Populations of Spp+ bacteria (strain Challis) on sucrose media switch to a soft noncohesive ... in human subacute bacterial endocarditis, however, is com-.
Vol. 57, No. 12
INFECTION AND IMMUNITY, Dec. 1989, p. 3945-3948
0019-9567/89/123945-04$02.00/0 Copyright © 1989, American Society for Microbiology
Spontaneous Switching of the Sucrose-Promoted Colony Phenotype in Streptococcus sanguis GINETTE TARDIF,lt MARK C. SULAVIK,2 GARTH W. JONES,1 AND DON B. CLEWELL3* Department of Biologic and Materials Sciences, School of Dentistry,2 Department of Microbiology and Immunology, School of Medicine,1 and The Dental Research Institute,3 The University of Michigan, Ann Arbor, Michigan 48109-0402 Received 19 June 1989/Accepted 30 August 1989
Streptococcus sanguis on media containing 3% sucrose gives rise to characteristic hard cohesive colonies (designated Spp+). Populations of Spp+ bacteria (strain Challis) on sucrose media switch to a soft noncohesive phenotype (designated Spp-) at a frequency of 10-4 to 10-3. Spp- bacteria switch back to Spp+ bacteria at a similar frequency. Successive rounds of Spp variation were observed. The Spp phenotypic switch was associated with changes in extraceliular glucosyltransferase activity. Streptococcus sanguis is one of the first bacteria to colonize newly erupted or freshly cleaned tooth surfaces (3, 30). Although strains of S. sanguis have been shown to induce low levels of fissure caries in gnotobiotic rats fed high-sucrose diets (8, 9), they are not considered significantly cariogenic in humans. The involvement of S. sanguis in human subacute bacterial endocarditis, however, is common (1), and the onset of such infections sometimes can be related to transient bacteremias deriving from dental trauma. The production of extracellular polysaccharides (i.e., glucans and fructans) by S. sanguis, as well as other oral streptococci such as Streptococcus mutans, is well known. The enzymes which use sucrose (i.e., glucosyltransferases [GTFs] and fructosyltransferases [FTFs]) to generate such polymers have been identified and at least partially characterized (4, 5, 10, 12-16, 18, 20, 22, 28). Extracellular glucan formed by S. mutans appears to enhance colonization on tooth surfaces in vivo, whereas the role of extracellular glucan produced by S. sanguis is not clear (11). When grown on solid media containing high concentrations of sucrose (i.e., 3 to 5%), colonies of S. mutans and S. sanguis exhibit characteristic colony morphologies. In the case of S. sanguis, the colonies are usually hard and cohesive and adhere firmly to the surface of the agar. It is presumed that glucan production is necessary for this colony type (2, 6, 23). One characteristic of the phenotype is that colonies do not exhibit their hard cohesive trait when >0.05% dextran (MW = 2 x 106) is included in the media. In approaching a genetic characterization of this phenomenon in S. sanguis Challis (24), we noticed that variants which did not exhibit the hard cohesive colony characteristic of the sucrose-promoted phenotype (Spp) appeared at a frequency of approximately 10-4 to 10-3. When grown on Todd-Hewitt broth (THB; Difco Laboratories) agar containing 3% sucrose, these variant (Spp-) colonies, in contrast to the hard cohesive (Spp+) colonies, were soft and easily removed from the agar surface and had a flatter and duller appearance under a stereoscopic microscope. When representative Spp- colonies were subcultured overnight in THB and cultivated again on THB agar containing sucrose, the cultures maintained their Spp- phenotype but gave rise to revertant Spp+ colonies at a frequency of about 10-4 to Corresponding author. t Present address: Biotechnology Research Institute, Montreal, Quebec, Canada H4P 2R2. *
