Activity of Novel Synthetic Peptides against Candida albicans

Candida spp. are the most common causes of fungal infections worldwide. Among the Candida species, Candida albicans remains the predominant species that causes invasive candidiasis in most countries. In this study, we used two peptides, KABT-AMP and uperin 3.6 as templates to develop novel antifungal peptides. Their anticandidal activity was assessed using a combination of MIC, time-killing assay and biofilm reduction assay. Hybrid peptides, KU2 and KU3 containing a mixed backbone of KABT-AMP and Uperin 3.6 demonstrated the most potent anticandidal activity with MIC values ranging from 8–16 mg/L. The number of Trp residues and the amphipathic structure of peptides probably enhanced the anticandidal activity of peptides. Increasing the cationicity of the uperin 3.6 analogues resulted in reduced MIC from the range of 64–128 mg/L to 16–64 mg/L and this was also correlated with the antibiofilm activity and killing kinetics of the peptides. Peptides showed synergistic effects when used in combination with conventional antifungals. Peptides demonstrated low haemolytic activity but significant toxicity on two normal human epithelial cell lines. This study provides us with a better understanding on the structure-activity relationship and the balance between cationicity and hydrophobicity of the peptides although the therapeutic application of the peptides is limited.

faces) molecules which have been found in most living organisms ranging from bacteria to plants, invertebrate and vertebrate species 13,14 . They act as the first line of defense and encompass direct antimicrobial activity against a broad spectrum of invading pathogens including bacteria, fungi, viruses and protozoa by using different modes of action as described previously 15,16 . Thus, AMPs can be potentially promising candidates for development of novel therapeutic agents against pathogens.
Several approaches such as template-based design, biophysical modelling, and computer-aided design have been proposed to design synthetic AMPs 17 . Template-based strategy is known as de novo design of peptide analogues based on natural AMPs with known antimicrobial activity. This approach is often more convenient and has been extensively applied by researchers to elucidate the structureactivity relationship of peptides and further enhance the antimicrobial potency of natural AMPs with low toxicity to mammalian cells 18,19 . In our study, KABT-AMP, a synthetic peptide 20 and uperin 3.6, a naturally occurring peptide 21 with known activity were selected and modified in an attempt to improve their antifungal activity. Physicochemical properties including size, residue composition, overall charge, hydrophobicity, and amphipathic structure were taken into consideration while designing synthetic AMPs.

Results
Peptide design. KABT-AMP and uperin 3.6 peptides were selected as templates for designing synthetic AMPs. Uperin 3.6 adopts a welldefined amphipathic a-helix, with hydrophobic side above and hydrophilic side below the polypeptide backbone. Synthetic modification performed by Chia et al had revealed that three lysine residues in the sequence of uperin 3.6 are crucial for antimicrobial activity 21 . To further enhance the antimicrobial activity of uperin 3.6, three lysine-substituted analogues of uperin 3.6 (Upn-lys4, Upn-lys5, and Upn-lys6) were designed by replacing less hydrophobic amino acids with cationic lysine residues so that activity variation depends solely on altering the charge. The overall charge of a parent peptide was increased to 14, 15 and 16 respectively without disrupting the amphipathic structure of the peptide. In addition, four hybrid peptides (KU1, KU2, KU3, and KU4) were designed by selecting and fusing peptide fragments from KABT-AMP and uperin 3.6. KU1 was designed by truncating the first nine residues at the N-terminus of KABT-AMP which was then fused to the 9 th to 17 th C-terminus of residual fragment (VVNVLKNLF-NH 2 ) of uperin 3.6. Likewise, KU2, KU3 and KU4 were also designed using the same approach by truncating first eleven, twelve and thirteen residues at N-terminus of KABT-AMP, respectively, and fused with the respective 11 th to 17 th , 12 th to 17 th and 13 th to 17 th C-terminus residues fragment of uperin 3.6 with amidation. The sequence and physicochemical properties of all designed peptides are provided in Table 1. The amphipathic orientation of the residues in helical structure of four hybrid peptides are displayed using the helical wheel projection (Figure 1).
