S. aureus ATCC12600, C. albicans SC5314 wild type, and C. albicans SC5314 als1/als3 (genotype: als1-1Δ::FRT/als1-2Δ::FRTals3-1Δ::FRT/als3-2Δ::FRT) (Van Dyck et al., 2021) were grown as described before (Pasman et al., 2024). In short, C. albicans and S. aureus strains were maintained from glycerol freezer stocks on Sabouraud dextrose/glucose agar supplemented with chloramphenicol (Sigma, 63567) and mannitol salt phenol red agar (MSA, Sigma, 89579), respectively. Single colonies were added to tryptic soy broth (TSB; Brunschwig Chemie, 211825) and cultured overnight at 37°C, 200 rpm. Cultures were rinsed with Dulbecco’s phosphate-buffered saline (DPBS; 137 mM NaCl, 2.7 mM KCl, 10 mM Na₂HPO₄, 1.8 mM KH₂HPO₄) and diluted to ~2 × 10⁶ CFU/mL for C. albicans and ~2 × 10⁷ CFU/mL for S. aureus in Dulbecco’s Modified Eagle’s Medium (Sigma D5030) supplemented with 2.5 g/L dextrose, 1× GlutaMAX (Gibco 35050061), and 1× MEM non-essential amino acids (Gibco 11140050), with a final pH of 7.3 (mDMEM-DMP) (Berman, 2012). Monocultures were generated by further diluting the culture in a 1:1 ratio with mDMEM-DMP, whereas co-cultures were constituted by combining (undiluted) monocultures in a 1:1 ratio. For buffered growth, mDMEM-DMP was supplemented with 100 mM of 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid (HEPES, Gibco 15630056).

Mono- and co-culture biofilms were grown by inoculating wells of a six-well plate (Corning 3506) with 3 mL of culture, prepared as described above, and incubating it stationarily for 2 h at 37°C. Each well was washed for three times using DPBS to remove non-adherent cells. Finally, fresh mDMEM-DMP was added to the wells and all plates were incubated stationary for 72 h at 37°C in a humidified environment. Concerning Transwell co-cultures, 1.5 mL of C. albicans monoculture was added to the inside of the 0.4 µm Transwell inserts (Thermo Scientific, 140660) whereas 1.5 mL of S. aureus monoculture was added to the inside of the well plate. Transwell co-cultures were further treated identical to normal cultures with half the volume added on either side of the membrane. Medium pH was measured by pH 2.0–9.0 strips (Supelco, 1.09584) inoculated with 40 µL of medium. Culture medium was sampled for downstream analysis, as described below. Biofilms were collected using cell scrapers (Greiner, 391-3010), pelleted, and stored at −80°C until further analysis using qPCR.

Genomic DNA was extracted from the collected cell pellets using a DNeasy PowerBiofilm Kit (Qiagen, 24000) according to the manufacturer’s protocol. To determine the amount of fungal and bacterial DNA, a quantitative polymerase chain reaction (qPCR) was performed using primers specific to the fungal 28S rRNA gene (forward: GCATATCAATAAGCGGAGGAA AAG; reverse: TTAGCTTTAGATGATTTACCACC; probe: 6FAM-CGGCGAGTG AAGCGGSAARAGCTC-BHQ) (Vollmer et al., 2008) and bacterial 16S rRNA gene (forward: TCCTACGGGAGGCAGCAGT; reverse: GGACTACCAGGGTATCTAATCCTGTT; probe: 6FAM-CGTATTACCGCGGCTGCTGGCAC-BHQ1) (Ciric et al., 2010). Fluorescence was measured using a LightCycler 480 System (Roche) and analyzed using the corresponding system software.

To concentrate the secreted proteins in the spent media, the medium collected from six wells of the same culturing conditions was pooled and filtered using a 0.2-µm polyethersulfone (PES) filter (Sarstedt, 83.1826.001). The filtrate was further concentrated with a 3-kDa Amicon ultra centrifugal filter (Pall, MAP003C37). The concentrated secretome samples were immediately stored without protease inhibitor supplementation for cytotoxicity assay, or supplemented with protease inhibitors (Roche, 11873580001) according to manufacture manual for proteomic analysis, or supplemented with RPMI 1640 to the original volume before protein extraction for macrophage activation experiments described below. Protein concentrations were determined using a Bradford protein assay, and prior to LC-MS proteomic analysis, protein concentrations were diluted to identical levels across all samples. All secretome samples were stored at −80°C until use.

