Nitric Oxide in Dental Pulp Tissue: From Molecular Understanding to Clinical Application in Regenerative Endodontic Procedures

NO production mechanism

In mammals, NO is predominantly produced by genuine immune system cells and vascular endothelial cells at varied concentrations (nM–μM) from intracellular five-electron oxidation of l-arginine with or without stimulation.^11,22,23 This reaction (viz. oxidation of l-arginine to citrulline) is catalyzed by a family of membrane-bound isoenzymes collectively termed NO synthases (NOSs). Based on the tissues in which they are most prominently expressed, the following three distinct genes encoded NOS isozymes have been identified to date: neuronal NOS (nNOS or NOS I), inducible/immune NOS (iNOS or NOS II), and endothelial NOS (eNOS or NOS III).^11,24

The nNOS form is abundantly expressed in the brain and peripheral nerves, while eNOS is present mainly in endothelial cells; however, nNOS and eNOS expression is not restricted to neurons and endothelial cells, respectively. iNOS has been identified in a wide range of cells, such as bone cells, glial cells, and peripheral blood cells.^12,22,25 eNOS and nNOS, which are constitutively produced in low levels (and thus are both called cNOS), are typically activated by intracellular calcium ion (Ca²⁺) influx and subsequent binding to calmodulin to synthesize low levels (on the picomolar to nanomolar level; pM–nM) of short-acting NO (seconds to minutes), which in turn promotes calmodulin binding to cNOS. In contrast, iNOS has a higher affinity for calmodulin at very low intracellular Ca²⁺ levels (below 40 nM), which thus acts in a Ca²⁺-independent manner, and can be induced at high levels for prolonged periods and can produce copious amounts of NO up to micromolar (mM) levels for sustained periods of time (hours to days) in response to different mediators, including inflammatory incentives such as bacteria lipopolysaccharide (LPS) and cytokines (e.g., IL-1β, TNF-α, or IFN-γ). In turn, NO can further affect the synthesis of the following chemokines: IL-1, IL-6, IL-12, TNF-α, and PG.^22,26 Similarly, NOS enzymes are responsible for endogenous NO synthesis; in turn, NO itself may negatively regulate NOS activity through feedback regulation.^27,28

All isoforms of NOS synthesize endogenous NO by consuming l-arginine as the substrate, with molecular oxygen (O₂), nicotine adenine dinucleotide phosphate (NADPH; the best-known enzyme in phagocytic cells that produces superoxide [O₂⁻] for bacterial destruction within the respiratory burst process), and other cofactors [such as flavin adenine dinucleotide (FAD), flavin mononucleotide (FMN), and (6R-)5,6,7,8-tetrahydroL-biopterin (BH₄), etc.] as cosubstrates, in a process that mainly involves oxidation of NADPH and reduction of molecular O₂.^23,29 Thus, NADPH is typically served as a histochemical marker for NOS activity. In addition to the above l-arginine NOS pathway, NO can also be produced through non-NOS pathways, that is, through a reductive nitrate–nitrite–NO pathway. ³⁰

Biological functions of NO and relevant mechanisms

NO can mediate multiple cellular activities, such as apoptosis, differentiation, neurotransmission, immune responses, vascular tone regulation, platelet aggregation, and angiogenesis (Fig. 1). ²³ Accordingly, the underlying chemical mechanisms of NO are complex processes that include numerous possible reactions exerting effects in two distinct categories: direct and indirect effects.

OPEN IN VIEWER

FIG. 1.

Examples of the biological effects of NO at different concentrations and the related signals. AKT, protein kinase B/PKB; cGMP, cyclic guanosine 3′,5′-monophosphate; EGFR, epidermal growth factor receptor; HIF-1α, hypoxia inducible factor-1α; HO-1, heme oxygenase-1; MAPK, mitogen-activated protein kinases; NF-κB, nuclear factor-kappa B; NO, nitric oxide; Nrf2, nuclear factor erythroid 2-related factor 2; PKG, protein kinase G; sGC, soluble guanylyl cyclase. Color images are available online.

The direct effects of NO, which happen at low steady-state levels of NO (nM), involve chemical reactions of NO with biological target molecules (e.g., direct reactions with transition metals in heme proteins to yield nitrosylated derivatives), which tend to be regulatory in nature but may be cytoprotective or even cytotoxic in some circumstances. ³¹ In contrast, the indirect effects, which arise at relatively high NO levels (mM), include reactions of NO with O₂ or O₂⁻ to form reactive nitrogen species (RNS) such as peroxynitrite anion (ONOO⁻) and S-nitrosothiols, which subsequently interact with biological targets and tend to be cytotoxic and to impair the activity of target enzymes and proteins, including transcription factors and mitochondrial enzymes, producing long-term cellular effects. Furthermore, the indirect effects can be subdivided into nitrosative stress and oxidative stress.^31,32

NO signaling transduction (NO/cGMP-dependent or -independent signaling)

NO signaling can be classified into classic cyclic guanosine 3′,5′-monophosphate (cGMP)-dependent or nonclassic cGMP-independent mechanisms. ²⁵ The cGMP-dependent pathway is thought to be a mode of “long-range” signaling, as the signal travels a relatively long distance from the NO sources. In contrast, the cGMP-independent pathway is referred to as a “short-range” scheme, since the action occurs relatively close to the NO source and often within a certain subcellular location. ³³ For instance, in cerebellar granular neurons, endogenous NO exerts its effects in a cGMP-dependent manner, while exogenous NO shows its effects in a cGMP-independent manner. ³⁴

