Bacterial strains and growth conditions

The biofilm-forming bacteria, including S. epidermidis ATCC 35984, Staphylococcus aureus ATCC 25923, Bacillus cereus ATCC 11778, Listeria monocytogenes 10403s, Pseudomonas aeruginosa ATCC 27853, Salmonella Typhimurium ATCC 13311, Escherichia coli ATCC 25922 and Acinetobacter baumanii MT strain, were used. All bacterial strains were grown on tryptic soy agar (TSA; BD Bacto™, USA) except E. coli and L. monocytogenes which were cultured on Luria agar (LA) and TSA supplemented with 0.6% Yeast Extract (TSAYE), respectively, at 37°C overnight. Subsequently, isolated colonies were inoculated and cultured in 5 mL of tryptic soy broth (TSB; BD Bacto™, USA), Luria Broth (LB), or TSB supplemented with 0.6% Yeast Extract (TSBYE) at 37°C for 16–18 h.

Peptide design and in silico characteristics

Using a lesson learned from peptides designed by a template-assisted approach, a novel antibiofilm peptide, BiF, was developed from the alignment of conserved sequences of 18 α-helical antibiofilm peptides obtained from APD3 (Antimicrobial Peptide Database, retrieved on 7 August 2019). To further enhance its antibiofilm activity, a lipid binding motif (or KILRR motif [17]) was hybridized at the C-terminus of BiF and transformed into hybrid BiF2 peptide. In accord with the results of helical wheel projections and 3D structures, BiF2 was subsequently modified by substituting in one and two hydrophobic amino acids at positions 5 and 7 with lysines (L5K, L7K), positively charged amino acid, to obtain two hybrid derivatives, BiF2_5K and BiF2_5K7K, respectively. Peptide sequences were in silico analysed by APD3: Antimicrobial Peptide Calculator and Predictor (https://aps.unmc.edu/prediction). Both hydrophobicity (<H>) and hydrophobic moment (<μH>), as well as helical wheel projection, were determined using the HeliQuest web server (https://heliquest.ipmc.cnrs.fr/). Moreover, I-TASSER (https://zhanglab.dcmb.med.umich.edu/I-TASSER/) was employed to predict the 3D structure of these peptides.

Peptide synthesis and preparation

The peptides were generated using solid-phase peptide synthesis with 9-fluorenylmethoxycarbonyl (Fmoc) as trifluoroacetate salts and amidation at the C-terminus (China Peptides, China). Reverse phase high-performance liquid chromatography (RP-HPLC; >98% purity) and mass spectrometry were employed to confirm the identity and purity of the peptide sequences, respectively. The peptides were dissolved in sterile deionized water and kept at -20°C until used.

Biofilm formation capacity of S. epidermidis ATCC 35984

The capability of S. epidermidis ATCC 35984 to form biofilm was evaluated by crystal violet (CV) staining as described previously [18]. Briefly, bacterial suspensions derived from an overnight culture were harvested and centrifuged at 2,500 × g for 5 min to collect the pellet. Then, TSB was used to resuspended and diluted the pellet to achieve a concentration of 106 CFU/mL at optimal density (OD) of 620 at 0.05. These diluted suspensions were subsequently dispensed into 96-well plates and statically incubated for 24 h at 37°C. Following the incubation period, both the culture media and planktonic cells were discarded, and the biofilm in each well was washed thrice with 0.85% NaCl. Subsequently, the biofilms were fixed using 95% ethanol and stained with a 0.1% (v/v) solution of CV solution (Sigma-Aldrich, USA) for 15 min. Excess dye was removed, and each well was washed thrice with 0.85% NaCl before resuspended in 95% ethanol to dissolve the bound dye. Absorbance was determined at 570 nm using a Multiskan SkyHigh microplate reader (Thermo Scientific™, Singapore). Wells containing only TSB medium without bacteria served as the negative control. All experiments were conducted with six technical replicates and repeated independently three times. The biofilm-forming ability of the bacteria was categorized into four groups based on the OD570 value: no biofilm production (OD570≤ ODc), weak biofilm production (ODc<OD570≤ 2 × ODc), moderate biofilm production (2 × ODc<OD570≤4 × ODc), and strong biofilm production (4 × ODc<OD570). The mean OD ± three standard deviations (SD) of the OD value of the negative control was defined as cut-off OD (ODc).

