Corynebacterial membrane vesicles disrupt cariogenic interkingdom assemblages

ABSTRACT

Polymicrobial diseases such as periodontal disease and caries pose significant treatment challenges due to their resistance to common approaches like antibiotic therapy. These infections exhibit increased resilience, due to microbial interactions that also disrupt host immune responses. Current research focuses on virulence and disease-promoting interactions, but less is known about interactions that could inhibit or prevent disease development. Normally human-associated microbiomes maintain homeostasis, preventing pathobionts from becoming dominant. In conditions like chronic disseminated candidiasis or severe early childhood caries (s-ECC), an overgrowth of microbes such as Candida albicans disrupts this balance. Typically, C. albicans coexists benignly within the microbial community but can become pathogenic, forming biofilms and interacting with other microbes such as cariogenic Streptococcus mutans. This interaction is particularly significant in s-ECC, where it exacerbates the disease’s progression and severity. Here, we present that Corynebacterium durum, itself and through its extracellular membrane vesicles disrupts interkingdom assemblages between C. albicans and S. mutans. Mechanistically the interaction interference occurs at the genetic level with downregulated HWP1 expression, a surface protein specifically induced in the presence of S. mutans promoting the interkingdom interaction. Additionally, we show that C. durum can impede C. albicans systemic virulence in the Galleria mellonella infection model. This suggests that oral corynebacteria may act as a beneficial commensal species, exerting antifungal effects within polymicrobial communities and opening new avenues for managing polymicrobial diseases.

IMPORTANCE

Polymicrobial diseases such as severe early childhood caries (s-ECC) lack effective treatment options. Prevention, requiring a deeper understanding of ecological processes before the onset of disease symptoms, could be a potential strategy. In this context, we investigated how relatively abundant oral biofilm Corynebacterium species, which are associated with oral health, can interfere with the interkingdom partnership of Streptococcus mutans and Candida albicans. This partnership is a significant driver of tooth decay in s-ECC due to synergistic activities that increase cariogenicity. Our study reveals that oral corynebacteria, through the production of extracellular membrane vesicles, can disrupt the S. mutans and C. albicans partnership by inhibiting fungal hyphae formation. Additionally, the fatty acid cargo within these vesicles exhibits antifungal properties, suggesting that corynebacteria play a role in shaping microbial dynamics within the oral biofilm.

INTRODUCTION

Polymicrobial diseases significantly impact human health, often leading to recurrent infections such as chronic wounds, ulcerative colitis, periodontal disease, and caries (1). Treatment of chronic polymicrobial diseases with antibiotics is impractical for daily, long-term preventive use due to the increasing prevalence of antimicrobial-resistant microbes (2). This is further exacerbated by the observation that mixed microbial communities are more resistant to antimicrobials and immune effectors (3). Moreover, interspecies interactions among disease-promoting communities can interfere with host immune responses (4). Although current research primarily focuses on understanding interspecies interactions that promote disease development or increase polymicrobial virulence, there is considerably less knowledge about microbial interactions that disrupt or prevent virulence (5). In a healthy state, human-associated microbiomes support homeostasis. A eubiotic microbial composition protects against invading pathogens through colonization resistance and prevents any members from becoming dominant, as observed in dysbiotic, polymicrobial diseases (6, 7).

Candida albicans is a common constituent of the human microbiota and is present at various mucosal sites within the host (8). Typically, C. albicans remains a benign member of the microbial community, coexisting with other microbes in mucosal and other biofilm-associated environments. However, it is also linked to several polymicrobial diseases, where C. albicans overgrowth leads to dysbiotic polymicrobial communities. These conditions range in severity from localized infections to systemic and potentially life-threatening chronic disseminated candidiasis (9, 10). The virulence of C. albicans is associated with its ability to form biofilms alongside other members of the human microbiota (8, 9). For instance, interkingdom interactions with S. mutans have been identified as a significant driver of severe early childhood caries, or s-ECC, which is characterized by a rapid and aggressive destruction of the smooth surfaces of teeth in the primary dentition. The role of S. mutans in the development of caries is well established; however, the interkingdom interaction with C. albicans has attracted increased interest due to its potential to augment virulence (9, 11–15). At the molecular level, the interaction between the two microbes is facilitated by the development of C. albicans hyphae, a process mediated by the hyphae-specific cell surface protein HWP1. HWP1 abundance is increased in the presence of S. mutans, allowing for a tight aggregation of S. mutans with C. albicans, which is even further enhanced by the presence of sucrose (16–18).

