A modular approach to synthetic RNA binders of the hepatitis C virus internal ribosome entry site

Article abstract of DOI:10.1002/cbic.201000177

(Figure Presented) DAPper ligands: The 3,5-diaminopiperidine (DAP) heterocycle has been developed as a structural mimetic of the RNA-recognizing pharmacophore of the 2-DOS scaffold. Here we describe the synthesis of novel modular DAP ligands that bind to a conformational target in the internal ribosome entry site RNA of the hepatitis C virus.

Full text of DOI:10.1002/cbic.201000177

DOI: 10.1002/cbic.201000177
A Modular Approach to Synthetic RNA Binders of the Hepatitis C Virus
Internal Ribosome Entry Site
Maia Carnevali,[a] Jerod Parsons,[a] David L. Wyles,[b] and Thomas Hermann*[a]
Natural products that target the RNA components of bacterial
ribosomes and thereby act as antibiotics that shut down mi-
crobial protein synthesis, have provided a rich source of inspi-
ration for the design and synthesis of small-molecule ligands
directed at RNA targets.[1,2] A prominent example of a privi-
leged scaffold for RNA recognition occurring in natural amino-
glycoside antibiotics is 2-deoxystreptamine (2-DOS),[3] which
contains a rigid framework of hydrogen-bond donors among
which the rigid cis-1,3 arrangement of amino groups is respon-
sible for selective interaction with structural motifs in RNA tar-
gets (Figure 1).[4] In an approach to reduce the complexity of
chemical library synthesis that involves the highly functional-
ized 2-DOS scaffold, we have recently developed the 3,5-diami-
nopiperidine heterocycle (DAP) as a structural mimetic of the
RNA-recognizing pharmacophore of the 2-DOS scaffold
(Figure 1). Structure-guided design had been applied to dis-
cover a series of antibacterial DAP-triazine derivatives that act
on the same ribosomal RNA target as the natural aminoglyco-
side antibiotics, which initially served as the inspiration for the
conception of the DAP compounds.[5,6]
Here, we describe the synthesis of a novel class of modular
ligands (1) that contain the DAP scaffold as the key moiety for
RNA recognition in nonribosomal targets (Figure 1).[7] Screen-
ing of modular DAP ligands against the subdomain IIa, an RNA
target in the internal ribosome entry site (IRES) of hepatitis C
Figure 1. A) Structures of RNA-binding scaffolds 2-deoxystreptamine (2-DOS)
and 3,5-diaminopiperidine (DAP) as well as modular RNA binders, 1[7] and 2
virus (HCV) (Figure 1), revealed a set of N-amido substituted a-
that contain the DAP moiety. B) Secondary structure of the internal ribosome
amino acid conjugates of DAP (2) as micromolar binders of
this RNA. We had previously shown that ligand-induced con-
entry site (IRES) of the hepatitis C virus (HCV).[19] The position of the subdo-
main IIa target is indicated by a box. C) Secondary structure of the IRES sub-
domain IIa. D) Three-dimensional structure of the subdomain IIa RNA as
determined by X-ray crystallography.[13] Mg2+ ions are shown as spheres. For
a ligand binding assay used here, the A54 residue has been replaced by a
fluorescent 2-aminopurine.
formational change in the subdomain IIa RNA disrupts the
function of the IRES and blocks viral protein synthesis; this ulti-
mately leads to inhibition of HCV in infected cells.[8]
The common building block 4 used in the synthesis of the
modular DAP compounds 2 was obtained from 2-chloro-3,5-di-
nitro-pyridine (3) following an established route,[9] which re-
quired, in the last step, high-pressure hydrogenation to reduce
the pyridine (Scheme 1). As the yield in this transformation crit-
ically depends on the hydrogen pressure (>1000 psi), we
sought to employ an alternative reduction procedure. Among
the explored methods, neither transformation of the pyridine
Scheme 1. Reagents and conditions: a) Pd/C, H2, MeOH, RT, 48 h, quant.;
b) (Boc)2O (4 equiv), Et3N (2 equiv), MeOH, 08C!RT, 12 h, 70%; c) AcOH
[a] M. Carnevali, J. Parsons, Dr. T. Hermann
(1 equiv), Rh/C (wet, 5 mol%), H2 (1600 psi), MeOH, 1108C, 24 h, 54%. Boc=
Department of Chemistry and Biochemistry, University of California San
Diego
tert-butoxycarbonyl.
