Synthesis and screening of stereochemically diverse combinatorial libraries of peptide tertiary amides

Article abstract of DOI:10.1016/j.chembiol.2013.01.013

Large combinatorial libraries of N-substituted peptides would be an attractive source of protein ligands, because these compounds are known to be conformationally constrained, whereas standard peptides or peptoids are conformationally mobile. Here, we report an efficient submonomer solid-phase synthetic route to these compounds and demonstrate that it can be used to create high quality libraries. A model screening experiment and analysis of the hits indicates that the rigidity afforded by the stereocenters is critical for high affinity binding.

Full text of DOI:10.1016/j.chembiol.2013.01.013

Chemistry & Biology
Article
Synthesis and Screening
of Stereochemically Diverse Combinatorial Libraries
of Peptide Tertiary Amides
Yu Gao1 and Thomas Kodadek1,
*
1Department of Chemistry, Scripps Florida, Jupiter, FL 33458, USA
*Correspondence: kodadek@scripps.edu
http://dx.doi.org/10.1016/j.chembiol.2013.01.013
SUMMARY
hundreds of suitable primary amines are available commercially
or synthesized readily, large one-bead one-compound (OBOC)
Large combinatorial libraries of N-substituted libraries of peptoids can be made easily using the split and
pool strategy (Lam et al., 1991). These libraries can be screened
peptides would be an attractive source of protein
on bead against labeled, soluble proteins (Alluri et al., 2003; Lim
et al., 2007; Wrenn et al., 2007; Xiao et al., 2007) or against cells
(Aina et al., 2005; Lee et al., 2010a; Udugamasooriya et al., 2008)
ligands, because these compounds are known to
be conformationally constrained, whereas standard
peptides or peptoids are conformationally mobile.
Here, we report an efficient submonomer solid-
phase synthetic route to these compounds and
demonstrate that it can be used to create high quality
libraries. A model screening experiment and analysis
to identify novel protein ligands. However, these ‘‘hits’’ do not
generally display high affinity, with rare exceptions (Zuckermann
et al., 1994). This may be due, in part, to the ‘‘floppiness’’ of
peptoids. N-substitution and the lack of a chiral center in
peptoids results in the loss of the preference for the trans amide
of the hits indicates that the rigidity afforded by the
stereocenters is critical for high affinity binding.
bond conformation characteristic of simple peptides. Moreover,
there is little preference for particular conformations around the
carbonyl-Ca or Ca-N bonds. Assuming that the peptoid binds
its target protein in a defined conformation, this means a signifi-
cant entropic penalty must be paid to form a complex, limiting
affinity. Peptoid floppiness may also render hit optimization
INTRODUCTION
Natural products constructed entirely or partially of N-methyl difficult.
peptide building blocks display many interesting activities.
Therefore, we and others have been interested in creating
Some, such as cyclosporine, are in use clinically. N-methylation peptoid (Chongsiriwatana et al., 2008; Gorske and Blackwell,
has the effect of both stabilizing the peptide backbone toward 2006; Huang et al., 2006; Kwon and Kodadek, 2008; Lee et al.,
proteolytic degradation and, by removing highly hydrated N-H 2010b, 2011; Shah et al., 2008; Shin et al., 2007; Stringer et al.,
bonds, generally improving the cell permeability and bioavail- 2011), or peptoid-like (Aquino et al., 2012; Sarma et al., 2011),
ability of these compounds. N-methylation can also strongly molecules with greater conformational constraints.
stabilize one or a few particular conformations of the backbone,
particularly in the context of cyclic structures.
Previous studies of N-methylated peptides have indicated
that they are far more constrained conformationally than
N-methylated peptides are a subfamily of peptide tertiary either peptoids or simple peptides. For example, the CD spec-
amides (PTAs) (Figure 1) in which the main-chain nitrogen is alky- trum of poly-N-methyl-L-alanine indicated a helical structure
lated. Another PTA subclass that has garnered considerable (Goodman and Fried, 1967), though X-ray crystallography of
attention is the peptoid scaffold (Simon et al., 1992) (Figure 1, (N-Me-Ala)6 shows a more extended, b strand-like conformation
R1 = H). Peptoids are oligomers of N-alkylated glycines. Like (Zhang et al., 2006). A striking indication of the much greater
N-methyl peptides, they are also stable to proteases and exhibit conformational constraints in these molecules have been
enhanced cell permeability relative to standard peptides (Kwon reports of conformational isomers of the same compound that
and Kodadek, 2007; Miller et al., 1994). In contrast to N-methyl can be separated by high-pressure liquid chromatography
´
peptides, there have been numerous reports of the construction (HPLC) (Alfredson et al., 1994; Teixido et al., 2005). This can
of large synthetic combinatorial peptoid libraries that have been be rationalized on the basis of nonbonded steric interactions.
screened successfully for protein ligands (Kodadek, 2010; For example, as shown in Figure 2A, the trans amide bond
´
Kodadek et al., 2004; Sanchez-Perez et al., 2003; Zuckermann
´
geometry will be strongly favored in N-methylalanine in order
and Kodadek, 2009). This is largely due to the ease of construct- to separate the two branched stereocenters, as is also true in
ing peptoid libraries via the ‘‘submonomer’’ route (Figure 1B, simple peptides. Moreover, an allylic 1,3 (A1,3) strain will act to
R1 = H) (Figliozzi et al., 1996; Uno et al., 1999; Zuckermann restrict rotation about the carbonyl-Ca and the Ca-nitrogen
et al., 1992). In this scheme, an activated ester of 2-bromoacetic bonds so as to keep the hydrogen atom at the chiral center in
acid (or the chloro analog; Burkoth et al., 2003) is added to an or near the plane of the substituted amide (Figure 2B). These
amine-displaying resin. The halide is then displaced by a nucleo- effects are seen clearly in the crystal structure of (N-Me-Ala)6
philic primary amine, providing a peptoid monomer. Because (Zhang et al., 2006).
