Move the side chain from the alpha-carbon to the backbone nitrogen and the molecule stops being a peptide. Peptoids are poly-N-substituted glycines: structurally similar to peptides, but with the side chains attached to the nitrogen rather than the alpha-carbon [1]. The atom count barely moves. Three properties of the chain change completely, and the rest of this article is consequences of those three.
This is a different axis of modification from D-amino acid substitution and the other alpha-carbon edits. Those keep the backbone amide N-H and keep a stereocentre at the alpha-carbon, and change their geometry. The peptoid change deletes both.
What moving the side chain actually does
The backbone amide becomes tertiary and loses its N-H. The achiral peptoid backbone lacks hydrogen-bond donors, which effectively prevents formation of the regular intrachain hydrogen bonds that stabilise peptide alpha-helical structures [2]; more generally there is no backbone NH available to form CO…HN hydrogen bonds at all [3]. Whatever holds a folded peptoid in a defined conformation, it is not the hydrogen-bond network you would draw for a peptide, and neither the alpha-helix nor the beta-sheet is available on the usual terms.
The backbone also loses its stereocentre. Peptoids have no backbone chirality [2, 3]; the alpha position of each residue is a glycine methylene. Chirality, where a design needs it, lives entirely in the side chain — (S)-N-(1-phenylethyl)glycine and (S)-N-(1-cyclohexylethyl)glycine are the standard examples — on a carbon that is not the electrophilic centre in the coupling step. The epimerisation that constrains coupling reagent choice, base loading and temperature in Fmoc peptide synthesis has no substrate here.
Third, because the amide is tertiary, the strong trans preference of a secondary peptide amide is gone. Both the cis and the trans conformations of peptoid backbone amides can be significantly populated, so a peptoid is a mixture of configurational isomers whose number doubles with every additional monomer residue [4]. Note precisely what that does and does not say. In an NMR study of N-acetylated peptoid monomers, dipeptoids and tripeptoids, resonance intensities indicated the cis conformation was more populated than is generally the case for the analogous secondary amide bond to proline in acyclic peptides — but the rate of interconversion in those peptoids was generally slower than rotation about amide bonds to proline, and on the inversion-magnetisation-transfer timescale only rotation about the amide bond to the C-terminal peptoid residue was observable [4]. Conformational mobility here is a statement about isomer populations. It is not a licence to assume fast exchange.
Submonomer synthesis: a two-step cycle and no monomer library
The synthesis is what makes the class cheap to explore. The backbone is grown one residue at a time by a two-step cycle. First, bromoacetic acid activated in situ with N,N'-diisopropylcarbodiimide acylates the resin-bound secondary amine. Second, nucleophilic displacement of the bromide by a primary amine introduces the side chain and regenerates a secondary amine for the next round [1]. The cycle is iterated until the desired chain length is reached, its coupling efficiency routinely exceeds 98%, and that efficiency supports a simple manual protocol giving polypeptoids up to 50-mers in excellent yields [1]. The route was reported in 1992 [5] and peptoid chemistry has been built on it since [6].
The consequence that matters for library design is that there is no monomer to make. Side-chain identity is set by whichever primary amine goes into the displacement step, so hundreds of readily available primary amines can be incorporated directly [1]. You are not preparing or buying a protected, activated, enantiopure building block for every new side chain, which is why the accessible sequence space is enormous [7]. The submonomer method is fully compatible with Fmoc peptide synthesis conditions and is readily automated on existing automated peptide synthesisers [7]. Microwave irradiation reduces the reaction time further, to a coupling requirement of about one minute per residue, with purity and yields as good as or better than standard conditions [8].
Where it gets harder is where the displacement step gets harder. Accessing many of the most useful and desirable side chains requires special consideration, including most amino-acid-like side chains, several important non-natural monomer classes, and peptoid conjugates and peptide-peptoid hybrids [7]. One documented alternative avoids the displacement entirely: N-alkylated arylsulfonamides have been used to prepare side-chain-protected N-substituted glycines from commercially available amines in four straightforward steps, and those N-substituted N-arylsulfonylglycines were then used as preformed monomers in solid-phase synthesis to introduce functionalised side chains into peptoid oligomers and peptide-peptoid hybrids [9]. Reach for a preformed-monomer route when the amine you want carries functionality that will not survive the displacement conditions.
