Food Chemistry 126 (2011) 1890–1895

Contents lists available at ScienceDirect

Food Chemistry

journal homepage: www.elsevier .com/locate / foodchem

Analytical Methods

Qualitative and quantitative analysis of curcuminoids in herbal medicinesderived from Curcuma species

Rui Li, Cheng Xiang, Min Ye ⇑, Hui-Fang Li, Xing Zhang, De-An Guo ⇑The State Key Laboratory of Natural and Biomimetic Drugs, School of Pharmaceutical Sciences, Peking University, 38 Xueyuan Road, Beijing 100191, PR China

a r t i c l e i n f o

Article history:Received 26 June 2010Received in revised form 1 November 2010Accepted 2 December 2010Available online 10 December 2010

Keywords:CurcuminoidsCurcuma spp.Herbal medicinesHPLC–UV–MS

0308-8146/$ - see front matter � 2010 Elsevier Ltd. Adoi:10.1016/j.foodchem.2010.12.014

⇑ Corresponding authors. Tel.: +86 10 82801516; faE-mail addresses: liruielite@gmail.com (R. Li),

Xiang), yemin@bjmu.edu.cn (M. Ye), honghaier888@na@126.com (X. Zhang), gda5958@163.com (D.-A. Gu

a b s t r a c t

A validated and sensitive HPLC–UV–MS method was developed for qualitative and quantitative analysisof curcuminoids in eight herbal medicines derived from four Curcuma species. The samples were sepa-rated on a YMC ODS-A C18 column with a gradient elution of acetonitrile and 0.1% formic acid. Curcumin,demethoxycurcumin and bisdemethoxycurcumin showed good linearity (r > 0.9998) in the concentrationranges of 4.88–625, 4.29–550 and 3.98–510 lg/mL, respectively. The results suggested that the contentsof three major curcuminoids in different herbal medicines varied significantly. Curcuminoids were onlydetected in Jianghuang, HuangsiYujin, and PengEzhu. Amongst them, Jianghuang contained the highestamounts of curcuminoids (40.36 mg/g), which were almost 20 times higher than HuangsiYujin(1.94 mg/g) and 400 times higher than PengEzhu (0.098 mg/g). Furthermore, amongst the Jianghuangsamples collected from different areas, samples from Sichuan Province contained remarkably higheramounts of curcuminoids (22.21–40.36 mg/g) than other cultivation regions.

� 2010 Elsevier Ltd. All rights reserved.

1. Introduction

Curcuminoids, represented by curcumin, demethoxycurcumin(DMC), and bisdemethoxycurcumin (BDMC), are the major bioactiveconstituents of Curcuma species including turmeric (Curcuma longaL.). Turmeric is a popular herbal medicine, and is also widely usedas food additives. Curcuminoids showed a wide range of biologicalactivities, including anti-inflammatory (Anand et al., 2008; Goel,Kunnumakkara, & Aggarwal, 2008; Kunnumakkara, Anand, & Aggar-wal, 2008), anti-tumorigenesis (Kunnumakkara et al., 2007; Shar-mila, Suthakar, Qinghe, & Rakesh, 2008; Shishodia, Amin, Lai, &Aggarwal, 2005), anti-angiogenesis (Li, Ahmed, Mehta, & Kurzrock,2007; Lin et al., 2007), anti-Alzheimer’s disease (Ishrat et al., 2009;Ma et al., 2009; Yanagisawa et al., 2010), and anti-diabetes (Merrellet al., 2009; Reddy, Sundari, Balamurugan, & Menon, 2009; Sanduret al., 2007). In addition, curcuminoids are non-toxic even at highdose, and exhibit great promise as therapeutic agents. Currently,curcumin is used in clinical trial for the treatment of various tumours(Hatcher, Planalp, Cho, Torti, & Torti, 2008). The analogues of curcu-min including demethoxycurcumin and bisdemethoxycurcuminpossess similar biological activities to curcumin (Jayaprakasha,Rao, & Sakariah, 2006; Sandur et al., 2007).

ll rights reserved.

x: +86 10 82802024.damaocheng@gmail.com (C.sina.com (H.-F. Li), zx10chi-o).

