Effects of extracellular DNA and DNA-binding protein on the development of a Streptococcus intermedius biofilm

Authors


Correspondence

Yoichiro Miyake, Department of Oral Microbiology, Institute of Health Biosciences, The University of Tokushima Graduate School, 3-18-15 Kuramoto-cho, Tokushima 770-8504, Japan.

E-mail: miyake@tokushima-u.ac.jp

Abstract

Aims

The aim of this study was to clarify the effects of homologous and heterologous extracellular DNAs (eDNAs) and histone-like DNA-binding protein (HLP) on Streptococcus intermedius biofilm development and rigidity.

Methods and Results

Formed biofilm mass was measured with 0·1% crystal violet staining method and observed with a scanning electron microscope. The localizations of eDNA and extracellular HLP (eHLP) in formed biofilm were detected by staining with 7-hydoxyl-9H-(1,3-dichloro-9,9-dimethylacridin-2-one) and anti-HLP antibody without fixation, respectively. DNase I treatment (200 U ml−1) markedly decreased biofilm formation and cell density in biofilms. Colocalization of eHLP and eDNA in biofilm was confirmed. The addition of eDNA (up to 1 μg ml−1) purified from Strep. intermedius, other Gram-positive bacteria, Gram-negative bacteria, or human KB cells into the Strep. intermedius culture increased the biofilm mass of all tested strains of Strep. intermedius, wild-type, HLP-downregulated strain and control strains. In contrast, the addition of eDNA (>1 μg ml−1) decreased the biofilm mass of all Strep. intermedius strains.

Conclusions

These findings demonstrated that eDNA and eHLP play crucial roles in biofilm development and its rigidity.

Significance and Impact of the Study

eDNA- and HLP-targeting strategies may be applicable to novel treatments for bacterial biofilm-related infectious diseases.

Introduction

Bacteria accumulate at the biological interface and form biofilms, which are a community of bacterial cells embedded in a self-produced polymeric matrix. This matrix constitutes about 90% of the biofilm mass and mainly consists of extracellular polysaccharides, proteins, lipids and nucleic acids (Flemming and Wingender 2010). Polysaccharides and proteins are important components as a critical element of the matrix, and extracellular DNA (eDNA) is a common component among various bacterial constituents of the extracellular polymeric substance (EPS) in the biofilm (Hall-Stoodley et al. 2004). Recently, the role of eDNA has been increasingly recognized in both biofilm structural stability and protection against antimicrobial agents (Whitchurch et al. 2002; Mulcahy et al. 2008; Vilain et al. 2009). Hydroxyl radicals are extremely toxic and readily damage proteins, membrane lipids and DNA (Imlay and Linn 1986; Imlay et al. 1988; Farr and Kogoma 1991), and H2O2 releases eDNA from Streptococcus sanguinis (Kreth et al. 2009). It has also been reported that eDNA serves as an important structural component of Streptococcus pneumoniae biofilms, and the addition of intact DNA leads to increases in both the biofilm mass and bacterial viability of biofilms (Carrolo et al. 2010). A previous report showed that the addition of DNase I into initial inocula at concentrations of 40–400 U ml−1 inhibited biofilm formation by Streptococcus intermedius, indicating that eDNA may play an important role in the structure of the Strep. intermedius biofilm (Petersen et al. 2004).

Biofilm development can be divided into three distinct stages: attachment of bacterial cells to a surface, growth of cells into a sessile biofilm and the detachment of cells from the biofilm into the surrounding medium. Previous scientific research has focused on the attachment of planktonic bacterial cells to surfaces and the subsequent growth of the biofilm; however, the detachment and dispersal of bacterial cells from biofilms have received less attention (Hall-Stoodley et al. 2004). The dispersal of bacterial cells from the biofilm can spread, colonize new surfaces and form biofilms; therefore, this is an essential stage of the biofilm life cycle (Kaplan 2010). While many bacterial cells can disperse from biofilms by passive processes, such as the erosion or sloughing of cells from the biofilm caused by fluid shear (Stoodley et al. 2001), the bacterial biofilm can also periodically undergo active dispersal events, and bacterial cells in sessile, matrix-encased biofilms convert en masse to planktonic bacteria (Costerton et al.1999; Hall-Stoodley et al. 2004).

Bacterial nucleoid-associated proteins have been documented as an accessory architectural factor in a variety of bacterial cellular processes. The fact that bacterial histone-like protein (HLP) also exists extracellularly has been known for c. 30 years (Goodman et al. 2011). Recently, other studies have reported that bacterial HU from other genera are also present in the extracellular milieu (Menozzi et al. 1996; Paramonova et al. 2009). Interestingly, Goodman et al. (2011) suggested that the members of HLP, HU and IHF significantly contribute to the structural integrity of eDNA.

