A single-pot synthesis of atovaquone: An antiparasitic drug of choice

Article abstract of DOI:10.1021/op500032w

The present article relates to a practical, economically viable, and validated at industrial scale, single-pot synthetic route for preparation of atovaquone, one of the most versatile antiparasitic drugs of choice used for the prophylaxis and treatment of diseases such as pneumocystis, toxoplasmosis, babesiosis, coccidiosis, and malaria. However, owing to the extremely poor yields of synthesis and very high doses of treatment (due to poor bioavailability) the cost of treatment with this drug is not affordable by the patients in need, particularly in the third world countries where these diseases are most prevalent. Unlike most of the reported processes which use 2-chloronaphthoquinone and pure trans-4-chlorophenyl cyclohexane carboxylic acid, our process is based on the decarboxylative alkylation of isomeric mixture of 4-chlorophenyl cyclohexane carboxylic acid with 1,4-naphthoquinone to give 42% overall yield of atovaquone, 10 times higher than from the reported process (4%) from the innovators of this drug.

Full text of DOI:10.1021/op500032w

Article
pubs.acs.org/OPRD
A Single-Pot Synthesis of Atovaquone: An Antiparasitic Drug of
Choice
Suneel Y. Dike,* Dharmendra Singh, Byju N. Thankachen, Brajesh Sharma, Pramil K. Mathur,
Swapnil Kore, and Ashok Kumar*
Ipca Laboratories Ltd., 123-AB, Kandivli Industrial Estate, Kandivli (W), Mumbai - 400067, Maharashtra, India
S
* Supporting Information
ABSTRACT: The present article relates to a practical, economically viable, and validated at industrial scale, single-pot synthetic
route for preparation of atovaquone, one of the most versatile antiparasitic drugs of choice used for the prophylaxis and treatment
of diseases such as pneumocystis, toxoplasmosis, babesiosis, coccidiosis, and malaria. However, owing to the extremely poor
yields of synthesis and very high doses of treatment (due to poor bioavailability) the cost of treatment with this drug is not
affordable by the patients in need, particularly in the third world countries where these diseases are most prevalent. Unlike most
of the reported processes which use 2-chloronaphthoquinone and pure trans-4-chlorophenyl cyclohexane carboxylic acid, our
process is based on the decarboxylative alkylation of isomeric mixture of 4-chlorophenyl cyclohexane carboxylic acid with 1,4-
naphthoquinone to give 42% overall yield of atovaquone, 10 times higher than from the reported process (4%) from the
innovators of this drug.
chlorophenylcyclohexane from its carboxylic acid (3) using
AgNO3 and ammonium persulfate following a method reported
by Jacobsen et al.22 and subsequent condensation with 2-
chloronaphthoquinone (2) to give chloroatovaquone (4).
Hydrolysis of 4 with methanolic KOH gives the required
product (1) in 4−4.5% overall yield in two steps (Scheme 1).
The first report on the free radical-mediated alkylation of 2-
hydroxy-1,4-naphthoquinone with cycloalkyl carboxylic acids,
in general, was published in 1948 by Fieser23 using diacyl
peroxide to give respective products in a range from 5 to 6%
(Scheme 2); however, it was not extended for the synthesis of
1.
Two more processes along similar lines using 2-hydroxy-1,4-
naphthoquinone (13) (Scheme 3)24 and 2,3-dichloro-1,4-
naphthoquinone (15) (Scheme 4)25 instead of 2, have recently
been reported to give atovaquone in 23% and 38% yields,
respectively. The main disadvantage of both processes is the
cost and nonavailability of starting materials [2-hydroxy-1,4-
naphthoquinone (13) and 2,3-dichloro-1,4-naphthoquinone
(15)] and lower overall yields of the final product.
An additional route which starts from 4-(4-chlorophenyl)-
cyclohexane oxalate (8)26 and 2, is reported to yield
chloroatovaquone (4) in 43% yield, and would ultimately
yield 25% or less of atovaquone if extrapolated using
Hudson’s21 hydrolytic transformation of 4 to atovaquone (1),
(Scheme 5). The most recent synthesis from GlaxoSmithK-
line27 (Scheme 6) and an additional route using Barton’s
method28 are novel from the chemistry point of view, but may
not prove to be economical as both use exotic reagents and
catalysts.
INTRODUCTION
■
About 3.3 billion people, almost half of the world’s population,
is at risk of malaria, and therefore, undoubtedly the last few
decades have seen significant progress towards treatment of
malaria and efforts to develop new antimalarial agents.
Artemisinins, having proven themselves to be life-saving
drugs,1−3 however, are not useful for malaria prophylaxis4,5
because of their incapability of acting upon the initial liver
stages of the malaria parasite. Malarone, a fixed dose
combination of atovaquone with proguanil hydrochloride
marketed by GlaxoSmithKline, on the other hand, stands
alone for its versatility because it is useful for the treatment as
well as casual prophylaxis of malaria in chloroquine-sensitive
and -resistant strains of Plasmodium falciparum.6 Atovaquone
acts by inhibiting the mitochondrial electron transport system
and thereby nucleotide biosynthesis,7−10 and proguanil by
synergistically blocking the alternate routes of electropotential
generation via adenine nucleotide carriers (ANC) and
decreasing the effective concentration of atovaquone required
for killing the parasite11−13 Atovaquone alone is also used
against Pneumocystis carinii14 and tachyzoite and cyst forms of
Toxoplasma gondii15,16 and is marketed as Mepron by
GlaxoSmithKline for the treatment of pneumonia in patients
intolerant to the first-line therapy using trimethoprim-
sulphamethoxazole (TMP-SMX).17 Treatment of babesiosis is
yet another important indication of atovaquone when given in
combination with azithromycin.18 However, the only factor
which has prevented Malarone from becoming first-line
treatment for malaria is its very high cost as compared to
other antimalarial drugs, including artemisinins. Indeed, there is
an unmet need to develop an economically viable synthesis of
atovaquone19 to make this drug affordable to patients in the
third world.
