Proc. Nat. Acad. Sci. USAVol. 70, No. 12, Part I, pp. 3376-3381, December 1973

Anaerobic Transport in Escherichia coli Membrane Vesicles*(fl-galactosides/D-lactate dehydrogenase/a-glycerol phosphate dehydrogenase/fumarate reductase/formate dehydrogenase/nitrate reductase)

WILHELMUS N. KONINGSt AND H. RONALD KABACKt§

t The Laboratorium voor Microbiologie, University of Groningen, Kerklaan 30, Haren, The Netherlands; and tThe Roche Instituteof Molecular Biology, Nutley, New Jersey 07110

Communicated by Sidney Udenfriend, July 26, 1973

ABSTRACT Anaerobic ,6-galactoside transport in wholecells and membrane vesicles from E. coli ML 308-225 iscoupled to the oxidation of a-glycerol-P or D-lactate withfumarate as an electron acceptor. Alternatively, anaerobic,6-galactoside transport may be coupled to the oxidation offormate utilizing nitrate as electron acceptor. Bothanaerobic electron-transfer systems are induced bygrowth of the organisms under appropriate conditions.Components of both systems are loosely bound to themembrane, necessitating the use of a modified procedurefor vesicle preparation in order to demonstrate anaerobictransport in vitro. Addition of ATP or an ATP-generatingsystem to vesicles prepared from anaerobically-grown cellsor inclusion of ATP or the ATP-generating system duringpreparation of vesicles does not stimulate transport. Theresults support the conclusion that active transportunder anaerobic conditions is coupled primarily to elec-tron flow.

Cytoplasmic membrane vesicles isolated from Escherichiacoli, as well as a number of other organisms, catalyze theactive transport of a wide variety of metabolites such as

amino acids, sugars, hexose phosphates, hydroxy-, keto-, anddicarboxylic acids, and potassium or rubidium (in the pres-

ence of valinomycin) (1-7). In E. coli and Salmonella ty-phimurium, these transport systems are coupled primarilyto the oxidation of D-lactate or reduced phenazine metho-sulfate (PMS) via a membrane-bound cytochrome chain withoxygen as the terminal electron acceptor (1, 2). The energy-

coupling site for transport in E. coli vesicles is localized in a

segment of the respiratory chain between D-lactate dehydro-genase and cytochrome b1 (1, 2), and evidence for an electron-transfer-coupling component mediating the coupling ofrespiratory energy to active transport has been presented (8).One aspect of these respiration-dependent transport sys-

tems that has not been studied is their possible relationshipto active transport under anaerobic conditions. Obligateanaerobes or facultative anaerobes growing under anaerobicconditions transport nutrients; moreover, 5-amino-levulinicacid- or heme-requiring mutants of E. coli presumably do notmanifest transport defects. Although ATP is not involved inactive transport under aerobic conditions (1,. 2, 9-15), evi-dence has been presented which is consistent with the interpre-tation that cells can utilize glycolytically-generated ATP to

Abbreviation: PMS, phenazine methosulfate.* This is paper XVII in the series, "Mechanisms of ActiveTransport in Isolated Bacterial Membrane Vesicles." The pre-

ceding paper is given in ref. 5.§ To whom reprint requests should be addressed.

drive transport under anaerobic conditions (10-16). It is alsopossible, however, that such cells might use the same generaltype of transport mechanisms as that used aerobically, withthe exception that an alternative electron acceptor is usedrather than the cytochrome chain and oxygen.

Recent experiments carried out by Butlin (13) demon-strate that mutants of E. coli which are deficient in calcium,magnesium-stimulated ATPase (17) are able to catalyzeactive serine or phosphate transport anaerobically in thepresence of fumarate. The experiments presented here con-firm these observations for fl-galactoside transport, anddemonstrate that anaerobic transport can also be coupled tonitrate reductase. A modified procedure for the preparationof membrane vesicles is described which allows the study ofthese anaerobic-transport mechanisms in vitro.

