doi:10.1111/iej.12474
Collagen-like peptide sequences inhibit bacterial invasion of root dentine
J. L. Brittan1, S. V. Sprague1, S. P. Huntley1, C. N. A. Bell1, H. F. Jenkinson1 & R. M. Love2 1 School of Oral and Dental Sciences, University of Bristol, Bristol, UK; and 2Department of Oral Rehabilitation, University of Otago, Dunedin, New Zealand
Abstract Brittan JL, Sprague SV, Huntley SP, Bell CNA, Jenkinson HF, Love RM. Collagen-like peptide sequences inhibit
bacterial
invasion
of
root
dentine.
International
Endodontic Journal, 49, 462–470, 2016.
Aim To investigate the effects of peptides derived from the sequence of collagen to inhibit penetration of human or bovine dentine by species of streptococci and enterococci. Methodology Blocks of human or bovine root dentine were infected for 14 days with bacterial cultures, in the presence or absence of various collagen-like peptide sequences. Invasion of dentinal tubules was determined from microscopic images of histochemically stained dentine thin sections. Extent of invasion was expressed as tubule invasion index (TI), or tubule invasion factor (TIF) which, in addition to the density of invasion, took into account the depth of invasion. Data were analysed by twoway ANOVA.
Introduction A wide variety of oral bacteria is associated with infections of the pulp, and root canal systems (Sundqvist 1994, Siqueira & R^ oßcas 2009, Sousa et al. 2013), leading to pulpitis and inflammatory periapical lesions. These infections are believed to arise principally following bacterial penetration of dentinal tubules (Olgart et al. 1974, Love & Jenkinson 2002). This occurs when dentine is exposed in the oral
Correspondence: Robert M. Love, Department of Oral Rehabilitation, University of Otago School of Dentistry, PO Box 647, Dunedin, New Zealand (e-mail: [email protected]).
462
International Endodontic Journal, 49, 462–470, 2016
Results Streptococcus gordonii, Streptococcus mutans and Enterococcus faecalis were associated with heavy invasion (TI >2.5, TIF >4) of human or bovine root dentinal tubules, with E. faecalis being the most penetrative. Incorporation of peptides Gly–Pro–Ala or Gly– Pro–Hyp into the in vitro model system significantly reduced (P < 0.05) dentine invasion by the three species of highly invasive organisms. Inhibition of bacterial invasion by the peptides was dose dependent, and the peptides did not inhibit bacterial growth in culture. Conclusion Specific collagen-like peptide sequences inhibited the invasion of dentine in vitro by a range of oral bacteria. The peptides likely act as competitive inhibitors blocking bacterial collagen receptors and could potentially allow for target-specific control of dentine infections. Keywords: bovine dentine, dentinal Enterococcus, proline peptides, Streptococcus.
tubules,
Received 23 November 2014; accepted 24 May 2015
cavity allowing bacteria to gain access to dentinal tubules typically via dental caries, cracks or fractures, or microleakage around restorations. The bacteria that initially invade dentinal tubules are primarily of the genera Enterococcus or Streptococcus (Love & Jenkinson 2002). Enterococci readily penetrate dentinal tubules (Love 2001, Chivatxaranukul et al. 2008) and are root canal pathogens with reduced susceptibility to killing by irrigants such as sodium hypochlorite (NaOCl) (Ørstavik & Haapasalo 1990) or calcium hydroxide (Hancock et al. 2001). Enterococcus faecalis produces a cell surface protein termed Ace, which is a collagen-binding adhesin mediating adherence of enterococci to dentine (Kowalski et al. 2006).
© 2015 International Endodontic Journal. Published by John Wiley & Sons Ltd
Brittan et al. Bacterial invasion of dentine
Ace also confers reduced sensitivity to disinfectants (Kayaoglu et al. 2008). Oral streptococci, such as ubiquitous Streptococcus gordonii, invade dentinal tubules by binding to intratubular collagen type I via bacterial cell surface proteins designated antigen I/II (AgI/II) family proteins (Jenkinson & Demuth 1997, Love et al. 1997). Streptococci also produce collagen-binding adhesins similar to Ace, such as CbdA (Moses et al. 2013), which may enhance survival of S. gordonii in instrumented root canals. The AgI/II proteins are able to bind other oral bacteria (Demuth et al. 1996) and may be responsible not only for promoting dentine infection, but also for establishing co-infections leading to mixed microbial communities (Love et al. 2000, Ozok et al. 2012). Thus, inhibitors of collagen-binding functions may in future be important for the control of infections caused by enterococci or streptococci, and for polymicrobial infections that are driven by primary streptococcal infections (Jenkinson 2011). In addition, inhibitors of AgI/II interactions with collagen type I could be useful in preventing dentine invasion. Inhibitors of oral mutans streptococcal colonization that have been successfully tested include antibodies that block AgI/II adhesin functions (Lehner et al. 1985) and adhesin-derived peptides that block bacterial colonization sites in the host (Kelly et al. 1999). Neither of these strategies has been thought of as applicable to preventing dentine infections because they are designed to target the microorganisms rather than the specific host site of infection. There are no products commercially available that work selectively at a molecular level to inhibit or prevent bacterial invasion of dentinal tubules. If it is necessary to seal dentine to prevent bacterial invasion, then materials such as glass–ionomer cement, resin, or a desensitizing mineral precipitant are used. Disinfection of radicular dentine is mainly performed using nonspecific antimicrobial agents, NaOCl and calcium hydroxide. However, NaOCl toxicity can cause bone and soft tissue necrosis (Lundy & Stanley 1969), and Ca(OH)2 is not so effective against robust microorganisms such as E. faecalis (Ørstavik & Haapasalo 1990). Patients with severe oral bacterial infections are almost always prescribed antibiotics. However, recent surveys suggest that there is increasing incidence of oral bacterial isolates that are resistant to the most frequently used antibiotics in dentistry (Rams et al. 2014) and that this may be responsible at least in part for higher treatment failure rates. In addition, the use of broad-range antibiotics unbalances the natural microbiota that is essential for the well-being of
© 2015 International Endodontic Journal. Published by John Wiley & Sons Ltd
the host. Side effects of antibiotic treatment include inflammatory reactions and overgrowth of opportunistic pathogens such as Candida fungi. There is much interest therefore in developing alternative approaches to inhibit or prevent oral microbial infections, and inhibitors that target bacterial dentine invasion mechanisms would be highly desirable in managing endodontic infections. Love et al. (1997) demonstrated that acid-soluble collagen or a digest of collagen type I inhibited penetration of dentinal tubules by S. gordonii DL1. They hypothesized that collagen fragments competitively inhibited the bacterial adhesins that interacted with intratubular collagen that normally enabled the bacteria to penetrate and grow inside the dentinal tubules. This raised the possibility that similar peptides, or mimetics, might be useful clinically in controlling or preventing root canal or pulpal infections. The purpose of this study was to investigate the effects of small defined peptides, based upon those found within collagen, on the invasion of dentine by streptococci and enterococci. The null hypothesis was that the small peptides would not influence invasion.
