the determination of fruit juice authenticity using high ... · the determination of fruit juice...

1

The Determination of Fruit Juice Authenticity Using High Resolution Chromatography, UV, Time-of-Flight MS, and Multivariate AnalysisMarian Twohig1, Antonietta Gledhill2, and Jennifer Burgess1

1Waters Corporation, Milford, MA, USA and 2 Waters Corporation, Manchester, UK

IN T RO DU C T IO N

The verification of food sources and authenticity is an activity that has increased

in importance over the last decade. The adulteration of food and beverages

has emerged as a growing problem that can pose potential threats to the health

of consumers and to the integrity of the industry1. Economic adulteration and

other types of counterfeiting in the food and consumer products industries are

estimated to cost between $10 and $15 billion USD per year.2

There are different types of adulteration that can occur, as illustrated in

Figure 1. While some types can be harmful to health3 (melamine is a good

example of this); others can be very misleading to the consumer – especially

if they believe the product is beneficial for certain health conditions. To date,

reported cases of fruit juice adulteration primarily include the addition of high-

fructose corn syrup (HFCS) or blending with other fruit juices.

Figure 1. Flow diagram showing various ways of food adulteration.

Dilution

Adulteration

SubstitutionUnapprovedenhancement

Failing todisclose

CounterfeitMislabeling

WAT E R S SO LU T IO NS

ACQUITY UPLC® System

ACQUITY UPLC PDA eλ Detector

Xevo® G2 QTof MS

MarkerLynx™ XS Application Manager

MassFragment™ Software

ACQUITY® HSS T3 Columns

K E Y W O R D S

Pomegranate, MVA, juice profiling,

adulteration, chlorogenic acid, food

authentication, arils, pulp

A P P L I C AT IO N B E N E F I T S ■■ Provides juice chemists with a powerful

system solution to comprehensively research

existing and new juice product prototypes

in a fast-paced marketplace.

■■ Allows easy and rapid differentiation

between authentic and adulterated

pomegranate juice samples.

■■ Multivariate statistical analysis (MVA)

enables visualization and interpretation

of complex MS data sets.

■■ QTof MS provides simultaneous accurate

mass information, for both the precursor

and fragment ions in a single injection.

Consumer demand for some fruit juices, such as pomegranate juice, has increased

significantly in recent years as clinical studies indicate that there may be positive

health benefits associated with the consumption of the fruit and the juice form.

These health benefits include the prevention against cancer and cardiovascular

disease.4-7 For this reason, pomegranate juice has gained a reputation for being

a super fruit – with demand often exceeding supply. As a result, the practice of

adulterating pomegrantite juice with lower quality juices has become prevalent.8

2The Determination of Fruit Juice Authenticity Using High Resolution Chromatography, UV, Time-of-Flight MS, and Multivariate Analysis

One of the issues the fruit juice industry faces is in the research and development

of new and exotic juice varieties. While many of the common fruit juices that have

been in the marketplace for many years are analytically well documented and

studied, this is not always the case for the newer juices that are now available.

For these types of products, there is a need for sophisticated testing methods to

help authenticate ingredients and finished products.

In this application note, we illustrate how the combination of Waters® Ultra

Performance Liquid Chromatography (UPLC®) System for high-resolution

separations, ACQUITY UPLC Photodiode Array (PDA), eλ Detector, and accurate

mass Xevo G2 QTof MS, with the appropriate chemometric software tools, allow

easy and rapid differentiation between authentic and adulterated pomegranate

juice samples.

Data analysis

Data analysis and trending were performed using MarkerLynx XS Application

Manager, which enables the user to perform multivariate analysis (MVA). Spectral

interpretation used EleComp, ChemSpider, and MassFragment Software to help

identify the structures of unknown marker compounds.

R E SU LT S A N D D IS C U S S IO N

All samples were analyzed using UPLC, PDA, and QTof MS. Typical chromatograms

are shown in Figure 2. Replicate injections of the pomegranate samples (in

random sequence order) were used to ensure that any experimental trends

observed were directly associated to the sample.

Figure 2. UPLC/PDA/QTof-MS chromatograms from the analysis of a pomegranate juice sample.

E X P E R IM E N TA L

Sample description

A description of the pomegranate samples is

provided in Table 1.

