Profi ling Hoodia Extracts by HPLC with Charged Aerosol Detection, Electrochemical Array Detection, and Principal Component Analysis
←
→
Page content transcription
If your browser does not render page correctly, please read the page content below
Profiling Hoodia Extracts by HPLC with Charged Aerosol Detection, Electrochemical Array Detection, and Principal Component Analysis Ian Acworth, Bruce Bailey, Marc Plante, Qi Zhang, and David Thomas Thermo Fisher Scientific, Chelmsford, MA, USA
Thermo Fisher Scientific, Chelmsford, MA Overview Methods Purpose: To combine fast and sensitive HPLC methods with chemometric analysis HPLC with Charged Aerosol Detectio to profile botanical extracts from the botanical supplement Hoodia gordonii. HPLC System: Thermo Scientific Methods: Hoodia extracts were analyzed by gradient reversed-phase high - DAD–3000RS D performance liquid chromatography with either a charged aerosol detector or - Corona™ ultra R photodiode array detector in series with an electrochemical array detector (the Nebulizer Temper spectro-electro array). Chromatographic data were interrogated by chemometric Power function: 1 software to identify patterns related to differences in plant species, origin or processing. Data collection ra Results: Unsupervised chemometric evaluation of the information rich data obtained Data Analysis: Thermo Scientific from HPLC with charged aerosol detection and electrochemical array detection Data System softw correctly identified differences in sample origin and processing. Column: Thermo Scientific Introduction Temp.: 2.6 µm 40 oC Injection Vol.: 10 µL Hoodigosides are oxypregnane steroidal glycosides abundant in Hoodia gordonii Flow Rate: 2.0 mL/min and related plants native to the deserts of southwestern Africa. These plants, used Mobile Phase A: Deionized water traditionally to ease hunger during long hunting expeditions, enjoy wide use today Mobile Phase B: Acetonitrile, Fishe in dietary supplements purported to aid in appetite suppression and weight loss. Gradient: 40%B to 47%B in Presented are two approaches to profile Hoodia related products. In the first approach, HPLC with Spectro-Electro Array De Hoodia extracts were analyzed by HPLC with charged aerosol detection. Eight hoodigosides isolated from dried plant material are separated within 15 min on a solid HPLC System: Thermo Scientific core C8 analytical column that delivers superb resolution with low backpressure. With - DAD–3000RS D low-nanogram sensitivity, the charged aerosol detector responds uniformly to all non- - CoulArray™ Ele volatile species including the target analytes, degradation products and impurities that 16-channel array may not possess a chromophore. Data Analysis: Thermo Scientific In the second approach, Hoodia extracts were analyzed by HPLC with an online System 6.8 and C electrochemical array detector that responds to redox active compounds such as polyphenols. Each of the sixteen sensors in the electrochemical array detects analytes Column: Thermo Scientific at a unique potential to yield picogram on-column sensitivity for diverse compounds 3 µm that vary widely in redox properties. The chromatograms resulting from this Temperature: 35 oC voltammetric approach often clearly resolve co-eluting compounds. Flow Rate: 0.65 mL/min Injection Vol.: 10 µL These two techniques provide complementary data that are used to quantify known Mobile Phase A: 20 mM Monobasic compounds. The complex patterns of known and unknown compounds revealed by ` 0.2% tetrahydrofu the detectors were also interrogated by pattern recognition software to support Mobile Phase B: 20 mM Monobasic inferences on product quality, authenticity, adulteration, or origin. Limits of detection, 10% tetrahydrofur linear and dynamic range, precision, and pattern recognition results are compared Mobile Phase C: Methanol, Optima for the two approaches. Gradient: FIGURE 1. HPLC with charged aerosol detection identifies hoodigosides in a Time Flow %A standard with precision and