Potential of Acacia Extract as a Collagen Crosslinker and Stabilizer of the Resin-dentin Interface Open Access

Dental adhesive systems have developed rapidly over the last few decades.¹  However, the durability of the resin-dentin bond interface has always been an important clinical issue since degradation occurs over time.^2,3 

An essential factor related to bonding is the hybrid layer, which corresponds to the region where the resin monomers infiltrate the demineralized dentin zone and polymerize in situ.⁴  Collagen fibrils can remain unprotected by resin monomers at the base of the hybrid layer; consequently, these fibrils undergo hydrolysis and enzymatic degradation.⁵  In addition, acid etching of dentin activates the collagenolytic activities of host-derived matrix metalloproteinases (MMPs) that degrade type I collagen fibrils at the base of the hybrid layer, affecting the longevity of the resin– dentin interface.^6,7 

Collagen fibrils are formed by bundles of crosslinked microfibrils organized by the staggering of collagen molecules. The intramolecular crosslinks (endogenous crosslinkers) are responsible for the biostability of the collagen molecule, while intermolecular and interfibrillar crosslinks enhance the tensile strength, viscoelasticity, and biostability characteristics of these fibrils.⁸  Some synthetic and natural compounds can undergo collagen crosslinking^9–11  and/or inhibit MMPs.¹² 

Chlorhexidine gluconate is a synthetic compound capable of inhibiting MMPs. In vitro studies have shown that 2% chlorhexidine gluconate can effectively maintain the bond strength at the resin-dentin interface.^6,13,14  Furthermore, the addition of chlorhexidine into adhesive systems may reduce resin-dentin interface degradation.¹⁵  However, another analysis showed that this trend did not remain for longer aging periods (>12 months).¹⁶  In addition, a 36-month follow-up randomized clinical trial concluded that pretreatment with 2% chlorhexidine digluconate did not promote further restoration retention in noncarious cervical lesions.¹⁷  Other chemical agents, such as carbodiimide¹⁸  and glutaraldehyde,¹⁹  are collagen cross-linkers; however, a concern related to the tissue cytotoxicity of glutaraldehyde arises.¹⁹ 

Most natural collagen crosslinkers are polyphenols that are widely found in fruits, vegetables, cereals, and beverages derived from them, such as wine or grape juice. These polyphenols have common chemical structures and can be classified into flavonoids, phenolic acids, lignins, and stilbenes.²⁰ 

Myricetin is a flavonoid with a low molecular weight; it diffuses deeply through dentin, acts as an MMP inhibitor and crosslinking agent, and preserves the resin-dentin bond interface.²¹  Lignin eucalyptus, a polyphenol-rich natural polymer, can perform collagen crosslinking and prevent the degradation of unprotected dentin collagen.²² 

Grape seed extract is a flavonoid that contains proanthocyanidins (PAs).²³  PAs are structured by monomeric building blocks of catechins, which are flavan-3-ol units.⁸  Several studies have shown that PA from grape seed extract increases collagen resistance to biodegradation,^24–26  inhibits the synthesis of several MMPs,^26,27  increases the mechanical properties of the demineralized dentin matrix,^28–30  and positively influences the durability of the bond strength at the resin-dentin interface.^26,31  Despite good results in laboratory studies, clinical studies have shown that the application of PA as a primer³²  or its addition to the adhesive solution³³  did not result in clinical advantages in the restorations of noncarious cervical lesions after 24 months of clinical service.

Since the degradation of bonding interfaces over time remains a challenge for restorative dentistry even using crosslinker agents, other polyphenols should be tested to analyze their ability to increase collagen resistance to collagenolytic degradation through the crosslinking effect.

A polyphenol found in abundance in nature is acacia extract. Acacia extract is obtained from the bark of Acacia mearnsii, which is a medium-sized tree from the Fabaceae (Leguminosae) family native to Australia. The quick adaptation of this tree to different environmental and climatic conditions has facilitated its introduction in many countries.³⁴  The industrial production of Acacia mearnsii is mainly used in leather tanning, in the production of adhesives for wood, and as a natural coagulation agent.^34,35  The bark of this tree is known for its high concentration of PAs, which has aroused great economic interest.³⁶  However, since the differences in the chemical structures of flavan-3-ol units (PA) derived from different sources dictate their reactivity, physical properties, and interactivity with collagen,²⁵  it is relevant to study the effect of acacia extract on the resin-dentin interface. A review of the literature reveals no previous studies that have evaluated the use of acacia extract in an adhesive procedure.