MATERIALS AND METHODS

Culture Media and Growth Conditions. E. coli ML 308-225(i-z-y+a+) was grown at 370 on minimal medium A (18)supplemented with 0.1% yeast extract (Difco) and the fol-lowing carbon sources, electron acceptors, and trace metals:(a) 0.5% glycerol; (b) 0.5% glycerol plus 10 mM sodiumfumarate; (c) 0.5% glucose; or (d) 0.5% glucose plus 50 mMpotassium nitrate, 1 IAM selenic acid (19-21), and 1 gM sodiummolybdate (19-21). Cultures were grown aerobically as de-scribed previously (18). For anaerobic cultures, 3-liter Er-lenmeyer flasks were filled to the top, tightly stoppered, andstirred slowly by means of a magnetic stirrer. All cultureswere harvested during the exponential phase of growth afteraddition of chloramphenicol to a final concentration of 50/Ag/ml. Cells were sedimented at approximately 13,000 X gfor 20 min, washed once in 50 mM potassium phosphatebuffer (pH 6.6) containing chloramphenicol at 50 $g/ml,and resuspended in 50 mM potassium phosphate b9fer(pH 6.6) containing 10 mM magnesium sulfate and chlor-amphenicol at 50 ,ug/ml.

Preparation of Spheroplasts. Spheroplasts were prepared asdescribed previously (18) with the exception that freshly har-vested cells were directly resuspended in 30 mM Tris . HCl(pH 8.0) containing 20% sucrose and 50 ,ug/ml of chloram-phenicol and converted to spheroplasts without prior washingin 10 mM Trise HCl (pH 8.0).

Preparation of Membrane Vesicles. Membrane vesicles wereprepared either by osmotic lysis as described previously (18)or by a modified procedure described as follows: approxi-

3376

Dow

nloa

ded

by g

uest

on

Aug

ust 2

7, 2

021

Anaerobic Transport in E. coli 3377

mately 2 g wet-weight of a spheroplast pellet obtained by cen-trifugation at 13,000 X g for 30 min was lysed by rapid dis-persion in 10 ml of 10 mM potassium phosphate buffer (pH6.6) containing 2 mM magnesium sulfate and pancreaticDNase (Worthington) and RNase (Worthington) at final con-centrations of 10 ,/g/ml each. Dispersion was achieved- bymeans of a hypodermic syringe fitted with an 18-gauge needle.The lysate was incubated at 370 for 30 min on a rotatingshaker platform. After 15 min, if the suspension was stillviscous, additional magnesium sulfate and DNase was addedto bring the final concentrations to 5 mM and 20 pug/ml, re-spectively, and the incubation was continued for 15 min.The suspension was then centrifuged at approximately 800X g for 1 hr in order to remove whole cells and unlysed sphero-plasts as judged by phase-contrast microscopy. The milky,opalescent supernatant from this centrifugation containingmembrane vesicles was carefully decanted, and centrifugedat approximately 46,000 X g for 1 hr. Finally, the mem-braneous pellet obtained from this high-speed centrifugationwas resuspended in 50 mM potassium phosphate buffer (pH6.6) to a protein concentration of approximately 5-10 mg/ml.Samples of 1.0 ml were rapidly frozen and stored in liquidnitrogen.

Lactose Transport was assayed under aerobic and anaerobicconditions as described previously (18, 22). Anaerobic incuba-tions were carried out under oxygen-free argon (less than 1ppm of oxygen), allowing 5-10 min incubation prior to addi-tion of electron donors, acceptors, and ['4C ]lactose.

Protein Was Determined according to the method of Lowry,et al. (23).

Materials. [1-4C ]Lactose (20 Ci/mol) was obtained fromAmersham-Searle. All other materials were reagent gradeobtained from commercial sources. Stock solutions (1 M) ofelectron donors and acceptors were neutralized to pH 7.0with sodium or potassium carbonate as indicated.

