Sol-gel-derived calcium silicate cement incorporating collagen and mesoporous bioglass nanoparticles for dental pulp therapy

Dental Materials 40 (2024) 1832–1842 Available online 24 August 2024 0109-5641/© 2024 The Authors. Published by Elsevier Inc. on behalf of The Academy of Dental Materials. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/). Sol-gel-derived calcium silicate cement incorporating collagen and mesoporous bioglass nanoparticles for dental pulp therapy Hazel O. Simila a,1 , Caroline Anselmi b,1 , Lais M. Cardoso b , Renan Dal-Fabbro b , Ana M. Beltr´ an c , Marco C. Bottino b,d,* , Aldo R. Boccaccini a,** a Institute of Biomaterials, University of Erlangen-Nuremberg, Cauerstraße 6, 91058 Erlangen, Germany b Department of Cariology, Restorative Sciences, and Endodontics, School of Dentistry, University of Michigan, Ann Arbor, MI, USA c Departamento de Ingeniería y Ciencia de los Materiales y del Transporte, Escuela Polit´ ecnica Superior, Universidad de Sevilla, 41011 Seville, Spain d Department of Biomedical Engineering, College of Engineering, University of Michigan, Ann Arbor, MI, USA ARTICLE INFO Keywords: Anti-bacterial agents Calcium silicate Collagen Dental pulp Mesoporous bioactive glass nanoparticles Odontogenic ABSTRACT Objective: Calcium silicate cements (CSCs) are often used in endodontics despite some limitations related to their physical properties and antibacterial efficacy. This study aimed to develop and demonstrate the viability of a series of CSCs that were produced by sol-gel method and further modified with mesoporous bioactive glass nanoparticles (MBGNs) and collagen, for endodontic therapy. Methods: Calcium silicate (CS) particles and MBGNs were synthesized by the sol-gel method, and their elemental, molecular, and physical microstructure was characterized. Three CSCs were developed by mixing the CS with distilled water (CS+H 2 O), 10 mg/mL collagen solution (CS+colH 2 O), and MBGNs (10 %) (CSmbgn+colH 2 O). The mixing (MT) and setting (ST) times of the CSCs were determined, while the setting reaction was monitored in real-time. Antibacterial efficacy against Enterococcus faecalis (E. faecalis) and regenerative potential on dental pulp stem cells (DPSCs) were also analyzed. Results: The CS+H 2 O displayed a ST comparable to commercial products, while CSmbgn+colH 2 O achieved the longest MT of 68 s and the shortest ST of 8 min. All the experimental CSCs inhibited the growth of E. faecalis. Additionally, compared to the control group, CSCs supported cell proliferation and spreading and mineralized matrix production, regardless of their composition. Significance: Tested CSCs presented potential as candidates for pulp therapy procedures. Future research should investigate the pulp regeneration mechanisms alongside rigorous antibacterial evaluations, preferably with multi-organism biofilms, executed over extended periods. 1. Introduction Calcium silicate cements (CSCs) have gained prominence in endodontics and dentin-pulp regeneration due to their superior biological and physical properties, with multiple CS-based commercial products being adopted in clinical practice [1]. These biocompatible materials support cell attachment and proliferation and induce mineralization, making them ideal for pulp capping, root-end filling, apexification procedures, and other regenerative endodontic procedures [2]. Their unique composition allows them to set in the presence of moisture, forming a hard, impermeable seal resistant to bacterial penetration. However, CSs could still benefit from improvements in handling, physicomechanical, and antibacterial properties [3]. The ability of new biomaterial formulations to inhibit microbial growth under biofilms involved in endodontic infections has been expressly recommended [4]. Although several studies have tested both commercially successful CSCs [5–8] and experimental CSC formulations [9] for antimicrobial efficacy, most of these studies used different methodologies, yielding results that are not directly comparable, hence complicating the ability to measure how these materials genuinely * Corresponding author at: Department of Cariology, Restorative Sciences, and Endodontics, School of Dentistry, University of Michigan, Ann Arbor, Michigan, USA. ** Corresponding author. E-mail addresses: [email protected] (M.C. Bottino), [email protected] (A.R. Boccaccini). 1 Hazel O. Simila and Caroline Anselmi have contributed equally to the work. Contents lists available at ScienceDirect Dental Materials journal homepage: www.elsevier.com/locate/dental https://doi.org/10.1016/j.dental.2024.08.006 Received 21 April 2024; Received in revised form 8 August 2024; Accepted 15 August 2024 Dental Materials 40 (2024) 1832–1842 1833 perform [10]. In addition, attempts to enhance the antibacterial properties of CSCs with compounds such as chlorhexidine [11], cetrimide [12], and titanium tetrafluoride [13] persist and constitute a valid research pursuit. Moreover, a review of CSCs suggests using porous materials and nanoparticles (NPs) in combination with CSCs as the next target for enhancing these cements’ physical and biological properties [3]. It has been proposed that nanotechnology could help fabricate CS- based materials of superior characteristics [14]. Sol-gel synthesis is one such method that is being used to optimize inorganic nanoparticles for biomedical applications. This technique synthesizes mesoporous bioactive glass nanoparticles (MBGNs), which are increasingly