1

Introduction

Current concepts relating to dental restorative materials exclusively used to maintain structure and function of teeth are presently undergoing drastic change. An increasing number of restorative dental materials actively interfere with regulatory pathways stimulating regenerative processes in periodontal and pulp tissues. Materials like tricalcium silicate cements recommended for vital pulp therapy were shown to induce odontogenic differentiation in dental pulp stem cells (DPSCs) leading to dental hard tissue formation. These materials were considered bioactive because of their regenerative properties . Yet, insight into the cellular mechanisms of these processes is very limited compared to the latest progress in research on the bioactivity or biocompatibility of restorative dental materials . Moreover, the characterization of substances as bioactive or biocompatible encompasses a much broader range of dental biomaterials than much of the dental community has indicated so far. Dental materials, or their compounds, are considered bioactive if they induce changes in cellular homeostasis as a consequence of their interaction with cellular receptors or regulatory proteins, and subsequently activate pathways leading to specific adaptive cell responses. In this respect, biocompatibility is not a property of a material but instead describes the complex interplay between the characteristics of a material and a specific target tissue. This makes it essential for biocompatibility to be considered when referring to a specific application since different materials interfere with various biological systems in a variety of ways .

This review focuses on cellular mechanisms activated by materials present in cavitated carious lesions. In particular, we analyzed the interaction between the specific effects initiated by lipopolysaccharide (LPS) released from cariogenic microorganisms and the influence of resin monomers as compounds of long lasting dental restorations. We first concisely describe the process of dental caries progression to explain the pathways in eukaryotic cells in response to LPS or LTA (lipoteichoic acid) stimulation in a clinical situation. These pathogen-associated molecular patterns (PAMP) activate the redox-sensitive transcription factor NF-κB, which in turn is part of the regulation of immune responses of various cell types in dental pulp tissues like the release of pro- or anti-inflammatory cytokines. There is also evidence that reactions of immune cells in response to NF-κB are controlled by the transcription factor Nrf2 (nuclear factor erythroid 2 [NF-E2]-related factor 2). It seems that the activation of NF-κB and Nrf2 pathways by LPS critically depends on oxidative stress . Likewise, oxidative stress as a result of the increased formation of reactive oxygen species (ROS) is produced in cells exposed to dental resin monomers like TEGDMA (triethyleneglycol dimethacrylate) or HEMA (2-hydroxyethyl methacrylate). Subsequently, the expression of Nrf2 and Nrf2-regulated enzymatic antioxidants is increased as an adaptive cell response . In addition, we consider various cellular pathways activated or influenced by ROS in monomer-exposed cells. Finally, we discuss the interaction of LPS- and monomer-stimulated cell responses based on crosslinks between NF-κB and Nrf2-related biocompatibility pathways. These cell reactions reflect the complex and multidimensional scenarios crucial in the clinical management of carious lesions.

2

Dental caries and the activity of cariogenic organisms

Dental caries is a multifactorial infectious disease of dental hard tissue caused by a complex biofilm of acidogenic and aciduric pathogenic microorganisms, among other species, which stimulate demineralization processes of enamel and dentin. Although the etiology of dental caries was initially related to the activity of Streptococcus mutans as a keystone pathogen, the current paradigm favors the collective interplay of a complex microbiome rather than the activity of a singular species . The multi-species community of cariogenic Gram-positive bacteria including mutans and non-mutans streptococci, actinomyces, lactobacilli, or bifidobacteria species of supposedly predominantely active organisms are accompanied by Gram-negative species like Prevotella , Porphyromonas or Fusobacterium species in a dynamic biofilm . Differential accumulation of acids through the fermentation of sugars and nutrition by a lively ecosystem of cariogenic bacteria produces an acidic environment which is considered a primary factor in changing the diversity of the microbial community . The continuous imbalance between demineralization and remineralization stimulates the degradation of dental hard tissues or structural proteins, and may ultimately create deep cavitation in a more anaerobic environment of a polymicrobial community . Ongoing debates on strategies for removing carious tissue and the questioning of conventional carious tissue removal are motivated by the clinical need to protect the dentin-pulp complex from further degradation, and prevent subsequent penetration of microorganisms or microbial products into the pulp cavity resulting in severe inflammation of pulp tissues followed by necrosis . However, as outlined below the influence of cariogenic microorganisms and their products on the various specialized cell types building the pulp tissue complex is much broader than simply inducing inflammatory responses.

