Antiviral amphipathic oligo- and polyribonucleotides: Analogue development and biological studies

Article abstract of DOI:10.1021/jm0203276

A series of novel N1 alkylated purine nucleic acids were polymerized either enzymatically or by automated synthesis to further establish the SAR requirements for HIV, RT, and HCMV activity. Out of the series, two constructs, 2′-O-methyl-1-allylinosinic ac

Full text of DOI:10.1021/jm0203276

1878
J . Med. Chem. 2003, 46, 1878-1885
An tivir a l Am p h ip a th ic Oligo- a n d P olyr ibon u cleotid es: An a logu e Develop m en t
a n d Biologica l Stu d ies
Robyn M. Hyde,† Arthur D. Broom,*,† and Robert W. Buckheit, J r.‡
Department of Medicinal Chemistry, 201 Skaggs Hall, University of Utah, 30 South 2000 East,
Salt Lake City, Utah 84112-5820, and TherImmune Research Corporation, 620 Professional Drive,
Gaithersburg, Maryland 20879
Received J uly 26, 2002
A series of novel N1 alkylated purine nucleic acids were polymerized either enzymatically or
by automated synthesis to further establish the SAR requirements for HIV, RT, and HCMV
activity. Out of the series, two constructs, 2′-O-methyl-1-allylinosinic acid phosphorothioate
33-mer (16) and an oligomer incorporating 1-propyl-6-thioinosinic acid residues (20), were found
to be highly active under all three assay conditions. SAR studies indicate that sulfur
incorporation, high molecular weight, and low steric bulk at N1 all can be important for activity.
In tr od u ction
The human immunodeficiency virus (HIV) is a geneti-
cally diverse infectious agent, and its ability to rapidly
develop resistance toward standard front line treatment
has left physicians scrambling for new therapeutic
options. Because of mutational changes within the HIV
genome, mostly resulting from the infidelity of reverse
transcriptase (RT),1,2 mutant particles are able to
withstand drug pressure, survive, and proliferate. It has
been documented that some resistant strains are in fact
more fit than wild-type HIV.3 Only three classes of
drugs, the nucleoside reverse transcriptase inhibitors
(NRTIs), the non-nucleoside RT inhibitors (NNRTIs)
and the protease inhibitors (PIs) are currently approved
for use in the U.S. Numerous resistant mutants to each
of these drugs have been characterized.4 For several
F igu r e 1. Anti-HIV active oligo- and polyribonucleotides.
drugs it has been shown that a single mutation will
greatly diminish efficacy.5,6 Salvage treatment after
methyl-6-thioinosinic acid) (1, Figure 1). As a group,
initial treatment failure is not always effective because
these “senseless” polymers, so-called because of their
within each of the three classes of FDA-approved drugs
there is significant cross-resistance.7,8 The change in
treatment regimen from one drug to another within the
same class may not always lead to viral suppression.
Because of the impressive mutation ability of HIV, it is
imperative that drugs with diverse structures and
unique targets be developed. For both first and second
line management of HIV infections, new drugs are
needed that lack cross-resistance with existing agents
and are slow to select resistant clones.
lack of Watson-Crick hydrogen bonding potential, form
a new class of anti-HIV agents. Their nucleic acid
macromolecular structure and proposed mechanism(s)
of action significantly differ from drugs currently used.
The pursuit of these macromolecules as opposed to small
molecules is warranted because (1) the proposed mech-
anism of action will be less likely to select for resistant
strains and (2) they have demonstrated equipotent
activity against HIV and human cytomegalovirus
(HCMV). A significant advantage over current HCMV
therapy would be realized if there were the ability to
treat both HIV and HCMV with a single agent. HCMV
is the most prevalent opportunistic infection (OI) ob-
served in HIV patients,13,14 and current therapy suffers
from toxicity limitations.15,16
Previous work identified nucleic acid polymers, espe-
cially those incorporating thiolated purines, that were
active against viral RT.9-11 It was hypothesized that
these constructs act as inhibitors of RT by binding as
“antitemplates” within the active site and blocking the
transcription of viral RNA.12 Studies in this lab focused
on the development of other novel oligo- and polyribo-
nucleotides as anti-HIV agents and identified several
potent inhibitors including the lead compound poly(1-
Until recently, the SAR that has guided analogue
development of the “senseless” macromolecules empha-
sized the requirement for amphipathicity and the ability
to form secondary structure in solution and highlighted
the preference for long polymer length and sulfur
incorporation.17-19 Unexpectedly, it was discovered that
poly(1-propargylinosinic acid) (3) had submicromolar
* To whom correspondence should be addressed. Phone: (801)581-
5764. Fax: (801)581-7087. E-mail: arthur.broom@hsc.utah.edu.
† University of Utah.
‡ TherImmune Research Corporation.
10.1021/jm0203276 CCC: $25.00 © 2003 American Chemical Society
Published on Web 04/04/2003

Oligo- and Polyribonucleotides
J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 10 1879
Sch em e 1a
F igu r e 2. Target oligo- and polyribonucleotides.
activity against HIV though it lacked both secondary
structure and sulfur.20
P r oject Ra tion a le
a
Reagents: (a) dimethylacetamide, 1,8-dizabicyclo[5.4.0]undec-
Proposed for this study were two series of compounds
based on derivative 3 that would be used to further
investigate the SAR of inosinic acid analogues. The first
series would include polymers (∼300-mers) prepared
enzymatically with polynucleotide phosphorylase
(PNPase) from nucleoside diphosphate derivatives modi-
fied at the N1 position with either the allyl or propyl
group. The second series would include oligomers (33-
mers) prepared by automated synthesis from protected
nucleoside derivatives with N1 modifications of pro-
pargyl, allyl, or propyl and additional modifications at
the 2′,6 and nonbridging oxygen positions (Figure 2).
Although the established SAR suggested that the
polymers of series 1 would be intrinsically more active
than the oligomers of series 2, the synthesis and
biological evaluation of oligomers were added to the
project to further study this factor and to exploit the
advantages of automated synthesis. Automated synthe-
sis of polymers is less expensive, and the chemistry
involved has lower substrate specificity. It was thought
that although some activity may be sacrificed by having
a shortened polymer, the reduction in activity may be
overcome by the incorporation of other groups into the
base, sugar, and backbone of the oligomers that would
not be possible if PNPase were used for polymerization.
Proposed were 2′-O-methyl and phosphorothioate sub-
stitutions to increase the stability of the prepared
oligoribonucleotides21,22 and to improve the delivery
kinetics and half-lives of the drugs. Also proposed was
a 6-thio substitution to satisfy the SAR preference for
sulfur. Earlier attempts to incorporate sulfur at the C6
position of the propargyl20 or allyl23 derivatives failed,
but it was expected that the synthesis of 2′-O-methyl-
1-propyl-6-thioinosinic acid would be possible.
