INFECTION AND IMMUNITY. OCt. 1986. p. 85-89 Vol. 54, No. 10019-9567/86/100085-05502.00/0Copyright t 1986. American Society for Microbiology

Role of Capsule and 0 Antigen in Resistance of Klebsiellapneumoniae to Serum Bactericidal Activity

JUAN M. TOMAS,* VICENTE J. BENEDi, BLANCA CIURANA, AND JUAN JOFRE

Departttnenito de Microbiologia, Facdltad de Biologit, Universidad de Barc(elona, 08071 Barcelona, Spain

Received 28 April 1986/Accepted 7 July 1986

The ability of Klebsiella pneumoniae strains to resist the bactericidal activity of serum was quantitated. TheK. pneumoniae strains tested included mutants lacking the capsular polysaccharide and mutants having amodified lipopolysaccharide structure. The last mutants were obtained as phage-resistant mutants, and theirlipopolysaccharide was characterized by sodium dodecyl sulfate-polyacrylamide gel electrophoresis andchemical analysis. Serum-resistant mutants derived from phage-resistant mutants (lipopolysaccharide mu-tants) were also characterized. Resistance to the bactericidal activity of complement was mediated by thelipopolysaccharide, especially by the 0-antigen polysaccharide chains. The capsular polysaccharide seemed notto play any important role in resistance to serum bactericidal activity in this bacterium.

The complement system of homeothermic animals playsan important role in protection against bacterial infections.In nonimmune hosts, this primary defense is mediatedlargely through activation of the alternative complementpathway, whereas in immune hosts, both classical andalternative pathways participate. Complement can produce adirect bactericidal reaction between bacter-ia and phagocyticcells (9).

Clearly, if an invading organism can resist the bactericidalactivity of complement in nonimmune serum, it has a sur-vival advantage and, hence, a pathogenic advantage. In thecase of gram-negative pathogens of homeothermic animals,this virulence property is conferred by either normal lipo-polysaccharide (LPS) or protein components of the outermembrane (OM) (4). In immune animals, the binding ofantibodies to cell surface components can overcome thisresistance.

Klebsiellai paeiiioioiae biotypes are widely recognized asopportunistic pathogens acting as agents of bacteremias andrespiratory and genitourinary infections, particularly in pa-tients under stress (6). Unlike other enterobacteria, K.pCnleuOniae contains a large capsular polysaccharide in-volved in colonization, adhesion, maintenance, and prolifer-ation of this species on its host (24). We isolated several K.pneunioniae bacteriophages whose receptor was the LPS(21, 26; J. M. Tomas and V. J. Benedi, manuscript inpreparation). By using K. pnieil/noniiae C3 LPS mutants(isolated as phage-resistant mutants) and capsule mutants,we examined the role of the capsular polysaccharide andLPS in the survival of this strain in serum.

MATERIALS AND METHODS

Bacteria, bacteriophages, and media. The strains used arelisted in Table 1. Bacteriophages FC3-1, FC3-2, FC3-3,FC3-6, and FC3-9 were partially characterized previously(21). The basal medium used for bacterial growth and phagepropagation was Luria broth (LB) (16) or LB with 1.5% agar.To prepare soft agar, we added 0.6% agar to LB.

Spontaneous mutants of K. pnelunzoniae C3 resistant tobacteriophage FC3-1 (a bacteriophage whose receptor isLPS) were previously described (26). Spontaneous mutants

* Corresponding author.

of K. pnieiumoniiae C3 resistant to bacteriophages FC3-2,FC3-3, FC3-6, and FC3-9 were isolated by spreading amixture containing 108 bacteria and 109 phage PFU on LBwith 1.5% agar. After 36 h at 30°C, colonies of phage-resistant mutants were picked. purified by streaking, andcross-streaked against the corresponding bacteriophage toconfirm resistance. Resistance to the other bacteriophageswas assayed by a spot test.