The cGMP-dependent pathway mediates the direct actions of NO; the most representative example involves eNOS-generated NO in vascular endothelial cells. Upon diffusion into the subendothelial space, NO undergoes nitrosylation and activates the heme enzyme-soluble guanylyl cyclase (sGC). ³¹ sGC is a heterodimer made up of two different alpha and beta subunits. In general, two alpha subunits (α₁ and α₂) and two beta subunits (β₁ and β₂) are recognized, but β₂ has not been identified at the protein level and does not generate a catalytically active sGC upon coexpression with the other subunits, thus indicating that the β₂ gene may be a pseudogene.^25,35,36 Most tissues possess the heterodimers, α₁β₁ (NO-GC1) and α₂β₁ (NO-GC2), which have been shown to exhibit enzymatic activity and to be potentially expressed in almost all cell types except platelets and erythrocytes; however, the ratios of the isoforms in individual cell types are not known. ³⁵ Under physiologically low NO concentrations, cNOS-generated NO reacts directly with the iron atom in the heme group of sGC to activate enzymatic conversion of guanosine 5′-triphosphate (GTP) into the intracellular second messenger cGMP. cGMP exerts its effects primarily by cGMP-dependent protein kinase (protein kinase G or PKG). ³⁷ Upon facilitating the formation of cGMP, the NO-sensitive enzyme sGC acts as a receptor for NO within the NO-cGMP signaling cascade, triggering a variety of cellular responses. ²⁵ Under pathological conditions (e.g., endothelial dysfunction), NO-dependent cGMP signaling is impaired; thus, novel NO-independent sGC stimulators could be useful as both pharmacological tools and therapeutics. ³⁸

In addition to activating sGC to produce cGMP, NO has also been implicated to play critical roles in normal physiology as well as in pathological changes by the cGMP-independent signaling pathway (e.g., through S-nitrosylation of cysteines and nitration of tyrosines), which is mainly involved in the indirect effects of NO.^25,27 The protein S-nitrosylation reaction (in which NO reacts with cysteine thiols to form S-nitrosylated proteins) is an important posttranslational modification and thus regulates protein structures and functions.^33,39–41 Tyrosine nitration is another modification induced by NO-derived species. NO reacts with O₂⁻ in a diffusion-limited manner to form ONOO⁻, which may cause significant oxidative damage on various target biomolecules, such as lipids, sugars, DNA, and proteins.^25,42

Expression of NOS and NO/sGC/cGMP signaling molecules in dental pulp tissue

Under physiological conditions, nNOS has been identified in human pulpal nerve fibers and rat molar odontoblasts (Table 1).^47,51 eNOS has been identified in both endothelial cells and odontoblasts within healthy human dental pulp ⁵² and rat molars.^47,53 These findings suggest that cNOS are physiological sources of endogenous NO in odontoblasts. It is generally recognized that iNOS is not expressed in healthy pulp tissue ⁴⁴ ; however, some researchers have detected weak iNOS protein expression in healthy pulp.^54,55

In inflamed dental pulp tissue, increased eNOS messenger RNA (mRNA) and protein levels with coexistent vasodilation were identified in human samples ⁴⁴ ; however, a later study found a decreased expression of eNOS mRNA in rat odontoblasts and DPCs at the early phase (1–3 days) after tooth preparation. ⁵³ Enhanced mRNA and protein levels of iNOS have also been identified in inflamed or hyperemic rat pulp tissue, which is mainly composed of macrophages, neutrophils,^3,53 leukocytes, and odontoblasts,^44,53 along with perivascular lymphocyte and erythrocyte infiltration. ⁵⁴ Notably, it is generally accepted that iNOS protein is absent from most cells under resting conditions, but that the iNOS gene is regulated by a range of inflammatory mediators, such as bacterial LPS, double-stranded RNA, and cytokines (e.g., IL-1, TNF-α, and IFN-γ).^22,56 For instance, in an LPS-induced experimental model of rat pulpitis, the gene expression of iNOS has been found to be dramatically increased 3 h and remain upregulated after 3 days despite a reduction by 9 h compared with that in untreated teeth, ⁵⁷ suggesting an essential role of iNOS induction in high concentrations of NO inside the inflamed pulp tissue. In addition, several investigations have indicated that nNOS and eNOS are also under the influence of expression regulation and that iNOS is present even under resting status.^58,59 Thus, it has become manifest that all three NOS isotypes are simultaneously expressed under both physiological and pathological states.

As mentioned above, since NADPH is a common histochemical marker for NOS, many studies have histochemically analyzed NADPH-diaphorase (NADPH-d) to determine the localization of NOS in dental pulp. It has been reported that NADPH-d-positive/NOS-immunoreactive perivascular and solitary varicose axons are present in the dental pulp and gingiva of cats and dogs. ⁶⁰ In healthy human pulp, strong NADPH-d activity has been identified in the odontoblastic layer and in DPCs, endothelial cells, and Schwann cells, while weak positivity has been detected in the other parts of the central area; in addition, NADPH-d staining density has been found to be more intense within carious teeth with chronically inflamed pulp than in teeth with normal pulp in the stromal area, as indicated by positive staining of inflammatory cells, angioblasts, fibroblast, collagen fibers, and endothelial cells in abundant, newly formed capillaries.^8,61

Similar to the case in other tissues, the NO-sensitive enzyme/receptor sGC and its product cGMP have also been identified in dental pulp tissue.^47,54 In healthy rat molar pulp tissue, the sGC α₁ and α₂ subunits are localized in odontoblasts and DPCs, while the β₁ subunits and cGMP are localized in odontoblasts, odontoblast processes, and DPCs.^47,54 Challenge of the rat molar dentin/pulp complex for up to 30 min using the NO donor (viz. synthetic NO-releasing compound) spermine-N-(2-aminoethyl)-N-(2-hydroxy-2-nitrosohydrazino)-1,2-ethylenediamine (NONOate) increases cGMP staining intensity in odontoblasts, while stimulation with the NOS inhibitor N^G-nitro l-arginine methyl ester (l-NAME) attenuates cGMP staining intensity, indicating an action of NO on the activation of sGC in odontoblasts; conversely, the critical enzyme sGC mediates the diverse biological actions of NO in healthy odontoblasts. ⁴⁷ Inflammation of the dentin/pulp complex decreases immunostaining of the α₁, α₂, and β₁ subunits in a subunit-specific manner (moderately decreasing β₁ subunit staining and much more strongly decreasing α₁ and α₂ subunit staining) compared with that in healthy dental pulp; these decreases are accompanied by desensitization of sGC to NO activation.^54,62 This suppression of sCG subunits may be regulated by inflammation-dependent generation of inflammatory mediators (e.g., bacteria LPS, IL-1, TNF-α), reactive oxygen species (ROS), and RNS, in both a NO-dependent and a NO-independent manner.^63–66

