Engineering of bioactive nanocomplexes on dental floss for targeted gingival therapy

Abstract Periodontitis induced by chronic subgingival infection is a ubiquitous disease that causes systemic inflammatory consequences and poses a negative impact on quality of life. The disease is treated and potentially prevented by patient's self‐care aimed at eliminating the oral pathogens from the region. Currently available products for interdental self‐care, including dental floss and interdental brush, have limited ability to prevent the disease. Here, we report a coated dental floss thread, termed “nanofloss,” which uses polyphenol‐based nanocoating to functionalize the floss thread with therapeutic agents. Multiple therapeutics can be integrated into the nanofloss including antibacterial small molecules and proteins. Flossing with nanofloss‐delivered therapeutic agents to the challenging subgingival region with long‐term retention even against the flushing action of the oral fluid in vivo. Our in vitro and in vivo studies demonstrate that chlorhexidine gluconate‐loaded nanofloss effectively treats the subgingival infection by Porphyromonas gingivalis. Collectively, the nanofloss offers a promising and easily usable tool for targeted self‐care of subgingival infection against periodontitis.

the plaque is fundamentally important for the prevention and treatment of periodontitis. Moreover, since pathogens easily recolonize the gum pocket, daily plaque removal is required for effective treatment.
The gold standard of daily plaque removal is the combination of brushing and interdental cleaning. 7 Normal toothbrushes have limited access to interdental regions and hence specially designed tools (e.g., interdental brush [IDB] and dental floss) are always required for adequate interdental cleaning. Among these, only IDB had yielded reduction of plaque and gingival inflammation, that too at a modest level. 8 However, most interdental spaces, especially in young adults, are too small to be accessed by IDBs, which significantly hampers their widespread adoption. Although dental floss is commonly used in daily care and is universally accepted, systematic reviews and metaanalyses have concluded that a majority of reported studies fail to demonstrate that flossing is generally effective in reducing plaque and moderating gum inflammation even when combined with toothbrushing. [8][9][10] This originates from the fact that mechanical force is the primary mode of action of the dental floss. Therefore, it is highly desirable and clinically essential to further enhance the functionalities of dental floss beyond the simple mechanical mechanism through the integration of bioactive therapeutic molecules on the surface of floss thread.
Here, we explored polyphenol-based surface functionalization of the floss thread to load a series of bioactive molecules on its surface, referred to as "nanofloss," to achieve targeted and sustained delivery of the active agents into the deep subgingival regions (Figure 1). Ver-

| Fabrication of nanofloss and characterization
The surface of silk thread was coated using a simple dipping process. Commercially available threads were dipped in water and treated with tannic acid (TA), a natural polyphenol, and iron (III).
The assembled metal-polyphenol superstructures on the surface   (Table S1). The coating of the CHG payload on the thread is thin and is not expected to impact the mechanical strength of the thread.

| Delivering small molecules by nanofloss
The ability of nanofloss to deliver agents into the subgingival region was demonstrated using both small and large molecules (Figure 4a).  Clear green line that matched with gingival sulcus, the narrow gap between the tooth and gingiva, confirmed that BSA was successfully delivered throughout the subgingival region by BSA-nanofloss.
Needle-insertion-mediated delivery of BSA into the sulcus faded away within 1 day. However, strong signal was observed in the sulcus after 1 day after flossing with BSA-nanofloss. No fluorescent signal was observed after flossing with BSA soaked naïve silk thread.

| Long retention of agents in subgingival region
Gingival crevicular fluid (GCF), an outflow under the gingival margin, is a major factor that makes it difficult for the administered agents to remain in the subgingival region. 11 However, as shown in Figure 4, both small and large molecules were maintained for prolonged times in the region against the flow. To reveal the underlying mechanisms of prolonged retention, rubbing test of nanofloss was carried out first.
Nanofloss, which is visibly purple, turned white after the rubbing test, while the rubbed site turned purple, thus demonstrating that the polyphenol-based nanocomplexes were transferred from the nanofloss to the application site ( Figure S1A,B). Next, the retention of the polyphenol-based nanocomplexes against the flow was observed by a slide-based test ( Figure S2). A microscope glass slide was rubbed with BSA (AF 488)-nanofloss or BSA soaked naïve silk thread, followed by washing with running water (500 μl, 100μl/s). The flow rate of water was much higher than previously reported values of GCF (0.008 μl/s). 12 The fluorescent signal was well preserved after exposure to running water, demonstrating that polyphenol-mediated nanocomplexes adhered to the slide against the water flow. Taken together, these data support the transfer of polyphenol-based nanocomplexes from the nanofloss to the subgingival region.

| Therapeutic efficacy of CHG-nanofloss
Since bacterial infection is the most important cause of periodontitis, antimicrobial activity of CHG-nanofloss was evaluated in vitro and in vivo. In vitro efficacy was studied using disk diffusion test. Growth inhibition of Porphyromonas gingivalis, the major periodontopathic bacteria, was measured on the agar plate (Figure 5a,b). Nanofloss itself,