Materials and methods Bacteria strains and growth media Bacterial strains utilized were Streptococcus gordonii DL1 (Challis), Streptococcus mutans NG8 and Enterococcus faecalis JH2-2. Bacteria were grown at 37 °C in a candle jar on brain–heart infusion agar, or in brain– heart infusion broth (37 g L 1; Difco Laboratories, Detroit, MI, USA) containing yeast extract (5 gL 1; Difco) (BHY) in screw-cap tubes or bottles at 37 °C. Cells were harvested by centrifugation (5000 g, 10 min) and suspended in one-tenth volume BHY medium containing 15% (v/v) glycerol at a density of ~1010 colony-forming units (CFU) mL 1. Aliquots were stored at 80 °C until required. Experimental cultures were produced by inoculation from stock suspension (50 lL) into BHY medium (20 mL) and were grown at 37 °C in closed tubes or bottles without shaking. Peptides Gly–Hyp, Gly–Pro, Gly–Pro–Ala, Gly–Pro– Hyp, Gly–Pro–Gly–Gly and Gly–Pro–Arg–Pro were obtained from Sigma-Aldrich (St. Louis, MO, USA).
Bacterial growth yields Bacterial cultures were prepared by inoculating stock suspension as above, and various concentrations of
International Endodontic Journal, 49, 462–470, 2016
463
Bacterial invasion of dentine Brittan et al.
the peptides (20 lL) were added to 180 lL portions of inoculated culture medium in microtitre plate wells. Growth yields of bacteria with or without peptides were determined by measuring optical density (at 600 nm) in a microplate reader (Synergy 2, BioTek, Winooski, VT, USA) of cultures after 16-h incubation at 37 °C.
Preparation of root dentine Noncarious, unrestored human canine or premolar teeth with single root canals were obtained with consent from patients who had undergone orthodontic extractions at the Bristol Dental Hospital. Bovine incisors were extracted from young (2.5 indicated heavy invasion, and TI 50 lm; and x3, where at least 5 tubules per section were invaded to a depth of 100 lm or greater. TI and TIF scores were normally distributed, and data were analysed by two-way ANOVA (GraphPad Prism 6.0, GraphPad Software Inc., La Jolla, CA, USA) with a significance level of 0.05.
Results Comparison of human and bovine dentine for invasion studies: light microscopy S. gordonii DL1 was associated with heavy invasion of both human and bovine dentine after 14 days as was S. mutans NG8 and E. faecalis JH2-2 (Fig. 1). The three bacterial species penetrated bovine dentine to a greater depth than in human dentine (data not shown). S. mutans cells formed a thin biofilm on the pulpal surface of both human and bovine dentine (Fig. 1). A graphical representation of the comparison of tubule invasion index (TI) and tubule invasion factor (TIF) for the invasion of bovine tubules by the three organisms is shown in Fig. 2. E. faecalis was clearly the most penetrative species into bovine dentine. This was exemplified by the higher TIF value of ~8.0 compared with the streptococci (Fig. 2). The TI
© 2015 International Endodontic Journal. Published by John Wiley & Sons Ltd
Brittan et al. Bacterial invasion of dentine
Figure 1 Bacterial penetration of human or bovine dentine showing heavy invasion by all species with S. mutans forming thin biofilms on the surface of both dentine types. Bars, 50 lm.
values for the three bacterial strains were more similar as they showed similar frequencies of tubule invasion, with depth of invasion not taken into account.
Effect of collagen peptides on tubule invasion Figure 3 shows the effects of different peptides glycine–proline–alanine (GPA), glycine–proline (GP), glycine–proline–arginine–proline (GPRP), glycine–proline–glycine–glycine (GPGG) and glycine–hydroxyproline (GO) (100 lg mL 1) on the invasion (TIF) of bovine dentinal tubules by S. gordonii DL1. All of the tripeptides and tetrapeptides tested were effective in reducing the invasion levels compared with the control (no peptide). Glycine–proline–alanine (GPA) was consistently the most reproducibly effective. Glycine– proline–glycine–glycine was also very effective. The dipeptide GP was not significantly inhibitory, whereas GO was inhibitory to invasion of bovine dentinal tubules by S. gordonii DL1.
Inhibition of human or bovine dentinal tubule invasion by collagen tripeptides
Figure 2 Comparison of TI (filled columns) and TIF (hatched columns) of bovine dentine. Error bars are SD, *P < 0.05; NS, not significant.
© 2015 International Endodontic Journal. Published by John Wiley & Sons Ltd
Incorporation of the tripeptide GPA (200 lg mL 1) into the invasion model medium resulted in inhibition of invasion of both human and bovine dentine by S. gordonii DL1, S. mutans NG8 and E. faecalis JH2-2 as visualized histologically (Fig. 4). The tripeptide
International Endodontic Journal, 49, 462–470, 2016
465
Bacterial invasion of dentine Brittan et al.
that the inhibitory effect of the tripeptide on S. gordonii invasion of bovine dentinal tubules was dose dependent. The TIF decreased from 5.11 0.87 with no GPA present to 1.66 0.43 in the presence of 0.2 mg GPA mL 1 (Fig. 6a). There was 43.9% reduction in invasion (TIF) at the lowest concentration of peptide tested (50 lg mL 1), and TIF values decreased with increasing peptide concentration up to 80% reduction in the maximum concentration tested. Comparison of the dose-dependent inhibition of dentinal tubule invasion by GPA with corresponding effects of GPA on the growth of S. gordonii DL1 showed a nearly 50% inhibition of dentinal tubule invasion at 50 lg mL 1 GPA (Fig. 6b). At the same concentration, there was little or no effect of the peptide on microbial growth rate or yield. At higher concentration (0.2 mg mL 1), there was a small, but not significant, reduction in growth of bacteria in suspension culture, whilst there was ~80% inhibition of dentinal tubule invasion. The GPA tripeptide did not therefore exert the inhibitory effects on dentinal tubule invasion by directly inhibiting bacterial growth.
Figure 3 Effects of different collagen-like peptides on the invasion (TIF) of bovine dentinal tubules by S. gordonii DL1. Error bars are SD, *P < 0.05, relative to control (DL1 column, untreated).
