The Phytochemical Profile and Anticancer Activity of Anthemis tinctoria and Angelica sylvestris Used in Estonian Ethnomedicine

The aerial parts of Anthemis tinctoria L. and Angelica sylvestris L. and the roots of A. sylvestris have been used as traditional anticancer remedies in Estonian ethnomedicine. The aim of this study was to investigate content of essential oils (by gas chromatography) and polyphenolic compounds (using two different methods of high performance liquid chromatography–mass spectrometry (HPLC–MS)) of both plant species, as well as the in vitro anti-cancer effects of their essential oils and methanolic extracts. The average (n = 5 samples) yield of essential oils was 0.15%, 0.13%, and 0.17%, respectively. The principal compounds of the essential oil from the aerial parts of A. tinctoria were palmitic acid (15.3%), p-cymene (12.6%), and α-muurolene (12.5%), and α-pinene (45.4%), p-cymene (15.5%), and β-myrcene (13.3%) in aerial parts of A. sylvestris, while isocaryophyllene oxide (31.9%), α-bisabolol (17.5%), and α-pinene (12.4%) were the main constituents in the roots. The most abundant phenolic compounds in aerial parts were the derivatives of caffeic acid, quinic acid, and quercetin; the main compounds in roots of A. sylvestris were chlorogenic acid, quinic acid, and naringenin. The strongest anticancer effects were observed in essential oils of A. sylvestris roots and aerial parts on human carcinoma in the mouth cells (KB, IC50 19.73 μg/mL and 19.84 μg/mL, respectively). The essential oil of A. tinctoria showed a strong effect on KB and LNCaP cells (27.75–29.96 μg/mL). The methanolic extracts of both plants had no effect on the cancer cells studied.


Introduction
According to the estimation of the World Health Organization, the approximately 12.7 million new cancer cases that occurred globally in 2008 will increase to 21.3 million by 2030 [1]. When plants with expressed anticancer activity used in Estonian ethnomedicine were studied [2], 44 species with potential anticancer properties were elicited, five of which Anthemis tinctoria L., Angelica sylvestris L., Pinus sylvestris L., Sorbus aucuparia L. and Prunus padus L. had not been previously described with respect to their tumoricidal activities in the scientific literature. Later, the anticancer activity of essential oil and methanolic extract from needles of P. sylvestris and bark of S. aucuparia were studied [3,4].
The focus of this study was on A. tinctoria (golden marguerite) and A. sylvestris (wild angelica). Both species, which grow naturally in Estonia, have been used only in local and neighboring folk medicine. Only a few studies have been published on their chemical composition, mainly on the composition of essential oils and polyphenols.
The roots of A. sylvestris contain 0.16% of essential oil, flowers 0.52%, leaves 0.08%, and seeds of A. sylvestris contain 1.1% of essential oil with the main components being limonene, α-pinene, myrcene, β-fellandrene, α-hamrigren, and β-sesquifellandrene [12][13][14]. The essential oil of roots contains mostly sesquiterpenoids, while the essential oil of seeds contains monoterpenoids [13,15]. Moderate amounts of phenolic compounds, and a small amount of flavonoids, have been found in the herb of A. sylvestris [16], but their exact composition is unknown.
The aim of this study was to investigate the chemical composition of A. tinctoria and A. sylvestris, as well as the anti-cancer effects of essential oils and methanolic extracts.

