Therapeutic implications of C. albicans-S. aureus mixed biofilm in a murine subcutaneous catheter model of polymicrobial infection

ABSTRACT Biofilm-associated polymicrobial infections tend to be challenging to treat. Candida albicans and Staphylococcus aureus are leading pathogens due to their ability to form biofilms on medical devices. However, the therapeutic implications of their interactions in a host is largely unexplored. In this study, we used a mouse subcutaneous catheter model for in vivo-grown polymicrobial biofilms to validate our in vitro findings on C. albicans-mediated enhanced S. aureus tolerance to vancomycin in vivo. Comparative assessment of S. aureus recovery from catheters with single- or mixed-species infection demonstrated failure of vancomycin against S. aureus in mice with co-infected catheters. To provide some mechanistic insights, RNA-seq analysis was performed on catheter biofilms to delineate transcriptional modulations during polymicrobial infections. C. albicans induced the activation of the S. aureus biofilm formation network via down-regulation of the lrg operon, repressor of autolysis, and up-regulation of the ica operon and production of polysaccharide intercellular adhesin (PIA), indicating an increase in eDNA production, and extracellular polysaccharide matrix, respectively. Interestingly, virulence factors important for disseminated infections, and superantigen-like proteins were down-regulated during mixed-species infection, whereas capsular polysaccharide genes were up-regulated, signifying a strategy favoring survival, persistence and host immune evasion. In vitro follow-up experiments using DNA enzymatic digestion, lrg operon mutant strains, and confocal scanning microscopy confirmed the role of C. albicans-mediated enhanced eDNA production in mixed-biofilms on S. aureus tolerance to vancomycin. Combined, these findings provide mechanistic insights into the therapeutic implications of interspecies interactions, underscoring the need for novel strategies to overcome limitations of current therapies.


Introduction
Polymicrobial infections caused by a combination of microorganisms are responsible for significant mortality and morbidity [1][2][3]. Biofilm-associated mixed infections, such as those on indwelling medical devices, are particularly challenging to treat due to their inherent heterogeneity, which can be further amplified by synergistic inter-species interactions [1,[4][5][6]. Therefore, understanding the molecular inter-species interactions will greatly aid in designing novel therapeutic strategies targeting mixed biofilm infections, specifically those involving fungi and bacteria.
Candida albicans is the most common human fungal pathogen causing diseases ranging from superficial mucosal to life-threatening systemic infections [7][8][9].
The majority of C. albicans infections are associated with its ability to form biofilms where adhesion of yeast cells to the substrate is followed by proliferation and formation of hyphae, resulting in a network of cells embedded in an extracellular matrix [10][11][12][13]. C. albicans biofilm matrix is complex, with major polysaccharide constituents being α-mannan, β-1,6-glucan, and β-1,3-glucan which has been linked to resistance of biofilm to antifungals [14][15][16][17]. However, in various niches in the host, C. albicans coexists with numerous bacterial species, most notably Staphylococcus aureus a major human bacterial pathogen that causes a wide variety of diseases [18][19][20][21][22][23][24]. S. aureus alone is a poor former of biofilms, however, together with C. albicans, this species forms a substantial biofilm where the fungus creates a scaffold for the bacteria [6,25]. The growing use of implanted medical devices is another reason why the incidence of Candida and staphylococcal infections has steadily increased, since the majority of these infections are emerging from biofilms formed on medical implants [26,27].
Our previous characterization of mixed C. albicans and S. aureus biofilms in vitro and in vivo demonstrated that S. aureus exhibits high affinity to the C. albicans hyphal form as these species co-adhere and interact synergistically, forming a dense and architecturally complex biofilm [25,28,29]. Furthermore, using a mouse model of oral co-infection, we demonstrated that onset of oral candidiasis may lead to systemic staphylococcal infection, stemming from a thick matrix of mixed biofilm formed on tongue tissue [30,31]. Our subsequent in vitro studies into the therapeutic implication of C. albicans-S. aureus interactions in biofilm demonstrated that C. albicans secreted matrix polysaccharides, primarily β-1,3-glucan, confers S. aureus with enhanced tolerance to antimicrobials by impeding drug penetration through the biofilm [29]. However, findings from the study also indicated that other factors may also contribute to the enhanced S. aureus tolerance to antimicrobials, which we subsequently identified to be farnesol, a quorum-sensing molecule secreted by C. albicans [32]. Specifically, we demonstrated that farnesol indirectly confers S. aureus tolerance to vancomycin by elicitng a stress response via activation of the thiol-based redox system.
In addition to secreted proteins and carbohydrates one major distinguishing feature of biofilms is the presence of secreted extracellular DNA (eDNA), particularly in staphylococcal biofilms where it is crucial for biofilm structural integrity and maintenance [33][34][35]. In fact, the extracellular matrix of S. aureus biofilm is primarily composed of eDNA and the polysaccharide intercellular adhesin (PIA) [36,37]. The formation of the eDNA matrix appears to be necessary for early staphylococcal biofilm development, as early DNase treatment profoundly impairs biofilm formation [33]. In S. aureus, the generation of the eDNA matrix is facilitated by autolysis of a subset of initial biofilm cells, regulated by the cidABC and lrgAB operons [38--38-40]. Gene regulation in S. aureus biofilms is complex and involves the activity of several major global regulators [41,42]; however, the environment in a biofilm can be altered by the presence of coinfecting microorganisms where polymicrobial interactions can result in augmented pathogenesis. To that end, the overall goal of this study was to elucidate the impact of inter-species interactions during a biofilmassociated infection in a host. Specifically, we used an established mouse model for in vivo grown biofilms to demonstrate the therapeutic implications of deviceassociated polymicrobial infections. Further, we aimed to provide lacking insights into the transcriptional changes governing the interactions between these diverse pathogens, particularly within the context of antimicrobial resistance.
