JOURNAL OF BACTERIOLOGY, Mar. 1978, p. 1478-14830021-9193/78/0133-1478$02.00/0Copyright © 1978 American Society for Microbiology

Vol. 133, No. 3

Printed in U.S.A.

New Major Outer Membrane Protein Found in anEscherichia coli tolF Mutant Resistant to

Bacteriophage TuIbJOHN FOULDS* AND TUU-JYI CHAI

National Institute ofArthritis, Metabolism and Digestive Diseases, National Institutes of Health, Bethesda,Maryland 20014

Received for publication 26 October 1977

Cell envelopes prepared from an Escherichia coli tolF strain selected asresistant to phage TuIb contained a new major outer membrane protein relatedto outer membrane proteins la and Ib. The strain that produces this protein is atolF par double mutant but contains an additional mutation leading to theproduction of the new major outer membrane protein. Antibiotic sensitivity lostas a result of the tolF mutation is regained in strains that contain the new majorouter membrane protein. This indicates that this protein functions to restore theselective permeability of the outer membrane to low-molecular-weight hydro-philic molecules.

The Escherichia coli outer membrane servesas the initial permeability barrier of the cell,allowing passage of hydrophilic molecules witha molecular weight less than about 700 (19). Thismembrane contains a set of three to four majorproteins with molecular weights from 35,000 to40,000, some of which may have a role in main-taining this permeability barrier. Several labo-ratories have described similar or identical setsof proteins, and a variety of nomenclatures haveevolved. These nomenclatures have been sum-marized by Henning and Hailer (10), Lugtenberget al. (14), and, more recently, by Bassford et al.(2).Sekizawa et al. (24) have suggested that the

major outer membrane proteins are synthesizedfrom precursor molecules. Thus, outer mem-brane proteins Ia and/or lb (matrix proteins)may be formed by the action of one or moretoluene-sensitive proteases acting on what hasbeen called promatrix protein.

Proteins Ia and Ib are closely related proteins(22) strongly associated with the peptidoglycanof the cell envelope (20). These proteins, to-gether with lipopolysaccharide, can inactivatecertain bacteriophages and are apparently partof the specific bacteriophage receptors for bac-teriophages TuIa and Tulb (5). Interaction withboth peptidoglycan and bacteriophages suggeststhat these proteins span the outer membrane.

Proteins Ia and lb may contribute directly tothe permeability of the outer membrane to low-molecular-weight hydrophylic substances. Lut-kenhaus has shown that E. coli B strains missingprotein Ia were markedly deficient in their abil-

ity to incorporate a variety of substrates whenthese substrates were present at low concentra-tions (15). Nakae has demonstrated reconstitu-tipp of selective permeability to low-molecular-weight substances using membrane vesicles con-taining outer membrane proteins similar to pro-teins Ia and lb prepared from Salmonella ty-phimurium (17) or E. coli (18). Together withthe likelihood that proteins Ia and lb span theouter membrane, these data suggest that pro-teins Ia and lb may be involved in the formationof transmembrane channels, or pores.We have isolated a mutant strain of E. coli

that contains a new peptidoglycan-associatedmajor outer membrane protein. We call thisprotein E. The mutant strain containing proteinE contains neither major outer membrane pro-teins Ia nor Ib and, thus, may be deficient inproteins required for the formation oftransmem-brane channels or pores. We find that a mutationat a locus separate from the loci leading to theloss of proteins Ia or Ib leads to the synthesis ofprotein E. The presence of protein E in the outermembrane serves to restore the selective perme-ability or transmembrane channels of the outermembrane.

