Current Approaches to Produce Durable Biomaterials: Trends in Polymeric Materials for Restorative Dentistry Applications

Abstract

Dental caries continues to be a public health issue, especially more evident in underserved populations throughout the U.S. Unfortunately, especially with an aging population, hundreds of thousands of resin composite restorations are replaced each year due to recurring decay and fracture. According to several cohort studies, the average life span of this type of restoration is 10 years or less, depending on the caries risk level of the patient and the complexity of the restorative procedure. Any new material development must depart from the simple restoration of form paradigm, in which the filling is simply inert/biocompatible. This review will discuss novel antibiofilm structures, based on a targeted approach specifically against dysbiotic bacteria. Biofilm coalescence can be prevented by using glycosyl transferase - GTF inhibitors, in a non-bactericidal approach. On the tooth substrate side, MMP-inhibiting molecules can improve the stability of the collagen in the hybrid layer. This review will also discuss the importance of testing the materials in a physiologically relevant environment, mimicking the conditions in the mouth in terms of mechanical loading, bacterial challenge, and the presence of saliva. Ultimately, the goal of materials development is to achieve durable restorations, capable of adapting to the oral environment and resisting challenges that go beyond mechanical demands. That way, we can prevent the unnecessary loss of additional tooth structure that comes with every re-treatment.

Graphical abstract

1. Introduction

Oral microbial infections leading to disease are prevalent in the general population. The most commonly observed oral pathologies are caries and periodontal disease, which can have serious systemic complications, including endocarditis [1, 2]. In the United States, for example, despite prevention campaigns, roughly 95% of the population aged 19–85 is affected by caries, whether newly formed or recurrent around restorations. Oral bacteria establish a resilient biofilm, firmly attached to the surface and protected from clearance by saliva and penetration by antimicrobial agents. Biofilm formation and survival are a result of early colonization of the protein-coated oral surfaces and coalescence of multi-species bacteria via the production of extracellular polymeric matrix (EPS), composed mostly of soluble and insoluble polysaccharides [3, 4].

As a result of the disease, teeth are left with some degree of destruction. The restorative material development paradigm so far has focused on merely restoring form and function and sealing the tooth. Biological considerations have been limited to producing materials that would not elicit a response from the host. Meanwhile, as mentioned, oral biofilm-derived disease, particularly dental caries and periodontal disease continues to be a public health issue, especially more evident in underserved populations. Even after teeth are restored, recurrent carious lesions, especially in interproximal areas where biofilm is more likely to accumulate [5], continue to be a challenge, and this has not changed in 40 years. The average lifespan of composite restorations is 10 years or less [6], with the main cause for replacement being secondary decay, likely influenced by biofilm formation and material/bonded interface degradation through hydrolysis and enzymatic attack. Thus, current materials are not effective at eliminating recurring disease, suggesting that any effective and long-lasting solution to this problem necessarily requires a multi-factorial approach, including the development of materials with not only biocompatibility and resistance to degradation but also biofilm-curtailing and substrate-preserving capabilities, as well as with regenerative potential.

Effective antibacterial approaches must preferably combine antifouling, bactericide, and/or EPS-disrupting strategies (kill, disrupt, and detach). However, currently available antimicrobial materials or topical medicaments are not ideal because they rely on broad-spectrum antibiotics. These are delivered via toothpaste/mouthwashes (chlorhexidine, triclosan) with only limited effect due to the transient nature of their contact with the bacteria. More recently, examples of broad-spectrum antimicrobials have been incorporated into restorative materials, aiming to disrupt biofilm formation for both dysbiotic and commensal bacteria [7]. These can vary from silver nanoparticles [8] or other compounds that can be made co-polymerizable within the networks of dental composites and adhesives [9–11]. Examples include polymerizable quaternary ammonium methacrylates (QAMs) [12], and more recently, polymerizable imidazolium methacrylate compounds [13]. Others have investigated the effectiveness of designer peptides capable of inhibiting bacterial attachment [14], including dentin impermeabilization function, which has also been proposed for use in scaffolds [15] and bone-anchored devices [16]. The main drawback is that those compounds indistinctively affect pathogenic and commensal organisms, mainly via contact-kill mechanisms. One additional issue is that oral bacteria are very resilient, and even though the surface of the material might delay colonization, the biofilm eventually gets re-established, with dysbiotic bacteria such as S. mutans being as likely to colonize oral surfaces as commensal bacteria such as S. gordonii [17]. Recent studies have proposed a targeted approach, capable of selectively modulating oral ecology on the material surface to favor commensal species and deplete dysbiotic ones. EPS-inhibiting strategies have been developed via the inactivation of glycosyltransferase (GTF), the cell membrane protein responsible for the polymerization of glucose into soluble and insoluble glucans [18]. These enzymes are conserved in both S. sanguinis (early colonizer, commensal) as well as in S. mutans (pathogenic), but are unique to each species, which makes them a viable target. Many small molecule compounds with high efficiency specifically against S. mutans GTFs have been designed, with nitro compounds showing the best specificity and efficiency at inhibiting S. mutans biofilm formation [19–21].

