Improvement of estradiol esters monitoring in bovine hair by dansylation and liquid chromatography/tandem mass spectrometry analysis in multiple reaction monitoring and precursor ion scan modes

Article abstract of DOI:10.1002/rcm.6160

Rationale: The control of forbidden anabolic practices in cattle in the European Union has become challenging since endogenous compounds such as estradiol derivatives can potentially be used as growth promoters. Due to the great difficulty in establishing a reference threshold value for endogenous steroids, the direct detection of steroid esters in hair is an efficient strategy for the detection of 'natural' steroid abuse in cattle. Methods: The present study aimed to develop and validate according to the current European standards a specific liquid chromatography/tandem mass spectrometry (LC/MS/MS) analytical strategy to monitor estrogen esters in bovine hair. The analysis was performed by positive ion electrospray ionisation (ESI+) after dansylation. Two acquisition modes were then assessed: single reaction monitoring and precursor ion scanning. Results: The results showed that the introduction of a dansylation step strongly improves the sensitivity of the detection of estradiol-17-esters by LC/(ESI+)-MS/MS. The CCα values are in the range 1-10 ng g^-1 after optimisation, except for estradiol decanoate for which the derivatisation is not efficient. In addition, this LC/MS/MS approach makes it possible to carry out a precursor ion scan to screen for the presence of these estradiol 17-esters in hair samples. Conclusions: Based on the specific product ions, i.e. m/z 255 in native conditions or m/z 171 after dansylation, this strategy has the advantage of detecting any (un)known estradiol ester and of giving access to the [M + H]⁺ ion of the suspected ester through only a single analysis.

Full text of DOI:10.1002/rcm.6160

Research Article
Received: 19 October 2011
Revised: 10 January 2012
Accepted: 12 January 2012
Published online in Wiley Online Library
Rapid Commun. Mass Spectrom. 2012, 26, 819–827
(wileyonlinelibrary.com) DOI: 10.1002/rcm.6160
Improvement of estradiol esters monitoring in bovine hair by
dansylation and liquid chromatography/tandem mass
spectrometry analysis in multiple reaction monitoring and
precursor ion scan modes
*
E. Bichon , A. Béasse, S. Prevost, S. Christien, F. Courant, F. Monteau and B. Le Bizec
LUNAM Université, Oniris, Laboratoire d’Etude des Résidus et Contaminants dans les Aliments (LABERCA), F-44307
Nantes, France
RATIONALE: The control of forbidden anabolic practices in cattle in the European Union has become challenging since
endogenous compounds such as estradiol derivatives can potentially be used as growth promoters. Due to the great
difficulty in establishing a reference threshold value for endogenous steroids, the direct detection of steroid esters in hair
is an efficient strategy for the detection of ’natural’ steroid abuse in cattle.
METHODS: The present study aimed to develop and validate according to the current European standards a specific
liquid chromatography/tandem mass spectrometry (LC/MS/MS) analytical strategy to monitor estrogen esters in
bovine hair. The analysis was performed by positive ion electrospray ionisation (ESI+) after dansylation. Two acquisition
modes were then assessed: single reaction monitoring and precursor ion scanning.
RESULTS: The results showed that the introduction of a dansylation step strongly improves the sensitivity of the
detection of estradiol-17-esters by LC/(ESI+)-MS/MS. The CCa values are in the range 1–10 ng g–1 after optimisation,
except for estradiol decanoate for which the derivatisation is not efficient. In addition, this LC/MS/MS approach makes
it possible to carry out a precursor ion scan to screen for the presence of these estradiol 17-esters in hair samples.
CONCLUSIONS: Based on the specific product ions, i.e. m/z 255 in native conditions or m/z 171 after dansylation, this
strategy has the advantage of detecting any (un)known estradiol ester and of giving access to the [M + H]+ ion of the
suspected ester through only a single analysis. Copyright © 2012 John Wiley & Sons, Ltd.
The use of growth promoters in cattle fattening is prohibited
in the European Union according to Directive 96/22/EC.[1]
Since the 1990s, the concentration levels of xenobiotics have
been decreasing and it seems that steroid misuse is shifting
toward natural hormones.[2] Currently, one of the most
difficult challenges is probably the differentiation of mimetic
compounds from the natural ones, where the suspicious
signal is hidden by that arising from the endogenous
production of the animal. This is mainly the case for testoster-
one and estradiol, but it also affects nandrolone and boldenone.
In the last decade, several analytical solutions have been
developed to improve the monitoring of these substances,
based on an easily available and non-invasive biological
matrix, e.g. urine. These analytical strategies are usually
based on targeted gas chromatography/tandem mass
spectrometry (GC/MSn) or liquid chromatography/tandem
mass spectrometry (LC/MSn) measurement methods.[3–11]
However, finding evidence of the administration of a natural
steroid still represents a very difficult challenge, mainly due
to the high variability of the endogenous metabolites in urine
between animals.[12] As a result, non-ambiguous analytical
strategies, also called confirmatory methods, were implemen-
ted in control laboratories. As an example, the confirmation of
estradiol and testosterone abuse in cattle is performed by
measuring the isotopic deviation (d13CV-PDB
) of these
compounds by gas chromatography coupled to combustion
isotope ratio mass spectrometry (GC/C/IRMS). In this case,
elucidation of the steroid origin provides unambiguous
confirmation of illegal administration.[13–15] In parallel, the
identification of some markers of administration, never found
in non-treated animals, is a powerful alternative method of
confirming the misuse of boldenone[16] or nandrolone in
cattle.[17] However, the corresponding metabolism has to be
completely elucidated before such efficient targeted analytical
strategies can be employed. In addition, the use of urine as
the biological sample has the drawback that xenobiotics are
rapidly eliminated through the excretion of metabolites in
urine. Detection sensitivity remains the major issue in
these different approaches, reducing the time window of
detectability to rarely over one week after administration.
