AB SCIEX Quantitation and Identification of Dicyandiamide in Milk and Other Protein-Rich Foods

LC-MS/MS Analysis of Emerging Food Contaminants

Quantitation and Identification of Dicyandiamide in Milk and other Protein-Rich Foods

Fanny Fu1 and André Schreiber2 1AB SCIEX Taipei (Taiwan), 2AB SCIEX Concord, Ontario (Canada)

Introduction

Recent issues with adulteration of food using nitrogen rich compounds to make the protein content of food appear higher than the actual value highlighted the need for both food manufacturers and regulatory agencies to utilize fast and accurate analytical techniques to proactively ensure product safety.

In 2007, melamine and cyanuric acid in wheat gluten added to pet food caused renal failure and sickened and killed large numbers of cats and dogs. In 2008, Chinese authorities discovered the adulteration of milk and infant formula with melamine by several Chinese producers. There were hundreds of thousands of victims and six confirmed deaths in China, as well as product recalls in many countries.1-4 Experimental In response to the melamine contamination a large number of analytical methods were developed for the detection of melamine Sample Preparation and its analogues, including several published by the United States Food and Drug Administration (FDA) that also targeted Simple liquid extraction of food samples was performed using 4 cyanuric acid.4-8 the following procedure :

However, the Kjeldahl method, the traditional standard technique · Add 10 mL of acetonitrile containing 2% formic acid to 1 g of a for measuring protein content by indirectly measuring the homogenized sample. nitrogen content in food, remains the most widespread · Mix thoroughly and sonicate for 10 minutes. methodology. As long as protein content in food is not · Centrifuge for 10 minutes. determined directly, economic adulteration with nitrogen rich · Transfer an aliquot of 50 μL of the extract into and compounds will continue to be a serious concern. autosampler vial and dilute with 950 μL acetonitrile resulting in Analytical methods to detect potential adulterants (non-protein a total dilution factor of 200. nitrogen sources), including amidinourea, ammelide, ammeline, Further dilution of the extract might be necessary if the sample is biuret, cyanuric acid, cyromazine, dicyandiamide, melamine, heavily contaminated. triuret, and urea (Figure 1) have been developed and validated to test milk products and bulk protein.4, 5 LC

Recently, traces of dicyandiamide were found in milk produced in The target compounds were separated using a normal phase New Zealand. Milk producers and government agencies moved gradient on a Hydrophilic Interaction Chromatography (HILIC) quickly to reassure there was no risk to health. Here we present column. LC separation was achieved using the Eksigent a fast, easy, and sensitive LC-MS/MS method for the detection of dicyandiamide and other nitrogen rich compounds in milk and ekspert™ ultraLC 100 system with a Phenomenex LUNA HILIC other protein-rich foods with limits of quantitation down to low 3u (100 x 2 mm) column with a mobile phase of acetonitrile and μg/kg. water containing 0.1% formic acid and 10 mM ammonium formate at a flow rate of 0.2 mL/min (Table 1). A sample volume of 10 μL was injected.

p 1

NH2 OH OH O NH2

N N N N N N N NH N N

H2N N NH2 HO N OH H2N N OH H2N N NH2 NH N NH2

NH O O O O O O NH O

CN H2N NH H2N NH2 H2N NH NH2 H2N NH NH NH2 H2N NH NH2

Figure 1. Potential adulterants (non-protein nitrogen sources), including melamine, cyanuric acid, ammelide, ammeline, cyromazine, dicyandiamide, urea, biuret, triuret, amidinourea, (top left to bottom right)

Table 1. LC gradient used for the separation of dicyandiamide and other Table 2. MRM transitions used for the detection of dicyanamide and potential adulterants other potential adulterants

Mobile phase A (%): Mobile phase B: Compound Polarity Q1 (amu) Q3 (amu) water with 0.1% formic acid 95% actetonitrile with 0.1% Time (min) and 10 mM ammonium formic acid and 10 mM Dicyandiamide 1 positive 85 68 formate ammonium formate Dicyandiamide 2 positive 85 43 0.0 0 100 Melamine 1 positive 127 85 2.0 0 100 Melamine 2 positive 127 68 2.1 50 50 Cyanuric acid 1 negative 128 42 4.3 50 50 Cyanuric acid 2 negative 128 85 4.4 0 100 Ammelide 1 positive 129 86 10.0 0 100 Ammelide 2 positive 129 70

Ammeline 1 positive 128 86

MS/MS Ammeline 2 positive 128 69

13 14 ® C N - The AB SCIEX QTRAP 5500 was used with the Turbo V™ 3 3 positive 133 89 Melamine source and an Electrospray Ionization (ESI) probe. The mass spectrometer was operated in Multiple Reaction Monitoring (MRM) mode using fast switching between negative and positive Results and Discussion polarity. Two selective MRM transitions were monitored for each analyte using the ratio of quantifier and qualifier ion for First, the limit of detection (LOD) and reproducibility were 13 15 identification (Table 2). C3 N3-melamine was used as an evaluated using injections of dicyandiamide standards and internal standard. spiked matrix samples.

