Parallel synthesis of H-pin polyamides by alkene metathesis on solid phase

Article abstract of DOI:10.1021/ja0213221

A small library of H-pin polyamides with variable aliphatic bridge lengths (CH₂)_n, where n = 4-8, connecting a central Py/Py pair was prepared via parallel synthesis with Ru-catalyzed alkene metathesis on solid phase as a complexity-generating cross-linking reaction. DNA binding affinities and sequence specificities were analyzed for each member of the library to determine the optimum linker length. An H-pin polyamide with a six-methylene bridge was found to have the highest affinity to its match site with high selectivity over a 1-bp mismatch site. The relationship between the number of methylenes in the linker (CH₂)_n and affinity is n = 6 > 4 > 7 > 5 > 8. These results indicate that 6 followed by 4 methylene-bridged polyamides represent the optimum spacer length for the H-pin motif in the DNA minor groove. Importantly, the H-pin is competitive with hairpin polyamides with respect to affinity and specificity. The metathesis-based convergent synthetic route to H-pin polyamides expands the scope of readily available DNA recognition motifs for small molecule-based gene regulation studies.

Full text of DOI:10.1021/ja0213221

Published on Web 03/29/2003
Parallel Synthesis of H-pin Polyamides by Alkene Metathesis
on Solid Phase
Bogdan Olenyuk, Cristian Jitianu, and Peter B. Dervan*
Contribution from the DiVision of Chemistry and Chemical Engineering, The Beckman Institute,
California Institute of Technology, Pasadena, California 91125
Received November 1, 2002; E-mail: dervan@caltech.edu
Abstract: A small library of H-pin polyamides with variable aliphatic bridge lengths (CH2)n, where n )
4-8, connecting a central Py/Py pair was prepared via parallel synthesis with Ru-catalyzed alkene
metathesis on solid phase as a complexity-generating cross-linking reaction. DNA binding affinities and
sequence specificities were analyzed for each member of the library to determine the optimum linker length.
An H-pin polyamide with a six-methylene bridge was found to have the highest affinity to its match site
with high selectivity over a 1-bp mismatch site. The relationship between the number of methylenes in the
linker (CH2)n and affinity is n ) 6 > 4 > 7 > 5 > 8. These results indicate that 6 followed by 4 methylene-
bridged polyamides represent the optimum spacer length for the H-pin motif in the DNA minor groove.
Importantly, the H-pin is competitive with hairpin polyamides with respect to affinity and specificity. The
metathesis-based convergent synthetic route to H-pin polyamides expands the scope of readily available
DNA recognition motifs for small molecule-based gene regulation studies.
Introduction
The “hairpin motif” connects the N and C termini of the two
strands with a γ-aminobutyric acid unit (γ-turn) to form a folded
The development of DNA-binding ligands capable of target-
ing a broad range of sequences with high affinity and sequence
specificity is essential for the regulation of gene expression by
chemical methods.1 Synthetic polyamides comprising N-meth-
ylpyrrole (Py), N-methylimidazole (Im), and 3-hydroxypyrrole
(Hp) amino acids represent a set of molecular recognition tools
with potential for transcriptional modulation,2a-e chromatin
modification,2f promoter scanning,2g and chromosome mapping.2h,i
Recognition of a specific DNA sequence by two antiparallel
polyamide strands depends on a code of side-by-side aromatic
amino acid pairs in the minor groove, usually oriented N to C
with respect to the 5′ to 3′ direction of the DNA helix.3
Enhanced affinity and specificity of polyamide DNA binding
is accomplished by covalently linking the antiparallel strands.4-6
linear chain.4 The “H-pin motif” connects the antiparallel strands
across a central ring/ring pair5 by a short, flexible polymethylene
bridge (Figure 1). The branched nature of H-pin polyamides
made solid-phase synthesis of this motif less convenient,
hindering a broad exploration of its potential regarding DNA-
binding properties and biological applications. In this paper, we
present a convergent synthesis of H-pin polyamides based on
alkene metathesis as the key step for cross-linking the two
polyamide strands. To investigate the optimum distance for the
flexible linker, we used this metathesis strategy to create a small
library of H-pin polyamides with incrementally increasing
aliphatic linker lengths, from (CH2)4 to (CH2)8 (Figure 2).6 The
parallel synthesis allows the preparation of five 10-ring H-pin
polyamides 1-5. A control hairpin polyamide ImPyPyPyPy-
γ-ImPyPyPyPy-â-Dp 6 was also synthesized by standard solid-
phase protocol.7 Subsequent DNase I footprint titration experi-
(1) For reviews: (a) Dervan, P. B. Bioorg. Med. Chem. 2001, 9, 2215-2235.
(b) Sharma, S. K.; Reddy, B. S. N.; Lown, J. W. Drug Future 2001, 26,
39-49. (c) Bailly, C.; Chaires, J. B. Bioconjugate Chem. 1998, 9, 513-
538.
(2) (a) Dickinson, L. A.; Gulizia, R. J.; Trauger, J. W.; Baird, E. E.; Mosier,
D. E.; Gottesfeld, J. M.; Dervan, P. B. Proc. Natl. Acad. Sci. U.S.A. 1998,
95, 12890-12895. (b) Mapp, A. K.; Ansari, A. Z.; Ptashne, M.; Dervan,
P. B. Proc. Natl. Acad. Sci. U.S.A. 2000, 97, 3930-3935. (c) White, C.
M.; Satz, A. L.; Bruice, T. C.; Beerman, T. A. Proc. Natl. Acad. Sci. U.S.A.
2001, 98, 10590-10595. (d) Janssen, S.; Durussel, T.; Laemmli, U. K.
Mol. Cell 2000, 6, 999-1011. (e) Janssen, S.; Cuvier, O.; Mu¨ller, M.;
Laemmli, U. K. Mol. Cell 2000, 6, 1013-1024. (f) Gottesfeld, J. M.;
Melander, C.; Suto, R. K.; Raviol, H.; Luger, K.; Dervan, P. B. J. Mol.
Biol. 2001, 309, 615-629. (g) Ehley, J. A.; Melander, C.; Herman, D.;
Baird, E. E.; Ferguson, H. A.; Goodrich, J. A.; Dervan, P. B.; Gottesfeld,
J. Mol. Cell Biol. 2002, 22, 1723-1733. (h) Gygi, M. P.; Ferguson, M.
D.; Mefford, H. C.; Lund, K. P.; O’Day, C.; Zhou, P.; Friedman, C.; van
den Engh, G.; Stolowitz, M. L.; Trask, B. J. Nucleic Acids Res. 2002, 30,
2790-2799. (i) Maeshima, K.; Janssen, S.; Laemmli, U. K. EMBO J. 2001,
20, 3218-3228.
(4) (a) Mrksich, M.; Parks, M. E.; Dervan, P. B. J. Am. Chem. Soc. 1994, 116,
7983-7988. (b) Trauger, J. W.; Baird, E. E.; Dervan, P. B. Nature 1996,
382, 559-561. (c) Herman, D. M.; Baird, E. E.; Dervan, P. B. J. Am. Chem.
Soc. 1998, 120, 1382-1391. (d) Swalley, S. E.; Baird, E. E.; Dervan, P.
B. J. Am. Chem. Soc. 1999, 121, 1113-1120.
(5) (a) Mrksich, M.; Dervan, P. B. J. Am. Chem. Soc. 1993, 115, 9892-9899.
(b) Dwyer, T. J.; Geierstanger, B. H.; Mrksich, M.; Dervan, P. B.; Wemmer,
D. E. J. Am. Chem. Soc. 1993, 115, 9900-9906. (c) Mrksich, M.; Dervan,
P. B. J. Am. Chem. Soc. 1994, 116, 3663-3664. (d) Greenberg, W. A.;
Baird, E. E.; Dervan, P. B. Chem. Eur. J. 1998, 4, 796-805.
(6) (a) Chen, Y. H.; Lown, J. W. J. Am. Chem. Soc. 1994, 116, 6995-7005.
(b) Chen, Y. H.; Yang, Y.; Lown, W. J. J. Biomol. Struct. 1996, 14, 341-
355. (c) O’Hare, C. C.; Mack, D.; Tandon, M.; Sharma, S. K.; Lown, J.
W.; Kopka, M. L.; Dickerson, R. E.; Hartley, J. A. Proc. Natl. Acad. Sci.
