Inhibition of HIV-1 fusion by hydrogen-bond-surrogate-based α helices

Article abstract of DOI:10.1002/anie.200704227

Inhibition with a twist: An artificial α helix obtained by replacing an N-terminal main-chain i and i+4 hydrogen bond with a carbon-carbon bond (see structure) inhibits gp41-mediated cell fusion. This work suggests that hydrogen-bond-surrogate-derived helices may provide a general class of scaffolds for the generation of leads against viral entry. (Chemical Equation Presented).

Full text of DOI:10.1002/anie.200704227

Angewandte
Chemie
DOI: 10.1002/anie.200704227
a-Helix Mimetics
Inhibition of HIV-1 Fusion by Hydrogen-Bond-Surrogate-Based
a Helices**
Deyun Wang, Min Lu, and Paramjit S. Arora*
Entry of HIV-1 into its target cells to establish an infection is
mediated by viral envelope glycoprotein (Env) and cell-
surface receptors (CD4 and a coreceptor,such as CXCR4 or
CCR5).[1] The mature Env complex is a trimer,with three
gp120 glycoproteins associated noncovalently with three
membrane-anchored gp41 subunits. Binding of gp120/gp41
to cellular receptors triggers a series of conformational
changes in gp41 that ultimately leads to the formation of a
postfusion trimer-of-hairpins structure and membrane
fusion.[2–4] The core of the postfusion trimer-of-hairpins
structure is a bundle of six a helices: three N-peptide helices
form an interior,parallel,coiled-coil trimer,while three
C-peptide helices pack in an antiparallel manner into hydro-
phobic grooves on the coiled-coil surface (Figure 1).[2–4] The
N-peptide region features a hydrophobic pocket targeted by
[5]
C-peptide residues W628,W631,and I635.
Peptides and
synthetic molecules that bind to the N-terminal hydrophobic
pocket and inhibit the formation of the six-helix bundle have
been shown to effectively inhibit gp41-mediated HIV
fusion.[6–11] Herein,we describe the structure–activity rela-
tionships of hydrogen-bond-surrogate-derived a helices that
inhibit HIV fusion in cell culture. Our work suggests that
hydrogen-bond-surrogate (HBS) helices can potentially func-
tion as in vivo inhibitors of protein–protein interactions
involved in viral fusion.
Figure 1. a) The HIV-1 gp41 core six-helix bundle with the N region in
goldandthe C region in blue. b) Interaction of C-peptide residues
W628, W631, and I635 with the N-peptide (PDB code: 1AIK).[2]
c) Helical wheel diagram showing interactions between C- and
N-peptide coiled-coil domains.
HBS a helices are obtained by replacing an N-terminal
main-chain i and i+4 hydrogen bond with a carbon–carbon
bond through
a
ring-closing metathesis reaction
(Figure 2).[12–14] The hydrogen-bond surrogate preorganizes
an a turn and stabilizes the peptide sequence in an a-helical
conformation. We have shown that hydrogen-bond-surrogate
a helices adopt stable a-helical conformations from a variety
[*] D. Wang, Prof. P. S. Arora
Department of Chemistry, New York University
New York, NY 10003 (USA)
Fax: (+1)212-260-7905
E-mail: arora@nyu.edu
Figure 2. Hydrogen-bond-surrogate (HBS) a helices feature a carbon–
carbon bondin place of an i and i+4 hydrogen bond. R = amino acid
side chain.
Prof. M. Lu
Department of Biochemistry
Weill Medical College of Cornell University
New York, NY 10021 (USA)
[**] P.S.A. is grateful for financial support from the National Institutes of
Health (NIH) (grant no.: GM073943). We thank the National
Science Foundation for equipment grants (grant nos.: MRI-0116222
andCHE-0234863) andthe National Center for Research Re-
sources/NIH for a Research Facilities Improvement Grant (grant
no.: C06 RR-16572). M.L. thanks the NIH (awardno.: AI42382) and
the Irma T. Hirschl Trust for financial support.
of short peptide sequences.[15] We have also demonstrated that
these artificial a helices can target their expected protein
receptor with high affinity.[16]
We began our gp41-targeting studies by mimicking the
14-residue C-peptide derived from gp41 that contains resi-
dues W628,W631,and I635 ( 1,Table 1). This sequence would
be expected to bind the gp41 hydrophobic pocket based on
Supporting information for this article is available on the WWW
under http://www.angewandte.org or from the author.
Angew. Chem. Int. Ed. 2008, 47, 1879 –1882
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1879

