Enhanced Virulence of Candida albicans by Staphylococcus aureus: Evidence in Clinical Bloodstream Infections and Infected Zebrafish Embryos

Coinfection with Candida and Staphylococcus results in higher mortality in animal studies. However, the pathogenesis and interplay between C. albicans and S. aureus in bloodstream infections (BSIs) is unclear. This study determines the clinical features and outcomes of mixed C. albicans/S. aureus (CA/SA) BSIs and biofilm formation on pathogenesis during coinfection. Demographics and outcomes for mixed BSIs and monomicrobial candidemia were compared. Compared to 115 monomicrobial C. albicans BSIs, 22 patients with mixed CA/SA BSIs exhibited a significantly higher mortality rate and shorter survival time. In vitro and in vivo biofilm analysis showed that C. albicans accounted for the main biofilm architecture, and S. aureus increased its amount. Antibiotic tolerance in S. aureus, which adhered to Candida hyphae observed by scanning electron microscope, was demonstrated by the presence of wild-type C. albicans co-biofilm. Upregulation in exotoxin genes of S. aureus was evidenced by quantitative RT-PCR when a co-biofilm was formed with C. albicans. Mixed CA/SA BSIs result in a higher mortality rate in patients and in vivo surrogate models experiments. This study demonstrates that the virulence enhancement of C. albicans and S. aureus during co-biofilm formation contributes to the high mortality rate.


Introduction
The global SENTRY Antimicrobial Surveillance Program from 2012-2017 demonstrated that Staphylococcus aureus, which was the most prevalent bacterial pathogen, caused 22.5% of bloodstream infections (BSIs), and Candida species caused 3.1% [1], with increasing prevalence during the last decade [2]. Mixed BSIs with bacteria and Candida results in a more severe prognosis and are associated with higher morbidity and mortality than monomicrobial or polybacterial BSIs [3][4][5].
S. aureus and C. albicans are both important lethal pathogens in nosocomial BSIs, and there is an increase in the incidence of mixed BSI, so the risk factors and prognosis of mixed C. albicans/S. aureus (CA/SA) BSIs in the population and the interplay between C. albicans and S. aureus in host pathogenesis are worthy of investigation.
Coinfection models involving Candida and Staphylococcus have been used in several animal studies [6][7][8][9][10][11][12]. These studies all show that mixed-infection with Candida and Staphylococcus increases mortality of the infected hosts, and the interplay between the bacterium and the fungus significantly increases the bacterial burden or virulence on hosts. It has been demonstrated that a biofilm of Candida protects bacteria from host defensiveness, and each microbial secretory factor affects their virulence [13,14].
Embryos of zebrafish are commonly used as a surrogate infection host and are used as a model to determine the biofilm activity of C. albicans in vivo. This model allows antifungal screening and genetic studies with the advantages of conserved immunity, prolific fecundity, and cost-effectiveness [15][16][17][18]. In order to determine the interplay between C. albicans and S. aureus in infected hosts without immune interference, a zebrafish embryo infection assay was used for this study.

Study Design and Definition
This retrospective study was conducted from 2003 to 2015 at Chang Gung Memorial Hospital at Linkou (CGMHL) and was approved by the institutional review board of the hospital (201701182B0 and 202101345B0). Adult hospitalized patients (≥18 years old) were recruited. Mixed CA/SA BSI cases were defined by a positive blood culture for C. albicans with S. aureus growing concomitantly from the same set. Data from adult patients who had monomicrobial C. albicans BSIs in 2010 were used for comparison.

Clinical Characteristics and Outcomes
Data for demographics, comorbidities, risk factors, and clinical outcomes were recorded. Comorbidities included heart failure, chronic lung diseases, diabetes mellitus, hepatic dysfunction (a serum total bilirubin level ≥ 2.0 mg/dL or liver cirrhosis), renal insufficiency (a serum creatinine level ≥ 2.0 mg/dL or a requirement for hemodialysis), hematological malignancies, and solid tumors. The risk factors collected within 30 days before the BSIs included abdominal surgery, central venous catheterization, parenteral nutrition, immunosuppressant administration (prednisolone ≥ 20 mg/day for more than 3 weeks, chemotherapeutic and other immunosuppressive drugs), and neutropenia (an absolute neutrophil count < 500 cells/mL). Severity of illness was assessed at the onset of BSIs using a sequential organ failure assessment (SOFA) score [19]. Clinical outcomes were assessed in terms of mortality, median survival days after BSIs, and length of stay in hospital. The clinical characteristics and outcomes for C. albicans candidemia were described in a previous study by the authors [20].

