|Home | About | Journals | Submit | Contact Us | Français|
The present study was conducted to carry out the phytochemical investigation of Tanacetum gracile Hook. f. & Thomson and to develop a method for the simultaneous quantification of the isolated compounds in the extracts of T. gracile growing in different locations. Cluster analysis rectangular similarity matrix was performed to understand the chemical fingerprinting variations in the extracts. High-performance liquid chromatography–electrospray ionization-quadrupole-time-of-flight-mass spectrometry (HPLC–ESI-QTOF-MS) was used to quantify four bioactive compounds, and separation of the compounds was achieved on a reverse-phase C8 column using a mobile phase of acetonitrile: 0.1% formic acid in water with a gradient elution by maintaining the flow rate of 300 μL/min. The QTOF–MS was operated using the electro-spray ionization technique with the positive ion polarity mode. The calibration curves of four marker compounds were linear over the concentration range of 3.12–100 ng/µL (R2> 0.996). A specific, accurate and precise HPLC–ESI-QTOF-MS method was optimized for the determination of kaempferol, ketoplenolide, tetramethoxyflavone and artemetin both individually and simultaneously. Quantification of these chemical markers in different extracts was carried out using this validated method. Kaempferol was isolated for the first time from T. gracile.
Since ancient time, humankind is depending on plants for health care, food and other day-to-day requirements in lives. There are several plants that are being used for healthcare system as a traditional medicine or as a source of bioactive constituents. The genus Tanacetum (Asteracea) is represented by almost 150–200 species, widespread mainly in Europe and western Asia (1, 2). Several species of the genus have been reported to have various biological activities (3). Tanacetum gracile Hook. f. & Thomson grows wild in Ladakh (India), Pakistan and Afghanistan in the alpine western Himalayas (4, 5). The height of the plant is 30–60 cm. The stem is hairy, slender and branched. Leaves are scattered, twice pinnately cut and 1.3–2.5 cm in size (6, 7). Traditional applications of this plant include the use of decoction of leaves and flowers in small doses against intestinal worms in children (8). Powder of the leaves is also used in curing obesity and sore throat (9). There are several reports in the literature about the chemical composition of essential oil of different Tanacetum species. Chemical composition of the essential oil of two Tanacetum species of the alpine region in Indian Himalayas was reported (10). Apart from this, a paper has been reported on induction of mitochondrial-dependent apoptosis by an essential oil from T. gracile (11).
As per our institute's mandate, investigations are being carried out on aromatic and medicinal plants for their chemical and biological evaluation. In an experiment conducted by Sabari Ghosal and Sadhna (12) with the aid of local Amchis and scientists of Indian Institute of Integrative Medicine, Jammu, cytotoxic activity of T. gracile was evaluated. On an average, the IC50 values of these components were between 6 and 60 μg/mL. Cytotoxicity of the different extracts was evaluated on PC-3 and HeLa cells by the MTT assay. Their study clearly demonstrates that the extracts of T. gracile could exhibit an antiproliferative effect on human prostate cancer and tumor cells. These results reveal the initial potential of this medicinal plant as a cytotoxic agent for therapy against human prostate cancer and tumor. As the plant has been least explored for its phyto-constituents when compared with its essential oil and there are only few reports available where β-thujone, ketoplenolide B, artemetin, 5-hydroxy-3,6,7,3′,4′-pentamethoxyflavone and 5,4′-dihydroxy-3,6,7,3′,4′-tetramethoxyflavone have been reported from this plant (13, 14), an attempt was made to investigate the chemical constituents and the percentage of those constituents responsible for its cytotoxicity. We were successful in isolating four marker compounds, such as ketoplenolide, kaempferol, artemetin and tetramethoxyflavone, from this plant. Out of the four constituents, kaempferol was isolated for the first time. Kaempferol, tetramethoxyflavone and artemetin are known to possess anticancer activities, active against various human cancer cell lines (15–20).
Thus our efforts resulted in the isolation of ketoplenolide, kaempferol, artemetin and tetramethoxyflavone from T. gracile, and their simultaneous quantification in various extracts collected from different geographical locations, chemical fingerprinting and development of a validated HPLC–QTOF-MS method because there is no method available in the literature where the above-said four markers have been simultaneously separated and quantified on the basis of HPLC–QTOF-MS. The proposed method was applied for the standardization of T. gracile extracts collected from nine different locations in Ladakh, and cluster analysis (CA) rectangular similarity matrix was performed to understand the chemical fingerprinting variations in the extracts. The results highlight the potential of this plant as a source of bioactive compounds, and the developed method can be applied for the quality control (QC) of the plant being used as a medicinal herb and in the formulations containing the active anticancer compounds, respectively.
