perons in f Cobalamin Biosynthetic O Salmonella typhimurium JORGE C. ESCALANTE-SEMERENA * AND JOHN R. ROTH University of Utah , Department of Biology , Salt Lake City , Utah 84112 Received 21 October 1986/Accepted 12 February 1987 Transcription of cobalamin ( cob ) biosynthetic genes in Salmonella typhimurium is repressed by cobalamin and by molecular oxygen .
These genes seem to be subject to catabolite-repression , and they are maximally expressed under conditions of anaerobic respiration of glycerol-fumarate .
A 215-fold increase in the expression of cob genes occurs when S. typhimurium shifts from aerobic-growth-on-glucose to anaerobic respiration of glycerol-fumarate under strictly anoxic growth-conditions .
Exogenous cyclic AMP substantially stimulates the transcription of cob-lac fusions during aerobic-growth .
However , cyclic AMP is not absolutely required for the expression of the pathway , nor does it mediate the aerobic control .
Cobalamin biosynthesis is not seen under aerobic-growth-conditions , even when transcription is stimulated by the addition of cyclic AMP .
Hence , additional control mechanisms triggered by the presence of molecular oxygen must operate independently from transcription effects on the cob operons .
The metabolic role of cobalamin ( vitamin-B12 ) in bacteria has been studied most extensively in bacteria of the genera Clostridium , Rhodopseudomonas , Rhizobium , Lactobacillus , and Propionibacterium ( 1 , 3-5 , 11 , 12 , 15 , 16 , 21 , 24 , 26 , 29 , 30 , 33 , 39 , 40 , 42 ) .
Relatively little is known about the general metabolic significance of cobalamin , how its synthesis is regulated , or why some bacteria produce such large quantities of this cofactor .
The finding that cobalamin is synthesized in Salmonella typhimurium ( 19 ) makes it possible to apply genetic methods of assessing the metabolic significance of cobalamin and the regulation of its synthesis .
What importance of cobalamin accounts for the maintenance of a pathway estimated to require 30 biosynthetic enzymes ?
Why does S. typhimurium synthesize cobalamin only under anaerobic conditions ?
Of all the chemical reactions known to be catalyzed by cobalamin-dependent enzymes ( 14 ) , only two have been shown to occur in the enterobacteria Escherichia coli and S. typhimurium .
One of these reactions is the last step in the biosynthesis of methionine ( 38 ) .
In this reaction the methyl group of N5-methyltetrahydrofolate is transferred to Lhomocysteine to yield methionine .
Methylcobalamin is the intermediate in this reaction .
The methyltransferase enzyme catalyzing this reaction is the product of the metH locus , which is thought to play an important role in the transcriptional regulation of the alternative methyltransferase ( the metE gene product ) , which is known to catalyze the same reaction in a cobalamin-independent way ( 18 , 25 , 35-37 ) .
The second cobalamin-dependent reaction known to occur in enterobacteria involves the catabolism of ethanolamine as a carbon or nitrogen source ( 6 , 30 ) .
This reaction is catalyzed by ethanolamine-ammonia lyase to yield acetate and ammonia .
This enzyme requires adenosyl-cobalamin as a coenreactions zyme ( 7 ) .
To date , no other cobalamin-dependent have been reported to occur in S. typhimurium .
Neither of the known reactions is essential , in that S. typhimurium mutants lacking the ability to synthesize cobalamin grow normally on glucose both aerobically and anaerobically .
These facts suggest that cobalamin may play a role in discovered .
metabolism that is yet to be To approach the metabolic significance of cobalamin we have investigated the biosynthetic genes , thinking that knowledge of their regulation might suggest clues to the general metabolic importance of this cofactor .
Mutants defective in cobalamin biosynthesis have been nutritionally classified into three groups ( CobI , CoblI , and CobIII ) ( 19 ) .
CobI mutants are defective in the synthesis of the intermediate cobinamide and can synthesize cobalamin only if cobinamide is provided .
Cobll mutants are defective in the synthesis of 5,6-dimethylbenzimidazole ( DMB ) but can make cobalamin if DMB is supplied .
CoblIl mutants fail to make cobalamin even if both DMB and cobinamide are supplied ; they are presumed to be defective in late steps of the pathway involved in joining cobinamide and DMB to form cobalamin .
The cob biosynthetic genes are located near the his operon at min 41 of the chromosome , and they are transcribed counterclockwise ( 19 ; R. Jeter and J. Roth , submitted for publication ) .
The cobI biosynthetic genes appear to comprise a single operon , since synthesis of cobinamide ( the cobI reactions ) occurs only under anaerobic conditions .
The cobII and cobIII operons are very close to each other but separated ( -10 kilobases ) from the cobI operon as judged by genetic linkage .
The number of operons involved in the cobII and cobIII clusters is uncertain .
Here we report the effect of growth-conditions on the transcription of the cob operons .
The results are interpreted to suggest that the most important role of cobalamin may be to promote anaerobic catabolism of nonfermentable carbon sources .
Bacteria , media , and growth strains used are derivatives of S. LT2 .
The typhimurium genotypes of all bacterial strains used in this study are listed in 1 .
Difco nutrient broth containing NaCl at a ( 0.8 % ) Table of 85 mM was used as medium .
final concentration complex The E medium of and Bonner supplemented with Vogel ( 41 ) mM was used as minimal 11 mM glucose or 22 glycerol medium .
No-carbon E medium ( 10 ) supplemented with MgSO4 ( 10 mM ) was used for studying the expression of lac operon fusions during fermentative growth on citrate .
Citrate ( trisodium salt ) was added to a final concentration of 40 mM .
Auxotrophic supplements were added at concentrations recommended elsewhere ( 10 ) .
