Journal of
MASS
L-K Zhang et al.
trast, UHPLC-MS and UHPLC-MS/MS are highly sensitive tech-
nique requiring typically less than 0.1μg of material for analysis.
The capabilities of separating a mixture containing highly varied
concentrations of analytes and structural characterization of impu-
rities have led to the increased use of UHPLC-MS. The goal of this
study is to develop a rapid and sensitive method for the structural
determination of the major impurities of doravirine. The study uti-
lizes UHPLC-HRMS/MS techniques to perform structure elucidation
of the unknown impurities. This approach was instrumental for im-
purity identification, and a total of five impurities of doravirine were
characterized using the developed method.
pattern of doravirine provides a basis for assessing structural as-
signment for the impurities. The ability of the Waters Xevo G2-XS
MS instrument to provide an exact-mass measurement for each
of the ions produced in tandem MS assists greatly in the investiga-
tion of the fragmentation mechanism of doravirine. The molecular
weight of doravirine was confirmed to be 425 Dalton (Da) by
LC/MS analysis (spectrum not presented). The elemental composi-
tion of doravirine was established by high-resolution MS (MH+:
C17H12ClF3N5O+3). The calculated exact mass and measured mass
were 426.0575 Da and 426.0577 Da, respectively. The mass
measurement error of À0.5ppm was achieved using internal
calibration. Collision-induced decomposition (CID) of the ESI-
produced MH+ precursor generates ions at m/z 315, 295, 260
and 112 (Fig. 1). The most abundant product ion (m/z 315.0145)
observed in the spectrum represents the protonated ion of
3-chloro-5-((2-oxo-4-(trifluoromethyl)-1,2-dihydropyridin-3-yl)oxy)
benzonitrile (Scheme 1). The isotopic pattern of the fragment ion
indicates it contains one chlorine atom. It could further decom-
pose to an ion of m/z 295.0082 by a loss of hydrogen fluoride
in the gas phase (Scheme 1). A minor fragment ion, m/z
260.0395, is generated from m/z 295.0082 ion by further loss of
a chorine radical. A complementary fragment ion of m/z
315.0145 is also observed at m/z 112.0497 (Fig. 1). The mass
measurement errors of these product ions are determined to
be À0.6 ppm, 0.3 ppm, À0.8 ppm and À7 ppm, respectively
(Scheme 1). Although the accurate mass measurement of the
m/z 112 fragment displays a relative large ppm error (À7 ppm,
À0.8 mmu accuracy), it is still sufficient by today’s standards for
unambiguous determination of a chemical formula of this low-
mass fragment.[11] Moreover, we performed the tandem MS anal-
ysis of the doravirine on a TheromFisher Orbitrap MS system and
confirmed the formula of the m/z 112 fragment by ultra-high re-
solving power MS analysis (Cal. 112.0505, Found: 112.0506. error:
0.9 ppm; spectrum not shown).
Experimental
Materials
Doravirine (3-chloro-5-((1-((4-methyl-5-oxo-4,5-dihydro-1H-1,2,4-
triazol-3-yl)methyl)-2-oxo-4-(trifluoromethyl)-1,2-dihydropyridin-3-
yl)oxy)benzonitrile) and the synthesized standards were pro-
vided by Merck Research Laboratories. The stock solution of
doravirine was prepared by dissolving the compound in
acetonitrile/water (80:20, v:v) at the concentration of 0.1mg/
ml. Formic acid (99.9%) was purchased from Sigma-Aldrich
Chemical Co. (St. Louis, MO, USA) and used without further pu-
rification. Acetonitrile and water were also obtained from
Sigma-Aldrich Chemical Co.
UHPLC-HRMS and UHPLC-HRMS/MS analysis
LC/ESI-MS and LC/ESI-MS/MS analysis were performed on a Waters
Xevo G2-XS Q-TOF mass spectrometer (Waters, Milford, MA)
coupled to an I-Class Waters Acquity UPLC system, operated in ei-
ther positive-ion or negative-ion mode. The UHPLC system
consisted of a binary pump system, a sample manager and a TUV
detector (Waters, Milford, MA). All UHPLC separations were per-
formed at 40 °C. The samples were analyzed on a reverse-phase
C18 column (Waters BEH C18, 100 mm× 2.1 mm, I.D., 1.7μm) using
A (Water/Formic acid; 99.9:0.1, v:v) and B (Acetonitrile/Formic Acid;
99.9:0.1, v:v) as mobile phases at a flow rate of 0.45 ml/min. The in-
jection volume was 0.5μl. The mobile phase gradient started at 5%
B, and linearly increased to 25% B in 1 min, then linearly increased
to 40% B over 5 min, and followed by linearly increased to 90% over
4 min. The tandem MS analysis of all the compounds was per-
formed on the same Waters Xevo G2-XS MS instrument in the
MS-MS mode. The collision energy was 20–35eV for all the analytes
and standards. Calibration was achieved using sodium formate and
the resolution of the time-of-flight was ~30 000, full width half max-
imum (FWHM) at m/z 556 (m/z 554, negative-ion mode). Leucine
enkephalin was used as the lock mass (m/z of 556.2772 [M + H]+;
554.2615 [M À H]À) for accurate mass calibration. MS3-like spectra
were generated by in-source fragmentation, followed by selection
of the first-generation product ion in the quadrupole and fragmen-
tation in the collision cell. The tandem mass spectral data for MK-
1439 was also obtained on a Thermo LTQ Orbitrap mass spectrom-
eter (Thermo Fisher Scientific) operated at a resolution of 30 000.
Doravirine impurity profile
To assess the purity of doravirine, a MS friendly UHPLC method was
developed as described in the experimental section. In order to de-
termine the molecular weight of these impurities, the ESI source pa-
rameters were optimized to reduce the background signals and to
minimize fragmentation reactions. The UHPLC-UV chromatogram
and the UHPLC-MS total ion chromatogram (TIC) were obtained
for the doravirine (Fig. 2). According to the UHPLC-UV chromato-
gram, a total of five major trace-level impurities are identified (A,
B, C, D and E), and the relative retention times (RRT) of the five com-
ponents were found to be 0.56, 0.83, 0.92, 1.10 and 1.35, respec-
tively. The molecular weights of these impurities are determined
to be 443, 459, 536, 314 and 546 Da, respectively.
Mass accuracy is critical in establishing compound identity.
Accurate mass measurement (i.e. that less than 1 ppm) can aid in
the characterization of the impurities by placing constraints on ele-
mental formulas of the unknowns.[9,10] Despite significant advances
of the HRMS, identification and structure elucidation of unknowns
remain challenging. Usually, high mass accuracy (1 ppm) alone still
cannot determine the molecular formulas of impurities. Thus, the
combination of accurate mass measurement and the isotope
pattern abundance errors using calculated isotopic distributions is
needed to obtain more reliable prediction of the molecular formulas
of unknown substances. Commercial software has been developed
to predict empirical formulas of unknowns based on both accurate
mass measurements and calculated isotopic distributions.[12]
Results and discussion
Doravirine structure characterization by high-resolution tan-
dem MS analysis
An initial step in elucidating impurity structures of doravirine is to
understand the fragmentation pattern of the drug substance.
Extensive tandem mass spectrometry analysis of the fragmentation
wileyonlinelibrary.com/journal/jms
Copyright © 2016 John Wiley & Sons, Ltd.
J. Mass Spectrom. 2016, 51, 959–968