The parotid and submaxillary salivary glands secrete human sialoperoxidase from birth at adult concentrations (9,10). Human sialoperoxidase occurs in different forms based on molecular weight and electrophoretic migration; sialoperoxidase may be free or linked to other salivary molecules, such as mucin (11–13). When unbound and monomeric, its molecular weight is ~78,000 (7), and its isoelectric point is 8–10 (11,14). This enzyme adsorbs different oral surfaces, while retaining its enzymatic activity (15–17), including enamel (15), dental plaque (18), salivary sediments (16) and bacteria such as streptococci (17). Other exocrine secretions, such as tears, sweat, airways or digestive secretions, also have a similar activity (usually referred to as lactoperoxidase and, at times, lacrimal peroxidase in tears). Salivary peroxidase, along with other proteins secreted by salivary glands (i.e. lysozyme and lactoferrin), participates in the antimicrobial protection of oral surfaces, such as mucosa and dental crowns.

Myeloperoxidase in saliva comes from the lysis of neutrophils entering the oral cavity by the dento-gingival sulci from the first dental eruptions (19–24). The leukocyte count in saliva decreases in edentulous patients and increases in oral inflammation processes (25). The number of neutrophils migrating into the oral cavity is ~10⁶ cells/min in healthy subjects, and higher in patients with periodontal disease (3,26,27). In the saliva, neutrophils undergo hypo-osmotic shock and release myeloperoxidase along with other antimicrobial proteins, such as lysozyme and lactoferrin; in a similar manner, the salivary glands secrete the same two proteins, in addition to sialoperoxidase. Myeloperoxidase represents up to 20–25% of the total peroxidase activity in total centrifuged resting saliva, while it predominates in the salivary sediment and in the liquid of the dento-gingival sulci, where it participates in antimicrobial tissue protection (3,23). The primary substrate for myeloperoxidase is chloride (Cl⁻) in the tissues [with the production of hypochlorite (OCl⁻)] and SCN⁻ in the saliva (with the production of OSCN⁻).

Lactoperoxidase, similarly to sialoperoxidase, catalyzes the oxidation of SCN⁻ in vivo and iodide (I⁻) in vitro, in the presence of H₂O₂. Easily purified from bovine milk, lactoperoxidase was first characterized and its enzymatic aspects were specifically documented by Aune and Thomas in 1977 and 1978 (28,29). Lactoperoxidase was commonly used to study the effects of a peroxidase-H₂O₂-SCN⁻ system on cariogenic bacteria, such as oral streptococci (30) or Lactobacillus acidophilus (31), on periodontopathogenic bacteria (32). The lactoperoxidase-SCN⁻ combination is used as a preservative for vegetables and as a supplement in several oral hygiene products. The lactoperoxidase-I⁻ combination allows for the radioactive labelling of proteins. Lactoperoxidase has the property of adsorbing different surfaces, including that of biomaterials, such as titanium (33).

Oral H₂O₂ comes from the salivary glands [DUOX family of oxidases, reviewed in (34)], neutrophils [NOX family of oxidases, reviewed in (35)] and particularly certain bacterial species present in the mouth through the action of pyruvate oxidase (36). H₂O₂ is one of the reactive oxygen species (ROS) synthesized by neutrophils during phagocytosis and secreted by the salivary glands as a trigger for sialoperoxidase activity. Determination of H₂O₂ in the saliva is challenging, given its rapid consumption by oral peroxidases and bacterial catalases. Most of the oral H₂O₂ comes from bacteria of the genus Streptococcus (37–39) and, to a much lesser extent, from optional anaerobes (40,41). Based on the equilibrium constant of the SCN⁻ peroxidation reaction (3.7×10³ M⁻¹) in the oral medium, the mean salivary H₂O₂ concentration has been estimated at 10 µM (42), the maximum dose tolerable by mammalian cells (43,44). At low concentrations (1–10 nM), H₂O₂ plays a role as a signaling molecule in bacteria-bacteria interactions (45,46). For example, H₂O₂ can promote the formation of Streptococcus parasanguinis biofilms co-cultured with Actinobacillus actinomycetemcomitans in vitro (47).

