MTA promotes the proliferation and survival of human DPSCs, bone marrow stromal/stem cells (BMSCs) and PDLSCs via the ERK signaling pathway (25), and also exhibit a positive effect on viability of human DPSCs (41). MTA at high concentrations (20 and 10 mg/ml) is toxic to human DPSCs but MTA at low concentrations (2.0, 1.0, 0.2 and 0.1 mg/ml) enhances viability of human DPSCs but has no effect lower concentrations (0.020 and 0.002 mg/ml) (42). Similarly, undiluted MTA extract slightly increase survival of human DPSCs, while 1/2 and 1/4 dilutions of MTA extract have no effect on cell viability (43). In addition, MTA at high concentrations (20 mg/ml) decreases proliferation of DPSCs under inflammatory conditions but has no effect at low concentrations (0.020 and 0.002 mg/ml) (44). Moreover, various commercial MTA extracts, such as Angelus MTA (Angelus Dental Products Industry) and Root MTA (University of Tabriz, Iran), show similar effects on human DPSCs but were more biocompatible when applied at lower concentration (1:2) and longer exposure times compared with MTA. These results suggested that the biocompatibility of MTA is dependent on not only dosage but also exposure time (45). In terms of time-dependent biocompatibility of MTA, the cytotoxicity of MTA decreased over time and the viability and proliferation of human DPSCs increased following two aging cycles, which further supported the aforementioned time- and concentration-dependent effects of MTA (46). Moreover, the proliferation and viability of DPSCs decreased significantly when in direct contact with MTA for the first day but subsequently raised after three days (47). The initial cytotoxicity of MTA to growth and viability of DPSCs (47-49) may be partly ascribed to the relatively rough surface of biomaterials (50) or leakage of components such as bismuth oxide (51) and Al (52) and Si ions (53). Higher levels of Si ion concentrations from the SiO₂ phase of materials may lead to hyperosmoticity and subsequently stimulates production of inflammatory cytokines (54). Furthermore, production of MTA during the hydration reaction and its exposure concentration may result in early slight cytotoxicity of MTA. Calcium hydroxide is produced when calcium silicate contained in MTA reacts with water and increases pH of the culture media (55,56). Basic pH environment and release of inorganic salts induced by high concentrations of MTA significantly decreases cell proliferation (57). Furthermore, 24-h set MTA promotes viability of human DPSCs, whereas 1-h set MTA exhibits an inhibitory effect after 7 days, which suggests that incompletely solidified MTA is cytotoxic (58). The subsequent rise in cell viability may be caused by hydroxyapatite layer formation on the hydrated bioceramic surface (59) controlled continuous production of calcium, silica and phosphate ions from bioceramics along with alkaline pH in the later stage (60,61). Regarding differentiation of human DPSCs, in the first week following induction with MTA in odontogenic differentiation medium, DPSCs began to change from spindle to rounded shape upon reaching confluency and then migrated to form clusters (62). DPSCs began to differentiate at day 7 (63) and mineralization was observed by Alizarin Red staining until day 21. Combination of MTA and odontogenic differentiation medium enhanced odontogenic differentiation of DPSCs but MTA extract-alone did not induce this (62). Treatment with MTA for 1 day affected more genes in uninduced DPSCs than in DPSCs induced by odontogenic differentiation medium, which suggested that MTA exhibits a greater stimulative effect on odontogenic differentiation of uninduced DPSCs compared with induced DPSCs (64). In addition, MTA at a concentration of 0.2 mg/ml displayed the strongest capacity to induce odontoblastic differentiation of human DPSCs via the p42/44 ERK signaling pathway (42). MTA-conditioned medium at the same concentration enhances the odonto/osteogenic capacity of DSPCs from inflammatory sites by activating the NF-κB pathway, as shown by significantly upregulated odonto/osteoblastic gene expression levels, such as ALP, RUNX2, OSX, OCN and DSPP (44). Consistent with the changes in genes associated with osteo/dentinogenic differentiation, MTA promotes mineralized nodule formation of human DPSCs (65,66). Increased secretion of angiogenic factor VEGF has been detected in human DPSCs induced by MTA (63), which in turn affects viability and function of DPSCs (67). MTA contributes to dentin bridge formation in endodontics (68,69). In addition, MTA is applied in pulp capping because of its excellent bioactivity, which has been confirmed by the elongated shape of DPSCs, formation of collagen fibers and calcified deposition in the presence of MTA in a model simulating indirect pulp capping (70). In previous studies, accelerants, including 5% CaCl₂ and 2.5% Na₂HPO₄, have been mixed with MTA to shorten setting time. Compared with MTA in the presence of distilled water, MTA in the presence of 5% CaCl₂ and 2.5% Na₂HPO₄ is more biocompatible and exhibits greater ability to promote odontoblastic differentiation of DPSC niches (48,71). Propolis, a natural alternative endodontic material produced by honeybees from tree resin, also enhances the ability of MTA to promote odontogenic differentiation and mineralization of DPSCs via the ERK pathway (72).