Content and Chemical Composition of Essential Oils
The average (n = 5 samples) yield of essential oils in the aerial parts of A. tinctoria and A. sylvestris, and in the roots of A. sylvestris was 0.15%, 0.13%, and 0.17%, respectively.The main components of the essential oil from the aerial parts of A. tinctoria were palmitic acid (15.3%), p-cymene (12.6%), and α-muurolene (12.5%), while the content of other compounds was less than 10% (Table 1). Ocimenes (Z and E), isoborneol, crysanthenyl acetate, humulene epoxide, 2-pentadecanone, and nerolidon acetate were found in the essential oil of A. tinctoria for the first time. The principal compounds in the essential oil from aerial parts were α-pinene (45.4%), p-cymene (15.5%), and β-myrcene (13.3%), while isocaryophyllene oxide (31.9%), α-bisabolol (17.5%), and α-pinene (12.4%) were the main constituents in root oil of the same plant. The essential oil of aerial parts of A. sylvestris contained more monoterpenes and cyclic monoterpenes but less sesquiterpenes and bicyclic sesquiterpenes than the oil hydrodistilled from the roots of A. sylvestris (Table 2).  (Tables 3 and 4). In contrast, 10 polyphenolic compounds were detected in A. sylvestris root extract, the most abundant of which were chlorogenic acid, quinic acid, and naringenin. Substances common for analyzed herbs and roots were naringenin, chlorogenic acid, quercetin, and coumarylquinic acids.
Roots of A. sylvestris contain a number of phenolic compounds in remarkable quantities that remain unidentified. Some of these have a UV absorption maximum near 306 nm that refers to hydroxystilbenes such as resveratrol or piceatannol and six compounds have a negative collision fragment with m/z = 201 ( Figures S1-S3).
The principal polyphenols in the aerial parts of A. tinctoria were caffeoylquinic (chlorogenic) and dicaffeoylquinic acids as well as several glycosides of quercetin and patuletin, in the aerial parts of A. sylvestris there were also chlorogenic and caffeoylquinic acids and different glycosides of quercetin. The principal polyphenols of A. sylvestris roots were neo-chlorogenic acid, caffeic acid glucosides, feruloylquinic acid, and naringenin together with naringenin chalcone. However, the total polyphenol content (TPC) of aerial parts of A.tinctoria was about 3.1 and 4.5 times higher than the TPC of aerial parts of A. tinctoria and roots of A. sylvestris, respectively, estimated by comparison of UV chromatogram areas at 280 nm. Total content of flavonol glycosides (TCF) of aerial parts of A. tinctoria was about 4 and 8 times higher than TCF of aerial parts of A. tinctoria and roots of A. sylvestris, respectively, estimated by comparison of UV chromatogram areas at 360 nm, the maximum of UV spectra of flavonols. Consequently, the TPC and, particularly, the TCF of the methanol extract of both aerial parts and especially of roots of A. sylvestris was too low to expect remarkable effects on cancer cells.
Phenolic compounds were identified using the selected MRM transition detection method and inhouse database. It was found that aerial parts accumulated most of phenolic compounds while in root samples some compounds were not detected and others were found in lower abundance. The most abundant compounds were phenolic acids (chlorogenic and dicaffeoylquinic acids). Flavonoid aglycones and their glycosides were detected in lower abundance. Qualitative profiles of selected phenolic compounds were similar between samples, although naringenin was detected only in A. sylvestris samples, and luteolin was detected only in A. tinctoria samples. It should be emphasized that soem of the phenolic compounds, such as neochlorogenic, and chlorogenic acids, rutin, and dicaffeoylquinic acids were confirmed by both LC-MS methods. The presence of caffeic acid, ferulic acid, quercetin, isorhamnetin (detected in the aerial parts of tested species), luteolin, naringenin (found in samples of A. sylvestris), and diosmetine derivatives was also confirmed by both methods. The inequality of methods used and the selected methodological conditions was the basis for supplementing the data obtained. The UPLC-MS/MS method provided more detailed data on luteolin derivatives. Luteolin-7-rutinoside was detected in all samples tested by UPLC-MS/MS, while other luteolin derivatives (luteolin-7-glucoside, luteolin-4-glucoside, and luteolin aglycone) were detected only in A. tinctoria aerial parts. The UPLC-based triple quadrupole MRM transitions were more abundant for detection of aglycone monomers as free ferulic acid, caffeic acid, and chlorogenic acid (which was detected in all samples), as well as quercetin and isorhamnetin aglycones (which were detected in aerial parts of A. sylvestris), and apigenin aglycone (found in all samples). The quercetin derivative quercitrin (quercetin 3-rhamnoside) was detected in all samples by UPLC-MS/MS as against flavone isorhamnetin-3-rutinoside, which was characterized only in the aerial parts of A. sylvestris.   Altogether, ion trap enabled us to identify and semi-quantify 47, and triple quadrupole 22, substances at least in one of the studied plant extracts. This difference between results can be explained by the greater versatility of ion trap, but triple quadrupole is in turn more sensitive, allowing lower concentrations of substances to be determined. It is therefore advisable to analyze any plant extract using two different types of mass spectrometry.