Strains and growth conditions. The Candida albicans wild-type strain SC5314 [43] and the methicillinresistant Staphylococcus aureus strain (USA300 LAC) were used in this study. S. aureus mutant derivatives of the reference WT strain defective in cell autolysis regulation (∆lrgA and ∆lrgB) were obtained from the Nebraska Transposon Mutant Library (University of Nebraska, NE). C. albicans was grown on yeast peptone dextrose broth (YPD) (Difco Laboratories) overnight at 30°C and S. aureus cultures were grown overnight in trypticase soy broth (TSB) (Difco) at 37°C. Cells were washed in Phosphate Buffered Saline (PBS) and resuspended in RPMI 1640-HEPES (Invitrogen) media to final cell density needed.
In vitro S. aureus vancomycin susceptibility testing in single and mixed biofilms. Single-and mixedspecies biofilms were grown in the wells of 96-well polystyrene flat-bottom plates as we previously performed [29]. C. albicans and S. aureus cell suspensions were adjusted to a final concentration of 1 × 10 6 cells/ ml in RPMI 1640-HEPES medium and 100 µl of cell suspensions were seeded in the wells individually or in combination. Plates were incubated for 90 mins at 37° C, then washed with PBS to remove non-adherent cells. Fresh medium (100 µl) was added to each well, and biofilms were allowed to form for 24 hrs at 37°C. Following incubation, wells were gently washed with PBS to remove non-sessile cells. Vancomycin (800 µg/ ml) was added to wells, and plates were incubated for 24 hrs at 37°C. Following incubation, wells were sonicated to disrupt biofilms, and cell suspensions were diluted and plated on S. aureus-specific and Candida-specific chromogenic medium (CHROMagar; DRG International, Inc.) for CFU (colony forming units) enumeration. Control wells with no vancomycin were included; vancomycin concentration used was determined based on our previously published work on mixed biofilm assays [29].
Animal studies. All animal experiments were conducted at the AAALAAC accredited Animal Facility of the University of Maryland, Baltimore and were approved by Animal Care and Use Committee. Threemonth-old female Balb/c mice were purchased from Envigo; animals received food and water ad libitum and were housed at a maximum of 5 mice per cage, weighed and closely monitored for any signs of distress throughout experimental periods.
Polymicrobial catheter infection mouse model. The previously described rodent model [44,45] was adopted, with modifications. Mice were immunosuppressed by the addition of dexamethasome (Sparhawk Laboratories; Lenexa, KS) in drinking water at final concentration of 0.4 mg/L one day prior to infection, which was maintained throughout the study. 5-8 fragments of polyurethane triple-lumen central venous catheters (1 or 0.5 cm in length) (Jorgensen Laboratories; Loveland, CO) were pre-coated overnight with fetal bovine serum (Gibco™; Waltham, MA) at 37° C. Catheters were then inoculated with single or mixedspecies cell suspensions of S. aureus and/or C. albicans in RPMI 1640-HEPES at final concentrations of 1 × 10 6 cells/ml for 90 mins at 37°C to allow for microbial adhesion to catheters. For each experimental set, in vitro-infected catheters were processed for assessment of microbial recovery to confirm standardized microbial adherence to catheters prior to implantation. For implantation in animals, following in vitro infection, fragments were rinsed with PBS and kept on ice until implanted. Mice were anesthesized by intraperitoneal injections (0.5 ml) of TBE solution (250 mg/kg body weight) and the dorsum was shaved; small incisions were aseptically made (Figure 2a-i), and a subcutaneous tunnel created allowing for insertion of 5-8 pieces of catheters (Figure 2a-ii). Incisions were sealed using 3M Vetbond™ tissue glue and lidocaine analgesic gel was applied to the incision area (Figure 2a-iii). Animals were placed in a supine position under a heating lamp and monitored until they recovered from anesthesia, and then daily for any developing clinical signs of distress. Animals were euthanized by CO 2 inhalation followed by cervical dislocation and catheters were harvested individually.
Scanning electron microscopy (SEM) analysis of explanted catheters for evaluation of in vivo-grown biofilms. In order to visualize the biofilms formed within the lumen of the catheters, representative fragments from each group were processed for SEM analysis. Explanted catheters were rinsed with PBS then fixed in 2% paraformaldehyde, 2.5% glutaraldehyde in phosphate buffer (pH 7.4). Catheters were post-fixed with 1% osmium tetroxide in PBS for 1 hr, washed with buffer, dehydrated in a series of graded ethyl alcohol, 30%, 50%, 70%, 90%, and 3 times in 100% for 10 mins each. Specimens were critical-point dried in CO 2 , mounted on SEM pin mounts, and sputter coated with platinum-palladium (EMS 150 T ES). SEM images were acquired using a Quanta 200 scanning electron microscope (FEI Co., Hillsboro, OR), and images were processed using Adobe Photoshop software.