A number of herbal medicines are derived from Curcuma species,including Jianghuang (the rhizomes of C. longa), HuangsiYujin (thetuberous roots of C. longa), PengEZhu (the rhizomes of C. phaeocaulis),LvsiYujin (the tuberous roots of C. phaeocaulis), WenEZhu (therhizomes of C. wenyujin), WenYujin (the tuberous roots of C. wenyu-jin), GuiEzhu (the rhizomes of C. kwangsiensis), and GuiYujin (thetuberous roots of C. kwangsiensis). All the above herbal medicinesare recorded in Chinese Pharmacopoeia (Chinese PharmacopoeiaCommission, 2010). According to the theory of traditional Chinesemedicine (TCM), pharmacological effects and clinical applicationsof these herbal medicines are different. The difference may be corre-lated with the contents of curcuminoids.

A number of methods have been established for the qualitativeand quantitative analysis of Curcuma species, including HPTLC(Zhang et al., 2008), GC–MS (Qin, Yang, Wang, & Li, 2007; Yang,Li, Zhao, Lao, & Wang, 2007), and HPLC (Jadhav, Mahadik, & Parad-kar, 2007; Jayaprakasha, Rao, & Sakariah, 2002; Wisut, Nutthapon,Sunibhond, & Pornchai, 2009). However, few studies have beenconducted to determine curcuminoids in different Curcuma spe-cies. Little is known about the difference in contents of curcumi-noids in different Curcuma species, or in the same species ofsamples collected from different cultivation regions. The aboveinformation is critically important for the quality control of relatedherbal medicines.

In the present study, a sensitive HPLC–UV–MS method wasestablished for the qualitative and quantitative analysis of curc-uminoids in eight herbal medicines derived from four Curcumaspecies. The contents of three major curcuminoids in 40 batches

R. Li et al. / Food Chemistry 126 (2011) 1890–1895 1891

of herbal samples collected around China were determined by thismethod.

2. Experimental

2.1. Materials and reagents

HPLC grade acetonitrile, methanol, and formic acid were fromJ.T. Baker (Phillipsburg, NJ, USA). De-ionised water was preparedby a Milli-Q system (Millipore, MA, USA). The reference com-pounds, curcumin, demethoxycurcumin (DMC), and bisdeme-thoxycurcumin (BDMC) (Fig. 1) were isolated from the rhizomesof C. longa (Jianghuang) by the authors. Their structures were estab-lished by 1H NMR, 13C NMR, and mass spectrometry (Pedersen,Rasmussen, & Lawesson, 1985). Purities of the reference com-pounds were above 98% determined by HPLC.

Commercial herbal materials belonging to eight herbal medi-cines derived from Curcuma species were collected from differentcultivation regions around China in 2007–2009 and were identifiedby the authors (see Table 3). All voucher specimens were depositedin School of Pharmaceutical Sciences, Peking University Health Sci-ence Center.

2.2. HPLC–UV–MS conditions

An Agilent series 1100 HPLC system including a quaternarypump, a variable wavelength detector, an autosampler, and a col-umn compartment was used. The separation was performed on aYMC ODS-A C18 reversed column (250 � 4.6 mm i.d., 5 lm)equipped with an Agilent Zorbax SB-C18 guard column(12.5 � 4.6 mm i.d., 5 lm) at 30 �C. The mobile phase consistedof acetonitrile (CH3CN, A) and 0.1% formic acid in water (B) usinga gradient program of 40–50% (A) in 0–30 min, 50–65% (A) in30–35 min, 65–70% (A) in 35–42 min, 70% (A) in 42–55 min, and70–100% (A) in 55–60 min. The mobile phase flow rate was1.0 mL/min. The UV detector was monitored at 270 nm for finger-printing analysis and 380 nm for quantitative analysis. LC/MS anal-ysis was performed on an LCQ Advantage ion-trap electrosprayionisation mass spectrometer (ThermoFisher, San Jose, CA, USA).

