Journal of Integrative Agriculture 2016, 15(9): 2163–2174

Available online at www.sciencedirect.com

ScienceDirect

RESEARCH ARTICLE

Analysis of 13 kinds of steroid hormones in raw milk using

modified QuEChERS method combined with UPLC-QTOF-MS

TAN Xin-tong1, LI Zeng-mei2, 3, DENG Li-gang2, 3, ZHAO Shan-cang2, 3, WANG Ming-lin1

1
Academy of Food Science and Engineering, Shandong Agricultural University, Tai’an 271018, P.R.China
2
Institute of Agricultural Quality Standards and Testing Technology Research, Shandong Academy of Agricultural Sciences, Jinan
250100, P.R.China
3
Key Laboratory of Test Technology on Food Quality and Safety of Shandong Province, Jinan 250100, P.R.China

Abstract
Thirteen kinds of steroid hormones in raw milk (cow, goat and buffalo milk) were analyzed with ultra performance liquid
chromatography-quadrupole time of flight mass spectrometry (UPLC-QTOF-MS) after extraction and cleanup with the
modified QuEChERS method. These steroid hormones included 17β-estradiol, estriol, estrone, diethylstilbestrol, proges-
terone, melengestrol acetate, megestrol acetate, chlormadinone acetate, 19-nortestosterone, metandienone, boldenone,
epitestosterone, and testosterone. The limits of detection for the raw milk basing on 3 times the signal to noise ratios
(S/N=3) was in range of 0.07-0.51 µg kg–1, and the limits of quantification (basing on S/N=10 method) covered the rang-
es from 0.23 to 1.7 µg kg–1. With matrix external standard method, the substances presented recoveries over the range
74.2–99.7%. Qualitative analysis was also done in the mass/mass spectrum (MS/MS) mode and each debris structure
of 13 kinds of steroid hormones was achieved. The methodology was then applied in real raw milk samples which were
collected in several areas of China and the progesterone was detected with high level.

