Mahmoud et al. J Nanobiotechnol (2018) 16:69 https://doi.org/10.1186/s12951-018-0396-4 Journal of Nanobiotechnology

RESEARCH Open Access BAR‑encapsulated nanoparticles for the inhibition and disruption of Porphyromonas gingivalis–Streptococcus gordonii bioflms Mohamed Y. Mahmoud4,5, Donald R. Demuth1,3*† and Jill M. Steinbach‑Rankins2,3,4,5*†

Abstract Background: Porphyromonas gingivalis adherence to oral streptococci is a key point in the pathogenesis of peri‑ odontal diseases (Honda in Cell Host Microbe 10:423–425, 2011). Previous work in our groups has shown that a region of the streptococcal antigen denoted BAR (SspB Adherence Region) inhibits P. gingivalis/S. gordonii interaction and bioflm formation both in vitro and in a mouse model of periodontitis (Daep et al. in Infect Immun 74:5756–5762, 2006; Daep et al. in Infect immun 76:3273–3280, 2008; Daep et al. in Infect Immun 79:67–74, 2011). However, high localized concentration and prolonged exposure are needed for BAR to be an efective therapeutic in the oral cavity. Methods: To address these challenges, we fabricated poly(lactic-co-glycolic acid) (PLGA) and methoxy-polyethylene glycol PLGA (mPEG-PLGA) nanoparticles (NPs) that encapsulate BAR peptide, and assessed the potency of BAR- encapsulated NPs to inhibit and disrupt in vitro two-species bioflms. In addition, the kinetics of BAR-encapsulated NPs were compared after diferent durations of exposure in a two-species bioflm model, against previously evaluated BAR-modifed NPs and free BAR. Results: BAR-encapsulated PLGA and mPEG-PLGA NPs potently inhibited bioflm formation (IC50 0.7 μM) and also disrupted established bioflms (IC50 1.3 μM) in a dose-dependent manner. In addition, BAR released= during the frst 2 h of administration potently inhibits= bioflm formation, while a longer duration of 3 h is required to disrupt pre- existing bioflms. Conclusions: These results suggest that BAR-encapsulated NPs provide a potent platform to inhibit (prevent) and disrupt (treat) P. gingivalis/S. gordonii bioflms, relative to free BAR. Keywords: Polymer nanoparticle, Poly(lactic-co-glycolic acid), Peptide delivery, Drug delivery, Porphyromonas gingivalis, Streptococcus gordonii, Periodontal disease, Oral bioflm

Background Tannerella forsythia, and Treponema denticola. Together Periodontal disease is a group of chronic infammatory these pathogens are known as the “red complex” [1]. diseases commonly caused by Porphyromonas gingivalis, Te progression of periodontal disease can cause tissue destruction and tooth loss, and if left untreated can con- tribute to systemic conditions of increased cancer risk, *Correspondence: [email protected]; [email protected] †Donald R. Demuth and Jill M. Steinbach-Rankins are Co-Senior authors cardiovascular disease, diabetes, rheumatoid arthritis, 1 Department of Oral Immunology and Infectious Diseases, University pulmonary disease, and obesity [2, 3]. of Louisville School of Dentistry, 501 S. Preston St, Louisville, KY 40202, Current periodontal treatments aim to reduce bacte- USA 2 Department of Bioengineering, University of Louisville Speed School rial plaque formation in the oral cavity using primar- of Engineering, 505 S. Hancock St., Room 623, Louisville, KY 40202, USA ily physical and chemical (antibiotic) methods [4, 5]. Full list of author information is available at the end of the article However, current antibiotic treatment strategies exhibit

© The Author(s) 2018. This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creat​iveco​mmons​.org/licen​ses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creat​iveco​mmons​.org/ publi​cdoma​in/zero/1.0/) applies to the data made available in this article, unless otherwise stated. Mahmoud et al. J Nanobiotechnol (2018) 16:69 Page 2 of 12