Microbes and Fish Strains
To evaluate the virulence enhancement of C. albicans biofilm by S. aureus, a virulent laboratory strain SC5314 and its biofilm-defective mutant HLC54 were chosen as well as a hemolysin-producing S. aureus ATCC 29213, which is the most commonly used strain in the clinical laboratory. C. albicans HLC54 (cph1/cph1 efg1/efg1) is a hyphae-defective mutant derived from virulent parental SC5314 [21]. A clinical isolate P004 with low biofilm was also used as the fungal pathogen [17]. All yeasts are caspofungin susceptible. A methicillinsusceptible S. aureus ATCC 29213 was used as the bacterial pathogen, which is vancomycin susceptible [22]. Wild-type zebrafish (Danio rerio) AB line (Zebrafish International Resource Center, Eugene, OR, USA), aged approximately 8-15 months, were maintained at 28 • C in a 10-h dark 14-h light cycle to collect fertilized eggs, which served as surrogate hosts.
All experiments were performed in accordance with standard guidelines for zebrafish studies [23].

In Vitro Biofilm Analysis
3-Morpholinopropane sulfonic acid (MOPS)-buffered RPMI-1640 (pH7.2) containing 10% fetal calf serum and 1.25% N-acetyl-D-glucosamine was used as a biofilm formation medium. For the metabolic reducing dye assay, overnight-cultivated microbes were refreshed for 4 h, 10 5 cfu of the cultures were seeded into microplates, and then, biofilm that was cultured at 37 • C for 4-h or 24-h was measured using a CCK-8 kit (cat. 96992, Sigma-Aldrich, St. Louis, MO, USA). For the other biofilm quantification method using safranin, sterile 13-mm hydrophilic filters (cat. AAWP01300, Millipore, Bedford, MA, USA) were used as an adherent surface, and biofilm-stained Gram's safranin was dissolved using 30% acetic acid and measured using OD 530 .

Zebrafish Embryo Infection Assay
The infection assay follows the method previously described [17]. Embryos were co-incubated with 5 × 10 5 yeast/mL C. albicans and/or 2 × 10 7 cfu/mL S. aureus in RPMI-1640 medium (Thermo Fisher Scientific, Waltham, MA, USA) at 120 rpm 30 • C for 4 h. After 4-h co-incubation, the infected embryos were washed thrice with egg water (0.03% sea salt) to remove non-adherent microbes and then transferred into fresh egg water supplemented with 0.5% YPD broth and were grown at 30 • C for 24 h. The survival rate was calculated as the percentage of the death by monitoring the heartbeat of embryos. All experiments were approved by the Biosafety Committee (No. 00417-2020092830581) and IACUC (CGU109-109) of the hospital and were conducted in a BSL-2 laboratory.

Microbial Enumeration
After 4-h adhesion, the microbial infected chorions were incubated for 20 h to allow biofilm maturation and were treated with antimicrobials (80 µg/mL vancomycin or 0.5 µg/mL caspofungin). The chorions were washed and transferred to microtubes and ground using a pestle. The resulting mixtures were colony counted for microbes on suitable agar plates with vancomycin (4 µg/mL) or caspofungin (4 µg/mL).

Scanning Electron Microscope (SEM)
Samples were washed and harvested after 24-h incubation and fixed in a mixture of 3% gutaraldehyde and 2% paraformaldehyde that was buffered to pH 7.4 using 0.1 M cacodylate buffer. Samples were washed with cacodylate buffer and treated with 1% osmium tetroxide. The cell pellets were washed thoroughly with 0.1 M cacodylate buffer and dehydrated gradually with ethanol until only 100% ethanol remained. After criticalpoint drying and mounting on an SEM stub, the samples were sputter-coated with a thin layer of gold, and images were recorded at an appropriate accelerating voltage.

