Mutations in trans That Affect Formate Dehydrogenase Gene Expression in Salmonella typhimurium Expression of the fdhF locus of Salmonella typhimurium is shown to be dependent upon ntrA and oxrB .
However , the oxrB8 mutation is pleiotropic and also affects the expression of hyd , pepT , and chiC , whereas a mutation in ntrA does not .
Insertional inactivation with TnWO and localized mutagenesis permitted the definition and partial characterization of two new genes , fdhS and fdhR , which appear to be involved in the positive regulation of fdhF expression .
Both genes were mapped to the 71-to 72-min region of the S. typhimurium chromosome with the gene order fdhS-crp-fdhR-rpsL .
Mutations in fdhS specifically affect fdhF expression without affecting the expression of the other anaerobically induced genes or enzymes that were tested , including hyd , pepT , chiC , nitrite reductase , sulfite reductase , and trimethylamine-N-oxide reductase .
Both fdhR and fdhS may be involved infdhF regulation vis-a-vis oxygen , since localized mutagenesis produced alleles of both genes that permitted the aerobic expression offdhF .
However , fdhR may more directly interact withfdhF , since insertional inactivation of fdhS does not abolish aerobic expression of fdhF in fdhR mutant strains .
Taken together , these results suggest that fdhS and fdhR act in concert under anaerobic conditions to activate fdhF transcription .
Much attention has been focused in recent years on detailing the molecular mechanisms involved in the genetic regulation of anaerobic metabolism in members of the family Enterobacteriaceae ( 33 ) .
In particular the mechanisms that enable Enterobacteriaceae to switch from aerobic respiration to anaerobic metabolism ( fermentation or anaerobic respiration ) have attracted much interest .
Under anaerobic-growth-conditions , these facultative bacteria are able to ferment glucose to produce lactate , acetate , ethanol , succinate , and formate ( 10 ) .
The formate that is produced can be converted into carbon dioxide and hydrogen by the formatehydrogen lyase complex .
This complex consists of a formate dehydrogenase ( coded for by fdhF ) , redox carriers , and hydrogenase isoenzyme 3 ( 6 ) .
Salmonella typhimurium and Escherichia coli contain at least three hydrogenase isoenzymes ( 16 , 17 ) .
The formate dehydrogenase gene ( fdhF ) is located min at 93 in S. typhimurium ( 27 ) , and several containing operons genes coding for hydrogenase synthesis and activity are probably located in the 58-59-min region , according to to recent studies performed with E. coli ( 6 , 28-30 ) .
Induction of fdhF synthesis requires the of formate and molyb-presence date and the absence of electron-acceptors like and oxygen nitrate ( 5 , 9 , 31 , 36 , 37 ) .
There is also need for acidic pH a an in the extracellular environment ( 23 ) .
At the transcriptional level , anaerobic expression of fdhF has been shown ( 4 ) to require an alternate enterobacterial sigma factor ( NtrA ) ( 14 ) and a cis-acting upstream activating sequence ( UAS ) ( 2 ) located at about 100 bp 5 ' to the transcription start site fdhF .
Deletion or mutagenesis of this region shows that it is absolutely required for formate induction of the fdhF gene ( 2 ) .
Indeed , it has not yet been possible to find physically separable sequences in or near the promoter region offdhF that respond to the three distinct stimuli ( oxygen , nitrate , and formate ) known to regulate fdhF expression .
In addition , the transcription of several genes involved in the synthesis of active hydrogenase has been shown to require NtrA ( 4 , 6 , 18 ) .
Unlike the case with other enterobacterial sigma factors , activation of transcription by NtrA has been shown to require accessory , DNA-binding , trans-acting factors that bind to specific UASs that have many of the properties of eucaryotic enhancers ( 7 , 8 , 26 ) .
The activity of these trans-acting factors is usually regulated by phosphor-ylation and dephosphorylation .
Since the fdhF UAS mediates repression by nitrate and oxygen and induction by formate and molybdate , it would appear to be the site at which a putative transcriptional activator , homologous to ntrC or nifA , would interact .
Additionally , it has been shown , by using a chimeric fdhF gene in which the fdhF UAS was exchanged for a nif UAS , that the UAS is solely for the responsible transcriptional regulation of fdhF ( 3 ) , again strongly implying the binding of a trans-activating factor at this side .
What are the genes whose products activate transcription offdhF ?
Do they act directly by binding at the fdhF UAS or indirectly as part of a regulatory cascade ?
Recent studies with E. coli ( 20 , 32 , 34 ) have demonstrated three factors that could potentially play a role , either direct or indirect , in the regulation of fdhF .
However , their exact role and at what step they act in the regulation offdhF are not clear .
One of these factors , HydG , shows a high degree of homology with the transcriptional activator NtrC from Klebsiella pneumoniae ( 34 ) .
It is apparently required for the activity of the labile hydrogenase isoenzyme 3 ( 34 ) .
However , it has yet to be determined what factors serve to activate HydG or whether HydG is involved in the regulation offdhF .