non-specifc activity, afecting benefcial organisms also investigated. In this study, we sought to develop a formu- present in the oral microbiome. Additional potential lation that encapsulates and prolongs the delivery of BAR, risks include the development of anti-bacterial resistant for durations relevant to oral delivery. BAR-encapsulated species, emergence of fungal opportunistic infections or PLGA NPs were characterized and evaluated in two-spe- Pseudomonas infection, and allergic reactions. Last, most cies bioflm inhibition and disruption models. In addition, current antibiotics have difculty penetrating periodon- the kinetics of BAR-encapsulated, relative to BAR surface- tal bioflms, and must be frequently administered, due to modifed NPs were assessed in a two-species model. their transient activity in the oral cavity [6–8]. Porphyromonas gingivalis has been found to be asso- Methods ciated with chronic periodontitis in 88% of sub-gingival Peptide synthesis plaque samples [9]. Moreover, P. gingivalis and S. gordo- BAR peptide is comprised of residues 1167 to 1193 of nii association enhances the disruption of host–microbe the SspB (Antigen I/II) protein sequence of S. gordonii homeostasis and induces population changes in the sub- ­(NH2-LEAAPKKVQDLLKKANITVKGAFQLFS -COOH) gingival bioflm, driving infammatory periodontal dis- [16]. To enable peptide quantifcation and detection, the eases [10–12]. Previous work in our group has shown epsilon amine of the underlined lysine residue of BAR was that P. gingivalis adherence to streptococci is driven by covalently reacted with 6-carboxyfuorescein to produce the interaction of the minor fmbrial antigen (Mfa) of P. fuorescent BAR (F-BAR). Both unlabeled and labeled pep- gingivalis and the streptococcal antigen I/II (AgI/II) [13, tides were synthesized by BioSynthesis, Inc. (Lewisville, 14]. From these studies, a peptide (designated BAR), was TX) and obtained with greater than 90% purity. developed that potently inhibits P. gingivalis/S. gordonii adherence in vitro and reduces P. gingivalis virulence in a BAR‑encapsulated and BAR surface‑modifed nanoparticle mouse model of periodontitis [15–17]. While efcacious, synthesis one of the challenges to free BAR administration is that BAR and F-BAR encapsulated poly(lactic-co-glycolic acid) it provides relatively transient inhibition of P. gingivalis PLGA and methoxy-polyethylene glycol (mPEG-PLGA) in the oral cavity. Moreover, to treat established bioflms, NPs were synthesized using a double emulsion technique relative to initial bioflm formation, higher concentra- [25, 26]. Briefy, BAR was encapsulated in PLGA car- tions of BAR are required. boxyl-terminated polymer (0.55–0.75 dL/g; LACTEL­ ®; Polymeric delivery vehicles provide one option to DURECT Corporation, Cupertino, CA, USA) or mPEG- address these challenges, by ofering prolonged and tar- PLGA (Mw ~ 5000:55,000 Da; ­PolySciTech®; Akina, Inc., geted delivery of active agents. In particular, for applica- IN, USA). One hundred milligrams of PLGA or mPEG- tion to the oral cavity, polymeric nanoparticles (NPs) are PLGA was dissolved in 2 mL methylene chloride (DCM) easy to fabricate and produce stable formulations. From overnight. Te next day, BAR was dissolved in 200 μL Tris a delivery perspective, polymeric NPs may ofer rapid EDTA (TE) bufer at a concentration of 43 μg BAR/mg degradation in the acidic environment of the oral cav- PLGA. Te resulting PLGA/DCM solution was vortexed ity, while providing mucoadhesive properties due to the while adding 200 μL of BAR peptide solution dropwise, electrostatic interactions between NPs and gingival epi- and the mixture was ultrasonicated. Next, 2 mL of the thelium [18–20]. Furthermore, for more labile molecules PLGA/DCM/BAR solution was added dropwise to 2 mL like biologics, polymers have the potential to protect of 5% (w/v) polyvinyl alcohol (PVA) while vortexing and the functionality of the active agent and provide tunable was subsequently sonicated. Te NP solution was added release and prolonged delivery, while enabling localiza- to 50 mL of 0.3% PVA for 3 h to evaporate residual DCM. tion of the active agent to target sites [19, 21]. In addition After evaporation, the NP solution was centrifuged at polymeric NPs may ofer a safer and more biocompatible 13,000 rpm at 4 °C and washed with distilled water twice. delivery method, relative to currently applied metallic F-BAR encapsulated NPs were synthesized similarly, but NPs that exhibit broad antimicrobial efect [22, 23]. were protected from light to avoid photobleaching. Previous work in our groups has demonstrated that NPs BAR surface-modifed NPs were synthesized similarly surface-modifed with BAR peptide more potently inhibit as above using a previously described double emulsion P. gingivalis adherence to S. gordonii, relative to an equi- technique [26–29]. Briefy, the 5% (w/v) polyvinyl alcohol molar administration of free BAR peptide in vitro [24]. (PVA) solution was mixed with 2 mL of 5 mg/mL avidin- Tis increased potency was attributed to a higher local- palmitate and the 2 mL PLGA/DCM solution was added ized dose of BAR, facilitating multivalent interactions dropwise to 4 mL PVA/avidin-palmitate while vortex- with P. gingivalis. While surface-modifed NPs provide ing. After the frst wash, the supernatant was discarded targeting efcacy, a method of delivering high concen- and the pelleted NPs were resuspended in 10 mL PBS trations of BAR for prolonged duration has not been for 30 min on a benchtop rotator, with biotinylated BAR Mahmoud et al. J Nanobiotechnol (2018) 16:69 Page 3 of 12

peptide at a molar ratio of 3:1 BAR:avidin (18.5 nmol/ cell culture plates were wrapped in aluminum foil to pro- mg) in PBS. After conjugation, the NPs were washed two tect the labeled cells from light and placed on a rocker times with distilled water by centrifugation at 13,000 rpm platform in the anaerobic chamber for 24 h. at 4 °C. After washing, both BAR-encapsulated and BAR Porphyromonas gingivalis cultures were optimized surface-modifed NPs, were suspended in 5 mL of dis- using a similar approach, utilizing a diferent fuorescent tilled water, frozen at − 80 °C, and lyophilized. label (20 μL of 4 mg/mL carboxyfuorescein–succiny- lester). P. gingivalis was incubated with the fuorescent NP characterization: NP morphology, size, BAR loading, dye for 30 min on a rocker platform and protected from controlled release light. Te same procedures were followed as performed with S. gordonii to determine cell concentration, with Unhydrated NP morphology, diameter, and size distri- slight adaptations. Te optical density of P. gingivalis was bution were determined by analyzing scanning electron adjusted from 0.8 to 0.4 O.D. (5 107 CFU/mL) by dilut- microscopy (SEM) images with NIH ImageJ software × ing P. gingivalis cultures with an equal volume of BAR (version 1.5a, imageJ.nih.gov). Dynamic light scattering NPs or free BAR. Te fnal concentration of BAR NPs or and zeta potential analyses were performed on hydrated free BAR ranged from 0.3 to 3 μM based on the previ- NPs to determine the hydrodynamic diameter and sur- ously determined IC50 of free BAR (1.3 μM). P. gingivalis face charge (Malvern, Malvern, UK). To determine BAR was incubated with BAR NPs or free BAR at 25 °C for loading and encapsulation efciency (EE), NPs were dis- 30 min before transferring to wells containing S. gordonii. solved in dimethyl sulfoxide (DMSO). Te quantity of Plates containing P. gingivalis and S. gordonii were sub- extracted F-BAR was determined by measuring fuores- sequently incubated for 24 h at 37 °C in anaerobic condi- cence (488/518 nm excitation/emission). For BAR-encap- tions [24]. Te following day, the supernatant was removed sulated NPs, in vitro release was measured by gentle and cells were washed with PBS. Adherent cells were fxed agitation of NPs in phosphate bufered saline (PBS, pH with 4% (w/v) paraformaldehyde and the cover glass was 7.4) at 37 °C. At fxed time points (1, 2, 4, 8, 24, 48 h), mounted on a glass slide. Bioflms were visualized using a samples were collected and the amount of BAR released Leica SP8 confocal microscope (Leica Microsystems Inc., from the NPs was quantifed as described above. Bufalo Grove, IL) under 60× magnifcation. Background noise was minimized using software provided with the Growth of bacterial strains Leica SP8 and three-dimensional z-stack bioflm images Porphyromonas gingivalis ATCC 33277 was grown in were obtained from 30 randomly chosen frames using a Trypticase soy broth (Difco Laboratories Inc., Livonia, z-step size of 0.7 μm. Images were analyzed with Voloc- MI, USA) supplemented with 0.5% (w/v) yeast extract, ity image analysis software (version 6.3; Perkin Elmer, 1 μg/mL menadione, and 5 μg/mL hemin. Te medium Waltham, MA, USA) to determine the ratio of green to was reduced for 24 h under anaerobic conditions (10% red fuorescence (GR), representing P. gingivalis and S. gor- CO­ 2, 10% H­ 2, and 80% N­ 2) and P. gingivalis was subse- donii, respectively. Control samples were used to subtract quently inoculated and grown anaerobically for 48 h at background levels of auto-fuorescence. Briefy, triplicate 37 °C. S. gordonii DL-1 was cultured aerobically without samples of S. gordonii alone were immobilized without shaking in brain–heart infusion broth (Difco Laboratories P.g or BAR in 12-well culture plates and the same proce- Inc.) supplemented with 1% yeast extract for 16 h at 37 °C. dures for dual-species bioflm were followed. S. gordonii- only coverslips were visualized and images were analyzed Bioflm inhibition assay using the previously mentioned approach. GR background To assess the efectiveness of BAR-encapsulated NPs to was subtracted using the following formula: GR sample or prevent the interaction of P. gingivalis with S. gordonii, control − GR S. gordonii-only. Each treatment group (BAR S. gordonii was harvested from culture and labeled with NPs or free BAR) was analyzed in triplicate and three inde- 20 μL of 5 mg/mL hexidium iodide for 15 min at room pendent frames were measured for each well. Te mean temperature. Following incubation, cells were centri- and variation (SD) between samples were determined fuged to remove unbound fuorescent dye. Subsequently, using analysis of variance (ANOVA) and diferences were the bacterial concentration was measured by the O.D. considered to be statistically signifcant when p < 0.05. Te at 600 nm from 20-fold diluted cultures of S. gordonii. percent inhibition of P. gingivalis adherence was calculated Te optical density of S. gordonii cells was adjusted to with the following formula: GR sample/GR control. 9 0.8 (1 × 10 CFU/mL) to obtain uniformity between cell counts in each well. After adjusting the optical density, Bioflm disruption assay 1 mL of S. gordonii cells was added to each well of 12-well Te same procedures utilized in the inhibition assay were culture plates containing a sterilized micro-coverslip. Te followed, except P. gingivalis was allowed to adhere to Mahmoud et al. J Nanobiotechnol (2018) 16:69 Page 4 of 12