Name Sample description Sample type

S1 Study sample – Adulterated Juice

S2 Study sample – Authentic Juice

S3 Study sample – Adulterated Juice

S4 Study sample – Authentic Juice

S5 Study sample – Unknown history

Juice

S6 Purchased-Blend Juice

Whole fruit Three whole pomegranates

Skin, arils, pulp

Table 1. Description of the pomegranate samples used for the analysis.

Pomegranate samples

Juice samples: Three pomegranate juice concen-

trate samples were obtained from a collaborator:

one authentic (S2) and two adulterated (S1 and

S3). Commercially available pomegranate juice

samples were also included in the experiment

(S4-S6). Consultation with industry experts

confirm that the sample labeled S4 is believed

to be 100% authenticpomegranate juice.

The juice concentrates were made up to single

dose (160 Brix). One degree Brix is one-gram of

fruit soluble solids in 100 grams of solution, and

thus represented the strength of the solution as a

percentage by weight (% w/w). All samples were

centrifuged, filtered, and diluted before analysis.

Pomegranate fruit: Whole fruit pomegranates

were dissected into three components: arils, pulp,

and skin. Each component was homogenized,

centrifuged, and filtered before analysis.

UPLC conditions LC system: ACQUITY UPLC

Column: ACQUITY® HSS T3,

2.1 x 100 mm, 1.8 μm

Column temp: 45 ˚C

Time1.00 2.00 3.00 4.00 5.00 6.00 7.00

%

1

1.00 2.00 3.00 4.00 5.00 6.00 7.00

%

1

1.00 2.00 3.00 4.00 5.00 6.00 7.00

AU

0.0

2.0e+1

4.0e+1

24_11_0361.14

4.514.20

3.51

4.86

24_11_036

BPI4.63e3

5.35

4.531.22

4.223.53

1.43 2.35

4.88

6.18

5.937.76

24_11_0361: TOF MS ES-

BPI3.82e4

5.35

1.22

4.533.531.79

6.44

5.93 6.72

24_11_036 4: Diode Array Range: 5.921e+1

24_11_0362: TOF MS ES-

24_11_036

2A: PDA -total wavelength

2B: QTof MS - High CE

2C: QTof MS - Low CE


UPLC conditions (cont.)

Flow rate: 0.4 mL/ min

Mobile phase A: 10 mM Ammonium

acetate in water

Mobile phase B: Acetonitrile

UPLC gradient:

Time (min) % Age A % Age B

0.00 99 1

0.75 99 1

2.00 95 5

3.00 95 5

6.50 45 55

8.50 10 90

9.00 10 90

9.10 99 1

UV conditions

Detector: ACQUITY UPLC

PDA Detector

Range: 210 to 500 nm

Sampling rate: 20 pts/s

Filter constant: 0.1

MS conditions

MS System: Xevo G2 QTof

Ionization mode: ESI negative (ESI-)

Analyzer mode: Resolution

Capillary voltage: 2.0 kV

Cone voltage: 25 V

Desolvation temp.: 450 ˚C

Desolvation gas: 900 L/Hr

Source temp.: 130 ˚C

MSE

Low energy CE: 4 eV High energy CE: 15 to 45 eV

Acquisition range: 50 to 1200 m/z

Scan time: 0.1 sec

Lock mass reference: Leucine enkephalin

Data from the Xevo G2 QTof MS provided simultaneous accurate mass

information, for both the precursor and fragment ions in a single injection. Two

separate functions from the MS were produced. It is this spectral information that

enables potential structural identification of unknown marker compounds.

The low collision energy chromatogram shown in Figure 2C provided the exact

mass precursor ion information, and the high collision energy, shown in Figure 2B

gave the exact mass fragment ion data. The ability to cycle between low and high

collision energy to obtain both precursor and fragment ion information within one

analysis is a unique feature of the Waters QTof MS systems – this capability

is called MSE.

The MS data were interpreted using a Multivariate Approach (MVA) using the

MarkerLynx XS Application Manager. The software performs automatic data

integration and alignment to generate Exact Mass Retention Time (EMRT) pairs.

These EMRT pairs can then be used for multivariate statistical analysis to enable

visualization and interpretation of complex MS data sets.9 The complete software

workflow used in these experiments is described in Figure 3.

Figure 3. Pomegranate juice profiling workflow.