Biofilm-formation inhibitory activity

The biofilm inhibitory activity of peptide derivatives was assessed by modified crystal violet (CV) staining [19]. Briefly, the pellet of S. epidermidis (ATCC 35984) was collected through centrifugation of bacterial suspensions grown overnight at 2,500 × g for 5 min. Subsequently, the pellet was resuspended and diluted with TSB to a final concentration of 106 CFU/mL. Then, 50 µL of these diluted suspensions was transferred to 96-well microplates containing same volume of peptide derivatives in 10 mM phosphate buffer saline (PBS; Caisson Labs, USA), with concentrations ranging from 0.98 to 250 µg/mL. Vancomycin was used as positive control while TSB medium without bacteria was used as negative control. After 24 h of incubation and reaching static growth at 37°C, culture media as well as planktonic cells were removed, then the biofilm in each well was washed thrice with 0.85% NaCl. For CV staining, biofilms were fixed with 95% ethanol and stained with a 0.1% (v/v) CV solution for 15 min. After removal of excess dye, each well was washed thrice with 0.85% NaCl and resuspended in 95% ethanol to dissolve bound dye. Absorbance was determined at 570 nm using a Multiskan SkyHigh microplate reader (Thermo Scientific™, Singapore). All experiments were done in six technical replicates and repeated independently three times. Minimum biofilm inhibitory concentrations at 50% and 90% (MBIC50 and MBIC90) refers to concentrations of the peptide at which the inhibition of biofilm was at least 50% and 90%, respectively [19, 20].

Antimicrobial activity

The minimal inhibitory concentrations (MICs) of peptide derivatives versus planktonic bacteria were assessed using a modification of the broth microdilution assay previously described in the National Committee for Clinical Laboratory Standards (NCCLS) [21]. In brief, bacteria at mid-logarithmic growth were collected by centrifugation at 2,500 × g for 5 min. Subsequently, the pellet was resuspended and adjusted in Mueller-Hinton Broth (MHB; BD Bacto™, USA) to an OD620 of 0.05. Fifty µL of bacterial suspensions was transferred to 96-well microplates. Later, 50 µL of peptide derivatives at 0.98–250 µg/mL in 10 mM PBS, was added to wells and incubated for 24 h at 37°C with continuous shaking at 220 rpm. After incubation, the minimum concentration of peptide that resulted in no bacterial growth observed by visible inspection after 24 h of incubation was defined as the MIC [19]. The experiments were done in three technical replicates and repeated independently.

To determine minimal bactericidal concentrations (MBCs) of the peptides, 50 µL of supernatant from non-turbid wells was plated on TSA, and the number of colony-forming units (CFUs) was observed after 24 h of incubation at 37°C. MBC was defined as the minimum concentration of peptide at which no bacterial colony was observed on the plate [19].

Hemolysis activity

The effect of peptides on human red blood cells (hRBCs) was assessed by observing the release of hemoglobin upon treated with peptide [16]. Briefly, the hRBCs were obtained from healthy volunteers and subsequently diluted in 10 mM PBS to a concentration of 2% hRBCs. An equal volume of 2% hRBCs and peptide at concentration of 0.98–250 µg/mL in 10 mM PBS mixtures were incubated for 1 h at 37°C. Following incubation, the suspension was centrifuged, then supernatants were collected and transferred to 96-well microplates for measurement of absorbance at OD of 405 nm (OD405). This study was performed in accordance with the Thammasat University Ethics Committee (COA No. 066/2562). Written informed consent was obtained from all volunteers.

Biofilm cell viability

The biofilm cell viability after treated with hybrid peptide BiF2_5K7K was examined by 3-(4,5-Dimethylthiazol-2-yl)-2,5-Diphenyltetrazolium bromide or MTT colorimetric assays [16]. Bacterial suspensions (106 CFU/mL in TSB) and peptide at concentration ranging from 0.98 to 250 µg/mL in 10 mM PBS were incubated in 96-well microtiter plates at 37°C for 24 h. After incubation, supernatant was discarded, followed by the addition of 0.4 mg/mL MTT (Sigma-Aldrich, USA) in 10 mM PBS to each well and incubated for 3 h at 37°C. Subsequently, supernatant was removed and dimethyl sulfoxide (DMSO; Fisher BioReagents™, UK) was added to dissolve the formed formazan crystals in wells. At a wavelength of 570 nm, the optical densities of the solutions were then determined by microplate reader.

