UPLC–PDA‐ESI–QTOF–MS/MS and GC‐MS analysis of Iranian Dracocephalum moldavica L.

Abstract Dracocephalum moldavica L. is a significant component in the Iranian food basket. This study aimed to investigate the bioactive compounds and biological activities of different extracts obtained from D. moldavica aerial parts. From the aerial parts, a crude methanolic (MeOH) extract and its four sub‐fractions, that is, petroleum ether (Pet), ethyl acetate (EtOAc), n‐butanol (n‐BuOH), and aqueous (water) extracts were obtained. The total phenolic and flavonoid contents as well as the antioxidant and cytotoxic activities of the extracts were determined. Moreover, the phytochemical profiles of the essential oil (EO) and of those extracts with the highest antioxidant activity measured by GC/MS and UPLC–PDA‐ESI–QTOF–MS/MS. Results showed that the highest concentrations of phenols and flavonoids as well as the most potent antioxidant potential according to the DPPH method were determined in the EtOAc and MeOH extracts with IC50 values of 22.0 and 34.4 µg.ml‐1, respectively. Quantitative analysis of these extracts was subsequently performed by UPLC–PDA‐ESI–QTOF–MS/MS. Both extracts contained mainly rosmarinic acid, caffeic acid, and 2‐hydroxycinnamic acid, which may be responsible for their high antioxidant activity. Moreover, none of the extracts showed cytotoxic effects against MCF7, SW48, and a normal cell line of mouse embryonic fibroblast cells (NIH/3T3) in the tested concentrations (up to 400 μg.ml‐1). Additionally, GC‐MS analysis showed that oxygenated monoterpenes (55.4%) were the main constituents of the EO of D. moldavica.


| INTRODUC TI ON
The daily intake of sufficient vegetables has an important role in preventing several diseases (Barends et al., 2019). D. moldavica (Moldavian balm) is a common edible vegetable used daily for the preparation of many Iranian dishes. It belongs to the Lamiaceae family, is up to 80 cm tall, and is native to central Asia (Yousefzadeh et al., 2018). D. moldavica preparations are used in food and in pharmaceutical industries as food additive, tea, and herbal remedy. Traditionally, the plant is applied as analgesic, anticonvulsive, anti-inflammatory, sedative, wound healing, and in the treatment of cardiovascular disorders (Yousefzadeh et al., 2013).
In the Mexican traditional medicine, it is used for the treatment of nervous diseases (Martinez-Vazquez et al., 2012), while in traditional Chinese medicine (TCM), it is mainly used in the treatment of liver disorders, headache, stomach problems, and congestion (Jiang et al., 2014). Furthermore, in TCM in a clinical trial the aqueous extract of D. moldavica was shown to be effective in the treatment of cardiovascular disease, asthma, fatigue insomnia, and neurasthenia (N. Yu et al., 2015).

Phytochemical investigations on the aerial parts of D. moldavica
have demonstrated the presence of several bioactive compounds, including terpenoids, phenolic compounds (rosmarinic and caffeic acid derivatives), flavonoids (kaempferol, quercetin, esculetin, diosmetin, acacetin, apigenin, luteolin, cirsimaritin, salvigenin, santa flavone, agastachoside, and their glycosides), alkaloids, iridoids, and coumarins (Sultan et al., 2008;Yang et al., 2014;Zeng et al., 2010). Phenolic compounds, especially phenolic acid derivatives, such as rosmarinic and caffeic acids, were associated with the high antioxidant potential of D. moldavica (Weremczuk-Jeżyna et al., 2013). Various analytical methods are developed for the identification and quantification of bioactive compounds in medicinal plants. However, in these samples, there are some limitations, including the complexity, the structural diversity, and the low content of bioactive compounds (Adnani et al., 2012). In this regard, the choice of an appropriate technique is important. The application of UPLC-ESI-MS in the identification of natural compounds has attracted much attention because of its high resolution for the separation of complicated samples, analysis speed, sensitivity, selectivity, specificity, and reduced solvent consumption . As it is a significant component in the Iranian food basket, D. moldavica was selected for this study. To the best of our knowledge, there is no comprehensive study on this edible

