Structure-based design, synthesis, and in vitro assay of novel nucleoside analog inhibitors against HIV-1 reverse transcriptase

Article abstract of DOI:10.1016/j.bmcl.2005.05.099

Crystal structure of HIV-RT in complex with a DNA template:primer and a dTTP leads us to design and synthesize a new class of nucleoside analog inhibitors containing a branched 3′-group against HIV-RT. An in vitro primer extension assay indicates that thr

Full text of DOI:10.1016/j.bmcl.2005.05.099

Bioorganic & Medicinal Chemistry Letters 15 (2005) 3775–3777
Structure-based design, synthesis, and in vitro assay
of novel nucleoside analog inhibitors against HIV-1
reverse transcriptase
Xianjun Liu, Wei Xie and Raven H. Huang*
Department of Biochemistry, School of Molecular and Cellular Biology, University of Illinois at Urbana-Champaign,
600 South Mathews Avenue, Urbana, IL 61801, USA
Received 7 April 2005; revised 11 May 2005; accepted 16 May 2005
Available online 1 July 2005
Abstract—Crystal structure of HIV-RT in complex with a DNA template:primer and a dTTP leads us to design and synthesize a
new class of nucleoside analog inhibitors containing a branched 30-group against HIV-RT. An in vitro primer extension assay indi-
cates that three out of five compounds are effective HIV-RT inhibitors.
ꢀ 2005 Elsevier Ltd. All rights reserved.
The human immunodeficiency virus (HIV) causes
acquired immunodeficiency syndrome (AIDS), which
is responsible for the death of millions of people
worldwide.1,2 Proliferation of HIV requires that its
RNA genome be converted into a double-stranded
DNA copy that is then integrated into the host gen-
ome. Reverse transcriptase (RT) is responsible for cat-
alyzing this conversion and therefore it is a primary
target for therapeutic agents, such as AZT.3,4 Howev-
er, if an RT inhibitor is administered alone, drug
resistance increases rapidly, compromising the effec-
tiveness of treatment.5,6 To circumvent this problem,
a combination therapy that usually includes three
inhibitors—two against RT and one against HIV pro-
tease—is employed, which substantially reduces the
death rate of AIDS patients.7 Despite this progress,
the battle against the AIDS epidemic is far from over,
as indicated by an ever-increasing population of peo-
ple infected by HIV and the continuing death caused
by the disease.8 Clearly, more and better drugs are
needed.
cates that the 30-hydroxyl group of the resident dTTP,
which is modified in the nucleoside analog drugs, pro-
jects into a pocket (Fig. 1). Surprisingly, this Ô30-pocketÕ
is large, certainly having enough space for a larger
group, such as the azido group of AZT. This may
We have reported a crystal structure of RT in complex
with a DNA template:primer and a dTTP.9 Inspection
of the structure surrounding the active site of RT indi-
Figure 1. Surface representation of
a structure surrounding the
polymerase active site of RT, with emphasis on the 30-pocket. The
protein part is shown in gray. dTTP is shown as a stick model and is
colored by individual atoms (oxygen, red; phosphate, yellow; nitrogen,
blue; carbon, white). Arrows indicate the top and bottom portions of
the 30-pocket.
Keywords: RT inhibitors; Nucleoside analogs; Structure-based design;
HIV-RT; Primer extension assay.
*
Corresponding author. Tel.: +1 217 333 3967; fax: +1 217 244
5858; e-mail: huang@uiuc.edu
0960-894X/$ - see front matter ꢀ 2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2005.05.099

