Rink amide resin, 9-fluorenylmethoxycarbonyl (Fmoc) amino acids, 6-HCl 1-hydroxybenzotriazole (HOBt), and N,N-dicyclohexylcarbodiimide (DCC) were purchased from AAPPTec (Louisville, KY, USA). Methanol, diethyl ether, N,N-dimethylformamide (DMF), absolute ethanol, dichloromethane (DCM), acetonitrile (ACN), isopropyl alcohol (IPA), trifluoroacetic acid (TFA), ethane-1,2-dithiol (EDT), and triisopropylsilane (TIS) were obtained from Merck KGaA (Darmstadt, Germany). 4-methylpiperidine, pyridine, Triton-X, potassium cyanide (KCN), phenol, ninhydrin, biotin-NHS, Supelclean LC-18 SPE columns, α-cyano-4-hydroxycinnamic acid, ethylenediaminetetraacetic acid (EDTA), Trypan blue, hydrocortisone, doxorubicin hydrochloride, resazurin sodium salt, 7,12-dimethylbenz[α]anthracene (DMBA), dexamethasone sodium phosphate and trypsin were obtained from Millipore-Sigma (St. Louis, MO, USA).

All bLFcin-derived peptides (Table 1) were manually synthesized by solid-phase peptide synthesis (SPPS) using a previously described Fmoc/tert-butyl (tBu) protection strategy [36]. Briefly, (a) Rink amide resin (0.46 meq/g) was swollen in DMF for 15 min with constant stirring and washed 3 times with DCM. (b) Fmoc removal reactions were carried out using 25% 4-methylpiperidine (Triton-X 1%) in DMF for 10 min at room temperature; this treatment was repeated once, and the resin was then washed 5 times with DMF and 3 times with DCM. (c) We performed the Kaiser test to verify free amine groups on the resin during subsequent Fmoc deprotection steps along the synthesis. (d) For the coupling reaction, 5 eq of each Fmoc-protected amino acid, 5 eq of DCC, and 5 eq of HOBt were dissolved in DMF and added to the resin with constant stirring for a minimum of 3 h. The coupling reaction was monitored with the Kaiser test until a negative result was obtained. (e) Cleavage of the full amino acid sequence and the deprotection of the sidechain protecting groups were carried out by adding a cleavage cocktail solution of TFA/water/TIS/EDT (92.5/2.5/2.5/2.5, v/v) for 3 h with constant stirring at room temperature (20–23°C). The solution from the resin and cleavage cocktail mixture was recovered by filtration, and the peptide was precipitated using 2 volumes of ice-cold diethyl ether. The dried pellet was resuspended in water for lyophilization; 500 µL of the solution was reserved for high-performance liquid chromatography (HPLC) and characterization. (f) Crude peptides were purified by reversed-phase solid-phase extraction using Supelclean C18 silica columns (5 g LC-18 SPE). After activating the column according to the manufacturer’s instructions, the crude peptide (50 mg/mL) was loaded onto the column, and a stepwise gradient from 0% to 90% ACN in water was applied to purify the peptide. The fractions were pooled based on their HPLC-analyzed purity (> 90%), and lyophilized. Peptide purity was determined by HPLC by injecting 10 µL of peptide solution (1 mg/mL) and eluting using a linear 2 mL/min, 10 min, 5% to 70% gradient of ACN 0.05% TFA in water 0.05% TFA, in an Agilent 1200 LC system equipped with a Chromolith C18 column (50 mm × 4.6 mm) and a UV detector set at 210 nm.

Molecular weights were confirmed by matrix-assisted laser desorption/ionization–time of flight (MALDI–TOF): MALDI–TOF mass spectrometry using an Ultraflex III TOF/TOF mass spectrometer (Bruker Daltonics, Bremen, Germany) operated in reflector mode, with an MTP 384 polished steel target plate (Bruker Daltonics) and α-cyano-4-hydroxycinnamic acid as the matrix. Spectra were acquired using 500 laser shots at 25–30% laser power.

The synthesis of the dimeric branched peptide bLFcin(20-25)₂ involved the double elongation of the L-lysine chain at both the N-alpha and N-epsilon termini. The tetrameric branched peptide bLFcin(20-25)₄ was produced through dimerization, which utilized oxidative disulfide bridging of the C-terminal Cys monomer in a solution of phosphate buffer (pH 8.0) and 10% DMSO (see Figure 1). HPLC was used to monitor the conversion of the dimer to the tetramer until the reaction was complete.

Peptides (purity > 90%) were dissolved in sterile phosphate buffered saline (PBS) to achieve final concentrations ranging from 100 to 6.25 μM (for most peptides) or from 50 to 3.125 μM (for the tetrameric peptide). For in vivo assays, the dimeric peptide was dissolved to 100 μM in sterile saline.