All above findings are indicative of a plastic pattern of NOS localization and expression that occurs in an isoform- and a pulp status-dependent manner. Research has revealed that the induction of NOS and signaling by relevant molecules are mediated by various proinflammatory molecules, but few studies have explored the role of cGMP in regulating microcirculation in the dental pulp. Hence, the roles of NOS and the NO/sGC/cGMP pathway in dental pulp circulation need to be clarified in the future, as these roles may have great significance for pulp revascularization.

nNOS (NOS I)/NO and dental pulp innervation and nociception

nNOS, which is predominantly expressed in neurons, produces NO that can act as an anterograde and retrograde messenger at synapses involved in nociceptive processing. Moreover, nNOS is also identified in the vascular endothelium and has been considered as an important contributor to the maintenance of cardiovascular system homeostasis. ⁶⁷ In addition to the involvement in physiological activities, recent studies have demonstrated that nNOS plays important roles in several mechanisms related to neuronal degeneration/plasticity and cardiovascular disorders.^68,69

Dental pulp tissue contains both sensory and autonomic nerves to fulfill its defensive and vasomotor functions. ⁶ The trigeminal nerve endings, pulpal nerve fibers, and odontoblasts are together responsible for the sensory innervation of the pulp and are all nNOS positive.^47,70,71 During human tooth germ development, nNOS is expressed in stellate reticulum cells, ameloblasts, and odontoblast clusters at much higher levels in the advanced stage than in the early stage. ⁷¹ A previous study reported that the growth of axons into molar tooth pulp parallels increases of NOS activity in rat undergoing inferior alveolar nerve transection, suggesting a favorable action of increased NOS activity in axotomized trigeminal ganglion neurons on the regeneration of injured neurons. ⁵⁰ These findings raise the possibility that the neuronal functions of NOS and NO may depend on instant circumstances such as the localization of these molecules and the statuses of the cells.

Under inflammatory conditions, both the numbers of nNOS-positive neurons and subsequent production of NO are clearly elevated in Sprague-Dawley rats with experimentally induced chronic pulp inflammation compared with healthy rats, ⁷² and the increased production of NO may facilitate neurotransmitter release and enhance nociceptive transmission. Peripheral injury due to local pulp inflammation can induce central sensitization and neuroplastic changes in the relative amounts and activity of nNOS in trigeminal nociceptive neurons through activation of glutamate receptors and N-methyl-d-aspartate (NMDA) receptors.^73–75 However, after a certain period of time, the nitrergic system shifts back to the initial state independently of the persistence of pulp inflammation. ⁷⁴ Another study has indicated that NO does not drive pulp inflammation through nNOS but may mediate the pulpal inflammatory events by affecting perilesional pulp tissue and the neighboring endothelial/vascular structures. ¹³

Thus, we speculated that nNOS/NO may play an important role in dental pulp neural transmission. However, it is not involved in neurogenic inflammation that is believed to be mediated by neuropeptides released from sensory nerves at the site of inflammation, such as substance P and ROS. ⁶ Overall, the existing studies have not clarified the action of nNOS in dental pulp inflammation, and there is a lack of relevant studies on the function of nNOS in pulp microcirculation.

eNOS (NOS III)/NO and dental pulp microcirculation

eNOS is mainly detected in vascular endothelial cells and is the major isoform regulating vascular function. ⁶⁷ Increases in intracellular Ca²⁺ due to vasodilator agonists, such as acetylcholine and bradykinin, induce activation of eNOS and enhanced synthesis of NO. NO liberated from the vascular endothelium can regulate vascular tone and blood flow through activating sGC/cGMP signal; it can also inhibit platelet aggregation and adhesion, leukocyte adhesion and migration, and vascular smooth muscle cell proliferation, thus preventing atherosclerosis.^11,12 In addition, eNOS exerts an anti-inflammatory effect through promotion of NO and regulation of cGMP signaling and nuclear transcription factor (nuclear factor-kappa B [NF-κB]) pathway, thus inhibiting the expression of inflammatory genes. ⁷⁶

The dental pulp microcirculatory vessels is characterized by a hierarchical structure, low compliance, firm and resilient tissue, and a lack of a true collateral blood supply. ⁶ As suggested above, endothelial cells, DPCs, and odontoblasts in healthy human dental pulp are all eNOS-positive cells, and evidence shows that NADPH-d reactivity in pulpal vessels supports the role of NO as a mediator of dental vascular homeostasis by NO/sGC/cGMP signaling.^12,47,52,77 A previous investigation using anesthetized rats reported that local application of l-NAME to a thinned dentine layer of the lower incisor could reduce the diameter of pulpal arterioles under both rest and bonding agent-induced hyperemic conditions. ⁷⁸ The authors hypothesized that the vasoconstriction was caused by the suppressive effect of l-NAME on eNOS in both endothelial cells and odontoblasts. However, l-NAME is not a specific NOS inhibitor; rather, it weakly inhibits all types of NOS and can even release NO.^79–81 Therefore, the above statement needs to be strengthened by further studies using a selective eNOS inhibitor. It is worth noting that different opinions exist regarding the role of eNOS in pulp flow. For example, another group concluded that l-NAME treatment alone has no obvious effects on pulpal blood flow in the feline lingual artery, nor does l-NAME pretreatment affect the actions of multiple vasodilators, such as substance P; these findings suggest that these processes are not mediated through an endothelium-dependent mechanism. ⁸² Moreover, previous studies have identified the effective regulation of neurons on dental pulp microcirculation under physiological and pathological conditions.^83,84 Similarly, eNOS-mediated pulp vasodilation is also a valuable response in neurogenic inflammation and involves subsequent release of NO produced in endothelial cells and of inflammatory mediators (e.g., substance P) from sensory nerve endings.^85,86

Angiogenesis, another critical physiological event associated with wound healing and tissue regeneration, involves complex cellular activities correlated with cell proliferation and migration.^87,88 The role of eNOS-produced endogenous NO in angiogenesis has been confirmed by the impaired angiogenetic capacity in eNOS-knockout mice in response to ischemia. ⁸⁹ Moderate concentrations of eNOS-dependent NO production can significantly regulate endothelial proliferation and migration induced by other key proangiogenic factors, such as vascular endothelial growth factor (VEGF), bradykinin, vitamin D, etc., and therefore contribute to angiogenesis and wound healing.^90–93 In dental pulp tissue, cNOS- and VEGF-mediated angiogenesis plays a key role during dental embryo morphogenesis. ⁷¹ Bradykinin has been found to exert dual effects on eNOS in the context of odontoblast activation and on protein kinase B (PKB/Akt)-dependent upregulation and extracellular signal-regulated kinase 1/2 (ERK1/2)-independent downregulation of eNOS. ⁹⁴

Besides its effects on vascularization, eNOS likely exerts an anti-inflammatory effect in healthy pulp and during the early phases of inflammation in hyperemic pulp by restraining the aggregation/adhesion on the endothelial surface of platelet and leukocyte and by increasing tissue perfusion.^8,47,95 Furthermore, eNOS is localized in rat and human odontoblasts and is a physiological source of NO in odontoblasts. NO can modulate nociceptive input in double directions: small amounts of peripherally generated NO are algesic, while large amounts of NO are analgesic.^44,96,97 Thus, it is possible that eNOS is also involved in nociceptive transmission in dental pulp.