Bacterial survival in fresh nonimmune serum. The survivalof logarithmic-phase bacteria in serum was measured. To 0.4ml of fresh nonimmune rabbit serum was added 0.1 ml of asuspension in phosphate-buffered saline of bacteria in thelogarithmic phase, adjusted by optical density to about 5 x107 cells per ml. Viable counts were made at 30°C up to 3 hby the Miles-Misra method (15) after serial dilution inphosphate-buffered saline. Results are expressed as thepercentage of inoculated bacteria that survived after a giventime. Controls (bacteria in phosphate-buffered saline)showed no significant changes in viable counts over theincubation period. In some assays, the same serum previ-ously heated at 56°C for 30 min or treated with 20 mM EDTAfor 1 h at 37°C to destroy complement activity was used.Serum was usually used on the day of collection.

Serum-resistant mutants were obtained as survivors aftertreatment with fresh normal serum for 3 h. After 36 h at30°C, colonies of serum-resistant mutants were picked,purified by streaking, and retested with fresh normal serumto confirm resistance.

Cell surface isolation and analyses. Cell envelopes wereprepared by French pressure cell lysis at 16,000 Pa of wholecells or were disrupted by sonication, followed by theremoval of unbroken cells at 10,000 x g for 10 min and bysedimentation of the membrane fraction at 100,000 x g for 2h. Cytoplasmic membranes were solubilized twice withsodium lauryl sarcosinate (8), and the OM fraction wassedimented twice at 100,000 x g for 2 h. OM proteins weresolubilized in 1% deoxycholate-2 mM EDTA (20). Mem-brane proteins were analyzed by sodium dodecyl sulfate-polyacrylamide gel electrophoresis by a modification (1) ofthe Laemmli procedure (12). Protein gels were routinelystained with Coomassie blue. Protein concentrations weredetermined by the Lowry procedure (13) with bovine serumalbumin as the standard.

85

on June 23, 2020 by guesthttp://iai.asm

.org/D

ownloaded from

86 TOMAS ET AL.

TABLE 1. Strains of K. pneumoniae used and their relevantcharacteristics and origin

Strain Relevant characteristics Origin

C3 Wild type; O1:K66 5, 11KT141 FC3-1 phage-resistant mutant 26KT701 FC3-2 phage-resistant mutant This studyKT702 FC3-3 phage-resistant mutant This studyKT703 FC3-6 phage-resistant mutant This studyKT707 FC3-9 phage-resistant mutant This studyKT717 Nonencapsulated strain; O1:K- (Frield 204) I. 0rskovKT142 Serum-resistant mutant from KT141 This studyKT719 Serum-resistant mutant from KT701 This studyKT720 Serum-resistant mutant from KT702 This studyKT721 Serum-resistant mutant from KT703 This studyKT723 Serum-resistant mutnat from KT707 This study

LPS was purified by the method of Westphal and Jann (29)as modified by Osborn (19). For chemical analyses, purifiedLPS was hydrolyzed with 1 N HCl for 2 h at 100°C.Colorimetric analyses of the 2-keto-3-deoxyoctulosonic acid(KDO) and L-glycero-D-manno-heptose (heptose) contentsof LPS were performed by the method of Osborn (19).Organic phosphate was assayed by the method of Bartlett(2). Monosaccharides were also analyzed as their alditolacetate derivatives by gas-liquid chromatography on a 3%SP-3840 column (Supelco Inc., Cras, Switzerland) at 225 and180°C. Alditol acetate monosaccharides were obtained by astandard procedure (bulletin 774A, Supelco). Alditol acetatecarbohydrate standards were either purchased from Supelcoor prepared by us.

Purified LPS was further analyzed by sodium dodecylsulfate-polyacrylamide gel electrophoresis and silver stainedby the method of Tsai and Frasch (28). Purified LPS fromEscherichia coli 055:B5 (smooth strain) for use as a standardwas purchased from Difco Laboratories, Detroit, Mich.

Antisera. (i) Anti-LPS serum. Immune serum was obtainedfrom adult New Zealand White rabbits previously injectedwith 50 p.g of purified LPS in Freund complete adjuvant,followed by two successive injections at 2-week intervals.After 2 weeks, the animals were bled, and serum wascollected. Fresh normal serum was obtained from untreatedanimals.

