The effectiveness of probiotics in the oral cavity was assessed when exposed to various enzymes, hydrogen peroxide, and saliva. The ability of probiotics to attach to human gingival fibroblast cells (HGFCs) and maxillary teeth was also analyzed. In the following section, the effect of each strain of Bifidobacterium and Lactobacillus spp., a combination of five Lactobacillus strains, a combination of five Bifidobacterium strains, and a combination of the most effective Lactobacillus and Bifidobacterium strains were analyzed on the expression of the main virulence genes in A. actinomycetemcomitans, P. gingivalis, F. nucleatum, and S. mutans. Analysis of the gene expression of IL-8 and IL-10 cytokines was also conducted on the supernatant of HGFC. Except for the competition and biofilm formation assays, which were carried out in duplicate, all other experiments were performed in triplicate.

Five Lactobacillus strains were isolated from the stool and five Bifidobacterium strains were taken from breast milk and infant feces. These strains were chosen for investigation in the current study, as outlined in Table 1. The study included Lactiplantibacillus plantarum (L. plantarum 42 and L. plantarum 156), Levilactobacillus brevis (L. brevis 205), Limosilactobacillus reuteri (L. reuteri 100), and Lacticaseibacillus rhamnosus (L. rhamnosus 195) (Rohani et al., 2015). Additionally, Bifidobacterium bifidum (Bifidobacterium bifidum 1001 and Bifidobacterium bifidum 1005), Bifidobacterium breve (Bifidobacterium breve 1015 and Bifidobacterium breve 1063), and Bifidobacterium longum (Bifidobacterium longum 1044) were also studied (Eshaghi et al., 2017). The current study involved the use of four pathogens, including A. actinomycetemcomitans JP2 genotype strain HK1651 and S. mutans purchased from the microbial collection bank of Shahid Beheshti University of Medical Sciences, faculty of dentistry, as well as P. gingivalis ATCC 33277 and F. nucleatum ATCC 27725 from Tehran University of Medical Sciences (kindly donated by Dr. Douraghi).

Probiotics were grown on Man-Rogosa-Sharpe (MRS) agar (Merck, Darmstadt, Germany) and incubated at 37°C for 48–72 h under microaerophilic conditions. A. actinomycetemcomitans and P. gingivalis were cultivated on brain heart infusion (BHI) agar (Merck, Darmstadt, Germany) supplemented with 0.5% defibrinated sheep blood and incubated at 37°C for 48–72 h under anaerobic conditions. F. nucleatum was grown on Brucella agar based medium (Condalab, Spain) supplemented with 5% horse serum (Baharafshan, Iran) and 10% defibrinated sheep blood (Baharafshan, Iran) and incubated at 37°C for 48–72 h in anaerobic conditions. S. mutans was cultured on blood agar (Merck, Darmstadt, Germany) and incubated at 37°C for 24 h under aerobic conditions. In all experiments, to achieve an optical density (OD) of 0.08–0.13 equivalent to 1.5 × 10⁸ colony forming units (CFUs)/mL, bacterial suspensions were first prepared by transferring fresh colonies of probiotics and pathogens grown on agar plates to MRS and BHI broth, respectively, and then incubated at 37°C under microaerophilic conditions.

Intact maxillary molars were provided by healthy women who have been replaced with gold teeth for cosmetic reasons. In order to investigate the adherence ability of probiotics to maxillary molars, suspensions of probiotics at 1.5 × 10⁸ CFU/mL were prepared. Then, 1.5 mL of the prepared suspension was placed into a 2-mL sterile Eppendorf tube containing a sterilized maxillary molar sample, and incubated at 37°C for 48 h under microaerophilic conditions. After incubation, the maxillary molar was rinsed three times with either normal saline or PBS to eliminate unattached bacteria. The maxillary molar was gently washed with PBS and transferred into a 2-mL sterile Eppendorf tube containing 1.5 mL of PBS. It could be either sonicated for 15 s at a frequency of 30 kHz and an output power of 7 W or centrifuged at 5,000 rpm and 4°C for 3–5 min to remove unbound and residual periopathogens (Pourhajibagher et al., 2020). Finally, 10 μL of Eppendorf content containing maxillary molar was placed on MRS agar. After incubation at 37°C for 72 h under microaerophilic conditions, the CFU/mL of the probiotic strain adhered to the maxillary molar was calculated based on the method described by Miles et al. (1938).

