bacterial respiration · 50 haddockandjones sulfur proteins of the nadh dehydrogenase...

BACTIumOWoGICAL Rzvizws, Mar. 1977, p. 47-99Copyright 0 1977 American Society for Microbiology

Vol. 41, No. 1Printed in U.S.A.

Bacterial RespirationBRUCE A. HADDOCK* AND COLIN W. JONES

Department ofBiochemistry, Medical Sciences Institute, University ofDundee, Dundee DD1 4HN, Scotland,*and Department ofBiochemistry, School ofBiological Sciences, University ofLeicester, Leicester LEI 7RH,

England, United Kingdom

INTRODUCTION.............................................................. 47

BACTERIAL ELECTRON TRANSPORT CHAINS ........... ................... 52Escherichia coli ....................................... 52

Aerobic electron transport................................................... 55

Anaerobic electron transport ...................... 59Paracoccuw denifirfcans ...................... 66

Aerobic electron transport.................................................. 66

Anaerobic electron transport ...................... 68Azotobacter vinelandii ...................... 68

Respiratory chain energy conservation ..................................... 69Respiratory protection of nitrogenase ....................................... 70

BACTERIAL ATPase COMPLEXES ....................................... 71Membrane-Bound ATPase Complexes ....................................... 71Morphology and location .......... ............................. 71Proton-translocating properties........................................ 72

Purification and Properties of the F,-ATPase .................................. 73

Isolation and general properties ........................................ 73Subunit composition.............................. 75Functional organization ofsubunits. 78

Purification and Properties of the Fo-F1 Complex ........... ................... 79Mutants Defective in Energy Transduction .................................... 80unc ATPase- mutants ......... .............................. 81unc ATPase+ mutants ........ ............................... 82

CONCLUSIONS ....... ..................................... 84LITERATURE CITED............................................ 85

INTRODUCTION

Bacteria can derive the energy they need forgrowth from a considerable number of diverseand varied reactions, and the particular reac-tions utilized by a given organism can changedepending upon the growth conditions em-ployed. Operationally, however, these differentreactions can be considered as examples ofjusttwo general methods for the conservation ofenergy. The first of these is the formation ofadenosine 5'-triphosphate (ATP) by substratelevel phosphorylation and two distinct classesof reaction can be distinguished:

(i) ADP + substrate P = ATP + substrate(ii) ADP + Pi + substrate X = ATP + substrate

+ X

where ADP is adenosine 5'-diphosphate, and Piis inorganic phosphate. Particular examplesare described in detail elsewhere (279), but itis of interest to note that only a relatively smallnumber of substrate level phosphorylationreactions have been identified, and all arecatalyzed by soluble enzymes present in the cellcytoplasm. The second general method of ATPsynthesis in bacteria is by oxidative or photo-

phosphorylation. In this case, ATP synthesis iscoupled to electron transport reactions which,in turn can be driven by light (in phototrophs)or by the oxidation of both organic compounds(in organoheterotrophs) and inorganic ions (inchemolithotrophs) of negative redox potential,linked to the reduction of electron acceptors ofmore positive redox potential. Although thereare differences in detail, the overall features ofelectron transport-dependent ATP synthesisare very similar in bacteria, in mitochondria,and in photosynthetic systems. Thus, in allcases, the enzymes responsible for oxidativephosphorylation are membrane bound and, inaddition, are asymmetrically organized in themembrane so as to catalyze vectorial chemicalreactions. The mechanism of oxidative phos-phorylation, that is, the way in which redoxreactions are linked to the synthesis of ATP,has been, and continues to be, a lively topic fordebate. To the three general models of energycoupling developed in the 1960s (the chemicaltheory, the chemiosmotic theory, and the con-formational theory) have now been added anumber of variations. It is not our intentionhere to review these various proposals in detailor to argue their relative merits and limita-

47

on October 16, 2020 by guest

http://mm

br.asm.org/

Dow

nloaded from

http://mmbr.asm.org/

48 HADDOCK AND JONES

tions. These aspects have been covered thor-oughly both by the original authors and byothers in a number of comprehensive reviews(e.g., references 30, 44, 138, 139, 160, 161, 377).However, it is of importance to discuss one ofthese theories in some detail, since it is fromthe chemiosmotic theory proposed by Mitchell(270, 272) that a unifying conceptual frameworkhas emerged to link the various energy-depend-ent functions in bacteria, mitochondria, chloro-plasts, muscle, and nerve.A characteristic feature of the chemiosmotic

theory is the consideration given to the asym-metrical orientation of membrane-bound en-zymes catalyzing vectorial reactions that bringabout the translocation of molecules, ions, andchemical groups across the membrane. In addi-tion, some of these reactions lead to the separa-tion of electrical charges within and across themembrane, and their recombination underliesthe performance of osmotic, chemical, and me-chanical work.

In its very simplest form, the chemiosmotichypothesis requires that a proton-translocatingelectron transport chain and a proton-translo-cating adenosine triphosphate (ATPase) coexistin a membrane that is essentially impermeableto most ions, including both OH- and H+ ions.The end result of either electron transport orATP hydrolysis is the generation across themembrane of gradients of both pH (ApH) andelectrical potential (A/v), with the soluble phaseon one side of the membrane alkaline and elec-trically negative relative to the other. The sumof these two components, in electrical units(usually millivolts), is known as the protonmo-tive force and, although these components arenot identical, they are all related and intercon-vertible as described by the expression:

AP = A -ZApH

(AP is the proton motive force in millivolts andis a measure of the combined electrical andchemical forces acting on the protons; Atp is theelectrical potential difference across the mem-brane; Z = 2.3 RTIF where R, T, and F havetheir usual meanings, and Z has a numericalvalue of 59 mV at 250C; and ApH is the pHdifference between the interior and the exte-rior). Methods are available for the measure-ment oftransmembrane electrochemical protongradients, and the results obtained have beenreviewed recently (339). It is important to real-ize that, according to the chemiosmotic theory,AP can be a function of ApH exclusively, of Atpexclusively, or a combination of the two, but itcan only be maintained as long as the mem-brane forms a topologically closed vesicle. Un-

der suitable conditions and when of the correctmagnitude, AP drives a variety of energy-linked processes across the membrane, e.g., re-versed electron transport through the respira-tory chain, ATP synthesis via the reversibleproton-translocating ATPase, and the accumu-lation of certain solutes via their respectivepermeases (154).The essential features of electron transport-

dependent ATP synthesis, insofar as they arerelevant to the theme of this review, are sum-marized in Fig. 1. Two protolytic reactions, in-volving the oxidation of a donor (DH2) and thereduction of an acceptor (A), are catalyzed byan enzyme complex, comprising an alternatingsequence of a hydrogen carrier and an electroncarrier, which is arranged across the mem-brane to form a proton-translocating oxidore-duction loop or segment. As drawn and origi-nally described by Mitchell, both donor andacceptor interact with the enzyme complex onthe same side of the membrane, and the netresult of the reaction is the appearance of 2 H+on the left of the membrane and the disappear-ance of 2 H+ from the right of the membrane.However, it must be appreciated that alterna-tive configurations are possible such that thedonor and acceptor interact with the enzymecomplex on different sides of the membrane (anillustrative example is given in Fig. 5) and alsothat stoicheiometries other than 2 H+ translo-cated per redox segment are to be expected forcertain reactions (e.g., the oxidation ofNO2- byNitrobacter winogradskyi [74]).The nature of the donor and acceptor is not

specified in the general scheme shown in Fig. 1.In practice, a variety of both physiological andnonphysiological reductants (e.g., reduced nic-otinamide adenine dinucleotide [NADHI, succi-nate, reduced N,N,N',N'-tetramethylphenyl-enediamine) and oxidants [e.g., oxygen,Fe(CN)63-] can interact with the membrane-bound electron transport chain. In addition, theoxidation of a particular reductant may involvethe action of more than one proton-translocat-ing segment. This is illustrated in Fig. 2, whichsummarizes some current views on the func-tional organization of the mitochondrial respi-ratory chain (128, 129, 275, 307, 413). Thescheme proposed in Fig. 2 must be consideredspeculative at this stage; however, it is in ac-cordance with much of the experimental evi-dence so far available. Most workers agree thatthe mitochondrial respiratory chain can be op-erationally divided into three separate regionsthat are responsible for proton translocationand concomitant ATP synthesis. One of theseregions requires a specific orientation of the

BACTERIOL. REV.


http://mm

br.asm.org/

Dow

nloaded from


BACTERIAL RESPIRATION

Le f t Membrane RightDH2

2H

A +2H+

2~~~~~~, <t~~~H2

ADP +Pi

0HtSW X\A ATP+2HHO02

FIG. 1. Schematic representation of a proton-translocating oxidoreduction segment of the electrontransport chain and of a proton-translocating ATP-ase. (Symbols used are defined in the text.)

NADH dehydrogenase within the membrane,with the proposal that a flavoprotein acts as thehydrogen carrier, and at least two iron sulfurcenters act as the electron carriers (129, 307).Studies with yeast mitochondria have shownthat the NADH dehydrogenase is modified un-der certain growth conditions such that it is nolonger capable of proton translocation or ATPsynthesis and, in some cases, is no longer sensi-tive to inhibition by the site-specific inhibitorspiericidin A and rotenone (73, 75, 129, 140, 147,307). The remainder of the respiratory chain isresponsible for ubiquinol oxidation and is func-tionally organized into two equivalent potentialsites of energy conservation. Recently, Mitchell(275) has proposed the protonmotive Q cycle todescribe the organization of the various redoxcarriers in this region of the respiratory chain.A modified version ofthese proposals is incorpo-rated in Fig. 2 (128). The important conceptualfeature to note is that ubiquinol acts as thehydrogen carrier for two separate electron-carrying limbs, one involving two type b cyto-chromes, the other involving c and a cyto-chromes.Although these suggestions must be consid-

ered speculative, and indeed difficult to verifyunequivocally, they are largely commensuratewith the known properties of the various redoxcarriers involved (see [128]). However, it should

be realized that the experimental evidence for atransmembrane orientation of these carriers asproposed in Fig. 2 is fragmentary. Certainly,the membrane has a defined sidedness, withoxidants and reductants interacting on specificsides of the membrane (for a review, see [160]),and convincing evidence has been presented tosuggest that the type c and a cytochromes forma complex that spans the membrane (176).However, similar evidence for the orientationof the other electron carriers, either the iron-

Inner membrane

FMN

Fe/S2e

Fe/S2e-l

2QH2~- -.---

tQe-

2Q-cyt.c,

10y.<cyt.C

cyt.aII

2Cucyta3

Matrix

.-NADH +H+

_2H+succinate

,2H+

/ 2 H+

tH20

FIG. 2. Proposed functional organization of themitochondrial electron transport chain. Abbrevia-tions: FMN, Flavoprotein; Fe/S, iron-sulfur protein;Q, ubiquinone; QH2, ubiquinol; cyt., cytochrome;and Cu, copper-containing redox proteins. Cyto-chromes bK and bT represent the two type b cyto-chromes differentiated by their spectral and thermo-dynamic properties (413). The scheme is based uponthose proposed previously (eg., references 128, 129,307) and is obviously an oversimplification since, forexample, it does not take into account the knownmultiplicity of the iron-sulfur proteins ofthe electrontransport chain (307).

I

.L

-

49VOL. 41, 1977

2H+ A


http://mm

br.asm.org/

Dow

nloaded from



sulfur proteins of the NADH dehydrogenase(307) or the type b cytochromes (128, 413), is notso convincing and is open to alternative inter-pretation.

According to Mitchell's proposals, as a conse-quence of electron transport through the respi-ratory chain, a protonmotive force is generatedacross the membrane which, in turn, when ofthe correct magnitude and in the presence ofADP and Pi, reverses the direction of a proton-translocating ATPase so as to bring about netsynthesis of ATP. It will be noticed that theoxidation ofNADH by oxygen results in the nettranslocation of six protons and that of succi-nate, for example, of four protons, which can beused for the synthesis of three or two moleculesof ATP, respectively. An additional proton-translocating enzyme complex is also found inthe inner mitochondrial membrane, the en-ergy-linked nicotinamide nucleotide transhy-drogenase, and it has been shown that twoprotons are translocated per NADPH moleculeoxidized by NAD+ (280). Although energy con-servation concomitant with transhydrogenaseactivity has been demonstrated in submito-chondrial particles by direct measurement ofATP synthesis (401) or implied by movement ofsynthetic anions (376), for thermodynamic rea-sons, it is likely that these are only transientphenomena, not found under physiological con-ditions, and occur only when the [NADPH]/[NADP+] and [NAD+]/[NADH] ratios are ex-ceptionally high, since the standard redox po-tentials (Eo') of these couples are so close.The ATPase complexes of energy-coupling

membranes have been the subject of intensiveinvestigation over the past decade and a half.At the beginning of this period, work was di-rected predominantly towards the mitochon-drial and chloroplast systems but, more re-cently, has been extended to an analysis of theATPase complex from several bacteria. The in-formation so far available suggests that theATPase complexes of bacteria, mitochondria,and chloroplasts share many common molecu-lar features (7, 358). In all cases, the ATPasecomplex can be operationally subdivided intotwo components (273), Fo and F., as representedschematically in Fig. 1. In this representation,the hydrophilic F1 component is that part of theenzyme complex whose polypeptides are re-sponsible for adenine nucleotide binding andwhich is still catalytically active as an ATPaseafter removal from the membrane. In contrast,the hydrophobic polypeptides that comprise theF0 component are envisaged as forming a poreor well through the membrane so as to allowthe passage of H+ and H2O to and from F1.

Overall, the complete membrane-bound en-zyme complex catalyzes the vectorial reaction:

ATPR + H20L + 2 H+R = ADPR + PiR + 2 H+L

Although the proposed stoicheiometry is 2 H+translocated per ATP molecule synthesized (orhydrolyzed) as drawn (281), it is experimentallydifficult to determine this --H+/P ratio withprecision, due to the presence ofcharged groupson all chemically reactive substrates and prod-ucts. Indeed, ATPase complexes from variousbiological membranes may well have differentstoicheiometries (references given in [274]).However, the model defines the general princi-ples and, as discussed in greater detail byMitchell (274), additional variations and so-phistications of this type of model are obviouslypossible.An important feature of the ATPase complex

is that the enzyme is fully reversible. Thus, itnot only acts as an ATP synthetase duringoxidative phosphorylation but, in addition, itcan catalyze the formation of a protonmotiveforce at the expense of ATP hydrolysis. Theimportance of the former reaction to aerobicsystems is self-evident. In contrast, the latterreaction is probably little utilized by intact mi-tochondria or aerobic bacteria in vivo, althoughit assumes a much greater importance in bacte-ria such as Escherichia coli and Streptococcusfaecalis, which will grow anaerobically in theabsence of added electron acceptors; under suchconditions, these organisms conserve energy bysubstrate level phosphorylation and musttherefore effect membrane energization, andhence active transport or reversed electrontransfer, via ATP hydrolysis (160, 161, 273, 371).The mechanism of action of the ATPase com-

plex, that is, how the protonmotive force can beutilized to drive ATP synthesis and converselyhow the hydrolysis of ATP brings about theelectrogenic movement of protons, is far fromcertain. Any proposed model for the functionalorganization of the ATPase complex must takeinto account additional features of the enzymediscussed in detail elsewhere (313). These fea-tures include the various exchange reactionscatalyzed by the complex, the presence ofboundnucleotides in the enzyme, and the fact thathydrolysis and synthesis of ATP differ bothkinetically and in their response to certain in-hibitors. A detailed description of the variousmechanisms that have been proposed for ATPsynthesis is outside the scope ofthis article, andthe reader is referred to Harold's recent review(161) for a critical appraisal of the current hy-potheses.A considerable amount of experimental evi-

BACTERIOL. REV.


http://mm

br.asm.org/

Dow

nloaded from


BACTERIAL RESPIRATION 51

dence has now accumulated from studies withbacteria, mitochondria, and chloroplasts insupport, at least in principle, of the chemios-motic theory for oxidative phosphorylation pro-posed by Mitchell. This is not the place to cata-logue all this evidence, but certain key observa-tions are worth noting: (i) there is an invariantcorrelation between the presence of an ener-gized membrane and the generation of a pro-tonmotive force across that membrane; (ii) ATPsynthesis can be supported by an artificiallygenerated protonmotive force; and (iii) isolatedproton pumps and ATPase complexes from dif-ferent sources can be reassembled in membranevesicles to form a complete system capable ofoxidative or photophosphorylation. Indeed, perhaps the greatest single achievement of thechemiosmotic theory has been the stimulusthat it has given to the development of newexperimental ideas and concepts. There re-main, however, certain unresolved questionsconcerning the ability of the chemiosmotic the-ory to account for, both qualitatively and quan-titatively, oxidative phosphorylation in intactcells and derived organelles. Perhaps the mostserious objection to the chemiosmotic interpre-tation of oxidative phosphorylation in mito-chondria is the discrepancy between the maxi-mal phosphorylation potential and the mea-sured protonmotive force. Thus, a comparisonof the electrochemical gradient generatedacross the mitochondrial inner membrane byelectron transport, with the chemical potentialagainst which ATP can be formed from ADPand Pi, indicates that a stoicheiometry of 2, forthe number of protons translocated per ATPmolecule synthesized is insufficient to accountfor the known phosphorylation capacity ofmito-chondria (304, 339). Recently, a report has ap-peared to suggest that the number of protonstranslocated per redox segment of the respira-tory chain is at least 3, and the suggestion hasbeen made that the value of 2, obtained previ-ously, was an underestimate caused by the un-recognized masking of H+ ejection by move-ments of endogenous phosphate (55).As originally proposed by Mitchell, the chem-

iosmotic hypothesis requires that the primaryevent in the energization of a coupling mem-brane is the translocation of protons from oneside of the membrane to the other and theestablishment of a protonmotive force acrossthat membrane (Fig. 1). However, as discussedin greater detail, first by Williams (415-417)and then Robertson and Boardman (331), theprimary event in energization is more likely tobe a charge separation reaction across, butwithin, the membrane phase. The appearanceofH+ in the left aqueous phase is envisaged as a

secondary and slower process that is not a nec-essary requirement for ATP synthesis, sincedirect proton transfer from the respiratory en-zymes to the ATPase can occur directly in thelipid phase. The proposals of Williams andMitchell are both clearly in accordance with theexperimental observation that, under appropri-ate conditions, the addition of a pulse of air-saturated buffer to an anaerobic suspension ofmitochondria results in a transient acidifica-tion of the suspending medium. They differ inthat, according to Williams, the presence of atransmembrane potential is not a necessarycondition for ATP synthesis, and it is the en-ergy of hydration of the "dry" proton producedin the lipid phase of the membrane that drivesthe synthesis of ATP by the ATPase. The twotheories also have different consequences inthat, according to Mitchell, electron transport-dependent ATP synthesis and solute transportare exclusively a property of topographicallyclosed vesicles in which two aqueous phases areseparated by a lipid phase of limited protonpermeability. According to Williams, electrontransport-dependent ATP synthesis does not re-quire a closed vesicle system provided that,following the initial charge separation reactioncatalyzed by a proton pump, the preferred routefor proton movement is through the ATPaseenzyme and not dissipatively by release fromthe membrane into the aqueous phase; natu-rally, it would only be possible to demonstrateenergy-dependent solute accumulation inclosed vesicles. These predictions from Wil-liams' theory have some experimental basis inthat electron transport-dependent ATP synthe-sis has been observed with enzyme complexesreassembled in an octane-water interface (e.g.,reference 421) and in reputedly nonvesicularmembrane fragments (e.g., reference 78).

Clearly, a great deal of controversy and spec-ulation still exists as to the precise mecha-nism(s) of oxidative phosphorylation and asso-ciated energy-linked functions in biologicalmembranes. The generalized picture that is be-ginning to emerge is as follows. First, the var-ious proteins and their component polypeptideshave a defined sidedness and orientation withinthe membrane and, indeed, some, in addition totheir catalytic activity, have a structural rolein maintaining membrane integrity (e.g., theF1 component of the ATPase complex, as dis-cussed later). Second, the redox proteins of theelectron transport chain are functionally orga-nized in the membrane so as to catalyze someform of charge separation reaction. Third (atthis time speculatively), this charge separationreaction can be used directly to synthesize ATPby a series of reactions that occur within the

VOL. 41, 1977


http://mm

br.asm.org/

Dow

nloaded from



lipid phase ofthe membrane. Alternatively, thecharge separation reaction can be used to estab-lish a protonmotive force across the membranethat, in turn, is responsible for the energy-dependent accumulation of solutes across thecoupling membrane by the various pathwayselaborated by Mitchell (270) and others (154,160, 161, 371). The concept that ATP synthesisand active transport differ, in that the former isessentially an intramembrane process reflectedindirectly by the magnitude of the protonmo-tive force, whereas the latter is a transmem-brane event directly dependent upon the pro-tonmotive force, requires more careful consid-eration and evaluation. Further information onthe functional organization ofthe various redoxcarriers into proton-translocating loops and onthe stoicheiometry of H+ translocation duringrespiration and ATP synthesis is obviously re-quired and should be forthcoming in the nearfuture. Also, the significance of protein-proteininteractions and conformational changes in thevarious respiratory complexes and in the ATP-ase complex needs to be properly evaluated be-fore the mechanism of oxidative phosphoryla-tion can be defined at the molecular level.

BACTERIAL ELECTRON TRANSPORTCHAINS

It is now firmly established that bacteria pos-sess membrane-bound electron transportchains that are very similar, in general terms,to their counterparts in both mitochondria andphotosynthetic systems, in that all result in netproton translocation across a membrane of lim-ited ion permeability during oxidation-reduc-tion reactions. The diversity of the individualmembrane-bound redox components found inbacteria, as well as the great variations in thephysiological reductants and oxidants utilizedby bacteria, are well documented in recent re-views (35, 131, 179, 206, 211, 232, 243, 261, 379,383, 411). However, comparatively little atten-tion has been directed towards assessing thefunctional organization of these carriers intoproton-translocating electron transport chains,with the notable exception of proposals madefor Micrococcus lysodeikticus (394). We havechosen a small number of organisms to discussin detail, partly because they have been studiedmore extensively than others and partly be-cause they serve to illustrate the general fea-tures and properties of the various electrontransport chains most commonly found in bac-teria. In this analysis, we have chosen to inter-pret the experimental evidence in accordancewith the proposals of Mitchell (e.g., Fig. 3).Therefore, in the various schemes that follow,

the redox carriers are arranged in oxidoreduc-tion loops or segments containing an alternatesequence of hydrogen and electron carriers. Itshould be remembered, however, that the ex-perimental evidence for these proposals on thefunctional organization of bacterial electrontransport chains is even more fragmentarythan the equivalent evidence for the mitochon-drial or photosynthetic systems. The schemesare presented in the hope that they will encour-age future thinking and experiments to be for-mulated towards understanding the vectorialorganization and functional activity of thosemembrane-bound components responsible forelectron transport-dependent ATP synthesis inbacteria.

Escherichia coliThe gram-negative bacterium E. coli is a

facultative anaerobe that is able to derive en-

Membrane Cytoplasm

NADH+H+

Fp

2H4'<

Fe/SF

,'-gpFp D-lactateFe/S succinate

2H

FIG. 3. Proposed functional organization of theredox carriers responsible for aerobic electron trans-port in E. coli. The scheme includes the variousroutes for aerobic electron transport in E. coli, withthe dashed lines indicating alternative pathways forreducing equivalents. Abbreviations: L,a-gp, L-a-glycerophosphate; otherwise, as in the legend to Fig.2.

BACTERIOL. REV.


http://mm

br.asm.org/

Dow

nloaded from



ergy for growth both fermentatively, via glycol-ysis, and oxidatively, using either oxygen or,under anaerobic conditions, fumarate andNO3- as terminal electron acceptors. The mech-anism of electron transport and oxidative phos-phorylation in E. coli has been the center ofconsiderable attention due largely to the easewith which a variety of mutants can be iso-lated; the subject has been reviewed recently(79, 133). The mutant strains so far availablefor the study of electron transport in E. coli aresummarized in Table 1. The ability to reversethe phenotypic effects resulting from some ofthese mutations has been of use in elucidatingthe functional activity of several of the redoxcomponents (Table 2).E. coli can synthesize a variety of redox car-

riers, depending upon the growth phase, theterminal electron acceptor, the carbon sourcefor growth, and the strain. In addition, altera-

tions to the redox components, synthesized un-der otherwise defined conditions, can beachieved by altering the growth-limiting nutri-ent in the medium. This technique has not yetbeen fully exploited, but some relevant exam-ples of its application are summarized in Table3.At least nine different cytochromes have

been identified in E. coli, two type c cyto-chromes (cytochromes c550 and c548), five type bcytochromes (cytochromes b556, b558, b562, b55°6and o), cytochrome a,, and cytochrome d (149,363). Not all of these cytochromes are mem-brane bound and involved in oxidative phos-phorylation. For example, cytochrome c,50 ispresent in the periplasmic space, has a mid-point potential (Em, pH 7.0) between -195 and-220 mV (122), and apparently accepts reduc-ing equivalents from a soluble form of formatedehydrogenase (340). A lively controversy ex-

TABLE 1. Mutants ofE. coli deficient in electron transport processes

Class of mutant

Ubiquinone defi-cient

Menaquinone defi-cient

Heme deficient

Iron deficient

Nitrate reductase(EC 1.7.99.4)

Dehydrogenase de-ficient

Gene designa-tion Comments References

ubiA-ubiH 132

menA,B

hemA-hemG Isolated as auxotrophs for 5-aminolaevu-(pop) linic acid or hematin (hem) or alterna-

tively as porphyrin-accumulating mu-tants (pop); confusion over gene nomen-clatures (see [346])

entA-entG Lesions in synthesis of enterochelinfep Lesions in transport of ferric-enterochelin

complexfesB Unable to hydrolyze ferric-enterochelin

complexchlA-chlG Isolated by three independent methods;

(nar) some mutants pleiotrophic (chlA,BpDE)involving loss of several of the followingactivities or components: soluble and par-ticulate formate dehydrogenase (EC1.2.2.1), hydrogenase (EC 1.98.1.1), cyto-chrome c., cytochrome bN.-, and ni-trate reductase (EC 1.7.99.4)

dld D-Lactate dehydrogenase (EC 1.1.2.4)

sdh Succinate dehydrogenase (EC 1.3.99.1)frd Fumarate reductase (EC 1.3.99.-)glpA L-a-Glycerophosphate dehydrogenase (EC

1.1.99.-); soluble enzyme required for an-aerobic growth with fumarate as electronacceptor

glpD L-a-Glycerophosphate dehydrogenase (EC1.1.99.5); particulate enzyme required forgrowth aerobically and used preferen-tially during anaerobic growth withNO3- as electron acceptor

nut Energy-linked transhydrogenase (EC1.6.1.1)

298, 427

38, 86, 149, 257, 319,320, 346, 347, 348,420

248, 419, 42880

238

See 136, 342

178, 372

89, 247, 382247, 381219

219, 265

Cited in 133

VOL. 41, 1977


http://mm

br.asm.org/

Dow

nloaded from



TABLE 2. Phenotypic restoration offunctionalactivity in various mutants ofE. coli

lesion Phenotypic restoration by: References

ubiB- Addition of Q-1 to membranes 85, 144, 316menA- Addition of MK-1 to mem- 298

braneshemA- Addition of 5-aminolaevulinic 143, 144,

acid to nongrowing cells, or 149, 332hematin + ATP to mem-branes

ent-, fep- Addition of citrate + Fe"' to 80growth medium

chID- Addition of molybdate to an- 135aerobic growth medium con-taining NO3-

TABLE 3. Alterations to electron transport processesin E. coli induced by changes in the growth limiting

nutrient

Growth-limiting Phenotypic effect Refer-nutrient ences

Iron (aerobically) Decreased levels of type b 326cytochrome and nonhemeiron

Iron (anaerobi- Decreased hydrogenase and 124cally) formate-hydrogen lyase

activitiesSulfate (aerobi- Alteration to iron-sulphur 317

cally) proteins, and synthesis ofalternative cytochromes

Molybdate (an- Decreased nitrate reductase 244aerobically) and formate dehydrogen-

ase activitiesSelenite (anaero- Decreased formate dehydro- 244

bically) genase activity

ists as to whether reduced cytochrome c550, inturn, donates electrons to hydrogenase, consti-tuting an overall formate-hydrogen-lyase activ-ity (76), or alternatively to nitrite reductase(100); neither pathway is coupled to ATP syn-thesis. An NADPH-sulfite reductase (EC1.8.1.2), containing siroheme as a prostheticgroup, has been isolated and characterized fromE. coli (286, 368-370). This enzyme also reducesnitrite, and attention has been drawn to thesimilarities ofthe E. coli NADPH-sulfite reduc-tase and the ferredoxin-nitrite reductase [EC1.7.7.1] of spinach (287). Clearly, much morework is required to understand the significanceof these various soluble redox systems to themetabolic activity of the cell, and they will notbe discussed further, but it is important to re-member that not all of the cytochrome speciesfound in intact cells are necessarily involved inelectron transport-dependent ATP synthesis.The various membrane-bound cytochromesthat have defined roles in oxidative phosphoryl-ation are described in detail below; however, it

should be noted that the intracellular locationand functional significance of other cyto-chromes, notably cytochrome cs48 and cyto-chrome b5,3, is not known at this time. Thus,cytochrome c.548 is present in low concentrationin E. coli (363), and its role is uncertain; inter-estingly, a type c cytochrome has been demon-strated in particles derived from cells grownunder conditions of sulfate limitation in achemostat, but has not been characterized(317). Cytochrome bm, which has been purified(186), sequenced (187), and extensively studied(e.g., reference 90), although its function re-mains obscure, is considered to be a solubleprotein by some workers (122), yet a considera-ble amount remains membrane bound duringparticle isolation (25).E. coli synthesizes both a benzoquinone,

ubiquinone-8, and a naphthoquinone, menaq-uinone-8 (311). In general, a four- to fivefoldhigher concentration of ubiquinone is found inaerobic cells than in cells grown anaerobically,whereas the content of menaquinone, and itsimmediate biosynthetic precursor demethylme-haquinone, are present at higher concentrationin anaerobically grown cells than in cells grownaerobically (41, 150). However, high concentra-tions of menaquinone are often, but not always(317), found in aerobically grown cells when theactivity of the aerobic respiratory chain(s) isimpaired, e.g., in ubiquinone-deficient mutants(85), in heme-deficient mutants (149), and inwild-type cells grown aerobically in the pres-ence of low concentrations of KCN (25). Thecontrol mechanisms regulating the synthesis ofthe two quinone species are not known, but it ispossible that each can substitute for the other,at least in part, to support functional electrontransport activity (85, 298).Of the other types of redox carriers typically

associated with respiratory chains, there is noconvincing evidence for the involvement of cop-per proteins in electron transport. There is aconsiderble amount ofnonheme iron in electrontransport particles, although, as in other sys-tems, presumably only a small fraction is asso-ciated with functional iron-sulfur proteins. Thepresence of several iron-sulfur proteins hasbeen suggested, based upon (i) their differentialreactivity with o-phenanthroline (46, 218) andthe more lipophilic iron chelator bathophenan-throline (88) and (ii) the detection of various g= 1.94 signals in electron paramagnetic reso-nance spectra ofmembranes and derived prepa-rations at 77°K (153, 169, 303) and at 12°K (317).Molybdenum-containing enzymes play a keyrole in anaerobic electron transport, as dis-cussed later.

