A universal biocatalyst for the preparation of base- and sugar-modified nucleosides via an enzymatic transglycosylation

Article abstract of DOI:10.1002/1522-2675(200207)85:7<1901::AID-HLCA1901>3.0.CO;2-C

The E. coli BMT-4D/1A cells have been selected according to Munch-Petersen et al. They carry two regulatory mutations (cytR and deoR) and are able to synthesize constitutively nucleoside-catabolizing enzymes, e.g., cells that possess high UPase and PNPase activities. The cells have been cross-linked by glutaraldehyde to afford a biocatalyst that retained high UPase and PNPase activities and was comfortable for repeated use. An incubation of 2′-deoxyguanosine (1) and 2-chloroadenine (2) (molar ratio 3:1) in K-phosphate buffer (10 mM; pH 7.0) in the presence of the biocatalyst at 65° for 7 h resulted in quantitative transformation of 2 into 2-chloro-2′-deoxyadenosine (4; cladribine) that was isolated in 81% yield (Scheme 1). Similarly, the reaction of guanosine (5) and 1,2,4-triazole-3-carboxamide (6) (molar ratio 1:1) in K-phosphate buffer (10mM; pH 7.0) in the presence of the biocatalyst at 60° for 30h led to the formation of 1-(β-D-ribofuranosyl)-1,2,4-triazole-3-carboxamide (8; ribavirin) in 90-92% yield (67-70% isolated yield) (Scheme 2).

Full text of DOI:10.1002/1522-2675(200207)85:7<1901::AID-HLCA1901>3.0.CO;2-C

Helvetica Chimica Acta Vol. 85 (2002)
1901
A Universal Biocatalyst for the Preparation of Base- and
Sugar-Modified Nucleosides via an Enzymatic Transglycosylation
by Vladimir N.Barai* a), Anatoli I.Zinchenko a), Ludmilla A.Eroshevskaya a), Elena N.Kalinichenko b),
b
Tamara I.Kulak b), and Igor A.Mikhailopulo*
)
a) Institute of Microbiology, National Academy of Sciences, 220141 Minsk, Acad. Kuprevicha 2, Belarus
(e-mail: zinbar@mbio.bas-net.by)
b) Institute of Bioorganic Chemistry, National Academy of Sciences, 220141 Minsk, Acad. Kuprevicha 5,
Belarus (phone/fax: ‡375/172/648148; e-mail: igormikh@iboch.bas-net.by)
Dedicated to Prof. Dr. Wolfgang Pfleiderer on the occasion of his 75th birthday
The E. coli BMT-4D/1A cells have been selected according to Munch-Petersen et al. They carry two
regulatory mutations (cytR and deoR) and are able to synthesize constitutively nucleoside-catabolizing
enzymes, e.g., cells that possess high UPase and PNPase activities. The cells have been cross-linked by
glutaraldehyde to afford a biocatalyst that retained high UPase and PNPase activities and was comfortable for
repeated use. An incubation of 2'-deoxyguanosine (1) and 2-chloroadenine (2) (molar ratio 3 :1) in K-
phosphate buffer (10 mm; pH 7.0) in the presence of the biocatalyst at 658 for 7 h resulted in quantitative
transformation of 2 into 2-chloro-2'-deoxyadenosine (4; cladribine) that was isolated in 81% yield (Scheme 1).
Similarly, the reaction of guanosine (5) and 1,2,4-triazole-3-carboxamide (6) (molar ratio 1 :1) in K-phosphate
buffer (10 mm; pH 7.0) in the presence of the biocatalyst at 608 for 30 h led to the formation of 1-(b-d-
ribofuranosyl)-1,2,4-triazole-3-carboxamide (8; ribavirin) in 90 92% yield (67 70% isolated yield)
(Scheme 2).
Introduction. Nucleoside analogues are a mainstream in the treatment of many
viral infections and play a significant role in the chemotherapy of different
malignancies [1 4]. Most of them are produced by chemical methods employing the
natural nucleosides as starting compounds, or by a convergent approach via the
condensation of the carbohydrate precursor and heterocyclic base. The latter strategy
suffers from many drawbacks, most serious of which are the low stereospecificity of the
glycosidic-bond formation and ambiguous regiospecificity, especially in the case of
purine bases. As a result, chemical schemes of nucleoside-analogue preparation are
rather complicated and accompanied by a tedious isolation of the individual desired
compounds [5].
