Evaluation of tissue response of experimental calcium silicate‐based endodontic sealers: An in vivo study

Abstract

The aim of this study was to evaluate whether the experimental sealers, composed of tricalcium silicate (CaSi) and CaSi supplemented with calcium hypochlorite (CaSi‐H), stimulate tissue repair. The tissue response of experimental materials was compared with commercially available materials Bio‐C sealer (BC sealer) and BioRoot (BROOT). Polyethylene tubes filled with materials or left empty (controls) were implanted in rats for 7, 15, 30 and 60 days. The number of fibroblasts, mast cells, collagen content, interleukin‐10 (IL‐10) and fibroblast growth factor‐1 (FGF‐1) immunoexpression were evaluated. Data were analysed with two‐way anova and Tukey's test, and linear regression. The fibroblast–FGF correlation was estimated using the Pearson correlation coefficient. In all groups, the number of fibroblasts and collagen content increased significantly over time. At all‐time points, CaSi‐H and control groups showed the highest values of fibroblasts, whereas the lowest values were observed in BROOT. At 60 days, no significant differences were detected among CaSi, CaSi‐H, BC sealer and controls. At all‐time points, the capsules of CaSi and CaSi‐H exhibited the highest values of FGF‐1 and IL‐10 immunoexpression. CaSi and CaSi‐H induced the formation of fibrous capsules, indicating that these sealers stimulate connective tissue repair.

INTRODUCTION

Calcium silicate‐based materials, commonly referred to as bioceramics, have been extensively developed due to their excellent biological and physicochemical properties. They also exhibit a wide diversity of chemical compositions, with varying proportions of active components and differences in their physical, chemical and biological properties, making them versatile for different clinical applications. The bioceramic materials are hydrophilic and exhibit dimensional stability and harden upon contact with moisture [1, 2].

The hydration of calcium silicate materials upon contact with tissue fluids is essential for their bioactivity and occurs in interconnected stages. Initially, water from the fluids penetrates into the material, reacting with the calcium silicates and triggering chemical reactions that result in the formation of calcium silicate gel and calcium hydroxide. The gel serves as an amorphous matrix that promotes cell adhesion and mineralisation, whereas the calcium hydroxide increases the local pH, creating an alkaline environment that inhibits microbial growth and stimulates cell proliferation, promoting the tissue remodelling [3, 4]. Simultaneously, ions released by the material, such as calcium and silicon, interact with phosphates present in the tissue fluids or extracellular matrix, promoting the precipitation of a hydroxyapatite layer [2], which is essential for tissue regeneration and the integration of the material with the surrounding tissues. Thus, the calcium silicate‐based materials have demonstrated bioactive potential, which may allow faster healing and tissue remodelling, thereby accelerating the repair process [5].

Among the commercially available bioceramic materials, Bio‐C sealer (BC sealer; Angelus) is a ready‐to‐use sealer composed of calcium silicates, tricalcium aluminate and iron oxide, which can influence the mechanical strength and final colour of the material. When compared to TotalFill BC (FKG Dentaire) and AH Plus (AHP; Dentsply DeTrey), BC sealer exhibits the shortest setting time and the highest solubility, but with low dimensional change [6]. In subcutaneous tissue of rats, BC sealer presented biocompatibility and bioactive potential [7]. In turn, BioRoot RCS (BROOT; Septodont) uses a formulation with tricalcium silicate and povidone in the powder, and a liquid solution containing calcium chloride dihydrate, which accelerates the setting reaction and may favour the formation of apatite [8, 9]. An in vitro study in human periodontal ligament cells showed that BROOT stimulated cell metabolism and demonstrated low cytotoxicity [8]. In healthy patients who had lower premolars sealed with BROOT, there were few spaces at the dentine‐sealant interface, indicating greater marginal adaptation of the material [9].

However, no endodontic sealer has all the properties necessary for an ideal material, highlighting the need for improvement in available materials. Thus, a tricalcium silicate‐based sealer (CaSi) was developed in an attempt to obtain a material that causes minimal tissue injury and, in addition, stimulates tissue cells to release mediators involved in connective tissue repair. Considering that the addition of antimicrobial agents to endodontic sealers can help control residual microorganisms after root canal preparation [10], another experimental sealer (CaSi‐H) was manufactured with the addition of calcium hypochlorite to CaSi. Although an in vivo study has demonstrated that these experimental materials (CaSi and CaSi‐H) are biocompatible and present bioactive potential [7], the study did not evaluate whether these materials stimulate the release of anti‐inflammatory mediators favouring the repair of connective tissue, as is expected from a repair material. In fact, although the sequence of inflammatory processes induced by calcium silicate‐based sealants is well‐documented [11, 12, 13], reports on the cellular and molecular mechanisms involved in the regression of the inflammatory process that culminate in tissue repair are still scarce [14, 15, 16].

Connective tissue repair requires the formation of collagen fibres by fibroblasts, whose proliferation is regulated by various factors, such as fibroblast growth factor‐1 (FGF‐1) [16]. Different cells release mediators that control connective tissue remodelling and repair. Among them, evidence suggests that mast cells release FGF‐1, which stimulates fibroblast proliferation and consequently promotes the formation of a fibrous capsule around biodentine and mineral trioxide aggregate materials implanted in the subcutaneous tissues of rats [16]. Interleukin‐10 (IL‐10), an anti‐inflammatory cytokine, plays a critical role in regulating and limiting excessive inflammation; thus, a marked increase in IL‐10 is often associated with regression of the inflammatory response and tissue repair [17].