Therefore, the aim of the present study was to evaluate different concentrations of acacia extract (1%, 2%, and 3%) on the following outcomes in vitro: a) interaction with collagen; b) resistance to enzymatic biodegradation; and c) bond strength to dentin after 24 hours and six months of storage. The study was conducted with the null hypotheses that the different concentrations of acacia extract on human dentin do not i) interact with collagen, ii) reduce collagen degradation, iii) influence the bond strength of the adhesive to dentin after 24 hours, and iv) influence the bond strength after six months of storage.

The study was approved by a local ethical committee (01867018.2.0000.5336). Forty-eight sound human third molars, extracted for therapeutic reasons, were obtained from patients aged between 17 and 25 years by signing a tooth donation consent form. The teeth were cleaned with a periodontal curette and disinfected in a 0.5% chloramine T solution for 24 hours. The teeth were then stored in distilled water at 4°C for up to two weeks. The roots of the teeth were embedded in self-cured acrylic resin (Jet Clássico, São Paulo, SP, Brazil) with the aid of a cylindrical Teflon matrix.

In 20 teeth, the occlusal, buccal, lingual, mesial, and distal enamel surfaces were removed with a double-sided diamond disc (Extec Corp., London, UK) under water irrigation using a cutting machine (Labcut 1010, Extec Corp.). After enamel removal, the dentin of 12 teeth was cut with a double-sided diamond disc under water irrigation to obtain a slice from the middle dentin. The remaining enamel at the edges was removed with a diamond tip under water cooling, resulting in 12 dentin discs approximately 1.0 mm thick and 6.0 mm in diameter. The other eight teeth were cut to obtain 40 beams measuring 1.5 mm in height x 1.5 mm in width x 3.0 mm in length. Five beams were obtained from each tooth. The dentin discs were used to evaluate the interaction between the acacia extract and the organic matrix (collagen), and the beams were used to evaluate the resistance to enzymatic biodegradation. The dentin discs and the beams were randomly distributed among the groups using the Randomizer Android app.

The acacia extract (Tanac, Montenegro, RS, Brazil) was diluted in distilled water (solvent) at three different concentrations: 1%, 2%, and 3%. Prior to each experiment (interaction methodology, resistance to enzymatic biodegradation, and microtensile bond strength), new and fresh solutions were formulated.

The dentin slices were demineralized in 10% phosphoric acid for 24 hours at room temperature. After demineralization, the slices were washed three times with deionized water and dried in a desiccator for 24 hours. The specimens were randomly divided into four groups (n=3) according to the concentration of acacia extract (1%, 2%, or 3%) and immersion in deionized water (control). The specimens were immersed in the acacia extract solutions or deionized water for 1 hour at room temperature.^8,37  After incubation, the slices were washed three times in deionized water and dried in a desiccator for 24 hours. The spectra of the dried specimens were obtained by a Fourier transform infrared (FTIR) spectrophotometer (Shimadzu, Toquio, Japan) at a resolution of 4 cm^-1. Each specimen was placed on an attenuated total reflectance (ATR) plate (Perkin-Elmer, Waltham, MA, USA). The ATR crystal is zinc selenite (ZnSe) with a transmission range between 650 and 4000 cm^-1.³⁸ 

The sample size per group was determined based on a pilot test using the following parameters: α = 0.05, a power of 80%, a difference of 7.25 % between groups, and an expected standard deviation of residuals of 4.68 %. The minimum sample size required was 10 specimens per group.

