Oxygen Regulation in Salmonella typhimurium KATHRYN L. STRAUCH , JEAN BOTTI LENK , BARBARA L. GAMBLE , AND CHARLES G. MILLER * Department of Molecular Biology and Microbiology , Case Western Reserve University , Cleveland , Ohio 44106 Received 30 Julyl984/Accepted 27 November 1984 Regulation by oxygen of the peptidase T ( pepT ) locus of Salmonella typhimurium was studied by measuring j8-galactosidase levels in strains containing apepT : : Mu dl ( Apr lac ) operon fusion .
I-Galactosidase was induced in anaerobic cultures and late-exponential and stationary-phase aerated cultures .
Peptidase T activity also was induced under these growth-conditions .
pepT + but not pepT strains will utilize as amino-acid sources the tripeptides Leu-Leu-Leu and Leu-Gly-Gly only when grown anaerobically .
Mutations at two loci , oxrA and oxrB ( oxygen regulation ) prevent induction of the pepT locus .
The oxrA locus is homologous to thefnr locus of Escherichia coli .
We have isolated 12 independent Mu dl insertions ( oxd : : Mu dl , oxygen dependent ) that show induction of,-galactosidase in anaerobic cultures and stationary-phase aerated cultures .
These insertions fall into nine classes based on map location .
All of the oxd : : Mu dl insertions are regulated by oxrA and oxrB and therefore define a global regulon that responds to oxygen limitation .
Peptidase T of Salmonella typhimurium is an aminotripeptidase that removes the N-terminal amino-acid from a variety of tripeptides ( 28 ) .
The enzyme was first identified as a band of hydrolytic activity towards tripeptides seen after native polyacrylamide gel electrophoresis of cell extracts and staining for peptidase activity ( 18 ) .
Mutants lacking peptidase T ( pepT ) were obtained by screening extracts of mutagenized colonies for those unable to hydrolyze the peptides Met-Ala-Ser and Met-Gly-Gly ( 28 ) .
This screen was carried out with a strain of S. typhimurium lacking several aminopeptidases ( peptidases N , A , and B ) which also are capable of hydro-lyzing tripeptides .
Peptidase T can not be specifically assayed in extracts of wild-type strains because of the tripept-idase activity of these other enzymes ( 18 ) .
The ability of peptidase T to allow-growth of S. typhimur-ium on peptides as amino-acid sources was tested in strains lacking peptidases N , A , and B. Compared with pepT + strains , the pepT mutants are not deficient in using as amino-acid sources any of the peptides which the enzyme hydro-lyzes in-vitro ( 28 ) .
Peptidase T is the only S. typhimurium peptidase studied which does not permit growth on at least one peptide substrate ( 17 , 18 , 29 , 30 ) .
The inability of peptidase T to allow-growth on tripeptides results from insufficient peptidase activity in the wild type .
Mutants ( pto , peptidase T overproducer ) that overproduce peptidase T are able to use peptidase T substrates as amino-acid sources ( 27 ) .
The pto mutations are tightly linked to pepT , are cis dominant , and increase 3-galactosidase expression from a pepT7 : : Mu dl ( Apr lac ) operon fusion ( pepT : : lac ) ( 2 , 27 ) .
The occurrence of mutants that overproduce peptidase T suggested that the enzyme might be regulated .
To identify a regulatory pattern for peptidase T we have used the pepT : : lac fusion to search for conditions that affect P-galactosidase levels .
In this communication we report that peptidase T levels are regulated by oxygen .
In addition , we describe the identification of other genes regulated by oxygen and two regulatory loci which control all of these genes .
MATERIALS AND METHODS Bacterial strains .
Strains used in this work are derivatives of S. typhimurium LT2 or Escherichia coli K-12 .
Strains used repeatedly are listed in Table 1 .
Media and growth-conditions have been described previously ( 28 ) .
Growth on peptides , electron-acceptors , and nitrogen sources was tested by plating 0.1 ml of an overnight culture in 2.5 ml of soft agar on a minimal plate and placing crystals of the test substrate on the soft agar .
Growth was scored after overnight incubation .
Chlorate resistance was tested by placing a filter paper disk saturated with 3 % KC103 on a lawn of the strain to be tested growing on nutrient agar supplemented with 0.2 % glucose .
Sensitive strains showed a zone of growth inhibition after overnight anaerobic incubation .
Culture growth was meas-ured by determining the optical density at 600 nm ( OD ' ) on a Gilford spectrophotometer .
For anaerobic-growth , liquid cultures either were grown in filled volumetric flasks with the medium overlaid with mineral oil or were sparged with 5 % C02-95 % N2 , and plates were incubated in Brewer jars under an atmosphere of 5 % C02-95 % N2 .
Cultures for P-galactosidase assays were grown in 50 ml of E medium plus glucose ( 0.4 % ) in a 125-ml screw cap Erlenmeyer flask incubated in a model G-76 Gyrotory shaker ( New Brunswick Scientific Co. , Inc. ) at a speed setting of 5 .
Mutagenesis with diethyl sulfate ( Sigma Chemical Co. ) was performed as described by Roth ( 24 ) .
Mutagenesis with Mu dl ( Apr lac cts62 ) was performed as described previously ( 6 , 27 ) .
Transduction with P22 ( HT 12/4 int-3 ) was performed by the method of Roth ( 24 ) .
To transduce pepT7 : : Mu dl ( Apr lac cts62 X ) into recipient strains , ca. 108 cells were infected at a multiplicity of infection of 10 PFU per cell and plated to give ca. 100 Ampr colonies .
The X mutation in the Mu dl prophage immobilizes it and allows it to be transduced into other strains without zygotic induction ( 1 , 28 ) .