3776
X. Liu et al. / Bioorg. Med. Chem. Lett. 15 (2005) 3775–3777
explain the effectiveness of AZT as an inhibitor of HIV-
RT but not human DNA polymerases, because human
DNA polymerases may not possess such a large 30-pock-
et. However, when we tried to model AZTTP, which is
an active form of AZT in vivo, into the position of
dTTP in the structure, we found that the azido group
could be placed either in the top portion or in the bot-
tom portion of the 30-pocket, which are oriented roughly
in the opposite direction (Fig. 1). We reasoned that a
nucleoside analog that contains a branched 30-group,
with a sp2 hybridized central atom to orient the two at-
tached groups in the opposite direction, might fit the 30-
pocket better, and therefore could be a more selective
RT inhibitor because it would accentuate further the
structural differences between the RT and human
DNA polymerases. This reasoning, together with the
consideration of limited size of the 30-pocket, led us to
design and synthesize compounds 2, 3, and 4 (Fig. 2).
In addition to these compounds, we included com-
pounds 5 and 6, which contain a 30-group that is one
methylene unit larger than that in 3 and 4, in our synthe-
sis for comparison. Our molecular modeling predicted
that the 30-pocket of HIV-RT should have enough room
to accommodate a small branched 30-group shown in 2,
3, or 4, but not the ones in 5 or 6 without any substantial
structural perturbation (data not shown).
To construct the desired nucleoside analogs with a
branched 30-group, the key intermediate aldehyde was
synthesized through a known protocol (1 in Fig. 2).10
All the new derivatives of 1 were synthesized according
to a general strategy illustrated in Figure 2. Oxidation of
aldehyde 1 to the corresponding carboxylic acid with io-
dine in aqueous alkaline solution,11 followed by detrity-
lation with trichloroacetic acid, resulted in 2. In an
attempt to convert the aldehyde into the amides 3 and
5, several methods were examined. The best result was
obtained by utilizing radical-mediated oxidation.12
Aldehyde 1 was first treated with NBS in the presence
of a catalytic amount of AIBN. The reaction mixture
was then cooled and treated with a slight excess of
ammonia or methylamine to give 3 and 5, respectively,
after detritylation. Ketones 4 and 6 were prepared first
by alkylation of the aldehyde with Grignard reagents
13
in the presence of CeCl3 and followed by oxidation.
Specifically, the reaction of MeMgBr or CH3CH2MgBr
in the presence of an equal-molar amount of CeCl3 with
1 in THF delivered the corresponding alcohols, which
were used directly in the next step reaction without fur-
ther purification. Dess–Martin oxidation in CH2Cl2 and
subsequent detritylation with trichloroacetic acid affor-
ded 4 and 6. In addition, compounds 2–6, together with
ddT and AZT, were converted into their corresponding
triphosphate derivatives using a published procedure to
carry out the in vitro assay (see supporting material for
experimental details).14
A primer extension assay, analogous to the Sanger
DNA sequencing,15 was employed to analyze the effica-
cy of these compounds as RT inhibitors.16 For this
purpose, a DNA template:primer was designed such
that there is only one A (colored in red in Fig. 3a) cen-
tered in the single-stranded region of the template. In
the presence of dNTP (mixture of four nucleotides,
Figure 2. Chemical synthesis of designed nucleoside analogs.
Reagents: (a) I2, KOH, CH3CN/H2O, CCl3COOH/H2O; (b) NBS,
AIBN, NH3, CCl3COOH/H2O; (c) MeMgBr, CeCl3, Dess–Martin
reagent, CCl3COOH/H2O; (d) NBS, AIBN, CH3NH2, CCl3COOH/
H2O; (e) CH3CH2MgBr, CeCl3, Dess–Martin reagent, CCl3COOH/
H2O.
Figure 3. An in vitro primer extension assay for the efficacy of compounds 2–6 as possible RT inhibitors. (a) Nucleotide sequence of the DNA
template:primer used in the assay. (b) Denaturing polyacrylamide gel electrophoresis analysis of reaction samples carried out under the conditions
indicated. designates the radiolabeled oligonucleotides. P, primer; SFP, synthetic full-length product; STP, synthetic terminated product.
*