In conclusion, eNOS located in endothelial cells of dental pulp may mediate local vasodilation, inflammation, angiogenesis, and nociception, but the exact functions of eNOS in odontoblasts are still unknown. It is assumed that eNOS-produced NO in odontoblasts regulates the vascular tone of adjacent vessels. Further studies should be carried out to clarify the role of eNOS in the crosstalk between endothelial cells and DPCs/odontoblasts during dental pulp neovascularization.

iNOS (NOS II)/NO and dental pulp inflammation

In comparison with cNOS, iNOS expression is absent in strictly resting cells; instead, it is induced by immunological irritation, such as proinflammatory cytokines and bacterial LPS, in a Ca²⁺-independent manner.^79,98 iNOS-produced NO plays distinct roles in cell activities; for example, it stimulates apoptosis, inhibits mitochondrial respiration, regulates oxidative phosphorylation, and exhibits cytotoxicity in target cells. ²² iNOS-produced NO has also been implicated in the pathogenesis of a range of human diseases, such as arteriosclerosis, diabetes, rheumatoid arthritis, cancer initiation, aging, and acute and chronic inflammation.^22,25 With regard to teeth, an example is the iNOS-mediated inflammation-induced bone resorption on the compression side during orthodontic tooth movement. ⁹⁹

Crosstalk between NOS isoforms in dental pulp tissue

Low amounts of NO (presumably in the nM range) produced by cNOS are thought to primarily contribute to the regulation of physiological events; high NO levels (μM) induced by enhanced iNOS expression also play key roles in protective host immune defense, but are nonetheless associated with pathological conditions (e.g., infectious diseases and sepsis). ¹³³ iNOS is a NO regulator that can be affected by NO donors, whereas cytokine-induced iNOS expression is mainly regulated by the transcription factor NF-κB.^27,133 Under physiological conditions, low levels of NO (pM–nM) synthesized by cNOS inhibit NF-κB. Soon after a pathological stimulus (e.g., bacterial LPS and/or a variety of proinflammatory cytokines) arrives, cNOS is phosphorylated at a tyrosine residue, thus resulting in enzymatic suppression and decrease of physiological NO levels. Under these conditions, NF-κB is liberated from the inhibitory action of NO; it translocates into the nucleus and participates in the transcription of numerous genes, including iNOS.^133–136 In summary, activation of NF-κB requires early declines in basal NO levels (nM range) that are dependent on signaling leading to the inhibition of cNOS, suggesting that the production of NO by cNOS is a suppressor for iNOS expression. Thus, crosstalk between cNOS and iNOS may exist in the context of inflammation. ¹³⁷

Regarding dental pulp tissue, some information from previous studies supports the possibility of crosstalk between cNOS and iNOS. First, in inflamed or hyperemic pulp tissue, decreased expression of eNOS mRNA in odontoblasts and DPCs has been detected ⁵³ ; in contrast, the iNOS gene and protein expressions of macrophages, neutrophils, and odontoblasts are significantly enhanced.^3,53 These results are consistent with the concept that the shift of the inflammatory process into irreversible pulpitis in humans decreases eNOS levels but increases iNOS expression, which may be explained by the negative feedback that is provoked by the large amounts of NO synthesized by the significantly upregulated iNOS.^44,138 Hence, we speculate that eNOS and iNOS are inversely regulated in pulp tissue. Moreover, as described above, low amounts of NO (nM) play a protective role by preventing accidental activation of NF-κB, while greater amounts (μM–mM) can induce iNOS expression and NF-κB activation. The NO donor NOC-18 at a concentration of 10 μM has been found to stimulate iNOS expression and NO production, promote odontogenic differentiation through the TNF-α/NF-κB axis, and induce tertiary dentin formation in vivo in rat DPCs, ⁴⁹ and 1 mM of SNP can stimulate IL-8 production through activation of NF-κB and MAPK. ¹³² Second, different NOS isoforms have different ligand efficacies. In healthy dental pulp of Wistar rats, stimulation of muscarinic acetylcholine receptors (mAChRs) with the orthosteric agonist pilocarpine (10⁻⁹ M) exerts no clear effect on cNOS expression, but pilocarpine at 10⁻⁷ M can trigger release of PGE₂ and NO through activation of nNOS and eNOS.^139,140 In contrast, in LPS-treated rat pulp, a low concentration of pilocarpine (10⁻⁹ M) can inhibit iNOS mRNA and NOS activity in vivo and in vitro^138,140,141 without exerting an obvious effect on cNOS expression, while a high concentration (10⁻⁷ M) produces a positive effect correlated with elevated cNOS gene levels without affecting iNOS expression. ¹⁴⁰ Thus, cNOS and iNOS might be inversely regulated in cells or tissues depending on cell status.

