171 , No. 9 Transcription of pfl Is Regulated by Anaerobiosis , Catabolite Repression , Pyruvate , and oxrA : pfl : : Mu dA Operon Fusions o Salmonella typhimurium f K. K. WONG , K. L. SUEN , AND H. S. KWAN * Department of Biology , The Chinese University of Hong Kong , Shatin , N.T. , Hong Kong Received 20 October 1988/Accepted 31 May 1989 Pyruvate formate-lyase ( EC 2.3.1.54 ) , a key enzyme in the anaerobic metabolism of Salmonella typhiunuium , catalyzes the conversion of pyruvate to acetyl coenzyme A and formate .
pfl : : Mu dA operon fusions were isolated for the study of transcriptional regulation .
pfl was transcribed both aerobically and anaerobically , but the activity increased about sixfold under anaerobic conditions .
The addition of pyruvate , formate , and acetate in nutrient broth did not have any effect on the anaerobic expression of pfl .
However , the addition of pyruvate to minimal glucose medium increased the anaerobic expression of pfl .
The expression of pfl varied in different growth media .
Anaerobic expression of pfl was lower when the culture was grown in minimal glucose medium than when it was grown in nutrient broth .
When Casamino Acids ( Difco Laboratories , Detroit , Mich. ) were added to minimal glucose medium , the expression ofpfl increased proportionally with the amount of Casamino Acids added .
The transcription of pfl was positively controlled by the oxrA gene product and was affected by both the cya and crp mutations .
However , mutations in genes affecting the cyclic AMP-cyclic-AMP-receptor-protein complex or oxrA could not completely abolish the anaerobic derepression ofpfl .
In merodiploid strains , pfl : : Mu dA/F ' pjt , the j-galactosidase activities were decreased .
The mutations gyrA , oxrC , and oxrE , which affected anaerobic metabolism , did not affect anaerobic expression of pfl .
Facultative anaerobic enteric bacteria such as Salmonella typhimurium and Escherichia coli carry out mixed acid fermentation during anaerobic-growth-on-glucose when electron-acceptors such as nitrate and trimethylamine oxide are not available .
Pyruvate formate-lyase ( PFL ; EC 2.3.1.54 ) is a key enzyme in fermentative growth and is responsible for converting pyruvate to acetyl coenzyme A and formate , which serve as substrates for the production of acetate , ethanol , hydrogen gas , and carbon dioxide ( 14 ) .
The activity of PFL is regulated by a reversible enzymatic interconversion of an active form to an inactive form in response to oxygen ( 6 ) .
Smith and Neidhardt ( 27 ) determined that the anaerobic steady-state level of PFL is 10 times higher than the aerobic level .
A temporal overproduction of PFL after a shift from aerobic to anaerobic-growth has also been dem-onstrated ( 27 ) .
The expression of the cloned pfl gene of E. coli , which is the structural gene encoding PFL , is derepressed 5-to 10-fold during anaerobiosis compared with the basal level in aerobically grown E. coli ( 22 ) .
However , it is unclear whether the derepression of pfl is regulated by oxygen per se or by the accumulation of some inducers .
The effects of various global control systems , such as catabolite-repression andfnr ( oxrA ) regulation ( 28 ) , on pfl transcription are still unclear .
We report here the use of pfl : : Mu dA operon fusions to study the regulation of transcription of pfl .
MATERIALS AND METHODS Bacteria and bacteriophages .
The bacterial strains used in this study are listed in Table 1 .
All S. typhimurium strains were derived from strain LT2 .
Transductions with P22 int4 phage have been described by Ely et al. ( 8 ) .
The Mu dl derivative Mu dA has been described by Hughes and Roth ( 12 ) .
Ampicillin , fusaric acid , tetracycline , chlortet-racyline , kanamycin sulfate , tryptophan , 5-bromo-4-chloro - 3-indolyl-fo-D-galactoside ( X-gal ) , and organic-acids were from Sigma Chemical Co. ( St. Louis , Mo. ) .
MacConkey agar , Bacto-Agar , nutrient broth ( NB ) , nutrient agar , and MacConkey agar base were purchased from Difco Laboratories ( Detroit , Mich. ) .
Complex medium consisted of 0.8 % NB .
Minimal glucose medium ( NCE medium with 1 % glucose ) E has been described by Gutnick et al. ( 10 ) .
Medium ( 31 ) 1 % with glucose was the minimal-medium used for genetic mapping .
