Strategies to prevent hydrolytic degradation of the hybrid layer—A review

Abstract

Objective

Endogenous dentin collagenolytic enzymes, matrix metalloproteinases (MMPs) and cysteine cathepsins, are responsible for the time-dependent hydrolysis of collagen matrix of hybrid layers. As collagen matrix integrity is essential for the preservation of long-term dentin bond strength, inhibition of endogenous dentin proteases is necessary for durable resin-bonded restorations.

Methods

Several tentative approaches to prevent enzyme function have been proposed. Some of them have already demonstrated clinical efficacy, while others need to be researched further before clinical protocols can be proposed. This review will examine both the principles and outcomes of techniques to prevent collagen hydrolysis in dentin–resin interfaces.

Results

Chlorhexidine, a general inhibitor of MMPs and cysteine cathepsins, is the most tested method. In general, these experiments have shown that enzyme inhibition is a promising approach to improve hybrid layer preservation and bond strength durability. Other enzyme inhibitors, e.g. enzyme-inhibiting monomers, may be considered promising alternatives that would allow more simple clinical application than chlorhexidine. Cross-linking collagen and/or dentin matrix-bound enzymes could render hybrid layer organic matrices resistant to degradation. Alternatively, complete removal of water from the hybrid layer with ethanol wet bonding or biomimetic remineralization should eliminate hydrolysis of both collagen and resin components.

Significance

Understanding the function of the enzymes responsible for the hydrolysis of hybrid layer collagen has prompted several innovative approaches to retain hybrid layer integrity and strong dentin bonding. The ultimate goal, prevention of collagen matrix degradation with clinically applicable techniques and commercially available materials may be achievable in several ways.

Introduction

About 50 vol.% of dentin is composed of minerals, the rest being type I collagen and non-collagenous proteins (30 vol.%) and water (20 vol.%) . During the bonding of composite restorations, the surface and subsurface mineral component of dentin is removed either totally by acid etching in etch-and-rinse (E&R) adhesives or partially with acidic primers or adhesives in self-etch (SE) adhesives. The exposed collagen matrix is then infiltrated with solvated adhesive resin comonomers ( Fig. 1 ). In reality, however, adhesive monomers are not able to fully encapsulate the exposed collagen matrix, leaving totally or partially exposed collagen fibrils at the bottom of the hybrid layer, lacking the protection of polymerized resin. The lack of resin protection and presence of water leaves demineralized collagen fibrils vulnerable to time-dependent hydrolytic degradation.

Fig. 1
Transmission electron microscope (TEM) image of undemineralized, unstained human tooth showing the dentin–adhesive interface created with 2-step E&R adhesive (Scotchbond 1XT, 3M ESPE). (A) On the top of the mineralized dentin (md) is the hybrid layer (hl), where the exposed dentin collagen mesh is infiltrated with adhesive monomers, creating a mechanical interlock between dentin-bound collagen and polymerized adhesive. On top of the hybrid layer, adhesive (a) forms a chemical bond with the restorative resin composite (c). Adhesive resin tags (rt) penetrate into dentinal tubules, sealing them and providing additional retention. (B) Higher magnification image from the area marked with box in (A), with collagen matrix readily seen in the hybrid layer, even in unstained sections.
Images courtesy of BDs Pekka Mehtälä and Dr. Saulo Geraldeli.

Degradation of collagen fibrils and hydrophilic resin components leads to destruction of the hybrid layer and loss of dentin bond strength over time. We recently reviewed in detail the presence, role and function of collagen-degrading enzymes matrix metalloproteinases (MMPs) and cysteine cathepsins in dentin . The increased knowledge of the function of these enzymes in the hybrid layer degradation has lead to extensive research activity aiming to prevent collagenolysis in the dentin–resin interface. This review focuses on different strategies that have been developed to control and prevent the hydrolytic enzyme-related loss of the hybrid layer collagen and bond strength.

