A practical synthesis of 8-hydroxyacyclovir and 9-(carboxymethoxymethyl) guanine, metabolites of acyclovir

Article abstract of DOI:10.3987/COM-05-10430

Improved methods for the synthesis of 8-hydroxyacyclovir and 9-(carboxymethoxymethyl)guanine, metabolites of acyclovir, were examined. The methods were found to be useful for practical preparation of 8-hydoxyacyclovir and 9-(caboxymethoxymethyl)guanine of high purity. Careful spectroscopic analysis of 8-hydroxyacyclovir in DMSO-d₆ suggested that it may be exist in an 8-oxo tautomer rather than an 8-hydroxy tautomer under the solution conditions.

Full text of DOI:10.3987/COM-05-10430

HETEROCYCLES, Vol. 65, No. 8, 2005
1967
HETEROCYCLES, Vol. 65, No. 8, 2005, pp. 1967 - 1973
Received, 15th April, 2005, Accepted, 27th May, 2005, Published online, 27th May, 2005
A PRACTICAL SYNTHESIS OF 8-HYDROXYACYCLOVIR AND
9-(CARBOXYMETHOXYMETHYL)GUANINE,
METABOLITES
OF
ACYCLOVIR
a
b
b
c
Teruo Kutsuma, Yasuo Sakai, Hiroyuki Ouchi, Noriaki Shiina, Takehiro
a
a
a,
Yamagishi, Shiroshi Shibuya, and Tsutomu Yokomatsu *
a
School of Pharmacy, Tokyo University of Pharmacy & Life Science, 1432-1
b
Horinouchi, Hachioji, Tokyo 192-0392, Japan. Toa Pharmaceuticals Co., Ltd.,
c
2
6 Sango, Toyama 939-3548, Japan. Junsei Chemical Co., Ltd., 1-6 Ohmano,
Koshigaya, Saitama 343-0844, Japan. E-mail: yokomatu@ps.toyaku.ac.jp
Abstract – Improved methods for the synthesis of 8-hydroxyacyclovir and
9
-(carboxymethoxymethyl)guanine, metabolites of acyclovir, were examined.
The methods were found to be useful for practical preparation of
-hydoxyacyclovir and 9-(caboxymethoxymethyl)guanine of high purity. Careful
spectroscopic analysis of 8-hydroxyacyclovir in DMSO-d suggested that it may
8
6
be exist in an 8-oxo tautomer rather than an 8-hydroxy tautomer under the
solution conditions.
9
-(2-Hydroxyethoxymethyl)guanine (1) (acyclovir), the acyclic nucleoside analogue of guanosine, is a
selective inhibitor of the replication of herpes simplex virus (HSV) types 1 and 2 and varicella-zoster
1
2
virus. Many synthetic and metabolic studies of acyclovir have been reported in the last decade. The
metabolic disposition of acyclovir has been investigated in humans and several experimental species
including mice, rats and dogs. In these metabolic studies, 8-hydroxy-9-(2-hydroxyethoxymethyl)guanine
(
8-hydroxyacyclovir) (2) and 9-(caboxymethoxymethyl)guanine (CMMG) (3) have been identified as
3
major metabolites for humans and animals.
O
O
N
O
N
N
N
N
N
N
HN
HN
H N
HN
H N
OH
N
H N
N
2
2
2
O
O
O
CO H
2
OH
OH
acyclovir (1)
8-hydroxy acyclovir (2)
CMMG (3)

1968
HETEROCYCLES, Vol. 65, No. 8, 2005
In the course of our investigation directed toward developing novel prodrugs of acyclovir, we required a
large quantity of 8-hydroxyacyclovir and CMMG of high purity for the metabolic studies. Although a
4
method for the synthesis of 8-hydroxyacyclovir (2) has been reported in the literature, in our experience,
the method could not be applied to large-scale synthesis. The structural analysis of 2 was not studied in
detail. Moreover, synthetic methods for CMMG were not disclosed in the literature. In this paper we
wish to describe experimental details for a practical synthesis of 8-hydroxyacyclovir and CMMG with
their full physical constants.
