Ot-at Microbioi Itntnutioi 1992: 7: 177-181

DNA probes for (detection of cariogenic Streptococcus mutans

M. Smorawinska, H. K. Kuramitsu Department of Pediatric Dentistry, University of Texas Heaitli Science Center, San Antonio, Texas, USA

Stnorawin.ska M, Kuratnitsu HK. DNA probes for detection of cariogenic Streptococcus tnutans. Oral Microbioi Ittttnitnot 1992: 7: 177-181. Streptococcus mutatis has been identified as the principal etiological agent in human dental caries, Thetefot-e, the specificity and sensitivity of 3 potential gene probes derived ft-om S. tnutans GS5 have been exatnined as potential reagents for detecting the cariogenic bacteria. The gene probes derived frotn the cloned gtfB, gtfD and///genes were exatnined in Southern blots with a panel of representative oral bacteria. The gtfB and ///genes wet-e appat-ently specific for S. tnutans under high-stt-ingency hybridization conditions. However, the gtfD gene fragtnent did cross-hybridize with the DNA IVotn other tnutans streptoeocci. These 1-esults suggest that gtfB and ///gene fragments tnay be used as specific probes for S. mutatis.

Dental caries is otie of the tnost cotntnon baeterial itifectious diseases afllictitig h u t n a n s (11), Streptococcus

tnutans

has been identified as the principal etiologic agent in this disease. Although there has been a significant declitie in the incidence of caries in the industrialized countries, i-oughly 20% of the populafion is affiicted with 60% of the cat-ious lesions (10). In addition, as tnore teeth are retained in adult populations thet-e is reason to believe that root surface carious lesions tnay t-etnain a significant ptoblein. In the face of the continuing declitie in the fi-equency of dental caries in childt-en, a rational approach toward further t-educing dental decay in the population should be based on identifying cat-ies-susceptible itidividuals rather than applying additional preventive tneasut-es on the genet-al populafion. Although tnany factot-s may be involved in caries susceptibility, salivary levels of 5, tnutans appear to provide a basis for identifying individuals at risk of developing caries (10), Curt-ently, convenient tnethods for quantifying S, tnutans levels in salivat-y satnples at-e not yet available, Howevet-, since DNA pt-obes can be used to quickly quantify pathogenic bacteria in clinical specimens (12), the developtnent of DNA probes specific for 5, mutans should pt-ove useful in this t-egard. Although DNA probes constructed from eht-otnosotnal DNA could be used to identify S. tnutans, the pt-esence of

other ot-al streptococci in human saliva (6) tnight lead to false-positive t-esults, Thet-efore, probes constructed frotn isolated S. mutans genes could obviate this probletn. Recently, several genes involved in extt-acellular polysacchat-ide synthesis have been isolated from the human ot-al isolate S. mutatis GS5 in E.sctiertchia coli: ftf coding for fructosyltransferase (FTF) (18), highly hotnologous gtfB and C genes coding for glucosyltransfet-ase-I (2) and glucosylti-ansferase-Sl (7) activities t-espeetively, and gtJT) coding for the glucosylti-ansferase-S enzytne (8), The present communication discusses the potential use of the///", and the gtfB and D genes as DNA probes for identifying and quantifying S. mutans. Material and methods Bacterial strains •(•

S. tnutatis ot-al isolates GS5 and UAIOI (frotn P. Caufield, Llniversity of Alabatna, Birtninghatn), Streptococcus sobritius 6715, Streptococcus gordonii Challis, Streptococcus rattus BHT, Streptococcus cricetus HS6 and E. coli HBlOl were tnaintained and grown as previously described (2) atid Bacillus sitbtitis was growti in LB tnediutn. DNA isolation

Chrotiiosomal DNA frotn 5. tnuians GS5 and other oral stt-eptococci was isolated as described pt-eviously (2). E. coti

Key words: Streptococcus tnutans, DNA probe; glucosyltransferase; fructosyltransferase H, K, Kuramitsu, Department of Pediatric Dentistry, University of Texas Health Science Center, San Antonio, TX 78284, USA Accepted for publication September 7, 1991

DNA was isolated by the standard lysozytne lysis procedure (15). B. subtitis DNA was isolated essentially as described for E, coli except that lysis was perfortned at 37X for 30 tnin and the lysate was treated with self-digested pronase (1.0 tng/tnl, Sigtna Chemical Co., St, Louis, MO) at 50"C for 30 tnin prior to phenol extt-action. Actinotttyces vi.scosus T14V chrotnosomal DNA was kindly supplied by Maria Yeutig (University of Texas, San Antonio). Porphyrotnonas gitigivatis ATCC 53977 DNA was provided by T, Kato of this laboratory. Plastnid DNA frotn E. coli stt-ains was isolated by the alkaline lysis procedut-e (15), DNA manipuiation