Enzyme inhibition

Chlorhexidine

The vast majority of the experiments aimed to improve the durability of dentin bonds using enzyme inhibition have been performed with chlorhexidine (CHX), a potent antimicrobial agent. CHX inhibits effectively MMP-2, -9 and -8 , and cysteine cathepsins . In 2004, Pashley et al. presented convincing evidence of its efficacy in inhibiting dentin collagenolytic enzymes. Since then, several studies have demonstrated that CHX can preserve the structural integrity of hybrid layer collagen matrix and reduce time-dependent reduction in dentin bond strength both in vivo and in vitro (Table 1: online supplementary data). With simplified E&R adhesives the loss of bond strength in control ( i.e. uninhibited) teeth over 1–2 years has been approximately 50%; teeth pretreated with CHX have shown 20–25% loss of bond strength (Table 1: online supplementary data). The effect of CHX on bond durability with 3-step E&R adhesives has been tested only in a few studies . While in all studies CHX resulted with slightly better long-term bond strengths, no statistically significant differences were found. The reason may be the relatively low bond strength loss in the controls, demonstrating again the better long-term function of 3-step E&R adhesives compared to their simplified 2-step versions.

Table 1

Experiments aiming to improve the durability of dentin bond strength by elimination of collagen degradation. Material names as presented in the article. Permanent teeth with microtensile bond strength testing method were used, unless otherwise mentioned.

With SE adhesives, fewer studies have been performed, and the results have been conflicting. However, if at least 0.5% CHX was used, preservation of dentin bond strength in aged samples can be seen , while concentrations lower than 0.1% may not be sufficient , with one study as an exception. Mobarak using Clearfil SE Bond (Kuraray) and 2 and 5% CHX concentrations in both carious and intact dentin demonstrated partial preservation of bond strength only in caries-affected dentin with 5% CHX after 24-month storage. The comparison with other studies is difficult because the microshear testing was used , while all the other studies used microtensile method. Sectioning of specimens into microtensile beams before storage reduces cross-sectional area of specimens, allowing faster water diffusion through the hybrid layer . This mimicks the in vivo conditions where dentinal fluid flow enters hybrid layer from the tubules due to pulpal hydrostatic pressure, a phenomenon called micropermeability . Indeed, composite restorations aged in vivo have comparable reduction in microtensile bond strengths to the specimens aged as beams in vitro , and in vitro introduced pulpal pressure is almost as effective as beam storage .

The importance of dentinal fluid flow may be questioned by the reduced permeability of caries-affected dentin with reduced fluid flow. Transparent caries-affected dentin can provide an excellent seal along the bonded interfaces even under simulated dentinal fluid flow, but the highly porous nature of the interface raises questions about the immediate and long term bond strengths . Carious dentin contains high collagenolytic activity . Partially demineralized caries-affected dentin that can be hundreds of micrometers thick may be prone to self-destruction via the release of endogenous collagenolytic enzymes similarly to that observed with an incompletely infiltrated hybrid layer produced by simplified E&R adhesives . The loss of dentin collagen matrix integrity has even been observed in intact dentin under the hybrid layer after prolonged (3 years) in vitro storage . Indeed, lower immediate and long-term bond strengths have been observed with caries-affected dentin as a substrate . The patency of tubules in caries-affected dentin may also vary, and the fluid flow under restorations may not necessarily be reduced significantly compared to intact dentin due to different levels of tubular occlusion . Taken together, the potential reduction of fluid flow through caries-affected dentin cannot be considered to reduce the overall loss of dentin bond strength.

The dose-dependency may be related to the mode of MMP-inhibiting action of CHX. Even though CHX binding rate to mineralized dentin is almost 80% lower that to demineralized dentin , low (0.05–0.2%) CHX concentrations are sufficient to completely inhibit the collagenolytic activity of untreated (undemineralized) dentin powder , while 0.5–2.0% concentrations are able to only partially inhibit the activity induced with acidic SE primer . Since MMPs require calcium to maintain their tertiary structure and zinc ions for their catalytic activity also in dentin , and CHX loses its MMP inhibition in the presence of calcium chloride , CHX-related MMP inhibition may be related to its chelating property and calcium ions released by the primer may be responsible for the loss of inhibition by CHX. This is supported by the finding that treating dentin powder with Clearfil SE Bond primer for 2 min instead of 20 s not only increased the collagenolytic activity, but also caused the loss of inhibition by 0.5% and 1.0% CHX, with only 2.0% CHX showing significant inhibition . Also the recent reports of chlorhexidine binding by dentin collagen suggest that collagen may compete with MMPs for CHX binding, requiring the use of relatively high CHX concentrations.

Incorporation of CHX at reasonably low (0.2–2.0%) concentration into methacrylate comonomers has no effect on water sorption, cause a slight decrease in conversion rate, but may even increase flexural strength and modulus of elasticity . Drug release from the polymerized adhesive is concentration-dependent and retains a slow steady-state level . Since the idea behind the incorporation of inhibitors into adhesive is their continuous release to prevent collagen degradation for a prolonged time, these results are promising.