In addition to bacterial disruption, dentin substrate reinforcement by treatment of the surface with collagen cross-linkers and MMP inhibitors has been proposed [22]. MMPs are over-expressed in carious lesions, and can stem from several different sources: 1 – fossilized enzymes encased in the mineralized tissue are exposed upon dentin demineralization by caries and/or acid etching procedures [23]; 2 – inflammatory response in the pulp increases the production of signaling factors as well as enzymes such as MMPs, whose concentration in dentin consequently increases [24]; 3 – biofilm derived proteases [25]. All of these sources contribute to the degradation of exposed collagen and compromise the adhesive interface. One of the strategies to prevent this issue is to use collagen crosslinkers, either based on natural compounds such as proanthocyanidins [26] and catechins [27], or chlorhexidine [28]. So far, these compounds have been largely proposed as pre-treatments, adding one step to the adhesive procedure [29, 30]. Their effectiveness has also been linked to somewhat impractical application times [31], though more recent studies have shown promising results with the direct incorporation in the material formulation [32], or drug-delivery systems such as nanostructures and hydrogels [33, 34]. The use of delivery systems is advantageous and growing in interest to allow for the controlled release of drugs such as chlorhexidine (CHX), which so far has only been shown to increase bond strength in the short term [35]. This occurs because CHX is water-soluble and only binds to dentin electrostatically (CHX positive charges interact with the negative charges of phosphate in hydroxyapatite), which results in gradual dissipation over time due to the competitive desorption mechanism with other cations [36]. The other commonly used strategy is the direct inhibition of degrading enzymes, including MMPs and cysteine-cathepsins [23, 37]. A growing body of literature has pointed to chlorhexidine - CHX [38], or other synthetic small molecules such as NNGH (N-hydroxy-2-[[(4-methoxyphenyl)sulfonyl](2-methylpropyl)amino]-acetamide) [39]. CHX, for example, besides being an efficient broad-spectrum antimicrobial agent, has already been demonstrated to be also capable of inhibiting endogenous gelatinases and collagenases present in human dentin, specially MMPs −2, −8, and −9 in vitro and in vivo [40], even at considerably low concentrations (0.001% to 0.2%, depending on the MMP) [29]. As for clinical applications, CHX was not shown to play a significant role as a pretreatment in the long-term preservation of noncarious cervical lesions [41].

Whatever the approach, the weakest link in resin composite restorations seems to be the bonded interface. The mechanical integrity of materials has significantly improved over the years, with far more advances in filler technology than in the organic phase. The concomitant biofilm colonization and substrate degeneration are inextricably connected, and over time, contribute to the instability of the interface. This review covers the most recent advances in the field, going beyond the strategies described so far.

2. Preventing bacterial-derived degradation

a. Materials resisting hydrolytic degradation – departing from methacrylates?

Methacrylate-based materials are intrinsically prone to hydrolysis due to the presence of labile ester bonds, and in fact their degradation in simple aqueous environment has been widely demonstrated by water sorption and solubility studies, as well as by the comparison of properties in dry and wet storage [42]. In the oral environment, other than demineralization and destruction of the dental substrate itself, bacterial colonization also affects restorative material integrity, accelerating the degradation of labile ester bonds at low pH, allied with the presence of esterases derived from the saliva and biofilm [43, 44]. Ester-containing methacrylate-based materials have overwhelmingly been the most used in dentistry since the inception of tooth-colored restorations, for several different reasons including the robust free-radical polymerization mechanism that can be triggered on command. From the outset, it is noteworthy that, though not without its faults, photopolymerization of dental materials is a very familiar procedure for practicing dentists, so any proposed modifications to materials and usage protocols need to balance the potential benefit of property improvement and the practicality of its use by the profession, with the ultimate goal of translating novel technologies to clinical reality. That being said, the past 10–15 years have seen both commercial and experimental developments using ester-free chemistry for restorative materials [9]. Amongst those, materials based on epoxy (the only one with a past commercial example), thiol-ene [45], alkyne-azide [30, 46], vinyl-sulfones [47, 48], vinyl-ethers [49], and acrylamides [50, 51] have been developed. These materials have varying advantages and disadvantages in terms of mechanical reinforcement, needed for unique photopolymerization procedures, and cytotoxicity, but the main goal of reduced degradation has been largely achieved by most of them [9, 52]. Divinyl-ethers have been able to stabilize bonds, and present much lower water sorption and solubility than HEMA-based controls [53]. Multi-acrylamides in particular are extremely stable even when tested at low pH (<2) and led to a four-fold increase in bond stability in vitro after aging for 6 months in water [51], and after simultaneous bacterial and mechanical challenge in a bioreactor [52]. This was true even though acrylamides are more hydrophilic than HEMA-based materials (log P between −0.21 and 0.30 and 0.50, respectively) [9], which may be explained, other than the monomer stability, by potential collagen reinforcement via supramolecular interactions, as demonstrated by the increase in dentin mechanical properties and reduction of collagen degradation by-products [54]. Neither of these two HEMA-free examples, however, are completely ester-free, since the fully formulated adhesives were still co-polymerized with dimethacrylates as the base monomer. This does demonstrate that the use of crosslinked networks, with hydrolytically stable components, is very promising and ripe with opportunities for commercial and clinical applications.

b. Materials resisting bacterial attachment – antifouling and antibacterial

Most dental materials claiming antibacterial characteristics to date have relied on cationic compounds, such as quaternary ammonium methacrylates (QAMs), whose mechanism of action is through contact kill [55, 56]. Even though surface charge can decrease bacterial attachment through a bactericidal effect, it does not prevent the formation of a biofilm, since adaptive mechanisms allow bacteria to attach to the dead early colonizers instead of directly to the surface, thus shielding them [57]. It has been demonstrated that stressful growth conditions can also trigger in situ biofilm remodeling by S. mutans to create a highly resilient biofilm [58]. This further emphasizes that surface charge alone may not be sufficient to prevent biofilm formation for some bacterial species. If one proceeds with the realization that secondary decay is not simply due to the presence of bacteria, but to the dysbiotic ecology at or near the margin of the restoration, then it becomes obvious that a targeted antibacterial strategy is required.