LC-MS/MS data was processed using the MultiQuant™ software version 2.1.

p 2

Figure 2 shows a chromatogram of dicyandiamide spiked into The MRM ratios calculated across the dynamic range for milk at 2 μg/kg with a Signal-to-Noise (S/N) of 54 and 13 for the identification were found well in between the expected 25% quantifier and qualifier ion, respectively. tolerance9 of the standard ratio of 0.392. The MRM ratios were automatically calculated and reported using the ‘Multicomponent’ query in the MultiQuant™ software.

XIC of +MRM (2 pairs): 85.000/68.000 Da ID: Dicyandiamide 1 from Sample 3 (DC 0.01 ppb Matrix) of 20130201-DC.wiff (Turbo Spray), Smoothed,... Max. 1.1e4 cps.

2.06 1.10e4 In a second step the method was extended to also detect other 1.05e4 1.00e4 2 μg/kg dicyandiamide in milk known potential adulterants. An example chromatogram is 9500.00 (0.01 ng/mL in final extract) 9000.00 shown in Figure 4.

8500.00

8000.00

7500.00 Dicyandiamide (retention time, RT=2.0 min), melamine

7000.00

6500.00 (RT=4.6 min), ammeline (RT=4.7 min), ammelide (RT=4.8 min) 6000.00 were detected in positive polarity and cyanuric acid (RT=2.1 min) 5500.00 ® I n t e s i y , c ps 5000.00 in negative polarity. The fast polarity switching of the QTRAP 4500.00

4000.00 5500 system was used to detect dicyandiamide and cyanuric

3500.00

3000.00 acid in a single run.

2500.00

2000.00 1.60

1500.00

1000.00 XIC of +MRM (9 pairs): 85.000/68.000 Da ID: Dicyandiamide ... Max. 1.9e4 cps. XIC of +MRM (9 pairs): 128.000/86.000 Da ID: Ammeline 1 fr... Max. 2972.1 cps. 500.00 2.05 4.69 0.00 1.9e4 0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0 2.2 2.4 2.6 2.8 3.0 3.2 3.4 3.6 3.8 4.0 4.2 4.4 4.6 4.8 5.0 2500 Time, min 1.5e4 2000

dicyandiamide 1500 ammeline Figure 2. LC-MS/MS chromatogram of 2 μg/kg dicyanamide spiked into 1.0e4 1000 I n t e s i y, c ps I n t e s i y, c ps milk with a concentration of 0.01 ng/mL in the final extract after 200x 5000.0 500 dilution 0.0 0 0.8 1.0 1.2 1.4 1.6 1.8 2.0 2.2 2.4 2.6 2.8 3.0 3.2 3.4 3.6 2.5 3.0 3.5 4.0 4.5 5.0 5.5 6.0 6.5 Time, min Time, min XIC of +MRM (9 pairs): 127.000/85.000 Da ID: Melamine 1 fro... Max. 2.2e4 cps. XIC of +MRM (9 pairs): 129.000/86.000 Da ID: Ammelide 1 fr... Max. 5073.4 cps.

4.66 4.77 2.2e4 5000 2.0e4 4000 1.5e4 melamine 3000 ammelide

1.0e4 2000

Figure 3 shows calibration lines for dicyandiamide spiked into I n t e s i y , c ps I n t e s i y , c ps 5000.0 1000 milk, extracted using the described procedure with a total dilution 0.0 0 3.0 3.5 4.0 4.5 5.0 5.5 6.0 6.5 3.0 3.5 4.0 4.5 5.0 5.5 6.0 6.5 7.0 Time, min Time, min factor of 200x. Extensive dilution is recommended to accurately XIC of -MRM (6 pairs): 128.000/42.000 Da ID: Cyanuric acid 2 from Sample 12 (5mix 1.0ppb (200ppb) neg) of 20130130-M.wiff (Turbo Spray), Smo... Max. 1.3e4 cps. 2.05 1.3e4 quantify the target analyte in matrix samples to minimize 1.2e4 1.0e4 possible ion suppression effects which cannot be compensated 8000.0 cyanuric acid 6000.0 I n t e s i y, c ps using an internal standard. 4000.0 1.44 2000.0

