Identification, synthesis, and strategy for minimization of potential impurities observed in raltegravir potassium drug substance

Article abstract of DOI:10.1021/op300077m

Multiple sources of anticipated degradation and process impurities of raltegravir potassium drug substance observed during the laboratory optimization and later during its bulk synthesis are described in this article. The impurities were monitored by UPLC, and their structures are tentatively assigned on the basis of fragmentation patterns in LC-MS and NMR spectroscopy. Most of the impurities are synthesized, and their assigned constitutions were confirmed by co-injection in UPLC. In addition to the formation, synthesis, and characterization, strategy for minimizing these impurities to the level accepted by ICH is also described. We feel that our study will be helpful to the generic industry for obtaining chemically pure raltegravir potassium.

Full text of DOI:10.1021/op300077m

Communication
pubs.acs.org/OPRD
Identification, Synthesis, and Strategy For Minimization of Potential
Impurities Observed In Raltegravir Potassium Drug Substance
Gulabrao D. Patil, Siddheshwar W. Kshirsagar, Shivnath B. Shinde, Pankaj S. Patil, Mangesh S. Deshpande,
Ashok T. Chaudhari, Swapnil P. Sonawane,* Golak C. Maikap, and Mukund K. Gurjar
API R & D Centre, Emcure Pharmaceuticals Ltd., ITBT Park, Phase-II, M.IDC Hinjewadi, Pune - 411057, India.
S
* Supporting Information
This study will help a synthetic organic chemist to understand
the potential impurities in raltegravir potassium synthesis and
thereby obtain the pure compound.
ABSTRACT: Multiple sources of anticipated degradation
and process impurities of raltegravir potassium drug
substance observed during the laboratory optimization
and later during its bulk synthesis are described in this
article. The impurities were monitored by UPLC, and their
structures are tentatively assigned on the basis of
fragmentation patterns in LC−MS and NMR spectrosco-
py. Most of the impurities are synthesized, and their
assigned constitutions were confirmed by co-injection in
UPLC. In addition to the formation, synthesis, and
characterization, strategy for minimizing these impurities
to the level accepted by ICH is also described. We feel that
our study will be helpful to the generic industry for
obtaining chemically pure raltegravir potassium.
To ensure that desired drug metabolism, safety and clinical
studies are not jeopardized by inconsistent purity or impurities
having potential harmful toxicological properties, when
raltegravir was synthesized by a commercial route (Scheme
1),4a−c many impurities were appearing in the drug substance.
As regulatory guidelines promulgated by the International
Conference on Harmonization (ICH)5 dictate rigorous
identification of impurities at levels of 0.1%, we were required
to identify a total of nine impurities appearing in the sample of
the drug substance. The literature reveals few impurities formed
due to degradation, incomplete reaction, or side reactions.6
However, no synthetic details have been reported. In this
context, the present study describes identification, synthesis,
characterization, and strategy for controlling these impurities.
INTRODUCTION
■
Raltegravir potassium (MK-0518), belongs to a new class of
compounds discovered for development as inhibitors of HIV
RESULTS AND DISCUSSION
■
The first part of this article elaborates the identification and
possible pathways for the formation of impurities, while in the
second part, strategies we adapted for minimizing these
impurities to the level accepted by ICH are described.
Being a popular approach for impurity identification, LC−
MS was used to identify impurites in raltegravir potassium by
which we observed molecular weights for impurities A−I
(Table 1). These structures were later confirmed by NMR,
mass, and UPLC co-injection.
Impurities A and B are the hydrolysis products of the
raltegravir precursors 9 and 14. However, the possibility of
disconnection of the oxadiazole amide bond as shown in
Scheme 2 cannot be ruled out. During our study we realized
that the impurities listed in Table 1, cannot be removed from
raltegravir potassium due to solubility issues; therefore, we
decided to remove these impurities from raltegravir as the
impurity profile of raltegravir remains unchanged when it is
converted to its potassium salt 1.
integrase, an enzyme catalyzing the integration of viral DNA
into the host genomic DNA, thus preventing further virus
replication.1−3 It was approved on October 12, 2007, and is
marketed as ISENTRESS as a complementary agent to existing
anti-retroviral therapies.
The primary objectives of process chemical research are the
development of efficient, scalable, and safe reproducible
synthetic routes to drug candidates within the developmental
space and acting as a framework for commercial production in
order to meet the requirement of various regulatory agencies.
Therefore, assessment and control of the impurities in a drug
substance and drug product are important aspects of drug
development for the development team to obtain various
marketing approvals. It is extremely challenging for an organic
chemist to identify the impurities which are formed in very
small quantities in a drug substance and wearisome if the
product is nonpharmacopeial. Thus, in our study we described
the formation, identification, synthesis, and characterization of
impurities found in the preparation of raltegravir potassium.
The final step of raltegravir involves the amide bond
formation in which amine 9 and oxadiazole carbonyl chloride
14 are reacted in the presence of N-methylmorpholine to give
pivaloyl protected raltegravir (Scheme 3). The reaction mixture
was directly treated with an aqueous base such as ammonia,
methyl amine, or potassium hydroxide to hydrolyze the pivaloyl
Received: March 19, 2012
Published: July 13, 2012
© 2012 American Chemical Society
1422
dx.doi.org/10.1021/op300077m | Org. Process Res. Dev. 2012, 16, 1422−1429

Organic Process Research & Development
Communication
Scheme 1. Synthesis of raltegravir
Scheme 2. Amide disconnection
Scheme 3. Final step of raltegravir
ester. Acidification provides the desired product, raltegravir
contaminated with impurities listed in Table 1.