streptococci in the absence of BAR peptide or BAR NPs and 3 μM) were applied to the bioflms. Te bioflms were to demonstrate the ability of BAR-encapsulated NPs to assessed 1, 2, and 3 h post-administration and visualized disrupt or “treat” pre-established bioflms. Te resulting as described above. P. gingivalis/S. gordonii bioflms were then treated for 3 h with free BAR or BAR-encapsulated NPs at various Results concentrations and processed and analyzed as described Nanoparticle characterization above. Te morphology, size, and zeta potential of BAR PLGA Inhibitory kinetics of BAR released from BAR‑encapsulated and mPEG-PLGA NPs were determined. Te morpholo- NPs gies of BAR-encapsulated PLGA and mPEG-PLGA NPs Due to the similar release properties of BAR from PLGA are shown in Fig. 1. Both PLGA and mPEG-PLGA NPs and mPEG-PLGA NPs, PLGA NPs were selected to fur- demonstrated spherical morphology with average unhy- ther assess the ability of NPs to release therapeutically drated diameters of 227.5 ± 23.0 nm and 243.1 ± 31.2 nm relevant concentrations of BAR at diferent time points. respectively (Table 1). In comparison, the average PLGA BAR NPs (1.3 μM) were incubated with gentle agi- hydrated diameters of PLGA and mPEG-PLGA NPs tation in PBS (pH 7.4) at 37 °C. After 1, 2, 4 and 8 h, the were 234.4 ± 19.2 nm and 278.9 ± 13.8 nm, respectively. NP suspension was centrifuged, and the supernatant was PLGA and mPEG-PLGA NPs had zeta potentials of collected for bioflm experiments. Te NPs were re-sus- − 13.1 ± 0.4 mV and − 5.9 ± 0.1 mV. pended with new PBS. P. gingivalis was incubated with BAR NP eluate for 30 min, and subsequently transferred Quantifcation of BAR loading and release to a well containing an S. gordonii bioflm. Te same bio- Te loading of BAR peptide in PLGA and mPEG-PLGA flm inhibition assay procedure detailed above was used NPs was determined using fuorescence spectroscopy, to visualize and analyze the samples. and the fuorescence was compared to a known standard Time‑dependent comparison between free BAR, of F-BAR. Loading experiments demonstrated that both BAR‑encapsulated, and BAR surface‑modifed NPs PLGA and mPEG-PLGA NPs highly encapsulated BAR with 19.0 0.1 and 16.1 0.2 µg of BAR per mg of NP, In addition to delivering high concentrations of BAR dur- ± ± ing the time frame of interest, the temporal evaluation of BAR activity against established bioflms was evaluated Table 1 Physical characterization of NPs and compared. Both BAR-encapsulated and BAR surface- NP type Unhydrated Hydrated Zeta potential (mV) modifed NPs were assessed due to their previously dem- diameter diameter (nm) (nm) onstrated efcacy. P. gingivalis was allowed to adhere to streptococci in the absence of peptide, then BAR (3 μM), PLGA NPs 227.5 23.0 234.4 19.2 13.1 0.4 ± ± − ± BAR-encapsulated, and BAR surface-modifed NPs (1.3 mPEG-PLGA NPs 243.1 31.2 278.9 13.8 5.9 0.1 ± ± − ±

Fig. 1 SEM images of BAR-encapsulated a PLGA NPs and b mPEG-PLGA NPs. Scale bars represent 1 μm Mahmoud et al. J Nanobiotechnol (2018) 16:69 Page 5 of 12

Table 2 The amount of BAR (μg) loaded in PLGA respectively, corresponding to encapsulation efciencies and mPEG-PLGA NPs (mg) of 44 and 37% (Table 2). NP type BAR input BAR output (μg/mg) Encapsulation To assess BAR release from the NPs, the fuorescence (μg/mg) efciency (%) of supernatant from 1, 2, 4, 8, 24, 48 h release time points PLGA NPs 43 19.0 0.1 44.2 was measured and compared to a known standard of ± mPEG-PLGA NPs 43 16.1 0.2 37.3 F-BAR in PBS. Release experiments demonstrated that ± 47% of encapsulated BAR (10.3 μg/mg) was released from PLGA NPs, while 56% of BAR (9.9 μg/mg) was released from mPEG-PLGA NPs within 24 h (Fig. 2).