Since the 1990s, several approaches to detect natural steroid
hormones administration have been described on sweat
extracts, and more widely on hair in drug testing,[18–20] doping
control,[21,22] and livestock production monitoring.[6,23] The hair
sample presents some advantages, in particular in being a
non-invasive sample easy to obtain, but also in allowing a
* Correspondence to: E. Bichon, ONIRIS, École nationale
vétérinaire, agroalimentaire et de l’alimentation Nantes-
Atlantique, Laboratoire d’Etude des Résidus et Contaminants
dans les Aliments (LABERCA), Atlanpole – La Chantrerie, BP
40706, F-44307 Nantes, France.
E-mail: laberca@oniris-nantes.fr
Rapid Commun. Mass Spectrom. 2012, 26, 819–827
Copyright © 2012 John Wiley & Sons, Ltd.

E. Bichon et al.
longer detection window after administration than urine.
Furthermore, in hair, the administered substance is still present
under its administered form.
estradiol, estrone and estriol.[33] The dansylation reaction
has already been successfully applied to ethinylestradiol
monitoring in hair[34] and in cerebrospinal samples.[35] The
use of atmospheric pressure photoionisation was also tested[36]
and the results were equivalent to those obtained with
ESI + with a sensitivity of detection close to 1 ng.mL–1 (lowest
calibration point on plasma samples).
Based on the experience acquired for steroid analysis, the aim
of the present study was to enhance steroid ester monitoring in
hair samples using LC/MS/MS. To this end, the derivatisation
reaction was adapted for the first time to estradiol-17-esters, the
acquisition parameters were defined and the entire analytical
methodology was validated according to 2002/657/EC analy-
tical criteria. The time window of detectability after the admin-
istration of estradiol benzoate was determined and compared
with published results. In addition, the feasibility of a wider
screening tool for the monitoring of estradiol esters in hair
was studied based on the specific product ions formed in the
MS/MS study of the dansylated derivatives.
In view of these major advantages, several developments
have been carried out to monitor steroid esters in hair either
by gas chromatography/tandem mass spectrometry (GC/
MS/MS)[24–26] or by liquid chromatography/tandem mass
spectrometry (LC/MS/MS).[27] A method inter-comparison
demonstrated reasonable equivalence between the LC/MS/
MS and GC/MS/MS methods for the detection of estradiol
benzoate and nandrolone decanoate.[28] The preparation
procedures include a sample size of 100–300 mg of ground hair,
with various washing treatments,[29] and an extraction with
MeOH, sometimes coupled to a reducing agent such as
tris(2-carboxyethyl)phosphine (TCEP). Extracts can then be
directly injected[24] or purified on an solid-phase extraction
(SPE) C18 cartridge[27,28] or on successive SPE NH2 and C18
cartridges[25] before analysis. The detection of estradiol
benzoate, for example, was still possible 15 days after
administration, at which time the concentrations had decreased
to the decision limit (CCa) values of previously described
analytical methods, i.e. 10 ng.g–1.[27,28]
EXPERIMENTAL
Nevertheless, longer detectability of steroid esters might
be expected in a matrix such as hair, and more modern
instruments can deliver higher sensitivity. LC/MS/MS
analysis does not seem to be the most suitable and
sensitive method for steroid determination, and GC/MS
is often preferred. Indeed, the ionisation efficiencies of
such compounds in electrospray ionisation (ESI) are
actually very low because these substances are not highly
polar or even non-polar. Therefore, derivatisation reactions
have often been recommended, particularly for estrogen
compounds[30–32] with dansylation of ethinylestradiol, or
with 2-fluoro-1-methylpyridinium-p-toluenesulfonate (FMPTS)
and pentafluorobenzyl bromide (PFBBr) derivatives for
Chemicals
The reference steroids (listed in Table 1) were from Steraloids
Inc. Ltd. (London, UK), Sigma–Aldrich (St. Quentin Fallavier,
France) and RIKILT (Wageningen, The Netherlands). Most of
the reagents and solvents were of analytical grade quality
and provided by VWR International (Pessac, France) and
Carlo-Erba Reagents (Rodano, Italy). The SPE columns were
from Carlo-Erba Reagents (silica: 1 g, aminopropyl: 0.5 g).
The derivatisation reagent (dimethylamino)naphthalene-
1-sulfonyl chloride (dansyl chloride) was purchased from
Fisher Bioblock Scientific (Illkirch, France).
Table 1. LC/MS/MS parameters used for detecting and identifying each target compound
Transition
CV
(V)
CE
(eV)
Transition
CV
(V)
CE
(eV)
(m/z)
(m/z)
norethindrone acetate
(IS)
341.2 > 281.2
30
15
norgestrel (ES)
313.2 > 245.2
509.2 > 170.8
530.2 > 170.8
395.2 > 105
30
45
45
30
30
70
25
70
30
70
25
70
30
50
30
70
30
15
40
40
25
20
60
8
60
20
60
15
60
20
60
20
60
20
estradiol-d3 (ES)*
ethinylestradiol (IS)*
stanolone benzoate (IS)
estradiol-17-benzoate
estradiol-17-benzoate*
estradiol-17-cypionate
estradiol-17-cypionate*
estradiol-17-decanoate
estradiol-17-decanoate*
estradiol-17-enanthate
estradiol-17-enanthate*
estradiol-17-propionate
estradiol-17-propionate*
estradiol-17-valerate
estradiol-17-valerate*
estradiol-3-benzoate
377.2 > 105.1
506.1 >155.8
397.2 > 255.2
630.3 > 170.8
255.2 > 133.1
660.3 > 170.8
385.2 > 255.1
618.2 > 170.8
255.2 > 133.1
562.2 > 155.8
255.2 > 133.1
590.2 > 170.8
377.2 > 105.1
377.2 > 135.1
506.1 >170.8
255.2 > 159.1
630.3 > 155.8
255.2 > 159.1
660.3 > 155.8
255.2 > 159.1
618.2 > 155.8
255.2 > 159.1
562.2 > 170.8
255.2 > 159.1
590.2 > 155.8
377.2 > 135.1
30
70
30
70
30
70
30
70
30
50
30
70
30
15
40
20
40
20
40
20
40
20
40
20
40
15
*Dansylated compound for LC/MS/MS analysis
CV: cone voltage; CE collision energy
wileyonlinelibrary.com/journal/rcm
Copyright © 2012 John Wiley & Sons, Ltd.