Considering that an endodontic sealer is expected to promote the formation of soft and mineralised connective tissues, our aim was to investigate whether the experimental sealers (CaSi and CaSi‐H), compared to BC sealer and BROOT, promote the proliferation of fibroblasts and recruitment of mast cells, increase in the collagen content and stimulate the production of IL‐10, favouring the connective tissue repair. The null hypothesis was that the experimental materials would not induce connective tissue repair when compared to Bio‐C sealer and BioRoot materials.

MATERIAL AND METHODS

Experimental procedure

This study is reported in accordance with ARRIVE guidelines 2.0 (Animal Research: Reporting of In Vivo Experiments). The study received approval from the Institutional Ethical Committee for Animal Research of Dental School of Araraquara (CEUA #19/2021; São Paulo State University—UNESP), adhering to national legislation regarding the use of animals.

Thirty‐two male Holtzman rats (Rattus norvegicus albinus), weighing approximately 250–300 g, were randomly distributed into one of four experimental groups (n = 6/group per period) receiving different materials (Table 1) and a control group (empty polyethylene tubes; n = 6 per period) (Figure 1). As recommended by the ISO standards (ISO‐10993‐6:2016), four polyethylene tubes were implanted in each animal, such that each tube contained a different sealer or control, placed in a quadrant rotation pattern. Thus, eight rats were needed to obtain six implants (necessary sample size) representing each sealer or control per period (Figure S1). The sample size was determined based on previous studies [15, 16], considering a minimum difference between experimental materials and control of 20% in the number of fibroblasts in the capsules. Considering a 90% test power and an alpha error of 0.05 for recognising significant differences, at least six samples per group in each period were necessary.

TABLE 1.

Chemical composition, manufacturer and proportions of the bioceramic materials used.

Materials/Manufacturer	Proportion	Composition
Calcium silicate‐based sealer (CaSi) aMineral Research Processing bSynth, Diadema cMerck dSigma‐Aldrich	1 g: 0.3 g (powder/liquid)	Powder: tricalcium silicatea, dicalcium silicatea, monobasic calcium phosphateb, calcium hydroxidec, zirconium oxidec and calcium tungstated. Liquid: polyethylene glycol 400d
Calcium silicate and hypochlorite‐based sealer (CaSi‐H) aMineral Research Processing bSynth, Diadema cMerck dSigma‐Aldrich	1 g calcium silicate + 0.2 g calcium hypochlorite: 0.3 g (powder/liquid)	Powder: tricalcium silicatea, dicalcium silicatea, monobasic calcium phosphateb, calcium hydroxidec, zirconium oxidec, calcium tungstated and calcium hypochlorited. Liquid: polyethylene glycol 400d
BioRoot RCS (BROOT) Septodont Batch: B27522	1 portion of powder (measurement on a scale) for 5 drops (250 µL) of liquid according to the manufacturer	Powder: tricalcium silicate, zirconium oxide and povidone. Liquid: dihydrate calcium chloride, sand and purified water
Bio C‐sealer (BC sealer) Angelus Batch: 57779	Ready to use	Tricalcium silicate, dicalcium silicate, tricalcium aluminate, calcium oxide, zirconia oxide, silicon oxide, polyethylene glycol and iron oxide

FIGURE 1.

Diagram illustrating the bioceramic materials implanted into the subcutaneous tissues of rats and the parameters evaluated after obtaining sections of the paraffin‐embedded samples. FGF‐1, fibroblast growth factor 1; IL‐10, interleukin‐10.

The rats were maintained in a room under 12 h light/12 h dark cycle at controlled temperature (23 ± 2°C) and humidity (55% ± 10%), with water and food provided ad libitum. Four rats were housed in each polyethylene cage, which was filled with a layer of white pine shavings.

The materials (Table 1) were inserted into polyethylene tubes (10.0 mm in length and 1.6 mm in diameter) and, immediately, implanted into dorsal subcutaneous tissues. To implant the polyethylene tubes, the animals were anaesthetised with intraperitoneal injection of ketamine hydrochloride (80 mg/kg body weight) and xylazine hydrochloride (8 mg/kg body weight), using an insulin syringe. The dorsal skin was shaved and disinfected with a 5% iodine solution. A 2 cm incision was made along the cranio‐caudal plane, and after tissue separation, the polyethylene tubes were placed in the subcutaneous pockets. The skin was then sutured with simple stitches using 4–0 silk thread.

After 7, 15, 30 and 60 days, the animals were sacrificed (eight rats per period) with an anaesthetic overdose (solution containing 240 mg/kg body weight of ketamine hydrochloride and 24 mg/kg body weight of xylazine hydrochloride). The dorsal skin was shaved and disinfected, and the portion containing the implants was excised; the implants were identified according to the groups, and, subsequently, the specimens (implants and surrounded tissues) were immersed in a 4% formaldehyde solution (prepared from paraformaldehyde) buffered with 0.1 mol/L sodium phosphate at pH 7.2 at room temperature. After 48 h, the specimens were dehydrated with increasing concentrations of ethanol, cleared with xylene and embedded in liquid paraffin at 60°C for 6 h. Longitudinal sections (6 µm thick) were made using a rotary microtome (Leica, model RM2125 RST) and disposable stainless‐steel knives (Leica, model 818). Non‐serial sections were stained with Masson's trichrome for morphological analysis and quantification of fibroblasts. In the sections subjected to the Alcian blue (AB) histochemical method, the number of AB‐positive mast cells was quantified. The content of collagen in the capsules was measured in picrosirius red‐stained sections. Non‐serial sections, adhered to silanised slides, were subjected to immunofluorescence reactions for detection of IL‐10 and FGF‐1. These parameters were evaluated in all specimens of each group (six specimens/group) and time points (7, 15, 30 and 60 days).