Streptococcus intermedius is a commensal bacterium and a member of the Streptococcus anginosus group (SAG). Among the SAG species, Strep. intermedius is the most common pathogen that is often isolated from oral infectious lesions, such as periodontitis, and fatal purulent infections in internal organs, especially brain and liver abscesses (Wagner et al. 2006; Erne et al. 2010). In addition, some clinical case reports also presented its ability in causing various kinds of infections such as infective endocarditis (Cunha et al. 2009). Streptococcus intermedius often causes chronic and/or recurrent infectious diseases depending on the biofilm life cycle.

Besides eDNA resulting from lysed or autolysed resident bacterial cells, eDNA within the biofilm can also originate from polymorphonuclear neutrophils, which release DNA (Brinkmann et al. 2004). In diseases with a biofilm component, biofilms formed in vivo are likely to be composed of eDNA of both host and bacterial origins (Goodman et al. 2011). In vivo studies showed that eDNA levels in the human lung are abundant (100–200 μg ml−1), even under normal physiological conditions, and that levels reach as high as 4 mg ml−1 in cystic fibrosis patients (Potter et al. 1969). Because Strep. intermedius has also been reported to have been isolated from a patient suffering from cystic fibrosis (Grinwis et al. 2010; Olson et al. 2010; Sibley et al. 2010), eDNA levels in the Strep. intermedius biofilm may also reach such high concentrations. Furthermore, as the habitat of Strep. intermedius is in the mouth and gastrointestinal tract, we assume that DNA derived from host cells in the oropharynx may also have an effect on Strep. intermedius biofilm development. To date, there have been no reports to demonstrate whether the external addition of homologous or heterologous DNA into the bacterial culture medium and HLP could affect the biofilm mass of Strep. intermedius. In this study, we examined the effects of the exogenous DNA (final concentration of 0·01–100 μg ml−1) of Strep. intermedius, Escherichia coli and human KB cells, a human carcinoma cell line of the oropharynx, and Si-HLP on the development of the Strep. intermedius biofilm.

Materials and methods

Bacterial strains, human cell line and culture conditions

We previously constructed an inducible antisense Si-hlp RNA-expressed Strep. intermedius (Si-HLP-downregulated Strep. intermedius) strain, BETAHT, by transforming into Strep. intermedius ATCC27335 as a wild-type (WT) strain with a Streptococci–E. coli shuttle plasmid harbouring the tetracycline-regulated antisense Si-hlp gene, a control strain, BETT, by transforming a parent plasmid into the WT strain, and another control strain, BETAXT, by transforming into the WT strain with an irrelevant antisense x fragment-inserted shuttle plasmid (Liu et al. 2008b). The WT, BETAHT, BETT and BETAXT strains of Strep. intermedius were grown in brain heart infusion (BHI) broth (Difco, Detroit, MI, USA) at 37°C under anaerobic conditions with AnaeroPack anaerobic atmosphere generation systems (Mitsubishi Gas Chemical Co. Inc., Tokyo, Japan). Streptococcus intermedius strains transformed with the plasmids were incubated in BHI containing 10 μg ml−1 erythromycin as a selective pressure and 20–60 ng ml−1 doxycycline for regulating antisense Si-hlp gene (Liu et al. 2008b). Staphylococcus aureus 209P was cultured in BHI under aerobic condition at 37°C. Escherichia coli strain K12 and Pseudomonas aeruginosa PAO1 were cultured in LB broth at 37°C under aerobic conditions.

The KB cell line (derived from a human oral epidermoid carcinoma; kindly provided by Dr. T. Okamoto, Hiroshima University School of Dentistry) was cultured in Dulbecco's modified Eagle's medium (DMEM) (Life Technologies, Grand Island, NY, USA) supplemented with 2 mmol l−1 l-glutamine, 10% (vol/vol) foetal bovine serum (JRH Biosciences, Lenexa, KA, USA), 50 IU ml−1 penicillin and 50 μg ml−1 streptomycin at 37°C in a water-saturated atmosphere of 95% air and 5% CO2.