Compared with human PDLSCs and tooth germ stem cells (TGSCs), human DPSCs exhibit better viability in the presence of both Biodentine and MTA (73). Moreover, Biodentine displays a superior ability to promote viability, adhesion and migration of human DPSCs compared with MTA. Human DPSCs spread on the surface of Biodentine show a spindle, polygonal and flattened morphology (74). Similar to MTA, high concentrations of Biodentine extract (20 mg/ml) exhibit slight cytotoxicity, whereas 0.2 mg/ml Biodentine enhances biological behaviors of human DPSCs, including cell proliferation, viability, migration, adhesion and mineralization formation. In addition, low concentrations of Biodentine (0.2 mg/ml) promote odontoblast differentiation and biomineralization of human DPSCs by activating ERK1/2 and JNK and attenuating the NF-κB pathway (75,76). Increased Alp activity and dentin matrix protein expression levels have been observed in human DPSCs stimulated with Biodentine (75-78). Furthermore, Biodentine significantly increases calcium deposition (79) and enhances the production of Ocn and Runx2 in human DPSCs stimulated with lipopolysaccharide (LPS) although there is no change in ALP expression levels (53). In terms of the inflammatory response, Biodentine does not affect high expression of IL-6 and IL-8 in DPSCs induced by LPS stimulation but decreases levels of the anti-inflammatory cytokine TGF-β1 (53). Although Biodentine medium is biocompatible with DPSCs, when in direct contact with DPSCs, Biodentine exhibits a slight toxic effect and delays closure of wound edges, which implies that direct contact between Biodentine and DPSCs leads to cell death or decreased proliferation (80,81). However, unaltered expression levels of actin, tubulin and vimentin indicate that Biodentine neither induces apoptosis, inflammation and genotoxicity nor impairs cellular architecture. The inhibitory effect of Biodentine on cell proliferation and migration may result from the decreased space for cell growth in a direct culture model and prolonged cell doubling time (80). Similarly, other studies have observed that the viability of human DPSCs in direct contact with Biodentine is initially decreased (49,77), which may be due to substantial calcium ion released from Biodentine in the first 3 h (82). Relatively large amounts of Ca and Si ions and the absence of Sr, Al and S in Biodentine extract contribute to the biocompatibility of Biodentine (74). Considering the increased release of Ca and Si ions (83) and microstructural (84,85) changes induced by acidic conditions, human DPSCs cultured in Biodentine stored in acidic environment display more spindle-shaped formation and higher adherent cell density compared with that in Biodentine stored in saline (58) Similar to MTA, Biodentine also increases gene expression levels of osteogenic and odontogenic markers, such as OPN and DSPP in human DPSCs when in direct contact with DPSCs (49,77). However, Biodentine promotes odontogenic differentiation of DPSCs more significantly than osteogenic differentiation, as indicated by detection of expression levels of odontoblastic marker DSPP and osteogenic gene markers ALP, COL1A1 and OPN (80). Moreover, Biodentine promotes biomineralization and secretion of extracellular mineral matrix in human DPSCs cultured with osteogenic medium for 21 days (77) and induces more mineralized nodules in the osteogenic medium compared with MTA, suggesting that increased calcium ion release, along with a neutral pH, promotes differentiation and mineralization of DPSCs and subsequently generates a greater number of structured dentin bridges (82). Similarly, when DPSCs are cultured with three-dimensional models, Biodentine induces higher viability compared with MTA. Furthermore, expression levels of COL1A1, ALP and DSPP in DPSCs on MTA and Biodentine are initially upregulated significantly and then decrease gradually until day 21; however, expression of RUNX2 in three-dimensional cultures remains lower than that in control group (86). The reason for this may be that COL1A1, ALP and DSPP are associated with initiation of dentinogenesis and mineralization. Low expression levels of RUNX2 contribute to odontoblast differentiation and cell maturation, whereas increased expression is observed during terminal odontoblast differentiation (87,88). MTA and Biodentine stimulated angiogenesis by improving the expression levels of VEGF in human DPSCs on day 14 (49). VEGF enhances the proliferation, migration, and tubulogenesis of endothelial cells close to microvessels, which regulates both vascularization and angiogenesis. Angiogenesis is a key factor for wound healing and tissue regeneration of damaged dental pulp (89-91). The pro-angiogenic capacity of bioceramics depends on dissolution products, such as Si, Mg and Ca ions, which induce secretion of angiogenic factors from cells. Dissolution of calcium ions and subsequent precipitation reactions on the surface of bioceramics leads to vascular penetration and osteoblastic differentiation. Interconnections and pore size of the scaffold also influence the size and amount of the blood vessels, which are necessary for the vascularization of bioceramic material (92).

iRoot BP Plus promotes proliferation, attachment, migration and mineralization of DPSCs compared with MTA (93). Furthermore, iRoot BP Plus releases more Si ions than MTA, which may explain why iRoot BP Plus induces greater apatite formation. iRoot BP Plus and MTA promote stretched and highly organized stress fibre assembly of DPSCs, which is indicative of reorganization of the actin cytoskeleton. Moreover, iRoot BP Plus and MTA enhance phosphorylation of both Paxillin and focal adhesion kinase (FAK) and increase protein expression levels of Vinculin, FAK and Paxillin in human DPSCs (94); this is associated with the formation of focal adhesions (95). Cytoskeleton reorganization and focal adhesion formation is essential for cell adhesion and migration (96). These results confirmed that both iRoot BP Plus and MTA promote attachment and migration of human DPSCs. Likewise, iRoot FS enhances proliferation, migration, and osteoblastic differentiation of human DPSCs (97). Additionally, iRoot FS displays superior ability than Biodentine to promote proliferation and migration of human DPSCs on day 7. However, iRoot FS showed no significant effect on osteogenic differentiation on day 7 (98), which implied that iRoot FS affects proliferation and migration of human DPSCs and later influenced osteoblastic differentiation. A longer experimental observation period should be used to investigate the bioactivity of calcium silicate-based bioceramics on human DPSCs.