In this study, we report that two of the most abundant oral commensal corynebacteria, Corynebacterium durum and Corynebacterium matruchotii (19, 20) can interfere with the interkingdom interactions between S. mutans and C. albicans. Particularly, C. durum has been recently characterized for its ability to form species-specific interactions with another health-associated oral microbe, Streptococcus sanguinis (21). These specific interactions have several beneficial effects for S. sanguinis, primarily through the secretion of extracellular membrane vesicles (EMVs). Content analysis of these EMVs has identified several fatty acids, including palmitic, stearic, and oleic acids (21). Interestingly, these fatty acids have been reported to possess antifungal activity (22–25). Consequently, we tested the hypothesis that C. durum can exert an antifungal effect on C. albicans within the context of s-ECC interactions with S. mutans and confirmed key results with C. matruchotii. The presented data suggest that C. durum and C. matruchotii, both prevalent oral corynebacteria, can act as health-supporting commensal species.

RESULTS

Corynebacterium-dependent disruption of C. albicans and S. mutans type strain interaction

C. albicans is prevalent in the oral cavity and is typically considered a benign component of the oral microbiota. Nonetheless, its pathogenic role in s-ECC has been well established. Particularly noteworthy is the interkingdom interaction with the cariogenic species S. mutans, which exhibits enhanced virulence due to the cooperative behavior of both species that facilitates various processes (9). Moreover, the close physical interaction between the two species relies on the formation of C. albicans hyphae, presumably induced by S. mutans (16). Incubation of the S. mutans type strain UA159 with C. albicans ATCC 14053 under conditions conducive to biofilm formation confirmed the induction of hyphae formation, leading to a close association between the two species (Fig. 1). However, the simultaneous inoculation of C. durum JJ1 with S. mutans and C. albicans in the presence of sucrose that would normally promote interkingdom interactions prevented C. albicans from forming hyphae and associating with S. mutans, which could still form aggregates independently. This indicates that C. durum can disrupt the interkingdom interaction between the two types of strains.

Fig 1.

Impact of C. durum on interkingdom interactions between C. albicans and S. mutans. Presented is the interaction between C. albicans (type strain ATCC 14053) and S. mutans, with and without the presence of C. durum. Each strain was cultured in brain heart infusion (BHI) medium under three conditions: (i) C. albicans alone, (ii) C. albicans with S. mutans, and (iii) C. albicans with both S. mutans and C. durum. All inoculations were performed in 24-well plates at similar ratios and cultured overnight. Cells were resuspended by gentle pipetting before an aliquot was transferred onto a microscopy slide for imaging. Displayed are representative images from three biological replicates, all yielding similar outcomes (as a reference, all three species are labeled in Fig. S3 for comparison of microbial morphology).

Isolation of new oral C. albicans strains

C. albicans strains exhibit phenotypic heterogeneity (26–28), which might affect the capacity of C. durum to disrupt their interkingdom interactions. Consequently, we aimed to isolate new oral strains to verify that the observed inhibition of interkingdom interactions between S. mutans and C. albicans is not exclusive to the type strain. From a collection of several new isolates, we selected four strains (Ca-05, Ca-08, Ca-12, and Ca-16) as well as another previously isolated low-passage clinical isolate (Ca-JM) for their ability to form hyphae in the presence of S. mutans (Fig. 2A and B). A portion of the new isolates exhibited relatively poor hyphae formation and were therefore excluded from subsequent experiments. We confirmed the identity of the new C. albicans strains on selective CHROMagar Candida plates and demonstrated their capacity for hyphae formation and interaction with S. mutans (Fig. 2A and B; Fig. S1).

