Vol. 58, No. 9INFECTION AND IMMUNITY, Sept. 1990, p. 2929-2934)0019-9567/90/092929-06$02.00/0

Oral Immunization with Recombinant Streptococcus lactis Carryingthe Streptococcus mutans Surface Protein Antigen GeneMASAAKI IWAKI,1* NOBUO OKAHASHI,2 ICHIRO TAKAHASHI,2 TAISEI KANAMOTO,2

YOSHIKO SUGITA-KONISHI,l KAGEAKI AIBARA,lt AND TOSHIHIKO KOGA2Department ofBiomedical Research on Foods' and Department of Dental Research,2 National Institute of Health,

Kamiosaki, Shinagawa-ku, Tokyo 141, Japan

Received 10 November 1989/Accepted 27 June 1990

A recombinant Streptococcus lactis strain which carries the structural gene for a surface protein antigen(PAc) of 190,000 daltons from Streptococcus mutans serotype c was constructed for development of an oralvaccine against dental caries. The gene from S. mutans MT8148 joined to shuttle vector pSA3 was successfullytransformed into S. lactis IL1403. A small amount of PAc was detected in the cell homogenate and cytoplasmicfraction of the recombinant S. lactis, but not in the culture supernatant of the recombinant, by Westernimmunoblotting and dot immunoblotting. The level of PAc-specific mRNA in the recombinant strain was lowerthan that in S. mutans MT8148. However, significant salivary immunoglobulin A and serum immunoglobulinG responses to PAc were induced in mice immunized orally with the recombinant S. lactis.

Streptococcus mutans has been strongly implicated as acausative organism of dental caries (7, 18). S. mutanspossesses a cell surface protein antigen of 190 kilodaltonsthat has been variously called antigen IIl (28), B (29), IF(10), P1 (6), and PAc (26). Here we refer to the 190-kilodalton protein antigen of S. mutans serotype c as PAc(26). Okahashi et al. (26) and Lee et al. (14) have cloned thestructural gene (pac) for the protein antigen. The completenucleotide sequence of the gene for PAc has been recentlydetermined (27). Streptococcus sobrinus produces a proteinantigen of 210 kilodaltons named SpaA (9) or PAg (25) whichshows serological cross-reaction with PAc. The proteinantigen gene of S. sobrinus has also been cloned (9, 23, 32).PAc has been successfully used as a vaccine to protect

monkeys against dental caries (15, 17). In addition, localpassive immunization with monoclonal antibodies raisedagainst PAc prevents the colonization by S. mutans ofanimal and human tooth surfaces (16, 19).

Lactic streptococci have been used for a long time forproduction of dairy products such as yogurt and cheese, andfrequent ingestion of a large number of viable bacterial cellsby a great number of the population of the world is awell-known occurrence. On the basis of their safety, lacticstreptococci might be one of the most suitable host organ-isms for oral vaccine production. Recently, protoplast trans-formation (12, 31), electroporation (8), and conjugal plasmidtransfer (11) have become techniques available for introduc-ing foreign DNA into lactic streptococci. Several Esche-richia coli-Streptococcus lactis shuttle vectors have beenconstructed (3, 5, 35) for stable maintenance and expressionof foreign genes in lactic streptococci.Here we describe the introduction and expression of the

pac gene in S. lactis by using an E. coli-S. lactis shuttlevector. Furthermore, we discuss whether the S. lactis pro-ducing recombinant PAc (rPAc) could be useful for thedevelopment of an oral vaccine against dental caries.

* Corresponding author.t Present address: Department of Food and Environmental San-

itation, Food and Drug Safety Center Hatano Research Institute,Hatano-Shi, Kanagawa 257, Japan.

MATERIALS AND METHODS

Bacterial strains and plasmids. The bacterial strains andplasmids used in this study are listed in Table 1. S. lactisIL1403 and E. coli HB101 strains were kind gifts from Danielvan der Lelie (University of Groningen, Groningen, TheNetherlands) and Keiji Yano (University of Tokyo, Tokyo,Japan), respectively. Plasmid pSA3 was a kind gift fromJoseph J. Ferretti (University of Oklahoma, Oklahoma City,Okla.).

Reagents. All restriction enzymes used in this study werepurchased from Toyobo Co. Ltd. (Osaka, Japan). T4 DNAligase was from Takara Shuzo Co. (Kyoto, Japan). Horse-radish peroxidase-conjugated anti-rabbit immunoglobulin G(IgG) was from Bio-Rad Laboratories (Richmond, Calif.).Alkaline phosphatase-conjugated rabbit anti-mouse IgA,IgM, and IgG were from Zymed Laboratories Inc. (SanFrancisco, Calif.). Other reagents were of the highest puritycommercially available.PAc and antiserum. Native PAc was purified from the

culture supernatant of S. mutans MT8148 grown in diffusatemedium of brain heart infusion broth (Difco Laboratories,Detroit, Mich.). Rabbit anti-PAc serum was prepared aspreviously described (24). The PAc was immunologicallyidentical to antigen B kindly supplied by R. R. B. Russell(Royal College of Surgeons of England, Kent, England).