Keywords: UPLC-QTOF-MS, raw milk, QuEChERS, steroid hormones, acidic alumina

precocious puberty, breast cancer in women and prostate

1. Introduction
Milk and dairy products are one of the most popular foods
in people’s life, due to the rich in protein, vitamins, calcium
and other essential nutrients. So its quality and safety
issues are extremely important. Studies have shown that
Received 4 November, 2015 Accepted 29 April, 2016
TAN Xin-tong, E-mail: tanxintong142536@163.com;
Correspondence WANG Ming-lin, Tel: +86-538-8246029,
Fax: +86-538-8249157, E-mail: mlwang1963@163.com
© 2016, CAAS. All rights reserved. Published by Elsevier Ltd.
doi: 10.1016/S2095-3119(16)61386-2
cancer incidence rising are ascribed the hormone residues
in foods (James et al. 2011; Key et al. 2011; Sandholm et al.
2012). Hypertension in men is related to androgen levels
(Reckelhoff et al. 1998; Wu and Schwartzman 2011). The
World Anti-Doping Agency listed the nandrolone, testos-
terone, methyltestosterone and other steroid hormones as
banned steroid doping (Berry 2008; Danaceau et al. 2008).
However, many dairy farmers abused drugs in violation of
the law in order to increase raw milk production, which led
to abnormal secretion of hormones in cows and increased
the hormone level in raw milk. These would enhance the
steroid hormones excessive risks in raw milk. It is neces-
sary to explore an efficient way to detect steroid hormones
in raw milk.
Many detection methods have been established for de-
2164 TAN Xin-tong et al. Journal of Integrative Agriculture 2016, 15(9): 2163–2174
tecting steroid hormones (Stanczyk and Clarke 2010). Ra-
dioimmunoassay and enzyme-linked immunosorbent assay
provide higher sensitivity, simple operation and are suitable
for large amount measurements (Avila-Poveda et al. 2013).
However, various hormones are prone to influence each
other, resulting in contamination and false-positive in the
samples. Recently, prevalent instrumental methods using
gas chromatography tandem mass spectrometry (Hansen
et al. 2011; Pérez et al. 2014), liquid chromatography tan-
dem mass spectrometry (Ingrand et al. 2003; Farke et al.
2011; Gao et al. 2013) showed higher sensitivity, good speci-
ficity and chromatographic resolution. But their pretreatment
processes were tedious and time-consuming. Time-of-flight
mass spectrometry (TOF-MS) allows the generation of
mass information with higher accuracy and precision, and
provides both the parent ions and their fragment information.
TOF-MS was a time-saving, high sensitivity and accuracy
method. Thus, it was used for determination of 13 kinds of
steroid hormones in the study.
QuEChERS is the acronym for a quick, easy, cheap,
effective, rugged and safe method, which was developed
by U.S. Department of Agriculture, Agricultural Research
Service, USA (Anastassiades et al. 2011), and often used
as a pretreatment method when detecting pesticide residues
in fruits and vegetables (Cieślik et al. 2011; Furlani et al.
2011; Pareja et al. 2011). In this experiment, we used the
QuEChERS method for handling steroid hormones in raw
milk samples. During the purification process, we found that
original purification adsorbents would reduce the recoveries
and could not achieve the best cleanup results. To obtain a
higher recovery and better purification effect, we adjusted
the QuEChERS cleanup adsorbents and compared the
purification results of different adsorbents.
In this study, we optimized ultra performance liquid chro-
matography-quadrupole time of flight mass spectrometry
(UPLC-QTOF-MS) conditions for analyzing 13 kinds of
steroid hormones in raw milk. The modified QuEChERS
method was used to pretreat the raw milk samples and
provide good analytical results for the target hormones.
2. Materials and methods
2.1. Materials and standards
Steroid hormones, its purity >99%, were obtained from
Germany DR Company. Acetonitrile, formic acid, ammo-
nia water, anhydrous magnesium, Cleanert PSA (primary
and secondary amine) and acidic alumina are all analytical
grade; and sodium citrate, sodium citrate dibasic sesqui-
hydrate are chromatographic grade. Deionized water was
used through the experiment.
The raw milk (cow, goat and buffalo milk) were collected
from Shandong and Shanxi provinces and Guangxi Zhuang
Autonomous Region, China, at three seasons of 2014. The
samples which we received were in frozen state and kept
in the refrigerator (–18°C) until analysis.
2.2. Instruments
The main instruments were as follows: acquity ultra perfor-
mance liquid chromatography, Xevo G2-S-flight mass spec-
trometer (Waters Corporation, U.S.), Heidolph L4000 rotary
evaporator (Heidolph Corporation, Germany), 3K30-speed
rotating centrifuge (Sigma Corporation, U.S.), IKA-MS3
vortex spin oscillator (IKA Industrial Equipment Corporation,
Germany), KQ-500E ultrasonic extractor (Kunshan Ultrason-
ic Instrument Co. Ltd. China), 0.22 μm organic membrane
(Pall Corporation, U.S.), single-channel pipette (100–1000
μL) (Eppendorf Corporation, Germany).
2.3. Chromatographic conditions
A Waters BEH C18 (2.1 mm×100 mm, 1.7 μm) was used for
separating steroid hormones. Column and sample chamber
temperature were set at 35 and 20°C, respectively. The flow
rate was 0.3 mL min–1. The mobile phase consisted of 0.1%
formic acid (A) and acetonitrile (B) in positive ion mode. A
gradient elution was applied as follows: Initial 65% A, 1 min
65% A, 10 min 50% A, 10 min 6 s 65% A, 12 min 65% A.
While in negative ion mode, the mobile phase consisted of
0.1% ammonia (A) and acetonitrile (B). A gradient elution
was applied as follows: Initial 65% A, 1 min 65% A, 5 min
50% A, 9 min 20% A, 10 min 65% A. The injection volume
was 2 μL.
2.4. Mass spectrum (MS) conditions
Electro spray ionization (ESI) was set in positive electro
spray ionization mode (ESI+) and negative electro spray
ionization mode (ESI-). MS parameters were as follows:
capillary voltage, 3.0 kV in ESI+ and 2.5 kV in ESI-; desol-
vation gas flow, 800 L h-1; ion source temperature, 120°C;
desolvation temperature, 420°C; cone voltage, 40 V; scan
range, 50–600 m/z.
2.5. Final sample pretreatment
10.0 g raw milk was mixed with 10 mL acetonitrile and ul-
trasonic extracted for 10 min. Then 6.5 g QuEChERS salt
(4.0 g anhydrous magnesium, 1.0 g sodium chloride, 1.0 g
sodium citrate, 0.5 g sodium citrate dibasic sesquihydrate)
was added. QuEChERS salts are commonly used as the
drying agent to promote rapidly and satisfactorily separating
water and extraction agent to get the analytes in QuEChERS
 TAN Xin-tong et al. Journal of Integrative Agriculture 2016, 15(9): 2163–2174 2165

extraction method and also could stable the PH value (Koe-
sukwiwat et al. 2008). After an intensive shaking of 2 min,
the mixtures were centrifuged for 5 min with 7 000 r min–1.
5 mL supernatant was applied to the modified QuEChERS
purification tube (900 mg anhydrous magnesium, 150 mg
PSA and 200 mg acidic alumina). The purified samples
were shaken for 2 min, then centrifuged for 5 min with 7 000
r min–1 speed again. The supernatant (0.5 mL) was mixed
with deionized water (0.5 mL) and passed through 0.22 µm
organic membrane.
3. Results and discussion
3.2. Optimization of sample cleanup
Sample cleanup was necessary to remove interferences and
impurities (Li et al. 2014). Here, two types of absorbents,
PSA and acidic alumina were tested for 13 kinds of steroid
hormones recoveries from raw milk samples (Fig. 2). From
Fig. 2-A, the cleanup performance was not obvious when
the amount of PSA was below 50 mg. The better results
were obtained with the PSA mass close to 150 mg. After
that, the recoveries were stabilized. From Fig. 2-B, we
could see that the highest recoveries for 13 steroid hor-

3.1. Optimization of samples extraction

80
The extraction effects of solvent methanol, acetonitrile, ethyl
acetate and acetone were compared. Experimental results Recovery (%) 60
showed that the salting out effect of methanol did not meet
the experimental requirements. The extraction efficiency 40
and recoveries of acetone were under 60%. While the re-
coveries of acetonitrile and ethyl acetate could reach more 20
than 80%, and met the requirements of our experiment
(Fig. 1). The extracts with ethyl acetate were turbid, which 0
Acetone Ethyl acetate Acetonitrile
increased the difficulty of purification. But acetonitrile could
precipitate protein, and simplify the purification process, thus
it was selected as the extraction agent. Fig. 1 Recoveries of three solvents.