Numerical density of fibroblasts

In all specimens, the number of fibroblasts was quantified in the capsules from sections stained with Masson's trichrome; three non‐serial sections were analysed for each specimen. Using a light microscope (BX51, Olympus), in each section, a standardised field (0.09 mm²), in the central portion of the capsule adjacent to the opening of the implanted tube was captured at ×695. The fibroblasts, identified considering their elongated/fusiform shape and ovoid nucleus, were computed using image analysis software (image pro‐express 6.0, Olympus). The number of fibroblasts per square millimetre of capsule was calculated by dividing the total number of fibroblasts by the total area per capsule of each specimen (0.27 mm²), as described in previous studies [15, 16]. The number of fibroblasts was obtained by a calibrated and blinded examiner; the quantitative analysis was performed twice at intervals of at least 4 weeks. This analysis was conducted in all specimens of each group (six samples/implants per group) at 7, 15, 30 and 60 days.

Numerical density of mast cells

The number of mast cells in the capsules was evaluated in all samples of each group at 7, 15, 30 and 60 days. For each specimen, two non‐serial sections were immersed in 1% solution of AB 8GX (Sigma‐Aldrich Chemie) dissolved in 3% aqueous acetic acid (pH 2.5) for 30 min. After rinsing in running water for 10 min, the sections were counterstained with Carazzi's haematoxylin.

In each section, two fields of the capsule adjacent to the opening of the implanted tube were captured at ×345 magnification (light microscope BX51, Olympus), totalling a standardised field of around 1.44 mm² per implant. Typical mast cells exhibiting AB‐positive granules (turquoise–blue colour) were counted, and the number of mast cells per mm² of capsule was calculated. The number of mast cells was counted by a calibrated and blinded examiner; the quantitative analysis was performed twice at intervals of at least 4 weeks.

Content of collagen in the capsules

The collagen content in the capsules was quantified in the sections obtained from six samples of each group per period. To estimate the content of collagen, the sections were subjected to the 0.1% picrosirius‐red solution and examined using a polarised light microscope (BX‐51, Olympus). In all specimens, the collagen content was estimated in three non‐serial sections per specimen (implant). Using a digital camera attached to the light microscope (BX51, Olympus) with polarised filters, a field of the capsule from each section was captured at ×695. To assess birefringent collagen content, images were captured under rigorously standardised parameters, including light intensity, diaphragm aperture, condenser position and exposure time. Image analysis software (imagej; National Institutes of Health) was employed to evaluate birefringence, defined by hue parameters: red/orange (2–38 and 230–256), yellow (39–51) and green [18, 19]. Subsequently, collagen quantity was expressed as the percentage of birefringent area relative to the total area. A calibrated and blinded examiner estimated the birefringent collagen content; this analysis was performed twice at intervals of at least 4 weeks.

Immunofluorescence for detection of FGF‐1 and IL‐10

The detection of FGF‐1 and IL‐10 was performed in sections of all samples of each group (six samples/implant per group in each period) at 7, 15, 30 and 60 days. After dewaxing and hydration, the sections were immersed in Tris‐EDTA buffer (pH 9.0) and heated for 30 min in a microwave oven at 80°C for antigen retrieval. After cooling, the sections were incubated for 20 min in non‐immune serum, and the sections were incubated overnight at 4°C in a humid chamber with the mouse monoclonal FGF‐1 antibody (Santa Cruz Biotechnology; code: sc55520; diluted 1:75) and mouse monoclonal IL‐10 antibody (Santa Cruz Biotechnology, code sc‐365858; diluted 1:100). After washing with 0.1 M Tris–HCl buffer, the sections were incubated with fluorochrome‐conjugated anti‐mouse secondary antibodies (Alexa Fluor 488; Life Technologies) for 60 min at room temperature. The nuclear staining was performed with DAPI, and the sections were mounted with Fluoromount G (EMS). Quantitative analysis of the immunofluorescence reactions was performed using a fluorescence microscope (Leica DM4000 B LED) and image analysis software las4 (Leica). In each specimen, two non‐serial sections were captured at ×40 magnification. In a standardised field (0.146 mm²), the number of FGF‐1 and IL‐10 immunofluorescent cells was counted, and the number of these cells per mm² was calculated. These analyses were conducted by a calibrated and blinded examiner; quantitative analyses were performed twice at intervals of at least 4 weeks.

Statistical analysis

The outcome variables (number of fibroblasts, number of mast cells, number of FGF‐1‐immunolabelled cells, number of IL‐10‐immunolabelled cells and amount of birefringent collagen content) were analysed using the graphpad prism 9.02 software (graphpad). The Shapiro–Wilk test was used to assess the normality of the data, and, subsequently, differences of each parameter among the groups in each period and the differences of each group among the periods were evaluated by two‐way anova analysis followed by the Tukey post‐test. The significance level was set at p < 0.05. All outcome variables were summarised using the mean value and the associated confidence interval.

A linear regression analysis was also performed considering the main effects of sealer and time. The control group at 7 days was adopted as the reference category for estimating the intercept and the β coefficients. The means of the raw data for each group‐time combination were used to calculate the relative effects. All analyses were conducted in r software (version 4. 1.0; r core team), and the lm() function from the stats package was used.

