Synthesis of a nucleoside phosphoramidate prodrug inhibitor of HCV NS5B polymerase: Phenylboronate as a transient protecting group

Article abstract of DOI:10.1021/op500042u

A synthetic process for 2′-C-methylcytidine-5′-[2-[(3-hydroxy- 2,2-dimethyl-1-oxopropyl)thio]ethyl-N-benzylphosphoramidate], a nucleotide prodrug inhibitor of hepatitis C virus NS5B polymerase, is described. The route developed was demonstrated on 100 g scale and featured the key application of phenylboronic acid as an effective transient means to protect the 2′,3′-hydroxyls of 2′-C-methylcytidine. This synthetic methodology resulted in a reduction in the number of isolations from five to two and an increase in the overall yield by 50% relative to the original unscalable discovery route. The synthesis and characterization of 2′-C- methylcytidine-2′,3′-O-phenylboronate is also provided.

Full text of DOI:10.1021/op500042u

Article
pubs.acs.org/OPRD
Synthesis of a Nucleoside Phosphoramidate Prodrug Inhibitor of
HCV NS5B Polymerase: Phenylboronate as a Transient Protecting
Group
Benjamin A. Mayes,* Jeevanandam Arumugasamy, Erkan Baloglu, David Bauer, Alan Becker,
Narayan Chaudhuri, G. Mark Latham, Jie Li, Steve Mathieu, F. Patrick McGarry, Elodie Rosinovsky,
Alistair Stewart, Christophe Trochet, Jingyang Wang, and Adel Moussa
Idenix Pharmaceuticals Inc., 320 Bent Street, Cambridge, Massachusetts 02141, United States
ABSTRACT: A synthetic process for 2′-C-methylcytidine-5′-[2-[(3-hydroxy-2,2-dimethyl-1-oxopropyl)thio]ethyl-N-benzyl-
phosphoramidate], a nucleotide prodrug inhibitor of hepatitis C virus NS5B polymerase, is described. The route developed was
demonstrated on 100 g scale and featured the key application of phenylboronic acid as an effective transient means to protect the
2′,3′-hydroxyls of 2′-C-methylcytidine. This synthetic methodology resulted in a reduction in the number of isolations from five
to two and an increase in the overall yield by 50% relative to the original unscalable discovery route. The synthesis and
characterization of 2′-C-methylcytidine-2′,3′-O-phenylboronate is also provided.
INTRODUCTION
■
Hepatitis C virus (HCV) is blood-borne pathogen chronically
infecting an estimated 3% of the global population, for which a
vaccine is not available.1 Treatment with boceprevir, telaprevir,
or simeprevir, the first direct acting antivirals (DAAs) to be
introduced, in combination with PEGylated interferon-α (IFN-
α) and ribavirin is limited to those patients presenting with
HCV genotype 1 infection.2 Greater potency, broader genotype
coverage, higher barrier to resistance, and improved safety
profile are clinical requirements which have driven efforts to
develop novel and more effective DAAs.3 Beyond these HCV
NS3/4A protease inhibitors, one potential therapeutic target is
the HCV NS5B RNA polymerase, of which 2′-C-methyl-
cytidine-5′-triphosphate is an effective inhibitor.4,5 Valopicita-
bine 1, an early prodrug of the parent nucleoside 2′-C-
methylcytidine 2, utilized a 3′-O-L-valine ester to improve
bioavailability and achieved proof of concept in the clinic
(Figure 1).6 Several subsequent prodrugs of 2′-C-methyl-
cytidine have employed phosphoramidate monoesters, diesters,
or cyclic phosphoramidate diesters with the intention of
delivering the key nucleoside-5′-monophosphate to the liver,
thus bypassing its inefficient initial formation from the free
nucleoside.4,7 2′-C-Methylcytidine prodrug 3 was identified as
one such candidate of interest.8 Bearing both S-acyl thioethyl
and benzyl phosphoramidate moieties, 3 is a mixture of
phosphorus diastereomers in a 1:1 ratio. A scalable synthetic
route was required for prodrug 3 to enable the rapid onset of
further preclinical studies. This article details the novel
chemical synthesis developed toward achieving this objective;
in particular, the use of phenylboronic acid (PBA) as an
efficient, transient protecting group for 2′-C-methylcytidine is
reported.
Figure 1. Valopicitabine 1, 2′-C-methylcytidine 2, and target
phosphoramidate 3.
chemistry procedure on milligram scale in 20% overall yield
from 2′-C-methylcytidine 2 (Scheme 1).8a The strategy
involved (1) sequential protection of the 2′,3′-OH and NH2
groups on the ribonucleoside with acetonide and dimethoxy-
trityl (DMTr), respectively, to avoid competitive phos-
phorylation and to improve the limited solubility of 2 in
organic solvents;7a (2) coupling the protected nucleoside 5
with the H-phosphonate salt intermediate 6, a convergent
building block which had been used to access prodrugs of a
variety of nucleosides; (3) transformation of the H-
phosphonate 7 to phosphoramidate 8; and (4) global
RESULTS AND DISCUSSION
■
Route 1: Original Discovery Synthesis. The phosphor-
amidate prodrug 3 was initially synthesized using the discovery
Received: February 5, 2014
© XXXX American Chemical Society
A
dx.doi.org/10.1021/op500042u | Org. Process Res. Dev. XXXX, XXX, XXX−XXX

Organic Process Research & Development
Article
a
Scheme 1. Discovery Chemistry Synthesis of Prodrug 3 (Route 1)
a
Conditions: (i) HC(OEt)3, acetone, p-TSA (86%); (ii) (a) TMSCl, pyridine, (b) DMTrCl, DMAP, (c) NH4OH, dioxane (81%); (iii) Me3CCOCl,
pyridine (94%); (iv) BnNH2, CCl4 (87%); (v) TFA, DCM (36%).
deprotection of DMTr, acetonide, and trityl groups to give the
free nucleoside phosphoramidate 3.
