Vol. 53, No. 1INFECTION AND IMMUNITY, JUlY 1986, p. 129-1340019-9567/86/070129-06$02.00/0Copyright © 1986, American Society for Microbiology

Expression of Mycoplasma pneumoniae Antigens in Escherichia coliL. B. TREVINO, W. G. HALDENWANG, AND J. B. BASEMAN*

The University of Texas Health Science Center at San Antonio, Department of Microbiology, San Antonio, Texas 78284

Received 30 October 1985/Accepted 7 April 1986

A genomic library of Mycoplasma pneumoniae was generated by using bacteriophage lambda EMBL3 as thevector. Screening of the library for the expression of M. pneumoniae protein antigens with adsorbed anti-M.pneumoniae serum revealed strong reactivity from a third of those clones which contained mycoplasma DNAinserts. Three of the most highly reactive clones were analyzed in detail and found to synthesize discretemycoplasma proteins. Two carried overlapping fragments of mycoplasma DNA which encoded a protein thatwas readily detected in Escherichia coli after infection with recombinant bacteriophage. The third clonecontained a novel mycoplasma DNA fragment which directed the synthesis of two additional mycoplasmaproteins. Further screening of the library with antiserum raised against the major M. pneumoniae adhesinprotein P1 (165 kilodaltons [kDa]) yielded one clone which produced an immunologically reactive protein of 140kDa. Adsorption of anti-Pl serum by this clone selected a population of antibodies that were reactive with M.pneumoniae adhesin P1 (165 kDa). These results demonstrate that immunologically active M. pneumoniaeproteins are synthesized in E. coli.

Mycoplasma pneumoniae is a procaryote which causesprimary atypical pneumonia in humans. Tissue colonizationand subsequent disease production require cytadherence ofthe organism to respiratory epithelial cells (9, 12, 14). Atrypsin-sensitive surface protein designated P1 is likely to bethe major adhesin of M. pneumoniae (7). The proper posi-tioning of P1 in the mycoplasma membrane appears to be aprecisely regulated process. This protein is sparsely distrib-uted over the entire surface of avirulent hemadsorption-negative mycoplasmas (1) which also typically lack severalother membrane proteins (5, 9), while P1 molecules aredensely clustered at the tip of virulent hemadsorption posi-tive mnycoplasmas (1).The mechanism(s) by which P1 and other proteins are

oriented in the mycoplasma membrane, as well as the role ofthese other proteins in cytadherence, is not known. Wechose to investigate these questions by isolating the struc-tural genes for mycoplasma surface proteins. These clonedDNAs could then be used to determine the primary structureof the proteins, their genetic organization, and the factorsinfluencing their expression. In this report we describe thecloning of M. pneumoniae DNA fragments into a bacterio-phage lambda vector (EMBL3) (4) and the subsequentgeneration in Escherichia coli of proteins which are immu-nologically reactive with antiserum raised against whole M.pneumoniae organisms. In addition, we isolated a clone thatis likely to carry the coding sequence for the presumptivemajor mycoplasma adhesin, protein P1.

MATERIALS AND METHODS

Organisms and growth conditions. M. pneumoniaeM129-B16 cultures were grown at 37°C in 32-oz (about946-ml) bottles containing 70 ml of Hayflick medium. After72 h, glass-attached mycoplasmas were washed three timeswith phosphate-buffered saline (PBS; 137 mM NaCl, 2.7 mMKC1, 4.6 mM Na2PO4, 1.5 mM KH2PO4) before they werescraped into PBS and pelleted at 10,000 x g for 20 min at40C. The pellets were stored at -700C until they were needed

* Corresponding author.

either for DNA extractions or as an antigen source forWestern blots.

E. coli LE392 (13) (F- hsdR5J4 supE44 supF58 lacYgalK2 galT22 metB trpRSS and A-) and NM539 (4) (supFhsdR [P2 cox]) were grown in L broth (13).