Digestion with t-estrietion endonucleases (Bethesda Research Laboratories, Gaithersbut-g, MD; Protnega Corp,, Madison, WI) was perfot-med as t-ecomtnended by the suppliers, DNA fragtnents wet-e purified frotn agarose gels by the GeneClean procedure as suggested by the ptovider (Bio 101, La Jolla, CA). DNA labeling

DNA ft-agments were labeled with 11deoxyuridine triphosphate (Biotin-11dUTP, 1.0 tnM, Bio-Rad Laboiatories, Richtnond, CA) or deoxycytidine triphosphate (dCT"P, 800 Ci/tntnol, Du Pont, Boston, MA) by the random

178

Smorawin.ska & Kuramitsu

priming procedure (15). To aehieve maximal Bio-11-dUTP ineorporation labeling was carried out as described by Bialkowska-Hobrzanska (3).

6

8

Hybridization

Southern blot transfers were routinely performed as described by Maniatis et al. (15). Hybridizations with biotinylated probes and posthybridization washes were carried out according to the protocol described for the BlueGene Nonradioacfive NA detection system (Bethesda Research Laboratories), After a final wash, hybridization reactions were visualized using streptavidinalkaline phosphatase conjugates and the chromogenie substrates nitroblue tetrazolium (NBT) and 5-bromo-4-chloro-3indolylphosphate (BCIP). Hybridizafions with "P-probes were also performed under high-stringency conditions (15). In each hybridization reaction lO*" cpm of DNA probe was added per tnl of hybridization solution. Preparation of DNA slot blots

S. mutans cells were lysed with mutanolysin as deseribed by Aoki et al, (2). DNA in the lysates was denatured by heating at 100 C for 10 min, mixed with an equal volume of 2 M atntnonium acetate and immediately filtered onto nitroeellulose membranes with a microfiltration tnanifold (Hybri-slot, Bethesda Research Laboratories). The membrane was air-dried and baked for 2 h at 80°C, The filter was then sequentially treated with RNase A (20 /(g/ml, Sigma) for 30 min at 37°C and with selfdigested pronase (1.0 mg/ml, Sigtna) for an additional 30 min in order to avoid nonspecific binding of the biotinylated DNA probe to cellular protein and RNA.

Eig. I. Southern blot analysis of chromosotiial DNA from representative oral bacterial species hybridized with the biotinylated 1,6 kb BatnHl fragment of the GS5 gtj'B gene, Cht-omosotiial DNA was cleaved with EcoR\, except P. gingivaiis, which was cleaved with Hitidlll. Lanes: I, S. tnutans GS5; 2, S. .salivarius HK; 3, S, sobritius 6715; 4. S. gordonii Challis; 5, A. viscosus T14V; 6, E. coli HBlOl; 7, P gitigivaiis ATCC 53977; 8, HitidUl digest of latnbda DNA (23,1, 9,4, 6,5, 4,3, 2.3, 2,0, 0,5 kb from the top).

When a probe cotitaining the 1,6 kb BamHI fragment from the gtj'B gene of S. tnutans GS5 isolated from plastnid pTS20 (2) was used in Southern blot analysis with chromosomal DNA frotn

a nutnber of cotntnon ot-al bacteria (Fig. 1), it was obset-ved that the probe was specific for S. tnutans under these hybridization conditions. It was of ititet-est that no cross-hybridization was deteet-

6

Results Specificity of the gtf probes

One of the unique characteristics of S, mutans strains is their ability to exhibit sucrose-dependent colonization of tooth surfaces (6), This property depends on the synthesis of water-insoluble glucans catalyzed by the glucosyltransferases (GTF) elaborated by these organisms (11). Therefot-e, one potential DNA probe which could prove highly specific for S, mutatis might be a fragment from one of the gtf genes present on the chromosotne of these organisms.

