These two authors contributed equally to this study.
A. Devaraj,
Center for Microbial Pathogenesis, The Research Institute at Nationwide Children's Hospital, and The Ohio State University College of Medicine, Columbus, OH, USA
Center for Microbial Pathogenesis, The Research Institute at Nationwide Children's Hospital, and The Ohio State University College of Medicine, Columbus, OH, USA
Center for Microbial Pathogenesis, The Research Institute at Nationwide Children's Hospital, and The Ohio State University College of Medicine, Columbus, OH, USA
Center for Microbial Pathogenesis, The Research Institute at Nationwide Children's Hospital, and The Ohio State University College of Medicine, Columbus, OH, USA
Correspondence: Steven D. Goodman, Center for Microbial Pathogenesis, The Research Institute at Nationwide Children's Hospital, 700 Children's Dr., Columbus, OH, 43205, USA Tel.: +1 614 355 2761; fax: +1 614 722 2817; E-mail: steven.goodman@nationwidechildrens.org
Periodontal disease exemplifies a chronic and recurrent infection with a necessary biofilm component. Mucosal inflammation is a hallmark response of the host seen in chronic diseases, such as colitis, gingivitis, and periodontitis (and the related disorder peri-implantitis). We have taken advantage of our recently developed rat model of human peri-implantitis that recapitulates osteolysis, the requirement of biofilm formation, and the perpetuation of the bona fide disease state, to test a new therapeutic modality with two novel components. First we used hyperimmune antiserum directed against the DNABII family of proteins, now known to be a critical component of the extracellular matrix of bacterial biofilms. Second we delivered the antiserum as cargo in biodegradable microspheres to the site of the biofilm infection. We demonstrated that delivery of a single dose of anti-DNABII in poly(lactic-co-glycolic acid) (PLGA) microspheres induced significant resolution of experimental peri-implantitis, including marked reduction of inflammation. These data support the continued development of a DNABII protein-targeted therapeutic for peri-implantitis and other chronic inflammatory pathologies of the oral cavity in animals and humans.
Biofilm-mediated diseases such as gingivitis and periodontitis are considered the most common chronic inflammatory conditions of humans (Offenbacher et al., 2007). Half of the adult population presents with periodontitis that is a major cause of tooth loss in adults. Peri-implantitis is a similar inflammatory osteolytic infection that affects dental implants (Heitz-Mayfield & Lang, 2010). Periodontitis and peri-implantitis have associated microbial biofilm etiology, including Aggregatibacter actinomycetemcomitans, Porphyromonas gingivalis, Treponema denticola, and Tannerella forsythia (Socransky et al., 1998). The initial host response against these pathogenic bacteria causes transient local inflammation, commonly called gingivitis. It is increasingly evident that unresolved acute inflammation leads to chronic inflammation and neutrophil-mediated destruction of the periodontal ligaments and bone surrounding the affected tooth (Van Dyke & Kornman, 2008; Baloul et al., 2011). Although periodontal disease is locally manifested on tooth-supporting structures, there is increasing evidence that chronic inflammation modulates a patient's risk for developing systemic diseases including cardiovascular diseases, atherosclerosis, diabetes mellitus, osteoporosis, and, possibly, rheumatoid arthritis that result from the entry of oral microbes, microbial antigens, cytokines, and other proinflammatory mediators into the circulation (Cairo et al., 2004; Chun et al., 2005; Couper et al., 2008; Janket et al., 2008; Kornman, 2008; Williams et al., 2008; Kaur et al., 2013).
More than 700 prokaryote species have been identified and sequenced as part of the human oral microbiome (Dewhirst et al., 2010). The host is able to maintain a homeostatic state between commensals and pathogenic flora in healthy individuals; however, when this interactive relationship that maintains homeostasis is lost, disease starts. A known factor in the initiation and progression of periodontal inflammatory diseases is an increase in the proportion of Gram-negative bacteria present in the oral microbiome (Teles et al., 2010). One of the most important etiological agents in chronic and aggressive periodontitis in humans is A. actinomycetemcomitans (Slots, 1984; Muller et al., 1998; Umeda et al., 1998; Nalbant & Zadeh, 2002; Papapanou, 2002). Multiple virulence factors including leukotoxins, chemotaxis inhibitors, immunosuppressive proteins, collagenase, lymphocyte suppressive factor, and bone resorption agents facilitate survival of A. actinomycetemcomitans in the oral cavity and enable it to circumvent the host's immune response (Fives-Taylor et al., 1999; Henderson et al., 2003).
Periodontal debridement in combination with several adjunctive therapies including local use of antimicrobial agents, povidone-iodine during debridement (Greenstein, 1999; Sahrmann et al., 2010), chlorhexidine rinses (Southard et al., 1989), subgingival gels, and systemic antibiotics have been used in the treatment of periodontitis with only moderate improvement in clinical outcomes. (Karlsson et al., 2008; Schwarz et al., 2008; Sgolastra et al., 2012a,b; Drisko, 2014). The limited efficacy of the current treatment options and the need for repeated treatments demand the need for development of novel methods to effectively treat periodontitis due to biofilms and prevent the associated tooth loss.