The human OSCC cell lines FaDu (ATCC^® HTB-43^™) and SCC9 (ATCC^® CRL-1629^™), and the human keratinocyte line HEK001 (ATCC^® CRL-2404^™) were purchased from the American Type Culture Collection (ATCC-Manassas, VA). Cell culture was performed strictly following the ATCC recommendations, and only one cell line was manipulated at a time to avoid cross-contamination. FaDu cells were cultured in RPMI medium (L0500, Biowest) supplemented with 10% fetal bovine serum (S1810, Biowest). SCC9 cells were cultured in DMEM/F12 medium (L0093, Biowest) supplemented with 400 ng/mL hydrocortisone and 10% FBS. HEK001 keratinocytes were cultured in serum-free KGM BulletKit medium containing the supplements included in Lonza Corporation^®’s kit CC 3107. All cells were maintained in a 5% CO₂ humidified atmosphere at 37°C. The culture media were changed every 2–3 days; cells were passaged twice a week and used between 6 and 10 generations. Trypan blue dye was used routinely to quantify cell viability and density. Mycoplasma infection was assessed when thawing a new vial from each cell master bank using the Mycoplasma Detection Kit MycoStrip^® (Invitrogen).

The cytotoxic activity of all peptides was evaluated using the resazurin assay, which measures cell viability. Cells were plated onto 96-well tissue culture plates (FaDu = 1.2 × 10⁴ cells/well; SCC9 = 1.0 × 10⁴ cells/well; and HEK001 = 1.4 × 10⁴ cells/well) and allowed to attach for 18 h until they reached 70–80% confluence. We then incubated the cells with 100, 50, 25, 12.5, and 6.25 µM bLFcin-derived peptides; the tetrameric peptide bLFcin(20-25)₄ was tested at 50, 25, 12.5, 6.25, and 3.125 µM. Doxorubicin (10 μM) was used as a positive control for cytotoxicity; 100% cell viability was established after the addition of PBS (10 µL) to untreated control cells. Cell micrographs were obtained using phase contrast microscopy (DMi1, Leica-microsystems). After treatment for 2, 6, and 24 h, cells were rinsed with PBS, and 10 μL of resazurin 440 μM with 90 μL of medium was added to reach a final concentration of 44 μM. Metabolically active cells converted resazurin (blue-purple) to resofurin (pink); fluorescence was measured at excitation and emission wavelengths of 535 nm and 595 nm, respectively (TECAN^® Genios). The fluorescence intensity of the treated cells was normalized to establish the percentage of viable cells relative to untreated counterparts. The half-maximal inhibitory concentration (IC₅₀) was calculated by plotting cell viability versus the log of peptide concentration (µM) and analyzed (GraphPad Prism 10 software) using a nonlinear regression (sigmoidal curve) fit.

Cells were treated with biotin-labeled bLFcin(20-25)₂ to evaluate its ability to permeate through the cell membrane. FaDu and HEK001 cell suspensions (2 × 10⁵ cells/mL) were treated with 100 μM biotin-labeled bLFcin(20-25)₂ for 2 h. Negative control cells were treated with PBS, and peptide control cells were treated with 100 μM biotin-bLFcin(20-25), the linear 6-residue peptide. Each cell suspension (100 μL) was seeded onto 96-well, rounded-bottom plates. The plates were then centrifuged (700 g × 3 min), the cells rinsed with PBS, and stained with 1:1,600 phycoerythrin (PE)-streptavidin conjugate (189737, Sigma-Aldrich) in PBS. After incubation for 30 min at 4°C in the dark, excess PE-streptavidin was removed by centrifugation, and the plates were washed with PBS. Cellular PE fluorescence was measured using a flow cytometer (BD FACSCanto II). We acquired a total of 10,000 events per sample. The abundance distribution of PE-streptavidin-stained cells was determined using the FACSDiva v7.0 software.

bLFcin(20-25)₂-treated and untreated FaDu cells were fixed on glass coverslips using 3% glutaraldehyde in PBS (pH 7.0) for 2 h, rinsed with deionized and 0.22 μm-filtered water, and fixed in 0.5% osmium tetroxide for 45 min. We then performed 3 additional 10-min washes with filtered, deionized water and dehydrated the samples using increasing ethanol concentrations: 2× 70%, 2× 95%, and 3× 100%. Coverslips were maintained in 100% ethanol until dry, then flushed with CO₂ at critical-point conditions using a SAMDRI^®-795 instrument and subsequently covered with gold nanoparticles in a Denton Vacuum Desk IV thin-film deposition system. SEM images were acquired using a JEOL JSM 6490-LV microscope.

