This biological chain reaction explains aspects of protein denaturation, blood clotting, and mitosis.

Elwood V. Jensen.

Recently it has become evident that a variety of phenomena involving proteins and peptides possess certain rather unusual features consistent with the assumption that existing disulfide bonds have been disrupted and new disulfide bonds have formed. Unlike the previously known cases of disulfide modification in proteins by reduction of these bonds to sulfhydryl groups followed by reoxidation of the latter to new disulfide linkages, the more subtle disulfide rearrangements here described take place in the absence of added reducing and oxidizing agents and appear to be chain-type reactions initiated in most cases by very small amounts of sulfhydryl compounds. Under appropriate conditions, the initiating sulfhydryl reacts with a disulfide group to form a new disulfide linkage, at the same time generating a new sulfhydryl group capable of reiterating the process (Fig. 1). Thus, a single sulfhydryl initiator can bring about the reaction of a large number of disulfide groups, and the occurrence of such processes, often under rather mild conditions, can exert far-reaching effects.

This article summarizes the reported examples of such sulfhydryl-disulfide interchange phenomena. Although most of these observations are concerned with in vitro transformations of proteins and peptides, it now appears that similar reactions are involved in certain physiological processes as well. It is hoped that the concept of a sulfhydryl-disulfide chain reaction may prove of value in elucidating the mechanisms of still other biological phenomena.

The author is associate professor in the Ben May Laboratory for Cancer Research, University of Chicago, Chicago, 111.

Protein Reactions

Most observations of sulfhydryl-disulfide interchange reactions have been concerned with phenomena accompanying protein denaturation, chiefly those of aggregation. The first indication that protein sulfhydryl groups play a role in the aggregation of denatured protein came from observations of the remarkable influence of the single sulfhydryl group of bovine plasma albumin on the nature of the clot or coagulum formed when solutions of this protein are heated (1). Clots formed by the thermal denaturation of ordinary bovine plasma albumin at neutral pH are opaque, friable, and synerizing, whereas those formed from albumin which has had its sulfhydryl group destroyed or blocked by treatment with an appropriate "sulfhydryl reagent" are transparent, firm, and nonsynerizing. Moreover, in the absence of the sulfhydryl group, solid gels are formed in solutions of much lower albumin concentration than when the sulfhydryl group is present. On the basis of existing concepts of gel structure (2), it was concluded that in the coagulum formed from sulfhydryl-containing albumin, the denatured protein chains lie in close side-by-side association and that the albumin sulfhydryl group must in some way bring about this type of aggregation (1).

The manner in which a sulfhydryl group is able to promote cross linking of protein molecules became apparent during subsequent experiments on the denaturation of proteins by urea (5). The previously known property of proteins such as plasma albumin, fibrinogen, y-globulin, and egg albumin to form clear firm gels when exposed to concentrated urea (4) was shown to depend on the presence of small amounts of protein sulfhydryl groups. Gelation in urea or in guanidine hydrochloride is favored by increased pH, inhibited by oxygen, eliminated by blockage of protein sulfhydryl groups, and restored to sulfhydryl-free protein by the addition of trace amounts either of sulfhydryl-containing proteins or of simple mercaptans. Since the observed phenomena appear to involve a stoichi-ometry quite different from that usually encountered in protein reactions, it was proposed that, under the conditions of protein denaturation, the sulfhydryl group initiates a chain reaction with disulfide groups in the manner illustrated in Fig. 1, leading to a regular three-dimensional gel network in the case of urea denaturation (and to side-by-side association of protein molecules in the case of thermal denaturation where no urea molecules are present to hold the protein units apart).

Subsequent measurements (5) of viscosity changes in more dilute bovine plasma albumin solutions clearly demonstrated that, in addition to the large immediate increase in viscosity following exposure of the protein to concentrated urea, there is a further gradual sulfhydryl-dependent viscosity rise which is influenced by the same factors as is the gelation in more concentrated albumin solutions, and which thus reflects the disulfide interchange reaction. Similarly, in the case of thermal denaturation, the results of viscosity and sedimentation measurements on dilute albumin solutions, as well as the effect of traces of mercaptans on the thermal coagulation of iodoacetamide-treated albumin, furnish strong support for the concept that lateral association by sulf-hydryl-initiated disulfide interchange takes place when albumin solutions are heated (6).

These considerations of a chain-type sulfhydryl-disulfide interchange reaction operating during conditions of protein denaturation have been confirmed and extended by a number of investigators. As part of an extensive study of the effect of urea denaturation on the viscosity and optical rotation of protein solutions, Kauzmann and his collaborators (7) have shown that, with both ovalbumin and bovine plasma albumin, an exchange reaction between sulfhydryl and disulfide groups is an important cause of aggregation during denaturation, especially at pH values above neutrality. Subsequent investigations of solubility changes during the course of urea denaturation of bovine plasma albumin permitted Kauzmann and Douglas (8) to distinguish between intramolecular disulfide exchange, which diminishes protein solubility only slightly, and intermolecular disulfide exchange leading to aggregation and a larger decrease in solubility. From the ability of reducing agents to decrease the light scattering of solutions of albumin which had been denatured by heat, shaking, or exposure to ethanol or urea, Halwer (9) concluded that the denatured protein is cross linked, at least partially, by intermolecular disulfide bonds, this cross linking being most pronounced in the case of the alcohol-treated protein.

From the effect of pH and of sulfhydryl reagents on the molecular weights of proteins in 1M urea, McKenzie, Smith, and Wake (10) concluded that, with bovine plasma albumin, the only cause of aggregation is sulfhydryl-disulfide interchange, which occurs under alkaline but not under acid conditions. With ovalbumin, aggregation appears to take place both by disulfide exchange and by hydrogen bonding, the latter being somewhat more important. Kolthoff and his coworkers (11) have studied viscosity changes of dilute bovine plasma albumin solutions after treatment with guanidine hydrochloride, and, in agreement with previously mentioned findings with urea, they have observed a gradual prolonged increase in viscosity, resulting from protein aggregation through sulfhydryl-disulfide interchange. After prolonged exposure of bovine plasma albumin to guanidine hydrochloride, the same authors observed a decrease in the reactivity of the protein sulfhydryl group toward silver or mercuric ions, which they attribute to the production, by sulfhydryl-disulfide interchange, of a sulfhydryl group less accessible to silver or mercury than the original one. Toward ferricya-nide, the reactivity relationship is reversed; Kolthoff and Anastasi (12) have shown that the original sulfhydryl group in native bovine plasma albumin is not oxidizable by ferricyanide but that exposure to concentrated urea or guanidine hydrochloride produces a new sulfhydryl group which, after removal of the denaturing agent, can be oxidized by ferricyanide to form a protein dimer.