BACTERIOL. REV.


http://mm

br.asm.org/

Dow

nloaded from


VOL. 41, 1977

Membrane particles from aerobically grownE. coli demonstrate both an energy-dependentand an energy-independent transhydrogenaseactivity. Energy for the reversal of electronflow from NADH to NADP+ can be derivedeither by ATP hydrolysis via the proton trans-locating ATPase (as in mitochondria) or by theoxidation of NADH, succinate, 1)-lactate, andascorbate (in the presence of phenazine metho-sulfate) by the electron transport chain (53, 181,389) and is a property of vesicles with an inside-out orientation with respect to the original cell(181). The energy-linked transhydrogenase ac-tivity has proved a popular assay system forinvestigating the restoration of functional pro-ton-translocating ATPase activity in variousmutants, deficient in such activity, as discussedlater. The activity of the energy-dependenttranshydrogenase in particles derived. fromcells grown aerobically on different carbonsources varied markedly (47), though the activ-ity in particles from anaerobically grown cellshas not been reported. The enzyme was notsubject to catabolite repression, but its synthe-sis was repressed by various mixtures of aminoacids in the growth medium, and it has beensuggested that the transhydrogenase has a rolein generating NADPH for biosynthesis (47).Although transhydrogenase activity in parti-cles is clearly established, it is of interest thatproton translocation associated with NADPHoxidation by whole cells of E. coli has not yetbeen reported.Evidence has accumulated over the last few

years to suggest that the various membrane-bound redox carriers ofE. coli are functionallyorganized into several proton-translocatingelectron transport chains, which serve for respi-ration-dependent ATP synthesis and for the ac-cumulation of various solutes. We shall discussthe aerobic and anaerobic electron transportsystems separately and propose severalschemes to describe the functional organizationof the various components in the cytoplasmicmembrane. It should be remembered, however,that although these schemes are in accordancewith the experimental evidence so far availa-ble, they are only working models and are pre-sented in the hope that they will serve for thedevelopment of more critical experimentaltests.Aerobic electron transport. When grown un-

der conditions of vigorous aeration in the pres-ence of a nonfermentable carbon source likeglycerol or succinate, the membrane-bound re-dox carriers include ubiquinone-8 and cyto-chromes b5m, b5a, and o. Cytochrome o is a typeb cytochrome that serves as the terminal oxi-


dase and is kinetically competent to support theobserved rates of respiration exhibited by cellsgrown under these conditions (145). It can beidentified spectrally as a type b cytochrome,absorbing maximally at 556 nm in reduced-minus-oxidized difference spectra (317), withthe ability to bind carbon monoxide (65) and, inaddition, it exhibits a high affinity for cyanide(321). Further evidence for the presence of anelectron transport chain involving only type bcytochromes is shown by the ability to restorefunctional electron transport activity in parti-cles from a hemA--deficient mutant by incu-bation with hematin and ATP (143, 144, 149,332).Assuming that 2 H+ are required per ATP

molecule synthesized by the ATPase (410),measurements of the stoicheiometry of respira-tion-driven proton translocation in intact cellssuggest that the electron transport chain is or-ganized into two equivalent energy conserva-tion segments in cells grown under these condi-tions (57, 197, 241). Furthermore an H+/0ratio of 4 for -malate oxidation, and an --H+/Oratio of 2 for the oxidation of succinate, 1)-lac-tate, and DL-a-glycerophosphate suggests thatone of these segments is associated specificallywith the NADH dehydrogenase [EC 1.6.99.3]region of the respiratory chain, and the other islocated in the region between the junction ofthe other flavin-linked dehydrogenase with therespiratory chain and the terminal oxidase, cy-tochrome o (241).

Alternative methods exist for assessing thenumber of energy conservation sites in E. coli.Thus, true molar growth yields with respect tooxygen consumption (Ymax) of <60 g of cellsper mole of oxygen consumed have recentlybeen determined for glycerol- and glucose-lim-ited continuous cultures ofE. coli (110, 166) bymeasuring in situ respiratory activity (Qo2)as a function of dilution rate (= specific growthrate; for a review ofthis approach, see reference384). Similar, or slightly lower, values havebeen determined for Bacillus megaterium D440(101) and Klebsiella (Aerobacter) aerogenes(173, 385), both of which also exhibit only twoproton-translocating respiratory segments.During anaerobic growth under glucose-limitedconditions, the latter organism exhibited a truemolar growth yield with respect to ATP utiliza-tion of 14.0 g of cells per mol ofATP consumed(385). Since yjax is equal to the product ofYIpx and N (the overall efficiency of aerobicenergy conservation), then, assuming Ya tobe the same under anaerobic and aerobic condi-tions, N is equivalent to s4.3 mol of ATP permol of oxygen consumed, i.e., the respiratory


http://mm

br.asm.org/

Dow

nloaded from



systems of E. coli and related organisms con-tain two energy-conserving segments. Further-more, since each of these segments translocatestwo protons per pair of reducing equivalentstransferred, these results indirectly confirmthat ATP synthesis via the reversible, proton-translocating ATPase occurs with an H+/ATP ratio of approximately two. Other investi-gations employing direct measurements ofphosphorylation coupled to NAD(P)H oxidationin whole cells have suggested the presence ofthree energy-coupling sites in E. coli (164), butthis approach has been severely criticized (400).Further evidence for two potential energy

conservation sites in the electron transportchain comes from an analysis of energy-linkedfunctions of particles derived from cells grownaerobically on nonfermentable substrates.Thus, in ammonia-treated particles, in whichthe NADH dehydrogenase activity is inhibited,the oxidation of succinate, 1-lactate, and ascor-bate (with phenazine methosulfate) can be usedto drive the energy-dependent transhydrogena-tion of NADP+ (53). In addition, an ATP-de-pendent, uncoupler-sensitive reduction ofNAD+ by succinate, n-lactate, and L,a-glycero-phosphate, has been demonstrated in particlesfrom cells grown under aerobic conditions (316,388). This energy-dependent reversal of elec-tron flow through the NADH dehydrogenaseregion of the electron transport chain requiresubiquinone, but not cytochromes (316), a con-clusion that is supported by the ability to dem-onstrate exogenous ubiquinone-dependent pro-ton translocation linked to the oxidation of en-dogenous substrates in cells of a heme-deficientmutant (146).A scheme summarizing these various experi-

mental observations and suggesting the func-tional organization of the membrane-bound re-dox carriers is given in Fig. 3. At this time,there is no experimental evidence for the rela-tive sidedness of the type b cytochromes in themembrane, and the possibility remains thatcytochrome b5, has a functional role in thiselectron transport chain (see above).

It is now well established that, under certaingrowth conditions, E. coli synthesizes alterna-tive redox carriers that function during aerobicelectron transport. It appears that two, proba-bly separate, modifications can occur to theelectron transport chain we have discussed: (i)the NADH dehydrogenase becomes alteredsuch that it is no longer proton-translocating;and (ii) the cytochrome b556 - cytochrome oelectron-carrying limb associated with ubi-quinol oxidation is replaced by a second elec-tron-carrying limb containing cytochrome bw

with an additional kinetically competent (145)oxidase, cytochrome d (65), which differs fromcytochrome o both spectrally and in having ahigher Ki for cyanide (321, 322). These addi-tional pathways for electron transport to oxy-gen are represented schematically by thedashed lines in Fig. 3.Evidence for the adaptive loss of proton

translocation associated with the NADH dehy-drogenase region of the electron transportchain has come from measurements of the sto-icheiometry of respiration-driven proton trans-location in sulfate-limited cells ofE. coli grownin continuous culture (317), where it was shownthat the oxidation of all substrates, including L-malate, was associated with an -- H+/O ratio ofabout 2. Further evidence supporting this con-clusion came from the inability to demonstrateenergy-dependent reversal of electron transportthrough the NADH dehydrogenase in particlesprepared from such cells (317). These experi-ments were performed with a low-sulfate-re-quiring strain of E. coli K-12 isolated afterprolonged growth of a prototroph under sulfate-limited conditions. More recently, the experi-ments have been repeated withE. coli Wgrownunder sulfate-limited conditions, where it wasshown that the oxidation of endogenous sub-strates and added L-malate was associated with- H+/O ratios approaching 4 (110). The reasonfor this discrepancy is not known, and althoughit may well be accounted for on the basis of astrain difference, further work is obviously re-quired. Additional evidence for a non-proton-translocating NADH dehydrogenase has comefrom measurement of oxygen-dependent protontranslocation in anaerobically grown cellswhere - H+/O ratios of about 2 were obtainedfor the oxidation of endogenous substrates (57)and added L-nmalate and n-lactate (J. A.Downie, Ph.D thesis, University of Dundee,Dundee, Scotland, U.K., 1974). Direct evidencefor the presence of two separate enzymes cata-lyzing NADH oxidation in E. coli is at presentlacking. Although two NADH dehydrogenaseactivities, differing in their ability to reducevarious dye acceptors, have been partially puri-fied fromE. coli (50, 169), it is not clear whetherthese are two distinct enzymes or merely twopreparations of the same enzyme with differentmolecular weights. It is possible that the pro-ton-translocating NADH dehydrogenase is con-verted into a non-proton-translocating form bythe loss or modification of one or a few polypep-tides that would be difficult to detect experi-mentally.The second alteration to the aerobic electron

transport chain apparently involves the synthe-

BACTERIOL. REV.


http://mm

br.asm.org/

Dow

nloaded from


VOL. 41, 1977

sis of an alternative terminal electron-carryinglimb involving cytochromes bm and d. Coor-dinate synthesis of cytochrome bm,8 and d inwild-type cells has been observed under thefollowing growth conditions: (i) during the late-exponential or stationary phase of aerobicbatch cultures growing on nonfermentable car-bon sources (322, 363); (ii) during aerobicgrowth in the presence of glucose (149); (iii)during anaerobic growth on either fermentablesubstrates or on glycerol with fumarate (145,149); however, in cells grown anaerobicallywith NO3- as terminal electron acceptor, thecytochrome d has altered spectral and kineticproperties (145), possibly the result of NO2-binding (B. A. Haddock, unpublished observa-tions); (iv) during growth on a nonfermentablesubstrate, like succinate, in the presence of lowconcentrations of cyanide sufficient to inhibitcytochrome o, but not cytochrome d, oxidaseactivity (25); and (v) in some strains duringsulfate-limited growth conditions in continuousculture (317, but see also reference 110). Inaddition, during aerobic growth of the respira-tory-deficient mutants lacking ubiquinone (85)and functional cytochromes (149) on glucose,high concentrations of (apo)cytochromes bmand d were observed. All these growth condi-tions also resulted in increased synthesis ofcytochrome a,. The role of cytochrome a, in E.coli is a mystery: in certain bacteria, e.g., Ace-tobacter species, cytochrome a, serves as a ter-minal oxidase (65, 260), but cytochrome a, isnot kinetically competent to support observedrespiration rates in E. coli (145). It is also ofinterest that, under all these growth condi-tions, except that of sulfate-limited growth incontinuous culture (317), the cells containedincreased levels of menaquinone. As discussedlater, menaquinone has been implicated as anobligatory carrier in fumarate-dependent an-aerobic electron transport and its function in,and metabolic significance to, aerobicallygrown cells has not been determined. Althoughcoordinate synthesis of cytochromes b5,8 and dis observed under a variety of apparently unre-lated growth conditions, the obligatory reduc-tion of cytochrome d by cytochrome bm has yetto be established; indeed, there is no evidencefor the orientation of the two carriers in themembrane, as proposed in Fig. 3. The experi-mental evidence so far available suggests thatthe stoicheiometry of respiration-driven protontranslocation associated with ubiquinol oxida-tion using either cytochrome bw with cyto-chrome o or cytochrome bw with cytochrome das the electron-carrying limb is similar andequal to 2 (57, 317).


In simple terms, it can be imagined that E.coli synthesizes two oxygen-dependent electrontransport chains. The first, shown by the solidlines in Fig. 3, is designed for maximum yieldof 2 mol of ATP synthesized per mole ofNADH oxidized under conditions of high aera-tion; the second, shown by the dotted lines inFig. 3, is designed primarily for the reoxidationof reduced coenzymes, allows for the synthesisof 1 mol ofATP per molecule ofNADH oxidizedand can operate at lower oxygen tensions. How-ever, at this time, there is no evidence to sug-gest that the synthesis of a non-proton-translo-cating NADH dehydrogenase and the addi-tional electron-carrying limb, containing cyto-chromes b5, and d, are coordinately controlledand necessarily function together. Clearly,the possession of these various pathways foroxygen-dependent electron transport must en-dow the bacterium with selective growth ad-vantages under certain conditions, but the na-ture of these advantages is obscure. It is tempt-ing to speculate that the possession of a proton-translocating NADH dehydrogenase is nor-mally necessary for growth on oxidizable sub-strates, where all ATP synthesis is generatedfrom electron transport, but that during fer-mentative growth, when ATP synthesis occursby glycolysis, a non-proton-translocating, andtherefore, non-energy-conserving NADH dehy-drogenase would be better suited for coenzymeoxidation. The possibility also exists that, dur-ing fermentation, NADH oxidation is linkedpreferentially to an additional proton-translo-cating reaction, involving fumarate reduction,by an electron transport pathway that is dis-cussed in the next section. The presence of twoterminal electron-carrying limbs with differentterminal oxidases that possess different affini-ties for nxygen, as suggested by their differentsensitivity to cyanide, would be useful for adap-tive growth at different oxygen tensions.

Little work has been performed to determinethe factors regulating either the synthesis orthe functional activity ofthe various redox com-ponents that can coexist in the cytoplasmicmembrane. Except for the relatively crude datagiven above there has been no systematic in-vestigation as to the type and amount of qui-none synthesized by a particular bacterialstrain under a variety ofdifferent growth condi-tions. A variety of mutants blocked in theirability to synthesize both ubiquinone and me-naquinone are known (Table 1), though no evi-dence for control genes regulating the amountof quinone produced has been found. Such in-formation is obviously required before the regu-lation of quinone synthesis can be discussed in


http://mm

br.asm.org/

Dow

nloaded from



detail. A regulator gene controlling porphyrinbiosynthesis has been identified (Table 1), andcytochrome gene dosage effects in F-lac and F-gal heterogenotes ofE. coli suggest that a num-

ber of cytochrome genes are clustered in a

2-min-long region of the chromosome betweenthe purE and supE genes (365). Whether thesegenes control the amount of heme produced or

the amount of apocytochrome synthesized hasnot been determined; certainly, the apoproteinsfor cytochromes b556, b558, o, and d are synthe-sized and incorporated into the cytoplasmicmembrane in the absence of heme synthesis(149). The synthesis of cytochrome b556, andprobably o, appears to be constitutive, thoughthe presence of cytochrome o in all cell typesmust be confirmed by kinetic analysis ratherthan by carbon monoxide studies alone, andcytochromes b558 and d show coordinate synthe-sis under a variety of at least superficially un-

related growth conditions listed above. Re-cently, a report has appeared on the stimula-tion of cytochrome synthesis in a cya- mutantofE. coli, grown on glucose, by the addition ofcyclic adenosine 3',5'-monophosphate (AMP) tothe growth medium (58). Specific type b cyto-chromes were not identified, and levels of cyto-chrome d were not recorded, but clearly a rolefor cyclic AMP has been observed and moreintensive work with both cya- and crp- mu-

tants should be performed.An additional control, that of the timing of

synthesis during the cell cycle, has again onlybeen briefly studied. Using synchronous cul-tures, it has been concluded that, for certainmembrane proteins, succinate dehydrogenaseand cytochrome b,, either the conversion of in-active to active forms, or the coordinate expres-sion ofthe genes responsible for their synthesis,is controlled by a protein that is only synthe-sized or activated in turn at a specific phase ofthe cell cycle (305, 306). Clearly, further work isrequired to identify the time in the cell cycle atwhich other well-characterized membrane pro-teins appear and to differentiate between thetimings of apoprotein synthesis and the appear-ance of holoenzyme with functional activity.The factors regulating the relative activity of

the various routes of electron transport, whenalternative pathways coexist in the membrane,are again poorly defined. It appears that ubi-quinone is an obligatory redox carrier for oxy-

gen-dependent electron transport, since mu-tants deficient in their ability to synthesizeubiquinone show negligible oxidase activities(85, 299). Pudek and Bragg have suggested, on

the basis of cyanide inhibition studies, that,when both cytochromes o and d are present,

they function simultaneously as oxidases (322),and that the relative flow of reducing equiva-lents through the two electron-carrying limbs isa function oftheir pool size in the membrane.The presence of alternative routes for elec-

tron transport in E. coli that involve differentredox carriers whose relative concentration andfunctional activity vary with changes in thegrowth conditions makes the interpretation ofsome previous work in the literature, where thegrowth conditions were not rigorously definedor controlled, difficult to interpret. For exam-ple, the functional relationship of crystallinecytochrome b, (95) and of the three thermody-namically characterized type b cytochromes,with midpoint potentials at pH 7.1 of about-50, + 110, and +220 mV (170), to the proposedredox carriers in Fig. 3 is not known. In addi-tion, the functional significance of variouspreparations derived from solubilized mem-branes (e.g., references 29, 49, 50, 142, 168, 169)is difficult to interpret. Indeed, of all the redoxproteins associated with these two aerobic respi-ratory chains, the only one to have been exten-sively purified and characterized to date, is the1-lactate dehydrogenase (125, 228, 366). Theseconsiderations also apply to an evaluation ofthe scheme proposed earlier for the aerobic elec-tron transport chain of E. coli (85). These con-clusions were based on experiments performedwith a ubiquinone-deficient mutant, grownwith glucose as carbon source, whose mem-branes contain all the cytochromes shown inFig. 3, and it was proposed that ubiquinone hada double location as a redox carrier, both beforeand after cytochrome b. These results canequally well be interpreted by the schemes pro-posed in Fig. 3 and by assuming that type bcytochromes exist on either side of a quinonepool that contains both ubiquinone and mena-quinone (see Fig. 4) in particles from thesecells.The need to control and define the growth

conditions of the cell cannot be overempha-sized. Glucose is commonly used as the pre-ferred carbon source for growth, since highyields of cell mass are obtained, but it is partic-ularly difficult to characterize and differentiatebetween the various electron transport chainsthat are synthesized and operate under theseconditions. During aerobic growth in the pres-ence of glucose, cells contain concentrations ofcytochromes b558 and d that are intermediatebetween those found in anaerobically growncells and those found in cells grown aerobicallyon a nonfermentable carbon source. Such cellsexhibit a decreased efficiency of oxidative phos-phorylation, which can be increased by the ad-

BACTERIOL. REV.


http://mm

br.asm.org/

Dow

nloaded from



(a) (b)Membrane Cytoplasm Membrane Cytoplasm

Fp NADH+H+F NADH+H+

Fe/S formate Fe/S formate

MK 2Hfumarote cyt.b fumorate

f umorote fumaratereductase reductose

succinate succinate

FIG. 4. Proposed functional organization of the redox carriers responsible for anaerobic electron transportwith fumarate as terminal acceptor in E. coli. Scheme (a) allows for the oxidation of reduced coenzymeswithout concomitant ATP synthesis and occurs in the heme-deficient mutant. In scheme (b) the additionalredox components convert this scalar sequence into a vectorial reaction catalyzing netH+ translocation linkedto oxidation-reduction. Abbreviations: MK, Menaquinone; otherwise, as in the legend to Fig. 2.

dition of cyclic AMP to the growth medium(165). In addition, energy-dependent reversal ofelectron flow through the NADH dehydrogen-ase region of the respiratory chain could not bedemonstrated in particles from glucose-growncells (316). This suggests that the synthesis of afunctional proton-translocating NADH dehy-drogenase is sensitive to catabolite repression.However, other growth conditions (e.g., sulfate-limited continuous culture of strain K-12 withglycerol as carbon source) are also known toresult in a similar phenotype and, clearly,further work is required to differentiate be-tween those changes that are regulated bycatabolite repression and those that are regu-lated by other mechanisms.A second poorly defined growth condition is

that in which the cells are allowed to becomeoxygen limited. Although under anaerobicgrowth conditions the cells produce ATP eitherby fermentation or by electron transport linkedto N03- or fumarate reduction, they apparentlysynthesize all the cytochromes shown in Fig. 3and, when supplied with oxygen, demonstrateoxygen-dependent proton translocation and as-sociated ATP synthesis, with an efficiency thatsuggests the operation of one potential site ofenergy conservation. Anaerobically grown

cells, when incubated under nongrowing condi-tions in the presence of oxygen, develop oxidaseactivities and demonstrate efficiencies of oxida-tive phosphorylation that are similar to aerobi-cally grown cells in a process that requires denovo protein synthesis (70). As described in thenext section, fumarate, produced from the ca-tabolism of glucose, can serve as a terminalacceptor for electron transport-dependent ATPsynthesis under anaerobic conditions. Cellsgrown aerobically but with limited oxygenavailability synthesize all the componentsshown in Fig. 3 and have properties which sug-gest that the NADH dehydrogenase is non-pro-ton translocating, with oxygen as the terminalelectron acceptor, but it remains to be estab-lished whether these cells have an oxygen-inde-pendent pathway for the reoxidation ofNADHlinked to the reduction of endogenously synthe-sized fumarate with concomitant ATP synthe-sis.Anaerobic electron transport. (i) To fuma-

rate. The presence of fumarate as an electronacceptor enables E. coli to grow anaerobicallyon glycerol and L-a-glycerophosphate as thesource of carbon. Under these conditions, ananaerobic L-a-glycerophosphate dehydrogen-ase, distinct from the enzyme already dis-

VOL. 41, 1977


http://mm

br.asm.org/

Dow

nloaded from



cussed, and fumarate reductase, distinct fromsuccinate dehydrogenase, are induced (Table 1;177, 220). So far, the following limited informa-tion is available about this electron transportchain: (i) the oxidation of L-a-glycerophosphateand formate with fumarate reduction can beused for the accumulation of sugars and aminoacids in cells (338, 375) and right-side-out vesi-cles (42, 266); (ii) NADH-, L-a-glycerophos-phate-, and formate-induced fumarate-depend-ent atebrin quenching has been observed ininside-out vesicles (148, 374); (iii) fumarate-de-pendent proton translocation with an -k H+/2eratio of about 1 has been reported for cellsoxidizing endogenous substrates (57); and (iv)an enzyme complex that catalyzes L-a-glycero-phosphate-dependent fumarate reduction (263)with concomitant ATP synthesis (264, 266) hasbeen isolated from the cytoplasmic membrane,but not extensively characterized. The use ofmutants has established that menaquinone isrequired for fumarate reduction (298) and hasshown that this electron transport chain plays akey role in uracil production in cells grownanaerobically on fermentable substrates by al-lowing the anaerobic oxidation of dihydrooro-tate to orotate. Although a heme-deficient mu-tant could grow anaerobically on glycerol withfumarate (374), suggesting that cytochromeswere not required for electron transport, fur-ther work showed that fumarate-dependent an-aerobic active transport could only be demon-strated in cells grown in the presence of 5-aminolaevulinic acid and hence functional cyto-chromes (375).The information so far available is summa-

rized in Fig. 4, which shows two schemes toaccount for these experimental observations.The schemes differ in the proposal that an uni-dentified cytochrome, possibly with other redoxcomponents, converts a membrane-bound elec-tron transport chain catalyzing a scalar se-quence of oxidation-reduction reactions (Fig.4a) into a vectorial reaction sequence (Fig. 4b)in which 2 H+ are translocated across the cyto-plasmic membrane for each fumarate moleculereduced. Thus, both pathways are capable ofcatalyzing the oxidation ofNADH and formate,but only the H+ translocating route establishesa protonmotive force that can be used subse-quently for ATP synthesis and other secondaryenergy-requiring processes. In cytochrome-defi-cient cells, fumarate allows glycolysis to con-tinue by acting as an electron acceptor for thereoxidation ofNADH, and only ATP, generatedby glycolysis and hydrolyzed by the ATPase,can be used for transport. In cytochrome-suffl-cient cells, the energy for transport can come

from either anaerobic electron transport withfumarate as acceptor or from ATP hydrolysis.Clearly, further work is required to validatethese proposals. It is also well established thatcells grown anaerobically on glycerol with fu-marate demonstrate oxygen-dependent protontranslocation and associated energy-linkedfunctions (42, 57, 148, 338, 374, 375). The pointat which electron flow from this anaerobic elec-tron transport chain interacts with the aerobicredox carriers and the factor(s) that regulatethe relative flow of reducing equivalents downthe two pathways, when both oxygen and fuma-rate are present, have yet to be determined.

(ii) To nitrate. As will be seen, this is thebest characterized bacterial proton-translocat-ing electron transport chain so far studied, andthe various methods used to determine thefunctional organization of the membrane-bound redox carriers serve as a model for fur-ther investigations. When incubated under an-aerobic conditions in the presence of NO3- andmolybdenum, E. coli synthesizes a membrane-bound electron transport chain that allowsgrowth on a variety of nonfermentable sub-strates, like n-lactate, as sole carbon source(e.g., 367). These growth conditions result inthe specific induction of two membrane-boundredox carriers cytochrome bNO,,P- and NO3- re-ductase, whose synthesis is repressed by thepresence of oxygen. Interestingly, and unlikeParacoccus denitriftcans, growth in the pres-ence of high concentrations of KNO3 (0.1 M)does not result in the coordinate induction ofnitrite reductase, which is only synthesized atlower concentrations of KNO3 (<10 mM) (77,418).

If potassium selenite is also included in thegrowth medium, high levels of a membrane-bound formate dehydrogenase activity are alsoinduced (Table 3). Formate dehydrogenase wasisolated from E. coli grown under these condi-tions (107). The purified enzyme had a molecu-lar weight of about 600,000 and containedheme, molybdenum, selenium, nonheme iron,and acid-labile sulfide. The enzyme containedthree polypeptides, a, f3, and 'y, in approxi-mately a 2:2:1 molar ratio, of molecular weight110,000, 32,000, and 20,000, respectively; onlythe a-polypeptide contained significantamounts ofselenium, but the distribution oftheother redox centers in the polypeptides was notreported. It has been argued that formate dehy-drogenase is the normal physiological donor ofreducing equivalents to NO3- reductase (341)and, indeed, formate dehydrogenase and N03-reductase can readily be separated from eachother (407); on subsequent recombination, in

BACTERIOL. REV.


http://mm

br.asm.org/

Dow

nloaded from



the presence of endogenous quinone, an in,vitro reconstruction offormate-dependent NO3-reduction can be achieved (106). However,other physiological reductants, e.g., NADH, i-

a-glycerophosphate, and Dlactate, also inter-act with this membrane-bound electron trans-port chain via their respective dehydrogenases,and N03--dependent proton translocation incells (57, 130) and particles (148), as well asN03--dependent solute uptake (42, 338) linkedto the oxidation of these compounds, has beendemonstrated. Although NO3- reductase syn-thesis is repressed by oxygen, 02-dependent cy-tochrome oxidation (145), proton translocation(130, 148) and transport (42) can still be demon-strated in cells grown anaerobically in the pres-ence of NO3-, indicating that aerobic electrontransport is still functional in these cells. Whenoxygen and NO3- are added simultaneously toan anaerobic suspension of cells grown underthese conditions, oxygen is reduced first, fol-lowed by NO3- (145).The nature of the quinone component in-

volved in anaerobic NO3- reduction is notknown. Under these growth conditions, bothubiquinone and menaquinone are synthesized.and, although ubiquinone analogues are moreefficient than menaquinone analogues in the invitro reconstruction experiments describedabove (106), ubiquinone-deficient mutants growjust as well as parental strains on glucose an-

aerobically in the presence of NO3- (85). Itseems likely, therefore, that either ubiquinoneor menaquinone functions under these condi-tions.The type b cytochrome involved in anaerobic

electron transport to NO3- is specifically in-duced by NO3- (341) and is genetically (341,342) and kinetically (145) distinguishable fromother type b cytochromes synthesized by E.coli. In addition, work with heme-deficient mu-tants has demonstrated that, although N03reductase is synthesized, and in part incorpo-rated into the cytoplasmic membrane duringanaerobic growth in the presence of glucosewith nitrate, apocytochrome bNJ3 is not stablein the absence of heme synthesis (72, 130, 217,252). Evidence has already been presented toshow that the apocytochromes of the aerobicelectron transport chain are synthesized and in-corporated into the cytoplasmic membrane inthe absence of heme synthesis (143, 149, 217,332) and the results, therefore, indicate that, ofall the membrane-bound cytochromes synthe-sized by E. coli, heme synthesis in some wayregulates the synthesis (or stability) of apocy-tochrome bM" specifically.The solubilization and purification of nitrate

reductase from the cytoplasmic membrane ofE.coli, as assayed by the anaerobic, N03--depend-ent reoxidation of reduced viologen dyes, hasbeen reported by several workers. A summaryof the various procedures used and the charac-teristics of the preparations obtained are givenin Table 4. Nitrate reductase is a molybdenum-containing iron sulfur protein composed of twononidentical subunits designated a and (3. Insome preparations, two (8 subunits are seen;however, one of these is apparently a proteo-lytic degradation product of the other (255; R.A. Clegg, personal communication). The nativeenzyme can exist as either a monomer (la:1/3)or a tetramer (4a:4,8). Active nitrate reductasecan also be prepared in association with a thirdsubunit, y, which is heme containing and pre-sumably equivalent to cytochrome bN3T (72,106, 107, 250); the three subunits appear to bepresent in a 1:1:1 molar ratio. The precise con-tent and subunit localization of the molybde-num and iron-sulfur center(s) in nitrate reduc-tase have not yet been determined; however,two groups have reported electron paramag-netic resonance spectra of purified enzymepreparations. DerVartarian and Forget (97)obtained complex spectra with one preparation(120) which were interpreted as showing sig-nals due to (i) Mov-MOUII interconversions and(ii) nitrogen hyperfine structure from an NOcomplex, perhaps with iron, in the enzyme.Neither of these features was observed in sub-sequent work with a less-degraded enzymepreparation (72), in which two signals due toMov were seen, both signals reducible by so-dium dithionite and oxidizable by nitrate; sig-nificantly, one of the signals showed interac-tion of Mov with a proton exchangeable withthe solvent (56).N03--dependent proton translocation associ-

ated with the oxidation of various substratesadded to starved cells has been reported, andthe results obtained indicate - H+/2e- ratios of4 for L-malate oxidation and 2 for the oxidationof glycerol, succinate, and 1-lactate (130).Measurements of the -- H+/2e- ratio with for-mate as the reductant were complicated by pHchanges associated with formate uptake andC02 production, but it was possible to concludethat the site offormate oxidation is on the inneraspect of the cytoplasmic membrane and the --

H+/NO3- ratio was greater than 2 (130). Oxy-gen-dependent proton translocation was alsoobserved in cells grown anaerobically withNO3- and -- H+/O ratios of 4, for the oxidationof L-malate and formate, and 2, for the oxida-tion of other substrates were obtained (130).These various experimental results are sum-

VOL. 41, 1977


http://mm

br.asm.org/

Dow

nloaded from


TABLE 4. Summary ofdata on the purification and properties of nitrate reductase (EC 1.7.99.4) from E. coli

Enzyme prepared by:

Alkali-acetone Precipitation Solubilization with deoxy- Triton X-100 solubiliza-Crecipitation of ikaline-heat from Triton X- cholate (Enoch and Lester tion (Clegg [72])b

Characteristic membranes, treatment 100 extracts of [106, 1071)asolubilization MacGregor et cells by anti-with sodium al. [255]; See bedllsyatoOdeoxycholate also [390]) (Maregf:or I II III I II III(Forget [120]) t250])

Native enzymeMol wtGel filtration 300 K 720 K I 740 K 230 KElectrophoresis 320 KUltracentrifuge 773.6 K 498 K 880 K c5OOK 220 K

Isoelectric point pH 4.2 More acidicthan I

Metal content Mo (1.5); Fe Mo (3.2); Fe(atoms/mol) (20); S (20)

Heme content 6.7 3.2 15(nmol/mg protein)

Subunits mol wta 142 K 142 K 155 K 155 K 155 K 150 K 150K 150K1 (32) 58 K (60 K) 60 K 63 K 63 K 63 K 67 K 67K 67K

(65K) (65K) (65K)v none 19.5 K 19K 19 K None None 20K None

Molar ratio subunits 1:1:0 1:1:2 Less y sub- 4:4:0 2:2:2 1:1:0a: (,18 +/2):Y unit

than Ia Preparations I and II were separated by electrophoresis and chromatography; preparation III was isolated from

preparation I after alkaline-heat treatment.b Various preparations, isolated from sucrose density gradients, were analyzed.

