Streptococcus constellatus is a part of normal commensal flora in the mouth, oropharynx and gastrointestical tract, and composes the Streptococcus anginosus Group (SAG), together with Streptococcus anginosus and Streptococcus intermedius [1,2]. S. constellatus has been cultured from dental caries and periodontitis lesions, but has also been found to cause purulent infections with abscess formation in various regions of the body, especially in thorax and head and neck regions, despite this organism appears to be innocuous at its natural habitat [3-5]. In contrast, S. anginosus and S. intermedius in SAG show tropism for abdomen and genitourinary tract, and for liver and Central Nerve System (CNS), respectively. There apparently appears to be species bias to some infections, but mechanisms behind such phenomena are not fully understood [6,7].

In this study, we found that S. constellatus strain H39 isolated from a subperiosteal abscess lesion due to the infection of root canal system of a tooth, has a capacity to produce viscous materials in spent culture media and form dense meshwork structures, a typical phenotype for biofilm-forming bacteria [8], around cells on agar plateS. This phenotype was also observed on type strains for S. constellatus. As far as we retrieved, only a small number of literatures are available regarding biofilm-phenotypic characterization of SAG [4,9,10]. Therefore, we also examined the capacity of this isolate and type strains of SAG, to form biofilm on abiotic materials.

Bacteria and tentative identification

A clinically isolated facultative anaerobic Gram-positive coccus capable of producing viscous materials in its spent culture medium (see below) was used. The isolate was recovered in pure culture from an odontogenic subperiosteal abscess lesion, and designated as strain H39. Cultures were maintained by weekly transfer on a trypticase soy agar plate supplemented with 5% sheep blood (TSA-B: BD BBL, Franklin Lakes, NJ, USA) in a 10% CO₂, 10% H₂, 80% N₂ atmosphere at 37°C in an anaerobic chamber (ANX-3, Hirasawa, Tokyo, Japan). Genomic DNA was extracted from strain H39, using MagExtractor (TOYOBO, Osaka, Japan), according to the manufacturer’s instructions. The nucleotide sequence of the 16S rRNA gene of the clinical isolate amplified with a primer pair of 27F (5’-AGAGTTTGATCCTGGCTCAG-3’) and 1525R (5’-AAAGGAGGTGATCCAGCC-3’) was identical [DNA Data Bank of Japan (DDBJ: http://www.ddbj.nig.ac.jp) accession: AB749301] and 99% homologous to S. constellatus ATCC 27823 (accession number: AF104676).

S. constellatus ATCC 27823, S. intermedius ATCC 27335 and S. anginosus ATCC 33397 were used as the type strains for SAG, and the cultures were maintained as described above.

Identification by phylogenetic analysis

Identification of organisms in SAG had long been problematic due to the considerable genetic similarity within the group and heterogeneous phenotypes on each species in SAG, but phylogenetic analysis using several housekeeping gene sequences is recently acknowledged as a powerful tool to divide SAG in three species [11-14]. Therefore, one of clinical isolates designated as strain H39 was subjected to phylogenetic analysis, using partial sequences of 16S rRNA, recN, groEL and tuf genes.

Briefly, strain H39 was grown anaerobically for 48 hours and the genomic DNA was extracted using MagExtractor (TOYOBO), as described above. Gene fragments of recN, groEL and tuf were amplified with primer pairs of Si-recN-left (5’-TGAATATGATGTTGGGAAGC-3’) and Si-recN-right (5’-ACAATGCTTGTCTTGCCTTC-3’), groELFW (5’-GGNGACGGNACNACNACNGCAACNGT-3’) and groELRV (5’-TCNCCRAANCCNGGYGCNTTNACNGC-3’), and tuf- FW (5’-AAYATGATIACIGGIGCIGCICARATGG-3’) and tuf-RV (5’-CCIACIGTICKICCRCCYTCRCG-3’). PCR was performed using GoTaq Master Mix (Promega, Tokyo, Japan) and MJ Mini Personal Thermal Cycler (Bio-Rad, Tokyo, Japan) for 30 cycles of 30 s at 95°C, 30 s at 50°C, and 2 min at 72°C, with a final extension at 72°C for 7 min. The amplified fragments were subsequently sequenced on the Applied Biosystem 3130xl DNA Analyzer (Applied Biosystems, Tokyo, Japan). The sequence data were assembled with the GENETYX Ver.7 program (Genetyx Ltd., Tokyo, Japan), and deposited in the DDBJ (accession number: AB749302-4). Phylogenetic analyses were performed by using obtained sequences of the housekeeping genes and the 16S rRNA gene and the data sets of recN, groEL, tuf and 16S rRNA genes for streptococci described elsewhere [11,12,14]. The nucleotide sequences of these genes were aligned by MAFFT (version 6) with FFT-NS-I option [15]. Then, a neighbor-joining tree was constructed using Jukes-Cantor distance matrices for each housekeeping gene.