SCN⁻ is secreted by the parotid, submaxillary, sublingual and minor salivary glands (48,49) at a concentration that is dependent on exogenous intake, mainly brassica vegetables (50), as well as the smoking habits of individuals (51), and on endogenous production coupled with the detoxification of food cyanide, such as cassava (52). The salivary concentration of SCN⁻ is 0.5–2 mM in non-smokers and may be up to 10 mM in heavy smokers (12,51,53); its concentrations in the gingival crevicular fluid is ~40 µM, but may be higher in smokers (12). Of note, the optimal levels for peroxidase activity are in the order of mM. To the best of our knowledge, no study to date has documented the harmful effect of smoking on oral health through the inhibition of the natural peroxidase systems. SCN⁻ has already been proposed as a vasodilator to combat hypertension or as an antimicrobial agent in combination with a peroxidase system. In this context, various studies [reviewed in (54)] have suggest the absence of SCN⁻ toxicity up to concentrations sufficient to activate the enzymatic production of OSCN⁻. A more significant increase in plasma SCN⁻ concentration is linked to the inhibition of thyroid function, particularly in cases of iodine deficiency. Indeed, SCN⁻ is a competitive inhibitor of the Na⁺/I⁻ symport in thyroid follicular cells. This competition between a halogen and SCN⁻ for peroxidases explains the etymology of the word ‘pseudohalogen’ to qualify for SCN⁻.

The ability of the salivary glands to concentrate I⁻ into the salivary compartment (20–100 times higher than the plasma concentration) has long been known (55). This capacity for concentrating iodine in saliva before intestinal re-absorption contributes to the preservation of the iodine pool for optimal thyroid function (56). Iodine deficiency, which is marked by lower urinary and salivary rates, has been correlated with an increase in the incidence of dental caries (57), but no association between iodine deficiency and oral peroxidases has been reported in these studies. Other studies have implicated iodine deficiency in immune dysfunctions (56). Excess iodine has been reported to induce alterations of the salivary glands, including lymphocytic infiltrations (58), in an animal model. In vitro, I⁻ is another substrate for sialoperoxidase, myeloperoxidase or lactoperoxidase in antimicrobial systems. The salivary concentration of I⁻ in humans is estimated to be insufficient for antimicrobial action in vivo. Indeed, the salivary levels reported in the literature are in the order of µM (1,59,60), while those of SCN⁻ are in the mM range. The ability of SCN⁻ to inhibit the uptake of I⁻ by the parotid cells in parallel with thyroid cells has long been known (61). The presence of SCN⁻ at higher concentrations in the saliva should limit the metabolism of I⁻ (oxidation, iodination of tyrosine and iodination of proteins) (12,62,63). However, certain studies have suggested a synergy between the two substrates (64), whereas others have documented the in vitro formation of iodine-SCN⁻ complexes as a more active/stable antimicrobial component, as compared with OI⁻ and OSCN⁻ alone (65–68).

OSCN⁻ is considered to be the major oxidant produced by oral peroxidases in the oral cavity. Concentrations of OSCN⁻ up to 300 µM have been reported in the literature for the whole saliva (12). However, methodological problems when collecting samples may lead to default errors due to the instability of the molecule and the large number of oxidizable targets in saliva and oral biofilms. As SCN⁻ is metabolized by both myeloperoxidase and sialoperoxidase, the production of OSCN⁻ alone cannot be used to characterize the activity of one of these two enzymes (76). OSCN⁻ at the doses required for an antibacterial effect does not appear to cause damage to the genetic material of the mucous membranes (77). However, selective oxidation of cell/tissue targets containing thiol groups could initiate significant cell damage, such as damage to gingival epithelial cells (78).

