Cyclic semipeptoids: Peptoid-organic hybrid macrocycles

Article abstract of DOI:10.1021/ol052867d

Cyclic semipeptoids 1 and 2 represent constrained, secondary structure mimics where the R¹ and R² side chains correspond to those of amino acids. Solid-phase syntheses and conformational analyses of these compounds are described.

Full text of DOI:10.1021/ol052867d

ORGANIC
LETTERS
2006
Vol. 8, No. 7
1259-1262
Cyclic Semipeptoids: Peptoid−Organic
Hybrid Macrocycles
Ernest Nnanabu and Kevin Burgess*
Texas A&M UniVersity, Chemistry Department, P.O. Box 30012,
College Station, Texas 77842
burgess@tamu.edu
Received November 26, 2005
ABSTRACT
Cyclic semipeptoids 1 and 2 represent constrained, secondary structure mimics where the R1 and R2 side chains correspond to those of
amino acids. Solid-phase syntheses and conformational analyses of these compounds are described.
Poly-N-alkylated glycine chains, peptoids,1 are easily con-
structed via solid-phase syntheses.2 A vast number of amines
are available to use as starting materials for these compound
types, so peptoids provide rapid access into diverse structures.
However, they are not rigid. The next step in the evolution
of this field is the production of conformationally constrained
peptoid analogues, to decrease the entropy loss on binding
and improve bioavailability. There have been some moves
in the literature toward this goal; Golebiowski and co-
workers3 and Liskamp et al.,4 for example, have combined
a dipeptide moiety with N-alkylation to produce the â-turn
analogues A and B, respectively. This Letter describes cyclic
peptoid/amino acid hybrids, compounds we call “semipep-
toids”, 1 and 2.
the context of 1 and 2, this means that the syntheses should
accommodate protected, functionalized amines. Conse-
quently, it was important to find conditions that were mild
and generally applicable to a diverse amine set. In the event,
two approaches were developed, as described below.
Solid-phase syntheses of peptoids often have been per-
formed by heating supported bromoacyl derivatives with
excess amine in DMF for approximately 80 min at 35 °C.
Preliminary experiments by us using this approach gave
variable results. Kodadek had previously reported that
microwave heating in a domestic instrument gave shorter
reaction times.6 We were reluctant to use domestic instru-
ments that have no temperature control and give nonuniform
heating in the reaction cavity because the data tend to be
hard to reproduce. Nevertheless, the prospect of dramatically
reducing the reaction times was appealing.
(1) Simon, R. J.; Kania, R. S.; Zuckermann, R. N.; Huebner, V. D.;
Jewell, D. A.; Banville, S.; Ng, S.; Wang, L.; Rosenberg, S.; Marlowe, C.
K.; Spellmeyer, D. C.; Tan, R.; Frankel, A. D.; Santi, D. V.; Cohen, F. E.;
Bartlett, P. A. Proc. Natl. Acad. Sci. 1992, 89, 9367.
(2) Patch, J. A.; Kirshenbaum, K.; Seurynck, S. L.; Zuckermann, R. N.;
Barron, A. E. In Pseudo-Peptides in Drug DiscoVery; Nielsen, P. E., Ed.;
Wiley-VCH: Weinheim, 2004; p 1-31.
(3) Golebiowski, A.; Klopfenstein, S. R.; Shao, X.; Chen, J. J.; Colson,
A.-O.; Grieb, A. L.; Russell, A. F. Org. Lett. 2000, 2, 2615.
(4) Wels, B.; Kruijtzer, J. A. W.; Liskamp, R. M. J. Org. Lett. 2002, 4,
2173.
Syntheses of 1 and 2 were both based on linear precursors.
Syntheses of these would be relatively straightforward for
compounds that have only simple N-alkyl and aryl substit-
uents. However, we are convinced that the best peptidomi-
metic designs for inducing diverse biological activities must
allow facile incorporation of functionalized amino acids.5 In
(5) Burgess, K. Acc. Chem. Res. 2001, 34, 826-35.
(6) Olivos, H. J.; Alluri, P. G.; Reddy, M. M.; Salony, D.; Kodadek, T.
Org. Lett. 2002, 4, 4057.
10.1021/ol052867d CCC: $33.50
© 2006 American Chemical Society
Published on Web 03/07/2006

