In this article we describe the use of the “biofilm potential” method to assess the ecological status of periodontal sulci with respect to the health and spreading tendencies of the biofilm communities growing in them. The data suggest that the biofilm potential is an accurate indicator of the microbiological health of the sulcus, and further suggest that the efficient delivery of antibacterial oxidants via the PerioProtect system*, which uses an oxidative chemical strategy rather than physical removal of the biofilms by scaling and root planning (SRP), is an effective treatment for periodontitis.
By C. Schaudinn, A. Gorur, Dr P.P. Sedghizadeh, Dr JW Costerton and Dr D. Keller

A biofilm disease
As microbial ecologists struggled with the dawning realisation that only a very small proportion (approximately 1 %) of the bacteria present in natural ecosystems actually gives rise to colonies when plated on agar media [1], they gradually abandoned culture methods in favour of direct observations [2]. In even the earliest of these publications [3, 4] it was noted that the predominant microbial population of all the aquatic ecosystems studied was attached to surfaces, and that the best way to study these populations was to insert a clean slide and observe the communities that developed on its surface. Microbial ecologists adopted the biofilm potential as a measure of the health of microbial communities in the ecosystem in question [2]. This approach is based on the colonisation of clean surfaces introduced into the ecosystem, and on the fact that healthy biofilms shed planktonic cells that colonise these surfaces in a finite length of time, and reproduce the sessile communities that are dominant in the ecosystem. If the ecosystem is healthy, and there are sufficient nutrients, freshly introduced surfaces are fully colonised by mature communities in 24 – 48 hours [5]. If the ecosystem is compromised by the lack of nutrients [6], or by the activity of biocides that kill planktonic cells and the most vulnerable cells of the biofilm communities [7], then the colonisation of the freshly introduced surfaces is retarded and incomplete [8]. Wecke et al inserted both plastic and gold “carriers” into periodontal pockets, to obtain biofilms that replicate the sessile bacterial populations on the tooth and gum surfaces [9], and demonstrated that these ex vivo communities closely resemble those seen on extracted teeth [10]. In this study we extended the use of this technique to monitor the effects of anti-biofilm therapy on the biofilm populations that colonise surfaces within the infected sulcus.

In their comprehensive review of chronic bacterial infections, Costerton et al presented evidence  that the bacteria that cause these infections grow in matrix-enclosed biofilms, within which they are protected from host defenses and antibiotics [11,12]. Direct microscopic evidence that periodontitis is caused by biofilm bacteria is presented in this paper, in Costerton’s book [13], and in other publications [14-17]. This places this chronic bacterial disease squarely in the category of biofilm diseases that currently constitute 65 – 80% of infections treated by health professionals in the developed world [11]. This perception offers a plausible explanation for the fact that a bacterial infection affecting 85% of adults in the USA is inherently resistant to intact host defenses and to antibiotics, even though it involves tissues that are open to physical intervention and to systemic vascular access [18].

Therapeutic strategies
If we take a global view of all biofilm infections, from simple gingivitis to device-related infections of the bone surrounding complex orthopaedic reconstructions [19], two therapeutic strategies have emerged that promise relief to desperate patients.The first of these strategies involves the physical removal of the bacterial biofilm from the colonised biomaterial and/or from the infected tissues, and the prevention of recolonisation, by the use of antibiotics to kill residual planktonic cells of the infecting species. This approach always gives a measure of relief in orthopaedic infections, but its success depends on the complete removal of the biofilm [19] and the selection of antibiotics so that all planktonic cells are killed, and in practice it gives complete resolution of the infection in only approximately 50% of device-related infections [20]. The scaling and root planning (SRP) treatment for periodontitis is very similar, in that residual biofilm left in crevices and on occluded surfaces will re-grow and spread to cover mechanically cleaned areas, and planktonic
cells that escape post-SRP treatments will colonise newly available surfaces [21].

The second strategy involves the use of non-specific chemical agents to kill all of the planktonic cells, and some of the biofilm cells, in a particular ecosystem. This strategy, which has proven to be very successful in industrial applications [22], and in the protection of various catheters from bacterial colonisation and consequent infection [23-25], depends on the alteration of the microbial ecology of an ecosystem so that biofilm formation is minimised. Biofilm bacteria show the same susceptibility to non-specific oxidising agents as their planktonic counterparts [26], so that industrial biocides [27] and “catheter lock” solutions [23-25,28] kill all of the planktonic cells and as many of the biofilm cells as their stochiometry allows. The regular application of 0.5 % bleach in the “Y” sets used to protect Tenchoff catheters from colonisation and infection has been successful for several years [25], in spite of the extreme susceptibility of the peritoneum to bacterial incursions.
The PerioProtect therapeutic system uses this concept, because it delivers peroxide and an antioxidant to the periodontal space at regular intervals, and kills the planktonic bacteria and enough of the biofilm bacteria to gradually alter the microbial ecology of this ecosystem. The direct result of this alteration in the microbial ecology of the periodontal space is a sharp reduction in the rate at which available surfaces are colonised; the relevant measurement is the rate of colonisation of an inert carrier material introduced into the area for a specified length of time. This measurement is clinically relevant because it is the inflammatory response of the gingival tissues to the presence of planktonic and detached biofilm cells that lies at the base of the aetiology of this, and all other, chronic biofilm infections [11].