Helices built by sterics, not by hydrogen bonds
Peptoids fold anyway. Oligomers N-substituted with alpha-chiral aromatic side chains form stable, chiral, polyproline-like helices in either organic or aqueous solution at chain lengths as short as five residues, and the adoption of that chiral secondary structure is driven primarily by the steric influence of the bulky chiral side chains [2]. Because the driver is side-chain sterics, the design rules are composition rules rather than hydrogen-bonding rules. A family of structured peptoids up to 36 residues long, built by the solid-phase protocol, included sequences with stable secondary structure despite the achirality of the backbone and its lack of hydrogen bond donors; differential scanning calorimetry and CD on those sequences showed the structure is highly stable and that unfolding is reversible and cooperative, with thermodynamic parameters similar to those obtained for the alpha-helix-to-coil transitions of peptides [10]. Cooperative, reversible unfolding without a backbone hydrogen-bond network is the single most useful fact about this class.
The geometry came from crystallography. The X-ray structure of an N-(1-cyclohexylethyl)glycine pentamer, the first reported for any peptoid, shows a helix with cis-amide bonds, approximately three residues per turn and a pitch of approximately 6.7 A, with backbone dihedral angles similar to those of a polyproline type I peptide helix [11]. CD spectra of analogous hexamers, dodecamers and pentadecamers built entirely from (S)-N-(1-cyclohexylethyl)glycine or (S)-N-(sec-butyl)glycine also carry polyproline type I features, which is the argument that the crystal conformer represents the major solution conformer [11]. In solution the same architecture had been read off a peptoid pentamer of five para-substituted (S)-N-(1-phenylethyl)glycine residues by NMR: the major species in methanol is a right-handed helix with cis-amide bonds, three residues per turn, pitch approximately 6 A [12]. Peptoids with alpha-chiral aromatic and alpha-chiral aliphatic side chains appear to adopt fundamentally similar helical structures in solution despite distinct CD spectra [11].
That last clause is worth holding onto, because CD is the most-used and most-overread peptoid measurement. Short peptoid helices of this class exhibit intense CD spectra that closely resemble those of peptide alpha-helices [2]. They are not alpha-helices, and the resemblance licenses nothing about backbone dihedrals. Settling what you actually have needs a crystal structure or an NMR structure [11, 12] and, if aggregation is a candidate explanation, a concentration series.
The composition rules themselves were mapped by CD across more than 30 heterooligomers carrying mixed chiral and achiral side chains. For that family, a composition of at least 50% alpha-chiral aromatic residues was necessary for stable helical structure in hexameric sequences; placing an alpha-chiral aromatic residue at the carboxy terminus stabilised the short helices, as did an aromatic face patterned by positioning aromatic residues with three-fold periodicity in the sequence; and extending chain length beyond 12 to 15 residues minimised the impact of placement, but not of percentage, on overall helical stability [2].
Chain length carries a separate threshold in homooligomers. In a CD series of organosoluble (R)-N-(1-phenylethyl)glycine homooligomers from 3 to 20 residues, spectral features evolved strikingly between 4 and 12 residues and then stopped changing between 13 and 20, which the authors read as a minimal chain length of about 13 residues for stable poly(Nrpe) helices; no concentration dependence was observed for an Nrpe hexamer, evidence that these helices remain monomeric in solution [13]. The two numbers are not in conflict. One is a composition requirement measured on mixed-side-chain hexamers, the other a length requirement measured on a single-monomer series, and they should be quoted with their experiments attached.
That concentration check earns its place, because peptoid order also arrives above the single-chain level. A single-chain amphiphilic 36-mer assembled from only three different commercially available monomers — hydrophobic, cationic and anionic — folds into a highly ordered nanosheet in which the hydrophobic phenylethyl side chains are buried in the sheet core while the ionic amine and carboxyl side chains align on the hydrophilic faces [1]. Periodic amphiphilicity is easy to build accidentally while patterning an aromatic face onto a helix. If your sequence has it, a CD or NMR spectrum recorded in aqueous buffer may be reporting on an assembly rather than on a folded monomer, and only a concentration series separates the two.
Controlling the rotamer equilibrium
Substantial cis/trans amide isomerisation gives rise to conformational heterogeneity, and that heterogeneity is the practical obstacle to a discretely folded peptoid [14]. Two side-chain-level tools have been developed against it, and they push in opposite directions.
N-aryl side chains enforce trans. Aniline derivatives and bromoacetic acid are used in the solid-phase synthesis of sequence-specific N-aryl glycine oligomers, and quantum-mechanical calculations substantiate a strong energetic preference for trans-amide bonds where an anilide group is present, a preference verified by X-ray crystallographic analysis and solution NMR [14]. Molecular modelling indicates those linear oligomers can adopt helices resembling polyproline type II, and high-resolution structures of macrocycles incorporating both N-alkyl and N-aryl glycine units confirm that trans-amides can be directed specifically to the N-aryl positions [14]. Note the reversal relative to the alpha-chiral aliphatic and aromatic helices above: that motif is all-cis, this one is trans.