marized in Fig. 5, and proposals are made forthe functional organization of the redox car-riers in the cytoplasmic membrane. The essen-tial features of the scheme are that (i) the poly-peptide components of both NADH and formatedehydrogenase are organized in the membraneto allow for the translocation of 2 H+ per qui-none molecule reduced and (ii) quinol (H car-rier) and cytochrome bN03 with NO3- reductase(e- carriers) are similarly arranged to give pro-ton translocation during oxidation-reduction.In the presence of oxygen, presumably cyto-chromes b5, and o or cytochromes b558 and d(Fig. 3) serve as electron carriers in place ofcytochrome b N03 and nitrate reductase. Thereare several unresolved questions concerningthis scheme including (i) the nature ofthe typeb cytochrome involved in the proton-translocat-ing formate dehydrogenase activity; (ii)whether this type b cytochrome is required forproton translocation associated with the NADHdehydrogenase; (iii) the relationship betweenthe proton-translocating NADH and formatedehydrogenase activities found in cells grownanaerobically with NO3- and the correspondingactivities found in cells grown both aerobicallyand also anaerobically with fumarate as termi-nal electron acceptor.Because of the relative simplicity of the ter-

minal electron-transporting limb involving cy-tochrome bN03 and NO3- reductase and theease with which the various polypeptides canbe isolated and identified, considerable atten-tion has been directed towards assessing thespatial orientation of the cytochrome bNW3-ni-trate reductase complex in the cytoplasmicmembrane. That the site of NO3- reduction,and presumably of binding to NO3- reductase,cannot be on the cytoplasmic face of the mem-brane (130) is suggested by the following exper-imental observations: (i) the observed rate ofreduced benzylviologen-dependent NO3- reduc-tion in whole cells or protoplasts is three ordersof magnitude higher than the observed rate ofpassive diffusion of NO3- into protoplasts sus-pended in isosmotic KNO3 containing valino-mycin; (ii) the failure to detect antiport systemsfor NO3- with NO2- or OH-; (iii) the lack ofeffect of a transmembrane pH gradient on thecompetitive inhibition by aside of NO3- reduc-tase in whole cells (see reference 310); and (iv)the absence of NO3- accumulation and, in fact,NO3- exclusion in a hemeless mutant ofE. coliin which membrane-bound NO3- reductasewas, nevertheless, active with artificial donors(217). Using (i) closed-vesicle preparations ofE.coli with a defined sidedness, protoplasts (sameorientation as the original cell) and vesicles

62 HADDOCK AND JONES BACTERIOL. REzV.


http://mm

br.asm.org/

Dow

nloaded from



Periplasmicspace

FIG. 5. Proposed functional organization of theredox carriers responsible for anaerobic electrontransport with NO3- as terminal electron acceptor inE. coli. Note the separate polypeptide components ofthe proton-translocating NADH- and formate-dehy-drogenases; it is not known whether some of thecomponents are common to both segments. Thescheme is modified from Garland et al. (128). Abbre-viations: Mo, Molybdenum-containing polypeptide;Se, selenium-containing polypeptide; a, (3, subunitsof nitrate reductase; otherwise as in the legends topreceding figures.

prepared by sonication (inside-out with respectto the original cell), (ii) a specific nonpenetrantlabel for tyrosine residues in exposed proteins(lactoperoxidase/H202-mediated incorporationof 125p) and (iii) the knowledge that antibody toN03- reductase specifically precipitates thepolypeptides of cytochrome bN03- (y) and N03-reductase (a + (), Boxer and Clegg were ableto show that the y subunit was located on theperiplasmic side of the membrane, and the a

subunit was located on the cytoplasmic side ofthe membrane; the relative sidedness of the (8subunit could not be determined with the samecertainty (43). In the light of these experimen-tal observations and in the absence of any di-rect evidence that N03- reductase is exposed onboth surfaces of the cytoplasmic membrane, ithas been proposed (128) that N03- approachesthe active center of N03- reductase via a cleftor well, as shown in Fig. 5. An important conse-quence of this proposal is that N03- reductase

must catalyze a vectorial reaction in conveyingboth e- and H+ across the membrane. Clearly,further work is required to test the validity ofthese proposals and to determine whether N03-reductase does indeed have a transmembraneorientation by using, for example, permeantand nonpermeant analogues ofviologen dyes asreductants of N03- reductase in vesicle prepa-rations of known sidedness.Mutants defective in N03- reductase activity

have been widely used in an attempt to charac-terize the factors regulating the synthesis andassembly of this anaerobic electron transportchain. As indicated in Table 1, such mutantshave been isolated by three general methods: (i)resistance to chlorate during anaerobic growthconditions, (ii) differential ability to grow onnonfermentable substrates aerobically and an-aerobically with N03-, and (iii) inability to re-.duce NO3- to NO2-, as judged by a dye overlaytechnique. Seven genetic loci have been identi-fied to date ofwhich four, the chA, chiB, chlD,and chME genes, are pleiotropic mutations in-volving not only the loss of membrane-boundformate-dependent N03 reductase activity butalso the loss of formate hydrogenlyase activity(formate dehydrogenase with hydrogenase),which is apparently located in the periplasmicspace (Table 1). All of these mutants are able togrow normally under aerobic conditions withboth fermentable and nonfermentable carbonsources. Considerable attention has been di-rected towards assessing the gene products de-fective in these various mutants. Four generalmethods have been used: (i) the phenotypic res-toration of functional activity by changing thegrowth conditions of the cell (135; Table 2); (ii)the identification of missing polypeptides inTriton-solubilized membranes from cells by so-dium dodecyl sulfate-polyacrylamide gel elec-trophoresis using point mutants (253) and dele-tion mutants (334), although a direct causalrelationship between a missing or an alteredpolypeptide and the loss offunctional activity isdifficult to demonstrate unambiguously; (iii) invitro complementation assays involving themixing of soluble extracts, i.e., nonsedimenta-ble on prolonged centrifugation, from the var-ious mutants and testing for the restoration ofparticulate N03- reductase activity as devel-oped by Azoulay and co-workers (e.g., refer-ences 26, 27, 256, 288, 329; but see also reference254) and (iv) the specific identification of miss-ing polypeptides from the cytochrome b °3-N03- reductase complex by challenging Tritonextracts of cells with antibody to purified N03-reductase (249).

Interpretation of the data provided by thesedifferent methods is difficult and further com-

VOL. 41, 1977


http://mm

br.asm.org/

Dow

nloaded from



plicated by the possibility that individual chlloci may be functionally subdivisible. Thus,Venables has shown by fine-structure geneticanalysis of chl loci that chIA, chMB, and chiEare divisible into two, three, and two comple-mentation groups, respectively, which, in thecases of chIA and chME may represent distinctgenes (404). Indeed, chME mutants exhibit avariety of phenotypes and, of three mutantsassayed for their ability to produce NO3- reduc-tase specifically, using the antibody precipita-tion technique, two made reduced amounts ofprecipitable enzyme and one made none (249).In addition, the variable amounts ofNO3 reduc-tase, formate dehydrogenase, and type b cyto-chrome that are retained by chiC mutants, aswell as the possibility that these amounts arerelated in a polarized manner to the map posi-tion of the chiC gene involved, have led to thesuggestion (141, 342) that chiC may comprisean operon specifying these three components,but evidence for genetic complementation be-tween chiC mutants is, unfortunately, lacking.An attempt has been made to summarize thepostulated primary lesions in those individualchl mutants that have been extensively charac-terized in several different laboratories (Table5). However, it should be remembered that de-tailed biochemical investigations have usuallyonly been performed with a single mutantstrain thought typical of a particular chl locus.As mentioned above, the evidence so far availa-ble indicates that this is an oversimplificationand clearly a more detailed genetic analysis ofeach individual locus by fine-structure geneticmapping, as well as a rigorous biochemicalanalysis of different isolates reputedly charac-teristic of a single locus, are required before theproposals made in Table 5 can be accepted with-out reservation.A heme-deficient mutant has also been used

to study the factors regulating the synthesis of

TABLE 5. Proposed primary lesions in various chIrmutants ofE. coli

Gene Suggested primary lesion Refer-ences

chlA Synthesis of Mo-cofactor 249, 254.chlB Association factor (FA) required for at- 249, 254,

tachment or insertion of Mo-cofactor 329into nitrate reductase

chiC Structural gene for nitrate reductase, 141, 249possibly a subunit

chMD Processing of Mo (transport, redox 135llevel, incorporation into Mo-cofactor)

chiE Structural gene for nitrate reductase, 249, 253possibly y subunit

chiF Structural gene for formate dehydro- 136genase(?)

chIG Not known 136

functional NO3- reductase. When this mutantis grown anaerobically in the presence of N03-and absence of heme synthesis, apocytochromebN03- is not present in the membrane, and the aand (3 subunits of NO3- reductase are over pro-duced and chiefly located in the cytoplasm,though some enzyme is still membrane boundand susceptible to proteolytic attack (251). Ifheme synthesis is allowed to proceed, by theaddition of 5-amino-levulinic acid to the growthmedium, then the N03- reductase present inthe cytoplasm is incorporated into the mem-brane in a stable form in parallel with theformation of a functional type b cytochrome(252).Therefore, the following information is so far

available concerning the factors regulating thesynthesis and assembly ofa functional formate-dependent N03- reductase activity: (i) the mo-lybdenum-containing enzymes, soluble andparticulate formate dehydrogenase and N03-reductase, require the gene products of thechiD gene (probably for the transport of Mo),the chIA gene (for the synthesis of a Mo-cofac-tor), and the chMB gene (for the attachment orinsertion of Mo-cofactor into the apoenzymes)for the synthesis of functional enzymic activity,though apoprotein synthesis apparently occursin the absence of the Mo-processing machinery;(ii) apocytochrome bNW- synthesis requires con-comitant heme synthesis, possibly the productof the chiE gene, and probably occurs in thecytoplasmic membrane; (iii) apo-N03- reduc-tase synthesis requires the product of the chiCgene, occurs in the cytoplasm, and functionalcytochrome bNW- is required for correct bindingand insertion into the membrane; (iv) onlywhen membrane bound and functional is N03-reductase protected from proteolytic attack.Although E. coli has been studied exten-

sively by molecular biologists in the past, it isonly comparatively recently that attention hasbeen directed towards elucidating the mecha-nism of oxidative phosphorylation in this bacte-rium. The experimental evidence indicates thatthe various redox carriers synthesized by E.coli are functionally organized in the mem-brane into several proton-translocating oxido-reduction segments. Those redox segments thathave so far been identified and, in part, charac-terized include the following.

(i) A proton-translocating NADH-ubiqui-none oxido-reductase activity, produced andutilized under aerobic growth conditions, whichis apparently cytochrome independent, re-quires flavoprotein and iron-sulfur protein(s)and, at least superficially, resembles the equiv-alent mitochondrial enzyme. Under certaingrowth conditions, this protein is either modi-

BACTERIOL. REV.


http://mm

br.asm.org/

Dow

nloaded from



fied or replaced by an alternative enzyme suchthat the associated redox reactions are nolonger linked to net proton translocation (Fig.3).

(ii) A proton-translocating NADH fumarateoxidoreductase activity produced and utilizedunder anaerobic growth conditions in the pres-ence of fumarate, which is apparently cyto-chrome dependent and requires menaquinonein addition, presumably, to flavoprotein andiron-sulfur protein(s) (Fig. 4b). A proton-trans-locating NADH dehydrogenase is also synthe-sized and functional during anaerobic growthin the presence of nitrate. It is not known, atthis time, if this activity is the result of acytochrome-independent route (as shown inFig. 3 and 5) or a cytochrome- and menaqui-none-dependent route (equivalent to thescheme shown in Fig. 4b without fumarate re-ductase) for electron transport; indeed, it is pos-sible that both of these proton-translocatingredox segments for NADH oxidation can func-tion simultaneously during anaerobic electrontransport to nitrate and, indeed, under certainconditions during aerobic electron transport.

(iii) A proton-translocating formate dehydro-genase activity that requires, as minimal com-ponents, iron-sulfur protein(s), molybdenum,selenium, and a type b cytochrome. Althoughmenaquinone has been implicated as a carrierin formate-dependent fumarate reduction (Fig.4b), it remains to be established whether thiscarrier is an obligatory component of the pro-ton-translocating formate dehydrogenase activ-ity produced and functional during anaerobicgrowth in the presence of nitrate (Fig. 5).

(iv) A proton-translocating ubiquinol-nitrateoxidoreductase activity (Fig. 5).

(v) Two proton-translocating ubiquinol-oxy-gen oxidoreductase activities that possess dif-ferent electron-carrying limbs (Fig. 3).The following general conclusions can be

reached concerning the functional organizationof the various redox carriers in these proton-translocating redox segments. First, as mini-mal components in those redox segments thatresult in the direct reduction of an exogenousacceptor (e.g., nitrate, fumarate, or oxygen),the hydrogen-carrying limb contains a quinonespecies, and the electron-carrying limb con-tains two components, a type b cytochrome anda terminal reaction center (nitrate reductase,fumarate reductase, cytochromes o and d) capa-ble of interacting with and reducing the accep-tor. Second, during oxygen- and nitrate-de-pendent electron transport, an additional en-ergy conservation site is available to the bacte-rium through the activity of NADH and for-mate dehydrogenase, which results in the re-

duction of an endogenous membrane-bound ac-ceptor, presumably ubiquinone. The individualredox carriers present in the hydrogen- andelectron-carrying limbs of these dehydrogen-ases remain uncertain and, presumably, alter-native components function in different en-zymes, or in different forms of the enzyme,depending upon the control mechanisms thatregulate their synthesis and activity as thegrowth conditions and metabolic demands onthe cell vary. Third, although electron trans-port in E. coli may appear superficially to bemore complex than in mitochondria, this ap-parent complexity is due to the presence of alter-native and parallel pathways for electron trans-port in this bacterium.

If each of the proposed proton-translocatingredox segments is considered as a distinctbuilding block, then each block can fuinctioneither separately (e.g., in NADH-dependent fu-marate reduction or 1)-lactate-dependent ni-trate reduction) or sequentially, such that re-ducing equivalents can pass from a block of lowredox potential (e.g., the NADH dehydrogen-ase) to a block of high redox potential (e.g.,ubiquinol-oxygen oxidoreductase), through car-riers of similar redox potential at the junctionof the two blocks, and thence to the terminalelectron acceptor. Thus, under any particulargrowth conditions, alternative and parallelpathways can exist in the membrane for thereoxidation of reduced coenzymes. For exam-ple, NADH oxidation, by oxygen, could proceedsimultaneously through one of the two proton-translocating redox segments (either menaqui-none dependent or menaquinone independent)or through the non-proton-translocating routeand then, via ubiquinone, through either thecytochrome b,56 with o or cytochrome b,% withd electron-carrying limbs to oxygen. Previousschemes in which the various redox carrierswere arranged in a linear sequence (85) areobviously too simple to account for the diversityof pathways available for electron transport inE. coli. The fourth general point to emergefrom these considerations is the need to identifyand characterize the redox carriers that aresynthesized by, and are functionally active in, aparticular strain of E. coli grown under con-trolled and defined conditions, before attempt-ing to interpret their metabolic significance tothe cell.Although much further work is obviously re-

quired, in the last few years sufficient progresshas been made, from a combined biochemicaland genetic approach, to suggest that E. colihas many advantages for studying the molecu-lar mechanism of respiratory driven protontranslocation and associated energy-linked

VOL. 41, 1977


http://mm

br.asm.org/

Dow

nloaded from



functions that are not found with classical mito-chondrial systems. It is of interest that thevarious redox carriers, which are synthesizedby E. coli and which are functionally activeduring aerobic electron transport, are orga-nized in the membrane into no more than twopotential energy conservation sites. AlthoughE. coli can synthesize two terminal electron-carrying limbs for the oxidation of ubiquinol byoxygen, it appears that this bacterium has notdeveloped an electron-carrying limb containingtypes c and a cytochromes, which is necessaryfor the functional activity of the additional po-tential site of energy conservation available tomitochondria and P. denitrificans (see below).

Paracoccus denitrificansThe gram-negative bacterium Paracoccus

denitrificans is a chemoorganotroph and facul-tative chemolithotroph, capable of using theoxidation of molecular H2 (116, 223, 233) andmethanol (87) as sole course of energy for auto-trophic growth. Many different organic com-pounds serve as sole carbon source for growth,but the metabolism is respiratory, never fer-mentative. Molecular 02, or N03-, and N02-,under anaerobic conditions (192, 234), serve asterminal electron acceptors. N03- is reduced tonitrous oxide and, ultimately, to molecular N2under anaerobic conditions.

Aerobic electron transport. When grownaerobically, P. denitrificans synthesizes anelectron transport chain which, in its compo-nents and their functional organization, moreclosely resembles the electron transport chainof the inner mitochondrial membrane than thatof any other bacterium (for a review, see refer-ence 193). The essential features that charac-terize the aerobic electron transport chain ofthis bacterium, which is located in the cytoplas-mic membrane (357), are summarized in Fig. 6aand include: an energy-dependent nicotinamidenucleotide transhydrogenase activity (23); anNADH dehydrogenase activity that shows en-ergy-dependent reversal (24), sensitivity to ro-tenone (183) and piercidin A (240) at low con-centration, and an iron-sulfur center(s) with acharacteristic g = 1.94 signal in electron para-magnetic resonance spectroscopy at 770K, de-creased in content under conditions of iron-lim-ited growth (184); ubiquinone-10 as the solefunctional quinone of the respiratory chain(357) and at least two kinetically and spectrallydistinguishable type b cytochromes, possiblytwo type c cytochromes, with cytochromes a +a3 serving as the terminal oxidase, althoughcupric copper, associated with the mitochon-drial cytochrome oxidase, was not present in P.

denitrificans particles (184, 240, 357, 364, 405,406). In addition, electron transport is sensitiveto low concentrations of antimycin A (357) andcarboxin (396). All these features are typical ofthe mitochondrial inner membrane, and indeedthe cytoplasmic membrane of P. denitrificansshows further similarities to this membrane inits content of phospholipids and fatty acids(193). It has been claimed that cytochrome o, atype b cytochrome that binds carbon monoxidein the original terminology and definition ofCastor and Chance (65), serves as a terminaloxidase in P. denitriftcans, as it does in certainmitochondria (349). However, there is no evi-dence from kinetic data, using a stopped-flowspectrophotometer, to suggest that a type bcytochrome is acting as a terminal oxidase in P.denitrificans (240).The cytoplasmic membrane of P. denitrifi-

cans is known to have a low proton conductance(355), and Scholes and Mitchell (354) haveshown that protons are ejected through the cy-toplasmic membrane after the addition of asmall amount of oxygen to a stirred anaerobicsuspension of cells. Limiting -*H+/O quotientsof 8 were obtained with cells oxidizing endoge-nous substrates (354). This suggests that theelectron transport chain of P. denitrificans,like that of mitochondria, is arranged in threeproton-translocating segments that can be usedto drive ATP synthesis, plus a fourth, the trans-hydrogenase, that probably cannot do so underphysiological conditions. This was confirmed insubsequent experiments using starved cells oxi-dizing known added substrates, where it wasshown that the oxidation of succinate or glyc-erol resulted in the translocation of 4 H+ per 0consumed, and the oxidation of Lascorbate, viaN,N,N',N'-tetramethyl-p-phenylene diamine.(TMPD), was associated with the translocationof about 2 H+ per 0 consumed (240). It was alsoshown in this work that, in cells harvested inthe early-exponential phase of batch growth,the oxidation of L-malate was associated withan -*H+/O ratio of about 8, which decreased toa value closer to 4 in cells harvested in thestationary phase of growth; the sensitivity of L-malate oxidation to piercidin A inhibition wassimilar in cells harvested from all phases of thegrowth cycle (240). This suggests that protontranslocation associated with both the transhy-drogenase and the NADH dehydrogenase canbe modified under certain growth conditions.The data also indicate that sensitivity towardspiercidin A and the occurrence of proton trans-location associated with the NADH dehydro-genase are distinct and separate features ofthisregion of the respiratory chain; this is a point of

BACTERIOL. REV.


http://mm

br.asm.org/

Dow

nloaded from



(a)aerobic electron transport

NADPH -- NADH*H+ .W

succinatecyt.b2 cyt.c1

10

2

_-P FP --_Fe/S_ 10_--

cyt.b565 cyt.c

L,.c -glycerophosph ate

2 °2 +2H

H20

(b anaerobic electron transport

'202 .2H+

2cyt.c

Fe/S;Mo Cyt.cvtredutae nitritereductase

-i ~~re ductase

2H++NO3 H20+NC0 2H++N0 H20+NO(?)FIG. 6. Proposed electron transport chains in Paracoccus denitrificans. Scheme (a) is the proposed linear

sequence ofredox carriers associated with aerobically grown cells. Although there is no experimental evidenceavailable atpresent, it is proposed that these carriers have a similar functional organization in the cytoplasmicmembrane to those ofthe inner mitochondrial membrane (shown in Fig. 2) and are so arranged as to allow thesynthesis of3 mol ofATP per mol ofNADH oxidized. Scheme (b) is a summary ofthe known redox carriersfound in cells grown anaerobically in the presence of NO3-. During these growth conditions, there is anincrease in the synthesis of cytochrome o (see text) which presumably can act as a terminal oxidase, yet itssignificance to the metabolic activity ofthe cell during anaerobic growth remains obscure. Nothing is known ofthe functional organization ofthese redox carriers in the membrane; indeed NO2- reductase is not membranebound, and the specific identity of the type b cytochrome has not been established. In addition, the number ofpotential sites of energy conservation associated with anaerobic electron transport has not been determined.The schemes are modified from John and Whatley (193) and the abbreviations used are defined in the legendto Fig. 2.

some controversy in analogous studies with mi-tochondria derived from the yeast Candidautilis, in which it has been claimed that thesephenomena are either independent (147, 183) orinterdependent (75, 140) properties. Of particu-lar interest is the fact that particles preparedfrom cells grown under iron-limited conditions,though containing only a very small amount ofnonheme iron, had normal NADH oxidase ac-tivity and performed NADH dehydrogenase-de-pendent ATP synthesis by oxidative phospho-rylation as effectively as normal preparations;unfortunately, the sensitivity ofNADH oxidaseactivity to rotenone or piercidin A was not re-ported (184).A yoax of approximately 73 g of cells per

mol of oxygen consumed has recently been de-termined for glycerol-limited continuous cul-tures ofP. denitrificans (C. Edwards and C. W.

Jones, unpublished data). This value is consid-erably higher than has been reported by otherworkers for carbon-limited cultures of this or-ganism (403) and is also very much higher thanhas been obtained with organisms whoseNADH oxidases contain only two proton-trans-locating segments (101, 110, 166, 385), thus sug-gesting that in P. denitrificans the terminalcytochrome c -* aa3 limb contributes towardsATP synthesis.One of the advantages of working with P.

denitrificans is the relative ease with whichsubcellular particles capable of oxidative phos-phorylation can be isolated (182, 192, 223). In-deed, phosphorylating particles, prepared fromthe cytoplasmic membrane of P. denitrificans,show a mitochondrial type of respiratory con-trol in oxygen electrode experiments that israrely observed in other phosphorylating bacte-

VOL. 41, 1977


http://mm

br.asm.org/

Dow

nloaded from



rial preparations (190, 191; see also references105, 195, 196). These particles necessarily havean orientation that is the reverse of that of thecytoplasmic membrane of the intact cell, thatis, they are inside-out vesicles. Recently, right-side-out vesicles, in which the vesicle mem-brane has the same orientation as the cytoplas-mic membrane of the intact cell, have beenprepared (60, 412) and, clearly, a comparativestudy of these two preparations will be of use inelucidating the relative sidedness ofthe variousredox components in the cytoplasmic mem-

brane.Anaerobic electron transport. P. denitrifi-

cans can grow anaerobically using either N03-or NO2- as a terminal electron acceptor, theultimate product being N2 gas (234). Under an-

aerobic growth conditions in the presence ofN03-, several alterations occur to the aerobicrespiratory chain which include: (i) the disap-pearance of cytochrome a + a3; (ii) increasedsynthesis of type b and c cytochromes, includ-ing cytochrome o, although the additional cyto-chromes have not been characterized exten-sively; (iii) the appearance of a membrane-bound N03- reductase activity; and (iv) thesynthesis of a soluble NO2- reductase activity(192, 234, 349, 357). The various results ob-tained by these investigators and the interrela-tionships between the different redox carriersare summarized in Fig. 6b.The respiratory^N03- reductase has been sol-

ubilized and purified from the cytoplasmicmembrane (119, 235) and shown to be a molyb-denum-containing' iron-sulfur protein (121).N02- reductase has also been purified from P.denitrificans. The enzyme is a two-heme cyto-chrome, containing both a type c (with an unu-

sual double a-band) and a d-like heme and, inaddition to its N02- reductase activity, it alsoexhibits an oxidase activity; the enzyme is notmembrane bound (236, 297). Interestingly, a"blue"-copper protein co-purifies with N02- re-

ductase in the early stages of enzyme isolation,though the functional involvement ofthe "blue"copper protein has not been determined (296).Although cytochromes a + a3 are no longer

pre~ent in cells grown anaerobically withN03t, the apparent oxidase activity of suchcells is similar' to that of aerobically growncells. It has been suggested, on the basis of a

carbon monoxide-binding assay, that cyto-chrome o synthesis is increased under anaero-bic growth conditions and could presumablyserve as the terminal oxidase (349); a kineticanalysis is required to support this proposal. Ifoxygen and N03- are added simultaneously toan anaerobic suspension of cells, grown anaero-

bically in the presence of NO3-, the oxygen isused first followed by the N03- (71). The controlmechanism(s) that regulates the relative flowof reducing equivalents down the various path-ways to the different terminal electron accep-tors has yet to be determined.

Comparatively little attention has been di-rected towards assessing the efficiency of oxida-tive phosphorylation associated with anaerobicelectron transport. In cells grown anaerobicallywith N03-, ATP synthesis from NADH-, butnot succinate-, dependent N03- reduction hasbeen demonstrated in phosphorylating parti-cles (192), and N03--dependent proton translo-cation has been observed in starved Whole cellsoxidizing unspecified endogenous substrates(H. G. Lawford and J. C. Cox, unpublishedobservations). There are no reports, to ourknowledge, of oxidative phosphorylation associ-ated with NO2- reduction; this is due, in part,to the experimental difficulties of growing cellsanaerobically in the presence of NO2-, which istoxic above a certain concentration, and in partto the fact that the soluble nitrite reductase islost during the preparation of phosphorylatingparticles (192).An attempt has been made to determine the

factors influencing the synthesis of the variousredox components and it appears that: (i) oxy-gen represses the synthesis of both N03- reduc-tase and NO2- reductase; (ii) N03-, but notNO2-, induces the synthesis of N03- reductase;(iii) N02-, either'added exogenously or derivedfrom N03- via N03- reductase, induces thesynthesis ofN02- reductase, so that cells grownanaerobically with N03- contain both NO3- re-ductase and NO2- reductase' activity; and (iv)cytochrome a + a3 synthesis is oxygen depend-ent (234, 249).As can be seen from the above brief review,

the electron transport chains ofP. denitrificanshave not, as yet, been explored in depth, butalready sufficient information is available(e.g., similarity to mitochondrial electrontransport chain, variability of redox carriersand efficiency of oxidative phosphorylationwith growth conditions, conspicuous lack ofwork with mutants) to indicate the profitabilityof more extensive investigations.

Azotobacter yinelandiiIn contrast to E. coli and P. denitrificans,

Azotobacter vinelandii (like its close relativesA. chrococcum and A. beijerinckii) is a nitro-gen-fixing, obligate aerobe; this .organism cantherefore use only molecular oxygen as a termi-nal oxidant for respiration. The energy-depend-ent, reductive assimilation of atmospheric ni-

BACTERIOL. REV.


http://mm

br.asm.org/

Dow

nloaded from


VOL. 41, 1977

trogen (174, 318, 423) is catalyzed by an enzymecomplex, nitrogenase, comprising two iron-sul-fur proteins -one of which contains molybde-num (61, 424). Purified nitrogenase from A.vinelandii, like that from various obligate orfacultative anaerobes, is readily inactivated byhigh concentrations of molecular oxygen. Thus,during growth, the organism is faced with theproblem ofmaintaining its intracellular oxygenconcentration at a level that is too low to inhibitnitrogen fixation, but which is high enough toallow adequate ATP synthesis. The respiratorysystdm of A. vinelandii, therefore, has a dualfunction, viz., the conservation of energy viaoxidative phosphorylation and the protection ofnitrogenase via the removal of excess oxygen.

Respiratory chain energy conservation. Therespiratory system ofA. vinelandii is located inthe extensively invaginated cytoplasmic mem-brane (174, 308) and is characterized by an ex-tremely rich complement of redox carriers.These include highly'active, flavin-dependentNADH, NADPH, and malate dehydrogenases(9, 200, 204), of which the former is known tocontain iron-sulfur centers (96), ubiquinone Q-8as the sole quinone component (200, 387), andat least six spectroscopically detectable cyto-chromes, viz., b.,0, c4, C5, a1, o, and d, pluscytochromes c551 and cms, which have been de-tected in low concentrations and are probablysubunits of the normally dimeric cytochromesC4 and c5, respectively (200, 201, 386, 395).

In addition, the particulate respiratory chaininteracts with a soluble nicotinamide nucleo-tide transhydrogenase (399) and also with var-ious soluble NAD+- and NADP+-linked dehy-drogenases (34, 200, 203, 204).

Respiration by A. vinelandii membranes isremarkably insensitive to classical inhibitors ofmitochondrial -electron transfer such as rote-none or antimycin A, but is readily inhibited


by 2-n-alkyl-4-hydroxyquinoline-N-oxide (167,201, 205) and, to a variable extent, by cyanide(194, 201, 214, 215,' 216). The ability of the latterto inhibit the oxidation of the artificial electronfdonors ascorbate-TMPD and ascorbate-2,6-dichlorophenol indophenol (DCPIP) much morereadily than the oxidation of physiological sub-strates (Ki = 0.5 juM versus <115 uM (194); seealso references 32, 201, 205, 207, 216, 386) hasled to the now generally accepted concept of abranched' respiratory' system in A. vinelandii(Fig. 7). This is fully supported by the differentsensitivities of the two branches towards car-bon monoxide (201),'the photodissociation pat-terns of the cytochrome oxidase-carbon monox-ide complexes (108), and the strikingly differentoxidation-reduction kinetics of the type -b and ccytochromes (201, 216). The cytochrome b dbranch thus resembles the terminal limb of theE. coli system (following the growth of thatorganism under oxygen-limited conditions; seeFig. 3), whereas the b-) C4, C5 -. a, o branch issimilar to the terminal cytochrome systems ofP. denitrificans (substituting a, for aa3; seeFig. 6) and of several major families of bacteria(199, 261). The qualitative response of this lat-ter pathway to cyanide has led to the sugges-tion (194) that cytoehrome oxidases o and a,may oxidize different type c cytochromes; sincethere is good evidence that cytochrome o isfunctionally associated with. C4 (216), cytochrome a, would therefore accept electronsfrom c5. However, recent claims that A. vine-landii particles devoid of cytochrome a, stillreadily oxidize reduced TMPD and DCPIP (215,282) and that this cytochrome may not be com-petent as a terminal oxidase in some bacterialsystems' (145) have cast doubts upon the in-volvement of cytochrome a, in the cytochromec-linked pathway.

In contrast, Kauffman and van Gelder have

Low [KcN]

I ¶flNADH-. 4Fe-S/FMN -------- o0

Malate Fe-S/Fp - Q mo d

NADPH F--S/ Fp 02FIG. 7. Respiratory system of A. vinelandii. Solid arrows represent pathways of electron transfer, the

broken arrow represents the cyanide-insensitive autooxidation ofcytochrome b, and the open arrow representsthe site of action of low -concentrations of cyanide. I, II, and III are the approximate sites of energy coupling(proton translocation) (after Downs and Jones [102D).


http://mm

br.asm.org/

Dow

nloaded from



recently provided convincing kinetic and spec-tral evidence that confirms the role of cyto-chrome d as the oxidase of the major terminalpathway (216). It has been proposed that theoxidized form of the enzyme exists in two differ-ent conformations, viz., an inactive conforma-tion that absorbs at 648 nm and a more open,active conformation (dx) to which cyanide bindspreferentially, the binding being greatly en-hanced under conditions in which the enzyme isturning over rapidly (214-216).A third terminal pathway to oxygen has re-

cently been proposed (216); this branch, whichis characterized by its extreme insensitivity tocyanide, carries only about 5% of the total elec-tron flux from NADH and many involve au-tooxidation of the type b cytochrome.The dehydrogenase end of the A. vinelandii

respiratory chain is particularly interesting,since it furnishes a rare example of an appar-ently allsostericNDPdehydroggnse (9).The latter is highly active, shows SigmnOidlsaturation kinetics with respect to NADPHI, ~~and is- competiey 1nhibited

by both adenine nucleotides (AMP > ADP >ATP: no loss of sigmoidicity) and NAD+ (withreversion to Michaelis-Menten kinetics). The[SI0.5 of the enzyme is severalfold higher thanthe Km values of the other membrane-bounddehydrogenases and thus ensures that the res-piratory chain rapidly oxidizes I{ADPW onlywhen the latter is present in abundance, aproperty that is probably of major importanceto the mechanism of respiratory protection (seebelow).