The first synthesis of atovaquone (1) reported by Hudson et
Received: January 29, 2014
Published: April 8, 2014
al.20,21 relies on the generation of a free radical of trans-4-
© 2014 American Chemical Society
618
dx.doi.org/10.1021/op500032w | Org. Process Res. Dev. 2014, 18, 618−625

Organic Process Research & Development
Article
Scheme 1. Hudson’s synthesis of atovaquone (1)
isolation of 4 from the crude product also contributes to the
lower yields of atovaquone in 6−6.5% yields, the formation of
dechlorinated product (a mixture of trans-6 and cis-6a isomers)
in the free radical-assisted reaction is a major source of yield
loss, and the formation of the cis-isomer (4a) occurs even when
pure trans-3 is used as an alkylation agent in the first step.
Free radical-mediated reactions in general are not very clean,
but the formation of 6 and 6a in high percentage was an
important observation and suggested that, if the formation of
these side products were avoided, then the overall yield would
be more than double. It was found to be a challenge to reduce
this to practice; however, rather than minimizing the formation
of 6, what appeared more interesting was to plan the synthesis
of atovaquone (1) through this intermediate 6 itself. This was a
novel approach that prompted the examination of the
condensation of 1,4-naphthoquinone (5) with 3 to prepare 6
and exploration of its transformation into atovaquone (1).
Gratifyingly,29 the reaction of 5 with 3 following Hudson’s21
conditions [Scheme 8; Example (i)] occurred smoothly. The
crude product obtained from the reaction (run in triplicate)
was analyzed to assess ratios and yields of 6 and 6a, which was
found to be in a range of 55−61% (Table 1). It is pertinent to
mention that the reaction of 2 and 3 gave a yield of <19%
under identical conditions (see Table 2). As detailed in Scheme
5, the crude product obtained from the reaction of 5 and 3 was
crystallized from acetonitrile affording 97% pure trans-(6) in
∼20% yield.
After establishing the feasibility of the synthesis of 6 from 5,
the next step was the transformation to atovaquone which was
envisioned by chlorinating 6 and subjecting the expected
dichloro product 7 to dehydrohalogenation under standard
conditions [(Scheme 9, Example (iii)].
The chlorination of 6 in acetic acid proceeded well to give
the dichloro product30 (7) with quantitative conversion of 6.
Though the isolation of the dichloro intermediate was not
necessary, it was isolated and analyzed by HPLC. The analysis
confirmed the presence of 7 as a mixture of two isomers in a
ratio of 30:60 along with ∼7% of monochlorinated product 4.
The presence of ∼7% dehydrohalogenated product (4) in the
reaction mass was encouraging and demonstrated the ease with
Scheme 2. Fieser’s free radical-mediated alkylation of 2-
hydroxy-1,4-naphthoquinone (13)
RESULTS AND DISCUSSION
■
The method reported by Hudson et al.20,21 appeared to be too
impractical for a large-scale preparation of atovaquone because
of its poor yields. However, the reported process was repeated
to examine and understand the factors responsible for lower
yields. In addition, pure samples of the various intermediates
were also obtained for quantitative monitoring of this
condensation step.
Reactions of 2 with 3 [Scheme 7; Example (x) below] were
performed in triplicate in the presence of powdered AgNO3
and ammonium persulfate in acetonitrile/water as per the
published method to get a semisolid mass after reaction work-
up. Assay using HPLC and pure samples of 4, 4a, 6, and 6a
(quantitative) revealed the crude product to be a mixture of
mainly trans-isomer (4) (8−8.5%), cis-isomer (4a) (9.5−10%),
the dechlorinated product as a mixture of 6 and 6a (13−14%),
and a substantial amount of unidentified polymeric material.
Crystallization of the crude product thus obtained under
reported conditions gave 97−98% pure 4 in 6−6.5% isolated
yields. Hydrolysis of the latter following Hudson’s21 conditions
gave pure atovaquone in 57−60% yield (maximum overall yield
on 3 was ∼4%).
The learnings gleaned from these experiments are that the
conditions and yields published by Hudson et al.21 are
reproducible, free radical-mediated alkylation of chloronaph-
thoquinone is responsible for the majority of the yield loss of
the target product (i.e., atovaquone), the poor efficiency in the
Scheme 3. Synthesis of atovaquone using 2-hydroxy-1,4-naphthoquinone (13)
619
dx.doi.org/10.1021/op500032w | Org. Process Res. Dev. 2014, 18, 618−625

Organic Process Research & Development
Article
Scheme 4. Synthesis of atovaquone using 2,3-dichloro-1,4-naphthoquinone (15)
Scheme 5. Williams’ synthesis of atovaquone (1)
Scheme 6. GlaxoSmithKline’s synthesis of atovaquone (1)
Scheme 7. Validation of Hudson’s condensation of 2 with 3
Scheme 8. Condensation of 1,4-naphthoquinone (5) with 4-(4-chlorophenyl) cyclohexane carboxylic acid (3)
which elimination of hydrochloric acid was taking place from 7.
