Combined Application of Antisense Oligomers to Control Transcription Factors of Candida albicans Biofilm Formation

Antisense oligomers (ASOs) have been little exploited to control determinants of Candida albicans virulence. Biofilm formation is an important virulence factor of C. albicans, that is regulated by a complex network of transcription factors (such as EFG1, BRG1 and ROB1). Thus, the main goal of this work was to project ASOs, based on the 2'-OMethyl chemical modification, to target BRG1 and ROB1 mRNA and to validate its application either alone or in combination with the EFG1 mRNA target, to reduce C. albicans biofilm formation. The ability of ASOs to control gene expression was evaluate by qRT-PCR. The effect on biofilm formation was determined by the total biomass quantification, and simultaneously the carbohydrates and proteins reduction on extracellular matrix. It was verified that all the oligomers were able to reduce the levels of gene expression and the ability of C. albicans to form biofilms. Furthermore, the combined application of the cocktail of ASOs enhances the inhibition of C. albicans biofilm formation, minimizing biofilm thickness by reducing the quantity of matrix content (protein and carbohydrate). So, our work confirms that ASOs are useful tools for research and therapeutic development on the control of Candida species biofilm formation.


Introduction
Candidiasis is the primary fungal infection, with a mortality rate of about 40% and high costs associated in hospitalized patients [1]. Candida albicans remains the most prevalent of all Candida species and its pathogenicity is facilitated by a number of virulence factors. It is assumed that one of the major contributions to C. albicans virulence is its versatility to adapt to a variety of different environments and to form surface-attached microbial communities, known as biofilms [2,3]. Biofilm formation is a sequential phenomenon that involves attachment, maturation, and detachment, and is regulated by a complex network of genes [3]. In this sense, EFG1, BRG1 and ROB1 are known as important transcription factors required for C. albicans biofilm formation [2]. EFG1 has an important role in cell adhesion, through the cell-wall gene regulation [4,5] and is required for biofilm maturation [6] and hyphal growth [7]. BRG1 is also required for normal biofilm growth [6,8] and is essential for invasive growth [9]. ROB1 is a zinc finger protein transcription factor required for spider model biofilm formation [6,8], epithelial invasive growth and normal colony morphology [8]. Antisense oligomers (ASOs) are short strands of nucleic acids that hybridize to a target complementary mRNA, specifically blocking its translation into protein, and hence, inhibiting its function [10]. At the moment, there are three generations of chemically modified ASOs [10,11], developed to enhance their nuclease resistance, to improve their delivery and biodistribution and to increase their binding affinity for target mRNA [10][11][12]. The 2'-OMethylRNA (2'OMe) is characterized by a sugar modification with the introduction of an oxygenated group. Although this chemical modification does not allow the recruitment of RNase H, the insertion of a central unmodified sequence, called as gapmer, can allow the RNase activity, overcoming this limitation [13,14]. The anti-EFG1 2'OMe oligomer was recently projected by our group, presenting a high ability to reduce C. albicans cells' filamentation in vitro [15] and in vivo status [16]. In this sense, the main goal of this work was to project new ASOs, based on 2'OMethyl chemical modification, to target BRG1 and ROB1 mRNA which are important transcription factors for biofilm formation, and to validate their application either alone or in combination with EFG1 2'OMe oligomer [15], to control C. albicans biofilm formation.

Materials and Methods
ASOs' design and synthesis The ASO to hybridize with EFG1 mRNA was projected by us in our previous work [15]. To design specific ASOs for C. albicans BRG1 and ROB1, the target regions of each gene were selected based on a search conducted at Candida Genome Database (http://www.candidagenome.org/cgi-bin/compute/ blast_clade.pl). Firstly, several gene sequences were aligned to make sure that conserved regions were used for the design. Also, a general BLAST search was performed to ensure that the sequences were not targeting any sequence at the human genome neither a similar region in another C. albicans genes. The sequences were selected considering its high specificity to C. albicans genome, non-binding against Homo sapiens genome and the number of nucleotides to include in the sequence [10]. In order to improve the oligomers hybridization and stability specific nucleotides of each sequence were chemically modified using the 2'ribose modification. The 2'OMe modification was selected since it is one of the most used for antisense applications [17,18]. A gapmer region was also introduced in both ASOs to increase the odds of activating the RNase H activity [19].
The ASOs were synthesized at the Nucleic Acid Center, University of Southern Denmark, using the standard phosphoramidite method on an automated nucleic acid synthesizer (PerSpective Biosystems Expedite 8909 instrument) as described previously by Araújo and colleagues [15].

