Synthesis of a neamine dimer targeting the dimerization initiation site of HIV-1 RNA

Article abstract of DOI:10.1021/ol701760k

(Chemical Equation Presented) A neamine dimer designed to bind to a specific sequence of HIV-1 RNA has been synthesized. Starting from neomycin B (1), a five-step synthesis efficiently provided a key protected neamine monomer 6 (28%). From the latter, cou

Full text of DOI:10.1021/ol701760k

ORGANIC
LETTERS
2007
Vol. 9, No. 22
4415-4418
Synthesis of a Neamine Dimer Targeting
the Dimerization Initiation Site of HIV-1
RNA
Anne Bodlenner,† Aure´lien Alix,† Jean-Marc Weibel,† Patrick Pale,*,† Eric Ennifar,‡
Jean-Christophe Paillart,‡ Philippe Walter,‡ Roland Marquet,‡ and
Philippe Dumas‡
Laboratoire de synthe`se et re´actiVite´ organique, Institut de Chimie, UniVersite´ Louis
Pasteur, 4 rue Blaise Pascal, 67070 Strasbourg, France, and Architecture et re´actiVite´
de l’ARN, UniVersite´ Louis Pasteur, CNRS, IBMC, 15, rue Rene´ Descartes,
67084 Strasbourg, France
ppale@chimie.u-strasbg.fr
Received July 24, 2007
ABSTRACT
A neamine dimer designed to bind to a specific sequence of HIV-1 RNA has been synthesized. Starting from neomycin B (1), a five-step
synthesis efficiently provided a key protected neamine monomer 6 (28%). From the latter, coupling reactions with activated diacids gave
dimers. After deprotection, a neamine dimer was obtained as the hexachlorohydrate salt 15 with 13% overall yield over nine steps.
Infectious viral particles of human immunodeficiency virus
(HIV) contain two positive RNA strands linked together near
their 5′ end.1 This organization is initiated by interactions
between a conserved six nucleotide self-complementary
sequence located in the dimerization initiation site (DIS), a
stem-loop promoting formation of a loop-loop “kissing
complex” (Figure 1a).2 Since the DIS is essential for efficient
viral replication at the RNA packaging and reverse transcrip-
tion stages,3 the dimerization step represents a new target
for the development of future anti-HIV drugs.
† Institut de Chimie, Universite´ Louis Pasteur.
‡ Architecture et re´activite´ de l’ARN, Universite´ Louis Pasteur.
(1) Paillart, J.-C.; Shehu-Xhilaga, M.; Marquet, R.; Mak, J. Nat. ReV.
Microbiol. 2004, 2, 461-472.
(2) (a) Skripkin, E.; Paillart, J.-C.; Marquet, R.; Ehresmann, B.; Ehres-
mann, C. Proc. Natl. Acad. Sci. U.S.A. 1994, 91, 4945-4949. (b) Laughrea,
M.; Jette, L. Biochemistry 1994, 33, 13464-13474.
(3) (a) Paillart, J.-C.; Berthoux, L.; Ottmann, M.; Darlix, J.-L.; Marquet,
R.; Ehresmann, B.; Ehresmann, C. J. Virol. 1996, 12, 8348-8354. (b)
Berkhout, B.; Van Wamel, J. L. B. J. Virol. 1996, 70, 6723-6732. (c)
Clever, J. L.; Parslow, T. G. J. Virol. 1997, 71, 3407-3414. (d) Laughrea,
M. J. Virol. 1997, 71, 3397-3406.
Figure 1. (a) Stereoview of the X-ray structure of the “kissing
complex” of HIV-1, made of two HIV RNA dimerization initiation
sites (DIS). (b) Concept of interfering with the RNA dimerization
step (N: neamine; the two DIS are in green and red).
10.1021/ol701760k CCC: $37.00
© 2007 American Chemical Society
Published on Web 10/04/2007

