follows: hooc-13hz-c12hrcooh

ACETIC ACID OXIDATION BY ESCHERICHIA COLI: QUANTITATIVESIGNIFICANCE OF THE TRICARBOXYLIC ACID CYCLE'

H. EARLE SWIM AND LESTER 0. KRAMPITZDepartment of Microbiology, School of Medicine, Western Reserve University, Cleveland, Ohio

Received for publication September 30, 1953

Krebs (1937) reported that acetate and a num-ber of other substrates were oxidized under ana-erobic conditions by nonproliferating suspensionsof Eschrichia coli when fumarate was added asan oxidant. These observations resulted from astudy of the role of fumarate in respiration, andno consideration was given to the mechanism ofacetate oxidation. The over-all reaction for theoxidation of acetate is illustrated in equation (1).CH3-COOH + 4HOOC-CH==CH-COOH+ 2H20 -* 4HOOC-CH.-CH2-COOH (1)

+ 2CO2The present studies were undertaken to deter-mine which mechanism or combination of mecha-nisms accounts for acetate oxidation in the pres-ence of fumarate. The following posibilities wereconsidered: (1) the direct oxidation of acetate tocarbon dioxide, (2) the oxidative condensation oftwo moles of acetate to yield succinate (Thun-berg condensation), and (3) the tricarboxylic acidcycle.The results of experiments with acetate-1-C4

showed that most of the isotope was located inthe succinate and that the residual fumarate andrespiratory carbon dioxide were essentially un-labeled. These data eliminated the possibilitythat acetate was oxidized directly to carbon di-oxide; thus, the most probable mechanism wasthe tricarboxylic acid cycle, but the possibilityremained that some of the acetate was metabo-lized by a Thunberg type condensation.

In the conventional tracer experiment, it isusually impossible to differentiate between suc-cinate that is formed from acetate via the tri-carboxylic acid cycle and that formed by a Thun-berg condensation. Both mechanisms give rise tocarboxyl labeled succinate from carboxyl labeledacetate and methylene labeled succinate from

1 This work was supported by a grant (Contractno. AT(30-1)-1050) from the Atomic Energy Com-mission. The radioactive isotope used in thesestudies was obtained on allocation from the AtomicEnergy Commission.

methyl labeled acetate. The succinate derivedfromC"H3,-COOH by a methyl to methyl condensa-tion would be labeled HOOC-Cl3H2--C"H--COOH (type a), whereas from the tricarboxylicacid cycle the labeling is HOOC-C3Hr Cl2Hr.COOH (type b) if the C4-acid condensing withthe acetate is unlabeled. This statement is valid,however, only for conditions under which re-cycling of succinate does not occur. In the courseof recycling, there is an equal chance for thelabeled methylene carbon of the succinate to be-come the carbonyl carbon of oxalacetate and,therefore, give rise to doubly labeled succinate(type a). The succinate produced under theseconditions would be a mixture of both types andthe relative quantities of each would depend onthe extent to which recycling occurs. Under theseconditions, type a succinate would be producedby both acetate condensation and the tricar-boxylic acid cycle. By employing fumarate as anoxidant, it has been found that recycling is virtu-ally eliminated, and, therefore, doubly (type a)and singly (type b) labeled succinate each wouldbe derived from separate pathways of acetatemetabolism. In addition to the labeled succinate,the mixture contains unlabeled succinate formedby the reduction of fumarate. It is apparent that,on the basis of chemical degradation and deter-mination of the C14 or C" content in the differentpositions, it is impossible to distinguish betweenthe individual types of succinate except on thebasis of concentration of isotope. Usually, as inthe present case, the C'4-succinate is diluted bycarbon from an unlabeled source (from fumarate),and, thus, the concentration of isotope in themethylene carbons does not exceed the criticalvalue in which one carbon is labeled.

It is possible to distinguish between the twoisotopic types (a and b) by measuring the massabundance of the molecules per se (Wood, 1952).The types of succinate to be considered are asfollows: HOOC-13Hz-C12HrCOOH_H2C"3=C"3H2, mass = 30; HOOC-C3Hz-CuH, COOH _ H2C'=C'2H2, mass 29;

426

SIGNIFICANCE OF KREBS CYCLE IN E. COLI

HOOCC12Hr-C12Hr-COOH -- H2C12=C'2H2,mass = 28. In order to obtain the molecules ingaseous form and to eliminate the mass effect ofthe carboxyl groups, the succinate was convertedto ethylene in such a manner as to obtain themethylene carbons as ethylene. It is possible bymass spectrometry to determine the respectiveamounts of doubly and singly labeled ethyleneand unlabeled ethylene in a mixture and, there-fore, the relative quantities of the correspondingspecies of succinate. This has been done in thepresent experiments, and the results showed thatno doubly labeled succinate (type a) is formedfrom acetate.