Scheme 1. Microwave-Accelerated Syntheses of
Peptide-Peptoid Hybridsa
Scheme 2. Alternative Approach to Microwave-Accelerated
Syntheses of Peptide-Peptoid Hybrids
In other cases, it was more convenient to prepare the
amines with a nosyl protecting group9 and to perform the
alkylation step under basic conditions. This was the route
used where the side-chain functionalized amines s and t were
involved (Scheme 2). Finally, the “monomer approach”10 was
used to install the side chains of a and g; i.e., FMOC-
sarcosine (N-methylglycine) and -glycine were coupled to
incorporate these units directly.
a The lower-case codes indicated correspond to one-letter codes
for the closest amino acid side chains. For instance, a methyl side
chain a resembles Ala or A and side chain k resembles Lys or K.
e′, k′, and r′ indicate masked side chains that will give e, k, and r
side chains on deprotection. Compounds involving a and g
fragments were introduced via the FMOC approach, and the rest
were made as indicated in the reaction above.
Scheme 3. Syntheses of Cyclic Semipeptoids 1 via SN2
Reactionsa
On the basis of the considerations outlined above, it
seemed attractive to investigate microwave-mediated ac-
celeration of these coupling steps in a scientific microwave
reactor designed to maintain constant reaction temperatures
(CEM: Discover). Our initial exploratory work focused on
longer reaction times and higher temperatures than were
necessary. However, it was quickly demonstrated that solid-
phase syntheses of peptoids on Rink-linker-functionalized7
polystyrene proceeded rapidly and efficiently if the acylation
and nucleophilic displacement steps were microwaved at 50
°C for 1 min (see Supporting Information). When this work
was nearly complete, Blackwell et al. also reported solid-
phase peptoid syntheses under microwave conditions.8 They
concluded that reactive, unhindered amines combine with
bromoacyl groups on polystyrene beads in approximately 1
min at room temperature, even without microwave excitation.
Our data are consistent with those findings because the
microwave energy input is relatively mild and it is applied
for a short period. However, sometimes we found that
microwaves were necessary.
a R1′ and R2′ indicate masked side chains.
The linear precursors 3 were converted to the cyclic
semipeptoids 1 via the approach shown in Scheme 3. This
featured a simple acylation of the N-terminus with 2-bro-
momethyl benzoyl chloride, deprotection of the 4-methoxy
trityl (Mmt) protecting group, and then microwave-assisted
SN2 ring closure. Finally, the compounds were cleaved into
solution, with concomitant removal of any protecting groups,
by rupture of the Rink linker.
Scheme 1 shows the use of our microwave conditions to
introduce a Cys and two N-alkylated Gly residues. This
approach was applied using the amines indicated in Schemes
1 and 2 in various combinations (see below).
(7) Rink, H. Tetrahedron Lett. 1987, 28, 3787.
(8) Gorske, B. C.; Jewell, S. A.; Guerard, E. J.; Blackwell, H. E. Org.
Lett. 2005, 7, 1521.
(9) Kan, T.; Fukuyama, T. Chem. Commun. 2004, 353.
(10) Zuckermann, R. N.; Kerr, J. M.; Kent, S. B. H.; Moos, W. H. J.
Am. Chem. Soc. 1992, 114, 10646.
1260
Org. Lett., Vol. 8, No. 7, 2006

Scheme 4. Syntheses of Cyclic Semipeptoids 2 via SNAr
Table 1. Purity and Yield Data for the Cyclic Semipeptoids 1
Reactionsa
a R1′ and R2′ indicate masked side chains.
a 254 nm.
became sharper at elevated temperatures, indicative of slow
exchange on the NMR time scale. Moreover, their limited
solubilities in most solvents prevented extensive variable-
temperature NMR experiments. These characteristics pre-
cluded detailed analyses of the molecular conformations by
2D NMR. Further, interpretation of the CD spectra of these
molecules cannot rely on the literature precedent that exists
for peptides. Nevertheless, despite all these restrictions, it
was possible to form some tenuous conclusions about the
conformational biases of these molecules.
Table 1 summarizes crude purities and yields for the
compounds made via the approach shown in Scheme 3.
Measurements of crude purities are somewhat dependent
upon the detection method (here UV and evaporative light
scattering {ELS}). Many of the compounds made via this
procedure would have passed an 85% purity cutoff to be
entered into a high-throughput preliminary screen.
Scheme 4 shows an alternative approach to cyclizing linear
peptide-peptoid hybrids. There are two key differences
between this and the SN2 strategy featured in Scheme 3. The
former leads to SNAr ring closures from homo-Cys, rather
than Cys, as the amino acid foundation. Homo-Cys was used
so that the final ring-closed products could support a C10
template arrangement in the template part of the molecule;
we have hypothesized that this frequently leads to â-turn
conformations in the encapsulated dipeptide or, in this case,
dipeptoid fragment.5 The Cys derivatives 1 can form this
C10 arrangement in the template, but homologation to homo-
Cys is required for the 2-fluoro-5-nitrobenzoic acid deriva-
tives to do so.
Compounds 1nk and 2ff were selected as illustrative
members of the series. Their conformational states were
Table 2. Purity and Yield Data for the Cyclic Semipeptoids 2
Yield and purity data for the semipeptoids 2 formed via
Scheme 4 are shown in Table 2. Overall, both parameters
tend to be higher for these products than for systems 1.
Full solution-state conformational analyses of the title
cyclic semipeptoids were not possible via routine methods,
for several reasons. First, these molecules lack some pivotal
spectroscopic markers. For instance, the peptoid part has no
NH/CRH couplings to report on dihedral angles and gives
no NH/CRH or NH/NH ROESY cross-peaks. In any case,
a 254 nm.
1
the H NMR spectra of these compounds were broad but
Org. Lett., Vol. 8, No. 7, 2006
1261

