JOURNAL OF BACrERIOLOGY, Apr., 1965Copyright a 1965 American Society for Microbiology Vol. 89, No. 4

Printed in U.S.A.

Coupling of Phosphorylation and Carbon DioxideFixation in Extracts of Thiobacillus thioparus1

EMMETT J. JOHNSON2 AND HARRY D. PECK, JR.3

Department of Microbiology, The University of Mississippi Medical Center, Jackson, Mississippi,and Biology Division, Oak Ridge National Laboratory, Oak Ridge, Tennessee

Received for publication 28 November 1964

ABSTRACTJOHNSON, EMMETT J. (University of Mississippi Medical Center, Jackson), AND

HARRY D. PECK, JR. Coupling of phosphorylation and carbon dioxide fixation in ex-tracts of Thiobacillus thioparus. J. Bacteriol. 89:1041-1050. 1965.-A cell-free system fromThiobacillus thioparus which fixes large quantities of C1402 in the presence of ribose-5-phosphate, adenosine triphosphate (ATP), and Mg++ has been described. The specificactivity (0.041 jumole of ribulose-1,5-diphosphate min-' mg-' protein) of the C02-fixingsystem approaches that of green plants, and is further evidence for the importance ofthe role of carboxydismutase in the thiobacilli. In addition to ATP, adenosine diphos-phate (ADP) and other nucleoside triphosphates served with varying degrees of effec-tiveness for the fixation of C1402 . The ATP requirement for C02 fixation was partiallyreplaced under aerobic conditions by a combination of S03-, P04-, and adenosine mono-phosphate (AMP). Phosphorylation and CO2 fixation were separated in time by firstincubating S03 and AMP aerobically, and then anaerobically introducing C1403- andribose-5-phosphate into the reaction mixture. During the first incubation, P3204 wasesterified into nucleotides, mainly ADP, and in the second incubation C1402 was fixed,with the concomitant utilization of almost equal amounts of the esterified phosphate.These data provide the first in vitro evidence for the mechanism of the coupling of C02fixation and phosphorylation in T. thioparus. The fixation of C1402 was shown to bealmost completely inhibited by AMP. This inhibition was not due to the conversion ofATP to ADP by adenylic kinase, or to the binding of magnesium by the nucleotide.The inhibition was specific for AMP, since other mononucleotides, adenosine, and ade-nine did not inhibit. The AMP regulation of CO2 fixation may represent a basic controlmechanism in autotrophic metabolism.

Chemosynthetic autotrophs exhibit a mode oflife unique among living organisms, derivingtheir total energy for growth from the oxidationof inorganic substrates and their total carbonfrom the fixation of C02. The earliest attemptsto explain the relationship between energyproduction and C02 fixation in chemosyntheticautotrophic bacteria are found in the postulationsof Vogler and Umbreit (1942) concerning theseparation in time of phosphorylation and C02fixation. They observed that, in the absence ofC02 fixation, whole cells of Thiobacillus thio-oxidans can oxidize available substrates and store

1 Presented in part at the 64th Annual Meetingof the American Society for Microbiology, Wash-ington, D.C., 3 May 1964.

2 Recipient of a Lederle Medical Faculty Award.Present address: Exobiology Division, Ames Re-search Center, NASA, Moffett Field, Calif.

8 Present address: Laboratoire de Chimie Bac-terienne, Centre National de la Recherche Scienti-fique, C.R.S.I.M., Marseille, France.

the energy in a form that subsequently, in theabsence of oxidizable substrates, can be utilizedto fix C02. Objections were raised by Baalsrudand Baalsrud (1952), because examination of thedata quantitatively revealed an irreconcilabledivergence from expectation on the basis oftheoretical considerations concerning the amountof C02 fixed and of orthophosphate (Pi) utilized.The proposals of Umbreit (1960) seem to begenerally correct, althou,gh perhaps not warrantedby the data existing at the time; however, evi-dence from experiments that include all thecritical variables is still lacking (Larsen, 1960).

Vogler, LePage, and Umbreit (1942) suggestedthe existence of oxidative phosphorylation,because they were able to show that, at lowconcentrations, 2,4-dinitrophenol stimulated therespiration of T. thiooxidans. As yet, however,there is no direct evidence for oxidative phos-phorylation in these organisms, and the exactrelationship between energy production and C02fixation is still largely a matter of conjecture.

1041

on March 1, 2021 by guest

http://jb.asm.org/

Dow

nloaded from

JOHNSON AND PECK

Indirect evidence for oxidative phosphorylationin thiobacilli has been reviewed by Vishniac andTrudinger (1962); however, oxidative phos-phorylation has not as yet been demonstrated tooccur in cell-free preparations of any thiobacilli.At present, only a substrate phosphorylationinsensitive to 2,4-dinitrophenol has been demon-strated in thiobacilli. This substrate-level phos-phorylation occurs during the oxidation of sulfiteto sulfate, and it might be employed to demon-strate the in vitro coupling of phosphorylationwith CO2 fixation (Peck, 1962). This pathway ofsulfite oxidation involves the enzyme, adeno-sine-5-phosphosulfate (APS) reductase, which, inaddition to reducing APS, brings about the oxida-tion of sulfite in the presence of adenosinemonophosphate (AMIP), with the formation ofAPS (Peck, 1961):

APS-reductaseSO3- + AMP APS + 2e (1)

The high-energy sulfate thus formed participatesin a reaction with Pi, catalyzed by the enzyme,adenosine diphosphate (ADP)-sulfurylase, andresults in the production of ADP and sulfate:

ADP-sulf urylaseAPS + Pi I ADP + SO4° (2)Adenylic kinase, which is also present in extractsof these organisms, will convert ADP to adenosinetriphosphate (ATP):

adenylic kinase2 ADP

I

AMP + ATP (3)The requisite enzymes for the fixation of CO2

withribulose-1,5-diphosphate (Ru-1,5-diP) havealso been demonstrated in several species ofthiobacilli (Santer and Vishniac, 1955; Trudinger,1956; Suzuki and Werkman, 1958; Jwatsuka,Kuno, and Maruyama, 1962). The enzymescatalyzing the fixation of CO2 appear to be similarto those described in green plants (Elsden,1962); beginning with ribose-5-phosphate(R-5-P), fixation involves the enzymes phos-phoriboisomerase, phosphoribulokinase, and Ru-1,5-P carboxydismutase (equations 4, 5, and 6).In order that CO2 fixation be dependent uponphosphorylation, it is necessary to begin withR-5-P or ribulose-5-phosphate (Ru-5-P) and CO2as shown in the following equations:

phosphoriboisomeraseR-5-P -'` Ru-5-P (4)

phosphoribulokinaseRu-5-P + ATP I ( )

Ru-1, 5-diP + ADP

Ru-1,5-diP + C02

ribulose-1 ,5-diphosphatecarboxydismutase

I

2 3-phosphoglyceric acid

As suggested previously (Peck, 1962), therequisite enzymes for oxidation of sulfite coupledto the production of high-energy phosphate, aswell as for the fixation of CO2 (see above), havebeen demonstrated in extracts of these organisms.Thus, it should now be possible to reproduce theexperiments involving simultaneous energyproduction and CO2 fixation and sequentialenergy production and CO2 fixation in a cell-freesystem. The present communication describesthe properties of a C02-fixing system from T.thioparus and the sequential coupling of phos-phorylation and CO2 fixation in a cell-free extractof this organism. In addition, a possible mecha-nism for the regulation of CO2 fixation is pre-sented.

MATERIALS AND METHODSCultural procedures and preparation of extracts.

T. thioparus (ATCC 8158) was grown in 40 liters ofthe inorganic medium described by Santer, Boyer,and Santer (1959), with the following modifications:Na2S203-5H20, 8.0 g; NH4Cl, 0.5 g. The thio-sulfate and phosphates were autoclaved separatelyand added aseptically just prior to inoculation.Cells were grown at 30 C with forced aeration for 5to 6 days, and the pH was adjusted when requiredby addition of a 10% solution of Na2CO3 . Afterharvesting the cells in a refrigerated Sharples cen-trifuge, a 25 to 50% suspension of the cell pastewas made in 0.33 M tris(hydroxymethyl)amino-methane (Tris)-HCl buffer, pH 7.1 to 7.5; thissuspension was passed once, or occasionally twice,through a French pressure cell (American Instru-ment Co., Inc., Silver Spring, Md.). If the extractwas too viscous, 20,ug of deoxyribonuclease wereadded, and the extract was allowed to stand 5 to 10min at 30 C. Later, it was observed that greaterretention of activity is achieved if the extract isheld in an ice bath until the gel-like materialliquifies, as indicated by visual inspection (usuallyseveral hours, but it can be left overnight). Theextract was then centrifuged at 20,000 X g for 30min. Where indicated, the clarified extract waschromatographed on a coarse Sephadex G-25 col-umn, 15 mm X 300 mm (16-ml capacity). In somecases, the Sephadex-treated extracts were agedapproximately 24 hr at 2 to 4 C to reduce endog-enous CO2 fixation.

Respiration and C02 fixation. Oxygen uptakewas measured with Warburg vessels (20 ml) at 30 C(gas phase, air) without NaOH in the center well.Double side arm flasks were used; one side armcontained Na2SO3 and MgCl2 , and the other con-tained R-5-P and Na2C'403. One side arm wassealed with a vented plug to change the gas phasewhen required. The second side arm was sealedwith a serum cap to allow additions of perchloricacid to stop the reaction and NaOH to absorb theunfixed CO2. The flasks were equilibrated 5 min

1042 J. BACTERIOL.

on March 1, 2021 by guest

http://jb.asm.org/

Dow

nloaded from

VOL. 89, 1965 PHOSPHORYLATION ANI) CO2 FIXATION IN T. THIOPARUS

before adding the substrate, and oxidation wasallowed to proceed for 10 min in air. In thoseexperiments in which oxidation was separatedfrom CO2 fixation, the flasks were flushed withhelium for 5 min after the oxidation period, andthen the R-5-P and Na2C1403 were tipped fromone sidearm for the fixation of CO2 . After 10 min,0.3 ml of 10% perchloric acid was injected into theside arm through the serum cap by use of a tuber-culin syringe with a 21-gauge needle. The perchlo-ric acid was then added to the main compartmentto stop the reaction and to release the unfixedC402 . Subsequently, 0.3 ml of 10% NaOH was in-jected into the side arm, and the flasks were shakenfor 10 min at 30 C so that the unfixed C'402 wouldbe absorbed. Radioactivity retained in the depro-teinized reaction mixtures was employed as ameasure of C'402 fixed.Phosphate measurements. In experiments in

which changes in nucleotide phosphates are re-ported, P32-labeled orthophosphate was includedin the reaction mixtures to facilitate analyses.Nucleotides were adsorbed onto charcoal fromprotein-free reaction mixtures. After washing thecharcoal four times with 0.05 M acetate buffer (pH5.0) and once with distilled water, the labile phos-phate was hydrolyzed (1 N HCl for 15 min at100 C). Radioactivity of the hydrolysate wasdetermined by means of a Packard liquid scintilla-tion spectrometer.

