Improved industrial syntheses of penciclovir and famciclovir using N2-acetyl-7-benzylguanine and a cyclic side chain precursor

Article abstract of DOI:10.1080/15257770600686279

We have established practical synthetic methods for penciclovir (PCV, 1) and famciclovir (FCV, 2) from N2-acetyl-7-benzylguanine (NAc7BnG, 3) and 6,6-dimethyl-5,7-dioxaspiro[2.5]octane-4,8-dione (4)-the latter being a more easily prepared cyclic precursor

Full text of DOI:10.1080/15257770600686279

This article was downloaded by: [Tulane University]
On: 29 January 2015, At: 05:49
Publisher: Taylor & Francis
Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered
office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK
Nucleosides, Nucleotides and Nucleic
Acids
Publication details, including instructions for authors and
subscription information:
http://www.tandfonline.com/loi/lncn20
Improved Industrial Syntheses of
Penciclovir and Famciclovir Using N2-
Acetyl-7-Benzylguanine and a Cyclic Side
Chain Precursor
Takayoshi Torii a , Keizo Yamashita a , Mitsuhiko Kojima a , Yumiko
Suzuki a , Toyoto Hijiya a & Kunisuke Izawa a
a Amino Science Laboratories , Ajinomoto Co., Inc. , Kawasaki-ku,
Kawasaki, Japan
Published online: 07 Feb 2007.
To cite this article: Takayoshi Torii , Keizo Yamashita , Mitsuhiko Kojima , Yumiko Suzuki , Toyoto
Hijiya & Kunisuke Izawa (2006) Improved Industrial Syntheses of Penciclovir and Famciclovir Using N2-
Acetyl-7-Benzylguanine and a Cyclic Side Chain Precursor, Nucleosides, Nucleotides and Nucleic Acids,
25:4-6, 625-634, DOI: 10.1080/15257770600686279
To link to this article: http://dx.doi.org/10.1080/15257770600686279
PLEASE SCROLL DOWN FOR ARTICLE
Taylor & Francis makes every effort to ensure the accuracy of all the information (the
“Content”) contained in the publications on our platform. However, Taylor & Francis,
our agents, and our licensors make no representations or warranties whatsoever as to
the accuracy, completeness, or suitability for any purpose of the Content. Any opinions
and views expressed in this publication are the opinions and views of the authors,
and are not the views of or endorsed by Taylor & Francis. The accuracy of the Content
should not be relied upon and should be independently verified with primary sources
of information. Taylor and Francis shall not be liable for any losses, actions, claims,
proceedings, demands, costs, expenses, damages, and other liabilities whatsoever or
howsoever caused arising directly or indirectly in connection with, in relation to or arising
out of the use of the Content.
This article may be used for research, teaching, and private study purposes. Any
substantial or systematic reproduction, redistribution, reselling, loan, sub-licensing,
systematic supply, or distribution in any form to anyone is expressly forbidden. Terms &

Conditions of access and use can be found at http://www.tandfonline.com/page/terms-
and-conditions

Nucleosides, Nucleotides, and Nucleic Acids, 25:625–634, 2006
Copyright ꢀ Taylor & Francis Group, LLC
C
ISSN: 1525-7770 print / 1532-2335 online
DOI: 10.1080/15257770600686279
IMPROVED INDUSTRIAL SYNTHESES OF PENCICLOVIR AND
FAMCICLOVIR USING N2-ACETYL-7-BENZYLGUANINE AND
A CYCLIC SIDE CHAIN PRECURSOR
Takayoshi Torii, Keizo Yamashita, Mitsuhiko Kojima, Yumiko Suzuki,
Toyoto Hijiya, and Kunisuke Izawa
AminoScience Laboratories, Ajinomoto Co.,
2
Inc., Kawasaki-ku, Kawasaki, Japan
We have established practical synthetic methods for penciclovir (PCV, 1) and famciclovir (FCV,
2
2) from N2-acetyl-7-benzylguanine (NAc7BnG, 3) and 6,6-dimethyl-5,7-dioxaspiro[2.5]octane-4,8-
dione (4)—the latter being a more easily prepared cyclic precursor of the diacetate side chain (5) used
in the conventional process. The coupling of 4 with 3 proceeded regioselectively at the N9 position
of guanine in good yield. The coupling product was then successfully transformed into the known
antiviral agents in a short number of steps.
Keywords Famciclovir; Industrial synthesis; Penciclovir
INTRODUCTION
PCV (1) and its pro-drug FCV (2) are found to be potent and highly
selective antiviral agents against both the herpes simplex virus (HSV) and
the varicella-zoster virus (VZV) (Figure 1).[1] It has also been reported that 1
and 2 exhibit anti–hepatitis B virus (HBV) activity.[2] PCV (1) and FCV (2) are
analogues of acyclovir, having an alkyl side chain at the N9 position. Boryski
et al. found that the N7 isomer of diacetylacyclovir could be converted to a 1:1
mixture of the N7 and N9 isomers under acidic conditions via an oxonium
ion intermediate.[3] By optimizing the reaction conditions, we eventually
succeeded in obtaining an almost pure N9 isomer of acyclovir in one pot.[4]
However, we could not use this same methodology to produce 1 and 2 since
they do not have an oxygen atom in their side chain.[5] Accordingly, we set
about establishing a new strategy for the N9 selective alkylation of guanine.
Received 19 January 2006; accepted 21 February 2006.
This article is dedicated to Professor Eiko Ohtsuka on the occasion of her 70th birthday.
Address correspondence to Kunisuke Izawa, AminoScience Laboratories, Ajinomoto Co., Inc.
1-1, Suzuki-cho, Kawasaki-ku, Kawasaki-shi, Kanagawa, 210-8681, Japan. E-mail: kunisuke izawa@
ajinomoto.com
625