2.3. Preparation of sample solutions

The C. longa (Jianghuang) samples were air-dried and pulverisedinto a fine powder. An aliquot of 0.3 g of the powder was extractedwith 30 mL of methanol in an ultrasonic bath (40 kHz, 300 W) atroom temperature for 60 min. The solution was filtered through

Fig. 1. Chemical structures, and (�)-ESI/MS and MS/MS spectra of three curcuminoids. (

a 0.45 lm membrane and a 15 lL aliquot was injected for HPLCanalysis. For the other herbal samples, the drugs were treated withthe same procedure as described previously, until the sample wasextracted in the ultrasonic bath. Then the methanol solution wasevaporated to dryness, dissolved with an appropriate quantity ofmethanol, and was finally diluted to 2 mL in a volumetric flask.The solution was filtered through a 0.45 lm membrane and a15 lL aliquot was used for HPLC analysis.

2.4. Preparation of standard solutions

The three reference compounds were accurately weighed,mixed, and dissolved in methanol, and then stored at 4 �C beforeHPLC analysis. The concentration of three compounds in stocksolution was 625 lg/mL (curcumin), 550 lg/mL (DMC) and510 lg/mL (BDMC), respectively.

2.5. Fingerprint software analysis

HPLC fingerprint similarity indices were calculated by CASEsoftware (Similarity Evaluation System for Chromatographic Fin-gerprint of Traditional Chinese Medicine, Version 2004A), whichwas recommended by Chinese Pharmacopoeia Commission (Liuet al., 2007).

3. Results and discussion

3.1. Optimisation of extraction method

In order to extract curcuminoids from the herbal samples effi-ciently, variables involved in this procedure were optimised,including extraction solvent (50% methanol, 100% methanol),extraction method (ultrasonication, reflux, percolation), andextraction time (30, 60, 90 min). Due to the water insolubility ofcurcuminoids, 100% methanol was more efficient to extract curc-uminoids from herbal samples than 50% methanol. Then extractionmethods were compared in parallel experiments using 100% meth-anol as the solvent. The ultrasonication showed the greatestextraction efficiency. Furthermore, the extraction time was opti-mised to be 60 min.

3.2. Optimisation of HPLC method

To meet the requirements for quantitative and fingerprint anal-ysis, the following HPLC parameters were examined, including dif-ferent columns (Agilent SB-C18, Waters Atlantis-C18, and YMC ODS-

A) Curcumin; (B) demethoxycurcumin (DMC); (C) bisdemethoxycurcumin (BDMC).

Table 1Regression equations, LOD and LOQ of three curcuminoids.

Analytes Regression equation Dynamic range (lg/mL) r LOD (ng/mL) LOQ (ng/mL)

Curcumin y = 50.9334x + 167.5680 4.88–625 0.9999 18.9 62.5DMC y = 69.8324x + 220.2898 4.29–550 0.9999 16.7 55.0BDMC y = 84.5488x + 300.6608 3.98–510 0.9998 15.4 51.0

Note: In the regression equation y = ax + b, x refers to the concentration of pure curcuminoids (lg/ml); y the peak area; r, the correlation coefficient; LOD, limit of detection;LOQ, limit of quantification.

1892 R. Li et al. / Food Chemistry 126 (2011) 1890–1895

A C18), column temperature (25, 30, and 35 �C), and UV wavelength(200–400 nm). The best chromatographic resolution was obtainedon YMC Pack ODS-A C18 column at 30 �C. The UV detector wasmonitored at 270 nm for fingerprinting analysis because the mostpeaks were observed under this wavelength, and was monitored at380 nm to achieve high sensitivity for curcuminoids.

3.3. Method validation of quantitative analysis

3.3.1. Calibration, the limits of detection and quantificationTo obtain the calibration curve, working solutions of six concen-

trations containing curcumin, DMC, and BDMC were analysed intriplicate. The calibration curves were established by plotting peakareas versus the concentration of each analyte. In the regressionequation y = ax + b, x refers to the concentration of pure curcumi-noids (lg/mL), y the peak area, and r the correlation coefficient.The limit of detection (LOD) and limit of quantification (LOQ) ofcurcuminoids were determined by injecting a series of standardsolutions until the signal-to-noise (S/N) ratio was three for LODand 10 for LOQ, respectively. Calibration curves of the three ana-lytes showed good linearity in relatively wide dynamic ranges (Ta-ble 1).