Pearson's correlation coefficient and its 95% confidence interval were used to estimate the linear correlation between the number of fibroblasts and FGF‐1‐immunopositive cells.

RESULTS

Morphological description and number of fibroblasts in the capsules

The analysis of Masson's trichrome‐stained sections revealed that the thick capsules exhibited dense cell population after 7 days of implantation (Figure 2A,E,I,M,Q). However, a marked reduction in the thickness of the capsules was observed in all implants over time (Figure 2A–T). Moreover, the analysis at higher magnification view of the capsules revealed significant alterations in the cellular population and extracellular matrix components of capsules around the materials over time (Figure 3A–T). At 7 days post‐implantation, the capsules predominantly consisted of inflammatory cells, with sparse fibroblasts exhibiting spindle/elongated shapes among several inflammatory cells and few blue‐stained fibrous components (Figure 3A,E,I,M,Q). At 15 days, an evident reduction in the cellular density accompanied by marked presence of fine collagen fibres (stained in blue) was seen in all groups (Figure 3B,F,J,N,R). At 30 (Figure 3C,G,K,O,S) and 60 (Figure 3D,H,L,P,T) days, the capsules exhibited a pronounced presence of fibroblasts (β = +13.4) and bundles of collagen fibres (β = +14.3) compared to Day 7 (Table 3), strongly stained in blue.

FIGURE 2.

Light micrographs of sections showing a general view of capsules (labelled C) around the CaSi (A–D), CaSi‐H (E–H), BROOT (I–L), BC sealer (M–P) and the control (CG) (Q–T) at 7, 15, 30 and 60 days. T, space of implanted polyethylene tube; M, muscle tissue. Masson's trichrome. Bars: 180 µm.

FIGURE 3.

Light micrographs of sections showing portions of capsules of CaSi (A–D), CaSi‐H (E–H), BROOT (I–L), BC sealer (M–P) and control (CG) (Q–T) at 7, 15, 30 and 60 days. Fb, fibroblasts; CF, collagen fibres; BV, blood vessel; inflammatory cells (arrows). Masson's trichrome. Bars: 18 µm.

TABLE 3.

Estimated β coefficients and their 95% CI's obtained from linear regression analysis of the outcomes number of fibroblasts, mast cells, fibroblast growth factor (FGF‐1)‐ and interleukin‐10 (IL‐10)‐immunolabelled cells per mm² and the percentage of birefringent collagen content in the capsules as a function of the sealer used (control, CaSi, CaSi‐H, BROOT and BC sealer) and the time elapsed (7, 15, 30 and 60 days).

Independent variable		Fibroblasts β (95% CI)	Mast cells β (95% CI)	FGF‐1 β (95% CI)	IL‐10 β (95% CI)	Collagen content β (95% CI)
	Constant (control, Day 7)	10.5 (157,178)	127.0 (123,129)	504.0 (437,450)	95.7 (86,119)	16.0 (14,18)
Sealer	Control (ref.)	–	–	–	–	–
	CaSi	1.7 (−1, 4)	48.2 (30, 66)	424.4 (380, 468)	47.3 (19, 75)	−1.0 (−2, 0)
	CaSi‐H	1.7 (−1, 4)	52.7 (33, 72)	464.6 (416, 513)	56.0 (29, 83)	0.4 (−1, 2)
	BROOT	−10.5 (−14, −6)	−62.7 (−81, −44)	73.9 (24, 123)	−34.0 (−48, −20)	−1.1 (‐3, 1)
	BC sealer	1.7 (−1, 4)	−2.0 (−15, 11)	375.0 (320, 430)	2.0 (−15, 19)	7.0 (5, 9)
Time	7 days (ref).	–	–	–	–	–
	15 days	64.7 (40, 89)	225.3 (180, 270)	−297.5 (−340, −255)	234.1 (200, 268)	12.9 (10, 15)
	30 days	7.5 (−3, 18)	358.8 (310, 407)	−383.5 (−428, −339)	398.6 (360, 437)	10.8 (7, 14)
	60 days	14.3 (3, 25)	468.5 (420, 517)	−412.0 (−450, 374)	487.0 (450, 524)	13.4 (10, 16)

From 7 to 60 days, there was a significant increase in the number of fibroblasts in the capsules of all groups (Table 2). No statistically significant difference was detected at any time point between the CaSi‐H and control samples, which showed the highest values of fibroblasts. The capsules around the CaSi specimens showed a statistically significant lower number of fibroblasts than seen for controls at 7 and 15 days, although no statistically significant differences were found between these groups at 30 and 60 days. At all periods, the CaSi‐H specimens exhibited a higher number of fibroblasts than did the BROOT and BC sealers, except at 60 days. At 60 days, the difference between CaSi‐H and BC sealer in the number of fibroblasts was not statistically significant. At 60 days, no statistically significant differences in the number of fibroblasts were observed between CaSi and BC sealer and CaSi‐H and BC sealer specimens, whereas BROOT specimens exhibited the lowest values (β = −10.5) (Tables 2 and 3).

TABLE 2.

Mean number (95% CI for mean) of fibroblasts, mast cells, fibroblast growth factor‐1 (FGF‐1)‐ and interleukin‐10 (IL‐10)‐immunolabelled cells per mm² and mean percentage (95% CI for mean) of birefringent collagen (CF) content in the capsules around the CaSi, CaSi‐H, BROOT, BC sealer and control groups after 7, 15, 30 and 60 days (n = 6/group and time).