(ii) Anti-capsule serum. Immune serum was obtained fromadult New Zealand White rabbits by repeated immunizationwith whole Formalin-killed K. pneumoniae C3 cells (wildtype) by the method of Edwards and Ewing (7). The serumwas rendered capsular polysaccharide (K antigen) specific

TABLE 2. Phage sensitivity of K. pneumoniae strainsSensitivity to phagea:

StrainFC3-1 FC3-2 FC3-3 FC3-6 FC3-9

C3 S S S S SKT141 R R R R RKT701 R R R R SKT702 R R R R SKT703 R R R R RKT707 R R R R RKT142 S S S S SKT719 S S S S SKT720 S S S S SKT721 S S S S SKT723 S S S S S

a R, Resistant; S, sensitive.

0.1

i 2 3 HFIG. 1. Survival of K. pneumoniae strains in fresh normal se-

rum. Symbols: O, strain C3 (wild type); A, strain KT717;S*, strainKT141; *, strain KT701; and *, strain KT707. Strain KT703showed survival in serum similar to that of strain KT141, and strainKT702 showed survival in serum similar to that of strain KT701. H,Hours.

by extensive adsorption with the KT717 strain (01; Frield204) as described by 0rskov and 0rskov (18).

Assay of antibody levels. Immune rabbit sera were testedwith a solid-phase enzyme-linked immunosorbent assay(ELISA) for LPS (0.1eug of antigen) and capsular polysac-charide (1 ,ug of antigen or 104 cells as antigen) indepen-dently. The developing antibody was a 1/50 dilution ofaffinity-purified goat anti-rabbit immunoglobulin G-alkalinephosphatase (27).

RESULTS

Mutants resistant to bacteriophages FC3-1, FC3-2, FC3-3,FC3-6, and FC3-9 occurred spontaneously at a frequency of1 x 10-6 to 5 x 10-6 and fell into a single class based on theirphage sensitivity patterns (Table 2). They are represented bystrains KT141, KT701, KT702, KT703, and KT707, respec-tively. The parent strain, K. pneumoniae C3, was sensitiveto all of the five related phages (FC3-1, FC3-2, FC3-3,FC3-6, and FC3-9). Serum-resistant mutants (KT142,KT719, KT720, KT721, and KT723) were obtained as de-scribed in Materials and Methods from the serum-sensitive,phage-resistant mutants (KT141, KT701, KT702, KT703,and KT707, respectively). All of the serum-resistant mutantswere sensitive to all of the five related phages (Table 2).

Survival in fresh nonimmune serum. Results of freshnonimmune serum survival experiments with phage-resist-ant mutant strains, strain KT717 (nonencapsulated), andstrain C3 (wild type) are shown in Fig. 1. Strain KT707showed a great decrease (under 1% survival) in viablenumbers after 1 h, strains KT701 and KT702 showed adecrease (under 1% survival) after 2 h, and strains KT141and KT703 showed a decrease (under 1% survival) after 3 h.No decrease was observed with strains C3 and KT717.To determine whether differences in serum survival

among strains were due to nutritional rather than bacteri-cidal factors, we measured survival in serum heated ortreated with 20 mM EDTA to destroy complement. Theseresults (Table 3) showed that there were no significant

INFECT. IMMUN.

on June 23, 2020 by guesthttp://iai.asm

.org/D

ownloaded from

K. PNEUMONIAE SURVIVAL IN SERUM 87

TABLE 3. Survival of K. pneumoniae in complement-depletedfresh normal serum'

% of inoculum surviving at:Strain

lh 2h 3h

C3 137 172 268KT717 121 164 201KT141 135 169 257KT701 131 170 251KT702 129 166 259KT703 136 173 270KT707 138 174 269

a Complement depletion was achieved by either heating the serum for 30min at 56°C or treating the serum with 20 mM EDTA for 1 h at 37C. The datarepresent the values obtained by heating the serum at 56°C for 30 min. Thevalues obtained by treating the serum for 1 h with 20 mM EDTA are verysimilar.

differences between the C3 and KT717 strains and thephage-resistant mutant strains (KT141, KT701, KT702,KT703, and KT707).No serum killing in fresh nonimmune serum survival

experiments was observed with strains KT142, KT719.KT720, KT721, and KT723 (serum-resistant mutants derivedfrom strains KT141, KT701, KT702, KT703, and KT707,respectively) (data not shown).