used as main constituents, fillers, or coatings to prepare a wide range of composites [15–18]. The rationale for utilizing MBGNs lies in their nanosized and mesoporous structure, which can upgrade the physical properties of the CSCs [3,14,16]. Moreover, their bioactivity promotes hydroxyapatite formation, improving dentin remineralization and reducing post-operative complications. Recently, several nature-inspired polymers, such as gelatin [19,20], silk fibroin [21], and polysaccharides, such as chitosan [22] or alginate [23], have been used to improve CSCs. Collagen, a natural polymer found in teeth, has been touted as holding potential for pulp-dentin complex regeneration given that it is highly biocompatible, has stem cell recruiting ability, has variable solubility, is biodegradable, and is of low immunogenicity [24,25]. An alternative to conventional collagen sources extracted from animal tissues is the Recombinant Collagen Protein (RCP). Notably, the integration of RCP in CSCs has yet to be explored, with limited reports highlighting this potential synergy [26–28]. Consequently, our research pioneered the development of innovative CSCs that incorporate MBGNs and RCP. The developed CSCs should possess ideal physicochemical properties and display antibacterial efficacy against E. faecalis while maintaining cytocompatibility with Dental Pulp Stem Cells (DPSCs). Most importantly, we aimed to demonstrate that the novel CSCs formulated from particle precursors of CS and MBGNs synthesized by the sol-gel method have comparable and desirable properties to those of other commercial CSCs produced by standard methods. Ultimately, we verified that the reported cements uphold the properties of CSCs, hence validating the choice of sol-gel synthesis and inclusion of the selected additives in the formulation of the tested CSCs. 2. Materials and methods 2.1. Materials The precursor CS and MBGNs were synthesized using the sol-gel method according to our previously published protocols [29]. Briefly, to synthesize CS, a solution of 200 μ L of 65 % nitric acid (HNO 3 ) (Sigma, Germany) catalyst in 15 mL of deionized (DI) water (MilliQ) was prepared. Then, 8.5 mL of tetraethyl orthosilicate (TEOS) (Sigma, Germany) was added, followed by 30 min of stirring under ambient conditions. Next, 26.5 g of calcium nitrate tetrahydrate (Sigma, Germany) was sequentially dissolved in small amounts. The final mixture was stirred for a further 1 h at 500 rpm before aging the sol in a 60 ◦C oven for 24 h, followed by drying for 24 h at 120 ◦C and calcination at 700 ◦C for 24 h. The final powder was milled in a grinding machine (Retsch PM 100, Retsch GmbH, Germany). On the other hand, preparation of MBGNs of nominal composition (mol%) 70SiO 2 -30CaO commenced with the dissolution of 5.6 g of cetrimonium bromide (CTAB) (Merck, Germany) in 264 mL of DI water at 35 ◦C. Approximately 10–20 min later, when the solution was no longer turbid, the temperature was reduced to 20 ◦C before adding 80 mL of ethyl acetate (EA)−99 % (Sigma, Germany), 56 mL of 1 M aqueous ammonia solution (VWR, France) and 28.8 mL of TEOS in 30 and 15 min intervals. Finally, 18.24 mg of calcium nitrate tetrahydrate (Ca (NO 3 )2.4 H 2 O) was added batch-wise, allowing complete dissolution between increments. The final solution was stirred for 4 h before centrifuging the resultant suspension at 7830 rpm for 15 min (Centrifuge 5430 R, Eppendorf, Germany) to retrieve the nanoparticles. DI water and 96 % ethanol were used to rinse the NPs, which were dried at 60 ◦C overnight and calcined at 700 ◦C for 3 h (2 ◦C/min). Zirconium dioxide (ZrO 2 ) was purchased from Sigma Aldrich (Germany), while the RCP powder was obtained from Biollagen™ (Jland, UK). The synthesized particles were characterized by scanning electron microscopy (SEM) (FESEM, Auriga Crossbeam, Carl Zeiss Microscopy, GmbH, Germany) and by bright-field transmission electron microscopy (BFTEM) with energy-dispersive X-ray analysis (EDX) (TEM, FEI Talos F200S microscope, Thermofischer, USA). Additional molecular and structural characterizations were undertaken by Fourier transform infrared spectroscopy (FTIR) (SHIMADZU, IRAffinity-1S spectrophotometer, Shimadzu Corp, Japan) and X-ray diffraction (XRD) (Miniflex 600 HR, Rigaku, Japan). For the formulation of the powder blends for cement preparation, CS and ZrO 2 powders were weighed and dry mixed in a milling device (Retsch PM 100) in a 7:3 ratio for 30 min at 450 rpm with reversal of the direction of rotation every 5 min, and a pause interval of 2 min. This formed the classical powder admixture. Blended powder (previously described) was mixed with MBGNs (10 wt%) using the same mixing parameters for a second alternative. The subsequent powder blends were used to produce the cements. 2.2. Cement preparation Cement was prepared using a powder-to-liquid ratio of 3:2 (w:v), with the liquid being either distilled water (MilliQ - ELGA DV 25 PURELAB option R7BP) or a solution of the RCP powder proportioned for dissolution in distilled water at concentrations of 10 mg/mL. The latter reactant was labeled as colH 2 O. Complete solubility was achieved in a 37 ◦C environment for 10 min. The respective powder and liquid were dispensed onto a mixing slab and spatulated until all the powder had been incorporated and uniform consistency obtained. Ultimately, three distinct types of cement were prepared: the first, CS+H 2 O, consists of a calcium silicate and zirconium oxide (CSZrO 2 ) admixture combined with water; the second, CS+colH 2 O, blends the CSZrO 2 admixture with a collagen solution to integrate collagen’s biological properties; and the third, CSmbgn+colH 2 O, incorporates MBGN into the CSZrO 2 admixture, further mixed with a collagen solution, leveraging the synergistic benefits of MBGN and collagen. For the comprehensive evaluation in subsequent experiments, these cements were utilized in their initial state as freshly prepared pastes and after the setting process to solidify, enabling analysis of both their immediate and final hardened properties. 