MTA exhibits concentration-dependent cellular compatibility with human PDLSCs; MTA at higher dilution exhibits better biocompatibility with human PDLSCs (99). Due to the biological characteristics of MTA, other MTA-based endodontic materials have been developed, including Endoseal MTA (ES; Maruchi Co., Ltd.), Nanoceramic Sealer (NCS; B&L Biotech USA, Inc.), Bioroot BC Sealer (BR; Septodont Holding) and MTA Fillapex (Angelus Dental Products Industry). Although human PDLSCs in the presence of these MTA-based bioceramics maintain high expression levels of MSC markers, including CD105, CD73 and CD90, the capacity of human PDLSCs to migrate, adhere and grow is higher when treated with BR at different concentrations compared with ES and NCS. ES and MTA Fillapex show cytotoxicity to human PDLSCs at 24, 48 and 72 h and give rise to worse cell attachment and spread, which may be caused by tungsten contained in MTA Fillapex (100-102). By contrast, 2 mg/ml MTA extract is the optimal concentration to markedly increase calcified nodule formation, Alp activity and odonto/osteogenic differentiation in human PDLSCs; these effects are mediated by activating NF-κB and MAPK signaling pathways (103). Likewise, 2 mg/ml Biodentine is the most biocompatible concentration to promote migration, attachment, and mineralization of human PDLSCs. Biodentine at low concentrations (2.00, 0.20 and 0.02 mg/ml) significantly enhances viability of human PDLSCs, while Biodentine at higher concentrations (20 mg/ml) exhibits cytotoxicity, which may be associated with high pH (104). High pH of Biodentine results in an increased concentration of iron and calcium ions in the extracellular environment (105). The inhibitory effect of Biodentine at high concentration (20 mg/ml) on the viability of human PDLSCs may be explained by increased or unbalanced ions levels, which generate a toxic effect on cells (106).

BMSCs are used as an in vitro model of MSCs associated with wound healing and tissue regeneration of alveolar bone. MTA promotes the adhesion and proliferation of human DPSCs, PDLSCs and BMSCs, which display an elongated morphology and are well-spread. Moreover, positive effects of MTA on proliferation are cell type-dependent (25). Human PDLSCs and DPSCs rapidly respond to MTA with significantly increased cell number by day 3, whereas human BMSCs stimulated with MTA show slow increase by day 5 (25). Furthermore, MTA promotes migration of human BMSCs significantly after 18 h and cells spread well on the surface of MTA after 24 h. MTA increases proliferation of human BMSCs when incubated in normal medium, whilst the same phenomenon was delayed by 7 days when cultured in differentiation medium (107). When human BMSCs are cultured in osteogenic differentiation medium, MTA does not stimulate osteogenic differentiation by day 10, as indicated by decreased mRNA and protein expression levels of osteogenic markers, such as RUNX2, OSX, MSX2 and OCN (108). A positive effect of MTA on viability of rat BMSCs is observed up to day 7, which is sooner than in human BMSCs. Simultaneously, MTA stimulates rat BMSC differentiation into osteoblast-like cells over 7 days, which is confirmed by enhanced ALP staining and upregulation of gene expression levels of bone morphogenetic protein 2 (BMP-2), ALP, bone sialoprotein (BSP) and osteocalcin (OC) (109). Low concentrations of MTA-conditioned media (2.000, 0.200, 0.020 and 0.002 mg/ml) are not toxic to BMSCs and 0.02 mg/ml MTA is the optimal concentration to upregulate odonto/osteogenic capacity of BMSCs originating from rat mandible; this finding was supported by enhanced Alp activity, calcified nodule formation and expression levels of odonto/osteoblastic genes in BMSCs, such as ALP, RUNX2, OSX, OCN and DSPP. However, MTA at lower dilution inhibits the proliferation of BMSCs (110). Likewise, 1:5 MTA medium decreases viability and osteogenic differentiation of primary BMSCs from the hind limb of mice, as indicated by decreased von Kossa staining and low expression levels of OCN and BSP (111). Furthermore, tricalcium aluminate, a primary composite of MTA, contains aluminum and exhibits significant toxicity to rat BMSCs from femur and tibia bone marrow in vitro, whereas good tissue compatibility has been observed with tricalcium or dicalcium silicate (112). In addition, bismuth oxide in MTA is cytotoxic to human BMSCs (113). Therefore, the negative effect of MTA on the viability of BMSCs at high concentrations may partly be caused by tricalcium aluminate and bismuth oxide. Investigation of the mechanism underlying how MTA affects behavior of BMSCs has demonstrated that the ERK signaling pathway is associated with the positive effect of MTA on proliferation (25) and odonto/osteogenic differentiation of BMSCs. JNK signaling pathway is also involved in the odonto/osteogenic capacity of BMSCs induced by MTA (110). In addition, iRoot BP Plus promotes osteogenic/odontogenic differentiation of BMSCs via the MAPK signaling pathway and autophagy (114).

Bioaggregate is non-toxic to human BMSCs throughout a culture period of 24 h (115). Similarly, Biodentine promotes proliferation of rat (116) and human BMSCs (117,118). Both Biodentine and MTA at high concentrations (1:2) show cytotoxicity to human BMSCs, while Biodentine at low concentrations (1:10 and 1:20) and long incubation periods exhibits an inferior ability to enhance proliferation and osteoblastic differentiation of human BMSCs compared with MTA (119). Moreover, Biodentine and MTA display an inhibitory effect on survival of human BMSCs in a concentration-dependent manner over 7 days but do not affect cell morphology (120). The cytotoxicity of Biodentine to human BMSCs may be ascribed to the calcium chloride contained in its liquid (121), which has been reported to exhibit less biocompatibility when added to MTA compared with MTA mixed with water (122). Therefore, Biodentine prepared following complete setting is more compatible with human BMSCs than MTA (120). High Alp activity with mineral deposits in rat BMSCs have been found in the presence of Biodentine after 12 days (116) but the capacity of Biodentine to induce osteogenic differentiation in human BMSCs is inferior to that of MTA (117,118). MTA exhibits earlier and more pronounced calcium deposits than Biodentine (120), which may be explained by the fact that MTA produces high pH and Alp activity and promotes production of high concentration of Ca ions (123).