MATERILS AND METHODS

E. coli, strain E-26, was cultured and harvestedas previously described (Swim and Krampitz,1954a). Freshly harvested cells were used in allexperiments. The usual Warburg technique wasused, and the isotope experiments were carriedout in 125 ml Warburg vessels containing 10 mlof reaction mixture. The reactions were termi-nated by the addition of sulfuric acid to pH 1.5,the cells removed by centrifugation, and the or-ganic acids determined by partition chroma-tography on celite (Swim and Krampitz, 1954a).The various acid fractions were collected andoxidized to carbon dioxide by chromic acid (VanSlyke and Folch, 1940). The Cu4-carbon dioxidewas precipitated as barium carbonate and as-sayed for radioactivity with an end window G-Mtube counter. Self-absorption corrections wereapplied. When C'3 was employed, the carbondioxide and ethylene were analyzed in a Niertype mass spectrometer (Nier, 1947), and theresults are reported in atoms per cent C's.

Acetate was degraded according to Phares(1951). Succinate was degraded by using the fol-lowing series of reactions:

HOOC-C*Hr-C*H2-COOH M. lactilyticus

C*Hs- C*H2-COOH HN)

C*H3 C*2 NH2 HCHO, HCOOH

CH.31gO(CH)S(C2*H5)N H (CH3)3(C2*HI)NI >

(CHs)3(C2*Hs)NOH 1 C

H2C*=C*H2 + (CH13)8N + H20

The succinate was decarboxylated by Micrococcuslactilyticus (Swim and Krampitz, 1954b) and theresulting propionate converted to ethylamine, asdescribed by Phares (1951). The ethylamine wasisolated as the hydrochloride and converted todimethylethylamine hydrochloride by heating ina sealed tube with formaldehyde and formic acidaccording to the Eschweiler-Clarke (Eschweiler,1905; Clarke et al., 1933) modification of theLeuckart reaction. The excess formaldehyde andformic acid were removed by distillation in vacuo.The dimethylethylamine hydrochloride was dis-solved in water, the pH adjusted to 12 with so-dium hydroxide, and the amine distilled into atrap containing ether and cooled in a dry ice-cellusolve mixture. The ether solution of di-methylethylamine was dried over sodium hydrox-ide pellets and distilled into a trap containingan equivalent quantity of dry methyl iodide.The mixture was warmed to room temperature,allowed to stand for 6 hours, and the trimethyl-ethylammonium iodide was removed by filtra-tion. Ethylene was obtained from the trimethyl-ethylammonium iodide and purified for massspectrometer analysis, as described by Wood(1952). A portion of the ethylene was oxidized tocarbon dioxide by passing through a tube con-taining copper oxide at 350 C.The acetate-1-C'4 was synthesized as described

by Sakami et al. (1947). Acetate-2-C'3 was syn-thesized from methyl iodide prepared from C"-carbon dioxide (Nystrom et al., 1948; Little andBloch, 1950).

RESULTS

Manometric studies. The rates at which carbondioxide is produced from lactate, acetate, anda-ketoglutarate by E. coli, under anaerobic con-ditions in the presence of fumarate, are imlustratedin figure 1. The results obtained with citrate plusfumarate are the same as for fumarate alonewhich indicates the former compound is notmetabolized.

Oxidation of acetate-l-&14. According to equa-tion (1), it would appear that acetate is oxidizedto carbon dioxide with the concomitant reductionof fumarate to yield succinate. Quantitative ex-periments were performed with acetate-i-C14 andunlabeled fumarate in order to test the possi-bility of direct oxidation. The results of a typicalexperiment are shown in table 1. The respiratorycarbon dioxide was not labeled to a significant

427-1954]

H. EARLE SWIM AND LESTER 0. KRAMPITZ

Figicoli inactantpensio6.8, 10phase,

Oxidat

AcetalSuccinFumaMalatRespir

and also show that there can be little recyclingof succinate via either the tricarboxylic or di-carboxylic acid cycles. The latter conclusion issupported by the low specific activities of theresidual fumarate and malate which indicatesthat succinate is not converted to these acids tol. Fumarate + loctot any appreciable extent. The fumarate and malate