Analyses of OMs. The OM protein profiles examined bysodium dodecyl sulfate-polyacrylamide gel electrophoresis(SDS-PAGE) of K. pneiumoniae C3 (wild type), KT717.KT141, KT701, KT702, KT703, and KT707 are shown inFig. 2. Similar patterns were seen for strains C3 (wild type)and KT717 (nonencapsulated). Some differences were ob-served between the phage-resistant mutants and the wildtype, the most important being the lack of one protein of 36kilodaltons in strain KT707 (presumably one of the porins in

92

.u66

-21

1 2 34 56 7FIG. 2. SDS-PAGE of OM proteins from K. pneium)oniae strains.

OM proteins were obtained as sodium lauryl sarcosinate-insolublematerial (8). The molecular size standards (21, 31, 45. 66. and 92kilodaltons) used were from Bio-Rad Laboratories. Richmond.Calif. Each lane contained 40 jig of protein. Lanes: 1, strain C3 (wildtype); 2, strain KT717; 3, strain KT141; 4, strain KT701; 5. strainKT702; 6, strain KT703; and 7, strain KT707.

.~~~ O-ANI

^^_=~~~= C OR E

1 2345 6 7 8910111213FIG. 3. SDS-PAGE of LPS from K. pneuinoniae strains. Purified

LPS was assayed by the method of Tsai and Frasch (28). Lanes: 1,strain C3 (wild type); 2, strain KT717; 3. strain KT141; 4. strainKT142; 5. strain KT701; 6, strain KT719; 7. strain KT702; 8. strainKT720; 9. strain KT703; 10, strain KT721; 11. strain KT707: 12,strain KT723; and 13, E. coli 055:B5 (smooth strain) used as acontrol for the LPS 0 antigen and core.

this strain). There were also alterations in a number ofbands, but there was no consistency from strain to strain.We also examined purified LPS from these strains by themethod of Tsai and Frasch (28) (Fig. 3). There was anapparently complete loss of 0 antigen in the purified LPSfrom the phage-resistant mutants, indicating that the 0antigen was lost rather than altered in structure. Identicalprofiles were seen for strains C3 (wild type) and KT717(nonencapsulated).The chemical analyses of purified LPS from K. pneiumo-

niae C3 (wild type) and the phage-resistant mutants areshown in Table 4. K. pneirnoniae C3 has been serotyped as01 on the basis of its LPS (I. 0rskov, personal communica-tion), and its chemical composition is in accordance with thisserotype. Strains KT141 and KT703 had LPS lacking riboseand arabinose but containing xylose, perhaps as a substitu-tion that stops LPS elongation (chemotype Ra). StrainisKT701 and KT702 had LPS lacking pentoses and somehexoses (maybe chemotype Rb or Rc). Strain KT707 hadLPS deficient in pentoses and hexoses but containing KDOand heptoses (chemotype Rd). All of these strains, asexpected, had increased quantities of KDO and heptoses(LPS inner core) per milligram of LPS in comparison withLPS from the wild type (smooth). These data are consistentwith the lack of LPS 0 antigen in these phage-resistantmutants.

Serum-resistant mutants (KT142, KT719, KT720, KT721,and KT723) had OM protein profiles identical to that of thewild-type strains, as determined by SDS-PAGE (data notshown). They also had LPS (with 0-antigen polysaccharide)identical to that of the wild-type strain (Fig. 3), as deter-mined by the method of Tsai and Frasch (28).Antibody assays. The ELISA for LPS revealed a 500-fold

decrease (antibody titration) in the response against purifiedLPS from phage-resistant mutants in comparison with puri-fied LPS from the wild type. No differences were observedbetween purified LPS from K. pneium1oniae C3 (wild type)and purified LPS from strain KT717 (nonencapsulated) orpurified LPS from serum-resistant mutants (KT142, KT719,KT720, KT721, and KT723).