Construction of pSMl. The construction of the chimericplasmid pPC41, containing the structural gene (pac) for PAcfrom S. mutans MT8148, was described previously (26). Thisrecombinant plasmid contains a 7.5-kilobase (kb) insert inpUC118 harboring the entire pac gene. The plasmid pPC41was digested with SphI and BamHI to isolate the 6.2-kbfragment containing the pac gene (Fig. 1). The fragment wasthen ligated to E. coli-S. lactis shuttle vector pSA3 that hadbeen cleaved with the same enzymes. The ligated DNA wasused to transform E. coli HB101 (20). Transformants wereselected on the basis of their anticipated antibiotic resistancephenotypes. Expression of the pac gene in recombinant E.coli cells was examined by colony immunoblotting withrabbit anti-PAc serum (26). All clones that expressed rPAcharbored 16.4-kb chimeric plasmids, and one of these plas-mids was termed pSM1. pSM1 plasmid DNA was preparedas previously described (20).

2929

on March 29, 2020 by guest

http://iai.asm.org/

Dow

nloaded from

2930 IWAKI ET AL.

TABLE 1. Bacterial strains and plasmids

Strain or Phenotype or genotype Referenceplasmid

S. lactisIL1403 Lac- 31IL1403(pSA3) Lac- Emr This paperIL1403(pSM1) Lac- PAc producer, Emr This paper

S. mutans MT8148 PAc producer, serotype c 24E. coli HB101 rB- mB- recAJ3 proA2 20

rpsL20 (Smr)

pSA3 E. coli-S. lactis shuttle, 5Tcr Cmr Emr

pPC41 pac (gene for PAc) Apr, 26pUC118 derivative

pSM1 pac Cmr Emr This paper

Transformation of S. lactis. Transformation of S. lactisIL1403 was performed essentially as described by Simon etal. (31). A 0.2- to 0.5-ml portion of overnight culture of theorganism was seeded into 10 ml of GM17 (M17glc) medium(31, 33), and the culture was grown at 300C to an opticaldensity at 550 nm of 0.22 to 0.25. Cells were then washedwith SMM-GM17 buffer (2x GM17 medium, 0.5 M sucrose,

0.02 M maleate, 0.02 M MgCl2, pH 6.5) and treated with 5 ,ugof lysozyme per ml. The resultant protoplast suspension wasmixed with pSMI plasmid DNA (1 [Lg) or pSA3 plasmidDNA (1 jLg), which was immediately followed by the addi-tion of 40% polyethylene glycol. After 2 min of exposure topolyethylene glycol, the protoplasts were again washed withSMM-GM17 and pour plated into GM17 agar plates contain-ing S ,ug of erythromycin per ml. After incubation at 30°C for

BamHB

Pil pPC41p -t10.6kb

Sphl

PstI

I BamHI+SphI

ppSA3 >_ 10.2kb

4. BamHI

BamHI+Sphl I4Ligation

BaH+Sh

Sphhz/'4;FIG. 1. Construction of pSM1. The recombinant plasmid pSM1

was constructed by joining the BamHI-SphI fragment of pPC41,which contains the pac gene, to the BamnHI-SphI fragment of E.coli-S. lactis shuttle vector pSA3. Open boxes indicate the sequencederived from S. mutans MT8148 containing the pac gene (indicatedby thick open boxes). Shaded and solid boxes of pSA3 and pSM1indicate drug resistance genes and replication and copy numbercontrol regions (in streptococci), respectively.

7 days, the colonies appearing on the plates were isolated.Plasmids in recombinant S. lactis were isolated as describedby Anderson and McKay (1). The plasmid DNA isolated wasfurther purified once or twice with CsCI-ethidium bromidebuoyant density ultracentrifugation.

Southern blot hybridization. The BamHI-SphI fragment ofpPC41 was radiolabeled by nick translation (20) with [32p]dCTP. Plasmid DNA from S. lactis and E. coli was digestedwith BamHI and SphI. The DNA fragments were separatedby agarose gel electrophoresis and then transferred to nitro-cellulose membranes (20). After hybridization was com-pleted, the filter was washed at 68°C in 0.1 x SSC (1 x SSC is0.15 M NaCl plus 0.015 M sodium citrate) and subjected toautoradiography with Kodak X-Omat AR film (EastmanKodak Co., Rochester, N.Y.) with overnight exposure at-800C.Preparation of cell fractions of S. lactis and S. mutans.