A 82 Testosterone
Epitestosterone
80 82 Chlormadinone acetate
Boldenone Megestrol acetate
Metandienone 81
Melengestrol acetate
80
Recovery (%)

78 19-nortestosterone 17β-estradiol
Recovery (%)

Progesterone 79 Estriol
76 78 Estrone
Diethylstilbestrol
77
74 76
75
72
74
70 73
72
0 50 100 150 200 0 50 100 150 200
Mass of PSA (mg) Mass of PSA (mg)
B Testosterone
98 Epitestosterone Chlormadinone acetate
96 Boldenone 96 Megestrol acetate
94 Metandienone 94 Melengestrol acetate
19-nortestosterone 17β-estradiol
92 92 Estriol
Recovery (%)

90 Progesterone 90 Estrone
Recovery (%)

88 88 Diethylstilbestrol
86 86
84 84
82
82 80
80 78
78 76
76 74
0 100 200 300 400 0 100 200 300 400
Mass of acidic alumina (mg) Mass of acidic alumina (mg)

Fig. 2 Recoveries of 13 steroid hormones with different mass of (A) primary and secondary amine (PSA) and (B) acidic alumina.
2166 TAN Xin-tong et al. Journal of Integrative Agriculture 2016, 15(9): 2163–2174
mones were achieved with 200 mg acidic alumina used.
That was because acidic alumina could absorb the fat of
raw milk. The modified QuEChERS cleanup absorbents
were optimized as 900 mg anhydrous magnesium, 150 mg
PSA and 200 mg acidic alumina. These absorbents could
decrease the inhibitory of matrix effects efficiently and got
a better recovery.
The cleanup time varied from 60 to 180 s for the target
compounds was also examined. From Fig. 3, it was obvious
that the extraction efficiency of all the steroid hormones first
increased with increasing cleanup time and reached the
highest recovery at about 120 s, and then decrease from 120
to 180 s. Therefore, we adjusted the cleanup time to 120 s.
3.3. UPLC-QTOF-MS methods development and
optimization
The development of the UPLC-QTOF-MS method involved
optimization of each individual component to obtain an
overall working system.
Optimization of chromatographic conditions Two WA-
TERS models columns (Waters, Massachusetts, USA),
UPLC BEH C18 (2.1 mm×100 mm, 1.7 μm) and UPLC
HSS T3 (2.1 mm×100 mm, 1.8 μm) were chosen for chro-
matographic separation. Both of them showed an efficient
separation, while the retention time of the target compounds
with T3 column were postponed for 4 to 5 min compared with
C18 column. Therefore, C18 column was used all through
the experiment. The standard chromatograms of 13 kinds
of steroid hormones were shown in Fig. 4.
Optimization of MS conditions The Q-TOF-MS determi-
nation was performed in ESI+/ESI- mode. And formic acid
(0.1%) and ammonia water (0.1%) was used as mobile
phase in ESI+ and ESI-, respectively. Thus, the ionization
efficiency of target compounds can be effectively improved
Chlormadinone acetate
100 Megestrol acetate
Melengestrol acetate
17β-estradiol
Estriol
Recovery (%)
Estrone
Diethylstilbestrol
80
60 120
Fig. 3 Recoveries of 13 kinds of steroid hormones at different cleanup time.
by 1–2 times.
Thirteen kinds of steroid hormones were dissolved in
matrix solution for MS and MS/MS analysis respectively. To
obtain the best ionization efficiency of the target compounds
and the suitable MS condition, the single impact factor was
examined. The capillary voltage, cone voltage, ion source
temperature, desolvation temperature, and desolvation
gas flow rate showed a great influence on the ionization
efficiency. The optimized MS conditions are summarized
in Section 2.4. The collision energy and debris information
under MS/MS mode are shown in Table 1.
The collision energy of 13 kinds of steroid hormones was
22-42 eV. The deviation of the theoretical and measured
debris mass was controlled in range of –1.1 to 1.8 mDa.
The possible debris structures of 13 steroid hormones were
shown in Table 2.
3.4. Assessment of matrix effects
Although a good cleanup results was got by modified QuEC-
hERS purification method, the matrix effects still could not
be completely eliminated. As reported (Tso and Aga 2010),
the matrix effects can be evaluated using the formula below:
(
The slope of the calibration curve with matrix
–1)×100%
The slope of the calibration curve without matrix
The matrix effects of 13 kinds of steroid hormones were
ranged from –39.57 to –16.87%. The negative value indi-
cates the inhibitory of matrix effects and the positive value
indicates the enhancement of matrix effects.
3.5. Linear range and limits of detection
The UPLC-QTOF-MS method was validated according to
the following design. The mixed steroid hormones samples
with different concentrations were injected into the analytical
Testosterone
Epitestosterone
Boldenone
100 Metandienone
19-nortestosterone
Progesterone
Recovery (%)
80
180 60 120 180
Time (s)
 TAN Xin-tong et al. Journal of Integrative Agriculture 2016, 15(9): 2163–2174 2167