Impurity G could be the source for impurity D and forms
when ammonia is used for quenching excess oxalyl chloride.4
Impurities E and F are the desfluoro analogues of raltegravir
and may be forming during −Cbz removal of 8 under acidic
conditions where as impurity I is forming due to condensation
of amine 9 with oxalyl chloride during the final step of
raltegravir in about 0.5%. This is a very crucial impurity and due
to solubility issues is extremely difficult to remove.
Synthesis and Control. Impurities A and B are the
degradation products, while impurity C is the hydrolysis
product of raltegravir and was synthesized by stirring raltegravir
with acetonitrile and aqueous KOH. The pure product was
isolated by column chromatography. The structure was
Plausible Pathways. The process impurities A, B, and H
are observed in about 0.7−2.0%. Impurity H may be due to the
carryover of byproduct 7 to the final step. Impurity G may be
forming due to the presence of residual ethyl oxalic acid in 13,
which was also converted to acid chloride and reacted with 9
during the reaction of 9 and 14 in about 0.5−1.2%. Further, it
was difficult to remove the impurity G from raltegravir
potassium, without significant loss in yield. Impurities C, D,
E, and F are the degradation products, and I is the process
impurity. The degradation pathway for impurity C due to
hydrolysis of the oxadiazole moiety is shown in Scheme 4.
1423
dx.doi.org/10.1021/op300077m | Org. Process Res. Dev. 2012, 16, 1422−1429

Organic Process Research & Development
Communication
Table 1
a
Waters Acquity UPLC; column: Acquity CSH (2.1 mm × 100 mm), 1.7 μ; flow rate: 0.4 mL/min, λ: 210 nm; injection vol: 3.0 μL; mobile phase-
A: 1 mL perchloric acid in 1000 mL water; mobile phase-B: ACN/water (65:35); Run time: 25 min; Raltegravir relative retention time: about 13.38
min.
confirmed by proton NMR, supported by 13C NMR and
DEPT-135 experiment, observing two sets of carbonyl due to
four amide groups.
A synthetic protocol for impurities D and G is shown in
Scheme 5. Impurity G can be synthesized by the process
disclosed by Benedetta et al.7 in which amine 15 was reacted
with ethyl oxalyl chloride (11). Impurity D was synthesized
1424
dx.doi.org/10.1021/op300077m | Org. Process Res. Dev. 2012, 16, 1422−1429

Organic Process Research & Development
Communication
Scheme 4. Formation of impurity C
Scheme 5. Synthesis of impurities D and G
Scheme 6. Synthesis of impurities E and F
Scheme 7. Synthesis of impurity H
from impurity G by passing ammonia gas in a pressure reactor
and methanol as the solvent almost quantitatively.
Therefore, impurity H was synthesized exclusively by the
route disclosed in Scheme 7.
Impurities E and F are synthesized using the synthesis
protocol of raltegravir (Scheme 1), except the 4-fluorobenzyl
amine was replaced with 4-methoxybenzyl amine and benzyl
amine, respectively (Scheme 6) in good yields and purity.
The probability of anticipated impurities due to positional
isomers of 4-fluorobenzyl amine such as 2-fluorobenzyl amine
and 3-fluorobenzyl amine was ruled out after evaluating the
synthetic route of 4-fluorobenzyl amine.
Impurity H is another potential process impurity observed in
raltegravir drug substance. After careful scrutiny of reported
routes, we envisaged that balancing the steric factors of the 4-
pyrimidone moiety may lead to selective O-methylation.
Modification of a literature procedure provided an efficient
process for the oxadiazole fragment 20 with high efficiency and
was robust and simple to perform with a direct crystallization to
afford the product 20 with good purity and yield. Formation of
compound 18 was confirmed by observing the IR signals due to
the nitrile group at 2243 cm−1 and the amide carbonyl at 1693
cm−1. The structure was supported by mass (M + 1; 195.1) and
proton NMR. Compound 18 was almost quantitatively
converted to amidoxime 19. Disappearance of IR signals due
to the nitrile group and appearance of two IR signals at 3491
cm−1 and 3456 cm−1 due to primary amine, supported by mass
(M + 1; 228.2) and proton NMR confirmed the structure of
1425
dx.doi.org/10.1021/op300077m | Org. Process Res. Dev. 2012, 16, 1422−1429

Organic Process Research & Development
Communication
Scheme 8. Synthesis of impurity I:
a
Table 2. Optimization of reaction condition
purity
raltegravir
b
sr
solvent
mole ratio of 13
mole ratio of C2O2Cl2
reaction time (h)
mole ratio of 9
imp-I
yield (%)
1
THF
NMP
ACN
ACN
ACN
ACN
ACN
ACN
ACN
ACN
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.2
1.2
1.2
1.2
1.0
0.9
0.9
0.9
0.9
0.9
2
2
2
4
4
4
4
4
2
1.0
1.0
1.0
1.0
1.0
1.0
0.9
0.8
0.8
0.8
0.81
1.23
1.51
0.54
0.37
0.18
0.10
0.07
0.17
0.07
92.48
97.22
95.33
96.37
96.55
96.72
97.95
98.55
98.11
98.58
78
65
80
81
79
78
88
92
91
92
2
3
4
5
6
7
8
9
10
6
a
b
All reactions performed at 0−5 °C, and N-methylmorpholine as the base. Yields calculated based on 9.