few examples specifically concerning prodrugs of 2′-C-methyl-
cytidine 2. The two main synthetic approaches, utilizing either
PIII or PV chemistry, have employed 2 protected with acetonide
or as the 2′,3′-hydroxyl-unprotected nucleoside to form 2′-C-
methylcytidine aryl phosphoramidates. The PIII coupling
methodology has been exemplified using diphenylphosphite
at 0 °C followed by installation of a functionalized amine using
Atherton−Todd conditions in 40% yield from the acetonide-
protected nucleoside 4 to give a 1:1 mixture of phosphorus
diastereomers.4,7a,11 The second approach has typically
Critical evaluation of this route highlighted several major
issues which were deemed to be unsuitable or impractical for
scale up. First, the two-step protection protocol of 2′-C-
methylcytidine resulted in a 30% loss of this valuable starting
material, and the use of DMTr protection was poor in terms of
atom economy. The additional expensive reagents, large
amounts of solvents and waste associated with these reactions,
and tedious chromatographic purifications made this procedure
unfavorable. Second, the H-phosphonate 7 was determined to
have poor thermal and hydrolytic stability. Third, carbon
tetrachloride was used in large quantity, functioning as both
reagent and solvent in the Atherton−Todd9 reaction to install
the benzylamine group to give 8. The target nucleoside
phosphoramidate 3 was isolated in a mere 36% yield by
sequential silica gel chromatography, reverse-phase silica gel
chromatography, then lyophilization. Alarmingly, multiple
attempts to scale up this deprotection reaction beyond 1 g
failed as the compound disintegrated to a complex mixture
before the acetonide was completely removed. Additional
identifiable side products observed by mass spectrometry were
found to have incorporated a trifluoroacetate ester at one of the
hydroxyl positions or to have displaced the benzylamine
moiety, giving a trifluoroacetic phosphoric anhydride. Attempts
to use other acids for the deprotection were also ineffective in
producing the desired phosphoramidate 3.10
t
employed a classical 5′-hydroxyl activation with BuMgCl
followed by reaction with a functionalized electrophilic PV
phosphorochloridate.12 In addition to being performed at
−78 °C, under high dilution conditions (<0.1 M), the isolated
phosphorus diastereomer ratios reportedly varied from 1:1 to
1:7.4,13 Relatively modest yields obtained (5−36%) have been
attributed to the poor solubility of unprotected nucleoside 2
under the reaction conditions.7b
Related PV chemistry has also been applied to produce
similar aryl aminoester phosphoramidate prodrugs of other
nucleosides. In the case of 6-O-methyl-2′-C-methylguanosine,
coupling of the nucleoside and phosphorochloridate using
tBuMgCl or N-methylimidazole produced 1:1 phosphorus
diastereomer ratios; however, the yields ranged from 22 to
25% and 11−28% under the two respective conditions.14,15
With the aim of forming a single phosphorus diastereomer, the
use of a phosphorochloridate reagent was replaced by an
t
Route 2: Synthesis with 2′- and 3′-Hydroxyls
Unprotected. Although there are numerous reports of the
formation of nucleoside phosphoramidates, there are relatively
isolable pentafluorophenyl analogue in the BuMgCl-mediated
coupling to form the aryl phosphoramidate prodrug of 2′-
deoxy-2′-fluoro-2′-C-methyluridine.16 Although the chemistry
B
dx.doi.org/10.1021/op500042u | Org. Process Res. Dev. XXXX, XXX, XXX−XXX

Organic Process Research & Development
Article
a
Scheme 2. Synthesis of 3 with 2′- and 3′-Hydroxyls Unprotected (Route 2)
a
Conditions: (i) (a) TMSCl, pyridine, (b) DMTrCl, DMAP, (c) TBAF, THF (93%); (ii) Me3CCOCl, pyridine (27%); (iii) BnNH2, CCl4 (75%);
(iv) TFA, DCM (71%).
was performed on the unprotected 2′-deoxynucleoside,
acetonide-protected adenosine was utilized as a model substrate
in the development to avoid the complications of a competing
hydroxyl group. In practice, 5−8% of the 3′,5′-bisphos-
phorylated uridine side product was produced; however, this
level was noted to be higher for the cytidine analogue.
Application of the benzyl phosphoramidate prodrug in 3 to
2′-C-methylguanosine has been reported using a related
pentafluorophenyl phosphorus ester coupling reaction with
tBuMgCl-mediated 5′-hydroxyl activation.17 Although the 2′,3′-
hydroxyls were unprotected, the use of 6-O-methyl-2′-C-
methylguanosine was presumably required to improve solubility
and avoid base phosphorylation.12b The racemic benzyl
pentafluorophenyl phosphoramidate intermediate was formed
in 38% yield, and the subsequent coupling yields with the two
separate phosphorus enantiomers were a disparate 59 and 36%,
respectively, a potential indication that a 1:1 phosphorus
diastereomer prodrug ratio may not be trivial to obtain via this
methodology.
Installation of the same benzyl phosphoramidate prodrug to
an unnatural, diaminopurine nucleoside analogue has been
reported using PIII chemistry with H-phosphonate salt
intermediate 6.18 The 2′,3′-hydroxyls were unprotected during
the coupling; however, the amino functions on the base were
protected with DMTr groups which presumably improved
solubility of the 2′-C-methyl ribonucleoside.
After consideration of these various alternate approaches, a
second strategy to make prodrug 3, which involved leaving the
nucleoside 2′- and 3′- hydroxyls free, was investigated to
circumvent the problematic acetonide removal of route 1,
thereby also having the advantage of avoiding the initial
acetonide protection step (Scheme 2). The solubility of 2′-C-
methylcytidine was improved by maintaining the DMTr amino
protection, while the PIII coupling/Atherton−Todd chemistry
was retained from route 1 to facilitate a 1:1 phosphorus
diastereomer ratio. This approach involved initial trimethylsilyl
protection of the 2′-, 3′-, and 5′-OH groups, prior to
introduction of DMTr onto the cytosine base, followed by
silyl cleavage using TBAF to give 9. Coupling of 9 with the H-
phosphonate salt intermediate 6 using pivaloyl chloride and
purification by silica gel chromatography resulted in a mere
27% yield of H-phosphonate 10.19 Similar isolated yields (25−
27%) were obtained using EDCI in place of pivaloyl chloride.
Introduction of the benzylamine group to give phosphor-
amidate 11 was performed using the original conditions in 75%
yield. Finally, removal of the DMTr and trityl groups gave the
desired nucleoside phosphoramidate 3 in a significantly
increased 71% deprotection yield. However, despite this
improved final deprotection step, the unprotected 2′,3′-OH
groups led to a decrease in overall yield to 13%. As a result, it
was apparent that protection of the 2′,3′-hydroxyl groups
would be advantageous to obtain reasonable yields. It was
envisioned that a highly labile moiety may potentially avoid
similar problems associated with the acetonide removal and
enable the efficient synthesis of the desired phosphoramidate
prodrug on larger scale.