1: TOF MS ESI+
4.83 301.217 0.0200Da
100
2.06e5
%

9.97
0
–0.00 2.00 4.00 6.00 8.00 10.00 12.00 (min)
1: TOF MS ESI+
8.69 397.238 0.0200Da
100
2.51e5
%

0
–0.00 2.00 4.00 6.00 8.00 10.00 12.00 (min)
1: TOF MS ESI+
4.52 275.201 0.0200Da
100
1.07e5
%

0
–0.00 2.00 4.00 6.00 8.00 10.00 12.00 (min)

1: TOF MS ESI+
8.54 405.183 0.0200Da
100
8.13e4
%

0
–0.00 2.00 4.00 6.00 8.00 10.00 12.00 (min)
1: TOF MS ESI+
8.49 385.238 0.0200Da
100
4.09e5
%

0
–0.00 2.00 4.00 6.00 8.00 10.00 12.00 (min)
1: TOF MS ESI+
8.57 315.232 0.0200Da
100
1.88e5
%

0
–0.00 2.00 4.00 6.00 8.00 10.00 12.00 (min)
1: TOF MS ESI+
4.12 287.201 0.0200Da
100
1.23e5
%

9.58
0
–0.00 2.00 4.00 6.00 8.00 10.00 12.00 (min)
(Continued on next page)
1: TOF MS ESI+
5.27 289.217 0.0200Da
Fig. 4100
The chromatograms of 13 kinds of standard steroid hormones. 1.08e5
5.91
%

4.14
0
–0.00 2.00 4.00 6.00 8.00 10.00 12.00 (min)
1: TOF MS ESI+
0.83 TIC
 0
–0.00 2.00 4.00 6.00 8.00 10.00 12.00 (min)
1: TOF MS ESI+
4.12 287.201 0.0200Da
100
1.23e5

%
2168 TAN Xin-tong et al. Journal of Integrative Agriculture 2016, 15(9): 2163–2174
9.58
0
–0.00 2.00 4.00 6.00 8.00 10.00 12.00 (min)

1: TOF MS ESI+
5.27 289.217 0.0200Da
100
1.08e5
5.91
%

4.14
0
–0.00 2.00 4.00 6.00 8.00 10.00 12.00 (min)
1: TOF MS ESI+
0.83 TIC
100
0.76 0.95 10.02 1.20e7
9.14 9.48
10.46
1.20 8.18 8.47 10.63
%

7.79 11.76
0.67 3.75 4.83 6.83 11.05
2.57 4.12 5.62

0
–0.00 2.00 4.00 6.00 8.00 10.00 12.00 (min)

1: TOF MS ESI–
5.77 269.154 0.0200Da
100
1.94e3
%

6.67 8.44
0
–0.00 2.00 4.00 6.00 8.00 10.00 12.00 (min)
1: TOF MS ESI–
100
4.51 271.17 0.0200Da
1.59e3
%

2.20
0
–0.00 2.00 4.00 6.00 8.00 10.00 12.00 (min)
1: TOF MS ESI–
100
1.37 287.165 0.0200Da
2.27e3
%

9.75
0
–0.00 2.00 4.00 6.00 8.00 10.00 12.00 (min)
1: TOF MS ESI–
6.67 267.139 0.0200Da
100
8.12e3

8.45
%

0
–0.00 2.00 4.00 6.00 8.00 10.00 12.00 (min)
1: TOF MS ESI–
0.75 TIC
100
0.07 9.77 9.89 9.97 3.37e5
7.17 7.35 7.86 8.01 9.27
6.16 6.66
%

0
–0.00 2.00 4.00 6.00 8.00 10.00 12.00 (min)

Fig. 4 (Continued from the preceding page)

 TAN Xin-tong et al. Journal of Integrative Agriculture 2016, 15(9): 2163–2174 2169