Carefully prepared respiratory membranepreparations from A. vinelandii exhibit elec-tron transfer-linked energy conservation, as ev-idenced by their abilities to catalyze either ATPsynthesis (10, 104), the quenching of atebrinfluorescence (103), or the active transport ofglucose (31, 32). It was concluded from thesestudies that a maximum of three energy-cou-pling sites are present, viz., at the level ofNADH dehydrogenase (site I), in the central Q

b region of the chain (site II) and on thecytochrome C4, c5 -- a, o terminal branch (siteIII); the cytochrome b -* d terminal branch andthe flavin-linked malate and NADPH dehydro-genases show no evidence of energy coupling.Classical respiratory control is detectable atsites I and II, but not at site II (105, 195, 196),thus reflecting the low energy conservation effi-ciency of site III in these membrane prepara-tions. In contrast, the activity of the uncoupledNADPH dehydrogenase is apparently modu-lated by the ambient energy charge (9).The recent measurement of respiration-

linked proton ejection by starved cells of A.vinelandii oxidizing added substrates (e.g., f-hydroxybutyrate, malate, or reduced TMPDand DCPIP) has indicated the presence of threeproton-translocating respiratory segments(102) at locations similar to those proposed forenergy conservation in isolated membranes.Furthermore, concentrations of cyanide suffi-cient to block the c4 c5 -- a, o branch (but notthe b -> d pathway) slightly decreased the--H+/O ratios that were observed during theoxidation of physiological substrates, thus con-firming that the cytochrome b -> d pathwaydoes not conserve energy and reinforcing theconcept that this branch constitutes the major,terminal route to oxygen. This latter conclusionis fully supported by the P/O ratios of approxi-mately 2 for the oxidation of NAD+-linked sub-strates that have been determined by directmeasurement of ATP synthesis after the addi-tion of oxygen to anaerobic whole-cell suspen-sions (225; but see also reference 28). It is clearfrom these results that the abilities of the ter-minal branches of the A. vinelandii respiratorysystem to translocate protons are dependentupon their redox carrier composition, particu-larly with respect to type c cytochromes. Thisconclusion is supported by the observationsthat a third proton-translocating segment ispresent in the respiratory system ofP. denitri-ficans, but not E. coli, and by the results ofcomparative studies on a variety of bacterialrespiratory systems (197).

Respiratory protection of nitrogenase. Thegrowth of Azotobacter under nitrogen-fixingconditions in the presence of a high PO2 is char-acterized by a higher respiratory activity(Qo,) and a lower molar growth yield (Yomaxor Ysubstrate) than during growth under con-ditions of oxygen limitation (11, 91, 92, 174,198, 289). Since both of these parametersare proportional to the overall efficiency of cel-lular energy conservation (predominantly res-piratory chain phosphorylation in this obligateaerobe), the above changes must reflect energywastage by the cell. The latter may occur eithervia the complete or partial failure of the respi-ratory system to couple electron transfer to theformation of the energized state (and subse-quently to the synthesis of ATP) or via theoperation of metabolic "slip reactions" (290,291), which spill excess energy by ATP hydroly-sis (either directly via ATPase, or indirectly viafutile metabolic cycles).The branched nature of the Azotobacter res-

piratory system is fully compatible with theformer possibility, since it would allow substan-tial variation in the efficiency of energy conser-

BACTERIOL. REV.


http://mm

br.asm.org/

Dow

nloaded from


VOL. 41, 1977

vation according to the exact route of electrontransfer. Thus, the exposure of oxygen-limitedpopulations of Azotobacter to excess oxygencauses a sudden cessation of nitrogen fixationand elicits a rapid increase in Q0,, togetherwith mobilization of reserve storage materialssuch as polysaccharide and poly-3-hydroxybu-tyrate, which are accumulated under oxygen-limited conditions (91, 188, 360). These changesare accompanied by the de novo synthesis bothof cytochrome d and of the NADH and NADPHdehydrogenases (99, 198), as well as the rapidloss of respiration-linked proton translocationand ATP synthesis at the level ofNADH dehy-drogenase (102, 198; C. W. Jones, unpublisheddata). It is clear, therefore, that these changeswould tend to lower the overall efficiency ofrespiratory chain energy conservation and thusproduce an increase in respiratory activity.Since the second proton-translocating segmentof the respiratory chain is still retained underexcess oxygen conditions (102), the efficiency ofenergy conservation would fall by a maximumof two-thirds, which would thus lead to a three-fold increase in Qo, (provided that the growthrate is unaltered). However, since the suddenexposure of nitrogen-fixing populations ofAzo-tobacter to excess oxygen leads both to a cessa-tion of growth and to at least a threefold in-create in Qo, (91, 198), it must be acknowl-edged that the observed changes in energy-coupling efficiency cannot alone account for theincrease in respiratory activity. The hypersen-sitivity of phosphate-limited cultures of A.chroococcum to oxygen (242) suggests that alarge adenine nucleotide pool is a prerequisitefor efficient respiratory protection and, thus,lends some weight to the idea that as yet uni-dentified ATP-spilling reactions may comprisepart of the overall mechanism of respiratoryprotection.

BACTERIAL ATPase COMPLEXESThe Mg2+-dependent, proton-translocating

ATPase complexes of bacteria have been in-tensively investigated in the last few years.Thus, the pioneering work of Abrams and hiscolleagues on the ATPase complex of the nor-mally fermentative bacterium Streptococcusfaecalis has recently been extended to aerobesand facultative anaerobes, principally Micro-coccus lysodeikticus (luteus) and E. coli (forreviews, see references 7, 343, 371).The energized state is viewed by the chemi-

osmotic hypothesis as a transmembrane pro-tonmotive force. In the previous section, wehave described how this force can be producedthrough the action of proton-translocating res-


b

N

£3~

a83

FIG. 8. Diagrammatic representation of the var-ious types of membrane-bound, energy-transducingATPases. (a) Mitochondrial; (b) bacterial; a,3,-y,8and e are the F1 subunits, and I is the ATPaseinhibitor protein.

piratory chains; in this section, we will discussthe properties ofbacterial ATPase complexes tosee to what extent these are compatible withtheir postulated role as reversible, proton-translocating systems.

Membrane-Bound ATPase ComplexesMorphology and location. Repeating struc-

tures of similar shape and size to the ATPasecomplex of the mitochondrial inner membranehave been observed in electron micrographs ofnegatively stained plasma membrane prepara-tions from M. lysodeikticus (137, 283) and fromseveral other species of bacteria (2, 21, 202).These structures were identified as ATPasecomplexes by their ability to form microscopi-cally detectable conjugates with ferritin-labeledantibodies to purified ATPase. In addition, ex-posure of the membranes to low-ionic-strengthbuffers yielded smooth membranes which nolonger reacted with these antibodies and whichwere essentially devoid of ATPase activity(309). The observations that protoplasts fromM. lyosdeikticus failed to bind ferritin-labeledantibodies, whereas membrane vesicles concen-trated them on one surface only, strongly sup-ported the concept that in whole bacteria, as inmitochondria, the ATPase is located on the in-ner surface of the coupling membrane.

It is now generally accepted that intact bacte-ria, or spheroplasts and protoplasts preparedfrom them, do not readily hydrolyze ATP orutilize ADP as a phosphoryl acceptor for oxida-tive phosphorylation. However, when sphero-plasts of E. coli were lysed by treatment withlow concentrations of detergents or made morepermeable by exposure to toluene, ATPase ac-tivity was substantially increased and was al-most totally sensitive to inhibition by ATPase-specific antibodies (126, 402). These resultsclearly suggested that the ATPase is located on


http://mm

br.asm.org/

Dow

nloaded from



the inner surface of the coupling membranewhich, in contrast to the mitochondrial innermembrane, appears to lack an adenine nucleo-tide translocase.

In mitochondria, the latter effects the in-ward, electrogenic transport of one molecule ofADP concomitant with the ejection of one mole-cule of ATP (222), thus balancing the require-ment of the ATP synthetase for a phosphorylacceptor against that of the cytoplasm for en-ergy. The absence of this transport system fromthe bacterial plasma membrane is clearly inaccordance with the relatively self-containedexistence of these predominantly free-living or-ganisms, which are able to exchange nutrientsand waste products, but not essential cofactors,with the surrounding medium.

If the orientation of the bacterial ATPasecomplex across the coupling membrane is in-deed asymmetric, then it should be possible toisolate two pure species of topologically closedvesicles analogous to submitochondrial parti-cles resulting from digitonin and sonic treat-ment (324), viz., right-side-out vesicles (inwhich the orientation of the membrane is thesame as that present in the intact cell, suchthat the ATPase remains on the inner surfaceand is thus inaccessible to exogenous adeninenucleotides) and inside-out vesicles (in whichthe orientation of the membrane has becomeinverted, such that the ATPase is fully ex-posed). Both types of vesicles have now beendetected in preparations of bacterial mem-branes, and there is also some evidence for thepresence of scrambled vesicles, i.e., those inwhich the membrane of a single vesicle exhibitsareas of right-side-out and inside-out orienta-tion (160, 371).In 1971, Kaback reported a method for the

isolation of predominantly right-side-out vesi-cles from E. coli, in which spheroplasts pre-pared by lysosyme-ethylenediaminetetraaceticacid treatment were subsequently subjected toosmotic shock in the presence of potassiumphosphate buffer (208). The resultant relativelylarge vesicles were osmotically sensitive andwere thus judged to be essentially intact (209).Freeze-cleave electron microscopy indicatedthat the vesicles were almost completely in theright-side-out orientation (17, 209), althoughtheir relatively high sensitivity to agglutina-tion by antibodies to purified ATPase has re-cently thrown some doubt on this conclusion(159). However, the predominantly right-side-out orientation of these vesicles was supportedby their ability to exhibit respiration-linkedproton ejection (328) and to drive the uptake ofvarious nutrients at the expense of electron

transfer to oxygen and other acceptors (40, 209);because of this latter property, they are oftenreferred to as active transport vesicles. Theobservations that these vesicles usually hydro-lyzed ATP rather slowly (221, 402, 410; but seereference 126), failed to utilize ATP as a sourceof energy for nutrient uptake (209), and did notcatalyze oxidative phosphorylation (221) sup-ported the concept that the ATPase was locatedon the inner surface of the vesicle membrane.Further support was added to this conclusionby the ability of Triton X-100 and other deter-gents to stimulate ATPase activity by destroy-ing the membrane permeability barrier to-wards adenine nucleotides, thus allowingATP to reach the inner surface (402, 410). Inaddition ATP-dependent amino acid uptakecould be induced by osmotically shocking ATPinto the vesicles (402). The results of similarexperiments with osmotically (prepared vesi-cles from Bacillus subtilis (156, 230, 231) andprotoplast ghosts from Mycobacterium phlei(21, 22) have suggested that these structuresare also predominantly, but by no means com-pletely, right-side-out and that the ATPase islocated on the inner surface. A similar locationfor the ATPase in whole cells ofM. lysodeikticushas also been deduced from studies on heteroge-neous populations of right-side-out, inside-out,and scrambled vesicles from this organism (137,309).

Disruption of whole cells, spheroplasts, orright-side-out vesicles ofE. coli by sonication orby exposure to shear forces (e.g., in a Frenchpressure cell or Ribi cell fractionator) yields apopulation of relatively small closed vesicles(17, 181, 259, 410); on the basis of evidence fromfreeze-cleave electron microscopy, these vesi-cles are predominantly in the inside-out config-uration (17, 105). Biochemical examination indi-cated that they did not catalyze energy-depend-ent amino acid uptake at a significant rate (181,259), although they readily catalyzed respira-tion-driven proton uptake (171), oxidative phos-phorylation (259), ATP hydrolysis (181), andATP-dependent transhydrogenation (157, 181);the latter two reactions were almost completelyinhibited by antibodies to ATPase (157, 181). Incontrast to right-side-out vesicles. ATPase ac-tivity was not stimulated by exposure to deter-gents (181). These results, together with thoseof less comprehensive experiments on similarlyprepared vesicles from other bacteria, convinc-ingly show that the ATPase is located on theouter surface of inside-out vesicles and thusconfirm its location on the inner surface of thecoupling membrane in whole cells.

Proton-translocating properties. There is

BACTERIOL. REV.


http://mm

br.asm.org/

Dow

nloaded from



considerable experimental evidence to suggestthat, in mitochondria and heterotrophic bacte-ria, two protons are extruded per pair of reduc-ing equivalents transferred through each oxi-dation-reduction segment of the respiratorychain, although this stoicheiometry can be de-creased by the use of certain artificial electrondonors and acceptors, and it may also be low insome chemolithotrophs. Thus, in order to com-ply with the P/O ratios that have been observedwith intact mitochondria, two protons must beretranslocated inwards for every molecule ofATP synthesized. Since the ATPase is reversi-ble, the same stoicheiometry (but of oppositedirection) should be observed during ATP hy-drolysis. In accordance with this prediction,--H+/P ratios (=-)H+/ATP) of approximately 2g ion of H+ per mol of ATP hydrolyzed havebeen determined for mitochondria (276, 281,391) and chloroplasts (63). Transmembrane pro-ton translocation concomitant with ATP hy-drolysis, albeit ofunknown stoicheiometry, hasalso been detected in chromatophores from pho-tosynthetic bacteria (356) and in reconstitutedmembrane vesicles from the thermophilic bac-terium PS3 (380, 425).The quantitative measurement of --H+/P ra-

tios associated with ATP hydrolysis is a techni-cally difficult operation, since it is necessary todistinguish between that part of the total pHchange that simply reflects ionization changesresulting from the chemical hydrolysis of ATP(up to 0.8 H+ per molecule ofATP [15]) and thatpart which results from transmembrane protonmovement and thus reflects the anisotropicproperties of the ATPase. With bacterial sys-tems, this problem is compounded by the ab-sence of an ATP-ADP translocase (126, 159,402), which makes the use of inside-out vesiclesobligatory.West and Mitchell (410) recently reported

that ATP hydrolysis by inside-out vesicles ofE.coli was accompanied by acidification of theexternal medium. However, when correctionwas made for the ionization effects of chemicalhydrolysis, it was clear that the reaction wasalso accompanied by inwardly directed protontranslocation (equivalent in intact cells to in-ward translocation during ATP synthesis). Bycomparing the initial rates of ATP hydrolysisand proton uptake, an -)H+/P ratio of approxi-mately 0.6 g ion ofH+ per mol ATP hydrolyzedwas obtained. Since this value was not cor-rected for the undoubted presence of imper-fectly sealed vesicles (which would allow trans-located protons to escape), the true --H+/P ra-tio is likely to be considerably higher. Indeed,an -*H+/P ratio that is approximately equal to

the --H+/2e ratio for one oxidation-reductionsegment (i.e. 2 g ion of H+ per mol of ATPhydrolyzed) can be predicted from comparisonsof molar growth yields of E. coli under aerobicand anaerobic conditions.The inherent proton conductance of energy-

coupling membranes is generally low (189, 270,354), although it can be increased substantiallyby the addition of uncoupling agents or by theinitiation ofknown proton-translocating events(e.g., the hydrolysis or synthesis of ATP andthe transport of certain nutrients). It has beenproposed that the transmembrane movement ofprotons concomitant with ATPase or ATP syn-thetase activity occurs via a specific channel inthe Fo complex of the coupling membrane (273).Removal of F1, either by physically stripping itfrom the membrane or by genetic deletion,would therefore be expected to increase protonconductance by exposing the entrance to theproton-conducting channel; this would lead to aconcomitant decrease in the efficiency of respi-ration-linked energy-dependent reactions. Thisprediction has been confirmed experimentallyusing F,-depleted membrane vesicles frombeef heart mitochondria (175, 271, 312); whenthe vesicles were subsequently treated witholigomycin of N,N' dicyclohexylcarbodiimide.(DCCD) or reconstituted by the addition of F1,their proton-conducting and energy-transduc-ing properties were substantially repaired.Thus, these results support the idea that F1 hasa structural, as well as a catalytic, role in themembrane-bound ATPase; in the absence of F1,this structural role can apparently be mimickedby oligomycin and DCCD which, by binding tothe oligomycin-sensitivity-conferring protein(OSCP) and the DCCD-sensitivity-conferringprotein (DSCP), respectively, presumably blockthe proton-translocating channel, which is pos-tulated to be present in FO, and thus preventdissipation of the protonmotive force.

Similar conclusions have also been drawnfrom the results of analogous experiments us-ing bacterial membranes which, as a result ofgenetic manipulation, have been depleted of F1or contain structurally defective of F1 or F0components (see below). In this respect, it isinteresting to note that the addition of purifiedF1 from S. faecalis to artificial phospholipidbilayers causes a large increase in electricalconductance, which is claimed to be compatiblewith the formation of discrete, but relativelywide, aqueous-filled channels (327).

Purification and Properties of the F1-ATPaseIsolation and general properties. Bacterial

ATPases, like those ofmitochondria and chloro-

VOL. 41, 1977


http://mm

br.asm.org/

Dow

nloaded from



plasts, can be released from the coupling mem-brane with relative ease (the various methodsare reviewed in detail in references 7 and 343).The most commonly used release procedure hasbeen a shock-wash process in which L-forms ofwhole bacteria, spheroplasts, protoplasts, or

isolated membrane preparations (3, 51, 94, 185,226, 267, 269, 278, 284, 333, 409) were washedseveral times in Mg2+-free, low-ionic-strengthbuffers [e.g., 2 mM tris(hydroxymethyl)amino-methane (Tris)-hydrochloride, pH 7.21 (226),such that the ATPase was dissociated from thecoupling membrane by the lack of cations.Alternative, and considerably more drastic,methods of release have involved extraction ofthe plasma membrane with organic solventssuch as n-butanol (which caused the enzyme toaccumulate in the aqueous phase) (344, 345) or

dissociation of the membrane with detergentssuch as Triton X-100 or sodium dodecyl sulfate(109, 157). The solubilized enzyme was thenpurified by relatively standard procedures in-volving combinations of ammonium sulfatefractionation, ion exchange or exclusion chro-matography, density gradient centrifugation,and polyacrylamide gel electrophoresis (51, 54,226, 227, 269). The resultant purified ATPasewas the headpiece (F1) component of the origi-nal, membrane-bound ATPase complex, themembrane (FO) components having been lostduring the purification procedure.Homogeneous preparations of purified F,

have now been obtained from S. faecalis (5a,

352, 353), B. megaterium (267, 268, 269), M.lysodeikticus (283, 285, 344), Bacillus stearo-thermophilus (151), Alcaligenes faecalis (14),Salmonella typhimurium (54), the thermo-philic bacterium PS3 (425), M. phlei (172), andfrom several strains ofE. coli (54, 127, 134, 157,294). They are all relatively large, multi-sub-unit proteins with molecular weights in therange 250,000 to 400,000 (Table 6) and are thussimilar in size to the solubilized ATPases ofmammalian and yeast mitochondria (molecu-lar weight, 340,000 to 360,000 [69, 237, 362, 397])and green plant chloroplasts (molecularweight, 325,000 [1111).

Analysis of F, from S. faecalis (353), B. meg-aterium (269), B. stearothermophilus (151), andM. lysodeikticus (343) showed that they con-tained very similar amino acid contents; allwere fairly rich in acidic amino acids, and theircontent of hydrophobic residues was strikinglyconstant: approximately 32%. Kobayashi andAnraku (227) reported that E. coli F, contained8 mol ofphosphorus per mol ofprotein, ofwhichonly a small proportion could be attributed tophospholipids; the remainder probably re-

flected the presence of other tightly bound,phosphorous-containing compounds such as ad-enine nucleotides. ATP, ADP, and inorganicphosphate (in the molar ratios 1.0:1.0:0.1 per

mol of enzyme) also appeared to be associatedwith the F, from S. faecalis (6). In this respect,these two bacterial enzymes resemble theircounterpart from beef heart mitochondria,

TABLE 6. Subunit size and composition ofpurified ATPases (Fda

Subunit mol wt (x 10-3) Subunit cor-Source Mol wt sition References

a 8, position

Rat liver mitochondria 340-380 62.5 57 36 12 7.5 af300y8e 68, 6953 50 28 12.5 7.5 af3,y 237

Beef heart mitochondria 53 50 25 12.5 7.5 a30ybe 361, 362ajy,8ye2 359

Yeast mitochondria 340 58 54 38 31 12 397Spinach chloroplasts (CF,) 325 59 56 37 17.5 13 aC,82ye2 39, 111, 245, 246, 293Escherichia coli 56.8 51.8 32 20.7 13 a3,83y 54

400 54 52 33 11 226,227360 60 56 35 13 157340 56 52 32 21 11.5 a2,2y28,12e2 408

Salmonella typhimurium 56.8 51.8 30.9 21.5 13.2 a,/3,y8e 54Alcaligenes faecalis 350 59 54 43 12 12, 13, 14, (see also

[313])Thermophilic bacterium PS3 380 56 53 32 11 15.5c 380, 425

(TFI)Micrococcus lysodeikticus 62 60 -& 344

345 52.5 47 41.5 28.5 af3,3yS 19, 20Streptococcus faecalis 385 60 55 37 20 12 a3P3sY 5a, 352, 353Bacillus megaterium 399 68 65 a3P3 268, 269

a The purified ATPases from B. stearothermophilus (molecular weights, 280,000 [151]) and M. phlei (molecular weight,250,000 [172]) are not included in this table, since little is known of their subunit composition.

b Several additional, minor bands are present in shock-wash preparations which are not present after extraction with n-butanol (123, 344).

e The authors refer to this subunit as 8' rather than e (425).

BACTERIOL. REV.


http://mm

br.asm.org/

Dow

nloaded from


VOL. 41, 1977

which contains 3 mol of tightly bound ATP and2 of ADP per mol of enzyme (163); it is possiblethat these bound nucleotides play an importantrole during ATP synthesis (45). Some contro-versy exists over the phospholipid content ofpurified F1; phospholipids were detected in theenzyme from E. coli (315) but not in that fromM. lysodeikticus (285). In the former case, re-moval of phospholipid by centrifugation or ionexchange chromatography caused a decrease inATPase activity, which could be partly restoredby the addition of a phospholipid extract fromthe whole organism.The ATPase activities of bacterial F1 prepa-

rations were extremely variable, although theywere generally higher than those of the mem-brane-bound ATPase complexes from whichthey were derived. All of the purified ATPaseswere found to be dependent upon divalentmetal ions for maximum activity; this require-ment was usually satisfied by either Mg2+ orCa2+, although, in the case of S. faecalis F1,only Mg2+ was effective (3). Na+ or K+ appearedto exert little stimulatory action upon the solu-bilized ATPases (94, 226), although they occa-sionally stimulated the membrane-bound en-zymes to a variable extent (3, 152, 409).The ability of bacterial F1 to hydrolyze nu-

cleotides is not limited to ATP. Indeed, all ofthe purified F1 preparations so far examinedshowed substantial activity with guanosine 5'-triphosphate (GTP) and also, to a lesser extent,with uridine 5'-triphosphate (UTP), and cyti-dine 5'-triphosphate (CTP). B. megaterium F1was unusual in that it exhibited as high anactivity with inosine 5'-triphosphate (ITP) aswith ATP (267); the E. coli enzyme hydrolyzedboth ITP and ADP at a significant rate (94,157), and the S. faecalis F1 hydrolyzed ITP anddeoxyATP (7). None of the other purified bacte-rial enzymes showed significant activity withITP, ADP, AMP, or Pi (see reference 343). All ofthe purified ATPases investigated exhibitedMichaelis-Menten kinetics with respect to ATPconcentration. There is some evidence that bothbacterial and mitochondrial F1 preparationshave significantly lower Km values for ATPthan do the corresponding membrane-boundcomplexes (155, 157, 277, 333, 352, 353), whichsuggests that the two forms exist in differentconformations; this conclusion is supported byevidence from electron microscopy. Those F1preparations that have been investigated forevidence of product inhibition all showed a lossof activity with increasing concentrations ofADP and Pi. The enzymes from S. faecalis (353),E. coli (157, 226, 333), and M. phlei (172) wereall inhibited competitively by ADP and were


inhibited either competitively or noncompeti-tively by Pi.In striking contrast to the membrane-bound,

bacterial ATPase complexes, most of the F1preparations examined so far have been foundto be cold labile (i.e., activity was rapidly lostafter storage at 0 to 40C (157, 227, 267); thisproperty is shared with motochondrial F1 (314),but not with the purified ATPases from PS3and M. phlei (150, 153). The precise molecularmechanism of cold-lability is not known, al-though there is good evidence that the ATPasesfrom E. coli (157, 227, 408) and beef heart mi-tochondria (314) undergo dissociation intosmaller fragments with, in the latter case, con-comitant release of the bound adenine nucleo-tides. The cold-lability of the E. coli F1 can beprevented by storage in the presence of highconcentrations of glycerol or methanol, whichpresumably help to maintain the quaternarystructure of the enzyme (226, 227, 408); thisprotective action is nullified by high salt con-centrations (e.g., 0.5 M Tris-hydrochloride or 2mM Tris-hydrochloride plus 0.5 M NaCl) (227).

Further differences between purified andmembrane-bound ATPases are exemplified bytheir differential sensitivities to inhibitors anduncoupling agents. Thus, the ATPase activityof bacterial F1 preparations, in contrast to thatof the membrane-bound enzymes, is neitherstimulated by uncouplers nor inhibited byDCCD (134, 162, 226, 333), although it remainssensitive to azide, since the latter binds to theheadpiece ofthe enzyme complex rather than toa component of the membrane-located FO; nei-ther the purified or membrane-bound bacterialATPases are inhibited by oligomycin. This phe-nomenon, in which a purified soluble enzymeexhibits a whole range of different propertiescompared with its membrane-bound counter-part, is called allotopy (324) and is apparentlycommon to all energy-transducing ATPases.Subunit composition. All the purified bacte-

rial ATPases that have so far been examinedcan be dissociated into at least two nonidenticalsubunits after exposure to polyacrylamide gelelectrophoresis in the presence of low concen-trations ofsodium dodecyl sulfate (see example,reference 227). Subunit stoicheiometry is subse-quently determined either by colorimetric anal-ysis of the stained gels or by radioactive analy-sis of the gels after electrophoresis of the 14C_labeled enzyme.

(i) E. coli and S. typhimurium. It has re-cently been reported that the purified ATPasesfrom E. coli (54, 127, 408) and S. typhimurium(54) contain five different subunits (termed a,,3, y, 8, and E, in order of decreasing molecular


http://mm

br.asm.org/

Dow

nloaded from



weight; see Table 6); other reports have con-cluded that the enzyme lacks a 8 subunit (157,227). These differences have been resolved byFutai et al. (127), who showed that, by slightlymodifying the isolation-purification procedurethat normally yielded an ATPase containingfour different subunits, it was possible to pro-duce an enzyme that also contained the 8 sub-unit. However, it has been claimed that differ-ent strains ofE. coli consistently yield purifiedATPases with either four or five subunits, re-gardless of the extraction and purification pro-cedures employed (371). The presence or ab-sence of the 8 subunit in the purified enzyme,therefore, probably reflects the strength withwhich the polypeptide is bound to the remain-der ofthe ATPase molecule. Since the 8 subunitappears to be necessary for energy coupling (seebelow), there seems to be little question thatthe F1 from these organisms is comprised of fivedifferent subunits in vivo (Fig. 8).

Until very recently, it appeared that the F1from E. coli, S. typhimurium, and mammalianmitochondria exhibited similar subunit stoi-cheiometries, viz., a3/33y8E (54, 68, 69, 237, 361,362). However, after a quantitative analysis ofthe sulfhydryl groups and disulfide bonds foundin beef heart F1 and its constituent subunits,Senior has recently suggested the stoicheiome-try a2gy2 plus unknown proportions of theand 8 components (359). Furthermore, the stoi-cheiometry a2/32Y281-22 has been proposed forE. coli F1 (408). It is possible, therefore, thatthese enzymes are structurally even moreclosely related to chloroplast CF1 (which is ag-gregated in the subunit ratio a2/2y8E2 [293])than was originally thought likely, but thisobviously requires more extensive investiga-tion. The purified ATPase from the thermo-philic bacterium PS3 also has five subunits (a,,/, y, 8, and 8'), albeit of unknown stoicheiome-try; the fifth subunit is referred to as the 8'component, since it is rather larger than the E

subunit of other ATPases. Because of its ther-mal stability (temperature maximum, 700C),the ATPase of PS3 is usually referred to as TF1(380, 425).

It is clear that both purified and membrane-bound ATPase from various sources exhibitsome degree of latency, i.e., their ATP-hydro-lyzing activities can be stimulated by a varietyof chemical or physical treatments. Thus, theATPase activity of coupling membranes fromeukaryotes appears to be at least partly con-trolled by a trypsin-sensitive component thatmay or may not be part of the purified ATPase,depending upon the isolation procedure em-ployed. In chloroplasts, this role appears to be

BACTERIOL. REV.

played by theE subunit of CF1 (295), whereas inmitochondria an additional, loosely bound com-ponent (the ATPase inhibitor protein) is proba-bly responsible (180, 323, 325, 361; but see alsoreference 224). Evidence is now accumulatingthat E. coli F, also contains a trypsin-sensitiveATPase inhibitor. Thus, Bragg and Hou (54)observed that maximum activation ofE. coli F1occurred when trypsin had digested all of the 8and E subunits, together with over 70%o of the ysubunit; the a and /8 components were littleaffected. It was concluded from the results ofthis and similar experiments (294) that one ormore of the smaller subunits (y, 8, or E) par-tially masks the ATPase activity of the en-zyme, whereas the larger a and /8 subunitshave a catalytic role. Further support for thepresence of an ATPase inhibitor in E. coli F1has been provided by Nieuwenhuis et al. (300),who reported that the activity of a solubilizedATPase from this organism was irreversiblyinhibited by high concentrations ofurea (proba-bly as a result of the complete dissociation ofthe enzyme into its constituent subunits) andthat the inactivated enzyme strongly inhibitedthe ATPase activity ofboth untreated and tryp-sin-activated Fl. These workers also isolated acrude protein fraction (molecular weight,12,000) from E. coli F1 which potently inhibitedthe ATPase activity of the intact enzyme. Al-though it has not been unequivocably identifiedas such, circumstantial evidence suggests thatthe active component of this fraction is the Esubunit. Thus, the ATPase inhibitor proteinfrom E. coli appears to more closely resembleits counterpart in chloroplasts rather than inmitochondria. This view is reinforced by theobservation that the ATPase inhibitor can beremoved from submitochondrial particles bypassage over a Sephadex G-50 column (325),whereas the inhibitor proteins oftheE. coli andchloroplast ATPases are resistant to this treat-ment and are therefore presumably bound rela-tively strongly to the remainder of the enzyme.The postulate that the a and /3 subunits ofE.

coli F1 have a catalytic function is supported bythe work of Gutnick and his colleagues (213,294), who observed that antibodies preparedagainst the combined a and /3 components ofE. coli F1 completely inhibited ATP-linkedtranshydrogenation (without having any effecton the respiration-linked reaction), ATP hy-drolysis, and ATP-P1 exchange in E. coli mem-branes. Antibodies against the a subunit alonewere also strongly inhibitory, but those pre-pared against the /8 or y subunits were muchless effective. It is likely, therefore, that the aand /3 subunits ofE. coli F1, like those of mito-


http://mm

br.asm.org/

Dow

nloaded from



chondrial F1 (229) and chloroplast CF1 (98), playa central catalytic role in ATP-dependent en-ergy-coupling reactions, as well as in oxidativephosphorylation.