Parameter	Period (days)	CaSi	CaSi‐H	BROOT	BC sealer	Control
Fibroblasts	7	132 (115,148) b;1	161 (154,167) a;1	129 (112,145) b;1	136 (114,158) b;1	168 (157,178) a;1
	15	114 (101,126) c;2	232 (210,254) a;2	141 (134,147) b;1	158 (141,174) b;1	236 (214,258) a;2
	30	298 (288,307) a;3	309 (299,318) a;3	196 (185,206) c;2	221 (211,230) b;2	313 (298,327) a;3
	60	322 (297,346) a;4	331 (320,341) a;4	268 (251,284) b;3	329 (300,357) a;3	358 (341,374) a;3
Mast cells	7	12 (9,14) a,1	13 (9,16) a,1	2 (‐14,18) b,1	13 (6,19) a,1	11 (5,16) a,1
	15	16 (6,25) a,2	18 (12,23) a,2	5 (3,6) c,2	18 (9,26) a,2	13 (4,22) b,1
	30	27 (13,40) b,3	37 (26,47) a,3	9 (7,11) d,3	23 (16,29) b,3	17 (10,23) c,2
	60	20 (15,24) a,4	19 (12,25) a,4	4 (0.9,7) b,4	20 (14,25) a,4	24 (16,31) a,3
FGF‐1	7	983 (951,1015) a,1	955 (913,997) a,1	590 (548,632) c,1	911 (870,952) b,1	444 (437,450) d,1
	15	733 (503,585) a,2	544 (503,585) b,2	494 (491,496) b,2	529 (483,574) b,2	238 (207,269) c,2
	30	427 (383,471) a,3	433 (381,485) a,3	316 (288,344) b,3	366 (314,418) b,3	112 (97,126) c,3
	60	155 (117,193) a,4	127 (106,147) a‐b,4	111 (88,133) b,4	194 (162,226) a,4	94 (87,100) c,3
IL‐10	7	133 (121,144) a,1	133 (123,138) a,1	66 (49,82) c,1	105 (93,116) b,1	103 (86,119) b,1
	15	444 (428,459) a,2	433 (421,444) a,2	205 (186,223) c,2	388 (373,402) b,2	391 (381,400) b,2
	30	577 (562,591) a,3	588 (572,603) a,3	344 (327,360) c,3	405 (386,423) b,3	480 (469,490) b,3
	60	672 (653,690) a,4	605 (582,627) b,3	466 (441,490) d,4	505 (484,526) c,4	587 (579,594) c,4
CF %	7	14.7 (13,15) b;1	15 (11,18) b;1	15 (13,16) b;1	21.8 (19,24) a;1	16.4 (14,18) b;1
	15	20.9 (20,21) a;2	20.6 (18,23) a;2	18.5 (16,20) c;2	22.8 (21,24) a;1	28.9 (26,31) b;2
	30	24.3 (23,25) b;3	25.8 (25,26) b;3	20.9 (18,23) c;3	28.2 (27,29) a;2	26.8 (25,28) a;2
	60	29.9 (28,31) a;3	26.8 (26,27) b;2	23.2 (22,24) c;3	27.8 (27,28) b;2	29.4 (28,30) a;2

Note: Two‐way anova followed by Tukey's test. Significance level was set at p < 0.05. Comparison among groups in the same period is indicated by superscript letters in the lines; same letters = no significant difference. a > b > c > d. The superscripted numbers represent the analysis of each group over time; same numbers = no statistically significant difference.

Numerical density of mast cells

The histochemical method revealed AB‐positive mast cells (turquoise–blue colour) in the capsules of all specimens. These cells, exhibiting strong staining in their cytoplasm, were frequently observed throughout the capsules, surrounding the implanted materials as well as in the control specimens (Figure 4A–J). The morphological and quantitative analyses revealed a differential pattern in the distribution of AB‐positive mast cells, according to the implanted materials and periods (Table 2).

FIGURE 4.

Light micrographs of sections showing portions of capsules of CaSi (A and B), CaSi‐H (C and D), BROOT (E and F), BC sealer (G and H) and control (CG) (I and J). The sections were submitted to the Alcian blue histochemical method and counterstained by haematoxylin. Round/ovoid mast cells stained by AB (turquoise colour) are present throughout the capsules. The inset of the outlined area in the general views shows mast cells (arrows) with cytoplasm strongly positive for AB. P, material particles; T, space of implanted polyethylene tube. Bars: 73 and 18 µm (insets).

According to Table 2, statistically significant differences in the number of AB‐positive mast cells were not seen in the capsules of CaSi, CaSi‐H and BC sealer specimens at 7, 15 and 60 days. In contrast, the number of mast cells was significantly greater in the capsules around the CaSi and CaSi‐H than in BROOT specimens at all times. At 60 days, no statistically significant difference was seen in the distribution and presence of mast cells in the capsules around the CaSi, CaSi‐H, BC and control specimens. From 7 to 30 days, the capsules around the implanted materials exhibited a gradual increase in the number of mast cells, which reduced significantly from 30 to 60 days. Otherwise, in the control specimens, a significant increase in the number of mast cells was seen over time.

The regression analysis showed that CaSi and CaSi‐H induced a moderately higher number of mast cells (CaSi: β = 48.2 and CaSi‐H: β = 52.7) than seen in controls, whereas BROOT promoted a markedly lower number (β = −62.7), and BC Sealer exhibited an almost neutral effect (Table 3).