U.S.A. 2002, 99, 72-77.
(3) (a) White, S.; Szewczyk, J. W.; Turner, J. M.; Baird, E. E.; Dervan, P. B.
Nature 1998, 391, 468-471. (b) Kielkopf, C. L.; White, S.; Szewczyk, J.
W.; Turner, J. M.; Baird, E. E.; Dervan, P. B.; Rees, D. C. Science 1998,
282, 111-115.
(7) (a) Baird, E. E.; Dervan, P. B. J. Am. Chem. Soc. 1996, 118, 6141-6146.
(b) Wurtz, N. R.; Turner, J. M.; Baird, E. E.; Dervan, P. B. Org. Lett.
2001, 3, 1201-1203.
9
10.1021/ja0213221 CCC: $25.00 © 2003 American Chemical Society
J. AM. CHEM. SOC. 2003, 125, 4741-4751
4741

A R T I C L E S
Olenyuk et al.
Figure 1. DNA binding model for two polyamide motifs utilizing antiparallel pyrrole-imidazole strands: asymmetric hairpin (a, b) and symmetric H-pin
(c). (Top) Circles with dots represent lone pairs of N3 of purines and O2 of pyrimidines. Circles containing H represent the N2 hydrogens of guanine.
Putative hydrogen bonds are illustrated by dotted lines. (Bottom) Solid circles represent imidazoles, open circles represent pyrroles, and diamonds denote
â-alanines. The aliphatic linkers of the H-pin and the γ-aminobutyric acid of the hairpin are depicted as curved lines. Curved lines with plus signs represent
N,N-[(dimethylamino)propyl]amine tails.
ments allowed the relationship of bridge distance to be correlated
with binding affinity and specificity.
phase synthesis of dimeric molecules10 and libraries of homo-
and heterodimeric compounds11 and in the solid-phase synthesis
of natural products.12
Cross-linking via intrasite interactions on solid support have
been demonstrated,13 albeit infrequently, since the introduction
of solid-phase chemistry. Despite the initial belief about the
paucity of these interactions, cross-reactivity between the
individual peptide chains, in some cases with excellent yields,
Results and Discussion
Parallel Synthesis via Metathesis on Resin. Solid-phase
syntheses of pyrrole-imidazole polyamides involve stepwise
addition of an activated ester of the corresponding protected
amino acid to the resin-bound amine at the end of the growing
polyamide chain.7 This approach, useful in the preparation of
linear hairpin polyamides, has limitations when applied to the
synthesis of certain branched structures, such as H-pin poly-
amides. In the synthetic strategy described here, we start with
simultaneous synthesis of the two C-terminal sections of the
H-pin on solid support, followed by the cross-linking reaction
of a middle monomer on resin. After the cross-linking step, the
synthesis can be continued with the appropriate terminal residues
being added to the growing chain in the usual manner. We chose
olefin metathesis using the Grubbs catalyst as the cross-linking
reaction,8 due to the ease of preparation of the alkenyl-pyrrole
precursors and the high functional group tolerance of the Ru
catalyst.9 This method has been used previously in the solution-
(8) (a) Grubbs, R. H.; Chang, S. Tetrahedron 1998, 54, 4413-4450. (b) Trnka,
T. M.; Grubbs, R. H. Acc. Chem. Res. 2001, 34, 18-29. (c) Grubbs, R. H.
Chimia 2002, 56, 22-25. (d) Blackwell, H. E.; Sadowsky, J. D.; Howard,
R. J.; Sampson, J. N.; Chao, J. A.; Steinmetz, W. E.; O’Leary, D. J.; Grubbs,
R. H. J. Org. Chem. 2001, 66, 5291-5302.
(9) Toste, F. D.; Chatterjee, A. K.; Grubbs, R. H. Pure Appl. Chem. 2002, 74,
7-10.
(10) (a) Nicolaou, K. C.; Hughes, R.; Cho, S. Y.; Winssinger, N.; Smethurst,
C.; Labischinski, H.; Endermann, R. Angew. Chem., Int. Ed. 2000, 39,
3823-3828. (b) Diver, S. T.; Schreiber, S. L. J. Am. Chem. Soc. 1997,
119, 5106-5109.
(11) (a) Bra¨ndli, C.; Ward, T. R. HelV. Chim. Acta 1998, 81, 1616-1621. (b)
Boger, D. L.; Chai, W. Y.; Jin, Q. J. Am. Chem. Soc. 1998, 120, 7220-
7225. (c) Giger, T.; Wigger, M.; Audetat, S.; Benner, S. A. Synlett 1998,
688-691. (d) Gierasch, T. M.; Chytil, M.; Didiuk, M. T.; Park, J. Y.; Urban,
J. J.; Nolan, S. P.; Verdine, G. L. Org. Lett. 2000, 2, 3999-4002. (e)
Blackwell, H. E.; Clemons, P. A.; Schreiber, S. L. Org. Lett. 2001, 3, 1592-
1592.
9
4742 J. AM. CHEM. SOC. VOL. 125, NO. 16, 2003

Synthesis of H-pin Polyamides by Alkene Metathesis
A R T I C L E S
Figure 2. (Top) Structures of 10-ring H-pin polyamides with aliphatic linkers of 4-8 methylenes. (Bottom) Schematic illustration of H-pin polyamide
bound to predetermined match site (symbols are as described for Figure 1).
Figure 3. Synthesis of the pyrrole precursors containing variable alkenyl chains.
was quickly established and became an obstacle in synthesis of
large peptide fragments. To circumvent these problems, resins
with lower loading (0.05-0.2 mmol/g) and a high degree of
cross-linking with divinylbenzene (10-20%) were employed,
sometimes in solvents and conditions that prevent maximal
swelling of the resin. Therefore, we chose commercially
available PAM resins with a low degree of cross-linking (∼1%
divinylbenzene) and high loads (0.88-1.2 mmol/g) with dichlo-
romethane as a solvent to obtain high cross-linking yields.
Similar conditions have been reported in solid-phase synthesis
of libraries of dimeric molecules with short peptide backbones.11e
When two different terminal alkenes with similar reactivities
toward the catalyst are employed in the olefin cross-metathesis,
the result is a mixture of three products: two homodimers and
one heterodimer. This often unwanted side effect of the typical
cross-metathesis reaction could be an advantageous step when
mixed products are desirable. Indeed, if several terminal alkenes
with variable chain lengths are used simultaneously, then upon
completion of the metathesis reaction, a mixture of homo- and
heterodimeric products resulting from all possible combinations
of chain lengths in precursors will be formed. Therefore, for
our purposes, a library of five different cross-linked products
can be generated from three common precursors.
(12) (a) Nicolaou, K. C.; Winssinger, N.; Pastor, J.; Ninkovic, S.; Sarabia, F.;
He, Y.; Vourloumis, D.; Yang, Z.; Li, T.; Giannakakou, P.; Hamel, E.
Nature 1997, 387, 268-272. (b) Schinzer, D.; Limberg, A.; Bauer, A.;
Bohm, O. M.; Cordes, M. Angew. Chem., Int. Ed. Engl. 1997, 36, 523-
524. (c) Meng, D. F.; Bertinato, P.; Balog, A.; Su, D. S.; Kamenecka, T.;
Sorensen, E. J.; Danishefsky, S. J. J. Am. Chem. Soc. 1997, 119, 10073-
10092. (d) May, S. A.; Grieco, P. A. Chem. Commun. 1998, 1597-1598.
(e) Biswas, K.; Lin, H.; Njargarson, J. T.; Chappell, M. D.; Chou, T.-C.;
Guan, Y.; Tong, W. P.; He, L.; Horwitz, S. B.; Danishefsky, S. J. J. Am.
Chem. Soc. 2002, 124, 9825-9832.
(13) (a) Kraus, M. A.; Patchornik, A. J. Am. Chem. Soc. 1971, 93, 7325-7327.
(b) Rapoport, H.; Crowley, J. I. Acc. Chem. Res. 1976, 9, 135-144. (c)
Yan, B.; Sun, Q. J. Org. Chem. 1998, 63, 55-58. (d) Yan, B. Combinator.
Chem. High Throughput Screen. 1998, 1, 215-229. (e) Lindsley, C. W.;
Chan, L. K.; Goess, B. C.; Joseph, R.; Shair, M. D. J. Am. Chem. Soc.
2000, 122, 422-423.