Similarly, numerous studies have suggested the existence of crosstalk between nNOS and eNOS. First, their anatomical distributions may provide some clues: (1) nNOS is mainly expressed in parasympathetic vasodilator neurons; (2) and that is also present in nonneuronal cells; (3) eNOS-NO-cGMP signaling molecules has been involved in neurotransmission.^47,97 Some typical studies are as following: in an experimental migraine model with stimulation of the rat trigeminal ganglion, most nNOS-immunoreactive nerve fibers were found to be associated with branches of the anterior meningeal artery ⁷⁰ ; another study has indicated the existence of NADPH-d-positive/NOS-immunoreactive perivascular and solitary varicose axons in the dental pulp and gingiva of cats and dogs that may regulate local blood flow ⁶⁰ ; in human dental pulp, parasympathetic innervations are distributed around small blood vessels. ¹⁴² Second, eNOS can produce NO to maintain the health of oral tissue circulation, not only under resting states but also under conditions subjected to physicochemical stimuli, which may be related to the pathophysiology of inflammatory dental diseases, such as pulpitis and gingivitis. ¹² Further research on the correlations between the cellular and subcellular localizations of nNOS, eNOS, sGC, and cGMP and alterations in cGMP staining after challenge with an exogenous NO delivery compound might elucidate the involvement of different NO-sGC-cGMP signaling molecules in neurotransmission modulation, vascular regulation, and dentinogenesis in the dentin/pulp complex, and their crosstalk in these processes. ⁴⁷

A more recent study reported that the knockout of nNOS in mice can increase the eNOS expression in DPCs and endothelial cells inside the dental pulp, and further induce positive expression of iNOS in odontoblasts that was negative in the control mice. ¹⁴ The above findings indicate that crosstalk may exist among three NOS isoforms in dental pulp tissue. However, compared with corresponding evidence for other tissues and cells, the available evidence supporting this notion for dental pulp is obviously unconvincing; therefore, additional high-quality studies should be conducted in the future.

In brief, healthy and inflamed dental pulp tissues have different NOS expression levels and localizations.^44,52,54 In healthy pulp, cNOS (eNOS and nNOS) is constitutively expressed in a Ca²⁺-dependent manner such that it can rapidly produce low levels of NO (pM–nM range), which play protective roles by maintaining pulpal homeostasis and regulating other physiological events. iNOS, an inducible Ca²⁺-independent isoform, is mainly expressed in inflamed pulp tissue as a result of external stimulation and synthesizes high levels of NO (μM–mM range) for sustained periods of time. These large amounts of NO are involved in inflammatory processes, including pulp defense and repair/regeneration. Evidently, crosstalk between NOS isoforms exists but is far from being understood and needs to be verified by extensive research.

Targeting NO signaling with NO donors, NO scavengers, and NOS inhibitors from NO-delivering biomaterials

With the progress in understanding the critical role of NO in tissue repair and regeneration under physiological and pathological conditions, such as cardiovascular, anti-inflammatory, antibacterial, antiparasitic, and anticancer processes, research focusing on NO-based therapies has rapidly increased. ¹⁴⁶ Most studies on the biological activity of NO have used various hybrid drug compounds, such as NO donors, NO scavengers, and NOS inhibitors, to elevate or suppress NO levels or activity in the extracellular microenvironment to manipulate cell behaviors.

NO donors (viz. NO-releasing compounds) preserve NO in their molecular structures and evoke biological activity after decomposing. Some examples include organic nitrates, SNP, furoxans, and compounds containing the N(O)NO functional group (e.g., the diethylamine-NO complex or spermine-NO complex). ¹⁴⁷ In numerous experimental studies, different NO donors have been adopted to monitor the biological effects of exogenous NO on DPCs, which have shown promotive effects on odontoblastic differentiation and subsequent reparative dentin formation in varying degrees (Table 2).^15,49,53 For example, an in vivo analysis has revealed that a NO-releasing biomimetic nanomatrix gel incorporating antibiotics (ciprofloxacin and metronidazole) can promote tooth revascularization with apexification in healthy beagles. ⁴⁸ Hence, NO donors may serve as a promising adjuvant in regenerative endodontics to promote the angiogenic and odontogenic events. However, only a few studies have explored the participation of exogenous NO in the cellular activities of DPCs, and the results are still controversial.^{15,49,124,148,149} The amounts of NO, cell type, and the states of cells may be pivotal determinants of the biological actions of NO. For example, low concentrations of NO tend to maintain the cell vitality although enhancing the expression of cytoprotective proteins such as HO-1; in these cases, NO is used as a signaling messenger. In contrast, high concentrations of NO appear to immediately cause cell damage before the production of cytoprotective proteins. ¹²⁴

Controlling bacterial progression and subsequent inflammatory reaction is a well-known prerequisite for pulp repair and regeneration. ⁷ NO is a highly potent radical with good antimicrobial property generated by many cell types, which acts as an inhibiter or killer on a variety of microorganisms under inflammatory conditions, such as E. faecalis and Treponema denticola. ⁴⁸ Its antibacterial property is primarily derived from interactions with O₂⁻ that form highly cytotoxic ONOO⁻ and S-nitrosylation of thiol residues that possesses strong oxidizing property and cytotoxicity. Thus far, macrophages and neutrophils have been the host cells most thoroughly investigated in terms of iNOS localization, and these cells possess antibacterial effect against aerobic and anaerobic bacteria through an iNOS-derived NO mechanism. ⁹⁸ Likewise, human odontoblasts have also been reported to produce endogenous NO by synthesizing large amounts of iNOS, thus exhibiting antibacterial activity against caries-related microorganisms, for example, S. mutans, through TLR-2 activation. ⁴⁶ In addition, recent studies reported that exogenous NO from a NO-releasing compound, such as pluronic F68-branched polyethylenimine-NONOates (F68-BPEI-NO) and NO-releasing silica nanoparticles, showed an adequate antimicrobial effect against specific periodontal pathogens with minimal cytotoxicity to human gingival fibroblasts.^108,150

In contrast to NO donors, synthetic NO scavengers, such as 3-methyl-1-phenyl-2-pyrazolin-5-one (Edaravone) and 2-phenyl-4,4,5,5-tetramethylimidazoline-1-oxyl-3-oxide (PTIOs), can directly counteract NO through direct radical/radical coupling interactions with NO and may play beneficial roles in therapeutic strategies for oxidative stress-related diseases. ²⁹ A recent study reported that PTIOs can negatively regulate the odontoblast differentiation of rat DPCs induced by the NO donor NOC-18. ⁴⁹ However, similar studies on the anti-inflammatory, proangiogenic, and antibacterial roles of NO scavengers in dental pulp are rare and need to be conducted in the future.