Vancomycin treatment of mice with single and mixed species catheter infection. To comparatively evaluate the therapeutic efficacy of vancomycin against S. aureus in animals with single or mixed catheter infections, animals were divided into a control (no therapy) group and experimental (therapy) group; within each group, 3 subgroups were included: implanted with catheters infected with S. aureus, C. albicans or in combination for a total of 6 groups with 5 animals in each group. Animals in the treated group were administered vancomycin twice daily by intraperitoneal injections (0.25 ml) at 55 mg/kg, and treatment was initiated on the 3 rd day post-catheter implantation and continued for 5 days, then animals were euthanized. Untreated control groups received sterile PBS injections, and mice with C. albicans monoinfected catheters were also treated with vancomycin as control. Harvested catheters were aseptically fragmented, sonicated in sterile PBS to detach biofilms and cell suspensions were diluted and plated on species-specific chromogenic media for CFU enumeration; each harvested catheter was processed individually. As pilot experiments indicated blockage of the catheters by host tissue and immune cells compromising vancomycin delivery into catheter lumen, therapy experiments were performed using longitudinally cut catheters implanted with lumen facing downwards, in direct contact with the tissue. Experiments were performed on 5 separate occasions and data combined.
Global transcriptional analysis of single and mixed-species catheter biofilms using RNAsequencing. To ensure sufficient RNA yield, initial experiments were performed to determine optimum cell densities for catheter inoculation prior to implantation. Based on measurement of RNA from extracted catheters, cell densities of 1 × 10 8 and 5 × 10 8 cells/ml for C. albicans and S. aureus, respectively, were selected for the initial in vitro catheter inoculation. Implantation of catheters was performed as described above and animals were euthanized on the 3rd-day post-catheter implantation; catheters were collected in RNAlater buffer, aseptically fragmented, sonicated in RNAse free water to detach biofilms, and cells from all catheters recovered from each mouse were pooled by centrifugation. In parallel, prior to extraction, suspensions were diluted and plated for CFU enumeration to assess microbial recovery. Pooled cells were snap-frozen on dry ice and ethanol, allowed to thaw at room temperature (RT), then incubated for 30 mins at 37°C in "digestion buffer" containing 200 µg/ml of lysostaphin (for S. aureus cell wall digestion) and/or 100 U/ml of lyticase (for C. albicans cell wall digestion) and RNAse inhibitor in TRIS-EDTA 1x buffer. RNA was extracted in 1 ml of TRI Reagent™ Solution (Ambion, Invitrogen; Carlsbad, CA) using bead-beating for 30 mins at RT followed by purification using Direct-zol RNA Miniprep kit (Zymo Research; Tustin, CA). Eluted RNA was analyzed in a Nanodrop Lite (Thermo Scientific). Total RNA was subjected to rRNA depletion with the Illumina Ribominus Depletion kit (bacteria module) and/or the Ribominus Eukaryote Kit. All RNA-seq libraries (strand-specific, paired end) were prepared with the TruSeq RNA sample prep kit (Illumina). Seventy-five nucleotides of the sequence were determined from both ends of each cDNA fragment using the HiSeq platform (Illumina). Sequencing reads were aligned to the reference genomes (Staphylococcus aureus USA300 TCH1516 uid58925; C. albicans strain SC5314) using HISAT [46] and alignment files were used to generate read counts for each gene; statistical analysis of differential gene expression was performed using the DEseq package from Bioconductor [47]. A gene was considered differentially expressed the p-value for differential expression was less than 0.05. Two biological samples from each experimental group were analyzed. Sequencing reads were aligned to the reference genomes (Staphylococcus aureus USA300 TCH1516 uid58925, C. albicans strain SC5314) In vitro evaluation of biofilm formation of the S. aureus lrgA and lrgB mutant strains. Biofilm formation of S. aureus knock-out mutants defective in cell autolysis regulation (∆lrgA and ∆lrgB) was comparatively evaluated to the WT strain. 24 hrs biofilms were evaluated based on metabolic activity measurement using the MTSassay as per manufacture directions (Promega; Madison, WI). Total biofilm biomass was evaluated by the crystal violet staining assay where 24 hr-grown biofilms were fixed in methanol for 20 mins, air dried and stained with crystal violet solution (0.2%) for 20 mins. The excess crystal violet was removed, plates were washed twice with PBS and cell-bound crystal violet was dissolved in 33% acetic acid. 100 µl of the extraction solution was transferred to 96-well plates and absorbance was measured at 595 nm.
In vitro evaluation of eDNA accumulation in singleand mixed-species biofilms. 24 hr-preformed single and mixed-species biofilms were grown as described above using 6-well microtiter plates and eDNA was quantified as previously described with modifications [48]. Briefly, biofilm supernatant was aspirated and formed biofilms were harvested by scraping and resuspended in 1 ml PBS. Suspensions were centrifuged for 10 mins at 5000 g to separate cells and the soluble biofilm matrix and the recovered supernatant was filtered using 0.22 µm Millex syringe filters. The amount of eDNA was quantified using the QuantiFluor® dsDNA System kit (Promega, Madison, WI) as per manufacturers instructions. Fluorescence was measured at 504 nm/531 nm (Ex/Em) in triplicates per sample in flat-bottom black-sides 96-well plates (Greiner, Kremsmünster, Austria) using Biotek Cytation 5 plate reader.
