Validated HPTLC method for the simultaneous quantification of diterpenoids in Vitex trifolia L

J. Sep. Sci. 2013, 36, 2373–2378 2373

Neerja TiwariAkhilesh Kumar YadavMadan Mohan Gupta

Analytical ChemistryDepartment, CSIR-CentralInstitute of Medicinal andAromatic Plants, Lucknow, India

Received February 4, 2013Revised May 9, 2013Accepted May 9, 2013

Research Article

Validated HPTLC method for thesimultaneous quantification of diterpenoidsin Vitex trifolia L.

Vitex trifolia L. is an important Indian medicinal plant with diverse pharmacological proper-ties. In a recent study, we reported the isolation and antitubercular activity evaluation of threenew diterpenoids from its leaves; here we have developed a validated rapid, simple, precise,and accurate high-performance TLC method for the simultaneous quantification of isolatedditerpenoids in V. trifolia. Diterpenoids, 6�,7�-diacetoxy-13-hydroxy-8(9),14-labdadien (A),13-hydroxy-5(10),14-halimadien-6-one (B), and 9-hydroxy-13(14)-labden-16,15-olide (C) wereseparated on silica gel 60F254 high-performance TLC plates using chloroform/acetone (98:2,v/v) as mobile phase. The quantitation of diterpenoids was carried out using densitomet-ric reflection/absorption mode at 610 nm after postchromatographic derivatization usinga vanillin/sulfuric acid reagent. A precise and accurate quantification can be performedfor compounds A and B in the linear working concentration range of 333–1000 ng/bandand for C in the range of 670–2000 ng/band with good correlations (r = 0.9984, 0.9991,and 0.9994, respectively). The method was validated for peak purity, precision, accuracy,robustness, LOD, and LOQ, as per the ICH guidelines. The method reported here is simple,reproducible and may be applied for the quantitative analysis of the above diterpenoids inthe leaves of V. trifolia.

Keywords: Diterpenoids / HPTLC / Vitex trifolia L.DOI 10.1002/jssc.201300131

1 Introduction

Vitex trifolia L. (Verbenaceae) is an important plant,widespread throughout Southeast Asia, Micronesia, Aus-tralia, and East Africa. It is commonly known as commonchaste tree (English), nochi (Kannada), and Jalanirgundi (San-skrit). The leaves of V. trifolia are useful for rheumatic pains,inflammation, sprains, and fever. The roots are used to treatfebrifuge, painful inflammations, cough, and fever. The flow-ers are used to treat fever and the fruits are used in amen-orrhea [1]. This plant is known to possess several pharmaco-logical properties viz., antipyretic [2], antibacterial [3], againstasthma and allergic diseases [4]. Aqueous and ethanol ex-tracts of leaf of V. trifolia show significant protection againstCCl4-induced liver damage. Phytochemical studies on thisplant showed the presence of terpenoids [5, 6], iridoid glyco-sides [7], lignans [8], and flavonoids [9, 10].

Correspondence: Dr. Madan Mohan Gupta, Analytical ChemistryDepartment, CSIR-Central Institute of Medicinal and AromaticPlants, Lucknow 226015, IndiaE-mail: [email protected]: +91-522-2342666

Abbreviations: DHL, 6�,7�-diacetoxy-13-hydroxy-8(9),14-labdadien; HHO, 13-hydroxy-5(10),14-halimadien-6-one;HLO, 9-hydroxy-13(14)-labden-16,15-olide; HPTLC, high-performance TLC

Some HPLC [11] and high-performance TLC (HPTLC)[12–14] methods are reported for the quantification of iri-doids. A method for the simultaneous quantification of diter-penoids is reported in V. agnus-castus [15]. In spite of con-taining complex mixture of diterpenoids, V. trifolia still lacksan analytical method based on diterpenoids. HPTLC is themethod commonly applied for the identification assay andstability study of herbal raw materials and formulations. Cur-rently HPTLC emerges as an alternative to HPLC because offar less solvent consumption, reduction of the analysis time,cost of analysis, besides being eco friendly. HPTLC also fa-cilitates repeated detection (scanning) of the chromatogramswith same or different parameters [16, 17]. In our previousstudy, we reported the isolation and antitubercular activityevaluation of three new diterpenoids from the methanolicextract of leaves of V. trifolia [18], here we have developed avalidated rapid, simple, precise, and accurate HPTLC methodfor the simultaneous quantification of isolated diterpenoidsin V. trifolia (Fig. 1).