Gene Expression Assays
The biofilm mass was dissolved and ground in REzol C&T reagent (Protech Technology, Taipei, Taiwan) and kept frozen before RNA extraction. To determine the expression of S. aureus toxin genes in the biofilm, total RNA extracted with MagNA Pure Compact RNA Isolation Kit (Roche Applied Science, Indianapolis, IN, USA) was two-step reversetranscribed (High Capacity cDNA Reverse Transcription Kit, Applied Biosystems, Foster City, CA, USA) to cDNA and analyzed by quantitative PCR run in ABI 7900 HT Real-Time PCR System: 10 min at 95 • C and 40 cycles of 15 sec at 95 • C, 1 min at 60 • C using 2× Gene Expression Master Mix (Applied Biosystems, Foster City, CA, USA). The primers (Mission Biotech, Taipei, Taiwan) and TaqMan probes (Thermo Fisher Scientific, Waltham, MA, USA) corresponding to S. aureus housekeeping gene rrsA and hemolysin genes hla and hlgB are listed in Table S1. The change in the expression was calculated using the comparative CT method, with rrsA as an endogenous control.

Statistical Analysis
Statistical analysis used SPSS 22.0 software (IBM, Armonk, NY, USA). Continuous variables are presented as mean ± standard deviation (SD) if the data are normally distributed and as a median and interquartile range (IQR) if the data are non-normally distributed. Categorical variables were compared using Pearson's chi-square test or Fisher's exact test and continuous variables were compared using a Student's t-test or a Mann-Whitney U test. Variables with a two-tailed p-value < 0.01 were included in a binary logistic regression model for a multivariate analysis. All tests were two-tailed, and a p-value of <0.05 represents statistical significance. Patients who did not survive more than 14 days were correlated with liver dysfunction (39.7% vs. 13.5%, p < 0.001), renal insufficiency (52.4% vs. 14.9%, p < 0.001), higher SOFA score (11 vs. 4, p < 0.001), and more mixed CA/SA BSIs (27.0% vs. 6.8%, p = 0.001) ( Table 2). Multivariate analysis showed that 14-day mortality was positively associated with a high SOFA score (OR 1.13, 95% CI 1.03-1.24) and mixed CA/SA BSIs (OR 3.47, 95% CI 1.05-11.50) ( Table 3).

Co-Biofilm of C. albicans and S. aureus In Vitro
The in vitro biofilm activity assay was performed using a metabolic reducing dye and Gram's safranin staining. The results for the two biofilm assays showed that there was significantly higher activity in the mixed-infection group than in the monomicrobial groups ( Figure 1). The difference of biofilm in 4-h adhesion between the mixed-infection group and others was more significant than that for the 24-h maturation, and the results showed that C. albicans accounted for most biofilm structure in terms of both the activity and the amount of biofilm. β-glucan in culture supernatants of C. albicans has been shown to potentiate drug resistance of S. aureus [24]. Biofilm enhancement in microbial supernatants was also studied, and a significant increase in S. aureus biofilm with C. albicans supernatants was observed in terms of the activity and the amount ( Figure S1). Figure 1. Analysis of in vitro microbial biofilm with C. albicans and S. aureus. Using microplates or hydrophilic membrane filters as an adherent surface, 10 5 cfu of C. albicans and/or S. aureus strains were seeded for 4-h adhesion and 24-h maturation of biofilms at 37 • C and analyzed using a CCK-8 kit (A) or Gram's safranin dye staining (B). Note: * p < 0.05, ** p < 0.01, *** p < 0.001, **** p < 0.0001, ns = not significant.