Tooth function with an emphasis on the preservation of vital pulp tissues is presently restored by the use of chemically diverse materials. It has been suggested that restorative sealing of a carious lesion is the first priority in order to preserve healthy and remineralizable hard tissues since bacteria are deprived of carbohydrates and their degrading activity thus arrested . Although acidification might be terminated, this strategy is still not controllable and is inferior and difficult to follow, especially in deep carious lesions since membrane-bound molecules like lipopolysaccharide (LPS) found in Gram-negative organisms or lipoteichoic acid (LTA) released from Gram-positive bacteria are very probably active independent of the vitality of microorganisms. It has been repeatedly emphasized that pulpal inflammatory responses to bacteria, or their compounds LPS or LTA invading through dentinal tubules, strongly depend on the remaining dentin thickness and tubular permeability. A relative wide distance of about two millimeters of sound dentin between the caries front and pulpal tissues has been estimated to be appropriate for maintaining pulp health . This crude estimation, however, is primarily based on the limited sensitivity of current histological techniques. Modern molecular methods may be able to detect disturbances of the tightly controlled homeostasis in pulp cells caused by bioactive compounds such as LPS or LTA at low doses.

2

Dental caries and the activity of cariogenic organisms

Dental caries is a multifactorial infectious disease of dental hard tissue caused by a complex biofilm of acidogenic and aciduric pathogenic microorganisms, among other species, which stimulate demineralization processes of enamel and dentin. Although the etiology of dental caries was initially related to the activity of Streptococcus mutans as a keystone pathogen, the current paradigm favors the collective interplay of a complex microbiome rather than the activity of a singular species . The multi-species community of cariogenic Gram-positive bacteria including mutans and non-mutans streptococci, actinomyces, lactobacilli, or bifidobacteria species of supposedly predominantely active organisms are accompanied by Gram-negative species like Prevotella , Porphyromonas or Fusobacterium species in a dynamic biofilm . Differential accumulation of acids through the fermentation of sugars and nutrition by a lively ecosystem of cariogenic bacteria produces an acidic environment which is considered a primary factor in changing the diversity of the microbial community . The continuous imbalance between demineralization and remineralization stimulates the degradation of dental hard tissues or structural proteins, and may ultimately create deep cavitation in a more anaerobic environment of a polymicrobial community . Ongoing debates on strategies for removing carious tissue and the questioning of conventional carious tissue removal are motivated by the clinical need to protect the dentin-pulp complex from further degradation, and prevent subsequent penetration of microorganisms or microbial products into the pulp cavity resulting in severe inflammation of pulp tissues followed by necrosis . However, as outlined below the influence of cariogenic microorganisms and their products on the various specialized cell types building the pulp tissue complex is much broader than simply inducing inflammatory responses.

Tooth function with an emphasis on the preservation of vital pulp tissues is presently restored by the use of chemically diverse materials. It has been suggested that restorative sealing of a carious lesion is the first priority in order to preserve healthy and remineralizable hard tissues since bacteria are deprived of carbohydrates and their degrading activity thus arrested . Although acidification might be terminated, this strategy is still not controllable and is inferior and difficult to follow, especially in deep carious lesions since membrane-bound molecules like lipopolysaccharide (LPS) found in Gram-negative organisms or lipoteichoic acid (LTA) released from Gram-positive bacteria are very probably active independent of the vitality of microorganisms. It has been repeatedly emphasized that pulpal inflammatory responses to bacteria, or their compounds LPS or LTA invading through dentinal tubules, strongly depend on the remaining dentin thickness and tubular permeability. A relative wide distance of about two millimeters of sound dentin between the caries front and pulpal tissues has been estimated to be appropriate for maintaining pulp health . This crude estimation, however, is primarily based on the limited sensitivity of current histological techniques. Modern molecular methods may be able to detect disturbances of the tightly controlled homeostasis in pulp cells caused by bioactive compounds such as LPS or LTA at low doses.