MATERIALS AND METHODSMicroorganisms and growth media. The micro-

organisms in this study are described in Table 1. Thecells were grown at 37°C usually in PPBE medium (8)containing protease peptone no. 3 (Difco Laboratories,Detroit, Mich.), beef extract (Difco), and NaCl. Forpropagation of all bacteriophages except phage Plvirand testing sensitivity to bacteriophages, tryptone me-dium was used. Tryptone medium contained 10 g of

1478

on Septem

ber 14, 2020 by guesthttp://jb.asm

.org/D

ownloaded from

NEW E. COLI MAJOR OUTER MEMBRANE PROTEIN 1479

Strain

E. coli K-12 derivativesJF568

JF703JF694Lin221PB105W1485F- PA2CSH75

BacteriophagesTulaTuIbPA2cHy2Hy8TuII*

TABLE 1. Microorganisms a

Genotype or comment

aroA357 ilv-277 metB65 his-53purE41cyc-I xyl-14 lac629 rpsL97 tsx63proC24

aroA+ tolF4 transductant of strain JF568JF703 parCGSC4861: phoA8 glpT6 reUAl tonA22nalAPhage PA2 lysogenara leu lacY proC purE gal trp his argGmalP rpsL xyl mtl ilv, metB or metA thi

labIblblblbII*

Source or reference

4

4This paper

222

Cold Spring Harbor Lab-oratory

552225

aGenetic symbols are described by Bachmann et al.b These bacteriophages do not grow on an E. coli strain whose outer membrane is missing the indicated

protein.

tryptone (Bacto, Difco) and 5 g of NaCl per liter. LBmedium (16) containing 2.5 mM CaCl2 was used forthe growth of bacteriophage Plvir. Minimal mediumfor transductions was that described by Vogel andBonner (27), supplemented with 0.2% glucose or 0.4%a-glycerol phosphate and growth requirements at ap-propriate concentrations. All media were solidifiedwith 1.5% agar.

Genetic techniques. Growth of bacteriophagePlvir and transduction procedures were accomplishedas described by Signer (25). naU transductants wereselected as described by Hane and Wood (9).

Sensitivity to bacteriophages and colicins.Bacteriophage sensitivity was tested by cross-streak-ing after first resuspending a portion of a single colonyof the strain to be tested in 0.1 to 0.2 ml of saline ortryptone broth. Sensitivity to colicins was determinedby placing about 5-,il portions of each of several serialdilutions of colicin solutions on the surface of a platecontaining PPBE agar medium inoculated with about2 x 107 exponentially growing cells of the strain to betested. These cells were spread over the surface of aPPBE agar plate by the method of Low (12). Afterovernight incubation at 37°C, the activity ofthe colicinwas determined as previously described (8) as thereciprocal of the highest dilution that completely in-hibited growth. Purified colicin L-JF246 was preparedas previously described (7). Crude preparations ofother colicins were prepared from mitomycin C-in-duced cells (8), and the activity of these preparationswas estimated by a spot test as described above.

Bacteriophage induction. Bacteriophage synthe-sis in lysogenic strains was induced by treatment witheither mitomycin C or UV irradiation. Mitomycin C(Sigma Chemical Co., St. Louis, Mo.) was added at afinal concentration of 2 gg/ml to a freshly growingculture in tryptone broth (approximately 2 x 108 cellsper ml). After a 3-h incubation at 37°C with vigorousaeration, a few drops of chloroform were added to killthe cells and promote the release of intracellular bac-

teriophage. Next, cells and cellular debris were re-moved by centrifugation (5,000 x g for 5 min), and thesupernatant fluid containing the bacteriophagewas stored at 4°C over chloroform. Induction by UVlight was accomplished by irradiation of freshly growncells (2 x 10' cells per ml) that had been collected bycentrifugation and resuspended at 2 x 109 cells per mlin saline. After irradiation (survival about 50%), thecells were diluted 10-fold into fresh tryptone broth,and bacteriophage were collected after a 3-h incuba-tion as described above.

Determination of antibiotic sensitivity. Anti-biotic sensitivity was determined by using antibioticsusceptibility test disks (Sensi-Discs, Baltimore Bio-logical Laboratory, Cockeysville, Md.) on PPBE agarmedium. Plates were inoculated with about 2 x 107cells by the method of Low (12), and up to 12 diskswere placed on the surface of the medium. Afterovernight incubation at 37°C, the diameter of the clearzones of inhibition was estimated with an accuracy of±0.5 mm. We scored strains as resistant if the zone ofinhibition was reduced more than 6 mm. The diameterof zones of inhibition on all strains scored as sensitivewas nearly identical for individual antibiotics.