Effect of biofilm DNAse treatment on vancomycin susceptibility of lrgA and lrgB S. aureus mutants. Effect of biofilm DNAse I treatment on vancomycin susceptibility was evaluated as previously described [33], with modifications. 24 hr-preformed biofilms of the WT, ∆lrgA and ∆lrgB strains were grown as described above; following removal of non-biofilm cells, DNAse I (0.625 mg/ml), vancomycin (200, 100, 50 and 25 µg/ml) or a combination of both were added to appropriate wells. Following 24 hrs incubation at 37° C, metabolic activity of biofilm cells was measured using MTS-assay. Control wells with no treatment were included; DNAse I optimal activity concentration was pre-determined in a dose-response assay (data not shown).
Effect of biofilm DNAse treatment on S. aureus vancomycin susceptibility in single and mixed biofilms with C. albicans. 24 hr-preformed single and mixed-species biofilms were grown as described above and, after removal of non-sessile cells, DNAse I (0.625 mg/ml), vancomycin (800 µg/ml) or a combination of both were added to appropriate wells with biofilms. Following 24 hrs of incubation at 37°C, wells were sonicated, and cell suspensions were diluted and plated on species-specific chromogenic media for CFU enumeration. Control wells with no treatment were included; DNAse I optimal activity concentration was pre-determined in a dose-response assay (data not shown) and the vancomycin concentration used was based on our previously published work on mixed biofilm assays [29].
Confocal scanning laser microscopy (CSLM) analysis of vancomycin diffusion through single and mixed-species biofilms with and without DNAse treatment. To visualize the effect of eDNA enzymatic digestion on vancomycin diffusion through single and mixed biofilms, confocal scanning laser microscopy was performed as we previously described [29], with modifications. Single-species S. aureus or mixed-species C. albicans-S. aureus biofilms were grown on glass coverslip-bottom dishes (MatTek Co., Ashland, MA), as described above, for 24 hrs. Following incubation, biofilms were treated with DNAse I (0.625 mg/ml) or fresh RPMI (control) and fluorescent vancomycin conjugated to BODYPI FL (1 µg/ml; 488/650) was added to both control and treated biofilms, and the system was incubated for 1 hr at 37°C. S. aureus cells were stained with concanavalin A (ConA) conjugated to Alexa 657 (50 µg/ml; 650/668 nm) and C. albicans with calcofluor-white (0.05%; 433/355). Biofilms were gently washed and immediately observed in an inverted confocal laser scanning microscope (Nikon T2i). At least three random fields were visualized for each sample, and representative images of the reconstructed Z-stacks are presented.
Data analysis. All statistical analysis was performed using GraphPad Prism 8.0 software. All in vitro experiments were performed on at least 3 separate occasions and in triplicate where applicable, and averages were used to present data. Multiple groups were compared by 2-way ANOVA with Tukey's multiple-comparison correction. The set of data obtained from WT and lrg mutants biofilm susceptibility to vancomycin alone or combined with DNAse was analyzed by 2-WAY ANOVA using the Multiple Effect analysis (Mixed Model). Student's unpaired t-test was used to compare differences between two samples; p values of <0.05 were considered to be significant. For in vivo experiments, the database for statistical analysis was built up by incremental addition of experimental data. Based on our analysis, a sample size of 5 mice per experimental group will have at least 80% power to detect difference of 80%. All experiments were performed on at least three separate occasions, analyzed separately and averages used to present data. The Kruskal-Wallis oneway analysis of variance test and Dunnet's Multiple Comparison Test were used to compare differences between multiple groups and statistical significance, respectively. P values <0.05 were considered to be significant.

Results
Mixed biofilm growth confers S. aureus with enhanced tolerance to vancomycin in vitro. C Singleand mixed-species biofilms were grown in 96-well plates using S. aureus USA300 and C. albicans SC5314 reference strains for 24 hrs and then treated with vancomycin (800 µg/ml) for 24 hrs at 37°C. Following incubation, wells were sonicated to disrupt biofilms, and cell suspensions were diluted and plated on S. aureus-specific and Candida-specific chromogenic medium (CHROMagar; DRG International, Inc.) for CFU (colony forming units) enumeration. Consistent with previous observations, significant S. aureus killing was seen in the vancomycin-treated single-species biofilms (****, p < 0.0001), which was significantly lower in mixed biofilms with C. albicans (**; p < 0.05) ( Figure 1).
Scanning electron microscopy and microbial recovery reveals formation of robust mixed biofilms within implanted catheters. In order to determine optimal in vivo biofilm growth period, initially, small groups of animals with implanted catheters were euthanized two, three, four, and six days post-implantation and catheters were evaluated for microbial recovery. Based on CFU counts, significantly less S. aureus and C. albicans were recovered from mixed-species Single-and mixed-species biofilms were grown for 24 hrs in the wells of 96-well plates then treated with vancomycin (800 µg/ml) for 24 hrs. Following incubation, biofilms cells were harvested and plated on species-specific chromogenic medium. Based on CFU values, results demonstrated significant efficacy for vancomycin against S. aureus in single-species biofilms with significantly reduced efficacy against S. aureus in mixed-species biofilm. *, p < 0.05; ***, p < 0,001. catheters after 2 days (***, p < 0.001 and **, p < 0.01; respectively), compared to from single-species ( Figure  2c) and for both species CFU were equivalent in single and mixed biofilms by day 3. Based on the kinetics study, CFU recovery 3 days post-implantation was determined to be optimal time-point, after which growth of both species reached a plateau. SEM representative images of lumens of explanted catheters revealed the presence of mature mixed biofilms after 8 days (Figure 2b). Comparatively, single-species S. aureus biofilms had less biomass than C. albicans-S. aureus mixed biofilms, whereas the C. albicans biofilms consisting primarily of hyphae were comparable between C. albicans single and mixed biofilms ( Figure 3).