Three N-alkenyl substituted pyrrole amino acids 11a-c were
synthesized as described in Figure 3. Nitration of the com-
mercially available 2-trichloroacetyl-1H-pyrrole 7, followed by
haloform reaction, provided 1,1,1-trimethylsilylethyl (TMSE)
ester 8. It was hydrogenated and converted into Boc-protected
amine 9, which was then separately alkylated with three alkenyl
bromides resulting in TMSE esters of alkenyl pyrroles 10a-c.
They were deprotected with TBAF to yield the acids 11a-c. A
mixture of corresponding activated esters 12a-c was added to
the growing chain containing two pyrrole monomers 13 attached
9
J. AM. CHEM. SOC. VOL. 125, NO. 16, 2003 4743

A R T I C L E S
Olenyuk et al.
ethanol.H-pinpolyamidesofsequencecomposition(ImPyPyPyPy-
â-Dp)2(CH2)4-8 1-5 were separated by semipreparative HPLC
and their masses were confirmed by MALDI-TOF mass
spectrometry. Remarkably, despite the structural similarity, each
compound was separated on reverse-phase C18 HPLC columns
with a relatively slow gradient (Figure 6). The prepared amounts
of 1-5 (100-200 nM of each H-pin polyamide compound per
100 mg of resin) were sufficient for further characterization as
well as DNA-binding studies.
Alternative Synthesis on Resin with Cross-Linked Mono-
mers. Having established the utility of metathesis-based cross-
linking for the synthesis of a library of H-pin polyamides, we
examined an alternative cross-linking strategy, which could be
scalable and compatible with standard conditions used in
automated solid-phase peptide synthesis where transition metal
chemistry might be more difficult to implement. Our alternative
approach involved preparation of the dimeric building block
where two pyrroles are linked through ring nitrogens (Figure
7). As in the case of the library synthesis, olefin metathesis
was used to prepare the dimer 17 from N-alkenyl-substituted
pyrrole monomer 16 (Figure 7), which provided the Boc-
protected amine 18, diacid 19, and finally the activated ester
20 after several straightforward synthetic transformations. Solid-
phase synthesis was performed as depicted in Figure 8. For
convenience, we used separately prepared BocPyPyCOOBt and
ImPyCOOBt building blocks7a that allowed us to reduce the
number of steps on solid phase, although stepwise addition of
one monomer at a time may be utilized. On high-loading resin
with a low cross-link rate (0.88 mmol/g, 1% divinylbenzene),
reaction of the dimeric residue 20 with resin 21 proceeds with
remarkable efficiency to produce cross-linked product 22.
Further deprotection and coupling steps were performed without
significant modifications to the standard polyamide protocol.7
Removal from the solid support followed by HPLC purification
provided the desired product 3. Notably, the described method
requires roughly half the synthetic steps compared to the
synthesis of a 10-ring hairpin. In addition, using high-load resin
provides a larger amount of product for a given amount of the
reagents and therefore is more economical and amenable to
large-scale synthesis.
Figure 4. (a) Preparation of a resin-bound mixture of three-ring polyamides
via standard solid-phase synthetic protocols utilizing Boc chemistry. (b) A
mixture of three pyrrole precursors with variable alkenyl chains lengths
was added in the final step.
to the â-alanine-PAM resin (Figure 4) in accordance with
previous solid-phase procedures for polyamides.7 Treatment of
alkenyl-derivatized resin 14 with 12-18 mol % of Grubbs’s
ruthenium benzylidene catalyst8,9 in dichloromethane under Ar,
followed by an additional 12 mol % catalyst and 12 h under Ar
and rigorous washing, produced a small library of cross-
metathesis products (Figures 5 and 6). The time course of the
reaction and the purity of the resulting library products were
monitored by cleaving a resin sample with N,N′-[(dimethylami-
no)propyl]amine (Dp) followed by HPLC analysis and identi-
fication of each individual peak by MALDI-TOF mass spec-
trometry. After satisfactory conversion of the individual chains
to cross-linked dimers, the resin was washed with dichlo-
romethane and treated with 1,3-bis(diphenylphosphino)propane
(dppp) in dichloromethane to remove catalyst decomposition
products, and the solid-phase synthesis was continued. After
deblocking, activated pyrrole monomer was added and the
coupling was carried out as before, followed by another
deblocking and coupling of an activated imidazole acid. The
solid-phase synthesis was concluded by cleaving the contents
of the resin with Dp (Figure 6). As expected, after cross-linking
the resin exhibited a lesser degree of swelling in organic
solvents, but it is worth noting that this did not preclude
successful coupling in the subsequent steps. After cleavage from
the solid support, the excess of cleaving diamine was removed
first in high vacuum and subsequently by passing the crude
mixture dissolved in 50% acetonitrile-0.1% aqueous TFA
through a Sep-Pak (Waters). The products were reduced by use
of a mixture of palladium acetate-ammonium formate in
Binding Affinities and Sequence Specificities. To compare
the binding affinity of the 10-ring H-pin polyamides 1-5 with
the variable methylene bridge as well as a control 10-ring hairpin
6 to their respective match and mismatch sites, a restriction
fragment of plasmid pDHN2 was used (Figure 9). It contains
three sites: a match 5′-ATGTTACAT-3′, a single bp mismatch
5′-ATGGTACAT-3′, and a double bp mismatch 5′-ATGGTC-
CAT-3′. Quantitative DNase I footprint titrations were per-
formed to determine the equilibrium association constants for
each polyamide-DNA complex (Ka) according to previously
published protocols.14,15
Comparison of the equilibrium association constants of 1-5
reveals that H-pin polyamide 3 (ImPyPyPyPy-â-Dp)2(CH2)6 has
the highest affinity in the library to its match site (Ka ) 4.3 ×
1010 M-1) and displays a 73-fold lower affinity (Ka ) 5.9 ×
(14) (a) Brenowitz, M.; Senear, D. F.; Shea, M. A.; Ackers, G. K. Methods
Enzymol. 1986, 130, 132. (b) Brenowitz, M.; Senear, D. F.; Shea, M. A.;
Ackers, G. K. Proc. Natl. Acad. Sci. U.S.A. 1986, 83, 8462. (c) Senear, D.
F.; Brenowitz, M.; Shea, M. A.; Ackers, G. K. Biochemistry 1986, 25,
7344.
(15) Trauger, J. W.; Dervan, P. B. Methods Enzymol. 2001, 340, 450.
9
4744 J. AM. CHEM. SOC. VOL. 125, NO. 16, 2003

Synthesis of H-pin Polyamides by Alkene Metathesis
A R T I C L E S
Figure 5. Synthesis of H-pin polyamides via on-resin cross-linking. (i) Ru catalyst Cl2(PCy3)2RudCHPh, DCM, 40 °C, 24-40 h, then second batch of Ru
catalyst, 40 °C, 24 h; (ii) TFA/DCM, 20 min, (iii) BocPyCOOBt, DMF, DIEA, 2 h; (iv) TFA/DCM, 20 min, (v) ImCOOBt, DMF, DIEA, 1 h; (vi) Dp, 60
°C, 6 h; (vii) ammonium formate, Pd(OAc)2, EtOH, 12 h.
108 M-1) for the 1-bp mismatch site (Figure 10). The (CH2)4-
containing the match sequence for symmetrical 10-ring aromatic
core plus two A/T sites downstream but only one A/T site
upstream. As a result, the control hairpin polyamide bound the
designed site with high affinity, as expected for the complete
match sequence interaction. H-pin polyamides, however, have
two positively charged tails. On the tested sequence, the first
tail was favorably positioned over the first two A/T sites,
whereas the second tail was situated over one A/T site and one
G/C site. The resulting unfavorable interaction of the positively
charged Dp moiety with the exocyclic guanine of the second
base pair diminished the binding affinity of the H-pin polyamide.
As a result, the observed equilibrium association constant (Ka
) 4.4 × 108 M-1) for the (CH2)4 bridged H-pin was lower than
observed in the current study, where a complete sequence match
of both tails, in addition to the aromatic core, was achieved.
Among all members of the small library, the C6-linked H-pin
was found to have the highest equilibrium association constant,
followed closely by the C4-linked H-pin. We did not observe a
gradual increase or decrease of the binding affinity as we
changed the bridge length. This may be attributed to the
linked H-pin polyamide 1 (ImPyPyPyPy-â-Dp)2(CH2)4 binds
the match site with 2-fold lower affinity, Ka ) 1.5 × 1010 M-1
.
