Quercetin as an auxiliary endodontic irrigant for root canal treatment: Anti-biofilm and dentin collagen-stabilizing effects

: (1) Background: Bacterial reinfection and root fracture are the main culprits related to root canal treatment failure. This study aimed to assess the utility of quercetin solution as an adjunctive endodontic irrigant that strengthen root canal dentin with commitment anti-biofilm activity and bio-safety. (2) Methods: Based on a noninvasive dentin infection model, dentin tubules infected with Enterococcus faecalis ( E. faecalis ) were irrigated with sterile water (control group), and 0, 1, 2, 4 wt% quercetin-containing ethanol solutions. The live and dead bacteria proportions within E. faecalis biofilms were analyzed using confocal laser scanning microscopy (CLSM). Elastic modulus and hydroxyproline release and X-ray photoelectron spectroscopy (XPS) characterization was tested on irrigant-treated demineralized dentin to evaluate irrigants’ biostability. The cytotoxicity of irrigants was tested by CCK-8 assay. (3) Results: Quercetin increased the proportion of dead bacteria volumes within E. faecalis , and improved the flexural strength of dentin collagen compared to control group. The XPS characterization revealed an increase in C-O peak area under both C1s and O1s narrow-scan spectra. The CCK-8 assay confirmed no cytotoxicity of quercetin solutions. (4) Conclusions: Quercetin exhibited anti-biofilm activity, collagen-stabilizing effect as well as cytocompatibility, supporting quercetin as a potential candidate for endodontic irrigant.


Introduction
Root canal treatment is the method of choice for treating oral diseases associated with pulp and periapical infection. Its success mainly relies on thorough chemo-mechanical debridement of the infected root canals through the effective use of disinfectants and instrumentation, followed by sealing of the enlarged, disinfected canals with appropriate root filling materials. However, the bacterial elimination efficacy of root canal treatment is often compromised due to various reasons, such as untouched residual debris, and incomplete root canal filling that would allow secondary bacterial colonization [1][2][3]. In an ocean of failure cases, Enterococcus faecalis (E. faecalis) has been reported to present in the root canal as a biofilm alone or as the predominant bacterial specie [4][5][6]. In this regards, besides meticulous instrumentation, auxiliary endodontic medicaments (e.g. calcium hydroxide) and irrigants (e.g. sodium hypochlorite) with antibacterial activity are adopted by clinicians during root canal treatment, in order to disinfect the canal system and prevent reinfection. However, the risk of using these disinfectants is that they will cause alterations in chemical and structural composition of the dentin (e.g. degradation of apatite-depleted dentin collagen) when they are used in high concentrations and/or for long periods [7,8]. In turn, these changes reduced the mechanical properties of dentin, such as microhardness [9][10][11], flexural strength [12,13], modulus of elasticity [13,14], and the fracture resistance under normal masticatory loads. This is considered to be another risk factor for the failure of root canal treatment. Therefore, in recent years, endodontic disinfectants that kill E. faecalis whilst not compromising the biomechanical properties of dentin are in great demand [15][16][17].
Quercetin ( Figure 1) is a natural bioflavonoid found in abundance in food and beverages such as broccoli, onions, tea, apples, and berries. It has many biological and pharmacological benefits, such as anti-inflammatory, anti-allergic and antioxidant properties, owing to its ability to remove reactive oxygen radicals and superoxide anions [18][19][20]. It has also been proved to be an effective antimicrobial agent against a wide range of pathogenic microorganisms, including E. faecalis [21][22][23][24][25][26]. In the field of biomaterial and bioengineering, quercetin has also been widely utilized for biomodification of scaffold materials due to its glutaraldehyde-comparable collagen-crosslinking capability and superior cytocompatibility [27]. Recently, it has been doped into dentin adhesive with the aim to improve the long-term performance of resin-dentin bonding interface, wherein quercetin protects the structural integrity of dentin collagen from biodegradation by crosslinking the apatite-depleted collagen [28]. The anti-biofilm and collagen-stabilizing effects of quercetin make it a potential candidate for endodontic irrigant. Therefore, the aim of this study was to explore the feasibility of quercetin solution as an auxiliary endodontic irrigant with both anti-biofilm activities against E. faecalis and dentin collagen-stabilizing capability. We evaluated the anti-biofilm effect of quercetin by confocal laser scanning microscopy (CLSM), based on an in vitro dentin infection model with intensive and ubiquitous E. faecalis infection in the dentin tubules [29]. The dentin collagen-stabilizing capability of quercetin solution was assessed in terms of biomechanical properties and biodegradation resistance of dentin collagen. Moreover, biomodification of demineralized dentin collagen by quercetin was characterized by X-ray photoelectron spectroscopy (XPS), and the cytotoxicity of quercetin irrigant was tested on human dental pulp cells (HDPs).