Experiments were performed on an Agilent LC-MS/MS system (Agilent Technologies, USA) equipped with an Agilent HPLC 1260 series consisting of a standard auto-sampler (G1329B), a quaternary pump (G1311B), a thermostatted column compartment (G1316A), a diode array detector VL (G1315D) interfaced to an Agilent G6540A quadrupole–time-of-flight detector with a dual ESI source. Agilent Mass Hunter Qualitative analysis (B.05.00) and Agilent Mass Hunter quantitative analysis were used for qualitative and quantitative analysis, respectively. An Agilent Mass Hunter workstation was used for data acquisition. For chromatographic analysis, an Agilent Poroshell 120 EC-C8 column (3.0 × 30 mm, 2.7 um) was employed. 0.1% formic acid in water [A] and acetonitrile [B] was used as mobile phase. Separations were carried out using a gradient as follows: initially B% was kept at 20 and then up to 10 min 40% B, 15 min 20% B and 20 min 20% B. The flow rate was constant at 0.3 mL/min with a run time of 20 min and an injection volume of 1 μL. The column temperature was maintained at 30°C. The sample was run in a positive mode with gas temperature 350°C, drying gas flow rate 12 L/min, nebulizer 40 psi, sheath gas flow rate 9 L/min, sheath gas temperature 350°C, Vcap voltage 4000 V, nozzle voltage 1,000 V, fragmentor voltage 110 V, and acquisition range was set between 100 and 1,000 to get good-quality MS spectra.
Infusion mass experiments were performed on the marker compounds using an external syringe pump to monitor the ions in the compounds. The selected ion monitoring (SIM) mode was used for quantification of the compounds. The ions detected for the compounds were as kaempferol at m/z 287.0515 [M + H]+, ketoplenolide at m/z 273.1428 [M + Na]+, 251.1609 [M + H]+, tetramethoxyflavone at m/z 397.0842 [M + Na]+, 375.1027 [M + H]+ and artemetin, m/z 411.0997 [M + Na]+, 389.1180 [M + H]+.
All chemicals and reagents used in the study were of high purity. LC–MS grade water, acetonitrile and methanol were purchased from Fisher Scientific, India. HPLC grade formic acid, purchased from Merck, India, was used as the mobile phase, whereas LC–MS grade methanol was used for sample preparation. All the four chemical markers, i.e., kaempferol (≥95%), ketoplenolide (≥90%), tetramethoxyflavone (≥95%) and artemetin (≥95%), used as standards in the study, were isolated from T. gracile. The purity of the standards was established by HPLC, whereas the compounds were identified on the basis of NMR and mass spectral data.
The samples of T. gracile, family Asteracea, were procured from Ladakh (India). The plant material was identified by Dr. S. Kitchlu, and a specimen voucher (no. 22,123) of the sample was deposited in herbarium of the institute. The powder of T. gracile whole plant (941.9 g) was taken and extracted with pet ether. The extraction process was repeated five times and solvent was evaporated to dryness under vacuum using a rotavapor. The yield of the extract was 4.08%. The marc (residue that remains after extraction with pet ether) was extracted using methanol. These extracts were subjected to bioactivity against different human cancer cell lines. Some of these extracts were exhibiting potent anticancer activity against the human lung (A549), colon (HCT-116), prostate (PC-3) and breast (MCF-7) carcinoma cell lines. The bioactive extracts were taken for detailed phytochemical investigations. Detailed investigation on anticancer activity of the extract along with the marker compounds will be published separately.
The pet ether extract (36 g) was subjected to silica (100–200 mesh size) column chromatography and first eluted with pet ether, gradually increasing the polarity by increasing the quantity of ethyl acetate. A total of 615 fractions, each of 100 mL, were collected and concentrated on water bath by using a rotavapor. These fractions were pooled by monitoring the TLC using different developing systems and finally a total of 15 groups were prepared. Fractions 204–221 were pooled and further subjected to re-chromatography by which a white compound was isolated. This compound was crystallized and re-crystallized in ethyl acetate to obtain a purified molecule coded as TG-1 (18 mg). Fractions 325–332 were pooled and re-chromatographed by which a dark yellow compound was obtained. This compound was crystallized and re-crystallized in methanol to obtain a purified molecule coded as TG-2 (15 mg). In addition, fractions 375–394 were subjected to repeated column chromatography, which resulted in a light yellow compound. Crystallization and re-crystallization of this compound in methanol were carried out to obtain a purified molecule coded as TG-3 (21.8 mg).
