HPLC-Determination of nine major Anthocyanins in red and rosé wines (Type-II)
OIV-MA-AS315-11 HPLC-Determination of nine major anthocyanins in red and rosé wine
Type II method
- Field of application
The analytical method concerns the determination of the relative composition of anthocyanins in red and rosé wine. The separation is performed by HPLC with reverse phase column and UV-VIS detection.
Many authors [3, 6-17] have published data on the anthocyanin composition of red wines using similar analytical methods. For instance Wulf et al. [18] have detected and identified 21 anthocyanins and Heier et al. [13] nearly 40 by liquid chromatography combined with mass spectrometry. The anthocyanin composition may be very complex, so it is necessary to have a simple procedure. Consequently this method only determines the major compounds of the whole anthocyanin fraction.
Member states are encouraged to continue research in this area to avoid any non scientific evaluation of the results.
- Principle
Separation of the five most important non acylated anthocyanins (see Figure 1, peaks 1-5) and four major acylated anthocyanins (see Figure 1, peaks 6-9).
Analysis of red and rosé wine by direct separation by HPLC by using reverse phase column with gradient elution by water/formic acid/acetonitrile with detection at 518 nm [1.2].
- Reagents and material
Formic acid (p.a. 98 %) (CAS 64-18-6);
Water, HPLC grade;
Acetonitrile, HPLC grade (CAS 75-08-8);
HPLC solvents:
Solvent A: Water/Formic acid/Acetonitrile 87 : 10 : 3 (v/v/v)
Solvent B: Water/Formic acid/Acetonitrile 40 : 10 : 50 (v/v/v)
Membrane filter for HPLC solvent degassing and for sample preparation to be analysed.
Reference products for peak identification.
The HPLC analysis of anthocyanins in wine is difficult to perform due to the absence of commercially available pure products. Furthermore, anthocyanins are extremely unstable in solution.
The following anthocyanin pigments are commercially available:
Cyanidol-3-glucoside (also couromanin chloride); M = 484.84 g/mol
Peonidol-3-glucoside; M = 498.84 g/mol
Malvidol-3-glucoside (also Oeninchloride); M = 528.84 g/mol
Malvidol-3,5-diglucoside (also Malvinchloride); M = 691.04 g/mol
- Apparatus
HPLC system with:
binary gradient pump, injection system for sample volumes ranging from 10 to 200 μl,
diode array detector or a UV detector with a visible range,
integrator or a computer with data acquisition software,
furnace for column heating at 40°C,
solvent degassing system,
analytical column, for example:
LiChrospher 100 RP 18 (5 μm) in LiChroCart 250-4 guard column: for example RP 18 (30-40 mm) in a cartridge 2 mm in diameter x 20 mm long
- Procedure
5.1. Preparation of samples
Clear wines are poured directly without any preparation into the sample vials of the automatic sample changer. Cloudy samples are filtered using a 0.45 μm membrane filter for HPLC sample preparation. The first part of the filtrate should be rejected.
Since the range of the linearity of absorption depending on the concentration of anthocyanins is large, it is possible to modulate the injection volumes between 10 and 200 μl depending on the intensity of the wine colour. No significant difference between the results obtained for different injection volumes was observed.
5.2. Analysis
HPLC conditions
The HPLC analysis is carried out in the following conditions:
Injection Volume: |
50 μl (red wine) up to 200 μl (rosé wine) |
Flow: |
0.8 ml/minute |
Temperature: |
40°C |
Run time: |
45 minutes |
Post time: |
5 minutes |
Detection: |
518 nm |
Gradient elution: |
Time (min) |
Solvent A % (v/v) |
Solvent B % (v/v) |
0 |
94 |
6 |
|
15 |
70 |
30 |
|
30 |
50 |
50 |
|
35 |
40 |
60 |
|
41 |
94 |
6 |
To check the column efficiency, the number of theoretical plates (N) calculated according to malvidol-3-glucoside should not be below 20,000, and the resolution (R) between peonidol-3-coumaryl glucoside and malvidolin-3-coumaryl glucoside should not be lower than 1.5. Below these values, the use of a new column is recommended.
A typical chromatogram is given in Figure 1, where the following anthocyanins are separated:
Peak-N° |
||
Group 1: “Nonacylated anthocyanidin-3-glucosides”: |
delphinidol-3-glucoside cyanidol-3-glucoside petunidol-3-glucoside peonidol-3-glucoside malvidol-3-glucoside |
1 2 3 4 5 |
Group 2: “Acetylated anthocyanidin-3-glucosides”: |
peonidol-3-acetylglucoside malvidol-3-acetylglucoside |
6 7 |
Group 3: “Coumarylated anthocyanidin-3-glucosides”: |
peonidol-3-coumarylglucoside malvidol-3-coumarylglucoside |
8 9 |
- Expression of results
Note that the values are expressed as relative amounts of the sum of the nine anthocyanins defined in this method.