The final concentrations of antibiotics in complex medium were 20 gug of tetracycline , 50 , ug of kanamycin , and 30 gug of ampicillin per ml .
The final concentrations of antibiotics in minimal-medium were 10 , ug of tetracycline , 125 , ug of kanamycin , and 15 , ug of ampicillin per ml .
Solid media contained 1.5 g of Bacto-Agar ( Difco Laboratories ) per liter .
When added to the culture medium , the final concentrations were 1.5 x 10-8 M cyano-cobalamin , 1.8 x 10-8 M cobinamide dicyanide , and 3 x 10-4 M DBM .
High aeration of the cultures was achieved by growing the cells in 5 ml of medium in a 125-ml culture flask ( Bellco Glass , Inc. , Vineland , N.J. ) at a speed setting of 8 in a Gyrotory shaker ( New Brunswick Scientific Co. ) .
Cell density was monitored with a Klett-Summerson photoelectric colorimeter ( Klett Manufacturing Co. , Inc. ) .
( i ) Growth under low-oxygen conditions .
All manipulations of anaerobic cultures were performed inside an anaerobic chamber ( Forma Scientific ; model 1024 ) whose atmosphere contained N2-H2-CO2 ( 90:5:5 ) .
Culture medium was rendered low in ( but not completely devoid of ) dissolved molecular oxygen by overnight degassing inside the anaerobic chamber .
Inocula were prepared by single-colony inoculation of 2 ml of low-oxygen medium in 13-by 100-mm culture tubes .
Culture tubes were stoppered with sterile rubber stoppers .
The rubber stoppers were individually wrapped in aluminum foil , autoclaved , and degassed over-night inside the anaerobic chamber before use .
Larger volumes of anaerobic cultures ( e.g. , 5 ml ) were obtained in Transcription of cobalamin ( cob ) biosynthetic genes in Salmonella typhimurium is repressed by cobalamin MATERIALS AND METHODS conditions .
All bacterial 225 125-ml culture flasks as described above , but fitted with rubber stoppers to maintain low-oxygen conditions .
All 5-ml cultures were started at a cell density of approximately 10 Klett units .
( ii ) Growth under anoxic conditions .
To completely remove 02 gas dissolved in the culture medium , we prepared it as described elsewhere ( 2 ) for growing strict anaerobes .
Briefly , the culture medium was brought to a boil under a blanket of 02-free N2 gas ; the boiled medium was brought into the anaerobic chamber and dispensed into tubes ( 5 ml each ) .
Then , the tubes were fitted with previously degassed rubber stoppers , removed from the chamber , and crimp-sealed with an aluminum seal .
The atmosphere in the headspace of the tubes ( 22 ml ) was exchanged for 02-free N2 gas to remove residual CO2 and H2 gas .
The pressure of gas inside the tubes was maintained at 101 kPa ( 101 kPa is equal to 1 atm ) .
For the expetiment showing expression of cob-lac fusions as a function of the partial pressure of oxygen , premeasured amounts of air were injected into the tubes after removal of an equivalent volume of N2 gas from the headspace of the tubes .
Partial pressure of oxygen was calculated by assuming 21 % 02 in air .
The tubes were then autoclaved and inoculated by means of syringes .
Cell turbidity was monitored with a Bausch & Lomb Spectronic 20 spectrophotometer at 650 nm TABLE 1 .
All transductional crosses were performed with a derivative of bacteriophage P22 which contains the mutation HT 10511 ( HT = high transducing ) , which increases the frequency of generalized transduction ( 31 , 32 ) , and the mutation int-201 , which prevents the formation of stable lysogens ( 34 ) .
All transductional crosses were performed at a multiplicity of infection of about 1 .
A rapid procedure for obtaining P22 phage lysates has been described ( 10 ) .
In most cases , phage and bacteria were mixed directly on solid media .
Crosses to be plated on kanamycincontaining media were first plated on drug-free medium to allow expression of the drug resistance phenotype and then replica printed onto selective medium .
Tetracycline and ampicillin resistance selections did not require any preincubation on drug-free medium .
Transductants were freed of phage by streaking on green indicator plates ( 10 ) .
Sensitivity to phage infection was tested by cross-streaking against P22 clear-plaque-forming mutant H5 .
Construction of cob : : Mu d11734 fusions .
The isolation of a transposition-defective derivative of the original Casadaban and Cohen Mu lac bacteriophage ( 8 ) has been reported ( 17 ) .
Operon fusions of the transposition-defective bacteriophage Mu d 18 ( hereafter referred to as Mu dA ) were obtained in each of the branches of the cobalamin biosynthetic pathway ( Jeter and Roth , submitted ) .
To further reduce the frequency of transposition of these insertion mutations , a deletion of the transposition genes was added to each insert by recomi-bination with Casadaban 's Mu dII1734 ( 9 ) .
This conversion ( originally described by Castilho et al. [ 9 ] ) occurs by homologous recombination between sequences present in Mu dA and in Mu dII1734 to yield Mu d11734 ( hereafter referred to as Mu dJ ) .
The resulting Mu dJ derivative carries kanamycin resistance instead of ampicillin resistance and is 11.3 kilobases in size instead of 37.2 kilobases .
More importantly , Mu dJ lacks the A and B Mu functions necessary for transposition , which prevents further transposition of the inserted material .
Beta-galactosidase activity was assayed as described by Miller ( 23 ) using CHCl3-sodium dodecyl sulfate to permeabilize whole-cells .
The final assay volume was 1.7 ml .
Enzyme activity was expressed as nanomoles of o-nitrophenyl-p-D-galactoside per minute per unit of optical density at 650 nm ( A650 ) .
All assays were performed in duplicate in early-log-phase cultures ( 70 to 80 Klett units ) .