Another factor , fhlA , is located in the 59-min region of the E. coli chromosome , codes for a protein of 78 kDa , FhlA ( 29 ) , and is capable , when cloned into a multicopy plasmid , of complementing a mutant strain that lacks formate-depen-dent expression offdhF and hyd-17 , a gene that is necessary for the activity of hydrogenase isoenzyme 3 .
The FhlA protein has been proposed to be a transcriptional activator for the production of the formate-hydrogen lyase complex , since it shows a high degree of homology with other transcriptional activators that function in concert with NtrA , such as NtrC , HydG , and Klebsiella pneumoniae NifA ( 20 , 32 ) .
The hyp operon , which contains fhlA and five other genes that are necessary for the activity of all three hydrogenases , is expressed both aerobically and anaerobically ( 19 ) .
What is the function of FhlA , and what factor in turn regulates its activity ?
FhlA has recently been implicated frilB in the regulation of the expression of ( 21 ) , which is located in the 95-min region of the E. coli chromosome .
frhIB has been suggested as a third potential regulator of fdhF , since it is required for the formate dehydrogenase activity associated with formate hydrogenlyase and since mutations in fhIB pleiotropically affect all three hydrogenase isoenzymes ( 21 ) .
FhlB is maximally expressed under anaerobic conditions in the presence of formate , and it has been proposed that formate interacts with the FhlA protein to initiate transcription of fhIlB ( 21 ) .
Thus , the picture of how fdhF is regulated in E. coli is far from clear .
As part of an ongoing interest in the regulation of anaerobic metabolism in S. typhimurium , we are attempting to define the genetic factors required for the regulation offdhF expression in this organism .
In S. typhimurium , very little is presently known about potential fdhF regulatory factors .
In this report , we present evidence for a set offdhF regulatory loci , which differ from flzA and JflB of E. coli and for which we propose the names fdhR and fdhS .
fdhR and fdhS have been mapped in the 71-to 72-min region of the S. typhimu-rium genome .
Our results suggest that at least one of these genes ( fdhR ) may be specifically and directly involved in the regulation of fdhF expression vis-a-vis oxygen , since mutations in this gene cause constitutive expression of fdhF under aerobic conditions .
MATERIALS AND METHODS Strains , media , and growth-conditions .
The bacterial strains used in this study are listed in Table 1 .
For the determination of the correction of auxotrophies , a minimal-medium ( M9 ) containing lactose ( 0.2 % ) or glucose ( 0.2 % ) was prepared as previously described ( 12 ) .
Amino acids and vitamins , when needed , were used at the following concentrations : arginine ( 0.6 mM ) , proline ( 2 mM ) , cysteine ( 0.3 mM ) , tryptophan ( 0.1 mM ) , phenylalanine ( 0.5 mM ) , para-aminobenzoate ( 0.1 mM ) , and tyrosine ( 0.1 mM ) .
MacConkey agar medium contained lactose ( 1 % ) or mannitol ( 1 % ) .
Tetrazolium agar was formulated as NB agar with the addition of 50 , ug of 2,3,5-triphenyl-2-tetrazolium chloride per ml and lactose ( final concentration , 1 % ) ( 22 ) .
Antibiotics were added at the following final concentrations : ampicillin ( 50 , ug/ml ) , kanamycin ( 25 , ug/ml ) , tetracycline ( 20 , ug/ml ) , and streptomycin ( 500 , ug/ml ) .
Minimal media were formulated with one-half of these concentrations .
When needed , MacConkey agar and tetrazolium agar contained 8 , ug of tetracycline per ml .
Strains were tested for TMAO ( trimeth-ylamine-N-oxide ) reductase by streaking on MacConkey agar base that contained glucose ( 0.15 % ) and TMAO ( 0.1 % ) and incubating anaerobically .
Strains were tested for H2S production from sulfite by using nutrient agar that contained FeCl2 ( 1 mM ) , glucose ( 10 mM ) , and Na2SO3 ( 12 mM ) ( 11 ) .
Strains were tested for nitrite reductase activity by scoring for anaerobic-growth on Gutnick medium ( 13 ) containing 0.2 % galactose and 10 mM NaNO2 .
When necessary , plates were incubated anaerobically at 37 °C in Gas-Pak jars .
I When strains are transductants , this is indicated by P22 ( 1 ) x Z , where Y is the strain used to prepare the phage lysate and Z is the recipient strain .
RnTnlO indicates a random pool of TnJO insertions .
* indicates that the phage were subjected to chemical mutagenesis as described in Materials and Methods .
SGSC , Salmonella Genetic Stock Center .
b The mutation in EB137 , formerly fhl-IOJ : : Mu dl lac ( Apr ) , has been redesignatedfdhFIOJ : : Mu dl lac ( Apr ) ( 37 ) .
' The mutation in EB138 [ hyd-I0J : : Mu dl lac ( Apr ) ] pleiotropically affects all three hydrogenases ( 17 ) .
d The Tn5 insertion in TN2064 has been redesignated zhb-895 : : TnS ( 25 ) .
Phage P22 HTJO5Iint4 lysates were prepared and transductions were performed as previously described ( 12 ) .