The MacConkey agar-glucose-trimethylamine oxide media ( MGT ) MGT and MGT1 have been described previously ( 15 ) .
The concentrations of glucose in MGT1 and MGT were 1 and 0.15 % , respectively ; and that of trimethyl-amine oxide was 0.1 % .
When they were grown aerobically , pfl mutants were dark red on MGT1 plates , while the wild-type strains was pink in the center .
When grown anaerobically , the pfl mutants formed red colonies on MGT , while the wild-type strains formed white colonies .
MacCon-key nitrate medium has been described by Barrett et al. ( 3 ) .
Under anaerobic conditions on MacConkey nitrate medium , oxrA mutants formed tiny dark red colonies , and the wildtype strains formed large white colonies .
The fusaric acid medium described by Maloy and Nunn ( 18 ) was used to select for tetracycline-sensitive derivatives from tetracy-cline-resistant parents .
Green and X-gal media have been described by Miller ( 19 ) .
When needed , potassium nitrate ( 20 % ) , glucose ( 40 % ) , and other carbon sources were autoclaved separately as stock solutions and added to the me-dium at final concentrations of 1 % .
Tryptophan was prepared as a 40 mM stock solution , filter sterilized , and added to the autoclaved medium at a final concentration of 1 mM .
Ampicillin and tetracycline were added to the autoclaved complex medium at concentrations of 50 and 20 , ugIml , respectively .
Ampicillin and tetracycline were added to the 490 Putative pfl : : Mu dA ( Apr lac ) mutants were purified on MGT1 and green plates .
Mu dA operon fusions were stabilized by transducing them into strains without a suppressor mutation ( 12 ) .
Determination of transcription orientation .
The tempera-ture-sensitive episome F ' ts lac TnJO ( 5 ) was used for the determination of the transcription orientation .
In this episome , the orientation of the lac operon relative to oriT is known .
After integration via lac homology into the pfl : : Mu dA fusion strain HSK21 at the nonpermissive temperature , one Hfr strain was obtained .
The direction of transcription of pfl in this Hfr strain was determined as described before ( 15 ) .
An anaerobic liquid culture was prepared by filling tubes to the top , and the tubes were stoppered with rubber stoppers after inoculation .
Growth was started by inoculating a 1 % overnight culture grown aerobically in nutrient broth into the tubes .
At time intervals , the A650 of the cultures was measured directly against a blank of growth medium .
Anaerobic incubation of plates wa TABLE 1 .
minimal-medium at concentrations of 25 and 10 , ug/ml , respectively .
Gas production from pyruvate and glucose was detected by inoculating the bacteria into NB with a durham tube in the presence or absence of 10 mM formate .
The tubes were then grown at 37 °C as standing cultures for 24 h and scored for the presence of gas in the durham tube .
Acid profiles were obtained by high-perfor-mance liquid chromatographic analysis of acid products from glucose as described by Guerrant et al. ( 9 ) .
D-Glucose was determined by the ortho-toluidine method ( 7 ) .
Isolation of pfl : : Mu dA operon fusions .
S. typhimurium TT8388 , which has a Mu dA ( Apr lac ) insertion in its F ' factor ( 12 ) , was used as the donor of Mu dA ( Apr lac ) to generate operon fusions .
P22 lysate prepared from strain TT8388 was used to transduce strain HSK2 to ampicillin resistance on aerobic MGT1 plates containing 50 , ug of ampicillin per ml .
After 24 h of incubation at 30 °C , the plates were examined for red colonies among pink-centered wild-type colonies .
RESULTS '' Aerobic growth was achieved by growing 20 ml of culture in a 250-ml conical flask with vigorous shaking .
Anaerobic-growth was achieved by incubating tubes that were filled to the top with medium and stoppered .
b Specific activity is expressed as nanomoles of o-nitrophenol per minute per unit .
When HSK21 was grown in minimal glucose medium , the anaerobic 3-galactosidase expression was lower than that in NB .
However , the addition of pyruvate to the minimal glucose medium increased pfl expression .
Moreover , the addition of Casamino Acids ( Difco ) to the minimal glucose medium increased both the growth-rate and the expression of pfl : : Mu dA .
Supplementation of 1 % Casamino Acids to minimal glucose medium increased the anaerobic 3-galactosidase expression to the anaerobic expression level in NB .
When HSK21 was grown in minimal glycerol nitrate medium , expression of the pfl : : Mu dA fusion was almost as low as that at the aerobic level .