The mixture of 7 and 4 on treatment with anhydrous sodium
acetate in acetic acid gave 4 as a single product. Hydrolysis of 4
with methanolic KOH, as previously presented, gave
quantitative yield of atovaquone (1) of high purity (overall
yield =13.5% on 3, see Scheme 9).
It was encouraging to obtain 3−4-fold increase in yield in
comparison to the original synthesis reported by Hudson;21
however, it was also very clear that if atovaquone was to replace
artemisinin as a first-line treatment for malaria, the synthesis
could not afford to lose a significant amount of product as its
unwanted isomer (6a), (almost 60% of the reaction products)
620
dx.doi.org/10.1021/op500032w | Org. Process Res. Dev. 2014, 18, 618−625

Organic Process Research & Development
Article
Being convinced that the acid-catalyzed isomerisation of 1a
to 1 should proceed; efforts continued and ultimately the
desired transformation proceeded in nearly quantitative yield at
ambient temperature (25−30 °C) when 90% H2SO4 was used.
After our discloser29 on the isomerisation of 1a to 1, another
report claiming a similar transformation appeared;33 however,
inferior results in terms of yield and purity of atovaquone were
obtained.
The current process takes care of most of the issues
highlighted above wherein the chlorination is carried out on the
mixture of 6 and 6a to get dichloro product (7) as a mixture of
isomers, which upon dehydrohalogenation, followed by
hydrolysis, yields atovaquone as a mixture of cis- trans-isomers
(1a and 1). Upon treatment with 90% aqueous sulfuric acid as
disclosed above, the mixture of 1a + 1 leads to the desired
productatovaquone (1) through a single-pot synthesis, in
42% overall yield on the mixture of trans- and cis-4-
chlorophenyl cyclohexanecarboxylic acid (3 and 3a), almost
10 times higher than reported by Hudson et al.21 [Scheme 12;
Example (xi)]
Table 1. Reaction of 4-(4-chlorophenyl) cyclohexane
carboxylic acid (3) and 1,4-naphthoquinone (5) following
the method given in Example (i)
ratio of
sr.
no.
example
no.
cis-isomer
(6a) %
trans-isomer HPLC assay of both isomers
(6) %
(6 + 6a) %
1
2
3
(i)-A
(i)-B
(i)-C
61.18
63.20
65.10
38.8
55.5
59.8
61.4
36.79
34.89
Table 2. Reaction of 4-(4-chlorophenyl) cyclohexane
carboxylic acid (3) and 2-chloro-1,4-naphthoquinone (2)
following the method given in Example (ix)
ratio of
sr.
no.
example
no.
cis-isomer
(4a) %
trans-isomer HPLC assay of both isomers
(4) %
(4 + 4a) %
1
2
3
(x)-A
(x)-B
(x)-C
51.85
50.67
45.87
48.15
49.32
54.13
18.9
16.8
18.2
CONCLUSION
■
The work reported in this paper discusses a novel and cost-
effective single-pot practical synthesis of atovaquone which
takes care of most of the issues related to the synthesis and
yield of this versatile antiparasitic drug, atovaquone. Unlike
most of the reported processes which use 2-chloronaphthoqui-
none and pure trans-4-chlorophenyl cyclohexane carboxylic
acid, our process is based on the decarboxylative alkylation of
an isomeric mixture of 4-chlorophenyl cyclohexane carboxylic
acid with 1,4-naphthoquinone to give 42% overall yield of
atovaquone, 10 times higher than that from the reported
process (4%) by the innovators of this drug.
invariably formed during reaction of 3 with 5. Since
epimerization in a free radical-mediated reaction is typical, it
was realized that it may not be possible to avoid the formation
of 6a in the condensation of 5 and 3. The only practical way to
improve overall yield of atovaquone would be utilizing the
formation of 6a and isomerization to the desired stereo-
chemistry (6). A literature search did not reveal any method to
convert 6a directly to 6 via isomerization. However, the
literature search led to reports claiming isomerization of cis-
atovaquone (1a) as well as acetyl atovaquone (2-[4-(4-
chlorophenyl)cyclohexyl]-3-acetoxy-1,4-naphthoquinone) to
1.24 Hence, the goal could in theory be achieved by
isomerization at the final step, provided a practical method to
convert 6a into 1a was available. Chlorination of 6a and
dehydrochlorination of the resulting dichloro intermediate (7)
led to 4a in 99% yield (Scheme 10). The basic hydrolysis of 4a
did indeed afford 1a. The overall yield of 1a starting from 6a
was 65%.
After establishing the successful transformation of 6a to 4a
and 1a, attention was focused on the isomerisation of 1a to 1.
Thermal conditions as reported by Verma et al.31were initially
investigated. Unfortunately, the reported results could not be
repeated. The conditions, (99% H2SO4), as disclosed by
Antonio Nardi et al.,24 also did not yield the desired results. It
was disappointing; however, the report on the formation of
unwanted products (Scheme 11) in a similar study32 on cis-2-
hydroxy-3-(4-butylcyclohexyl)-1,4-naphthoquinone (9) at 55
°C further substantiated our findings that these conditions are
somewhat not suitable for the desired transformation.
