High-Resolution UPLC-MS Profiling of Anthocyanins and Flavonols of Red Cabbage (Brassica oleracea L. var. capitata f. rubra DC.) Cultivated in Egypt and Evaluation of Their Biological Activity

In this paper, biological investigations and a high-resolution UPLC-PDA-ESI-qTOF-HRMS technique were employed for Brassica oleracea L. var. capitata f. rubra DC. (red cabbage) of the family Brassicaceae (Cruciferae), cultivated in Egypt, for the first time. The positive ionization mode is usually performed to identify anthocyanins. However, this technique cannot differentiate between anthocyanins and corresponding non-anthocyanin polyphenols. Thus, the negative ionization mode was also used, as it provided a series of characteristic ions for the MS analysis of anthocyanins. This helped in identifying five kaempferol derivatives for the first time in red cabbage, as well as nine—previously reported—anthocyanins. For the biological investigations, the acidified methanolic extract of fresh leaves and the methanolic extract of air-dried powdered leaves were examined for their antioxidant, antimicrobial, and anticancer activities. The freshly prepared phenolic extract was proven to be more biologically potent. Statistical significance was determined for its anticancer activity in comparison with standard doxorubicin.


Introduction
Anthocyanins are brightly colored pigments that have attracted the interest of many researchers in recent decades. They are found in different natural sources (fruits and vegetables) which may vary in color and/or composition. Unlike the anthocyanins present in different fruits and vegetables, the anthocyanins of red cabbage leaves are highly conjugated with a variety of sugars and/or acyl groups; with the lack of pure standards, it would be a challenge to fulfill their characterization and identification [1][2][3][4]. Several studies on the characterization and identification of secondary metabolites in complex plant extracts showed that liquid chromatography coupled to photodiode array detection is very effective for anthocyanins analysis [5,6].
At an acidic pH, anthocyanins exist in a positively charged cationic form; hence, the positive ionization mode has typically been used for the mass analysis of anthocyanins. However, anthocyanins cannot be distinguished from corresponding flavonol glycosides in the positive ionization mode using mass spectrometry. In negative ionization mode, a series of ions, e.g., [M − 2H + H 2 O]and [M − 2H] − doublet ions, were found to be characteristic for the MS analysis of anthocyanins [7]. Therefore, the negative ionization mode was also used in this analysis. The anthocyanins profiling of the leaves of Brassica oleracea L. var. capitata f. rubra growing in Sweden, the United States, and China was previously investigated [1,[8][9][10]. All the previously reported anthocyanin aglycones were cyanidin, glycosylated mainly with glucose and/or sophorose (diglucoside), and acylated with various aromatic and aliphatic acids (caffeoyl, p-coumaroyl, feruloyl, p-hydroxybenzoyl, sinapoyl, and oxaloyl). Red cabbage growing in India, China, and many other countries was investigated for its antioxidant, antimicrobial, and anticancer activities [11][12][13]. On the contrary, nothing was traced concerning the anthocyanins content and/or the biological effects of red cabbage cultivated in Egypt.
Considering the increasing numbers of cancer victims in Egypt, especially those not yet in their twenties [14], and because of the numerous side effects of the current therapies, it was necessary to find an alternative natural remedy with minimal (or hopefully with no) side effects. Being promising natural pharmaceuticals, it was of interest to define the phytochemical composition of the anthocyanins content in Egyptian red cabbage leaves, as well as the co-existing phenolics, using high-resolution (UPLC-MS) profiling, and to examine its pharmacological activity as an anticancer, antioxidant, and antimicrobial agent in attempt to use it as a natural original remedy in such fields.
The methanolic extract of air-dried powdered leaves as well as the acidified methanolic extract of fresh leaves of red cabbage were investigated for their antioxidant, antimicrobial, and anticancer biological activities. The antioxidant activity was measured using the ferric reducing antioxidant power (FRAP) method and the antimicrobial activity was determined using the agar well diffusion assay. Minimum inhibitory concentration (MIC) was also determined against the examined organisms. The anticancer activity was determined using a sulforhodamine-B (SRB) assay and a statistical significance analysis was performed in comparison with standard doxorubicin.