7-ene, allyl bromide; (b) dimethylacetamide, 1,8-dizabicyclo[5.4.0]un-
dec-7-ene, propyl bromide; (c) POCl3, PO(OEt5); (d) dimethylform-
amide, carbonyl diimidazole, (nBuNH3+)OPO3H2-; (e) E. coli
PNPase, Tris-HCl, pH 9, 37 °C.
Sch em e 2a
a
Reagents: (a) dimethylformamide, NaH, iodomethane; (b)
H2O, acetic acid, sodium nitrite; (c) dimethylacetamide, 1,8-
dizabicyclo[5.4.0]undec-7-ene, propargyl bromide; (d) dimethylac-
etamide, 1,8-dizabicyclo[5.4.0]undec-7-ene, allyl bromide; (e) di-
methylacetamide, 1,8-dizabicyclo[5.4.0]undec-7-ene, propyl bro-
mide; (f) pyridine, DMTrCl; (g) acetonitrile, diisopropylammonium
tetrazolide, 2-cyanoethyl tetraisopropylphosphorodiamidite; (h)
automated RNA synthesis.
ride in chilled tributylamine.24 Activation of the 5′-
monophosphate with carbonyldiimidazole followed by
treatment with tri(n-butyl)ammonium phosphate gave
the 5′-diphosphate derivatives 8a and 8b.25 Polymeri-
zation with PNPase20 at basic pH afforded poly(1-
allylinosinic acid) (9a ) with an approximate molecular
weight of 1 × 105 as estimated by size exclusion HPLC.18
1-Propylinosine diphosphate (8b) was not a substrate
for PNPase, and poly(1-propylinosinic acid) was not
formed.
The starting material for oligonucleotide synthesis,
2′-O-methylinosine (11), was obtained in two steps.
Adenosine was alkylated with iodomethane after a 1 h
NaH treatment in dimethylformamide26 to give 2′-O-
methyladenosine (10). This purified product was dia-
zotized with nitrous acid generated in situ and solvo-
Ch em istr y
Except for those derivatives incorporating sulfur, the
strategies used to synthesize each polymer in series 1
and oligomer in series 2 were identical; only the alky-
lating reagent was changed to afford the different
analogues (Schemes 1 and 2). Polymer synthesis began
with the alkylation of inosine under basic conditions
with either allyl bromide or propyl bromide to give
1-allylinosine (6a ) and 1-propylinosine (6b), respec-
tively. Each nucleoside was converted to its 5′-mono-
phosphate derivative in reaction with phosphoryl chlo-

1880 J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 10
Ta ble 1. SAR Relationship
Hyde et al.
compd
R
R2
X
Y
amphipathicity
2° structure
sulfur
chain length
320
9a
15a
15b
15c
16
CH2CCH
CH2CHCH2
CH2CCH
CH2CHCH2
CH2CH2CH3
CH2CHCH2
CH2CH2CH3
H
H
O
O
O
O
O
O
S
O
O
O
O
O
S
X
X
X
X
X
X
X
∼300-mer
∼300-mer
33-mer
33-mer
33-mer
CH3
CH3
CH3
CH3
CH3
X
X
X
X
X
33-mer
33-mer
20
O
Sch em e 3a
Ta ble 2. Activity of Target Compounds against HIV in
CEM-SS Cells and against Cell-Free RT
compd
HIV EC50 (µM)
RT IC50 (µM)
320
9a
15a
15b
15c
16
0.13
0.94
0.26
0.32
4.97
0.61
0.14
0.0223
9.6
0.15
50.5
6.7
0.039
0.82
20
ammonia under pressure to remove the acetate protect-
ing groups. Nucleoside was then derivatized at the 3′
and 5′ positions appropriately for automated oligonucle-
otide synthesis. Owing to low yields at each of the
synthetic steps, there was not enough product to syn-
thesize a thio homopolymer, and instead, a copolymer
using both 2′-O-methylinosinic acid and 2′-O-methyl-6-
thioinosinc acid residues was made. Automated synthe-
sis was conducted as before.
a
Reagents: (a) pyridine, acetic anhydride; (b) pyridine, Lawes-
son’s reagent, reflux; (c) methanolic ammonia; (d) pyridine,
DMTrCl; (e) acetonitrile, diisopropylammonium tetrazolide, 2-cya-
noethyl tetraisopropylphosphorodiamidite; (f) automated RNA
synthesis.
Resu lts a n d Discu ssion
lyzed to give 11.27 Starting material 11 was alkylated
using propargyl, allyl, or propyl bromide under basic
conditions to give 12a , 12b, and 12c, respectively. Each
modified nucleoside was then prepared for automated
synthesis by protecting the 5′-hydroxyl group with
dimethoxytrityl and derivatizing the 3′-hydroxyl as its
cyanoethyldiisopropylphosphoroamidite.28 The 33-mer
oligomers were synthesized by automated synthesis
from a solid support resin conjugated to a 3′-thymidine
using a 25 min coupling time and oxidation with tert-
butyl peroxide. Oligomer 16 with a phosphorothioate
backbone was made by automated synthesis similarly
to 15b except a Beaucage reagent29 was used for
oxidation instead of tert-butyl peroxide. All oligomers
were removed from the resin by a 1 h treatment with
ammonium hydroxide at room temperature and purified
from their failure sequences.30
To synthesize the 6-thio nucleoside precursor 2′-O-
methyl-1-propyl-6-thioinosine (17), 12c was treated with
acetic anhydride to protect the 3′ and 5′ hydroxyl groups
and dry crude product was refluxed in pyridine with
Lawesson’s reagent (Scheme 3). After removal of excess
pyridine from the reaction at termination, the crude oil
was dissolved in methanol and treated with methanolic
An ti-HIV Activity. In comparison with the previ-
ously established SAR, each of the constructs made for
this project had two of the highlighted structural
characteristics. Each was amphipathic, having a hydro-
philic backbone and hydrophobic base, and each had one
additional characteristic (Table 1). None of the macro-
molecules fully adhered to the established SAR. It is
not known if the secondary structure observed in the
oligomers 15a -c is significant. Though melting transi-
tions were observed, the Tm was below 11 °C and was
not highly cooperative.31
All senseless macromolecules were tested for their
ability to protect cells from the pathogenicity of HIV.
Additionally, they were tested for their ability to inhibit
the polymerization activity of RT in a cell-free environ-
ment. As shown in Table 2, most constructs had sub-
micromolar activity against HIV in a CEM-SS cell line
using wild-type RF virions initially acquired from the
NIH AIDS Research and Reference Reagent Program.
The exception was 2′-O-methyl-1-propylinosinic acid 33-
mer (15c) that had activity roughly equivalent to 5 µM.