i. Effect of the acquired salivary pellicle

In the realm of “bioactive” materials, the ones that can resist bacterial attachment and/or demonstrate antibacterial effect are of particular interest in restorative dentistry. As already mentioned, the pathway to improving the longevity of adhesive restorations necessarily involves attention to the biological interactions of the material with the oral cavity, one of which is the colonization by bacteria. Biofilm formation is initiated by the attachment of early colonizers to the acquired salivary pellicle (ASP), a proteinaceous coating that immediately forms on the surfaces exposed to the oral environment. The ASP has protective functions, with many of its proteins/peptides having some antimicrobial role (myeloperoxidases, statins, statherins) [59, 60] [61], and serving as a barrier involved in the remineralization-demineralization homeostasis of the tooth [62, 63]. It may also pose complications when surface functionalization is designed, with the concern being that the salivary coating may decrease the efficacy of antimicrobial moieties engineered on the surface of materials, especially for those that rely on strict contact-kill mechanisms [64, 65]. For example, materials based on quaternary ammonium functionalities are known for their charge concentration-dependent antimicrobial mechanism. In solution, the size of the side chain on the quaternized nitrogen determines the degree of micellization, which directly correlates with the antimicrobial potential, especially in gram-positive species [66]. On the surface of materials, quaternary ammonium methacrylates (QAM) were shown to reduce the colonization of S. mutans, monotonically correlated with both the surface charge concentration and the length of the side chain [67, 68]. The standing theory is that the positive charge on the surface of the material attracts the bacteria’s negatively charged surface, and upon contact, the hydrophobic side chain interacts with the bacterial envelope to cause cell lysis – this is known as the “lancet theory” [68]. However, the side chain has a theoretical length that is much smaller than the thickness of the bacterial envelope (Figure 1), in addition to being hydrophobic, flexible, and prone to folding, therefore not likely to interact with the hydrophilic exterior of the bacterial envelope – not unless there was an existing pore [69]. The fact is, for its longevity in the scientific literature as a potent bacterial inhibitor, the true antibacterial mechanism of QAMs when immobilized on surfaces has not been fully elucidated. It can be hypothesized that the function of the side chain in materials is similar to what happens when these molecules are dissolved in aqueous environments: the hydrophobicity of the side chain may cause phase-separation within the relatively hydrophilic methacrylate monomers, which then leads to charge concentration and bacterial killing [70]. One study has shown that the size of the side chain directly correlates with the antibacterial efficiency, with 16–18 carbons being the ideal length against S. mutans [71]. Whatever the mechanism, it has been demonstrated by several reports in the literature that the coating with ASP only modestly decreases the antimicrobial activity of QAMs [72, 73].

Figure 1.

Within seconds to minutes of exposure to saliva in the mouth, the acquired enamel pellicle layer (ASP) starts to form on clean enamel surfaces (1). Proline-rich proteins such as statins have a calcium-binding domain which helps them attach to the hydroxyapatite crystal on the enamel surface. The other saliva-derived constituents that are selectively adsorbed to the tooth surface include amylase, lysozyme, statins, peroxidase, statherin, and mucin 2, all of which form an electron-rich base that adheres to the enamel via filamentous structures (2). Within an hour, protein-protein interactions intensify and more proteins from salivary glands secretion, gingival cervical fluids oral mucosa, and microorganisms are deposited on the enamel. Lipids and carbohydrates are also added to the ASP which controls the permeability of the growing structure (3). The mature ASP is heterogeneous with a knotted globular surface that is rich in proteins (4).

A more recent development in the evolution of antimicrobial dental materials is the modulation of the ASP [60, 63, 74–76]. Over 3,000 proteins have been identified in human saliva so far, amongst the most abundant being amylase, proline-rich proteins (PRP), and mucins (MUC5 and MUC7), which have bioadhesion roles, in addition to the antimicrobial peptides already mentioned [77, 78]. The pellicle reaches a thickness of 10–20 nm in just a few minutes after formation and more aggregation of molecules continues to occur 30–45 min after the initial attachment (Figure 1). The dense and electron-rich basal layer, closest to the tooth structure, is dominated by protein–protein interactions. The proteins continue to aggregate towards maturation (around 90–120 min), with the thickness of the salivary pellicle ultimately reaching 30–1000 nm in size, depending on location in the oral cavity, and within the same tooth [62]. While the ASP provides some amount of protection against bacterial attachment, oral bacteria have evolved to adhere to specific receptors in the pellicle layer to initiate biofilm formation. For example, while mucins present in the ASP are often considered to be a non-immune defense mechanism against bacterial biofilms [79], many of them may also facilitate the adhesion of specific bacteria onto the tooth surface [80]. Other studies show that S. mutans can use mucins for nutrition to enhance its growth and survival [81]. Many factors play a crucial role in how the proteins are absorbed on the surface of dental materials, such as the pH and the ionic strength of the medium, surface free energy, charge, polarity, and roughness [82, 83]. Recent publications have investigated how the material properties influence the composition of the ASP and of the subsequent biofilm ecology attached to them. For example, when comparing gold, stainless steel, alumina and zirconia, it was observed that the work of adhesion, final pellicle thickness and the biofilm were similar in all groups [84]. As far as the pellicle composition, differences were noted when comparing metal bracket surfaces and hydroxyapatite using proteomic analysis, which revealed that only a small portion of the proteins attached were common to both groups [85].

Once the pellicle is formed, commensal, early-colonizer bacteria attach to the surface (Figure 2). The initial adhesion is modulated by cell-surface-associated adhesins, then the early biofilm produces extracellular polymeric substance (EPS), which favors adhesion of diversified species, and also functions as a binding and protective medium, forming 3D structures where cell-cell signaling takes place. Biofilm formation and maturation can be disrupted at any of these stages, with different approaches and prognosis [86].

Figure 2.

