Antiviral oligo- and polyribonucleotides containing selected triazolo[2,3-a]purines

Article abstract of DOI:10.1021/jm9802057

Several amphipathic (hydrophobic base, hydrophilic backbone) polyribonucleotides have recently been shown to have potent antiviral activity against HIV and human cytomegalovirus (HCMV). The working hypothesis developed during these studies was that the ab

Full text of DOI:10.1021/jm9802057

4958
J . Med. Chem. 1998, 41, 4958-4964
An tivir a l Oligo- a n d P olyr ibon u cleotid es Con ta in in g Selected
Tr ia zolo[2,3-a ]p u r in es|
Mayoka G. Tutonda,† Robert W. Buckheit, J r.,‡ Vijai K. Agrawal,†,§ and Arthur D. Broom*,†
Department of Medicinal Chemistry, 20 South 2030 East, 295 Biomedical Polymers Research Building, University of Utah,
Salt Lake City, Utah 84112-9453, Southern Research Institute, Frederick Research Center, Frederick, Maryland 21701, and
Aral Biosynthetics, Inc., 230 West 2855 South, Salt Lake City, Utah 84115
Received April 2, 1998
Several amphipathic (hydrophobic base, hydrophilic backbone) polyribonucleotides have recently
been shown to have potent antiviral activity against HIV and human cytomegalovirus (HCMV).
The working hypothesis developed during these studies was that the ability to form an ordered,
non-hydrogen-bonded array in solution was an important criterion for activity. To explore
further the role of structure and molecular size on the inhibition of virus replication, one new
polynucleotide and two 32-mer oligonucleotides based on the triazolo[2,3-a]purine ring system
have now been prepared. High-molecular-weight polynucleotide 4a (PTPR) and sulfur-
containing 32-mer 5b (TTPR) were moderately active against HIV but showed greater potency
against HDMV than ganciclovir. Both 4a and 5b gave clear evidence of cooperative melting
behavior, whereas inactive 32-mer 5a showed no such behavior.
Ch a r t 1. Prototype Polynucleotides
In tr od u ction
Human acquired immunodeficiency syndrome (AIDS)
is an expanding pandemic for which no curative therapy
is available. Current estimates indicate that over 20
million adults are alive and infected with HIV or have
AIDS, and the World Health Organization projects that
40 million HIV-positive individuals will be alive in the
year 2000.1 At present, only two loci in the HIV
replication cycle, reverse transcriptase and viral pro-
tease, are targeted by approved drugs. The former
target is inhibited by nucleoside analogues such as
AZT,2 ddC,3 ddI,4 3TC,5 and D4T.6 The recent advent
of nonnucleoside reverse transcriptase inhibitors and
protease inhibitors has given rise to combination pro-
tocols which markedly reduce viral load and, in some
cases at least, lead to the generation of undetectable
viral loads in patients.6 However, resistance to mono-
therapy and combination therapy is developing,7,8 and
it remains very clear that new drugs having unique
structures and targets are urgently needed.
Human cytomegalovirus (HCMV) infections are wide-
spread in AIDS patients and lead to substantial mortal-
ity and morbidity. Systemic (visceral) manifestations
of HCMV are usually managed with ganciclovir, cido-
fover, and foscarnet, at great expense and with consid-
erable toxicity.9 CMV retinitis is a leading cause of
blindness in AIDS patients and is poorly controlled by
systemic therapy; intraocular therapy has met with
some success but remains risky and somewhat contro-
versial.10 New drugs having increased potency, novel
mechanisms of action, and, importantly, equipotent
activity against HIV and CMV would be a useful step
forward in treating both AIDS and opportunistic cyto-
megalovirus infections simultaneously.
For a number of years, this laboratory has been
involved in studies on the synthesis and biological
activity of novel polynucleotides, especially those con-
taining thiopurines. These studies led to the demon-
stration that poly(1-methyl-6-thioguanylic acid) (PMTG,
1) and poly(1-methyl-6-thioinosinic acid) (PMTI, 2)
(Chart 1) are potent inhibitors of HIV11and HCMV
replication and cytopathicity. This observation has
prompted further studies designed to explore the pa-
rameters responsible for antiviral activity in order to
elucidate the mechanism of action and to optimize the
therapeutic potential of this class of compounds.
The working hypothesis on the structural basis of
antiviral activity has been that amphipathic character
(hydrophobic base/hydrophilic backbone) and the ability
to form a highly ordered, non-hydrogen-bonded array
in solution are prerequisites for antiviral activity. As
a test of this hypothesis, the synthesis and biological
evaluation of poly(1-amino-6-thioinosinic acid) (PATI,
| This paper is dedicated to our colleague and friend Prof. Dr.
Gottfried (Friedl) Heinisch on the occasion of his 60th birthday.
† University of Utah.
‡ Frederick Research Center.
§ Aral Biosynthetics, Inc.
10.1021/jm9802057 CCC: $15.00 © 1998 American Chemical Society
Published on Web 11/12/1998

Antiviral Oligo- and Polyribonucleotides
J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 25 4959
Ch a r t 2. Poly- and Oligonucleotides Containing
Triazolo[2,3-a]purines
Sch em e 1a
a
(i) Ac2O, pyridine; (ii) CH3I, K2CO3, DMF; (iii) NH3, MeOH;
(iv) Lawesson’s reagent.
Resu lts a n d Discu ssion
Ch em istr y. The key starting material for the req-
uisite polynucleotide and oligonucleotide precursors was
5H(7H)-9-oxo-3-(â-D-ribofuranosyl)-1,2,4-triazolo[2,3-a]-
purine (6), which was prepared as previously described13
by cyclization of 1-aminoguanosine using phosphoryl
chloride in DMF.
3; Chart 1) was recently reported.12 This polymer,
which has a much more hydrophilic base than PMTI,
was completely devoid of solution secondary structure
and antiviral activity or cytotoxicity.12 Since increased
hydrophilicity in the base abolished secondary structure
and antiviral activity, the next logical step in testing
the hypothesis was to enhance the hydrophobicity of the
base while retaining a normal phosphodiester backbone.