SYNTHESIS OF 4-HYDROXYACYCLOVIR
Robins et al. reported that 8-hydroxyacyclovir (2) could be prepared from 8-bromoacyclovir (4) by
4
treatment with sodium acetate in glacial acetic acid, followed by neutralization with 0.1N NaOH.
However, our attempted application of the reported procedure for the synthesis in large scale was not
successful, due to production of a mixture of 2 and the corresponding acetates in varying ratios, which
were not readily separated. Then, we examined a modified procedure for obtaining 2 in large scale
(
Scheme 1).
Treatment of 8-bromoacyclovir (4) with sodium acetate in glacial acetic acid at reflux for 6 h, followed
by neutralization with powdered NaHCO gave the acetate (5) in 75% yield. The results suggest that the
3
initial-formed 8-acetoxyacyclovir is transformed to 5 via concomitant migration of the acetyl group to the
primary hydroxyl group under the conditions. In this procedure, we readily obtained multi-gram
quantities of 5 of high purity after single recrystallization of the crude materials. When 5 was treated with
6
2
-aminoethanol in boiling water for 5 h, hydrolysis of the acetate group cleanly occurred to give the
requisite 8-hydroxyacyclovir (2), which could be isolated in 96% yield after single recrystallization of the
crude materials from 25% aqueous acetic acid. By using the step-wise method via the acetate (5), we
obtained several grams of extremely pure 8-hydroxyacyclovir (2) in 72% yield for two steps.
O
O
N
N
AcONa/AcOH
N
N
1
)
HN
H N
HN
H N
Br
OH
2
)
0.1 N NaOH
N
2
N
2
O
O
OH
OH
4
2
O
2
-aminoethanol
N
N
HN
water/100°C
1
) AcONa/AcOH
) NaHCO3
OH
2
H N
N
2
O
OAc
5
Scheme 1

HETEROCYCLES, Vol. 65, No. 8, 2005
1969
Although compound (2) was believed for a long time albeit without substantiating evidence to be in the
8
8
-hydroxy tautomer, theoretically, 2 would exist in an equilibrium between the 8-hydroxy tautomer and
-oxo tautomer (oxo-2) (Scheme 2). Structural analysis of 8-hydroxy-2′-deoxyguanosine (8-OH-dG) has
7
been reported by Culp and co-worker. They found that chemical shifts of 5-C for 8-OH-dG in solution
of pH 6.0–14 are markedly dependent upon the pH value. The 5-C resonance exhibited a significant
downfield chemical shift of ca. 10 ppm from 101 ppm with increasing pH, while the remaining base
carbons exhibit minimal effects in this pH range. These marked shift in 5-C indicates that increasing
basicity of the solution increases the 8-enolate form of 8-OH-dG in an equilibrium. Then we carefully
1
3
analyzed the C NMR spectrum of 2 to elucidate the tautomeric structures in comparison with that of
1
3
acyclovir and CMMG (Table 1). In the C NMR spectrum (DMSO-d ) of 2, 5-C and 6-C resonated at δ
6
9
8.6 and 153.7, respectively. The corresponding carbons for acyclovir (1) resonated at δ 116.5 and 156.9,
respectively. Although no significant differences in the chemical shifts for 6-C were observed between 2
and acyclovir, the signal due to 5-C of 2 is significantly shifted upfield by ca. 18 ppm as compared with
that of acyclovir. The 5-C of 2 resonates at approximately the same chemical shift to that of 8-OH-dG in
7
a solution of pH 6.0. While precise tautomeric ratios for 2 and oxo-2 remain unclear at this stage, the
results delineated in this section strongly suggest that 8-oxo tautomer (oxo-2) would be a presumed major
tautomer for compound (2) under the solution conditions. In a similar manner, compound (5) was also
estimated to exist as 8-oxo tautomer (oxo-5). A clear understanding of the pH-dependent tautomeric
7
equilibrium for compounds (2 and 5) must await further experimentation.