Table 1 Collision energy and debris information of 13 kinds of steroid hormones in the mass/mass spectrum (MS/MS) mode
Estrogen Collision energy Elemental composition Theoretical mass Experimental mass ESI mode
17β-estradiol 40 C13H11O 183.0810 183.0804 ESI-
C10H9O 145.0653 145.0649
Estriol 42 C12H11O 171.0810 178.0821 ESI-
C10H9O 145.0653 145.0642
Estrone 36 C11H11O 159.0810 159.0809 ESI-
C10H9O 145.0653 145.0644
Diethylstilbestrol 40 C16H23O 237.0619 237.0618 ESI-
C14H9O 209.0603 209.0610
Testosterone 35 C 7H 9O 109.0653 109.0654 ESI+
C 6H 9O 97.0653 97.0649
Epitestosterone 30 C 7H 9O 109.0653 109.0647 ESI+
C 6H 9O 97.0653 97.0662
Boldenone 22 C10H15 135.1174 135.1178 ESI+
C 8H 9O 121.0653 121.0652
Metandienone 30 C20H27O 283.2026 283.2027 ESI+
C8H90 121.0653 121.0659
19-nortestosterone 26 C18H25O 109.0653 109.0649 ESI+
C 7H 9O 257.1935 257.1933
Chlormadinone acetate 25 C21H25O2 309.1861 309.1863 ESI+
C21H25O2Cl 345.1646 345.1664
Megestrol acetate 33 C19H23O 267.1752 267.1750 ESI+
C22H29O2 325.2191 325.2196
Melengestrol acetate 28 C20H23O 279.1749 279.1750 ESI+
C23H29O2 337.2168 337.2166
Progesterone 30 C 6H 9O 97.0653 97.0649 ESI+
C 7H 9O 109.0686 109.0688

system. We draw standard curve and calculated the linear
correlation coefficient. The y-axis is the parent ion chromato-
gram peak area and the x-axis is the steroid hormone con-
centration. The limits of detection for each steroid hormone
were determined basing on 3 times the signal to noise ratios
(S/N=3), while the limits of quantification were determined
basing on 10 times the signal to noise ratios (S/N=10). The
linearity of the analytical response obtained for all 13 steroid
hormones across the studied range of concentrations (1–5
µg kg-1) was excellent. Table 4 includes the results ob-
tained for the linear equation, linear correlation coefficients,
LOD and LOQ from the different linear correlation curves
evaluated along the complete validation study. Correlation
coefficients of the linear correlation curves are higher than
0.995 in all cases. The limits of detection basing on S/N=3
method in raw cow milk, raw goat milk and raw buffalo milk
were in range of 0.07–0.51, 0.08–0.53 and 0.07–0.50 µg
kg–1, respectively. While the limits of quantification basing
on S/N=10 in the three raw milk were in range of 0.23-1.7,
0.26-1.7 and 0.23-1.7 µg kg–1, respectively.
Compared to GB22973-2008 (2009), the limits of detec-
tion of our method were lower, and we could analyze 13
kinds of steroid hormones by treating one sample. While,
the method of GB22973-2008 could only analyze 3 kinds
of steroid hormones.
3.6. Recoveries and precision
The accuracy and reproducibility of the assessed quantita-
tive method were evaluated by means of recovery tests on
blank milk samples spiked at three concentration levels (5,
10, 50 µg kg–1), with five replicates at each level. Recovery
values and the relative standard deviations (RSD) obtained
on each matrix and spiking level is indicated in Table 5. The
average recoveries of steroid hormones were in range of
74.2–99.8%. The RSD values were lower than 11.2%. All
this values indicated that the method achieved the satisfied
results.
3.7. Determination of real samples
The real milk samples were collected at three different
months (June, September and October) in 2014. The
raw cow milk and raw goat milk samples were collected
in Shandong and Sichuan provinces, China, while the raw
buffalo milk samples were collected in Guangxi Zhuang
Autonomous Region, China. Ten samples were collected
at each time. The screening of all the raw milk samples
was performed by UPLC-QTOF-MS analysis with modified
QuEChERS method mentioned in Section 2.5.
The progesterone was detected in raw cow milk and
2170 TAN Xin-tong et al. Journal of Integrative Agriculture 2016, 15(9): 2163–2174

Table 2 Parent ion and debris structures of 13 kinds of steroid hormones

Estrogen Parent ion structures and debris structures
17β-estradiol OH

HO HO
HO

OH
Estriol
OH

HO HO HO

Estrone O

HO HO
HO

Diethylstilbestrol HO
HO HO

OH OH
OH

Testosterone OH

OH

O O

Epitestosterone
OH

OH

O O

OH
Boldenone

O
O
OH
Metandienone

O O
O

OH
19-nortestosterone

O O
O

Chlormadinone acetate O
O O
O O

O
Cl O O
Cl

Megestrol acetate O O
O
O

O
O
O

Melengestrol acetate O O
O O
O

O
O

Progesterone O

O
O
 TAN Xin-tong et al. Journal of Integrative Agriculture 2016, 15(9): 2163–2174 2171