The protein-rich acquired enamel pellicle (ASP) is a prerequisite for bacterial attachment (1). Early colonizers or pioneer colonizers attach to the ASP and create a non-mature or basic biofilm. The extracellular polymeric substances (EPS) matrix secreted by bacteria provides the biofilm with surface adhesion properties The bacteria present in non-mature biofilms depend on both the material surface and the environment but in general, early colonizers are seen to consist of Streptococcus with S. sanguinis, S. oralis, S. gordonii, S. mitis, S. mutans, and S. sobrinus being active participants in biofilm formation. Actinomyces naeslundii, Lactobacillus acidophilus, and Capnocytophaga ochracea are also seen in the early biofilms. As facultative anaerobes or organisms that grow both in the presence and absence of oxygen, they are versatile organisms that make ATP by aerobic respiration if oxygen is present but are capable of switching to fermentation if oxygen is absent (2). Once the early colonizers are established within the basic biofilms, secondary colonizers join the biofilm by attaching to primary colonizers and increasing the diversity of the biofilm in a process called co-aggregation. During secondary colonization, the biofilm composition shifts from abundant Gram-positive species to predominantly Gram-negative species and includes Gram-negative bacteria such as Fusobacterium nucleatum, Porphyromonas gingavalis, and Veillonella spp.(3). In the presence of sugars and carbohydrates the biofilm shifts to a more cariogenic mode and form mature biofilms with high microbiological variability. The structural and spatial organization of EPS provides cariogenic biofilm resistance against external disturbances and a stronger structure can now withstand mechanical forces. In addition to providing mechanical stability, the matrix causes the microenvironment to create areas of low pH, which are inaccessible to saliva and therefore add to the acidic environment around the tooth structure.

ii. Small molecule releasing antimicrobials vs. tethered antimicrobials

In the case of materials whose antimicrobial mechanism relies on the release of small molecules, maintenance of mechanical properties and therapeutic effects over time are of concern [65, 87]. Silver nanoparticles, chlorhexidine and other small molecules are all examples of compounds that have been added to restorative materials, including one commercial example containing QAM. One recent study has demonstrated that farnesol-containing micelles are effective in reducing cross-kingdom biofilm formation in solution [70], designed for topical delivery rather than material loading, which depends on patient compliance. Materials loaded with farnesol showed a significant decrease in biofilm formation, which was accompanied by a decrease in mechanical properties and short-lived drug release [88, 89]. Polyphenols (PP) have also been extensively studied for their antibacterial abilities and have been found to inhibit specific virulence factors associated with cariogenicity of oral bacteria such as Streptococcus mutans. PP such as epigallocatechin gallate (EGCG), quercetin, gallic acid, ethyl gallate, and emodine (Figure 3) have been shown to reduce the acidogenicity and aciduric level of S. mutans and inhibit its growth and glucosyltransferases activity, especially in planktonic cultures.

Figure 3.

Polyphenols (PP) are commonly used for their antibacterial properties. As the most abundant PP, flavonoids such as epigallocatechin gallate (EGCG) commonly found in tea are known for their antibacterial activity and are known to act against both bacteria as well as fungi and viruses.

More recently, ecology-based strategies for biofilm modification have been introduced, with the rationale of preserving commensal bacteria by reducing the attachment capabilities of virulent species, without an antibacterial effect [19]. Computational models have been used to predict the attachment of small molecules in the catalytic domain of glycosyltransferases (GTF), which mediate the production of glucans/dextrans, essential components of the biofilm matrix. After screening a library of existing drugs, a few compounds have been identified to selectively inhibit the activity of GTF-C produced by pathogenic S. mutans [19]. By specifically inhibiting its activity in a dysbiotic species, these compounds have been shown to selectively impair virulent bacterial colonization and biofilm production. Biofilm formation by other microbial species, even ones bearing a GTF enzyme such as S. sanguinis, are not subject to inhibition, since each species bears structurally distinct GTF enzymes. This targeted approach exploits the inherent ability of the commensal bacteria to antagonize S. mutans growth, aiming at switching the composition of the biofilm to favor non-virulent bacteria, relieving it of the inherent limitations posed by typical stressor-induced antibacterial approaches. In addition, for the commensal initial colonizers, GTF-mediated mechanisms are not primarily used for tooth surface attachment [90]. Therefore, this selectively reduces S. mutans biofilm formation [19, 90].

As effective as these small molecules have been proven to be in culture medium, delivery systems and maintenance of material physical integrity are still elusive [91, 92]. For those reasons, a lot of effort has been placed in developing antimicrobial surfaces, with the active principle being tethered to it. Examples include the already mentioned quaternary ammonium methacrylates [93], quaternary thiazole salts [70], and imidazolium methacrylate [13], amongst others. As an example, one of those studies showed that the inclusion of imidazolium moiety at 2 wt% eliminated S. mutans biofilm formation with minimal cytotoxicity without compromising the mechanical integrity of the restorative material [13]. Another approach is the incorporation of polymerizable antifouling molecules, such as 2-methacryloyloxyethyl phosphorylcholine (MPC), which essentially create a bacteria-repellent surface [94]. Very recent studies demonstrate excellent antibacterial effects with minimal disruption of mechanical properties in denture bases [95], glass ionomer cements [21], and self-adhesive orthodontic resins [96]. Finally, dual-function surfaces, capable of switching from bacteria-repellent to bacteria-killing surfaces as a function of the pH of the environment have been described with efficiency in combating gram-negative bacterial colonization (E. coli), with repeated switching cycles [97]. While these are promising results, these moieties are still broad-spectrum antibiotics and may lead to the selection of resistant strains, at least locally.