Glycine-Proline-Hydroxyproline (GPO) had similar inhibitory effects, which suggested that the nature of the third amino acid residue may not be crucial for inhibitory properties (Fig. 4). A graphical representation of the extent of inhibition by GPA or GPO is shown in Fig. 5. By TI measurements, the three microorganisms invaded human dentine to similar levels (Fig. 5a), and tripeptides GPA and GPO had similar statistically significant (P < 0.05) inhibitory effects. By TIF measurements, E. faecalis appeared to be more invasive of bovine dentinal tubules than S. gordonii or S. mutans, but this was not statistically significant (P = 0.092) (Fig. 5b). Tripeptide GPA significantly (P < 0.05) inhibited tubule invasion of all bacterial species tested but was more effective at blocking S. gordonii invasion of bovine dentine than S. mutans or E. faecalis (Fig. 5b).
GPA inhibited invasion but not bacterial growth The significant results with GPA tripeptide and S. gordonii DL1 were further analysed and showed
466
International Endodontic Journal, 49, 462–470, 2016
Discussion Three species of Gram-positive cocci that commonly infect dentine and pulp were used in the dentine invasion model, and all three species were able to invade human root dentine to levels previously shown (Love et al. 1997, Love 2001). Bovine root dentine was also incorporated into the study because this was easier to prepare and many more sections could be obtained from a single tooth root, and has been utilized for other invasion studies (Assouline et al. 2001). Bovine radicular dentine contains a lower density of dentinal tubules, and the tubules have larger diameter to those in human dentine (Lopes et al. 2009). The two types of dentine showed similar patterns of invasion with the three microorganisms tested. However, there was deeper invasion of bovine dentinal tubules by all three species of bacteria which was consistent with other studies and may be related to the larger diameter of tubules in bovine dentine (Lopes et al. 2009), which would facilitate better diffusion of nutrients (Sigusch et al. 2014). These observations confirmed that bovine radicular dentine was an appropriate substrate to investigate bacterial invasion of dentine. Previous studies have shown that S. gordonii bacterial cells interacted strongly with bundles of collagen type I and formed chains along the lengths of the
© 2015 International Endodontic Journal. Published by John Wiley & Sons Ltd
Brittan et al. Bacterial invasion of dentine
Figure 4 Inhibition of bacterial invasion of dentine by the presence of GPA or GPO tripeptides. Bars, 50 lm.
(a)
(b)
Figure 5 Comparison of bacterial invasion of dentine in the presence of GPA or GPO tripeptides. (a), TI values for human dentine; controls (open column), GPA (grey shaded), GPO (black filled). (b), TIF values for bovine dentine; controls (black columns, GPA (grey hatched). Error bars are SD, *P < 0.05, relative to controls (untreated).
© 2015 International Endodontic Journal. Published by John Wiley & Sons Ltd
International Endodontic Journal, 49, 462–470, 2016
467
Bacterial invasion of dentine Brittan et al.
(a)
(b)
Figure 6 Dose-dependent inhibitory effect of GPA tripeptide on invasion of S. gordonii DL1. (a) TI of bovine dentine. (b) growth yields (filled column) and invasion (hatched column) expressed as % untreated wild type. Error bars are SD, *P < 0.05, relative to controls (untreated).
fibres (Love et al. 1997, Heddle et al. 2003). This is the means by which it has been proposed that streptococci are able to grow along the lengths of dentinal tubules (e.g. as seen in Fig. 1) by adhering to intratubular collagen (Love & Jenkinson 2002). It has also been shown that collagen-derived peptides were sensed by the bacterial cells and affected the transcription of genes encoding cell surface proteins that interacted with collagen type I (Love et al. 1997, Heddle et al. 2003). Additionally, a collagen type 1-digest inhibited penetration of dentinal tubules by S. gordonii DL1 (Love et al. 1997) which was hypothesized to have resulted from collagen fragments blocking the bacterial adhesins recognizing dentinal collagen. These observations led us to investigate whether specific synthetic peptides based upon the collagen primary amino acid sequence might be effective in modulating penetration of dentinal tubules by bacteria. To develop this, the effects of tripeptides and tetrapeptides containing the common collagen sequence motif glycine–proline-Xaa, where X is any amino acid residue that fits into the repeating structural model for collagen (Fig. S1), were tested. The results
468
International Endodontic Journal, 49, 462–470, 2016
revealed that GPX peptides were inhibitory, but that GP alone was not sufficient to inhibit invasion unless P was replaced by O. Additionally, inhibition of invasion by GPA was not by direct inhibition of bacterial growth, suggesting that competitive inhibition of bacterial collagen-binding sites was a likely mode of action. The data presented raise an interesting question as to whether a collagen-like tripeptide sequence, for example GPA, or dipeptide GO, might be useful as target-specific inhibitors of dentine infection by bacteria. These peptides are potentially nontoxic (hydroxyproline is already utilized as a topical repair agent) and would have minimal side effects to patients. The actual concentrations of peptide required clinically would have to be tested empirically. The concentrations that were tested in vitro are likely to be orders of magnitude more than necessary for the bacterial challenge in vivo. Unlike antibiotics, the peptides would also not perturb the general composition or activities of the host microbiota, except at the target site for disease prevention. These peptides are found naturally within components of human tissue proteins and are products to which the host would not normally have
© 2015 International Endodontic Journal. Published by John Wiley & Sons Ltd
Brittan et al. Bacterial invasion of dentine
any adverse responses. They are not bactericidal and thus would not induce development of bacterial resistance. It can be suggested that the peptides work synergistically by adversely modulating bacterial gene expression (Love et al. 1997) and by blocking the proper recognition by the bacteria of the host collagen. As the peptide does not generally inhibit bacterial growth, it could work at the site of potential infection, preventing the invasive spread of infection. Additionally, the peptides have the potential to interfere with mixed species biofilm formation because inhibiting streptococci, which are primary colonizers, could help prevent a more pathogenic community from developing (Love et al. 2000). If these peptides, or derivatives such as mimetics, were to be of use clinically, they would need to show stability within the oral environment. This aspect of their potential has been investigated by incubating GPA peptide in human whole saliva at 37 °C for up to 16 h. Under these conditions, there was no significant loss of peptide concentration as measured by mass spectrometry (Fig. S2). Potential applications of such peptides could be (i) inhibition of bacterial invasion around a restoration and into dentinal tubules by topical application, (ii) incorporation into intracanal materials or medicaments, (iii) inclusion within periodontal intra-oral pocket material to inhibit recolonization of the exposed root dentine after it is disinfected or root planed: this might stop the pocket dentine acting as a reservoir for recolonization of the periodontal pocket, and (iv) incorporation into medicaments to inhibit bacterial invasion of wounds at sites of endodontic, periodontal or oral surgery treatment. Application of collagen-like peptide sequences to prevent colonization or invasion would enhance tissue regeneration, especially in periodontal clinical practice of fibrous (collagen) regeneration. Their application to membranes used in guided tissue regeneration around natural teeth or implants would be useful, as tissue regeneration will not occur if the area is infected.
