Comprehensive and highly sensitive urinary steroid hormone profiling method based on stable isotope-labeling liquid chromatography-mass spectrometry

Article abstract of DOI:10.1021/ac301984t

Steroid hormones are crucial substances that mediate a wide range of vital physiological functions of the body. Determination of the levels of steroid hormones plays an important role in understanding the mechanism of the steroid hormone-related diseases. In this study, we present a novel targeted metabolic profiling method based on the introduction of an easily protonated stable isotope tag to a hydroxyl-containing steroid hormone with a synthesized derivatization reagent, deuterium 4-(dimethylamino)-benzoic acid (d ₄-DMBA), and liquid chromatography-mass spectrometry (LC-MS). Different from other reported derivatization reagents that have been used to enhance the sensitivities for estrogens or androgens, our method is comprehensive with the capability of covering hydroxyl-containing androgens, estrogens, corticoids, and progestogens. Furthermore, the nonderivatized steroid hormones (e.g., 17α-hydroxyprogesterone, progesterone, and androstenedione) were not destroyed during the derivatization process, and their levels could still be obtained in one LC-MS run. We were able to detect 24 steroid hormones at subng/mL levels (the lower limit of detection could reach 5 pg/mL for estrone and 16α-hydroxy estrone, which is equivalent to 0.1 pg on column) with maximum sensitivity enhancement factors of more than 10 ³- to 10⁴-fold after derivatization. The method was successfully applied to the measurement of free (unconjugated) steroid hormones in urine samples of males, females, and pregnant women. Because the significant role the steroid hormone pathway plays in humans, a comprehensive, sensitive, specific, and accurate method for profiling the steroid hormone metabolome shall offer new insights into hormone-related diseases.

Full text of DOI:10.1021/ac301984t

Article
pubs.acs.org/ac
Comprehensive and Highly Sensitive Urinary Steroid Hormone
Profiling Method Based on Stable Isotope-Labeling Liquid
Chromatography−Mass Spectrometry
Weidong Dai, Qiang Huang, Peiyuan Yin, Jia Li, Jia Zhou, Hongwei Kong, Chunxia Zhao, Xin Lu,
and Guowang Xu*
CAS Key Laboratory of Separation Science for Analytical Chemistry, Dalian Institute of Chemical Physics, Chinese Academy of
Sciences, Dalian 116023, China
*
S Supporting Information
ABSTRACT: Steroid hormones are crucial substances that mediate a
wide range of vital physiological functions of the body. Determination
of the levels of steroid hormones plays an important role in
understanding the mechanism of the steroid hormone-related diseases.
In this study, we present a novel targeted metabolic profiling method
based on the introduction of an easily protonated stable isotope tag to
a hydroxyl-containing steroid hormone with a synthesized derivatiza-
tion reagent, deuterium 4-(dimethylamino)-benzoic acid (d -DMBA),
4
and liquid chromatography−mass spectrometry (LC-MS). Different
from other reported derivatization reagents that have been used to enhance the sensitivities for estrogens or androgens, our
method is comprehensive with the capability of covering hydroxyl-containing androgens, estrogens, corticoids, and progestogens.
Furthermore, the nonderivatized steroid hormones (e.g., 17α-hydroxyprogesterone, progesterone, and androstenedione) were
not destroyed during the derivatization process, and their levels could still be obtained in one LC-MS run. We were able to detect
2
4 steroid hormones at subng/mL levels (the lower limit of detection could reach 5 pg/mL for estrone and 16α-hydroxy estrone,
3
4
which is equivalent to 0.1 pg on column) with maximum sensitivity enhancement factors of more than 10 - to 10 -fold after
derivatization. The method was successfully applied to the measurement of free (unconjugated) steroid hormones in urine
samples of males, females, and pregnant women. Because the significant role the steroid hormone pathway plays in humans, a
comprehensive, sensitive, specific, and accurate method for profiling the steroid hormone metabolome shall offer new insights
into hormone-related diseases.
teroid hormones are among the most important species of
endogenous metabolites, regulating the life action for
Therefore, attention needs to be paid to the risk of a
pathological result obtained from the measurement of single
hormone metabolites. In contrast, detecting the steroid
hormone metabolome is helpful for holistically understanding
the alteration of the steroid hormone pathway, and leads to a
reduction in the risk of false positive/negative diagnosis error.
However, due to their extremely low abundance in body fluids,
the detection of steroid hormones is still a very challenging
S
human beings by participating in the maintenance of secondary
sexual characteristics, modulation of gene transcription, and
regulation of endocrine and immune functions, etc. Disorders
of the steroid hormones homeostasis are related to various
diseases. Recent studies showed that the steroid hormones were
implicated in the process of cancer diseases. It was reported
that breast cancer is often correlated with high circulating and
11
task.
1
−3
urinary endogenous estrogens levels.