One of the targets for the treatment of biofilm-mediated infectious diseases is the extracellular matrix that contains an abundant amount of extracellular DNA (eDNA) (Flemming & Wingender, 2010). Extracellular DNA plays an integral role as a structural component of the biofilm and has been believed to be an impenetrable barrier to both the immune system and antimicrobials (Nickel et al., 1985; Slinger et al., 2006; Flemming & Wingender, 2010). Therapeutic approaches for bacterial infections have failed in ‘biofilm-related diseases’, partially due to the protection rendered by the biofilm matrix. Interestingly, a family of proteins referred to as DNABII proteins have been ubiquitously found in the biofilm matrix of multiple human bacterial pathogens (Winters et al., 1993; Lunsford et al., 1996; Kim et al., 2002; Boleij et al., 2009) and identified as critical to maintaining the structural integrity of eDNA in the biofilm matrix (Goodman et al., 2011; Justice et al., 2012; Gustave et al., 2013; Novotny et al., 2013; Brockson et al., 2014). This family includes the ubiquitous eubacterial histone-like protein (HU) and integration host factor (IHF) found in α- and γ-proteobacteria (Rouviere-Yaniv & Gros, 1975; Swinger & Rice, 2004), with A. actinomycetemcomitans possessing three DNABII alleles (D7S_00047, ihfA; D7S_00171, ihfB; D7S_00989, HU). We have shown hyperimmune polyclonal antiserum against one of the family members IHF (isolated from Escherichia coli, anti-IHFEc), to effectively disrupt pre-formed biofilms of several pathogenic bacteria in vitro (Goodman et al., 2011; Novotny et al., 2013; Brockson et al., 2014) and even disperse polymicrobial sputum aggregates recovered from cystic fibrosis patients ex vivo (Gustave et al., 2013). Furthermore, we have found that immunization of chinchillas with E. coli IHF in a model of experimental otitis media results in significantly more rapid resolution of disease and clearance of non-typeable Haemophilus influenzae biofilms from the middle ear (Goodman et al., 2011). In a mouse model of urinary tract infection, we have shown, using mutants of uropathogenic E. coli strain UTI89 that lacks either subunit of IHF, is defective in colonization of the mouse bladder and kidney (Justice et al., 2012). More recently we have also demonstrated that DNABII proteins play a crucial role in uropathogenic E. coli biofilm development (Devaraj et al., 2015) and also identified eDNA and DNABII proteins in polymicrobial otorrhea solids (Idicula et al., 2016). These studies underscore the importance of DNABII proteins in maintaining the structural stability of eDNA in the extracellular matrix and the therapeutic potential of strategies that target these proteins to mediate biofilm disruption in multiple biofilm-mediated diseases.
In this study we employed a previously developed animal model (Freire et al., 2011) of a disease state that is closely related to periodontitis, namely peri-implantitis. This model takes advantage of an A. actinomycetemcomitans biofilm pre-formed on the surface of titanium implants, to investigate the therapeutic potential of hyperimmune antiserum directed against anti-IHFEc. We encapsulated anti-IHFEc in poly(lactic-co-glycolic acid) (PLGA) microspheres as a means to limit diffusion of the antibody away from the site of infection as well as to mediate sustained release at that site. PLGA is biodegradable and has gained attention in drug delivery systems because of several attractive properties such as minimal systemic toxicity, protection of cargo from degradation, capacity for sustained release of drug, compatibility with a wide array of small molecules and macromolecules, and the ability to target contents to a specific tissue/organ (Danhier et al., 2012). Most importantly, PLGA is approved by the US Food and Drug Administration for drug delivery in humans and has been used in a myriad of in vivo preclinical studies including cerebral and cardiovascular diseases, osteoporosis, diabetes, cancer, inflammation, and regenerative medicine (Danhier et al., 2012). Minocycline-loaded PLGA microspheres registered under the name Arestin™ (Meinberg et al., 2002; Renvert et al., 2008) is currently used as an adjunct to scaling and root planing in limiting but not reversing the bacterial nidus that causes periodontitis.
In this study, we demonstrated that treatment of a pre-formed A. actinomycetemcomitans biofilm with anti-IHFEc encapsulated in PLGA microspheres resulted in significant disruption of the biofilms in vitro. Furthermore, using our recently developed rat model of A. actinomycetemcomitans biofilm-induced oral osteolytic lesion (peri-implantitis), we have demonstrated that a single dose of anti-IHFEc in PLGA microspheres applied topically to implants that carried pre-formed A. actinomycetemcomitans biofilms facilitated rapid resolution of the pathogenic state including a significant reduction in inflammation and increased stability of the implants within bone in the oral cavity. Our studies emphasize the therapeutic potential of an anti-DNABII approach in A. actinomycetemcomitans-induced osteolytic infections, including peri-implantitis and periodontitis.
Methods
Aggregatibacter actinomycetemcomitans culture conditions for in vivo experiments
Strain D7S-1 of A. actinomycetemcomitans was originally recovered from a patient with aggressive periodontitis (Wang et al., 2002; Chen et al., 2010). Strain D7S-1 serotype A was grown for 48 h in modified tryptic soy agar (mTSB; 3% tryptic soy broth with 0.6% yeast extract; 1.5% agar), 5–10 colonies were transferred to 5 ml of liquid mTSB then mixed by vigorous vortexing to disperse the bacteria. The suspension was transferred to 2 ml of fresh mTSB at 1 : 20 dilution and incubated at 37°C in an atmosphere of 5% CO2.