Scheme 9
19. Michael addition of amidoxime 19 to DMAD followed by
thermal rearrangement furnished the synthesis of key 20. The
structure was confirmed by proton NMR and mass spectros-
copy. After obtaining the key hydroxypyrimidinone 20,
methylation of 20 was attempted using Mg(OCH3)2/CH3I in
DMSO solvent.4 Unexpectedly, methylation of 20 gave a clean
desired O-methyl product 21. Two methyl signals appearing at
δ 3.97 and δ 3.88 due to ester and −OMe group supported by
mass spectra (M + 1; 338.2) confirmed the structure. UPLC
and NMR analysis revealed that product 22 was formed in this
reaction albeit in minor quantities. Compound 21 was then
reacted with 4-fluorobenzyl amine to furnish the synthesis of
impurity H. This impurity can be controlled by repeated
crystallization of 6 from toluene. This impurity is highly soluble
in toluene; it can even be removed in the final stage by
crystallization of raltegravir from toluene or methanol.
Impurity I, another critical impurity, was synthesized by the
reaction of 15 with 0.5 equivalent of oxalyl chloride in
acetonitrile at lower temperature. A pure sample of impurity I
was obtained by crystallization from acetonitrile (Scheme 8).
The structures of all these impurities were confirmed by
proton NMR and mass spectroscopy, whereas relative retention
times were established using UPLC.
oxalyl chloride and oxadiazole potassium salt 13, in such a way
that the excess oxalyl chloride should not be available in the
reaction media. After detailed study we observed that the
reaction time for the reaction of 13 with oxalyl chloride to form
14 and mole ratio of 14 with respect to 9 are the important
aspects to reduce the formation of impurity I. Table 2
summarizes some valuable experiments.
The above study indicates that entry 8 of Table 2 was the
best experiment in this series and acetonitrile was the best
solvent for this reaction. Entry 8 of Table 2 also shows that
sufficient time should be given for the conversion of 13 to 14.
Additionally the mole ratio of 9 is also important in minimizing
the critical impurity I to the level acceptable to ICH. When
product is obtained by the conditions reported in entry 8 of
Table 2 and purified from methanol, all impurities were
reduced to below 0.1%, and all organic volatile impurities
(OVI) were found well within ICH criteria, particularly
acetonitrile was found below 5 ppm by GC-HS.
Impurities A and B are controlled by selecting the mole
ratios given in entry 8 of Table 2. We observed that in the
reaction performed using 0.8 equiv of 9, the unreacted
oxadiazole moiety, i.e. impurity A, being highly water soluble,
gets removed during workup.
Control. During our study, we observed that among all
impurities, impurity I is the most critical impurity and it can be
controlled to the level 0.1% during the final step of
condensation between oxadiazole carbonyl chloride 14 and
amine 9. This was achieved by controlling the mole ratios of
Impurity C is the degradation product, and it forms up to
0.2% during alkaline hydrolysis of pivaloyl raltegravir to
raltegravir; this impurity gets removed in methanol purification.
Impurities D and G are the byproducts and appear in
raltegravir due to carryover of ethyl oxalic acid during the
1426
dx.doi.org/10.1021/op300077m | Org. Process Res. Dev. 2012, 16, 1422−1429

Organic Process Research & Development
Communication
synthesis of 14. Therefore 13 was purified using 3 volumes of
methanol at ambient temperature to remove the residual ethyl
oxalic acid.
Impurities E and F are controlled by performing −Cbz
removal at 10−15 °C using 10% dry Pd/C. Impurity H is the
process impurity, and during process optimization we observed
that, if the % of 7 is below 5% in 6, then this impurity does not
appear in raltegravir. Therefore, 6 was purified using refluxing
toluene to lower the level of byproduct 7 below 5% (Scheme
9).
7.35 (m, 2H), 7.11−7.15 (m, 2H), 6.83 (br s, 1H), 4.46 (d, 2H,
J = 5.64 Hz), 3.37 (s, 3H), 1.64 (s, 6H); IR 3667, 1682, 1346,
1223 cm−1; MS (CI): calcd For C18H20FN5O5 (M + H)/z:
406.3, found: (M + H)/z: 406.3.
Synthesis of Impurity E. To a slurry of hydroxypyr-
imidone 64 (45 g, 0.119 mol) and methanol (315 mL) was
added benzyl amine (19.26 g, 0.179 mol) over 30 min. The
resultant solution was aged at 60−65 °C for 4−5 h. The
conversion was monitored by TLC (CHCl3/MeOH, 9:1). After
TLC showed the disappearance of 6, the reaction mixture was
cooled to 50−55 °C, and acetic acid (13.5 mL) was added.