Study design
Sterile 13 mm Thermanox plastic coverslips were shaped with sterile razor blades so that they would fit into individual periodontal pockets. After this shaping, which was conducted in a laminar flow chamber, the inserts were dipped in 70% ethanol and stored in sterile 24 well plates until use. Thermanox (http://www.nuncbrand.com/page/en/303.aspx) is commonly used as an attachment surface for the cultivation of mammalian cell lines. Based on the patient’s impressions, perio trays were made, in accordance with FDA regulations for a laboratory registered with the FDA, to coincide with the specific disease conditions of the patient. The tray was worn in accordance with the scope and magnitude of disease, and wearing instructions were modified as healing occured. The patient shown in Figure 1 was instructed to wear the trays for 20 minutes, four times a day. Prior to the tray delivery, small, sterile polycarbonate carriers (Thermanox) were inserted in three periodontal pockets: 26 mesial buccal, 26 distal buccal and 36 mesial buccal and supragingivally attached to the tooth surface with PeriAcryl (GlueStitch Corporation, Canada) for 48 hours. New sets of carriers were placed for 48 hours at the same sites after 2, 7, 12 and 17 days, respectively [Figure 2]. During the time the carriers were inserted the patient did not use the Perio-Tray but was allowed to brush and floss except at the indicated sites.
When the carriers were removed the tooth side of each was identified with a score mark. The carriers were removed and fixed in 2.5% (v/v) glutaraldehyde in 0.1 M cacodylate buffer (pH 7) at 4° C for 24 hours, washed in PBS (pH 7.0) buffer, dehydrated in a graded ethanol line critical point dried (EMS 850), mounted on a stub, sputter coated with 20 nm platinum and examined with a scanning electron microscope. For bacterial enumeration, the morphotypes were counted on an area of 10x10 μm2 of each colony, and the counts were multiplied by the area and the height of the micro-colony. The counts of a certain morphotype, per carrier, were expressed as the sum of all micro-colonies on both sides of the carrier. The timetable for the periochart examinations, the placement and removal of the carriers and the application of the PerioProtect therapy is illustrated in Figure 2.
In the treatment period between days 0 and 2, 1.5 % peroxide was administered, using the PerioProtect trays, four times per day. In the treatment periods between days 4 - 7, days 9 -12, and days 14 – 17, 1.5 % peroxide was administered with a subclinical dose of Sumycin syrup, twice per day. At day 14 new PerioProtect trays were fitted, because the patient’s gums were less swollen than at the beginning of the course of treatment.

Results before and after treatment
When the polyolefin carriers were removed from three sulci [Figure 1] in the patient’s mouth, prior to any treatment [Day 0 in Figure 2], an SEM showed that both the tooth and gum surfaces were colonised by luxuriant biofilms composed of bacterial cells of many different morphotypes [Figure 3].

When the biofilms on the carriers from sites 26 mb, 26 db, and 36 mb were analysed, in terms of the number of cells of six distinct morphotypes present on the 6 mm2 surface area [Figure 4], it was obvious that the colonisation was very luxuriant and that the three sulci differed very radically regarding the communities that had developed. The biofilm on the carrier from site 26 mb was composed predominantly of cocci, with smaller numbers of short rods and fusiform cells, while the biofilm on the carrier from site 36 mb showed the same morphotypes in different proportions. The biofilm on the carrier from site 26 db contained spiral treponemas, and cells with the unique Selenomonas morphotype of curved cells with a tuft of flagella, as well as cells of the three morphotypes seen in the other sulci. We noted that the microbial biofilms in sulci adjacent to the same tooth can vary, very substantially, in the bacterial morphotypes that constitute these sessile communities.

When treatment with peroxide was initiated, using the PerioProtect delivery system, a 99% decrease in the number of colonising bacteria was seen at site 26 mb after two days of treatment. By day 7 [Figures 2 and 4] the biofilm potential at all three sites had decreased to between 0.2 % and 2.0 % of that of the untreated sulci, and the morphotypes present in these much less luxuriant biofilms were reduced to short rods and cocci. A further reduction in cell numbers was seen after 12 days of treatment. After 17 days of treatment the carrier from site 26 mb showed no bacterial cells, and only 0.02 % of the cells seen before treatment at site 36 mb were counted. An SEM of the gum side of the carrier from site 26 mb showed the complete absence of bacteria and the presence of human epithelial cells and rhomboid sumycin crystals, which indicated that the PerioProtect system had delivered this antioxidant deep into the infected sulcus [Figure 5]. An SEM of the tooth side of the carrier from site 36 mb showed that very large areas of the carrier remained uncolonised, and that the small micro-colonies that were formed consisted of coccoid cells and of a branched and polymorphic Actinomyces morphotype that had not previously been seen at this or at any of the other sites [Figure 6].

