Differences in bioactivity of three endemic Nepeta species arising from main terpenoid and phenolic constituents

Methanol extracts of three endemic Nepeta species were analyzed for their main secondary metabolites, terpenes and phenolics, and further investigated for antioxidant capacity and embryonic toxicity in zebrafish. UHPLC/DAD/(±) HESI-MS/MS analysis showed that the dominant compound in N. rtanjensis was trans,cis-nepetalactone, the cis,trans isomer of this monoterpene lactone was dominant in N. sibirica, while nepetalactone was detected only in traces in N. nervosa. In all investigated species, rosmarinic acid was the dominant phenolic compound, while other identified phenolic acids (chlorogenic, neochlorogenic and caffeic) were present in considerably lower amounts. ABTS and DPPH assays showed that the methanol extracts of N. rtanjensis, N. sibirica and especially N. nervosa possessed strong antioxidant activities, with the FRAP assay revealing high ferric-reducing abilities for all three tested species. Such a strong antioxidant potential, especially as manifested in the DPPH and FRAP assays, can be attributed to phenolic acids, and in the first place to rosmarinic acid. Increased lethality of zebrafish embryos in any of the treatments was not observed, but several toxic effects on embryonic development were recorded, such as pericardial and yolk sac edema. As in other Nepeta species, the three studied endemic species possessed a great potential for food conservation or as medicinal supplements if applied in optimized concentrations; however, alternative sources of plant material (e.g. field cultivation) should be established bearing in mind their vulnerability in nature.


INTRODUCTION
Oxidative stress can lead to the development of disorders and diseases in humans.Scientists have become interested in natural sources that can provide active compounds to prevent or reduce oxidative stress-induced damage in cells [1].The constantly produced free radicals can damage DNA, proteins and lipids, and the resulting cell damage can lead to accelerated ageing and many oxidative stress-related chronic conditions, including cancer, cardiovascular diseases, diabetes, obesity, joint diseases and Alzheimer's disease [2].In recent years, there is growing interest in the use of natural compounds with antioxidant properties derived from aromatic herbs for improving human health.Previous studies showed that aromatic plants with high contents of phenolic compounds are an important source of natural antioxidants [3,4].Numerous pharmacological effects have been attributed to Nepeta species, including antiinflammatory [5], analgesic [6], cytotoxic [7], antiviral [8] and antimicrobial activities [9].Furthermore, phytotoxic activities, attractant effects on cats and aphids and repellent activity against a variety of insects and other pests were reported for Nepeta species [10][11][12].
The pharmacological activities of Nepeta species are usually ascribed to numerous secondary metabolites, in the first place iridoid monoterpenes -nepetalactones, and phenolic compounds.Previous phytochemical studies on species belonging to the genus Nepeta revealed that monoterpenes, diterpenes, triterpenes and phenolics (flavonoids, phenolic acids and their derivatives) were the major secondary metabolites.Among monoterpenoids, the most frequently determined are nepetalactones [13], which appear as the main constituents in most species.Nepetalactone is actually a cyclopentanoid monoterpene with two fused rings, a cyclopentane and a lactone.There are eight stereoisomers, four diastereomers and their corresponding enantiomers.With some exceptions, the (7S)-diastereomers are found in natural sources [14].Nepetalactone isomers show different biological activities [15].
Many Nepeta species were tested for antioxidant activity and it has been shown that phenolic compounds, in the first place phenolic acids, played an important role in the antioxidant assays [3,5,16].Tepe et al. [3] have demonstrated a positive correlation between antioxidant activity and the amount of total phenolic compounds in Nepeta flavida.
The Nepeta species in this study included N. rtanjensis Diklić & Milojević, an endemic and critically endangered perennial in Serbia [17], N. sibirica L., which is endemic to central Asia, Mongolia and southern Siberia [18], and N. nervosa Royle ex Bentham, a plant species endemic to Kashmir [19].To the best of our knowledge, these species have not previously been described for their antioxidant activities and zebrafish embryo toxicity.Because of the specific phytochemical composition, i.e. qualitative composition of nepetalactone isomers, they represent interesting objects for investigating the biological activities of their extracts.As previously reported, N. rtanjensis contains much higher amounts of trans,cis-nepetalactone than cis,trans-nepetalactone [13,17], while N. sibirica possesses high amounts of cis,trans-nepetalactone [13,18].N. nervosa contains nepetalactones only and in trace amounts [13,20].Considering that these species are endemic and extremely endangered, as in the case of N. rtanjensis, plant material for the experiments was obtained by micropropagation.

