170 , No. 1 Identification of a Repressor Gene Involved in the Regulation of NAD De Novo Biosynthesis in Salmonella typhimurium NING ZHU , * BALDOMERO M. OLIVERA , AND JOHN R. ROTH Department of Biology , University of Utah , Salt Lake City , Utah 84112 Mutations at the nadi locus affect expression of the first two genes of NAD synthesis , nadA and nadB , which are unlinked .
Genetic data imply that the regulatory effects of nadl mutations are not due to indirect consequences of physiological alterations .
Two types of mutations map in the nadI region .
Common null mutations ( nadi ) show constitutive high-level expression of the nadB and nadA4 genes .
Rare nadl mutations cause constitutive low-level expression of nadB and nadA .
Some nadlP mutations shut off the expression of the biosynthetic genes sufficiently to cause a nicotinic acid auxotrophy .
Spontaneous revertants of auxotrophic nadl mutants have a NadI-phenotype , including some with deletions of the nadI locus .
The nadI locus encodes a repressor protein acting on the unlinked nadA and nadB genes .
NAD ( H ) and NADP ( H ) are the major electron carriers in cellular metabolism .
In the enteric bacteria Salmonella typhimurium and Escherichia coli , NAD also serves as a substrate for DNA ligase ( 20 , 23 ) .
The current understanding of the NAD metabolic pathways in S. typhimurium is summarized in Fig. 1 ( for a review , see reference 11 ) .
The first two enzymes of the de novo pathway are encoded by the nadB gene at 55 min ( 11 ) and the nadA gene at 17 min ( 10 , 19 ) , respectively .
When E. coli is in the of grown presence NAD the cells exhibit decreased capacity for precursors , a quinolinic acid synthesis ( 5 , 6 , 24 ) .
Also , studies with nadA-lacZ and nadB-lacZ operon fusions in S. typhimurium demonstrated transcriptional control of these in genes response to exogenous nicotinic acid ( NA ) ( 7 , 12 , 13 ) .
Regulatory mutants have been isolated which express nad : : lac fusions at a constitutively high level in the presence of a high concentration of NA nicotinamide ( 7 ) .
Three or kinds of obtained which unlinked the mutants were are to nadA and nadB nadI , the serB locus genes ; one , maps near ( 7 ) .
The nadl mutations include ambers and insertions .
Previously , Holley al. ( 13 ) isolated similar that et mutants were designated nadR ; these mutations the map at same position as nadI mutations .
Since the `` nadR '' nomenclature has been used previously describe class of mutations in to a the nadB region ( 27 , 28 ) , the regulatory mutations mapping designated nadI ( 7 ) .
near serB have been causing constitutive high-level expression of Mutations a do necessarily affect the regulatory mechanism of gene not the pool size of relevant mutation that alters that gene .
A a signal-molecule indirectly alter expression .
Some may gene nadB do constitutive mutations for the nadA and genes appear to cause derepression by indirect means .
These mutations impair the biosynthetic genes nadD ( 7 ) and nadE ( K. Hughes , B. M. Olivera , and J. R. Roth , unpublished results ) , which encode NA mononucleotide adenylyltransferase and NAD synthetase , respectively ( 15 ) ( Fig. 1 ) .
Mu-enzymatic activities tations that reduce the level of these cause derepression of the nadA and nadB genes , presumably cell .
by limiting the level of NAD and NADP in the The nadl mutations map in or near a gene , pnuA ( 7 , 13 ) , involved in nicotinamide mononucleotide ( NMN ) transport almost ( Fig. 1 ) ( 10 , 18 ) .
The nadI and pnuA mutations are certainly in the same operon ( 7 ) or may even affect the same gene ( N. Zhu , unpublished data ) .
Genetic evidence is presented that the nadI locus encodes a repressor protein which is directly responsible for regulation of both the nadA and nadB genes .
MATERIALS AND METHODS Bacterial strains .
All strains used in this study and their sources are listed in Table 1 .
All strains used are derived from S. typhimurium LT2 .
MudA refers to a conditional transposition-defective de - ( 1 ) rivative of Casadaban 's original Mu dl ( Lac Ampr ) phage which forms operon fusions ( 16 ) .
Mu dJ refers to a transposition defective `` mini Mu '' bacteriophage , Mu dl-1734 ( Lac Kan ) , constructed by Castilho et al. ( 2 , 3 ) .
This phage is deleted for transposition functions and carries kanamycin resistance .
The E medium of Vogel and Bonner ( 30 ) , supplemented with 0.2 % glucose , was used as minimal-medium .
Difco nutrient broth ( NB ; 8 g/liter ) , with NaCl added ( 5 g/liter ) , was used as rich-medium .
Difco agar was added at a final concentration of 1.5 % for solid medium .
Auxotrophic requirements were included in media at final concentrations described by Davis et al. ( 8 ) , except as otherwise indicated in the text .
Antibiotics were added to media at the following final concentrations : ampicillin ( sodium salt ) , 30 , ug/ml in NB in E and 15 , ug/ml medium ; tetracycline hydrochloride , 20 pug/ml in NB and 10 , ug/ml in E medium ; kanamycin sulfate , 50 in NB and 125 in E medium .
All antibiotics , ug/ml , ug/ml were obtained from Sigma Chemical Co. .
Media containwere always prepared fresh before use .
5-ing ampicillin Bromo-4-chloro-3-indolyl-3-D-galactoside ( X-Gal ) dissolved in N,N-dimethylformamide ( 20 mg/ml ) was added to media at a final concentration of 25 , ug/ml .
The generalized high-frequency , transducing bacteriophage P22 mutant HT10511 int-201 was used for all transductional crosses .
This phage was derived by G. Roberts ( unpublished results ) from the P22 HT10511 phage of Schmieger ( 26 ) .
To select for the inheritance of the Kanr marker of Mu dJ , the transduction mixture of cell and phage was preincubated on NB plates overnight before being replica-printed to selective plates .
In all other crosses , selective plates were spread directly with 2 x 108 cells and 108 to 109 phage .
Transductants were purified , and phagefree clones were isolated by being streaked nonselectively onto green indicator plates ( 4 ) .
Transductants containing plasmids were purified on NB plates containing ampicillin .
Phage-free clones were then tested for phage sensitivity by cross-streaking with P22 H5 phage , a clear plaque mutant of P22 .
Hydroxylamine mutagenesis of P22 transducing phage was done as described by Davis et al. ( 8 ) .
The general method is that of Hong and Ames ( 14 ) .
P-Galactosidase activity was determined as described by Miller , with sodium dodecyl sulfate-chloro-form-permeabilized cells ( 21 ) .