TnJO insertion mutants that decreased expression of fdhFlOl : : Mu dl ( Apr ) were isolated from a random pool of TnlO insertions kindly provided by John Roth .
Specifically , phage P22 lysates of these pools were used to transduce strain EB137 , and transductants wer screened on MacConkey-tetracycline plates .
When grown anaerobically on MacConkey medium , strain EB137 produces bright red colonies .
Therefore , transductants with an apparently decreased color were kept and further characterized .
In some cases , the identical procedure was followed with tetrazolium agar , except that the expected color change was reversed .
Localized mutagenesis ( 15 ) was used to isolate mutants altered in the regulation offdhF as follows : P22 grown on PH254 ( selection for aerobic expression ) PH456 or for absence-of-nitrate repression ) ( selection was mutagen-ized and used to infect EB137 on , respectively , aerobically incubated minimal lactose medium containing kanamycin or anaerobically incubated minimal lactose medium containing nitrate ( 10 mM ) and tetracycline .
Since S. typhimurium is naturally Lac-and sincefdhF101 : : Mu dl is solely expressed under anaerobic conditions in the absence-of-nitrate ( 1 ) , only transductants possessing putative mutations allowing the altered expression of fdhF should grow under the selection conditions used .
Deletion mutants were generated by using the previously described medium for isolation of tetracy-cline-sensitive mutants ( 20 ) .
Assays of 0-galactosidase activity .
Anaerobic culture densities were monitored with a Spectronic 20 spectrophotometer ( Milton Roy Co. ) .
Aerobic culture densities were monitored with a Hewlett-Packard Diode Array Spectro-photometer 8452A .
Inocula for liquid cultures were aerated overnight at 275 rpm in 2.5 ml of LB-medium ( 23 ) that was buffered ( pH 6.5 ) with 80 mM 2 - ( N-morpholino ) ethanesulfo-nic acid ( MES ) and contained 0.2 % glucose .
Anaerobic-growth was initiated by inoculation ( 84 i.l ) of screw-cap tubes ( 13 by 100 mm ) filled to the top ( 8.4 ml ) with LB-MES-KOH ( pH 6.5 ) and glucose ( 80 mM ) .
Cultures were placed in an incubator at 37 °C .
When needed , formate ( final concentration , 0.2 % ) or nitrate ( final concentration , 10 mM ) was added to the medium .
Aerobic growth was initiated by inoculating 50 [ LI of an overnight culture into loosely capped tubes ( 14 by 120 mm ) that contained 5 ml of LB-MES-KOH ( pH 6.5 ) and glucose ( 80 mM ) .
Cultures were incubated at 37 °C in a shaker ( 275 rpm ) .
Tests indicated that under these conditions the cultures were not 02 limited ( data not shown ) .
3-Galactosidase activities of aerobic and anaerobic cultures were determined with mid-exponential-phase cultures .
Cells were lysed as previously described ( 24 ) , and,-galactosidase activities were measured as described by Miller ( 22 ) .
All assays were conducted in duplicate .
The reported values are averaged from at least three independent experiments .
Activities are expressed in terms of nanomoles of o-nitrophenol per minute per unit of optical density at 600 nm .
RESULTS Mutations that negatively affect fdhF expression .
Previously , it was reported that in E. coli the expression offdhF ( 4 ) and various hydrogenase genes ( 6 , 18 , 30 ) required a functional ntrA .
We have confirmed that fdhF expression in S. typhimurium also requires ntrA by using transposon insertional inactivation , i.e. , introduction of ntrA209 : : TnlO .
As is the case for E. coli , ntrA may be specific for fermentative pathways , since insertional inactivation of ntrA did not affect the expression of phs ( necessary for the production of hydrogen sulfide ) , chiC ( the nitrate reductase operon ) , and pepT ( an anaerobically induced peptidase ) .
To search for potential regulators that could act in concert with NtrA to activate fdhF expression , we used transposon mutagenesis .
This approach has at least two advantages : the creation of a null phenotype and the introduction of a marker Background hyd-101 0.2 % formate , 80 mM glucose phs-101 80 mM glucose pep77 80 mM glucose chiC101 10 mM nitrate , 80 mM glucose a All strains were grown under anaerobic-growth-conditions in LB-MES-KOH ( pH 6.5 ) with the indicated additions .
b OD6. , optical density at 600 nm .
that facilitates further genetic analysis .
Therefore , pools of random TnJO insertions were screened for a decrease infdhF expression by transducing strain EB137 ( fdhFlOl : : Mu dl ) with phage lysates of these pools and selecting for tetracycline resistance on MacConkey-tetracycline plates .
Colonies showing the desired phenotype , i.e. , a paler color than that of the parent strain EB137 , were isolated , and the phenotype ( decreased fdhFJOJ : : Mu dl expression ) was verified by , B-galactosidase assays .
One insertion that showed an apparent decrease in fdhF expression was retained for further characterization .
As shown below , this insertion is in a new locus for which we propose the name fdhS .
Thus , this insertion will be referred to asfdhSlOl : : TnlO .