SHEDs originate from deciduous teeth and regenerate bone and dentin, but not dentin/pulp-like complexes as human DPSCs do (124). Due to the good porous microstructures in MTA or Biodentine, SHEDs attach and spread well on the surface of MTA and Biodentine, which helps SHEDs maintain mesenchymal properties in the presence of MTA or Biodentine with positive expression of CD105, CD90 and CD73 (125). Moreover, the capacity of SHEDs to adhere and proliferate is enhanced by MTA or Biodentine after 48 h (125). MTA and Biodentine exhibit a comparable ability to promote migration of SHEDs. In addition, the viability and proliferation of SHEDs cultured with 1 mg/ml MTA- or Biodentine-conditioned medium is similar to that of negative control during the whole incubation period, which implies that both MTA and Biodentine at 1 mg/ml are non-toxic to SHEDs (126). However, MTA shows greater potential to promote odontogenic differentiation compared with Biodentine (126), whereas Biodentine has better capability to promote proliferation and calcified matrix deposition in SHEDs than MTA (125,127). Furthermore, Biodentine affects behavior of SHEDs in a concentration-dependent manner. Biodentine at low concentrations (2.00, 0.20 and 0.02 mg/ml) stimulates proliferation, viability and migration of SHEDs, whereas high concentrations of Biodentine (20 mg/ml) exhibit slight cytotoxicity to SHEDs. Changes in the concentration of Biodentine have no impact on the adhesion ability of SHEDs (128). Similarly, Biodentine at higher dilutions (1:16 and 1:32) is more effective in promoting proliferation, odontogenic differentiation and biomineralization of SHEDs, which may be because Sr and Si are gradually released from Biodentine as the concentration of Biodentine decreases over time (129). By contrast, fresh mixed MTA impairs the viability and migration of SHEDs and enhance apoptosis over 7 days. Furthermore, the cytotoxicity of MTA to SHEDs is more apparent when SHEDs directly contact MTA (130). The different results may be associated with the preparation of MTA. For example, freshly mixed MTA is frequently used in endodontics whereas bioceramic eluate or aged bioceramics were used in the aforementioned in vitro studies (86). However, the cytotoxicity of these bioceramics generally decreased as the bioceramic set and pH changed of medium over time. Freshly mixed MTA caused severe damage to cells due to the initially high concentration of calcium hydroxide and subsequent raise in pH to 12.5 after mixing for 3 h (131,132). Although iRoot BP Plus possesses a similar capacity to MTA in terms of SHEDs proliferation, it displays more prominent capacity to enhance adhesion, migration and osteogenesis of SHEDs compared with MTA (93).

Both MTA and Biodentine have been shown to promote the proliferation, odontoblastic differentiation and biomineralization of SCAPs over 14 days (133,134). However, Schneider et al (135) found MTA induces early short-term proliferation of SCAPs over 5 days and promotes the migration of SCAPs after 6 h. Saberi et al (136) discovered that the cytotoxicity of both complete set MTA and Biodentine to SCAPs decreased over time. By contrast, Miller et al (137) revealed that incompletely set MTA inhibits proliferation of SCAPs, whereas Biodentine promotes proliferation of SCAPs. The difference in results may be due to the method of cytotoxicity assessment, contact between cells and material, concentration of material and assessment time points. MTA affects survival of SCAPs in concentration-dependent manner. MTA at lower concentrations (0.02, 0.20 and 2.00 mg/ml) exhibits excellent biocompatibility with SCAPs; however, proliferation of SCAPs is inhibited and normal morphological cells disappeared when treated with MTA at higher concentrations (10 and 20 mg/ml) (138). Low concentrations of MTA or Biodentine (2.00, 0.20 and 0.02 mg/ml) enhance Alp activity and osteoblastic/odontoblastic differentiation in SCAPs, while high concentrations of MTA or Biodentine (20 mg/ml) exhibit a negative effect (138,139). MTA extract enhances the ability of osteogenic medium to induce mineralization and increase expression of mineralization-associated genes, such as Ocn (140). In comparison with MTA, SCAPs treated with Biodentine display greater odontoblastic differentiation, as demonstrated by positive alizarin red staining and expression of genes encoding DMP-1, DSPP, OCN and matrix extracellular phosphoglycoprotein (133,134). Biodentine enhances expression of odontoblast specific marker DSPP, while MTA promotes osteoblastic differentiation of SCAPs by increasing expression of the osteoblastic marker integrin-binding sialoprotein (137). Both MTA and Biodentine enhance the secretion of pro-inflammatory cytokines, such as IL-1α, IL-1β and IL-6 (139), and MTA activates NF-κB signaling pathway, which affects the odonto/osteogenic differentiation of SCAPs (141,142). In addition, p38 and ERK signaling pathways serve an essential role in odontoblastic/osteoblastic differentiation of SCAPs stimulated by MTA (138). MTA and Biodentine enhance the angiogenic potential of SCAPs; these bioceramics promote the expression of angiogenic genes in human SCAPs, such as VEGFA and c-fos induced growth factor (FIGF) (134), which induce endothelial cell proliferation, migration and differentiation, and promote formation of endothelial tubules (143-146).

