Apple extract (AE) preparation
Apple fruits (Malus domestica cv. Anna) were purchased in a ripe state (Fig. 1) from local producers located in San Marcos, Tarrazú, San José, Costa Rica (permit R-CM-ITCR-002–2021-OT). Whole fruits (without seeds) were rinsed in water, sliced, freeze-dried (Bench Top FDB-8602, OPERON), grinded (1 mm, MF-10 basic IKA, WERKE), and preserved at − 20 °C until extraction. The powdered material was extracted (1:10, plant material to solvent) by three consecutive water–ethanol macerations at room temperature, followed by ethanol elimination in a vacuum evaporator (40 °C, R-300, BUCHI). The identity of all fruits collected was authenticated by Botanist M.Sc. Elizabeth Arnáez-Serrano, Biology Department, Costa Rica Institute of Technology and were previously confirmed with the support of the Costa Rican National Herbarium [19].
Phytochemical analysis
Preliminary phytochemical screening was evaluated qualitatively by colorimetric, foaming, and precipitation reactions [21, 22], including detection of flavonoids (Shinoda test), steroids and terpenoids (Liebermann-Buchard test), alkaloids (Dragendorff test), and saponins (foam test).
High-performance Thin Layer Chromatography (HPTLC) was performed using the freeze-dried AE powder dissolved in methanol (1 mg mL−1). The method was developed using a CAMAG HPTLC system (CAMAG, Muttenz, Switzerland). Sample application (10 µL) was done on a 20 × 10 cm HPTLC glass plate coated with Silica gel F254 (Merck, Darmstadt, Germany). Samples were applied to the HPTLC plate with the aid of the CAMAG Automatic Sample Applicator 4 (ATS4) as 8 mm bands. The first application position (x axis) was 20 mm and the sample bands were applied at a distance of 8 mm from the base of the HPTLC plate (application position y). The distance between tracks was 11 mm. Chromatography development was done in a CAMAG Automatic Developing Chamber 2 (ADC 2) with the following settings: 30 s pre-drying, 20 min saturation with filter paper, humidity control with MgCl2 (10 min, 33% relative humidity), 70 mm migration distance (solvent front position), 5 min drying time, 10 ml of mobile phase for HPTLC development and 25 mL of mobile phase for HPTLC chamber saturation. The mobile phase was comprised of ethyl acetate: water: formic acid (100: 30: 20). The chromatograms were exposed to UV light (254 nm and 366 nm) and white light. The derivatization of the chromatograms with the natural product reagent (2-aminoethyl diphenyl borinate), anisaldehyde reagent, phosphomolybdic acid, fast blue salt B and the Dragendorff reagent were performed with the CAMAG Derivatizer. If the derivatization process required heating, The CAMAG TLC Plate Heater III was used. All reagents were purchased from SIGMA; prepared and applied according to the CAMAG Derivatizer instructions. Pre- and post-derivatization, images were taken, processed, and documented with the CAMAG TLC Visualizer. All HPTLC steps were programmed and monitored with the CAMAG VisionCATS software (version 2.5). Image comparison views were created with this software as well.
Total phenolic content (TPC) in the apple extract (AE) was determined using a modified Folin-Ciocalteu spectrophotometric method as described by Rojas-Garbanzo et al. [23]. Briefly, AE were incubated with Folin-Ciocalteu reagent and sodium carbonate solution (75 g/L) at 50 °C for 15 min. The absorbance was measured at λ = 620 nm (FLUOstar OPTIMA, BMG LABTECH). TPC was calculated by comparing against an external calibration curve of gallic acid (10–80 mg GAE L−1, r2 = 0.9909). Results were expressed as µg of gallic acid equivalents (GAE) per gram of AE. TPC are shown as the mean ± SD (n = 3).
Cell culture
The cell lines used in this study were NIH-3T3 (ATCC CRL-1658™, murine skin fibroblasts, passages 51 to 54), NCI-H460 (ATCC HTB-177™, human lung carcinoma, passages 46 to 50), and MCF7 (ATCC HTB-22™, human breast adenocarcinoma, passages 18 to 21). NIH-3T3 were cultured in DMEM (4.5 g L−1, GIBCO) and NCI-H460 cells were cultured in RPMI (GIBCO). Both DMEM and RPMI media were supplemented with 10% fetal bovine serum (FBS, SIGMA), 2% L-glutamine (4 mM, GIBCO), 1% sodium pyruvate (0.11 mg mL−1; SIGMA) and 1% antibiotics (1 × 104 IU mL−1 penicillin and 1 × 104 µg mL−1 streptomycin; GIBCO). The same supplemented DMEM was used for culturing MCF7 cells, with additional 10 µg mL−1 insulin (SIGMA). All cells were maintained at 95% humidity and 5% CO2 at 37 °C.
Cytotoxicity by MTT assay
Cells lines NIH-3T3, NCI-H460, and MCF7 were seeded onto 96-well plates (1 × 105 cells cm−2) and treated for 24 h with increasing concentrations of the apple extract (AE) diluted in the respective culture medium. Cell viability was measured by the MTT (3-(4,5-dimethylthiazolyl-2)-2,5-diphenyltetrazolium bromide) assay as described by Calvo-Castro et al. [24]. The formazan crystals were dissolved in ethanol 95%, and the absorbance was measured at 570 nm (FLUOstar OPTIMA, BMG LABTECH). Cell viability was normalized as relative percentages in comparison to untreated controls. The half-maximal inhibitory concentration (IC50) for each extract in each cell line was determined from dose–response linear dispersion curves. Data is shown as the mean ± SD (n = 3). Normality was tested with the Kolmogorov–Smirnov test and D’Agostino-Pearson test. Differences between AE treatments in the three cell lines were analyzed by two-way ANOVA followed by a Bonferroni post-test. All statistical analyses were performed using the software package Graph-Pad Prism (version 8.01; GraphPad Software, USA).
Scratch wound healing assay
NIH-3T3 fibroblasts were seeded on 24-well plates (1 × 105 cells cm−2). Confluent monolayers were scraped using a standard 1 mL micropipette tip, leaving an empty and homogeneous cross-shaped area [25], followed by treatment for 72 h with non-cytotoxic AE concentrations (38 and 58 µg GAE mL−1). The lesion was photographed by light microscopy every 24 h, and percentage wound area over time was calculated using Image J (https://www.imagej.nih.gov/ij/). Data is shown as the mean ± SD (n = 3).