626
T. Torii et al.
FIGURE 1 PCV (1) and FCV (2).
Normally, it is difficult to introduce an alkyl side chain at the N9 position
of guanine with high regio-selectivity. In order to achieve selective N9 alky-
lation, 2-amino-6-halopurines are commonly used as starting materials for
coupling with alkyl halide side chains.[6] However, 2-amino-6-halopurines are
not ideal candidates for large scale synthesis due to their high mutagenicity.
In addition, to obtain highly pure N9 compounds, N7 compounds should be
eliminated from the N9/N7 mixture by means of column chromatography.
To overcome these problems, we established and reported a new method-
ology involving the N9 selective alkyation of NAc7BnG (3), which can be
readily prepared from guanosine.[7]
Geen et al. reported that the commercially available cyclic side chain
precursor 4 was most suitable for coupling with 2-amino-6-chloropurine for
the synthesis of 1 and 2.[6f] The diacetate 5, which is commonly used for the
synthesis of 1 and 2 as the side chain precursor, can be prepared in 7 steps
from ethylene glycol in 45% yield.[6c,8] On the other hand, the side chain
precursor 4[9] can be prepared from 1,2-dibromoethane in 74% yield in only
3 steps (Figure 2).[9,10] We concluded that the method using 4 as a side chain
precursor would be better than the method using 5 despite the fact that it
requires extra 2 steps to obtain 1 and 2 from NAc7BnG 3. We wish to describe
herein a practical synthesis of 1 and 2 using NAc7BnG 3 and the side chain
precursor 4.[11]
RESULTS AND DISCUSSION
Firstly, we studied the coupling reaction of NAc7BnG 3 with the side
chain precursor 4 (see Table 1). The coupling reaction of 3 with 4 scarcely
proceeded at all using NaHCO3 in DMF at room temperature under the same
conditions as the coupling of 2-amino-6-chloropurine with 4 as reported by
FIGURE 2 Side chain precursors 4 and 5.

Improved Industrial Syntheses
627
TABLE 1 Coupling of NAc7BnG 3 and the Cyclic Analogue Side Chain Precursor 4
Isolated yield
of crude 6 (%) of 6 (%)
Net yield
Purity of
crude 6 (wt%)
a
b
Entry
Solvent (M)
Temp. (◦C)
4 (eq)
1
2
3
4
5
6
7
8
DMF (0.6)
NMP (1.6)
DMSO (1.3)
DMF (0.5)
DMF (0.5)
DMF (0.5)
60
60
60
60
60
60
60
70
1.0
1.0
1.0
1.0
2.0
79
57
45
49
92
135
148
74
53
50
27
38
61
76
86
74
67
87
60
78
66
56
58
99
3.0
DMF (0.5)
5:1 DMF-H2O (1.5)
1.5 + 1.5c
1.2
aNet yields were determined by HPLC a bsolute calibration method.
bPurity of crude 6 were calculated by the isolated yield of crystals 6 and net yield of 6.
cCompound 4 was added twice.
Geen et al.[6f] On the other hand, the coupling product 6 was observed in
DMF at 60◦C without using a base (entry 1), which were the same conditions
as the coupling of 3 with 5.[12] It had been reported that the coupling of
pyridine with 4 proceeded quantitatively at room temperature.[9] However,
the coupling of 3 with 4 had to be carried out at over 60◦C, probably because
of the lower nucleophilicity of 3. Although the yield of 6 was improved from
38 to 86% by adding three equivalents of 4, it is still desirable to avoid the
use of excessive amounts of 4 for reasons both of cost and environmental
protection. After several unsuccessful attempts, we discovered that the yield
of 6 was substantially improved to 74% by adding a small amount of water
using only 1.2 equivalents of 4 (entry 8). The reason for the improvement
of the yield is not clear at this point in time. After the coupling reaction,
compound 6 was obtained in crystalline form in good yields by adding ethyl
acetate and filtration. When an extra amount of 4 was used for the coupling
reaction, the purity of 6 was only 58% (entry 7). In the case of entry 8,
however, compound 6 was obtained in 99% purity by the simple purification
method described above; the major impurity of the crude crystal 6 being in
all probability the polymerization product of 4. Nevertheless, we have faced
severe problems such as poor filterability of 6 in the scale-up study of entry
8. The main reason for the poor filterability might be that the reaction was
obliged to be carried out under highly concentrated conditions in order
to maintain the yield of 6 as same level as entry 7. Therefore, we selected
to use the crude crystals of 6 without further purification for the next step
because the crude crystals of 6 have better physical property and the impurity