Anticandidal activity of designed peptides. The MIC of the reference and designed peptides against C. albicans, C. krusei and C. parapsilosis were determined as shown in Table 2. Conventional antifungal agents such as fluconazole and amphotericin B were included in this study as controls. Among the hybrid peptides, KU2 and KU3 demonstrated the most potent antifungal activity, showing MIC values ranging from 8 to 16 mg/L against four Candida strains. KU1 and KU4 also displayed good activity, with MIC values ranging from 16 to 32 mg/L and 16 to 64 mg/L respectively. All four KU hybrid peptides showed improved anticandidal activity when compared to their parent peptides KABT-AMP (MIC of 32 to 64 mg/L) and uperin 3.6 (MIC of 64 to 128 mg/L).
In terms of uperin 3.6 analogues, substitution of a single Lysine residue into the parent peptide (Upn-lys4) did not improve the antifungal activity against C. albicans strains and C. krusei. Upn-lys5 and Upn-lys6 displayed two-fold reduction in MIC values against C. albicans and C. krusei (MIC of 32 to 64 mg/L) suggesting that replacement of two lysine residues was required to improve the antimicrobial activity of the parent peptide by two-fold. With the exception of C. parapsilosis, the antimicrobial activity of uperin 3.6 was improved by twofold as the number of lysine residue substituted into template peptide increased. As for the standard antifungal drugs, amphotericin B has the most potent antimicrobial activity with MIC values ranging from 0.5 to 1 mg/L against C. albicans and non-albicans species. For fluconazole, two C. albicans strains and C. parapsilosis showed MIC values ranging from 1 to 2 mg/L while C. krusei, a species that is intrinsically resistant to fluconazole, demonstrated MIC value of 64 mg/L.
Biofilm reduction assay. Biofilms formation is often associated with antifungal resistance as compared to planktonic cells and requires drug concentrations of 30-2000 times the corresponding MIC values to reduce 50% of the biofilms metabolic activity 22 . The susceptibility of the fungal biofilms to the designed peptides was assessed by using XTT reduction assay, which enables quantitation of the number of living cells in 24 hour-old biofilms after treatment. Based on the results shown in Figure 3, the biofilm metabolic activity of which Candida decreased as the concentration of peptides increased from 13 to 43 the planktonic MIC. KU4 with a BEC-2 value of 96 mg/L demonstrated the highest ability to eradicate biofilms among the hybrid peptides. The BEC-2 value for uperin 3.6 was 192 mg/L while its analogues, upn-lys4, upn-lys5 and upn-lys6 showed BEC-2 values of 192 mg/L, 128 mg/L and 96 mg/L, respectively. The parent peptide KABT-AMP exhibited good antibiofilm activity with a BEC-2 value of 64 mg/L. In contrast, the conventional antifungal agent, amphotericin B, displayed the most potent antibiofilm activity with a BEC-2 value of ,1 mg/L while fluconazole had a BEC-2 value of .4 mg/L (.4 3 MIC) ( Table 3).
Growth inhibitory effect of the peptide-peptide and peptideantifungals combination. The in vitro synergism effects of the peptide-peptide and peptide-antifungals combination were www.nature.com/scientificreports SCIENTIFIC REPORTS | 5 : 9657 | DOI: 10.1038/srep09657 identified by using a modified checkerboard method. Most of the peptide-peptide combinations showed no synergistic effect against C. albicans SC5314 and ATCC 90028 strains and were in the additive or indifferent range (Supplementary data: Table S1). All the peptideamphotericin B combinations demonstrated synergistic effects with FIC indices of less than 0.5 when tested against C. albicans SC5314 strain. In contrast, there are only four peptide-fluconazole combinations (KU4, KABT-AMP, Upn-lys6 and uperin 3.6) that displayed synergistic effects against SC5314 strain. However, most of the peptides inhibited the growth of ATCC 90028 strain when   tested in combination with amphotericin B except Upn-lys6 and uperin 3.6. In term of fluconazole, KU4 was the only peptide displaying a synergistic effect when used in combination with fluconazole against both strains.