Considerable progress in the preparation of nucleoside analogues under consid-
eration was achieved by combination of chemical methods and biochemical trans-
formations (for reviews, see [6]). The enzymatic equilibrium-transfer reaction of the
sugar moiety of one nucleoside to the heterocyclic base was discovered by Krenitsky in
the 60×s [7][8]. It was established that this equilibrium reaction is catalyzed by purine
nucleoside phosphorylase (PNPase) in the case of purine nucleosides and purine bases,
and by uridine (thymidine) phosphorylases (UPase or TPase) in the case of pyrimidine
nucleosides and pyrimidines. More than a decade later, it was shown that this reaction
can be used for the preparation of base- and sugar-modified nucleosides [9 13].

1902
Helvetica Chimica Acta Vol. 85 (2002)
Employment of enzymes or the whole cells as biocatalysts results in a principal
improvement of key stages of the preparation of nucleoside analogues, viz., allowing
fulfillment of the stereo- and regioselective formation of the glycosidic bond. From the
published enzymatic nucleoside syntheses, it becomes obvious that the application of
chemical-enzymatic methodology for the preparation of nucleosides of medicinal
importance may essentially contribute to an improvement of cost-effectiveness ratios
vs. current chemical processes. At the same time, despite the widespread appreciation
of benefits of −chemical biotechnology×, a number of problems should be overcome in
order to displace synthetic methods, which have, in some cases, a highly established
position. Biocatalysts should possess higher enzymatic efficiency and operational
stability, and need to be more cost-effective.
Two approaches have successfully been employed for the pentofuranosyl-transfer
reaction catalyzed by nucleoside phosphorylases: one of these makes use of the purified
enzymes, the other is based on the use of the whole bacterial cells displaying the UPase
and/or TPase, and PNPase activities as a biocatalyst [6]. During the course of our
studies directed towards the search of bacterial cells manifesting the high UPase or/and
TPase and PNPase activities, we have selected three strains of E. coli. The first of them,
E. coli BM-11, displayed a rather broad specificity in relation to the bases and sugar-
modified nucleosides [14] and was successfully employed for the synthesis of a variety
of base- and sugar-modified nucleosides [15 19]. It was found that the utilization of
cells cross-linked by glutaraldehyde (GA) offers some experimental advantages over
unmodified E. coli BM-11 cells [17 19]. Furthermore, the E. coli BM-11 cells display,
besides the high UPase or/and TPase and PNPase activities, high cytidine deaminase
(CDase) activity that enables the use of cytosine nucleosides as donors of pentofur-
anose moiety in the enzymatic transglycosylation. These findings have been success-
fully used in the synthesis of 9-(b-d-arabinofuranosyl)guanine (ara-G), employing 1-
(b-d-arabinofuranosyl)cytosine (ara-C) as a donor of the arabinofuranose residue and
guanosine (Guo) or 2'-deoxyguanosine (dGuo) as depot forms of guanine via a cascade
of enzymatic reactions [18]. The second strain, E. coli BMT-1D/1A, proved to be
somewhat more efficient in the synthesis of a) thymidine, using either i) dGuo, 2'-
deoxycytidine (dCyd), or 2'-deoxyadenosine (dAdo), or ii) a mixture of the same 2'-
deoxynucleosides resulting from an enzymatic hydrolysis of DNA as donors of 2-
deoxyribofuranose moiety [20], b) dAdo [21], c) a panel of ribo- and 2-deoxyribofur-
anosides of 1-deaza- and 3-deazapurines [22] and benzimidazoles [22][23], and d) 2-
chloro-2'-deoxyadenosine (cladribine) [24]. Finally, a slightly modified strain, E. coli
BMT-4D/1A, was found to be essentially more efficient vs. the parent E. coli BMT-1D/
1A in the synthesis of 2'-deoxy-2'-fluoroguanosine [25] and 2,6-diamino-9-(3-deoxy-b-
d-erythro-pentofuranosyl)purine [26].