Samples were prepared and measured according to Šimkovicová et al. (2024) (Šimkovicová et al., 2024). In short, samples were thawed, reduced, and alkylated by incubation with tris-(2-carboxyethyl)phosphine (10 mM) and chloroacetamide (30 mM) for 30 min, 70°C. Next, samples were prepared for mass spectrometry analysis using the single-pot, solid-phase-enhanced sample preparation (SP3) protocol (Hughes et al., 2014). Soluble protein recovery was optimized by ensuring no detergents were added to the samples and precipitation time was extended to 30 min (room temperature). Beads used for washing were air-dried and resuspended in ammonium bicarbonate (100 mM) after which trypsin (Sequencing Grade Modified) was added at a protease-to-protein ratio of 1:50 (w/w) at 37°C. Formic acid was added to the overnight digestion at a final concentration of 1% and pH of 2. Finally, peptides were recovered using a magnetic separator device.

Samples were separated by reversed-phase chromatography using an UltiMate 3000 RSLCnano UHPLC system (Thermo Scientific, Germeringen, Germany), and peptides were separated using a 75 μm × 250 mm analytical column (C18, 1.6 μm particle size, Aurora, IonOpticks, Australia), maintained at 50°C, and operated at a flow rate of 400 nL/min with 3% solvent B for 3 min (solvent A: 0.1% formic acid in water, solvent B: 0.1% formic acid in acetonitrile, ULCMS-grade, Biosolve). Next, a multi-stage gradient was applied (17% solvent B at 21 min, 25% solvent B at 29 min, 34% solvent B at 32 min, 99% solvent B at 33 min, kept at 99% solvent B till 40 min). For equilibration, the system was returned to initial conditions (t = 40.1 min) for 58 min. Eluted peptides were electrosprayed by a captive spray source via the column-associated emitter and were analysed by a TIMS-TOF Pro mass spectrometer (Bruker, Bremen, Germany) operated in PASEF mode for standard proteomics acquisition. MS/MS scans were initiated 10× with a total cycle time of 1.16 s, a target intensity of 2 × 10⁴, an intensity threshold of 2.5 × 10³, and a charge state range of 0–5. Active exclusion was enabled for a period of 0.4 min and precursors reevaluated when the ratio of current intensity:previous intensity exceeded 4.

LC-MS data were processed using MaxQuant software (version 1.16.14.0) using standard settings, i.e., trypsin/p as the enzyme allowing for two missed cleavages with carbamidomethylation at cysteine as a fixed modification and oxidation at methionine as a variable modification searching the proteome databases of: Candida_albicans_UP000000559 and Saureus_UP000008816. MaxQuant outputs were used for subsequent analysis using Perseus (version 2.0.7.0). Proteins that are only identified by peptides carrying one or more modified amino acids as well as reverse and potential contaminant proteins were removed from the dataset. Remaining data were log2 transformed, and proteins that were not measured in both samples of at least one condition were removed. Next, remaining proteins were annotated using the 2019_11 release of the Max Planck Institute of Biochemistry annotation database (C. albicans SC5314 and S. aureus NCTC 8325), after which the dataset was split into two sets: one set containing all C. albicans proteins and one set containing all S. aureus proteins. For both sets, missing values were imputed based on the low end of the corresponding normal distribution (width = 0.3, down shift = 1.8) and all values were subtracted with the most frequent value. Protein differences were tested on significance using ANOVA with a permutation-based FDR of 0.05 and 250 number of randomizations after which non-significant proteins were removed. Principal component analysis was used to identify sample group differences based on the remaining proteins. Finally, remaining data were normalized using Z-score normalization and significantly differing proteins between monoculture and co-culture conditions were identified using volcano plots based on a Pearson correlation with an FDR of 0.05. Using the virulence factor database (Liu et al., 2022), Aureo Wiki (Fuchs et al., 2018), UniProt (UniProt Consortium, 2018), STRING database (Szklarczyk et al., 2023), and various studies (Calderone and Fonzi, 2001; Sorgo et al., 2010; Zecconi and Scali, 2013), extracellular (ECVFs) and non-extracellular (N-ECVFs) virulence factors were identified. Results were visualized using Microsoft Excel (version 2402, build 17328.20282).