For transduction to antibiotic resistance , the phage and bacteria were mixed and preincub-ated for at least 30 min before plating on selective medium .
Hfr crosses were performed by broth mating as described by Sanderson et al. ( 25 ) .
F ' episomes were transferred from strain to strain by plate matings as described by Sanderson et al. ( 25 ) .
Mapping of Mu dl insertions was performed by transduction to Tetr with phage lysates grown on various strains , each containing a TnJO insertion at a known site on the chromosome as described by Maurer et al. ( 16 ) .
The donor strains were a collection of ca. 50 strains , each carrying a TnJO insertion at a different locus .
Approximately 5 x 108 67 TN2063 leuBCD485 pepN90 pepA16 pepBlI pepPI pepQl leuBCD485 pep77 : : Mu dl ( Apr lac cts62 X ) pep77 : : Mu dl ( Apr lac cts62 X ) pto-2 zce-850 : : TnJO pep77 : : Mu dl ( Apr lac cts62 X ) zce-850 : : Tnl O pepT7 : : Mu dl ( Apr lac cts62 X ) pto-13 pep77 : : Mu dl ( Apr lac cts62 X ) oxrAl pep77 : : Mu dl ( Apr lac cts62 X ) oxrA + zda-888 : : TnIO pep77 : : Mu dl ( Apr lac cts62 X ) oxrAl zda-888 : : TnJO pep77 : : Mu dl ( Apr lac cts62 X ) leuBCD485 pepT7 : : Mu dl ( Apr lac cts62 X ) leuBCD485 oxrAl zda-888 : : TnlO pepT7 : : Mu dl ( Apr lac cts62 X ) leuBCD485 oxrA2 : : TnlO pep77 : : Mu dl ( Apr lac cts62 X ) leuBCD485 oxrB8 zxx-895 : : Tn5 zce-850 : : TnJO 75 % a The insertion is cotransducible with pepT ( 28 ) .
The insertion zda-888 : : Tn ) O is 15 % cotransducible with oxrA ( see the text ) .
The insertion zxx-895 : : TnS is 45 % cotransducible with oxrB8 .
TN1304 TN1379 TN1668 TN1672 TN1686 TN1893 TN 1895 TNI909 TN1910 TN1989 TN2062 TN2064 recipient cells were spread on a nutrient agar tetracycline plate , and drops of phage lysates were spotted on this lawn .
The plates were incubated at 30 °C for 4 h and then at 42 °C overnight .
Transductants which lost the Mu dl prophage became temperature resistant and formed colonies on the selective plates .
The results of these crosses were confirmed by single transductional crosses , performed by selecting Tetr and scoring for the Lac and Amp phenotypes conferred by the Mu dl prophage .
Insertions of a TnJO derivative TnJOA16A17 ( 7 ) near Mu dl fusions were identified in a similar way by using P22 transducing lysates grown on a set of ca. 300 strains carrying random insertions of this transposable element ( A. Kukral and R. Maurer , manuscript in preparation ) .
3-Galactosidase was measured by the method of Miller ( 20 ) .
Hydrolysis of peptides by crude soluble extracts ( 28 ) was measured by high-pressure liquid chromatography of trinitrophenyl derivatives to analyze reaction products ( T. H. Carter , Ph.D. thesis , Case Western Reserve University , Cleveland , Ohio , 1982 ) .
Met-Gly-Gly was obtained from Bachem .
N-Acetyl-Ala3 was from Sigma Chemical Co. .
The protein concentration was determined by the method of Lowry et al. ( 15 ) , with bovine serum albumin as the standard .
A. B. 300 12701 150 0 10 o 1.0 0 o 10 4l ( 0 0 50 0.1 n l 0 RESULTS Regulation of peptidase T .
An operon fusion of the lac structural genes to the pepT locus ( pepT : : lac ) was used to search for conditions that regulate synthesis of peptidase T. Strain TN1668 ( pepT : : Iac ) was grown in various mnedia , and 3-galactosidase activity was measured .
We found no evidence that,-galactosidase levels were significantly affected by growth on alternative carbon sources or nitrogen sources , by starvation for carbon or nitrogen , or by changes in growth-rate ( data not shown ) .
However,,-galactosidase levels were dependent on the growth phase of the cultures from which cells were harvested .
Stationary-phase cells showed significantly higher enzyme levels than did exponential cells .
The levels of f-galactosidase throughout the growth cycle therefore were measured ( Fig. 1 ) .
During exponential-growth in minimal glucose medium , the P-galactosidase activity remained constant at ca. 25 U .
During the last few doublings before the culture entered stationary-phase , the specific activity increased sixfold to ca. 150 U ( Fig. 1 ) .
In strain TN1672 which contains the pto-2 mutation , , B-galactosidase remained constant throughout the growth cycle at ca. 300 U ( Fig. 1 ) .
Strain TN1668 ( pepT : l : ac ) showed increased expression of p-galactosidase in stationary-phase when grown in minimal-medium containing either glucose or glycerol as the carbon source .
To determine whether peptidase T production is regulated in this fashion as well , extracts of strain TN1304 ( leu pepN pepA pepB pepP pepQ ) were prepared from cultures at several optical densities , and the specific activity of peptidase T was assayed .
The specific activity of peptidase T increased 15-fold between the exponential phase and the stationary-phase ( Table 2 ) .
The specific activity of another peptidase , dipeptidyl , carboxypeptidase ( 30 ) , was assayed in the same extracts .
The levels of dipeptidyl carboxypeptidase did not differ significantly in exponential-and stationary-phase cells ( Table 2 ) .
Effect of nutrient and 02 limitation on pepT expression .
It seemed possible that limitation for some nutrient might be the signal for increasing peptidase T levels late in exponential phase .