ASOs' Cytotoxicity
The ASOs' cytotoxicity was evaluated against the 3T3 cell line (Fibroblast cells, Embryonic tissue, Mouse from CCL 163, American Type Culture Collection) as described in a previous work [15]. For that, 3T3 cells were grown in Dulbecco's Modified Eagle's Medium (D-MEM, Biochrom, Germain) supplemented with 10% of fetal bovine serum (FBS, Sigma Aldrich) and 1% of antibiotic containing P/S (penicillin and streptomycin) (Biochrom, Germain). After detachment, a suspension with 1 9 10 5 cells mL -1 was added to a 96-well plate and the cells grew until attaining 80% of confluence. Before the cytotoxicity assay, the wells were washed twice with PBS. Individual and combined concentrations of ASOs (40 and 100 nM) were prepared in DMEM medium and 50 lL of each concentration was added to each well. A negative control was performed by adding 50 lL of DMSO to the cells and a positive control by adding 50 lL of D-MEM medium. The plates were incubated for 24 h, at 37°C, and 5% of CO 2 .

Microorganism and growth conditions
The Candida strain used in this study was the reference C. albicans SC5314, belonging to the Candida collection of the Biofilm group at the Centre of Biological Engineering. Its identity was confirmed using a chromogenic medium, CHROMagar TM Candida, through the distinction of colonies' colours and by PCR-based sequencing with specific primers (ITS1 and ITS4) [20]. The mutant strains C. albicans efg1DD (HLC52) and brg1DD (SN76) were also included in this study in order to validate the ASOs performance.
For all experiments, the yeast strains were subcultured on Sabouraud dextrose agar (SDA; Merck, Germany) and incubated for 24 h at 37°C. Cells were then inoculated in Sabouraud dextrose broth (SDB; Merck, Germany) and incubated overnight at 37°C, 120 rpm. After incubation, the cell suspensions were centrifuged for 10 min at 3000 g at 4°C and washed twice with phosphate-buffered saline (PBS; pH 7, 0.1 M). Pellets were suspended in 5 mL of Roswell Park Memorial Institute medium (RPMI, pH 7, Sigma, St Louis, USA), and the cellular density was adjusted for each experiment using a Neubauer chamber (Marienfild, Land-Konicshofem, Germany) to 1 9 10 6 cells Ml -1 .

ASOs' effect on biofilm formation
Candida albicans biofilms were developed in RPMI on 24-wells polystyrene microtiter plates (Orange Scientific, Braine-l'Alleud, Belgium). For the individual and combined tests, 500 lL of ASOs, each at a final concentration of 40 nM were prepared in RPMI (concentration selected based on cytotoxicity assays) and added to 500 lL of C. albicans suspension at 2 9 10 6 cells mL -1 . The positive control was prepared with 1 mL of the same yeast cell concentration on RPMI. The biofilms were incubated at 37°C under agitation of 120 rpm. After 24 h of biofilm formation, the supernatants were removed and the biofilms were washed twice with PBS to remove the non-adherent cells, and subsequently characterized [21].

ASOs' effect on gene expression
Reverse transcription-quantitative PCR (qRT-PCR) was used to determine the effect of ASOs on EFG1, BRG1 and ROB1 expression. For that, C. albicans biofilms were developed as described previously, and then the biofilm cells were removed for RNA extraction. Briefly, the developed biofilms were scraped from the microtiter plate wells in the presence of 1 mL of PBS and sonicated (Ultrasonic Processor, Cole-Parmer, Vernon Hills, Illinois) for 30 s at 30 W to separate the yeast cells from matrix [21]. Cells were collected from the suspension by centrifugation for 5 min at 6000 g and 4°C, and then washed once with PBS. RNA extraction was performed using the PureLink RNA Mini Kit, according to the manufacturer's instructions (Invitrogen, Carlsbad, CA, USA) [15]. To avoid potential DNA contamination, samples were treated with DNase I (Deoxyrybonuclease I, Amplification Grade, Invitrogen) and the RNA concentration was determined by optical density measurement (NanoDrop 1000 Spectrophotometer Thermo ScientificÒ). The complementary DNA (cDNA) was synthesized using the Xpert cDNA Synthesis Mastermix (Grisp, Porto, Portugal) in accordance with the manufacturer's instructions, and qRT-PCR (CFX96, Biorad) was performed on a 96-well microtiter plate using Eva Green Supermix (Biorad, Berkeley, USA). Each reaction was performed in triplicate and mean values of relative expression were determined by the 2 -DDCq method. The expression of each gene was normalized using the ACT1 Candida reference gene [22]. Non-transcriptase reverse (NRT) controls were included in each run. The primers were designed using the Primer 3 web-based (Table S1).