9500 Gilman Drive, La Jolla, CA 92093 (USA)
Fax: (+1)858-534-0202
E-mail: tch@ucsd.edu
to an N-oxide followed by ammonium formate treatment[10]
nor reduction by lithium triethylborohydride[11] afforded the
[b] Dr. D. L. Wyles
Division of Infectious Diseases, Department of Medicine
desired piperidine, which led us to pursue the original route to
University of California San Diego
9500 Gilman Drive, La Jolla, CA 92093 (USA)
the building block 4. Synthesis of the DAP compounds 2 com-
menced with coupling of 4 with N-Cbz and side-chain protect-
ed a-l-amino acids 5 (Scheme 2). Commercially available pro-
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/cbic.201000177.
1364
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemBioChem 2010, 11, 1364 – 1367

tween the stem termini, which follows from an in-
crease in the interhelical angle in IIa. We have recent-
ly used this assay to demonstrate that a benzimid-
azole inhibitor of HCV IRES-driven translation acts by
such conformational induction at the IIa target.[8]
Titrations of 2AP54 labeled IIa RNA with modular
DAP compounds resulted in fluorescence quenching
at low micromolar concentrations for three ligands
(2ab, 2bb, and 2cb), all of which contained a 4-ace-
tylbenzoic acid moiety (Table 1). None of the other
tested compounds showed an appreciable change of
2AP fluorescence. The dose-dependent decrease of
2AP fluorescence observed for the 4-acetylbenzoic
acid DAP derivatives (Figure 2) suggested that these
compounds bind to the IIa RNA at the internal loop
region, where 2AP54 is located, and thereby shield
the fluorescent label from solvent exposure. The fact
that a residual signal was observed even at the high-
est concentrations tested suggested that the 2AP la-
beled target remained in solution and that the fluo-
rescence decrease was not due to nonspecific RNA
aggregation. Binding was in competition with salt;
this indicates electrostatic contributions to the
ligand-RNA interaction (Figure 2). All titrations were
performed in a background of 100 mmMg2+ to
ensure that the IIa RNA was properly folded. Our pre-
vious studies on the three-dimensional structure and
metal binding of the IIa target revealed two tightly
bound Mg2+ ions in close proximity of 2AP54; these
ions are essential for the stabilization of the l-shaped
Scheme 2. Reagents and conditions: a) Et3N (7 equiv), HATU (1.1 equiv), HOAT (1.2 equiv),
CH2Cl2, 08C!RT, 5 h, 96%; b) Pd/C, H2, MeOH, RT, 24 h; c) 7 (1 equiv), Et3N (7 equiv),
HATU (1.1 equiv), HOAT (1.2 equiv), CH2Cl2, 08C!RT, 3.5 h, 70% (2 steps); d) deprotec-
tion; for 5a=N-Cbz-l-Lys(Boc) and 5b=N-Cbz-l-Arg(Boc)2: HCl/dioxane (4m), MeOH (1:2
v/v), 08C!RT, 5 h, 50% after RP HPLC; for 5c=N-Cbz-l-Ser(TBDMS) and 5d=N-Cbz-l-
Thr(TBDMS): TBAF (3 equiv), THF, 08C!RT, 4 h; then Boc deprotection as above.
Cbz=carboxybenzyl; HATU=2-(7-Aza-1H-benzotriazole-1-yl)-1,1,3,3-tetramethyluronium
hexafluorophosphate; HOAT=1-hydroxy-7-aza-benzotriazole; TBAF=tetra-n-butylammo-
nium fluoride; TBDMS=tert-butyldimethylsilyl; THF=tetrahydrofuran. Spectroscopic
characterization of the final products 2 is provided in the Supporting Information.
tected amino acids were used except for fully Boc-protected l-
Table 1. Structure–activity relationships for modular DAP compounds.