Carrilho et al. examined all the fractured samples bonded using the 2-step E&R adhesive (Single Bond, 3M ESPE). In controls, the failures at the bottom of the hybrid layer ( Fig. 2 A ) increased dramatically after 6-months of storage ( Fig. 3 ). CHX significantly reduced the failures at the bottom of the hybrid layer when compared to the control group, and completely eliminated the cohesive failures in dentin in aged samples ( Fig. 3 ). The cohesive failures in the adhesive layer ( Fig. 2 B) and composite in CHX-treated aged samples increased significantly, while in controls, clear reductions were seen especially in cohesive composite failures ( Fig. 3 ). These findings indicate that with E&R adhesives, the weakest link in the bonded complex was the area of the totally or partially denuded collagen at the bottom or below the hybrid layer where the fibrils lack appropriate protection by adhesive ( Fig. 2 C). Inhibiting collagenolytic enzymes using CHX eliminates or slows down collagen degradation, and shifts the major site of failure elsewhere. This conclusion has later been supported by other studies .

Fig. 2
(A) SEM image of the fracture occurring at the bottom of the hybrid layer. Dentinal tubules are mostly exposed, with few dentinal tubules containing remaining resin tags. Partially degraded collagen at the bottom of the hybrid layer gives can be seen (asterisk). (B) SEM image of a cohesive fracture localized in the middle of the hybrid layer. Dentinal tubules are completely filled by resin tags (black arrow), and intertubular dentin is covered by adhesive (asterisk). (C) Schematic presentation of resin-bonded acid-etched dentin covered with resin composite. The acid-etched tubules no longer contain peritubular dentin, making the tubules twice their normal diameter. Resin tags extend down from the adhesive layer. The tags are hybridized with the surrounding demineralized dentin as they pass through the hybrid layer. There is no such hybridization of the resin tags as it passes into mineralized dentin. As poorly infiltrated hybrid layers age, the collagen fibrils degrade and disappear. In such hybrid layers, water replaces the collagen. The spaces in the composite are due to hydrolysis of nanofillers of silica from the resin composite. These, too become filled with water.
Figures A and B reproduced from Carrilho et al. , with permission.

Fig. 3
The effect of 2% CHX pretreatment on the distribution of failure modes (in percentage) in vitro , as observed with SEM. External inhibitor indicates the absence or presence of a protease inhibitor cocktail used in incubation medium (artificial saliva, AS). In immediate testing, no differences between the fracture modes were detected. After 6-month incubation, statistically significant increase in failures located at the bottom of the hybrid layer was seen in control group, but not in CHX group. External inhibitors in AS significantly reduced the failures at the bottom of the hybrid layer in controls, indicating partial elimination of endogenic enzyme function; respective effect in CHX-treated samples was non-significant.
Data from Carrilho et al. .

Enzyme inhibition

Chlorhexidine

The vast majority of the experiments aimed to improve the durability of dentin bonds using enzyme inhibition have been performed with chlorhexidine (CHX), a potent antimicrobial agent. CHX inhibits effectively MMP-2, -9 and -8 , and cysteine cathepsins . In 2004, Pashley et al. presented convincing evidence of its efficacy in inhibiting dentin collagenolytic enzymes. Since then, several studies have demonstrated that CHX can preserve the structural integrity of hybrid layer collagen matrix and reduce time-dependent reduction in dentin bond strength both in vivo and in vitro (Table 1: online supplementary data). With simplified E&R adhesives the loss of bond strength in control ( i.e. uninhibited) teeth over 1–2 years has been approximately 50%; teeth pretreated with CHX have shown 20–25% loss of bond strength (Table 1: online supplementary data). The effect of CHX on bond durability with 3-step E&R adhesives has been tested only in a few studies . While in all studies CHX resulted with slightly better long-term bond strengths, no statistically significant differences were found. The reason may be the relatively low bond strength loss in the controls, demonstrating again the better long-term function of 3-step E&R adhesives compared to their simplified 2-step versions.

Table 1

Experiments aiming to improve the durability of dentin bond strength by elimination of collagen degradation. Material names as presented in the article. Permanent teeth with microtensile bond strength testing method were used, unless otherwise mentioned.

