Involvement of the Type IX Secretion System in Capnocytophaga ochracea Gliding Motility and Biofilm Formation

Abstract

Capnocytophaga ochracea is a Gram-negative, rod-shaped bacterium that demonstrates gliding motility when cultured on solid agar surfaces. C. ochracea possesses the ability to form biofilms; however, factors involved in biofilm formation by this bacterium are unclear. A type IX secretion system (T9SS) in Flavobacterium johnsoniae was shown to be involved in the transport of proteins (e.g., several adhesins) to the cell surface. Genes orthologous to those encoding T9SS proteins in F. johnsoniae have been identified in the genome of C. ochracea; therefore, the T9SS may be involved in biofilm formation by C. ochracea. Here we constructed three ortholog-deficient C. ochracea mutants lacking sprB (which encodes a gliding motility adhesin) or gldK or sprT (which encode T9SS proteins in F. johnsoniae). Gliding motility was lost in each mutant, suggesting that, in C. ochracea, the proteins encoded by sprB, gldK, and sprT are necessary for gliding motility, and SprB is transported to the cell surface by the T9SS. For the ΔgldK, ΔsprT, and ΔsprB strains, the amounts of crystal violet-associated biofilm, relative to wild-type values, were 49%, 34%, and 65%, respectively, at 48 h. Confocal laser scanning and scanning electron microscopy revealed that the biofilms formed by wild-type C. ochracea were denser and bacterial cells were closer together than in those formed by the mutant strains. Together, these results indicate that proteins exported by the T9SS are key elements of the gliding motility and biofilm formation of C. ochracea.

INTRODUCTION

Capnocytophaga ochracea is a Gram-negative, rod-shaped bacterium that demonstrates gliding motility when cultured on solid agar surfaces (1–3). C. ochracea was first isolated from human periodontitis lesions (4–6); however, subsequent studies have indicated that this bacterium is present in dental plaque from periodontally healthy sites (7). Although a recent metatranscriptome analysis has shown that putative virulence factors of C. ochracea are upregulated in patients with periodontitis (8), the role of C. ochracea in the pathogenesis and progression of periodontal disease remains controversial (9–13). C. ochracea is also reported to be involved in several systemic diseases and to produce an immunosuppressive factor (14, 15). C. ochracea has been implicated in focal infections, such as sepsis and purpura fulminans (16, 17), and associations between high levels of antibodies to C. ochracea and coronary heart disease (18) and a potential relationship with Sjögren's syndrome (19) have been reported.

To clarify the involvement of C. ochracea in these diseases, investigation of the colonization strategies used by this bacterium is essential. C. ochracea colonizes tooth surfaces by forming a biofilm and synergizing its growth with that of Fusobacterium nucleatum, which acts as a bridge between antecedent bacteria on the tooth surface and late-colonizing periodontopathic bacteria (20). However, the factors involved in biofilm formation by this bacterium remain unknown. For Escherichia coli, it is thought that motility facilitates biofilm expansion by enabling growing cells to migrate across the surface on which they are growing (21). For Pseudomonas aeruginosa, flagella and pili are known to play important roles in the aggregation of bacterial cells and the formation of microcolonies (22). Thus, we hypothesized that the gliding motility of C. ochracea is an important aspect of its biofilm formation.

Recently, a novel protein secretion system, the type IX secretion system (T9SS), was identified in Porphyromonas gingivalis and other members of the phylum Bacteroidetes. For example, in P. gingivalis, which is a prominent periodontal pathogen, several T9SS proteins (e.g., PorK, PorL, and PorT) were found to be crucial for the secretion of major proteases (23–26). T9SS proteins have also been found in Flavobacterium johnsoniae (phylum Bacteroidetes) (23, 25, 27, 28), and orthologous genes encoding T9SS proteins have been found in the genome of C. ochracea (28).

In F. johnsoniae, SprB (colony-spreading protein B) allows the bacterium to attach to agar and glass surfaces and to move via gliding motility (29). F. johnsoniae mutants deficient in genes encoding T9SS proteins (e.g., gldK, gldL, and sprT) are unable to attach to and to glide across glass surfaces, presumably as a result of their inability to secrete SprB and other adhesins at their cell surfaces (30). Since C. ochracea harbors orthologous genes predicted to encode T9SS proteins such as GldK and SprT (28), it is possible that proteins exported by the T9SS are involved in the gliding motility and biofilm formation of C. ochracea.

Here we constructed C. ochracea mutants deficient in the genes orthologous to gldK, sprT, or sprB in F. johnsoniae, and we investigated the role of the T9SS in gliding motility and biofilm formation. Our results indicated that proteins exported by the T9SS in C. ochracea are involved in gliding motility and biofilm formation.

MATERIALS AND METHODS

Bacterial strains, plasmids, and growth conditions.