Conclusion The study has shown that specific tripeptides and a dipeptide, the sequences of which are found in collagens, are able to inhibit bacterial penetration of dentine. These peptides, or derivatives of them, might thus allow for target-specific control of dentine infections, with potentially minimal side effects to the subject or to the normal oral microbiota.
© 2015 International Endodontic Journal. Published by John Wiley & Sons Ltd
Acknowledgements We would like to thank The University of Bristol Veterinary School staff associated with the Langford Abattoir for their assistance with provision of bovine teeth. We thank Nicola West, Martin Addy, Nick Banfield and Peter Shellis for helpful discussions. This study was supported in part by a grant from the University of Bristol Enterprise Development Fund.
References Assouline LS, Fuss Z, Mazor Y, Weiss EI (2001) Bacterial penetration and proliferation in root canal dentinal tubules after applying dentin adhesives in vitro. Journal of Endodontics 27, 398–400. Brown JH, Brenn L (1931) A method for the differential staining of gram-positive and gram-negative bacteria in tissue sections. Bulletin of the Johns Hopkins Hospital 48, 69–73. Chivatxaranukul P, Dashper SG, Messer HH (2008) Dentinal tubule invasion and adherence by Enterococcus faecalis. International Endodontic Journal 41, 873–82. Demuth DR, Duan Y, Brooks W, Holmes AR, McNab R, Jenkinson HF (1996) Tandem genes encode cell-surface polypeptides SspA and SspB which mediate adhesion of the oral bacterium Streptococcus gordonii to human and bacterial receptors. Molecular Microbiology 20, 403–13. Hancock HH III, Sigurdsson A, Trope M, Moiseiwitsch J (2001) Bacteria isolated after unsuccessful endodontic treatment in a North American population. Oral Surgery, Oral Medicine, Oral Pathology Oral Radiology and Endodontics 91, 579–86. Heddle C, Nobbs AH, Jakubovics NS et al. (2003) Host collagen signal induces antigen I/II adhesin and invasin gene expression in oral Streptococcus gordonii. Molecular Microbiology 50, 597–607. Jenkinson HF (2011) Beyond the oral microbiome. Environmental Microbiology 13, 3077–87. Jenkinson HF, Demuth DR (1997) Structure, function and immunogenicity of streptococcal antigen I/II polypeptides. Molecular Microbiology 23, 183–90. Kayaoglu G, Erten H, Ørstavik D (2008) Possible role of the adhesin Ace and collagen adherence in conveying resistance to disinfectants on Enterococcus faecalis. Oral Microbiology and Immunology 23, 449–54. Kelly CG, Younson JS, Hikmat BY et al. (1999) A synthetic peptide adhesion epitope as a novel antimicrobial agent. Nature Biotechnology 17, 42–7. Kowalski WJ, Kasper EL, Hatton JF, Murray BE, Nallapareddy SR, Gillespie MJ (2006) Enterococcus faecalis adhesin, Ace, mediates attachment to particulate dentin. Journal of Endodontics 32, 634–7. Lehner T, Caldwell J, Smith R (1985) Local passive immunization by monoclonal antibodies against streptococcal
International Endodontic Journal, 49, 462–470, 2016
469
Bacterial invasion of dentine Brittan et al.
antigen I/II in the prevention of dental caries. Infection and Immunity 50, 796–9. Lopes MB, Sinhoreti MAC, Gonini J unior A, Consani S, McCabe JF (2009) Comparative study of tubular diameter and quantity for human and bovine dentin at different depths. Brazilian Dental Journal 20, 279–83. Love RM (2001) Enterococcus faecalis - mechanism for its role in endodontic failure. International Endodontic Journal 34, 399–405. Love RM, Jenkinson HF (2002) Invasion of dentinal tubules by oral bacteria. Critical Reviews in Oral Biology and Medicine 13, 171–83. Love RM, Chandler NP, Jenkinson HF (1996) Penetration of smeared or nonsmeared dentine by Streptococcus gordonii. International Endodontic Journal 29, 2–12. Love RM, McMillan MD, Jenkinson HF (1997) Invasion of dentinal tubules by oral streptococci is associated with collagen recognition mediated by the antigen I/II family of polypeptides. Infection and Immunity 65, 5157–64. Love RM, McMillan MD, Park Y, Jenkinson HF (2000) Coinvasion of dentinal tubules by Porphyromonas gingivalis and Streptococcus gordonii depends upon binding specificity of streptococcal antigen I/II adhesin. Infection and Immunity 68, 1359–65. Lundy T, Stanley HR (1969) Correlation of pulpal histopathology and clinical symptoms in human teeth subjected to experimental irritation. Oral Surgery, Oral Medicine, and Oral Pathology 27, 187–201. Moses PJ, Power DA, Jesionowski AM, Jenkinson HF, Pantera EA Jr, Vickerman MM (2013) Streptococcus gordonii collagen-binding domain protein CbdA may enhance bacterial survival in instrumented root canals ex vivo. Journal of Endodontics 39, 39–43. Olgart L, Br€annstr€ om M, Johnson G (1974) Invasion of bacteria into dentinal tubules. Experiments in vivo and in vitro. Acta Odontologica Scandinavica 32, 61–70.
470
International Endodontic Journal, 49, 462–470, 2016
Ørstavik D, Haapasalo M (1990) Disinfection by endodontic irrigants and dressings of experimentally infected dentinal tubules. Endodontics & Dental Traumatology 6, 142–9. Ozok AR, Persoon IF, Huse SM et al. (2012) Ecology of the microbiome of the infected root canal system: a comparison between apical and coronal root segments. International Endodontic Journal 45, 530–41. Rams TE, Degener JE, van Winkelhoff AJ (2014) Antibiotic resistance in human chronic periodontitis microbiota. Journal of Periodontology 85, 160–9. Sigusch BW, Kranz S, Klein S et al. (2014) Colonization of Enterococcus faecalis in a new SiO/SiO2-microtube in vitro model system as a function of tubule diameter. Dental Materials 30, 661–8. Siqueira JF Jr, R^ ocßas IN (2009) Diversity of endodontic microbiota revisited. Journal of Dental Research 88, 969– 81. Sousa EL, Gomes BP, Jacinto RC, Zaia AA, Ferraz CC (2013) Microbiological profile and antimicrobial susceptibility pattern of infected root canals associated with periapical abscesses. European Journal of Clinical Microbiology and Infectious Disease 32, 573–80. Sundqvist G (1994) Taxonomy, ecology, and pathogenicity of the root canal flora. Oral Surgery Oral Medicine and Oral Pathology 78, 522–30.
Supporting Information Additional Supporting Information may be found in the online version of this article: Figure S1. Ball and stick model of the internal Gly-Pro-Xaa repeats within collagen Figure S2. Stability of GPX in human whole saliva.
