This study was approved by IMU Joint-Committee on Research and Ethics under the research project ERGS/1/2013/SKK11/IMU/03/01.
The effectiveness of CPN as an intracanal medicament was evaluated against the strain E. faecalis (ATCC 29212) in human tooth model. The effectiveness of CPN was also assessed using SEM and CLSM. Another experiment was carried out to evaluate the effectiveness of CPN as an intracanal medicament against E. faecalis isolates from patients with failed root canal treatment.
Preparation of ethanolic extracts of Malaysian propolis
Malaysian propolis was collected from bee farm, Pahang, Malaysia with the following geographical coordinates: north latitude 3.8126°, east latitude 103.3256° and height of 12 m above sea level. There was no permission required to collect Propolis.
The extraction method used in this study was similar to the method explained by Jacob et al. . Propolis was manually cut into small pieces, 40 g were weighed using a weighing balance (Pyrometro, Malaysia) and divided equally into four pieces of ten grams each. Just after, in a flask, 20% (w/v) extract of propolis was prepared using 80% ethanol under constant agitation in a rotary shaker (Certomat Model S II, Sartorius, Goettingen, Germany) at 200 rpm, 37 °C for 48 h. This was later centrifuged (Eppendorf Model 5810 R, Hamburg, Germany) at 3000 rpm for 15 min, filtered through Whatman no.1 filter paper and subjected to reduced pressure using a rotary evaporator (Buchi Rotavapor R-215, Flawil, Switzerland) at the set pressure 175 mBar, temperature 52 °C and speed 95 rpm to remove the solvent. The ethanolic extract of propolis was then stored in a glass container and left for three days to allow evaporation of the residual solvent resulting in extracts of propolis (final weight/initial weight × 100). Stock solutions of 1 mg/ml of the extracts were prepared to use in further experiments. Saline with 0.1% DMSO was used to prepare the stock solution of propolis. To study the content of Malaysian propolis, reversed phase high performance liquid chromatography (RP-HPLC) analysis was carried out. The flavonoids such as pinocembrin (5.90 µg/ml), kaempferol (5.88 µg/ml) and quercetin (1.43 µg/ml) were identified to be in the highest concentration in Malaysian propolis .
Preparation and characterization of CPN
CPN were prepared by ionotropic gelation of chitosan with sodium TPP according to the method reported by Koukaras et al. . Stock solutions of 0.2% w/v of chitosan and 0.15% w/v sodium TPP were prepared by mixing in 1% v/v acetic acid and distilled water, respectively. The pH of both solutions was adjusted to between pH 5.0 and 5.5 by adding acetic acid. The different concentrations of ethanol extract of propolis was dissolved in chitosan solution with continuous stirring. The chitosan solutions containing propolis was added into the TPP solution and continuously stirred at 400–600 rpm at 37 °C. The nanoparticles were formed spontaneously due to ionic interaction. Following this, the formed nanoparticles were separated by centrifugation at 11,000 rpm for 25 min and the supernatants were discarded. CPN were resuspended in purified water for further characterization.
In this study, CPN (0.2% w/v chitosan and 1 mg/ml propolis) was used with an average particle size of 107.74 ± 0.53 nm, zeta potential of 45.2, polydispersity index of 0.225, and encapsulation efficiency of 88.8%. The shape of nanoparticles was observed using transmission electron microscopy. It was spherical in shape with a smooth surface similar to study done by Ong et al. .
Antibacterial effect of CPN as an intracanal medicament against E. faecalis (ATCC 29,212) in human tooth model
Preparation of dentine block specimens
In this study, the experiments were carried out in extracted human tooth model, a modification of Haapasalo & Orstavik tooth model in which bovine teeth were used. This provided a better simulation to clinical settings to assess the antibacterial effectiveness of intracanal medicaments in the dentinal tubules. This protocol is similar to study done by Chua et al. 
A total of 240 sound human teeth, including maxillary anterior teeth and mandibular canines with complete root formation were included in this study. The teeth were cleaned and stored in saline during all procedures to avoid dehydration. A low-speed diamond disc (Bredent®, Wittighausen, Senden, Germany) mounted on a milling machine under water cooling was used to section the teeth between cementoenamel junction and the apical third of the root to obtain 6 mm of the middle third of the root. Peeso Reamer no. 2 (Mani®, Utsunoniya, Tochigi, Japan) in a low-speed hand piece (Kavo, Charlotte, North Carolina, USA) was used to standardise the internal diameter of root canals to 0.9 mm. The dentine blocks were subjected to sonic irrigation (EndoActivator, Dentsply, Weybridge, Surrey, UK) using 5.25% NaOCl (Clorox®, Oakland, California, USA) and then 17% EDTA (Calasept®, Nordiska Dental, Ängelholm, Skåne Country, Sweden) for two minutes to remove smear layer. The dentine block specimens were thoroughly rinsed with sterile saline after each irrigation. Following this, the dentine blocks were sterilised by autoclave (LTE®, Oldham, Lancashire, UK) at 121 °C for 20 min. In order to prevent any contact of E. faecalis and medicament with the external surface, nail varnish was applied to the outer surface of the specimen. Petri dishes containing wax with a flat surface were prepared, and surface was disinfected using 70% ethanol and later air dried in a sterile biosafety cabinet before use. All experiments were done in the laminar hood after the ultraviolet sterilization. The dentine block specimens were placed upright with the apical ends fixed to the petri dishes with wax, using a thin small square of sterilised parafilm (Parafilm M®, Brand, Wertheim, Baden-Württemberg, Germany) obliterating the apical orifice to prevent any softened wax from entering the root canals.