Plant material and in vitro culture establishment
Seeds of three Nepeta species − N. rtanjensis, N. sibirica and N. nervosa, were used to establish the in vitro cultures.Seeds of N. sibirica and N. nervosa were provided by Grugapark Essen (Germany), while those of N. rtanjensis were collected in August 2004 from greenhouse-grown plants.After 10 min of surface sterilization in a 20% solution of commercial bleach, the seeds were rinsed 5 times with sterile distilled water and transferred onto 9-cm Petri dishes, each containing 20 mL solid ½ MS [21] culture medium, supplemented with 20 g L -1 sucrose, 7 g L -1 agar (Torlak, Serbia) and 100 mg L -1 myo-inositol (Merck, Germany).Seedlings were transferred into 370-mL glass jars with transparent polycarbonate covers.Each jar contained 100 mL of ½ MS culture medium.Plants were grown in a growth chamber under long day conditions (16/8 h light/dark cycle), at 25±2°C.A photon flux rate of 32.5 μmol m -² s -¹ was provided to the plant cultures by white fluorescent tubes.Plant material for the analyses and assays was obtained by micropropagation, using nodal segments with one pair of leaves as explants.Following several cycles of subculturing and micropropagation, sufficient plant material was obtained for further analyses.

Preparation of plant methanol extracts
Shoots of four-week-old micropropagated plants were used for the analysis of secondary metabolites.Plant material was air-dried in a thin layer at room temperature until constant weight was attained.The plants were weighed (dry wt) and powdered in liquid nitrogen (LN) until further use.Each sample (0.2 g, dried and powdered) was extracted with 10 mL of methanol (AppliChem GmbH, Germany) in an ultrasonic bath for 15 min.After centrifugation for 20 min at 10000 g, the supernatants were filtered through 0.2-µm cellulose filters (Agilent Technologies, Santa Clara, CA) and stored at 4°C until use.
A TSQ Quantum Access Max triple-quadrupole mass spectrometer equipped with a heated electrospray ionization (HESI) source was used as previously described [13], with slight modifications.The vaporizer temperature was kept at 350°C, and the ion source settings were as follows: spray voltage was 4000 V, sheet gas (N 2 ) pressure was 50 AU, ion sweep gas pressure was 0 AU and the auxiliary gas (N 2 ) pressure was 20 AU, capillary temperature was 270°C, skimmer offset was 0 V.The mass spectrometry data were initially acquired in both the positive and negative modes, in the m/z range from 100-1000.Full scanning (FS) and product ion scanning (PIS) mass spectrometric scanning modes were conducted for the qualitative analysis of targeted compounds in samples.Collision-induced fragmentation experiments were performed using argon as the collision gas, with the collision energy was set to 30 eV for both phenolic acids and nepetalactones.The selected reaction monitoring (SRM) experiment for quantitative analysis was performed using two MS2 fragments for each compound, which were previously defined as dominant in the PIS experiments.
Phenolic acids were identified by direct comparisons with commercial standards and literature data.Standards of caffeic acid, chlorogenic acid and rosmarinic acid were purchased from Sigma-Aldrich (Steinheim, Germany).Essential oils of N. cataria, containing 90% of cis,trans-nepetalactone and N. rtanjensis containing 72% of trans,cis-nepetalactone and 16% of cis,trans-nepetalactone were used as standards for nepetalactone diastereomer identification and quantification, as previously described [13].Stock standard solutions of phenolics were prepared by dissolving 1 mg of a pure compound in 1 mL of methanol.The stock solution was further diluted with methanol to obtain a concentration of 100 μg mL -1 , and subsequent calibration levels were obtained by diluting the solution with methanol.Working standard solutions were prepared at 15 levels, in a concentration range from 100 μg mL -1 to 0.001 μg mL -1 .Regressions were calculated for each of the calibration curves and they all showed good linearity with correlation coefficients between r=0.990 and 0.999, p<0.001.The total amount of each compound in a sample was evaluated by calculating the peak areas and is expressed as µg 100 mg -1 dry wt.
The solvents for HPLC-MS analyses (acetonitrile and formic acid) were LC-MS grade, obtained from Fisher Scientific (Loughborough, UK).The methanol used for plant extraction (HPLC grade) was purchased from AppliChem (Cheshire, CT, USA).Ultrapure water was generated by deionization (Millipore, Billerica, USA).