3. Fumarate +cketoglutarie were labeled exclusively in the carboxyl carbons,4. Fumarate and, therefore, the specific activity of the indi-

vidual carboxyl carbons was 20.5 and 23.0 cpmper um. Judged on this basis, it seems likely thatthe carbon dioxide was derived from the carboxylcarbons of these acids and that the fumarate andmalate were in isotopic equilibrium with the in-tracellular acids which were being metabolized.The C14-activity of the original acetate was di-

0 30 60 90 120 luted 47 per cent which indicated that part of the

MINUTES fumarate was converted to acetate. The quantityof carbon dioxide produced was in excess of the

ure 1. Anaerobic oxidation by E cherichia acetate utilized due to the endogenous formationthe presence of fumarate. Volulme of re- of crodixe. Siial,frec irml

ts 2.0 ml containing 0.5 ml 10 per cent sus- .. .f

n of E. coli, 100 Am phosphate buffer, pH of acetate utilized, 4.8 micromoles of succinate0 pm sodium fumarate, 10 jAm substrate. Gas were produced. Equation (1) represents the theo-helium. Temperature, 30 C. retical and does not consider the fumarate that

is reduced by endogenous reactions. Part of theTABLE 1 fumarate was converted to malate by the activeTABLE1 fumarase of E. coli. When fumarate was replaced

tion of acetate-i-C14 by Escherichia coli in with malate, identical results were obtained.the presence of fumarate Although the results showed that acetate was

S -P:R E not oxidized to carbon dioxide, it was consideredCOXPOUND QUANTITY ACTIVITY OF ADDED possible that part of the acetate could be utilized

c by a Thunberg condensation and that in some;IN cpm PerON way the oxidation of fumarate to carbon dioxide

te......... 24.3 866 12.7 was linked with the utilization of acetate.2 It waslate......... 370 362 81.2 possible, however, to determine the extent ofrate......... 31.8 41 0.8 Thunberg condensation by mass analysis of thee......... 57.6 46 1.6 succinate formed from acetate-2-C03 since thereratory C02.- 162 23 2.2 is virtually no recycling in this system.

Volume of reactants, 10 ml containing 3.0 ml10 per cent suspension of E. coli, 0.9 mm potas-sium phosphate buffer, pH 6.8, 500,M sodiumfumarate, 100 pm sodium acetate-l-Ct4 (1,650 cpmper umM); 2.0 ml 3 N sodium hydroxide in centerwell and 4 mm of sulfuric acid in side arm. Gasphase, helium. Temperature, 30 C. Time of incu-bation, 2 hours.

extent, and 95 per cent of the isotope removedfrom the acetate pool was located in the succinate.These data eliminate the possibility that acetateis oxidized directly to carbon dioxide and water

2 The following is an example for purposes ofillustration only: (a) 2 fumarate -* 2CO2 +"C2" + succinate; (b) 2 acetate + "C2" -- "C2-H2" + succinate; (c) fumarate + "C2-H2" -."C6"; (d) "0w" + 5 fumarate -. 2CO2 + 6 suc-cinate.

8 fumarate + 2 acetate -- 4CO2 + 8 succinate

The "C2" might be a specific hydrogen acceptorfor the Thunberg condensation. The "C2-H2"but not the "C2" might be specific for the conden-sation reaction to yield the "Ce" which is oxidizedto carbon dioxide and succinate.

428 [VOL. 67


Changes in the specific activity of the respiratorycarbon dioxide with time. In order to obtain fur-ther information on the question of recycling,acetate-i-C1' was oxidized in the presence of un-labeled fumarate, and the specific activity of therespiratory carbon dioxide was determined atvarious time intervals during the incubation pe-riod. Acetate-i-C1' is converted to carboxyl la-beled succinate which upon oxidation would giverise to fumarate with the corresponding isotopedistribution. Since the carboxyl carbons of fu-marate are precursors of the carbon dioxide, theC14 content of the latter is indicative of the extentto which succinate oxidation occurs. The resultsof this experiment are presented in table 2. Thesuccinate formed from acetate is not oxidized toan appreciable extent as shown by the low spe-cific activity of the respiratory carbon dioxide.Of particular interest is the fact that the C14content of the respiratory carbon dioxide did notincrease after 15 minutes which indicates thatsuccinate is not oxidized after this time. It ap-pears, therefore, that the system contains a smallquantity of oxygen or some other oxidant whichis capable of oxidizing succinate, and after thisis reduced, no further oxidation of succinate oc-curs.Mass analysis of types of succinate. There are

three equations which may be set up to charac-terize the mixture of succinate. These equationswere derived by Wood (1952) and tested by com-paring experimental values with known valuesobtained from mixtures of synthetic ethylene.The equations are listed here for purposes of