2.3. Characterization of the setting reaction After dispensing the powder and liquid in the desirable proportions onto a mixing slab, a countdown timer was activated, and the amount of time elapsed from the beginning of mixing until a final workable consistency was recorded. The setting time was determined according to Standard ISO 9917–2. At the end of mixing, the cement paste was loaded into a plastic mold measuring 2 mm in height and 6 mm in diameter. The assembly was placed inside a plastic bag with a moist piece of gauze inside a humid oven at 37 ◦C. After 5 min, the cement inside the mold was indented with the small, 1 mm wide Gilmore tip weighing 113.4 ± 0.5 g (Gilmore apparatus, TESTING, Bluhm & Feuerherdt GmbH, Germany). Initial setting time (IST) was recorded as the period from the onset of mixing until the indentor left no mark. On the other hand, the final setting time (FST) was checked using the same steps but applying the 2 mm Gilmore tip, weighted with 453.6 ±0.5 g. During the initial 5 min, checks were performed every 3 min, every 1 min for IST, and every 5 min for FST, up to 1 h. The setting time values of each cement were obtained by arithmetic mean of four repetitions of the test for each cement sample. FTIR and XRD also monitored the setting cements at baseline and every 60, 120, 240, 360, and/or 420 min. H.O. Simila et al. Dental Materials 40 (2024) 1832–1842 1834 2.4. Antibacterial properties Enterococcus faecalis (E. faecalis; ATCC 19433) was used to evaluate the antimicrobial potential of these experimental cements using the agar diffusion test and optical density (OD) measurements through indirect contact tests. As a first step, the bacteria strain at a cell density of 3 ×10 8 CFU mL −1 was cultured in Brain and Heart Infusion (BHI; Difco Laboratories Inc. MI, USA) broth overnight and used as per the methodology below. 2.4.1. Agar well diffusion method Twenty milliliters of fluid BHI agar were dispensed into the required number of well plates and allowed to solidify. Then, 100 μ L of the previously cultured bacteria suspension was measured and spread onto individual BHI agar plates to form a lawn of bacterial growth. A sterile punch measuring 5 mm in diameter was used to create three wells in each agar plate, into which freshly prepared cement was filled. A filter paper disc impregnated with 2 % chlorhexidine gluconate (CHX) was also placed on the same plate to act as a positive control. The plates were incubated at 37 ◦C for 48 h. The antimicrobial activity of the experimental CSCs was determined by measuring the zone of inhibition diameter (mm) with the aid of ImageJ (Wayne Rasband, National Institute of Mental Health, NIH, MD, USA). Two measurements representing the widest and narrowest dimensions of the halo were taken for each specimen, yielding six measurements per sample group (n =6). 2.4.2. Optical density measurements through indirect tests Cement was mixed and placed in a single layer on the side wall of a 96-well plate held perpendicularly. This was done in 4 wells, representing A wells (Fig. 1) for each sample group. The setup was transferred to a sterile cabinet and sterilized by UV light for 60 min. Concurrently, a stock solution with an ideal OD of 0.015 =1 ×10 7 CFUs/mL was prepared from the bacteria suspension grown overnight. 10 μ L of bacterial suspension from the stock solution was carefully placed onto the set CSC material and allowed to dry for 30 min, after which 245 μ L of BHI broth was added into the well and shaken gently for 2 min. Fifteen microliters of the BHI broth from each of the A wells was transferred to 4 corresponding wells, which contained 215 μ L of fresh BHI medium and were used for the indirect contact tests (n =4). Four wells containing the bacteria suspension without cement coating served as a positive control. Additional wells, with cement coating but filled with bacteria-free BHI medium, were also used as controls for turbidity arising from cement degradation. OD was measured at 600 nm hourly or every 2 h for up to 12 h. The experiment was repeated with cement aged for 24 h. From the recorded OD data of the experimental cement samples co-cultured with bacteria, the average OD of the matching bacteria-free wells was subtracted to determine the final turbidity values of the experimental CSCs. The mean absorbance values of all the OD measurements taken throughout the 12-hour evaluation period were also computed from the four replicates of each CSC multiplied by the 13 time points (n =52). 2.5. Cell culture studies Two methods were used to assess cell viability. An initial indirect assessment using cement extract of different concentrations was performed, followed by a direct assessment using the transwell approach (Falcon, 8 µm pore size, Corning, NY, USA). Dental pulp stem cells (DPSCs; Lonza; Cat no. PT-5025) that had been sub-cultured until passage seven and maintained by established protocols were utilized. Cell culture flasks measuring 75 cm 2 were used to grow the cells in a humidified environment at 37 ◦C and 5 % CO 2 . Alpha minimum essential medium ( α -MEM) supplemented with 15 % fetal bovine serum (FBS) and 1 % penicillin/streptomycin was selected. Unless otherwise stated, all the reagents were used as purchased from Gibco® (Thermo Fisher Scientific, CA, USA). 