General Aspects of Steroid-Protein Interaction

About 35 years ago B. Brunelli published the first paper on the binding of a steroid hormone to serum proteins (1). Its title, “Sulla ‘Funzione Veicolante’ delle Proteine Plasmatiche per l’Ormone Follicolare”, i. e., on the carrier function of the plasma proteins for the estrogenic hormone, clearly shows the influence of the concept of the vehicle function of the serum proteins which had just been developed by H. Bennhold (2). The program of our present meeting illustrates to what extent the small door which Brunelli opened in Pisa in 1934 has been widened. New doors have been found and opened after the first one, and large new areas have been discovered in which interactions between steroid hormones and proteins play essential roles.

Figure 1 shows the main areas of steroid-protein interactions. Binding of steroids to serum proteins has been studied most extensively, and we know more about the properties of these complexes than about those of any others. The present discussion will be mainly concerned with the area of steroid-protein interactions, involving non-covalent bonding with different serum proteins. Perhaps of greatest biological interest is the interaction of steroid hormones with the receptor proteins of target tissues (3–6); this aspect clearly involves the mechanism of action of this important class of vertebrate hormones. However, our knowledge of the chemical nature of the receptor proteins is still limited.

A special type of steroid binding to proteins is that to steroid-specific enzymes(7). Here the steroid plays the role of a substrate to an enzyme, and at the same time may be considered a ligand to a protein which has a high and specific affinity for the particular steroid. Obviously, the study of steroid-enzyme complexes is of interest for more than one reason, and results in recent years on the 3- and 17 β– hydroxysteroid dehydrogenases, and especially the Δ5-3-ketosteroid isomerase have demonstrated their usefulness for the investigation of fundamental problems of steroid-protein interaction. The Δ5-3-ketosteroid isomerase of bacterial origin is the most highly purified steroid enzyme known today, it is also one of the most active catalytic proteins known. The homogeneous crystalline enzyme (Table I)1 has a molecular weight of 40,800 and is composed exclusively of amino acids. It is free of cyst(e)ine and tryptophan residues. According to Talalay (7), the isomerase has 3 steroid-binding sites. These properties make the protein highly suitable for a study of the relationship between chemical structure of the binding site and its affinity for steroids. We may continue to expect valuable results on these fundamental problems from the analysis of steroid-enzyme interactions. A promising approach is seen in the affinity-labeling studies in Warren’s laboratory (8).

As may be expected from any fundamental research in a significant area, the study of steroid-protein interaction has contributed important fringe benefits to the field of steroid hormones (Fig. 1). One of them is the use of the binding proteins as reagents to measure very small quantities of steroid hormones. Displacement of radiolabeled steroid by the test steroid in a competitive binding system introduced more than 6 years ago by Beverlv Murphy (9) is the basis for many procedures of highly sensitive steroid determinations. Increased specificity has been obtained by rigorous purification of the steroid prior to binding analysis. The principle of this competitive binding method has also been applied for an assay of the binding protein.

A field of research closely related to the serum proteins may be considered as a second fringe benefit, namely the study of immunoproteins. Immunoproteins have been produced against protein-conjugated steroid hormones as antigens (10). These reactive antibodies have certain properties in common with specific steroid-binding serum proteins and with cellular receptor proteins. The unique design of these studies is to let nature produce binding sites which are tailor-made for a given steroid. The immunoglobulins can be applied for specific hormonal inactivation and for sensitive radio-immunoassays. The further development of this more recent application of steroid-protein interaction can be looked forward to with great interest and expectation.

Binding of steroid hormones to serum proteins was originally considered important for transport A closer look at steroid solubilities in aqueous media showed, however, that such function is not necessary for steroid hormones under most conditions because of sufficiently high water solubility. Now there is renewed interest in a possible transport function of steroid-binding serum proteins in connection with the entrance of the hormone into specific target cells and permeation through nuclear membranes. This question and that of the chemical relationship between the specific carriers in the serum and the receptor proteins, as well as other problems are waiting to be answered.

Another consequence of steroid hormone interaction with serum proteins is perhaps more obvious. It has been found in all cases investigated so far that the formation of the protein complex suppresses the biological activity of the steroid hormone (Table II). This was shown many years ago for the corticosteroid hormones by Slaunwhite and Sandberg(11,12), and subsequently by other investigators. The hormonal activity of progesterone is equally suppressed by complex formation with the three serum proteins, albumin, α1–acid glycoprotein (AAG or orosomucoid), and corticosteroid-binding globulin (CBG or transcortin). It follows clearly as a consequence of the inactivation, that changes in the concentration of the binding proteins, especially those with high binding affinity and low capacity, would provide a regulatory mechanism of hormonal function. Similarly, protein interaction may protect the circulating steroid hormones from chemical or enzymatic attack.

The steroid-protein complexes best known today are those between the steroid hormones and three serum proteins, which are available in pure form and are relatively well characterized: albumin, α1–acid glycoprotein and corticosteroid-binding globulin. Table III shows that these three types of steroid-binding proteins have molecular weights of the same order of magnitude. It should be mentioned that according to unpublished results from Dr. Baulieu’s laboratory, the sex steroid-binding β–globulin, which has a high affinity for testosterone and for estradiol, also has a molecular weight of 52,000 (19); about twice this size has been reported from other laboratories2

The table shows the great differences in the concentration of the steroid-binding proteins in the blood serum. A comparison with the hormone concentrations, which are given only as orders of magnitude, shows that the CBG capacity for the corticoids and other steroid hormones is not much greater than the normal steroid level. A quantitative relationship of this order would appear to be required for efficient regulation of...