Background

Micromeria fruticosa (L.) Druce spp. brachycalyx P.H. (M. fruticosa spp. brachycalyx) is grown in South Anatolia. This species is known as “tas nanesi” and has the smell of peppermint due to its essential oil components. M.fruticosa (Lamiaceae) is widely used in traditional medicine in the form of herbal tea against disorders such as heart diseases, headaches and skin infections. Chromatographic (HPLC, GC, GC/MS) studies have shown that the essential oil of M. fruticosa contains linalool, pulegone, piperitenone components majorly [1,2,3,4].

Rhus coriaria L. (R. coriaria) is grown widely in Africa, South Anatolia, the Mediterranean region and West Asia. It is also common in the Mediterranean and South-eastern of Turkey. This plant, which called sumac, is used as a spice and sauce. It is known that it is traditionally used in diseases such as stomach ailments, hypertension, diuresis and diabetes. It is also known to be used in cancer treatment [5]. It is known that the extracts from the fruits of sumac contain organic acids (malic, citric, tartaric fumaric), apigenin neohesperidoside-I, myricetin-glucosides, quercetin-3-O-rhamnoside, tannins and terpenoids [6]. The leaves of the plant have gallic acid, myricetin, quercetin, kaempferol and high levels of tannins [7]. In addition, the essential oil from plants fruits is known to contain carvacrol and β caryophyllene, α-pinene, cembrene α-terpineol [8]. It is known that phenolic compounds, which constitute the widest class of phytochemical compounds contained in both plants, have various bioactive properties. For this reason, it is very important to determine these compounds responsible for the activity in plants qualitatively or quantitatively. When looking through the literature, it is clear that the liquid chromatography–mass spectrometry approach is one of the most widely used methods for analysing phenolic chemicals in medicinal plants [9, 10].

The literature review revealed that there were some studies on the content of R. coriaria fruit, but no studies on the chemical content of M. fruticosa spp. brachycalyx. In addition, the ADMET properties of these phenolic compounds were examined in detail for the first time. Therefore, the primary purpose of this study is to determine the phenolic contents of both plants by LC-QTOF-MS/MS. In addition, pharmacokinetic properties such as drug-like and toxicity of polyphenols analysed in plants were estimated for the first time using open-source software, pkCSM and SwissADME.

Methods

Plant materials and preparation of extracts

M. fruticosa spp. brachycalyx and R. coriaria were collected from Kahramanmaraş in Turkey. The identification of the plants has been established by Dr. I. Senkardes from Marmara University, Pharmacy Faculty. The Marmara University herbarium code MARE-19184 and MARE-19185 were assigned to the M. fruticosa spp. brachycalyx and R. coriaria, respectively.

The aerial parts of plant samples were dried at room temperature and pulverized with a mechanical grinder. The samples (50 g) were extracted with methanol (for 48 h × 2; 400 mL) at room temperature. After the solvents were filtered through Whatman No. 1 paper, the filtrate was evaporated to dryness by rotary evaporator (Heidolph Hei-Vap Presicion ML/G1) at 40 °C and 350 mbar. The raw extracts were kept at 4 °C in the refrigerator. The extracts (10 mg) were dissolved in 3 mL of methanol–water solution (2:1 v/v). The filtrates were then filtered using 0.2-m Millipore syringe filter, and then 10 μL samples were injected to LC system.

LC-ESI-Tandem MS analysis

The polyphenolic compounds of the extracts from aerial parts of plants were determined by LC-ESI-tandem MS technique. An Agilent 6530 was used to separate and analyse polyphenolic compounds. The chromatographic separation was performed on reverse phase Agilent Poroshell C18 (3 × 150 mm, 2.7 μm) analytical column. The column temperature was set to 30 °C. The separation was carried with a gradient binary mixture of solvent A (0.1% aqueous acetic acid) and solvent B (0.1% acetic acid acetonitrile) at a flow rate of 0.4 mL/min:0–5 min 10% B; 2–5 min 10–50%B;5–9 min 50% B; 9–10 min 50–90% B; 10–12 min 50–90% B; 12–18 min 10–90% B; 18–25 min 10–90% B; 25–25.01 min 90–10% B and stop time is 30.00 min. The full mass and fragmentation spectra of the polyphenols were generated by the electrospray ionization with quadrupole time-of-flight analyser in negative ion mode. The use of helium as collision gas and nitrogen was used as nebulizing gas.