The relationship between the number of methylenes in the linker
(CH2)n and affinity is n ) 6 > 4 > 7 > 5 > 8 (Table 1). The
analogous hairpin polyamide 6 (ImPyPyPyPy-γ-ImPyPyPyPy-
â-Dp) binds the match site with slightly lower affinity (Ka )
2.5 × 1010 M-1) than H-pin polyamide 3. In the sequence
context, specificity of the H-pin polyamide 3 was found to be
higher than that of control hairpin 6. In addition, nonspecific
DNA binding for H-pin 3 appears at 50 nM, whereas hairpin 6
reveals nonspecific binding at lower concentrations (∼10 nM).
The first DNA-binding studies of H-pin polyamides were
completed before the sequence specificities of the polyamide
tails (â and Dp) were rigorously established. At the C-terminus,
the tail units â and Dp specify for A/T with an energetic penalty
for G/C, presumably due to steric reasons.4d In an early study,
10-ring H-pin polyamides with (CH2)4 linkers were tested on a
DNA restriction fragment derived from a plasmid (pJK7)
9
J. AM. CHEM. SOC. VOL. 125, NO. 16, 2003 4745

A R T I C L E S
Olenyuk et al.
Figure 7. Preparation of Boc-protected bispyrrole dimer and its activated
ester.
binding polyamides and other libraries of dimeric small
molecules with gradually increased linker lengths. Each H-pin
polyamide shows subnanomolar DNA binding affinity for its
match site with the optimum bridges being (CH2)6, closely
followed by (CH2)4. The current study is restricted to H-pin
polyamides with C2 symmetry, but we are optimistic that the
synthetic strategy can be modified to allow convergent synthesis
of unsymmetrical H-pins. This report sets the stage for the
synthesis of H-pins lacking â-Dp tails capable of targeting pure
GC binding sites without the inherent A/T-specific interaction
of the hairpin γ-turn, as well as issues surrounding application
of H-pin polyamides in gene regulation studies.
Experimental Section
General. N,N′-dicyclohexylcarbodiimide (DCC), N-hydroxybenzo-
triazole (HOBt), Boc-â-alanine-(4-carboxamidomethyl)-benzyl-ester-
copoly(styrene-divinylbenzene) resin (Boc-â-PAM-resin), and 2-(1H-
benzotriazole-1-yl)-1,1,3,3-tetramethyluronium hexafluorophosphate
(HBTU) were purchased from Peptides International.
Figure 6. (a) Illustration of the cross-metathesis products of five different
lengths generated via a combination of three alkenyl chains in precursors.
(b) Analytical HPLC trace after cleavage of resin-bound precursor mixture
14 with Dp. (c) Analytical HPLC trace of the crude mixture of products
obtained after the reduction step.
N,N-Diisopropylethylamine (DIEA) and N,N-dimethylformamide
(DMF) and dimethyl sulfoxide (DMSO) were purchased from Applied
Biosystems. Reagent-grade dichloromethane (DCM) was from EM, and
trifluoroacetic acid (TFA) was from Halocarbon. All other chemicals
were obtained reagent-grade from Aldrich (unless otherwise stated) and
used without further purification. 1H NMR spectra were recorded on a
Varian Mercury 300 instrument. Chemical shifts are reported in parts
per million downfield from the signal for Me4Si, with reference to the
solvent residual signal. UV spectra were measured on a Hewlett-Packard
model 8452A diode array spectrophotometer. Matrix-assisted, laser
constrained conformation that each aliphatic chain may have
in the dimer upon binding to DNA.
Conclusions
We developed a convergent solid-phase parallel synthesis of
a library of H-pin polyamides with variable methylene linker
lengths by olefin cross-metathesis. The reported synthetic
strategy may be used for simultaneous preparation of DNA-
9
4746 J. AM. CHEM. SOC. VOL. 125, NO. 16, 2003

Synthesis of H-pin Polyamides by Alkene Metathesis
A R T I C L E S
Figure 8. Synthesis of H-pin polyamide 3 via on-resin cross-linking through peptide bond formation. (i) 80% TFA/DCM, 20 min; BocPyPyCOOBt, DMF,
DIEA, 1 h, 37 °C; 80% TFA/DCM, 20 min; (ii) 15, DMF, DIEA, 12 h; (iii) 80% TFA/DCM, 20 min; (iv) ImPyCOOBt, DMF, DIEA, 4 h, 37 °C; (v) Dp,
6 h, 37 °C.
Figure 9. Illustration of the EcoRI/PVuII restriction fragment with the BamHI/HindIII insertion sites indicated. The sequence of the synthesized insert from
the plasmid pDHN2 is shown with the polyamide binding sites boxed and in boldface type: match site A, double mismatch site B, and single mismatch site
C. Ball and stick models are shown as in Figure 1.
desorption/ionization time-of-flight (MALDI-TOF) mass spectrometry
was carried out on Voyager DE-Pro mass spectrometer from PerSeptive
Biosystems. Electrospray ionization mass spectrometry was carried out
on an LC-Q mass spectrometer from Finnigan Mat. Fast atom
bombardment mass spectrometry was carried out on a Jeol mass
spectrometer. HPLC analysis was performed on a Beckman Gold system
using a Rainin C18, Microsorb-MV, 5 µm, 300 × 4.6 mm reversed-
phase column in 0.1% (v/v) TFA with acetonitrile as eluent and a flow
rate of 1.0 mL/min, gradient elution 1.70% acetonitrile/min unless
otherwise stated. Preparatory HPLC was carried out on a Beckman
HPLC apparatus with a Waters DeltaPak 100 × 25 mm, 100 µm C18
column, 0.1% (w/v) TFA, 0.25% acetonitrile/min unless otherwise
stated. Water (18 MΩ) was obtained from a Millipore MilliQ water
purification system, and all buffers were 0.2 µm filtered.
Glycogen (20 mg/mL), dNTPs (PCR nucleotide mix), and all
enzymes (unless otherwise stated) were purchased from Roche Diag-
nostics and used with their supplied buffers. Plasmid pUC19 was from
New England Biolabs. Deoxyadenosine [R-32P]triphosphate and thy-
midine [R-32P]triphosphate were from NEN. Calf thymus DNA
(sonicated, deproteinized) and DNase I (7500 units/mL, FPLC pure)
were from Amersham Pharmacia Biotech. HEPES was from Sigma.
9
J. AM. CHEM. SOC. VOL. 125, NO. 16, 2003 4747

A R T I C L E S
Olenyuk et al.
Figure 10. (a) Storage phosphor autoradiogram from quantitative DNase I footprint titration experiment with H-pin 3 on the radiolabeled restriction fragment
derived from plasmid pDHN2. Lane 1, intact DNA; lanes 2-22, DNase I digestion products in the presence of 50 nM, 20 nM, 10 nM, 5 nM, 2 nM, 1 nM,
500 pM, 200 pM, 100 pM, 50 pM, 20 pM, 10 pM, 5 pM, 2 pM, 1 pM, 0.5 pM, 0.2 pM, 0.1 pM, 0.05 pM, 0.02 pM, and 0.01 pM polyamide, respectively;
lane 23, DNase I standard; lane 24, A reaction. (b) Storage phosphor autoradiogram from quantitative DNase I footprint titration experiment with control
hairpin 6 on the same restriction fragment. Lane 1, intact DNA; lanes 2-18, DNase I digestion products in the presence of 50 nM, 20 nM, 10 nM, 5 nM,
2 nM, 1 nM, 500 pM, 200 pM, 100 pM, 50 pM, 20 pM, 10 pM, 5 pM, 2 pM, 1 pM, 0.5 pM, and 0.2 pM polyamides, respectively; lane 19, DNase I standard;
lane 20, A reaction. Footprints near the 5′-end of the gel are tail mismatch sequences 5′-CTGTTTCCT-3′ and 5′-TTGTTATCC-3′ from the pDHN2 plasmid.
Tris-HCl, dithiothreitol (DTT), RNase-free water (used for all DNA
manipulations), and 0.5 M EDTA were from U.S. Biochemical Corp.
Ethanol (200 proof) was from Equistar, and 2-propanol was from
Mallinckrodt. Calcium chloride, potassium chloride, and magnesium
chloride were from Fluka and purchased in Microselect quality.
