acetohydroxy-acid isomeroreductase ( 2 , 11 ) .
The ilvC gene constitutes a separate transcriptional unit from ilvGEDA ( 4 ) , the expression of which is induced by the presence of the substrates of the isomeroreductase , a-acetolactate and a-acetohydroxybutyrate ( 1 , 2 ) .
Action of these inducers , as in Escherichia coli K-12 ( 22 ) , requires the participation of a regulatory protein encoded by the ilv Y gene .
The S. typhimurium ilvY gene was mapped between ilvA and ilvC by analysis of cloned fragments of the gene cluster ( 6 ) .
The ilvY gene product has an essential role in expression of ilvC as demonstrated by the isolation of an ilv Y mutant that displays a trans-recessive IlvC-phenotype ( 5 ) .
The ilv YC genes constitute a distinctive regulatory system that employs positive control , which is uncommon for biosynthetic genes ( 21 ) and sharply contrasts with the control of the ilvGEDA operon .
In addition , the apparently contiguous relationship between the regulatory gene ilv Y and its target gene ilvC is unusual , and this propinquity may allow for more complex interactions between ilv Y and ilvC .
This work presents further characterization of the S. typhi-murium ilvYC genetic system .
MATERIALS AND METHODS Bacterial strains and plasmids .
The bacterial strains and plasmids used are shown in Tables 1 and 2 , respectively .
The primary S. typhimurium ilv + recombinant plasmid was pDU1 ( 6 ) ; the vector was a derivative of RSF2124 , and the pertinent genotype of the chromosomal insert was ilvGEDA Y+C ' .
pDU5 is a derivative of pDU1 that contains a 6.0-kilobase ( kb ) Sall ilvEDA Y + fragment inserted in t Present address : Department of Medicine , University Hospital , University of Washington , Seattle , WA 98105 .
The pKO galK vectors pKO5 and pKO6 are derivatives of pKO1 ( 19 ) and contain different arrays of cloning sites proximal to galK .
The galK gene is expressed only when a properly oriented promoter is inserted within one of these sites .
The liquid medium contained Davis-Mingioli salts ( 7 ) modified by the omission of citrate and supplemented with 0.5 % glucose .
Repressing medium contained 100 mg of L-valine , 100 mg of L-leucine , and 50 mg of L-isoleucine per liter .
Limitation for isoleucine or limitation for leucine was achieved by decreasing the concentration of that amino-acid to 5 mg/liter .
Limitation for valine was effected by substituting 25 mg of glycyl-L-valine for valine .
Additional amino-acid requirements were met with 50 mg of the appropriate amino-acid per liter except for tryptophan , which was used at 10 mg/liter .
The thiamine concentration was 10 mg/liter .
All plasmid-containing organisms were grown in media containing 40 mg of ampicillin per liter .
Transformants were selected on nutrient agar ( Difco Laboratories ) or MacConkey agar ( Difco ) containing 0.5 % D-galactose ; both media contained 40 mg of ampicillin per liter .
All restriction endonucleases were purchased from either New England Biolabs or Bethesda Research Laboratories .
T4 DNA ligase was from Bethesda Research Laboratories .
BAL-31 nuclease was prepared from Alt-eromonas espejiani BAL-31 as described previously ( 9 ) .
Plasmid DNAs were prepared from chlo-ramphenicol-amplified cultures as described previously ( 6 ) .
The rapid screening of transformants was performed as described by Birnboim and Doly ( 3 ) .
Construction and characterization of recombinant plasmids .
All restriction endonuclease digestions were performed under the conditions suggested by Maniatis et al. ( 17 ) .
Subcloning experiments were performed with a vector/fragment ratio of 1:1 .
T4 DNA ligation reactions were performed as described by Weiss et al. ( 23 ) at a DNA concentration of 25 , ug/ml .
The ligase concentration was 1 to 2 U/ml for staggered-end DNA and 100 U/ml for blunt-end DNA .
BAL-31 nuclease digestions were generally performed under the low-salt conditions described by Legerski ( 15 ) .
Digestion of X DNA was performed to quantify the BAL-31 nuclease activity and to determine appropriate digestion rates as suggested previously ( 9 ) .
After BAL-31 nuclease digestion , the DNAs were extracted once with phenol-chloroform ( 1:1 [ vol/vol ] ) and twice with watersaturated diethyl ether .
Residual ether was boiled away at 50 °C , and the DNAs were ethanol precipitated , collected by centrifugation , and suspended in an appropriate volume of T4 ligase buffer .
Transformation was performed as described by Lederberg and Cohen ( 14 ) .
Electrophoresis was generally performed in a horizontal apparatus with 1.0 % agarose gels and with the E buffer described by Loening ( 16 ) .
Electropho-resis of small fragments for molecular weight determinations was performed in either 1.4 % agarose gels or 5 % polyacryl-amide gels with a Tris-borate buffer ( 10 ) .
Molecular weight standards were either X phage or pBR322 restriction fragments .
Preparation of cells for assay .
All cells were grown at 37 °C in the defined media described above .
When isomeroreductase activity was measured , the cells were grown as 125-ml batch cultures , and most of the culture was used to prepare a cell extract ( 4 ) .