Fig 2.

Characterization of newly isolated C. albicans strains in interkingdom interactions with S. mutans. (A) Identification of newly isolated C. albicans strains was conducted using CHROMagar Candida, a chromogenic medium designed for the qualitative, direct detection and presumptive identification of Candida species. The C. albicans type strain ATCC 14053 and C. krusei type strain ATCC 14243 served as reference strains. (B) The interaction between the selected C. albicans strains and S. mutans was assessed under aerobic conditions in brain heart infusion (BHI) medium, both with and without the addition of sucrose. All tested C. albicans strains demonstrated hyphal growth in the presence of S. mutans. Scale bar = 20 µm.

C. durum and C. matruchotii are able to interfere with C. albicans and S. mutans interkingdom interaction

We subsequently confirmed the inhibitory effect of C. durum on the interkingdom interaction between C. albicans and S. mutans using the new isolates Ca-JM (Fig. 3) and Ca-05, Ca-08, Ca-12, and Ca-16 (Fig. S2). In the presence and absence of sucrose, Ca-JM was capable of forming hyphae when co-cultured with S. mutans, but the association with S. mutans was significantly more pronounced in sucrose-supplemented conditions (Fig. 3A and B). We observed large, intertwined aggregates of S. mutans and C. albicans hyphae. The induction of hyphae formation was clearly triggered by S. mutans, as cultures of C. albicans alone did not exhibit any hyphae formation (Fig. 3A and B; Fig. S1). Upon the addition of C. durum, the inhibition of hyphae formation observed previously with C. albicans ATCC 14053 was also noted with Ca-JM (Fig. 3A and B). Isolates Ca-05, Ca-08, Ca-12, and Ca-16 were also tested, and C. durum similarly interfered with these isolates (Fig. S2) confirming the broad-spectrum phenotypic intervention by C. durum. The presence of all three species in overnight culture is exemplified with Ca-05 (Fig. S3). Moreover, when another frequently isolated oral Corynebacterium species, C. matruchotii, was co-incubated with Ca-JM and S. mutans, inhibition of hyphae formation was again observed (Fig. 3A and B). We also quantified the influence of C. durum and C. matruchotii on cell numbers by plating and CFU counting for S. mutans and C. albicans (Fig. 3C through F). The incubation of S. mutans with either C. durum or C. matruchotii in BHI alone did not exhibit a difference in CFU counts (Fig. 3C). The addition of sucrose resulted in a statistically significant reduction in S. mutans CFU counts with C. matruchotii and a slight reduction in CFU counts in co-cultures with C. durum, compared to the S. mutans alone control (Fig. 3E). C. albicans numbers were statistically lower whether sucrose was present or not (Fig. 3D and F), confirming a Corynebacterium-dependent growth inhibition effect.

Fig 3.

Impact of C. durum and C. matruchotii on S. mutans-induced C. albicans hyphae formation. C. albicans low passage isolate Ca-JM was cultured under various conditions: alone, with S. mutans, with S. mutans and C. durum, or with S. mutans and C. matruchotii. Inoculations were performed at consistent ratios and cultures were grown overnight. Cells were then resuspended by gentle pipetting and an aliquot was placed on a microscopy slide for imaging. (A) Growth without sucrose. (B) Growth with sucrose. Both panels (A and B) display representative images from three biological replicates with similar results: scale bar = 20 µm. (C) Colony-forming unit (CFU) of S. mutans grown without sucrose. (D) CFU of C. albicans grown without sucrose. (E) CFU of S. mutans grown with sucrose. (F) CFU of C. albicans grown with sucrose. For panels (C–F), controls were cultures of single species, and data are presented as average and standard deviation for n = 3 replicates.