DNA purification from bacteria and human cells

Streptococcus intermedius ATCC27335, E. coli K12, Staph. aureus 209P, Ps. aeruginosa PAO1 or KB cells were suspended in TE buffer (10 mmol l−1 Tris-HCl [pH 8·0], 1 mmol l−1 EDTA). Only Strep. intermedius and Staph. aureus 209P were treated with mutanolysin (final: 250 units ml−1) for 1 h at 37°C. Each cell suspension was then treated with proteinase K (final 100 μg ml−1) and SDS (final 0·12%) for 2 h at 50°C. DNA purification from cells was performed using phenol/chloroform extraction and ethanol precipitation methods. Briefly, DNA was extracted with an equal amount of 50 mmol l−1 Tris-HCl (pH 8·0)-saturated phenol and was precipitated with sodium acetate (final: 0·3 mol l−1) and 2·5 volumes of ethanol. Extracted DNA was treated with RNase A (final: 10 μg ml−1) for 1 h at 37°C, followed by re-extraction with phenol-chloroform (1 : 1) and precipitation with sodium acetate and ethanol. Purified DNA was finally dissolved in TE buffer. The purity and concentration of purified DNA was assessed by measuring the absorbance at 260 and 280 nm and agarose gel electrophoresis.

Quantification of the biofilm mass

A crystal violet biofilm assay was used to quantify the biofilm mass of Strep. intermedius as previously described (Moscoso et al. 2006). Aliquots of a 1 : 40 dilution of an overnight bacterial culture (1·0 × 107 CFU ml−1) were inoculated into the wells of a 96-well plate and incubated anaerobically at 37°C using the AnaeroPack system for 24 or 48 h. Formed biofilms were washed with PBS without disturbing the adherent biofilm, stained with 50 μl 0·1% crystal violet, incubated at room temperature for 15 min, and excess stain was removed by three gentle washes with PBS (pH 7·2). After being dried, the stained biofilm was extracted from the well by adding 50 μl ethanol and was determined by measuring the absorbance of the extract at 540 nm with a microplate reader (model 680; Bio-Rad, Hercules, CA, USA). Sterile BHI broth was substituted for bacterial cultures in control experiments.

To determine the inhibitory effect of DNase I on biofilm formation, DNase I (Roche, Mannheim, Germany) was added to the initial inoculum to a final concentration of 200 U ml−1 and incubated for 24 and 48 h to form a biofilm. The formed biofilm mass was quantified using the crystal violet biofilm assay as described above. In addition, to determine the effect of eDNA on biofilm stability, the 24-h-cultured Strep. intermedius WT biofilm was treated with DNase I (200 U ml−1) for 24 h and then DNase I-treated biofilm mass was quantified using the crystal violet staining.

To investigate the effect of DNA on biofilm formation, various concentrations of DNA purified from Strep. intermedius, Staph. aureus, E. coli, Ps. aeruginosa (0·01−100 μg ml−1) or KB cells (0·01−10 μg ml−1) were added into 96-wells plate containing 1·0 × 107 CFU ml−1 of all Strep. intermedius strains (WT, BETT, BETAHT and BETAXT) and incubated anaerobically at 37°C for 48 h. The 0·1% crystal violet biofilm assay was performed to quantify the biofilm mass as described above.

To investigate the rigidity of biofilm, formed biofilm mass before washing and retained biofilm mass after washing were quantified by the 0·1% crystal violet biofilm assay. The biofilm removal percentage was calculated using the following equations:

Biofilm removal percentage = (OD540 nm before washing − OD540 nm after washing)/OD540 nm after washing) × 100%.

Scanning Electron Microscopy (SEM) observation

The Strep. intermedius suspension (1·0 × 107 CFU ml−1) was added to each well of a 24-well culture plate with a type I collagen coating coverslip (Celldesk LF1; Sumitomo Bakelite Co., Tokyo, Japan) and was anaerobically incubated for 48 h at 37°C. After incubation, the coverslips were removed, rinsed with distilled water and fixed with 2·5% glutaraldehyde solution for 1 h at room temperature. Samples were then rinsed three times with distilled water and were dehydrated through a graded series of ethanol solutions to 100% ethanol. All samples were air-dried and were coated with Au ions for SEM analysis. SEM was carried out with a Miniscope TM-1000 (Hitachi High-Technologies Corp., Tokyo, Japan).

eDNA staining

7-hydoxyl-9H-(1,3-dichloro-9,9-dimethylacridin-2-one) (DDAO; Invitrogen, Carlsbad, CA, USA) was used to stain eDNA. Before using DDAO to stain the Strep. intermedius biofilm, we first confirmed the effectiveness of DDAO in staining eDNA. Streptococcus intermedius cells were disrupted by mixing with glass beads (Ф = 100 μm) for 10 min and nondisrupted cells were used as a control. DDAO was added into the cell sample at a concentration of 2 μmol l−1 and the sample was incubated for 30 min. The sample was then observed using a BIOREVO BZ-9000 microscope (KEYENCE Co., Osaka, Japan). After confirmation of the effectiveness of DDAO, we determined the effect of DNase I treatment on eDNA in the Strep. intermedius biofilm by staining with DDAO. The 24-h-cultured biofilm of Strep. intermedius WT was treated with DNase I (200 U ml−1) for 24 h at 37°C. Without fixation, the biofilm was then stained with DDAO and observed using a confocal fluorescence microscope as described above.

