Acid-Base Chemistry of Ionizable Side Chains

The pKa values of ionizable side chains are not a museum piece of textbook chemistry; they are load-bearing numbers in modern biology and medicine. In enzymology, catalytic triads (Ser-His-Asp in chymotrypsin, trypsin, and thousands of related hydrolases) function because a perturbed histidine pKa near neutral pH lets it shuttle between protonated and neutral forms exactly at the pH where catalysis occurs, acting as a general base to activate serine's hydroxyl. In hemoglobin, the Bohr effect — the pH-dependence of oxygen affinity — arises because specific histidine residues (notably His146 of the β chain) change protonation state as blood becomes more acidic in respiring tissue, driving a conformational shift that promotes oxygen release exactly where it is needed. In drug design, ionizable side chains lining a binding pocket dictate whether a candidate molecule will be protonated or neutral upon binding, directly affecting binding affinity, membrane permeability, and oral bioavailability; computational pKa prediction (via Poisson–Boltzmann electrostatics or empirical tools like PROPKA) is now a routine step in structure-based drug design. In protein engineering, deliberately shifting a catalytic residue's pKa — by mutating nearby residues to alter the local electrostatic environment — is a standard strategy for tuning enzyme pH-optima for industrial applications (detergent proteases engineered to remain active at high pH, for example). In structural biology, cryo-EM and X-ray crystallography routinely need to assign protonation states to Asp, Glu, and His residues near active sites to correctly interpret catalytic mechanisms, since standard diffraction data at typical resolution cannot see hydrogen atoms directly. And in the eventual construction of full protein Titration Curves of Amino Acids and the calculation of a protein's Isoelectric Point — the pH at which its net charge is exactly zero — every one of these seven side-chain equilibria is summed, residue by residue, across the entire polypeptide.

Ask the tutor: How do these seven individual side-chain equilibria combine to produce a protein's full titration curve and its isoelectric point?

Each ionizable side chain is an acid or a conjugate acid in the Brønsted–Lowry sense: it can donate a proton to water, generating a conjugate base and a hydronium ion. The equilibrium is written generically as

for carboxylic acids (Asp, Glu), the phenolic OH of Tyr, and the thiol of Cys, or as

for the protonated nitrogen bases: the imidazolium of His, the ε-ammonium of Lys, and the guanidinium of Arg. The distinction matters because it determines which side of the equilibrium is positively charged. For carboxylic acids and Cys/Tyr, the protonated form is neutral and the deprotonated form is negatively charged. For the nitrogen bases, the protonated form is positively charged and the deprotonated form is neutral. The strength of each acid is captured by its acid dissociation constant $K_a$, almost always reported as $\text{p}K_a = -\log_{10} K_a$. A low pKa means a strong tendency to give up the proton (a strongly acidic group); a high pKa means the group clings to its proton and only ionizes at high pH. The seven side chains span an enormous range — from Asp's carboxyl at pKa ≈ 3.9 up to Arg's guanidinium at pKa ≈ 12.5, nearly nine orders of magnitude in acid strength. What makes each group acidic or basic to the degree it is comes down to resonance stabilization of the conjugate base or acid. The carboxylate of Asp/Glu is stabilized by delocalization of the negative charge over two equivalent oxygens (Y-aromaticity-like resonance), lowering its pKa relative to a simple alcohol. The imidazole ring of His is aromatic in both protonated and neutral forms, but protonation converts it to a symmetric, doubly-stabilized aromatic cation (imidazolium), giving intermediate acidity. The guanidinium cation of Arg is exceptionally stable because the positive charge is delocalized over three nitrogens by resonance — this is why Arg's pKa is so high: the protonated (cationic) form is deeply stabilized and very reluctant to release its proton.

The consequence at physiological pH (≈ 7.4) is a striking asymmetry: Asp and Glu, sitting roughly 3.5 pH units above their pKa's, are overwhelmingly deprotonated and negatively charged. Lys and Arg, sitting roughly 3–5 units below their pKa's, are overwhelmingly protonated and positively charged. His straddles the transition, with its pKa (6.0) close enough to physiological pH that a modest local perturbation — a nearby carboxylate, a buried hydrophobic pocket, a hydrogen bond — can shift the equilibrium fraction protonated from 10% to 90%. This is precisely why histidine is nature's favorite general acid/base catalyst.