10'. Frequencies of phase variations from the wild-type Challis strain to Spp- variants and from the Spp- strain, CH97, to Spp+ strains were reproducible (Table 1). Repeated rounds of Spp variation were also apparent (Table 1). The Spp phase variation was also observed in S. sanguis G9B (data not shown). The production of extracellular GTF and FTF in culture supernatants of both Spp+ and Spp- derivatives was examined. Two methods were used to analyze GTF and FTF: the ['4C-glucose]sucrose incorporation procedure of Kuramitsu to detect total extracellular GTF activity (soluble and insoluble glucan-synthesizing activity [18]) and activity gel assays (21, 26, 32) to detect both extracellular GTF and FTF activities. Batch cultures of Spp+ and Spp- Challis derivatives, each derived from its previous opposite phenotype, were grown at 37°C to late log phase in 100 ml of THB with 1 mM phenylmethylsulfonyl fluoride to inhibit protease activity. The final cell densities, as determined with a Klett-Summerson colorimeter (no. 54 filter), were 150 Klett units. Viability counts of 1.25 x 109 CFU/ml, cell dry weights of 0.4 mg/ml, and supernatant protein concentrations of 30 mg/ml (based on A280) did not differ significantly among cultures (Student's t test). For the [14C-glucose]sucrose incorporation GTF activity assay procedure, 0.3 ml of filter-sterilized supernatant was assayed as described by Kuramitsu (18). For the GTF and FTF activity gel assays, 0.05 ml of culture supernatant was mixed with an equal volume of 2.5 x loading buffer containing sodium dodecyl sulfate (19). The mixtures were heated for 5 min in a boiling water bath and stored at -20°C. Samples were subjected to sodium dodecyl sulfate-8.75% polyacrylamide gel electrophoresis as described by Laemmli (19). The gel was then incubated overnight at 37°C in 0.5% Triton X-100 and 3% sucrose to facilitate GTF and FTF enzymatic formation of glucan and fructan, respectively (26). The gel was stained for glucan and fructan bands by the periodic acid-Schiff method (32). Molecular weight standards (Bio-Rad Laboratories) were run on the same gel (data not shown). The results of the [14C-glucose]sucrose incorporation procedure (Fig. 1A) show that levels of extracellular GTF activity of Spp+ culture supernatants were more than threefold higher (P > 0.05 by Student's t test) than the GTF activity levels of Spp- culture supernatants. The attenuation of extracellular GTF levels of wild-type strains correlates with the Spp- phenotype. The GTF activity levels of the Spp- variants CH1A8 and CHlCl were notably lower than 3945
INFECT. IMMUN.
NOTES
3946
TABLE 1. Starting strain/variant
strain(s) Challis/Spp- variantSa
Spp+
+ Spp- frequencies for S. sanguis Challis and Challis variants Variants/total Type of
variation Spp+ - Spp-
2/5,500 8/4,500
11/6,500 8/6,500
4/,000 Mean
+
SD
CH97/Spp+ variantSa
SPP-
SPP+
8/3,600 2/5,400 7/6,000 10/6,000
9/6,500
3.6 1.8 1.7 1.2 5.7
x 10-4 x 10-3
1.1
X
10-3
2.2 3.7 1.2 1.7 1.4
x x
1o-4
x x
10-3 10-3
x 10-3 x 10-3 x 1o-4
0.6 x10-3
10x 10-3
1.4 x 10-3 ± 0.7 x 10-3
Mean ± SD
Challis/CH1A8b CH1A8/CH1B1b CH1B1/CH1C1 C11b,C CH1C1/CH1D1 D6bc
Frequency
colonies
Spp+ Spp-+
Spp-
SPP+
1/7,000 1/7,925 11/12,925
SPP-
SP
6/23,031
Spp
Spp-
Reversible phenotypic variation. Variant(s) derived sequentially from strain Challis. c CHlCl-C1l represents 11 variants strains beginning with CHlCl and ending with CHlC1l.