Determination of total phenolics and total flavonoids
The total phenolics concentration (TPC) was measured by the Folin-Ciocalteu assay [22], while the total flavonoid concentration (TFC) of the samples was deter-mined according to Karadeniz et al. [23], as previously described by Šiler et al. [24].The TPC values were calculated from a standard calibration curve based on gallic acid (Sigma Aldrich, Germany), and expressed as mmol of gallic acid equivalents (GAE) per 1 g of dry wt (mmol GAE g -1 dry wt).The TFC in each extract was calculated from the standard curve based on rutin hydrate (Sigma Aldrich, Germany).The results are expressed as mmol of rutin hydrate equivalents (RE) per 1 g of dry wt (mmol RE g -1 dry wt).

Antioxidant activity
Preparation of N. rtanjensis and N. cataria essential oil and phenolic acid standard solutions For antioxidative activity, the essential oil of N. rtanjensis (which contained 72% of trans,cis-nepetalactone and 16% of cis,trans-nepetalactone) and the essential oil of N. cataria (composed of 77% cis,trans-nepetalactone) were diluted in methanol (AppliChem, Cheshire, SAD) to obtain a final nepetalactone concentration of 4%.Preparation of essential oils and GC/MS analysis were described previously by Dmitrović et al. [11].N. rtanjensis EO was isolated from air-dried aerial parts by hydrodistillation for 2 h in a Clevengertype apparatus.The essential oil of N. cataria was a gift from Dr. M. Birkett (Biological Chemistry and Crop Protection Department, Rothamsted Research, Harpenden, UK), and is prepared by steam distillation using cyclohexane as the cosolvent.Working solutions of chlorogenic and rosmarinic acids were prepared by adjusting their concentrations to 1 mM.

ABTS radical scavenging activity
The ABTS radical cation decolorization assay was carried out as previously described by Re et al. [25], with some modifications [24].ABTS •+ radical scavenging activity (%) was calculated using the formula: where A sample is the absorbance of the solution when the extract has been added, and A control is the absorbance of the ABTS •+ solution without added extract.A methanol solution of Trolox was used for construction of the calibration curve and the results are expressed as reducing activity in mmol Trolox equivalents per gram of dry sample wt (mmol TE per g -1 dry wt).All analyses were performed in triplicate.

DPPH radical scavenging activity
DPPH assay was performed as previously described by Brand-Williams et al. [26], with slight modifications reported earlier [24].The radical scavenging activity was calculated as the percentage of DPPH • discoloration using the equation: where A sample represents the absorption of the solution when extract has been added, and A control is the absorbance of the DPPH • solution without the addition of extract.Trolox was used as the reference compound.All the analyses were run in triplicate and mean values were calculated.Results are expressed as mmol TE per g -1 .

Ferric reducing/antioxidant power (FRAP)
The ferric reducing/antioxidant power (FRAP) assay was carried out according to the procedure described by Benzie and Strain [27], with some modifications described by Šiler et al. [24].The methanol solution of Trolox was used for construction of the calibration curve and the results are expressed as Trolox equivalents antioxidant capacity (mmol TE per g -1 dry wt).

