u n i ve r s i t y o f co pe n h ag e n

Københavns Universitet

Non-biased investigation of standardized Ginkgo biloba preparations by 1H NMR-based metabolomics and HPLC-PDA-MS-SPE-NMRAgnolet, Sara; Jaroszewski, Jerzy W.; Verpoorte, Robert; Stærk, Dan

Publication date:2009

Document versionPublisher's PDF, also known as Version of record

Citation for published version (APA):Agnolet, S., Jaroszewski, J. W., Verpoorte, R., & Stærk, D. (2009). Non-biased investigation of standardizedGinkgo biloba preparations by

1H NMR-based metabolomics and HPLC-PDA-MS-SPE-NMR. Poster session

presented at ULLA Summerschool, Copenhagen, Denmark.

Download date: 28. jan.. 2020

F A C U L T Y O F P H A R M A C E U T I C A L S C I E N C E S

U N I V E R S I T Y O F C O P E N H A G E N

Non-biased investigation of standardized Ginkgo biloba preparations by 1H NMR-Based Metabolomics and HPLC-PDA-MS-SPE-NMR

Sara Agnolet1, Jerzy W. Jaroszewski1, Robert Verpoorte2, and Dan Stærk1,3

BACKGROUND

Extracts of Ginkgo biloba leaves are among the best selling herbal preparations worldwide and are used for improvement of the peripheral and central blood circulation, improvement of memory, symptomatic treatment of dementia, Alzheimer’s disease as well as for their antioxidant and neuro-protective activities.

Commercial preparations are very complex mixtures prepared from raw leaf extract by a series of extraction and prepurification steps. During this process the extracts are enriched in some compounds while other compounds are removed.

The final Ginkgo biloba extract (EGb 761) are standardized to contain 24% flavonoids and 6% terpene trilactones (TTL’s; 3.1% ginkgolides and 2.9% bilobalide) which are considered the active constituents. However the current quality control does not assure the same distribution of individual TTL’s and flavonoid glycosides from batch-to-batch or brand-to-brand and other constituents that may possess pharmacological activity are not included in the standardization.

OO

H3C

H

HOO

O

R2

tBu

H

O

R1

OHO

H

O O

OHO

O

O

tBuO

O

H

H

OH

O

OOH

HO

OHR2

OR1

R1 R2 Ginkgolide A H H Ginkgolide B OH H Ginkgolide C OH OH Ginkgolide J H OH

Bilobalide

R2 Kaempferol H Quercetin OH Isorhamnetin OMe R1 = glycoside

RESULTS AND DISCUSSION

SAMPLE PREPARATIONNMR ANALYSIS DATA PREPROCESSING DATASET FOR PCA ANALYSIS

METHODOLOGY

Samples of all tablets and capsules were prepared in triplicate by extraction of pulverized material with 70% methanol. The dry residue was dissolved in acetone-d6, and 1H NMR spectra of all extracts were acquired at a resonance frequency of 600 MHz. All 1H-NMR spectra were reduced to 850 sequencially integrated spectral regions (buckets) of 0.01 ppm width from 0.5 to 9 ppm using Bruker AMIX software. PCA analysis were performed using SIMCA-P software on centered data.

2.963.38

2.81 2.962.73 2.77

3.19

2.551.99

3.84

1.702.13 2.20

2.452.88 3.06

0.89

0.81

0.980.99

0.76 0.77

1.10

0.80

0.87

1.10

0.65

0.66 0.680.60

1.111.24

0.620.51

0.66 0.56

0.78 0.76

0.46

0.801.25

0.21

1.12

1.09 1.040.97

0.520.41

0.49 0.47

0.37 0.36

0.45

0.51 0.40

0.27

0.41

0.28 0.37 0.29

0.550.55

1.09 0.96 1.06 1.021.36 1.34

0.791.35 1.49

0.58

2.111.84 1.69 1.70

0.94 0.75

0.340.44

0.0

1.0

2.0

3.0

4.0

5.0

6.0

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Preparations

Rel

ativ

e pe

rcen

tage

of T

TL´s

BB GC GB GJ GA

R2 = 0.9927

R2 = 0.9942

R2 = 0.9955 R2 = 0.9918

R2 = 0.90380.00

0.50

1.00

1.50

2.00

2.50

3.00

3.50

0.00 0.50 1.00 1.50 2.00 2.50 3.00 3.50Integrals

Buc

kets

BilobalideGinkgolide CGinkgolide BGinkgolide JGinkgolide A

1Department of Medicinal Chemistry, Faculty of Pharmaceutical Science, University of Copenhagen; 2Division of Pharmacognosy, Section Metabolomics, Institute of Biology, Leiden University; 3Department of Basic Science and Environment, Faculty of Life Sciences, University of Copenhagen

