The used Salvia verbenaca over ground parts were harvested at Ain Aouda in the Rabat region, Morocco. Species identification was graciously done by Pr. Fennane from the Scientific Institute, Mohammed V University in Rabat. A voucher sample of the species plant was placed in the National Herbarium of this institution. The freshly collected parts were dried under shade, crushed and stored at –20 °C until use.

The used analytical grade extraction solvents and HPLC grade chromatographic solvents were bought from Sigma-Aldrich (Steinheim, Germany).

The dried Salvia verbenaca plant material was powdered. An amount of 200 g of the obtained powder was soaked under stirring for one day in a mixture of EtOH/H${}_2$O (70/30). For a thorough extraction, this procedure was twicely repeated. The organic solvent of the obtained extract was evaporated and the aqueous mixture was extracted with hexane and AcOEt solvents. The two obtained organic mixtures were concentrated through evaporation and the obtained extracts were suspended in the initial HPLC eluent before injection. The extraction of phenolic compounds through maceration by hydro-alcoholic mixture (either ethanol or methanol) is a conventional extraction technique widely known and has been used in several published works. After evaporation of the alcoholic solvent, the further fractionation of the aqueous solution with other solvents with different polarities beginning with the non-polar one such as hexane and followed by others such as ethyl acetate are also known and reported.

The total amounts of phenolic compounds of the instigated extracts were found using the Folin–Ciocalteu reagent with gallic acid as standard [15]. Results were given as mg of gallic acid equivalents per gram extract (mg GAE/g extract).

Flavonoids amount was determined by spectrophotometry using the aluminum chloride reagent with quercetin as standard [15]. Results were then given as mg of quercetin equivalents per gram extract (mg QE/g extract).

The phytochemical composition of the extracted samples was explored through ultra high performance liquid chromatography coupled to mass spectrometry. Analysis were carried out on an Agilent 1290 Infinity LC apparatus set (Agilent Technologies, Santa Clara, CA, USA) equipped with a column chamber, an auto-sampler and a binary pump. An Inertsil C18-4 column (3 $\mu$m, 150 $\times$ 2.1 mm i.d) kept at 30 °C was employed as separating stationary phase.

The LC separation was achieved using a binary gradient of solvent A (MeOH 20% in water containing 5 mM formate and 0.1% formic acid and solvent B (MeOH 90% in water containing 5 mM formate and 0.1% formic acid. Separation mobile phase began with 10% B and then incremented during 23 minutes to 95%. This was kept for 5 minutes and then followed by a washing and conditioning of the column. The mobile phase flow rate was 0.4 mL/min.

The used separation chromatographic system was coupled to an Agilent 6530 Quadrupole Time of Flight Mass Spectrometer using Agilent Mass Hunter Software version B.05.01 (B5125) (Agilent Technologies, Santa Clara, CA, USA). The MS detector was equipped with an ESI interface using the following conditions: capillary voltage, 3.5 kV; nebuliser, 45 psig; drying gas (nitrogen) flow rate, 8.0 L/min; sheath gas flow rate, 11.0 L/min; gas temperature, 375 °C; fragmentor, 175 V; skimmer voltage, 65 V; OCT 1 RF Vpp, 750 V. Spectra were recorded from m/z 100 to 3000.

In addition to the above analysis technique, the investigated Salvia verbenaca extracts were also explored using an UHPLC-DAD-ESI-Q-ToF/MS platform consisting of a Waters ACQUITY LC unit coupled to both DAD (Diode-Array Detector) and MS (Mass Spectrometry) detectors. Separations were performed on a C18 reversed phase column (Acquity UPLC BEH 1.7 $\mu$m, 100 $\times$ 1) at 35 °C. Elution conditions were as follow: phase A (1% acetic acid in water containing acetonitrile 10%), phase B (1% acetic acid in acetonitrile containing water 20%), isocratic elution 15% B for 10 min, from 15 to 30% B in 2 min, from 30 to 50% B in 13 min, from 50 to 80% B in 5 min, from 80 to 90% B in 5 min, from 90 to 95% B in 3 min followed by washing and re-equilibrating of the column. Injection volume was 5 $\mu$L and flow rate was 0.08 mL/min.

