Multi-gram synthesis of a nucleotide-competing reverse transcriptase inhibitor

Article abstract of DOI:10.1016/j.tetlet.2013.01.129

An efficient multi-gram synthesis of a nucleotide-competing reverse transcriptase inhibitor is reported. The synthesis features a chiral auxiliary-assisted alcohol resolution, a Mitsunobu reaction involving a carbamate, followed by a lithium-iodide exchange/Weinreb ketone synthesis tandem. These chemical transformations were optimized in order to increase the yield of the synthesis. The route is concluded by a late-stage palladium-catalyzed cyanation followed by a pyrimidine-2-one ring formation.

Full text of DOI:10.1016/j.tetlet.2013.01.129

Tetrahedron Letters 54 (2013) 2303–2307
Contents lists available at SciVerse ScienceDirect
Tetrahedron Letters
journal homepage: www.elsevier.com/locate/tetlet
Multi-gram synthesis of a nucleotide-competing reverse transcriptase inhibitor
⇑
Martin Duplessis , Louis Morency , Clint James, Joannie Minville, Patrick Deroy, Sébastien Morin,
Bounkham Thavonekham, Martin Tremblay, Ted Halmos, Bruno Simoneau, Yves Bousquet, Claudio Sturino
Boehringer Ingelheim (Canada) Ltd, Research and Development, 2100 Cunard Street, Laval (QC), Canada H7S 2G5
a r t i c l e i n f o
a b s t r a c t
Article history:
An efficient multi-gram synthesis of a nucleotide-competing reverse transcriptase inhibitor is reported.
The synthesis features a chiral auxiliary-assisted alcohol resolution, a Mitsunobu reaction involving a car-
bamate, followed by a lithium-iodide exchange/Weinreb ketone synthesis tandem. These chemical trans-
formations were optimized in order to increase the yield of the synthesis. The route is concluded by a
late-stage palladium-catalyzed cyanation followed by a pyrimidine-2-one ring formation.
Ó 2013 Published by Elsevier Ltd.
Received 17 December 2012
Revised 28 January 2013
Accepted 29 January 2013
Available online 8 February 2013
Keywords:
HIV
Reverse transcriptase inhibitor
NcRTI
Multi-gram synthesis
Mitsunobu reaction
1. Introduction
excellent antiviral potency, favorable cross-resistance profile, and
good pharmacokinetic profile. Based on its favorable in vitro and
According to the World Health Organization, an estimated 34
million people were infected with HIV worldwide in 2010, mostly
in Sub-Saharan Africa.1 Fortunately, progress in highly active anti-
retroviral therapy over the past 20 years has reduced the mortality
caused by the AIDS epidemic.2 While a vaccine cure aimed at the
eradication of the disease is the most desirable goal, supplement-
ing the antiviral arsenal with novel drugs aiming at highly vali-
dated viral targets is also needed. In particular, inhibition the HIV
reverse transcriptase is a key component in reducing viral load
and suppressing the infection.3
Our research group focused on a new class of reverse transcrip-
tase inhibitors, nucleotide-competitive reverse transcriptase inhib-
itors (NcRTIs).4 These non-nucleosidic molecules bind at, or very
close to, the active site of the enzyme,5 thereby competing with
incoming deoxyribonucleotide triphosphates (dNTP). This mode
of inhibition prevents elongation of the proviral DNA strand, but
does not lead to chain termination. As such, NcRTIs could comple-
ment currently available NRTI and NNRTI regimens, especially in
the context of the emergence of drug-resistant HIV strains.
Following a high-throughput screening campaign, a new chem-
otype exhibiting NcRTI mechanism was discovered. Extensive lead
optimization led to the discovery of inhibitor 1 that demonstrated
in vivo profiles, this molecule was selected as a candidate for tox-
icological and pharmacological studies, requiring a multi-gram
synthesis of the active pharmaceutical in high purity. However,
the highly functionalized polycyclic molecule 1 posed several syn-
thetic challenges (Fig. 1). Starting from azabenzofuran 2, the fol-
lowing chemical steps were required: mono-coupling of pyrazole
at the C-8 position, cyanation at position C-7, introduction of opti-
cally pure 2-ethyltetrahydropyran-4-yl at N-1, without erosion of
stereochemical integrity, and introduction of an N-methylpyrazole
at the C-4 position. The present communication describes the opti-
mization of these key transformations, leading to the multi-gram
synthesis of 1.