UPLC and Xevo G2 QTof:High resolution chromatographic separation

High sensitivity and accurate mass

MS/MS measurement:Acquire standards and compare standard results with samples

MassFragment:Evaluate proposed structure and searched compound

MSE and ChemSpider:Structual elucidation of identified marker compounds

MarkerLynx XS:Process and extract peaks

Use chemometrics to ID marker compounds


Figure 4 shows the six juice samples interpreted using Principal Component Analysis (PCA). The scores

plot helps to visualize the patterns, trends, and similarities between samples. Each point in the scores plot

represents a single injection. The repeat injections of each sample set are very close, indicating

good repeatability.

Figure 4. The scores plot obtained on analysis of six of the pomegranate juice samples using UPLC/QTof-MA MS data. (Each color represents a sample, and each sample was injected six times).

-800

-700

-600

-500

-400

-300

-200

-100

0

100

200

300

400

500

600

700

800

-1200 -1100 -1000 -900 -800 -700 -600 -500 -400 -300 -200 -100 0 100 200 300 400 500 600 700 800 900 1000 1100 1200

t[2]

t[1]

S1S2S3S4S5S6

S6S6S6S6S6S6S6S6S6

S5S5S5S5S5S5S5S5

S4S4S4S4S4S4S4

S3S3S3S3S3S3

S3

S2S2S2S2S2S2S2S2S2

S1S1

S1S1S1S1S1S1

EZinf o 2 - untitled67 (M4: PCA-X) - 2011-05-08 21:07:12 (UTC-5)

Scores Comp[1] vs. Comp[2] colored by Sample Group

S6S6S6S6S6S6S6S6S6

S5S5S5S5S5S5S5S5

S4S4S4S4S4S4S4

S3S3S3S3S3S3

S3

S2S2S2S2S2S2S2S2S2

S1S1

S1S1S1S1S1S1

Authentic juice

Juice Blend

Adulterated juice

Pomegranate juice (unverified

authenticity)


To explain the relationship between the potential markers (EMRTs) and relate these to the score plots, one

approach (when looking at multiple samples) is to use a loadings plot, as shown in Figure 5. It is then possible

to highlight EMRT’s in the loadings plot and track their abundance using a variable trend plot.

Using both the loadings plot and the variable trend plot, it can be seen that two EMRTs, with m/z 353.0868

at retention times (Rts) 3.18 and 4.34 min are absent in the known authentic samples (S2 and S4), and S5

(a sample of unverified authenticity), as shown in Figure 6.

-2000

0

2000

4000

6000

8000

S3 S3 S3 S3 S3 S3 S2 S2 S2 S2 S2 S2 S2 S2 S2 S1 S1 S1 S1 S1 S1 S1 PW PW PW PW PW PW PW NSI

NSI

NSI

NSI

NSI

NSI

NSI

NSI

LW LW LW LW LW LW LW

3.16

_353

.086

8

Sample Group

Authentic Juice

S6S3 S1

S4 S5

A: 3.18_353.0868

Juice Blend

S2

Figure 5. Loadings plot obtained from the analysis of the for the six pomegranate juice samples (UPLC/QTof-MS data): Each symbol in the loadings plot represents a single EMRT pair: where 4.34 is retention time and 353.0867 is the exact mass (m/z).

Figure 6. Variable trend plots for two possible markers (with m/z 353.0868) at Rt 3.18 min (A) and Rt 4.34 min (B).

-0.1

0.0

0.1

0.2

0.0 0.1 0.2 0.3

p[2]

p[1]

Loadings Comp[1] vs. Comp[2]

EZinf o 2 - Neg_Pomegranate_juice_data_N6_samples1 (M17: PCA-X) - 2011-03-24 11:05:45 (UTC-5)

3.18_353.0868

4.34_353.0867

0

1000

2000

3000

S3 S3 S3 S3 S3 S3 S2 S2 S2 S2 S2 S2 S2 S2 S2 S1 S1 S1 S1 S1 S1 S1 PW PW PW PW PW PW PW NSI

NSI

NSI

NSI

NSI

NSI

NSI

NSI

LW LW LW LW LW

4.34

_353

.086

4

Sample Group

Authentic Juice

S6S3

S1

S4 S5

B: 4.34._353.0868

S2

Juice Blend


Identification of chlorogenic acid

The presence of these two EMRT pairs at Rts 3.18 and 4.34 min was clearly evident in the deliberately

adulterated juice samples and the juice blend. The elemental composition for both components was determined

to be C16H17O9. Searching the formula in ChemSpider (an online database) proposed chlorogenic acid as the

top hit. Following the workflow described in Figure 3, MassFragment Software was used to determine whether

the fragments seen in the MSE data matched possible fragments for this compound. The results indicated that

chlorogenic acid matched well.