628
T. Torii et al.
SCHEME 1 Debenzylation of 6.
contained in the crude crystals of 6 could be removed easily at the following
step.
Next, we studied the debenzylation of 6 (Scheme 1). The coupling prod-
uct 6 was transformed into the debenzylated compound 7 under a hydrogen
atmosphere in the presence of Pd catalyst and base. The reaction mixture
changed from a slurry to a solution during the progress of the reaction. At
that time, several products were observed in the reaction mixture. We con-
sidered that a part of the acetyl group of 7 at the N2 position had already
undergone deprotection under these reaction conditions. Therefore, the
mixture was further treated under strong basic conditions for completion
of the deacetylation of 7. After neutralization of the reaction mixture, com-
pound 8 was obtained as crystals in 86% yield from 6. However, we failed
to isolate pure 8 when we subsequently scaled up this synthesis due to its
inherently poor filterability. Therefore, the reaction mixture containing 8
was then further treated under acidic conditions at pH 3.5 to obtain the di-
carboxylate 9. Fortunately, compound 9 exhibited good filterability and was
obtainable in a 90% isolated yield from 6. Based on HMBC NMR studies,

Improved Industrial Syntheses
629
SCHEME 2 Preparation of PCV 1.
compound 9 was identified as the N9 alkylated compound. Thus, the pro-
ton signal of 9 at 4.00 ppm (H-1ꢁ) showed HMBC correlation peaks with the
purine base carbons only at 137.63 (C-8) and 151.48 ppm (C-4) indicating
that the side chain was connected to the N9 nitrogen atom of purine base.
When the coupling product 6 was debenzylated under acidic conditions aim-
ing at the deprotection of the isopropylidene group at the same time, the
decarboxylated compound 10 was produced as a byproduct. Since it was very
difficult to remove 10 from 9, we did not pursue the deprotection of 6 under
acidic conditions.
The dicarboxylic acid 9 thus obtained was converted to the diester 11 in
95% yield by treating with thionyl chloride in methanol. Penciclovir 1 was ob-
tained by the sodium borohydride reduction of 11 in 77% yield (Scheme 2).
This method using NAc7BnG 3 as a starting material gave 1 in 57% overall
yield and 4 isolation steps. It may thus be considered as superior to that using
2-amino-6-chlorpurine as a starting material which affords 1 in 23% yield and
4 steps.[6f] Attempts at the direct reduction of 8 did not succeed since 8 is
poorly soluble in most solvents.
Next, we turned our attention to the synthesis of famciclovir 2 from 11.
The diester 11 was treated with phosphorus oxychloride in the presence of
tetraethylammonium chloride and triethylamine at 80◦C to give 12 in 70%
yield. Compound 12 was transformed into 2 in 53% yield by the procedure
described in the literature (Scheme 3).[6f] This method using NAc7BnG 3 as
SCHEME 3 Preparation of FCV 2.