The biofilm formation ability of probiotics was evaluated according to Mirzaei et al. (2022) using a microtiter plate assay. Briefly, the suspension of fresh probiotics was cultured in MRS broth supplemented with 2.5% glucose and incubated at 37°C for 12–36 h with 5% CO₂. Then standard suspensions (1.5 × 10⁸ CFU/mL) of probiotics were prepared, and 200 μL of suspension containing probiotic-MRS broth was added to a 96-well microplate (NEST, China) and incubated in an incubator with 5% CO₂ at 37°C for 36 h. After incubation, the suspension was gently removed from the well, and washed three times with PBS and air-dried for 30 min. Then, 200 μL of methanol (Mojallali, Iran) was added to each well to stabilize the biofilm. After 15 min, the solution was aspirated, and the plate was air-dried at room temperature. The wells were stained with 200 μL of crystal violet (0.05%) for 5 min, and the solution was aspirated, and the wells were washed three times with PBS and left to dry naturally at room temperature for 30 min. Finally, 200 μL of ethanol (95%) (Mojallali, Iran) was added to each well and incubated at 37°C for 30 min in a shaker incubator (IKA, Germany). The content of each well was transferred to the corresponding well in another microplate, and the absorbance was measured at 492 nm using a microplate reader (BioTek, United States). All experiments were performed in triplicate. The classification of probiotic isolates based on their optical absorbance (OD) is showed in Table 2.

Aggregation assay was performed according to Del Re et al. (2000). Certain probiotics with the ability to form aggregates, can effectively inhibit the formation of dental plaque by biofilm-producing bacteria. Briefly, probiotic isolates were incubated overnight in MRS broth medium in an incubator with 5% CO₂ at 37°C. After washing the cells with phosphate buffered saline (PBS), the probiotic suspension was prepared with PBS to achieve a concentration of 1.5 × 10⁸ CFU/mL. The ability of the current probiotics to induce aggregation was investigated by measuring the decrease in optical absorbance of bacterial suspensions at 600 nm due to aggregation and diffusion. The aggregation capacity (AC) was determined using the following formula, where ODt and OD0 are the optical absorbance at the end (8 h) and initial times, respectively (Bosch et al., 2012).

The methods employed for co-aggregation and aggregation assays were identical. Briefly, probiotic isolates were incubated overnight in MRS broth in an incubator with 5% CO₂ at 37°C. After washing the cells with phosphate buffered saline (PBS), the probiotic suspension was prepared with PBS to achieve a concentration of 1.5 × 10⁸ CFU/mL. The capability of the probiotics to co-aggregate with pathogens was investigated by combining equal quantities of probiotics and pathogens. After 8 h, the turbidity of the mixture was examined and compared to that of individual suspensions of each pathogen and each probiotic as the control groups (MacDonald et al., 2021). The co-aggregation capacity was reported as the percentage reduction in optical absorbance and calculated according to Handley et al. (1987) using the following formula: Where X and y indicate each of the two strains in the control tubes, (x + y) represents the mixture, and A represents absorbance at a wavelength of 600 nm.

The HGFC cells with the NCBI code of C165 were obtained from the Cell Bank Department of the Pasteur Institute of Iran. The HGFC were grown in DMEM (Dulbecco’s Modified Eagle Medium) (Biosera, France) supplemented with 15% heated-inactivated fetal bovine serum (FBS) (Biosera, France) and 2% glutamine (Bio-IDEA, Iran). Then, the cells were incubated in a 5% CO₂ incubator at 37°C. HGF cells were seeded in 6-and 12-well plates (SPL, Korea) in order to conduct adhesion and competition assays, respectively.