2. Synthesis of optically pure tetrahydropyran 4
In order to develop an efficient synthesis of tetrahydropyran
alcohol 4, we decided to embark on a racemic synthesis followed
by a chiral auxiliary-assisted separation. Strict timelines, combined
with synthetic challenges and the need for a high level of optical
purity (>98.5% ee) precluded the pursuit of a stereoselective
synthesis.
The epimer of alcohol 4 (epi-4) was prepared using a Prins reac-
tion between propionaldehyde and 3-buten-1-ol, according to a
previously reported procedure, in good yield (Scheme 1).6 The cis
diastereomer was obtained with 95:5 selectivity; the undesired
trans isomer could be removed by chromatography during the
enantioresolution process (vide infra). The stereochemistry of
epi-4 at the alcohol center had to be inverted to obtain the desired
⇑
Corresponding authors. Tel.: +1 450 682 4640.
E-mail addresses: martin.duplessis@boehringer-ingelheim.com (M. Duplessis),
Bruno.simoneau@boehringer-ingelheim.com (B. Simoneau).
This article is dedicated to the memory of Dr. Louis Morency (1979–2012), fellow
scientist, colleague and friend.
0040-4039/$ - see front matter Ó 2013 Published by Elsevier Ltd.
http://dx.doi.org/10.1016/j.tetlet.2013.01.129

2304
M. Duplessis et al. / Tetrahedron Letters 54 (2013) 2303–2307
O
Chiral THP -
Resolution
N-1 - Mitsunobu
N
N
OH
4
OEt
B(OR)2
O
C-8 - Suzuki
C-4 - Carbanion
3
O
addition
O
NH
N
N
1
N
Cl
Cl
N
O
N
4
8
7
N
O
OEt
X
O
NC
N
2
N
N
C-7 - cyanation
N
1
5
Figure 1. Retrosynthetic approach to compound 1.
stereochemical outcome in the subsequent Mitsunobu reaction
involving carbamate 2 (vide infra). To this end, a Mitsunobu reac-
tion with isonicotinic acid led to complete inversion of stereo-
O
O
Resolution
chemistry.
A
simple two-step acid–base extraction was
performed, affording compound 5. Isonicotinate ester 5 was hydro-
lyzed under basic conditions followed by vacuum distillation to af-
ford hydroxypyran rac-4 as a racemic mixture.
OH
OH
( )-4
4
O
We planned on obtaining optically pure tetrahydropyran 4
through chromatographic separation after coupling of a chiral aux-
iliary. Several chiral auxiliaries (amino acids, Mosher’s ester, Trost’s
ester) were screened, but no practical level of separation of the
enantiomeric mixture could be achieved. Finally, Harada’s chiral
auxiliary 6 was selected for its demonstrated enantioresolution
capability7 (Scheme 2). Due to the fact that the element imparting
chirality is far removed from the chiral auxiliary, the separation of
O
OH
iii)
i), ii)
6
O
O
7a and 7b proved challenging using standard 40–63 lm silica.
O
However, the diastereomeric mixture comprising 7a and 7b was
successfully separated with the help of medium-pressure chroma-
tography, using high performance silica cartridges.8
O
O
O
+
O
O
After cleavage of the ester under basic conditions, an acid–base
separation was performed, at which point chiral auxiliary 6 was
7b
7a
Scheme 2. Enantioresolution of chiral tetrahydropyran 4. Reagents and conditions:
(i) DCC (1.1 equiv), DMAP (0.5 equiv), CSA (0.1 equiv), DCM; (ii) separation of
diastereomers (highly spherical silica), 25:75 EtOAc:Hex; (iii) NaOMe, 70 °C, then
water, 80 °C; 60–70% yield.
O
O
O
O
O
i)
ii)
+
OH
H
isolated and recycled, with recovery greater than 80%. The optically
pure tetrahydropyran was obtained in good yield, typically be-
tween 60% and 70%.