630
T. Torii et al.
a starting material gave famciclovir 2 in 27% overall yield, which might be bet-
ter than the method using 2-amino-6-chlorpurine as a starting material (12%
yield) even though it requires extra 2 steps.[6f] It also has additional advan-
tages in as much as it avoids the use of the mutagenic 2-amino-6-chloropurine
and the need to purify intermediates by chromatogrophical means.
CONCLUSION
We discovered that the cyclic side chain precursor 4 may be readily cou-
pled with NAc7BnG 3 at the N9 position selectively. The coupling product 6
is easily and efficiently converted to known antiviral agents PCV 1 and FCV 2.
This is also a highly practical synthetic method in that it avoids the need for
purification by chromatogrophical means.
EXPERIMENTAL
All reagents were purchased and used without further purification.
Thin-layer chromatography (TLC) was conducted on precoated TLC plates
(Merck 60F254). High-performance liquid chromatography (HPLC) was
performed with a Hitachi L-6000 pump and L-4000 UV detector system using
a YMC-ODS-A column. Melting points were measured with the Bu¨chi B-545.
NMR spectra were obtained on a BRUKER Advance-400 MHz spectrome-
ter. All [1]H-NMR spectra were measured in DMSO-d6 solvent, and chemical
shifts are reported as δ values in parts per million relative to tetramethylsi-
lane (δ 0.00) as an internal standard. All [13]C-NMR spectra were measured
in DMSO-d6 solvent, and chemical shifts are reported as δ values in parts per
million relative to DMSO-d6 (δ 39.5) as an internal standard. Infrared (IR)
spectra were recorded on an ASI React IR 1000 FT-IR spectrometer with an
ATR sampling system and are reported in wavenumber (cm−1). Mass spectra
(MS) and High-resolution mass spectra (HRMS) were obtained with JEOL
JMS-700V (JEOL datum Ltd.).
Crude 2-Acetyl-7-benyl-9-[2-(2,2-dimethyl-4,6-dioxo-1,3-dioxan-5-yl)ethyl]guani-
nium inner salt (6).[7a] A mixture of N2-acetyl-7-benzylguanine (3) (497 mg,
1.68 mmol) and 6,6-dimethyl-5,7-dioxaspiro[2,5]octane-4,8-dione (4) (430
mg, 2.53 mmol) in DMF (3 ml) was stirred for 24 h at 60◦C. Then to the
mixture was added 4 (430 mg, 2.53 mmol), which was stirred for 21 h at
60◦C. AcOEt (6 ml) was added to the mixture, filtered off and washed
with AcOEt (10 ml). After drying, compound 6 (1.12 g, 58 wt%, 86%) was
obtained as a pale yellowish powder.
2-Acetyl-7-benyl-9-[2-(2,2-dimethyl-4,6-dioxo-1,3-dioxan-5-yl)ethyl]guaninium
inner salt (6). A solution of N2-acetyl-7-benzylguanine (3) (5.72 g,
20.0 mmol) in DMF (11.1 ml) and water (2.2 ml) was warmed to 70◦C.
Then to the mixture was added 6,6-dimethyl-5,7-dioxaspiro[2,5]octane-
4,8-dione (4) (4.08 g, 24.0 mmol) and stirred 8 h at 70◦C. DMF (6.7 ml)

