Articulo6 gene cloning of glucosyltransferase

INFECTION AND IMMUNITY,0019-9567/00/$04.0010

May 2000, p. 2475–2483 Vol. 68, No. 5

Copyright © 2000, American Society for Microbiology. All Rights Reserved.

Purification, Characterization, and Molecular Analysis of theGene Encoding Glucosyltransferase from Streptococcus oralis

TAKU FUJIWARA,1* TOMONORI HOSHINO,1 TAKASHI OOSHIMA,1 SHIZUO SOBUE,1

AND SHIGEYUKI HAMADA2

Departments of Pedodontics1 and Oral Microbiology,2 Osaka University Faculty of Dentistry,Suita-Osaka 565-0871, Japan

Received 22 November 1999/Returned for modification 21 December 1999/Accepted 19 January 2000

Streptococcus oralis is a member of the oral streptococcal family and an early-colonizing microorganism inthe oral cavity of humans. S. oralis is known to produce glucosyltransferase (GTase), which synthesizes glucansfrom sucrose. The enzyme was purified chromatographically from a culture supernatant of S. oralis ATCC10557. The purified enzyme, GTase-R, had a molecular mass of 173 kDa and a pI of 6.3. This enzyme mainlysynthesized water-soluble glucans with no primer dependency. The addition of GTase markedly enhanced thesucrose-dependent resting cell adhesion of Streptococcus mutans at a level similar to that found in growing cellsof S. mutans. The antibody against GTase-R inhibited the glucan-synthesizing activities of Streptococcus gordoniiand Streptococcus sanguis, as well as S. oralis. The N-terminal amino acid sequence of GTase-R exhibited nosimilarities to known GTase sequences of oral streptococci. Using degenerate PCR primers, an 8.1-kb DNAfragment, carrying the gene (gtfR) coding for GTase-R and its regulator gene (rgg), was cloned and sequenced.Comparison of the deduced amino acid sequence revealed that the rgg genes of S. oralis and S. gordoniiexhibited a close similarity. The gtfR gene was found to possess a species-specific nucleotide sequence corre-sponding to the N-terminal 130 amino acid residues. Insertion of erm or aphA into the rgg or gtfR gene resultedin decreased GTase activity by the organism and changed the colony morphology of these transformants. Theseresults indicate that S. oralis GTase may play an important role in the subsequent colonizing of mutansstreptoccoci.

Streptococci formerly classified as Streptococcus sanguishave recently been subclassified into at least five distinct ge-netic groups. These groups have been assigned the speciesnames S. sangius sensu stricto, Streptococcus gordonii, Strepto-coccus oralis, Streptococcus mitis, and Streptococcus parasanguis(14, 15) and are collectively called sanguis (group) strepto-cocci. These streptococci are early-colonizing microorganismsin the oral cavity of neonates as well as on adult cleaned toothsurfaces (17). The distribution of these species varies amongoral sites and changes as dental plaque matures (6, 23). Incontrast, mutans streptococci colonize the oral cavity only afterthe eruption of teeth (8).

Mutans streptococci and Streptococcus salivarius (29) havemultiple glucosyltransferases (GTases) encoded by multiple gtfgenes, e.g., gtfB, gtfC, and gtfD, in Streptococcus mutans (10,16). These enzymes synthesize water-soluble and/or -insolubleglucans from sucrose. They contribute to the development ofdental plaque and, eventually, to the initiation of dental caries.Recent studies indicate that adhesive glucan is synthesizedfrom sucrose in concert with these GTases in S. mutans (7).

S. oralis, S. gordonii, and S. sanguis are known to possessGTases and produce extracellular polysaccharide from sucrose(36). However, only a limited number of investigations ofGTase from sanguis group streptococci have been performed.Recently, the gene encoding S. gordonii strain Challis GTase(gtfG) has been cloned and sequenced (35), and a regulatorygene, rgg, has been described as a positive transcriptional reg-ulator (30, 31, 33). Similar positive regulatory functions have

been identified in the rgg gene of S. pyogenes (3) as well asLactococcus lactis (27).

Since S. oralis is an earlier colonizer in the oral flora (6, 25,32), the infection and colonization of mutans streptococci maybe affected by the presence of S. oralis, and the glucan synthe-sized by S. oralis GTase may function as a substratum foradhesion of the bacteria. In addition, the prevalence of sanguisgroup streptococci was found to be different between caries-active and caries-inactive individuals (24).

In this study, we purified a GTase protein from S. oralis anddetermined its immunochemical properties and contributionto the sucrose-dependent cellular adherence of S. mutans. Inaddition, a gene encoding S. oralis GTase (designated gtfR)and its regulatory gene (rgg) were cloned and sequenced.

MATERIALS AND METHODS

Bacterial strains and growth media. S. oralis ATCC 10557 was used in most ofthe experiments. For comparison, S. oralis SK23 and ATCC 9811, S. sanguisATCC 10556, ST3, and ST7, S. gordonii ATCC 10558, F90A, and SK51, S. mitisSK24 and ATCC 903, S. mutans MT8148, S. sobrinus 6715, and S. salivarius HHTwere selected from our culture collection. Organisms were routinely cultured inbrain heart infusion (BHI) broth (Difco Laboratories, Detroit, Mich.) or mitissalivarius (MS) agar (Difco). Escherichia coli XL-2 (Stratagene Ltd., Cambridge,United Kingdom) was cultured in Luria-Bertani (LB) medium aerobically.Erythromycin, kanamycin, and ampicillin (Wako Pure Chemicals, Osaka, Japan)were added to LB medium to produce final concentrations of 500, 30, and 100mg/ml, respectively. Erythromycin (5 mg/ml) and kanamycin (250 mg/ml) wereadded to MS agar for selection of the S. oralis transformants.