F. nucleatum, P. gingivalis, A. actinomycetemcomitans, and S. mutans, along with probiotic strains, were prepared in a solution containing 1.5 × 10⁸ CFU/mL. The equal volumes (200 μL) of the probiotic and the pathogen were cultured in TSB (tryptic soy broth) (Merck, Darmstadt, Germany) with 1% sucrose, then transferred to a 96-well plate and incubated in an incubator with 5% CO₂ at 37°C. The control group was considered for each pathogen and probiotic. Following the incubation, the suspension was carefully removed from the wells, and the wells were rinsed three times with PBS, before being left to dry for 30 min. Then, 200 μL of methanol was added to each well to stabilize the biofilm. After 15 min, the solution was aspirated, and the plate was air-dried at room temperature. The wells were stained with 200 μL of crystal violet (0.05%) and incubate for 5 min. Then, the solution was aspirated, and the wells were washed three times with PBS and left to air dry for 30 min at room temperature. Finally, 200 μL of ethanol (95%) was added to each well, and left to incubated at 37°C for 30 min in a shaker incubator. The content of each well was transferred to its equivalent well in the same microplate, and the absorbance was measured at 570 nm using a microplate reader.

GraphPad software version 9 was used for statistical analysis of the data. Paired-sample t-test was used to show significant differences between the groups receiving probiotics and the control group. Dunnet’s two-sided t-test was used to investigate the effect of probiotics on gene expression and biofilm formation of periodontal pathogens in the groups receiving probiotics either as prevention or as treatment, which allows multiple comparisons. The statistical significance level of the data was considered to be less than 0.05 (<0.05).

Although the number of probiotic strains exposed to the saliva and teeth after 24 h was significantly decreased compared to the control group, as shown in Table 4, probiotics displayed tolerance to saliva and showed weak adherence to the teeth (p > 0.05).

All probiotics were able to form biofilm. In Table 5, the probiotic isolates are classified based on biofilm production ability as strong, medium, weak, and non-biofilm producers. None of the probiotics showed strong abilities in producing biofilms, the maximum and minimum levels of biofilm production by probiotics were measured at 0.144 and 0.112, respectively. There was no significant difference among probiotics in biofilm production (p > 0.05).

All probiotics showed moderate to weak auto-aggregation ability (Table 6); however, no significant difference was observed in aggregation formation between probiotics (p > 0.05). The present study examined the co-aggregation of probiotics with oral pathogens, and measured the reduction in optical absorbance of bacterial tubes after 8 h. The findings revealed that all probiotics were able to co-aggregate with S. mutans, P. gingivalis, F. nucleatum, and A. actinomycetemcomitans. Strong co-aggregation was observed between B. bifidum 1001, L. plantarum 156, L. reuteri 100, and L. brevis 205 with S. mutans; B. bifidum 1005 and L. reuteri 100 with P. gingivalis; L. plantarum 156 and L. plantarum 42 with F. nucleatum; and B. longum 1044 and L. plantarum 156 with A. actinomycetemcomitans (Table 6).

L. plantarum 42 and B. breve 1,015 in the prevention group and L. rhamnosus 195, L. plantarum 156, and B. longum 1,044 in the treatment group significantly reduced the biofilm production ability of A. actinomycetemcomitans (p < 0.05), compared to the control group that did not receive any treatment. Furthermore, intra-group comparison showed no difference between the probiotics in the prevention and treatment groups in reducing A. actinomycetemcomitans biofilm production (p > 0.05). All probiotics in the prevention group were able to significantly reduce the biofilm production ability of P. gingivalis in comparison to the treatment group (p = 0.01). However, L. reuteri 100, B. bifidum 1001, B. bifidum 1005, B. breve 1015, and B. breve 1063 in the treatment group were able to reduce P. gingivalis biofilm formation, but it was not significant (p > 0.05). When comparing the effects of probiotics on F. nucleatum biofilm formation, it was observed that L. plantarum 42, L. reuteri 100, B. bifidum 1001, and B. longum 1044 in the prevention group, as well as L. plantarum 42, L. brevis 205, L. plantarum 156, B. bifidum 1001, B. bifidum 1005, B. breve 1015, and B. longum 1044 in the treatment group, reduced F. nucleatum biofilm formation in compare to the control group (p > 0.05).

The group of S. mutans that received probiotics for prevention showed a significant decrease in biofilm production compared to the control group (p = 0.04). In contrast, the difference in biofilm formation was not significant between the treated and untreated S. mutans groups (p = 0.05). No significant difference was observed following an intra-group comparison between either S. mutans isolates receiving probiotics as a prevention or treatment group (p > 0.05). Notably, L. plantarum 156 and B. bifidum 1001 in the prevention group and L. plantarum 42 and B. longum 1044 in the treatment group resulted in the highest reduction in S. mutans biofilm formation. All results are shown in Figure 5.