Improved Industrial Syntheses
631
and AcOEt (140 ml) were added to the mixture and stirred for 13 h at
room temperature. The mixture was filtered off and washed with AcOEt
(110 ml). After drying, compound 6 (6.81 g, 74%) was obtained as colorless
1
crystals. H-NMR (400 MHz, DMSO-d6)δ 1.36 (6H, s, CH3 × 2), 2.20 (3H,
s, Ac), 2.64 (2H, t, J = 6.2 Hz, H-2ꢁ), 4.22 (2H, t, J = 6.2 Hz, H-1ꢁ), 5.64
(2H, s, CH2), 7.35–7.38 (3H, m, aryl), 7.45–7.48 (2H, m, aryl), 9.68 (1H,
s, H-8); 13C-NMR (100 MHz, DMSO-d6)δ 24.22, 24.56, 26.18, 40.73, 46.46,
51.80, 67.18, 99.52, 110.29, 128.48, 128.83, 129.10, 134.72, 140.44, 148.41,
150.80, 165.76, 174.37; IR (neat) 3080, 3000, 1737, 1704, 1565, 1410, 1248,
1206, 1187, 1135, 851, 774, 695 cm−1; mp 194.5–195.4◦C; MS (FAB+) m/z
454 (MH+); HRMS (FAB+) calc for C22H24N5O6 (MH+), 454.1727, found
454.1747.
5-[2-(Guanin-9-yl)ethyl]-2,2-dimethyl-1,3-dioxacyclohexane-4,6-dione (8).
A
mixture of 6 (2.40 g, 57 wt%, 3.0 mmol), Na2CO3 (0.643 g, 6.0 mmol),
and 5% Pd-C (0.277 g, 46.9 wt%) was stirred at 50◦C in CH3OH (10 ml) with
water (10 ml) under hydrogen atmosphere of 1 atom pressure for 24 h. To
the reaction mixture was added 4 M NaOH aq (2.3 ml) and CH3OH (6 ml)
and it was then filtered through Celite and the filter bed washed with 50%
CH3OH aq (10 ml). To the combined filtrates evaporated to 5.8 was added
2 M NaOH aq (1.0 ml) warmed to 60◦C for 2 h. The solution was cooled
to 0◦C, neutralized to pH 7.0 with 2 M HCl aq., and stirred for 14 h at 0◦C.
The mixture was filtered off and washed with water (1 ml). After drying,
compound 8 (0.836 g, 86%) was obtained as colorless crystals. 1H-NMR (400
MHz, DMSO-d6)δ1.39 (6H, s, CH3 × 2), 2.64 (2H, t, J = 7.3 Hz, H-2ꢁ), 3.90
(2H, t, J = 7.3 Hz, H-1ꢁ), 6.44 (2H, br, NH2), 7.50 (1H, s, H-8), 10.49 (1H,
br, NH); 13C-NMR (100 MHz, DMSO-d6)δ 25.73, 26.26, 42.72, 68.35, 99.20,
116.77, 137.54, 151.50, 153.60, 157.26, 165.50; IR (neat) 3330, 3176, 1686,
1644, 1553, 1403, 1260, 1196, 1111, 930, 777, 691 cm−1; mp 247◦C (dec.);
MS (FAB−) m/z 320 (MH−); HRMS (FAB−) calc for C13H14N5O5 (MH−),
320.1000, found 320.0990.
2-[2-(Guanin-9-yl)ethyl]malonic acid (9).[1] A mixture of 6 (33.46 g, 61 wt%,
45 mmol), NaHCO3 (3.78 g, 45.0 mmol), and 5% Pd-C (4.05 g, 46.9 wt%,
0.9 mmol) was stirred at 50◦C in CH3OH (225 ml) with water (75 ml) under
hydrogen atmosphere of 1 atom pressure for 19 h. The reaction mixture
was then filtered through Celite, the filter bed washed with 50% CH3OH aq
(130 ml), and the combined filtrates evaporated to 70 g. To the solution was
added 4 M NaOH aq (26 ml) which was warmed to 60◦C for 2 h. The solution
was acidified to pH 2.5 with 6 M HCl aq, water added (130 ml) and stirred for
5 h at 65◦C. The solution was cooled to 0◦C, its pH being controlled to 3.5 with
4M NaOH aq. It was then stirred for 3 h at 0◦C and the precipitate filtered
off, washed with cold 50% CH3OH aq (18 ml) and CH3OH (15 ml). After
1
drying, compound 9 (11.81 g, 90%) was obtained as colorless crystals. H-
NMR (400 MHz, DMSO-d6)δ 2.15-2.21 (2H, m, H-2ꢁ), 2.95 (1H, t, J = 6.6 Hz,
H-3ꢁ), 4.00 (2H, t, J = 7.4 Hz, H-1ꢁ), 6.45 (2H, br, NH2), 7.63 (1H, s, H-8),