It is clear that the 8 subunit of the E. coliATPase is not required for ATP hydrolysis,since its removal from the enzyme (after treat-ment of the latter with trypsin) was accompa-nied by an increase rather than a decrease inhydrolytic activity (54). On the other hand,there is reasonably good evidence that the 8subunit facilitates the binding of F, to the cou-pling membrane and is therefore required foroxidative phosphorylation and ATP-dependentenergy transduction; this is based upon the ob-servations that only the five-subunit F1 wasable to reconstitute ATP-dependent transhy-drogenation when added to F,-deficient mem-branes (48, 127, 294) and that the inactive four-subunit preparation could be reactivated by theaddition of a combined 8, e subunit fraction(378). It appears, therefore, that the 8 subunitof the E. coli ATPase is functionally analogousto the OSCP of mitochondria (in the sense thatit acts as the stalk component of the entirecomplex), although it does not, of course, confersensitivity to oligomycin. In contrast, the 8 sub-unit of chloroplast CF, does not appear to havea binding function (295) and the role of thissubunit in eukaryotic ATPases is unclear.

Little is known about the function of the ysubunit, although it has been suggested that inmitochondrial F1 it serves to bind the 8 and esubunits together to form a composite, "super-subunit" ofa size comparable with each ofthe aand (3 subunits (229). There is some evidencefrom the use of an E. coli mutant which con-tains an ATPase with a defective y subunit,that the presence of the latter is obligatory foreffective energy transduction in this organism,although it is apparently not essential for ATP-ase activity (48) (see below).

(ii) A. faecalis. The purified ATPase fromA.faecalis appears to contain only three subunits(a, 3, fy) plus a fourth component which, solelyon the basis of its very low molecular weight,must be classed as the e rather than 8 subunit(14); the relative proportions of the constituentsubunits are not known. It is possible, by anal-ogy with E. coli, that in vivo the enzyme alsocontains a weakly bound 8 subunit which is lostduring the purification procedure, but there iscurrently no experimental evidence to supportthis. The hydrolytic activity of the A. faecalisenzyme is strongly enhanced by trypsin, thussupporting the concept that one of the subunitsacts as an ATPase inhibitor.

(iii) M. lyIsodeikticus. The properties and

subunit composition of the F1 from M. lysodeik-ticus appear to depend very much upon theprocedures employed for its extraction and pu-rification. Thus, the enzyme exhibited the stoi-cheiometry a3,83 when isolated using n-butanol(344), but the gentler shock-wash procedureyielded a preparation that contained two addi-tional, smaller polypeptides (19, 20, 123, 344),viz., a loosely bound component (y) and anintegral subunit (8), to give the overall stoi-cheiometry a3833y8; no e subunit was detectedin either of these preparations. The observa-tions that both membrane-bound and crude,soluble ATPases from M. lyosdeikticus could beactivated by trypsin, whereas both prepara-tions ofpure F1 were insensitive to this enzyme(19, 20, 239, 285), suggested that the trypsin-sensitive component was lost during purifica-tion; it is likely that this is the e subunit.Similarly, since only the ATPase purified fromshock-wash extracts was capable of binding torespiratory membranes, it is possible thateither the y or, more probably, the 8 subunitattaches the F1 to the membrane in vivo.

(iv) S. faecalis. Until very recently, the F1from S. faecalis was thought to consist only of aand (3 subunits, plus an additional polypeptidecalled nectin (Latin, nectere = to bind) (4, 33,353). However, it has now been reported thattwo minor subunits (8 and e) are also presentsuch that the enzyme has the overall composi-tion a39y8E (5a) and is thus similar to the F1from most other bacterial and mitochondrialsources. Interestingly, when the enzyme is iso-lated and purified in the absence ofmagnesiumions, it appears to lack the 8 subunit and is nolonger capable of binding to F1-depleted mem-branes. It has been concluded therefore thatMg2+ is probably an integral component of theenzyme, since it is apparently required for theattachment of the 8 subunit to the remainingpolypeptides. Furthermore, it is likely that the8 subunit is involved in binding the enzyme tothe receptor site on the coupling membrane(either alone or in association with Mg2l); thissubunit thus has a similar role in the F1 fromboth S. faecalis and E. coli. The earlier reportthat nectin is involved in the binding of F1 tothe membrane (33) has been rationalized by theclaim that it may be a functionally active dimerof the 8 subunit (5a).

(v) B. megaterium. The ATPase that Mirskyand Barlow (269) purified from B. megateriumafter shock-wash treatment of cytoplasmicmembranes contained only a and (3 subunits(in the ratio a363) and was apparently devoid ofadditional polypeptide components. However,in view of the complex subunit-composition of

VOL. 41, 1977


http://mm

br.asm.org/

Dow

nloaded from



all the' other bacterial ATPases so far investi-gated, it is likely that minor polypeptides arepresent inB. megaterium F, in vivo but are lostduring the purification of the enzyme.Functional organization of subunits. The

chemical properties and intermolecular ar-rangement of the constituent subunits of bacte-rial F, are currently receiving increasing exper-imental attention, since only by resolving theseproblems can the mode of 'action of these com-plex enzymes be properly understood.

Central to these studies is the use of chemi-cal-modifying agents that inhibit ATPase activ-ity by forming covalent linkages with specificamino acid residues and thus yield informationon the possible involvement of these residues inthe active site of the enzyme or in the bindingadjacent subunits. 7-Chloro-4-nitrobenzo-2-oxa-1,3-diazole (NBD-Cl) has been used extensivelyfor this purpose by Radda and his colleagues(113-115). This reagent reacts with the phenolicoxygen of tyrosine and with the sulfhydrylgroup of cysteine; the two reactions can be dis-tinguished from the different absorption spec-tra of the resultant adducts (1, 115) or by com-paring the observed inhibition with that causedunder similar conditions by N-ethyl maleimideor iodacetamide, both of which react only withcysteine (113).NBD-Cl readily inhibited the ATPase activ-

ity of F. from beefheart mitochondria (115) andE. coli (294). In both cases, the inhibition wasovercome by treating the covalent enzyme-in-hibitor complex with the sulfhydryl reagent di-thiothreitol to remove the inhibitory group.There is good evidence that NBD-Cl reacts witha tryosine residue of mitochondrial F1 (114,115), and this is probably also true for the E.coli enzyme; in the latter case, the site of inhi-bition was identified as the /8 subunit throughthe use of 3H-labeled NBD-Cl (294). This re-agent also inhibited ATP hydrolysis and ADP-dependent respiratory control by respiratorymembranes from P. denitrifleans (113); modifi-cation of a tyrosine residue again appeared tobe responsible. Since these effects were re-versed by the relatively nonpenetrating sulfhy-dryl reagent glutathione (as well as by dithio-threitol), it was concluded that the reactivetyrosine residue(s) was not buried within theATPase. These experiments thus strongly im-plicate tyrosine in the active site of the P.dentrificans ATPase.The problem ofdetermining the arrangement

of the subunits in bacterial ATPase has beentackled principally via electron microscopyand, more latterly, through the use of proteincross-linking reagents. The former has been

applied with some success to purified F1 fromM. lysodeikticus (283), B. megaterium (185),and S. faecalis (353), all of which appear asplanar hexagonal arrangements of six subunitsthat surround any other polypeptides* thatmight be present. In contrast, Bragg and Houhave recently used protein cross-linking re-agents, particularly dithiobis (succinimidylpropionate; DSP), to examine the architectureof the purified, five-subunit ATPase from E.coli (54). This reagent cross-links the e-aminogroups of lysine residues and thus can be usedto investigate the proximity and location ofsuch groups within the same and adjacent sub-units; very usefully, the cross-linkage can sub-sequently be cleaved by reduction with mercap-toethanol or dithiothreitol. Exposure ofE. coliF1 to DSP resulted in the rapid loss of ATP-hydrolyzing activity, the partial disappearanceof the a and (3 subunits, and the concomitantformation of a new, high-molecular-weightcomponent (Y) plus traces of other, even largercomponents; the -y, 8 and e subunits disap-peared completely (54). Treatment of Y withdithiothreitol regenerated a and /3 subunits inequal proportions, whereas treatment of thelarger adducts yielded y subunits in addition tothe a and /8 components. These results sug-gested that the latter were arranged within theenzyme in such a manner that they could easilybe cross-linked without the concomitant forma-tion of aa or /,3/ dimers, and they thereforesupport the concept of alternating subunits ar-ranged in the form of a hexagon. The ability ofthe y subunit to cross-link with a and ,3 sub-units supports its location in the central hole ofthe hexagon. Furthermore, the fact that an E.coli ATPase can be isolated which is deficient inthe 8 subunit (48, 157, 294, 336) indicates thatthe latter is not necessary for attaching the yand e components to the remainder of the mole-cule, although it is clearly required to attach F1to the coupling membrane.Two models have recently been proposed by

Kozlov and Mikelsaar (229) for the structure ofa3/&3y8c ATPases. The first envisages that al-ternating a and /3 subunits form the outer hex-agonal structure and that the 'y, 8, and E sub-units comprise the center of the hexagon; thesecond suggests that a planar trimer of /8 sub-units is packed above a similar arrangement ofa subunits, and that the remaining componentscomprise a stalk that presumably assists in theattachment ofthe enzyme to the coupling mem-brane. A slightly modified version of the firstmodel is fully compatible with the chemical andmorphological properties ofE. coli F1 with thesubunit ratio a3,33'y8E (54) (Fig. 9a). In contrast,

BACTERIOL. REV.


http://mm

br.asm.org/

Dow

nloaded from



Wi) (ii)

b

p p

aYE G:Y

42

FIG. 9. Possible arrangement ofsubunits in the F,of E. coli. (a) Assuming the structure a43,3y8e, (i)transverse section through the ad33 hexagon (ii) sideview (after Bragg and Hou [54D); (b) assuming thestructure aYA02281-2,2 (after Vogel and Steinhart[408D).

neither model can explain the behavior of E.coli F1 as reported by Vogel and Steinhart (408).This enzyme was dissociated by freezing andthawing into two major subunits of dissimilarsize, namely, IA (molecular weight, 100,000;subunits aye) and II (molecular weight, 54,000;subunit 13), but could be reconstituted to showATPase activity by mixing IA and II in equimo-lar proportions. These results confirmed thetheory that both a and (8 subunits were re-quired for catalytic activity and were consid-ered to be commensurate with the stoicheiome-try a232VY281-2E2, the subunits being arranged asshown in Fig. 9b.

Purification and Properties of the F.-F,Complex

Although the membrane-bound ATPasesfrom mitochondria and bacteria are readily in-hibited by DCCD and other carbodiimides, pur-ified F1 preparations are insensitive to this typeof inhibitor (36, 162, 333). This observation indi-cates that the DCCD-sensitive component is

not one of the F1 subunits, but is presumably amembrane component of the ATPase complexthat is eliminated during the solubilization-purification procedure. In confirmation of this,Beechey and his colleagues have shown thatDCCD binds covalently to a proteolipid compo-nent, DSCP (molecular weight, 10,000), of themitochondrial Fo complex (36, 66, 67, 330). Asecond major component of mitochondrial Fo,OSCP, is apparently absent from bacterialATPases and its structural role, at least in themore complex bacterial enzymes, is probablytaken over by the 8 subunit of F1 (see above).Resolution of the component proteins of FO, andof the structural relationship between Fo andF1, are fundamental prerequisites to a properunderstanding of the mode of action of reversi-ble ATPases, particularly since F0 is probablyinvolved in the transfer of protons across thecoupling membrane and is thus intimately as-sociated with the mechanism of energy trans-duction.

Several attempts have recently been made toisolate a DCCD-sensitive ATPase complex (i.e.,an Fo-F, complex) from bacterial systems. Bysolubilizing respiratory membranes of E. coliwith low concentrations of cholate or deoxycho-late, Neiuwenhuis et al. (302) isolated a low-activity ATPase that showed some sensitivityto DCCD. This crude enzyme preparation,which still contained a substantial amount ofcytochrome, was activated severalfold by theaddition of soybean phospholipids, which alsorendered it almost completely sensitive toDCCD. Hare (158) reported generally similarresults using a partially purified, deoxycholate-solubilized ATPase complex from the same or-ganism and attributed the stimulatory and sen-sitizing properties of the added phospholipids(of which phosphatidyl choline, phosphatidylserine, and phosphatidyl ethanolamine werethe most potent) to their abilities to replacethose that were lost during the solubilizationprocedures. It is interesting to note that phos-pholipids are also required for maximal ATP-ase activity in mitochondrial and other bacte-rial F0-F, preparations (210, 380, 425) and possi-bly also in purified F1 (315).

Detailed investigations of the composition ofthe partially purified, DCCD-sensitive ATPasecomplex from E. coli indicated the presence of12 polypeptide components, 5 of which wereidentified as the a to e subunits of F1, since theycomigrated with the purified subunits on so-dium dodecyl sulfate-polyacrylamide gels andwere immunoprecipitated by F,-specific anti-body (158). The latter also precipitated twoother polypeptides, 4 (molecular weight, 29,000)

VOL. 41, 1977


http://mm

br.asm.org/

Dow

nloaded from



and n (molecular weight, 9,000), ofwhich n wasidentified as the DCCD-sensitivity-conferringprotein (DSCP) through its ability to form acovalent adduct with 14C-labeled DCCD (117,118, 158); other workers have claimed a molecu-lar weight of 12,000 to 13,000 for DSCP (18). Thenature and function of the five remaining com-ponents are unclear.A DCCD-sensitive ATPase complex has also

been purified from the thernophilic bacteriumPS3 after extraction of its membranes with Tri-ton X-100 (380, 425). Sodium dodecyl sulfate-polyacrylamide gel electrophoresis of this com-plex yielded only eight subunits, five of whichcorresponded to the subunits of TF,; the threeremaining TFo components were hydrophobicand exhibited molecular weights of 19,000,13,500, and 5,400. No attempt was made todetermine which of the TFo subunits was re-sponsible for binding DCCD. The addition ofphospholipids to the TFo-TF, complex stimu-lated its ATPase activity and also caused theformation of inside-out membrane vesicles. Thelatter exhibited ATP-dependent energization(as measured by the quenching of 1-anilino-naphthalene-8-sulfonate fluorescence) and alsocatalyzed the uptake of protons concomitantwith ATP hydrolysis; both of these reactionswere abolished by DCCD and uncouplingagents, and neither reaction was carried out bythe nonvesicular TFo-TF, complex. It was latershown (426) that vesicles reassembled fromphospholipids, TFo-TF, and purple membranes(bacteriorhodopsin) from Halobacterium halo-bium exhibited light-dependent ATP synthesisthat was sensitive to DCCD and uncouplingagents. These elegant experiments thus showthat a relatively simple, eight-subunit ATPasecomplex can catalyze a two-way translocationof protons across the coupling membrane thatis sufficient to drive ATP synthesis in onedirection and membrane energization at theexpense of ATP in the other.DCCD-resistant mutants of E. coli (RF-7)

and S. faecalis (sf-dcc-8) have recently beenisolated after selection for growth on plates con-taining growth medium supplemented withnormally lethal concentrations of DCCD (8,117). Both mutants were unimpaired with re-spect to energy transduction, as evidenced bytheir normal growth characteristics and bytheir capacity for catalyzing ATP-dependentmembrane reactions, but their membrane-bound ATPases were up to 100-fold less sensi-tive to inhibition by DCCD than were thosefrom the wild-type organisms.Abrams and his colleagues (5, 162) have re-

ported that the inhibition of the membrane-

bound ATPase of S. faecalis by DCCD andother carbodiimides probably occurs via the for-mation of a covalent compound between theinhibitor and an undissociated carboxyl groupon a so-called carbodiimide-sensitizing factor(CSF), which was tentatively identified as theF,-binding protein, nectin (33). However, inlater experiments, the location of CSF wasmore precisely identified by measuring the sen-sitivity to DCCD of hybrid ATPase complexesreassembled from F1, nectin, and depletedmembranes isolated from mutant and wild-typecells (8). It was observed that the ATPase com-plex that had been reassembled from mutantF., mutant nectin, and wild-type membraneswas sensitive to DCCD, whereas any complexthat contained mutant membranes was insensi-tive to this inhibitor. The results of these ele-gant experiments thus indicated that DCCDsensitivity was a property of the coupling mem-brane, rather than of nectin as previouslyclaimed, and that the mutation was probablyexpressed as a defective CSF.By similarly reassembling ATPase hybrids of

E. coli using components isolated from thewild-type and DCCD-resistant (RF-7) strains,Fillingame showed that the sensitivity of thisorganism to DCCD was also a property of itsmembrane fraction (117, 118). Furthermore,since membranes from this mutant and from E.coli mutant K-12DG7/1 (18) both contained aDSCP that was no longer capable of binding[14C]DCCD, it was concluded that DCCD resist-ance was caused by an alteration in the struc-ture of this proteolipid such that its functionalgroups were no longer accessible to the inhibi-tor. The mutation in RF-7 was mapped at ap-proximately 73.5 min on the chromosome andwas 90% cotransducible with the uncA gene(see below).There is, therefore, general agreement that

DCCD sensitivity requires the presence of acompetent, DCCD-binding proteolipid (DSCP,CSF) in the Fo complex of the membrane-boundATPase, although preliminary experimentswith some energy-transduction mutants of E.coli suggest that a second component may alsobe required (213). In the light of these results,the observed sensitivities ofthe solubilized (butnot purified) ATPases of Proteus P18 L-form(278) and E. coli (109) to DCCD implies thatthey also contain associated membrane pro-teins which presumably include DSCP.

Mutants Defective in Energy Transduction

The problem ofdetermining the physiologicalfunctions of the individual polypeptide compo-

BACTERIOL. REV.


http://mm

br.asm.org/

Dow

nloaded from


VOL. 41, 1977

nents of F1 and F0 has recently been approachedthrough the isolation and analysis of mutantstrains that are defective in energy transduc-tion. Because of its ready susceptibility to ge-netic manipulation, E. coli has been used al-most exclusively in these studies, and two gen-eral classes of mutants have been isolated fromthis organism which appear to be useful in thisrespect: unc- ATPase- and unc- ATPase+(317). Phenotypically, both classes grow aerobi-cally on glucose but not on nonfermentable car-bon sources such as succinate; aerobic molargrowth yields with respect to glucose consump-tion (Y,10,,e) are midway between those ex-hibited by the wild-type organism under aero-bic and anaerobic conditions. These growth pat-terns are compatible with genetic lesions re-sulting in defective ATPase complexes suchthat both classes ofmutants are uncoupled, i.e.,they are defective in oxidative phosphorylation,but their capacity for respiration is unim-paired.The properties of energy-coupling mem-

branes prepared from these mutants have beeninvestigated mainly by detailed analysis of thefollowing reactions: (i) ATP hydrolysis (ATPaseactivity); (ii) ATP synthesis via oxidative phos-phorylation; (iii) transhydrogenation or activetransport at the expense ofenergy derived fromATP hydrolysis or respiration; (iv) energiza-tion of the membranes by ATP hydrolysis orrespiration, as measured by the quenching offluorescent dyes such as 9-amino-6-chloro-2-methoxyacridine (ACMA) or atebrin. Addi-tional information has also been obtained bymeasuring the sensitivities of the above reac-tions to inhibition by DCCD and from attemptsto reconstitute hybrid ATPase complexes frommutant and parental components. However,not all ofthese assays have been applied to eachunc- mutant, and many of these strains are asyet incompletely characterized with respect totheir energy-transducing properties. As a re-sult, it is often extremely difficult to make auseful comparison of closely related mutants orto relate with any confidence their phenotypicproperties to a defect in an individual polypep-tide subunit of the ATPase complex.unc- ATPase- mutants. Several mutants of

E. coli that fall into the unc- ATPase- categoryhave been investigated in some detail: uncA(strain AN 249) and unc-405 (strain AN 285)(79, 81, 82), various MDA strains (93), unc-1 7(393, 422), Nlm (212, 301), DL-54 (16, 52, 359),NR-70 (335, 336), NR-76 (337) and uncA103c(350). Generally speaking, membrane vesiclesprepared from these mutants failed to catalyzeATP hydrolysis, oxidative phosphorylation, or


ATP-dependent reactions such as transhydro-genation or active transport. On the otherhand, respiration and respiration-linked activetransport or transhydrogenation were gener-ally unimpaired, although there were a num-ber of exceptions to this. Thus, respiration-linked active transport was defective in mem-brane vesicles from DL-54, N14, and NR-70 (16,335, 336), transhydrogenation was of only lowactivity in DL-54 (52), and membrane energiza-tion as measured by the quenching of ACMAfluorescence was defective in N,14 (301); how-ever, in all cases, the defective reactions couldbe largely restored by the addition of DCCD ornormal F1 to the membranes.

It is clear from these observations that unc-ATPase- mutants cannot use ATP as a sourceof energy for membrane-associated reactions,although they can still, with some exceptions,use the high energy state that is produced atthe expense of electron transfer. This is nicelyillustrated by the reports that whole cells ofunc-1 7, maintained under anaerobic conditionsin the presence of glucose, failed to exhibittheir expected motility or to catalyze eitheractive transport or DNA synthesis, unless alsosupplied with an electron acceptor such as oxy-gen or nitrate (258, 393, 422). Thus, the growthof unc- ATPase- strains on glucose (but notsuccinate) is possible because substrate-levelphosphorylation provides ATP, whereas elec-tron transfer to oxygen or nitrate furnishes theenergized state that is necessary for drivingessential energy-dependent membrane func-tions such as active transport. Unc- ATPase-strains will not grow anaerobically on glucosein the absence of an added electron acceptor.Much of the work that has been carried out

with these mutants has been directed towardsdetermining the nature of the ATPase- lesion:e.g., do these mutants lack the ability to hydro-lyze ATP because the ATPase is absent fromthe coupling membrane or because the enzyme,although bound to the membrane, is defectivein catalysis? In this respect, Gibson and hiscolleagues have obtained compelling evidencethat the uncA lesion is expressed as a defectivemembrane-bound F1, since agarose-gel chroma-tography of a low-ionic-strength wash fromuncA membranes yielded a protein peak thatwas characteristic of F1 yet showed no ATPaseactivity (79). Furthermore, both oxidative phos-phorylation and various ATP-dependent reac-tions could be reactivated in ATPase-depleteduncA membranes by the addition of F1 from thewild-type organism or from an unc- ATPase+mutant (79, 81). Since F1 from either of thesetwo sources failed to reactivate intact uncA


http://mm

br.asm.org/

Dow

nloaded from



membranes, it was concluded that the defectiveF1 was still attached to these membranes andoccupied the available binding sites. In con-trast, the low-ionic-strength wash from mem-branes of unc-405 contained no protein peakcorresponding to F., and both washed and in-tact membrane preparations could be reacti-vated by the addition of parental F1; it wasconcluded, therefore, that this mutant hadprobably synthesized an altered F1 that was nolonger capable of binding to the coupling mem-brane (79, 82). Finally, neither intact or washedmembranes from mutant N144 could be reacti-vated by the addition of parental F1, an obser-vation which suggested that this mutation hadproduced a catalytically defective F1 that wasmore firmly bound to the membrane than thewild-type enzyme (301). The MDA1, MDA2, andMDA3 mutants isolated by Kepes and his col-leagues exhibited ATPase-binding and reacti-vation properties which suggested very closesimilarities with the uncA, unc-405, and N144mutants, respectively (93).

It is likely that the genetic lesion in the uncAand MDA1 mutants has led to the synthesis of adefective catalytic subunit (a or (8 or both),while leaving the binding capacity of the F1unimpaired (which is expressed via the 8 sub-unit). On the other hand, the ability of F1 tobind to the coupling membranes is apparentlydecreased in the unc-405 and MDA2 mutants(and probably also in DL-54 and NR-70), whichsuggests that the affinity of the 8 subunit of F1for its binding site of the membrane may bediminished. The evidence suggests that theMDA3 and N1,. mutants are also characterizedby a point mutation in the 8 subunit, but onethat has pleiotropic effects, i.e., it enhances thebinding of F1 to the membrane but, in doing so,eliminates its catalytic properties.Although the mutations in DL-54, NR-70,

NR-76, and N1,. clearly decreased the efficiencywith which membranes prepared from thesestrains utilized the energized state produced byrespiration, this defect could be repaired by theaddition of DCCD (16, 52, 301, 335-337). Fur-thermore, membrane vesicles prepared fromNR-70, NR-76, and DL-54 were much morepermeable to protons than were parental vesi-cles and, in both cases, the defect could again berepaired by the addition of DCCD (16, 336). Itmust be concluded, therefore, that the stalkcomponent of F1 plays an important role inmaintaining the integrity of the coupling mem-brane, presumably by preventing the uselessdissipation of the transmembrane proton gra-dient. It has been suggested that the effect of amutation in which the F1 is either physically

deformed or missing from the membrane (e.g.,in mutants DL-54, NR-70, NR-76, and N1,4, butnot uncA) is to cause the internal exit of theproton-translocating channel in Fo to becomeexposed, with the resultant loss of respiration-linked energization (16). The ability of DCCDto repair these membranes suggests that theDCCD-binding protein is located in Fo suchthat it forms a transmembrane proton-translo-cating channel; DCCD would thus inhibit oxi-dative phosphorylation in the intact ATPasecomplex and restore energization to the F1-de-fective complex by preventing the passage ofprotons through Fo. The molecular weight ofDSCP is fully compatible with a transmem-brane location, but the ionophoric properties ofthe purified protein have yet to be investigated.

It is clear that the various unc- ATPase-mutants that have been isolated from E. coliare not the expression of a single genetic defect,although they all map between 73 and 74 minon the chromosome and are partly cotransduci-ble with the ilv locus. Instead, they all probablyreflect point mutations in a cluster of structuralgenes that code for the five polypeptide compo-nents of F.unc- ATPase+ mutants. The most inten-

sively studied E. coli mutants of this genotypeare unc-B (AN-283) (79, 81), BV4, A144 and K11(212, 301), MDB (93), uncD (392), BG-31 (373)etc-15 (48, 68). Membrane vesicles preparedfrom these mutants exhibited ATP hydrolysis,yet failed to catalyze either oxidative phospho-rylation or ATP-dependent membrane energi-zation; their capacity for using respiration togenerate a utilizable energized state was ex-tremely variable. However, many of these mu-tants were capable of growing anaerobically onglucose with the addition of exogenous elec-tron acceptors. This paradox was subsequentlyresolved by the observation that these mutantscontained a highly active fumarate reductaseand thus presumably generated an energizedmembrane state via anaerobic respiration tofumarate produced by endogenous metabolism(338); genetic deletion of fumarate reductaseeliminated this capacity for anaerobic growth.The ATPase activity of membranes from mu-

tants uncB, BV4 A144, and K11 was at least 50%of that exhibited by parental membranes but,whereas most of the cellular ATPase activity ofthe parent and uncB strains was associatedwith the coupling membrane, over 90% of thetotal activity of strains BV4, K11, and A144 waslocated in the supernatant fraction (213). Theseapparently contradictory figures are explainedby the observation that the total ATPase activ-ity of each of these latter three mutants was up

BACTERIOL. REV.


http://mm

br.asm.org/

Dow

nloaded from



to fivefold higher than in the parent strain.Furthermore, the F1 was so weakly attached tothe coupling membrane that it could be re-moved simply by recentrifuging the particles in50 mM Tris-sulfate buffer (pH 7.8) + 10 mMMgSO4, i.e., it was not necessary, as was thecase with parental membranes, to lower theionic strength and Mg2+ concentration of thewash medium in order to effect the release ofthe ATPase.The addition of solubilized ATPase from

uncB membranes to depleted membranes fromthe parent or uncA strains brought about resto-ration of both oxidative phosphorylation andATP-dependent transhydrogenation; in con-trast, depleted membranes from the uncB mu-tant could not be reactivated by parental F, (79,81). Similarly, crude solubilized ATPase prepa-rations from mutants K11 and A144 were able torestore both the ATP- and respiration-depend-ent quenching of ACMA fluorescence whenadded to depleted parental membranes,whereas the addition of parental supernatantsto depleted mutant membranes was again with-out effect. In addition, the response of sodiumdodecyl sulfate-treated membranes from K11and A1. to antibodies prepared against the a,,(, a + (3, and y subunits of parental F1 wasidentical to that of similarly treated parentalmembranes (213, 301). It was concluded thatmutants uncB, Kl, and A144 contain a normalF1 and that the genetic lesion in each of theseorganisms is expressed as a defect in an as yetunidentified component of the membrane-bound Fr-F1 complex (possibly DSCP) thatleads to inefficient energy transduction; in K11and Al,., this defect is also expressed as a weakbinding of F, to the coupling membrane.The ATP-dependent, energy-transducing

properties ofmutant BV4 were generally similarto those of K11 and A1l,; in contrast, BV4 wasuniquely capable of using respiration to drivethe active transport of amino acids (213). Hy-bridization experiments of the type describedabove indicated that BV4 membranes were ap-parently normal (since they could be reconsti-tuted by parental ATPase) and suggested thatthe genetic lesion was probably expressed as adefective ATPase (since solubilized ATPasefrom BV4 failed to reconstitute depleted paren-tal membranes). The defect in the ATPase wasclearly one that did not abolish its capacity forATP hydrolysis or alter its immunodiffusionprofile, although it did affect the ability of theATPase to act as a coupling factor and to bindwith the membrane. The properties probablyreflect a relatively small structural alterationin the y or 8 subunit of F, rather than in either

of its catalytic subunits. It has been concludedthat mutant MDB may also have a defectivesubunit in the stalk region of its ATPase, sincethe phenotype of this mutant is very similar tothat of BV4 (93).The ATPase activities of membranes pre-

pared from mutants BV4, K11, and A,,. weremarkedly resistant to DCCD but, whereas ad-dition of this inhibitor to membranes from Kiland A144 restored defective respiration-linkedreactions, it had no similar restorative effect onmembranes from BV4. However, the variousATP-dependent reactions that could be recon-stituted via the formation of active ATPasehybrids from membrane and supernatant frac-tions were all sensitive to DCCD. These obser-vations have led to the suggestion that confer-ral of sensitivity to DCCD is dependent uponthe presence of two polypeptides: DSCP (thecomponent of Fo that binds the inhibitor) and asecond, as yet unidentified, component of theATPase complex which is involved in the bind-ing of F1 to the coupling membrane and whichappears in the supernatant fraction of thesemutants (213).

Strain BG-31 contained a highly active,DCCD-resistant ATPase, but was capable ofmembrane energization only at the expense ofrespiration (373). This lesion was located in theFo region which appeared to lack a polypeptide(molecular weight, 54,000) that was present inthe parent organism. In this case it was con-cluded that DCCD resistance was probably areflection ofan alteration in the binding of F1 tothe coupling membrane, such that it preventedaccess ofDCCD to its binding site, rather thanto a lesion in DSCP per se.Mutant etc-15 exhibited slightly lowered

ATPase activity, but was strikingly deficient inits ability to catalyze active transport andtranshydrogenation at the expense of eitherrespiration or ATP (48, 178). Since electro-phoresis of the mutant F1 indicated that thegenetic lesion was expressed as a defective ysubunit (48), it was concluded that this subunitwas not essential for ATPase activity, althoughit was necessary for efficient energy transduc-tion. In the latter context, the y subunit mayneed to be present in order for the F1 to bindmost efficiently to the coupling membrane andthus block the wasteful dissipation of theenergized state produced by respiration.The uncD mutant is a unique member of the

unc ATPase+ class in that its ATPase activity,in contrast to that of the wild-type organism,was stimulated much more strongly by Ca2+than by Mg2+ (392). However, the addition ofCa2+ to membrane vesicles did not restore their

VOL. 41, 1977


http://mm

br.asm.org/

Dow

nloaded from



capacity for either oxidative phosphorylation orATP-dependent transhydrogenation. It wastentatively concluded that the lesion leads to analteration in a subunit of F1 which is responsi-ble for imposing divalent ion specificity andwhich is also necessary for energy transductionto and from ATP.

All of the unc- ATPase+ mutants that havebeen examined resemble the unc- ATPase- andDCCD-resistant mutants, insofar as they mapat approximately 73.5 min on the E. coli chro-mosome and are partly cotransducible with theilv and asn loci. Since some of these mutantsare defective in components of F0, whereas oth-ers probably contain altered subunits of F1, it isclear that the structural genes for the variouspolypeptides that comprise these two couplingfactors lie close together on the chromosome. Incontrast, Cox et al. (83) have described a possi-ble regulatory mutant ofE. coli (AN-295) whichmaps at approximately 77 min and is partlycotransducible with the argH gene. Mem-branes prepared from this mutant catalyzedATP-dependent transhydrogenation and con-tained an apparently normal ATPase with anactivity that was several times higher thanthat of wild-type membranes; F1 purified fromthe mutant successfully reconstituted ATP-de-pendent transhydrogenation in depleted uncAmembranes. Genetic evidence indicated thatthe high ATPase activity reflected derepressionof the enzyme, but is is not known whether thisis the direct result of a mutation in a regulatorgene or whether it is merely an indirect effect ofa mutation elsewhere.