Biofilm biomass analysis
The ASOs effect on biofilm formation was determined by crystal violet (CV) methodology [21]. For that, firstly the biofilms were fixed with 500 lL of methanol, for 15 min. After the methanol removal, the biofilms were dried at room temperature, and then 500 lL of CV (1% v/v) were added to each well. The stain was aspirated after 5 min and its excess was removed by washing the biofilms twice with sterile ultra-pure water. Finally, 500 lL of acetic acid (33% v/v) were added to each well to release and dissolve the CV stain. The absorbance of the CV solutions was then measured, at 570 nm, and the results presented as absorbance per unit area (Abs CV cm -2 ).
Simultaneously, the number of cultivable cells on C. albicans biofilm was also estimated using the colony forming units (CFUs) counting methodology [23]. Briefly, the developed biofilms were scraped from the microtiter plate wells in the presence of 1 mL of PBS and disrupted was previously described to separate the yeast cells from the matrix [21]. The cells suspensions were then collected by centrifugation for 5 min at 6000 g and 4°C and serial decimal dilutions (in PBS) were plated onto SDA. Agar plates were incubated for 24 h at 37°C and the total enumerated and presented per unit area (Log CFU cm -2 ) [23].

Biofilm structure analysis
In order to study the oligomers' effect in the biofilm thickness the confocal laser scanning microscopy (CLSM) was used. For that, the 24 h biofilms were stained with 1% (v/v) of Calcofluor white (Sigma-Aldrich, St Louis, MO, USA) for 10 min at room temperature in the dark and then observed with a CLSM (Olympus BX61, Model FluoView 1000, Portugal). The excitation line 405 and the emission filters BA 430-470 (blue channel) were used, and images were acquired with the program FV10-ASW 4.2 (Olympus).

Biofilm matrix analysis
The combined effects of ASOs on biofilm matrix composition were evaluated as described by Silva et al. [23]. For that, after 24 h, the biofilms were scraped from the wells and then disrupted as described before. The suspensions were vortexed for 2 min and centrifuged at 5,000 g for 5 min. The pellets were dried at 37°C until a constant weight was obtained. The matrix-containing supernatants were filtered through a 0.2 mm nitrocellulose filter and then the protein and total carbohydrate contents were estimated as described next.

Total protein quantification
The protein content was measured using the BCA Kit (Bicinchoninic Acid, Sigma-Aldrich, St Louis, Missouri) and bovine serum albumin (BSA) as standard [21]. Briefly, 0.2 mL of BCA solution was added to 25 lL of matrix sample and incubated for 30 min at 37°C. Then, the absorbance was determined in a microplate reader at 562 nm. The protein concentration was extrapolated from a calibration curve (abs 0.009 x [protein] þ 0.1685) performed with standard concentrations of BSA. The results were normalized with the dry weight of biofilm cells, previously determined, and presented as mg of protein per g of biofilm (mg g biofilm -1 ).

Total carbohydrate quantification
The total carbohydrate content was estimated using the phenol-sulfuric method [24] and glucose as standard. Briefly, 0.5 mL of phenol (50 g L -1 ) and 2.5 mL of sulfuric acid (95-97%) were added to 0.5 mL of matrix sample. The solution was vortexed for 30 s and incubated for 15 min at room temperature. The absorbance was determined in a microtiter plate reader at 490 nm. The concentration of carbohydrate was extrapolated from a calibration curve (Abs 0.2955 9 [carbohydrate] þ 0.114) performed with standard glucose concentrations. The results were normalized with the dry weight of biofilm cells and presented as mg of carbohydrate per g of biofilm (mg g biofilm -1 ).

Statistical analysis
All experiments were performed in triplicate and in a minimum of three independent assays. Data are expressed as the mean ± standard deviation (SD) of a least three independent experiments. Results were compared using t test analysis with a confidence level of 95%. All tests were performed with GraphPad Prism 6Ò (GraphPad Software, CA, USA).