E. coli K-12 was used as a positive and negative control for the expression of the lacZ gene .
A culture grown in the presence of the gratuitous inducer isopropyl -,3-D-thiogalactopyranoside ( 1 mg/ml ) was used as the positive control .
A culture grown in the absence of isopropyl-p-D-thiogalactopyranoside was used as the negative control .
Typical values for the amount of beta-galactosidase synthesized by E. coli K-12 in the presence of IPTG ranged between 1,100 and 1,300 U of activity per A650 unit .
In the absence of the inducer , 3 to 5 U of activity per A650 unit were recorded .
Isopropyl-p-D-thiogalactopyranoside , o-nitro-phenyl-p-D-galactoside , cyclic AMP sodium salt ( cAMP ) , antibiotics , trimethylamine-N-oxide , and other chemicals were purchased from Sigma Chemical Co. ( St. Louis , Mo. ) .
Dimethyl sulfoxide was purchased from EM Science ( Cherry Hill , N.J. ) .
RESULTS Regulation of cobalamin synthetic genes by growth-conditions .
The effect of molecular oxygen on the transcriptional regulation of cob-lac fusions was investigated .
Separate strains carrying operon fusions of the lacZ gene to a gene in each one of the branches of the cobalamin biosynthetic pathway ( cobI , cobII , and cobIII ) were tested .
Table 2 illustrates the regulatory effect of molecular oxygen on the expression of cob genes as measured by the synthesis of beta-galactosidase .
These assays were performed under the low-oxygen conditions described in Materials and Methods .
The expression of the fusions was assayed for cells grown under conditions of anaerobic fermentation of glucose and anaerobic respiration of glycerol-fumarate .
These results were compared to those obtained for cells grown aerobically on glucose or glycerol as the energy source .
All the assays presented in Table 2 are on cells unable to synthesize cobalamin , and cultures were grown without cobalamin in the growth medium .
L-Methionine was added to the culture medium to satisfy the auxotrophic requirement of all the strains .
Several points should be noted .
Maximum expression of the cob genes was seen under conditions of anaerobic respiration with fumarate as the electron-acceptor .
Compared with this level , expression is strongly reduced in the presence of glucose ( glucose minus oxygen ) or oxygen ( glycerol plus oxygen ) .
Also a small but repeatable glucose effect is seen in aerobic cells grown with glucose versus cells grown with glycerol as the carbon source .
The effect was twofold on cobI fusions .
The cobII and cobIII fusions remained unaffected .
This effect suggested the possibility of cAMP control , which was tested and is discussed below .
Fusions to the cobI operon routinely show a wider range of expression than cobII and cobIII fusions ; that is , under repressed conditions , cobI fusions are shut off more tightly , and under induced conditions they are expressed more highly .
This observation was made with three independently isolated fusions in the cobI operon ( strains TT10852 , TT10875 , and TT10876 ; Table 1 ) ; these fusions are known t map at different locations in the operon ( Jeter and Roth , submitted ) .
Only one fusion to cobI , cobII , or cobIII is presented , since no difference in their regulatory response was found in parallel experiments .
Under conditions of anaerobic respiration ( glycerol-fumarate ) , the level of transcription of a cobI fusion ( strain TT10852 ) was 22-fold higher than the level measured under fermentative growth-on-glucose .
The same pattern of regulation was seen for fusions to cobII and cobIII operons .
However , the levels for these operon fusions were only seven-to eightfold higher on glycerol-fumarate versus glucose ( Table 2 ) .
The most dramatic increase in the transcription of a cobI-lac fusion was seen for anaerobic respiration of glycerol-fumarate , where over 400 U of beta-galactosidase activity was produced .
This represents an increase of more than 2 orders of magnitude .
Compared with the level seen for aerobic-growth-on-glucose , the stimulatory effect of anaerobic expression on cobII and cobIlI fusions was 9-and 12-fold , respectively .
Under strictly anoxic growth-conditions , the expression of cob-lac fusions varied according to the electron-acceptor available to the cell ( Table 3 ) .
Maximum expression was observed with fumarate , followed by trimethylamine-N-oxide , dimethyl sulfoxide , and nitrate .
A 215-fold increase in the expression of the cobI-lac fusion was observed when we compared transcription level under stricly anoxic conditions on glycerol-fumarate to that measured under aerobic-growth-on-glucose .
This value is slightly higher than that seen in TABLE 2 .
Regulation by molecular oxygen of cob-lac fusions Beta-galactosidase activity '' during-growth on : Relevant genotype Glucose Glycerol + , -0 O + 0 , + Fumarate TT10852 cob-24 : : lac Cobl-4 20 10 430 TT10857 cob-62 : : 1 ac CobII-30 40 40 280 TT10858 cob-66 : : Iac CoblIl-20 30 30 240 a Values represent beta-galactosidase activity measured in early log phase cultures .
Culture conditions , and substrate concentrations are described under Materials and Methods .
Culture medium containing low amounts of oxygen was obtained by overnight degassing of the medium inside an anaerobic chamber whose atmosphere contained 95 % N2 , 5 % H2 , and 5 % CO2 .
No further efforts were made to remove residual oxygen from the medium .
Values shown represent the means of duplicate determinations .
A unit of activity is defined as nanomoles of a-nitrophenyl-p-D-galactoside per minute per A650 unit .
All strains carry a metE205 mutation .
Strain Phenotype of fusion TABLE 3 .