Formamide was from Gibco, and premixed Tris-borate-EDTA for
gel running buffer was from Sigma. Bromophenol blue and xylene
cyanol FF were from Acros Organics. 1,1,1-Trimethylsilylethyl 4-ni-
tropyrrole-2-carboxylate 8 was prepared according to the published
procedure.16
9
4748 J. AM. CHEM. SOC. VOL. 125, NO. 16, 2003

Synthesis of H-pin Polyamides by Alkene Metathesis
A R T I C L E S
Table 1. Equilibrium Association Constantsa for H-pin Polyamides
1-5 and Control Hairpin 6 on a Restriction Fragment Derived from
Plasmid pDHN2
(2 × 50 mL) and then dried over anhydrous MgSO4 and filtered.
Removal of solvent in vacuo provided the crude product 10a, which
was purified on silica gel with a gradient of 0-5% EtOAc-hexane as
an eluent, affording 2.80 g (78%) of product. 1H NMR (CDCl3, TMS):
δ 7.12 (br s, 1H), 6.62 (d, 1H, J ) 1.8 Hz), 6.42 (br s, 1H), 5.83-6.02
(m, 1H), 5.05 (dd, 1H, J ) 17.7, 1.6 Hz), 5.03 (dd, 1H, J ) 10.2, 1.5
Hz), 4.87 (d, 2H, J ) 5.7 Hz), 4.26 (t, 2H, J ) 8.4 Hz), 1.47 (s, 9H),
1.03 (t, 2H, J ) 8.4 Hz), 0.03 (s, 9H). 13C NMR (CDCl3, TMS): δ
160.8, 152.9, 134.3, 122.4, 119.5, 118.2, 116.6, 107.8, 80.1, 62.0, 50.9,
28.4, 17.4, -1.3. The silyl ester 10a was deprotected as above, affording
1.83 g (90%) of 11a. 1H NMR (CDCl3, TMS): δ 7.25 (br s, 1H), 6.75
(s, 1H), 6.36 (br s, 1H), 5.88-6.03 (m, 1H), 5.14 (d, 1H, J ) 10.2
Hz), 5.03 (d, 1H, J ) 17.1, 0.9 Hz), 4.88 (d, 2H, J ) 5.7 Hz), 1.49 (s,
9H). 13C NMR (CDCl3, TMS): δ 165.1, 152.8, 134.0, 122.8, 119.8,
118.3, 116.9, 109.5, 80.2, 51.2, 28.4. FAB-MS calcd for C13H18N2O4
(M+ •): 266.1266. Found: 266.1267.
match
5′-ccATGTTACATtc-3′
1-bp mismatch
5′-ccATGGTACATtc-3′
specificity over
1-bp mismatch
polyamide
1
2
3
4
5
6
1.5 (( 0.18) × 1010
4.9 (( 0.25) × 109
4.3 (( 0.20) × 1010
1.1 (( 0.35) × 1010
2.1 (( 0.50) × 109
2.5 (( 0.23) × 1010
5.0 (( 0.15) × 109
2.5 (( 0.28) × 108
5.9 (( 0.22) × 108
4.6 (( 0.38) × 109
1.8 (( 0.48) × 108
1.4 (( 0.20) × 109
30
20
73
24
12
18
a The reported equilibrium association constants (in reciprocal molar
units) are the mean values obtained from three DNase I footprint titration
experiments. Assays were carried out in the presence of 10 mM Tris-HCl,
10 mM KCl, 10 mM MgCl2, and 5 mM CaCl2 at pH 7.0 and 22 °C (See
Figure 10 and Supporting Information).
4-[(tert-Butoxycarbonyl)amino]-1-but-3-enylpyrrole-2-carboxyl-
ic acid (11b). A solution of 1,1,1-trimethylsilylethyl 4-[(tert-butoxy-
carbonyl)amino]pyrrole-2-carboxylate (3.2 g, 9.8 mmol) in dry acetone
(30 mL), was treated sequentially with 1-bromo-2-butene (5.6 g, 42
mmol), anhydrous K2CO3 (5.2 g, 38 mmol), and tetrabutylammonium
iodide (300 mg, 0.812 mmol) and stirred at reflux for 24 h. The solvent
was removed in vacuo, and 10b was isolated and purified as above.
Yield: 2.5 g (58%). 1H NMR (CDCl3, TMS): δ 7.12 (br s, 1H), 6.60
(d, 1H, J ) 2.1 Hz), 6.23 (br s, 1H), 5.65-5.88 (m, 1H), 5.05 (dd, 1H,
J ) 16.8, 1.8 Hz), 5.02 (dd, 1H, J ) 10.2, 1.5 Hz), 4.29, (t, 2H, J )
8.4 Hz), 4.28 (d, 2H, J ) 8.1 Hz), 2.49 (d, 2H, J ) 14.4, 7.2 Hz), 1.49
(s, 9H), 1.06 (t, 2H, J ) 8.4 Hz), 0.06 (s, 9H). 13C NMR (CDCl3,
TMS): δ 160.7, 152.9, 134.2, 122.1, 119.0, 118.3, 116.8, 107.8, 79.6,
61.8, 48.4, 35.8, 28.3, 17.3, -1.4. The silyl ester was deprotected as
General Procedure for Deprotection of 1,1,1-Trimethylsilylethyl
Esters (10). The solution of 2.0 mmol an appropriate precursor in 2.0
mL of dry THF was placed under argon and cooled to 0 °C on an ice
bath. To this solution, TBAF (4.0 mL of 1 M in THF) was added via
a syringe with stirring. The solution was allowed to slowly warm to
ambient temperature, and the mixture was stirred overnight under Ar.
The solvent was removed under reduced pressure, and the residue was
dissolved in Et2O, washed with 0.5 M citric acid (3 × 100 mL) and
saturated NaCl solution (4 × 100 mL), dried (anhydrous MgSO4),
filtered, and evaporated in vacuo. The products were dissolved in a
minimum amount of dichloromethane and precipitated with cold
petroleum ether. The solvent was removed by filtration and the obtained
white fine powder solid was dried in vacuo.
1,1,1-Trimethylsilylethyl 4-[(tert-Butoxycarbonyl)amino]pyrrole-
2-carboxylate (9). The 1,1,1-trimethylsilylethyl 4-nitropyrrole-2-car-
boxylate 816 (7.50 g, 29.3 mmol) was dissolved in 35 mL of 6:1 EtOAc/
MeOH and treated with Pd/C (2.0 g, 10%) under Ar. The mixture was
pressurized with H2 (100 psi) and stirred at room temperature for 40 h.
The reaction mixture was filtered through a short pad of Celite and the
solvent was removed in vacuo. The mixture was dissolved in cold
diethyl ether (150 mL) and precipitated with HCl (1 M in diethyl ether).
The precipitate was collected, washed with diethyl ether, and dried in
vacuo. The crude amine salt was then suspended in 15 mL of EtOAc
and 7.5 mL of DIEA followed by 11.0 g (50.4 mmol) of di-tert-butyl
dicarbonate. The mixture was stirred at 37 °C for 12 h and then diluted
to 30 mL with EtOAc. The mixture was washed with 0.5 M citric acid
(2 × 200 mL) followed by sodium bicarbonate (3 × 100 mL) and
saturated sodium chloride (2 × 100 mL), dried over anhydrous MgSO4,
and filtered. The solvent was removed in vacuo and the residue was
dissolved in 8 mL of dichloromethane. Cold petroleum ether was added
to the product and the precipitate was stored at -20 °C for 4 h. The
off-white powder was collected by filtration, washed with the minimum
amount of cold petroleum ether, and dried in vacuo. Yield was 7.28 g
(76%) for two steps. 1H NMR (CDCl3, TMS): δ 7.84 (br s, 1H), 7.12
(br s, 1H), 6.62 (d, 1H, J ) 1.8 Hz), 6.42 (br s, 1H), 4.26 (t, 2H, J )
8.4 Hz), 1.47 (s, 9H), 1.03 (t, 2H, J ) 8.4 Hz), 0.03 (s, 9H). FAB-MS
calcd for C15H26N2O4Si (M+ •): 326.1662. Found: 326.1676.