Route 3: Phenylboronic Acid Protection of 2′-C-
Methylcytidine. Boric20 and boronic acid21 containing
protecting groups have been used for the protection of
carbohydrates and other 1,2-diols; however, their application
in nucleoside chemistry has received only minimal attention.22
It was envisaged that the hydrolytic lability of a 1,2-cyclic
boronate would be an advantageous characteristic in this
instance. Phenylboronic acid is a relatively inexpensive reagent
C
dx.doi.org/10.1021/op500042u | Org. Process Res. Dev. XXXX, XXX, XXX−XXX

Organic Process Research & Development
Article
which has been shown to regioselectively form 2′,3′-cyclic
boronates with unmodified nucleosides, including natural
cytidine.22a It was reasoned that its introduction to give a 2′-
C-methylcytidine-2′,3′-cyclic boronate would behave similarly,
and that its facile removal in a one-pot process would provide
an efficient, transient protection strategy, whilst beneficially
improving the solubility of 2. It was also anticipated that
protection of the cytosine −NH2 may be unnecessary, thus
eliminating the requirement for the uneconomical DMTr
group, as demonstrated with other 2′-C-methylcytidine
phosphoramidate systems using PIII coupling reagents.7a
Scheme 4. Formation of H-Phosphonate Salt Intermediate
6
a
Formation of 2′-C-methylcytidine-2′,3′-O-phenylboronate
12 was initially achieved on a 1 g scale by refluxing 2′-C-
methylcytidine 2 in acetonitrile with 2 equiv of PBA (Scheme
3). The resulting white suspension was cooled and filtered to
a
Conditions: (i) TrCl, DMAP, TEA, DCM; (ii) NaOH, H2O, DME;
(iii) (a) CDI, DCM, (b) HO(CH2)2SH; (iv) H3PO3, DCM,
Me3CCOCl, TEA (66% overall yield from 13).
Scheme 3. Formation of 2′-C-Methylcytidine-2′,3′-
phenylboronate 12
Modification of the original procedure used at 3 g scale to
form thioester 16 began by reducing the molar equivalents of
carbonyldiimidazole (CDI) by 20% to 1.05 and mercapto-
ethanol by 12% to 1.15 with no detriment to the reaction
profile. The high-boiling DMF/toluene solvent system was
replaced by dichloromethane which also enabled a reduction in
volume of 60%. Addition of CDI was performed portionwise to
allow for controlled evolution of CO2 gas. Mercaptoethanol was
then added at such a rate to control the exotherm below 25 °C.
Workup consisted of sequential washing with acid, base, and
water. The concentrated crude thioester was then taken up in
dichloromethane and triethylamine, avoiding the use of
pyridine in the H-phosphonate formation. The amount of
phosphorous acid was reduced by 50% to 2 equiv and that of
pivaloyl chloride by 73% to 1.5 equiv. The exotherm on
addition of pivaloyl chloride was controlled via cooling to
maintain an internal temperature below 25 °C. TEA salt
formation was achieved using further triethylamine, thus
eliminating the use of the expensive triethylammonium
bicarbonate buffer, and workup was further simplified to a
single water wash. Concentration of the organics and drying via
acetonitrile azeotropic distillation provided H-phosphonate 6
with <0.4% water. As the next reaction solvent, residual
acetonitrile presented no concern. This material was of
satisfactory quality (∼90% LCAP26) to be used directly in
the coupling step, thereby eliminating the need for purification
by silica gel chromatography. As a result, production of
intermediate 6 was achieved with 66% overall yield from
hydroxyester 13 in multi-kilogram batches.
Synthesis of Phosphoramidate Prodrug 3. The
synthetic procedure developed to produce phosphoramidate 3
is outlined in Scheme 5. With access to H-phosphonate
intermediate 6, PIII chemistry was employed with the
transiently 2′,3′-O-boronate-protected nucleoside: benzyl
phosphoramidate formation under Atherton−Todd conditions
was again anticipated to provide the desired 1:1 ratio of
phosphorus diastereomers.
Protection of the high-value 2′-C-methylcytidine 2 with PBA
was achieved in refluxing pyridine while removing the
azeotrope. Evaporation of the pyridine gave a golden oil
which was stable enough to be stored overnight under vacuum
or inert atmosphere if required.27 Analysis by proton NMR
indicated greater than a 97:3 ratio of boronate 12 to starting
material 2 on 150 g scale.
give the first example of a 2′-C-methylnucleoside-2′,3′-cyclic
boronate, isolated as a white solid in 75% yield.23 The
regioselectivity of the 2′,3′-cyclic boronate was confirmed by
2D NMR spectrometry. Sensitivity of 12 to hydrolytic cleavage
was apparent from specific rotation24 data recorded at various
time points; initially, a specific rotation of +15.6 was observed
in DMSO containing 0.1% water. After 28 h, the same solution
yielded a specific rotation under similar conditions of +78.1,
corresponding to a slow but appreciable hydrolysis to 2′-C-
methylcytidine, for which [α]D = +137.9.25
20
Removal of the water byproduct was required to obtain full
conversion of 2 to the boronate 12. Either anhydrous sodium
sulfate addition or by switching solvent to pyridine and
subsequently distilling off the pyridine−water azeotrope
1
resulted in 97−99% conversion, as indicated by H NMR.
For further development, the pyridine method was utilized and
the boronate was taken directly to the next step.
Formation of H-Phosphonate Salt Intermediate 6. The
medicinal chemistry synthesis,8a which was originally performed
to give single gram quantities of H-phosphonate salt
intermediate 6, was modified and scaled to 7 kg batch size
(Scheme 4).
Modifications for scale up to 1.8 kg of hydroxyester 13
included decreasing the amounts of dichloromethane by 74% to
10 volumes and trityl chloride by 20% to 0.95 equiv. One hour
reaction time was found to be sufficient rather than 15 h, and
the workup was simplified by using only a single aqueous wash.
Dimethoxyethane was used in preference to the carcinogenic
1,4-dioxane in the subsequent saponification, with a 60%
reduction in total reaction volume. After switching solvents to
dichloromethane and acidification with 5 N HCl, crystallization
from dichloromethane/n-heptane provided crystalline trityl
acid 15 in 73% yield over two steps.
D
dx.doi.org/10.1021/op500042u | Org. Process Res. Dev. XXXX, XXX, XXX−XXX

Organic Process Research & Development
Article
a
Scheme 5. Synthesis of Prodrug 3 Using PBA Transient Protection Strategy (Route 3)
a
Conditions: (i) (a) PhB(OH)2, pyridine, reflux, (b) H-phosphonate 6, EDCI·HCl, MeCN, (c) BnNH2, CCl4, (d) 20% w/w citric acid (aq) (53%);
(ii) AcCl, EtOH (61%).
After redissolution in pyridine, the boronate was added to a
solution of H-phosphonate salt 6 in acetonitrile and the
carbodiimide coupling agent EDCI at 45 °C generated a 7:1
ratio of H-phosphonate diester to 2′-C-methylcytidine 2 by
HPLC analysis.28 Neither extended reaction time, elevated
reaction temperature, nor the addition of further EDCI was
beneficial, potentially due to cleavage of the PBA protection on
prolonged coupling. Pivaloyl chloride was also effective as a
coupling agent; however, as in the case of route 2, these
reactions were less reproducible and robust, primarily due to
higher sensitivity to temperature and moisture. Although H-
phosphonate formation using boronate-protected 12 was
inferior to using acetonide 5, it was dramatically superior to
using the 2′,3′-O-unprotected nucleoside 9. This lower
conversion was more than compensated, however, by avoiding
both the initial inefficient protection of 2 and, ultimately, the
problematic deprotection of acetonide 8.