sensitivity; overlay of seven injections, 10 mg/L (min) (mL/min) 0 0.65 92 2 0.65 92 30 0.65 0 45 0.65 0 45 0.65 92 50 0.65 92 Sample Preparation 50 mg powder from a commercially ava tube. One 1 mL methanol was added a supernatant was transferred to a 5 mL two times with fresh methanol and the with methanol and mixing, an aliquot w 13,000 rpm for 10 min and the superna Data Analysis Principal components analysis (PCA) a pattern differences in chromatograms r time. Data were aligned by using the P Windows 3.0 software and imported int modeling software version 4.5 for PCA 2 Profiling Hoodia Extracts by HPLC with Charged Aerosol Detection, Electrochemical Array Detection, and Principal Component Analysis
tific, Chelmsford, MA Methods Results methods with chemometric analysis HPLC with Charged Aerosol Detection Chromatography pplement Hoodia gordonii. Method 1: HPLC-Charged aerosol detection HPLC System: Thermo Scientific™ Dionex™ UltiMate™ 3000 RSLC system with a: dient reversed-phase high - DAD–3000RS Diode Array Detector and a A mixture of eight hoodigosides eluted with bas charged aerosol detector or - Corona™ ultra RS™ Charged Aerosol Detector: the Accucore C8 column under an aqueous/solv ochemical array detector (the Nebulizer Temperature: 25 °C acetonitrile over 15 min (Figure 1). After separa re interrogated by chemometric Power function: 1.00 washed for an additional 10 min with 95% aceto s in plant species, origin or processing. Data collection rate: 20 Hz hydrophobic compounds that might be present i of the information rich data obtained Data Analysis: Thermo Scientific™ Dionex™ Chromeleon™ Chromatography ample response can be seen in this overlay of s lectrochemical array detection Data System software, 7.1 SR1 and diluent blank. nd processing. The chromatograms obtained from sample extra Column: Thermo Scientific™ Accucore™ C8, 4.6 × 150 mm; 2.6 µm peaks eluting before, during and after the retent Temp.: 40 oC standards. Figure 2 shows example chromatogr Injection Vol.: 10 µL Hoodia gordonii mixed parts. In most cases, the des abundant in Hoodia gordonii Flow Rate: 2.0 mL/min sufficiently to permit their accurate identification western Africa. These plants, used Mobile Phase A: Deionized water mixtures. Some of the extracts included peaks xpeditions, enjoy wide use today Mobile Phase B: Acetonitrile, Fisher Scientific™ Optima™ LCMS hoodigosides. Note that in many of the extracts e suppression and weight loss. Gradient: 40%B to 47%B in 2.5 min; to 95%B in 15 min; hold 10 min. compounds that eluted between 15 and 25 min related products. In the first approach, but were well detected by the charged aerosol d HPLC with Spectro-Electro Array Detection obtained from standard injections ranging from arged aerosol detection. Eight re separated within 15 min on a solid The calibration curves spanned four orders of m HPLC System: Thermo Scientific™ Dionex™ UltiMate™ 3000 RS system with a data and yielded a correlation coefficient of > 0. solution with low backpressure. With - DAD–3000RS Diode Array Detector tector responds uniformly to all non- - CoulArray™ Electrochemical Array Detector: FIGURE 2. HPLC-CAD chromatograms of me radation products and impurities that 16-channel array from 0 to +900 mV versus Pd in +60 mV steps gordonii, Hoodia gordonii (from Namibia), an Data Analysis: Thermo Scientific™ Dionex™ Chromeleon™ Chromatography Data nalyzed by HPLC with an online System 6.8 and CoulArray 3.0 software. edox active compounds such as electrochemical array detects analytes Column: Thermo Scientific™ Acclaim™ C18, 3 × 150 mm; n sensitivity for diverse compounds 3 µm ograms resulting from this Temperature: 35 oC uting compounds. Flow Rate: 0.65 mL/min Injection Vol.: 10 µL ata that are used to quantify known Mobile Phase A: 20 mM Monobasic sodium phosphate, 3% acetonitrile, unknown compounds revealed by ` 0.2% tetrahydrofuran, pH 3.35 ecognition software to support Mobile Phase B: 20 mM Monobasic sodium phosphate, 50% acetonitrile, ration, or origin. Limits of