With SE adhesives, fewer studies have been performed, and the results have been conflicting. However, if at least 0.5% CHX was used, preservation of dentin bond strength in aged samples can be seen , while concentrations lower than 0.1% may not be sufficient , with one study as an exception. Mobarak using Clearfil SE Bond (Kuraray) and 2 and 5% CHX concentrations in both carious and intact dentin demonstrated partial preservation of bond strength only in caries-affected dentin with 5% CHX after 24-month storage. The comparison with other studies is difficult because the microshear testing was used , while all the other studies used microtensile method. Sectioning of specimens into microtensile beams before storage reduces cross-sectional area of specimens, allowing faster water diffusion through the hybrid layer . This mimicks the in vivo conditions where dentinal fluid flow enters hybrid layer from the tubules due to pulpal hydrostatic pressure, a phenomenon called micropermeability . Indeed, composite restorations aged in vivo have comparable reduction in microtensile bond strengths to the specimens aged as beams in vitro , and in vitro introduced pulpal pressure is almost as effective as beam storage .

The importance of dentinal fluid flow may be questioned by the reduced permeability of caries-affected dentin with reduced fluid flow. Transparent caries-affected dentin can provide an excellent seal along the bonded interfaces even under simulated dentinal fluid flow, but the highly porous nature of the interface raises questions about the immediate and long term bond strengths . Carious dentin contains high collagenolytic activity . Partially demineralized caries-affected dentin that can be hundreds of micrometers thick may be prone to self-destruction via the release of endogenous collagenolytic enzymes similarly to that observed with an incompletely infiltrated hybrid layer produced by simplified E&R adhesives . The loss of dentin collagen matrix integrity has even been observed in intact dentin under the hybrid layer after prolonged (3 years) in vitro storage . Indeed, lower immediate and long-term bond strengths have been observed with caries-affected dentin as a substrate . The patency of tubules in caries-affected dentin may also vary, and the fluid flow under restorations may not necessarily be reduced significantly compared to intact dentin due to different levels of tubular occlusion . Taken together, the potential reduction of fluid flow through caries-affected dentin cannot be considered to reduce the overall loss of dentin bond strength.

The dose-dependency may be related to the mode of MMP-inhibiting action of CHX. Even though CHX binding rate to mineralized dentin is almost 80% lower that to demineralized dentin , low (0.05–0.2%) CHX concentrations are sufficient to completely inhibit the collagenolytic activity of untreated (undemineralized) dentin powder , while 0.5–2.0% concentrations are able to only partially inhibit the activity induced with acidic SE primer . Since MMPs require calcium to maintain their tertiary structure and zinc ions for their catalytic activity also in dentin , and CHX loses its MMP inhibition in the presence of calcium chloride , CHX-related MMP inhibition may be related to its chelating property and calcium ions released by the primer may be responsible for the loss of inhibition by CHX. This is supported by the finding that treating dentin powder with Clearfil SE Bond primer for 2 min instead of 20 s not only increased the collagenolytic activity, but also caused the loss of inhibition by 0.5% and 1.0% CHX, with only 2.0% CHX showing significant inhibition . Also the recent reports of chlorhexidine binding by dentin collagen suggest that collagen may compete with MMPs for CHX binding, requiring the use of relatively high CHX concentrations.

Incorporation of CHX at reasonably low (0.2–2.0%) concentration into methacrylate comonomers has no effect on water sorption, cause a slight decrease in conversion rate, but may even increase flexural strength and modulus of elasticity . Drug release from the polymerized adhesive is concentration-dependent and retains a slow steady-state level . Since the idea behind the incorporation of inhibitors into adhesive is their continuous release to prevent collagen degradation for a prolonged time, these results are promising.

Carrilho et al. examined all the fractured samples bonded using the 2-step E&R adhesive (Single Bond, 3M ESPE). In controls, the failures at the bottom of the hybrid layer ( Fig. 2 A ) increased dramatically after 6-months of storage ( Fig. 3 ). CHX significantly reduced the failures at the bottom of the hybrid layer when compared to the control group, and completely eliminated the cohesive failures in dentin in aged samples ( Fig. 3 ). The cohesive failures in the adhesive layer ( Fig. 2 B) and composite in CHX-treated aged samples increased significantly, while in controls, clear reductions were seen especially in cohesive composite failures ( Fig. 3 ). These findings indicate that with E&R adhesives, the weakest link in the bonded complex was the area of the totally or partially denuded collagen at the bottom or below the hybrid layer where the fibrils lack appropriate protection by adhesive ( Fig. 2 C). Inhibiting collagenolytic enzymes using CHX eliminates or slows down collagen degradation, and shifts the major site of failure elsewhere. This conclusion has later been supported by other studies .