Lyophilized cells (0.7 g) were mixed with 10 g of glass beads(0.17 to 0.18 mm in diameter) and suspended in 10 ml ofdistilled water. The mixtures were homogenized 10 times for1.5 min with a mechanical cell homogenizer (model MSK; B.Braun Apparatebau, Melsungen, Federal Republic of Ger-many). The homogenates contained cell walls and solublematerials (cell homogenate). The cell homogenate was fur-ther centrifuged at 10,000 x g for 20 min at 40C, and thesupernatant was used as a cytoplasmic fraction.Dot immunoblotting assay. S. mutans and S. lactis were

grown in brain heart infusion broth. Serial dilutions ofculture supernatants and cytoplasmic fractions of S. mutansand S. lactis and PAc from the culture supernatant of S.mutans MT8148 were prepared in 50 mM Tris hydrochloridecontaining 0.15 M sodium chloride (pH 7.5; TBS). Thesamples (0.5 ml) were placed into wells of the 48-well Bio-Dot SF microfiltration apparatus (Bio-Rad) with a sheet ofnitrocellulose membrane and filtered through the membraneby gravity flow. Proteins on the membrane were fixed in 0.1ml of 0.25% (vol/vol) glutaraldehyde for 15 min (30). Freeprotein-binding sites were blocked by immersion in 0.1 Mglycine in TBS for 10 min at room temperature. The nitro-cellulose sheet was treated with rabbit anti-PAc serum. Theantibody bound to the immobilized antigens on the sheet wasdetected by solid-phase immunoassay with horseradish per-oxidase-conjugated goat anti-rabbit IgG (2). Quantitation ofthe color developed was done with an ACD-25-DX densi-tometer (Atto Co., Tokyo, Japan). The concentration of PAcin the samples was determined by comparison with a stan-dard curve of known concentrations of PAc from the culturesupernatant of S. mutans MT8148.Sodium dodecyl sulfate-polyacrylamide gel electrophoresis

and Western immunoblot. Proteins were electrophoresed in7.5% polyacrylamide gels as described by Laemmli (13). Thegels were stained with Coomassie brilliant blue or trans-ferred to a nitrocellulose membrane by the electrophoreticblotting technique (2). After transfer, the membrane wastreated with rabbit anti-PAc serum. The antibody bound tothe immobilized replica proteins on the membrane wasdetected by solid-phase immunoassay with horseradish per-oxidase-conjugated goat anti-rabbit IgG (2).RNA dot blot. S. lactis and S. mutans were grown in 100

ml of Todd-Hewitt broth (Difco) supplemented with 20 mMDL-threonine. The cultures were grown to a concentration ofapproximately 109 cells per ml. The cells were harvested bycentrifugation and suspended in 10 ml of brain heart infusionbroth supplemented with 30% (wt/vol) raffinose. The cellsuspensions were incubated with 2 mg of lysozyme per mlfor 30 min at 37°C and then with N-acetylmuramidase SG

INFECT. IMMUN.

on March 29, 2020 by guest

http://iai.asm.org/

Dow

nloaded from

IMMUNIZATION WITH RECOMBINANT S. LACTIS 2931

(0.1 mg/ml; Seikagaku Kogyo, Tokyo, Japan) for 30 min at37°C. The cells were then harvested by centrifugation andsuspended in 3 ml of 4 M guanidine thiocyanate containing0.05% (wt/vol) of N-lauroylsarcosine sodium salt, 0.1%(vol/vol) 2-mercaptoethanol, and 25 mM sodium citrate. Thecells were disrupted by drawing the suspension through an18-gauge needle. After the suspensions were clarified bycentrifugation, cesium chloride (1.2 g) was added and dis-solved. The solutions were then layered onto 1.4 ml of 5.7 Mcesium chloride containing 0.1 M ethylenediamine tetraace-tic acid (pH 7.0) in a polyallomer tube (Beckman Instru-ments, Inc., Fullerton, Calif.) and centrifuged at 30,000 rpmfor 16 h in an SW50.1 rotor (Beckman). The pelleted RNAwas dissolved in 0.4 ml of 0.3 M sodium acetate (pH 7.0) andprecipitated with 2.5 volumes of ethanol. The RNA prepa-rations were washed in 70% (vol/vol) ethanol and stored indistilled water at -70°C.RNA dot blot analysis was performed as described by

Thomas (34). The RNA preparations (5 ,ug) were incubatedin 0.01 M sodium phosphate buffer (pH 7.0) containing 1 Mglyoxal at 50°C for 1 h. Serial twofold dilutions of themixture were prepared in distilled water. Samples (0.2 ml)were placed into wells of the 48-well Bio-Dot SF microfil-tration apparatus with a sheet of nitrocellulose membraneequilibrated with 3 M sodium chloride and 0.3 M sodiumcitrate. The blot was dried, baked for 2 h at 80°C, and treatedwith 20 mM Tris hydrochloride (pH 8.0) for 5 min at 100°C.It was prehybridized and hybridized as described by Thomas(34). The 1.5-kb PstI fragment of the pac gene, which coversthe middle region of PAc, was radiolabeled as describedabove and used as the probe.

