Interference of a sulfate conjugate in quantitative liquid chromatography/tandem mass spectrometry through in-source dissociation

Full text of DOI:10.1002/rcm.4580

RAPID COMMUNICATIONS IN MASS SPECTROMETRY
Rapid Commun. Mass Spectrom. 2010; 24: 1817–1819
Published online in Wiley InterScience (www.interscience.wiley.com) DOI: 10.1002/rcm.4580
early eluting salts from entering the LC/MS interface.
Conditions for MS analysis of biochanin A included an ion
spray voltage of ꢀ4500 V, and a temperature of 3508C.
Nebulizer and curtain gas flow were 10 mL/min and 8 mL/
min, respectively. The fragment was induced with a collision
energy of ꢀ30 eV. The optimized declustering potential (DP),
focusing potential and collision cell exit potential were ꢀ60,
ꢀ175 and ꢀ30 V, respectively. The MS was performed in a
negative ion mode in multiple reaction monitoring (MRM)
mode. The m/z ratios of precursor ion and product ion of
biochanin A were 283 and 268, respectively. A XTerra MS
C18 column (2.1 ꢁ 150mm i.d., 3.5 mm; Waters Corporation,
Milford, MA, USA) was used with a flow rate of 200 mL/min.
The mobile phase we used initially consisted of 60%
acetonitrile with 0.1% formic acid (mobile phase B) and
40% of water with 0.1% formic acid (mobile phase A). A peak
with a clear shoulder was observed in the channel for
biochanin A (m/z 283!268), as shown in Fig. 1(A). As
discussed by Liu et al.,4 the use of a deteriorated column or an
inappropriate reconstitution solvent is usually the common
reason contributing to a peak shoulder. However, neither of
the reasons was applicable to our case since we used a brand
new column and appropriate reconstitution solution. After
the composition ratio of mobile phases A and B was adjusted
to 50:50, the shoulder was resolved into a second peak, as
shown in Fig. 1(B). Since glucuronide conjugates are the most
commonly reported interferences and glucuronidation has
been reported to be one of the major pathways regarding the
metabolism of biochanin A, we hypothesized that this
interference peak was a biochanin A glucuronide conjugate.
However, no peak was observed when we added the
RCM
Letter to the Editor
Dear Editor
Interference of a sulfate conjugate in quantitative liquid
chromatography/tandem mass spectrometry through in-
source dissociation
Over the last 15–20 years, high-performance liquid chroma-
tography (HPLC) with tandem mass spectrometry (MS/MS)
has played an increasingly important role in the quantifi-
cation of analytes in complex biological matrices, due to its
superior sensitivity and specificity. The separation of
analytes in LC/MS/MS is achieved first by their physio-
chemical properties by the chromatographic separation and
secondly by the mass-to-charge resolution provided by the
mass spectrometer. Due to this specificity of the mass
spectrometer, multiple compounds can be quantified sim-
ultaneously within a very short analytical run time. This
property of LC/MS/MS has been appreciated especially by
the pharmaceutical industry: chromatographic run times of
less than 3 min are commonly applied to allow the high
throughput of samples.1,2 MS/MS provides an accurate
specific method for most analytes by coupling a specific
precursor ion with a unique fragment ion. There are cases
however where interferences can arise. In-source dissoci-
ation of metabolites can potentially produce ions identical to
the precursor ions of the compound under study.1–3 The
most commonly reported interferences that can undergo in-
source dissociation are different types of glucuronide
conjugates, including acyl glucuronides, carbamoyl glucur-
onides and N- and O-glucuronides.3–5 In addition to glucur-
onides, in-source dissociation of N-oxide metabolites has also
been documented.6 In this communication, we report the
interference arising from a sulfate metabolite in LC/ESI-MS/
MS. To the best of our knowledge, there have been no reports
on interference generated from a sulfate conjugate. The
sulfate conjugate was almost undetected in our investigation
because it was eluted almost at the same time as the parent
drug under the chromatographic conditions initially used.