Zebrafish husbandry, eggs collection, and embryo exposure
Adult, wild-type Tubingen zebrafish (Danio rerio) were used in all experiments and were maintained in fish medium (2 mM CaCl 2 , 0.5 mM MgSO 4 , 0.7 mM NaHCO 3 , 0.07 mM KCl).Fish were maintained at a temperature of 26±1°C under continuous water aeration and filtering, and under artificial light with a 12 h dark/light cycle.Female and male individuals were always kept apart and regularly fed twice a day with commercial dry-flake food (TetraMin™ flakes; Tetra Melle, Germany) supplemented with Artemia nauplii.The day before spawning, males and females at a ratio of 2:1 were placed in a breeding tank before the onset of darkness and left undisturbed overnight.At the onset of light, the separators were removed from the breeding tanks.The eggs were collected after 30 min, rinsed twice from debris using fresh fish medium and transferred into Petri dishes with the fish medium.
Within 1.5 h post-fertilization (hpf), fertilized eggs were selected under a binocular stereomicroscope (PXS-VI, Optica) and transferred into 24-well plates, 10 embryos per well, in a fish medium volume of 2 mL.The extracts were diluted 100x and 1000x in fish medium and added to the wells.The fish medium, 1% methanol and 0.1% methanol were used as controls.The embryos were then incubated at 28±0.5°C.Each experiment was performed in duplicate or triplicate and repeated three times from independent breedings.Fish embryo toxicity assay was performed if fertilization rates were ≥90%.An assay was considered valid if the overall survival of embryos in negative controls was ≥90% until hatching.The apical endpoints for the assessment of embryotoxicity (mortality) and teratogenicity were evaluated at 24, 48, 72 and 96 hpf using an inverted microscope (CKX41; Olympus, Tokyo, Japan).
All embryos were inspected for morphological characteristics at different developmental stages as described by Kimmel [28].Lethal and teratogenic effects were recorded according to OECD (236) guide-lines for testing of chemicals.Teratogenic effects were recorded if the fingerprint endpoint was observed in at least 50% of all embryos showing teratogenic effects in the test group and if a concentration-response was present.

225
a -numbers of peaks according to Fig. 1. S -confirmed by standard; l -confirmed by references.Masses used in the selected reaction monitoring experiments for the quantification of compounds are presented in bold.
nolic acids was also confirmed by their UHPLC-DAD spectra.UHPLC-DAD chromatograms of N. rtanjensis, N. sibirica and N. nervosa methanol extracts are presented in Fig. 1, while the retention times and λ max values of the corresponding compounds are shown in Table 1.

Total phenolic and total flavonoid contents in the methanol extracts of N. rtanjensis, N. sibirica and N. nervosa
Table 3 shows the total phenolic and total flavonoid contents in the methanol extracts of the three studied Nepeta species.The methanol extract of N. rtanjensis was characterized by a higher content of total phenolics (~18 mmol GAE g -1 dry wt) than N. nervosa (~14 mmol GAE g -1 dry wt) and N. sibirica (~13 mmol GAE g -1 dry wt).The content of total flavonoids followed the same trend.The highest amount of total flavonoids was recorded in the methanol extract of N. rtanjensis (around 1 mmol RE g -1 dry wt), while the extract of N. nervosa (~0.8 mmol RE g -1 dry wt) contained more total flavonoids than the extract of N. sibirica (~0.6 mmol RE g -1 dry wt).