It was reasoned that extending the purine hydrophobic,
planar surface into a triazolopurine should provide
enhanced opportunity for stacking interactions and
facilitate formation of an ordered array in solution. An
additional important consideration entered into this
design process. It was previously shown11 that de-
creased chain length of PMTI to an average chain length
of approximately 36 bases resulted in a compound which
exhibited no secondary structure and had no biological
activity. It was anticipated that increasing the planar
surface of the bases might provide shorter compounds
which would still retain the desired properties. If this
turned out to be true, it would open up the area of
oligonucleotide chemical synthesis as a source of anti-
viral oligonucleotides, obviating the requirement that
nucleoside diphosphates be substrates for polynucleotide
phosphorylase.
Syn th esis of P olyn u cleotid e P r ecu r sor s. Some-
what surprisingly, nucleoside 6 was found to be quite
acidic. The pKa, as measured by UV, was approximately
4.5 for ionization of the triazolo ring proton, compared
to about 10 for triazole itself.14 Obviously, the nucleo-
side base would be fully anionic at pH 9; hence,
diphosphates of this and related compounds would
undoubtedly be very poor substrates for polynucleotide
phosphorylase, which has a pH optimum at about 9. It
was decided to explore the regioselectivity of methyla-
tion of the triazolopurine in order to determine whether
N5 could be selectively alkylated. Because 6 was
essentially insoluble in DMF, the nucleoside was first
converted to its tri-O-acetyl derivative 7 using acetic
anhydride in pyridine (Scheme 1). Although isolation
of this compound was difficult because of the acidity of
the heterocyclic base, it was possible to isolate the tri-
O-acetyl derivative by precipitation from an aqueous
solution at pH 4.5. Using the methylation procedure
applied to a derivative of the Y nucleoside by Ueda,15
in which alkylation is carried out by methyl iodide in
dimethylformamide in the presence of potassium car-
bonate, only a single N-methyl derivative was obtained.
The structure of this compound was unequivocally
In this paper, the synthesis and characterization of
polynucleotide 4a and oligonucleotides 5a and 5b (Chart
2) are described. Activity against HIV and against
human cytomegalovirus (HCMV) are also reported.

4960 J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 25
Tutonda et al.
Ta ble 1. Selected 1H and 13C NMR Chemical Shifts for
Compound 8a
Sch em e 3a
position
δ 1H
δ
13C
2
8.18
138.5
148.4
147.1
143.2
118.0
18.5
3a
4a
6
9a
CH3
H1′
8.80
3.72
6.20
86.1
a
Chemical shift data and HETCOR experiments were per-
formed in DMSO-d6 using a Varian Unity 500-MHz spectrometer.
Sch em e 2a
a
(i) POCl3, P(OEt)3; (ii) tris(tetrabutylammonium)pyrophosphate,
carbonyl diimidazole.
established as the N5-methyl on the basis of HMQC and
HMBC experiments.
Chemical shift assignments for carbons and protons
used in defining the position of the methyl group are
given in Table 1. The aromatic protons attached to C2
and C6 resonated at lowest field, δ 8.18 and 8.80.
Heteronuclear multiple quantum correlation (HMQC)
spectroscopy revealed that the proton resonating at
higher field (δ 8.18) was coupled to a carbon resonating
at δ 138.5 and the δ 8.80 proton to the δ 143.2 carbon.
The carbon shift at δ 138.5 was assigned to C2 by virtue
of a heteronuclear multiple bond correlation (HMBC)
experiment which demonstrated three-bond coupling
with the proton at H1′(δ 6.20). The methyl protons (δ
3.72) were strongly three-bond-coupled to C6 (δ 143.3)
and quaternary carbon C4a (δ 147.1); conversely, H6
was coupled only to C4a and the methyl group. These
data conclusively establish the sole site of methylation
to be N5.
Treatment of protected nucleoside 8 with Lawesson’s
reagent afforded the 9-thione 10 (Scheme 1). Deacyla-
tion of compounds 8 and 10 using methanolic ammonia
provided the unprotected nucleosides 9 and 11. Both
nucleosides 9 and 11 were converted to their respective
5′-diphosphates 12 and 13 using the phosphorylation
procedure of Yoshikawa16 and the diphosphate synthesis
of Hoard and Ott17 (Scheme 2).
Syn th esis of Oligon u cleotid e P r ecu r sor s. Nu-
cleoside 9 was 2′(3′)-O-methylated using the approach
of Ts’o and colleagues18 employing methyl iodide and
sodium hydride in DMF. Both the 5,2′-O-dimethyl and
5,3′-O-dimethyl isomers were isolated from the reaction
mixture by silica gel column chromatography. The
requisite 5,2′-O-dimethyl derivative 14 was separated
by crystallization from the mixture in an overall yield
of 34%. The usual COSY NMR proof of substitution
site, in which one may “walk” from H1′ to H2′ to H3′ to
3′-OH, failed in this case because of complete magnetic
a
(i) CH3I, DMF, NaH; (ii) Ac2O, pyridine; (iii) Lawesson’s
reagent, DMF; (iv) NH3, MeOH.
equivalence of the 2′- and 3′-protons, even at 500 MHz.
The structure of 14 was confirmed using 2D NOESY
spectroscopy; only in the presumed 2′-isomer did the
O-methyl group show a cross-peak with H1′ as would
be expected only if substitution were in the 2′-position.
Final structure proof was achieved by the smooth
conversion of 14 to its 3′,5′-(tetraisopropyldisiloxane)
derivative using the Markiewicz reagent as applied by
Robins.19 Such a conversion cannot be accomplished
with the 3′-O-methyl derivative. Compound 14 was
acetylated, and its 3′,5′-di-O-acetyl derivative 15 was
converted to the 9-thione 16 in 76% yield. Deacylation
of 16 afforded compound 17. Nucleosides 14 and 17
were converted to their respective 5′-O-dimethoxytrityl
derivatives 18 and 19 and then to phosphoramidites 20
and 21 in the usual way.20
P olym er iza tion of 5′-Dip h osp h a tes. Polymeriza-
tion of the 5′-diphosphate 12 (X ) O) was undertaken
using Micrococcus luteus polynucleotide phosphorylase
as previously described11,12 to give PTPR (4a ) in about
25% isolated yield (Scheme 3). The average molecular
weight, as estimated by size exclusion HPLC,11 was of
the order of 40 000-50 000. Enzymatic digestion of
PTPR using a mixture of venom phosphodiesterase and
alkaline phosphatase gave only the starting nucleoside
9, thus attesting to the homogeneity of the polynucle-
otide. The polymer exhibited fairly cooperative melting
behavior with a Tm of 24 °C.