E. faecalis inoculation
E. faecalis (ATCC 29212) were suspended in 20.0 ml of tryptic soy broth (TSB) (BD DifcoTM, NJ, USA). The cell suspension was adjusted to match the turbidity of 1.5 × 108 CFUs /ml (equivalent to 0.5 McFarland standards). The E. faecalis inoculum were transferred into the dentine block specimens using sterile 5.0 ml syringe (Terumo®, Somerset, New Jersey, USA) with 30-gauge needles (Terumo, Somerset, New Jersey, USA) in a sterile laminar flow hood. The coronal part of the dentine blocks was then sealed immediately using parafilm (Parafilm M®, Brand, Wertheim, Baden-Württemberg, Germany). Following this inoculation, the dentine block specimens were incubated for 21 days at 37 °C. The root canals were replenished with E. faecalis inoculum every three days to supply nutrients to bacteria and prevent their death.
Intracanal medicament placement
Following the inoculation period, 240 dentine blocks were randomly divided into eight groups (n = 30) according to the intracanal medicament placed: group I: saline, group II: chitosan, group III: propolis100 µg/ml (P100), group IV: propolis 250 µg/ml (P250), group V: chitosan-propolis nanoparticle100 µg/ml (CPN100), group VI: chitosan-propolis nanoparticle 250 µg/ml (CPN250), group VII: calcium hydroxide (CH) and group VIII: 2% chlorhexidine gel (2% CHX) (Consepsis V®, Ultradent, UT, USA).
Each group was further divided into three subgroups based on the time period (day one, three and seven) of the intracanal medicament placed. The intracanal medicaments were placed in the canal using a sterile 5.0 ml syringes (Terumo®, NJ, USA) and gel etchant needle tip (Kerr®, CA, USA) until the canals were completely filled. Thereafter, the coronal orifices were sealed using Parafilm (Parafilm M®, Brand, Wertheim, Germany). The blocks were kept in incubator at 37 °C for the experimental period of one, three and seven days.
Dentinal shavings collection
At the end of one, three and seven days, the dentine blocks were removed from the petri dishes and the canals were dried with sterile paper points. Samples of dentinal shavings were collected from all groups after one day of exposure, after three days of exposure and after seven days of exposure. Dentinal shavings were collected using peeso reamer (Mani®, Utsunoniya, Tochigi, Japan) size no. 4 equivalent to 1.3 mm diameter followed by size no. 6 equivalent to 1.7 mm diameter using a low speed handpiece (Kavo®, Charlotte, North Carolina, USA). Only one stroke was made to standardize the volume of dentinal debris collected.
The collected dentinal shavings were transferred into a micro-centrifuge tube (Axygen, NY, USA) containing 1 ml sterile tryptic soy broth (TSB) (BD DifcoTM, NJ, USA). A sterile microtip was used to take 100 µl of broth containing dentinal shavings and transferred to another tube containing 900 µl sterile tryptic soy broth (TSB) (BD DifcoTM, NJ, USA). The content of each tube was then serially diluted from 10–1 until 10–4. Subsequently, 300 µl of the diluted dentinal shavings was streaked uniformly using a L-shaped glass rod and triplicated. These tryptic soy agar plates (BD DifcoTM, NJ, USA) were incubated at 37 °C for 24 h. Following the incubation, the colonies were counted, and readings were tabulated.
Total numbers of CFUs were calculated to determine the remaining viable microbial population. The SPSS computer software version 21.0 (SPSS Inc., Chicago, Illinois, USA) was used to perform statistical analysis. Mean CFUs were compared between the groups and subgroups. Additionally, mean difference in CFUs between the groups based on different time periods and dentinal tubules depths was compared.
The data distribution was assessed for normality and was found that it did not follow a normal distribution. Therefore, non-parametric tests including Kruskal–Wallis test and Mann Whitney U test were used to compare CFUs between the groups and subgroups of intracanal medicaments and endodontic irrigants at different time periods and depths of dentinal tubules. Probability values of p < 0.05 were set as the reference for statistically significant results.