Rapid Commun. Mass Spectrom. 2012, 26, 819–827

Monitoring of estradiol esters in bovine hair
Sample preparation
determined with the calibration curve from the average of
the noise intensity for the 20 samples at the retention time
of the analyte + 2.33 times the associated standard deviation.
The detection capability (CCb) was calculated from the
intensity value of CCa + 1.64 times the standard deviation of
the within-laboratory reproducibility on each level of the
calibration curve.
A 100 mg hair sample was ground for about 15 min,
sonicated for 1 h with 5 mL of methanol and incubated
overnight at 50 ꢀC. After centrifugation, the methanolic
extract was removed, evaporated and prepared for successive
purifications on NH2 and SiOH cartridges.[25] The purified
extract was then reconstituted in 50 mL of a derivatisation
reagent made up of 10 mg dansyl chloride diluted in 10 mL
of acetonitrile (ACN)/Na2CO3 (10 mM) (50:50, v/v) and held
at 60 ꢀC for 40 min. The dansylated steroids were extracted in
hexane after liquid/liquid extraction with 1.5 mL hexane
against 1.5 mL water. The organic solvent was then
evaporated and the final extract was reconstituted in 50 mL
of an ACN/H2O mixture (50:50, v/v).
RESULTS AND DISCUSSION
Optimisation of ESI + MS acquisition parameters
After a brief study in negative ionisation mode, theoreti-
cally more appropriated to the ionisation of estrogens,
due to their phenol moiety, the positive mode was selected
for this study. Under these conditions, the protonated
molecule, the [M + H]+ ion, was always detected for all
the compounds considered, at a better sensitivity than
was obtained for the compounds in negative ion mode
(data not shown).
First, an acquisition in SCAN mode was carried out
after optimising the capillary and cone voltages to reach
the best intensity for the [M + H]+ ion. Then, a product
ion scan of the [M + H]+ ion revealed the major product
ions associated to a specific collision energy applied in
the T-wave collision cell. After a manual ramp of collision
energies, two collision energy values were selected per
compound, at which product ion scans were recorded.
The product ion mass spectra are presented in Fig. 1, for
each of the seven studied estradiol esters. Finally, an
MRM method was created with at least two diagnostic
transitions per compound (cf. Table 1).
LC/MS/MS system
An Acquity UPLCW system (Waters, Milford, MA, USA) was
used to perform reversed-phase liquid chromatography on a
C18 and a phenyl column (Acquity BEH C18 and Acquity
BEH Phenyl, 1.7 mm, 2.1 Â100 mm) at 50 ꢀC. The elution
solvents were 0.1% formic acid in H2O (A) and 0.1% formic
acid in ACN (B). The mobile phase composition was 30:70
A/B (v/v): between 0 and 2 min, 0:100 from 8 to 9 min, and
back to the initial conditions until 12 min have elapsed. The
flow rate was 600 mL min–1. The injection volume was 5 mL.
Data were acquired on a triple quadrupole XEVO-TQMS
mass spectrometer (Waters) operating in positive electrospray
ionisation (ESI+) mode. Nitrogen was used as the desolvation
gas at a flow rate of above 900 L.h–1. The source and the
desolvation temperatures were set at 150 and 450 ꢀC,
respectively. The capillary potential was set at 3.5 kV. Argon
was used as the collision gas at a flow rate of 0.15 mL min–1
for Multiple Reaction Monitoring (MRM) and Precursor ion
Scan (PS) experiments. For MRM acquisition, the collision
energies and cone voltages were optimised for each
compound (Table 1). For the PS mode, the cone voltage,
collision energy, scan range and product ions were
respectively 30 V, 15 eV, from m/z 254 to 500 and m/z 255 for
underivatised estradiol esters, and 70 V, 50 eV, from m/z 500
to 670 and m/z 171 for dansylated estradiol esters. For these
optimisations, each standard was directly introduced at a
concentration of 10 ng/mL in H2O/ACN/formic acid
(50:50:0.1, v/v/v) with a flow rate of 10 mL/min. At this
flow rate, the desolvation temperature and gas flow were,
respectively, decreased to 300 ꢀC and 600 L h–1. TargetLynx
(Waters) software was used for the integration and
assignment of all chromatographic peaks acquired in MRM
mode. All the analytes were identified according to the
Decision 2002/657/EC,[37] on the basis of retention times
and transition ratios.
