1. Introduction
Over the last decade, the concept of food sustainability has received increased attention mainly owing to the growth of the world’s population and the intensification of pressure on the global food system. Thus, the food system has faced up to a novel challenge: to discover new food ingredients with functional properties of bioactive compounds present in natural matrices [
1]. Besides, for the reason that the food processing industry generates an enormous amount of food waste worldwide, food sustainability is also based on the minimization of negative effects of food byproducts on the environment [
2]. Currently, there is a growing interest in the recovery of food waste, and upgrading it into high-value byproducts for use as ingredients in new product development [
3]. To obtain health-promoting phytochemicals such as vitamins, minerals, antioxidants, or amino acids from food byproducts, several studies have already been carried out on citrus peels, grape seeds, mango peels, pepper stalks, or pomegranate marcs [
4,
5,
6,
7]. Furthermore, the main topic of research and innovation funded by the European Union in the Horizon 2020 framework program is developing scientific projects about the valorization of agriculture and food byproducts. The objective of these projects is to achieve extracts, enriched fractions, and isolated compounds, which would be subsequently integrated into food or pharmaceutical product formulations [
8].
Plant-derived foods such as vegetables contain high levels of natural antioxidants and bioactive dietary components, so their byproducts could be a valuable source for new product formulations [
9]. The genus
Brassica (Brassicaceae or Cruciferae family), in general, and broccoli, in particular, are rich sources of nitrogen-sulfur derivatives (glucosinolates and isothiocyanates), and they possess high values of phenolics (chlorogenic and sinapic acid derivatives, and flavonols) and essential nutrients (vitamins and minerals) [
10,
11]. Broccoli (
Brassica oleracea L. var.
italica) is one of the most widely consumed vegetables. A diet rich in broccoli has been reported to play an essential role in the prevention of chronic diseases, such as cancer and cardiovascular diseases [
12]. However, the main edible part of broccoli is the florets, which make up only 30% of the vegetable’s biomass. This means that around 70% of the plant (mainly leaves) is discarded as non-edible parts. These residues are not valuable commercially, require an additional cost of disposal, and have a detrimental impact on the environment [
13]. Moreover, broccoli’s flowering is triggered under the optimal temperatures of 18 to 25 °C. When the temperature exceeds 30 °C, initial floret development is disrupted. Taiwan is located in tropical and subtropical regions, consequently reducing the quality and quantity of broccoli produced during summer [
14]. Hence, it might be essential for the food industry to take advantage of broccoli byproducts such as stems, leaves, roots, and seeds as functional product materials, rather than only using the florets.
To date, the use of broccoli byproducts has been applied to flour and fiber [
15]. On the other hand, scientific research of chemical composition and biological activities mostly focused on broccoli florets, whereas investigations regarding bioactive potentials of other broccoli parts were generally limited. Several studies demonstrated that broccoli leaves and seeds had good nutritional values, as well as notable antioxidant and antiproliferative capacities [
16,
17,
18]. Nevertheless, to our knowledge, there are no reports available about the mechanism of cancer cell death such as apoptosis induction or cell cycle arrest, and other biologically active properties such as anti-inflammatory or antimicrobial effects of these non-edible parts.
Therefore, the present study was designed to investigate and compare polyphenolic contents and diverse bioactivities of edible and non-edible parts of broccoli on various in vitro models. It is the first to examine the apoptotic and antibacterial activities of broccoli leaves and seeds. Significantly, this study aimed to validate these byproduct extracts as biologically active substances for functional food and nutraceutical applications.