Enhancement of Virulence of C. albicans Biofilm by S. aureus on Zebrafish Embryos
Within 4-h incubation, the microbial loads for S. aureus and C. albicans SC5314 grown on the surface of egg chorion were, respectively, calculated to be 1.5 × 10 6 cfu and 3.5 × 10 6 cfu per egg (data not shown). After 24-h development, no matter whether S. aureus was present, heavy and thick biofilms were observed on chorions of the SC5314 group and the lowbiofilm P004 group ( Figure 2K,L,O,P), but there was no obvious biofilm observed in neither S. aureus nor hyphae-defective HLC54-infected groups ( Figure 2J,M,N).
Co-cultivation of C. albicans significantly increased the bacterial load in the co-biofilm on eggs from 2.3 × 10 7 to 6.9 × 10 7 in the SC5314 group and 2.6 × 10 8 in the HLC54 group (Table 4). Significantly fewer embryos in the co-infected groups survived and this reduction was much greater for wild-type SC5314: from 35% to 7% ( Figure 2L,P). After 4-h adhesion, bound microorganisms, including S. aureus ATCC 29213, C. albicans hyphaedefective mutant HLC54, low biofilm clinical strain P004, and wild-type SC5314 strains, were cultivated at 30 • C for 24 h to form mature biofilms (A-P). Photographs were taken during the 4-h adhesion and 24-h maturation periods. The survival rate for infected embryos was measured and is shown in parentheses with the number (n) of eggs that was used for each experiment. Around 30 embryos were used for each test. The heart-rate of fish embryos was measured to determine the survival rate.

Protection of S. aureus against Antibiotics in a Co-Biofilm with Hyphal C. albicans
The S. aureus loads after vancomycin treatment were significantly increased in cobiofilm with wild-type SC5314, while the protection declined when formed with a hyphaedefective HLC54 (Table 4). However, no obvious protective effect was observed in the co-biofilms for C. albicans against caspofungin, either SC5314 or HLC54 (data not shown). The scanning electron microscope (SEM) results showed that S. aureus almost adhered to C. albicans hyphae (Figure 3), but yeast cells did not express molecules for intimate interaction with S. aureus to shield from drug damage. There was no obvious change in the biofilm morphology for co-infected embryos that were treated with vancomycin, which eliminated non-hyphal protected S. aureus (Figure 4).  . The morphology of mixed biofilm that are treated with S. aureus and C. albicans under vancomycin. After 4-h adhesion and 24-h maturation, fish embryos that were infected with S. aureus ATCC 29213, a hyphae-defective mutant HLC54, or wild-type SC5314 strains individually or mixed infected were photographed using a stereotactic microscope. Rows 2 and 4 share the same preparation of infected embryos but involve supplementation of 80 µg/mL vancomycin.

Changes in S. aureus Toxin Genes in A Co-Biofilm
S. aureus toxin genes were augmented by C. albicans in murine infection models, including hla (alpha hemolysin) and hlgB (gamma hemolysin) [25]. If S. aureus formed cobiofilms with C. albicans on a filter membrane after 24-h maturation, there was a significant increase in expression of hla (7.06 ± 2.19 fold increase) and hlgB (8.12 ± 2.64 fold increase) in SC5314 group but not in hyphae-defective HLC54 group (hla: 1.54 ± 0.30 and hlgB: 2.09 ± 0.81) ( Figure 5). Figure 5. The upregulation of S. aureus toxin genes hla and hlgB in a mixed biofilm that is formed by S. aureus and C. albicans SC5314. After 24-h maturation of biofilms on filter membranes, the in vitro biofilms that were formed with S. aureus only (SA, white bars), S. aureus co-cultivated with C. albicans SC5314 (SA + SC5314, gray bars), or with hyphae-defective mutant HLC54 (SA + HLC54, black bars) were washed twice, and total RNA was isolated and reverse transcribed to determine gene expression. Note: * p < 0.05, ** p < 0.01, ns = not significant.