PS acquisition mode strategy to screen for estradiol esters
For all the estrogen esters considered, the m/z 255 product
ion is present from the lowest collision energies (from 8 to
25 eV) (cf. Fig. 1). This ion is specific to the steroid
considered, i.e. estradiol here, and represents [Mfree steroid
+
H – H2O]+. Nevertheless, m/z 255 is not the base peak of
the estradiol benzoate product ion spectrum, where the
main fragmentation pathway leads to the benzoyl ion at
m/z 105. As the m/z 255 ion is characteristic of the
estradiol-ester fragmentation, a screening strategy based
on this specific ion can easily be set up to monitor
estradiol esters in a very specific manner. Therefore, the
precursor ion scan of m/z 255 was tested. Spiked (between
10 and 25 ng.g–1 of all the estrogen esters considered) and
blank hair samples were then studied by LC/MS/MS
(data not shown). The performance of this method was
found to be limited because of the wide variations in the
intensity of the m/z 255 ion for the seven compounds
of interest and because of the presence of a range of
interferences which complicated the interpretation. Very
poor signal-to-noise ratios were obtained, probably due
to a huge matrix effect on m/z 255 associated with an
acquisition mode which was not sufficiently sensitive.
Nevertheless, this strategy is theoretically very promising
and it could be implemented with some modifications
(see } Precursor ion scan mode for a screening method of
(un)known steroid esters).
Validation protocol
The validation procedure was based on Decision 2002/657/
EC.[37] We analysed 20 blank hair samples (free of estrogen
esters) collected from porcine and bovine species, exhibiting
various colors, from dark to white. One sample pool of all
the samples was also used to build the calibration curve on
sample extracts for each batch extracted (6 batches in the
global validation extracted on 6 different days and spiked
from 2.5 to 200 ng/g). The decision limit (CCa) was
Rapid Commun. Mass Spectrom. 2012, 26, 819–827
Copyright © 2012 John Wiley & Sons, Ltd.
wileyonlinelibrary.com/journal/rcm

E. Bichon et al.
377
Estradiol 3-benzoate
Product ion Scan
m/z 377 coll 10 eV
135
of Li et al.[31] The comparison of all the results was based on
the dansylation recoveries to determine the derivatisation
efficiency. In LC/(ESI+)-MS/MS, the [dansylated steroid
+ H]+ ion was always formed in the interface and yielded,
after collision, the specific dansyl product ions m/z 171 and
156 (loss of a methyl group from the dansyl moiety)[30–32]
(cf. Fig. 2). Hence, two transitions were then selected for each
of the dansylated estrogen ester studied, from the [M + H]+
ion to the two product ions (cf. Table 1). Furthermore, the
dansyl-estradiol esters were stored for several days in the
fridge and reinjected without loss of sensitivity, thus demon-
strating the stability of these derivatised compounds. (Fig. 3)
The stability of the reagent was investigated because a
variation in color between samples and series (more or
less yellow in color) was sometimes observed. Several
derivatisation mixtures (dansyl chloride/sodium carbonate/
acetonitrile) were prepared to test the duration of the
reactivity of the reagent, over a week (D7), a full day (D2)
and only several minutes before use (D0). The derivatisation
was performed for 10 min at 60 ꢀC on all the standards of
interest, each derivative solution being tested in triplicate.
The dansylation recoveries were calculated in reporting each
relative intensity (analyte to internal standard) to the highest
one obtained during these assays. These recoveries (Fig. 4)
appeared to be linked to the day when the derivatisation
reagent was prepared, with an efficient recovery obtained
on D0. This led to the conclusion that the derivatisation
reagent has to be recently prepared to obtain the best
recoveries.
100
105
359
0
273
Estradiol decanoate
100
Product ion Scan
m/z 427 coll 25 eV
427
345
255
255
0
Estradiol valerate
Product ion Scan
357
100
m/z 357 coll 25 eV
159
339
0
329
Estradiol propionate
100
Product ion Scan
m/z 329 coll 25 eV
255
255
135
247
311
0
Estradiol enanthate
100
385
Product ion Scan
m/z 385 coll 8 eV
0
255
Estradiol cypionate
100
Product ion Scan
m/z 397 coll 20 eV
397
0
105
Estradiol 17-benzoate
Product ion Scan
100
Dansylation does not, however, seem to be suitable for
estradiol-17-decanoate, which gave recoveries below 50%
(not reported on Fig. 4). Fortunately, the estradiol-3-benzoate
derivative is not affected and degraded by the reaction, even
when the reagent is most active at D0.
m/z 377 coll 20 eV
377
135
359
255
0
m/z
100 125 150 175 200 225 250 275 300 325 350 375 400 425
Figure 1. Product ion spectra obtained after fragmentation of
each [M + H]+ ion of estrogen esters in (ESI+)-MS/MS.
The reaction time was also optimised with the testing, in
triplicate, of six different reaction times. The dansylated
solution was added to the estrogen ester mixture and held
Derivatisation as a strategy to enhance performance in the
detection of estradiol-17-esters
ꢀ
at 60 C for 0 min, 10 min, 15 min, 20 min, 30 min, 45 min
and 60 min. The dansylation recoveries are presented in Fig. 5,
calculated in the same way as for the recoveries linked to
the reagent preparation. As a first observation, the contact
time seems relatively well correlated with the recovery of
dansylation. Moreover, for estradiol decanoate, the maximum
contact time is required to obtain sufficient recovery yield.
Since a dansyl chloride reagent has been widely used to
enhance the detection of estrogens in several matrices,[33–36]
it was decided to study its ability to improve the detection
of estrogen esters in the present work. The first assay was
carried out on standard solutions according to the protocol
Figure 2. Scheme of dansyl choride derivatisation reaction on estradiol esters
(R is an alkyl group) and illustration of the MS fragmentation of the protonated
dansyl-estradiol ester with the resulting two product ions.
wileyonlinelibrary.com/journal/rcm
Copyright © 2012 John Wiley & Sons, Ltd.
Rapid Commun. Mass Spectrom. 2012, 26, 819–827

Monitoring of estradiol esters in bovine hair
Figure 3. Typical diagnostic chromatograms obtained for estradiol esters in LC/MS/MS in a spiked hair sample
(top) and a blank hair sample (bottom) in their native form (without derivatisation, left) and after dansylation (right).
