The Role of Glycoside Hydrolases in S. gordonii and C. albicans Interactions

ABSTRACT

Candida albicans can coaggregate with Streptococcus gordonii and cocolonize in the oral cavity. Saliva provides a vital microenvironment for close interactions of oral microorganisms. However, the level of fermentable carbohydrates in saliva is not sufficient to support the growth of multiple species. Glycoside hydrolases (GHs) that hydrolyze glycoproteins are critical for S. gordonii growth in low-fermentable-carbohydrate environments such as saliva. However, whether GHs are involved in the cross-kingdom interactions between C. albicans and S. gordonii under such conditions remains unknown. In this study, C. albicans and S. gordonii were cocultured in heart infusion broth with a low level of fermentable carbohydrate. Planktonic growth, biofilm formation, cell aggregation, and GH activities of monocultures and cocultures were examined. The results revealed that the planktonic growth of cocultured S. gordonii in a low-carbohydrate environment was elevated, while that of cocultured C. albicans was reduced. The biomass of S. gordonii in dual-species biofilms was higher than that of monocultures, while that of cocultured C. albicans was decreased. GH activity was observed in S. gordonii, and elevated activity of GHs was detected in S. gordonii-C. albicans cocultures, with elevated expression of GH-related genes of S. gordonii. By screening a mutant library of C. albicans, we identified a tec1Δ/Δ mutant strain that showed reduced ability to promote the growth and GH activities of S. gordonii compared with the wild-type strain. Altogether, the findings of this study demonstrate the involvement of GHs in the cross-kingdom metabolic interactions between C. albicans and S. gordonii in an environment with low level of fermentable carbohydrates.

IMPORTANCE Cross-kingdom interactions between Candida albicans and oral streptococci such as Streptococcus gordonii have been reported. However, their interactions in a low-fermentable-carbohydrate environment like saliva is not clear. The current study revealed glycoside hydrolase-related cross-kingdom communications between S. gordonii and C. albicans under the low-fermentable-carbohydrate condition. We demonstrate that C. albicans can promote the growth and metabolic activities of S. gordonii by elevating the activities of cell-wall-anchored glycoside hydrolases of S. gordonii. C. albicans gene TEC1 is critical for this cross-kingdom metabolic communication.

INTRODUCTION

Dental plaque biofilm is a complex microbial community that accumulates on the dental hard tissues and is associated directly with oral pathologies, especially dental caries (1, 2). After the acquired enamel pellicle is formed from salivary or nonsalivary proteins, the early dental plaque biofilm formation begins with the colonization of pioneer organisms on the pellicle, where Streptococcus spp. play a predominant role (3, 4). These pioneer colonizers facilitate sequential integration of other species and the establishment of biofilms (5). Candida albicans, as an oral opportunistic pathogen, has been identified in the etiology of dental caries, periodontitis, and other oral infectious diseases (6). C. albicans in the dental plaque shows active cross-kingdom interactions with a diverse spectrum of oral commensal bacteria, such as Streptococcus gordonii, Streptococcus oralis, and Streptococcus sanguinis (4). S. gordonii is one of the most predominant pioneer colonizers. S. gordonii promotes oral colonization of C. albicans by providing alternative adhesive surfaces in early-stage biofilm formation (7). The cross-kingdom communication between C. albicans and S. gordonii involves direct interaction through C. albicans surface proteins (e.g., Als3, Hwp1) and S. gordonii cell-wall-associated polypeptides (SspA, SspB) (8) and indirect cross-talk through quorum-sensing signals, such as farnesol from C. albicans (9). In addition, they may also share metabolic benefits (10). This fungal-bacterial interaction is mutualistically beneficial, facilitating the survival, colonization, and persistence of both species.

Dental biofilm formation is affected by the salivary environment (11, 12). Saliva facilitates the clearance of dietary fermentable carbohydrates and, on the other hand, also nourishes oral organisms with degradable glycans (13). Oral microbes residing in this environment live with a limited level of fermentable carbohydrates but a high level of dietary glycoproteins. Glycoside hydrolase (GH) is a category of carbohydrate-active enzymes (CAZymes) that are mainly involved in the cleavage of glycosidic bonds of glycans (14). S. gordonii possesses three types of cell-wall-anchored GHs, including GH2, GH20, and GH85, which can forage, metabolize, and transport salivary oligosaccharides, such as glycosylated basic proline rich protein (gPRP), thus contributing to the accommodation of S. gordonii in the fermentable sugar-limited environment such as saliva (15, 16). When environmental glucose is rare, expression of C. albicans genes related to hyphal formation and cell wall protein synthesis is upregulated in dual-species biofilms, while genes of S. gordonii relevant to fructose metabolism are upregulated (17), indicating the involvement of metabolic communication in the cross-kingdom interactions. In addition, the fact that C. albicans can only bind basic proline rich protein (bPRP) that is already adhered to S. gordonii may also suggest the potential involvement of metabolic communication between these two microorganisms (18). A variety of models have been developed to investigate the direct mutual interaction in traditional nutrient-rich microbiological media; however, whether the cross-talk also involves metabolic communication under the fermentable sugar-limited condition is still unclear.