0.0 0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0 2.2 2.4 2.6 2.8 3.0 3.2 3.4 3.6 3.8 4.0 4.2 4.4 Time, min Coefficients of regression were determined to be greater than 0.997 for both transitions. Figure 4. Quantitation of five potential adulterants (non-protein nitrogen sources) in a single run using fast polarity switching with the AB SCIEX QTRAP® 5500 system

6.5e5

6.0e5 dicyandiamide (1 to 40 μg/kg) 5.5e5 Figure 5 shows example calibration lines for melamine (positive

5.0e5 polarity) and cyanuric acid (negative polarity). All calibration lines 4.5e5 had r-values of greater than 0.998.

4.0e5

3.5e5 a e

r Note that the spiked matrix contained traces (< 10 μg/kg) of A 3.0e5 cyanuric acid and the calibration line does not go through zero. 2.5e5

2.0e5

1.5e5

1.0e5

5.0e4

0.0e0 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08 0.09 0.10 0.11 0.12 0.13 0.14 0.15 0.16 0.17 0.18 0.19 Concentration (ng/mL) Figure 3. Calibration lines for dicyanamide spiked into milk and analyzed after 200x dilution

p 3

Summary melamine The method and data presented here showcase the fast, easy, and accurate solutions for the analysis of dicyandiamide and other nitrogen rich compounds in milk and other protein rich foods by LC-MS/MS. The AB SCIEX QTRAP® 5500 systems provide excellent sensitivity and selectivity for this analysis, with cyanuric acid minimal sample preparation allowing maximized throughput for the analysis of many samples in a short time period.

Dicyandiamide was quantified in milk samples. Automatic MRM

ratio calculation in MultiQuant™ software was used for compound identification. Figure 5. Calibration lines for melamine and cyanuric acid spiked into milk and analyzed after 200x dilution References

1 C.A. Brown et al.: ‘Outbreaks of Renal Failure Associated Milk samples were analyzed using the developed method and with Melamine and Cyanuric Acid in Dogs and Cats in 2004 tested positive for dicyandiamide. The ‘Multicomponent’ query and 2007’ J. Vet. Diagn. Invest. 19 (2007) 525-531 2 was used to automatically calculate ratio of quantifier and H. Xin and R. Stone: ‘Tainted Milk Scandal. Chinese Probe qualifier ion for identification (Figure 6). Unmasks High-Tech Adulteration with Melamine’ Science 322 (2008) 1310-1311 3 Y.C. Tyan et al.: ‘Melamine Contamination’ Bioanal. Chem. 395 (2009) 729-735 4 S. MacMahon et al.: ‘A Liquid Chromatography–Tandem Mass Spectrometry Method for the Detection of Economically Motivated Adulteration in Protein-containing Foods’ J. Chromatogr. A 1220 (2012) 101-107 5 S. Turnipseed: ‘Determination of Melamine and Cyanuric Acid Residues in Infant Formula using LC-MS/MS’ FDA LIB 4421 (2008) 1-18 6 M. Smoker and A.J. Krynitsky: ‘Melamine and Cyanuric Acid Residues in Foods’ FDA LIB 4422 (2008) 1-28 7 T. Sakuma et al.: ‘A New, Fast and Sensitive LC-MS/MS Method for the Accurate Quantitation and Identification of

Melamine and Cyanuric Acid in Pet Food Samples’ Figure 6. Milk samples tested positive for dicyandiamide, the Application Note AB SCIEX (2010) # 1283110-01 ‘Multicomponent’ query was used to automatically calculate MRM ratios 8 E. Braekevelt et al.: ‘Determination of Melamine, Ammeline, for compound identification Ammelide and Cyanuric Acid in Infant Formula Purchased in Canada by Liquid Chromatography-Mass Spectrometry’

Food Additives & Contaminants Part A Chem. Anal. Control Expo. Risk Assess. 28 (2011) 698-704 9 Document N° SANCO/12495/2011 ‘Method Validation and Quality Control Procedure for Pesticide Residues Analysis in Food and Feed’ (2011)

For Research Use Only. Not for use in diagnostic procedures.

Publication number: 7170213-01

Headquarters International Sales 353 Hatch Drive Foster City CA 500 Old Connecticut Path, Framingham, MA 01701 USA For our office locations please call the division Phone 508-383-7700 headquarters or refer to our website at www.absciex.com www.absciex.com/offices