Inhibition (or prevention) of P. gingivalis/S. gordonii bioflm formation BAR-encapsulated PLGA and mPEG-PLGA NPs were functionally evaluated to determine their potential to inhibit P. gingivalis adherence to S. gordonii after 24 h, relative to free BAR. As shown in Fig. 3 and Additional fle 1, P. gingivalis adherence was signifcantly reduced in the presence of BAR-encapsulated PLGA and mPEG- PLGA NPs. Adherence was inhibited by 39% at the low- est administered concentration (0.3 μM), 59% at 0.7 μM, Fig. 2 Cumulative release of BAR as a function of mass (μg BAR and reached maximum inhibition (94%) at the high- per mg NP, open symbols) and percent of total BAR loaded (closed est concentration of PLGA NPs tested (3 μM). Simi- symbols) over 48 h lar inhibitory results were observed for mPEG-PLGA

Fig. 3 BAR-encapsulated PLGA NPs prevent P. gingivalis adherence to S. gordonii. Bioflms were visualized with confocal microscopy and the ratio of green (P. gingivalis) to red (S. gordonii) fuorescence in z-stack images was determined using Volocity image analysis software. Each grid represents 21 μm Mahmoud et al. J Nanobiotechnol (2018) 16:69 Page 6 of 12

NPs, where P.g/S.g bioflm formation was inhibited by disrupted pre-existing dual-species bioflms by ~ 25% 37, 55, and 92% at concentrations of 0.3 μM, 0.7 μM with the lowest administered concentration (0.3 μM), and 3 μM respectively. Te ability of BAR-encapsulated 40% with 0.7 μM, and 85% with 3 μM of BAR-encapsu- NPs to inhibit bioflm formation was dose-dependent lated PLGA NPs. Similar trends were observed for the (IC50 = 0.7 μM) with no statistically signifcant difer- disruption of pre-existing bioflms with 0.3, 0.7, and ences between PGLA and mPEG-PLGA BAR-encap- 3 μM mPEG-PLGA NPs (20%, 38%, and 80% disrupted). sulated NPs (p > 0.05). Moreover these results indicate Overall the IC50 values of PLGA and mPEG-PLGA that a lower concentration of BAR is required if incor- (~ 1.3 μM) NPs for bioflm disruption were not statisti- porated within NPs, relative to free BAR administration cally diferent (p > 0.05, Fig. 5b); demonstrating statisti- (IC50 = 1.3 μM) (Figs. 3 and 5a). cally signifcant improvements in efcacy relative to free BAR (p < 0.05). Disruption (or treatment) of P. gingivalis/S. gordonii bioflms Inhibitory activity of BAR released from BAR‑encapsulated To determine whether BAR peptide is capable of dis- NPs rupting pre-existing P. gingivalis/S. gordonii bioflms, To determine the inhibitory potential of BAR-encap- dual-species bioflms were formed in PBS in the absence sulated NPs, as a function of release duration, strep- of BAR peptide for 24 h, and were subsequently incu- tococcal cells were immobilized and P. gingivalis was bated for 3 h with BAR-encapsulated PLGA or mPEG- incubated with eluate released from 1.3 μM BAR- PLGA NPs. Various molar concentrations of BAR NPs encapsulated PLGA NPs at 1, 2, 4, and 8 h. BAR-encap- ranging from 0.3 to 3 μM were tested. Te bioflms were sulated PLGA NPs were selected due to their similar visualized and the percent inhibition was calculated release and inhibitory properties, relative to mPEG- as described above. As shown in Fig. 4 and Additional PLGA NPs. As shown in Fig. 6, BAR released during fle 2, BAR-encapsulated PLGA and mPEG-PLGA NPs the first 2 h, potently inhibited biofilm formation (68%

Fig. 4 BAR-encapsulated PLGA NPs disrupt pre-established P. gingivalis–S. gordonii bioflms. Bioflms were visualized with confocal microscopy and the ratio of green (P. gingivalis) to red (S. gordonii) fuorescence in z-stack images was determined using Volocity image analysis software. Each grid represents 21 μm Mahmoud et al. J Nanobiotechnol (2018) 16:69 Page 7 of 12

Fig. 5 Comparison of the concentration of BAR-encapsulated PLGA and mPEG-PLGA NPs needed to a inhibit or b disrupt P. gingivalis/S. gordonii bioflms

Fig. 6 BAR-encapsulated PLGA NPs inhibit P. gingivalis adherence to S. gordonii after diferent durations of release. Bioflms were visualized with confocal microscopy and the ratio of green (P. gingivalis) to red (S. gordonii) fuorescence in z-stack images was determined using Volocity software. Each grid 21 μm =

and 32%, respectively), whereas BAR released after 4 Time‑dependent comparison of free BAR, and 8 h provided less potent inhibition of biofilm for- BAR‑encapsulated, and BAR surface‑modifed NP bioflm mation (25% for both time points). These results indi- disruption cate that BAR-encapsulated NPs release an inhibitory Previous studies demonstrated that BAR surface- dose of peptide for at least 2 h. modifed PLGA NPs potently disrupt pre-established Mahmoud et al. J Nanobiotechnol (2018) 16:69 Page 8 of 12

