Early childhood caries (ECC), formerly referred to as nursing bottle tooth decay, remains a significant chronic childhood infectious disease in developing as well as industrialized countries (1). It is defined as “the presence of one or more decayed, missing (due to caries), or filled tooth surfaces in primary dentition in children under the age of six” (1). If a child younger than 3 years of age has any sign of smooth-surface caries; a child from ages three through five has one or more cavitated, missing, or filled smooth surfaces in primary maxillary anterior teeth; or a child from ages three through five has a decayed, missing or filled score larger than the child's age, the child is considered to have severe ECC (S-ECC) (1). ECC is a major public health problem, causing tooth pain and loss, masticatory dysfunction, poor nutritional status, disrupted growth and development, as well as impaired learning ability in children. It also constitutes a substantial economic burden, particularly for individuals from low socioeconomic backgrounds due to high treatment costs, emergency room visits, and hospitalizations (1, 2).

The oral microbiome plays a critical role in the etiology of dental caries (2). Streptococcus mutans and Lactobacillus species were consistently associated with the initiation and progression of ECC. However, latest development in molecular microbiology provide novel insights into the complexity of the oral microbiome and the association of other microbial species with the risk of developing ECC. In the last two decades, increasing evidence indicates that Candida, the only genus of fungi demonstrated to reach a significant biomass in the oral cavity, thrives in lower oral pH and ECC (3). At the species level, many studies have shown that the prevalence and abundance of Candida albicans is positively associated with the incidence and severity of ECC, suggesting that C. albicans may also constitute a major etiological risk of ECC (2, 4). The present review aimed to provide an update of the current knowledge on the association of C. albicans with ECC, its individual virulence and synergistic interactions with S. mutans. Recent advances in the development of approaches to disrupt C. albicans-S. mutans cross-kingdom biofilms in ECC prevention and treatment were also summarized.

C. albicans is a polymorphic yeast from the Candidaceae family and is a normal resident of the skin, vaginal tract, oral cavity, and gastrointestinal tract in most healthy humans (5). However, conversion of this microorganism into an opportunistic pathogen occurs when there is dysbiosis in the microbial community or alterations in host immunity, and as such, is considered a pathobiont. Multiple host and environmental factors may affect the growth of C. albicans in the oral cavity, such as host age, diet, geographic location, socioeconomic status, gender, immunosuppression, and antibiotic use (4). An overgrowth of C. albicans in the oral cavity usually causes mucosal infection and leads to oral candidiasis or thrush. Virulence factors involved in the pathogenesis of C. albicans infection include biofilm formation, evasion of host innate immunity, yeast to hyphae transition, and production of candidalysin (6).

Although commonly present in the oral cavity, it was reported previously that C. albicans could not colonize to the tooth surfaces effectively on its own, instead, it mainly adheres to oral mucosa and acrylic surfaces, causing mucosal infection (7). However, others showed that C. albicans is capable of adhering to enamel and dentine effectively, and that dental plaque biofilms from children with ECC frequently contain C. albicans (8). These findings have led to the postulation that C. albicans is involved in caries development. Although some pilot clinical studies with limited subjects did not observe significant differences in the prevalence of oral C. albicans between children with or without dental caries (9, 10), multiple lines of evidence have shown that the presence of C. albicans in saliva and dental plaque is strongly associated with dental caries, especially ECC. In this regard, a number of cross-sectional studies have revealed that the prevalence of oral C. albicans in ECC children is significantly higher than that in caries-free children (4, 11–14). In addition, several studies indicate that both the prevalence and carriage of oral C. albicans is positively correlated with the severity of ECC (12, 15–17). Furthermore, a higher C. albicans detection rate was noted in plaque samples collected close to carious lesions than those collected from sound tooth surfaces (18). Moreover, a 2018 systematic review based on fifteen cross-sectional studies (4) indicated that children with oral C. albicans have over five times greater risk of developing ECC than those children without oral C. albicans carriage, reinforcing the belief that oral C. albicans may be considered as a potential risk factor for ECC. In line with these reports, it was found that children with oral thrush in the first year of life are three times more prone to have ECC (19). Interestingly, on the other hand, it has also been suggested that decayed teeth that harbor C. albicans may act as a reservoir for recurrent oral and non-oral candidiasis (4), highlighting a bidirectional relationship between soft and hard tissue oral Candida infection.