Table 3 Assessment of matrix effects for 13 kinds of steroid hormones

Slope of calibration curve without Slope of calibration curve with matrix Matrix effect (%)
Estrogens
matrix (cow milk, goat milk, buffalo milk) (cow milk, goat milk, buffalo milk)
17β-estradiol 22.41 14.22/15.26/16.14 –36.55/–31.91/–27.98
Estriol 24.55 18.24/19.12/17.45 –25.70/–22.12/–28.92
Estrone 32.14 23.52/25.36/26.37 –26.82/–21.10/–17.95
Diethylstilbestrol 22.81 15.35/14.65/15.40 –32.71/–37.52/–32.49
Testosterone 10.28 7.849/6.918/7.779 –23.65/–32.70/–24.33
Epitestosterone 9.782 7.235/7.004/7.352 –26.04/–28.39/–24.84
Boldenone 11.25 7.662/8.163/8.178 –31.89/–27.44/–27.31
Metandienone 16.56 11.38/12.14/11.15 –31.28/–26.69/–32.67
19-nortestosterone 16.18 12.85/13.45/12.35 –20.58/–16.83/–23.67
Chlormadinone acetate 7.895 4.771/5.176/6.438 –39.57/–34.44/–31.12
Megestrol acetate 24.52 17.52/18.46/19.21 –28.54/–24.72/–21.66
Melengestrol acetate 21.44 14.14/15.35/14.46 –34.05/–28.41/–32.57
Progesterone 22.48 16.31/17.21/18.34 –27.48/–23.44/–18.42

Table 4 Linear equations, correlation coefficient, limits of detection (LOD) and limits of quantification (LOQ) obtained along the
complete validation study for 13 kinds of steroid hormones
Correlation LOD (µg kg–1) LOQ (µg kg–1)
Estrogens Linear equation
coefficient (cow milk, goat milk, buffalo milk) (cow milk, goat milk, buffalo milk)
17β-estradiol Y=14.22X+312.1 0.997 0.31/0.34/0.32 1.0/1.1/1.1
Estriol Y=18.24X+152.6 0.999 0.28/0.30/0.28 0.92/1.0/0.92
Estrone Y =23.52X+453.6 0.995 0.33/0.36/0.33 1.1/1.2/1.1
Diethylstilbestrol Y =15.35X+458.2 0.995 0.37/0.40/0.34 1.2/1.3/1.1
Testosterone Y =7.849X+34.22 0.997 0.21/0.26/0.23 0.70/0.86/0.76
Epitestosterone Y =7.235X+40.18 0.999 0.51/0.0.53/0.50 1.7/1.7/1.7
Boldenone Y =7.662X–2.851 0.995 0.28/0.30/0.29 0.93/0.99/0.96
Metandienone Y =10.38X+26.07 0.997 0.13/0.15/0.11 0.43/0.50/0.36
19-nortestosterone Y =12.85X+26.07 0.996 0.21/0.22/0.21 0.70/0.73/0.70
Chlormadinone acetate Y =4.771X+28.21 0.996 0.30/0.32/0.31 1.0/1.1/1.0
Megestrol acetate Y =17.52X+930.2 0.998 0.07/0.08/0.07 0.23/0.26/0.23
Melengestrol acetate Y =14.14X+12.45 0.999 0.11/0.14/0.12 0.37/0.46/0.39
Progesterone Y =16.31X+985.6 0.998 0.19/0.22/0.19 0.63/0.73/0.63

Table 5 Recoveries (Rec) and Relative Standard Deviation (RSD) of steroid hormonefrom spiked blank milk samples
Raw cow milk Raw goat milk Raw buffalo milk
Estrogens 5 µg kg–1 10 µg kg–1 50 µg kg–1 5 µg kg–1 10 µg kg–1 50 µg kg–1 5 µg kg–1 10 µg kg–1 50 µg kg–1
Rec. RSD Rec. RSD Rec. RSD Rec. RSD Rec. RSD Rec. RSD Rec. RSD Rec. RSD Rec. RSD
17β-estradiol 77.8 6.5 83.2 4.2 88.9 4.3 77.7 7.9 80.2 3.8 85.6 3.3 76.2 9.2 87.6 4.2 90.3 2.3
Estriol 79.5 9.9 89.6 3.8 92.5 3.6 79.6 8.4 83.6 2.9 88.7 2.5 80.6 7.3 88.6 3.5 91.3 3.1
Estrone 82.3 10.4 88.7 4.4 94.3 2.5 80.4 9.5 85.2 2.4 92.6 2.9 77.2 11.2 85.2 3.8 95.2 2.9
Diethylstilbestrol 76.6 7.9 91.2 3.1 91.6 3.4 79.4 10.2 90.3 3.4 95.4 3.8 78.2 6.2 86.3 4.6 96.1 3.5
Testosterone 80.6 8.7 88.3 5.2 96.5 3.8 81.9 7.6 89.3 2.8 96.3 5.2 77.2 7.6 90.2 2.5 98.7 2.8
Epitestosterone 80.4 9.6 89.4 5.6 95.3 4.2 82.6 4.9 91.2 3.3 95.7 3.4 79.6 6.9 92.5 3.8 98.5 3.6
Boldenone 75.5 7.9 83.6 5.2 94.2 3.9 80.5 8.6 90.7 4.2 92.1 2.8 80.4 7.1 90.7 2.9 99.7 2.6
Metandienone 79.6 11.2 86.9 4.4 92.1 4.2 81.2 6.9 84.5 3.9 93.2 2.9 74.2 6.5 93.0 4.2 99.5 3.5
19-nortestosterone 76.4 10.3 90.5 6.2 92.3 5.0 78.4 9.6 86.2 4.5 94.2 3.4 90.4 6.4 85.6 4.8 92.1 2.8
Chlormadinoneaccetae 77.1 6.6 93.6 2.9 98.4 2.9 79.4 6.2 92.5 4.2 96.5 3.3 77.8 9.6 86.7 3.6 93.1 3.2
Megestrol acetate 81.5 6.4 90.2 4.5 96.2 3.8 76.5 7.3 94.5 5.2 94.2 2.0 79.2 8.7 89.3 2.8 96.2 3.4
Melengestrol acetate 77.3 8.6 91.6 4.9 95.6 3.6 77.2 5.5 92.3 2.2 90.8 1.8 80.5 7.5 88.6 2.9 94.2 2.6
Progesterone 82.5 7.4 92.4 3.0 99.8 2.6 81.2 6.3 94.2 1.9 99.8 1.6 83.2 7.8 85.2 3.5 95.6 2.1