Total Content of Different Compounds
Quantitative analysis of polyphenols was performed mainly, with the exception of chlorogenic acid, by substance groups. As can be seen from Tables 5 and 6, most of the phenolic compounds in the three studied plant materials were quinic acid derivatives (chlorogenic acids, di-and caffeoylquinic acids, and feruloylquinic acid). The second biggest group, which is completely absent in the roots of A. sylvestris, were glycosides of different flavonols (quercetin, myricetin, isorhamnetin, kaempferol, and patuletin). Phenolic acid glycosides (caffeic and coumaric acid), flavanons (naringenin and eriodictyol), flavones (luteolin), and stilbenols(pinosylvin) were also represented, mostly in small quantities. In particular, the herb of A. tinctoria was distinguished by its abundance in polyphenols both qualitatively and quantitatively. The aerial parts of A. tinctoria contain much more total phenolics, total chlorogenic acids, and total flavonols than the herb of A. sylvestris whose roots showed the lowest concentrations of the mentioned compounds (Table 5). A similar result was estimated when we analyzed total quinic acid derivatives; the roots studied did not contain flavonols ( Table 6). The total phenolics and chlorogenic acids were calculated where the ratio between plant material and solvent was 1:10.

Discussion
In previous studies, the essential oil content of A. tinctoria inflorescences was 0.10% [5] and 0.10-0.14% [17]. The amount of essential oil obtained in the present study from the aerial parts was slightly higher (0.15%). The content of essential oil in the A. sylvestris leaves was 0.08%, in flowers 0.52%, in fruits 1.1%, and in roots 0.16% [15]. In the present study, the content of essential oil found in the herb was about 2 times higher than in the leaves and about 3.5 times lower than in the flowers and similar to the yield of essential oil measured in roots of A. sylvestris.
The three main compounds of the essential oil from A. tinctoria (palmitic acid, p-cymene, and α-muurolol (Table 1) were also found in our earlier study of the same species of Estonian origin [7]. They were not identified in the oil from flowerheads of A. tinctoria cultivated in the Slovak Republic [6] and were not mentioned among of principal compounds, which were 1,8-cineole (7.9%), β-pinene (7.3%), and decanoic acid (5.4%) [18]. Thus, A. tinctoria essential oil does not contain just one dominant component, rather, three major components have typically been found in all of these studies. A. tinctoria plants seems to be present in several essential-oil chemotypes.
The main compounds in the essential oil of aerial parts and seeds of A. sylvestris collected from Serbia were limonene (66.6 and 75.3%) and α-pinene (19.0 and 9.6%) [12,14]. In the present study, we did not found limonene, but the concentration of α-pinene (12.4%) was similar to that in the mentioned papers. The content of limonene was lower (5.6%) and the concentration of α-pinene higher (25.6%) in the fruits of A. sylvestris grown in Turkey. A study by Agalar et al. (2020) [15] found that the content of limonene in leaves, flowers, fruits, and roots was 0.4-4.2%. The different results from several authors may depend on different varieties (var. vulgaris or var. elatior) [13]. Thus, the absence of limonene may be influenced by the specific chemotype of A. sylvestris growing in Estonia.
In the aerial parts of A. Tinctoria, 15 flavonoid aglycones were identified [11]. Similarly to the present study, caffeic acid, rutin, naringenin, chlorogenic acid, patuletin, and quercetin have been previously found in A. tinctoria inflorescences. However, apigenin-7glucoside and rosmarinic acid were not detected in the current study although they were detected by previous authors [8,9].