RESULTS

Whole Cells. Anaerobic lactose transport by E. coli ML308-225 grown under each of the conditions described inMethods is shown in Fig. 1. Cells induced for a-glycerol-Pdehydrogenase and fumarate reductase exhibit a marked in-crease in lactose transport when D,-a-glycerol-P and fuma-rate are added to the incubation medium (panel 1). Similarly,cells induced for formate dehydrogenase and nitrate reductaseexhibit a marked increase in lactose transport in the presenceof formate and nitrate (panel 3). Addition of a-glycerol-Pand fumarate to the latter elicits no stimulation of lactosetransport (panel 3), nor does addition of formate and nitratehave any effect on lactose transport by cells grown anaerobi-cally in the presence of glycerol and fumarate (data not shown).When the cells are grown either aerobically on glycerol as acarbon source (panel 2) or anaerobically with glucose as acarbon source (panel 4), addition of a-glycerol-P and fumarateor formate and nitrate causes no stimulation of lactose uptakeover endogenous levels. These data demonstrate that anaero-bic ,B-galactoside transport by whole cells can be coupled toeither of two specific, inducible redox systems-a-glycerol-Pdehydrogenase :fumarate reductase or formate dehydroge-nase :nitrate reductase.

080-

wOC°5 60

40E

20

0 1 2 30 2 30 2 30 2 3MIN

FIG. 1. Anaerobic lactose transport by E. coli ML 308-225cells grown under various conditions. Cells grown as indicatedin the panel headings of the figure and as described in detail inMaterials and Methods were assayed for lactose transport underargon as described previously (18, 22) and in Materials andMethods. Where indicated, DL-a-glycerol-P (sodium salt) andsodium fumarate or potassium formate and potassium nitratewere added to the reaction mixtures to 10 mM final concentra-tions. 0, no additions; O,a-glycerol-P + fumarate; A, formate +nitrate.

Membrane Vesicles. In order to investigate more preciselythe coupling between,-galactoside transport and these an-aerobic electron-transfer systems, membrane vesicles wereprepared from cells grown under the conditions describedabove. Initial attempts to obtain transport under anaerobicconditions with membrane vesicles prepared by the usualmethod (18) were negative, as were studies with vesiclesprepared from spheroplasts lysed in the presence of ATP oran ATP-generating system consisting of ADP, creatine-P,and creatine-P kinase. Since the standard procedure for prep-aration of membrane vesicles involves lysis of spheroplastsby dilution into 300-500 volumes of hypotonic buffer, followedby extensive washing in EDTA-containing buffer (18), thepossibility was considered that one or more factors necessaryfor anaerobic transport might be removed from the vesiclesby these drastic conditions. For these reasons, the alternativemethod described in Materials and Methods was devised. Incontrast to the previous method of preparation (18), thismethod involves lysis of spheroplasts in much smaller volumesin the presence of magnesium sulfate, and extensive washingis avoided. Although yields of membrane vesicles obtainedusing this procedure are considerably diminished, these vesi-cles, shown in Fig. 2, do not differ significantly from thoseprepared according to the original procedure (1, 2, 18). Asshown, the preparation consists essentially of intact unitmembrane-bound sacs. The sacs are surrounded by a single65- to 70-& membrane, and there is no recognizable internalstructure. Moreover, these preparations catalyze anaerobic,B-galactoside transport when the appropriate electron donorsand acceptors are added to the assay mixtures.The data presented in Fig. 3 demonstrate that vesicles

prepared from E. coli MIL 308-225 grown anaerobically onglycerol and fumarate catalyze anaerobic lactose transportin the presence of a-glycerol-P or D-lactate with fumarate asan electron acceptor. In the absence of added electron donorsor acceptor, relatively high endogenous uptake is observed,

Proc. Nat. Acad. Sci. USA 70 (1973)

Dow

nloa

ded

by g

uest

on

Aug

ust 2

7, 2

021

3378 Biochemistry: Konings and Kaback

FIG. 2. Electron micrograph of membrane vesicles isolated from E. coli ML 308-225 grown anaerobically in the presence of glyceroland fumarate. Cells were grown anaerobically and membrane vesicles prepared as described in Materials and Methods. Thin sections ofglutaraldehyde-fixed membrane vesicles were counter-stained with uranyl acetate as described previously (37). The electron micrographshown was obtained by Dr. Milas Boublik and Mr. Frank Jenkins of The Roche Institute of Molecular Biology. Magnification X 88,000.