P. gingivalis/S. gordonii bioflms [24]. To compare the niche, and is thus an ideal target for therapeutic interven- temporal efect resulting from the administration of the tion [5]. Previous studies have shown that BAR peptide newly formulated BAR-encapsulated NPs, relative to free inhibits bioflm formation by P. gingivalis and S. gordo- BAR or previously tested BAR surface-modifed NPs, two nii in vitro and reduces the virulence of P. gingivalis in concentrations of BAR-encapsulated and BAR surface- a murine model of infection [15–17]. While efcacious, modifed PLGA NPs were compared with free BAR after BAR efectiveness was limited by the duration of expo- 1, 2, and 3 h administration to pre-established bioflms. sure within the oral cavity, and necessitated a higher As shown in Table 3 and Fig. 7, free BAR (3 μM) mini- concentration to disrupt previously established bioflms mally disrupted pre-existing bioflms during the frst [15–17]. In previous work we sought to address these hour of application (23%), and demonstrated a slight challenges by synthesizing BAR surface-modifed NPs increase in disruption after 2 h (44%). After 3 h, free to multivalently inhibit bioflm formation [24]. Te goal BAR (3 μM) disrupted 69% of the pre-existing bioflm. of this study was to develop, characterize, and compare In comparison, administration of the same equimolar BAR-encapsulated NPs that release BAR within a time concentration of BAR-encapsulated NPs (1.3 and 3 μM) frame relevant to delivery in the oral cavity. disrupted the established bioflm during the frst hour of Nanoparticle characterization revealed that PLGA exposure by 32% and 38%, respectively and demonstrated and mPEG-PLGA BAR-encapsulated NPs exhibited even more potent disruption (47% and 52%) after 2 h. spherical morphologies and average particle diameters of Te maximum disruption for 1.3 and 3 μM doses (66% 234.4 ± 19.2 nm and 278.9 ± 13.8 nm, with respective zeta and 77%, respectively) was achieved after 3 h exposure to potentials of − 13.1 ± 0.4 mV and − 5.9 ± 0.1 mV. Tese bioflms. Comparatively, both 1.3 and 3 μM BAR surface- values are in agreement with expected values for these modifed NPs disrupted pre-existing bioflms within 1 h polymeric NPs [24–26, 28, 29]. Both PLGA and mPEG- by 43% and 49%, respectively, and induced more potent PLGA NPs were synthesized with 43 μg of BAR per mg bioflm disruption (59% and 69%) after 2 h exposure, NP, corresponding to loading concentrations deemed fea- demonstrating statistically signifcant disruption, rela- sible for bioflm inhibition with free BAR [15–17]. PLGA tive to disruption induced by free BAR peptide. Te high- and mPEG-PLGA NPs demonstrated relatively high pep- est levels of disruption (71% and 83% respectively) were tide loading with 19.0 ± 0.1 and 16.1 ± 0.2 µg BAR per mg achieved after 3 h BAR surface-modifed NP administra- of NP respectively. tion. Overall, BAR surface-modifed NPs were statisti- In addition to high loading, PLGA and mPEG-PLGA cally more efective than free BAR (p < 0.05) in disrupting NPs released 40% and 48% of BAR within the frst 4 h, established bioflms after 1, 2, and 3 h administration. with no statistically signifcant diferences between release However, no statistical diferences were observed for profles. Te NP formulations were designed to achieve BAR-encapsulated NPs (p > 0.05), relative to BAR sur- therapeutic concentrations of BAR in the oral cavity for face-modifed NPs or free BAR peptide after 1, 2, or 3 h a minimum of 2 h. Tis initial window of 2 h release was administrations. targeted as we envision formulating NPs in a mouth rinse or toothpaste product. Ideally, in future formulations, we Discussion seek to tailor the release of peptide for up to 12 h since we Porphyromonas gingivalis has been identifed as a envision these formulations may be applied once or twice “keystone” pathogen involved in the initiation and daily, to exert immediate efect over a number of hours. progression of periodontal infammatory disease, by dis- To assess the functionality of BAR-NPs, the inhibi- rupting host-microbe homeostasis and inducing popula- tion and disruption concentrations of BAR-encapsulated tion changes in the subgingival bioflm. Tis disruption PLGA and mPEG-PLGA NPs were determined against and colonization is initially prompted by the association dual-species bioflms. As shown in Figs. 3 and 4, BAR of P. gingivalis with oral streptococci in the supragingival NPs demonstrated potent inhibition and disruption with

Table 3 Percent disruption of pre-existing bioflms with diferent treatment groups and concentrations Time (h) % Disruption of pre-formed bioflms Free BAR (3 μM) BAR-mod NPs BAR-mod NPs (3 μM) BAR-encap NPs BAR-encap (1.3 μM) (1.3 μM) NPs (3 μM)

1 22.6 0.2 43.4 0.2 48.9 0.1 32.3 0.1 37.7 0.1 ± ± ± ± ± 2 44.4 0.2 59.2 0.1 68.7 0.1 46.6 0.2 52.4 0.2 ± ± ± ± ± 3 69.0 0.0 71.2 0.1 83.4 0.0 66.1 0.1 77.0 0.0 ± ± ± ± ± Mahmoud et al. J Nanobiotechnol (2018) 16:69 Page 9 of 12

Fig. 7 Disruption of established P. gingivalis/S. gordonii bioflms after diferent exposure times to BAR surface-modifed NPs, BAR-encapsulated NPs and free BAR. Bioflms were visualized with confocal microscopy and the ratio of green (P. gingivalis) to red (S. gordonii) fuorescence in z-stack images was determined using Volocity software. Each grid 21 μm =

4 h of administration (Fig. 6). Moreover, the temporal IC50 s = 0.7 μM and 1.3 μM, respectively, with negli- gible diferences observed between PLGA and mPEG- dependence of free BAR, BAR-encapsulated, and BAR PLGA NPs (Additional fle 1 and Additional fle 2). To surface-modifed NPs to disrupt pre-established bioflms explore the temporal efect of BAR released from PLGA (treatment) was measured after 1, 2, and 3 h applica- NPs on bioflm inhibition (prevention) in greater depth, tion. As shown in Fig. 7 and Table 3, BAR-encapsulated the efcacy of BAR-encapsulated NPs was assessed in and BAR surface-modifed NPs achieved moderate bio- a dual-species bioflm after 1, 2, 4 and 8 h post-appli- flm inhibition within 1 h in a dose-dependent manner; cation. Sufcient BAR release was achieved, relating however, similar concentrations of free BAR required to inhibitory concentrations of 1.3 μM during the frst prolonged exposure of up to 3 h to achieve more potent Mahmoud et al. J Nanobiotechnol (2018) 16:69 Page 10 of 12