Figure 4. Recoveries yield obtained for the dansylation derivatisation reaction (n = 3) for five
estradiol-17-esters derivatised with a reagent prepared on the day of derivatisation (0), 2 days before
(2) and 7 days before (7). Estradiol-3-benzoate was also introduced into the mixture to control its
stability (the average is represented by a square and the min-max difference by a vertical line).
ꢀ
Thus, the final protocol was to hold the sample at 60 C for
40 min, to guarantee a sufficient recovery of derivatisation
(above 30%) for all the estrogen esters within an acceptable
time compatible with routine analysis requirements.
Finally, to control the derivatisation step, several potential
internal standards were tested, among which were D3-estradiol
and ethinylestradiol. The best choice would have been an
isotopic dilution of estrogen esters, but this type of compound
Rapid Commun. Mass Spectrom. 2012, 26, 819–827
Copyright © 2012 John Wiley & Sons, Ltd.
wileyonlinelibrary.com/journal/rcm

E. Bichon et al.
Figure 5. Recoveries of dansylation for six estradiol-17-esters derivatised at 60 ꢀC with
a contact time from 0 to 60 min (three replicates per point). Estradiol-3-benzoate was
introduced as non-reactive analyte.
is not commercially available. The results showed that only
ethinylestradiol presented the same behavior as the steroid
esters, with similar recoveries after the extract purification
(not shown). Thus, ethinylestradiol was selected as an
additional internal standard, able to follow all the analytical
process, and D3-estradiol as an external standard to monitor
only the derivatisation step.
Subsequently, derivatisation was tested on spiked samples
(cf. Fig. 3). First, the internal and external standards,
respectively norethindrone acetate and norgestrel, were not
affected by the dansylation and kept a very good peak shape
with the same sensitivity (S/N) as without derivatisation.
Secondly, the specificity was strongly improved for the
majority of the steroid esters after dansylation, allowing easy
detection from a sample concentration of 2.5 ng g–1, with a
good S/N, as it is shown for estradiol propionate, valerate,
enanthate and cypionate (cf. Fig. 3). Rambaud et al.[25]
obtained estradiol-benzoate concentrations below 20 ng g–1
10 days after administration. Thus, improved sensitivity of
detection of such a compound could be obtained by this pro-
tocol. Even if the selected product ions in the MRM mode are
specific to the dansyl moiety, i.e. m/z 171 and 156, the chroma-
togram for the dansylated fraction appeared the most
relevant in term of sensitivity and specificity. However,
estradiol-decanoate remains the exception, showing its best
sensitivity in LC/MS/MS analysis without derivatisation.
Because of this low dansylation yield, estradiol-decanoate
might be better monitored with the non-phenol-steroid
esters.
Validation results
The CCa and CCb values are presented in Table 2. The
CCa values appear more homogenous for the esters of
interest (from 1 to 10 ng g–1) for the LC/MS/MS method if
estradiol-17-decanoate is not taken into account. Based on
the results shown in Fig. 3 for this latter compound as well
as the relatively poor results obtained for the derivatisation
recoveries, an analysis of this compound under its native form
is recommended. The linearity results were obtained from cali-
bration curves built in the range 2.5–200 ng g–1 with at least 10
spiked samples.
Table 2. Decision limit (CCa) and detection capability (CCb) (in ng g–1) of estradiol esters in hair samples calculated from 20
different blank samples and from the standard deviation of the within-laboratory reproducibility on spike samples. Slope,
intercept and determination coefficient for each calibration curve built from at least 10 spike samples (2.5–200 ng g–1)
Transition 1
Transition 2
CCa
(ppb)
CCb
(ppb)
CCa
(ppb)
CCb
(ppb)
a
b
R²
a
b
R²
estradiol-17-benzoate*
estradiol-17-cypionate*
estradiol-17-decanoate*
estradiol-17-enanthate*
estradiol-17-propionate*
estradiol-17-valerate*
estradiol-3-benzoate
2.58
6
9
51
189
11
5
0.001
0.001
0.006 0.766
0.000 0.995
3
7
5
54
287
12
5
0.001
0.001
0.007
0.008
0.095
0.007
0.002
0.002 0.933
0.000 0.997
À0.039 0.930
À0.008 0.957
À0.028 0.993
0.000 0.997
0.001 0.993
139
1
0.013 À0.086 0.936
0.014 À0.011 0.962
0.148 À0.039 0.992
243
1
1
1
1
4
0.001
0.000 0.995
1
3
1
1
0.024 À0.008 0.999
7
19
*Dansylated compound for LC/MS/MS analysis
wileyonlinelibrary.com/journal/rcm
Copyright © 2012 John Wiley & Sons, Ltd.
Rapid Commun. Mass Spectrom. 2012, 26, 819–827

Monitoring of estradiol esters in bovine hair
MRM acquisition mode
PS acquisition mode
m/z 660
6.84
6.85
100
0
100
Estradiol cypionate
0
2.00
2.00
2.00
4.00
4.00
4.00
4.00
4.00
4.00
4.00
6.00
8.00
2.00
4.00
4.00
4.00
4.00
4.00
4.00
4.00
6.00
8.00
8.00
8.00
8.00
8.00
8.00
m/z 630
6.70
6.71
100
0
100
Estradiol énanthate
0
6.00
5.81
8.00
2.00
6.00
5.82
m/z 618
100
0
100
Estradiol valérate
0
6.00
6.00
6.00
6.00
6.00
8.00
2.00
6.00
6.00
6.00
6.00
6.00
m/z 590
4.72
4.73
100
0
100
Estradiol propionate
0
2.00
2.01
8.00
2.00
m/z 530
2.00
100
100
0
Ethinylestradiol
0
2.00
1.85
8.00
2.00
m/z 509
1.85
100
100
0
Estradiol-d3
0
2.00
1.87
8.00
2.00
m/z 506
1.86
100
100
0
Estradiol-17-benzoate
Time
8.00
0
Time
8.00
2.00
2.00
Figure 6. MRM (left) versus PS extracted chromatogram (right) using the Xevo TQMS
mass spectrometer for estradiol-17-esters, estradiol-d3 and ethinylestradiol (0.2 ng injected).
