comparison of the capabilities of accelerated solvent extraction and sonication as extraction...

98 E. WARBURTON ET AL.

© Crown Copyright 2007. Reproduced with the permission of the Controller of HMSO. Phytochem. Anal. 18: 98–102 (2007)Published by John Wiley & Sons, Ltd. DOI: 10.1002.pca

Phytochemical AnalysisPhytochem. Anal. 18: 98–102 (2007)Published online 29 January 2007 in Wiley InterScience (www.interscience.wiley.com) DOI: 10.1002/pca.955

Comparison of the Capabilities of Accelerated SolventExtraction and Sonication as Extraction Techniques for theQuantification of Kavalactones in Piper methysticum (Kava)Roots by High Performance Liquid Chromatographywith Ultra Violet Detection

EMMA WARBURTON,* PAUL L NORRIS and HEIDI GOENAGA-INFANTELGC Ltd, Queens Road, Teddington, Middlesex, TW11 0LY, UK

Received 6 April 2006; Revised 6 October 2006; Accepted 19 October 2006

Abstract: A conventional extraction technique of sonication has been compared, in terms of extraction efficiency, extraction timeand amount of solvent, with the more novel technique of accelerated solvent extraction for the extraction of kavain from the pow-dered roots of Piper methysticum (Kava) with acetone. The extracts were analysed using high-performance liquid chromatographywith ultra violet detection. The effects of varying solvent volume and extraction time upon the quantity of kavain extracted withsonication, and the effects of varying temperature upon the kavain extraction efficiency by ASE, were investigated. ASE was foundto be more efficient with respect to time and solvent volume required; however, a good agreement was found between the kavainconcentration obtained using both extraction techniques. Published by John Wiley & Sons, Ltd.

Keywords: Accelerated solvent extraction; kavain; dihydrokavain; methysticin; dihydromethysticin; yangonin; desmethyoxyyan-gonin; Piper methysticum; Piperaceae.

INTRODUCTION

The Kava plant (Piper methysticum G. Forst.; Pipera-ceae) has been used in herbal medicine over the last2000 years. This plant, which is native to the westernpacific and grows mainly on the Islands of Tonga,Vanuatu, Fiji and Samoa (Wyk and Wink, 2004), isknown for its calming and relaxing properties and hastraditionally been used to treat sleep disturbances,anxiety and stress (Whitton et al., 2003; Wyk andWink, 2004). The kavalactones are known to be theconstituents responsible for the anxiolytic properties ofthe plant, with the highest kavalactone concentrationsfound in the plant roots (Dragull et al., 2003; Loew andFranz, 2003; Whitton et al., 2003; Bilia et al., 2004).However, a number of reported cases of hepatotoxictyhave questioned the safety of kava-containing productsand have led to many countries prohibiting its sale formedicinal purposes (Whitton et al., 2004; Wyk and Wink,2004). Although up to 18 kavalactones have been iden-tified (Lebot et al., 1987), only six of them, including

kavain (1), dihydrokavain (2), desmethoxyyangonin (3),yangonin (4), methysticin (5) and dihydromethysticin(6), have been widely investigated since they are con-sidered to be the constituents most responsible for theanxiolytic properties of this plant (Dragull et al., 2003;Bilia et al., 2004).

Techniques capable of measuring and identifying thekavalactones are required for the characterisation ofP. methysticum in medicinal products. Separation anddetection of kavalactones has been carried out usingtechniques such as GC, HPLC, supercritical fluidchromatography (SFC) and micellar electrokineticchromatography (Bilia et al., 2004). The kind ofprocedure used for the complete extraction of thekavalactones is of paramount importance for theaccurate determination of these constituents.

Numerous techniques have been reported for theextraction of the kavalactones from the P. methysticmplant and include cold percolation with ethyl acetate(Dragull et al., 2003), Soxhlet extraction with ethylacetate (Smith et al., 1984), extraction with carbondioxide (Gaub et al., 2004), shaking overnight (Shaoet al., 1998) and the traditional approach of strainingthe powdered material through muslin cloth (Duveand Prasad, 1984). Sonication with organic solvents(He at al., 1997; Kubatova et al., 2001; Bobeldijk et al.,2005) has become a universal approach because of itssimplicity and the high extraction efficiency achievedfor the kavalactones in comparison with that obtained

PhytochemicalAnalysis

* Correspondence to: E. Warburton, LGC Ltd, Queens Road,Teddington, Middlesex, TW11 0LY, UK.E-mail: [email protected]/grant sponsor: Department of Trade and Industry of theUnited Kingdom as part of the National Measurement System ValidAnalytical Measurement (VAM) program.