Water (360 mL) was added, and the precipitated solid was
stirred at 25−35 °C for 30−45 min and filtered; the cake was
washed with 1:1 of methanol/water (2 × 50 mL) and dried at
40 °C under reduced pressure to afford 16 (50 g, 98.42%
purity) in 93% yield.
CONCLUSION
■
For the better understanding of the synthetic pathway of an
active pharmaceutical ingredient (API) it is necessary to
identify all the impurities formed/anticipated. In this regard, we
have synthesized and characterized all potential impurities of
raltegravir potassium. In addition to this we have also
demonstrated the strategy for minimization of these impurities
to the level accepted by the International Conference on
Harmonization (ICH).
Mp; 178−183 °C; 1H NMR (400 MHz, CDCl3) δ 11.94 (br
s, 2H), 7.78 (br s, 1H), 7.39−7.26 (m, 10H), 5.00 (s, 2H), 4.60
(d, 2H), 3.64 (s, 3H), 1.66 (s, 6H); IR 3312, 1716, 1678, 1593,
1249 cm−1; MS (CI): calcd For C24H26N4O5 (M + H)/z:
451.4, found: (M + H)/z: 451.2.
To a clean autoclave was charged compound 16 (49 g, 0.108
mol) in a mixture of 10% Pd/C (0.98 g), methanesulfonic acid
(10.79 g, 0.114 mol), and methanol (392 mL). The content
was hydrogenated at 50−55 °C for 2−3 h at 2 kg/cm2
hydrogen pressure. The conversion was monitored by TLC
(CHCl3/MeOH, 9:1). After completion, the reaction mixture
was cooled to 25−35 °C, and hydrogen was replaced with
nitrogen. The Pd/C was removed by Celite filtration, and to
the filtrate was added 1 N NaOH to adjust the pH 7−8. The
precipitated amine was filtered and dried to yield 2-(2-
aminopropan-2-yl)-N-benzyl-1,6-dihydro-5-hydroxy-1-methyl-
6-oxopyrimidine-4-carboxamide (amine) (28 g, 98.22% purity)
in 82% yield.
EXPERIMENTAL SECTION
■
All materials were purchased from commercial suppliers. Unless
specified otherwise, all reagents and solvents were used as
supplied by manufacturers. Melting points were determined by
open air capillary with an X-6 melting point apparatus, Beijing
1
Tech instrument Co. Ltd., and are uncorrected. Varian H
NMR spectra (400 MHz) and 13CNMR spectra (100 MHz)
were recorded in CDCl3, DMSO-d6, and mass spectra were
determined on API-2000LCMS mass spectrometer, Applied
Biosciences.
Synthesis of Impurity C. To a suspension of raltegravir
(20 g, 0.045 mol) in acetonitrile (100 mL) was added 50%
aqueous KOH solution (10 mL), and the reaction mixture was
stirred at ambient temperature for 4−5 h. The solid was filtered
and purified using flash chromatography to give impurity C (10
g, 99.44% purity) in 50% yield.
1
Mp: 182−185 °C; H NMR (400 MHz, DMSO-d6) δ 9.70
(br s, 1H), 7.34−7.21 (m, 5H), 6.23 (s, 1H), 4.50 (d, 2H), 3.76
(s, 3H), 3.33 (br s, 2H), 1.53 (s, 6H); IR 3036, 1620, 1572,
1533, 1305 cm−1; MS (CI): calcd For C16H20N4O3 (M + H)/z:
317.3, found: (M + H)/z: 317.2.
The compound was purified using The Reveleris Flash
system on 330 g of prepacked 40 μ silica gel cartridge supplied
by Grace-USA, the mobile phase was CHCl3/methanol (9:1).
A slurry of oxadiazole K salt 134 (25.41 g, 0.153 mol) in
acetonitrile (198 mL) and DMF (0.3 mL) was cooled to 0−5
°C, and oxalyl chloride (18.12 g, 0.142 mol) was added,
keeping the internal temperature 0−5 °C under inert
atmosphere. The slurry was aged for 1−2 h at 0−5 °C. A
slurry of amine (22 g, 0.069 mol) and THF (594 mL) was
cooled to 0−5 °C, and NMM (16.87 g, 0.167 mol)was added.
The acetonitrile slurry of 14 was added slowly maintaining the
reaction temlperature 0−5 °C. The slurry was maintained for 1
h, and 25% aqueous ammonia (13.2 mL) was added to adjust
the pH 8−9. The reaction mixture was acidified to pH 2−3 by
2 HCl and evaporated under reduced pressure to give a syrupy
mass. A mixture of isopropanol (154 mL) and water (528 mL)
was added, and the precipitated solid was filtered and dried
under vacuum to yield impurity E (13 g, 98.56% purity) in 46%
yield.
1
Mp: 292−296 °C; H NMR (400 MHz, DMSO-d6) δ 12.19
(s, 1H), 10.38 (s, 1H), 9.85 (s, 1H), 9.41 (s, 1H), 9.06 (t, 1H),
7.36−7.4 (m, 2H), 7.15−7.19 (m, 2H), 4.51 (d, 2H, J = 6.32
Hz), 3.42 (s, 3H), 1.86 (s, 3H), 1.70 (s, 6H); IR 3273, 3007,
2947, 1728, 1670, 1354, 1223 cm−1; MS (CI): calcd For
C20H23FN6O6 (M + H)/z: 463.44, found: (M + H)/z: 463.1.