O
O
6
5
6
H
N
N
N
5
4
HN
H N
8
HN
8
OH
O
2
N
4
2
N
O
H N
2
N
2
O
OR
OR
2
5
R=H
R=Ac
oxo-2 R=H
oxo-5 R=Ac
Scheme 2
1
3
a
Table 1. Selected C NMR spectral data of acyclovir derivatives
chemical shifts, ppm
compound
2
-C
4-C
5-C
116.5
113.7
98.6
6-C
8-C
acyclovir (1)
153.9
151.1
152.4
151.0
151.5
148.6
147.8
147.4
156.9
153.9
153.7
152.1
137.9
134.9
151.3
153.4
CMMG
8
-OH-acyclovir (2)
8
-OH-acyclovir acetate (5)
98.5
a
All spectra were obtained in DMSO-d
6

1970
HETEROCYCLES, Vol. 65, No. 8, 2005
SYNTHESIS OF 9-(CABOXYMETHOXYMETHYL)GUANINE (CMMG)
5
Although CMMG was proved to be a major metabolite of acyclovir by Schaeffer et al., Wellcome
Research Laboratories, to the best of our knowledge, details of synthetic methods for CMMG are not yet
disclosed in scientific literature. We, therefore, examined a synthetic route for CMMG via
N-glycosylation of guanine with ethyl (chloromethoxy)acetate (Scheme 3). The N-glycosylation of
silylated guanine (TMSG) with ethyl (chloromethoxy)acetate has been reported by Russian chemists to
8
give ethyl ester (7) of CMMG. However, in our experimentation, the reported procedure was very
difficult to isolate 7 of high purity owing to production of a large quantity of inseparable gelatinous
products. Then we examined modified conditions for N-glycosylation of TMSG with ethyl
(
chloromethoxy)acetate.
Although the Russian chemists used dichloroethane as a solvent for the N-glycosylation reaction, we
examined several low-boiling solvents rather than dichloroethane to minimize formation of the gelatinous
products. We found the gelatinous compounds were minimally formed when TMSG, prepared from
guanine and hexamethyldisilazane (HMDS) in toluene, was treated with ethyl (chloromethoxy)acetate in
toluene at 70ºC for 12 h. The crude mixture was chromatographed on silica gel to give the desired 7 in
1
4.2% yield. We also isolated 1,9-bis(ethoxycarbonylmethoxymethyl)guanine (8) in 12.4% yield as a
gelatinous product. Although the yield of 7 was modest, the method was available for synthesis of several
grams quantities of 7. Finally, compound (7) was hydrolyzed with aqueous NaOH to give the requisite
CMMG (3) in quantitative yield.
O
O
N
O
N
N
1) HMDS
HN
H N
N
N
N
N
HN
N
EtO C
O
H
+
2
2
)ClCH OCH CO Et
2
2
2
N
H
N
2
3) H O
H N
2
2
N
2
O
CO Et
O
CO Et
2
2
6
7
8
O
N
OTMS
1
) separation
) NaOH
N
N
HN
N
2
N
H N
2
O
CO H
2
N
TMSHN
N
TMS
3
TMSG
Scheme 3
EXPERIMENTAL
All melting points are uncorrected. The NMR spectra were measured using 300 or 400 MHz
1
3
6 4
spectrometers with DMSO-d as the solvent and SiMe as the internal standard. The assignment of C

HETEROCYCLES, Vol. 65, No. 8, 2005
1971
carbon signals is based on DEPT data. IR spectra were recorded as a KBr pellet. Anhydrous toluene was
purchased from Kanto Chemical Co., Ltd. and used without further purification. Mass spectra were
recorded using electrospray ionization (ESI) techniques.