suggesting that these preparations are not as depleted ofendogenous energy reserves as those prepared using theprevious method (18), despite their appearance under theelectron microscope. This suggestion is consistent with theobservations that addition of either a-glycerol-P or fumaratealone causes significant stimulation of lactose uptake overendogenous levels (left-hand panel). Although data will not bepresented, it is also noteworthy that no stimulation is observedon addition of formate and/or nitrate, ATP, or when thevesicles are prepared in the presence of ATP or an ATP-generating system (ADP, creatine-P, and creatine-P kinase).The data presented in Fig. 4 represent transport assays

carried out with the identical vesicle preparations underaerobic assay conditions. As reported previously (1, 2), D-lactate, a-glycerol-P, and particularly ascorbate-PMS causemarked stimulation of both the initial rate and steady-statelevel of lactose accumulation. Addition of fumarate alone orwith D-lactate, a-glycerol-P (Fig. 4), or ascorbate-PMS(data not shown) has no significant effect.When these membrane vesicles are washed in 50 mM po-

tassium phosphate buffer (pH 6.6) containing 10 mM mag-

nesium sulfate, anaerobic transport in the presence of a-glycerol-P and fumarate is markedly diminished, and com-plete loss of activity is observed after a second wash. How-ever, no loss of activity is observed when the vesicles areassayed under aerobic conditions in the presence of D-lactate,a-glycerol-P, or ascorbate-PMS. These findings provide astrong indication that one or more components of this an-aerobic system are loosely bound to the vesicle membrane.Attempts to restore anaerobic-transport activity with a crudesoluble fraction from sonicated whole cells, flavin adeninedinucleotide, flavin mononucleotide, ubiquinone Qi or Q6,menaquinoneMK 1, or combinations thereof were negative.Membrane vesicles prepared from cells induced for formate

dehydrogenase and nitrate reductase also catalyze anaerobiclactose transport (Fig. 5). As expected, these vesicles requireformate and nitrate rather than a-glycerol-P and fumarate;however, the vesicles exhibit relatively low specific activitiesunder either anaerobic or aerobic conditions. As shown in theleft-hand panel of Fig. 5, addition of formate and nitratecauses stimulation of lactose uptake relative to control sam-ples incubated in the absence of electron donor and acceptor.

Pr&. Nat. Acad. Sci. USA 70 (1973)

Dow

nloa

ded

by g

uest

on

Aug

ust 2

7, 2

021

Anaerobic Transport in E. coli 3379

w21NMIN ~ N

.JL0

EL

0E 2 4 0 2 4

MIN

FIG. 3. Anaerobic lactose transport by E. coli ML 308-225membrane vesicles isolated from cells grown anaerobically inthe presence of glycerol and fumarate. Membrane vesicles wereisolated as described in Materials and Methods and transportassays were carried out as described previously (18, 22) and inMaterials and Methods. Where indicated, D,L-a-glycerol-P(sodium salt), lithium D-lactate, and/or sodium fumarate wereadded to the reaction mixtures at 10 mM final concentrations.0, no additions; *, a-glycerol-P + fumarate; A, fumarate; 0,ca-glycerol-P; A, D-lactate + fumarate; V, D-lactate.

Formate alone has no effect, while addition of nitrate alonecauses transient stimulation of lactose uptake. Although notshown, no other combination of electron donors and acceptorstested stimulates lactose uptake above endogenous levels(e.g., D-lactate plus nitrate, a-glycerol-P plus nitrate, a-glycerol-P plus fumarate), nor does ATP or the ATP-generat-ing system. In contrast to vesicles prepared from cells grownanaerobically with glycerol and fumarate, aerobic transportcatalyzed by these vesicles is markedly diminished (Fig. 5,

AEROBIC

50

40

< 2)40LuEA

0 20-E

10

0 1 2345 0 12 34 5MIN

FIG. 4. Aerobic lactose transport by E. coli ML 308-225membrane vesicles isolated from cells grown anaerobically inthe presence of glycerol and fumarate. Membrane vesicles wereisolated as described in Materials and Methods, and transportof lactose was carried out under aerobic conditions as describedpreviously (18, 22) and in Materials and Methods. Where indi-cated, D,L-a-glycerol-P (sodium salt), lithium D-lactate, and/orsodium fumarate was added to the reaction mixtures in 10 mMfinal concentrations. Sodium ascorbate and PMS, where indi-cated, were added at final concentrations of and 0.1 mM, respec-tively, and these reactions were carried out under oxygen (22).0, a-glycerol-P; *, a-glycerol-P + fumarate; A, fumarate; 0,no additions; 0, ascorbate-PMS; V, D-lactate; A, D-lactate +fumarate.