Communications
Table 1: Binding affinities and cell-fusion-inhibition properties of pep-
tides and HBS a helices.
[a]
CompoundSequence
Kd [mm][b]
EC50 [mm][c]
C-peptide
NNMTWMEWDREINNYTSLI
–
–
–
1
2
3
4
5
6
7
8
9
AcMTWMEWDREINNYT
XK MTWL MEWDREINNYT
XK WAAL WDKKI
XK AAAL WEEWDKKI
XK WAAL WDREINNYT
XK MTWL EEWDKKIEEYT
XK EMAL WEEWDKKIEEYT
XK NEML TWEEWDKKIEEYT
XK MTWL EEWDKKIEEYTKKI
AcMTWEEWDKKIEEYTKKI
37.4Æ14.8[d]
46.6Æ14.6[d]
>500
@200
@200
@200
@200
@200
@200
@200
42.7Æ7.50
@200
>500
>500
7.50Æ1.70
146Æ47.3
<5.00[e]
<5.00[e]
<5.00[e]
10
[a] X: pentenoic acid residue; Ac: acetyl. Residues that occupy a or d
positions in the heptad are shown in bold. [b] Binding affinity for IZN17
[17]
as calculatedfrom a fluorescence polarization assay.
[c] Inhibitory
activity of peptides in a gp41-mediated cell–cell fusion assay monitoring
syncytia formation.[18] [d] Precise measurements of the Kd value were not
possible because the peptide aggregates at high concentrations. [e] The
competitive binding assay (see the Supporting Information) does not
allow accurate estimates of Kd values much lower than binding values of
the fluorescent probe.
the crystal structure (Figure 1b)[2] but has been previously
shown to be ineffective in cell–cell fusion assays.[8]
We utilized a previously described fluorescence polar-
ization assay to determine the in vitro binding affinity of
unconstrained peptides and HBS helices for a stable model of
the gp41 N-terminal three-strand coiled-coil,IZN17,which
contains the binding site for residues W628,W631,and
I635.[17] A fluorescein-labeled peptide,containing residues
628–641 of gp41,was used as a probe. The competitive
displacement of this probe by HBS helices afforded the
dissociation constant (Kd) value for each peptide (Table 1 and
Figure 3b).[11] Details of the fluorescence polarization assay
are included in the Supporting Information. The fluorescein-
labeled C-peptide derivative (suc-MTWMEWDREIN-
NYTCFlu; suc: succinimidyl) bound IZN17 with a Kd value
Figure 3. a) Circular dichroism spectra of 1, 2, 5, and 9 in 10%
trifluoroethanol (TFE) in phosphate-bufferedsaline (PBS). b) Determi-
nation of peptide binding to IZN17 by a fluorescence polarization
assay. c) Inhibition of gp41-mediated cell–cell fusion by HBS a helix 9
andpeptide 10.
[8,11]
of 24 mm,within the range of previously reported values.
The competition assay provided a binding affinity of 37 mm for
peptide 1,which is also in the range of reported values,
although aggregation of the peptide made it difficult to obtain
accurate Kd values.
The limited solubility of short C-peptides in aqueous
solution has been proposed as a key reason for their
inactivity.[19] To improve the solubility and helicity of con-
strained C-peptides,we incorporated charged residues at
positions not expected to be involved in binding to IZN17,
based on the work of Otaka,Fujii,and co-workers. [19] HBS a
helix 3 is the shortest peptide designed to contain the Trp/Trp/
Ile motif necessary for binding. Two lysine residues were
added within this sequence to improve the solubility of the
peptide. We designed HBS 4 such that the Trp/Trp/Ile motif
would lie outside of the HBS constraint,in order to probe the
potential steric effects of this constraint. Pairs of glutamic acid
and lysine residues were placed at the i and i+4 positions to
further stabilize the a-helical conformation through potential
salt-bridging interactions.[20,21] Neither of these two HBS
helices bound IZN17,a result suggesting that more contact
points are needed to target the coiled-coil complex. HBS 5
Circular dichroism spectroscopy suggested that 1 is
unstructured or very weakly helical in 10% TFE in PBS.[8]
We conjectured that stabilization of this peptide in helical
conformation with the HBS approach may increase its helicity
and affinity for IZN17. HBS a helix 2 is roughly four times
more helical than 1 (Figure 3a) but does not bind the target
protein with higher affinity (Table 1). This outcome suggests
that the aggregation state of the peptide is influencing the
binding affinity for the target. Our results with the short
peptide and constrained helix mirror those observed by Kim
and co-workers,[8] and this prompted us to design a small
library of HBS a helices to determine sequences that bind
IZN17 with higher affinity. A representative selection of HBS
helices prepared in this project is shown in Table 1.
1880
www.angewandte.org
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2008, 47, 1879 –1882