In vitro, OI⁻ decreases the survival rate of some microbial species at a lower concentration, as compared with that observed for OSCN⁻, especially in Candida albicans (79). Some authors (80) have been able to demonstrate that ¹²⁵I only binds to a limited number of membrane molecules of Candida albicans, which is particularly sensitive to OI⁻. However, the hypohalous ion present in the oral cavity is OSCN⁻ rather than OI⁻, as the substrate SCN⁻ at the origin of the former is 1,000 times more abundant, as compared with I⁻ at the source of the latter. The salivary SCN⁻ concentration reaches the mM level, while OI⁻ barely reaches concentrations in the µM range. At equal (pseudo)halide concentrations, lactoperoxidase has a higher affinity for SCN⁻ than for I⁻. However, experimentally, the substitution of SCN⁻ by I⁻ increases the innate antiviral defences of the respiratory mucosa against adenoviruses and the respiratory syncytial virus (81). Oral administration of KI vs. NaI increases iodemia, allowing for the accumulation of I⁻ in the mucous secretions of the upper respiratory tract in humans (81) as well as animals (82,83). Although it has already been demonstrated that I⁻ accumulates in submandibular glands in hamsters (84), no similar observations have been reported in humans. It would be interesting to be able to specifically assess salivary OI⁻ in excretory ducts, such as the Stenson's duct, where the oral biofilms do not extend, except in the case of infectious parotitis. Finally, the success of topical formulations based on povidone-iodine to reduce bacterial growth in the oral cavity in ventilated patients may be attributed to the increase in I⁻ available for the patient's peroxidases (85).

The two substrates of peroxidase, SCN⁻ and I⁻, have been mostly tested independently in suspensions, and less frequently in combination (64). Moreover, few articles reporting that the activity of lactoperoxidase produces a mixture of OSCN/OI⁻ in the presence of a mixture of SCN⁻/I⁻ substrates have effectively analyzed the products generated under their experimental conditions (1,64,86). Mixtures of the two substrates, SCN⁻ and I⁻, however, appear to lead to the formation of a more stable oxidative complex over time (65,66,87–90). Under well-defined in vitro experimental conditions (high ionic strength; KI:KSCN ratio, ~4.5; neutral or acidic buffer; presence of H₂O₂), lactoperoxidase produces stable iodine-SCN⁻ complexes, such as I₂(SCN)⁻, objectified by carbon-13 (C13) nuclear magnetic resonance spectroscopy (66) with antimicrobial properties against Candida (67). Oxidant concentrations are then less toxic to epithelial cells in the mouth, as compared with chlorhexidine (67). The resistance of epithelial cells may be explained by the presence of bacteria covering their surface and forming a protective biofilm, and/or by the detoxification of enzymatic activities in certain commensal species present in the oral environment (91). The use of these complexes for rinsing dentures ex vivo further limits the possible effect on epithelial cells in vivo. The efficacy of the iodine-SCN⁻ complexes prepared without peroxidase activity has already been demonstrated in vitro on Pseudomonas aeruginosa and Staphylococcus aureus, both in suspension and in biofilms (92). Other studies will have to analyze the interference between this solution of iodine-SCN⁻ complexes and oral microbial ecology.

Oral peroxidases control the oral microbiota while detoxifying the environment of H₂O₂. By oxidizing thiol groups (29) and nicotinamide adenine dinucleotide phosphate (NAD(P)H) (94), OSCN⁻ exerts antimicrobial, antimycotic and antiviral effects (1,2). In particular, the bacterial hexokinase activity, which initiates the Emden-Meyerhoff pathway, decreases in the presence of OSCN⁻. Acid production by dental plaque is reduced in the presence of OSCN⁻ (95). In vitro, OSCN⁻ slows the growth of cariogenic bacteria belonging to the genera Streptococcus or Lactobacillus (1). Periodontopathogenic bacteria are also inhibited (32,96–98), as are yeasts of the Candida genus, which are responsible for mucosal pathologies (99). Some viruses are sensitive to this oxidant, such as the herpes virus (100), HIV (101) or influenza virus (102,103). Antisepsis research suggests using peroxidase-generated OSCN⁻ as a putative prophylactic agent to prevent contamination with SARS-CoV-2 (104).