3. Discussion
Every year, the agri-food industry generates a massive amount of food waste around the world. It leads to environmental and food security issues owing to the impacts related to the production and treatment [
2]. The valorization of agriculture and food byproducts is an objective of the European Union to support sustainable development [
8]. Previously, these agri-food residues were applied for animal feeds, compost, or essential oil extractions [
1,
3]. Presently, there is an increasing interest in the research associated with providing bioactive components of food byproducts as new ingredients for functional product formulations. Several studies investigated the chemical composition and biological activities of tomato leaves, orange peels, hazelnut shells, mango peels, or pomegranate marcs [
4,
5,
6,
7,
23,
24]. For example, orange peels revealed an anti-inflammatory bowel potential after 15 days of application in mice treated with dextran sodium sulfate [
4]. The isolated proanthocyanidins from grape seeds possessed significant antioxidant and anti-inflammatory properties [
6]. The extracts of pomegranate internal membranes by green technologies were used to enrich a cosmetic hydrophilic gel, as a model cosmetic [
7]. Broad bean pods were demonstrated as a rich source of bioactive ingredients with antimicrobial, enzymes inhibitory, and anti-diabetic properties [
25].
For functional product formulations,
Brassica vegetables and their byproducts could be a valuable source of bioactive compounds and health benefits [
9,
18,
26]. Numerous publications, counting our previous study, analyzed the bioactive composition and the potentially beneficial bioactivities of broccoli florets [
9,
12,
27,
28,
29]. So far, the functional properties of broccoli byproducts have not been well documented. Some studies revealed the nutritional value, antioxidant, and antiproliferative activities of broccoli leaves and seeds [
16,
17,
18]. Particularly, a report indicated that broccoli roots contained high levels of glucosinolates and related hydrolysis products, as well as provided information of the glucosinolate metabolome and transcriptome for distinct tissues of broccoli [
30]. In the present study, LE and SE showed substantial polyphenolic concentrations, along with considerable antioxidant and cytotoxic activities on various in vitro models (
Table 1,
Table 2,
Table 3 and
Table 4). Additionally, corresponding with potential cytotoxicity towards cancer cells, our study confirmed the selective activity of broccoli extracts on the normal liver cells (FL83B). Nonetheless, this test was not previously detailed for all broccoli tissues.
Tumors are caused by the disorder of cell proliferation and the obstruction of cell apoptosis. Searching for bioactive compounds, which can induce cell apoptosis, would be an important strategy for potential chemotherapeutic agents [
20]. Cancers are the consequence of cell-cycle dysregulation. Targeting the checkpoint signaling pathway, which generally leads to an arrest at the G1/S or G2/M phases, would be another effective therapeutic strategy [
19]. Loss of MMP is closely linked to the activation of caspase-9, which plays a critical role in the initiation and maintenance of apoptosis [
20]. Therefore, in this study, to evaluate if the cytotoxicity of broccoli extracts was the result of the induction of apoptosis or cell cycle arrest, Caco-2 cells were exposed to FE, LE, and SE and cell cycle distribution and MMP were analyzed using flow cytometry. The results showed significant increases in cell percentage with subG1 DNA content, obvious cell cycle arrests, as well as notable decreases of MMP (
Figure 1 and
Figure 2). To date, a few authors, along with our earlier publication [
28], reported the mechanism of cancer cell death by the effects of broccoli, but focused on florets and sprouts only. A study presented the high efficacy of broccoli florets and sprouts on viability inhibitory and proapoptotic induction in human cancer cell lines [
31]. Another study revealed the apoptotic role of isothiocyanate from broccoli florets in experimental lung carcinogenesis in Swiss albino mice [
32].
Many reports demonstrated that the antimicrobial activity of plant extracts was attributed to the presence of bioactive compounds such as phenols, glucosinolates, or organic acids. Plant phenolics and extracts rich in such substances could inhibit the growth of various bacterial pathogens [
33]. Over the last years, the detrimental effect on microorganisms was extensively studied on broccoli florets [
34,
35,
36,
37]. For instance, the antimicrobial activity of broccoli florets was evaluated using flow cytometry to identify dead cells probably owing to cell disruption and effusion of internal contents [
36]. The capacity of florets against pathogenic bacteria, yeast, and fungi was also proven by the antimicrobial peptides [
37]. However, there is no study available about the antimicrobial potential of broccoli leaves and seeds. This is the first study to present the antibacterial potency of LE and SE towards foodborne pathogens by the IZDs and MICs (
Table 5 and
Table 6). Briefly, together with our previous findings on broccoli sprouts [
28], the present study indicated that both edible and non-edible parts of broccoli were effective for different biological activities.