TABLE 2

Changes in the specific activity of respiratory carbondioxide with time

CARBON DIOXIDMTIME INMINUTS

Quantity Specific activity

pM Cm PeF pm

15 13.3 15.830 26.9 16.760 76.6 16.4120 196 17.0

Volume of reactants, 10 ml containing 2.5 ml10 per cent suspension of Escherichia coli, 1.0mm phosphate buffer, pH 6.8, 400 iM sodium fu-marate, 100 pM sodium acetate-i-C4 (1,650 cpmper pm); 2 ml 3 N sodium hydroxide in center welland 4 mm of sulfuric acid in side arm. Gas phase,helium. Temperature, 30 C. Duplicate flasks wereemployed for each analysis.

of C" in the methyl carbon of acetate; the labeledsource (atoms per cent C" per 100).The procedure used in studying the type of

succinate formed from acetate-2-C3 may be sum-marized as follows. The succinate was convertedto ethylene, and the ratios H2C0=Cl3H=/H2C"=C'2H2 and H203-C2H2/H2C02C02Hwere calculated from the relative abundance ofmases 30, 29, and 28, as determined by the massspectrometer. In addition, the average per centC3 in a combusted sample of ethylene was meas-ured. The average F was calculated from theinitial and final concentration of C13 in the ace-

(H2CI3--Cl32/H2C'==CnH100 -100F'D + FS(H2C1BC1H2/H,02=C1H2) 100 (1- F)2D + 0.99(1- F)S + 0.98(1- D -S)

(H,12==C13H,/H,Cll=ClSH,)100 = 200F(1 - F)D + (1 - F)S + 99FS + 1.98(1 - D -S)(H2lA---C3H3H2C%==IM000 a -(1 - F)'D + 0.99(1 - F)S + 0.98(1 - D - 5)

Average atoms per cent C" in ethylene2FD+FS+ 0.01S+ 0.02(1-D-S) X 100 (4)

2

clarity, and the following symbols are employed:D fraction of ethylene in the mixture derived

from succinate formed by acetate condensation(double labeling process). S = fraction of ethy-lene in the mixture derived from succinate formedby the tricarboxylic acid cycle (single labelingproce). N = fraction of ethylene in the mixturederived from succinate formed by the reductionof fumarate (nonlabeling process). F = fraction

tate. The values D and S were calculated then byequations (2) and (3) or (2) and (4). The valueforN was obtained by subtracting the sum of Dand S from unity.

Oxidation of acetate-2-Cu. The experiment was

conducted in the type of flask previously de-scribed (Swim and Krampitz, 1954a), and a pro-tocol appears beneath table 3. The respiratorycarbon dioxide was collected in a sodium hydrox-

(2)

(3)

19541 429


TABLE 3Oxidation of acetate-3-C03 by Escherichia coli in

the presence of fumarate

COMPOUND QUANT CcoISC NT

mx per cent

-CH:--CH2;c 7.98 5.84*Succinate

-COOH 7.98 1.26Acetate -CH3 1.07 39.0Fumarate 0.58 1.48Malate 1.02 1.42Respiratory C02 3.80 1.13

* C1 content of the carbon dioxide obtained bycombusting a sample of ethylene (cf Methods).

Volume of reactants, 200 ml containing 25 ml ofa 20 per cent suspension of E. coli, 20 mm phos-phate buffer, pH 6.8, 10.0 mm of sodium fumarate,2.6 mm of sodium acetate-2-Cl0 (54.4 per centC1'). Gas phase, helium. Temperature, 30 C.

All the reactants except acetate-2-C0' weremixed and incubated for 20 minutes. Acetate-2-C's was added then and the fermentation con-tinued for an additional 80 minutes and termi-nated by adding 10 mm of sulfuric acid.

ide bubbler attached to the apparatus. The re-sidual acids were separated on a 40 g celite col-nmn, degraded, and assayed for CU content. Thegeneral experimental data are shown in table 3and are esentially the same as for the 014 experi-ments in table 1. The fumarate contained 0.36per cent excess C0 (1.45 - 1.09) which indicatesthat the incorporation of isotope into this com-pound is not eliminated by allowing the mixtureto incubate for 20 minutes prior to the additionof acetate-2-0C (ef protocol, table 3).The average CU content of the methyl carbon

of acetate during the experiment was 46.7 per

cent (.4 2 39.0) As was pointed out by Wood

(1952), it is desirable in this type of experimentto maintain the isotope content of the sourcecompound (acetate) as nearly constant as pos-sible. The C1 content of the original acetate wasdiluted 28 per cent as compared with 40 to 50per cent for the 014 experiments (table 1). Theamount of dilution was reduced by increasingthe acetate to fumarate ratio and incubating themixture for a shorter period of time. It was notpractical to reduce the isotope dilution, still fur-ther, by using more acetate because of the limited