Therefore , pepT expression was measured in cultures containing limiting concentrations of nutrients .
Cultures with limiting glucose , ammonia , or leucine did not show elevated levels of p-galactosidase expressed from the pepT promoter before or after cessation of growth ( Fig. 2 ; data not shown ) .
Because the 02 supply is frequently cited as an important factor limiting growth , we tested the effect of anaerobiosis on expression of-galactosidase front the pepT : : lac fusion .
Exponentially growing cells from an-aerobic cultures contained P-galactosidase levels comparable to those of statidhary-phase cells from aerobic cultures ( ca. 400 U ) .
It therefore is likely that oxygen limitation .0 .
1-Galactosidase levels during the bacterial growth cycle .
Strains were grown in E medium plus 0.4 % glucose .
Samples for 3-galactosidase assay were removed at 0.5-h intervals and kept on ice until assayed by the method of Miller ( 20 ) .
( A ) TN1668 , pto + pepT : : lac ; ( B ) TN1672 , pto-2 pepT : : lac during late-exponential phase causes increased pepT expression in aerobic cultures .
50 0.1 n l 0 .01 ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ I I. 0 p The growth properties of pepT + and pepT strains also reflect regulation of peptidase T by oxygen levels .
When grown anaerobically , pepT + strains are able to utilize peptide substrates of peptidase T as amino-acid sources .
These tripeptides did not serve as amino-acid sources during aerobic-growth ( Table 3 ) .
Only growth on tripeptides , not dipeptides or tetrapeptides , was affected by anaerobiosis .
This suggests that the cell does not contain a group of anaerobic peptidases induced by 02 limitation .
An E. coli peptidase mutant , CM89 ( 19 ) , shows the same effect of anaerobiosis on peptide utilization , indicating that E. coli contains a tripeptidase regulated in the same manner as peptidase T of S. typhimurium .
Regulatory loci for pepT .
We have sought mutations that alter regulation of pepT .
We expected that such mutants might be detectable by screening mutagenized cultures of the pepT : : lac fusion strain for colonies with an altered appearance on lactose indicator plates .
When grown on MacConkey-lactose agar , colonies of the pepT : : lac strain TN1668 had dark-red centers and light peripheries ( fisheye colonies ) .
Strains carrying a pto mutation formed uniformly red colonies .
This suggested that cells in the center of the wild-type colony are expressing relatively high 1Bgalactosidase levels ( as are cells from a stationary-phase liquid culture or an anaerobic culture ) , whereas cells on the periphery of the colony contain relatively low 1Bgalactosidase levels ( as do cells from aerated , exponential-ly-growing liquid cultures ) .
We expected that mutants with altered regulation of the pepT locus could be detected on MacConkey agar as colonies that are uniformly red ( indicating high , B-galactosidase levels in all cells ) or that lack the red center ( indicating low , B-galactosidase levels in all cells ) .
Such strains might contain mutations linked to the pepT locus ( e.g. , pto or lac mutations ) or mutations unlinked to pepT .
Unlinked mutations might represent alterations in trans-acting factors that control transcription of pepT .
To isolate mutations affecting pepT : : lac expression , strain TN1668 was mutagenized with diethyl sulfate and plated on MacConkey agar .
Colonies that differed in appearance from the parent were picked and purified .
Mutant phenotypes included uniformly red colonies , colonies with darker , lighter , or smaller red centers , pink colonies , and white 0 0 0 0.51 -- I 0f , 50 c cn a C ~ ~ ~ 40 w 30 4 20 10 ( 4-J 3 4 5 6 7 8 4 TIME ( hrs ) FIG. 2 .
13-Galactosidase levels in glucose-limited culture .
Strain TN1668 ( pto + pepT : : lac ) was grown in NCE mediut plus 0.04 % glucose .
Samples were removed at 1-h intervals and kept on ice until assayed for 13-galactosidase by the method of Miller ( 20 ) .
Few completely white colonies were picked and characterized because they were expected to be lacZ or lac Y mutants .
From ca. 30,000 colonies screened , 125 mutants were tested for the presence of mutations that were unlinked to pepT .
This was done by transducing each of these strains to Tetr on MacConkey-lactose-tetracycline agar with phage grown on a strain ( TN1686 ) containing a TnlO insertion 75 % linked to pepT : : lac .
If a recipient strain carries a mutation linked to pepT , some of the transductants from this cross should regain the wild-type ( fisheye ) colony appearance .
For strains with mutations unlinked to pepT , all of the transductants should retain the mutant phenotype .
To verify that these strains contained mutations unlinked to pepT , each strain was used as a donor in a transduction cross with LT2 ( wild type , no lac fusion ) as recipient , and Ampr transduct-ants were selected .
This cross ( mutant donor x wild-type recipient ) should move the pepT : : lac fusion into a clean genetic background and separate it from any unlinked mutations that affect expression of , B-galactosidase .
Mutations linked to pepT should give a mixture of wild-type and mutant phenotypes reflecting coinheritance of pepT : : lac and the linked mutation .
Unlinked mutations should give only wildtype ( fisheye ) colonies because the pepT : : lac fusion and the mutation altering expression of 3-galactosidase can not be cotransduced .
Four of the strains tested formed uniformly red colonies on MacConkey-lactose medium .
All of these strains had mutations linked to the pepT locus .
Mapping data for one of these mutants , TN1893 , is shown in Table 4 .
These strains probably carry pto mutations .
No unlinked mutation was found that elevates , B-galactosidase production .
One mutant ( TN1895 ) which had decreased 1-galactosidase levels ( pink colony ) carried a mutation unlinked to pepT ( Table 4 ) .
We have called the mutation in this strain oxrAl ( oxygen regulation ) .