The dentin beams were demineralized in 10% phosphoric acid for 24 hours at room temperature to expose the organic matrix (collagen). Afterward, the specimens were washed three times with deionized water and dried in a desiccator for 72 hours. After drying, the specimens were randomly divided into four groups (n=10) and weighed individually on a digital precision scale (Shimadsu, Kyoto, Japan) to obtain the dry mass. After weighing, each specimen was placed inside an Eppendorf tube, which received 1 mL of the following solutions according to the group: group 1 (control)–deionized water; group 2–1% acacia extract; group 3–2% acacia extract; and group 4–3% acacia extract. The specimens remained immersed in the solutions for 1 hour at room temperature. Afterward, the acacia extract solution was removed from the tube with a syringe. Then, the specimens were washed three times in deionized water and digested in 1 mL of type I bacterial collagenase solution (200 μg/mL Clostridium histolyticum) (Sigma-Aldrich, Saint Louis, MO, USA) in HEPES buffer at a pH value of 7.4. The Eppendorf tubes with the specimens were first agitated in a vortex (Glas-Col, Terre Haute, IN, USA) for 20 minutes and then stored in a culture oven (Biomatic, Porto Alegre, RS, Brazil) at 40°C for 24 hours. After this period, the collagenase solution was removed with a pipette, and the specimen inside the Eppendorf tube was washed three times with deionized water. The specimens were then dried in a desiccator for 72 hours and weighed again to obtain the dry mass after incubation.

The collagen biodegradation rates (R) were determined by the following formula: R (%) = 100 – (M₂ x 100)/M₁; where M₁ is the dry mass before incubation and M₂ is the dry mass after bacterial collagenase digestion.⁸ 

According to a previous study, the mean μTBS to dentin (± standard deviation) of Adper Scotchbond Multipurpose Plus is 32.44 ± 6.07 MPa.³⁹  To detect a difference equivalent of 1.9 standard deviations between the groups tested, using a two-sided test with α = 0.05 and a power of 80%, the estimated minimum sample size required for each group was seven teeth.

In 28 sound human third molars, the occlusal enamel surface was removed with a double-sided diamond disc mounted on a cutting machine with a water cooling system, exposing a flat surface of occlusal dentin. The dentin surface was finished on a polisher (DPU-10, Panambra, São Paulo, SP, Brazil) with 600-grit silicon carbide sandpaper for 30 seconds. The teeth were randomly divided into four groups (n=7) using the Randomizer Android app.

Group 1 (control): Adper Scotchbond Multipurpose Plus adhesive system (3M, St. Paul, MN, USA): The dentin was conditioned with 37% phosphoric acid for 15 seconds, followed by rinsing for 20 seconds. The excess water was removed with absorbent paper. The primer was applied to dentin, followed by a light jet of air for 5 seconds. Then, the adhesive was applied and light cured for 10 seconds with an LED unit (Radii-cal, SDI, Vic., Australia). A Filtek Z250 resin composite (3M/ESPE, St. Paul, MN, USA) block that was approximately 6 mm high was built in three increments. Each increment was light cured for 40 seconds. The light intensity of the light-curing device was monitored with a radiometer (Ecel, São Leopoldo, RS, Brazil) to maintain an intensity of 900-1,000 mW/cm², and the tip distance to specimen was standardized in approximately 2 mm.

Group 2 (1% acacia extract): The dentin was conditioned, rinsed, and dried as described in group 1. Next, 1% acacia extract was actively applied with a microbrush for 1 minute. The excess solution was removed with absorbent paper, followed by the application of a primer, adhesive, and resin composite as described in group 1.

Group 3 (2% acacia extract): Group 3 underwent the same procedure as group 2, but with 2% acacia extract being applied instead of 1% acacia extract.

Group 4 (3% acacia extract): Group 4 underwent the same procedure as group 2, but with 3% acacia extract being applied instead of 1% acacia extract.

The adhesive procedures were performed by a single operator.

The teeth/resin composite block sets were stored in distilled water at 37°C for 24 hours. After storage, the teeth/composite resin sets were cut parallel to the long axis of the tooth into approximately 0.90 x 0.90 mm transverse sections in a Labcut 1010 cutting machine with a double-sided diamond disc at a speed of 480 rpm under water cooling. In this manner, beam-shaped specimens were obtained, in which the upper half was made of composite resin and the lower half was made of dentin, measuring approximately 0.80 mm².^39,40  From each tooth, 12 to 18 beams were obtained. Half of the specimens of each tooth were submitted to the microtensile bond strength (μTBS) test immediately after cutting, and the other half were stored in distilled water at 37°C for a period of 6 months. Distilled water was changed weekly.