Maternal transmission has been considered one of the factors that contribute to early detection of Candida spp. in children. Studies have shown that oral colonization of C. albicans in infants was positively correlated with mother's oral C. albicans carriage, and that more than 60% of children with S-ECC carry the same C. albicans strains as their mothers (15, 20). In addition, the mothers of S-ECC children have a significantly higher oral C. albicans detection rate than the mothers of caries-free children, suggesting that maternal carriage of oral C. albicans might be a risk prediction for children's C. albicans carriage and ECC development.

Based on the amplified DNA band lengths determined with a specific PCR primer designed to span a transposable intron region in the 25S rRNA gene, previous studies were able to classify clinical C. albicans strains into four to five genotypes, including genotypes A, B, C, D, and E (21, 22). It was found that genotypes A, B, and C are typically detected in the dental plaque of children, genotype D is found in the periodontal pocket of patients with periodontitis, while genotype E is rarely present in the oral cavity (13). Others found that genotype A is mostly predominant in plaque biofilms of children with S-ECC, followed by genotypes C and B (11). In addition, genotype A was predominantly associated with proximal lesions, whereas genotypes B and C were mainly isolated from occlusal cavities.

Interestingly, by comparing pooled plaque mycobiome of 40 children with ECC with the same number of caries-free children, a recent study showed that it was C. dubliniensis, but not C. albicans, that dominated the mycobiome of children with caries (23). Using a site-specific analysis of the mycobiome associated with ECC, another study showed a trend for decreased mycobiome diversity as caries severity increased and found that both C. dubliniensis and C. albicans were positively correlated with ECC, with C. albicans being only associated with severe disease, while C. dubliniensis increasing steadily as caries severity increased (24). C. dubliniensis is closed related to C. albicans but is less pathogenic in mucosal disease models (3). Further studies are needed to determine if C. dubliniensis plays a contributory role in caries development (3).

Ecological theory of dental caries states that caries lesion-associated biofilm have high amounts of acidogenic and aciduric microorganisms. S. mutans, Lactobacilli and Candida spp. are believed to be the major acid producing organisms associated with ECC. Although analysis of the number and biomass of microorganisms associated with dental caries revealed that S. mutans dominates in number followed by Lactobacilli and C. albicans, the biomass of C. albicans is much larger than that of Lactobacilli and S. mutans (25–27). Clinical studies have shown that the presence of C. albicans in the oral cavity is associated with a highly acidogenic and aciduric bacterial community in S-ECC (11, 20). Despite being lower in abundance, C. albicans dissolved hydroxyapatite crystals at a 20-fold faster rate than S. mutans (26). Others have noted that when the biofilm pH fell below 5.5, acidification by S. mutans dropped substantially and stopped at pH around 4.2, whereas Lactobacilli and C. albicans continued to acidify even at pH 4 (11, 25). Additionally, the main organic acid produced by C. albicans is pyruvic acid, and pyruvic acid is more potent than lactic acid in decreasing the pH of an already intensely acidified environment (25). In terms of aciduricity, C. albicans is highly aciduric and can survive at a pH 4 or even 3 in the presence of glucose and sucrose (8, 11). The acid tolerance ability of C. albicans has been postulated to favor its high frequency in dentinal caries, the highly acidic part of a decayed tooth (25). However, others have reported that C. albicans does not invade carious human dentine (28). Therefore, the role of C. albicans in dentinal caries pathologies need to be further investigated.

Interestingly, different results have been reported related to the acidogenicity and aciduricity of C. albicans isolated from caries-free and caries-active children. It was reported that Candida is more acidogenic and aciduric in nature when isolated from children with ECC than those isolated from children without caries (11). In contrast, others showed that the yeast isolated from children without caries caused significant higher percentage of demineralization in vitro in the presence of sucrose, compared with those isolated from children with caries (29). It was speculated that C. albicans from children without caries may metabolize sucrose faster to adapt to a sucrose-rich environment, whereas C. albicans from individuals with caries should already accustom to an environment rich in sucrose. In addition, it was also reported that biofilms originating from children with and without caries have similar cariogenicity when subjected to the same cariogenic challenge (30).