Ca9-22 (gingival squamous carcinoma) and HO1N1 (buccal epithelial carcinoma) cells were cultured in DMEM + 10% foetal bovine serum (FBS; Sigma) at 37°C, 5% CO₂. Cells reached confluence within 10 days and were washed using PBS, detached by trypsinization (0.05% trypsin) for 3 min at 37°C, spun down, and diluted to 1 × 10⁵ cells/mL using DMEM/F12 (Gibco) supplemented with 10% FBS. Cells were inoculated (1 × 10⁵ cells/well) into wells of a 24-well plate (Greiner) and incubated for 24 h at 37°C, 5% CO₂. Ca 9–22 and HO1N1 cells were exposed to secretome protein isolates (1:10 diluted in mDMEM-DMP, 100 mM HEPES), acquired as described above but without protease inhibitor supplementation, and incubated for 24 h at 37°C, 5% CO₂. Medium was collected and stored at −20°C until cytotoxicity assay. Cytotoxicity was determined using a Roche LDH Kit PLUS (Cat. No. 04 744 934 001) according to the manufacturer’s protocol. Medium of cells lysed with 1% Triton-X 100 in mDMEM-DMP (100 mM HEPES) was used as positive control, whereas medium of unexposed macrophages was utilized as negative control.

THP-1 monocytes were maintained in culture in Roswell Park Memorial Institute medium (RPMI 1640, Gibco, 11875093) supplemented with 10% foetal bovine serum (Sigma F9665), 1% GlutaMAX (Gibco, 35050061), and 1% Pen-strep (Gibco, 15140-122). THP-1 monocytes were seeded in either 96- or 24-well plates at a density of 0.5 × 10⁶ cells/mL and differentiated into M0 macrophages by incubating the cells for 72 h at 37°C, 5% CO₂ in RPMI 1640 supplemented with 100 nM phorbol 12-myristate 13-acetate (PMA, Millipore 500582). Following a 72-h incubation, cells were washed with Dulbecco’s phosphate-buffered saline (DPBS, Gibco, 14190177, 37°C) and used for downstream application. Differentiated macrophages were imaged at 10× magnification using a Leica DM13000 B coupled to a ZEISS Axiocam MRc.

THP-1 M0 macrophages, differentiated in 96-well plates, were incubated in RPMI 1640 supplemented with various dilutions of the conditioned media (1:25, 1:50, 1:100, and 1:200, diluted in RPMI 1640) for 6 h at 37°C, 5% CO₂. Macrophages in RPMI 1640 served as the negative control, whereas macrophages treated with 1% Triton-X 100 for 10 min before medium collection served as the positive control. Following incubation, medium was collected and LDH activity was quantified using an LDH-Glo™ Cytotoxicity Assay (Promega, J2380), measured with a Synergy Mx (Bio Tek). Cytotoxicity was determined as follows: Percentage of cytotoxicity (%) = 100 * ((Experimental LDH release – negative control)/(Positive control LDH release – negative control)).

Macrophages were lysed by incubating in 350 µL of 1% Triton-X 100 (Bio-Rad) in DPBS for 10 min. Subsequently, 350 µL of buffer RLT in the RNeasy Plus Mini Kit (Qiagen, 74134) supplemented with 1% β-mercaptoethanol was added, mixed with pipetting, harvested in 1.5-mL tubes, and stored at −70°C until RNA isolation. The total RNA was then isolated by following the manufacturer’s protocol. The RNA purity was measured using a DeNovix DS-11+ spectrophotometer, with RNA integrity number of all samples 9.9–10.0 except for one at 9.7 due to the high concentration of total RNA.

The mRNA enrichment and library preparation was carried out by the Dutch Genomics Service & Support Provider at the University of Amsterdam. The NEBNext Poly(A) mRNA Magnetic Isolation Module (New England Biolabs) was used to perform a poly-A enrichment using 1 μg total RNA. RNA-Seq libraries were generated according to the manufacturers’ protocols using the NEBNext Ultra II Directional RNA Library Prep Kit for Illumina and NEBNext Multiplex Oligos for Illumina (Unique Dual Index Primer Pairs) (New England Biolabs). The size distribution of the libraries with indexed adapters was assessed using a 2200 TapeStation System with Agilent D1000 ScreenTapes (Agilent Technologies). The libraries were quantified on a QuantStudio 3 Real-Time PCR System (Thermo Fisher Scientific) using the NEBNext Library Quant Kit for Illumina (New England Biolabs) according to the instructions of the manufacturer. The libraries were clustered and sequenced (75 bp) on a NextSeq 550 System (Illumina) using a NextSeq 500/550 High Output Kit v2.5 (75 Cycles) (Illumina).