Preparation of glucosyltransferases. S. oralis ATCC 10557 was cultured in 5liters of dialyzed TTY medium (12) at 37°C to an optical density of 0.8 at 550 nm.The culture supernatant was collected by centrifugation and adjusted to 60%saturation with ammonium sulfate. The precipitate was dissolved in 10 mMsodium phosphate buffer (NaPB) (pH 6.5) and then dialyzed against the samebuffer. The crude sample was applied to a Q Sepharose FF (Pharmacia BiotechAB, Uppsala, Sweden) column (bed volume, 10 ml) and eluted with a lineargradient of 0 to 1.0 M NaCl in the same buffer. Active fractions were pooled,dialyzed against 10 mM potassium phosphate buffer (KPB) (pH 6.0), applied to

* Corresponding author. Mailing address: Department of Pedodon-tics, Osaka University Faculty of Dentistry, 1-8 Yamadaoka, Suita-Osaka, 565-0871, Japan. Phone: 81-6-6879-2962. Fax: 81-6-6879-2965.E-mail: [email protected].

2475

a Bio-Scale CHT10-I column (bed volume, 10 ml; Bio-Rad Laboratories, Her-cules, Calif.), and then eluted with a 10 to 500 mM KPB linear gradient.

GTase samples from other streptococci were obtained from the culture su-pernatants of test strains by 50% saturation ammonium sulfate precipitation.Cell-associated GTase (CA-GTase) was extracted from centrifuged cells of S.mutans with 8 M urea followed by ammonium sulfate precipitation (11).

Generation of antiserum. Antisera were prepared by repeated intramuscularinjections of rabbits with the purified GTase from S. oralis ATCC 10557 sus-pended in Freund’s complete adjuvant (Difco) followed by immunization withthe antigen suspended in Fruend’s incomplete adjuvant (Difco). The antibody toS. oralis GTase was purified from rabbit antiserum by repeated 33% saturationwith ammonium sulfate.

Glucan synthesis assay. GTase activity was determined using [glucose-14C]sucrose with or without primer dextran T10, as described previously (11). Briefly,reaction mixtures composed of GTase, 10 mM [glucose-14C]sucrose (11.47 GBq/mmol), and 0 or 20 mM dextran T10 in 20 ml of 50 mM KPB (pH 6.0) were

incubated for 1 h at 37°C, spotted on a filter paper square (1.5 by 1.5 cm), anddried in air. The filters were washed with methanol or distilled water and thenimmersed in scintillation fluid to estimate the amount of total [14C]glucan orwater-insoluble [14C]glucan. Kinetic constants were determined by Lineweaver-Burk analyses of the glucan synthesis rates.

Determination of pI and optimum pH. The pI was determined by analyticalisoelectric focusing using a PhastSystem (Pharmacia) with a PhastGel IEF3-9(Pharmacia). After electrophoresis, the gel was incubated for 1 h at 37°C in 10mM NaPB (pH 6.5) containing 5% sucrose, 2% Triton X-100, and 0.05% NaN3.The enzyme activity was visualized by periodic acid-Schiff staining. The optimumpH of GTase was determined by measuring the GTase activity in 50 mM KPB(pH 5.0 to 7.5).

SDS-PAGE and Western blotting. Sodium dodecyl sulfate-polyacrylamide gelelectrophoresis (SDS-PAGE) and Western blot analyses were carried out asdescribed previously (9). Briefly, GTase samples and E. coli cells carrying therecombinant plasmid were suspended in SDS gel-loading buffer (26) and boiledfor 5 min. Proteins separated by SDS-PAGE were transferred onto a polyvinyli-dene difluoride membrane (Immobilon; Millipore). After being blocked with 5%bovine serum albumin, the membrane was reacted with the rabbit antibody to S.oralis GTase at 37°C for 1 h, and the antibody which was bound to the proteinband(s) was detected by a solid-phase immunoassay.

Effects of S. oralis GTase on the sucrose-dependent adhesion of S. mutansresting cells. S. mutans strain MT8148 cells grown in BHI broth were washed at0°C with 0.1 M KPB (pH 6.0) containing 0.05% NaN3. The centrifuged cells wereresuspended in the same buffer containing 1% sucrose and then adjusted to anoptical density of 1.0 at 550 nm. Aliquots (3 ml) of the cell suspension were mixedwith various amount of S. oralis GTase and incubated at 37°C for 18 h at a 30°angle. Next, the culture tubes were vigorously vibrated with a Vortex mixer for3 s. The degree of cell adhesion was determined by reading the optical density at550 nm and expressed as the percentage of total cell mass. To assess the adhesionof S. mutans growing cells, the organism was grown at 37°C for 18 h at a 30° anglein BHI broth containing 1% sucrose. The percent adhesion was determined asdescribed above.

Amino acid sequence. S. oralis GTase was subjected to SDS-PAGE and blottedonto ProBlott membranes (Applied Biosystems, Foster City, Calif.). The GTaseband was excised from several lanes and subjected to sequencing using an ABI477A/120A protein sequencer (Applied Biosystems).

FIG. 1. Chromatographic purification of GTase-R from S. oralis ATCC10557. (A) Separation of ammonium sulfate-precipitated GTase (60% satura-tion) by anion-exchange chromatography on a Q Sepharose FF column (bedvolume, 10 ml). GTase was eluted with a linear gradient of 0 to 0.3 M NaCl. (B)Further purification of GTase-R containing fractions from the elution profileshown in panel A on a Bio-Scale CHT10-I column (bed volume, 10 ml). Elutionwas done with a 10 to 500 mM KPB linear gradient. A280, optical density at 280nm.