Cytotoxicity of peptides on normal human cells. Haemolytic assay was carried out to determine the ability of the peptides to disrupt the membrane integrity of mammalian cells. The hybrid peptide KU1 showed haemolytic activity (.10%) at concentrations lower than the MIC doses. KU2 and KU3 also displayed moderate haemolytic activity with HC 10 and HC 50 of 7.33 6 0.73 mg/L and 57.71 6 9.80 mg/L for KU2 and 5.65 6 0.85 mg/L and 55.20 6 11.90 mg/ L for KU3. KU4 was the hybrid peptide with the least haemolytic activity (HC 10 5 49.38 6 5.14 mg/L and HC 50 . 256 mg/L). Neither uperin 3.6 nor its analogs exhibited toxicity to human red blood cells (HC 10 and HC 50 of more than 256 mg/L) ( Table 4). The cytotoxicity of the designed peptides against human vaginal (VK2/E6E7) and esophagus (Het-1A) epithelial cell lines was assessed after 24, 48 and 72 hours of treatment by using MTS assay (Table 5). Significant toxic effects of the hybrid peptides and uperin 3.6 analogs were observed against VK2/E6E7 and Het-1A cell lines. Hybrid peptides and uperin 3.6 analogs reduced the cell viability to 50% at concentrations below the MIC doses. Uperin 3.6 and its analogs were highly toxic to both cell lines with low IC 50 values although they did not exert haemolytic activity on human erythrocytes.

Molecular interaction of designed peptides with SAP1, SAP 5 and exo-beta-(1, 3)-glucanase
Results obtained from rigid docking performed by Autodock Vina software indicated that all peptides bound to three selected target proteins respectively were within the negative binding affinity range (Supplementary data: Table S2). Peptides with greater negative binding affinity range indicate stronger binding affinities with target proteins. In comparison with peptides in respective series, KU3 and Upn-lys5 with lower MIC values showed higher binding affinities for three target proteins and hence, they were selected for further analysis. Parent peptide KABT-AMP and uperin 3.6 were included as a control (Supplementary data: Table S2).
The total interaction energy of the target proteins and peptides, which is the sum of the van der Waals (VDW) and electrostatic energy within 3 Å , is tabulated (Supplementary data: Table S3, S4 and S5). KU3 shows strong binding interactions with the target proteins as compared to other peptides. For sap1, KABT-AMP showed slightly stronger interaction than KU3 mainly due to the electrostatic interaction with Asp37, Asp191 and ASP214. KU3 was found to exhibit interaction with Glu10, Asp32, Asp37, Asp54, Asp86, Glu295 and Asp308 of sap5 while for Exg, KU3 has interactions with Asp145, Asn191, Glu192, Gly291, Gly292, Asn299 and Asp318 residues. All peptides demonstrated interactions with Glu192 and Glu292 of Exg since both residues are catalytic sites. From the result, Upn-lys5 showed higher total interaction energy than Uperin3.6 for sap1 (2453.603 kcal/mol), sap5 (21004.134 kcal/mol) and Exg (2627.185 kcal/mol).

Discussion
The increased incidence of invasive fungal infections and the emergence of antifungal drug resistance in Candida spp., particularly against echinocandins and azoles, have prompted the development of alternative antifungal agents 12 . Antimicrobial peptides (AMPs) can be promising candidates for novel antifungal agents due to their broad-spectrum activities against pathogens and low chances of developing antimicrobial resistance 23 . Some naturally occurring antifungal peptides such as polymyxin, melittin and protegrin demonstrated potent antimicrobial activity, but their therapeutic application is limited due to the toxicity effect on mammalian cells [24][25][26] . Therefore, sequence modification of natural peptides with known activity is essential to enhance the antimicrobial activity and reduce the toxicity of peptides. In this study, KABT-AMP, a synthetic peptide and uperin 3.6, a natural peptide isolated from the Australian toadlet, Uperoleia mjobergii were selected as a template to design a new class of peptides 20,21 . Physicochemical parameters such as cationicity, hydrophobicity and amphipathicity were taken into consideration for peptide design (Table 1).