The main finding of this study is the durability of probiotics isolated from breast milk and infant feces in exposure to various enzymes as well as their binding ability to gingival fibroblast cells. To determine the most resistant probiotic in the oral cavity, the concentrations of protease, lipase, and α-amylase enzymes, tested in this study were adjusted near their natural concentrations in saliva, and the results showed all probiotics were stable in the oral cavity and saliva. Probiotics that possess the capability to produce biofilm, can establish long-term colonization in the gut, effectively limiting the growth of harmful bacteria. All probiotics demonstrated weak biofilm formation and aggregation. The development of biofilm and aggregation by probiotics is primarily influenced by their ability to adhere to surfaces, and all probiotics examined in this study showed the ability to attach to maxillary molars and gingival fibroblast cells (D'Alessandro et al., 2021). Nevertheless, probiotics showed weak adhesion capabilities, Lactobacillus spp. demonstrated great adhesiveness compared to other probiotic strains. This supports the idea that probiotics originated from fecal samples are the same as to those found in the mouth (Rohani et al., 2015). In order to establish long-term colonization of probiotics in the oral cavity, a strong and permanent adherence is required to colonization, however, probiotics remain for a short period in the oral cavity, and most of them establish a reversible adherence with cells that can be easily detached (Van Holm et al., 2023).

Biofilm serves as a niche for protecting microorganisms (Jayathilake et al., 2017). This study revealed, S. mutans exhibited a robust ability to form biofilms, although P. gingivalis showed weaker biofilm formation capability. Targeting biofilm formation in biofilm-producing pathogens is the most important objective in the management of caries and periodontal disease. An efficient probiotic hinders the growth of pathogens in dental plaque through biofilm formation and co-aggregation with pathogens. Our probiotics reduced the biofilm formation of A. actinomycetemcomitans, P. gingivalis, F. nucleatum, and S. mutans in the prevention and treatment groups. A significant decrease in biofilm formation of S. mutans was observed, which was confirmed by a significant reduction in the GtfB transcript. This is in accordance with observations made by Ahmed et al. (2014) and Hasan et al. (2015) that demonstrated the reduction in Gft enzyme expression plays an important role in preventing biofilm formation and developing caries in L. rhamnosus. These claims highlight the potential of Lactobacillus spp. as a promising alternative to antibiotics due to their anti-biofilm activity (Ahmed et al., 2014), and suppress the expression of GtfB, GtfC, and GtfD in S. mutans (Hasan et al., 2015). Although the main mechanism of biofilm inhibition by probiotics is still not fully understood, competing for nutrients, disrupting the attachment of pathogens, and producing antimicrobial peptides have been described as potential mechanisms (Barzegari et al., 2020). Some probiotics, like L. plantarum, reduce the biofilm mass by positively-charged D-Alanine in lipoteichoic acid (LTA). LTA has an inhibitory effect on the production of extracellular polymeric substances (EPS) (Ahn et al., 2018). The cationic amphiphilic structure of LTA also prevents aggregation by binding biofilm-associated genes to bacterial DNA (Anunthawan et al., 2015). In addition to LTA, probiotics such as L. fermentum can limit the biofilm production in S. mutans by using antibiofilm substances like biosurfactants (Tahmourespour et al., 2011). It is similar to the suppression of biofilm formation in Actinomyces naeslundii and Staphylococcus aureus through rhamnolipid and lipopeptide production by Burkholderia thailandensis and Bacillus subtilis, respectively (Rivardo et al., 2009; Elshikh et al., 2017). Interestingly, probiotics prevent the production of biofilm by other pathogenic bacteria due to their biofilm formation ability. According to the Ramos et al. study (Ramos et al., 2012), the supernatant of L. plantarum reduced the formation of P. aeruginosa biofilm without changing its matrix composition. By viability assay, they showed lactic acid solution prepared from L. plantarum not only has bacteriostatic and bactericidal properties on biofilm and planktonic forms of P. aeruginosa, but also suppresses the expression of virulence factors which are regulated by quorum sensing (QS), like elastase, rhamnolipid, and pyocyanin (Durant et al., 2000; Ramos et al., 2012). Differences in bacterial source, LTA structure, and probiotic strain led to controversies in biofilm inhibition results (Ryu et al., 2008).