Immunofluorescence for detection of FGF‐1 and IL‐10

FGF‐1 immunofluorescence (green colour) was observed in the cytoplasm and/or nuclei of several cells in the capsules of all implants. Although FGF‐1‐immunostained cells were observed at all‐time points, a gradually lower number of FGF‐1‐immunostained cells in the capsules was observed over time (ranging from β = −297.5 to β = −412.0) (Figure 5A–T; Tables 2 and 3). No immunostained cells were observed in the sections used as negative controls (data not shown). The quantitative analysis showed that the capsules around the implanted materials presented an accentuated FGF‐1 immunoexpression in comparison with control specimens at all‐time points (Table 2). Moreover, a differential pattern in the immunolabelling for FGF‐1 was also seen among the different materials. At 7 days, the number of FGF‐1 immunolabelled cells was statistically significantly greater in the CaSi (mean: 983 cells/mm²) and CaSi‐H (mean: 955 cells/mm²) than in BROOT (mean: 590 cells/mm²) and BC sealer (mean: 911 cells/mm²). At 15 days, no statistically significant differences were seen among CaSi‐H (mean: 544 cells/mm²), BROOT (mean: 494 cells/mm²) and BC sealer (mean: 529 cells/mm²), which presented lower immunoexpression than in CaSi specimens (mean: 733 cells/mm²). The mean values were significantly greater in the CaSi (mean: 427 cells/mm²) and CaSi‐H (mean: 433 cells/mm²) than in BROOT (mean: 316 cells/mm²) and BC sealer (mean: 366 cells/mm²) at 30 days. At 60 days, the greatest mean values of FGF‐1 immunolabelled cells were seen in CaSi (mean: 155 cells/mm²) and BC sealer (mean: 194 cells/mm²). Furthermore, no statistically significant difference was found between CaSi‐H (mean: 127 cells/mm²) and BROOT (111 cells/mm²) at 60 days. In the control specimens, the immunoexpression for FGF‐1 varied from 444 cells/mm² (at 7 days) to 94 cells/mm² at 60 days. However, in the regression analysis using the 7‐day control group as the reference, CaSi (β = 424.4) and CaSi‐H (β = 464.6) exhibited the highest values, whereas BROOT showed the lowest response (β = 73.9) (Table 3).

FIGURE 5.

Light micrographs of sections showing portions of capsules from the CaSi (A–D), CaSi‐H (E–H), BROOT (I–L), BC sealer (M–P) and control (CG) (Q–T) at 7, 15, 30 and 60 days. Sections were submitted to immunofluorescence for detection of FGF‐1 (green colour) and nuclear staining with DAPI (blue colour). Immunolabelled cells (arrows); BV, blood vessel; asterisks, material particles; T, space of implanted polyethylene tube. Bars: 18 µm.

Regarding IL‐10, the immunolabelling was observed in several cell types, including fibroblasts, mast cells and endothelial cells in the capsules around the specimens at all periods (Figure 6A–T). In the sections used as negative controls, immunolabelled cells were not observed (data not shown). As shown in Table 2, from 7 to 60 days, the number of IL‐10‐immunolabelled cells increased significantly with increasing time in the capsules of all implants, showing marked increase at 15 (β = 234.1), 30 (β = 398.6) and 60 days (β = 487.0) (Table 3). However, the IL‐10 immunoexpression was statistically significantly greater in the CaSi and CaSi‐H specimens than in BROOT, BC sealer and control specimens at all‐time points. At 60 days, the immunoexpression was greater in CaSi (mean: 672 cells/mm²) than in CaSi‐H (mean: 605 cells/mm²). At all‐time points, the lowest values were observed in the BROOT specimens, whereas no statistically significant differences between BC sealer and control specimens were detected (Table 2).

FIGURE 6.

Light micrographs of sections showing portions of capsules from the CaSi (A–D), CaSi‐H (E–H), BROOT (I–L), BC sealer (M–P) and control (CG) (Q–T) at 7, 15, 30 and 60 days. Sections were submitted to immunofluorescence for detection of IL‐10 (green colour) and nuclear staining with DAPI (blue colour). Immunolabelled cells (arrows); BV, blood vessel; asterisks, material particles; T, space of implanted polyethylene tube. Bars: 18 µm.

Collagen content in the capsules

The analysis of picrosirius‐red stained sections under polarised light revealed thin birefringent collagen fibres (in red colour) in the capsules around all implants at 7 days (Figure 7A,E,I,M,R). An evident increase in the birefringent collagen was seen in the capsules around all the implants over time (Figure 7A–U), as shown by regression linear analysis (Table 3). According to the quantitative analysis (Table 2), the amount of collagen was statistically significantly greater in the capsules containing CaSi and CaSi‐H than in BROOT specimens at 15, 30 and 60 days, although no statistically significant differences in the content of birefringent collagen were seen among CaSi, CaSi‐H and BROOT specimens at 7 days. Moreover, there was no statistically significant difference in the birefringent content between the CaSi and CaSi‐H, on the one hand, and control specimens at 7 days. At 60 days, the highest values were observed in the CaSi (mean: 29.9%) and CG (mean: 29.4%), followed by CaSi‐H (mean: 26.8%) and BC sealer (mean: 27.8%). The capsules around all implants had a significant increase in the content of birefringent collagen over time.

FIGURE 7.

Light micrographs showing portions of sections of capsules adjacent to the implanted tubes after 7 (A, E, I, M, R), 15 (B, F, J, N, S), 30 (C, G, K, O, T) and 60 (D, H, L, P, U) days of implantation. Sections were stained with picrosirius‐red and photographed under polarised light. Birefringent collagen fibres are mainly seen in red colour. T, space of implanted polyethylene tube. Bars: 20 µm.