Fig. 2
(A) SEM image of the fracture occurring at the bottom of the hybrid layer. Dentinal tubules are mostly exposed, with few dentinal tubules containing remaining resin tags. Partially degraded collagen at the bottom of the hybrid layer gives can be seen (asterisk). (B) SEM image of a cohesive fracture localized in the middle of the hybrid layer. Dentinal tubules are completely filled by resin tags (black arrow), and intertubular dentin is covered by adhesive (asterisk). (C) Schematic presentation of resin-bonded acid-etched dentin covered with resin composite. The acid-etched tubules no longer contain peritubular dentin, making the tubules twice their normal diameter. Resin tags extend down from the adhesive layer. The tags are hybridized with the surrounding demineralized dentin as they pass through the hybrid layer. There is no such hybridization of the resin tags as it passes into mineralized dentin. As poorly infiltrated hybrid layers age, the collagen fibrils degrade and disappear. In such hybrid layers, water replaces the collagen. The spaces in the composite are due to hydrolysis of nanofillers of silica from the resin composite. These, too become filled with water.
Figures A and B reproduced from Carrilho et al. , with permission.

Fig. 3
The effect of 2% CHX pretreatment on the distribution of failure modes (in percentage) in vitro , as observed with SEM. External inhibitor indicates the absence or presence of a protease inhibitor cocktail used in incubation medium (artificial saliva, AS). In immediate testing, no differences between the fracture modes were detected. After 6-month incubation, statistically significant increase in failures located at the bottom of the hybrid layer was seen in control group, but not in CHX group. External inhibitors in AS significantly reduced the failures at the bottom of the hybrid layer in controls, indicating partial elimination of endogenic enzyme function; respective effect in CHX-treated samples was non-significant.
Data from Carrilho et al. .

EDTA

Chelation of calcium and zinc with ethylene diamine tetraphosphonic acid (EDTA) inactivates dentinal MMPs and preserves dentin mechanical properties . EDTA as an etchant has also been suggested to create a hybrid layer that would be more resistant to degradation and produce higher immediate bond strengths with experimental ethanol wet bonding technique . However, the hybrid layer collagen resistance in these studies was tested with exposure to sodium hypochloride (NaOCl) . NaOCl dissolves completely or partially exposed proteins regardless of the presence, absence or inhibition of collagen bound-enzymes. Since long-term aging studies, allowing dentin endogenous collagenolytic enzyme function, have not yet been done, durability of bond strengths remains to be demonstrated. Washing EDTA away after demineralization leads to significant reduction in mechanical properties of EDTA-demineralized dentin beams accompanied with degradation of collagen already in 1-week incubation , indicating the MMP inhibition is reversible . Also the time needed for efficient EDTA-etching for restorative purposes limits its clinical use in adhesive dentistry .

Synthetic MMP inhibitors

Tetracyclines are antibiotics with cationic chelating properties, and inhibit MMP extracellularly . CMTs are chemically modified tetracyclines that lack antibacterial activity, but have retained their MMP-inhibition capacity . Doxycycline strongly decreases dentin matrix degradation , and CMT-3 (aka Metastat, COL-3, one of the most potent CMTs) is particularly effective in inhibiting MMPs in dentinal caries lesions . Bisphosphonates (BPs) are pyrophosphate analogs with a high affinity for hydroxyapatite crystals and they are used to treat conditions involving increased bone resorption, e.g. Paget’s disease and osteoporosis. Hydroxamate-based BPs, such as Batimastat and Galardin, also inhibit MMPs by chelating active-site zinc . Zoledronic acid (zoledronate) is effective in inhibiting carious dentin MMPs .

Galardin (aka GM6001 or Ilomastat) has a collagen-like backbone that binds to the MMP active site and a hydroxamate structure which chelates the MMP catalytic domain zinc . Galardin has potent inhibitory activity against MMP-1, -2, -3, -8 and -9 . Galardin has been shown to reduce the loss of bond strength comparable to the effect of CHX (Table 1: online supplementary data). The effect was questioned by three studies, where Galardin and Batimastat (aka BB94) and SB-3CT (thiol-based selective gelatinase MMP inhibitor) failed to preserve the bond strength in aged samples. In these studies, MMP inhibitors were used in very low (5–10 μmolar) concentrations that have been shown to be effective against soluble enzymes, while Breschi et al. used 0.2 mM Galardin (the highest possible concentration to achieve a saturated water solution). Since dentin MMPs in the hybrid layer remain bound to collagen, the concentration of inhibitors required to complete inhibition may be much higher in dentin than with unbound soluble MMPs.