detection, 10% tetrahydrofuran, pH 3.45 recognition results are compared Mobile Phase C: Methanol, Optima LCMS Gradient: on identifies hoodigosides in a Time Flow %A %B %C Curve ay of seven injections, 10 mg/L (min) (mL/min) 0 0.65 92 5 3 5 2 0.65 92 5 3 7 (concave) 30 0.65 0 97 3 7 (concave) FIGURE 3. PCA score plot of CAD data, only for PCA analysis 45 0.65 0 97 3 5 45 0.65 92 5 3 5 Factor 2 (8.2%) 50 0.65 92 5 3 5 3126 Sample Preparation 50 mg powder from a commercially available capsule was placed into a 2 mL centrifuge Factor 3 (4.6%) 2925 tube. One 1 mL methanol was added and the mixture vortexed for 10 min. The HG4 supernatant was transferred to a 5 mL volumetric flask. Extraction was repeated two times with fresh methanol and the extracts combined. After bringing to volume HG5 3558 2915 HG6 HG1 with methanol and mixing, an aliquot was centrifuged in a 1 mL centrifuge tube at 2821 2927 13,000 rpm for 10 min and the supernatant transferred to a glass HPLC autosampler vial. Data Analysis Figure 3 shows the PCA score plot of the hoodi Principal components analysis (PCA) and cluster analysis (HCA) were used to investigate aerosol detection. Capsule samples HG1 and H pattern differences in chromatograms represented as detector response versus retention and sample vial 2915 containing Opuntia leptoc time. Data were aligned by using the Pattern Recognition Wizard in CoulArray for primary difference between these clusters is ide Windows 3.0 software and imported into Pirouette® Comprehensive Chemometrics (A description of the pattern recognition analysis modeling software version 4.5 for PCA and HCA. Thermo Scientific Poster Note • PN70540_E 03/13S 3
Results Method 2: HPLC-Spectro-Electro Array Chromatographic results: n Chromatography The Spectro-Electro Array makes use o Method 1: HPLC-Charged aerosol detection data. While UV detection provides iden ™ Dionex™ UltiMate™ 3000 RSLC system with a: ode Array Detector and a A mixture of eight hoodigosides eluted with baseline resolution within 10 min from components in a sample, EC array det S™ Charged Aerosol Detector: the Accucore C8 column under an aqueous/solvent gradient ranging from 40–95% measuring compounds missed by UV. ture: 25 °C acetonitrile over 15 min (Figure 1). After separation was achieved, the column was resolve compounds that co-elute chrom 00 washed for an additional 10 min with 95% acetonitrile to ensure complete elution of compound reacting across the array pr e: 20 Hz hydrophobic compounds that might be present in sample extracts. Good precision and for analyte identification/authentication. ample response can be seen in this overlay of seven injections of a 10 mg/L standard generating chemical fingerprints of sup ™ Dionex™ Chromeleon™ Chromatography and diluent blank. compounds that are essential to flavor, are, 7.1 SR1 the Spectro-Electro Array was optimize The chromatograms obtained from sample extracts are very complex, with many components in the samples. Illustrative ™ Accucore™ C8, 4.6 × 150 mm; peaks eluting before, during and after the retention time window of the hoodigoside obtained for a few of the samples analy standards. Figure 2 shows example chromatograms obtained from extracts of dried these extracts contain many species th Hoodia gordonii mixed parts. In most cases, the individual hoodigosides were resolved sufficiently to permit their accurate identification and quantification in the sample mixtures. Some of the extracts included peaks that coeluted with the standard FIGURE 3: HPLC-ECD Array chroma Scientific™ Optima™ LCMS hoodigosides. Note that in many of the extracts there were several hydrophobic extract #3165 2.5 min; to 95%B in 15 min; hold 10 min. compounds that eluted between 15 and 25 min that are not well detected at 220 nm, 1.40 but were well detected by the charged aerosol detector. Calibration curves were 4-hydroxybenzaldehyde 9.76m 4-hydroxybenzyl alcohol 3.87m 1.20 ection obtained from standard injections ranging from 0.10 to 100 mg/L (not shown). salicylic