Biofilm eradication activity

The eradication capacity of BiF2_5K7K against preformed biofilm was examined as previously described with some modifications [22, 23]. In brief, suspensions of S. epidermidis with a final density of 106 CFU/mL were added to 96-well microplates, then covered with 96-peg lids (Nunc™ Immuno TSP Lids, Denmark). Following 24 h of incubation at 37°C, the peg lids were washed with 0.85% NaCl to remove culture media and non-adherent cells. Afterwards, the biofilms adherent to peg lids were incubated at 37°C for 24 h with different concentrations of peptide derivatives. The peg lids were subsequently washed and placed on new plates containing PBS. The biofilms on the lid were removed using a water bath sonicator (Bandelin SONOREX™, Germany). Biofilms were subsequently diluted and then placed on TSA. Minimum biofilm eradication concentrations at 50% and 90% (MBEC50 and MBEC90) refer to the concentration of the peptide at which established biofilm was eradicated by at least 50% and 90%, respectively [19, 20].

Serum stability

The stability of BiF2_5K7K in the presence of human serum was assessed by evaluating the MIC values, as previously described [24]. Human serum was subjected to heat inactivation at 56°C for 30 min. Bacterial suspensions were then adjusted to a concentration of 106 CFU/mL in MHB supplemented with 25% and 50% human serum. These suspensions were mixed with two-fold serially diluted concentrations of BiF2_5K7K (ranging from 0.98 to 250 µg/mL) and incubated at 37°C for 24 h. MICs were subsequently determined as described above. The study protocol was ethically approved by the Thammasat University Ethics Committee (COA No. 066/2562). All volunteers provided written informed consent.

In vivo toxicity assay

The toxicity of peptide BiF2_5K7K against human lung fibroblast MRC-5 cell lines (ATCC CCL-171) was investigated by MTT colorimetric assay [16]. Briefly, MRC-5 cells were cultured in Minimum Essential Medium [MEM; supplemented with 10% heat-inactivated FBS and 1% Penicillin-Streptomycin] in 5% CO2 at 37°C. Following an overnight incubation, MRC-5 cells were diluted to 10,000 cells/well and placed in 96-well microplates. Subsequently, they were exposed to BiF2_5K7K at 0.98 to 250 µg/mL. After incubation for 24 h, MTT solution (0.4 mg/mL) was added to each well and incubated for an additional 3 h under 5% CO2 at 37°C. The supernatant was then removed and 100 µL DMSO added, and absorbance was measured at 570 nm using a microplate reader.

Circular dichroism spectroscopy analysis

The conformational change of BiF2_5K7K in a membrane mimicking environment was determined by measuring CD spectra of the peptide at wavelengths 190 to 250 nm with a scanning speed of 10 nm/min by a Jasco-815 spectropolarimeter (Jasco, Japan). As previously described [25], the CD spectra were obtained using a quartz cell with a 1 mm path length at a temperature of 25°C. At a final concentration of 0.2 mg/mL, BiF2_5K7K was dissolved in three distinct membrane-mimicking environments, 10 mM PBS, 30 mM sodium dodecyl sulfate (SDS; Sigma-Aldrich, Japan), and 50% (v/v) 2,2,2-trifluoroethanol (TFE; Merck, USA). PBS served as the aqueous surrounding, while the SDS micelles mimic the negatively charged environment of bacterial membranes featuring an external surface with negative charges and an internal hydrophobic environment resembling phospholipid chains. TFE was utilized to simulate the hydrophobic compartment of microbial membranes as the fluorinated environment exhibits limited interaction with the molecule and induce intra-molecular interactions that frequently result in induced structuring, such as the formation of an alpha-helix structure [26]. An average of three scans were performed for each condition.

Time-killing analysis

The time-dependent bactericidal activity of BiF2_5K7K against planktonic S. epidermidis was determined as previously described [23]. In brief, mid-log phase bacteria in MHB were exposed to BiF2_5K7K at MIC and/or MBC for 24 h at 37°C. The bacterial suspensions were obtained at specific time point (0, 2, 4, 6, 8 and 24 h), and 10-fold serially diluted with PBS and then plated on TSA. Untreated bacterial suspension was served as a control. After incubation, the number of CFUs was determined by colony count.

Confocal laser scanning microscopy

An Amsterdam active attachment (AAA) model was used to assess the effectiveness of BiF2_5K7K to inhibit biofilm formation as previously described [22]. In a 24-well microplate, various concentrations of BiF2_5K7K were incubated overnight with 107 CFU/mL of S. epidermidis. The plate was then covered with a sanitary stainless-steel lid containing double-glass coverslips. Following 24 h of incubation at 37°C, the biofilms were washed, and then 50 mM green Alexa-ConA (Sigma-Aldrich, USA) was applied for 30 min to stain the biofilms. After staining, biofilms were fixed for 3 h with 2.5% glutaraldehyde. Biofilm images were acquired using a confocal laser scanning microscope (CLSM; Carl-Zeiss/LSM800, Zeiss, Germany). The fluorescence intensity of each test was analysed using ZEN 2.1 (blue edition) software.

