• We are sorry, but NCBI web applications do not support your browser and may not function properly. More information
Logo of pnasPNASInfo for AuthorsSubscriptionsAboutThis Article
Proc Natl Acad Sci U S A. Feb 26, 2008; 105(8): 2794–2799.
Published online Feb 19, 2008. doi:  10.1073/pnas.0708254105
PMCID: PMC2268539
Applied Biological Sciences

Peptoids that mimic the structure, function, and mechanism of helical antimicrobial peptides


Antimicrobial peptides (AMPs) and their mimics are emerging as promising antibiotic agents. We present a library of “ampetoids” (antimicrobial peptoid oligomers) with helical structures and biomimetic sequences, several members of which have low-micromolar antimicrobial activities, similar to cationic AMPs like pexiganan. Broad-spectrum activity against six clinically relevant BSL2 pathogens is also shown. This comprehensive structure–activity relationship study, including circular dichroism spectroscopy, minimum inhibitory concentration assays, hemolysis and mammalian cell toxicity studies, and specular x-ray reflectivity measurements shows that the in vitro activities of ampetoids are strikingly similar to those of AMPs themselves, suggesting a strong mechanistic analogy. The ampetoids' antibacterial activity, coupled with their low cytotoxicity against mammalian cells, make them a promising class of antimicrobials for biomedical applications. Peptoids are biostable, with a protease-resistant N-substituted glycine backbone, and their sequences are highly tunable, because an extensive diversity of side chains can be incorporated via facile solid-phase synthesis. Our findings add to the growing evidence that nonnatural foldamers will emerge as an important class of therapeutics.

Keywords: antibiotics, peptidomimetics, structure–activity studies

Natural antimicrobial peptides (AMPs) defend a wide array of organisms against bacterial pathogens and show potential as supplements for or replacements of conventional antibiotics, because few bacteria have evolved resistance to them (13). Many AMPs kill bacteria by permeabilization of the cytoplasmic membrane, causing depolarization, leakage, and death (4), whereas others target additional anionic bacterial constituents (e.g., DNA, RNA, or cell wall components) (2, 5). Amphipathic secondary structures in which residues are segregated into hydrophobic and cationic regions (Fig. 1 A and B) are the hallmark of most AMPs (6). Regardless of their final target of killing, AMPs must interact with the bacterial cytoplasmic membrane, and their amphipathicity is integral to such interactions (1, 2, 7). Additionally, their cationic nature imparts AMPs with some measure of selectivity, because mammalian cell membranes are largely zwitterionic. The precise nature of AMP–membrane interactions remains controversial and actively debated; a variety of mechanisms have been proposed, including the carpet (4), barrel-stave pore (4), toroidal pore (8), and aggregate (9) models. Nevertheless, a considerable number of structure–activity investigations have elucidated how the physicochemical properties of these molecules relate to their biological activities (4, 7, 1022).

Fig. 1.
NMR structure of magainin-2 in DPPC micelles [Protein Data Bank (PDB) ID Code 2MAG]. (A and B) Parallel (A) and perpendicular (B) to its helical axis. (C and D) Similar views of a model structure of ampetoid 1, using backbone [var phi] and ψ angles ...

Although AMPs have been actively studied for decades (2325), they have yet to see widespread clinical use (1). This is due in part to the vulnerability of many peptide therapeutics to rapid in vivo degradation, which dramatically reduces their bioavailability. Nonnatural mimics of AMPs can circumvent the proteolytic susceptibility of peptides while retaining their beneficial features. The short (<40-aa) cationic, linear, α-helical class of AMPs, including the well known magainins (23), are especially amenable to mimicry. β-Peptide mimics of these AMPs have been successfully created that show antibacterial and nonhemolytic in vitro activity (26). Poly-N-substituted glycines (peptoids) comprise another class of peptidomimetics that differ from peptides only in that peptoid side chains are attached to the backbone amide nitrogen rather than to the α-carbon (27), making them protease-resistant (28). More than any of the other peptidomimetic systems under study (29), including β-peptides (26, 30), β-peptoids (31), oligoureas (32), and oligo(phenylene ethynylene)s (33), peptoids (3436) are particularly well suited for AMP mimicry because they are easily synthesized on solid phase (by using conventional peptide synthesis equipment) with access to diverse sequences at relatively low cost (27). Any chemical functionality available as a primary amine can be incorporated via a submonomer synthetic method (27); thus, peptoids are highly and finely tunable.

The poly-N-substituted glycine structure of peptoids precludes both backbone chirality and intrachain hydrogen bonding; nevertheless, peptoids can be driven to form helical secondary structures via a periodic incorporation of bulky, α-chiral side chains (3740). Previous work has shown that incorporation of homochiral side chains can give rise to polyproline type-I-like helices with a periodicity of approximately three monomers per turn and a helical pitch of 6.0–6.7 Å (3739, 41, 42). The threefold periodicity of the peptoid helix facilitates the design of facially amphipathic structures similar to those formed by many AMPs; for example, the trimer repeat (X-Y-Z)n forms a peptoid helix with three faces, composed of X, Y, and Z residues.