In this study, we investigated the involvement of GHs in the cross-kingdom interaction of C. albicans and S. gordonii. We found that C. albicans promoted the planktonic growth and biofilm formation of S. gordonii in the fermentable carbohydrate-limited environment and elevated the activities of cell-wall-anchored GHs of S. gordonii. In addition, TEC1, a gene of C. albicans related to biofilm formation, hyphal growth, and virulence (19, 20), was involved in this cross-kingdom communication. The results provide a new perspective in the investigation of cross-kingdom interaction between oral commensal streptococci and opportunistic pathogenic fungi.

RESULTS

C. albicans promotes the planktonic growth and biofilm formation of S. gordonii under a fermentable carbohydrate-limited condition.

We used heart infusion broth (HIB) to mimic the fermentable carbohydrate-limited condition as in saliva. The numbers of CFU of S. gordonii and C. albicans in the cocultures were 11.59-fold higher and 15.17-fold lower than those in individual cultures, respectively (Fig. 1A). However, in 2% sucrose-supplemented HIB (HIBS), the levels of planktonic growth of both S. gordonii and C. albicans were similar to those of monocultures (Fig. 1A). Notably, the planktonic growth of S. gordonii in HIB was even higher than that in HIBS. Consistently, coculture of S. gordonii and C. albicans resulted in elevated biofilm formation in HIB compared to their monocultures (Fig. 1B). As visualized by confocal laser scanning microscopy (CLSM), S. gordonii or C. albicans monocultures in HIB formed biofilms with limited biomass (Fig. 1C). The bacterial community was more symmetrical than that of S. gordonii monocultures in HIBS (Fig. 1C). Filamentous fungal cells were observed in both HIB and HIBS. In the dual-species biofilms, the biovolume of S. gordonii was remarkably higher, along with a decreased biovolume of C. albicans in HIB (Fig. 1C and D). A relatively structurally uniform biofilm was formed, compared with densely packed aggregates observed in HIBS. In HIBS, the total biomass of the mixed biofilm was increased along with an increase of C. albicans although no statistics difference was detected (Fig. 1D). We further investigated the coaggregation of the two microbes in HIB. S. gordonii and C. albicans rapidly formed large aggregates in HIB, while forming relatively small clumps with more turbid suspensions in HIBS, suggesting an elevated coaggregation of S. gordonii and C. albicans in HIB relative to HIBS (Table 1). These results indicate a strikingly different bacterial-fungal interaction in environments with high- or low-level of fermentable carbohydrates.

FIG 1.

Planktonic growth and biofilm formation of S. gordonii in cocultures are elevated in HIB. (A) CFU of planktonic S. gordonii and C. albicans monocultures and cocultures in HIB and HIBS; (B) Total biomass of monospecies and dual-species biofilms estimated by crystal violet assay; (C) Representative CLSM images of monospecies and dual-species biofilms. Cyan, cocultured C. albicans. Scale bar = 100 μm. (D) Quantitative analysis of biofilm biovolume of S. gordonii and C. albicans. Each circle represents an individual replicate (n ≥ 3). Data are presented as mean ± SD. ***, P<0.001; **, P<0.01; *, P<0.05; n.s., not significant. S.g, S. gordonii in monocultures; C.a, C. albicans in monocultures; S.g in co, S. gordonii in cocultures; C.a in co, C. albicans in cocultures.

TABLE 1.

Coaggregation assay between S. gordonii Challis DL1 and C. albicans SC5314 in HIB or HIBS

C. albicans	Result for S. gordonii on:
	HIB	HIBS
HIB	+++	/ a
HIBS	/ a	++

^a/, indicates that the coaggregation ability of HIB-cultured S. gordonii with HIBS-cultured C. albicans or HIBS-cultured S. gordonii with HIB-cultured C. albicans was not examined.