Scheme 1. Synthesis of a Key Monomer Derived from Neamine 6
The DIS “kissing complex” crystal structure was solved,
of an N1 amide-linked neamine dimer which is capable of
interacting with HIV-1 RNA. To maintain and even extend
the interactions due to the N1 amino group, amide linkages
4-6 Å long were chosen to connect two neamine units at
amine N1.
revealing a coaxial interaction of two stem-loops, two purines
bulging out and stacking on one another, leaving a cavity in
each stem-loop (Figure 1a).4 This feature led us to imagine
blocking the DIS dimer with dimeric molecules able to
simultaneously bind to each RNA strands. Over-stabilization
of the DIS “kissing complex” could interfere with packaging
and reverse transcription of the genomic RNA, thus interfer-
ing with viral replication (Figure 1b). Drug design as well
as biological and crystallographic studies5,6 revealed that
neamine tethered at N1 by a 4-6 Å chain would fit into the
DIS “kissing complex” and thus interfere with it (Figure 2).
Surprisingly, only a few dimeric aminoglycosides have
been described, but none of them are linked via N1.7
Moreover, approaches aiming at distinguishing the N1 amine
from the other neamine functions are scarce.8 As they proved
unsatisfactory in our hands, we relied on selective formation9
of cyclic carbamates and on their selective opening9 expect-
ing that a strained carbamate opens up more readily than an
unstrained one (Scheme 1).
The synthesis of the key protected neamine monomer 6
started with methanolysis of commercially available neo-
mycin B trisulfate 1 to give neamine 2 as the tetrahydro-
chloride salt.10 The four amine groups of neamine were easily
protected as benzyloxycarbamates using slightly modified
conditions,11 triethylamine being used here as base for
purification reasons. A basic treatment of 3 with NaH under
carefully controlled conditions allowed us to convert two of
the four benzyloxycarbamates groups into five- and six-
membered cyclic carbamates, producing compound 4 in good
yield. The position of these carbamates were clearly estab-
lished by HMBC correlations. It is worth noting that this
Figure 2. Stereoview of the X-ray structure of the neamine-DIS
complex. The two DIS RNA strands are represented in green and
blue, and the yellow sphere is a potassium cation. As expected,
neamine occupies the RNA stem-loop cavity of the “kissing
complex”, and interactions are mostly localized around the glu-
cosamine moiety, with amine N1 pointing away.
(7) (a) Michael, K.; Wang, H.; Tor, Y. Bioorg. Med. Chem. 1999, 7,
1361-1371. (b) Agnelli, F.; Sucheck, S. J.; Marby, K. A.; Rabuka, D.;
Yao, S.-L.; Sears, P. S.; Liang, F.-S.; Wong, C.-H. Angew. Chem., Int. Ed.
2004, 43, 1562-1566. (c) Liu, X.; Thomas, J. R.; Hergenrother, P. J. J.
Am. Chem. Soc. 2004, 126, 9196-9197.
(8) (a) Umezawa, S.; Ikeda, D.; Tsuchiya, T. J. Antibiot. 1973, 26, 304-
306. (b) Haddad, J.; Kotra, L. P.; Liano-Sotelo, B.; Kim, C.; Azucena, E.
F.; Meizheng, J. L.; Vakulenko, S. B.; Chow, C. S.; Mobashery, S. J. Am.
Chem. Soc. 2002, 124, 3229-3237. (c) Li, J.; Chen, H.-N.; Chang, H.;
Wang, J.; Chang, C. W. T. Org. Lett. 2005, 7, 905-910.
(9) For a review on the chemistry of carbamates, see: Agami, C.; Couty,
F. Tetrahedron 2002, 58, 2701-2724.
(10) (a) Byron, E.; Leach, E.; Teeters, C. M. J. Am. Chem. Soc. 1951,
73, 2794-2797. (b) Dutcher, J.; Donin, M. J. Am. Chem. Soc. 1952, 74,
3420-3422. (c) Grapsas, I.; Cho, Y. J.; Mobashery, S. J. Am. Chem. Soc.
1994, 59, 1918-1922.
(11) Park, W. K. C.; Auer, M.; Jaksche, H.; Wong, C.-H. J. Am. Chem.
Soc. 1996, 118, 10150-10155.
On this basis, we designed, synthesized, and studied dimeric
neamine derivatives, and herein, we report the first synthesis
(4) (a) Ennifar, E.; Dumas, P. J. Mol. Biol. 2006, 356, 771-782. (b)
Ennifar, E.; Walter, P.; Ehresmann, B.; Ehresmann, C.; Dumas, P. Nat.
Struct. Biol. 2001, 8, 1064-1068.
(5) Ennifar, E.; Paillart, J.-C.; Marquet, R.; Ehresmann, B.; Ehresmann,
C.; Dumas, P.; Walter, P. J. Biol. Chem. 2003, 278, 2723-2730.
(6) Ennifar, E.; Paillart, J.-C.; Bodlenner, A.; Walter, P.; Weibel, J.-M.;
Aubertin, A.-M.; Pale, P.; Dumas, P.; Marquet, R. Nucleic Acids Res. 2006,
34, 2328-2339.
4416
Org. Lett., Vol. 9, No. 22, 2007