Antiserum. Production of hyperimmune rabbit sera di-rected against whole M. pneumoniae organisms and M.pneumoniae P1 protein has been described previously (8).Rabbit anti-M. pneumoniae, anti-Pl, and preimmune serawere extensively adsorbed with E. coli LE392 (intact cellsand lysates) before they were used in immunological screen-ing of the clone bank. Test sera were incubated at 4°Covernight with intact E. coli LE392 at a concentration of 7.5x 1011 E. coli cells per 1 ml of serum. Aggregates werepelleted at 10,000 x g for 20 min at 4°C. Adsorbed sera werethen incubated as described above with a lysate from 7.5 x1011 E. coli cells per 1 ml of serum and centrifuged at 40,000x g for 60 min at 4°C to remove aggregated antibody andcellular debris. E. coli LE392 lysates were prepared bypassing cells washed and suspended in PBS containing 1 mMphenylmethylsulfonyl fluoride thru a French pressure celltwice at 15,000 lb/in2.DNA extraction. Pellets of M. pneumoniae were sus-

pended in 2.7 ml of PBS, lysed by the addition of 0.3 ml of10% sodium dodecyl sulfate (SDS), and incubated with 10 ,ugof RNase for 30 min at 37°C. Preparations were extractedthree times with an equal volume of redistilled phenol(equilibrated with 100 mM Tris [pH 8.0]-10 mM EDTA][TE]) followed by dialysis overnight at 4°C against a total of6 liters of sterile TE.

Bacteriophage EMBL3 DNA was prepared by extractingthe phage particles three times with an equal volume ofphenol as described above and three times with an equalvolume of ether. Bacteriophage DNA preparations wereethanol precipitated and suspended in TE.

Construction of genomic library. M. pneumoniae DNA (80,ug) was partially digested with Sau3A enzyme and fraction-ated by centrifugation through a linear gradient (10 to 40%)of sucrose (13). Gradient fractions containing DNA frag-ments of 11 to 20 kilobase pairs (kbp) were combined anddialyzed against TE.

129

on May 5, 2018 by guest

http://iai.asm.org/

Dow

nloaded from

130 TREVINO ET AL.

Bacteriophage EMBL3 DNA was digested with BamHIand EcoRI. The resultant DNA preparations were phenolextracted, isopropanol precipitated, and suspended in TE.The endonuclease-digested EMBL3 vector DNA was

added to the sized M. pneumoniae fragments at a ratio of 3:1(1.5 p.g of phage DNA arms to 0.5,ug of mycoplasma DNAinsert). The DNA was coprecipitated with ethanol, sus-

pended in 10pul of TE, and ligated overnight at 14°C with T4DNA ligase (Bethesda Research Laboratories, Gaithers-burg, Md.). Recombinant DNA was packaged to produceviable phage with a lambda in vitro packing system (Be-thesda Research Laboratories), according to the instructionsof the manufacturer.

Immunological screening of the clone bank. Recombinantphage were plated with E. coli LE392 to produce approxi-mately 500 plaques per plate (13). Following plaque forma-tion the plates were cooled to 4°C and overlaid with nitro-cellulose disks for transfer of proteins. Disks were removedfrom plates, dried at room temperature, and washed oncewith 500 mM NaCl-20 mM Tris (pH 7.4) (TBS). Disks were

then processed through the immunoblot procedure (seebelow) for screening of M. pneumoniae protein-producingclones. Adsorbed rabbit anti-M. pneumoniae serum andadsorbed rabbit anti-Pl monospecific serum served as theantibody probes.On isolation of individual plaques, clones were spotted in

duplicate on E. coli lawns. Proteins from the plaques were

transferred onto nitrocellulose disks and processed as de-scribed above.

Gel electrophoresis and Western blotting. M. pneumoniaeprotein (2 mg) was suspended in 0.3 ml of PBS, and an equalvolume of 100 mM Tris (pH 6.8)-2% SDS-20% glycerol-2%2-mercaptoethanol-0.02% bromophenol blue buffer (SPbuffer) was added. Samples were boiled for 5 min andelectrophoresed on a 7.5% polyacrylamide gel prior toelectrophoretic transfer to nitrocellulose paper (11, 16).After protein transfer, the nitrocellulose was cut into stripsand reacted with selectively adsorbed antibody reagents bythe immunoblot technique.Immunoblot. Nitrocellulose blots were blocked in 1.5%

bovine serum albumin (BSA)-1.5% gelatin in TBS for 3 to 4h prior to incubation with the primary antibody. Adsorbedrabbit sera (anti-M. pneumoniae, anti-Pl, or preimmune)were diluted 1:50 (preadsorbed volume) in TBS and centri-fuged at 10,000 x g for 60 min at 4°C to remove aggregates.An equal volume of 1.5% BSA-1.5% gelatin in TBS was