eDNA and HLP colocalization

To observe the distribution of genomic DNA, eDNA and Si-HLP in the Strep. intermedius biofilm, Hoechst 33342, DDAO and anti-Si-HLP peptide antibody were used, respectively. The biofilm of Strep. intermedius WT (1·0 × 107 CFU ml−1) was formed as described above. The formed biofilm was blocked with PBS-BSA (1·5%) without fixation and then reacted with rabbit anti-Si-HLP peptide antibody for 1 h at room temperature. After washing with PBS, the biofilm reacted with Alexa fluor 488-labelled anti-rabbit IgG (Invitrogen) for 1 h, followed by eDNA staining with 2 μmol l−1 DDAO for 30 min and continued by genomic DNA staining with 10 μg ml−1 Hoechst 33342 for 30 min. The biofilm was then observed using a confocal fluorescence microscope (model BZ-9000).

Growth of Streptococcus intermedius

DNA (1·0 and 10 μg ml−1) purified from the Strep. intermedius WT strain was added into 1·0 × 107 CFU ml−1 of Strep. intermedius WT and incubated anaerobically for 12 h. The growth of Strep. intermedius was monitored every 2 h by measuring the absorbance of the culture at OD600 nm.

Statistical analysis

All statistical analyses were performed using the unpaired Student's t-test. Differences were considered significant when the probability value was <5%.

Results

Effect of DNase I treatment on biofilm formation and its stability

To determine the effect of DNase I on biofilm formation, DNase I (200 U ml−1) was added to the initial Streptococcus intermedius WT inoculum and incubated for 24 and 48 h to form a biofilm. As shown in Fig. 1a, Strep. intermedius WT biofilm mass with the DNase I treatment was significantly lower in both the 24- and 48-h cultures than in the untreated control. SEM observations also showed that cell density with the DNase I treatment was markedly lower in the biofilm than in the untreated control (Fig. 1b). Next, to determine the effect of eDNA on biofilm stability, the 24-h-cultured Strep. intermedius WT biofilm was treated with DNase I (200 U ml−1) for 24 h. As shown in Fig. 1(c), biofilm mass with the DNase I treatment was significantly lower in the 24-h-cultured Strep. intermedius WT biofilm than in the untreated control. Before using DDAO to stain eDNA in the Strep. intermedius WT biofilm, we first confirmed the usefulness of DDAO for eDNA-specific staining in a planktonic Strep. intermedius WT culture (Fig. 1d: I and II). We observed that eDNA stained with DDAO in the 24-h-cultured Strep. intermedius WT biofilm was markedly decreased by the DNase I treatment (Fig. 1d: III and IV). These results suggest that eDNA may induce biofilm formation and plays an important role in the stability of the formed biofilm.

Figure 1.

Effect of the DNase I treatment on Streptococcus intermedius ATCC27335 biofilm formation (a, b) and its stability (c, d). (a, b) DNase I (200 U ml−1) was added to the initial Strep. intermedius wild-type (WT) inoculum and incubated for 24 and 48 h to form a biofilm. After staining with crystal violet, the formed biofilm mass was quantified by measuring the absorbance at 540 nm as shown on the y axis (a). Scanning electron microscopy images of Strep. intermedius WT cells after the 48-h incubation with DNase I (b). (c) The 24-h-cultured Strep. intermedius WT biofilm was treated with DNase I (200 U ml−1) for 24 h, and then biofilm mass was quantified by measuring the absorbance at 540 nm as shown on the y axis after staining with crystal violet. (d) Strep. intermedius cells were disrupted by mixing with glass beads (Ф = 100 μm) for 10 min and followed by staining with DDAO. Nondisrupted cells were used as a control. After washing, nondisrupted cells (I) and glass beads in the cell lysate (II) were observed by confocal fluorescence microscopy. The 24-h-cultured biofilm of Strep. intermedius WT was treated with DNase I (200 U ml−1) for 24 h and then stained with DDAO (IV). Nontreated cells were used as a control (III). After washing, the Strep. intermedius WT biofilm was observed by confocal fluorescence microscopy. The results are representative of four different experiments demonstrating similar results. Data are the mean and SD of four independent experiments. Asterisks show significant differences between the indicated groups (*< 0·01).