A 2-ml sample of each culture was processed for galactokinase assay ( 19 ) .
In all cases , the cells were first grown in repressing medium and then inoculated into various media to effect limitation for each branchedchain amino-acid .
The cells were generally harvested after about 8 h of growth .
Limitation for the specific amino-acid was manifest when the cells reached a certain maximal optical density of less than that seen with fully supplemented medium and was confirmed by measurement of derepressed levels of ilv enzymes ( 4 ) .
Galactokinase was measured by the method of McKenny et al. ( 19 ) , and the units were expressed as nanomoles of galactose phosphorylated per minute per unit of absorbance at 650 nm .
The a-acetohydroxyacid isomeroreductase was measured by the method of Arfin et al. ( 1 ) with oa-acetohydroxybutyrate as the substrate .
The isomeroreductase units were nanomoles of NADPH reduced per minute per milligram of protein .
ot-Acetolactate and a-acetohydroxybutyrate were prepared by saponification of the corresponding methyl-esters , which were the gift of Frank Armstrong .
D - [ 14C ] galactose was purchased from Amersham Corp. ( product no .
Strains Genotype Strain Source ( reference ) E. coli K-12 DU650 F-A ( ilvEDA YC ) leuB5 hsdR hsdM + AtrpE5 F-galK leu pro F-galK recA pro Laboratory collection McKenney et al. ( 19 ) McKenney et al. ( 19 ) C600 N100 S. typhimurium LT2 DU2 F-hsdR hsdM ilvDl8 ilvA228 leuA409 F-hsdR hsdM ilvA ilvY702 leuA409 recAl F-ilv + rho-Ill metE338 hisC2124 trpE49 Laboratory collection ( 5 ) Laboratory collection ( 5 , 6 ) Housley et al. ( 12 ) DU703 AA111 RESULTS ilvC-directed galK expression .
We initially cloned an EcoRI fragment from the S. typhimurium LT2 chromosome that was ilvGEDA Y + but lacked an intact gene ilvC as shown by complementation analysis and enzyme assay ( 6 ) .
An S. typhimurium BamHI-EcoRI ilvDA Y+C ' subclone is shown in Fig. 1 .
Approximately 0.9 kb of DNA contained in this fragment is distal to the ilvY gene , and , because of the contiguous location of ilvC to ilvY , we concluded that some of the ilvC gene must be present between the Sall and EcoRI sites .
If the S. typhimurium ilvC gene is transcribed in the same direction as that of ilvGEDA ( that is , from ilvG to ilvA ) , then the ilvC promoter probably resides within the Sall ¬ TABLE 2 .
Plasmids Plasmid pDU1 Pertinent genotype ilvGEDA Y+C ' rrnC ' Source ( reference ) Original S. typhimurium ilv , clone ( 6 ) pDU5 ilvEDA Y + Sall fragment of pDU1 inserted into pBR322 ( 6 ) pKO vector ( 19 ) pKO6 ( BamHI-EcoRI ) galK ilvDA Y+C ' gaIK + pDU60 pDU61 BamHI-EcoRI fragment of pDU1 inserted into pKO6 Deletion of internal Xhol-SalI fragment from pDU60 ilvDA + Ailv Y ivC ' galK + ilvD + AilvA Y ilvC ' - galK + pDU62 pDU63 Deletion of internal BgllI fragment from pDU60 Deletion of BamHI-BgIII and BglII fragments from pDU60 AilvDA Y ilvC ' - galK + ilvDA + AilvY ilvC ' - galK + pDU60 .43 pDU64 BAL-31 nuclease-mediated deletion of ca. 0.4 kb of the N-terminal region of ilvY Inversion of the internal XhoI-SalI fragment of pDU60 ilvDA Y + ilvC-galK + ( BamHI ) galK pKO5 pKO vector ( 19 ) pDU71 ilvA ' ilvY + galK ilh ` Y + ilvA ' galK BgIll fragment from pDU60 inserted into B BgIII fragment from pDU60 inserted into B orientation from pDU71 amHI site of pKO5 pDU72 amHI site of pKO5 but in opposite pDU71 .1 ilvA ' ilvY ' - galK + BAL-31 nuclease-mediated fusion of ilvY to gal EcoRI region .
This was demonstrated by insertion of the BamHI-EcoRI fragment into the galK vector pKO6 ( Fig. 1 ) .
Expression of galK from the resulting plasmid , pDU60 , was studied under conditions known to modulate transcription of the ilvC gene ( Table 3 ) .
In strain C600 ( pDU60 ) , galK was induced by the acetohydroxyacids a-acetohydroxybutyrate and a-acetolactate .
The less-efficient induction of ilvC by a-acetolactate has been noted previously for an E. coli K-12 in-vitro-transcription system ( 24 ) .
Since the products of the acetohydroxyacid synthases are inducers of ilvC expression , aThe modified Davis-Mingioli medium described in the text was used .
Repressing medium was supplemented with either 50 mg of L-proline per liter ( for strain C600 ) or 10 mg of L-tryptophan per liter ( for strain DU650 ) .
