Backbone Modifications and Peptidomimetics: What Each Change Buys and What It Costs

Proteases recognise their substrates through both backbone and side-chain contacts, but the bond they hydrolyse is a backbone amide. That is the opening backbone modification exploits: change the repeating unit and the bond the enzyme is built to cleave is no longer presented to it, whatever the side chains offer. Resistance still has to be demonstrated for the enzyme in question rather than assumed. The difficulty is that the same backbone also supplies the hydrogen bonds that hold secondary structure together and much of what keeps a peptide in water. Every modification is therefore a trade, and the useful question is not which backbone is best but which property you can afford to give up.

Four places to cut

Peptoids: the largest change to the backbone, and the best characterised

A peptoid is, in its inventor's words, simply a regioisomer of a peptide: the side chain is moved one atom, from the α-carbon to the amide nitrogen5. The design intent was explicit — keep the diversity and spacing of side chains and a polar backbone, while engineering out protease degradability. Moving the substituent onto the nitrogen removes both the hydrogen bond donor and the backbone stereocentre, and leaves tertiary amides that adopt cis and trans conformations far more freely than a peptide bond does1. The result is an achiral, flexible backbone carrying no amide NH donors2; the termini and any donor-bearing side chains remain available.

The backbone is consequently resistant to proteolysis: in a comparison of homologous L-amino acid, D-amino acid and N-substituted glycine oligomers against a panel representing the major protease classes — chymotrypsin, elastase, papain, pepsin, trypsin and carboxypeptidase A — each L-peptide was readily cleaved by the corresponding enzyme, while the peptoid oligomers were cleaved minimally or not at all6. This biostability is what makes peptoids attractive for antimicrobial and other applications where serum stability is the limiting factor3. Two consequences follow that are less often mentioned.

• Structure has to be rebuilt, not inherited. With no backbone hydrogen bonding and no backbone chirality, a peptoid cannot fold the way its parent peptide did. It can still fold — periodic incorporation of bulky α-chiral side chains drives stable helices with intense CD spectra4 — but the folding is driven by steric and hydrophobic effects in the side chains rather than by the backbone. Transposing a sequence residue-for-residue and expecting the same shape does not work.
• Water solubility becomes a real problem. The loss of backbone NH donors contributes to poor aqueous solubility in many peptoids2, although solubility depends heavily on the side chains chosen and water-soluble peptoids are well known. It is worth considering when planning the first solubility test.

The compensation is synthetic. Peptoids are built by a submonomer cycle rather than from protected monomers, so the side chain is introduced from a primary amine — a large and inexpensive chemical space, far wider than the twenty proteinogenic side chains and most unnatural amino acids combined1. For library work that economics is decisive.

The milder options

N-methylation removes one donor at one position rather than all of them. It confers local protease resistance, raises lipophilicity and can improve permeability, which is why it is the central lever in macrocycle design — see macrocycle membrane permeability. Because it is positional, it is also reversible as a design decision: a scan of single N-methyl substitutions tells you which positions tolerate the loss and which do not, in the same way an alanine scan maps side-chain contributions. The failure mode is methylating a residue whose NH makes a binding contact.

β-peptides insert an additional carbon into the repeat unit. They resist proteases and fold into well-defined helices of their own, but those helices have different geometry and periodicity from an α-helix. A β-peptide is best thought of as a new foldamer that can be designed to present a similar array of side chains, not as a drop-in replacement for a helical peptide.

Thioamides replace a carbonyl oxygen with sulphur: a single-atom change that alters hydrogen bonding and geometry only slightly, which is what makes them useful as site-specific probes and as a minimal perturbation rather than a wholesale redesign.

D-amino acids deserve a mention here even though they leave the backbone connectivity untouched. Inverting stereochemistry at a position disrupts protease recognition without removing any hydrogen bond donor, which makes a D-scan a sensible first experiment before committing to a more invasive backbone change — see D-amino acid peptides.

What changes in the analysis

Modified backbones break assumptions built into routine characterisation, and it is worth knowing which.

• Amino acid analysis relies on hydrolysis to free amino acids. A peptoid or a β-peptide does not give the expected residues, so peptide content has to be determined another way — see net peptide content.
• Circular dichroism reference spectra are derived from α-peptide secondary structure. Peptoid helices give strong CD signals, but those signals are not interpreted with the same reference set.
• Mass spectrometry fragmentation is less predictable: tertiary amides fragment differently, and sequence assignment from MS/MS cannot be assumed — see reading a peptide mass spectrum.
• Retention behaviour shifts. Removing backbone hydrogen bonding generally increases retention on reversed phase, so a gradient developed for the parent peptide will usually need redeveloping.

Building the modified sequence

LifeTein synthesises peptides with modified backbones: N-methyl amino acids at specified positions, D-amino acid and retro-inverso sequences, β-amino acids, peptoid and peptide–peptoid hybrid oligomers, and other non-natural building blocks. If you are scanning positions rather than making one compound, tell us the parent sequence and which positions to vary and we will quote the series.

References

1. Zuckermann RN. Peptoid origins. Biopolymers (Peptide Science) 2011;96(5):545–555. doi:10.1002/bip.21573 (first-hand account of the design rationale: moving the side chain from the α-carbon to the amide nitrogen, the loss of the backbone NH donor and of backbone chirality, the resulting cis/trans freedom, and the submonomer approach)
2. Culf AS, Ouellette RJ. Solid-phase synthesis of N-substituted glycine oligomers (α-peptoids) and derivatives. Molecules 2010;15(8):5282–5335. doi:10.3390/molecules15085282 (review covering peptoid chemistry from its inception in 1992, including the consequences of a backbone with no hydrogen-bond donors)
3. Chongsiriwatana NP, Patch JA, Czyzewski AM, Dohm MT, Ivankin A, Gidalevitz D, Zuckermann RN, Barron AE. Peptoids that mimic the structure, function, and mechanism of helical antimicrobial peptides. Proceedings of the National Academy of Sciences USA 2008;105(8):2794–2799. doi:10.1073/pnas.0708254105
4. Sanborn TJ, Wu CW, Zuckermann RN, Barron AE. Extreme stability of helices formed by water-soluble poly-N-substituted glycines (polypeptoids) with α-chiral side chains. Biopolymers 2002;63(1):12–20
5. Zuckermann RN, Kerr JM, Kent SBH, Moos WH. Efficient method for the preparation of peptoids [oligo(N-substituted glycines)] by submonomer solid-phase synthesis. Journal of the American Chemical Society 1992;114(26):10646–10647. doi:10.1021/ja00052a076; and Simon RJ, Kania RS, Zuckermann RN, Huebner VD, Jewell DA, Banville S, et al. Peptoids: a modular approach to drug discovery. Proceedings of the National Academy of Sciences USA 1992;89(20):9367–9371. doi:10.1073/pnas.89.20.9367 — the original peptoid reports
6. Miller SM, Simon RJ, Ng S, Zuckermann RN, Kerr JM, Moos WH. Comparison of the proteolytic susceptibilities of homologous L-amino acid, D-amino acid, and N-substituted glycine peptide and peptoid oligomers. Drug Development Research 1995;35(1):20–32; and Miller SM et al. Proteolytic studies of homologous peptide and N-substituted glycine peptoid oligomers. Bioorganic & Medicinal Chemistry Letters 1994;4(22):2657–2662. doi:10.1016/S0960-894X(01)80691-0