The UPLC system was connected to a quadrupole time of flight (QToF SYNAPT-G2 HDMS) mass spectrometer with an electrospray ionization (ESI) source operating in positive and negative modes. Nitrogen was used as desolvation and cone gas at a flow rate of 20 $\mu$L/min. Source and desolvation temperatures were respectively 120 °C and 450 °C and capillary voltage was set to 2 kV. MS data acquisitions (from m/z 50 to 1500) were performed in continuous mode using a MassLynxTM software Version 4.1 (Waters Inc., Milford, MA, USA). DAD spectra were recorded from 190 to 500 nm on PDA (Photo Diode Array) detector (Waters Acquity, UPLC LG 500 nm).

The antioxidant activity of S. verbenaca extracts was explored using DPPH${}^{\mathrm{•}}$ and ABTS assays according to previously reported evaluations methods [16, 17]. The DPPH${}^{\mathrm{•}}$ radical has an intense violet color with an absorption maximum at 715 nm which became yellow in presence of an antioxidant. 50 $\mu$L of sample extract or standard (TROLOX) with varying concentrations were added to 5 mL of the same volume of DPPH${}^{\mathrm{•}}$ methanolic solution (0.004%). TROLOX and samples’ extracts were used at concentrations of 0.01, 0.02, 0.05, 0.15 and 0.2 mg/mL. The samples were incubated for 30 min in the dark at room temperature. A decrease in absorbance was measured at 517 nm. Methanol and TROLOX were used as negative and positive controls respectively. The DPPH${}^{\mathrm{•}}$ scavenging activity was determined in percentage and the half maximal inhibitory concentrations (IC${}_{50}$) were graphically found.

The antioxidant activity of Salvia verbenaca extracts were also determined using the trolox equivalent antioxidant capacity test with the ABTS decolorization assay.

In the occurrence of an antioxidant, the transformation of the ABTS radical to the free form is accompanied with a diminution of the absorbance at 734 nm. A volume (2.5 mL) of the ABTS solution was blended with 200 $\mu$L of each extract and 300 $\mu$L of methanol. After 8 min in the dark at room temperature, the absorbance was assessed at 734 nm. Methanol and TROLOX were respectively used as negative and positive controls. The ABTS scavenging activity was determined in percentage and the half maximal inhibitory concentrations (IC${}_{50}$) were graphically determined.

A discussion on the relation between the chemical composition of the hexane and ethyl acetate extracts and their observed activities will be presented. In addition to the antioxidant activity, the discussion will concern also the cytotoxicity [11], the acute and subchronic oral and dermal toxicity [12] in addition to the effects of the extracts for healing of second-degree burns wounds [10]. We have previously performed and reported these activities [10, 11, 12] and the relation of the observed results with the phytochemical composition herein presented will be discussed in this work.

The obtained results were given as means $\pm$ standard errors. One-way analysis of variance (ANOVA) and tuckey’s multiple comparison test were used to compare group means. The differences were considered significant, when p 0.05.

Salvia verbenaca extracts (hexane, AcOEt) were analyzed by UHPLC coupled to both DAD and MS detectors. Structural characterization of products for which standards were accessible was performed by injecting the latter in the same conditions and comparison with the obtained chromatographic and spectroscopic UV and MS data. Structural elucidation of the others was initiated using the obtained UV which permit to determine the phytochemical family since each one exhibits a characteristic UV–vis spectrum [30]. This was followed by the interpretation of the obtained MS spectra and the further observed fragmentations giving useful information on each compound molecular weight and also on the presence and the nature of substituents. The compound molecular mass was assigned from MS spectra obtained through low collision energy, since in such conditions the protonated/deprotonated molecules are easily obtained. Mass spectra with several fragment signals were recorded through high collision energy experiments allowing the occurrence of fragmentations and providing useful structural information related to the aglycon and glycosides moieties. The compounds evidenced in Salvia verbenaca extracts were split into phenolic acids, phenolic diterpenes and flavonoids which structural elucidation is discussed below.