TnJO and TnS insertions linked to oxrA were isolated .
In transduction crosses with these insertions the oxrAl allele behaves as a mutation at a single locus ( Table 4 ) .
A TnJO insertion into the oxrA locus also has been isolated .
strain In another attempt to isolate regulatory mutants , a with a chromosomal duplication of pepT7 : : Mu dl ( unpublished data ) was used to reduce the occurrence of lacZ and lacY mutations .
The strain was mutagenized and screened on MacConkey agar for pink or white colonies .
Mutants were tested by transduction to determine whether they contained mutations linked to pepT or oxrA .
In this experi ment another oxrA allele and a mutation at another locus , oxrB , were isolated .
TnS insertions linked to oxrB have been isolated , and in crosses using this TnS insertion the oxrB8 allele also behaves as a mutation at a single locus .
The effect of oxr mutations on P-galactosidase expressed from the pepT promoter was measured .
The oxrA and oxrB mutants failed to induce,-galactosidase as the cultures entered stationary-phase or during anaerobic-growth ( Table 5 ) .
To map the oxrA locus we used a TnlO insertion ( zda-888 : : TnJO ) 15 % linked to oxrA .
Hfr strains with origins of transfer near oxrA were constructed with F ' lac zzfr : TnlO episomes as described by Chumley et al. ( 5 ) .
These Hfr 's have an origin of transfer at the site of the chromosomal TnJO insertion near oxrA .
With Hfr 's constructed at the TnJO near oxrA , the only region of the map at which a discontinuity in the gradient of transfer ( determined by the number of recombinants formed with particular auxotrophs ) occurred was the interval between pyrC ( 22 map units ) and trp ( 34 map units ) ( Table 6 ) .
This indicates that the origin of the Hfr 's lies between pyrC and trp .
To further define the position of oxrA , the TnlO insertion near oxrA was introduced into an Hfr with its origin at 0 map units and a clockwise direction of transfer .
This strain TN2015 ( HfrK4 serAl3 rfa-3058 zda-888 : : TnlO ) was mated to auxo-trophic mutants .
Prototrophic recombinants were selected and scored for Tetr .
The linkage of the TnWO insertion to auxotrophic markers was : proAB25 , 7 map units , 6 % ( 3/50 ) ; pyrD13 , 21 map units , 14 % ( 7/50 ) ; trp43 , 34 map units , 84 % ( 42/50 ) ; his-644 , 42 map units , 54 % ( 27/50 ) ; and argH88 , 88 map units , 2 % ( 1/50 ) .
The high linkage to trp suggests that oxrA maps near the trp end of this interval of the Salmonella map .
The oxrAl mutation was not cotransducible by P22 with purBJ3 , pyrF696 : : TnJO , trp-2451 : : TnJO , or the following transposon insertions : zce-862 : : TnJO , zce-864 : : TnJO , and zce-859 : : TnS ( linked to purB ; 27 ) ; zcf-845 : : TnJO and zcf ¬ ~ ~ ~ a * * ~ 1.0 ; 0 0 0 0.1 o ~ ~ ~ ~ ~ ~ ~ ~ Vf .
Peptidase T and dipeptidyl carboxypeptidase levelsa Dipeptidyl carboxypeptidase hceydrlyAia3 hdrmoloyfepaise ( > kmol of alanine per min/mg ) 0.12 0.048 ( 1.0 ) 0.082 ( 1.0 ) 0.44 0.117 ( 2.4 ) 0.103 ( 1.25 ) 1.02 0.743 ( 15.5 ) 0.094 ( 1.15 ) a Extracts were prepared from 1-liter cultures grown in E medium plus 0.4 % glucose .
The extract for testing cells in exponential-growth was made from two pooled 1-liter cultures with optical densitities of 0.089 and 0.15 ( averaged to 0.12 ) .
Met-Gly-Gly hydrolysis was assayed in 50 mM Tris-hydrochloride ( pH 7.5 ) containing 10 mM Met-Gly-Gly and 1 mM MnCI .
N-Acetyl-Ala3 hydrolysis was assayed in 0.1 M sodium barbital ( pH 8.1 ) containing 10 mM N-acetyl-Ala3 and 0.1 mM CoCl2 .
The values within parentheses measure relative activity .
b The extracts used for these assays contain a dipeptidase ( peptidase D ) hydrolysis of N-acetyl-Ala-Ala-Ala by which converts Ala-Ala ( the product of dipeptidyl carboxypeptidase ) to Ala. .
Small amounts of Ala-Ala ( less than 5 % of the Ala levels ) were detected in the reaction products .
Ppia T Met-Gly-Gly hydrolysis ( p.mol of methionine per min/mg ) ODwo of culture + + CM89 ( pepT + ) a TN1298 , leu48S pepN90 pepAI6 pepBII pepPI pepQI supQ309A ( proAB pepD ) ; TN1420 , leu485 pepN90 pepA16 pepBlI pepPI pepQI supQ302A ( proAB pepD ) ; CM89 , E. coli K-12 leu-9 Apro-lac met thyA pepNI02 pepAII pepBl pepQlO .
847 : : TnS ( linked to dcp ; 30 ) ; and zcd-6 : : TnJO , zcd-7 : : TnJO , zcd-8 : : TnJO , zcd-JO : : TnJO , and zcd-2 : : TnJO ( linked to pncA , gdhA , or nit ; 23 ) .
Identification of other genes regulated by anaerobiosis .
We wondered whether other genes might show a pattern of regulation similar to that of pepT .
In an attempt to identify such genes we screened independent Mu dl ( Apr lac ) insertion strains for those which formed fisheye colonies on MacConkey-lactose medium .