Antiviral Oligo- and Polyribonucleotides
J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 25 4961
Ta ble 2. Virus Inhibition and Cytotoxicity Values (µM)a
Sch em e 4a
HIV
HCMVb
IC50
CD50
TI
IC50
CD50
TI
PMTI (2)
PTPR (4a )
TPR (5a )
TTPR (5b)
DDC
0.08c
1.0
>7
1.7
>1.0c >12.5 0.15
>3.2 >21
>7.9 >18
>2.5
>2.5 0.43
6.7
>22.8
>22.3
>3.4
>7.9
>7
>4.1 2.8
0.0015 >0.1
>67
ganciclovir
6.3
>125
>20
a
See text for cell lines, virus strains, and assay conditions; see
b
ref 11 (HIV) and ref 21 (HCMV). HCMV studies were performed
by Drs. J ohn Huffman and Robert Sidwell at the Institute for
Antiviral Research, Utah State University, Logan, UT. c As previ-
ously reported.11
clovir. Since HCMV is a DNA virus of the herpes class
and has no reverse transcriptase, elucidation of the
mechanism of this potent inhibition must await further
investigation. Regardless of the mechanism of action,
however, the ability of these amphipathic oligo- and
polyribonucleotides to inhibit both HIV and HCMV
opens an exciting new area of therapeutic investigation.
a
(i) DMTRCl, pyridine; (ii) â-cyanoethyl tetraisopropylphos-
phorodiamidite, diisopropylammonium tetrazolide, CH3CN.
Despite modifications in the source of polynucleotide
phosphorylase, divalent counterion, pH, and tempera-
ture, no conditions could be found under which 5′-
diphosphate 13 (X ) S) was a substrate for the enzyme,
emphasizing once more the importance of examining
oligonucleotides prepared by chemical synthesis.
Oligon u cleotid e Syn th esis. Oligonucleotides 5a
and 5b were synthesized from their respective 5′-O-
(dimethoxytrityl)-3′-O-phosphoramidites 20 and 21
(Scheme 4) on an automated DNA synthesizer as 32-
mers having a single thymidine residue on the 3′-
terminus. The crude mixtures were analyzed by SAX
HPLC using a 60-min linear gradient elution with 1 mM
KH2PO4 in formamide/water (6:4) (pH 6.3) and satu-
rated KH2PO4 in formamide/water (6:4) (pH 6.3). Pu-
rification was performed on a gel filtration column using
Sephadex 50-80 with aqueous elution. Purity of the
compounds was then checked by SAX HPLC using the
same phosphate buffer system. Both oligomers exhib-
ited fairly cooperative melting transitions at 19.0 °C for
8a and 32.6 °C for 8b. This significant difference in
melting temperature emphasizes the important role of
hydrophobicity/polarizability of sulfur compared to oxy-
gen in stabilizing hydrophobic stacking interactions of
single-stranded oligonucleotides.
Biologica l Resu lts. The polynucleotide 4a and the
oligonucleotides 5a and 5b were tested for anti-HIV
activity as previously described.11 As noted above, the
polynucleotide 4a was found to be active against HIV
IIIB in CEM-SS cells with an IC50 of ∼0.4 µM and was
nontoxic (Table 2). The oligomer containing the 9-thio-
triazolopurine 5b was active with an IC50 of 1-2 µM in
the same system and was nontoxic up to 100 µg/mL,
whereas the 9-oxo oligomer 5a was devoid of activity or
cytoxicity.
Perhaps the most striking and unexpected observa-
tions in this study relate to activity against human
cytomegalovirus (HCMV). Antiviral activity was de-
termined in the laboratories of Drs. Robert Sidwell and
J ohn Huffman, Utah State University, using a plaque
reduction assay with strain AD-169 of HCMV in MRC-5
cells as previously described.21 As shown in Table 2,
compounds 2, 4a , and 5b are approximately equipotent
inhibitors of HCMV and HIV and are somewhat more
potent against HCMV than the positive control ganci-
Exp er im en ta l Section
1H and 31P NMR spectra were recorded with an IBM AF
200-MHz FT-NMR or a Varian Unity 500-MHz spectrometer.
UV spectra and melting curves were recorded with a Hewlett-
Packard 8452A diode array spectrophotometer equipped with
a Peltier variable temperature device; circular dichroism
spectra were obtained using a J asco 700A CD spectropolarim-
eter equipped for variable temperature studies. FAB and EI
mass spectra were recorded with a MAT 95 spectrometer.
MALDI mass spectra were obtained using a Per Septive
Biosystems MALDI-TOF mass spectrometer Voyager - DE
STR Bio Spectrometry workstation. Electrospray mass mea-
surements were conducted using a Fisons Instrument Quattro
II mass spectrometer in
a solvent of 0.4 M 1,1,1,3,3,3-
hexafluoro-2-propanol adjusted to pH 7 with triethylamine,
and methanol. Whatman plates, precoated with silica gel 60
containing fluorescent indicator F254, were used for thin-layer
chromatography, and silica gel 60 (Mallinckrodt SilicAR, 60-
200 mesh) was employed for column chromatography.
HPLC experiments were carried out on a Hitachi L6200
pump equipped with an L3000 photodiode array. Reverse-
phase chromatography utilized Rainin Microsorb-MV C8 and
C18 columns. Purity determinations were made by peak
integration of the HPLC traces using 50 mM KH2PO4, pH 5
(solvent A), MeOH (solvent B), 50 mM (NH4)HCO3, pH 7.0
(solvent C), and/or 40% aqueous CH3CN (solvent D). For
strong anion-exchange chromatography, a Partisil 10 SAX
WCS analytical column was employed. Size separations were
carried out using a BioRad SEC 125 column. The synthesis
of polynucleotides was performed using M. luteus polynucle-
otide phosphorylase obtained from Sigma.