acid 15.3m 1.00 valillic acid 8.93m Response (µA) The calibration curves spanned four orders of magnitude, had a quadratic fit to the [900 mV] syringealdehyde 11.6m 0.80 [840 mV] vanillin 11.0m [780 mV] ™ Dionex™ UltiMate™ 3000 RS system with a [720 mV] syringic acid 9.22m data and yielded a correlation coefficient of > 0.999 for each standard hoodigoside. [660 mV] ferulic acid 14.9m sinapic acid 14.2m 0.60 [600 mV] [540 mV] ode Array Detector [480 mV] 0.40 [420 mV] [360 mV] [300 mV] trochemical Array Detector: FIGURE 2. HPLC-CAD chromatograms of methanolic extracts from Hoodia [240 mV] 0.20 [180 mV] [120 mV] [60 mV] om 0 to +900 mV versus Pd in +60 mV steps gordonii, Hoodia gordonii (from Namibia), and Opuntia leptocaulis 0.00 [0 mV] 0.00 10.00 ™ Dionex™ Chromeleon™ Chromatography Data oulArray 3.0 software. ™ Acclaim™ C18, 3 × 150 mm; FIGURE 4: HPLC-ECD Array chroma capsule extract HG4 1.40 4-hydroxybenzyl alcohol 3.75m 1.20 sodium phosphate, 3% acetonitrile, 1.00 valillic acid 8.92m Response (µA) [900 mV] an, pH 3.35 syringealdehyde 11.6m 0.80 [840 mV] vanillin 11.0m [780 mV] [720 mV] syringic acid 9.21m [660 mV] ferulic acid 14.9m sodium phosphate, 50% acetonitrile, 0.60 sinapic acid 14.4m [600 mV] [540 mV] [480 mV] 0.40 [420 mV] n, pH 3.45 [360 mV] [300 mV] [240 mV] 0.20 [180 mV] LCMS [120 mV] [60 mV] 0.00 [0 mV] 0.00 10.00 FIGURE 5: HPLC-ECD Array chroma %B %C Curve tablet extract HG6 4-hydroxybenzyl alcohol 3.75m 5 3 5 4-hydroxybenzaldehyde 1.40 hydroxyphenyl acetic acid 7.79m 1.20 5 3 7 (concave) salicylic acid 14.6m 1.00 FIGURE 3. PCA score plot of CAD data, only hoodigoside response were used Response (µA) 97 3 7 (concave) 3,4-dihydroxybenzoic acid 4.54m [900 mV] 0.80 [840 mV] valillic acid 8.94m [780 mV] for PCA analysis syringic acid 9.23m [720 mV] [660 mV] ferulic acid 15.0m 0.60 sinapic acid 14.1m [600 mV] 97 3 5 [540 mV] [480 mV] catechin 8.11m MPG-D046 0.40 [420 mV] [360 mV] [300 mV] 5 3 5 Factor 2 (8.2%) [240 mV] 0.20 [180 mV] [120 mV] [60 mV] 0.00 [0 mV] 5 3 5 0.00 10.00 3126 Pattern Recognition Analysis lable capsule was placed into a 2 mL centrifuge Factor 3 (4.6%) 2925 2926 Factor 1 (85.5%) Chromatographic data from either the c d the mixture vortexed for 10 min. The HG4 2799 array detector were pre-processed to a olumetric flask. Extraction was repeated recognition wizard in CoulArray 3.0 sof xtracts combined. After bringing to volume HG5 3558 2915 HG6 HG1 for PCA and HCA. PCA is a mathemati s centrifuged in a 1 mL centrifuge tube at 2821 2927 transformation to convert a set of obse ant transferred to a glass HPLC autosampler vial. 3165 set of values of linearly uncorrelated va 3229 used as a tool in exploratory data analy Figure 3 shows the PCA score plot of the hoodigosides data acquired by charged greatest variance by any projection of t d cluster analysis (HCA) were used to investigate aerosol detection. Capsule samples HG1 and HG5, tablet samples HG-6 and 3558, (Factor 1), the second greatest varianc presented as detector response versus retention and sample vial 2915 containing Opuntia leptocaulis form a distinct cluster. The and so on. Thus samples with different ttern Recognition Wizard in CoulArray for primary difference between these clusters is identified as hoodigoside content. electrochemical profile are distinguishe Pirouette® Comprehensive Chemometrics (A description of the pattern recognition analysis is provided below.) position in three- dimensional space. T and HCA. to investigate authenticity and differenc or capsule. 4 Profiling Hoodia Extracts by HPLC with Charged Aerosol Detection, Electrochemical Array Detection, and Principal Component Analysis