To confirm the identity of this potential marker, a standard of chlorogenic acid was analyzed and this standard

eluted at a retention time of 3.18 min, shown in Figure 7A, which was the same retention time as one of the

peaks in the samples, shown in Figure 7B.

Figure 7. Extracted ion chromatograms (m/z 353.0873) for chlorogenic acid [M-H]- in a standard (7A) and an adulterated pomegranate juice sample, S3 (7B). High CE MSE spectra for standard (7C), and sample (7D) are also shown.

Time2.00 4.00 6.00

%

0

100

2.00 4.00 6.00

%

0

1002_12_031 1: TOF MS ES-

353.087 0.0030Da1.04e4

3.18

2_12_035 1: TOF MS ES- 353.087 0.0030Da

1.04e43.18

4.34

Time2.00 4.00 6.00

%

0

100

2.00 4.00 6.00

%

0

1002_12_031 1: TOF MS ES-

353.087 0.0030Da1.04e4

3.18

2_12_035 1: TOF MS ES- 353.087 0.0030Da

1.04e43.18

4.34

Chlorogenic acid Extracted ion

chromatogram

Adulterated juiceExtracted ion

chromatogram

Chlorogenic acidMSE spectrum

Adulterated juiceMSE spectrum

C16H17O9 0.1 mDa,0.4 PPM

A

B

C

D


The UV spectra of the standard chlorogenic acid and the peak in the sample were compared. The UV spectra taken

across the entire peak width was inconclusive due to co-elution of other compounds in the samples. The MS data

was then investigated and the low energy precursor ions and high energy fragment exact mass ions provided

excellent confirmation that the component at Rt 3.18 min was chlorogenic acid, as shown in Figures 7C and 7D.

The MSE data for the component at Rt 4.34 min revealed that while it shared some common fragments with

chlorogenic acid, it also showed additional fragments. This suggests that this component may be a possible

isomer of chlorogenic acid (data not shown).

A closer look at the adulterated samples (S3 and S1) and the juice blend (S6), show that they all contain

chlorogenic acid (m/z 353.0873 at 3.18 min) and possible isomers at 4.34 min and 3.09 min.

Analysis of other fruit juices showed that chlorogenic acid was also identified in authentic peach, cranberry,

and blueberry juice samples (data not shown).

Eleven additional pomegranate juice samples, all claiming to be made from 100% authentic pomegranate

juice or juice concentrate were purchased and analyzed. The LC/MSE data revealed the presence of chlorogenic

acid in five of the samples tested. Two of those five samples were made from imported juice and juice

concentrate from Turkey. Poyrazoglu et.al have reported the presence of chlorogenic acid in 13 different

pomegranate varieties from four growing regions in Turkey using LC/UV.10

Whole fruit experiments

Experiments were then performed on whole fruit samples to determine whether chlorogenic acid is present in

pomegranate. Three whole fruit pomegranates (all believed to be grown in North America) were purchased and

dissected into three components: arils, pulp, and skin. The pomegranate segments were extracted and analyzed.

Using the same analytical conditions as for the juice, chlorogenic acid could not be found in any part of the

pomegranates, as shown in the extracted ion chromatograms in Figure 8.