Expression of cob-lac fusions during fermentation and anaerobic respiration Beta-galactosidase activity '' during-growth on : Strain genotype Relevant Phenotype of fusion TMAOd DMSOe Glycerol ( R/F ) ( R/F ) 750 ( 18.8 ) 570 ( 14.3 ) 520 ( 10.4 ) 440 ( 8.8 ) 490 ( 9.8 ) 380 ( 7.6 ) Glucose Citrateb Fumarate Nitrate ( R/FI ) ( R/F ) 860 ( 21.5 ) 680 ( 13.6 ) 620 ( 12.4 ) 480 ( 12.0 ) 300 ( 6.0 ) 170 ( 3.4 ) TT10852 cob-24 : : lac CobI-40 110 TT10857 cob-62 : : Iac CobII-50 100 TT10858 cob-66 : : Iac CoblII-50 80 a Values represent the average of duplicate determinations of enzymatic activity .
b All cultures were grown in anoxic NCE medium supplemented with MgSO4 ( 10 mM ) and trisodium-citrate ( 40 mM ) .
c R/F is the ratio of units of enzyme activity per A650 unit measured during-growth under anaerobic respiration divided by the number of enzyme units per A650 unit measured during fermentative growth-on-glucose .
All strains carry an metE205 mutation .
All culture were grown in E medium under strictly anoxic conditions .
The concentration of electron-acceptors in the culture medium was 10 mM .
Concentrations of carbon sources in the culture medium : glucose ( 11 mM ) and glycerol ( 22 mM ) .
e DMSO , Dimethyl sulfoxide .
Table 2 because of traces of oxygen present in the early experiments ( see below ) .
The growth-rate of S. typhimurium under each set of conditions was glucose > glycerol-nitrate > glycerol-fumarate > glycerol-trimethylamine-N-oxide > glycerol-dimethyl sulfoxide ( data not shown ) .
We found no correlation between growth-rate and expression of the fusions .
Unlike growth of E. coli ( 22 ) , fermentative growth of S. typhimurium on citrate does not require the presence of a cosubstrate , and under these growth-conditions the expression of all the cob-lac fusions was higher than the expression of the same fusions during fermentative growth-on-glucose .
This , like the stimulation of transcription by aerobic-growth-on-glycerol ( Table 2 ) , suggested that cob genes might be subject to catabolite-repression .
The results reported in Table 2 and 3 suggested that cob genes might be subject to catabolite-repression and raised the further possibility that the respiration-fermentation differences might be mediated by changes in the cAMP levels .
To test the involvement of cAMP in cobalamin biosynthesis , we constructed strains carrying both cob-lac fusions and mutations affecting adenylate cyclase ( cya - ) , the cAMP-receptor-protein ( crp - ) , or both ( Table 4 ) .
The addition of cAMP to glucose-supplemented medium resulted in increased levels of transcription of all the fusions tested under both aerobic and anaerobic-growth-conditions .
Aerobically and anaerobically , a cobI - / ac cya + TABLE 4 .
Role of cAMP in the expression of cob-lac fusions Beta-galactosidase activity '' during-growth on : Glucose Glucose + cAMP Glycerol Strain Phenotype of fusion Relevant genotype +02 10 +02 10 -02 70 +02 - ° 2 20 6 6 20 40 50 50 40 30 40 40 30 + Fumarate TT10852 cob-24 : : lac Cobl-TT10853 cob-24 : : lac cya : : TnJO Cobl-TT10854 cob-24 : : lac crp-773 : : TnJO Cobl-TT11297 cob-24 : : lac crp * -771 cya : : TnIO Cobl-TT10857 cob-62 : : Iac CobII-TT10859 cob-62 : : Iac cya : : TnlO CobII-TT10861 cob-62 : : 1 ac crp-773 : : TnJO CobII-TT11295 cob-62 : : Iac crp * -771 cya : : TnJO CobII-TT10858 cob-66 : : lac CoblIl-TT10860 cob-66 : : Iac cya : : TnJO CoblIl-TT10862 cob-66 : : Iac crp-773 : : TnJO Coblll-TT11296 cob-66 : : lac crp * -771 cya : : TnJO CobIII-a One J3-galactosidase unit = 1 nmol of ONPG per min per A650 unit .
b NG , No growth 4 3 430 NG NG 350 280 NG NG 320 240 NG NG 290 NGb NG 10 40 NG NG 30 30 NG NG 30 40 10 50 90 80 80 70 70 60 40 60 20 4 20 50 50 40 40 30 40 20 30 3 4 30 30 40 30 20 30 30 20 Relevant genotype ef ta crp + strain showed a three-to fourfold increased transcription of the fusion in the presence of exogenous cAMP .
The cobI-lac cya : : TnJO double mutant expressed basal levels of transcription of the fusion in the absence of cAMP ; these levels were stimulated six-to sevenfold by the addition of cAMP ( Table 4 ) .
A cobl-lac crp : : TnJO double mutant showed the same behavior as the cobI-lac cya : : TnJO , except that the addition of cAMP failed to stimulate transcription of the fusion ( Table 4 ) .
These results suggested a strong cAMP effect on cob transcription .
To test whether cAMP mediates the stimulation of transcription seen during anaerobic respiration , a crp * mutation was tested .
This mutation alters the Crp protein so it can activate transcription even in the absence of cAMP ( 13 , 27 , 28 ) .
A strain was constructed that had a crp * mutation , a cob fusion , and a TnJO insertion in cya ( Table 4 ) .
In this strain , which can not make cAMP , the expression of the fusion was similar to that measured for a cobI-lac cya + crp + strain in either the presence or absence of molecular oxygen .
When glycerol was used as the source of energy and fumarate was substituted for molecular oxygen as the final electron-acceptor , a 44-fold increase in the expression of the fusion was recorded .
As expected , strains containing cya or crp mutations failed to grow on glycerol .
However , the strain carrying the cobl-lac , crp * , and cya : : TnJO mutations transcribed the fusion as efficiently as the crp + cya + cobI-lac strain during aerobic-growth .