Plant Material
Three heads of Brassica oleracea L. var. capitata f. rubra DC (red cabbage) were obtained at full maturity in July 2016 from Nubaria farms (30 • 40 16.428 N 30 • 4 33.024 E), Alex-Cairo desert highway after permission was obtained from the landowner. Plant collection complied with relevant national and international guidelines and legislation. Identification was performed through the courtesy of Dr. A. Abd-Elmogali, a specialized taxonomist at the Agricultural Research Centre, Giza, Egypt and a voucher specimen (voucher code: BOCR375) was deposited in the herbarium of the department of Pharmacognosy, Faculty of Pharmacy, October 6 University, Giza, Egypt.

Preparation of the Extracts
The total alcoholic extract of B. oleracea L. var. capitata f. rubra DC was prepared by the classical method; air-dried powdered leaves (100 g) were extracted using methanol (cold maceration) until exhaustion. The solvent was then removed by vacuum distillation at a temperature not exceeding 40 • C and the residue was saved for biological investigations.
On the other hand, fresh leaves were used for preparing the phenolic extract. Fresh leaves (100 g) were cut using a knife into small pieces (5-7 mm), and then macerated in 1 L of methanol/HCl (98:2 v/v) for 24 h at 4 • C. The powder was extracted until exhaustion by repeating the process of extraction under the same conditions 3 times. The acidified methanolic extracts were combined, filtered, freeze-dried, and separated into two parts, one for the UPLC-MS analysis and the other for the biological investigation. All experiments involving plants adhered to relevant ethical guidelines.
For UPLC-MS analysis, 25 mg of the freeze-dried phenolic extract was homogenized with 2.5 mL of 70% MeOH containing 5 µg/mL of umbelliferone (as the internal standard for relative quantification using UPLC-MS) using a Turrax mixer (11,000 RPM) for five 20 s periods. Each mixing period was separated by a cooling period of 1 min to prevent heating. The extract was then vortexed vigorously and centrifuged at 3000× g for 30 min to remove plant debris. Then, 500 µL was aliquoted and placed on a (500 mg) C18 cartridge which was preconditioned with methanol and water. Samples were eluted using 3 mL of 70% MeOH and 3 mL of 100% MeOH. The eluent was evaporated under a nitrogen stream and the obtained dry residue was resuspended in 500 µL of methanol. Three microliters were used for analysis.
Tentative identification was done by comparing the retention time, UV-vis spectra and high-resolution MS spectrometry (accurate mass and MS/MS fragmentation patterns) of the compounds detected with the reported data [1,[8][9][10] and searching in the existing phytochemical dictionary of natural products database (Compact Reinforced Composite (CRC), Wiley).

Antioxidant Determination
The total antioxidant capacity (TAC) of both extracts was determined using the ferric reducing antioxidant power (FRAP) method. The extracts were prepared in a concentration of 50 µg/mL. Ferrous sulfate heptahydrate (FeSO 4 · 7H 2 O) was purchased from Sigma-Aldrich (St. Louis, MO, USA) and was used as a standard at a concentration of 1000 µM [15]. Antioxidant activity was expressed as the concentration of antioxidants with a ferric reducing ability equivalent to that of 1 mM/L FeSO 4 · 7H 2 O.

Antimicrobial Determination
The antimicrobial activity was determined using the agar well diffusion assay. The extracts and all the standards were prepared as 100 µg/mL. Sterile plates were inoculated with the fresh cultures of Staphylococcus aureus ATCC 6538, Bacillus subtilis ATCC 6633, Candida albicans ATCC 10231, Escherichia coli ATCC 8739, Pseudomonas aeruginosa ATCC 9027, and Aspergillus niger ATCC 16404, obtained from the microbiology laboratory of the Faculty of Pharmacy, October 6 University, Giza, Egypt. Antimicrobial activity was determined by measuring the zones of inhibition (in mm) using ampicillin as a standard against gm +ve bacteria, streptomycin as a standard against gm −ve bacteria, and clotrimazole as an antifungal standard [16]; all standards were purchased from Sigma-Aldrich, Germany. The minimum inhibitory concentrations (MIC) were estimated against each of the tested organisms in triplicate.