The area corresponding to the adhesive bond of each specimen was first measured in mm² using a digital caliper (Mitutoyo, Suzano, SP, Brazil), with a maximum stated error of 0.01 mm, and then submitted to the μTBS test. The specimens were individually attached to the microtensile device using a cyanoacrylate-based adhesive (Superbonder Gel-Loctite, São Paulo, SP, Brazil) associated with an accelerator (Zip Kicker, Pacer, Rancho Cucamonga, CA, USA). The test was performed on the EMIC DL-2000 universal testing machine (EMIC, São José dos Pinhais, PR, Brazil) with a cross-head speed of 0.5 mm/min and a 50 N load cell. The results were obtained in MPa.

After the test, the failures were observed under a stereoscopic magnifying glass (Olympus Corp, Tokyo, Japan) at 40× magnification and classified as follows: adhesive (failure at the adhesive/dentin interface), cohesive (failure exclusively in the dentin or resin composite), and mixed (adhesive interface failure including cohesive failure in dentin and/or resin composite).

Descriptive analyses of the interactions among different concentrations of acacia extract and collagen and the types of failures obtained after the μTBS test were conducted.

The collagen biodegradation rate data were submitted to one-way ANOVA and Tukey post hoc test.

The mean μTBS (MPa) value of all beams from the same tooth was calculated for statistical purposes. Pretest failures were not included in the average value for each tooth. The mean values of μTBS for each group were expressed as the mean of the seven teeth used per group. μTBS data were subjected to two-way ANOVA (treatment x storage time) with blocking, followed by the Tukey test. The significance level was 5%.

Representative FTIR spectra of untreated specimens and specimens treated with acacia extract at concentrations of 1%, 2%, and 3% are shown in Figure 1. FTIR spectra ranging from 1800 cm^-1 to 1200 cm^-1 revealed that the most prominent changes between untreated and treated collagen occurred in the amide II region at ~1530 cm^-1 and at ~1400 cm^-1. Treatment with acacia extract at the three concentrations induced a shift in aromatic vibration at ~1530 cm^-1 in the amide II region and a reduction in the peak intensity at ~1400 cm^-1.

According to ANOVA, a significant difference occurred in the collagen biodegradation rates among the groups (p<0.05). According to the Tukey test, group 1 (control) obtained the highest biodegradation rate (100%), which was significantly greater than that of the other groups. The biodegradation rates of groups 2 (24%) and 3 (23%) did not differ significantly. Group 4 (17%) obtained the lowest biodegradation rate, which did not differ significantly from that of group 3 (Table 1).

According to the two-way ANOVA, the treatment factor was not significant (p=0.318), the storage factor was significant (p=0.0001), and the treatment x storage interaction was significant (p=0.009). According to the Tukey test, the mean μTBS did not differ significantly between groups in the immediate evaluation (after 24 hours). However, after 6 months of storage, relatively high μTBS mean values were obtained for group 2 (17.17 MPa) and group 3 (17.38 MPa), which were not significantly different from each other and were higher than group 1 (14.06 MPa). Group 4 obtained an intermediate μTBS mean value (15.50 MPa), which was not significantly different from the values of group 1, group 2, and group 3 (Table 2).

Only one group exhibited a significant decrease in μTBS after six months of storage. In the other groups, μTBS did not exhibit a significant difference between 24 hours and six months of storage.

The failures were predominantly mixed in all groups at 24 hours and six months of storage. There was an increase in the number of pretest failures in the control group after 6 months of storage (Figure 2).