3.1.1 Phenolic Acids and Derivatives

Nine polyphenols in the category of phenolic acids were detected in the explored Salvia verbenaca samples presenting UV spectral properties of hydroxycinnamic acid adducts. Six of them, namely, caffeic acid 1, p-coumaric acid 2, ferulic acid 3, 3,4-dimethyl-caffeic 4, chlorogenic acid 5 and rosmarinic acid 9 were identified by comparing their characteristic UV and MS characteristic data with those of previously reported data [27, 31, 32, 33].

Through UHPLC-QToF-MS in the negative ionization mode, compound 1 gave a signal at m/z 179 (C${}_9$H${}_7$O${}_4$) corresponding to the deprotonated molecular ion [M-H]${}^{-}$ of caffeic acid. This was confirmed by the presence of another signal at m/z 135 corresponding to the loss of carboxylic acid moiety to form the product ion [M-H-CO${}_2$]${}^{-}$. This cleavage agrees with the known cleavage pattern of hydroxycinnamic acids as their deprotonated molecular ions [M-H]${}^{-}$ are generally characterized by the loss of a CO${}_2$ (44 Da) from the carboxylic acid group giving the corresponding [M-H-CO${}_2$]${}^{-}$ ion [31, 32, 33].

In a comparable manner, the mass characteristics of compound 2 exhibited the deprotonated molecule [M-H]${}^{-}$ ion signal at m/z 163. The main produced fragments signals were located at m/z 145 and m/z 119 corresponding to loss of water and carbon dioxide molecules respectively from the precursor ion in agreement with the proposed p-coumaric acid molecule.

Ferulic acid 3 and 3,4-dimethoxycaffeic acid 4 have also been detected. Their characterizations were identified by matching their spectral results with those of literature data [31]. Compound 3 molecular ion signal was located at m/z 193 corresponding to the ion [M-H]${}^{-}$. Other fragment signal was located at m/z 149 corresponding to the loss of carboxylic acid group to form the [M-H-CO${}_2$]${}^{-}$ ion as indicated above. Another signal was observed at m/z 179 corresponding to the loss of methyl confirming the structure proposed for compound 3.

In addition to the phenolic acid indicated above, chlorogenic acid 5 was also detected giving deprotonated molecule [M-H]${}^{-}$ ion signal at m/z 353 in its MS spectrum recorded in the negative ion mode. The major produced fragment ions obtained in negative mode was located at m/z 191 corresponding to the deprotonated quinic acid moiety in agreement with previously reported data of the proposed chlorogenic acid structure [34, 35].

Besides these five phenolic acids, three additional phytochemicals 6–8 presenting spectral characteristics of phenolic acid derivatives were also evidenced in the explored extracts. This was deduced from their ultraviolet spectra which were similar of hydroxycinnamic acid derivatives [30]. Compound 6 which MS spectrum showed a fragment ion signal at m/z 181 was concluded to be caffeic acid derivative [31, 36, 37].

The mass spectrum of compound 9 in the negative ion mode (Fig. 2), showed a signal at m/z 359 corresponding to the deprotonated molecular ion [M-H]${}^{-}$ of rosmarinic acid as confirmed by its characteristic UV spectrum. The fragment ion signals observed at m/z 197, 179 and 161 and the proposed fragmentation pathways represented in Fig. 2 were all consistent with the proposed rosmarinic acid structure [27].

Fig. 2.

Mass spectrum of compound 9 with the proposed corresponding cleavages.