Experimental irrigant preparation and tooth collection
Due to the poor solubility in water, we selected pure ethanol as the solvent for quercetin in our study [30]. Quercetin powder (Sigma-Aldrich, St. Louis, MO, USA) was directly added to 100 % ethanol, followed by water-bath heating at 37 ℃ for 15 min. All experiments are conducted based on the following 5 groups: sterile water (control group), 0 wt% (ethanol group), 1 wt%, 2 wt% and 4 wt% quercetin groups.
All methodologies applied in this study were performed in accordance with the relevant guidelines and regulations. Human single-rooted teeth (antibacterial study and XPS characterization) and third molars (biomechanical study) were collected with informed consent. Different fresh and intact human teeth were obtained for the following purposes: (a) 10 single root premolars for antimicrobial test; (b) 40 dentin beams (1.2 × 0.8 × 6.0 mm 3 ) prepared from the third molars for biomechanical test; and (c) 5 premolars for XPS examination. All teeth were stored in saline solution at 4 ℃ for no more than one month..

Antibacterial test
2.2.1. Sample preparation As previously described [29], each single-rooted tooth was horizontally sectioned 1 mm below the cementoenamel junction using a low-speed diamond saw (SYJ-150, Kejin, Shenyang, China) at 300 rpm under water cooling. Then, the root was cut in parallel with the occlusal surface of the teeth to obtain a cylindrical root dentin block with a uniform length of 4 mm. The root canals inside the dentin block were enlarged to a diameter of 1.5mm a size 6 Gates Glidden drill (Tulsa Dentsply, Tulsa, OK, USA) at 300 rpm under water cooling. A groove was made in the middle of each cylindrical dentin block using a low-speed handpiece with a round bur. Then the dentin block was fractured with a blade and a hammer into 2 semi-cylindrical halves. The semi-cylindrical halves were ground to 2 mm thickness by 600-grit silicon paper. The specimens were then shaped using a low-speed handpiece with a fine carbide bur at 300 rpm under water cooling to obtain refined semi-cylindrical specimens with a size approximately 4 × 4 × 2 mm 3 , so that they can be placed in a 0.45 μm pore size filter tube (Pall Corporation, Ann Arbor, MI) [29].
The specimens were rinsed with 5.25 % sodium hypochlorite and 6 % citric acid (pH 4.0) solution in an ultrasonic bath (Sankei Giken Industry Co Ltd, MIE, Tokyo, Japan) for 4 minutes to remove the smear layer, and were rinsed in sterile water for 1 minute. Each specimen was placed in a filter tube with the canal side up. The gap between the specimen and the inner walls was sealed with composite resin (Kerr Co, Orange, CA, USA).