In order to avoid isolation of the H-phosphonate 17,
investigations were made into the possibility of executing the
benzyl phosphoramidate formation directly on this reaction
mixture. After 4 h reaction time, the H-phosphonate mixture
was cooled to 15−20 °C and excess benzylamine (6.2 equiv)
was added followed by 4.6 equiv of carbon tetrachloride,
resulting in almost immediate conversion of the P−H to
phosphoramidate. Addition of ethyl acetate and aqueous citric
acid effected complete and rapid removal of the PBA protection
to produce the 2′,3′-OH nucleoside phosphoramidate 18 in
situ. Pyridine, benzylamine, PBA, and EDCI-related urea
byproducts were successfully removed after consecutive citric
acid and aqueous bicarbonate washes.29 Purity of the crude
product was upgraded by passing through a short silica plug
(3:1 w/w silica/crude) to give 18 with 98.5% LCAP purity and
the required 1:1 phosphorus diastereomer ratio in 53% yield
from 2′-C-methylcytidine.30
using acetyl chloride in ethanol,31 thereby avoiding the
dramatic degradation of the final product which plagued the
initial scale up efforts. Treatment of the phosphoramidate 18
with this HCl solution at 60 °C resulted in complete
deprotection of the trityl group in <1 h. In contrast to the
original acetonide, DMTr and trityl deprotection of 8 with
TFA, which was unscalable over 1 g, this procedure was shown
to be successful at as high as 250 g phosphoramidate batch size.
After addition of solid sodium bicarbonate and subsequent
trituration with TBME to remove the trityl byproduct, the
residual crude product was purified by reverse-phase silica
chromatography (8:1 w/w silica: crude) to give a 61% yield of
prodrug 3. Overall, this route 3 procedure required only two
product isolations, at the penultimate and final compound
stages, and was scaled up successfully to provide nucleoside
phosphoramidate 3 in 32% yield from 2′-C-methylcytidine.32
Route Comparison via Selected Green Metrics. It is
noteworthy to assess the improvements gained in synthetic
efficiency of the scaled up methodology over the original
discovery procedure. Fundamental green metrics33 not only
provide an assessment of the environmental impact of synthetic
routes but also give an approximate relative indication of their
associated throughput, cost and timeimportant factors in the
rapid, first time scale up of early lead drug candidates. Selected
green metrics are presented for the three routes to 3 (Table 1).
The original acetonide route actually out-performed the
second in terms of overall yield primarily due to the poor
reaction yield during the H-phosphonate coupling associated
with the unprotected 2′,3′-hydroxyls. The cumulative process
mass intensity34 for route 1, however, was determined to be
30% higher than that for route 2. In contrast, despite the
modest absolute values, the boronate route showed marked
improvements over the original discovery method in all
calculated metrics: the number of isolations was reduced
from five to two; both the overall yield and atom economy35
were increased by half, and the cumulative process mass
intensity was reduced by 82%. Although the use of CCl4 in the
Atherton−Todd reaction was not completely obviated, a 97%
reduction in the amount of CCl4 required per gram of final
Using this approach, removal of only a single, labile trityl
group was required to obtain the final phosphoramidate
prodrug 3. The absence of the robust acetonide (and DMTr)
protection allowed a critical switch of acid from 10% TFA in
the discovery synthesis to 3 equiv of dry HCl, generated in situ
E
dx.doi.org/10.1021/op500042u | Org. Process Res. Dev. XXXX, XXX, XXX−XXX

Organic Process Research & Development
Article
(1H, d, J3′,4′ 5.2 Hz, H-3′), 5.78 (1H, d, J5,6 7.5 Hz, H-5), 6.16
(1H, br s, H-1′), 7.30 (2H, 2 × br s, NH2), 7.44 (2H, t, J 7.6
Hz, 2 × Ar-Hmeta), 7.55 (1H, dt, J 7.6 Hz, J 1.3 Hz, Ar-Hpara),
7.75 (2H, dd, J 7.6 Hz, J 1.3 Hz, 2 × Ar-Hortho), 7.80 (1H, d, J6,5
7.5 Hz, H-6); 13C NMR (100 MHz, DMSO-d6) δ 20.23 (CH3),
60.49 (C-5′), 84.84 (C-3′), 85.12 (C-4′), 91.23 (C-2′), 92.34
(C-1′), 93.75 (C-5), 127.89 (2 × Ar-Cmeta), 131.77 (Ar-Cpara),
134.52 (2 × Ar-Cortho), 141.50 (C-6), 154.84, 165.53 (C-2, C-4).
Ar-Cipso not observed. Anal. Calcd for C16H18N3O5B: C, 56.00;
H, 5.29; N, 12.25; B, 3.15. Found: C, 55.95; H, 5.37; N, 12.20;
B, 2.99%.
Table 1. Comparison of Selected Green Metrics for the
Three Routes to 3
atom
economy
(%)
number of
isolations
overall
yield (%)
process mass
a
route
intensity (g/g)
route 1,
5
4
2
20
13
32
25.5
26.8
38.0
511
394
93
acetonide
route 2, free
2′,3′-OH
route 3,
boronate
a
Cumulative process mass intensity of substrates, reagents, and
2,2-Dimethyl-3-(trityloxy)propanoic Acid (15). To a
reactor were charged methyl 2,2-dimethylhydroxypropionate
13 (1.82 kg, 13.77 mol), dichloromethane (18.2 L), trityl
chloride (3.65 kg, 13.10 mol), DMAP (0.168 kg, 1.37 mol), and
triethylamine (1.95 kg, 19.27 mol), and the mixture was heated
to 40−45 °C. The mixture may be held at that temperature for
1−24 h without detriment to the reaction profile. After being
cooled to 20−25 °C, dichloromethane (3.64 L) and water (7.28
L) were added and the biphasic mixture was stirred for 10 min.
The organic layer was separated and distilled under vacuum at
38−42 °C. The crude golden oil methyl 2,2-dimethyl-3-
(trityloxy)propanoate 14 (assumed 13.77 mol) was transferred
using 1,2-dimethoxyethane (16 L) to a reactor containing
NaOH (1.103 kg, 27.58 mol) dissolved in water (12.12 L).