As noted above , the colony appearance of the pepT : : lac strain ( red center , light edge ) and of mutants with altered pepT : : lac expression ( pto , uniformly red ; oxrA , uniformly pink ) suggests that colony appearance on MacConkey-lactose agar is a reflection of the pepT regulatory pattern .
If this is correct , strains containing lac fusions to other genes regulated in the same way might form similar colonies on MacConkey-lactose agar .
In one experiment ca. 800 Ampr insertion mutants of TN1379 ( leuBCD485 ) were selected on MacConkey lactose-ampicillin agar .
Most of the colonies showing any red color seemed to be darker in the center of the colony .
Consequently , the choice of colonies that most ' closely resembled the pepT : : lac strain was somewhat subjective .
' Colonies were chosen for study that had a light edge and a red center , a sharp boundary between these two areas of the colony , and no precipitation of bile-salts around isolated colonies .
These strains were assayed for P-galactosidase during exponential and stationary-phase ( Table 7 ) .
Of 12 strains with the fisheye appearance , 4 showed an induction ratio ( stationary-phase/exponential phase ) of at least 6 .
All four isolates that showed a high induction ratio were unable to grow on minimal lactose medium ( Table 7 ) .
Of the strains that did not show significant induction of 13-galactosidase , only one was unable to use lactose as the sole carbon source .
The Lac phenotype presumably reflects the low levels of Pgalactosidase during exponential-growth .
All of the Lacstrains had less than 15 U of activity in exponential phase .
Leu-Leu-Leu Leu-Leu-Leu-Leu Leu + + + TN1298 ( pepT + ) + + + + TN1420 ( pepT ) + + CM89 ( pepT + ) a TN1298 , leu48S pepN90 pepAI6 pepBII pepPI pepQI supQ309A ( proAB pepD ) ; TN1420 , leu485 pepN90 pepA16 pepBlI pepPI pepQI supQ302A ( proAB pepD ) ; CM89 , E. coli K-12 leu-9 Apro-lac met thyA pepNI02 pepAII pepBl pepQlO .
Transductional mapping of mutations affecting P-galactosidase expression from pepT7 : : Mu dl ( Apr lac ) a Colony types on MacConkey agar Donor ( genotype , phenotype ) Recipient Selection Fisheye Red Pink TN1668 ( pepT : : lac , fisheye ) LT2 Ampr 37/37 0 0 TN1893 ( pepT : : lac pto-13 , red ) LT2 Ampr 9/40 31/40 0 TN1895 ( pepT : : Iac oxrAl , pink ) LT2 Ampr 40/40 0 0 TN1686 ( pepT : : Iac zce-850 : : TnIO , fisheye ) TN1895 Tetr 0 0-100/100 TN1909 ( pepT : : lac zda-888 : : TnJO , fisheye ) TN1895 Tetr 7/60 0-53/60 TN1910 ( pepT : : lac oxrAl zda-888 : : TnlO , pink ) TN1668 Tetr 207/238 0-31/238 a Selection for Ampr requires inheritance of pepT : : Iac .
The insertion zce-850 : : TnIO encodes Tetr and is 75 % cotransducible with pepT .
The insertion zda-888 : : TnlO was isolated on the basis of its linkage to oxrAl The pepT : : lac strain ( ca. 30 U in exponential phase ) shows only weak growth on minimal lactose medium ( 27 ) .
Effect of the oxrAl mutation .
To test whether these fusions are regulated by oxrA , the oxrAl mutation was introduced by cotransduction with a linked TnJO insertion .
Phage grown on TN1910 ( oxrAl zda-888 : : TnJO ; 15 % linked ) was used to transduce each of the lac fusion strains to Tetr on MacConkey lactose-tetracycline and the transductants agar , were scored for colony type .
Some strains showed only the fisheye colony type , similar to the recipient strains in the cross .
Other strains showed two types : fisheye colonies resembling the recipient strain and pink colonies virtually indistinguishable from pepT : : lac oxrAl strains .
All four strains with high induction ratios ( > 6 ) for , B-galactosidase gave pink colonies after introduction of the oxrAl mutation ( Table 7 ) .
One strain ( TN1932 ) showed no induction of Pgalactosidase but did give pink transductants after introduction of the oxrAl mutation ( Table 7 ) .
Because induction ratios were measured in minimal glucose medium but colony appearance was scored on MacConkey agar , a rich-medium , it seemed possible that the gene carrying the Mu dl insertion in TN1932 might show a high induction ratio only when grown in a rich-medium .
To test this , 3-galactosidase levels were measured in cells grown in nutrient broth .
TN1932 grown in this medium had 3.6 U of 3-galactosidase during exponential phase and 221 U during stationary-phase .
Therefore , all strains showing high induction ratios ( > 6 ) respond to the oxrA mutation , and no strain with a low induction ratio showed any oxrA effect .
All strains with high induction ratios also gave pink transductants after introduction of the oxrB8 mutation .
Because the criteria for choosing strains with colony appearances similar to pepT : : lac strains were somewhat subjective , the screening process may have missed some lac fusion strains that induce P-galactosidase in stationary-phase .
In a second experiment all Mu dl insertion mutants of TN1379 that showed any red color ( from pink to deep red ) on MacConkey agar were transduced to Tetr with phage on the oxrAl zda-888 : : TnJO strain ( TN1910 ) .
Of 887 Ampr colonies , 53 ( ca. 5 % ) expressed enough P-galactosidase to be detected on MacConkey plates .
All 53 strains that expressed P-galactosidase were tested for growth on minimal lactose medium and for the effect of introducing the oxrAl mutation ( formation of pink colonies ) .
Four strains ( TN1942 , TN1943 , TN1944 , and TN1945 ) were found that expressed less Pgalactosidase when the oxrAl mutation was introduced into the strain .