3.3.2. Precision, reproducibility and stabilityIntra- and inter-day variations were used to determine the pre-

cision. For intra-day test, the samples were analysed for six timeswithin the same day, while for inter-day test, the samples wereexamined twice per day for three consecutive days. Reproducibilitywas evaluated by extracting and analysing five replicates of thesame batch of sample with the established method. For stabilitytest, the same sample solution was analysed every 6 h for 48 h atroom temperature. The RSD of the three analytes of intra- and in-ter-day was less than 1.70%, which demonstrated good precision ofthis method. The method exhibited good reproducibility with RSDless than 3.0% and all analytes were found to be rather stable with-in 48 h with RSD less than 1.03%. Detailed data are available inTable S1 of the Supplementary Material.

3.3.3. AccuracyRecovery test was used to evaluate the accuracy of this method.

Three different concentrations (high, middle, low) of three curc-uminoids were added to approximately 0.15 g of the C. longa

Table 2Recoveries of three curcuminoids. (n = 3).

Analytes Original amount (mg) Spiked (mg)

Curcumin 3.346 0.9083.377 2.4503.348 3.128

DMC 1.391 0.9021.404 1.9261.392 2.812

BDMC 1.325 0.9361.337 1.8521.326 2.856

(Jianghuang) sample C1, and then extracted and analysed as de-scribed for normal samples. The average recoveries were calcu-lated by the formula: recovery (%) = (amount found�originalamount)/amount spiked � 100%. The present quantitative methodhad satisfactory accuracy with the overall recoveries of three ana-lytes ranging from 94.8% to 104.7% (Table 2).

3.4. Qualitative analysis

3.4.1. Identification of curcuminoids in different Curcuma species withHPLC–MS

The established analytical method was applied to identify threemajor curcuminoids (curcumin, DMC, and BDMC) in eight herbalmedicines derived from four Curcuma species. Different sampleswere collected and analysed by LC–MS method. Curcuminoidswere detected in only three herbal medicines, including Jianghu-ang, PengEzhu, and HuangsiYujin. The structures were unambigu-ously identified by comparing with the retention time and MSspectra of reference standards. The (�)-ESI mass spectra gave char-acteristic quasi-molecular ions of curcumin ([M�H]� ion at m/z367), DMC ([M�H]� ion at m/z 337), and BDMC ([M�H]� ion atm/z 307). MS/MS analysis exhibited characteristic fragmentationbehaviours of curcuminoids identical to previous reports: m/z367 ? 217 ? 173 for curcumin; m/z 337 ? 217 ? 173 and m/z337 ? 187 ? 143 for DMC; m/z 307 ? 187 ? 143 for BDMC (Jiang,Somogyi, Jacobsen, Timmermann, & Gang, 2006a; Jiang, Timmer-mann, & Gang, 2006b), as depicted in Fig. 1. In addition, no bisde-methoxycurcumin (BDMC) was detected in PengEzhu, and nocurcuminoids were detected in the other five herbal medicines(Fig. 2).

3.4.2. HPLC fingerprinting analysis of C. longa (Jianghuang) samplescollected from different cultivation regions

A total of 17 batches of Jianghuang samples were collected fromdifferent cultivation regions in China, and their HPLC fingerprintswere obtained. The HPLC method was fully validated by determin-ing the RSD of peak area and retention time of six common peaks.The method exhibited good precision by analysing the same sam-ple solution for six times (RSD of peak area and retention time low-er than 4.47% and 0.23%, respectively). The method reproducibilitywas also satisfactory (RSD of peak area and retention time lowerthan 5.0% and 0.51%, respectively). The sample solutions were sta-

Found (mg) Recovery (%) RSD (%)