Many studies report the multiple anticancer properties of plant-derived polyphenols, including inhibitory effects on the proliferation of cancer cells, tumor expansion, angiogenesis, inflammation, and metastasis. Overall, polyphenolic compounds are attractive molecules for cancer treatment [19]. Flavonols quercetin and kaempferol, that are represented mostly by various glycosides derivatives in the aerial parts of both A. tinctoria and A. sylvestris, have been shown to have various anticancer effects [19]. Quercetin has been reported to reduce both the risk and progression of cancer through free-radical scavenging activity, protecting cells from oxidative stress, inflammation, and DNA damage due to its antioxidant properties and modulating growth of many cancer cell lines by blocking cell-cycle progression and tumor-cell proliferation and by inducing apoptosis [20]. A group of polyphenols that includes apigenin, quercetin, curcumin, resveratrol, EGCG, and kaempferol has been shown to regulate signaling pathways that are central for cancer development, progression, and metastasis [21]. Chlorogenic acid has also shown anticancer activity [22]. The fact that these compounds did not demonstrate an anticancer effect in our tests can be explained by the fact that the extracts tested contained polyphenols, not chlorogenic acids, in various glycosidic forms, which usually have lower biological activities than the corresponding aglycones [23]. It will be necessary to repeat the experiments using enzymatically hydrolyzed plant extracts. Other explanations, for example differences in cancer cell lines, should also be considered.
The strongest anticancer effects, with IC 50 less than 20 µg/mL, had the essential oils from aerial parts and roots of A. sylvestris on KB (human carcinoma in the mouth) cells. All other results showed moderate or weak activity to different cancer cells ( Table 5). The same study of ethnomedicinal traditions in cancer therapy [2] also showed the use of chaga mushroom (Inonotus obliquus), chamomile (Chamomilla recutita), marigold (Calendula officinalis), and Scots pine (Pinus sylvestris), which have been previously studied in the context of possible anti-cancer effects in vitro. The methanolic chaga muchroom extract exhibited the strongest cytotoxic effects against promyelocytic leukemia cells (HL-60) and lung adenocarcinoma (LU-1, 32.2 and 38.0 µg/mL, respectively), and modest cytotoxic effects against colon adenocarcinoma (SW480), liver hepatocellular carcinoma (HepG2), oral epidermoid carcinoma (KB), and prostate cancer (LNCaP, 41.3-57.7 µg/mL) [24]. The cytotoxic activity of methanolic extract of chamomile (Chamomilla recutita) flowers on SK-MEL-2 (melanoma cells, IC 50 value 40.7 µg/mL) was approximately twofold higher than on KB cells (IC 50 value 71.4 µg/mL). With the marigold (Calendula officinalis) flower extracts, the anticancer activity was more than 100 µg/mL in both cell lines studied [2]. The essential oil from the needles of P. sylvestris showed the stronger cytotoxic effect to both negative and positive breast cancer cell lines (MCF7 and MDA-MB231, both IC 50 29 µg/mL) than pine methanolic extract (IC50 42 and 80 µg/mL, respectively) [4]. In this context, the potency of the anti-cancer activity of A. sylvestris against different cancer cell lines can be considered remarkable.
α-Pinene has a weak anti-cancer effect, but its use is limited due to its toxicity to normal body cells [25]. A study of liver carcinoma cells has shown moderate anti-cancer activity of pinene [26]. β-Myrcene has been shown to inhibit specific types of breast cancer cells [27]. In vitro experiments with α-bisabolol on pancreatic cancer cells showed strong anti-cancer activity [28]. The rather high concentration of these three terpenoids in the essential oil of A. sylvestris is interesting and could have an effect on other cancer cells not yet studied.
To the best of the author's knowledge, there are no previous studies on the polyphenolic composition of A. sylvestris, which has been studied, by us, for the first time. The novelty of the study is also the detection of the anticancer activity of A. tinctoria and A. sylvestris. Ocimenes (Z and E), isoborneol, crysanthenyl acetate, humulene epoxide, 2-pentadecanone, and nerolidon acetate were found in the essential oil of A. tinctoria for the first time. Additionally, the MS analysis was performed by two distinctive methods.