2.5.1. Cell viability by extract method Initially, a non-cytotoxic concentration for indirect cell study was determined by co-culturing different concentrations of eluates (100 %, 50 %, 25 %, and 12.5 %) from the CSCs with DPSCs for 24 h. To prepare the sample eluates, pre-weighed powders for the three experimental cements were autoclaved at 160 ◦C for 2 h while the respective reactant fluids were filtered using 0.22 µm filters. The cement pastes were prepared in an aseptic environment using sterile instruments. After mixing the cement on a glass slab, the paste was filled into a sterile 1 mL syringe, and 0.3 mL of cement was dispensed to the bottom of a 12-well plate (Corning). A sterile disposable plastic spatula was used to compact and spread the cement until it was evenly distributed and covered the entire surface of the well bottom. For each sample group, three fresh cement wells were prepared and incubated for 30 min after the setting time. Three mL of α -MEM supplemented with 1 % Penicillin- Streptomycin, without fetal bovine serum, were added. The setup was incubated at 37 ◦C, 5 % CO 2 , and 95 % air in a humidified atmosphere for 24 h, allowing the medium to interact with the freshly mixed cement. The eluates were then collected, filtered, and diluted to assess cell viability. To determine the metabolic activity of the DPSCs in response to the four extract concentrations, the alamarBlue® (Invitrogen) assay was performed (n =9). For each specimen, 1 ×10 4 DPSCs were seeded in 100 μ L of culture medium in 96-well plates and incubated for 24 h to allow cell attachment. The culture medium was then replaced with extract dilutions (100 %, 50 %, 25 %, and 12.5 %) and incubated for 24 h. The extracts were replaced by 100 μ L of AlamarBlue® solution (1:10) in a serum-free medium. After a further 3 h of incubation, the formation of resorufin was quantified by measuring fluorescence using a microplate reader at excitation and emission wavelengths of 560 nm and 590 nm, respectively (SpectraMax iD3, Molecular Devices LLC, CA, USA). Cells on the culture medium were used as a negative control, and a blank control of the culture medium and no cells were also included. Cell viability values were presented in percentage using the negative control as 100 %. Fig. 1. Schematic of the experimental setup for the indirect contact tests (top) and the 96-well plates (bottom). H.O. Simila et al. Dental Materials 40 (2024) 1832–1842 1835 2.5.2. In vitro analysis by transwell method Cement discs measuring 6 mm in diameter and 3 mm in height were produced from silicone molds and incubated for 1 h to set in a humidified environment. Before evaluation, the samples were sterilized by UV light on each side for 1 h. Concurrently, 3 ×10 4 DPSCs were seeded in 24-well plates (Corning) and incubated for 24 h. The culture medium was then replaced before a transwell insert containing individual cement discs (n =10) was placed inside the well and cultured for up to 21 days before carrying out subsequent assays. During the experiments, the culture medium was refreshed every two days. 2.5.3. Direct cell viability and morphology Cell viability was evaluated after 1, 3, and 7 days using the alamarBlue® assay as previously described. Cell morphology, focusing on the cytoskeleton and nucleus, was investigated after 1 and 2 days of exposure of the DPSCs to the CSCs. Specific manufacturer protocols were followed by staining the cell’s nucleus with 4 ′ , 6-diamidino-2-phenylin- dole (DAPI) (Thermo Scientific, CA, USA) and actin cytoskeleton with ActinRed 555 (Invitrogen). Representative images were captured by direct fluorescence microscopy (Echo Revolve, BICO Company, CA, USA) at 10 ×magnification. The fluorescence images were analyzed qualitatively. 2.5.4. Mineralization assay Induction of mineralized matrix formation by the CSCs (n =8) was evaluated after 14 and 21 days by Alizarin red staining. Two control groups were used: basal media as negative control and odontogenic medium (0.1 μ M dexamethasone, 10 mM glycerophosphate, and 50 μ g/ mL ascorbic acid) as positive control. For the CSCs, a transwell insert containing the representative CSC disks was placed in 24-well culture plates (Corning) and maintained with a basal medium, which was refreshed every two days. At the time point of interest, the cement +insert was removed from the wells, and the cells were fixed using ethanol (70 %) and stained with 40 mM of Alizarin Red solution (ARS; pH 4.2; Sigma-Aldrich) for 20 min under dynamic conditions to stain the calcium deposits. Next, the unabsorbed staining solution was aspirated, and distilled water was used to rinse out any remaining ARS three times. Microscopy and visualization of the mineral deposits at a magnification of 4 ×was performed using a light microscope (Echo Revolve). Subsequently, the mineralized matrix was solubilized using a 10 mM solution of Cetylpyridinium chloride (pH 7.0) (Sigma-Aldrich). 100 μ L aliquots were transferred into 96 well plates in duplicates (Corning), and the absorbance was measured at 570 nm (SpectraMax iD3). The cells maintained with basal medium and odontogenic medium without CSCs were used to compare the influence of the CSCs on mineralization, with the basal medium acting as the control and reported as 100 %. 