Angewandte
Chemie
contains Y638 in addition to the Trp/Trp/Ile residues; Y638
occupies the d position in the heptad and is expected to
directly contact the N-terminal three-strand coiled-coil (Fig-
ure 1c). The HBS 5 sequence is similar to that of the parent
compounds 1 and 2 but is missing the N-terminal M626 and
T627 residues. This compound also showed negligible affinity
for IZN17,which was surprising because we had expected 5 to
bind IZN17 with affinity similar to that observed for 2. CD
spectroscopy suggests that 2 and 5 are equally helical
(Figure 3a). The low binding affinity of 5 relative to that of
2 highlights the potential role of T627,which occupies the g
position in the heptad and was not projected to contribute
significantly to binding (Figure 1c). Reintroduction of the
M626 and T627 residues and incorporation of charged
residues at positions not expected to be involved in binding
provided HBS 6,which bound IZN17 with a higher affinity
than the parent HBS a helix 2. The role of T627 in the binding
of HBS helices is also highlighted by HBS 7,in which this
residue is replaced with an alanine residue.[22] HBS 7 bound
IZN17 with 20-fold lower affinity than 6.
the proteolytic stability of HBS a helices as compared to that
of their unconstrained counterparts.[16]
In summary,through rational design and synthesis,we
have developed an artificial a helix, 9,that inhibits gp41-
mediated cell fusion. As the formation of coiled-coil assem-
blies is a prerequisite for the fusion of several classes of
viruses to their host cells,[27] this work suggests that HBS
helices may be effective scaffolds for the generation of small-
molecule inhibitors or antigens against these viruses.[28,29]
Received: September 14,2007
Published online: January 25,2008
Keywords: helical structures · inhibitors · molecular recognition ·
.
peptidomimetics · protein–protein interactions
[1] D. M. Eckert,P. S. Kim, Annu. Rev. Biochem. 2001, 70,777 – 810.
[2] D. C. Chan,D. Fass,J. M. Berger,P. S. Kim, Cell 1997, 89,263 –
273.
HBS helices 8 and 9 were prepared to explore the effect of
additional contact points (residues that occupy a or d
positions in the heptad) on the amino and carboxy ends.
HBS helices 8 and 9 both include five residues that occupy a
or d positions in the heptad and bind IZN17 with Kd values
< 5 mm. The values for 8 and 9 represent the lower limits in the
present 1ZN17 assay because competitive-binding analysis
does not allow accurate estimates of Kd values much lower
than the binding affinity of the fluorescent probe (see the
Supporting Information).[23] Although the 1ZN17 assay did
not provide accurate values for the best HBS helices,it
allowed us to qualitatively evaluate various constructs
described herein. We decided that a combination of the
1ZN17 binding assay and a gp41-mediated cell–cell fusion
assay would provide a better gauge of the effectiveness of
HBS helices as inhibitors of HIV fusion.[18]
Cell–cell fusion (that is,syncytium formation) was assayed
by coculturing CHO[HIVe] (clone 7d2) cells expressing
HXB2 envelope and Tat[24] with U373-MAGI cells in the
presence of different concentrations of the peptide inhibitors.
Cell fusion allows the expression of nuclear b-galactosidase
from the U373-MAGI indicator cell line and can be quanti-
tated by monitoring b-galactosidase activity.
[3] W. Weissenhorn,A. Dessen,S. C. Harrison,J. J. Skehel,D. C.
Wiley, Nature 1997, 387,426 – 430.
[4] K. Tan,J. Liu,J. Wang,S. Shen,M. Lu,
Proc. Natl. Acad. Sci.
USA 1997, 94,12303 – 12308.
[5] D. C. Chan,C. T. Chutkowski,P. S. Kim, Proc. Natl. Acad. Sci.
USA 1998, 95,15613 – 15617.
[6] C. T. Wild,D. C. Shugars,T. K. Greenwell,C. B. McDanal,T. J.