added, producing a final serum dilution of 1:100. Blots were

incubated with the diluted serum preparation overnight at

room temperature with shaking, followed by three 10-minwashes with TBS. Horseradish peroxidase-conjugated goat

anti-rabbit immunoglobulin G (IgG) diluted 1:2,000 in TBScontaining 0.75% BSA-0.75% gelatin was added to the blotsand incubated with shaking for 3 to 4 h at room temperature.Blots were washed three times for 10-min periods with TBSprior to substrate development (6).DNA analysis of recombinant clones. The recombinant

phage DNA was isolated by a rapid, small-scale plate lysatemethod (13). The DNA was digested with Sall restrictionendonuclease prior to electrophoresis on a 0.5% agarose

mini-gel.Electrophoretic and blotting analysis of clones. Proteins

were harvested from plate lysates by scraping soft agarose

overlays from the plates, passing them through a 22-g needleinto a Corex tube, and eluting with 4 ml of SM buffer (13) for2 h at 4°C. The agarose was pelleted by centrifugation at

10,000 x g for 15 min at 4°C prior to trichloroacetic acid

precipitation of the supernatant by the addition of coldtrichloroacetic acid, for a final concentration of 10%. Sam-ples were incubated at4°C overnight prior to centrifugationat 10,000 x g for 20 min at 4°C. Supernatants were dis-carded; and pellets were washed twice with 1 ml of PBS,suspended in 200p,l of SP buffer, and neutralized with 1,ul of5 N NaOH. Samples were boiled for 5 min, and solubilizedproteins were electrophoresed and transferred to nitrocellu-lose. Blots were processed by the immunoblot technique,using as the primary antibodies rabbit anti-M. pneumoniae,anti-Pl, or preimmune adsorbed sera at 1:100 in 0.75%BSA-0.75% gelatin.

Selective antibody adsorption with phage lysates. Proteinsfrom plate lysates of individual clones were transferred tonitrocellulose disks and used to selectively adsorb antibodiesgenerated against M. pneumoniae proteins. Preblocked blotswere incubated with primary antibody reagents, as describedabove for the immunoblot procedure, and washed threetimes with TBS to remove nonspecifically bound antibodies.Selectively bound antibodies were eluted from the blots byincubation for 6 to 8 h at room temperature in 8 ml of elutionbuffer (200 mM glycine, 200 mM NaCl, 0.1% BSA, [pH 2.8]).The enriched antibody solution was dialyzed for 6 to 8 h in 50mM citric acid-50 mM Na2HPO4 (pH 5.5) and for 12 to 16 hin two changes of PBS. An equal volume of 1% gelatin wasadded to the antibody preparation prior to its use as theprimary antibody in the immunoblot technique (3, 10).

RESULTS

Construction of a genomic library. A genomic library ofrecombinant E. coli clones expressing M. pneumoniae pro-teins was prepared by ligating Sau3A-generated M. pneumo-niae DNA fragments to EMBL3 DNA which had beendigested with both BamHI and EcoRI to decrease thepossibility of reforming the original vector (4). Packaging ofthe recombinant EMBL3-M. pneumoniae DNA yielded 2 x105 PFU/2 ,ug of DNA when plated on the E. coli P2lysogenic strain NM539. Amplification of the library on thishost generates clones enriched for those containing foreignDNA inserts (4). On screening the library with preadsorbedanti-M. pneumoniae serum, 33% of the clones containing M.pneumoniae DNA inserts produced proteins which wereimmunologically reactive. Reactivity among specific clonesranged from very intense color reactions to very light spotsor no reaction at all (Fig. 1A). Several of the positive cloneswere picked from the prescreened plate and spotted induplicate onto fresh E. coli lawns. Plaques were againscreened with preadsorbed anti-M. pneumoniae serum. Re-activity among positive duplicate plaques was equally in-tense, while the negative plaques reacted very weakly or notat all (Fig. 1B). Preimmune adsorbed rabbit serum wasnonreactive with the genomic library.