Compound Component 2AP assay[a] EC50 [mM] [mm] FRET assay[b] IFRET
arginine, which was synthesized by alkylation of N-Cbz-l-orni-
thine with N,N’-Boc2-(S)-methylisothiourea.[12] Selective depro-
tection of the coupling products gave DAP-amino acid conju-
gates 6, which were further treated with carboxylic acids 7 to
furnish the final products 2 after deprotection of the inter-
mediates 8 and HPLC purification.
2
5
7
100 mmMg2+ 1 mmMg2+ 100 mm 1 mm
2aa
2ab
2ac
2ba
2bb
2cb
2cc
2dc
5a
5a
5a
5b
5b
5c
5c
5d
7a
n.a.
6.3ꢀ0.6
n.a.
n.a.
7.2ꢀ1.2
n.d.
270ꢀ100
n.d.
n.d.
92ꢀ30
n.a.
n.a.
n.a.
n.a.
n.a.
n.a.
n.a.
n.a.
n.a.
n.a.
1.3
n.a.
n.a.
1.1
n.a.
n.a.
n.a.
7b
7c
7a
7b
RNA binding of the modular DAP compounds 2 was tested
in two different fluorescence assays that we had previously de-
veloped for the subdomain IIa target in the HCV IRES. In the
first assay, an oligonucleotide construct was used in which an
adenine residue at position 54 in the internal loop of IIa
(Figure 1) is replaced by the fluorescent base analogue 2-ami-
nopurine (2AP) as a sensitive probe of the RNA folding state.[13]
An increase of the fluorescence intensity upon addition of a
ligand to IIa-2AP54 indicates unfolding, or destabilization, of
the internal loop, while a signal decrease suggests further
compaction of the RNA fold or, more likely, additional shielding
of the 2AP label by a bound ligand. The second assay, which is
based on fluorescence resonance energy transfer (FRET) be-
tween cyanine dyes attached at the termini of a IIa oligonu-
cleotide construct,[8] was used to measure the impact of ligand
binding on the interhelical angle between the base paired
stems as a probe of the RNA conformation at the internal
loop. A decrease in FRET intensity signals a larger distance be-
7b 68ꢀ8
7c
7c
n.a.
n.a.
n.d.
n.d.
[a] 2AP EC50, concentration required for 50% of the 2AP fluorescence de-
crease triggered by compound interaction with 2AP54 labeled IIa RNA, in
the presence of 100 mm or 1 mmMg2+, in an assay that was described
previously.[13] [b] FRET IFRET, relative change of the FRET intensity of Cy3/
Cy5 5’-labeled IIa RNA in the presence 100 or 1000 mm compound, mea-
sured in an assay that was described previously.[8] Compounds were
tested for optical interference with the fluorescent labels for both the
2AP and FRET assays and no significant interference was found. n.a.=no
activity; n.d.=not determined.
RNA fold (Figure 1). For the DAP ligands that showed binding
to the IIa target, the affinity was substantially reduced in the
presence of 1 mmMg2+ (Table 1); this suggests direct competi-
tion with one, or both, of the structural Mg2+ ions. Frame-
ChemBioChem 2010, 11, 1364 – 1367
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chembiochem.org
1365

Figure 3. Model for the binding of small-molecule ligands to the HCV IRES
IIa RNA. The subdomain IIa RNA adopts a magnesium-stabilized bent fold
that plays a key role for the correct positioning of the IRES on the ribosome.
Flexibility of the l-shaped IIa structure might be required for release of the
ribosome after translation initiation (center). Benzimidazole inhibitors of
IRES-driven translation (left) induce a widened angle in the IIa target, which
leads to disruption of IRES function and inhibition of viral protein synthesis
in infected cells.[8] Modular DAP compounds (right) bind in competition with
structural Mg2+ ions and immobilize the l-shaped RNA, which might provide
another mechanism for inhibition of IRES function. See text for discussion.
