Effects of Bacillus subtilis on Candida albicans : biofilm formation, filamentation and gene expression

Michelle Peneluppi SILVA1, Patrícia Pimentel de BARROS1, Adeline Lacerda JORJÃO1, Rodnei Dennis ROSSONI1, Juliana Campos JUNQUEIRA1, Antonio Olavo Cardoso JORGE1. 1 – Univ Estadual Paulista (Unesp) Institute of Science and Technology, São José dos Campos Department of Biosciences and Oral Diagnosis – São José dos Campos – SP Brazil. doi: 10.14295/bds.2019.v22i2.1692 UNIVERSIDADE ESTADUAL PAULISTA “JÚLIO DE MESQUITA FILHO”


INTRODUCTION
C andida albicans is a commensal fungus found on mucosal surfaces of the oral cavity, gastrointestinal and genitourinary tracts. This yeast causes severe and recurrent mucosal infections as well as fatal invasive infections in both immunocompromised and immunocompetent individuals [1].
Some virulence factors such as dimorphism and the ability to adhere and form biofilm on medical devices and/or the host mucosal epithelium, enhance the pathogenicity of C. albicans. The cells in biofilms exhibit distinct phenotypes that are similar to their free-living counterparts, including extreme resistance to many antimicrobial agents. Advances in genetic manipulation and expression profiling have helped define the regulatory pathways and mechanisms that govern C. albicans biofilm development [1,2,3]. Several classes of genes, such as specify transcription factors (EFG1, BCR1, TEC1) and adhesion (HWP1, ALS3), control biofilm development on abiotic and biotic surfaces [4,5].
In the recent past, microbial secondary metabolites have engrossed great attention of researchers as antifungal compounds. Most of the antifungal compounds currently in use are synthetic derivatives with known serious side effects and toxicity [6]. Emergence of drug resistant Candida has augmented the desire for new biocompatible antifungal drugs [7,8].
The genus Bacillus includes a great diversity of industrially important strains, including Bacillus subtilis. This spore-forming bacterium has been established as industrial bacteria in the production of biological indicators for sterilization, in studies of biodefense and astrobiology methods as well as disinfection agents, in treatment evaluation and as potential adjuvants or vehicles for vaccines, among other applications. The metabolic diversity and lack of reported incidenceof pathogenicity makes this bacterium the most suitable for the development of marketable products [9].
Bacillus strains could produce some metabolites that inhibit pathogens, since some Bacillus species used in commercially available products have the ability to produce antimicrobials, such as aminocumaim A [10] and bacteriocin [11,12]. Zheng and Slavik [11] isolated and characterized a bacteriocin (molecular weight of approximately 34 kDa and a pI value of approximately 47) of B. subtilis 22. Firstly, the precipitated supernatant from the B. subtilis 22 culture had inhibitory activity in nine strains of Gram-negative and Gram-positive bacteria by agar diffusion assay. After partial purification of supernatant, the bacteriocin was named bacillocin 22 making this peptide potentially useful against the food-borne pathogens. In a recent study, Mayer and Kronstad [13] have identified the soil bacterium, Bacillus safensis, as a potent inhibitor of virulence factor production by two major fungal pathogens of humans, Cryptococcus neoformans, and C. albicans. They determined that the anti-virulence factor mechanism was, at least in part, baseded on production of bacterial chitinases that target and destabilize the fungal cell surface.
In this context, in an attempt to investigate new compounds and/or metabolites with antifungal properties produced by a sporulated bacillus, the current study was evaluate the effect of B. subtilis on C. albicans biofilm formation and filamentation by evaluating the gene expression of ALS3, HWP1, BCR1, EFG1 and TEC1.

Microorganisms and growth conditions
In this study, we used the reference strains of B. subtilis (ATCC 19659) and C. albicans (ATCC 18804); these strains were plated on Brain Heart Infusion Agar (Difco, Detroit, USA) and Sabouraud Dextrose Agar (Difco, Detroit, USA), respectively, and were incubated at 37 °C for 24 h. After growth, bacterial colonies were transferred and cultured in Brain Heart Infusion Broth (BHI) (Difco Laboratories Inc., Detroit, MI, USA) and the yeast in Yeast Nitrogen Base -YNB (Difco Laboratories Inc., Detroit, MI, USA) at 37 °C for 24 h.