Electrophoretic analysis of immunologically reactive pro-teins. To further characterize the M. pneumoniae protein-producing clones, plate lysates were prepared from three ofthe highly reactive clones and processed for SDS-polyacrylamide gel electrophoresis. No differences wereobserved between the protein profiles of recombinantbacteriophage lysates and that obtained from a EMBL3control when the resulting gels were stained with Coomassieblue (data not shown). However, Western blots of theseprofiles screened with adsorbed anti-M. pneumoniae serumshowed distinct reactive proteins with various molecularweights. Two of the recombinant phage (clones 12 and 28)led to the synthesis of a protein of 29.3 kilodaltons (kDa),

INFECT. IMMUN.

on May 5, 2018 by guest

http://iai.asm.org/

Dow

nloaded from

M. PNEUMONIAE ANTIGEN EXPRESSION IN E. COLI 131

A,

FIG. 1. Screening of recombinant clones with antimycoplasma sera. Recombinant phage clones were screened with preadsorbed rabbitanti-M. pneumoniae serum at a 1:100 dilution as the primary antibody and horseradish peroxidase-conjugated goat anti-rabbit IgG as thesecondary antibody as described in the text. (A) Reactivity of 500 clones randomly plated on E. coli; (B) reactivity of individual clones spottedin duplicate onto E. coli lawns.

with one of them (clone 28) coding for the synthesis of anadditional protein of 123 kDa (Fig. 2, lanes A and C). A thirdclone (clone 27) generated proteins of 67.5 and 26 kDa (Fig.2, lane B). None of the proteins from a control EMBL3lysate reacted with the antibody (Fig. 2D). In Fig. 2E isdisplayed the pattern of immunologically reactive proteins ofM. pneumoniae, which is similar to total M. pneumoniaeprotein profiles (1, 8). Thus, distinct mycoplasma proteinsare being synthesized in E. coli infected with the recombi-nant phage DNAs.

A B C D E

Analysis of cloned M. pneumoniae DNA inserts. DNAanalysis of the recombinants was performed to examine thesize and possible relatedness of the cloned inserts. Becausethe BamHI site used as the cloning site lies 6 bases inside ofthe SalI sites on EMBL3 (4), the recombinant DNA prepa-rations were digested with Sall endonuclease. This digestionresulted in the cleavage of the EMBL3 phage arms awayfrom the complete M. pneumoniae DNA insert containing 6additional bases on either end. In Fig. 3 is shown restrictedDNA from the clones which synthesized the protein de-tected in Fig. 2 electrophoresed on a 0.5% agarose gel.Lanes A, B, and C in Fig. 3 contain M. pneumoniae DNAinserts of approximately 12, 12.2, and 17 kbp, respectively,

A B C D

123 -

67.5-

29.3 - .mw26 -

upOIWF 17 -12 -

8.2 -

c..

_F : ,,~~~~~

I;

.N

FIG. 2. Western blot analysis of recombinant phage proteinswhich were immunologically reactive with rabbit anti-M. pneumo-niae serum. Proteins from plate lysates of individual recombinantphage were electrophoresed on a 7.5% polyacrylamide gel, electro-phoretically transferred onto nitrocellulose paper, and screenedwith preadsorbed anti-M. pneumoniae serum (see text). Lanes A, B,and C, profiles of recombinant clones 12, 27, and 28, respectively;lane D, control profile containing the wild-type phage EMBL3; laneE, M. pneumoniae protein profile. The numbers on the left indicatethe sizes of the visualized proteins in kilodaltons.

4-

- 19.4--EMBL3ARMS

- 9.2 -

FIG. 3. Restriction enzyme analysis of DNA from recombinantclones. The lanes depict SalI-digested recombinant DNAs followingelectrophoresis on a 0.5% agarose gel. Lanes A, B, C, digestedDNAs from clones 12, 27, and 28 respectively; lane D, digestedvector DNA (molecular mass, 19.4, 9.2, 8.4, and 4 kilobases). Thenumbers indicate the sizes of fragments in kilobase pairs.

VOL. 53, 1986

on May 5, 2018 by guest

http://iai.asm.org/

Dow

nloaded from

132 TREVINO ET AL.

A

*

P ......