Results and Discussion. The present study was undertaken in order to develop a
practical synthesis of 2-chloro-2'-deoxyadenosine (4) and 1-(b-d-ribofuranosyl)-1,2,4-
triazole-3-carboxamide (8; ribavirin), using dGuo or Guo as donors in the enzymatic
transglycosylation of 2-chloroadenine (2; 2Cl-Ade) and 1,2,4-triazole-3-carboxamide
(6; TCA), respectively. A choice of these analogues is undoubtedly evoked by their
medicinal importance (see, e.g., the leading references [27] and [28], resp.). The E. coli
BMT-4D/1A cells have been cross-linked by glutaraldehyde (GA-cells) to afford a

Helvetica Chimica Acta Vol. 85 (2002)
1903
biocatalyst that retained high UPase and PNPase activities and was suitable for
repeated use. The cross-linking of the cells by GA seems to be advantageous over the
alginate gel-entrapped cells used by Holy and co-workers [29] for a related biocatalyst
because the GA cells retain operational stability and catalytic activity up to 658 in
phosphate buffer. Moreover, the GA cells have been shown to be more efficient vs.
unmodified cells in the synthesis of nucleosides [18] and may be used upto five ten
times in reactions performed at 60 658 [30].
The efficient chemical method for the synthesis of cladribine consists in the
stereospecific glycosylation of sodium salt of 2,6-dichloropurine with 2-deoxy-3,5-di-
O-p-toluoyl-a-d-ribofuranosyl chloride leading to the formation of the corresponding
N9-b-d-nucleoside as the main product (58%), along with the N9-a-d-isomer (13%)
[31][32]. Treatment of the former with NH3 afforded cladribine (71%) [31][32]. Two
enzymatic approaches have been used for the preparation of cladribine. First of them
consists in the direct transfer of the 2-deoxyribofuranose residue from thymidine to
2Cl-Ade catalyzed by partially purified trans-N-deoxyribosylase [33]. The other, which
seems to be advantageous over the former, makes use of PNPase and UPase of E. coli
cells (Table 1).
In the present study, we have essentially optimized the conditions of the reaction
and employed the more efficient biocatalyst GA-E. coli BMT 4D/1A (Table 1). The
enzymatic transdeoxyribosylation studied here includes two coupled reversible
reactions, which occur in situ: i) the phosphorolysis of the glycosyl donor, dGuo, in
the presence of inorganic phosphate, resulting in the formation of guanine (Gua) and
the key intermediate 2-deoxy-a-d-ribofuranose-1-O-phosphate (3), and ii) the
condensation of the latter with 2Cl-Ade with the release of phosphate affording the
desired cladribine (4; Scheme 1). Both reactions are catalyzed by PNPase. From the
experimental viewpoint, the reaction was conducted in K-phosphate buffer (10 mm; p H
Scheme 1
Biocatalyst: glutaraldehyde-treated E. coli BMT 4D/1A cells.

1904
Helvetica Chimica Acta Vol. 85 (2002)
Table 1. Enzymatic Synthesis of Cladribine (4)
Donor
Catalyst ([mg/ml]) Time/Temp. Yield of the
Bioreactor
Ref.
(donor/acceptor mm
concentrations)
([h/8])
reaction [%]
(isolated yield [g/l/h]
productivity
[%] of 4)
2'-Deoxyadenosinea) (30 : 5) E. coli ATCC 5275
(ca. 10)
5 (37)
16 (37)
21 (60)
56 (b))
0.16
0.18
0.04
[34]
[29]
[35]
2'-Deoxyuridinec) (20 : 15)
E. coli SPTÀ
(ca. 12)
(67)
2'-Deoxyguanosine (15 : 5)
GA-E. coli
BMT 1D/1A
(ca. 0.2)
65 (54)
2'-Deoxyguanosine (120 : 40) GA-E. coli
7 (65)
95 (81)
1.55
This study
BMT 4D/1A
(ca. 0.8)
a) The 2-chloroadenosine was used as a source of 2Cl-Ade after in situ phosphorolysis by PNPase, followed by
the enzymatic reaction of 2Cl-Ade with 2-deoxy-a-d-ribofuranose-1-O-phosphate resulting from the dAdo
phosphorolysis, eventually affording cladribine; an excess of dAdo was transformed into 2'-deoxyinosine by the
treatment with adenosine deaminase (ADase) for 30 min at 308. b) The isolated yield of individual cladribine
c
after crystallization was not indicated. ) 2Cl-Ade was not substrate of PNPase of E. coli SPTÀ, and 2-chloro-N6-
(dimethylaminomethylene)adenine was, therefore, used as an acceptor of the 2-deoxypentofuranose residue;
alginate gel-entrapped cells were used as a biocatalyst.