Regulation of pfl by oxrA .
The oxrA mutation affects the expression of many genes encoding components of anaerobic respiration systems , such as nitrate reductase ( 16 ) , but it does not seem to affect genes of the fermentative pathway ( 13 ) .
Since the pfl gene product is involved in both fermentation and anaerobic respiration , it was necessary to investigate whether it is controlled by oxrA .
To answer this question , we moved an oxrA : : TnJO insertion into the pfl : : Mu dA fusion and assayed for the 3-galactosidase activity of the anaerobically grown culture .
In the oxrA background , anaerobic transcription of the fusion was lowered by about twofold , but the aerobic transcription was unaffected and the anaerobic transcription was still threefold higher than aerobic transcription .
Thus , pfl is under the global regulation of oxrA , but the oxrA gene product is not absolutely essential for the anaerobic induction of pfl .
The effect of catabolic-repression on the regulation of pfl was investigated in cya pfl : : Mu dA and crp pfl : : Mu dA double mutants .
Isogenic pairs were studied .
Both cya and crp mutations lowered the expression of pfl : : Mu dA by about twofold ( Table 3 ) .
The effect of cya could be reversed by the addition of 5 mM cyclic AMP ( cAMP ) to the growth medium , indicating that the absence of cAMP affects pfl transcription .
The regulation of pfl b cAMP seemed to be via the cAMP-receptor-protein ( CRP ) because the pfl : : Mu dA strain which carried both cya and crp mutations did not have,3-galactosidase activities different from those of fusions with either mutation alone ( Table 3 ) .
On the other hand , cAMP did not appear to exert its effect on pfl transcription via the oxrA gene product , although the OxrA ( Fnr ) protein has homology with CRP ( 26 ) .
The results given in Table 3 indicate that the effects of cya and oxrA are independent of each other .
However , the transcription of pfl : : Mu dA in the cya oxrA double mutation background was still derepressed by anaerobiosis , albeit to a lesser extent .
Therefore , in addition to oxrA and catabolite-repression , other factors must be involved in the aerobic-anaerobic control of pfl transcription .
We constructed merodiploids to study the regulation of pfl : : Mu dA expression in a Pfl + background .
The F ' 114 episome , which carries a fragment of the E. coli chromosome containing the wild-type pfl structural gene , was conjugated into aroA derivatives of pfl : : Mu dA .
The F ' episome was maintained by complementation of the aroA mutation in the chromosome .
The exoconjugant became Gas ' , indicating that the Pfl function is complemented in the merodiploid .
Other mutations were then transduced into the merodiploid to study their effects on pfl expression .
In the merodiploid with the Pfl + background , anaerobic expression ofpfl : : Mu dA was decreased ( Table 4 ) .
The oxrA mutation further decreased pfl : : Mu dA expression in the merodiploid , but pfl expression was still higher under anaerobic conditions .
It was previously observed that Ackmutants had the same phenotype as Pfl-mutants ( 20 ) , suggesting that in ack mutants pfl is not expressed .
Comparison of the isogenic merodiploid pair , which differed only ip that one was Ack-and the other was Ack + , revealed that the ack mutation did not affect pfl expression .
On the contrary , the merodiploid with the ack mutation had an increased pfl expression , reaching a level similar to that ih the haploid .
Perhaps both pfl and ack mutations caused an accumulation of an inducer which induced pfl expression .
Pyruvate seemed to be the inducer , as exogenous pyruvate added to the minimal glucose medium stimulated pfl expression ( Table 2 ) .
Regulation by gyrA , oxrC , and oxrE .
gyrA is suggested to be a regulatory gene for the expression of anaerobic genes because gyrA mutants can not grow anaerobically on nutrient agar plates ( 34 ) and anaerobic induction of E. coli formate dehydrogenase ( hydrogenase-linked ) is enhanced by gyrase inhibition ( 2 ) .
We tested the effect of the gyrA mutation on pfl expression .
A gyrA derivative of pfl : : Mu dA was isolated by its resistance to nalidixic-acid ( 20 , ug/ml ) .
pfl expression was not affected in this strain ( Table 3 ) .
Jamieson and Higgins ( 13 ) have described a mutation , oxrC , which decreases phosphoglucose isomerase activity and affects the transcription of fhl and hyd .
We have isolated an oxrC mutant strain ( H. S. Kwan and K. K. Wong , unpublished data ) which has the same phenotype as that of the oxrC mutant strain isolated by Jamieson and Higgins ( 13 ) .
