Why some peptides dissolve badly: charge, hydrophobicity and gelling
Net charge, isoelectric point, hydrophobic residue content and beta-sheet self-assembly decide whether a peptide dissolves clear, turns cloudy or gels. The chemistry, with worked examples from published sequences.
Whether a peptide dissolves clear, turns cloudy or sets into a gel is decided by three properties of its sequence: its net charge at the pH of the solvent, the share of its residues that are hydrophobic, and its tendency to assemble into beta-sheets. Manufacturers sort their solubility guidance by net charge because charge is the property that can be changed with pH. This is chemistry, not a procedure: we do not publish reconstitution guidance, and nothing here names a volume or a step.
Net charge and the isoelectric point
Water dissolves what it can surround with its own dipoles, and it surrounds charged groups best. A peptide's charges come from its ends and its side chains. JPT's solubility note gives the counting rule: "Assign a value of +1 to each basic residue (K, R, H and NH2 at the N-terminus)" and "a value of −1 to each acidic residue (D, E, and COOH at the C-terminus)." An acetylated N-terminus or an amidated C-terminus removes that terminal charge, which is one reason those modifications appear on a sequence line at all (how to read a peptide sequence covers the notation).
The sum of those values is the net charge, and it depends on pH: acidic groups are neutral at low pH and negative above their pKa, basic groups positive at low pH and neutral above theirs. The pH at which the positives and negatives cancel is the isoelectric point, pI. Near it a peptide has the least reason to stay apart from its neighbours. Zapadka and colleagues, reviewing the physical stability of peptide drugs, state the rule directly: "Generally, the higher the net charge the slower the aggregation and the lower the net charge the higher the propensity to aggregate," and note that "around the isoelectric point of peptides and proteins, where net charges are essentially zero, then monodisperse and quasi-amorphous aggregates have been observed instead of amyloid-like fibrils." That amorphous aggregate is what a cloudy solution usually is.
Why manufacturers sort their guidance by charge
Bachem's solubility note classifies a peptide as basic when the "number of basic amino acids including the N-terminal amino group > number of acidic amino acids," and as acidic in the reverse case. The logic that follows is pure acid-base chemistry. Solubility rises as the pH moves away from the pI, which is why manufacturers' solubility notes are organised by charge class, and why peptides "containing a high proportion of polar uncharged amino acids and/or hydrophobic amino acids" form a separate class.
The same sequence sets two limits on that logic. Low pH is also the condition under which the bond after an aspartate hydrolyses, and high pH the condition under which a free cysteine thiol oxidizes to a disulfide; freeze-thaw, aliquots and the residues that oxidize covers both. That is why manufacturer guidance is written per peptide, and why the same molecule can carry different notes from different suppliers.
Hydrophobic content
Kyte and Doolittle's 1982 hydropathy scale assigns each residue a value, and averaging it over a sequence gives a single number for how water-avoiding the chain is overall. JPT's note draws a practical line: peptides "with > 25% hydrophobic aa" form their own solubility class. GenScript's design guide gives the matching rule for people designing sequences: "Hydrophobic amino acid content is suggested to be kept below 50% with at least one charged residue incorporated within every five amino acids."
The hydrophobic residues are the ones with aliphatic or aromatic side chains: leucine, isoleucine, valine, phenylalanine, tryptophan, tyrosine, methionine, alanine. In water they do what oil does, and a peptide with a run of them will bury that run against another copy of itself rather than expose it. Proline is the exception that helps: Bachem notes that "dispersed proline residues disrupt secondary structures," which keeps a chain from settling into the flat, layerable shape described next.
Aggregation, beta-sheets and gelling
The most stable way for two extended peptide chains to interact is side by side, hydrogen-bonded backbone to backbone, in a beta-sheet. Layer enough sheets and the assembly is a fibril. Zapadka's review describes the forces: the fibrils "are stabilized by the hydrophobic effect, van der Waals interactions," aromatic ring-to-ring contacts "and hydrogen bonds." The sequence signature of that tendency is short: "A series of five or more residues which all have a high intrinsic propensity form what is known as an aggregation-prone region, APR," and prediction tools such as TANGO and AGGRESCAN score sequences for exactly these regions using hydrophobicity and beta-sheet propensity.
Three variables control how fast it happens. Concentration, first: the review calls the concentration dependence "long-established through numerous kinetic studies on the formation of amyloid fibrils, amorphous aggregates and adsorption on surfaces." pH, second, and not always in the expected direction: GLP-1 forms fibrils at neutral pH, and "at pH > 8.0, GLP-1 follows the more common nucleation-polymerization kinetics ... however, at pH 7.5, the kinetics reverse"; glucagon aggregates both at pH 2.0 and at pH 8.5 to 9.7; salmon calcitonin forms fibrils at neutral pH. Ionic strength, third, because dissolved salt screens the charges that keep chains apart.
A fibril-forming peptide at high enough concentration does not just precipitate; the fibrils can tangle into a network that holds the solvent, and the solution thickens or sets. That is what the word "gelling" describes. It is a physical state, reversible in some systems and not in others, and it is the visible end of the same chemistry that starts with a low net charge and a hydrophobic stretch.