7.5) at 658 and separated into two steps. First, the reaction was continued for 4 h,
affording the main quantity of 4 (75 80%), then the reaction mixture was stored at
‡ 48 for 16 18 h, the biocatalyst as well as dGuo and 2Cl-Ade were filtered off,
whereas the most of 4 remained in the filtrate. The second stepwas conducted under
the same conditions in a fresh buffer with the biocatalyst, dGuo, and 2Cl-Ade filtered
off previously, giving rise to the formation of an additional quantity of 4 (15 20%).
Finally, 4 was purified by silica-gel column chromatography to give, after crystallization
from EtOH, the desired compound in 81% isolated yield.
Ribavirin was first synthesised in the 1972 by the chemical glycosylation method
[36]. The synthesis is multistepand leads to the formation of two regioisomers that
have been separated into individual compounds by chromatography. It should be
stressed that only one isomer, ribavirin, displayed a broad spectrum of antiviral
activity, whereas the other regioisomer is devoid of any biological activity. As distinct
from the chemical synthesis, biotechnological process consists in the enzymatic
regioselective transfer of the ribofuranose moiety of readily available and inexpensive
natural ribonucleosides to the heterocyclic base (6, TCA), affording the desired
ribavirin (8; Scheme 2). Data on the biotechnological methods are collected in
Table 2.
As in the case of the cladribine (4) synthesis, the enzymatic transribosylation of
TCA consisted in two coupled reversible reactions: i) the phosphorolysis of the glycosyl
donor, Guo, in the presence of inorganic phosphate, resulting in the formation of Gua
and the key intermediate a-d-ribofuranose-1-O-phosphate (7), and ii) the condensa-
tion of the latter with TCA with the release of phosphate, affording the desired
ribavirine (8; Scheme 2). Again, both reactions were catalyzed by PNPase. However,

Helvetica Chimica Acta Vol. 85 (2002)
1905
Scheme 2
Biocatalyst: glutaraldehyde-treated E. coli BMT 4D/1A cells.
Table 2. Enzymatic Synthesis of Ribavirin (8) by the Transglycosylation of TCA (6)
Donor (donor/acceptor Catalyst ([mg/ml])
Time/Temp. Yield of the
Bioreactor
Ref.
mm concentrations)
[h/8]
reaction [%]
productivity
(isolated yield [g/l/h]
[%] of 8)
Uridine (200 : 200)
Orotidine (300 : 300)
Guanosine (300 : 300)
Enterobacter aerogenes
AJ-1125 (ca. 20)
Erwinia carotovora
AJ-2992 (ca. 20)
Brevibacterium acetilicum 96 (60)
ATCC-954 (ca. 10)
PNPase (100 units
per 1.0 mmol
of N7-Guo)
96 (60)
48 (60)
55 (a))
61 (a))
73 (a))
44 (a))
0.28
0.93
0.56
0.06
[37]
[38]
[39]
[40]
N7-Methylguanosine
(1.0 : 0.25)
48 (25)
Guanosine (300 : 300)
GA-E. coli
30 (60)
90 92
2.26
This study
BMT 4D/1A (ca. 15)
(67 70)
a) The isolation of individual ribavirin was not described.
an application of GA-E. coli BMT 4D/1A as a biocatalyst led to displacement of the
total equilibrium to the formation of 8 (90 92%). Purification of 8 was performed by
‡
double ion-exchange chromatography (first Dowex 50 Â 8, H form, and then Dowex
1 Â 8, OHÀ form) to give, after crystallization from EtOH, the desired compound in
67 70% isolated yield. Structure of the synthesized compound 8 was established by
1
comparison of the m.p., UV, and H-NMR data with those published in [36].
Conclusions. We have developed a very efficient biocatalyst for the preparation of
modified nucleosides. The biocatalyst represents the glutaraldehyde-cross-linked cells

1906
Helvetica Chimica Acta Vol. 85 (2002)
of E. coli BMT 4D/1A strain containing very active PNPase, UPase, and TPase. The
synthetic potential of the biocatalyst is manifested by the practical preparation of two
medicinal drugs cladribine and ribavirin. Taking into account that the biocatalyst was
earlier successfully used for the preparation of 2'-deoxy-2'-fluoroguanosine [25] and
2,6-diamino-9-(3-deoxy-b-d-erythro-pentofuranosyl)purine [26], it may be considered
as the universal efficient tool for the preparation of modified nucleosides, employing
sugar-modified pyrimidine nucleosides as donors of the pentofuranose residue.