TABLE 4Relative abundance of doubly and singly labeled

succinate from acetate-20-C"s

RELATIV ABUNDANCE EQUA- SOURCZS OF SUCCINATECOMPARED TO TIONS (ETHYLENET)HC1-=Cl3H2 USED

lFOR

M,CI= H2C'== CALCU- C13:=C18 C12=Cl' CIr=:CICl'H CUHM LATIONS D S N

per cn per cen pe cen per cenw per cent

0.15 13.12 2, 3 0.20 21.2 78.6

2, 4 0.18 21.1 78.7

supply and considerable cost of carbon with sucha high C13 content.Mass analysis of types of succinate formed from

acetate-S-C"3. The calculated values for the rela-tive abundance of doubly and singly labeled andunlabeled succinate are shown in table 4. Thefigures for the relative abundance of H2C"2=C"H2 and H2C13=C"2H2 relative to H2C2=Cl2H2,as determined by the mass spectrophometer,were corrected for fragmentation (Stephenson,1951; Wood, 1952). The calculations (equations2 and 3) indicate that 21.2 per cent of the suc-cinate was derived from acetate by a single label-ing process and 0.2 per cent by a double labelingprocess. The remaining 78.6 per cent was pro-duced by the reduction of fumarate, a nonlabelingprocess. It is apparent from these calculatedvalues that, for every 21.4 (21.2 + 0.2) mole-cules of succinate which are derived from acetate,21.2 are singly labeled and 0.2 is doubly labeled.The fact that approximately one per cent of thelabeled succinate is doubly labeled does not dem-onstrate synthesis from acetate per se. The fuma-rate contained 0.36 per cent excess C"3 which indi-cates that sufficient recycling occurs to accountfor all of the doubly labeled succinate. On thisbasis, it is concluded that acetate condensationto succinate does not occur under these conditionsand that the tricarboxylic acid cycle accountsquantitatively for the metabolism of acetate andfumarate.The data presented in tables 1 and 3 show that,

if acetate is metabolized via the tricarboxylicacid cycle, between 4.5 and 5 moles of succinateare formed for each mole of acetate utilized, de-pending on the conditions of the experiment.Thus, 1 out of 4.5 to 5 of the succinates wouldcontain acetate carbon (20 to 22 per cent), and

430 [VOL. 67


78 to 80 per cent would be formed by a nonlabel-ing process through the reduction of fumarate.The results obtained by measuring the relativeabundance of the molecules per se (table 4) arein excellent agreement with the values predictedfrom carbon balance studies.

DISCUSSION

The present studies are in complete agreementwith the earlier work (Swim and Krampitz,1954a) which showed that acetate oxidation byE. coli occurs via the tricarboxylic acid cycle. Theoxidation of acetate in the presence of fumaratemay be considered to proceed as follows. Theprimary step involves a dismutation of fumarateto yield oxalacetate and succinate. The oxalace-tate condenses with acetate to yield citrate, andthe latter is converted oxidatively to succinatewith the removal of two carbons, as carbon di-oxide, with the concomitant reduction of therequired quantity of fumarate. When no acetateis added, the oxalacetate produced from fuma-rate has only endogenous acetate to condensewith, and the over-all reaction occurs at a slowrate. In addition, oxalacetate is a well known in-hibitor of succinic dehydrogenase (Pardee andPotter, 1948) and in the absence of acetate couldconceivably accumulate in a sufficient quantityto inhibit the reduction of fumarate. Accordingto this series of reactions, the respiratory carbondioxide is equivalent to the acetate utilized andis formed from tertiary carboxyl carbon of citrateand the alpha carboxyl of a-ketoglutarate whichare derived from the carboxyl carbons of fuma-rate-malate. In the absence of an oxidant with asufficiently high oxidation potential to acceptelectrons from succinate, acetate oxidation doesnot proceed beyond this point. In such a system,where recycling does not occur, all of the succi-nate synthesized from acetate-2-C03 is singly la-beled, and, therefore, the tricarboxylic acid cycleaccounts quantitatively for acetate utilization.The Thunberg type condensation of acetate toyield succinate is excluded as an additional mech-anism for acetate oxidation since doubly labeledsuccinate was not formed. The mass analysistype of experiment permits not only detection ofthe two types of succinate but also the quantita-tive significance of each reaction leading to theirformation. Such data are not provided by theusual tracer experiments (Swim and Krampitz,1954a).