QUANTIFICATION OF KAVALACTONES IN PIPER METHYSTICUM (KAVA) 99


with more sophisticated methods. However, the relativelylong extraction time required with this method resultsin a slow sample throughput. Moreover, the largeamount of solvent required for a qualitative extractionof the kavalactones with this method as previouslyreported may hamper detection of minor constituentsowing to analyte over-dilution. Denery et al. (2004)preliminarily examined the effect of a few parametersof accelerated solvent extraction (ASE), an automatedextraction technique that uses less solvent in a shorterperiod of time, on the extraction efficiency of kaval-actones from P. methysticum using methanol at 60°C.The results of studying the effect of extraction cycletime and number of cycles on the quantitative extrac-tion of the six main kavalactones suggested that thereis not a significant effect of these experimental para-meters on the extraction efficiency of the compoundsunder investigation. However, a complete optimisation ofthe extraction process by varying key extraction factorssuch as temperature, solvent nature and extractiontime was not accomplished in this work. Therefore noindication was given about the percentage of kavainthat can be extracted from P. methysticum with thismethod. Owing to the use of P. methysticum plantsfrom different origins, each with differing kavalactoneconcentrations, and the lack of reference materialscertified for kavalactone content, the real efficiency ofthese extraction methods is unknown and the resultsbetween studies cannot be easily compared.

This paper provides the first systematic comparisonof one conventional extraction procedure of sonicationto the alternative ASE technique in terms of extractionefficiency, extraction time and amount of solvent, for thequantification of the kavalactones in P. methysticumroot powder based on HPLC-UV measurements. Theeffect of extraction time and solvent volume upon the

sonication technique, and the effect of temperatureupon ASE for the quantitative extraction of kavainhave been investigated.

EXPERIMENTAL

Instrumentation. An accelerated solvent extractor modelASE 200 from Dionex (Sunnyvale, CA, USA) anda sonication bath (Kerry Ultrasonics, Skipton, UK)were used for the extraction of kavain from the kavaroot powder. HPLC-UV measurements were performedusing either an Agilent Technologies (Warrington, UK)1100 HPLC-UV–vis system equipped with diode arraydetector (DAD) and Chemstation software (Warrington,UK) or a Waters (Elstree, UK) 2695 HPLC-UV–vissystem equipped with a 996 photodiode array detector(PAD) and Millenium32 software (Elstree, UK). Bothinstruments were equipped with a quaternary pump,a vacuum degasser and a refrigerated autosampler.

Reagents, standards and samples. Kavalactone stand-ards were purchased from Aapin Chemicals (Abingdon,UK). For quantitative analysis, a range of calibrationstandards were prepared by serial dilutions of a multi-standard stock solution containing 2 mg/mL of eachkavalactone in acetone. Methanol, acetonitrile, acetone(Fisher Scientific, Loughborough, UK) and isopropanol(Sigma-Aldrich UK Ltd, Poole, UK) were of HPLC grade.De-ionised water (18 MΩ/cm) was obtained from anElga water purification unit (Marlow, UK).

The powdered root of P. methysticum, purchased fromFresh and Wild (Nadi, Fiji, South Pacific), was dried,thoroughly homogenised, sieved (<500 µm) and storedin an airtight secondary container at room temperatureprior to analysis.



Accelerated solvent extraction. Aliquots of P. methysticumroot powder (0.5 g) were accurately weighed into 11 mLextraction cells as described by Denery et al. (2004).The cell content was extracted with acetone using fiverepeat 12 min cycles; this was repeated three timesusing fresh solvent each time (a total time of 180 minper sample). After extraction, the cells were purgedwith nitrogen for 100 s to allow the complete recoveryof the acetone. Samples were extracted at temperaturesof 50, 100 and 150°C at 1500 psi. A minimum of tworeplicates for each extraction temperature were prepared.

Sonication. Aliquots of P. methysticum root powder(0.5 g) were accurately weighed into amber glass vials.In order to study the effect of extraction time upon thequantity of kavain extracted from the root powder withsonication, 15 mL of acetone were added to each vialand the samples were sonicated, at a controlled tempera-ture, for time periods of between 0 and 1245 min. Tostudy the effect of solvent volume upon the quantity ofkavain extracted from the root powder with sonication,5, 15, 25 or 35 mL of acetone were added to each vial,and the samples sonicated at a controlled temperaturefor a time period of 1080 min. After sonication thesamples were removed from the sonic bath, centrifugedand the supernatant removed. A fresh amount of ace-tone was added to the remaining solid of the samplesextracted with difference solvent volumes and the extrac-tion procedure repeated twice. Prior to sonication, allvials were thoroughly sealed to minimise solvent evapora-tion. A minimum of two replicates for each time period/solvent volume were prepared.