ADMET prediction

The ADMET word is an abbreviation for absorption, distribution, metabolism, excretion and toxicity. ADMET studies, one of the cheminformatics computers programs, provide us with very important data on whether a chemical compound can be used as a medicine or not without conducting experimental studies. In this study, pkCSM, a free online web server (http://structure.bioc.cam.ac.uk/pkcsm) [11], was used to predict the pharmacological properties of compounds from R. coriaria and M. fruticosa spp. brachycalyx. Pharmaceutical values (lipophilicity, size, polarity, insolubility, insaturation and flexibility) of compounds and their radar charts were produced using SwissADME (http://www.swissadme.ch) [12]. The ADMET properties of 18 compounds which extracted from R. coriaria and M. fruticosa spp. brachycalyx were calculated by computer, and the partition coefficients (log P) of all compounds in this study were found.

Results

Chromatographic separation

LC-ESI-tandem MS analysis was used to qualitatively analyse of polyphenolic compounds in R. coriaria and M. fruticosa spp. brachycalyx (Figs. 1 and 2). Major polyphenolic compounds were analysed by comparing the molecular weights of these compounds and their fragments with the fragments provided by standard compounds and literature information. The analysis of gallic, syringic, quinic and caffeic acid, and rutin in the LC MS system was analysed with authentic standards by comparing their molecular weights, retention times and the mass/charge ratios of the fragment ions they gave after fragmentation. The other polyphenols were also tentatively identified by LC-QTOF-MS/MS with negative ionization because lacking reference standards. By comparing spectra of compounds that have previously recorded mass fragmentation patterns in full scan mode (MS) and MS/MS modes, the compounds' identities were confirmed in the literature (Additional file 1).

Fig. 1
figure 1

MS base peak and blank chromatogram of R. coriaria methanol extract

Full size image
Fig. 2
figure 2

MS base peak and blank chromatogram of M. fruticosa spp. brachycalyx methanol extract

Full size image

In this study, it was analysed that the methanol extract from aerial parts mainly contained phenolic acids, flavonoids, coumarin and tannins. The molecular weights and fragments of these compounds are shown in Tables 1 and 2.

Table 1 Identification of polyphenols in R. coriaria by LC-ESI-tandem MS data
Full size table
Table 2 Identification of polyphenols in M. fruticosa spp. brachycalyx by LC-ESI-tandem MS data
Full size table

When the results in Table 1 were evaluated, it was found that galloyl hexoside dimer was formed in at 8.67 min [13]. The formation of dimeric moieties is a common situation in LC–MS/MS studies and is associated with the geometric structures of molecules. Galloyl hexoside comes in at 3.96 min and eludes after the dimer formation (8.67 min). Galloyl hexoside and dihexoside [14] have been formed in at 4.00 and 8.72 min, respectively. The sugar portion was separated and gave a fragment of gallate at a molecular weight of 170 g/mol. MS/MS fragments show us an anthocyanin aglycone and sugar molecule in at 3.9293 min as malvidin-3,5-O-diglucoside [15]. It gives 655.19 a molecule that has broken a proton. By separating the glucose part from the molecule, a fragment ion of 331 molecular mass was formed, which is in the structure of malvidin. The molecular ion peak gives [M–H]− ion at m/z 169.0157 as a gallic acid, and fragment ion peak gives at m/z 125 ([gallic acid–H–CO2]–). Syringic acid-O-hexoside was formed in at 13.89 min [16]. The aglycone as well as the presence of the O-hexoside moiety was proven by the [M–H–162]− m/z 197 ion and by the m/z 182 [syringic acid–H–CH3]−, m/z 169 [syringic acid–2CH3]− and m/z 153 [syringic acid–H–CO2]− ions. Also, aglycon syringic acid was formed in at 17.39 min with [M–H]− at m/z 197. Based on the comparison of their MS2 spectra with reported literature, base peak which gave [M–H]− value at m/z 301.0310 was tentatively identified as tricetin [17].