Infection of Dentin
E. faecalis ATCC 29212 (American Type Culture Collection, Manassas, VA, USA) was cultured on brain-heart infusion (BHI) agar (Becton-Dickinson, Sparks, MD) plates at 37 ℃ overnight. Then, the bacteria were suspended in BHI broth and standardized spectrophotometrically to 3 × 10 6 colony-forming units (CFU)/mL (OD405 = 0.05). Then 500 μL of the bacterial suspension was added into each filter tube wherein dentin specimen had been placed. The tubes were centrifuged at forces of 1400 g, 2000 g, 3600 g, and 5600 g in a sequence twice each for 5 minutes. After dumping the last centrifugal solution, the upper compartment was refilled with a fresh 500 μL of bacterial suspension before the next force of centrifugation. Ten filter tubes containing the dentin specimens were incubated in sterile BHI broth in air at 37 ℃ for 1 week.
After incubation, the infected specimens were taken out from the filter tube and the surrounding composite was removed. The specimens were rinsed by sterile water for 1 min and air-dried. Then, nail varnish was used to seal the cemental sides of the dentin pieces. Subsequently, a total of 20 specimens were randomly distributed into 5 groups with different irrigants as previously described: control group, ethanol group, 1 wt%, 2 wt% and 4 wt% quercetin groups. A droplet (50 μL) of irrigant was placed on the pulp side of dentin pieces for 3 minutes. Then the specimens were rinsed with sterile water for 1 minute and fractured vertically through the root canal into two halves to expose a fresh surface of longitudinally fractured dentinal tubules for CLSM examination.

Confocal laser scanning microscopy (CLSM) examination
The fractured dentin specimens were stained with LIVE/DEAD BacLight Bacterial Viability Kit (Molecular Probes, Eugene, OR, USA) according to the manufacturer's instructions. The excitation/emission wavelengths of SYTO 9 (green fluorescent dye that stains live bacteria) and propidium iodide (PI, red fluorescent dye that stains dead bacteria) were 480/500 nm and 490/634 nm, respectively. The mounted specimens were observed by CLSM (LSM 710, Zeiss, Oberkochen, Germany) using a 20 × lens with an addi-tional zoom of 2 ×. Under the microscope, the border of root canal for each specimen was first located, and then five randomly selected locations starting from the root canal border were scanned. For any scanned location, one image stack containing 20 slices were acquired at 0.5-μm z-step (the distance between two adjacent images of a stack). Three-dimensional reconstructions and volume calculation of all image stacks were performed using Imaris 7.2 software (Bitplan Inc, St Paul, MN, USA). The proportion of dead bacteria within the E. faecalis biofilms was represented by the volume ratio of red fluorescence to green and red fluorescence.

Modulus of elasticity of demineralized dentin before and after irrigation
Forty dentin beams (1.2 × 0.8 × 6.0 mm 3 ) free of notches and cracks were sectioned from root dentin of third molars as previously described [31]. They were completely demineralized in 0.5 M ethylenediaminetetraacetic acid (EDTA) (pH= 7.5) for 7 days at 37 °C and were thoroughly rinsed with distilled water for 10 min. The demineralized beams were divided into 5 groups and immersed in 2-mL irrigants for 3 min, followed by water-rinsing for 1 minute. Bacterial collagenase solution was prepared by dissolving bacterial collagenase (from Clostridium histolyticum, 125 collagen digestion units per mg solid; Sigma-Aldrich, Merck KGaA, Darmstadt, Germany) in 0.4 M ammonium bicarbonate-buffered solution (pH = 7.9) at a concentration of 0.1 % (m/m). The beams were incubated in bacterial collagenase solution individually in 96-well plate for 48 h at 37° C. The Modulus of the elasticity of all beams was estimated at baseline, after 3-minutes treatment in their irrgant solution and after 48-hours enzymatic hydrolysis through three-point bending flexure test. The cross-head of the universal tester is used at a rate of 1mm/min in a controlled range of up to 10 % of stain [32]. The corresponding modulus of elasticity was calculated according to the formula using pressure-displacement curves.