2
-Amino-8-bromo-9-[(2-hydroxyethoxy)methyl]-1,9-dihydro-6H-purin-6-one (4)
To a stirred suspension of bromine (10.55 g, 66 mmol) in water (300 mL) was added a solution of
acyclovir (13.52 g, 60 mmol) in water (2.3 L) at 40–45ºC. The mixture was stirred at rt for 30 min and
cooled with ice-cold water. The precipitates were filtered by suction and recrystallized from water to give
4
1
4
3
1
(15.9 g, 87%) as pale-yellow crystals: mp>300ºC (lit., mp ~280ºC). H NMR (300 MH, DMSO-d
6
) δ
.42-3.51 (m, 4H, 2xCH ), 4.68 (t, J=5.1 Hz, 1H, OH), 5.29 (s, 2H, NCH O), 6.62 (s, 2H, NH ), 10.73 (s,
2
2
2
1
3
H, NH). C NMR (100 MHz, DMSO-d
6
) δ 155.5 (6-C), 154.2 (2-C), 152.9 (4-C), 120.9 (8-C), 116.6
+
(
(
5-C), 72.4 (N-CH
2
-O), 70.8 (OCH
2
CH
2
OH), 59.9 (OCH
2
CH
2
OH). MS (ESI) m/z 304 (MH ). HRMS
+
ESI) calcd for C
8
H
11
N O
5 3
Br (MH ): 304.0045. Found: 304.0045.
2
-[(2-Amino-6,8-dioxo-1,6,7,8-tetrahydro-9H-purin-9-yl)methoxy]ethyl acetate (5)
A mixture of 4 (12.16 g, 40 mmol) and AcONa (24.61 g, 300 mmol) in AcOH (650 mL) was stirred under
reflux for 6 h. The volatile components of the mixture were removed in vacuo. The resulting white solids
3
were dissolved in hot water and neutralized by powdered NaHCO . The solution was cooled by ice water
and the resulting precipitates were collected and recrystallized from water to give 5 (7.53 g, 78%) as
1
colorless crystals: mp 268-269ºC (decomp). H NMR (300 MHz, DMSO-d
6
) δ 1.96 (s, 3H, CH
), 5.00 (s, 2H,
), 10.66 (s, 1H, NH), 10.69 (s, 1H, NH). C NMR (75.5 MHz, DMSO-d ) δ
70.1 (C=O), 153.4 (8-C), 152.1 (6-C), 151.0 (2-C), 147.4 (4-C), 98.5 (5-C), 68.6 (NCH O), 66.6
). IR (KBr) 1723, 1689, 1647, 1602 cm . MS (ESI) m/z 284
3
CO), 3.69
(
t with small splits, J=4.7 Hz, 2H, CH
2 2
), 4.06 (t with small splits, J=4.7 Hz, 2H, CH
1
3
NCH O), 6.50 (s, 2H, NH
2
2
6
1
2
–
1
2 2 2 2 3
(O-CH CH ), 62.9 (CH CH O), 20.6 (CH
+
(
MH ). Anal. Calcd for C H N O •H O: C, 39.86; H, 5.02; N, 23.25. Found: C, 39.88; H, 5.22; N,
1
0
13
5
5
2
2
3.47.
2
-Amino-9-[(2-hydroxyethoxy)methyl]-7,9-dihydro-1H-purine-6,8-dione (8-hydroxyacyclovir) (2)
A suspension of 5 (5.5 g, 19.4 mmol) and 2-aminoethanol (2.33 g, 36.6 mmol) in water (85 mL) was
heated under reflux for 5 h. The mixture was cooled. The resulting precipitates were collected and
recrystallized from 25% aqueous acetic acid (16 mL) to give 2 (4.45 g, 95.5%) as colorless flakes: mp
4
1
2
2
1
89-290ºC (decomp) (lit., mp 260ºC (decomp)). H NMR (300 MHz, DMSO-d
xCH ), 4.61 (t, J=5.7 Hz, 1H, OH), 4.99 (s, 2H, NCH O), 6.49 (s, 2H, NH ), 10.63 (s, 1H, NH), 10.68 (s,
H, NH). C NMR (75.5 MHz, DMSO-d ) 153.7 (6-C), 152.4 (2-C), 151.3 (8-C), 147.8 (4-C), 98.6 (5-C),
6
) δ 3.43–3.50 (m, 4H,
2
2
2
1
3
6

1972
HETEROCYCLES, Vol. 65, No. 8, 2005
–
1
7
0.8 (N-CH
2
-O), 68.9 (O-CH
2
-CH
2
), 60.1 (CH
2
-OH). IR (KBr) 1727, 1704, 1645, 1598 cm . MS (ESI)
+
+
m/z 242 (MH ). HRMS (ESI) calcd for C
8
H
12
N
5
O
4
(MH ): 242.0889. Found: 242.0901. Anal. Calcd for
8 11 5 4 2
C H N O •3/4H O: C, 37.72: H, 4.95; N, 27.50. Found: C, 37.97; H, 5.10; N, 27.56.