Further DME (3.6 L) was added, and the reaction mixture was
heated to 75 °C for 15 h and monitored by TLC (5% MeOH in
DCM). After concentration to approximately 17 L reaction
volume and cooling to 15 °C, DCM (15 L) was added and the
mixture was acidified with HCl (5 N, 7.32 kg) keeping the
internal temperature <15 °C. The organic layer was separated
and washed with water (10 L), and DCM was partially distilled
at 40 °C to give a volume of approximately 15 L. Heptanes (30
L) were added, and the mixture was cooled from 40 to −10 °C
at 3 °C/h. The solids were isolated by filtration on a Nutsche
filter equipped with a 50 μm frit and dried under argon to give
3.61 kg of acid 15 as fine off-white crystals in 73% yield: m/z
(ESI−) 259.33 [TrO]− 100%, 359.28 [M − H]− 50%; 1H NMR
(400 MHz, CDCl3) δ 1.25 (6H, s, C(CH3)2), 3.20 (2H, s,
CH2O), 7.22−7.32 (9H, m, 9 × Ar-H), 7.45−7.48 (6H, m, 6 ×
Ar-H); 13C NMR (100 MHz, CDCl3) δ 22.52 (C(CH3)2),
43.50 (C(CH3)2), 69.52 (CH2O), 86.46 (CPh3), 127.00 (3 ×
Ar-Cpara), 127.77, 128.75 (6 × Ar-Cortho, 6 × Ar-Cmeta), 143.82
(3 × Ar-Cipso), 182.65 (COOH).
Triethylammonium 2-(2,2-dimethyl-3-(trityloxy)-
propanoylthio)ethyl Phosphonate (6). To a reactor were
charged 2,2-dimethyl-3-(trityloxy)propanoic acid 15 (3.61 kg,
10.02 mol) and dichloromethane (21.3 L). A slurry of 1,1′-
carbonyldiimidazole (1.706 kg, 10.52 mol) in DCM (2.6 L) was
charged portionwise (CO2(g) evolution). The reaction mixture
was stirred not less than 1 h at 20−25 °C. The mixture was
cooled to 15−20 °C, and mercaptoethanol (0.90 kg, 11.52 mol)
was added to maintain internal temperature at <25 °C. The
mixture was stirred for 1.5 h at which point 15 was not
observed by TLC analysis (5% MeOH in DCM). The mixture
was washed with HCl (5 N, 2 × 7 L) followed by aqueous
K2CO3 (10%, 10 L) and water (10 L). The lower organic layer
was distilled under vacuum at 38−42 °C to give thioester 16:
1H NMR (400 MHz, CDCl3) δ 1.22 (6H, s, C(CH3)2), 3.06
(2H, t, CH2S), 3.17 (2H, s, CH2OTr), 3.70 (2H, t, CH2OH),
7.21−7.30 (9H, m, 9 × Ar-H), 7.40−7.42 (6H, m, 6 × Ar-H).
The crude golden oil 16 (assumed 10.02 mol) was transferred
to a reactor with dichloromethane (12.7 L). Phosphorous acid
reaction solvents.
product 3 was achieved. In practice, relative to the unscalable
discovery synthesis, route 3 demanded smaller quantities of raw
materials, allowed for larger batch sizes, and generated less
waste, resulting in a significantly quicker, cheaper, and greener
preclinical production campaign.
CONCLUSION
■
A novel synthetic approach to 2′-C-methylcytidine phosphor-
amidate 3 was developed. In contrast to the original procedure,
which was not viable beyond 1 g due to the harsh conditions
required to remove the acetonide, the new scalable route
featured the application of phenylboronic acid as an alternative,
mild, efficient, and transient protecting group. Accordingly, the
first example of a 2′-C-methylnucleoside-2′,3′-O-phenyl-
boronate is reported. Major advantages of the new route
included scalability, reduction in the number of steps from
seven to five and number of isolations from five to two, increase
in the overall yield by 50%, and significant improvement in
basic green efficiency metrics. Additionally, synthesis of the H-
phosphonate salt 6 was optimized and scaled up to provide this
key intermediate in multi-kilogram batch sizes.
EXPERIMENTAL SECTION
■
General. 2′-C-Methylcytidine was obtained commercially or
synthesized in-house.36 All reactions were performed under an
argon atmosphere using anhydrous solvents unless otherwise
stated. Reduced pressure distillations were performed between
250 and 25 mbar. HPLC spectra were obtained on an Agilent
1100 series instrument equipped with a UV detector at 254 nm
and an Agilent Zorbax Eclipse XDB C8 reverse-phase column
(4.6 mm × 75 mm, 3.5 μm) at 25 °C; flow rate 1.4 mL/min;
mobile phase A, acetonitrile; B, 0.01 M ammonium acetate
aqueous buffer; run time 10 min; gradient, 5 to 80% A over 5.5
min, hold at 80% A for 4.5 min.
2′-C-Methylcytidine-2′,3′-O-phenylboronate (12). 2′-
C-Methylcytidine 2 (1.0 g, 3.89 mmol) was suspended in
anhydrous MeCN (10 mL) at 20−25 °C. Phenylboronic acid
(0.948 g, 7.77 mmol) was charged in one portion, and the
mixture was heated at reflux (82 °C) for 2 h. The suspension
was cooled to 20−25 °C. The solids were collected by vacuum
filtration and dried under vacuum at 30−35 °C to afford 1.01 g
(75%) of phenylboronate 12 as a white solid: mp 245−246 °C;
m/z (ESI−, direct injection) 342.2 [M − H]− 100%; νmax (KBr,
cm−1) 3424, 3213, 1651, 1611, 1355; [α]D = +15.62 (c 1.0,
20
DMSO containing 0.1% H2O) at t = 0; [α]D20 = +78.13 (c 1.0,
DMSO containing 0.1% H2O) at t = 28 h; 1H NMR (400 MHz,
DMSO-d6) δ 1.21 (3H, s, CH3), 3.74 (1H, ddd, J5′,5″ 12.1 Hz,
J
5′,OH‑5′ 5.6 Hz, J5″,4′ 4.2 Hz, H-5′), 3.82 (1H, ddd, J5″,5′ 12.1 Hz,
J5″,OH‑5′ 4.5 Hz, J5″,4′ 3.4 Hz, H-5″), 4.05 (1H, m, H-4′), 4.65
F
dx.doi.org/10.1021/op500042u | Org. Process Res. Dev. XXXX, XXX, XXX−XXX

Organic Process Research & Development
Article
(1.65 kg, 20.06 mol) was charged followed by triethylamine
(3.0 L, 21.56 mol), keeping the internal temperature at <25 °C.
A solution of pivaloyl chloride (1.86 L, 15.10 mol) in DCM
(1.8 L) was added to maintain internal temperature at <25 °C,
and the mixture was stirred at 20−25 °C for 2 h. Additional
triethylamine (2.79 L, 20.06 mol) was charged to the reaction.