Hydroxyproline (HYP) release assay
After aforementioned 48-h collagenase incubation, the dry mass of each beam was measured and the HYP content in the collagenase incubation medium was analyzed by using an HYP kit (Jiancheng, Nanjing, China). The HYP content was used to estimate the percentage of degraded collagen, based on the knowledge that 90 % of the dry mass of demineralized dentin consists of type I collagen and that dentin collagen contains 9.6 % of hydroxyproline [33]. The absorbance was measured at 550 nm with a spectrophotometer (Tecan Group, Männedorf, Switzerland) and the HYP released from each beam was represented by dividing the HYP by the dry weight of dentin beam (μg HYP/mg dentin).

Biomechanical Test
Demineralized dentin discs (4 ×4 ×1 mm 3 ) were treated by different irrigants for 3 min each, and were rinsed with deionized water for 1 min. Then, the samples were dried for 12 h in a vacuum pump (Value FY-4C-N, Hangzhou, China). X-ray Photoelectron Spectrometer (ESCALAB250Xi, Thermo Fisher Scientific, Waltham, USA) with monochromatic Al Kα X-ray source was used to characterize the elemental distribution of C1s and O1s on the dried discs. Five sites of the fractured surfaces of each specimen were randomly selected for the analysis. The acquired data are analyzed using XPS PEAK 4.1 software for peaks of each element. The high resolution of C1s and O1s spectra was curve-fitted based on Gaussian-Lorentzian composite function with the background subtracted in Shirley mode.

Cytotoxicity evaluation with CCK-8 assay
Human dental pulp cells (HDPs) were a gift from the Department of Stomatology, Union Hospital, Huazhong University of Science and Technology. The cells were cultured in α-mem (HyClone, Logan, UT, USA) and 100 μg/mL of penicillin G-streptomycin at 37 ℃ with 5 % CO2 in a humidified environment. The third passage of HDPs were seeded into 96-well plates (5000 cells per well) for 24 h, followed by a 24-hour treatment with 100-fold dilutions of the previous test solutions in α-mem. Thereafter, the Cell Counting Kit-8 (CCK-8) was performed according to the manufacturer's protocol for 4 h to assess cell viability (%). The experiment was performed in quintuplicate.

Statistical analysis
Data are presented as mean ± standard deviation. Statistical analysis was performed using SPSS (version 25.0) software package at the 5 % level of significance. All data were submitted to one-way analysis of variance (ANOVA) and complemented by Tukey's multiple comparison. Groups labeled with the same letters are not statistically different (P > 0.05).