In vivo vancomycin therapy reduces S. aureus recovery from single species but not from mixedspecies infected catheters. After confirming the suitability of the model for in vivo biofilm development, vancomycin therapy was administered twice daily for 5 days starting 3-days post-catheter implantation (Figure 4a). Based on microbial recovery from explanted catheters, vancomycin treatment resulted in significantly less S. aureus recovery from catheters from animals with single-species infection (****, p < 0.0001), but had no impact on S. aureus recovery from animals with mixed-species catheter infections (Figure 4b).
Vancomycin had no effect on C. albicans recovery under all conditions tested (suppl. Fig S1).
C. albicans modulates expression of S. aureus biofilm regulators in mixed-species biofilms. Transcriptomic analysis of S. aureus recovered from explanted catheters revealed that 134 genes were differentially expressed between the multi-species biofilms and the S. aureus alone biofilms. These included 43 genes with higher expression in the multi-species biofilm and 91 genes with lower expression in the multi-species biofilm (Table 1 and S1). The most notable modulations were in S. aureus keyregulatory genes involved in biofilm formation, including sarU, icaB, icaC, lrgA, lrgB and nuc (Table 1). While the ica genes involved in the production of PIA (polysaccharide intercellular adhesin) were upregulated, the anti-holin like repressors of cell autolysis (lrgA and B) and the DNA nuclease (nuc) were downregulated. An overall downregulation of toxins and hemolysins was seen in S. aureus from mixed-biofilms including genes for gamma-hemolysins, alpha-hemolysin, leucocidin and staphylokinase; similarly, immunomodulatory and adhesion genes were also found to be down-regulated in S. aureus in mixed biofilms. In contrast, capsular polysaccharide biosynthesis was significantly induced by the presence of C. albicans. Although 114 C. albicans genes were found to be differentiallyexpressed between C. albicans alone and mixed-species biofilms (Table S2), Gene Ontology Analysis (GO) did not detect significant enrichment among the differentiallyexpressed genes. Therefore, we focused our downstream analyses on S. aureus genes that were significantly regulated in the mixed-species biofilms.

S. aureus lrgB mutant biofilms exhibit accumulation of eDNA in the ECM and enhanced susceptibility to vancomycin upon DNAse treatment of biofilm.
Compared to the WT strain, analysis of 24 hr-biofilms demonstrated reduced metabolic activity (**, p < 0.01; Figure 5a) for the lrgB mutant but comparable biomass (Figure 5b), and slightly reduced susceptibility to higher vancomycin concentrations (WT x ∆lrgB: p = 0.0147 using Multiple Effect analysis; Figure 5c). In contrast, the lrgA mutant exhibited similar biofilmforming ability as the WT strain and increased susceptibility was observed for reduced vancomycin concentrations (WT x ∆lrgA: p = 0.0008 using Multiple Effect analysis; Figure 5c). DNAse treatment of biofilms had no significant impact on the susceptibility of the WT or the ∆lrgA strains to vancomycin (Figure 5c); in contrast, DNAse biofilm treatment significantly increased the susceptibility of the ∆lrgB strain to vancomycin (*, p < 0.05; **, p < 0.001; Figure 5c).

C. albicans-S. aureus mixed-species biofilm exhibit
increased eDNA content compared to single-species biofilms. eDNA from single species and mixed biofilm was recovered and quantitatively compared using a fluorescence-based kit. As expected, eDNA content recovered from mixed biofilms was significantly higher than that recovered from biofilms formed by S. aureus or C. albicans alone (Figure 6a).

DNAse treatment of mixed S. aureus-C. albicans biofilm enhances vancomycin efficacy against S. aureus.
To determine whether increase in eDNA contributes to the enhanced S. aureus tolerance to vancomycin in mixed biofilm, vancomycin susceptibility testing was performed in the presence of DNAse. Based on CFU counts, DNAse treatment of mixed biofilms significantly sensitized S. aureus to vancomycin (*, p < 0.05) (Figure 6b). In contrast, DNAse did not impact S. aureus susceptibility to vancomycin in single-species biofilm. DNAse alone had no significant

. Efficacy of vancomycin therapy in mice with catheters with single or mixed-species infection. (a) Timeline of infection and treatment.
Vancomycin was administered (55 mg/kg) 3 days post-catheter implantation twice daily for five days and microbial recovery from explanted catheters was evaluated. (b) Following explantation, catheters were sonicated to disrupt biofilms, and cell suspensions were diluted and plated on species-specific chromogenic media for assessment of microbial recovery. Based on CFU counts, vancomycin treatment resulted in significant decrease in S. aureus recovery from catheters infected only with S. aureus, whereas vancomycin had no effect on S. aureus from catheters co-infected with C. albicans. ****, p < 0.0001; ns, not significant. effect on S. aureus viability compared to untreated controls.