The effects of this mutation on the expression offdhF and other anaerobically induced genes were determined ( Table 2 ) .
The insertional inactivation offdhS caused a strong reduction offdhF expression ( -10-fold ) and a smaller decrease in phs expression ( -2-fold ) .
However , assays of , B-galactosidase activities of strains carrying fdhS101 : : TnJO showed that the TnJO insertion had no effect on the expression of chlC1Ol , pepT7 , and hyd-101 .
( The hyd-101 operon fusion does not appear to be in a hydrogenase structural gene , because it pleiotropically affects all three hydrogenases [ 16 , 17 ] .
) All strains carrying fdhS101 : : TnJO were incapable of gas production .
Thus insertional inactivation offdhS has the same effects as does insertional inactivation of ntrA on the fusions tested .
Previously , it was reported that oxrB8 , for which no counterpart in E. coli has yet been described , decreased the anaerobic expression of the peptidase pepT as well as several unidentified operon fusions ( 35 ) .
Since preliminary mapping data had placed fdhS in the same chromosomal region as oxrB , it was of interest to determine whether oxrB8 would have the same effect as a mutation infdhS .
We found that the expression of fdhF was reduced about 10-fold in a strain carrying oxrB8 ( Table 2 ) but that , unlike fdhS101 : : TnJO , it does not seem to specifically affect fdhF expression because strains carrying oxrB8 also showed decreased expression of hydlOI , pepT7 , and chIC101 .
Thus , the spectrum of effects shown by oxrB8 is different than that shown by fdhSJOJ : : TnlO : oxrB8 is pleiotropic , whereas fdhS appears to be specific for fdhF .
Furthermore , ntrA is required for the expression of fdhF but not for the other genes studied , strongly suggesting that ntrA and fdhS are working in concert to activate fdhF expression .
Insertional inactivation of fdhS or ntrA does not directl affect TMAO , nitrite , fumarate , or sulfite reductase .
The question arose as to whetherfdhS and/or ntrA was involved in the regulation of other anaerobic genes .
There are a variety of anaerobic metabolic pathways whose regulation in S. typhimurium is presently not understood .
For example , unlike fnr mutants of E. coli , oxrA mutants of S. typhimu-rium have been reported to be capable of anaerobic-growth-with-fumarate as the electron-acceptor or nitrite as the nitrogen source ( 35 ) .
Likewise , the genes whose products regulate anaerobic sulfite reduction have yet to be identified .
Therefore a variety of tests were carried out comparing the phenotypes of strains carrying either fdhSJOJ : : TnJO or ntrA209 : : TnJO with the wild-type strain , S. typhimurium LT2 ( Z ) .
By using the appropriate indicator media ( see Materials and Methods ) , it was determined that insertional inactivation offdhS had no effect on either TMAO or sulfite reduction .
Likewise , a strain carrying fdhS101 : : TnJO was still capable of anaerobic-growth with nitrite as the sole nitrogen source .
On the other hand , growth of strains carrying mutations in either fdhS or ntrA was greatly decreased over that of the wild type on glycerol-fumarate medium .
However , these effects are probably secondary , since essentially the same result ( poor growth ) was obtained with strain EB137 ( fdhFJOJ : : Mu dl ) .
Thus , fdhSlOJ : : TnlO appears to affect fdhF expression specifically with essentially no direct effect on TMAO reductase , nitrite reductase , fumarate reductase , or sulfite reductase ( data not shown ) .
fdhS is located near min 72 of the Salmonella map .
The fdhS gene was mapped in the 71-to 73-min region of the S. typhimurium genome by P22 transduction analysis with the following markers : argD , cysG , rpsL , crp , aroE , and zhb-895 : : TnS .
The results , summarized in Fig. 1 , are presented in kilobase pairs ; we used the formula given by Sanderson and Roth ( 27 ) to take into account the influence of the sizes of the TnJO and Tn5 elements on apparent transduction frequencies .
These results show that fdhS is highly linked ( 98.4 % ) to crp .
It was also found to be linked ( 40 % ) to a streptomycin resistance marker ( rpsL ) near min 71.5 and linked ( 41 % ) to argD near min 72 .
Three-factor crosses ( data not shown ) established the gene order fdhS-crp-rpsL .
fdhS was unlinked to either cysG ( min 73 ) or aroE ( min 71 ) .
Since crp and fdhS were found to be highly linked , they were tested for possible identity by two different methods .
First , a crp mutant strain was tested for its ability to produce gas under anaerobic conditions with nutrient agar stabs containing glucose ( final concentration , 0.5 % ) ; production of gas was normal .
Likewise , fdhS mutant strains were tested for their crp character by examining their ability to fermen mannitol on MacConkey-mannitol plates ; they were capable of fermenting mannitol .
These results strongly suggest that fdhS and crp are not the same gene .
However , the possibility that fdhS is a highly unusual allele of crp has not yet been rigorously excluded .
No analogous mutations have yet been mapped to this position in E. coli .
Mutations that allow aerobic expression of fdhF .
As detailed above , a null mutation in fdhS caused a drastic reduction in anaerobicfdhF expression .