Preparation of celi envelopes. Cell envelope ma-terials were prepared as previously described (3) bydifferential centrifugation of cells disrupted in aFrench pressure cell.Polyacrylamide gel electrophoresis in SDS.

Samples of cell envelope materials were heated at100°C for 5 min in sodium dodecyl sulfate (SDS)-containing buffer as previously described (3). Slabpolyacrylamide gel electrophoresis was performed asdescribed by Lugtenberg et al. (13). Gels were stainedas described by Fairbanks et al. (6) with Coomassiebrilliant blue G and destained by soaking at 37°C inseveral changes of 10% acetic acid.

RESULTSIsolation of toIF, phage TuIb-resistant

VOL. 133, 1978

on Septem

ber 14, 2020 by guesthttp://jb.asm

.org/D

ownloaded from

1480 FOULDS AND CHAI

mutants. Outer membrane materials preparedfrom strain JF703 (toiF) contain protein Ib butno protein Ia as determined by gel electropho-resis (4). Mutants of strain JF703 missing proteinIb as well as Ia were selected by first mixing 109to 1010 phage TuIb with 108 cells of strain JF703and plating the mixture on tryptone agar plates.After overnight incubation at 370C, at least twotypes of colonies appeared. Large mucoid colo-nies, accounting for about one-third of the total,were not further analyzed. From the remainingcolonies, 192 were picked and purified by single-colony streaks. These were next tested for sen-sitivity to phages Tula, TuIb, and TuII*. Toinfect a host, these phages require the presenceof major outer membrane proteins Ia, Ib, andII*, respectively (5). Like the parental strain, all192 strains were resistant to phage Tula andsensitive to phage TuII*. About two-thirds (123of 192) were resistant to phage TuIb. Of these,eight were chosen for further analysis.Major outer membrane proteins missing

in phage TuIb-resistant mutants. SDS-poly-acrylamide gel electrophoresis of cell envelopematerials prepared from each of these eightstrains showed that seven of these eight strainswere missing major outer membrane proteins laand Ib. These seven strains invariably had in-creased amounts of a few higher-molecular-weight proteins (80,000 to 100,000). StrainJF693, previously described (4), is an example ofthis group.Of the eight tolF phage TuIb-resistant mu-

tants studied, cell envelope materials preparedfrom one mutant, strain JF694, contained noprotein Ia or Ib but did contain substantialamounts of a new major outer membrane pro-tein. This protein, which we call protein E, mi-grated more slowly than protein Ia after SDS-polyacrylamide gel electrophoresis, with an ap-parent molecular weight of about 40,000 as com-pared to about 37,500 for protein Ia.

Protein E is tightly associated with the pep-tidoglycan layer of the cell envelope, for it is notextracted after incubation of cell envelope ma-terials with 2% SDS at 600C as described byRosenbusch (20) (data not shown). These andother data that demonstrate conclusively thatprotein E is associated with the outer membranefraction of the cell envelope will be presentedseparately (Chai and Foulds, manuscript inpreparation).

Figpure 1 shows stained gels after electropho-resis of cell envelope materials prepared fromeach of several strains. Major outer membraneproteins la and lb were present in materialsprepared from strain JF568. Similar prepara-tions from strain JF703 contained protein Ib butno la, whereas strain JF694 contained neither

Ialb

Z* -4

A B C DFIG. 1. Stained SDS-polyacrylamide gel after elec-

trophoresis of cell envelope materials prepared fromthe following strains: JF568 (A), JF703 (B), JF694(C), and W1485F PA2c (D). Thepositions ofproteinsIa, Ib, and II* are indicated. The arrow indicates theposition ofpeptidoglycan-associated protein E. Thefigure represents only a portion of the gel.