Some bacteria have developed various strategies to resist the ROS produced by neutrophils (105). Similarly, commensal oral surface bacteria, such as Streptococcus sanguis (one of the first colonizers of the dental plaque) have the biochemical equipment to protect themselves from OSCN⁻. By contrast, Streptococcus mutans (a cariogenic bacterial species) does not have this detoxifying system (91). Thus, peroxidases operate more like ecological selectors by allowing commensal bacteria capable of reducing OSCN⁻ to survive in the oral environment. Moreover, peroxidases coupled to a system detoxifying the environment of hypohalous compounds (of the thioredoxin type) participate in the protection of tissues against oxidants resulting from the metabolism of activated oxygen, such as the superoxide anions or H₂O₂ (106). There appears to be an interconnection between resistance to oxidative stress and the ability to form biofilms via quorum-sensing molecules, such as farnesol (107). On the other hand, non-active lactoperoxidase, after the depletion of the substrate, can promote the in vitro growth of certain anaerobes (98). At the same time, a clinical study (108) suggested a rebound in periodontitis after a phase of clinical improvement, possibly due to an accumulation of lactoperoxidase without substrate. The association of a peroxidase system with another innate non-immune defence factor further strengthens the inhibition of some bacteria in vitro. For example, lysozyme improves the inhibition of glucose metabolism by the peroxidase system in Streptococcus mutans (109).

Sialoperoxidase, similar to lactoperoxidase in vitro, adsorbs enamel surfaces irreversibly (15) while retaining its activity. This property contributes to controlling the formation of dental plaque and regulating the bacterial microflora attached to the surfaces of the oral cavity. The activity of peroxidases present in dental plaque depends on the availability of its two substrates (SCN⁻ and H₂O₂) in the thickness of oral biofilms. SCN⁻ diffuses deeper into the biofilms from the salivary film that covers them. At the same time, H₂O₂ relies on the oxygen supply allowing certain bacteria to produce it in situ. Certain studies (12,106) have described a virtuous cycle summarizing the beneficial effect of sialoperoxidase on oral microflora control. After a supply of food sugars, commensal bacteria of the genus Streptococcus grow and produce more H₂O_2, which activates the production of OSCN⁻ in the presence of sialoperoxidase and SCN⁻, thereby limiting bacterial growth or even killing microorganisms. Theoretically, these mechanisms work less effectively deeper inside thick biofilms, given the consumption of peroxidase substrates in the most superficial layers. Similarly, mouth rinsing with solutions containing an OSCN⁻-generating system is likely to further impact the more superficial layers of biofilms. These remarks suggested that a peroxidase system may act both on the formation of biofilms and on already formed biofilms.

In addition to protecting the walls of the oral cavity, sialoperoxidase may also protect the excretory ducts of the salivary glands, which are rarely infected when they open into an oral cavity abundantly colonized by microbial germs; however, to the best of our knowledge, no scientific study has yet considered this physiological aspect.

The uncontrolled production of activated oxygen species (ROS: Superoxides, H₂O₂, free radicals, hypohalous compounds) or reactive nitrogen species (RNS: Compounds derived from nitric oxide and superoxides, such as peroxynitrite, nitrogen dioxide and dinitrogen trioxide) is implicated in various oral pathologies (110). These oxidative derivatives damage cells and are considered to be mutagenic and carcinogenic. They are controlled in the salivary environment by scavengers (mainly uric acid) and antioxidant enzymes (bacterial catalase and oral peroxidases). Lactoperoxidase transforms cytotoxic H₂O₂ into a less toxic product, as suggested by in vitro experiments on fibroblast cultures (44). The coupling of peroxidases to a bacterial thioredoxin system helps protect tissues from oxidants produced from peroxidase activities. In tissues (including gingival), catalase and glutathione peroxidase help remove H₂O₂.