Antioxidant activity of N. rtanjensis, N. sibirica and N. nervosa methanol extracts
Considering that antioxidants can act through different mechanisms, the antioxidant potential of N. rtanjensis, N. sibirica and N. nervosa was determined using three different assays: DPPH, ABTS and FRAP.Table 4 shows Trolox equivalent antioxidant capacity (TEAC) values of N. rtanjensis, N. sibirica and N. nervosa methanol extracts, and of the dominant reference chemicals present in the tested species: the essential oil of N. rtanjensis (containing 72% of trans,cis-nepetalactone and 16% of cis,trans-nepetalactone), the essential oil of N. cataria (containing 90% of cis,transnepetalactone), and standards of rosmarinic and chlorogenic acids.
The methanol extracts of all three studied Nepeta species exhibited scavenging activities against both DPPH • and ABTS •+ radicals and possessed a high capacity for the reduction of Fe 3+ in the FRAP assay (Table 4).The highest neutralizing activity of ABTS •+ radicals was characteristic of the N. nervosa methanol extract (with a TEAC value of 1.217 mmol TE g -1 dry wt), while the methanol extract of N. sibirica displayed the lowest activity (0.706 mmol TE g -1 dry wt).The methanol extracts of N. rtanjensis and N. nervosa (0.835 and 0.870 mmol TE g -1 dry wt, respectively) showed higher DPPH • radical scavenging activities than N. sibirica (0.768 mmol TE g -1 dry wt).In the FRAP assay, the methanol extract of N. sibirica showed the lowest capacity for reduction of Fe 3+ (1.935 mmol TE g -1 dry wt), while no statistically significant difference between the methanol extracts of N. rtanjensis and N. nervosa was observed (TEAC values of 2.939 and 2.968 mmol TE g -1 dry wt, respectively).The tested essential oils showed similar ABTS •+ and DPPH • radical scavenging activities and no activity in the FRAP assay (Table 4).Among the tested phenolic acids, rosmarinic acid showed a higher neutralizing capacity of ABTS •+ and DPPH • radicals (TEAC values were 1.695 and 40.326 mmol TE g -1 respectively) than chlorogenic acid (TEAC values were 0.823 and 23.311 mmol TE g -1 , respectively).The FRAP test showed that rosmarinic All determinations of antioxidant capacity by ABTS, DPPH, and FRAP assays were conducted in triplicate.The reported value for each sample was calculated as the mean of three measurements.Within each parameter, for methanol extracts and reference chemicals separately, values with the same letter are not significantly different at the p≤0.05 level according to the LSD test.Abbreviations: TE -Trolox Equivalents; DPPH -1,1-diphenyl-2picrylhydrazil; ABTS -2,2'-azino-bis(3-ethylbenzothiazoline-6sulphonate); FRAP -ferric reducing/antioxidant power; EO -essential oil.acid, as in other antioxidant tests, exhibited a higher potential for the reduction of Fe 3+ (1.523 mmol TE g -1 ) than chlorogenic acid (1.403 mmol TE g -1 ).