C. albicans promotes the GH activities of S. gordonii under fermentable carbohydrate-limited condition.

To investigate the metabolic interactions under the carbohydrate-limited condition, seven common glycosides were selected as metabolic substrates to evaluate GH activities of both species in HIB. Different levels of GH activities were detected for the most of glycosides in monocultures of S. gordonii (Fig. 2A), among which GH activities for β-d-glucopyranoside (Glu), N-acetyl-β-d-glucosaminide (GlcNAc), and β-d-galactopyranoside (Gal) were the most robust. GH activity was almost undetectable in C. albicans monocultures, indicating the weaker glycometabolic ability of this species. In the S. gordonii-C. albicans cocultures, increased GH activities against Glu, Gal, Nacetyl-β-d-galactosaminide (GalNAc), α-l-fucopyranoside (Fuc), β-d-mannopyranoside (Man), and a-d-N-acetylneuraminic acid sodium salt hydrate (NANA) were detected compared with S. gordonii monocultures (Fig. 2A). GH2, GH20, and GH85 are cell-wall-anchored GHs of S. gordonii, which are essential in salivary glycan foraging and bacterial growth (15). When cocultured with C. albicans in HIB, the expression of GH2 (SGO_RS07290) and GH85 (SGO_RS01020) of S. gordonii was upregulated (Fig. 2B), indicating that C. albicans activates cell surface GHs of S. gordonii in the environment with low levels of fermentable carbohydrates. We further grew S. gordonii in the spent culture medium of C. albicans. S. gordonii showed elevated growth in C. albicans spent culture medium relative to its monoculture or coculture with C. albicans (Fig. 2C). The spent medium of C. albicans also promoted GH activities of S. gordonii for Glu and Gal (Fig. 2D). Moreover, the GH-promoting ability of C. albicans was also observed in clinical isolates, as all the tested clinical isolates of C. albicans promoted the GH activities of S. gordonii for Glu and Gal (Fig. 2E). All these data suggest that C. albicans can promote the GH activities of S. gordonii under the condition of limited fermentable carbohydrates.

FIG 2.

GH activities of S. gordonii are elevated in the presence of C. albicans. (A) Enzyme activities of S. gordonii and C. albicans monocultures and cocultures in HIB; (B) expression of cell-wall-anchored GHs of S. gordonii determined by reverse transcription real-time quantitative PCR (RT-qPCR); (C) CFU of S. gordonii cultured in C. albicans SC5314 spent culture medium; (D) GH activities of planktonic S. gordonii in C. albicans SC5314 spent culture medium. Each circle represents an individual replicate (n ≥ 3). (E) GH activities of S. gordonii cocultured with C. albicans clinical isolates. GH activities of S. gordonii monocultures were set to 1. Data are presented as mean ± SD. ***, P<0.001; **, P<0.01; *, P<0.05; n.s., not significant. S.g, S. gordonii in monocultures; C.a, C. albicans in monocultures; S.g in co, S. gordonii in cocultures; C.a spent, spent culture medium of C. albicans.

C. albicans promotes the growth of S. gordonii under a fermentable carbohydrate-limited condition in a TEC1-dependent manner.

We further investigated the genes of C. albicans that are involved in its promotive effects on the growth and GH activities of S. gordonii. We first screened a C. albicans mutant library using 4-methylumbelliferyl-β-d-glucopyranoside (4-MU-Glu) and 4-methylumbelliferyl-β-d-galactopyranoside (4-MU-Gal) as GH substrates. We identified six C. albicans mutant strains with the most reduced GH-promotive abilities compared to the SC5314 strain, including the crz2Δ/Δ, fcr3Δ/Δ, stp1Δ/Δ, upc2Δ/Δ, ash1Δ/Δ, and tec1Δ/Δ strains (Fig. 3A). Among these six C. albicans mutant strains, the crz2Δ/Δ and stp1Δ/Δ strains showed significantly reduced planktonic growth and biofilm formation in HIB (see Fig. S1A and B in the supplemental material) and thus were excluded for further validation.

FIG 3.

GH activities of S. gordonii in the presence of C. albicans mutants. (A) GH activities of S. gordonii cocultured with C. albicans mutant strains as measured by 4-methylumbelliferyl substrate assay (substrates are 4-MU-Glu and 4-MU-Gal). GH activities of S. gordonii cocultured with C. albicans SC5314 were set to 1, and the results were clustered by GH activities of Glu and Gal. (B) GH activities and (C) planktonic growth of S. gordonii in C. albicans spent culture medium. Each circle represents an individual replicate (n = 3). Data are presented as mean ± SD. ***, P<0.001; n.s., not significant.