The evidence of phenolic acids in Salvia genus has been formerly described in the literature. Caffeic acid, p-coumaric acid, ferulic acid and rosmarinic acid in addition to several derivatives of phenolic acids have been previously reported [18, 19, 20, 21, 22, 23, 24]. Our results corroborate the presence of such adducts in Salvia verbenaca. Rosmarinic acid has also been previously reported in extracts of Salvia verbenaca [9]. Its detection in the present work confirm thus its occurrence in the Moroccan sample. The present work confirmed also the occurrence of rosmarinic acid in vascular plants particularly in the Lamiaceae family [38, 39]. Plant extracts containing rosmarinic acid have been reported to possess dietary, cosmetic, pharmaceutical and beverage production applications [40]. This phytochemical phenolic acid derivative has also been reported to have various biological properties such as antioxidant antiviral, anti-inflammatory, antimutagenic, antibacterial and above all antioxidant [38, 39, 41]. The antioxidant activity of rosmarinic acid is due to the presence of four hydroxyl groups in its molecule [42]. An anti-HIV property of rosmarinic acid has also been reported in the literature [43]. Its presence in the studied Salvia verbenaca could justify the observed biological activities of the investigated plant extracts [11].

Following the identification of the phytochemicals occurring in the instigated Salvia verbenaca samples, the relative percentages of these compounds have been determined using UHPLC chromatography connected to diode array and ESI-ToF-MS detectors. An example of the obtained profile is presented in Fig. 3 and the obtained results for the studied extracts are gathered in Table 2. The relative quantitative analysis of the detected phytochemicals has been made from the area % calculation procedure which gives the area of each chromatographic peak as a percentage of the total peaks area. This method assumes that the detected phytochemicals answer in a similar manner to the detector. Even if these criteria are may be not guaranteed, this method gives at least a useful approximation of the relative amounts of the evidenced phytochemicals.

Fig. 3.

UHPLC chromatographic profile of Salvia verbenaca (ethyl acetate extract).

The obtained results showed that the Salvia verbenaca hexane extract contained six compounds representing a percentage of 92.24% of the total peaks area. This extract was dominated by genkwanin as major compound (39.25%), followed by phenolic acid derivatives (38.54%).

The quantitative analysis of the Salvia verbenaca ethyl acetate extract showed the presence of 16 phytochemicals representing a percentage of 98.74% of the total peaks area (Table 2). The major compounds were luteolin (23.71%) and 3,4-dimethyl caffeic acid (23.71%). These were followed by p-coumaric acid, rosmarinic acid and naringenin with similar percentage (8.28%).

The antioxidant properties of the Salvia verbenaca samples were assessed using DPPH${}^{\mathrm{•}}$ and ABTS procedures. The use of different incremented concentrations of each sample exhibited a dose-dependent antioxidant activity and allowed the IC${}_{50}$ to be determined and to be compared to that of the used positive control.

The obtained results presented in Table 3 revealed higher antioxidant activity on DPPH${}^{\mathrm{•}}$ for the two explored Salvia verbenaca extracts with results of 0.0925 $\pm$ 0.0003 and 0.0744 $\pm$ 0.0001 mg/mL for the hexane and ethyl acetate extracts respectively. These were close to the results obtained for the used positive controm sample (0.0621 $\pm$ 0.0001 mg/mL).

In term ABTS model assay, the ethyl acetate sample exhibited the highest potent activity with an IC${}_{50}$ of 0.1338 $\pm$ 0.0003 mg/mL and hexane extract with an IC${}_{50}$ of 0.1867 $\pm$ 0.0002 mg/mL which was moderately superior than that exhibited with Trolox (0.0604 $\pm$ 0.0002 mg/mL). The difference observed between hexane and ethyl acetate extracts and Trolox was statistically significant (p-value 0.0001) for the two tested assays DPPH${}^{\mathrm{•}}$ and ABTS.

The difference obtained between the two tests (DPPH${}^{\mathrm{•}}$ and ABTS) allowed us to argue that the use of each test should meet specific purposes. The fluctuations in the results obtained by the methods used can be explained by the chemical composition of the extracts tested. The identified phytochemicals could be behind the antioxidant activity of Salvia verbenaca instigated using DPPH${}^{\mathrm{•}}$ and ABTS procedures.