Six-week-old male Syrian golden hamsters (Mesocricetus auratus) were procured from the National Institute of Health of Colombia (Bogotá, Colombia). We housed 2 animals per cage in individually ventilated filter-capped polypropylene cages (RAIR OneCage^® system) at a controlled temperature (20–24°C), with a 12-h dark/light cycle; drinking water and a dry food (LabDiet 5010) were provided ad libitum. All animal experiments complied with the ARRIVE guidelines and the Guide for the Care and Use of Laboratory Animals. They were carried out following the EU directive 2010/63 for the protection of animals used for scientific purposes, and were approved by the Ethics Committee of the Sciences Faculty of the National University of Colombia (approved main project identification: RC# 678 of 2014).

The hamsters’ right pouches were painted with a 0.5% DMBA oil solution 3 times a week for 12–16 weeks. Immunosuppression was achieved by administering daily subcutaneous dexamethasone (1 mg/kg) for 7 days, and repeated every 3 weeks during the carcinoma induction period [35]. Peptides were administered intratumorally once macroscopic lesions were evident. Animals that developed oral carcinoma were divided into 2 groups: Group A (control): 5 animals were treated daily with saline (50 μL) for 15 days; and Group B: 5 animals were treated with bLFcin(20-25)₂ (50 μL, 100 μM) daily for 15 days.

Once macroscopic lesions were detected, tumor dimensions (length and width) were measured daily for 15 consecutive days with a digital caliper. The tumor volume was calculated using the formula V = 0.52 × length × width² [37]. After 15 days of treatment, the hamsters were euthanized by intraperitoneal injection of 300 μL Euthanex^® (390 mg/mL sodium pentobarbital and 50 mg/mL sodium diphenylhydantoin), and buccal pouches were procured for standard histological analyses. Tumors were fixed in buffered 3.7% formaldehyde solution, processed for routine H&E staining, and evaluated by light microscopy using an Olympus CX21 microscope.

Cytotoxic assays were conducted in cell lines in triplicate (n = 3). The IC₅₀ values were ascertained via non-linear sigmoidal fit regression and are expressed as the mean ± standard error of the mean (s.e.m.). To assess the in vivo anticancer efficacy of bLFcin(20-25)₂, the tumor volume growth ratio was plotted against the duration of treatment (days). Statistical comparisons between the distinct groups were conducted using One-way ANOVA and Tukey’s test. A p-value of < 0.05 was considered statistically different. All statistical analyses were carried out using GraphPad Prism version 10.

bLFcin(17-31) is a linear peptide with demonstrated antibacterial activity; however, shorter sequences obtained by depleting the amino acids at the carboxyl-terminus and conserving the amino-terminus unaltered, e.g., bLFcin(17-26), still show potent cytotoxic activity [38]. More recently, it has been studied how not only the length of immune peptides influences their bioactivity, but also how their shape and configuration play a critical role. Molecular dynamics simulations revealed that greater antibacterial activity of multiple antigenic peptides, here referred to as ‘branched’, results from stronger adsorption to and interaction with bacterial membranes, causing greater membrane deformation compared to linear peptides [39]. For this reason, peptides derived from bLFcin(17-31) were generated by SPPS using an orthogonal Fmoc/tBu protection strategy, as previously described [36, 40] and detailed in the Materials and methods section. The tetrameric peptide was obtained by oxidative disulfide bonding in solution of two dimeric monomers, and it is illustrated in Figure 1. Purification via solid-phase extraction and subsequent characterization resulted in peptides with purity greater than 90%, and their masses were confirmed by MALDI–TOF analyses. A complete list with the corresponding HPLC chromatograms and m/z mass spectrograms is available in the supplementary material (Table S1). Table 1 presents the sequences of the synthetic bLFcin-derived peptides obtained for this study; please note that all synthetic peptides are C-terminal amides.

We next compared the ability of the peptides listed in the 3 groups shown in Table 1 to induce cytotoxicity in oral cancer cells compared to normal keratinocytes, HEK001. Our results are systematically presented below.

The strong cytotoxic effect of the tetrameric peptide on normal keratinocytes raised concerns about pursuing further experimentation with this candidate, particularly given its reported hemolytic activity (H_max = 49.1% of red blood cells at 100 µM and 37°C) [41]. In contrast, the dimeric peptide exhibits a much lower hemolytic index (H_max = 5.6%) and has been reported to display cytotoxic activity against multiple cancer cell lines, including colon, breast, cervix, and prostate [42], making it a more promising candidate for further evaluation against oral cancer cell lines. Although the palindromic peptide showed higher cytotoxicity toward FaDu cells, it has also been reported to be hemolytic (H_max = 24.8%). Despite a higher IC₅₀ of the tetramer and palindromic peptides, the modest cytotoxicity of bLFcin(20-25)₂ toward FaDu cells prompted its selection as a branched peptide model. The dimeric peptide bLFcin(20-25)₂ comprises 2 copies of the RRWQWR minimal sequence linked to a lysine by its α-amino and ε-amino termini. The cytotoxicity of the two-branched peptide was evaluated microscopically in FaDu cells (Figure 3, left panel); we observed cell membrane lysis, loss of adhesion, and expulsion of intracellular content after a 2-h treatment (100 μM). After 6 h and 24 h, the irreversible damage caused to the integrity of FaDu cells prevented their recovery. In contrast, although HEK001 keratinocytes exhibited morphological changes after 6 h of bLFcin(20-25)₂ exposure (Figure 3, right panel, second row), including apparent detachment and membrane blebbing, they recovered after 24 h. Lastly, SCC9 cells were highly resistant to the dimeric peptide (Figure 3, middle panel). The SCC9 cell line is reported to be significantly more chemo-resistant than similar SCC lines, such as SCC4 or SCC15 [43]. While miRNA expression and its downstream pathways contribute to these resistance mechanisms, evaluating the molecular pathways specifically related to bLFcin-derived peptides is beyond the scope of this study.