360 Chemistry & Biology 20, 360–369, March 21, 2013 ª2013 Elsevier Ltd All rights reserved

Chemistry & Biology
Conformationally Restrained Peptide Tertiary Amide
Figure 1. Synthesis of Peptide-like Oligo-
mers
Solid-phase submonomer synthesis of peptoids
(when R1 = H). See also Figures S1 and S2.
in PTAs is required. This meant accessing
deuterated amino acids, which can be
accomplished through transaminase-
catalyzed hydrogen-deuterium exchange
Of course, the same arguments would apply to molecules using D2O as the donor. Following the published procedure
containing other N-substituents besides methyl. Thus, it would (Oshima and Tamiya, 1961), more than 90% of the L-alanine
be of great interest to construct combinatorial libraries of was labeled with deuterium on both the a and b carbons.
diverse N-alkylated peptides and investigate them as sources When the procedure was repeated three times, >98% isotopic
of bioactive molecules, but this has never been done. N-methyl- purity was obtained (see Figures S1 and S2). This simple proce-
ated, N-protected amino acids are available (Biron and Kessler, dure can be performed on a large scale (>30 g) because L-
2005), providing the building blocks required for the synthesis of alanine, D2O, and transaminase are all inexpensive.
OBOC libraries of N-methylated peptides by standard peptide
Isotopically labeled L-alanine was then converted to the
bond couplings. But to apply a standard peptide synthesis bromide as described above. After distillation under reduced
approach to libraries with different N-substituents would entail pressure, the desired acid displayed high stereochemical purity
the maintenance of a huge number of different N-substituted, (>95% enantiomeric excess) as determined by HPLC of the cor-
Fmoc-protected amino acid building blocks. To address this responding (+)-camphorsultam amide (see Figure S1D).
important goal, we therefore considered a different solution: to
create libraries of PTAs via submonomer synthesis in which Submonomer Synthesis of Optically Pure N-Substituted
chiral 2-bromo carboxylic acids are employed in place of 2-bro- Alanines
moacetic acid (Figure 1B, R1 s H).
The peptide bond-forming step in the submonomer synthesis of
Here, we report a relatively general procedure for the creation peptoids usually employs a carbodiimide, such as diisopropyl-
of high-quality OBOC libraries of diverse N-substituted alanines carbodiimide (DIC), to activate bromoacetic acid. Although the
and also demonstrate that the same chemistry can be applied to use of this reagent can be problematic in peptide synthesis
include residues other than methyl at the alpha carbon. More- because of racemization at the a-carbon, this was not antici-
over, we demonstrate that these libraries are indeed a source pated to be an issue in PTA synthesis via the submonomer route
of high-affinity ligands for a protein target. Characterization of because amino acid racemization involves formation of an inter-
one of these PTA-protein complexes provides strong evidence mediate oxazalone that cannot form in this class of molecules
for the importance of conformational constraints in supporting (Goodman and Levine, 1964; Yuan et al., 2002). To confirm
high-affinity binding.
this, R and S bromopropionic acid were condensed with a chiral
amine ((S)-(-)-1-phenylethylamine) under these conditions. No
evidence for racemization was observed in the nuclear magnetic
resonance spectrum of the final products (see Figure S2).
To determine if oligomers prepared via the submonomer route
RESULTS AND DISCUSSION
Synthesis of Chiral Submonomers
To construct PTAs via a peptoid-like, submonomer synthetic indeed display the anticipated conformational constraints, we
route (Zuckermann et al., 1992), chiral 2-bromo carboxylic constructed N-methyl-(S)-alanine oligomers of various lengths
acids must be employed in place of 2-bromoacetic acid (Figure 1, on Rink amide resin. As shown in Figure 3, an increased circular
RsH). These building blocks can be prepared from a-amino dichroism signal was observed as the length of the oligomer
acids by treatment with sodium nitrate in the presence of KBr increased from a dimer to a tetramer to a hexamer, suggesting
and HBr (see Figure S1A available online) (Izumiya and the buildup of specific conformers as the length of the molecule
Nagamatsu, 1952; Tanasova et al., 2009). This reaction increases. The CD spectrum of the hexamer shown in Figure 3 is
proceeds with retention of stereochemistry, presumably essentially identical with the published spectrum of the same
because, after conversion of the amine to a diazonium group, compound made by traditional amino acid couplings (Zhang
a reactive three-membered lactone is formed that is then opened et al., 2006). Thus, this chemistry provides a route to highly struc-
by bromide ion (Izumiya and Nagamatsu, 1952).
To make full use of both enantiomers of a given 2-bromo
tured oligomers.
carboxylic acid in the construction and screening of PTA combi- Synthesis and Characterization of a Combinatorial
natorial libraries, a method to decode stereochemistry in each Library
monomer is required. Because tandem mass spectrometry is With the appropriate conditions in hand, we proceeded
the most convenient method to elucidate the structure of to construct a combinatorial library of trimeric PTAs following
‘‘hits’’ after screening one-bead one-compound (OBOC) combi- an invariant linker. The synthesis was conducted on TentaGel
natorial libraries (Astle et al., 2010; Paulick et al., 2006; Thakkar beads, which is the preferred resin for subsequent screening
et al., 2009), isotope encoding of the absolute stereochemistry (Alluri et al., 2003). 2-bromoacetic acid, (S)-2-bromopropanoic
Chemistry & Biology 20, 360–369, March 21, 2013 ª2013 Elsevier Ltd All rights reserved 361

Chemistry & Biology
Conformationally Restrained Peptide Tertiary Amide
Figure 2. Conformational Analysis of Peptide Tertiary Amides Using a Simple Model Compound in which All of the Substituents Are Methyl
Groups
(A–C) The bond whose conformation is considered is highlighted in red.
(A) Amide bond conformations.
(B) Carbonyl-Ca bond conformations.
(C) Ca-N bond conformations. The conformations indicated as favorable are seen in the crystal structure of hexa-N-methylalanine (Zhang et al., 2006).
acid-d4, and (R)-2-bromopropanoic acid) were employed releasing the compound using CNBr, the progress of the reaction
as submonomers along with 16 amines (see Figure 4). The theo- was checked by MALDI mass spectrometry. This monitoring is
retical diversity of the library was 110,592 compounds. The important to create a high-quality library. The acylation step in
amines were chosen carefully such that any combination of PTA synthesis, in particular, can be significantly slower than is
amine and acid submonomers could be identified uniquely by the case for peptoid synthesis. Thus, although the conditions re-
tandem mass spectrometry upon gas-phase cleavage of the ported above for DIC-mediated coupling work well much of the
amide bond. Methionine was added to the beads first, using time, this depends to some extent on the N-substituent. The blind
standard peptide coupling conditions (N,N,N0,N0-tetramethyl- use of a single set of reaction conditions without monitoring the
O-(1H-benzotriazol-1-yl)uronium hexafluorophosphate [HBTU], reaction can result in a significant fraction of incomplete coupling.