Thus fdhS appeared to be necessary for the positive control offdhF expression .
We reasoned that if this were indeed the case and if fdhS were involved in the response to the lack of oxygen , then it might be possible to find mutations in fdhS that allow the aerobic expression of fdhF .
To search for such mutations , localized mutagenesis ( 12 , 15 ) was carried out in this region with a P22 phage lysate of strain PH254 ( zhb-895 : : TnS fdhFJOl : : Mu dl ) .
The mutagenized phage lysate was used to transduce strain EB137 ( fdhFOJ : : Mu dl ) , and transductants capable of expressing fdhFJOJ : : Mu dl aerobically were selected on minimal lactose-kanamycin medium ( S. typhimu-rium is naturally Lac - ) .
The mutants obtained could be classified into three groups according to the linkage of the mutation allowing aerobicfdhF expression to thefdhS : : TnJO insertion .
Group I contains five strains that have a mutation highly ( 97 to 100 % ) linked tofdhS01 : : TnJO , and presumably were in the fdhS gene .
Group II contains four strains that have a mutation that is 55 to 85 % linked to fdhSJOl : : TnJO .
The results of P22 transductional linkage analysis together with the results of,-galactosidase activity measurements ( see below ) suggest that these group II mutations lie in a second gene regulatingfdhF .
We propose the name fdhR for this locus .
Mapping of these mutations with P22 established the gene order fdhS-crp-fdhR-rpsL ( Fig. 1 ) .
This gene order was confirmed by three-factor crosses ( data not shown ) .
Finally , group III contains two strains with mutations that are 15 to 20 % linked to fdhSlO0 : : TnJO .
These mutations have not been further characterized .
To quantitate the effect offdhS ( group I ) and fdhR ( group II ) mutations on fdhF expression , P-galactosidase assays of mid-exponential-phase cultures of two isolates of each locus ( fdhS , PH360 and PH437 ; fdhR , PH359 and PH361 ) were carried out ( Table 3 ) .
As expected from the selection procedure used , the aerobic 3-galactosidase activities of the mutant strains examined were considerably higher than that of the parent fusion ( EB137 ) .
Interestingly , these mutations also diminished the anaerobic induction offdhF .
Even more striking was the marked alleviation of nitrate repression of fdhF expression in these mutants .
Nitrate caused a 12-fold decrease in fdhF expression in the parent strain EB137 , whereas there was no decrease found with nitrate with strains PH360 , PH437 , and PH359 and fdhF expression in strain PH361 was only decreased 1.4-fold .
Thus , the effect of the mutations in these strains was to cause , under the conditions used here , an apparent constitutive expression of fdhF .
The major difference observed between the mutations infdhS ( strains PH360 and PH437 ) and the mutations infdhR ( strains PH359 and PH361 ) that were examined was the higher level of fdhF expression in strains with fdhR mutations .
Thus , under aerobic conditions , fdhF expression by strains PH361 and PH359 was 8-to 11-fold greater than that of strain EB137 , whereas fdhF expression by strains PH360 and PH437 was only 4-to 5-fold greater .
( These differences are statistically significant .
) These results suggest that both fdhR and fdhS may be involved , either directly or indirectly , in the regulation of fdhF expression vis-a-vis oxygen .
The introduction of fdhSlOl : TnlO into group II strains PH359 and PH361 had no effect on aerobic fdhF expression as ascertained with P-galactosidase assays ( compare the results for strains PH359 and PH502 with those for strains PH361 and PH503 in Table 3 ) .
This suggests that the fdhR locus plays a more direct role than does the fdhS locus in the regulation of fdhF ; i.e. , fdhR is epistatic to fdhS .
Mutations that relieve anaerobic nitrate repression offdhF expression .
As mentioned above , mutants selected for aerobic fdhF expression were also found to be affected in anaerobic nitrate repression .
One possible explanation for this effect is that these mutations are such that the gene product offdhR orfdhS is locked into an active configuration and no longer responds to its normal effectors .
Thus it was of interest to directly select for mutants that had lost anaerobic nitrate repression of fdhF expression .
To do this , localized mutagenesis of the fdhR ( fdhS ) region was carried out with a P22 phage lysate of PH456 ( zhb-6755 : : TnJO ) .
The mutagen-ized phage lysate was used to transduce strain EB137 ( fdhFIOl : : Mu dl ) , and transductants were selected on an anaerobically incubated lactose minimal-medium containing nitrate ( 10 mM ) and tetracycline .
Four putative mutants ( strains PH457 through PH460 PH361 fdhFJOJ : : Mu dlfdhRJ02 PH422 fdhFlOJ : : Mu dl Dpll8 [ lys ( serA cysGJ542 : : Tn5 ) ( serA cysG + ) ilv ] PH463 fdhF101 : : Mu dl fdhR101 DP118 [ lys ( serA cysGJ542 : : TnS ) ( serA cysG + ) i1v ] PH464 fdhF101 : : Mu dl fdhR102 DP118 [ lys ( serA cysG1542 : : TnS ) ( serA cysG + ) i1v ] a Cultures were grown aerobically in Gutnick medium ( 12 ) containing 0.4 % glucose , 20 mM NH4Cl , and 12.5 p.g of kanamycin per ml to the midexponential phase , and P-galactosidase activity was determined as described under Materials and Methods .