The P-galatosidase activity is reported as nanomoles per minute per optical density unit ( at 650 nm ) of cells .
Construction and manipulation of a tandem duplication .
The procedure for constructing the tandem duplication that covers the serB-nadI ( pnuA ) - thrA region is presented in Fig. 2 .
It is one example of a general method of constructing duplications ( 15 ; M. Schmid and J. R. Roth , manuscript in preparation ; D. Hillyard and J. R. Roth , manuscript in preparation ) .
The donor strain ( TT11433 ) carries a wild-type thr operon and a TnJO insertion ( zaa-1868 ) which is linked to the thr operon on the side clockwise from thr operon .
The recipient strain ( TT11434 ) carries a pyrB692 : : TnJO insertion and a thr470 : : Mu dA insertion mutation .
When the recipient is transduced to Thr + by using P22 transducing phage grown on the donor strain , one class of Thr + transductants is a merodiploid which carries a tandem duplication of the region from the pyrB692 : : TnlO to the zaa-1868 : : TnlO ( see Fig. 2a ) .
This merodiploid transductant class is identified by the following criteria .
( i ) The transductants are Thr + Ampr Tcr .
( ii ) The Tcr phenotype is unstable , since the TnJO element is located at the novel join point between the two copies of the duplicated region ; these two copies can recombine , resulting in the loss of the duplication and TnWO join point .
( iii ) The heterozygous state of the duplicated region ( e.g. thr + l thr : : Mu dJ ) is also unstable owing to the recombination between the two copies ; therefore , these strains give rise to Thr-Kanr and Thr + Kans segregants .
The merodiploid state can be moved into other strains by transducing the novel join point ( TnJO ) selectively into those strains ( Fig. 2b ) .
The merodiploid state can be maintained by selecting for Tcr .
A general strategy for using these merodiploids in complementation tests is the following .
First , the novel join point is transduced into a series of mutants to be tested , thus creating strains which are homozygous diploid for the recipient allele .
The second mutation can be moved into these diploids by either directly selecting the second mutation , if it has a selectable phenotype , or using a linked selective marker .
The existence of two different alleles in the strains constructed for the complementation test can be examined by a segregation test .
Cloning of nadA and nadB genes .
Phage lambda clones of nadA and nadB genes were selected from an EcoRI bank of S. typhimurium DNA cloned into Xgt7 ( 8 ) .
Clones carrying the nadA and nadB genes were selected by complementation of E. coli nadA and nadB mutants .
These lambda clones also complement S. typhimurium nadA and nadB mutations .
The initial genome inserts of nadA and nadB clones are about 7.5 and 5.5 kilobases , respectively .
Both genes were subcloned into pBR322 plasmid as PstI-EcoRI fragments .
To further subclone these two genes , exonuclease BAL 31 was used to digest the cloned fragments sequentially from two ends ; they were ligated into the multiple-cloning site of the pUC9 plasmid .
The final subclones of the nadA ( pZT352 ) and nadB ( pZT349 ) genes used in this study are about 1.5 and 2.5 kilobases , respectively .
They are able to complement nadA or nadB mutants of both E. coli and S. typhimurium .
11 RESULTS Evidence for negative control of the nadA and nadB genes .
Previous work has demonstrated that expression of nadA and nadB genes increases under conditions of pyridine limitation ( 5-7 , 13 ) .
Both genes are expressed at a high constitutive level in nadI mutants , suggesting that the nadI gene might encode a repressor protein .
To determine whether the existence of a repressor is likely , the effect of gene dosage on the expression of these genes was tested .
I a repressor exists , it should be possible to titrate the repressor in-vivo by providing a high dosage of one of the regulated genes .
As a consequence , the target gene , and other genes subject to the same repressor , would escape repression .
We have used plasmids carrying either the nadA or nadB gene in these gene dosage tests .
Both plasmids cause constitutive expression of the nadA : : Mu dJ or nadB : : Mu dJ fusion present in the host chromosome .
This suggests that the nadA and nadB genes are negatively regulated by a common repressor molecule .
These experiments are outlined below , and results are presented in Table 2 .
The S. typhimurium nadB and nadA genes have been cloned into the multiple-cloning site of plasmid pUC9 ( pZT349 and pZT352 ) ( N. Zhu , L. Bossi , and R. Gesteland nadC nadE nadB F CAs/C sCO nadA co. .
H nadD COOH COOH * CH w N m IA O8icOpo3 DHAP NADP rI1 : J Ad ~ ~ 5 ~ ~ cIN Ad 6 R-V-V-R R - ® -0 V-R o 2 N 3 4 I R - ® NaMN N a A D N A D QA - o-T-o-c ?
( Asp pncB tii pNAj \  NM NMN N 4 ¬ pruA .
NA NM NMN FIG. 1 .
NAD metabolic pathway of S. typhimurium .
The enzymes included are 1 , L-aspartate ( Asp ) oxidase ; 2 , quinolinic acid ( QA ) synthetase ; 3 , quinolinic acid phosphoribosyl transferase ; 4 , nicotinic acid mononucleotide ( NaMN ) adenylyltransferase ; 5 , NAD synthetase ; 6 , NAD kinase ; 7 , DNA ligase ; 8 , NMN deamidase ; 9 , NMN glycohydrolase ; 10 , nicotinamide ( NM ) deamidase ; 11 , NA phosphoribosyl transferase .
Abbreviations : DHAP , dihydroxylacetone phosphate ; IA , iminoaspartate ; PRPP , 5-phosphoribosyl-1-pyrophosphate .
Genetic loci corresponding to enzymatic steps are indicated above the reaction arrows .
List of strains Genotype Straina TT8785 ... .
nadA219 : : Mu dA TT8793 ... .
nadB227 : : Mu dA TT10120 ... .
nadB499 : : Mu dJ TT11334 ... .
nadA219 : : Mu dJ TT11335 ... .
nadB499 : : Mu dJ srl-202 : : TnJO recAl TT11336 ... .
nadA219 : : Mu dJ srl-202 : : TnJO recAl TT11347 ... .
srl-202 : : TnlO recAl nadB499 : : Mu dJ ( pZT349 ) TT11348 ... .
srl-202 : : TnlO recAl nadB499 : : Mu dJ ( pZT352 ) TT11349 ... .
srl-202 : : TnlO recAl nadA219 : : Mu dJ ( pZT349 ) TT11350 ... .
srl-202 : : TnlO recAl nadA219 : : Mu dJ ( pZT352 ) TT11356 ... .
nadB499 : : Mu dJ serB1463 : : TnJO nadIS507 TT11357 ... .