PS mode for a screening method of (un)known steroid
esters
the [M + H]+ ion is provided, allowing us to determine the
chemical formula of each compound. However, this kind of
result requires a significant time for manual interpretation.
The screening of unknown steroid esters was then
investigated using the precursor ion scan (PS) acquisition
mode and improved with the derivatisation step as described
above.
3.64
1
2
3
5
6
7
4
2.64
The development of this PS method was based on the m/z
171 ion which is specific to the dansylated group. At first,
detection in MRM mode was compared with that in PS mode
on an injected 0.2 ng standard mixture (Fig. 6). As expected,
the sensitivity is lower in PS mode and estradiol-decanoate
was not detected at this level (not shown). However, all the
other estradiol-17-esters were detected when the specific
[M + H]+ precursor ion of each compound of interest was
selected. Moreover, the internal and external standards, i.e.
ethinylestradiol and estradiol-d3, were also detected in this
acquisition mode, the recommendation of the EU criteria
being respected (Decision 2002/657/EC).
After these first results obtained on standard solutions, a
sample obtained from the French national control plan was
analysed. Internal and external standards were added at an
equivalent of 0.75 ppb in the hair sample. The chromatogram
and the associated unknown spectra are presented in Fig. 7.
Estradiol-d3 and ethinylestradiol were easily detected in the
hair sample, as for the standard. Five other compounds were
also detected from which the precursor ion of interest could
be extracted. Additional investigations have to be led by the
analyst at this stage to confirm the origin of these signals, i.
e. comparison with the blank samples, and comparison with
other standard spectra. In this case, these precursor ions were
not identified as belonging to previously known steroid
esters. Therefore, this method appears to be a very interesting
tool for studying the wide range of residues monitored, and
potentially identifying new estradiol-17-esters. In parallel,
1.85
2.01
3.13
3.93
4.00
5.13
7.73
8.00
0
Time
2.00
512.33
3.00
5.00
6.00
7.00
100
7
513.21
534.86 552.28
652.76
590.60
625.52 631.50
618.35
664.72
m/z
0
500
520
540
560
580
600
620
640
660
617.28
618.28
100
0
6
582.36
620.61
620
569.33
667.55
539.70
540
606.33
600
510.63
500
662.64
m/z
m/z
m/z
m/z
520
520
560
560
560
560
580
640
640
660
605.20
607.21
100
5
0
500
540
580
600
620
620
620
660
591.23
592.11
593.24
100
4
633.19
529.07533.72
577.01
580
0
500
520
540
600
640
660
577.20
578.21
100
3
579.27
0
500
520
540
580
600
640
660
Figure 7. Unknown hair total ion current (top) in PS
mode (from m/z 500 to 675; precursor ion scan of m/z 171).
1: estradiol-d3 dansylated, 2: ethinylestradiol, and asso-
ciated mass spectra from 3 to 7 retention times (unknown
compounds).
Rapid Commun. Mass Spectrom. 2012, 26, 819–827
Copyright © 2012 John Wiley & Sons, Ltd.
wileyonlinelibrary.com/journal/rcm

E. Bichon et al.
strategy for detection of anabolic steroids in bovine urine
using gas chromatography tandem mass spectrometry.
Anal. Chim. Acta 2007, 586, 43.
An efficient solution to help the analyst would be a new
feature of the software where, when a precursor ion of
interest is detected above a threshold limit, the instrument
carries out an MRM acquisition under the same conditions
as the PS acquisition (i.e. energy in the collision cell, product
ion to monitor) during at least the peak width time. This
protocol would allow improvements in terms of sensitivity
and analyst comfort, and also in screening efficiency. Such a
mode would make it possible to discard a major part of
the false compliant samples in detecting many of the
estrogen-17-esters.
[5] A.A.M. Stolker, T. Zuidema, M.W.F. Nielen. Residue
analysis of veterinary drugs and growth-promoting agents.
Trends Anal. Chem. 2007, 26, 967.
[6] L. Rambaud, F. Monteau, Y. Deceuninck, E. Bichon,
F. André, B. Le Bizec. Development and validation of a
multi-residue method for the detection of a wide range
of hormonal anabolic compounds in hair using gas
chromatography-tandem mass spectrometry. Anal. Chim.
Acta 2007, 586, 93.
In parallel, the yeast bioassays described by Becue et al.[38]
could be compared with our present screening method. The
main advantage of the precursor ion scan method would be
the direct access not only to the molecular mass of the ester,
but also to that of its native steroid. Currently, the main
drawback is that this approach is not ready for use in
an extended routine analysis because of the processing
difficulties experienced by the operator. A development in
the software would be helpful to extract the significant
precursor ions from the noise observed for each total ion
current (TIC).
[7] M. Gasparini, M. Curatolo, W. Assini, E. Bozzoni,
N. Tognoli, G. Dusi. Confirmatory method for the determi-
nation of nandrolone and trenbolone in urine samples
using immunoaffinity cleanup and liquid chromatogra-
phy-tandem mass spectrometry. J. Chromatogr. A 2009,
1216, 8059.
[8] P. Regal, C. Nebot, B. Vazquez, A. Cepeda, C. Fente.
Determination of the hormonal growth promoter 17
alpha-methyltestosterone in food-producing animals: Bovine
hair analysis by HPLC-MS/MS. Meat Sci. 2010, 84, 196.