Sex hormones,
The current methods for measuring endogenous steroid
hormones have involved immunoassay, gas chromatography−
mass spectrometry (GC-MS), and liquid chromatography−
especially androgens, are involved in the etiology of prostate
4
−6
cancer.
Steroid hormones are also involved in the homeo-
stasis of immune/inflammatory diseases such as rheumatoid
12
7
,8
mass spectrometry (LC-MS). Although immunoassay can
arthritis and systemic lupus erythematosus. Quantitatively
measuring a series of steroid hormones (ideally the steroid
hormone metabolome) will play a vital role in understanding
disease progression and individual biochemical responses to the
disease, as well as in achieving better personalized medicine.
In previous studies, analytical methods to quantify one or
several steroid hormones have been widely employed in
analyze one hormone sensitively, it suffers from poor specificity,
accuracy, and reproducibility because of matrix interferences
13,14
and the high lot-to-lot variability of antibodies.
The
specificity of the measurements of steroid hormones signifi-
cantly improves when using GC-MS and LC-MS. However,
9
,10
pathological studies and clinic diagnoses.
However, steroid
Received: July 18, 2012
hormone levels can vary a great deal from one individual to
another and can also be affected by the menstrual cycle.
Accepted: October 30, 2012
Published: October 30, 2012
©
2012 American Chemical Society
10245
dx.doi.org/10.1021/ac301984t | Anal. Chem. 2012, 84, 10245−10251

Analytical Chemistry
Article
because of limited sensitivity and labor-intensive and time-
consuming sample preparation, the GC-MS approach has not
become a routine method for measuring steroid hormones in
Dr. Ehrenstorfer (Augsburg, Germeny). Dichloromethane of
HPLC grade was purchased from Tedia (Fairfield, OH, USA).
Acetonitrile and methanol of HPLC grade were purchased from
Merck (Darmstadt, Germany). 4-Dimethylaminopyridine
(DMAP) and 1-ethyl-3-(3-dimethylaminopropyl) carbodiimide
(EDC) were purchased from Sigma Chemical Co. (St. Louis).
4-(Dimethylamino)-benzoic acid (DMAB) was purchased from
Alfa Aesar (Ward Hill, MA, USA).
12,14
clinical laboratories.
LC-MS provides good specificity, high
sensitivity, and the ability to measure multiple analytes.
Therefore, LC-MS is a preferable choice for profiling the
15
steroid hormone metabolome. A number of LC-MS-based
metabolomic studies have been carried out to analyze as many
1
3,16−20
steroid hormones as possible.
However, it is unable to
Synthesis of d -4-(Dimethylamino)-benzoic Acid.
4
profile the steroid hormone metabolome with only one LC-MS
Figure 1A shows the synthetic scheme for d -4-(dimethylami-
4
platform. It is necessary to use various types of ionization
1
5,21
sources and polarities to achieve satisfactory sensitivity.
Derivatization is a widely used approach to improve
ultraviolet (UV), fluorescence (FL), and mass spectrometry
(
MS) signals. In LC-MS-based studies, the derivatization
reagent reacts with a functional group of the targeted
compounds, such as an amino, carboxyl, hydroxyl, or carbonyl
group, to improve column retention and ionization efficiency
22
and to overcome the switch of the ionization mode. Dansyl
chloride is the most popular derivatization reagent used for
esterifying the hydroxyl group to enhance the sensitivity for the
1
7−19
measurement of estrogens
and unactivated alcohols
Figure 1. Reaction schemes for (A) synthesis of the stable isotope-
including testosterone, 6β-OH testosterone, and hydrocorti-
labeling reagent, d -DMAB, and (B) steroid hormones derivatized by
2
3
4
sone. Some novel derivatization reagents, such as picolinic
DMAB.
9
24
acid, N,N-dimethylglycine, 1,2-dimethylimidazole-4-sulfonyl
2
5
26
chloride, and Sanger’s reagent and analogue, have also been
used to react with steroid hormones to introduce a high
ionization tendency moiety for sensitivity enhancement.
Serious matrix effects caused by endogenous metabolites
occur during the measurement of low abundance metabolites
no)-benzoic acid (d -DMAB). The detailed procedures of the
4
synthesis and purification are described in the Supporting
Information. The confirmation and purity was tested against
commercial 4-(dimethylamino)-benzoic acid by thin layer
chromatography (TLC), UV, LC-UV-MS, and nuclear
magnetic resonance (NMR).
Urine Sample Collection. The urine samples were
collected from volunteers after a morning of fasting. The
volunteers included five males, five nonpregnant females, and
two pregnant women (3 and 4 months pregnant). The ages of
the 12 volunteers ranged from 23 to 28 years. The urine
samples were centrifuged at 10 000 rpm for 10 min. The
supernatants were stored at −20 °C.
Three pooled urine samples were prepared (before the
steroid hormone extraction and derivatization) by pooling the
five samples from the male volunteers into the first sample, the
five samples from the female volunteers into the second sample,
and the two samples from the pregnant volunteers into the
third sample.