Similarly, fractions 409–445 from methanolic extract resulted in a yellow compound. Crystallization and re-crystallization of this compound in methanol were carried out to obtain a purified molecule coded as TG-4 (8 mg).
All the compounds (TG-1 to TG-4) have been characterized by NMR and mass spectrometry (13) as ketoplenolide, artemetin, tetramethoxyflavone and kaempferol. Kaempferol has been isolated for the first time from T. gracile.
Reference solutions of kaempferol (Stock I), ketoplenolide (Stock II), tetramethoxyflavone (Stock III) and artemetin (Stock IV) were prepared by weighing 1 mg of each compound. The quantities were transferred to 1 mL volumetric flasks, and dissolved and diluted suitably with LC–MS grade methanol. All the reference solutions (1 mg/mL) were covered with aluminum foil and sealed with paraffin film to avoid loss due to evaporation. Stocks I, II, III and IV were mixed together, and diluted suitably with methanol. One microliter of this solution was used to achieve six calibration standards (CAL STD) containing combination of chemical markers.
Three QC standards (lower quality control (LQC): 5 ng/mL; medium quality control (MQC): 45 ng/mL; higher quality control (HQC): 90 ng/mL each of chemical markers) were prepared. Three different QC standards (LQC: 10 ng/mL; MQC: 40 ng/mL; HQC 80 ng/mL each of chemical markers) were used for recovery studies.
A 5 mg of each from the 18 extracts (9 petroleum ether extracts and 9 methanolic extracts prepared from the plant material collected from different geographical locations) was accurately weighed and dissolved in 1 mL of LC–MS grade acetonitrile and methanol (1:1). The solutions were subjected to sonication at 30°C for 15 min. All the samples were filtered through Millipore Millex-GN Nylon (0.2 μm) filters. The filtered samples were injected into the system for LC–MS analysis.
The descriptive statistics of all the nine Tanacetum extracts using four chemical markers was performed using the SPSS, ver. 22, software. For CA rectangular similarity, matrix was generated using Euclidean distant coefficient and un-weighted pair group method with arithmetic mean (UPGMA)-based dendrogram was plotted. Two-tailed significant Pearson's correlation coefficient (r) was calculated at P < 0.01 and P < 0.05. Same data were validated by principal component analysis (PCA) following Kaiser's rule and unrotated factor solution (SPSS, ver. 22).
The proposed method was validated to meet the acceptance criteria as per the guidelines of the International Conference on Harmonization of Technical Requirements for Registration of Pharmaceuticals for Human Use (ICH) (21). The specificity of the method was established by comparing mobile phase with those spiked with the chemical markers to find out interference from endogenous components. The CAL STD solutions were utilized for establishment of linearity and range (linear least-squares regression with a weighting index of 1/x). The precision and accuracy parameters were ascertained in LQC, MQC and HQC samples (six replicates each in three sets) on the same day and on 3 consecutive days. The intra- and inter-assay accuracy (% RE) of the method was determined by getting absolute error from subtracting the true value with the measured value and relative error as follows: relative error (% RE) = [(absolute error/true value] × 100. The intra-and inter-assay precision (% relative standard deviation or RSD) of the method was calculated from mean measured concentrations as follows: % RSD = (SD of mean measured conc./mean measured conc.) × 100.
Recovery (RSD) studies were also performed by using three different concentration (LQC, MQC and HQC) samples (six replicates each in three sets) on the same day and on 3 consecutive days for ensuring the optimum performance of the method.
Optimum chromatographic separation of chemical markers was assessed by eluting the extract on an Agilent Poroshell 120 EC-C8 (3.0 × 30 mm, 2.7 µm) column with a mobile phase combination of acetonitrile: 0.1% formic acid in water with a gradient elution at initially B% was kept 20 and then up to 10 min 40% B, 15 min 20% B and 20 min 20% B. The flow rate was constant at 300 µL/min with a run time of 20 min.
The gradient elution used in this experiment allowed the separation of all the compounds studied and baseline separation was achieved in the case of all the chemical markers and a full scan in positive mode was used for all the chemical markers. The resulting chromatograms showed the peaks of the compounds.