We further grew S. gordonii in the spent culture medium of C. albicans fcr3Δ/Δ, upc2Δ/Δ, ash1Δ/Δ, and tec1Δ/Δ strains. We found that the promotive effects of tec1Δ/Δ strain on the GH activities and growth of S. gordonii were significantly reduced compared to those of the C. albicans SC5314 strain (Fig. 3B and C). When cocultured with tec1Δ/Δ strain, the levels of expression of the S. gordonii GH2, GH20, and GH85 genes were significantly lower than those of S. gordonii cocultured with C. albicans SC5314 (Fig. 4A). Consistently, the spent culture medium of the tec1Δ/Δ strain also showed a significantly decreased ability to induce the GH expression of S. gordonii compared with that of the SC5314 strain (Fig. 4B). The tec1Δ/Δ strain also showed a reduced ability to promote the biofilm formation of S. gordonii (Fig. 4C and D) and a decreased coaggregation ability (+ [small uniform aggregates]) in HIB compared with the SC5314 strain (+++ [large coaggregates that settled rapidly, leaving some turbidity in the supernatant fluid]), as described in Materials and Methods. All of these data indicate a critical role of TEC1 in the S. gordonii-C. albicans interaction under the fermentable carbohydrate-limited condition.

FIG 4.

C. albicans promotes the growth of S. gordonii in a TEC1-dependent manner. (A) Expression of cell-wall-anchored GHs of S. gordonii cocultured with C. albicans; (B) expression of cell-wall-anchored GHs of S. gordonii cultured in the spent culture medium of C. albicans; (C) representative CLSM images of S. gordonii-C. albicans dual-species biofilms. Scale bar = 100 μm. (D) Quantitative analysis of biofilm biovolume of S. gordonii in monocultures and cocultures. Each circle represents an individual replicate (n ≥ 3). Data are presented as mean ± SD. ***, P<0.001; **, P<0.01; *, P<0.05; n.s., not significant.

DISCUSSION

C. albicans is the most common opportunistic pathogenic fungus in humans (21). C. albicans is associated with various oral infectious diseases, including dental caries and oral mucositis (22, 23). In early-stage biofilm formation, commensal bacteria such as S. gordonii facilitate the adherence and persistence of fungi, which is critical for the pathogenesis of C. albicans. C. albicans interacts with S. gordonii through physical interactions, quorum-sensing molecular diffusion, and possibly, metabolic communications. The current study demonstrates that in the fermentable carbohydrate-limited environment, C. albicans promoted the growth and biofilm formation of S. gordonii and elevated the enzymatic activities of cell-wall-anchored GHs of S. gordonii, likely partly dependent on the expression of C. albicans gene TEC1.

Saliva facilitates bacterial and fermentable carbohydrate clearance in the oral cavity, but it also provides nutrients to support bacterial growth in dental plaque biofilms (13, 24). The concentration of glucose in saliva is approximately 10 to 100 μM, which is too low to support the growth of multiple species in the oral cavity (24). Moreover, the fermentable carbohydrate level fluctuates along with the host’s circadian cycle, diet changes, and life events, such as fasting and diseases. Most polypeptides in saliva are glycosylated (25). To gain sustainable growth in saliva, oral microbes work together to forage, degrade, and transport free carbohydrates from glycoproteins by GHs (26). HIB is characterized as a medium with no additive fermentable sugars in comparison with classic brain heart infusion (BHI) broth or yeast extract-peptone-dextrose (YPD) medium, whereas, beef heart infusion and tryptose within the formula of HIB provide nitrogenous compounds, amino acids, and other essential growth nutrients for microbes. Considering the oscillatory level of fermentable carbohydrates in saliva, HIB was adopted to mimic the condition of low fermentable carbohydrates but abundant nitrogen source as in saliva. Sucrose is the most widely used sweetener in food and pharmaceuticals (27). The white sugar and brown sugar used daily both originate from sucrose (28). To simulate the oral condition of a high-sugar diet, 2% sucrose-supplemented HIB was used. Interestingly, the planktonic growth of S. gordonii in HIB was higher than that in HIBS. The difference in cell growth may be attributed to several reasons. First, cells in HIBS were highly aggregated and less dispersed than those in HIB. Second, the pH is lower in HIBS (pH 4.2 [data not shown]) than that in HIB (pH 5.6 [data not shown]) after cultures, while it is hard for S. gordonii to increase growth under acidic pH (29). Finally, cells grew faster in HIBS with fewer nutrients for sustainable activity. According to the CAZy database (http://www.cazy.org/) (30), S. gordonii possesses a series of intracellular GHs and three extracellular GHs (GH2, GH20, and GH85); the latter are effective in degrading Glu, Gal, GlcNAc, and Man, favoring its growth in HIB like in whole saliva (15, 16). In this study, we confirmed that both C. albicans and S. gordonii grew and formed biofilms under conditions with a minimum (HIB) or high (HIBS) level of fermentable carbohydrates.