All of these strains had the fisheye phenotype and showed no growth or only marginal growth on minimal lactose medium .
The 3-galactosidase levels of these strains in exponential and stationary-phase were measured ( Table 8 ) .
All four strains showed high induction ratios .
These four strains also gave pink colonies on MacConkey-lactose agar after introduction of the oxrB8 mutation .
The P-galactosidase levels of strains with oxrA controlled lac fusions were measured during anaerobic-growth ( Table 9 ) .
All strains tested showed,-galactosidase levels comparable to or higher than those of stationary-phase aerobic cultures .
The loci to which lac is fused in these strains therefore have been designated oxd ( oxygen dependent ) .
Mapping of oxd loci .
The oxd : : Mu dl ( Apr lac ) insertions were tested for cotransduction with a set of ca. 50 different TnJO insertions at known sites on the chromosome .
The map positions of two oxd : : Mu dl insertions were identified ( Table TABLE 6 .
Mapping of oxrA by conjugation crosses with Hfr 's with transfer origins near oxrA No .
of colonies when mated with : '' Donor TT1333 TT459 ( trp : : TnIO ) 340 ( pvrC : : Tn/0 ) 20 TN1999 ( leuBCD485 zda-888 : : TnlO1 F'ts 114 lac + zzf-20 : : TnIO ) TN2000 ( leuBCD485 zxx-888 : : Tn1l0 F'ts 114 lac + zzf-21 : : TnlO ) TN2001 ( leuBCD485 zxx-888 : : TnlO ) / F'ts 114 lac + zzf-22 : : TnlO ) a TT1333 and TT459 were obtained from J. Roth .
Characterization of Mu dl ( Apr lac ) insertion mutants with the fisheye appearance on MacConkey agar P-Galactosidase ( U ) ' NCE-oxrAI Fold in-lactose efExponential Stationary crease growth '' fect ' Isolate 4 ( TN1932 ) 5.5 5.0 1 - + 8 ( TN1933 ) 6.0 35 6 - + 9 53 102 2 + -10 50 140 3 + -11 250 255 1 + -12 69 25 < 1 + -13 ( TN1934 ) 6.5 63 10 - + 14-110-100 1 + -15 170 130 < 1 + -16 ( TN1935 ) 13 180 14 - + 17-320-175 < 1 + -18 ( TN1936 ) 7.0 235 34 - + l19 470 900 2 + - TN1668 ( pepT ) 33 717 22 + / - + a A stationary-phase culture was diluted at least 104 into 100 ml of E medium plus glucose plus leucine late in the afternoon and incubated at 30 °C overnight in a shaking water bath .
The next morning the OD6 , , w of the culture was -0.2 .
The culture was assayed for 3-galactosidase when the optical was and for density s0 .2 ( exponential ) assayed,-galactosidase again on the next day ( stationary ) .
b Growth was tested by streaking strains on NCE agar plus 0.2 % lactose + 0.4 mM leucine .
C Each strain was transduced to Tetr on MacConkey-lactose agar with phage grown on strain TN1910 ( zda-888 : : TnJO oxrAI ) .
+ , Transductants were of two colony types , one fisheye and one pink ; - transductants resembled the recipient strain ( fisheye ) .
d This strain was uniformly red on MacConkey plates TABLE 5 .
Effect of regulatory mutations on pepT : : Iac expression during stationary-phase and anaerobic-growth Strain Cultuire 136w 3-Galactosidase condition OD ,0 ( U ) TN1989 oxrA + Aeration 0.187 23 Aeration 2.39 156 C02/N , 0.099 381 TN2062 oxrAI Aeration 0.180 24 Aeration 2.98 39 CO2/N2 0.153 30 TN2063 oxrA2 : : TnlO Aeration 0.292 25 Aeration 3.59 23 C02/N2 0.148 30 TN2064 oxrB8 Aeration 0.276 25 Aeration 3.49 26 C02/N2 0.099 50 10 ) .
Strain TN1935 contains the mutation oxdAS : : Mu dl , which is linked to metC ( 64 map units ) , and strain TN1942 contains the mutation oxdB7 : : Mu dl , which is linked to mel ( 93 map units ) .
All other oxd : : Mu dl insertions are not cotransducible with pepT , mel , or metC .
To gain better information about how many different loci are represented among our fusions , we tested all of them for linkage to a group of ca. 300 random insertions ofTnJOA16A17 isolated by Kukral and Maurer ( manuscript in preparation ) .
TnJOA16A17 insertions linked to all but one oxd fusion ( oxd-7 ) were found .
The oxd-3 , oxd-9 , and oxd-10 fusions were linked to the same TnJOA16A17 insertions with approximately the same linkage frequency .
All other oxd fusions were linked to unique TnJOA16A17 insertions .
Growth on anaerobic respiratory substrates .
When grown anaerobically , E. coli induces several proteins that allow-growth on nonfermentative carbon sources ( e.g. , glycerol ) with electron-acceptors other than oxygen ( e.g. , nitrate , fumarate , or trimethylamine N-oxide ) .
A regulatory locus , fnr ( also called nirA or nirR ) , which prevents induction of these proteins has been described previously ( 3 , 13 , 21 ) .
The fnr locus maps near trp ( 3 ) , in approximately the same region as oxrA .
We therefore tested the oxr and oxd mutants for anaerobic-growth on these electron-acceptors .
The oxrA mutants were unable to use nitrate as an electron-acceptor or a nitrogen source but could use fumarate and trimethylamine N-oxide as electron-acceptors and nitrite as a nitrogen source .
The oxrB8 mutant utilized all these compounds .