The expression of fusions in strains harboring a coblI-lac or a coblIl-lac fusion and the abovementioned cAMP-related functions was not affected at all ( Table 4 ) .
These results demonstrate that stimulation of transcription under conditions of anaerobic respiration is not mediated by variation of the cAMP levels .
This conclusion is supported further by measurements of the expression of the fusions under anoxic conditions ( Table 5 ) .
Our results show that , although exogenous cAMP greatly stimulates the expression of fusions in cells growing fermentatively on glucose or citrate , we could not achieve the levels seen for cells growing under conditions of anaerobic respiration of glycerol and fumarate .
Assuming adequate transport of cAMP , we would expect higher enzyme levels if the induction were mediated by cAMP .
This can be seen when S. typhimurium ferments citrate .
Under these conditions , the presence of exogenous cAMP ( 5 mM ) stimulates the expression of all the cob-lac fusions , but even the most responsive cobI fusion reaches only 60 % of the level measured when cells respire glycerol-fumarate .
If the regulatory effect of oxygen were exerted through variations in the levels of cAMP alone , we would expect the addition of cAMP to result in higher levels of expression at a level near that seen for cells respiring glycerol-fumarate .
These experiments do not rule out the possibility that oxygen represses the synthesis of the Crp protein .
If this were the case , one might expect that anoxia would derepress Crp * levels and could thus stimulate expression of a Crp-dependent operon even in a cya mutant background .
The results described in Table 4 suggested a requirement for cAMP for transcription of cob genes .
Thus , it was predicted that strains of S. typhimurium with a metE cob ' genotype might require the cya and crp gene products to synthesize B12 and show methionine-independent growth anaerobically via the MetH enzyme .
This prediction was tested by growing appropriate strains carrying a metE mutation anaerobically in minimal-medium ; under these conditions cells must synthesize cobalamin to satisfy their methi-onine requirement ( through use of the cobalamin-dependent MetH enzyme ) .
The doubling times of anaerobically grown cultures are shown in Table 6 .
( i ) With no additions to the culture medium all the strains grew .
However , the cya and crp mutants grew 39 and 25 % slower than the wild-type strain , respectively .
( ii ) Addition of methionine to the me-dium did not increase the growth-rate of the cya and crp mutants .
( iii ) Exogenous cAMP increased the growth-rate of the cya mutant to that of the cya + strain .
( iv ) Exogenous cAMP had a deleterious effect on the growth-rate of the crp mutant , resulting in a decrease of almost 50 % in the growth-rate of this strain under that measured for the isogenic crp + strain .
Thus , the basal level of transcription of cob genes seen in the absence of crp and cya functions appears to be sufficient to provide cobalamin for methionine synthesis .
Beta-galactosidase activity ' during-growth on : e fuo usiv Gl cose Gl cose + cA P Citra eb Citr Glycerol sinotpnbuoufMyteate + cAMP lfumarate TT11297 cya : : TnJO crp * -771 cob-24 : : lac Cobl-40 110-90-480-810 TT11295 cya : : TnlO crp * -771 cob-62 : : Iac Cobll-50 80-110-200-660 TT11296 cya : : TnlO crp * -771 cob-66 : : lac CoblII-50 90-80-230-540 a Units of enzyme activity have been defined in Materials and Methods .
Cells were grown in anoxic E culture medium prepared as described in Materials and Methods .
b All culture were grown in anoxic NCE medium supplemented with MgSO4 ( 10 mM ) and trisodium-citrate ( 40 mM ) .
Nitrogen gas was the only gas present in the headspace of the culture tubes .
All strains carry a metE205 mutation .
Effect of cAMP on B12-dependent growtha Doubling time ( min ) anaerobically in the presence of : Strain Relevant genotype No addition L-Methionine cAMP TR6583 metE205 cob ' 60 65 62 TT10855 metE205 cob ' cya : : TnJO 98 102 62 TT10856 metE205 cob ' crp : : TnlO 80 85 115 TT11775 metE205 cob + crp * -771 cya : : TnJO 60 60 62 a Strains were grown low-oxygen in E medium supplemented-with-glucose ( 11 mM ) as the energy source .
The concentrations of added compounds in the culture medium : L-methionine ( 0.5 mM ) and cAMP ( 5 mM ) .
Growth conditions were described in Materials and Methods .
Cell growth was monitored with a Klett-Summerson photocolorimeter ( 540 nm ) .
Doubling times were obtained from plots of Ketts units versus time ( minutes ) The slight effect of cya and crp mutations on the growth-rate seem due to impairment of other metabolic functions required for anaerobic-growth .
Regulation by molecular oxygen .
Figure 1 illustrates the regulatory effect of oxygen on the of cob-lac fusions .
Expression was exquisitely sensitive to oxygen .
Providing oxygen at pressures greater than 2 kPa ( < 2 % of 1 atm of pure 02 , or about 10 % of atmospheric oxygen ) resulted in a sharp decrease in the transcription of the fusions .
To investigate whether this effect of oxygen was specific to fusions to cob operons , a control experiment was performed with a fusion to the gene encoding for nicotinic acid phosphoribosyl transferase ( pncB ) .
The results of this control experiment are also shown in Fig. 1 .
In this case , unlike the cob-lac fusions , increasing levels of oxygen resulted in slightly increased transcription of the pncB-lac fusion .
The above results explain the variability in the maximum levels of expression of the same fusions that was seen in early experiments ( Tables 2 , 4 , and 7 ) .
In later experiments ( Tables 3 and 5 ) , 02 has been excluded completely by boiling the culture medium under oxygen-free nitrogen gas ( see Materials and Methods ) ; that is , if dissolved oxygen is removed to different degrees from the culture medium , the expression of the fusions will vary accordingly .