0.914 100.6 2.152.550 104.13.131 100.10.868 96.2 2.561.949 101.22.759 98.10.935 99.8 2.091.876 101.32.775 97.2

min5 10 15 20 25 30 35 40 45 50

mAU

0

500

min5 10 15 20 25 30 35 40 45 50

mAU

0

200

400

min5 10 15 20 25 30 35 40 45 50

mAU

0

5

10

min5 10 15 20 25 30 35 40 45 50

mAU

0

min5 10 15 20 25 30 35 40 45 50

mAU

0

min5 10 15 20 25 30 35 40 45 50

mAU

0

20

40

min5 10 15 20 25 30 35 40 45 50

mAU

0

5

10

min5 10 15 20 25 30 35 40 45 50

mAU

0

min5 10 15 20 25 30 35 40 45 50

mAU

0

200

A

B

C

D

E

F

G

H

I C2, tR= 24.4C1, tR= 26.5

C3, tR=22.5

Fig. 2. The HPLC chromatograms (380 nm) of different herbal medicines. (A) C. longa (Jianghuang); (B) C. longa (HuangsiYujin); (C) C. phaeocaulis (PengEzhu); (D) C. phaeocaulis(LvsiYujin); (E) C. kwangsiensis (GuiEzhu); (F) C. kwangsiensis (GuiYujin); (G) C. wenyujin (WenEzhu); (H) C. wenyujin (WenYujin); (I) Mixed standards. C1, curcumin; C2,demethoxycurcumin (DMC); C3, bisdemethoxycurcumin (BDMC).

R. Li et al. / Food Chemistry 126 (2011) 1890–1895 1893

ble within 24 h, with RSD of peak area and retention time lowerthan 2.46% and 0.17%, respectively. Detailed data are given in Ta-bles S2–S4 in the Supplementary Material.

The representative HPLC fingerprints are shown in Fig. 3. P1–P6were selected as common peaks. Amongst them, P1 (BDMC,tR = 22.5 min), P2 (DMC, tR = 24.4 min) and P3 (curcumin,tR = 26.5 min) were unambiguously characterised by comparingwith reference standards. By the CASE software, 12 out of the 17

0 10 20 30 40

mAU

0

50

100

150

200

250

300

P2P1

P3

P

Fig. 3. HPLC chromatograms of 17 batches of C. longa detected at 270 nm. Common peakcurcumin.

batches showed similarity indices of >95%. The above results sug-gested that the chemical compositions of Jianghuang samples fromdifferent cultivation regions were similar.

3.5. Quantitative analysis of curcuminoids

The HPLC quantitative method described above was used todetermine curcuminoids in different Curcuma species. A total of

min50 60 70 80

4P5

P6

12

34

56

78

910

1112

1314

1516

17

s: P1–P6. P1, bisdemethoxycurcumin (BDMC); P2, demethoxycurcumin (DMC); P3,

Table 3Contents (mg/g) of three curcuminoids in eight herbal medicines derived from different Curcuma species.

Samples Source Cultivation regions Curcumin DMC BDMC Total

C1 C. longa (Jianghuang) rhizomes Pengzhou, Sichuan 22.28 9.26 8.82 40.36C2 Leshan, Sichuan 14.04 6.74 3.57 24.35C3 Zhejiang 4.18 1.08 0.40 5.66C4 Guangxi 8.53 3.14 2.00 13.67C5 Jianwei, Sichuan 15.30 4.59 3.55 23.44C6 Yunnan 12.29 4.71 2.76 19.76C7 Xichang, Sichuan 13.58 5.67 2.96 22.21C8 Xinjiang 8.50 2.24 1.70 12.44C9 Yuanan 10.55 3.30 2.77 16.62C10 Chongqing 8.14 2.15 1.60 11.89C11 Wenjiang, Sichuan 19.36 6.01 4.37 29.74C12 Chengdu, Sichuan 14.68 7.10 9.50 31.28C13 Chengdu, Sichuan 15.60 5.93 3.79 25.32C14 Guizhou 5.05 1.34 0.80 7.19C15 Shuangliu, Sichuan 12.88 5.77 4.82 23.47C16 Leshan, Sichuan 17.76 5.55 4.54 27.85C17 Chengdu, Sichuan 19.19 6.10 4.97 30.26H1 C. longa (HuangsiYujin) roots Wenjiang, Sichuan 0.21 0.10 nd 0.31H2 Pengzhou, Sichuan 1.29 0.47 0.18 1.94H3 Leshan, Sichuan 0.12 0.03 nd 0.15H4 Chengdu, Sichuan 1.05 0.30 0.07 1.42H5 Chengdu, Sichuan 1.14 0.33 0.08 1.55P1 C. phaeocaulis (PengEzhu) rhizomes Jianwei, Sichuan 0.026 0.004 nd 0.030P2 Leshan, Sichuan 0.028 0.004 nd 0.032P3 Wenjiang, Sichuan 0.072 0.026 nd 0.098L1, L2, L3 C. phaeocaulis (LvsiYujin) roots Chengdu, Sichuan nd nd nd ndWE1, WE2, WE3 C. wenyujin (WenEzhu) rhizomes Guangxi nd nd nd ndW1, W2, W3 C. wenyujin (WenYujin) roots Wenzhou, Zhejiang nd nd nd ndGE1, GE2, GE3 C. kwangsiensis (GuiEzhu) rhizomes Guangxi nd nd nd ndG1, G2, G3 C. kwangsiensis (GuiYujin) roots Wenzhou, Zhejiang nd nd nd nd