Unlike S. mutans, Candida does not metabolize sucrose efficiently due to the lack of invertase activity, and grows at much higher rate when cultures in glucose or fructose than in sucrose (31). Glucose is an essential factor for Candida to produce high amount of acid. In a glucose limited environment, C. albicans is comparatively less acidogenic than Lactobacilli (25). In addition, it was suggested that carious activity of C. albicans may depend not only on the presence of fermentable carbohydrates, but also on the proportion of sucrose and glucose in the diet (32).

In addition to the acidogenicity and aciduricity, C. albicans produces multiple proteolytic enzymes such as proteases, hemolysins, phospholipases, collagenases (8, 33). These extracellular enzymes are particularly active in an acidic milieu, and may play a significant role in dentinal caries progression via destroying dentinal collagen, or assisting the yeast to penetrate deep into dentine through dentinal tubules (8). It has been shown that the activity of aspartyl proteinases (Saps) in dental plaque biofilm in the S-ECC group were significantly higher than those in caries-free group (33, 34). More studies are necessary to understand whether C. albicans from caries-free and caries-active children have identical or different virulence.

A biofilm is a an assemblage of surface-associated microbial cells that is enclosed in a self-produced extracellular polymeric matrix (35). Microorganisms living within a biofilm have increased resistance to environmental stress, such as antibiotics and the host immune response. Candida mostly exists in a polymicrobial environment, and such a heterogeneous biofilm population is a crucial and clinically important element for the growth, proliferation and survival of C. albicans. It also enhances bacterial colonization and biofilm formation by interacting with environmental and host factors, which enhances biofilm virulence and drug protection/resistance (36). Recent studies suggest that C. albicans may act as an essential “keystone” component in oral biofilms (37).

S. mutans is one of the most predominant microbial pathogens associated with ECC. C. albicans is frequently co-isolated with S. mutans from the dental plaque biofilms. Studies have demonstrated that the abundance of salivary S. mutans in infants positively correlates with infants' C. albicans levels (38). In addition, the emergence of S. mutans is much higher in infants who have early colonization of oral Candida compared to those who are free of oral Candida (38). Similarly, the adherence of S. mutans to oral biofilm and tooth surfaces also increases with the prevalence of Candida species (38).

The association of S. mutans and C. albicans may result in highly cariogenic biofilms that are readily associated with ECC. Yeast, without co-existence with S. mutans showed weak ability to colonize smooth surfaces of teeth, and the average number of DMFT/DMFS in individuals with co-existence of S. mutans and C. albicans were higher than those infected with S. mutans alone (15). In addition, the total count of C. albicans and S. mutans in the supragingival dental plaque of children with ECC increases with an increase in the percentage of active carious lesions and the severity of dental caries (39). Furthermore, co-existence of C. albicans and S. mutans in saliva and dental plaque is also strongly associated with caries recurrence in children (40).

In vitro studies have illustrated that the C. albicans-S. mutans co-species biofilms present greater 3D complexity, and are more resistant to stress conditions (41). The co-species biofilms not only exhibit protection against antibacterial drugs, but also protect yeast cells from antifungal agents (42). Saliva contains a wide variety of antimicrobial peptides (AMPs) that resists S. mutans. Thus, in the presence of purified saliva, S. mutans fail to form mature biofilms (43). However, the co-species biofilms showed rapid maturation, maintained acidogenicity in saliva, and caused severe enamel demineralization in vitro, while S. mutans single species biofilms displayed poor development, failed to create an acidic environment, and caused minimal damage to the enamel surface (43, 44). In addition, the co-species biofilms also have a positive effect on the survival of C. albicans, increasing its biomass, thereby increasing the overall biofilm formation. However, it is necessary to highlight that the synergistic interactions between C. albicans and S. mutans may be variable and tend to be dictated by environmental conditions and population density (45). Furthermore, other species of oral microbiota may also influence the synergistic interactions between C. albicans and S. mutans (46, 47).

Animal studies further showed that co-infection with S. mutans and C. albicans can cause more severe and extensive caries in rats exposed to a cariogenic diet than the rats infection either species alone (44). Others have shown that exposure to C. albicans alone significantly increases the advanced fissure lesions in rats than non-infected controls in the presence of 40% of sucrose or 40% glucose; however, co-infection with C. albicans and S. mutans did not increase the incidence of occlusal caries (32). Although these animal studies suggest that C. albicans could have a pathological role in caries development, the potential effect of indigenous microbiome members, either enriched or depleted by a cariogenic diet in dental plaque of the animal models, has not been considered (3). Recent studies have shown that a high-sucrose diet is associated with a significant reduction in indigenous enterococci in a murine candidiasis model (48).