Extraction of Urinary Steroid Hormones. An optimized
SPE procedure was applied for the removal of the bulk of the
polar solutes in urine. The SPE extraction cartridge (Waters
Oasis HLB, 30 mg) was activated with 2 mL of methanol and
was then washed with 2 mL of water. The mixed solution of 0.5
mL of urine and 0.4 mL of methanol was loaded onto the SPE
extraction cartridge. Washing with 3 mL of 45% methanol prior
to the elution with 1.5 mL of 90% methanol caused the eluted
solution to be lyophilized pending derivatization.
(
e.g., steroid hormones) in LC-MS-based complex sample
studies. Stable isotope internal standards are widely used to
overcome the matrix effects for the accurate absolute/relative
quantification. However, it is impossible to obtain all the stable
isotope internal standards of a metabolome. The stable isotope-
tagged derivatization is an alternative way to introduce a stable
isotope-tagged moiety to the endogenous compounds and
therefore perform as stable isotope internal standards. This
27−32
isotope labeling method has been applied in metabolomics
3
3−35
and proteomics
studies.
Urine is a sample that is very easy and noninvasive to obtain.
Detecting urinary steroid hormones has shown potential in
36
disease biomarker discovery. In this study, we present a
method that integrates LC-MS with solid phase extraction
(
SPE) and derivatization to profile free (unconjugated)
endogenous steroid hormones in urine. Deuterium 4-
dimethylamino)-benzoic acid was synthesized as a novel
(
derivatization reagent to label steroid hormones for sensitivity
enhancement and stable isotope-based quantification.
EXPERIMENTAL SECTION
■
Chemicals. Estrone (E ), β-estradiol (E ), 2-hydroxy
1
2
estrone (2-OH E ), 16α-hydroxy estrone (16α-OH E ), estriol
1
1
(
E3), testosterone, dehydroepiandrosterone (DHEA), 5a-
androstan-17b-ol-3-one (5α-DHT), 5b-androstan-17b-ol-3-one
Derivatization. The derivatization scheme for the steroid
hormones using DMAB in the presence of DMAP and EDC is
displayed in Figure 1B. The procedure was optimized as
follows: 80 μL of 2 mg/mL DMAB solution, 80 μL of 2 mg/
mL DMAP solution, and 80 μL of 5 mg/mL EDC solution
were added into the lyophilized sample. Dichloromethane was
used as the solvent. After spirited vortex, the solution was
transferred into a glass vial and sealed. The vial was kept at 75
°C for 5 h. The dichloromethane was then evaporated at 45 °C.
(
(
5β-DHT), mesterolone, pregnenolone, deoxycorticosterone
DOC), 17α-hydroxypregnenolone (17OH-Preg), corticoster-
one, cortexolone, tetrahydrodeoxycortisol (THS), hydrocorti-
sone, androsterone, progesterone, 17α-hydroxyprogesterone
(
17-OHP), D-androsten-3b,17b-diol (5-androstenediol), 5a-
tetrahydrocorticosterone (5a-THB), and aldosterone were
purchased from the International Laboratory (South San
Francisco, CA, USA). Androstenedione was purchased from
1
0246
dx.doi.org/10.1021/ac301984t | Anal. Chem. 2012, 84, 10245−10251

Analytical Chemistry
Article
Figure 2. Structures and the related steroid hormones biosynthesis pathway of the 24 selected standards. THS, 5a-THB, and mesterolone were listed
with a dashed line frame because they were not found in the given pathway from the KEGG database.
Figure 3. Extracted ion chromatogram (EIC) of the 21 DMAB derivatized steroid hormones in urine spiked with 20 ng/mL standards mixture by
LC-MS. Identification: 1, Aldosterone; 2, hydrocortisone; 3, E ; 4, corticosterone; 5, 5a-THS; 6, cortexolone; 7, THS; 8, DOC; 9, E ; 10, E ; 11,
3
2
1
testosterone; 12, 5-androstenediol; 13, 2-OH E ; 14, androsterone; 15, 17OH-Preg; 16, DHEA; 17, 5α-DHT; 18, 5β-DHT; 19, 16α-OH E ; 20,
1
1
mesterolone; 21, pregnenolone.
1
0247
dx.doi.org/10.1021/ac301984t | Anal. Chem. 2012, 84, 10245−10251

Analytical Chemistry
Article
Table 1. LLOD, Intrabatch Precision, Interbatch Precision, Linear Range, Recovery, and Stability within 48 h of Urinary Steroid
Hormones Measurement, Including SPE Extraction, Derivatization, and LC-MS Analysis Steps
recovery (%)
steroid
hormones
LLOD
(pg/mL)
intrabatch precision interbatch precision linear range
regression
coefficient (r)
4 ng/
mL
8 ng/ 16 ng/ stability within
no.