3

Pulp tissue responses to cariogenic microorganisms

Dental pulp tissue is a complex network of specialized cells which dynamically interact to build dental hard tissues, maintain pulp viability and respond to invading pathogenic microorganisms or their products . Cariogenic bacteria and pathogen-associated molecular patterns (PAMP) like peptidoglycan, LPS or LTA will first interact with a cell layer of dentin-forming odontoblasts at the periphery of the pulp. Beyond this first line of defense fibroblasts, lymphocytes, macrophages, dendritic, endothelial, and neural cells as well as pulpal stem cells build tissues involved in initial and adaptive pulp responses . It has become evident that PAMPs like LPS induce a wide variety of biological effects, which are seemingly contradictory, in different cell types of the dental pulp. LPS released from Gram-negative bacteria was associated with pulpitis and necrosis in human dental pulps . At the same time LPS may stimulate regeneration of pulp tissues through the promotion of cell adhesion, migration and by the induction of odontoblastic differentiation in pulp stem cells . It has been suggested that cytokines and other proinflammatory mediators including ROS may directly stimulate dental tissue repair responses . As a transcription factor the NF-κB protein complex is probably associated with these processes, although its precise function still needs to be elucidated . Likewise, signaling through NF-κB is most likely involved in the immune responses of dental pulp stem cells or odontoblasts . However, interaction of dental resin monomers with the influence of LPS on cell viability and the induction of inflammatory responses as described below has only been analyzed in macrophages so far. Acute and chronic activation of the innate immune response in odontoblasts is followed by reactions of tissue resident immune cells in later stages of bacterial infection . It has been shown that LPS and LTA stimulate the release of pro- or anti-inflammatory cytokines including TNFα, IL-6 or IL-10 from various cell types .

Immunocompetent cells of the dental pulp such as macrophages or odontoblasts respond to PAMPs like LPS or LTA through the function of pattern recognition receptors (PRR), including NLR proteins ( nucleotide-binding domain , leucine-rich repeat containing ), RLR proteins ( retinoic acid-inducible gene-I like receptors ), CLR proteins ( C-type lectin receptors ) and the conserved family of Toll-like receptors (TLR) located on the outer surface of immunocompetent cells. PPR vary in the activation of intracellular signaling pathways and in tissue-specific expression levels . According to the current view, the synthesis of pro- and anti-inflammatory cytokines and interferones (IFN) is a primary cell response regulated at the transcriptional level, while phagocytosis, autophagy, or processing of cytokines are driven through PRRs at the non-transcriptional level . Yet, the immediate onset of the LPS-stimulated release of TNFα from macrophages, and the concurrent inhibition of this effect in the presence of a dental resin monomer, suggest a transcription-independent mechanism as well .

LTA is a major compound of the cell wall of Gram-positive cariogenic bacteria built from hydrophobic glycolipids linked with anionic polyglycerolphosphate . The amphiphilic polymer is synthesized in vast amounts by cariogenic microorganisms and released in an acidic environment. Although not clear until recently, LTA has been identified as a ligand of TLR2 . It is hypothesized that LTA diffuses across dentinal tubules and binds to TLR2, which then forms a dimer complex with TLR6 (TLR2/TLR6). Expression of TLR2 has been identified in odontoblasts underneath the interphase with caries lesions but was absent in odontoblasts of healthy pulps. There is growing evidence that LTA stimulates the expression of TLR2 in odontoblasts, dendritic cells and macrophages in inflamed dental pulp tissues triggering initial inflammatory responses .

LPS is a glycolipid and the pathogenic compound of many Gram-negative cariogenic bacteria. As a strong inflammatory PAMP, LPS specifically binds first to LPS binding protein (LBP) on the cell surface, is then transferred to CD14 ( cluster of differentiation 14 ), which in turn interacts with TLR4 complexed with MD-2 ( myeloid differentiation protein-2 ) activating multiple signaling pathways of the innate immune responses . Then, signaling through MyD88 leads to the activation of the redox-sensitive transcription factor NF-κB . Finally, activation of MyD88/NF-κB signaling pathways by LTA through TLR2/TLR6 or LPS through TLR4, with the subsequent release of pro- and anti-inflammatory cytokines, has been detected in odontoblasts and macrophages .

Yet, the synthesis of cytokines and chemokines as an early immune response is differentially stimulated by LPS and LTA. In odontoblasts LTA activates the expression of TLR2 and, among others, stimulates the release of cytokines and chemokines like IL-6, CXCL1, CXCL2, CXCL8 or CXCL10, while LPS increases the expression of IL-1β, TNF-α, CCL20, or IL-8 for example . In macrophages, LPS again activates the release of pro-inflammatory cytokines such as IL-1, IL-6, or TNF-α, or anti-inflammatory compounds like IL-10 or TGF-β. LTA is less effective than LPS but still induces TNF-α, IL-6 or IL-8 expression .