CONCLUSIONS

The work described in this review allows usto draw several general conclusions regardingbacterial oxidative phosphorylation. Thus, it isclear that, in spite of the enormous diversitythat is encountered in bacterial respiratorychains, there exist striking similarities be-tween proton-translocating redox segmentsfrom different systems. In contrast to mitochon-dria, however, the number of potential energyconservation sites available to, and utilized by,different bacteria vary widely. Indeed, for anyone bacterium the number of such sites canchange, depending upon the growth conditionsemployed. It should also be apparent that, withappropriate experimental techniques, oxidativephosphorylation in bacteria can be studied aseasily, specifically, and conveniently as in mi-tochondria. There are many advantages instudying oxidative phosphorylation in bacterialsystems; for example, the availability of a wide

range of mutants and the possibility of effectingalterations to the various components synthe-sized by the cell in response to changes in thegrowth conditions. The ability to vary the com-ponents responsible for oxidative phosphoryla-tion at the discretion of the investigator is atechnique not generally available in equivalentstudies with mitochondria. Yeast and otherfungi can be used, but interpretation of theresults obtained is complicated by the fact thatcertain mitochondrial enzyme complexes con-tain polypeptides synthesized from the mito-chondrial, as well as the nuclear, DNA. Thepotential use of these approaches in bacteriahas by no means been fully exploited in theanalysis of the control mechanisms that serveto regulate the synthesis and functional activ-ity of different electron transport chains. Inaddition, their application to the more funda-mental problem of the mechanism of oxidativephosphorylation remains a challenge for thefuture.

It should also be remembered that certainbacteria, for example, chemolithotrophs, alkal-ophiles, and acidophiles, have had to developunusual enzymes and reaction pathways in or-der to exploit a wide variety of potential elec-tron donors and acceptors and, thereby, adaptto life in somewhat exotic ecosystems. The iden-tification and characterization of these novelcomponents and their role in oxidative phos-phorylation should prove of great scientific in-terest.The final point we should like to consider

briefly is the possible significance of the in-creasing order of complexity of the electrontransport chains, described in this review, tothe evolutionary development of more efficientpathways for the conservation of energy. Thus,examples have been given of electron transportchains in which the oxidation ofone molecule ofNADH is linked to the synthesis of one (in thepresence of fumarate, Fig. 4) two (in the pres-ence of oxygen, Fig. 3, or nitrate, Fig. 5), three(in the presence of oxygen, Fig. 2 and 6), orbetween one and three (according to the oxygenconcentration, Fig. 7) molecules of ATP. Thecomplexity of biological mechanisms has in-creased gradually through time and, althoughit would be foolhardy to assume any direct evo-lutionary significance in the examples given inthis review, a comparative study of thebranched respiratory chains of bacteria, interms of their individual redox components andfunctional activity, will clearly be informativein unravelling the evolutionary development ofthe mitochondrial respiratory chain and itsability to conserve energy.

BACTERIOL. REV.


http://mm

br.asm.org/

Dow

nloaded from



ACKNOWLEDGMENTSWe would like to extend our thanks and apprecia-

tion to our many colleagues and co-workers whohave helped in a variety ofways to make this reviewpossible. Our work described in this review wasfinanced by the Science Research Council, UnitedKingdom, through grants B/RG/58518 to (B.A.H.)and B/SR/59102 (to C.W.J.).

LITERATURE CITED1. Aboderin, A. A., E. Boedefeld, and P. L. Luisi.

1973. Reaction of chicken egg white lysozymewith 7-chloro4-nitrobenz-2-oxa-1,3-diazole.Biochim. Biophys. Acta 328:20-30.

2. Abram, D. 1965. Electron microscope observa-tions on intact cells, protoplasts, and the cyto-plasmic membrane of Bacillus stearother-mophilus. J. Bacteriol. 89:855-873.

3. Abrams, A. 1965. The release of bound adeno-sine triphosphatase from isolated bacterialmembranes and the properties of the solubi-lised enzyme. J. Biol. Chem. 240:3675-3681.

4. Abrams, A., and C. Baron. 1967. The isolationand subunit structure of Streptococcal mem-brane ATPase. Biochemistry 6:225-229.

5. Abrams, A., and C. Baron. 1970. Inhibitoryaction of carbodiimides on bacterial mem-brane ATPase. Biochem. Biophys. Res. Com-mun. 41:858-863.

5a.Abrams, A., C. Jensen, and D. H. Morris. 1976.Role of Mg2+ ions in the subunit structure andmembrane-binding properties of bacterial en-ergy-transducing ATPase. Biochem. Biophys.Res. Commun. 69:804-811.

6. Abrams, A., A. E. Nolan, C. Jensen, and J. B.Smith. 1973. Tightly bound adenine nucleo-tide in bacterial membrane ATPase. Bio-chem. Biophys. Res. Commun. 55:22-29.

7. Abrams, A., and J. B. Smith. 1974. Bacterialmembrane ATPase, p. 395429. In P. D. Boyer(ed.), The enzymes, vol. X, 3rd ed. AcademicPress Inc., New York and London.

8. Abrams, A., J. B. Smith, and C. Baron. 1972.Carbodiimide-resistant membrane adenosinetriphosphatase in mutants of Streptococcusfaecalis. I. Studies on the mechanism of resist-ance. J. Biol. Chem. 247:1484-1488.

9. Ackrell, B. A. C., S. K. Erickson, and C. W.Jones. 1972. The respiratory chain NADPHdehydrogenase ofAzotobacter vinelandii. Eur.J. Biochem. 26:387-392.

10. Ackrell, B. A. C., and C. W. Jones. 1971. Therespiratory system of Azotobacter vinelandii.1. Properties of phosphorylating respiratorymembranes. Eur. J. Biochem. 20:22-28.

11. Ackrell, B. A. C., and C. W. Jones. 1971. Therespiratory system of Azotobacter vinelandii;2. Oxygen effects. Eur. J. Biochem. 20:29-35.

12. Adolfsen, R., J. A. McClung, and E. N. Moud-rianakis. 1975. Electrophoretic microhetero-geneity and subunit composition of the 13Scoupling factors of oxidative and photosyn-thetic phosphorylation. Biochemistry 14:1727-1735.

13. Adolfsen, R. and E. N. Moudrianakis. 1971.Purification and properties oftwo soluble cou-pling factors of oxidative phosphorylationfrom Alcaligenes faecalis. Biochemistry 10:2247-2253.

14. Adolfsen, R., and E. N. Moudrianakis. 1974.Molecular microheterogeneity and subunitcomposition of the 13S coupling factors of oxi-dative and photosynthetic phosphorylation.Fed. Proc. 33:1330.

15. Alberty, R. A. 1968. Effect ofpH and metal ionconcentration on the equilibrium hydrolysisof adenosine triphosphate to adenosine di-phosphate. J. Biol. Chem. 243:1337-1343.

16. Altendorf, K., F. M. Harold, and R. D. Simoni.1974. Impairment and restoration of the ener-gised stated in membrane vesicles ofa mutantofEscherichia coli lacking adenosine triphos-phatase. J. Biol. Chem. 249:45874593.

17. Altendorf, K. H., and L. A. Staehelin. 1974.Orientation ofmembrane vesicles from Esche-richia coli as detected by freeze-cleave elec-tron microscopy. J. Bacteriol. 117:888-899.

18. Altendorf, K., and W. Zitzmann. 1975. Identi-fication of the DCCD-reactive protein of theenergy transducing adenosine triphosphatasecomplex from Escherichia coli. FEBS Lett.59:268-272.

19. Andreu, J. M., J. A. Albendea, and E. Mu-noz. 1973. Membrane adenosine triphospha-tase of Micrococcus lysodeikticus; molecularproperties of the purified enzyme unstimu-lated by trypsin. Eur. J. Biochem. 37:505-515.

20. Andreu, J. M., and E. Munoz. 1975. Micro-coccus lysodeikticus ATPase; purification bypreparative gel electrophoresis and subunitstructure studied by urea and sodium dodecyl-sulphate gel electrophoresis. Biochim. Bio-phys. Acta 387:228-233.

21. Asano, A., N. S. Cohen, R. F. Baker, and A. F.Brodie. 1973. Orientation of the cell mem-brane in ghosts and electron transport parti-cles of Mycobacterium phlei. J. Biol. Chem.248:3386-3397.

22. Asano, A., H. Hirata, and A. F. Brodie. 1972.Factors required for activation of oxidativephosphorylation in protoplast ghosts of Myco-bacterium phlei Biochem. Biophys. Res. Com-mun. 46:1340-1346.

23. Asano, A., K. Imai, and R. Sato. 1967. Oxida-tive phosphorylation in Micrococcus denitrifi-cans. II. The properties of pyridine nucleotidetranshydrogenase. Biochim. Biophys. Acta143:477-486.

24. Asano, A., K. Imai, and R. Sato. 1967. Oxida-tive phosphorylation in Micrococcus denitrifi-cans. III. ATP-supported reduction of NAD+by succinate. J. Biochem. (Tokyo) 62:210-214.

25. Aschroft, J. R., and B. A. Haddock. 1975. Syn-thesis of alternative membrane-bound redoxcarriers during aerobic growth ofEscherichiacoli in the presence ofpotassium cyanide. Bio-chem. J. 148:349-352.

26. Azoulay, E., P. Couchoud-Beaumont, and J.M. Lebeault. 1972. etude des mutants chlor-

VOL. 41, 1977


http://mm

br.asm.org/

Dow

nloaded from



ate-r6sistants chez Escherichia coli K12. IV.Isolement purification, et 6tude de la nitrate-reductase reconstitut6e in vitro par comple-mentation. Biochim. Biophys. Acta 256:670-680.

27. Azoulay, E., J. Puig, and P. Couchoud-Beau-mont. 1969. etude des mutants chlorate-r6-sistants chezEscherichia coli K12. I. Reconsti-tution in vitro de l'activit6 nitrate-reductaseparticulaire chez Escherichia coli K12. Bio-chim. Biophys. Acta 171:238-252.

28. Baak, J. M., and P. W. Postma. 1971. Oxida-tive phosphorylation in intact Azotobacter vi-nelandii. FEBS Lett. 19:189-192.

29. Baillie, R. D., C. Hou, and P. D. Bragg. 1971.The preparation and properties of a solubi-lized respiratory complex from Escherichiacoli. Biochim. Biophys. Acta 234:46-56.

30. Baltscheffsky, H., and M. Baltscheffsky. 1974.Electron transport phosphorylation. Annu.Rev. Biochem. 43:871-897.

31. Barnes, E. M. 1972. Respiration coupled glu-cose transport in membrane vesicles fromAzotobacter vinelkndii. Arch. Biochem. Bio-phys. 152:795-799.

32. Barnes, E. M. 1973. Multiple sites for the cou-pling of glucose transport to the respiratorychain of membrane vesicles from Azotobactervinelandii. J. Biol. Chem. 248:8120-8124.

33. Baron, C., and A. Abrams. 1971. Isolation of abacterial membrane protein, nectin, essentialfor the attachment of ATPase. J. Biol. Chem.246:1542-1544.

34. Barrera, C. R., and P. Jurtshuk. 1970. Charac-terisation of the highly active isocitrate(NADP+) dehydrogenase of Azotobacter vine-landii. Biochim. Biophys. Acta 220:416-429.

35. Bartsch, R. G. 1968. Bacterial cytochromes.Annu. Rev. Microbiol. 22:181-200.

36. Beechey, R. B. 1974. Structural aspects of mi-tochondrial adenosine triphosphatase. Bio-chem. Soc. Trans. 2:466-471.

37. Beechey, R. B., and K. J. Cattell. 1973. Mito-chondrial coupling factors. Curr. Top. Bioe-nerg. 5:306-357.

38. Beljanski, M., and M. Beljanski. 1957. Sur laformation d'enzymes respiratoires chez unmutant d'Escherichia coli streptomycine-r6-sistant et auxotrophe pour l'hemine. Ann.Inst. Pasteur (Paris) 92:396-412.

39. Berzborn, R. J. 1972. Tennung von Unterein-heiten des kopplungsfaktors 1 der Chloroplas-ten (CF1) und deren immunologische charak-terisierung. Hoppe-Seyler's Z. Physiol. Chem.353:693.

40. Bhattacharyya, P., W. Epstein, and S. Silver.1971. Valinomycin-induced uptake of potas-sium in membrane vesicles from Escherichiacoli. Proc. Natl. Acad. Sci. U.S.A. 68:1488-1492.

41. Bishop, D. H. L., K. P. Pandya, and H. K.King. 1962. Ubiquinone and vitamin K in bac-teria. Biochem. J. 83:606-614.

42. Boonstra, J., M. T. Huttunen, W. K. Konings,and H. R. Kaback. 1975. Anaerobic electron

BACTERIOL. REV.

transport in Escherichia coli membrane vesi-cles. J. Biol. Chem. 250:6792-6798.

43. Boxer, D. H., and R. A. Clegg. 1975. A trans-membrane-location for the proton-translocat-ing reduced ubiquinone-nitrate reductase seg-ment of the respiratory chain of Escherichiacoli. FEBS Lett. 60:54-57.

44. Boyer, P. D. 1974. Conformational coupling inbiological energy transductions. Biochim.Biophys. Acta 13:289-301.

45. Boyer, P. D., R. L. Cross, and W. Momsen.1973. A new concept for energy coupling inoxidative phosphorylation based on a molecu-lar explanation of the oxygen exchange reac-tions. Proc. Natl. Acad. Sci. U.S.A. 70:2837-2839.

46. Bragg, P. D. 1970. Reduction of nonheme ironin the respiratory chain of Escherichia coli.Can. J. Biochem. 48:777-783.

47. Bragg, P. D., P. L. Davies, and C. Hou. 1972.Function of energy-dependent transhydrogen-ase in Escherichia coli. Biochem. Biophys.Res. Commun. 47:1248-1255.

48. Bragg, P. D., P. L. Davies, and C. Hou. 1973.Effect of removal or modification of subunitpolypeptides on the coupling factor and hydro-lytic activities of the Ca2+ and Mg2+-activatedadenosine triphosphatase of Escherichia coli.Arch. Biochem. Biophys. 159:664-670.

49. Bragg, P. D., and C. Hou. 1967. Reduced nico-tinamide adenine dinucleotide oxidation inEscherichia coli particles. I. Properties andcleavage of the electron transport chain.Arch. Biochem. Biophys. 119:194-201.

50. Bragg, P. D., and C. Hou. 1967. Reduced nico-tinamide adenine dinucleotide oxidation inEscherichia coli particles. II. NADH dehydro-genases. Arch. Biochem. Biophys. 119:202-208.

51. Bragg, P. D., and C. Hou. 1972. Purification ofa factor for both aerobic-driven and ATP-driven energy-dependent transhydrogenasesofEscherichia coli. FEBS Lett. 28:309-312.

52. Bragg, P. D., and C. Hou. 1973. Reconstitutionof energy-dependent transhydrogenase inATPase-negative mutants ofEscherichia coli.Biochem. Biophys. Res. Commun. 50:729-736.

53. Bragg, P. D., and C. Hou. 1974. Energizationof energy-dependent transhydrogenase ofEscherichia coli at a second site of energyconservation. Arch. Biochem. Biophys.163:614-616.

54. Bragg, P. D., and C. Hou. 1975. Subunit com-position, function and spatial arrangement inthe Ca2+ and Mg2+ ATPases ofEscherichia coliand Salmonella typhimurium. Arch. Bio-chem. Biophys. 167:311-321.

55. Brand, M. D., B. Reynafarje, and A. L. Lehn-inger. 1976. Stoichiometric relationship be-tween energy-dependent proton ejection andelectron transport in mitochondria. Proc.Natl. Acad. Sci. U.S.A. 73:437-441.

56. Bray, R. C., S. P. Vincent, D. J. Lowe, R. A.Clegg, and P. B. Garland. 1976. Electron par-amagnetic resonance studies on the molybde-


http://mm

br.asm.org/

Dow

nloaded from



num of nitrate reductase from E. coli K12.Biochem. J. 155:201-203

57. Brice, J. M., J. F. Law, D. J. Meyer, and C. W.Jones. 1974. Energy conservation in Esche-richia coli and Klebsiella pneumoniae. Bio-chem. Soc. Trans. 2:523-526.

58. Broman, R. L., W. J. Dobrogosz, and D. C.White. 1974. Stimulation of cytochrome syn-thesis in Escherichia coli by cyclic AMP.Arch. Biochem. Biophys. 162:595-601.

59. Bryan-Jones, D. G., and R. H. Whittenbury.1969. Haematin-dependent oxidative phos-phorylation in Streptococcus faecalis. J. Gen.Microbiol. 58:247-260.

60. Burnell, J. N., P. John, and F. R. Whatley.1975. The reversibility of active sulphatetransport in membrane vesicles ofParacoccusdenitrificans. Biochem. J. 150:527-536.

61. Burns, R. C., R. D. Holsten, and R. W. F.Hardy. 1970. Isolation by crystallisation oftheMo-Fe protein of Azotobacter nitrogenase.Biochem. Biophys. Res. Commun. 39:90-99.

62. Butlin, J. D., G. B. Cox, and F. Gibson. 1973.Oxidative phosphorylation in Escherichia coliK-12; the genetic and biochemical characteri-sation of a strain carrying a mutation in theunc B gene. Biochim. Biophys. Acta 292:366-375.

63. Carmeli, C. 1970. Proton translocation inducedby ATPase activity in chloroplasts. FEBSLett. 7:297-300.

64. Carreira, J., J. A. Leal, M. Rojas, and E.Munoz. 1973. Membrane ATPase of Esche-richia coli K12; selective solubilisation of theenzyme and its stimulation by trypsin in thesoluble and membrane-bound states. Bio-chim. Biophys. Acta 307:541-556.

65. Castor, L. N., and B. Chance. 1959. Photo-chemical determinations of the oxidases ofbacteria. J. Biol. Chem. 234:1587-1592.

66. Cattell, K. J., I. G. Knight, C. R. Lindop, andR. B. Beechey. 1970. The isolation of dicyclo-hexylcarbodiimide-binding proteins from mi-tochondrial membranes. Biochem. J. 117:1011-1013.

67. Cattell, K. J., C. R. Lindop, I. G. Knight, andR. B. Beechey. 1971. The identification of thesite of action of NN-dicyclohexylcarbodi.im-ide as a proteolipid in mitochondrial mem-branes. Biochem. J. 125:169-177.

68. Catterall, W. A., W. A. Coty, and P. L. Peder-sen. 1973. ATPase from rat liver mitochon-dria. III. Subunit composition. J. Biol. Chem.248:7427-7431.

69. Catterall, W. A., and P. L. Pedersen. 1971.ATPase from rat liver mitochondria. I. Purifi-cation, homogeneity and physical properties.J. Biol. Chem. 246:4987-4994.

70. Cavari, B. Z., Y. Avi-Dor, and N. Grossowicz.1968. Induction by oxygen of respiration andphosphorylation of anaerobically grown Esch-erichia coli. J. Bacteriol. 96:751-759.

71. Chance, B. 1955. Intracellular reaction kinet-ics. Faraday Spec. Discuss. Chem. Soc.20:205-216.

72. Clegg, R. A. 1976. Purification and some prop-erties of nitrate reductase (EC 1.7.99.4) fromEscherichia coli K12. Biochem. J. 153:533-541.

73. Clegg, R. A., and P. B. Garland. 1971. Non-haem iron and the dissociation of piercidin Asensitivity from site 1 energy conservation inmitochondria from Torulopsis utilis. Bio-chem. J. 124:135-154.

74. Cobley, J. G. 1976. Reduction of cytochromesby nitrite in electron-transport particles fromNitrobacter winogradskyi. Proposal of a mech-anism of H+ translocation. Biochem. J. 156:493-498.

75. Cobley, J. G., S. Grossman, T. P. Singer, andH. Beinert. 1975. Piercidin A sensitivity, site1 phosphorylation, and reduced nicotinamideadenine dinucleotide dehydrogenase duringiron-limited growth ofCandida utilis. J. Biol.Chem. 250:211-217.

76. Cole, J. A., and J. W. T. Wimpenny. 1966. Theinter-relationships of low redox potential cy-tochrome csu and hydrogenase in facultativeanaerobes. Biochim. Biophys. Acta 128:119-425.

77. Cole, J. A., and J. W. T. Wimpenny. 1968.Metabolic pathways for nitrate reduction inEscherichia coli. Biochim. Biophys. Acta162:39-48.

78. Cole, J. S., and M. I. H. Aleem. 1973. Electrontransport-linked compared with proton-in-duced ATP generation in Thiobacillus novel-lus. Proc. Natl. Acad. Sci. U. S. A. 70:3571-3575.

79. Cox, G. B., and F. Gibson. 1974. Studies onelectron transport and energy-linked reac-tions using mutants of Escherichia coli.Biochim. Biophys. Acta 346:1-25.

80. Cox, G. B., F. Gibson, R. K. J. Luke, N. A.Newton, I. G. O'Brien, and H. Rosenberg.1970. Mutations affecting iron transport inEscherichia coli. J. Bacteriol. 104:219-226.

81. Cox,- G. B., F. Gibson, and L. McCann. 1973.Reconstitution of oxidative phosphorylationand the adenosine triphosphate-dependenttranshydrogenase activity by a combinationof membrane fractions from unc A- and uncB- mutant strains of Escherichia coli K12.Biochem. J. 134:1015-1021.

82. Cox, G. B., F. Gibson, and L. McCann. 1974.Oxidative phosphorylation in Escherichiacoli K12; an uncoupled mutant with alteredmembrane structure. Biochem. J. 138:211-215.

83. Cox, G. B. F. Gibson, L. M. McCann, J. D.Butlin, and F. L. Crane. 1973. Reconstitu-tion of the energy-linked transhydrogenaseactivity in membranes from a mutant strainof Escherichia coli K12 lacking magnesiumion- or calcium ion-stimulated adenosine tri-phosphatase. Biochem. J. 132:689-695.

84. Cox, G. B., N. A. Newton, J. D. Butlin, and F.Gibson. 1971. The energy-linked transhydro-genase reaction in respiratory mutants ofEscherichia coli K12. Biochem. J. 125:489-

VOL. 41, 1977


http://mm

br.asm.org/

Dow

nloaded from



493.85. Cox, G. B., N. A. Newton, F. Gibson, A. M.

Snoswell, and J. A. Hamilton. 1970. Thefunction of ubiquinone in Escherichia coli.Biochem. J. 117:551-562.

86. Cox, R., and H. P. Charles. 1973. Porphyrin-accumulating mutants ofEscherichia coli. J.Bacteriol. 113:122-132.

87. Cox, R. B., and J. R. Quayle. 1975. The auto-trophic growth of Micrococcus denitrificanson methanol. Biochem. J. 150:569-571.

88. Crane, R. T., I. L. Sun, and F. L. Crane. 1975.Lipophilic chelator inhibition of electrontransport in Escherichia coli. J. Bacteriol.122:686-690.

89. Creaghan, I. T., and J. R. Guest. 1972. ambermutants of the a-ketoglutarate dehydrogen-ase gene of Escherichia coli K12. J. Gen.Microbiol. 71:207-220.

90. Czerwinski, E. W., and F. S. Mathews. 1974.Location of the iron atom and the non-crys-tallographic symmetry elements in cyto-chrome b562. J. Mol. Biol. 86:49-57.

91. Dalton, H., and J. R. Postgate. 1969. Effect ofoxygen on growth of Azotobacter chroococ-cum in batch and continuous cultures. J.Gen. Microbiol. 54:463-473.

92. Dalton, H., and J. R. Postgate. 1969. Growthand physiology of Azotobacter chroococcumin continuous culture. J. Gen. Microbiol.56:307-319.

93. Daniel, J., M.-P. Roisin, C. Burstein, and A.Kepes. 1975. Mutants ofEscherichia coli K12unable to grow on non-fermentable carbonsubstrates. Biochim. Biophys. Acta 376:195-209.

94. Davies, P. L., and P. D. Bragg. 1972. Proper-ties of a soluble Ca2+- and Mg2+- activatedATPase released from Escherichia coli mem-branes. Biochim. Biophys. Acta 266:273-284.

95. Deeb, S. S., and L. P. Hager. 1964. Crystallinecytochrome b1 from Escherichia coli. J. Biol.Chem. 239:1024-1031.

96. DerVartanian, D. V., and R. Bramlett. 1970.Electron paramagnetic resonance studies of95Mo-enriched NADH dehydrogenase iso-lated from iron deficientAzotobacter vinelan-dii. Biochim. Biophys. Acta 220:443-448.

97. DerVartanian, D. V., and P. Forget. 1975. Thebacterial nitate reductase, EPR studies onthe enzyme A of Escherichia coli K12.Biochim. Biophys. Acta 379:74-80.

98. Deters, D. W., E. Racker, N. Nelson, and H.Nelson. 1975. Partial resolution of the en-zymes catalysing photophosphorylation. XV.Approaches to the active site of coupling fac-tor 1. J. Biol. Chem. 250:1041-1047.

99. Drozd, J., and J. R. Postgate. 1970. Effects ofoxygen on acetylene reduction, cytochromecontent and respiratory activity ofAzotobac-ter chroococcum. J. Gen. Microbiol. 63:63-73.

100. Douglas, M. W., F. B. Ward, and J. A. Cole.1974. The formate hydrogenlyase activity ofcytochrome c552-deficient mutants of Esche-richia coli K-12. J. Gen. Microbiol. 80:557-560.

101. Downs, A. J., and C. W. Jones. 1975. Energyconservation in Bacillus megaterium. Arch.Microbiol. 105:159-167.

102. Downs, A. J., and C. W. Jones. 1975. Respira-tion-linked proton translocation in Azotobac-ter vinelandii. FEBS Lett. 60:42-46.

103. Eilermann, L. J. M. 1970. Oxidative phospho-rylation in Azotobacter vinelandii. Atebrineas a fluorescent probe for the energised state.Biochim. Biophys. Acta 216:231-233.

104. Eilermann, L. J. M., H. G. Pandit-Hoven-kamp, and A. H. J. Kolk. 1970. Oxidativephosphorylation in Azotobacter vinelandiiparticles; phophorylation sites and respira-tory control. Biochim. Biophys. Acta 197:25-30.

105. Eilermann, L. J. M., H. G. Pandit-Hoven-kamp, M. Van der Meer-Van Buren, A. H. J.Kolk, and M. Feenstra. 1971. Oxidativephosphorylation in Azotobacter vinelandii;effect of inhibitors and uncouplers on P/Oratio, trypsin-induced ATPase and ADP-stimulated respiration. Biochim. Biophys.Acta 245:305-312.

106. Enoch, H. G., and R. L. Lester. 1974. The roleof a novel cytochrome b-containing nitratereductase and quinone in the in vitro recon-struction of formate-nitrate reductase activ-ity of E. coli. Biochem. Biophys. Res. Com-mun. 61:1234-1241.

107. Enoch, H. G., and R. L. Lester. 1975. Thepurification and properties of formate dehy-drogenase and nitrate reductase from Esche-richia coli. J. Biol. Chem. 250:6693-6705.

108. Erickson, S. K., and H. Diehl. 1973. The termi-nal oxidases of Azotobacter vinelandii. Bio-chem. Biophys. Res. Commun. 50:321-327.

109. Evans, D. J. 1970. Membrane Mg2+ (Ca2+)-acti-vated ATPase released from Escherichia colimembranes. J. Bacteriol. 104:1203-1212.

110. Farmer, I. S., and C. W. Jones. 1976. Theenergetics ofEscherichia coli during aerobicgrowth in continuous culture. Eur. J. Bio-chem. 67:115-122.

111. Farron, F. 1970. Isolation and properties of achloroplast coupling factor and heat-acti-vated ATPase. Biochemistry 9:3823-3828.

112. Faust, D. J., and P. J. Vandemark. 1970. Phos-phorylation coupled to NADH oxidation withfumarate in Streptococcus faecalis 10CL.Arch. Biochem. Biophys. 137:393-398.

113. Ferguson, S. J., P. John, W. J. Lloyd, G. K.Radda, and F. R. Whatley. 1974. Selectiveand reversible inhibition of the ATPase ofMicrococcus denitrificans by 7-chloro-4-nitro-benz-2-oxa-1,3 diazole. Biochim. Biophys.Acta 357:457-461.

114. Ferguson, S. J., W. J. Lloyd, and G. K. Radda.1974. A specific and reversible inactivation ofsoluble ox heart mitochondrial adenosine tri-phosphatase. Biochem. Soc. Trans. 2:501-502.

115. Ferguson, S. J., W. J. Lloyd, and G. K. Radda.1974. An unusual and reversible modifica-tion of soluble beefheart mitochondrial ATP-ase. FEBS Lett. 38:234-236.

BACTERIOL. REV.


http://mm

br.asm.org/

Dow

nloaded from



116. Fewson, C. A., and D. J. D. Nicholas. 1961.Respiratory enzymes in Micrococcus denitri-ficans. Biochim. Biophys. Acta 48:208-210.

117. Fillingame, R. H. 1975. Identification and pu-rification of the factor conferring carbodi-imide sensitivity to the membrane-associ-ated ATPase ofE. coli. Fed. Proc. 34:577.

118. Fillingame, R. H. i975. Identification of thedicyclohexylcarbodiimide-reactive proteincomponent of the adenosine 5'-triphosphateenergy-transducing system of Escherichiacoli. J. Bacteriol. 124:870-883.

119. Forget, P. 1971. Les nitrate-rdductase bacter-iennes. Solubilisation, purification et pro-prietes de l'enzyme A de Micrococcus denitri-ficans Eur. J. Biochem. 18:442-450.

120. Forget, P. 1974. The bacterial nitrate reduc-tases. Solubilization, purification and prop-erties of the enzyme A of Escherichia coliK12. Eur. J. Biochem. 42:325-332.

121. Forget, P., and D. V. DerVartanian. 1972. Thebacterial nitrate reductases: EPR studies onnitrate reductase A from Micrococcus deni-trificans. Biochim. Biophys. Acta 256:600-606.

122. Fujita, T. 1966. Studies on soluble cytochromesin Enterobacteriaceae; Cytochrome bw2 andc,.,. J. Biochem. (Tokyo) 60:329-334.

123. Fukui, Y., and M. J. R. Salton. 1972. Commonpeptides in Micrococcus lysodeikticus mem-brane proteins. Biochim. Biophys. Acta288:65-72.

124. Fukuyama, T., and E. J. Ordal. 1965. Inducedbiosynthesis of formic hydrogen lyase iniron-deficient cells of Escherichia coli. J.Bacteriol. 90:673-680.

125. Futai, M. 1973. Membrane D-lactate dehydro-genase from Escherichia coli. Purificationand properties. Biochemistry 12:2468-2474.

126. Futai, M. 1974. Orientation ofmembrane vesi-cles from Escherichia coli prepared by differ-ent procedures. J. Membrane Biol. 15:15-28.

127. Futai, M., P. C. Sternweis, and L. A. Heppel.1974. Purification and properties of reconsti-tutively active and inactive ATPase fromEscherichia coli. Proc. Natl. Acad. Sci. U. S.A. 71:2725-2729.

128. Garland, P. B., R. A. Clegg, D. Boxer, J. A.Downie, and B. A. Haddock. 1975. Proton-translocating nitrate reductase of Esche-richia coli, p. 351-358. In E. Quagliariello, S.Papa, F. Palmieri, E. C. Slater, and N. Sili-prandi (ed.), Electron transfer chains andoxidative phosphorylation. North Holland-American Elsevier, Amsterdam.

129. Garland, P. B., R. A. Clegg, J. A. Downie, T.A. Gray, H. G. Lawford, and J. Skyrme.1972. Is the NADH dehydrogenase looped?,p. 105-117. In S. G. van den Bergh, P. Borst,L. L. M. van Deenen, J. C. Riemersma, E. C.Slater, and J. M. Tager (ed.), Mitochondria:biogenesis and bioenergetics. Biomem-branes: molecular arrangements and trans-port mechanisms. North Holland-AmericanElsevier, Amsterdam.