Correlation between fibroblasts and FGF

The Pearson correlation analysis revealed significantly negative correlations between the number of fibroblasts and the immunoexpression of FGF‐1, as indicated by the Pearson correlation coefficients: CaSi (r = −0.84), BC sealer (r = −0.86), CaSi‐H (r = −0.94), BROOT (r = −0.98) and control (r = −0.96).

DISCUSSION

This study evaluated the tissue response induced by experimental calcium silicate‐based sealers in comparison with two commercial sealers, BioRoot (Septodont) and Bio‐C sealer (Angelus), with a focus on tissue repair. The experimental cements consisted of tricalcium silicate, dicalcium silicate, monobasic calcium phosphate and calcium tungstate, featuring a composition more focused on calcium silicate, with the addition of an antimicrobial agent (CaSi‐H), and without the stabilisers and additives often present in the composition of commercial calcium silicate sealers. The findings of the present study show that the experimental sealers stimulate the immunoexpression of IL‐10 and promote an increase in the number of fibroblasts and collagen content, indicating that experimental sealers induce the connective tissue repair.

During setting reaction of calcium silicate‐based endodontic sealers, calcium and hydroxyl ions (OH−) are released, providing an alkaline pH to the microenvironment. In fact, it has been demonstrated that after 3 h of immersion in distilled water, the medium containing CaSi, the same experimental material evaluated in the present study, exhibited a pH of 10.1 [20], whereas the medium containing BROOT exhibited pH 11.7 [21]. After 7 days of immersion in distilled water, Bio‐C sealer maintained an alkaline pH of 9.7 [6]. The alkalinisation of the microenvironment promotes the recruitment of inflammatory cells and induces the release of several factors and cytokines [7, 17, 19, 22]. In fact, BROOT capsules with a marked presence of inflammatory cells exhibited the lowest immunoexpression of FGF‐1 and IL‐10, markers associated with regression of the inflammatory reaction and rearrangement of connective tissue. The experimental sealers (CaSi and CaSi‐H) evaluated in the present study had polyethylene glycol as vehicle, as this polymer allows diffusion of Ca²⁺ and OH⁻ [23], in addition to improving the physicochemical properties of silicate materials [24]. A previous study demonstrated that the setting time of the experimental material (CaSi) using polyethylene glycol as a vehicle was approximately 785 min [2]. Despite the long setting time, the experimental materials (CaSi and CaSi‐H) induced the recruitment of inflammatory cells after 7 days of implantation in the subcutaneous tissues, whereas the inflammatory reaction reduced significantly over time, as previously demonstrated [7].

In the present study, the experimental materials stimulated the FGF‐1 immunoexpression as the capsules adjacent to the CaSi and CaSi‐H showed higher values of FGF‐1 than in control specimens. FGF‐1 is a potent mitogen factor that exerts strong effects on various cell types, promoting the angiogenesis and wound healing [21, 25, 26]. Our findings showed a differential pattern in the immunoexpression of FGF‐1 in the capsules, supporting the concept that this factor plays a role in the control of the connective response to calcium silicate‐based sealers. This idea is reinforced by the enhanced FGF‐1 immunoexpression detected in the capsules around sealers compared to control specimens at all‐time points. Significant reduction in the FGF‐1 immunoexpression over time suggests that this mitogenic factor may be associated with the onset of the inflammatory process. In the initial phase of inflammatory reaction, FGF‐1 stimulates the proliferation of vascular cells, leading to an increase in the profiles of blood vessels, as observed in the granulation tissue [16]. The potential of bioceramic materials to stimulate cells to release several factors and cytokines has been widely demonstrated. An in vitro study demonstrated that BioRoot RCS, in direct contact with human periodontal ligament cells, induced the release of angiogenic and osteogenic growth factors, such as vascular endothelial growth factor, fibroblast growth factor‐2 and bone morphogenetic protein‐2, suggesting that this bioceramic material may stimulate angiogenesis and bone formation [8]. Here, the enhanced FGF‐1 immunoexpression in the capsules formed in response to bioceramic materials, including the CaSi and CaSi‐H materials, indicates that these materials stimulate the proliferative activity of fibroblasts, mediating the regression of the inflammatory reaction and the rearrangement of connective tissues. In fact, it has been demonstrated that FGF‐1 stimulates fibroblast proliferation and migration [16], increasing the number of fibroblasts and subsequent collagen fibres in the capsules [15, 16]. Furthermore, a direct relation between FGF‐1 and Ki‐67, a marker of cell proliferation, was also demonstrated in the capsules around bioceramic materials implanted in the subcutaneous tissues [15, 16].

Our results revealed a negative correlation between immunoexpression of FGF‐1 and the number of fibroblasts in all groups, indicating, therefore, that accentuation of FGF‐1 may stimulate the proliferation of fibroblasts, leading to a gradual increase in the number of these cells in the capsules, as demonstrated in the present study. Thus, as the fibroblast population increases in the capsules, FGF‐1 immunoexpression decreases, thereby controlling the population of these cells. As immunolabelling for FGF‐1 was mainly observed in fibroblasts, it is likely that this factor may act through an autocrine pathway. Thus, it is reasonable to suggest that the accentuated immunoexpression of FGF‐1 at 7 and 15 days may stimulate the fibroblast proliferation, leading to the increase in the number of these cells and contributing for rearrangement of connective tissue, as demonstrated in the present study.