5H(7H)-9-Oxo-3-(â-D-r ibofu r a n osyl)-1,2,4-tr ia zolo[2,3-
a ]p u r in e (6). This compound was prepared from 1-ami-
noguanosine as previously described.5
2′,3′,5-Tr i-O-a cetyl-5H(7H)-9-oxo-3-(â-D-r ibofu r a n osyl)-
1,2,4-tr ia zolo[2,3-a ]p u r in e (7). Compound 6 (13 g, 42 µmol)
was added to a mixture of acetic anhydride (126 mL) and
pyridine (169 mL) and stirred for 2 h at room temperature.
The clear solution was evaporated, and the residue was
repeatedly coevaporated with toluene until the odor of pyridine
was no longer present. The residue was partially dissolved
in cold water, and the pH of the suspension was adjusted from
nearly 2 to 4.5 by slow addition of ammonium hydroxide; then
the suspension was cooled at 4 °C overnight. The precipitate
was filtered, washed with ice water and then cold ethanol, and
then dried under vacuum to give 17 g of 7 (93%). 1H NMR
(DMSO-d6): δ 8.73 (s, 1, H6); 8.19 (s, 1, H2); 6.14 (d, 1, H1′);
5.90 (t, 1, H2′); 5.53 (t, 1, H3′); 4.34 (m, 3, H4′ and H5′,5′′); 2.1
(s, 3, COCH3); 2.03 (2s, 6, COCH3). EI HRMS (M - HOAc)+:

4962 J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 25
Tutonda et al.
calcd 374.09748, obsd 374.09808. Purity was >99% as judged
overall yield, based upon starting nucleoside 9). 1H NMR
(D2O): 8.42 (s, 1, H6); 8.20 (s, 1, H2); 5.98 (d, 1, H1′); 4.67 (m,
1, H2′); 4.52 (t, 1, H3′); 4.24 (m, 1, H4′); 4.13 (m, 2, H5′,5′′);
3.65 (s, 3, NCH3). 31P NMR (D2O): -5.5 (d); -9.75 (d). FAB
MS: 503 (M + Na - H)-.
5-Meth yl-3-(â-D-r ibofu r a n osyl)-1,2,4-tr ia zolo[2,3-a ]p u -
r in e-9-th ion e-5′-d ip h osp h a te (13). Yield: 30.7% (two steps,
overall yield, based on starting nucleoside 11). 1H NMR
(D2O): 8.67 (s, 1, H6); 8.42 (s, 1, H2); 6.05 (d, 1, H1′); 4.66 (m,
1, H2′); 4.56 (m, 1,H3′); 4.26 (m, 1, H4′); 4.13 (m, 2, H5′,5′′).
31P NMR: -5.43 (d); -9.86 (d). FAB MS (glycerol): 519 (M +
Na - H)-.
P oly[5-m eth yl-9-oxo-3-(â-D-r ibofu r an osyl)-1,2,4-tr iazolo-
[2,3-a ]p u r in ylic a cid ] (P TP R, 4a ). A solution made from
the following components was incubated for 5 h at 37 °C: Tris‚
HC1 (pH 9, 2 M), 0.666 mL; MgCl2 (0.1 M), 0.666 mL;
2-mercaptoethanol (0.28 M), 0.666 mL; PNPase (M. luteus),
10 mg (∼400-900 units); water, 2.44 mL; diphosphate 12, 100
mg. After incubation, the reaction mixture was diluted with
H2O, and the pH was adjusted to approximately 7.5 with very
dilute aqueous HCl. The solution was then extracted 12 times
with 60 mL of chloroform/isoamyl alcohol (5:2) to give a clear
aqueous solution. The aqueous layer was dialyzed against 0.1
M HCl for 24 h and then H2O for 48 h. Lyophilization of the
aqueous solution afforded PTPR (4a ) (28 mg, 28%) as a fluffy
solid. UV: λmax 270 nm, ꢀmax 8340 (0.1 N HCl).
HPLC analysis using a size exclusion column (BioRad SEC
125), isocratically eluted with an aqueous solution containing
NaH2PO4 (50 mM), Na2HPO4 (50 mM), NaCl (150 mM), and
NaN3 (10 mM), pH 6.8, at a flow rate of 1 mL/min gave a broad
peak, retention time: 7.06 min.
En zym a tic Degr a d a tion of P TP R (4a ). A solution made
from venom phosphodiesterase, 150 µL; MgCl2 (0.1 M), 150
µL; Tris‚HCl (2 M, pH 9); phosphatase, alkaline, 50 µL; PTPR
(4a ) (2 mg/1 mL of H2O), 550 µL; HCl (0.1 M), was incubated
at 37 °C overnight. After incubation, the solution was diluted
to 3 mL with HPLC water and extracted six times with
chloroform/isoamyl alcohol (5:2) to give a clear solution. HPLC
analysis on reverse-phase C18 column isocratically eluted with
KH2PO4 buffer (50 mM)/MeOH (85:15) gave a peak at retention
time: 5.29 min, which corresponds to the starting nucleoside
9 as confirmed by a co-injection experiment.
5,2′-O-Dim eth yl-9-oxo-3-(â-D-r ibofu r a n osyl)-1,2,4-tr ia -
zolo[2,3-a ]p u r in e (14). A solution of 9 (5.796 g, 18 mmol)
in 180 mL of dry dimethylformamide was cooled to ice
temperature. NaH (600 mg; 24 mmol, 60% dispersion in
mineral oil) was washed with benzene (3 × 20 mL) and added
to the solution. The reaction mixture was stirred vigorously
for 1 h; then methyl iodide (1 mL, ∼16.2 mmol) in 10 mL of
dimethylformamide was added in two portions of 5 mL each
(at a 45-min interval). The reaction mixture was stirred for 4
h at 0 °C in an ice bath. The dimethylformamide was
evaporated in vacuo and the residue dissolved in a minimum
amount of methanol and dried on silica gel. The compound
was purified by chromatography using a silica gel column
eluting with a methanol gradient in chloroform (8-15%). The
fractions collected initially below the yellow band contained
mostly the 2′-OCH3 compound (2.52 g). Those fractions that
were eluted and collected later (1.94 g) contained a mixture
of 2′-OCH3 and 3′-OCH3 in a ratio of 2.2:1. Ethanol recrys-
tallization of the first pool of fractions gave 2.05 g of pure 2′-
OCH3 compound 14 (34%, isolated yield). The reaction was
repeated on a larger scale starting from 14.5 g of 9 to give
6.12 g of 2′-OCH3 compound 14. 1H NMR (D2O): δ 8.37 (s, 1,
H6); 8.00 (s, H, H2); 5.88 (d, 1, H1′); 4.40 (m, 2, H2′ and H3′);
4.07 (m, 1, H4′); 3.76 (m, 2, H5′,5′′); 3.59 (s, 3, NCH3); 3.36 (s,
3, OCH3). FAB HRMS (MH+): calcd 337.12604, obsd 337.12880.