The raw sequencing data were subjected to quality control with FastQC v0.11.9 (Andrews, 2010) and MultiQC version 1.21 (Ewels et al., 2016). Subsequently, the reads were trimmed with Trimmomatic v0.39 (Bolger et al., 2014). Post-trimming quality control indicated that all samples were of similar high quality. The reads were aligned (without soft clipping) to the human reference genome (GRCh38.111) using HISAT2 version 2.2.1 (Kim et al., 2015). HTSeq-count version 2.0.5 (Anders et al., 2015) was used to count the amount of reads per gene and determine the gene expression values. The gene expression data were checked for quality and subsequently normalized and analysed with DESeq2 (Love et al., 2014). The technical replicates were collapsed and statistical tests controlling for batch effects were performed. The differences between the monocyte control and the medium control were assessed by comparing differentially expressed genes (DEGs), i.e., genes with a FDR-adjusted p-value < 0.01 and a log2 fold change above 1 or below −1, with previously identified DEGs following THP-1 monocyte differentiation (Liu et al., 2023). DEGs were identified between all conditions and visualized using Microsoft Excel (version 2403). Expression of M1 and M2 polarization related genes was identified according to previously published gene sets for differential gene expression (Mills and Ley, 2014; Murray, 2017; Shapouri-Moghaddam et al., 2018; Rynikova et al., 2023). Finally, to identify pathways containing overrepresented genes, a Gene Set Enrichment Analysis (GSEA) was performed using WebGestalt 2019 (Liao et al., 2019) with a KEGG-based pathway analysis, a weighted set cover redundancy removal, and a significance level cutoff of FDR-adjusted p-value < 0.01. WebGestalt output pathways were visualized using both Microsoft Excel (version 2403) and Python (version 3.8.5). Gene ratios were determined by dividing the number of enriched genes in a pathway by the total number of genes of that pathway.

All data were tabulated and visualized using Microsoft Excel (version 2403). When applied, data normality was tested using a Shapiro–Wilk test, and group comparison was performed using either a one-way ANOVA (normally distributed data) or Kruskal–Wallis test (non-normally distributed data) combined with a Tukey post analysis using Prism graph pad (8.3.0). All conditions were tested in at least three biological replicates and technical duplicates.

To investigate the effect of C. albicans and S. aureus co-culturing on the secreted proteins, we measured the levels of the proteins in the spent media collected from different co-culture and monoculture conditions (see Methods). Following mass spectrometry analysis, we detected 183 C. albicans proteins in both samples of at least one culturing condition. Principal component analysis (PCA) revealed a clear separation between co-cultures and monocultures ( Figure 1A ), indicating that C. albicans-secreted protein profiles were changed by the presence of S. aureus. Of the 21 known C. albicans ECVFs ( Supplementary Table S1 ), 14 ECVFs were detected in both samples of at least one culturing condition and 12 ECVFs were significantly increased in co-culture compared with monoculture ( Figure 1B ; Supplementary Table S1 ). Among these, five ECVFs, i.e., Plb1, Sap4, Sap6, Sod5, and Csa2, were significantly increased only in wild-type co-culture compared with monoculture, but this increase was abolished in ALS1/ALS3 ΔΔ/ΔΔ or Transwell setting ( Supplementary Table S1 ), suggesting that their secretion was strongly dependent on Als1/Als3-mediated binding, whereas secreted aspartic protease 5 (Sap5) and glycosidase Crh11 was independent of Als1/Als3-mediated binding and other types of physical interaction ( Supplementary Table S1 ). Interestingly, the remaining five ECVFs, i.e., Sap9, glycosidase Utr2, glucan 1,3-beta-glucosidase Xog1, PR-1 protein homolog Rbe1, and Rbt4, were significantly increased during both regular and Transwell-separated co-culturing with wild-type C. albicans but were not changed following co-culturing with C. albicans ALS1/ALS3 ΔΔ/ΔΔ ( Supplementary Table S1 ), indicating that their release was strongly dependent on soluble factors and independent of Als1/Als3-mediated binding ( Figure 1B ; Supplementary Table S1 ). These significantly increased C. albicans ECVFs contribute to damaging functions such as proteolysis, immune evasion, cell wall modelling, and iron acquisition, suggesting that co-culturing C. albicans with S. aureus promotes virulence of C. albicans. It is worth noting that C. albicans candidalysin, a known virulence factor, does not contribute to the virulence increase in co-cultures since C. albicans candidalysin was detected but did not significantly change following co-culturing with S. aureus ( Supplementary Table S1 ).