Here, we report that the in vitro activities of certain antimicrobial peptoids (“ampetoids”) are similar to those of many AMPs, in that ampetoids exhibit broad-spectrum antibacterial activity and low mammalian cytotoxicity. Structure–activity relationships observed in a library of rationally designed ampetoids are wholly analogous to those that describe many AMPs. Using synchrotron radiation to probe interactions between ampetoids and lipid layers, we have also found that AMP–ampetoid analogy extends to the molecular level.


Initial Antibacterial Activity and Selectivity Screening.

We synthesized 13 ampetoid analogs to determine whether peptoids are affected by structural and sequence modifications in a manner consistent with AMP activities (13, 17, 21, 43). The designs for ampetoids in this library were derived from two antibacterial and selective amphipathic dodecamers, 1 and 2, from previous work (34). Peptoid 1 [H-(NLys-Nspe-Nspe)4-NH2] is composed of 2/3 Nspe, a peptoid analog of Phe, and 1/3 NLys, a peptoid analog of Lys (see Fig. 1 C and D for a model structure of peptoid 1 and Fig. 2 for the structures of peptoid monomers). Peptoid 2 [H-(NLys-Nssb-Nspe)4-NH2] contains 1/3 Ile-like Nssb monomers in place of Nspe. The variant sequences were designed to effect changes in chirality, length, hydrophobicity, charge, and amphipathicity. All compounds were tested for antibacterial activity against representative BSL1 Gram-negative (Escherichia coli JM109) and Gram-positive (Bacillus subtilis BR151) bacterial strains. As an initial measure of selectivity, we determined the lytic activity of the peptoids against human erythrocytes. Table 1 summarizes the sequences we synthesized, the solvent composition at RP-HPLC elution as a relative measure of molecular hydrophobicity, and antibacterial and hemolytic activities [full hemolysis curves are shown in supporting information (SI) Fig. 8]. Ten of the 15 peptoids exhibited low-micromolar MICs against both E. coli and B. subtilis, demonstrating that nonnatural peptoid oligomers can be as active as some antimicrobial peptides (MICs for pexiganan (44)—a selective AMP analog of magainin-2—and the bee-venom AMP melittin (45) are shown in Table 1).

Fig. 2.
Peptoid monomer side-chain structures, with full and shorthand names.
Table 1.
Antibacterial and hemolytic activities of ampetoids and AMPs

We defined the selectivity ratio (SR) (Table 1) as the quotient of the 10% hemolytic dose (HD10) and the E. coli MIC; thus, the SR is an estimate of a compound's tendency to kill bacteria rather than mammalian cells. Ampetoid 1 had an SR of 6.0. As expected, melittin (well known to be cytotoxic) had a low SR of 1.3. Six of 13 peptoid variants were more selective than 1 (i.e., SRs >6), with SRs approaching that of pexiganan (i.e., SR of 1-Pro6 = 20, SR of pexiganan = 24).

Broad-Spectrum Antibacterial Activity Against BSL2 Bacterial Pathogens.

Broad-spectrum activity against a variety of pathogenic bacteria is a hallmark of AMPs. Accordingly, we tested ampetoid 1 against six clinically relevant biosafety-level-2 (BSL2) bacterial pathogens. The MICs for these microorganisms, summarized in Table 2, demonstrated promising activity, comparable with that of pexiganan, which was optimized for pharmaceutical applications (44).

Table 2.
Broad-spectrum activity of 1 against BSL2 CLSI quality control strains, compared to MICs of pexiganan as determined by Ge et al. (44)

Ampetoid Selectivity Evaluated by Using Cultured Mammalian Cells.

We evaluated the biocompatibility of selected oligomers with A549 lung epithelial cells by using the MTS assay (46). Fig. 3 shows that the 10% metabolic inhibitory dose (ID10) and 50% inhibitory dose (ID50) of these selected peptoids compared favorably to that of pexiganan and melittin; ampetoids 1-Pro6 and 1-NHis6,12, with MICs <7 μM, had ID10s 10 and 20 times that of pexiganan. Interestingly, nonselective peptoid 115mer exhibited cytotoxicity no worse than pexiganan.

Fig. 3.
Cytotoxicity data for selected peptoids and comparator peptides against A549 lung epithelial cells.

Structure–Activity Studies.

Structure–activity relationships derived from our library of ampetoids are discussed below, according to the primary physicochemical parameter that was altered. We used circular dichroism (CD) spectroscopy to compare the helicities of the ampetoids by monitoring the intensity of spectral extrema, particularly at 190 and 220 nm (3739, 41, 42, 47). We performed CD in 10 mM aqueous Tris buffer (pH 7.4), alone and in the presence of two types of small unilamellar vesicles (SUVs) composed of model binary lipid mixtures: (i) anionic E. coli membrane-mimetic SUVs [POPE/POPG (7:3 mole ratio)] (48) and (ii) zwitterionic erythrocyte membrane-mimetic SUVs [POPC/cholesterol (CH) (1:1 mole ratio)] (49).