Purity was >99% as judged by HPLC (C18) using A:B (85:15)
and C8 using C:D (60:40).
by HPLC (C18) using A:B (75:25) and C8 using C:D (75:25).
2′,3′,5′-Tr i-O-a cet yl-5-m et h yl-9-oxo-3-(â-D-r ib ofu r a n o-
syl)-1,2,4-tr ia zolo[2,3-a ]p u r in e (8). To a solution of 7 (8.68
g, 20 mmol) in dimethylformamide (100 mL) were added
powdered potassium carbonate (4.15 g, 30 mmol) and methyl
iodide (4.26 g, 30 mmol). The suspension was stirred for 18 h
at room temperature and passed through a layer of Celite. The
white precipitate was carefully washed with dimethylforma-
mide until the filtrate exhibited no UV absorption. The filtrate
was then evaporated to an oil, and the product was purified
by chromatography using a silica gel column eluted by a
gradient of methanol in chloroform (0-5%). The pure fractions
were pooled and evaporated to give 6.27 g of 8 (69%). 1H NMR
(DMSO-d6): δ 8.8 (s, 1, H6); 8.16 (s, 1, H2); 6.18 (d, 1, H1′);
5.93 (t, 1, H2′); 5.85 (t, 1, H3′); 4.33-4.24 (m, 3, H4′, H5′, H5′′),
3.71 (s, 3, NCH3); 2.11, 2.06, 1.95 (s, 3 each, COCH3). FAB
HRMS (MH+): calcd 449.14209, obsd 449.14148. Purity was
>99% as judged by HPLC (C18) using A:B (75:25) and C8
using C:D (60:40).
5-Meth yl-9-oxo-3-(â-D-r ibofu r a n osyl)-1,2,4-tr ia zolo[2,3-
a ]p u r in e (9). Compound 8 (4.48 g, 10 mmol) was placed in a
stainless steel bomb and treated with methanolic ammonia
saturated at -20 °C (200 mL). The mixture was allowed to
stand at room temperature for 18 h. The ammonia was
vented, and the contents were transferred into a beaker and
allowed to stand. White crystals were then collected by
filtration and washed with cold MeOH. Yield: 2.97 g, 92%.
1H NMR (DMSO-d6): δ 8.8 (s, 1, H6); 8.2 (s, 1, H2); 5.87 (d, 1,
H1′); 5.46 (d, 1, OH); 5.24 (d, 1, OH); 5.02 (t, 1, OH); 4.55 (m,
1, H2′); 4.14 (m, 1, H3′); 3.94 (m, 1, H4′); 3.63-3.59 (m, 5, H5′,
H5′′ and NCH3). FAB HRMS (MH+): calcd 323.11039, obsd
323.10898. Purity was >99% as judged by HPLC (C18) using
A:B (90:10) and C8 using C:D (90:10).
2′,3′,5′-Tr i-O-acetyl-5-m eth yl-3-(â-D-r ibofu r an osyl)-1,2,4-
tr ia zolo[2,3-a ]p u r in e-9-th ion e(10). Compound 8 (6 g, 13.4
mmol) was dissolved in anhydrous DME (50 mL) under argon.
Lawesson’s reagent (3.30 g, 8.57 mmol) was added to the
solution. The suspension was refluxed for 1 h at which time
TLC in CHCl3/CH3OH (9:1) indicated that all starting material
had been consumed. The mixture was allowed to cool and was
poured into a beaker containing water (250 mL), and brine
(50 mL) was added. The milky aqueous solution was extracted
with CHCl3 (6 × 100 mL). The chloroform extract was washed
with water (100 mL) and dried over sodium sulfate, and the
dried extract was concentrated in vacuo to give the crude
product. It was purified by silica gel chromatography using
2% CH3OH in chloroform as eluent to give 3.59 g of 10 (58%).
1H NMR (DMSO-d6): δ 8.8 (s, 1, H6); 8.2 (s, 1, H2); 5.87 (d, 1,
H1′); 5.46 (d, 1, OH); 5.24 (d, 1, OH); 5.02 (t, 1, OH); 4.55 (m,
1, H2′); 4.14 (m, 1, H3′); 3.94 (m, 1, H4′); 3.63-3.59 (m, 5, H5′,
H5” and NCH3). FAB HRMS (MH+): calcd 465.11924, obsd
465.12088. Purity was 94-96% as judged by HPLC (C18)
using A:B (65:35) and C8 using C:D (60:40).
5-Meth yl-3-(â-D-r ibofu r a n osyl)-1,2,4-tr ia zolo[2,3-a ]p u -
r in e-9-th ion e (11). Compound 10 (2.65 g, 5.57 mmol) was
placed in a stainless steel bomb and treated with methanolic
ammonia saturated at -20 °C (100 mL). The mixture was
allowed to stand at room temperature for 18 h. The ammonia
was vented, and the contents were concentrated to one-third
volume. The crystals were filtered, washed with methanol,
and recrystallized with methanol/water to give compound 11
(1.35 g, 70%). 1H NMR (DMSO-d6): δ 9.02 (s, 1, H6); 8.4 (s,
1, H2); 5.95 (d, 1, H1′); 5.5 (d, 1, OH); 5.25 (d, 1, OH); 5.02 (t,
1, OH); 4.55 (m, 1, H2′); 4.14 (m, 1, H3′); 3.93 (m, 1, H4′); 3.74
(s, 3, NCH3); 3.61 (m, 2, H5′, 5”). FAB MS (MH+): 339. Anal.