OH
epi-4
N
5
The absolute configuration of 4 was determined by comparison
of the 1H NMR spectral data of compounds 7a and 7b. Chiral aux-
iliary 6 is known to induce strong, unambiguous anisotropic shifts.
iii)
In the case of 7a and 7b,
Dd values were consistent with (2S,4R)
O
and (2R,4S) configurations, respectively.7,9
3. Study of the Mitsunobu reaction
OH
rac-4
The Mitsunobu reaction represents an efficient way of creating
carbon–heteroatom bonds with complete preservation of optical
purity.10 This transformation was clearly the procedure of choice
Scheme 1. Synthesis of rac-4. Reagents and conditions: (i) H2SO4; 80%; (ii)
isonicotinic acid, triphenylphoshine, DIAD, THF; (iii) NaOH, MeOH; >90% (two-
steps).

M. Duplessis et al. / Tetrahedron Letters 54 (2013) 2303–2307
2305
Table 1
for the introduction of tetrahydropyran 4 on the core of the mole-
Mitsunobu reaction optimization
cule. However, in previous studies conducted on different interme-
diates, we realized that this key connection presented a significant
challenge (Scheme 3). Due to the weakly acidic nature of the carba-
mate proton, the Mitsunobu reaction was slow and subject to side
reactions.
Subtle electronic effects had a critical impact on the outcome of
the Mitsunobu reaction (Scheme 3). In the case of substrate 8a
bearing an ester at the 2-position of the azabenzofuran, the reac-
tion proceeded quite efficiently over 16 h. In contrast, Weinreb
amide 8b reacted at a much lower rate, reaching less than 40% con-
version. Finally, the electron-withdrawing ketone 8c did not show
any reactivity under these conditions. 1H NMR analysis of the car-
bamate proton of 8a showed a chemical shift of 9.47 ppm, while
the poor susbtrates 8b and 8c showed chemical shifts at
9.76 ppm and 10.06 ppm, respectively.11 This observation was
indicative of the relevance of electronic effects in the Mitsunobu
reaction.
OEt
O
N
O
O
4, TPP
NH
DIAD, THF
N
Br
O
N
OEt
Br
O
O
O
N
O
N
O
8b
9b
Entry
Tetrahydropyran (equiv)
% Conversion after 17 h
1
2
3
4
5
1.0
1.2
1.5
1.8
2.0
32
40
28
28
16
Due to the slow rate of the Mitsunobu reaction involving carba-
mates of type 8, tetrahydropyran alcohol 4 was subject to a non-
constructive reaction pathway, leading to low levels of conversion.
In order to overcome that difficulty, the stoichiometry of the alco-
hol was adjusted for problematic substrate 8b. It was found that
1.2 equiv of tetrahydropyran alcohol 4 yielded a 40% level of con-
version (Table 1, entry 2), representing the optimal conditions after
significant efforts. While the remaining starting material 8b could
be isolated and re-submitted to the reaction conditions, it was
realized that the only path forward to a multi-gram synthesis of
inhibitor 1 was to use a substrate of type 8a, bearing an ester at
the 2-position. The key findings from the optimization study
(Scheme 3 and Table 2) were used to allow an efficient synthesis
of inhibitor 1.
4. Introduction of C-4 pyrazole
Initially, we planned to introduce the C-4 pyrazole by the addi-
tion of an organolithium reagent to an ester. Early on, we realized
that the preformation of the lithiated pyrazole species was not a
viable option, due to the poor outcome of the reaction. The pyra-
zole anion was believed to be unstable, hence, the order of addition
O
OEt
O
O
NH
N
OEt
Br
O
>80% conversion
Br
O
O
OEt
N
O
OEt
N
O
8a
9a
OEt
NH
O
O
N
OEt
O
Br
<40% conversion
O
N
Br
O
O
N
O
N
O
N
O
9b
8b
OEt
NH
O
O
N
OEt
O
No reaction
Br
Br
O
O
O
N
N
N
N
N
N
9c
8c
Scheme 3. Impact of electronic effects on the conversion level of Mitsunobu reaction. Reagents and conditions: Tetrahydropyran 4, triphenylphosphine, DIAD (dropwise
addition), THF, 0 °C.