One final consideration regarding functionalized surfaces is the fact that, while such approaches are promising, there is a need for solutions that target the continual build-up of biofilms on the surfaces within the oral cavity. Release-based mechanisms are limited in the quantities of drugs or active molecules that can be delivered within the oral cavity, and contact-killing is plagued by the declining efficacy that comes from the buildup of dead bacteria on potent surfaces. Dead bacteria that result from contact-killing approaches or as a result of bactericidal molecules often accumulate on the surfaces and that, allied with increased EPS production derived from the metabolic stress from antimicrobial compounds [98], provides a buffer between the antibacterial surface and bacteria, allowing biofilm to grow eventually. The accumulation of dead bacteria can also potentially hamper the release from stimuli-responsive carriers as they can change the environmental cues that the carriers are exposed to. Therefore, there is a need for novel solutions that periodically disrupt the biofilm formation, the pellicle layer, and a mechanism by which dead bacteria and debris can be removed from the surface of the tooth [99].

iii. Antimicrobial peptides

As a host defense against infections, antimicrobial peptides (AMP), also called Host Defense Peptides, can be classified in antibacterial, antiviral, and/or antifungal. Typically made up of 12–50 amino acids, they tend to be unstructured in free solution, partition into distinct regions, and then folding into their final configurations when portioning into biological membranes. The distinct structures then contain hydrophilic and hydrophobic amino acid residues aligned on opposite sides of the helical structure. The amino acid backbone, size, and the ability of AMP to all contribute to its function as an antibacterial agent. AMPs are generally classified based on their origin, activity, constituents, and structure (Figure 4).

Figure 4.

Naturally occurring Antimicrobial Peptides or Host Defense Peptides are classified based on the source of origin, their pathogenic activity, the amino acid classification they fall under, and their inherent structure.

While not completely understood, the mode of action of AMPs seems to be the inhibition of cell growth and proliferation by inducing cell lysis via cell membrane permeabilization and the prevention of essential nutrients to cells. In the oral cavity, cathelicidin, defensins, and histatin form the major groups of natural AMPs that have been identified [100]. LL-37, for example, has been shown to kill Gram-negative species in the oral cavity via the neutralization of their lipopolysaccharides (LPS) [101]. However, as AMP comes from a variety of sources, with structures and varying levels of potency, they are not always selective to bacteria and can attack host cells. Wei et al. evaluated histidine AMP-temporin-GHa (GHa) that had been modified to reduce cytotoxicity against host cells using database-assisted design and found that the antimicrobial against planktonic S. mutans activity was compromised [102]. Zhong et al. examined the two peptides temporin-GHc (GHc) and temporin-GHd (GHd) produced by the frog Hylarana guentheri and studied their antimicrobial activity against the oral pathogen S. mutans via increasing cell membrane permeability. In addition to preventing S. mutans attachment to surfaces, the AMP also downregulated the GTF genes that affect the role in S. mutans biofilm formation, with very low cytotoxicity [103].

High biocompatibility and success with select AMPs have led to a plethora of designer peptides being developed to combat biofilms in dentistry and other applications [104]. Much like the peptides naturally occurring in the acquired pellicle with antimicrobial functions [105], certain molecules reduce microcosm biofilm attachment and can be applied to virtually any surface, including titanium implants [106]. Using naturally occurring small molecules or host defense peptides from the innate immune system as templates, a range of synthetic peptides have been developed (Figure 5). Several examples of modified natural peptides have demonstrated enhanced bactericidal potency over S. mutans, A. viscosus, P. aeruginosa as well as Candida albicans [107]. Modification strategies include synthesizing truncated AMPs to increase potency [108], synthesizing peptides with D-enantiomer amino acids and retro-inverse peptides to increase resistance to enzymatic degradation [109], or creating hybrid peptides from different AMPs [110]. For example, combining Cecropin A with LL-37 resulted in a hybrid peptide with enhanced antimicrobial properties with minimal hemolytic potential. While the net positive charge on the hybrid peptide would enable attachment to the bacterial membrane, the presence of the α-helix fragments can drive it deep within the bacterial membrane [111]. Finally, peptides generated de novo that are co-immobilized on the surface of biomedical materials have been developed with anti-microbial and osteoinductive properties, triggered by cleavage by MMP-9 [72]. These same amphipathic peptides have been successfully applied to dentin coatings combining impermeabilizing and antimicrobial function [14], as well as have been used in biomimetic antimicrobial scaffolds for bone regeneration [96]. Their antimicrobial mechanism has been shown to involve self-assembly in amyloid-like nanofibrils upon interactions with components of the outer membrane of Gram-negative bacteria and the cell wall of Gram-positive bacteria, respectively [70]. In particular, interactions with peptidoglycans have been reported, which in turn favor interaction with the much thicker cell wall layer in Gram-positive bacteria [112]. These hold great promise in the specific targeting of dysbiotic species in the oral environment.

Figure 5.

(left) Host Defense Peptides or AMPs are modified to deliver designer peptides that can be synthesized in high yields at low costs with enhanced antibacterial efficacy. (right) The specific designer peptide with combined antimicrobial and anti-MMP functions (from Ye et al, 2019, DOI: 10.1039/c8nr07334a).

iv. Surface micropatterning and switching molecules

Introducing nanometer to micrometer-scale patterns on material surfaces has been found to influence bacterial attachment and growth, without the need for chemical modification [113]. Factors such as the geometry, feature density and size, and surface roughness all contribute to bacteria’s ability to adhere to and grow on the surface [114]. These microtopographies have the potential to regulate cell behavior and differentiation. Current research in this field aims to understand how patterns affect cell adhesion, promote differentiation, modify integrin expression, enhance focal contacts, reorganize the cytoskeleton, impact proliferation, increase differentiation, and enhance susceptibility to hormones and growth factors. Microtopographies in the range of 500 nm to 2 μm have been found to influence bacterial adhesion and proliferation in antifouling studies. In implant dentistry, micro and/or nano-patterned implant surfaces have been studied for their ability to alter biofilm formation and drive osteointegration. Various nanopatterns, such as nanogrooves, nanopores, and nanopillars, have shown enhanced cell proliferation and control over cell orientation (Figure 6). Different nanotube diameters have been found to influence bacterial adhesion, with some studies showing superior antiadhesion results with 80 nm nanotubes. However, further research is needed to determine the optimal combination of nanopattern, size, and geometry for effective antifouling outcomes [113, 115].