Chromatography and electrophoresis. The pro-tein-free reaction mixtures were neutralized with30% KOH, and the KCl04 was removed by cen-trifugation. The samples were then frozen andthawed several times, and any further precipitatewas similarly removed. The reaction mixtureswere then extracted with 5 ml of petroleum etherto remove volatile acids, evaporated to dryness,and brought back into solution with 0.2 to 0.3 mlof glass-distilled water. The sugar phosphateswere chromatographed on Whatman no. 1 filter-paper strips or sheets with descending develop-ment at 20 to 25 C. The solvent systems used wereisopropyl ether-formic acid (90%, w/v; 3:2) and80% ethanol-0.8% acetic acid. The former systemwas run for 6 to 8 hr and the latter for 12 to 13 hr.The chromatograms were sprayed with a mixtureof perchloric acid, HCl, and ammonium molybdate(Hanes and Isherwood, 1949), and the sugar phos-phates were located with ultraviolet light. Elec-trophoresis of nucleotides was carried out asdescribed by Robbins and Lipmann (1958), by useof 0.03 M citrate buffer (pH 5.5). For electrophore-sis, nucleotides were adsorbed on acid-washedcharcoal (100 mg per 5 ml), which was then washedfive times with glass-distilled water before elutingthe nucleotides with two 3-ml portions of am-moniacal ethanol. The eluates were combined,evaporated to dryness, dissolved in 0.2 ml of glass-distilled water, and subjected to high-voltagepaper electrophoresis. Nucleotides were alsochromatographed on diethylaminoethyl (DEAE)-cellulose paper sheets at 20 to 25 C and on What-man no. 1 filter-paper sheets at 20 to 25 C in iso-

butyric acid-NH3-water (577:38:385; Jacobson,1964).

Radioactivity measurements. Radioactivity wasmeasured in a liquid scintillation spectrometer(Packard Instrument Co., LaGrange, Ill.) in 10ml of the following solvent: 2,5-diphenyloxazole,4 g; 1,4-bis-2'(5'-phenyloxazolyl) benzeno, 100mg; toluene, 700 ml; absolute ethanol, 300 ml.P3204 and BaC'403 were purchased from the OakRidge National Laboratory, and BaC14O3 wasconverted to Na2C'403 .

RESULTSRequirements for C02 fixation. Before testing

the ability of extracts from T. thioparus to couplephosphorylation with CO2 fixation, it was neces-sary to demonstrate that the extracts had thecapacity for fixing reasonable amounts of C02,beginning at some point in the pathway of CO2fixation that would include at least one energy-requiring step. The work of Santer and Vishniac(1955) showed that small amounts of CO2 (2.5inmmoles) were fixed by extracts of T. thioparuswith Ru-1,5-diP, but not with R-5-P, as the ac-ceptor. However, Trudinger (1956) was able todemonstrate the fixation of C1402 with R-5-P inextracts of T. denitrificans, and similar resultswere obtained with extracts of T. thiooxidans(Suzuki and Werkman, 1958; Iwatsuka et al.,1962). Beginning with R-5-P, the system shouldnecessarily be ATP-dependent (equations 5 and6). Table 1 shows that our extract from T.thioparus fixes relatively large amounts of C1'02(0.041 umole of Ru-1,4-diP min-'mg-' protein)in the overall system. This compares with a spe-cific activity of 0.08 to 0.09 for Ru-1, 5-diPcarboxydismutase in the soluble extract ofspinach leaves (Racker, 1962). If phosphoribulo-kinase was rate-limiting in our extracts of T.thioparus, the specific activity of the carboxylasereported here would be a minimal rather than amaximal value. The system is completely de-pendent on Mg+ and has a strict requirementfor ATP and R-5-P. Ethylenediaminetetraceticacid (EDTA) and NaF were initially added toreaction mixtures because they are required formaximal activity in the phosphorylating system(Peck and Fisher, 1962). Omission of EDTA didnot appreciably affect the fixation, probably be-cause Mg+ was in excess. However, omission ofNaF resulted in an increase in fixation, and it wasconcluded that NaF inhibits this system. Theaddition of fructose-1,6-diphosphate, glucose-6-phosphate, and 3-phosphoglyceric acid (3-PGA) to the system resulted in a decrease in theamount of C"402 fixed. The first two compoundsprobably shunted R-5-P into pathways of utiliza-tion other than CO2 fixation, thus decreasing theamount of R-5-P available for the fixation process.

1043

on March 1, 2021 by guest

http://jb.asm.org/

Dow

nloaded from

JOHNSON AND PECK

TABLE 1. Requirements for CO2 fixation in extractsof Thiobacillus thioparus

System ~~~~~C1402System fixed

JAmolesComplete*............................... 2.73Minus Mg ................................ 0.00Minus ATP ............................... 0.02Minus R-5-P.............................. 0.03Minus EDTA............................. 2.62Minus NaF............................... 3.52Plus fructose-i, 6-diphosphate ....... ...... 2.01Plus glucose-6-phosphate................. 2.20Plus 3-phosphoglyceric acid............... 1.82Minus R-5-P + fructose-1,6-diphosphate.. 0.12Minus R-5-P + glucose-6-phosphate...... 0.03Minus R-5-P + 3-phosphoglyceric acid.... 0.02

* The complete system contained in ,Amoles:Tris (pH 8.0), 330; EDTA, 10; NaF, 10; MgC2,30; ATP, 10; R-5-P, 10; Na2C403, 5 (106 counts permin per jsmole); 0.5 ml of a Sephadex-treatedextract (8.5 mg of protein). The following sub-stances were added in the amount of 10 ,umoles:fructose-1,6-diphosphate, glucose-6-phosphate,and 3-phosphoglyceric acid. The total volumeof the reaction mixture was 2.7 ml. The incuba-tion period was 10 min at 30 C under helium, andthe reaction was stopped with 0.3 ml of 10%perchloric acid.

The reduction of fixation effected by 3-PGA maybe due to product inhibition. None of thesecompounds could replace R-5-P in the fixationsystem, as no significant amount of C102 wasfixed in its absence.

Chromatography and radioautography of theneutralized aqueous phase of ether-extractedprotein-free reaction mixtures revealed only tworadioactive spots corresponding to R-5-P andfructose-1, 6-diphosphate. Short pulses withlabeled carbonate were not done; therefore, itmight not be expected that the label would befound in 3-PGA, as 3-PGA is rapidly convertedinto other intermediates of hexose synthesis.The compounds found in this system are knownintermediates of the pathway of hexose synthesisfrom C02 (Bassham et al., 1954), and it is knownfrom the work of others (Trudinger, 1956; Suzukiand Werkman, 1958) that extracts of thiobacillicontain many of the enzymes involved in hexosemetabolism.