Dentinal blocks (n = 3 per group) were prepared using the same method as mentioned above under the dentine block specimens for SEM analysis before and after treatment. E. faecalis (ATCC 29,212) was cultured in 10 ml TSB (BD DifcoTM, NJ, USA) added with 8% sucrose with pH 7.4 and a minimal amount of xylitol (0–2%) at 37 °C for 48 h. This broth was incubated at 37 °C for 24 h. After centrifugation using 4000 rpm for 15 min, each cell pellet was washed thrice with sterile phosphate buffered solution (0.01 M, pH 7.2). Thereafter, it was re-suspended in 10 ml of the growth medium to adjust its concentration similar to 0.5 McFarland units (108 cells/ml) before use. The bacterial inoculum was mixed in five millilitres of TSB (BD DifcoTM, NJ, USA) and transferred into to root canal using sterilised syringe for a period of 21 days. The bacterial inoculation was similar to the method previously described in human tooth model used in this study. After 21 days, intracanal medicaments were placed according to the groups mentioned above. Two parallel grooves were created using a diamond disc onto the external surfaces of the dentin specimen in mesio-distal direction to facilitate a split fracture. Final splitting was done using chisel and hammer. Following this, all specimens were dehydrated in ascending grades of ethanol for 20 min each and immediately transferred into the pressure chamber of the critical point drying machine (CPD 30; Leica). All specimens were mounted on aluminium stubs using double-sided conductive tape and 30 nm-thick layer gold sputtering was done for two minutes. Following this, the specimens were examined using SEM (Philips/FEI XL30 FEG SEM, Japan) at an accelerated voltage of 5 kV at different magnifications and images were evaluated. Different magnifications and images were observed to evaluate the qualitative reduction of E. faecalis. Four-score scale system based on percentage of residual isolated microbial cells was used to assess the microbial coverage on SEM images of the canal walls . The scores were defined as clean dentine or residual isolated microbial cells covering less than 5% of the dentine, covering 5%—33% of the dentine, 34%—66% of dentine and 67%—100% of the dentine.
This analysis was conducted to evaluate the effectiveness of CPN250 and CPN100 as intracanal medicaments by assessing the viability profile. The proportion of live and dead bacteria was determined by fluorescent staining followed by imaging. The protocol used in this study was similar to done by Dawood et al. 
After the disinfection solution regimen, the specimen (n = 1 in each group) was rinsed in 0.1% by weight fluorescein for 24 h. Specimen were thereafter rinsed with deionised water and examined using CLSM (Leica Fluoview FV 1000, Olympus, Tokyo, Japan) equipped with a 60 × /1.4 NA oil immersion lens using 488 nm argon/helium and a 633 nm krypton ion laser illumination in reflection as well as fluorescence modes. Reflected and fluorescence signals were detected using a photomultiplier tube to a depth of 20 μm and then converted to single-projection images for better visualisation and qualitative analysis. Stacks of fluorescent images of the biofilm were obtained and examined using BioimageL software (v.2.0. Malmő, Sweden). This software provides information on the structure of the biofilm, including green-stained indicating live bacteria and red-stained indicating dead bacteria and volume on a two-dimensional x–y section based on colour segmentation algorithms written in MATLAB.
Antibacterial effect of CPN as an intracanal medicament against E. faecalis isolates from clinical samples
Ten patients aged between 20 and 60 years were selected from those who attended the IMU Oral Health Centre, Kuala Lumpur, Malaysia, needing endodontic retreatment. A detailed medical and dental history were obtained from each patient. Patients who have systemic disease or have received antibiotic treatment during the last three months were excluded from the study to minimise any risk of bias. Ten teeth from ten different patients with failed root canal treatment were included in this experiment. Failure of root-canal treatment was determined on the basis of clinical examination such as presence of pain, tenderness, swelling, sinus opening and mobility and radiographical examinations such as persistence of periapical lesion and root resorption.
After explaining the complete process of investigation including the method of sample collection, a written informed consent was obtained. Thereafter, the retreatment procedure was carried out. An access cavity was prepared under syringe irrigation using sterile high-speed diamond bur. Root-filling material was removed by rotary instrumentation and K-files (Dentsply-Maillefer, Ballaigues, Switzerland) in a crown-down technique without the use of chemical solvent, accomplished by irrigation with sterile saline. Following this, a sterile paper point (Dentsply-Maillefer, Ballaigues, Switzerland) was then introduced into the full length of the canal and retained in position for one minute for sampling. Culture procedure was done using the selective E. faecalis plates (Slanetz Bartley Agar (m-Enterococcus A.), Liofilchem, Italy) and the bacteria were grown and identified.
To prepare the E. faecalis inoculum, these isolates were suspended in 20.0 ml of TSB (BD DifcoTM, NJ, USA) and adjusted to match the turbidity of 1.5 × 108 CFUs /ml (equivalent to 0.5 McFarland standards) similar to the method describe above. Thereafter, one ml of this E. faecalis suspension was transferred into the Eppendorf tube containing 50 µl of each medicament according to these eight groups Group I: Saline, Group II: Chitosan, Group III: P100, Group IV: P250, Group V: CPN100, Group VI: CPN250, Group VII: CH and Group VIII: 2% CHX and incubated. After day one, three and seven, the content of each tube was serially diluted as described above in this study. 300 µl of the diluted shavings was streaked evenly using a L-shaped glass rod and triplicated. Thereafter, these plates were incubated at 37 °C for 24 h, bacteria were grown CFUs were calculated.