Figure 2. Titrations of 2AP54 labeled IIa RNA with DAP compound 2ab in
the absence and presence of salt. The assay was performed as previously de-
scribed.[13] For the titration in 100 mm NaCl, 100 mmMg2+ was present as
well. Error bars represent ꢀ1s of triplicate titrations. EC50 values for all
tested DAP compounds are listed in Table 1. IF =relative fluorescence intensi-
ty.
works of amino groups, which are positively charged under
physiological conditions, play an important role for target rec-
ognition of aminoglycoside antibiotics that interact with RNA
often by exploiting structural electrostatic complementarity
with divalent metal-ion binding sites.[14] Similarly, binding data
derived from the 2AP54 assay suggest that the DAP scaffold in
the modular ligands described here participates in interactions
with Mg2+ sites at the IIa target.
mRNA binding cleft.[16] Correct positioning of the IIb hairpin
and the viral mRNA at the ribosome depends critically on the
L-shaped fold of the domain II.[18] Intrinsic flexibility in the sub-
domain IIa might be required for release of the ribosome from
the IRES during translation initiation. We have previously dem-
onstrated that IRES-binding benzimidazole inhibitors of the
HCV replicon act by conformational induction of a widened in-
terhelical angle in the subdomain IIa (Figure 3), which facili-
tates the undocking of the hairpin IIb from the ribosome and
ultimately leads to inhibition of IRES-driven translation in HCV-
infected cells.[8] In a distinct mode of action, the IIa RNA-bind-
ing modular DAP ligands described here might represent a
second class of novel IRES inhibitors that affect HCV translation
by arresting subdomain IIa in a 908 bent state and thereby in-
terfering with translation initiation (Figure 3).
The IIa-binding DAP compounds 2ab, 2bb and 2cb were in-
active in the FRET assay at micromolar concentrations in which
binding occurred in the 2AP assay (Table 1); this indicates that
these ligands interact with the RNA target without inducing a
conformational change. For two of the ligands, 2ab and 2bb,
both of which contain a basic amino acid (Lys or Arg), a small
increase of the FRET signal was observed at millimolar concen-
trations. As optical interference of the DAP compounds with
the fluorescent cyanine dyes was ruled out, the increased FRET
intensity was likely due to immobilization of the IIa RNA target
in the L-shaped fold upon ligand binding. Comparison of struc-
tural data from an NMR study of the IRES domain II[15] and crys-
tal structure analysis[13] suggests that the bent subdomain IIa
shows some flexibility, attested by an interhelical angle that
averages at >908 in solution and that is fixed at 908 in the
crystal. The DAP ligands are able to bind and arrest the IIa RNA
in the 908 bent state in solution, thereby increasing the overall
intensity of the FRET signal. This immobilizing action of the
modular DAP ligands might provide a mechanistic basis for a
novel class of HCV IRES inhibitors (Figure 3).[7]
Preliminary testing of one of the IIa RNA-binding DAP com-
pounds in a cell based HCV replicon assay revealed that the
ligand 2ab induced dose-dependent reduction in HCV IRES-
driven translation of a luciferase reporter (Figure 4). The inhibi-
tion of reporter expression in infected human cells was specific
for the replicon construct and not due to cytotoxicity. Future
work will be focused on the iterative optimization of the mod-
ular DAP compounds to improve RNA target binding and HCV
replicon inhibition.
Cryo-electron microscopy investigations of the HCV IRES
bound to the 40S ribosomal subunit[16] and whole 80S ribo-
somes[17] revealed an overall extended shape of the RNA with
the domain II located at one end, and adopting a hook-like
structure that directs the highly conserved terminal hairpin IIb
toward the ribosomal E site in proximity of the active site.[16] In-
teraction of domain II with the ribosome induces a conforma-
tional change in the 40S head[17] that leads to closure of the
Acknowledgements
This work was supported in part by the National Institutes of
Health, Grant Nos. AI72012 (T.H.), CA132753 (T.H.) and AI069989
(D.L.W.). Support of the NMR facility by the National Science
Foundation is acknowledged (CRIF grant CHE-0741968). We
thank Prof. Yitzhak Tor for providing access to a semipreparative
HPLC instrument.