2 Materials and methods

2.1 Reagents and chemicals

The markers 6�,7�-diacetoxy-13-hydroxy-8(9),14-labdadien(DHL; A), 13-hydroxy-5(10),14-halimadien-6-one (HHO; B),and 9-hydroxy-13(14)-labden-15,16-olide (HLO; C) used for

C© 2013 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim www.jss-journal.com

2374 N. Tiwari et al. J. Sep. Sci. 2013, 36, 2373–2378

Figure 1. Chemical structures of A,B, and C.

analysis, were of 99% purity (determined by HPLC and spec-tral analysis). Markers A–C were isolated in the lab fromV. trifolia and identified by evaluation of NMR and otherspectral properties [18]. HPTLC plates (Silica gel 60F254), sil-ica gel (60–120 mesh) for column chromatography and HPLCgrade solvents were purchased from m/s Merck (Mumbai, In-dia). Plant material of V. trifolia was collected from Lucknowin March 2009 and authenticated by Taxonomy and Pharma-cognosy Department of our institute (CSIR-CIMAP specimenvoucher no. 9417).

2.2 Standard solution and sample preparation

Extraction was optimized as solvents, techniques, times, andtemperatures. Different solvents (hexane, chloroform, ace-tone, and methanol) (Table 1) with different techniques likecold percolation, microwave extraction, sonication, and hotextraction methods (Table 2) were tried to optimize the ex-traction method. For content determination, amount of plantparts were adjusted to perform the quantification above LOQlevel. Different plant parts (leaves, stem, fruit, and root) havebeen taken for content determination to find out highest con-tent of markers (Table 3). The extracts were concentrated ona rotary evaporator to afford different extracts. The standardsolutions of markers A–C were prepared by dissolving accu-rately weighed 1.0 mg in 1 mL of methanol as a stock solution(1 mg/mL) and stored at 4�C. These standards were further

Table 1. Extraction efficiency of different solvent for analyzedditerpenesa)

Solvent Amount of compounds quantifiedb) (mg/100 gm) ± SD

DHL (A) ± SD HHO (B) ± SD HLO (C) ± SD

Hexane 5.15 ± 0.36 ND NDChloroform 8.11 ± 0.13 18.09 ± 0.15 NDAcetone 14.08 ± 0.24 13.59 ± 0.38 17.30 ± 0.06Methanol 22.25 ± 0.50 18.63 ± 0.44 19.90 ± 0.53

a) Microwave extraction at 50�C and 350 W for 10 min.b) Dried plant material.

diluted and mixed as per the requirements to obtain a de-sired concentration for quantification. The concentrations ofreference standards used for calibration were in the range of333–1000 ng/band for A and B and in the range of 670–2000ng/band for C. The known volumes of mixed standards werespotted on the TLC plates to prepare six point calibrationcurves.

2.3 Chromatographic separation

Samples of each standard solution and V. trifolia leaf extractwere spotted as bands of 5 mm width on silica gel 60F254 TLCplates (10 × 10 cm, Merck, Germany). Compounds were spot-ted at 15 mm from both the lower and left edge, and 15 mm


J. Sep. Sci. 2013, 36, 2373–2378 Other Techniques 2375

Table 2. Extraction efficiency of different techniques for analyzedditerpenesa)

Techniquesc) Amount of compounds quantifiedb)

(mg/100 gm) ± SD)


Cold 21.99 ± 0.68 31.51 ± 0.36 16.48 ± 0.25Sonication 20.15 ± 0.35 29.35 ± 0.30 15.11 ± 0.35Hot 19.84 ± 0.32 23.26 ± 0.29 12.21 ± 0.25Microwave 22.25 ± 0.50 18.63 ± 0.44 19.90 ± 0.53

a) Dried plant material.b) Methanol was used as solvent.c) Cold, kept for 24 h at 30�C; sonication, for 30 min at 50�C; hot,boiling water bath for 30 min; microwave, 50�C and 350 W for10 min.