Synthesis of Impurity D. Compound 154 (15 g, 0.044
mol) was added to a solution of triethyl amine (13.5 g, 0.134
mol) in dichloromethane (150 mL) under nitrogen atmos-
phere. The reaction mixture was cooled to 0−5 °C, and ethyl
oxalyl chloride (9.3 g, 0.067 mol) was added dropwise. The
conversion was monitored by TLC (CHCl3/MeOH, 9:1); after
TLC showed disappearance of 15, 1 N HCl (100 mL) was
added, and the organic layer was separated. The organic layer
was washed with water, dried over sodium sulfate, and
evaporated under reduced pressure to give a syrupy mass,
which was treated with methanolic ammonia at 2−3 kg/cm2
ammonia pressure in an autoclave for 5−6 h. After completion
of the reaction, methanol was evaporated under reduced
pressure to give impurity D (17 g, 97.04% purity) in 93% yield
1
Mp: 132−135 °C; H NMR (400 MHz, DMSO-d6) δ 12.24
(s, 1H), 8.85 (s, 1H), 9.06 (t, 1H), 7.24−7.35 (m, 5H), 4.54 (d,
2H, J = 6.44 Hz), 3.49 (s, 3H), 2.56 (s, 3H), 1.75 (s, 6H); IR
3385, 3244, 3059 and 2940, 1688 and 1674, 1605 and 1555,
1360 cm−1; MS (CI): calcd For C20H22N6O5 (M + H)/z:
427.43, found: (M + H)/z: 427.2.
1
Mp: 243−247 °C; H NMR (400 MHz, DMSO-d6) δ 10.73
Synthesis of Impurity F. To a slurry of hydroxypyr-
imidone 6 (45 g, 0.119 mol) and methanol (315 mL) was
(br s, 1H), 9.08 (br s, 1H), 7.99 (br s, 1 H), 7.77 (s, 1H), 7.32−
1427
dx.doi.org/10.1021/op300077m | Org. Process Res. Dev. 2012, 16, 1422−1429

Organic Process Research & Development
Communication
added 4-methoxybenzyl amine (24.66 g, 0.179 mol) over 30
min. The resultant solution was aged at 60−65 °C for 4−5 h.
The conversion was monitored by TLC (CHCl3/MeOH, 9:1).
After TLC showed disappearance of 6, the reaction mixture was
cooled to 50−55 °C, and acetic acid (13.5 mL) was added.
Water (360 mL) was added, and the precipitated solid was
stirred at 25−35 °C for 30−45 min and filtered; the cake was
washed with a 1:1 mixture of methanol and water (2 × 50 mL)
and dried at 40 °C under reduced pressure to afford 17 (53.3 g,
98.02% purity) in 93% yield.
added, and the organic layer was separated. The organic layer
was washed with water, dried over sodium sulfate, and
evaporated under reduced pressure to give impurity G (17 g,
99.05% purity) in 89% yield.
1
Mp: 176−180 °C; H NMR (400 MHz, DMSO-d6) δ 12.18
(br s, 1H), 9.46 (s, 1H), 9.05 (t, 1H), 7.37−7.40 (m, 2H),
7.14−7.19 (m, 2H), 4.50 (d, 2H), 4.20−4.25 (q, 2H), 3.45 (s,
3H), 1.68 (s, 6H), 1.26 (t, 3H); IR 2999, 1686, 1636 and 1570,
1350, 1032 cm−1; MS (CI): calcd For C20H23FN4O6 (M + H)/
z: 435.42, found: (M + H)/z: 435.2.
Mp: 182−185 °C; 1H NMR (400 MHz, CDCl3) δ 11.87 (br
s, 2H), 7.70 (br s, 1H), 7.27−7.24 (m, 7H), 6.90−6.88 (d, 2H),
5.08 (s, 2H), 4.53 (d, 2H), 3.81 (s, 3H), 3.64 (s, 3H), 1.65 (s,
6H); IR 3362, 3331, 1712, 1676, 1535, 1251 cm−1; MS (CI):
calcd For C25H28N4O6 (M + H)/z: 481.5, found: (M + H)/z:
481.3.
Synthesis of Impurity H. To a suspension of 13 (10 g,
0.060 mol) and DMF (0.5 mL) in DCM (50 mL) was added
oxalyl chloride (9.1 g, 0.0723 mol) dropwise at 0−5 °C. N-
Methylmorpholine (8.52 g) was introduced while maintaining
the temperature between 0 and 5 °C. A solution of aminonitrile
2 (7.6 g, 0.0904 mol) in DCM (10 mL) was added dropwise at
0−5 °C. The reaction was monitored by TLC (CHCl3: MeOH,
9:1). After completion of the reaction, salts were filtered, and
the filtrate was completely evaporated under reduced pressure.
To the resulting residue was added isopropanol (10 mL), and
the mixture was again evaporated under reduced pressure. The
residue was triturated with isopropanol (20 mL) to precipitate
the solid which was filtered and dried under vacuum to give a
pale-yellow compound 18 (8.3 g, 71%). Purity by HPLC >
99%, mp 142−145 °C; 1H NMR (400 MHz, CDCl3) δ 7.34 (br
s, 1 H), 2.64 (s, 3 H), 1.84 (s, 6 H); 13C NMR (100 MHz,
CDCl3) δ 166.7, 158.4, 152.8, 119.9, 47.4, 27.1, 11.4; MS (CI):
Calcd for C8H10N4O2 m/z 195.1 (M + 1). Found: m/z 195.1.