Studies have shown that probiotics exert antimicrobial properties through the production of enzymes, acids, bacteriocins, and hydrogen peroxide (Duncker et al., 2011). In addition to preventing biofilm formation, the present study also showed L. plantarum 42, L. reuteri 100, and B. bifidum 1001 have inhibitory impacts on gene transcripts of RcpA, Mfa1, RgpA, Fap2, and GtfB in treated A. actinomycetemcomitans, P. gingivalis, F. nucleatum, and S. mutans compared to the untreated control group and other probiotics. The aforementioned genes are involved in biofilm formation (Smith et al., 2016), colonization, auto-aggregation, and the interaction of pathogens with the surrounding (Zijnge et al., 2012).

Probiotics also regulate host-related factors and modulate inflammation in the oral cavity. In this study, probiotics were effective in decreasing IL-8 and increasing IL-10 in HGFCs in both prevention and treatment groups. This is similar to the Kaci et al. (2011, 2014) findings which showed S. salivarius reduced TNF-α, IL-1, and IL-8 in stimulated intestinal epithelial cells and HT-29 by a small heat-resistant protein produced by some probiotics. Probiotics also modulate inflammation by producing short-chain fatty acids (SCFA) as postbiotic compounds like propionate and butyrate. Postbiotics produced by probiotics not only suppress the NF-κB pathway and interfere with suppression of Treg function (Vinolo et al., 2011), but also stimulate the production of anti-inflammatory cytokines and Treg differentiation (Kespohl et al., 2017).

The efficacy of our probiotics did not meet significant results as expected, which was similar to the Bosch et al. (2012) study. Lack of significant impressiveness by probiotics arises from their limited potential in biofilm formation, aggregation, and adhesion to HGFCs, which determine the eligibility of a candidate probiotic for use in the treatment of oral diseases (Sorroche et al., 2012). Moreover, the pathogen-specific inhibitory effect of probiotics and the presence of thick peptidoglycan in Gram-positive bacteria (Yang et al., 2021), should be considered which require a higher dosage of probiotics and a cell-free culture supernatant to significantly prevent the growth of harmful bacteria (Ding et al., 2021). In this context, Yang et al. (2021) showed that cell-free culture supernatant of L. reuteri had greater antimicrobial activity than live strains, due to the production of secondary bioactive metabolites (Ren et al., 2018), and the reduction of intracellular ATP required for the growth of oral pathogens. This evidences encourages us to prioritize the use of cell-free culture supernatants in future studies. In the present study, L. reuteri represents great function in tolerating oral conditions, binding to gingival fibroblast cells, and reducing biofilm formation, virulence gene expression, and inflammatory responses. The beneficial effect of L. reuteri and its survival in rough conditions were already proven and attributed to the production of inulin-type fructansucrase (Duncker et al., 2011), reutericyclin, reuteran, and reuterin (Sun et al., 2022).

To determine whether the origin of the probiotic isolation involved in lack of significant results or not, the functional equivalences and differences of oral and fecal probiotics should be addressed. Shimabukuro et al., in an animal model, evaluated the management of periodontitis by B. breve and B. bifidum, which were extracted from fecal samples. They showed complete clearance of P. gingivalis from oral biofilms and improvement in alveolar bone destruction by B. bifidum (Shimabukuro et al., 2021). The study evaluated the adherence ability and the modulatory effect of the immune system of L. salivarius AR809 extracted from the oral cavity of a healthy individual (Jia et al., 2019). The efficient adherence of oral isolated probiotics to pharyngeal epithelial FaDu cells and modulating the host’s immune response by enhancing IL-10 production and reducing the expression of TNF-α and IL-1B were observed. Despite being less studied about the effect of probiotics on dental caries disease and the need for further comprehensive studies, recent studies have shown probiotics isolated from oral and fecal sources to be promising candidates for clinical application (Shokryazdan et al., 2017).