4. Materials and Methods
4.1. Plant Materials and Preparation of Crude Extracts
Broccoli (Brassica oleracea L. var. italica) samples, Green King variety, were obtained from Known-You Seed Co. Ltd., Kaohsiung, Taiwan (22°39′39.8″ N 120°25′32.0″ E) and its experimental station, Tainan, Taiwan (23°04′46.5″ N 120°17′43.4″ E), in four periods, including November 2017, December 2017, March 2018, and November 2018. These are the end of the harvesting stages of broccoli, after being cultivated in conventional cool seasons in Taiwan. The samples were divided into three groups, comprising florets (a main edible part), leaves, and seeds (non-edible parts). Seeds in the original intact package were purchased directly from the company. Florets were harvested in 60–90 days after transplanting. Broccoli heads, which reached 4–8 inches in diameter with the dense and firm flower clusters, were collected. Thereafter, the side shoots stopped growing; this is a sign the plant is no longer producing, leaves were picked as unused byproduct. Mature leaves without signs of mechanical damage were selected. Each sample group was obtained from various plants for each biological replication.
Samples were dried in a TR 120 drying oven (Nabertherm GmbH, Lilienthal, Germany) at 40 °C, grounded into fine powders, and stored at −20 °C until extraction. Firstly, 10 g of sample powder was transferred into a flask containing separately 100 mL of each solvent (70% ethanol, 70% ethanol, or hot water) to reach the concentration of 0.1 g dry weight (DW)/mL. Ethanol and methanol extractions were performed by a SR-2 DW shaker (Taitec, Koshigaya-shi, Saitama-ken, Japan) for 24 h at 30 °C. Hot water extraction was undertaken by adding samples into boiling distilled water for 15 min. Extracts were centrifuged at 10,000 rpm for 10 min and supernatants were filtered through Whatman No.1 filter paper (Merck KGaA, Darmstadt, Germany). The filtrates were collected and stored at 4 °C for further polyphenolic content and antioxidant determinations [
38,
39]. Subsequently, the filtrates were concentrated to dryness using a RV8 rotary evaporator (IKA Works Guangzhou, Guangzhou, China) and an Alpha 1-2 LDplus freeze dryer (Martin Christ GmbH, Osterode, Germany). The lyophilized extracts were stored at 4 °C for further anticarcinogenic and antibacterial experiments.
4.2. Antioxidant Properties
4.2.1. Antioxidant Activity Assays
For antioxidant properties, DPPH (2,2-diphenyl-1-picrylhydrazyl) radical scavenging, ABTS (2,2’-azino-bis-3-ethylbenzothiazoline-6-sulphonic acid) radical cation decolorization, and reducing power assays were utilized. Ascorbic acid solution (0.5 mg/mL) was employed as a positive standard.
The DPPH assay was performed as reported earlier [
40]. Then, 1 mL of extracts (0.1 g DW/mL) was mixed with 4 mL of DPPH methanolic solution (0.1 mM), and incubated for 60 min to allow for complete reaction. The absorbance of mixtures was read at 517 nm, and results were calculated as the absorbance inhibition (%) of a blank composed of DPPH and methanol as extract solvent.
The ABTS radical cation was produced by mixing ABTS stock solution (7 mM) with potassium persulfate (2.45 mM) and kept at room temperature in the dark for 16 h before use. Then, 1 mL of diluted ABTS solution (0.70 absorbance at 734 nm) was mixed with 10 µL of extracts (0.1 g DW/mL). The absorbance of the mixtures was read at 415 nm. The antioxidant capacity was calculated based on the calibration curve and expressed as Trolox equivalents (µmol TE/g DW) [
41].
The potency of extracts to reduce iron (III) was tested, as reported earlier [
40]. Then, 1 mL of extracts (0.1 g DW/mL) reacted in 0.5 mL of 0.2 M phosphate-buffered saline (PBS) and 0.5 mL of 1% potassium ferricyanide (Fe
3+). The reaction was stopped by adding 0.5 mL of 10% trichloroacetic acid. After centrifugation, 0.25 mL of 0.1% ferric chloride was added to 1 mL of supernatants to form ferric ferrous complex with an absorption maximum at 700 nm. The absorbance indicated reducing power.