Table 3. Percent content of analyzed diterpenes in different plantparts of V. trifolia (VT)

Plant Parta) Amount of compounds quantifiedb)

(mg/100 gm) ± SD


VT-L 22.25 ± 0.50 18.63 ± 0.44 19.90 ± 0.53VT-S 14.48 ± 0.30 10.35 ± 0.35 14.45 ± 0.35VT-R ND 11.26 ± 0.29 13.55 ± 0.40VT-F ND ND 12.33 ± 0.24

a) L, leaves; R, root; S, stem; F, fruit.b) Dried plant material.

space between two spots using a semiautomatic applicator de-vice (Linomat 4-Camag) under continuous nitrogen currentwith a constant application rate of 5 �L/s. Different mobilephases were used to separate DHL, HHO, and HLO from V.trifolia leaf extracts. The plates were developed in a normalchromatographic chamber presaturated for 15 min with mo-bile phase. Quantification was performed at 610 nm usingTLC Scanner 3-CAMAG. The images of plates were taken us-ing Reprostar 3-Camag in white light. All these instrumentswere controlled by winCats 1.4.2.8121 software.

2.4 Method validation

2.4.1 Specificity

The specificity, expressed as the ability of the method to as-sess the analyte clearly in the presence of other compoundsfrom analyzing sample, was confirmed by comparing the Rf

of analytes and standards. The spots for DHL, HHO, andHLO in samples were confirmed by comparing the Rf withthe reference standards (Figs. 2 and 3). Peak resolutions andquantitation of the fortified peaks (see recovery) of individ-ual A, B, and C were taken in additional indications of thespecificity (Table 4). Coelution of compounds was checked bycomparison of spectra (Fig. 4). Also, the bands correspond-

Figure 2. Photodocumentation after derivatization with anisalde-hyde acetic acid reagent; track assignment as: track 1, 3, and 5artificial mixture of standards, track 2 and 4 methanolic extract ofV. trifolia.

ing to spots A, B, and C were recovered by dissolving inmethanol. Spot purity was confirmed by NMR spectroscopicdata of recovered samples (A–C).

2.4.2 Linearity

For providing linearity six calibration points were analyzedover the range of 333–1000 ng/band for A and B and in therange of 670–2000 ng/band for C. The calibration graphswere constructed by plotting the peak area versus concen-tration of A–C, and the regression equations were calcu-lated using the least square method in Graph PAD Prism 4(Table 5). Standard solution was applied in triplicates and an-alyzed using the TLC method described in the experimentalpart.

2.4.3 LOD and LOQ

LOD and LOQ were calculated for compounds using thelinear regression equations over a concentration range of333–1000 ng/band for A and B and in the range of 670–2000 ng/band for C. The following equations were applied:LOD = 3 Sy.x/b & LOQ = 10 Sy.x/b, where Sy.x is the standarddeviation of the Y-value distribution around the regressionline and b is the best fit value of slope of calibration curve.The LOD and LOQ values are listed in Table 5. Aliquots ofthe standard working solutions of markers were applied intriplicate to the plates.

2.4.4 Precision

The precision was also determined as intra-assay preci-sion (repeatability) and inter-assay precision (reproducibil-ity). Intra-assay precision was determined by analyzing sixindividual spots (333–1000 ng/band) for each compound ap-plied onto different plates on the same day and intermediate



Figure 3. Digital scanning profile of methanolic extract of V. trifolia (1) and artificial mixture of standards (2).

Table 4. Recovery study to evaluate accuracy of the method

Compounds Amount in the Amount Amount %sample (ng) added (ng) found (ng) Recovery

222.5 116.5 350DHL (A) 222.5 333 569 102.73

222.5 449.5 689186.3 116.5 299

HHO (B) 186.3 333 510 99.10186.3 449.5 638199 116.5 320

HLO (C) 199 333 535 101.15199 449.5 658

precision was determined by six repeated assays of referenceover a period of 6 days (one plate per day only). Both precisionwas expressed in terms of % RSD (Table 5).

2.4.5 Robustness

Robustness is a measure of the capacity of a method to re-main unaffected by small but deliberate variations in themethod conditions, and is an indication of the reliability ofthe method. The robustness of the method was studied bydetermining the effects of small variation of mobile phasecomposition, chamber saturation period, derivatizing time,and heating time (Table 6).