Fmoc-Met-OH, DIEA) to facilitate selective release of the
In tubes in which the reaction had not proceeded to >90%
compounds from the beads after the screen by treatment with completion using the standard procedure, a second round of
CNBr. Following the methionine, several invariant peptoid resi- coupling was done by adding a 2:1 mixture of the bromopropa-
dues were added. The two Nlys residues facilitate display of noic acid and DIC (after a preactivation time of 10 min, final
the molecule in aqueous solution, whereas the alkyne and furan concentration of activated bromopropanoic acid at 1 M). Pro-
side chains facilitate subsequent labeling of the compound longed reaction time (24 hr) provided a satisfactory yield in these
(Hintersteiner et al., 2009) and immobilization on a microarray difficult reactions. With respect to assessing the progress of the
(Astle et al., 2010) if desired. Moreover, the additional mass reactions by mass spectrometry, it is noteworthy that the amide
serves to move the molecular ion and important fragment peaks bond on the C-terminal side of the PTA fragments easily and is
out of the range of the matrix peaks, which facilitates compound much more acid labile than in standard peptoids (see Figure S5).
characterization.
This results in the production of an abundant b ion on the MALDI
Following linker synthesis, the library was split in three, treated mass spectrum even without high-energy tandem mass spec-
with DIC and (R)-2-bromopropanoic acid, (S)-2-bromopropanoic trometry fragmentation. The same effect has been noticed previ-
acid-d4, or bromoacetic acid, respectively. The beads were then ously in the mass spectrometric analysis of N-methyl peptides.
mixed together and then split in 16 equal parts. One of the amines (Vaisar and Urban, 1998).
was added to each tube. The same procedures were then
After the final step, the entire library was washed thoroughly,
repeated two more times in order to form the trimer. The acylation and 40 beads were picked randomly. The beads were treated
step was monitored by the chloranil test, and the amine substitu- with CNBr to release the compounds from the beads, which
tion was monitored by the silver acetate test (details are were then analyzed by tandem MALDI mass spectrometry. We
described in the Supplemental Experimental Procedures). In successfully sequenced 37 out of 40 beads for quality control
addition, individual beads were sampled at each step. After (see Figure S4 for some examples). All 16 of the amines and all
362 Chemistry & Biology 20, 360–369, March 21, 2013 ª2013 Elsevier Ltd All rights reserved

Chemistry & Biology
Conformationally Restrained Peptide Tertiary Amide
Figure 3. CD Spectrum of the N-Methyl-Alanine Dimer, Tetramer, and Hexamer Synthesized via the Submonomer Route
(A) Spectrum obtained in an aqueous solution (PBS buffer pH = 7).
(B) Spectrum obtained in a trifluoroethanol solution. All spectra were recorded at 0.3 mM compound, path length 10 mm, at 298ꢀK.
See also Figure S3.
three of the acids were found in these spectra, suggesting that all picked up magnetically displayed a red glow (data not shown),
of the submonomers had incorporated in this synthesis. consistent with their being PX4-4 ligands. These validated hits
However, we observed no trimers in which a methyl group was were released from the beads with CNBr and sequenced by
present at all three positions, and some dimers were also found. tandem mass spectrometry.
This is clear evidence that in the demanding library synthesis
The most promising hit from the bead screen (as judged by the
format, the reaction conditions employed resulted in some level intensity of the red halo after quantum dot addition), compound
of failure when attempting to string together three chiral centers 1, was resynthesized on Rink amide resin with a fluorescein tag,
in a row, despite all of our efforts. As will be described cleaved, and purified by HPLC. As shown in Figure 5A, this
below, a modified coupling protocol has solved this problem. compound proved to have chiral centers at the first and third
Nonetheless, because most compounds in the library could be variable positions of the library (highlighted with ovals in
sequenced, we decided to proceed with a screening experiment Figure 5A), whereas there was no Ca substituent at the second
to probe the idea that the greater conformational constraints in position.
the PTA relative to peptoids might result in higher affinity ligands.
The affinity of compound 1 for PX4-4 was determined by fluo-
rescence anisotropy. As shown in Figure 5B, PTA fluorescein-
conjugated 1 exhibited a saturable binding curve when titrated
with increasing concentrations of PX4-4, indicating a KD of
Isolation and Characterization of Protein Ligands from
the PTA Library
The library shown in Figure 4 was screened for ligands to approximately 5 mM. In contrast, little binding was observed
a single-chain variable fragment (scFv) antibody called PX4-4, when fluorescein-tagged 1 was mixed with bulk IgG antibodies,
which was derived from a patient with the skin blistering disease indicating selectivity for PX4-4.
Pemphigus vulgaris (Yamagami et al., 2010). Approximately
The peptoid analog of 1, compound 2, which lacks any chiral
250,000 beads were incubated with 10 mM of purified scFv centers, was also made. As shown in Figure 5B, this compound
PX4-4 in the presence of 1 mg/ml Escherichia coli lysate as had a much lower affinity for PX4-4. The large difference in
a diverse source of competitor proteins. To isolate beads that affinity of PTA 1 and peptoid 2, which differ only in the presence
display ligands that captured significant amounts of the target or absence of the methyl groups at the chiral centers, is striking.
antibody, magnetic beads displaying protein L were added to Although we cannot rule out the possibility that the Ca methyl
the mixture. Protein L will bind tightly to the scFv antibody groups in PTA 1 contact the scFv antibody directly, these data
through the kappa light chains. After a brief incubation, a power- strongly support the idea that the conformational constraints
ful magnet was employed to segregate the beads that had also afforded by the presence of the chiral centers strongly stabilizes
affixed to the magnetic protein L-displaying beads (Astle et al., binding of the small molecule to the antibody fragment.
2010), and 33 beads were isolated as possible hits.