Fitting with the SAR, the polymer 3 was more active
than the oligomers 15a -c; however, activity was res-
cued in the oligomer series by the incorporation of sulfur

Oligo- and Polyribonucleotides
J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 10 1881
Ta ble 3. Activity of Target Compounds against HCMV in
MRC-5 Cells
for the anti-HCMV activity, full understanding of the
SAR is not possible.
compd
EC50 (µM)
Con clu sion s
ganciclovir
320
15a
9.4
0.019
0.56
Though the inosinic acid based compounds synthe-
sized for this project did not strictly adhere to the
previously established SAR, the most active compounds
retained significant anti-viral activity. The data do not
clearly support RT inhibition, however, which was the
prevailing hypothesis regarding the mechanism of ac-
tion. The previously established SAR may have identi-
fied those compounds that do primarily act by RT
inhibition. These active constructs that are at variance
with the SAR may primarily act by some other mech-
anism of action. There is some indication that the
mechanism of action does include a specific interaction
because steric bulk at N1 limits the inhibitory capabili-
ties of the constructs. Additionally, these studies show
that sulfur is important, though it is not known exactly
what role sulfur plays in any interaction. Other mech-
anisms of action have been proposed for HIV activity
such as inhibition of virion attachment and uptake and
interference with viral uncoating.33 Further analogue
development and viral assays will be required to fully
understand the SAR and mechanism of action.
15b
15c
20
inactive
inactive
1.3
into the construct. Of those compounds made for this
project, the sulfur compounds 16 and 20 were the most
active.
In general, the propargyl analogues were more active
than the allyl or propyl analogues. It is more likely that
the effective space occupied by each group contributed
to the activity difference rather than to the degree of
hydrophobicity. The propargyl group has limited flex-
ibility and therefore a smaller effective steric volume
than the two other groups. Correspondingly, the pro-
pargyl derivatives were the most active. This may
indicate that the mechanism of action requires the base
to fit into a constrained active site. This observation
would be consistent with the hypothesis that these
constructs act as RT inhibitors.
On the other hand, although the whole-cell activities
of the most active constructs were approximately equiva-
lent, the RT activity was not. Generally, the RT activity
was poor except for constructs 15a and 16. This may
indicate that the current compounds act primarily by
mechanisms of action other than RT inhibition. Clearly,
9a , 15b, 15c, and 20 cannot act as RT inhibitors in the
cell. The IC50 values were higher than the EC50 values,
which precludes them from being realistic RT inhibitors
in the in vivo system.
Each of the polymers was relatively nontoxic to cells.
Though the CD50 has not been well explored for each
compound, cell viability was only observed to drop below
96% for 15a (50%, 50.3 µM) and 15c (77%, 16 µM).32
An ti-HCMV Activity. A very curious quality about
the “senseless” compounds is that they have potent
activity against HCMV. HCMV, a DNA virus, has little
in common with HIV, an RNA virus. Though the two
viruses are enveloped and have similar viral entry
processes, the remainder of the life cycle is quite
different. Of the senseless macromolecules tested, 15a
and 20 had activity against HCMV and that activity was
nearly equipotent with HIV activity. Even though 20
was only weakly active, it was still severalfold more
potent than ganciclovir, the current first-line therapeu-
tic agent (Table 3).
Until now, the SAR profile for anti-HCMV activity of
the senseless compounds has been loosely related to the
SAR for the anti-HIV activity. Of the previously made
compounds, those that had anti-HIV activity also had
anti-HCMV activity. This study demonstrates for the
first time dissimilarity in antiviral activity. Both 15b
and 15c had quite good anti-HIV activity; however, they
were inactive against HCMV. The most reasonable
explanation is that the cellular target has a more
stringent requirement regarding maximal steric volume
allowed at the N1 position of the purine. This cannot
be the whole story, though, since the inclusion of sulfur
in the purine with 25 salvages the activity. Until further
work is done that elucidates the mechanism of action
Exp er im en ta l Section
Gen er a l. Pyridine was dried by distillation over CaH2.
Anhydrous dimethylacetamide (DMAC), dimethylformamide
(DMF), and methanol (MeOH) were purchased from Aldrich.
All other chemicals were purchased as ACS or reagent grade
and used without further purification. Thin-layer chromatog-
raphy was conducted on silica gel coated Whatman aluminum
backed flexible plates containing fluorescent indicator F254
.
Spots were visualized with short- and long-wavelength UV as
appropriate. Slide-A-Lyzer cassettes (MWCO 10,000) from
Pierce were used for dialysis. Gravity and flash silica gel
chromatography were performed with particle size 63-200 gel
from Selecto Scientific and 170-400 from Fisher, respectively.
Adsorption chromatography was performed with Amberlite
XAD-4 resin from Sigma. Reverse-phase chromatography was
performed with Poly-Pac II cartridges from Glen Research.
Cation exchange and anion exchange chromatography were
performed with Bio-Rad resins AG 50W-X8, 200-400 mesh
and AG 1-X8, 200-400 mesh, respectively. Gel filtration
chromatography was performed with Bio-Gel P2 (fine) from
Bio-Rad. Ion exchange and gel filtration chromatography used
Bio-Rad Econo-Columns, Wiz pump/diluter/dispenser, ISCO
type 6 optical unit (254 nm) and ISCO UA-5 absorbance
monitor. HPLC chromatography was preformed with a Hitachi
L6200 pump equipped with a L3000 photodiode array. Size
exclusion and reverse-phase HPLC were accomplished with
the Bio-Rad Bio-Sil SEC-125 (Bio-Sil 125 guard) and Rainin
Microsorb-MV 100 Å C8 columns, respectively. High-resolution
mass spectra were recorded with a MAT 95 spectrometer.
Proton, carbon, and phosphorus NMR were recorded with
either a Brucker AF200 MHz, Varian Mercury 400 MHz FT-
NMR or Unity 500 MHz NMR spectrometer. Phosphorus NMR
spectra were referenced to an external phosphoric acid stan-
dard. UV spectra and Tm measurements were recorded with
a Hewlett-Packard 8452A diode array spectrometer equipped
with a Peltier variable-temperature controller.
HP LC P r otocol. Size exclusion (Bio-Rad Sec-125) param-
eters are the following: 100%; phase a, 0.05 M phosphate
buffer + 0.15 M NaCl + 0.01 M NaN3. Ion exchange (partisil-
SAX) parameters are the following: 0-100%, 50 min; phase
a, 0.01 M acetic acid, 6 mM potassium phosphate monobasic,
pH 5; phase b, 0.6 M potassium phosphate monobasic, pH 4.
Reverse-phase (C8) parameters are the following: 0-60%, 45
min; phase a, acetonitrile; phase b, 0.1 M ammonium acetate.

1882 J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 10
Hyde et al.