[HPLC analysis; Phenomenex Gemini (4.6 mm × 150 mm)
3 μ, column; buffer: 0.01 M Na2HPO4 pH 3.0 by
orthophosphoric acid, eluent: (A) Buffer: ACN (85:15), (B)
ACN/MeOH/H2O (55:10:35); flow rate: 1.5 mL/min;
detector: 210 nm]
To a clean autoclave compound 17 (53 g, 0.110 mol) was
charged methanesulfonic acid (11.12 g, 0.115 mol) and
methanol (424 mL) in a mixture of 10% Pd/C (1.06 g). The
content was hydrogenated at 50−55 °C for 2−3 h at 2 kg/cm2
hydrogen pressure. The conversion was monitored by TLC
(CHCl3/MeOH, 9:1). After completion, the reaction mixture
was cooled to 25−35 °C, and hydrogen was replaced with
nitrogen. The Pd/C was removed by Celite filtration, and to
the filtrate was added 1 N NaOH to adjust the pH 7−8. The
precipitated amine was filtered and dried to yield N-(4-
methoxybenzyl)-2-(2-aminopropan-2-yl)-1,6-dihydro-5-hy-
droxy-1-methyl-6-oxopyrimidine-4-carboxamide (amine) (31 g,
98.62% purity) in 81% yield.
1
Mp: 190−194 °C; H NMR (400 MHz, DMSO-d6) δ 9.50
(br s, 1H), 7.24−7.22 (d, 2H), 6.89−6.87 (d, 2H), 6.16 (s, 1H),
4.43 (d, 2H), 3.84 (s, 3H), 3.72 (s, 3H), 3.32 (br s, 2H), 1.52
(s, 6H); IR 3489, 3271, 1693, 1539, 1516, 1309, 1244 cm−1;
MS (CI): calcd For C17H22N4O4 (M + H)/z: 347.3, found: (M
+ H)/z: 347.2.
A mixture of the amide 18 (8 g, 0.0412 mol) and isopropanol
(20 mL) was warmed to 60 °C, and then 50% (w/w) aqueous
solution of hydroxylamine (3.3 mL, 0.0495 mol) was added
over 25 min. The solution was heated at 60 °C for 30 min and
cooled to 0−5 °C; heptane (20 mL) was added over 30 min.
The resultant slurry was filtered, and the cake was washed with
heptane (8 mL). The residue was dried at 45−50 °C under
vacuum to afford compound 19 (9.2 g, 98%). Purity by HPLC
A slurry of oxadiazole K salt 13 (29.21 g, 0.176 mol) in
acetonitrile (252 mL) and DMF (1.12 mL) was cooled to 0−5
°C, and oxalyl chloride (20.82 g, 0.164 mol) added, keeping the
internal temperature 0−5 °C under inert atmosphere. The
slurry was aged for 1−2 h at 0−5 °C. A slurry of amine (28 g,
0.08 mol) and THF (756 mL) was cooled to 0−5 °C, and
NMM (19.39 g, 0.191 mol) was added. The acetonitrile slurry
of 14 was added slowly, maintaining the reaction temperature
0−5 °C. The slurry was maintained for 1 h, and 25% aqueous
ammonia (16.8 mL) was added to adjust the pH 8−9. The
reaction mass was acidified to pH 2−3 by 2 N HCl (126 mL).
The reaction mixture was distilled under reduced pressure, and
a mixture of isopropanol (196 mL) and water (755 mL) was
added. The precipitated solid was filtered and dried under
vacuum to yield impurity F (25 g, 98.14% purity) in 67% yield.
1
> 97%, mp 175−178 °C; H NMR (400 MHz, DMSO-d6) δ
9.36 (br s, 1H), 8.75 (s, 1H), 7.43 (br s, 1 H), 5.58 (s, 2 H),
2.55 (s, 3 H), 1.55 (s, 6 H); 13C NMR (100 MHz, DMSO-d6) δ
165.6, 158.9, 155.1, 151.7, 54.9, 24.9, 10.7; MS (CI): Calcd for
C8H13N5O3 m/z: 228.22 (M + H). Found: m/z: 228.1.
[HPLC analysis; Phenomenex Gemini (4.6 mm × 150 mm)
3 μ, column; buffer: 0.01 M Na2HPO4 pH 3.0 by
orthophosphoric acid, eluent: (A) buffer/ACN (85:15), B)
ACN/MeOH/H2O (55:10:35); flow rate: 1.5 mL/min;
detector: 210 nm]
1
Mp: 280−287 °C; H NMR (400 MHz, DMSO-d6) δ 11.56
(t, 1H), 9.79 (br s, 1H), 7.21 (d, 2H, J = 8.64), 6.86 (d, 2H, J =
8.64), 4.38 (d, 2H, J = 5.8), 3.72 (s, 3H), 3.39 (s, 3H), 2.56 (s,
3H), 1.70 (s, 6H); IR 2999, 1686, 1636 and 1570, 1350, 1032
cm−1; MS (CI): calcd For C21H24N6O6 (M − H)/z: 455.45,
found: (M − H)/z: 455.2.