To more thoroughly probe the influence of the stereochemistry
To confirm that these beads were true hits, they were stripped at each position of PTA 1, a small library was synthesized in
of protein by incubation in a trypsin solution at 37ꢀC for 1 hr and which the amines were held invariant as found in 1, but all
then washed thoroughly. The beads were then incubated with possible combinations of the three acid submonomers were
10 mM of purified scFv PX4-4 again for 30 min and washed employed at each alpha position. In order to ensure the efficient
gently. A 1:200 dilution of red quantum dots conjugated to synthesis of PTA trimers, we performed an extensive optimiza-
protein L was then added. After 30 min of incubation, brightly tion of acylation conditions and conditions for amination. Of
glowing beads were picked manually using a low-power fluores- the eight coupling reagents examined, the triphosgene reagent
cence microscope to visualize them. Seventeen of the hits bis(trichloromethyl) carbonate (BTC) developed by Jung and
Chemistry & Biology 20, 360–369, March 21, 2013 ª2013 Elsevier Ltd All rights reserved 363

Chemistry & Biology
Conformationally Restrained Peptide Tertiary Amide
Figure 4. Structure of the Library Con-
structed: The Invariant Linker Is Boxed
The acid submonomers are indicated as ‘‘X’’. The
amine submonomers employed are shown at the
bottom of the figure. Protecting groups employed
on the acids, alcohols, and diaminobutane are not
shown. See also Figures S4 and S5 and Table S1.
centers can have a profound effect on
binding affinity. It is interesting that the
compounds with the S,S,R or S,R,R
configurations (the same as compound
1 in the first and third positions but with
a chiral center at position 2) were not
high-affinity ligands for PX4-4 (see Table
S2). This indicates that a stereocenter at
the second position does not allow the
molecule to access the bound
conformation.
To address the selectivity of com-
pounds A14, A7, A68, A41, and A24 for
PX4-4, the FP experiment was repeated
using a mixture of human IgG as we did
previously for compound 1. A14, A7,
A68, and A24 showed no binding with
whole-human IgG, whereas A41, the
analog of peptoid 2, had a similar affinity
colleagues (Thern et al., 2002) was found to be the best by far, for these random antibodies as it did for PX4-4 (Figure 6). This
providing superior yields with no racemization (see Figure S6).
implies that greater conformational flexibility is associated with
Several beads were chosen randomly from the library, and the binding promiscuity, which has been noted previously in studies
compounds were released and analyzed by mass spectrometry. of N-methylated peptides (Doedens et al., 2010; Fiacco and
Gratifyingly, in this case, several of the compounds were trimers Roberts, 2008).
with chiral centers at each position. Indeed, close to the
expected ratio of compounds was observed (see Figure S6), Beyond Oligo-N-Substituted Alanines
demonstrating that the use of the BTC-based protocol is appro- The synthetic route employed to access the chiral bromides
priate for the synthesis of high-quality PTA libraries.
works well with other amino acids besides alanine (Izumiya
Ninety-six beads were chosen randomly from the library and and Nagamatsu, 1952; Tanasova et al., 2009). Thus, we con-
placed individually into the wells of a 96-well filter plate. Each ducted a small, preliminary study of whether bulkier substituents
well was treated with fluorescein azide and a copper catalyst at the a-carbons in 1 might increase the affinity of the PTA ligand
to attach fluorescein to the alkyne handle (Hintersteiner et al., for PX4-4. Four fluorescein-labeled compounds were synthe-
2009), and then the compounds were cleaved from beads. The sized in which a benzyl or an isobutyl group was substituted
beads were filtered out, and the soluble compounds were for the chiral methyl groups of 1. The crude HPLCs of three of
used for the fluorescence polarization (FP) assay. All 27 possible these compounds were quite clean, whereas the fourth had
structures were found in this collection of molecules by tandem some level of impurities (see Figures S7 and S8). Their affinities
MS sequencing. Semiquantitative binding constants for the for PX4-4 were measured by fluorescence polarization spectros-
PX4-4 antibody for all 27 of the compounds were determined copy. As shown in Figure S7, none of the compounds bound
by titrating the compounds with the protein in the microwell PX4-4 as tightly as the parent compound. The affinities were
format and monitoring the increase in FP (see Table S2). The reduced between 2- to 20-fold for the four compounds. Although
six highest affinity compounds, which included A48, identical improved PX4-4 ligands were not discovered in this small exper-
to hit 1 in the variable region, were resynthesized and purified iment, it shows that PTAs other than oligo-alanines can be ac-
to allow more accurate binding data to be acquired. All six com- cessed via this chemistry.
pounds showed an increase in FP when titrated with PX4-4 (Fig-
ure 6). Most of the curves approached saturation, allowing for an Conclusions
accurate determination of the KD. In contrast, compound A41, We report the synthesis of diverse libraries of peptide tertiary
corresponding to peptoid 2, was not saturated in the protein amides (PTAs) using the submonomer approach first developed
concentration range tested (100 nM to 400 mM) (Figure 6).
for peptoid synthesis (Figliozzi et al., 1996; Zuckermann et al.,
These data confirmed that compound 1 is the highest affinity 1992). This involves the use of chiral, 2-bromo acids as submo-
PX4-4 ligand of the six and that the chirality of the stereogenic nomers in place of 2-bromoacetic acid, several of which can be
364 Chemistry & Biology 20, 360–369, March 21, 2013 ª2013 Elsevier Ltd All rights reserved

Chemistry & Biology
Conformationally Restrained Peptide Tertiary Amide
Figure 5. Structure and Characterization of
a Ligand for the scFv PX4-4
(A) Structure of a fluorescein-labeled derivative of
one of the screening hits, compound 1, and the
peptoid analog 2, which lacks the methyl groups at
the chiral centers. The positions of the chiral centers
in 1 are highlighted in the silver ovals, as well as the
analogous carbons in 2.
(B) Fluorescence polarization assay employing the
fluorescein-tagged compounds and the indicated
proteins. PTA 1 binds the PX4-4 scFv with a KD of
approximately 3 mM but binds much less well to
a collection of IgG antibodies, indicating selectivity
for PX4-4. The peptoid 2 binds poorly to PX4-4.
See also Figure S6.
such as Jung’s BTC reagent (Thern et al.,
2002; Videnov et al., 1996), allows the
synthesis of high-quality libraries by split
and pool synthesis. In this initial report,
we have mostly limited the substitution at
the a-carbon to methyl (i.e., the synthesis
of N-alkylated oligoalanines), but, as
shown in Figures S7 and S8, other groups
can be accommodated at this position.
The impetus behind the development of
this chemistry was the need to create
libraries of oligomers with greater confor-
mational constraints than simple peptoids.
Peptoid libraries are a useful source of
protein ligands, but the primary hits from
these libraries tend to be of modest
affinity. This is likely due, in part, to the
fact that peptoids are quite floppy mole-
cules that must sacrifice a good deal of
entropy in order to assume their bound
conformation. Although various clever
strategies have been reported to confor-
mationally constrain peptoids (Yoo and
Kirshenbaum, 2008), none of these have
been applied to the synthesis of large
combinatorial libraries, probably because
of the extremely high efficiency demanded
of the chemistry for this application (step-
wise yields in excess of 95%).