wE" 4-

WUEQ E A

0EC1

0 1 2 3 4 5 0 1 2 3 4 5MIN

FIG. 5. Anaerobic and aerobic lactose transport by E. coli ML308-225 membrane vesicles isolated from cells grown anaero-bically in the presence of glucose and nitrate. Membrane vesiclesisolated as described in Materials and Methods were assayed forlactose transport under anaerobic (left-hand panel) or aerobic(right-hand panel) conditions as described previously (18, 22) andin Materials and Methods. Where indicated, potassium formate,lithium D-lactate, D,I.a-glycerol-P (sodium salt), and/or potas-sium nitrate were added at final concentrations of 10 mM.Sodium ascorbate and PMS, where indicated, were added at finalconcentrations of 20 and 0.1 mM, respectively, and these reac-tions were carried out under oxygen (22). 0, no additions; @,formate; V, nitrate; A, formate + nitrate; V, D-lactate; 0, a-glycerol-P; E, ascorbate-PMS.

right-hand panel). Despite some stimulation by formate plusnitrate or ascorbate-PMS, the specific activity of these mem-brane vesicles is only about 5-10% of that exhibited by vesi-cles prepared from cells grown anaerobically on glycerol andfumarate (compare Figs. 4 and 5).D-Lactate- or ascorbate-PMS-dependent aerobic lactose

transport by membrane vesicles is strongly inhibited by anumber of electron-transfer inhibitors (1, 2). Potassium cya-nide, for instance, at a concentration of 10mM inhibits lactoseuptake to the extent of 85-90%. In contrast, a-glycerol-Pdehydrogenase:fumarate reductase-coupled anaerobic lactosetransport is only moderately inhibited by cyanide concentra-

2 20

< )

DPo

UJ EWE

E)0 E 10

0

E

V- int.-0 2 3 4 5 0 2 3 4 5

MIN

FIG. 6. Effect of cyanide on anaerobic lactose transport byE. coli ML 308-225 membrane vesicles isolated from cells grownanaerobically in the presence of glycerol and fumarate (left-handpanel) or glucose and nitrate (right-hand panel). The experimentsshown were carried out anaerobically as described in Figs. 3 and5. Where indicated potassium cyanide was added to the incuba-tion mixtures at 20mM final concentration, and the samples were

incubated under argon for 5 min prior to addition of electrondonors and acceptors. 0, a-glycerol-P + fumarate; A, a-glycerol-P + fumarate + KCN; *, no additions; A, KCN; U, formate +nitrate; O., formate + nitrate + KCN.