nadB499 : : Mu dJ serBJ463 : : TnlO nadIS5O8 TT11358 ... .
nadB499 : : Mu dJ serB1463 : : TnJO nadIS509 TT11359 ... .
nadB499 : : Mu dJ serBJ463 : : TnlO nadIS510 TT11360 ... .
nadB499 : : Mu dJ serB1463 : : TnlO nadIS511 TT11379 ... .
nadB499 : : Mu dJ serBJ463 : : TnJO nadPS513 TT11385 ... .
nadB499 : : Mu dJ serB1463 : : TnlO nadl519 TT11389 ... .
nadB499 : : Mu dJ serB1463 : : TnlO nadPS523 TT11392 ... .
nadB499 : : Mu dJ serB1463 : : TnJO nadPS526 TT11394 ... .
nadB499 : : Mu dJ serB1463 : : TnJO nadP528 TT11395 ... .
nadB499 : : Mu dJ serB1463 : : TnlO nadPl51 nadBc529b TT11396 ... .
nadB499 : : Mu dJ serB1463 : : TnlO nadPS51 nadBc530 TT11397 ... .
nadB499 : : Mu dJ serB1463 : : TnlO nadPS51 nadBc531 TT11398 ... .
nadA219 : : Mu WJ serBJ463 : : TnJO nadPl51 nadAc532b TT11399 ... .
nadA2J9 : : Mi dJ serBJ463 : : TnlO nadPS511 nadAc533 TT11400 ... .
nadB499 : : Mu dJ nadBc529 TT11401 ... .
nadB499 : : Mu dJ nadBc530 TT11402 ... .
nadB499 : : Mu dJ nadBc531 TT11403 ... .
nadA219 : : Mu dJ nadAc532 TT11404 ... .
nadA219 : : Mu dJ nadAc533 TT11408 ... .
nadB499 : : Mu dJ serB1463 : : TnJO nadI534 TT11409 ... .
nadB499 : : Mu dJ serB1463 : : TnJO nadI535 TT11410 ... .
nadB499 : : Mu dJ serB14 .63 : : TnlO nadI536 TT11411 ... .
nadA2I9 : : Mu dJ serB1463 : : TnJO nadI534 TT11412 ... .
nadA2J9 : : Mu dJ serB1463 : : TnlO nadI535 TT11413 ... .
nadA2l9 : : Mu dJ serB1463 : : TnlO nadI536 TT11416 ... .
nadB499 : : Mu dJ nadP51l srl-202 : : TnJO recAl TT11418 ... .
nadA2J9 : : Mu dJ nadIS5l1 srl-202 : : TnlO recAl TT11422 ... .
nadB499 : : Mu dJ nadP51 srl-202 : : TnlO recAi ( pZT349 ) TT11424 ... .
nadA219 : : Mu dJ nadP51 srl-202 : : TnlO recAl ( pZT349 ) TT11428 ... .
nadB499 : : Mu dJ nadJI511 srl-202 : : TnlO recAl ( pZT352 ) TT11430 ... .
nadA219 : : Mu dJ nadP51 srl-202 : : TnlO recAl ( pZT352 ) TT11433 ... .
pyrB692 : : TnlO thr470 : : Mti dA TT11435 ... .
DUP728 [ ( zaa-1868 ) TnlO ( pyrB692 ) ] c TT11441 ... .
nadB227 : : Mu dJ serB9 pncA278 : : TnlOd ( Cam ' ) d TT1144 ... ... 3 ... .
nadB227 : : Mu dJ pncA278 : : TnlOd ( Cam ' ) DUP728 [ ( serB9 zaa-1868 ) TnlO ( pyrB692 s TT11449 ... .
nadB227 : : Mu dJ pncA278 : : TnlOd ( Camr ) DUP728 [ ( nadP51l zaa-1868 ) TnlO ( pyrB69 TT11450 ... .
nadB227 : : Mu dJ pncA278 : : TnlOd ( Cam ' ) DUP728 [ ( zaa-1868 ) TnlO ( pyrB692 serB9 ) ] TT11453 ... .
nadB227 : : Mu dJ pncA278 : : TnlOd ( Cam ' ) DUP728 [ ( nadI262 zaa-1868 ) TnlO ( pyrB692 TT11744 ... .
nadB499 : : Mu dJ nad1262 TT13239 ... .
nadB499 : : Mu dJ srl-202 : : TnlO recAl ( pNC9 ) TT13240 ... .
nadA219 : : Mu dJ srl-202 : : TnJO recAl ( pNC9 ) TT13242 ... .
nadB499 : : Mu dJ nadIl511 srl-202 : : TnlO recAl ( pNC9 ) TT13244 ... .
nadA219 : : Mu dJ nadIS5JJ srl-202 : : TnlO recAl ( pNC9 ) a All strains used in this study are derivatives of S. typhimurium LT2 .
b nadBc and nadAc designate constitutive mutations linked to the nadB and nadA genes , respectively .
' Duplication nomenclature is as described by Schmid and Roth ( 25 ) .
d TnJOd ( Camr ) designates a transposition-defective derivative of transposon TnlO ( T. Elliott and J. R. Roth , personal communication ) .
erB9 ) ] 2 serB9 ) ] serB9 ) ] TT11424 nadA2J9 : : Mu dJ nadPl511 ( pZT349 ) ( nadB ) TT11430 nadA2J9 : : Mu dJ nadPSlJ ( pZT352 ) ( nadA ) a B , Blue on X-Gal plate ; W , white on X-Gal plate .
pUC9 is nadA and nadB b a plasmid used to clone the genes .
cThe plasmids pZT349 ( nadB clone ) and pZT352 ( nadA clone ) were transduced into the strains by transducing phage P22 .
The pUC9 plasmid is a pBR322-derived cloning vector and exists in high copy number in cells ( 29 ; Zhu et al. , unpublished data ) .
These plasmids can be used to provide many copies of either nadB or nadA in-vivo .
The plasmids were introduced by P22-mediated transduction into hosts carrying in their chromosome either a nadA : : Mu dJ or a nadB : .
Mu dJ fusion and a recA mutation ( TT11338 and TT11341 ) .
The results were scored qualitatively by the color of colonies on media with X-Gal ( Table 2 ) .
In each case , the presence of the plasmid causes derepression of the chromosomal nad : : lac fusion .
The simplest explanation of these results is that a repressor is involved in the regulation of the nadA and nadB genes .
When either gene is present in high copy number , repressor is sequestered , resulting in derepression of the chromosomal fusions .
We provide evidence that the nadI gene encodes this repressor protein .
Isolation of low-level constitutive mutants ( nadr ) .