EXPERIMENTAL SECTION
■
Example (i): trans-2-[4-(4-Cyclohexyl)chlorophenyl]-
1,4-naphthoquinone (6). To a stirred solution of silver
nitrate (2.13 kg, 12.54 mol) dissolved in 30 L water, trans-4-(4-
chlorophenyl)cyclohexane carboxylic acid (3), (15.0 kg, 62.89
mol) was added. To this was added acetonitrile (75 L) under
stirring. The reaction mixture was heated to reflux, and 1,4-
naphthoquinone (5), (12.0 kg, 75.94 mol) was added in three
lots. Ammonium persulfate (35.85 kg, 157.2 mol) dissolved in
90 L water was added over a period of 1 h to the stirred
solution and maintained under reflux for half an hour. The
reaction mass was cooled to 30−32 °C and extracted with 2 ×
30 L methylene chloride. The organic layer was first washed
with 30 L water and then with 2 × 30 L 10% sodium carbonate
aqueous solution, followed by water until the pH became
neutral. The reaction mass was distilled to remove methylene
chloride and was stirred in 9.0 L acetonitrile and filtered. The
Scheme 9. Transformation of 6 to 4 by classical chlorination/dechlorination method
621
dx.doi.org/10.1021/op500032w | Org. Process Res. Dev. 2014, 18, 618−625

Organic Process Research & Development
Article
Scheme 10. Transformation of unwanted isomer (6a) into cis-isomer of atovaquone (1a)
a
Scheme 11. Hudson’s study on H2SO4-catalyzed degradation of cis-2-hydroxy-3-(4-butylcyclohexyl)-1,4-naphthoquinone (9)
a
cis-trans-2-Hydroxy-3-(4-butylcyclohexyl)-1,4-naphthoquinone (9), 3-tert-butyl-1,2,3,4-tetrahydrobenzo(b)naphtho[2,3-d]furan-6−11-quinone
(10), 9-tert-butyl-6b,7,8,9,10,10a,-hexahydrobenzo-[b]naphtho[2,1-d]furan-5,6,quinine (11), 2-hydroxynaphthoquinone (12).
Scheme 12. Single-pot synthesis of atovaquone (1) from naphthoquinone (5) and isomeric mixture of 4-chlorophenyl
cyclohexane carboxylic acid (3 and 3a)
solid thus obtained was heated to reflux in 2 × 14 L acetonitrile,
cooled to 30−32 °C to crystallize, and filtered to get 6. Yield:
4.6 kg (20.8%); mp 147−149 °C (uncorrected). IR (KBr) ν
780, 823, 1091, 1259, 1305, 1490, 1591, 1661, 2856, 2939,
3497 cm−1; 1H NMR (CDCl3 with 0.03% TMS, 400 MHz): δH
8.08−8.15 (2H, m, H13 and H16), 7.74−7.77 (2H, m, H14
and H15), 7.28−7.30 (2H, d, J = 8.5 Hz, H19 and H21), 7.18−
7.20 (2H, d, J = 8.5 Hz, H18 and H22.), 6.81 (1H, d, J = 1.2
Hz, H12), 2.99−3.05 (1H, tt, H4), 2.55−2.62 (1H, tt, H1),
2.01−2.05 (4H, m, H3 and H5), 1.42−1.72 (4H, m, H2 and
H6). 13C NMR (CDCl3 with 0.03% TMS, 100 MHz) δ 185.8,
184.8, 155.7, 145.3, 133.7, 133.1, 132.5, 131.9, 131.7, 128.5,
128.2, 126.8, 126.0, 43.4, 36.2, 33.9, 32.2. HRMS calcd for
C22H19O2Cl [M + H]+ 351.11518, found 351.11544. The
HPLC analysis using hypersil BDS, acetonitrile/water/meth-
anol/ortho-phosphoric acid (525:300:175:5), 3 mL/min, 25
°C, shows purity more than 97% with cis-isomer being less than
0.5%.
90 L aqueous solution obtained from the above experiment was
added to it over a period of 1 h. and maintained under reflux for
half an hour. The reaction mixture was then cooled to 30−32
°C and extracted with 3 × 30 L methylene chloride. The
organic layer was first washed with 50 L water followed by
washing with 3 × 30 L 10% sodium carbonate aqueous
solution, followed by water until the pH became neutral. The
organic layer was distilled to eliminate methylene chloride and
was stirred in 12.0 L acetonitrile and filtered. The solid thus
obtained was further leached in 14 L acetonitrile to get 6 of
97.5% purity. Yield: 4.46 kg (19.98%). Mp 147−149 °C
(uncorrected).