Calcd: C, 42.60; H, 4.17; N, 24.84; S, 9.48. Obsd: C, 42.54;
H, 4.25; N, 24.91; S, 9.23.
The nucleoside-5′-diphosphates 12 and 13 were prepared
from their respective nucleosides 9 and 11 according to the
phosphorylation procedure of Yoshikawa16 and the diphosphate
synthesis of Hoard and Ott.17
5-Meth yl-9-oxo-3-(â-D-r ibofu r a n osyl)-1,2,4-tr ia zolo[2.3-
a ]p u r in e-5′-d ip h osp h a te (12). Yield: 35.4% (two steps,
3′,5′-Di-O-a cetyl-5,2′-O-d im eth yl-9-oxo-3-(â-D-r ibofu r a -
n osyl)-1,2,4-tr ia zolo[2,3-a ]p u r in e (15). Compound 14 (5.38
g, 16 mmol) was added to a mixture of acetic anhydride (48
mL) and pyridine (64 mL). The reaction mixture was stirred
at room temperature for 2 h. The clear solution was evapo-

Antiviral Oligo- and Polyribonucleotides
J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 25 4963
rated to dryness, and a few pieces of ice were added to destroy
excess acetic anhydride. The mixture was evaporated and
coevaporated twice with toluene (50 mL). The residue was
dissolved in CHCl3 (200 mL), and the chloroform solution was
washed with cold 0.1 N aqueous HCl (2 × 50 mL), cold water
(5 × 50 mL), cold aqueous 5% NaHCO3 (2 × 50 mL), and cold
water (2 × 50 mL). The chloroform extract was dried over
Na2SO4 and evaporated to give 5.85 g of 15 (87%). 1H NMR
(DMSO-d6): δ 8.8 (s, 1, H6); 8.2 (s, 1, H2); 6.00 (d, 1, H1′);
5.56 (d, 1, H2′); 4.81 (m, 1, H3′); 4.32 (m, 3, H4′ and H5′,5′′);
3.68 (s, 3, NCH3); 3.31 (s, 3, OCH3); 2.14 and 2.02 (s, 3 each,
COCH3). FAB HRMS (MH+): calcd 421.14717, obsd 437.12444.
Purity was >99% as judged by HPLC (C18) using A:B (75:25)
and C8 using C:D (60:40).
3′,5′-Di-O-a cetyl-5,2′-O-d im eth yl-3-(â-D-r ibofu r a n osyl)-
1,2,4-tr ia zolo[2,3-a ]p u r in e-9-th ion e (16). Compound 15
(3.00 g, 7.2 mmol) was dissolved in anhydrous DME under
argon. Lawesson’s reagent (1.78 g) was added to the solution.
The suspension was refluxed for 2 h at which time TLC in
CHCl3/MeOH (9:1) indicated that the all starting material had
been consumed. The mixture was cooled and was poured into
a beaker containing 270 mL of water. The milky aqueous
solution was extracted with chloroform (3 × 100 mL). The
chloroform extract was washed with water (20 mL) and dried
over Na2SO4. The dried extract was concentrated in vacuo to
give the crude product. It was purified by silica gel chroma-
tography using 2% MeOH/CHCl3 as eluent to give 2.39 g of
16 (76%).1H NMR (DMSO-d6): δ 9.03 (s, 1, H6); 8.44 (s, 1, H2);
6.03 (d, 1, H1′); 5.56 (m, 1, H2′); 4.82 (t, 1, H3′); 4.32 (m, 3,
H4′ and H5′, 5′′); 3.76 (s, 3, NCH3); 3.29 (s, 3, OCH3); 2.14 and
2.01 (s, 3 each, COCH3). FAB HRMS (MH+): calcd 437.12433,
obsd 437.12444. Purity was at least 95% as judged by HPLC
(C18) using A:B (65:35) and C8 using C:D (60:40).
5,2′-O-Dim et h yl-3-(â-D-r ib ofu r a n osyl)-1,2,4-t r ia zolo-
[2,3-a ]p u r in e-9-th ion e (17). Compound 16 (2.20 g, 5.045
mmol) was placed in a stainless steel bomb and heated with
methanolic ammonia saturated at -20 °C (100 mL) and the
mixture stood at room temperature for 18 h. The ammonia
was vented, and the contents were transferred to a beaker and
allowed to stand. The white crystals were collected by filtra-
tion and washed with cold methanol. Yield of 17: 1.56 g, 88%.
1H NMR (DMSO-d6): δ 9.02 (s, 1, H6); 8.45 (s, 1, H2); 6.01 (d,
1, H1′); 5.32 (d, 1, OH); 5.09 (t, 1, OH); 4.32 (m, 2, H2′ and
H3′); 3.97 (m, 1, H4′); 3.61 (m, 2, H5′,5′′); 3.73 (s, 3, NCH3),
3.33 (s, 3, OCH3). FAB HRMS (MH+): calcd 353.10320, obsd
353.10435. Purity was >99% as judged by HPLC (C18) using
A:B (85:15) and C8 using C:D (60:40).
5′-(Dim eth oxytr ityl)-5,2′-O-d im eth yl-9-oxo-3-(â-D-r ibo-
fu r a n osyl)-1,2,4-tr ia zolo[2,3-a ]p u r in e (18). Compound 14
(1.53 g, 4.6 mmol) was dried by coevaporation with pyridine
(3 × 20 mL), dissolved in 20 mL of anhydrous pyridine, and
cooled to 0 °C in an ice bath. Dimethoxytrityl chloride (1.86
g, 5.5 mmol) was added to the above solution, and the reaction
was warmed to room temperature. It was stirred for 2 h at
which time TLC in 5% MeOH/CHCl3 indicated complete
disappearance of 14. Methanol (3 mL) was added to quench
the reaction. The solvent was evaporated under vacuum, and
the residue was dissolved in chloroform (35 mL). The chlo-
roform solution was washed with 5% aqueous NaHCO3 (2 ×
15 mL) and water (15 mL) and dried over Na2SO4. The
chloroform solution was evaporated, and the product was puri-
fied by silica gel chromatography with a gradient of methanol
in chloroform (0-5%) containing triethylamine (0.1%) to
prevent degradation of tritylated product. Yield of 18: 2.26
g, 71%. 1H NMR (DMSO-d6): δ 8.78 (s, 1, H6); 8.12 (s, 1, H2);
7.36-6.77 (m, 13, Ar); 6.02 (d, 1, H1′); 5.34 (d, 1, OH); 4.36
(m, 2, H2′ and H3′); 4.06 (m, 1, H4′); 3.70 (s, 6, OCH3); 3.50 (s,
3, NCH3); 3.30 (s, 3, OCH3). FAB MS: 639 (MH+).