632
T. Torii et al.
10.52 (1H, br, NH); 13C-NMR (100 MHz, DMSO-d6)δ 29.55, 41.63, 48.00,
116.95, 137.63, 151.48, 153.82, 157.16, 172.29; IR (neat) 3122, 2744, 1706,
1603, 1383, 1185, 839, 687 cm−1; mp 258◦C (dec); MS (FAB-) m/z 280 (MH−);
HRMS (FAB-) calc for C10H10N5O5 (MH−), 280.0687, found 280.0670.
4-(Guanin-9-yl)ethyl]butanoic acid (10). A mixture of 6 (2.40 g, 57 wt%,
3.0 mmol), 2 M HCl aq (1.66 ml, 3.3 mmol), and 5% Pd-C (0.277 g, 46.9
wt%) was stirred at 50◦C in CH3OH (10 ml) with water (8.34 ml) under
hydrogen atmosphere of 1 atom pressure for 24 h. To the reaction mixture
was added 4 M NaOH aq (3.1 ml) and CH3OH (8 ml) and it was then filtered
through Celite and the filter bed washed with 50% CH3OH aq (10 ml). The
combined filtrates were evaporated to 4.9 g, then 2 M NaOH aq (0.7 ml)
added and it was warmed to 60◦C for 2 h. The solution was cooled to 0◦C,
neutralized to pH 5.2 with 2 M HCl aq and then stirred for 14 h at 0◦C. The
mixture was filtered off and washed with water (1 ml). The solid (0.509 g)
was purified by chromatography on silica gel (CH2Cl2/CH3OH, 4:1) to give
9 (0.396 g, 47%) and 10 (16.5 mg, 2.3%) as colorless crystals. 1H-NMR (400
MHz, DMSO-d6)δ 1.92-2.00 (2H, m, H-2ꢁ), 2.21 (2H, t, J = 7.4 Hz, H-3ꢁ), 3.96
(2H, t, J = 7.0 Hz, H-1ꢁ), 6.49 (2H, br, NH2), 7.73 (1H, s, H-8), 10.01 (1H, br,
NH); 13C-NMR (100 MHz, DMSO-d6)δ 25.19, 31.06, 42.54, 116.57, 137.72,
151.45, 153.95, 157.00, 174.00; IR (neat) 3348, 3170, 2672, 1702, 1412, 1189,
778, 758, 695, 656 cm−1; mp 275.4–277.4◦C; MS (FAB−) m/z 236 (MH−);
HRMS (FAB−) calc for C9H10N5O3 (MH−), 236.0789, found 236.0807.
Dimethyl 2-[2-(guanin-9-yl)ethyl]malonate (11).[11] To a CH3OH (40 ml) so-
lution in an ice bath, thionyl chloride (6.15 ml, 80.0 mmol) was added. After
adding, the mixture was warmed to rt. To the solution was added 9 (11.62 g,
40.0 mmol) and it was heated for 3.5 h at 40◦C and then for 22.5 h at 45◦C.
After cooling to room temperature, CH3OH (100 ml) was added and the
mixture was cooled in an ice bath. The solution was neutralized with 4 M
NaOH aq. (ca. 39.5 ml), heated for 0.5 h at 55◦C, and then cooled to 16◦C
until crystals began to separate. The solid was collected on filter and washed
with 50% CH3OH aq (15 ml), CH3OH (15 ml). The crystals were suspended
in 50% CH3OH aq (180 ml) and stirred 2 h at rt. The solid was collected on
filter and washed with 50% CH3OH aq (20 ml), CH3OH (15 ml). After dry-
ing, compound 11 (11.74 g, 95%) was obtained as colorless crystals. 1H-NMR
(400 MHz, DMSO-d6)δ 2.35 (2H, q, J = 7.1 Hz, H-2ꢁ), 3.65 (6H, s, Ac × 2),
3.66 (1H, t, J = 7.2 Hz, H-3ꢁ), 4.16 (2H, t, J = 6.9 Hz, H-1ꢁ), 7.24 (2H, br,
NH2), 8.90 (1H, s, H-8), 11.66 (1H, br, NH); 13C-NMR (100 MHz, DMSO-d6)δ
27.90, 42.44, 48.47, 52.97, 109.53, 137.49, 150.37, 154.24, 155.62, 169.11; IR
(neat) 3465, 3182, 2634, 1731, 1642, 1605, 1395, 1248, 1185, 1164, 911, 778,
687 cm−1; mp 226◦C (dec); MS (FAB+) m/z 310 (MH+); HRMS (FAB+) calc
for C12H15N5O5 (MH+), 310.1151, found 310.1170.
Penciclovir (1). To a mixture of 11 (3.25 g, 10.0 mmol) and NaBH4 (1.14
g, 30 mmol) in 2-propanol (26.5 ml), water (6.6 ml) was added over 5 min.