2.6. Statistical analysis Both descriptive and inferential data analyses were performed. For the contact tests on E. faecalis, the median value for the OD measurements at each time point was presented in the form of line plots. Mean and SD values of the inhibition zones, turbidometry measurements for the 12-hour evaluation period of the microbial contact tests, cell viability assays, and mineralization assays were computed and subjected to ANOVA with Tukey’s post-hoc test (p <0.05). Before this, data was submitted to normality and homoscedasticity tests. The latter results were presented in the form of column graphs. 3. Results 3.1. Particle synthesis Both powders obtained after calcination were white. SEM and TEM imaging results confirm the irregularity and agglomeration associated with CS particles, while the MBGNs possess a spherical shape with surface mesopores (Fig. 2). According to the EDX data, the MBGNs of the SiO 2 -CaO (nominal composition) contain a higher Si content. In contrast, CS particles, including tricalcium silicates (3CaO.SiO 2 ) and dicalcium silicates (2CaO.SiO 2 ), possess a higher Ca proportion, which is reflective of their stoichiometry. The particle sizes and analysis of the relevant Si-O-Ca, O-Si-O, and Si-O-Si vibrations in the FTIR spectra of both synthesized particles and pertinent peaks of the XRD patterns of the CS have been extensively reported in our previous work, where the size of the MBGNs was confirmed to be 172 ±21 nm, while the CS particles formed micron-sized agglomerates of NPs whose dimensions ranged between 115–240 nm [29]. 3.2. Characterization of the setting reaction The length of time required to manipulate the experimental CSCs is depicted in Fig. 3. CS+H 2 O and CS+colH 2 O had the shortest mixing times of 27 and 28 s, respectively, while that of CSmbgn+colH 2 O was 68 s. Setting time ranged from 8 min to 23 min and was the shortest for CSmbgn+colH 2 O and the longest for pure cement (CS+H 2 O). Only the mixing (p <0.0001) and setting (p <0.0001) times of CSmbgn+colH 2 O differed from the other two CSCs in a statistically significant manner. When considering the FTIR data of the setting CSCs, silicate peaks associated with the 400–1300 cm −1 region are almost standard across all three cements for the duration of the monitoring. The split peaks in the 800–1200 cm −1 region correspond to the stretching vibrations of Si- O bands with potential overlap of some carbonate bonds due to exposure to air [30]. The peak at around 484 cm −1 can be assigned to either dimeric or oligomeric silicate species as hydration occurs [31]. For all cements, the broad vibrational OH peak that spans the 2800–3800 cm −1 region develops a small peak at around 3640 cm −1 . A similar peak appearance in this region has been attributed to a hydrated CS compound called Jennite. Calcium hydroxide Ca(OH) 2 is also associated with this vibration. XRD analysis clearly shows a reduction in the peak at 28.2◦and a concurrent appearance and increase in the size of the peak at 29.3◦. These peaks likely correspond to the metastable forms of calcium silicate. Another peak at 18◦associated with Ca(OH) 2 is present in small amounts soon after mixing and within the first hour but disappears at later time points, except in the unmodified CSC, which is present even after 4 h (Fig. 4). 3.2.1. Antibacterial Inhibition zone by agar diffusion method The agar diffusion test results depicted in Fig. 5A reveal that, although all samples formed a halo, it was obscured by the opaqueness resulting from cement diffusion in all the CSCs. According to the graphical representation of these results, although CS+H 2 O and CS+colH 2 O inhibited the growth of E. faecalis, similarly to the CHX control, the antimicrobial effectiveness was statistically lower for CSCs modified with collagen and MBGNs (p <0.0001). Since most of these experimental materials resulted in visible diffusion in the agar medium, the inhibition zone outcome was cautiously interpreted and supplemented by the turbidity measurements reported in the next section. 3.2.2. Optical density measurements The additional testing by indirect contact tests and real-time measurement of OD performed for both freshly mixed cement and those aged for 24 h are summarized in Fig. 5B-C. All the experimental cements inhibited bacterial growth to various extents. The CS+colH 2 O had the greatest inhibitory effect for the freshly mixed cement, even though this effect was not of statistical significance. More specifically, Fig. 5C, demonstrates that in the initial 2 h of the test, all the fresh cements have a uniform inhibitory effect. The similarity is maintained for CS+colH 2 O and CSmbgn+colH 2 O for up to 7 h before their distinct ranks develop from the 8th to 11th hour. In the aged CSCs, the unmodified CS+H 2 O consistently has fewer viable bacteria throughout the exponential phase of the bacteria growth. These results are mirrored by the results of the statistical analyses performed on the cumulative OD readings presented H.O. Simila et al. Dental Materials 40 (2024) 1832–1842 1836 in Fig. 5D, revealing that the developed CSCs inhibited the growth of E. faecalis, whereas aging for 24 h seems to result in greater potency of CS+H 2 O and CSmbgn+colH 2 O. In general, testing the cement when freshly mixed or after 24 h did not alter the antimicrobial activity in a statistically significant manner (p =0.0082). Additionally, the interaction between the CSCs and the testing time points was insignificant (p =0.2411). 