Compound 4, having two amines (N1, N6′) distinguished
from the two others (N3, N2′) as cyclic carbamates, was
protected with a TBS group for solubility reasons. As
expected from the known reactivity of the 5-hydroxyl
group,11 only the silylation at the 3′ position was observed
in the conditions used (see the Supporting Information). The
trans-fused five-membered carbamate of the resulting 5 was
selectively opened by treatment with barium hydroxide,12
leading to 6 protected except at amine N1 and alcohols O5
and O6. The position of the silylation as well as the
regioselectivity of the carbamate opening were again estab-
lished by HMBC correlations. The remaining free hydroxyl
groups could also be selectively protected, but the higher
nucleophilicity of the amino group led us to directly attempt
to introduce the required linker through amide bond forma-
tion.
Table 1. Dimerization of the Key Monomer
Diacids, three to four carbons long, were used for the
dimerization step, either directly in peptide coupling or as
N-hydroxysuccinimide esters13 or pivaloyl anhydrides,14
known to be selective for amines in the presence of alcohols
(Table 1). Dimerization with malonic acid 7a or its bis-
(pivaloyl) anhydride 7b or bis(succinimidyl)diester 7c did
not yield the expected product. Decomposition often oc-
curred, except with 7b, which gave the N-pivaloyl product
10 (entry 1). In contrast, reactions between 6 and succinic
acid 8a or its activated diester 8c selectively provide the
expected dimer 11 in good yields (entries 2 and 4). The
anhydride 8b also gave the N-pivaloyl product 10 (entry 3).
Fumaric acid 9a was also investigated, since its E double
bond induces a shorter chain length and should avoid intra-
molecular side reactions. Peptide coupling and the use of
the anhydride 9b gave almost the same good results (entries
5 and 6). Rewardingly, the diester 9c provided the dimer 12
with an excellent yield (entry 7).
a Yield of chromatographically pure products; condn, conditions; piv,
pivaloyl; Su, N-succinimidyl; DIPEA, N,N-diisopropylethylamine; HOBt,
1-hydroxybenzotriazole; DIPC, diisopropylcarbodiimide.
sequence provided 4 in excellent overall yield and without
chromatography, improving other routes to this compound.12
Scheme 2. Synthesis of Neamine Dimer 15
Org. Lett., Vol. 9, No. 22, 2007
4417

Deprotection of the carbobenzyloxy groups (Cbz) in 11
or 12 could not be achieved under conventional hydro-
genolysis conditions. Extensive investigations revealed that
treatment with sodium in liquid ammonia led to deprotection
of the four Cbz groups and also to hydrogenation of the
double bond in 12 (Scheme 2). Desilylation also proved
cumbersome and only acidic treatment with hydrogen
chloride in methanol provided an efficient and clean reaction,
cleaving both silyl groups in quantitative yield. The two
remaining 6-membered cyclic carbamates were removed by
treatment with barium hydroxide in carefully controlled
conditions, leading to the hexamine, which was then treated
with 1 M aqueous hydrochloric acid in order to recover good
water solubility for biological tests and reliable NMR
analysis. The corresponding hexahydrochloride salt 15 was
obtained with an overall yield of 13% over nine steps from
neomycin B.
Biochemical studies relying on a selective Pb2+ induced
cleavage15 of the DIS loop showed that 15 binds to DIS, as
anticipated. It indeed protected DIS from Pb2+ cleavage
(Figure 3a). Dimer 15 also stabilizes the kissing complex at
37 °C in the context of 615 nucleotide RNA fragments
(Figure 3b). Competition experiments between long and short
RNA DIS fragments (615 and 311 nucleotides, respectively),
performed in the presence or absence of either neamine or
15, revealed that 15 is more efficient in preventing het-
erodimer formation (RNA615/RNA311) than neamine, in-
dicating homodimer overstabilization, as expected.
Figure 3. (a) Denaturating gel showing inhibition of Pb2+-induced
DIS cleavage by dimer 15. Control experiments without Pb2+
or without dimer are indicated as -Pb and 0 respectively.
(b) Stabilization of the kissing complex by 15 and neamine.
Radiolabeled RNA 1-615 was incubated in dimerization buffer,
and a 5-fold excess of unlabeled RNA1-311 was added to the
solution in absence (“negative” curve) or in presence of dimer 15
or neamine.
Acknowledgment. We thank the CNRS and the French
Ministry of Research as well as the ANRS for financial
support.
Supporting Information Available: General procedure
for the preparation of unknown compounds, 1H and 13C NMR
spectra of all compounds, and procedure for competition
experiments are provided. This material is available free of
charge via the Internet at http://pubs.acs.org.
In conclusion, we developed an efficient route toward a
new class of dimeric compounds, designed to interfere with
the HIV DIS function, and showed that a representative dimer
interacts with DIS. Work is in progress to evaluate the
potential antiviral activity of this new family of compounds.
OL701760K
(12) (a) Horton, D.; Weckerle, W. Carbohydr. Res. 1975, 44, 227-240.
(b) Grewe, R.; Herberg, H. W. Chem. Ber. 1959, 92, 1195-1205.
(13) (a) Kawaguchi, H.; Naito, T.; Nakagawa, S.; Fujisawa, K.-I. J.
Antibiot. 1972, 25, 695-708. (b) Katsarava, R. D.; Kharadze, D. P.;
Avalishvili, L. M. Makromol. Chem. 1986, 187, 2053-2062. (c) Tanaka,
H.; Nishida, Y.; Furuta, Y.; Kobayashi, K. Bioorg. Med. Chem. Lett. 2002,
12, 1723-1726.
(14) (a) Kocian, O.; Stransky, K.; Zavada, J. Collect. Czech. Chem.
Commun. 1982, 47, 1356-1366. (b) Myers, A. G.; Gleason, J. L.; Yoon,
T.; Kung, D. W. J. Am. Chem. Soc. 1997, 119, 656-673.
(15) Paillart, J.-C.; Westhof, E.; Ehresmann, C.; Ehresmann, B.; Marquet,
R. J. Mol. Biol. 1997, 270, 36-49.
4418
Org. Lett., Vol. 9, No. 22, 2007