The bacterial strains and plasmids used in the present study are listed in Table 1. C. ochracea ATCC 27872 (wild-type strain) was cultured and maintained, using standard methods, at 37°C on blood agar plates containing tryptic soy agar (Becton Dickinson, Sparks, MD) supplemented with hemin (5 μg/ml), menadione (0.5 μg/ml), and 10% horse blood (Nippon Bio-Test Laboratories, Tokyo, Japan), under anaerobic conditions (80% N₂, 10% H₂, and 10% CO₂) in an anaerobic chamber (ANX-3; Hirasawa, Tokyo, Japan). C. ochracea mutants were cultured and maintained on blood agar plates containing 10 μg/ml erythromycin (Sigma-Aldrich, St. Louis, MO). C. ochracea strains were also grown at 37°C in tryptic soy broth (Becton Dickinson) supplemented with hemin (5 μg/ml) and menadione (0.5 μg/ml), under anaerobic conditions. E. coli DH5α was grown at 37°C in Luria-Bertani agar (Wako Pure Chemical Industries, Osaka, Japan), under aerobic conditions, and the plasmid-transformed strains of E. coli were grown in Luria-Bertani agar containing 25 μg/ml kanamycin (Sigma-Aldrich).

TABLE 1.

Bacterial strains and plasmids used in the present study

Strain or plasmid	Descriptiona	Reference or source
Bacterial strains
Escherichia coli DH5α	Strain used for gene cloning	TaKaRa Bio Inc.
Capnocytophaga ochracea strains
ATCC 27872	Wild type	ATCC
ΔgldK mutant	ermF-ermAM insertion mutation in Coch_0809 of ATCC 27872; Emr	This study
ΔsprT mutant	ermF-ermAM insertion mutation in Coch_1748 of ATCC 27872; Emr	This study
ΔsprB mutant	ermF-ermAM insertion mutation in Coch_0203 of ATCC 27872; Emr	This study
Plasmids
pVA2198	ermF-ermAM	27
pCR2.1-TOPO	3.9-kb vector for cloning PCR products; Kmr Ampr	Invitrogen

^aAntibiotic resistance phenotypes: Em^r, erythromycin resistant; Km^r, kanamycin resistant; Amp^r, ampicillin resistant.

Construction of ΔgldK, ΔsprT, and ΔsprB mutant strains.

The genomic nucleotide sequence of C. ochracea ATCC 27872 was obtained from the GenBank database (accession no. NC_013162). The C. ochracea sequences of gldK and sprT (Coch_0809 and Coch_1748, respectively) were obtained from the National Center for Biotechnology Information database (www.ncbi.nlm.nih.gov). The C. ochracea ortholog of the F. johnsoniae gene encoding the gliding motility protein SprB (31) was searched against the whole-genome sequence of C. ochracea in the National Center for Biotechnology Information database by means of a BLAST search. The C. ochracea DNA sequence obtained from the search was designated sprB (Coch_0203).

The primers used in the present study are listed in Table 2. To construct the ΔsprT mutant, the upstream and downstream sequences of the target gene were amplified by means of PCR from chromosomal DNA of C. ochracea ATCC 27872 with the primer pairs SprT-F1/SprT-R1 and SprT-F2/SprT-R2, respectively. The ermF-ermAM cassette was amplified from the pVA2198 plasmid (32) by using the primers EMD2 and EMU2. The ermF-ermAM cassette was inserted into the upstream and downstream fragments of the target genes by using an overlap extension PCR method (33).

TABLE 2.

Primers used in the present study

Primer	Sequencea
SprT-F1	5′-ACAGGATTAGATATCTCAGAAGGAA-3′
SprT-R1	5′-TGTTGCAAATACCGATGAGCTGCCCATACAAATTCAATGCTCCT-3′
SprT-F2	5′-CGTTACTAAAGGGAATGTAGCCCAATAGCCCTTGGACAAGTAACCTT-3′
SprT-R2	5′-CATTAGAAATGTTCTCGAATGAGAACTC-3′
EMD2	5′-GCTCATCGGTATTTGCAACATCATAG-3′
EMU2	5′-CTACATTCCCTTTAGTAACGTGTAACTTTC-3′
SprB-F1	5′-TTATAACGTAACGACTGACCCATTT-3′
SprB-R1	5′-AAGGGAATGTAGAATTATAGAGCTCAACGTGCCTATG-3′
SprB-F2	5′-ATACCGATGAGCAAAAGGAATATAATTCTGCCCAAAG-3′
SprB-R2	5′-ATCTACCACGAACAAGCGTATAGAG-3′
EMSprB-F	5′-AGCTCTATAATTCTACATTCCCTTTAGTAACGTGTAACTTTC-3′
EMSprB-R	5′-TATATTCCTTTTGCTCATCGGTATTTGCAACATCATAG-3′
GldK-F1	5′-GATGCCTACATTCAGTGTTGCCAAT-3′
GldK-R1	5′-TGTTGCAAATACCGATGAGCAGCTAGTACTAATAGTACAAGTAGCATTAC-3′
GldK-F2	5′-CGTTACTAAAGGGAATGTAGGCTCCTTACGGTATGACGCTTATTCCGAG-3′
GldK-R2	5′-TCATAGTAGACATATATTCATATGAATTCG-3′

^aUnderlined letters indicate overlap regions. All primers are from this study.

Similarly, for the ΔsprB mutant, the upstream and downstream sequences of sprB were amplified by means of PCR from the chromosomal DNA with the primer pairs SprB-F1/SprB-R1 and SprB-F2/SprB-R2, respectively. For the ΔgldK mutant, the upstream and downstream sequences of the target gene were amplified by means of PCR from the chromosomal DNA with the primer pairs GldK-F1/GldK-R1 and GldK-F2/GldK-R2, respectively. The ermF-ermAM cassette was amplified from the pVA2198 plasmid with the primers EMSprB-F and EMSprB-R (for sprB) or EMD2 and EMU2 (for gldK) and then inserted into the upstream and downstream fragments as described for the ΔsprT mutant.