© 2015 International Endodontic Journal. Published by John Wiley & Sons Ltd
Collagen-like peptide sequences inhibit bacterial invasion of root dentine
J. L. Brittan1, S. V. Sprague1, S. P. Huntley1, C. N. A. Bell1, H. F. Jenkinson1 & R. M. Love2 1 School of Oral and Dental Sciences, University of Bristol, Bristol, UK; and 2Department of Oral Rehabilitation, University of Otago, Dunedin, New Zealand
Abstract Brittan JL, Sprague SV, Huntley SP, Bell CNA, Jenkinson HF, Love RM. Collagen-like peptide sequences inhibit
bacterial
invasion
of
root
dentine.
International
Endodontic Journal, 49, 462–470, 2016.
Aim To investigate the effects of peptides derived from the sequence of collagen to inhibit penetration of human or bovine dentine by species of streptococci and enterococci. Methodology Blocks of human or bovine root dentine were infected for 14 days with bacterial cultures, in the presence or absence of various collagen-like peptide sequences. Invasion of dentinal tubules was determined from microscopic images of histochemically stained dentine thin sections. Extent of invasion was expressed as tubule invasion index (TI), or tubule invasion factor (TIF) which, in addition to the density of invasion, took into account the depth of invasion. Data were analysed by twoway ANOVA.
Introduction A wide variety of oral bacteria is associated with infections of the pulp, and root canal systems (Sundqvist 1994, Siqueira & R^ oßcas 2009, Sousa et al. 2013), leading to pulpitis and inflammatory periapical lesions. These infections are believed to arise principally following bacterial penetration of dentinal tubules (Olgart et al. 1974, Love & Jenkinson 2002). This occurs when dentine is exposed in the oral
Correspondence: Robert M. Love, Department of Oral Rehabilitation, University of Otago School of Dentistry, PO Box 647, Dunedin, New Zealand (e-mail: [email protected]).
462
International Endodontic Journal, 49, 462–470, 2016
Results Streptococcus gordonii, Streptococcus mutans and Enterococcus faecalis were associated with heavy invasion (TI >2.5, TIF >4) of human or bovine root dentinal tubules, with E. faecalis being the most penetrative. Incorporation of peptides Gly–Pro–Ala or Gly– Pro–Hyp into the in vitro model system significantly reduced (P < 0.05) dentine invasion by the three species of highly invasive organisms. Inhibition of bacterial invasion by the peptides was dose dependent, and the peptides did not inhibit bacterial growth in culture. Conclusion Specific collagen-like peptide sequences inhibited the invasion of dentine in vitro by a range of oral bacteria. The peptides likely act as competitive inhibitors blocking bacterial collagen receptors and could potentially allow for target-specific control of dentine infections. Keywords: bovine dentine, dentinal Enterococcus, proline peptides, Streptococcus.
tubules,
Received 23 November 2014; accepted 24 May 2015
cavity allowing bacteria to gain access to dentinal tubules typically via dental caries, cracks or fractures, or microleakage around restorations. The bacteria that initially invade dentinal tubules are primarily of the genera Enterococcus or Streptococcus (Love & Jenkinson 2002). Enterococci readily penetrate dentinal tubules (Love 2001, Chivatxaranukul et al. 2008) and are root canal pathogens with reduced susceptibility to killing by irrigants such as sodium hypochlorite (NaOCl) (Ørstavik & Haapasalo 1990) or calcium hydroxide (Hancock et al. 2001). Enterococcus faecalis produces a cell surface protein termed Ace, which is a collagen-binding adhesin mediating adherence of enterococci to dentine (Kowalski et al. 2006).
© 2015 International Endodontic Journal. Published by John Wiley & Sons Ltd
Brittan et al. Bacterial invasion of dentine
Ace also confers reduced sensitivity to disinfectants (Kayaoglu et al. 2008). Oral streptococci, such as ubiquitous Streptococcus gordonii, invade dentinal tubules by binding to intratubular collagen type I via bacterial cell surface proteins designated antigen I/II (AgI/II) family proteins (Jenkinson & Demuth 1997, Love et al. 1997). Streptococci also produce collagen-binding adhesins similar to Ace, such as CbdA (Moses et al. 2013), which may enhance survival of S. gordonii in instrumented root canals. The AgI/II proteins are able to bind other oral bacteria (Demuth et al. 1996) and may be responsible not only for promoting dentine infection, but also for establishing co-infections leading to mixed microbial communities (Love et al. 2000, Ozok et al. 2012). Thus, inhibitors of collagen-binding functions may in future be important for the control of infections caused by enterococci or streptococci, and for polymicrobial infections that are driven by primary streptococcal infections (Jenkinson 2011). In addition, inhibitors of AgI/II interactions with collagen type I could be useful in preventing dentine invasion. Inhibitors of oral mutans streptococcal colonization that have been successfully tested include antibodies that block AgI/II adhesin functions (Lehner et al. 1985) and adhesin-derived peptides that block bacterial colonization sites in the host (Kelly et al. 1999). Neither of these strategies has been thought of as applicable to preventing dentine infections because they are designed to target the microorganisms rather than the specific host site of infection. There are no products commercially available that work selectively at a molecular level to inhibit or prevent bacterial invasion of dentinal tubules. If it is necessary to seal dentine to prevent bacterial invasion, then materials such as glass–ionomer cement, resin, or a desensitizing mineral precipitant are used. Disinfection of radicular dentine is mainly performed using nonspecific antimicrobial agents, NaOCl and calcium hydroxide. However, NaOCl toxicity can cause bone and soft tissue necrosis (Lundy & Stanley 1969), and Ca(OH)2 is not so effective against robust microorganisms such as E. faecalis (Ørstavik & Haapasalo 1990). Patients with severe oral bacterial infections are almost always prescribed antibiotics. However, recent surveys suggest that there is increasing incidence of oral bacterial isolates that are resistant to the most frequently used antibiotics in dentistry (Rams et al. 2014) and that this may be responsible at least in part for higher treatment failure rates. In addition, the use of broad-range antibiotics unbalances the natural microbiota that is essential for the well-being of
© 2015 International Endodontic Journal. Published by John Wiley & Sons Ltd
the host. Side effects of antibiotic treatment include inflammatory reactions and overgrowth of opportunistic pathogens such as Candida fungi. There is much interest therefore in developing alternative approaches to inhibit or prevent oral microbial infections, and inhibitors that target bacterial dentine invasion mechanisms would be highly desirable in managing endodontic infections. Love et al. (1997) demonstrated that acid-soluble collagen or a digest of collagen type I inhibited penetration of dentinal tubules by S. gordonii DL1. They hypothesized that collagen fragments competitively inhibited the bacterial adhesins that interacted with intratubular collagen that normally enabled the bacteria to penetrate and grow inside the dentinal tubules. This raised the possibility that similar peptides, or mimetics, might be useful clinically in controlling or preventing root canal or pulpal infections. The purpose of this study was to investigate the effects of small defined peptides, based upon those found within collagen, on the invasion of dentine by streptococci and enterococci. The null hypothesis was that the small peptides would not influence invasion.