To study the effect of cobalamin on the regulation of its biosynthetic pathway we measured the levels of beta-galactosidase synthesized during anaerobic-growth ( glycerol-fumarate ) in the presence or absence of cyanocobalamin in the culture medium .
Operon fusions in each of the three branches were tested ( Table 7 ) .
Transcription of the fusion in cobI was reduced sixfold ; expression of the cobII fusion was reduced approximately threefold ; and the expression of the cobIII fusion was reduced twofold .
Also shown in Table 7 is the effect of cobinamide , DMB , or cobinamide and DMB on the expression of cob-lac fusions .
The expression of cobI remained unaffected by the presence of DMB , but it showed some reduction in the presence of cobinamide .
The addition of cobinamide and DMB together resulted in a reduction of approximately 50 % .
This is interpreted to mean that B12 can be readily synthesized from these precursors and that of the can cobI repression operon occur .
Practically no effect of cobinamide or DMB individually on fusions to branches II and III was recorded .
Larger repressive effects of B12 have recently been seen ( Dan I. Andersson , unpublished results ) in cells grown anaerobically on glycerol-fumarate .
Role of biosynthetic intermediates in gene regulation .
The regulatory interactions between branches of the pathway might be mediated by levels of particular biosynthetic intermediates .
For this purpose strains were constructed which carried both the branch I cob-24 : : 1 ac fusion and a TnJO element inserted in either branch II or III .
No change was recorded in the amount of beta-galactosidase synthesized in either of the double mutants tested ( strains TT10874 and TT12231 ) ( data not shown ) .
Similarly , we tested the effect of cysG mutations .
A mutation in the cysG locus has a CobI-phenotype ( 19 ) , since the substrate for the branch I of the pathway is not made .
Thus , in a cysG mutant neither cobinamide nor any of its precursors should be present .
The regulation of fusions in any of the branches of the pathway remained unaffected when a cysG mutation was present in the background ( strains TT10863 , TT10864 , and TT10865 ) KPa PARTIAL 02 FIG. 1 .
Effect of 02 on the expression of cob-lac fusions .
Key : * , cob-24 : : lac ( cobI ) ; 0 , cob-66 : : Iac ( cobl ) h ; A , cob-62 : : lac ( coblI ) ; A , pncB-252 : : 1 ac .
All cultures were grown in E minimal-medium supplemented with glycerol ( 22 mM ) as the source of carbon and energy and fumarate ( 10 mM ) the final electron as acceptor .
The medium was made anoxic as described under Materials and Methods .
L-Methionine was added to a final concentration of 0.5 mM .
Oxygen ( air ) was injected into the sealed tubes after removal of an equivalent volume of gas from the headspace of the tubes prior to autoclaving .
Each point is the average of duplicate measurements .
Ideal gas behavior was assumed in determining the partial pressure of oxygen .
The rate of decrease in the expression of the fusions as a function of the partial pressure of oxgen was ( enzyme units per A650 unit per kilopascal of 02 ) : cob-24 : : lac , 150 ; cob-62 : : lac , 140 ; cob-66 : : lac , 135 .
Regulation by cobalamin of cob-lac fusions Strain Relevant genotype Phenotype of Beta-galactosidase activity ' in the presence of : fusion No addition DMB CBi CBi + DMB B12 TT10852 cob-24 : : lac Cobl-280 240-210-130 50 TT10857 cob-62 : : lac CobII-200 190-190-220 70 TT10858 cob-66 : : lac CoblIl 110-150-140-150 50 a Values represent the average of two separate determinations .
A unit of activity is defined under Materials and Methods .
Concentrations : DBM ( 3 x 10-4 M ) , cobinamide ( CBi ) ( 1.8 x 10-8 M , and cyanocobalamin ( B12 ) ( 1.5 x 10-8 M ) .
Cultures were grown under low-oxygen conditions on E medium containing glycerol ( 22 mM ) as the source of energy , and fumarate ( 10 mM ) as the electron-acceptor .
Enzyme assays were performed on early-log-phase cultures .
All the strains carry a metE20S mutation ( data not shown ) , suggesting that intermediates of the pathway have subtle ( if any ) regulatory effects on the synthesis of B12 .
The introduction of a mutation in the oxrA locus ( 20 , 37a ) into strains carrying cob-lac fusions had no effect on the transcription of the latter ( data not shown ) .
DISCUSSION Cobalamin biosynthesis in S. typhimurium is regulated by cobalamin and by molecular oxygen .
Our findings indicate that the regulation of the pathway is primarily exerted on cobI , i.e. , the biosynthesis of the corrin ring .
We have shown that transcription all the branches of the pathway is negatively controlled by cobalamin .
Our data suggest that cobalamin , and not biosynthetic intermediates , is responsible for the regulation of the pathway .
This conclusion is based on our results , which show that cysG mutants ( which fail to make the substrate for branch I of the pathway ) still show normal regulation of all the branches of the pathway .
Although it is largely unclear how the presence of molecular oxygen regulates the biosynthesis of this macromole-cule , it is clear that the transcription of cob genes is strongly inhibited by even low levels of 02 ( Fig. 1 ) .
However , there are further complications to the regulatory effect ( s ) by molecular oxygen .
This is clearly illustrated by our results , which show that anaerobiosis is most effective in stimulating the transcription of cob genes when the cell is respiring to an alternative electron-acceptor instead of fermenting .
This observation can be explained by suggesting that the transcription of cob genes is somehow entrained with the changes that occur when the cell shifts from fermentation to respiration , e.g. , oxidative phosphorylation , membrane potential , reducing conditions , level of cAMP , etc. .
On the one hand , we have shown that exogenous cAMP stimulates transcription of cob genes when the cell is fermenting glucose or citrate .