Antibodies
Hyperimmune rabbit anti-IHFEc was described previously (Goodman et al., 2011; Gustave et al., 2013).
Encapsulation of the antibody in PLGA
The PLGA microspheres were created using a modified double emulsion technique (Beer et al., 1998). First, 1% polyvinyl alcohol (PVA; 31,500–50,000 molecular weight; Sigma-Aldrich, St Louis, MO) was dissolved in 10 ml phosphate-buffered saline (PBS) for 1 h with stirring at 65°C. Next, a 20% PLGA solution (40,000–70,000 molecular weight; Sigma-Aldrich) was made in 1 ml dichloromethane by spinning for 10 min until the PLGA was dissolved. To form the microspheres, the PLGA solution was spun while 100 μl of either undiluted naive rabbit serum or anti-IHFEc was added and then vortexed for an additional 30 s. The PLGA-anti-IHFEc or PLGA-naive serum suspension was then slowly pipetted into 2 ml of the PVA solution while spinning for an additional 30 s. The PLGA/PVA was then quickly added to 100 ml of 5% isopropyl alcohol with continuous stirring, then stirred for an additional 3 h. The mixture was transferred to 50 ml centrifuge tubes and spun at 515 ×g for 10 min at 4°C. The supernatant was removed and the microspheres were washed four times with 25 ml PBS. After the final wash, 1 ml of PBS was added to the microspheres and 100 μl aliquot stocks were prepared. The microspheres were stored at 4°C for immediate use or at −20°C for long-term storage.
Quantification of immunoglobulin G in PLGA microspheres
The Easy-Titer IgG assay kit (ThermoFisher Scientific, Waltham, MA) was used to determine the amount of total immunoglobulin G (IgG) encapsulated within the microspheres. PLGA microspheres containing anti-IHFEc were melted at 60°C for 30 min to release and quantify encapsulated IgG as per the manufacturer's instructions. This assay was repeated twice, with an average value of 734 ng anti-IHFEc IgG ml−1.
In vitro biofilm assay
In vitro biofilm assay was as previously described (Goodman et al., 2011). The A. actinomycetemcomitans was cultured in mTSB for 48 h at 37°C, in a humidified atmosphere containing 5% CO2. The culture was mixed by vigorous vortexing for 1 min to break apart the aggregates, then diluted 1 : 10 in mTSB and 200 μl of this bacterial suspension was inoculated into each well of an eight-well chamber slide (ThermoFisher Scientific). After 16 h of incubation at 37°C in 5% CO2, the medium in each well was replaced with fresh medium. After an additional 8 h incubation period, the medium was removed and replaced with one of the following: pre-warmed medium, medium that contained a 1 : 50 dilution of serum (naive or anti-IHFEc) or a 1 : 10 dilution of microspheres (average value of 734 ng anti-IHFEc IgG ml−1) that contained either naive serum or anti-IHFEc. Chamber slides were then incubated for an additional 16 h as described above. After incubation with serum or microspheres that contained serum, the medium was carefully removed; the biofilm was washed twice with 0.9% saline and stained with LIVE/DEAD® stain (Molecular Probes, Eugene, OR) for 15 min at room temperature as per the manufacturer's instructions. To determine the distribution of extracellular IHF, fixed A. actinomycetemcomitans biofilms were incubated with anti-IHFEc and then treated with goat anti-rabbit IgG conjugated to AlexaFluor® 647 (Life Technologies, Carlsbad, CA). The A. actinomycetemcomitans biofilms were visualized with LIVE/DEAD® stain. The biofilms were then imaged using a 63× objective on a Zeiss 510 Meta-laser scanning confocal microscope (Carl Zeiss, Oberkochen, Germany).
Biofilm imaging and quantification
After biofilms were washed twice with 0.9% saline, samples were fixed with a solution of 1.6% paraformaldehyde, 0.025% glutaraldehyde, and 4.0% acetic acid in phosphate buffer at pH 7.4. The biofilms were imaged using a 63× objective on a Zeiss 510 Meta-laser scanning confocal microscope (Carl Zeiss, Oberkochen, Germany). All in vitro biofilm assays were repeated a minimum of three times on separate days and all individual biofilm assays were performed in duplicate on each assay day. The efficacy of treatment was assessed based on differences in biofilm height, thickness, and biomass as determined by COMSTAT analysis (Heydorn et al., 2000). Values of percent reduction in biofilm architectural parameters upon exposure to immune serum were compared with those obtained upon exposure to naive serum.
Animals
Sprague–Dawley, virgin, 6-week-old female rats (n = 83) (Charles River Laboratories, Hollister, CA) were housed in a laboratory at 20–24°C under a 12 h light/12 h dark cycle and fed ad libitum (Purina Laboratory Rodent Chow, Purina Milk, Richmond, IN). Animal care was in accordance with the NIH Guide for the Care and Use of Laboratory Animals prepared by the Institute of Laboratory Animal Resources, National Research Council, all applicable government regulations, and university policies governing the care and use of laboratory animals. All protocols and procedures were approved by the Institutional Animal Care and Use Committee of the University of Southern California and are in accordance with the Panel on Euthanasia of the American Veterinary Medical Association.