Note: Samples C1–C17, Jianghuang; H1–H5, HuangsiYujin; P1–P3, PengEzhu; L1–L3, LvsiYujin; WE1–WE3, WenEzhu; W1–W3, WenYujin; GE1–GE3, GuiEzhu; G1–G3, GuiYujin;DMC, demethoxycurcumin; BDMC, bisdemethoxycurcumin. nd: not detected.

1894 R. Li et al. / Food Chemistry 126 (2011) 1890–1895

40 batches of different herbal samples were collected and ana-lysed. The data are shown in Table 3. Amongst these herbs, theamounts of curcuminoids varied significantly. Jianghuang con-tained the highest amounts of curcuminoids (sample C1,40.36 mg/g), which was 20-fold higher than HuangsiYujin (sampleH2, 1.94 mg/g) and 400 times higher than that from PengEzhu(sample P1, 0.098 mg/g). Although Jianghuang and HuangsiYujinare derived from rhizomes and roots of the same plant species(C. longa), respectively, the amounts of curcuminoids were signifi-cantly different. Therefore, C. longa could be considered as thedesirable botanical source of curcuminoids.

Amongst different batches of C. longa (Jianghuang) samples, thecontents of curcuminoids also varied remarkably, 4.18–22.28 mg/gfor curcumin, 1.08–9.26 mg/g for DMC, and 0.40–9.50 mg/g forBDMC. Five of the seventeen batches did not meet the requirementof Chinese Pharmacopoeia that curcumin content should not beless than 1% (w/w). However, it is noteworthy that samples culti-vated in Sichuan Province contained the highest amounts of curc-uminoids, with the contents of curcumin ranged from 22.28 to12.88 mg/g (2.23–1.29%). Samples cultivated in Guangxi, Xinjiang,Chongqing, Guizhou and Zhejiang provinces only contained 8.53–4.18 mg/g (0.85–0.42%, w/w) of curcumin. This result was in accor-dance with the traditional Chinese medicine (TCM) theory thatSichuan is the authentic cultivation region for C. longa (Jianghuang).

4. Conclusion

In summary, the qualitative and quantitative analysis of curc-uminoids in eight herbal medicines derived from four Curcumaspecies were conducted for the first time. Our results demon-strated that curcuminoids were only detected in Jianghuang (therhizomes of C. longa), HuangsiYujin (the tuberous roots of C. longa),and PengEzhu (the rhizomes of C. phaeocaulis). Amongst them,

Jianghuang contained the highest amounts of curcuminoids. Fur-thermore, Jianghuang samples collected from Sichuan Provincecontained much higher amounts of curcuminoids than other areas.Therefore, Sichuan Province could be considered as the desirablecultivation region for the production of Jianghuang. The methodestablished in this study could be used for the quality control ofherbal medicines derived from Curcuma species.

Acknowledgement

This research was supported by the 985 project of Peking Uni-versity (Grant No. 985-2-119-121).

Appendix A. Supplementary data

Supplementary data associated with this article can be found, inthe online version, at doi:10.1016/j.foodchem.2010.12.014.