eDNA and HLP colocalization in the biofilm

Ours and other previous studies reported that Si-HLP is released outside cells as well as being localized in the intracellular compartment without cell lysis, and the HLPs of Helicobacter pylori and Streptococcus pyogenes have been detected in the culture supernatant and on the bacterial cell surface (Lei et al. 2000; Kim et al. 2002; Severin et al. 2007; Liu et al. 2008a). Our recent report further indicated that recombinant Si-HLP can bind DNA and alter the structural conformation of DNA in vitro (Liu et al. 2008b). We next determined the localizations of HLP and eDNA in the Strep. intermedius biofilm using confocal fluorescence microscopy. As shown in Fig. 2, we observed that eDNA and extracellular HLP (eHLP) were present and abundant in a 2-day-old biofilm of Strep. intermedius by immunofluorescent staining without fixation. Some HLP molecules were also colocalized with intracellular DNA and eDNA in the Strep. intermedius biofilm. This observation showing the colocalization of eHLP and eDNA suggests that HLP may form a complex with eDNA and play some roles in biofilm formation and its stability.

Figure 2.

Fluorescence microscopic observations of genomic DNA, extracellular DNA and Si-HLP in Streptococcus intermedius biofilms. A total of 1·0 × 107 CFU ml−1 Strep. intermedius ATCC27335 was incubated in brain heart infusion broth for 48 h at 37°C. The formed biofilm was blocked with PBS-BSA (1·5%) without fixation and then reacted with rabbit anti-Si-HLP antibody for 1 h at room temperature. After washing with PBS, the biofilm reacted with Alexa fluor 488 anti-rabbit IgG as the secondary antibody for 1 h, followed by eDNA staining with 2 μmol l−1 DDAO for 30 min and continued by genomic intracellular DNA staining with 10 μg ml−1 Hoechst 33342. The stained biofilm was then observed using a confocal fluorescence microscope. A2 and A3 show the images of y- and x axis cross-sections, respectively. All images show that Si-HLP, eDNA and intracellular DNA are stained green, red and blue, respectively. White arrows and white arrowheads show the colocalizations of histone-like protein with eDNA and genomic intracellular DNA, respectively.

Effect of Si-HLP on biofilm formation by Streptococcus intermedius

We previously demonstrated that Si-HLP is essential for cell viability and normal growth using gene knockout mutation and tet-regulation system-based antisense-mediated gene silencing (Liu et al. 2008b). We further determined whether Si-HLP could affect biofilm formation by Strep. intermedius. The Si-HLP-downregulated strain (BETAHT) formed significantly less biofilm mass than all control strains (WT, BETT and BETAXT) and biofilm mass was dependent on Si-HLP expression levels under the control of doxycycline (Fig. 3). This result also suggests that HLP plays a role in biofilm formation.

Figure 3.

Effect of Si-HLP on biofilm formation by Streptococcus intermedius. The Si-HLP-downregulated strain (BETAHT), Strep. intermedius wild-type (WT) and two control transformant strains (BETT and BETAXT) were anaerobically cultured in brain heart infusion medium containing 10 μg ml−1 erythromycin and 20–60 ng ml−1 doxycycline for 24 and 48 h. The formed biofilm mass was quantified by measuring the absorbance at 540 nm as shown on the y axis after staining with crystal violet. Asterisks show significant differences from WT and control transformant strains (*< 0·01). (image_n/jam12202-gra-0001.png) WT; (image_n/jam12202-gra-0002.png) BETT; (image_n/jam12202-gra-0003.png) BETAHT; (image_n/jam12202-gra-0004.png) BETAXT.

Effect of DNA addition on biofilm formation

To determine the role of eDNA in Strep. intermedius WT biofilm formation, Strep. intermedius WT was incubated for 48 h to form a biofilm with purified Strep. intermedius DNA at various concentrations. Moreover, to investigate the rigidity of biofilm, formed biofilm mass before washing and retained biofilm mass after washing were quantified, and the biofilm removal percentage was calculated. Streptococcus intermedius DNA increased the retained biofilm mass of Strep. intermedius WT strain in a dose-dependent manner up to 1·0 μg ml−1 by crystal violet biofilm assay after washing. In contrast, the higher concentrations of Strep. intermedius DNA at 10 and 100 μg ml−1 decreased the biofilm mass of Strep. intermedius WT strain (Fig. 4a). Interestingly, this decreased biofilm mass at higher concentrations of DNA may be causally related to its fragile structure because we observed that the adherent biofilm was broken off during washing, and the biofilm removal percentage at 10 and 100 μg ml−1 was extremely high (Fig. 4b).

Figure 4.