The PCR-fused fragments from sprT and sprB were cloned into pCR2.1-TOPO (Invitrogen, Carlsbad, CA), and the plasmids were transformed into E. coli DH5α. Transformants were selected on Luria-Bertani agar plates containing kanamycin (25 μg/ml). Recombinant plasmid DNA was isolated and digested with EcoRI. The fragments obtained (sprT or sprB) and the PCR-fused fragments of gldK were introduced into C. ochracea ATCC 27872 by means of electroporation, as described below. Mid-logarithmic-phase C. ochracea ATCC 27872 cells were harvested from 100 ml of culture medium, washed three times with ice-cold distilled water, and suspended in 0.2 ml of 10% glycerol. Then, 10 μg of the DNA fragments was mixed with 40 μl of the cells, and the mixture was incubated on ice for 1 min before being transferred to a 0.1-cm electroporation cuvette (Bio-Rad Laboratories, Hercules, CA). Electroporation was performed with a Gene Pulser II (Bio-Rad) at settings of 1.8 kV, 25 μF, and 250 Ω. Immediately after electroporation, the cells were suspended in 1 ml of tryptic soy broth and incubated overnight at 37°C under anaerobic conditions. The cells were then plated on blood agar plates containing 10 μg/ml erythromycin and were incubated for 5 to 7 days at 37°C under anaerobic conditions. Correct gene replacement in erythromycin-resistant mutants was confirmed by means of PCR (data not shown).

Growth properties of the mutant strains in liquid culture.

C. ochracea strains were grown for 2 days at 37°C on blood agar plates and were inoculated into tryptic soy broth. After reaching the early stationary phase, the culture was diluted with fresh medium to an optical density at 660 nm (OD₆₆₀) of 0.1. Samples of the diluted cultures (10 ml) were then incubated at 37°C under anaerobic conditions. Bacterial growth was monitored by measuring OD₆₆₀ with a spectrophotometer (Mini Photo 518R; Taitec, Tokyo, Japan) at predetermined time points.

Microscopic observation of bacterial movement on solid agar and glass surfaces.

The movement of bacterial cells on solid agar surfaces (i.e., colony spreading) was examined by using phase-contrast microscopy as described previously, with minor modifications (34). In brief, cells in the early stationary phase were suspended in fresh medium to an OD₆₆₀ of 1.0. Five-microliter samples of the cell suspensions were then spotted on glass slides covered with a thin layer of tryptic soy agar (agar content, 3%) supplemented with hemin (5 μg/ml), menadione (0.5 μg/ml), and 0.1% yeast extract (Becton Dickinson). After incubation for 5 days at 37°C under anaerobic conditions, colony spreading was observed by using a stereomicroscope (SMZ800; Nikon, Tokyo, Japan) equipped with a digital camera.

The movement of C. ochracea over a glass surface was also examined by means of phase-contrast microscopy. C. ochracea strains were grown for 2 days at 37°C, under anaerobic conditions, in tryptic soy broth supplemented with 0.1% yeast extract (TSBYE), were dripped on tryptic soy agar supplemented with 0.1% yeast extract, and were incubated for 2 days at 37°C under anaerobic conditions. After incubation, the edges of the colonies were scraped off and resuspended in TSBYE. Tunnel slides were prepared as described previously, by using 5-mm-thick double-sided tape (NW-5; Nichiban, Tokyo, Japan) to hold a glass coverslip above a glass slide (35). Cells suspended in TSBYE were introduced into the tunnel slides, and fresh TSBYE was used to wash away unattached cells. Cell motility was observed under an inverted microscope. Images were recorded at 5-s intervals for 10 min by using a charge-coupled-device camera (Cool-SNAP EZ; Photometrics, Tucson, AZ) and were analyzed by using MetaMorph image analysis software (Molecular Devices, Downingtown, PA). Rainbow traces of cell movements were made by using ImageJ software (version 1.44p) (http://imagej.nih.gov/ij) and the Color FootPrint macro (29).

Western blot analyses.

To prepare polyclonal antibodies against C. ochracea SprB, a peptide corresponding to residues 5271 to 5284 with an added N-terminal cysteine residue (CAGDYWYVLKTHEDG) was synthesized by means of 9-fluorenylmethoxy carbonyl (Fmoc) chemistry and was conjugated to keyhole limpet hemocyanin by using maleimidobenzoic acid N-hydroxysuccinimide ester purchased from Sigma-Aldrich. Rabbit immunization with the peptide conjugate, bleeding, and serum isolation were outsourced to Sigma-Aldrich. C. ochracea cells were grown to early stationary phase at 37°C in tryptic soy broth. The cells were then pelleted by centrifugation at 8,000 × g for 30 min, and the culture supernatants were collected. The supernatants (20 μg of protein) were separated by means of sodium dodecyl sulfate-polyacrylamide gel electrophoresis (5% gels; Cosmo Bio, Tokyo, Japan) under reducing conditions. After electrophoresis, the proteins were transferred onto polyvinylidene difluoride membranes (Bio-Rad), and SprB was detected by using the polyclonal antibodies (1:100 dilution).

Crystal violet biofilm formation assay.