In vivo inflammation model
Titanium implant screw (1.2 × 4.5 mm) surfaces were modified by grit blasting with Al2O3 (100 μm) and HCl etching (pH 3, 20 min, 80°C) to produce a rough surface before being inoculated with A. actinomycetemcomitans as previously described (Freire et al., 2011). Briefly, the heads of implants (supragingival portion) were partially submerged in either sterile medium (mTSB) or mTSB inoculated with A. actinomycetemcomitans D7S-1 followed by incubation for 2 days at 37°C with 5% CO2 (Fig. 1). Rats were anesthetized with isoflurane and either two A. actinomycetemcomitans biofilm-inoculated implants or two control implants were transmucosally placed into the hard palate adjacent to the maxillary alveolar ridge on either side of the oral cavity (n = 32). Seven days post-surgery, PLGA microspheres (20 μl) containing naive or anti-IHFEc antisera (734 ng anti-IHFEc IgG ml−1) were applied topically at the site of implant insertion. After 3 days of treatment, peri-implant mucosal inflammation and bleeding were evaluated blindly. Tissue inflammation was classified clinically on a scale of 0–3 as follows: 0, no bleeding; 1, slight bleeding; 2, moderate bleeding; 3, severe bleeding (Renvert et al., 2008). After evaluation, rats were euthanized by CO2 asphyxiation and the skulls were harvested and stored in 10% buffered formalin for histological analysis.
Schematic representation of the animal model used to investigate the host response to implants that carried pre-formed Aggregatibacter actinomycetemcomitans (Aa) biofilms. (I) The heads of the titanium implants were partially submerged in sterile medium (mTSB) or mTSB inoculated with A. actinomycetemcomitans for 2 days at 37°C with 5% CO2 (to produce a uniform biofilm on the surface of the implant head; see Supplementary material, Fig. S3). (II) Rats were anesthetized and either Aa biofilm-inoculated implants or control implants were transmucosally placed into the palate adjacent to the maxillary alveolar ridge. (III) Seven days post-surgery animals were treated topically with poly(lactic-co-glycolic acid) (PLGA) microspheres containing naive serum or antiserum directed against Escherichia coli integration host factor (anti-IHFEc). After 3 days of treatment, clinical evaluation of peri-implant mucosal inflammation, bleeding, and implant mobility was recorded.
Implant stability
To investigate the stability of the implants surgically placed into the hard palate of the rats, clinical assessment of relative mobility of the implant was determined with a periodontal probe as well as through reverse torque analysis of the implants. Seven days after surgical placement of either sterile implants or those with pre-formed A. actinomycetemcomitans biofilms into the bone of the hard palate of a separate cohort of animals (n = 19), the implants were treated with either naive serum (n = 5) or anti-IHFEc (n = 8) for 3 days. At the end of 3 days, the animals were anesthetized and the implants were analyzed with a periodontal probe by naked eye by a single observer (MOF) for relative ability to move them in both a coronal or palatal plane. The implants were considered unstable if they were positive for any movement (Newman et al., 2002). For reverse torque analysis, animals that were surgically implanted with either control implants or those with pre-formed A. actinomycetemcomitans biofilms and treated with anti-IHFEc (n = 3) or naive serum (n = 3) were euthanized by CO2 asphyxiation after 3 days and samples were fixed in 10% formalin. After maxillary tissue dissection and stabilization, a counterclockwise (reverse) force was applied to the implant with the aid of a computerized torque driver to remove the implant. The torque required to remove the implant was recorded (Jividen & Misch, 2000).
Histology
Bone-implant specimens (n = 32) were fixed with 10% neutral buffered formalin (Richard-Allan Scientific, Kalamazoo, MI) for 24 h at 4°C. Each tissue was trimmed in the coronal plane and the resulting samples were processed to an individual Technovit 7200 VLC acrylic resin block (Heraeus Kulzer, Germany). One ground and polished section was created per block from the longitudinal through the midline of the implant. Samples were dehydrated in graded ethanol (70, 95, and 100%) and embedded in paraffin. Five-micrometer sections were cut and stained with hematoxylin & eosin (Zijnge et al., 2010). Stained histological slides were imaged using a Zeiss Axio Cam (Carl Zeiss, Oberkochen, Germany) with a 10× objective.
Statistics
All data were analyzed and graphed using GraphPad Prism (GraphPad Software, Inc., La Jolla, CA). Means and standard error of the mean (SEM) were evaluated by t-test (Figs 2 and 5) and Mann–Whitney U-test (Figs 3 and 4). A P-value ≤ 0.05 was considered significant.
Antiserum directed against Escherichia coli integration host factor (anti-IHFEc) antibodies diminished pre-formed Aggregatibacter actinomycetemcomitans (Aa) biofilms in vitro. The Aa biofilms were pre-formed in vitro for 24 h in modified tryptic soy broth and treated with one of the following: (A) medium only, (B) naive serum, or (C) anti-IHFEc. In addition to direct treatment, similar treatments were performed with (D) poly(lactic-co-glycolic acid) (PLGA) microspheres that contained either (E) naive serum (F) or anti-IHFEc. Representative images of Aa biofilms are shown. COMSTAT analysis was employed to calculate maximum height (G), average thickness (H) and biomass (I). (J) Colony-forming units (CFU) ml−1 in biofilm and planktonic state were enumerated. All in vitro biofilm assays were repeated a minimum of three times on separate days, and all individual biofilm assays were carried out in duplicates on each assay day. Data are presented as mean values ± standard deviation. A P-value ≤ 0.05 was considered significant. *P = 0.01 to P = 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001.