Experimental Part
1. General. High-performance liquid chromatography (HPLC) was carried out with the Waters apparatus
(Waters, USA) (column Nova-Pac C-18 (3.9 Â 300 mm, Waters); Waters Pump Control Module, Waters 515
HPLC pump; Waters 996 photodiode array detector operating at 208 and 254 nm for analysis of ribavirin and
cladribine, resp.) and an isocratic elution with 3% MeCN (ribavirin) and 7% MeCN (cladribine) in H2O (v/v) at
a flow rate 0.7 ml/min (time of analysis was 15 and 30 min, resp.). Guanosine and 2'-deoxyguanosine were
purchased from Fluka (Switzerland). Column chromatography (CC): silica gel 60 H (70 230 mesh ASTM;
Merck, Germany). TLC: Silufol UV254 aluminum sheets silica gel 60 F254 (Merck, Germany). Solvent systems for
TLC: CHCl3/EtOH 4 :1 (v/v) (A), BuOH/25% aq. NH2 7:2 (v/v) (B). M.p.: Boetius apparatus (Germany); not
corrected. UV Spectra were measured with Specord M-400 spectrometer (Carl Zeiss, Germany). 1H- and
13C-NMR spectra were measured at 200.13 and 50.325 MHz, resp., at 238 on an AC-200 spectrometer equipped
with an Aspect 3000 data system (Bruker, Germany); d values are in ppm downfield from internal Me4Si (1H,
13C); coupling constants J are given in Hz; assignments of proton resonances were confirmed, when possible, by
selective homonuclear decoupling experiments.
2. Microorganisms. The whole E. coli BMT-4D/1A cells were used as the source of PNPase. The E. coli
BMT-4D/1A cells have been selected according to the technique offered by Munch-Petersen et al. [41]. They
carry two regulatory mutations (cytR and deoR) and are able to synthesize constitutively nucleoside-
catabolizing enzymes. The biomass of E. coli BMT-4D/1A cells possessing high PNPase activities was prepared
as described in [18].
3. Glutaraldehyde Treatment. Cell suspension (4% dry weight) in 0.1m phosphate buffer (pH 7.0) was cooled
to 48. A 25% aq. soln. of GA was added with stirring to give a final concentration of 0.75%. Afterwards, the
suspension was incubated for 15 min at the same temp. Then the cross-linked cells were washed free of GA by
centrifugation five times with 30 volumes of 10 mm phosphate buffer (pH 7.0).
4. Synthesis of 2-Chloro-2'-deoxyadenosine (ˆCladribine; 4). To the heterogeneous mixture (250 ml;
10 mm K-phosphate buffer, pH 7.5) consisting of dGuo (1; 8.0 g, 0.03 mol) and 2Cl-Ade (2) [42] (1.7 g, 0.01 mol)
at 658, the biocatalyst (final concentration 0.08% that corresponds to ca. 200 mg of the cells calculating for dry
weight) was added, and the reaction mixture was incubated at 658 for 4 h with gentle stirring. The formation of
the product was monitored by TLC on Silufol UV254 plates (A; double elution): the Rf values for dGuo, 2Cl-Ade,
and 4 are 0.13, 0.33, and 0.23, resp. The products were eluted from the TLC plates with H2O and quantified
spectrophotometrically; under indicated conditions, the formation of 4 attained 75 80%. The reaction mixture
was stored at ‡ 48 for 16 18 h, the formed precipitate, which contained the biocatalyst, as well as dGuo and
2Cl-Ade according to TLC, was filtered off and washed with cold (‡48) H2O (2  40 ml). The filtrate and
washings were combined (first H2O phase), the residue was suspended in K-phosphate buffer (50 ml; 10.0 mm,
pH 7.5), and the mixture was incubated again at 658 for 3 h with gentle stirring. The formation of an additional
quantity of 4 (15 20%; determined spectrophotometrically) was monitored.