The biofilms formed by the C. ochracea strains were examined in 96-well polystyrene plates (Sumitomo Bakelite, Tokyo, Japan) as described previously, with minor modifications (20). Stationary-phase cultures of C. ochracea were adjusted to an OD₆₆₀ of 0.1 with fresh tryptic soy broth. Cells were then added to 96-well plates (100 μl/well) and incubated for 48 h at 37°C under anaerobic conditions. Following incubation, the wells were washed three times with 200 μl of distilled water and stained with 50 μl of 0.1% crystal violet for 15 min at room temperature. After the crystal violet solution was removed and each well was washed twice with distilled water, the biomass-associated crystal violet was extracted with 200 μl of 99.5% ethanol. The extracted biomass-associated crystal violet (100 μl/well) was transferred to a new microtiter plate, and the OD₅₉₅ was measured with a microplate reader (Spectra MAX M5e; Molecular Device, Sunnyvale, CA). Total biomass was calculated by using the following equation, which was described previously (36): total biomass = absorbance of crystal violet-stained biofilm at 595 nm/absorbance of total growth (including biofilm and planktonic cells) at 660 nm.

Confocal laser scanning microscopic analysis of biofilms.

The biofilms produced by the C. ochracea strains were also examined by means of confocal laser scanning microscopy. One milliliter of cell suspension (OD₆₆₀ of 0.1) was inoculated onto a plastic coverslip (Sumitomo Bakelite), in a 12-well polystyrene microtiter plate (BD Falcon, Franklin Lakes, NJ), at 37°C under anaerobic conditions. After 48 h, the medium was removed and the wells were washed three times with phosphate-buffered saline (PBS) (10 mM, pH 7.4; 2 ml/well) to remove unattached bacteria. The biofilms on the coverslips were fixed overnight at 4°C with 2% paraformaldehyde and 2.5% glutaraldehyde in PBS and were stained with the nucleic acid stains SYTO9 and propidium iodide by using the Live/Dead BacLight bacterial viability kit (Molecular Probes, Eugene, OR) according to the manufacturer's instructions. The biofilms were incubated in the dark at room temperature for 15 min, washed two times, and observed by means of confocal laser scanning microscopy, using a LSM5 DUO microscope (Carl Zeiss MicroImaging Inc., Göttingen, Germany) with a 63×/1.40 oil immersion objective. A series of z-stack images were scanned in increments of 0.1 μm, using excitation wavelengths of 489 and 532 nm. The images were analyzed using Zen 2009 (Carl Zeiss MicroImaging Inc.) and Imaris 7.0.0 (Bitplane AG, Zurich, Switzerland) software. Each z-stack image was further analyzed and quantified for biomass (total mass of living matter in a given unit area), average biofilm thickness, and maximum biofilm thickness by using the COMSTAT program (37).

Scanning electron microscopic analysis of biofilms.

Samples were adjusted and incubated as described for the confocal laser scanning microscopic analysis. After 48 h, the medium was removed and the wells were washed once with PBS. Cells attached to the coverslips were fixed overnight at 4°C with 2% paraformaldehyde and 2.5% glutaraldehyde in PBS. The samples were washed twice with PBS and dehydrated in ethanol. The samples were then dried at the critical point of t-butyl alcohol, coated with gold/palladium by using a SC500A sputter coater (Bio-Rad), and observed under a scanning electron microscope (SU6600; Hitachi, Tokyo, Japan).

Biofilm stability.

The stability of the biofilms produced by C. ochracea was examined by means of the crystal violet biofilm formation assay described above, with modification of the number of washes after removal of the medium, as described previously (38).

Biofilm detachment assay.

Enzymatic treatment was carried out as described previously, with minor modifications (39, 40). Biofilms (produced by 100 μl of bacterial culture) were grown for 48 h in the wells of 96-well polystyrene microtiter plates, as described for the crystal violet biofilm formation assay. The biofilms were washed with 200 μl of distilled water and then treated overnight at 37°C with 200 μl of 0.5 mg/ml proteinase K (Roche Applied Sciences, Indianapolis, IN), DNase I (Roche Applied Sciences), or the carbohydrate-modifying agent sodium metaperiodate (Sigma-Aldrich) in PBS. Control wells received an equal volume of PBS. After treatment, the biofilms were washed twice with 200 μl of distilled water, stained with 50 μl of crystal violet, and then quantitated as described for the crystal violet biofilm formation assay.

Microscopic observation of bacterial attachment to a glass surface.

Attachment of the C. ochracea strains to a glass surface was examined by using tunnel slides, as described previously (41). In brief, a stationary-phase culture of C. ochracea was adjusted to an OD₆₆₀ of 0.5 with fresh tryptic soy broth, and 40 μl of the cell suspension was added to a tunnel slide. After 5 min, 200 μl of medium was added to wash away unattached cells. Cells attached to the slides were visualized under an inverted microscope (IX83; Olympus, Tokyo, Japan).

Autoaggregation assay.

