INTRODUCTION

Top Abstract Introduction Materials and Methods Results and Discussion References

In the past, the science of microbiology has developed mainly from studies focused on free-floating (planktonic) cells living in batch culture. While a wealth of information about basic microbial growth and physiology has been gained from these studies, it is now widely accepted that in most natural, industrial, and medical environments, the majority of bacteria exist in highly structured, surface-attached communities (5, 6). These structured, microbial communities, known as biofilms, develop on virtually every material that contacts naturally occurring fluids, including those of medical and industrial importance. Mounting evidence indicates that bacterial biofilms are responsible for many persistent and chronic bacterial infections in humans. Biofilm infections may be caused by a single species or by a mixture of species of bacteria and fungi and are associated with such diseases as dental caries, periodontitis, otitis media, musculoskeletal infections, cystic fibrosis pneumonia, native valve endocarditis, and bacterial prostatitis (5, 6, 32). In addition to infections of living tissues, microbial biofilms foul implanted medical devices, such as catheters, artificial cardiac pacemakers, prosthetic heart valves, and orthopedic appliances. Chronic infection of medical devices can lead to incidences of acute sepsis and death, particularly in immunocompromised patients (6), and is the greatest problem affecting the success of biomedical implants within the human body (22).

One of the most common biofilms, dental plaque, is found in the human oral cavity. Total accumulation of microorganisms in plaque on all teeth can range from up to a few milligrams to more than gram (21). The sanguis streptococci are primary colonizers of the tooth surface (2, 25, 36). In addition to comprising a major portion of dental plaque, the sanguis streptococci serve as a substratum for the subsequent adhesion of other plaque bacteria. These later-arriving bacteria include species that are the causative agents of caries and periodontal diseases, some of the most prevalent infections afflicting humans (21).

The sanguis streptococci and related organisms are also a common cause of native heart valve (18, 42) and late prosthetic valve (28) endocarditis. Not all bacteria are capable of causing endocarditis, but these bacteria appear to have specific properties that enable them to colonize a modified valve surface (1, 12, 24, 27). Once the valve surface is colonized, formation of the biofilm vegetation proceeds by a series of complex steps (10, 39). Small surface aggregates of bacteria begin to form, and eventually rounded colonies tightly packed with large numbers of bacteria occur. When the bacterial population reaches a certain size, a layer of fibrin and platelets is deposited over the bacteria. The colonized valve surface is thus transformed into a biofilm vegetation with an organized appearance (10, 23, 40). Because these infections are often persistent and difficult to treat, infective endocarditis is a serious disease resulting in substantial morbidity and mortality despite modern antimicrobial and surgical treatment (9). Since cells grown on surfaces within biofilms are in a physiological state that differs markedly from that of their planktonic counterparts (5, 6, 37), new therapies that target the biofilm phenotype are needed.

Factors involved in the transition from a free-floating, planktonic existence to that of a surface-attached, sessile community are complex, and little is known about the molecular mechanisms of biofilm formation of oral streptococci. Recently, several investigators have demonstrated the power of transposon mutagenesis coupled with a convenient assay for biofilm formation in identifying genes required for bacterial biofilm formation (30, 33, 44, 26). Mutants are tested in 96-well microtiter plates, and biofilm formation within the well is detected by staining with crystal violet. A modification of this assay was used in the present study to begin characterization of biofilm formation by the oral bacterium Streptococcus parasanguis, a primary colonizer of the tooth surface and a known causative agent of bacterial endocarditis (1).

Previous studies with S. parasanguis FW213 have determined that adherence of this organism to an in vitro tooth surface model, saliva-coated hydroxlapatite (SHA), is mediated by long peritrichous surface fimbriae (13, 16). Wild-type, fimbriated FW213 binds to SHA, but isogenic mutants lacking these fimbriae do not. Recent analyses of genetic determinants encoding the biogenesis of these fimbriae revealed that Fap1, a high-molecular-weight glycoprotein, is essential for long fimbrial formation in S. parasanguis FW213 (45, 46). Insertional inactivation of fap1 results in loss of the long peritrichous surface fimbriae and a significant reduction in adhesion of FW213 to SHA (45, 46). In this report, we present data showing that S. parasanguis fap1 mutants are also defective in biofilm development.