NOS inhibitors are other important agents that can potentially be used to modulate some pathological disorders (e.g., mood disorders, ischemic stroke, arthritis, pancreatitis) by selectively targeting all NOS isoforms.^58,146 Nonselective NOS inhibitors, such as l-NAME and N^G-monomethyl-l-arginine (l-NMMA), have been used to verify the roles of NO derived from all NOS isoforms. However, as there are locational and functional differences among the three NOS isoforms, achieving selective inhibition using selective NOS inhibitors is important to decrease the risk of side effects. For example, the nNOS inhibitor compound 24 exhibits potency in the nM range with a superior selectivity index (472-fold and 239-fold higher selectivity for nNOS than for eNOS and iNOS, respectively), ¹⁴⁶ while a NOS inhibitor named aminoguanidine exhibits a 50-fold greater effect on the enzymatic activity of iNOS than on that of eNOS and nNOS and thus can reduce the intensity of inflammatory responses in periapical lesions in vivo. ¹⁵¹

As previously discussed, NO can cause strong relaxation of the smooth muscle of blood vessels and can enhance vascular permeability to promote healing in inflamed pulp; however, if overproduction of NO continues for a long time, these phenomena might become harmful to the low-compliance pulp tissue. Thus, administration of an NOS- or iNOS-specific inhibitor may bring benefit on pulp healing or regeneration.^3,72 Notably, NO may be induced by different pathways independent of endothelial cells, and NOS inhibitors also exert multiple nonspecific effects, including antagonism of mAChRs, generation of O₂⁻, suppression of cytochrome reduction, and restraint of endothelium-independent relaxation induced by cyclic adenosine monophosphate (cAMP) or amiloride. ⁵⁸ In addition, as mentioned above, iNOS-related NO can modulate the M1/M2 macrophage ratio ¹¹³ ; the reductions of this ratio may diminish the inflammatory activities and accelerate tissue repair process in the tooth suffering root resorption. ¹⁵² Thus, we may speculate that employing NO donors or NOS inhibitors may be a plausible therapeutic strategy to deal with orthodontic root resorption.

In summary, NO-based therapies using different types of NO donors, scavengers, and NOS inhibitors have shown some potential for clinical applications. However, few of them have been translated into practice because of drawbacks such as burst release, low payloads, and untargeted delivery. Currently, various biomaterials that have the potential to load and deliver NO in a sustained and/or controlled manner have been investigated that comprise hydrogels, nanoparticles and microspheres, nanoporous membrane, electrospun mats, and surface coatings; the combination of NO donors and these matrices may provide an occasion not only to control the release profiles of the donors but also to govern their cellular performances.^24,32,153

Stimulation of NO signals through mineral trioxide aggregate-like cements

Biomaterials are foreign to the host, and the immune interactions between materials and tissues are very important for their biocompatibility evaluation and application. ¹⁶ Thus, copious basic and clinical evaluations of the immune response and odontoblastic differentiation of DPCs in response to different endodontic bioactive materials have been conducted.

As mentioned above, increased NO and PGE₂ production through upregulation of iNOS and COX2 is a major inflammatory response in dental pulp.^154,155 The NOS/NO system is the major regulator of many cellular functions, including survival, migration, differentiation, apoptosis, etc.; this system participates in various physiological actions, such as vasodilation, angiogenesis, nociception, antithrombosis, and scavenging of ONOO⁻ (anti-inflammatory processes).^11,156 IL-1 is a typical multifunctional cytokine involved in immune responses, hematopoiesis, and inflammation with regard to both innate and adaptive immunity ¹⁵⁷ ; it is also a predominant cytokine in pulpal inflammation. ¹²⁰ The most salient property of IL-1 (mainly IL-1α and IL-1β) is its ability to activate iNOS and COX2, which accounts for the production of NO and PGE₂. ¹⁵⁸ HO-1 possesses a cytoprotective effect against oxidative damage and inflammatory responses and thus may contribute to good clinical outcomes through a sequence of healing events.^123,159 Hence, iNOS/NO, PGE_2, IL-1 and HO-1 have frequently been selected as pivotal markers for evaluation of the biological characteristics/immunocompatibility of endodontic bioactive cements.

Mineral trioxide aggregate (MTA), mainly comprising calcium and silicate elements, has been developed in gray (GMTA) and white (WMTA) forms and is widely used as an bioactive endodontic material, including in various vital pulp therapies. ¹⁶⁰ Human DPCs cultured on a calcium silicate cement substrate have been found to exhibit significant upregulation of proangiogenic factors (von Willebrand factor and angiopoietin-1 [vWF and Ang-1, respectively]) and to exhibit downstream eNOS protein expression and NO release. ¹⁶¹ Minamikawa et al. ¹²¹ found that MTA (ProRootMTA; Dentsply/Tulsa, Tulsa, OK) can increase iNOS and COX2 gene expression through activation of NF-κB signaling in rat DPCs, thereby stimulating the phosphorylation and degradation of IκB as well as the release of PGE₂. Similar results obtained with human DPCs have indicated that MTA can increase COX2 expression. ¹⁶²

Although calcium silicate ceramics are attracting increasing interest in the bone and dental fields, they still have some drawbacks, such as the inherent brittleness of ceramics, which prevents their mechanical properties from simulating those of cortical bone; their extended setting time; and their sandy consistency, which makes the materials difficult to handle.^163,164 Hence, many strategies have been adopted to overcome these drawbacks and improve the biological performances of calcium silicate cements. For example, calcium silicate/gelatin composites with increased initial mechanical strength and biological functions that can effectively inhibit iNOS and IL-1 expression but activate IL-10 expression in human DPCs have been developed. ¹⁶⁵ In endodontic practice, various calcium silicate-based materials (MTA-like materials), such as Portland cement (Brasseler USA, Savannah, GA), BioAggregate (Innovative Bioceramix, Vancouver, Canada), Biodentine (Septodont, Saint Maur des Fosses, France), and iRoot BP (Innovative Bioceramix), have been developed for use; the manufacturers claim these products have characteristics similar to those of MTA without exhibiting its shortcomings. ¹⁶⁶