The periocharts that had been prepared prior to treatment and at day 12, by clinicians blinded to the experimental design, showed reductions from 6mm to 5mm at site 26 mb, from 6mm to 4mm at site 26 db, and from 5mm to 3 mm at site 36 mb [Figure 7]. It is also important to note that the PerioProtect trays used from day 1 until day 12 were replaced by newly fitted trays (at day 13) because the reduction in swelling due to the treatment had caused the original trays to fit poorly.

Discussion
If the biofilms in a system are robust, planktonic cells will be released from all of the communities that make up the sessile population, and the biofilms that form on the new surface will represent those that predominate in the ecosystem. If however the sessile communities in the system are compromised, by lack of nutrients or by the action of antibacterial agents, the biofilms will be stressed and very few planktonic cells will be available to carry out the colonisation of the fresh surface. The usefulness of the Biofilm Potential is indicated by the observation [Figure 3] that the biofilm that formed on the carrier in the sulcus at site 26 mb covers almost all of both surfaces of this device, and consists of a very thick community in which cells of three distinct morphotypes are discernable [Figure 4]. Each morphotype may be characteristic of many different species of bacteria, and biofilms composed of the same morphotypes cannot be assumed to be composed of the same species, but the presence of different morphotypes in a biofilm is unequivocal proof that the communities are composed of different species. Manual counting of cells of each distinct morphotype [Figure 4] allows us to estimate the number of bacteria in the communities that have formed in 48 hours, and shows that different locations, even on the same tooth (tooth 26) may have very different sessile communities. The presence of very large numbers of cells of the spiral Treponema morphotype in the db position on tooth 26, and the absence of these inherently mobile 29 cells in the mb position of the same tooth, indicate that each site develops its own distinct biofilm population. Because different sites in the sulcus surrounding an individual tooth vary profoundly in their microbial population, and develop different adherent communities, longitudinal studies of treatment efficacy must always examine the same sites.

If we consider the control of the mixed species biofilms that cause periodontitis, in the context of biofilm control in other medical conditions, the physical removal of these adherent communities by scaling and root planning (SRP) is consistent with other standards of practice. The transitory nature of the benefits that accrue from SRP are probably explained by the fact that this physical removal can never be complete, because of the local geometry of cracks and grooves in the tooth surface, and biofilm removal must be complete if re-growth is to be avoided [19]. In general terms, the persistence of periodontitis can be attributed to inadequate supragingival and subgingival biofilm control over time [30].

The alternative method for biofilm control is sustained treatment of these attached communities, using agents that kill planktonic cells and the most vulnerable of biofilm cells. In this sustained attack on biofilms, oxidative agents are most successful because they both kill and remove bacterial cells [31], and hydrolyse and remove matrix material [32]. Antibiotics and biocides like the quaternary ammonium compounds are much less successful, because they kill planktonic cells and some sessile bacteria, but the dead cells and pervasive matrix material remain on the surface [33], and provide an optimal conditioning layer for re-colonisation. In this study of three sites in the mouth of a single individual, the intensive use of peroxide in the first seven days of the trial reduced the biofilm potential, by 2 – 3 logs, the mainstay and reduced the number of morphotypes in the site (26 db) with the most diversity from five to three [Figure 4].

The general decrease in the biofilm potential can be attributed to the killing of planktonic cells, and of the most vulnerable sessile organisms. The elimination of the Treponema morphotype at site 26 db may have resulted from the fact that these very mobile spiral cells are not fully integrated into or protected by the biofilm community. Our detailed studies of Treponema in the bovine rumen showed that they move in and out of classic biofilms, harvesting metabolites as they move, but never remain in the structured sessile community or benefit from its collective protection [34]. The further 4 – 7 log reduction in the biofilm potential seen after a cumulative 17 days of treatment using peroxide in the PerioProtect regimen, is consistent with the proposed mechanism of this treatment, in that there are very few bacteria present in the sulcus that are capable of colonising the surface of the carriers. At site 36 mb the few bacteria that were able to form small microcolonies on the tooth side of the carrier [Figure 6] were either coccoid, or of the Actinomyces morphotype, whose presence in thicker biofilms at the same site may have been masked by the larger number of cells of other morphotypes at earlier stages of this treatment. The presence of human epithelial cells on the gum surface of the carrier recovered from site 26 mb at day 17 indicates that the tissue that lines the sulcus may proliferate, and shed its own colonising cells, when the bacteria in the ecosystem have been essentially controlled by sustained peroxide delivery. The presence of rhomboid sumycin crystals on all of the surface areas of this carrier offer direct proof that the PerioProtect  system delivers its reagents to all areas of the sulcus.