Quaternary ammonium group

Both E&R and SE adhesives have been shown to activate dentin MMPs, and they may at least partially be responsible for the gelatinolytic activity observed in the hybrid layer . Adhesive monomers that would possess enzyme-inhibiting properties offer an appealing alternative to prevent hydrolytic degradation of hybrid layer collagen.

Polymerizable quaternary ammonium methacrylates (QAMs), especially 12-methacryloyloxydodecylpyridinium bromide (MDPB) have been incorporated into SE primers because they possess antimicrobial properties and can copolymerize with adhesive monomers . Similar to CHX, these compounds are cationic, water-soluble, but unlike CHX they may not leach out of bonded interfaces. QAMs inhibit soluble MMP-9 as or more effectively as Galardin, and almost completely inhibited the demineralized dentin collagen degradation . MDPB (a component of Clearfil Protect Bond and Clearfil Protect SE) proved to be among the most effective . In vitro and clinical experiments have also indicated that QAMs (namely MDPB in Clearfil Protect Bond) may inhibit collagenolytic enzymes in the hybrid layer . However, other studies have reported reductions in bond strength comparable to other adhesives (Table 2: online supplementary data), so it may be too early to make any definitive conclusions of the clinical efficacy of MDPB in the preservation of hybrid layer.

Table 2

Percentage change of the bond strength between immediate and aged values in the experiments in which Clearfil Protect Bond has been evaluated.

Benzalkonium chloride (BAC) is a mixture of alkylbenzyldimethylammonium chlorides of various alkyl chains. It is a cationic surface-acting agent with a quaternary ammonium group used as antimicrobial agent and surfactant . BAC-containing etchants can be used with E&R adhesives without affecting immediate bond strength to enamel or dentin . Tezvergil-Mutluay et al. demonstrated that 0.5% BAC concentrations completely inhibited soluble MMP-2, -8 or -9, and produced significant reduction in demineralized dentin collagen degradation.

Other approaches to eliminate collagen degradation

Even though the concept of chemical bonding with SE adhesives to dentin hydroxyapatite is, in a strict sense, not aiming to inhibit dentin enzymatic function, it has been suggested to preserve the long-term collagen integrity in the hybrid layer . Functional monomers in “mild” SE adhesives, such as 10-methacryloyloxydecyl dihydrogen phosphate (10-MDP), 4-methacryloxyethyl trimellitic acid (4-MET) and 2-(methacryloyloxyethyl)phenyl hydrogenphosphate (phenyl-P) lead to a chemical bond to calcium ions of the hydroxyapatite crystals. “Mild” SE adhesives interact only superficially with dentin, forming a thin (even submicrometre) hybrid layer. The use of “mild” SE adhesives is believed to minimize nanoleakage, leave a substantial amount of hydroxyapatite around the collagen fibrils to mask the collagen cleavage site and keep the enzymes “fossilized” . Thus, the collagen could not be degraded.

Even though the TEM analyses in long-term studies in vitro and in vivo indicate good preservation of the hybrid layer with the “mild” SE adhesives , there is evidence that even with 10-MDP – that is currently considered to form the most stable chemical bond with hydroxyapatite – the bond strengths do decrease with time, both in vitro and in vivo (Table 1: online supplementary data). Apparently the weakest zone in aged samples with SE adhesives is located immediately below the hybrid layer observed with TEM , as the cohesive fractures of dentin (when observed with SEM) increased significantly in an in vivo study . It is possible that the loss of collagen integrity still occurs at the base of the hybrid layer due to voids and nanoleakage that is practically undetectable with current TEM techniques. This is supported by the study by Kim et al. , in which alkaline glycine buffer incubation caused the basal hybrid layer to disappear, creating a 1–2 μm gap between the intact top of the hybrid layer and the mineralized dentin base that was readily detectable with SEM and confocal microscopy in spite of seemingly perfect hybrid layers in their respective TEM images. As the acidic monomers require water for their ionization and etching, the monomers at the bottom of the hybrid layer might be less polymerized, causing the hydrolytic degradation of the hydrophilic monomers with subsequent exposure and hydrolysis of collagen, and finally a loss of bond strength.