We next compared the ability of biotin-conjugated linear bLFcin(20-25) and dimeric bLFcin(20-25)₂ to elicit morphological changes and interact with FaDu and HEK001 cells using flow cytometry (FC) (Figure 4). We exposed these cells to each biotinylated peptide for 2 h, then stained them with a streptavidin-PE (SAPE) conjugate, rinsed them, and analyzed them by FC. Morphological alterations (granularity) were assessed by changes in the side-scatter signal (SSC-A [side scatter (for sample A)] vs. FSC-A [forward scatter (for sample A)], Figure 4). We found that treatment with bLFcin(20-25)₂ (bottom panel) increased SSC-A values, whereas the short linear motif bLFcin(20-25) (middle panel) maintained SSC-A values similar to those of untreated cells (top panel). These observations suggest that the dimeric peptide promotes internal structural changes.

We complemented our morphological analyses with fluorescence-based quantification using the “count vs. PE-A” histogram. After treatment of the cells with the dimeric peptide, the morphological alterations and potential poration of the cell membrane caused an internalization of the biotin-peptide; these cells were not permeabilized prior to the SAPE conjugate tagging, indicating that the fluorescent population had interacted and retained the biotinylated substrate, the peptide, even after the washes subsequent to the fluorescent labeling. Cells that internalized the biotin-peptide were tagged with the SAPE conjugate, enabling detection via FC due to the fluorescence properties of PE. We found that 98.4% of FaDu cells and 96.9% of HEK001 cells retained the biotinylated peptide within their structure (Figure 4, bottom panels). In contrast, untreated cells or those treated with the linear peptide RRWQWR (Figure 4, top and central panels) displayed minimal fluorescence, indicating limited peptide-cell interaction. In summary, our FC analyses confirmed dimeric peptide interaction and demonstrated its ability to induce morphological changes. However, further studies are needed to evaluate the peptide’s intracellular uptake by healthy HEK001 cells and its implication in selective cytotoxicity.

We next investigated the effects of the dimeric peptide bLFcin(20-25)₂ on the cell membrane of carcinogenic FaDu cells using SEM. Following a 2-h treatment, we observed cell membrane alterations, including pore formation, irregularities, and vacuolization (Figure 5). As expected, untreated cells remained completely attached to the slide surface and exhibited an extended, homogeneous, and smooth-appearing cytoplasm.

While our initial studies in FaDu cells established key features of the dimeric peptide activity in a cellular model of human oral cancer, this is a multistage, tissue- and host-dependent process that cannot be fully modeled in vitro; we employed a previously described methodology to induce oral carcinoma in hamsters [35]. The first animals exhibiting tumors (Group A) served as negative controls and received daily, intratumoral injections of 50 µL saline. Animals in Group B were treated with 100 µM bLFcin(20-25)₂ for 15 days (see macroscopic findings in Figure S2). No animals died before the scheduled end of the treatment, and the weight of the animals was registered every week (Figure S3).

Dimer peptide-treated animals developed much smaller tumors compared to the placebo group, as reflected by statistically significant differences in tumor volume (Figure 6). These comparisons were made by assessing volume change ratios, calculated as volume v_n/volume v₀, where v_n was the volume on the day of measurement, and v₀ was the volume on the first day of treatment.

Microscopic analyses of tissue samples from animals euthanized at the end of each treatment showed the histological features of the carcinoma stage with abundant cell and nuclear pleomorphism and loss of stratification, contrary to the continuous and distinguished epithelial cell layers characteristic of healthy tissue. The microscopic imaging of peptide-treated tumors revealed no signs of cancer lesion regression or significant differences in tissue morphology, compared to the mucosa of saline-treated negative controls. However, the images evidenced substantial lymphocytic cell recruitment to tumors in animals treated with the dimeric peptide dissolved in saline solution (Figure 7). Future studies should investigate the inflammatory markers that characterize the host immune response to peptide administration.