Effect of DNA addition on biofilm rigidity (a, b) and formation (c) by Streptococcus intermedius. (a, b) Various concentrations of Strep. intermedius DNA were added into 1·0 × 107 CFU ml−1 of Strep. intermedius wild-type (WT) strain. After the 48-h incubation, formed biofilm mass before washing and retained biofilm mass after washing were quantified by the 0·1% crystal violet biofilm assay. Symbols (* and #) indicate significant differences vs the control (< 0·01). (c) Various concentrations of exogeneous DNA purified from Staphylococcus aureus 209P, Escherichia coli K12, Pseudomonas aeruginosa PAO1 and KB cells as well as homologous Strep. intermedius DNA were added into 1·0 × 107 CFU ml−1 of Strep. intermedius WT, BETT, BETAHT and BETAXT strains, and all Strep. intermedius strains were anaerobically cultured in brain heart infusion medium containing 10 μg ml−1 erythromycin and 40 ng ml−1 doxycycline. The crystal violet biofilm assay was performed to quantify the formed biofilm mass after the 48-h incubation. Asterisks show significant differences from WT and control transformant strains (*< 0·01). (image_n/jam12202-gra-0003.png) Befor Washing; (image_n/jam12202-gra-0002.png) After Washing; (image_n/jam12202-gra-0001.png) WT; (image_n/jam12202-gra-0002.png) BETT; (image_n/jam12202-gra-0003.png) BETAHT; (image_n/jam12202-gra-0004.png) BETAXT.

We further determined whether heterologous eDNA could also induce biofilm formation by the addition of DNA purified from other bacteria, such as Staph. aureus, E. coli and Ps. aeruginosa, or KB cells as well as homologous Strep. intermedius DNA. All tested DNA increased the biofilm mass of all Strep. intermedius strains, including the Si-HLP-downregulated strain BETAHT, in a dose-dependent manner up to 1·0 μg ml−1, but the higher concentrations of all tested DNA decreased the biofilm mass of all Strep. intermedius strains (Fig. 4c). This result suggests that eDNA, regardless of the origin of DNA, may promote biofilm formation and affect the rigidity of the formed biofilm.

SEM observations

As shown in Fig. 5, we also observed that the addition of 1·0 μg ml−1 of Strep. intermedius DNA dramatically increased the biofilm mass of both Strep. intermedius WT and Si-HLP-downregulated BETAHT strains and Strep. intermedius cell density in both biofilms. However, the addition of 100 μg ml−1 of Strep. intermedius DNA markedly decreased the biofilm mass of both strains and cell density in both biofilms, indicating that biofilms formed at higher concentrations of eDNA become structurally weakened. These observations of structural changes to the biofilm formed in the presence of DNA correlated with the results of biofilm mass quantification formed in a culture with eDNA and after the DNase I treatment (Figs 1 and 4). We further observed similar images for biofilms formed by the addition of DNA purified from E. coli (data not shown). These results also suggest that eDNA, regardless of the origin of DNA, may affect biofilm formation and the rigidity of the formed biofilm.

Figure 5.

Scanning electron microscopy images of the Streptococcus intermedius biofilm formed in the absence or presence of Strep. intermedius DNA. DNA (1·0 and 100 μg ml−1) purified from the Strep. intermedius wild-type (WT) strain was added into 1·0 × 107 CFU ml−1 of Strep. intermedius WT and Si-HLP-downregulated BETAHT strains that were then incubated anaerobically in brain heart infusion medium containing 40 ng ml−1 doxycycline for 48 h. The formed biofilms were then observed under SEM. Images are representative of each biofilm.

Effect of DNA addition on the growth of Streptococcus intermedius

We finally determined whether the addition of DNA could affect the growth rate of Strep. intermedius. Figure 6 shows that the presence of DNA at a high concentration (10 μg ml−1) inhibited the growth of Strep. intermedius, whereas 1 μg ml−1 DNA had no effect on Strep. intermedius growth.

Figure 6.

Effect of DNA addition on the growth of Streptococcus intermedius. DNA (1·0 and 10 μg ml−1) purified from the Strep. intermedius wild-type (WT) strain was added into 1·0 × 107 CFU ml−1 of Strep. intermedius WT that was then incubated anaerobically for 12 h. The growth of Strep. intermedius was monitored every 2 h by measuring the absorbance of the culture at OD600 nm. Asterisks show significant differences from control and 1·0 μg ml−1 groups (*< 0·01). (image_n/jam12202-gra-0005.png) 0 μg ml−1 (control); (image_n/jam12202-gra-0006.png) 1 μg ml−1; (image_n/jam12202-gra-0007.png) 10 μg ml−1.

Discussion

This study successfully demonstrated that eDNA plays roles in Strep. intermedius biofilm formation and the rigidity of the formed biofilm. Our first results show that DNase I treatment markedly decreased biofilm formation as well as cell density in Strep. intermedius biofilms and degraded eDNA in the matrix of the mature biofilm (Fig. 1). These results indicate that eDNA plays essential roles in Strep. intermedius biofilm formation and its structural strength. Our findings are in agreement with the first report showing that eDNA is required for the initial establishment of a Ps. aeruginosa biofilm (Whitchurch et al. 2002) and another report suggesting that eDNA is important for the development of an Strep. intermedius biofilm (Petersen et al. 2004).