The concentration of a-acetohydroxybutyrate ( AHB ) or a-acetolactate ( AL ) was 2 mM .
b Nanomoles of galactose phosphorylated per minute per unit of ansorbance at 650 nm .
The galK assay was as described by McKenny et al. ( 19 ) .
c Low-level galK expression from pKO6 results from plasmid promoters .
d Expression of the chromosomal galK gene in strain DU650 can not be detected under these conditions .
a Modified Davis-Mingioli medium described in the text was used .
Repressing medium was supplemented with either 50 mg of L-proline per liter ( for strain C600 ) or 10 mg of L-tryptophan per liter ( for strain DU650 ) .
The concentration of a-acetohydroxybutyrate ( AHB ) was 2 mM .
b See Table 3 , footnote b. derepression of the synthases by valine limitation ( ilvB ) or leucine limitation ( ilvB and ilvI ) is correlated with an endo-genous induction of ilvC .
Accordingly , pDU60 galK expression was also induced under these conditions as shown by assay of strains DU650 A ( ilvEDAYC ) ( pDU60 ) and C600 ilv + ( pDU60 ) .
These results demonstrate that the pDU60 galK gene is expressed from the ilvC promoter ( Pc ) and that the transcription of ilvC is in the same direction as that of the ilvGEDA operon .
Furthermore , the induction of galK in the DU650 A ( ilvEDAYC ) background confirms that the BamHI-EcoRI fragment contains the intact ilv Y gene .
A more precise location for Pc and its relationship to the contiguous ilvY gene was determined by deleting specific restriction endonuclease fragments from pDU60 .
These de-letion derivatives were compared with pDU60 by agarose gel electrophoresis to confirm their compositions .
Expression of galK from each plasmid was measured in strains C600 and DU650 , and the responsiveness of galK to induction by cxacetohydroxybutyrate was determined ( Table 4 ) .
Previous analysis demonstrated that the ilvY gene is located between XhoI and Sall ( 6 ) .
When the XhoI-SalI fragment was deleted from pDU60 , the resulting plasmid pDU61 conferred inducible galK expression in strain C600 ilv + and constitutive expression in strain DU650 A ( ilvEDAYC ) , in which ilvY was absent from both the plasmid and chromosome .
This result confirms the location of ilv Y within the XhoI-SalI fragment and indicates that Pc is entirely contained within the 0.88-kb SalI-EcoI region .
De-letion of the internal BgIII fragment or both the BamHI-BglII and BglII fragments from pDU60 ( pDU62 and pDU63 , respectively ) also resulted in derivatives which expressed inducible galK activity in strain C600 ilv + and constitutive activity in strain DU650 A ( ilvEDA YC ) .
These results dem-onstrate that the Pc sequence must be contained within the 0.42 kb BglII-EcoRI region of the ilv insert .
Furthermore this region includes the cis-active regulatory information through which the acetohydroxyacids and the ilvY gene product effect induction of the ilvC gene .
The intact ilv Y gene is contained within the 1.0-kb XhoI-SalI segment of the S. typhimurium ilv cluster ( Fig. 1 ) .
This fragment is only slightly larger than the predicted ilvY structural gene sequence ( 0.95 kb ) based upon the 35,000-dalton E. coli K-12 ilvY gene product ( 22 ) .
Therefore , limited digestion with exonuclease BAL-31 at either end of the XhoI-SalI segment should result in de-letions extending into the ilvY transcriptional unit .
If this digestion were done so as to fuse the BAL-31 nuclease-digested ilvY gene to galK , then ilvY-directed galK expression should occur with fusions at either the XhoI or SalI termini of ilvY , depending upon the direction of transcription of ilv Y .
The BAL-31 nuclease-mediated fusion of galK to the ilv Y transcriptional unit also affords the opportunity to study the regulation of ilv Y expression by measuring ilv Y-directed galK activity under various conditions .
The 2.4-kb ilvY + fragment generated by BglII endonuclease digestion of pDU60 ( Fig. 1 ) was inserted in both orientations into the BamHI site of the galK vector pKO5 ( Fig. 2 ) .
The resulting plasmids , pDU71 and pDU72 , showed no galK expression .
The plasmids were cleaved with either Sall ( pDU71 ) or XhoI ( pDU72 ) endonucleases and then subjected to limited digestion with BAL-31 nuclease .
The nucleasetreated plasmids were restored to closed circular form by blunt-end ligation and used to transform strain N100 to Ampr on MacConkey-galactose agar .
Ampr transformants that were red on the indicator medium , indicating expression of galK , were obtained only with the pDU71 DNA preparation ( 25 red Ampr transformants per 120 Ampr transformants ) .
Analysis of the partially purified plasmids from the red transformants demonstrated that each plasmid contained a deletion extending into ilvY but not as far as the XhoI site .
Enzyme assay confirmed that the deletion plasmids expressed the galK gene ; the galactokinase activity of these plasmids ranged from 150 to 400 U. On the other hand , all 100 Ampr transformants obtained with the pDU72 DNA preparation were white .
Analysis of 20 of these transform-ants indicated that many contained plasmids with deletions extending into ilvY without affecting the distal Sall site ; these deletions also did not extend into the plasmid-encoded galK gene .
Enzyme assay of these transformants showed no galK expression .