Since the C02-fixing system was being em-ployed essentially to assay for the formation ofhigh-energy phosphate, it was of interest todetermine the nucleotide specificity of the reac-tion in these extracts. Although Hurwitz et al.(1956) indicated that the purified phosphoribulo-kinase from spinach is highly specific for ATP

(Table 2), all of the nucleotides tested supportedC02 fixation. In addition, the activity of thesenucleotides was approximately the same as ATP,except for cytidine triphosphate (CTP). ADP,the product of the substrate phosphorylation,also stimulated C02 fixation, although not as

effectively as ATP. Whether the utilization of a

variety of nucleotides is an indication of thenucleotide nonspecificity of this phosphoribulo-kinase or a reflection of nucleotide kinases is notknown and must await purification of the enzyme.Nevertheless, this result did indicate that any

high-energy phosphate could be utilized for thefixation of C02 .

Properties of CO2-fixing enzymes. Before testingthe ability of the extract to couple sulfite oxida-tion and ATP formation to C02 fixation, it was

necessary to determine the experimental response

of the C02-fixing enzymes to a range of concentra-tions of ATP, because the amount of ATP pro-

duced from sulfite oxidation in these extractswould probably be small. The activity was

proportional to the concentration of ATP up toabout 5,umoles of ATP, at which point the systemappeared to be saturated with respect to ATP(Fig. 1). This response of the C02-fixing systemindicated that it could be utilized essentially as

an assay for ATP. A study of the amount of C02fixed as a function of time (Fig. 2) showed thatthe activity was linear for the duration of theexperiment. Consequently, a 10-min incubationperiod for C02 fixation was selected. The relation-ship between C02 fixation and ATP utilization isless than stoichiometric and probably results fromthe hydrolysis of the added ATP, since NaF wasomitted from this assay system. With NaF andlower concentrations of ATP, there is a 1:1

TABLE 2. Nucleotide specificity of CO2 fixationin extracts of Thiobacillus thioparus*

Additionst Cfix4

smolesNone.. 0.02ATP.. 2.62ITP.. 2.42GTP.. 2.15CTP.. 1.18UTP.. 2.49dATP.. 2.37

* The system was the same as in Table 1,except that 10 ,umoles of each nucleotide wereadded; 0.5 ml of a Sephadex-treated extract wasthe source of enzyme (9.2 mg of protein).

t Symbols: ITP, inosine triphosphate; GTP,guanosine triphosphate; UTP, uridine triphos-phate.

1044 J. BACTERIOL.

on March 1, 2021 by guest

http://jb.asm.org/

Dow

nloaded from

VOL. 89, 1965 PHOSPHORYLATION AND CO2 FIXATION IN T. THIOPARUS

2.

2.,

2.a)00E-

LLIx

u

0

()

1.

1.

0.

0.4

.8 0-

.40

.0

0

,6

.2 0

~80

.4

5 1.IE

o1.1U]

x

N10

orO

o2~ 4 6ATP (PLM)

FIG. 1. Relationship between ATP coand C1402 fixation in a cell-free systembacillus thioparus. The complete systesame as in Table 1, except that fluoride t0.5 ml of a Sephadex-treated extract wassource (6.6 mg of protein).

relationship between the amount of Iand C02 fixed.

Sulfite-dependent CO2 fixation. If thePeck and Fisher (1962) can be coupCO2-fixing system just described, itpossible to study the simultaneous andcoupling of sulfite-dependent phosrwith C02 fixation, as well as the paitheir respective interrelationships. Sircoupling will be used to describe tsystem when phosphorylation and Care proceeding at the same time.coupling will refer to the system Mphorylation and C02 fixation are setime; i.e., S03 is incubated with e:AMP, and Pi, and, subsequently, th(made anaerobic to stop the phosrprior to the addition of R-5-P and Csystem.

In the initial experiments, it was nto demonstrate any significant couplirthe substrate phosphorylation and

, FIG. 2. Relationship between incubation time8 10 and C1402 fixation in a cell-free system from Thio-

bacillus thioparus. The complete system was thesame as in Table 1, except that fluoride was omitted;

mcentration 0.5 ml of a Sephadex-treated extract was the source

from Thio- of enzyme (5.7 mg of protein).em was thewas omitted;tas omitted fixing enzymes, either simultaneously or sequen-the enzyme

tially. The reaction mixture for the substratephosphorylation contained a high concentrationof NaF (5 X 10-2 M) and AMP (5 X 10-3 M),

iTP added both of which might interfere with the optimaloperation of the C02-fixing system. Sulfite, the

a system of other component of the reaction mixture, was

)led to the readily demonstrated to have no effect on C02should be fixation in these extracts. As the role of AMP[sequential in the coupling of phosphorylation and C02)horylation fixation should be catalytic, the AMP concentra-rticulars of tion was reduced to 5 X 10-4 M. Although NaFmultaneous is required for maximal activity of the substratethe overall phosphorylation, it was previously shown to02 fixation inhibit C02 fixation (Table 1). Consequently, theSequential concentration of NaF that would allow phos-vhen phos- phorylation without radically reducing theparated in amount of C02 fixed was determined. In addition,xtract, air, for the demonstration of significant coupling,a system is either simultaneous or sequential, it was neces-

)horylation sary to treat extracts with MnCl2 (0.02 M, final402 to the concentration) to reduce the endogenous C02-

fixing activity. The precipitate formed by MnCl2lot possible was removed by centrifugation, and the extractng between was passed over a Sephadex column, as describedthe C02- in Materials and Methods. This treatment