Bismuth oxide (Bi₂O₃; a radiopacifier)-containing Portland cement (BPC) is able to elevate nitrite (a marker of NO production) concentrations and iNOS mRNA expression that associates with HO-1 expression in DPCs. Interestingly, BPC-induced HO-1 expression in DPCs was found to attenuate its own cytotoxic effects. ¹⁶⁷ Several novel calcium silicate cements (e.g., Angelus-MTA; Angelus, Londrina, Brazil; Ortho-MTA; BioMTA, Seoul, Korea) have been developed that use zirconium oxide instead of bismuth oxide to alleviate tooth discoloration from MTA; these all reduce the expression levels of iNOS mRNA, COX2 protein, proinflammatory mediator (such as NO, PGE₂, TNF-α, IL-1β, IL-6, IL-8, and matrix metalloproteinase [MMP; MMP-1, MMP-2, MMP-8, and MMP-9]) mRNA, and intracellular ROS through ROS-dependent MAPK and Nrf2 activation and subsequent HO-1 expression. ¹⁶⁸ However, one study observed higher IL-1 and iNOS levels in human DPCs cultured on two cements, that is, radiopaque dicalcium silicate cement and WMTA, than in those cultured on a tissue culture plate, possibly because of the initial high pH values, the heat produced by material setting reaction, and the generation of proinflammatory cytokines, such as IL-1 and IL-6, during the procedure.^154,169

Although they exhibit various biological effects, nearly all MTA-like cements are highly biocompatible under odontogenic and angiogenic conditions, and we are still unable to recommend one specific material over the others. ¹⁷⁰ MTA-like materials exert anti-/proinflammatory effects through efficiently regulating the production of NOS mRNA and protein (eNOS and iNOS), COX2 protein, NO and PGE₂, and macrophage polarization that may contribute to their excellent clinical performance, at least in part.^{111,161,168,170–173} However, it is unclear which NOS isoforms and signaling pathways are related to this process. More detailed data related to the effects of various MTA-like materials (viz., calcium silicate cements) on DPCs can be gleaned from a review by Emara et al. ¹⁷⁰

Other potential drugs that affect NO signals

Sufficient control of infection and survival of stem/progenitor cells are two primary goals of all regenerative endodontic procedures, but traditional agents such as sodium hypochlorite, Ca(OH)₂, and antibiotic pastes have some drawbacks and cannot reliably achieve these two goals simultaneously. ⁷ For instance, the high pH of Ca(OH)₂ makes it potentially cytotoxic and causes it to tend to dissolve soft tissues and impair macrophages phagocytosis of E. faecalis, thereby resulting in chronic inflammation and cell necrosis. ¹⁷⁴ Thus, various newer agents or drugs that affects NO signals and have cytoprotective and anti-inflammatory effects may be potentially adopted in pulp regeneration therapy.

MMPs are known to degrade the extracellular matrix, thus affecting cell apoptosis, migration, differentiation, and inflammatory processes. MMP-3 has been reported to downregulate LPS-induced NO synthesis, inactivate or counteract proinflammatory mediators, and promote clearance of inflammatory cells; its application to injured pulp tissue can induce angiogenesis and wound healing,^57,175 which suggests the potential utility of MMP-3 in the control and elimination of pulpal inflammation.

Amino acids are well documented to interact with NOS/NO in the context of inflammation and to participate in the modulation of immune activities. ¹³⁴ Glutamine is a nonessential amino acid that is synthesized by organisms as needed. Defined levels of exogenous glutamine (10 mM) have been shown to modulate immunosuppressive properties of bone marrow mesenchymal stem cells (BMSCs). ¹⁷⁶ Thus, relevant studies have been conducted to determine its role in pulp tissue activity and have reported that it can promote growth, migration, and differentiation in human DPCs through the BMP-2 and MAPK pathways. ¹⁷⁷ In addition, glutamine can reduce LPS-induced iNOS and COX2 protein expression as well as the release of NO, PGE_2, and other cytokines (viz. IL-1β, TNF-α, and IL-8) in a dose-dependent fashion, thereby exerting an anti-inflammatory effect by blocking the MAPK (p38, ERK, and JNK) pathways, with activation of MAPK phosphatase 1 (MKP-1) and subsequent inhibition of NF-κB translocation, in LPS-treated human DPCs. ¹⁰³

Some natural compounds (e.g., davallialactone, sappanchalone, and pachymic acid) extracted from plants and fungi exhibit anti-inflammatory effects by stimulating NO signals and are now emerging as significant agents for pulpitis treatment. Davallialactone shows anti-inflammatory and antioxidant effects that decreases the level of inflammatory molecules such as iNOS and COX2 in LPS-induced human DPCs through inactivation of ERK1/2 followed by blockade of NF-κB translocation from the cytosol through the nuclear membrane. ¹²⁸ Sappanchalone is a bioactive flavonoid isolated from Caesalpinia sappan L. that exhibits many biological activities, including neuroprotective and anti-inflammatory activities. In human DPCs, 5–40 μM sappanchalone can increase the protein expression of HO-1 and HO activity in a concentration-dependent manner through activation of JNK and nuclear translocation of Nrf2, which results in the suppression of LPS-induced inflammatory mediators, decreasing iNOS and COX2 protein expression, NO and PGE₂ production, and IL-1β, IL-6, IL-12, and TNF-α levels. ¹²⁵ Epigallocatechin gallate, a major green tea polyphenol, has various pharmacological effects on cell vitality. It has also been reported to serve as a scavenger of ROS and regulator of B cell lymphoma 2 (Bcl-2) family, and thus exert a protective effect against NO-induced apoptosis in human DPCs. ¹⁷⁸ A more recent study suggested that Schisandrin C, a natural compound extracted from Schisandra chinensis, can suppress the LPS-induced formation of NO and other inflammatory mediators (IL-1β, TNF-α, ICAM-1, VCAM-1, MMP-2/9, and ROS) in human DPCs and thus has the potential to be used as an anti-inflammatory agent in the context of pulpitis to reduce inflammation and oxidation through downregulation of the MAPK (ERK1/2 and JNK) and NF-κB pathways and to promote mitochondrial biogenesis through upregulation of p-Akt- and Nrf-2-mediated HO-1 expression. ¹⁷⁹ Therefore, the favorable anti-inflammatory abilities of these natural compounds might enable their use in treatments for multiple oral diseases such as pulpitis and periodontitis.