If the nadl gene encodes a repressor , mutants with mutations analogous to the super repressor ( lacr ) of the lactose operon ( 22 ) might be recovered .
Such nadP mutations , which would cause repression of nadA and nadB genes under all growth-conditions , would be expected to be rare ( compared with nadl null alleles ) and dominant to the nadIl allele in complementation tests .
In a search for such nadP mutants , we subjected the nadI region to localized mutagenesis and screened for mutants that express a nadB : : Mu dJ fusion at a constitutive low level .
Such mutants were isolated , and the mutations were mapped and characterized as described below .
To achieve local mutagenesis , a strain containing serBJ463 : : TnJO ( TT10120 ) was used as a donor .
This TnlO insertion is about 85 % linked to the pnuA-nadl locus .
The recipient camres a nadB499 : : Mu dJ lacZ operon fusion TT10738 nadB499 : : Mu dJ 320 6 TT11356 nadB499 : : Mu dJ nadPS507 5 3 TT11357 nadB499 : : Mu dJ nadPS508 4 3 TT11358 nadB499 : : Mu dJ nadPSO9 4 3 TT11359 nadB499 : : Mu dJ nadPS50 4 2 TT11360 nadB499 : : Mu dJ nadPS51 4 3 TT11744 nadB499 : : Mu dJ nadI-262 373 372 a The strains were grown in minimal media supplemented with serine and the indicated concentrations of NA .
( TT10738 ) far from the nadI region .
The selection for tetracycline resistance was done on minimal X-Gal plates supplemented with a low concentration of NA ( 10 6 M ) ; on this medium most transductants form blue colonies , but a few white colonies were found .
These mutants express the nadB : : lac fusion at very low level under either repressing ( 2 x iO-4 M NA ) or derepressing ( 10-6 M NA ) conditions .
The putative nadP mutations show the expected linkage to the serB locus .
When P22 phage grown on five nadIP serB : : TnJO double mutants ( TT11351 to TT11355 ) was used to transduce nadA : : lac and nadB : : lac fusion strains ( TT10738 and TT11334 ) on minimal X-Gal plates containing 10-6 M NA plus tetracycline and serine , 80 to 90 % of the Tcr transductants show the phenotype expected for nadP mutants and form white colonies on X-Gal plates with either high or low levels of NA .
The,3-galactosidase activities of the putative nadP transductants are presented in Table 3 .
The results agree with those observed in the color test on X-Gal plates .
Thus , these mutations are coinherited with serB1463 : : TnlO at about the same frequency as nadI mutations ( 7 ) , but , in contrast to the derepression of nadA and nadB seen for nadl mutants , they express both nadB-lac and nadA-lac fusions at a constitutive low level .
Some nadP mutations cause pyridine auxotrophy .
To test the phenotype of nadIP mutations in a wild-type genetic background , without a nadA : : lac or nadB : lac insertion , five nadP alleles ( TT11356 to TT11360 ) were transduced into wild-type strain LT2 by selecting for inheritance of the linked serBl463 : : TnJO marker .
Three of the five mutants ( TT11357 , TT11358 , and 1T11360 as donors ) gave 80 to 90o Tcr transductants that are NA auxotrophs ( Table 4 ) .
The a Phages grown on these donors were used to transduce wild-type recipient strain LT2 selecting inheritance of Tcr ( conferred by the serBI463 : : TnlO insertion ) .
The Tcr transductants were picked and patched on NB plates containing tetracycline and then replica printed to minimal plates other two donors gave no auxotrophic transductants and have not been studied further .
We presume that the superrepression phenotype of these three nadls alleles is sufficiently strong that de novo synthesis of NAD is prevented .
No NA auxotrophic mutations have previously been isolated near serB ; we presume that this is due to the rarity of nadP mutations ( see below ) .
Frequency of nadIP and nadl mutations .
The nadP ( constitutive low-level expression ) mutations and the nadl ( constitutive high-level expression ) mutations show approximately the same linkage to serB , but have opposite phenotypes .
The relative frequency of these two mutant types has been examined .
The expectation is that the nadI class would include null mutations and thus be common , while the nadP mutations would produce a repressor with very specific properties and thus be rare .
Mutagenized P22 transducing phage , grown on the serB1463 : : TnJO strain ( TT10120 ) , was used to transduce nadB499 : : Mu dJ ( TT10738 ) , selecting for the inheritance of tetracycline resistance on nutrient agar medium .
The Tcr colonies were replica printed to minimal glucose plates containing X-Gal plus serine and tetracycline supplemented with a low level ( 106 M ) or a high level ( 2 x 1O ' M ) of NA .
White colonies on the low-NA plates ( nadP ) and blue colonies on the high-NA plates ( nadI ) were scored .
Putative nadPr mutants were found at a frequency of i0 ' ( 17 from about 160,000 Tcr transductants ) ; nadI mutants were found at a frequency 2 x 10-2 ( 3,300 from 160,000 Tcr transductants ) .
The ratio of the former to the latter is 1:200 .
Thus , as predicted , the frequency of the constitutively high mutations ( nad !
) is significantly higher than that of the constitutively low mutations ( nadP ) .
The rarer nadP mutants recovered from this experiment all show constitutive low expression of nadA and nadB genes and are 80 to 90 % linked to the serB locus ( data not shown ) .
Effect of nadA and nadB copy number on nadlP mutants .
If the nadP mutations owe their phenotype to a mutant repressor active even under conditions of pyridine limitation , then these phenotypes ought to be corrected by providing , in high copy number , either of the target sequences ( nadA or nadB ) with which this protein is thought to interact .
To test this , we introduced the nadA or nadB plasmid into strains carrying a nad : - Mu dJ fusion , a nadP mutation , and a recA mutation ( TT11415 to TT11418 ) .
This was done by comparing their relative level of lacZ expression on indicator media .
In all cases , introduction of the high-copy-number plasmid relieved the low-constitutive phenotype of the nadP mutant and , as expected , led to a high constitutive expression of whichever fusion was present .
Qualitative scoring of I8-galactosidase for some of these strains is shown in Table 2 .
Spontaneous Nad + revertants of nadP mutants .
If nadP mutants owe their phenotype to production of an active super repressor , then this phenotype should be lost by the introduction of a null mutation in the nadI gene .
Since some nadIP mutants have an auxotrophic phenotype in an LT2 background , revertants can be positively selected .
The frequency of Nad + revertants , their phenotype , and the linkage of the secondary mutation to the parental nadP mutation have been examined .
Overnight cultures of three auxotrophic nadP mutants ( TT11366 , TT11367 , and TT11368 ) were washed and plated on the minimal-medium plates containing serine .