1045

3 6 9 12 15 18TIME (min)

on March 1, 2021 by guest

http://jb.asm.org/

Dow

nloaded from

JOHNSON AND PECK

probably removed ribonucleic acids, therebypreventing the endogenous formation of high-energy phosphate by the action of polynucleotidephosphorylase. Under the conditions shown inTable 3, it was possible to partially replace addedATP by the phosphorylation accompanying theoxidation of sulfite.For the simultaneous system, both the sulfite

and C02-fixing components were added after theequilibration period, and the activities wereallowed to proceed in air for 10 min at 30 C.For the sequential system, the oxidation ofsulfite and the phosphorylation events wereallowed to proceed in air at 30 C for 10 min. Theair was then displaced with helium, and the com-ponents for CO2 fixation were tipped in from thesecond side arm and allowed to incubate 10 minat 30 C.The total activity under these conditions, both

endogenous and exogenous, was greater when theevents occurred simultaneously than when theyoccurred sequentially; however, the increase overthe endogenous was similar in each case. Themost striking result was the low activity of thecoupled system which was less than 10% of theactivity observed with ATP and considerablyless than the activity of the substrate phos-phorylation. This low activity may have been dueto the hydrolysis of ATP; however, this did not

TABLE 3. Fixation of C1402 coupled to the oxidationof sulfite in a cell-free system from Thiobacillus

thioparus

System* Oxygen C1402 fixeduptake

,uiters jAmoles

Endogenous (-Na2SO3)Simultaneous 0.08Sequential ............. 7 0.04

Exogenous (+Na2SO3)Simultaneous .......... 20 0.17Sequential............. 32 0.12

* The complete system contained in jAmoles:Tris (pH 8.0), 330; AMP, 1; Na2HPO4, 5; EDTA,10; NaF, 10; Na2SO3, 10; MgCl2, 60; R-5-P, 10;Na2C'403, 5 (106 counts per min per,umole); 1 mlof Sephadex-treated extract (18 mg of protein).Incubation was at 30 C for 10 min in air for thesimultaneous system, into which both the sulfiteand C02-fixing components were added at thesame time. In the sequential system, the oxida-tion of sulfite was allowed to proceed in air at30 C for 10 min. The air was then displaced withhelium, and the components for CO2 fixationwere tipped in from the second side arm and al-lowed to incubate for 10 min at 30 C. The reac-tion was stopped with 0.3 ml of 10% perchloricacid.

seem to be the explanation, because the amountsof CO2 fixed in the simultaneous and sequentialexperiments were almost identical. Furthermore,increasing amounts of NaF did not appreciablyaffect the amount of CO2 fixed in the presence ofSO3, although it did decrease the endogenousfixation of CO2 (Table 4). These results suggestedthat some component of the reaction mixture wasinhibiting either the phosphorylation or CO2fixation.A critical relationship between AMP stimula-

tion and inhibition of the sulfite-dependentC@02 fixation is apparent from the evidencepresented in Table 5. The small amount ofendogenous activity in the absence of AMP wasstimulated by the addition of 1 ,umole of AMP,but severely inhibited (95%) by the presence of5 Amoles of AMP. Likewise, the sulfite-dependentC1402 fixation was stimulated approximatelythreefold by the presence of 1 ,umole of AMP,but the presence of 5,moles of AMP producedalmost complete inhibition of the coupling.Because of the delicate balance between the levelof AMP which is required and that which in-hibits, it was concluded that the probabilitywas small of making the coupled system asactive as the ATP-promoted system.

Effect of AMP on the C02-fixing system. Theeffect of increasing concentrations of AMP onthe ATP-dependent CO2 fixation, under condi-tions where ATP is not limiting CO2 fixation (asevidenced by the fact that only 0.15 jAmole of theavailable 1 ,umole was used in the control), isshown in Table 6. Inhibition of CO2 fixation wasevident at an AMP-ATP ratio of 1; at a ratio of5:1, activity was almost completely inhibited.The addition of ADP to this system did not in-

TABLE 4. Effect of fluoride on the fixation of C1402coupled to the oxidation of sulfite

System* NaF Oxygen C1402 fixeduptake

;&moles pJiters smolesMinus Na2SO3 0 0.03

5 0.0110 0.0120 0.01

Plus Na2SO3 0 34 0.215 26 0.1510 34 0.1520 29 0.16

* The components were the same as in Table 3,except for the variation in fluoride concentration;1 ml of a Sephadex-treated extract (20 mg ofprotein) was used. Incubation time was 10min at 30 C in air for the oxidation, and 10 minat 30 C under helium for the CO2 fixation.

1046 J. BACTERIOL.

on March 1, 2021 by guest

http://jb.asm.org/

Dow

nloaded from

VOL. 89, 1965 PHOSPHORYLATION AND C02 FIXATION IN T. THIOPARUS

hibit C02 fixation, and ADP was actually capa-ble of supporting C02 fixation in the absence ofATP. Since ADP increased the extent of C02fixation in the presence of rate-limiting concen-trations of ATP, the inhibition appeared to bespecific for the mononucleotide.

Inhibition was not due to the conversion ofATP to ADP by the action of adenylic kinase,since chromatography of reaction mixtures,completely inhibited by AMP, indicated thatATP was still present and little or no ADP hadbeen formed. In the inhibited system, essentiallyall of the ATP originally added could be re-covered, and in the uninhibited svstem a littleless than half of the ATP had disappeared. Theseresults, plus the fact that ADP will supportC02 fixation, demonstrated that the AMP in-hibition of C02 fixation is not due to the removalof ATP by the action of adenylic kinase.As ATP is probably present only in limiting

TABLE 5. AMP inhibition of C'402 fixation coupledto the oxidation of sulfite

System* AMP Oxygen C1402 fixeduptake

limoles ;diters ILmoles

Minus Na2SO3 0.0 2 0.041.0 3 0.065.0 3 0.0010.0 3 0.00

Plus Na2SO3 0.0 32 0.071.0 31 0.205.0 32 0.0310.0 32 0.00

* The components were the same as Table 3,except for the variation in AMP concentration;1 ml of a Sephadex-treated extract (15.6 mg ofprotein) was used. Incubation time was 10 minat 30 C in air for the oxidation, and 10 min at30 C under helium for the fixation.