The ability to promote galK expression by fusion of galK to ilv Y , then , can occur by BAL-31 digestion only at the Sall terminus of the ilv Y gene .
The ilv Y promoter must be located near the XhoI site , so that transcription of ilv Y proceeds in the direction from XhoI to SalI .
Therefore , the ilv Ygene is transcribed in the same direction as the other TABLE 4 .
Expression of galK from deletion derivatives of pDU60 Galacto-kinase sp act ( gaIK ) b 250 Strain Medium ' C600 ilv + ( pDU61 ilvDA + AilvY ilvC ' - galK + ) Repressing Repressing + AHB 630 DU650 A ( ilvEDA YC ) ( pDU61 ) Repressing 200 Repressing + AHB 220 / C600 ilv + ( pDU62 ilvD + A [ ilvA Y ] ilvC ' - galK + ) Repressing 220 Repressing + AHB 580 i/vA ilyr \ i/vC a ( ptlIJ ( hoI So/I I Bg/IE KfpnI Xhd Sol/I Bggll EcoRI DU650 A ( ilvEDA YC ) ( pDU62 ) Repressing 250 Repressing + AHB 240 / / / nd Hl/id II 8omHI C600 ilv + ( pDU63 A [ ilvDA Y ] ilvC ' - Repressing 300 galK + ) Repressing + AHB 600 0 1 3 2 4 5 5.6 kb I DU650 A ( ilvEDA YC ) ( pDU63 ) Repressing 380 Repressing + AHB 280 FIG. 1 .
pDU60 ilvDA Y + ilvC ' - galK .
The vector is pKO6 .
Expression of galK from the S. typhimurium ilvC promoter Galacto-kinase sp act ( galK ) b 10 10 8 Strain Medium ' C600 ilv + ( pKO6 gaIK ) c Repressing Repressing + AL Repressing + AHB C600 ilv + ( pDU60 ilvDA Y + ilvCgalK + ) Repressing 80 Repressing + AL 400 Repressing + AHB 710 DU650 A ( ilvEDA YC ) ( pDU60 ) d Repressing 43 Limiting valine 660 Repressing + AHB 280 a Media are described in the text .
The media were supplemented with either 10 mg of L-tryptophan per liter ( for strain DU650 ) or 50 mg of L-proline per liter ( for strain C600 ) .
The concentration of a-acetolactate ( AL ) was 5 mM , and that of a-acetohydroxybutyrate ( AHB ) was 2 mM .
b See Table 3 , footnote b. two transcriptional units contained in the S. typhimurium ilv cluster , namely , ilvGEDA and ilvC .
The relationship between the ilv Y promoter and the XhoI site was more closely examined by cleaving two distinct pDU71-derived ilvY-galK fusion plasmids with XhoI endonuclease and then introducing small deletions about the XhoI site with BAL-31 nuclease .
BAL-31-degenerated de-letions as small as 50 base pairs destroyed ilvY-mediated galK expression ; galactokinase activity encoded by these plasmids was less than 10 U , which is the background level observed with the pKO5 vector .
Therefore , the maximum distance between the XhoI site and the ilvY promoter is 50 base pairs .
We have recently exploited the proximity of the XhoI site to the ilvY promoter and N-terminal coding sequence by fusing the X bacteriophage leftward promoter ( PL ) to ilv Y at an SstI site that overlaps with XhoI .
The ilv Y gene product is then under regulation by the temperature-sensi-tive cI857 lambda repressor .
Heat induction of the PL-ilVY plasmid results in overproduction of the gene product ilv Y to ca. 5 % of the total cellular protein .
We have used the PL-ilV Y system to determine that the subunit molecular weight of the S. typhimurium ilvY gene product is 34,000 ( unpublished data ) .
One representative ilvY-galK fusion plasmid , pDU71 .1 , was selected for preliminary studies on the regulation of ilv Y expression .
This fusion lacks about 0.5 kb of the ilv Ygene as determined by agarose gel electrophoresis ( data not shown ) .
Strain DU650 ( pDU71 .1 ) was assayed under conditions of excess branched-chain amino-acids or under limitation for isoleucine , valine , or leucine ; the ilvY-directed galK expression was not multivalently regulated by the branched-chain amino-acids ( Table 5 ) .
In addition , the pDU71.1-determined galK expression was not affected by the presence of a-acetohydroxybutyrate or acetolactate in either the DU650 A ( iIvEDA YC ) or C600 ilv + backgrounds , thus providing no evidence for autogenous regulation via the ilv Y gene product and the acetohydroxyacids .
The ilvY promoter , therefore , is constitutive with regard to ilv-specific regulation .
The ilvY promoter appears to be of low to intermediate strength when compared with the galK-determined expression of other procaryotic promoters in the pKO system ( 19 ) .
Is the ilvC gene transcribed from the ilvY promoter ?