Autoaggregation of the C. ochracea strains was examined by using the assay described previously (20, 42). In brief, cells grown to stationary phase at 37°C under anaerobic conditions were washed twice with PBS and once with a coaggregation buffer consisting of 1 mM Tris-HCl (pH 8.0), 0.1 mM CaCl₂, 0.1 mM MgCl₂, and 150 mM NaCl. Cells were resuspended to an OD₆₆₀ of 0.5 in the same buffer. One milliliter of each cell suspension was placed in a cuvette, and the OD₆₆₀ values at 0 and 120 min were determined with a spectrophotometer. The rate of autoaggregation was calculated by using the following equation: rate of autoaggregation = 100 − [(OD₆₆₀ at 120 min/OD₆₆₀ at 0 min) × 100].

Statistical analysis.

Each experiment was performed independently at least twice. Results are expressed as mean values with standard deviations (SDs). One-way analysis of variance (ANOVA) with the Tukey-Kramer multiple-comparison test was used to determine differences between groups in the crystal violet biofilm formation assay. The Kruskal-Wallis test followed by Dunn's multiple-comparison test was used in the other experiments. P values of <0.05 were considered statistically significant. Statistical calculations were performed with the Prism software package (version 6.04; GraphPad Software, La Jolla, CA).

RESULTS

Growth of the ΔgldK, ΔsprT, and ΔsprB mutants.

To determine the effects of gldK, sprT, or sprB ortholog deficiency on the growth of C. ochracea, the growth of the wild-type and mutant strains of C. ochracea was monitored under anaerobic conditions at 37°C. Significant differences in growth between the wild-type and mutant strains were observed during the late logarithmic and early stationary phases (Fig. 1).

FIG 1.

Growth curves for wild-type and ortholog-deficient mutant strains of Capnocytophaga ochracea. Wild-type C. ochracea ATCC 27872 (closed black squares), ΔgldK mutant (open light-gray squares), ΔsprT mutant (open gray circles), and ΔsprB mutant (closed dark-gray triangles) strains were grown at 37°C in tryptic soy broth supplemented with hemin and menadione under anaerobic conditions. Data are presented as means ± SDs (n ≥ 8). ***, P < 0.001 (ΔgldK mutant versus wild-type strain); ‡, P < 0.01 (ΔsprT mutant versus wild-type strain); §, P < 0.05 (ΔsprB mutant versus wild-type strain).

Movement of the mutant strains on solid agar and glass surfaces.

To investigate the effects of gldK, sprT, or sprB ortholog deficiency on gliding motility, colony spreading of the wild-type and mutant strains on a solid agar surface was examined under a phase-contrast microscope (Fig. 2A). The wild-type strain formed colonies that exhibited marked spreading across the solid agar surface, whereas the mutant strains formed colonies that exhibited little spreading. The movement of the wild-type and mutant strains on a glass surface was also examined. The wild-type strain exhibited movement over the glass surface, but the mutant strains did not (Fig. 2B). Together, these results indicate that SprB is crucial for the movement of C. ochracea across solid surfaces and that SprB is secreted via the T9SS.

FIG 2.

Effects of gldK, sprT, or sprB ortholog deficiency on gliding motility in Capnocytophaga ochracea. (A) Effects of gldK, sprT, or sprB ortholog deletion on colony spreading on a solid agar surface. Representative photographs of the colonies and photomicrographs of the edges of the colonies of each strain are shown. Black arrow, cells spreading away from the colony; scale bars, 100 μm. (B) Effects of gldK, sprT, or sprB ortholog deletion on gliding motility on a glass surface. Rainbow traces of cell movements were created with image analysis software. Nonmotile cells are shown in white. Scale bars, 10 μm. WT, wild-type C. ochracea ATCC 27872; ΔgldK, gldK ortholog-deficient C. ochracea mutant; ΔsprT, sprT ortholog-deficient C. ochracea mutant; ΔsprB, sprB ortholog-deficient C. ochracea mutant.

Involvement of SprT in the secretion of SprB.

In F. johnsoniae, T9SS proteins such as GldK, GldL, and SprT are necessary for secretion of the gliding motility adhesin SprB (30). Therefore, we examined the role of SprT in the secretion of SprB in C. ochracea by means of Western blotting (Fig. 3). SprB was detected in the culture supernatant of the wild-type strain but not in the supernatant of the ΔsprT mutant, indicating that the ΔsprT mutant did not secrete SprB. The size of the band detected was approximately equal to the estimated molecular mass of SprB (564,410.11 Da).

FIG 3.

Western blot analysis of the culture supernatants of wild-type Capnocytophaga ochracea ATCC 27872 and the ΔsprT mutant. The culture supernatants were subjected to sodium dodecyl sulfate-polyacrylamide gel electrophoresis and Western blot analysis with an antibody against SprB. Lane M, molecular size marker; lane WT, wild-type C. ochracea ATCC 27872; lane ΔsprT, sprT ortholog-deficient C. ochracea mutant. The size of the band observed for wild-type C. ochracea was approximately equal to the estimated molecular mass of SprB (564,410.11 Da).

Biofilm formation by the mutant strains.