The colonization, excessive growth, and pathogenicity of C. albicans are closely related to oral microbial dysbiosis and oral infectious diseases (6, 31). Saliva forms a vital microenvironment for close contact and translocation of oral microorganisms (32). The current study shows that the growth and biofilm formation of S. gordonii were remarkably elevated in HIB in coculture, while those of C. albicans were relatively inhibited compared with the organism’s monocultures. However, with the supplementation with 2% sucrose, such commensalism/amensalism disappeared, indicating distinct metabolic cross-feedings between these two microorganisms in the presence of differed levels of fermentable carbohydrates. In synthetic saliva, the metabolic activity of the S. gordonii-C. albicans mixed biofilms is higher than the sum of monocultures (33). GHs have been identified to support the growth of S. gordonii under conditions with limited fermentable carbohydrates (15). Consistently, we also detected GH activities of S. gordonii in HIB. More importantly, we have demonstrated that C. albicans promoted GH activities of S. gordonii under the fermentable sugar-limited condition. A previous study also indicated the enhanced biofilm accretion of S. gordonii when coinoculated with C. albicans under conditions of salivary flow. Glucosyltransferase of S. gordonii may mediate the initial binding of the mixed biofilm (34). Therefore, besides recognizing and adhering to S. gordonii through direct interaction (7), C. albicans also metabolically interacts with S. gordonii in a fermentable carbohydrate-limited environment. This metabolic interaction is beneficial to S. gordonii via elevated utilization of polysaccharides and glycoproteins, resulting in its growth advantages in the mixed biofilm grown in HIB. Intriguingly, we observed inhibited growth and biofilm formation of C. albicans in the mixed biofilm grown in HIB. We speculate that the overgrowth of S. gordonii may compete with the carbon source of C. albicans and exert oxidative stress via hydrogen peroxide production under the fermentable sugar-limited condition (35), thus inhibiting candidal growth in the mixed consortium. In addition, the competence-stimulating peptide released from S. gordonii can also inhibit C. albicans biofilm formation (36), contributing to the decreased abundance of C. albians in the cocultures. It is possible that sacrifice of growth and promotion of streptococcal growth provide more potential adhesive niches to help C. albians survive along with S. gordonii under a low level of fermentable carbohydrates.

C. albicans can survive and grow in saliva without additional glucose (37), although little is known about how C. albicans senses and takes up glycans or glycoproteins. According to the CAZy database, C. albicans is rich in intracellular GHs but expresses only several glycosylphosphatidylinositol (GPI)-anchored cell wall GHs. Pga4 and Pga5 as 1,3-β-glucanosyltransferase genes (38, 39), play active roles in fungal cell wall biosynthesis (40). Xog1p, Exg2p, and Spr1p are exo-β-1,3-glucanases that hydrolytically liberate the glucose, gentiobiose, and other oligosaccharides from β-glucan (41, 42), the polysaccharide mainly found within the cell wall (43). Consistently, the GH activities of the seven common glycosides were almost undetectable in C. albicans in the current study. C. albicans cannot utilize salivary mucins as growth substrates: instead, the nitrogen-containing nutrient may determine the growth of C. albicans in the fermentable carbohydrate-limited environment (37). In this study, the spent culture medium of C. albicans promoted the growth and GH activities of S. gordonii, suggesting that C. albicans may provide extracellular substances to support the metabolism and growth of S. gordonii, which in turn may compete with and inhibit the growth of C. albicans under the nutrient-scarce condition. However, whether this contact-free promotive effect on S. gordonii involves the exosome of C. albicans is worth further investigation.

To further investigate the critical role of C. albicans in promoting the growth and GH activities of S. gordonii, we screened a C. albicans mutant library by measuring the promotive effect of each strain on the GH activities of S. gordonii. We identified the tec1Δ/Δ strain of C. albicans, which had significantly reduced ability to promote GH activities, growth, and biofilm formation of S. gordonii under the fermentable sugar-limited condition. TEC1 encodes the hypha-related transcriptional regulator, which is extensively involved in biofilm formation, hyphal growth, and virulence of C. albicans (19, 20, 44, 45). Tec1 is one of the core regulators required for biofilm formation of C. albicans and plays a crucial role in regulating dual-species biofilm formation with S. gordonii (46). The previous study indicated that the expression of TEC1 was upregulated in the presence of S. gordonii (35). Moreover, the dual-species biofilm formation of C. albicans and S. gordonii significantly decreased after TEC1 deletion under the fermentable carbohydrate-affluent condition (47). In this study, we also detected that the tec1Δ/Δ strain showed a significantly weakened ability to promote the GH activities of S. gordonii and biofilm formation of mixed biofilm under the fermentable sugar-limited condition. We believed that TEC1 is closely involved in the cross-kingdom metabolic communications between C. albicans and S. gordonii. However, the exact mechanisms need further investigation.

