Validation the Method for Determination of

Melamine and Investigation Its trace in Milk

from Vietnam by LC-MS/MS

.
ABSTRACT

This work describes a rapid, selective, and sensitive method by using liquid
chromatography-tandem mass spectrometry (LC-MS/MS) to detect melamine (MEL) in milk
and dairy products. The optimal conditions of liquid chromatographic separation extraction
and mass spectroscopy of MEL have also been examined. The linear range for analyte
detected by the method was 0.5÷100.0 ng/mL, with correlation coefficients was 0.999. Mean
recoveries of the method in the real samples at three spike levels (low, medium, and high)
were within the range of 98.5% ÷102.5% (n = 7). LOD, LOQ values of the method were 10
and 30 ng/mL, respectively. The influence of the matrix effect on the accuracy, repeatability,
and recovery of the process was insignificant. The proposed method was used to quantify
the content of this compound in various real samples, which were collected in Ho Chi Minh
City-Vietnam in 2020.

Keywords: LC-MS/MS, tandem, melamine, matrix effect

1. INTRODUCTION

Melamine (MEL) is a synthesis organic most commonly observed in the form of white
crystals rich in nitrogen (named 2,4,6-triamino-1,3,5-triazine, C3H6N6). It is created in large
amounts fundamentally for use in the synthesis of melamine-formaldehyde resins for the
production of laminates, plastics, coatings, commercial filters, glues or adhesives, and
dishware and kitchenware [1](WHO, 2009). MEL matured a topic of discussion in 2007 when
veterinary scientists confirmed that pet food infection of melamine was the cause of
hundreds of pet deaths [2]. The ingestion of MEL may lead to renal failure, kidney stones,
and other health problems [3]. U.S. Food and Drug Administration (FDA) described that MEL
and cyanuric acid concentrate and interact in the urine-filled renal microtubules when they
are absorbed into the bloodstream. Then, they crystallize and procedure numerous round,
yellow crystals, which in turn block and damage the renal cells that line the tubes causing
the kidneys to malfunction[4]. In 2008, high concentrations of MEL were announced in
contaminated Chinese infant formula. More than 51,900 infants and young children in China
were hospitalized for urinary problems, possible kidney stones, possible renal tube
blockages, and related to the consumption of melamine-contaminated infant formula and
related dairy products [5]. After that time, melamine was detected in liquid milk and yogurts,
powdered milk, cereal products, confectionaries, cakes and biscuits, protein powders, frozen
desserts, and some processed foodstuffs. These foods included ammonium bicarbonate,
dried whole egg, fresh hen eggs, nondairy creamer, animal feed, and animal feed
ingredients[6]. Many countries have set maximum residue limits (MRL) for melamine in
various products to protect public health and food safety. For example, the US FDA set the
MRL of MEL in milk and dairy products and milk foods as 0.25 mg/kg and stressed that
infant formula sold to US consumers must be utterly free of MEL. The European Union (EU)
set the MRL of MEL in dairy products and high-protein foods at 2.5 mg/kg. The Ministry of
Health of China published new dairy safety standards and emphasized that food should not
be tainted with MEL. Ministry of Health of Vietnam announced and stressed that food should
not be contaminated with MEL [7]. Many determinatoion methods for MEL have been
developed, such as liquid chromatography[8], immunoassay, GC-MS, ELISA test, and liquid
chromatography-MS[9].
The MEL-contaminated milk scandal has happened over the past ten years, but to
strengthen the control from the state management agencies and raise the awareness of the
Vietnamese people about the quality of milk in general and contaminated milk MEL in
particular, the testing and analysis of this banned substance is necessary. Therefore, we
have validated the method for analyzing MEL by reversed-phase LC-MS/MS in milk and
dairy products. The procedure was then used to analyze 30 samples of dairy milk purchased
from major retailers in Vietnam.

2. MATERIAL AND METHODS

2.1. Chemicals and Reagents

All reagents were of analytical grade. Melamine (99%) standard was purchased from Sigma
Andrich (USA). Acetonitrile (ACN), formic acid (FA), HCl (36-38%), Amoniac 30%, and
Dichloromethane (CH2Cl2) were supplied from Merck (Darmstadt, Germany). Methanol was
of HPLC grade and acquired from J. T. Baker (Phillipsburg, USA). MgSO4, NaCl, and C18
powder were obtained from Waters (Milford, MA, USA).