2306
M. Duplessis et al. / Tetrahedron Letters 54 (2013) 2303–2307
Table 2
the reproducibility of the reaction. In addition, chromatographic
purification of 12a was complicated by the presence of this by-
product.
Based on experiments previously conducted on similar scaf-
folds, we expected that the use of a Weinreb amide would remedy
the bis-addition issue.13 It had been demonstrated that the addi-
tion of an organolithium species to a Weinreb amide led to the for-
Optimization of the C-4 addition
N
N
(4 equiv.)
O
N
O
N
I
11
O
O
N
n-BuLi (2 equiv.)
N
N
N
mation of
a stable chelate which did not undergo further
OEt
O
OEt
O
N
N
reaction.14 Unfortunately, the reaction of the lithiated pyrazole
on 10b provided only a marginal increase in yield (entry 3). It
was inferred from mass spectrometry data that the electrophilic
nitrile group at the C-7 position was interfering with the reaction.
The order of the synthetic steps was thus reordered: the pyrazole
lithiation-addition was performed on the cyanation precursor,
chloride 10c. Substrate 10c reacted cleanly with the lithiated pyr-
azole to afford 12b with a 74% yield (entry 4). Subsequently, it was
found that tert-butyllithium provided superior reproducibility (en-
try 5) and provided the best path forward to multi-gram scale up.
THF, -78oC
O
R
O
X
X
N
N
10a: X = CN, R = OEt
10b: X = CN, R = NMeOMe
10c: X = Cl, R = NMeOMe
12a: X = CN
12b: X = Cl
Entry
R
X
Yielda (%)
1b
2
3
OEt
OEt
CN
CN
CN
Cl
decomp.
60
NMeOMe
NMeOMe
NMeOMe
63
74
74
4
5c
Cl
5. Multi-gram synthesis of advanced lead 1
a
b
c
Isolated yields after silica gel chromatography.
Reaction performed with 4-bromo-1-methylpyrazole.
Reaction performed with tert-BuLi (4 equiv).
A multi-gram synthesis of the advanced lead 1 was undertaken
to support toxicological studies (Scheme 4). Extensive optimization
of the key chemical steps performed previously allowed for an effi-
cient process (vide supra).
Starting from 70 g of dichloro azabenzofuran 2,13 a Suzuki–
Miyaura cross-coupling with boronate 13 was completed. The
reaction was performed in three separate batches to afford 55 g
of intermediate 14. Coupling of the pyrazole proceeded exclusively
at the pyridine-activated C-8 position. The Mitsunobu reaction to
introduce tetrahydropyran 4 was performed in two batches of
27.5 g. The transformation proceeded smoothly despite the limited
solubility of 14 in THF. Purification of 15 presented a significant
bottleneck, since four separate chromatographic separation runs
had to be undertaken. A total of 58 g of 15 with >90% purity were
obtained.
of the reagents was adjusted accordingly, with the lithiating agent
(n-butyllithium) being added dropwise to a mixture of ethyl ester
10a and halopyrazole 1112. The lithiated species underwent nuleo-
philic addition as it was formed, minimizing the problem of
decomposition and affording a more efficient transformation.
In the case of compound 10a, the bromopyrazole proved inef-
fective in the reaction, leading to extensive decomposition (Table 2,
entry 1). Previous work showed that the use of the iodopyrazole,
which undergoes lithium–halogen exchange more readily than
bromide, increased the yield significantly. When iodopyrazole 11
was used, product 12a was formed in 60% yield. However, the
over-addition by-product was also formed, in up to 25% (Entry
2). This side reaction raised concerns about the scalability and
Ester 15 was satisfactorily converted into Weinreb amide 10c
using the preformed lithium anion of N,O-dimethylhydroxylamine.