Next, we expanded our investigation to non-extracellular virulence factors (N-ECVFs), as cytoplasmic proteins in C. albicans were showed to be released extracellularly via extracellular vescicles (Gil-Bona et al., 2015). We identified 42 N-ECVFs, of which 14 C. albicans N-ECVFs were significantly changed and mainly contributed to co-culture-based (PC1, 37.5%) and ALS1/ALS3 deletion-based (PC2, 22.6%) separation ( Figure 1C ). Most of the 14 significantly changed C. albicans N-ECVFs were related to adhesion or cell wall remodeling. For instance, hyphae-regulated cell wall protein 1 (Hyr1) and heat shock protein 70 (Ssa1) were significantly increased in wild-type co-culture versus C. albicans monoculture, but this change was attenuated in als1/als3 ΔΔ/ΔΔ co-culture and Transwell settings ( Supplementary Table S2 ). Additionally, thioredoxin peroxidase (Ahp1) and glycolipid 2-alpha-mannosyltransferase 1 (Mnt1), involved in the reduction of hydrogen peroxide and O-glycosylation of cell wall, respectively, were significantly decreased in wild-type and Transwell co-culture, and this decrease was enhanced in ALS1/ALS3 ΔΔ/ΔΔ co-culture. These results suggest that Hyr1, Ssa1, Ahp1, and Mnt1 are influenced by both Als1/Als3-dependent C. albicans–S. aureus interaction and other S. aureus factors. In addition, cell wall protein (Rhd3) and cell wall remodeling-related proteins Ihd1, Sun41, Pga4, Als1, Als3, Mp65, and Bgl2 were significantly increased in wild-type co-culture ( Supplementary Table S2 ), which was attenuated by the physical separation and completely abolished in ALS1/ALS3 ΔΔ/ΔΔ co-culture ( Supplementary Table S2 ). Moreover, the enolase (Eno1) level was not altered during wild-type co-culturing, whereas its level was decreased in ALS1/ALS3 ΔΔ/ΔΔ and Transwell co-cultures ( Supplementary Table S2 ). Together, these results suggest that the Als1/Als3-dependent and independent physical interaction between C. albicans and S. aureus promotes the levels of non-extracellular virulence factors.

Co-culturing also influenced the release of virulence factors by S. aureus. We detected 930 S. aureus proteins in at least one culturing condition. Principal component analysis indicates a clear separation between monocultures and co-cultures ( Figure 1D ), suggesting that the presence of C. albicans has substantial impact on the proteins released by S. aureus. It is worth noting that no separation was observed among three co-culture conditions ( Figure 1D ), indicating that deletion of ALS1/ALS3 or lack of physical interaction with C. albicans did not influence the S. aureus secretome compared with wild-type co-culture. Of the 50 known S. aureus ECVFs, 27 were detected in both samples of at least one culture condition and 20 were found statistically significantly changed ( Figure 1E , Supplementary Table S3 ). Of the 20 changed ECVFs, hemolysins including alpha hemolysin (Hly/Hla), Hlb, and HlgA-C were increased in all conditions ( Supplementary Table S3 ). During the co-culture, C. albicans can maintain the pH ( Supplementary Figure S1A ) from decreasing caused by S. aureus. Here, we observed that buffering the pH alone increased the secretion of these hemolysins to a similar extent as the C. albicans, and this increase was maintained despite the lack of physical contact or Als1/Als3-mediated binding ( Figure 1E ; Supplementary Table S3 ). This result, together with the maintained pH during C. albicans–S. aureus co-culture ( Supplementary Figure S1A ), suggests that C. albicans-maintained pH likely contributed to the increased release of the hemolysins. On the other hand, other S. aureus ECVFs were not influenced by pH but by other factors during co-culturing. For instance, cysteine proteinase staphopain B (SspB), serine protease-like protein A-F (SplA-F), lipase (Lip), phospholipase C (Plc), delta hemolysin (Hld), and leukotoxin D (LukD) were significantly more present in all co-culture conditions, but significantly less present or unaltered during buffered monoculture ( Supplementary Table S3 ). Because ALS1/ALS3 deletion and separated growth did not deviate from wild-type co-culture results, these results are independent of Als1/Als3 and physical binding.