Neither strain fermented glucose , but both strains fermented fructose and both strains had only about 1/10 of the wild-type phosphoglucose isomerase activity .
Our oxrC mutation did not affect pfl transcription ( Table 3 ) .
We moved another mutation , oxrE , which causes a deficiency in formate-nitrate reduction ( J. S. Tang and E. L. Barrett , Abstr .
1986 , K120 , p. 213 ) , into pfl : : Mu dA and determined the P-galactosidase activity .
oxrE did not affect pfl expression .
Effects of oxygen and growth media on the expression of pfl : : Mu dA ( lac ( ) in HSK21 carried out in anaerobic jars ( GasPak ; BBL Microbiology Systems , Cockeysville , Md. ) by using an atmosphere of 95 % H2-5 % CO2 .
Aerobic growth of the liquid culture was achieved by growing 20 ml of culture in a 250-ml conical flask with vigorous shaking .
The 3-galactosidase assay we performed has been described by Miller ( 19 ) .
Toluene-treated cell cultures were used directly for the assay .
For the phosphoglucose isomerase assay , the rate of production of D-fruc-tose-6-phosphate from D-glucose-6-phosphate by sonicated crude cell extracts was measured .
Quantitative determination of D-fructose-6-phosphate was done by the method of Roe et al. ( 23 ) .
Protein was determined by the method of Bradford ( 4 ) by using protein assay reagents ( Bio-Rad Laboratories , Richmond , Calif. ) .
Construction of merodiploid strains .
An F ' episome carrying the wild-type pfl and aroA transferred from E. genes was coli KL725 to an S. typhimurium strain with the genotype aroA5330 pfl : : Mu dA .
Mating was performed as described previously ( 19 ) .
Transconjugants were purified and checked for F ' and gas production .
The F ' episome was maintained by keeping the merodiploid strains in minimal-medium and selecting for AroA + strains .
Other strain constructions were done by P22 transduction .
, - Galacto - Basal medium and Supplement growth-condition ' sidase sp act '' NB Aerobic None None 270 1,900 1,630 1,780 1,690 Anaerobic Pyruvate , 0.1 % Formate , 0.05 % , Acetate , 0.1 % Minimal glucose Aerobic None Pyruvate , 0.05 % 227 223 Pyruvate , 0.1 % Pyruvate , 0.5 % None Casamino-acids , 0.1 % Casamino-acids , 1.0 % Pyruvate , 0.05 % Pyruvate , 0.1 % Pyruvate , 0.5 % 287-199-606 1,490 2,290 1,240 1,500 1,090 Anaerobic Minimal glycerol ( anaerobic ) Nitrate , 0.1 % 396 Isolation and characterization of pfl : : Mu dA fusion .
Four pfl : : Mu dA fusions isolated red colonies MGT1 were as on plates that were incubated aerobically .
They were stabilized by transduction into HSK516 , which had no suppressor mutation .
These mutants had the same phenotype as HSK3 ( pfl ) ; that is , they all formed red colonies on anaerobic MGT plates , produced gas only in the presence of formate , and produced less H2S in peptone-iron agar .
All insertions were cotransducible with zbj : : TnlO , which was linked to pfl ( 24 ) , with cotransduction frequencies of 70 to 94 % .
Strain HSK21 , which was chosen for further studies , could not produce formate from glucose ( data not shown ) .
Highperformance liquid chromatographic determination of organic-acids in cultures of HSK21 grown anaerobically in minimal glucose medium showed that HSK21 did not accumulate formate .
When compared with the parental strain HSK516 , PFL mutant HSK21 produced 10-fold less pyruvate , 5-fold less fumarate , slightly less succinate and acetate , but 2-fold more lactate , which constituted about 90 % of the total organic-acids in the culture of HSK21 .
The decreased anaerobic-growth-rates of strains HSK21 and HSK3 were observed in minimal glucose medium ( the generation time increased to 8 h ) .
The growth-rate , however , could be increased by the addition of trimethylamine oxide or nitrate , indicating that HSK3 and HSK21 are both deficient in fermenting glucose efficiently but have normal anaerobic respiration .
The map location and phenotype of HSK21 thus indicate that this strain carries a pfl : : Mu dA operon fusion .