FIG. 2. Physical characteristics of GTase-R. (A) SDS-PAGE of GTase prep-arations at different stages of purification. Lanes: 1, culture supernatant; 2,ammonium sulfate precipitate; 3, pooled active fractions from Q-ion-exchangechromatography; 4, pooled active fraction from CHT-10 hydroxylapatite chro-matography; M, molecular mass markers. (B) Isoelectric focusing-PAGE ofGTase-R. Lane: 1, pI markers (3.50 to 8.65); 2, Purified GTase-R visualized byPAS staining. (C) Effect of pH on GTase activity.

TABLE 1. Purification of S. oralis GTase

Preparationa step Total amt of protein(mg)

Total GTase activity(U)

GTase sp act(U/mg)

Recovery(%)

Purification(fold)

Culture supernatant 736 140 0.19 100 1Ammonium sulfate precipitation 180 117 0.65 84.0 3.4Q Sepharose fraction 5.5 4.0 0.72 2.9 3.8CHT-I fraction 0.3 2.4 8.00 1.7 42.0

a S. oralis ATCC 10557 was grown in 5 liters of dialyzed TTY medium to an optical density of 0.8 at 550 nm. The culture supernatant was concentrated by a 60%saturation of ammonium sulfate. The enzyme fraction was purified on a Q Sepharose FF column followed by a Bio-Scale CHT10-I column.

2476 FUJIWARA ET AL. INFECT. IMMUN.

DNA manipulations. Restriction enzymes, ligase, and other DNA-modifyingenzymes were purchased from New England Biolabs (Beverly, Mass.) or Takara(Kyoto, Japan). Manipulations of DNA with these enzymes were performed asrecommended by the manufacturers. All other DNA manipulations were carriedout using standard protocols (26).

Chromosomal DNA isolation and Southern blot analysis. Organisms weregrown in BHI broth for 18 h at 37°C, collected, and then washed by centrifuga-tion. Cells (750 mg [wet weight]) were suspended in 5 ml of 50 mM NaCl–10 mMTris-HCl (pH 7.4) and then digested with mutanolysin (0.25 mg/ml; DainipponPharmaceutical Co., Osaka, Japan) for 1 h at 50°C, and N-lauroyl sarcosine (finalconcentration, 1.5%) and EDTA (final concentration, 10 mM) were added tolyse the cells. The lysate was treated with RNase (0.3 mg/ml; Wako) and pro-teinase K (0.3 mg/ml; Merck, Darmstadt, Germany). The DNA was purified fromthe cell lysate by phenol and phenol-chloroform extractions and then collected byethanol precipitation.

Southern blot analysis was carried out as a standard procedure. Briefly, chro-mosomal DNA from the test organisms was digested with EcoRI, separated byelectrophoresis on a 0.8% agarose gel, and transferred onto a nylon membrane(Hybond-N; Amersham, Little Chalfont, United Kingdom). Next, the DNA wascross-linked to the membrane by UV radiation. A 397-bp DNA fragment cor-responding to positions 54 to 186 in the deduced amino acid of the gtfR gene wasamplified by PCR and used as a probe. The membrane was then hybridizedstringently with the 32P labeled probe.

PCR. PCR was performed in reaction mixtures containing 50 mM KCl, 10 mMTris-HCl (pH 8.3), 1.5 mM MgCl2, 200 mM deoxyribonucleoside triphosphate,1.0 mM primer, template DNA (,10 ng/ml), and AmpliTaq Gold DNA poly-merase (0.025 U/ml; Applied Biolystems). Amplification was performed in aGene AmpPCR System 2400 apparatus (Perkin-Elmer) as specified by the man-ufacturer. Degenerate PCR was performed as follows: a preincubation step at95°C for 9 min followed by 30 cycles of a denaturation step at 94°C for 30 s, aprimer-annealing step at 36°C for 30 s, and an extension step at 60°C for 30 s.Long PCR was performed using a TaKaPa LA PCR kit Ver 2.1 (Takara), asrecommended by the manufacturer.

Cloning and sequencing of the GTase gene. Two sets of genomic libraries wereconstructed by cloning EcoRI- or KpnI-digested S. oralis ATCC 10557 chromo-somal DNA into plasmid pMW119 (Nippon Gene) or pUC19 (Takara) and thentransforming them into competent E. coli. In addition, a vector named pGEM-TEasy (Promega, Madison, Wis.) was used for cloning of the PCR products.

For a DNA-sequencing template, plasmid DNA and PCR products wereprepared using a Wizard Plus Minipreps DNA purification system (Promega)and a Centricon 100 spin column (Millipore, Bedford, Mass.), respectively. Thedideoxy dye termination reaction was performed with an ABI PRISM cycle-sequencing kit (Perkin Elmer) in a GeneAmp 2400 thermal cycler. The productswere then analyzed using an automated DNA sequencer model 373 (AppliedBiosystems). The homology search, multiple-sequence alignment, and phyloge-netic tree creation were performed with the BLAST, FASTA, and CLUSTAL Wprograms on the DDBJ “supernig” computer system.

Expression of recombinant GTase. E. coli carrying a recombinant plasmid wasgrown in LB broth (3 ml) to an optical density of 0.6 at 550 nm. Cells werecollected by centrifugation, suspended in 100 ml of 10 mM NaPB (pH 6.0), anddisrupted by sonication. The sonic supernatant was then separated and examinedfor glucan synthesis.