Recent work indicates that the E. coli K-12 ilvY gene is transcribed in the same direction as ilvC ( J. Falk and H. E. Umbarger TA BL E 5 .
ilvY-directed galK expression from pDU71 .1 Galacto-kinase sp act ( gaIK ) b Repressing 360 XhoI Se/I wI loI 5g / BPI Strain Mediuma pDU7l pDUI2 DU650 A ( ilvEDA YC ) ( pDU71 .1 ilvA ' Y ' galK + ) Limiting isoleucine Limiting valine Limiting leucine Repressing + AL Repressing + AHB 280-320-270-300 320 FIG. 2 .
ilvY-containing subclones pDU71 and pDU72 .
The 2.4-kb BglII fragment of pDU60 was inserted in either direction into the BamHI site of pKO5 .
The orientations were confirmed by digesting the plasmids with AvaI endonuclease , which generates two distinct restriction profiles .
C600 ilv + ( pDU71 .1 ) Repressing 310 Repressing + AL 270 DU650 A ( ilvEDA YC ) ( pDU64 Repressing 200 ilvDA Y ilvC-gaIK ) Repressing + AHB 800 a Modified Davis-Mingioli medium was used , and the conditions were as described previously ( 4 ) .
The concentration of a-acetolactate ( AL ) or a-acetohydroxybutyrate ( AHB ) was 2 mM .
b See Table 3 , footnote b. c The ilvY gene of pDU64 is opposite in orientation from the native ilvY gene .
The insertion of transposon TnJO or phage Mu into the chromosomal ilvY gene prevented trans complementation by a plasmid-borne , intact ilvY gene .
It was concluded that the ilvC gene is dependent upon transcription from the contiguous ilv Y promoter and that the ilvY gene acts as an antiterminator at a site between ilv Y and ilvC to allow transcription to proceed into the ilvC gene .
The cis-dominant effect of ilvY : : TnJO or ilvY : : Mu insertion mutations on ilvC expression was explained by the ability of these elements to effect termination of transcription initiated at the ilvY promoter .
Two S. typhimurium ilvY mutants have been described that arose by imprecise excision of TnJO from ilvA : : TnJO insertion mutations ( 5 , 6 ) .
Chromosomal alterations associated with imprecise excision of TnWO are either inversions or deletions ( 13 ) .
Since the pertinent gene order of the S. typhimurium chromosome is i1vA-py-i1v Y-i1vC , then ilvA : : TnJO-generated mutations affecting ilvY expression would necessarily disrupt the association between py and ilvC by inverting py or deleting py .
However , both S. typhimurium ilvY mutants could be complemented in trans by appropriate F ' ilv or recombinant plasmids ( 5 , 6 ) .
In view of the apparent difference between the S. typhimurium and E. coli K-12 ilvY-ilvC systems , we used the S. typhimurium ilv YC-galK plasmids to define further the relationship between ilv Y and ilvC .
Deletion of the entire ilv Y gene from pDU60 did not affect ilvC-galK expression other than to impose a trans-recessive ilvY requirement for acetohydroxyacid induction ( Table 4 ) .
This is particularly striking with pDU63 , in which the 0.42-kb BglII-EcoRI fragment directs acetohydroxyacid-responsive galK expression , localizing Pc within this short sequence .
If ilvC is transcribed independently from i1vY , then small deletions removing the ilvY promoter should not affect Pc function .
This was demonstrated by BAL-31 nuclease digestion of pDU60 at the SstI site , which is in close proximity to py .
A total of 50 derivatives were obtained which lacked ilvY function but retained ilvC-directed galK expression .
One representative plasmid , pDU60 .43 , contains a 0.7-kb deletion , of which 0.4 extends into i1vY .
In strain DU650 A ( ilvEDA YC ) , pDU60 .43 galK expression was constitutive , reflecting the absence of ilvY function ( Table 6 ) .
Enzyme assay of strain C600 ilv + ( pDU60 .43 ) , however , showed that galK expression was responsive to induction by added acetohydroxyacids or by endogenous induction , varying from 70 U under repressing conditions to 320 U with a-acetolactate or 260 U in response to leucine limitation .
Deletion of the ilv Y promoter and N-terminal coding region , therefore , has no effect on ilvC-directed galK activity .
Dissociation of the Pc sequence from the ilvY transcriptional unit can also be effected by inverting the ilvY gene with respect to ilvC .
The ilvY structural gene and promoter are contained within the XhoI-SaiI fragment .
Both XhoI and Sall restriction endonucleases generate staggered ends with the sequence 5 ' - TCGA , but XhoI-Sall hybrid sites are resistant to cleavage with either enzyme ( 17 ) .
Inversion of ilv Y was accomplished by digesting pDU60 with both XhoI and SalI endonucleases and then ligating the fragments .
The derivative of pDU60 containing ilvY in the opposite orientation , designated pDU64 , was recognized by its identical size and its resistance to digestion by either XhoI or SalI .
Enzyme assay of strain DU650 A ( ilvEDA YC ) ( pDU64 ) showed an a-acetohydroxybutyrate-inducible galK activity , indicating that the activity of Pc persisted despite the inverted orientation of ilv Y .
In fact , inversion of the XhoI-SalI fragment was associated with an enhancement of basal Pc activity .
Whether this results from a cis-active modulator of Pc activity which is present near the Sall site is considered below .
The E. coli K-12 ilv Ygene product is proposed to act as an antiterminator that permits readthrough transcription from ilv Y to i1vC .