Example (iii): 2-[4-(4-Chlorophenyl)cyclohexyl]-2,3-
dichloro-2,3-dihydro-1,4-naphthoquinone(7). To a
stirred suspension of 2-[4-(4-chlorophenyl)cyclohexyl]-1,4-
naphthoquinone (6), (4.5 kg, 12.83 mol) in glacial acetic acid
(22.5 L) was passed chlorine gas at about 20 °C. After
completion, the reaction mass was filtered, washed with 0.5 L
glacial acetic acid and then with water until the pH of the
washing became neutral. The product was dried at 30−32 °C;
Yield 4.65 kg (85.9%). The HPLC analysis using hypersil BDS,
acetonitrile/water/methanol/ortho-phosphoric acid
(525:300:175:5), 3 mL/min, 25 °C, showed it to be a mixture
of two isomers in ratio of 30% and 60%, respectively, along with
contamination of ∼7% monochlorinated product (4). The
Example (ii): Recycling of Silver Nitrate. To the stirred
aqueous solution (180 L) obtained from the above experiment
were gradually added silver nitrate (1.07 kg, 6.27 mol) and
trans-4-(4-chlorophenyl)cyclohexane carboxylic acid (3, 15.0
kg, 62.89 mol). The reaction mixture was heated to reflux, and
1,4-naphthoquinone (2, 12.0 kg, 75.94 mol) was added in three
lots. Ammonium persulfate (35.85 kg, 157.2 mol) dissolved in
622
dx.doi.org/10.1021/op500032w | Org. Process Res. Dev. 2014, 18, 618−625

Organic Process Research & Development
Article
crude product as such was used in the next step without further
purification.
was neutral. The solvent was then concentrated to get 155 g of
2-[4-(4-chlorophenyl)cyclohexyl]-3-chloro-1,4-naphthoqui-
none (4 + 4a) which was dissolved in 4.5 L methanol. A
solution of potassium hydroxide (225 g, 4.02 mol) in water
(1550 mL) was added dropwise under heating over a period of
1 h. The reaction mass was then refluxed for 2 h and cooled to
30−32 °C and filtered. The filtrate was acidified with 50%
aqueous hydrochloric acid to precipitate the solid mass as a
mixture of trans- and cis-atovaquone (1 + 1a), yield 125 g
(81.3%).
Example (vii): Preparation of cis-Atovaquone (1a)
from cis-2-[4-(4-chlorophenyl) cyclohexyl]-1,4-naphtho-
quinone (6a). To a stirred suspension of cis-2-[4-(4-
chlorophenyl) cyclohexyl]-1,4-naphthoquinone (6a), (5.0 g,
0.014 mol) in glacial acetic acid (37.5 mL) was passed chlorine
gas at about 20 °C under continuous stirring, and the reaction
was monitored by TLC. After the complete disappearance of
the starting material, nitrogen gas was purged to remove excess
chlorine from the reaction mass. Anhydrous sodium acetate
(1.47 g) was then added, and the reaction mass was heated to
reflux for 90 min. The reaction mass was cooled and quenched
into a mixture of methylene chloride and water. The methylene
chloride was washed with water until neutral pH. The solvent
was then concentrated to get 5.5 g cis 2-[4-(4-chlorophenyl)
cyclohexyl]-3-chloro-1,4-naphthoquinone (4a) which was
dissolved in 165 mL methanol. A solution of potassium
hydroxide (8 g, 0.14 mol) in water (55 mL) was added
dropwise under heating over a period of 1 h. The reaction mass
was then refluxed for 2 h and cooled to 30−32 °C and filtered.
The filtrate was acidified with 50% aqueous hydrochloric acid
to precipitate a solid mass which on recrystallization from
acetonitrile gave 3.4 g (65%) cis-atovaquone (1a).
Example (viii): Preparation of Atovaquone (1). To a
solution of silver nitrate (14.17 g, 0.08 mol) dissolved in 200
mL water were added trans-4-(4-chlorophenyl)cyclohexane
carboxylic acid (3) [100 g (0.42 mol)] and acetonitrile (500
mL). The solution was heated to reflux followed by addition of
1,4-naphthoquinone (5), (80 g, 0.51 mol). A solution of
ammonium persulfate (239 g; 1.05 mol) in water (600 mL) was
added dropwise to the above solution, and reflux was continued
for half an hour. The reaction mass was then cooled to 30−32
°C and stirred with ethyl acetate (400 mL) and the lower
aqueous layer was allowed to settle and was removed; the upper
ethyl acetate layer was washed with 10% sodium carbonate
aqueous solution, and then with water until neutral pH. The
organic layer was concentrated to afford a solid mass which was
dissolved in glacial acetic acid (375 mL) in the same pot, and
chlorine gas was passed at about 20 °C under continuous
stirring; the reaction was monitored by TLC. After the
complete disappearance of the starting material, nitrogen gas
was purged to remove excess chlorine from the reaction mass.
Anhydrous sodium acetate (52.3 g) was then added, and the
reaction mass was heated to reflux for 90 min. The acetic acid
was distilled out completely to get 2-[4-(4-chlorophenyl)-
cyclohexyl]-3-chloro-1,4-naphthoquinone (4 + 4a) which was
dissolved in 2.25 L methanol. A solution of potassium
hydroxide (225g, 4.02 mol) in water (1550 mL) was added
dropwise under reflux over a period of 1 h. The reaction mass
was then refluxed for 2 h, cooled to 30−32 °C, and acidified
with 50% aqueous hydrochloric acid to precipitate a solid mass
as a mixture of trans- and cis-atovaquone (1 + 1a), yield 130 g
(84.6%).