Improved Industrial Syntheses
633
The solution was stirred for 2.5 h at 40◦C and cooled to room temperature.
The solution was quenched by 6 M HCl aq (4.5 ml) and stirred for 1.5 h at
room temperature. After adding 25% NaOH aq (10 ml) and water (15 ml),
the solution was evaporated until the 2-propanol was removed and it was
then neutralized to pH 6.0 with 6 M HCl aq, cooled to 0◦C, and stirred for
2.5 h. The precipitate was filtered off, washed with cooled water (2 ml × 2),
and crude 1 (2.18 g, 7.84 mmol, 78%) was then obtained as a colorless solid.
The solid was crystallized from water to afford 1 (2.09 g, 77%) as colorless
crystals. 1H-NMR (400 MHz, DMSO-d6)δ 1.41–1.48 (1H, m, H-3ꢁ), 1.67–1.74
(2H, m, H-2ꢁ), 3.32–3.38 (2H, m, H-4ꢁ-a), 3.39–3.46 (2H, m, H-4ꢁ-b), 4.00
(2H, t, J = 7.4 Hz, H-1ꢁ), 4.41 (2H, t, J = 5.2 Hz, OH × 24), 6.41 (2H, br,
NH2), 7.68 (1H, s, H-8), 10.50 (1H, br, NH); 13C-NMR (100 MHz, DMSO-d6)δ
29.15, 41.11, 41.41, 61.67, 116.94, 137.68, 151.46, 153.76, 157.17; IR (neat)
3411, 3128, 2881, 2667, 1675, 1602, 1397, 1196, 1054, 992, 781, 753 cm−1
;
mp 269.6–272.3◦C; MS (FAB+) m/z 254 (MH+); HRMS (FAB+) calcd for
C10H16N5O3 (MH+), 254.1253, found 254.1230.
Dimethyl 2-[2-(2-amino-6-chloropurin-9-yl)ethyl]malonate (12).[6f] To a mix-
ture of 11 (0.62 g, 2.0 mmol) and tetraethylammonium chloride (0.665 g, 4.0
mmol) in CH3CN (5 ml), N,N-dimethylaniline (0.134 ml, 1.0 mmol), and
phosphorus oxychloride (0.835 ml, 9.0 mmol) were added. The solution was
stirred for 1 h at 80◦C, and evaporated. The residue was diluted by CH2Cl2
(9 ml). After cooling to 0◦C, to the solution was added 1 M NaOH aq (5 ml)
and water (1 ml); it was then stirred vigorously at 0◦C, and separated. The
water layer was extracted by CH2Cl2 (4 ml). The combined CH2Cl2 layer
was washed with sat. NaHCO3 aq (6 ml) and water (6 ml) and evaporated
in vacuo. The residue was crystallized from CH3OH to afford 12 (0.462 g,
70%) as a pale yellowish powder. 1H-NMR (400 MHz, DMSO-d6)δ 2.33 (2H,
q, J = 7.0 Hz, H-2ꢁ), 3.60 (6H, s, Ac × 2), 3.54 (1H, t, J = 7.2 Hz, H-3), 4.12
(2H, t, J = 6.8 Hz, H-1ꢁ), 6.90 (2H, br, NH2), 8.08 (1H, s, H-8); 13C-NMR
(100 MHz, DMSO-d6)δ 28.25, 41.30, 48.79, 52.88, 123.69, 143,47, 149.65,
154.49, 160.08, 169.12; IR (neat) 3462, 3311, 3207, 1739, 1629, 1609, 1563,
1312, 1214, 1152, 973, 913, 783, 643 cm−1; mp 140.6–142.9◦C; MS (FAB+)
m/z 328 (MH+); HRMS (FAB+) calc for C12H15ClN5O4 (MH+), 328.0813,
found 328.0819.
REFERENCES
1. (a) Jarvest, R.L.; Harnden, M.R. Antiviral guanine derivatives, EP patent 141927, 1984. (b) Harnden,
M.R.; Jarvest, R.L. Purine derivatives and their pharmaceutical use, EP patent 182024, 1985. (c)
Tippie, M.A.; Martin, J.C.; Smee, D.F.; Matthews, T.R.; Verheyden, J.P.H. Antiherpes simplex virus
activity of 9-[4-hydroxy-3-(hydroxymethyl)-1-butyl]guanine. Nucleosides & Nucleotides 1984, 3, 525–
535.
2. It was discontinued to develop FCV, 2 for the treatment of HBV infection due to poor clinical
performance. See SmithKline Beecham plc, Press Release, April 20, 1999.