Regarding the regulation of biofilm formation, it has been recently reported that E. coli H-NS, the histone-like nucleoid structuring protein, plays important roles in regulating biofilm formation (Dalai et al. 2009). We previously demonstrated that Si-HLP could be released from bacteria without cell lysis as well as being localized in the intracellular compartment (Liu et al. 2008a). Here, we observed that abundant Si-HLP and DNA were colocalized in the matrix of the 24-h-cultured biofilm (Fig. 2). This finding indicates that Si-HLP and DNA are the important components of the matrix in a biofilm and suggests that Si-HLP may bind to eDNA and form an eDNA-eHLP complex. Therefore, this eDNA-eHLP complex may play roles in biofilm formation and its structural strength.

To date, the role or function of bacterial HLP in biofilm formation has not been investigated. In our previous study to verify the essentiality of Si-hlp, we constructed an inducible antisense Si-hlp RNA-expressed Strep. intermedius strain (BETAHT) by transforming into the WT strain with a Streptococci–E. coli shuttle plasmid harbouring the inserted Si-hlp gene between the tetR/O promoter and Ω fragment in an antisense orientation and demonstrated that doxycycline-induced Si-hlp antisense RNA expression specifically inhibited Si-HLP protein expression driven by the chromosomal Si-hlp locus (Liu et al. 2008b). Regarding this tet-regulation system-based antisense-mediated gene silencing, base pairing between sense mRNA and complementary antisense RNA has been considered to passively block the processing or translation of mRNA, or result in the recruitment of nucleases that promote mRNA destruction (Brantl 2002; Huntzinger et al. 2005). Using these doxycycline-regulated antisense RNA expression techniques, we demonstrated that the Si-HLP-downregulated strain (BETAHT) formed significantly less biofilm mass and biofilm mass was dependent on Si-HLP expression levels (Fig. 3). This result indicates that HLP as well as eDNA plays an important role in biofilm formation.

By adding Strep. intermedius DNA at a low concentration (up to 1 μg ml−1) as exogenous DNA supplementation, the biofilm mass of all tested strains was increased in a dose-dependent manner, and the cell density in the formed biofilms of both Strep. intermedius WT and Si-HLP-downregulated strains at 1 μg ml−1 of DNA was also increased (Figs 4 and 5). Our results correspond with a previous report showing that DNA addition enhanced Strep. pneumoniae biofilm mass in a dose-dependent manner and suggest that eDNA is essential for the enhancement of biofilm growth and has an important role in biofilm architecture. (Carrolo et al. 2010). Interestingly, we demonstrated that the biofilm mass of the Si-HLP-downregulated strain was also increased by the addition of purified DNA, but was still significantly less than other WT and control strains (Fig. 4). Our previous report has shown that the Si-HLP-downregulated strain grows significantly more slowly with prolonged lag and logarithmic phases than WT and control strains, and this growth inhibition results from the induction of antisense Si-hlp RNA expression controlled with the tetR/O-inducible promoter (Liu et al. 2008b). Moreover, we found that the ATP assay and cell numbers counted as CFU showed that the doxycycline-induced Si-HLP-downregulated strain displayed lower amounts of intracellular ATP and lower numbers of living cells than those of control strains when their culture reached the same value of OD600, respectively. Therefore, growth inhibition of the Si-HLP-downregulated strain may be one of the reasons for the lower ability of this strain to form a normal biofilm.

Intriguingly, we further demonstrated that the addition of heterologous DNAs (up to 1 μg ml−1) led to more robust biofilm formation in all tested Strep. intermedius strains in a dose-dependent manner (Fig. 4). These results suggest that enhancements in biofilm formation may not be dependent on homologous DNA or be species specific and that all kinds of DNA may increase the biofilm mass formed in a dose-dependent manner.