1-Allyllin osin e (6a ). Allyl bromide (8.0 mL, 91.5 mmol) and
1,8-dizabicyclo[5.4.0]undec-7-ene (DBU) (14.0 mL, 91.9 mmol)
were added to a suspension of inosine (18.7 g, 69.78 mmol) in
DMAC (400 mL). The reaction mixture was stirred at room
temperate under argon overnight, at which time TLC in
chloroform/methanol (85:15) demonstrated the disappearance
of starting material. The reaction was terminated by addition
to 3.5 L of ether/hexane (1:1). The resulting suspension was
placed at -15 °C overnight. While still chilled, the solvent was
decanted from the resulting gum that was then dissolved in
methanol. The solution was evaporated under reduced pres-
sure onto silica gel. The product was isolated by silica gel
chromatography using chloroform/methanol (95:5). Isolated
yield was 18.0 g (83.7%) of a white foam. UV ꢀmax 9403 at 250
aqueous/organic interface. The aqueous solution was dialyzed
against 0.1 M NaCl (12 L, 24 h) and H2O (24 L, 48 h).
Lyophilization of the aqueous solution gave poly(1-allylinosinic
acid) (11.5 mg, 10.1%) as a fluffy, white solid. UV ꢀmax 7477 at
250 nm (0.1 M NaCl, 0.1 M phosphate buffer, pH 6.8); HPLC
retention time 7.3 min (size exclusion protocol); homogeneous
by HPLC.
2′-O-Meth yla d en osin e (10). The reaction of adenosine (18
g, 67.4 mmol) in 430 mL of DMF treated with NaH (2.9 g)
and iodomethane (3.1 mL, 50.0 mmol) as described earlier34
yielded 12.6 g (66.5%) of a white foam. FABMS (glycerol and
H2O) m/z (MH+) calcd for C11H16N5O4 282.1, found 282.
2′-O-Meth ylin osin e (11). 2′-O-Methyladenosine 10 (7.6 g,
26.9 mmol) was reacted with 400 mL of H2O, 18.2 mL of acetic
acid, and 14.6 g of sodium nitrite (211.6 mmol) to give 11 as
described earlier.35 The product was isolated by adsorption
chromatography using ethanol (0-30%). The isolated yield was
6.1 g (80.4%) of a white solid. FABMS (thioglycerol [1% TFA],
MeOH), m/z (M + Na+) calcd for C11H14N4O5Na 305.1, found
305.
2′-O-Meth yl-1-p r op a r gylin osin e (12a ). To a solution of
2′-O-methylinosine (11) (6.0 g, 18.8 mmol) in 120 mL of DMAC
were added DBU (3.0 mL, 19.7 mmol) and propargyl bromide
(2.4 mL, 16.6 mmol) dropwise. The solution was stirred at room
temperature for 18 h, at which time TLC showed complete
disappearance of starting material. The reaction mixture was
treated generally as for 6a . The isolated yield was 4.5 g (75.2%)
of a white foam. Purity was determined by comparison of the
peak height of starting material H1′ and product H1′ observed
in the NMR spectra. UV ꢀmax 6424 at 250 nm; 1H NMR (DMSO-
d6) δ 8.51 (1H, s, H2), 8.42 (1H, s, H8), 5.98 (1H, d, J ) 5.8
Hz, H1′), 5.29 (1H, d, J ) 5.5 Hz, 3′-OH), 5.10 (1H, t, J ) 10.9
Hz, 5′-OH), 4.84, 4.83 (2H, d, J ) 2.7, CH2), 4.30 (1H, q, J )
8.5, H2′), 4.26 (1H, t, J ) 10.6, H3′), 3.96 (1H, q, J ) 11.3,
H4′), 3.55 (1H, m, H5′), 3.53 (3H, m, OCH3), 3.45 (1H, m, H5′′),
3.40 (1H, t, J ) 5.1, CH); HRMS (glycerol and MeOH) m/z
(MH+) calcd for C14H17N4O5 321.119 89, found 321.119 99;
purity 95% by NMR.
5′-O-(4,4′-Dim et h oxyt r it yl)-2′-O-m et h yl-1-p r op a r gyli-
n osin e (13a ). To a solution of 4.5 g of dry product (12a ) (14.1
mmol) in 66 mL of pyridine was added 6.0 g of DMTrCl (17.8
mmol). The reaction mixture was vigorously stirred for 6 h
until TLC in CHCl3/MeOH (97:3) showed disappearance of
starting material. One volume of MeOH was added and
evaporated to a thick oil. The oil was dissolved in 100 mL of
CHCl3, washed three times with 15 mL of both saturated
NaHCO3 and H2O, dried over anhydrous Na2SO4, and evapo-
rated onto silica gel. The product was purified by flash
chromatography under argon by gradient elution of ethyl
acetate from 0% to 100% in hexanes in the constant presence
of 1% triethylamine. Product was recovered at 100% ethyl
acetate and evaporated to a white foam by coevaporation with
CH2Cl2. The isolated yield was 3.2 g (36.5%). HRMS (glycerol
and MeOH), m/z (MH+) calcd for C35H35N4O7 623.250 57, found
623.251 62; Rf ) 0.80 (CHCl3/MeOH, 95:5); Rf ) 0.30 (EtOAC/
hexanes, 9:1); single spot by TLC.
5′-O-(4,4′-Dim eth oxytr ityl)-2′-O-m eth yl-1-a llylin osin e
(13b). The product was prepared in two steps generally as
described for 12a and 13a . Dry product 11 (5.0 g. 17.9 mmol)
was suspended in 100 mL of DMAC. To the reaction mixture
were added DBU (2.7 mL, 17.7 mmol) and allyl bromide (2.2
mL, 25.17 mmol). The reaction mixture was stirred vigorously
under argon overnight and was then doubled in volume with
methanol and evaporated by oil pump onto silica gel. Product
1
nm; H NMR (DMSO-d6) δ 8.38 (1H, s, H2), 8.36 (1H, s, H8),
6.00 (1H, m, CH), 5.86 (1H, d, J ) 5.8 Hz, H1′), 5.48 (1H, d, J
) 6.1 Hz, 2′-OH), 5.20 (1H, d, J ) 4.8 Hz, 3′-OH), 5.10 (2H,
m, CH2), 5.06 (1H, m, 5′-OH), 4.64, 4.63 (2H, d, J ) 5.5, CH2),
4.48 (1H, q, J ) 16.7, H2′), 4.13 (1H, q, J ) 13.3, H3′), 3.94
(1H, q, J ) 11.6, H4′), 3.63 (1H, m, H5′), 3.56 (1H, m, H5′′);
FABMS (glycerol and EtOAc), m/z (MH+) calcd for C13H17N4O5
309.119 89, found 309.118 16; HPLC retention time 32.5 min
(C18, 0-15%, 25 min, phase a 0.05 M phosphate buffer, phase
b MeOH); homogeneous by HPLC.
1-Allylin osin e-5′-m on op h osp h a te (7a ). The dry solid 6a
(10 g, 32.5 mmol) was converted to 11b as described earlier
by Yoshikawa,24 yielding 14.1 g (83.0%) of white solid. The
product was used without further purification. Samples for
HRMS were obtained by cation exchange chromatography
conversion of the compound to its NH4+ salt. HRMS (glycerol
and EtOAc) m/z (M - 3NH3 - H) calcd for C13H16PN4O8
387.070 58, found 387.068 70.