In addition to S. mutans, studies have shown that C. albicans may also synergize with Actinomyces to increase the biomass and cariogenic virulence of the C. albicans-Actinomyces dual species biofilm (49). However, decreased levels of salivary/plaque Actinomyces were identified in children with S-ECC with increased abundance of C. albicans (20). How C. albicans influences the composition and diversity of oral biofilm and the role of multiple species biofilms on C. albicans-S. mutans interactions and caries development require further investigation.

Various mechanisms are proposed to be associated with the synergistic effect between S. mutans and C. albicans in the pathogenesis of dental caries. One critical mechanism is the co-adhesion of S. mutans and C. albicans to tooth surfaces via glucan synthesis. One of the key virulence factor of S. mutans in dental caries is its ability to convert sucrose into a wide range of soluble and, especially, insoluble extracellular polysaccharides (EPS), namely glucan (7). The formation of glucan is catalyzed by the exoenzymes glucosyltransferases (Gtfs). EPS creates an acidic environment due to its diffusion-limiting barrier, and forms the foundation for dental biofilms. EPS and Gtfs have a pivotal role in the S. mutans-C. albicans partnership within dental plaque biofilms (36, 50). Microscopic in situ analysis of intact clinical biofilm samples from subjects with caries provides further evidence that the interactions between the bacteria and yeast is mediated by extracellular EPS (51). C. albicans has multiple Gtf binding sites on the cell wall. Mannans on the C. albicans surface have been identified as key binding sites for GtfB, and the presence of S. mutans significantly upregulates genes associated with C. albicans mannan production (52). Therefore, C. albicans utilizes Gtf to adhere to tooth surfaces indirectly via its interaction with S. mutans. In addition, C. albicans also upregulates gtfB gene expression by S. mutans, leading to an increase in EPS production (53). This matrix forms a scaffold for microbial adhesion and acts as a gradient by modulating chemical and nutrition diffusion. Additionally, the matrix blocks access of saliva to the interior of the biofilm and prevents acid within biofilm from diffusing outward, thereby resulting in acidification and demineralization of the teeth (7). Moreover, as yeast is highly acidogenic, acid production is even more enhanced in this cospecies biofilm.

Another mechanism involved in the synergistic interactions between C. albicans and S. mutans is metabolic communication/cross-feeding, wherein one organism uses metabolites secreted by another for nutrition. As described above, C. albicans significantly enhances S. mutans carbohydrate utilization and glucan biosynthesis in the mixed biofilm by upregulating Gtf expression (31). This modification is critical for S. mutans survival and proliferation. On the other hand, S. mutans can metabolize sucrose to release free glucose and fructose, allowing C. albicans to utilize the monosaccharide efficiently in the mixed biofilm. This cross-feeding from S. mutans therefore compensates the inefficiency of C. albicans to utilize sucrose, leading to an enhanced fungal growth and acid production under cariogenic conditions (31). The reduced environmental pH in turn favors S. mutans survival. In addition, lactate is one of the major metabolites detected in the C. albicans and S. mutans co-cultures, and it favors the growth of C. albicans by acting as a carbon donor thereby reducing oxygen tension (50). The reduced level of oxygen tension, in turn, favors the growth of S. mutans. Thus, C. albicans and S. mutans benefit mutually from a symbiotic bacterial-fungal sugar metabolism, leading to an enhanced biofilm virulence under cariogenic conditions.

Quorum sensing is also important in S. mutans-C. albicans interactions within the co-species biofilm (53, 54). C. albicans-derived farnesol, a quorum-sensing molecule, has a main role in the yeast morphological switching by inhibiting hyphae production (55). Farnesol and its derivatives/analogs usually exhibit anti-biofilm, anti-bacterial and fungicidal activity (55). However, farnesol at concentrations of 25–50 μM was found to enhance Gtf activity and S. mutans microcolony development (53). It was only when the farnesol production reached a higher concentration (>100 μM), inhibition of the growth of S. mutans occurred. The presence of S. mutans tends to control the level of farnesol production by C. albicans, which may contribute to hyphal formation typically seen in the co-species biofilm. In addition, farnesol can incorporate into S. mutans cell membrane due to its fatty acid-like structure (53, 56). These observations suggest that farnesol is an important modulator in resolving the potential antagonism between S. mutans and C. albicans, and that a well-controlled mechanism exists between S. mutans and C. albicans to maintain farnesol at levels that promote the symbiotic relationship in the cross-kingdom biofilm.