130. Garland, P. B., J. A. Downie, and B. A. Had-

dock. 1975. Proton translocation and the res-piratory nitrate reductase of Escherichiacoli. Biochem. J. 152:547-559.

131. Gel'man, N. S., M. A. Lukoyanova, and D. A.Ostrovskii. 1967. Respiration and phospho-rylation of bacteria. Plenum Press, NewYork.

132. Gibson, F. 1973. Chemical and genetic studieson the biosynthesis of ubiquinone by Esche-richia coli. Biochem. Soc. Trans. 1:317-326.

133. Gibson, F., and G. B. Cox. 1973. The use ofmutants ofEscherichia coli K12 in studyingelectron transport and oxidative phosphoryl-ation. Essays Biochem. 9:1-29.

134. Giordano, G., C. Riviere, and E. Azoulay.1975. Membrane reconstitution in Chl-r mu-tants of Escherichia coli K12. VII. Purifica-tion of the soluble ATPase of supernatantextracts and kinetics of incorporation intoreconstituted particles. Biochim. Biophys.Acta 389:203-218.

135. Glaser, J. M., and J. A. De Moss. 1971. Pheno-typic restoration by molybdate of nitrate re-ductase activity in chiD mutants of Esche-richia coli. J. Bacteriol. 108:854-860.

136. Glaser, J. H., and J. A. De Moss. 1972. Com-parison of nitrate reductase mutants ofEsch-erichia coli selected by alternative proce-dures. Mol. Gen. Genet. 116:1-10.

137. Gorneva, G. A., and I. D. Ryabova. 1974. Mem-brane orientation in vesicles from Micrococ-cus lysodeikticus cells. FEBS Lett. 42:271-274.

138. Green, D. E. 1974. The electromechanochemi-cal model for energy coupling in mitochon-dria. Biochim. Biophys. Acta 346:27-78.

139. Greville, G. D. 1969. A scrutiny of Mitchell'schemiosmotic hypothesis. Curr. Top. Bioe-nerg. 3:1-78.

140. Grossman, S., J. G. Cobley, T. P. Singer, andH. Beinert. 1974. Reduced nicotinamide ade-nine dinucleotide dehydrogenase, piercidinsensitivity, and site 1 phosphorylation in dif-ferent growth phases of Candida utilis. J.Biol. Chem. 249:3819-3826.

141. Guest, J. R. 1969. Biochemical and geneticstudies with nitrate reductase C-gene mu-tants of Escherichia coli. Mol. Gen. Genet.105:285-297.

142. Gutman, M., A. Schejter, and Y. Avi-Dor.1968. The preparation and properties of themembranal DPNH dehydrogenase fromEscherichia coli. Biochim. Biophys. Acta162:506-517.

143. Haddock, B. A. 1973. The reconstitution of oxi-dase activity in membranes derived from a 5-*aminolaevulinic acid-requiring mutant ofEscherichia coli. Biochem. J. 136:877-884.

144. Haddock, B. A., and J. A. Downie. 1974. Thereconstitution of functional respiratorychains in membranes from electron-trans-port deficient mutants of Escherichia coli asdemonstrated by quenching of atebrin fluo-rescence. Biochem. J. 142:703-706.

145. Haddock, B. A., J. A. Downie, and P. B. Gar-land. 1976. Kinetic characterization of the

VOL. 41, 1977


http://mm

br.asm.org/

Dow

nloaded from



membrane bound cytochromes ofEscherichiacoli grown under a variety of conditions byusing a stopped-flow dual-wavelength spec-trophotometer. Biochem. J. 154:285-294.

146. Haddock, B. A., J. A. Downie, and H. G. Law-ford. 1974. The function of ubiquinone respi-ration studies in cytochrome-deficient mu-tants ofEscherichia coli. Proc. Soc. Gen. Mi-crobiol. 1:50.

147. Haddock, B. A., and P. B. Garland. 1971. Ef-fect of sulphate-limited growth on mitochon-drial electron transfer and energy conserva-tion between reduced nicotinamide adeninedinucleotide and the cytochromes in Toru-lopsis utilis. Biochem. J. 124:155-170.

148. Haddock, B. A., and M. W. Kendall-Tobias.1975. Functional anaerobic electron trans-port linked to the reduction of nitrate andfumarate in membranes from Escherichiacoli as demonstrated by quenching of atebrinfluorescence. Biochem. J. 152:655-659.

149. Haddock, B. A., and H. U. Schairer. 1973.Electron transport chains ofEscherichia coli:reconstitution of respiration in a 5-aminolae-vulinic acid requiring mutant. Eur. J. Bio-chem. 35:34-45.

150. El Hackimi, Z., 0. Samuel, and R. Azerad.1974. Biochemical study on ubiquinone bio-synthesis in Escherichia coli. I. Specificity ofparu hydroxybenzoate polyprenyltransfer-ase. Biochimie 56:1239-1247.

151. Hachimori, A., N. Muramatsu, and Y. Nosoh.1970. Studies on an ATPase of thermophilicbacteria. I. Purification and properties.Biochim. Biophys. Acta 207:426-437.

152. Hafkenscheid, J. C. M., and S. L. Bonting.1969. Studies on (Na+-K+)-activated ATPase.XXIII. A Mg2+-ATPase in Escherichia coli,activated by monovalent cations. Biochim.Biophys. Acta 178:128-136.

153. Hamilton, J. A., G. B. Cox, F. D. Looney, andF. Gibson. 1970. Ubisemiquinone in mem-branes from Escherichia coli. Biochem. J.116:319-320.

154. Hamilton, W. A. 1975. Energy coupling in mi-crobial transport. Adv. Microbial Physiol.12:2-55.

155. Hammes, G. G., and D. W. Hilborn. 1971.Steady state kinetics of soluble and mem-brane-bound mitochondrial ATPase. Bio-chim. Biophys. Acta 233:580-590.

156. Hampton, M. L., and E. Freese. 1974. Expla-nation for the apparent inefficiency of re-duced nicotinamide adenine dinucleotide inenergizing amino acid transport in mem-brane vesicles. J. Bacteriol. 118:497-504.

157. Hanson, R. L., and E. P. Kennedy. 1973. En-ergy-transducing adenosine triphosphatasefrom Escherichia coli: purification, proper-ties, and inhibition by antibody. J. Bac-terial. 114:772-781.

158. Hare, J. E. 1975. Purification and characterisa-tion of an ATPase complex from membranesof Escherichia coli which is sensitive toDCCD. Biochem. Biophys. Res. Commun.66:1329-1337.

159. Hare, J. F., K. Olden, and E. P. Kennedy.1974. Heterogeneity of membrane vesiclesfrom Escherichia coli and their subfractiona-tion with antibody to ATPase. Proc. Natl.Acad. Sci. U. S. A. 71:4843-4846.

160. Harold, F. M. 1972. Conservation and transfor-mation of energy by bacterial membranes.Bacteriol. Rev. 36:172-230.

161. Harold, F. M. 1977. Membranes and energytransduction in bacteria. Curr. Top. Bioe-nerg, in press.

162. Harold, F. M., J. R. Baarda, C. Baron, and A.Abrams. 1969. Inhibition of membrane-bound adenosine triphosphatase and cationtransport in Streptococcus faecalis by NN'-dicyclohexylcarbodi-imide. J. Biol. Chem.244:2261-2268.

163. Harris, D. A., J. Rosing, and E. C. Slater.1974. Interaction of ox heart mitochondrialadenosine triphosphatase with nucleotides.Biochem. Soc. Trans. 2:86-87.

164. Hempfling, W. P. 1970. Studies of the efficiencyof oxidative phosphorylation in intact Esche-richia coli. Biochim. Biophys. Acta 205:169-182.

165. Hempfling, W. P., and D. K. Beeman. 1971.Release of glucose repression of oxidativephosphorylation in Escherichia coli B bycyclic adenosine 3',5'-monophosphate. Bio-chem. Biophys. Res. Commun. 45:924-930.

166. Hempfling, W. P., and S. E. Mainzer. 1975.Effects of varying the carbon source limitinggrowth on yield and maintenance character-istics of Escherichia coli in continuous cul-ture. J. Bacteriol. 123:1076-1087.

167. Henderson, P. J. F. 1973. Steady state enzymekinetics with high affinity substrates or in-hibitors; a statistical treatment of dose-re-sponse curves. Biochem. J. 135:101-107.

168. Hendler, R. W., and A. H. Burgess. 1972. Res-piration and protein synthesis in Escherichiacoli membrane-envelope fragments. VI. Sol-ubilization and characterization of the elec-tron transport chain. J. Cell Biol. 55:266-281.

169. Hendler, R. W., and A. H. Burgess. 1974. Frac-tionation of the electron transport chain ofEscherichia coli. Biochim. Biophys. Acta357:215-230.

170. Hendler, R. W., D. W. Towne, and R. I. Shra-ger. 1975. Redox properties of b-type cyto-chromes in Escherichia coli and rat liver mi-tochondria and techniques for their analysis.Biochim. Biophys. Acta 376:42-62.

171. Hertzberg, E. L., and P. C. Hinckle. 1974.Oxidative phosphorylation and proton trans-location in membrane vesicles prepared fromEscherichia coli. Biochem. Biophys. Res.Commun. 58:178-184.

172. Higashi, T., V. K. Kalra, S. H. Lee, E. Bogin,and A. F. Brodie. 1975. Energy-transducingmembrane-bound coupling factor ATPasefrom Mycobacterium phlei. I. Purification,homogeneity and properties. J. Biol. Chem.250:6541-6548.

173. Harrison, D. E. F., and J. E. Loveless. 1971.

BACTERIOL. REV.


http://mm

br.asm.org/

Dow

nloaded from


VOL. 41, 1977

The effect of growth conditions on respira-tory activity and growth efficiency in facul-tative anaerobes grown in chemostat cul-ture. J. Gen. Microbiol. 68:35-43.

174. Hill, S., J. W. Drozd, and J. R. Postgate. 1972.Environmental effects on the growth of ni-trogen-fixing bacteria. J. Appl. Chem. Bio-technol. 22:541-558.

175. Hinckle, P. C., and L. L. Horstman. 1971.Respiration-driven proton transport in sub-mitochondrial particles. J. Biol. Chem.246:6024-6028.

176. Hinckle, P., and P. Mitchell. 1970. Effect ofmembrane potential on equilibrium poise be-tween cytochrome a and cytochrome c in ratliver mitochondria. J. Bioenerg. 1:45-60.

177. Hirsch, C. A., M. Rasminsky, B. D. Davis, andE. C. C. Lin. 1963. A fumarate reductase inEscherichia coli distinct from succinate de-hydrogenase. J. Biol. Chem. 238:3770-3774.

178. Hong, J. S., and H. R. Kaback. 1972. Mutantsof Salmonella typhimurium and Escherichiacoli pleiotropically defective in active trans-port. Proc. Natl. Acad. Sci. U.S.A 69:3336-3340.

179. Hono, T., and M. D. Kamen. 1970. Bacterialcytochromes. II. Functional aspects. Annu.Rev. Microbiol. 24:399-428.

180. Horstmann, L. L., and E. Racker. 1970. Par-tial resolution ofthe enzymes catalysing oxi-dative phosphorylation. XXII. Interactionbetween mitochondrial ATPase inhibitorand mitochondrial ATPase. J. Biol. Chem.245:1336-1344.

181. Houghton, R. L., R. J. Fisher, and D. R. San-adi. 1975. Energy linked and energy-inde-pendent transhydrogenase activities inEscherichia coli vesicles. Biochim. Biophys.Acta 396:17-23.

182. Imai, K., A. Asano, and R. Sato. 1967. Oxida-tive phosphorylation in Micrococcus denitri-ficans. I. Preparation and properties of phos-phorylating membrane fragments. Biochim.Biophys. Act 143:462-476.

183. Imai, K., A. Asano, and R. Sato. 1968. Oxida-tive phosphorylation in Micrococcus denitri-ficans. IV. Further characterization of elec-tron transfer pathway and phosphorylationactivity in NADH oxidation. J. Biochem.(Tokyo) 63:207-218.

184. Imai, K., A. Asano, and R. Sato. 1968. Oxida-tive phosphorylation in Micrococcus denitri-ficans. V. Effects of iron deficiency on respi-ratory components and oxidative phospho-rylation. J. Biochem. (Tokyo) 63:219-225.

185. Ishida, M., and S. Mizushima. 1969. Mem-brane ATPase of Bacillus megaterium. I.Properties ofmembrane ATPase and its solu-bilised form. J. Biochem. (Tokyo) 66:33-43.

186. Itagaki, E., and L. P. Hager. 1966. Studies oncytochrome bm ofEscherichia coli. I. Purifi-cation and crystallization of cytochrome b.2-J. Biol. Chem. 241:3687-3695.

187. Itagaki, E., and L. P. Hager. 1968. The aminoacid sequence of cytochrome b52 of Esche-richia coli. Biochem. Biophys. Res. Commun.


32:1013-1019.188. Jackson, F. A., P. J. Senior, and E. A. Dawes.

1976. Regulation of the tricarboxylic acid cy-cle in Azotobacter beijerinckii grown with ox-ygen limitation. Proc. Soc. Gen. Microbiol.3:85.

189. Jeacocke, R. E., D. F. Niven, and W. A. Ham-ilton. 1972. The protonmotive force in Staph-ylococcus aureus. Biochem. J. 127:57P.

190. John, P., and W. A. Hamilton. 1970. Respira-tory control in membrane particles from Mi-crococcus denitrificans. FEBS Lett. 10:246-248.

191. John, P., and W. A. Hamilton. 1971. Release ofrespiratory control in particles from Micro-coccus denitriffcans by ion-translocating an-tibiotics. Eur. J. Biochem. 23:528-532.

192. John, P., and F. R. Whatley. 1970. Oxidativephosphorylation coupled to oxygen uptakeand nitrate reduction in Micrococcus denitri-ficans. Biochim. Biophys. Acta 216:342-352.

193. John, P., and F. R. Whatley. 1975. Paracoccusdenitrificans and the evolutionary origin ofthe mitochondrion. (London) Nature 254:495-498.

194. Jones, C. W. 1973. The inhibition ofAzotobac-ter vinelandii terminal oxidases by cyanide.FEBS Lett. 36:347-350.

195. Jones, C. W., B. A. C. Ackrell, and S. K. Er-ickson. 1971. Respiratory control in Azoto-bacter vinelandii membranes. Biochim. Bio-phys. Acta 245:54-62.

196. Jones, C. W., B. A. C. Ackrell, and S. K. Er-ickson. 1971. Some parameters affecting res-piratory control in Azotobacter vinelandiimembranes. FEBS Lett. 13:33-35.

197. Jones, C. W., J. M. Brice, A. J. Downs, and J.W. Drozd. 1975. Bacterial respiration-linkedproton translocation and its relationship torespiratory-chain composition. Eur. J. Bio-chem. 52:265-271.

198. Jones, C. W., J. M. Brice, V. Wright, and B. A.C. Ackrell. 1973. Respiratory protection ofnitrogenase in Azotobacter vinelandii. FEBSLett. 29:77-81.

199. Jones, C. W., and D. J. Meyer. 1977. The dis-tribution of cytochromes in bacteria. InHandbook of microbiology. C. R. C. Press,Cleveland, in press.

200. Jones, C. W., and E. R. Redfearn. 1966. Elec-tron transport in Azotobacter vinelandii.Biochim. Biophys. Acta 113:467-481.

201. Jones, C. W., and E. R. Redfearn. 1967. Thecytochrome system ofAzotobacter vinelandii.Biochim. Biophys. Acta 143:340-353.

202. Jones, C. W., and E. R. Redfearn. 1967. Prepa-ration of red and green electron transportparticles from Azotobacter vinelandii.Biochim. Biophys. Acta 143:354-362.

203. Jurtshuk, P., A. J. Bednarz, P. Zey, and C. H.Denton. 1969. L-Malate oxidation by the elec-tron transport fraction of Azotobacter vine-landii. J. Bacteriol. 98:1120-1127.

204. Jurtshuk, P., and L. Harper. 1968. OxidationofD(-) lactate by the electron transport frac-tion of Azotobacter vinelandii. J. Bacteriol.


http://mm

br.asm.org/

Dow

nloaded from



96:678-686.205. Jurtshuk, P., A. K. May, L. M. Pope, and P. R.

Aston. 1969. Comparative studies on succi-nate and terminal oxidase activity in micro-bial and mammalian electron transport sys-tems. Can. J. Microbiol. 15:797-807.

206. Jurtshuk, P., T. J. Mueller, and W. C. Acord.1975. Bacterial terminal oxidases. C. R. C.Crit. Rev. Microbiol. 3:399-468.

207. Jurtshuk, P., and L. Old. 1968. Cytochrome coxidation by the electron transport fractionof Azotobacter vinelandii. J. Bacteriol.95:1790-1797.

208. Kaback, H. R. 1971. Bacterial membranes.Methods Enzymol. 22:99-120.

209. Kaback, H. R. 1972. Transport across isolatedbacterial cytoplasmic membranes. Biochim.Biophys. Acta 265:367416.

210. Kagawa, Y., and E. Racker. 1966. Partial reso-lution of the enzymes catalysing oxidativephosphorylation. IX. Reconstruction of oligo-mycin-sensitive adenosine triphosphatase.J. Biol. Chem. 241:2467-2474.

211. Kamen, M. D., and T. Horio. 1970. Bacterialcytochromes. I. Structural aspects. Annu.Rev. Biochem. 39:673-700.

212. Kanner, B. I., and D. L. Gutnick. 1972. Use ofneomycin in the isolation ofmutants blockedin energy conservation in Escherichia coli. J.Bacteriol. 111:287-289.

213. Kanner, B. I., N. Nelson, and D. L. Gutnick.1975. Differentiation between mutants ofEscherichia coli K12 defective in oxidativephosphorylation. Biochim. Biophys. Acta396:347-359.

214. Kauffman, H. F., and B. F. Van Gelder. 1973.The respiratory chain ofAzotobacter vinelan-dii. I. Spectral properties of cytochrome d.Biochim. Biophys. Acta 305:260-267.

'215. Kauffman, H. F., and B. F. Van Gelder. 1973.The respiratory chain ofAzotobacter vinelan-dii. II. The effect ofcyanide on cytochrome d.Biochim. Biophys. Acta 314:276-283.

K( 216Kauffman, H. F., and B. F. Van Gelder. 1974.The respiratory chain ofAzotobacter vinelan-dii. III. The effect of cyanide in the presenceof substrates. Biochim. Biophys. Acta333:218-227.

217. Kemp, M. B., B. A. Haddock, and P. B. Gar-land. 1975. Synthesis and sidedness of mem-brane-bound respiratory nitrate reductase(EC 1.7.99.4) inEscherichia coli lacking cyto-chromes. Biochem. J. 148:329-333.

218. Kim, I. C., and P. D. Bragg. 1971. Properties ofnonheme iron in a cell envelope fraction fromEscherichia coli. J. Bacteriol. 107:664-670.

219. Kistler, W. S., and E. C. C. Lin. 1971. Anaero-bic L-a-glycerophosphate dehydrogenase ofEscherichia coli: its genetic locus and itsphysiological role. J. Bacteriol. 108:1224-1234.

220. Kistler, W. S., and E. C. C. Lin. 1972. Purifica-tion and properties of the flavin-stimulatedanaerobic L-a-glycerophosphate dehydrogen-ase ofEscherichia coli. J. Bacteriol. 112:539-547.

221. Klein, W. L., and P. D. Boyer. 1972. Energisa-tion of active transport by Escherichia coli.J. Biol. Chem. 247:7257-7265.

222. Klingenberg, M. 1970. Metabolite transport inmitochondria; an example for intracellularmembrane function. Essays Biochem. 6:119-159.

223. Knobloch, K., M. Ishaque, and M. I. H. Aleem.1971. Oxidative phosphorylation in Micrococ-cus denitrificans under autrotrophic growthconditions. Arch. Microbiol. 76:114-124.

224. Knowles, A. F., and H. S. Penefsky. 1972. Thesubunit structure of beef heart mitochon-drial ATPase; physical and chemical proper-ties of isolated subunits. J. Biol. Chem.247:6624-6630.

225. Knowles, C. J., and L. Smith. 1970. Measure-ments of ATP level of intact Azotobacter vi-nelandii under different conditions.Biochim. Biophys. Acta 197:152-160.

226. Kobayashi, H., and Y. Anraku. 1972. Mem-brane-bound adenosine triphosphatase ofEscherichia coli. I. Partial purification andproperties. J. Biochem. (Tokyo) 71:387-399.

227. Kobayashi, H., and Y. Anraku. 1974. Mem-brane bound ATPase of Escherichia coli. II.Physicochemical properties ofthe enzyme. J.Biochem. (Tokyo) 76:1175-1182.

228. Kohn, L. D., and H. R. Kaback. 1973. Mecha-nism of active transport in isolated bacterialmembrane vesicles. XV. Purification andproperties of the membrane bound D-lactatedehydrogenase from Escherichia coli. J. Biol.Chem. 248:7012-7017.

229. Kozlov, I. A., and H. N. Mikelsaar. 1974. Onthe subunit structure of soluble mitochon-drial ATPase. FEBS Lett. 43:212-214.

230. Konings, W. N. 1975. Localisation of mem-brane proteins in membrane vesicles of Ba-cillus subtilis. Arch. Biochem. Biophys.167:570-580.

231. Konings, W. N., A. Bisschop, M. Veenhuis,and C. A. Vermeulen. 1973. New procedurefor the isolation of membrane vesicles ofBa-cillus subtilis and an electron microscopystudy of their ultrastructure. J. Bacteriol.116:1456-1465.

232. Konings, W. N., and J. Boonstra. 1977. Anaer-obic electron transfer and active transport inbacteria. Curr. Top. Membr. Transp., inpress.

233. Kornberg, H. L., J. F. Collins, and D. Bigley.1960. The influence of growth substrates onmetabolic pathways in Micrococcus denitrifi-cans. Biochim. Biophys. Acta 39:9-24.

234. Lam, Y., and D. J. D. Nicholas. 1969. Aerobicand anaerobic respiration in Micrococcusdenitrificans. Biochim. Biophys. Acta 172:450-461.

235. Lam, Y., and D. J. D. Nicholas. 1969. A nitratereductase from Micrococcus denitrificans.Biochim. Biophys. Acta 178:225-234.

236. Lam, Y., and D. J. D. Nicholas. 1969. A nitratereductase with cytochrome oxidase activityfrom Micrococcus denitrificans. Biochim.Biophys. Acta 180:459-472.

BACTERIOL. REv.


http://mm

br.asm.org/

Dow

nloaded from



237. Lambeth, D. O., and H. A. Lardy. 1971. Purifi-cation and properties of rat liver mitochon-drial.ATPase. Eur. J. Biochem. 22:355-363.

238. Langman, L., I. G. Young, G. E. Frost, H.Rosenberg, and F. Gibson. 1972. Entero-chelin system of iron transport in Esche-richia coli: mutations affecting ferric-entero-chelin esterase. J. Bacteriol. 112:1142-1149.

239. Lastras, M., and E. Munoz. 1974. Membraneadenosine triphosphatase of Micrococcus ly-sodeikticus: effect ofmillimolar Mg2+ in mod-ulating the properties of the membrane-bound enzyme. J. Bacteriol. 119:593-601.

240. Lawford, H. G., J. C. Cox, P. B. Garland, andB. A. Haddock. 1976. Electron transport inaerobically grown Paracoccus denitrificans:kinetic characterization of the membrane-bound cytochromes and the stoicheiometry ofrespiration-driven proton translocation.FEBS Lett. 64:369-374.

241. Lawford, H. G., and B. A. Haddock. 1973.Respiration-driven proton translocation inEscherichia coli. Biochem. J. 136:217-220.

242. Lees, H., and J. R. Postgate. 1973. The behav-iour of Azotobacter chrococcum in oxygen-and phosphate-limited chemostat culture. J.Gen. Microbiol. 75:161-166.

243. Lemberg, R., and J. Barrett. 1973. Bacterialcytochromes and cytochrome oxidases, p.217-326. In Cytochromes. Academic PressInc., New York and London.

244. Lester, R. L., and J. A. De Moss. 1971. Effectsof molybdate and selenite on formate andnitrate metabolism in Escherichia coli. J.Bacteriol. 105:1006-1014.

245. Lien, S., R. J. Berzborn, and E. Racker. 1972.Partial resolution of the enzymes catalysingphotophosphorylation. IX. Studies on thesubunit structure of coupling factor 1 fromchloroplast. J. Biol. Chem. 247:3520-3524.

246. Lien, S., and E. Racker. 1971. Preparation andassay of chloroplast coupling factor CF,.Methods Enzymol. 23:547-555.

247. Lo, T. C. Y., M. K. Rayman, and B. D. San-wal. 1972. Transport of succinate in Esche-richia coli. I. Biochemical and genetic stud-ies oftransport in whole cells. J. Biol. Chem.247:6323-6331.

248. Luke, R. K. J., and F. Gibson. 1971. Locationof three genes concerned with the conversionof 2,3-dihydroxybenzoate into enterochelinin Escherichia coli K-12. J. Bacteriol.107:557-562.

249. MacGregor, C. H. 1975. Synthesis of nitratereductase components in chlorate resistantmutants of Escherichia coli. J. Bacteriol.121:1117-1121.

250. MacGregor, C. H. 1975. Solubilization ofEsch-erichia coli nitrate reductase by a mem-brane-bound protease. J. Bacteriol.121:1102-1110.

251. MacGregor, C. H. 1975. Anaerobic cytochromeb, in Escherichia coli: association with andregulation of nitrate reductase. J. Bacteriol.121:1111-1116.

252. MacGregor, C. H. 1976. Biosfnthesis of mem-

brane-bound nitrate reductase in Escherichiacoli: evidence for a soluble precursor. J. Bac-teriol. 126:122-131.

253. MacGregor, C. H., and C. A. Schnaitman.1971. Alterations in the cytoplasmic mem-brane proteins of various chlorate-resistantmutants of Escherichia coli. J. Bacteriol.108:564-570.

254. MacGregor, C. H., and C. A. Schnaitman.1973. Reconstitution of nitrate reductase ac-tivity and formation of membrane particlesfrom cytoplasmic extracts of chlorate-resist-ant mutants ofEscherichia coli. J. Bacteriol.114:1164-1176.

255. MacGregor, C. H., C. A. Schnaitman, D. E.Normansell, and M. G. Hodgins. 1974. Puri-fication and properties of nitrate reductasefrom Escherichia coli K12. J. Biol. Chem.249:5321-5327.

256. Marcot, J., and E. Azoulay. 1971. Obtention etetude de doubles mutants chlorate-r6sistantschez Escherichia coli K12. FEBS Lett.13:137-139.

257. McConville, M., and H. P. Charles. 1975. Isola-tion of 'haemin permeable' mutants andtheir use in the study ofthe genetics ofhaembiosynthesis in Escherichia coli K12. Proc.Soc. Gen. Microbiol. 3:14-15.

258. Mevel-Ninio, M. T., and R. C. Valentine. 1975.Energy requirement for biosynthesis ofDNAin Escherichia coli; role of membrane-boundenergy-transducing ATPase (coupling fac-tor). Biochim. Biophys. Acta 376:485-491.

259. Mevel-Ninio, M., and T. Yamamoto. 1974.Conversion of active transport vesicles ofEscherichia coli into oxidative phosphoryla-tion vesicles. Biochim. Biophys. Acta357:63-66.

260. Meyer, D. J., and C. W. Jones. 1973. Reactivitywith oxygen ofbacterial cytochrome oxidasesa,, aa3 and o. FEBS Lett. 33:101-105.

261. Meyer, D. J., and C. W. Jones. 1973. Distribu-tion of cytochromes in bacteria: relationshipto general physiology. Int. J. Syst. Bacteriol.23:459-467.

262. Mickelson, M. N. 1969. Phosphorylation andthe reduced nicotinamide adenine dinucleo-tide oxidase reaction in Streptococcus agalac-tiae. J. Bacteriol. 100:895-901.

263. Miki, K., and E. C. C. Lin. 1973. Enzyme com-plex which couples glycerol-3-phosphate de-hydrogenation to fumarate reduction inEscherichia coli. J. Bacteriol. 114:767-771.

264. Miki, K., and E. C. C. Lin. 1973. The couplingof glycerol-3-phosphate dehydrogenation tofumarate reduction in E. coli: an energyyielding reaction. Fed. Proc. Am. Soc. Exp.Biol. 32:632.

265. Miki, K., and E. C. C. Lin. 1975. Electrontransport chain from glycerol-3-phosphate tonitrate in Escherichia coli. J. Bacteriol.124:1288-1294.

266. Miki, K., and E. C. C. Lin. 1975. Anaerobicenergy yielding reaction associated withtranshydrogenation from glycerol-3-phos-phate to fumarate by an Escherichia coli sys-

VOL. 41, 1977


http://mm

br.asm.org/

Dow

nloaded from



tem. J. Bacteriol. 124:1282-1287.267. Mirsky, R., and V. Barlow. 1971. Purification

and properties of the ATPase from the cyto-plasmic membrane of Bacillus megateriumKM. Biochim. Biophys. Acta 241:835-845.

268. Mirsky, R., and V. Barlow. 1972. Alternativepurification of the membrane-bound ATPasefrom Bacillus megaterium KM, and someproperties. Biochim. Biophys. Acta 274:556-562.

269. Mirsky, R., and V. Barlow. 1973. Molecularweight, amino acid composition and otherproperties of membrane-bound ATPase fromBacillus megaterium KM. Biochim. Biophys.Acta 291:480-488.

270. Mitchell, P. 1966. Chemiosmotic coupling inoxidative and photosynthetic phosphoryla-tion. Biol. Rev. Cambridge Philos. Soc.41:445-502.

271. Mitchell, P. 1967. Proton-translocation phos-phorylation in mitochondria, chloroplastsand bacteria: natural fuel cells and solarcells. Fed. Proc. 26:1370-1379.

272. Mitchell, P. 1968. Chemiosmotic coupling andenergy transduction. Glynn Research Ltd.,Bodmin, United Kingdom.

273. Mitchell, P. 1973. Cation-translocating adeno-sine triphosphatase models: how direct is theparticipation of adenosine triphosphate andits hydrolysis products in cation transloca-tion? FEBS Lett. 33:267-274.

274. Mitchell, P. 1974. A chemiosmotic molecularmechanism for proton-translocating adeno-sine triphosphatases. FEBS Lett. 43:189-194.

275. Mitchell, P. 1975. The proton motive Q cycle: ageneral formulation. FEBS Lett. 59:137-139.

276. Mitchell, P., and J. Moyle. 1968. Proton trans-location coupled to ATP hydrolysis in ratliver mitochondria. Eur. J. Biochem. 4:530-539.

277. Mitchell, P., and J. Moyle. 1971. Activationand inhibition of mitochondrial adenosinetriphosphatase by various anions and otheragents. J. Bioenerg. 2:1-11.

278. Monteil, H., J. Schoun, and M. Guinard. 1974.A Na+K+-activated, Mg2+-dependent ATP-ase released from Proteus L-form membrane.Eur. J. Biochem. 41:525-532.

279. Morris, J. G. 1975. The physiology of obligateanaerobiosis. Adv. Microbial Physiol. 12:169-246.

280. Moyle, J., and P. Mitchell. 1973. The proton-translocating nicotinamide-adenine dinu-cleotide (phosphate) transhydrogenase of ratliver mitochondria. Biochem. J. 132:571-585.

281. Moyle, J., and P. Mitchell. 1973. Proton trans-location quotient for the adenosine triphos-phatase of rat liver mitochondria. FEBSLett. 30:317-320.

282. Mueller, T. J., and P. Jurtshuk. 1972. Solubi-lisation of cytochrome oxidase from Azoto-bacter vinelandii. Fed. Proc. 31:3825.

283. Munoz, E., J. H. Freer, D. J. Ellar, and M. J.R. Salton. 1968. Membrane-associated ATP-ase activity from Micrococcus lysodeikticus.