Measurement of viscosity

The viscous material-producing ability of clinical isolates was evaluated by measuring the viscosity of spent culture media, as described elsewhere [16]. Arbitrarily selected eight clinical strains and type strains of SAG were grown in trypticase soy broth (BBL Microbiology Systems, Cockeysville, MD, USA), supplemented with 0.5% yeast extract (BBL Microbiology Systems) (TSB-Y) for 48 h. The viscosity was measured as shearing stress between a rotor and a rotor shaft at 50 rpm, 20°C, using a rotary viscometer (Toki-sangyo, Tokyo, Japan), in accordance with the manual provided by the manufacturer. The measurement was performed five times on each sample, and the results of four independent cultures were averaged to give the final value. Statistical differences between the bacterial cultures and control medium were determined using an unpaired t-test, with the level of significance set at P<0.05.

For SEM studies, cells scraped from 1-day-old cultures were collected on a piece of filter paper (Glass fiber GA55, Toyo Roshi, Tochigi, Japan), fixed with 2% glutaraldehyde in 0.1 M phosphate buffer (Mitsubishi Chemical Medience, Tokyo, Japan) for 2 hours, and with 1% OsO₄ in 0.1M phosphate buffer for 1 h at 4°C and dehydrated through an ethanol series and 2-methyl-2-propanol, followed by platinum ion coating (E-1030, HITACHI, Tokyo, Japan). Specimens were examined with a scanning electron microscope (S-4800, HITACHI), at an accelerating voltage of 3 kV.

Biofilm formation assay

The ability to form biofilm was determined, as described previously [17]. Briefly, the seed cultures of S. constellatus strains H39 and ATCC 27823, S. intermedius ATCC 27335 and S. anginosus ATCC 33397 were prepared as described above, and diluted to an O.D. of 0.1 at 620 nm in the same medium. Next, 150 μl diluted culture was transferred to three sterile polystyrene microtiter plate wells (IWAKI, Tokyo, Japan) per strain. Sterile TSB-Y was used as a control. The plates were prepared in duplicate and incubated at 37°C for 24 h. Unbound cells were removed by inversion of the microplate and tapping on absorbent paper towel. Adhered cells were then stained with 0.1% crystal violet for 15 min. Excess stain was removed by washing with PBS. The stained wells were recorded photographically using a camera (PC1089, Canon, Tokyo, Japan) set. The crystal violet was then solublized by the addition of 95% ethanol, and absorbance at 570 nm was measured by spectrophotometer U-2000 (Hitachi). Each strain was assayed in three wells on each plate and the whole experiment was repeated three times.

Identification of strain H39

The nucleotide sequence of 16S rRNA gene of strain H39 (accession number: AB749301) was identical and highly homologous to that of S. constellatus ATCC 27823 (accession number: AF104676).

Phylogenetic analysis using the partial gene sequences of 16S rRNA gene (accession number: AB749301), recN (accession number: AB749304), groEL (accession number: AB749303) and tuf (accession number: AB749302) deposited in DDBJ (http://www.ddbj.nig.ac.jp) produced trees with a similar topology. A sister relationship between strain H39 and a type strain of S. constellatus was strongly supported by high bootstrap values (Figure 1). Therefore, strain H39 was identified as S. constellatus.

Figure 1: Phylogenetic tree based on partial sequences of 16S rRNA (A), recN (B), groEL (C) and tuf (D) genes. Phylogenetic analyses using these four housekeeping genes produced trees with a similar topolygy. A sister relationship between strain H39 and a type strain of S. Treptococcus constellatus was strongly supported by high bootstrap values.

Viscous material productivity

Strain H39 and type strains of SAG were inoculated into TSB-Y and grown for 48 h. The viscosities of the spent culture media were measured using a rotary viscometer. Strain H39 produced materials in vitro that were highly viscous, as compared with the control medium (Figure 2). The viscosity of spent culture medium of S. constellatus ATCC 27823 was as high as those of the clinical isolateS. The viscosities of S. intermedius ATCC 27335 and S. anginosus ATCC 33397 were similar to that of control medium, correlated to its cell surface structures (Figure 2).

Figure 2: Viscosity of the spent culture media obtained from S. Treptococcus constellatus strains H39 and ATCC 27823, S. Treptococcus intermedius ATCC 27335 and S. Treptococcus anginosus ATCC 33397. Viscosity of the enriched tryptic soy broth medium was measured as a control. Bars indicate standard deviations.

Biofilm phenotype on strain H39 and type strains of SAG

SEM observation revealed that strain H39 showed surfaceassociated dense meshwork-like structures around the cells (Figure 3). Similar surface structure was observed on S. constellatus ATCC 27823, but not on S. intermedius ATCC 27335 and S. anginosus ATCC 33397 (Figure 3).

Figure 3: Scanning electron micrographs showing surface structures of S. Treptococcus constellatus strains H39 (H39) and ATCC 27823 (27823), S. Treptococcus intermedius ATCC 27335 (27335) and S. Treptococcus anginosus ATCC 33397 (33397). S. Treptococcus constellatus strains H39 and ATCC 27823 had dense meshwork structures surrounding the cell surfaces, but other strains lacked this phenotype. Thick fibers behind S. anginosus ATCC 33397 cells are glass fibers of filter paper (33397). Bars=2 μm.