N-methyl peptides are known to have
significant conformational constraints
derived easily from amino acids in high yield and optical purity (Chatterjee et al., 2009; Goodman and Fried, 1967; Zhang
(Izumiya and Nagamatsu, 1952). Moreover, at least in the case et al., 2006), resulting from both a strong preference for the trans
of the alanine-derived 2-bromoacid employed in this study, amide bond geometry as well as potent allylic 1,3 strain effects
a simple transaminase-mediated deuteration protocol provides that bias the conformations of the other two types of bonds in
isotopically labeled material to allow the absolute stereochem- the molecule. Although chemistry exists to access Fmoc-pro-
istry at the chiral center of the PTA to be determined by mass tected, N-methyl amino acid monomers (Biron and Kessler,
spectrometry. This, in turn, allows both the R and S acids to be 2005), only small libraries of N-methyl peptides have been
used as submonomer diversity elements at any given position created by parallel synthesis (Ovadia et al., 2011). To the best
in the chain. As anticipated from studies of N-methyl peptide of our knowledge, large libraries of these molecules have never
synthesis, the peptide bond couplings are more difficult than is been created synthetically and screened for protein ligands,
the case for simple peptoids due to greater steric crowding, though there has been considerable activity in the creation of
but the use of highly potent carboxylate-activating agents, such libraries using biological approaches (Kawakami et al.,
Chemistry & Biology 20, 360–369, March 21, 2013 ª2013 Elsevier Ltd All rights reserved 365

Chemistry & Biology
Conformationally Restrained Peptide Tertiary Amide
A
B
C
Figure 6. Binding Isotherms for A48 and the Five Other Highest Affinity Members of the Library in which the Stereochemistry at Each
Alpha-Carbon of 3 Was Altered: Compound A48 Is Compound 3 with a Fluorescein Label
(A) General structure of the tested compounds and the KD determined by FP assay after resynthesis as shown in (B). The KD values are given in micrometers.
(B) Fluorescent polarization assay of a titration of fluorescein-labeled compound with PX4-4.
(C) Titration of fluorescein-labeled compound with a collection of human IgG antibodies (C). Note the different scales of the y axes in (B) and (C). The binding of the
peptoid A41 to PX4-4 and whole-human IgG is similar.
See also Table S2 and Figures S7 and S8.
2008; Subtelny et al., 2008). Thus, the PTA library described here be that compounds A14 and A7 (Table S2) can place the side
is good example of a large, chemically synthesized, and confor- chains in something close to the three-dimensional arrangement
mationally constrained library of N-alkylated peptides.
achieved by A48 in its bound conformation, despite their
To determine if this type of library would indeed be superior to different backbone stereochemistry. Efforts to obtain crystal
a collection of simple peptoids, we created a library with three structures are ongoing.
variable positions and screened it against an autoantibody of
In conclusion, this work makes accessible libraries of con-
interest. Sixteen amines and three bromoacids (R- and S-2-bro- formationally constrained N-alkylated peptide oligomers that
mopropionic acid and bromoacetic acid) were employed as promise to be exceptional ligands for biological targets of
submonomers. Thus, the library contained 3,375 peptoids interest.
(163) among the 110,592 PTAs, yet no simple peptoids were iso-
lated as hits from the screen. All of the hits had at least one chiral SIGNIFICANCE
center. Moreover, characterization of the best hit, compound 1,
demonstrated that it evinced a high affinity for the target anti- Peptide tertiary amides (PTAs; also called N-alkylated
body (KD = 5 mM), whereas the peptoid analog, 2, had a much peptides) have long been known to be conformationally con-
lower affinity for the autoantibody and also demonstrated strained. This suggests that libraries of these compounds
more promiscuous binding behavior. Finally, an analysis of all might be a superior source of protein ligands than the
possible stereoisomers of hit 1 revealed that the compound corresponding libraries of relatively ‘‘floppy’’ peptoids or
isolated was indeed the highest affinity ligand among this peptides. However, chemistry to create high-quality PTA
groups and that the vast majority of stereoisomers evinced libraries was lacking. This report demonstrates that large
much lower affinity for the target. Only two other compounds one-bead one-compound libraries of N-alkylated alanines
bound to PX4-4 with an affinity within 10-fold of A48. Although can be constructed by a peptoid-like submonomer synthesis
it is impossible to understand exactly what this means in the using either enantiomer of 2-bromopropionic acid as
absence of a structure of the PTA-antibody complexes, it may a building block. A model screening experiment suggests
366 Chemistry & Biology 20, 360–369, March 21, 2013 ª2013 Elsevier Ltd All rights reserved

Chemistry & Biology
Conformationally Restrained Peptide Tertiary Amide
using DCM, DMF, and then THF, respectively for 5 times each, and then 2:1
THF/DIPEA (750 ml THF, 375 ml DIPEA, 2.2 mmol) was added to the beads, fol-
lowed by gentle shaking. 2,4,6-Trimethylpyridine (356 ml, 2.7 mmol) was added
to the cold solution of bromopropionic acid with BTC, and white precipitation
formed following the addition. The white suspension was then applied to the
beads, and the reaction vessel was put on a shaker for 2 hr at room tempera-
ture. The solution in the vessel should be a pale yellowish suspension during
the whole course of the reaction. A darker color is an indication of excessive
heat released during the initial addition of the acid chloride solution. It can
be solved by further cooling of the acid chloride solution and the beads. The
beads were washed with DCM five times when the reaction was done and
then with DMF for five times. A chloranil test was used to monitor the comple-
tion of the reaction. All three portions of beads were then pooled together, and
the beads were incubated with 2 M solution of the corresponding amine in
DMF at 60ꢀC overnight. The completion of the reaction was monitored by
chloranil and silver acetate test.
that these molecules will indeed provide higher affinity
binding to protein targets than unconstrained peptoids.
EXPERIMENTAL PROCEDURES
Synthesis of Chiral Bromoacid
D-alanine (8.9 g, 0.1 mol) and KBr (11.9 g, 0.1 mol) was dissolved in 100 ml, of
a 30% HBr water solution and kept in ꢁ15ꢀC dry ice/ethylene glycol bath.