Acacia extract at different concentrations was applied to human dentin of extracted teeth as a biomodifying agent. The aims of the study were to determine if this treatment strengthens the dentin collagen exposed by acid etching and controls dentin biodegradation.^10,41,42 

The FTIR spectra of the specimens treated with acacia extract demonstrated shifts in aromatic vibration at ~1530 cm^-1 in the amide II region, indicating hydrogen bonding interactions.⁴³  A drop in peak intensity at ~1400 cm^-1 was also observed, which was caused by the dehydration of type I collagen.^38,44,45  Polyphenols, such as PAs present in acacia extract above 45% (w/w), could displace water between collagen microfibrils and create new hydrogen bonds between fibrils. Then, PA would dehydrate and improve the biological stability of the tissue by forming imine C = N bonds as a probable mechanism of action.²⁵  Therefore, it could be inferred that the PA contained in acacia extract crosslinked collagen. Thus, the first null hypothesis was rejected.

The exact mechanism of interaction between PAs and collagen is not completely understood. However, different types of interactions have been proposed, such as covalent bonding,¹⁰  hydrogen bonding,^12,46  ionic bonding,⁴⁷  and hydrophobic interactions.²⁰  Hydrogen bonds between the protein amine carbonyls in collagen and phenolic hydroxyls of PA have been considered the most important forces for stabilizing PA-treated collagen fibrils.²⁵  PA could also induce covalent bonds in type I collagen, increasing the forces of interactions between collagen fibrils.³⁷ 

PAs are structured by monomeric building blocks catechins, which are flavan-3- ol units that could be linked by an additional ether bond (C-O) or one or more C-C bonds to form oligomers and polymers. The oligomerization degree could influence PA interactions with collagen at distinct levels.⁸  PAs with high oligomeric forms would be responsible for both intramolecular crosslinking, providing biostability to the collagen molecule, and intermolecular and interfibrillar crosslinking, improving mechanical properties.⁸  Then, the application of crosslinkers to dentin collagen could increase the stiffness^29,48  and ultimate tensile strength of demineralized dentin.²⁸  Additionally, Polassi and others⁴⁹  showed that solution obtained from Acacia decurrens induced an increase of 70.8% of the dentin elastic modulus. As a consequence, collagen resistance to biodegradation could increase, making collagen fibrils increasingly resistant to enzymatic challenges.^{24–26,38,48}  This cross-linking effect justified the biodegradation rates found in the present study.

In the control group, the specimens were not treated with acacia extract; instead, they were immersed in deionized water, and the biodegradation rate was 100%. The specimens treated with acacia extract exhibited a significantly reduced biodegradation rate, demonstrating that all concentrations of acacia extract were effective in improving collagen sustainability under collagenolytic stress. Thus, the second null hypothesis was rejected. However, comparing the three concentrations, 3% acacia extract led to a significantly lower collagen biodegradation rate than 1% acacia extract. Seemingly, the higher the concentration was, the lower the biodegradation. Therefore, it would be interesting to investigate whether concentrations higher than 3% could favor complete protection of collagen against collagenolytic degradation.

In this study, the immersion of the specimens in the acacia extract was standardized for 1 hour, which is not realistic or applicable in the clinic. Relatively short times, such as 10 seconds or 1 minute, would be feasible for clinical practice. However, as this study was the first involving an evaluation of acacia extract as a crosslinking agent, immersion for 1 hour was stipulated to allow sufficient contact time between the collagen and solutions. We estimated that short times could favor collagen crosslinking, as demonstrated in the study by Liu and others,³⁸  in which 10 seconds of treatment with PA from grape seed was effective. Therefore, for future studies, short immersion times should be evaluated.

The acacia extract was applied to dentin after etching with 37% phosphoric acid. This would add an additional clinical step to the adhesive system protocol, which could be criticized and considered a clinically negative aspect. However, as an initial study, applying acacia extract as a separate step was simpler than trying to incorporate it into the adhesive system, since PA could influence the degree of monomer/polymer conversion and polymerization kinetics in adhesive systems⁵⁰  due to its radical scavenging ability.⁵¹  Other studies incorporated the tested compounds in adhesive systems⁵²  and in phosphoric acid,⁵³  and this testing could be conducted in future studies.

