Thiol-Disulfide Exchange (Molecular Biology)

Thiol Groups and disulfide bonds undergo a spontaneous chemical reaction, in which the thiol group displaces one sulfur atom of the disulfide bond in an Sn type of reaction:

This reaction is involved in very many biological phenomena, especially those involving the thiol groups of cysteine residues of proteins, thioredoxin, glutaredoxin, protein disulfide isomerase, glutathione, coenzyme A, dihydrolipoamide, and lipoic acid. This thiol-disulfide exchange reaction is unique in physical organic chemistry in its ability to break and reform a strong, directional, covalent, disulfide bond in water at room temperature reversibly at a rapid rate that can be varied over many orders of magnitude by controlling the pH. Nevertheless, it is one of the most specific chemical reactions used in molecular biology, and thiol groups and disulfide bonds tend not to react with other groups on proteins, nucleic acids, etc.

It is the ionized form of the thiol group that is the reactive species, so the rate of the reaction varies with the pH below the pKa of the thiol group; typical biological thiol groups of cysteine residues have pKa values in the region of 9. The reaction can be quenched by acidification, making the thiolate anion insignificant. Alternatively, such mixtures can be trapped irreversibly by rapidly reacting all free thiol groups with reagents such as iodoacetamide, iodoacetate, or N-ethylmaleimide. The thiol and disulfide species present in both equilibrium and kinetic trapped mixtures can be analyzed chemically, for example, by HPLC or NMR, under acidic conditions. The rate of a simple thiol-disulfide exchange reaction can be measured if the disulfide bonds reacting and being generated differ in their spectrophotometry properties. For example, cyclic disulfide bonds in 5- or 6-membered rings, such as that formed in dithiothreitol, absorb at 280 to 330 nm, whereas most linear disulfide bonds do not. Dynamic 1H-NMR lineshape analysis and spin-transfer method have also been used to measure rate constants for simple model systems.

The three sulfur atoms involved in the reactions are generally labeled as the nucleophile (nuc), the central atom (c), and the leaving group (lg). The reaction occurs by the ionized thiolate anion nucleophile attacking one of the two sulfur atoms of the disulfide bond, which becomes the central atom, optimally along the axis of the S-S bond. The transition state for the reaction is expected to be a linear arrangement of the three sulfur atoms, equally spaced, and with the negative charge of the thiolate anion spread symmetrically, with more on the terminal atoms than on the central one. The rate constant for the reaction depends upon the electron affinities of each of the three sulfur atoms, which are conveniently measured by their pKa values when thiol groups (1, 2). The second-order rate constant for the intermolecular reaction between model compounds can be predicted by these three pKa values:

Of course, it is not possible to measure just one thiol-disulfide exchange reaction, such as that of Eq. 2, for the thiolate anion will also react with the other sulfur atom of the disulfide bond, and the combined rates will be measured as any initial rate. Subsequently, the mixed disulfides formed will also react further with other thiol groups, leading to complex mixtures of disulfide and thiol species. The second-order rate constant for reaction of a typical, fully ionized thiol group with a typical cysteine disulfide bond is approximately 20s-1M-1 at 25°C. The optimum pKa for the attacking thiol group, to give the largest rate, is the same as the pH of the reaction. A thiol with a greater pKa value will have less ionized form; with a lower pKa value, the thiol is ionized, but its nucleophilicity is lower.

The above considerations also apply equally to the reverse of the reaction, and the expected equilibrium constant can be estimated from Eq. 4 for the reaction in both directions. The equilibrium constant is pH-dependent if the attacking and leaving sulfur atoms have different pKa values; it varies over the pH interval between the two thiol pKa values. The equilibrium favors the thiol group with the lower pKa value; just the opposite would have been predicted considering only the effect of ionization of the two thiol groups on mass action.

Of course, the reaction occurs more rapidly than predicted if there are positive charges near the disulfide bond, to attract the attacking thiolate anion, or if the disulfide bond is strained; for example, the strained disulfide bond of lipoic acid reacts considerably more rapidly than otherwise expected (1), which is accounted for about 3.8 kcal/mol of conformational strain in the 5-membered ring (the CSSC dihedral angle is only about 30° and not the favored 90°). Conversely, the reaction is inhibited by the presence of negative charges near the disulfide bond, if either the thiol or disulfide group is buried and inaccessible, or if there are bulky substituents adjacent to the sulfur atoms. Electrostatic effects on the equilibria of thiol-disulfide exchange reactions are small in magnitude, but occur in the expected direction. Formation of a mixed disulfide with unlike charges close on the two moieties is favored, while that with like charges is disfavored. Such effects are substantial only when the charged groups are on adjacent residues, and their magnitude can be decreased in magnitude by electrostatic screening with high salt concentrations.

The thiol-disulfide exchange reaction is routinely used in molecular biology when protein disulfide bonds (P S) are reduced by reagents such as dithiothreitol or b-mercaptoethanol. In this case, two sequential thiol-disulfide exchange reactions are necessary, proceeding through a mixed disulfide between the reagent (RSH) and the protein:

Some proteins, such as thioredoxin, glutaredoxin, and protein disulfide isomerase (PDI), participate in thiol-disulfide exchange reactions much more rapidly than expected from the pKa values of their active-site thiol groups (Eq.(3)). This is believed to be due to these proteins tending to bind noncovalently the molecules that they react with, as an enzyme binds its substrate, and to them actually stabilizing the transition state for the reaction, again like an enzyme. When present in small quantities, such proteins appear to catalyze thiol-disulfide exchange reactions between other thiol and disulfide compounds. They do this by reacting more rapidly with both the thiol and disulfide compounds, so they do not catalyze the direct reaction between them but instead provide a more rapid alternative reaction pathway (3).

Of course, thiol-disulfide exchange reactions are readily reversible, and the above reactions can be used to add disulfide bonds to a protein (see Protein Folding In Vitro).

The thiol-disulfide exchange reaction can be simplified if a mono- or dioxide form of the disulfide is used. The oxidation increases the reactivity, but only the nonoxidized sulfur atom undergoes thiol-disulfide exchange:

Consequently, the reaction stops at this stage and will be stoichiometric.