Nad + revertant colonies were scored .
The spontaneous revertant frequencies in these nadP mutants are high ( 3 x 10-5 to 5 x 10-6 ) .
Six revertants of each of the above nadP strains were examined .
Several revertants are Ser - , Tcs , and PnuA - , and when tested further , these phenotypes were not separable in genetic crosses , suggesting that they are deletions .
Since the pnuA and nadI genes are known to map at the same genetic locus , the deletions probably extend from the serB locus to the pnuA-nadl region and remove the TnJO element inserted in serB .
( These deletions may be generated by the TnJO element .
) Similar deletions were also found in the spontane-ous Nad + revertants of a nadP strain without serB : : TnlO .
The regulatory phenotypes of both Ser-Tcs and Ser-Tcr revertants were tested by transducing either a nadB : : Mu dJ fusion ( from TT10738 ) or a nadA : : Mu dJ fusion ( from TT11334 ) into these strains .
All Nad + revertants of nadP mutants show constitutive high expression of nadA or nadB .
These results suggest that the nadIs auxotrophy is corrected by nadI null mutations .
Levels of P-galactosidase were assayed for several revertants of nadf mutants for which the nadl region had been transduced into a genetic background that includes a nadA : : Mu dJ or nadB : : Mu dJ fusion ( TT10738 or TT11334 ) .
The results ( Table 5 ) show that the revertants ( which have mutations tightly linked to the parental nadP mutations ) show constitutive high-level expression of both nadB and nadA , as do the nadI mutations described previously ( 7 ) .
Expression of nad : : lac at a low constitutive level by nadP mutants TABLE 2 .
Gene dosage effect of the expression of nadA and nadB genes in a nadI or nadP background Color ' on X-Gal plate with : 10-2 106 2 x 1-6 M NA M NA B W B W B B , B-Galactosidase activity ( U ) on Strain Relevant genotype Strain Relevant genotype mediuma with : 10-6 2 x 10-4 M NA M NA TT11335 TT13239 TT11347 TT11348 TTi336 TT13240 TT11349 TT11350 nadB499 : : Mu dJ nadB499 : : Mu dJ ( pUC9 ) b nadB499 : : Mu dJ ( pZT349 ) c ( nadB ) nadB499 : : Mu dJ ( pZT352 ) ( nadA ) nadA219 : : Mu dJ nadA2l9 : : Mu dJ ( pUC9 ) nadA219 : : Mu dJ ( pZT349 ) ( nadB ) nadA219 : : Mu dJ ( pZT352 ) ( nadA ) B B B W B W B B B B W W nadPSll nadB499 : : Mu dJ TT11416 TT13242 TT11422 W W nadB499 : : Mu dJ nadPSl1 ( pUC9 ) nadB499 : : Mu dJ nadPSl1 ( pZT349 ) ( nadB ) nadB499 : : Mu dJ nadPS511 ( pZT352 ) ( nadA ) B B B B TT11428 W W TT11418 nadA219 : : Mu dJ nadPSJJ TT13244 nadA219 : : Mu dJ nadPS51 ( pUC9 ) W W B B B B TABLE 4 .
Transduction of nadP mutations into strain LT2 No .
of Tcr Donor Relevant genotype transductantsa auxotrophs 0 44 41 0 38 TT11356 TT11357 TT11358 TT11359 1T11360 serB1463 : : TnlO nadPS507 serB1463 : : TnlO nadPS508 serB1463 : : TnlO nadPSO9 serB1463 : : TnIO nadPS50 serB1463 : : TnJO nadPSJ1 50 49 50 50 50 TABLE 5 .
Expression of nadA : : lac and nadB : : lac fusions in prototrophic revertants of nadP Relevant genotype 13-Galactosidase activity ( U ) in medium with : 10-6 M NA 2 x iO-M N Strain nadl revertant allele A nad : : lac allele nadP allele nadPS508 ( nadPS508 ) a ( nadPS508 ) a TT11357 TT11408 TT11411 nadB499 : : Mu dJ nadB499 : : Mu dJ nadA2J9 : : Mu dJ 4 3 295-256-370-316 nadI534 nadIS34 TT11358 nadB499 : : Mu dJ nadPSO9 4 3 TT11409 nadB499 : : Mu dJ ( nadPSO9 ) a 328 335 FT11412 nadA219 : : Mu dJ ( nadISO9 ) a 450 473 nadIS35 nadIS35 TT11360 nadB499 : : Mu dJ nadlSlla 4 3 TT11410 nadB499 : : Mu dJ ( nadISJJ ) a nadIS36 57 23 TT11413 nadA219 : : Mu dJ ( nadPl5I ) a nadIS36 245 123 TT11744 nadB499 : : Mu dJ nadI262 373 372 a The nadP allele indicated in parentheses is the parental nadP allele ( auxotrophic ) whose presence is inferred ; the strain acquired the nadI mutation when it reverted to prototrophy These data suggest that the auxotrophic nadIs mutants become prototrophic when they acquire a nadl mutation .
nadI mutations with different effects on nadA and nadB genes .
If nadl encodes a repressor that acts to control both the nadA the nadB genes , one might expect that some nadI mutants would produce a repressor that could distinguish between the two control regions .
This is not expected if nadI mutations cause derepression indirectly ( e.g. , they might cause a reduction of the NAD pool and indirectly lead to derepression ) .
Data in Table 5 describe a nadl mutant allele that distinguishes between nadA and nadB regions .
Mutant nadI536 ( which was selected as a prototrophic revertant of nadPSll and presumably carries both mutations ) shows constitutive expression of both nadA and nadB fusions at a level higher than that shown by the parent nadIs mutant .
However , the constitutive level is fivefold higher for nadA than for nadB .
Revertants of the other nadP mutants do not show this differential effect on the two target genes ( Table 5 ) .
Isolation of temperature-sensitive nadP mutants .
Evidence presented thus far is consistent with , but does not directly address , the idea that nadI and nadli mutations are alleles of a single gene .
This idea is directly supported by isolation of temperature-sensitive mutants that show a Nadls phenotype at one temperature and a Nadl-phenotype at another temperature .
Seventeen nadP mutants ( TT11378 to TT11394 ) , which form white colonies on X-Gal plates containing 10-6 M NA , were isolated at 30 °C .
The mutations were transduced into a unmutagenized background by crossing with the parental recipient strain TT10738 ( nadB : : Mu dJ ) selecting for inheritance of mutation serB1463 : : TnlO ( Tcr ) .
Transductants which formed white colonies at 30 °C on X-Gal plates containing a low concentration of NA were tested further .
The cotransduction frequency of these mutations to serB1463 : : TnlO is 80 to 90 % .