In addition to the above-mentioned mechanisms, in vitro studies found that S. mutans-derived AgI/II is also important for incorporating C. albicans into the two-species biofilms and increasing acid production (57). Therefore, with these highly complex symbiotic communications between S. mutans and C. albicans, it is plausible that this cross-kingdom biofilm has an enhanced virulence under cariogenic conditions, resulting in increased risk of ECC. More studies are needed to better understand the molecular mechanisms that govern the cross-kingdom interactions between C. albicans and S. mutans in ECC, which may help define novel approaches to prevent and treat ECC.

Due to the association of the C. albicans-S. mutans biofilm with ECC, various approaches have been attempted to disrupt this cross-kingdom interactions. Numerous in vitro studies have tested the efficacy of using antimicrobial agents, especially plant extracts or plant-derived compounds to inhibit the cross-kingdom biofilm formation and virulence. Curcumin, a food-grade natural product extracted from the root of turmeric, was found to be able to downregulate Gtf and quorum sensing-related gene expression of S. mutans, reduce EPS production, and decrease biofilm biomass and viability in C. albicans-S. mutans dual-species biofilms (58). The expression of genes related to C. albicans adhesion and aggregation in biofilm, the agglutinin-like sequence (Als) family members als1 and als2, was also suppressed after curcumin treatment. Polyphenol extracts from green tea or cranberries have also been shown to effectively inhibit the acidogenicity and metabolic activity of C. albicans-S. mutans biofilms, decrease EPS and microbial biovolumes, and disrupt biofilm structure (59, 60). Thymol, the major constituent of thyme essential oil, is another nature product that showed ability to diminish the C. albicans-S. mutans dual-species biofilm formation and virulence (61). In addition, thymol was found to be effective in diminishing C. albicans-S. mutans dual-species virulence in the invertebrate model Galleria mellonella (61). Other agents, such as chitosan nanoparticles (62), gymnemic acids (63), Rhamanus prinoides (gesho) stem extracts (64) and candy derived from Melaleuca cajuputi essential oil (65), have also been shown to be effective in the inhibition of C. albicans-S. mutans dual-species biofilm formation. However, the efficacy of these antimicrobial agents in inhibiting cariogenic biofilms and preventing ECC in vivo needs to be further verified.

Lactobacillus spp. are part of human microbiome and are natural competitors of Candida in the vaginal environment (66). This concept has prompted studies to probe the effect of Lactobacilli in preventing or treating ECC by targeting the C. albicans-S. mutans biofilm. Studies have shown that L. salivarius can inhibit the dual-species biofilm formed in vitro with clinical isolates from dental plaque of children with ECC (47). The addition of L. salivarius decreased biofilm biomass, S. mutans and C. albicans abundance, and fungal morphological transformation. It was also found that L. plantarum 108 culture supernatants can downregulate the expression of S. mutans Gtf genes, C. albicans hyphal specific genes, inhibited biofilm formation, and reduced the pre-formed biofilm in vitro (67). In addition, L. plantarum CCFM8724 was found to be capable of preventing and treating C. albicans-S. mutans-induced caries in a rat model, and it exhibited a better inhibitory effect than 0.02% chlorhexidine (68). A number of clinical trials have also shown the beneficial effect of using probiotic bacteria in caries reduction (69). These probiotic strains typically do not colonize the oral cavity permanently, either following early-in-life interventions or in individuals with a mature microbiota. In contrast, other clinical studies have shown a positive correlation between C. albicans and Lactobacillus spp. in ECC. Yeast and Lactobacilli have been co-isolated from advanced caries lesions, and the presence of C. albicans in dental plaque is associated with increased abundance of several Lactobacillus spp., as well as the risk of ECC (20). In addition, L. casei was shown to stimulate C. albicans hyphal growth in vitro, which in turn supported the coaggregation of Lactobacilli and biofilm development (70). Therefore, it is possible that the interactions between Lactobacilli and Candida may depend on the environment they cohabit. In addition, a strain-specific relationship may exist between Lactobacilli and Candida.