ASOs' characterization
The ASOs targeting BRG1 and ROB1 were designed taking into account important characteristics in order to increase the binding affinity for the target mRNA and its stability, namely the melting temperatures achieved (around 39-42°C) and the guanine-cytosine (GC) content (around 50-60%). The ASOs were projected with sizes between 14 and 18 nucleotides in order to improve the hybridization performance and also its specificity. Three or four nucleotides in the end of the each ASOs sequence were altered with 2'OMe modification to increase the stability and the central regions were constituted by a gapmer of DNA nucleotides in order to ensure compatibility with RNase H [17]. The ASOs' sequences and characteristics are presented in Table 1. The sequence and characteristics of the ASO targeting the EFG1 mRNA was previously described in Araújo et al. [15].

Evaluation of ASOs' cytotoxicity
In order to infer about the cytotoxicity of the ASOs, the viability of the 3T3 fibroblasts cells in the presence of 40 and 100 nM of anti-BRG1 and anti-ROB1 2'OMe oligomers in individual and combined status (also with anti-EFG1) was determined using the MTS assay (Fig. 1). The results revealed that either in individual and combined status, the ASOs did not present cytotoxicity in both concentrations tested (40 nM and 100 nM), since under these conditions the relative cell viability was higher than 70% [25]. The anti-EFG1 2'OMe absence of cytotoxicity was confirmed in our earlier study [15].

ASOs' effect on target gene expression
The ASOs effect was evaluated in terms of EFG1, BRG1 and ROB1 expression reduction by qRT-PCR, in C. albicans biofilm cells incubated in the presence of the individual and combined oligomers in comparison to their absence (Fig. 2). Figure 2A presents the levels of EFG1, BRG1 and ROB1 expression on C. albicans biofilm cells incubated individually with ASO. It was clear that all ASOs were able to control the respective gene expression, although at different levels. The anti-EFG1 2'OMe was capable to reduce around 75% the levels of EFG1 expression (P value \ 0.001), anti-ROB1 approximately 70% (P value \ 0.05) and anti-BRG1 around 30% (P value \ 0.05). Figure 2B presents the levels of Table 1 Characterization of anti-EFG1 [15], anti-BRG1 and anti-ROB1 2'OMe ASOs, with the respective size and GC (guanine and cytosine) content each gene expression on C. albicans biofilm cells treated with the different combinations of ASOs. Figure 2B shows that, the expression level of EFG1genes was reduced around 50-60%, however only with significantly level of reduction in case of biofilms incubated with the cocktail of the three oligomers (P value \ 0.05). In the case of BRG1, the highest reduction (around 70%) was observed when the biofilms were treated with the combination of anti-BRG1 ? anti-ROB1 2'OMe (P value \ 0.05). Concerning the ROB1 gene, the result was similar for all the combinations of ASOs tested with a reduction of around 50-70% (P value \ 0.05).

ASOs' effect on C. albicans biofilm formation
The ASOs effects was evaluated in terms of ability to control C. albicans biofilm formation under an individual and combined status by total biomass quantification (Fig. 3). It was clear that all ASOs individually (Fig. 3A) were able to reduce C. albicans biofilm biomass, in around 20%, with the highest reduction found in the presence of anti-BRG1 (around 30%). Furthermore, the combined presence of ASOs enhanced the individual potential of each one in around 10% (Fig. 3B). The C. albicans biofilms formed in the presence of individual and combined of ASOs were also analysed in terms of number of cultivable cells, but no differences were observed among them with values of around 3logCFU/cm 2 of biofilm without reduction ( Figure S1). In addition, it was also tested the effect of the anti-EFG1 and anti-BRG1 2'OMe in the biofilm formation of their respective deletion mutant strains (C. albicans efg1DD and brg1DD). No effect was observed either in terms of total biomass quantification or cultivable cells enumeration ( Figure S2).

ASOs' effect on biofilm thickness
In order to analyze the influence of ASOs on C. albicans biofilms thickness, the biofilms were further visualized by CLSM (Fig. 4). The treatment of C. albicans biofilms with the ASOs, individual and combined status, caused a significant decrease on biofilm thickness of around of 25-50%. The highest reduction was observed in the biofilms treated with the cocktail of the three ASOs (P value \ 0.05).

ASOs' effect on C. albicans biofilm matrix composition
The content of the matrix of C. albicans biofilms developed in presence of combined ASOs were also analyzed. Table 2 summarize the values of total protein and carbohydrate contents obtained. It was evident that in all conditions tested, there was a significant impact on the production of C. albicans biofilm matrix components. Indeed, all ASOs caused a decrease of total carbohydrate content of around 30-50%, with the highest value (60%) observed for the cocktail of the three ASOs (P value \ 0.05). Regarding the matrix protein content, the presence of the ASOs induced a slight impact in its production (of around 20-30%).