GLYCEROL + FUMARATE GLUCOSE + NITRATEI I~~~~

_- / 8

~~~~6

.4

2

Proc. Nat. Acad. Sci. USA 70 (1978)

11-11,

FI

Dow

nloa

ded

by g

uest

on

Aug

ust 2

7, 2

021

3380 Biochemistry: Konings and Kaback

tions as high as 20 mM, while the endogenous activity is al-most completely abolished (Fig. 6, left-hand panel). On theother hand, formate dehydrogenase:nitrate reductase-coupledanaerobic lactose transport is completely abolished by 20mM cyanide, as is the endogenous activity (Fig. 6, right-handpanel).

DISCUSSIONExperiments presented in this paper demonstrate clearlythat anaerobic ,3-galactoside transport in whole cells andmembrane vesicles can be coupled to at least two specific,inducible electron-transfer systems which do not requireoxygen as a terminal electron acceptor. In one case, anaerobiclactose transport may be coupled to the oxidation of eithera-glycerol-P or D-lactate with fumarate as an electron ac-ceptor; and in another, to the oxidation of formate with ni-trate as electron acceptor. These findings, as such, providestrong support for the conclusion based on previous experi-ments (1, 2, 5) that active transport in E. coli is coupledprimarily to electron flow.Both of the anaerobic electron-transfer systems used in

these studies have been investigated in some detail in E. coli.The a-glycerol-P dehydrogenase:fumarate reductase systemis particulate, requires both flavin adenine dinucleotide andflavin mononucleotide (24-28), and has most recently beenshown to be capable of catalyzing the phosphorylation ofADP to ATP (29). Although data have not been presentedhere, spectral measurements carried out during the course ofthese investigations indicate that a cytochrome of the b typemay also be involved in the coupling between a-glycerol-Pdehydrogenase (and D-lactate dehydrogenase) and fumaratereductase. The formate dehydrogenase:nitrate reductasesystem is also particulate, involves cytochromes of the b type,is capable of catalyzing ATP formation, and, in addition tonitrate, requires selenium and molybdate for induction ofmaximal activity (30-33).

In order to obtain membrane vesicles that can catalyzeanaerobic lactose transport by means of either of these sys-tems, a modified procedure was devised which apparentlyallows sufficient retention of one or more components of theseanaerobic electron-transfer systems. Thus, membrane vesiclesprepared by the usual means or washed vesicles preparedusing the procedure described here are devoid of activity.Moreover, attempts to reactivate these vesicles were unsuc-cessful. In any case, both the rate and extent of anaerobicca-glycerol-P dehydrogenase (or D-lactate dehydrogenase):fumarate reductase-coupled lactose transport by vesicles pre-pared by the procedure described are comparable to those ob-tained under aerobic conditions with vesicles prepared fromaerobically or anaerobically grown cells. On the other hand,vesicles prepared from cells induced for the formate dehydro-genase: nitrate reductase system exhibit markedly diminishedspecific activities under anaerobic and aerobic conditions.Since precautions were not taken in these studies to preparevesicles under anaerobic conditions or in the dark, this relativelack of activity may be due to the oxygen and/or light labilityof some components of this system (34). In a similar context,the difference in cyanide sensitivity exhibited by the twopreparations may be explained by the ability of cyanide tocomplex selenium and/or molybdate with resultant inhibitionof the formate dehydrogenase: nitrate reductase-coupledtransport system (30).Another aspect of these experiments which bears on the