(%, n = 6)
(%, n = 6)
(pg/mL)
mL
mL
48 h (%)
1
2
3
4
5
6
7
8
9
aldosterone
hydrocortisone
E3
200
20
10.2
2.3
11.8
5.2
200−20000
10−20000
10−5000
10−20000
100−5000
10−20000
100−20000
5−20000
0.967
0.995
0.999
0.996
0.999
0.996
0.996
0.998
0.993
0.997
0.999
0.998
0.970
0.995
0.998
0.999
0.995
0.993
0.994
0.999
0.986
134.9
137.4
106.1
102.7
119.4
100.1
106.6
103.5
108.5
106.8
88.5
109.5
94.9
88.8
91.1
106.0
90.0
90.9
91.8
95.4
93.7
95.5
92.3
7.9
101.4
102.2
86.0
4.6
3.1
5.9
3.6
6.6
6.5
5.5
5.5
6.3
7.7
6.9
10.8
5.1
4.2
9.7
7.3
6.7
6.4
5.5
12.4
14.8
10
13.3
3.8
21.6
4.3
corticosterone
5a-THB
cortexolone
THS
10
91.8
100
10
7.0
9.1
107.3
92.2
3.4
4.9
50
9.1
14.8
3.6
89.5
DOC
10
1.3
90.9
E2
10
5.1
9.3
10−20000
5−20000
94.5
10
11
12
13
14
15
16
17
18
19
20
21
E1
5
4.3
5.1
94.0
testosterone
5-androstenediol
20
13.6
4.9
19.3
4.1
20−10000
20−20000
500−20000
500−10000
500−10000
100−10000
50−10000
500−10000
100−10000
5−20000
100.6
98.2
10
97.7
a
2-OH E1
50
7.4
6.9
8.7
11.8
androsterone
17OH-Preg
DHEA
500
500
100
100
100
50
16.4
13.7
16.1
12.5
14.6
13.8
7.0
19.9
16.6
14.5
20.3
18.2
20.0
7.7
96.6
96.5
91.6
101.2
90.4
96.3
96.1
99.5
99.0
109.9
102.5
112.2
105.3
101.4
107.4
96.8
99.5
121.8
95.2
5α-DHT
5β-DHT
93.2
mesterolone
16α-OH E1
pregnenolone
97.0
5
112.3
104.8
500
11.8
16.9
500−10000
105.5
a
2
-OH E with very low recoveries could only be used for semi-quantitative analysis.
1
The residue was dissolved with 200 μL of 50% acetonitrile
pending LC-MS analysis.
at least one hydroxyl moiety in their chemical structure that was
expected to be esterified by DMAB (Figure 1B).
LC-MS Analysis. The MS instrument LTQ-Orbitrap XL
Thermo Fisher Scientific, Rockford, IL, USA) settings were as
The results show that all of the hydroxyl-containing steroid
hormones could be derivatized by DMAB in the presence of
DMAP and EDC in dichloromethane, with the exception of 17-
OHP. The failure of 17-OHP to react with DMAB was
attributed to the steric hindrance of the hydroxyl moiety. The
extracted ion chromatogram (EIC) of the 21 DMAB
derivatized steroid hormones by LC-MS is displayed in Figure
(
follows: capillary temperature 325 °C, source voltage 4.5 kV,
and capillary voltage 49 V for ESI+ analysis. The mass range
was set at 260−700 m/z. The resolution of the Orbitrap was set
at 15 000. Chromatographic separation was performed on a
BEH C18 column (10 cm × 2.1 mm, 1.7 μm; Waters, Milford,
MA, USA) using a Thermo Fisher Accela LC 1000 system with
an injection volume of 20 μL. The linear gradient program was
set as follows with mobile phase (A) 0.1% formic acid solution
and phase (B) acetonitrile: 0 min, 20% B; 8 min, 30% B; 33
min, 75% B; 35 min, 98% B; 41 min, 98% B; 41.01 min, 20% B;
3
. All of the derivatized steroid hormones were separated well
in the retention time range from 19 to 37 min. Isomers, for
example, 5α-DHT and 5β-DHT with a difference only in the
orientation of one hydrogen atom, could be separated on the
baseline. Because of the feasibility of the derivatization of
hydroxyl-containing steroid hormones by DMAB, we can
expect that obtaining a detailed pathway including more steroid
hormones can be accomplished by enlarging the steroid
hormone standards library.
4
8 min, 20% B. The flow rate was 0.25 mL/min.
RESULTS AND DISCUSSION
■
Coverage of DMAB-Derivatized Steroid Hormones.
Determination of Nonderivatized Steroid Hormones.
A solution containing a mixture of progesterone, 17-OHP, and
androstenedione (200 ng/mL) was tested to investigate the
percentage of nonderivatized steroid hormones that are
destroyed when subjected to the derivatization procedure.