Purified corynebacterial extracellular membrane vesicles are also able to interfere with C. albicans and S. mutans interkingdom interaction

We previously discovered that EMVs purified from corynebacterial culture supernatants retain their biological activity (21). Therefore, we investigated the ability of purified EMVs from both C. durum and C. matruchotii to inhibit C. albicans growth and hyphae formation. EMVs (approximately 1.5 × 10¹¹) were added to dual-species S. mutans/C. albicans cultures at the time of inoculation. Growth was visualized after overnight incubation. The EMVs from both species were capable of inhibiting hyphae formation, as shown in Fig. 4A and B. Enumeration of C. albicans numbers by colony-forming unit (CFU) counting revealed a statistically significant inhibition of C. albicans CFUs (Fig. 4C and D), similar to that observed with the addition of corynebacterial cells. We confirmed that the inhibition of C. albicans by C. durum EMV is dose-dependent. Higher dilutions of C. durum EMV result in reduced inhibition, with 1.5 × 10¹¹ EMV exhibiting strong inhibitory effects, while 1 × 10⁸ EMV shows no effect on C. albicans (Fig. S4).

Fig 4.

Interference of C. durum and C. matruchotii EMVs with the interkingdom interaction between S. mutans and C. albicans. A low passage isolate of C. albicans (Ca-JM) was cultured under various conditions: alone, with S. mutans, with S. mutans plus C. durum EMVs, and with S. mutans plus C. matruchotii EMVs. Microbial inoculations were carried out at similar ratios. Purified EMVs (100 µL; approx. 1.5 × 10¹¹ particles) from overnight cultures of C. durum or C. matruchotii were added at the start of incubation. Following overnight incubation, cells were gently resuspended by pipetting and an aliquot was placed on a microscopy slide for imaging. (A) Growth without sucrose. (B) Growth with sucrose. (C) C. albicans CFU count without sucrose. (D) C. albicans CFU count with sucrose. The images presented are representative of three biological replicates with consistent outcomes. Control groups consisted of single-species cultures. Data are shown as mean and standard deviation based on three experiments. Scale bar = 20 µm.

Inhibition of C. albicans hyphae formation is specific to corynebacteria

To assess whether the observed inhibition of C. albicans hyphae formation by C. durum and C. matruchotii was due to the experimental setup, which involved three different species in the growth medium, we replaced corynebacteria with Neisseria elongata or Actinomyces oris. Both N. elongata and A. oris are components of the oral and nasopharyngeal biofilm, with A. oris belonging to the same class of bacteria, Actinomycetia, as C. durum and C. matruchotii. Notably, N. elongata is a Gram-negative species. As depicted in Fig. 5A, the presence of either species did not affect the hyphae formation of C. albicans or its association with S. mutans. The presence of N. elongata and A. oris was confirmed by plating aliquots on BHI plates and through microscopic examination (Fig. S5). Additionally, we evaluated fatty acid concentrations in the supernatants of A. oris and N. elongata. Compared to the supernatants from C. durum and C. matruchotii, no increase in fatty acid content was detected (Fig. 5B), further indicating that neither species influenced the interkingdom interaction between S. mutans and C. albicans.

Fig 5.

Lack of interference by oral species N. elongata and A. oris with S. mutans/C. albicans interkingdom interactions. (A) Triple-species cultures involving N. elongata, A. oris, and the S. mutans/C. albicans interkingdom pair, grown in brain heart infusion (BHI) broth, both with and without sucrose. The images displayed represent three biological replicates with consistent results. Scale bar = 20 µm. (B) Comparison of total fatty acids in supernatants from cultures of C. durum (Cd), C. matruchotii (Cm), A. oris (Ao), and N. elongata (Ne) grown in BHI and BHI supplemented with sucrose. BHI alone served as the control. Data are presented as means ± standard deviations for three replicates.

Interference of C. albicans hyphae formation is dependent on HWP1 downregulation by C. durum

Hyphae formation in C. albicans is a rigorously regulated process, and during interaction with S. mutans, HWP1 acts as a crucial modulator, showing increased expression in dual-species growth scenarios (16). HWP1, a surface protein, also plays roles in biofilm formation, interactions with hosts, and virulence, and is exclusively found on the hyphae surface (29). To investigate whether C. durum can disrupt hyphae formation through HWP1 modulation and subsequent S. mutans aggregation, we measured HWP1 expression in both co-cultures and triple cultures. As illustrated in Fig. 6A, co-cultures of S. mutans and C. albicans significantly increased HWP1 expression by approximately sevenfold, aligning with previous reports (13). Interestingly, HWP1 expression in the S. mutans/C. albicans/C. durum triple culture was similar to that in the C. albicans monoculture, suggesting that C. durum interferes with C. albicans hyphae formation by modulating its genetic control.