Hydrophobicity is one of the crucial parameters responsible for the antimicrobial activity of AMPs. It facilitates the AMP-membrane interaction and governs the extent of partition of AMPs into the cell membrane. Studies showed that enhancement of the overall hydrophobicity increased the antimicrobial potency of AMPs [27][28][29] . Based on GRAVY score, KU1 had the highest hydrophobicity (positive GRAVY score) among the hybrid peptides. However, the MIC values obtained show that KU2 was more potent than KU1 although KU2 is more hydrophilic (negative GRAVY score). By observing the helical projection of the peptides, we found that the enhancement of anticandidal activity of KU2 can be due to the number of tryptophan (Trp) residues incorporated in the peptide sequences. Tryptophan (Trp) is a hydrophobic residue that has a strong preference for the interfacial region of the lipid bilayers of the yeast cell membrane 30 . Replacement of Val 10 at KU1 by Trp residue resulted in three Trp residues (Trp 3 , Trp 6 and Trp 10 ) positioned at the nonpolar face of KU2 and hence, enhanced the anchoring of the peptide to membrane (Figure 1b). By comparing KU2 and KU3, substitution of Asn 12 by Lys residue at the hydrophilic face has no effect on the anticandidal activity (Figure 1c). KU4 demonstrated the least anticandidal potency among the hybrid peptides as a result of the substitution  of hydrophobic Val 13 by Lys residue which reduced the hydrophobicity of the peptide (Figure 1d). These results are consistent with a previous study 18 . Cationic residues on the peptide promote the electrostatic attraction of the peptides to the anionic phospholipids in fungal membranes such as phosphatidylserine and phosphatidylinositol or wall components such as mannoproteins in yeast 31,32 . According to the MIC results of uperin 3.6 lysine-substituted analogues, substitution of a single lysine residue into the parent peptide (Upn-lys4) exhibited no improved activity against C. albicans strains and C. krusei. Upn-lys5 and Upn-lys6 displayed two-fold increase in potency against C. albicans as one or two Lys residues were substituted into the parent peptide. This suggested that further increase in the number of positive charge residue on Upn-lys6 is required as it may  (Figure 2a). In term of uperin 3.6 analogues, the killing effects were enhanced as the number of cationic lysine residues increased (Figure 2b). Previous studies on magainin analogues reported that increased cationicity increased the translocation of peptides across the membrane 33 . However, uperin 3.6 with 13 net charge displayed an exception in this case. It exhibited fungicidal effect after 4 hours of test period which is almost similar to the killing kinetic demonstrated by Upn-lys6. This is probably due to the presence of negatively charged aspartic acid, D in uperin 3.6, which may contribute to the electrostatic interaction of the peptide with yeast cell membrane.
C. albicans is the most common fungal species associated with biofilm formation 34 . Mature Candida biofilms display a complex three-dimensional structure consisting of a dense network of yeasts, hyphae and pseudohyphae and are frequently encountered on the surface of most implanted medical devices 34 . Candida biofilms are more resistant to host defense mechanism and antifungal agents such as fluconazole and amphotericin B. The resistance can be up to 1000fold greater than planktonic cells, depending on both drug and yeast species 35,36 . For AMPs, Burrows et al demonstrated that the most potent peptide in their studies, dF 21-10K, was less effective at 5 3 MIC and a concentration of 10 3 MIC was required to eradicate biofilm cells completely 37 . In our study, although KU1, KU2 and KU3 showed potent anticandidal activity against planktonic cells, they were unable to reduce biofilm viability to 50% even at 4 3 MIC (BECs . 4 3 MIC) ( Table 3). In contrast, KU4 and KABT-AMP with higher MIC values demonstrated potent antibiofilm activity. In term of uperin 3.6 analogues, we found that the BEC-2 values reduced as the number of lysine substituted increased. Hence, we suggest that the enhanced antibiofilm activity of the peptides against C. albicans may have a correlation with the increased cationicity of the peptides.