reaction mixture was stirred at room temperature for 18 h at
which time TLC in 5% MeOH/CHCl3 indicated complete
disappearance of 18. The reaction mixture was diluted with
saturated aqueous NaHCO3 and extracted with dichloro-
methane (2 × 70 mL). The dichloromethane extract was
washed with brine (3 × 20 mL), dried over Na2SO4, and
evaporated to a foam. The product was purified by flash
chromatography using CHCl3/CH3CN (8:2) as eluent to give
20 (1.05 g, 80%). 31P NMR: δ 151.229 and 151.547.
5′-(Dim eth oxytr ityl)-5,2′-O-d im eth yl-3-(â-D-r ibofu r a n o-
syl)-1,2,4-tr ia zolo[2,3-a ]p u r in e-9-th ion e (19). Compound
17 (1.4 g, 3.97 mmol) was dried by coevaporation with pyridine
(2 × 15 mL) and was suspended in pyridine (15 mL).
Dimethoxytrityl chloride (2.16 g, 6.4 mmol) was added to the
above suspension, and the reaction mixture was stirred at
room temperature. After 8 h, the suspension turned to a clear
solution and the TLC in CHCl3/MeOH (9:1) indicated that all
starting material had been consumed. Methanol (3.7 mL) was
added to quench the reaction. The solvent was evaporated in
vacuo and the residue dissolved in CHCl3 (35 mL). The chlo-
roform solution was washed with 5% aqueous NaHCO3 (2 ×
15 mL) and brine (15 mL) and dried over Na2SO4. The dried
chloroform solution was evaporated, and the product was puri-
fied by silica gel column chromatography with a gradient of
methanol in chloroform (0-3%) containing triethylamine
(0.1%). Yield of 19: 2.24 g, 86%. 1H NMR: δ 9.01 (s, 1, H6);
8.33 (s, 1, H2); 7.37-6.77 (m, 13, Ar); 6.06 (d, 1, H1′); 5.36 (d,
1, OH); 4.38 (m, 2, H2′ and H3′); 4.08 (m, 1, H4′); 3.70 (s, 6,
OCH3); 3.58 (s, 3, NCH3); 3.30 (s, 3, OCH3). FAB MS: 655
(MH+).
5′-(Dim eth oxytr ityl)-5,2′-O-d im eth yl-3-(â-D-r ibofu r a n o-
syl)-9-t h io-1,2,4-t r ia zolo[2,3-a ]p u r in e-3′-O-p h osp h or a -
m id ite-9-th ion e (21). Purified acetonitrile (27 mL) and O-(â-
cyanoethyl) N,N,N′,N′-tetraisopropylphosphorodiamidite (0.84
mL) were added to 19 (1.5 g, 2.29 mmol) and dry diisopropyl-
ammonium tetrazolide (195 mg). The reaction mixture was
stirred at room temperature for 18 h at which time TLC in
CHCl3/MeOH (95:5) indicated that the starting material had
been consumed. The reaction mixture was diluted with satu-
rated aqueous NaHCO3 and extracted with dichloromethane
(2 × 100 mL). The dichloromethane extract was washed with
brine (3 × 25 mL) and dried over Na2SO4, and the dried solu-
tion was evaporated to foam. The product was then purified
by flash chromatography using CHCl3/CH3CN (9:1) as eluent
to give 21 (1.52 g, 78%). 31P NMR: δ 151.542 and 151.375.
TP R 32-m er 5a . TPR 32-mer was synthesized from 5′-O-
DMTr-3′-O-phosphoramidite (20) on a 1-mmol scale synthesis
using an automated DNA synthesizer to afford 10.26 mg of
crude 5a . The purification was performed by gel filtration
using a Sephadex 50-80 column and eluting with deionized
and sterilized water. Pure fractions as judged by SAX HPLC
were pooled and freeze-dried to give 6.35 mg of material 5a .
UV: ꢀmax 6800 at 268. MALDI mass spectrum: calcd 12 986,
obsd 2 993. Electrospray MS: calcd 12 986, obsd 12 985.
TTP R 32-m er 5b. TTPR 32-mer was synthesized from 5′-
O-DMTr-3′-O-phosphoramidite (21) on a 1-mmol scale using
an automated DNA synthesizer to afford 7.32 g of crude. The
product was purified by gel filtration using a Sephadex 50-80
column to give 4.76 mg of pure TTPR 32-mer 5b. The purity
of the compound was checked by SAX HPLC using the same
system as described for 5a . UV: ꢀmax ∼6000 at 328. MALDI
mass spectrum: calcd 13 500, obsd 13 517.
Ack n ow led gm en t. The authors gratefully acknowl-
edge support of these studies by PHS/NIH/NIAID Grant
R01 AI 27692. Support for the FAB and EI mass
spectrometry facilities was provided by NSF Grant CHE
9002690 and a University of Utah Institutional Funds
Committee grant. The Health Science Center Nuclear
Magnetic Resonance, Electrospray/MALDI Mass Spec-
trometry, and DNA Synthesis facilities are supported
in part by a Cancer Center Support grant to the
Huntsman Cancer Institute, P30 CA 42014.
5′-(Dim eth oxytr ityl)-5,2′-O-d im eth yl-9-oxo-3-(â-D-r ibo-
fu r a n osyl)-1,2,4-t r ia zolo[2,3-a ]p u r in e-3′-O-p h osp h or a -
m id ite (20). Purified acetonitrile (18 mL) and O-(â-cyanoet-
hyl) N,N,N′,N′-tetraisopropylphosphorodiamidite (0.56 mL,
1.70 mmol) were added to 18 (1 g, 1.56 mmol) and dried
diisopropylammonium tetrazolide (133 mg, 0.78 mmol). The

4964 J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 25
Tutonda et al.