Estimation of plasmid stability. Strains of S. lactis and E.coli harboring plasmids were precultured overnight in GM17broth and LB broth (20) containing erythromycin (5 ,ug/ml)and chloramphenicol (10 ,ug/ml), respectively. A 0.2-miportion of overnight culture was seeded in 10 ml of mediumwith or without antibiotics. After overnight cultivation, thecultures were diluted 104- to 106-fold with sterile water andplated onto BL agar (Eiken Chemical Co., Tokyo, Japan, forS. lactis) or LB agar (for E. coli) plates. After replica platingof 100 colonies on the plates with or without antibiotics, thenumbers of colonies which were resistant and sensitive toantibiotics were scored.

Oral immunization. Male BALB/c mice, 8 weeks old, wereused. Before immunization, whole cells (109) of S. lactisIL1403(pSM1) and IL1403(pSA3) were suspended in 0.5 mlof 0.5% Formalin. Three consecutive daily intragastric im-munizations were performed for groups of mice (eight ornine animals per group) with intubation needles. After 4 and5 weeks, booster immunizations were done with the sameantigens.One week after the last challenge, saliva and serum

samples were collected. Mice were anesthetized intraperito-neally with pentobarbital, and pilocarpine-stimulated salivaand blood were subsequently collected. After clotting of theblood, serum was collected by centrifugation. These sampleswere stored at -70°C until use and were assayed by enzyme-linked immunosorbent assay (ELISA) for their titers of PAc-specific antibodies (36). Serial twofold dilutions of sampleswere incubated in ELISA plates coated with PAc purifiedfrom the culture supernatant of S. mutans MT8148. Theplates were subsequently treated with alkaline phosphatase-conjugated rabbit anti-mouse IgA, IgM, or IgG antibodiesfor the determination of titer of PAc-specific IgA, IgM, orIgG, respectively, followed by addition of p-nitrophenyl-phosphate substrate solution. The ELISA antibody titer was

13

kb

(1 , A_

6._

1.9-

FIG. 2. Agarose gel electrophoresis (A) and Southern hybridiza-tion (B) with the BamHI-SphI fragment of the plasmids from theEmr transformants. Lambda DNA digested with HindIlI was usedas molecular weight markers. Lanes 1, 2, and 3, Cesium chloride-purified plasmid DNA of pSM1 from S. lactis IL1403, pSM1 from E.coli HB101, and pSA3 from E. coli HB101, respectively; lanes 4 and5, BamHI- and SphI-digested pSM1 DNA from S. lactis IL1403(pSM1) and E. coli HB101, respectively; lanes 6, pSA3 from E. coliHB101 digested with BamHI and SphI; lanes 7, pPC41 digested withBamHI and SphI.

expressed as the reciprocal, log2 unit, of the highest dilutiongiving an optical density at 405 nm of 0.1 after developmentfor 1 h (serum) or 16 h (saliva) at room temperature.

Statistical analysis. Statistical comparison was performedby using Student's t test. Differences from the control valueswere considered significant if the P value was <0.05.

RESULTS

Construction of pSM1. The chimeric plasmid pPC41 con-tains the entire structural gene for PAc from S. mutansMT8148 (26). A BamHI-SphI fragment of pac gene, lackingthe 3'-terminal portion which presumably codes for themembrane anchor domain of PAc (27), was inserted betweenthe SphI and BamHI sites of an E. coli-S. lactis shuttlevector, pSA3, to inactivate its tetracycline resistance (Tcr)gene (Fig. 1). The resultant chimeric plasmid pSM1 thuscontained the pac gene, the erythromycin resistance (Emr)gene, and the chloramphenicol resistance (Cmr) gene.

Transformation of S. lactis IL1403 by the plasmid pSM1. S.lactis IL1403 was transformed with pSM1 by the protoplasttransformation method described by Simon et al. (31). Al-though the transformation efficiency in this study was lowerthan that reported by Simon et al. (31), over 20 Emrtransformants were obtained. All transformants tested har-bored a single plasmid identical to pSM1 by agarose gelelectrophoresis patterns (data not shown).