Results
Bacteria penetration from root canal to dentin tubules after centrifugation and the anti-biofilm effect of each irrigant against E. faecalis are shown in Figure 2. The histogram represented the death cell rate of E. faecalis biofilm that grown infected dentin tubules. There were statistically significant differences between the experimental and the control groups (P < 0.05). Both the pure ethanol and quercetin solutions showed obvious anti-biofilm effects, and as the concentration of the quercetin solution increased, the proportion of dead volumes of E. faecalis raised from 19.13 % to 57.8 %, with a statistically significant difference (P < 0.05). The 4 wt% group (57.8 %) showed the strongest anti-biofilm activity (P < 0.05).  Figure 3 shows that all quercetin groups had a significant higher elastic modulus, and less loss of modulus and HYP release than that of control groups (P < 0.05). Demineralized dentin beams in the control and pure ethanol group had no significant change of modulus (P > 0.05) and had most reduction in elastic modulus after enzymatic degradation (P < 0.05). However, dentin beams in the quercetin groups, in concentration between 1 to 4 wt%, had more increase in elastic modulus after irrigant treatment (P < 0.05), and the less loss of elastic modulus reduced after biodegradation (P < 0.05), ranging from 55.5 % to 21.93 %. After 48-h collagenase incubation, dentin beams from control group released 520.9 ± 115.6 μg of HYP/mg dentin, whilst beams irrigated by quercetin demonstrated lower HYP release (P < 0.05), ranging from 88.15 ± 37.6 to 296.9 ± 45.6 μg of HYP/mg dentin. The improvement of elasticity and biodegradation resistance of dentin was quercetion concentration-dependent (P < 0.05). The XPS analysis characterized the chemical bonds of dentin collagen after quercetin biomodification as a surface-specific technique (Figure 4). The C1s spectra were curve-fitted into three main peaks at 284.7, 285.9 and 288.0 eV, which are associated with carbons in the aliphatic chain of amino acid side chains (C-C/C-H), carbons at the α-position of peptidic residues (C-O/C-N) and carbons in peptidic carbonyl and in carboxyl (C=O), respectively [34]. The O1s spectra exhibits two main peaks at 531.0 and 532.2 eV, which are dominantly attributed to the ketonic oxygen (C=O) and hydroxyl oxygen (C-O), respectively [35]. The detailed parameters of each peak of C1s and O1s spectra were listed in Table 1. Due to the abundance of hydroxyl groups in quercetin molecules, the C-O peak areas of both C1s and O1s spectra increased, in a quercetion concentration-dependent manner. The relative area of C-O peak under C1s from in the quercetin groups increased from 13.0 % to 46.4 %, and the area of C-O peak under O1s increased from 26.2 % to 68.8 %, which indicated that quercetin with abundant phenolic hydroxyls was grafted onto dentin collagen.   Using human dental pulp cells, we investigated the cytotoxicity of quercetin solution ( Figure 5). Following a 24-h exposure, the quercetin irrigants presented acceptable cytotoxicity compared with the control group (P > 0.05), even at the highest concentration tested. Figure 5. Evaluation of HDPs viability exposed to different quercetin-containing mediums.