O
N
O
N
OEt
NH
OEt
NH
O
O
O
O
N
N
N
N
N
OEt
O
OEt
O
i)
ii)
iii)
N
N
N
N
N
Cl
Cl
O
O
O
O
O
O
OEt
OEt
OEt
N O
Cl
Cl
Cl
2
14
15
10c
N
N
O
O
O
O
N
B
O
O
O
O
N
N
iv)
v)
vi)
N
N
N
OEt
O
N
OEt
O
N
N
N
N
13
N
N
N
I
O
O
NC
Cl
O
NC
N
N
N
N
N
N
N
12b
12a
1
11
Scheme 4. Synthesis of 1. Reagents and conditions: (i) Boronate 13 (1.6 equiv), cesium fluoride (0.25 equiv), sodium bicarbonate (2.4 equiv), Pd(OAc)2. (7 mol %),
triphenylphosphine (15 mol %), dioxane:water (4:1), 90 °C, 16 h; 69–83%; (ii) tetrahydropyran 4, triphenylphosphine, DIAD (dropwise add.), THF, 0 °C, 16 h; 67–73%; (iii) N,O-
dimethylhydroxylamine hydrochloride (3.5 equiv), n-butyllithium (2.5 M in hexanes, 7 equiv), THF, À60 °C, 15 min, then 15 in THF added, À60 °C, 30 min; 83%; (iv) iodide 11
(5 equiv), tert-butyllithium (1.7 M in pentane, 4 equiv), THF, À78 °C, 45 min; 74%; (v) Zn(CN)2 (4.6 equiv), Pd(P(t-Bu)3)2 (20 mol %) DMA, microwave 15 °C, 30 min; 74%; (vi)
ammonia (gas), ammonium acetate (60 equiv), NMP, 130 °C, 4 h; 70%.

M. Duplessis et al. / Tetrahedron Letters 54 (2013) 2303–2307
2307
Weinreb amide 10c was then submitted to the optimized lithium–
halogen exchange/addition protocol to afford ketone 12b. The larg-
est scale on which this protocol was performed was 8 g of 10c.
Considering the necessity for cryogenic conditions and the high
reactivity of the organolithium species, we deemed this transfor-
mation too delicate to risk a further increase in reaction scale. After
combination and chromatographic purification, 37 g of compound
12b was obtained.
We would also like to thank Norman Aubry for his efforts on the
1H NMR assignment of compounds 7a and 7b. We are also grateful
to Steven Laplante, Colette Boucher, Michael Little, Pascal Turcotte,
Sylvain Bordeleau, and Angelo Filosa for NMR and analytical
support.
Supplementary data
The palladium-catalyzed cyanation of the aryl chloride was
done using Zn(CN)2 as the cyanide source. The bis(tri-tert-butyl-
phoshine)palladium catalyst was used because of its high effective-
ness with aryl chlorides and its ease of handling on a large scale.15
The reaction required microwave heating to proceed efficiently. It
was absolutely necessary to achieve near 100% conversion for the
cyanation reaction, since the separation of 12b and 12a was te-
dious. After chromatographic purification, the material could be
further purified by trituration, to achieve a yield of 27 g of 12a with
>97% purity.
The final cyclization step to form the pyrimidine-2-one ring was
performed using ammonium acetate in N-methylpyrrolidinone at
130 °C. The reaction mixture was degassed thoroughly with nitro-
gen prior to heating. Ammonium acetate is known to decompose to
acetamide and water at high temperatures,16 so a relatively short
reaction time was desirable. Ammonia gas was bubbled into the
reaction mixture to ensure high concentrations of ammonia. After
reaction completion, the compound was precipitated by addition
of water to the reaction mixture and filtered. Trituration of the
crude material with diethyl ether followed by treatment with acti-
vated charcoal (SX-Ultra, 100% w/w) afforded inhibitor 1. We pro-
vided 17.4 g of purified (>98% homogeneity by HPLC) active
pharmaceutical ingredient for pre-development pharmacological
and toxicological profiling. The overall yield of 1 starting from car-
bamate 2 was 17% over six-steps, for an average yield of 74% per
synthetic step.
Supplementary data (detailed experimental procedures and
spectroscopic data for compounds 7a, 7b, 14 and 1) associated
with this article can be found, in the online version, at http://
dx.doi.org/10.1016/j.tetlet.2013.01.129. These data include MOL
files and InChiKeys of the most important compounds described
in this article.
References and notes
1. Progress Report 2011: Global HIV/AIDS Response; World Health Organization;
www.who.int/hiv/pub/progress_report2011/global_facts/en/index.html.