Transformation of S. oralis. S. oralis was subjected to transformation as re-ported previously (9). Briefly, the recipient organisms were cultured in Todd-Hewitt broth (Difco) supplemented with 10% heat-inactivated horse serum(Gibco, Grand Island, N.Y.) for 18 h. The culture was diluted 1:40 with the broth(10 ml) and then incubated for another 1.5 h at 37°C, and the donor DNA wasadded to a final concentration of 25 mg/ml. The culture was further incubated for2 h, concentrated approximately 10-fold by centrifugation, and then spread onMS agar plates containing antibiotics. The plates were incubated in a CO2incubator for 2 to 3 days at 37°C, and possible transformant colonies were pickedup for further examinations.

Construction of the insertional mutants. Recombinant plasmid pYT303 orpYT311 carrying the 830- or 1,070-bp fragment of the erythromycin resistancegene (erm) from pVA838 (20) or the kanamycin resistance gene (aphA) fromtransposon Tn1545 (2) was used. A subclone, pTHR8, carrying a 1.5-kb SphI-PstIinsert containing the rgg gene was generated from pTH171. A 2.5-kb DNAfragment containing the center portion of the gtfR gene was amplified by PCRand cloned to generate pTH808. pTHR8 or pTH808 was restricted with ApaI orHindIII to be linear at a unique site. The linear plasmid was then blunted andligated with the erm or aphA cassette to yield pTHR805 or pTH818. After beingmade linear at the unique PstI site, the plasmid was introduced into S. oralisATCC 10557 by transformation to allow an allelic exchange.

Statistical analysis. Differences between S. oralis GTase concentrations and S.mutans resting-cell adhesion were determined by analysis of variance with sub-sequent use of the Tukey-Kramer multiple-comparisons test. Significance levelswere taken at P , 0.01.

FIG. 3. Effects of GTase-R on the cellular adhesion of S. mutans. S. mutanscells grown in BHI broth were resuspended at 0°C in 0.1 M KPB (pH 6.0) with0.05% NaN3 containing 1% sucrose. The cell suspension was mixed with increas-ing amounts of GTase-R and incubated at 37°C for 18 h at a 30° angle (restingcells). As a positive control, S. mutans was grown in BHI broth with 1% sucroseat 37°C for 18 h at a 30° angle (growing cells). Numbers of adhesive cells werethen determined. Data are expressed as means and standard deviations of trip-licate experiments. The asterisks indicate statistical significance (P , 0.01) fromthe sucrose value for resting cells incubated in sucrose containing KPB withoutGTase-R.

TABLE 2. Effects of anti-S. oralis ATCC 10557 GTase antibody onthe GTase activities of various oral streptococci

GTase origina

GTase activity (dpm)b

% InhibitionWithout anti-GTase-R

With anti-GTase-R

S. oralisATCC 10557 4,930.3 6 12.8 831.8 6 14.2 83.1SK23 4,997.9 6 49.7 1,065.1 6 42.2 78.7ATCC 9811 5,022.0 6 37.6 1,128.8 6 36.7 77.5

S. sanguisATCC 10556 5,000.0 6 100.0 1,305.4 6 59.5 73.9ST3 5,027.1 6 25.8 1,244.4 6 36.1 75.3ST7 5,053.5 6 21.8 1,243.8 6 29.6 75.4

S. gordoniiATCC 10558 4,975.8 6 37.9 949.4 6 23.3 80.990A 4,950.2 6 32.9 1,468.6 6 30.9 70.3SK51 4,956.5 6 40.9 1,007.3 6 58.9 79.7

S. mutansMT8148 cell-associated 4,996.4 6 53.3 4,687.9 6 63.8 6.2MT8148 cell-free 4,964.7 6 72.5 3,431.3 6 61.6 30.9

S. sobrinus6715 5,006.9 6 138.4 3,732.1 6 194.9 25.5

S. sobrinusHHT 4,989.9 6 66.9 3,662.2 6 76.5 26.6

a GTase was from the concentrates of test strain culture supernatants, exceptfor the cell-associated GTase from S. mutans. The cell-associated GTase wasextracted from the cells of S. mutans with 8 M urea.

b The GTase fraction (1 mU) was reacted with the antibody to GTase-R (32 mgof protein) at 37°C for 30 min or left unreacted. The reaction mixture was thenincubated with [glucose-14C]sucrose at 37°C for 1 h, and the amount of synthe-sized [14C]glucan was measured. Data are expressed as means 6 standard devi-ations of triplicate experiments.

VOL. 68, 2000 GLUCOSYLTRANSFERASE OF S. ORALIS 2477

Nucleotide sequence accession numbers. The nucleotide sequences of the rggand gtfR gene have been deposited in the DDBJ database under accession no.AB025228.

RESULTS

Purification of S. oralis GTase. GTase was purified from theculture supernatant of S. oralis ATCC 10557 by ammoniumsulfate precipitation followed by anion-exchange and hydroxy-lapatite chromatography (Fig. 1). The recovery of the purifiedGTase preparation was 1.7%, and the degree of purificationwas 42-fold. The specific activity of the purified enzyme,

GTase-R, was 8.0 mU/mg of protein (Table 1). SDS-PAGE ofGTase-R gave a single protein band with a molecular mass of173 kDa. The optimum pH and pI values were 6.5 and 6.3,respectively (Fig. 2). The Km value was determined to be 2.49mM. Glucan synthesized by GTase-R from sucrose was largelywater soluble (89.7%), and its production was not enhanced inthe presence of the primer dextran T10.