Compared with MTA, iRoot FS exhibits similar biocompatibility with human SCAPs but possesses markedly stronger capacity to enhance migration and osteo/odontogenesis differentiation of human SCAPs, and mineralized nodule formation via the Wnt/β-catenin signaling pathway (147). iRoot FM at low concentrations (0.5 mg/ml) increases proliferation and osteo/odontoblastic differentiation of SCAPs, whereas there is no marked effect on SCAPs stimulated with iRoot FM at high concentrations (1.0 and 2.5 mg/ml). Moreover, mineralized nodule formation and expression of DMP-1 and ALP are enhanced by iRoot FM compared with Ca(OH)₂. However, iRoot FM at different concentrations has no impact on morphology of SCAPs (148).

TGSCs, a popularized stem cell source derived from wisdom teeth, display MSC properties and can differentiate into endothelial or epithelial cells in dental tissue regeneration (149,150). Consistent with human DPSCs and PLSCs, MTA and Biodentine exhibit no cytotoxicity to TGSCs (73). Nonetheless, viability and odontogenic differentiation of TGSCs are inhibited slightly when in direct contact with MTA, which has been confirmed by decreased numbers of attached cells and Alp activity (151). MTA and Biodentine induce angiogenesis of TGSCs by promoting the release of angiogenic growth factors (platelet-derived growth factor, fibroblast growth factor-2 and VEGF) and enhancing tube formation of human umbilical vein endothelial cells (73). Furthermore, iRoot SP exhibits good biocompatibility with human TGSCs and promote their attachment (152). iRoot SP possesses an inferior capacity to MTA in terms of inducing odontogenic differentiation of human TGSCs and hard tissue deposition; human TGSCs in the presence of MTA exhibit higher Alp activity and enhanced odontoblastic differentiation compared with those in the presence of iRoot SP (153).

Proksch et al (156) found that MTA impairs proliferation, osteogenic differentiation and extracellular matrix mineralization of primary human osteoblasts derived from the alveolar bone; this could be restored by addition of fluoride to MTA. Similarly, MTA inhibits proliferation and COL1 gene expression in bone marrow osteoblasts from rat femur (157). MTA inhibits proliferation and differentiation of rat primary calvarial osteoblasts, as demonstrated by decreased calcified nodule formation and osteoblastic differentiation (158). When primary osteoblasts are cultured with MTA, the highest ionic concentration in areas around MTA results in an inhibitory effect on primary osteoblasts in the central area (158). High dissolution rate of MTA and large release of calcium hydroxide from MTA contribute to highest ionic concentration in areas around MTA (159). Excess calcium concentration induces cell death by apoptosis or necrosis (160), whereas slight increases in extracellular Ca²⁺ concentration (161) and continuous low levels of calcium ion release, along with slower change in pH, promote proliferation and differentiation of osteoblasts (162,163). The decreased viability of primary human osteoblasts induced by MTA may be due to the cumulative effect of the release of toxic components, such as Bi and Al ions, following contact with bioceramic extract (164). These different results may be due to different methods and concentrations of bioceramics used in the experiments. MTA exhibit higher biocompatibility with osteoblasts cultured in three-dimensional culture systems and are attracted toward the material compared with cells cultured directly on materials or in extract. Increased percentage of mature osteoblasts or osteocytes with high expression levels of green fluorescent protein and osteogenic genes, including ALP, BSP and OSX, in primary mouse mandibular osteoblasts demonstrates that MTA promotes differentiation of primary mouse osteoblasts in a three-dimensional cell culture model without differentiation medium, which is required for osteoblastic differentiation in a two-dimensional cell culture model (165,166).

MTA exhibits minimal cytotoxicity to the human osteoblast MG-63 cell line (167). MTA promotes viability and increases mRNA expression levels of COL1 and OCN in MC3T3-E1 cells induced by MTA, accompanied by enhancement of Alizarin Red-S staining and Alp activity, which suggests that MTA promotes bone matrix formation and mineralization of MC3T3-E1 cells (168-171). Moreover, when incubated on the surface of set MTA, a thick mineralized matrix layer covered with multilayered flat cells is observed in MC3T3-E1 cells after three weeks (169). The mechanism underlying MTA-induced osteoblastic differentiation, mineralization and bone matrix formation in MC3T3-E1 cells in vitro may be associated with the activation of activating transcription factor 6 by MTA and the subsequent endoplasmic reticulum stress response (170). Moreover, Ca, Si and P ions released from MTA (157,172) and alkaline pH value of MTA extract (171) contribute to the bioactivity of MTA. Alkaline pH neutralizes lactic acid, which dissolves mineralization of dentin, and promotes formation of hard tissue by activating alkaline phosphatase (173). MTA with high surface pH exhibits cytotoxicity to MG-63 cells initially (174); the inhibitory effect of MTA on viability of MG-63 cells weakens with the setting of MTA (174), which may be due to the fact that the expression of pro-inflammatory cytokines in MG-63 cells are increased by MTA during the first 24 h but decreased over time (167). Growth factor BMPs promote osteoblastic differentiation and biomineralization (175); the mixture of BMPs and MTA attenuates the cytotoxic effect of MTA to MG-63 cells and inflammation response of tissue to MTA (174). MTA promotes the adhesion, spreading, proliferation and collagen secretion in Saos-2 cells (176,177). Moreover, MTA enhances osteogenic differentiation, Alp activity and calcified nodule formation in Saos-2 cell line (178). By contrast, Modareszadeh et al (179) reported that MTA does not affect survival and Alp activity in Saos-2 cell line. The inconsistency between these studies may due to differences in the preparation of MTA, experimental methods and assessment of Alp activity. Bioaggregate shows no cytotoxicity to MC3T3-E1 and markedly increases expression levels of mineral-associated genes, including COL1, OCN and OPN, in MC3T3-E1 cells compared with MTA (180).