B

.. -P

FIG. 4. Western blot analysis of a clone that reacted with anti-Plserum. Proteins from a plate lysate which reacted with anti-Plserum were screened (as described in the legend to Fig. 2) withpreadsorbed monospecific rabbit anti-Pl serum at a 1:100 dilution.Lane A, protein profile of recombinant clone 58; lane B, M.pneumoniae protein profile. The proteins that reacted with theantiserum are depicted as P1* in the recombinant clone profile andP1 in the mycoplasma profile.

in addition to the EMBL3 phage arms (19.4 and 9.2 kbp).The DNA insert from lane B (Fig. 3) contains an internal Sallsite producing two bands (4 and 8.2 kbp) on Sall digestion.The analysis of the proteins generated from these clonesrevealed a protein with a common molecular mass (29.3 kDa)coming from clones 12 and 28. To investigate the possibilitythat the 12-kbp fragment cloned in clone 12 could be acomponent of the 17-kbp fragment of clone 28, we digestedboth cloned DNAs with an additional restriction endonucle-ase (HindIII) and discovered that each clone contained DNAfragments of common mobility when electrophoresedthrough agarose gels (data not shown). We verified that theinsert in clone 12 was indeed a part of the insert in clone 28by radioactively labeling the 12-kbp mycoplasma DNA in-sert of clone 12 and hybridizing it to SalI restriction endo-nuclease-generated DNA fragments from clones 12 and 28that had been electrophoretically separated on agarose gelsand transferred to nitrocellulose. The radioactively labeledprobe specifically hybridized to both mycoplasma DNAinserts. We conclude that clones 12 and 28 represent inde-pendent isolates of a common region of the mycoplasmachromosome.

Screening recombinants with anti-Pl polyclonal serum.Because recombinant clones were capable of synthesizingimmunologically reactive M. pneumoniae proteins in E. coli,we screened the library with adsorbed anti-Pl polyclonalserum in the hope of identifying the structural gene for thisprimary mycoplasma adhesin. Among 2,500 to 3,000plaques, one plaque (clone 58) produced a reaction with theanti-Pi serum (data not shown). The plaque was spotted induplicate on a fresh E. coli lawn and screened as describedabove. The duplicate plaques reacted with equal intensity tothe anti-Pl serum. Proteins synthesized by E. coli followinginfection with this recombinant phage clone were analyzedfurther to determine the molecular weight of the protein(s)

reactive with antisera raised against whole M. pneumoniaeand protein P1.

Plate lysates of clone 58 were harvested, and proteinswere electrophoresed on a 7.5% polyacrylamide gel andblotted onto nitrocellulose paper prior to immunologicalscreening with adsorbed serum directed against P1. Onemajor band was observed which corresponded to a protein,P1*, with an apparent molecular mass of 140 kDa (Fig. 4).Authentic P1 migrates in this gel system with an apparentmolecular weight of 165 kDa. Thus, this cloned DNA islikely to either be directing the synthesis of a protein withP1-like antigenicity but an altered mobility during SDS-polyacrylamide gel electrophoresis or a protein that reactswith a contaminating activity in our polyclonal sera. Todistinguish between these possibilities, proteins synthesizedby E. coli infected with clone 58 were used to adsorbimmunoglobulin populations that reacted with them from theanti-Pl sera. The selectively bound antibodies were used asprobes in an immunoblot against total M. pneumoniae pro-teins. Antibodies enriched by clone 58 bound to a proteinband that comigrated with P1, while no anti-Pl antibodieswere adsorbed by wild-type EMBL-3 phage (Fig. 5). Inaddition, large-scale phage lysates of clone 58 were pre-pared, and detergent-solubilized proteins were electropho-retically separated and used to adsorb immunoglobulin prep-arations as described above. Protein P1* (140 kDa) was cutfrom the nitrocellulose, and bound antibodies were elutedfrom the P1* strips for immunoblotting against total M.pneumoniae proteins. Only a single protein band, P1 (165kDa), was immunologically detected.

A B C

FIG. 5. Selective adsorption of anti-Pl antibody by proteinsfrom a recombinant clone. Proteins from recombinant clone 58 werebound to nitrocellulose and incubated with preadsorbed anti-Plserum. Bound antibodies were eluted, dialyzed, and reacted withWestern blot strips containing M. pneumoniae protein profiles, asdescribed in the text. Lane A, reaction of the antibodies whichbound to clone 58 with the M. pneumoniae protein profile; lane B,reaction of the antibodies which bound to EMBL3 with the M.pneumoniae protein profile; lane C, M. pneumoniae protein profilereacted with anti-Pl serum.