proteins Ia nor Ib. Strain JF694 contained anadditional protein, labeled protein E in Fig. 1.The electrophoretic mobility of protein E is

close to that of protein 2 found by Schnaitmanet al. (23) in E. coli strains lysogenized by bac-teriophage PA2. Our experience indicates thatprotein E migrates somewhat more rapidly thanprotein 2. We have tried to show that strainJF694 is not a PA2 lysogen. Strain JF694 is nota normal PA2 lysogen, for no detectable bacte-riophage were found after UV or mitomycin C

J. BACTERIOL.

4dwk

on Septem

ber 14, 2020 by guesthttp://jb.asm

.org/D

ownloaded from

NEW E. COLI MAJOR OUTER MEMBRANE PROTEIN 1481

induction under conditions in which similartreatment of strain W1485 PA2 yielded lysatescontaining more than 109 bacteriophage per ml.This does not eliminate the possibility thatstrain JF694 could be lysogenized by a defectivebacteriophage. Unfortunately, we could not ex-

amine this possibility by using techniques forrescue of a putative bacteriophage PA2 hostrange gene by superinfection of strain JF694with phage Xvir as described by Bassford et al.(2) because this strain is resistant to Xvir (Table4). We have isolated two new bacteriophagesthat require the presence of protein E for ad-sorption (Chai and Foulds, manuscript in prep-

aration). Strain JF694 is sensitive to these bac-teriophages, whereas all other strains describedin Table 1, including strain W1485F- PA2, areresistant. These data indicate that the protein Eis not identical with protein 2.Genetic characterization of strain JF694.

Strain JF694 still contains the tolF allele as

indicated by the following results. BacteriophagePlvir grown on this strain was used to transducestrain JF568 (aroA). Examination of the aroA+transductants showed that 53% were resistant tobacteriophage TuIa and colicin A as expectedfor tolF strains (Table 2, cross 1).The data in Table 2 demonstrate that strain

JF694 apparently carries a mutation leading tothe loss of protein lb located close to gipT andnalA.When bacteriophage Plvir prepared on strain

JF694, was used to transduce glpT+ into strainLin221, 10% of these transductants were resist-ant to phage TuIb (Table 2, cross 2). SDS-gelelectrophoretic analysis of cell envelope materi-als prepared from four of these glpT+, phageTuIbr transductants showed that all four strainswere missing outer membrane protein Ib. Thesepreparations did not contain protein E.When bacteriophage Plvir prepared on strain

PB105 was used to prepare nalidixic acid-resist-ant transductants of strain JF694, 43% of thesetransductants were sensitive to bacteriophageTuIb. SDS-gel electrophoresis analysis of cell

envelope materials prepared from five nalA,phage TuIb-sensitive transductants showed five

of five contained outer membrane protein Ib.These preparations still contained protein E.The cotransduction frequencies of the mutationin strain JF694 leading to the loss of protein Ibwith nalA and gipT agree well with those de-scribed by Bassford et al. for the par loci (2)that are involved in the expression of outermembrane protein Ib. Therefore, we believe thatstrain JF694 contains a par mutation. This issupported by the resistance of strain JF694 tobacteriophage PA2c (Table 4), for par mutantsare resistant to this bacteriophage (2). StrainJF694 does not contain a mutation at the ompBlocus which affects proteins Ia or Ib, for no malPtransductants examined were resistant to eitherbacteriophage TuIa or bacteriophage TuIb (Ta-ble 2, cross 4).These data indicate that the mutation leading

to the synthesis of protein E is distinct fromboth the tolF mutation leading to the loss ofprotein Ia and the par mutation leading to theloss of protein Ib.

Using conjugation and transduction with bac-teriophage Plvir, we have located the mutationleading to the synthesis of protein E on the E.coli chromosome. This mutation is located nearilvE at about 82.7 min on the linkage map de-scribed by Bachmann et al. (1) (Foulds and Chai,manuscript in preparation).