Aggregatibacter actinomycetemcomitans (Aa) biofilm effects on surrounding tissue. Control implants or Aa biofilm-inoculated implants were placed into the anterior region of the first molars in maxillae on the right and left sides of the oral cavity. Representative images of peri-implant tissues with control (A) or Aa biofilm-inoculated implants (B) 7 days after surgical placement of titanium implants are illustrated. Clinical evaluation of peri-implant mucosal inflammation and bleeding were recorded from 1 to 7 days and tissue inflammation (mean ± standard deviation) was classified clinically with score values: 0–3 (0, no bleeding; 1, slight bleeding; 2, moderate bleeding; 3, severe bleeding). (C) Inflammation and bleeding score after 7 days post-surgery. Statistical significance was evaluated by Mann–Whitney U-test; *P = 0.01 to P = 0.05.
Effect of antiserum directed against Escherichia coli integration host factor (anti-IHFEc) on Aggregatibacter actinomycetemcomitans (Aa) biofilm Induced peri-implant mucosal inflammation. Two Aa biofilm-inoculated implants were placed into the anterior region of the first molars in maxillae on the right and left sides of the oral cavity. (A, C) Representative images of peri-implant tissues 7 days after surgical placement of titanium implants and before treatment are illustrated. Representative images of the tissue response of rat mucosa 3 days after treatment with either naive serum (B) or anti-IHFEc (D) encapsulated in poly(lactic-co-glycolic acid) (PLGA) microspheres are illustrated. (E) After 3 days of treatment, clinical evaluation of peri-implant mucosal inflammation and bleeding, were recorded and tissue inflammation was classified clinically with score values: 0–3 (0, no bleeding; 1, slight bleeding; 2, moderate bleeding; 3, severe bleeding). Inflammation and bleeding score is represented as mean ± standard deviation. Statistical significance was evaluated by Mann–Whitney U-test; **P < 0.01.
Biomechanical evaluation of stability of implants in bone of the rat oral cavity. (A) Two Aggregatibacter actinomycetemcomitans (Aa) biofilm-inoculated implants were placed into the anterior region of the first molars on the right and left sides of the oral cavity. Seven days after placement of the biofilm-inoculated implant, the implants were treated with either naive serum or antiserum directed against Escherichia coli integration host factor (anti-IHFEc) encapsulated in poly(lactic-co-glycolic acid) (PLGA) microspheres. The clinical stability of the implant was evaluated by moving the implants coronally or palatally with a periodontal probe in anesthetized animals. The implants were considered stable if no movement was observed. (B) Animals that were surgically implanted with Aa biofilm-inoculated implants and treated with anti-IHFEc (n = 3) or naive serum (n = 3) were euthanized after 3 days and the stability of the implant was evaluated by reverse torque analysis. Representative images of hematoxylin & eosin staining of the bone (dark red) upon treatment with naive serum (C) or anti-IHFEc (D) are shown. Data are represented as mean ± standard deviation. *P = 0.01 to P = 0.05.
Results
Hyperimmune antiserum to IHFEc disrupted pre-formed A. actinomycetemcomitans biofilms in vitro
We have previously demonstrated that hyperimmune anti-IHFEc disrupts biofilms formed by multiple known human pathogens under laboratory conditions and that sequestration of extra-bacterial IHF from the bulk media surrounding the extracellular matrix increases sensitivity of bacteria to antimicrobials (Goodman et al., 2011). To now determine whether IHF is within the extracellular matrix, we performed immunofluorescence by incubating fixed A. actinomycetemcomitans biofilms with anti-IHFEc followed by goat anti-rabbit IgG conjugated to AlexaFluor® 647 (Invitrogen). The A. actinomycetemcomitans biofilms were further visualized with LIVE/DEAD® stain. We observed labeling (white) throughout the biofilm (see Supplementary material, Fig. S1) that confirmed the presence of IHF within the extracellular matrix of A. actinomycetemcomitans biofilms. To test the ability of anti-IHFEc to disrupt pre-formed A. actinomycetemcomitans biofilms, we treated in vitro pre-formed A. actinomycetemcomitans biofilms (Fig. 2A) with either naive rabbit serum (Fig. 2B) or anti-IHFEc at an optimal dilution of 1 : 50 (see Supplementary material, Fig. S2) for 16 h (Fig. 2C). As shown in Fig. 2B, when compared with a biofilm incubated with naive serum, treatment with anti-IHFEc significantly disrupted the A. actinomycetemcomitans biofilms (Fig. 2C). Quantification of biofilm parameters with COMSTAT revealed that treatment with anti-IHFEc significantly decreased the maximum height by 18%, average thickness by 47% and biomass by 45%, respectively (Fig. 2G–I). The enumeration of bacteria in the planktonic and biofilm states revealed that although the total bacteria (planktonic + biofilm) upon treatment with anti-IHFEc was comparable to that in the presence of naive serum, there was a significant increase in planktonic bacteria in the presence of anti-IHFEc compared with treatment with naive serum (Fig. 2J). This outcome suggested that the treatment with anti-IHFEc simply released bacteria from the biofilm to the planktonic state (Fig. 2J). In a parallel experiment, in which we treated dental implants with an A. actinomycetemcomitans biofilm with either medium or anti-IHFEc, we also demonstrated disruption of A. actinomycetemcomitans biofilm upon treatment with anti-IHFEc (see Supplementary material, Fig. S3) suggesting that this treatment was not dependent on the attached surface.