Example (iv): 2-[4-(4-Chlorophenyl)cyclohexyl]-3-
chloro-1,4-naphthoquinone (4). The crude product ob-
tained from the above experiments (4.5 kg, 10.70 mol) was
suspended in glacial acetic acid (36 L) and anhydrous sodium
acetate (1.32 kg) was added to the reaction mixture. The latter
was heated to reflux for 1 h, and 124 L water was added to the
cooled reaction mixture. The precipitated product was filtered
off, dried at 65 °C, and recrystallized from 360 L acetonitrile to
get 4. Yield: 3.66 kg (89.0%); mp185−187 °C; IR (KBr) ν 830,
848, 1083, 1282, 1458, 1566, 1593, 1668, 3071 cm−1; 1H NMR
(CDCl3 with 0.03% TMS, 400 MHz): δ 8.10−8.16 (m, 2H),
7.73−7.80 (m, 2H), 7.28−7.30 (d, 2H), 7.19−7.21 (d, 2H),
3.32−3.40 (tt, 1H), 2.66−2.74 (tt, 1H), 2.30−2.41 (m, 2H),
2.01−2.05 (m, 2H), 1.80−1.85 (m, 2H), 1.53−1.64 (m, 2H).
The HPLC analysis using hypersil BDS, acetonitrile/water/
methanol/ortho-phosphoric acid (525:300:175:5), 3 mL/min,
25 °C, confirmed the purity of trans-isomer to be more than
99.5%.
Example (v): Preparation of Atovaquone (1). 2-[4-(4-
Chlorophenyl)cyclohexyl]-3-chloro-1,4-naphthoquinone (4),
3.5 kg, 9. 09 mol) was suspended in 105 L methanol, and
potassium hydroxide (5.1 kg, 91.07 mol) dissolved in 35 L
water was added under heating to 65−68 °C over a period of
20 min. Further, the reaction mixture was refluxed for 45 min
and cooled to 30−32 °C and filtered. The filtrate was acidified
with 50% aqueous hydrochloric acid to precipitate the product,
which was filtered, dried, and recrystallized from 300 L
acetonitrile to obtain 1. Yield: 2.83 kg (85%); mp 219−221
°C; IR (KBr) ν 728, 822, 1089, 1278, 1347, 1369, 1490, 1595,
1
1659, 2855, 2926, 3377 cm−1; H NMR (CDCl3 with 0.03%
TMS, 400 MHz): δ 8.08−8.16 (m, 2H), 7.68−7.80 (m, 2H),
7.55 (s, 1H), 7.27−7.29 (d, 2H), 7.18−7.20 (d, 2H), 3.15−3.23
(tt, 1H), 2.62−2.69 (tt, 1H), 2.16−2.26 (m, 2H), 1.97−2.01
(m, 2H), 1.77−1.80 (m,2H), 1.54−1.65 (m, 2H). The HPLC
analysis using hypersil BDS, acetonitrile/water/methanol/
ortho-phosphoric acid (525:300:175:5), 3 mL/min, 25 °C,
confirmed the purity of atovaquone to be more than 99.8%,
without a detectable amount of cis-isomer and no other single
impurity more than 0.05%.
Example (vi): Preparation of Atovaquone (1). To a
solution of silver nitrate (14.17 g, 0.08 mol) dissolved in 200
mL water were added trans-4-(4-chlorophenyl)cyclohexane
carboxylic acid (3) 100 g (0.42 mol) and acetonitrile (500
mL). The solution was heated to reflux followed by addition of
1,4-naphthoquinone (5), (80 g, 0.51 mol). A solution of
ammonium persulfate (239 g; 1.05 mol) in water (600 mL) was
added dropwise to the above solution and reflux continued for
half an hour. The reaction mass was then cooled to 30−32 °C
and extracted with methylene chloride. The organic layer was
first washed with water, followed with 10% sodium carbonate
aqueous solution, and then with water until the pH was neutral.
The organic layer was concentrated to afford a solid mass which
was dissolved in glacial acetic acid (745 mL), and chlorine gas
was passed at about 20 °C under continuous stirring, and the
reaction was monitored by TLC. After the complete
disappearance of the starting material, nitrogen gas was purged
to remove excess chlorine from the reaction mass. Anhydrous
sodium acetate (52.3 g) was then added, and the reaction mass
was heated to reflux for 90 min. The reaction mass was cooled
and quenched into a mixture of methylene chloride and water.
The methylene chloride was washed with water until the pH
623
dx.doi.org/10.1021/op500032w | Org. Process Res. Dev. 2014, 18, 618−625

Organic Process Research & Development
Article
Example (ix): Epimerization of Mixture of Atova-
quone (1 + 1a). The cis-/trans-isomeric mixture (130 g)
obtained from the above Example (viii) was added to 1140 mL
of 90% sulfuric acid and stirred at 28−30 °C. The reaction was
monitored by HPLC, and when the cis-isomer had reached a
level of less than 0.5%, the reaction mixture was quenched into
ice−water, filtered, and washed with water until neutral pH.
The crude product was purified by column chromatography on
silica gel (1% ethyl acetate/hexane) affording 64.5 g
atovaquone (yield 42% from 3). The HPLC analysis using
hypersil BDS, acetonitrile/water/methanol/ortho-phosphoric
acid (525:300:175:5), 3 mL/min, 25 °C, confirmed the purity
of atovaquone to be more than 99.8%, without a detectable
amount of cis-isomers and no other single impurity more than
0.05%.