Southern blot analysis of pSM1 isolated from E. coli and S.lactis. The plasmid isolated from S. lactis IL1403 harboringpSM1 was digested with BamHI and SphI. The digestiongenerated two fragments of 6.2 and 10.4 kb (Fig. 2A, lane 4).The restriction fragments were subjected to Southern blot-ting and hybridized with the radiolabeled BamHI-SphI frag-ment of pPC41 containing the pac gene. Only the 6.2-kb

VOL. 58, 1990

on March 29, 2020 by guest

http://iai.asm.org/

Dow

nloaded from

2932 IWAKI ET AL.

TABLE 2. Production of PAc by S. lactis IL1403(pSM1)"

Amt of PAc in:Strain Cell homogenate Culture supernatant

(,ug/mg of dry wt) (,ug/ml of culture)

S. lactisIL1403 0 0IL1403(pSM1) 1.8 ± 0.3 0.002

S. mutans MT8148 10.9 ± 0.2 5.5 ± 0.6a Samples were blotted onto membrane filters with a Bio-Dot SF apparatus

(Bio-Rad). Rabbit anti-PAc serum was adsorbed with whole cells of S. lactisIL1403 before use. Each value represents the mean + standard deviation fortriplicate assays.

fragment hybridized to the probe (Fig. 2B). The restrictionand hybridization patterns of pSM1 from S. lactis IL1403(pSMl) were the same as those of pSM1 isolated from E.coli.

Expression of the pac gene in S. lactis. The amount of PAcproduced by S. lactis IL1403(pSM1) and S. mutans MT8148was measured with dot immunoblotting with anti-PAc anti-body. A large amount of the antigen was detected in culturesupernatant and cell homogenate of S. mutans MT8148(Table 2). A small amount of rPAc was detected in the cellhomogenate of S. lactis IL1403(pSMl) but not in the culturesupernatant of the recombinant organism. Cell homogenatesand cytoplasmic fractions of S. lactis IL1403 and IL1403(pSMl) were analyzed by sodium dodecyl sulfate-polyacryl-amide gel electrophoresis and Western blotting. PAc was notfound in the cell homogenate and cytoplasmic fraction of S.lactis IL1403 by sodium dodecyl sulfate-polyacrylamide gelelectrophoresis and Western blotting (Fig. 3). A smallamount of 190-kilodalton rPAc was detected in the cellhomogenate and cytoplasmic fraction of S. lactis IL1403(pSM1) (Fig. 3).

Transcription of the pac gene in S. lactis. To investigate the

A1 BI

4I 4

kDa

60-

4,

FIG. 3. Sodium dodecyl sulfate-poly-acrylamide gel electropho-resis (A) and Western blotting with anti-PAc antiserum (B) of cellfractions from S. lactis strains. Lanes 1, PAc (1 Rg) from S. mutansMT8148; lanes 2, cell homogenate (250 jig) of S. lactis IL1403; lanes3, cytoplasmic fraction (250 F.g) of S. lactis IL1403; lanes 4, cellhomogenate (250 ,ug) of S. lactis IL1403(pSM1); lanes 5, cytoplas-mic fraction (250 [.g) of S. lactis IL1403(pSM1). kDa, Kilodaltons.

oLIt|2

I 2S 5 L I 0

6.125 Lt

FIG. 4. Dot blot analysis of RNA isolated from S. lactisIL1403(pSM1) (lane 1) and S. mutans MT8148 (lane 2). Samples ofRNA (0.3125 to 5 ,ug) were placed into wells. The 1.5-kb PstIfragment of the pac gene was used as a probe.

transcription level of the pac gene in the recombinant S.lactis, the amount of PAc-specific mRNA was measured byRNA dot blot analysis with the radiolabeled 1.5-kb PstIfragment of the pac gene. The level of PAc-specific mRNAin S. lactis IL1403(pSM1) was four to eight times lower thanthat in S. mutans MT8148 (Fig. 4). This result indicates thatthe poor expression of the pac gene in S. lactis IL1403(pSM1) might be attributed to the poor transcription level ofthe gene in the host cell.

Stability of pSM1. The stability ofpSM1 in S. lactis IL1403was investigated after cultivation of the recombinant organ-ism with or without 5 ,ug of erythromycin per ml for 24 or 48h. The plasmid pSM1 as well as the vector pSA3 wererelatively stable in S. lactis IL1403 in the presence of 5 ,ug oferythromycin per ml (Table 3). Only 2 to 3% of the totalclones lost resistance to erythromycin, while 22 to 28% ofthe clones lost resistance in the absence of erythromycin.These results suggest that poor production of rPAc by S.lactis IL1403(pSM1) was not likely because of the loss ofplasmids during cultivation with erythromycin.Immunization of mice with the recombinant S. lactis. Al-

though the expression level ofpac in S. lactis IL1403(pSM1)was lower than that in S. mutans MT8148, the recombinantorganism was effective in inducing PAc-specific antibodies inmice when given intragastrically. The immunization sched-ule is shown in Fig. 5. Tables 4 and 5 show the production ofPAc-specific immunoglobulins in mice immunized with therecombinant organism. Mice immunized with Formalin-killed recombinant cells (109) induced significant salivaryIgA and serum IgG responses to PAc.