efect. Tese results demonstrate that BAR-encapsulated against S. gordonii. Although this study demonstrated NPs provide a feasible alternative to free BAR and BAR promise utilizing S. gordonii as a live oral vaccine, to surface-modifed NPs to target dual-species oral bioflms date there are few formulations available to localize or and provide rapid onset of action. Together, these stud- sustain biologic administration to the oral cavity [40]. ies indicate that BAR-encapsulated NPs may serve as a In other work, PLGA NPs encapsulating a novel anti- short-term delivery formulation to enhance BAR deliv- infammatory agent (15d-PGJ2), demonstrated promise ery and potency in the oral cavity. Moreover, by encap- in reducing infammatory response and bone resorption sulating versus surface-modifying NPs with BAR, these in mouse model of periodontitis after daily administra- NPs may ofer the potential to specifcally target NPs tion [44], demonstrating the feasibility of combined bio- with modifcations that can complement BAR activity logic and delivery vehicle against oral pathogens. Despite to engage with these or other bacterial species in future this recent progress in the delivery of biological agents work. for oral applications, currently few biological agents in To date, a variety of polymeric nanoparticle formula- combination with delivery vehicles have been developed tions have been developed for oral delivery; however, to inhibit keystone-specifc interactions during the initial these vehicles have primarily focused on the delivery of stages of periodontal disease [24]. non-specifc active agents such as antibiotics [30–39]. In addition to progress in the development of vehicles Antibiotics such as chlorhexidine [30, 31], minocycline to encapsulate antibiotic and biological agents in poly- [32, 33], clarithromycin [36], vancomycin [34], doxy- meric delivery vehicles, polymeric platforms have also cycline [37], and tetracycline [35, 38, 39] are among the been surface-modifed with a variety of molecules includ- antibiotics that have been incorporated into a variety of ing RGD [33], chitosan [31, 36], tertiary amines bearing polymeric vehicles [30–37] to provide sustained-delivery, two t-cinnamaldehyde substituents [45], dimethyl-octyl prolong activity, exert antibacterial activity, and decrease ammonium [45], and BAR peptide [24] to increase the antibiotic cytotoxicity [30–37]. Yet, despite antibiotic mucoadhesivity (and in the latter case, specifcity) of oral choice, primary concerns of antibacterial resistance and delivery formulations. A variety of polymers have been cytotoxicity remain [30, 31, 35]. While chitosan and modifed with biological ligands to impart enhanced PLGA NPs that encapsulated chlorhexidine dihydro- therapeutic efect [24, 33]. As one example, the deliv- chloride (CHX) demonstrated strong adherence to tooth ery of antibiotic minocycline-loaded poly(ethylene gly- surfaces and sustained-release for 48 h in neutral pH con- col)–poly(lactic acid) (PEG–PLA) nanoparticles have ditions, moderate cytotoxicity due to CHX was observed targeted oral epithelial cells by surface-modifcation in human gingival fbroblasts [31]. Similar studies seeking with RGD peptides. Surface-modifcation of PEG–PLA to ameliorate periodontal infection caused by A. actino- NPs increased epithelial cell attachment and maintained mycetemcomitans and P. nigrescens with PLGA lovas- efective drug concentrations in gingival fuid for more tatin-chitosan-tetracycline NPs demonstrated potent than 2 weeks in vivo, relative to unmodifed minocycline inhibition up to 1 week after administration. However, NPs. Similarly, chitosan-modifed polyvinyl caprolactam- signifcantly elevated alkaline phosphatase was observed polyvinyl acetate-polyethylene glycol graft copolymer in cells treated with 0.1% or 0.3% tetracycline-loaded (Soluplus) and poly-(dl-lactide-co-glycolide) nanopar- nanoparticles on days 7 and 9 [35]. Overall, these stud- ticles loaded with clarithromycin, increased antibacte- ies have shown that delivery vehicles have the potential rial efcacy and provided sustained-release against oral to increase antibiotic efectiveness by decreasing the con- bioflms [36]. Although this study demonstrated efective centration required. However, bacterial resistance, non- treatment of periodontitis, the limitations of antibiotic specifc targeting, and cytotoxicity concerns with chronic delivery still pose challenges [33]. Surface modifcation use suggest that the development of more specifcally of nanoparticles has imparted new attributes to tar- acting active agents will ofer safer alternatives for bioflm get active agents to oral-specifc niches. We expect that inhibition. combining our current work, with surface functionaliza- More recently, specifcally targeted biological agents tion demonstrated in our previous study [24], may con- have been investigated to treat periodontal diseases. fer additional advantages in targeting keystone species by Delivery of thyA gene [40], Punica granatum extract [41], providing prevention and treatment via adhesion and a H. madagascariensis leaf extract [42], miR-146a [43], localized release-mediated platform. and the anti-infammatory agent 15d-PGJ2 [44] have Taken together, our results demonstrate that BAR- been investigated to vaccinate against and target peri- encapsulated NPs achieve more potent in inhibition odontal diseases. Recent work assessed the delivery of an and disruption than equimolar free BAR administra- oral vaccine comprised of an auxotrophic complemen- tion. We believe that incorporation of BAR peptide in tation of the thyA gene to produce an immune response NPs provides gradual release of BAR peptide, while Mahmoud et al. J Nanobiotechnol (2018) 16:69 Page 11 of 12

BAR-modifcation ofers a platform to provide a higher Availability of data and materials The datasets used and/or analyzed during the current study are available from localized concentration of BAR in the oral cavity via the corresponding author on reasonable request. All data generated or ana‑ multivalent interactions. BAR-encapsulated NPs ofer lyzed during this study are included in this published article and its additional a platform to improve efcacy, and potentially longev- information fles. ity in the oral cavity compared to the transient activity Consent for publication of free BAR. Tese experimental results will be helpful Not applicable. in developing NPs in therapeutic formulations such as Ethics approval and consent to participate toothpaste, mouth rinse or chewing gum. Future stud- Not applicable. ies may focus on developing blended polymeric NPs to more gradually release inhibitory concentrations Funding We gratefully acknowledge support from grants R01DE023206 (DRD) and for 8–12 h. Moreover, combining this platform with R21DE025345 (DRD and JMSR) from the National Institute for Dental and surface functionality to provide mucoadhesive or spe- Craniofacial Research. cifc interactions with gingival tissue may be pursued to enhance the targeting potential. Ongoing and future Publisher’s Note work in our groups seeks to assess the efcacy of both Springer Nature remains neutral with regard to jurisdictional claims in pub‑ lished maps and institutional afliations. BAR-modifed and BAR-encapsulated NPs in a murine model of periodontitis. Received: 26 April 2018 Accepted: 5 September 2018