Compared with intact standards solution (200 ng/mL), the
recoveries (n = 3) of progesterone, 17-OHP, and androstene-
dione were 95.6%, 102.1%, and 102.0%, respectively. This result
indicated that these nonderivatized steroid hormones were not
destroyed during the process of derivatization. In other words,
the initial levels of progesterone, 17-OHP, and androstenedione
could still be obtained in the derivatized urine in one LC-MS
run. Therefore, the method presented herein has the capability
of profiling all the steroid hormones listed in Figure 2.
Fortunately, the nonderivatized steroid hormones progester-
one, 17-OHP, and androstenedione had the lowest lower limit
The metabolic pathway shown in Figure 2 was modified
according to the pathway of steroid hormone biosynthesis in
the Kyoto Encyclopedia of Genes and Genomes database
(
KEGG, http://www.genome.jp/kegg-bin/show_
pathway?ko00140+C00280). It represents the core metabolic
pathway of steroid hormones biosynthesis containing most of
the well-known steroid hormones. Mesterolone, 5a-THB, and
THS are listed in Figure 2 with a dashed line frame because
they were not found in the given pathway from the KEGG
database. To establish an analytical method to cover as many
steroid hormones as possible, 24 steroid hormone standards
(
listed in Figure 2) covering the species of estrogen, androgen,
glucocorticoid, mineralocorticoid, and progestogen were used
to evaluate the performance of the derivatization and the LC-
MS analysis. All the selected hormone standards, with the
exception of the progesterone and androstenedione, contained
1
0248
dx.doi.org/10.1021/ac301984t | Anal. Chem. 2012, 84, 10245−10251

Analytical Chemistry
Article
Figure 4. Results of relative quantification based on D/H stable isotope DMAB labeling: (A) Extracted ion chromatogram of heavy and light labeled
estrone; (B) mass spectrum of the ion pair of heavy and light labeled estrone; (C) intensity ratios of 21 heavy/light labeled steroid hormones with
the amount of 1:2, 1:1, and 2:1 (the number of the Y-axis represents the steroid hormone in Table 1); (D) ion pairs of light/heavy labeled estrone in
1:10, 1:3, 1:1, 3:1, and 10:1 mixed solution showed a good linear regression.
of detections (LLODs, 20 pg/mL) when compared with the
other intact steroid hormones.
1000, 2000, 5000, 10 000, and 20 000 pg/mL were subjected to
SPE extraction, derivatization, and LC-MS analysis steps to
investigate the LLODs and linear ranges for the 21 derivatized
steroid hormones. The results (Table 1) show that the 21
derivatized steroid hormones were all detected at subng/mL
levels. The LLODs ranged from 5 to 500 pg/mL, equivalent to
Enhancement of Sensitivity. The detection sensitivities
for intact steroid hormones and DMAB-derivatized steroid
hormones by LC-MS are compared in Supplemental Table S1.
Because the baseline of the EIC was horizontal for the high-
resolution mass spectrometer, the signal-to-noise ratio (S/N)
cannot be calculated. Herein, the lowest concentration
corresponding to the signal that can be detected by the EIC
was defined as the LLOD. The LLODs of the 21 steroid
hormones without SPE extraction are summarized (Supple-
mental Table S1). The LLODs of the DMAB-derivatized
steroid hormones ranged from 5 to 500 pg/mL (equivalent to
0
.1 to 10 pg being injected onto the column.
The derivatized steroid hormones show good linear behavior
with concentration within a range of 2−4 orders of magnitude.
The regression coefficients ranged from 0.967 to 0.999 (Table
1
).
A 20 μL portion of a steroid hormone mixture solution (200
ng/mL) was spiked into 500 μL of urine and was then
subjected to the SPE extraction-derivatization-LC-MS analysis
procedure to evaluate the intrabatch and interbatch precision (n
0
.1−10 pg on column), while the LLODs of their
corresponding intact steroid hormones ranged from 50 pg/
mL to >200 ng/mL by LC-MS. The sensitivity, therefore, was
improved by 1- to 10 000-fold using the derivatization method.
For example, for the improvement in LLOD for 5-
=
6). The relative standard deviation (RSD) for the intrabatch
precision ranged from 1.3% to 16.4%, with an average RSD of
.1% for the 21 steroid hormones. The RSD for the interbatch
precision ranged from 3.6% to 21.6%, with an average RSD of
9
androstenediol, 5a-THB, and 16α-OH E , the improvement
1
folds of LLOD was observed to be more than 3−4 orders of
magnitude. There were two major aspects that contributed to
the improved sensitivity after the DMAB derivatization: (1) the
introduction of an easily protonated tertiary amine moiety by
derivatization with DMAB greatly increased the ionization
efficiency for steroid hormones and (2) the decrease in the
polarity and the elution with a higher percentage of organic
phase led to an increase in the nebulization efficiency.
1
2.1% for the 21 steroid hormones.