Fig 6.

Expression of C albicans strain JM HWP1 and hyphae formation on spider agar. (A) Expression of HWP1 gene in C. albicans (strain JM) was assessed in dual cultures with S. mutans (+Sm) and triple cultures with S. mutans and C. durum (+Sm + Cd), relative to the single culture of C. albicans without S. mutans (−Sm). Data are presented as means ± standard deviations (n = 3). (B) Hyphae formation of C. albicans JM on spider agar. An aliquot of 15 µL from an overnight culture was inoculated onto spider agar plates, with similar volumes of either S. mutans or C. durum placed on each side. A control inoculum of C. albicans JM was positioned at a greater distance. Plates were incubated for 7 days, with photographs taken post-incubation. (C) Close-up view of the interface between C. albicans/S. mutans and C. albicans/C. durum. This demonstrates that hyphae formation is inhibited near C. durum. Images shown are representative of three biological replicates, all yielding comparable results.

Hyphae formation can also be observed on spider agar (30). Aliquots of S. mutans and C. durum were inoculated near C. albicans and incubated for a specified period (7 days). Hyphae formation was visualized through photography. Figure 6B illustrated significant hyphal outgrowth from the control C. albicans colony (bottom colony in Fig. 6B, as well as Fig. S6 for Ca-05, Ca-08, and Ca-12). Near the interface with S. mutans, hyphae appeared to extend close to the edge of the S. mutans colony. However, hyphae formation seemed to halt at a distinct distance from C. durum, creating a hyphae-free zone between the two species, as shown in Fig. 6C.

C. durum is able to interfere with C. albicans virulence

C. durum demonstrates potential in modulating C. albicans virulence, particularly by inhibiting hyphae formation—a critical virulence factor necessary for attachment and biofilm formation. To investigate this interaction further, we employed the Galleria mellonella systemic infection model (31), monitoring survival rates over a 7-day period (see Fig. 7). Injection of 20 mL from a C. durum overnight culture (approximately 2 × 10⁸ CFU/mL) did not affect larvae survival, with survival rates of up to 100% similar to those seen in the phosphate-buffered saline (PBS) control group. In contrast, the highest mortality rates occurred following the injection of the same volume of an overnight culture of C. albicans (approximately 4 × 10⁸ CFU/mL) or its fivefold dilution. All larvae infected with C. albicans succumbed within a day post-injection (refer to Fig. S7). Notably, a significant improvement in survival was observed when the larvae were co-injected with undiluted C. durum and the fivefold diluted C. albicans culture; 25% of these co-infected larvae survived after 7 days (Fig. 7). These findings suggest that C. durum can significantly reduce C. albicans virulence in an in vivo systemic infection model.

Fig 7.

Interference of C. durum with C. albicans virulence in the G. mellonella infection model. Presented are the survival rates of larvae 7 days after infection under various conditions. Compared are the survival percentages of G. mellonella larvae following single infections with C. albicans and dual infections with both C. albicans and C. durum. For the dual infection setup, C. albicans was used both undiluted and at a 1:5 dilution. Larvae injected with PBS served as the control. N stands for undiluted inoculum. Data are presented as means ± standard deviations for each condition, with each consisting of 10 larvae. The graph displays representative results from two biological replicates.