One oxd mutant , TN1943 ( oxd-8 : : Mu dl ) was unable to use nitrate or trimethylamine N-oxide as electron-acceptors .
This phenotype suggested that the oxd-8 mutation might TABLE 9 .
f3-Galactosidase levels during anaerobic-growth ' Strain 3-Galactosidase ( U ) TN1668 ... 773 TN1933 ... 185 TN1934 ... 53 TN1935 ... 213 TN1936 ... 408 TN1942 ... 270 TN1943 ... 318 TN1944 ... 233 TN1945 ... 85 a Cells ` were harvested from cultures growing exponentially ( OD6o < 0.38 ) in E medium plus 0.4 % glucose plus 0.4 mM leucine , overlaid with mineral oil .
affect one of the chl loci .
Mutations at these loci confer chlorate resistance because they are pleiotropically deficient in several molybdo-enzymes which reduce chlorate to the toxic compound chlorite .
All of the oxd fusions therefore were tested for chlorate resistance .
Only oxd-8 : : Mu dl was chlorate resistant .
In a transduction cross with a TnJOAM6A17 linked to oxd-8 as donor and oxd-8 : : Mu dl , all of the Lac-ampicillin-sensitive recombinants regained chlorate sensitivity .
The chlorate-resistant phenotype therefore results from the oxd mutation and not from a secondary mutation .
Straina DISCUSSION The overall response of cells to several major physiological stresses or changes has been studied genetically with the Mu dl ( Apr lac ) vector to identify genes induced by DNA damage ( 10 ) or phosphate-starvation ( 31 ) and biochemically by two-dimension gel electrophoresis to identify polypeptides whose synthesis is regulated by heat-shock ( 8 , 14 ) or growth-rate ( 22 ) .
These phenomena have been termed global regulation .
We believe that the group of genes ( oxd ) defined by our Mu dl ( Apr lac ) insertions comprise another global regulon characterized by high levels of gene products in anaerobic cultures or in stationary-phase aerobic cultures Expression of these genes is reduced by mutations at the oxrA and oxrB loci .
Mu dl insertions into these genes occur at a frequency of 1 in 200 among all Mu dl lysogens .
Based on map position , 10 different loci are represented among the 12 Mu dl fusions .
Only one class contains more than one member , so it is likely that several other anaerobically induced loci remain undiscovered .
By using two-dimen-sional gels , Smitha and Neidhardt ( 26 ) observed in E. coli 18 polypeptides that are present at higher steady-state levels in anaerobically grown cells .
Most of these polypeptides showed 2-to 5-fold increases in anaerobiosis , whereas our anaerobic/aerobic ratios were considerably higher ( 5-to 60-fold ) .
It is possible that the products of the oxd loci defined by our Mu dl insertions are not resolved by the two-dimensional gel system used by Smith and Neidhardt .
However , a more likely explanation is that their aerobic reference culture , grown to an OD420 of 1.0 , was already partially induced for these gene products .
We find that an OD420 of 1.0 corresponds to an OD600 of 0.53 .
At this cell density peptidase T levels have reached 50 % of the maximum under our culture conditions .
The induction ratio for P-galactosidase expressed from the pepT promotor shows some variability in different experiments ( Fig. 1 and Tables 5 , 7 , and 8 ) .
Most of this variability is probably due to variations in the rates of aeration as cultures enter stationary-phase .
We have found that the most reproducible measure of the induction ratio is anaerobic exponential phase/aerobic exponential phase .
This can be calculated as 381 U/23 U = 16.7 for pepT ( Table 5 ) .
This proves to be the best method for determining induction ratios of the oxd loci as well .
The pepT gene is the only one of this group of genes for which a gene product is known .
The physiological significance of this mode of regulation for an aminotripeptidase is not clear .
Perhaps tripeptides are present at high levels in some natural anaerobic environment , and peptidase T is induced to degrade them .
Clearly the levels of peptidase T are sufficiently high in cells growing anaerobically to allow the use of some tripeptides that can not be used by aerobically growing peptidase-deficient cells .
An alternative possibility is that peptidase T plays some role in regulating the response to anaerobiosis , perhaps by processing a peptide signal-molecule .
The observation that pepT mutants show normal regulation of the oxd : : Mu dl ( Ampr lac ) fusions ( K. Strauch , unpublished data ) argues against this possibility .
Because a multiply-peptidase-deficient E. coli K-12 strain uses certain tripeptides only when grown anaerobically , this organism seems to contain a similarly regulated tripeptidase , presumably tripeptidase TP ( 9 ) .
Regulation of tripeptidase activity by oxygen levels therefore is conserved in E. coli and S. typhimurium .
Some proteins have been observed previously to be regulated much like peptidase T. Under conditions of oxygenlimited growth and in stationary-phase , cytochromes b558 , a1 , and d are present in higher levels than in well-aerated , exponentially growing cells ( 10 ) .
Other respiratory proteins involved in electron transport to acceptors other than 02 also are expressed during anaerobiosis .
We have considered the possibility that our regulatory locus , oxrA , corresponds to an E. coli locus that pleiotropically affects several an-aerobic respiratory pathways .
Mutations at this locus ( variously called fnr , nirA , or nirR ) prevent the expression of several anaerobic respiratory proteins ( fumarate reductase , nitrate reductase , nitrite reductase , trimethylamine N-oxide reductase , and formate hydrogenlyase ) .
These E. coli mutants have lost the ability to use nitrate or fumarate as electron-acceptors or nitrite as a nitrogen source ( 13 , 21 ) .
The fnr mutations map at 29.5 min on the E. coli map , in approximately the same region as oxrA in S. typhimurium .
Strains carrying oxrA mutations can not use nitrate as an electron but do acceptor or as a nitrogen source , they use fumarate as an electron-acceptor and nitrite as a nitrogen source .