Synthesis of Impurity G. Compound 15 (15 g, 0.044 mol)
was added to a solution of triethyl amine (13.5 g, 0.134 mol) in
dichloromethane (150 mL) under nitrogen atmosphere. The
reaction mixture was cooled to 0−5 °C, and ethyl oxalyl
chloride (9.3 g, 0.067 mol) was added dropwise. The
conversion was monitored by TLC (CHCl3/MeOH, 9:1),
after TLC showed disappearance of 15; 1 N HCl (100 mL) was
A slurry of the amidoxime derivative 19 (9 g, 0.0396 mol)
and methanol (63 mL) was cooled to 15−25 °C. DMAD (6.5
g, 0.0455 mol) was added over 30 min, maintaining the batch
temperature between 15 and 25 °C. The resultant solution was
aged at 25 °C for 2−3 h. The solution was concentrated under
reduced pressure and solvent switched, by feed and bleed, at
constant volume to xylenes (27 mL). The final batch
temperature was maintained below 70 °C during the solvent
switch. The mixture was heated to 125 °C for 2 h, raised to 135
°C for 6 h, and cooled to 60 °C. Methanol (7.7 mL) was added
and stirred for 1 h, and the MTBE (28 mL) was introduced.
The solution was cooled to 0−5 °C for 2 h, filtered, washed
1428
dx.doi.org/10.1021/op300077m | Org. Process Res. Dev. 2012, 16, 1422−1429

Organic Process Research & Development
Communication
with 9:1 of MTBE/methanol (2 × 16 mL), and dried to afford
cm−1; MS (CI): calcd For C34H36F2N8O8 (M + H)/z: 723.71,
found: (M + H)/z: 723.3
1
20 (9 g, 66%). Purity by HPLC >97%, mp 234−237 °C; H
NMR (400 MHz, DMSO-d6) δ 12.74 (s, 1 H), 10.35 (s, 1 H),
9.12 (s, 1 H), 3.81 (s, 3 H), 2.58 (s, 3 H), 1.59 (s, 6 H); 13C
NMR (100 MHz, DMSO-d6) δ 166.6, 166.1, 160.2, 159.2,
153.2, 152.9, 145.6, 128.3, 56.6, 52.9, 26.2, 11.3; MS (CI):
Calcd for C13H15N5O6 m/z: 338.29 (M + H). Found: m/z:
338.2.
ASSOCIATED CONTENT
■
S
* Supporting Information
Additional characterization data of 15−21 and impurities C−I.
This material is available free of charge via the Internet at
http://pubs.acs.org.
[HPLC analysis: Phenomenex Gemini (4.6 mm × 150 mm)
3 μ, column; buffer: 0.01 M Na2HPO4 pH 3.0 by
orthophosphoric acid, eluent: (A) buffer/ACN (85:15), (B)
ACN/MeOH/H2O (55:10:35); flow rate: 1.5 mL/min;
detector: 210 nm]
AUTHOR INFORMATION
Corresponding Author
*Fax: +91-20-39821445. E-mail: Swapnil.Sonawane@emcure.
co.in.
■
To a mixture of hydroxypyrimidinone 20 (8.5 g, 0.025 mol)
and DMSO (68 mL) was added a solution of Mg(OMe)2 (4.35
g, 0.0504 mol) in MeOH (57 mL). Excess MeOH was then
evaporated under vacuum at 40−45 °C. After cooling to
ambient temperature, MeI (14.24 g, 0.1 mol) was added
dropwise, and the mixture was stirred at 20−25 °C for 2 h, then
at 60 5 °C for 5 h and cooled to 20−25 °C. HCl (2 M, 85
mL) was added followed by 5 wt % sodium bisulfite (10 mL).
Water (170 mL) was added over 40 min, and the slurry was
stirred for 40 min; the resultant slurry cooled to 0−5 °C. The
crystalline product was collected by filtration and washed with
water (85 L). The product was dried under vacuum to give 21
(8 g, 98.5% purity) in 91% yield.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
We thank Mr. Samit S. Mehta, Emcure Pharmaceuticals Ltd.,
for generous support and constant encouragement. Finally, it is
a pleasure to acknowledge the reviewers and editor for valuable
suggestions.
REFERENCES
■
(1) Pace, P.; Spieser, S.; Summa, V. Bioorg. Med. Chem. Lett. 2008, 18,
3865−3869.
(2) Croxtall, D. J.; Williamson, K, A.; Perry, C. M. Drugs 2008, 68,
131−138.
To a slurry of hydroxypyrimidone 21 (7 g, 0.019 mol) and
methanol (50 mL) was added 4-fluorobenzyl amine (3.73 g,
0.029 mol) over 30 min. The resultant solution was aged at
60−65 °C for 4−5 h. The conversion was monitored by TLC
(CHCl3/MeOH, 9:1). After TLC showed the disappearance of
21, the reaction mixture was cooled to 50−55 °C, and acetic
acid (2 mL) was added. Water (77 mL) was added, and the
precipitated solid was stirred at 25−35 °C for 30 to 45 min and
filtered, the cake was washed with 1:1 of methanol/water (2 ×
50 mL) and dried at 40 °C under reduced pressure to give
impurity H (7 g, in 83% yield). Impurity H (7 g, 0.015 mol)
was suspended in methanol (70 mL), and potassium tert-
butoxide (1.76 g, 0.015 mol) was added at 0−5 °C; the reaction
mixture was stirred for 2 h, filtered, and dried to give pure
potassium salt of impurity H (7 g, 99.12% purity) 92% yield.