Additional fles References 1. Hussain M, Stover CM, Dupont AP. gingivalis in periodontal disease and atherosclerosis—scenes of action for antimicrobial peptides and comple‑ Additional fle 1. BAR-encapsulated mPEG-PLGA NPs prevent P. gingivalis ment. Front Immunol. 2015;6:45. adherence to S. gordonii. Bioflms were visualized with confocal micros‑ 2. Genco RJ, Van Dyke TE. Prevention: reducing the risk of CVD in patients copy and the ratio of green (P. gingivalis) to red (S. gordonii) fuorescence with periodontitis. Nat Rev Cardiol. 2010;7(9):479–80. in z-stack images was determined using Volocity image analysis software. 3. Lundberg K, Wegner N, Yucel-Lindberg T, Venables PJ. Periodonti‑ Each grid 21 μm. tis in RA-the citrullinated enolase connection. Nat Rev Rheumatol. = Additional fle 2. BAR-encapsulated mPEG-PLGA NPs disrupt pre- 2010;6(12):727–30. established P. gingivalis–S. gordonii bioflms. Bioflms were visualized with 4. Ellen RP. Periodontal care for community-dwelling older adults. J Prosthet confocal microscopy and the ratio of green (P. gingivalis) to red (S. gordonii) Dent. 1994;72(5):500–6. fuorescence in z-stack images was determined using Volocity image 5. Kaldahl WB, Kalkwarf KL, Patil KD, Molvar MP, Dyer JK. Long-term evalu‑ analysis software. Each grid 21 μm. ation of periodontal therapy: I. Response to 4 therapeutic modalities. J = Periodontol. 1996;67(2):93–102. 6. Drisko CH. Non-surgical pocket therapy: pharmacotherapeutics. Ann Peri‑ Authors’ contributions odontol. 1996;1(1):491–566. MYM performed the experiments, statistical analysis, and drafted the manu‑ 7. Walker CB. The acquisition of antibiotic resistance in the periodontal script. JMSR conceived of the study, participated in its design and coordina‑ microfora. Periodontology. 1996;2000(10):79–88. tion, and drafted the manuscript. DRD conceived of the study, participated in 8. Allaker RP, Ian Douglas CW. Non-conventional therapeutics for oral infec‑ its design and coordination, and drafted the manuscript. All authors read and tions. Virulence. 2015;6(3):196–207. approved the fnal manuscript. 9. Datta HK, Ng WF, Walker JA, Tuck SP, Varanasi SS. The cell biology of bone metabolism. J Clin Pathol. 2008;61(5):577–87. https​://doi.org/10.1136/ Author details jcp.2007.04886​8. 1 Department of Oral Immunology and Infectious Diseases, University 10. Honda K. Porphyromonas gingivalis sinks teeth into the oral microbiota of Louisville School of Dentistry, 501 S. Preston St, Louisville, KY 40202, USA. and periodontal disease. Cell Host Microbe. 2011;10(5):423–5. 2 Department of Bioengineering, University of Louisville Speed School of Engi‑ 11. Hajishengallis G, Lamont RJ. Beyond the red complex and into more neering, 505 S. Hancock St., Room 623, Louisville, KY 40202, USA. 3 Department complexity: the polymicrobial synergy and dysbiosis (PSD) model of peri‑ of Microbiology and Immunology, University of Louisville School of Medicine, odontal disease etiology. Mol Oral Microbiol. 2012;27(6):409–19. Louisville, KY 40202, USA. 4 Department of Pharmacology and Toxicology, 12. Hajishengallis G, Liang S, Payne MA, et al. Low-abundance bioflm species University of Louisville School of Medicine, Louisville, KY 40202, USA. 5 Center orchestrates infammatory periodontal disease through the commensal for Predictive Medicine, University of Louisville, 505 S. Hancock St, Louisville, KY microbiota and complement. Cell Host Microbe. 2011;10(5):497–506. 40202, USA. 13. Lamont RJ, El-Sabaeny A, Park Y, Cook GS, Costerton JW, Demuth DR. Role of the Streptococcus gordonii SspB protein in the development of Porphy- Acknowledgements romonas gingivalis bioflms on streptococcal substrates. Microbiology. This work was supported by grants R01DE023206 and R21DE025345 from the 2002;148(Pt 6):1627–36. National Institute for Dental and Craniofacial Research. The authors declare no 14. Park Y, Simionato MR, Sekiya K, et al. Short fmbriae of Porphyromonas conficts of interest. We will freely provide reagents and/or data generated in gingivalis and their role in coadhesion with Streptococcus gordonii. Infect this study to academic investigators according to the policies of the University Immun. 2005;73(7):3983–9. of Louisville. An abstract comprised of parts of this study was presented on 15. Daep CA, James DM, Lamont RJ, Demuth DR. Structural characteriza‑ March 24, 2017 at the International Association of Dental Research Meeting in tion of peptide-mediated inhibition of Porphyromonas gingivalis bioflm San Francisco, CA. formation. Infect Immun. 2006;74(10):5756–62. 16. Daep CA, Lamont RJ, Demuth DR. Interaction of Porphyromonas gingivalis Competing interests with oral streptococci requires a motif that resembles the eukaryotic The authors declare they have no competing interests. Mahmoud et al. J Nanobiotechnol (2018) 16:69 Page 12 of 12