| Preparation of the extracts
The aerial parts were washed with tap water and dried. For extraction of plant materials, all solvents were purchased from Dr. Mojallali Industrial Chemical Complex Co. 400 g dried material was powdered and macerated in methanol (analytical grade, 99.5%) for 24 hr (3 times, 1 L) at room temperature. The obtained extract was filtered using filter papers (Whatman ® No.1, Merck) and the organic solvent concentrated under a vacuum. Then, the entire extract was suspended in water (50 ml) and partitioned with Pet (200 ml), EtOAc (200 ml), and n-BuOH (200 ml), successively. Afterward, the solvents were evaporated under reduced pressure to get the different subfractions. To prepare the EO, the aerial parts of the plant were subjected to hydrodistillation (Clevenger-type apparatus, Pyrexfan Co) for 3 hr. The obtained EO was dried over anhydrous sodium sulfate (Merck) and stored in the dark until further testing.
The antioxidant capacity was then calculated using the following Equation (1):

| β-carotene linoleic acid bleaching (BCB) assay
The BCB assay was conducted according to the standard method (Kulisic et al., 2004). In brief, β-carotene (0.1 mg, ≥93%, Merck) was dissolved in 0.5 ml chloroform and mixed with 10 mg linoleic acid (≥99%, Merck) and 100 mg Tween-40. Then, the chloroform was evaporated at 50℃, distilled water (25 ml) was added and the mixture sonicated for 1 min. An initial absorbance was recorded at 470 nm (time =0 min). Aliquots of the β-carotene/linoleic acid solution (200 μl) were mixed with the prepared extracts (50 μl) and incubated at 50℃. The absorbance was measured at 470 nm after 120 min incubation. Antioxidant activity of the extracts was calculated by Equation (2): where A A(0) and A A(120) are the absorbances of sample at times 0 and 120 min, while A C(0) and A C(120) are the absorbances of control after 0 and 120 min.

| Cytotoxic activity
Human breast cancer cell line MCF7, colorectal cancer cell line SW48, and a normal cell line mouse embryonic fibroblast cells NIH 3T3 were provided by the National Cell Bank of Iran (Pasteur Institute). They were kept with 10% (v/v) fetal bovine serum (FBS) (Gibco), penicillin/streptomycin at 100 IU/ml and 2 mM L-glutamine.
Cultures were incubated with 5% CO 2 in a humidified atmosphere at 37ºC. The cytotoxic effect of the prepared extracts was assessed using the AlamarBlue ® (BioSource Invitrogen) proliferation assay.
Briefly, cells were seeded in 96-well plates at a density of 1 × 10 4 .

| Chemical profiles and phytochemical content
The ESI operating conditions for MS spectra acquisition in negative mode were as follows: capillary voltage, 2.6 kV; cone, 40 V; desolvation temperature 500℃; and source temperature, 150℃. The desolvation and cone gas flow rates were 0 and 1,000 L/h, respectively. Nitrogen (99.80% N28, Air Liquide, Auderghem, Belgium) was used for both desolvation and cone gas. Sample analysis was done independently in MS E acquisition (E is the collision energy) applying a full scan mode (50-1200 m/z range), in 1 s scan time. The precursor mass spectra acquisition was done in two continuous modes, a no collision energy mode, and a high collision energy (15-35 eV).
Leucine enkephalin (Sigma-Aldrich) was used as internal reference (LockSpray ™ ) to calibrate the ESI source. The data were acquired by a MassLynx™ 4.1 software (Waters).

| Sample preparation
Plant extract, 4 mg, was dissolved in 2.0 ml water/methanol (1:1; v/v) and then mixed for 10 min. Then, the sample was filtered using a membrane filter (0.20μm) prior to injection.

| Identification and quantification of compounds
Compounds were identified and quantified in accordance to the retention times and mass spectral data (mass-to-charge (m/z), molecular peaks and their fragmentation) of the calibration standards.
The analyte concentration was calculated using calibration curves of pure standards (Sigma-Aldrich). Stock solution of each pure calibration standard (1 mg.ml -1 ) was prepared in methanol, and dilutions were made at 6 levels (1, 5, 10, 25, 50, 100 μg.ml -1 ) for the calibration curves. Results were expressed as μg.g -1 pure extract. The quantification was done in duplicate.

| Total phenolic (TPC) and total flavonoid contents (TFC)
The total phenolic content (

MS analysis
Among the extracts of D. moldavica, the EtOAc one exhibited the strongest scavenging activity with an IC 50 value of 22.0 ± 2.1 µg.

| Cytotoxic activity
Extracts of D. moldavica in a concentration range from 50 to 400 μg.  Abbreviation: ND, not detected.