TABLE 3. Stability of plasmidsa

Erythromycin content Culture Ems coloniesStrain in culture medium period appearing

(,ug/ml) (h) (%)

S. lactisIL1403(pSM1) 0 24 25

0 48 275 24 3

IL1403(pSA3) 0 24 220 48 285 24 2

E. coli HB101(pSA3) 0 24 245 24 2

"Bacterial strains were cultured for the period indicated and plated ontosolid medium without antibiotics. After replica plating of 100 isolated colo-nies, Em' clones were counted.

INFECT. IMMUN.

on March 29, 2020 by guest

http://iai.asm.org/

Dow

nloaded from

IMMUNIZATION WITH RECOMBINANT S. LACTIS 2933

Age 8wk

fl?12wk 13wk 14wk

At

Triple daily Triple daily Singlechallenges challenges challenge

SacrificeSerumSaliva

FIG. 5. Immunization schedule. BALB/c mice were given For-malin-killed whole cells (109) of S. lactis IL1403(pSM1) andIL1403(pSA3) intragastrically. First and second immunizations wereeach carried out as three consecutive intragastric intubations, fol-lowed by one booster intubation (arrows).

DISCUSSIONS. lactis has been safely used for food fermentation

throughout history. The organism, therefore-, might be one ofthe most suitable hosts for oral vaccine production. In thisstudy, we constructed an S. lactis strain producing anexogenous protein, a surface protein antigen (PAc) of S.mutans, for development of an oral vaccine against dentalcaries. Recently, Curtiss et al. (4) have also constructedr-ecombinant Salmonella typhimurium strains expressing asurface protein antigen (SpaA) of S. sobrinus for oral immu-nization.

Formalin-killed whole cells of S. lactis IL1403(pSM1)proved an effective immunogen in mice. Oral administrationof the recombinant (109 cells) induced salivary IgA responseto PAc. Salivary IgA is reported to act as a protective agentagainst dental caries in experimental animals (22). Thus, therecombinant organism is expected to- be useful in oralvaccination against dental caries. Moreover, the recombi-nant organism was able to induce serum IgG responses toPAc, suggesting that the recombinant organism was alsoeffective in inducing systemic immune response. Challengestudies with S. mutans in the presence of high levels ofsucrose to measure colonization or caries prevention are inprogress.The 3'-truncated pac gene of S. mutans was expressed in

S. lactis IL1403. However, the amount of rPAc in the cellhomogenate of the recombinant strain was about sixfoldsmaller than that in the fraction of S. mutans MT8148. TheS. mutans-derived DNA fragment used in this study containsthe DNA sequence approximately 1.7 kilobase pairs up-stream from the initiation codon of the pac (27). Transcrip-tion of the pac gene may be controlled by this putative

TABLE 4. Serum PAc-specific antibody responses in BALB/cmice immunized orally with S. lactis IL1403(pSM1)

and IL1403(pSA3)a

mmunogenNo. of Log2 ELISA antibody titer of:mice IgA IgG IgM

None 9 <4 <4 <4IL1403(pSA3) 9 <4 4.3 ± 0.5 <4IL1403(pSM1) 8 <4 8.4 ± 0.1b 4.1 ± 0.1

a Samples were diluted 1:16, and the log2 ELISA antibody titer of eachclass of immunoglobulins was measured by the ELISA method (as describedin Materials and Methods) after developing at room temperature for 1 h.Values are means + standard errors.

b p < 0.05 versus the nonimmunized control group and IL1403(pSA3)group.

TABLE 5. Salivary PAc-specific antibody responses in BALB/cmice immunized orally with S. lactis IL1403(pSM1)

and IL1403(pSA3)a

No. of Log2 ELISA antibody titer of:Immunogen micemle IgA IgG- IgM

None 9 3.6 t 0.1 3.4 ± 0.2 3.3 ± 0.1IL1403(pSA3) 9 3.2 ± 0.1 3.1 ± 0.1 3.8 ± 0.3IL1403(pSM1) 8 6.3 ± 0.3b 3.1 ± 0.1 3.6 ± 0.1

" Samples were diluted 1:8, and the log2 ELISA antibody titer of each classof immunoglobulins was measured by the ELISA method (as described inMaterials and Methods) after developing at room temperature for 16 h. Valuesare means + standard errors.bp < 0.05 versus the nonimmunized control group and IL1403(pSA3)

group.

promoter sequence that might not work effectively in S.lactis. In fact, the amount of PAc-specific mRNA wassmaller than that of the wild-type PAc producer. Lack of 207base pairs and 69 amino acids at the 3'- and C-terminal ends,respectively, might also cause inefficient transcription, andwith the resultant immature mRNA, translation might alsobe inefficient. Whether the poor production of rPAc in therecombinant strain is due to promoter inefficiency or to thelack of a 3' end may be elucidated by measuring thepromoter strength in S. lactis in the future.PAc is known to exist both in a form apparently closely

associated with peptidoglycan of S. mutans strains and intheir culture supernatants (21, 24, 29). Because S. lactisIL1403(pSM1) was assumed to produce the protein antigenlacking the anchor domain, we expected that most of theantigen would be produced extracellularly. However, PAcwas not found in the culture supernatant, although pSM1 hada putative 5' signal sequence of the pac gene (27). Atpresent, there is no satisfactory explanation for this finding.