When the results in Table 2 were evaluated, it was determined that molecular ion peak gives [M–H]− ion at m/z 683.2282 as a hexose polymer as the literature [18]. Caffeic acid showed [M–H]− value at m/z 179.0329. The fragment ions at m/z 161 and 135 by losses of a H2O molecule and a CO2 molecule were found, respectively. Rutin showed that [M–H]− value at m/z 609.2080 gave product ions at m/z 301 by losses rutinose. The molecular ion peak gives [M–H]− ion at m/z 353.8721 as a 5-O-caffeoylquinic acid as the literature [19]. The compound in at 16.69 min is thought to be a methoxy coumarin in the light of the relevant literature [21]. Deprotonated molecular ions at m/z 191 and fragment ions at m/z 173 [M–H–H2O]−, 127 [M–H–H2O–H2O–CO]− were generated during the peak 12.95th minute. As a result, it was characterized as quinic acid. The compound in at 18.25 min had the [M–H]− ion at m/z 461 which yielded the fragment ion at m/z 285 ([M–H]−—176, loss of one glucuronyl unit). As a result, kaempferol monoglucuronide was tentatively identified [22]. Malonyl-mono caffeoylquinic acid (malonyl-monocqa) was tentatively identified in at 18.44 min based on comparison of their MS2 spectra with reported literature [23].

In silico ADMET profiling of phenolic compounds from plant

The pharmacokinetics of compounds were predicted by the parameters of absorption, distribution, metabolism, excretion and toxicity as shown in Tables 3 and 4. The Caco-2 permeability values of all compounds were predicted to be low. Galloyl hexoside, syringic acid, gallic acid, tricetin, hexose polymer, 5-O-caffeoylquinic acid, caffeic acid and 6,8-dimethoxy-7-hydroxycoumarin were predicted to have high absorbed the intestinal absorption (human). All compounds were not predicted to be permeable skin. Galloyl hexoside, gallic acid, hexose polymer, caffeic acid and 6,8-dimethoxy-7-hydroxycoumarin were predicted to have not P-glycoprotein substrate. All of compounds were predicted to have not inhibitory effects. Galloyl hexoside, malvidin-3,5-diglucoside, galloyl-di-O-hexoside, tricetin, 5-O-caffeoylquinic acid, kaempferol-3-O-glucuronide and rutin had high the volume of distribution. All compounds were predicted to be poorly distributed to the blood–brain barrier, and they unable to penetrate the Central Nervous System. It is estimated that p450 enzymes, mostly found in the liver, do not metabolize the analysed compounds. It is also predicted that the same molecules are not substrates for this enzyme. It is estimated that no analyses compounds are a substrate for organic cation transport protein 2. Not all compounds analysed are predicted to have mutagenic and minnow toxicity effects. Syringic acid-O-hexoside was predicted to be hepatotoxicity effect. None of the compounds were predicted to have the potential to inhibit hERG I. However, malvidin-3,5-diglucoside, gallic acid, galloyl-di-O-hexoside and rutin have hERG II inhibitory effects. None of the compounds were predicted to have skin sensitization. When the log P values of all compounds are examined, it is estimated that the molecules except tricetin, caffeic acid, syringic acid and dimethoxy-hydroxycoumarin are more hydrophilic. It can be seen from the values in Tables 3 and 4 that these four molecules with more lipophilic properties are absorbed more easily too.

Table 3 The ADMET parameters of polyphenols from R. coriaria via pkcsm software
Full size table
Table 4 The ADMET parameters of polyphenols from M. fruticosa spp. brachycalyx via pkcsm software
Full size table

The relationship between drug permeability and lipophilicity in brain capillaries is shown in Fig. 3 [24]. As can be seen in the figure, it is estimated that the uptake of many substances into the brain will be limited due to their low octanol/water distribution coefficient. Among the compounds, caffeic and syringic acid and methoxy coumarin derivatives are predicted to be more hydrophobic, so they can pass the brain barrier more easily than the others.

Fig. 3
figure 3

Drug permeability in brain capillaries (log pc) as a function of partition coefficient (log p o/w)

Full size image

The biggest problem that can be encountered in oral administration of a drug molecule is bioavailability. The bioavailability radar device is used for oral bioavailability estimation, such as lipophilicity, size, polarity, solubility, flexibility, and saturation to determine drug affinity. In Fig. 4, some pharmacological properties of the compounds from M. fruticosa spp. brachycalyx and R. coriaria were predicted by SwissADME. Figure 4 presents an oral bioavailability radar field based on the lipophilicity, molecular size, polarity, TPSA and water solubility criteria of 18 compounds analysed in plants. As shown in the figure, syringic acid, hexose polymer and dimethoxy-hydroxycoumarin are predicted to have suitable physicochemical profiles for oral bioavailability. However, polar values of ten other molecules and saturation values of three ones indicate that they fall outside the desired range for bioavailability.