Method 2: HPLC-Spectro-Electro Array. FIGURE 6. PCA score plot for ECD array dat Chromatographic results: The Spectro-Electro Array makes use of both spectrophotometric and electrochemical data. While UV detection provides identification and quantification of the major eline resolution within 10 min from components in a sample, EC array detection is incredibly sensitive and is capable of Factor 2 (10.8%) ent gradient ranging from 40–95% measuring compounds missed by UV. Furthermore, the EC array can voltammatrically 3229 ion was achieved, the column was resolve compounds that co-elute chromatographically, and the redox behavior of a 2915 nitrile to ensure complete elution of compound reacting across the array provides qualitative information that can be used for analyte identification/authentication. The CoulArray is particularly suitable for 29 HG5 n sample extracts. Good precision and even injections of a 10 mg/L standard generating chemical fingerprints of supplements containing endogenous electroactive compounds that are essential to flavor, color and stability. The HPLC method used with 3558 the Spectro-Electro Array was optimized to resolve the earlier-eluting more polar Factor 3 (6.3%) acts are very complex, with many components in the samples. Illustrative examples of HPLC-ECD Array chromatograms ion time window of the hoodigoside obtained for a few of the samples analyzed are shown in Figure 3–5. As can be seen, ams obtained from extracts of dried these extracts contain many species that are electrochemically active. individual hoodigosides were resolved and quantification in the sample that coeluted with the standard FIGURE 3: HPLC-ECD Array chromatogram of Hoodia gordonii mixed parts HG1 HG6 there were several hydrophobic extract #3165 that are not well detected at 220 nm, 1.40 etector. Calibration curves were 4-hydroxybenzaldehyde 9.76m A score plot from the HPLC-Charged aerosol d 4-hydroxybenzyl alcohol 3.87m 1.20 0.10 to 100 mg/L (not shown). salicylic acid 15.3m 1.00 discussed in that section. A score plot from PC luteolin 25.8m valillic acid 8.93m Response (µA) agnitude, had a quadratic fit to the [900 mV] syringealdehyde 11.6m ethyl vanillin 16.2m 0.80 [840 mV] data is shown in Figure 6. Samples 2821, 2799 cavarcrol 31.9m vanillin 11.0m [780 mV] [720 mV] syringic acid 9.22m 999 for each standard hoodigoside. [660 mV] ferulic acid 14.9m sinapic acid 14.2m 0.60 [600 mV] eugenol 29.1m PCA analysis because of problems with the chr [540 mV] [480 mV] 0.40 [420 mV] [360 mV] major groups among the remaining samples. O [300 mV] thanolic extracts from Hoodia [240 mV] 0.20 [180 mV] [120 mV] the non-hoodia species Opuntia leptocaulis. An [60 mV] 0.00 nd Opuntia leptocaulis [0 mV] 0.00 10.00 20.00 Retention time (minutes) 30.00 processed into capsules or tablets. The third se to authentic Hoodia gordonii dried plant materia unsupervised chemometric analysis correspon FIGURE 4: HPLC-ECD Array chromatogram of Hoodia commercial product the samples, allowing us to verify the accuracy capsule extract HG4 combined chromatographic/chemometric appro more or less automated way to quickly screen Conclusion 1.40 4-hydroxycoumarin 18.8m 4-hydroxybenzyl alcohol 3.75m 1.20 1.00 valillic acid 8.92m Response (µA) • Gradient HPLC with charged aerosol or ele [900 mV] syringealdehyde 11.6m ethyl vanillin 16.8m 0.80 [840 mV] vanillin 11.0m 32.1m [780 mV] 31.9m [720 mV] syringic acid 9.21m [660 mV] ferulic acid 14.9m approaches that generate information-rich 0.60 sinapic acid 14.4m luteolin 25.8m cavarcrol [600 mV] eugenol 29.1m thymol [540 mV] [480 mV] 0.40 [420 mV] reveal analytes overlooked by UV detection quercetin 26.4m [360 mV] [300 mV] [240 mV] 0.20 [180 mV] [120 mV] [60 mV] 0.00 [0 mV] 0.00 10.00 20.00 30.00 Retention time (minutes) • Chemical profile data can be imported into principal components analysis used to read FIGURE 5: HPLC-ECD Array chromatogram of Hoodia commercial product tablet extract HG6 • The combined chromatographic/chemomet 4-hydroxybenzyl alcohol 3.75m to known test samples, but can be readily a 4-hydroxybenzaldehyde 1.40 4-hydroxycoumarin 18.3m samples. hydroxyphenyl acetic acid 7.79m 1.20 salicylic acid 14.6m 1.00 hoodigoside response were used References Response (µA) 3,4-dihydroxybenzoic acid 4.54m [900 mV] 0.80 [840 mV] valillic acid 8.94m [780 mV] hesperidin 18.7m syringic acid 9.23m [720 mV] [660 mV] ferulic acid 15.0m 0.60 sinapic acid 14.1m [600 mV] [540 mV] [480 mV] catechin 8.11m 0.40 [420 mV] 1. Avula, B.; Wang, Y.