Each mutant strain showed significantly attenuated biofilm formation, compared with that of the wild-type strain (Fig. 4A and B). The amounts of crystal violet-associated biomass for the ΔgldK, ΔsprT, and ΔsprB strains at 48 h were 49%, 34%, and 65% of that for the wild-type strain, respectively, indicating that the T9SS was involved in biofilm formation. To investigate further the effects of gldK, sprT, or sprB ortholog deficiency on biofilm formation, we examined the three-dimensional structures of the biofilms by means of confocal laser scanning microscopy and the COMSTAT program. The wild-type biofilm appeared denser than the mutant biofilms (Fig. 4C), and the biomass, average thickness, and maximum thickness of the ΔgldK, ΔsprT, and ΔsprB biofilms were significantly smaller than those of the wild-type biofilm (Fig. 4D). Furthermore, the amounts of crystal violet-associated biomass and the average and maximum thicknesses of the ΔgldK and ΔsprT biofilms were significantly smaller than those of the ΔsprB biofilm (Fig. 4A and D). Together, these results indicate not only that SprB secreted via the T9SS is involved in biofilm formation but also that, because biofilm formation was not completely attenuated, other proteins are involved in biofilm formation by C. ochracea.

FIG 4.

Effects of gldK, sprT, or sprB ortholog deficiency on biofilm formation by Capnocytophaga ochracea. (A) Total biomass of biofilms. Cells were incubated for 48 h at 37°C under anaerobic conditions, and the amount of biofilm produced was quantified by measuring the OD₅₉₅ following crystal violet staining. Total biomass was calculated by using the following equation: total biomass = absorbance of crystal violet-stained biofilm at 595 nm/absorbance of total growth (including biofilm and planktonic cells) at 660 nm. Data are presented as means ± SDs (n = 18). *, P < 0.05; ***, P < 0.001. (B) Representative images of the results of the crystal violet biofilm formation assay. (C) Confocal laser scanning microscopic analysis of C. ochracea biofilms. C. ochracea strains were incubated for 48 h at 37°C under anaerobic conditions and were stained with the nucleic acid stains SYTO9 and propidium iodide. Images are presented as orthographic projections. (D) COMSTAT analyses of the three-dimensional structures of the biofilms produced by the wild-type and mutant strains of C. ochracea. Data are presented as means ± SDs (n = 19). *, P < 0.05; **, P < 0.01; ***, P < 0.001. WT, wild-type C. ochracea ATCC 27872; ΔgldK, gldK ortholog-deficient C. ochracea mutant; ΔsprT, sprT ortholog-deficient C. ochracea mutant; ΔsprB, sprB ortholog-deficient C. ochracea mutant. Scale bars, 10 μm.

Scanning electron microscopic analysis of the structure of biofilms.

Bacterial biofilms are usually supported by a matrix of biopolymers known as extracellular polymeric substances (EPSs) (43). We next used a scanning electron microscope to analyze the structures of the biofilms formed by the C. ochracea strains. The wild-type strain formed a denser biofilm, with the cells packed together more closely, than did the mutant strains (Fig. 5).

FIG 5.

Scanning electron microscopic analysis of the structures of the C. ochracea biofilms. The biofilms were cultured at 37°C for 48 h before fixation. Scale bars, 20 μm for ×2,000 magnification and 2 μm for ×20,000 magnification. WT, wild-type C. ochracea ATCC 27872; ΔgldK, gldK ortholog-deficient C. ochracea mutant; ΔsprT, sprT ortholog-deficient C. ochracea mutant; ΔsprB, sprB ortholog-deficient C. ochracea mutant.

Stability of the biofilms produced by the mutant strains.

To investigate the stability of the biofilms produced by the mutants, the crystal violet biofilm formation assay was used but with different amounts of washing (Fig. 6). After a single wash, the decreases in the amounts of crystal violet-associated biomass were significantly greater for the ΔgldK and ΔsprT strains than for the wild-type strain. After three washes, the decreases in the amounts of crystal violet-associated biomass for all mutant strains were significantly greater than that for the wild-type strain. These results indicate that deletion of gldK, sprT, or sprB in C. ochracea reduces the stability of the C. ochracea biofilm.

FIG 6.

Effects of sprT or sprB ortholog deficiency on biofilm stability. Capnocytophaga ochracea biofilms were incubated for 48 h at 37°C under anaerobic conditions and then were washed a predetermined number of times. Black bar, 0 washes; white bar, 1 wash; light-gray bar, 3 washes. Percentages indicate the amounts of biomass remaining after each wash, calculated by using the following equation: (biomass remaining after each wash/biomass without washing) × 100. Data are presented as means ± SDs (n = 18). *, P < 0.05; ***, P < 0.001. WT, wild-type C. ochracea ATCC 27872; ΔgldK, gldK ortholog-deficient C. ochracea mutant; ΔsprT, sprT ortholog-deficient C. ochracea mutant; ΔsprB, sprB ortholog-deficient C. ochracea mutant.

Enzyme treatment of biofilms.

EPSs can be proteins, nucleic acids, or polysaccharides (44, 45). Therefore, to characterize the structure of the biofilm further, we treated mature C. ochracea biofilms with proteinase K, DNase I, or NaIO₄. Treatment with proteinase K significantly reduced the amounts of biofilm produced by the wild-type, ΔsprT, and ΔsprB strains, indicating that biofilms produced by these strains were predominantly proteinaceous (Fig. 7). DNase I treatment slightly reduced the amounts of biofilm produced by each strain, although not statistically significantly.

FIG 7.

