What is Isoelectric point?

The isoelectric point (pI) is the pH at which a peptide carries no net charge. It is set by the dissociation constants (pKa) of the ionisable side chains of seven amino acids (Asp, Glu, His, Cys, Tyr, Lys, Arg) together with the free N-terminal amine and C-terminal carboxyl ^[1]. Below the pI the peptide is net positive; above it, net negative.

Why it matters at the bench

• Solubility. With little net charge, molecules lose the electrostatic repulsion that keeps them apart. In engineered variants of ribonuclease Sa with pI values of 3.5, 6.4 and 10.2, the pH of minimum solubility moved with the pI, while the pH of maximum activity and maximum stability did not ^[3]. This is the logic behind the solubility troubleshooting guide: move the pH away from the pI before reaching for organic solvents.
• Aggregation. The same loss of repulsion makes self-association more likely, so a solution held near its pI is more prone to aggregation.
• Analysis. Charge-based methods such as capillary isoelectric focusing and two-dimensional gel electrophoresis separate molecules by pI ^[1]. Deamidation converts neutral Asn to acidic Asp or isoAsp and lowers the pI, which is why deamidated forms can be resolved by charge.

Worked example: estimating pI from a sequence

Using average pKa values measured in folded proteins (Asp 3.5, Glu 4.2, His 6.6, Lys 10.5, C-terminus 3.3, N-terminus 7.7) ^[2], the net charge can be summed across pH until it crosses zero:

BPC-157 is therefore an anion in neutral buffer, but in dilute acid near pH 3.5 it would carry almost no net charge. Ipamorelin's C-terminal amide removes the only carboxyl, so its net charge stays positive at every pH and only approaches zero in strongly alkaline conditions; it has no true pI. Reading the ionisable groups straight off the sequence gives most of this picture before any arithmetic.

Common misunderstandings

• "The pI is a fixed, measured property." A calculated pI depends on the pKa set behind it. The ribonuclease Sa study found that estimates built from model-compound pKa data can be wrong by more than one pH unit ^[3], and the averages above come from folded proteins, so for short peptides they are approximations ^[2]. Dedicated predictors narrow the error: IPC 2.0 reported a root-mean-square deviation of 0.222 pH units for peptides, against 0.405 for previous algorithms ^[1].
• "A peptide is insoluble at its pI." Solubility is lowest near the pI, not zero. Short, hydrophilic peptides may stay in solution there; for hydrophobic sequences it is often decisive.
• "End modifications are cosmetic." N-terminal acetylation removes a positive charge and C-terminal amidation removes a negative one, and either can move the pI by several units, as ipamorelin shows.
• "The salt form changes the pI." Counter-ions such as trifluoroacetate are not part of the peptide and do not change its pI, although residual acid can lower the pH of an unbuffered solution.

See peptide solubility for how pI combines with hydrophobicity and length.