Steroid hormone mixture solutions of different volumes (10,
2
0, and 40 μL at 200 ng/mL) were spiked into 500 μL urine
samples and were then subjected to the SPE extraction-
derivatization-LC-MS analysis procedure to evaluate the
recovery. The recoveries of the 21 steroid hormones with
spiked standard levels of 4, 8, and 16 ng/mL in urine ranged
from 88.5% to 137.4%, from 88.8% to 109.5%, and from 86.0%
Analytical Characteristics of the Method. Steroid
hormone mixture solutions of 5, 10, 20, 50, 100, 200, 500,
to 112.2%, respectively. The steroid hormone 2-OH E was the
1
1
0249
dx.doi.org/10.1021/ac301984t | Anal. Chem. 2012, 84, 10245−10251

Analytical Chemistry
Article
only exception, with very low recoveries that could only be used
for semiquantitative analysis.
A 20 μL portion of the steroid hormone mixture solution
(200 ng/mL) was spiked into 500 μL of urine. After being
derivatized, the sample was analyzed at 0, 24, and 48 h to test
the stability at 4 °C within 2 days (n = 3). The RSD of the 21
derivatized steroid hormones were all <15%, with an average
RSD of 6.9%. This result indicated that the derivatized urine
sample was stable while waiting for the LC-MS run in the
autosampler within 48 h.
Stable Isotope-Based Quantification. After the SPE
extraction and lyophilization, 8, 16, and 32 ng/mL steroid
hormone mixtures were derivatized with d -DMAB. The
4
derivatized mixtures were then mixed with an equal volume
of a DMAB-derivatized 16 ng/mL steroid hormone mixture.
The mixed solution contained 1:2, 1:1, and 2:1 d -DMAB/
4
DMAB derivatized steroid hormones. These mixed solutions
were then analyzed to investigate the MS intensity ratios of ion
Figure 5. Scheme of the procedure for the absolute quantification of
free (unconjugated) urinary steroid hormones of male, female, and
pregnant female volunteers.
pairs of d -DMAB derivatized (heavy labeled) and DMAB
4
derivatized (light labeled) steroid hormones.
Here, we can look at estrone as an example. In Figure 4A, the
ion pair of the d -DMAB and the DMAB derivatized steroid
hormone was found in one LC-MS run. DMAB-estrone was
The analysis by LC-MS detected 17 derivatized free steroid
hormones and 3 nonderivatized steroid hormones (17-OHP,
androstenedione, and progesterone) in the nonhydrolyzed
urine samples (the EICs of the detected steroid hormones are
shown in Supplemental Figure S2). The levels of these steroids
were absolutely quantified (the results of the concentrations are
shown in Supplemental Table S2). As expected, the urine
samples from the males had higher levels of androgens, while
females excreted more estrogens than males. The pregnant
women had many significant changes in urinary steroid
hormones, offering a snapshot of the alteration in the steroid
hormone pathway during pregnancy (Supplemental Figure S3).
Pregnant women excreted much greater amounts of estrone,
4
determined to have an m/z = 418, while d -DMAB-estrone
4
showed an m/z = 422. The difference in m/z values indicates
the replacement of four hydrogen atoms by four deuterium
atoms (Figure 4B). The feasibility of stable isotope labeling-
based relative quantification was evaluated in two ways: (1) 1:2,
1
:1, and 2:1 steroid hormones were derivatized with d -DMAB
4
and DMAB (the results are shown in Figure 4C) and (2) the
d4-DMAB and DMAB derivatized solutions were mixed with
volume ratios of 1:10, 1:3, 1:1, 3:1, and 10:1 (the results are
shown in Figure 4D). After the SPE extraction, lyophilization,
derivatization, and mixing, the intensity ratios of ion pairs were
approximately 1 (ranging from 0.90 to 1.14) with an average
ratio of 1.06 for the 21 steroid hormones when the D/H = 1:1
estradiol, estriol, 16α-OH E , and progesterone. These steroid
1
hormones were increased more than 30-fold when compared to
nonpregnant women. In particular, the level of free urinary
estriol reached 187 ng/mL, increasing 510-fold compared with
nonpregnant women. This result is consistent with the
description that “during pregnancy, estriol constitutes 60−
70% of the total estrogens, increasing to 300−500-fold in
relation to non-pregnant women” in HMDB.
(
Figure 4C). Additionally, the quantitative ratios were
approximately 2 (except for the 3.31 observed for 2-OH E1,
which could only be used for semiquantitative analysis) and 0.5
when the D/H = 2:1 and 1:2, respectively. In Figure 4D, the
intensity ratios of the ion pairs in the 1:10, 1:3, 1:1, 3:1, and
1
0:1 mixed d -DMAB/DMAB derivatized solutions showed a
4
good linear regression. With use of estrone as an example, the
R value was 0.9998 (the linear regression of the other 20
2
CONCLUSIONS
■
steroid hormones is shown in Figure S1 of the Supporting
Information). The linear regressions indicated that good
relative quantification results can be obtained based on the
stable isotope-labeling strategy.
These results show that the heavy labeled urine sample
mixed with the light labeled steroid hormone standards of
known concentrations as stable isotope internal standards can
be used to provide accurate absolute measurements of steroid
hormones.