Strains PH422 , PH463 , and PH464 were derived from duplication strain FT2398 ( obtained from J. R. Roth ) by transduction .
OD6. , optical density at 600 nm .
were isolated and further characterized .
In all cases linkage ( 42 to 58 % ) of the mutations in these strains to zhb-6755 : : TnJO was confirmed .
The results of , B-galactosidase assays conducted on cultures grown aerobically , anaerobically , and anaerobically with nitrate are shown in Table 3 .
As might be expected from the selection procedure used , anaerobic expression offdhF in the presence of nitrate was increased five-to eightfold over that found with the parent strain ( EB137 ) .
Additionally , in all cases the aerobic expression of fdhF was also 3.5-to 6.5-fold greater than that of the parent strain .
These increases might be at least partly explained by a general overall increase in the expression of fdhF .
For example , even when these mutant strains ( PH457 through PH460 ) were incubated anaerobically in the absence-of-nitrate , their,-galactosidase activities increased two-to fivefold .
The linkage to zhb-6755 : : TnJO found for the mutations in these strains was essentially the same as that in the mutations in strains PH359 and PH361 described above .
This indicated that the mutations in strains PH457 through PH460 are in fdhR .
This was confirmed by fine structure mapping with P22 ( Fig. 1 ) .
Aerobic regulation in diploid strains .
Since fdhR interacts more directly than does fdhS with fdhF , we were interested in studying the effects of the presence of the wild-typefdhR allele ( fdhR + ) on fdhF expression when mutant alleles of fdhR were also present .
To do this , strains that were diploid for the fdhR locus , i.e. , fdhRJ01IfdhR + fdhRJ021fdhR + , were constructed by transduction of the required genes into strain TT2398 ( obtained from J. R. Roth ) , which carries a duplication of the chromosomal region between 62 and 73 min .
Diploid strains were maintained and cultured for Pgalactosidase assays by simultaneously selecting for kana-mycin resistance and Cys + .
( Diploidy was verified by segregation of the appropriate markers [ data not shown ] .
) As might be expected , the aerobic expression of fdhF : : Mu dl was unchanged in the fdhR + IfdhR + diploid strain PH422 ( Table 4 ) .
For unknown reasons , the P-galactosidase activities of strains PH359 and PH361 were significantly higher in minimal-medium than in buffered LB .
The P-galactosidase activities of aerobically grown strains diploid for mutant and wild-type alleles offdhR ( PH463 , PH464 ) were 60 % of those of haploid strains ( PH359 , PH361 ) .
These results demonstrate that fdhR101 and fdhR102 are partially dominant over fdhR + .
DISCUSSION The current knowledge of the physiology of fdhF regulation and the molecular architecture of the region upstream of the FdhF coding region presents two salient features : ( i ) NtrA is required for transcription initiation at the fdhF promoter , and ( ii ) another factor that presumably binds at the UAS is also required .
In addition , all of the regulatory signals ( absence-of-oxygen , nitrate concentration , and formate induction ) are integrated at the level of the UAS , the additional factor , or both .
The present study has attempted to further elaborate upon these points .
As detailed in the introduction , recent studies with E. coli have implicated three potential regulatory genes , hydG ( 34 ) , .
fhlA ( 31 ) , andfhlB ( 20 ) , in the regulation offdhF expression .
What is their mode of action , and with what metabolic signals might they interact ?
when cloned into a hydG , multicopy plasmid , causes the expression of hydrogenase 3 in a pleiotropic hydrogenase mutant ( 34 ) .
Whether it controls hydrogenase 3 expression in wild-type cells or has any effect on fdhF expression is presently unknown .
Likewise , the specific role of a second potential regulator , JhIA , in the control of fdhF expression is presently unclear .
Although fhiA has been described as part of the hyp operon that is expressed both aerobically and anaerobically ( 19 ) , studies with JhlA : : cat fusions have suggested that it is only expressed anaerobically ( 29 ) .
Thus , whether one of the signals for fdhF expression is required for JhIA expression in the absence-of-oxygen or whether oxygen interacts with FhlA has yet to be determined .
A potential role for FhlA in sensing formate has been proposed , since it appears to be necessary for the anaerobic , formate-mediated induction of JhlB ( 21 ) .
Mutational analysis indicates that an intactjhlB is absolutely necessary for fdhF expression .
Thus , any model for fdhF regulation must includefhlA andfhlB .
Obviously one potential role is that of a formate sensor .
However , all the signals known to regulate fdhF may not interact with JhlA and JhhlB .
Since JIB expression is not repressed by nitrate , nitrate repression offdhF expression must occur either by interaction with FhlB or via another set of regulatory factors .
Further work will be required to obtain a clear picture of fdhF regulation in E. coli .
Much less is known about fdhF regulation in S. typhimu-rium .