Using enantiomeric side chains (Nrpe in place of Nspe; Fig. 2) (42), we created a left-handed helical analog of 1, called 1enantiomer, as evidenced by its mirror-image CD spectrum compared with that of 1 (Fig. 4A) (47). The activities and selectivities of 1 and 1enantiomer were congruent (Table 1).

Fig. 4.
CD spectra of variants of 1 (A) and variants of 2 (B) in 10 mM Tris buffer (pH 7.4).


We created length variants of 1 ranging from 6 to 15 monomers, all with the same threefold sequence repeat and 1:3 charge-to-length ratio. The shortened variants, 16mer and 19mer (≈12 Å and 18 Å in length, respectively), were both active antimicrobials and more selective than 1 (Table 1). The lengthened variant, 115mer (≈30 Å in length), was substantially more hemolytic (SR = 0.55) and cytotoxic (Fig. 3) than 1 and slightly less antibacterial. The helicities of these compounds in buffer were congruent (SI Fig. 9).


We modulated hydrophobicity independently of length by replacing hydrophobic Ile-like Nssb and Phe-like Nspe monomers in 1 and 2 with bulkier and more hydrophobic Nsmb and Nsna, respectively (Fig. 2). The heightened molecular hydrophobicity of 2-Nsmb2,5,8,11 and 2-Nsna6,12 led to increases in both antibacterial and hemolytic activities (Table 1 and SI Fig. 8B). 1-Nsna6,12, however, showed no enhancement of antibacterial activity relative to 1, but was much more hemolytic (SR = 1.2) (SI Fig. 8B).

We also created two variants of 1 with reduced hydrophobicity. Two evenly spaced hydrophobic Nspe residues in 1 were replaced by NHis, an achiral peptoid analog of histidine (Fig. 2) that is polar yet predominantly uncharged at physiological pH, yielding 1-NHis6,12. This oligomer exhibited substantially decreased in vitro hemolysis (SR >31; Table 1) and decreased cytotoxicity (Fig. 3), with only a slight reduction in antibacterial activity against B. subtilis compared with 1. A second reduced-hydrophobicity variant, 1-Pro6, was created by replacing the Nspe at position 6 with L-proline. Although Pro is well known to destabilize peptide α-helices, 1-Pro6 and 1 were similarly helical in buffer (Fig. 4A) because l-Pro is well accommodated in right-handed type-I-polyproline-like peptoid helices (SI Fig. 10). 1-Pro6 (SR = 24) exhibited less hemolysis and less cytotoxicity than 1, with similar antibacterial activity (Table 1, Fig. 3).


We substituted NLys monomers in 1 with glutamate-like NGlu (Fig. 2) to create 1-NGlu4,10 and 1-NGlu1,4,7,10. Zwitterionic 1-NGlu4,10 had significantly reduced activity against B. subtilis compared with 1 and was inactive against E. coli (Table 1), likely because of the absence of favorable electrostatic interactions with anionic bacterial membranes; however, it was quite hemolytic (SR <0.17). The fully anionic variant, 1-NGlu1,4,7,10 was devoid of both antibacterial and hemolytic activity.


We created a terminally, rather than facially, amphipathic isomer of 1 with block-like architecture (1block), and a scrambled sequence of 2 designed to preclude global amphipathicity (2scrambled) (Table 1). The terminal segregation of cationic and hydrophobic residues in the 1block sequence ensures a strongly amphipathic structure independent of the facial organization of residues along a helix. 1block was slightly less antibacterial and more hemolytic than 1 (SR = 2.6), although it has the same monomer composition. The CD spectra of 2scrambled in buffer and lipid environments (Figs. 4B, ,55B, and and66B) were nearly congruent to those of 2, suggesting that 2scrambled forms a structured helix that, because of its scrambled sequence, has low global amphipathicity relative to 2. Peptoid 2scrambled exhibited antibacterial activity but no hemolysis up to 120 μM.

Fig. 5.
CD spectra of variants of 1 (A) and variants of 2 (B) in 5 mM POPC/cholesterol (1:1) SUVs suspended in 10 mM Tris buffer (pH 7.4).
Fig. 6.
CD spectra of variants of 1 (A) and variants of 2 (B) in 5 mM POPE/POPG (7:3) SUVs suspended in 10 mM aqueous Tris buffer (pH 7.4).

X-Ray Reflectivity Studies of Ampetoid Orientation in Lipid Layers.