These nadP transductants can be divided into several classes on the basis of their phenotypes at elevated temperatures .
The results of P-galactosidase assays for strains representative of each class are presented in Table 6 .
The most informative class is represented by mutation nadPS526 in line 5 of Table 6 .
The strain carrying this mutation shows constitutive low-level expression of the nadB fusion at 300C .
At 37 °C , expression is regulated in response to pyridine supplementation .
At 400C , the mutant shows constitutive high-level expression characteristic of a nadI null allele .
Below we will show that the NadI5 phenotype ( at 300C ) is dominant to NadI + and that the Nadl-phenotype ( at 400C ) is recessive to NadI + .
Other classes of nadl mutants are also presented in Table 6 .
The most common class , including 6 of the 17 mutants , is shown in line 2 .
This class shows constitutive low-level expression at all temperatures .
Line 3 presents a mutant class that shows a NadIs phenotype at 30 and 37 °C but regulates expression at 40 °C .
Another mutant type ( line 4 ) regulates at both 37 and 40 °C .
Line 6 presents a mutant that shows no regulated expression , but increases expression in response to higher temperatures .
As a control , a typical nadI null mutant is presented in line 7 .
The existence of these mutants and the fact that at 40 °C NadI-phenotypes are so frequent among mutants isolated as NadPs types at 30 °C ( 5 of 17 ) suggest that both phenotypes are caused by mutations in the same gene .
Results presented above show that nadIs mutations are rare compared with nadl mutations and that these two kinds of mutations seem to be different alleles of the same gene .
The results suggest that the nadP mutations might be dominant alleles that produce a functional but altered repressor , while the nadI type might be null mutations and therefore would be expected to be recessive .
Dominance tests confirm these expectations : nadlY mutations are dominant to the wild type , and nadI mutations are recessive .
Complementation tests were made by using tandenm duplications to form the necessary merodiploids as discussed in Materials and Methods and as shown in Fig. 2 .
The duplicated segment starts within the pyrB gene ( 98 min ) and extends to just beyond the thr locus ( 100 min ) .
A series of nadP and nadl mutations were transduced into this duplication strain by selecting for Ser + transductants which inherit the ser + locus ( and very frequently the donor nadI allele ) in one copy of the duplication .
Thus , the strains are heterozygous for serB and for nadI , with the general structure ( serB + nadI ) TnlO ( serB nadI + ) .
These merodip-loids permit the testing of dominance and allow simple verification of the genotype of the diploid , since segregants which arise by recombination between the two copies are of two types : Tcs Ser + Nad and Tcs Ser-Nad + .
The three nadP mutations tested ( nadIS08 , nadIS09 , and nadIS11 ) were each transduced into the Ser-merodiploid selecting Ser + transductants ; the resulting strains ( nadlI nadI ) ( TT11445 , TT11447 , and TT11449 ) form white colo-nies on X-Gal plates containing low concentrations of NA .
This is a NadIP phenotype , which demonstrates that in all cases tested , the nadls mutation is dominant to the nadI + allele carried in the other copy of the duplication .
The structure of the merodiploids used for these tests was verified by isolating Tcs segregants that had lost the duplication .
Segregants from the strains used for the dominance tests fell into two classes : Ser + NadPs and Ser-NadI + .
This verifies that both alleles of nadI were present in the diploid .
The three nadI mutations tested ( nadI260 , nadI261 , and nadI262 ) were each transduced into the Ser-merodiploid selecting Ser + transductants .
All transductants ( nadl/nadI ) ( TT11451 , TT11452 , and TT11453 ) showed normal regulation of nadB as judged by colony color on X-Gal plates containing a high or low concentration of NA .
Thus , the nadl mutations tested are all recessive to the wild-type allele ( nadI ) present in the other copy of the duplicated segment .
The structure of the diploids on which this conclusion is based was demonstrated by allowing the duplication to segregate and testing the phenotypes of the haploid segregants .
Some segregants show constitutive high-level expression of the nadB : : Mu dJ fusion , whereas others show regulated expression .
This demonstrates that the diploid did carry the mutant nadl allele as well as the wild-type nadIl allele .
The,-galactosidase activities of some of the merodi-ploids and their segregants have been measured .
The results ( Table 7 ) agree with what was found with X-Gal plates .
Additional evidence on dominance of nadP mutations comes from the dominance test of mutation nadPS526 ( Ts ) ; this mutation has a Nadls phenotype at 30 °C , is NadI + at 37 °C , and is Nadl-at 42 °C ( Table 6 , line 5 ) .
The diploid strain ( TT11454 ) [ nadPS526 ( Ts ) / nadI ) shows constitutive low-level expression of the nadB : : Iac fusion at 30 °C ( the NadIP phenotype at 30 °C is dominant to NadI + ) ; at 40 °C , expression of the fusion is regulated normally in response to NA concentration ( the NadI-phenotype at 40 °C is recessive to NadI + ) .
Isolation of putative cis-regulatory mutations mapping near nadA and nadB .
To further explore interactions between the nadl repressor and the genes regulated by this repressor , attempts have been made to isolate cis-acting regulatory mutants .
The cis-acting regulatory or operator mutations are expected to map near the nadA and nadB genes and to be rare .
We used the Kanr determinants of nadB499 : : Mu dJ and nadA219 : : Mu dJ as selectable markers and performed localized mutagenesis of each region .
P22 transducing phage was grown on nadA : : Mu dJ and nadB : : Mu dJ strains ( TT10738 and TT11334 ) ; the free phage was mutagenized with hydroxylamine as described in Materials and Methods .
The mutagenized lysate was used to transduce a nadP mutant strain ( TT11368 ) selecting for kanamycin resistance on minimal X-Gal plates containing a low concentration ( 10-6 M ) of NA .
Most transductants form white colonies on this medium owing to the nadP ( super-repressor ) mutation present in the recipient .
Rare blue colonies were picked as candidates for mutants that had become insensitive to the nadP repression ; some of these show a constitutively high level of,-galacto-sidase .
Three putative cis-regulatory mutants from the nad B499 : : Mu dJ donor ( TT11395 , TT11396 , and TT11397 ) and two from the nadA2J9 : : Mu dJ donor ( TT11398 and TT11399 ) have been tested ; each is highly linked ( 97 to 99 % cotransduction ) to the nadA or nadB fusion .
The new regulatory mutations and their linked fusion were transduced into a wild-type ( nadI ) background ( TT11400 to TT11404 ) , and regulation checked by assaying 3-galactosidase ( Table was 8 ) .
The mutations linked to the fusion cause the of escape transcription from repression by both the nadI and the nadI repressor .