Sensitivity to colicins, antibiotics, andbacteriophages. The data summarized in Ta-ble 3 show that strains JF703 and JF694 (bothof which carry the tolF mutation) have identicalpatterns of colicin sensitivity. Introduction ofthe par mutation and a spontaneous mutationleading to the production of protein E in strainJF694 does not affect this pattern.The data in Table 4 show that the bacterio-

phage sensitivity of strain JF694 has been al-tered. The resistance of strain JF694 to phagesPA2, TuIb, and hy2 can be explained by the lossof outer membrane protein lb that serves as a

receptor for these phages (2, 5). We do notunderstand why strain JF694 is resistant tophages Xvir and hy8. Perhaps insertion of pro-tein E in the outer membrane reduces theamount or interferes with the function of the

TABLE 2. Genetic analysis by phage Plvir transduction

Donor no. Donor Recipient Selected marker Recombinantcl. Cotranaduction(no. scored) frequency (%)

1 JF694 JF568 aroA+ (192) aroA' tolFa 532 JF694 Lin221 glpT (192) glpTI Tulbr 103 PB105 JF694 nalA (192) naLA TuIb8 434 JF694 CSH74 malP' (192) malP+ Tular <0.5

malP+ TuIbr <0.5

atolF recombinants are identified as resistant to 1,000 U of colicin A per ml and/or resistance to bacteriophageTula.

VOL. 133, 1978

on Septem

ber 14, 2020 by guesthttp://jb.asm

.org/D

ownloaded from

TABLE 3. Colicina sensitivity of strain JF694Colicin (U/ml)

StrainA B C D El E2 E3 K L M

JF568 20,000 1,280 1,600 160 8,000 2,560 1,024 640 1,000 1,280JF703 <10 1,280 400 160 8,000 40 <10 <10 40 1,280JF694 <10 1,280 400 160 8,000 40 <10 <10 <10 1,280a Colicins were prepared as previously described (8).

TABLE 4. Bacteriophage sensitivity of strain JF694Bacteriophage

StrainTuIa TuIb TuII* PA2c hy2 hy8 Avir

JF568 Sa S S S S R SJF703 R S S S S S SJF694 R R S R R R Ra S, Sensitive; R, resistant.

lamB gene product that is responsible both forthe adsorption of bacteriophage lambda andtransport of maltose (26). All three strains listedin Table 2 were sensitive to phages K3 andBF23.

It would be interesting to examine the outermembrane proteins in a derivative of strainJF694 lysogenic for bacteriophage PA2. Bothprotein 2 (23) and protein E are major outermembrane proteins closely associated with thepeptidoglycan layer of the cell envelope. Thepresence of one may interfere with the other.However, because strain JF694 is resistant toboth bacteriophages PA2 and hy8 (Table 4), wewill have to wait untilwe transduce the mutationleading to the production of protein E into aPA2 lysogen.

Insertion of the protein E in the outer mem-brane of strain JF694 restores sensitivity to allthe antibiotics to which tolF strains are resistant(Table 5). This suggests that the protein E canserve to restore the selective permeability bar-rier of the outer membrane.

DISCUSSIONWe have called the peptidoglycan-associated

outer membrane protein found in strain JF694protein E for we believe the structural gene forthis protein is distinct from other loci involvedin the expression of peptidoglycan-associatedproteins Ta, Ib, and 2. We have located this geneon the E. coli chromosome and will propose thatthe locus be called ompE (Foulds and Chai,manuscript in preparation). Other properties ofthis protein suggest that it is related to thematrix protein. We find that the protein E re-mains associated with the peptidoglycan afterincubation of cell envelope preparations in SDS-containing buffer at 600C as described by Rosen-busch (20), as do proteins Ia and Tb. Purified

preparations of this protein produce a singleband of precipitation on Ouchterlony double-diffusion plates by using antibody produced inresponse to purified matrix protein. This precip-itin band is identical to that produced by purifiedprotein Ia. Furthermore, comparison of theamino acid composition of purified protein Ewith that of proteins Ia and Ib shows all three tobe nearly identical. All three proteins have thesame N-terminal amino acid (alanine). Detailsof these data will be reported separately (Chaiand Foulds, manuscript in preparation).Many of the properties of the protein E are