In summary, the current study revealed the GH-related cross-kingdom communications between S. gordonii and C. albicans under the low-fermentable-carbohydrate condition. C. albicans can promote the growth and metabolic activities of S. gordonii by elevating the activities of cell-wall-anchored GHs of S. gordonii. C. albicans gene TEC1 is critical for this cross-kingdom metabolic communication.

MATERIALS AND METHODS

Microbial growth condition.

C. albicans strains used in this study are listed in Table 2. The S. gordonii Challis DL1 and C. albicans strains were obtained from the State Key Laboratory of Oral Diseases (Sichuan University). All of the strains were maintained in the glycerol stock at −80°C and thawed on ice before inoculation. The purity of culture was checked by microscopic observation (Nikon, Tokyo, Japan). S. gordonii and C. albicans were grown to the mid-exponential phase in BHI (BD, New York, NY, USA) and YPD (1% yeast extract, 2% peptone, 2% glucose), respectively. For planktonic monocultures and cocultures, S. gordonii and C. albicans strains were cultured alone or together in HIB (2% beef heart infusion, 2% tryptose, and 1% sodium chloride) (BD, New York, NY, USA) with a low level of fermentable carbohydrates and in HIB supplemented with 2% sucrose (HIBS) as a fermentable carbohydrate-affluent medium. Totals of 1.0 × 10⁶ CFU·mL⁻¹ of S. gordonii and 1.0 × 10⁵ CFU·mL⁻¹ of C. albicans were prepared for monocultures and cocultures. Planktonic cultures were incubated at 37°C aerobically in an orbital shaker (150 rpm) for 24 h.

TABLE 2.

C. albicans strains used in this study

Strain name	Disrupted gene(s)	Reference for gene knockout protocol
SC5314	Reference strain
ZCF28	zcf28Δ/Δ	50
CPH1	cph1Δ/Δ	51
CNC11	hog1Δ	52
NOT3	not3Δ/Δ	53
ECE1	ece1Δ/Δ	54
HDA2	hda2Δ/Δ	55
CWT1	cwt1Δ/Δ	56
ZCF5	zcf5Δ/Δ	20
CNC13	hog1Δ/Δ	52
HAP43	hap43Δ/Δ	57
NDT80	ndt80Δ/Δ	58
ZCF21	zcf21Δ/Δ	59
THI20	thi20Δ/Δ	20
TYE7	tye7Δ/Δ	60
SKO1	sko1Δ/Δ	61
ACE2	ace2Δ/Δ	50
RLM1	rlm1Δ/Δ	62
RAS1	ras1Δ/Δ	63
GCN4	gcn4Δ/Δ	64
IRO1	iro1Δ/Δ	65
FGR13	fgr13Δ	66
BRG1	brg1Δ/Δ	67
CTA4	cta4Δ/Δ	68
EFH1	efh1Δ/Δ	69
MDR1	mdr1Δ/Δ	70
BDF1	bdf1Δ	71
ZCF32	zcf32Δ/Δ	72
ZCF11	zcf11Δ	66
HAP31	hap31Δ/Δ	73
ZCF39	zcf39Δ/Δ	50
MAC1	mac1Δ/Δ	74
NRG1	nrg1Δ/Δ	75
CAS5	cas5Δ/Δ	76
MCM1	mcm1Δ	77
GAL4	gal4Δ/Δ	78
PHR1	phr1Δ/Δ	79
PHR2	phr2Δ/Δ	80
BRE1	bre1Δ	81
ZPR1	zpr1Δ/Δ	20
CRZ1	crz1Δ/Δ	82
NDH51	ndh51Δ/Δ	83
ECM22	ecm22Δ/Δ	84
ZCF24	zcf24Δ/Δ	85
TAF14	taf14Δ/Δ	50
FGR17	fgr17Δ/Δ	84
ZCF38	zcf38Δ/Δ	20
ZCF6	zcf6Δ/Δ	20
HAP5	hap5Δ/Δ	86
RIM8	rim8Δ/Δ	87
RBF1	rbf1Δ/Δ	88
CPH2	cph2Δ/Δ	89
FKH2	fkh2Δ/Δ	90
TPK1	tpk1Δ/Δ	91
TPK2	tpk2Δ/Δ	91
TEC1	tec1Δ/Δ	92
ASH1	ash1Δ/Δ	93
UPC2	upc2Δ/Δ	94
STP1	stp1Δ/Δ	95
FCR3	fcr3Δ/Δ	50
CRZ2	crz2Δ/Δ	82
HAC1	hac1Δ/Δ	96
UME6	ume6Δ/Δ	97
MIG1	mig1Δ/Δ	98
EFG1	efg1Δ/Δ	99
MNL1	mnl1Δ/Δ	100
SPT20	spt20Δ/Δ	101
CK43A	cekΔ	102
GCF1	gcfΔ	103
Clinical isolatesa
2393		This study
6781		This study
22019		This study
6845		This study
3407		This study
2315		This study
5326		This study
4125		This study
1178		This study
2290		This study
10231		This study