INFECT. IMMUN.

on May 5, 2018 by guest

http://iai.asm.org/

Dow

nloaded from

M. PNEUMONIAE ANTIGEN EXPRESSION IN E. COLI 133

We conclude that the protein P1* synthesized from clone58 DNA has antigenic determinants in common with authen-tic P1 and probably contains at least a portion of the P1structural gene.

DISCUSSION

To establish an alternate approach for understanding thestructure-function properties of M. pneumoniae adhesinprotein P1 and other cytadherence related proteins, weconstructed a M. pneumoniae genomic library, using lambdaphage EMBL3 as the vector. This library provides thematerial for examining and manipulating the structural genesthat encode these proteins and for generating M. pneuimo-niae proteins in an E. coli host.DNA fragments from another mycoplasma species (M.

hyorhinis) have been successfully cloned with lambda phageCharon 4 as the vector (15). We therefore chose an E.coli-bacteriophage lambda system for our initial cloningexperiments and constructed a genomic library containingM. pneuimoniae DNA fragments of approximately 15 kbp inlambda phage EMBL3 (2).The M. pneumoniae genome consists of approximately

800 kb of DNA. If we assume that our partial digestions withSau3A resulted in the generation of random DNA fragmentsand if the average cloned fragment is approximately 15 kb,then it can be calculated that approximately 230 cloneswould be needed to have a 99% probability of including anygiven portion of the mycoplasma genome within the clonebank (2). The library described in this report consists of 5 x105 independent clones containing M. pneiumoniae DNAfragments. Therefore, the yield of recombinant phage shouldeasily cover the entire mycoplasma genome, allowing arealistic anticipation for the expression of most M. pneumo-niae proteins. Immunological screening of the clones withM. pneumoniae antiserum demonstrated strong reactivity of33% of the DNA-containing clones. Further screening withanti-Pl serum identified one strongly reactive clone (clone58) which we assumed would carry the structural gene forP1. Western blot analysis, however, revealed a protein of140 kDa (P1*) instead of an anticipated 165-kDa protein (P1).It is therefore formally possible that the 140-kDa protein isnot P1 but instead a distinct mycoplasma protein that reactswith a contaminating activity in the anti-Pl polyclonal sera.

This is unlikely given the observation that adsorption ofanti-Pl antiserum by the proteins synthesized from clone 58and especially by the 140-kDa protein enriched for antibod-ies reactive with a mycoplasma protein that migrated asauthentic P1 in SDS-polyacrylamide gel electrophoresis. Ata minimum, P1 and the 140-kDa proteins must share com-mon antigenic sites. If we assume that P1 and the 140-kDaproteins are products of the same structural gene, whatmight account for the difference in apparent molecular mass?Possible causes could be that (i) the insert may not containthe entire gene, (ii) the protein is not processed completely inan E. coli host, (iii) the protein is processed entirely but isdegraded by E. coli, or (iv) the mycoplasma P1-codingsequence uses a triplet code that is read as a terminationsignal in E. coli (17), leading to the synthesis of a truncatedprotein.The clone that carries the putative P1 structural gene

contains an 18-kb mycoplasma DNA insert. This insert is 2to 3 times larger than the estimated 5- to 6-kbp fragmentrequired for coding a 165-kDa protein. The clone expressesthe 140-kDa protein, in addition to several other proteins (60,55, 40, 39, and 28 kDa) detected with antiserum against

whole M. pneumoniae. If these proteins are each indepen-dent gene products, it would suggest that a large portion ofthe DNA insert is expressed in E. coli.

Aside from the clone that may carry the structural gene forP1, we identified clones that code for other mycoplasmaproteins. The finding that two of three clones, picked on thebasis of the fact that they reacted strongly with antiseraagainst whole myoplasmas, had common insert sequencessuggests that our screening technique identifies the struc-tural genes for major immunogenic proteins. We are cur-rently subcloning the DNA fragments derived from ourprimary clones to localize the coding sequences for P1 andother mycoplasma proteins on the DNAs. These DNAsshould be useful in analyzing the properties of P1 andaccessory proteins as they relate to the phenomenon ofmycoplasma cytadherence.