Quantifying the Steroid Hormones in Male, Female,
and Pregnant Volunteer Urine Samples. To test the new
method’s utility for the quantification of steroid hormones in
actual clinical samples, urine samples from five males, five
nonpregnant females, and two pregnant females were analyzed.
As shown in Figure 5, the pooled urine from male, female, and
pregnant women was heavy labeled (n = 3), while 4 ng/mL
steroid hormone standards were light labeled. The heavy
labeled urine and light labeled standards were mixed with a
volume ratio of 3:1 and then analyzed by LC-MS.
There is mounting evidence showing that steroid hormones
play important roles in the development and progression of
various diseases. Unfortunately, few approaches offer a
comprehensive and highly sensitive means of detecting steroid
hormones. In this work, we developed a stable isotope labeling-
based quantitative profiling technique for targeted analysis of
steroid hormones by LC-MS. The hormones studied include
hydroxyl-containing androgens, estrogens, corticoids, and
progestogens. Deuterium-4-(dimethylamino)-benzoic acid was
synthetized as a novel derivatization reagent for labeling steroid
hormones to improve the sensitivity 1- to >10 000-fold with
LLODs ranging from 5 to 500 pg/mL. The method showed
good intrabatch precision (average RSD = 9.1%), interbatch
precision (average RSD = 12.1%), and stability within 48 h
(average RSD = 6.9%), and 2−4 orders of magnitude of linear
range. Furthermore, the accurate relative and absolute
quantification can be achieved using a stable isotope-labeling
strategy. We also noted that a small number of nonderivatized
steroid hormones were not destroyed during the derivatization
1
0250
dx.doi.org/10.1021/ac301984t | Anal. Chem. 2012, 84, 10245−10251

Analytical Chemistry
Article
process, and that their quantitative information can be obtained
simultaneously in one LC-MS run. Therefore, our approach
using a stable isotope-labeling method combined with SPE and
LC-MS was able to obtain quantitative profiling of the steroid
hormone metabolome with subng/mL LLODs. Furthermore,
the method was successfully applied to analyze the free urinary
steroid hormones of males, females, and pregnant women. Our
future work will focus on building a larger library of derivatized
steroid hormones and applying this method to take a snapshot
of the steroid hormone pathway in hormone-related disease
studies, such as ovarian cancer, prostate cancer, hepatocellular
carcinoma, and many others.
(15) Soldin, S. J.; Soldin, O. P. Clin. Chem. 2009, 55, 1061−1066.
(
̈ ̈
16) Jantti, S. E.; Tammimaki, A.; Raattamaa, H.; Piepponen, P.;
Kostiainen, R.; Ketola, R. A. Anal. Chem. 2010, 82, 3168−3175.
17) Xu, X.; Veenstra, T. D.; Fox, S. D.; Roman, J. M.; Issaq, H. J.;
Falk, R.; Saavedra, J. E.; Keefer, L. K.; Ziegler, R. G. Anal. Chem. 2005,
7, 6646−6654.
18) Xu, X.; Roman, J. M.; Issaq, H. J.; Keefer, L. K.; Veenstra, T. D.;
Ziegler, R. G. Anal. Chem. 2007, 79, 7813−7821.
19) Huang, J.; Sun, J.; Chen, Y.; Song, Y.; Dong, L.; Zhan, Q.;
(
7
(
(
Zhang, R.; Abliz, Z. Anal. Chim. Acta 2012, 711, 60−68.
(20) Guo, T.; Taylor, R. L.; Singh, R. J.; Soldin, S. J. Clin. Chim. Acta
2006, 372, 76−82.
(
21) Flores-Valverde, A. M.; Hill, E. M. Anal. Chem. 2008, 80, 8771−
8
779.
(
22) Xu, F.; Zou, L.; Liu, Y.; Zhang, Z.; Ong, C. N. Mass Spectrom.
ASSOCIATED CONTENT
Supporting Information
This material is available free of charge via the Internet at
http://pubs.acs.org.
■
Rev. 2011, 30, 1143−1172.
*
S
(
23) Tang, Z.; Guengerich, F. P. Anal. Chem. 2010, 82, 7706−7712.
(24) Jiang, X.; Ory, D. S.; Han, X. Rapid Commun. Mass Spectrom.
2007, 21, 141−152.
(
25) Xu, L.; Spink, D. J. Chromatogr. B 2007, 855, 159−165.
AUTHOR INFORMATION
Corresponding Author
E-mail: xugw@dicp.ac.cn. Fax: +86-411-84379559.
Notes
(26) Nishio, T.; Higashi, T.; Funaishi, A.; Tanaka, J.; Shimada, K. J.
■
*
Pharmaceut. Biomed. Anal. 2007, 44, 786−795.
(
(
(
27) Guo, K.; Ji, C.; Li, L. Anal. Chem. 2007, 79, 8631−8638.
28) Guo, K.; Li, L. Anal. Chem. 2009, 81, 3919−3932.