Eig. 2. Southern blot analysis of chromosomal DNA fiotri lept-esentativc otal bacteria hybridized with the biotinylated 739 bp FcoR\ lVagment of the GS5 /(/gene, Chrotnosomal lDNA was digested with EcoRl, except P. gingivalis, which was digested with BaniH\. Lanes: 1. S. sobritius 6715; 2, S. gordonii Challis; 3, S. salivarius; 4, P. gingivali.s; 5, S. mutatis UAlOl; 6, •S, mutatis GS5,

S. mutans probes ed with chromosotnal DNA from S. sobrinus 6715, which also synthesizes insoluble glucan (5), Another gtf getie involved in glucan synthesis by S. mutans is the gtfD getie, which codes for the GTF-S enzytne synthesizing pritnarily water-soluble glueans (8), When a DNA fragtnent frotn this getie was used in a Southet-n blot analysis with chrotnosomal DNA from representative oral baeteria, it was observed that the probe reacted only with mutans streptococcal (19) DNA (Fig. 5) but not with the other organistns tested (data not shown). No cross-hybridization was observed with DNA from S. gordotiii, a species also capable of synthesizitig water-soluble glueans (9). Specificity of the ftf probe

Several ot-al microorganistns including S. mutatis are able to synthesize fructose polymers collectively called fruetans (6), Therefore, it was of interest to detet-mine whether the/// getie from S. tnutans GS5 could be used as a specific probe for these organisms. Southern blot analysis (Fig. 2) itidicated that the ///pt-obe was specific for only S. tnutans among the organistns tested. In addition, this probe did not react with 2 other bacteria capable of synthesizing fructatis. A, viscosus and B. subtilis (data not shown). Therefore, the /// probe appears to be as specific for S. mutans as the gtj'B probe.

1 2

3

4

5

/•7i,', 4. Southern blot analysis of chromosotnal DNA frotn different mutans streptococci hybridized with the biotinylated 739 bp fragtrtent of the/r/gene, Chrotnosotnal DNA was digested with EcoR]. Lanes: 1, HindUl digest of lambda DNA; 2, S, mutatis GS5; 3, S. rattus BHT; 4, S. cricetus HS6; 5, S. sobrinus 6715,

12

3

4

Eig. 5. Southern blot analysis of chrotnosotnal DNA from diffetent tnutans streptococci hybridized with the biotinylated L77 kb fragtnent of thegZ/T) gene, Laties: 1, HiiidUl digest of lambda DNA; 2, 5, .sobritius 6715; 3, S, rattus BHT; 4, S, cricetus HS6; 5, S, tnutans GS5,

Eig 3. Southern blot analysis of chrotnosomal DNA from different tnutans stt-eptococci hybt-idized with the biotinylated 1,6 kb BamHI fragtnent of the gtfB gene. Chromosomal DNA was digested with EeoRl. Lanes: 1, S. .sobrinus; 2, S. cricetus HS6; 3, S. rattus BHT; 4, S. tnutatis GS5; 5, Hind\\\ digest of latnbda DNA,

179

(Fig, 3) and///(Fig. 4) probes appears to be specific for S. tnutatis serotype c strains sitice no cross-hybridization was detected with DNA from S. rattus BHT (serotype b), S, cricetus HS6 (serotype a), or S, sobrinus 6715 (serotype g). By contrast, the gtfD probe reacted with all of the mutatis streptococci tested (Fig. 5). Identical patterns were observed for S, cricetus and 5. sobritius but those for S, ttiutans and S. rattus were distinet. Therefore, both the gtfB and / / / probes could be used to distinguish S. mutatis from the 3 other major serotypes of mutans streptococci. Sensitivity of DNA probes

It was of interest to detennine the sensitivity of the DNA probes, sinee 5, tnutam'-specific probes could possibly be used in the clinieal setting. Therefore, the probes tnust be able to deteet levels of this organistn that at-e not-tnally present in clinical specimens such as saliva. Caries-susceptible individuals harbor > 10' S. niutanslval of saliva atid salivary samples from nonsusceptible subjects contain around 10'' or less S. mutans/ml (10), When various levels of purified S. tnutans GS5 chromosomal DNA were probed with the 1.6 kb gtfB DNA fragment (Fig. 6), it was observed that > 3 ng of DNA eould be readily detected under the conditions used. In addition, levels as low as 300 pg could be barely deteeted. No significant difference was deteeted when either "P or biotinylated probes were used. When chromosomal DNA equivalent to that present in lO'-lO' GS5 cells (20) was extracted and probed with the gtfB fragment, it was observed that DNA frotn > lO*" cells could be readily detected with either "P or biotinylated probes but that DNA frotn 2x10' cells was weakly detected. Discussion