We carried out liquid (aqueous buffer) surface specular x-ray reflectivity (XR) studies using synchrotron radiation to investigate the membrane orientation and depth of penetration of ampetoid 1 in a model lipid layer that mimics the outer leaflet of the cell membrane (Fig. 7) (50, 51). X-rays reflected off of the monolayer yielded an electron density profile perpendicular to the interface, allowing determination of the layer thickness and the presence and orientation of added molecules. The experimental data were represented as a stack of slabs, each with a uniform thickness, electron density, and interface roughness (52).

Fig. 7.
X-ray reflectivity data and corresponding fit for DPPG monolayer before (circles; Inset Upper) and after (squares; Inset Lower) peptoid 1 was injected into the subphase.

XR data for a pure DPPG (anionic) film (Fig. 7, circles) fit well with a two-slab model, yielding a hydrocarbon tail density (ρts) of 0.99 and a hydrocarbon tail slab thickness (L1) of 17.9 Å as well as a head-group electron density (ρhs) of 1.54 and head-group slab thickness (L2) of 5.7 Å. These data are in good agreement with previous DPPG monolayer x-ray work (51). The XR profile changed dramatically after 1 was introduced (Fig. 7, squares) and fit a four-slab model corresponding to (i) lipid tails without 1ts = 0.96, L1 = 12.1 Å), (ii) lipid tails with partial insertion of 1t+ps = 1.05, L2 = 2.8 Å), (iii) lipid headgroups with 1 fully inserted (ρh+ps = 1.33, L2 = 7.0 Å), and (iv) peptoid 1 alone, protruding beyond the DPPG head groups (ρps = 1.16, L2 = 3.6 Å). This electron density profile was consistent with insertion of 1 through the lipid head groups and partially into the lipid tail region. Assuming that 1 retained its helical structure in the monolayer, the data suggest that 1 inserted at an angle of ≈56° between the interface normal and the long helical axis of the peptoid.


Previous work has shown that peptoids can be used to create compounds with antimicrobial activities similar to AMPs (34). In this study, we exploited the facile synthesis and high propensity for helix formation of certain peptoids to create and study a library of ampetoid variants that suggests strong functional and mechanistic analogy between ampetoids and AMPs and that demonstrates the ampetoids' potential for development into clinically useful antibiotics.

The equivalent activities of 1 and 1enantiomer demonstrate that ampetoid mechanism does not depend on overall handedness or on stereospecific interactions with receptors or enzymes, an attribute that has also been observed for many AMPs (43). Evidence that ampetoids interact with and insert into membranes is provided by x-ray reflectivity studies (Fig. 7). Furthermore, the depth of insertion of peptoid 1—through the head groups and partially into the lipid tails—demonstrates that 1 interacts simultaneously with hydrophobic and hydrophilic lipid moieties; thus, as with AMPs, the amphipathic structure of 1 is integral to its interactions with membranes. The orientation of 1 at an angle of ≈56° to the interface normal suggests that 1 does not operate through a barrel-stave mechanism, because that would require a transmembrane configuration (5). Although it cannot be concluded that 1 exhibits identical mechanistic behavior to natural, α-helical antimicrobial peptides, these x-ray results demonstrate ampetoid-lipid interactions consistent with those seen for AMPs such as pexiganan (53) and LL-37 (51).

We have found that ampetoid activity also adheres to trends relating structure and function in a manner analogous to AMPs—physicochemical properties of selective ampetoids are consistent with those of selective AMPs and nonselective peptoids exhibit close similarities to cytotoxic AMPs. Numerous structure–activity studies of a wide variety of AMPs have delineated the physicochemical characteristics that give rise to selective antibacterial activity or nonselective cytotoxicity. Regardless of structural class (i.e., α-helix, β-sheet, loop, or extended), nonselective AMPs typically (i) are very hydrophobic, such that their interactions with membranes are governed primarily by the hydrophobic effect (1015), and (ii) have a well defined amphipathic structure (4, 11, 1417). In contrast, the antibacterial activity of selective AMPs depends on (i) high net cationic charge (1214, 18) (although excessive cationic charge can also lead to hemolytic activity) (18, 19) and (ii) only moderate hydrophobicity (16, 19, 21, 22). Perhaps counterintuitively, a well defined amphipathic structure is not necessary for selective antimicrobial activity (7, 16); destabilization of AMP secondary structure often leads to improvements in selectivity (15, 20).

In our library, 1-NGlu4,10, 115mer, 1-Nsna6,12, and 1block were all less selective than peptoid 1 (SRs <6.0) (Table 1). Consistent with properties of nonselective AMPs, these compounds are all either more hydrophobic (according to RP-HPLC elution time) and/or less charged (1-NGlu4,10) than 1. Furthermore, they are all as (or more) helical than 1 in erythrocyte-mimetic POPC/CH SUVs (Fig. 5), indicative of their well defined membrane-bound amphipathic structures in that lipid environment.