Experimental and manufacturing guidelines for organismscontaining recombinant genes to be taken orally in viableform have yet to be established. Therefore, we used therecombinant organism killed with Formalin in our immuni-zation experiments. Even after being killed, the organismhad enough antigenic activity to induce specific immuno-globulin production in mice via the oral immune system. Thepotency of this recombinant organism as an oral vaccinewhen administered in viable form will be evaluated in thefuture.

ACKNOWLEDGMENTS

This work was supported in part by the Japan Health SciencesFoundation (grant 1-5-2) and by a grant-in-aid for DevelopmentalScientific Research (grant 63870086) from the Ministry of Education,Science and Culture in Japan.

LITERATURE CITED1. Anderson, D. G., and L. L. McKay. 1983. Simple and rapid

method for isolating large plasmid DNA from lactic strepto-cocci. Appl. Environ. Microbiol. 46:549-552.

2. Burnette, W. N. 1981. "Western blotting": electrophoretictransfer of proteins from sodium dodecyl sulfate-polyacryl-amide gels to unmodified nitrocellulose and radiographic detec-tion with antibody and radioiodinated protein A. Anal. Bio-chem. 112:195-203.

3. Chopin, M.-C., A. Chopin, A. Rouault, and D. Simon. 1986.Cloning in Streptococcus lactis of plasmid-mediated UV resis-tance and effect on prophage stability. Appl. Environ. Micro-biol. 51:233-237.

4. Curtiss, R., III, R. M. Goldschmidt, N. B. Fletchall, and S. M.

VOL. 58, 1990

on March 29, 2020 by guest

http://iai.asm.org/

Dow

nloaded from

2934 IWAKI ET AL.

Kelly. 1988. A virulent Salmonella typhimurium Acya Acrp oralvaccine strain expressing a streptococcal colonization and vir-ulence antigen. Vaccine 6:155-160.

5. Dao, M. L., and J. J. Ferretti. 1985. Streptococcus-Escherichiacoli shuttle vector pSA3 and its use in the cloning of strepto-coccal genes. Appl. Environ. Microbiol. 49:115-119.

6. Forester, H., N. Hunter, and K. W. Knox. 1983. Characteristicsof a high molecular weight extracellular protein of Streptococ-cus mutans. J. Gen. Microbiol. 129:2779-2788.

7. Hamada, S., and H. D. Slade. 1980. Biology, immunology, andcariogenicity of Streptococcus mutans. Microbiol. Rev. 44:331-384.

8. Harlander, S. K. 1987. Transformation of Streptococcus lactisby electroporation, p. 229-233. In J. J. Ferretti and R. CurtissIII (ed.), Streptococcal genetics. American Society for Micro-biology, Washington, D.C.

9. Holt, R. G., Y. Abiko, S. Saito, M. Smorawinska, J. B. Hansen,and R. Curtiss III. 1982. Streptococcus mutans genes that codefor extracellular proteins in Escherichia coli K-12. Infect. Im-mun. 38:147-156.

10. Hughes, M., S. M. Machardy, A. J. Sheppard, and N. C. Woods.1980. Evidence for an immunological relationship betweenStreptococcus mutans and human cardiac tissue. Infect. Im-mun. 27:576-588.

11. Klaenhammer, T. R., and R. B. Sanozky. 1985. Conjugal trans-fer from Streptococcus lactis ME2 of plasmids encoding phageresistance, nisin resistance and lactose-fermenting ability: evi-dence for a high-frequency conjugative plasmid responsible forabortive infection of virulent bacteriophage. J. Gen. Microbiol.131:1531-1541.

12. Kondo, J. K., and L. L. McKay. 1984. Plasmid transformation ofStreptococcus lactis protoplasts: optimization and use in mo-lecular cloning. Appl. Environ. Microbiol. 48:252-259.

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

14. Lee, S. F., A. Progulske-Fox, and A. S. Bleiweis. 1988. Molecularcloning and expression of a Streptococcus mutans major surfaceprotein antigen, P1 (I/II), in Escherichia coli. Infect. Immun.56:2114-2119.