90% sulfuric acid and stirred at 28−30 °C for 4 h. The reaction
was monitored by HPLC, and when the cis-isomer (1a) was
less than 0.5%, the reaction mixture was quenched into ice−
water, filtered, and washed with water until neutral pH. The
crude product was purified by column chromatography on silica
gel (1% ethyl acetate/hexane) affording atovaquone (1) (126
g). Yield: 41%. HPLC purity: 99.8%.
ASSOCIATED CONTENT
■
S
* Supporting Information
1H NMR for structures no. 1, 4 and 6; details of HRMS
measurements and HRMS spectrum structure no. 6, 13C NMR
spectrum and 2D NMR spectrum for structure no. 6. This
material is available free of charge via the Internet at http://
pubs.acs.org.
Example (x): 2-Chloro-[4-(4-chlorophenyl)cyclohexyl]-
1,4-naphthoquinone (4). A mixture of 2-chloro-1,4-naph-
thoquinone (2), (8.07 g, 0.04 mol), trans 4-(4-chlorophenyl)-1-
cyclohexane carboxylic acid (3), (10 g, 0.04 mol), and silver
nitrate (2.14 g,0.01 mol) was heated to vigorous reflux in 80
mL acetonitrile. A solution of ammonium persulfate (24.0 g,
10.11 mol) in 100 mL water was added over a period of 1.0 h.
The mixture was refluxed for 3 h, then cooled in ice for 30 min,
extracted with 2 × 50 mL boiling MDC, washed with water,
and distilled off when a yellow−brown mass was obtained. The
HPLC and mass spectra of this material showed the presence of
cis- and trans-isomers of 2-chloro-[4-(4-chlorophenyl) cyclo-
hexyl]-1,4-naphthoquinone (4a and 4) as well as the mixture of
cis- and trans-isomers of 2-[4-(4-chlorophenyl)cyclohexyl]-1,4-
naphthoquinone (6 and 6a).
Example (xi): Preparation of Atovaquone (1) from a
mixture of 3 and 3a. To a solution of silver nitrate (19.33 g,
0.11 mol) dissolved in 272 mL water were added cis- and trans-
4-(4-chlorophenyl)cyclohexane carboxylic acid [200 g (0.8385
mol)] (3 + 3a) and acetonitrile (680 mL). The solution was
heated to reflux followed by addition of 1,4-naphthoquinone
(5) (108.11 g, 0.68 mol). A solution of ammonium persulfate
(325.3 g; 1.43 mol) in water (816 mL) was added dropwise to
the above solution, and reflux was continued for 30 min. The
reaction mass was then cooled to 30−32 °C and extracted with
methylene chloride. The organic layer was first washed with
50% hydrochloric acid followed by 10% sodium carbonate
aqueous solution, then 50% hydrochloric acid, and then with
water until neutral pH. The organic layer was concentrated to
dryness, glacial acetic acid (1.02L) was added, and chlorine gas
was passed at about 20 °C. After the complete disappearance of
the starting material (as monitored by TLC), nitrogen gas was
purged to remove excess chlorine from the reaction mass.
Anhydrous sodium acetate (63.09 g, 0.77 mol) was then added,
and the reaction mass was heated to reflux for 120 min. The
reaction mass was cooled and quenched into a mixture of
methylene chloride and water. The methylene chloride was
then washed with water until neutral pH. The solvent was
concentrated to obtain the residue containing a mixture of
compound 4 and 4a, and the residue was then stirred with 6.39
L methanol. A solution of potassium hydroxide (306.4 g, 5.47
mol) in water (2.038 L) was added dropwise under heating
over a period of 2 h. The reaction mass was then refluxed for
2.5 h and cooled to 30−32 °C and filtered. The filtrate was
neutralized with 50% aqueous hydrochloric acid to precipitate a
mixture of 1 and 1a. It was then filtered and washed with water
until neutral pH and dried in an oven at 65 °C. Dry atovaquone
(1+1a) (207.6 g) was obtained which was added to 2.076 L of
AUTHOR INFORMATION
■
Corresponding Authors
*E-mail: suneel.dike@gmail.com.
*E-mail: ashok.kumar@ipca.com.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
The authors acknowledge the contribution of Dr. Gaurav Sahal
for editing the manuscript.
REFERENCES
■
(1) Sutherland, C. J.; Ord, R.; Dunyo, S.; Jawara, M.; Drakeley, C. J.;
Alexander, N.; Coleman, R.; Pinder, M.; Walraven, G.; Targett, G.
PLoS Med. 2005, 2, e92.
(2) Mclntosh, H. M.; Olliaro, P. Med. Trop (Marseille) 1998, 58 (3
Suppl.), 61−62.
(3) Gabor, A. B. Pharmacol. Ther. 2001, 90 (2−3), 261−265.
(4) Olumese, P. Guidelines for the Treatment of Malaria; World
Health Organization: Geneva, 2006.
(5) Lucy, C. O.; Chris, J. D.; Teun, B.; Christopher, J. W.; Azra, C. G.
PLoS Med. 2008, 5 (11), e226.
(6) Marra, F.; Salzman, J. R.; Ensom, M. H. Ann. Pharmacother. 2003,
37 (9), 1266−1275.
(7) Fry, M.; Pudney, M. Biochem. Pharmacol. 1992, 43, 1545−1553.
(8) Srivastava, I. K.; Rottenberg, H.; Vaidya, A. B. J. Biol. Chem. 1997,
272, 3961−3966.