The main challenges in the widespread application of probiotics for oral treatment are the introduction of single probiotic strains and the limited global access to probiotics (Spacova et al., 2020). Another important consideration in probiotic use is patient acceptance. However, following the use of mouthwash-containing probiotics, patient compliance was evaluated which more than 95% of volunteers were satisfied (Nisha et al., 2023). The effectiveness of probiotics also depends on the delivery method in terms of fluctuations of temperature, oxidation, pH level, and resistance to decomposition by enzymes. Various methods have been introduced to overcome delivery limitations, among them oral delivery is more common due to patient compliance, cost-effectiveness, and easy application for prevention and treatment purposes (Baral et al., 2021). Oral delivery of probiotics is achieved by dietary supplements, oral suspensions, lozenges, mouthwash, granules, capsules, oral films, and tablets for the treatment of oral diseases (Lee et al., 2017; Nie et al., 2023). Mouthwash-containing probiotics contribute to reducing plaque formation and compete with pathogens for adhesion sites, thereby regulating plaque ecology in normal flora (Meurman and Stamatova, 2007). Chewing gums are another common formulation of probiotics for oral diseases which are designed by probiotics (Ribeiro et al., 2020). Probiotic based gums successfully prevent S. mutans growth, and manage periodontal diseases (Krasse et al., 2006). The simple way to administer of probiotics is through dietary supplements in the form of granules to achieve a gradual and slow release of probiotics (Shirbhate et al., 2023).

Despite the safety history of probiotics for clinical application, a rarely occurring bacteremia in high-risk individuals was reported following consumption of probiotics (Doron and Snydman, 2015). Therefore, it is essential to properly examine the characteristics of microbial therapy with probiotics. The general safety of probiotics is described as their stability against rough conditions like gastric juice and enzymes, and their the binding ability, while functional safety is determined by having an antagonistic ability against pathogens, modulating immune responses, and selective activity in stimulating or suppressing the growth of certain bacteria in the oral cavity (Shirbhate et al., 2023). In addition to the survival of probiotics in the oral cavity, we have previously confirmed that present probiotics are resistant at pH = 2 and 0.4% bile salt, do not harbor any antibiotic resistance genes, or induce hemolysis (Rohani et al., 2015; Eshaghi et al., 2017). Also, the safety of present probiotics based on their mutagenicity, genotoxicity, and possible adverse effects was shown in the Darbandi et al. (2022) study, in which no mutations, genotoxicity in cell line examination, or adverse effects in animal models like mortality, abnormality, and weight change were observed in any dose range of 2000, 1,000, and 500 mg/kg. The examination of probiotics to produce lactic acid is important. Bosch et al. (2012) excluded 7 out of 46 probiotic candidates due to lactic acid production and the risk of dental caries.

The long-term effect of probiotics in the oral cavity is not well-known, but the short-term effects of probiotics showed they have significant anti-caries impacts, and significant changes in the alpha diversity of the oral microbiome (Dassi et al., 2018). Hradicka et al. (2023) studied the safety concerns of long-term probiotics in people who have taken probiotics daily for more than 10 years. They showed the long-term administration of probiotics does not have a significant effect on health, and causes gut microbiota alteration, and a significant increase in serum biochemical parameters, lipid metabolism, and inflammatory response. Meanwhile, the short-term administration of probiotics has significant effects on strengthening the gut-liver axis pathway and stimulating the innate immune response. This claim emphasizes the importance of the duration of probiotic consumption in the outcome of the clinical application.

For future prospective research, evaluation of the safety of probiotics, the composition of the microbiota before and after the administration of probiotics, the optimal dosage, the duration of administration, and determining whether preventive or treatment purposes were recommended to be considered. Despite the various animal model studies that have been conducted to investigate caries-related disease, rats are not an ideal representation of humans. Therefore, probiotics should be studied in all human groups, especially high-risk populations such as immunocompromised individuals and patients with severe underlying diseases. Also, it is important to consider the characteristics of the participants based on gender and lifestyle habits (smoking, drugs, alcohol, and lifestyle) in a long-term clinical trial.

The current study did not investigate the production of lactic acid, unpleasant volatile compounds, hydrogen sulfide, and soluble glucan by probiotics, as well as the scanning electron microscope (SEM) analysis that could provide more insight into microbial communities. These experiments will be examined in upcoming studies. In this study, different combinations of probiotics in cocktails were used to determine their efficacy against oral pathogens. The key distinction of the current study from similar studies lies in the source of probiotics utilized. While many studies have focused on the effects of commercially available oral probiotic tablets or isolated strains, the present study employed probiotics derived from feces and breast milk, which has shed light on their potential for oral treatment. Further studies are needed to examine their protective role in the mouth and confidently introduce all or some of them as potential candidates for the treatment of dental caries disease.