If it is accepted that the enzymatic degradation of collagen is not completely eliminated in the deepest parts of the hybrid layer, even with the “mild” SE adhesive functional monomers, the need for enzyme inhibition is still apparent. This assumption is actually supported by some studies. Adding CHX into the self-etching primer of Clearfil SE Bond significantly improved the 12-month bond strength . Also, the use of the self-etching primer containing MDPB (Clearfil Protect Bond) with MMP-inhibitory activity (see details above) has been shown to preserve the dentin bond strength both in vivo and in vitro .

Collagen cross-linking

Even though CHX inhibits both MMPs and cysteine cathepsins , the potential disadvantage is that CHX may leach out of hybrid layers within 18–24 months . To find a permanent solution, inactivation of endogenous proteases of dentin matrix using cross-linking agents is an interesting option. Covalent cross-links produced with external cross-linkers are very stable, and may inactivate the active sites of dentin proteases by reducing the molecular mobility of the active site or by changing negatively charged ionized carboxyl groups into positively charged amides. Even though the results thus far have been promising, problems remain to be solved before clinical applications are available. Glutaraldehyde works well but is toxic. Grape-seed extract is also effective , and an increase in immediate dentin bond strength may be achievable even in reduced treatment times , but it stains dentin brown and the durability of long-term bond strength remains to be examined. Carbodiimide, a cross-linking agent with very low cytotoxicity, has shown promising results in eliminating dentin collagen degradation and preserving dentin bond strength with clinically acceptable procedure time (Table 1: online supplementary data). Low-dose riboflavin has also been tested successfully, but the need for separate curing device (UVA) or separate curing of the cross-linker with blue light irradiation for optimal outcome make further research necessary for a clinically acceptable technique.

Ethanol wet bonding

To ensure proper hybridization of wet collagen matrix, increasing concentrations of hydrophilic and ionic monomers have been added to new adhesives . These polymers are vulnerable to water sorption and/or hydrolysis due to the presence of ester linkages , which significantly weakens the mechanical properties of adhesive with time . Therefore, hydrolytic degradation of adhesives is the other “weak link” in bond strength durability. Brackett et al. demonstrated the water-related loss of nano-fillers within a water-rich zone in the adhesive layer in 12 months in vivo irrespective of degradation of the hybrid layer collagen or its preservation with CHX. Since most studies with CHX and other enzyme inhibitors show some loss of bond strength (Table 1: online supplementary data), part of bond strength loss may be attributed to the water sorption or degradation of the adhesive monomers or polymerized hydrophilic adhesives.

The problem of water hydrolysis of ester-bonds in adhesive polymers and peptide bonds in collagen might be eliminated if water could be excluded from the bonded interface. This has been the aim in ethanol wet bonding , where ethanol is used to chemically dehydrate acid-etched demineralized dentin matrices to reduce collagen hydrophilicity and facilitate the infiltration of more hydrophobic monomers to dentin . Ethanol wet bonding coaxes hydrophobic monomers into a demineralized collagen matrix with limited matrix shrinkage . Infiltration of hydrophobic monomers decreases water sorption/solubility and resin plasticization, but it has been suggested that the elimination of residual water also reduces or eliminates enzyme-catalyzed hydrolytic collagen degradation , which would also contribute to improved durability of the resin bonds . This assumption is supported by the recent study , in which relatively thin hybrid layer created with ethanol wet bonding demonstrated almost complete absence of nanoleakage, as well as excellent durability of bond strength (Table 1: online supplementary data).

Even if ethanol wet bonding may still be considered more as a “bonding philosophy” , it offers both important information of the role of solvent or intrinsic water in the hybrid layer degradation process and interesting potential for the clinically useable bonding techniques to be developed.

Remineralizing the hybrid layer collagen

The loss of protection by the mineral phase in acid-etched dentin renders collagen vulnerable to enzymatic degradation in several ways. Acid-etching activates inactive endogenous proteases bound to collagen, the enzymes themselves are freed from their “fossilized” state, and chemical and/or mechanical damage may result with the exposure of the critical collagen cleavage site, facilitating collagen molecule cleavage. The mechanical damage may be especially important in the hybrid layer. Collagen fibrils in poorly infiltrated hybrid layers, being unsupported by resin, undergo various degrees of irreversible mechanical disruption under occlusal loads. This part of collagen fibril network (modulus of elasticity of 10–100 MPa) has a much lower modulus of elasticity than resin-infiltrated fibrils (modulus of elasticity of 2.5 GPa), allowing it to strain more than resin-enveloped fibrils and causing microdisruptions . For these reasons, remineralization of the hybrid layer collagen fibrils to restore mechanical properties to its mineralized state is an attractive approach to improving the durability of resin–dentin bonds.