1-Allylin osin e-5′-d ip h osp h a te (8a ). Monophosphate 7a (7
g, 13.4 mmol) was converted to 8a as described by Hoard and
Ott.25 The reaction mixture was quenched by doubling the
volume with MeOH, and using minimal heat, the solution was
evaporated to dryness. The resulting residue was dissolved in
200 mL of H2O, and after addition of 8.4 g of lithium acetate,
the pH was adjusted to 12.0 with 3 N LiOH while being chilled
in an ice bath. The solution was immediately filtered through
Celite, and the filtrate was adjusted to pH 7.0 with 1 N HCl
while chilled in an ice bath. After the filtrate was concentrated
to approximately 100 mL, 8a was precipitated with addition
of 1 L of ethanol/acetone (1:1). The precipitate was isolated
by centrifugation and dried in
a vacuum oven at room
temperature. Crude diphosphate was purified by anion ex-
change chromatography using LiCl (0-0.3 M) in 0.003 M HCl.
Pooled fractions containing product as identified by ion-
exchange HPLC were adjusted to pH 7.0 with 1 M HCl and
concentrated to approximately 300 mL by rotary evaporation.
Diphosphate was precipitated in 2.4 L of ethanol/acetone
(1:1) and, following 48 h of refrigeration, was isolated by
centrifugation and dried in a vacuum oven at room tempera-
ture. When H2O was used as eluent, diphosphate was con-
verted to its Na+ salt by cation exchange chromatography and,
after lyophilization, desalted by gel filtration. The product was
isolated as a white fluffy solid (1.1 g, 15.4%). Samples for
HRMS were obtained by cation exchange chromatography
conversion of the compound to its NH4+ salt. HRMS (glycerol
and MeOH) m/z (M - 3NH3 - H) calcd for C13H16PN4O8
467.036 91, found 467.036 26; HPLC retention time 25.7 min
(ion exchange protocol); homogeneous by HPLC.
P oly(1-a llylin osin ic a cid ) (9a ). A solution comprising the
following components was incubated at 37 °C for 6 h with
gentle rocking: 1.0 mL of Tris-HCl (pH 9.0, 2 M); 1.0 mL of
MgCl2 (0.1 M); 1.0 mL of 2-mercaptoethanol (2%); 4.0 mL of
H2O; 75 IU of PNPase (E. coli) in 0.6 mL of buffer containing
50% glycerol, 5 mM Tris, and 150 mg of diphosphate 8a . After
incubation was complete as determined by size exclusion
HPLC,18 the reaction mixture was diluted to 20 mL with H2O
and adjusted to pH 7.0 with 0.1 M HCl. All protein was
removed from the aqueous reaction mixture by several extrac-
tions with chloroform/isoamyl alcohol (5:2). Extraction was
repeated until no precipitated protein was apparent at the
was eluted from
(95:5).
a silica gel column with CHCl3/MeOH
Oil (12b) was coevaporated three times with pyridine to
remove all traces of water. To the suspension of oil in pyridine
(85 mL) was added dimethoxytrityl chloride (7.8 g, 23.0 mmol).
The reaction was completed, and the product was purified as
described for 13a . The isolated yield was 4.8 g (43.0%). HRMS
(glycerol [1% TFA] and MeOH), m/z (MH+) calcd for C35H37N4O7
625.266 22, found 625.263 98; Rf ) 0.42 (CHCl3/MeOH, 95:5);
Rf ) 0.20 (EtOAC/hexanes, 9:1); homogeneous by TLC.

Oligo- and Polyribonucleotides
J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 10 1883
5′-O-(4,4′-Dim et h oxyt r it yl)-2′-O-m et h yl-1-p r op ylin o-
sin e (13c). Product was prepared from 11 (3.0 g. 10.6 mmol)
as described for 13b. Isolated yield was 2.3 g (34.7%). FABMS
(glycerol and MeOH), m/z (MH+) calcd for C35H39N4O6
627.264 12, found 627.266 88; Rf ) 0.40 (CHCl3/MeOH, 95:5);
Rf ) 0.18 (EtOAC/hexanes, 9:1); single spot TLC.
125 69, found 12 570.5; HPLC retention time 20.2 min (reverse
phase protocol); homogeneous by HPLC.
2′-O-Meth yl-1-p r op ylin osin ic Acid 33-m er (15c). 33-mer
was synthesized using 14c on
a 1 µmol scale with an
automated DNA synthesizer as described for 15a and 15b. The
isolated yield was 7.9 mg of fluffy white solid (62.7%). UV ꢀmax
7060 at 252; MALDI calcd 12 565, found 12 565.9; HPLC
retention time 24.4 min (C8 protocol); homogeneous by HPLC.
5′-O-(4,4′-Dim et h oxyt r it yl)-2′-O-m et h yl-1-p r op a r gyl-
in osin e-3′-O-(2-cya n oeth yl)-N,N-d iisop r op ylp h osop h or -
a m id ite (14a ). To a vigorously stirring solution of product (1.2
g, 1.93 mmol) (13a ) in 15 mL of acetonitrile was added 91.3
mg of diisopropylammonium tetrazolide and 1 g of 2-cyano-
ethyl tetraisopropylphophorodiamidite (3.22 mmol). The salt
was prepared by adding diisopropylamine (1 mL, 7.14 mmol)
to a solution of tetrazole (206 mg, 2.94 mmol) in 11 mL of dry
acetonitrile and isolated by filtering the precipitate that formed
after 10 min. The reaction mixture was stirred overnight under
argon and verified to be complete by TLC in CHCl3/MeOH (97:
3). The reaction volume was doubled with saturated NaHCO3
and extracted three times with 25 mL of ethyl acetate. The
organic layer was washed two times with 25 mL of brine and
once with H2O, dried over anhydrous Na2SO4, and evaporated
to a thick oil. The product was purified by flash chromatog-
raphy under argon by gradient elution of ethyl acetate from
0% to 100% in hexanes in the constant presence of 1%
triethylamine. Product was recovered at 80-100% ethyl
acetate and evaporated to a white foam by coevaporation with
CH2Cl2. The isolated yield was 1.3 g (81.9%). 31P NMR (200
MHz, CD3CN) δ 151.96, 151.18; HRMS [3-nitrobenzyl alcohol
(NBA) and CHCl3], m/z (MH+) calcd for C44H52PN6O8 823.358 43,
found 823.361 59; Rf ) 0.72, 0.6 (CHCl3/MeOH, 95:5); Rf )
0.75, 0.67 (EtOAC/hexanes, 9:1)
2′-O-Met h yl-1-a llylin osin e P h osp h or ot h ioa t e 33-m er
(16). 33-mer was synthesized with 14b on a 1 µmol scale with
an automated DNA synthesizer as described for 15a ; however,
tetraethylthiuram disulfide was used for oxidation to create
the phosphorothioate backbone. The isolated yield was 6.4 mg
(38.1%) of a white fluffy solid. UV ꢀmax 8775 at 252 nm; MALDI
(M + H) calcd 13 048, found ∼13 000. HPLC retention time
22.7 min (broad peak) (reverse-phase protocol); homogeneous
by HPLC.