Materials and methods Bacteria strains and growth media Bacterial strains utilized were Streptococcus gordonii DL1 (Challis), Streptococcus mutans NG8 and Enterococcus faecalis JH2-2. Bacteria were grown at 37 °C in a candle jar on brain–heart infusion agar, or in brain– heart infusion broth (37 g L 1; Difco Laboratories, Detroit, MI, USA) containing yeast extract (5 gL 1; Difco) (BHY) in screw-cap tubes or bottles at 37 °C. Cells were harvested by centrifugation (5000 g, 10 min) and suspended in one-tenth volume BHY medium containing 15% (v/v) glycerol at a density of ~1010 colony-forming units (CFU) mL 1. Aliquots were stored at 80 °C until required. Experimental cultures were produced by inoculation from stock suspension (50 lL) into BHY medium (20 mL) and were grown at 37 °C in closed tubes or bottles without shaking. Peptides Gly–Hyp, Gly–Pro, Gly–Pro–Ala, Gly–Pro– Hyp, Gly–Pro–Gly–Gly and Gly–Pro–Arg–Pro were obtained from Sigma-Aldrich (St. Louis, MO, USA).
Bacterial growth yields Bacterial cultures were prepared by inoculating stock suspension as above, and various concentrations of
International Endodontic Journal, 49, 462–470, 2016
463
Bacterial invasion of dentine Brittan et al.
the peptides (20 lL) were added to 180 lL portions of inoculated culture medium in microtitre plate wells. Growth yields of bacteria with or without peptides were determined by measuring optical density (at 600 nm) in a microplate reader (Synergy 2, BioTek, Winooski, VT, USA) of cultures after 16-h incubation at 37 °C.
Preparation of root dentine Noncarious, unrestored human canine or premolar teeth with single root canals were obtained with consent from patients who had undergone orthodontic extractions at the Bristol Dental Hospital. Bovine incisors were extracted from young (2.5 indicated heavy invasion, and TI 50 lm; and x3, where at least 5 tubules per section were invaded to a depth of 100 lm or greater. TI and TIF scores were normally distributed, and data were analysed by two-way ANOVA (GraphPad Prism 6.0, GraphPad Software Inc., La Jolla, CA, USA) with a significance level of 0.05.
Results Comparison of human and bovine dentine for invasion studies: light microscopy S. gordonii DL1 was associated with heavy invasion of both human and bovine dentine after 14 days as was S. mutans NG8 and E. faecalis JH2-2 (Fig. 1). The three bacterial species penetrated bovine dentine to a greater depth than in human dentine (data not shown). S. mutans cells formed a thin biofilm on the pulpal surface of both human and bovine dentine (Fig. 1). A graphical representation of the comparison of tubule invasion index (TI) and tubule invasion factor (TIF) for the invasion of bovine tubules by the three organisms is shown in Fig. 2. E. faecalis was clearly the most penetrative species into bovine dentine. This was exemplified by the higher TIF value of ~8.0 compared with the streptococci (Fig. 2). The TI
© 2015 International Endodontic Journal. Published by John Wiley & Sons Ltd
Brittan et al. Bacterial invasion of dentine
Figure 1 Bacterial penetration of human or bovine dentine showing heavy invasion by all species with S. mutans forming thin biofilms on the surface of both dentine types. Bars, 50 lm.
values for the three bacterial strains were more similar as they showed similar frequencies of tubule invasion, with depth of invasion not taken into account.
Effect of collagen peptides on tubule invasion Figure 3 shows the effects of different peptides glycine–proline–alanine (GPA), glycine–proline (GP), glycine–proline–arginine–proline (GPRP), glycine–proline–glycine–glycine (GPGG) and glycine–hydroxyproline (GO) (100 lg mL 1) on the invasion (TIF) of bovine dentinal tubules by S. gordonii DL1. All of the tripeptides and tetrapeptides tested were effective in reducing the invasion levels compared with the control (no peptide). Glycine–proline–alanine (GPA) was consistently the most reproducibly effective. Glycine– proline–glycine–glycine was also very effective. The dipeptide GP was not significantly inhibitory, whereas GO was inhibitory to invasion of bovine dentinal tubules by S. gordonii DL1.
Inhibition of human or bovine dentinal tubule invasion by collagen tripeptides
Figure 2 Comparison of TI (filled columns) and TIF (hatched columns) of bovine dentine. Error bars are SD, *P < 0.05; NS, not significant.
© 2015 International Endodontic Journal. Published by John Wiley & Sons Ltd
Incorporation of the tripeptide GPA (200 lg mL 1) into the invasion model medium resulted in inhibition of invasion of both human and bovine dentine by S. gordonii DL1, S. mutans NG8 and E. faecalis JH2-2 as visualized histologically (Fig. 4). The tripeptide
International Endodontic Journal, 49, 462–470, 2016
465
Bacterial invasion of dentine Brittan et al.
that the inhibitory effect of the tripeptide on S. gordonii invasion of bovine dentinal tubules was dose dependent. The TIF decreased from 5.11 0.87 with no GPA present to 1.66 0.43 in the presence of 0.2 mg GPA mL 1 (Fig. 6a). There was 43.9% reduction in invasion (TIF) at the lowest concentration of peptide tested (50 lg mL 1), and TIF values decreased with increasing peptide concentration up to 80% reduction in the maximum concentration tested. Comparison of the dose-dependent inhibition of dentinal tubule invasion by GPA with corresponding effects of GPA on the growth of S. gordonii DL1 showed a nearly 50% inhibition of dentinal tubule invasion at 50 lg mL 1 GPA (Fig. 6b). At the same concentration, there was little or no effect of the peptide on microbial growth rate or yield. At higher concentration (0.2 mg mL 1), there was a small, but not significant, reduction in growth of bacteria in suspension culture, whilst there was ~80% inhibition of dentinal tubule invasion. The GPA tripeptide did not therefore exert the inhibitory effects on dentinal tubule invasion by directly inhibiting bacterial growth.
Figure 3 Effects of different collagen-like peptides on the invasion (TIF) of bovine dentinal tubules by S. gordonii DL1. Error bars are SD, *P < 0.05, relative to control (DL1 column, untreated).