Immunological properties of GTase-R. Western blot analy-ses revealed that the rabbit antibody to GTase-R reactedstrongly with GTase preparations from other sanguis strepto-cocci and cell-free GTase (CF-GTase) but not with CA-GTasefrom S. mutans. Further, the enzyme activity of GTase-R wasmarkedly inhibited by the antibody to GTase-R. The antibodystrongly inhibited S. sanguis and S. gordonii GTase, as well asS. mutans CF-GTase. S. sobrinus and S. salivarius GTases wereonly weakly inhibited, while S. mutans CA-GTase was notaffected by the antibody (Table 2). The inhibition of glucansynthesis exhibited a similar pattern to the reactivity whenanalyzed by Western blotting.

Effects of GTase-R on the adhesion of S. mutans. The effectsof GTase-R on the sucrose-dependent adhesion of S. mutansresting cells are shown in Fig. 3. Cells incubated withoutGTase-R adhered to the glass surface only loosely, and ap-proximately 60% of the cells were easily removed by vibration.However, the addition of a small amount of GTase-R (1 mU/ml) resulted in firm adhesion of the S. mutans cells. Thisadhesion was as strong as that of the S. mutans cells grown insucrose-containing medium.

Amino acid sequencing of GTase-R. The N-terminal aminoacid sequence of GTase-R was determined to be DDVKQVVVQEPATAQTSGPGQQ. This sequence did not show anysimilarity to other reported sequences, including GTases fromS. gordonii and other oral streptococci, by BLAST and FASTAhomology searching.

Cloning of the GTase-R gene (gtfR) by PCR with degenerateprimers. A schematic diagram of the GTase-R gene and clon-ing strategy is shown in Fig. 4. PCR was done using the de-generated oligonucleotides corresponding to the N-terminalamino acid sequence VKQVVV (59GTNAARCARGTNGTNGT39, forward primer) and GPGQQ (59YTGYTGNCCNGGNCC39, reverse primer). A 60-bp gene fragment was clonedinto a pGEM-T Easy vector, resulting in the recombinant plas-mid pTHN01. The sequence of the insert of pTHN01 wasconsistent with the N-terminal amino acid sequence. Then a60-mer oligonucleotide primer (59GTAAAGCAGGTTGTAGTTCAAGAACCTGCTACAGCTCAGACTAGTGGTCCCGGTCAGCAA39) was synthesized and used for hybridiza-tion. The E. coli transformants carrying the EcoRI-digested

FIG. 4. Genetic map and cloning strategy of the rgg and gtfR genes.

FIG. 5. Western blot analysis and glucan synthesis activity of recombinantGTase-R. (A) The recombinant proteins and native GTase-R were separated bySDS-PAGE and blotted onto a polyvinylidene difluoride membrane. The blotwas reacted with the antibody to GTase-R. (B) The sonic supernatants of E. colicells carrying the recombinant plasmid were used for the glucan synthesis assay.GTase activity was determined as the amount of [14C]glucan synthesized from[glucose-14C]sucrose. Data are expressed as means and standard deviations oftriplicate experiments.


insert were screened by colony hybridization with this primer.The recombinant plasmids pTH121 and pTH171, with 1.4- and6.1-kb S. oralis chromosomal inserts, respectively, were iso-lated.

A sequenced analysis of pTH171 revealed that an openreading frame (ORF) composed of the 861-bp nucleotide waspresent, and an incomplete reading frame without the termi-nation codon was identified 89 bp downstream of the ORF.This frame encoded the N-terminal amino acid sequence ofGTase-R; however, the molecular mass of the recombinantprotein expressed by pTH171 was only 128 kDa (Fig. 5). Toclone the C-terminal region, the KpnI-digested library wasscreened and pTH181, carrying a 1.8-kb insert with a termina-tion codon, was obtained. The complete nucleotide sequenceof the gtfR gene was determined by reconstruction of the DNAsequences from the inserts of pTH171 and pTH181. To con-firm the sequence, primers for amplification of the whole gtfRgene were synthesized and long PCR amplification was per-formed. The PCR product was cloned into a pGEM-T Easyvector to yield pTH275. The recombinant protein expressed bypTH275 had a molecular mass of 175 kDa and glucan-synthe-sizing activity (Fig. 5).

Nucleotide sequences. The nucleotide sequence determinedin this study is shown in Fig. 6. The ORF in pTH171 wascomposed of 861 bp and encoded a polypeptide of 287 aminoacids. This gene exhibited a high degree of homology (74%) tothe rgg gene of S. gordonii (accession no. M89776), which has

been reported to be a regulatory gene of GTase. We thereforedesignated it rgg. A multiple alignment of the deduced aminoacid sequence of rgg revealed that the rgg gene of S. oralisexhibited a 76% homology to the S. gordonii rgg gene, whereasthe homology of S. oralis rgg to the S. pyogenes and L. lactis rgggenes was only 21%.

The gtfR gene was composed of 4,728 bp, which encoded apolypeptide composed of 1,575 amino acids with a predictedmolecular mass of 177 kDa and a pI of 5.58. The alignment ofthe deduced amino acid sequence of the gtfR gene and the gtfGgene encoding S. gordonii GTase is shown in Fig. 7. The gtfRgene displayed 79.9% homology to the gtfG gene. The N-terminal sequence from amino acids 55 to 186, deduced fromthe nucleotide sequence data of gtfR, was completely differentfrom that of other streptococcal species. These findings weresupported by Southern hybridization analyses, which indicatedthat no hybridized bands were detected, except for strains of S.oralis, when the PCR-amplified DNA fragment correspondingto the N-terminal 140 amino acid residues of gtfR was used asa probe (Fig. 8).