In addition to the direct treatment of A. actinomycetemcomitans biofilms with anti-IHFEc, we tested the effect of exposure to anti-IHFEc that had been encapsulated in biodegradable PLGA microspheres as a possible means to limit diffusion of the antibody and target the antibody for sustained release at the site of infection for later use in our animal model (Freire et al., 2011). Consistent with the previous observation, in vitro pre-formed A. actinomycetemcomitans biofilms were disrupted by anti-IHFEc encapsulated in PLGA microspheres, in comparison to the treatment with naive serum encapsulated in PLGA microspheres (Fig. 2B,C,E–I). These data further demonstrated that PLGA microspheres could be used as a mode of delivery of the antibody in an in vivo model of experimental peri-implantitis.
Aggregatibacter actinomycetemcomitans induced robust inflammation in a rat model of biofilm-mediated peri-implantitis
To characterize the host response to A. actinomycetemcomitans, we established a 48 h biofilm on the supragingival portion of the implant (Fig. 1; and see Supplementary material, Fig. S3) and transmucosally placed either a control implant with no exogenous biofilm (implant incubated in sterile medium) or implants with pre-formed A. actinomycetemcomitans biofilms into the hard palate adjacent to the maxillary alveolar ridge (Fig. 1). Implants were blindly evaluated for inflammation and bleeding every day for 7 days. As shown in Fig. 3A, the control implants without pre-formed A. actinomycetemcomitans biofilm presented with healthy peri-implant mucosal tissue, as noted by the absence of inflammation. Conversely, clinical manifestations of inflammation including bleeding, redness, and swelling of the mucosal tissue surrounding implants were observed in the case of implants with pre-formed A. actinomycetemcomitans biofilms (Fig. 3B,C). Although inflammation and bleeding due to the surgical procedure ceased 2 days after surgery and healthy mucosa was maintained for up to 7 days in those animals implanted with control implants (n = 16; Fig. 3C), those that received implants with pre-formed A. actinomycetemcomitans biofilms (n = 16) exhibited severe peri-mucosal inflammation throughout the 7-day observation period (Fig. 3C). Taken together with our previous studies, these data demonstrated that the implants on which A. actinomycetemcomitans biofilms were present induced inflammation with tissue destruction and could thereby be employed to investigate the host immune responses to biofilms formed by A. actinomycetemcomitans.
Aggregatibacter actinomycetemcomitans biofilm-induced peri-implant mucosal inflammation was resolved by treatment with anti-IHFEc
Next, we employed the rat model described above to evaluate the ability of anti-IHFEc to resolve mucosal inflammation caused by the implants with pre-formed A. actinomycetemcomitans biofilms. First, inflammation was clinically evaluated 7 days after the surgical placement of implants into the oral cavity of rats (Fig. 4A,C,E). At the end of 7 days, treatment with a single topical dose of either naive or anti-IHFEc encapsulated in PLGA microspheres was applied to the implant and the surrounding tissue. Mucosal inflammation was evaluated 3 days after treatment. Although the implants with A. actinomycetemcomitans biofilms that were treated with naive serum demonstrated persistence of inflammation as shown in Fig. 4B, the implants with A. actinomycetemcomitans biofilms that were treated with anti-IHFEc showed a significant decrease in inflammation (Fig. 4D). The severity of the bleeding, erythema, and edema after treatment with either naive or anti-IHFEc was also scored and the results demonstrated that although implants that carried pre-formed A. actinomycetemcomitans biofilms that had been treated with naive serum encapsulated in PLGA microspheres received moderate to high inflammation scores (1.8 ± 1.0, n = 5), those implants treated with anti-IHFEc encapsulated in PLGA microspheres showed no signs of inflammation (Fig. 4E) (n = 8). These data demonstrated the efficacy of topical treatment of anti-IHFEc encapsulated in PLGA microspheres against A. actinomycetemcomitans biofilm-mediated mucosal inflammation in a rat model of experimental peri-implantitis.
Treatment of A. actinomycetemcomitans biofilm-coated implants with anti-IHFEc increased the stability of the implant in a rat model of experimental peri-implantitis
Titanium implants are biocompatible and bioinert (Beder & Eade, 1956; Branemark et al., 1977). We have demonstrated that the inflammation induced during implantation was transient and resolves within 2 days (Fig. 3C). In contrast, naturally occurring or experimentally induced pathogenic biofilm formation on implants is associated with progressive loss of stability leading to implant failure (Hultin et al., 2002; Renvert et al., 2008; Freire et al., 2011). Also, any micro-movement resulting from continuous inflammation interferes with collagen deposition of the connective tissue that is responsible for mucosal healing and bone deposition that is required for bone healing (Vandamme et al., 2007). Here, we evaluated the effect of treatment of anti-IHFEc encapsulated in PLGA microspheres on the stability of implants with pre-formed A. actinomycetemcomitans biofilms in the oral cavity of rats. As described above, 7 days after surgical placement of the A. actinomycetemcomitans biofilm-coated implants into the oral cavity, the implants were treated with a single dose of either naive or anti-IHFEc encapsulated in PLGA microspheres for 3 days. The clinical stability of the implant was evaluated by application of light lateral force with a periodontal probe in anesthetized animals. The implants were considered stable if no movement was observed. As shown in Fig. 5A, although all of the implants treated with anti-IHFEc encapsulated in PLGA microspheres were stable, only about 40% of those treated with naive serum encapsulated in PLGA microspheres were stable.