TABLE 6. Inhibition of A TP-dependent C02fixation by AMP*

ATP AMP C1402 fixed

Amoles Mmoles ;moles

0.0 0.0 0.021.0 0.0 0.151.0 0.5 0.151.0 1.0 0.131.0 2.0 0.121.0 5.0 0.041.0 10.0 0.01

* The complete system was the same as in Fig.1; 0.2 ml of a Sephadex-treated extract (1.7 mgof protein) was used.

concentrations in the coupled system, it was ofinterest to determine whether AMP also in-hibited C02 fixation under conditions where ATPwas limiting. The data presented in Table 7indicate that the pattern of inhibition is the sameas that obtained with nonlimiting concentrationsof ATP, except that inhibition is evident at a

slightly lower ratio of AMP-ATP. This resultagain indicates the difficulty in obtaining an

active, coupled system, due to the inhibitoryaction of AMP.Although a number of nucleotide triphosphates

are capable of supporting C02 fixation (Table 8),only AMP, of the mononucleotides tested, in-hibited C02 fixation. The specificity of the in-hibition indicated that the inhibition does notresult from binding of Mg++, or possibly othercomponents, by the added nucleotide. In agree-

ment with this conclusion was the additionalobservation that added Mg++ does not relieve the

TABLE 7. Inhibition of A TP-dependent C02 fixationby AMP with limiting concentrations of ATP*

ATP AMP C1402 fixed

pmoles ,umoles umoles

0.0 0.0 0.030.5 0.0 0.5110.0 0.0 2.240.5 0.25 0.420.5 0.5 0.330.5 1.0 0.170.5 2.0 0.060.5 5.0 0.05

* The complete system was the same as in Fig.1; 0.5 ml of a Sephadex-treated extract (4.2 mgof protein) was used.

TABLE 8. Mononucleotide specificity of theinhibition of C02 fixation*

Per cent inhibition of C02 fixationAddition

5 JAmoles 10 umoles

AMP ................. 90 95IMP................. 13 6GMP................. 5 20CMP ................. 9 7UMP ................. 12 7Adenosine ............. 11 15Adenine ............... 13 12

* The reaction mixture was the same as de-scribed in Fig. 1, except that 1 ,umole of ATPwas employed. Inhibitors were added at con-centrations of 5 and 10 /Amoles; 0.3 ml of a Sepha-dex-treated extract (2.3 mg of protein) was umed.

1047

on March 1, 2021 by guest

http://jb.asm.org/

Dow

nloaded from

JOHNSON AND PECK

AMP inhibition. This specific inhibition of a

major energy-utilizing system by AMP mayindicate that the level of AMP in these cellscontrols the activity of the C02-fixing enzymes,and, consequently, the size of the ATP pool.Since the utilization of ATP for all other syn-theses must await CO2 fixation, which providesthe organic precursors for all other syntheses inan autotroph, it is possible that uncontrolledCO2 fixation and the concomitant decrease inthe ATP pool could limit the available ATP forother essential anabolic reactions. This could beespecially critical in the absence of phosphoryla-tion to restore the diminishing ATP pool. Theproper balance between the various ATP-requir-ing activities would be expected to play a signifi-cant role in the well-being of the cell. If so, thenthis AMP control of CO2 fixation could have a

functional significance through its influence inmaintaining the necessary energy balance re-

quired for the survival of the cell.Sequential coupling of phosphorylation and

C02 fixation. To measure the changes of high-energy phosphate in the absence of C'402 fixationand in the presence of C1402 fixation, a series ofexperiments with P32O4- and C'402 were carriedout with the coupled system. To measure theesterification of phosphate, duplicate reactionswere run, and esterified phosphate was deter-mined before and after CO2 fixation. For themeasurement of the esterification of P32O4-before CO2 fixation, the extract was incubatedfor 10 min in air in the presence of SO3, P3204m,

and AMP; esterified phosphate was then deter-mined. For the measurement of the esterificationof P3204= after CO2 fixation, So3= P3204-,- andAMP were incubated with the extract for 10 minin air at 30 C. After 10 min, the air was displacedwith helium, R-5-P plus Na2C1403 was added,and C1402 fixation was allowed to proceed for 10min. A third set of reaction mixtures identical tothose just described, but with unlabeled phos-phate, was treated the same as the set withP32047 and Na2C1403 to measure the C1402 fixed.In the absence of CO2 fixation, 172 m,moles ofphosphate were esterified, and the esterificationwas dependent on both sulfite and AMP as

shown in column 1 of Table 9. When CO2 was

being fixed, 68 mgmoles of this labeled phosphatedisappeared from the pool of high-energy phos-phate (Table 9, column 2). In the absence ofsulfite, there was actually a slight increase in thelabile phosphate pool that may have been due toa reaction between high-molecular-weight en-

dogenous polyphosphates or ribonucleic acidsand AMP. In the absence of AMP, this could notoccur, and, consequently, there was no change in

TABLE 9. Sequential formation and utilization ofhigh-energy phosphate for C02 fixation

Totallabile A labilephos- phosphate

System* phate C0oSystem ~~~~~~~~fixationminus plus C02

mjAmoles miAmoles mole-moles

Complete ................ 172 -68 84Minus Na2SO3 ........... 63 +23 15Minus AMP.............. 57 0 20

* The complete system contained in ,umoles:Tris (pH 8.0) 330; EDTA, 10; NaF, 10; MgCl2,30; Na2HP3204, 5 (4 X 10' counts per min per,umole); Na2SO3, 10; 0.5 ml of a Sephadex-treatedextract (8.7 mg of protein). For CO2 fixation,10 Amoles of R-5-P and 5 Amoles of Na2CO3 (106counts per min per ,umole) were added. The flaskswere incubated in air for 10 min, flushed with Hefor 5 min, and, for CO2 fixation, incubated underHe for 10 min with R-5-P, 10 /Amoles, andNa2C1403, 5 ,umoles (106 counts per min per/Amole). Labile phosphate was measured as P3204,which was adsorbable on acid-washed charcoaland hydrolyzable by treatment with 1 N HCI at100 C for 15 min.

the pool under these conditions. The data incolumn 3 (Table 9) show the amount of C1402fixed during the utilization of high-energy phos-phate indicated in column 2. If one applies acorrection for the largest amount of C@402 fixedin the absence of AMP, the amount of C1402fixed is 64 m,umoles when 68 m,umoles of high-energy phosphate were utilized. Or, if one makesa correction for the total high-energy phosphateinvolved and total C402 fixed, the values are 91m,umoles of high-energy phosphate and 84mAmoles of C1402. This stoichiometry is practi-cally identical with the theoretical requirementbeginning with R-5-P.The P3204-- esterified during the oxidation was

found-by paper electrophoresis and chromatog-raphy, followed by radioautography-to bemainly present in both ADP and ATP. In agree-ment with the mechanism of Peck and Fisher(1962), the larger amount of label was found inthe ADP.