In contrast, 19mer, 1-NHis6,12, and 1-Pro6 all had antibacterial activities comparable to 1, but enhanced selectivities (SRs >6.0). These peptoids are all more hydrophilic than 1, and all have a net charge of at least +3. Thus, ampetoids are selectively active provided they have a net positive charge and are sufficiently but not excessively hydrophobic, consistent with observations of selective AMPs (4, 11). Additionally, retention of antibacterial activity and enhancement of selectivity in the shortened 19mer is analogous to behavior observed in studies of truncated AMPs (10, 16, 21, 22).

The effect of length on ampetoid potency and selectivity can largely be attributed to differences in hydrophobicity (RP-HPLC retention time, Table 1), which increases proportionally with length. Because the 12mer 1 was more antibacterial than longer and shorter analogs, these results suggest the existence of an optimal hydrophobicity at which antibacterial activity is maximized; added hydrophobicity increases only hemolytic activity. This conclusion is also supported by the hydrophobicity variants, because moderately hydrophobic 2-Nsmb2,5,8,11 and 2-Nsna6,12 are both more antibacterial and more hemolytic than 2, whereas the strongly hydrophobic 1-Nsna6,12 (Table 1 and SI Fig. 8B) shows enhancement of hemolytic but not antibacterial activity relative to 1 (additional discussion in SI Text).

Although they are more hemolytic than 2, peptoids 2-Nsna6,12 and 2-Nsmb2,5,8,11 are still more selective (SRs of 7.6 and 16, respectively) than 1, and their physicochemical properties are consistent with this finding. They are highly charged (+4), moderately hydrophobic, and exhibit CD spectra consistent with low helicity (Figs. 4B, ,55B, and and66B). In general, CD spectra in erythrocyte-mimetic POPC/CH SUVs (Fig. 5) reveal a correlation between hemolytic activity and helicity. 1-Nsna6,12, and 1-NGlu4,10 are the most hemolytic ampetoids (Table 1 and SI Fig. 8) and are the most helical in POPC/CH SUVs (Fig. 5A). Conversely, 1-NHis6,12, 2, 2-Nsmb6,12, and 2scrambled are less hemolytic (Table 1 and SI Fig. 8), and less helical (Fig. 5) than 1.

CD spectra of ampetoids in bacteria-mimetic POPE/POPG SUVs (Fig. 6), however, show that the extent of helicity is poorly correlated with antimicrobial activity. For example, 1-Nsna6,12 is less helical than 1 (Fig. 6A), but the two compounds have similar antibacterial potencies (Table 1). Also, 2-Nsmb2,5,8,11 is similarly helical to 2 (Fig. 6B), yet 2-Nsmb2,5,8,11 is much more antibacterial (Table 1). Together, these results suggest that a defined helical structure is important only as a means to organize an amphipathic structure and is required for hemolytic, but not antibacterial, activity. Based on this conclusion, 1block, which has a well defined terminally amphipathic structure independent of its helicity, should be hemolytic, and it is (SR = 2.6). Likewise, a helical but poorly amphipathic compound such as 2scrambled should be antimicrobial, but selective; indeed, 2scrambled kills bacteria, but exhibits no hemolysis up to 120 μM.

In summary, these results suggest that antibacterial activity among AMPs and ampetoids alike depends on moderate hydrophobicity and net cationic charge, whereas hemolytic activity is associated primarily with high hydrophobicity and a strongly amphipathic structure, regardless of helical content. The relationships between structure and function in ampetoids are empirically analogous to those in AMPs. X-ray reflectivity studies, which show that ampetoid 1 is membrane-active and adopts a stable membrane-bound orientation, demonstrate molecular-level analogy between AMPs and ampetoids.

As both structural and functional mimics of AMPs, ampetoids act via mechanisms similar to those that have been refined over the course of evolution; thus, much of the knowledge amassed in the AMP field should be transferable to the development of peptoid antibiotics. Peptoids have greater potential than peptides to be used as pharmaceuticals and in biomaterials because of their improved stability, bioavailability, and highly tunable side-chain chemistry. Because peptides' potential for toxicity is a major obstacle limiting their clinical use (1), ampetoids' low cytotoxicity observed here relative to the AMP pexiganan further emphasizes their therapeutic potential. The results reported herein will aid in the rational design and optimization of ampetoids and other nonnatural oligomers as antimicrobials in the future.

Materials and Methods

Synthesis and Purification.

Peptoids were synthesized on an ABI 433A peptide synthesizer or a parallel synthesis robot on Rink amide resin according to the submonomer method (27, 54). Briefly, the amide on the nascent chain is bromoacetylated, followed by SN2 displacement of bromide by a primary amine to form the side chain (see SI Text for amines used). Peptides were synthesized by using standard Fmoc chemistry. After synthesis, oligomers were cleaved and deprotected in trifluoroacetic acid (TFA)/triisopropylsilane/water (95:2.5:2.5 by vol.) for 10 min. Compounds were purified to >97% homogeneity by RP-HPLC on a C18 column with a linear acetonitrile/water (0.1% TFA) gradient. Mass spectrometry was used to confirm the molecular weight of the purified product.