Toxicity evaluation of N. rtanjensis, N. sibirica and N. nervosa methanol extracts
An increase in lethality by any of the treatments compared to the controls was not observed.There were several toxic effects on development that were observed after the treatment of zebrafish embryos with all three extracts, but only pericardial and yolk sac edema were observed in 50.7±2.5% of all treated embryos, and only at the 48 hpf stage and in 100x diluted extracts (Fig. 4).We observed premature hatching (increased hatching rate) in embryos treated with 100x dilution of N. nervosa extracts, with 59.3%±4.2% of embryos hatching at 48 hpf as compared to only 14.3%±3.4% hatching in all other treatments and controls.
In addition to the phenolic compounds in Nepeta species, one more important group of biologically active secondary metabolites are monoterpenoids.The subfamily Nepetoideae is divided into two groups based on the presence or absence of nepetalactone [33].The largest number of Nepeta species possess some of the eight stereoisomers of nepetalactone [29].
In the present study, UHPLC/DAD/(+)HESI-MS/MS analysis confirmed previous phytochemical studies which revealed the tran,cis-isomer of nepetalactone as the major monoterpenoid in N. rtanjensis [9,13,17,34], and cis,trans-nepetalactone as the major isomer in N. sibirica [9,13,18].On the other hand, species such as N. bracteata [35], N. cilicia [36], N. curviflora [37] and N. daenensis [38] lack nepetalactone.In the case of N. nervosa, some researchers report no nepetalactone [9] while others detected this monoterpene lactone in trace amounts [13,20].Differences in reported data might be due to the detection limitations of the analytical methods applied and of the devices' performances, extraction protocols, plant growth conditions, developmental stages of plants, etc.In the present study, nepetalactone was not detected in the methanol extracts of N. nervosa plants grown in vitro, either by UHPLC-DAD or UHPLC/(+)HESI-MS 2 analysis.
Previous studies have shown that aromatic plants containing large amounts of phenolic compounds are a good source of natural antioxidants [3].The vast array of medicinal properties of phenolic compounds, such as anticancer, antimutagenic and protective activities, may be associated with their antioxidant capacity, during which the elimination of free radicals and prevention of lipid peroxidation occur [39].The present study reports for the first time the antioxidant properties of N. rtanjensis, N. sibirica and N. nervosa methanol extracts.Since antioxidants may act through different mechanisms, the antioxidant potential of the methanol extracts was determined using three differ- ent assays, DPPH, ABTS and FRAP.The DPPH assay is widely used to test the ability of compounds to act as free radical scavengers or hydrogen donors and is applicable in the evaluation of antioxidant activity of lipophilic compounds.The ABTS decolorization assay is, however, applicable to both lipophilic and hydrophilic antioxidants.The preformed radical monocation ABTS •+ is generated by the oxidation of ABTS with potassium persulfate and is reduced in the presence of hydrogen-donating antioxidants [25].The FRAP assay measures the absorption change that appears when the ferric-tripyridyltriazine complex (TPTZ-Fe 3+ ) is reduced to the ferrous-tripyridyltriazine (TPTZ-Fe 2+ ) in the presence of antioxidant compounds [40].
Generally, a considerable antioxidant activity in the ABTS assay was recorded for methanol extracts of all three Nepeta species analyzed here.The N. nervosa methanol extract was the most efficient, followed by the N. rtanjensis and N. sibirica extracts.In the case of DPPH and FRAP assays, N. rtanjensis and N nervosa methanol extracts were more efficient than those of N. sibirica.The more efficient antioxidant activity of N. nervosa and N. rtanjensis methanol extracts might be attributed to the higher content of phenolic compounds, especially of rosmarinic and chlorogenic acids.N. sibirica is characterized by a lower phenolic content than the other two species and, therefore, by lower DPPH • and ABTS •+ radical scavenging activities and FRAP capacity.
When major chemical constituents of the methanol extracts were tested separately, rosmarinic acid showed the highest neutralizing capacity to DPPH • and ABTS •+ radicals and a high capacity for iron reduction in the FRAP assay.The antiradical activity of phenolic compounds directly depends on the number of hydroxyl groups, but also on their molecular structure which determines the availability of phenolic hydrogens and the potential for stabilization of the resulting phenoxyl radicals formed by hydrogen donation [41].Rosmarinic acid is a dimer of caffeic acid and contains two aromatic rings, both carrying two ortho-hydroxyl groups.Rosmarinic acid, with its four hydroxyl groups, showed a stronger scavenging potential than other phenolic acids that have only one or two hydroxyl groups [42].Fadel et al. [43] have proposed a possible mechanism of antioxidative activity of rosmarinic acid according to which its presence in the cell membrane in association with polar lipids could prevent lipid peroxidation in the presence of reactive oxygen species, without disturbing the structure and permeability of the membrane.
As stated before, chlorogenic acid was the second most abundant phenolic acid in the analyzed samples, while caffeic acid was present in significantly lower amounts than chlorogenic acid, especially in N. rtanjensis and N. nervosa.On the other hand, caffeic acid was reported to possess stronger antioxidative activity in comparison to chlorogenic acid [42].This can be explained by the fact that the esterification of quinic acid decreases the activity of the acid-phenol in chlorogenic acid.Furthermore, the introduction of a second hydroxyl group in the ortho position is known to increase the antioxidative activity, which is the case for caffeic acid (ortho-diphenols).
Many authors have ascribed the antioxidant potential of Lamiaceae species to rosmarinic acid, as in the case of N. menthoides [4], Salvia virgata, S. staminea and S. verbenaca [44].Some Nepeta species have also been investigated for their antioxidant activity, and phenolic compounds were presumed to play an important role there.Kraujalis et al. [30] showed that the antioxidant potential of the methanol extract of N. cataria, N. cataria var.citriodora, N. transcaucasica and N. bulgaricum originated from rosmarinic acid, which is the dominant phenolic compound in these species.The same authors showed that there was a high degree of correlation between the amount of rosmarinic acid in the tested extracts and their antioxidant potential.Yazici et al. [16] determined the total phenolic compounds and antioxidant potential of N. italica, N. cilicia and N. caesarea, and the species which contained the highest amounts of phenolic compounds (N.italica) displayed the highest antioxidant potential.However, besides phenolic acids, other phenolic compounds such as flavonoids also affect the total antioxidant potential of an extract.Flavonoids were considerably less abundant in the studied Nepeta species and therefore were not qualitatively analyzed.However, taking into account the content of total flavonoids, the contribution of these compounds to the overall antioxidant capacity should not be ignored.