Another approach proposed is to use oral commensal Streptococci to disrupt the S. mutans and C. albicans synergy in biofilms. S. parasanguinis, a mitis group Streptococcus and one of the most abundant commensals in the oral cavity, inhibits the growth of S. mutans in vitro (71). This commensal Streptococcus can also disrupt S. mutans-C. albicans synergy in the three species biofilms in vitro by altering the global metabolic signature, impairing S. mutans GTF activity, and blocking C. albicans from binding glucan (46). However, it was shown that S. parasanguinis is overrepresented in the saliva from children experiencing recurring decay within 6 months after being treated for ECC (40). Others have also observed more abundant S. parasanguinis present in dental plaque from caries-active children compared to caries-free children (72). Further studies are needed to elucidate the role of S. parasanguinis in the pathogenesis of ECC.

Most recently, a binding mechanism-based non-microbicidal approach using mannan-degrading enzymes was evaluated for the possibility of disrupting S. mutans-C. albicans cross-kingdom biofilm interactions (73). The enzymes were found to efficiently degrade mannans on C. albicans cell wall surfaces, therefore impairing S. mutans Gtf-to-mannan binding mechanism in the biofilm, leading to weakened biofilm mass, population, mechanical stability, acidity, and significantly decreased human tooth enamel demineralization in vitro. No microbicidal effect and cytotoxic effect on gingival keratinocytes were noted, and the enzymes were stable in human saliva, suggesting a potential application of this targeting intervention on C. albicans-S. mutans interkingdom ligand-receptor binding interactions in preventing and/or treating ECC. Further understanding of the proteins/structures involved in the interkingdom biofilm interactions may reveal more potential targets against the pathogenic biofilm in ECC.

Additionally, the C. albicans-S. mutans biofilm builds up tenaciously on various dental restorative materials with higher numbers of C. albicans being found on rough restorative surfaces. In addition, different components in different restorative materials may have different effects on biofilm formation. The presence of heavy metals in restorative materials exhibits antimicrobial properties, for example, amalgam tends to liberate ions that interfere with the microorganism's ability to adhere to the tooth surfaces and form biofilm in vitro and in vivo (74, 75). Similarly, in vitro studies showed that glass ionomer cements reduced biofilm formation by these organisms owing to their fluoride releasing property (75). Furthermore, biointeractive dental materials incorporated with antimicrobial and ion releasing/recharging formulations have been explored for the efficacy in interfering C. albicans-S. mutans biofilm formation. Incorporation of dimethylaminohexadecyl methacrylate (DMAHDM) and amorphous calcium phosphate nanoparticles (NACPs) into dental sealants showed promising effect on inhibiting C. albicans-S. mutans dual-species biofilm formation and preventing enamel mineral loss in vitro (76). Thus, using these selective materials may have beneficial effect in inhibiting C. albicans-S. mutans biofilm formation. More in vivo studies including animal studies, microscopic in situ evaluation of the biomass and metabolic state of intact clinical biofilm samples from subjects with ECC, and randomized clinical trials, are necessary to further explore and verify the effectiveness of different approaches in inhibiting C. albicans-S. mutans biofilm formation and in controlling ECC.

There is increasingly evident from cross-sectional studies that the presence and accumulation of C. albicans in dental plaque biofilms is associated with the incidence and the severity of ECC, indicating that detection of an enriched level of C. albicans in saliva and dental plaques may serve as an indicator of the risk of ECC. Although animal studies have suggested that C. albicans may have a pathogenic role in caries development, it's still a matter of debate whether the presence and accumulation of C. albicans in ECC is merely a consequence of the adaptation of C. albicans to a cariogenic oral environment, or it plays an active role in the initiation and progression of dental caries. Longitudinal clinical studies, including interventional clinical trials targeting C. albicans, are necessary to ascertain if C. albicans indeed contributes in a significant manner to the initiation and progression of ECC. In addition, further work is needed to understand the influence of other bacteria and fungi of oral microbiota on C. albicans-S. mutans interactions in ECC.

The idea of this manuscript was conceived by PZ. LM wrote the manuscript. JS, JJ, and PZ critically reviewed the manuscript. All authors have read and approved the final manuscript.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.