buffalo milk. Other kinds of steroid hormones were not found Table 6. The chromatogram and spectrum of the positive
in them. The detection rate and content range are listed in sample in which the progesterone was detected are shown
2172 TAN Xin-tong et al. Journal of Integrative Agriculture 2016, 15(9): 2163–2174

in Fig. 5. No other kind of steroid hormones were detected 2010), the contents of progesterone in milk were approxi-
in all raw goat milk samples. mately same. Meanwhile, detection of one sample using
Compared with other previously reports (Chen et al. our method can be completed in only 20 min. This proposed

Table 6 Detection rate and content range of samples

June September October
Sample Detection rate Content range Detection rate Content range Detection rate Content range
(%) (µg kg–1) (%) (µg kg–1) (%) (µg kg–1)
Raw goat milk (Shandong) 0 0 0 0 0 0
Raw cow milk (Shandong) 100 1.35–4.28 100 1.78–5.52 100 1.02–4.32
Raw goat milk (Shanxi) 0 0 0 0 0 0
Raw cow milk (Shanxi) 100 1.81–5.89 100 2.51–6.13 100 1.61–5.08
Raw buffalo milk (Guangxi) 20 0.93–2.23 30 1.82–3.21 50 0.87–2.81

1: TOF MS ESI+
8.52 315.232 0.0200Da
100
1.65e3
%

8.03 9.30 10.24 10.65

3.24 4.31 5.355.795.867.027.14 7.79
1.00 1.13 1.84 2.42 2.92 4.63 8.82 11.86
0 Time
–0.00 2.00 4.00 6.00 8.00 10.00 12.00 (min)

1: TOF MS ESI+
315.2332 4.98e3
100
%

315.1537
314.2802 314.4489 315.1027 315.2493
314.7171
314.5968
314.9094 315.6464 315.7086315.7847
0 m/z
315

1: TOF MSMS 314.23ESI+

109.0686 1.74e4
100

97.0682
%

123.0839 147.1204
40.550452.3993 67.0572 85.068795.0888 159.1200173.1376 185.1404
0 m/z
20 30 40 50 60 70 80 90 100 110 120 130 140 150 160 170 180 190 200

Fig. 5 The chromatogram and mass spectrum (MS) of real samples.