2.2. Instrumentation

The method validation was conducted on liquid chromatography (LC) system, including the
column ZORBAX 300SB-C reversed polarized phase inversion liquid chromatography
column, 250mm x 4.6mm; particle size 5μm and thermostat autosampler (TSQ 7000,
Thermo Quest Finnigan Bremen, Germany). The equipment is combined with the Waters
TQD three quadrupole mass spectrometers with APCI ion source. Selected reaction
monitoring (SRM) was applied for data acquisition.

2.3. Chromatographic and MS conditions

For liquid chromatographic separation, the binary mobile phases were A(H 2O- HCOOH
0.1%) B (ACN:H2O (40:60)). Table 1 illustrates the mobile phase program for the loading
pumps. The flow rate was maintained at 0.6 mL/min during the whole chromatographic
analysis process. Both standard and sample solutions were held at 10°C in the sample tray.
A 5.0 µL of standard or samples was injected into the apparatus system via an autosampler.
Using the MeOH and deionized water (1: 1, v: v), cleaned triplicate the needle and the
sample loop in the autosampler. Selected Reaction Monitoring (SRM) spectra obtained in
positive ion mode were applied to identify the specified analyte.
 Table 1: The mobile phase gradient program

Flow rate
Time (min) A(%) B(%)
(mL/min)
0.00 0.6 90 10
0.20 0.6 90 10
3.00 0.6 0 100
4.50 0.6 0 100
5.00 0.6 90 10
7.00 0.6 90 10

2.3. Sample treatment

Sample collection

Samples of powdered milk, pasteurized liquid milk, sweetened condensed milk were
collected from supermarkets in Ho Chi Minh City, Vietnam, from October to November 2020.

Sample extraction

0.5 g of the homogenized sample was placed in a 10 mL test tube. 3 mL of distilled water
were added to a test tube and shaken well with a vortex machine for 3 minutes. Then, the
sample was ultrasound for 5 minutes. 6 mL of acetonitrile was added and shaken vigorously
for 3 minutes, followed by ultrasound for 5 minutes. 1 mL of 1M HCl was added and shaken
vigorously for 3 minutes, using ultrasound for 5 minutes. The sample was centrifuged at
4000 rpm for 10 minutes. 5.0 mL of solution was placed in a 30 mL test tube with a lid. Next,
3 mL of water and 15 mL of dichloromethane were added and shaken thoroughly for 2
minutes using a vortex machine. The test tubes were centrifuged at 2000 rpm for 5 minutes.
The upper layer was withdrawn and transferred to a 10-mL test tube. 2.5 mL
dichloromethane and 0.1M HCl were added. The sample was shaken well for 2 minutes and
centrifuged at 2000 rpm for 5 minutes. Finally, the supernatant was removed by pipette and
pooled into the original 10 mL test tube.

Solid-phase extraction

SPE Strata Screen-C 200mg/3mL (Phenomenex) column was connected to SPE extractor.
3.0 mL of methanol and 3 mL H2O were passed through the column. The extract is then
passed through the column and eluted with 1.0 mL (x2) 0.1 N HCl and 0.5 mL (x 2)
methanol. The MEL was eluted with 5 ml of 5% ammonia in methanol in a 50 ml flask. The
sample was then vacuum-evaporated at 45-50 ° C to dryness. The residue was then
dissolved with 1.0 mL of acetonitrile: water / 1: 4 (v / v) mixture. Finally, the solution was
filtered to a 0.45 m membrane filter into a vial for quantification.

2.4. Method validation

Standards Preparation

A stock solution of 100 μg/ml concentration of MEL was set in DIW, from which a standard
o
solution 10 μg/ml was obtained in DIW by dilution and kept at 4 C. Calibration mixtures of
concentration levels 5.0, 10.0, 20, 50.0, 100.0 and 200.0 ng/mL were freshly made in
ACN/DIW (8:2). Standard solutions were injected into the LC-MS / MS machine under the
optimum conditions selected in order of low to a high concentration. The calibration curve
equation was established by the relationship between the area of the peaks and the
concentration of standards.
MEL is considered positive in the test sample if it fully meets the criteria of Commission
Decision 2002/657/EC[10]. Analyte signal with two mass transfer modes for each substance
analysis and two mass transfer modes for the corresponding internal standards must be
appeared with a signal-to-noise ratio per ion must be ≥ 3: 1. The analyte's relative retention
time corresponds to the mean relative retention time of the calibration solution within ± 2,5%
tolerance. The peak area ratio between the different mass transfer reactions of each analyte
is within the permissible range specified in Commission Decision 2002/657/EC.