Biochim. Biophys. Acta 150:531-533.284. Munoz, E., M. S. Nachbar, M. T. Schor, and

M. J. R. Salton. 1968. Adenosine triphospha-tase of Micrococcus lysodeikticus; selectiverelease and relationship to membrane struc-ture. Biochem. Biophys. Res. Commun.32:539-546.

285. Munoz, E., M. J. R. Salton, M. H. Ng, and M.T. Schor. 1969. Membrane ATPase of Micro-coccus lysodeikticus; purification, propertiesof the "soluble" enzyme and properties of themembrane-bound enzyme. Eur. J. Biochem.7:490-501.

286. Murphy, M. J., L. M. Siegel, H. Kamin, and D.Rosenthal. 1973. Reduced nicotinamide ade-nine dinucleotide phosphate-sulfite reduc-tase of Enterobacteria. II. Identification of anew class of heme prosthetic group: an iron-tetrahydroporphyrin (isobacteriochlorin type)with eight carboxylic acid groups. J. Biol.Chem. 248:2801-2814.

287. Murphy, M. J., L. M. Siegel, S. R. Tove, andH. Kamin. 1974. Siroheme: a new prostheticgroup participating in six-electron reductionreactions catalyzed by both sulfite and ni-trate reductase. Proc. Natl. Acad. U.S.A.71:612-616.

288. Mutaftschiev, S., and E. Azoulay. 1973. Mem-brane reconstitution in chl-r mutants ofEscherichia coli K12. VI. - Morphologicalstudy of membrane assembly during comple-mentation between extracts of chl-r mu-tants. Biochim. Biophys. Acta 307:525-540.

289. Najai, S., and S. Aiba. 1972. Reassessment ofmaintenance and energy uncoupling in thegrowth ofAzotobacter vinelandii. J. Gen. Mi-crobiol. 73:531-538.

290. Neijssel, O. M., and D. W. Tempest. 1975. Theregulation of carbohydrate metabolism inKlebsiella aerogenes NCTC 418 organismsgrowing in chemostat culture. Arch. Micro-biol. 106:251-258.

291. Neijssel, 0. M., and D. W. Tempest. 1976.Bioenergetic aspects of aerobic growth ofKlebsiella aerogenes NCTC 418 in carbon-limited and carbon-sufflicient chemostat cul-ture. Arch. Microbiol. 107:215-221.

292. Nelson, N. D., W. Deters, H. Nelson, and E.Racker. 1973. Partial resolution of the en-zymes catalysing photophosphorylation; XIIIProperties of isolated subunits of couplingfactor 1 from spinach chloroplasts. J. Biol.Chem. 248:2049-2055.

293. Nelson, N., A. Kamienietzky, D. W. Deters,and H. Nelson. 1975. Subunit structure andfunction of CF,, p. 149-154. In E. Quagli-ariello, S. Papa, F. Palmieri, E. C. Slater,and N. Siliprandi (ed.), Electron transferchains and oxidative phosphorylation. NorthHolland-American Elsevier, Amsterdam.

294. Nelson, N., B. I. Kanner, and D. L. Gutnick.1974. Purification and properties of Mg2+-Ca2+ ATPase from Escherichia coli. Proc.Natl. Acad. Sci. U.S.A. 71:2720-2724.

295. Nelson, N., H. Nelson, and E. Racker. 1972.

BACTERIOL. REV.


http://mm

br.asm.org/

Dow

nloaded from



Partial resolution of the enzymes catalysingphotophosphorylation. XII. Purification andproperties of an inhibitor isolated from chlo-roplast coupling factor 1. J. Biol. Chem.247:7657-7662.

296. Newton, N. A. 1967. A soluble cytochrome con-taining c-type and a2-type haem groups fromMicrococcus denitrificans. Biochem. J.105:21C-23C.

297. Newton, N. A. 1969. The two haem nitratereductase of Micrococcus denitrificans.Biochim. Biophys. Acta 185:316-331.

298. Newton, N. A., G. B. Cox, and F. Gibson. 1971.The function ofmenaquinone (vitamin K2) inEscherichia coli K-12. Biochim. Biophys.Acta 244:155-166.

299. Newton, N. A., G. B. Cox, and F. Gibson. 1972.Function of ubiquinone in Escherichia coli: amutant strain forming a low level of ubiqui-none. J. Bacteriol. 109:69-73.

300. Nieuwenhuis, F. J. R. M., J. A. M. van derDrift, A. B. Voet, and K. Van Dam. 1974.Evidence for a naturally occurring ATPase-inhibitor in Escherichia coli. Biochim. Bio-phys. Acta 368:461-463.

301. Nieuwenhuis, F. J. R. M., B. I. Kanner, D. L.Gutnick, P. W. Postma, and K. Van Dam.1973. Energy conservation in membranes ofmutants of Escherichia coli defective in oxi-dative phosphorylation. Biochim. Biophys.Acta 325:62-71.

302. Nieuwenhuis, F. J. R. M., A. A. M. Thomas,and K. Van Dam. 1974. Solubilisation bycholate or deoxycholate of a DCCD-sensitiveATPase complex from Escherichia coli. Bio-chem. Soc. Trans. 2:512-513.

303; Nicholas, D. J. D., P. W. Wilson, W. Heinen,G. Palmer, and H. Beinert. 1962. Use ofelectron paramagnetic resonance spectros-copy in investigations of functional metalcomponents in micro-organisms. Nature(London) 196:433-436.

304. Nicholls, D. G. 1974. The influence of respira-tion and ATP hydrolysis on the proton elec-trochemical gradient across the inner mem-brane of rat liver mitochondria as deter-mined by ion distribution. Eur. J. Biochem.50:305-315.

305. Ohki, M. 1972. Correlation between metabo-lism of phosphatidylglycerol and membranesynthesis in Escherichia coli. J. Mol. Biol.68:249-264.

306. Ohki, M., and H. Mitsui. 1974. Defective mem-brane synthesis in an E. coli mutant. Nature(London) 252:64-66.

307. Ohnishi, T. 1973. Mechinism of electron trans-port and energy conservation in the site Iregion of the respiratory chain. Biochim.Biophys. Acta 301:105-128.

308. Oppenheim, J., and L. Marcus. 1970. Correla-tion ofultra structure inAzotobacter vinelan-dii with nitrogen source for growth. J. Bacte-riol. 101:286-291.

309. Oppenheim, J. D., and M. J. R. Salton. 1973.Localisation and distribution of Micrococcus

lysodeikticus membrane ATPase determinedby ferritin labelling. Biochim. Biophys. Acta298:297-322.

310. Palmieri, F., and M. Klingenberg. 1967. Inhi-bition of respiration under the control ofazide uptake by mitochondria. Eur. J. Bio-chem. 1:439-446.

311. Pandya, K. P., and H. K. King. 1966. Ubiqui-none and menaquinone in bacteria: a com-parative study of some bacterial respiratorysystems. Arch. Biochem. Biophys. 114:154-157.

312. Papa, S., F. Guerrieri, L. Rossi-Bernadi, andJ. M. Tager. 1970. Effect of oligomycin onproton translocation in submitochondrialparticles. Biochim. Biophys. Acta 197:100-103.

313. Pedersen, P. L. 1975. Mitochondrial ATPase.J. Bioenerg. 6:243-275.

314. Penefsky, H. S., and R. Warner. 1965. Partialresolution of the enzymes catalysing oxida-tive phosphorylation. VI. Studies on themechanism of cold-inactivation of mitochon-drial adenosine triphosphatase. J. Biol.Chem. 240:4694-4702.

315. Peter, H. W., and J. Ahlers. 1975. Phospholipidrequirements of ATPase of Escherichia coli.Arch. Biochem. Biophys. 170:169-178.

316.. Poole, R. K., and B. A. Haddock. 1974. En-ergy-linked reduction of nicotinamide-ade-nine dinucleotide in membranes derivedfrom normal and various respiratory-defi-cient mutant strains ofEscherichia coli K12.Biochem. J. 144:77-85.

317. Poole, R. K., and B. A. Haddock. 1975. Effectsofsulphate-limited growth in continuous cul-ture on the electron transport chain and en-ergy conservation in Escherichia coli K12.Biochem. J. 152:537-546.

318. Postgate, J. R. 1974. Pre-requisites for biologi-cal nitrogen fixation in free-living hetero-trophic bacteria, p. 663-686. In A. Quispel(ed.), The biology of nitrogen fixation. NorthHolland, Amsterdam.

319. Powell, K. A., R. Cox, M. McConville, and H.P. Charles. 1973. Genetical and biochemicalanalysis of porphyrin biosynthesis in Esche-richia coli. Hoppe-Seyler's Z. Physiol. Chem.354:842-843.

320. Powell, K. A., R. Cox, M. McConville, and H.P. Charles. 1973. Mutations affecting por-phyrin biosynthesis in Escherichia coli. En-zyme 16:65-73.

321. Pudek, M. R., and P. D. Bragg. 1974. Inhibi-tion of the respiratory chain oxidases ofEscherichia coli. Arch. Biochem. Biophys.164:682-693.

322. Pudek, M. R., and P. D. Bragg. 1975. Reactionof cyanide with cytochrome d in respiratoryparticles from exponential phase Escherichiacoli. FEBS Lett. 50:111-113.

323. Pullman, M. E., and G. C. Monroy. 1963. Anaturally occurring inhibitor of mitochon-drial ATPase. J. Biol. Chem. 238:3762-3769.

324. Racker, E. 1967. Resolution and reconstitution

VOL. 41, 1977


http://mm

br.asm.org/

Dow

nloaded from



of the inner mitochondrial membrane. Fed.Proc. 26:1335-1340.

325. Racker, E., and L. L. Horstman. 1967. Partialresolution of the enzymes catalysing oxida-tive phosphorylation. XIII. Structure andfunction of submitochondrial particles com-pletely resolved with respect to coupling fac-tor 1. J. Biol. Chem. 242:2547-2551.

326. Rainnie, D. J., and P. D. Bragg. 1973. Theeffect of iron deficiency on respiration andenergy-coupling in Escherichia coli. J. Gen.Microbiol. 77:339-349.

327. Redwood, W. R., D. C. Gibbes, and T. E.Thompson. 1973. Interaction of a solubilisedmembrane ATPase with lipid bilayer mem-branes. Biochim. Biophys. Acta 31:10-22.

328. Reeves, J. P. 1971. Transient pH changes dur-ing D-lactate oxidation by membrane vesi-cles. Biochem. Biophys. Res. Commun.45:931-936.

329. Riviere, C., G. Giordano, J. Pommier, and E.Azoulay. 1975. Membrane reconstitution inchl-r mutants of Escherichia coli K12. VIII.Purification and properties of the FA factor,the product of the chl B gene. Biochim. Bio-phys. Acta 389:219-235.

330. Roberton, A. M., C. T. Holloway, I. G. Knight,and R. B. Beechey. 1966. The effect of dicy-clohexylcarbodi-imide on reactions involvedin respiratory chain phosphorylation. Bio-chem. J. 100:78P.

331. Robertson, S. N., and N. K. Boardman. 1975.The link between charge separation, protonmovement and ATPase reaction. FEBS Lett.60:1-6.

332. Rockey, A. E., and B. A. Haddock. 1974. Therole of adenosine triphosphate in the recon-stitution of functional cytochromes in mem-branes derived from a 5-aminolaevulinicacid-requiring mutant of Escherichia coli.Biochem. Soc. Trans. 2:957-960.

333. Roisin, M. P., and A. Kepes. 1973. The mem-brane ATPase ofEscherichia coli. II. Releaseinto solution, allotopic properties and recon-stitution of membrane ATPase. Biochim.Biophys. Acta 305:249-259.

334. Rolfe, B., and K. Onodera. 1972. Genes, en-zymes and membrane proteins of the nitraterespiration system of Escherichia coli. J.Membr. Biol. 9:195-207.

335. Rosen, B. P. 1973. Restoration of active trans-port in an Mg2+-adenosine triphosphatase-deficient mutant of Escherichia coli. J. Bac-teriol. 116:1124-1129.

336. Rosen, B. P. 1973. f-Galactoside transport andproton movements in an adenosine triphos-phate-deficient mutant of Escherichia coli.Biochem. Biophys. Res. Commun. 53:1289-1296.

337. Rosen, B. P., and L. W. Adler. 1975. The main-tenance of the energised membrane stateand its relation to active transport in Esche-richia coli. Biochim. Biophys. Acta 387:23-36.

338. Rosenberg, H., G. B. Cox, J. D. Butlin, and S.

BACTERIOL. REV.

J. Gutowski. 1975. Metabolite transport inmutants of Escherichia coli K12 defective inelectron transport and coupled phosphoryla-tion. Biochem. J. 146:417-423.

339. Rottenberg, H. 1975. The measurement oftransmembrane electrochemical proton gra-dients. J. Bioenerg. 7:61-74.

340. Ruiz-Herrera, J., and A. Alvarez. 1972. Aphysiological study of formate dehydrogen-ase, formate oxidase and hydrogen lyasefrom Escherichia coli K-12. Antonie vanLeeuwenhoek; J. Microbiol. Serol. 38:479-491.

341. Ruiz-Herrera, J., and J. A. De Moss. 1969.Nitrate reductase complex ofEscherichia coliK-12; participation of specific formate dehy-drogenase and cytochrome b, components innitrate reduction. J. Bacteriol. 99:720-729.

342. Ruiz-Herrera, J., M. K. Showe, and J. A. DeMoss. 1969. Nitrate reductase complex ofEscherichia coli K-12. Isolation and charac-terization of mutants unable to reduce ni-trate. J. Bacteriol. 97:1291-1297.

343. Salton, M. J. R. 1974. Membrane-associatedenzymes in bacteria. Adv. Microbial Phys-iol. 11:213-283.

344. Salton, M. J. R., and M. T. Schor. 1972. Sub-unit structure and properties of two forms ofATPase released from Micrococcus lysodeik-ticus membranes. Biochem. Biophys. Res.Commun. 49:350-357.

345. Salton, M. J. R., and M. T. Schor. 1974.Release and purification of Micrococcus lyso-deikticus ATPase from membranes extractedwith n-butanol. Biochim. Biophys. Acta345:74-82.

346. Sasarman, A., P. Chartrand, R. Proschek, M.Desrochers, D. Tardif, and C. Lapointe.1975. Uroporphyrin-accumulating mutant ofEscherichia coli K-12. J. Bacteriol. 124:1205-1212.

347. Sasarman, A., M. Surdeanu, and T. Horodni-ceanu. 1968. Locus determining the synthe-sis of 8-aminolaevulinic acid in Escherichiacoli K-12. J. Bacteriol. 96:1882-1884.

348. Sasarman, A., M. Surdeanu, G. Szegli, T. Hor-odniceanu, V. Greceanu, and A. Dumi-trescu. 1968. Hemin-deficient mutants ofEscherichia coli K-12. J. Bacteriol. 96:570-572.

349. Sapshead, L. M., and J. W. T. Wimpenny.1972. The influence of oxygen and nitrate onthe formation of the cytochrome pigments ofthe aerobic and anaerobic respiratory chainof Micrococcus denitrificans. Biochim. Bio-phys. Acta 267:388-397.

350. Schairer, H. U., and B. A. Haddock. 1972. 38-galactoside accumulation in a Mg2+-, Ca2+-activated ATPase deficient mutant ofE. coli.Biochem. Biophys. Res. Commun. 48:544-551.

351. Schatz, G., H. S. Penefsky, and E. Racker.1967. Partial resolution of the enzymes cata-lysing oxidative phosphorylation. XIV. In-teraction of purified mitochondrial ATPase


http://mm

br.asm.org/

Dow

nloaded from


VOL. 41, 1977

from bakers yeast with submitochondrialparticles from beef heart. J. Biol. Chem.242:2552-2560.

352. Schnebli, H. P., and A. Abrams. 1970. Mem-brane ATPase from Streptococcus faecalis;preparation and homogeneity. J. Biol.Chem. 245:1115-1121.

353. Schnebli, H. P., A. E. Vatter, and A. Abrams.1970. Membrane ATPase from Streptococcusfaecalis; molecular weight, subunit structureand amino acid composition. J. Biol. Chem.245:1122-1127.

354. Scholes, P., and P. Mitchell. 1970. Respira-tion-driven proton translocation in Micrococ-cus denitrificans. J. Bioenerg. 1:309-323.

355. Scholes, P., and P. Mitchell. 1970. Acid-basetitration across the plasma membrane ofMi-crococcus denitrifscans; factors affecting theeffective proton conductance and the respira-tory rate. J. Bioenerg. 1:61-72.

356. Scholes, P., P. Mitchell, and J. Moyle. 1969.The polarity of proton translocation in somephotosynthetic microorganisms. Eur. J. Bio-chem. 8:450-454.

357. Scholes, P. B., and L. Smith. 1968. Composi-tion and properties of the membrane-boundrespiratory chain system ofMicrococcus den-itrificans. Biochim. Biophys. Acta 153:363-375.

358. Senior, A. E. 1973. The structure of mitochon-drial ATPase. Biochim. Biophys. Acta301:249-277.

359. Senior, A. E. 1975. Mitochondrial adenosinetriphosphatase. Location of sulphydrylgroups and disulphide bonds in the solubleenzyme from beef heart. Biochemistry14:660-664.

360. Senior, P. J., G. A. Beech, G. A. F. Ritchie,and E. A. Dawes. 1972. The role of oxygenlimitation in the formation of poly-/8-hydrox-ybutyrate during batch and continuous cul-ture of Azotobacter beijerinckii. Biochem. J.128:1193-1201.

361. Senior, A. E., and J. C. Brooks. 1970. Studieson the mitochondrial oligomycin-insensitiveATPase. I. An improved method of purifica-tion and behaviour of the enzyme in solu-tions of various depolymerising agents.Arch. Biochem. Biophys. 140:257-266.

362. Senior, A. E., and J. C. Brooks. 1971. Subunitcomposition ofthe mitochondrial oligomycin-insensitive ATPase. FEBS Lett. 17:327-329.

363. Shipp, W. S. 1972. Cytochromes ofEscherichiacoli. Arch. Biochem. Biophys. 150:459-472.

364. Shipp, W. S. 1972. Absorption bands of multi-ple b and c cytochromes in bacteria detectedby numerical analysis of absorption spectra.Arch. Biochem. Biophys. 150:482-488.

365. Shipp, W. S., M. Piotrowski, and A. E. Fried-man. 1972. Apparent cytochrome gene doseeffects in F-lac and F-gal heterogenotes ofEscherichia coli. Arch. Biochem. Biophys.150:473-481.

366. Short, S. A., H. R. Kaback, T. Hawkins, andL. D. Kohn. 1975. Immunochemical proper-

BACTERIAL RESPIRATION

ties of the membrane-bound-D-lactate dehy-drogenase from Escherichia coli. J. Biol.Chem. 250:4285-4290.

367. Showe, M. K., and J. A. De Mos. 1968. Locali-zation and regulation of synthesis of nitratereductase in Escherichia coli. J. Bacteriol.95:1305-1313.

368. Siegel, L. M., and P. S. Davis. 1974. Reducednicotinamide adenine dinucleotide phos-phate-sulfite reductase ofEnterobacteria. IV.The Escherichia coli hemoflavoprotein; sub-unit structure and dissociation into hemo-protein and flavoprotein components. J. Biol.Chem. 249:1587-1598.

369. Siegel, L. M., P. S. Davis, and H. Kamin. 1974.Reduced nicotinamide adenine dinucleotidephosphate-sulfite reductase of Enterobac-teria. m. The Escherichia coli hemoflavopro-tein: catalytic parameters and the sequenceof electron flow. J. Biol. Chem. 249:1572-1586.

370. Siegel, L. M., M. J. Murphy, and H. Kamin.1973. Reduced nicotinamide adenine dinu-cleotide phosphate-sulfite reductase ofEnter-bacteria. I. The Escherichia coli hemoflavo-protein: molecular parameters and pros-thetic groups. J. Biol. Chem. 248:251-264.

371. Simoni, R. D. and P. W. Postma. 1975. Theenergetics of bacterial active transport.Annu. Rev. Biochem. 44:523-554.

372. Simoni, R. D., and M. K. Shallenberger. 1972.Coupling of energy to active transport ofamino acids in Escherichia coli. Proc. Natl.Acad. Sci. U.S.A. 69:2663-2667.

373. Simoni, R. D., and A. Shandell. 1975. Energytransduction in Escherichia coli; genetic al-teration of a membrane polypeptide of the(Ca2+, Mg2+)-ATPase complex. J. Biol.Chem. 250:9421-9427.

374. Singh, A. P., and P. D. Bragg. 1975. Reducednicotinamide adenine dinucleotide depend-ent reduction of fumarate coupled to mem-brane energization in a cytochrome-deficientmutant of Escherichia coli K12. Biochim.Biophys. Acta 396:229-241.

375. Singh, A. P., and P. D. Bragg. 1976. Anaerobictransport of amino acids coupled to the glyc-erol-3-phosphate-fumarate oxidoreductasesystem in a cytochrome-deficient mutant ofEscherichia coli. Biochim. Biophys. Acta423:450-461.

376. Skulachev, V. P. 1971. Energy transformationin the respiratory chain. Curr. Top. Bioe-nerg. 4:127-190.

377. Slater, E. C. 1974. Electron transfer and en-ergy conservation. Biochim. Biophys. ActaLibrary 13:1-20.

% 378. Smith, J. B., and P. C. Sternweis. 1975. Resto-ration of coupling factor activity to Esche-richia coli ATPase missing the delta subunit.Biochem. Biophys. Res. Commun. 62:764-771.

379. Smith, L. 1968. The respiratory chain system ofbacteria, p. 55-122. In T. P. Singer (ed.),Biological oxidations. Interscience Publish-

97


http://mm

br.asm.org/

Dow

nloaded from



ers Inc., New York.380. Sone, N., M. Yoshida, H. Hirata, and Y. Ka-

gawa. 1975. Purification and properties ofdicyclohexylcarbodiimide-sensitive adeno-sine triphosphatase from a thermophilic bac-terium. J. Biol. Chem. 250:7917-7923.

381. Spencer, M. E., and J. R. Guest. 1973. Isolationand properties of fumarate reductase mu-tants of Escherichia coli. J. Bacteriol.114:563-570.

382. Spencer, M. E., and J. R. Guest. 1974. Proteinsof the inner membrane of Escherichia coli:identification of succinate dehydrogenase bypolyacrylamide gel electrophoresis with sdhamber mutants. J. Bacteriol. 117:947-953.

383. Stouthamer, A. H. 1976. Biochemistry and ge-netics of nitrate reductase in bacteria. Adv.Microbial Physiol. 14:315-375.

384. Stouthamer, A. H., and C. Bettenhaussen.1973. Utilization of energy for growth andmaintenance in continuous and batch cul-tures of microorganisms. A reevaluation ofthe method for the determination of ATPproduction by measuring molar growthyields. Biochim. Biophys. Acta 301:53-70.

385. Stouthamer, A. H., and C. Bettenhaussen.1975. Determination of the efficiency of oxi-dative phosphorylation in continuous cul-tures of Aerobacter aerogenes. Arch. Micro-biol. 102:187-192.

386. Swank, R. T., and R. H. Burms. 1969. Purifica-tion and properties of cytochromes c ofAzoto-bacter vinelandii. Biochim. Biophys. Acta180:473-489.

387. Swank, R. T., and R. H. Burris. 1969. Restora-tion by ubiquinone ofAzotobacter vinelandiireduced nicotinamide adenine dinucleotideoxidase activity. J. Bacteriol. 98:311-313.

388. Sweetman, A. J., and D. E. Griffiths. 1971.Studies on energy-linked reactions. Energy-linked reduction of oxidized nicotinamide-adenine dinucleotide by succinate in Esche-richia coli. Biochem. J. 121:117-124.

389. Sweetman, A. J., and D. E. Griffiths. 1971.Studies on energy-linked reactions. Energy-linked transhydrogenase reactions in Esche-richia coli. Biochem. J. 121:125-130.

390. Taniguchi, S., and E. Itagaki. 1960. Nitratereductase of nitrate respiration type from E.coli. I. Solubilization and purification fromthe particulate system with molecular char-acterization as a metallo-protein. Biochim.Biophys. Acta 41:263-279.

391. Thayer, W. S., and P. C. Hinkle. 1973. Stoichi-ometry of adenosine triphosphate-drivenproton translocation in bovine heart submi-tochondrial particles. J. Biol. Chem.248:5395-5402.

392. Thipayathasana, P. 1975. Isolation and proper-ties ofEscherichia coli ATPase mutants withaltered divalent metal specificity for ATPhydrolysis. Biochim. Biophys. Acta 408:47-57.

393. Thipayathasana, P., and R. C. Valentine.1974. The requirement for energy-transduc-

ing ATPase for anaerobic motility in Esche-richia coli. Biochim. Biophys. Acta 347:464-468.

394. Tikhonova, G. V. 1974. Some properties ofApm + generators in Micrococcus lysodeikti-cus membranes. Biochem. Soc. Spec. Publ.4:131-143.

395. Tissieres, A., and R. H. Burris. 1956. Purifica-tion and properties of cytochromes C4 and c5from Azotobacter vinelandii. Biochim. Bio-phys. Acta 20:436-437.

396. Tucker, A. N., and T. T. Lillich. 1974. Effect ofthe systemic fungicide carboxin on electrontransport function in membranes of Micro-coccus denitrificans. Antimicrob. AgentsChemother. 6:572-578.

397. Tzagoloff, A., and P. Meagher. 1971. Assemblyof the mitochondrial membrane system. V.Properties of a dispersed preparation of therutamycin-sensitive ATPase of yeast mito-chondria. J. Biol. Chem. 246:7328-7336.

398. Tzagoloff, A., and P. Meagher. 1972. Assemblyof the mitochondrial membrane system. VI.Mitochondrial synthesis of subunit proteinsof the rutamycin-sensitive adenosine tri-phosphatase. J. Biol. Chem. 247:594-603.

399. van den Broek, H. W. J., J. S. Santema, J. H.Wassink, and C. Veeger. 1971. Pyridine nu-cleotide transhydrogenase. I. Isolation, puri-fication and characterisation of the transhy-drogenase from Azotobacter vinelandii. Eur.J. Biochem. 24:31-45.

400. Van der Beek, E. G., and A. H. Stouthamer.1973. Oxidative phosphorylation in intactbacteria. Arch. Mikrobiol. 89:327-339.

401. Van de Stadt, R. J., F. J. R. M. Niewenhuis,and K. Van Dam. 1971. On the reversibilityof the energy-linked transhydrogenase.Biochim. Biophys. Acta 234:173-176.

402. Van Thienen, G., and P. W. Postma. 1973.Coupling between energy conservation andactive transport of serine in Escherichia coli.Biochim. Biophys. Acta 323:429-440.

403. Van Verseveld, H. W., and A. H. Stouthamer.1976. Oxidative phosphorylation in Micrococ-cus denitrificans. Calculation of the P/O ra-tio in growing cells. Arch. Microbiol.107:241-247.

404. Venables, W. A. 1972. Genetic studies withnitrate reductase-less mutants of Esche-richia coli. I. Fine structure analysis of thenar A, nar B and nar E loci. Mol. Gen.Genet. 114:223-231.

405. Vernon, L. P. 1956. Bacterial cytochromes. I.Cytochrome composition of Micrococcus den-itrificans and Pseudomonas denitrificans. J.Biol. Chem. 222:1035-1044.

406. Vernon, L. P., and F. G. White. 1957. Termi-nal oxidases of Micrococcus denitrificans.Biochim. Biophys. Acta 25:321-328.

407. Villarreal-Moguel, E. I., V. Ibarra, J. Ruiz-Herrera, and C. Gitler. 1973. Resolution ofthe nitrate reductase complex from the mem-brane Escherichia coli. J. Bacteriol.113:1264-1267.

BACTERIOL. REV.-


http://mm

br.asm.org/

Dow

nloaded from



408. Vogel, G., and R. Steinhart. 1976. ATPase ofEscherichia coli: purification, dissociationand reconstitution ofthe active complex fromisolated subunits. Biochemistry 15:208-216.

409. Weibull, C., J. W. Greenawald, and H. Low.1962. The hydrolysis of adenosine triphos-phate by cell fractions of Bacillus megate-rium. I. Localisation and general character-istics of the enzyme activities. J. Biol. Chem.237:847-852.

410. West, I. C., and P. Mitchell. 1974. The proton-translocating adenosine triphosphatase ofEscherichia coli. FEBS Lett. 40:1-4.

411. White, D. C., and P. R. Sinclair. 1971.Branched electron transport systems in bac-teria. Adv. Microbial. Physiol. 5:173-211.

412. White, D. C., A. N. Tucker, and H. R. Kaback.1974. Relationship between amino acidtransport and electron transport by mem-brane vesicles of Micrococcus denitriflcans.Arch. Biochem. Biophys. 165:672-680.

413. Wikstrom, M. K. F. 1973. The different cyto-chrome b components in the respiratorychain of animal mitochondria and their rolein electron transport and energy conserva-tion. Biochim. Biophys. Acta 301:155-193.

414. Wilkinson, B. J., M. R. Morman, and D. C.White. 1972. Phospholipid composition andmetabolism of Micrococcus denitrificans. J.Bacteriol. 112:1288-1294.

415. Williams, R. J. P. 1969. Electron transfer andenergy conservation. Curr. Top. Bioenerg.3:79-156.

416. Williams, R. J. P. 1970. Electron transfer, con-formation changes and energy conservation,p. 7-23. In M. J. Tager, S. Papa, E. Quagli-ariello, and E. C. Slater (ed.), Electrontransport and energy conservation. Adria-tica Editrice, Bari.

417. Williams, R. J. P. 1975. Proton-driven phos-phorylation reactions in mitochondrial andchloroplast membranes. FEBS Lett. 53:123-125.

418. Wimpenny, J. W. T., and J. A. Cole. 1967. Theregulation of metabolism in facultative bac-teria. III. The effect of nitrate. Biochim. Bio-

phys. Acta 148:233-242.419. Woodrow, G. C., I. G. Young, and F. Gibson.

1975. Mu-induced polarity in the Escherichiacoli K-12 ent gene cluster: evidence for a gene(entG) involved in the biosynthesis ofentero-chelin. J. Bacteriol. 124:1-6.

420. Wulff, D. 1967. 8-Aminolaevulinic acid-requir-ing mutant from Escherichia coli. J. Bac-terial. 93:1473-1474.

421. Yaguzhinsky, L. S., L. I. ,Boguslavsky, A. G.Volkov, and A. B. Rakhmaninova. 1976.Synthesis of ATP coupled with action ofmembrane protonic pumps at the octane-wa-ter interface. Nature (London) 259:494-496.

422. Yamamoto, T. H., M. Mevel-Ninio, and R. C.Valentine. 1973. Essential role of membraneATPase on coupling factor for anaerobicgrowth and anaerobic active transport inEscherichia coli. Biochim. Biophys. Acta314:267-275.

423. Yates, M. G., and C. W. Jones. 1974. Respira-tion and nitrogen fixation in Azotobacter.Adv. Microbial. Physiol. 11:97-135.

424. Yates, M. G., and K. Planque. 1975. Nitrogen-ase from Azotobacter chrococcum; purifica-tion and properties of the component pro-teins. Eur. J. Biochem. 60:467-476.

425. Yoshida, M., N. Sone, H. Hirata, and Y. Ka-gawa. 1975. A highly stable adenosine tri-phosphatase from a thermophilic bacterium;purification, properties and reconstitution.J. Biol. Chem. 250:7910-7916.

426. Yoshida, M., N. Sone, H. Hirata, Y. Kagawa,Y. Takeuchi, and K. Ohno. 1975. ATP syn-thesis catalyzed by purified DCCD-sensitiveATPase incorporated into reconstituted pur-ple membrane vesicles. Biochem. Biophys.Res. Commun. 67:1295-1300.

427. Young, I. G. 1975. Biosynthesis of bacterialmenaquinones. Menaquinone mutants ofEscherichia coli. Biochemistry 14:399-406.

428. Young, I. G., L. Langman, R. K. J. Luke, andF. Gibson. 1971. Biosynthhesis of the iron-transport compound enterochelin: mutantsof Escherichia coli unable to synthesize 2,3-dihydroxylbenzoate. J. Bacteriol. 106:51-57.

VOL. 41, 1977


http://mm

br.asm.org/

Dow

nloaded from


bacterial respiration · 50 haddockandjones sulfur proteins of the nadh dehydrogenase...

Documents