Mutants shown in Table 8 have not lost all control of expression ; small differences in the P-galactosidase levels are generated in response to changes in NA concentration .
These data support the idea that these mutations have changed operator sites , which retain some affinity for repressor .
Although these mutants are presumed to affect operator sites of nadA and nadB genes , we have not yet demonstrated the expected cis-dominance behavior .
Effect of nadP ( Ts ) mutations on nadB : : Iac fusion 1-Galactosidase activity ( U ) at : 37 °C 2 x 1O-4 M NA 6 3 3 9 36-128-369 nadl allelca 300C 40C 10-6 M NA 258 3 6 31-120-126-300 10-6 M NA 2 x 1O-4 M NA 2 x 10-4 M NA 10-6 M NA nadI + 417 9 nadPS53 4 3 nadPS528 6 4 nadISl9 8 6 nadIS526 11 8 nadP523 18 11 nadI262 482 568 a All strains carry the fusion nadB499 : : Mu dJ 4 3 169 3 63-120-253-251 235 3 26-182-276-247 a thr + thr + zaa : : TnI 0 TnlO pyrB Donor ( TT1 1435 ) ( novel join point ) b Donor ( TT l 1433 ) MudA thr pyrB ' TnlO pyrB ' pgrB + serB-nadI + thr + Recipient ( TTl 1434 ) Recipient Select TetR thr + serB-nadI ~ ~ ~ ~ I ~ transduced fragment TnlO % pyrrB 1l thr serB-nedIl copy copy 11 MudA Copy 11 - , Copy r f pgrB + pgrB serB-nadI + thr + serB-nadI LL .
thr + TnlO L Tn1 O TnlO .
novel joi n point Tandem Duplication ( TT 1 1435 ) .
Construction and transduction of a tandem duplication structure .
( a ) Construction of a tandem duplication which duplicates a region from pyrB to thr and has a TnJO transposon at the join point .
( b ) Transduction of the join point of the tandem duplication to a new strain .
The structure of the tandem duplication is regenerated when inheritance of the join point is selected ; all other duplicated material is of recipient origin .
Dominance test of nadP and nadI mutations P-Galactosidase activity ( U ) in medium with : 2 x 10-4MNA Relevent genotype of merodiploid [ ( Copy 1 ) * TnlO * ( Copy 2 ) ] Strain 10-6MNA 285 4 T711443 TT11449 nadB227 : : Mu dJ ( serB9 nadl + ) * TnJO * ( serB9 nadI + ) nadB227 : : Mu dJ ( serB + nadPS51 ) * TnJO * ( serB9 nadI + ) nadB227 : : Mu dJ serB + nadPSIJ nadB227 : : Mu dJ serB9 nadI + 2 3 3 3 Tcs segregant Tcs segregant 9 359 4 301-402-374 TT11450 nadB227 : : Mu dJ ( serB + nadI + ) * TnlO * ( serB9 nadI + ) nadB227 : : Mu dJ serB + nadI + nadB227 : : Mu dJ serB9 nadI + 9 Tcs segregant Tcs segregant 6 5 336-441-440 TT11453 nadB227 : : Mu dJ ( serB + nadI262 ) * TnlO * ( serB9 nadI + ) nadB227 : : Mu dJ serB + nadI262 nadB227 : : Mu dJ serB9 nadI + 337 1 Tcs segregant Tcs segregant TABLE 8 .
Effect of putative operator mutations on expression of the nadA and nadB genes P-Galactosidase activity ( U ) in medium with : 10-6 2 x 20-4 M NA M NA 320 6 4 3 377-191-134 95 Strain Relevent genotype TT10738 TT11360 nadB499 : : Mu dJ nadB499 : : Mu dJ nadP511 TT11400 FT11395 nadBc529 nadB499 : : Mu dJa nadBc529 nadB499 : : Mu dJ nadJS511 TT11401 nadBc530 nadB499 : : Mu dJ 366 204 TT11396 nadBc530 nadB499 : : Mu dJ nadP511 130 88 TT11402 nadBc531 nadB499 : : Mu dJ 408 184 TT11397 nadBc531 nadB499 : : Mu dJ nadP511 116 87 DISCUSSION The results presented here provide genetic evidence for the existence of a repressor for the NAD biosynthetic pathway of S. typhimurium .
This repressor , encoded by the nadl gene , appears to be directly responsible for the transcriptional control of the nadB and nadA genes , two biosynthetic genes that are not linked to each other or to the nadI gene .
The nadI gene was first discovered because mutations at that locus cause derepression of the nadA and nadB genes .
Genetic analysis has revealed that nadI appears to be in an operon with the pnuA gene , which is known to encode a product involved in the for NMN ( 7 , 9 , 13 , transport system 18 ) .
The mechanism by which mutations in this region cause derepression of the de novo pathway was not previously known ; the relationship between nadI and pnuA also was unclear .
The first suggestion of has from in-vivo a repressor come dosage experiments with cloned nadB nadA gene a or gene .
The high number of either of these copy one two genes results in derepression of chromosomal nadB and nadA We have pursued the possibility that genes .
a repressor is encoded in the nadI A specific of which gene .
type mutant expresses nadA and nadB at a constitutively low level was isolated on the basis of the assumption that mutations in rare could result in noninducible like a repressor gene a repressor super-repressor mutations ( lacd ) in the lac system ( 17 , 22 ) .
Because of this analogy , we have designated these mutations nadP .
Several lines of evidence suggest that nadlY ( super-re-pressor ) and nadI ( high constitutive ) mutations affect the same gene .
( i ) Both nadlY and nadl mutations show approximately the same cotransduction frequency with TT11403 nadAc532 nadA2J9 : : Mu dJa 511 279 TT11398 dJ 236 154 nadAc532 nadA219 : : Mu nadP511 TT11404 nadAc533 nadA219 : : Mu dJ 493 266 TT11399 nadAc533 nadA2J9 : : Mu dJ nadPS511 198 153 a See Table 1 , footnote b. serBJ463 : : TnJO .
( ii ) Selection for reversal of the Nadls phenotype yields nadI mutations which are tightly linked to the original nadP mutations .
( iii ) Some temperature-sensi-tive nadP mutations isolated at 30 °C show constitutive high-level ( Nadl - ) expression of nadA-lac and nadB-lac fusions at higher temperatures .
The nadP mutations , which are rare , are dominant to nadI + , and the common nadI mutations are recessive to the wild-type allele .
The data also suggest that the nadI protein interacts directly with nadA and nadB sequences .
One of the spontaneous revertants of nadP511 ( nadI536 ) causes constitutive expression of the nadB-lac and nadA-lac fusions at significantly different levels .