References

Anand, P., Thomas, S. G., Kunnumakkara, A. B., Sundaram, C., Harikumar, K. B., Sung,B., et al. (2008). Biological activities of curcumin and its analogues (congeners)made by man and mother nature. Biochemical Pharmacology, 76, 1590–1611.

Chinese Pharmacopoeia Commission (2010). Pharmacopoeia of the People’s Republicof China (pp. 247–248). Beijing: Chinese Medical Science and Technology Press.

Goel, A., Kunnumakkara, A. B., & Aggarwal, B. B. (2008). Curcumin as ‘‘Curecumin’’:From kitchen to clinic. Biochemical Pharmacology, 75, 787–809.

Hatcher, H., Planalp, R., Cho, J., Torti, F. M., & Torti, S. V. (2008). Curcumin: Fromancient medicine to current clinical trials. Cellular and Molecular Life Sciences, 65,1631–1652.

Ishrat, T., Hoda, M. N., Khan, M. B., Yousuf, S., Ahmad, M., Khan, M. M., et al. (2009).Amelioration of cognitive deficits and neurodegeneration by curcumin in ratmodel of sporadic dementia of Alzheimer’s type (SDAT). EuropeanNeuropsychopharmacology, 19, 636–647.

Jadhav, B. K., Mahadik, K. R., & Paradkar, A. R. (2007). Development and validation ofimproved reversed phase-HPLC method for simultaneous determination of

R. Li et al. / Food Chemistry 126 (2011) 1890–1895 1895

curcumin, demethoxycurcumin, and bis-demethoxycurcumin.Chromatographia, 65, 483–488.

Jayaprakasha, G. K., Rao, L. J. M., & Sakariah, K. K. (2002). Improved HPLC method forthe determination of curcumin, demethoxycurcumin, andbisdemethoxycurcumin. Journal of Agricultural and Food Chemistry, 50,3668–3672.

Jayaprakasha, G. K., Rao, L. J., & Sakariah, K. K. (2006). Antioxidant activities ofcurcumin, demethoxycurcumin and bisdemethoxycurcumin. Food Chemistry,98, 720–724.

Jiang, H., Somogyi, A., Jacobsen, N. E., Timmermann, B. N., & Gang, D. R. (2006a).Analysis of curcuminoids by positive and negative electrospray ionization andtandem mass spectrometry. Rapid Communications in Mass Spectrometry, 20,1001–1012.

Jiang, H., Timmermann, B. N., & Gang, D. R. (2006b). Use of liquid chromatography–electrospray ionization tandem mass spectrometry to identify diarylheptanoidsin turmeric (Curcuma longa L.) rhizome. Journal of Chromatography, 1111, 21–31.

Kunnumakkara, A. B., Anand, P., & Aggarwal, B. B. (2008). Curcumin inhibitsproliferation, invasion, angiogenesis and metastasis of different cancers throughinteraction with multiple cell signaling proteins. Cancer Letters, 269, 199–225.

Kunnumakkara, A. B., Guha, S., Krishnan, S., Diagaradjane, P., Gelovani, J., &Aggarwal, B. B. (2007). Curcumin potentiates antitumor activity of gemcitabinein an orthotopic model of pancreatic cancer through suppression ofproliferation, angiogenesis, and inhibition of nuclear factor-kB-regulated geneproducts. Cancer Research, 67, 3853–3861.

Li, L., Ahmed, B., Mehta, K., & Kurzrock, R. (2007). Liposomal curcumin with andwithout oxaliplatin: Effects on cell growth, apoptosis, and angiogenesis incolorectal cancer. Molecular Cancer Therapeutics, 6, 1276–1282.

Lin, Y. G., Kunnumakkara, A. B., Nair, A., Merritt, W. M., Han, L. Y., Armaiz-Pena, G.N., et al. (2007). Curcumin inhibits tumor growth and angiogenesis in ovariancarcinoma by targeting the nuclear factor-kB pathway. Clinical Cancer Research,13, 3423–3430.