- r

VOL. 54, 1986

TIGEN

on June 23, 2020 by guesthttp://iai.asm

.org/D

ownloaded from

88 TOMAS ET AL.

TABLE 4. Chemical composition of purified LPS from K. pneutnoniae C3 (wild type) and several phage-resistant mutants

LPS from Amt (tLmol/mg of LPS) of:strain: KDOa. Heptosesa. Hexosesh Riboseh Arabinoseh Xyloseh Organic

phosphate'

C3 0.035 0.28 0.10 0.14 0.13 0 0.72KT141 0.102 0.71 0.10 0 0 0.12 0.73KT701 0.086 0.63 0.06 0 0 0 0.72KT702 0.088 0.63 0.05 0 0 0 0.73KT703 0.101 0.69 0.11 0 0 0.12 0.74KT707 0.112 0.79 0 0 0 0 0.73

aAssayed by a colorimetric method (10).h Assayed by gas-liquid chromatography.Assayed by the method of Bartlett (2).

The presence of capsular polysaccharide in these strains(Table 5) was detected by the ELISA as described inMaterials and Methods. All the strains tested had capsularpolysaccharide in similar amounts, except for strain KT717(nonencapsulated).

DISCUSSIONK. pneilEnoniiae C3 has been serotyped as K66 for its K

antigen and 01 for its LPS 0 antigen (I. 0rskov, personalcommunication). Bacteriophages FC3-1, FC3-2, FC3-3,FC3-6, and FC3-9 were previously characterized (21), andwe found that LPS alone is the receptor for all these K.pnfeiumoniiae phages (26; Tomis and Benedi, in preparation).Mutants resistant to these phages were isolated, and all ofthem had alterations in LPS structure. The common alter-ation in LPS composition was the lack of the 0 antigen in allof the phage-resistant mutants. This point was concludedfrom the LPS gels, the chemical composition of LPS fromthe wild type and from the mutants, and finally from theELISA experiments with purified LPS.The chemical composition of LPS from these mutants

suggested three different chemotypes. The mutants selectedfor FC3-1 and FC3-6 phage resistance (KT141 and KT703),as described before (26), appear to have xylose substituted intheir remaining core oligosaccharides, and maybe this inser-tion of xylose stops the elongation of the LPS in these

TABLE 5. Presence of capsular polysaccharide and high- andlow-molecular-weight LPS fractions in OMs

LPShStrain Capsular

polysaccharide' High mass Low mass

(O antigen) (core)

C3 + + +KT717 - + +KT141 + - +KT701 + - +KT702 + - +KT703 + - +KT707 + - +KT142 + + +KT719 + + +KT720 + + +KT721 + + +KT723 + + +

" Capsular polysaccharide was detected with anti-capsule serum assayed bythe ELISA. We used 104 prewashed cells as antigen. -, Negative in theELISA with a 10-2 antibody dilution; +. Positive in the ELISA with a 10-4antibody dilution.

I LPS fractions were detected by SDS-PAGE of purified LPS stained by themethod of Tsai and Frasch (28) and by chemical composition analyses. +.Presence; -, absence.

mutants, preventing the insertion of the 0 antigen into theLPS core. We suggest that these mutants belong tochemotype Ra. The mutants selected for FC3-2 and FC3-3phage resistance (KT701 and KT702) lack pentoses andsome hexoses from their LPS composition; they are LPScore mutants probably belonging to chemotype Rb or Rc.The mutants selected for FC3-9 phage resistance (KT707)lack all the hexoses and pentoses from their LPS composi-tion, only containing KDO and heptoses; they are LPS coremutants belonging to chemotype Rd. It seems clear thatthere are at least three different types of LPS core in thesephage-resistant mutants.The changes observed in OM protein composition be-

tween the wild type and the phage-resistant mutants arepossibly due to LPS differences. LPS and OM proteininteractions have been well documented (1, 3, 23). The mostevident change in OM protein composition was observed instrain KT707, which lacks one of the major outer membraneproteins in K. pneilinoniae (36 kilodaltons), presumably oneof the porins in this strain. KT707 is, to our knowledge, thefirst presumable porin mutant described in K. pneumoniaeand may be very useful in transport studies, as in E. coli orSalmtionella tvphimnurium (17).