nuclear receptor box protein-protein interaction domain. Infect Immun. 33. Yao W, Xu P, Zhao J, et al. RGD functionalized polymeric nanoparticles 2008;76(7):3273–80. targeting periodontitis epithelial cells for the enhanced treatment of peri‑ 17. Daep CA, Novak EA, Lamont RJ, Demuth DR. Structural dissection and odontitis in dogs. J Colloid Interface Sci. 2015;458:14–21. in vivo efectiveness of a peptide inhibitor of Porphyromonas gingivalis 34. Benoit DS, Koo H. Targeted, triggered drug delivery to tumor and bioflm adherence to Streptococcus gordonii. Infect Immun. 2011;79(1):67–74. microenvironments. Nanomedicine. 2016;11(8):873–9. 18. de Freitas LM, Calixto GM, Chorilli M, et al. Polymeric nanoparticle-based 35. Lee BS, Lee CC, Wang YP, et al. Controlled-release of tetracycline and photodynamic therapy for chronic periodontitis in vivo. Int J Mol Sci. lovastatin by poly(d, l-lactide-co-glycolide acid)-chitosan nanopar‑ 2016;17(5):769. ticles enhances periodontal regeneration in dogs. Int J Nanomed. 19. Makadia HK, Siegel SJ. Poly lactic-co-glycolic acid (PLGA) as biodegrad‑ 2016;11:285–97. able controlled drug delivery carrier. Polymers. 2011;3(3):1377–97. 36. Takahashi C, Akachi Y, Ogawa N, et al. Morphological study of efcacy 20. Patel A, Patel M, Yang X, Mitra AK. Recent advances in protein and of clarithromycin-loaded nanocarriers for treatment of bioflm infection peptide drug delivery: a special emphasis on polymeric nanoparticles. disease. Med Mol Morphol. 2017;50(1):9–16. Protein Pept Lett. 2014;21(11):1102–20. 37. Wayakanon K, Thornhill MH, Douglas CW, et al. Polymersome-mediated 21. Singh R, Lillard JW Jr. Nanoparticle-based targeted drug delivery. Exp Mol intracellular delivery of antibiotics to treat Porphyromonas gingivalis- Pathol. 2009;86(3):215–23. infected oral epithelial cells. FASEB J. 2013;27(11):4455–65. 22. de Sousa FF, Ferraz C, Rodrigues LK, Nojosa Jde S, Yamauti M. Nanotech‑ 38. Madhumathi K, Sampath Kumar TS. Regenerative potential and anti- nology in dentistry: drug delivery systems for the control of bioflm- bacterial activity of tetracycline loaded apatitic nanocarriers for the dependent oral diseases. Curr Drug Deliv. 2014;11(6):719–28. treatment of periodontitis. Biomed Mater. 2014;9(3):035002. 23. Elsabahy M, Wooley KL. Design of polymeric nanoparticles for biomedical 39. Sindhura Reddy N, Sowmya S, Bumgardner JD, Chennazhi KP, Biswas delivery applications. Chem Soc Rev. 2012;41(7):2545–61. R, Jayakumar R. Tetracycline nanoparticles loaded calcium sulfate 24. Kalia P, Jain A, Radha Krishnan R, Demuth DR, Steinbach-Rankins JM. composite beads for periodontal management. Biochem Biophys Acta. Peptide-modifed nanoparticles inhibit formation of Porphyromonas 2014;1840(6):2080–90. gingivalis bioflms with Streptococcus gordonii. Int J Nanomed. 40. Lee SF, Hulbah M, Halperin SA. Development of a gene delivery system in 2017;12:4553–62. Streptococcus gordonii using thymidylate synthase as a selection marker. J 25. Steinbach JM. Protein and oligonucleotide delivery systems for vaginal Microbiol Methods. 2016;125:43–8. microbicides against viral STIs. Cell Mol Life Sci CMLS. 2015;72(3):469–503. 41. Tiyaboonchai W, Rodleang I, Ounaroon A. Mucoadhesive polyethylen‑ 26. Steinbach JM, Weller CE, Booth CJ, Saltzman WM. Polymer nanoparticles imine-dextran sulfate nanoparticles containing Punica granatum peel encapsulating siRNA for treatment of HSV-2 genital infection. J Control extract as a novel sustained-release antimicrobial. Pharm Dev Technol. Release. 2012;162(1):102–10. 2015;20(4):426–32. 27. Martin DT, Steinbach JM, Liu J, et al. Surface-modifed nanoparticles 42. Moulari B, Lboutounne H, Chaumont JP, Guillaume Y, Millet J, Pellequer Y. enhance transurothelial penetration and delivery of survivin siRNA in Potentiation of the bactericidal activity of Harungana madagascariensis treating bladder cancer. Mol Cancer Ther. 2014;13(1):71–81. Lam. ex Poir. (Hypericaceae) leaf extract against oral bacteria using poly 28. Sims LB, Curtis LT, Frieboes HB, Steinbach-Rankins JM. Enhanced uptake (d, l-lactide-co-glycolide) nanoparticles: in vitro study. Acta Odontol and transport of PLGA-modifed nanoparticles in cervical cancer. J Nano‑ Scand. 2006;64(3):153–8. biotechnol. 2016;14:33. 43. Liu L, Shu S, Cheung GS, Wei X. Efect of miR-146a/bFGF/PEG-PEI nano‑ 29. Steinbach JM, Seo YE, Saltzman WM. Cell penetrating peptide-modifed particles on infammation response and tissue regeneration of human poly(lactic-co-glycolic acid) nanoparticles with enhanced cell internaliza‑ dental pulp cells. Biomed Res Int. 2016;2016:3892685. tion. Acta Biomater. 2016;30:49–61. 44. Napimoga MH, da Silva CA, Carregaro V, et al. Exogenous administration 30. Cai X, Han B, Liu Y, Tian F, Liang F, Wang X. Chlorhexidine-loaded amor‑ of 15d-PGJ2-loaded nanocapsules inhibits bone resorption in a mouse phous calcium phosphate nanoparticles for inhibiting degradation and periodontitis model. J Immunol. 2012;189(2):1043–52. inducing mineralization of type I collagen. ACS Appl Mater Interfaces. 45. Zaltsman N, Ionescu AC, Weiss EI, Brambilla E, Beyth S, Beyth N. Surface- 2017;9(15):12949–58. modifed nanoparticles as anti-bioflm fller for dental polymers. PLoS 31. Chronopoulou L, Nocca G, Castagnola M, et al. Chitosan based nano‑ ONE. 2017;12(12):e0189397. particles functionalized with peptidomimetic derivatives for oral drug delivery. New Biotechnol. 2016;33(1):23–31. 32. Yao W, Xu P, Pang Z, et al. Local delivery of minocycline-loaded PEG–PLA nanoparticles for the enhanced treatment of periodontitis in dogs. Int J Nanomed. 2014;9:3963–70.

Ready to submit your research ? Choose BMC and benefit from:

• fast, convenient online submission • thorough peer review by experienced researchers in your field • rapid publication on acceptance • support for research data, including large and complex data types • gold Open Access which fosters wider collaboration and increased citations • maximum visibility for your research: over 100M website views per year

At BMC, research is always in progress.

Learn more biomedcentral.com/submissions