DISCUSSION

The interkingdom association between S. mutans and C. albicans is clinically relevant due to its synergistic cariogenic activity (11, 14, 15, 32–34). This synergistic virulence has been explored in vitro and involves several major mechanisms:

1. Metabolically, S. mutans possesses exo-enzymatic activity that hydrolyzes sucrose into fructose and glucose (35). Sucrose significantly contributes to S. mutans cariogenicity due to glycosyltransferase (Gtf)-mediated hydrolysis and glucan synthesis, a polymeric substance that facilitates the formation of tenacious biofilms by S. mutans (36, 37). While C. albicans can metabolize sucrose intracellular via an alpha-glucosidase maltase, sucrose uptake seems to be repressed in the presence of glucose (38). Since Gtf hydrolyzing activity from S. mutans would release glucose from sucrose extracellularly, sucrose is less efficiently taken up and metabolized in C. albicans when compared to that of S. mutans (39). Therefore, the hydrolysis of sucrose by S. mutans enables synergistic carbohydrate metabolism providing C. albicans with the metabolically preferred glucose (40). Interestingly, GtfB plays a specific role in this process; as an exoenzyme, it can bind to the mannan layer of the C. albicans cell wall (41), exerting its hydrolytic activity to release fructose and glucose in close proximity to C. albicans (42), thus enabling swift uptake and metabolism. Consequently, mixed-species biofilms can further reduce the culture pH compared to S. mutans single-species biofilms, contributing to their synergistic cariogenicity (43).

2. Synergistic biofilm formation between the two species leads to an increased accumulation of biomass and subsequent fitness increase (14). Interkingdom biofilm formation might also facilitate the distribution of both species since it enables group-level surface motility (14). Synergistic biofilm formation was also observed in the presence of saliva with added sucrose (43), suggesting that synergistic activities are potentially as effectfive in vivo.

3. Relevant to our study is the role of hyphae formation in the process of community development. Numerous studies have shown that S. mutans can induce the expression of HWP1 (12, 16), a C. albicans surface protein exclusive to hyphae (29). Our work builds off and confirms existing research, demonstrating hyphae formation in the presence of S. mutans using several newly isolated C. albicans strains. Kim et al. highlighted the significance of hyphae formation when exposed to saliva, using a specific mutant, C. albicans efg1ΔΔ, which is incapable of developing hyphae (43). Although S. mutans and C. albicans efg1ΔΔ cells managed to grow together, the resulting biofilm was structurally flat, and the biomass was significantly reduced compared to the wild type. Interestingly, the biofilm produced by C. albicans efg1ΔΔ and S. mutans still effectively lowered the pH without compromising acidogenicity (43). GtfB binding to C. albicans efg1ΔΔ was comparable to the wild type, suggesting that the metabolic benefits in sucrose’s presence remain intact (41), thereby explaining the potential for pH reduction. Therefore, merely targeting hyphae formation to curb the synergistic cariogenic effect of S. mutans and C. albicans may be insufficient as demineralization could theoretically still occur. Nonetheless, C. albicans efg1ΔΔ alone proves to poorly construct biofilms (44), implying that hyphae formation is essential for biofilm development under certain conditions. In general, hyphae formation is regarded as a virulence factor. However, the direct role of C. albicans hyphae in caries development has not yet been demonstrated. A recent in vitro study revealed that both hyphae-deficient and hyper-filamentous C. albicans oral isolates can still form biofilms with S. mutans, promoting enhanced sucrose metabolism and increasing biofilm acidogenicity. This suggests that the caries-associated enhancement in virulence of S. mutans-C. albicans biofilms at the enamel surface likely occurs independently of hyphae formation (45).