Furthermore, we have been able to find revertants fromthe phage-resistant mutants (which were serum sensitive) byselecting for resistance to serum bactericidal activity. Theserum-resistant mutants (KT142, KT719, KT720, KT721,and KT723) derived from the phage-resistant mutants(KT141, KT701, KT702, KT703, and KT707, respectively)had LPS identical to that of the wild-type strain. Thisidentity was concluded from the LPS gels, the phage sensi-tivity patterns, and the results obtained with the ELISA forpurified LPS. They also had the same OM protein composi-tion as the wild-type strain (data not shown).That the ability of some pathogenic strains to resist

complement killing clearly resides in their surface structurehas already been described (22). Indeed, it would appear thatK. pneiumoniiae has a double barrier, the capsular polysac-charide and the LPS, able to prevent the access of mem-brane lytic complement to their membrane targets. Thisstudy has shown that the major contribution of LPS toresistance to serum bactericidal activity is mediated by the 0antigen, as for nonencapsulated strains from other entero-bacteria (14, 25). The results obtained with the phage-resistant mutants (LPS mutants) indicated that the loss of thelong 0-antigen polysaccharide chains allows for the accessof complement components to their targets, with concomi-tant bactericidal effects. Regaining the 0-antigen polysac-charide chains was a common characteristic of the serum-resistant mutants obtained from the LPS mutants (phage-resistant mutants). Also, there seemed to be degrees of

INFECT. IMMUN.

on June 23, 2020 by guesthttp://iai.asm

.org/D

ownloaded from

K. PNEUMONIAE SURVIVAL IN SERUM 89

resistance to serum bactericidal activity, depending on theLPS core length, as can be seen in the phage-resistantmutants. The mutation in the deepest part of the LPS coreresulted in high sensitivity to complement killing (strainKT707, chemotype Rd), while strains KT701 and KT702(chemotype Rb or Rc) were less sensitive than strain KT707;strains KT141 and KT703 (chemotype Ra) were less sensi-tive than all the other ones.The role of capsular polysaccharide in resistance to serum

bactericidal activity in K. pneumoniae seems to be nullbecause strain KT717, with the same serotype for LPS(serotype 01) but without capsular polysaccharide, showedresistance very similar to that of the wild-type strain (encap-sulated). Also, no changes in K antigen were observed in thephage-resistant mutants (LPS mutants). The complete LPSstructure is obviously sufficient to confer resistance tocomplement killing, while capsular polysaccharide by itselfclearly allows for the access of complement components totheir targets.

This study has defined an important role for LPS in thesensitivity of K. pneumoniae C3 (encapsulated strain) toserum. We also demonstrated that capsular polysaccharidedoes not play any important role in survival in serum.Furthermore, this study provides an interesting model forclarifying the contribution of LPS and capsular polysaccha-ride to different aspects of the pathogenicity of K. pneumo-niae.

ACKNOWLEDGMENT

J. M. Tomas was supported by a postdoctoral fellowship fromC.I.R.I.T. (Generalitat de Catalunya).

LITERATURE CITED1. Ames, G. F. L., E. N. Spudich, and H. Nikaido. 1974. Protein

composition of the outer membrane of Salmonella typhimurium:effect of lipopolysaccharide mutations. J. Bacteriol. 117:406-416.

2. Bartlett, G. R. 1959. Phosphorus assay in column chromatogra-phy. J. Biol. Chem. 234:466-468.

3. Beher, M. G., and C. A. Schnaitman. 1981. Regulation of theOmpA outer membrane protein of Escherichia coli. J. Bacteriol.147:972-985.

4. Buchanan, T. M., and W. A. Pearce. 1979. Pathogenic aspects ofthe outer membrane components of Gram-negative bacteria, p.475-514. In M. Inouye (ed.), Bacterial outer membranes. JohnWiley & Sons, Inc., New York.

5. Clotet, R. C., J. Guinea, and R. Pares. 1968. Segregaci6n deaminoacidos por una cepa de C. intermedius. Microbiol. Esp.21:155-173.

6. Cruickshank, R., J. P. Duguid, B. P. Marmion, and R. H. A.Swain. 1975. In Medical microbiology, 12th ed., vol 2. ChurchillLivingstone Inc., New York.

7. Edwards, P. R., and W. H. Ewing. 1972. Identification ofEnterobacteriaceae, 3rd ed. Burgess Publishing Co., Minneap-olis.

8. Filip, C., G. Fletcher, J. L. Wulif, and C. F. Earhart. 1973.Solubilization of the cytoplasmic membrane of Escherichia coliby the ionic detergent sodium-lauryl-sarcosinate. J. Bacteriol.115:717-722.