The possibility of readthrough transcription from the S. typhimurium ilvY gene was examined with pDU71 , which contains the intact ilv Ygene and an additional 0.45 kb of DNA sequence beyond the Sall site .
Because of the propinquity of the C-terminal end of ilv Y to SalI and the Pc sequence to the BglII terminus of the pDU71 insert , it is likely that the ilv Y transcriptional terminator resides within this region .
If antitermination occurs , readthrough transcription of the adjacent galK gene should be detected as increased galK expression .
The pDU71-determined galK activity was increased to 15 U when measured in a S. typhimurium rho mutant , AA111 ( 12 ) .
This level of activity was much lower than that seen with ilvY-galK fusion plasmids such as pDU71 .1 , suggesting that most transcription from py was still properly terminated .
Expression of galK was not enhanced by the presence of ax-acetohydroxyacids in the strain AA111 rho background , measuring 10 U with oa-acetolactate and 15 U with a-acetohydroxybutyrate .
Therefore , readthrough transcription from the S. typhimurium ilvY gene can occur at low levels in a rho mutant , but this activity is not enhanced by ilvC-specific regulatory signals .
The results with ilvC-galK fusion plasmids demonstrate that the S. typhimurium promoter Pc is distinct and is active independently of the ilvY transcriptional unit .
The a-ace-tohydroxyacids and ilv Y gene product modulate the level of transcription initiated at Pc-Readthrough transcription of ilvC from the ilv Y transcription ` unit probably can occur but this does not contribute significantly to the apparent activity of the ilvC promoter .
trans complementation of an S. typhimurium ilvY mutant .
The ilv Y702 mutation was generated by imprecise excision of TnWO from strain TT69 ilvA : : TnlO .
The phenotype of strain DU702 ilvA ilv Y702 is an isoleucine-valine bra-dytrophy which is explained by the low-level , noninducible ilvC expression permitted by the ilv Y702 mutation ( see Table 7 ) .
The ilv Y702 mutation was transduced into a readily transformable S. typhimurium strain DU2 resulting in strain DU703 , and various ilv recombinant plasmids were introduced for complementation studies .
The original pDU a Modified Davis-Mingioli medium was supplemented with 10 mg of Ltryptophan per liter .
Minimal medium additionallly contained 50 mg of L-leucine per liter .
b Isomeroreductase activity is expressed as nanomoles of NADPH oxidized per minute per milligram of protein .
ilvGEDA Y + plasmid conferred inducibility to ilvC expression , indicating trans complementation by the plasmid ilv Y gene ( Table 7 ) .
The same result occurred with pDU60 ilvDA Y + - pc-galK , but ilvC expression was not inducible with pDU63 pc-galK since both plasmid and chromosome lack functional ilv Y genes .
These results confirm the genetic assignm , ent of the ilv Y702 mutation and further demonstrate the trans-recessive nature of S. typhimurium ilv Y mutations .
The effects of plasmids containing intact ilvY but lacking the Pc sequence were also examined .
One plasmid , pDU5 , contains a Sall ilvEDA Y + insert that ends at the Sall site immediately distal to ilv Y. Enzyme assay of strain DU703 ilvY702 ( pDU5 ) showed the expected inducible expression of ilvC .
The pDU71 plasmid contains the intact ilv Y gene as well as the distal 0.45-kb region between Sall and BgIII .
No induction of ilvC occurred in strain DU703 ( pDU71 ) , suggesting that the presence of the additional sequence between Sall and BgiII prevents trans-active complementation by ilvY .
This effect is apparently obviated by the presence of the contiguous Pc ( contained within the BglII-EcoRI region ) since trans-complementation does occur with pDUl and pDU60 .
The lack of trans complementation is also seen with pDU72 , ruling out some orientation-specific effect peculiar to the pKO system ( data not shown ) .
The SaiI-BglII region probably does not act as a negative regulator of ilv Y expression because normal levels of the ilv Y gene product encoded by pDU71 or pDU72 are detected in maxicell preparations ( unpublished data ) .
The possible role of the SaiI-BglII region as a cis-active regulatory element is considered below .
DISCUSSION The results presented herein demonstrate that the S. typhimurium ilvC gene is transcribed from a promoter located within the 0.42-kb BglII-EcoRI region of the cloned ilv fragment .
This sequence is completely sufficient for ace-tohydroxyacid-inducible galK expression in the ilvC-galK fusion plasmids .
The specific 4ctivities of the purified E. coli TABLE 6 .
ilvC-directed galK expression in the absence of transcription of a contiguous ilv Y gene Galacto-kinase sp act ( galK ) b 80 120 Strain MediUMa DU650 A ( ilvEDA YC ) ( pDU60 .43 ilvDA 1AilvY iIvC ' - galK ) C600 ilv + ( pDU60 .43 ) Repressing Limiting valine Repressing 70 Repressing + AL Limiting leucine 320 260 K-12 galactokinase and S. typhimurium LT2 ac-ace-tohydroxyacid isomeroreductase enzymes have been reported previously ( 11 , 25 ) .