Biofi lm formation
Suspensions of each microorganism sample were adjusted to 10 7 cells/mL using a spectrophotometer (B582, Micronal, São Paulo, Brazil). The formation of monotypic and mixed biofilms was based on the methodology described by Seneviratne et al. [14] and Bridier et al. [15] with modifications. Biofilms were formed in the bottom of a 96-well microtiter plate (TPP®, Trasadingen, Switzerland), by pipetting 100 µL of a standardized suspension containing 10 7 cells/mL of each microorganism, C. albicans and B. subtilis, associated in each well to the formation of a mixed biofilm. For monotypic biofilm formation, 100 µL of only one microorganism and 100 µL BHI broth were pipetted in each well. The plates were incubated under shaking at 75 rpm rotation (Quimis, Diadema, Brazil) for 90 min at 37 °C with only the microorganisms to facilitate initial adhesion.
After adhesion phase, the suspension of non-adhered microorganisms was aspirated, and each well was washed with 200 µL of 0.85% NaCl to remove the non-adhered cells; 200 µL of BHI broth was added in all formed biofilms, both monotypic and mixed groups. All plates were incubated at 37°C for 48 h under shaking at 75 rpm rotation. After 24 h, the broth was changed and after 48 h, the biofilms were aspirated and washed with 0.85% NaCl to remove the non-adhered cells. Biofilms were evaluated in relation to cell viability (quantification of the colony forming units -CFU/mL) and gene expression in interspecies interaction. For each of these evaluations the experiment was performed in triplicate at two different times with 10 trials for each experimental group, resulting in 180 trials.

Quantifi cation of viable cells (CFU/ mL)
After washing the biofilm, 200 µL of 0.85% NaCl was added in each well for homogenization during 30 s in an ultrasonic homogenizer (Vibra Cell-Sonics & Materials, Inc. Newtown, Connecticut, USA), with 25% amplification to detach the biofilms. Serial dilutions of suspensions were performed, and 100 µL aliquots of each dilution were plated in the culture medium, according to the microorganism: BHI Agar supplemented with fluconazole for B. subtilis and HiCrome Candida Differential Agar (Himedia, Mumbai, India) for C. albicans. The plates were incubated at 37 °C for 48 h to count the CFU/mL and converted into logarithm to base 10 (Log).

C. albicans fi lamentation production
The filamentation assay was performed in 24-well microtiter plates (TPP®, Trasadingen, Switzerland) following a methodology described previously with modifications [16]. The experimental groups formed were: C. albicans +PBS (Phosphate Buffered Saline) (control group) and C. albicans + B. subtilis (interaction group). After 18 hours of growth C. albicans and B. subtilis were centrifuged to prepare standard suspensions corresponding to 10 7 cells/mL in a spectrophotometer as previously described. In 24-well cell culture plates, 1 mL distilled water supplemented with 10% bovine fetal serum was added to a 100 µL C. albicans suspension. Then, a 100 µL B. subtilis suspension was added to each well. The control group used 100 µL PBS.
The plate was incubated at 37°C for 6 h and 50 µL of the inoculum contained in each well was placed on microscopic glass slides with 10 fields previously marked on the back and observed under an optical microscope with a magnification of 400 x. The images were analyzed in relation to C. albicans morphology and hyphae quantification in 10 microscopic fields per slide and classified as scores from 0 to 4 attributed to each field according to the number of hyphae present: score 0: no hyphae; score 1: 1 to 20 hyphae; score 2: 21 to 40 hyphae; score 3: 41 to 50 hyphae; and score 4: more than 50 hyphae. Each experimental group had 5 trials at two different times, totaling 20 tests. The percentage of germination was calculated considering the mixed group formed by C. albicans-B. subtilis in relation to monotypic group of C. albicans.