SUV Preparation.

Lipid mixtures, either POPE/POPG (7:3) or POPC/CH (1:1), were dissolved in chloroform, dried under N2, and lyophilized overnight. The lipid film was hydrated with 10 mM Tris·HCl (pH 7.4) at 40°C for 1 h. The resulting multilamellar vesicle suspension was vortexed and then sonicated at 40°C until the solution clarified to make SUVs, which were used within 6 h.

CD Spectroscopy.

CD measurements were performed on a Jasco 715 spectropolarimeter, by using a quartz cylindrical cell (path length = 0.02 cm), with 50 μM peptoid in 10 mM Tris·HCl (pH 7.4) and 5 mM lipids when SUVs were used. Scans were conducted at 100 nm/min between 185 and 280 nm with 0.2-nm data pitch, 1-nm bandwidth, 2-s response, 100-mdeg sensitivity, and 40 accumulations.

Antibacterial Assays.

MICs were determined according to CLSI M7-A6 protocols in a 96-well microtiter plate. In test wells, 50 μl of bacterial inoculum (1 × 106 CFU/ml) in Mueller–Hinton broth (MHB) was added to 50 μl of peptoid solution in MHB (prepared by 1:2 serial dilutions). Positive controls contained 50 μl of inoculum and 50 μl of MHB without peptoid. The MIC was defined as the lowest concentration of peptoid that completely inhibited bacterial growth after incubation at 35°C for 16 h. MIC values reported were reproducible between three independent experimental replicates, each consisting of two parallel trials. Broad-spectrum MIC testing was performed against BSL2 pathogens by Nova Biologicals.

Hemolysis Assays.

Erythrocytes were isolated from freshly drawn, heparanized human blood and resuspended to 20 vol% in PBS (pH 7.4). In a 96-well microtiter plate, 100 μl of erythrocyte suspension was added to 100 μl of peptoid solution in PBS (prepared by 1:2 serial dilutions) or 100 μl of PBS in the case of negative controls. One-hundred percent hemolysis wells contained 100 μl of blood cell suspension with 100 μl of 0.2 vol% Triton X-100. The plate was incubated for 1 h at 37°C, and then each well was diluted with 150 μl of PBS. The plate was then centrifuged at 1,200 × g for 15 min, 100 μl of the supernatant from each well was transferred to a fresh microtiter plate, and A350 was measured. Percentage of hemolysis was determined as (AA0)/(AtotalA0) × 100, where A is the absorbance of the test well, A0 the absorbance of the negative controls, and Atotal the absorbance of 100% hemolysis wells, all at 350 nm.

MTS Assays.

A549 carcinoma-derived lung epithelial cells (CCL-185; American Type Culture Collection) were cultured in Ham's F12K media. A peptoid solution plate (100 μl per well) was prepared by serial dilution of aqueous peptoid stocks in media. Peptoid solutions were transferred to a 96-well plate of day-old cell monolayers containing 100 μl per well media with ≈5,000 cells per well. MTS reagent (Promega) (40 μl per well) was added to each well, and the plate was incubated at 37°C for 3 h, after which absorbance at 490 nm was read. Percentage of inhibition = [1 − (AAtestblank)/(AcontrolAblank)] × 100, where A is the absorbance of the test well and Acontrol the average absorbance of wells with cells exposed to media and MTS (no peptoid). Atestblank (media, MTS, and peptoid) and Ablank (media and MTS) were background absorbances measured in the absence of cells. The average of six replicates are reported.

Specular X-Ray Reflectivity.

XR experiments were carried out at the 9-ID beamline at the Advanced Photon Source, Argonne National Laboratory (Argonne, IL). The custom-built Langmuir trough was mounted in a He-filled, sealed canister and equipped with a moveable single barrier. The surface pressure was measured by using a Wilhelmy plate. Constant-pressure insertion experiments were performed at room temperature on Dulbecco's PBS (D-PBS) subphase. DPPG (Avanti Polar Lipids) was dissolved to a known concentration in 65/35 (vol/vol) chloroform/methanol, then spread at the air–buffer interface by using a glass syringe; organic solvent was allowed to evaporate for 10 min. The monolayer was compressed to the surface pressure thought to occur in cell membranes, 30 mN/m, and XR was performed on the pure lipid layer. Then, peptoid 1 dissolved in D-PBS was injected into the subphase to a total concentration of 6.3 μM (twice the E. coli MIC), and allowed to insert for ≈45 min, after which XR measurements were again collected. The x-ray reflectivity (XR) profile was determined by the Fourier transform of the gradient of the electron density perpendicular to the interface. XR measurements were carried out over a range of angles corresponding to qz values of ≈0–0.6 Å−1, where qz = (4π/λ)sin(α), λ is the wavelength, and α is the angle.