-H.; Pawar, R.S.; Shu MPG-D046 [360 mV] [300 mV] [240 mV] 0.20 [180 mV] A rapid method for chemical fingerprint a [120 mV] [60 mV] 0.00 [0 mV] 0.00 10.00 20.00 30.00 genera, and dietary supplements using U Retention time (minutes) 2008, 48, 722–731. Pattern Recognition Analysis 2. Gamache, P.; Acworth, I.; Lynch, M.; and Detection for HPLC In the analysis of Jui 2926 Factor 1 (85.5%) Chromatographic data from either the charged aerosol detector or the Spectro-Electro Adulteration of Fruit Juice Beverages; Na 2799 array detector were pre-processed to align peak retention times by using the pattern USA, Inc.1995; Vol. 1, pp 120-144. recognition wizard in CoulArray 3.0 software and then imported into Pirouette software for PCA and HCA. PCA is a mathematical procedure that uses an orthogonal transformation to convert a set of observations of possibly correlated variables into a Acknowledgements set of values of linearly uncorrelated variables called principal components. It is mostly 3165 We would like to thank Drs. Ikhlas Khan and Bh 3229 used as a tool in exploratory data analysis and for making predictive models. The Natural Products Research at The University o gosides data acquired by charged greatest variance by any projection of the data is found in the first coordinate the plant materials and hoodigoside standards G5, tablet samples HG-6 and 3558, (Factor 1), the second greatest variance is found on the second coordinate (Factor 2), aulis form a distinct cluster. The and so on. Thus samples with different patterns in their chromatographic or © 2013 Thermo Fisher Scientific Inc. All rights reserved. Piro ntified as hoodigoside content. electrochemical profile are distinguished from each other based on their relative All other trademarks are the property of Thermo Fisher Scien is provided below.) position in three- dimensional space. The relationships among these data were used intended to encourage use of these products in any manners others. to investigate authenticity and differences in processing, e.g., dried plant parts, tablet, or capsule. Thermo Scientific Poster Note • PN70540_E 03/13S 5
FIGURE 6. PCA score plot for ECD array data from channels 1–16. both spectrophotometric and electrochemical ication and quantification of the major ction is incredibly sensitive and is capable of Factor 2 (10.8%) 3165 urthermore, the EC array can voltammatrically 3229 atographically, and the redox behavior of a 2915 vides qualitative information that can be used The CoulArray is particularly suitable for 2925 HG5 ements containing endogenous electroactive MPG olor and stability. The HPLC method used with 3558 2926 Factor 1 (72.7%) to resolve the earlier-eluting more polar Factor 3 (6.3%) 3126 2927 examples of HPLC-ECD Array chromatograms ed are shown in Figure 3–5. As can be seen, t are electrochemically active. ogram of Hoodia gordonii mixed parts HG1 HG6 A score plot from the HPLC-Charged aerosol detection data is shown in Figure 3 and discussed in that section. A score plot from PCA analysis of the ECD array (CoulArray) luteolin 25.8m data is shown in Figure 6. Samples 2821, 2799 and HG4 were excluded as outliers before cavarcrol 31.9m eugenol 29.1m PCA analysis because of problems with the chromatography. PCA readily identifies three major groups among the remaining samples. One group contains only sample 2915, from the non-hoodia species Opuntia leptocaulis. Another group corresponds to the samples 20.00 Retention time (minutes) 30.00 processed into capsules or tablets. The third set, within the black circle, corresponds to authentic Hoodia gordonii dried plant material. In each case, the results from the unsupervised chemometric analysis correspond to plainly observable characteristics of ogram of Hoodia commercial product the samples, allowing us to verify the accuracy of the approach. The real power of this combined chromatographic/chemometric