1
above. Yield: 1.7 g (92%). H NMR (CDCl3, TMS): δ 7.23 (br s,
1H), 6.74 (s, 1H), 6.33 (br s, 1H), 5.72-5.85 (m, 1H), 5.06 (d, 1H, J
) 8.1 Hz), 5.04 (d, 1H, J ) 17.4, 1.5 Hz), 4.88 (t, 2H, J ) 6.3 Hz),
2.50 (d, 2H, J ) 14.4, 6.9 Hz), 1.50 (s, 9H). 13C NMR (CDCl3, TMS):
δ 165.1, 152.8, 134.3, 122.6, 120.1, 117.3, 116.2, 109.5, 80.1, 48.9,
35.9, 28.5. FAB-MS calcd for C14H20N2O4 (M+ •): 280.1423. Found:
280.1417.
4-[(tert-Butoxycarbonyl)amino]-1-pent-4-enylpyrrole-2-carboxyl-
ic acid (11c). A solution of 1,1,1-trimethylsilylethyl 4-[(tert-butoxy-
carbonyl)amino]-pyrrole-2-carboxylate (3.45 g, 1.06 mmol) in 30 mL
of dry acetone was treated sequentially with 1-bromo-3-pentene (5.30
g, 35.6 mmol), tetrabutylammonium iodide (300 mg, 0.812 mmol), and
anhydrous K2CO3 (5.20 g, 37.6 mmol) and stirred at reflux for 40 h.
The solvent was removed in vacuo, and 10c was isolated and purified
as above. Yield: 2.97 g (71%). 1H NMR (CDCl3, TMS): δ 7.06 (br s,
1H), 6.92 (br s, 1H), 6.60 (d, 1H, J ) 2.1 Hz), 5.61-5.73 (m, 1H),
4.91 (dd, 1H, J ) 21.0, 1.5 Hz), 4.87 (dd, 1H, J ) 10.5, 0.9 Hz), 4.20
(t, 2H, J ) 9.0 Hz), 4.14 (t, 2H, J ) 8.8 Hz), 1.94 (d, 2H, J ) 14.0,
7.2 Hz), 1.70-1.76 (m, 2H), 1.38 (s, 9H), 0.96 (t, 2H, J ) 8.4 Hz),
-0.04 (s, 9H). 13C NMR (CDCl3, TMS): δ 160.8, 153.0, 137.3, 122.3,
119.1, 118.4, 115.0, 107.9, 79.5, 61.8, 48.4, 30.6, 30.5, 28.3, 17.3, -1.4.
1
Deprotection of the silyl ester yielded 1.77 g (80%) of 11c. H NMR
(CDCl3, TMS): δ 7.24 (br s, 1H), 6.73 (s, 1H), 6.31 (br s, 1H), 5.72-
5.86 (m, 1H), 5.04 (d, 1H, J ) 16.8, 2.1 Hz), 4.99 (d, 1H, J ) 11.1
Hz), 4.25 (t, 2H, J ) 7.0 Hz), 2.04 (d, 2H, J ) 14.0, 7.5 Hz), 1.80-
1.89 (m, 2H), 1.50 (s, 9H). 13C NMR (CDCl3, TMS): δ 165.1, 152.9,
137.4, 122.6, 121.1, 116.2, 115.2, 109.4, 82.3, 48.8, 30.7, 30.6, 28.5.
FAB-MS calcd for C15H22N2O4 (M+ •): 294.1580. Found: 294.1575.
4-[(tert-Butoxycarbonyl)amino]-1-allylpyrrole-2-carboxylic acid
(11a). A solution of 1,1,1-trimethylsilylethyl 4-[(tert-butoxycarbonyl)-
amino]pyrrole-2-carboxylate 9 (3.2 g, 9.8 mmol) in dry acetone (30
mL), was treated sequentially with allyl bromide (5.00 g, 41.3 mmol),
anhydrous K2CO3 (5.20 g, 37.6 mmol), and tetra-n-butylammonium
iodide (300 mg, 0.812 mmol). The heterogeneous mixture was stirred
at reflux for 12 h. The solvent was removed in vacuo and the residue
was partitioned between Et2O (150 mL) and water (150 mL). The layers
were separated and the organic phase was washed with water (2 ×
100 mL), saturated NaHCO3 (3 × 100 mL), and saturated NaCl solution
1,1,1-Trimethylsilylethyl (4-nitro-1-but-3-enyl)pyrrole-2-carboxyl-
ate (16). To a solution of 1,1,1-trimethylsilylethyl 4-nitropyrrole-2-
carboxylate 8 (15.3 g, 59.8 mmol) in 50 mL of acetone was added
10.1 g (74.8 mmol) of 1-bromo-2-butene, followed by 20.6 g (0.15
mmol) of anhydrous potassium carbonate and 1.07 g (2.90 mmol) of
tetrabutylammonium iodide. The mixture was stirred at reflux for 15 h
and cooled to room temperature, and the precipitate was filtered out.
The solvent was removed in vacuo and the residue was partitioned
(16) Rucker, V.; Foister, S.; Melander, C.; Dervan, P. B. J. Am. Chem. Soc.
2003, 125, ASAP article.
9
J. AM. CHEM. SOC. VOL. 125, NO. 16, 2003 4749

A R T I C L E S
Olenyuk et al.
between diethyl ether (300 mL) and water (200 mL). The organic phase
was separated, washed with water (2 × 200 mL) and saturated NaCl
(200 mL), dried over anhydrous MgSO4, filtered, and evaporated. The
product was purified by column chromatography on silica with hexane/
mmol). The solution was stirred overnight to ensure the complete
formation of the activated ester. The solution was then filtered into a
stirring mixture of 100 g of water and 50 g of ice. The precipitate was
collected, washed with a minimum amount of cold water, and dried in
vacuo. Yield: 79%. 1H NMR (DMSO-d6, TMS): δ 9.42 (s, 1H), 8.12
(d, J ) 8.1 Hz, 1H), 7.75 (d, J ) 8.1 Hz, 1H), 7.59 (t, J ) 8.1 Hz,
1H), 7.54 (s, 2H), 7.51 (t, J ) 8.1 Hz, 1H), 7.50 (s, 1H), 7.19 (s, 1H),
1
EtOAc (10:1) as the eluent. Yield: 12.7 g (68%). H NMR (CDCl3,
TMS): δ 7.58 (d, J ) 2.1 Hz, 1H), 7.37 (d, J ) 2.1 Hz, 1H), 5.73 (m,
1H), 5.03 (m, 1H), 5.03 (d, J ) 17.9 Hz, 1H), 4.40 (t, J ) 6.6 Hz,
2H), 4.34 (t, J ) 6.9 Hz, 2H), 2.52 (m, 2H), 1.09 (t, J ) 6.9 Hz, 2H),
0.04 (s, 9H). 13C NMR (CDCl3, TMS): δ 160.2, 135.3, 133.2, 126.9,
122.6, 118.6, 113.0, 63.5, 50.0, 35.5, 17.7, -1.2. FAB-MS calcd for
C14H23N2O4Si (MH+): 311.1427. Found: 311.1422.
4.15 (t, J ) 7.2 Hz, 2H), 1.57 (m, 2H), 1.44 (s, 9H), 1.18 (m, 2H). 13
C
NMR (DMSO-d6, TMS): δ 155.5, 152.6, 142.6, 129.0, 127.0, 125.0,
124.3, 123.2, 119.7, 111.3, 110.1, 109.1, 78.9, 48.4, 34.5, 30.8, 28.1.
FAB-MS calcd for C38H45N4O8 (MH+): 769.3421. Found: 769.3419.
Solid-Phase Synthesis. Boc-Py-Py-â-Ala-PAM resin 13 was pre-
pared from Boc-â-Ala-PAM resin (0.88 mmol/g) according to the
standard solid-phase synthetic protocol as described by Baird and
Dervan.7a A mixture of acids 11a-c (150 mg each) was simultaneously
activated with DCC (220 mg, 1.07 mmol) and HOBt (300 mg, 2.22
mmol) in 2 mL of DMF. The mixture was stirred at room temperature
for 1 h. The precipitate was filtered off and the solution containing a
mixture of activated esters 12a-c was mixed with 1 mL of dry DIEA.
Dried Boc-Py-Py-â-Ala-PAM resin (100 mg) was deblocked (80% TFA
in DCM, 1 × 30 s + 1 × 20 min) and washed (DCM), and activated
esters 12a-c were added. After 4 h at room temperature, the reaction
vessel was drained and the resin was rigorously washed with DMF,
DCM, methanol, and diethyl ether and then dried overnight in vacuo.