DNAse treatment significantly disrupts mixedspecies biofilm impacting vancomycin distribution. Confocal laser scanning microscopy (CLSM) was performed using fluorescent vancomycin to visualize the effect of DNAse treatment on vancomycin diffusion through single and mixed biofilms. S. aureus cells were stained with concanavalin A (ConA; magenta) and C. albicans with calcofluor-white (blue). Vancomycin penetration through the biofilms is seen in white and green (overlapping of green in high intensity generates the white color). Compared to untreated S. aureus single-species biofilm, biofilm z-stack reconstructions demonstrated that DNAse treatment resulted in a more dispersed distribution of vancomycin (green/white) throughout the layers of S. aureus biofilm (Figure 6c). In contrast, DNAse treatment of mixed C. albicans-S. aureus biofilms resulted in significant disruption in the biofilm structure, and increase in vancomycin presence (green/white) (Figure 6d).

Discussion
Mixed fungal-bacterial infections tend to be the most complex and challenging to treat; however, the impact of these interactions on therapy remains largely understudied. Candida and staphylococci species are the most common etiological agents isolated from deviceassociated nosocomial infections, and co-infection with these species poses a significant therapeutic challenge [23,24,49]. Therefore, it has become crucial to understand the mechanisms of inter-species interactions within the context of polymicrobial infections. However, a major challenge in investigating polymicrobial biofilm-associated infections is the development of suitable animal models that account for host factors, and allow for evaluation of efficacy of treatment strategies under in vivo conditions. Therefore, to substantiate our in vitro findings on enhanced S. aureus antimicrobial tolerance in mixed biofilms with C. albicans (Figure 2), we modified the established mouse subcutaneous catheter model, which proved to be suitable for sustaining polymicrobial biofilm formation, as demonstrated by SEM analysis of explanted catheters (Fig. 2, 3) [45,50]. Vancomycin therapy was implemented at clinically relevant therapeutic doses over a period of five days in mice with single-species or mixed-species infected catheters. Comparative assessment of S. aureus recovery from explanted catheters demonstrated inefficacy for vancomycin against S. aureus in mice with coinfected catheters (Figure 4). Vancomycin is one of the few antibiotics that have remained effective against methicillin-resistant S. aureus, and development of resistance to vancomycin is relatively rare [51]. Therefore, the demonstration of failure of vancomycin against mixed C. albicans-S. aureus device infection carries significant clinical implications. Notably, although single-species S. aureus biofilms were statistically susceptible to vancomycin treatment, the level of response was variable among the experimental sets. It is important to note however, that although the subcutaneous catheter infection model is an appropriate and reproducible model to evaluate foreign device biofilm  infections and anti-biofilm strategies [45,50], one disadvantage is lack of blood flow, which can result in inefficient drug delivery to the biofilm within the catheter. In fact, an initial pilot study indicated no effect for vancomycin on S. aureus recovery from single-species biofilms, supporting this concern (data not shown), and therefore, to circumvent this problem, for the therapy experiments, the catheters were cut longitudinally prior to inoculation and implantation in the animals. Neverthelss, despite this limitation, the findings from using the model for in vivo-grown polymicrobial biofilms serve as a proof-of-principle underscoring the challenges associated with treatment of deviceassociated fungal-bacterial infections.
To provide insights into the mechanisms contributing to the observed therapeutic implications of mixed biofilm infections, we performed RNA-seq analysis to delineate changes in gene expression during polymicrobial infections. The most significant finding from the comparative transcriptome analysis of S. aureus was the down-regulation of the repressors of autolysis, lrgA and lrgB (Table 1 and Table S1), during mixed catheter infection. eDNA is essential for bacterial biofilm formation and stability, and the majority of eDNA in the biofilm matrix is a result of cell lysis, which are critical for biofilm attachment during the initial stages of development [40]. Regulatory mechanisms in S. aureus controlling cell death and lysis during biofilm development are mediated by the opposing functions of the cid and lrg operons encoding holin-and antiholinlike proteins, respectively, [36,39]. The lrg operon encoding LrgA and LrgB proteins are the inhibitory counterparts to cid, and a repressor of murein hydrolase activity that hydrolyzes components of the cell wall involved in autolysis. Therefore, down-regulation of lrgAB promotes the activity of the cidA-encoded holin allowing increased activity of murein hydrolases, cell lysis and eDNA production. In fact, the lrgAB mutant was shown to exhibit increased biofilm matrixassociated eDNA, consistent with its role as an inhibitor of cidA-mediated lysis [36,40,52,53]. Similarly, staphylococcal thermonuclease (nuc), which was downregulated during mixed catheter infection, was also shown to be involved in biofilm development, and a nuc mutant was reported to form thicker biofilms containing increased levels of eDNA [40]. Therefore, cell lysis and genomic DNA release during biofilm development is regulated via opposing activities of the cid and lrg gene products, while staphylococcal thermonuclease functions to degrade the eDNA, possibly as a mean to promote biofilm dispersal.