( The oxrB mutant is not deficient in utilizing any of these substrates .
) Although these phenotypes are clearly different from that of E. colifnr mutants , it is possible that S. typhimurium contains furmarate and nitrite reductase activities that are not repressed by oxygen .
Multiple trimethyl-amine N-oxidase reductases , some of which are constitutive , are present in S. typhimurium ( 12 ) .
Higgins and co-workers ( 9a ) have shown that the E. coli fnr + gene can complement oxrAl in S. typhimurium for expression of , B-galactosidase from the pep T : : lac fusion , indicating that oxrA and fnr are homologous loci .
This locus must have a much broader regulatory significance than previously recognized , including the regulation of at least one gene ( pepT ) having no obvious role in respiration .
Casadaban , M. J. , and S. N. Cohen .
Lactose genes fused to exogenous promoters in one step using a Mu-lac bacteriophage : in-vivo probe for transcriptional control sequences .
Chippaux , M. , V. Bonnefoy-Orth , J. Ratouchniak , and M.-C .
Operon fusions in the nitrate reductase operon and study of the control gene nirR in Escherichia coli .
Chippaux , M. , D. Giudici , A. Abou-Jaoude , F. Casse , and M. C. Pascal .
A mutation leading to the total lack of nitrite reductase activity in Escherichia coli K12 .
Chumley , F. G. , R. Menzel , and J. R. Roth .
Hfr formation directed by TnlO .
Csonka , L. N. , M. M. Howe , J. L. Ingraham , L. S. Pierson III , and C. L. Turnbough , Jr. 1981 .
Infection of Salmonella typhwith Mu imurium coliphage dl ( Apr lac ) : construction ofpyr : : lac gene fusions .
Foster , T. J. , M. A. Davis , D. E. Roberts , K. Takeshita , and N. Kleckner .
Genetic organization of transposon TnlO .
Herendeen , S. L. , R. A. VanBogelen , and F. C. Neidhardt .
Levels of major proteins of Escherichia coli during-growth at different temperatures .
Tripeptide-specific aminopeptidase from Escherichia coli AJO05 .
Jamieson , D. J. , and C. F. Higgins .
Anaerobic and leucine-dependent expression of a peptide transport gene in Salmonella typhimurium .
Kenyon , C. J. , and G. C. Walker .
Expression of the E. coli uvrA gene is inducible .
Kranz , R. G. , and R. G. Gennis .
Isoelectric focusing and crossed immunoelectrophoresis of heme-proteins in the Esch-erichia coli cytoplasmic membrane .
Kwan , H. S. , and E. L. Barrett .
Roles for menaquinone and the two trimethylamine oxide ( TMAO ) reductases in TMAO respiration in Salmonella typhimurium : Mu d ( Apr lac ) insertion mutations in men and tor .
Lambden , P. R. , and J. R. Guest .
Mutants of Escherichia coli K12 unable to use fumarate as an anaerobic electron-acceptor .
Lemaux , P. G. , S. L. Herendeen , P. L. Bloch , and F. C. Neidhardt .
Transient rates of synthesis of individual polypeptides in E. coli following temperature shifts .
H. , N. J. Rosebrough , A. L. Farr , and R. J. Randall .
Protein measurement with the Folin phenol reagent .
Maurer , R. , B. C. Osmond , E. Shekhtman , A. Wong , and D. Botstein .
Functional interchangeability of DNA replication genes in Salmonella typhimurium and Escherichia coli demonstrated by a general complementation procedure .
McHugh , G. L. , and C. G. Miller .
Isolation and characterization of proline peptidase mutants of Salmonella typhimur-ium .
Miller , G. , Peptidase Salmonella typhimurium .
Miller , C. G. , and G. Schwartz .
Peptidase-deficient mutants of Escherichia coli .
Experiments in molecular genetics .
Cold Spring Harbor Laboratory , Cold Spring Harbor , N.Y. 21 .
Newman , M. , and J. A. Cole .
The chromosomal location and pleiotropic effects of mutations of the nirA ' gene of Escherichia coli K12 : the essential role of nirA + in nitrite reduction and in other anaerobic redox reactions .
Pedersen , S. , P. L. Bloch , S. Reeh , and F. C. Neidhardt .
Patterns of protein synthesis in E. coli : a catalog of the amount of 140 individual proteins at different growth-rates .
Rosenfeld , S. A. , S. M. Dendinger , C. H. Murphy , and J. E. Brenchley .
Genetic characterization of the glutamat dehydrogenase gene ( gdhA ) of Salmonella typhimurium .
Genetic techniques in studies of bacterial metabolism .
Sanderson , K. E. , H. Ross , L. Ziegler , and P. H. Maikelai .
F + , Hfr , and F ' strains of Salmonella typhimurium and Salmo-nella abony .
Smith , M. W. , and F. C. Neidhardt .
Proteins induced by anaerobiosis in Escherichia coli .
Strauch , K. L. , T. H. Carter , and C. G. Miller .
Overproduction of Salmonella typhimurium peptidase T. J. Bacteriol .
Strauch , K. L. , and C. G. Miller .
Isolation and characterization of Salmonella typhimurium mutants lacking a tripeptidase ( peptidase T ) .
Vimr , E. R. , L. Green , and C. G. Miller .
Oligopeptidase-deficient mutants of Salmonella typhimurium .
Vimr , E. R. , and C. G. Miller .
Dipeptidyl carboxypept-idase-deficient mutants of Salmonella typhimurium .
Wanner , B. L. , and R. McSharry .
Phosphate-controlled gene expression in Escherichia coli K12 using Mudl-directed lacZ-fusions .