Effects of enzyme treatment on the biofilms produced by Capnocytophaga ochracea. (A) Total biomass of biofilms. Biofilms were grown for 48 h at 37°C under anaerobic conditions, treated with 0.5 mg/ml proteinase K, DNase I, or sodium metaperiodate (NaIO₄), and then incubated overnight at 37°C. After being washed, the biofilms were stained with 0.1% crystal violet. Data are presented as means ± SDs (n = 9). *, P < 0.05; **, P < 0.01; ***, P < 0.001. (B) Representative photographs of the results of crystal violet staining after enzymatic treatment of C. ochracea biofilms. WT, wild-type C. ochracea ATCC 27872; ΔsprT, sprT ortholog-deficient C. ochracea mutant; ΔsprB, sprB ortholog-deficient C. ochracea mutant.

Attachment to a glass surface and autoaggregation of the mutant strains.

The adherence of bacteria to a solid surface is the first step in biofilm formation (44). Therefore, the effects of gldK, sprT, or sprB ortholog deficiency on the ability of C. ochracea to attach to a glass surface were examined by means of phase-contrast microscopy. Wild-type and ΔsprT strains exhibited comparable adherence to the glass surface; however, the ΔgldK and ΔsprB strains exhibited significantly less adherence than did the wild-type cells (Fig. 8).

FIG 8.

Effects of gldK, sprT, or sprB ortholog deficiency on the attachment of Capnocytophaga ochracea to a glass surface. The number of cells attached to the coverslip was counted. Data are presented as means ± SDs (n = 15). ***, P < 0.001. WT, wild-type C. ochracea ATCC 27872; ΔgldK, gldK ortholog-deficient C. ochracea mutant; ΔsprT, sprT ortholog-deficient C. ochracea mutant; ΔsprB, sprB ortholog-deficient C. ochracea mutant.

Autoaggregation of the mutant strains was also examined, because it may play an important role in cellular adherence and biofilm formation (46). However, the autoaggregation rates of the wild-type, ΔgldK, ΔsprT, and ΔsprB strains were 6.7% ± 4.3%, 2.8% ± 2.5%, 5.7% ± 2.2%, and 7.0% ± 1.3%, respectively (data are presented as means ± SDs), indicating that autoaggregation was not affected by the deletion of sprT or sprB.

DISCUSSION

Bacterial surface molecules such as proteins are important for the attachment of bacterial cells to surfaces and for biofilm formation, because they interact directly with biotic or abiotic surfaces (47, 48). The T9SS is a major protein transport system for bacteria in the phylum Bacteroidetes, and it has been shown to be essential for the secretion of surface motility adhesins in F. johnsoniae (30). Here we demonstrated that GldK, SprT, and SprB were involved in the biofilm formation of C. ochracea.

Motility is an important factor for biofilm formation by Gram-negative bacteria. For E. coli, highly motile strains display biofilm structures that extend vertically, whereas strains with less motility display smoother microcolonies (49). A recent study with Flavobacterium spp. suggests that gliding motility is important for bacterial attachment to and colonization of plant surfaces (50). As C. ochracea demonstrates gliding motility, it is possible that this motility plays an important role in colonization on the tooth surface. In the present study, we showed that deletion of the C. ochracea orthologs of gldK, sprT, and sprB resulted in defective gliding motility on solid surfaces (Fig. 2). In F. johnsoniae, SprB is transported by the T9SS, is propelled along a left-handed helical loop on the cell surface, and is involved in gliding motility (25, 29). Similarly, the results of the present study indicate that SprB of C. ochracea is transported by the T9SS (Fig. 3) and is involved in gliding motility (Fig. 9). However, the detailed mechanism for the involvement of SprB in the gliding motility of C. ochracea, such as the movement of SprB on the cell surface, is unclear. The reductions in biofilm formation, relative to that of the wild-type strain, were greater for the ΔgldK and ΔsprT mutant strains than for the ΔsprB strain, even after correction for bacterial growth to avoid the effects of differences in growth rates (Fig. 4). This suggests that gliding motility is partly involved in biofilm formation by C. ochracea and that proteins exported by the T9SS other than SprB also play certain roles in biofilm formation.

FIG 9.

Model of the involvement of GldK, SprT, and SprB in attachment, gliding, and biofilm formation by C. ochracea. SprB is translocated to the cell surface and extracellular milieu via the T9SS. Subsequently, SprB attaches to and glides on solid surfaces (see the text for details). CM, cytoplasmic membrane; OM, outer membrane.

Many bacteria that belong to the phylum Bacteroidetes secrete extracellular and surface proteins via the T9SS. The present data indicate that GldK and SprT, which are components of the T9SS (30), participate in biofilm formation. Several proteins with conserved C-terminal domain (CTD) regions, which are required for secretion by the T9SS, were detected in the genome sequence of C. ochracea (data not shown); these may be secreted by the T9SS. In the present study, ΔgldK, ΔsprT, and ΔSprB mutants showed a general trend for reductions in biofilm formation (Fig. 4) and adherence (Fig. 8), suggesting that the proteins secreted by the T9SS are involved in biofilm formation. It is important to note that the level of reduction of the ΔgldK mutant was somewhat low compared with that of the ΔsprT mutant. GldK is required for secretion by the F. johnsoniae T9SS, and SprT also has important roles in T9SS-mediated secretion. In F. johnsoniae, knockout of GldK completely inactivated T9SS function, but SprT knockout only partially inhibited it (30). This may be one reason for the differences in the reductions of biofilm biomass and thickness in C. ochracea. These results collectively suggest that GldK and SprT participate in biofilm formation via T9SS-secreted proteins, including SprB (Fig. 9). Further study by proteomic analysis is necessary to clarify the locations of these proteins in the three-dimensional structure of the T9SS.