Our study also explores the effects of C. durum and C. matruchotii cells and their EMVs on the interkingdom assembly of S. mutans and C. albicans, noting two primary impacts: significant interference with hyphae formation and decreased CFU numbers. The reduction in CFU, relative to the dual-species system of S. mutans/C. albicans, is likely due to the antifungal properties of fatty acids (22, 23, 25) found in the corynebacterial EMVs (21), such as oleic, palmitic, and stearic acid, though the exact mechanisms of hyphae suppression remain unknown. Interestingly, Honorato et al. reported that extracellular vesicles from C. albicans can disrupt their own hyphae formation (46). A detailed analysis identified sesquiterpenes, diterpenes, and fatty acids as cargo, albeit with shorter chain lengths than those in the corynebacterial EMVs (46), suggesting that hyphae regulation by its vesicles could be a natural control mechanism in C. albicans. The authors speculated that EVs could regulate morphogenesis during infection, as the yeast form may be advantageous during dissemination (46). This raises intriguing possibilities within the context of a diverse, health-associated oral multispecies biofilm, where corynebacteria might typically regulate the morphological state of C. albicans through EMV secretion, promoting a balanced presence in the oral biofilm. This homeostatic state may be disrupted under certain conditions, such as when S. mutans numbers increase as is the case in s-ECC, facilitating and supporting interkingdom interactions that enable C. albicans to counteract the corynebacterial EMV-imposed suppression of hyphae formation. This hypothesis is supported by next-generation sequencing studies revealing a decreased presence of corynebacteria in s-ECC (20, 47–49), directly connecting the interkingdom interaction of C. albicans and S. mutans on the ecological level with corynebacterial abundance. Therefore, an ecological approach to reinstate oral health in a community by either supplying corynebacterial species or their EMVs might be a viable option and warrants further investigation.

We also observed that C. durum can mitigate the virulence of C. albicans, as demonstrated in the G. mellonella systemic infection model. This observation aligns with findings from other studies. Specifically, the inhibition of C. albicans hyphae formation by its own membrane vesicles significantly increased larval survival in the same infection model (46). After pretreatment with these membrane vesicles, only 5–10% of the larvae died within a week, compared to an 80% mortality rate in the untreated control (46). Our C. albicans clinical isolate triggered 100% mortality within just 1 day. However, co-infection with C. durum increased survival to about 20%, presumably due to interference with hyphae formation. However, we cannot exclude the possibility that additional C. durum metabolites contributed to increased larval fitness. Regardless, these data strongly support C. durum’s role as a beneficial commensal species.

Moreover, our study in conjunction with existing literature on oral bacterial-fungal interactions (50–53), highlights a broader challenge: effectively investigating the interactions among microbial species in relation to health and disease. Health and disease are influenced by a myriad of factors, and the introduction of any additional relevant species can significantly change the dynamics of microbial interactions. The increasing complexity of microbial multispecies approaches poses experimental challenges, but also provides crucial opportunities to uncover novel mechanisms that may promote health or contribute to disease.

MATERIALS AND METHODS

Microbial strains and growth conditions

Streptococcus mutans UA159 (54), N. elongata (subsp. glycolytica) ATCC 29315, and A. oris (lab culture collection) were routinely grown anaerobically (90% N₂, 5% CO₂, and 5% H₂) as static cultures; and C. durum JJ1 (21) and C. matruchotti ATCC 14266 aerobically (5% CO₂) at 37°C in brain heart infusion medium (Bacto BHI; Becton Dickinson & Co., Sparks, MD, USA) or on BHI agar plates. C. albicans ATCC 14053 and Candida krusei ATCC 14243 were routinely grown aerobically (5% CO₂) at 37°C in brain heart infusion medium (Bacto BHI; Becton Dickinson & Co., Sparks, MD, USA) or on BHI agar plates. Biofilms were grown in 24-well tissue culture plates (avantor, Radnor, PA, USA) by inoculating equal ratios of cells. Biofilms were grown as indicated in BHI as static cultures at 37°C. Sucrose was added to the media from a filter-sterilized 50% (wt/vol) stock solution when indicated at a final concentration of 2.5%. Hyphae formation was tested on spider agar plates prepared as described previously (30).

CFU counting

Cells were resuspended, sonicated to disrupt cell chains and aggregates (55), serial diluted, and plated on BHI and CHROM Candida agar plates. Colony morphologies between all strains are easily distinguishable visually.

C. albicans isolation

C. albicans strains were isolated from dental plaque of volunteers by plating on Sabouraud dextrose agar and further confirmed on CHROMagar Candida plates (CHROMagar; Paris, France). The isolation was institutionally approved (IRB protocol 00026979, OHSU).