X. Liu et al. / Bioorg. Med. Chem. Lett. 15 (2005) 3775–3777
Table 1. Quantitative data of primer extension assays
3777
Acknowledgments
Reaction time (min)
Inhibitors
This work was supported by the start-up fund from
University of Illinois at Urbana-Champaign and a grant
from American Cancer Society, Illinois Division. We
thank R. Switzer and D. Kranz for a critical reading
of the manuscript.
ddTTP AZTTP 2TP 3TP 4TP
5
20
60
37.1
37.7
35.4
14.4
15.0
14.5
15.0 15.1 15.7
14.7 14.9 14.7
14.6 15.2 16.3
The number is the percentile of the intensity of the terminated product
divided by that of the full-length product, averaged from three inde-
pendent experiments.
Supplementary data
Supplementary data associated with this article can be
found in the online version at doi:10.1016/
j.bmcl.2005.05.099.
dATPs, dCTP, dGTP, and dTTP) and HIV-RT, the
primer was fully extended (Fig. 3b, lane 4, and com-
pare it to lane 2). Addition of ddTTP or AZTTP to
the reaction mixtures resulted in chain termination, as
indicated by the appearance of a band with the same
mobility as the molecular marker (Fig. 3b, lanes 5
and 6, and compare them to lane 3). This is consistent
with the fact that both ddTTP and AZTTP are sub-
strates of RT and are chain terminators. Addition of
the triphosphate derivatives of compounds 2, 3, and
4 (named 2TP, 3TP, and 4TP, respectively) also result-
ed in the appearance of chain-terminated band of the
primer extension (Fig. 3b, lanes 7, 8, and 9). These re-
sults clearly indicate that 2TP, 3TP, and 4TP are sub-
strates of RT. As the amounts of AZTTP, 2TP, 3TP,
and 4TP used in these reactions were the same, the fact
that the ratio of the full-length product to the terminat-
ed product was roughly the same in these lanes (Table
1) indicates that 2TP, 3TP, and 4TP are as effective
substrates of RT as AZTTP in vitro. Interestingly, no
chain termination products were observed when 5TP
or 6TP was added to the reaction mixture (Fig. 3b,
lanes 10 and 11), indicating that they are not substrates
of RT. Considering that 5 and 6 contain a 30-group
that is only one methylene unit larger than that in 3
and 4, the selectivity of RT is remarkable.
References and notes
1. Barre-Sinoussi, F.; Chermann, J. C.; Rey, F.; Nugeyre, M.
T.; Chamaret, S.; Gruest, J.; Dauguet, C.; Axler-Blin, C.;
Vezinet-Brun, F.; Rouzioux, C.; Rozenbaum, W.; Mon-
tagnier, L. Science 1983, 220, 868.
2. Gallo, R. C.; Salahuddin, S. Z.; Popovic, M.; Shearer, G.
M.; Kaplan, M.; Haynes, B. F.; Palker, T. J.; Redfield, R.;
Oleske, J.; Safai, B., et al. Science 1984, 224, 500.
3. Mitsuya, H.; Yarchoan, R.; Broder, S. Science 1990, 249,
1533.
4. Havlir, D. V.; Richman, D. D. Ann. Intern. Med. 1996,
124, 984.
5. Oxford, J. S.; al-Jabri, A. A.; Stein, C. A.; Levantis, P.
Methods Enzymol. 1996, 275, 555.
6. Moyle, G. J. J. Antimicrob. Chemother. 1997, 40, 765.
7. Palella, F. J., Jr.; Delaney, K. M.; Moorman, A. C.;
Loveless, M. O.; Fuhrer, J.; Satten, G. A.; Aschman, D. J.;
Holmberg, S. D. N. Engl. J. Med. 1998, 338, 853.
8. UNAIDS AIDS Epidemic Update; UNAIDS: www.una-
ids.org, 2004.
9. Huang, H.; Chopra, R.; Verdine, G. L.; Harrison, S. C.
Science 1998, 282, 1669.
10. Swyze, E. E.; Sanghvi, Y. S. Syn. Lett. 1997, 7, 859.
11. Yamada, S.; Morizono, D.; Yamamoto, K. Tetrahedron
Lett. 1992, 33, 4329.
In conclusion, we have designed and synthesized five
nucleoside analogs containing a branched 30-group. An
in vitro primer extension assay indicates that the three
analogs with smaller 30-group are effective RT inhibi-
tors. We are currently in the process of arranging for
them to be tested in an HIV-infected cell line for their
anti-HIV activity in vivo. The study described here pro-
vides a blueprint for the development of new classes of
nucleoside analog inhibitors. As in the case of other
RT inhibitors, it is likely that drug resistance will arise
if one of the inhibitors described here is administered
alone. However, the risk of drug resistance can be re-
duced with the adoption of the combination therapy de-
scribed previously. More importantly, because of the
change of chemical structures at the 30-position in our
inhibitors as compared to the current drugs, the resis-
tance profile, if it were to develop, could very well be dif-
ferent because the profile of drug resistance has been
shown to be correlated to the chemical structure of the
drug.9 Therefore, this new class of inhibitors may help
circumvent the current drug resistance profiles and,
together with currently available HIV-RT drugs, may
12. Marko, I. E.; Mekhalfia, A. Tetrahedron Lett. 1990, 31,
7237.
13. Imamoto, T.; Takiyama, N.; Nakamura, K.; Hatajima, T.;
Kamiya, Y. J. Am. Chem. Soc. 1989, 111, 4392.
14. Switzer, C. Y.; Moroney, S. E.; Benner, S. A. Biochemistry
1993, 32, 10489.
15. Church, G. M.; Kieffer-Higgins, S. Science 1988, 240, 185.
16. HIV-1 RT was overexpressed and purified according to a
published procedure.9 The DNA primer was radiolabeled
at the 50-end by T4 polynucleotide kinase and [c-32P]ATP.
The radiolabeled DNA primer was annealed to the DNA
template using a standard procedure. For a typical primer
extension reaction, 15 lL of DNA duplex containing
45 nM DNA primer and 30 nM of DNA template was
mixed with 6 lL of dNTP (containing 100 lM of each
dGTP, dATP, dCTP, and dTTP), 6 lL of 200 lM XTP
(X = ddT, AZT, 2, 3, 4, 5, or 6), 6 lL of RT (0.3 mg/mL),
and 3 lL of 0.1 mM DTT in a 1· DNA sequenase buffer.
The extension reaction was allowed to proceed at 25 ꢁC.
Twelve microliters of reaction mixture was taken out at
the reaction time points of 5, 20, and 60 min, and the
reaction was terminated by addition of 5 lL of DPAGE
loading buffer and heating of the sample at 90 ꢁC for
5 min. Five microliters of the mixture was analyzed by
denaturing PAGE.
allow for
therapies.
a broader spectrum of combinatorial