general problem of anaerobic-transport mechanisms is thefailure to obtain any stimulation whatsoever of lactose trans-port with ATP or ATP-generating systems in any of thepreparations studied. Previous studies (9-15) provide strongsupport for the conclusion that ATP does not play a role inactive transport under aerobic conditions; however, a numberof investigators have provided evidence that ATP may beinvolved in anaerobic transport (10-16). Although it is cer-tainly possible that membrane vesicles prepared under anyconditions lose coupling factors that are necessary for ATP-driven transport, experiments carried out with anaerobicallygrown, cold-shocked cells in which nucleoside triphosphateshave been shown to be substrates for RNA polymerase (35)have also been negative with regard to ATP-driven transport(14, Konings, W. N. & Kaback, H. R., unpublished experi-ments). In view of these findings, the possibility should beconsidered that the involvement of ATP in anaerobic trans-port is indirect, mediated perhaps by reverse electron flow(36).

We thank Dr. Milas Boublik and Mr. Frank Jenkins of TheRoche Institute of Molecular Biology for the electron microscopicstudy presented in this paper, and Mr. Leonard Kushnins forhis excellent technical assistance during the course of thesestudies.

1. Kaback, H. R. (1972) Biochim. Biophys. Acta 265, 367416.2. Kaback, H. R. & Hong, J. -S. (1973) in CRC Critical Re-

views in Microbiology, eds., Laskin, A. I. & Lechevalier, H.(Chemical Rubber Co., Cleveland, Ohio), Vol. 2, pp. 333-376.

3. Konings, W. N. & Freese, E. (1972) J. Biol. Chem. 247,2408-2418.

4. Matin, A. & Konings, W. N. (1973) Eur. J. Biochem. 34,58-67.

5. Reeves, J. P., Shechter, E., Weil, R. & Kaback, H. R.(1973) Proc. Nat. Acad. Sci. USA 70, 2722-2726.

6. MacLeod, R. A., Thurman, P. & Rogers, H. J. (1973) J.Bacteriol. 113, 329-340.

7. Stinnett, J. D., Guymon, L. F. & Eagon, R. G. (1973)Biochem. Biophys. Res. Commun. 52, 284-290.

8. Hong, J. -S. & Kaback, H. R. (1972) Proc. Nat. Acad. Sci.USA 69,3336-3340.

9. Prezioso, G., Hong, J. -S., Kerwar, G. K. & Kaback, H. R.(1973) Arch. Biochem. Biophys. 154, 575-582.

10. Schairer, H. U. & Haddock, B. A. (1972) Biochem. Biophys.Res. Commun. 48, 544-551.

11. Klein, W. L. & Boyer, P. D. (1972) J. Biol. Chem. 247,7257-7265.

12. Berger, E. A. (1973) Proc. Nat. Acad. Sci. USA 70, 1514-1518.

13. Butlin, J. D. (1973) Ph.D. thesis, Australian NationalUniversity, Canberra K, Australia.

14. Parnes, J., R. & Boos, W. (1973) J. Biol. Chem. 248, 4429-4435.

15. Or, A., Kanner, B. I. & Gutnick, D. L. (1973) FEBS Lett.,in press.

16. Abrams, A. & Smith, J. B. (1971) Biochem. Biophys. Res.Commun. 44, 1488-1495.

17. Butlin, J. D., Cox, G. B. & Gibson, F. (1971) Biochem. J.124, 75-81.

18. Kaback, H. R. (1971) in Methods in Enzymology, ed.,Jakoby, W. D. (Academic Press, New York), Vol. XXII,pp.99-120.

19. Lester, R. L. & DeMoss, J. A. (1971) J. Bacteriol. 105,1006-1014.

20. Shum, A. C. & Murphy, J. C. (1972) J. Bacteriol. 110, 447-451.

21. Enoch, H. G. & Lester, R. L. (1972) J. Bacteriol. 110, 1032-1040.

22. Konings, W. N., Barnes, E. M., Jr. & Kaback, H. R. (1971)J. Biol. Chem. 246, 5857-5861.

Proc. Nat. Acad. Sci. USA 70 (1973)

Dow

nloa

ded

by g

uest

on

Aug

ust 2

7, 2

021

Proc. Nat. Acad. Sci. USA 70 (1973)

23. Lowry, 0. H., Rosebrough, N. J., Farr, A. L. & Randall,R. J. (1951) J. Biol. Chem. 193, 265-275.

24. Hirsch, C. A., Rasminsky, M., Davis, B. K. & Lin, E. C. C.(1963) J. Biol. Chem. 238, 3770-3774.

25. Kistler, W. S. & Lin, E. C. C. (1971) J. Bacteriol 108, 1224-1234.

26. Kistler, W. S. & Lin, E. C. C. (1972) J. Bacteriol. 112, 539-547.

27. Spencer, M. E. & Guest, J. R. (1973) J. Bacteriol. 114,563-570.

28. Miki, K. & Lin, E. C. C. (1973), J. Bacteriol. 114, 767-771.29. Miki, K. & Lin, E. C. C. (1973) Fed. Proc. Fed. Amer. Soc.

Exp. Biol. Abstr. 2353.30. Itagaki, E., Takeshi, F. & Sato, R. (1962) J. Biochem.

(Tokyo) 52, 131-141.

Anaerobic Transport in E. coli 3381

31. Ota, A., Yamanaka, T. & Okunuki, K. (1964) J. Biochem.(Tokyo) 55, 131-141.

32. Ruiz-Herrera, J. & DeMoss, J. A. (1969) J. Bacteriol. 99,720-729.

33. Ruiz-Herrera, J., Showe, M. K. & DeMoss, J. A. (1969)J. Bacteriol. 97, 1291-1297.

34. Kearny, J. J. & Sagers, R. D. (1972) J. Bacteriol. 109, 152-161.

35. Rau6, H. A. & Cashel, M. (1973) Biochim. Biophy. Acta312, 722-736.

36. Fisher, R. J. & Sanadi, D. R. (1971) Biochim. Biophys.Acta 245, 34-41.

37. Kaback, H. R. & Stadtman, E. R. (1966) Proc. Nat. Acad.Sci. USA 55, 920-927.

Dow

nloa

ded

by g

uest

on

Aug

ust 2

7, 2

021