LOD and LOQ

The detection limit (LOD) of an analytical procedure is the smallest analyte content that this
procedure can detect with the statistical confidence that the sensitivity of the procedure can
be assessed. Likewise, the quantitative limit (LOQ) is the minimum content of the analyte
that can be quantified using this procedure[11]. The test was carried out as follows: MEL
standard solutions were added to the blank milk sample so that the concentration of these
substances in the sample was relatively low. Samples were treated and performed analysis
on the equipment under the conditions in section 2.3 to find a sample solution with a known
minimum standard concentration Cmin giving a signal: 3 <T = S / N <10.
LOD and LOQ were determined by the following formula: LOD = 3.S / N and LOQ = 3.LOD

Recovery (R%)

The recovery of the method is determined by the standard addition technique[12]. MEL
concentration was added to the sample at three levels of 15, 30, and 100 ng / L. Samples
were processed and measured in section 2.3. Recovery (R%) is determined by the formula:

Where:
CRe: total concentration of the sample and concentration of the added standard
Cm: concentration of the sample analyzed
Cc: concentration of the added standard

Repeatability

The evaluation of the Intra-assay imprecision and inaccuracy was conducted using an
experimental model of previous works [13]. These tests were evaluated by examining three
quality controls 15, 30, and 100 ng/L of MEL as six replicates during a single day. The mean,
standard deviation, and coefficient of variation values were measured for each quality
control. The inaccuracy of the assessments for each quality control was determined as the
distinction between the mean measured concentration and the nominal concentration as a
percentage of the nominal concentration. Inter-assay imprecision was evaluated in six
assays run on separate days with two quality controls containing MEL concentrations within
the operating range. This data was again shown as the coefficient of variation.

Matrix effect.

The evaluation of the matrix eff ect was conducted using an experimental model of previous
works [14, 15]. The matrix eff ect was examined by measuring the analyte's analytical signal
in the postextraction spiked solution and the analyte standard in a neat solution. The test
was conducted on three samples with the protocol as mentioned above. The final solution
was spiked at three levels: 15.0, 30.0, and 100 ng/L f the standard MLN. The matrix effect
was defined by the following equation:

ME (%) = X/Y×100

Where: X, Y are the chromatographic peak area of the standard in neat solution and peak
area of the standard spiked into sample solution after extraction, respectively.

3. RESULTS AND DISCUSSION

3.1. Chromatographic and MS conditions

The conditions for LC chromatography have been established with the C18 ZORBAX
300SB-CN column (250 x 4.6 mm, particle size 5.0 m), mobile phase A (H2O- HCOOH
0.1%), and B (ACN- HCOOH 0.1%) and 5.0 µL injection volume. The flow rate was set at 0.6
mL/min, MEL retention times was 4.6 min (RSD= 1.21 %). This result is entirely consistent
with the previous works.[16]

RT: 4.66
MA: 77379 NL:
SN: 72 52663
TIC F: cAPCI
SRM ms2
127.00
67.7-68.3
MS 5.00 PPB
R
el
ati
ve
A
b
u
n
da
nc
e NL:
RT: 4.68 52663
MA: 227937 TIC F: cAPCI
SN: 120 SRM ms2
127.00
84.7-85.3
MS 5.00 PPB

Time (min)

Fig 1. LC-MS/MS chromatogram of MEL its production: 127.0/85 and 127.0/68.

Mass spectrometric data were acquired in atmospheric pressure chemical ionization (APCI),
a widely used ionization technique in mass spectrometry, positive mode, using the selected
reaction monitoring (SRM) function. The instrument operates with the following parameters:
curtain gas, 25 (manufacturers units); source gas 1, 45; source gas 2, 45; CAD gas pressure
high; and nebulizer current, 3.0. Two SRM transitions were monitored for MEL. Collision
energy and collision exit potential settings were optimized for each transition during infusion
of the MEL solution. The optimization results were as follows: original ion of m/z 127 and
product ions of m/z 85, m/z 68. The m/z of 85 was chosen for the quantitation analysis. This
data is quite related to previous studies (Fig.1). This result is quite similar to the previous
researches[8, 16]

3.2. Method validation

Linearity

Linearity test solutions were adjusted from the stock MEL solution at six concentration levels
ranging from 5.0, 10.0, 20, 50.0, 100.0 and 200.0 ng/mL. A calibration curve was received by
plotting the peak area vs. concentration. A linear calibration plot was obtained over the
4 3
calibration range of 5.0-200 ng/L with regression equation y = 1.344.10 x + 1.74.10 and
correlation coefficient (r) of 0.9997. The RSD % values of the repeated injections are < 5%
within each level.