Evaluation of ALS3, HWP1, BCR1, EFG1 and TEC1 gene expression with qPCR
The evaluation of C. albicans virulence genes was based on the methodology described by Barros et al. [17] with modifications. Biofilms were formed in 24-well microtiter plates (TPP®, Trasadingen, Switzerland) with the same concentration and conditions as in the "Biofilm formation" section. The expression of ALS3, HWP1, BCR1, EFG1 and TEC1 genes was evaluated in relation to four reference genes, ACT1 (β-Actin), RIP1 (Ubiquinol Cytochrome-c Reductase Complex Component), PMA1 (Adenosine Triphosphatase) and LSC2 (Succinyl-CoA Synthetase β-Subunit Fragment), described in the literature [18], in all experimental groups. The results were analyzed at http:// www.leonxie.com/referencegene.phpe, and the reference gene RIP1 was selected for this experiment. Transcribed cDNA was amplified for the relative quantification of ALS3, HWP1, BCR1, EFG1 and TEC1 gene expression levels in relation to the concentration of the selected reference gene, RIP1. Quantitative real-time PCR was conducted using the Platinum® SYBR® Green qPCR SuperMix-UDG Kit (Applied Biosystems, Framingham, MA, USA) in the StepOnePlus™ apparatus (Applied Biosystems, Framingham, MA, USA). The 2 -DDCT method was used to analyze the relative changes in gene expression from the quantitative RT-qPCR experiment [19].

Statistical analysis
Student's t-test was used to compare the CFU/mL results from the in vitro biofilm formation assay and the relative quantification of gene expression. The scores obtained from the filamentation analysis were compared using the Mann-Whitney test. All analyses were performed using the GraphPad Prism 5 Program (GraphPad Software, Inc., USA) and 5% level of signifcance was adopted.

RESULTS
The CFU/mL count of C. albicans in monotypic biofilms showed higher values compared to mixed biofilms formed by associations between C. albicans and B. subtilis ( Figure 1). Comparing the results obtained in monotypic biofilms with mixed biofilms, there was a reduction of 1.0 log of C. albicans with a statistically significant difference (p = 0.001). B. subtilis was not affected in the mixed biofilm compared to monotypic (p = 0.7002). The results, expressed in mean and standard deviation, for CFU/mL counts of C. albicans in monotypic biofilms were 5.38 ± 0.149 log and 4.37 ± 0.183 in mixed biofilms. For B. subtilis, the means were 5.33 ± 0.202 log in monotypic biofilms and 5.11 ± 0.179 in mixed biofilms.
In relation to the filamentation test, the results showed reduction of morphological transition (yeast/hyphae) for C. albicans in the presence of B. subtilis (C. albicans and B. subtilis interaction), with a statistically significant reduction (p=0.0001) compared to the control group with PBS ( Figure 2).
To elucidate the inhibitory mechanisms of B. subtilis on C. albicans biofilms, this study was extended to analyze C. albicans gene expression, concentrating on genes involved in biofilm formation and filamentation. The expression levels of the adhesion genes (ALS3 and HWP1) and transcriptional regulatory genes (TEC1, BCR1 and EFG1) were quantifed in C. albicans cells from monotypic and mixed biofilms using qPCR. In C. albicans and B. subtilis association, ALS3, HWP1, BCR1, TEC1 and EFG1 genes were downregulated when compared to the C. albicans control group with statistically significant differences (p=0.0001) (Figure 3). Among all of the analyzed genes, the ALS3 and HWP1 genes were the most affected, achieving 111.1-and 333.3-fold decreases in the C. albicans biofilms associated with B. subtilis, respectively. These results agreed with those obtained for the CFU/mL counts and filamentation assay and suggest that B. subtilis exerts inhibitory effects on C. albicans by interfering with yeast adherence during biofilm formation and reducing its development during filamentation.   The units in the Y-axis were calculated based on the 2 -DDCT method, and are expressed as the means and standard deviations. Each gene was normalized and compared with the expression of monotypic biofilms of C. albicans (control groups) using the reference gene RIP1. Student's t-test was used to compare gene expression between the groups (p ≤ 0.05). ***indicates p ≤ 0.0001.