Compared with MTA, Biodentine displays good cytocompatibility with primary human osteoblasts, indicated by enhancement of cell viability, attachment and proliferation (164,181). Similar biocompatibility of MTA and Biodentine has been observed with human osteoblast-like cell line MG63; both enhance viability, adhesion and proliferation of MG63 cells, which may be because Biodentine and MTA have similar surface roughness, heterogeneous morphology and particle size (182). Biodentine and MTA both exhibit positive effects on viability and calcification of MC3T3-E1 cells (183). Biodentine and MTA show dose-dependent effects on viability of Saos-2 cells. Biodentine or MTA at lower concentrations (1:4 and 1:8) result in higher viability of Saos-2 cells. Furthermore, Biodentine stimulates proliferation and migration of Saos-2 cells and induces expression of ALP and mineralization (184).

Compared with MTA, iRoot BP Plus induces greater cytotoxicity to primary human osteoblasts but is still considered as biocompatible because cell viability in the presence of iRoot BP Plus remained >70% compared with that in the control group (185). When MTA or iRoot BP Plus is applied in acidic pH conditions, the secretion of Ca and Si ions is enhanced and apatite formation is decreased. However, cell attachment of MC3T3-E1 on these bioceramics is not affected significantly in an acidic environment. Furthermore, MTA decreases viability, whereas iRoot BP Plus increases survival of MC3T3-E1 cells, which suggests that, compared with MTA, iRoot BP Plus may be more suitable as root-end filling material under inflammatory acidic conditions when used in endodontics (186). By contrast, iRoot FS exhibits better biocompatibility with human osteoblast-like MC3T3-E1 and MG63 cells by promoting their attachment and proliferation (187,186). iRoot FS shows better biocompatibility than MTA or iRoot BP Plus because certain toxic metal substances, such as bismuth (189), aluminium and manganese, are excluded in iRoot FS to enhance its compatibility. On the other hand, smaller particle size on the surface of iRoot FS results in higher cell attachment and subsequent proliferation (188). iRoot SP is non-cytotoxic to MG63 cells and enhances osteoblastic differentiation, which is beneficial to healing inflammatory periapical tissue (190).

The migration and fusion of osteoclast precursors is key to osteoclast formation. MTA and Bioaggregate inhibit bone resorption and osteoclast differentiation via preventing the migration and fusion of osteoclast precursors, including mouse bone marrow macrophages (191) and RAW264.7 macrophages (192-194). In addition, MTA inhibits osteoclastogenesis in a dose-dependent manner in the co-culture of mouse bone marrow cells with primary osteoblast cells. Furthermore, MTA suppresses expression of osteoprotegerin in primary osteoblast cells without affecting receptor activator of NF-κB ligand (RANKL) expression levels (195,196). Moreover, MTA solution at low concentration (20%) impairs phosphorylation of c-Src, decreases expression levels of genes encoding MMP-9 and cathepsin K and disrupt formation of actin rings. MTA solution at high concentration (50%) upregulate expression levels of Bim to increase apoptosis of osteoclasts (196). The mechanism underlying MTA-induced inhibition of osteoclastogenesis is associated with attenuation of the autophagic pathway, as demonstrated by decrease in autophagic vacuole and expression levels of autophagic genes and proteins (194). Bioaggregate inhibits osteoclastogenesis via the NF-κB/RANK signaling pathway by decreasing expression levels of Rank, TNF receptor-associated factor 6, NF-κB and nuclear factor of activated T cells 1 (192,193). Bioaggregate and MTA possess comparable ability to decrease osteoclast numbers and attenuate bone resorption (191). Bioaggregate or MTA inhibit osteoclast differentiation and bone resorption due to activation of autophagy in osteoclast differentiation; MTA inhibits osteoclast differentiation via inhibition of the autophagic pathway (194). Compared with MTA, Biodentine exhibits a lower inhibitory effect on osteoclast differentiation and activity of murine bone marrow macrophages by inhibiting ERK1/2 and NF-κB signaling pathways (197). Moreover, the inhibitory effect on osteoclast differentiation and activity of both MTA and Biodentine is similar to that of alendronate, which has been reported to prevent root resorption by inhibiting macrophages (197), which suggests the application of calcium silicate-based bioceramics as treatment to prevent root resorption in endodontics (198,199). iRoot SP shows more cytotoxicity to RAW264.7 cells than MTA but possesses a similar ability to inhibit osteoclastogenesis (200). The aforementioned studies suggested that calcium silicate-based bioceramics attenuate osteoclast differentiation and the primary mechanism is associated with their bioactive elements. Bioactive elements contained in these bioceramics, such as Ca, Mg, Si and Sr, enhance osteoblastic differentiation and suppress RANKL-induced osteoclastogenesis (201-204). A significant increase in Si and Sr ions has been observed in extracts of calcium silicate-based bioceramics (192,205). Si ions enhance the viability, adhesion, differentiation, mineralization and angiogenesis of osteoblasts via the Wnt/β-catenin and MAPK signaling pathways (206-210). Meanwhile, the effect of Si ion on surface roughness is characterized by increased adhesion and proliferation of human osteoblast cell lines (211). Furthermore, Si and Sr ions suppress RANKL-mediated osteoclastic differentiation and bone resorption by inhibiting expression levels of cathepsin K, tartrate-resistant acid phosphatase and c-Fos (205,212,213). In addition, Si and Sr ions create alkaline conditions, which neutralize lactic acid from osteoclasts and promotes accumulation of mineralized components of teeth (214,215). Therefore, the bioactive elements exhibit synergistic effects on osteogenesis, osteoclastogenesis and angiogenesis of associated cells in endodontics (216).