If it is assumed that the copy number of pDU60 ( a pBR322 derivative ) is 30 per cell , then , based upon the activities of the purified enzymes , the strain C600 ( pDU60 ) - encoded galK expression ( Table 3 ) corresponds to a haploid , noninduced ilvC activity of 3 U ( nano-moles of NADPH oxidized per minute per milligram of protein ) and an induced level of 25 U. Typical isomeroreductase levels for S. typhimurium ilvG mutants ( which are most similar to E. coli K-12 because of the cryptic nature of ilvG in that organism ) range from 10 to 40 U for basal expression and 100 to 500 U upon induction ( D. Primerano and R. 0 .
The induction ratio of ilvC-di-rected galK expression , therefore , is similar to that seen with the native ilvC gene , although the normalized galK activity is about 10-to 20-fold lower than the haploid ilvC levels .
Since the galK gene sequence contains its own ribosome-binding site and should be translated with invariant efficiency regardless of the contiguous promoter ( 19 ) , one explanation for this discrepancy is that the galK mRNA sequence is translated with less efficiency than the ilvC sequence , thus producing fewer protein products from each galK mRNA molecule than from ilvC mRNA .
Although the 0.42-kb BgiII-EcoRI fragment contains the intact Pc element , the adjacent sequence also appears to modulate Pc activity .
For instance , each deletion derivative of pDU60 , such as pDU61 [ A ( Xho-Sa ( I ) ] , pDU62 ( ABgiII ) , or pDU63 [ A ( BamHI-BglII ) ] , demonstrates a higher basal activity of pc than does pDU60 .
If the region immediately 5 ' to Pc is altered by inversion ( pDU64 ) , an increase in Pc also occurs , although the promoter remains respopsive to acetohydroxyacid induction .
It appears , therefore , that the region defined by the Sall-BglII termini may exert a negative modulation of Pc activity , reducing the basfl pc level .
Negative modulation is circumvented by the putative induction complex composed of the ilvY gene product and acetohydroxyacids , and a large component of the ilvC gene induction ratio can be ascribed to action at this negative regulatory region .
Some facilitation of Pc activity must also occur at a site closer to Pc since all of the p1DU60 derivatives ( both deletion and inversion ) retain some capacity for induction by acetohydroxyacids .
This characteristic of pDU60 derivatives is reflected by the phenotype of the ilv Y702 mutation which was generated by an iivA : : TnJO-mediated chromosomal alteration ( probably , a deletion mutation ) .
This mutation results in a fivefold elevation in the basal expression of ilvC ( Table 6 ) , and when ilvY is provided in trans , normal induction of ilvC can occur .
We infer that the ilv Y702 deletion extends into the negative regulation region .
Negative modulation of Pc does not require intact ilvY gene product since small deletions of the N-terminal region of ilv Y do not alter Pc basal activity ( Table 6 , pDU60 .43 ) .
Similarly , we have isolated one iivA : : TnJO-generated ilvY mutation which shows low basal Pc activity ( 5 ) .
An additional effect of the interface region between ilvY and ilvC is suggested by the inability of certain ilv Y + plasmids containing the SalI-BgiII fragment to act in trans ( Table 7 , pDU71 and pDU72 ) .
If the SaiI-BgiII region is altered by deletion or if it is coupled with the coptiguous Pc sequence , then the ilv Y gene can act in trans to effect ilvC induction ( Table 7 , pDU5 or pDU6 and pDU60 , respectively ) .
Although these observittions are as yet preliminary , we are intrigued by the finding that this region also appears to exert negative modulation ofPc and may be a site at which the ilv Y gene product acts to increase Pc activity .
On possibility is that the ilv Y gene product avidly binds to this sequence in the absence of acetohydroxyacids but does not interact with Pc .
The binding of the coinducers to the ilvY gene product could then permit the induction complex to ihteract with the contiguous and Pc sequence increase transcription of ilvC .
The activated induction complex may then be trans active .
In the absence of the contiguous Pc sequence ( as in pDU71 and pDU72 ) , the necessary transcriptional interaction to form and then release activated complex cotUld not occur .
The nature of the ilvY-ilvC interface is currently under investigation .
Analysis of the E. coli K-12 ilvY-ilvC system employing ilvY insertion mutations has led to the conclusion that ilvC expression is dependent upon transcription initiated at the ilv Y promoter .
This is in clear distinction to the S. typhimur-in which it is that ilv Y and ilvC ium system , apparent constitute separate transcriptional units .
It is certainly possible that these two species utilize disparate control mechanisms to govern ilvC expression .
Another interpretation is that the TnJO and Mu insertion mutations in the E. coli ilvY gene are located within the interface region between ilv Y and ilvC .
These mutations could be positioned so as to permit binding of the ilvY gene product but prevent its interaction with the contiguous ilvC promoter .
This would negate cisactive induction of ilvC well induction since the as as trans putative release of activated induction complex after interaction with Pc could not occur .
Further analysis of both the E. coli K-12 and S. typhimurium ilvY-i1vC systems is needed to clarify these possibilities .