By the use of 0.1% formic acid, the ionic strength decreased and the signal-to-noise ratio increased.
Optimum chromatographic separation of kaempferol, ketoplenolide, tetramethoxyflavone and artemetin was achieved by acetonitrile: 0.1% formic acid in water with a gradient elution. All the chemical markers were added simultaneously in the samples and the resulting chromatograms showed a retention time of 7.24, 9.26, 10.96 and 13.28 min for kaempferol, ketoplenolide, tetramethoxyflavone and artemetin, respectively (Figure 1A: total ion chromatogram (TIC); Figure 1B–E: extracted ion chromatogram (EIC)). A full scan in positive ion mode was used for all the chemical markers. During direct infusion, the mass spectra of the major product ions in the positive mode for SIM mode were used to monitor the ions of the compounds.
The method was found to be specific: mobile phase when compared with mobile phase spiked with chemical markers did not show any interference at the respective retention times of each chemical markers.
The limit of detection (LOD) and limit of quantitation (LOQ) for kaempferol, ketoplenolide, tetramethoxyflavone and artemetin are given in Table I.
The combined recovery for kaempferol, ketoplenolide, tetramethoxyflavone and artemetin was carried out in LQC, MQC and HQC samples. The recovery % of kaempferol was 99.20 (from LQC), 98.50 (from MQC) and 97.47 (from HQC). The recovery % of ketoplenolide was 98.45 (from LQC), 102.30 (from MQC) and 103.10 (from HQC). For tetramethoxyflavone, it was 101.00 (from LQC), 98.25 (from MQC), 97.53 (from HQC) and for artemetin, the recovery % was found to be 100.50 (from LQC), 99.05 (from MQC) and 98.33 (from HQC) (Table II).
The intra-assay accuracy in terms of % RE was in the range of −1.20 to −0.15 for kaempferol, −1.96 to 0.8 for ketoplenolide, −0.82 to 0.6 for tetramethoxyflavone and −1.74 to −0.62 for artemetin. Inter-assay % RE was in the range of −2.00 to −1.28 for kaempferol, 0.0 to 0.97 for ketoplenolide and −0.8 to 0.24 tetramethoxyflavone and −2.4 to 0.88 for artemetin.
Intra-assay precision (% RSD) was in the range of 1.1 to 2.25 for kaempferol, 0.01 to 2.48 for ketoplenolide, 0.80 to 1.29 for tetramethoxyflavone and 1.58 to 2.07 for artemetin. The inter-assay % RSD was in the range of 1.51 to 2.88 for kaempferol, 1.20 to 2.35 for ketoplenolide, 0.85 to 1.78 for tetramethoxyflavone and 1.56 to 2.35 for artemetin (Table III).
ESI–QTOF-MS fingerprinting similarity evaluations of the extracts were assessed by using cluster analysis that helps to understand the concentration of individual chemical markers in the extract growing at different altitudes.
Our investigation on the inter-relationships between markers present in Tanacetum extract showed very strong correlation (P < 0.01) of ketoplenolide with tetramethoxyflavone and artemetin and strong correlation (P < 0.05) between tetramethoxyflavone and artemetin. These data were further validated by PCA for complete and proper classification of the elemental data, using Kaiser's rule (Eigen value > 1) and no rotation parameters. PCA yielded one component explaining a cumulative variation of 67.47%. Principal component 1 (PC1) could significantly correlate ketoplenolide with tetramethoxyflavone and artemetin (Table V) as was also seen in Pearson's correlation analysis (Table IV). Correlation between principle factor loadings and markers are presented in Table V, with loading values >0.85 are considered significant and marked bold.
The quantified data of the nine extracts were computed for similarity values (Table VI). All the nine sample extracts were investigated for a clustering pattern using Euclidean distance and between groups. The resulted dendrogram classified all the nine samples into two major clusters (I and II). Cluster I had the maximum samples (seven), while Cluster II had only two samples (7 and 9). At a critical value of 12.5, Cluster I was subdivided into IA and IB with three samples each (Figure 2).
In recent times MS coupled with HPLC (22, 23) is generally considered as a better analytical technique in terms of quality data, chromatographic resolution and increased detection limits with greater sensitivity. The HPLC–QTOF-MS method was used for the simultaneous determination and quantification of kaempferol, ketoplenolide, tetramethoxyflavone and artemetin in various extracts of T. gracile. Results are shown in Table VII. TICs of nine extracts indicating the presence of all the four compounds have been shown in Figure 3.