	MATERIALS AND METHODS

Top Abstract Introduction Materials and Methods Results and Discussion References

Bacterial strains and media. Bacterial strains used in this study are listed in Table 1. The wild-type S. parasanguis strain used was FW213 (4). All other strains are derivatives of FW213 and were generated by gene replacement techniques. Strains were grown statically in the presence of 5% CO₂ at 37°C in Todd-Hewitt broth (TH; Difco Laboratories, Detroit, Mich.). Biofilm formation of S. parasanguis FW213 was assessed in each of the following media: TH, TH supplemented with 0.2% yeast extract (TH+YE), TH supplemented with glucose to give a final concentration of 1% (TH+Glu), brain heart infusion (BHI; Difco Laboratories), chemically defined medium (CDM [41]), Trypticase soy broth (TSB; Becton Dickinson and Company, Sparks, Md.), and TSB supplemented with 0.2% yeast extract (TSB+YE). The effect of glucose concentration on FW213 biofilm formation was examined by supplementing TH, BHI, CDM, or TSB with glucose to a final concentration of 1% (wt/vol).

Biofilm formation assay. The biofilm formation assay used in this study was adapted from the method of O'Toole and Kolter (31) and is based on the ability of bacteria to form biofilms on solid surfaces, such as polystyrene or polyvinyl chloride. S. parasanguis strains grown overnight in TH were washed once in sterile distilled (dH₂O) and diluted 1:200 in fresh medium; 200 µl was inoculated into wells of a non-tissue culture-treated polystyrene flat-bottom 96-well microtiter plate (Nunc 269787; Krackeler Scientific, Inc., Albany, N.Y.). Non-tissue culture-treated polyvinyl chloride round-bottom 96-well plates (Falcon 3911; Becton Dickinson Labware, Franklin Lakes, N.J.) were also used in initial studies. Wells filled with only growth media were included as negative controls. Plates were incubated at 37°C for 16 h either aerobically in 5% CO₂ or anaerobically. Before biofilm quantification, growth of wild-type and mutant strains was assessed by measuring the absorbance of cultures in the wells at 490 nm with an EL311 automated microplate reader (BIO-TEK Instruments, Inc., Winooski, Vt.). Media, including any unattached bacteria, were then decanted from the wells, and any remaining planktonic cells were removed by rinsing with dH₂O. Wells were air dried, and adherent bacteria were stained for 15 min with a 0.5% (wt/vol) solution of crystal violet (Fisher Scientific, Pittsburgh, Pa.). After rinsing with dH₂O, bound dye was released from stained cells using either 95% ethanol, ethanol-acetone (80:20), or 30% glacial acetic acid. This allowed indirect measurement of biofilms formed on both the bottom and sides of the well. Biofilm formation was quantified by measuring absorbance of the solution at 562 nm with a microplate reader (BIO-TEK Instruments).

Time course assay. Early-stationary-phase cultures of S. parasanguis were diluted 1:50 in TH+Glu, and 200 µl of subculture was inoculated into the wells of a polystyrene microtiter plate and allowed to incubate at 37°C, 5% CO₂ from 1 to 14 h. Before biofilm quantification, bacterial growth at each time point was assessed by measuring the absorbance of cultures in the wells at 490 nm with a microplate reader. Wells were subsequently rinsed and stained with crystal violet, and biofilm formation was quantified as described above.

Phase-contrast microscopy. Direct visualization of S. parasanguis biofilm formation on vinyl coverslips (Structure Probe, Inc., West Chester, Pa.) over time was achieved using phase-contrast microscopy. For earlier time points (1 to 6 h), 12 ml of a culture, grown to mid-logarithmic phase in TH+Glu, was transferred to a 50-ml Corning centrifuge tube containing a vinyl coverslip. Incubation was continued without shaking at 37°C, 5% CO₂. At chosen time points, planktonic cells were removed from the coverslip by rinsing, and attached cells were examined by phase-contrast microscopy using a Nikon Eclipse E400 microscope. Digitized images were captured using a SpotRT monochrome camera driven by Spot version 3.0.1 (AppleEvent) software (Diagnostic Instruments, Inc., Sterling Heights, Mich.). For later time points (15 to 18 h), late-logarithmic-phase cultures of S. parasanguis were diluted 1:200 in TH+Glu. Twelve milliliters of each subculture was added to a 50-ml Corning centrifuge tube containing a plastic coverslip and incubated at 37°C, 5% CO₂ for 15 to 18 h. Plastic coverslips were then removed and rinsed, and remaining cells were visualized as described above.

Quantification of Fap1. Surface expression of Fap1 was determined by a whole-bacterial-cell enzyme-linked immunosorbent assay (bactELISA [11]) as described previously (45). Incubation with mouse monoclonal antibody F51 (13), which is specific to the Fap1 subunit of S. parasanguis fimbriae, was used as the primary antibody in the assay. Primary antibody was detected using anti-mouse horseradish peroxidase-conjugated secondary antibody (Jackson ImmunoResearch Laboratories, Inc., West Grove, Pa.). Color development was quantified by measurement of absorbance at 490 nm with an automated microplate reader (BIO-TEK Instruments).