Although the addition of exogenous cAMP stimulates the transcription of the pathway when the cell is fermenting glucose or citrate , this increase is only a fraction , 14 and 60 % respectively , of the levels observed under anaerobic respiration of glycerol-fumarate ( Table 5 ) .
Thus , there must be other factors modulating the expression of this pathway .
Regulation in response to molecular oxygen is independent of intracellular cAMP levels .
This is seen in a strain which carries an inactive adenylate cyclase and can therefore make no cAMP .
This strain also carries a crp * mutation which allows transcription of cAMP-regulated genes or operons in the absence of the nucleotide ( 13 , 27 , 28 ) .
In this strain , which should be unable to modulate its cAMP effects , molecular oxygen still shows the full range of regulation of all branches of the pathway .
Also , it should be noted that although exogenous cAMP stimulates substantial transcription of the pathway under aerobic conditions , growth fails to occur under conditions that demand the biosynthesis of cobalamin .
In other words , functional B12 molecules are not synthesized , not even the minimal amounts needed to satisfy the methionine auxotrophy through the function of the MetH enzyme .
This suggests the existence of additional mechanisms whereby molecular oxygen prevents cobalamin biosynthesis or , alternatively , that anoxia is required to avoid the inactivation of 02-labile intermediate or enzymes ( or both ) of the pathway .
It seems likely that the stimulatory effect of cAMP on the biosynthesis of cobalamin under anaerobic conditions is a secondary consequence of the global metabolic effect on metabolism caused by this nucleotide especially under anaerobic conditions .
Our results shown in Table 6 support this idea .
The addition of methionine , which is the only auxotrophic requirement of strain TT10855 ( which carries cya : : TnJO ) , fails to reduce the doubling time of the culture .
Moreover , the sole addition of cAMP to the culture medium is enough to reduce the doubling time to wild-type behavior .
As expected this stimulatory effect ( s ) of cAMP is mediated by the crp gene product ( Table 6 ) .
The deleterious effect of cAMP on the anaerobic-growth of a crp strain is not understood at this point .
This result illustrates the importance of the role of cAMP in the anaerobic-growth of S. typhimurium .
Since all the strains deficient in the synthesis of cAMP or the Crp protein ( strains TT10855 , and TT10856 ) grew in the absence of exogenous cAMP or L-methionine , it follows that cAMP is not required for the synthesis of cobalamin .
In other words , cAMP may contribute to establishing the physiological conditions that favor cobalamin biosynthesis , but may not participate in the process directly .
It is important to notice that transcription levels of the cobI operon are quite low ( 20 U of beta-galactosidase activity ) under conditions of anaerobic fermentation of glucose , yet this low level is sufficient to satisfy the cells ' methionine biosynthetic requirement .
This suggests that rather little B12 is required for methionine .
This low level of transcription is also adequate to supply B12 for use of ethanolamine as a nitrogen source ( unpublished results ) .
This raises the question of why transcription increases to such high levels ( 800 U of beta-galactosidase activity ) under conditions of anaerobic respiration .
We suppose that under conditions of anaerobic respiration some unidentified functions exist that require B12 .
It seems that these functions are dispensable , since deletion mutants of S. typhimurium lacking the entire cobalamin biosynthetic pathway grow normally on glycerol-fumarate .
The results presented here suggest some clues to the metabolic importance of B12 .
The biosynthetic genes are very highly transcribed under conditions of anaerobic respiration , suggesting that the major value of B12 is realized under these conditions .
Although it is not clear whether there is a direct correlation between anaerobic respiration and cAMP levels , our data show that if the levels of this cyclic nucleotide are increased by providing an excess of it during fermentative growth , the level of transcription of cob genes approaches that observed when the cell is respiring anaerobically .
Assuming that cAMP signals a shortage of carbon source , under both aerobic and anaerobic conditions , the stimulatory effect of cAMP might be interpreted to mean that B12 is particularly valuable under conditions of limiting carbon sources .
We therefore suggest that the major value of B12 may prove to be in the catabolism of poor carbon sources that can only be utilized under conditions of anaerobic respiration .
We thank P. W. Postma and C. G. Miller for providing us with requested strains .
Research work of the authors is supported by Public Health Service grant GM-34804 to J.R.R. from the National Institutes of Winchell Cancer Fund Health and by Damon Runyon-Walter postdoctoral fellowship DRG-811 to J.C.E.-S .
Balch , W. E. , and R. S. Wolfe .
New approach to the cultivation of methanogenic bacteria : 2-mercaptoethanesulfonic acid-dependent growth of Methanobacterium ruminantium in a pressurized atmosphere .
Battersby , A. R. , M. J. Bushell , C. Jones , N. G. Lewis , and A. Pfenniger .
Identity of the fragment extruded during ring contraction to the corrin macrocycle .
Battersby , A. R. , and E. McDonald .
Biosynthesis of the corrin macrocycle , p. 107-144 .
In D. Dolphin ( ed .
John Wiley & Sons , Inc. , New York .
Battersby , A. R. , E. McDonald , M. Thompson , and V. Y. Bykovsky .
Biosynthesis of vitamin-B12 : proof of A-B structure for sirohydrochlorine by its specific incorporation into cobyrinic acid .
The clostridial fermentations of choline and ethanolamine .
I. Preparation and properties of cell-free extracts .
The clostridial fermentations of choline and ethanolamine .
Requirement for a cobinamide coenzyme by an ethanolamine deaminase .
Casadaban , M. J. , and S. N. Cohen .
Lactose genes fused to exogenous promoters in one step using Mu-lac bacteriophage .
In vivo probe for transcriptional control sequences .
Castilho , B. A. , P. Olfson , and M. J. Casadaban .