15. Lehner, T. 1985. Immunization against dental caries. Vaccine3:65-68.

16. Lehner, T., J. Caldwell, and R. Smith. 1985. Local passiveimmunization by monoclonal antibodies against streptococcalantigen I/II in the prevention of dental caries. Infect. Immun.50:796-799.

17. Lehner, T., M. W. Russell, J. Caldwell, and R. Smith. 1981.Immunization with purified protein antigens from Streptococcusmutans against dental caries in rhesus monkeys. Infect. Immun.34:407-415.

18. Loesche, W. J. 1986. Role of Streptococcus mutans in humandental decay. Microbiol. Rev. 50:353-380.

19. Ma, J. K.-C., R. Smith, and T. Lehner. 1987. Use of monoclonalantibodies in local passive immunization to prevent colonizationof human teeth by Streptococcus mutans. Infect. Immun.55:1274-1278.

20. Maniatis, T., E. F. Fritsch, and J. Sambrook. 1982. Molecularcloning: a laboratory manual. Cold Spring Harbor Laboratory,

Cold Spring Harbor, N.Y.21. McBride, B. C., M. Song, B. Krasse, and J. Olsson. 1984.

Biochemical and immunological differences between hydropho-bic and hydrophilic strains of Streptococcus mutans. Infect.Immun. 44:68-75.

22. Morisaki, I., S. M. Michalek, C. C. Harmon, M. Torii, S.Hamada, and J. R. McGhee. 1983. Effective immunity to dentalcaries: enhancement of salivary anti-Streptococcus mutans an-tibody responses with oral adjuvants. Infect. Immun. 40:577-591.

23. Ogundipe, J. O., and R. G. Holt. 1989. Molecular and immuno-chemical characterization of recombinant Escherichia coli con-taining the spaA gene region of Streptococcus sobrinus. Infect.Immun. 57:1906-1915.

24. Ohta, H., H. Kato, N. Okahashi, I. Takahashi, S. Hamada, andT. Koga. 1989. Characterization of a cell-surface protein antigenof hydrophobic Streptococcus mutans strain GS-5. J. Gen.Microbiol. 135:981-988.

25. Okahashi, N., T. Koga, and S. Hamada. 1986. Purification andimmunochemical properties of a protein antigen from serotype gStreptococcus mutans. Microbiol. Immunol. 30:35-47.

26. Okahashi, N., C. Sasakawa, M. Yoshikawa, S. Hamada, and T.Koga. 1989. Cloning of a surface antigen gene from serotype cStreptococcus mutans. Mol. Microbiol. 3:221-228.

27. Okahashi, N., C. Sasakawa, M. Yoshikawa, S. Hamada, and T.Koga. 1989. Molecular characterization of a surface proteinantigen gene from serotype c Streptococcus mutans, implicatedin dental caries. Mol. Microbiol. 3:673-678.

28. Russell, M. W., L. A. Bergmeier, E. D. Zanders, and T. Lehner.1980. Protein antigens of Streptococcus mutans: purificationand properties of a double antigen and its proteinase-resistantcomponent. Infect. Immun. 28:486-493.

29. Russell, R. R. B. 1979. Wall-associated protein antigens ofStreptococcus mutans. J. Gen. Microbiol. 114:109-115.

30. Russell, R. R. B., and K. Smith. 1986. Effect of subculturing onlocation of Streptococcus mutans antigens. FEMS Microbiol.Lett. 35:319-323.

31. Simon, D., A. Rouault, and M.-C. Chopin. 1986. High-efficiencytransformation of Streptococcus lactis protoplasts by plasmidDNA. Appl. Environ. Microbiol. 52:394-395.

32. Takahashi, I., N. Okahashi, C. Sasakawa, M. Yoshikawa, S.Hamada, and T. Koga. 1989. Homology between surface proteinantigen genes of Streptococcus sobrinus and Streptococcusmutans. FEBS Lett. 249:383-388.

33. Terzaghi, B. E., and W. E. Sandine. 1975. Improved medium forlactic streptococci and their bacteriophages. Appl. Microbiol.29:807-813.

34. Thomas, P. 1983. Hybridization of denatured RNA transferredor dotted to nitrocellulose membrane. Methods Enzymol. 100:255-285.

35. van der Vossen, J. M. B. M., J. Kok, and G. Venema. 1985.Construction of cloning, promoter-screening, and terminator-screening shuttle vectors for Bacillus subtilis and Streptococcuslactis. Appl. Environ. Microbiol. 50:540-542.

36. Voller, A., and D. Bidwell. 1986. Enzyme-linked immunosorbentassay, p. 99-109. In N. R. Rose, H. Friedman, and J. L. Fahey(ed.), Manual of clinical laboratory immunology, 3rd ed. Amer-ican Society for Microbiology, Washington, D.C.

INFECT. IMMUN.

on March 29, 2020 by guest

http://iai.asm.org/

Dow

nloaded from