The parent compound we investigated here is biochanin
A, one of the most commonly consumed isoflavones.7 The
samples were collected in transport buffer (HBSS) when we
evaluated the transport of biochanin A across MDCK
(Madin-Darby canine kidney) cell monolayers.
precursor to product ion transition for biochanin
A
glucuronide (459!283) to our MS condition. Therefore, this
interference peak either is a non-glucuronide metabolite or is
a glucuronide conjugate undergoing complete in-source
dissociation under our current MS conditions. To further
investigate whether this interference was a biochanin A
glucuronide conjugate, an aliquot (50 mL) of sample was
incubated with an equal volume of buffer containing no
enzyme or b-glucuronidase (in 350 mM phosphate buffer,
pH 6.8). No change of either peak was observed when the
sample was incubated with 50 U glucuronidase, indicating
that this interference peak is not a glucuronide conjugate
(data not shown). For those compounds with phenolic
group(s), in addition to O-glucuronidation, O-sulfation is
also a common metabolic pathway. Therefore, we speculated
that the interference peak observed might be a sulfate
metabolite. Consistent with our hypothesis, enzyme
hydrolysis of transport samples using 5 U sulfatase resulted
in the disappearance of the interference peak and an increase
in the biochanin A aglycone peak, indicating that the inter-
fering peak was a sulfate metabolite. In addition, we added
the precursor to product ion transition for biochanin A
sulfate (363!283) to our MS method and a peak was
observed at the same retention time as our interference peak
observed in the 283!268 channel (shown in Fig. 2). The
fragment at m/z 283 was strong evidence that this peak
contained biochanin A. The precursor ion of 363 indicated
LC/MS/MS was performed using an Applied Biosystems
API 3000 triple-quadruple tandem mass spectrometer linked
to a TurboIonspray interface and a Shimadzu Prominence
liquid chromatograph. To minimize the ion suppression
caused by the high concentration of salts in the transport
buffer (HBSS), the flow from the LC column was diverted to
waste for the first 2 min using a diversion valve to prevent the
that this metabolite contained
a monosulfate moiety
Copyright # 2010 John Wiley & Sons, Ltd.

1818 Letter to the Editor
Figure 1. MRM chromatograms (negative ion mode) of bio-
chanin A sample when the mobile phase consisted of (A) 60%
of mobile phase B (acetonitrile with 0.1% formic acid) and
40% of mobile phase A (water with 0.1% formic acid); or (B)
50% of mobile phase B and 50% of mobile phase A. Biochanin
A was detected under the channel of 283/268.
Figure 2. MRM chromatograms (negative ion mode) of an
aliquot of a 50 mL sample (from transport study) incubated
with an equal volume of buffer containing (A) no enzymes; (B)
5 U of sulfatase. The red line represents the chromatogram
obtained under MRM of 283/268 (for the measurement of
biochanin A). The blue line represents the chromatogram
obtained under MRM of 363/283 (for the measurement of
biochanin A sulfate). Biochanin A was eluted at about 6.3 min
and biochanin A sulfate was eluted at about 4.7 min.
(molecular weight (mw) 80). The proposed biochanin A
sulfate conjugate formation (in MDCK cells) and in-source
dissociation during the ionization process (in MS) are shown
in Fig. 3.
There are reports in the literature of different MS
parameters as the cause of the in-source dissociation. For
example, Yan et al.5 reported that source temperature had
little effect on the in-source glucuronide dissociation and the
presence of glucuronide interference was dependent on the
cone voltage. However, the data from another research
group4 clearly showed temperature-dependent in-source
dissociation: the carbamoyl glucuronide conjugate displayed
slight in-source fragmentation at 3008C and much more
fragmentation was obtained when the temperature was
increased to 4008C. To investigate whether in-source dis-
sociation of the sulfate conjugate was also sensitive to certain
MS parameter(s) in our current study, we further examined
the stability of biochanin A sulfate at different MS source
parameters using two MS instruments from different
manufacturers. Instrument 1 was a Thermal Scientific TSQ
quantum mass spectrometer and instrument 2 was an App-
lied Biosystems API 3000 triple-quadruple tandem mass
spectrometer. The electrospray ionization (ESI) method
was used in both instruments. As shown in Fig. 4, the in-
source dissociation of the sulfate metabolite is dependent on
the temperature in instrument 1. A higher degree of in-source
fragmentation was observed as capillary temperature was
increased. When samples were analyzed using instrument
2, the in-source dissociation of the sulfate conjugate demon-
strated both temperature- and DP-independence (data not
shown). Our results indicated that the stability of the sulfate
conjugate may be dependent on not only certain MS
parameters, but also different instruments. It should be
noted that the ionization process used in the current study is
ESI,
a process usually operated at a relatively low
temperature. Since atmospheric pressure chemical ionization
(APCI) is also a widely used ionization method and usually
requires higher temperatures than ESI, extra caution should
Copyright # 2010 John Wiley & Sons, Ltd.