Isotope from the carboxyl carbon of acetatedoes not appear in the respiratory carbon whichconfirms the Ogston hypothesis (Ogston, 1948) asapplied to citrate. Kalnitsky et al. (1943) re-ported that cell-free extracts of E. coli producedsuccinate from pyruvate by carbon dioxide fixa-tion. When acetate--3Cu was added to this sys-tem, a small amount of isotope was found in thecarboxyl carbons of succinate, and the C13 con-tent of the residual carbon dioxide was not inexcess of the normal value. Acetate oxidation inthe presence of pyruvate, under conditions wherecarbon dioxide fixation occurs, is analogous tothe fumarate system, and their data are explainedreadily on the basis of the tricarboxylic acid cycle.The evidence presented by Ajl and Kamen

(1951) and Ajl (1951a, b) for the occurrence ofacetate condensation in E. coli, strain E-26, isbased primarily on the results obtained from iso-tope carrier experiments. Objections to the valid-ity of this type of evidence have been noted pre-viously (Swim and Krampitz, 1954a).

It is generally accepted that the tricarboxylicacid cycle is the main pathway of acetate oxida-tion in animal tissues. This conclusion is based,in part, on the rate at which intermediates of thecycle are metabolized and on the results obtainedby employing malonate as an inhibitor (Krebs,1943). The results of isotope studies also indicatethat this pathway is of considerable quantitativeimportance (Wood, 1946). Many microorganisms,on the other hand, do not satisfy the criteria onwhich the cycle was based in animal tissues. Thishas led a number of investigators to conclude thatthe tricarboxylic acid cycle either does not occuror is of little quantitative importance in thesemicroorganisms (cf Ajl, 1951b and Krebs et al.,1952, for reviews).

It seems likely that rates of oxidation of inter-mediates and malonate inhibition may not bereliable criteria for excluding the cycle. For exam-ple, in the present studies the resting-cell sus-pensions of E. coli oxidize acetate at a rapid ratewhen either fumarate or oxygen is employed asan oxidant, whereas a-ketoglutarate is oxidizedat a slower rate and citrate is not metabolized;also, malonate does not effectively inhibit ace-tate oxidation (Lenti, 1946). Citrate and a-keto-glutarate, however, have been shown to be inter-mediates (Swim and Krampitz, 1954a), and thepresent results indicate that the tricarboxylicacid cycle accounts quantitatively for acetate

4311954]


oxidation. Permeability is an important factor inthe utilization of citrate by E. coli since driedcells (Lara and Stokes, 1952) and cell-free prepa-rations (Swim and Krampitz, 1952) oxidize thiscompound at an appreciable rate. However, theuse of treated cells to evaluate the role of the tri-carboxylic acid cycle in normal cells is extremelyhazardous because enzymes may be inactivatedpartially or completely and, thus, make it impos-sible to compare the metabolic activities oftreated and untreated preparations. There areindications, however, that in addition to perme-ability other factors may be involved in the me-tabolism of citrate by E. coli. For example, citrateis utilized by growing cells in a medium contain-ing glucose (Vaughn et al., 1950) and by restingsuspensions when grown on a citrate-glucose me-dium (Szulmajster et al., 1952). As yet we do notknow the complete picture of active intermediatesof the tricarboxylic acid cycle. Just as it is impos-sible to compare the rates of metabolism of indi-vidual components of anaerobic glycoly&s unlessall phosphate and electron acceptors and donorsare provided in optimal concentrations, it may beimpossible likewise to compare citrate utilizationwith that of acetate unless intermediary acceptorsare mtained. Malonate inhibits succinic dehy-drogenase, but it is conceivable that it may not beeffective on a dehydrogenase which converts suc-cinyl-coenzyme A to fumaryl-coenzyme A. It isclear that observations based on rates and onmalonate inhibition must be interpreted withconsiderable reserve.

Objections could be raised against the presentstudies on the basis that acetate oxidation, in thepresence of fumarate, may not be comparable tothe aerobic system. There is no reason to believethat acetate condensation could occur underaerobic conditions and not under anaerobic con-ditions in the presence of fumarate. For example,the calculated E' for acetate condensation is-0.109 volts (Barron et al., 1950), whereas theE' for the fuimarate systems is 0.00 volts (Kalc-kar, 1941). If acetate condensation occurs aerobi-cally, it would be necessary to postulate a ratherspecific electron transfer mechanism which iscapable of transferring electrons to oxygen butnot to fumarate. At the present time we are notaware of any data which support such a proposal.The metabolic patterns of nonproliferating sus-