Plant Material
The herbs from A. tinctoria were collected in July 2019 from Võrumaa, Estonia (57.

Hydrodistillation of Essential Oil
The essential oils were isolated from dried aerial parts of A. tinctoria, as well as from aerial parts and roots of A. sylvestris by the modified hydrodistillation method described in the European Pharmacopoeia [29] using 20-45 g of plant material, a 500 mL roundbottomed flask, and 400 mL water as the distillation liquid, and 0.5 mL of xylene in the graduated tube was added to take up the essential oil. The distillation time was 3 h at a rate of 3-4 mL/min. To improve consecutive chromatographic analyses, hexane was used instead of xylene. Each plant material was distilled five times to obtain a sufficient amount of essential oil for anti-cancer studies.

Making of Methanolic Dry Extracts
The methanolic extracts of the plant materials were prepared by double maceration by adding 200 mL of methanol to 10 g of crushed sample, which was allowed to stand in the dark at room temperature for 72 h. After filtration, the remaining plant material was returned to the flask and poured into 100 mL of methanol and allowed to stand in the dark for 24 h. After filtration, the two methanolic extracts were combined. The solvent was evaporated on a rotary evaporator in a water bath at 60 • C with a flask rotation of 70 rpm. The pressure was initially reduced to 300 mmHg, then the pressure was slowly increased to 10 mmHg until all the methanol had evaporated. The resulting dry extracts were stored in the freezer, and redissolved in 5 mL of methanol before analysis.

Gas-Chromatografic Analysis of the Essential Oils
The GC analysis of main compounds of essential oil was performed using Agilent (Santa Clara, CA, USA) GC 7890a chromatograph with software Agilent Open Lab CDS Chem Station and with FID on two fused silica capillary columns with stationary phases DB-5 and HP-Innowax (both 30 m × 0.25 mm, Agilent,). The carrier gas was hydrogen with split ratio 1:150 and the flow rate of 30 mL/min was applied. The temperature program was from 50 to 250 • C at 2.92 • C/min, and the injector temperature was 250 • C.
The identification of the oil components was accomplished by comparing their retention indices using Agilent Open Lab CDS Chem Station software. The content (%) of the components in the essential oils was determined from the mean retention times and peak areas of four parallel chromatograms. Components were identified by DB-5 column retention indices compared to databases and literature data. Detection was performed by ion trap instrument (1100 series LC/MSD Trap-XCT, Agilent) with electrospray ionization in negative mode. Carrier gas was dry Helium (5.6 atm).
HPLC-MS/MS spectra were analyzed using HPLC-2D-ChemStation software. Polyphenolic components were identified by HPLC retention time and MS/MS fragments, comparing them with literature data and the inhouse database. More details are available in our previous paper (Rusalepp et al., 2017).

Reagents
The reagents and reference substances were purchased from Sigma-Aldrich, Steinheim, Germany. Purified de-ionized water was prepared with the Milli-Q ® (Millipore, Bedford, MA, USA) water purification system.