Compared with SuperEBA and Vitrebond, MTA exhibits decreased suppression of mitochondrial activity in the rat DPC RPC-C2A cell line (218). Furthermore, MTA significantly promotes proliferation, odontogenic differentiation and mineralization of human DPCs but inhibits secretion of lL-1β and IL-6 (219,220). In order to improve the bioactivity potential of MTA, the growth factor FGF-2 has been added to MTA to enhance its effect on proliferation and osteogenic differentiation of human DPCs (221). Set and fresh MTA display similar biocompatibility with human DPCs. In addition, MTA increases expression of the angiogenic factors VEGF and angiogenin (222). In comparison with MTA, Bioaggregate and Biodentine possess equal biocompatibility with human DPCs. Moreover, Bioaggregate, Biodentine and MTA enhance mRNA expression levels of ostogenic/odontogenic genes, such as ALP, OPN, OCN, DSPP and DMP-1, increase Alp activity and promote mineralization nodule formation due to activation of the MAPK signaling pathway induced by these calcium silicate-based bioceramics (223,224). In addition, Bioaggregate exhibits superior capacity to MTA in terms of adhesion, attachment and migration of human DPCs (225). Moreover, compared with MTA, Bioaggregate induces enhanced mineralization and odontoblastic differentiation in human DPCs (226). In terms of osteogenic differentiation of human DPCs, Bioaggregate displays stronger potential than MTA (223). Both Biodentine and MTA promote mineralization by increasing secretion of TGF-β1 from human DPCs, which mediates mineralization-associated cellular activity and subsequent dentin bridge formation (227,228). Biodentine enhances cell proliferation, viability, migration, adhesion, odontoblastic differentiation and biomineralization of the immortalized murine DPC OD-21 cell line (229). iRoot BP Plus exhibits a higher proliferation rate of human DPCs compared with MTA during the whole culture period (230). Moreover, iRoot BP Plus promotes migration and upregulates the expression of focal adhesion molecules in human DPCs via the ERK 1/2, JNK and Akt signaling pathways (231). In addition, iRoot BP Plus possesses stronger potential than MTA to enhance the mineralization and odontoblastic differentiation of human DPCs (226).

Given that Ca(OH)₂ and MTA are used in pulp capping, Kierat et al (232) compared cytotoxicity to human DPFs; MTA was less cytotoxic to human DPFs compared with Ca(OH)₂. Likewise, Zakerzadeh et al (233) compared the biocompatibility of MTA and Biodentine with human DPFs; MTA and Biodentine did not display cytotoxicity or genotoxicity at 0-1,000 µg/ml concentration. Similarly, Biodentine does not influence the morphology, proliferation or cell integrity of human DPFs, but affects cell survival in a concentration- and time-dependent manner (234). Moreover, higher concentration of Biodentine suppress differentiation of DPFs by decreasing production of COL1 and TGF-β1 (234), whereas Biodentine at lower concentrations stimulate TGF-β1 secretion (228). Both TGF-β1 and COL1 exhibit notable effects on collagen synthesis, mineralization and hard dental tissue generation (228,235-237). TGF-β1 and FGF-2 are also involved in proliferation, differentiation and migration of DPFs (228,238,239). Giraud et al investigated the effect of Biodentine on injured human DPFs; Biodentine increased cellular survival and migration, as well as the secretion of FGF-2 and TGF-β1 in lipoteichoic acid-stimulated and physically injured human DPFs. Moreover, injured DPFs exhibit decreased levels of pro-inflammatory cytokines, such as IL-6, and inflammatory cell recruitment in the presence of Biodentine compared with that in the presence of resin-containing TheraCal (240). TheraCal inhibits proliferation of human DPFs while Biodentine has no effect on proliferation of human DPFs. In addition, decreased release of pro-inflammatory cytokine IL-8 and higher mineralization have been observed in human DPFs induced by Biodentine compared with TheraCal (78). In addition, complement, particularly the C5a fragment, is involved in initiation of inflammation (241) and recruitment of DPSCs in regeneration (242). Due to the fact that C5a is produced by local DPFs when dental pulp is injured, Giraud et al compared several pulp capping materials and their effect on release of C5a in human DPFs. The results revealed that C5a secretion of injured human DPFs was not affected by Biodentine, whereas TheraCal increased C5a secretion. Furthermore, compared with TheraCal, Biodentine significantly decreased recruitment of THP-1 cells involved in inflammation by affecting C5a/C5a receptor interactions (243). These results suggested that calcium silicate-based Biodentine affects the balance between initial pulp inflammatory reaction and subsequent pulp healing by affecting complement activation following pulp injury. Calcium silicate-based bioceramics shift the balance toward pulp healing, whereas resin-containing materials shift the balance toward pulp inflammation (78,240).