	RESULTS AND DISCUSSION

Top Abstract Introduction Materials and Methods Results and Discussion References

S. parasanguis biofilm formation on polystyrene. A number of recent studies have used molecular and genetic approaches to identify genes important for biofilm formation (reviewed in reference 34). One genetic approach that has been particularly fruitful in identifying genes critical for biofilm formation uses a convenient macroscopic assay to screen for biofilm-defective mutants (31). This assay is based on the ability of bacteria to form biofilms on surfaces such as polystyrene (16). Bacteria are cultured in the wells of plastic tissue culture plates, and the presence or absence of a biofilm is detected by staining the wells with crystal violet. In the present study, characterization of S. parasanguis FW213 biofilm formation was initiated by investigating the ability FW213 to form biofilms on plastic surfaces.

View larger version (62K): [in this window] [in a new window]   FIG. 1.   Bacterial growth and biofilm formation of S. parasanguis FW213 under different growth conditions. Cells grown in TH to early stationary growth phase were washed once in sterile water and subcultured (1:200) into either TH, TH+YE, BHI broth, CDM, TH+Glu, TSB, or TSB+YE. These cultures were then grown in polystyrene microtiter dishes at 37°C for 16 h either aerobically in 5% CO2 (A) or anaerobically (B). Growth (light bars) and biofilm formation (dark bars) were quantified as optical density at 490 and 562 nm, respectively. Assays were performed in quadruplicate; mean values and standard deviations are shown. A representative row of crystal violet-stained microtiter plate wells is shown above each graph.

View larger version (41K): [in this window] [in a new window]   FIG. 2.   Effect of glucose on S. parasanguis growth and biofilm formation. Cells grown in TH to early stationary growth phase were washed once in sterile water and then subcultured (1:200) into the media indicated. The percentage of glucose in each medium is given in parentheses. Cultures were grown aerobically (5% CO2) at 37°C in polystyrene microtiter dishes for 16 h. Representative crystal violet-stained wells are shown above the graph. Growth (light bars) and biofilm formation (dark bars) are quantified below as optical density at 490 and 562 nm, respectively. Assays were performed in quadruplicate; mean values and standard deviations are shown.

Environmental factors affect S. parasanguis biofilm formation. Previous studies indicate that the nutritional content of the growth medium can regulate biofilm development (5, 26, 31). In general, bacteria tend to adhere to available surfaces and form mature biofilms in environments that provide sufficient nutrients but will not adhere to surfaces in environments that are nutrient deficient. S. parasanguis FW213 biofilm formation was tested in a variety of growth media under aerobic (Fig. 1A) and anaerobic (Fig. 1B) conditions. Under both aerobic and anaerobic conditions, growth was about 1.2- and 1.6-fold greater in CDM and TH+Glu, respectively, than in TH. By contrast, biofilm formation increased significantly, about 2.6-fold in CDM and 5.1-fold in TH+Glu, under aerobic conditions but by a much smaller increment under anaerobic conditions. The final concentration of glucose in both CDM and TH+Glu is 1.0%, about four to five times greater than that in TH, BHI, or TSB.

Analysis of S. parasanguis mutants. A number of mutants of FW213 available in our laboratory (Table 1) were examined in the above assay to determine their ability to form a biofilm (Fig. 3). In the biofilm assay, all of the mutants grew to the same extent as wild-type FW213 (Fig. 3). However, two of the mutants, VT1393 (45) and VT1428 (46), displayed severe defects in biofilm formation (Fig. 3). Both mutant strains had a growth rate indistinguishable from that of wild type in liquid medium (data not shown). Both VT1393 and VT1428 contain an inactivated copy of fap1. VT1393 has an insertion mutation in the 5' non repetitive region of fap1 (45), while VT1428 has an insertion mutation at a site just upstream of the 3' cell wall sorting signal (46). Previous studies in our laboratory have demonstrated that Fap1 is essential for formation of long fimbriae by S. parasanguis FW213 and that Fap1 is a structural subunit of the long fimbriae (45, 46). Long fimbriae are absent on the surface of Fap1-deficient strains, while monoclonal antibodies localize Fap1 to the long fimbriae and detect Fap1 as a major component of purified fimbriae (45, 46).