Inactivation of the rgg or gtfR gene. Inactivation of the chro-mosomal rgg or gtfR gene of S. oralis ATCC 10557 was per-formed by insertion of the erm or aphA gene. The colonymorphology and glucan synthesis activity of representativetransformants are shown in Fig. 9. The rgg mutant had a flatand dull appearance without the zooglealic zone, while the gtfRmutant had smaller colonies with a more transparent appear-

FIG. 6. DNA sequence and deduced amino acid sequence of the rgg and gtfR genes. The putative promoter (210 and 235) and ribosome binding sites (SD) areshown. The sequence corresponding to the identified N-terminal amino acid sequence is marked. Regions of dyad symmetry are indicated by arrows.


FIG. 7. Alignment of the deduced amino acid sequences of GtfR and GtfG of S. gordonii (accession no. U12643). Identical amino acid sequences are indicated byshaded boxes.


ance. The GTase activity of both mutants had decreased toabout 10% that of the parent strain.

DISCUSSION

Several methods of comparing the amino acid sequences ofGTases from various oral streptococci have revealed thatGTase possesses two functional domains. One is a catalyticdomain that is composed of approximately 800 amino acidresidues located in the N-terminal region, while the other is aglucan binding domain that includes a large number of re-peated units in the C-terminal region. The former contains

common putative active-site peptides involved in sucrose hy-drolysis (22, 28). Molecular modeling analyses have suggestedthat the core region contains a cyclically permuted form of the(a/b)8-barrel structure (4, 19). Recently, circular dichroismanalysis has verified the presence of that structure (21). Directrepeats of the glucan binding domain are also found in theC-terminal region of the ligand binding proteins in some gram-positive organisms, including S. mutans glucan binding protein,Clostridium difficile ToxA and ToxB, and lysins from S. pneu-moniae and its bacteriophage (37). Structure-function relation-ship studies have revealed that GTases synthesizing water-insoluble and/or water-soluble glucans exhibit an almostidentical amino acid sequence at the putative active sites. The

FIG. 7—Continued.


putative active sites of gtfR are the same as those of geneencoding GTases that synthesize water-soluble glucan (28)(Fig. 7). Deletion of glucan binding domain direct repeatsaffects the activity and localization of GTase (13), as well as itsglucan production (1). Our finding that the recombinant pro-tein of pTH171 lacking the C-terminal region of gtfR did notexhibit GTase activity (Fig. 5) accords with the finding re-ported by Vickerman et al. (34) using S. gordonii GTase.

It seems quite clear from the GTase antibody inhibition datathat there are three groups of organisms with similar levels ofinhibition (Table 2). The enzymes from S. oralis, S. sanguis, andS. gordonii strains are all inhibited by 75 to 80%; those from S.sobrinus, S. salivarius, and the cell-free enzyme from S. mutansare all inhibited by 25 to 30%; and the cell-associated enzymefrom S. mutans is inhibited by 6%. These differences in inhi-bition should reflect differences in the nucleotide sequence ofGTase gene.

Southern blot analysis indicated that the N-terminal 130amino acid residues were conserved exclusively in S. oralis (Fig.8). Thus, this region is thought to be a species-specific se-quence for S. oralis. Classification of sanguis streptococci hasbeen difficult. In fact, DNA-DNA hybridization studies haverevealed that many strains which had previously been identi-fied phenotypically as S. mitis, S. oralis, or S. sanguis were notcorrectly classified (5). Moreover, 16S rRNA sequencing anal-

ysis has indicated that S. mitis, S. pneumoniae, and S. oralisexhibited .99% homology in nucleotide sequencing (14).Thus, the 59 region of gtfR can be used as a useful probe inPCR amplification for the rapid and exact classification of S.oralis.

S. oralis also possessed the rgg gene immediately upstream ofgtfR. The presence of an rgg-like gene in strains of S. oralis andS. sanguis was previously reported (33). Moreover, the de-duced amino acid sequences of the rgg gene from S. oralis andS. gordonii were very similar. An S. oralis mutant strain inwhich the rgg gene was inactivated displayed a soft-colonyphenotype and markedly reduced GTase activity (Fig. 9).These results indicate that the rgg gene of S. oralis is a positivetranscriptional regulator of gtfR. Similar findings have beenreported for S. gordonii (31). However, the putative RNAsecondary structures at the junction of the rgg and gtf genes inS. oralis were different from those in S. gordonii. These resultsclearly indicate that the regulatory mechanism of the rgg geneis different for S. gordonii and S. oralis.

It is interesting that the colony morphologies of our rgg andgtfR mutants were different, even though both mutants exhib-ited minimal GTase activity (Fig. 9). Recently, rgg-like geneshave been identified in some bacterial species; the rgg gene inS. pyogenes positively regulates the expression of cysteine pro-teinase (3, 18), while that of L. lactis has been claimed to beglutamate-g-aminobutyrate antiporter and glutamate decar-boxylase (27). Those two test strains have no GTase, and it isof interest to know if the rgg-like genes regulated the expres-sion of proteins other than GTase. These findings, coupledwith the results of the present study, may suggest that the rgggene of S. oralis can regulate a gene(s) other than gtfR.

Sucrose-dependent adhesion is an important pathogenictrait of mutans streptococci. S. mutans produces three GTases:GTase-I, GTase-SI, and GTase-S, which are encoded by thegtfB, gtfC, and gtfD genes, respectively. GTase-I is present inassociation with the cell surface, whereas GTase-S is releasedextracellularly. GTase-SI can be coextracted from cells by 8 Murea treatment and is more likely to play an important role incellular adhesion (9). S. mutans adheres firmly to solid surfacesby the cooperative action of these GTases in the presence ofsucrose in vivo. However, firm cellular adhesion was obtainedonly when S. mutans was grown in a sucrose-containing brothmedium. In this study, we revealed that S. mutans resting cellsfirmly adhered to a glass surface in the presence of S. oralisGTase and sucrose (Fig. 3). The maximum adhesion was al-most equivalent to that of growing S. mutans cells in terms ofadhesion strength and macroscopic features. However, thepresence of excess amounts of GTase-R resulted in a surfeit ofsoluble-glucan synthesis, which in turn may interfere with thecell adhesion of S. mutans and cariogenic dental-plaque for-mation. These results indicate that S. oralis GTase may play asignificant role in the formation of dental plaque in vivo.