The effect of BiF2_5K7K against mature biofilm was also investigated. In short, overnight cultures of S. epidermidis were diluted to 107 CFU/mL with TSB and further incubated at 37°C for 24 h to form biofilms. Next, wells were rinsed with sterile water and mature biofilms was treated with BiF2_5K7K at concentration of MBEC50 and further incubated for 24 h. Subsequently, biofilms were washed again, then stained with green Alexa-ConA and LIVE/DEADTM BacLight™ Bacterial Viability Kit (Invitrogen™, Thermo Fisher, USA) for 15 min. After staining, 2.5% glutaraldehyde was used to fixed biofilms, then observed under CLSM as described above.

Transcriptional analysis

The expression level of S. epidermidis biofilm formation gene was investigated using a real-time reverse transcription-polymerase chain reaction [27, 28]. In brief, total RNA of S. epidermidis biofilm, both untreated and treated with BiF2_5K7K (at MBIC50 for 24 h), was extracted by a RNeasy PowerBiofilm Kit (Qiagen, USA) and immediately reverse transcribed to cDNA by iScript reserve transcription supermix (Biolabs, UK). RNAprotect Bacteria Reagent (Qiagen, USA) was used to preserve RNA during isolation. Remaining DNA was removed by treatment with DnaseI (Biolabs, UK). Real-time RT-PCR was done in duplicate using iTaq Universal SYBR Green Supermix (Bio-Rad, USA) with 5 ng of cDNA and 10 mM of forward and reverse primers [29, 30] (Table S1) using a CFX96 Real-Time PCR Detection System (Bio-Rad, USA). The cycling condition was started with an initial denaturation at 95°C for 2 min, followed by 50 amplification cycles consisting of 95°C for 30 s, 56°C for 1 min, and 72°C for 1 min. Subsequently, a melting curve was generated from 65°C to 95°C at intervals of 1 s at the end of each run. To analyse the relative quantitation of gene expression, 2− ΔΔCt method was used, and 16S ribosomal subunit (rRNA) was served as the reference or housekeeping gene.

Inhibitory activity of BiF2_5K7K on segments of silicone catheters

The capacity of BiF2_5K7K to prevent S. epidermidis biofilm formation on a silicone catheter was determined using CV staining and colony count as previously described [31, 32] with some modifications (Fig. 1). CV staining was used to evaluate the development of biofilm on BiF2_5K7K-coated catheters and the surrounding area, while the viability of bacteria and the number of planktonic cells adherent to the catheter was assessed by colony count. Briefly, a sterile silicone catheter (Norta®, Malaysia) was sectioned into 50 mm segments which were then coated with BiF2_5K7K at concentrations of 250, 500, 1,000 or 2,000 mg/mL for 1 h by submerging the catheter segments, followed by overnight air-drying. The pre-coated catheters were subsequently placed in a 24-well microplate containing S. epidermidis suspension (\(\sim\)107 CFU/mL in TSB) and incubated at 37°C. Following 24 h of biofilm formation, supernatant was collected for the colony count assay, and the wells were rinsed thrice with 0.85% NaCl.

To quantitate biofilm prevention, catheter segments were stained for 15 min with a 0.1% CV solution, then rinsed with 0.85% NaCl and resolved in 95% ethanol. Biofilm formation on catheter segments was measured by a microplate reader at absorbance of 570 nm. Wells were also stained with the CV solution, as previously described.

To evaluate the viability of sessile and planktonic S. epidermidis bacterial cells were removed from catheter segments by vortexing for 10 s, sonicating for 10 min, and vortexing again for 10 s. Cell suspensions were diluted and subsequently plated on TSA. The planktonic cells in wells were evaluated by ten-fold serially diluting collected supernatants and then plated on agar. After incubations, the CFUs were counted.

Fig. 1

Workflow of peptide coating on catheters

Statistical analysis

The experiment was performed at least three times. The results were displayed as the mean ± standard deviation (SD). One-way ANOVA or Student’s t-test were used to analyse the data via GraphPad Prism (version 9.0). *p-value < 0.05, **p-value < 0.01, ***p-value < 0.001, and ****p-value < 0.0001 were considered significant.