DISCUSSION
Biofilm growth is a main model of microbial life and up to 80% of all microorganisms are sessile attached to the surface of biofilm communities. Within this group of microorganisms, 65 to 80% can cause microbial infections in humans from pathogenic biofilm formation [20]. The capability of biofilm formation and morphological transformation from yeast to hyphae are important to C. albicans pathogenicity [21]. Biofilms create an area of higher antimicrobial resistance for the residing microorganisms [20]. Thus, the interactions between bacteria and fungi can influence survival and proliferation, which can be beneficial for microorganisms because it increases resistance to antimicrobials or antagonistic that lead to death [22].
In this study, the results demonstrated that C. albicans was reduced in association with B. subtilis during biofilm formation. The quantification of CFU/mL showed a reduction of 87.44% in C. albicans cells when compared to the control group. Thus, it demonstrated that in mixed biofilm the microorganisms interacted competitively and antagonistic, suggesting an inhibitory effect of B. subtilis on C. albicans. A possible explanation for the reduction of biofilm formation by C. albicans during the microbial interaction is the fact that Bacillus spp. secrete antimicrobial agents, such as bacteriocins, which have effects for suppressing the growth of competitive and opportunistic microorganisms [8,9]. One example is Subtilisin A, which is a bacteriocin isolated from B. subtilis [23]. B. subtilis in mixed biofilms did not present a significant reduction in relation to the monotypic biofilm. According to López et al.
[24] B. subtilis uses a mechanism called cannibalism to slow its sporulation, and these cannibal cells secrete two toxins, skf and sdp, in which they can interfere with the growth of other microorganisms, constituting as a defense mechanism of B. subtilis.
Furthermore, we analyzed hyphae formation in C. albicans, and the results showed a significant reduction in the number of hyphae when compared to the control group of C. albicans, which confirms the results previously obtained by the inhibitory effect exerted by B. subtilis on C. albicans biofilm. Ability of Candida to reversibly switch between budding yeast and filamentous form is crucial for establishing infection. This phenotypic plasticity is tightly regulated by several environmental cues and internal signaling pathways [25]. The hyphal formation is related to gene expression that encodes virulence factors, such as the hyphae wall proteins HWP1 and ALS3 [25].
To confirm the results obtained in the CFU/ mL and filamentation assays, we analyzed the expression of ALS3, HWP1, BCR1, EFG1 and TEC1, which are C. albicans virulence genes responsible for biofilm formation and morphogenesis of yeast, by the relative quantification method qPCR [21,25]. All of these genes were significantly downregulated compared to monotypic biofilms, demonstrating the reduction effect of B. subtilis on morphogenesis and biofilm formation in C. albicans. HWP1 and ALS3 were the most downregulated genes in the presence of B. subtilis, and this result can be associated with the decrease in the CFU counts, the hyphae formation interfering in biofilm formation. Our findings are in agreement with Rautela et al. [8] that evaluated the influence of cyclic lipopeptides, fengycin and iturin, produced by Bacillus amyloliquefaciens strain AR2 in C. albicans biofilms. They showed that the fungicidal activity was concentration dependent and sub-minimal concentrations of the lipopeptide showed a decrease in the hydrophobicity of the cell surface, formation of germinative tubes and reduced of the hyphae specific genes, such as HWP1 and ALS3.
Our results show the potential antifungal effect of B. subtilis, since this species interferes in the transcriptional analysis of C. albicans. Recently, Subramenium et al. [26] evaluated the antibiofilm effect of B. subtilis, a marine bacterial isolate from Palk Bay against C. albicans. This bacterium was able to reduce the biofilm of C. albicans by up to 90%, the filamentation and production of hydrolytic enzymes. In addition, the genes ERG11, FLU1, CDR2 and CDR1 (involved in drug resistance mechanisms) were downregulated when associated with B. subtilis.
The interactions between bacteria and fungi are highly complex and many factors need to be analyzed together, such as microorganism virulence associated with environmental factors and interactions between different species of microorganisms present in certain niches, requiring proper development of in vitro and in vivo models to characterize these interactions. The results from the experimental conditions described in this study demonstrated that B. subtilis inhibited C. albicans, suggesting that this bacterium produced compounds and/or metabolites with antifungal activity.

CONCLUSION
In conclusion, this study demonstrated that B. subtilis reduced the biofilm formation and filamentation of C. albicans by negatively regulating the ALS3, HWP1, BCR1, EFG1 and TEC1 genes that are essential for the production of biofilm and hyphae.