In the current work, 1H NMR-based metabolomics was used for non-biased investigation of the global composition of 16 commercially-available Ginkgo biloba preparations. In addition, hyphenation of HPLC, PDA, MS, and NMR with an automated solid-phase extraction (SPE) device, i.e., HPLC-PDA-MS-SPE-NMR, was used for identification of eight major flavonoid glycosides.

HPLC ANALYSIS

2255 4433

77

6611

88 O

OOH

HO

OH

R1

OR2

235

7

1' 3'4'

1: R1 = OH R2 = H2: R1 = H R2 = Rha(2 1)Glc(6 1)Coumaroyl ester3: R1 = OH R2 = Rha(2 1)Glc(6 1)Coumaroyl ester4: R1 = H R2 = Rha(2 1)Glc5: R1 = OCH3 R2 = Glc(6 1)Rha6: R1 = H R2 = Glc(6 1)Rha7: R1 = OH R2 = Rha(2 1)Glc8: R1 = OH R2 = Glc(6 1)Rha

HPLC-PDA-MS-SPE-NMR was used for the isolation of 8 selected flavonoids. Peaks were trapped 8 times on a SPE cartridges using threshold-based absorbance at 254 nm for selection of analytes.

Lack of commercially available reference substances for flavonol glycosides is a well know problem in the analysis of flavonoids.

HPLC-PDA-MS-SPE-NMR has constituted a fast and reliable method for the identification of the major flavonoids in Ginkgo biloba extracts.

CONCLUSIONS1H-NMR based metabolomics is able to discriminate between the various preparations according to their global composition. The analysis disclosed differences in the content of flavonol glycosides and terpene trilactones, as well differences in other constituents normally not considered in the standardization procedure. 1H-NMR based metabolomics constitute an attractive method for global characterization of standardized Ginkgo biloba preparations.

PCA analysis was performed on the entire spectrum. The preparations results separated based on their relative content of shikimic acid and ginkgolides. Discrimination between relative content of single TTL’s was possible. Sample 4 and 10 cluster together for the presence of citric acid as excipient.

More Shikimic Acid More Ginkgolides

More Ginkgolides C, J and Bilobalide

More Ginkgolides A and B

Citric Acid

OO

O O

O

OO

R2

HOR1

R3

H

H

13 8

1012

5

13

1416 17 O

OO

OOO

OHHO

H

H

Ginkgolides and BilobalideShikimic Acid

Citric Acid

2

3

4

5

6

1

HO

OH

OH

COOH

Citric acid constitute a frequently used excipient in pharmaceutical preparations. However, as demonstrated in a recent publication (Larsen et al., 2009), citric acid can form potentially harmful reaction products with constituents of pharmaceutical preparations and its addition in complex extract like Ginkgo biloba extracts should therefore preferably be avoided.

Quercetin

Protocatechuic acid

COOH

OHR

Ginkgolic Acid R = CxHy

Quercetin

A separated model was performed in the aromatic region of the spectra and further characterization of the samples based on the relative amount of TTLsand flavonoids was possible.

The same dataset but scaled on the total intensity, enabled the The same dataset but scaled on the total intensity, enabled the identification of ginkgolic acid in sample 8 and observation of identification of ginkgolic acid in sample 8 and observation of quercetin as almost the only flavonoid in sample 9, clear quercetin as almost the only flavonoid in sample 9, clear symptom of fortification.symptom of fortification.

A measure of the relative amount of the single TTLs (normalized to sum of 6%) can be easily obtained from the buckets and results equivalent to the manual integration of the peaks. The relative amount of TTL’s confirm the results of PCA analysis.

ACTIVE CONSTITUENTS OF GINKGO BILOBA

STRUCTURE IDENTIFICATION