Matthews, Proc. Natl. Acad. Sci. USA 1994, 91,9770 – 9774.
[7] M. Ferrer,T. M. Kapoor,T. Strassmaier,W. Weissenhorn,J. J.
Skehel,D. Oprian,S. L. Schreiber,D. C. Wiley,S. C. Harrison,
Nat. Struct. Biol. 1999, 6,953 – 960.
[8] S. K. Sia,P. A. Carr,A. G. Cochran,V. N. Malashkevich,P. S.
Kim, Proc. Natl. Acad. Sci. USA 2002, 99,14664 – 14669.
[9] J. T. Ernst,O. Kutzki,A. K. Debnath,S. Jiang,H. Lu,A. D.
Hamilton, Angew. Chem. 2002, 114,288 – 291; Angew. Chem. Int.
Ed. 2002, 41,278 – 281.
[10] G. Frey,S. Rits-Volloch,X. Q. Zhang,R. T. Schooley,B. Chen,
S. C. Harrison, Proc. Natl. Acad. Sci. USA 2006, 103,13938 –
13943.
[11] O. M. Stephens,S. Kim,B. D. Welch,M. E. Hodsdon,M. S. Kay,
A. Schepartz, J. Am. Chem. Soc. 2005, 127,13126 – 13127.
[12] R. N. Chapman,P. S. Arora, Org. Lett. 2006, 8,5825 – 5828.
[13] R. N. Chapman,G. Dimartino,P. S. Arora, J. Am. Chem. Soc.
2004, 126,12252 – 12253.
[14] G. Dimartino,D. Wang,R. N. Chapman,P. S. Arora, Org. Lett.
2005, 7,2389 – 2392.
[15] D. Wang,K. Chen,J. L. Kulp III,P. S. Arora, J. Am. Chem. Soc.
2006, 128,9248 – 9256.
[16] D. Wang,W. Liao,P. S. Arora, Angew. Chem. 2005, 117,6683 –
6687; Angew. Chem. Int. Ed. 2005, 44,6525 – 6529.
[17] D. M. Eckert,P. S. Kim, Proc. Natl. Acad. Sci. USA 2001, 98,
11187 – 11192.
[18] D. M. Eckert,V. N. Malashkevich,L. H. Hong,P. A. Carr,P. S.
Kim, Cell 1999, 99,103 – 115.
Inhibition of gp41-mediated cell fusion by short peptides
is a challenging feat and has only been accomplished with a
handful of synthetic peptides.[8,11,18] We found that only HBS a
helix 9 inhibited cell fusion,with an EC
value of 43 mm
50
(Figure 3c).[25] This value is comparable to those measured
for side-chain-constrained a helices,[8] cyclic d-peptides,[18]
aromatic foldamers,[9] and b-peptide foldamers.[11] The other
HBS a helices, 2–8,did not provide any hint of cell-fusion
inhibition at concentrations up to 200 mm. Unconstrained
peptide 10,which bound IZN17 with a similar affinity to that
of 9,remained ineffective in the cell-culture assay. This result
potentially reflects the proteolytic instability of the uncon-
strained peptide,as stabilization of peptides in a-helical
conformation is expected to enhance their resistance to
proteases.[26] We have previously reported improvements in
[19] A. Otaka,M. Nakamura,D. Nameki,E. Kodama,S. Uchiyama,
S. Nakamura,H. Nakano,H. Tamamura,Y. Kobayashi,M.
Matsuoka,N. Fujii, Angew. Chem. 2002, 114,3061 – 3064;
Angew. Chem. Int. Ed. 2002, 41,2937 – 2940.
[20] S. Marqusee,R. L. Baldwin, Proc. Natl. Acad. Sci. USA 1987, 84,
8898 – 8902.
[21] Z. S. Shi,C. A. Olson,A. J. Bell,N. R. Kallenbach, Biopolymers
2001, 60,366 – 380.
[22] Glutamic acid residues were placed at the N termini of 7 and 8 to
potentially stabilize the helix macrodipoles; however,incorpo-
Angew. Chem. Int. Ed. 2008, 47, 1879 –1882
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
1881

Communications
ration of this residue did not effect the binding affinity of the
peptides.
[26] J. D. Tyndall,T. Nall,D. P. Fairlie, Chem. Rev. 2005, 105,973 –
999.
[23] M. H. Roehrl,J. Y. Wang,G. Wagner, Biochemistry 2004, 43,
16056 – 16066.
[24] K. Kozarsky,M. Penman,L. Basiripour,W. Haseltine,J.
Sodroski,M. Krieger, J. Acquired Immune Defic. Syndr. 1989,
2,163 – 169.
[27] D. S. Dimitrov, Nat. Rev. Microbiol. 2004, 2,109 – 122.
[28] E. P. English,R. S. Chumanov,S. H. Gellman,T. Compton,
Biol. Chem. 2005, 281,2661 – 2667.
[29] N. E. Shepherd,H. N. Hoang,V. S. Desai,E. Letouze,P. R.
Young,D. P. Fairlie, J. Am. Chem. Soc. 2006, 128,13284 – 13289.
J.
[25] T. Mosmann, J. Immunol. Methods 1983, 65,55 – 63.
1882
www.angewandte.org
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2008, 47, 1879 –1882