DIscussIoNWhether substrate phosphorylation (Peck and

Fisher, 1962) is the only mechanism for produc-tion of high-energy phosphate in this organism isas yet an unanswered question. That it mightsatisfy the energetic requirements of the cell hasbeen suggested by Peck (1962); however, on the

1048 J. BACTERIOL.

on March 1, 2021 by guest

http://jb.asm.org/

Dow

nloaded from

VOL. 89, 1965 PHOSPHORYLATION AND CO2 FIXATION IN T. THIOPARUS

basis of theoretical ATP requirement and ob-served cell yield, it has been indicated that thissubstrate phosphorylation cannot supply all theenergy for growth (Vishniac and Trudinger,1962). This fact-plus the known existence ofcytochromes (Cook and Umbreit, 1963;Trudinger, 1961; Aubert et al., 1958), coenzymeQ (Cook and Umbreit, 1963), and the inhibitionof CO2 fixation and ATP formation in T. thioparusby 2,4-dinitrophenol (Kelley and Syrett, 1963;Kelley and Syrett, 1964)-suggests the existenceof oxidative phosphorylation in these organisms,but, as yet, no direct evidence has been presented.In any case, the phosphorylating system of Peckand Fisher (1962) and the C02-fixing systemdescribed here provide the minimal requirementsfor studies on the relationship between phos-phorylation and CO2 fixation, particularly thesequential coupling of phosphorylation and CO2fixation.Although considerable discussion has persisted

over the significance of the data presented byUmbreit and co-workers concerning the separa-tion in time of phosphorylation and CO2 fixationin T. thiooxidans, the generalizations made fromthese data have been substantiated in otherorganisms. It is not intended to discuss the con-troversy here, as this aspect has recently beenably reviewed by Larsen (1960). The demonstra-tion that, in extracts, a phosphorylation involvedin the oxidation of reduced sulfur compounds toS04 can be stoichiometrically coupled with CO2fixation supports the applicability of the gen-eralization concerning separation in time ofphosphorylation and CO2 fixation to the thio-bacilli, albeit the present work was done with T.thioparus and a different substrate.

Ru-1, 5-diP carboxydismutase is the criticalenzyme involved in autotrophic CO2 fixation.Analysis of metabolites after short exposure ofcells of T. denitrificans to C'402 indicates that theCalvin cycle functions in these organisms (Mil-haud, Aubert, and Millet, 1956). Furthermore,data indicating the presence of carboxydismutasein extracts of T. thioparus, T. denitrificans, andT. thiooxidans have been published; however,specific activities were either low or could not becalculated. The specific activity of the C02-fixing system reported here, approaching that ofgreen plants, is further evidence for the im-portance of the role of carboxydismutase inthiobacilli.

It is not known whether the utilization of allthe nucleoside triphosphates tested for CO2fixation is a reflection of nucleotide nonspecificity,or an indication of kinase activity allowing for thetransfer of phosphate from all of the nucleoside

triphosphates tested to AMP or ADP. A distinc-tion between these two possibilities will not bepossible until the phosphoribulokinase is suffi-ciently purified to test directly the nucleotidespecificity. Purification of this enzyme and thecarboxydismutase is underway.

Autotrophic organisms, by definition, arecapable of synthesizing their total cellular ma-terial from CO2 and are, therefore, considered tobe the most biosynthetically complete of knownorganisms. Imposed on this biosynthetic metabo-lism is an energy-producing system, chemosyn-thetic or photosynthetic, and a very potentC02-fixing system. The C02-fixing system mustutilize a comparatively large fraction of the ATPproduced by the cell and might be thought of asan "ATP-sink." During periods of energy dep-rivation, this system, uncontrolled, couldrapidly deplete accumulated ATP and endog-enous energy reserves, and possibly result in thepremature death of the cells. Thus, it might bepostulated that there exists a mechanism forcontrolling CO2 fixation, and thereby the level ofintracellular ATP during periods of energyshortage or deprivation.The inhibition of the C02-fixing system by

AMP may represent such a control mechanismfor maintaining a critical level of ATP. If the invitro results are applicable to the intact cells,when 65 to 70% of the adenosine nucleotides arein the form of AMP, the C02-fixing system iscompletely inhibited, thus permitting the cells tomaintain the level of ATP necessary for otheressential reactions in the cell. This inhibitiondoes not represent a Mg++-AMP effect (Lardyand Parks, 1956), and it is highly specific forAMP-other nucleotides and ADP being ineffec-tive. Further studies will be required to deter-mine the locus of the inhibition, the type ofinhibition, and the composition of the adenosinenucleotide pool in whole cells under conditions ofenergy deprivation.

ACKNOWLEDGMENTThis investigation was supported by the U.S.

Atomic Energy Commission under contract withthe Union Carbide Corp.

LITERATURE CITEDAUJBERT, J. P., G. MILHAUD, C. MONCEL, AND J.MILLET. 1958. Existence, isolement et propridt6sphysico-chimiques d'un cytochromec de Thio-bacillus denitrificans. Compt. Rend. 246:1616-1619.