Metathesis. The entire batch of dried resin from above step was
placed in a 100 mL round-bottom flask, and 3 mL of dry DCM was
added under Ar. Ruthenium catalyst (20 mg) was added to the
heterogeneous mixture, which was stirred at reflux for 24 h at 40 °C
under an Ar stream to remove the formed ethylene. A fresh portion of
catalyst (10 mg) was then added and the stirring continued for another
24 h. The resin was collected and washed with 10% dppp in DCM,
then with DCM, and finally with methanol. The extent of cross-linking
was monitored by reverse-phase HPLC with a 0.1% TFA gradient in
acetonitrile (0-60% in 35 min). The Boc group was deprotected with
25% TFA in DCM (2 × 10 min) and the synthesis was continued as
described previously.7a Both activated esters of BocPyOBt and ImOBt
were coupled as above. The mixture was cleaved from the solid support
with Dp at 37 °C for 12 h. The excess of Dp was removed in vacuo
and the residue was dissolved in absolute ethanol and passed through
a Waters Sep-Pak. The EtOH solution was then treated with NH4HCO2
(150 mg) and Pd(OAc)2 (25 mg) and stirred for 24 h at room
temperature. After reduction was completed, the deposited Pd was
filtered off and the solvent was removed in vacuo. The residue was
dissolved in a mixture of 20% MeOH + 0.1% TFA/H2O (v/v) and
purified by reverse-phase HPLC on a Varian semipreparative C18
column. Each separated product was analyzed twice on analytical HPLC
on a 100 Å C18 analytical column with gradients of 0.1% TFA-
acetonitrile of 0-60% in 35 min and 0-55% in 60 min.
1,6-Di-[4-nitro-2-[[(1,1,1-trimethylsilylethyl)oxy]carbonyl]-N-pyr-
rolyl]-3-hexene (17). [1,1,1-Trimethylsilylethyl (4-nitro-1-but-3-enyl)]-
pyrrole-2-carboxylate 16 (12.3 g, 39.6 mmol) was dissolved in dry
dichloromethane (90 mL) and the solution was purged with Ar for 15
min. Ru catalyst (0.760 g, 1.98 mmol, 2.5 mol %) was added to the
solution and stirred at reflux under Ar for 24 h. Another 0.76 g of
catalyst was added, and the stirring continued for another 12 h until
reaction was complete as determined by TLC. The solvent was removed
in vacuo and the residue was chromatographed on silica with hexane/
EtOAc ) 5:1 as an eluent. Two isomers (E and Z) were collected and
used in the next step without further separation. Yield for E + Z
1
isomers: 8.71 g (74%). H NMR (CDCl3, TMS): δ 7.65 (d, J ) 2.1
Hz, 1H), 7.54 (d, J ) 2.1 Hz, 1H), 7.40 (s, 2H), 5.52 (m, 1H), 5.39
(m, 1H), 4.34 (m, 8H), 2.48 (m, 4H), 1.09 (t, J ) 6.9 Hz, 9H), 0.07 (s,
18H). 13C NMR (CDCl3, TMS): δ 160.2, 129.0, 128.0, 126.9, 126.7,
122.6, 113.0, 63.7, 50.0, 34.5, 29.3, 17.7, -1.2. FAB-MS calcd for
C26H40N4O8Si2 (M+ •): 592.2384. Found: 592.2372.
1,6-Di-[4-[(tert-butoxycarbonyl)amino]-2-[[(1,1,1-trimethylsilyl-
ethyl)oxy]carbonyl]-N-pyrrolyl]hexane (18). A solution of 17 (8.60
g, 14.5 mmol) in 2:1 EtOAc/EtOH (300 mL) was treated with 3.00 g
of 10% Pd/C and stirred under H2 (400 psi) at room temperature for
15 h. After reaction was complete, the solution was filtered through a
pad of Celite and concentrated in vacuo. The product was precipitated
with HCl (40 mL, 1 M in diethyl ether), collected by filtration under
N2, and dried in vacuo to afford crude amine hydrochloride in 89%
yield. The amine salt was suspended in dioxane (150 mL) and 10%
aqueous potassium carbonate (150 mL) and treated with Boc2O (6.0 g,
28 mmol) in dioxane (30 mL) over a period of 20 min. The reaction
was further stirred for 6 h, then cooled to 0 °C, and the precipitated
product was collected and dried in vacuo. Yield: 65%. 1H NMR
(CDCl3, TMS): δ 7.08 (br s, 1H), 6.61 (d, J ) 2.1 Hz, 1H), 6.29 (br
s, 1H), 4.27 (t, J ) 8.1 Hz, 2H), 4.20 (t, J ) 6.6 Hz, 2H), 1.68 (m,
2H), 1.52 (s, 9H), 1.27 (m, 2H), 1.05 (t, J ) 8.1 Hz, 2H), 0.05 (s, 9H).
13C NMR (CDCl3, TMS): δ 161.1, 153.2, 122.2, 119.5, 118.7, 108.0,
62.2, 49.2, 31.7, 28.7, 27.7, 26.6, 17.7, -1.1. FAB-MS calcd for
C36H62N4O8Si2 (M+ •): 734.4106. Found: 734.4112.
1,6-Di-[4-[(tert-butoxycarbonyl)amino]-2-carboxy-N-pyrrolyl]-
hexane (19). Trimethylsilyl ester 18 (5.0 g, 6.8 mmol) was dissolved
in 20 mL of dry THF, placed under Ar, cooled to 0 °C, and treated
with 20 mL of 1 M TBAF in THF. The solution was allowed to warm
to room temperature and stirred for 12 h. The solvent was removed
under reduced pressure, and the compound was dissolved in 100 mL
of DCM. The organic layer was washed with 1 M citric acid (3 × 100
mL), saturated sodium bicarbonate (3 × 100 mL), and saturated NaCl
(2 × 100 mL), dried over anhydrous MgSO4, filtered, and concentrated
Cross-Linking. Boc-PyPy-â-Ala-PAM resin 13 was prepared from
Boc-â-Ala-PAM resin (100 mg, 0.88 mmol/g) as described above. After
deblocking (80% TFA in DCM, 1 × 30 s + 1 × 20 min) and washing
(DCM), activated ester 20 (68 mg, 0.088 mmol) was added in 0.5 mL
of DMF and 0.25 mL of DIEA. The extent of cross-linking was
monitored by reverse-phase HPLC. The resin was shaken at 37 °C for
12 h, drained, and washed with DMF. Further coupling steps, cleavage
from the solid support, and purification by HPLC were carried out as
previously desribed.7a Yield of 3: 48 mg, 30% recovery.
1
in vacuo to provide the product as a tan solid. Yield: 90%. H NMR
(ImPyPyPyPy-â-Dp)2(CH2)4 (1). 1H NMR (CDCl3, TMS): δ 10.45
(s, 2H, arom NH), 9.98 (s, 2H, arom NH), 9.97 (s, 2H, arom NH),
9.90 (s, 2H, arom NH), 9.4 (br s, 2H, CF3COOH), 8.05 (m, 4H, aliph
NH), 7.40 (s, 2H, CH), 7.29 (m, 2H, CH), 7.27 (m, 2H, CH), 7.15 (m,
4H, CH), 7.04 (m, 4H, CH), 7.00 (m, 4H, CH), 6.86 (m, 2H, CH),
4.31 (br m, 4H, linker CH2), 3.97 (s, 6H, NCH3), 3.83 (s, 6H, NCH3),
3.82 (s, 6H, NCH3), 3.78 (s, 6H, NCH3), 3.36 (m, 4H, CH2), 3.08 (m,
4H, CH2), 2.98 (m, 4H, CH2), 2.72 (d, J ) 4.8 Hz, 4H, N(CH3)2), 2.33
(t, J ) 7.2 Hz, 4H, aliph CH2), 1.70 (m, 8H, aliph CH2 + linker CH2).
(DMSO-d6, TMS): δ 12.02 (br s, 1H), 9.03 (s, 1H), 7.05 (s, 1H), 6.6
(s, 1H), 4.15 (t, J ) 7.2 Hz, 2H), 1.57 (m, 2H), 1.41 (s, 9H), 1.18 (m,
2H). 13C NMR (DMSO-d6, TMS): δ 161.4, 152.5, 122.7, 118.6, 117.6,
107.7, 78.3, 47.8, 31.2, 28.2, 25.6. FAB-MS calcd for C26H39N4O8
(M+ •): 534.2676. Found: 534.2679.