ACKNOWLEDGMENTS

We thank Rose Garza for secretarial assistance and C. Su and V.Tryon for technical advice and assistance.

This research was supported by Public Health Service grant Al18540 from the National Institute of Allergy and Infectious Diseases.

LITERATURE CITED1. Baseman, J. B., R. M. Cole, D. C. Krause, and D. K. Leith.

1982. Molecular basis for cytadsorption of Mycoplasmna pneu-moniae. J. Bacteriol. 151:1514-1522.

2. Clarke, L., and J. Carbon. 1976. A colony bank containingsynthetic ColEl hybrid plasmids representative of the entire E.coli genome. Cell 9:91-99.

3. Engleberg, N. C., E. Pearlman, and B. I. Eisenstein. 1984.Legionella pneumnophila surface antigens cloned and expressedin Escherichia coli are translocated to the host cell surface andinteract with specific anti-Legionella antibodies. Infect. Immun.160:199-203.

4. Frischauf, A., H. Lehrack, A. Poustka, and N. Murray. 1983.Lambda replacement vectors carrying polylinker sequences. J.Mol. Biol. 170:827-842.

5. Hansen, E. J., R. M. Wilson, and J. B. Baseman. 1979. Two-dimensional gel electrophoretic comparison of proteins fromvirulent and avirulent strains of Mycoplasma pneumoniae.Infect. Immun. 24:468-475.

6. Hawks, R., E. Niday, and J. Gordon. 1982. A dot-immunobind-ing assay for monoclonal and other antibodies. Anal. Biochem.119:142-147.

7. Hu, P. C., A. M. Collier, and J. B. Baseman. 1977. Surfaceparasitism by Mycoplasma pneumoniae of respiratory epithe-lium. J. Exp. Med. 145:1328-1343.

8. Krause, D. C., and J. B. Baseman. 1983. lnhibition ofMycoplasma pneuinoniae hemadsorption and adherence to res-piratory epithelium by antibodies to a membrane protein. Infect.Immun. 39:1180-1186.

9. Krause, D. C., D. K. Leith, R. M. Wilson, and J. B. Baseman.1982. Identification of Mycoplasma pneumnoniae proteins asso-ciated with hemadsorption and virulence. Infect. Immun.35:809-817.

10. Krause, D. C., H. H. Winkler, and D. 0. Wood. 1985. Cosmidcloning of Rickettsia prowazekii antigens in Escherichia coliK-12. Infect. Immun. 47:157-165.

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

12. Lipman, R. P., W. A. Clyde, Jr., and F. W. Denny. 1969.Characteristics of virulent, attenuated, and avirulent Myco-plasma pneumnoniae strains. J. Bacteriol. 100:1037-1043.

13. Maniatis, T. F., E. F. Fritsch, and J. Sambrook. 1982. Molecularcloning: a laboratory manual, p. 64-65, 371-372. Cold SpringHarbor Laboratory, Cold Spring Harbor, N.Y.

14. Powell, D. A., P. C. Hu, M. Wilson, A. M. Collier, and J. B.

VOL. 53, 1986

on May 5, 2018 by guest

http://iai.asm.org/

Dow

nloaded from

INFECT. IMMUN.

Baseman. 1976. Attachment of Mycoplasma pneumoniae torespiratory epithelium. Infect. Immun. 13:959-966.

15. Taylor, A. M., M. A. McIntosh, J. Robbins, and K. S. Wise.1983. Cloned genomic DNA sequences from Mycoplasmahyorhinis encoding antigens expressed in Escherichia coli. Proc.Natl. Acad. Sci. USA 80:4154-4158.

16. Towbin, H., T. Staehelin, and J. Gordon. 1979. Electrophoretic

transfer of proteins from polyacrylamide gels to nitrocellulosesheets: procedure and some applications. Proc. Natl. Acad. Sci.USA 76:4350-4354.

17. Yamao, F., A. Muto, Y. Kawauchi, M. Iwami, S. Iwagami, Y.Asumi, and S. Osawa. 1985. UGA is read as tryptophan inMycoplasma capricolum. Proc. Natl. Acad. Sci. USA82:2306-2309.

134 TREVINO ET AL.

on May 5, 2018 by guest

http://iai.asm.org/

Dow

nloaded from