(11) Broom, A. D.; Agrawal, V. K.; Tutonda, M. G.; Fain, H. D.;
Buckheit, R. W. Unusual single-stranded polyribonucleotides as
potent anti-HIV agents. J . Med. Chem. 1995, 38, 3253-3257.
(12) Tutonda, M. G.; Fain, H. D.; Buckheit, R. W.; Broom, A. D. The
role of hydrophobic versus hydrophilic base character in the anti-
HIV activity of purine-containing polyribonucleotides. Nucleo-
sides Nucleotides 1997, 16 (1&2), 173-182.
Refer en ces
(1) Mertens, T. E.; Low-Beer, D. HIV and AIDS: where is the
epidemic going? Bull. WHO 1996, 74 (2), 121-129.
(2) Fischl, M. A.; Richman, D. D.; Grieco, M. H.; Gottlieb, M. S.;
Volberding, P. A.; Laskin, O. L.; Leedom, J . M.; Groopman, J .
E.; Mildvan, D.; Schooley, R. T.; J ackson, G. G.; Durack, D. T.;
King, D. The efficacy of azidothymidine (AZT) in the treatment
of patients with AIDS and AIDS-related complex. N. Engl. J .
Med. 1987, 317, 185-191.
(3) Yarchoan, R.; Thomas, R. V.; Allain, J . P.; McAtee, N.; Dubinsky,
R.; Mitsuya, H.; Lawley, T. J .; Safai, B.; Myers, C. E.; Perno, C.
F.; Klecker, R. W.; Wills, R. J .; Fischl, M. A.; McNeely, M. C.;
Pluda, J . M. L.; Leuther, M.; Collins, J . M.; Broder, S. Phase I
studies of 2′,3′-dideoxycytidine in severe human immunodefi-
ciency virus infection as a single agent and alternative with AZT.
Lancet 1988, I, 76-81.
(4) Cooley, T. P.; Kunches, L. M.; Saunders, C. A.; Ritter, J .; Perkins,
C. J .; McLaren, C.; Path, M. R. C.; McCaffrey, R. P.; Liebman,
H. A. Once-daily administration of 2′,3′-dideoxyinosine (ddI)
in patients with the acquired immunodeficiency syndrome or
AIDS-related complex. N. Engl. J . Med. 1990, 322, 1340-
1345.
(5) Staszewski, S.; Hill, A. M.; Bartlett, J .; Eron, J . J .; Katlana, C.;
J ohnson, J .; Sawyer, W.; McDade, H. Reductions in HIV-1
Disease Progression for Zidovudine Lamivudine Relative to
Control Treatments. AIDS 1997, 11, 477-83.
(6) Mathe, G.; Colasante, V.; Morette, C.; Hallard, M.; Blanquet,
D. Will killing the last HIV-1 particle cure AIDS patients?
Biomed. Pharmacother. 1996, 50, 473-479.
(7) Craig, C.; Moyle, G. The development of resistance of HIV-1 to
zalcitabine. AIDS 1997, 11, 271-279.
(8) Iversen, A. K. N.; Shafer, R. W.; Wehrly, K.; Winters, M. A.;
Mullins, J . I.; Chesebro, B.; Merigan, T. Multidrug-resistant
human immunodeficiency virus type 1 strains resulting from
combination antiretroviral therapy. J . Virol. 1996, 70 (2), 1086-
1090.
(9) Katlama, C. Consequences for the management of cytomega-
lovirus. AIDS 1996, 10, 543-546.
(10) Engstrom, R. E.; Holland, G. N. Local therapy for cytomega-
lovirus retinopathy. Am. J . Ophthalmol. 1995, 120 (3), 376-
385.
(13) Anderson, G. L.; Rizkalla, B. H.; Broom, A. D. Synthesis of some
tricyclic nucleosides related to the “Y” base of tRNA. J . Org.
Chem. 1974, 39, 937-939.
(14) Gilchrist, T. L. Heterocyclic Chemistry, 2nd ed.; J ohn Wiley &
Sons: New York, 1992; p 286.
(15) Ueda, T.; Boryski, J . Synthesis from Guanosine via 4-Demeth-
ylwyosine. In Nucleic Acid Chemistry, Part 4; Townsend, L. B.,
Tipson, R. S., Eds.; J ohn Wiley & Sons: New York, 1991; pp
248-251.
(16) Yoshikawa, M.; Kato, T.; Takenishi, T. Studies of Phosphoryla-
tion III. Selective phosphorylation of unprotected nucleosides.
Bull. Chem. Soc. J pn. 1969, 42, 3505-3508.
(17) Hoard, D. E.; Ott, D. G. Conversion of Mono- and Oligodeoxyri-
bonucleotides to 5′-Triphosphates. J . Am. Chem. Soc. 1965, 87,
1785-1788.
(18) Yano, J .; Kan, L. S.; Ts’o, P. O. P. A simple method of the
preparation of 2′-O-Methyladenosine - Methylation of adenosine
with methyl iodide in anhydrous alkaline medium. Biochim.
Biophys. Acta 1980, 629, 178-183.
(19) Robins, M. J .; Wilson, J . S.; Sawyer, L.; J ames, M. N. G. Nucleic
acid related compounds. 41. Restricted furanose conformations
of 3′,5′-O-(1,1,3,3,-tetraisopropyldisilox-1,3-diyl)nucleosides pro-
vide a convenient evaluation of anomeric configuration. Can. J .
Chem. 1983, 61, 1911-1920.
(20) Kierzek, R.; Caruthers, M. H.; Longfellow, C. E.; Swinton, D.;
Turner, D. H.; Freier, S. M. Polymer-Supported RNA Synthesis
and its application to test the nearest-neighbor model for duplex
stability. Biochemistry 1986, 25, 7840-7846.
(21) Huffman, J . H.; Sidwell, R. W.; Morrison, A. G.; Coombs, J .;
Reist, E. J . Structure-activity relationship of phosphonic acid
analogues of acyclovir or ganciclovir against human cytomega-
lovirus in MRC-5 cells. Nucleosides Nucleotides 1994, 13 (1-3),
607-613.
J M9802057