Induction of ilvC in an S. typhimurium ilvY mutant by certain S. typhimurium ilvY + recombinant plasmids Isomero-reductase sp act ( ilvC ) b Repressing 50 Limiting leucine 45 Strain Medium ' DU703 ilvY702 leuA409 DU703 ( pDU1 ilvGEDA Y+C ' ) Repressing 60 Minimal 420 DU703 ( pDU60 ilvDA YC ' - galK + Repressing 40 Limiting leucine 200 DU703 ( pDU63 Repressing 40 A [ ilvDA Y ] ivC ' - galK + Limiting leucine 42 DU703 ( pDU5 ilvDA Y + ) Repressing 50 Minimal 190 DU703 ( pDU71 ilvA'Y + ) Repressing 50 Limiting leucine 50 As always , we thank borothy Thompson and Dayle Wilkins for their considerable assistance .
This work was supported in part by Public Health Service grant GM12551 from the National Institutes of General Medical Sciences .
D.L.B. was a trainee in the Medical Scientist Training Program and was funded by Public Health Service grant GM07171 from the Natiortal Institutes of Health .
Arfin , S. M. , B. Ratzkin , and H. E. Umbarger .
The metabolism of valine and isoleucine in Escherichia coli .
The role of ihduction in the derepression of acetohydroxy-acid isomeroreductase .
Arfin , S. M. , and H. E. Umbarger .
Purification and properties of acetohydroxy-acid isomeroreductase of Salmo-nella typhimurium .
Birnboim , H. C. , and J. Doly .
A rapid alkaline extraction procedure for screening recombinant plasmid DNA .
Blazey , D. L. , and R. 0 .
Genetic organization of the Salmonella typhimurium ilv gene cluster .
Blazey , D. L. , and R. 0 .
Gene ilvY of Salmonella typhimurium .
Blazey , D. L. , R. Kim , and R. 0 .
Molecular cloning and expression of the ilvGEDA Y genes from Salmonella typhimurium .
Davis , B. D. , and E. S. Mingioli .
Mutants of Escherichia coli requiring either methionine or vitamin-B12 .
Burns , and H. E. Umibarger .
Control of isoleucine , valine and leucine biosynthesis .
Gray , H. B. , D. A. Ostrander , J. L. Hodnett , R. J. Legerski , and D. L. Robberson .
Extracellular nucleases of Pseu-specific domonas BAL-31 .
I. Characterization of single-strand deoxyribonuclease and double-strand deoxyribonuclease activities .
Greene , P. J. , M. C. Betlach , H. W. Boyer , and H. M. Goodman .
The EcoRI restriction endonuclease .
Hofler , J. G. , C. J. Decedue , G. H. Lunginbuhl , J. A. Reynolds , and R. 0 .
The subunit structure of a-ace-tohydroxyacid isomeroreductase from Salmonella typhimur-ium .
Housley , P. R. , A. D. Leavitt , and H. J. Whitfield .
Genetic analysis of a temperature-sensitive Salmonella typhimurium rho mutant with an altered rho-associated polycytidylate-dependent adenosine triphosphatase activity .
Kleckdier , N. , K. Reichardt , and D. Botstein .
Inversions and deletions of the Salmonella chromosome generated by the translocatable tetracycline resistance element TnJO .
Lederberg , E. M. , and S. N. Cohen .
Transformation of Salmonella typhimurium by plasmid deoxyribonucleic acid .
Hodnett , and H. B. Gray .
Legerski , R. J. , J. L. lular nucleases of Pseudomonas BAL-31 .
Use of the double-strand deoxyribonuciease activity as the basis of a convenient method for mapping of fragments of DNA produced by cleavage with restriction enzymes .
The determination of the molecular weight of ribonucleic acid by polyacrylamide gel electrophoresis .
The effects of changes in conformation .
Maniatis , T. , E. F. Fritsch , and J. Sambrook .
Molecular cloning : a laboratory manual .
Cold Spring Harbor Laboratory , Cold Spring Harbor , N.Y. 18 .
McCorkle , G. M. , T. D. Leathers , and H. E. Umbarger .
Physical organization of the ilvEDAC genes of Escherichia coli K-12 .
McKenney , K. , H. Shimatake , D. Court , U. Schmeissner , C. Brady , and M. Rosenberg .
In J. Chirikian and T. Papas ( ed .
) , Gene amplification and analysis , vol .
Taillon , M. P. , D. Amanda-Gotto , and R. P. Lawther .
The DNA sequence of the promoter-attenuation of the ilvGEDA operon of Salmonella typhimurium .
Amino acid biosynthesis and its regulation .
Watson , M. D. , J. Wild , and H. E. Umbarger .
Positive control of ilvC expression in Escherichia coli K-12 : identification and mapping of regulatory gene ilvY .
Weiss , B. , A. Jacquemin-Sablon , T. R. Live , G. C. Fareed , and C. C. Richardson .
Enzymatic breakage and joining of deoxyribonucleic acid .
Further purification and properties of polynucleotide ligase from Escherichia coli infected with bacteriophage T4 .
Wild , J. , J. M. Smith , and H. E. Umbarger .
In vitro synthesis of,-galactosidase with ilv-lac fusion deoxyribonucleic acid as template .
Wilson , D. B. , and D. S. Hogness .
The enzymes of the galactose operon in Escherichia coli .
The frequencies of translation of the terminal cistrons of the operon .