Supplementary Material

Supporting Information:


We thank Dr. David M. Steinhorn for hemolysis assays, Chao Liu and Ivan Kuzmenko for x-ray studies, Pamela Focia for model structure of peptoid 1, Joanna Burdette for cell culture, and Phillip B. Messersmith for helpful comments. N.P.C. was supported by a Department of Homeland Security Fellowship. J.A.P. was supported by National Institutes of Health (NIH) National Research Service Award Molecular Biophysics Training Grant 5 T32 GM08382-10. A.M.C. was supported by a 3M graduate research fellowship. A.E.B. acknowledges support from Northwestern University's Institute for Bioengineering and Nanoscience in Advanced Medicine, the Dreyfus Foundation, a DuPont Young Investigator award, NIH Grants 1 R01 HL67984 and 1 R01 AI072666, and National Science Foundation Grant L474530-A2//CHE-0404704. Work at the Molecular Foundry was supported by the Office of Science, Office of Basic Energy Sciences, of the U.S. Department of Energy under Contract DE-AC02-05CH11231.


The authors declare no conflict of interest.

This article is a PNAS Direct Submission.

This article contains supporting information online at www.pnas.org/cgi/content/full/0708254105/DC1.


1. Hancock RE, Sahl HG. Nat Biotechnol. 2006;24:1551–1557. [PubMed]
2. Jenssen H, Hammill P, Hancock REW. Clin Microbiol Rev. 2006;19:491–511. [PMC free article] [PubMed]
3. Peschel A, Sahl H-G. Nat Rev Microbiol. 2006;4:529–536. [PubMed]
4. Shai Y. Biopolymers (Peptide Sci) 2002;66:236–248. [PubMed]
5. Brogden KA. Nat Rev Microbiol. 2005;3:238–250. [PubMed]
6. Zasloff M. Nature. 2002;415:389–395. [PubMed]
7. Pag U, Oedenkoven M, Papo N, Oren Z, Shai Y, Sahl HG. J Antimicrob Chemother. 2004;53:230–239. [PubMed]
8. Matsuzaki K. Biochim Biophys Acta. 1998;1376:391–400. [PubMed]
9. Wu M, Maier E, Benz R, Hancock REW. Biochemstry. 1999;38:7235–7242. [PubMed]
10. Blondelle SE, Houghten RA. Biochemistry. 1992;31:12688–12694. [PubMed]
11. Chen Y, Mant CT, Farmer SW, Hancock REW, Vasil ML, Hodges RS. J Biol Chem. 2005;280:12316–12329. [PMC free article] [PubMed]
12. Frecer V. Bioorg Med Chem. 2006;14:6065–6074. [PubMed]
13. Maloy WL, Kari UP. Biopolymers. 1995;37:105–122. [PubMed]
14. Park Y, Kim HN, Park SN, Jang SH, Choi CH, Lim HT, Hahm KS. Biotechnol Lett. 2004;26:493–498. [PubMed]
15. Shin SY, Lee SH, Yang ST, Park EJ, Lee DG, Lee MK, Eom SH, Song WK, Kim Y, Hahm KS, et al. J Peptide Res. 2001;58:504–514. [PubMed]
16. Bessalle R, Gorea A, Shalit I, Metzger JW, Dass C, Desiderio DM, Fridkin M. J Med Chem. 1993;36:1203–1209. [PubMed]
17. Shai Y, Oren Z. Peptides. 2001;22:1629–1641. [PubMed]
18. Bessalle R, Haas H, Goria A, Shalit I, Fridkin M. Antimicrob Agents Chemother. 1992;36:313–317. [PMC free article] [PubMed]
19. Dathe M, Nikolenko H, Meyer J, Beyermann M, Bienert M. FEBS Lett. 2001;501:146–150. [PubMed]
20. Song YM, Yang ST, Lim SS, Kim Y, Hahm KS, Kim JI, Shin SY. Biochem Biophys Res Commun. 2004;314:615–621. [PubMed]
21. Feder R, Dagan A, Mor A. J Biol Chem. 2000;275:4230–4238. [PubMed]
22. Kustanovich I, Shalev DE, Mikhlin M, Gaidukov L, Mor A. J Biol Chem. 2002;277:16941–16951. [PubMed]
23. Zasloff M. Proc Natl Acad Sci USA. 1987;84:5449–5453. [PMC free article] [PubMed]
24. Ganz T, Selsted ME, Szklarek D, Harwig SS, Daher K, Bainton DF, Lehrer RI. J Clin Invest. 1985;76:1427–1435. [PMC free article] [PubMed]
25. Payne JW, Jakes R, Hartley BS. Biochem J. 1970;117:757–766. [PMC free article] [PubMed]