Quantification of kavalactones by HPLC-UV. Separationof the kavalactones in the undiluted ASE extracts andsonication supernatants were performed on a ZorbaxC8 (Agilent Technologies, Warrington, UK) column (250× 4.6 mm i.d. 5 µm particle size) at ambient tempera-ture in isocratic mode using a mobile phase composedof water:isopropanol:acetonitrile (65:15:20, v/v/v), eachcontaining 0.1% formic acid, at a flow rate of 0.5 mL/min. Quantitation by UV (detection wavelength 246 nm)was carried out using external calibration and peakarea measurements of the chromatographic signal.The amount of kavalactone is expressed as % w/w ofthe kava root powder (not on a dry weight basis).Precisions for the HPLC-UV system (as %RSD) for themeasurement of each kavalactone were calculated byreplicate injections (n = 4) of a kava root extract andwere found to be 2, 3, 3, 2, 6 and 5% for kavain,dihydrokavain, methysticin, dihydromethysticin, yanoninand desmethoxyyangonin, respectively. The error asso-ciated with the measurement of each kavalactone hasbeen reported as ± %RSD value. Linear calibration(correlation coefficient value >0.99) using a minimum offive data points was achieved over the range 0–2 mg/mLfor all kavalactones. The limit of detection and limit

of quantitation of kavain were found to be 0.04 and0.13 µg/mL, respectively, calculated according to IUPACguidelines. For all calculations the density of acetonewas taken as 785.0 kg/mL.

RESULTS AND DISCUSSION

The separation of the six kavalactones present in theacetone extracts of P. methysticum root obtained usingHPLC-UV as described above was similar to that re-ported by Ganzer and Khan (1999). These conditionswere used to evaluate the extraction efficiency ofkavain by ASE and sonication. Acetone was chosenas the extracting solvent as it has led to the highestrecoveries of kavain from the solid material whenusing sonication as a conventional method of extraction(Kubatova et al., 2001). The effects of the key para-meters were initially studied on the extraction effi-ciency of the six kavalactones under investigation.More detailed data were then obtained for kavain,which is the major component of the extract and forwhich good baseline separation from the other com-ponents in the extract was achieved.

Accelerated solvent extraction

Temperatures of 50, 100 and 150°C extracted a total of1.46 ± 0.03, 1.48 ± 0.03 and 1.48 ± 0.03% w/w kavain,respectively, from the root powder after three consecu-tive extractions with fresh solvent on the same solid.The results presented in Fig. 1 suggest that the extrac-tion temperature does not appear significantly to affectthe total quantity of kavain, or the other kavalactones,when using ASE. Investigation of the proportion ofkavain extracted in each extraction cycle (Table 1) sug-gests that a quantitative extraction of this compoundcan be achieved with only one extraction (extractiontime of 60 min).

Extraction by sonication

The effect of time upon the quantity of kavain extractedfrom P. methysticum root powder by sonication in asingle extraction using 15 mL acetone is illustrated inFig. 2. As can be observed, there was a slight increasein the amount of kavain extracted with increasingextraction time up to 45 min, at which point a plateauwas reached. At this time 1.45 ± 0.03% w/w kavain wasextracted. A similar effect was also observed for theother kavalactones (not shown). In order to guaranteea quantitative extraction of kavain, sonication for atleast 45 min per cycle and three consecutive extractioncycles on the same material (total extraction time of135 min) were found to be required.

QUANTIFICATION OF KAVALACTONES IN PIPER METHYSTICUM (KAVA) 101


Figure 1 Comparison of the amount of kavalactones extracted from Piper methysticum root powder by ASE at different extractiontemperatures.

The effect of solvent volume upon the quantity ofkavain extracted from the P. methysticum root powderby sonication using three consecutive extractions with5, 15, 25 and 35 mL acetone was studied. Sonicationwith 5 and 15 mL acetone extracted in total 1.46 ± 0.03and 1.47 ± 0.03% w/w of kavain from the kava rootpowder respectively. This is comparable to the total

amount of kavain extracted with ASE. Increasing thesolvent volume from 5 to 35 mL did not lead to a signifi-cant increase of the total quantity of kavain extracted,using three extraction cycles. A volume of 5 mL ofacetone was found to be sufficient for the quantitativeextraction of kavain when using 0.5 g of sample andthree extraction cycles (a total volume of 15 mL acetone).