2.4.6 Accuracy

The accuracy of the method was tested by performing recoveryof the compounds in the sample at three levels. V. trifoliamethanolic extract was spiked with known amount of eachcompound and analyzed (Table 4). The recovery percentage

was calculated by using the formula: recovery (%) = (amountfound–original amount)/amount spiked × 100.

3 Results and discussion

3.1 Optimization of extraction procedure

To optimize the extraction of analytes, effects of differentextraction solvents and techniques were studied. Powderedplant parts were extracted with 40 mL different solvents (hex-ane, chloroform, acetone, and methanol) by microwave ex-traction method at 50�C and 350 watt for 10 min to choosemost suitable solvent for further analysis. Methanol was pre-ferred due to high extractive value (Table 1). Cold percolationfor 24 h at 30�C was performed by allowing the plant ma-terial in contact with the solvent; while hot extraction wasperformed for 15, 30, and 60 min in a boiling water bath. Ul-trasonic extraction was performed for 15, 30, and 60 min at50�C and microwave extraction was applied for 10, 15, 20 minat 50, 75, and 100�C (all at 350 W). The extraction efficiencyanalysis of the obtained extracts revealed that the amounts ofmarkers varied with the extraction techniques. It is evidentthat microwave extraction for 10 min at 50�C (350 W) withmethanol is the better choice for extraction of analyzed diter-penes from plant matrix due to comparative high extractionefficiency and less time and solvent consumption (Table 2).Also, the leaves were found to contain highest concentrationof analyzed diterpenes (Table 3).

3.2 Method validation

The developed TLC method was validated according to ICHguidelines for the following parameters: specificity, linearity,range, LOD, LOQ, precision, robustness, and accuracy. The


J. Sep. Sci. 2013, 36, 2373–2378 Other Techniques 2377

Figure 4. Spectra of compound A (1), B (2), and C (3) in visible range (400–800 nm).

Table 5. Statistical analysis of calibration curves in the HPTLC determination of analyzed diterpenesa)

Parameters DHL (A)b) HHO (B)b) HLO (C)c)

Rf value 0.74 0.57 0.51Slope 7.653 ± 0.07678 5.186 ± 0.03993 1.713 ± 0.0109095% confidence limits of slope 7.490 to 7.816 5.101 to 5.271 1.690 to 1.736Intercept 3760 ± 99.67 1494 ± 51.84 2660 ± 28.2995% confidence limits of intercept 3549 to 3971 1384 to 1604 2600 to 2720Goodness of fit (Sy,x) 185.5 96.46 52.63Correlation coefficient (r) 0.9984 0.9991 0.9994LOD (ng/band) 72.72 55.80 92.17LOQ (ng/band) 242.39 186.00 307.24Repeatabilitya) (% RSD) 1.19 1.03 0.98Reproducibilitya) (% RSD) 1.21 1.09 1.27Specificity/peak purity

R (s, m)d)

Standard track 0.9997 0.9989 0.9999Sample track 0.9998 0.9989 0.9898

R (m, e)e)

Standard track 0.9998 0.9989 0.9989Sample track 0.9984 0.9995 0.9991

a) Number of concentration levels 6.b) Concentration range 333–1000 ng.c) Concentration range 670–2000 ng.d) Correlation of spectrum at start of peak with spectrum at center of peak.e) Correlation of spectrum at center of peak with spectrum at end of peak.



Table 6. Robustness testing of the HPTLC method

Parameters % RSD of peak area

DHL (A) HHO (B) HLO (C)

Mobile phase composition 0.46 0.86 0.78Chamber saturation time 0.21 0.26 0.26Derivatization time 0.87 0.97 0.76Heating time 1.12 1.21 0.97