NaNO2 (10.35 g, 0.15 mmol) were dissolved in 15 ml water and slowly dripped
in the above solution under argon atmosphere. The reaction was allowed to
proceed for 3 hr and to warm from ꢁ15ꢀC to room temperature. It was then
put under vacuum for 30 min. Product was extracted by diethyl ether
(25 ml 3 3). Organic phases were combined and dried over Na2SO4. Then
the solvent was evaporated under vacuum, and the crude product was further
purified by distillation at 115ꢀC, high vacuum. The pure product was obtained
as a colorless oil in 82% yield.
SUPPLEMENTAL INFORMATION
Isotopic Labeling of Alanine
L-alanine (300 mg, 3.36 mmol) was dissolved in 10 ml of D2O, a-ketoglutarate
(10 mg, 0.068 mmol) was then added as a cosubstrate, and the whole system
was then warmed to 37ꢀC, and the pD was adjusted to 8.5ꢂ8.7 with NaOD.
Alanine transaminase (0.1 mg, EC 2.6.1.2 from pig heart, Roche Diagnostics,
Indianapolis, IN, USA) was added and incubated at 37ꢀC overnight with mild
shaking. Ninety percent of D2O was recovered by distillation, and L-alanine-
d4 was obtained by lyophilization.
Supplemental Information includes Supplemental Experimental Procedures,
eight figures, and two tables, and can be found with this article online at
http://dx.doi.org/10.1016/j.chembiol.2013.01.013.
ACKNOWLEDGMENTS
This work supported by a contract from the National Institutes of Health Heart,
Lung, and Blood Institute (NO1-HV-00242). We thank Prof. John Stanley
(University of Pennsylvania) for providing the expression vector for PX4-4.
Library Synthesis with DIC
Tentagel beads (1 g, 160 mm, ꢂ500,000 beads, 0.52 mmol/g, cat# HL 12 162,
Rapp-Polymere, Tuebingen, Germany) were swelled in dimethylformamide
(DMF) for 2 hr before use. DMF was used as the solvent unless otherwise
mentioned. Fmoc-Met-OH (0.77 g, 2.08 mmol) was coupled on the beads
using HBTU (0.77 g, 2.08 mmol) and N,N-Diisopropylethylamine (DIPEA)
(0.45 ml, 2.6 mmol) for 3 hr. Fmoc was deprotected by 20% piperidine for
30 min. Beads were washed thoroughly with DMF after each step. The beads
were split in three portions after deprotection. In a conical tube, 2 ml 1 M DIC
solution and 2 ml 1 M corresponding bromoacid (2-Bromoacetic acid, (S)-2-
bromopropanoic acid-d4 or (R)-2-bromopropanoic acid) solution were added
together for preactivation for 5 min. The 4 ml combined solution was then
added to one portion of beads and gently shook till completion. The reaction
was monitored by chloranil test, a clear negative result after 5 min indicates
the amine was acylated by the corresponding bromoacid. The beads were
then thoroughly washed and pooled together, then split in 16 portions. Each
portion was incubated with one of the amines listed in Table S1. Amine was
used as 2 M DMF solution with an incubation time of 12 hr at 50ꢀC. Silver
acetate test and chloranil test were used to monitor the completion of the reac-
tion. When reaction was complete, the beads were washed and pooled
together. The acylation and amination steps were then repeated two more
times for forming the trimer library.
Received: August 31, 2012
Revised: December 16, 2012
Accepted: January 4, 2013
Published: March 21, 2013
REFERENCES
Aina, O.H., Marik, J., Liu, R., Lau, D.H., and Lam, K.S. (2005). Identification of
novel targeting peptides for human ovarian cancer cells using ‘‘one-bead one-
compound’’ combinatorial libraries. Mol. Cancer Ther. 4, 806–813.
Alfredson, T.V., Bruins, P.W., Maki, A.H., and Excoffier, J.L. (1994). Conformer
interconversion in the LC analysis of triostin A and its under-N-methylated
synthetic analogue. J. Chromatogr. Sci. 32, 132–138.
Alluri, P.G., Reddy, M.M., Bachhawat-Sikder, K., Olivos, H.J., and Kodadek, T.
(2003). Isolation of protein ligands from large peptoid libraries. J. Am. Chem.
Soc. 125, 13995–14004.
Aquino, C., Sarkar, M., Chalmers, M.J., Mendes, K., Kodadek, T., and
Micalizio, G.C. (2012). A biomimetic polyketide-inspired approach to small-
molecule ligand discovery. Nat. Chem. 4, 99–104.
Astle, J.M., Simpson, L.S., Huang, Y., Reddy, M.M., Wilson, R., Connell, S.,
Wilson, J., and Kodadek, T. (2010). Seamless bead to microarray screening:
rapid identification of the highest affinity protein ligands from large combinato-
rial libraries. Chem. Biol. 17, 38–45.
Peptoid Linker Synthesis with DIC
The peptoid linker was synthesized by standard peptoid synthesis methods
using microwave condition and 2 M solutions of DIC, bromoacetic acid, and
the corresponding amines in DMF.
Biron, E., and Kessler, H. (2005). Convenient synthesis of N-methylamino acids
compatible with Fmoc solid-phase peptide synthesis. J. Org. Chem. 70, 5183–
5189.
Library Synthesis with BTC
Tentagel beads with Rink linker (1 g, ꢂ100,000 beads, 0.27 mmol/g, cat. # MB
250 230, Rapp-Polymere) were used in order to avoid long exposure to acid
during cleavage. Beads were first treated by 20% piperidine for 30 min in
DMF in order to remove Fmoc and then the linker part was synthesized as
described above. The beads were then split in three equal portions, coupled
with 2-Bromoacetic acid, (S)-2-bromopropanoic acid-d4, and (R)-2-bromo-
propanoic acid, respectively. For bromoacetic acid, 2 M solution of DIC and
2 M solution of bromoacetic acid were used as described above. For bromo-
propanoic acids, BTC was used as a coupling reagent. BTC (92.1 mg,
0.31 mmol) was dissolved in 5 ml anhydrous tetrahydrofuran (THF) in a glass
vial. Bromopropionic acid (89 ml, 0.95 mmol) was then added to the vial, and
the whole vial was kept in a ꢁ20ꢀC freezer for 15 min. Beads were washed
Burkoth, T.S., Fafarman, A.T., Charych, D.H., Connolly, M.D., and
Zuckermann, R.N. (2003). Incorporation of unprotected heterocyclic side
chains into peptoid oligomers via solid-phase submonomer synthesis.