9. Frank, M. F. 1979. The complement system in host defense andinflammation. Rev. Infect. Dis. 1:483-501.

10. Hanson, R. S., and J. A. Phillips. 1981. Chemical composition,p. 328-364. In P. Gerhardt, R. 0. E. Murray, R. N. Costilow,E. W. Nester, W. A. Wood, N. R. Krieg, and G. B. Phillips(ed.), Manual of methods for general bacteriology. AmericanSociety for Microbiology, Washington, D.C.

11. Imperial, J., R. Pares, and J. A. Cole. 1982. Defective nitrateassimilation by a derivative of Klebsiella pneumoniae strain C3(formerly Citrobacter intermedius C3) which has lost theisocitrate dehydrogenase plasmid. FEMS Microbiol. Lett.13:247-252.

12. Laemmli, U. K. 1970. Cleavage of structural proteins during theassembly of the head of bacteriophage T4. Nature (London)227:680-685.

13. Lowry, 0. H., N. J. Rosebrough, A. L. Farr, and R. J. Randall.1951. Protein measurement with the Folin phenol reagent. J.Biol. Chem. 193:265-275.

14. Lugtenberg, B., and L. van Alphen. 1983. Molecular architec-ture and functioning of the outer membrane of Escherichia coliand other Gram-negative bacteria. Biochim. Biophys. Acta737:51-115.

15. Miles, A. A., and S. S. Misra. 1938. The estimation of thebactericidal power of the blood. J. Hyg. 38:732-749.

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

17. Nikaido, H., and M. Vaara. 1985. Molecular basis of bacterialouter membrane permeability. Microbiol. Rev. 49:1-32.

18. 0rskov, F., and I. 0rskov. 1978. Serotyping of Enterobacteria-ceae with special emphasis on K antigen determination, p.37-38. In J. R. Norris and T. Bergen (ed.), Methods in micro-biology, vol. 11. Academic Press, Inc., New York.

19. Osborn, M. J. 1966. Preparation of lipopolysaccharide frommutant strains of Salmonella. Methods Enzymol. 8:161-164.

20. Randall-Hazelbauer, L., and M. Schwartz. 1973. Isolation of thebacteriophage lambda receptor from Escherichia coli. J. Bacte-riol. 116:1436-1446.

21. Regue, M., J. Tomas, R. Pares, and J. Jofre. 1981. Isolation andpartial characterization of phages infecting Citrobacter interme-dius C3. Curr. Microbiol. 5:153-156.

22. Rogers, H. J., H. R. Perkins, and J. B. Ward. 1980. Microbialcell walls and membranes. Chapman & Hall, Ltd., London.

23. Smit, J., K. Kamio, and H. Nikaido. 1975. Outer membrane ofSalmonella typhimurium: chemical analysis and freeze-fracturestudies with lipopolysaccharide mutants. J. Bacteriol.124:942-958.

24. Sutherland, I. W. 1978. Surface carbohydrates of the procary-otic cell. Academic Press, Inc., New York.

25. Taylor, P. W. 1983. Bactericidal and bacteriolytic activity ofserum against gram-negative bacteria. Microbiol. Rev.47:46-83.

26. Tomas, J., and J. Jofre. 1985. Lipopolysaccharide-specificbacteriophage for Klebsiella pneumoniae C3. J. Bacteriol.162:1276-1279.

27. Trust, T. J., W. W. Kay, E. E. Ishiguro, J. T. Buckley, andT. W. Pearson. 1982. Properties of A protein, a virulence factoron the surface of Aeromonas salmonicida. Dev. Comp. Immu-nol. 6(Suppl. 1).

28. Tsai, C. M., and C. E. Frasch. 1982. A sensitive silver stain fordetecting lipopolysaccharide in polyacrylamide gels. Anal.Biochem. 119:115-119.

29. Westphal, O., and K. Jann. 1965. Bacterial lipopolysaccharides:extraction with phenol-water and further applications of theprocedure. Methods Carbohydr. Chem. 5:83-91.

VOL. 54, 1986

on June 23, 2020 by guesthttp://iai.asm

.org/D

ownloaded from