View larger version (57K): [in this window] [in a new window]   FIG. 3.   Growth and biofilm formation of S. parasanguis FW213 and its isogenic mutant strains. S. parasanguis strains FW213 (wild type), VT930 (fimA), VT1393 (fap1), VT1428 (fap1), VT1346 (pepO), and VT1354 (recA) were grown in TH to early stationary growth phase, washed once in sterile water, and then subcultured (1:200) into TH+Glu. Cultures were incubated in polystyrene microtiter dishes at 37°C, 5% CO2 for 17 h. Crystal violet-stained wells are shown above the graph. Growth (light bars) and biofilm formation (dark bars) are quantified below as optical density at 490 and 562 nm, respectively. Assays were performed in quadruplicate; mean values and standard deviations are shown.

Time course assay of biofilm development of wild-type FW213 and of a fap1 mutant. The difference in biofilm formation between wild-type FW213 and the fap1 mutants prompted us to study biofilm development in these strains in more detail. Biofilm development by both wild-type FW213 and the fap1 mutant VT1393 was studied over a period of 14 h (Fig. 4). The wild-type strain displayed a time-dependent increase in biofilm formation that reached a plateau, as detected by crystal violet staining after about 10 h. By contrast, FW213 cells lacking Fap1, and hence the long peritrichous fimbriae, were severely hindered in the ability to form biofilms and displayed only low levels of crystal violet staining over the course of the experiment. The apparent differences in biofilm-forming competency could not be explained by unequal growth, as growth of the fap1 mutant was equivalent to that of the wild type at each time point (data not shown).

View larger version (30K): [in this window] [in a new window]   FIG. 4.   Biofilm development over 14 h for S. parasanguis FW213 and fap1 mutant VT1393. Cells were grown in TH to early stationary growth phase, washed once in sterile water, and diluted 1:50 in TH+Glu. Cells were then grown in polystyrene microtiter dishes, and at the times indicated, biofilm formation was assayed. (A) Representative crystal violet-stained wells at each time point for FW213 and the fap1 mutant. (B) Quantification of crystal violet staining. Assays were performed in quadruplicate; mean values and standard deviations are shown.

Microscopic analysis of wild-type and fap1 mutant biofilms. The detection of biofilm formation in the above studies was done indirectly by determining the amount of crystal violet-stained cells that attach to wells of polystyrene microtiter dishes. Biofilm formation, however, is considered a complex developmental process (5, 7, 30, 31, 44) that can be broken down into a series of observable steps (5, 44). These steps are likely to include attachment and immobilization on a surface, cell-to-cell attachment, microcolony formation, formation of a confluent biofilm, and formation of a characteristic three-dimensional biofilm structure. To outline steps involved in biofilm formation by S. parasanguis FW213, we used an experimental system for the development of three-dimensional biofilms as reported by Watnick and Kolter (44). We also used this system to help identify the role(s) of Fap1 and fimbriae during FW213 biofilm formation.

View larger version (124K): [in this window] [in a new window]   FIG. 5.   Phase-contrast micrographs comparing biofilm formation of wild-type S. parasanguis FW213 and the fap1 mutant VT1393 on plastic coverslips over time. (A to C) FW213 biofilms taken at 1, 3, and 18 h after inoculation, respectively; (D to F) fap1 mutant biofilms taken at comparable times. The arrow indicates one of many microcolonies that form over the course of the experiment. Bar = 30 µm.

Effect of glucose on surface expression of Fap1. Given our findings that Fap1 and fimbriae play a role in FW213 biofilm formation and that glucose promotes biofilm formation in FW213, we asked whether glucose concentration in the growth medium affected the surface expression of Fap1 and fimbriae. Utilizing a bactELISA with a monoclonal antibody specific to Fap1, we determined that an increase in glucose concentration did not increase surface expression of Fap1. This result suggests that factors other than Fap1 and fimbriae were responding to the increased glucose concentration and were contributing to FW213 biofilm formation in these experiments.

Conclusions. This study was undertaken to investigate whether S. parasanguis FW213 can form biofilms in vitro and to determine some of the environmental parameters that affect biofilm formation. It was determined that S. parasanguis FW213 can form biofilms on solid surfaces, such as polystyrene and polyvinyl-chloride, under a range of growth conditions. Glucose promoted biofilm formation in FW213, suggesting that glucose is an environmental signal regulating FW213 biofilm development. This result supports the idea that bacteria form biofilms under favorable nutrient conditions. It does not rule out, however, the possibility that other factors associated with an increase in glucose concentration, such as pH and/or osmolarity, play a role in biofilm formation.