1,6-Di-[4-[(tert-butoxycarbonyl)amino]-2-[[(1-hydroxybenzotria-
zolyl)oxy]carbonyl]-N-pyrrolyl]hexane (20). A solution of 19 (2.94
g, 5.50 mmol) in 15 mL of DMF was sequentially treated with
1-hydroxybenzotriazole (3.00 g, 22.2 mmol) and DCC (2.27 g, 11.0
9
4750 J. AM. CHEM. SOC. VOL. 125, NO. 16, 2003

Synthesis of H-pin Polyamides by Alkene Metathesis
A R T I C L E S
MALDI-TOF-MS (monoisotopic) calcd for C76H97N26O12 (MH+):
1565.8. Found: 1565.9.
was precipitated with 2-propanol (1.5 volumes). The pellet was washed
with 75% ethanol, lyophilized to dryness, and then resuspended in 10
mL of RNase-free H2O. Chemical sequencing reactions were performed
according to published protocols.19
(ImPyPyPyPy-â-Dp)2(CH2)6 (3). 1H NMR (CDCl3, TMS): δ 10.49
(s, 2H, arom NH), 9.97 (s, 2H, arom NH), 9.94 (s, 2H, arom NH),
9.89 (s, 2H, arom NH), 9.45 (br s, 2H, CF3COOH), 8.05 (m, 4H, aliph
NH), 7.41 (s, 2H, CH), 7.27 (m, 2H, CH), 7.27 (m, 2H, CH), 7.16 (m,
4H, CH), 7.06 (m, 4H, CH), 7.03 (m, 4H, CH), 6.87 (m, 2H, CH),
4.27 (br m, 4H, linker CH2), 3.98 (s, 6H, NCH3), 3.92 (s, 6H, NCH3),
3.83 (s, 6H, NCH3), 3.78 (s, 6H, NCH3), 3.36 (m, 4H, CH2), 3.09 (m,
4H, CH2), 2.99 (m, 4H, CH2), 2.72 (d, J ) 4.8 Hz, 4H, N(CH3)2), 2.33
(t, J ) 7.3 Hz, 4H, aliph CH2), 1.70 (m, 8H, aliph CH2 + linker CH2),
1.24 (m, 4H, linker CH2). MALDI-TOF-MS (monoisotopic) calcd for
C76H97N26O12 (MH+): 1593.8. Found: 1593.7.
Quantitative DNase I Footprint Titrations. All reactions were
carried out in a volume of 400 µL according to the published
procedure.15 We note explicitly that no carrier DNA was used in these
reactions until after DNase I cleavage. The concentration of radiolabeled
DNA was calculated to remain below the concentration of polyamide
in all but the most diluted reaction. Polyamide stock solution (or water
for reference and intact lanes) was added to an assay buffer where the
final concentrations were 10 mM Tris‚HCl buffer (pH 7.0), 10 mM
KCl, 10 mM MgCl2, 5 mM CaCl2, and 8 kcpm of 3′-radiolabeled DNA.
The solutions were allowed to equilibrate for 18-24 h at 22 °C.
Cleavage was initiated by the addition of 10 µL of a DNase I solution
(diluted with 1 mM dithiothreitol to 1.5 units/mL) and allowed to
proceed for 7 min at 22 °C. The reactions were stopped by adding 50
µL of a solution containing 2.25 M NaCl, 150 mM EDTA, 0.6 mg/mL
glycogen, and 30 µM base-pair calf thymus DNA and then precipitated
with EtOH (2.1 volumes) and centrifuged at 14 krpm for 25 min. The
pellets were washed with 75% ethanol, resuspended in 15 µL of RNase-
free H2O, lyophilized to dryness, and then resuspended in 10 µL of
100 mM Tris-borate-EDTA/80% formamide loading buffer, denatured
at 90 °C for 10 min, and cooled in ice. A 5 µL sample (ca. 4.5 kcpm)
was loaded onto a prerun (30 min at 2000 V) 8% denaturing
polyacrylamide gel (5% cross-link, 7 M urea) at 2000 V for 1.5 h. The
gels were dried in vacuo at 80 °C and then exposed to a storage
phosphor screen (Molecular Dynamics) at 22 °C in the dark for 12-
24 h. Storage phosphor screens were then scanned with a Storm 820
phosphorimager or Typhoon 8900 variable-mode imager (Molecular
Dynamics). Equilibrium association constants were determined as
previously described.14,15
(ImPyPyPyPy-â-Dp)2(CH2)5 (2). MALDI-TOF-MS (monoisotopic)
calcd for C76H97N26O12 (MH+): 1579.8. Found: 1579.9.
(ImPyPyPyPy-â-Dp)2(CH2)7 (4). MALDI-TOF-MS (monoisotopic)
calcd for C76H97N26O12 (MH+): 1607.8. Found: 1607.9.
(ImPyPyPyPy-â-Dp)2(CH2)8 (5). MALDI-TOF-MS (monoisotopic)
calcd for C76H97N26O12 (MH+): 1621.8. Found: 1621.3.
Preparation of 3′- and 5′-End-Labeled DNA Restriction Frag-
ments. All cloning was carried out according to published procedures.17
Plasmid pDHN2 was constructed by inserting the hybridized inserts
5′-GAT CCT TAA GTT CGT GGG CCA TGG TAC ATT CGT GGC
CAT GGT CCA TTC GTG GGC CAT GTT ACA TTC GA-3′ and
5′-AGC TTC GAA TGT AAC ATG GCC CAC GAA TGG ACC ATG
GCC ACG AAT GTA CCA TGG CCC ACG AAC TTA AG-3′ into
the BamHI/HindIII polycloning site in pUC19.18 The insert was obtained
by annealing complementary synthetic oligonucleotides and was then
ligated to the large BamHI/HindIII restriction fragment of pUC19 with
T4 DNA ligase. The ligated plasmid was then used to transform
Escherichia coli XL-1 Blue Supercompetent cells. Colonies were
selected for R-complementation on 25-mL Luria-Bertani agar plates
containing 50 mg/mL ampicillin and treated with X-Gal and IPTG
solutions and grown overnight at 37 °C. Well-defined white colonies
were transferred into 100 mL of Luria-Bertani medium containing 50
mg/mL ampicillin. Cells were harvested after overnight growth at 37
°C. Medium-scale plasmid purification was performed with Qiagen
purification kits. The presence of the desired insert was determined by
dideoxy sequencing. DNA concentration was determined at 260 nm
by use of the relation 1 OD unit ) 50 mg/mL duplex DNA. Plasmid
pDHN2 was simultaneously cut with EcoRI and PVuII and then
radiolabeled by 3′ fill-in with 1 mCi of [R-32P]dATP, 1 mCi of [R-32P]-
dTTP, and the Klenow fragment of DNA polymerase II at 37 °C for
25 min. The product mixture was purified on a 7% nondenaturing
preparative polyacrylamide gel (5% cross-link), and the desired
fragment was isolated after visualization by autoradiography. The DNA
Acknowledgment. We thank the National Institutes of Health
(GM-27681) for research support and a postdoctoral fellowship
to B.O. (F32 GM-19788). C.J. is grateful to California Institute
of Technology for a Summer Undergraduate Research Fellow-
ship.
Supporting Information Available: Two figures showing
quantitative DNase I footprint titration experiments for com-
pounds 1, 2, 4, and 5 in Table 1 and equilibrium binding
isotherms for 3 and 6 (PDF). This material is available free of
charge via the Internet at http://pubs.acs.org.
JA0213221
(17) Sambrook, J.; Fritsch, E. F.; Maniatis, T. Molecular Cloning; Cold Spring
Harbor Laboratory: Cold Spring Harbor, NY, 1989.
(18) Nguyen, D. H. Ph.D. Dissertation, California Institute of Technology,
Pasadena, CA, 2002.
(19) (a) Iverson, B. L.; Dervan, P. B. Nucleic Acids Res. 1987, 15, 7823. (b)
Maxam, A. M.; Gilbert, W. S. Methods Enzymol. 1980, 65, 499.
9
J. AM. CHEM. SOC. VOL. 125, NO. 16, 2003 4751