Further, the evidence reported here suggests that S. oralisGTase strongly contributes to the establishment of oral bacte-rial biofilms, and therefore a more precise description of itsmechanism should be sought.

ACKNOWLEDGMENTS

We thank Toshiyuki Miyata (National Cardiovascular Center Re-search Institute, Suita-Osaka, Japan) for technical assistance with theamino acid sequence.

This work was supported in part by a grant-in-aid from the Ministryof Education, Science and Sports of Japan (11470451).

REFERENCES1. Abo, H., T. Matsumura, T. Kodama, H. Ohta, K. Fukui, K. Kato, and H.

Kagawa. 1991. Peptide sequences for sucrose splitting and glucan binding

FIG. 8. Southern blot analysis of chromosomal DNA from various oral strep-tococci digested with EcoRI. As a probe, the PCR-amplified DNA fragmentcorresponding to the N-terminal sequence of gtfR was used. Lanes: 1, S. mutansMT8148; 2, S. sobrinus 6715; 3, S. salivarius HHT; 4, S. sanguis ATCC 10556; 5,S. sanguis ST3; 6, S. sanguis ST7; 7, S. oralis ATCC 10557; 8, S. oralis SK23; 9, S.oralis ATCC 9811; 10, S. gordonii ATCC 10558; 11, S. gordonii F90A; 12, S.gordonii SK51; 13, S. mitis SK24; and 14, S. mitis ATCC 903.

FIG. 9. Colonial morphology and GTase activity of S. oralis ATCC 10557(A), and its erythromycin-resistant rgg-deficient (B) and kanamycin-resistantgtfR-deficient (C) mutants. GTase activity was determined by the synthesis of[14C]glucan from [glucose-14C]sucrose. Data are expressed as means 6 standarddeviations of triplicate experiments.


within Streptococcus sobrinus glucosyltransferase (water-insoluble glucansynthetase). J. Bacteriol. 173:989–996.

2. Caillaud, F., C. Carlier, and P. Courvalin. 1987. Physical analysis of theconjugative shuttle transposon Tn1545. Plasmid 17:58–60.

3. Chaussee, M. S., D. Ajdic, and J. J. Ferretti. 1999. The rgg gene of Strepto-coccus pyogenes NZ131 positively influences extracellular SPE B production.Infect. Immun. 67:1715–1722.

4. Devulapalle, K. S., S. D. Goodman, Q. Gao, A. Hemsley, and G. Mooser.1997. Knowledge-based model of a glucosyltransferase from the oral bacte-rial group of mutans streptococci. Protein Sci. 6:2489–2493.

5. Ezaki, T., Y. Hashimoto, N. Takeuchi, H. Yamamoto, S. L. Liu, H. Miura, K.Matsui, and E. Yabuuchi. 1988. Simple genetic method to identify viridansgroup streptococci by colorimetric dot hybridization and fluorometric hy-bridization in microdilution wells. J. Clin. Microbiol. 26:1708–1713.

6. Frandsen, E. V., V. Pedrazzoli, and M. Kilian. 1991. Ecology of viridansstreptococci in the oral cavity and pharynx. Oral Microbiol. Immunol. 6:129–133.

7. Fujiwara, T., S. Kawabata, and S. Hamada. 1992. Molecular characteriza-tion and expression of the cell-associated glucosyltransferase gene fromStreptococcus mutans. Biochem. Biophys. Res. Commun. 187:1432–1438.

8. Fujiwara, T., E. Sasada, N. Mima, and T. Ooshima. 1991. Caries prevalenceand salivary mutans streptococci in 0-2-year-old children of Japan. Commu-nity Dent. Oral Epidemiol. 19:151–154.

9. Fujiwara, T., M. Tamesada, Z. Bian, S. Kawabata, S. Kimura, and S.Hamada. 1996. Deletion and reintroduction of glucosyltransferase genes ofStreptococcus mutans and role of their gene products in sucrose dependentcellular adherence. Microb. Pathog. 20:225–233.

10. Fujiwara, T., Y. Terao, T. Hoshino, S. Kawabata, T. Ooshima, S. Sobue, S.Kimura, and S. Hamada. 1998. Molecular analyses of glucosyltransferasegenes among strains of Streptococcus mutans. FEMS Microbiol. Lett. 161:331–336.

11. Hamada, S., T. Horikoshi, T. Minami, N. Okahashi, and T. Koga. 1989.Purification and characterization of cell-associated glucosyltransferase syn-thesizing water-insoluble glucan from serotype c Streptococcus mutans.J. Gen. Microbiol. 135:335–344.

12. Hamada, S., and M. Torii. 1978. Effect of sucrose in culture media on thelocation of glucosyltransferase of Streptococcus mutans and cell adherence toglass surfaces. Infect. Immun. 20:592–599.

13. Kato, C., and H. K. Kuramitsu. 1991. Molecular basis for the association ofglucosyltransferases with the cell surface of oral streptococci. FEMS Micro-biol. Lett. 63:153–157.

14. Kawamura, Y., X. G. Hou, F. Sultana, H. Miura, and T. Ezaki. 1995. De-termination of 16S rRNA sequences of Streptococcus mitis and Streptococcusgordonii and phylogenetic relationships among members of the genus Strep-tococcus. Int. J. Syst. Bacteriol. 45:406–408.