Drug combination therapy which involves the use of two or more drugs with separate mode of action can potentially improve the current available antimicrobial treatment. Combination regimens, in certain cases could potentially improve the efficacy, reduce toxic side effects and the cost of each individual drug and limit the development of antifungal resistance as shown in monotherapy 38 . A number of studies reported that AMPs exhibited synergistic effect with conventional antifungal agents [38][39][40] . In the current study, all peptide-amphotericin B combinations demonstrated synergistic activity (FICI # 0.5) when tested on C. albicans SC5314 strain. For C. albicans ATCC 90028 strain, all peptides exhibited synergistic effects with amphotericin B with the exception of Upn-Lys6 (FICI 0.58) and uperin 3.6 (FICI 0.67). With regard to fluconazole, there were four peptide-fluconazole combinations (KU4, KABT-AMP, Upn-lys6 and uperin 3.6) and only one KU4-fluconazole combination displayed synergistic effects against C. albicans SC5314 strain and ATCC 90028 strain, respectively. In general, two compounds that interact in a synergistic way most probably exert antimicrobial activity in a different mechanism of action. Amphotericin B demonstrated fungicidal effect by binding to the fungal cell membrane ergosterol moiety and causing porosity on the fungal membrane. Fluconazole is a fungistatic azole-based antifungal which interferes with the key enzyme involved in the ergosterol biosynthesis and thereby inhibits the formation of ergosterol. In term of AMPs, various mechanisms of action have been proposed which include the ability of AMPs to damage cell membrane, interact with intracellular targets such as inhibition of cell wall, nucleic acid and protein synthesis, and induction of apoptosis 41 . Therefore, as the peptides synergistically interacted with fluconazole and amphotericin B, this might indicate that their mode of actions differ from both antifungals, and the synergistic effects could be attributed to the combination of two different mechanism of actions.
Although the peptides designed in this study demonstrated improved anticandidal activity, their therapeutic potential was hindered by toxicity. KU1, KU2 and KU3, which had the greatest hydrophobicity among hybrid peptides, demonstrated significant toxicity to red blood cells with low HC 10 and HC 50 values. Peptides with higher hydrophobicity tend to penetrate deeper into the hydrophobic core of red blood cell membrane and cause stronger haemolytic activities. In term of KU4, less haemolytic activity was observed due to the overall low hydrophobicity. Besides that, we found that nonpolar helix face of KU4 was disrupted by Lys 13 residue and hence, reducing the penetration of the peptide into the cell membrane 27,42 . In term of uperin 3.6 analogues, no haemolytic activity was observed although they exhibited greater overall hydrophobicity compare to other hybrid peptides. Hence, the haemolytic activity was probably attributed to the tryptophan residues present in the hybrid peptides 18 . However, uperin 3.6 analogues induced significant toxicity against normal vaginal cell and esophagus epithelial cell lines although they did not lyse the erythrocytes. We propose that the mechanism of action of uperin 3.6 and its analogues could be organelle or nucleus dependent as they did not show toxicity effect on enucleated red blood cells 18 . Therefore, further investigations on  these peptides are needed to elucidate the mechanisms of action of these peptides to kill Candida cells. Molecular docking investigations enable clarifications of the binding mode of the peptides with the possible target proteins (sap1, sap5 and Exg). According to docking results, all peptides bound well with the target proteins since they demonstrated negative binding affinities to the target proteins (Supplementary data: Table S2). The interaction energy which is the sum of van Der Waals (VDW) and electrostatic interaction determines the interaction between peptides and target proteins. Based on the results of the interaction energy obtained, we suggest that KU3, KABT-AMP, Upn-lys5 and uperin 3.6 exhibit stronger interaction with sap5 as compared to sap1 and Exg.

Conclusion
This study provides us a better understanding of the structure-activity relationship of antifungal peptides although the therapeutic applications of the peptides are limited due to toxicity. Several parameters such as hydrophobicity, the number of tryptophan residues, amphipathicity and cationicity are crucial for anticandidal activity of the peptides. We demonstrated that the antibiofilm activity and the killing kinetics are correlated with the cationicity of the peptides. The designed peptides also showed synergistic effects when used in combination with conventional antifungals and hence, this may result in reduced probability of resistance development in Candida spp. against the conventional antifungals. Further investigations are needed to elucidate the mechanism of action and mode of killing of the peptides.

Methods
Yeast strains and growth conditions. C. albicans strains SC5314 and ATCC 90028, C. krusei ATCC 6258 and C. parapsilosis ATCC 22019 were used in the study. The strains were stored at 280uC in nutrient broth supplemented with 2.0% glycerol stock. Prior to experimentation, each strain was streaked for single colonies on Sabouraud Dextrose (SDA) agar plates and incubated overnight at 37uC.