The biocatalyst was removed from the reaction mixture by centrifugation (8000 Â g, 5 min). The precipitate
was suspended in H2O (40 ml) with stirring for 15 min, and the biocatalyst was removed as described above. Two
supernatants were combined with the first H2O phase, evaporated to dryness, and the crude 4 was purified by
silica-gel CC as described in [23]. The fractions containing individual 4 were combined and evaporated to
dryness, the residue was crystallized from EtOH to afford 4 (2.3 g, 81% based on 2Cl-Ade taken into reaction)
as a white powder that was identical with the earlier synthesized compound [23] in all respects (TLC, UV, and
1H-NMR). HPLC: tR 15.39 min (purity 99.60%).
5. 1-(b-d-ribofuranosyl)-1,2,4-triazole-3-carboxamide (8). The reaction mixture (100 ml; 10 mm
K-phosphate buffer, pH 7.0) consisting of the biocatalyst (final concentration 1.5% that corresponds ca.
15 mg/ml of the cells calculating for dry weight), Guo (5; 8.5 g, 0.03 mol), and TCA (6) [43] (3.36 g, 0.03 mol)

Helvetica Chimica Acta Vol. 85 (2002)
1907
was incubated at 608 for 30 h with gentle stirring. The formation of the product was monitored by TLC on Silufol
UV254 plates (B): the Rf values for TCA, 8, and Guo are 0.22, 0.15, and 0.11, resp. The products were quantified
spectrophotometrically, and, under indicated conditions, the formation of the desired compound 8 attained 90
92%. The cells were removed from the reaction mixture by centrifugation (8000 Â g, 10 min), washed with H2O
(2 Â 40 ml) with the subsequent withdrawal of the cells by centrifugation, and the combined supernatants were
‡
loaded onto a column (4.0 Â 20.0 cm; Dowex 50 Â 8, 200 400 mesh; H form). The column was washed with
H2O, slightly acidic (pH 3.0 4.0) ribavirin-containing fractions (TLC control) were combined (ca. 300 ml), and
loaded onto a column (2.5 Â 20.0 cm; Dowex 1 Â 8, 200 400 mesh; OHÀ form). The column was washed with
H2O, the ribavirin-containing fractions (TLC control) were combined (ca. 1.0 l), and evaporated to dryness. The
residue was dissolved under reflux in EtOH (70 80 ml), filtered through a glass filter, allowed to cool to r.t. and
left in a refrigerator overnight. Crystalline product was filtered off, washed with cold EtOH, and dried at 608
under vacuum to give 8 (4.91 5.13 g, 67 70%; TLC- and HPLC pure). M.p. 176 1788. UV (H2O): lmax 207
(11600) ([36]: m.p. 166 1688 (from aq. EtOH); 174 1768 (from EtOH)). HPLC: tR 3.99 min (purity 99.46%).
1H-NMR ((D6)DMSO): 8.90 (s, HÀC(5)), 7.90 (br. s, 1 H, 1 H of the CONH2 grouppresumably H-bonded with
N(2)); 7.64 (br. s, 1 H, non-H-bonded H of the CONH2 group); 5.82 (d, J(1',2') ˆ 3.9, HÀC(1')); 5.62 (d,
J(HÀC(2', OHÀC(2')) ˆ 5.5, OHÀC(2')); 5.24 (br. s, J(HÀC(3'), OHÀC(3')) ˆ 5.5, OHÀC(3')); 4.98 (t,
J(HÀC(5'), OHÀC(5')) ˆ 5.5, OHÀC(5')); 4.38 (m, J(2',3'') ˆ 4.6, HÀC(2')); 4.18 (m, J(3',4'') ˆ 4.8, HÀC(3'));
3.98 (m, J(4',5') ˆ J(4',5'') ˆ 5.0, HÀC(4')); 3.60 (m, J(5',5'') ˆ 12, HÀC(5')); 3.48 (m, HÀC(5'')).
REFERENCES
[1] E. De Clercq, Nature Rev. 2002, 1, 13, and refs. cit. therein; E. De Clercq, Int. J. Antimicrob. Agents 2001, 18,
309; E. De Clercq, J. Clin. Virol. 2001, 22, 73.
[2] E. De Clercq, Current Med. Chem. 2001, 8, 1543; E. De Clercq, Clin. Microbiol. Rev. 2001, 14, 382; E. De
Clercq, Nucleosides Nucleotides Nucleic Acids 2000, 19, 1531.