1366
www.chembiochem.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemBioChem 2010, 11, 1364 – 1367

[1] J. R. Thomas, P. J. Hergenrother, Chem. Rev. 2008, 108, 1171–1224.
[2] T. Hermann, Biopolymers 2003, 70, 4–18.
[3] G. F. Busscher, F. P. Rutjes, F. L. van Delft, Chem. Rev. 2005, 105, 775–791.
[4] T. Hermann, Curr. Opin. Struct. Biol. 2005, 15, 355–366.
[5] Y. Zhou, V. E. Gregor, Z. Sun, B. K. Ayida, G. C. Winters, D. Murphy, K. B.
Simonsen, D. Vourloumis, S. Fish, J. M. Froelich, D. Wall, T. Hermann, An-
timicrob. Agents Chemother. 2005, 49, 4942–4949.
[6] Y. Zhou, V. E. Gregor, B. K. Ayida, G. C. Winters, Z. Sun, D. Murphy, G.
Haley, D. Bailey, J. M. Froelich, S. Fish, S. E. Webber, T. Hermann, D. Wall,
Bioorg. Med. Chem. Lett. 2007, 17, 1206–1210.
[7] T. Hermann, M. Carnevali, WO 2009/099897 2009.
[8] J. Parsons, M. P. Castaldi, S. Dutta, S. M. Dibrov, D. L. Wyles, T. Hermann,
Nat. Chem. Biol. 2009, 5, 823–825.
[9] Y. Zhou, D. Vourloumis, V. E. Gregor, G. Winters, T. Hermann, B. Ayida, Z.
Sun, D. Murphy, K. B. Simonsen, WO 2005/028467 2005.
[10] B. Zacharie, N. Moreau, C. Dockendorff, J. Org. Chem. 2001, 66, 5264–
5265.
[11] B. E. Blough, F. I. Carroll, Tetrahedron Lett. 1993, 34, 7239–7242.
[12] J. Izdebski, T. Gers, D. Kunce, P. Markowski, J. Pept. Sci. 2005, 11, 60–64.
[13] S. M. Dibrov, H. Johnston-Cox, Y. H. Weng, T. Hermann, Angew. Chem.
2007, 119, 230–233; Angew. Chem. Int. Ed. 2007, 46, 226–229.
[14] T. Hermann, E. Westhof, J. Mol. Biol. 1998, 276, 903–912.
[15] P. J. Lukavsky, I. Kim, G. A. Otto, J. D. Puglisi, Nat. Struct. Biol. 2003, 10,
1033–1038.
[16] C. M. Spahn, J. S. Kieft, R. A. Grassucci, P. A. Penczek, K. Zhou, J. A.
Doudna, J. Frank, Science 2001, 291, 1959–1962.
[17] D. Boehringer, R. Thermann, A. Ostareck-Lederer, J. D. Lewis, H. Stark,
Structure 2005, 13, 1695–1706.
Figure 4. Biological activity of DAP compound 2ab in human Huh-7.5 cells.
A) Dose-dependent translation inhibition of luciferase reporter under HCV
IRES control in the BM4-5 FEO HCV replicon.[20] B) Cytotoxicity as determined
in a colorimetric viability assay. The luciferase reporter replicon assay and cy-
totoxicity testing were performed as described before.[8] IL =relative lumines-
cence intensity.
[18] P. J. Lukavsky, Virus Res. 2009, 139, 166–171.
[19] J. S. Kieft, K. Zhou, R. Jubin, M. G. Murray, J. Y. Lau, J. A. Doudna, J. Mol.
Biol. 1999, 292, 513–529.
[20] D. L. Wyles, K. A. Kaihara, F. Vaida, R. T. Schooley, J. Virol. 2007, 81,
3005–3008.
Keywords: antibiotics · antiviral agents · proteins · RNA
recognition · viruses
Received: March 17, 2010
Published online on June 16, 2010
ChemBioChem 2010, 11, 1364 – 1367
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chembiochem.org
1367