Fluoride-containing adhesives have been suggested to eliminate the time-dependent decrease in dentin bond strength . However, the ability of fluoride in adhesive resin to remineralize the collagen matrix has been questioned , at least with acid-etched dentin or with more aggressive SE adhesives . Remineralization of collagen requires seed apatite crystallites that determine the orientation of remineralized crystalline lattices. This limitation severely restricts the remineralization of hybrid layers that contain little or no residual apatite . Interestingly, several studies have been performed where the preservation of long-term bond strength has been attributed to fluoride in the adhesive primer of Clearfil Protect Bond , an adhesive containing MDPB monomer with MMP-inhibiting properties (Table 2: online supplementary data). So, at least part of the preservation of dentin bond strength durability with this adhesive resin may be attributable to the inhibition of collagen degradation by MDPB. It is necessary to prevent collagen fibril hydrolysis if the fibrils are to mineralize.

Remineralization by adhesives containing bioactive particles

Hydrophilic biodegradable polymers and bioactive silicates have been proposed as useable materials for remineralizing scaffolds. With calcium/sodium phosphosilicate (Bioglass 45S5, BAG)-containing adhesive, significantly lower degrees of collagen degradation were seen compared with neat-resin . Ionic dissolution products, when exposed to biological fluids, may be responsible for the biological activity of the tested inorganic materials . However, the hybrid layer remineralization that is based on the degradation of the adhesive components could lead to the reduction of the mechanical properties of adhesive.

Biomimetic remineralization – imitating nature

A new approach to resin–dentin bonding is to perfect the resin–dentin bonds after they have been made, using remineralizing reagents to backfill the water-filled voids with nanometer-sized apatite crystallites. There are two forms of remineralization: (1) the residual mineral crystals can serve as templates for appositional regrowth of apatite crystallites by epitaxial growth . However, remineralization of dentin will not occur in fully demineralized dentin where seed apatite crystals are absent, and noncollagenous, developmental phosphoproteins necessary to attract calcium to collagen may have been extracted. (2) One of the mechanisms for apatite nucleation in the absence of seed crystals is to include polyanions such as polyacrylic acid or polyaspartic acid. These polyanions bind to collagen and serve as templates for stereo specific calcium binding and promote apatite nucleation. In biomimetic remineralization of resin–dentin bonds, one covers a polymerized resin–dentin bond with a “therapeutic” flowable composite that contains a source of amorphous calcium phosphate (ACP). Biomimetic polyanionic analogs of noncollagenous phosphoproteins are included to bind to collagen and serve as templates to induce nucleation and growth of apatite in completely demineralized regions of collagen. Polyanions like polyacrylic acid or polyaspartic acid are included to cluster around nanoprecursors of ACP to prevent it from growing into crystals that are too large to fit into the gap zone of collagen. The key to remineralization of dentin is to create “fluidic” nanoprecursors that can pass through polymerized dental adhesives. This process is thought to use water-filled voids ( i.e. water trees) that extend from the surface of the adhesive layer, through the thickness of the layer into contiguous water filled spaces in the hybrid layer. Water-trees become saturated with nanoprecursors and ACP and mineralize. The process is self-limiting.

Hybrid layers created by E&R adhesives have been shown to be remineralizable using a biomimetic mineralization approach . Apatite crystallites were detected in both extrafibrillar and intrafibrillar compartments after remineralization in phosphoric acid etched resin-bonded dentin. SE adhesives also have water-filled defects that can be remineralized , but the apatite deposition is largely limited to the intrafibrillar spaces, because the extrafibrillar spaces were better filled with adhesives ( Fig. 4 ). Even the water-filled spaces ( e.g. water trees) in many adhesive layers become filled with apatite nanocrystals , although they have no hierarchical order because they contain no collagen.

Nov 25, 2017 | Posted by in Dental Materials | Comments Off on Strategies to prevent hydrolytic degradation of the hybrid layer—A review

VIDEdental - Online dental courses

Get VIDEdental app for watching clinical videos