Glycine-Proline-Hydroxyproline (GPO) had similar inhibitory effects, which suggested that the nature of the third amino acid residue may not be crucial for inhibitory properties (Fig. 4). A graphical representation of the extent of inhibition by GPA or GPO is shown in Fig. 5. By TI measurements, the three microorganisms invaded human dentine to similar levels (Fig. 5a), and tripeptides GPA and GPO had similar statistically significant (P < 0.05) inhibitory effects. By TIF measurements, E. faecalis appeared to be more invasive of bovine dentinal tubules than S. gordonii or S. mutans, but this was not statistically significant (P = 0.092) (Fig. 5b). Tripeptide GPA significantly (P < 0.05) inhibited tubule invasion of all bacterial species tested but was more effective at blocking S. gordonii invasion of bovine dentine than S. mutans or E. faecalis (Fig. 5b).
GPA inhibited invasion but not bacterial growth The significant results with GPA tripeptide and S. gordonii DL1 were further analysed and showed
466
International Endodontic Journal, 49, 462–470, 2016
Discussion Three species of Gram-positive cocci that commonly infect dentine and pulp were used in the dentine invasion model, and all three species were able to invade human root dentine to levels previously shown (Love et al. 1997, Love 2001). Bovine root dentine was also incorporated into the study because this was easier to prepare and many more sections could be obtained from a single tooth root, and has been utilized for other invasion studies (Assouline et al. 2001). Bovine radicular dentine contains a lower density of dentinal tubules, and the tubules have larger diameter to those in human dentine (Lopes et al. 2009). The two types of dentine showed similar patterns of invasion with the three microorganisms tested. However, there was deeper invasion of bovine dentinal tubules by all three species of bacteria which was consistent with other studies and may be related to the larger diameter of tubules in bovine dentine (Lopes et al. 2009), which would facilitate better diffusion of nutrients (Sigusch et al. 2014). These observations confirmed that bovine radicular dentine was an appropriate substrate to investigate bacterial invasion of dentine. Previous studies have shown that S. gordonii bacterial cells interacted strongly with bundles of collagen type I and formed chains along the lengths of the
© 2015 International Endodontic Journal. Published by John Wiley & Sons Ltd
Brittan et al. Bacterial invasion of dentine
Figure 4 Inhibition of bacterial invasion of dentine by the presence of GPA or GPO tripeptides. Bars, 50 lm.
(a)
(b)
Figure 5 Comparison of bacterial invasion of dentine in the presence of GPA or GPO tripeptides. (a), TI values for human dentine; controls (open column), GPA (grey shaded), GPO (black filled). (b), TIF values for bovine dentine; controls (black columns, GPA (grey hatched). Error bars are SD, *P < 0.05, relative to controls (untreated).
© 2015 International Endodontic Journal. Published by John Wiley & Sons Ltd
International Endodontic Journal, 49, 462–470, 2016
467
Bacterial invasion of dentine Brittan et al.
(a)
(b)
Figure 6 Dose-dependent inhibitory effect of GPA tripeptide on invasion of S. gordonii DL1. (a) TI of bovine dentine. (b) growth yields (filled column) and invasion (hatched column) expressed as % untreated wild type. Error bars are SD, *P < 0.05, relative to controls (untreated).
fibres (Love et al. 1997, Heddle et al. 2003). This is the means by which it has been proposed that streptococci are able to grow along the lengths of dentinal tubules (e.g. as seen in Fig. 1) by adhering to intratubular collagen (Love & Jenkinson 2002). It has also been shown that collagen-derived peptides were sensed by the bacterial cells and affected the transcription of genes encoding cell surface proteins that interacted with collagen type I (Love et al. 1997, Heddle et al. 2003). Additionally, a collagen type 1-digest inhibited penetration of dentinal tubules by S. gordonii DL1 (Love et al. 1997) which was hypothesized to have resulted from collagen fragments blocking the bacterial adhesins recognizing dentinal collagen. These observations led us to investigate whether specific synthetic peptides based upon the collagen primary amino acid sequence might be effective in modulating penetration of dentinal tubules by bacteria. To develop this, the effects of tripeptides and tetrapeptides containing the common collagen sequence motif glycine–proline-Xaa, where X is any amino acid residue that fits into the repeating structural model for collagen (Fig. S1), were tested. The results
468
International Endodontic Journal, 49, 462–470, 2016
revealed that GPX peptides were inhibitory, but that GP alone was not sufficient to inhibit invasion unless P was replaced by O. Additionally, inhibition of invasion by GPA was not by direct inhibition of bacterial growth, suggesting that competitive inhibition of bacterial collagen-binding sites was a likely mode of action. The data presented raise an interesting question as to whether a collagen-like tripeptide sequence, for example GPA, or dipeptide GO, might be useful as target-specific inhibitors of dentine infection by bacteria. These peptides are potentially nontoxic (hydroxyproline is already utilized as a topical repair agent) and would have minimal side effects to patients. The actual concentrations of peptide required clinically would have to be tested empirically. The concentrations that were tested in vitro are likely to be orders of magnitude more than necessary for the bacterial challenge in vivo. Unlike antibiotics, the peptides would also not perturb the general composition or activities of the host microbiota, except at the target site for disease prevention. These peptides are found naturally within components of human tissue proteins and are products to which the host would not normally have
© 2015 International Endodontic Journal. Published by John Wiley & Sons Ltd
Brittan et al. Bacterial invasion of dentine
any adverse responses. They are not bactericidal and thus would not induce development of bacterial resistance. It can be suggested that the peptides work synergistically by adversely modulating bacterial gene expression (Love et al. 1997) and by blocking the proper recognition by the bacteria of the host collagen. As the peptide does not generally inhibit bacterial growth, it could work at the site of potential infection, preventing the invasive spread of infection. Additionally, the peptides have the potential to interfere with mixed species biofilm formation because inhibiting streptococci, which are primary colonizers, could help prevent a more pathogenic community from developing (Love et al. 2000). If these peptides, or derivatives such as mimetics, were to be of use clinically, they would need to show stability within the oral environment. This aspect of their potential has been investigated by incubating GPA peptide in human whole saliva at 37 °C for up to 16 h. Under these conditions, there was no significant loss of peptide concentration as measured by mass spectrometry (Fig. S2). Potential applications of such peptides could be (i) inhibition of bacterial invasion around a restoration and into dentinal tubules by topical application, (ii) incorporation into intracanal materials or medicaments, (iii) inclusion within periodontal intra-oral pocket material to inhibit recolonization of the exposed root dentine after it is disinfected or root planed: this might stop the pocket dentine acting as a reservoir for recolonization of the periodontal pocket, and (iv) incorporation into medicaments to inhibit bacterial invasion of wounds at sites of endodontic, periodontal or oral surgery treatment. Application of collagen-like peptide sequences to prevent colonization or invasion would enhance tissue regeneration, especially in periodontal clinical practice of fibrous (collagen) regeneration. Their application to membranes used in guided tissue regeneration around natural teeth or implants would be useful, as tissue regeneration will not occur if the area is infected.