The identification and quantification of pathogenic bacteria frotn clinical specimens is eritical for both the diagnosis Specificity of the ftf and gtf probes among and prevetitioti of itifeetious diseases mutans streptococci (12). For bacterial infections, tradidonal S. ttiutatis is otie tnetnber of a group of identificatioti proeedures involving oral stt-eptococci known collectively as growth of the organisms are both labormutans streptococci that share sitnilar ititensive and titne-consuming. The reproperties (II), Thet-efot-e, it was of cent development of DNA probes for interest to determine whether the gtf identifying pathogenic bacteria (3, 4, 16) and / / / probes frotn serotype c GS5 should help to obviate these litnitations. could be used to readily discriminate In addition, several probes specific for between these organistns. Both the gtfB potential oral bacterial pathogens have

180

St7iorawinska & Kuramitsu

B

A 1

2

1

3

Eig. 6. Slot blot hybridization of S, tnutans GS5 chromosotnal DNA with the 1,6 kb BamHI fragment from the gtJM gene. A, Probe labeled with biotin-11-dUTP, Lanes: 1, Whole cell DNA preparations: a, 10' cells; b, 10'' cells; c, 2 x 10' cells. 2, purifted S. mutans DNA: a, 30 ng; b, 3 ng; c, 300 pg, B, Probe labeled with ('-P) dCTP Lanes: 1, purified S. mutans DNA; a, 30 ng; b, 3 ng; c, 300 pg, 2, Whole cell DNA preparations: a, 10' cells; b, 10'' cells; c, 2 x 10' cells.

been developed recetitly and show promise for rapid identification of these organisms (17). Sitice a variety of different approaches have implicated S. tnutans as the principle etiologic agent in hutnan dental caries (11) and salivary levels of these organisms appear to be a reasonable predictor of caries susceptibility (10), the quantification of these organisms in saliva samples may serve to identify candidates for anti-caries prophylaxis. However, quantification of S. mutans usually requires the growth of saliva samples on selective mediutn under anaerobie conditions (10). Therefore, it is unlikely that such measurements could be carried out conveniently in the dentist's office. By contrast, detection kits based on DNA probes are now being developed for routine identificafion of pathogetiic baeteria in the clinical setting. Although differentiation of S. tnutans from other streptococci based on chromosomal DNA probes has been previously reported (13), it is likely that probes based on individual gene fragments should be more specific. The present results indicate that use of DNA probes based on the gtJH and ftf genes of S, mutans GS5 ean readily distinguish S. mutans frotn other common oral microorganistns, including S. .sobrinus and A. viscosus, whieh express sitnilar enzymes. Since S. sobrinus strains also express a GTF-I enzytne that shares approximately 60% amino acid sequence

tration in 1 tnl of saliva tnay be near the tht-eshold for discritninating between high and low risk of caries susceptibility (10), the sensitivity of these S. tnutans probes would need to be increased sotnewhat before use iti the clinical setting. In addition, the effects of salivary components on such deteetion systetns needs to be evaluated. However, because of the rapid development of DNA probes for diagnostic purposes, it is likely that the t-equired sensitivity can be obtained by itnproving the detection systetn. Therefore, it is t-easonable to assutne that it should be possible to develop S. tnutatis detection systetns based on DNA probes for use in dental pracfice.

Acknowledgement

This investigation was supported in part by grants DE03258 and DE09039 frotn the National Institutes of Health,