1.4 1.3 8.5 2.6
x x x
10-4 10-4
10-4
X 10-s
a
b
that of strain Challis but returned to wild-type levels in the Spp+ variants CHlBl and CH1D2 (Fig. 1A). Although these experiments used substrate with only the glucose moiety labeled, uniformly labeled sucrose gave rise to similar data. In addition, insoluble polysaccharide (i.e., polysaccharide that collected on filter membranes without prior methanol precipitation [18]) was found to vary similarly and represented about 10 to 20% of the total polysaccharide. The production of extracellular GTF and/or FTF activity by the same Spp+ and Spp- culture supernatant derivatives described above was examined by activity gel assays. S. mutans LM7 (serotype e) was used as a control; the sizes of its GTF and FTF enzymes have been reported previously (21). Strain LM7 excreted a GTF of 161 kilodaltons (kDa) and two FTFs of 96 and 83 kDa (Fig. 1B, lane 6) corresponding to the bands of activity described by Mukasa (21). Extracellular GTF activity was confined to a single band of 170 kDa for S. sanguis Challis and its derivatives (Fig. 1B, lanes 1 through 5). Supernatants of the Spp+ variants (Fig. 1B, lanes 3 and 5) have GTF activity bands of greater intensity than those of the Spp- variants (Fig. 1B, lanes 2 and 4); these results are qualitatively similar to the results from the [14C]glucose incorporation assays of the S. sanguis Challis Spp variant strains shown in Fig. 1A. No extracellular FTF activity was detected in S. sanguis Challis by a similar activity gel method (26) in which raffinose was provided as a substrate instead of sucrose. The control 83and 96-kDa FTF activity bands of S. mutans LM7 were apparent (data not shown). It is important to note reports by Keevil et al. (16, 17) indicating that a high Na+/K+ ratio in the medium, which is the case for the THB medium used in the present study, can greatly reduce FTF activity of S. sanguis. The GTF band of strain Challis resembles the 174-kDa activity band reported for S. sanguis 10558 by Grahame and Mayer (13), who observed significant aggregation of this enzyme. They also reported that proteolytic activity from the culture could result in a 156-kDa form. In our experiments, proteolytic activity was minimized by
CHlD1-D6 is abbreviated similarly.
growing the cells in a protease inhibitor (phenylmethylsulfonyl fluoride), and sodium dodecyl sulfate was used to eliminate aggregate formation during gel electrophoresis. The presence of 5 or 500 mg of dextran (Sigma Chemical Co.; average molecular weight of 10,000) per ml as a primer for polymer formation in the gel made no difference in the appearance of the bands (data not shown), a result consistent with other reports indicating that dextrans do not appear to prime the GTFs of S. sanguis (16, 17, 31). Indeed, highermolecular-weight dextrans were reported to be inhibitory. Therefore, using two different assays for GTF activity, we show a correlation between the Spp phase variation and levels of extracellular GTF. We believe this variation represents a novel phenomenon distinct from that reported by Donkersloot and Flatow (7), which described variants of S. sanguis Challis (which contained the S. mutans plasmid pVA380). Their variants of strain Challis displayed increased levels of extracellular GTF in comparison with wild-type Challis levels, whereas our variants (Spp-) had decreased levels of extracellular GTF. Also, their variants were distinguishable by expression of different hemolytic activities on Columbia horse blood agar. We observed no differences in hemolytic activity on Columbia horse blood agar between our Spp+ and Spp- variants. Finally, their variants were distinguishable by the presence of different chain lengths, whereas we observed no differences in chain length. The nature of the Spp phase variation remains to be determined. The involvement of DNA rearrangements which somehow control GTF expression is one reasonable possibility analogous to mechanisms of gram-negative bacterial phase and antigenic variation (25). Alternatively, mechanisms such as those involving frameshift mutations (27) could be involved. A recombinational phenomenon involving two gtf genes in S. mutans GS-5 is believed to be related to spontaneously appearing colonization-defective mutants, but the phenomenon is not known to be reversible
VOL. 57, 1989
NOTES
3947
This work was supported by Public Health Service grant DE02731 from the National Institutes of Health.