Bacterial adherence is another important factor for biofilm formation (51). Previous studies showed that ΔgldK and ΔsprT mutants of F. johnsoniae failed to attach to a glass surface (25, 30). In contrast, the ability of the ΔgldK mutant of C. ochracea to adhere to a glass surface decreased in the present study, but this decrease was much smaller than that for the ΔgldK mutant of F. johnsoniae. In addition, the ability of the ΔsprT mutant of C. ochracea to adhere to a glass surface was comparable to that of the wild-type strain (Fig. 8), suggesting that C. ochracea possesses other secretion systems that translocate adhesins to the cell surface and that C. ochracea GldK and SprT participate more in the maintenance of the three-dimensional architecture of the biofilm than in bacterial adhesion. Our results also indicate that autoaggregation does not play a major role in biofilm formation by C. ochracea.

Generally, biofilms cannot be removed easily by fluid shear stress (44). In the present study, however, the amounts of ΔgldK and ΔsprT biofilms remaining after washing were dramatically decreased, compared with the wild-type biofilm (Fig. 6). This result confirms that, in C. ochracea, GldK and SprT are involved in the maintenance of the biofilm architecture. In contrast, the amount of biofilm formed by the ΔsprB mutant was significantly decreased after three washes but not after one wash (Fig. 5 to 7), suggesting that, in C. ochracea, SprB also participates in the maintenance of the biofilm architecture. However, the degree of involvement of SprB in the maintenance of the biofilm architecture is likely less than that of GldK or SprT, which is reasonable, given that SprB is one of the proteins secreted via the T9SS. The present results also indicate that there are other substances secreted via the T9SS that are crucial for the maintenance of the biofilm architecture, because biofilm formation was not completely attenuated in the mutants.

EPSs provide mechanical stability for bacterial biofilms, keep bacterial cells in close proximity, mediate bacterial adhesion to surfaces, and form a cohesive, three-dimensional, polymer network that interconnects and transiently immobilizes cells within the biofilm (43). Our scanning electron microscopic analysis indicated that, compared with biofilms produced by the mutant strains, the wild-type biofilm was denser and bacterial cells were closer together (Fig. 5). Therefore, these results suggest that the differences in EPSs are responsible for the observed differences in biofilms. Moreover, treatment with proteinase K significantly reduced the total mass of the C. ochracea biofilm, whereas treatment with DNase I or NaIO₄ did not (Fig. 7), indicating that the main components of the biofilm were proteinaceous and the proteinaceous substances were responsible for the strength and amounts of the biofilm produced by C. ochracea (Fig. 9). We treated the biofilm with NaIO₄ to investigate the existence of polysaccharide in EPSs, because it was reported that C. ochracea produces exopolysaccharide containing large amounts of mannose, with smaller quantities of glucose, galactose, glucuronic acid, and glucosamine (52, 53). NaIO₄ oxidizes acidic polysaccharides and destroys their mannose residues (54). The results suggest that the mannose-rich exopolysaccharide was not the major component in the biofilm architecture. However, the results are not sufficient to rule out the involvement of other types of polysaccharides in the biofilm architecture. Further analysis with a staining procedure is necessary.

Altered biofilm formation by T9SS-component mutants has been reported for Tannerella forsythia, an important periodontal pathogen (36, 55, 56). The T. forsythia mutants lacked the surface layer (S-layer), which contains the cell surface glycoproteins TfsA and TfsB (36, 56). The S-layer is known to participate in T. forsythia adhesion to and invasion of host cells and in T. forsythia-mediated suppression of proinflammatory cytokine expression. A T. forsythia mutant deficient in the S-layer exhibited decreased hemagglutination and increased biofilm formation (36, 56). These findings contradict our observation that the amounts of biofilm formed by the C. ochracea mutant strains were smaller than the amount formed by the wild-type strain. This difference may be attributable to whether the bacteria possess an S-layer, but further analysis is required to understand the differences in biofilm formation between the T. forsythia and C. ochracea mutant strains. Analysis of the biofilms formed by T9SS-component mutants of other members of the phylum Bacteroidetes may also help us to understand in detail the role of the T9SS in biofilm formation.

In the present study, we demonstrated the involvement of the T9SS in C. ochracea biofilm formation. C. ochracea is found mainly in the oral cavity and is known to coaggregate with F. nucleatum (1, 20, 51). In dental plaque biofilms, F. nucleatum is a core bacterium for the growth of periodontopathic bacteria, including Porphyromonas, Fusobacterium, and Prevotella spp. It was recently reported that Capnocytophaga spp. increase their dominance within the bacterial community as dental plaque matures (57). Furthermore, a number of other bacteria, including major periodontal pathogens, have been found to harbor T9SS-related genes (28). Therefore, inhibitors of the T9SS represent a potential means of controlling the development of periodontopathic biofilms. Further clarification of the roles of the T9SS in periodontal pathogens is now needed.

Funding Statement

Japan Society for the Promotion of Science (JSPS) provided funding to Kazuyuki Ishihara via a Grant-in-Aid for Scientific Research (C) (no. 24592778). JSPS provided funding to Atsushi Saito via a Grant-in-Aid for Scientific Research (C) (no. 25463228).