Discussion
An essential step in successful root canal treatment is to minimize the number of microorganisms in the root canal and dentin tubules through mechanical instrumentation and the use of irrigants with antimicrobial properties. However, due to the complexity of the root canal system (e.g., isthmus, lateral canals, root canal side, traffic branch, root separation and root bifurcation), it is difficult to remove all bacteria hidden in the dentin tubules by routine mechanical removal or endodontic irrigants [36][37][38]. In addition, lurking in the root canal system, bacteria can withstand harsh environmental conditions and exhibits resistance to many antibiotics [2,39]. E. faecalis is physically and ecologically strong, and easy to penetrate into dentin tube [40,41]. Although several endodontic irrigants have been investigated, some compounds demonstrate many adverse effects such as compromised efficiency, cytotoxicity to mammal cells, allergic potential and even reduction in the mechanical properties of dentins. Some studies have confirmed that chlorhexidine solution cannot infiltrate into dentinal tubules wherein a great number of infectious bacteria could hide for survival, but can decrease the microhardness of dentin [9,37,42]. Sodium hypochlorite can penetrate the dentin and degrade the collagen matrix, eventually leading to root fracture [43,44]. Given this, the search for an adjunctive endodontic irrigant with strong antibacterial properties, dentin collagen-stabilizing ability and good biocompatibility is paramount. In previous studies, quercetin has been proven to promote the long-time resin-dentin bonding [28,45]. In this study, we evaluated the utility of quercetin as auxiliary endodontic disinfectant and its effect on mechanical properties and degradation resistance of dentin. In order to eliminate the effect of ethanol on the experimental results, 100 % ethanol was utilized as negative control since it was used to dissolve quercetin. Ethanol has a strong permeability and can infiltrate into the inner parts of bacteria and fungi to denature proteins crucial for their survival. Previous studies have proved that 95 % ethanol had antibacterial effects against mono-and multispecies biofilms as an endodontic irrigant [46]. In addition, ethanol can improve the sealing ability of the root canal obturation decrease the leakage as a final irrigant, and has no effect on endodontic irrigants' antimicrobial properties [47][48][49]. Our experiments also show that ethanol solutions at appropriate concentrations are not significantly cytotoxic to HDPs, indicating that ethanol is a suitable solvent for our study.
By establishing an in vitro noninvasive model of infected root dentin tubules, we for the first time demonstrated the antibacterial ability of quercetin-containing ethanol solutions against E. faecalis biofilms hidden in the dentin tubules ( Figure 2). With increased concentration, the antibacterial effect of the quercetin solution was enhanced with deeper penetration into the dentin tubules (P < 0.05). Quercetin is a natural polyphenolic flavonoid that presents in many kinds of plants. It has a strong antimicrobial ability against various Gram-positive bacteria, Gram-negative bacteria, and viruses [23,50]. It has been reported that quercetin can change the membrane potential to hinder bacteria's ability to synthesize adenosine triphosphate (ATP) and transporting material, thereby killing the bacteria [51]. However, it is hard to guarantee the antimicrobial ability in infected dentin tubes in vivo. Before the development of this noninvasive model, dentin block culture had been used as a very common method to simulate the presence, action, and eradication of bacteria in dentin tubules. However, it is difficult or impossible to obtain substantial, evenly distributed bacterial presence in dentin tubules with this method [52,53]. The application of this new model allows predictable, intensive and deep penetration of bacteria in dentin tubules, greatly aiding the study of dentin disinfection. In combination with the CLMS test, we not only directly visualized the bacteria, but also identified live and dead bacteria in infected dentins, which verified this in vitro noninvasive model as a suitable detection method with clinical relevance [5,54].
Our results showed that quercetin irrigant increased the mechanical properties and the biodegradation resistance ( Figure 3A, B). Owing to the abundant phenolic hydroxyls on its molecules, quercetin may interact with the collagen in root dentin via hydrogen bonds, van der Waals forces, electrostatic forces, and hydrophobic forces, thus maintaining the stability of exposed collagen matrix after irrigant application [55], and improve the mechanical properties of demineralized dentin [56]. In current work, we utilized XPS to confirm the quercetin-biomodificaiton of dentin collagen through the increased C-O peak area under the C1s and O1s spectra ( Figure 4 and Table 1) [34,57]. We also confirmed the resistance to collagenase degradation of quercetin-treated dentin by HYP release assay. Dentin collagen treated by high concentrations (i.e. 2 wt%, 4 wt%) of quercetin showed the least HYP release ( Figure 3C). Flavonoids, such as quercetin, inhibit both free and collagen-bound proteolytic enzymes in dentin [58] and down-regulate endogenous protease expression [59,60] to inactivate the protease, preventing free access of collagenase to sites containing collagen chains, thus increasing collagen's resistance to enzymatic degradation [61]. In addition, quercetin crosslinks with the exposed dentin to provide a mechanical barrier to the collagen matrix, with protection of the remaining dentin against acid attack erosion [62], abrasion [63] and erosion plus abrasion [64]. Since endodontic irrigants work directly in contact with human tissues in the clinic, they should be biocompatible to human cells and tissues. Quercetin is known to have good biosafety and low cytotoxicity [65]. According to the CCK-8 results ( Figure 5), quercetin-containing ethanol solution with low concentration had no any cytotoxic effect on human dental pulp cells. But the other results argued the cytotoxic of quercetin with high concentration [66]. However, a previous study has emphasized that irrigants generally do not flow all the way to the root end in the positive pressure of irrigation technique [67]. Therefore, it is proper to use quercetin as an auxiliary endodontic irrigant under proper isolation condition and technical conditions.

Conclusions
Quercetin-containing ethanol irrigants showed bactericidal effect against E. feacalis biofilms in dentin tubules and can maintain the mechanical strength of demineralized dentin and enhance biodegradation resistance. This proof-of-concept study demonstrated the utility of quercetin as a safe and reliable adjunctive root canal irrigant. Further clinical studies are still needed to evaluate its effectiveness in combination with other root canal disfectants, as well as the biosafety in vivo.