Using transposon and localized mutagenesis , we have in the present study identified two potential factors , fdhS and fdhR , that are implicated in the regulation offdhF expression in S. typhimurium and that differ from the previously described E. coli genes .
These genes were mapped to the 71-to 72-min region of the Salmonella chromosome with the gene orderfdhS-crp-fdhR-rpsL .
No analogous mutations have yet been mapped to this region of the E. coli chromosome .
Two lines of evidence implicatefdhS and fdhR in the regulation of S. typhimuriumfdhF expression .
( i ) A TnJO insertion infdhS specifically diminishes fdhF expression without affecting the expression of chiC , pepT , or hyd .
Also , fdhS mutant strains are unaffected in TMAO , nitrite , or sulfite reduction .
In addition , the introduction of a TnJO insertion in ntrA has the same effect .
( ii ) Localized mutagenesis of this region produces mutations ( mapped to fdhR and fdhS ) that permit , in mutant strains , the aerobic expression of fdhF or its anaerobic expression in the presence of nitrate .
Both types of selection procedure used , the presence of oxygen and the presence of nitrate under anaerobic conditions , produced alleles of fdhR with increased aerobic expression and decreased anaerobic repression by nitrate offdhF expression .
However , the nitrate selection procedure yielded mutant that still showed anaerobic induction of fdhF , whereas the oxygen selection procedure did not .
These differences merit further study .
One possible explanation is that the increased anaerobic induction of fdhF in these strains is due to the increased expression of fdhR .
Merodiploid analysis showed that mutant alleles of fdhR that show increased expression of fdhF aerobically were partially dominant tofdhR + .
Among several possible explanations for the lack of full dominance are ( i ) mutationally altered FdhR and FdhR + competitively binding to a site involved in fdhF regulation and ( ii ) autoregulation of the fdhR locus with FdhR + decreasing aerobic expression of fdhR101 and fdhR102 .
Further experiments will be necessary to differentiate between these and other possibilities .
Nevertheless , these results provide further evidence for the participation offdhR in the positive control of fdhF expression .
The roles offdhS , fdhR , fhIlA , andfhllB in the regulation of fdhF expression are presently unclear .
jhIA has been shown to possess an ntrA-dependent promoter ( 31 ) and may be expressed only in the absence-of-oxygen ( 29 ) .
Thus , it is possible that fdhR and fdhS act in concert to regulate fhIlA expression .
Another possibility that can not be ruled out at present is that fdhR interacts directly with the fdhF UAS to activate transcription .
According to this model , fluB ( or fhIlA ) could communicate the formate status of the cell to fdhR and fdhS could communicate tofdhR the oxygen status ( and possibly the nitrate status ) of the cell .
Such a model is consistent with a number of observations .
Both fdhR and fdhS appear to be involved in the anaerobic expression of fdhF , since mutant alleles of both genes can be obtained that permit the aerobic expression of fdhF .
fdhR may interact with fdhF more directly than does fdhS , since insertional inactivation offdhS does not abolish the aerobic expression of fdhF in fdhR strains .
Single-point mutations in fdhR can cause apparent constitutive expression of fdhF , suggesting that three of the signals known to affectfdhF expression ( the absence-of-oxygen or nitrate and the presence of formate ) interact at or before fdhR .
Although this model can explain the effects of all presently known regulatory factors , obviously at present other models , or the interaction with other factors , such as hydG ( 34 ) , can not be excluded .
Further studies are in progress to characterize fdhS and fdhR on a molecular level .
We thank Ericka Barrett for her generous supply of strains .
Nancy Tremblay and Isabelle Chenier are gratefully acknowledged for their help in the initial isolation and characterization of the mutants .
We thank Marie Lesage and Johanne Plante for skillful aid in the preparation of the manuscript .
This study was supported by the Medical Research Council of Canada ( MT-10424 ) and by the Comite d'attribution des fonds internes de recherche de l'Universite de Montreal .
A.F. was supported by a studentship from Le fonds pour la formation de chercheurs et l'aide a la recherche .
ADDENDUM IN PROOF oxrB8 has been shown to be a mutant allele of rpoA ( M. J. Lombardo , D. Bagga , and C. G. Miller , Abstr .
Barrett , E. L. , H. S. Kwan , and J. Macy .
Anaerobiosis , formate , nitrate , and pyrA are involved in the regulation of formate hydrogen lyase in Salmonella typhimurium .
Birkmann , A. , and A. Bock .
Characterization of a cis regulatory DNA element necessary for formate induction of the formate dehydrogenase gene ( fdhF ) of Escherichia coli .
Birkmann , A. , H. Hennecke , and A. B6ck .
Construction of chimaeric promoter regions by exchange of the upstream regulatory sequences from fdhF and nif genes .
Birkmann , A. , R. G. Sawers , and A. Bock .
Involvement of ntrA gene product in the anaerobic metabolism of Escherichia coli .
Birkmann , A. , F. Zinoni , G. Sawers , and A. Bock .
Factors affecting transcriptional regulation of the formate hydrogen lyase pathway of Escherichia coli .
Bohm , R. , M. Sauter , and A. B6ck .