Similar to C. albicans, we also detected S. aureus N-ECVFs. Of the 28 detected N-ECVFs, 17 were significantly changed ( Figure 1F ). Buffering strongly impacted N-ECVF levels, evidenced by the fact that 11 and 2 N-ECVFs in buffered monoculture were significantly higher and lower than those in unbuffered monoculture, respectively ( Figure 1F ). This pH-mediated effect seemed to be divergently interfered with other factors from C. albicans. For example, the buffering increase of d-alanine-d-alanyl carrier protein ligase (DltA) and enolase (Eno) was further enhanced by the presence of C. albicans, and this enhancement is independent of Als1/Als3 or physical proximity ( Figure 1F ; Supplementary Table S4 ). For those proteins that were decreased by buffering, i.e., ClfB, Ebp, Oata, Srap/Sasa, IsdA, IsdC, IsdE, SpA, SdrC, and ClfA, the decrease was attenuated or even reserved by the presence of C. albicans ( Supplementary Table S4 ), despite that C. albicans maintained pH comparably with HEPES buffering ( Supplementary Figure S1A ). These results suggest that other C. albicans-derived soluble factors have stronger effects in influencing these proteins compared with pH. Together, these results suggest that both C. albicans-derived pH maintenance has a broader impact than that of Als1/Als3-mediated binding on N-ECVFs of S. aureus.

Previous in vivo murine studies have reported that oral inoculation of C. albicans potently promotes co-invasion and dissemination of co-inoculated S. aureus (Carlson, 1982; Carlson, 1983; Carlson, 1988; Nash et al., 2014; Kong et al., 2015; Schlecht et al., 2015; Nash et al., 2016; Allison et al., 2019; Van Dyck et al., 2021), suggesting that epithelium in the oral cavity is disrupted by co-infection. Reasoning that the increase in ECVF and N-ECVF in C. albicans–S. aureus co-culture would increase the damaging potential, we next sought to test whether the co-culture induces higher cytotoxicity to oral squamous cells. To test this, we exposed human gingival squamous Ca 9–22 and human buccal mucosa squamous HO1N1 cells to undiluted mono- and co-culture secretomes of C. albicans wild-type, C. albicans ALS1/ALS3 ΔΔ/ΔΔ, and/or S. aureus (cultured in either buffered or unbuffered medium) for 24 h and determined cytotoxicity by measuring lactate dehydrogenase activity in the medium. All monoculture secretomes, i.e., C. albicans wild-type, C. albicans ALS1/ALS3 ΔΔ/ΔΔ, and S. aureus (unbuffered), induce similar levels of cytotoxicity as the negative control (~20%–30%; Figure 2 ). In contrast, all tested co-culture secretomes and S. aureus buffered induced higher cytotoxicity to both Ca 9–22 cells and HO1N1 cells ( Figure 2 ). These results are in agreement with the observed elevated levels of ECVFs in co-culture secretomes compared with monoculture. Interestingly, the secretome of buffered S. aureus monocultures showed higher cytotoxicity to Ca 9-22 (92%) and HO1N1 (79%) ( Figure 2 ), despite that the levels of ECVFs in buffered S. aureus were comparable or lower than that in co-cultures ( Supplementary Table S3 ).

Next, we sought to investigate the effects of co-culturing on macrophages. Macrophages are first-line defense mechanisms during the infection in oral mucosa and are expected to encounter C. albicans and S. aureus during initial co-invasion of the mucosa. We differentiated THP-1 monocytes into M0 macrophages and verified the differentiation by comparing the macrophage and monocyte transcriptomes, as well as macrophage morphology ( Supplementary Figure S2 ). Secretome of monocultures and co-cultures induced different levels of cytotoxicity to these macrophages ( Figure 3A ). Surprisingly, C. albicans monoculture showed similar cytotoxicity as the medium control. Similar to oral cells, co-cultures induced higher cytotoxicity than S. aureus and C. albicans monocultures, and the cytotoxicity is dose-dependent, with 1-to-200 dilution showing the same level of cytotoxicity as the medium ( Figure 3A ). We next sought to identify the molecular response in macrophages that were exposed to a non-cytotoxic level of secretome. Transcriptomic analysis revealed that S. aureus alone significantly increased the expression of 93 and decreased 8 ( Figure 3B ), with genes related to M1 macrophage polarization were mostly upregulated ( Supplementary Figure S3 ), and M2 polarization-related genes were hardly affected ( Supplementary Figure S4 ). Consistently, multiple proinflammatory pathways were increased in S. aureus secretome-exposed macrophages ( Figure 3C ). In contrast, C. albicans secretome did not significantly change the transcription of any genes in macrophages ( Figure 3D ), which is in agreement with the cytotoxicity data ( Figure 2 ).