Plasmid insertion mutagenesis and lac gene fusion with mini-Mu bacteriophage transposons .
Davis , R. W. , D. Botstein , and J. R. Roth .
Cold Spring Harbor Laboratory , Cold Spring Harbor , N.Y. 11 .
Amidation of , and ( R ) -1-amino-2-propanol attachment to the corrin ring during vitamin-B12 biosynthesis by Clostridium tetanomorphum extracts .
Fyfe , J. H. , and H. C. Friedmann .
Garges , S. , and S. Adhya .
Sites of allosteric shift in the structure of the cyclic-AMP-receptor-protein .
Mechanisms of coenzyme B12-dependent rearrangements .
Holriegi , V. , L. Lamm , J. Rowold , J. A. Horig , and P. Renz .
Horig , J. A. , and P. Renz .
[ 5 - ' 5N ] Riboflavin as precursor in the biosynthesis of the 5,6-dimethylbenzymidazole moiety of vitamin-B12 .
Hughes , K. T. , and J. R. Roth .
Conditionally transposi-tion-defective derivative of Mu dl ( Ap lac ) .
Kung , H.-F. , C. Spears , R. C. Greene , and H. Weissbach .
Regulation of the terminal reactions in methionine biosynthesis by vitamin-B12 and methionine .
Jeter , R. M. , B. M. Olivera , and J. R. Roth .
Salmonella typhimurium synthesizes cobalamin ( vitamin-B12 ) de novo under anaerobic-growth-conditions .
Lambden , P. R. , and J. R. Guest .
Mutants of Escherichia coli K12 unable to use fumarate as an anaerobic electron-acceptor .
Lamm , L. , G. Heckmann , and P. Renz .
Lutgens , M. , and G. Gottschalk .
Why a co-substrate is required for anaerobic-growth of Escherichia coli on citrate .
Experiments in molecular genetics .
Cold Spring Harbor Laboratory , Cold Spring Harbor , N.Y. 24 .
Mombelli , L. , C. Nussbaumer , H. Weber , G. Muller , and D. Arlgoni .
Nature of the volatile fragment generated during formation of the corrin ring 25 .
Mulligan , J. T. , W. Margolin , J. H. Krueger , and G. C. Walker .
Mutations affecting regulation of the methionine bio-synthetic genes .
Isolation by use of met-lac fusions .
Nussbaumer , C. , M. Infeld , G. Worner , G. Muller , and D. Arigoni .
Biosynthesis of vitamin-B12 : mode of incorporation of factor III into cobyrinic acid .
Postma , P. W. , and J. W. Lengeler .
Phosphoenolpyruvate : carbohydrate phosphotransferase system of bacteria .
Postma , P. W. , and B. J. Scholte .
Regulation of sugar transport in Salmonella typhimurium , p. 249-257 .
In E. Quagliariello , F. Palmieri , S. Papa , and M. Klinkenberg ( ed .
) , Function and molecular aspects of biomembrane transport .
Elsevier Biomedical Press , Amsterdam .
Rasetti , V. , A. Pflatz , C. Kratky , and A. Eschenmosser .
Ring contraction of hydroporphinoid complexes .
Scarlet , F. A. , and J. M. Turner .
Microbial metabolism of amino-alcohols .
Rthanolamine catabolism in Escherichia coli and Klebsiella aerogenes .
A method of detection of phage mutants with altered transducing ability .
Schmieger , H. , and H. Backhaus .
The origin of DNA in transducing particles in P22-mutants with increased transduction frequencies ( HT-mutants ) .
Scott , A. I. , N. E. MacKenzie , P. J. Santander , and P. E. Fagerness .
Timing of the methylations steps between urogen III and cobyrinic acid .
Scott , J. F. , J. R. Roth , and S. W. Artz .
Regulation of the histidine operon does not require hisG enzyme .
Shoeman , R. , T. Coleman , B. Redfield , R. C. Greene , A. A. Smith , I. Saint-Girons , N. Brot , and H. Weissbach .
Regulation of methionine synthesis in Escherichia coli : effect of metJ gene product and S-adenosyl-methionine on the in-vitro expression of the metB , metL , and metJ genes .
Shoeman , R. , B. Redfield , T. Coleman , N. Brot , H. Weissbach , R. C. Greene , A. A. Smith , I. Saint-Girons , M. M. Zakin , and G. N. Cohen .
Regulation of the methionine regulon in Escherichia coli .
Shoeman , R. , B. Redfield , T. Coleman , R. C. Greene , A. A. Smith , N. Brot , and H. Weissbach .
Regulation of methioine synthesis in Escherichia coli : effect of the metJ gene product and S-adenosyl-methionine in the expression of the metF gene .
Strauch , K. L. , J. B. Lenk , B. L. Gamble , and C. G. Miller .
Oxygen regulation in Salmonella typhimurium .
Taylor , R. T. , and H. Weissbach .
In D. Boyer ( ed .
Academic Press , Inc. , New York .
Turner , S. P. D. , M. H. Block , Z.-C .
Sheng , S. C. Zimmerman , and A. R. Battersby .
Synthesis relevant to vitamin-B12 biosynthesis : synthesis of ( ± ) Faktor-I-octamethyl ester .
Uzar , H. C. , and A. R. Battersby .
Biosynthesis of vitamin-B12 : order of the later C-methylation steps .
Vogel , H. J. , and D. M. Bonner .
Acetylornithase of Escherichia coli : partial purification , and some properties .
Yeliseev , A. A. , A. V. Pushkin , N. I. Zaltseva , Z. G. Evstigneeva , V. Y. Bykhovsky , and W. L. Kretovich .
Glutamine as the source of amide groups in vitamin-B12 biosynthesis .