Peptide synthesis. Peptides were synthesized to .90% purity by Mimotopes Pty Ltd (Clayton, Victoria, Australia. http://www.mimotopes.com/). The characterization and purity estimation of the peptides were performed by using electrospray mass spectrometry (ESMS) and reverse phase high performance liquid chromatography (RP-HPLC). All peptides were dissolved in sterilized water to 10 mg/ml and stored in aliquots at 220uC.
MIC determinations. Antifungal activity of the designed peptides was determined according to a standardized broth microdilution method (Clinical and Laboratory Standards Institute (CLSI) document M27-A2). Briefly, the yeast colonies from 24hour-old cultures of Candida species were picked and resuspended in 5 mL of sterile 0.145 mol/L saline and adjusted to cell density of 1 3 10 6 to 5 3 10 6 cells/ml. The yeast stock suspension was then diluted to obtain a starting inoculum of 5.0 3 10 2 to 2.5 3 10 3 cells/ml. Conventional antifungals such as amphotericin B (Sigma, USA) and fluconazole (Sigma, USA) were included in this study as controls and to compare the anticandidal activity with the designed peptides. The drugs were dissolved in DMSO and prepared to a stock concentration of 25 mg/L and 50 mg/L, respectively. Peptides and the antifungals were then serially diluted in RPMI 1640 growth medium buffered with morpholinepropanesulfonic acid (MOPS) in volume of 100 mL per well, giving final concentrations ranging from 256 mg/L to 0.5 mg/L in sterile Ubottomed 96-well polypropylene microplates. 100 mL of standardized yeast suspension was then added to each well. Plates were incubated for 48 hours at 37uC. The minimum inhibitory concentration (MIC) was defined as the lowest concentration that inhibited 90% growth of Candida species. The MICs were determined three times and each time in duplicate.
Time killing assay. Killing kinetics of the peptides were determined by diluting the peptides with RPMI medium to a final concentration of 2 times the previously determined MIC for the strain to be tested. A fungal suspension was adjusted to a 0.5 McFarland turbidity standard with cell density of 1 3 10 6 to 5 3 10 6 cells/ml. 100 mL of adjusted yeast suspension was exposed to 900 mL of RPMI medium containing peptide, yielding a starting inoculum of approximately 10 5 cells/ml. The solutions were incubated at 37uC with agitation. At the predetermined time interval of 0, 1, 2, 4 and 6 hours, 10 mL or 30 mL of the samples were removed, serially diluted with  phosphate-buffered saline (PBS) and plated on SDA plates to allow colony counts after 24 hours of incubation. Controls for yeast growth and antifungal agents were also performed. The lower limit of accurate and reproducible quantitation was 100 CFU/ml. Results were obtained from three independent experiments.
Biofilm reduction assay. Biofilms of C. albicans were formed on commercially available pre-sterilized, polystyrene, flat bottomed 96 well microtitre plates as described by Jin et al with modification 43 . Firstly, a standardized yeast suspension (10 7 cells/ml) of C. albicans SC5314 was prepared by suspending colonies from 24-hours  www.nature.com/scientificreports old culture in RPMI 1640 medium and adjusted to an optical density of 0.38-0.39 at 520 nm. 100 mL of yeast suspension was dispensed into each well of a microtitre plate using a multichannel pipette and the plates were incubated in a shaking incubator at 37uC with 75 rpm for 90 minutes to allow adherence of yeast on the surface of each well. After the adhesion phase, the non-adherent cells were removed and each well was washed twice with 150 mL PBS. 100 mL of RPMI 1640 medium was transferred to each washed well and the plates were incubated at 37uC in a shaking incubator at 75 rpm for 24 hours to allow biofilm formation. Following biofilm phase, the medium was aspirated and each well was washed twice gently with 200 mL PBS to remove non-adherent cells. Residual PBS was removed by inverting the plates over an absorbent paper before addition of peptides. 200 mL of each peptide with concentrations ranging from 1 to 4 times of the MIC determined previously was added to respective wells and the plates were incubated as described above for 24 h. Antifungal agent-free wells and biofilm-free wells were included as positive and negative controls. After treatment with peptides, the medium was removed and each well was washed twice with 200 mL PBS. The biofilm formation was quantified by using XTT reduction assay as described below. Each experiment was repeated three times in duplicate. XTT reduction assay was performed according to the method adapted from Jin et al. 44 Briefly, XTT (Sigma) was dissolved to a concentration of 1 mg/ml using sterile PBS. XTT solution was then filter-sterilized using a 0.22 mm-pore-size filter and kept at 270uC prior to use. 0.4 mM menadione solution was prepared by dissolving menadione (Sigma) with acetone. The menadione solution was filter-sterilized and kept at 270uC before use. Prior to each assay, XTT solution was thawed and mixed with menadione solution at a volume ratio of 2051. After washing, a total volume of 200 mL of XTT-menadione mixture (158 mL of PBS, 40 mL of XTT and 2 mL of menadione) was added to each prewashed biofilm and control wells. The plates were incubated in the dark for 2 hours at 37uC. 100 mL of the solution was transferred to new wells and the colorimetric change in the solution was measured using a microtiter plate reader at 490 nm. Antibiofilm activity of the peptides and conventional antifungals was expressed as the biofilm-eradicating concentration 2 (BEC-2), which is defined as the minimum concentration of the peptides resulting in 50% reduction of the biofilm viability compared to the growth control.