Fig. 4
figure 4

Bioavailability radar of the polyphenols based on physicochemical indices ideal for oral bioavailability

Full size image

Discussion

In one a previous study, phenolic content of water and ethanol extract of R. coriaria was analysed by LC–MS/MS and flavonoid, phenolic acid and galloyl compounds [25] were determined. In this study, we started our research with the knowledge that the biological activity results made with this plant are stronger in the methanol extract [4], and therefore, we conducted tandem MS analysis of the methanol extract of the plant. In contrast to the above-mentioned study, we analysed the sugary structures of malvidin, syringic acid and galloyl compounds, syringic acid aglycon and tricetin in R. coriaria. There is also a publication on the leaf of R. coriaria [26]. In this publication, the authors determined that this species contains galloyl compounds. As a result of our tandem MS studies, it was proven that this species contains galloyl compounds as well as some phenolic acids and tricetin. As shown in Table 1, a total of nine compounds were analysed in R. coriaria; seven compounds were analysed tentatively based on literature information, while the other two compounds were analysed based on spectral and chromatographic information given by standard substances.

In a study, leaf parts of M. fruticosa L. extracted with 80% methanol and the phytometabolites in the extract were widely analysed in untargeted mode by LC–MS/MS and a total of 215 compounds were identified tentatively [27]. In another study, phenolic compounds of M. graeca (L.) Benth. ex Rchb were analysed by HPLC [28]. In another study, alpha amylase and tyrosinase enzyme inhibition and antioxidant activity of different extracts of M. nervosa were studied and the phenolic compounds in its content were analysed by LC–MS/MS [29]. As can be understood from the studies in the literature, there are very limited studies on the phenolic compound analysis on Micromeria species. Phenolic compounds of M. fruticosa spp. brachycalyx were analysed by us for the first time in tandem MS. As can be seen in Table 2, a total of nine compounds were analysed in the plant; six compounds were analysed tentatively in the light of literature information, while the other three compounds were analysed in the light of spectral and chromatographic information given by standard substances.

In a study, methanol extracts of Syzygium cumini (black plum) seeds and Allium cepa (onion) peels were analysed by GC MS and it was found that there were 20 phytocomponents in the plant, and four of these compounds had drug-like pharmacokinetic properties [30]. There are publications in the literature on the pharmacokinetic properties of the natural compounds using the in silico ADMET process [31,32,33,34]. Therefore, evaluation of the in silico ADME properties of biologically active compounds before in vivo and clinical studies in drug or drug raw material design and development based on a plant would be a great time saving and a good data source. The ADMET predictions of phenolic compounds show that some of these compounds are easily absorbed and do not have toxic effects, suggesting that this species can be used as a natural medicinal and nutritional source in the future after detailed analyses. The hydrophilic or hydrophobic properties of the compounds analysed in the chromatography system are shown with octanol–water partition coefficients. These partition coefficients are related to the distribution of compounds in the body. The high coefficient indicates that the molecule is hydrophobic and dispersed into the hydrophobic areas of the cell, while the low coefficient indicates that the molecule is hydrophilic and can be dispersed into aqueous areas such as blood serum [35].

While only a small fraction of lipophilic substances can cross the blood–brain barrier, most polar and hydrophilic substances cannot be transported to the brain without a special delivery system [36]. Among the compounds, caffeic and syringic acid and methoxy coumarin derivatives are predicted to be more hydrophobic, so they can pass the brain barrier more easily. Besides, the bioactive compound of the plants was predicted to be drug-like and non-mutagenic with in silico ADMET. These data support data for future analysis that can lead to their therapeutic use of the plants.

Conclusions

As a result, in this study, the phytochemical contents of these two species, which are used by the public for food and medical purposes, were analysed with tandem MS. According to the findings obtained, totally eighteen phenolic compounds, which of nine were analysed in R. coriaria and the other nine in the M. fruticosa spp. brachycalyx, were tentatively identified based on MS determination and fragmentation pattern. As a result of chromatographic analysis, it was determined that R. coriaria is rich in tannins and M. fruticosa spp. brachycalyx is rich in phenolic acid and flavonoid. Besides, the ADMET properties of these phenolic compounds contained in these two species were estimated in silico. The fact that some compounds in both plants are well absorbed, have drug-like properties and do not have toxic effects supports the traditional use of these two species.