Time2.00 4.00 6.00 8.00

%

0

100

2.00 4.00 6.00 8.00

%

0

100

2.00 4.00 6.00 8.00

%

0

100

2.00 4.00 6.00 8.00

%

0

100

2_12_0313.18

4.33

2_12_005 1: TOF MS ES- 353.087 0.0030Da

2.59e3

2_12_004 1: TOF MS ES- 353.087 0.0030Da

2.59e3

2_12_003 1: TOF MS ES- 353.073 0.0030Da

2.59e3

0.79

Skin

Pulp

Arils

Time2.00 4.00 6.00 8.00

%

0

100

2.00 4.00 6.00 8.00

%

0

100

2.00 4.00 6.00 8.00

%

0

100

2.00 4.00 6.00 8.00

%

0

100

2_12_0313.18

4.33

2_12_005 1: TOF MS ES- 353.087 0.0030Da

2.59e3

2_12_004 1: TOF MS ES- 353.087 0.0030Da

2.59e3

2_12_003 1: TOF MS ES- 353.073 0.0030Da

2.59e3

0.79Time

2.00 4.00 6.00 8.00

%

0

100

2.00 4.00 6.00 8.00

%

0

100

2.00 4.00 6.00 8.00

%

0

100

%

0

100

2_12_031

2_12_005 1: TOF MS ES- 353.087 0.0030Da

2.59e3

2_12_004 1: TOF MS ES- 353.087 0.0030Da

2.59e3

2_12_003 1: TOF MS ES- 353.073 0.0030Da

2.59e3

0.79

8.000

8.00

2_12_031

2_12_005 1: TOF MS ES- 353.087 0.0030Da

2.59e3

2_12_004 1: TOF MS ES- 353.087 0.0030Da

2.59e3

2_12_003 1: TOF MS ES- 353.073 0.0030Da

2.59e3

Chlorogenicstandard

acid

Figure 8. Extracted ion chromatograms (m/z 353.0873) of a chlorogenic acid standard, and the skin, pulp and arils, of a whole pomegranate.


These additional experiments suggest that North American pomegranates (grown and harvested during

this time) do not contain any detectable amount of chlorogenic acid. It also suggests that the presence of

chlorogenic acid in North American pomegranate juice may indicate the blending of other juices. To further

understand the detection of chlorogenic acid in pomegranate varieties from different regions, a much broader

mass spectrometric study is warranted.

EMRT 3.43_353.1448

Another EMRT of interest, 3.43_353.1448 was also investigated using the same workflow as described

previously. The elemental composition was determined to be C14H26O10. The extracted ion chromatograms

of this m/z from the juice samples used in the study revealed it to be present in all pomegranate samples but

at varying concentrations. For the juice blend (S6), it was present at a low level, as shown in Figure 9.

However for the authentic juices, the levels were higher with the highest levels found in the authentic

pomegranate juice (S2).

0

20

40

60

80

100

120

140

160

S1 S2 S3 S4 S5 S6

3.43_353.1448

Figure 9. Variation of the levels of component 3.43_353.1448 in the pomegranate juice samples.


Extracted ion chromatograms of m/z 353.1448 in the whole fruit extracts (arils, pulp, and skin) shown in

Figure 10A, revealed that this marker was only present in the arils, and that it was present in the arils of all

three pomegranates, shown in Figure 10B.

Figure 10A. Extracted ion chromatograms for marker 3.43_353.1448 in the arils, pulp and skin of a whole pomegranate. Figure 10B. The marker was found in the arils of all pomegranates used in the study.

The uniqueness of this compound to pomegranate, and only within the arils of pomegranate suggests

that it could be a valuable marker for the quality of pomegranate juice. However, additional

experiments are required to identify this compound and study the seasonal, varietal, and geographical

variations of the compound.

Summary of the results

Chlorogenic acid was absent in the authentic pomegranate juices analyzed in this study. Similarly

it was not found in three whole fruit pomegranates.the adulterated samples used in this study,

chlorogenic acid and two possible isomers were found to be present. This compound could, therefore

be a potential marker compound for determining pomegranate juice authenticity.

Time2.00 4.00 6.00

%

0

100

2.00 4.00 6.00

%

0

100

2.00 4.00 6.00

%

0

100

2_12_0031: TOF MS ES-

353.145 0.0050Da2.44e4

3.43

2_12_0041: TOF MS ES-

353.145 0.0050Da2.44e4

2_12_0111: TOF MS ES-

353.145 0.0050Da2.44e4

Arils

Pulp

Skin

Time2.00 4.00 6.00

%

0

100

2.00 4.00 6.00

%

0

100

2.00 4.00 6.00

%

0

100

1: TOF MS ES- 353.145 0.0050Da

2.44e43.43

2_12_009 1: TOF MS ES- 353.145 0.0050Da

2.44e4

3.43

28_2_023 1: TOF MS ES- 353.145 0.0050Da

2.44e4

3.43

Arils fraction Pomegranate 1



A B

Waters Corporation 34 Maple Street Milford, MA 01757 U.S.A. T: 1 508 478 2000 F: 1 508 872 1990 www.waters.com

References

1. Roberts MT. Cheaters Shouldn’t Prosper and Consumers Shouldn’t Suffer: The Need for Government Enforcement Against Economic Adulteration of 100% Pomegranate Juice and Other Imported Food Products. Michael T. Roberts is Special Counsel for Roll Law Group PC. (2010).