(9) Biagini, G. A.; Vinayavejakul, P.; O’Neill, P. M.; Bray, P. G.;
Ward, S. A. Antimicrob. Agents Chemother. 2006, 50, 1841−1851.
(10) Seymour, K. K.; Yeo, A. E.; Rieckmann, K. H. Ann. Trop. Med.
Parasitol. 1997, 91, 603−609.
(11) Srivastava, I. K.; Vaidya, A. B. Antimicrob. Agents Chemother.
1999, 43 (6), 1334−1339.
(12) Painter, H. J.; Morrisey, J. M.; Mather, M. W.; Vaidya, A. B.
Nature. 2007, 446, 88−91.
(13) Painter, H. J.; Morrisey, J. M.; Vaidya, A. B. Antimicrob. Agents
Chemother. 2010, 54 (12), 5281−5287.
(14) Baggish, A. L.; Hill, D. R. Antimicrob. Agents Chemother. 2002,
46, 1163−1173.
(15) Romand, S.; Pudney, M.; Derouin, F. Antimicrob. Agents
Chemother. 1993, 37, 2371−2378.
(16) Guttendge, W. E.; Latter, V. S. Hudson, A. T. Naphthoquinones
for the treatment and prophylaxis of Pneumocystis carinii infections.
EP0362996 B1. 27 Apr 1994; Chem. Abstr. 1990, 113, 152064.
(17) Hughes, W. T.; LaFon, S. W.; Scott, J. D.; Masur, H. J. Infect.
Dis. 1995, 171, 1295−1301.
(18) Wittner, M.; Lederman, J.; Tanowitz, H. B.; Rosenbaum, G. S.;
Weiss, L. M. Am. J. Trop. Med. Hyg. 1996, 55, 219−222.
624
dx.doi.org/10.1021/op500032w | Org. Process Res. Dev. 2014, 18, 618−625

Organic Process Research & Development
Article
(19) Nixon, G. A.; Moss, D. M.; Shone, A. E.; Lalloo, D. G.; Fisher,
N.; O’Neill, P. M.; Ward, S. A.; Biagini, G. A. J. Antimicrob. Chemother.
2013, 1−9.
(20) Hudson, A. T. Randall, A. W. Naphthoquinone derivatives.
EP123238. 1 July, 1992; Chem. Abstr. 1985, 102, 113082.
(21) Hudson, A. T. Latter, V. Medicaments. EP0580185A1. 26 Jan,
1994; Chem. Abstr. 1990, 113, 152064.
(22) Jacobsen, N.; Torrsell, K. J. Liebigs Ann. Chem. 1972, 736, 135−
147.
(23) Fieser, L. J. Am. Chem. Soc. 1948, 70 (10), 3174−3215.
(24) Antonio, N.; Mara, S.; Annibale, S. Stefano, M. Process for the
preparation of trans-2,3-disubstituted naphthoquinones. U.S. Patent
7,842,840 B2. 30 Nov, 2010; Chem. Abstr. 2008, 149, 576277.
(25) Saralya, S. S.; Shasikumar, S. H.; Shashipraba; Kanakamajalu, S.;
Koottungalamadhom, R. R.; Ananathalakshmi V.; Govindarajalu, J.;
Rao, K. S. Nagarajan, K. Preparation of naphthoquinone compounds
using 2,3-dihalonaphthoquinone. U.S. Patent Appl. 0004024 A1. 6 Jan
2011; Chem. Abstr. 2009, 151, 425364.
(26) Williams, D. R.; Clark, M. P. Tetrahedron Lett. 1998, 39, 7629−
7632.
(27) Dwyer, A. N.; Gordon, A. Urquhart, M. Novel process. WO/
2012/080243 A2. 21 June, 2012; Chem. Abstr. 2012, 157, 133525.
(28) Zhu, F.; Qiao, H. Bekhazi, M. A process for preparing
atovaquone and associate intermediates. WO/2010/001379 A1. 7 Jan,
2010; Chem. Abstr. 2010, 152, 119262.
(29) Kumar, A.; Dike, S. Y.; Mathur, P. K.; Thankachen, B. N.;
Sharma, B.; Kore, S. S. Buchude, V. S. Process for preparation of
atovaquone and novel intermediates thereof. U.S. Patent 7,847,127 B2.
7 Dec, 2010; Chem. Abstr. 2010, 152, 429385.
(30) Romily, C. A.; Peter, B. D.; Mare, D. L.; Larsen, D. S. J. Chem.
Soc., Perkin Trans. 1983, 2, 271−279.
(31) Verma, S. S.; Patel, D. J. Dwivedi, S. D. Process for preparation
of atovaquone and the conversion of cis-isomer to trans-isomer. WO/
2008/122988 A1. 16 Oct, 2008; Chem. Abstr. 2008, 149, 471210.
(32) Hudson, A. T.; Randall, A. W. Eur. J. Med. Chem. 1986, 271−
275.
(33) Zhu, F. Bekhazi, M. Process for the epimerization of atovaquone
isomer, atovaquone intermediates and mixture thereof. U.S. Patent
0144347 A1. 16 June, 2011; Chem. Abstr. 2010, 152, 153712.
625
dx.doi.org/10.1021/op500032w | Org. Process Res. Dev. 2014, 18, 618−625