2′-O-Meth yl-1-p r op yl-6-th ioin osin e (17). To a solution of
2′-O-Me-1-propylinosine (17c, 2.1 g, 6.5 mmol) dissolved in 25
mL of pyridine was added 10.5 mL of acetic anhydride. TLC
in CHCl3/MeOH (90:10) indicated that the reaction was
complete in 3 h. Crude reaction mixture was coevaporated
three times with toluene and then brought up in 50 mL of
CHCl3. The organic layer was washed three times with 15 mL
of 0.1 N HCl and twice with 15 mL of H2O. The aqueous
solution was back-extracted once with 15 mL of CHCl3, and
then the entire CHCl3 solution was dried over Na2SO4 and
rotary-evaporated to a thick oil.
Protected nucleoside was coevaporated three times with dry
pyridine and then dissolved in 50 mL of pyridine and refluxed
with 1 g of Lawesson’s reagent (2.5 mmol) overnight. TLC in
CHCL3/MeOH (90:10) showed the disappearance of all staring
material. Once cooled, the reaction mixture was evaporated
three times with toluene to remove all pyridine. The resulting
gum was dissolved in 50 mL of methanol, and insoluble black
material was removed by filtration. The filtrate was evapo-
rated, dissolved in 100 mL of MeOH, doubled in volume with
methanolic ammonia, placed in a Parr bomb, and sealed. The
bomb was left sealed for 48 h, and then pressure was slowly
released. The methanol solution was dried onto silica gel.
5′-O-(4,4′-Dim eth oxytr ityl)-2′-O-m eth yl-1-a llylin osin e-
3′-O-(2-cya n oet h yl)-N,N-d iisop r op ylp h osop h or a m id it e
(14b). Dry product 13b (2.5 g, 4.0 mmol) was treated as
described for 14a . The isolated yield was 2.3 g (69.8%) of a
white fluffy foam. 31P NMR (DMSO-d6) δ 150.71, 150.53;
HRMS (glycerol) m/z (MH+) calcd for C44H54PN6O8 825.374 08,
found 825.372 07; Rf ) 0.69, 0.58 (CHCl3/MeOH, 95:5); Rf )
0.60, 0.55 (EtOAC/hexanes, 9:1).
5′-O-(4,4′-Dim eth oxytr ityl)-2′-O-m eth yl-1-pr opylin osin e-
3′-O-(2-cya n oet h yl)-N,N-d iisop r op ylp h osop h or a m id it e
(14c). The product was prepared from 13c (1.0 g, 1.6 mmol)
generally as described for 14a and 14b. The isolated yield was
1.2 g (90.7%). 31P NMR (DMSO-d6) δ 150.72, 150.52; HRMS
(NBA and EtOAc) m/z (MH+) calcd for C44H56PN6O8 827.389 73,
found 827.390 97; Rf ) 0.67, 0.60 (CHCl3/MeOH, 95:5); Rf )
0.65, 0.52 (EtOAC/hexanes, 9:1).
Product was purified by silica gel chromatography in
a
gradient from 0% to 2% MeOH in CHCl3. The isolated yield
was 0.6 g (27.1%) of a light-brown solid. Purity was determined
by comparison of the peak height of starting material H1′ and
product H1′ observed in the NMR spectra. UV ꢀmax 10 745 at
234 nm, 24 970 at 322 nm; 1H NMR (DMSO-d6) δ 8.79 (1H, s,
H2), 8.54 (1H, s, H8), 5.99 (1H, d, J ) 5.5 Hz, H1′), 5.29 (1H,
d, J ) 5.5 Hz, 3′-OH), 5.10 (1H, t, J ) 11.3 Hz, 5′-OH), 4.51,
4.49 (2H, t, J ) 14.7, CH2), 4.31 (1H, q, J ) 13.7, H2′), 4.26
(1H, t, J ) 10.6, H3′), 3.96 (1H, q, J ) 11.3, H4′), 3.64 (1H, m,
H5′), 3.56 (1H, m, H5′′), 3.33 (3H, s, OCH3), 1.81, 1.78 (2H, q,
J ) 22.9, CH2), 0.92, 0.90, 0.88 (3H, t, J ) 14.7, CH3); FABMS
(glycerol [1% TFA] and MeOH) m/z (MH+) calcd for C14H20N4O4S
341.128 35, found 341.128 75; purity 86% by NMR.
5′-O-(4,4′-Dim eth oxytr ityl)-2′-O-m eth yl-1-pr op yl-6-th io-
in osin e (18). Dimethoxytrityl chloride (0.65 g, 1.9 mmol) was
added to a suspension of the starting material 17 (0.50 g, 1.5
mmol) in 20 mL of pyridine, and the reaction was carried out
as generally described for 13c. The isolated yield was 0.5 g
(51.9%) of a light-brown foam. FABMS (glycerol [1% TFA] and
MeOH) m/z (MH+) calcd for C35H39SN4O6 643.259 03, found
643.257 93; Rf ) 0.43 (CHCl3/MeOH, 95:5); Rf ) 0.60 (EtOAC/
hexanes, 9:1); single spot by TLC.
5′-O-(4,4′-Dim eth oxytr ityl)-2′-O-m eth yl-1-pr op yl-6-th io-
in osin e-3′-O-(2-cya n oeth yl)-N,N-d iisop r op ylp h osop h or a -
m id ite (19). Product 18 (0.5 g, 0.80 mmol) in acetonitrile (15
mL) was treated as generally described for 14c with 60 mg of
diisopropylammonium tetrazolide salt and 0.5 g of 2-cyano-
ethyl tetraisopropylphophorodiamidite (1.61 mmol). The iso-
lated yield was 0.4 g (59.4%) of a light-brown foam. 31P NMR
(DMSO-d6) δ 150.71, 150.57; HRMS (glycerol) m/z (MH+) calcd
for C44H56PSN6O7 843.366 88, found 843.368 46; Rf ) 0.69, 0.62
(CHCl3/MeOH, 95:5); Rf ) 0.5, 0.39 (EtOAC/hexanes, 9:1);
purity >95% by TLC.
2′-O-Meth yl-1-p r op a r gylin osin ic Acid 33-m er (15a ). 33-
mer oligo was synthesized using 14a on a 1 µmol scale with
an automated DNA synthesizer. The oligomer chain was
initially extended from a 3′-T residue attached to a controlled
pore glass (CPG) solid support. Modifications to the standard
protocol included extension of the coupling time to 25 min,
oxidation of the backbone with tert-butyl peroxide, and removal
of product from the solid support by a 1 h treatment with NH4-
OH at room temperature. Crude DMTr-ON product 15a was
dried by speed evacuation at room temperature and purified
by reverse-phase chromatography. Collected product was
lyophilized and converted to its Na+ salt by ion exchange
chromatography with sodium acetate (1.0 M) and by dialysis
with 0.1 M NaCl (24 h, 8 L). The oligomer was washed (H2O,
48 h, 16 L) and lyophilized. The isolated yield was 6.7 mg
(54.2% based on a theoretical 1 µm yield) of a white fluffy solid.