3. Carlsson, J., H. Grahnen, G. Jonsson, and S. Wikner. 1970. Establishment of Streptococcus sanguis in the mouths of infants. Arch. Oral Biol. 15:1143-1148. 4. Carlsson, J., E. Newbrun, and B. Krasse. 1969. Purification and properties of dextransucrase from Streptococcus sanguis. Arch. Oral Biol. 14:469-478. 5. Ciardi, J. 1983. Purification and properties of glucosyltransferases of Streptococcus mutans: a review, p. 51-64. In R. J. Doyle and J. E. Ciardi (ed.), Glucosyltransferases, glucans, sucrose and dental caries (a special supplement to Chemical Senses). Information Retrieval Limited, Washington, D.C. 6. de Stoppelaar, J. D., J. van Houte, and C. E. de Moor. 1967. The presence of dextran-forming bacteria, resembling Streptococcus bovis and Streptococcus sanguis, in human dental plaque. Arch. Oral Biol. 12:1199-1201. 7. Donkersloot, J. A., and U. Flatow. 1983. Glucosyltransferase activity of Streptococcus sanguis transformed with plasmids from Streptococcus mutans V380, p. 105-112. In R. J. Doyle and J. E. Ciardi (ed.), Glucosyltransferases, glucans, sucrose, and dental caries (a special supplement to Chemical Senses). Information Retrieval Limited, Washington, D.C. 8. Drucker, D. B., A. P. Shakespeare, and R. M. Green. 1984. In-vivo dental plaque-forming ability and relative cariogenicity of the bacteria Streptococcus mitis and Streptococcus sanguis I and II in mono-infected gnotobiotic rats. Arch. Oral Biol. 29:1023-1031. 9. Dummer, P. M. H., and R. M. Green. 1980. A comparison of the ability of strains of streptococci to form dental plaque-like deposits in vitro with their cariogenicity in gnotobiotic rats. Arch. Oral Biol. 25:245-249. 10. Furuta, T., T. Koga, T. Nisizawa, N. Okahashi, and S. Hamada. 1985. Purification and characterization of glucosyltransferases from Streptococcus mutans 6715. J. Gen. Microbiol. 131:285293. 11. Gibbons, R. J. 1983. Importance of glucosyltransferase in the colonization of oral bacteria, p. 11-19. In R. J. Doyle and J. E. Ciardi (ed.), Glucosyltransferases, glucans, sucrose, and dental caries (a special supplement to Chemical Senses). Information Retrieval Limited, Washington, D.C. 12. Gibbons, R. J., and M. Nygaard. 1968. Synthesis of insoluble dextran and its significance in the formation of gelatinous deposits by plaque-forming streptococci. Arch. Oral Biol. 13: 1249-1262. 13. Grahame, D. A., and R. M. Mayer. 1985. Purification and comparison of two forms of dextransucrase from Streptococcus sanguis. Carbohydr. Res. 142:285-298. 14. Hamada, S., and H. D. Slade. 1980. Biology, immunology, and cariogenicity of Streptococcus mutans. Microbiol. Rev. 44: 331-384. 15. Hamada, S., M. Torii, S. Kotani, and Y. Tsuchitani. 1981. Adherence of Streptococcus sanguis clinical isolates to smooth surfaces and interaction of the isolates with Streptococcus mutans glucosyltransferase. Infect. Immun. 32:364-372. 16. Keevil, C. W., A. A. West, N. Bourne, and P. D. Marsh. 1983. Synthesis of a fructosyltransferase by Streptococcus sanguis. FEMS Microbiol. Lett. 20:155-157. 17. Keevil, C. W., A. A. West, N. Bourne, and P. D. Marsh. 1984. Inhibition of the synthesis and secretion of extracellular glucosyl- and fructosyltransferase in Streptococcus sanguis by sodium ions. J. Gen. Microbiol. 130:77-82. 18. Kuramitsu, H. K. 1975. Characterization of extracellular glucosyltrap'sferase activity of Streptococcus mutans. Infect. Immun. 12:738-749. 19. Laemmli, U. K. 1970. Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature (London)
LITERATURE CITED 1. Bayliss, R., C. Clarke, C. M. Oakley, W. Somervile, A. G. W. Whitfield, and S. E. J. Young. 1983. The microbiology and pathogenesis of infective endocarditis. Br. Heart J. 50:513-519. 2. Carlsson, J. 1965. Zooglea-forming streptococci, resembling Streptococcus sanguis, isolated from dental plaque in man. Odontol. Revy 16:348-358.