- Limit of detection and limit of quantification
The limit of detection (LD) and the limit of quantification (LQ) are estimated following the instructions in the resolution OENO 7-2000 “Estimation of the Detection and Quantification Limits of a Method of Analysis“. Along the line of the ”Logic Diagram for Decision-Making” in N° 3 the graph approach has to be applied following paragraph 4.2.2.
For this purpose a part of the chromatogram is drawn out extendedly enclosing a range of a tenfold mid-height width (w½) from an anthocyan relevant peak.
Furthermore two parallel lines are drawn which just enclose the maximum amplitude of the signal window. The distance of these two lines gives , expressed in milli Absorption Units (mAU).
The limit of detection (LD) and the limit of quantification (LQ) depend on the individual measurement conditions of the chemical analysis and are to be determined by the user of the method. The Annex gives an example of its determination with the following results:
hmax = 0,208 [mAU]; LD = 3 x 0,208 [mAU] = 0,62 [mAU].
LQ = 10 x 0, 208 [mAu] = 2,08 [mAU].
Recommendation:
With combined data out of the whole Anthocyanin composition such as the sum of Acylated Anthocyanins or the ratio of Acetylated to Coumarylated Anthocyanins the calculation should not be carried out in cases where one of the components is below the limit of quantification (LQ).
On the other hand measurements below the limit of quantification (LQ) are not devoid of information content and may well be fit for purpose [1].
Bibliography:
- Thompson, M.; Ellison, S.L.R. ; Wood, R., Harmonized Guidelines for Single-Laboratory Validation of Methods of Analysis, Pure Appl. Chem. (2002) 74: 835- 855
- Fidelity parameters
The repeatability (r) and the reproducibility (R) values for the nine anthocyanins are given in Table 2 and depend on the amount of the peak area. The uncertainty measurement of a particular peak area is determined by the value of r and R which corresponds to the nearest value given in Table 2.
The values made up of validation data can be calculated by following the appropriate statistical rules. To calculate the total error (sr) for example of the sum of acetylated anthocyanins, the variances (sr2) of specific the total error of ratios, for example, that of acetylated to coumarylated anthocyanins the square of relative errors (=sr/ai) are to be added. By using these rules, all the fidelity values can be calculated by using the data in Table 2.
|
Annex A : Bibliography
- Marx, R., B. Holbach, H. Otteneder; Determination of nine characteristic Anthocyanins in Wine by HPLC; OIV, F.V.N° 1104 2713/100200
- Holbach, B., R. Marx, M. Ackermann; Bestimmung der Anthocyanzusammensetzung von Rotwein mittels Hochdruckflüssigkeitschromatographie (HPLC). Lebensmittelchemie (1997) 51: 78 – 80
- Eder, R., S. Wendelin, J. Barna; Auftrennung der monomeren Rotweinanthocyane mittels. Hochdruckflüssigkeitschromatographie (HPLC).Methodenvergleich und Vorstellung einer neuen Methode. Mitt. Klosterneuburg (1990) 40: 68-75
- ISO-5725-2: 1994 “Accuracy (trueness and precision) of measurement methods and results - Part 2: Basic method for the determination of repeatability and reproducibility”
- Otteneder, H., Marx, R., Olschimke, D.; Method-performance study on the determination of nine characteristic anthocyanins in wine by HPLC. O.I.V. F.V.N° 1130 (2001)
- Mattivi F.; Scienza, A.; Failla, O.; Vika, P.; Anzani, R.; Redesco, G.; Gianazza, E.; Righetti; P. Vitis vinifera - a chemotaxonomic approach: Anthocyanins in the skin. Vitis (special issue) 1990, 119-133
- Roggero, I.P.; Larice, I.L.; Rocheville-Divorne, C.; Archier, P.; Coen, V. Composition Antocyanique des cepages. Revue Francaise d’Oenologie 1998, 112, 41-48
- Eder, R.; Wendelin, S; Barna, J. Classification of red wine cultivars by means of anthocyanin analysis. Mitt. Klosterneuburg 1994, 44, 201-212