similarity with the g(fB gene product (1), it was of interest that under the References relatively stringent hybridization con1, Abo H, iVlatsutnura T, Kodatna T et al, ditiotis used in the current investigation Peptide sequences for sucrose splittitig the gtJB probe did not react with S, and glucan binding within Streptococcus sobritius DNA (Fig, 1). Therefot-e, this sobrinus glucosyltransferase (watet--insolprobe could be used to readily distinuble glucan synthetase), ,1 Bactet-iol 1991: guish between S, mutans and S. sobrittus 173: 989-996, in dental plaque satnples. Such a distinc2, Aoki H, Shiroza T, Hayakawa M. Sato tion has not always been made in asS. Kuratnitsu HK, Cloning of a Streptosessing the role of mutans streptococci coccus tnutans gene coding for insoluble in dental caries inifiation at different glitcan synthesis. Infect Itntnun 1986: 53: 587-594, sites on human teeth (11). 3, Bialkowska-Hobrzanska H, Detection of The inability of the gtJD gene to hyenterotoxigenic Eseheriehia coli by dotbridize with 5, gordonii DNA as well as blot hybridization with biotinylated the lack of reactivity of the///"getie with DNA probes, J Clin Mict-obiol 1987; 25: A. viscosus DNA probably indicates 338-343, relatively low sequence similarity be4, Danbara H, Identification of entetotoxitween these comparable genes. Howgenic Escherictiia eoli by colony hybridiever, the sequetice of the genes coding zation using biotitiylated LTlh, STla, and STIb entetotoxin probes. In: Macafor the GTF-S activity in S. gordonii as rio AJL, de Macario EC, ed, Getie well as the FTF activity of A. viscosus probes for bacteria. San Diego; Acahave yet to be reported, demic Press, 1990: 167-178, Sinee it is likely that the use of DNA 5, Fukui K, Moriyatiia T, Miyake Y, Mizutprobes in the clinical setting will involve ani K, Tanaka O, Purificatioti and propnonradioaetive rather than ''P-labeled erties of glucosyltransferase tesponsible probes, it was of interest that the relafor water insoluble glucan synthesis frotn tive sensifivities of the biotin and -"PStreptococcus tnutans. Infect Imtnun labeled probes used in the pt-esent inves1982: 37: 1-9, tigation were generally sitnilar (Fig, 6), 6, Hatnada S, Slade HD, Biology, itntnunology, and cariogenieity oiSteptocoecus However, as noted previously (20), the mutans. Mictobiol Rev 1980: 44: t-adiolabeled probe was somewhat tnore 331-384, sensitive than biotitiylated probes. 7, Hanada N, Kutatnitsu HK, Isolation Under the present conditions, 2 x 10' 5. and characterization of the Streptococcus tnutatis cells were weakly detected by mutans gtfC gene, coding for synthesis either probe. This is similar to the sensiof both soluble and insoluble glucans. vitity of other detection systems based Infect Itntnun 1988: 56: 1999-2005, on gene probes (14). Since this concen8, Hanada N, Kuratnitsu HK., Isolation

S. mutans probes

9,

10, 11,

12,

13,

and chatacterization of the Streptococcus mutatis gtJT> gene, coditig for pritner-dependent soluble glucan synthesis. Infect Immun 1989: 57: 2079-2085, Huang S, Lee HC, Mayer RM, The purification and pi-operties of dextransucrase frotn Streptococcus sanguis ATCC 10558, Carbohydr Res 1979: 74: 287-300, Krasse B, Caries risk, Chicago: Quintessetice Publishing, 1985, Loesehe WJ, Role oi Streptococcus mutans in hutnan dental decay, Microbioi Rev 1986: 50: 353-380, Lowe JB, Clinical application of gene probes in hutnan genetic disease, tnalignaticy, and infectious disease, Clin Chim Acta 1986: 157; 1 32, Lutnb SM. Russell RRB, Identification

14,

15,

16,

17,

of oral stt-eptococci by DNA probes, J Dent Res 1989: 68, p, 929, Malouin F, Bt-yan LE, Sliewciw P et al, DNA pt-obe technology for rapid detection of Haetnophitus itijtuetizae in clinical specitnens, j Clin Mictobiol 1988: 26: 2132-2138, Maniatis T, Ft-itsch EF, Satnbtook J, Molecular cloning: a laboratory tnanual. Cold Spt-ing Hatbot-, NY: Cold Spring Hat-bor Laboratoty, 1989, Moseley SL, Echeverria P. Serivvatana J et al, Identiftcation of enterotoxigenic Escherictiia coii by colony hybridization using thtee enterotoxin gene pt-obes, J Infect Dis 1982: 145: 863-869, Russell RRB, Getietic analysis and genetic probes for oral bacteria. In; Fergu-

181

son DB, ed. Aspects of oral tnolecular biology. Basel: Karger. 1991: 57-76, 18, Sato S, Kuratnitsu HK, Isolation and chai-actet-ization of a fructosyltt-ansferase gene from Streptococcus tnutans GS-5, Infect Immun 1986: 52: 166-170, 19, Shitnatnura A, Tsumot-i H, Mukasa H, Three kinds of extracellular glucosyltransferases frotn Streptococcus mutatis 6715 (serotype g), FEBS Lett 1983: 157: 79-81, 20, Wethetall BL, Johnson AM, Nucleic acid probes for Caniphylobactcr species. In: Macat-io AJL. de Macario EC, ed. Gene probes for bacteria, San Diego; Acadetnic Press. 1990: 255-293,