The product was collected for MALDI by reverse-phase HPLC
chromatography. UV ꢀmax 8015 at 250 nm; MALDI (M + 2Na)
calcd 124 88, found 124 92; HPLC retention time 23.3 min
(reverse-phase protocol); homogeneous by HPLC.
2′-O-Meth yl-1-a llylin osin ic Acid 33-m er (15b). 33-mer
was synthesized using 14b on
a 1 µmol scale with an
automated DNA synthesizer as described for 15a . The isolated
yield was 6.9 mg (55.2%) of a fluffy white solid. The sample
for MALDI evaluation was collected by HPLC (reverse-phase
protocol). UV ꢀmax 6855 at 250 nm. MALDI (M + 3Na) calcd

1884 J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 10
Hyde et al.
(5) De J ong, M.; Veenstra, J .; Stilianakis, N. I.; Schuurman, R.;
Lange, J . M. A.; De Boer, R. J .; Boucher, C. A. B. Host-parasite
dynamics and outgrowth of virus containing a single K70R
amino acid change in reverse transcriptase are responsible for
the loss of human immunodeficiency virus type 1 RNA load
suppression by Zidovudine. Proc. Natl. Acad. Sci. U.S.A. 1996,
5501-5506.
2′-O-Meth yl-1-p r op yl-6-th ioin osin ic Acid /2′-O-Meth yl-
1-p r op ylin osin ic Acid Cop olym er 33-m er (20). 33-mer was
synthesized as described for 16c using both 14c and 19 on a
1 µmol scale with an automated DNA synthesizer. Isolated
yield was 6.2 mg (48.4%) of a fluffy light-orange solid. MALDI
(M + 4Na) calcd 12 912, found 12 928; HPLC retention time
27.7 min (reverse-phase protocol); homogeneous by HPLC.
Degr a d a tion to Sta r tin g Nu cleosid e. To solutions of 9a ,
15a -c, 16, and 20 in 0.1 M NaCl (100 µL, 2 mg/mL) was added
65 µL of a solution comprising the following: 33 µL of Tris-
HCl (2 M, pH 9.0), 44 µL of MgCl2 (0.1 M), 44 µL of venom
phosphodiesterase, and 22 µL of alkaline phosphatase. The
solution was incubated at 37 °C with gentle rocking for 18 h
and then diluted to 5 mL with NaCl (0.1 M). TLC in
chloroform/methanol (85:15) confirmed degradation of 9a to
6a as the sole nucleoside product. TLC of the oligomers 15a -
c, 16, and 20 confirmed degradation to their respective
nucleoside precursors including the 3′-T residue.
Biologica l Assa ys. An ti-HIV. Sou th er n Resea r ch In -
st it u t e. CEM-SS cells passaged in T-75 flasks were split
(1:2) 24 h prior to the time of infection. J ust prior to infection,
cells were washed twice with media, pelleted, and resuspend
with media to a concentration of 5 × 104 cells per milliliter.
Cell viability was evaluated by trypan blue exclusion, and any
culture found to have less than 95% viability was discarded.
For HIV analysis, an amount of 50 µL of the cell culture was
added to drug-containing wells and to cell control wells in a
96-well plate.
For the standard HIV assay, HIV-1RF particles initially
acquired from the NIH AIDS Research and Reference Reagent
Program were grown in CEM-SS cells for stock pools that were
subsequently frozen at -80 °C. J ust prior to infection, the
particles were slowly thawed, suspended in media, and diluted
to a concentration that would effect 85-95% cell kill in 6 days.
For analysis, an amount of 50 µL was added to drug-containing
wells and virus control wells in a 96-well plate. Additional tests
with other virus strains are conducted by the same protocol
just using the other particles for infection.
For HIV analysis, the test compound was brought up in 0.1
M NaCl, PBS, or H2O to predetermined concentrations. The
compound was aliquoted into wells at 0.32, 1.0, 3.2, 10, 32,
and 100 µL. Drug was added to wells to provide duplicate
measurements of cells + drug and drug + media and triplicate
measurements of cells + drug + media at each drug concen-
tration.
(6) Isaka, Y.; Miki, S.; Kawauchi, S.; Suyama, A.; Sugimoto, H.;
Adachi, A.; Miura, T.; Hayami, M.; Yoshie, O.; Fujiwara, T.; Sato,
A. A single amino acid change at Leu-188 in the reverse
transcriptase of HIV-2 and SIV renders them sensitive to non-
nucleoside reverse transcriptase inhibitors. Arch. Virol. 2001,
146, 743-755.
(7) O’Brian, W. A. Resistance against reverse transcriptase inhibi-
tors. Clin. Infect. Dis. 2000, 30, S185-S192.
(8) Erickson, J . W.; Gulnik, S. V.; Markowitz, M. Protease inhibi-
tors: resistance, cross-resistance, fitness and the choice of initial
and salvage therapies. AIDS 1999, 13, S189-S204.
(9) Chandra, P.; Ebner, U.; Steel, L. K.; Laube, H.; Gericke, D.;
Mildner, B. Bardos, T. J .; Ho, Y. K.; Go¨tz, A. Molecular
approaches to inhibit oncogenesis by RNA tumor viruses. Ann.
N. Y. Acad. Sci. 1977, 284, 444-462.
(10) Arya, S. K.; Helser, T. L.; Carter, W. A. Polyxanthylic and
polyguanylic acid inhibition of murine leukemia virus activities.
Mol. Pharmacol. 1976, 12, 844-853.
(11) De Clercq, E.; Billiau, A.; Hattori, M.; Ikehara, M. Inhibition of
oncornavirus functions by poly(2-methylinosinic acid). Nucleic
Acids Res. 1975, 2, 2305-2313.
(12) Bardos, T. J . Antimetabolites: molecular design and mode of
action. Top. Curr. Chem. 1974, 52, 63-98.
(13) Freeman, W. R. New developments in the treatment of CMV
retinitus. Ophthalmology 1996, 103, 999-1000.
(14) Cheung, T. W.; Teich, S. A. Cytomegalovirus infection in patients
with HIV infection. Mt. Sinai J . Med. 1999, 66, 113-124.
(15) J acobsen, M. A. Update (1995) on clinical trials of antiviral
therapy and prophylaxis for AIDS-related cytomegalovirus
disease. Scand. J . Infect. Dis. Suppl. 1995, 99, 96-99.
(16) Walmsley, S.; Tseng, A. Comparative tolerability of therapies
for cytomegalovirus retinitis. Drug Saf. 1999, 21, 203-224.