Figure 6.

A few of the lithography techniques we envision can be used in restorative dentistry to impart patterns onto the surface of the tooth structure (A). Examples of microscale patterns that can inhibit bacterial attachment include the Sharklet^® pattern which is made up of millions of microscopic features arranged in a distinct diamond pattern (B1). Nanoscale patterns such as the one shown in B2 can be used to prevent bacteria from attaching and proliferating (B2).

As the name implies, shape-shifting materials (or stimuli-responsive materials or shape memory materials) are materials that can change their shape or surface characteristics in response to an external cue, including temperature, electric fields, magnetic fields, exposure to specific wavelengths of light, local changes in pH, or even the introduction of molecules such as specific enzymes [116]. Utilizing changes in local pH to release molecules encapsulated within nanocarriers such as micelles, core-shell NPs, nanogels, and polymer-drug complexes have been explored in multiple biomedical applications including dentistry and there are several excellent review articles on this topic [95, 117]. By polymerizing a coating made from acrylated azobenzenes on the surface of dental resins and activating them periodically using visible light (430–480 nm), bacterial biofilms can be removed via the periodic activation of the coating layer [118] (Figure 7). Using magnetic field-directed assembly of nanoparticles, microrobots composed of iron oxide nanoparticles can be used to produce structures resembling toothbrush bristles that can slide between teeth like floss and clean surfaces [115]. Dubbed surface topography-adaptive robotic superstructures or STARS, the nanoparticles could also release antimicrobials to kill bacteria. One additional example takes advantage of molecules capable of switching between an antibacterial and antifouling conformation, based on carboxybetaine chemistries [115]. This type of molecule presents an acid/base-driven equilibrium between two chemical states: a lactone ring structure and a zwitterionic open-ring structure. The anti-fouling properties of the zwitterionic state can be reversibly switched to the bactericidal properties of the lactone ring, depending on the pH of the environment [119]. This can be particularly useful in instances where acidogenic bacteria are present, such as those composing the oral biofilm. The remarkable nature of this material is that it can kill bacteria on contact in an acidic environment, but become self-sterilizing and bacterial resistant at neutral pH [120]. Such approaches can be used to clean surfaces periodically in the oral cavity, and while they are not currently designed to reside within restorations, such changes can be envisioned as part of a self-contained biofilm removal system.

Figure 7.

Covalently-tethered azobenzenes undergo cyclic trans-cis-trans isomerization when exposed to multiple wavelengths of visible light simultaneously (A). The azobenzene can be placed as the ultimate layer of a composite restoration (B1). Along with the ASP, biofilms will form on the surface of the structure (B2). Upon exposure to light from a visible light source, the azobenzene layer can be triggered to switch between its trans and cis states rapidly. This movement can be used to disrupt and disperse biofilms on the surface of materials. (B3) (right) Carboxybetaine chemistry undergoes pH-driven switching from linear (zwitterion, antifouling) to lactone ring (positively charged, antimicrobial) conformations. The switching pH is determined by the degree of substitution on the molecule (Thorpe-Ingold effect) (C).

3. Preventing substrate degradation

Modern adhesive dentistry still relies on modification of the tooth structure to ensure the anchoring of restorative materials and sealing of the tooth/restoration interface [121, 122]. The current standard of care can include pre-treatment of the substrate with cleansing substances (chlorhexidine, CHX for example), total acid etching before the application of adhesives, or the use of self-etching materials [123]. Especially in dentin, and in particular in caries-affected dentin, the use of those techniques has shown inconsistent clinical outcomes [41]. For example, in total-etch adhesives, the demineralization of dentin reduces its mechanical properties by 3–4-fold [124]. A decrease in stiffness comes with increased mobility of the collagen bundles such that the collagen peptides are now readily accessible to matrix metalloproteinases (MMPs) from dentin, as well as from other sources such as salivary glands and gingival fluid, all of which contribute to dentine matrix degradation [125]. This modified substrate is then used to anchor the dental adhesive, with which it forms a hybrid layer, therefore consisting of a resin-infiltrated collagen interface [126]. This layer is prone to degradation, as already mentioned, which then contributes to marginal staining and caries development underneath the restoration. Therefore, methods by which it can be stabilized are actively being explored [124]. Commonly, collagen stabilization approaches involve either direct crosslinking [127, 128] or inhibition of metalloproteinase degradation [129], often using compounds that can accomplish both to some extent [130]. Antimicrobial compounds such as quaternary ammonium, have also shown inhibitory effects on MMP-9 when tested in solution [131].