(3) Summa, V.; Petrocchi, A.; Bonelli, F.; Crescenzi, B.; Donghi, M.;
Ferrara, M.; Fiore, F.; Gardelli, C.; Gonzalez Paz, O.; Hazuda, D. J.;
Jones, P.; Kinzel, O.; Laufer, R.; Monteagudo, E.; Muraglia, E.; Nizi, E.;
Orvieto, F.; Pace, P.; Pescatore, G.; Scarpelli, R.; Stillmock, K.;
Witmer, M. V.; Rowley, M. J. J. Med. Chem. 2008, 51, 5843−5855.
(4) (a) Crescenzi, Benedetta, C.; Cristina, G.; Ester, M.; Federica, O.;
Paola, P.; Giovanna, P.; Alessia, P.; Marco, P.; Michael, R.; Rita, S.;
Summa, V.; Emanuela, N. U. S. Patent 7,169,780. 2007. (b) Belyk, K.
M.; Morrison, H. G.; Jones, P.; Summa, V.; Cooper, V. B.; Mahajan, A.
J.; Kumke, D. J.; Tung, H.; Wai, L.; Pruzinsky, V.; Pye, P.; Angelaud,
R.; Mancheno, D.U. S. Patent 20060122205. 2006. (c) Humphrey, G.
R.; Pye, P. J.; Zhong, Y. Li.; Angelaud, R.; Askin, D.; Belyk, K. M.;
Maligres, P. E.; Mancheno, D. E.; Miller, R. A.; Reamer, R. A.;
Weissman, S. A. Org. Process Res. Dev. 2011, 15, 73−83.
(5) ICH Q3A Impurities in New Drug Substances, R2; International
Conference on Harmonisation of Technical Requirements for
Registration of Pharmaceuticals for Human Use (ICH): Geneva,
Switzerland, October 2006; http://www.ich.org/fileadmin/Public_
Web_Site/ICH_Products/Guidelines/Quality/Q3A_R2/Step4/
Q3A_R2__Guideline.pdf.
(6) Quality Guidelines; International Conference on Harmonisation
of Technical Requirements for Registration of Pharmaceuticals for
Human Use (ICH): Geneva, Switzerland, http://www.ich.org/
products/guidelines/quality/article/quality-guidelines.html.
(7) Novak, T. J.; Grinberg, N.; Hartman, B.; Marcinko, S.; DiMichele,
L.; Mao, B. J. Pharm. Biomed. Anal. 2010, 51, 78−83.
(8) Benedetta, C.; Cristina, G.; Ester, M.; Emanuela, N.; Federica, O;
Paola, P.; Giovanna, P.; Alessia, P.; Marco, P.; Michael, R.; Rita, S.;
Summa V. World Patent WO/03/035077, 2003.
1
Mp: 277−281 °C; H NMR (400 MHz, DMSO-d) δ 12.19
(br s, 1H), 9.78 (br s, 1H), 7.32−7.36 (m, 2H), 7.10−7.14 (m,
2H), 4.48 (d, 2H), 3.82 (s, 3H), 2.59 (s, 3H), 1.70 (s, 6H); IR
3323, 2992 and 2943, 1703, 1643 and 1556, 1348, 1223 cm−1;
MS (CI): calcd For C20H20FKN6O5 (M − k)/z: 445.1, (M +
H)/z: 483.51, found: (M − k)/z: 445.1, (M + H)/z: 483.1
Synthesis of Impurity I. To a solution of oxalyl chloride
(1.9 g, 0.014 mol) and N-methylmorpholine (1.7 g, 0.016 mol)
in acetonitrile (100 mL) was added 15 (10 g, 0.029 mol) lot-
wise over 30 min. The resultant solution was aged at ambient
temperature for 8−10 h. The conversion was monitored by
TLC (CHCl3/MeOH, 9:1). After TLC showed the disappear-
ance of 15, the reaction mixture was cooled to 0−5 °C, and
water (100 mL) was added; the precipitated solid was stirred at
25−35 °C for 30−45 min and filtered; the cake was washed
with acetonitrile (2 × 50 mL) and dried at 40 °C under
reduced pressure to give impurity I (15 g, 99.48% purity) in
72% yield.
1
Mp: 315−318 °C; H NMR (400 MHz, DMSO-d6) δ 12.11
(br s, 2H), 11.04 (br s, 2H), 9.21 (s, 2H), 7.35−7.31 (m, 4H),
7.14−7.10 (m, 4H), 4.45 (d, 4H), 3.35 (s, 6H), 1.64 (s, 12H);
IR 3323, 2992 and 2943, 1703, 1643 and 1556, 1348, 1223
1429
dx.doi.org/10.1021/op300077m | Org. Process Res. Dev. 2012, 16, 1422−1429