Compared with S. aureus monoculture secretome, THP-1 M0 macrophages exposed to co-culture secretome showed 467 significantly upregulated genes and 87 downregulated genes ( Figure 4A ), including higher transcription of genes related to M1 polarization, pro-inflammatory cytokines and chemokines ( Supplementary Figure S3 ). Despite the activation of macrophages by S. aureus monoculture, 29 pro-inflammatory pathways were further enriched in co-culture secretome-exposed macrophages ( Figure 4B ), with the majority of these pathways (TNF signaling, TLR signaling, NFKB signaling, and cytokine–cytokine receptor interaction) upregulated as a result of exposure to S. aureus secretome versus medium. In agreement, co-culture secretome-exposed macrophages showed enrichment of the NOD2-like receptor signaling pathway, a known activating pathway of these pro-inflammatory pathways. Interestingly, pathways related to ribosomal and oxidative phosphorylation proteins were significantly negatively affected by exposure to co-culture secretome, indicating metabolic downregulation. When compared with C. albicans monoculture secretome, co-culture secretome-exposed macrophages showed 784 upregulated DEGs and 312 downregulated DEGs ( Figure 4C ) and enriched the proinflammatory pathways that are vastly overlapping with those when co-culture vs. S. aureus monoculture ( Figure 4D ).

Reasoning that Als1/Als3 plays an important role in the virulence factor pattern in C. albicans–S. aureus co-culture, we hypothesize that depletion of ALS1/ALS3 will significantly alleviate the proinflammatory effects induced by wild-type C. albicans–S. aureus co-culture. To test this hypothesis, we compared the transcriptome of macrophages exposed to the wild-type co-culture secretome versus C. albicans ALS1/ALS3 ΔΔ/ΔΔ–S. aureus co-culture secretome. Surprisingly, only two genes were significantly changed ( Figure 5A ), indicating that ALS1/ALS3 gene deletion has marginal impact on the gene transcription in macrophages. Compared with S. aureus secretome, ALS1/ALS3 ΔΔ/ΔΔ co-culture secretome showed 97 upregulated genes and 2 downregulated genes in macrophages ( Figure 5B ). Despite that there were less significantly changed genes in ALS1/ALS3 ΔΔ/ΔΔ co-culture versus S. aureus compared with wild-type co-culture versus S. aureus, similar proinflammatory pathways ( Figure 5C ), including the TLR2 and NOD2-like receptor signaling pathways, were enriched in ALS1/ALS3 ΔΔ/ΔΔ co-culture, affirming that ALS1/ALS3 deletion does not affect the proinflammatory effects of the co-culture.

ECVFs are essential for various pathogenic processes of C. albicans, such as hyphal formation, damaging host cells, invasion, and immune evasion. Previously, Als3, Csa2, Rbt4, Sap4, and Sap6 proteins were found to be enriched in the secretome of N-acetylglucosamine-induced hyphal growth over yeast and contribute to the virulence of C. albicans pathogenesis, and transcription of SOD5 was increased in hyphal growth (Naglik et al., 2003; Sorgo et al., 2010; Sorgo et al., 2011; Röhm et al., 2013; Okamoto-Shibayama et al., 2014). Similarly, we showed that S. aureus-promoted C. albicans secreted higher levels of Csa2, Rbt4, Sap4-6, and Sod5. Deletion of RBT4 (Naglik et al., 2003; Martchenko et al., 2004; Sorgo et al., 2010; Sorgo et al., 2011; Röhm et al., 2013; Okamoto-Shibayama et al., 2014), SOD5 (Martchenko et al., 2004), SAP4, and SAP6 ( Braun et al., 2000) significantly attenuates or completely diminishes C. albicans lethality in animal models, partially due to the depletion of RBT4-mediated resistance to leucocyte attack (Naglik et al., 2003; Martchenko et al., 2004; Sorgo et al., 2010; Sorgo et al., 2011; Röhm et al., 2013; Okamoto-Shibayama et al., 2014), SAP4-6-mediated resistance to macrophage killing (Borg-von Zepelin et al., 1998), and SOD-aided protection against intracellular neutrophil killing (Martchenko et al., 2004). In addition to these proteins, we found that hyphae-related proteins (Bailey et al., 1996; Martin et al., 2013), such as Als1, Als3, Ihd1, and Hyr1, were higher in co-cultures over monoculture. This is in agreement with microscopic/morphological characterization, showing increased hyphae in co-culture (data not shown). Furthermore, the co-culture promotes the level of Xog1 in the secretome. Xog1 is an exo-1,3-beta-glucanase essential for reducing beta-glucan epitope exposure (β-glucan masking) to immune cells, hence reducing the phagocytotic interaction and enhancing immune evasion (Ballou et al., 2016; Childers et al., 2020). Two other cell wall crosslinking enzymes Crh11 and Utr2 were also upregulated during co-culturing. Crh11 and Utr2 were shown to impact β-glucan masking mildly (Ballou et al., 2016). Together, these results support the contribution of these proteins in C. albicans–S. aureus co-culture to the pathogenesis during coinfection.