Ethyl (chloromethoxy)acetate
To a stirred suspension of ethyl glycolate (26 g, 250 mmol) and paraformaldehyde (6.76 g, 325 mmol) in
toluene (700 mL) was passed a stream of dry hydrogen chloride for 15 min at –10ºC. Then the mixture
2 4
was treated with Na SO (50 g) and stirred at the same temperature for 12 h. The temperature was
elevated to 0ºC. After being stirred for an additional 12 h, the solids were filtered. The volatile
components of the filtrates were removed in vacuo at below 30ºC to leave the title compound (23.5 g,
1
6
1.6%) as a colorless liquid. H NMR (300 MHz, CDCl
3
3
) δ 1.29 (t, J=7.2 Hz, 3H, CH ), 4.20 (q, J=7.2
Hz, 2H, OCH CH ), 4.28 (s, 2H, OCH CO-), 5.54 (s, 2H, ClCH O). This liquid was used for the next
2
3
2
2
reaction without further purification.
Ethyl
(2-ethoxy-2-oxoethoxy)methyl]-6-oxo-1,6-dihydro-9H-purin-9-yl}methoxy)acetate (8).
A mixture of guanine (6) (12.86 g, 85 mmol), hexamethyldisilazane (116.5 mL, 550 mmol), and
NH SO (1.0 g) in toluene (120 mL) was refluxed for 2 days. The mixture was cooled to rt. The
(2-amino-6-oxo-1,6-dihydro-9H-purin-9-yl)acetate
(7)
and
Ethyl
({2-amino-1-
[
(
4
)
2
4
resulting precipitates were filtered. The volatile components of filtrates were removed in vacuo to leave a
yellow oil, which was treated with ethyl (chloromethoxy)acetate (13 g, 85 mmol) in toluene (120 mL) at
7
0ºC for 20 h. The mixture was neutralized by Et
vacuo. The resulting yellow solid was heated in boiling water (60 mL) for 1 h. After being cooled to rt,
the solid materials (21.3 g) were collected and treated with a mixture of EtOH (50 mL) and CHCl (200
3
N (8.7 g, 86 mmol) and the solvent was removed in
3
mL) at reflux. The solid materials were removed by centrifugal separation. The supernatant was
concentrated and the residue was chromatographed on silica gel. Elution with CHCl /EtOH(4:1) gave 8
3
1
(
4.03 g, 12.4%) as white crystals: mp 191–193ºC (from 70%EtOH). H NMR (300 MHz, DMSO-d
6
) δ
CO), 4.28 (s, 2H,
), 8.10 (s. 1H, 8-H). Anal. Calcd
: C, 46.99; H, 5.52; N, 18.27. Found: C, 46.83; H, 5.59; N, 17.70. Successive elution with
/EtOH (7:3) gave white solids (3.92 g), which was recrystallized from water to give 7 (3.45 g,
1
.12–1.19 (m, 6H, 2xCH
3
), 4.02–4.13 (m, 4H, 2xOCH
2 3 2
CH ), 4.26 (s, 2H, OCH
OCH CO), 5.54 (s, 2H, NCH
2
2
O), 5.71 (s, 2H, NCH O), 7.15 (s, 2H, NH
2
2
21 5 7
for C15H N O
CHCl
3
8
1
1
4.2%) as colorless crystals: mp 223–224ºC (lit., mp 211–213ºC). H NMR (300 MHz, DMSO-d
.14 (t, J=7.2 Hz, 3H, CH ), 4.05 (q, J=7.2 Hz, 2H, OCH CH ), 4.19 (s, 2H, OCH O), 5.39 (s, 2H,
O), 6.49 (s, 2H, NH ), 7.80 (s, 1H, 8-H), 10.61 (s, 1H, NH ). C NMR (100 MHz, DMSO-d ) δ
69.4 (COOEt), 156.8 (6-C), 153.9 (2-C), 151.5 (4-C), 137.7 (8-C), 116.5 (5-C), 71.9 (N-CH -O), 65.9
6
) δ
1
3
2
3
2
1
3
NCH
2
2
2
6
1
2

HETEROCYCLES, Vol. 65, No. 8, 2005
1973
–
1
(
2 2 3 3
O-CH CO), 60.4 (OCH CH ), 13.9 (CH ). IR (KBr) 1764, 1687, 1631, 1575 cm . MS (ESI) m/z 268
+
+
(
MH ). HRMS (ESI) calcd for C10
H
14
N
5
O
4
(MH ): 268.1022. Found: 268.1066. Anal. Calcd for
C
10
H
13
N
5
O
4
•H
2
O: C, 42.10; H, 5.30; N, 24.56. Found: C, 42.31; H, 5.49; N, 24.30.