^aAll C. albicans clinical isolates are available by State Key Laboratory of Oral Diseases (Sichuan University) upon request.

CFU enumeration.

The numbers of cells from planktonic monocultures and cocultures were determined by enumerating CFU. Cells were harvested by centrifugation (4°C, 4,000 rpm, 8 min). The supernatant was removed, and the cells were washed and resuspended in phosphate-buffered saline (PBS). Cells were serial diluted and inoculated to BHI or YPD agar plates and grew for 48 h in an aerobic incubator. The colonies of C. albicans and S. gordonii were enumerated based on the difference in morphology on the agar plates.

Preparation of saliva.

This study was approved by the Institutional Review Board at West China Hospital of Stomatology, Sichuan University (WCHSIRB-D-2017-031). Unstimulated whole saliva from two healthy adult volunteers was collected on ice and briefly clarified by centrifugation (15,000 × g for 30 min). The supernatant was sterilized through a syringe filter with a 0.22-μm-pore membrane and then stored at −80°C.

Biofilm formation quantification.

Twenty-four-well plates or μ-Slide 8 wells (Ibidi, Martinsried, Germany) were pretreated with sterile saliva at 37°C for 1 h. For monospecies and dual-species biofilms, each well of μ-Slide 8 wells was inoculated with 1.0 × 10⁵ CFU·mL⁻¹ of S. gordonii and/or 1.0 × 10⁴ CFU·mL⁻¹ of the C. albicans strains, and each well of 24-well plates was inoculated with 1.0 × 10⁶ CFU·mL⁻¹ of the S. gordonii and/or 1.0 × 10⁵ CFU·mL⁻¹ of C. albicans strains. The biofilms were allowed to develop up to 16 h under aerobic and static conditions at 37°C using HIB or HIBS as the culture medium.

The total biomass of biofilms was measured using the crystal violet assay. Briefly, the biofilms were rinsed to remove nonadhered cells, fixed with methanol, and then stained with 0.5% crystal violet (J&K, Beijing, China). The wells were filled with 33% acetic acid at 37°C for 30 min, and the absorbance of each well was measured by optical density at 590 nm (OD₅₉₀). For CLSM visualization, the biofilms were gently rinsed with PBS, stained, and incubated with 2.5 μM SYTO9 (Thermo Scientific, Waltham, MA, USA) for 15 min and then calcofluor white stain (comprising 1 mg/mL calcofluor white M2R and 0.5 mg/mL Evans blue [Sigma, St. Louis, MO, USA]) for 1 min in the dark. The biofilms were then rinsed gently for complete elution of the dye. CSLM was applied using a FV3000 inverted confocal microscope (Olympus, Tokyo, Japan). The excitation/emission wavelengths were 405/461 nm for blue fluorescence and 488/520 nm for green fluorescence. Three-dimensional images were reconstructed, and the biovolume was estimated by IMARIS 7.2.3 (Bitplane, Zurich, Switzerland).

Enzyme assays.

4-Methylumbelliferyl substrates were purchased from Sigma (St. Louis, MO, USA), including 4-methylumbelliferyl-β-d-galactopyranoside (4-MU-Gal), 4-methylumbelliferyl-β-d-glucopyranoside (4-MU-Glu), 4-methylumbelliferyl-β-d-mannopyranoside (4-MU-Man), 4-methylumbelliferyl-α-l-fucopyranoside (4-MU-Fuc), 4-methylumbelliferyl-N-acetyl-β-d-glucosaminide (4-MU-GlcNAc), 4-methylumbelliferyl-N-acetyl-β-d-galactosaminide (4-MU-GalNAc), and 2′-(4-methylumbelliferyl)-α-d-N-acetylneuraminic acid sodium salt hydrate (4-MU-NANA). The 4-methylumbelliferyl substrate assay was performed as previously reported, with minor modification (15, 48). Planktonic cells were centrifuged and washed with PBS. The cell suspension was comprised of S. gordonii only, C. albicans strains only, and cocultures of S. gordonii and C. albicans SC5314, mutant strains, or clinical isolates. The OD₆₀₀ of each sample was adjusted to 1 (±0.05). Thirty microliters of OD-adjusted cells, 70 μL PBS, and 100 μL of 4-MU substrate working solution (100 μM) were added sequentially to a black plate (Jingan, Shanghai, China). The plate was placed immediately into a Victor multilabel plate reader (Thermo Scientific, Waltham, MA, USA) using a preset program (excitation at 365 nm, emission at 445 nm, 1 s of measurement, and 5 repeats with 5-min intervals). The relative fluorescence units (RFU) were normalized with cell counts and measuring the time period and illustrated with GraphPad 8.0 (Prism, San Diego, CA, USA).