Nucleotide sequence and expression of an operon in Escherichia coli coding for formate hydrogen lyase components .
Buck , M. , S. Miller , M. Drummond , and R. Dixon .
Upstream activator sequences are present in the promoters of nitrogen fixation genes .
Buck , M. , J. Woodcock , W. Cannon , L. Mitchenall , and M. Drummond .
Positional requirements for the function of nif-specific upstream activator sequences .
Pascal , and F. Casse .
Formate hydrogen lyase system in Salmonella typhimurium LT2 .
The fermentation pathways of Escherichia coli .
Clark , M. A. , and E. L. Barrett .
The phs gene and hydrogen sulfide production by Salmonella typhimurium .
Davis , R. W. , D. Botstein , and J. R. Roth .
Cold Spring Harbor Laboratory , Cold Spring Harbor , N.Y. 13 .
Gutnick , D. , J. M. Calvo , T. Klopotowski , and B. N. Ames .
Compounds which serve as the sole source of carbon or nitrogen for Salmonella typhimurium LT-2 .
Wong , K. Sei , J. Keener , and S. Kustu .
Products of nitrogen regulatory genes ntrA and ntrC of enteric bacteria activate gInA transcription in-vitro : evidence that the ntrA product is a factor .
Hong , J.-S. , and B. N. Ames .
Localized mutagenesis of any small specific region of the bacterial chromosome .
Jamieson , D. J. , and C. F. Higgins .
Two genetically distinct pathways for transcriptional regulation of anaerobic gene expression in Salmonella typhimurium .
Jamieson , D. J. , R. G. Sawers , P. A. Rugman , D. H. Boxer , and C. F. Higgins .
Effects of anaerobic regulatory mutations and catabolite-repression on regulation of hydrogenase metab-olism and hydrogenase isoenzyme composition in Salmonella typhimurium .
Lutz , S. , R. Bohm , A. Beier , and A. Bock .
Characterization of divergent NtrA-dependent promoters in the anaerobically expressed gene cluster coding for hydrogenase 3 components of Escherichia coli .
Lutz , S. , A. Jacobi , V. Schlensog , R. Bohm , G. Sawers , and A. Bock .
Molecular characterization of an operon ( hyp ) necessary for the activity of the three hydrogenase isoenzymes in Escherichia coli .
Maloy , S. R. , and W. D. Nunn .
Selection for loss of tetracycline resistance by Escherichia coli .
Maupin , J. A. , and K. T. Shanmugam .
Genetic regulation of formate hydrogen lyase of Escherichia coli : role of the fn1A gene product as a transcriptional activator for a new regulatory gene , fhIB .
Experiments in molecular genetics .
Col Spring Harbor Laboratory , Cold Spring Harbor , N.Y. 23 .
Pecher , A. , F. Zinoni , C. Jatisatien , R. Wirth , H. Hennecke , and A. Bock .
On the redox control of synthesis of anaerobically induced enzymes in Enterobacteriaceae .
Putnam , S. L. , and A. L. Koch .
Complications in the simplest cellular enzyme assay : lysis of Escherichia coli for the assay of j3-galactosidase .
Quale , R. , and C. G. Miller .
Reitzer , L. J. , and B. Magasanik .
Transcription of glnA in E. coli is stimulated by activator bound to sites far from the promoter .
Sanderson , K. E. , and J. R. Roth .
Linkage map of Salmonella typhimurium , edition VII .
Sankar , P. , J. H. Lee , and K. T. Shanmugam .
Cloning of hydrogenase genes and fine structure analysis of an operon essential for H2 metabolism in Escherichia coli .
Sankar , P. , J. H. Lee , and K. T. Shanmugam .
Geneproduct relationships ofJhIA and fdv genes of Escherichia coli .
Sankar , P. , and K. T. Shanmugam .
Biochemical and genetic analysis of hydrogen metabolism in Escherichia coli : the 31 .
Schiensog , V. , A. Birkmann , and A. Bock .
Mutations in trans which affect the anaerobic expression of a formate dehydrogenase ( fdhF ) structural gene .
Schlensog , V. , and A. Bock .
Identification and sequence analysis of the gene encoding the transcriptional activator of the formate hydrogenlyase system of Escherichia coli .
Nitrate respiration in relation to facultative metabolism in enterobacteria .
Stoker , K. , W. N. M. Reinders , L. F. Oltmann , and A. H. Stouthamer .
Initial cloning and sequencing of hydHG , an operon homologous to ntrBC and regulating the labile hydrog-enase activity in Escherichia coli K-12 .
Strauch , K. L. , J. B. Lenk , B. L. Gamble , and C. G. Miller .
Oxygen regulation in Salmonella typhimurium .
Wu , L. F. , and M.-A .
Regulation of the fdhF gene encoding the selenopeptide for benzyl viologenlinked formate dehydrogenase in Escherichia coli .
Zinoni , F. , A. Beier , A. Pecher , R. Wirth , and A. Bock .
Regulation of the synthesis of hydrogenase ( formate hydrogen-lyase linked ) of E. coli .