To further determine the stability of the implants the torque required to remove the implant was evaluated after the rats were euthanized. As shown in Fig. 5B, although the implants treated with naive serum exhibited a lower implant removal torque (0.5 N cm−2 ± 0.2, n = 3) consistent with reduced implant stability, implants treated with anti-IHFEc exhibited about a four-fold greater torque required to remove the implant (1.8 N cm−2 ± 0.4, n = 3). To investigate the effect of treatment of implants that contained pre-formed A. actinomycetemcomitans biofilms with anti-IHFEc on bone healing, histological analysis of the peri-implant tissues was performed. As shown in representative images (Fig. 5C), although bone surrounding the implant with A. actinomycetemcomitans biofilm that had been treated with a single dose of naive serum exhibited extensive disintegration, bone surrounding the implant that had been treated with a single dose of anti-IHFEc (Fig. 5D) was intact. These data demonstrated that treatment with anti-IHFEc allowed resolution of inflammation with healing that resulted in increased stability of implants that contained pre-formed A. actinomycetemcomitans biofilms and further underscored the therapeutic potential of anti-IHFEc in an animal model of experimental peri-implantitis.
Discussion
Homeostasis is the key to maintaining a healthy stable environment between the commensal microbiota and the host. Pathogenic bacterial biofilms such as those formed by A. actinomycetemcomitans disrupt this homeostasis causing dysbiosis and resulting in disease (Slots, 1984; Fives-Taylor et al., 1999; Nalbant & Zadeh, 2002; Henderson et al., 2003; Cairo et al., 2004; Freire et al., 2011; Bezerra Bde et al., 2012). Uncontrolled mucosal inflammation is a hallmark sign of chronic diseases of the oral cavity, including oral cancer, tonsillitis, gingivitis, peri-implantitis, and periodontitis. As unresolved acute inflammation results in chronic inflammation and subsequent destruction of the periodontal ligament and bone, it is imperative to control and resolve the inflammation to prevent tooth loss. Hence there is an urgent need for the development of targeted therapies to minimize pathogenic oral flora and thereby facilitate the reversal of dysbiosis caused by the presence of a dominant pathogenic biofilm.
Animal models of periodontitis often involve the introduction of bacteria into the oral cavity through oral gavage or direct injection of the pathogenic bacteria into the oral tissues (Trombone et al., 2009). The pathogenic bacteria used for inoculation are typically in the planktonic state (Kinane & Hajishengallis, 2009) and hence retention of these bacteria in the oral cavity is greatly diminished, as these pathogenic bacteria must compete with the existing oral microbiota and resist the host immune response. Although retention of bacteria can be achieved by the placement of cotton or silk ligatures around the teeth or implants (Lindhe et al., 1992; Berglundh et al., 2007; Albouy et al., 2009; Duarte et al., 2010), the tissue destruction caused by the placement of the ligature fails to mimic in vivo pathogenesis (Karimbux et al., 1998; Wilensky et al., 2009). To circumvent these barriers, we have previously developed an animal model wherein we employed titanium implants as a colonizing surface for A. actinomycetemcomitans biofilm formation in vitro, subsequently implanting these biofilm-coated implants transmucosally in the rat jaw (Freire et al., 2011). This technique allows for successful colonization and establishment of A. actinomycetemcomitans biofilm in the absence of antagonistic commensals and the immune response before placement in submucosal space in addition to eliciting a subsequent robust immune response characterized by inflammation and bleeding (Fig. 3) that persisted for weeks and resulting in exfoliation of the implant (Freire et al., 2011). In the present study we employed this model to investigate the effect of a hyperimmune polyclonal antiserum against one of the DNA binding proteins, IHF on an A. actinomycetemcomitans biofilm and demonstrated that a single dose of topical application of the antiserum encapsulated in PLGA microspheres promoted the resolution of the disease state including demonstrating greater stability of the implants in the oral cavity suggesting that this approach could serve as an effective treatment of periodontitis.