Spent culture medium assay.

A total of 1.0 × 10⁵ CFU·mL⁻¹ of C. albicans SC5314 or each of the mutant strains was cultured in HIB for 24 h aerobically with shaking. After being centrifuged at 4°C, the precipitate was removed, and 9 mL of supernatant was sterilized with a syringe filter. One milliliter of 10-fold-concentrated HIB was sterilized with a syringe filter and added to the supernatant of C. albicans to form the spent culture medium of C. albicans. A total of 1.0 × 10⁶ CFU·mL⁻¹ of S. gordonii was added to the spent culture medium of C. albicans SC5314 or mutant strains and cultivated for 24 h aerobically at 37°C with shaking. The CFU enumeration, RT-qPCR, and 4-MU substrate assay were performed to determine the planktonic growth, specific gene expression, and glycometabolic activities of cells.

Coaggregation assay.

The S. gordonii, C. albicans SC5314, and tec1Δ/Δ strains were cultured individually in HIB or HIBS under their respective growth conditions. Each sample was centrifuged and washed in coaggregation buffer (1 mM Tris-HCl [pH 8.0] containing 0.15 M NaCl, 0.1 mM MgCl₂, and 0.1 mM CaCl₂) (7) 3 times and resuspended in coaggregation buffer at an OD₆₀₀ of 1 (±0.05). One hundred microliters of S. gordonii suspension was mixed with 100 μL of C. albicans strain suspension. The mixture was vortexed, stood, observed, and scored from − to ++++ based on the following standards: −, no visible aggregates; +, small uniform aggregates; ++, definite coaggregates, but suspension remained turbid without immediate settling of coaggregates; +++, large coaggregates that settled rapidly, leaving some turbidity in the supernatant fluid; ++++, clear supernatant and large coaggregates that settled immediately.

RNA extraction and quantitative PCR.

Total RNA of planktonic monocultures and cocultures was extracted as previously reported (49). cDNA was generated using the RT reagent kit with gDNA Eraser (TaKaRa, Tokyo, Japan) according to the manufacturer’s instructions. Quantitative PCR was performed using TB Green (TaKaRa, Tokyo, Japan) in a CFX 96 real-time PCR detection system (Bio-Rad, Hercules, CA, USA). A total of 12.5 μL of TB Premix Ex Taq, 10 μM each forward and reverse primer, and 2 μL of cDNA comprised the qPCR mixture. After 30 s of incubation at 95°C, amplification was performed for 40 cycles, including 95°C (5 s) for denaturing and 60°C (30 s) for annealing. The threshold cycle (C_T) method was used for comparison of levels of expression of different genes. The expression level of target genes was normalized with S. gordonii 16S rRNA and calculated by the 2−^ΔΔCT method. The primer sequences are listed in Table 3.

TABLE 3.

Primer sequences

Primer	Sequence (5′→3′)a
S. gordonii 16S rRNA	F: AGACACGGCCCAGACTCCTAC
	R: CTCACACCCGTTCTTCTCTTACAA
SGO_RS07290	F: CCTTACGATTACGGTCGTTTCT
	R: CTCCATTAGCCTCACCGATTT
SGO_RS02020	F: CCAATGCTCAGGGTTGGTATTA
	R: CTGTGGTGATGCCAGGTTATAG
SGO_RS01020	F: TCGTCTTTCTCTGGGTCTCTAT
	R: GTTTCCCGTGGCCAGTATTA

^aF, forward; R, reverse.

Statistical analysis.

All data, except for the C. albicans strain screen, are presented as the mean ± standard deviation (SD) of results from at least two independent experiments performed in triplicate. Comparisons between different groups were performed by one-way analysis of variance (ANOVA) followed by Tukey’s multiple-comparison test or Student's t test in different contexts. P values of <0.05 were considered statistically significant.