This suggests that the mutant repressor distinguishes between the two normal control regions .
Constitutive mutations linked to nadB-lac and nadA-lac appear to alter the site at which the regulatory protein acts ; these mutants retain slight sensitivity to control , which is still affected by the difference between the nadI + and nadIP backgrounds .
The results presented strongly support the idea that the nadI gene encodes a repressor affecting the transcription of the nadA and nadB genes .
These results do not identify the that act as a The corepressor might signal compound .
finding that leaky nadE ( NAD synthetase ) mutants grow with elevated transcription of nadA and nadB genes suggests that the final in the compounds pathway , NAD or NADP , may provide the regulatory signal ( K. T. Hughes , B. M. Olivera , and J. R. Roth , unpublished results ) .
Our results predict that a repressor binds a pyridine nucleotide and interacts with sequences near the nadA and nadB genes to control transcription of these biosynthetic genes .
This protein may be bifunctional and play a direct role in NMN transport , which would explain the PnuA-phenotype of nadIP mutants ( unpublished data ) .
Alternatively , the nadI repressor may be a monofunctional protein encoded in the same operon as the pnuA gene .
We thank R. Gesteland , in whose laboratory the cloning of nadA and nadB genes was done , and K. Hughes for advice and enlightening discussions TT11402 nadBc531 nadB499 : : Mu dJ 408 184 TT11397 nadBc531 nadB499 : : Mu dJ nadP511 116 87 This work was supported by Public Health Service grants GM23408 ( to J. R. Roth ) and GM25654 ( to B. M. Olivera ) from the National Institutes of Health .
ADDENDUM IN PROOF The data included here were presented previously ( Abstr .
Since this submitted , similar data paper was were published by Foster et al. ( J. W. Foster , E. A. Holley-Guthrie , and F. Warren , Mol .
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 .
Castliho , B. A. , P. Olfson , and M. J. Casadaban .
Plasmid insertion mutagenesis and lac gene fusion with mini-Mu bacteriophage transposons .
Chaconas , G. , F. J. deBru `` n , M. J. Casadaban , J. R. Lupski , T. J. Kwok , R. M. Harshey , M. S. DuBow , and A. I. Bukhari .
In vitro and in-vivo manipulations of bacteriophage Mu DNA : cloning of Mu ends and construction of mini-Mu 's carrying selectable markers .
Chan , R. K. , D. Botstein , T. Watanabe , and Y. Ogata .
Specialized transduction of tetracycline resistance by phage P22 in Salmonella typhimurium .
Properties of a high frequency transducing lysate .
Chandler , J. L. R. , and R. K. Gholson .
De novo biosynthesis of nicotinamide in Escherichia coli : excretion of quinoli-nic acid by mutants lacking quinolinate phosphoribosyl transferase .
Chandler , J. L. R. , and R. K. Gholson .
Studies on the biosynthesis of NAD in Escherichia coli .
Precursors of quinolinic acid in-vitro .
Cookson , B. T. , B. M. Olivera , and J. R. Roth .
Genetic characterization and regulation of the nadB locus of Salmonella typhimurium .
Davis , R. W. , D. Botstein , and J. R. Roth ( ed .
Cold Spring Harbor Laboratory , Cold Spring Harbor , N.Y. 9 .
Foster , J. W. , D. M. Kinney , and A. G. Moat .
Pyridine nucleotide cycle of Salmonella typhimurium : isolation and characterization of pncA , pncB , and pncC mutations and utilization of exogenous nicotinamide adenine dinucleotide .
Foster , J. W. , and A. G. Moat .
Mapping and characterization of the nad genes in Salmonella typhimurium LT-2 .
Foster , J. W. , and A. G. Moat .
Nicotinamide adenine dinucleotide biosynthesis and pyridine nucleotide cycle metab-olism in microbiol systems .
Holley , E. A. , and J. W. Foster .
Bacteriophage P22 as a vector for Mu mutagenesis in Salmonella typhimurium : isolation of nad-lac and pnc-lac gene fusions .
Holley , E. A. , M. P. Spector , and J. W. Foster .
Regulation of NAD biosynthesis in Salmonella typhimurium : expression of nad-lac gene fusions and identification of a nad regulatory locus .
Hong , J.-S. , and B. N. Ames .
Localized mutagenesis of any specific small region of the bacterial chromosomes .
Hughes , K. , D. Ladika , J. R. Roth , and B. M. Olivera .
An indispensable gene for NAD biosynthesis in Salmonella typhi-murium .
Hughes , K. T. , and J. R. Roth .
Conditional transpositiondefective derivative of Mu d ) ( Amp , Lac ) .
Jobe , A. , A. D. Riggs , and S. Bourgeois .
Kinney , D. M. , J. W. Foster , and A. G. Moat .
Pyridine nucleotide cycle of Salmonella typhimurium : in-vitro demonstration of nicotinamide mononucleotide deamidase and characterization of pnuA mutants defective in nicotinamide mononucleotide transport .
Langley , D. , and J. R. Guest .
Biochemical and genetic characteristics of deletion and other mutant strains of Salmo-nella typhimurium LT-2 lacking a-keto acid dehydrogenase complex activities .
DNA ligase : structure , mechanism , function .
Experiments in molecular genetics .
Cold Spring Harbor Laboratory , Cold Spring Harbor , N.Y. 22 .
Millson , C. , D. Perrin , M. Cohn , F. Jacob , and J. Monod .
Non-inducible mutants of the regulatory gene in the `` lactose '' system of Escherichia coli .
Olivera , B. M. , and I. R. Lehman .
Diphosphopyridine nucleotide : a cofactor for the polynucleotide-joining enzyme from Escherichia coli .
Saxton , R. E. , V. Rocha , R. J. Rosser , A. J. Andreoli , M. Shimoyoma , A. Sosska , J. L. R. Chandler , and R. K. Gholson .
A comparative study of the regulation of nicotinamide adenine dinucleotide biosynthesis .
Schmid , M. B. , and J. R. Roth .
Genetic methods for analysis and manipulation of inversion mutations in bacteria .
A method for detection of phage mutants with altered transducing ability .
Characterization of the nadR locus in Esche-richia coli .
Tritz , G. J. , and J. L. R. Chandler .
Recognition of a gene involved in the regulation of nicotinamide adenine dinucleotide biosynthesis .
Vieira , J. , and J. Messing .
The pUC plasmids , an M13mp7-derived system for insertion mutagenesis and sequencing with synthetic universal primers .
Vogel , H. J. , and D. M. Bonner .
Acetylornithase of Escherichia coli : partial purification and some properties .