2. Palella, F. J., Jr.; Delaney, K. M.; Moorman, A. C., et al N. Engl. J. Med. 1998, 338,
853.
3. (a) Pau, A. K. Reverse Transcriptase Inhibitors in HIV/AIDS Therapy; Humana
Press: Totowa, NJ, 2006; (b) Waters, L.; John, L.; Nelson, M. Int. J. Clin. Pract.
2007, 61, 105.
4. (a) Jochmans, D. Virus Res. 2008, 134, 171; (b) Jochmans, D.; Wigerinck, P. Int.
Patent Appl. WO2007088214, 2007.; (c) DeRoy, P.; Duplessis, M.; Edwards, P. J.;
Faucher, A.-M.; Halmos, T.; James, C.; Lacoste, J.-É.; Malenfant, É.; Minville, J.;
Morency, L.; Morin, S.; Sturino, C.; Tremblay, M.; Yoakim, C.; U.S. Patent Appl.
US20100261714, 2010.
5. Freisz, S.; Bec, G.; Radi, M.; Wolff, P.; Crespan, E.; Angeli, L.; Dumas, P.; Maga,
G.; Botta, M.; Ennifar, E. Angew. Chem., Int. Ed. 2010, 49, 1805.
6. Hanschke, E. Chem. Ber. 1955, 88, 1053; For a review on the Prins reaction:
Pastor, I. M.; Yus, M. Curr. Org. Chem. 2007, 11, 925.
7. (a) Taji, H.; Watanabe, M.; Harada, N.; Naoki, H.; Ueda, Y. Org. Lett. 2002, 4,
2699; (b) Harada, N.; Watanabe, M.; Kuwahara, S.; Sugio, A.; Kasai, Y.;
Ichikawa, A. Tetrahedron: Asymmetry 2000, 11, 1249; (c) Ichikawa, A.; Hiradate,
S.; Sugio, A.; Kuwahara, S.; Watanabe, M.; Harada, N. Tetrahedron: Asymmetry
1999, 10, 4075.
8. A Combiflash RfÒ (Teledyne IscoÒ) system with an Isco RediSep Rf GoldÒ
column was used for the separation of 7a and 7b.
In conclusion, optically pure tetrahydropyran 4 was obtained
through an enantioresolution using an effective chiral auxiliary.
This pyran alcohol was introduced into carbamate 2 through a
Mitsunobu reaction. A lithium–halogen exchange/Weinreb ketone
synthesis protocol was optimized to allow the efficient introduc-
tion of a pyrazole at the C-4 position. All these advances allowed
for an efficient multi-gram synthesis of advanced lead 1, which
was used for pre-development pharmacological and toxicological
profiling.
9. See the Supplementary data section for
a detailed analysis of the
stereochemical assignment of 7a and 7b.
10. Reviews: (a) Swamy, K. C. K.; Kumar, N. N. B.; Balaraman, E.; Pavan Kumar, K. V.
P. Chem. Rev. 2009, 109, 2551; (b) Golantsov, N. E.; Karchava, A. V.; Yurovskaya,
M. A. Chem. Heterocycl. Compd. 2008, 44, 263.
11. 1H NMR spectra was collected on
a 400 MHz instrument. Samples were
dissolved in DMSO-d6. Data on compound 8a were collected on a set of three
samples with different levels of dilution to insure reproducibility of the
carbamate proton chemical shift.
12. Compound 11 was synthesized by WuXi according to known procedures. See
Supplementary data section for description of the synthesis.
13. Nahm, S.; Weinreb, S. M. Tetrahedron Lett. 1981, 22, 3815.
14. Qu, B.; Collum, D. B. J. Org. Chem. 2006, 71, 7117.
15. (a) Dai, C.; Fu, G. C. J. Am. Chem. Soc. 2001, 123, 2719; (b) Littke, A.; Soumeillant,
M., ; Kaltenbach, R. F., III; Cherney, R. J.; Tarby, C. M.; Kiau, S. Org. Lett. 2007, 9,
1711.
Acknowledgments
The authors would like to acknowledge the work of the chem-
istry team at WuXi for their work in the scale-up of carbamate 2.
16. Olszak-Humienik, M. Thermochemica. Acta 2001, 107.