4.2.2. Determination of Total Phenolic, Flavonoid, and Vitamin C Contents
The total phenolic content (TPC) was determined by the Folin–Ciocaltecu method, as previously described [
28]. The absorbance of the reaction mixture was measured at 765 nm using a SPECTROstar microplate reader (BMG labtech, Ortenberg, Germany). Gallic acid standards were applied to generate a calibration curve. TPC was expressed as gallic acid equivalent (mg GAE/g DW).
The total flavonoid content (TFC) was determined with aluminum chloride (AlCl
3), as previously described [
28]. The absorbance of the reaction mixture was measured at 490 nm. TFC was expressed as catechin equivalent (mg CE/g DW).
The vitamin C content (VCC) was determined using the Folin–Ciocalteu reagent, as previously described [
28]. The absorbance of the reaction mixture was measured at 760 nm. VCC was expressed as ascorbic acid equivalent (mg AA/g DW).
4.2.3. High-Performance Liquid Chromatography (HPLC) Analysis of Phenolic Compounds
The phenolic composition of ethanolic extracts was analyzed using an HPLC system (Hitachi Chromaster, Tokyo, Japan) according to the method previously described [
28]. Sample compounds were separated in a NUCLEODUR
® C
18 HTec column (250 × 4.6 mm, the particle size of 5 µm, Macherey-Nagel, Düren, Germany) at 25 °C by a flow rate of 1 mL/min. The mobile phase consisted of a mixture of 0.1% trifluoroacetic acid solution (A) and 100% methanol (B) with the following gradient: from 0–3 min, 10% B; from 3–20 min, 30% B; from 20–30 min, 40% B; from 30–50 min, 60% B; and from 50–60 min, 20% B. The eluted peaks were detected at 280, 320, and 360 nm. Phenolic compounds were identified and quantified by comparing the chromatographic behavior and retention times in specific UV spectra with external standards and reported data.
4.3. Anticancer Properties
4.3.1. Cell lines and Culture Conditions
The selected cell lines for the cytotoxicity test were procured from the Food Industry Research and Development Institute (Hsinchu, Taiwan). Hepatocellular carcinoma (HepG2) and colorectal adenocarcinoma (Caco-2) cells were cultivated in Dulbecco’s modified Eagle’s medium, while lung carcinoma (A549) and normal liver (FL83B) cells were maintained in F12K medium. Cells were grown at standard conditions (37 °C, 5% CO2, 95% humidity) in the media supplemented with 10% fetal bovine serum, penicillin (100 U/mL), streptomycin (100 μg/mL), and sodium bicarbonate (1.5 g/L).
4.3.2. Cytotoxic Assay
The cytotoxic activity was performed as described earlier [
42]. Cells were seeded at 1 × 10
4 cells per well in 96-well plates. After 24 h of incubation, cell media was removed and attached cells were stimulated with four concentrations of each extract (0.063–0.500 mg/mL) for 24 and 48 h. Cisplatin (0.006–0.050 mg/mL), a traditional chemotherapeutic agent, was treated similarly, as a positive control. Then, 20 μL of MTT (3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide) solution (5 mg/mL) was transferred to each well. After 2 h, supernatants were replaced with 100 µL of DMSO (dimethyl sulfoxide), and optical densities were measured at 570 nm. The results were expressed as a percentage of viable cells with respect to untreated cells, which was defined as 100%. The 50% inhibitory concentration (IC
50) was calculated using the Graphpad Prism 5 software (GraphPad Software, San Diego, CA, USA).