Chequerboard antifungal analysis. Peptide-peptide and peptide-antibiotic interactions were assessed in sterile U-bottomed 96-well polystyrene microplates using a chequerboard method with minor modification. 50 mL of eight serial twofold dilutions of peptide and ten serial twofold dilutions of antifungal agent with beginning concentration at 4 3 MIC were prepared in each well. A fixed concentration of 0.25 3 MIC of the peptide was added in a volume of 50 mL into each column. Each well of the plates was inoculated with 5.0 3 10 2 to 2.5 3 10 3 cells/ml of the yeast suspension. The plates were incubated at 37uC for 48 hours. The fractional inhibitory concentration (FIC), defined as MIC of the drug used in combination divided by the MIC of drug tested alone, was determined 45  The FICI # 0.5 was interpreted as synergism; while FICI . 0.5-4 was an indication of no interaction; and FICI . 4 was interpreted as antagonism.
Haemolytic assay. Freshly collected human blood with heparin was centrifuged to remove the buffy coat. The erythrocytes were washed three times with PBS, centrifuged for 10 min at 2000 rpm at 10uC, and resuspended in PBS to 4% (vol/vol). Peptides were serially diluted in PBS to a concentration ranging from 256 mg/L to 0.5 mg/L (100 mL) in U-bottomed 96-well polystyrene microplates and 100 mL of the erythrocyte suspension was added to each well. PBS and 0.1% Triton X-100 were used as agents to induce 0 and 100% haemolysis, respectively. Plates were incubated for 1 hour at 37uC and centrifuged at 2000 rpm for 10 min. The supernatant (100 mL) was transferred to a 96-well flat-bottomed polystyrene plate, and the release of haemoglobin was monitored by measuring the absorbance at 560 nm using a microplate reader. Results were repeated three times and each time in duplicate. The percentage of haemolysis was calculated using the following formula: where A p and A 100 represent the absorbance of erythrocytes treated with peptide and 0.1% Triton X-100. A 0 is the absorbance of erythrocytes without treatment.  www.nature.com/scientificreports concentration required for 50% reduction in cell viability. Results were recorded by averaging three repeated experiments and each experiment was performed in duplicate.

Molecular docking study
The Protein Data Bank (PDB) files of Sap1 (2QZW), Sap5 (2QZX) and exo-beta-(1, 3)-glucanase (1CZ1) with respective resolution of 2.05 Å , 2.50 Å , and 1.85 Å were obtained from the RSCB protein data bank. Tertiary structure of the designed and template peptides was predicted using peptide tertiary structure prediction server (Pepstr, http://www.imtech.res.in/raghava/pepstr/) 46 . All selected target proteins and peptides underwent minimization by applying the CHARMm forcefield with Memony Rone partial charge using Discovery Studio Client 2.5, San Diego: Accelrys Software Inc 47 . The docking procedure was conducted by Autodock Vina with rigid docking 48 . Complex structures with strong binding affinity were further minimized and analyzed. The interaction energy of the complex within 3 Å was explored by interaction energy protocol of Discovery Studio Client 2.5, San Diego: Accelrys Software Inc.