pensions of cells, however, do not necessarily re-

flect those operating during growth. Certainstrains of E. coli will grow on a medium contain-ing acetate as the sole source of carbon, and,therefore, the tricarboxylic acid cycle may notaccount for all the acetate metabolized underthese conditions. The findings of Cutinelli et al.(1951), who studied the synthesis of amino acidsbyE. coli grown on a medium in which ClSI3HC14-OO was the only source of carbon, are in accordwith theview that the tricarboxylic acid cycle sup-plies the carbon skeletons in the synthesis of thesecompounds. This would require, in addition tothe tricarboxylic acid cycle, a net synthesis ofC4-acids from acetate by a mechanism which isunknown. Foster et al. (1949) have shown thatRhisopus nigrVcans can convert ethanol to fuma-rate (65 per cent yield) and have concluded thatthe latter is synthesized by a direct condensationof C2 units. Utter and Wood (1951) have pointedout, however, that a beta condensation of a C2with a Ci formed from the methyl carbon of theethanol coupled with carbon dioxide fixationcould account for the fumarate synthesis as wellas the isotope pattern obtained when C14-ethanolwa employed. It is clear that the finding of anisotope pattern which fits the Thunberg con-densation or that certain bacteria can utilizeacetate as the sole source of carbon for growth isnot sufficient to prove this mechanism of con-densation or a dicarboxylic acid cycle. It is pos-sible that a major part of the oxidation is via thetricarboxylic acid cycle after the C4-acid isformed. In view of the present results which indi-cate that a Thunberg condensation does notoccur, the possibility that a Ci + C2 condensa-tion may play a role in the synthesis of C4-acidsby growing cultures of E. coli is attractive.

ACKNOWLEDGMENTS

The authors wish to express their sincerethanks to Dr. H. Friedman for the mass pec-trometer analyses and to Dr. H. G. Wood for hisvaluable discussion and criticism of this work.

The anaerobic metabolism of acetate by Esche-richia coli, in the presence of fumarate as anoxidant, was studied with the aid of labeled ace-tate. Experiments with acetate-i-C1 showed thatsuccinate was an end product of the oxidationand that the respiratory carbon dioxide was not

432 [voL. 67


labeled. The latter was derived from the carboxylcarbons of fumarate and was equivalent quantita-tively to the acetate utilized. By mass analysisof the succinate formed from acetate-S2Cu, it waspoWible to distinguish between succinate synthe-sized from acetate via the tricarboxylic acid cycleand that formed by oxidative condensation(Thunberg condensation). To accomplish this,the succinate was degraded in such a manner asto obtain the methylene carbons as ethylene. Theethylene was subjected to mass spectrographicanalysis and found to contain the mass types cor-responding to singly and nonlabeled succinate.Doubly labeled ethylene (succinate) was not pres-ent in excess of the normal value. These data arepresented as evidence that the tricarboxylic acidcycle accounts quantitatively for acetate oxida-tion by nonproliferating suspensions of this or-ganism and that a methyl to methyl condensa-tion of acetate to yield succinate does not occur.

REFERENCES

AJL, S. J. 1951a Studies on the mechanism ofacetate oxidation by bacteria. V. Evidencefor the participation of fumarate, malate,and oxalacetate in the oxidation of aceticacid by E8cherichia coli. J. Gen. Physiol.,34, 785-794.

AJL, S. J. 1951b Terminal respiratory patternsin microorganisms. Bact. Revs., 15, 211-244.

AJL, S. J., AND KAMEN, M. D. 1951 Studies onthe mechanism of acetate oxidation by Esch-erichia coli. J. Biol. Chem., 189, 845-857.

BARRON, E. S. G., ARDAO, M. I., AND HEARON,M. 1950 The mechanism of acetate oxida-tion by Corynebacterium creatinovorans.Arch. Biochem., 29, 130-153.

CLARKE, H. T., GILLESPIE, H. B., AND WEMISSHAUS,S. Z. 1933 The action of formaldehyde onamines and amino acids. J. Am. Chem.Soc., 55, 4571-4587.

CTUTINELLI, C., EHRENSVIRD, G., REIo, L.,SALUSTE, E., AND STJERNHOLM, R. 1951Acetic acid metabolism in Escherichia coli.I. General features and the metabolic connec-tion between acetate and glutamic acid, as-partic acid, glycine, valine, serine and thre-onine. Acta Chem. Scand., 5, 353-371.

ESCHWEILER, W. 1905 Ersatz von an Stickstoffgebunden Wasserstoffatomen durch dieMethylgruppe mit Hiilfe von Formaldehyd.Ber. deut. chem. Ges., 38, 880-882.