Rapid Commun. Mass Spectrom. 2010; 24: 1817–1819
DOI: 10.1002/rcm

Letter to the Editor 1819
Figure 3. Proposed biochanin A sulfate formation (in MDCK cells) and in-source dissociation in the
ionization process (in MS). In-source dissociation of sulfate from biochanin sulfate in ESI generates
a fragment ion which is identical to the precursor ion of biochanin A.
be exercised when APCI is used to analyze samples
containing both target compound and sulfate conjugates.
Sulfotransferase-catalyzed conjugation is one of the most
important biotransformation reactions and various endo-
biotics and xenobiotics are substrates of sulfotransferases
(SULTs).8,9 It has been reported that many of the substrates
metabolized by UDP-glucuronosyltransferases (UGTs) are
also substrates for SULTs.10,11 Previously, glucuronide
metabolites, but not sulfate metabolites, were reported to
interfere with the parent compound thorough in-source
dissociation. One possible reason is that in many of the
previous studies, samples were collected after incubation of a
drug in microsomal preparations, a widely used in vitro
system that does not contain SULTs. Our samples were
collected from transport buffer when the transport of
biochanin A across MDCK cell monolayers was investigated.
It should be noted that Caco-2 and MDCK cells, both of
which have SULT expression, have been used extensively in
studies to characterize drug metabolism and drug per-
meability.12
In conclusion, potential interferences from sulfate con-
jugates have been investigated and extra caution should be
taken when establishing conditions for a LC/MS/MS assay
to detect a new compound that can be metabolized mainly by
UGT and/or SULT.
Acknowledgements
Supported in part by a grant from the Susan G. Komen Breast Cancer
Foundation (BCTR0601385).
Guohua An, Donna M. Ruszaj and Marilyn E. Morris*
Department of Pharmaceutical Sciences, School of Pharmacy
and Pharmaceutical Sciences, University at Buffalo, State
University of New York, Amherst, NY 14260-1200, USA
*Correspondence to: M. E. Morris, Department of Pharmaceutical
Sciences, 517 Hochstetter Hall, University at Buffalo, State Univer-
sity of New York, Amherst, NY 14260-1200, USA.
E-mail: memorris@buffalo.edu
REFERENCES
1. Romanyshyn L, Tiller PR, Hop CE. Rapid Commun. Mass
Spectrom. 2000; 14: 1662.
2. Matuszewski BK, Constanzer ML, Chavez-Eng CM. Anal.
Chem. 1998; 70: 882.
3. Jemal M, Ouyang Z. Rapid Commun. Mass Spectrom. 2000; 14:
1757.
4. Liu DQ, Pereira T. Rapid Commun. Mass Spectrom. 2002; 16: 142.
5. Yan Z, Caldwell GW, Jones WJ, Masucci JA. Rapid Commun.
Mass Spectrom. 2003; 17: 1433.
6. Korfmacher WA. Using Mass Spectrometry for Drug Metab-
olism Studies. CRC Press: Boca Raton, 2004.
7. Lissin LW, Cooke JP. J. Am. Coll. Cardiol. 2000; 35: 1403.
8. Gamage N, Barnett A, Hempel N, Duggleby RG, Windmill
KF, Martin JL, McManus ME. Toxicol. Sci. 2006; 90: 5.
9. Kauffman FC. Drug Metab. Rev. 2004; 36: 823.
10. Moon YJ, Sagawa K, Frederick K, Zhang S, Morris ME. Aaps
J. 2006; 8: E433.
11. Shelnutt SR, Cimino CO, Wiggins PA, Ronis MJ, Badger TM.
Am. J. Clin. Nutr. 2002; 76: 588.
12. Balimane PV, Chong S, Morrison RA. J. Pharmacol. Toxicol.
Methods. 2000; 44: 301.
Figure 4. MRM chromatograms (negative ion mode) of a
sample containing biochanin A and its sulfate metabolite in
ESI mode using the Thermal Scientific TSQ Quantum Ultra
mass spectrometer. The sulfate metabolite displays (bottom)
slight in-source dissociation at 2008C; (middle) moderate in-
source dissociation at 3008C; and (top) extensive in-source
dissociation at 4008C. The chromatograms were obtained
under MRM of 283/268.
Received 26 March 2010
Revised 11 April 2010
Accepted 14 April 2010
Copyright # 2010 John Wiley & Sons, Ltd.
Rapid Commun. Mass Spectrom. 2010; 24: 1817–1819
DOI: 10.1002/rcm