2. Kearney AT and Grocery Manufacturers Association. Consumer Product Fraud: Deterrence and Detection, (2010).

3. Vai T, Jones PR, Sparkman OD . Rapid and unambiguous identification of melamine in contaminated pet food based on mass spectrometry with four degrees of confirmation. J. Anal. Toxicol. (2007) 31, 1319-1325.

4. Seeram NP et al. In vitro antiproliferative, apoptotoic and antioxidant activities of punicalagin, ellagic acid, and a total pomegranate tannin extract are enhanced in combination with other polyphenols as found in pomegranate juice. J Nutr Biochem. 16, 6:360-7, 2005.

5. Mehta RE, Lansky EP. Breast cancer chemopreventive properties of pomegranate ( Punica granatum) fruit extracts in a mouse mammary organ culture. Eur J Cancer Prev.13, 4:345-8, 2004.

6. Lansky EP et al. Pomegranate ( Pumica granatum) pure chemicals show possible synergistic inhibition of human PC-3 prostate cancer cell invasion across Matrigel. Invest New Drugs.23, 2:121-2, 2005.

7. Aviram M et al. Pomegranate juice consumption for 3 years by patients with carotid artery stenosis reduces common carotid intima-media thickness, blood pressure and LDL oxidation. Clin Nutr. 23, 3:423-33, 2004.

8. Zhang Y, Krueger D, Durst R, Lee R, Wang D, Seeram N , Heber D. International multidimensional authenticity specification (IMAS) algorithm for the detection of commercial pomegranate juice adulteration. J. Agric. Food Chem. 2009, 57, 2550-2557.

9. Eriksson L. Book “Multi- and Megavariate Data Analysis: Basic Principles and Applications”, 2nd edition, 2008.

10. Poyrazoglu E, Gokmen V, Artuk N. Organic acids and phenolic compounds in pomegranates (punica granatum) grown in Turkey. J. Food. Comp. Anal. (2002) 15, 567-575.

CO N C LU S IO NS

Economic adulteration of food and beverage products is a problem

that has increased in magnitude because it has a lucrative outcome

for unscrupulous members of the food manufacturing chains.

■■ Consumer demand for more exotic juices introduces analytical

challenges for juice chemists.

■■ The ACQUITY UPLC System, ACQUITY UPLC PDA Detector, and

Xevo G2 QTof MS, combined with a complete software workflow

based on Markerlynx XS functionality, provide a powerful

system solution for the analysis and determination of fruit juice

authenticity in a research laboratory. Here these tools enabled

the detection of compounds that have the potential to be used as

indicators of a product’s authenticity.

■■ The ACQUITY UPLC System provides high resolution

chromatographic data for complex samples such as

juice products.

■■ The ACQUITY UPLC PDA Detector and Xevo G2 QTof MS are

complementary instruments for this type of analysis. This

combination allows juice chemists to comprehensively research

existing and new juice product prototypes in a fast-paced

marketplace.

■■ MSE functionality allowed the acquisition of low energy

precursor (MS) and high energy product ions in a single run. The

fragment data together with exact mass measurement provided

added confidence and accuracy for structural elucidation.

■■ Intelligent analytical methods, such as the MarkerLynx XS

profiling workflow presented in this application note, provide

reliable and highly detailed information that can help in the

process of food authentication.

Waters, ACQUITY UPLC, ACQUITY, UPLC, and Xevo are registered trademarks of Waters Corporation. MarkerLynx, MassFragment, and The Science of What’s Possible are trademarks of Waters Corporation. All other trademarks are the property of their respective owners.

©2011 Waters Corporation. Produced in the U.S.A. August 2011 720004010EN AO-PDF

the determination of fruit juice authenticity using high ... · the determination of fruit juice...

Documents