simulated in a medium of dielectric constant 45 (correspond-
ing to DMSO) using the quenched molecular dynamics
(QMD)11,12 technique as described by us in several other
publications.13,14 Temperature coefficients were measured for
the NH protons in DMSO-D6; steeper slopes than -3.0 ppm
are assumed to be indicative of solvent shielding and/or
H-bonding.15 Finally, CD spectra were recorded for these
molecules dissolved in an aqueous methanol medium of
about the same dielectric as DMSO.
Figure 1a shows the most favored conformer for 1nk
in the molecular simulation experiments. The simulated
CdO‚‚‚HN in the template part of this molecule was 4.7 Å,
i.e., somewhat longer than in ideal â-turns. Additionally, the
temperature coefficient for this proton was -4.7 ppb K-1,
providing no evidence for solvent shielding or H-bonding.
Nevertheless, the simulated structure indicates that turnlike
conformations are accessible. The CD spectrum of this
compound (Figure 1b) is similar to those expected for a type
2 â-turn in a peptide, but contributions from the absorption
of the aromatic groups in compound 2ff cannot be ignored.16-18
The distance between the two â-atoms in the peptoid side
chains is very similar to that which we measured in ideal
type 1 â-turns (5.3 vs 5.2 Å), so the cyclic semipeptoid
template at least presents the side chains at the correct spacial
separation to mimic turns.
The preferred simulated conformation of 2ff (Figure 1c)
gave a much closer CdO‚‚‚HN contact (2.0 Å), and this was
consistent with the temperature coefficient experimentally
observed for this compound (-1.2 ppb K-1). The CD
spectrum of this compound was reminiscent of a type 1
â-turn. The two â-atoms in the peptoid side chains were
separated by 5.36 Å in the preferred simulated conformer,
about that observed in â-turns.
The research described in this paper illustrates a route19
to combine the enormous diversity of amine-derived side
chains in peptoids with amino acids in convenient solid-phase
syntheses. Two routes are demonstrated to cyclize linear
semipeptoids with an array of side chains that can resemble,
in terms of structure and diversity, those found in the protein
amino acids. These constrained compounds might be ex-
pected to have, on aggregate, superior properties with respect
to loss of entropy on binding and bioavailability, relative to
conformationally flexible peptides and peptoids. Further, the
Figure 1. (a) Simulated favored conformation of 1nk. (b) CD
spectra of 1nk and 2ff at 0.1 mg/mL in 35% MeOH(aq) with 1%
NaHCO3. (c) Simulated favored conformation of 2ff.
limited data collected on their conformational analyses
indicate that molecules of this type may be designed to mimic
â-turns.
(11) O’Connor, S. D.; Smith, P. E.; Al-Obeidi, F.; Pettitt, B. M. J. Med.
Chem. 1992, 35, 2870.
Acknowledgment. Financial support for this project was
provided by the NIH (CA82642) and the Robert A. Welch
Foundation. TAMU/LBMS Lab headed by Dr S. Tichy
provided mass spectrometric support, and the Lab for
Molecular Simulations (Dr L. Thompson) supported our
simulations work. NMR instrumentation at TAMU was
supported by a grant from the National Science Foundation
(DBI-9970232) and the TAMU System.
(12) Pettitt, B. M.; Matsunaga, T.; Al-Obeidi, F.; Gehrig, C.; Hruby, V.
J.; Karplus, M. Biophys. J. Biophys. Soc. 1991, 60, 1540.
(13) Moye-Sherman, D.; Jin, S.; Ham, I.; Lim, D.; Scholtz, J. M.;
Burgess, K. J. Am. Chem. Soc. 1998, 120, 9435.
(14) Feng, Y.; Wang, Z.; Jin, S.; Burgess, K. Chem.-Eur. J. 1999, 5,
3273.
(15) Ohnishi, M.; Urry, D. W. Biochem. Biophys. Res. Commun. 1969,
36, 194.
(16) Manning, M. C.; Illangasekare, M.; Woody, R. W. Biophys. Chem.
1988, 31, 77.
(17) Bush, C. A.; Sarkar, S. K.; Kopple, K. D. Biochemistry 1978, 17,
4951.
Supporting Information Available: Details of optimiza-
tion studies, synthetic procedures, and protocol/data for
conformational analyses. This material is available free of
charge via the Internet at http://pubs.acs.org.
(18) Perczel, A.; Hollosi, M. In Circular Dichroism and the Conforma-
tional Analysis of Biomolecules; Fasman, G. D., Ed.; Plenum Press: New
York and London, 1996; p 362-364.
(19) Peptoid and peptides have been mixed in the same molecule before,
in a much larger system: Shankaramma, S. C.; Moehle, K.; James, S.;
Vrijbloed, J. W.; Obrecht, D.; Robinson, J. A. Chem. Commun. 2003, 1842.
OL052867D
1262
Org. Lett., Vol. 8, No. 7, 2006