J. Am. Chem. Soc. 125, 8841–8845.
Chatterjee, J., Ovadia, O., Gilon, C., Hoffman, A., Mierke, D., and Kessler, H.
(2009). N-methylated cyclic pentapeptides as template structures. Adv. Exp.
Med. Biol. 611, 109–110.
Chongsiriwatana, N.P., Patch, J.A., Czyzewski, A.M., Dohm, M.T., Ivankin, A.,
Gidalevitz, D., Zuckermann, R.N., and Barron, A.E. (2008). Peptoids that mimic
the structure, function, and mechanism of helical antimicrobial peptides. Proc.
Natl. Acad. Sci. USA 105, 2794–2799.
Chemistry & Biology 20, 360–369, March 21, 2013 ª2013 Elsevier Ltd All rights reserved 367

Chemistry & Biology
Conformationally Restrained Peptide Tertiary Amide
Doedens, L., Opperer, F., Cai, M., Beck, J.G., Dedek, M., Palmer, E., Hruby,
V.J., and Kessler, H. (2010). Multiple N-methylation of MT-II backbone amide
bonds leads to melanocortin receptor subtype hMC1R selectivity: pharmaco-
logical and conformational studies. J. Am. Chem. Soc. 132, 8115–8128.
Oshima, T., and Tamiya, N. (1961). Mechanism of transaminase action.
Biochem. J. 78, 116–119.
Ovadia, O., Greenberg, S., Chatterjee, J., Laufer, B., Opperer, F., Kessler, H.,
Gilon, C., and Hoffman, A. (2011). The effect of multiple N-methylation on
intestinal permeability of cyclic hexapeptides. Mol. Pharm. 8, 479–487.
Fiacco, S.V., and Roberts, R.W. (2008). N-Methyl scanning mutagenesis
generates protease-resistant G protein ligands with improved affinity and
selectivity. ChemBioChem 9, 2200–2203.
Paulick, M.G., Hart, K.M., Brinner, K.M., Tjandra, M., Charych, D.H., and
Zuckermann, R.N. (2006). Cleavable hydrophilic linker for one-bead-one-
compound sequencing of oligomer libraries by tandem mass spectrometry.
J. Comb. Chem. 8, 417–426.
Figliozzi, G.M., Goldsmith, R., Ng, S.C., Banville, S.C., and Zuckermann, R.N.
(1996). Synthesis of N-substituted glycine peptoid libraries. Methods Enzymol.
267, 437–447.
´
Sanchez-Perez,
´
A.M., Montoliu, C., and Felipo, V. (2003). Trialkylglycines:
a new family of compounds with in vivo neuroprotective activity. CNS Drug
Goodman, M., and Levine, L. (1964). Peptide synthesis via active esters. IV.
Racemization and ring-opening reactions of opticlly active oxazolones.
J. Am. Chem. Soc. 86, 2918–2922.
Rev. 9, 263–274.
Sarma, B.K., Yousufuddin, M., and Kodadek, T. (2011). Acyl hydrazides as
peptoid sub-monomers. Chem. Commun. 47, 10590–10592.
Goodman, M., and Fried, M. (1967). Conformational aspects of polypeptide
structure. XX. Helical poly-N-methyl-L-alanine. Experimental results. J. Am.
Chem. Soc. 89, 1264–1267.
Shah, N.H., Butterfoss, G.L., Nguyen, K., Yoo, B., Bonneau, R., Rabenstein,
D.L., and Kirshenbaum, K. (2008). Oligo(N-aryl glycines): a new twist on struc-
tured peptoids. J. Am. Chem. Soc. 130, 16622–16632.
Gorske, B.C., and Blackwell, H.E. (2006). Tuning peptoid secondary structure
with pentafluoroaromatic functionality:
a new design paradigm for the
Shin, S.B., Yoo, B., Todaro, L.J., and Kirshenbaum, K. (2007). Cyclic peptoids.
construction of discretely folded peptoid structures. J. Am. Chem. Soc. 128,
J. Am. Chem. Soc. 129, 3218–3225.
14378–14387.
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., et al. (1992).
Peptoids: a modular approach to drug discovery. Proc. Natl. Acad. Sci. USA
89, 9367–9371.
Hintersteiner, M., Kimmerlin, T., Kalthoff, F., Stoeckli, M., Garavel, G., Seifert,
J.M., Meisner, N.C., Uhl, V., Buehler, C., Weidemann, T., and Auer, M. (2009).
Single bead labeling method for combining confocal fluorescence on-bead
screening and solution validation of tagged one-bead one-compound
libraries. Chem. Biol. 16, 724–735.
Stringer, J.R., Crapster, J.A., Guzei, I.A., and Blackwell, H.E. (2011).
Extraordinarily robust polyproline type I peptoid helices generated via the
incorporation of a-chiral aromatic N-1-naphthylethyl side chains. J. Am.
Chem. Soc. 133, 15559–15567.
Huang, K., Wu, C.W., Sanborn, T.J., Patch, J.A., Kirshenbaum, K.,
Zuckermann, R.N., Barron, A.E., and Radhakrishnan, I. (2006). A threaded
loop conformation adopted by a family of peptoid nonamers. J. Am. Chem.
Soc. 128, 1733–1738.
Subtelny, A.O., Hartman, M.C., and Szostak, J.W. (2008). Ribosomal synthesis
of N-methyl peptides. J. Am. Chem. Soc. 130, 6131–6136.
Izumiya, N., and Nagamatsu, A. (1952). Walden inversion of amino acids. VI.
The synthesis of D-Surinamine (N-methyl-D-tyrosine). Bull. Chem. Soc. Jpn.
25, 265–267.
Tanasova, M., TYang, Q., Olmsted, C.C., Vasileiou, C., Li, X., Anyika, M., and
Borhan, B. (2009). An unusual conformation of alpha-haloamides due to
cooperative binding with zincated porphyrins. Eur. J. Org. Chem. 25,
4242–4253.
Kawakami, T., Murakami, H., and Suga, H. (2008). Messenger RNA-pro-
grammed incorporation of multiple N-methyl-amino acids into linear and cyclic
peptides. Chem. Biol. 15, 32–42.
´
Teixido, M., Albericio, F., and Giralt, E. (2005). Solid-phase synthesis and char-
acterization of N-methyl-rich peptides. J. Pept. Res. 65, 153–166.