Evaluation of Phenolic Compound Profile by UPLC-MS/MS Conditions
Chromatographic separation of plant samples was performed on an Aquity H-class UPLC system (Waters, Milford, MA, USA) under conditions described by González-Burgos et al. [30]. A YMC Triart C18 (100 × 2.0 mm 1.9 µm) column was used for separation of phenolic compounds. The column temperature was maintained at 40 • C. Gradient elution was performed with mobile phase consisting of 0.1% formic acid water solution (solvent A) and acetonitrile (solvent B) with the flow rate set to 0.5 mL/min. A linear gradient profile was applied with following proportions of solvent A: 0 to 1 min-95%, 5 min.-70%, 7 min. 50%, 7.5 to 8 min. 0%, 8.1 to 10 min. 95%. MS and MS/MS analyses of separated peaks of phenolic compounds were carried out with triple quadrupole tandem mass spectrometer (Xevo, Waters, USA). An electrospray ionisation source (ESI) was used to obtain negative ions. Electrospray ionization was applied for analysis with the following settings: capillary voltage-2 kV, source temperature-150 • C, desolvation temperature-400 • C, desolvation gas flow-700 L/h, cone gas flow-20 L/h. Waters (USA) IntelliStart software was used for development of a specific collision energy and cone voltage for each compound of interest.

Cytotoxicity
Human cancer cell lines including hepatocellular carcinoma HepG2, gastric carcinoma MKN7, colon carcinoma SW480, prostate carcinoma LNCaP, and carcinoma in the mouth KB cells were cultured in DMEM or RPMI-1640 cell culture medium, both supplemented with 10% fetal bovine serum. Cells were cultivated at 37 • C in a humidified atmosphere containing 5% carbon dioxide.
The methanolic dry extracts were dissolved in dimethyl sulfoxide (DMSO) to prepare 4 mg/mL stock solutions that were later mixed with cell culture medium to achieve the desired concentrations. The final test concentrations were 0.8, 4, 20, and 100 µg/mL.
The effects of essential oils and extracts on viability of malignant cells were determined by sulforhodamine B cytotoxic assay [31]. Briefly, cells were grown in 96-well microtiter plates with each well containing 190 µL of medium. After 24 h, 10 µL of test samples dissolved in DMSO were added to each well. One plate with no samples served as a day 0 control. The cells were continuously cultured for an additional 48 h, fixed with trichloroacetic acid, and stained with sulforhodamine B, followed by the determination of optical densities at 515 nm using a Microplate Reader (BioRad, Hercules, CA, USA). The percentage of growth inhibition was calculated using the following equation: where OD is optical density or absorbance values. The potent anticancer agent ellipticine was used as a positive control.

Conclusions
Z-and E-ocimenes, isoborneol, crysanthenyl acetate, humulene epoxide, 2-pentadecanone, and nerolidon acetate were found in the essential oil of A. tinctoria for the first time.
The total polyphenol content of aerial parts of A.tinctoria was about 3.1 and 4.5 times higher than of aerial parts of A. tinctoria and roots of A. sylvestris, respectively. Total content of flavonol glycosides of aerial parts of A. tinctoria was about 4 and 8 times higher than in aerial parts of A. tinctoria and roots of A. sylvestris.
The principal polyphenols in the aerial parts of both A. tinctoria and A. sylvestris were caffeoylquinic and dicaffeoylquinic acids as well as several glycosides of quercetin and patuletin. In the aerial parts of A. Sylvestris, chlorogenic and caffeoylquinic acids and different glycosides of quercetin were also found.
The strongest anticancer effects were found in A. sylvestris roots' and aerial parts' essential oils on KB cells (IC 50 19.73 µg/mL and 19.84 µg/mL, respectively), and strong to moderate effects on other cell lines with IC 50 ranges 24.69-38.06 µg/mL. The essential oil of A. tinctoria showed a strong effect on KB and LNCaP cells (IC 50 between 27.75 µg/mL and 29.96 µg/mL, respectively). The methanolic extract of aerial parts of both plants had no effect on cancer cells (IC 50 > 100 µg/mL). The essential oils of A. tinctoria and A. sylvestris may have some perspective in development of natural anticancer compounds.