- Arozarena, I.; Casp, A.; Marin, R.; Navarro, M. Differentiation of some Spanish wines according to variety and region based on their anthocyanin composition. Eur. Food Res. Technol. 2000, 212, 108-112
- Garcia-Beneytez, E.; Revilla, E.; Cabello, F. Anthocyanin pattern of several red grape cultivars and wines made from them. Eur. Food Res. Technol. 2002, 215, 32-37
- Arozarena, I.; Ayestarán, B.; Cantalejo, M.J.; Navarro, M.; Vera, M.; Abril, K.; Casp, A. Eur. Food Res. Technol. 2002, 214, 313-309
- Revilla, E.; Garcia-Beneytez, E.; Cabello, F.; Martin-Ortega, G.; Ryan, J-M. Value of high-performance liquid chromatographic analysis of anthocyanins in the differentiation of red grape cultivars and red wines made from them. J. Chromatogr A 2001, 915, 53-60
- Heier, A.; Blaas, W.; Droß, A.; Wittkowski, R.; Anthocyanin Analysis by HPLC/ESI-MS, Am.J.Enol.Vitic, 2002, 53, 78-86
- Arozarena, I.; Casp, A.; Marin, R.; Navarro, M. Multivariate differentiation of Spanish red wines according to region and variety. J. Sci. Food Agric, 2000, 80, 1909-1917
- Anonymous. Bekanntmachung des Bundesinstituts für gesundheitlichen Verbraucherschutz und Veterinärmedizin. Bundesgesundheitsbl. Gesundheitsforsch. Gesundheitsschutz, 2001, 44, 748
- Burns, I.; Mullen, W.; Landrault, N.; Teissedre, P.-L.; Lean, M.E.I.; Crozier, A. Variations in the Profile and Content of Anthocyanins in Wines made from Cabernet Sauvignon and hybrid grapes. J. Agric. Food Chem. 2002, 50, 4096-4102
- Otteneder, H.; Holbach, B.; Marx, R.; Zimmer, M. Rebsortenbestimmung in Rotwein mittels Anthocyanspektrum. Mitt. Klosterneuburg, 2002, 52, 187-194
-
L.W. Wulf and C.W. Nagel; High-Pressure liquid chromatographic separation of Anthocyanins of Vitis vinifera.
Am.J.Enol.Vitic 1978, 29, 42-49
Annex B Statistical results
Method performance study and evaluation
17 laboratories from 5 European Nations participated in the validation study of the method under the coordination of the German Official State Laboratory for Food Chemistry in Trier. The participants are listed in Table 3. An example of a chromatogram is presented in Figure 1 and the detailed results are given in Table 2.
The statistical evaluation followed the Resolution 6/99 and the Standard ISO 5725-1944 [4.5].
The chromatograms sent back with the results sheets fulfilled all requirements concerning the performance of the analytical column. No laboratory had to be completely eliminated, for example, because of a wrong peak identification.
The outlier values were searched using Dixon and Grubbs outlier testing according to the procedure for “Harmonised Protocol – IUPAC 1994” and the OIV Resolution OENO 19/2002. The values of sr, sR, r and R were calculated for 9 major anthocyanins at 5 content levels. For analytical results, the values of the closest levels should be used.
In order to have a global vision of the method performance, all the values RSDr- et RSDR- gathered are grouped by range of areas in the following table:
Table 1: Summary of the results of the method performance study
Range of relative peak areas*[%] |
Range of RSDr |
Range of RSDR |
>0.4 – 1.0 |
6.8 - 22.4 |
20.6 - 50.9 |
>1.1 – 1.5 |
4.2 - 18.1 |
11.8 - 28.1 |
>1.5 – 3.5 |
2.1 – 7.7 |
10.6 - 15.6 |
>3.5 – 5.5 |
2.7 – 5.7 |
18.7 – 7.5 |
>5.5 – 7.5 |
2.4 – 3.9 |
6.5 - 10.0 |
>10 – 14 |
1.1 – 2.9 |
3.7 - 9.2 |
>14 – 17 |
1.0 - 3.9 |
3.2 - 5.4 |
>50 – 76 |
0.3 - 1.0 |
2.1 - 3.1 |
* independent of anthocyanin |
This leads to the conclusion that repeatabilities and reproducibilities depend on the total sum of the relative peak areas. The higher they are, the better are RSDr and RSDR. For anthocyanin contents close to the detection limit (e.g. Cyanidin-3-glucoside) with small relatives areas (less than 1%) the RSDr et RSDR values can rise significantly. For anthocyanin whose relative areas are more than 1%, the RSDr and RSDR values are reasonable.