[3] J. A. Montgomery, Publ. Am. Inst. Hist. Pharm. 1997, 16, 185.
[4] M. A. Weiss, Current Oncol. Rep. 2001, 3, 217; V. Gandhi, W. Plunkett, S. Weller, M. Du, M. Ayres, C. O.
Rodriguez Jr., P. Ramakrishna, G. L. Rosner, J. P. Hodge, S. O×Brien, M. J. Keating, J. Clin. Oncol. 2001, 19,
2142.
[5] H. Vorbruggen, C. Ruh-Pohlenz, − Handbook of Nucleoside Synthesis×, John Wiley & Sons, Inc., 2001, pp
646.
[6] J. R. Hanrahan, D. W. Hutchinson, J Biotechnol. 1992, 23, 193; D. W. Hutchinson, Trends Biotechnol. 1990,
8, 348; T. Utagawa, J. Mol. Catalysis B: Enzymatic 1999, 6, 215.
[7] T. A. Krenitsky, Mol. Pharmacol. 1967, 3, 526.
[8] T. A. Krenitsky, J. Biol. Chem. 1968, 243, 2871.
[9] T. Utagawa, H. Morisawa, T. Nakamatsu, A. Yamazaki, S. Yamanaka, FEBS Lett. 1980, 119, 101.
[10] T. Utagawa, H. Morisawa, T. T. Miyoshi, F. Yoshinaga, A. Yamazaki, K. Mitsugi, FEBS Lett. 1980, 109, 261.
[11] T. A. Krenitsky, G. W. Koszalka, J. V. Tuttle, Biochemistry 1981, 20, 3615.
[12] J. L. Rideout, T. A. Krenitsky, G. W. Koszalka, N. K. Cohn, E. Y. Chao, G. B. Elion, V. S. Latter, R. B.
Williams, J. Med. Chem. 1982, 25, 1040.
[13] T. A. Krenitsky, G. A. Freeman, S. R. Shaver, L. M. Beacham 3rd, S. Hurlbert, N. K. Cohn, L. P. Elwell,
J. W. Selway, J. Med. Chem. 1983, 26, 891.
[14] A. I. Zinchenko, L. A. Eroshevskaya, V. N. Barai, I. A. Mikhailopulo, Nucleic Acids Res. Symp. Ser. No. 18,
1987, 137.
[15] I. A. Mikhailopulo, E. I. Kvasyuk, V. I. Lyakhovets, I. L. Popov, L. A. Eroshevskaya, V. N. Barai, A. I.
Zinchenko, Nucliec Acids Res. Symp. Ser. No. 14, 1984, 291.
[16] L. A. Eroshevskaya, V. N. Barai, A. I. Zinchenko, E. I. Kvasyuk, I.A. Mikhailopulo, Antibiot. Med.
Biotechnol. 1986, 31, 174; Chem. Abstr. 1986, 104, 166850t.
[17] E. N. Kalinichenko, V. N. Barai, S. B. Bokut, V. V. Romanova, A. I. Zinchenko, G. Herrmann, I. A.
Mikhailopulo, Biotechnol. Lett. 1989, 11, 621.
[18] A. I. Zinchenko, V. N. Barai, S. B. Bokut, E. I. Kvasyuk, I. A. Mikhailopulo, Appl. Microbiol. Biotechnol.
1990, 32, 658.
[19] G. V. Zaitseva, E. I. Kvasyuk, E. V. Vaaks, V. N. Barai, S. B. Bokut, A. I. Zinchenko, I. A. Mikhailopulo,
Nucleosides Nucleotides 1994, 13, 819.

1908
Helvetica Chimica Acta Vol. 85 (2002)
[20] A. I. Zinchenko, V. N. Barai, S. B. Bokut, N. V. Dudchik, Yu. V. Belyaeva, I. A. Mikhailopulo, Biotechnol.
Lett. 1990, 12, 341.
[21] A. I. Zinchenko, V. N. Barai, S. B. Bokut, N. V. Dudchik, E. I. Kvasyuk, I. A. Mikhailopulo, Biotechnol.
Lett. 1991, 13, 87.
[22] I. A. Mikhailopulo, A. I. Zinchenko, S. B. Bokut, N. V. Dudchik, V. N. Barai, E. N. Kalinichenko, H.