For the μTBS test, acacia extract at different concentrations was applied to dentin for 1 minute, as longer times than this would be clinically inappropriate. After this time, the acacia extract was not rinsed; only the excess solution was removed with absorbent paper. Therefore, it is possible that the remaining solution partially mixed with the primer and was incorporated into the hybrid layer. At the 24-hour evaluation, there was no significant difference in the μTBS between the control and experimental groups, showing that the acacia extract, regardless of the concentration, did not interfere with the bonding to dentin. This result was positive. If the acacia extract had negatively affected the μTBS, the application of this compound to dentin would be contraindicated. Thus, the third null hypothesis was not rejected.

At the six-month evaluation, μTBS decreased significantly only in the control group. The collagen fibrils can remain unprotected by resin monomers at the base of the hybrid layer, favoring hydrolytic and enzymatic degradation,⁵⁴  and justifying the decrease in μTBS for the control group. However, in the specimens treated with acacia extract, μTBS did not significantly decrease. These findings corroborated the results of the interaction by FTIR analysis and the biodegradation rate, showing that the acacia extract decreased the degradation at the resin-dentin interface. Thus, the fourth null hypothesis was rejected.

Different factors could explain the stability of μTBS for groups treated with acacia extract: (1) the PA contained in acacia extract changed the mechanical properties of collagen, increasing the stiffnesses of collagen fibrils,^28,48,49  and making them increasingly resistant to biodegradation;²⁴  (2) PA favored the inhibition of MMPs,⁴¹  which had been activated by the low pH of acids;⁵⁵  and (3) the ability of acacia extract solutions to modify the contact angle of exposed collagen fibrils allowed better wetting of the primer and adhesive.⁴⁹  Liu and Wang²⁴  suggested that the synergistic action of two or more factors contributed to the stability of the μTBS at the resin-dentin interface during the six-month storage period.

Regarding the failure mode, the number of adhesive failures increased in the control group after six months of storage; this phenomenon was not observed in the groups treated with acacia extract. In addition, the number of pretest failures was relatively high in the control group. This evidence suggests greater degradation of the resin-dentin interface in the specimens not treated with the acacia extract due to the absence of the collagen crosslinking effect.

In this study, a multi-bottle etch-and-rinse adhesive system was applied. To date, practitioners tend to choose multimode universal adhesives using the self-etch strategy on dentin due to their being less technique-sensitive and more user-friendly than other options.²  Thus, it would be relevant to test acacia extract with this category of adhesive system, since they exhibited degradation of the collagen fibrils along the thin hybrid layers.⁵⁶ 

Water was used as a solvent to prepare the acacia extract solution. However, the solvents in polyphenol treatment solutions could affect the biostability and collagen crosslinking interactions¹¹  since the quantity of probable hydrogen bonding was dependent on the solvent.⁵⁷  If acetone or ethanol solvents were used, new sites of hydrogen bonding continued to be available on the PA molecule to interact with collagen.⁵⁷  Ethanol, as a protic solvent, raised the level of interpeptide hydrogen bonding between adjacent collagen fibrils and disrupted collagen structure by substituting water with the solvent molecule.⁵⁸  According to Hagerman and Klucher,⁵⁹  ethanol seems to be the best solvent since it could stimulate PA-collagen interactions and prolong its stability by diminishing the dielectric constant of the media. Therefore, it would be relevant to evaluate the acacia extract diluted in ethanol. In addition, most modern adhesive systems contain ethanol or a combination of water/ethanol as solvent.⁶⁰ 

The limitations of the present study are related to the short storage time (six months) of the specimens to assess the μTBS. Furthermore, immersion of the specimens in the acacia extract for 1 hour to evaluate the interactions by FTIR and the resistance rate to enzymatic biodegradation is not clinically feasible; thus, short times should be tested. In addition, the ability of acacia extract to inhibit MMPs should be investigated. These are topics for future studies.

Within the limitations of this laboratory study, it can be concluded that acacia extract at concentrations of 1%, 2%, and 3% interacted with human dentin collagen, reduced collagen biodegradation, and favored the stabilization of the dentin-resin interface after six months of storage without compromising the dentin-resin bond strength.

This study was financed in part by the Coordenaçâo de Aperfeiçoamento de Pessoal de Nível Superior, Brasil (CAPES), Finance Code 001.