also shared by outer membrane protein Ic re-cently described by Henning et al. (11). Com-parison of their electrophoretic mobility showswhat may be a discrepancy. In our gel system,protein Ic migrates just ahead of protein Ta,whereas protein E migrates somewhat moreslowly than protein Ta.We do not believe that the protein E is iden-

tical with protein a described by Lugtenberg etal. (14). This protein is reportedly present insmall amounts in several E. coli strains (4, 21)and should render these strains sensitive to thebacteriophages that specifically require the pres-ence of the protein E. These strains are notsensitive to these bacteriophages. In addition,protein a is not associated with the peptidogly-can (20) (Foulds and Chai, unpublished data) asis the protein E.The promatrix protein described by Sekizawa

et al. (24) has many properties in common withprotein E. Both are closely associated with pep-tidoglycan and migrate more slowly than proteinIa in SDS-gel electrophoresis. However, com-parison of the electrophoretic mobility of puri-fied, radioactively labeled protein E with radio-actively labeled toluene-treated cell envelopescontaining promatrix protein demonstratesthese proteins are distinct (Sekizawa, personalcommunication).A tolFpar double mutant has been previously

isolated and described (2). This strain containeda major outer membrane protein that had anelectrophoretic mobility similar to protein Ia.We suggest that this protein is related to proteinE or protein Ic rather than protein Ia. A similarconclusion has been reached by Schnaitman(personal communication).

1482 FOULDS AND CHAI J. BACTERIOL.

on Septem

ber 14, 2020 by guesthttp://jb.asm

.org/D

ownloaded from

NEW E. COLI MAJOR OUTER MEMBRANE PROTEIN 1483

TABLE 5. Antibiotic sensitivity of strain JF694Antibiotic

Strain Chlortet- Oxytetracy- Tetracy- Genta- Kanamy- Cephalo- Chloram-racycine cline cline mycin cin thin phenicol Polymyxin B

JF568 Sa S S S S S S SJF703 R R R R R R R RJF694 S S S S S S S Sa S, Sensitive; R, resistant.

ACKNOWLEDGMENTSWe thank Ulf Henning and Carl Schnaitman for providing

phages TuIa and PA2c and for information on these micro-organisms before publication. We thank Jun Sekizawa forcomparison of the electrophoretic mobility of protein E withpromatrix protein. We also appreciate the prompt response ofU. Henning to our request for a sample of a strain containingouter membrane protein Ic to allow comparison of the electro-phoretic mobilities of proteins Ia, Ic, and E.

LITERATURE CITED1. Bachmann, B. J., K. B. Low, and A. L. Taylor. 1976.

Recalibrated linkage map of Escherichia coli K-12.Bacteriol. Rev. 40:116-167.

2. Bassford, P. J., Jr., D. L. Diedrich, C. L. Schnaitman,and P. Reeves. 1977. Outer membrane proteins ofEscherichia coli. VI. Protein alteration in bacterio-phage-resistant mutants. J. Bacteriol. 131:608-622.

3. Chai, T., and J. Foulds. 1974. Demonstration of a miss-ing outer membrane protein in tolG mutants of Esche-richia coli. J. Mol. Biol. 85:465-474.

4. Chai, T., and J. Foulds. 1977. Escherichia coli K-12 tolFmutants: alterations in protein composition of the outermembrane. J. Bacteriol. 130:781-786.

5. Datta, D. B., B. Arden, and U. Henning. 1977. Majorproteins of the Escherichia coli outer cell envelopemembrane as bacteriophage receptors. J. Bacteriol.131:821-829.

6. Fairbanks, G. T., L. Steck, and D. F. H. Wallach. 1971.Electrophoretic analysis of the major polypeptides ofthe human erythrocyte membrane. Biochemistry10:2606-2617.

7. Foulds, J. 1972. Purification and partial characterizationof a bacteriocin from Serratia marcescens. J. Bacteriol.110:1001-1009.

8. Foulds, J., and C. Barrett. 1973. Characterization ofEscherichia coli mutants tolerant to bacteriocin JF246:two new classes of tolerant mutants. J. Bacteriol.116:885-892.