The study of the relation between the observed antioxidant activities of the explored extracts and the polyphenols and flavonoids contents showed that the importance of such metabolites in enhancing the antioxidant activities. Thus, the ethyl acetate extract which was the richer extracts either in total polyphenols contents and in total flavonoid contents showed the higher antioxidant activity. This indicated the importance of such phytochemicals in the antioxidant activity observed with the two used methods.

The medicinal properties of plant are obviously related to their composition. The biological activities of plant extracts are evidently linked and dependent of the extract’s chemical content.

In our previous research investigation on Salvia verbenaca, some biological activities of its extracts have been explored. Thus, the cytotoxicity and the ability of the plant to accelerate second degree burn healing in addition to its acute and subchronic oral and dermal toxicity of the hexane and ethyl acetate extracts have been investigated [10, 11, 12]. Among the obtained results, and in relation with burn healing, the ethyl acetate extract was as efficient as silver sulfadiazine which constitutes the basic principle of the most widely used topical treatment against injury [10]. The phytochemical investigation conducted in the present work showed that the ethyl extract contained various phenolic compounds such as flavonoids and phenolic acids which are recognized and reported to have different biological activities. Our results indicated that the flavone luteolin was the major compound of the ethyl acetate extract and the total percentage of the identified flavonoids was 42.55%. This demonstrated the role played by flavonoids as wound healing dressings and justify the use of flavonoids in different formulation for the treatment against burns. It is obviously evident that the multiple identified flavonoids in the extract could act in a synergistic effect or combined effect and the observed effect could not be obviously due to this or that compound even if it constitutes the major one. The positive role played by flavonoids in wound healing could be explained by the fact that they possess all the activity involved against burn such inflammation, proliferation and re-epithelialization. In addition to flavonoids, the ethyl acetate was also rich in phenolic acids which are also recognized for their properties in wound management [56]. The global observed action of the extracts involved obviously the total phytochemicals which could act in synergetic manner.

In addition to the ability of Salvia verbenaca extracts to accelerate the healing process, its cytotoxic effect was also explored against two cancer cell lines, Vero and RD, using the colorimetric MTT assay [11]. The obtained results showed that the effect observed for the ethyl acetate extract was higher than that of the hexane. This is also attributed to richness of this extract in phenolic acids (caffeic acid, p-coumaric acid, ferulic acid, chlorogenic acid and rosmarinic acid) and flavonoids (quercetin 3-O-glucoside, luteolin, apigenin and naringenin) constituting more than 90% of the total phytochemical composition. This confirmed the fact that polyphenols in general and flavonoids in particular have a range of biological activities including antioxidant, anti-inflammatory, and anti-tumorigenic properties. Flavonoids have been reported to be active at different stages of cancer development inducing cellular cytotoxicity [57]. Reported data suggest that flavonols may provide protection against oxidatively induced cellular damage in vivo [57]. Flavones such as luteolin and apigenin have also been reported to have strong anticancer [57]. Our findings are in agreement with these statements as these flavonoids were detected as both apigenin and luteolin were detected in the ethyl acetate extract.

Concerning the antioxidant activity, examination of the obtained results gathered in Table 3, showed a relatively higher effect of the ethyl acetate extract compared to the hexane one. As indicated above, this was quite obvious given the relatively higher total polyphenols and total flavonoids contents of the two extracts. The obtained results agreed also with the qualitative phytochemical composition of the two extracts. Indeed, Table 2 showed the presence of flavonoids pertaining to flavones and flavonols such as apigenin, luteolin and quercetin derivatives which are recognized as potent antioxidant agents. The structures of these compounds contained a double bond between the carbons 2 and 3, conjugated with a 4 oxo group in the ring C and the presence of additional hydroxyl groups at the C-3, C-3’, C-4’ positions. These structural characteristics are known to improve the radical scavenging capacity of flavonoids and to increase their antioxidant activity [58]. The presence of other compounds such as rosmarinic acid which was detected in the ethyl acetate extract could also enhance its antioxidant activity.