In this study, we encapsulated anti-IHFEc in PLGA microspheres as a means to limit diffusion of the antibody and target the antibody for sustained release at the site of infection in our animal model. There are potentially two mechanisms by which antiserum encapsulated in PLGA microspheres could mediate debulking of A. actinomycetemcomitans biofilm. One possibility is that the degradation of PLGA microspheres resulted in release of the antiserum at the site of infection. PLGA microspheres degrade by hydrolysis or by enzyme-mediated cleavage of the backbone ester bonds (Park, 1995) that contribute to the release of the cargo. The process of degradation of these microspheres is influenced by several factors including the ratio of glycolic to lactic acid, crystallinity, average molecular weight of the polymer, type of drug encapsulated, and other environmental factors, and therefore the rate of release of the encapsulated molecule can range from a few days to several weeks (Makadia & Siegel, 2011). The second possibility is that the microspheres simply served as a ‘sponge’ to sequester the IHF from the A. actinomycetemcomitans biofilm. Indeed, we have recently demonstrated, using a transwell system, that treatment of in vitro pre-formed non-typeable H. influenzae (a related bacterium from the same family Pasteurellaceae) biofilms with anti-IHFEc causes release of the resident bacteria by sequestering free/unbound IHF in the vicinity of the biofilm without the need for direct contact (Brockson et al., 2014). It is this sequestration, which causes a shift in equilibrium between the free and bound IHF, that eventually results in destabilization of the biofilm. It is possible that the PLGA microspheres containing anti-IHFEc served as a ‘sponge’ to sequester free IHF in a biofilm analogous to the transwell system to mediate disruption of the A. actinomycetemcomitans biofilm. The fact that a single dose of the antibody encapsulated in these microspheres was capable of disrupting pre-formed A. actinomycetemcomitans biofilm in vitro (Fig. 2) indicates the potential use of these microspheres for delivery of the antibody for a robust and rapid therapeutic effect. Importantly, this platform mimics Arestin™, a PLGA-based therapeutic containing an antimicrobial that is already an accepted approach in stabilizing the progression of periodontal disease (Meinberg et al., 2002; Renvert et al., 2008). As we have already shown that antimicrobials are synergistic for biofilm resolution with anti-IHFEc (Goodman et al., 2011; Brandstetter et al., 2013; Novotny et al., 2013; Brockson et al., 2014), future experiments will test a combined version of Arestin with anti-IHFEc for improved efficacy and eradication of periodontal biofilms.
Inflammation is fundamentally a host defense mechanism; however, if uncontrolled, it leads to tissue destruction and damage. Traditionally non-steroidal anti-inflammatory drugs are employed to inhibit inflammatory responses (Van Dyke & Serhan, 2003). In recent years pro-resolving molecules such as lipoxins, resolvins, and protectins that act as agonists to stimulate the resolution of inflammation have been actively investigated for the treatment of periodontitis (Hasturk et al., 2007; Serhan & Chiang, 2008). Although these molecules could effectively control inflammation, they fail to target the underlying cause of the inflammatory response that is mediated by pathogenic biofilms. In our animal model we have directly targeted the pathogenic biofilm with a hyperimmune antiserum directed against one of the members of the DNABII family and demonstrated that a single dose of the anti-IHFEc encapsulated in PLGA microspheres effectively resolved the inflammation in just 3 days (Fig. 4). Since this approach directly targets the pathogenic biofilm and resolves inflammation associated with the early stages of a disease state that mimics periodontitis, this approach holds great potential as a therapeutic as well as a prophylactic to the disease state before significant bone loss.
Periodontitis progresses from inflammation of gingiva without any bone loss termed as gingivitis, to inflammation of gingiva and the surrounding tissues with moderate to severe bone loss that ultimately results in tooth loss. Biofilms of A. actinomycetemcomitans have been shown to promote osteoclastogenesis and promote bone resorption in humans and in experimental models (Fives-Taylor et al., 1999; Schreiner et al., 2003; Freire et al., 2011; Bezerra Bde et al., 2012). The treatment of biofilm-inoculated implants with anti-IHFEc encapsulated in PLGA microspheres increased the stability of the implant in the oral cavity as measured by the reverse torque analysis (Fig. 5B). Whereas histological analysis of implants treated with naive serum revealed bone disintegration, intact bone was evident in implants treated with anti-IHFEc (Fig. 5C,D). These data suggest that treatment with anti-IHFEc could potentially ameliorate inflammation and so accelerate healing. Future work will include a more thorough analysis of this restoration process, including the relative changes in cytokines and other proinflammatory markers.
The DNABII family of proteins has been shown to be critical for the stability of biofilm matrix formed by several pathogenic bacteria both in vitro and in vivo (Goodman et al., 2011; Novotny et al., 2013; Brockson et al., 2014). We have previously demonstrated that antiserum directed against one of the DNABII proteins results in release of the resident bacteria and these released bacteria are more sensitive to the action of both antibiotics and the immune system (Goodman et al., 2011; Novotny et al., 2013). Hence anti-IHFEc encapsulated in PLGA microspheres has a great potential in the treatment of periodontitis, as it could potentially enhance the ability of the immune system to clear the bacterial biofilms that cause the disease state as well as quiet the host's inflammatory response that results in collateral tissue damage.
Acknowledgements
This work was supported by NIH grant R01DC011818 to SDG and LOB, NIH grant DE012212 to CC and HZ and Ideas Empowered Program, University of Southern California to SDG. We thank Joseph Jursicek for imaging the histology slides and Lauren Mashburn-Warren for critical reading of the manuscript.
Author contributions
MOF, JSD, CC, LOB, HHZ, and SDG conceived and designed the experiments; MOF, AD, JSD, AY, and JBN performed the experiments; MOF, AD, AY, and JBN analyzed the data; HHZ, CC, SDG, and LOB contributed reagents/materials/analysis tools; and MOF and AD wrote the paper.
Conflicts of interest
LOB and SDG own equity in a company, ProclaRx, outside the submitted work. In addition, LOB and SDG have a patent Compositions and methods for the removal of biofilms issued to ProclaRx.
2041-1014/asset/olalertbanner.jpg?v=1&s=cc9a807060d56e047c44f552cf5f4b548907907d)