Discussion
This study demonstrates that the clinical outcome strongly supports the results for an in vitro enhanced virulence for candidemia by S. aureus. Within the monomicrobial C. albicans BSI group, more patients were found with parenteral nutrition and immunosuppressants, so this population was more immunocompromised. However, in comparison with this group, 22 mixed CA/SA BSI patients, enrolled from 264 mixed CA/bacterial BSIs (2003)(2004)(2005)(2006)(2007)(2008)(2009)(2010)(2011)(2012)(2013)(2014)(2015), were associated with a higher SOFA score, increased 14-day and 30-day mortalities, and shorter median survival days. This demonstrates that S. aureus increases the virulence of C. albicans in BSI patients.
Zhong et al. reported that patients with mixed Candida albicans/bacterial bloodstream infections in China undergo mechanical ventilation for longer and stay longer in the ICU than patients with monomicrobial C. albicans BSIs, but there is no difference in 28-/60-day mortality [5]. The difference in mortality for mixed BSIs in different studies may be due to different bacterial pathogens. Staphylococcus is associated with higher mortality for animal hosts [6][7][8][9][10][11], but Enterococcus spp. negatively affects each other's virulence [26][27][28].
In terms of comorbidity, there is a higher incidence of renal insufficiency in mixed BSI group (Table 1). Patients who did not survive for 14 days had a higher prevalence of liver dysfunction and renal insufficiency and had almost four-times higher incidence of mixed CA/SA BSIs ( Table 2). The authors demonstrated previously that a higher prevalence of renal insufficiency was associated with high-biofilm C. albicans isolates [17]. Herein, in vivo zebrafish embryo infection assay exhibited embryos co-infected with S. aureus and high-biofilm C. albicans survived less. Therefore, renal insufficiency is closely associated with mixed CA/SA BSI. In a 12-year period autopsies review, the most frequently involved deep parenchymal organ in candidiasis was the kidney [29]. Shin et al. reported that severe renal failure plays a major role in lethality for mice infected with C. albicans [30]. In an intravenous mouse infection model, kidneys were demonstrated to be the main target of C. albicans, though the upregulation of hyphae-associated genes was measured in the liver without visible hyphal invasion [31].
Our study results show that S. aureus significantly increases biofilm formation of C. albicans in the in vitro abiotic assay and in vivo surrogate zebrafish model. The increased biofilm of C. albicans and enhanced virulence of S. aureus are the reason for higher mortality in mixed BSIs. For the biofilm model with zebrafish eggs, only wild-type C. albicans provides S. aureus with protection against vancomycin rather than hyphae-defective mutant (HLC54). SEM images show that there is tight contact between the hyphae of C. albicans and S. aureus in a thick and multilayer biofilm, in which there is a matrix that is composed of an entangled polymeric substance during maturation that shields against antimicrobials.
This study shows that the SOFA score was significant higher within mixed CA/SA BSI patients, so interaction between these two pathogens probably exacerbates tissue damage and erosion during hyphal invasion. Our embryo infection result suggested C. albicans cobiofilm protects S. aureus against antimicrobials, and the major toxin genes of S. aureus are significantly upregulated, including hla and hlgB, resulting in high mortality for C. albicans SC5314-infected embryos. An upregulation of virulence factors in S. aureus, including hla, hlgB, enterotoxin family protein, staphylocoagulase, staphylococcal protein A, nucleases, intercellular adhesion proteins, fibronectin-binding proteins, drug-resistant genes glmU, murC, and murD, and penicillin-binding proteins, was demonstrated by Hu et al. using cutaneous abscess and peritonitis murine models [25]. The increase in toxin production, especially exterotoxins, damages tissues and induces severe inflammation, which can cause multi-organ failure.
The study has some limitations. First, we acknowledge that insufficient information about follow-up blood cultures to confirm pathogen clearance was obtained in CA/SA BSI patients in this single-center retrospective study. However, these were critically ill patients with high mortality, and 12 of them (12/22 = 54.5%) died within seven days after mixed CA/SA BSIs occurred. Second, the zebrafish has recently become an extremely powerful model organism in the context of Candida infections and host-pathogen interactions [32]. However, the drawback of the model is that the embryos were collected and grown at 30 • C, which does not allow accurate mimicking of human infection.

Conclusions
Mixed CA/SA BSIs are associated with high morbidity and mortality in our hospital. The zebrafish embryo in vitro infection assay exhibits antibiotic tolerance in S. aureus if it attaches to the hyphae of C. albicans. Using in vitro assays, the production of S. aureus toxins in Candida co-biofilm possibly involves pathogenesis.
Supplementary Materials: The following are available online at https://www.mdpi.com/article/10 .3390/jof7121099/s1, Table S1: The primer sets and TaqMan probes for the gene expression assay; Figure S1: The enhancement of S. aureus biofilm using culture supernatants of C. albicans.