S. aureus biofilm formation is a two-step process that requires the adhesion of bacteria to a surface followed by cell-cell adhesion or intercellular aggregation. In staphylococci, the main molecule responsible for intercellular adhesion is the polysaccharide intercellular adhesin (PIA), also called poly-N-acetylglucosamine (PNAG), which together with other polymers such as teichoic acids forms the major part of the extracellular biofilm matrix [54,55]. PIA biosynthesis is mediated by the intercellular adhesin (ica) locus, icaADB and C, with the icaB and icaC genes encoding for extracellular membrane proteins, and icaC proposed to be a putative PIA exporter [36]. Transcriptome analysis of S. aureus recovered from mice with mixed catheter infection demonstrated both icaB and icaC to be upregulated. Regulation of icaADBC is complex and numerous regulatory factors have been implicated, many of which are well-known global transcriptional regulatory factors like SarA and sigmaB [54]. SarA also appears to enhance biofilm formation more directly, by increasing ica expression [54,56]. Interestingly, SarU, a member of the SarA family proteins was found to be up-regulated during mixed catheter infection. Combined, the downregulation of the lrg operon and nuc, concomitant with the upregulation of the icaADBC-encoded PIA indicate transcriptional activation of the S. aureus biofilm formation regulatory network, induced by the presence of C. albicans during mixed biofilm growth.
It is well established that biofilms are physiologically different from planktonic cells and cells in a biofilm have altered metabolic activity. In a comparative transcriptome analysis to determine which genes are up-regulated in S. aureus biofilm, a study by Resch et al. [57] found that, although the cell envelope appeared to be a very active compartment in biofilm cells as indicated by upregulation of PIA, virulence factors such as toxins and proteases were up-regulated under planktonic growth conditions but not during biofilm growth. These findings are in line with ours, where a set of host cell-damaging proteins were found to be down-regulated during mixed species infection including gamma-hemolysins (γhemolysin) HlgAB and HlgCB. Interestingly, staphylococcal alpha-hemolysin (α-hemolysin) encoded by the gene hla, also plays a major role in the biofilm formation and appears to be primarily required for cell-to-cell interactions. In fact, a mutant allele of hla was shown to initially aid in colonizing a substratum. In addition to hemolysins, S. aureus secretes bi-component cytotoxins, Panton-Valentine leukocidins, encoded by the genes lukSF-PV composed of two co-transcribed open reading frames (lukS-PV and lukF-PV), which were also found to be down-regulated during mixed catheter infection [58,59]. Expression and secretion of these factors play specific roles in certain S. aureus life stages and diseases, and are tightly controlled by a number of regulatory systems. Most notable is the accessory gene regulator (agr), a well-characterized two-component regulatory system that plays a critical role in the up-regulation and down-regulation of protease and exotoxins, respectively, [60][61][62]. Interestingly, SarU, a positive activator of agr expression, is a member of the SarA family proteins; sarA is a transcriptional regulator that represses extracellular proteases and therefore, the observed elevated sarU expression during mixed catheter infection may result in additional amplification of agr signal, and explain the observed down-regulation of exotoxins in our infection model [33,63].
Interestingly, in contrast to our findings, recent studies by Todd et al. [64,65] aiming to elucidate observed enhanced mortality during polymicrobial C. albicans and S. aureus intra-abdominal infection, identified C. albicans-mediated increase in S. aureus secreted alpha-and delta-toxins during polymicrobial infection. Although these factors are important virulence factors in S. aureus crucial to its ability to cause systemic and disseminated infections, down-regulation of these factors might contribute to survival, persistence, and growth in a biofilm environment, as was seen in our biofilm infection model. Nevertheless, the contrasting regulation of S. aureus exotoxins mediated by C. albicans between the two animal models highlights the complex interactions between these two pathogens, and support the notion of niche-specific augmented pathogenesis in a co-infected host.
It is important to note that previous studies by Peters et al. [66] exploring the mechanism by which C. albicans augments S. aureus infection during polymicrobial intrabdominal infection, correlated lethal outcomes with exacerbation of host inflammatory responses. These findings are interesting as our transcriptome analysis of mixed-species catheter infection seemed to indicate potential dampening of host immune responses via down-regulation of superantigen like proteins in S. aureus, such as the immunodominant surface antigen B (IsaB) which was shown to elicit immune response during septicemia [67]. These findings are in line with the overall perception that biofilms tend to be resistant to host immune defense system [68]. Interestingly, in addition to contributing to biofilm matrix, PIA molecules, which we found to be up-regulated during mixed catheter infection, also serve to shield cells from neutrophil phagocytosis, thus significantly contributing to biofilm resistance from elimination by innate host defense [26]. Most significantly however, was the noted up-regulation of S. aureus capsular polysaccharide biosynthesis during mixed catheter infection. The production of capsular polysaccharide by S. aureus has been proposed as an anti-phagocytic evasion strategy by inhibiting neutrophil phagocytosis [58]. Serotype 5 and 8 capsular polysaccharides predominate among clinical S. aureus isolates and animal studies have revealed that staphylococcal capsules enhance virulence by impeding phagocytosis. Furthermore, the capsule was also shown to modulate S. aureus adherence to surfaces in vitro and in vivo, suggesting that it also promotes bacterial colonization, and persistence in the infected host [69] In light of the findings