LOD and LOQ

LOD and LOQ of the method were determined at a signal-to-noise ratio of 3:1 and 10:1,
respectively, by injecting a series of diluted solutions with known concentrations. A precision
study was also carried at the LOQ level by injecting six individual preparations, and RSD %
of the peak area was calculated. The LOD was determined to be 1.0 ng/mL and 1.5 ng/mL
for baby milk and liquid milk, respectively, and the LOQ was settled to be 3.0 mg/kg and 4.5
ng/mL for baby milk and fluid milk, respectively. The RSD % of the precision study
conducted at the LOQ level was within 5%.

Table 2. The percentage recovery and matrix effect of the method at three levels of
concentration (n=6)

Matrix match
Recoveries
Spiked
Samples
(ng/mL) Mean ME RSD
Mean
RSD% (%) (%)
(%)
15.00 106.31 6.12 105.31 2.11
Children Powder
30.00 95.07 6.01 88.06 8.01
Milk
100.00 106.21 5.31 101.31 6.31
15.00 103.41 8.99 105.41 3.91
Sweeten Milk 30.00 96.51 6.71 99.58 6.71
100.00 93.11 6.48 93.11 4.29
15.00 103.381 5.12 107.31 7.12
Pasteurised milk 30.00 99.81 6.89 109.82 6.81
100.00 92.98 5.15 103.91 5.16
Percentage recovery, matrix effect

The percent recovery was expressed as the percentage of the standard recovered from the
sample matrix. All concentrations used in the examination were within the limits of the
analytical curve because samples all contained MEL. Table 2 shows that the percent
recoveries were higher than 90%, which designated that this method is reliable. Table 2
shows that at three sample backgrounds with different fat content, ME values of MEL in the
range from 88.06% to 109.82% were in an agreeable range. RSD values at various
concentrations were suitable values. These values proved that the influence of the matrix
sample on selectivity and recovery was negligible.

Repeatability

The RSD precision within-day was between 1.45 and 4.63 %. The accuracy of the method
within-day was between 95.0 and 103.5 %. The accuracy of the day-to-day data of this study
was from 95.1 to 103.8 %. The stability of the samples was found to be at least 7 days.

The LOD, LOQ, recovery, ME, and precision of this method for the measurement of MEL in
some dairy products were compared with other published techniques[8][9]. These data
indicated that a sensitive and verified HPLC-MS/MS method for determining melamine
residue in milk and dairy products was developed. The recommended approach was
sensitive, reliable, and accurate and allowed the detection of melamine residues at levels as
low as 10.1 to 12.3 μg/ kg in various dairy products. The method can be used for the routine
judgment of melamine residues in various dairy products

3.3. Application to actual samples

The proposed validated LC-MS/MS method was applied to the determination of MEL in three
different batches of the compound solution. Satisfactory results were obtained as shown in
Table 3; The results showed that MEL was not present in the samples analyzed. These data
proves that the management of milk quality in Vietnam has been much better than ten years
ago. Or, fraudulent manufacturing firms no longer exist in the dairy market.

Table 3. Result of MEL in milk samples analyses

*
Samples Found
Children Powder milk 1 ND
Children Powder milk 2 ND
Children Powder milk 3 ND
Children Powder milk 4 ND
Children Powder milk 5 ND
Children Powder milk 6 ND
Children Powder milk 7 ND
Sweeten milk 1 ND
Sweeten milk 2 ND
Sweeten milk 3 ND
Sweeten milk 4 ND
Pasteurized milk 1 ND
Pasteurized milk 2 ND
Pasteurized milk 3 ND
Pasteurized milk 4 ND
Pasteurized milk 5 ND
*
ND: none detection, values are expressed as the mean ± SD (n = 3).
4. CONCLUSIONS
In summary, this work evaluated the analytical method for MEL. The optimization conditions
of HPLC-MS / MS machine using APCI ionization source, mobile phase composition, milk
sample preparation have been thoroughly investigated. This method showed advantages
such as good qualitative and quantitative, wide linear range, high sensitivity and good
selectivity, high recovery efficiency, and relatively low detection limit and quantification limit.
The method has also been applied to analyze 20 types of milk in the Vietnamese market.
The results showed that these samples did not contain melamine.