Liu, A., Lin, Y., Yang, M., Guo, H., Guan, S., Sun, J., et al. (2007). Development of thefingerprints for the quality of the roots of Salvia miltiorrhiza and its relatedpreparations by HPLC–DAD and LC–MSn. Journal of Chromatography, B:Analytical Technologies in the Biomedical and Life Sciences, 846, 32–41.

Ma, Q., Yang, F., Rosario, E. R., Ubeda, O. J., Beech, W., Gant, D. J., et al. (2009). Beta -Amyloid oligomers induce phosphorylation of tau and inactivation of insulinreceptor substrate via c-Jun N-terminal kinase signaling: Suppression byomega-3 fatty acids and curcumin. Journal of Neuroscience, 29, 9078–9089.

Merrell, J. G., McLaughlin, S. W., Tie, L., Laurencin, C. T., Chen, A. F., & Nair, L. S.(2009). Curcumin-loaded poly(e-caprolactone) nanofibres: Diabetic wound

dressing with anti-oxidant and anti-inflammatory properties. Clinical andExperimental Pharmacology and Physiology, 36, 1149–1156.

Pedersen, U., Rasmussen, P. B., & Lawesson, S. O. (1985). Synthesis of naturallyoccurring curcuminoids and related compounds. Liebigs Annalen der Chemie, 8,1557–1569.

Qin, N. Y., Yang, F. Q., Wang, Y. T., & Li, S. P. (2007). Quantitative determination ofeight components in rhizome (Jianghuang) and tuberous root (Yujin) of Curcumalonga using pressurized liquid extraction and gas chromatography–massspectrometry. Journal of Pharmaceutical and Biomedical Analysis, 43, 486–492.

Reddy, B. V., Sundari, J. S., Balamurugan, E., & Menon, V. P. (2009). Prevention ofnicotine and streptozotocin treatment induced circulatory oxidative stress bybis-1,7-(2-hydroxyphenyl)-hepta-1,6-diene-3,5-dione in diabetic rats.Molecular and Cellular Biochemistry, 331, 127–133.

Sandur, S. K., Pandey, M. K., Sung, B., Ahn, K. S., Murakami, A., Sethi, G., et al. (2007).Curcumin, demethoxycurcumin, bisdemethoxycurcumin, tetrahydrocurcuminand turmerones differentially regulate anti-inflammatory and anti-proliferativeresponses through a ROS-independent mechanism. Carcinogenesis, 28,1765–1773.

Sharmila, S., Suthakar, G., Qinghe, C., & Rakesh, K. S. (2008). Curcumin sensitizesTRAIL-resistant xenografts: molecular mechanisms of apoptosis, metastasis andangiogenesis. Molecular Cancer, 7, 16.

Shishodia, S., Amin, H. M., Lai, R., & Aggarwal, B. B. (2005). Curcumin(diferuloylmethane) inhibits constitutive NF-kappaB activation, induces G1/Sarrest, suppresses proliferation, and induces apoptosis in mantle celllymphoma. Biochemical Pharmacology, 70, 700–713.

Wisut, W., Nutthapon, J., Sunibhond, P., & Pornchai, R. (2009). A simple isocraticHPLC method for the simultaneous determination of curcuminoids incommercial turmeric extracts. Phytochemical Analysis, 20, 314–319.

Yanagisawa, D., Shirai, N., Amatsubo, T., Taguchi, H., Hirao, K., Urushitani, M., et al.(2010). Relationship between the tautomeric structures of curcumin derivativesand their Abeta-binding activities in the context of therapies for Alzheimer’sdisease. Biomaterials, 31, 4179–4185.

Yang, F. Q., Li, S. P., Zhao, J., Lao, S. C., & Wang, Y. T. (2007). Optimization of GC–MSconditions based on resolution and stability of analytes for simultaneousdetermination of nine sesquiterpenoids in three species of Curcuma rhizomes.Journal of Pharmaceutical and Biomedical Analysis, 43, 73–82.

Zhang, J. S., Guan, J., Yang, F. Q., Liu, H. G., Cheng, X. J., & Li, S. P. (2008). Qualitativeand quantitative analysis of four species of Curcuma rhizomes using twicedevelopment thin layer chromatography. Journal of Pharmaceutical andBiomedical Analysis, 48, 1024–1028.