Table 1 Percentage of kavain extracted in each extraction cycle by ASE

Extraction Extraction% of total kavain in the root powder

temperature (°C) sample First extract Second extract Third extract

50 1 99.78 0.13 0.092 98.74 1.17 0.09

100 1 99.80 0.15 0.052 99.85 0.11 0.04

150 1 98.74 1.26 0.002 99.70 0.16 0.14

Figure 2 Effect of extraction time upon the amount of kavalactones extracted from Piper methysticum root powder in a singleextraction by sonication with acetone.



Using acetone volumes of 5 and 15 mL, 91 and 96%respectively of the total amount of kavain extractedusing three extraction cycles was extracted in the firstextract. These results suggest that the extraction effi-ciency in the first cycle could be improved by increas-ing the volume of acetone used, although the totalquantity of kavain extracted from the kava root powderby sonication seems to be independent of the totalsolvent volume used. Acetone volumes >15 mL shouldbe avoided since they may lead to kavain over-dilution,which may hamper the detection of this compoundby UV.

With respect to solvent volume, ASE also offeredsuperior performance over the sonication approach;the amount of solvent required using ASE was reducedby a factor of 2 compared with that required usingsonication. Also, the optimal time required for thequantitative extraction of kavain by sonication is ap-proximately 2.5 times longer than that required usingASE.

AcknowledgementsThe work described in this paper was supportedunder contract with the Department of Trade andIndustry of the United Kingdom as part of the NationalMeasurement System Valid Analytical Measurement(VAM) program.

REFERENCES

Bilia L, Scalise L, Bergonzi MC, Vinciere FF. 2004. Analysis of kavalac-tones from Piper methysticum (kava-kava). J Chromatogr B 812:203–214.

Bobeldijk I, Boonzaaijer G, Spies-Faber EJ, Vaes WHJ. 2005. Deter-mination of kava lactones in food supplements by liquid chromato-graphy atmospheric pressure chemical ionisation tandem massspectrometry. J Chromatogr A 1067: 107–114.

Denery J, Dragull L, Tang CS, Li QX. 2004. Pressurized fluid extractionof carotenoids from Haematococcus pluvialis and Dunaliella salina

and kavalactones from Piper methysticum. Anal Chim Acta 501:175–181.

Dragull K, Yoshida WY, Tang C. 2003. Piperidine alkaloids from Piper

methysticum. Phytochemistry 63: 193–198.Duve RN, Prasad J. 1984. Efficacy of extraction of constituents in the

preparation of yaqona beverage. Part 3: Major active constituents.Fiji Agric J 46: 11–16.

Ganzera M, Khan IJ. 1999. Analytical techniques for the determinationof lactones in Piper methysticm. Chromatographia 50: 654–659.

Gaub M, Brocke A, Roos G, Kovar KA. 2004. High performance liquidchromatography coordination ion spray–mass spectrometry (HPLC-CIS/MS): A new tool for the analysis of non-polar compoundclasses in plant extracts using the example of Piper methysticum

forst. Phytochem Anal 15: 300–305.He X, Lin L, Lian LZ. 1997. Electrospray high performance liquid

chromatograpy mass spectrometry in phytochemical analysis ofkava (Piper methysticum) extract. Planta Med 63: 70–74.

Kubatova A, Miller D, Hawthorne SD. 2001. Comparison of subcriticalwater and organic solvents for extracting kava lactones from kavaroot. J Chromatogr A 923: 187–194.

Lebot V, Merlin M, Lindstrom L. 1987. Kava, the Pacific Drug. YaleUniversity Press: New Haven.

Loew D, Franz G. 2003. Quality aspects of traditional and industrialkava-extracts. Phytomedicine 10: 610–612.

Shao Y, He K, Zheng B, Zheng Q. 1998. Reversed phase high perfor-mance liquid chromatographic methods for quantitative analysis ofthe six major kavalactones in Piper methysticum. J Chromatogr A

825: 1–8.Smith RM, Thakrar H, Arowolo TA, Shafi AA. 1984. High-performance

liquid chromatography of kava lactones from Piper methysticum.J Chromatogr 283: 303–308.

Whitton PA, Lau A, Salisbury A, Whitehouse J, Evans CS. 2003. Kavalactones and the kava-kava controversy. Phytochemistry 64: 673–679.

Wyk B, Wink M, 2004. Medicinal Plants of the World, 1st edn. Briza:South Africa.

comparison of the capabilities of accelerated solvent extraction and sonication as extraction...

Documents