3.2.3. Indirect cell viability by extract method The effect of the experimental collagen-containing CSCs on DPSCs is summarized in Fig. 6A. All of the cements resulted in improved cell viability at 100 % concentration relative to the control (DPSCs in culture medium) (p <0.0001). Subsequent dilutions resulted in the maintenance of cell viability comparable to that of the control group, and the highest viability was observed in the CS+colH 2 O. The CSCs containing collagen or collagen combined with MBGNs presented increased cell viability compared to the control group but with no statistically significant differences between these compositions. 3.2.4. Direct cell viability by transwell method Tests of viability by the transwell method showed the combined effect of the MBGNs and collagen on cellular activity, as assessed by the alamarBlue® assay. After 1 and 3 days of incubation, incorporating the additives did not impact cell viability compared to the control. However, at 7 days, markedly lower viability is observed for the reference CSC that reacted with water only and the CSC that contains MBGNs and reacted with RCP solution Fig. 6B. Across all samples, cell viability increases at 3 days. It remains relatively the same even after 7 days except for the reference CSC, and the CSC modified with collagen and MBGNs, which decreased cell viability (p <0.0001). 3.2.5. Cell morphology The cell morphology appears almost similar for all the cultured cells, thus suggesting that the different compositions of CSCs allowed cell Fig. 2. Results of the FTIR, XRD, SEM, and TEM-EDX characterizations of the two particles synthesized by the sol-gel method. SEM micrographs reproduced from reference [29] under a Creative Commons Attribution License (CC BY). Fig. 3. Mixing and initial setting times of the experimental cement. ANOVA/Tukey post-hoc; n =4; p <0.05. H.O. Simila et al. Dental Materials 40 (2024) 1832–1842 1837 spreading and adhesion after 1 or 2 days of incubation. This behavior agrees with the quantification of mitochondrial activity by alamarBlue®. 3.2.6. Mineralization assay Despite the similarities observed across the CSC samples regarding cell morphology and viability, significant differences in mineralization at 14 and 21 days were found (Fig. 7A). At the 2-week timepoint, all cements showed statistically higher mineralized nodule formation than the control and osteogenic medium. This pattern was further enhanced after 1 week in a statistically significant manner (p <0.0001). It is notable that during 14 days, the inclusion of MBGN into the CSCs did not improve the ability to mineralize, whereas, at 21 days, the unmodified CSC and CSCmbgn+colH 2 O emerged as the best performing in terms of mineralized matrix formation. However, although relative differences could be highlighted at 21 days among the tested formulations, all 3 CSCs exhibited a lower mineralized matrix formation than the osteogenic medium (positive control). The quantitative results are further echoed by the appearance of the mineralized nodules captured by light microscopy in Fig. 7B. 4. Discussion This study investigates the in vitro characterization, setting Fig. 4. Real-time setting data captured by FTIR (left) and XRD analyses (right). The pink band on the FTIR diagram corresponds to typical split peaks of Si-O-Ca and Si-O-Si in CS. Selected bands and peaks of interest are described in the text. H.O. Simila et al. Dental Materials 40 (2024) 1832–1842 1838 behavior, antibacterial and cellular effects of novel inorganic calcium silicate cements (CSCs) containing collagen and mesoporous bioactive glass nanoparticles (MBGNs) for endodontic applications, motivated by the need to sustain good handling, physicomechanical, and antibacterial properties of the CSCs formulated from particles synthesized by the solgel method. Sol-gel has become a popular route for producing bioactive glass nanoparticles due to the ability to control the composition and morphology of the desired particles [32]. The 10 % RCP was preferred over native collagen since the properties from such a commercial source are standardized, and recombinant technology ensures the product remains unencumbered with the risk of interspecies disease transmission [33]. The concentration was arbitrarily selected because of the desire to maintain a low-viscosity fluid reactant with Newtonian properties similar to distilled water. The reference CSC+H 2 O exhibited setting properties comparable to those of its commercial counterparts. Setting times range among available products from 15 min to 2 h and 45 min [3]. Additives like methylcellulose, tricalcium aluminate, strontium salts, and calcium chloride have resulted in faster setting times, while calcium sulfate has been linked to slower reactions and extended setting times [14]. In our case, the rheological assessments revealed that the inclusion of MBGNs in the experimental CSCs significantly prolonged the mixing time, likely because of difficulties in adequately moistening the cement admixture before adequate mixing, with CSmbgn+colH 2 O exhibiting the longest mixing time among the materials tested. The reduced setting time of the CSCmbgn+colH 2 O can be rationalized by the nucleating effects of the nanosized MBGNs, lowering the pH and accelerating the hydration