[9] M.E. Malone, C. Elliot, G. Kennedy, L. Regan,
H.F. De Brabander, L. Van Ginkel, S. Sterk, A. Bergwerff.
Development of a rapid method for the analysis of syn-
thetic growth promoters in bovine muscle using liquid
chromatography tandem mass spectrometry. Anal. Chim.
Acta 2009, 637, 112.
CONCLUSIONS
[10] C. Blasco, C. Van Poucke, C. Van Peteghem. Analysis of
meat samples for anabolic steroids residues by liquid
chromatography/tandem mass spectrometry. J. Chromatogr.
2007, 1154, 230.
[11] J. Scarth, A. Clarke, J. Hands, P. Teale, R. Macarthur, J. Kay.
Validation of a quantitative multi-residue urinary assay for
the detection of androgen, oestrogen and progestagen abuse
in the bovine. Chromatographia 2010, 71, 241.
[12] M.W.F. Nielen, J.J.P. Lasaroms, M. Essers, M. Sanders,
H. Heskamp, T. Bovee, J.H.A. van Rhijn, M. Groot, C. Van
Peteghem, A. Townshend. The ultimate veal calf reference
experiment: Hormone residue analysis data obtained by
gas and liquid chromatography tandem mass spectrometry.
Anal. Chim. Acta 2007, 586, 30.
[13] C. Buisson, M. Hebestreit, A.P. Weigert, K. Heinrich,
H. Fry, U. Flenker, S. Banneke, S. Prevost, F. Andre,
W. Schaenzer, E. Houghton, B. Le Bizec. Application of
stable carbon isotope analysis to the detection of
17beta-estradiol administration to cattle. J. Chromatogr. A
2005, 1093, 69.
Due to the lack of a threshold value for endogenous steroids
in breeding animals, the detection of steroid esters in hair
can be considered as one of the key pieces of information
needed to detect the synthetic origin of the administered
parent compound. A suitable method for the analysis of
estrogen-17-esters in bovine hair based on LC/MS/MS has
been developed. This method has been validated and it has
been shown that a significant improvement in sensitivity
can be obtained by the introduction of a derivatisation
reagent for the phenol moiety. Furthermore, a new tool
could be developed to screen for the presence of hormone
esters in hair samples. Even if a software improvement is
required, a precursor ion scan method could be adapted to
carry out a large screening. Based on the specific product
ion of the steroid considered, this strategy presents the
advantage of detecting any (un)known steroid ester and
giving access to the [M + H]+ ion of the considered ester in
only a single analysis.
[14] E. Bichon, F. Kieken, N. Cesbron, F. Monteau, S. Prévost,
F. André, B. Le Bizec. Development and application of
stable carbon isotope analysis to the detection of cortisol
administration in cattle. Rapid Commun. Mass Spectrom.
2007, 21, 2613.
REFERENCES
[15] M. Hebestreit, U. Flenker, C. Buisson, F. Andre, B. Le Bizec,
H. Fry, M. Lang, A.P. Weigert, K. Heinrich, S. Hird, W.
Schänzer. Application of stable carbon isotope analysis to
the detection of testosterone administration to cattle.
J. Agric. Food Chem. 2006, 54, 2850.
[1] Council Directive 96/22/EC concerning the prohibition on
the use in stockfarming of certain substances having a
hormonal or thyrostatic action and of beta-agonists, and
repealing Directives 81/602/EEC, 88/146/EEC and 88/
299/EEC. Off. J. Eur. Commun. 1996, L125, 3.
[2] B. Le Bizec, P. Marchand, D. Maume, F. Monteau, F. André.
Monitoring anabolic steroids in meat-producing animals.
Review of current hyphenated mass spectrometric
techniques. Chromatographia 2004, 59, S3.
[3] H. Noppe, B. Le Bizec, K. Verheyden, H. De Brabander.
Novel analytical methods for the determination of
steroid hormones in edible matrices. Anal. Chim. Acta
2008, 611, 1.
[16] B. Destrez, E. Bichon, L. Rambaud, F. Courant, F. Monteau,
G. Pinel, J.-P. Antignac, B. Le Bizec. Criteria to distinguish
between natural situations and illegal use of boldenone,
boldenone esters and boldione in cattle: 2. Direct
measurement of 17[beta]-boldenone sulpho-conjugate in
calf urine by liquid chromatography-high resolution and
tandem mass spectrometry. Steroids 2008, 74, 803.
[17] G. Dervilly-Pinel, L. Rambaud, P. Sitthisack, F. Monteau,
H. Armstrong, G. Kennedy, B. Le Bizec. 5a-Estrane-3b,
17b-diol and 5b-estrane-3a,17b-diol: Definitive screening
[4] S. Impens, J. Van Loco, J.M. Degroodt, H.F. De Brabander,
C. Van Peteghem, A. Townshend. A downscaled multi-residue
wileyonlinelibrary.com/journal/rcm
Copyright © 2012 John Wiley & Sons, Ltd.
Rapid Commun. Mass Spectrom. 2012, 26, 819–827

Monitoring of estradiol esters in bovine hair
biomarkers to sign nandrolone abuse in cattle? J. Steroid
Biochem. Mol. Biol. 2011, 126, 65.
[18] R.L. DuPont, W.A. Baumgartner. Drug testing by urine and
hair analysis: complementary features and scientific issues.
Forensic Sci. Int. 1995, 70, 63.
[19] P. Kintz. Caractérisation formelle de l’usage de testostérone
et de ses esters par analyse des cheveux. Applications au
contrôle du dopage. Science & Sports 2000, 15, 162.
[20] P. Kintz. Testing for anabolic steroids in hair: a review. Legal
Med. 2003, 5, S29.