Signaling through the activation of mitogen-activated protein kinases (MAPK) like p38/JNK is an alternative pathway to NF-κB . Noteworthy is that both NF-κB and p38/JNK signaling is controlled by the formation of ROS. While activation of p38 though ROS is firmly established, it has been reported that signaling through NF-κB may be either activated or repressed by ROS depending on subcellular compartments or the origin of cells. For instance, oxidation of a specific cysteine residue in p50 by ROS or S -nitrosylation by NO (nitric oxide) produced by the NF-κB target iNOS prevents DNA binding of NF-κB . It seems that the generation of NO by iNOS is crucial for the function of NF-κB in LPS- or LTA-stimulated cells . Odontoblasts activated through TLR2 as well as LPS- or LTA-stimulated macrophages produce a vast amount of NO. Moreover, it has been suggested that the generation of NO by odontoblasts could inhibit replication of a variety of dentin-invading microorganisms as an essential defense mechanism . As a non-toxic molecule, NO is an intercellular messenger in vertebrate cells but its reaction with superoxide anions (O ₂ ⁻ ) produced by LPS-stimulated NADPH oxidase 2 (Nox2) activity results in the formation of the reactive oxidant peroxynitrite (ONOO ⁻ ), which immensely adds to oxidative stress . In this respect, the expression of inflammatory cytokines or chemokines through the redox-sensitive transcription factor NFκ-B is associated with oxidative stress as well . It has been previously shown that LPS-generated oxidative stress activates the NF-κB pathway and enhances the expression of the stress-responsive protein HO-1 (hemeoxygenase 1) . Increased expression of HO-1 has also been observed in response to LTA-induced ROS generation, obviously mediated through TLR2/MyD88 signaling and probably the activation of Nrf2. HO-1 functions in the cellular defense particularly against various forms of ROS- or RNS-generating systems and other stressors including cytokines or endotoxin . It has also been suggested, however, that HO-1 inhibits immune reactions and suppresses cytokine- or LPS-induced activation of NF-κB . Most relevant, HO-1 expression is not only induced through NF-κB but is a direct target of the redox-sensitive transcription factor Nrf2 . In this respect, it seems that HO-1 is a link and crucial regulator of the independent pathways through NF-κB and Nrf2, and might be a key protein in the analyses and understanding of the interference of dental monomers with the clinically relevant LPS-stimulated expression of pro- and anti-inflammatory cytokines as discussed below.

4

Activation of transcription factors Nrf2 and NFκB—function of oxidative stress

Moderate amounts of ROS are signal molecules in the redox-sensitive NFκB pathways in LPS- or LTA-stimulated cells . Balanced redox homeostasis, in turn, is controlled by multidimensional adaptive mechanisms including the redox-sensitive transcription factor Nrf2 as a crucial regulator of cell responses to oxidative stress . As an adaptive cell response to oxidative stress, Nrf2 is released from a Nrf2/Keap-1 protein complex, and translocates to the nucleus to induce the expression of a network of enzymatic antioxidants at the transcriptional level through the binding to AREs (antioxidant responsive elements) in the promotor region of the corresponding genes . Notably, there is increasing evidence that Nrf2 also controls cell reactions of the innate immune system. It has been shown that Nrf2 downregulates inflammatory responses through the inhibition of the LPS-stimulated formation of ROS required for the activation of the NF-κB pathway . Although crosstalk mechanisms are extremely complex, oxidative stress is central for the activation of pathways through both redox-sensitive transcription factors NF-κB and Nrf2 . Expression of the stress protein HO-1 as a direct target of Nrf2 could be the link in a negative feedback mechanisms between Nrf2 and NF-κB . Activation of NF-κB and inflammatory cell responses are most likely inhibited through the Nrf2-induced expression of HO-1 and reduced levels of ROS . It appears as if the reciprocal regulation of the redox-sensitive activities of NF-κB and Nrf2 is tightly regulated by the temporal and local redox status essentially controlled by Nrf2. Both pathways are influenced by a resin monomer-induced disturbance of the cellular redox homeostasis.

5

Bioavailability of dental resin monomers and adaptive cell responses