reaction [34–36]. These results contrast those reported for Biodentine™, whose setting time was prolonged by adding BG, possibly due to the larger and less soluble melt quench glass used in that study [37]. Collagen addition in itself did not affect the mixing and setting times observed. Real-time FTIR monitoring confirmed silicate peaks characteristic of calcium silicate hydrate (CSH) [31] and modest Ca(OH) 2 formation. At the same time, XRD indicated potential transformation of CS phases, highlighting the metastable nature of reactive species during CS hydration and hardening [30,31,38]. It must be pointed out that the study’s evaluation of setting by determining the mixing and setting times while monitoring the chemical changes in real-time is comprehensive and should be encouraged for novel hydraulic cement compositions. Notably, limiting the chemical monitoring to 6 h was inadequate. Furthermore, frequent analysis during the first hour could have better elucidated any differences among the CSCs. Given the prevalence of E.faecalis in root canal reinfections and the difficulty of eradicating it, this evaluation emphasizes the importance of inhibiting this infection in endodontic procedures [39]. However, this should not diminish the importance of testing the antimicrobial effects against other microorganisms in dental infections. Specifically, cariogenic bacteria like Streptococcus spp. must be inhibited by cements used in pulp therapy, as caries are the primary precursor to pulpal infections [40,41]. Moreover, Porphyromonas gingivalis and Prevotella intermedia that cause periodontal infections should also be studied in light of the possibility of endo-perio lesions [42]. In the present study, two evaluation approaches were used to assess antibacterial potential since this combination addresses the limitations inherent in any single method, ensuring a comprehensive assessment [43,44]. For example, a material that is insoluble or whose active molecules cannot diffuse through agar may be interpreted to lack inhibitory effects using the agar diffusion test (ADT). Yet, the contrary may be observed in contact tests, which are not Fig. 5. (A) The antimicrobial capabilities of the experimental CSCs were determined by the agar diffusion method using chlorhexidine (CHX) as the positive control. ANOVA/Tukey’s post-hoc; n =6; p <0.05. (B) Comparative effect of the experimental CSCs, when freshly mixed or aged for 24 h, on the growth of E. faecalis for 12 h. Optical density results of the indirect contact test comparing the performance of the freshly mixed (C) CSCs and those aged for a day (D). Different letters indicate a statistical difference from each other. ANOVA/Tukey’s post-hoc; n =52; p <0.05. H.O. Simila et al. Dental Materials 40 (2024) 1832–1842 1839 dependent on material solubility or diffusivity [45]. The current study reports that all the cements fell within the 10.00–13.00 mm range, confirming that the MBGN additives did not impact the antibacterial effectiveness of the experimental CSCs, especially when compared to the antibacterial effectiveness reported for commercial alternatives like MTA Angelus (13.16 mm), and Biodentine™ (14.66 mm) in this testing strategy [6,46]. The antibacterial effects observed among the freshly mixed cement are consistent with those of MTA, a CSC type of cement that causes an increase in pH of up to 12.9 during the initial 3 h of setting. This alkalinity is above the 11.5 pH that is known to curtail the survival of E. faecalis [7,47]. Whereas the CS+colH 2 O sample had higher efficacy according to the indirect contact test with fresh cement, unmodified CSC emerged top when aged for 24 h and in the ADT. Discrepancies in antibacterial results between the different CSCs are not unusual, and several authors have also reported such mismatches, which can be attributed to the material chemistry, sample topography, active antibacterial agent, mechanism of action, microorganism tested, test method applied, contact time and sample handling. Detection of Ca(OH) 2 in the setting cements by FTIR and XRD confirms the expected setting reaction and correlates with the antibacterial effects and cell viability. Generally, the toxic effects on E. faecalis can be attributed to the Ca(OH) 2- -induced alkaline pH [48]. Elevated pH disrupts the bacteria membrane, inhibits enzymatic activity, and interferes with cellular processes [7]. Protein denaturation and damage to both DNA and the cytoplasm also explain these effects [49]. In addition to the high pH environment, Ca 2+ ions released from the setting CSCs and the Ca(OH) 2 are also destructive to E. faecalis as they may adhere to the cell membrane of the bacteria, chelate essential nutrients required by the microorganism and alter the osmotic pressure [50,51]. Overall, evaluating the microbial effect using three different modalities provides robust data. However, this study’s limitations include not testing the effect on other microorganisms or monitoring actual pH changes. Additionally, placing the CSCs on the sidewall of the wells for OD monitoring may have contributed to the inconsistent results across different antimicrobial evaluations. Cytotoxicity assays were conducted to understand further the potential of CSCs in dentin-pulp regeneration and mineralization. These assays confirmed that mineralization occurred in all the CSCs, with minimal improvements by collagen and inclusion of MBGNs. The amount of RCP and MBGNs added to the CSCs did not elicit