Conclusion The study has shown that specific tripeptides and a dipeptide, the sequences of which are found in collagens, are able to inhibit bacterial penetration of dentine. These peptides, or derivatives of them, might thus allow for target-specific control of dentine infections, with potentially minimal side effects to the subject or to the normal oral microbiota.
© 2015 International Endodontic Journal. Published by John Wiley & Sons Ltd
Acknowledgements We would like to thank The University of Bristol Veterinary School staff associated with the Langford Abattoir for their assistance with provision of bovine teeth. We thank Nicola West, Martin Addy, Nick Banfield and Peter Shellis for helpful discussions. This study was supported in part by a grant from the University of Bristol Enterprise Development Fund.
References Assouline LS, Fuss Z, Mazor Y, Weiss EI (2001) Bacterial penetration and proliferation in root canal dentinal tubules after applying dentin adhesives in vitro. Journal of Endodontics 27, 398–400. Brown JH, Brenn L (1931) A method for the differential staining of gram-positive and gram-negative bacteria in tissue sections. Bulletin of the Johns Hopkins Hospital 48, 69–73. Chivatxaranukul P, Dashper SG, Messer HH (2008) Dentinal tubule invasion and adherence by Enterococcus faecalis. International Endodontic Journal 41, 873–82. Demuth DR, Duan Y, Brooks W, Holmes AR, McNab R, Jenkinson HF (1996) Tandem genes encode cell-surface polypeptides SspA and SspB which mediate adhesion of the oral bacterium Streptococcus gordonii to human and bacterial receptors. Molecular Microbiology 20, 403–13. Hancock HH III, Sigurdsson A, Trope M, Moiseiwitsch J (2001) Bacteria isolated after unsuccessful endodontic treatment in a North American population. Oral Surgery, Oral Medicine, Oral Pathology Oral Radiology and Endodontics 91, 579–86. Heddle C, Nobbs AH, Jakubovics NS et al. (2003) Host collagen signal induces antigen I/II adhesin and invasin gene expression in oral Streptococcus gordonii. Molecular Microbiology 50, 597–607. Jenkinson HF (2011) Beyond the oral microbiome. Environmental Microbiology 13, 3077–87. Jenkinson HF, Demuth DR (1997) Structure, function and immunogenicity of streptococcal antigen I/II polypeptides. Molecular Microbiology 23, 183–90. Kayaoglu G, Erten H, Ørstavik D (2008) Possible role of the adhesin Ace and collagen adherence in conveying resistance to disinfectants on Enterococcus faecalis. Oral Microbiology and Immunology 23, 449–54. Kelly CG, Younson JS, Hikmat BY et al. (1999) A synthetic peptide adhesion epitope as a novel antimicrobial agent. Nature Biotechnology 17, 42–7. Kowalski WJ, Kasper EL, Hatton JF, Murray BE, Nallapareddy SR, Gillespie MJ (2006) Enterococcus faecalis adhesin, Ace, mediates attachment to particulate dentin. Journal of Endodontics 32, 634–7. Lehner T, Caldwell J, Smith R (1985) Local passive immunization by monoclonal antibodies against streptococcal
International Endodontic Journal, 49, 462–470, 2016
469
Bacterial invasion of dentine Brittan et al.
antigen I/II in the prevention of dental caries. Infection and Immunity 50, 796–9. Lopes MB, Sinhoreti MAC, Gonini J unior A, Consani S, McCabe JF (2009) Comparative study of tubular diameter and quantity for human and bovine dentin at different depths. Brazilian Dental Journal 20, 279–83. Love RM (2001) Enterococcus faecalis - mechanism for its role in endodontic failure. International Endodontic Journal 34, 399–405. Love RM, Jenkinson HF (2002) Invasion of dentinal tubules by oral bacteria. Critical Reviews in Oral Biology and Medicine 13, 171–83. Love RM, Chandler NP, Jenkinson HF (1996) Penetration of smeared or nonsmeared dentine by Streptococcus gordonii. International Endodontic Journal 29, 2–12. Love RM, McMillan MD, Jenkinson HF (1997) Invasion of dentinal tubules by oral streptococci is associated with collagen recognition mediated by the antigen I/II family of polypeptides. Infection and Immunity 65, 5157–64. Love RM, McMillan MD, Park Y, Jenkinson HF (2000) Coinvasion of dentinal tubules by Porphyromonas gingivalis and Streptococcus gordonii depends upon binding specificity of streptococcal antigen I/II adhesin. Infection and Immunity 68, 1359–65. Lundy T, Stanley HR (1969) Correlation of pulpal histopathology and clinical symptoms in human teeth subjected to experimental irritation. Oral Surgery, Oral Medicine, and Oral Pathology 27, 187–201. Moses PJ, Power DA, Jesionowski AM, Jenkinson HF, Pantera EA Jr, Vickerman MM (2013) Streptococcus gordonii collagen-binding domain protein CbdA may enhance bacterial survival in instrumented root canals ex vivo. Journal of Endodontics 39, 39–43. Olgart L, Br€annstr€ om M, Johnson G (1974) Invasion of bacteria into dentinal tubules. Experiments in vivo and in vitro. Acta Odontologica Scandinavica 32, 61–70.
470
International Endodontic Journal, 49, 462–470, 2016
Ørstavik D, Haapasalo M (1990) Disinfection by endodontic irrigants and dressings of experimentally infected dentinal tubules. Endodontics & Dental Traumatology 6, 142–9. Ozok AR, Persoon IF, Huse SM et al. (2012) Ecology of the microbiome of the infected root canal system: a comparison between apical and coronal root segments. International Endodontic Journal 45, 530–41. Rams TE, Degener JE, van Winkelhoff AJ (2014) Antibiotic resistance in human chronic periodontitis microbiota. Journal of Periodontology 85, 160–9. Sigusch BW, Kranz S, Klein S et al. (2014) Colonization of Enterococcus faecalis in a new SiO/SiO2-microtube in vitro model system as a function of tubule diameter. Dental Materials 30, 661–8. Siqueira JF Jr, R^ ocßas IN (2009) Diversity of endodontic microbiota revisited. Journal of Dental Research 88, 969– 81. Sousa EL, Gomes BP, Jacinto RC, Zaia AA, Ferraz CC (2013) Microbiological profile and antimicrobial susceptibility pattern of infected root canals associated with periapical abscesses. European Journal of Clinical Microbiology and Infectious Disease 32, 573–80. Sundqvist G (1994) Taxonomy, ecology, and pathogenicity of the root canal flora. Oral Surgery Oral Medicine and Oral Pathology 78, 522–30.
Supporting Information Additional Supporting Information may be found in the online version of this article: Figure S1. Ball and stick model of the internal Gly-Pro-Xaa repeats within collagen Figure S2. Stability of GPX in human whole saliva.
© 2015 International Endodontic Journal. Published by John Wiley & Sons Ltd