4.3.3. Cell Cycle Analysis
For flow cytometry analysis of DNA content, a total of 1 × 10
6 Caco-2 cells were seeded in a 25 cm
2 culture flask. After an overnight incubation, cells were exposed to extracts (0.200 mg/mL) and cisplatin (0.020 mg/mL) for 24 and 48 h. Cells were collected, washed with PBS, and fixed in cold 70% ethanol overnight at −20 °C. Cells were then washed twice in PBS at a centrifugation speed of 1500 rpm, resuspended in 1 mL of propidium iodide (PI) solution containing PI (0.050 mg/mL), 0.5% Triton X-100, and RNAse A (0.050 mg/mL), and incubated in the dark for 30 min at 37 °C [
43]. The fluorescence intensity was analyzed by the FACS Calibur flow cytometer (BD Biosciences, San Jose, CA, USA). The results were expressed in a histogram as a total percentage of cells from four different cell cycle phases using the BD FACSDiva 8.0 software (BD Biosciences, San Jose, CA, USA).
4.3.4. Assessment of Mitochondrial Membrane Potential (MMP)
DiOC
6 (3,3′-dihexyloxacarbocyanine iodide), a cell-permeable fluorescent lipophilic dye, was used to assess MMP as reported [
20]. After treatment with extracts (0.200 mg/mL) and cisplatin (0.002 mg/mL) for 6, 12, 24, and 48 h, Caco-2 cells were collected, washed with PBS, and stained with 1 mL of DiOC
6 solution (0.5 µg/mL) in the dark for 30 min at 37 °C. The fluorescence intensity was analyzed by the FACS Calibur flow cytometer. The results were expressed in a histogram as a percentage of positive cells for the given dye, which represented MMP (ΔΨ), using the BD FACSDiva 8.0 software.
4.4. Antibacterial Properties
4.4.1. Bacterial Strains and Culture Conditions
Two strains of gram-negative bacteria (Staphylococcus aureus CCRC 11863 and Bacillus subtilis CCRC 14199) and two strains of gram-positive bacteria (Salmonella typhimurium CCRC 12497 and Escherichia coli CCRC 11634) were used in this study. All these foodborne pathogens were obtained from the Culture Collection and Research Center (CCRC), Hsinchu, Taiwan. To obtain subcultures for further sample treatment, S. aureus and E. coli were cultured in the Tryptic soy broth, whereas B. subtilis and S. typhimurium were cultivated in the nutrient broth at 37 °C for 18 h.
4.4.2. Agar Diffusion Method
The agar well diffusion method was firstly employed to evaluate antibacterial activity [
44]. Agar plates were inoculated by spreading a 1% bacterial culture suspension (1 × 10
6 CFU/mL) over the entire agar surface. On the agar plate surface, the wells with a diameter of 9 mm were then punched by a sterile cork borer, and 100 µL of extract solution (diluting extracts with 20% DMSO to reach a concentration of 50 mg/mL) was introduced into each well. Standard antibiotics (ampicillin and amoxicillin, 0.1 mg/mL) and negative control (20% DMSO) were included. Plates were incubated under suitable conditions depending upon the tested bacteria at 37 °C for 16 h. The results were expressed as the diameter (mm) of inhibition zones using a caliper.
4.4.3. Broth Microdilution Method
The minimum inhibitory concentration (MIC) was determined by broth microdilution method [
45]. Multiple two-fold dilutions of extracts (0.1–50 mg/mL) in the nutrient broth were performed in 96-well plates. Ampicillin and amoxicillin (0.1 mg/mL) were included as positive controls. A log-phase culture of the bacteria was diluted and inoculated into plates to a final concentration of 1 × 10
5 CFU/mL. The plates were incubated at 37 °C for 16 h. Then, 20 μL of MTT (5 mg/mL) was transferred to each well as a colorimetric indicator of bacterial growth for 2 h. The MIC (mg/mL) was defined as the lowest concentration of extracts at which the tested bacteria showed no visible growth.
4.5. Statistical Analysis
All experiments were carried out in three replicates (n = 3). The results are presented as means ± standard deviation (SD). Statistical analysis was performed using SPSS 22.0 software (SPSS Inc., Chicago, IL, USA). The data were analyzed using one-way analysis of variance (ANOVA) with Tukey’s post hoc test for comparing mean significant differences between samples (p ≤ 0.05).