In contrast to the effect of eDNA supplementation at lower concentrations (up to 1 μg ml−1), the addition of DNA at higher concentrations (10 and 100 μg ml−1) significantly decreased Strep. intermedius biofilm mass in a dose-dependent manner, and the cell density in the formed biofilm with the addition of 100 μg ml−1 was markedly decreased (Figs 4 and 5). We suggested that this opposite effect may be due to the growth inhibitory properties of higher concentrations of DNA because we found that the addition of 10 μg ml−1 Strep. intermedius DNA led to a 20% inhibition in the Strep. intermedius growth rate (Fig. 6). In accordance with this result, a previous study showed that high concentrations of DNA (5 mg ml−1 or more) had a toxic effect on the growth of Ps. aeruginosa by acting as a cation chelator and subsequently induced cell lysis (Mulcahy et al. 2008). Although recent reports have shown the effect of exogenous DNA on enhancing biofilm formation, we here suggest that different DNA concentrations may have had different effects on biofilm formation and this may be part of the potential of bacteria to survive in unfavourable environments. Paramonova et al. (2009) reported that the increase in eDNA contents in the Candida albicans biofilm led to a decrease in biofilm strength and the formed biofilm could be more easily removed. Our present data also showed that the biofilm removal percentage at 10 and 100 μg ml−1 was extremely high because the adherent biofilm was broken off during washing (Fig. 4b). Therefore, a similar mechanism may have occurred in the Strep. intermedius biofilm, and it could be considered that the biofilm formed with a high concentration of DNA (>10 μg ml−1) has low rigidity and is therefore easy to remove by fluid shear stress of washing and the dispersed bacterial cells then attach to another site.

Our previous report also showed that the Si-HLP-downregulated strain, BETAHT, largely lost its surface hydrophobicity as a result of alterations in cell surface components and luxS gene expression was downregulated in the BETAHT strain (Liu et al. 2008b). It has been reported that luxS plays an important role in biofilm formation by Strep. intermedius (Ahmed et al. 2008, 2009). Considering these findings, we proposed that Si-hlp encoding histone-like DNA-binding protein may be involved in the biofilm development of Strep. intermedius by regulating the expression of bacterial surface components and bacterial quorum sensing communication. Therefore, further studies to identify the genes involved in biofilm development and determine their expression levels are needed and are currently under investigation.

It has recently been reported that autolysins (bacterial murein hydrolases) of Gram-positive bacteria, such as Enterococcus faecalis and Staphylococcus epidermidis, are implicated in biofilm formation, apparently through the mediation of bacterial lysis with the subsequent eDNA release (Qin et al. 2007; Thomas et al. 2008, 2009; Guiton et al. 2009). It has been demonstrated that DNA release displayed in the stationary phase of liquid cultures of pneumococcal cells can occur through cell lysis, rather than through a specific mechanism of secretion, and this release depends on the major autolytic N-acetylmuramyl-l-alanine amidase, LytA and the autolytic lysozyme, LytC, and it has also been shown that the competence-dependent release of DNA occurred by autolysis (Tomasz et al. 1988; Steinmoen et al. 2002; Moscoso and Claverys 2004). Moreover, recent interesting reports have shown that Strep. pneumoniae biofilm formation is influenced by the presence of eDNA, LytA mutants have a decreased capacity to form biofilms, and LytA-induced pneumococcal lysis may be related to biofilm formation through the release of eDNA (Moscoso et al. 2006; Hall-Stoodley et al. 2008). Previous studies have reported that eDNA present in bacterial biofilms results from cell lysis (Imlay and Linn 1986; Imlay et al. 1988; Farr and Kogoma 1991; Kreth et al. 2009; Perry et al. 2009) or is a product of direct secretion from intact cells (Whitchurch et al. 2002). However, the origin of eDNA in the Strep. intermedius biofilm is still unclear and is currently under investigation.

This is the first report to demonstrate that both homologous and heterologous DNA addition directly affects Strep. intermedius biofilm development and its rigidity and suggests that all kinds of DNA present at infection sites can increase bacterial biofilm formation. Moreover, our present results clearly show that bacterial histone-like DNA-binding protein (HLP) also plays crucial roles in biofilm development by forming a complex with eDNA. Regarding the recognition of bacterial infection, it has been known that the bacterial DNA, one of pathogen-associated molecular patterns, activates transcription factors, including NF-κB and interferon regulatory factors via TLR9-mediated and TLR-independent signalling pathways, and induces the production of proinflammatory cytokines and type I interferon (Hemmi et al. 2000; Takeuchi and Akira 2007). In addition, we recently reported that bacterial HLP initiates and exacerbates proinflammatory reactions during bacterial infection, as well as its physiological role in bacterial growth through DNA binding (Liu et al. 2008a). Considering these immunological findings and biofilm as a pathogenic factor, the contents of eDNA and HLP in bacterial biofilms can be used as indicators of the severity of infection. Furthermore, targeting HLP and eDNA may create a novel strategy to fight micro-organism-caused infectious diseases, especially those related to biofilm formation.

Acknowledgements

This work was supported by a Grant-in-Aid for Scientific Research from the Japan Society for the Promotion of Science (21592387 and 24592872). The authors declare that they have no conflicts of interest.

Ancillary