The complementary approach tunes local noncovalent interactions rather than sterics alone. A survey of monomeric peptoid model systems produced general strategies using n-to-pi*, steric and hydrogen-bonding interactions to regulate backbone amide rotameric equilibria, with the required chemical functionality typically confined to the peptoid side chains, preserving the side-chain N-alpha-carbon chirality known to engender peptoid structure, and fully compatible with standard peptoid synthesis techniques [15]. Extended to oligomers, that logic produced peptoids containing 1-naphthylethyl side chains which effectively eliminate trans-amide rotamers from the backbone, reported as the most conformationally homogeneous class of peptoid structures then available in terms of amide rotamerism [15]. The same survey found that solvent affects these side-chain-backbone interactions [15], so a rotamer ratio measured in one solvent does not transfer to another without re-measurement.
Protease resistance, and what a protease assay licenses
The practical consequence of the backbone change shows up in direct comparisons. In a side-by-side evaluation of peptoid derivatives of short lipophilic antimicrobial peptides, trypsin rapidly degraded one of the evaluated peptides while having no effect on the comparable peptoids [16]. At the level of the class, review coverage of the field describes peptoids as highly active in biological systems while resistant to proteolytic decay [6].
Two cautions before that becomes a general claim in your own write-up. A single-protease result is a single-protease result: it is one enzyme with one specificity acting in vitro, and it says nothing about a broad-specificity aminopeptidase, about a crude lysate, or about non-enzymatic hydrolysis. And the same comparison series shows the substitution is not free. In those in vitro potency assays peptides generally retained better antimicrobial activity than their peptoid counterparts, although certain peptoids were more effective against Gram-negative bacteria; peptoid MG10 gave a minimum inhibitory concentration of 6.3 micrograms per millilitre against Pseudomonas aeruginosa, compared with 100 micrograms per millilitre for its peptide counterpart [16]. Stability was gained across the series; potency moved in both directions depending on the organism.
Analysis
Nothing exotic is required. Sequence-specific peptoid oligomers encoded by solid-phase submonomer synthesis are purified by high-performance liquid chromatography and characterised by electrospray ionisation mass spectrometry, with matrix-assisted laser desorption/ionisation mass spectrometry used on the assembled products [17]. Reversed-phase retention tracks the side-chain hydrophobicity you designed in, and accurate mass confirms composition.
What that pair does not give you is order. Mass reports the sum of the side chains installed, not the sequence in which they were installed. Because every residue goes on by the same two-step chemistry, the characteristic submonomer failure mode is an incomplete cycle, and the resulting deletion product differs from the target by one residue's mass and is resolvable; a transposition of two different side chains is isobaric with the target and is not. Sequence-level confirmation needs fragmentation or an orthogonal synthesis of the suspected impurity.
Conformational heterogeneity is the other analytical nuisance, and it is intrinsic rather than a sign of a bad prep. A peptoid populating multiple slowly interconverting rotamers is a mixture on the NMR timescale [4]. In the pentamer NMR work above, multiple configurational isomers were observed and, because of extensive signal overlap, only the major all-cis isomer could be examined in detail [12]. Plan for that: a peptoid NMR assignment is usually an assignment of the dominant rotamer, and the population of the rest is part of the result, not noise to be subtracted.
What the substitution costs
The flexibility that makes peptoids easy to build is the same flexibility that costs binding. The conformational flexibility of peptoids — traced to the lack of backbone chirality, the absence of backbone NH donors able to form CO…HN hydrogen bonds, and the facile cis-trans isomerisation of the tertiary amide bonds — is linked to substantial entropic loss during binding to a biomacromolecule, and most peptoid ligands discovered so far possess moderate affinity towards their biological targets [3]. That is the honest summary of the trade, and it explains why so much of the design literature above — bulky alpha-chiral side chains, N-aryl residues, n-to-pi* tuning — is directed at buying back conformational order that a peptide gets for free from its own backbone.
The bench decision follows from that. Ask whether the interaction you are mimicking is read through the backbone. If it is — a strand docking into a groove by main-chain hydrogen bonds, a recognition motif that reads amide NH donors — the peptoid deletes the feature the interaction depends on, and the substitution is not conservative in any sense. If the interaction is read through side chains presented on a rigid scaffold, a peptoid helix is a plausible display, and the synthetic economy is real: no protected monomer to prepare per side chain, no alpha-carbon to epimerise, and hundreds of primary amines to draw on [1].