[
(2-Amino-6-oxo-1,6-dihydro-9H-purin-9-yl)methoxy]acetic acid (3).
Ethyl ester (7) (2.58 g, 9 mmol) was heated for 10 min in boiling water. After being cooled to rt, the
mixture was treated with 10% NaOH (9 mL, 22.5 mmol) for 4 h under stirring. The mixture was
3
neutralized with MeSO H (2.28 g, 24 mmol). The resulting white solids were washed with water and
f
filtered. The solid was recystallized from 25%AcOH to give 3 (2.13 g, 99%). mp>300ºC. R =0.56
1
(
(
(
CHCl /MeOH=3:1). H NMR (300 MHz, DMSO-d ) δ 4.12 (s, 2H, OCH O), 5.38 (s, 2H, NCH O), 6.51
3
6
2
2
1
3
s, 2H, NH ), 7.82 (s, 1H, 8-H), 10.61 (s, 1H, NH), 12.75 (s, 1H, COOH). C NMR (75.5 MHz) δ 168.1
2
CO
2
H), 153.9 (4-C), 151.1 (6-C), 148.6 (8-C), 134.9 (2-C), 113.7 (5-C), 69.1 (NCH
2
O), 62.9 (OCH
2
CO).
–
1
+
+
IR (KBr) 1727, 1648, 1602 cm . MS (ESI) m/z 240 (MH ). HRMS (ESI) calcd for C
8
H
10
N
5
O
5
(MH ):
2
8 9 5 4
40.0733. Found: 240.0735. Anal. Calcd for C H N O : C, 40.17; H, 3.79; N, 29.29. Found: C, 40.08; H,
3
.92; N, 28.97.
REFERENCES
1
.
D. M. Richards, A. A.Carmine, E. N. Brogden, R. C. Heel, T. M. Speight, and G. S. Avery, Drugs,
983, 26, 378.
1
2.
3.
4.
5.
J. J. O’Brien and D. M. Campoli-Richards, Drugs, 1989, 37, 233.
R. J. Whitley and J. W. Gnann, Jr., N. Engl. J. Med., 1992, 327, 782.
M. J. Robins, P. W. Hatfield, J. Balzarin, and E. Declercq, J. Med. Chem., 1984, 27, 1486.
H. J. Schaeffer, L. Beauchamp, P. de Miranda, G. B. Elion, D. J. Bauer, and P. Collins, Nature,1978,
2
72, 583.
6.
7.
8.
B. Lehmann, DE 19604101A (1997).
S. J. Culp, B. P. Cho, F. F. Kadlubar, and F. E. Evans, Chem. Res. Toxicol., 1989, 2, 416.
I. L. Shchaeleva, S. V. Kochetkova, T. L. Tsilevich, G. V. Vladyko, L. V. Korochenko, E. I. Boreko,
I. P. Smirnov, A. A. Khorlin, B. P. Gottikh, and V. L. Florent’ev, Bioorg. Khim. (Eng.Trans.), 1990,
1
5, 523.