FOsTER, J. W., CASON, S. F., ANTHONY, D. S.,DAviS, J. B., JEFFERSON, W. E., AND LONG,M. V. 1949 Aerobic formation of fumaricacid in the mold Rhizopus nigricane: Syn-thesis by direct C,2 condensation. Proc.Natl. Acad. Sci. U. B., 35, 663-672.

KALCKAR, H. M. 1941 The nature of energeticcoupling in biological syntheses. Chem.Revs., 28, 71-178.

KALNITSKY, G., WOOD, H. G., AND WERKMAN,C. H. 1943 C02-fixation and succinic acidformation by a cell-free enzyme preparationof Escherichia coli. Arch. Biochem., 2, 269-281.

KRzns, H. A. 1937 The role of fumarate in therespiration of Bacterium coli commune. Bio-chem. J. (London), 31, 2095-2124.

KKMBs, H. A. 1943 The intermediary stages inthe biological oxidation of carbohydrate.Advances in Enzymol., 3, 191-252.

KSEBS, H. A., GUIN, S., AND EGGLESTON, L. V.1952 The pathway of oxidation of acetate inbaker's yeast. Biochem. J. (London), 51,614-628.

LARA, F. J. S., AND STOKES, J. L. 1952 Oxida-tion of citrate by Escherichia coli. J. Bact.,63, 415-4.

LENTI, C. 1946 Oxidation of pyruvic acid byBacterium coli communis. Boll. soc. ital.biol. sper., 20, 532-533.

LITmE, H. N., AND BLOCH, K. 1950 Studies onthe utilization of acetic acid for the biologicalsynthesis of cholesterol. J. Biol. Chem.,183,33-46.

NIER, A. 0. 1947 A mass spectrometer for iso-tope and gas analysis. Rev. Sci. Instr., 18,398-411.

NYSTROM, R. F., YANKO, W. H., AND BROWN,W. G. 1948 Reduction of organic com-pounds by lithium aluminum hydride. III.Halides, quinones, miscellaneous nitrogencompounds. J. Am. Chem. Soc., 70, 3738-3740.

OGSTON, A. G. 1948 Interpretation of experi-ments on metabolic processes, using isotopictracer elements. Nature, 162, 963.

PARDEE, A. B., AND POTTER, V. R. 1948 In-hibition of succinic dehydrogenase by oxal-acetate. J. Biol. Chem., 176, 1085-1094.

PHa&Rs, E. F. 1951 Degradation of labelledpropionic and acetic acids. Arch. Biochem.and Biophys., 33, 173-178.

SAxAmi, W., EVANS, W. E., AND GURIN, S. 1947The synthesis of organic compounds labelledwith isotopic carbon. J. Am. Chem. Soc.,69, 1110-1112.

43319541


STEPENSON, D. P. 1951 On the mass spectraof propanes and butanes containing C1i.J. Chem. Phys., 19, 17-21.

SWIm, H. E., AND KRAMPITZ, L. 0. 1952 Acetateoxidation by E8cherichia coli. FederationProc., 11, 296.

Swim, H. E., AND KRAMPITZ, L. 0. 1954a Aceticacid oxidation by Escherichia coli: Evidencefor the occurrence of the tricarboxylic acidcycle. J. Bact., 67, 419-425.

SWIM, H. E., AND KRAMPITZ, L. 0. 1954b un-

published data.SZULMASTER, J., GRUNBERG-MANAGO, M., AND

DELAVIER, C. 1952 Utilisation du citratepar Escherichia coli. Experimentia, 8, 26-28.

UTTER, M. F., AND WOOD, H. G. 1951 Mech-

anisms of fixation of carbon dioxide by hetero-trophs and autotrophs. Advances in En-zymol., 12, 41-151.

VAN SLYKE, D. D., AND FotJCH, J. 1940 Mano-metric carbon determination. J. Biol. Chem.,136, 509-541.

VAUGHN, R. H., OSBORNE, J. T., WEDDING, G.T., TABACHNICK, J., BEISEL, C. G., ANDBRAXTON, T. 1950 The utilization of citrateby Escherichia coli. J. Bact., 60, 119-127.

WOOD, H. G. 1946 The fixation of carbon dioxideand the interrelationships of the tricarboxylicacid cycle. Physiol. Revs., 26, 198-246.

WOOD, H. G. 1952 A study of carbon dioxidefixation by mass determination of the types ofC"3-acetate. J. Biol. Chem., 194, 905-931.

434 [VOL. 67

follows: hooc-13hz-c12hrcooh

Documents