Vidovic Zdrilic et al (111) found that MTA-conditioned medium promotes healing of injured periapical tissue in vivo, whereas MTA inhibits viability and biomineralization in mouse PDLCs in vitro. Eluate extract from MTA results in decreased viability and increased death of human PDLCs compared with those in a control group (medium-only) (244). By contrast, when human PDLCs are incubated with set MTA, MTA enhances mineralization and BMP-2 expression levels, which is caused by gradual production of Ca²⁺ from MTA and the interaction of BMP2/BMP-2 receptors in human PDLCs (245). These contradictory results may be due to different preparation methods of MTA. Set MTA releases calcium gradually and calcium at low concentrations stimulates cells mildly, whereas MTA-conditioned medium leads to immediate release of calcium at high concentrations; this may exert a negative effect on bioactivity of MTA to PDLCs. Compared with MTA, Bioaggregate displays better biocompatibility with human PDLCs, which may be due to the absence of aluminium in Bioaggregate. Moreover, the primary components of Bioaggregate, calcium silicate, calcium hydroxide and hydroxyapatite, stimulate proliferation of human PDLCs (246). In addition, Bioaggregate significantly increases mineralization in PDLCs in a concentration- and time-dependent manner. Furthermore, Bioaggregate enhances differentiation of PDLCs, particularly osteogenic differentiation, via microRNA-146a upregulation, as demonstrated by increased expression levels of ALP, DMP1 and BMP and decreased cementum protein 1 expression levels in PDLCs (247). In comparison with MTA, Biodentine exhibits equally good biocompatibility with human PDLCs (55,181). Moreover, Biodentine displays greater potential in terms of attachment and proliferation of human PDLCs compared with MTA, which may be due to its hydroxyl apatite-like surface characteristics and greater release of calcium and silicon ions compared with MTA (181). Luo et al compared bioactivity of Biodentine and iRoot FS to human PDLCs and showed that both Biodentine and iRoot FS increased the adhesion of human PDLCs. iRoot FS possesses superior ability to Biodentine to promote viability, proliferation and osteoblastic differentiation of human PDLCs (248). In order to clarify the bioactivity of iRoot SP to human PDLCs, Chang et al investigated the biocompatibility and bioactivity of four root canal sealers: iRoot SP, Sealapex (Kerr Corporation), ARS (Dentsply-Sankin KK) and MTA Fillapex. The results revealed that all sealers were non-toxic to human PDLCs and increased Alp activity and formation of mineralization nodules. However, compared with Sealapex, an epoxy resin-based material, the other three sealers induced lower inflammatory response and enhanced osteogenic differentiation in PDLCs via the integrin-mediated signaling pathway (249).

Human PDLFs exhibit decreased proliferation rate when cultured on the surface of MTA compared with culturing on the surface of a coverslip (250); PDLFs on the surface of MTA have a rounded morphology with blunted extensions, while PDLFs on the surface of glass coverslips show good attachment and spreading (251). Balto investigated the effect of MTA surface characteristics on attachment of human PDLFs; human PDLFs did not attach to fresh MTA and the surface of cells appeared less smooth and exhibited more vacuoles. By contrast, human PDLFs on the surface of set MTA were round and flattened with smooth surfaces and attached well to MTA (252). Similarly, Bonson et al reported that human PDLFs exposed to washed MTA possess greater proliferation capacity those exposed to fresh MTA. Moreover, compared with fresh MTA, washed MTA exhibits stronger potential to induce osteogenic differentiation of human PDLFs (253). Compared with other endodontic materials, such as Diaket (ESPE; 3M), Super-EBA (Harry J Bosworth Company) and amalgam, MTA also displays better biocompatibility with human PDLFs and does not induce apoptosis and necrosis of human PDLFs (254-257). Compared with these other materials in root perforation models in vitro, MTA also results in higher viability in human PDLFs and induces mRNA expression levels of COL1 and RUNX2 in human PDLFs, which suggests that MTA has potential to induce osteogenic differentiation of PDLFs, which is key for periodontal regeneration (105). Likewise, MTA-conditioned medium at low concentrations (0.5, 5.0 and 50.0 µg/ml) possesses superior capacity to formocresol and ferric sulphate in maintaining the viability of human PDLFs, whereas MTA at higher concentrations (5,000 µg/ml) shows slight cytotoxicity to human PDLFs (258,259). Bioaggregate displays biocompatibility with human PDLFs comparable to that of MTA, whereas viability of human PDLFs in the presence of Biodentine is slightly decreased compared with MTA (260-262). By contrast, Akbulut et al (263) reported that Biodentine possesses better biocompatibility with human PDLFs. This discrepancy may be associated with chemical composition of the material, assessment time point and surface characteristics. In terms of chemical composition, calcium chloride used in the liquid of Biodentine as an accelerator decreases the setting time (264) and results in early production of calcium hydroxide, which contributes to relatively decreased cell survival after 24 h in the presence of Biodentine (265). Moreover, zirconium oxide in Biodentine is non-toxic to murine PDLFs, but bismuth oxide in MTA has no impact on cell growth (266). MTA supports higher cell viability during the first 24 h but decreases cell viability to 80% later (251,260). Biodentine maintains lower cell viability during the first 24 h but viability increases gradually over time (261). Human PDLFs attach well to the surfaces of both MTA and Biodentine and maintain their original morphology. Nevertheless, more cell aggregates have been observed on the surface of Biodentine, whereas human PDLFs tend to show greater spread and elongation on MTA (267). Moreover, expression levels of Integrin β1 and Vinculin, which are associated with focal contacts between human PDLFs and bioceramics, are higher in human PDLFs treated with Biodentine than in those treated with MTA, which suggests that the surface characteristics of Biodentine promote the adhesion and survival of human PDLFs more strongly compared with MTA (268).