Conclusions
We conclude that the PerioProtect treatment system delivers biofilm control agents to the infected sulcus very effectively. In this single case, in which peroxide was delivered using the PerioProtect trays, this oxidising agent was seen to reduce the biofilm potential very substantially, in the three sulci that were monitored using the flexible plastic carriers. The tetracycline crystals viewed on the SEM analysis [Figure 5] demonstrate the capability of delivering medications to the gingival sulcus. Future studies, using this same biofilm potential measurement, will be used to “fine tune” the solutions chosen for delivery by this simple but novel technology to provide a non-surgical alternative that can be used in combination with scaling and root planning (SRP).

References
1. Colwell RR et al. Biotechnol 1985;3:817
2. Costerton JW et al. Ann Rev Microbiol 1987;41:435-64.
3. Henrici AT. J Bacteriol 1933;25:277 – 86.
4. Zobell CE, Allen EC. J Bacteriol 1935;29:239 – 51.
5. Geesey GG . Limnol and Oceanogr 1978;23:1214-23.
6. Kjelleberg S. Starvation in Bacteria. New York: Plenum Press 1993.
7. Cusack F, Lappin-Scott HM Costerton JW. J. Indust. Microbiol. 1988;2:329-35.
8. Costerton JW et al. Ann Rev Microbiol 1995;49:711-45.
9. Wecke J et al. FEMS Microbiology Letters 2000;191:95 – 101.
10. Vrahopoulos TP, Barber PM, Newman HN. J Periodontol 1995;66:113-24.
11. Costerton JW, Stewart PS, Greenberg EP. Science 1999;284:1318-1322.
12. Nickel JC et al. Antimicrobial Agents & Chemotherapy, 1985;27:619-24.
13. Costerton JW. The Biofilm Primer. Hiedelberg: Springer 2007:1 – 200.
14. Kuramitsu HK et al. Microbiol Mol Biol Rev 2007;71:653 - 70
15. Lamont RJ, Jenkinson HF. Adhesins as an ecological determinant in the oral cavity In Kuramitsu HK Ellen RP(eds.), Oral bacterial ecology: the molecular basis. Horizon Science Press, Wymondham, United Kingdom 2000:131-168.
16. Offenbacher S. Ann Periodont 1996;1:821 – 78.
17. Palmer RJ et al. Infect Immun 2001;69:5794 – 804.
18. Peterson PE. Dent Oral Epidemiol 2003;31:3 – 24.
19. Costerton JW. Clin Orthop Relat Res 2005;437:7 - 11.
20. Khoury AE et al. ASAIO Transactions 1992;38:M174-M178.
21. Shiloah J et al. Periodontol 1998;69:1364-72.
22. Keevil CW, Mackerness CW, Colbourne JS. Internat Biodeterioration 1990;26:169 – 79.
23. Dasgupta MK et al. Peritoneal Dialysis Bull 1987;7:168-173.
24. Dasgupta MK et al. Am J Nephrol 1990;10:353-58.
25. Dasgupta MK, Larabie M. Peritoneal Dialysis International. 2001;21:S213 – S217.
26. Poole K. J Applied Microbiol 2002;92:55S – 64S.
27. Stickler DJ. J Applied Microbiol 2002;92:163S –170S.
28. Anwar H, Dasgupta MK, Costerton JW. Antimicrob Agents and Chemother 1990;34:2043 – 46.
29. Lux R et al. Infect Immun 2001;69:6276 – 283.
30. Driscko CH. Periodontol 2001;25:77 – 88.
31. Glaze WH, Kang JW, Chapin DH. Ozone Science and Engineering 1987;9:335 – 52.
32. Denyer SP. Internat Biodeterioration and Degradation. 1995;36:227 – 45.
33. Sung H-W, Hsu H-L, Hsu C-S. J Biomed Materials Res 1998;37:376 – 83.
34. Kudo H, Cheng K-J, Costerton JW. Can J Microbiol 1987;33:244-48.

The authors
Christoph Schaudinn, MSc1,, Amita Gorur MSc1,
Parish P. Sedghizadeh DDS, MS1, J. William Costerton PhD, FRCS1 and Duane Keller DDS2*

1Center for Biofilms
School of Dentistry
University of Southern California
925 West 34th Street
Los Angeles, CA, USA

2 St Louis Health and Wellness
Keller Professional Group
3955 Bayless Avenue
St Louis, MO, USA

*Corresponding author
drdkeller@sbcglobal.net
Tel +1 314 638 4190