The transcription orientation of pfl : : Mu dA was determined to be counterclockwise on the genetic map .
pfl : : Mu dA fusion strains expressed,-galactosidase aerobically in NB , and the activity was derepressed about sixfold when they were grown anaerobically in NB ( Table 2 ) .
Supplementation of pyruvate , formate , and acetate in NB did not affect the anaerobic expression of pfl in strain HSK21 .
When HSK21 was grown aerobically , it had a differential rate of 320 U/A650 unit .
A shift of the aerobic culture to anaerobiosis increased the differential rate of 3-galactosidase expression 10-fold to BL TA E 3 .
Effects of different gene mutations on the expression of the pfl : : Mu dA ( lac ) operon fusiona P-Galactosidase sp act under the following growth-conditions ' : Relevant mutation , in addition to pfl : : Mu dA ( lac ) Strain Aerobic Anaerobic HSK21 None 270 1,900 HSK1101 cya : : TnlO 192 693 HSK1120 crp : : Tn1O 167 710 HSK1121 Acya crp : : TnJO 192 791 HSK1105 oxrA : : TnlO 313 577 HSK1116 AoxrA 201 751 HSK1119 AoxrA cya : : TnJO 161 273 HSK1122 Acya 176 704 HSK1123 Acya oxrA : : TnJO 191 282 HSK1109 gyrA 279 1,790 HSK1110 oxrC + zxx : : TnlO 275 1,630 HSK1111 AoxrC zxx : : TnJO 266 1,660 HSK1132 oxrE + zda : : TnIO 194 1,570 HSK1133 oxrE zda : : TnlO 169 1,240 a Cultures were grown to the stationary-phase in N Broth and harvested for , B-galactosidase assays .
Aerobic growth was achieved by growing 20 ml of culture in a 250-ml conical flask with vigorous shaking .
Anaerobic-growth was achieved by incubating tubes that filled the with medium were to top and stoppered .
b Specific activity is expressed as nanomoles of o-nitrophenol per minute per A650 unit .
DISCUSSION Our results indicate that the transcription of pfl is regulated at the following three levels : ( i ) derepression by anwhen the pfl : : Mu dA mutant was grown in minimal glucose medium in which the anaerobic expression of pfl reached the level of pfl expression in rich-medium .
The addition of Casamino Acids had a similar effect , presumably because Casamino Acids contain amino-acids which can be converted to pyruvate during-growth of the bacteria .
Expression of pfl in merodiploids was also derepressed by pyruvate , but a higher pyruvate ( 1 % ) concentration was required ( Table 4 ) .
The F ' pfl + episome probably lowered pfl expression by providing PFL , which converted pyruvate to formate and acetyl coenzyme A .
An additional ack mutation could change the expression of pfl in the merodiploid to the full level , probably because the ack mutation somehow inhibits PFL activity , causing pyruvate to accumulate and thus derepressing pfl .
It is still not clear , however , whether pyruvate acts directly as an inducer or acts indirectly by releasing catabolite-repression .
The catabolic reduction charge , defined as NADH / ( NADH + NAD + ) , has been suggested to affect anaerobic metabolism ( 1 , 32 ) .
The conversion of pyruvate to lactate can regenerate NAD + from NADH and can lower the catabolic reduction charge .
The effect of pyruvate may be related to the catabolic reduction charge .
We have isolated a mutation , pflR ( 33 ) , which eliminates the anaerobic induction of pfl .
The primary defect caused by the pflR mutation is a deficiency in lactate production .
A deficiency in lactate production would probably affect the regeneration of NAD + .
It seems that a high catabolic reduction charge affects the anaerobic expression of pfl .
Alternatively , it is possible that lactate is a signal for anaerobic induction of pfl .
We have observed that when the wild type is grown anaerobically in minimal glucose medium with nitrate , the final concentration of lactate is negligible ( Kwan and Wong , unpublished data ) , and the anaerobic level of expression of pfl in nitrate medium is comparable to the aerobic level .
Furthermore , Pecher et al. ( 22 ) have shown that the aerobic level of expression of the cloned pfl structural gene on growth at the expense of D-lactate is twice as high as that at the expense of D-glucose .
It is possible that lactate somehow titrates some repressors and causes the derepression of pfl .
The pleiotropic mutations oxrC , gyrA , and oxrE , which are known to affect anaerobic metabolism , did not affect the expression of pfl .
Since the oxrC mutant did not grow in minimal-medium , we tested its effect on pfl expression in NB .
Thus , if an inducing signal is generated during glycolysis , the inducing signal may be counteracted by catabolites that are produced during-growth in NB .