approach is that it can be implemented in a more or less automated way to quickly screen unknown samples. Conclusion 4-hydroxycoumarin 18.8m • Gradient HPLC with charged aerosol or electrochemical array detection are simple 32.1m thymol 31.9m approaches that generate information-rich chemical profile data. Both detectors can luteolin 25.8m cavarcrol eugenol 29.1m reveal analytes overlooked by UV detection alone. quercetin 26.4m 20.00 30.00 Retention time (minutes) • Chemical profile data can be imported into pattern recognition software and with principal components analysis used to readily identify product authenticity. ogram of Hoodia commercial product • The combined chromatographic/chemometric approach was verified by applying it to known test samples, but can be readily applied to quickly screen unknown 4-hydroxycoumarin 18.3m samples. References hesperidin 18.7m 1. Avula, B.; Wang, Y.-H.; Pawar, R.S.; Shukla, Y.J.; Smillie, T.J.; Khan, I.A. A rapid method for chemical fingerprint analysis of Hoodia species, related 20.00 30.00 genera, and dietary supplements using UPLC–UV–MS J. Pharm. Biomed. Anal. Retention time (minutes) 2008, 48, 722–731. 2. Gamache, P.; Acworth, I.; Lynch, M.; and Matson, W. Coulometric Array Detection for HPLC In the analysis of Juice Products in Methods to Detect arged aerosol detector or the Spectro-Electro Adulteration of Fruit Juice Beverages; Nagy, S., Wade, R., Eds.; AGSCIENCE gn peak retention times by using the pattern USA, Inc.1995; Vol. 1, pp 120-144. ware and then imported into Pirouette software al procedure that uses an orthogonal ations of possibly correlated variables into a Acknowledgements ables called principal components. It is mostly We would like to thank Drs. Ikhlas Khan and Bharathi Avula of the National Center for is and for making predictive models. The Natural Products Research at The University of Mississippi, Oxford, MS, for providing e data is found in the first coordinate the plant materials and hoodigoside standards used in this study. is found on the second coordinate (Factor 2), atterns in their chromatographic or © 2013 Thermo Fisher Scientific Inc. All rights reserved. Pirouette is a registered trademark of Informetrix, Inc. from each other based on their relative All other trademarks are the property of Thermo Fisher Scientific and its subsidiaries. This information is not e relationships among these data were used intended to encourage use of these products in any manners that might infringe the intellectual property rights of others. s in processing, e.g., dried plant parts, tablet, PO70540_E 03/13S 6 Profiling Hoodia Extracts by HPLC with Charged Aerosol Detection, Electrochemical Array Detection, and Principal Component Analysis
www.thermoscientific.com/dionex ©2013 Thermo Fisher Scientific Inc. All rights reserved. ISO is a trademark of the International Standards Organization. Pirouette is a registered trademark of Informetrix, Inc. All other trademarks are the property of Thermo Fisher Scientific Inc. and its subsidiaries. This information is presented as an example of the capabilities of Thermo Fisher Scientific Inc. products. It is not intended to encourage use of these products in any manners that might infringe the intellectual property rights of Thermo Fisher Scientific, Sunnyvale, CA others. Specifications, terms and pricing are subject to change. Not all products are available in all countries. Please consult USA is ISO 9001:2008 Certified. your local sales representative for details. Australia +61 3 9757 4486 Denmark +45 70 23 62 60 Japan +81 6 6885 1213 Switzerland +41 62 205 9966 Austria +43 1 333 50 34 0 France +33 1 60 92 48 00 Korea +82 2 3420 8600 Taiwan +886 2 8751 6655 Belgium +32 53 73 42 41 Germany +49 6126 991 0 Netherlands +31 76 579 55 55 UK/Ireland +44 1442 233555 Brazil +55 11 3731 5140 India +91 22 2764 2735 Singapore +65 6289 1190 USA and Canada +847 295 7500 China +852 2428 3282 Italy +39 02 51 62 1267 Sweden +46 8 473 3380 PN70540_E 03/13S
You can also read