Anticancer Determination
The cytotoxicity investigation of the two extracts was carried out using a sulforhodamine-B (SRB) assay following the method reported by [17]. The potential cytotoxicity was determined against HeLa, MCF-7, and HepG-2 cell lines obtained from the National Research Centre, Giza, Egypt, using doxorubicin as a standard. Doxorubicin was supplied from Pharma-

Statistical Analysis
All statistical analyses were performed using GraphPad Prism version 9.2.0 to calculate IC 50 and the level of significance was set at p > 0.05. Quantitative data were described as mean ± standard deviation (SD). GraphPad Prism version 9.2.0 was also used to create multiple bar charts of the cytotoxic activity and cell viability.

Identification of Compounds
(Arapitsas et al. 2008) used the peak spectral characteristic λ vis (at 520 nm), λ 440 (at 440 nm), λ acyl (at 286 nm), and their corresponding absorptivities to identify the mono-or bioside anthocyanins (E 440 /E vis absorptivity ratio of 29-35% for a monoside and 15-24% for a bioside) and E acyl /E vis absorptivity ratio to determine the degree of aromatic acylation (53-69% for monoacylation and 98-128% for diacylation). However, these reported absorptivity ratios are not absolute and vary according to the extraction and analysis conditions used. This is because the spectral characteristics of anthocyanins strongly depend on pH, and dramatically change using different conditions and/or solvents for extraction [18][19][20][21].
In the present study, both positive ( Figure S1 in Supplementary Material) and negative ( Figure S2) ionization modes were performed. The interpretation of the high resolution UPLC-MS data of all the identified compounds was carried out depending on the previously reported data and the phytochemical dictionary of natural products database.
Nine anthocyanins (A 1 -A 9 ) and five phenolics (K 1 -K 5 ) were identified in the leaves of red cabbage cultivated in Egypt. All the identified compounds had a characteristic peak at m/z 287 in the positive ionization mode (m/z 285 in the negative ionization mode) representing the cyanidin and/or kaempferol aglycone, which is further glycosylated with different sugars and/or acids (Table 1). In addition to the characteristic peak at m/z 285, compounds (A 1 -A 9 ) showed both [M − 2H + H 2 O]and [M − 2H]doublet ions in the negative ionization mode, so they were identified as anthocyanins [7]. All detected anthocyanins (except for A 7 ) mainly consisted of the cyanidin-3-glucoside, which is further acylated in different ways.  Compounds A 4 , A 5 , and A 6 were acylated with coumaric, ferulic, and sinapic acids. Compound A 4 was tentatively identified as cyanidin-3-(p-coumaroyl)-sophoroside-5glucoside. When [8] tabulated their results concerning the red cabbage cultivated in Sweden, they mistakenly identified an anthocyanin with a molecular weight of m/z 919 and a similar fragmentation pattern to that of compound A 4 as cyan-3-(p-coumaroyl)-glucoside- Although the phytochemical dictionary of natural products database showed the previous identification of cyanidin-3-feruloyl-glucosyl-galactoside (glycosylation and acylation at C 3 only), the retention data indicated that the glycosylation occurs at both C 3 and C 5 . This is because an anthocyanin glycosylated and/or acylated at C 3 only would have a free C 5 , and consequently an earlier retention time than that of A 7 (Figure 1). Thus, compound A 7 was tentatively identified as cyanidin-3-(feruloyl)-5-glucoside [6]. side (glycosylation and acylation at C3 only), the retention data indicated sylation occurs at both C3 and C5. This is because an anthocyanin glycosyla ylated at C3 only would have a free C5, and consequently an earlier reten that of A7 (Figure 1). Thus, compound A7 was tentatively identified as cy loyl)-5-glucoside [6].  [1,6,9].
For compounds K1-K5, the anthocyanins characteristic [M − 2H + H2O] doublet ions were absent in the negative ionization mode, so they were te tified as kaempferol derivatives. All these kaempferol derivatives were ten fied for the first time in the red cabbage leaves by using the high resolu data (accurate mass and MS/MS fragmentation patterns). All the isolated compounds (A 1 -A 9 and K 1 -K 5 ) are compiled in Table 1, with their retention time data, mass, tandem mass, relative degree of bond unsaturation (RDB), λ acyl , E acyl /E vis , λ vis , E 440 /E vis , and tentative identification.