 TAN Xin-tong et al. Journal of Integrative Agriculture 2016, 15(9): 2163–2174
analytical method is easy, convenient and time-saving.
4. Conclusion
In this work, a simple and rapid method for the analysis of
13 kinds of steroid hormones in raw milk samples was de-
veloped basing on modified QuEChERS purification coupled
with UPLC-QTOF-MS determination. The modified QuEC-
hERS cleanup method may provide a potential reference in
detection of steroid hormones or compounds with the similar
structure in milk or meat. The low limits of detection and
high recovery of most of steroid hormones were obtained
with the developed method. Furthermore, a high level of
progesterone was detected validly in raw cow milk of several
areas of China by using the modified QuEChERS method
combined with UPLC-QTOF-MS.
Acknowledgements
This research was supported by the Natural Science Foun-
dation of Shandong Province, China (ZR2015CM015) and
the Special Fund for Agro-Scientific Research in the Public
Interest, China (201403071). The authors show great ap-
preciation to Shandong Academy of Agricultural Sciences,
China, for supporting the laboratory equipments.
References
Anastassiades M, Lehotay S J, Stajnbaher D, Schenck F J.
2003. Fast easy multiresidue method employing acetonitrile
extraction/partitioning, “dispersive solid-phase extraction”
for the determination of pesticide residues in produce.
Journal of Aoac International, 86, 412-431.
Avila-Poveda O H, Montes-Pérez R C, Benitez-Villalobos F,
Rosas C. 2013. Development and validation of a solid-
phase radioimmunoassay for measuring progesterone
and testosterone in octopus gonad extracts. Malacologia,
56, 121–134.
Berry D A. 2008. The science of doping. Nature, 454, 692–693.
Cieślik E, Sadowska-Rociek A, Ruiz J M M, Surma-Zadora
M. 2011. Evaluation of QuEChERS method for the
determination of organochlorine pesticide residues in
selected groups of fruits. Food Chemistry, 125, 773–778.
Chen X H, Sheng X F, Jin M C. 2010. Simultaneous
determination of eight progesterone derivatives residues
in milk by ultra-fast liquid chromatography coupled with
tandem mass spectrometry. Chinese Journal of Health
Laboratory Technology, 11, 2671–2675. (in Chinese)
Danaceau J P, Scott Morrison M, Slawson M H. 2008. Quantitative
confirmation of testosterone and epitestosterone in human
urine by LC/Q-ToF mass spectrometry for doping control.
Journal of Mass Spectrometry, 43, 993–1000.
Farke C, Rattenberger E, Roiger S U, Meyer H H. 2011.
2173
Bovine colostrum: Determination of naturally occurring
steroid hormones by liquid chromatography-tandem mass
spectrometry (LC-MS/MS). Journal of Agricultural and Food
Chemistry, 59, 1423–1427.
Furlani R P Z, Marcilio K M, Leme F M, Tfouni S A V. 2011.
Analysis of pesticide residues in sugarcane juice using
QuEChERS sample preparation and gas chromatography
with electron capture detection. Food Chemistry, 126,
1283–1287.
Gao W, Stalder T, Foley P, Rauh M, Deng H, Kirschbaum C.
2013. Quantitative analysis of steroid hormones in human
hair using a column-switching LC-APCI-MS/MS assay.
Journal of Chromatography (B), 928, 1–8.
GB/T 22973-2008. 2009. Determination of melengestrol
acetate,chlormadinone acetate and megestrol acetate
residues in milk and milk powder—HPLC-MS-MS method.
China Standards Press. pp. 20-21
Hansen M, Jacobsen N W, Nielsen F K, Björklund E,
Styrishave B, Halling-Sørensen B. 2011. Determination of
steroid hormones in blood by GC-MS/MS. Analytical and
Bioanalytical Chemistry, 400, 3409–3417.
Ingrand V, Herry G, Beausse J, de Roubin M R. 2003. Analysis
of steroid hormones in effluents of wastewater treatment
plants by liquid chromatoraphy-tandem mass spectrometry.
Journal of Chromatography (A), 1020, 99–104.
James R E, Lukanova A, Dossus L, Becker S, Rinaldi S,
Tjønneland A, Olsen A, Overvad K, Mesrine S, Engel P.
2011. Postmenopausal serum sex steroids and risk of
hormone receptor-positive and -negative breast cancer: a
nested case-control study. Cancer Prevention Research,
4, 1626–1635.
Key T, Appleby P, Reeves G, Roddam A, Helzlsouer K, Alberg
A, Rollison D, Dorgan J, Brinton L, Overvad K. 2011.
Circulating sex hormones and breast cancer risk factors in
postmenopausal women: Reanalysis of 13 studies. British
Journal of Cancer, 105, 709–722.
Koesukwiwat U, Sanguankaew K, Leepipatpiboon N. 2008.
Rapid determination of phenoxy acid residues in rice by
modified QuEChERS extraction and liquid chromatography-
tandem mass spectrometry. Analytica Chimica Acta, 626,
10–20.
Li Y F, Qiao L Q, Li F W, Ding Y, Yang Z J, Wang M L. 2014.
Determination of multiple pesticides in fruits and vegetables
using a modified quick, easy, cheap, effective, rugged
and safe method with magnetic nanoparticles and gas
chromatography tandem mass spectrometry. Journal of
Chromatography (A), 1361, 77–87.
Pareja L, Cesio V, Heinzen H, Fernández-Alba A R. 2011.
Evaluation of various QuEChERS based methods for the
analysis of herbicides and other commonly used pesticides
in polished rice by LC-MS/MS. Talanta, 83, 1613–1622.
Pérez R A, Albero B, Tadeo J L, Molero E, Sánchez-Brunete C.
2014. Analysis of steroid hormones in water using palmitate-
coated magnetite nanoparticles solid-phase extraction
and gas chromatography-tandem mass spectrometry.
2174 TAN Xin-tong et al. Journal of Integrative Agriculture 2016, 15(9): 2163–2174
Chromatographia, 77, 837–843.
Reckelhoff J F, Zhang H, Granger J P. 1998. Testosterone
exacerbates hypertension and reduces pressure-natriuresis
in male spontaneously hypertensive rats. Hypertension,
31, 435–439.
Stanczyk F Z, Clarke N J. 2010. Advantages and challenges
of mass spectrometry assays for steroid hormones. The
Journal of Steroid Biochemistry and Molecular Biology,
121, 491–495.
Tso J, Aga D S. 2010. A systematic investigation to optimize
simultaneous extraction and liquid chromatography
tandem mass spectrometry analysis of estrogens and their
conjugated metabolites in milk. Journal of Chromatography
(A), 1217, 4784–4795.
Wu C C, Schwartzman M L. 2011. The role of 20-HETE in
androgen-mediated hypertension. Prostaglandins & Other
Lipid Mediators, 96, 45–53.
(Managing editor WENG Ling-yun)