statistically significant differences, especially during the initial days of testing cell viability by the transwell method. The modest enhancement by CS- colH 2 0 after 7 days may arise from collagen’s stimulation of bioactive cues [52]. The ARS assay indicated that after 14 days, the addition of MBGN or RCP did not significantly enhance the apatite-forming capabilities of the reference CSC. Generally, the presence of calcium and silicon in all CSCs plays a pivotal role in driving cell metabolism, proliferation, gene expression, and mineralization processes [53,54]. Still, the mechanisms through which Ca 2+ does this are better understood and well-elucidated [53,55]. The calcium hydroxide detected during the setting correlates to the Fig. 6. A) Indirect cell viability by alamarBlue® assay representing eluate effect on DPSCs at 4 concentrations during a 24-hour evaluation period. Dissimilar letters indicate a significant difference between cement samples. ANOVA/Tukey’s post-hoc; n =9; p <0.05). B) Direct cell viability (Transwell method) by alamarBlue® assay at 1, 3, and 7 days. Different small letters indicate significant differences between groups within each time point. ANOVA/Tukey’s post-hoc; n =10; p <0.05). C) DPSCs morphology by fluorescence microscopy after 1 and 2 days of indirect contact with CSCs. Cell nuclei are stained in blue (DAPI), and actin filaments are stained in red (ActinRed 555). The scale bar represents 200 µm (10 ×magnification). H.O. Simila et al. Dental Materials 40 (2024) 1832–1842 1840 cell biology results since Ca(OH) 2 produces an alkaline environment that promotes the precipitation of calcium phosphate (Ca-P), thus explaining the mineralized nodule formation of our experimental CSCs [56,57]. Moreover, it has been established that gene expression of DSP, RUNX, OPN [54], and appropriate biomolecules such as fibroblast growth factor 2 (FGF-2) and alkaline phosphatase (ALP), an enzyme that induces proliferation, are upregulated by Ca 2+ [58,59]. This cascade of biological activities is linked to the influx of Ca 2+ into the cell, which activates non-specific Ca 2+ channels that direct odontogenic differentiation. 5. Conclusion The present study focused on validating CSCs produced by the sol-gel method and examining the effect of RCP combined with MBGNs on the antibacterial effect against E. faecalis and the mineralization potential of DPSCs, showing that cell viability was not fundamentally impaired, mineralization was best for the reference cement and the CSmbgn+colH2O, and antibacterial activity did not compromise the regenerative potential, thereby verifying the utility of sol-gel synthesis as a green approach for producing bioactive NPs for CS-based endodontic cements. Acknowledging study limitations, future work varying MBGNs and collagen volumes would clarify optimal amounts for improvements to cements. Investigating long-term degradation, sealing, microleakage, and antimicrobial testing with biofilms on tooth substrates will offer deeper insights into clinical scenarios. CRediT authorship contribution statement H.O.S, M.C.B, and A.R.B conceived the presented work. A.M.B supervised the collection, analysis, and interpretation of TEM-EDX data. C. A., L.M.C, R.D.F, and H.O.S performed all experiments and analyzed data. H.O.S: writing—original draft preparation. C.A., L.M.C, R.D.F: writing—review and editing. M.C.B and A.R.B provided the overall resources and support needed to undertake the study. M.C.B and ARB supervised the overall work. All authors revised the manuscript and contributed to the final version. All authors have read and agreed to the published version of the manuscript. Acknowledgments Hazel Simila is grateful to the Kenyan-German Postgraduate Training Program for providing a Deutscher Akademischer Austauschtdienst (DAAD) scholarship (Funding Program Number. 57460842) for the pursuit of doctoral studies in Germany. Doctoral study leave was granted by the University of Nairobi, Kenya. We also thank the central research services CITIUS of the Universidad de Sevilla (Spain) for the use of the BFTEM facilities. We extend our gratitude to Dr. Kai Zheng of Nanjing Medical University, China, for the donation of the collagen powder and Prof. J. Christopher Fenno of the University of Michigan for the use of the microbiology facilities. References [1] Guivarc’h M, Jeanneau C, Giraud T, Pommel L, About I, Azim A, et al. An international survey on the use of calcium silicate-based sealers in non-surgical endodontic treatment. Clin Oral Invest 2020;24:417–24. https://doi.org/10.1016/ j.euf.2020.05.015. [2] Palma PJ, Ramos JC, Martins JB, Diogenes A, Figueiredo MH, Ferreira P, et al. Histologic evaluation of regenerative endodontic procedures with the use of chitosan scaffolds in immature dog teeth with apical periodontitis. J Endod 2017; 43:1279–87. https://doi.org/10.1016/j.joen.2017.03.005. [3] Primus CM, Tay FR, Niu L-N. Bioactive tri/dicalcium silicate cements for treatment of pulpal and periapical tissues. Acta Biomater 2019;96:35–54. https://doi.org/ 10.1016/j.actbio.2019.05.050. Fig. 7. A) Quantification of mineralized matrix formation by Alizarin red staining after 14 and 21 days. Different letters demonstrate significant differences between sample groups within each time point. ANOVA/Tukey’s post-hoc; n =10; p <0.05. B) Mineralized matrix formation (Alizarin red stain) after 14 and 21 days of indirect contact with experimental CSCs. Oste. Med.=osteogenic media. H.O. Simila et al.