a. Reinforcing the substrate - Direct collagen crosslinking

The use of external crosslinking agents that can provide additional tether points within bundles of collagen via inter and intra-molecular bonds dates back to the introduction of a glutaraldehyde-based product several decades ago, which was intended to be used as a pre-treatment step before adhesive application [132, 133]. The principle for dentin reinforcement relies on the tissue fixation properties of this compound, via interaction with collagen through its amine groups, thus improving the mechanical properties of this tissue [22]. Indeed, it has been shown that glutaraldehyde compounds decrease the collagen solubilization in the hybrid layer and improve the cohesive strength of the fibrils [134]. Toxicity and staining concerns have prevented widespread adoption of this protocol, despite the in vitro evidence of improved bond strength [135]. In addition, clinical studies have shown no significant influence of glutaraldehyde-based desensitizers in reducing post-operative sensitivity after restoration placement [136, 137]. More recently, natural polyphenolic compounds based on proanthocyanidins such as grape seed extract, genipin, and other plant extracts have been explored [138, 139]. These compounds have also been proposed as a pre-treatment before the application of the adhesive, with excellent results in terms of bond preservation, as shown by the increased dentin elastic modulus [140, 141], decreased collagen degradation as measured by the released hydroxyproline [142], nanoscale dynamic mechanical analysis (nano-DMA), and mass loss of the dentin when in contact with the exogenous proteases. Though the advantage of low toxicity cannot be overlooked, one potential impediment to the clinical application of such compounds was the extended application time for effectiveness, as initially reported [143]. This prompted the addition of proanthocyanidin compounds (PAC) to the acid step of total etch experimental adhesives, with reported stabilization of the bond strength after 12 months storage [144], and prevention of degradation of dentin specimens [145]. Ever since, a lot of progress has been made in the elucidation of mechanism of crosslinking for these natural compounds, with recent reports demonstrating the clear dependence of structural features of the PAC on its reinforcing effect, with specific interflavan linkages affecting low-energy conformations of the molecule, in turn affecting their potential for dentin biomodification [146]. Flavan-3-ol-terminated PAC such as catechin and epicatechin were shown to have the biomodification activity modulated by different combinations of their C- or EC-terminals, but with differential effects at the nano- or macro-scale [147]. While the exact mechanism of reinforcement has not been completely elucidated, it is known that hydrogen bonding and even covalent interactions between the collagen and the different PACs are at play [148]. One last remark on phenolic compounds such as PACs, which are natural antioxidants, is their potential inhibition of free-radical polymerization [149], which was not demonstrated to happen at least in one case where PACs were used as dentin pre-treatments prior to adhesive application [149]. Other known crosslinkers include riboflavin, mostly known as vitamin B2, a molecule that can produce free radicals upon UV light exposure (445 nm, 270, and 366 nm) [150]. Riboflavin has been extensively shown to be a potent therapeutic agent in the treatment of keratoconus, a disease where collagen structure of the cornea is degraded, causing loss of visual acuity [151]. More recently, riboflavin used in dental application has also been shown to enhance the strength of the resin-dentin interface [152, 153]. It is also noteworthy that some compounds have the dual function of providing collagen crosslinking and MMP inhibition [154], which will be detailed in the next section.

b. Reducing collagen degradation – metalloproteinase inhibition

Twenty-four MMPs have been identified to be of interest in the dental substrate [155]. They comprise a family of structurally similar, genetically distinct enzymes which typically consist of an N-terminal propeptide, a catalytic domain, a hinge region, and a C-terminal hemopexin domain. Some MMPs also contain a fibronectin-like domain which forms the main binding sites for collagen, elastin, and gelatin and are a key component that mediates the cleavage of collagen and elastin [156]. Collagen turnover is a physiological process, triggered any time the enzymes are activated, which in the context of the dentin substrate, takes place when partial elimination of the mineral content is achieved with acid etching, provided that the pH remains neutral after the procedure [157]. This interaction slowly hydrolyzes the collagen matrix, reducing the strength and long-term performance of the restoration. The pH-dependent increase in MMP activity after acid etching has been demonstrated with in situ zymography and other colorimetric assays [54]. Competitive or non-specific inhibition of the zinc binding MMPs has been attempted with active site-specific or allosteric inhibitors in many applications, including dental, but also in the prevention of metastatic cancer, periodontal disease, osteoarthritis and multiple sclerosis [158]. In Dentistry, chlorhexidine is a well-known inhibitor, with current, albeit not widespread, clinical utility [159]. Commercial solutions of chlorhexidine digluconate are available for use before adhesive application [160], and have demonstrated improved marginal integrity over at least 36 months in clinical trials [161]. The survival of composite restorations placement in non-carious cervical lesions observed through a randomized clinical trial was not influenced by the pretreatment of dentin substrate with CHX. While CHX is a well-known non-specific inhibitor of MMPs, its effect is temporary as the electrostatic interactions it offers to engage the collagen-binding sites are overcome by the constituents of saliva, such that the inhibition of MMPs is weakened within months. A slow release of CHX loaded into clay was shown to inhibit MMPs more efficiently with bone strength maintained for up to 18 months after instillation. Tetracyclines and ethylenediaminetetraacetic acid (EDTA) have also been shown to inhibit MMPs and slow down collagen degradation [162]. In addition, alcohols with four methylene groups have been shown to efficiently inhibit MMPs in a dose-dependent manner. However, ethanol, which is routinely used as a part of the wet bonding technique in adhesive dentistry had limited efficacy in preventing MMPs from degrading collagen. More recently, synthetic MMP inhibitors such as galardin have been evaluated, and shown MMP activity inhibition at low concentrations for up to 12 months. Similarly, quaternary ammonium compounds (QACs) such as 12-methacryloyloxy-dodecyl-pyridinium bromide and benzalkonium chloride, which were primarily incorporated within adhesive resins for their antibacterial properties, have been shown to inhibit MMPs non-specifically [131]. Drug repurposing from other applications has also gained attention. One such example is the compound NNGH (N-hydroxy-2-[[(4-methoxyphenyl)sulfonyl](2-methylpropyl)amino]-acetamide), which is FDA-approved for use in preventing metastatic cancer [163], and has recently been proposed for use in dental applications [164]. Unlike chlorhexidine and polyphenols, these molecules have the advantage of being effective in nanomolar concentrations, which remains true across multiple MMP types [164].