29) Guo, K.; Li, L. Anal. Chem. 2010, 82, 8789−8793.
The authors declare no competing financial interest.
(30) Guo, K.; Bamforth, F.; Li, L. J. Am. Soc. Mass Spectrom. 2011, 22,
3
39−347.
ACKNOWLEDGMENTS
(31) Shortreed, M. R.; Lamos, S. M.; Frey, B. L.; Phillips, M. F.;
Patel, M.; Belshaw, P. J.; Smith, L. M. Anal. Chem. 2006, 78, 6398−
6
(
(
■
The study has been supported by the National Basic Research
Program of China (No. 2012CB518303) from the State
Ministry of Science & Technology of China, State Key Science
403.
32) Huang, X.; Regnier, F. Anal. Chem. 2008, 80, 107−114.
33) Boersema, P. J.; Raijmakers, R.; Lemeer, S.; Mohammed, S.;
&
Technology Project for Infectious Diseases
Heck, A. J. R. Nat. Protoc. 2009, 4, 484−494.
(
2817−2825.
(
(
(
2012ZX10002011, 2012ZX10002009), the Key Foundation
No. 20835006), and the Creative Research Group Project
No. 21021004) from the National Natural Science Foundation
of China. We also thank Mingqian Tan, Ying Duan, and Jing
Wu for help with organic synthesis.
34) Xiang, F.; Ye, H.; Chen, R.; Fu, Q.; Li, L. Anal. Chem. 2010, 82,
(35) Wiese, S.; Reidegeld, K. A.; Meyer, H. E.; Warscheid, B.
Proteomics 2007, 7, 340−350.
(
36) Arlt, W.; Biehl, M.; Taylor, A. E.; Hahner, S.; Libe, R.; Hughes,
B. A.; Schneider, P.; Smith, D. J.; Stiekema, H.; Krone, N.; Porfiri, E.;
Opocher, G.; Bertherat, J.; Mantero, F.; Allolio, B.; Terzolo, M.;
Nightingale, P.; Shackleton, C. H. L.; Bertagna, X.; Fassnacht, M.;
Stewart, P. M. J. Clin. Endocr. Metab. 2011, 96, 3775−3784.
REFERENCES
■
(
1) Hulka, B. S.; Liu, E. T.; Lininger, R. A. Cancer 1994, 74, 1111−
124.
1
(
2) Clemons, M.; Goss, P. N. Engl. J. Med. 2001, 344, 276−285.
(3) Blair, I. A. Steroids 2010, 75, 297−306.
(
4) Gann, P. H.; Hennekens, C. H.; Ma, J.; Longcope, C.; Stampfer,
M. J. J. Natl. Cancer Inst. 1996, 88, 1118−1126.
5) Hoffman, M. A.; DeWolf, W. C.; Morgentaler, A. J. Urol. 2000,
63, 824−827.
6) Attard, G.; Reid, A. H. M.; Olmos, D.; de Bono, J. S. Cancer Res.
009, 69, 4937−4940.
7) Cutolo, M.; Sulli, A.; Villaggio, B.; Seriolo, B.; Accardo, S. Ann.
Rheum. Dis. 1998, 57, 573−577.
8) Karlson, E. W.; Chibnik, L. B.; McGrath, M.; Chang, S.-C.;
(
1
(
2
(
(
Keenan, B. T.; Costenbader, K. H.; Fraser, P. A.; Tworoger, S.;
Hankinson, S. E.; Lee, I. M.; Buring, J.; De Vivo, I. Arthrit. Res. Ther.
2
009, 11, R97.
9) Yamashita, K.; Miyashiro, Y.; Maekubo, H.; Okuyama, M.;
Honma, S.; Takahashi, M.; Numazawa, M. Steroids 2009, 74, 920−926.
10) Janzen, N.; Sander, S.; Terhardt, M.; Steuerwald, U.; Peter, M.;
Das, A. M.; Sander, J. Steroids 2011, 76, 1437−1442.
11) Ceglarek, U.; Werner, M.; Kortz, L.; Korner, A.; Kiess, W.;
Thiery, J.; Kratzsch, J. J. Steroid Biochem. Mol. Biol. 2010, 121, 505−
12.
12) Holst, J. P.; Soldin, O. P.; Guo, T.; Soldin, S. J. Clin. Lab. Med.
004, 24, 105−118.
13) Koal, T.; Schmiederer, D.; Pham-Tuan, H.; Ro
M. J. Steroid Biochem. Mol. Biol. 2012, 129, 129−138.
14) Kushnir, M. M.; Rockwood, A. L.; Roberts, W. L.; Yue, B.;
Bergquist, J.; Meikle, A. W. Clin. Biochem. 2011, 44, 77−88.
(
(
(
̈
5
(
2
(
̈
hring, C.; Rauh,
(
1
0251
dx.doi.org/10.1021/ac301984t | Anal. Chem. 2012, 84, 10245−10251