Similarly to C. albicans, the co-culturing also promotes the virulence potential of S. aureus by increasing the secretary level of cytolytic, proteolytic, or lipolytic proteins. We found that C. albicans-mediated pH maintenance is a critical factor in regulating these proteins. This is consistent with previous observations, where C. albicans tended to maintain a neutral pH during co-culturing ( Supplementary Figure S1A ) (Todd et al., 2019b; Pasman et al., 2024) and, thereby, promoted the production and secretion of alpha hemolysin (hla) (Todd et al., 2019b), and hlb and hlg. This is likely due to the C. albicans-mediated activation of the P3 promoter of the S. aureus agr system, which increases the expression of RNAIII that is essential for the production and secretion of these hemolysing toxins (Novick et al., 1993; Dunman et al., 2001). While we and others confirmed the contribution of pH in regulating S. aureus virulence factors, other unknown C. albicans-derived factors also contributed to the elevated virulence potential of S. aureus. For example, the ECVFs were significantly decreased in buffered monocultures over unbuffered monoculture but significantly increased in co-culture over unbuffered monoculture. These results further highlighted the nuanced regulation of the secreted virulence factors and further studies are warranted to identify these C. albicans-derived factors.

As many virulence factors from both C. albicans and S. aureus were substantially increased in co-culturing, it is not surprising that the secretome from co-culture exhibited higher cytotoxicity toward Ca 9–22 and HO1N1 cells compared with that from the monocultures. Consistently, C. albicans–S. aureus co-culturing showed higher cytotoxicity toward keratinocytes NOK-si and HaCat cells, albeit to a different extent (de Carvalho Dias et al., 2017). This discrepancy might be due to the different sensitivity of these cell lines.

An important strategy for pathogenic microbes is to exploit host heme/iron sources during the infection. The increase in both C. albicans and S. aureus iron acquisition potential may also explain the high lethality of co-infection in vivo in animal models (Carlson, 1983). For C. albicans, Csa2 is critical in the uptake of hemoglobin and heme proteins and their utilization as an iron source (Okamoto-Shibayama et al., 2014) and this process is assisted by Rbt5, Pga7, Frp1, and Frp2 (Kuznets et al., 2014; Nasser et al., 2016; Roy et al., 2022). We found that co-culturing increases the secreted level of Csa2, but not Rbt5, Pga7, Frp1, and Frp2 in C. albicans. For S. aureus, the Isd (iron-regulated surface determinant) system is essential for hemoglobin and heme binding, uptake, and iron release (Caza and Kronstad, 2013). Of all the isd proteins, the levels of cell wall anchored isdA, isdB, isdC, isdE, and isdH, but not intracellular isdI and isdG, were significantly higher in co-culture, supporting the potential contribution of these proteins in the in vivo pathogenesis. Nonetheless, a previous study found that none of isd proteins was significantly changed in co-culture over monoculture (Peters et al., 2010).

While we systemically elucidated the reciprocal effects of C. albicans and S. aureus on their virulence factor secretion, cytotoxicity, and proinflammatory effects, there are several limitations of this study. First, the co-culturing conditions did not perfectly match the in vivo microenvironment, despite our effort to develop a completely defined medium to remove unknown artificial co-founding factors in the culturing media (Pasman et al., 2024). Future investigation may consider more physiologically relevant in vitro systems such as the reconstructed human gingiva–microbe interaction model (Zhang et al., 2022). In addition, all the culturing was performed under an oxygenated environment. Similar to our study, most studies carried out in vitro C. albicans cultures in a regular CO₂ incubator that maintains 5% CO₂. These conditions do not necessarily represent the in vivo physiology, since commensal C. albicans cells reside in the oral cavity and large intestine, where the oxygen tension can be relatively low. Furthermore, it must be noted that lowered biofilm DNA concentrations were observed regarding the C. albicans ALS1/ALS3 ΔΔ/ΔΔ conditions with respect to its wild-type strain ( Supplementary Figure S1 ). However, no differences were observed between C. albicans wild-type and ALS1/ALS3 ΔΔ/ΔΔ in secretome protein concentrations ( Supplementary Figure S1 ). Finally, we used THP-1-derived macrophages instead of primary macrophages or primary monocyte-derived macrophages. Despite that THP-1 cells are widely used to model human monocytes and macrophages (Mohd Yasin et al., 2022), it has been shown that THP-1 macrophages may show different response patterns as the primary macrophages (Hoppenbrouwers et al., 2022).