After 10 min, water (5 L) was added. The settled phases were
separated, and the organic layer was distilled under vacuum at
38−42 °C. After coevaporation with acetonitrile (2 × 5 L), H-
phosphonate salt 6 containing residual acetonitrile (7.7 kg,
132%) was obtained as a pourable golden oil of sufficient
quality (>90% LCAP, <0.4% water content) to be used in the
next step. An analytical sample of 6 was obtained after washing
1% TEA in DCM solution of 6 with an aqueous mixture of 3%
TEA in brine followed by drying (Na2SO4), filtration, and
concentration to an oil. Purification using silica gel chromatog-
raphy (1% TEA in DCM then 1% TEA, 10% MeOH in DCM
as eluent) gave 6 as a clear, colorless, viscous oil: m/z (ESI−)
483.1 [M − NEt3 − H]− 100%; 967.4 [2(M − NEt3) − H]−
25%; νmax (film, cm−1) 3388 (br), 2980, 1647 (br), 2603, 1676
yellow foam. The crude residue was dissolved in DCM (1 L)
and purified on a silica gel (2.3 kg) plug. Elution was carried
out using an increasing gradient of MeOH in DCM; 4% (7 L),
5% (3 L), 6% (2 L), 7% (10 L). Evaporation of the appropriate
fractions gave 254 g (53% yield; 98.5% LCAP) and 73 g (13%
yield; 87.6% LCAP, repurified with subsequent batches) of
trityl phosphoramidate 18 with combined yield 65%: 1H NMR
(400 MHz, DMSO-d6) two diastereomers δ 0.95 (3H, s, CH3),
1.14 (6H, s, (CH3)2C), 3.06 (2H, s, CH2OTr), 3.10 (2H, m,
CH2S), 3.59 (1H, br m, H-3′), 3.83−4.02 (5H, m, H-4′, CH2O,
CH2Ph), 4.09−4.13 (1H, m, H-5′), 4.14−4.28 (1H, m, H-5″),
5.11 (1H, s, OH-2′), 5.30 (1H, br d, J 6.6 Hz, OH-3′), 5.70
(1H, d, J 7.4 Hz, H-5), 5.72−5.79 (1H, m, P-N-H), 5.94 (1H,
br s, H-1′), 7.19−7.35 (22H, m, 20 × Ar-H, NH2), 7.55, 7.59 (2
× 0.5H, 2d, J 7.5 Hz, H-6); 31P NMR (162 MHz, DMSO-d6)
two diastereomers δ 9.68−9.95 (1P, m).
2′-C-Methylcytidine-5′-[2-[(3-hydroxy-2,2-dimethyl-1-
oxopropyl)thio]ethyl-N-benzylphosphoramidate] (3).
Trityl phosphoramidate 18 (246 g, 0.296 mol) was dissolved
in anhydrous EtOH (3.5 L), and acetyl chloride (62.6 mL, 0.88
mol) was added resulting in an exotherm from 18 to 27 °C.
The mixture was heated to 60−65 °C for 45 min at which point
it was cooled to <25 °C. Solid sodium bicarbonate (1.04 kg)
was added portionwise, bringing the pH to 5.5−6. The mixture
was filtered through Celite and washed with EtOH (7 L) prior
to concentration of the filtrate at 35 °C. The residue was
triturated with TBME (3 L) for 1 h and then filtered to obtain
185 g of crude product 3 (93% LCAP). The crude material was
dissolved in MeCN (58 mL), water (164 mL), and aqueous
NaHCO3 (saturated, 170 mL) and purified by chromatography
on reverse-phase silica (octadecyl, 40 μm, 1.5 kg, prewashed
with a gradient of 100% MeCN to 100% water). Elution was
carried out using a stepwise gradient of MeCN in water: 3, 10,
15, and held at 25% to obtain the product. Evaporation of the
appropriate fractions gave 106 g (61% yield; 98.5% LCAP) of
phosphoramidate 3 as a white solid: m/z (ESI+) 587.12 [M +
H]+ 100%; 1173.62 [2M + H]+ 80%; νmax (KBr, cm−1) 3343
1
(s); H NMR (400 MHz, CDCl3) δ 1.20 (6H, s, (CH3)2C),
1.30 (9H, t, J 7.3 Hz, N(CH2CH3)3), 3.02 (6H, q, J 7.3 Hz,
N(CH2CH3)3), 3.15 (2H, s, CH2OTr), 3.18 (2H, t, J 6.9 Hz,
CH2S), 3.96 (2H, dt, J 8.6 Hz, J 6.9 Hz, CH2OP), 6.87 (1H, d,
1JH−P 617 Hz, P-H), 7.20−7.24 (3H, m, 3 × Ar-Hpara), 7.27−
7.31 (6H, m, 6 × Ar-Hmeta), 7.39−7.43 (6H, m, 6 × Ar-Hortho),
12.71 (1H, br s, Et3N+-H); contains 0.4 mol MeOH by proton
NMR; 13C NMR (100 MHz, CDCl3) δ 8.50 (3 × NCH2CH3),
22.87 (C(CH3)2), 29.55 (3JC−P 6.8 Hz, CH2S), 45.38 (3 ×
NCH2CH3), 50.86 (C(CH3)2), 62.32 (2JC−P 4.3 Hz, CH2OP),
69.87 (CH2OTr), 86.35 (CPh3), 126.93 (3 × Ar-Cpara), 127.71
(3 x Ar-Cmeta), 128.78 (3 × Ar-Cortho), 143.86 (Ar-Cipso), 204.52
(COS); 31P NMR (162 MHz, CDCl3) δ 4.33 (1P, s). Anal.
Calcd for [C32H44NO5PS.(0.4 CH3OH)]: C, 65.01; H, 7.68; N,
2.34; S, 5.36; P, 5.29. Found: C, 64.57; H, 7.56; N, 2.10; S,
5.12; P, 5.30.
2′-C-Methylcytidine-5′-[2-[(3-trityloxy-2,2-dimethyl-1-
oxopropyl)thio]ethyl-N-benzylphosphoramidate] (18).
Pyridine (2.5 L) was charged to 2′-C-methylcytidine 2 (150
g, 0.583 mol) at 20−25 °C. Phenylboronic acid (78 g, 0.626
mol) was added, and the mixture was heated at reflux (115 °C)
for 3 h. The pyridine−water azeotrope was then distilled at
atmospheric pressure, removing 1.2 L of distillate. The mixture
was cooled to 20−25 °C, and the pyridine was evaporated
under vacuum to obtain a golden oil. Analysis of the crude
material by 1H NMR indicated greater than 97% conversion to
the desired phenylboronate 12, which was stable stored under
argon at 20−25 °C for >15 h. H-Phosphonate salt intermediate
6 (0.615 kg, 1.049 mol) was dissolved in acetonitrile (3 L) at
20−25 °C. A solution of 12 (assumed 0.583 mol) in pyridine
(250 mL) was charged to the mixture followed by EDCI·HCl
(0.570 kg, 2.973 mol), and the reaction was stirred at 41−46 °C
for 4 h. The reaction was cooled to <20 °C, and benzylamine
(395 mL, 3.61 mol) was added followed by carbon
tetrachloride (260 mL, 2.69 mol) controlling the internal
temperature to <20 °C. After 1 h, EtOAc (1 L) was charged to
the mixture and acidified to pH 4 with aqueous citric acid (20%
w/w, 3 L) to effect cleavage of the cyclic boronate. The
separated aqueous phase was extracted with EtOAc (2.5 L).