(17) 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.
(18) 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.
(19) Tutonda, M. G.; Buckheit, R. W.; Agrawal, V. K.; Broom, A. D.
Antiviral oligo and polyribonucleotides containing selected tria-
zolo[2,3-a]purines. J . Med. Chem. 1982, 25, 4958-4964.
(20) Hyde, R.; Buckheit, R. W., J r.; Broom, A. D. “Senseless” antiviral
polyribonucleotides: poly(1-propargyl-inosinic acid). Nucleosides,
Nucleotides Nucleic Acids 2002, 21, 45-54.
(21) Manoharan, M. 2′-Carbohydrate modifications in antisense
oligonucleotides therapy: importance of conformation, configu-
ration and conjugation. Biochim. Biophys. Acta 1999, 1489, 117-
130.
Cell viability was determined at the end of the testing period
by staining with 3-(4,5-dimethylthiazol-2-yl)-5-(3-carboxymethox-
yphenyl)-2-(4-sulfophenyl)-2H-tetrazolium, inner salt (MTS).
Briefly, 20 µL of MTS was added to all wells and incubated
overnight. The sealed plates were inverted several times, and
dye absorbance was read at 490 nm with a plate reader.
Calculations from the absorbance readings provided the
percent cytopathic effect (CPE) reduction, percent cell viability,
IC50, and therapeutic concentration 50% (TC50) values.
An ti-RT Assa y. Sou th er n Resea r ch Cen ter . Reverse
transcriptase activity was determined using generally the
process described earlier by Boyer et al.36
(22) Eckstein, F. Phosphorothioate oligodeoxynucleotides: What is
their origin and what is unique about them? Antisense Nucleic
Acid Drug Dev. 2000, 10, 117-121.
(23) Unpublished data.
(24) 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.
(25) Hoard, D. E.; Ott, D. G. Conversion of mono- and oligodeoxyri-
bonucleotides to 5′-triphosphates. J . Am. Chem. Soc. 1965, 87,
1785-1788.
(26) Yano, J .; Kan, L. S.; Ts’O, P. O. P. A simple method of the
preparation of 2′-O-methyladenosine. Biochim. Biophys. Acta
1980, 629, 178-183.
(27) Zambito, A. J .; Hendrzak, E.; Babson, R. D. 4(or 5)-(3,3-Dimethyl-
1-triazeno)imidazole-5(or 4)carboxamide: an improved procedure
adaptable to large-scale production. In Nucleic Acid Chemistry
Part One, Improved and New Synthetic Procedures, Methods,
and Techniques; Townsend, L. B., Tipson, R. S., Eds.; J ohn Wiley
& Sons: New York, 1978; pp 3-15.
(28) 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.
(29) Iyer, R. P.; Egan, W.; Regan, J . B.; Beaucage, S. L. 3H-1,2-
Benzodithiole-3-one 1,1-dioxide as an improved sulfurizing
reagent in the solid-phase synthesis of oligodeoxyribonucleoside
phosphorothioates. J . Am. Chem. Soc. 1990, 112, 1253-1254.
(30) The propargyl and allyl groups are unstable to the standard
overnight treatment with ammonium hydroxide at 55 °C.
(31) Data not shown.
An ti-HCMV Activity. In stitu te for An tivir a l Resea r ch ,
Uta h Sta te Un iver sity. Anti-HCMV assays were performed
as generally described earlier.37,38,39
Refer en ces
(1) Preston, B. D.; Poiesz, B. J .; Loeb, L. A. Fidelity of HIV-1 reverse
transcriptase. Science 1998, 242, 1168-1171.
(2) Roberts, J . D.; Bebenek, K.; Kunkel, T. A. The accuracy of reverse
transcriptase from HIV-1. Science 1988, 242, 1171-1173.
(3) Kosalaraksa, P.; Kavlick, M. F.; Maroun, V.; Le, R.; Mitsuya,
H. Comparative fitness of multi-dideoxynucleoside-resistant
human immunodeficiency virus type 1 (HIV-1) in an in vitro
competitive HIV-1 replication assay. J . Virol. 1999, 73, 5356-
5363.
(4) D’Aquila, R. T.; Schapiro, J . M.; Brun-Ve´zinet, F.; Clotet, B.;
Conway, B.; Demeter, L. M.; Grant, R. M.; J ohnson, V. A.;
Kurtizkes, D. R.; Loveday, C.; Shafer, R. W.; Richman, D. D.
Drug resistance mutations in HIV-1. Top. HIV Med. 2002, 10,
11-15.
(32) Data not shown

Oligo- and Polyribonucleotides
J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 10 1885
(33) Buckheit, R. W., J r.; Lackman-Smith, C.; Snow, M. J .; J alliday,
S. M.; White, E. L.; Ross, L. J .; Agrawal, V. K.; Broom, A. D.
PMTI, a broadly active unusual single-stranded polyribonucle-
otide, inhibits human immunodeficiency virus replication by
multiple mechanisms. Antiviral Chem. Chemother. 1999, 10, 23-
32.
(34) Yano, J .; Kan, L. S.; Ts’O, P. O. P. A simple method of the
preparation of 2′-O-methyladenosine. Biochim. Biophys. Acta
1980, 629, 178-183.
of human immunodeficiency virus type 1 reverse transcriptase.
J . Virol. 1993, 67, 2412-2420.
(37) Sidwell, R. W.; Huffman, J . H.; Barnard, D. L.; Reist, E. J . Effect
of phosphonic acid analogs of acyclovir and ganciclovir on in vitro
cytomegalovirus infections. Nucleotides Nucleosides 1989, 8,
833-836.
(38) Barnard, D. L.; Huffman, J . H.; Sidwell, R. W.; Reist, E. J .
Selective inhibition of cytomegaloviruses by 9-(3′-ethylphosphono-
1-hydrozymethyl-1-propyloxymethyl)guanine. Antiviral Res. 1993,
22, 77-89.
(39) Huffman, J . H.; Sidwell, R. W.; Morrison, A. G.; Coombs, J .;
Reist, E. J . Structure-activity-relationship of phosphonic acid
analogs of acyclovir or ganciclovir against human cytomegalo-
virus in MRC-5 Cells. Nucleosides Nucleotides 1994, 13, 607-613.
(35) Zambito, A. J .; Hendrzak, E.; Babson, R. D. 4(or 5)-(3,3-Dimethyl-
1-triazeno)imidazole-5(or 4)carboxamide: an improved procedure
adaptable to large-scale production. In Nucleic Acid Chemistry
Part One, Improved and New Synthetic Procedures, Methods,
and Techniques; Townsend, L. B., Tipson, R. S., Eds.; J ohn Wiley
& Sons: New York, 1978; pp 3-15.
(36) Boyer, P. L.; Currens, M. J .; McMahon, J . B.; Boyd, M. R.;
Hughes, S. H. Analysis of nonnucleoside drug-resistant variants
J M0203276