As pyruvate can be generated from the degradative products of amino-acids , our data do not necessarily conflict with the fact that oxrC ( pgi ) is required for the generation of an inducing signal .
Yama-moto and Droffner ( 34 ) have suggested that DNA supercoiling plays a significant role in anaerobic gene expression .
Their conclusion was based on the isolation of strict aerobic mutants containing mutations in gyrA and gyrB which lead to little or no DNA gyrase activity .
Our gyrA mutant may not have completely lost gyrase activity , and thus , the effect was not obvious .
The effect of supercoiling on the expression of pfl should be tested with gyrase inhibitors such as coumer-mycin .
None of the mutations , when tested alone , could completely eliminate the anaerobic derepression of pfl , indicating that multiple signals are involved .
Since this paper was submitted for publication , it was described in another report ( 25 ) that ( i ) there is anaerobic induction ( 12-fold ) and pyruvate stimulation ( 2-fold ) of the pfl gene from E. coli K-12 ; ( ii ) complete anaerobic induction requires a functionalfnr gene product , but the dependence i LITERATURE CITED 16 .
Kwan , H. S. , and K. K. Wong .
A general method for isolation of Mu dl8 ( Apr lac ) operon fusions in Salmonella typhimurium LT2 from TnlO insertion strains : chlC : : Mu dl-8 .
Coupling to an energized membrane : role of ion-motive gradients in the transduction of metabolic energy , p. 222-243 .
In F. C. Neidhardt , J. L. Ingraham , K. B. Low , B. Magasanik , M. Schaechter , and H. E. Umbarger ( ed .
) , Esche-richia coli and Salmonella typhimurium : cellular and molecular biology .
American Society for Microbiology , Washington , D.C. 18 .
Maloy , S. R. , and W. D. Nunn .
Selection for loss of tetracycline resistance by Escherichia coli .
Experiments in molecular genetics .
Cold Spring Harbor Laboratory , Cold Spring Harbor , N.Y. 20 .
Pascal , M.-C. , M. Chippaux , A. Abou-Jaoude , H. P. Blasch-kowski , and J. Knappe .
Mutants of Escherichia coli K12 with defects in anaerobic pyruvate metabolism .
Patrick , J. M. , and W. J. Dobrogosz .
The effect of cyclic AMP on anaerobic-growth in Escherichia coli .
Pecher , A. , H. P. Blaschkowski , J. Knappe , and A. Bock .
Expression of pyruvate formate-lyase of Escherichia coli from the cloned structural gene .
Roe , J. H. , J. H. Epstein , and N. P. Goldstein .
A photometric method for the determination of inulin in plasma and urine .
Sanderson , K. E. , and J. R. Roth .
Linkage map of Salmonella typhimurium , edition VI .
Sawers , G. , and A. Bock .
Anaerobic regulation of pyruvate formate-lyase from Escherichia coli K-12 .
Shaw , D. J. , D. W. Rice , and J. R. Guest .
Homology between CAP and Fnr , a regulator of anaerobic respiration in Escherichia coli .
Smith , M. W. , and F. C. Neidhardt .
Proteins induced by anaerobiosis in Escherichia coli .
Strauch , K. L. , J. B. Lenk , B. L. Gamble , and C. G. Miller .
Oxygen regulation in Salmonella typhimurium .
Unden , G. , and A. Duchene .
On the role of cyclic AMP and the Fnr protein in Escherichia coli growing anaerobically .
Unden , G. , and J. R. Guest .
Cyclic AMP and anaerobic gene expression in Escherichia coli .
Vogel , H. J. , and D. M. Bonner .
Acetylornithase of Escherichia coli : partial purification and some properties .
Wimpenny , J. W. T. , and A. Firth .
Levels of nicotinamide adenine dinucleotide and reduced nicotinamide adenine dinucle-otide in facultative bacteria and effect of oxygen .
Wong , K. K. , and H. S. Kwan .
A novel mutation causing the deficiency in lactate production of Salmonella typhimurium during mixed acid fermentation , p. 293-300 .
In S. T. Chang , K. Y. Chan , and N. Y. S. Woo ( ed .
) , Recent advances in applied biology and biotechnology .
The Chinese University Press , Hong Kong .
Yamamoto , N. , and M. L. Droffner .
Mechanisms determining aerobic or anaerobic-growth in the facultative anaerobe Salmonella typhimurium .