The combined organics were washed with aqueous citric acid
(10% w/w, 3 L) then twice with aqueous NaHCO3 (saturated,
5 L then 2 L). The organic phase was dried (Na2SO4), filtered,
and evaporated at 25−35 °C to give 712 g of crude 18 as a
20
(br), 1647 (br), 1616, 1495; [α]D = +55.01 (c 1.0, DMSO);
1H NMR (400 MHz, DMSO-d6) two diastereomers δ 0.94
(3H, 2s, CH3), 1.11 (6H, s, (CH3)2C), 3.04 (2H, m, J 6.4 Hz,
CH2S), 3.44 (2H, d, J 5.0 Hz, CH2OH), 3.60 (1H, br m, H-3′),
3.82−4.01 (5H, m, H-4′, CH2O, CH2Ph), 4.07−4.12 (1H, m,
H-5′), 4.13−4.24 (1H, m, H-5″), 4.94 (1H, t, J 5.0 Hz,
CH2OH), 5.07 (2 × 0.5H, 2s, OH-2′), 5.26 (1H, t, J 6.8 Hz,
OH-3′), 5.64−5.76 (1H, m, P-N-H), 5.69, 5.70 (2 × 0.5H, 2d,
2 × J 7.6 Hz, H-5), 5.93 (1H, br s, H-1′), 7.13−7.20 (2H, 2 ×
br s, NH2), 7.20−7.25 (1H, m, Ar-H), 7.28−7.35 (4H, m, 4 ×
Ar-H), 7.53, 7.57 (2 × 0.5H, 2d, J 7.6 Hz, H-6); 13C NMR (100
MHz, DMSO-d6) two diastereomers δ 19.81 (CH3), 21.79
(C(CH3)2), 28.17, 28.24 (CH2S), 44.18 (PhCH2), 51.62
(C(CH3)2), 63.74, 63.79 (CH2O), 64.21, 64.51 (C-5′), 68.29
(CH2OH), 72.41, 72.57 (C-3′), 77.80, 77.85 (C-2′), 79.47, (C-
4′), 91.66, (C-1′), 93.82 (C-5), 126.68, 127.09, 128.08, 128.09
(5 x Ar-C), 140.34, 140.38, 140.40 (Ar-Cipso, C-6), 155.12,
165.21 (C-2, C-4), 203.85 (COS); 31P NMR (162 MHz,
DMSO-d6) two diastereomers δ 9.71, 9.91 (1P, 2s, ratio
1.00:1.07). Anal. Calcd for C24H35N4O9PS: C, 49.14; H, 6.01;
N, 9.55; S, 5.47; P, 5.28. Found: C, 48.74; H, 5.83; N, 9.41; S,
5.81; P, 5.33.
AUTHOR INFORMATION
Corresponding Author
*E-mail: mayes.ben@idenix.com.
■
G
dx.doi.org/10.1021/op500042u | Org. Process Res. Dev. XXXX, XXX, XXX−XXX

Organic Process Research & Development
Article
diastereomer mixture. Tran, K.; Eastgate, M. D.; Janey, J.; Chen, K.;
Rosso, V. W. Patent Appl. WO2014008236, 2014.
(16) Ross, B. S.; Reddy, P. G.; Zhang, H.-R.; Rachakonda, S.; Sofia,
Notes
The authors declare no competing financial interest.
M. J. J. Org. Chem. 2011, 76, 8311−8319.
ACKNOWLEDGMENTS
(17) Ross, B. S.; Sofia, M. J.; Reddy, P. G.; Rachakonda, S.; Zhang,
H.-R. WO2011123668, 2011.
■
Ricerca Biosciences are acknowledged for demonstrating the H-
phosphonate salt chemistry on scale in their facility.
(18) Butler, T.; Cho, A.; Kim, C. U.; Saunders, O. L.; Zhang, L.
WO2009132135, 2009.
(19) In process checks ranged widely from 12 to 52% liquid
chromatography area percentage (LCAP).
REFERENCES
■
(20) Bjorsvik, H.-R.; Priebe, H.; Cervenka, J.; Aabye, A. W.;
Gulbrandsen, T.; Bryde, A. C. Org. Process Res. Dev. 2001, 5, 472−478.
(21) Dahlhoff, W. V.; Koster, R. J. Org. Chem. 1977, 42, 3151−3157.
(22) (a) Yurkevich, A. M.; Kolodkina, I. I.; Varshavskaya, L. S.;
Borodulina-Shvets, V. I.; Rudakova, I. P.; Preobrazhenskii, N. A.
Tetrahedron 1969, 25, 477−484. (b) Sekine, M.; Aoyagi, M.; Ushioda,
M.; Ohkubo, A.; Seio, K. J. Org. Chem. 2005, 70, 8400−8408.
(c) Jegourel, D.; Delepee, R.; Breton, F.; Rolland, A.; Vidal, R.;
Agrofoglio, L. A. Bioorg. Med. Chem. 2008, 16, 8932−8939.
(23) PBA protection of other unnatural nucleosides has also been
performed and will be reported in due course.
(1) (a) Lee, L. Y.; Tong, C. Y. W.; Wong, T.; Wilkinson, M. Int. J.
Clin. Pract. 2012, 66, 342−355. (b) Shepard, C. W.; Finelli, L.; Alter,
M. J. Lancet Infect. Dis. 2005, 5, 558−567.
(2) American Association for the Study of Liver Diseases Hepatitis C
Guidelines 2012. Gaetano, J. N.; Desai, A. P.; Reau, N. Clin. Med.
Insights: Ther. 2012, 4, 39−56.
(3) Thompson, A.; Patel, K.; Tillman, H.; McHutchison, J. G. J.
Hepatol. 2009, 50, 184−194.
(4) Gardelli, C.; Attenni, B.; Donghi, M.; Meppen, M.; Pacini, B.;
Harper, S.; Di Marco, A.; Fiore, F.; Giuliano, C.; Pucci, V.; Laufer, R.;
Gennari, N.; Marcucci, I.; Leone, J. F.; Olsen, D. B.; MacCoss, M.;
Rowley, M.; Narjes, F. J. Med. Chem. 2009, 52, 5394−5407.