| DISCUSSION
Periodontitis requires lifelong maintenance by both patients and dental professionals. 13 Even after receiving active periodontal therapy, residual deep periodontal pockets, the deep subgingival region, represent a significant risk factor for the relapse of inflammation leading to further disease progression. 14,15 Despite substantial research, the current treatment options fail to control the progression of the disease.
In this study, we report a modular strategy of engineering silk thread surface taking advantages of self-assembled polyphenol-based nanocomposites, referred to as nanofloss, for an advanced self-care of subgingival region. This technology enables the delivery and retention of various agents in the subgingival region, and CHG-loaded nanofloss effectively treated the bacteria infection.
Polyphenol-based bioactive nanocomplexes leverage the interactions between polyphenol-metal and polyphenol-agents. This polyphenol-based strategy for surface-functionalization is simple and inexpensive, and thus it has been utilized to coat various substances, including particles and cells, for various applications including drug delivery, wound healing, and imaging [16][17][18][19][20][21] and hence was selected for design of nanofloss. Successful coating of the thread surface opens another interesting application of this technology.
Another valuable feature of the polyphenol-based nanocomplexes described here is that they could be transferred from the thread to the tissue surface. The presence of polyphenol in the nanocomplexes of active agents enabled their retention on the tissue surface.
Dental floss is an ideal tool for targeted drug delivery to the subgingival region for prevention and treatment of periodontitis. Dental floss is already a common tool for personal care and is easily accessible to the broad population. Topical delivery of drugs for the treatment of periodontitis is a major challenge due to limited tissue penetration. Current products intended for periodontitis need to be inserted into periodontal pockets directly by dentists, which requires patients to visit the clinic on a regular basis. 11,22,23 Although dental floss is commonly used for tooth cleaning, little has been published on its use as a topical drug carrier. Kaewiad et al. showed that dental floss impregnated with antibacterial agent inhibit bacterial growth on agar plates, 24

| X-ray photoelectron spectroscopy
XPS measurement was completed on the Nexsa XPS system from Thermo Scientific. The probe for the measurement was aluminum K-α X-ray line with energy at 1.4866 keV and x-ray spot size was set at 150 μm. XPS data were taken after the chamber pressure was sitting at 5 E-8 mBar or lower. Flood gun, which supplies both low-energy electron and ion, was used throughout the entire experiment for sample surface charge compensation. Both survey spectrum (wide) and

| Liquid chromatography/mass spectrometry
The concentration of CHG on nanofloss and delivered to the subgingival tissues were measured by LC/MS as described previously with slight modifications. 32 The measurements were undertaken by Agilent 1290 Infinity II LC/MSD XT, an ultra-high-performance liquid chromatography analysis system equipped with an electrospray ionizationmass spectrometry detector, using a ZORBAX RRHD C18 column

| Animals
Male Wistar rats weighing 276-300 g were acclimated in an animal room for 3 days and provided with regular chow and sterile water throughout the experiment. All experiments were performed according to the approved protocols by the Institutional Animal Care and Use Committee of the Faculty of Arts and Sciences, Harvard University, Cambridge.

| Drug delivery to subgingival region
The delivering ability of nanofloss and pharmacokinetics of the delivered drug in the tissues were studied in live rats using two different types of agents, CHG and BSA. CHG (0.05%)-nanofloss or commercialized floss immersed with CHG were inserted into the deep subgingival region of rats' upper front teeth. After 5-times flossing, the front teeth and gingival tissues were collected at several time points and placed into vials, which contain 50% methanol for LC/MS study to measure delivered CHG in the tissues. The LC/MS study was performed as described above.
Upper front teeth were flossed by BSA (Alexa 488)-nanofloss and the interdental area was observed by fluorescent stereomicroscope chronologically up to 2 days. Naïve silk thread was used as a negative control. In addition, naïve silk thread was soaked into BSA solution, rinsed in water, and then applied for the front teeth. As a positive control, BSA solution was insert into the gingival sulcus using a needle. The stereomicroscope images were split by color channels by Ima-geJ (NIH). Bright green fluorescent areas on the stereomicroscope images, which were split by color channels were measured by Ima-geJ (NIH).

| Frequency test
Frequency of drug release from nanofloss was studied. Rats' upper front teeth were flossed by 10 cm of CHG-nanofloss or commercialized floss with CHG for one to five times. After the flossing, each thread was placed into 50% methanol and the concentrations of CHG that remained on the threads were measured by the LC/MS. The LC/MS preparation is as described above. Concentrations of CHG before flossing were set as 100%.

| Bacteria culture
For all bacterial studies, 25 μl of P. gingivalis W83 from frozen stock was added to 5 ml modified GAM broth and incubated in an anaerobic jar for 48 h at 37 C until reached to 1 Â 10 9 CFU (colony forming units)/ml. AnaeroPack-Anaero was placed inside the jar to maintain anaerobic conditions during the culture period.

| In vitro antimicrobial test
Disk diffusion test was conducted to study in vitro antimicrobial effect of CHG-nanofloss. Naïve silk thread, nanofloss, CHG (0.05%)-nanofloss, and silk thread soaked in CHG (0.05%) solution (CHG [0.05%] soaked thread) were obtained for this experiment. All threads were rinsed in water and dried out before using. P. gingivalis suspension (1 Â 10 8 CFU/700 μl) was spread onto blood agar plate. The threads were placed onto the agar gel and incubated for 8 days under the anaerobic condition. Bactria growth-inhibition zone around the threads were measured by ImageJ.

| Statistical analysis
All data are presented as mean ± SEM. Statistical analyses were performed using GraphPad Prism (GraphPad Software, Inc., San Diego, CA, USA). Comparison between two groups was conducted using Mann-Whitney U test. Comparisons among multiple groups were conducted using Kruskal-Wallis test followed by Bonferroni post hoc test. Statistical significance was assumed at *p < 0.05, **p < 0.01.