In addition to phenolic compounds, in the methanolic extracts of N. rtanjensis and N. sibirica, monot-erpenoid compounds, trans,cis-nepetalactone and cis,trans-nepetalactone, respectively, were identified, and they accounted for a significant share of the total secondary metabolites in the methanol extracts.To test the antioxidant activity of nepetalactones, we analyzed N. rtanjensis and N. cataria essential oils, which consisted of high amounts of trans,cis-nepetalactone and cis,trans-nepetalactone, respectively.Our results showed that in the DPPH test these two essential oils displayed significantly lower antioxidant potentials than the standards of rosmarinic and chlorogenic acid.There was no statistical difference between the two essential oils tested.In the case of the ABTS assay, the essential oils displayed similar ABTS •+ radical scavenging activity to chlorogenic acid.Neither the N. rtanjensis nor N. cataria essential oil could reduce iron in the FRAP assay, indicating that trans,cis-nepetalactone and cis,trans-nepetalactone did not contribute to the FRAP capacity of the corresponding methanol extracts.Our results are in agreement with published literature data that reported a very weak antioxidant potential of Nepeta essential oils in comparison to extracts containing phenolic compounds [3,5,45,46].Mahboubi et al. [46] showed that the essential oil of N. persica, with cis,trans-nepetalactone (80%) as the dominant compound, exhibited a very weak antioxidant potential, while the ethanol extract of the same species showed a high capacity for reduction of DPPH • radicals.Also, the essential oil of N. cataria, in which the dominant compound contained is cis,trans-(70%) and trans,cis-nepetalactone (6%), did not exhibit a capacity for DPPH • radical reduction, while the methanol extract of this species, which contained a very small amount of total phenolic compounds, showed a weak antioxidant potential [45].However, the essential oil of N. flavida had significant antioxidant activity, which was attributed to 1,8-cineole and linalool, both compounds being present in the oil at relatively high amounts [3].The essential oils of the species studied here contained 1,8-cineol and linalool in concentrations below one percent [17,18].The N. nervosa methanol extract, which contained no nepetalactone, generally showed slightly higher antioxidant activity than N. rtanjensis, and significantly higher activity than N. sibirica.This was further confirmation that the contribution of nepetalactones in the overall antioxidant potential of the analyzed methanol extracts was not of crucial importance.Phenolics most likely represent the major antioxidants in the samples.Synergistic or antagonistic actions between the individual components of extracts, which largely depend on their structure, i.e. between different phenolic compounds and between phenolic compounds and nepetalactone, should not be neglected.Iacopini et al. [47] showed that the antioxidant capacities of catechin, epicatechin, quercetin and rutin differed when applied as separate compounds in comparison with the assay when they were used as a mixture, leading the authors to conclude that the phenols can have synergistic or antagonistic effects.Furthermore, the minor constituents that are present in the methanol extracts of N. rtanjensis, N. sibirica and N. nervosa and which have not been analyzed here (e.g.flavonoids and other terpenoids) contribute to the bioactive potential of the extracts to a certain extent.A high total amount of phenolic compounds in an extract is not necessarily indicative of high antioxidant activity, but it should be taken into consideration that synergistic or antagonistic interactions may exist [48].
Natural products represent a significant reservoir of unexplored chemical diversity for early-stage drug discovery.Due to the strong antioxidant potentials of the tested species, they could be of great interest to the food and pharmaceutical industries.However, prior to use in the human diet, it is necessary to examine their potential toxicity.Zebrafish have been used predominantly in developmental biology and molecular genetics, but their value in toxicology and drug discovery has been recognized [49].To evaluate the toxicity of a compound or a plant extract, it is essential to identify the endpoints of toxicity and their dose-response relationships, to elucidate the mechanisms of toxicity and to determine the toxicodynamics of the compound/ extract [50].When the extracts of N. rtanjensis, N. sibirica and N. nervosa were diluted 100x and applied to zebrafish embryos, they developed a severe phenotype, consisting of pericardial and yolk sac edema, which may lead to reduced heart size, impaired cardiovascular function resulting in reduced cardiac output and ischemia of peripheral tissues.Only the extract of N. nervosa induced premature hatching (increased hatching rate).To our knowledge, this is the first attempt to characterize N. rtanjensis, N. sibirica and N. nervosa methanol extracts for their embryotoxicity.No Nepeta species were previously analyzed for their zebrafish embryo toxicity.A literature search revealed N. juncea cytotoxicity in the brine shrimp lethality bioassay [51], in which the water fraction was more effective than the chloroform one, while n-hexane and n-butanol fractions showed no significant cytotoxic activity.The methanol extracts of N. scrophularioides [52] and N. praetervisa [53] also showed significant cytotoxicity in the brine shrimp lethality assay.Numerous investigations revealed cytotoxicity of Nepeta species against animal and human cancer cell lines [54][55][56].Shakeri et al. [57] showed cytotoxic effects of the essential oil of Nepeta ucrainica L. spp.kopetdaghensis on the human ovarian carcinoma A2780 cell line and human breast adenocarcinoma MCF-7 cell line, while Dar et al. [58] showed that the essential oil of Nepeta govaniana inhibited the growth of lung and colon cancer cell lines.The hatching rate can be very heterogenous (it normally occurs between 48 and 72 hpf), but an increased hatching rate could be the result of changed physicochemical properties due to treatment or the influence of the treatment on embryonic motility inside the chorion or the hatching gland of zebrafish embryos [59].The observed increased hatching rate after the treatment with 100x diluted N. nervosa warrants further investigation into which constituents of the extract could cause the differences.
Being a plant genus of particular interest to pharmaceutical and food industries [29,54], the presented results contribute to an improved understanding of the chemical diversity of catnips.The three studied endemic species were shown to possess great potential for food conservation or as medicinal supplements if applied in optimized concentrations.However, alternative sources of plant material (e.g.field cultivation) should be established bearing in mind their vulnerability in nature.