Biofilm formation assay

The ability to form biofilm was investigated for S. constellatus strains H39 and ATCC 27823, S. intermedius ATCC 27335 and S. anginosus ATCC 33397, using crystal violet microtiter plate assay. Strain H39 was consistently able to form biofilm on flat-bottomed polystyrene microtiter plates (Figure 4). Type strains of S. constellatus and S. intermedius were able to form biofilm on abiotic materialS. The capacity to form biofilm on S. constellatus ATCC 27823 seemed to be comparable to that of strain H39, but S. intermedius ATCC 27335 was weaker than that of strain H39. S. anginosus ATCC 33397 could not form biofilms on microtiter plates (Figure 4).

Figure 4: Biofilm formation on microtiter plates. Biofilm production of S. Treptococcus constellatus strains H39 and ATCC 27823, S. Treptococcus intermedius ATCC 27335 and S. Treptococcus anginosus ATCC 33397 on a polystyrene microtiter plate: a set of microtitier plate wells (three wells for each strain) stained with 0.1% crystal violet solution after 24 h of incubation (A). The quantitative analysis of biofilm production as measuring the optical density of destained biofilms at 570 nm (B). The data are from one of three independent experiments. Bars indicate standard deviations.

Clinical studies [3,18] have shown that S. constellatus has the capacity to cause invasive purulent infection relatively in superficial areas through the body, while S. intermedius and S. anginosus in SAG exhibit a marked association with the infection in the central nervous system and gastrointestinal tracts, respectively. This organism is not frequently isolated from abscess lesions deep inside of the body, but several clinical reports showed the association of S. constellatus with brain abscesses [19,20], endocarditis [21-23], and sepsis after dental treatment [24]. The remarkable propensity to cause abscess or purulent infection observed on three species in SAG remind us an idea that there might be a unique interaction between the organisms in SAG and human neutrophils. In this regards, Wanahita et al. [25] reported that SAG organisms stimulated less chemotaxis than did Staphylococcus aureus, in contrast to viridans Streptococci (Streptococcus mitis, Streptococcus parasanguinis and Streptococcus sanguinis), which caused greater chemotaxis than did S. aureus. Interestingly, SAG and viridans Streptococci were equally engulfed by human neutrophils than S. aureus, but killed more slowly and less completely. Our previous studies showed that the resistance of biofilm-forming Prevotella intermedia strains to phagocytosis by human neutrophils are more significant than those of strains lacking biofilm phenotype [16,17,26]. We hypothesize that biofilm bacteria, being held together by exopolysaccharides as biofilm matrices, might present a huge physical challenge for phagocytosing neutrophils. As a consequence of these neutrophils being frustrated by their inability to phagocytose this bacterial mass, this might trigger the unregulated release of bactericidal compounds that could cause tissue injury [27], as shown in the inflammatory pathway associated with lung injury or chronic wounds [28,29]. Further investigation is needed to clarify whether inconsistency of biofilm phenotype in SAG gives influence of their interaction, with human neutrophils and propensity to cause abscesses.

As described above, only a few studies have done regarding biofilm phenotype on S. constellatus. Landrygan-Bakri et al. [10] investigated the adherence of SAGs to the matrix proteoglycans and fibronectin. All tested strains of S. intermedius in their study adhered strongly to gingival dermatan sulphate proteoglycans, as well as chondroitin sulphate proteoglycans (except one clinical isolate). In contrast, two strains of S. anginosus and all S intermedius test strains showed strong adherence to fibronectin. S. constellatus was less adherent to extracellular matrices of human tissue than S. intermedius [10]. Regarding the cell surface structures of organisms in SAG, biofilm-like structures were observed on a clinical strain of S. constellatus isolated from a severe recurrent form periodontitis lesion [4]. Petersen et al. [9] showed that the quorum sensing mediated by competence-stimulating signaling peptide favored the biofilm mode of growth of S. intermedius, in which the organism does not show meshwork structures on the cell surface. In this study, S. intermedius ATCC 27335 was able to form biofilm on microtiter plates, but lacked dense meshwork structures Presumably there might be different systems to induce biofilm growth between S. constellatus and S. intermdius .

In this study, we examined one clinical isolate and three type strains for SAG, and we found diversity of cell surface structures between the organisms in SAG. In addition to the previously known various properties of SAG which cause ambiguity in determining pathogenic potential of organisms in SAG, the phenotypic difference found in this study might influence their pathogenicity and tropism in the human body. It is known that S. constellatus is sometimes co-isolated with Eikenella species [30], or Prevotella species [31,32], so the capacity to form single/multi species biofilm on S. constellatus should be further investigated.

We would like to acknowledge Mr. Hideaki Hori for his excellent assistance with the electron microscopy. Part of this study was performed at the Institute of Dental Research, Osaka Dental University. This study was supported in part by Osaka Dental University Joint Research Fund (B08-01) and JSPS KAKENHI Grant Number 23792118, 23592724, 24791975 and 24592790.