Kodadek, T. (2010). Synthetic receptors with antibody-like binding affinities.
Thakkar, A., Cohen, A.S., Connolly, M.D., Zuckermann, R.N., and Pei, D.
(2009). High-throughput sequencing of peptoids and peptide-peptoid hybrids
by partial edman degradation and mass spectrometry. J. Comb. Chem. 11,
294–302.
Curr. Opin. Chem. Biol. 14, 713–720.
Kodadek, T., Reddy, M.M., Olivos, H.J., Bachhawat-Sikder, K., and Alluri, P.G.
(2004). Synthetic molecules as antibody replacements. Acc. Chem. Res. 37,
711–718.
Thern, B., Rudolph, J., and Jung, G. (2002). Total synthesis of the nemati-
cidal cyclododecapeptide omphalotin A by using racemization-free triphos-
gene-mediated couplings in the solid phase. Angew. Chem. Int. Ed. 41,
2307–2309.
Kwon, Y.U., and Kodadek, T. (2007). Quantitative evaluation of the relative
cell permeability of peptoids and peptides. J. Am. Chem. Soc. 129, 1508–
1509.
Kwon, Y.U., and Kodadek, T. (2008). Encoded combinatorial libraries for the
construction of cyclic peptoid microarrays. Chem. Commun. 2008, 5704–
5706.
Udugamasooriya, D.G., Dineen, S.P., Brekken, R.A., and Kodadek, T. (2008). A
peptoid ‘‘antibody surrogate’’ that antagonizes VEGF receptor 2 activity.
J. Am. Chem. Soc. 130, 5744–5752.
Lam, K.S., Salmon, S.E., Hersh, E.M., Hruby, V.J., Kazmierski, W.M., and
Knapp, R.J. (1991). A new type of synthetic peptide library for identifying
ligand-binding activity. Nature 354, 82–84.
Uno, T., Beausoleil, E., Glodsmith, R.A., Levine, B.H., and Zuckermann, R.N.
(1999). New submonomers for poly N-substituted glycines (peptoids). Tet.
Lett. 40, 1475–1478.
Lee, J., Reddy, M.M., and Kodadek, T. (2010a). Discovery of an orexin
Vaisar, T., and Urban, J. (1998). Gas-phase fragmentation of protonated
mono-N-methylated peptides. Analogy with solution phase acid-catalyzed
hydrolysis. J. Mass Spectrom. 33, 505–524.
receptor positive potentiator. Chem. Sci. 1, 48–54.
Lee, J.H., Meyer, A.M., and Lim, H.-S. (2010b). A simple strategy for the
construction of combinatorial cyclic peptoid libraries. Chem. Commun. 46,
8615–8617.
Videnov, G.I., Kaiser, D., Kempter, C., and Jung, G. (1996). Synthesis of natu-
rally occurring, conformationally restricted oxazole- and thiazole-containing
di- and tripeptide mimetics. Angew. Chem. Int. Ed. 35, 1503–1506.
Lee, J.H., Kim, H.S., and Lim, H.S. (2011). Design and facile solid-phase
synthesis of conformationally constrained bicyclic peptoids. Org. Lett. 13,
5012–5015.
Wrenn, S.J., Weisinger, R.M., Halpin, D.R., and Harbury, P.B. (2007).
Synthetic ligands discovered by in vitro selection. J. Am. Chem. Soc. 129,
13137–13143.
Lim, H.-S., Archer, C.T., and Kodadek, T. (2007). Identification of a peptoid
inhibitor of the proteasome 19S regulatory particle. J. Am. Chem. Soc. 129,
7750–7751.
Xiao, X., Yu, P., Lim, H.-S., Sikder, D., and Kodadek, T. (2007). Design and
synthesis of a cell-permeable synthetic transcription factor mimic. J. Comb.
Chem. 9, 592–600.
Miller, S.M., Simon, R.J., Ng, S., Zuckermann, R.N., Kerr, J.M., and Moos,
W.H. (1994). Proteolytic studies of homologous peptide and N-substituted
glycine peptoid oligomers. Bioorg. Med. Chem. Lett. 4, 2657–2662.
Yamagami, J., Payne, A.S., Kacir, S., Ishii, K., Siegel, D.L., and
Stanley, J.R. (2010). Homologous regions of autoantibody heavy chain
368 Chemistry & Biology 20, 360–369, March 21, 2013 ª2013 Elsevier Ltd All rights reserved

Chemistry & Biology
Conformationally Restrained Peptide Tertiary Amide
complementarity-determining region
3
(H-CDR3) in patients with pem-
Zuckermann, R.N., and Kodadek, T. (2009). Peptoids as potential therapeu-
phigus cause pathogenicity. J. Clin. Invest. 120, 4111–4117.
tics. Curr. Opin. Mol. Ther. 11, 299–307.
Yoo, B., and Kirshenbaum, K. (2008). Peptoid architectures: elaboration, actu-
Zuckermann, R.N., Kerr, J.M., Kent, S.B.H., and Moos, W.H. (1992).
Efficient method for the preparation of peptoids (oligo N-substituted glycines)
by submonomer solid-phase synthesis. J. Am. Chem. Soc. 114, 10646–
10647.
ation, and application. Curr. Opin. Chem. Biol. 12, 714–721.
Yuan, Z., Blomberg, D., Sethson, I., Brickmann, K., Ekholm, K., Johansson, B.,
Nilsson, A., and Kihlberg, J. (2002). Synthesis and pharmacological evaluation
of an analogue of the peptide hormone oxytocin that contains a mimetic of an
inverse gamma-turn. J. Med. Chem. 45, 2512–2519.
Zuckermann, R.N., Martin, E.J., Spellmeyer, D.C., Stauber, G.B., Shoemaker,
K.R., Kerr, J.M., Figliozzi, G.M., Goff, D.A., Siani, M.A., Simon, R.J., et al.
(1994). Discovery of nanomolar ligands for 7-transmembrane G-protein-
Zhang, S., Prabpai, S., Kongsaeree, P., and Arvidsson, P.I. (2006). Poly-N-
methylated alpha-peptides: synthesis and X-ray structure determination of
beta-strand forming foldamers. Chem. Commun. (Camb.) Feb 7, 497–499.
coupled receptors from
a diverse N-(substituted)glycine peptoid library.
J. Med. Chem. 37, 2678–2685.
Chemistry & Biology 20, 360–369, March 21, 2013 ª2013 Elsevier Ltd All rights reserved 369