In addition, some specific drugs used for other diseases, such as bone tumors, immune diseases, and metabolic diseases (diabetes), have also emerged as potential agents for the control of dental pulp inflammation. For instance, zoledronic acid, a third-generation amino bisphosphonate that is used for the treatment of bone metastatic cancers at high doses, can modify immune cell profiles in rat dental pulp, thereby increasing the number of macrophages and the expression levels of some proinflammatory markers, such as iNOS, IL-1β, and TNF-α, independent of the NF-κB pathway. ¹⁸⁰ Another example is metformin, a drug for treatment of type 2 diabetes, has been found to facilitate the healing of apical periodontitis by regulating the iNOS expression and NO production. ¹⁸¹

NOS activities and mechanical stress

In addition to the above chemotherapeutic strategies, physiotherapeutic strategies, such as mechanical stress and photodynamic therapy, have been considered as candidate approaches for pulp disease treatment. Mechanical loading is key to tissue growth, especially bone remodeling; fluid flow through the bone canalicular system generates shear stress, which stimulates the anabolic activity of osteoblasts and osteocytes that governs an adaptive process of bone healing and remodeling. ¹⁵⁶ Hence, fluid shear stress within the physiological magnitude is usually applied by pulsating fluid flow in vitro to mimic in vivo loading through the lacunar/canalicular network on osteocytes. Relatively high fluid shear stress values, that is, those between 0.5 and 2 Pa, can result in the production and fluctuation of signal molecules such as intracellular Ca²⁺, eNOS-mediated NO and PGE₂, which are required for cell anabolic responses; the latter two are also regulated by the mechanosensitive gene COX2. ¹⁸²

Similar to bone cells, DPCs are also sensitive to mechanical stress, which can significantly promote cell proliferation and odontogenic differentiation. ¹⁸³ Previous studies demonstrated that pulsating fluid flow can stimulate higher levels of NO, PGE₂, and COX2 gene expression in mature DPCs than in immature DPCs within 5 min, suggesting the following two possibilities: (1) that the rapid stimulation of NO production results from the activity of eNOS (Ca²⁺-dependent NOS) but not iNOS and (2) that there is a relationship between cell maturation stage and cell mechanosensitivity.^184,185 However, iNOS expression (typical of inflammatory conditions) in pulp subjected to orthodontic forces was present but had no obvious change compared with control dental pulp, which could be explained as a response to extractive stresses. ¹⁸⁶ Moreover, mechanical irritants can provoke a range of inflammatory responses in dental pulp, such as the production of diverse cytokines and chemokines related with immune responses.^6,184 Another illuminating case is that occlusal forces can promote periodontal ligament healing through inducing NOS expression and NO production.^187,188 Thus, exertion of optimum mechanical stress (e.g., occlusal stimuli) on DPCs or odontoblasts combined with the use of specific bioactive materials [e.g., Ca(OH)₂ and MTA] may be a possible method to stimulate dentin formation through activation of NOS/NO signaling.

Triggering of NO production by phototherapy in pulp inflammation

Phototherapy or low-level light therapy, also termed photodynamic therapy or photobiomodulation, involves the use of low-powered lasers in the far-red to near-infrared spectrum (600–1000 nm) or light-emitting diodes (LEDs) of <500 mW to reduce inflammatory reaction, relieve pain, and promote tissue repair and has emerged as the novel therapeutic method in a range of medical and health fields. ¹⁸⁹ Although early studies identified mitochondrial cytochrome c oxidase (COX) in the electron transport chain as an endogenous photoreceptor for phototherapy, the molecular and cellular mechanisms underlying the effects of phototherapy remain unclear. It is said that the extracellular release of NO, ATP, or some specific growth factors may activate autocrine signaling such as the secretion of IL-1 by macrophages, resulting in the beneficial actions of phototherapy.^190–192 iNOS/NO system may contribute to the mechanism of phototherapy by affecting mitochondrial physiology.^31,193 Light can dissociate bound NO from mitochondrial COX, which restores oxygen consumption (because oxygen can rebind COX) and resumes respiratory chain activity, inducing ATP synthesis and Ca²⁺ signal. Recent investigations have demonstrated that phototherapy may increase the bioavailability of NO by releasing it from intracellular stores, mainly heme proteins.^191,192 Notably, during cell stress and subsequent inflammatory defense response, high levels of NO exert an inhibitory effect on COX activity and thereby suppress tissue repair events.^190–192

Phototherapy has been applied for the treatment of many oral conditions, including multiple oral mucosal diseases, temporomandibular joint disorders, orofacial pain, postextraction paresthesia, endodontic infections, periodontal diseases, caries, and peri-implantitis.^194,195 With regard to dental pulp, most studies have stated that phototherapy can promote the viability of DPCs ¹⁹⁶ and modulate the expression of cytokines, growth factors, and other inflammatory mediators. ¹⁹⁵ A previous study indicated that 653 nm LED irradiation exhibited a promotive effect on cell growth and responses relevant to tissue repair in rat DPCs by significantly increasing ATP, NO, (likely not induced by NOS but rather released from COX), and mitochondrial metabolic activity. ¹⁹⁰ In a later study on LPS-stressed human DPCs, an 855 nm infrared LED stimulation was shown to obviously suppress NO and ROS production, while human DPC metabolism was not significantly affected. ¹⁰⁴ Hence, phototherapy can serve as a future clinical tool for biomodulation of inflammatory processes in pulp tissue favoring tissue repair.

Exploration of NOS/NO involved stem cell and gene therapy in regenerative endodontics

In addition to the abovementioned treatment methods, some NOS/NO-mediated therapeutic strategies for which relevant studies are lacking in dental pulp tissue may be applied to dental pulp repair and regeneration. For example, NO is believed to participate in the immunomodulation, epigenetic regulation, and multilineage differentiation of stem cells and thus has been used as a therapeutic agent for numerous diseases through its interaction with various stem cell types (e.g., neuronal stem cells in stroke neuron damage repair, endothelial progenitor cells in neural vessel repair, adipose-derived mesenchymal stem cells in calvarial defect repair, and BMSCs in multipotent tissue repair).^107,197 In addition to external NO supply from exogenous NO donors and novel NO carriers, gene-based internal NO regulation through insertion of the NOS gene into various types of stem cells or suppression of NOS gene appears to be a plausible alternative to control NO production in situ and has been tested in several types of cardiovascular diseases and neurological disorders.^197–199 However, there is no relevant research on dental pulp tissue. Thus, further work is required to understand the in vitro and in vivo effects and mechanisms of NOS gene-mediated cell therapy and to translate its benefits into clinical practice for regenerative endodontics.