Certain studies [reviewed in (12)] considered that the enzymatic peroxidation of mutagenic and carcinogenic molecules bound to diet as another function of oral peroxidases. However, the related data are currently insufficient to assess its actual significance in the detoxification of toxic molecules introduced into the oral cavity, through food or other means.

The use of peroxidase systems may be considered as topical administration to protect oral surfaces in vivo as a prophylactic measure. Recommending these galenic preparations may not be of value as a curative approach, as the infection itself indicates that a deregulated oral flora has overtaken physiological systems, including peroxidase. Reserving or even developing these galenic preparations as prophylactic treatment would avoid the appearance of strains resistant to antimicrobials used continuously to prevent the occurrence of certain infections (for example, oral candidiasis in patients with AIDS).

Peroxidase systems present a paradox: Although they have been shown to be active in vitro for a long time, the definitive evidence of their effects in vivo is still lacking. A possible explanation is the actual availability of H₂O₂ in the oral environment, which is rich in bacterial catalase. In this case, the increased availability of H₂O₂ can compensate for its consumption by other enzymatic activities (141,142), thus allowing the exogenous lactoperoxidase to be active. Another explanation would be to consider a transient metabolic suppression demonstrated, for example, by assaying ATP, but not observed on culture media.

Peroxidases adsorb different biomaterials used in dentistry: Titanium and resin are two examples. Ex vivo use may be employed to disinfect removable prostheses by short-circuiting in vivo use; the intra-oral administration is indeed considered by some to be dangerous, due to cytotoxicity and potential immunogenicity. The toxicity of peroxidase systems on the oral mucosa appears to be limited by the oral biofilms covering them. However, the accumulation of inactive lactoperoxidase in biofilms may favor certain anaerobic bacteria, such as Porphyromonas gingivalis (98). On the other hand, the accumulation of I⁻ during regular use would not be advised for some patients. With regards to health and safety, it should be noted that the FAO recommends the SCN⁻/lactoperoxidase system for the decontamination of milk and vegetables for food sales (113).

Few reports have analyzed the oral toxicity of peroxidase systems when recommended in oral care products. Indirectly, these products may be beneficial for patients who have aphthous ulcers secondary to the toxicity of lauryl sulfate, often used as a detergent in toothpaste (143). Indeed, the incorporation of such enzymes in oral hygiene products implies the no-use of detergent foaming agents, which have the ability to denature proteins, and therefore enzymes.

Numerous investigations have analyzed the effects of peroxidase systems on the various pathogens present in the oral cavity, including viruses. In vivo, oral peroxidases defend the gingival sulcus against microbial invasion (myeloperoxidase) and help regulate the microflora adhering to the surfaces of the oral cavity (sialoperoxidase). Incorporating lactoperoxidase with other salivary antimicrobial proteins in salivary substitutes may mimic the antimicrobial effect of the salivary exocrine fluid (144). The beneficial effect of oral sprays with such peroxidase preparations in combating dryness of the oral mucosa in patients on respirator treatment has already been proposed. However, their benefit in daily oral hygiene has not yet been quantified or even demonstrated by clinical studies. Certain studies have suggested considering peroxidases as ecological selectors directing the oral microflora towards a Gram-positive cocci microflora poor in cariogenic or periodontopathogenic germs, thereby preserving the mucosa and protecting it from yeast overgrowth, among others. Thus, the administration of an efficient lactoperoxidase system in a toothpaste also containing lactoferrin and lysozyme may prevent oral dysbiosis (145). A few studies have demonstrated an inverse association between the activity of sialoperoxidase and the depth of the periodontal pockets (146). On the other hand, the administration of inactive peroxidase without substrate seems to disturb in vitro the balance between commensals and periodontopathogens to the latter's advantage (98,147). The peroxidase systems' clinical benefit in oral hygiene must still be demonstrated on a large scale through evidence-based practice.