15. Kilian, M., L. Mikkelsen, and J. Henrichsen. 1989. Taxonomic study ofviridans streptococci: description of Streptococcus gordonii sp. nov. andemended descriptions of Streptococcus sanguis (White and Niven 1946),Streptococcus oralis (Bridge and Sneath 1982), and Streptococcus mitis (An-drewes and Horder 1906). Int. J. Syst. Bacteriol. 39:471–484.

16. Kuramitsu, H. K. 1993. Virulence factors of mutans streptococci: role ofmolecular genetics. Crit. Rev. Oral Biol. Med. 4:159–176.

17. Long, S. S., and R. M. Swenson. 1976. Determinants of the developing oralflora in normal newborns. Appl. Environ. Microbiol. 32:494–497.

18. Lyon, W. R., C. M. Gibson, and M. G. Caparon. 1998. A role for triggerfactor and an rgg-like regulator in the transcription, secretion and processingof the cysteine proteinase of Streptococcus pyogenes. EMBO J. 17:6263–6275.

19. MacGregor, E. A., H. M. Jespersen, and B. Svensson. 1996. A circularly

permuted a-amylase-type a/b-barrel structure in glucan-synthesizing glu-cosyltransferases. FEBS Lett. 378:263–266.

20. Macrina, F. L., R. P. Evans, J. A. Tobian, D. L. Hartley, D. B. Clewell, andK. R. Jones. 1983. Novel shuttle plasmid vehicles for Escherichia-Streptococ-cus transgeneric cloning. Gene 25:145–150.

21. Monchois, V., J. H. Lakey, and R. R. B. Russel. 1999. Secondary structure ofStreptococcus downei GTF-I glucansucrase. FEMS Microbiol. Lett. 177:273–248.

22. Mooser, G., S. A. Hefta, R. J. Paxton, J. E. Shively, and T. D. Lee. 1991.Isolation and sequence of an active-site peptide containing a catalytic aspar-tic acid from two Streptococcus sobrinus alpha-glucosyltransferases. J. Biol.Chem. 266:8916–8922.

23. Nyvad, B., and M. Kilian. 1987. Microbiology of the early colonization ofhuman enamel and root surfaces in vivo. Scand. J. Dent. Res. 95:369–380.

24. Nyvad, B., and M. Kilian. 1990. Microflora associated with experimental rootsurface caries in humans. Infect. Immun. 58:1628–1633.

25. Pearce, C., G. H. Bowden, M. Evans, S. P. Fitzsimmons, J. Johnson, M. J.Sheridan, R. Wientzen, and M. F. Cole. 1995. Identification of pioneerviridans streptococci in the oral cavity of human neonates. J. Med. Microbiol.42:67–72.

26. Sambrook, J., E. F. Fritsch, and T. Maniatis. 1989. Molecular cloning: alaboratory manual, 2nd ed. Cold Spring Harbor Laboratory Press, ColdSpring Harbor, N.Y.

27. Sanders, J. W., K. Leenhouts, J. Burghoorn, J. R. Brands, G. Venema, andJ. Kok. 1998. A chloride-inducible acid resistance mechanism in Lactococcuslactis and its regulation. Mol. Microbiol. 27:299–310.

28. Shimamura, A., Y. J. Nakano, H. Mukasa, and H. K. Kuramitsu. 1994.Identification of amino acid residues in Streptococcus mutans glucosyltrans-ferases influencing the structure of the glucan product. J. Bacteriol. 176:4845–4850.

29. Simpson, C. L., N. W. Cheetham, P. M. Giffard, and N. A. Jacques. 1995.Four glucosyltransferases, GtfJ, GtfK, GtfJ and GtfM, from Streptococcussalivarius ATCC 25975. Microbiology 141:1451–1460.

30. Sulavik, M. C., and D. B. Clewell. 1996. Rgg is a positive transcriptionalregulator of the Streptococcus gordonii gtfG gene. J. Bacteriol. 178:5826–5830.

31. Sulavik, M. C., G. Tardif, and D. B. Clewell. 1992. Identification of a gene,rgg, which regulates expression of glucosyltransferase and influences the Sppphenotype of Streptococcus gordonii Challis. J. Bacteriol. 174:3577–3586.

32. Tappuni, A. R., and S. J. Challacombe. 1993. Distribution and isolationfrequency of eight streptococcal species in saliva from predentate and den-tate children and adults. J. Dent. Res. 72:31–36.

33. Vickerman, M. M., M. C. Sulavik, and D. B. Clewell. 1995. Oral streptococciwith genetic determinants similar to the glucosyltransferase regulatory gene,rgg. Infect. Immun. 63:4524–4527.

34. Vickerman, M. M., M. C. Sulavik, P. E. Minick, and D. B. Clewell. 1996.Changes in the carboxyl-terminal repeat region affect extracellular activityand glucan products of Streptococcus gordonii glucosyltransferase. Infect.Immun. 64:5117–5128.

35. Vickerman, M. M., M. C. Sulavik, J. D. Nowak, N. M. Gardner, G. W. Jones,and D. B. Clewell. 1997. Nucleotide sequence analysis of the Streptococcusgordonii glucosyltransferase gene, gtfG. DNA Seq. 7:83–95.

36. Willcox, M. D., M. Patrikakis, and K. W. Knox. 1995. Degradative enzymesof oral streptococci. Aust. Dent. J. 40:121–128.

37. Wren, B. W. 1991. A family of clostridial and streptococcal ligand-bindingproteins with conserved C-terminal repeat sequences. Mol. Microbiol.5:797–803.

Editor: E. I. Tuomanen


Health & Medicine

Articulo6 gene cloning of glucosyltransferase