Counter-ions and salt form
The anion that came with the peptide matters too: AmbioPharm notes that salt choice can affect a peptide's stability, solubility and secondary structure. Trifluoroacetate and acetate are the two forms a research vial is likely to carry; TFA salt or acetate salt explains where each comes from and how much of the mass it can be.
Reading published sequences for aggregation
The table applies the rules above to published sequences. The net-charge column is the JPT count at a pH where the acidic groups are deprotonated and the amines protonated, with histidine counted as uncharged; it is arithmetic on the sequence, not a measurement.
| Peptide | Sequence | Basic sites | Acidic sites | Reading |
|---|---|---|---|---|
| TP-AMR1 | Lipidated amylin analogue, 4409 Da | (full sequence not published) | The parent hormone amylin has, in the words of the Novo Nordisk chemists who developed the analogue, "a high propensity toward the formation of amyloid fibrils, which makes it a challenging drug design effort"; the analogue was engineered to be "stable, lipidated." | |
| AOD-9604 | YLRIVQCRSVEGSCGF, cyclic via Cys-Cys | N-terminus, two Arg | Glu, C-terminus | A net positive charge, a disulfide ring, and a run of aliphatic and aromatic residues (Leu, Ile, Val, Phe, Tyr): the ring limits how far the chain can extend into a sheet, the hydrophobic residues raise its predicted aggregation propensity |
| BPC-157 | GEPPPGKPADDAGLV | N-terminus, Lys | Glu, two Asp, C-terminus | Net negative; a 2026 review notes that its "three consecutive Pro residues at positions 3-5 disfavour β-sheet conformation" |
| Epitalon | AEDG | N-terminus | Glu, Asp, C-terminus | A short, strongly acidic peptide with no hydrophobic stretch |
| Selank | TKPRPGP | N-terminus, Lys, Arg | C-terminus | Net positive, two prolines, no hydrophobic run |
| Semax | MEHFPGP | N-terminus (His weakly) | Glu, C-terminus | Close to neutral at neutral pH; short and proline-broken |
| MOTS-c | MRWQEMGYIFYPRKLR | N-terminus, three Arg, Lys | Glu, C-terminus | Net positive, but the Tyr-Ile-Phe-Tyr stretch is the kind of hydrophobic run the APR predictors look for |
The pattern is the one the manufacturers' notes encode. Short, charged, proline-broken sequences are predicted by these rules to have low aggregation propensity. Long sequences with a hydrophobic stretch and a documented fibril-forming parent (TP-AMR1, and the GLP-1 class generally) are the ones formulation chemists spend years on, and they are the peptides whose manufacturer solubility notes carry the most chemistry.
Common questions
What decides whether a peptide dissolves clear or turns cloudy?
Three properties of its sequence: its net charge at the pH of the solvent, the share of its residues that are hydrophobic, and its tendency to assemble into beta-sheets. This is chemistry, not a procedure, and we do not publish reconstitution guidance.
What is the isoelectric point of a peptide?
It is the pH at which a peptide's positive and negative charges cancel. Near it a peptide has the least reason to stay apart from its neighbours, and a published review notes that around the isoelectric point, where net charges are essentially zero, amorphous aggregates have been observed. That amorphous aggregate is what a cloudy solution usually is.
How is the net charge of a peptide counted from its sequence?
By adding up its charged groups. JPT's rule assigns +1 to each basic residue, meaning K, R, H and the amine at the N-terminus, and −1 to each acidic residue, meaning D, E and the acid at the C-terminus. The sum is the net charge, and it depends on pH. An acetylated N-terminus or an amidated C-terminus removes that terminal charge.
Why do some peptides form a gel in solution?
Because their fibrils can build a network. Extended chains hydrogen-bond side by side into beta-sheets, and enough layered sheets make a fibril. At high enough concentration the fibrils can tangle into a network that holds the solvent, and the solution thickens or sets. Gelling is a physical state, reversible in some systems and not in others.
Sources
- Peptide Solubilization, JPT, undated
- Peptide solubility, Bachem technical note, undated
- Custom Peptide Technical Resource, GenScript, undated (PDF)
- Which salt form should I choose for my peptide?, AmbioPharm FAQ, undated
- Zapadka KL, Becher FJ, Gomes dos Santos AL, Jackson SE, review of factors affecting the physical stability (aggregation) of peptide drugs, Interface Focus, 2017
- Kyte J, Doolittle RF. A simple method for displaying the hydropathic character of a protein. J Mol Biol, 1982
- Kruse T et al., the development paper for a long-acting amylin analogue, J Med Chem, 2021
- Mateescu DM et al., review of BPC-157 formulation and biopharmaceutical challenges, Pharmaceutics, 2026
- AOD-9604 (CID 71300630), PubChem
- Patel K, Borchardt RT. Chemical pathways of peptide degradation. II. Pharm Res, 1990
For laboratory research use only. Not a drug, not a supplement, and nothing here is a claim about what any of this material does in a person or an animal.
