In most peptide work β-sheet formation is a problem to be suppressed: it stalls couplings on the resin, precipitates peptides out of buffer, and turns a clean synthesis into a difficult one. Self-assembling peptide materials take the same physics and use it deliberately: a sequence designed so that the sheets form only when told to, in a geometry that produces fibrils rather than a precipitate. This page covers how the trigger works, what the resulting gels can do that chemically crosslinked ones cannot, and the practical points that decide whether a batch gels reproducibly.
The best-studied design is the β-hairpin class, of which MAX1 is the archetype. It is a twenty-residue peptide with the sequence (VK)4-VDPPT-(KV)4-NH2: alternating valine and lysine on both arms, with a turn-promoting unit in the middle1. That alternation is the whole trick. Folded into a hairpin, all the valines land on one face and all the lysines on the other, giving an amphiphilic sheet with one hydrophobic face and one charged face.
In low-salt water the peptide populates an ensemble with little stable hairpin structure and stays soluble, because the lysine charges repel each other strongly enough to prevent the hairpin closing. Raise the ionic strength to physiological levels — the published work uses 150 mM salt at pH 7.4 — and that repulsion is screened; the peptide folds, and the folded hairpins immediately stack through their hydrophobic faces3. The product is not an amorphous precipitate but branched clusters of uniform, semiflexible, β-sheet-rich nanofibrils with a defined cross-section of about 3 nm2.
Because the trigger is electrostatic, it can be moved. Point substitutions that change the net charge shift the conditions required, and a redesigned variant, MAX1(K15E), undergoes thermally triggered gelation at ordinary physiological buffer conditions — pH 7.4, 150 mM NaCl, 37 °C4. Other systems use pH instead: the widely used RADA16-I gels spontaneously a couple of hours after dissolution in water, and the process can be accelerated by adjusting pH8.
The crosslinks here are physical — fibrils entangled and associated, not covalently bonded. That single structural fact produces the property these materials are used for.
Apply enough strain and the physical crosslinks break, viscosity falls, and the solid flows: it can be pushed through a syringe needle. Stop the strain and the crosslinks reform. MAX1 gels flow under high strain and begin to stiffen again as soon as the stress stops. How much comes back immediately depends on how hard and how long they were sheared: under a mild oscillatory strain the modulus recovers close to half its value almost at once, whereas after sustained high-rate shear the immediate figure is only a few percent and rises with time. Around 80% of the equilibrium modulus is typically back within about half an hour, and 65–95% after two hours depending on the shear history6. For scale, a 2 wt% gelatin hydrogel takes over four hours to recover 80% of its equilibrium modulus after the same treatment7.
The practical consequence is that the gel can be made before it is used. A covalently crosslinked gel is commonly formed in place, which means doing reactive chemistry in the presence of whatever it is meant to contain. Covalent gels can also be prepared in advance; whether such a gel survives injection and recovers afterwards depends on how it is designed. A shear-thinning physical gel can be assembled under controlled conditions, loaded with cells or a protein, and then delivered through a needle, re-forming where it lands. In these physical gels there is no chemical crosslinker and no curing step, so the payload is not exposed to reactive species — though compatibility with the specific formulation still has to be checked, since self-assembling systems differ.
Three variables account for most of the irreproducibility in this work.
Self-assembling sequences are demanding to make for exactly the reason they work. A sequence of alternating valine and lysine is a textbook difficult sequence: it aggregates on the resin, which is the same β-sheet propensity that makes the final material. Expect low crude purity relative to an ordinary peptide of the same length, and expect the countermeasures from long peptide synthesis — low-loading resin, backbone protection, pseudoproline units where the sequence allows.
Two further points matter more here than elsewhere. Counterion identity is not cosmetic: residual trifluoroacetate changes the ionic environment the assembly trigger responds to, so a TFA salt and an acetate or hydrochloride salt of the same peptide can gel differently — see removing TFA salt and TFA content analysis. And peptide content, not net weight, sets the concentration: weighing out 10 mg of a lyophilised powder that is 70% peptide and 30% salt and water gives a markedly different wt% than intended, which is a common reason a gel that worked in one lab fails in another. See net peptide content.
| Design principle | Alternating polar and non-polar residues give an amphiphilic sheet face |
| Trigger | Ionic strength, pH or temperature — set by the charged residues |
| Structure | Branched β-sheet nanofibrils, not amorphous precipitate |
| Key property | Shear-thinning and self-healing — injectable without a crosslinker |
| Reproducibility | Concentration above the CGC, mixing method, and equilibration time |
| Specify on order | Counterion and peptide content, not just purity |
LifeTein synthesises self-assembling and aggregation-prone sequences, including alternating hydrophobic/charged designs, with the counterion exchanged to acetate or hydrochloride where the TFA salt would interfere, and peptide content reported by nitrogen determination or amino acid analysis so that a weight-percent formulation is quantitatively reliable. Tell us the trigger conditions you intend to use and we will supply the peptide in a form compatible with them.