9. Hane, M. W., and J. H. Wood. 1969. Escherichia coliK-12 mutants resistant to nalidizic acid: genetic map-ping and dominance studies. J. Bacteriol. 99:238-241.

10. Henning, U., and I. Haller. 1975. Mutants of Esche-richia coli K-12 lacking all major proteins of the outercell envelope membrane. FEBS Lett. 55:161-164.

11. Henning, U., W. Schmidmayr, and I. Hindennach.1977. Major proteins of the outer cell envelope mem-brane of Escherichia coli K-12: multiple species ofprotein I. Mol. Gen. Genet. 154:293-298.

12. Low, B. 1973. Rapid mapping of conditional and auxo-trophic mutations in Escherichia coli K-12. J. Bacteriol.113:798-812.

13. Lugtenberg, B., J. Meijers, R. Peters, P. van der

Hoek, and L. van Alphen. 1975. Electrophoretic res-olution of the major outer membrane protein of Esch-erichia coli K-12 into 4 bands. FEBS Lett. 58:254-258.

14. Lugtenberg, B., R. Peters, H. Bernheimer, and W.Berendsen. 1976. Influence of cultural conditions andmutations on the composition of the outer membraneproteins of Escherichia coli. Mol. Gen. Genet.147:251-262.

15. Lutkenhaus, J. F. 1977. Role of a major outer membraneprotein in Escherichia coli. J. Bacteriol. 131:631-637.

16. Miller, J. H. 1972. Experiments in molecular genetics.Cold Spring Harbor Laboratory. Cold Spring Harbor,N.Y.

17. Nakae, T. 1976. Outer membrane of Salmonella. Isolatingof protein complex that produces transmembrane chan-nels. J. Biol. Chem. 251:2176-2178.

18. Nakae, T. 1976. Identification of the outer membraneprotein of Escherichia coli that produces transmem-brane channels in reconstituted vesicle membranes.Biochem. Biophys. Res. Commun. 71:877-884.

19. Nakae, T., and H. Nikaido. 1975. Outer membrane as adiffusion barrier in Salmonella typhimurium. Penetra-tion of oligo- and polysaccharides into isolated outermembrane vesicles and cells with degraded peptidogly-can layers. J. Biol. Chem. 250:7359-7365.

20. Rosenbusch, J. P. 1974. Characterization of the majorenvelope protein from Escherichia coli. Regular ar-rangement on the peptidoglycan and unusual dodecyl-sulfate binding. J. Biol. Chem. 249:8019-8029.

21. Sarma, V., and P. Reeves. 1977. Genetic locus (ompB)affecting a major outer membrane protein in Esche-richia coli K-12. J. Bacteriol. 132:23-27.

22. Schmitges, C. J., and U. Henning. 1976. The majorproteins of Escherichia coli outer cell envelope mem-brane. Heterogeneity of protein I. Eur. J. Biochem.63:47-52.

23. Schnaitman, C. A., D. Smith, and M. Forn de Salas.1975. Temperate bacteriophage which causes the pro-duction of a new major outer membrane protein byEscherichia coli. J. Virol. 15:1121-1130.

24. Sekizawa, J., S. Inouye, S. Halegona, and M. Inouye.1977. Precursors of major outer membrane proteins ofEscherichia coli. Biochem. Biophys. Res. Commun.77:1126-1133.

25. Signer, E. R. 1966. Interaction of prophages at the attnsite with the chromosome of Escherichia coli. J. Mol.Biol. 15:243-255.

26. Szmeleman, S., M. Schwartz, J. T. Silhavy, and W.Boos. 1976. Maltose transport in Escherichia coli K-12. Eur. J. Biochem. 65:13-19.

27. Vogel, H. J., and D. M. Bonner. 1955. Acetylornithinaseof Escherichia coli: partial purification and some prop-erties. J. Biol. Chem. 218:97-106.

VOL. 133, 1978

on Septem

ber 14, 2020 by guesthttp://jb.asm

.org/D

ownloaded from