The challenge with these inhibitors, much like with the antimicrobial compounds described earlier, is to ensure long-term effectiveness [165]. Strategies to overcome this challenge include the use of nanocarriers [33, 165, 166] or the immobilization of the compounds at the surface of materials, the latter being underexplored. Nanoparticles and nanotubes have been tested as CHX carriers with mixed results [33, 167], mainly due to inconsistent loading and release of the drug [168]. In addition, in some instances, the use of nanotubes may hamper mechanical properties [169]. Other more effective carriers have been proposed such as lysolipid-containing thermosensitive liposomes (LTSLs), with sustained bioactivity in cancer applications [170] or others used for the treatment of inflammatory diseases [171]. The direct functionalization of materials with these compounds is a logical next step, but it does not come without its roadblocks. One known example is 12-methacryloyloxy-dodecyl-pyridinium bromide, which can copolymerize with adhesive resins and has demonstrated collagen-protecting activity in addition to its antimicrobial properties [172]. However, the collagen-protecting mechanism is still unknown, and it may not be related to specific enzymatic inhibition. Conceivably, for specific interaction with the binding site, the inhibitor would need to retain mobility – and that is inherently impaired once the compound gets immobilized at the surface of the material, or within the crosslinked network. Therefore, any compound that is engineered to be active at the surface would require a specific linker to retain enough degrees of freedom to allow for interaction with the enzyme without appreciable loss of effectiveness [173].

4. Polymerization stress – do we still care? Evaluating new materials in physiologically-relevant conditions

Over the last two decades, significant research and development effort has been placed in designing esthetic restorative materials with lower polymerization shrinkage and shrinkage stress, either by using monomers with intrinsically lower molar shrinkage coefficients (ring-opening polymerization mechanisms [174, 175], or with higher molecular weight [176], or by using formulations with direct stress reduction mechanisms such as delayed gelation/vitrification [177] or covalent adaptable networks [178]. In spite of successful in vitro studies demonstrating reduced gap formation at the interface of simulated restorations, the clinical performance of most commercially-available materials failed to show improvement compared to traditional formulations [179]. More recently, bulk-fill materials with stress-relieving features have demonstrated clinical performance similar to conventional, incremental materials, with the advantage of fewer application steps [180]. Even though the relationship between interfacial stress and gap formation is well-documented, clinical evidence demonstrating that materials with reduced gap formation result in reduced restoration interface failure is lacking. Clinical failure is then ascribed to other clinically-relevant factors, such as patient diet, oral microbiome and hygiene habits [181]. This explains why in clinically-relevant conditions, certain patients with stress-derived interfacial gaps do not develop secondary caries [182]. That being said, even if the mere presence of gaps does not ascertain secondary caries formation, a gap still needs to exist for bacterial colonization to take place [183], regardless of whether it is derived from interfacial stress development or adhesive layer degradation. Therefore, one important point when developing materials, whether they are intended to have bioactivity function or not, is using testing conditions that are relevant to the physiology of the oral cavity.

Innovative devices have been developed in the past 15–20 years that allow for some approximation of the cyclic nature of mechanical loading, bacterial challenges, pH changes, etc, present in the mouth. While no system is flawless, these at least approximate testing conditions to the oral cavity, which is more clinically-relevant than the methods that call for specimen storage in water at 37°C. One example of such device combining mechanical loading and bacterial challenges has been used to characterize the adhesive interface of simulated preparations restored with low-stress composites in combination with degradation-resistant adhesives, using confocal imaging as the end-point evaluation [52]. With this apparatus, it was possible to conclude that the material combination resulted in complete preservation of the bonded interface, in contrast with severe gaps and demineralization observed with a conventional composite/adhesive combination after testing under mechanical and bacterial challenges. Under less harsh conditions (water storage at body temperature), those differences were not evident [52]. While it was not the intention of this review to provide a comprehensive list of the different types of oral simulators available, it is important to note that several do exist. One example combines four-point bending fatigue of beam specimens under biofilm challenge and use bulk mechanical properties as the quantitative measure [184]. Other bioreactors that allow for active culture of bacterial under simulated saliva flow (drip flow reactor) [185, 186] have also been used.

5. Conclusions and future outlook

While esthetic, direct restorative materials have improved precipitously in the last several decades, the longevity of restorations built with them still falls short. To prevent extended damage to the tooth structure, which is onerous for the patient in both health and financial terms, there is still a need for judicious design of materials that are better able to interact with the oral environment to increase their survival. Rather than being simply inert and mechanically sound, it has become apparent that materials should actively integrate with the oral microbiome and host structures. Not explored in this review are other aspects related to the role of the immune system on biofilm development, as well as the strides made in the area of strategies to disrupt quorum sensing. The future of dental restorative materials has been driven by advancements in materials science, technology, and a better understanding of dental biology. Key trends include not only the topics discussed in this review, but also self-healing and responsive materials, regenerative materials, 3D printing of crowns, bridges and other types of restorations, and more. Digital workflows in dental practices are fast becoming a reality, and this will certainly also contribute to accelerate material development. As these innovations continue to evolve, they promise to make dental restorations more durable, functional, and aesthetically pleasing, while also improving patient outcomes and comfort.

Clinical Significance:

While proper restorative technique and patient education in terms of diet and oral hygiene are crucial factors in increasing the longevity of esthetic direct restorations, materials better able to resist and interact with the conditions of the oral environment are still needed. Reproducing the success of dental amalgams with esthetic materials continues to be the Holy Grail of materials development.

Highlights.

1. The main reasons why adhesive restoration fail include secondary decay and fracture. Factors leading to secondary decay are discussed here.

2. Different classes of antimicrobial compounds have been proposed for use in dental materials. The challenges with clinical translation of research findings are explored.

3. Dentin substrate reinforcement is another avenue to provide stability to the bonded interface. Natural compounds have been vastly explored for this application, and novel strategies are emerging.