[21] D. Thieme, P. Anielski, J. Grosse, H. Sachs, R.K. Mueller.
Identification of anabolic steroids in serum, urine, sweat
and hair: Comparison of metabolic patterns. Anal. Chim.
Acta 2003, 483, 299.
[22] M. Villain, M. Chèze, A. Tracqui, B. Ludes, P. Kintz. Testing
for zopiclone in hair application to drug-facilitated crimes.
Forensic Sci. Int. 2004, 145, 117.
[23] M. Gratacos-Cubarsi, M. Castellari, A. Valero, J.A. Garcia-
Regueiro. Hair analysis for veterinary drug monitoring in
livestock production. J. Chromatogr. B 2006, 834, 14.
[24] P. Anielski, D. Thieme, A. Schlupp, J. Grosse, F. Ellendorff,
R. Mueller. Detection of testosterone, nandrolone and
precursors in horse hair. Anal. Bioanal. Chem. 2005, 383,
903.
[25] L. Rambaud, E. Bichon, N. Cesbron, F. André, B. Le Bizec.
Study of 17[beta]-estradiol-3-benzoate, 17[alpha]-methyltes-
tosterone and medroxyprogesterone acetate fixation in
bovine hair. Anal. Chim. Acta 2005, 532, 165.
[26] J. Segura, S. Pichini, S.H. Peng, X. de la Torre. Hair analysis
and detectability of single dose administration of
androgenic steroid esters. Forensic Sci. Int. 2000, 107, 347.
[27] M.W.F. Nielen, J.J.P. Lasaroms, P.P.J. Mulder, J. Van Hende,
J.H.A. van Rhijn, M.J. Groot. Multi residue screening of
intact testosterone esters and boldenone undecylenate in
bovine hair using liquid chromatography electrospray
tandem mass spectrometry. J. Chromatogr. B 2006, 830,
126.
dansyl chloride to enhance electrospray ionization:
Application in trace analysis of ethinylestradiol in rhesus
monkey plasma. Anal. Chem. 2002, 74, 4136.
[31] W. Li, Y.-H. Li, A.C. Li, S. Zhou, W. Naidong. Simultaneous
determination of norethindrone and ethinyl estradiol in
human plasma by high performance liquid chromatography
with tandem mass spectrometry--experiences on developing
a highly selective method using derivatization reagent for
enhancing sensitivity. J. Chromatogr. B 2005, 825, 223.
[32] F. Zhang, M.J. Bartels, J.C. Brodeur, E.L. McClymont,
K.B. Woodburn. Quantitation of 17a-ethinylestradiol in
aquatic samples using liquid-liquid phase extraction, dansyl
derivatization, and liquid chromatography/positive
electrospray tandem mass spectrometry. Rapid Commun.
Mass Spectrom. 2004, 18, 2739.
[33] Y.-H. Lin, C.-Y. Chen, G.-S. Wang. Analysis of steroid
estrogens in water using liquid chromatography/tandem
mass spectrometry with chemical derivatizations. Rapid
Commun. Mass Spectrom. 2007, 21, 1973.
[34] S. Pedreira, M. Lolo, B.I. Vázquez, C.M. Franco,
A. Cepeda, C. Fente. Liquid chromatography–electrospray
ionization–mass spectrometry method in multiple reaction
monitoring mode to determine 17a-ethynylestradiol residues
in cattle hair without previous digestion. J. Agric. Food Chem.
2007, 55, 9325.
[35] H.P. Nguyen, L. Li, J.W. Gatson, D. Maass, J.G. Wigginton,
J.W. Simpkins, K.A. Schug. Simultaneous quantification of
four native estrogen hormones at trace levels in human
cerebrospinal fluid using liquid chromatography-tandem
mass spectrometry. J. Pharmaceut. Biomed. Anal. 2011, 54,
830.
[36] F. Zhang, M.J. Bartels, D.R. Geter, M.S. Carr,
L.E. McClymount, T.A. Marino, G.M. Klecka. Simultaneous
quantitation of testosterone, estradiol, ethinyl estradiol, and
11-ketotestosterone in fathead minnow fish plasma by
liquid chromatography/positive atmospheric pressure
photoionization tandem mass spectrometry. Rapid Commun.
Mass Spectrom. 2009, 23, 3637.
[37] Commission Decision 2002/657/EC of 12 August 2002
implementing Council Directive 96/23/EC concerning the
performance of analytical methods and the interpretation
of results (Text with EEA relevance) (notified under
document number C(2002) 3044). Off. J. Eur. Commun.
2002, L221, 8.
[38] I. Becue, T. Bovee, C. Van Poucke, M. Groot, M. Nielen,
C. Van Peteghem. Applicability of a yeast bioassay in the
detection of steroid esters in hair. Anal. Bioanal. Chem. 2011,
399, 1031.
[28] E. Duffy, M.H. Mooney, C.T. Elliott, M. O’Keeffe. Studies on
the persistence of estradiol benzoate and nortestosterone
decanoate in hair of cattle following treatment with growth
promoters, determined by ultra-high-performance liquid
chromatography-tandem mass spectrometry. J. Chromatogr.
A 2009, 1216, 8090.
[29] P. Aqai, A.A.M. Stolker, J.J.P. Lasaroms. Effect of sample
pre-treatment on the determination of steroid esters in hair
of bovine calves. J. Chromatogr. A 2009, 1216, 8233.
[30] M.R. Anari, R. Bakhtiar, B. Zhu, S. Huskey, R.B. Franklin,
D.C. Evans. Derivatization of ethinylestradiol with
Rapid Commun. Mass Spectrom. 2012, 26, 819–827
Copyright © 2012 John Wiley & Sons, Ltd.
wileyonlinelibrary.com/journal/rcm