20. Loesche, W. J. 1986. Role of Streptococcus mutans in human dental decay. Microbiol. Rev. 50:353-380. 21. Mukasa, H. 1986. Properties of Streptococcus mutans glucosyltransferases, p. 121-132. In S. Hamada et al. (ed.), Molecular microbiology and immunobiology of Streptococcus mutans. Elsevier Science Publishers B.V., Amsterdam. 22. Newbrun, E. 1971. Dextransucrase from Streptococcus sanguis,
A 3.0
2.0
E
-CC,-L
U._ O-
=:
1 .0
0
1
B.
2
3
4
5
6
kDa 200
-.- a
b
_1Ig , 116 -*
d-.C
92
66
-o
FIG. 1. (A) GTF activity of culture supernatants measured by the incorporation of ['4C]glucose from [14C-glucose]sucrose into methanol-insoluble glucan. GTF units are as defined by Kuramitsu (18) and are expressed as micromoles of [14C]glucose incorporated into methanol-insoluble glucan per minute per milliliter of supernatant. Symbols: FJ, Spp+; Spp-. (B) Polymer synthesis by GTF and FTF activities in culture supernatants of S. sanguis Challis and its Spp+ and Spp- derivatives. Lane 1, S. sanguis Challis (Spp+); lane 2, S. sanguis CH1A8 (Spp-); lane 3, S. sanguis CHlBl
(Spp+); lane 4, S. sanguis CHlCl (Spp-); lane 5, S. sanguis CH1D2 (Spp+); lane 6, S. mutans LM7 (control). In order to distinguish
between GTF and FTF activity bands, a similar gel was incubated in raffinose (a substrate used by FTF but not by GTF), resulting in only fructan polymer bands (data not shown). Activity bands are as follows: a, 170-kDa GTF activity band; b, 161-kDa GTF activity band; c, %-kDa FTF activity band; d, 83-kDa FTF activity band.
(29). A genetic analysis of the reversible GTF-related Spp switch reported here is presently in progress.
227:680-685.
3948
NOTES
further characterization. Caries Res. 5:124-134. 23. Niven, C. F., Jr., Z. Kiziuta, and J. C. White. 1946. Synthesis of a polysaccharide from sucrose by Streptococcus s.b.e. J. Bacteriol. 51:711-716. 24. Pakula, R., E. Hulanicka, and W. Walczak. 1958. Transformation reactions between streptococci, pneumococci, and staphylococci. Bull. Acad. Pol. Sci. Ser. Sci. Biol. 6:325-328. 25. Plasterk, R. H. A., and P. V. De Putte. 1984. Genetic switches by DNA inversions in prokaryotes. Biochim. Biophys. Acta 782:111-119. 26. Russell, R. R. B. 1979. Use of Triton X-100 to overcome the inhibition of fructosyltransferase by SDS. Anal. Biochem. 97: 173-175. 27. Stibitz, S., W. Aaronson, D. Monack, and S. Falkow. 1989. Phase variation in Bordetella pertussis by frameshift mutation in a gene for a novel two-component system. Nature (London) 338:266-269.
INFECT. IMMUN.
28. Tsumori, H., A. Shimamura, and H. Mukasa. 1985. Purification and properties of extracellular glucosyltransferase synthesizing 1,3-a-D-glucan from Streptococcus mutans serotype a. J. Gen. Microbiol. 131:553-559. 29. Ueda, S., and H. K. Kuramitsu. 1988. Molecular basis for the spontaneous generation of colonization-defective mutants of Streptococcus mutans. Mol. Microbiol. 2:135-140. 30. van Houte, J., R. J. Gibbons, and A. J. Pulkkinen. 1971. Adherence as an ecological determinant for streptococci in the human mouth. Arch. Oral Biol. 16:1131-1141. 31. West, A. A., C. W. Keevil, P. D. Marsh, and D. C. Ellwood. 1984. The effect of ionophores on growth and glucosyltransferase production of Streptococcus sanguis. FEMS Microbiol. Lett. 25:133-137. 32. Zacharius, R. M., T. E. Zell, J. H. Morrison, and J. J. Woodlock. 1969. Glycoprotein staining following electrophoresis on acrylamide gels. Anal. Biochem. 30:148-152.