Every one of these equilibria, whatever chemical group is involved, obeys the same acid dissociation constant expression. For a generic acid HA:

where:

• $K_a$ — acid dissociation constant (mol/L), specific to the chemical group
• $[\text{A}^-]$ — concentration of the deprotonated (conjugate base) form
• $[\text{HA}]$ — concentration of the protonated (acid) form
• $[\text{H}^+]$ — hydrogen ion concentration set by solution pH

Taking $-\log_{10}$ of both sides and rearranging gives the Henderson–Hasselbalch equation, the single most useful relation in this article:

where:

• $\text{pH} = -\log_{10}[\text{H}^+]$ — the solution's acidity
• $\text{p}K_a = -\log_{10}K_a$ — the group's intrinsic acid strength
• the log ratio — relative populations of base and acid forms at equilibrium

Reading: when pH = pKa, the ratio [A⁻]/[HA] = 1, meaning the group is exactly half-ionized. Each unit that pH exceeds pKa multiplies the base/acid ratio by 10; each unit below pKa divides it by 10. This single relation, applied group by group, predicts the ionization state of every side chain in a protein at any pH. It is more convenient for charge-balance calculations to solve directly for the fraction of molecules in the protonated form, $f_{\text{HA}}$, or deprotonated form, $f_{\text{A}^-}$, as an explicit function of pH.

where:

• $f_{\text{HA}}, f_{\text{A}^-}$ — dimensionless fractions summing to 1 for any given group
• the exponent $\text{pH}-\text{p}K_a$ — how many pH units the solution sits above the group's pKa

For basic groups (His, Lys, Arg), the same algebra applies with BH⁺ playing the role of HA:

Finally, the average net charge contributed by a single ionizable side chain at a given pH is a weighted sum over its two protonation states, with sign convention $+1$ for a cationic form and $-1$ for an anionic form:

This is the building block from which the whole-protein net charge — central to the eventual Isoelectric Point calculation — is assembled by summing over every ionizable residue.

Example 1 — Foundational: fraction of glutamate side chain ionized at pH 7.4. Glutamate's side-chain carboxyl has pKa ≈ 4.1 (values in the literature range 4.0–4.3; we use 4.1 as the standard Lehninger figure). At physiological pH 7.4:

So 99.95% of glutamate side chains are deprotonated (carboxylate, charge −1) at physiological pH, and only 0.05% remain protonated (neutral –COOH). This confirms the common shorthand "Glu is negatively charged at physiological pH" — but it is a probabilistic statement about an equilibrium, not an absolute rule; roughly 1 in 2000 glutamates is momentarily protonated at any instant.

Limiting case check: at pH = pKa = 4.1 exactly, the formula gives $f_{\text{A}^-} = 10^0/(1+10^0) = 1/2$ — the defining condition of pKa, confirming the calculation framework before applying it elsewhere.

Example 2 — Applied: net charge on a lysine side chain buried near a hydrophobic patch (perturbed pKa). Free lysine's ε-amino group has an intrinsic pKa of about 10.5. Suppose this particular lysine sits in a nonpolar cavity within a folded protein, where desolvation of the charged ammonium form is energetically costly; empirically, buried Lys residues can have their pKa depressed by several units (values as low as 6–8 have been measured by NMR in some proteins, e.g., staphylococcal nuclease mutants). Suppose spectroscopic titration reveals this particular buried Lys has an apparent pKa of 8.2 rather than 10.5. What fraction is protonated (cationic, +1) at pH 7.4, the physiological reference point?

So even with a dramatically depressed pKa, this lysine remains 86.3% protonated at physiological pH — still predominantly cationic, but with a meaningfully increased population (13.7%) of the neutral amine compared to a surface-exposed Lys (which, using pKa 10.5, would give $10^{7.4-10.5}=10^{-3.1}=7.94\times10^{-4}$, so $f_{\text{BH}^+}=1/(1.000794)=0.99921$ — essentially fully protonated). The comparison shows how burial can shift a charge state by two orders of magnitude in ionization fraction even though the residue remains "mostly protonated" in both cases — the qualitative label survives, but the quantitative population shifts by a factor of roughly 172 in the neutral-form fraction ($0.1368/0.00079 \approx 173$).