We have developed a simple and reliable HPLC–QTOF-MS-MS method. This method has high degree of reproducibility, accuracy, sensitivity and also provides short separation time (20 min). The proposed method can also be used for the quantification of individual chemical markers for routine analysis. This proposed method for the determination of kaempferol, ketoplenolide, tetramethoxyflavone and artemetin by HPLC–QTOF-MS has not been reported, prior to this investigation, in which chemical markers have been quantified on the basis of their major ions. The major ions observed in the positive ion ESI spectra are as follows: kaempferol at m/z 287.0515 [M + H]+ (Figure 4A), ketoplenolide at m/z 273.1428 [M + Na]+, 251.1609 [M + H]+ (Figure 4B), tetramethoxyflavone at 397.0842 [M + Na]+, 375.1027 [M + H]+ (Figure 4C) and artemetin at m/z 411.0997 [M + Na]+, 389.1180 [M + H]+ (Figure 4D). Four chemical markers used as a standard in the present study showed separate peaks in the EIC (Figure 1B–E) and TIC (Figure 1A). Kaempferol, ketoplenolide, tetramethoxyflavone and artemetin appeared in the EIC at 7.22, 9.24, 10.96 and 13.28 min, respectively.
The gradient method comprising of water with 0.1% formic acid and acetonitrile was found to be the most suitable method for the separation of kaempferol, ketoplenolide, tetramethoxyflavone and artemetin with optimum baseline separation. 0.1% formic acid was used as an ionizing agent to get maximum resolution and improved peak symmetry. However, when methanol and water with the same gradient parameters were used, separation of the four markers was not up to the mark. It led to the merging of peaks with poor resolution and poor peak shape. The method so developed led to the separation of the compounds on the basis of their hydrophobic and hydrophilic behavior. Kaempferol, being the most polar of all, was eluted first followed by ketoplenolide, then tetramethoxyflavone and artemetin, comparatively hydrophobic in nature. Since the three major components were not detected in the methanolic extracts, only pet ether extracts were taken for further quantification and analysis.
Kaempferol, ketoplenolide, tetramethoxyflavone and artemetin were quantified in various extracts of the plant and results are presented in Table VII. The table shows that artemetin is the major constituent in Extracts 6, 7 and 8, 0.927, 1.787 and 1.101%, respectively, while Extract 9 was rich with ketoplenolide. Tetramethoxyflavone was found in good concentration in Extracts 9 and 6 while the amount of this compound was not significant in other extracts. Kaempferol was not obtained in a significant amount in the all extracts (Figure 5).
Extract 9, collected from the Thiksay region at an altitude of 3,452 m, was found to possess a maximum percentage of ketoplenolide (1.51%), tetramethoxyflavone (0.73%) and artemetin (1.19%). Similarly, Extract 7, collected from the Mahuy region at an altitude of 4,152 m, contained 1.36% of ketoplenolide, 0.41% tetramethoxyflavone, 1.78% of artemetin and 0.07% of kaempferol. The Ganglas region with an altitude of 3,900 m represented minimum quantities of the aforesaid analytes with 0.006, 0.08 and 0.01% of ketoplenolide, tetramethoxyflavone and artemetin, respectively. Whereas the above-mentioned three markers were present in all the extracts collected from various geographical locations, kaempferol was present only in the extracts collected from the Khardung, Hemis and Mahuy regions with varying altitudes. It was not even detectable in the rest of the extracts.
Thus, the method was successfully applied in the identification and quantification of the isolated markers. The method was also validated in terms of specificity, accuracy, precision and sensitivity of the chemical markers, and utilized for the determination of kaempferol, ketoplenolide, tetramethoxyflavone and artemetin either individually or simultaneously in different extracts of T. gracile collected from nine different locations of different altitudes (Table VIII).
A specific, accurate and precise HPLC–QTOF-MS-MS method for the determination of kaempferol, ketoplenolide, tetramethoxyflavone and artemetin both individually and simultaneously was optimized. Quantification of these chemical markers in the extracts was carried out by using this validated method.
The authors are thankful to the Director, CSIR-IIIM, Jammu, for providing necessary facility, CSIR for financial support under project BSC 0110 for carrying out the study. N.S. is thankful to the Department of Science and Technology, New Delhi, for the award of INSPIRE Fellowship.
Conflict of interest statement. The authors have declared no conflict of interest.