REFERENCES
1. WHO/FAO (2009) Toxicological and health aspects of melamine and cyanuric
acid.
2. Tyan Y-C, Yang M-H, Jong S-B, et al (2009) Melamine contamination. Anal
Bioanal Chem 395:729–735. https://doi.org/10.1007/s00216-009-3009-0
3. Chen Z, Yan X (2009) Simultaneous Determination of Melamine and 5-
Hydroxymethylfurfural in Milk by Capillary Electrophoresis with Diode Array
Detection. J Agric Food Chem 57:8742–8747.
https://doi.org/10.1021/jf9021916
4. Suchý P, Straková E, Herzig I, et al (2009) Toxicological risk of melamine and
cyanuric acid in food and feed. Interdiscip Toxicol 2:.
https://doi.org/10.2478/v10102-009-0010-6
5. Xu X, Ren Y, Zhu Y, et al (2009) Direct determination of melamine in dairy
products by gas chromatography/mass spectrometry with coupled column
separation. Anal Chim Acta 650:39–43.
https://doi.org/10.1016/j.aca.2009.04.026
6. Gossner CM-E, Schlundt J, Ben Embarek P, et al (2009) The Melamine
Incident: Implications for International Food and Feed Safety. Environ Health
Perspect 117:1803–1808. https://doi.org/10.1289/ehp.0900949
7. Guo Z, Gai P, Hao T, et al (2011) Determination of melamine in dairy
products by an electrochemiluminescent method combined with solid-phase
extraction. Talanta 83:1736–1741.
https://doi.org/10.1016/j.talanta.2010.12.013
8. Filazi A, Sireli UT, Ekici H, et al (2012) Determination of melamine in milk and
dairy products by high performance liquid chromatography. J Dairy Sci
 95:602–608. https://doi.org/10.3168/jds.2011-4926
9. Lutter P, Savoy-Perroud M-C, Campos-Gimenez E, et al (2011) Screening
and confirmatory methods for the determination of melamine in cow’s milk
and milk-based powdered infant formula: Validation and proficiency-tests of
ELISA, HPLC-UV, GC-MS and LC-MS/MS. Food Control 22:903–913.
https://doi.org/10.1016/j.foodcont.2010.11.022
10. Decision 96/23/Ec Commission (2002) 96/23/EC COMMISSION DECISION
of 12 August 2002 implementing Council Directive 96/23/EC concerning the
performance of analytical methods and the interpretation of results (notified
under document number C(2002) 3044)(Text withEEA relevance)
(2002/657/EC). 96/23/Ec Comm Decis 29
11. b_Association of Analytical Communities International (2013) Guidelines for
single laboratory validation of chemical methods for dietary supplements and
botanicals. AOAC Int 1–38
12. Baenas N, Salar FJ, Domínguez-Perles R, García-Viguera C (2019) New
UHPLC-QQQ-MS/MS method for the rapid and sensitive analysis of ascorbic
and dehydroascorbic acids in plant foods. Molecules 24:1–10.
https://doi.org/10.3390/molecules24081632
13. Thang NQ, Tan L Van, Thi N, Phuong K (2016) The ICP-MS validated for
measuring arsenic levels in muscle tissues of freshwater fishes in Ho Chi
Minh City. 54:439–442. https://doi.org/10.15625/0866-7144.2016-00343
14. Matuszewski BK, Constanzer ML, Chavez-Eng CM (2003) Strategies for the
Assessment of Matrix Effect in Quantitative Bioanalytical Methods Based on
HPLC−MS/MS. Anal Chem 75:3019–3030. https://doi.org/10.1021/ac020361s
15. Tran QH, Nguyen TT, Pham KP (2020) Development of the High Sensitivity
and Selectivity Method for the Determination of Histamine in Fish and Fish
Sauce from Vietnam by UPLC-MS/MS. Int J Anal Chem 2020:1–9.
https://doi.org/10.1155/2020/2187646
16. Finete V de LM, Gouvêa MM, Marques FF de C, Pereira Netto AD (2015)
Validation of a method of high performance liquid chromatography with
fluorescence detection for melamine determination in UHT whole bovine milk.
Food Control 51:402–407. https://doi.org/10.1016/j.foodcont.2014.12.001