Reduction check: setting the perturbed pKa back to the free-amino-acid value of 10.5 exactly recovers the unperturbed surface-lysine result above, confirming the calculation is a genuine generalization rather than a coincidence.

Example 3 — Competition-level: coupled ionization of a catalytic Asp–His pair (charge-relay logic). In a serine-protease-like active site, a buried aspartate hydrogen-bonds to a histidine, which in turn deprotonates a serine during catalysis (the classic Asp-His-Ser charge relay). Suppose experimental pH-rate profiles indicate the following apparent pKa values in the folded enzyme (shifted from free-amino-acid values by the local environment): Asp, pKa = 2.5 (lowered from the free-amino-acid value of 3.9 because its carboxylate is stabilized by a buried hydrogen bond to protonated His); His, pKa = 7.0 (raised from the free value of 6.0 because of an adjacent hydrophobic patch destabilizing the charged imidazolium's desolvation less than expected — a modest positive shift). At pH 6.2 (near the reported optimum for many such enzymes), find (a) the fraction of Asp deprotonated, (b) the fraction of His protonated, and (c) the combined average charge of the pair, and (d) determine at what pH the pair's total charge would cross exactly through zero, if it does within the physiological range. (a) Asp:

Asp is 99.98% deprotonated at pH 6.2 — essentially fully ionized, consistent with its very low apparent pKa. (b) His:

His is 86.3% protonated at pH 6.2 (below its raised pKa of 7.0, so still mostly cationic). (c) Combined average charge:

The pair carries a net charge of approximately −0.14 at pH 6.2 — mostly canceling (Asp⁻ paired with His-H⁺, an ion pair stabilizing the folded structure by roughly one unit of formal charge, net near-neutral), with a small residual negative charge. (d) Crossing point where $\langle z_{\text{tot}}\rangle=0$: We need

For pH well above 2.5 (which any biologically relevant pH satisfies by several units), $10^{\text{pH}-2.5}\gg 1$, so the right-hand side ≈ 1 to excellent approximation (at pH 6, error is $10^{-3.5}\approx 3\times10^{-4}$, negligible). The equation simplifies to

which is only satisfied as pH → −∞ in the exact sense — meaning within any biologically accessible pH range (0–14), the sum never exactly reaches zero because Asp is essentially always fully ionized (contributing close to −1) while His's contribution varies smoothly from 0 to 1. Solving numerically for where the two terms are equal in magnitude (His fully protonated matching Asp's near-total ionization) shows the crossing would require His at pH ≈ pKa,His = 7.0 exactly cancelling only if Asp were also exactly 50% ionized, which never co-occurs given the 4.5-unit pKa gap. The physically meaningful conclusion: this Asp–His pair never becomes exactly neutral on average across the physiological pH range — it is always net negative to some degree, because Asp's ionization saturates far faster (lower pKa) than His's protonation state changes, an asymmetry that is itself mechanistically important: it keeps a small residual negative electrostatic field oriented toward the His imidazolium across the whole catalytic pH window, reinforcing the charge-relay geometry rather than letting it collapse to neutrality.

Lawrence Joseph Henderson, a Harvard physiologist trained originally in chemistry, derived the equation relating pH to the ratio of conjugate acid and base forms in 1908, while studying the buffering behavior of blood plasma bicarbonate — years before "pH" itself was even a standard unit (Sørensen introduced the pH scale in 1909). Henderson's original relation was expressed in terms of concentrations and equilibrium constants without logarithms; it was the Danish biochemist Karl Albert Hasselbalch who, in 1916, recast Henderson's expression in the now-familiar logarithmic form using the newly available pH notation, producing the equation still taught in every biochemistry course as the Henderson–Hasselbalch equation. Henderson's deeper interest was in physiological homeostasis — how the blood, a solution of extraordinary chemical complexity, could hold its pH so remarkably constant despite continuous metabolic acid production; his 1913 book The Fitness of the Environment extended this thinking into a broader argument about the unique suitability of water, carbon, and their compounds for life. The application of the same buffering mathematics to individual amino-acid side chains — treating each ionizable group in a protein as its own independent miniature buffer with its own pKa — followed naturally once amino acid structures and protein primary sequences began to be worked out through the mid-twentieth century, culminating in the systematic tabulation of side-chain pKa values (by potentiometric titration and later NMR) that underlies every modern textbook table, including the ones used throughout this article.