from the transcriptional analysis indicating enhanced S. aureus autolysis in mixed biofilms, and given the implication of biofilm eDNA on staphylococcal resistance to antimicrobials, as follow-up studies, we comparatively evaluated the accumulation of eDNA in single-and mixed-species biofilms. Additionally, the effect of eDNA on S. aureus response to vancomycin was also evaluated using DNA enzymatic digestion, and mutant strains lacking lrg operon genes. Quantification of eDNA in biofilms indicated a comparable level of eDNA produced by each species in their respective biofilms. In comparison, the total amount of eDNA in the mixedspecies biofilms was significantly higher (Figure 6a). Given that both species produced eDNA and the assay is not specific, it remains speculative at this time to confirm increase in S. aureus eDNA specifically in mixed biofilm. It is noteworthy however that eDNA level in mixed biofilms was greater than the combined eDNA concentrations from single-species biofilms (0.2350 ± 0.0046 versus 0.216 ± 0.00989; p = 0.2592; data not shown). Nevertheless, DNAse treatment of mixed biofilms significantly sensitized S. aureus to vancomycin, but had little impact on S. aureus susceptibility in single-species biofilm (Figure 6b). The impact of DNAse on mixed biofilms was further demonstrated using confocal laser scanning microscopy where fluorescently-labeled vancomycin diffusion through DNAse treated biofilms caused significant disruption to the mixed biofilm structure, resulting in a more dispersed and accessible distribution of vancomycin (Figure 6c, D). Comparative evaluation of biofilm formation using the lrg mutant strains paradoxically demonstrated reduced metabolic activity in lrgB mutant biofilm, but no reduction in biofilm biomass, indicating decrease in cell viability due to increase in cell lysis, without impacting biofilm due to increase in eDNA content (Figure 5a-C). Theses findings were substantiated by the reduced susceptibility of the lrgB mutant to vancomycin, which was reversed upon DNAse treatment (Figure 5c). It is important to mention that the downregulation of the lrg operon during in vivo grown mixed staphylococcal-C. albicans biofilm and inc12qqrease in biofilm eDNA content was previously reported by Pammi et al. [33], using a similar model. However, in that study, microarrays were used to delineate changes in Staphylococcus epidermidis gene expression induced by C. albicans. Although the authors posited that the enhanced release of eDNA due to increased autolysis would affect antibiotic tolerance, antimicrobial susceptibility assays were not performed in vitro or in vivo. However, the key role for eDNA in mediating staphylococcal tolerance to antimicrobials in single-species biofilm was clearly demonstrated by subsequent studies by Doroshenko et al. [70] and Brackman et al. [48]. In terms of therapeutic implications of the findings from our study, it is important to note an interesting study by Siala et al. [71] demonstrating that the antifungal caspofungin improves the activity of fluoroquinolones against S. aureus biofilms in vitro as well as in a similar subcutaneous biofilm model system as used in our study. Subsequent sequence analysis revealed a shared homology for IcaA with β-1-3-glucan synthase, the enzyme responsible for synthesis of β-1-3-glucan in C. albicans, and the target for caspofungin. The findings from the study did in fact show that caspofungin treatment destructured the S. aureus biofilm, and reduced polymerization of matrix polysaccharides increasing drug penetration into biofilms. Hence, the authors proposed that IcaA inhibitors could be useful in the treatment of S. aureus biofilm infections, substantiating the significance of our findings demonstrating up-regulation of the ica operon in mixed biofilms. However, as caspofungin is the drug of choice for treatment of C. albicans-biofilm infections, the findings from the study are of even more significance in light of ours identifying an important role for eDNA in the C. albicans-mediated enhanced S. aureus tolerance to antimicrobials. In fact, eDNA has been described as a promising new target for staphylococcal anti-biofilm treatments [35] and therefore, it is conceivable to speculate that the application of combination DNasecaspofungin treatment may prove effective against mixed C. albicans-staphylococci biofilm infections.
Combined with the findings from our previous studies [29,32], we propose that in a mixed biofilm, the C. albicans polysaccharide matrix, the increase in S. aureus matrix components, and interactions via secreted chemical moleucles, create a protective environment advantageous to S. aureus (Figure 7). Therefore, the work presented provides novel mechanistic insights into interspecies interactions in biofilms, with important clinical relevance. We expect the novel findings from this study to further our understanding of the enhanced pathogenesis and resistance of biofilmassociated polymicrobial infections, which may contribute to the development of novel therapeutic strategies targeting these complex infections.  [29], we proposed a "Barrier Model" whereby the C. albicans biofilm matrix, consisting of hyphae and secreted cell wall polysaccharides (α-mannan, β-1,3-glucan and β-1,6 glucan), impedes vancomycin diffusion through the mixed-species biofilm, providing embedded S. aureus cells with protection against antimicrobials. However, a subsequent study [32], indicated a key role for the C. albicans secreted quorum sensing molecule farnesol in the enhanced S. aureus tolerance to vancomycin, via activation of a bacterial stress response. In this current study, we demonstrate that C. albicans induces transcriptional modulations in key S. aureus regulatory genes involved in biofilm formation, via down-regulation of the lrg operon, repressor of autolysis, and up-regulation of the ica operon, resulting in increase in eDNA and extracellular polysaccharide (PIA/PNAG) production, respectively. Therefore, in a mixed biofilm, the C. albicans polysaccharide matrix, combined with increase in S. aureus matrix components, and interactions via secreted chemical molecules, create a protective environment advantageous to S. aureus.