Rf values of A, B, and C were 0.74, 0.57, and 0.51, respec-tively. The absence of the interfering or overlapping peaksin the analyzed samples at 610 nm after derivatization withanisaldehyde/acetic acid reagent could be observed, whichindicates that the method was specific. A linear relation-ship was found for the 333–1000 ng/band concentrationrange for A and B and for the 670–2000 ng/band concen-tration range for C. The coefficients of correlation (r) 0.9984,0.9991, and 0.9994 revealed a good linearity response for de-veloped method (Table 5). LOD were in the range of 72.72,55.80, and 92.17 ng/band and LOQ were found to be 242.39,186.00, and 307.24 ng/band for A, B, and C, respectively(Table 5). The results show a high sensitivity for the in-vestigated compounds. The intra- and inter-assay precisionswere in the range of 0.98–1.19% and 1.09–1.27%, respec-tively (Table 5). These results demonstrate that the proposedmethod has a good precision. The results obtained duringrobustness tests indicate that no significant change in Rf

occurred, demonstrating the method to be robust. The ac-curacy was expressed as percentage recovery of the com-pound in the sample. One may see in Table 3 that goodrecoveries, 99.1–102.7%, were obtained for all the test mark-ers. Taking into account that the validation parameters werefound within the recommended limits, the proposed methodmay be used for the quantification of test markers from realsamples.

4 Concluding remarks

The new HPTLC method for the simultaneous quanti-tative determination of recently isolated diterpenoids inV. trifolia extract has been developed and validated. The pro-posed method was found to be simple, precise, and accurateand has the advantage of rapidity and economy. Moreover,the chromatographic patterns evidenced that the extractiontechniques influence only the quantitative results, not thequalitative ones. The developed method demonstrates once

again that using appropriate resources, TLC combined withphotodocumentation and image processing can be a powerfultool.

We are thankful to Director, CIMAP, for keen interest duringthe course of work. One of the authors, Neerja Tiwari, is thankfulto CSIR (New Delhi) for the award of Senior Research Fellowship.

The authors have declared no conflict of interest.

5 References

[1] Chopra, R. N., Nayar, S. L., Chopra, I. C., Glossary ofIndian Medicinal Plants, CSIR, New Delhi 1956.

[2] Ikram, M., Khattak, S. G., Gilani, S. N., J. Ethnopharma-col. 1987, 19, 185–187.

[3] Hossain, M. M., Paul, N., Sohrab, M. H., Rahman, E.,Rashid, M. A., Fitoterapia 2001, 72, 695–697.

[4] Ikwati, Z., Wahyuono, S., Maeyama, K., J. Ethnopharma-col. 2001, 75, 249–256.

[5] Ono, M., Sawamura, H., Ito, Y., Mizuki, K., Nohara, T.,Phytochemistry 2000, 55, 873–877.

[6] Ono, M., Ito, Y., Nohara, T., Chem. Pharm. Bull. 2001, 49,1220–1222.

[7] Ono, M. Ito, Y., Kubo, S., Nohara, T., Chem. Pharm. Bull.1997, 45, 1094–1096.

[8] Gu, Q., Zhang, X. M., Zhou, J., Qiu, S.-X., Chen, J. J., J.Asian Nat. Prod. Res. 2008, 10, 499–502.

[9] Ramesh, P., Nair, A. G., Ramachandran Subramanian, S.S., Fitoterapia 1986, 57, 282–283.

[10] Nair, A. G., Ramesh, P., Subramanian, S. S., Curr. Sci.1975, 44, 214–216.

[11] Lokhande, P. J., Verma, J. K., Acta Chromatogr. 2010, 22,591–597.

[12] Hoberg, E., Meier, B., Sticher, O., Phytochem. Anal. 2000,11, 327–329.

[13] Lokhande, P. J., Verma, J. K., J. Planar Chromatogr.-Mod.TLC 2009, 22, 225–228.

[14] Tiwari, N., Yadav, D., Singh, S. C., Gupta, M. M., J. Liq.Chromatogr. Rel. Technol. 2011, 34, 1925–1937.

[15] Hoberg, E., Meier, B., Sticher, O., Planta Med. 2000, 66,352–355.

[16] Yadav, D., Tiwari, N., Gupta, M. M., J. Sep. Sci. 2011, 34,286–291.

[17] Khan, I., Sangwan, P. L., Dhar, J. K., Koul, S., J. Sep. Sci.2012, 35, 392–399.

[18] Tiwari, N., Thakur, J. P., Saikia, D., Gupta, M. M., Phy-tomedicine 2013, 20, 605–610.


Documents

Validated HPTLC method for the simultaneous quantification of diterpenoids in Vitex trifolia L