At all periods, the immunoexpression of IL‐10, an anti‐inflammatory cytokine, was significantly greater in the capsules surrounding the experimental materials (CaSi and CaSi‐H) than in other specimens, including the control specimens, indicating that these materials may stimulate IL‐10 production by connective tissue cells, favouring the connective tissue repair. In contrast, our findings suggest that BROOT inhibits the IL‐10 production, as the capsules surrounding these specimens exhibited lower numbers of IL‐10‐immunostained cells compared to that observed in control specimens. However, the BioRoot Flow, which constitutes a modified formulation of BioRoot RCS sealer containing propylene glycol, calcium carbonate and aerosol, induced cell differentiation and promoted the expression of key genes related to bone formation when added to the culture medium with human periodontal ligament stem cells [27]. In addition, human periodontal ligament stem cells co‐cultured with BioRoot Flow exhibited overexpression for bone sialoprotein and cementum protein 1, suggesting that this bioceramic material can induce periodontal repair and regeneration [28]. Here, the increased immunoexpression of IL‐10 in the capsules around the CaSi and CaSi‐H specimens was accompanied by a marked increase in the number of fibroblasts and collagen content in these capsules, corroborating the hypothesis that the experimental materials stimulated IL‐10‐mediated connective tissue remodelling.

The increase in IL‐10 can also be explained by the production of this cytokine by mast cells, as these cells exhibiting cytoplasm strongly immunolabelled for IL‐10 were observed in the capsules of all implants in the different time points. IL‐10 exerts beneficial effects in maintaining tissue homeostasis [17], and mast cells play a crucial role in the reorganisation and repair of connective tissue in the healing process. When activated in response to injury or inflammation, these cells release mediators that participate in controlling inflammation and mediating tissue repair. Mast cells secrete growth factors, such as platelet‐derived growth factor and transforming growth factor beta, which stimulate the migration and proliferation of fibroblasts [29, 30]. During the remodelling phase, mast cells continue to influence the reorganisation of the extracellular matrix, ensuring that healing occurs in a balanced and functional manner [15]. Mast cell knockout mice exhibited an absence of fibrous capsule around subcutaneously implanted biomaterials, demonstrating that these cells play a key role in collagen formation [31], corroborating our results. A gradual increase in the number of mast cells was observed in parallel to the increase in the content of bundles of birefringent collagen in the capsules around calcium silicate‐based materials, indicating that these cells may stimulate the proliferation of fibroblasts and the collagen formation [16]. In the present study, the highest number of mast cells observed in the capsules of experimental materials (CaSi and CaSi‐H) and BC sealer was accompanied by the highest number of fibroblasts and intense immunoexpression of FGF‐1 and IL‐10, reinforcing the idea that mast cells play a pivotal role in the remodelling of the connective tissue. This idea is supported as the lower number of mast cells in the capsules observed after 60 days of implantation of BROOT was accompanied with the lowest number of fibroblasts and FGF‐1‐ and IL‐10‐immunolabelled cells.

Thus, the rearrangement of the main structural components of connective tissue, evidenced by the increase in the number of fibroblasts and content of collagen fibres, indicates that experimental materials implanted subcutaneously in rats allowed the connective tissue repair. Furthermore, our findings reinforce the concept that mast cells may have an important role in the complex cascade involved in the regression of the inflammatory reaction and re‐establishment of the structural organisation of healthy connective tissue. These data highlight new endodontic materials, able to improve tissue repair. The efficacy of bioceramic materials in promoting tissue repair may lead to better clinical outcomes, reducing complications and improving the long‐term health to the treated teeth. In the present study, the experimental sealers (CaSi and CaSi‐H) demonstrated superior repair response compared to BioRoot RCS in promoting fibroblast proliferation, mast cell recruitment, collagen deposition and production of FGF‐1 and IL‐10. The addition of calcium hypochlorite to the experimental material did not impair its tested biological properties.

However, it is important to emphasise that this study has limitations. Although the subcutaneous implant model is useful for investigating the complex cascade of cellular and molecular events involved in the tissue response induced by bioceramic materials, the clinical use of these materials requires further studies to confirm whether contact of these materials with dental pulp and periodontal ligament promotes tissue repair. Furthermore, our results were obtained in a controlled laboratory animal model and in healthy connective tissue without inflammation.

AUTHOR CONTRIBUTIONS

Conceptualisation: Paulo Sérgio Cerri, Mário Tanomaru‐Filho, Juliane Maria Guerreiro‐Tanomaru. Experiment: Evelin Carine Alves Silva, Lucas de Andrade Rodrigues, Juliane Maria Guerreiro‐Tanomaru. Methodology: Evelin Carine Alves Silva, Lucas de Andrade Rodrigues, Estela Sasso‐Cerri, Paulo Sérgio Cerri. Formal analysis: Paulo Sérgio Cerri, Evelin Carine Alves Silva, Lucas de Andrade Rodrigues. Interpretation of results: Evelin Carine Alves Silva, Paulo Sérgio Cerri. Funding acquisition: Paulo Sérgio Cerri. Project administration: Paulo Sérgio Cerri, Mário Tanomaru‐Filho, Juliane Maria Guerreiro‐Tanomaru. Writing—original draft: Evelin Carine Alves Silva, Lucas de Andrade Rodrigues. Writing—review and editing: Paulo Sérgio Cerri, Evelin Carine Alves Silva, Mário Tanomaru‐Filho, Juliane Maria Guerreiro‐Tanomaru, Estela Sasso‐Cerri.

CONFLICT OF INTEREST STATEMENT

The authors declare no conflicts of interest.

Supporting information

Supporting information

DATA AVAILABILITY STATEMENT

All data are available on request from the authors.