Antioxidant Activity
The total antioxidant capacity (TAC) of the total alcoholic extract and the phenolic extract of red cabbage leaves was investigated using the FRAP assay. The phenolic extract showed about 96% antioxidant activity compared to the ferrous sulfate heptahydrate standard, while the total alcoholic extract showed no significant antioxidant activity ( Figure S3). This may be due to the decomposition of natural polyphenols and anthocyanins after air-drying.

Antimicrobial Activity
The antimicrobial activity of the total alcoholic extract and the phenolic extract of red cabbage leaves against Aspergillus niger, Bacillus subtilis, Candida albicans, Escherichia coli, Pseudomonas aeruginosa, and Staphylococcus aureus compared to standard ampicillin, streptomycin, and clotrimazole was determined ( Figure S4). A significant antimicrobial activity was detected for the phenolic extract against Staphylococcus aureus, Bacillus subtilis, and Escherichia coli with MICs of about 4 µg/mL, 5 µg/mL, and 5 µg/mL, respectively. Furthermore, it had moderate antimicrobial activity against Candida albicans and Aspergillus niger with an MIC of about 10 µg/mL, but had no effect against Pseudomonas aeruginosa. Once again, the total alcoholic extract showed no significant antimicrobial activity. The minimum inhibitory concentrations of the phenolic extract and the standards against the aforementioned organisms were determined and recorded in Table S1.

Anticancer Activity
The anticancer activities of the total alcoholic extract and the phenolic extract of red cabbage leaves (in comparison with standard doxorubicin) were investigated against the HeLa, MCF-7, and HepG-2 cell lines, and the results are depicted in Table 2. The total alcoholic extract and the phenolic extract of red cabbage leaves showed a high potency against the HeLa cell line. The half inhibitory concentrations (IC 50 ) were 22.78 µg/mL and 17.71 µg/mL, respectively, and results were found to be statistically significant when compared to the standard drug doxorubicin (IC 50 11.38 µg/mL) ( Figure 2). The total alcoholic extract gave weak activity both against the MCF-7 cell line with 5.3% activity and the HepG-2 cell line with 10.2% activity (IC 50 47.84 µg/mL and 69.11 µg/mL, respectively) when compared to standard doxorubicin ( Figure 2). The phenolic extract showed weak to moderate activity against the MCF-7 cell line with 11.0% activity and the HepG-2 cell line with 33.4% activity (IC 50 22.89 µg/mL and 21.08 µg/mL, respectively) in comparison with standard doxorubicin (Figure 2).
The IC 50 of the total alcoholic extract and the phenolic extract against the HeLa and MCF-7 cell lines were found to be statistically significant when compared to doxorubicin. The total alcoholic extract also showed statistical significance against the HepG-2 cell line in comparison with the phenolic extract and/or the standard doxorubicin. In addition, both extracts showed statistical significance to each other against MCF-7, but no significance was present for the HeLa cell line. The statistical significance of cell growth inhibition was determined using six different concentrations (compared to doxorubicin) and is shown in Figures S5-S7. The total alcoholic extract gave weak activity both against the MCF-7 cell line with 5.3% activity and the HepG-2 cell line with 10.2% activity (IC50 47.84 µg/mL and 69.11 µg/mL, respectively) when compared to standard doxorubicin ( Figure 2). The phenolic extract showed weak to moderate activity against the MCF-7 cell line with 11.0% activity and the HepG-2 cell line with 33.4% activity (IC50 22.89 µg/mL and 21.08 µg/mL, respectively) in comparison with standard doxorubicin (Figure 2).
The bioactivity assessment in the aforementioned therapeutic fields was conducted and showed that the phenolic extract of red cabbage leaves is efficient as an antioxidant; with 96% activity when compared to the ferrous sulfate heptahydrate standard, and had a significant antimicrobial activity against Staphylococcus aureus, Bacillus subtilis, and Escherichia coli, but had only a moderate effect against Aspergillus niger and Candida albicans, and no effect against Pseudomonas aeruginosa. Moreover, it had a very significant anticancer activity, especially against the human cervical (HeLa) cell line. Further studies are recommended to explore its possible future use in the aforementioned medical fields.