Discussion
In the last decades, ASOs have been successfully applied for the treatment of many genetic human diseases [26][27][28]. However, the antisense technology has been poorly explored to control fungal virulence determinants [15,16]. Candida albicans remains the most prevalent of all fungi in the world [1] and one of its most problematic virulence factors is the ability to adapt to a variety of different habitats and to form surface attached microbial communities, known as biofilms [2]. The presence of biofilms has been identified as the main cause of an increased resistance to antifungal agents, due to the high density of cells present in the biofilm structure and the production of the protective extracellular matrix (34, 35). Candida albicans biofilm development is a fascinating intricate process involving finely altered genes expression and requiring complex and well-coordinated regulation [2,29]. Relevant transcription factors such as EFG1, BRG1 and ROB1 are involved in this clinical phenomenon [29]. Through this work, the performance of ASOs designed to hybridize specifically to BRG1, ROB1 and EFG1 [15] transcription factors (Table 1), was validated regarding their ability to control gene expression and to reduce C. albicans biofilm formation when applied in an individual and combined way. It has been described that the nucleic acid mimics must be designed with a GC content of approximately 50 to 60% in order to increase the binding affinity for target mRNA and stability in the human body [17,30]. Furthermore, several studies have shown that ASOs with sizes between 12 and 20nt (nucleotides) usually present a good hybridization performance [17,30].
Considering these features, the anti-BRG1 2'OMe sequence was designed with three 2'OMe modifications in each end with 50% of GC, and a total of 14nt. The anti-ROB1 2'OMe was designed with four 2'OMe chemical modifications, 38.9% of GC, and a total of 18nt ( Table 1).
The in vitro ASOs efficacy was evaluated at 40 nM since none of the combinations showed cytotoxicity in concentrations of 40 nM (Fig. 1). The results demonstrated the capacity of the individual and combined 2'OMe oligomers to control EFG1, BRG1, and ROB1 expression of 24 h C. albicans biofilm cells (Fig. 2). In the case of anti-BRG1 and anti-ROB1, the effect on gene expression was similar when applied individually or in combination with the other ASOs. Only in the anti-EFG1 2'OMe, the effect on gene expression was lower when the ASOs were combined comparatively with the effect of the individual ASO (Fig. 2B).The differences observed in terms of ASOs performance can be related to the different levels of expression of each gene ( Figure S3).
Moreover, the in vitro results also demonstrate the capacity of all the 2'OMe oligomers to reduce C. albicans biofilm formation by about 20% (Fig. 3A). Interestingly, it can be noticed that the cocktail of ASOs enhances the individual potential o each one in around 10% (Fig. 3B). Despite of knowing the importance of EFG1, BRG1 and ROB1 on regulation of C. albicans biofilm formation, the relevant role of other genes in this phenomenon is also assumed [2,29] thus, it would not be expected to obtain a total reduction on C. albicans biofilm.
Furthermore, the CSLM images (Fig. 4) reinforce the results concerning the levels of biofilm reduction once the application of the cocktail of ASOs decreased C. albicans biofilm thickness in around 50%. In addition to the cells that are present in the biofilm of C. albicans, it also produces an extracellular matrix, which act as a protection for biofilm cells from phagocytic cells and also as barrier to antifungal agents [31,32]. Candida albicans biofilm matrices are composed mainly by carbohydrates and proteins [32]. To note, ASOs enhance the control of C. albicans biofilm formation, by reducing the amount of matrix content ( Table 2). In fact, it was noticed a reduction of protein and carbohydrate contents in around 30 and 50%, respectively. This in an important result since the carbohydrate and protein contents of the biofilm matrix have been described as kidnappers/blockers of the action of antifungals action on yeast biofilms [33].
In this sense, this work revealed the importance of the application of the antisense therapy to prevent and control biofilm of Candida species and thus, the oligomers could become a new source of molecules to be used as adjuvant therapy to control candidiasis.

Conclusions
Our data reported a novel cocktail of oligomers that can control biofilm formation of C. albicans, which is an important virulence factor involved in infection. Furthermore, the combined application of the new cocktail of ASOs enhances the control of C. albicans formation, minimizing biofilm thickness by reducing the levels of matrix components. Thereby, this work provides valuable information for future research to control Candida infections.