634
T. Torii et al.
3. (a) Boryski, J.; Golankiewicz, B. Transglycosylation of guanine nucleosides. Nucleosides & Nu-
cleotides 1987, 6, 385–386. (b) Boryski, J.; Golankiewicz, B. Application of the transpurination
reaction to synthesis of acyclic guanosine analogues. Nucleosides & Nucleotides 1989, 8, 529–536.
4. (a) Shiragami, H.; Koguchi, Y.; Izawa, K. Process for producing acyclic nucleosides, EP patent 532878,
1992. (b) Shiragami, H.; Koguchi, Y.; Tanaka, Y.; Takamatsu, S.; Uchida, Y.; Ineyama, T.; Izawa, K. Syn-
thesis of 9-(2-hydroxyethoxymethyl)guanine (acyclovir) from guanosine. Nucleosides & Nucleotides
1995, 14, 337–340.
5. Geen et al. developed a unique N7 to N9 isomerization procedure by using the α-acetoxyfuran
side chain precursor, though the yield of their method was low in terms of transforming it to PCV
and FCV. See Geen, G.R.; Kincey, P.M.; Spoors, P.G. Regioselective alkylation of guanines using
2-acetoxytetrahydrofurans. Tetrahedron Letters 2001, 42, 1781–1784.
6. (a) Pandit, U. K.; Grose, W.F.A.; Eggelte, T.A. New class of nucleoside analogues. Synthesis of
N1-pyrimidinyl- and N9-purinyl-4ꢁ-hydroxy-3-(hydroxymethyl)butanes. Synthetic Communications
1972, 2, 345–351. (b) Harnden, M.R.; Jarvest, R.L. An improved synthesis of the antiviral acyclonucle-
oside 9-(4-hydroxy-3-hydroxymethylbut-1-yl)guanine. Tetrahedron Letters 1985, 26, 4265–4268. (c)
Geen, G.R.; Grinter, T.J.; Kincey, P.M.; Jarvest, R.L. The effect of the C-6 substituent on the regioselec-
tivity of N-alkylation of 2-aminopurines. Tetrahedron 1990, 46, 6903–6914. (d) Geen, G.R.; Kincey,
P.M.; Choudary, B.M. Regiospecific Michael additions with 2-aminopurines. Tetrahedron Letters
1992, 33, 4609–4612. (e) Choudary, B.M.; Geen, G.R.; Grinter, T.J.; MacBeath, F.S.; Parratt, M.J. In-
fluence of remote structure upon regioselectivity in the N-alkylation of 2-amino-6-chloropurine: Ap-
plication to the synthesis of penciclovir. Nucleosides & Nucleotides 1994, 13, 979–996. (f) Choudary,
B.M.; Geen, G.R.; Kincey, P.M.; Parratt, M.J.; Dales, J.R.M.; Johnson, G.P.; O’Donnell, S.; Tudor,
D.W.; Woods, N. A. direct approach to the synthesis of famciclovir and penciclovir. Nucleosides &
Nucleotides 1996, 15, 981–994. (g) Freer, R.; Geen, G.R.; Ramsay, T.W.; Share, A.C.; Slater, G.R.;
Smith, N.M. A new route to famciclovir via palladium catalyzed allylation. Tetrahedron 2000, 56,
4589–4595. (h) Toyokuni, T.; Walsh, J.C.; Namavari, M.; Shinde, S.S.; Moore, J.R.; Barrio, J.R.;
Satyamurthy, N. Selective and practical synthesis of Penciclovir. Synthetic Communications 2003,
33, 3897–3905.
7. (a) Shiragami, H.; Uchida, Y.; Izawa, K.; Yamashita, K.; Katayama, S. Process for producing purine
derivatives, EP patent 827960, 1998. (b) Izawa, K.; Shiragami, H.; Practical syntheses of antiviral
nucleosides. Pure & Applied Chemistry 1998, 70, 313–318.
8. (a) Caldwell, J.J.; Colman, R.; Kerr, W.J.; Magennis, E.J. Novel use of a selenoalkyne within untra-
ditionally mild Do¨tz benzannulation processes; Total synthesis of a Calceolaria anadina L. natural
hydroxylated naphtoquinone. Synlett 2001, 1428–1430. (b) Rastetter, W.H.; Phillion, D.P.; Syntheses
of four thiol-substituted crown ethers. Journal of Organic Chemistry 1981, 46, 3204–3208.
9. Danishefsky, S.; Singh, R.K.A. Highly activated cyclopropane for homoconjugate reactions. Journal
of the American Chemical Society 1975, 97, 3239–3241.
10. (a) Dmowski, W.; Wolniewicz, A. Selective reactions of 1,1-cycloalkanedicarboxylic acids with SF4. A
route to 1,1-bis(trifluoromethyl)cycloalkanes, 1-fluoroformyl-1-(trifluoromethyl)cycloalkanes and 1-
(trifluoromethyl)-1-cycloalkanecarboxylic acids. Journal of Fluorine Chemistry, 2000, 102, 141–146.
(b) Abell, P.I.; Tien, R.; Kinetics of the decarboxylation of some 1,1-cycloalkanedicarboxylic acids.
Journal of Organic Chemistry 1965, 30, 4212–4215.
11. We had already reported preliminary results of this study. See Hijiya, T.; Yamashita, K.; Kojima, M.;
Uchida, Y.; Katayama, S.; Torii, T.; Shiragami, H.; Izawa, K. An economical synthesis of famciclovir.
Nucleosides & Nucleotides 1999, 18, 653–654.
12. Izawa, K.; Shiragami, H.; Yamashita, K. Process for producing purine derivatives, EP patent 728757,
1996.