Figure 1: Separation of 9 anthocyanins in red wine |
|
Table 2: Results of the method performance study
Anthocyanin |
sample 1 |
sample 2 |
sample 3 |
sample 4 |
sample 5 |
|
Delphinidol-3-glucoside |
||||||
n |
14 |
14 |
16 |
15 |
16 |
|
mean |
6.75 |
14.14 |
3.45 |
16.68 |
3.54 |
|
sr |
0.163 |
0.145 |
0.142 |
0.142 |
0.108 |
|
RSDr(%) |
2.4 |
1.0 |
4.1 |
0.8 |
3.1 |
|
r |
0.46 |
0.41 |
0.40 |
0.40 |
0.30 |
|
sR |
0.544 |
0.462 |
0.526 |
0.704 |
0.490 |
|
RSDR(%) |
8.1 |
3.3 |
15.2 |
4.2 |
13.8 |
|
R |
1.52 |
1.29 |
1.47 |
1.97 |
1.37 |
|
Cyanidol-3-glucoside |
||||||
n |
16 |
17 |
16 |
15 |
14 |
|
mean |
2.18 |
1.23 |
0.61 |
1.46 |
0.34 |
|
sr |
0.086 |
0.053 |
0.043 |
0.110 |
0.031 |
|
RSDr(%) |
4.0 |
4.3 |
7.1 |
7.5 |
9.2 |
|
r |
0.24 |
0.15 |
0.12 |
0.31 |
0.09 |
|
sR |
0.460 |
0.211 |
0.213 |
0.180 |
0.158 |
|
RSDR(%) |
21.2 |
17.2 |
34.9 |
12.3 |
46.7 |
|
R |
1.29 |
0.59 |
0.60 |
0.50 |
0.44 |
|
Petunidol-3-glucoside |
||||||
n |
15 |
17 |
16 |
14 |
15 |
|
mean |
10.24 |
14.29 |
5.75 |
12.21 |
6.19 |
|
sr |
0.233 |
0.596 |
0.157 |
0.097 |
0.196 |
|
RSDr(%) |
2.3 |
4.2 |
2.7 |
0.8 |
3.2 |
|
r |
0.65 |
1.67 |
0.44 |
0.27 |
0.55 |
|
sR |
0.431 |
0.996 |
0.495 |
0.469 |
0.404 |
|
RSDR(%) |
4.2 |
7.0 |
8.6 |
3.8 |
6.5 |
|
R |
1.21 |
2.79 |
1.39 |
1.31 |
1.13 |
|
Peonidol-3-glucoside |
||||||
n |
16 |
15 |
17 |
17 |
16 |
|
mean |
11.88 |
6.23 |
13.75 |
7.44 |
4.12 |
|
sr |
0.241 |
0.166 |
0.144 |
0.232 |
0.174 |
|
RSDr(%) |
2.0 |
2.7 |
1.0 |
3.1 |
4.2 |
|
r |
0.68 |
0.47 |
0.40 |
0.65 |
0.49 |
|
sR |
0.981 |
0.560 |
1.227 |
0.602 |
0.532 |
|
RSDR(%) |
8.3 |
9.0 |
8.9 |
8.1 |
12.9 |
|
R |
2.75 |
1.57 |
3.44 |
1.69 |
1.49 |
|
Malvidol-3-glucoside |
||||||
n |
16 |
15 |
17 |
16 |
16 |
|
mean |
55.90 |
55.04 |
76.11 |
52.60 |
61.04 |
|
sr |
0.545 |
0.272 |
0.251 |
0.298 |
0.377 |
|
RSDr(%) |
1.0 |
0.5 |
0.3 |
0.6 |
0.6 |
|
r |
1.53 |
0.76 |
0.70 |
0.83 |
1.06 |
|
sR |
2.026 |
2.649 |
2.291 |
1.606 |
1.986 |
|
RSDR(%) |
3.6 |
4.8 |
3.0 |
3.1 |
3.3 |
|
R |
5.67 |
7.42 |
6.41 |
4.50 |
5.56 |
|
n |
= N° of laboratories retained after eliminating outliers |
|||||
sr |
= standard deviation of repeatability |
|||||
RSDr(%) |
= relative standard deviation of repeatability |
|||||
r |
= repeatability |
|||||
sR |
= standard deviation of reproducibility |
|||||
RSDR(%) |
= relative standard deviation of reproducibility |
|||||
R |
= reproducibility |
|||||
Table 2: Results of the method performance study
Anthocyanin |
sample 1 |
sample 2 |
sample 3 |
sample 4 |
sample 5 |
|
Peonidol-3-acetylglucoside |
||||||
n |
14 |
16 |
14 |
16 |
||
mean |
1.16 |
1.44 |
0.59 |
3.74 |
||
sr |
0.064 |
0.062 |
0.059 |
0.215 |
||
RSDr(%) |
5.5 |
4.3 |
10.1 |