(5) The recent regulatory approval of sofosbuvir has clinically
validated this target: http://www.fda.gov/downloads/
AdvisoryCommittees/CommitteesMeetingMaterials/Drugs/
AntiviralDrugsAdvisoryCommittee/UCM371877.pdf (Accessed 25
November 2013).
20
(24) Values of [α]D (c 1.0, DMSO containing 0.1% H2O).
(25) Data qualitatively illustrate the general trend for the specific
rotation to tend toward the value for 2′-C-methylcytidine over time.
Concentrations differ.
(26) Liquid chromatography area percentage.
(27) Although not performed, concentration to 1/10th volume and
telescoping directly to the coupling reaction is anticipated to be
feasible.
(28) Boronates 12 and 17 were completely deprotected during the
reverse-phase HPLC analysis; therefore, the respective 2′,3′-diols were
monitored.
(29) It remains to be determined whether arylboronic acids in
general represent a genotoxic hazard. O’Donovan, M. R.; Mee, C. D.;
Fenner, S.; Teasdale, A.; Phillips, D. H. Mutat. Res. 2011, 724, 1−6.
Beyond preclinical development, the potential for PBA to be a
genotoxic impurity in 3 would need to be considered. Methods to
control residual arylboronic acids in APIs are discussed in http://www.
andersonsprocesssolutions.com/controlling-residual-arylboronic-acids-
as-potential-genotoxic-impurities-in-apis/#_ednref1 (Accessed 01
April 2014).
(6) Pierra, C.; Amador, A.; Benzaria, S.; Cretton-Scott, E.; D’Amours,
M.; Mao, J.; Mathieu, S.; Moussa, A.; Bridges, E. G.; Standring, D. N.;
Sommadossi, J.-P.; Storer, R.; Gosselin, G. J. Med. Chem. 2006, 49,
6614−6620.
(7) (a) Donghi, M.; Attenni, B.; Gardelli, C.; Di Marco, A.; Fiore, F.;
Giuliano, C.; Laufer, R.; Leone, J. F.; Pucci, V.; Rowley, M.; Narjes, F.
Bioorg. Med. Chem. Lett. 2009, 19, 1392−1395. (b) Meppen, M.;
Pacini, B.; Bazzo, R.; Koch, U.; Leone, J. F.; Koeplinger, K. A.; Rowley,
M.; Altamura, S.; Di Marco, A.; Fiore, F.; Giuliano, C.; Gonzalez-Paz,
O.; Laufer, R.; Pucci, V.; Narjes, F.; Gardelli, C. Eur. J. Med. Chem.
2009, 44, 3765−3770.
(8) (a) Sommadossi, J.-P.; Gosselin, G.; Pierra, C.; Perigaud, C.;
Peyrottes, S. Patent Appl. WO2008082601, 2008. (b) Sommadossi, J.-
P.; Gosselin, G.; Pierra, C.; Perigaud, C.; Peyrottes, S. Patent Appl.
US20090169504, 2009.
(30) A further 13% yield with 88% LCAP purity was also obtained
(which was combined and repurified with subsequent batches) to give
a total yield of 65% from 2′-C-methylcytidine.
(9) Atherton, F. R.; Openshaw, H. T.; Todd, A. R. J. Chem. Soc. 1945,
660−663. Georgiev, E. M.; Kaneti, J.; Troev, K.; Roundhill, D. M. J.
Am. Chem. Soc. 1993, 115, 10964−10973.
(31) Highly exothermic.
(32) This PBA transient protection methodology has since
successfully been applied in the synthesis of other nucleoside
phosphoramidates, to be reported in due course.
(33) Constable, D. J. C.; Curzons, A. D.; Cunningham, V. L. Green
Chem. 2002, 4, 521−527.
(34) Cumulative process mass intensity defined as [total mass of
substrates, reagents, and reaction solvents (g)]/[mass of isolated
product (g)]. Does not include contributions due to workup, isolation,
and purification of products. See PMI calculator: http://www.acs.org/
content/acs/en/greenchemistry/industry-business/pharmaceutical.
html (Accessed 07 April 2014).
(10) For example, p-TsOH and HCOOH.
(11) Attenni, B.; Donghi, M.; Gardelli, C.; Meppen, M.; Narjes, F.;
Pacini, B. Patent Appl. WO2008142055, 2008.
(12) (a) Hayakawa, Y.; Aso, Y.; Uchiyama, M.; Noyori, R.
Tetrahedron Lett. 1983, 24, 1165−1168. (b) Uchiyama, M.; Aso, Y.;
Noyori, R.; Hayakawa, Y. J. Org. Chem. 1993, 58, 373−379.
(13) (a) MacCoss, M.; Olsen, D. B.; Donghi, M.; Gardelli, C.;
Harper, S.; Meppen, M.; Narjes, F.; Pacini, B. WO2007095269, 2007.
(b) Narjes, F.; Gardelli, C.; Donghi, M.; Attenni, B.; Durette, P. L.
WO2008085508, 2008.
(14) (a) McGuigan, C.; Madela, K.; Aljarah, M.; Gilles, A.; Brancale,
A.; Zonta, N.; Chamberlain, S.; Vernachio, J.; Hutchins, J.; Hall, A.;
Ames, B.; Gorovits, B.; Ganguly, B.; Kolykhalov, A.; Wang, J.;
Muhammad, J.; Patti, J. M.; Henson, G. Bioorg. Med. Chem. Lett. 2010,
20, 4850−4854. (b) McGuigan, C.; Madela, K.; Aljarah, M.; Gilles, A.;
Battina, S. K.; Ramamurty, C. V. S.; Rao, C. S.; Vernachio, J.; Hutchins,
J.; Hall, A.; Kolykhalov, A.; Henson, G.; Chamberlain, S. Bioorg. Med.
Chem. Lett. 2011, 21, 6007−6012. (c) Chamberlain, S.; Hutchins, J.;
Madela, K.; McGuigan, C.; Vernachio, J.; Aljarah, M.; Gilles, A. Patent
Appl. WO2010081082, 2010.
(35) Li, C.-J.; Trost, B. M. Proc. Natl. Acad. Sci. U.S.A. 2008, 105,
13197−13202. Atom economy defined as [FW product (g mol−1)]/
[FW of all reactants used in reaction (g mol−1)] × 100%.
(36) According to the procedure described in the following patent:
Storer, R.; Moussa A.; Chaudhuri, N.; Waligora, F. Patent Appl.
WO2004052899, 2004.
(15) A recent addition to the patent literature demonstrated that
using 2 equiv of the uronium reagent HATU with a functionalized
phosphoric acid could be used to form this prodrug in 90% yield on 1
g scale. The addition of quinine resulted in a 6.7:1 phosphorus
H
dx.doi.org/10.1021/op500042u | Org. Process Res. Dev. XXXX, XXX, XXX−XXX