BAALSRUD, K., AND K. S. BAALSRUD. 1952. Therole of phosphate in C02 assimilation of Thio-bacilli. Phosphorus Metab., Symp., Baltimore,Johns Hopkins Univ., McCollum-Pratt Inst.Contrib. 23 2:544-576.

1049

on March 1, 2021 by guest

http://jb.asm.org/

Dow

nloaded from

JOHNSON AND PECK

BASSHAM, J. A., A. A. BENSON, L. D. KAY, A. E.HARRIS, A. T. WILSON, AND M. CALVIN. 1954.The path of carbon in photosynthesis. XXI. Thecyclic regeneration of carbon dioxide acceptors.J. Am. Chem. Soc. 76:1760-1770.

Coon, T. M., AND W. W. UMBREIT. 1963. Theoccurrence of cytochrome and coenzyme Q inThiobacillus thiooxidans. Biochemistry 2:194-196.

ELSDEN, S. R. 1962. Photosynthesis and litho-trophic carbon dioxide fixation, p. 1-40. In, I. C.Gunsalus and R. Y. Stanier, [ed.], The bacteria,vol. 3. Academic Press, Inc., New York.

HANES, C. S., AND F. A. ISHERWOOD. 1949. Separa-tion of the phosphoric esters on the filter paperchromatogram. Nature 164:1107-1112.

HURWITZ, J., A. WEISSBACH, B. L. HORECKER,AND P. Z. SMYRNIOTIS. 1956. Spinach phos-phoribulokinase. J. Biol. Chem. 218:769-783.

IWATSUKA, H., M. KUNO, AND M. MARUYAMA.1962. Studies on the metabolism of a sulfuroxidizing bacterium. II. The system of CO2fixation in Thiobacillus thiooxidans. Plant CellPhysiol. (Tokyo) 3:157-166.

JACOBSON, K. B. 1964. Chromatographic separa-

tion of nucleotides and nucleosides. J. Chro-matog. 14:542-543.

KELLEY, D. P., AND P. J. SYRETT. 1963. Effect of2:4-dinitrophenol on carbon dioxide fixation bya Thiobacillus. Nature 197:1087-1089.

KELLEY, D. P., AND P. J. SYRETT. 1964. Inhibitionof formation of adenosine triphosphate in Thio-bacillus thioparus by 2:4-dinitrophenol. Nature202:597-598.

LARDY, H. A. AND R. E. PARKS, JR. 1956. Influenceof ATP concentration on rates of some phos-phorylation reactions, p. 584-587. In 0. H.Gaebler [ed.], Enzymes: units of biologicalstructure and function. Academic Press, Inc.,New York.

LARSEN, H. 1960. Chemosynthesis, p. 613-648. InW. Ruhland led.], Handbuch der PflanzenPhysiologie, vol. 5, part II. Springer-Verlag,Berlin, Germany.

MILHAUD, G., J. P. AUBERT, AND J. MILLET. 1956.Metabolisme du carbone dans la chimiauto-trophic cycle d'assimilation de l'anhydridecarbonique. Compt. Rend. 243:102-105.

PECK, H. D., JR. 1961. Evidence for the reversi-bility of the reaction catalyzed by adenosine5'-phosphosulfate reductase. Biochim. Biophys.Acta 49:621-624.

PECK, H. D., JR. 1962. Comparative metabolismof inorganic sulfur compounds in microorgan-isms. Bacteriol. Rev. 26:67-94.

PECK, H. D., JR., AND E. FISHER, JR. 1962. Theoxidation of thiosulfate and phosphorylationin extracts of Thiobacillus thioparus. J. Biol.Chem. 237:190-197.

RACKER, E. 1962. Ribulose diphosphate carboxyl-ase from spinach leaves, p. 266-270. In S. P.Colowick and N. 0. Kaplan [ed.], Methods inenzymology, vol. 5. Academic Press, Inc., NewYork.

ROBBINS, P. W., AND F. LIPMANN. 1958. Separationof the two enzymatic phases in active sulfatesynthesis. J. Biol. Chem. 233:681-685.

SANTER, M., J. BOYER, AND V. SANTER. 1959.Thiobacillus novellus. I. Growth on organic andinorganic media. J. Bacteriol. 78:197-202.

SANTER, M., AND W. VISHNIAC. 1955. CO2 incor-poration by extracts of Thiobacillus thioparus.Biochim. Biophys. Acta 18:157-158.

SUZUKI, I., AND C. H. WERKMAN. 1958. Chemo-autotrophic carbon dioxide fixation by extractsof Thiobacillus thiooxidans. II. Formation ofphosphoglyceric acid. Arch. Biochem. Biophys.77:112-123.

TRUDINGER, P. A. 1956. Fixation of carbon dioxideby extracts of the strict autotroph Thiobacillusdenitrificans. Biochem. J. 64:274-286.

TRUDINGER, P. A. 1961. Thiosulphate oxidationand cytochromes in Thiobacillus X. I. Frac-tionation of bacterial extracts and propertiesof cytochromes. 2. Thiosulphate-oxidizing en-zyme. Biochem. J. 78:673-686.

UMBREIT, W. W. 1960. Sulfur (and selenium) bac-teria, p. 682-686. In W. Ruhland [ed.1, Handbuchder Pflanzenphysiologie, vol. 5, part II.Springer-Verlag, Berlin, Germany.

VISHNIAC, W., AND P. A. TRUDINGER. 1962. Carbondioxide fixation and substrate oxidation in thechemosynthetic sulfur and hydrogen bacteria.Bacteriol. Rev. 26:168175.

VOGLER, K. G., G. A. LEPAGE, AND W. W. UM-BREIT. 1942. The respiration of Thiobacillusthiooxidans on sulfur. J. Gen. Physiol. 26:89-102.

VOGLER, K. G., AND W. W. UMBREIT. 1942. Studieson the metabolism of the autotrophic bacteria.III. The nature of the energy storage materialactive in the chemosynthetic process. J. Gen.Physiol. 26:157-167.

1050 J. BACTERIOL.

on March 1, 2021 by guest

http://jb.asm.org/

Dow

nloaded from