5.8 |
||
0.18 |
0.17 |
0.17 |
0.60 |
|||
sR |
0.511 |
0.392 |
0.272 |
0.374 |
||
RSDR(%) |
43.9 |
27.2 |
46.4 |
10.0 |
||
R |
1.43 |
1.10 |
0.76 |
1.05 |
||
Malvidol-3-acetylglucoside |
||||||
n |
16 |
17 |
17 |
16 |
||
mean |
5.51 |
4.84 |
3.11 |
15.07 |
||
sr |
0.176 |
0.167 |
0.088 |
0.213 |
||
RSDr(%) |
3.2 |
3.4 |
2.8 |
1.4 |
||
r |
0.49 |
0.47 |
0.25 |
0.60 |
||
sR |
0.395 |
0.366 |
0.496 |
0.617 |
||
RSDR(%) |
7.2 |
7.6 |
16.0 |
4.1 |
||
R |
1.11 |
1.02 |
1.39 |
1.73 |
||
Peonidol-3-coumarylglucoside |
||||||
n |
16 |
14 |
17 |
16 |
||
mean |
1.26 |
0.90 |
0.89 |
1.32 |
||
sr |
0.130 |
0.046 |
0.060 |
0.058 |
||
RSDr(%) |
10.3 |
5.1 |
6.8 |
4.4 |
||
r |
0.36 |
0.13 |
0.17 |
0.16 |
||
sR |
0.309 |
0.109 |
0.204 |
0.156 |
||
RSDR(%) |
24.5 |
12.2 |
23.0 |
11.8 |
||
R |
0.86 |
0.31 |
0.57 |
0.44 |
||
Malvidol-3-coumarylglucoside |
||||||
n |
17 |
17 |
17 |
16 |
||
mean |
4.62 |
2.66 |
4.54 |
4.45 |
||
sr |
0.159 |
0.055 |
0.124 |
0.048 |
||
RSDr(%) |
3.4 |
2.1 |
2.7 |
1.1 |
||
r |
0.45 |
0.15 |
0.35 |
0.13 |
||
sR |
0.865 |
0.392 |
0.574 |
0.364 |
||
RSDR(%) |
18.7 |
14.7 |
12.6 |
8.2 |
||
R |
2.42 |
1.10 |
1.61 |
1.02 |
||
n |
= N° of laboratories retained after eliminating outliers |
|||||
sr |
= standard deviation of repeatability |
|||||
RSDr(%) |
= relative standard deviation of repeatability |
|||||
r |
= repeatability |
|||||
sR |
= standard deviation of reproducibility |
|||||
RSDR(%) |
= relative standard deviation of reproducibility |
|||||
R |
= reproducibility |
|||||
Table 3: List of participants
ABC Labor Dahmen, Mülheim/Mosel |
D |
Chemisches Landes- und Staatliches Veterinäruntersuchungsamt Münster |
D |
Institut für Lebensmittelchemie Koblenz |
D |
Institut für Lebensmittelchemie Speyer |
D |
Institut für Lebensmittelchemie Trier |
D |
Institut für Lebensmittelchemie und Arzneimittel Mainz |
D |
Labor Dr. Haase-Aschoff, Bad Kreuznach |
D |
Labor Dr. Klaus Millies, Hofheim-Wildsachsen |
D |
Labor Heidger, Kesten |
D |
Landesveterinär- und Lebensmitteluntersuchungsamt Halle |
D |
Staatliche Lehr- und Forschungsanstalt für Landwirtschaft, Weinbau und Gartenbau, Neustadt/Weinstraße |
D |
Staatliches Institut für Gesundheit und Umwelt, Saarbrücken |
D |
Staatliches Medizinal-, Lebensmittel- und Veterinäruntersuchungsamt, Wiesbaden |
D |
Laboratoire Interrégional de la D.G.C.C.R.F de Bordeaux, Talence/France |
F |
Unidad de Nutricion y Bromotologia, Facultad de Farmacia, Universidad de Salamanca, Salamanca/Espana |
E |
University of Glasgow, Div. of Biochem. and Molek. Biology |
UK |
Höhere Bundeslehranstalt und Bundesamt für Wein- und Obstbau, Klosterneuburg |
A |
Laboratories
D (13); A (1); F (1); E (1); UK (1)