26. Porter EA, Weisblum B, Gellman SH. J Am Chem Soc. 2002;124:7324–7330. [PubMed]
27. Zuckermann RN, Kerr JM, Kent SBH, Moos WH. J Am Chem Soc. 1992;114:10646–10647.
28. Miller SM, Simon RJ, Ng S, Zuckermann RN, Kerr JM, Moos WH. Bioorg Med Chem Lett. 1994;4:2657–2662.
29. Gante J. Angew Chem Int Ed. 1994;33:1699–1720.
30. Seebach D, Overhand M, Kuhnle FNM, Martinoni B, Oberer L, Hommel U, Widmer H. Helv Chim Acta. 1996;79:913–941.
31. Hamper BC, Kolodziej SA, Scates AM, Smith RG, Cortez E. J Org Chem. 1998;63:708–718. [PubMed]
32. Violette A, Averlant-Petit MC, Semetey V, Hemmerlin C, Casimir R, Graff R, Marraud M, Briand J-P, Rognan D, Guichard G. J Am Chem Soc. 2005;127:2156–2164. [PubMed]
33. Arnt L, Tew GN. J Am Chem Soc. 2002;124:7664–7665. [PubMed]
34. Patch JA, Barron AE. J Am Chem Soc. 2003;125:12092–12093. [PubMed]
35. Seurynck SL, Patch JA, Barron AE. Chem Biol. 2005;12:77–88. [PubMed]
36. Statz AR, Meagher RJ, Barron AE, Messersmith PB. J Am Chem Soc. 2005;127:7972–7973. [PubMed]
37. Wu CW, Sanborn TJ, Huang K, Zuckermann RN, Barron AE. J Am Chem Soc. 2001;123:6778–6784. [PubMed]
38. Wu CW, Kirshenbaum K, Sanborn TJ, Patch JA, Huang K, Dill KA, Zuckermann RN, Barron AE. J Am Chem Soc. 2003;125:13525–13530. [PubMed]
39. Armand P, Kirshenbaum K, Falicov A, Dunbrack RL, Dill KA, Zuckermann RN, Cohen FE. Folding Design. 1997;2:369–375. [PubMed]
40. Sanborn TJ, Wu CW, Zuckerman RN, Barron AE. Biopolymers. 2002;63:12–20. [PubMed]
41. Armand P, Kirshenbaum K, Goldsmith RA, Farr-Jones S, Barron AE, Truong KTV, Dill KA, Mierke DF, Cohen FE, Zuckermann RN, et al. Proc Natl Acad Sci USA. 1998;95:4309–4314. [PMC free article] [PubMed]
42. Kirshenbaum K, Barron AE, Goldsmith RA, Armand P, Bradley EK, Truong KTV, Dill KA, Cohen FE, Zuckermann RN. Proc Natl Acad Sci USA. 1998;95:4303–4308. [PMC free article] [PubMed]
43. Bessalle R, Kapitkovsky A, Gorea A, Shalit I, Fridkin M. FEBS Lett. 1990;274:151–155. [PubMed]
44. Ge YG, MacDonald DL, Holroyd KJ, Thornsberry C, Wexler H, Zasloff M. Antimicrob Agents Chemother. 1999;43:782–788. [PMC free article] [PubMed]
45. Fennell JF, Shipman WH, Cole LJ. Proc Soc Exp Biol Med. 1968;127:707–710. [PubMed]
46. Mosmann T. J Immunol Methods. 1983;65:55–63. [PubMed]
47. Wu CW, Sanborn TJ, Zuckermann RN, Barron AE. J Am Chem Soc. 2001;123:2958–2963. [PubMed]
48. Kruijff BD, Killian JA, Rietveld AG, Kusters R. In: Lipid Polymorphism and Membrane Properties. Epand RM, editor. Vol 44. London: Academic; 1997.
49. Dodge JT, Phillips GB. J Lipid Res. 1967;8:667–675. [PubMed]
50. Gidalevitz D, Ishitsuka YJ, Muresan AS, Konovalov O, Waring AJ, Lehrer RI, Lee KYC. Proc Natl Acad Sci USA. 2003;100:6302–6307. [PMC free article] [PubMed]
51. Neville F, Cahuzac M, Konovalov O, Ishitsuka Y, Lee KYC, Kuzmenko I, Kale GM, Gidalevitz D. Biophys J. 2006;90:1275–1287. [PMC free article] [PubMed]
52. Als-Nielsen J, McMorrow D. Elements of Modern X-ray Physics. Vol 43. New York: Wiley; 2002.
53. Ramamoorthy A, Thennarasu S, Lee DK, Tan A, Maloy L. Biophys J. 2006;91:206–216. [PMC free article] [PubMed]
54. Burkoth TS, Fafarman AT, Charych DH, Connolly MD, Zuckermann RN. J Am Chem Soc. 2003;125:8841–8845. [PubMed]

Articles from Proceedings of the National Academy of Sciences of the United States of America are provided here courtesy of National Academy of Sciences
PubReader format: click here to try