No. Name of compound Concentration [μg 100 mg -1 fresh weight]
Peak numbers correspond to the numbers indicated in Table1and in Fig.1.
Chlorogenic acid (C 16 H 18 O 9 ) was quantified in the methanol extracts of the three Nepeta species using the referent product ion m/z [M-H-C 9 H 6 O 3 ] -at 191, resulting from the loss of caffeoyl moiety.The minor MS 2 ion of chlorogenic acid with m/z [M-H-C 9 H 6 O 3 -H 2 O-CH 2 O 2 ] -at 127 can arise from deprotonation and subsequent decarboxylation of the ion m/z [M-H-C 9 H 6 O 3 -H 2 O] -at 173, which is a product of dehydration of quinic acid (m/z 191) (Fig 2b).Rosmarinic acid (C 18 H 16 O 8 ), the ester of caffeic acid and 3,4-dihydroxyphenyl lactic acid, showed the major product ion at [C 9 H 7 O 4 -H 2 O] -at m/z 161, resulting from the loss of a water molecule from the caffeic part (m/z 179) of rosmarinic acid.Fragment m/z [C 9 H 7 O 4 -CO 2 -2H] -at 133 occurred after the loss of the CO 2 group from the caffeic part of the molecule and its deprotonation (Fig 2c).

Table 3 .
Total phenolic and total flavonoid contents in the methanol extracts of N. rtanjensis, N. sibirica and N. nervosa revealed spectrophotometrically.