Bacterial membrane vesicle isolation and purification

Bacterial EMVs were isolated as described previously with minor modifications (56–58). In brief, C. durum or C. matruchotii were cultured in BHI with agitation at 37°C with 180 rpm agitation overnight. Cells from 1,000 mL cultures were pelleted and the supernatants were filtered through 0.45-mm pore size cellulose acetate filters (VWR International). Filtered supernatants were then concentrated using Vivaspin 20 ultracentrifugation units (100 kDa MWCO, GE Healthcare) prior to further centrifugation at 15,000 × g (15 min, 4°C) to remove cell debris and aggregates. The remaining supernatants were then centrifuged at 100,000 × g (1 h, 4°C) to isolate crude EMVs. The supernatants were discarded, and the pellets were suspended in 1 mL of 10 mM HEPES + 0.15 M NaCl.

C. albicans gene expression

Quantitative real-time reverse transcription polymerase chain reaction assay (qRT-PCR) was conducted to validate gene expression of HWP1 in C. albicans. RNA extraction was performed as published previously (59–61). In brief, cells were harvested, and pellets were resuspended in chilled TRI reagent (Sigma-Aldrich) and stored at −80°C. Bacterial cells were lysed in a Precellys Evolution homogenizer in the presence of 0.1 mm zirconia beads (BioSpec Products). Chloroform was added, RNA was precipitated, and then washed prior to being resuspended in nuclease-free, molecular-grade water (Corning). RNA was then treated with DNaseI (Turbo DNA-free, Ambion) and cleaned with the Qiagen RNeasy kit (Qiagen, Hilden, Germany). cDNA was synthesized and qRT-PCR was then conducted as published previously (61). Oligonucleotide primers and probes for qRT-PCR were designed with Primer3 software (62). Negative controls and the resultant cDNA were quantitatively amplified using Applied Biosystems PowerTrack SYBR Green Master Mix and a QuantStudio 3 Real-Time PCR System (Thermo Fisher Scientific, Wilmington, DE, USA). Each 20 µL of PCR reaction comprised cDNA template, 10 µM of each primer, and 2 × SYBR-Green mix (SYBR-Green and Taq DNA Polymerase). Three replicates were set up, and relative gene expression was determined using the comparative ΔΔCt method (63). C. albicans housekeeping gene for gene expression comparisons was ACT1 (encoding beta-actin) (64). The following oligonucleotides were used for gene expression: ACT1n3F: TGCTGAACGTATGCAAAAGG, ACT1n3R: TGAACAATGGATGGACCAGA, HWP1n1F: TCTACTGCTCCAGCCACTGA, and HWP1n1R: CCAGCAGGAATTGTTTCCAT.

Free fatty acid analysis

EnzyChrom Free Fatty Acid Assay Kit (BioAssay Systems, Hayward, CA, USA) was used to measure free fatty acid content in bacterial supernatant samples. Bacterial supernatants collected from overnight cultures were filtered through 0.45-mm pore size cellulose acetate filters (VWR International). The supernatants were concentrated using Vivaspin 20 ultracentrifugation units (100 kDa MWCO, GE Healthcare) and the free fatty acid concentrations were determined following the manufacturer’s instructions.

G. mellonella infection model

G. mellonella with a similar body size and weight (200–300 mg) were selected and divided into groups of 10 per experimental condition. Microbial cultures were grown in BHI overnight and the optical density as well as the CFU/mL were determined by plating on BHI plates. The proleg area for injection was sterilized with ethanol (70%) prior to the injection of 20 µL of the microbial suspension into the proleg of each larva using a 27-gauge standard hypodermic needle. Post-injection larvae were kept in a standard petri dish at 37°C and survival as well as pigmentation changes were recorded over a 7-day time interval.

Statistical analysis

The statistical significance of the difference between experimental groups was determined by Student’s t test (two-tailed). P values less than 0.05 were considered significant.

SUPPLEMENTAL MATERIAL

The following material is available online at https://doi.org/10.1128/aem.00885-24.

Supplemental figures. aem.00885-24-s0001.pdf.

Figures S1 to S7.

ASM does not own the copyrights to Supplemental Material that may be linked to, or accessed through, an article. The authors have granted ASM a non-exclusive, world-wide license to publish the Supplemental Material files. Please contact the corresponding author directly for reuse.