Rosemeyer, F. Seela, Biotechnol. Lett. 1992, 14, 885.
[23] I. A. Mikhailopulo, Z. Kazimierczuk, A. I. Zinchenko, V. N. Barai, V. V. Romanova, L. A. Eroshevskya,
Nucleosides Nucleotides 1995, 14, 477.
[24] I. A. Mikhailopulo, A. I. Zinchenko, Z. Kazimierczuk, V. N. Barai, S. B. Bokut, E. N. Kalinichenko,
Nucleosides Nucleotides 1993, 12, 417.
[25] A. I. Zinchenko, V. N. Barai, N. I. Pavlova, E. I. Boreko, I. A. Mikhailopulo, Nucleosides Nucleotides 1999,
18, 687.
[26] V. N. Barai, A. I. Zinchenko, L. A. Eroshevskaya, E. V. Zhernosek, E. De Clercq, I. A. Mikhailopulo, Helv.
Chim. Acta 2002, 85, 1893.
[27] K. R. Rai, Semin. Oncol. 1998, 3, 19; C. Tortorella, M. Rovaris, M. Filippi, Current Opin. Investig. Drugs
2001, 2, 1751.
[28] S. S. Lee, M. Sherman, J. Viral Hepat. 2001, 8, 202; S. P. Lawrence, Adv. Intern. Med. 2000, 45, 65.
[29] I. Votruba, A. Holy, H. Dvorakova, J. G¸nter, D. Hockova, H. Hrebabecky, T. Cihlar, M. Masojidkova,
Coll. Czech. Chem. Comm. 1994, 59, 2303.
[30] A. I. Zinchenko, L. A. Eroshevskaya, V. N. Barai, Biotechnol. (Moscow) 1990, 5, 36.
[31] Z. Kazimierczuk, H. B. Cottam, G. R. Revancar, R. K. Robins, J. Am. Chem. Soc. 1984, 106, 6379.
[32] G. E. Wright, C. Hildebrand, S. Freese, L. W. Dudicz, Z. Kazimierczuk, J. Org. Chem. 1987, 52, 4617.
[33] M.-C. Huang, K. Hatfield, A. W. Roetker, J. A. Montgomery, R. L. Blakley, Biochem. Pharmacol. 1981, 30,
2663; D. A. Carson, D. B. Wasson, R. Taetle, A. Yu, Blood 1983, 62, 737.
[34] W. A. Blank, K. J. Elder, W. P. Gati, A. R. P. Paterson, M. A. Pickard, J. S. Wilson, Biotechnol. Lett. 1992,
14, 669.
[35] I. A. Mikhailopulo, A. I. Zinchenko, Z. Kazimierczuk, V. N. Barai, S. B. Bokut, E. N. Kalinichenko,
Nucleosides Nucleotides 1993, 12, 417.
[36] J. T. Witkowski, R. K. Robins, R. W. Sidwell, L. N. Simon, J. Med. Chem. 1972, 15, 1150.
[37] H. Shirae, K. Yokozeki, K. Kubota, Agric. Biol. Chem. 1988, 52, 1233.
[38] H. Shirae, K. Yokozeki, K. Kubota, Agric. Biol. Chem. 1988, 52, 1499.
[39] H. Shirae, K. Yokozeki, K. Kubota, Agric. Biol. Chem. 1988, 52, 1777.
[40] W. J. Hennen, C.-H. Wong, J. Org. Chem. 1989, 54, 4692.
[41] A. Munch-Petersen, P. Nygaard, K. Hammer-Jespersen, N. Fill, Eur. J. Biochem. 1972, 27, 208.
[42] G. B. Brown, V. S. Weliky, J. Org. Chem. 1958, 23, 125; G. Elion, G. Hitchings, J. Am. Chem. Soc. 1956, 78,
3508.
[43] G. I. Cipens, V. Grinsteins, Latvijas PSR Zinatnu Akad. Vestis, Kim. Ser. 1965, 2, 204; Chem. Abstr. 1965, 63,
13243f.; G. Cipens, V. Grinsteins, Khim. Geoterotsikl. Soedin., Akad. Nauk Latv. SSR 1965, 4, 624; Chem.
Abstr. 1966, 64, 5074f.
Received May 23, 2002