Interaction between rat alpha 1 fetoprotein and fluorescent derivatives of estrone in relation to the position and type of the fluorescent label

A series of derivatives of estrone with fluorescent dyes (dansyl or coumarin) coupled at different positions on the steroid molecule, have been synthesized. These derivatives were tested for their quantum yields and their binding properties were determined with respect to rat alpha 1 fetoprotein. Derivatives at C-3 (estrone-3-hemisuccinate-dansyl-cadaverine and estrone-3-dansyl) compete with estrone for binding to the fetal protein; however derivates of estrone at C-6 (estrone-6-carboxymethyloxime-dansyl-cadaverine) and at C-17 (estrone-17-carboxymethyloxime-coumarine, estrone-17-carboxymethyloxime-dansyl-cadaverine and estrone-17-dansyl-hydrazine) compete poorly or not at all. The association constant of the radioactive derivative estrone-3-dansyl [3H] with rat alpha 1 fetoprotein was measured directly: the same number of high affinity binding sites (0.6) as that for estrone was found with an apparent association constant of 3.7 X 10(6) M-1. In addition to the high affinity binding sites, a low affinity class of binding sites was found which corresponds to the binding of the dansyl fraction of the fluorescent steroid derivative.     

Binding of steroid hormones and anabolic agents to bovine sex-hormone binding globulin

Binding of some selected steroids and anabolic agents to bovine sex-hormone binding globulin (SHBG) was investigated. SHBG binding affinities, relative to the reference hormone 5 alpha-dihydrotestosterone, were estimated for the compounds. The results demonstrate that binding of steroid hormones to SHBG is facilitated by the 17 beta-hydroxyl group, possibly involving hydrogen binding, and by the methyl group at C-19 of the steroid moiety. Structural modifications at C-17 of a steroid molecule involving esterification, epimerization or reduction of the 17 beta-hydroxyl group, or introduction of a bulky 17 alpha group have the effect of decreasing the SHBG binding affinity of the steroid molecule.     

Substrate recognition at the cytoplasmic and extracellular binding site of the lactose transport protein of Streptococcus thermophilus

The lactose transport protein (LacS) of Streptococcus thermophilus catalyzes the uptake of lactose in an exchange reaction with intracellularly formed galactose. The interactions between the substrate and the cytoplasmic and extracellular binding site of LacS have been characterized by assaying binding and transport of a range of sugars in proteoliposomes, in which the purified protein was reconstituted with a unidirectional orientation. Specificity for galactoside binding is given by the spatial configuration of the C-2, C-3, C-4, and C-6 hydroxyl groups of the galactose moiety. Except for a C-4 methoxy substitution, replacement of the hydroxyl groups for bulkier groups is not tolerated at these positions. Large hydrophobic or hydrophilic substitutions on the galactose C-1 alpha or beta position did not impair transport. In fact, the hydrophobic groups increased the binding affinity but decreased transport rates compared with galactose. Binding and transport characteristics of deoxygalactosides from either side of the membrane showed that the cytoplasmic and extracellular binding site interact differently with galactose. Compared with galactose, the IC(50) values for 2-deoxy- and 6-deoxygalactose at the cytoplasmic binding site were increased 150- and 20-fold, respectively, whereas they were the same at the extracellular binding site. From these and other experiments, we conclude that the binding sites and translocation pathway of LacS are spacious along the C-1 to C-4 axis of the galactose moiety and are restricted along the C-2 to C-6 axis. The differences in affinity at the cytoplasmic and extracellular binding site ensure that the transport via LacS is highly asymmetrical for the two opposing directions of translocation.     

Conservation of the DNA binding domain and other properties between porcine and rat glucocorticoid receptors

Activated glucocorticoid receptor protein (GCR) was partially purified from porcine liver cytosol by sequential chromatography on phosphocellulose and DNA-cellulose using a modification of a protocol developed for purification of rat GCR. This partially purified preparation, when separated by SDS-polyacrylamide gel electrophoresis and immunoblotted, indicated that a Mr = 94,000 protein band cross-reacts with a monoclonal antibody against rat GCR. A nitrocellulose filter binding assay showed that both the partially purified porcine and rat GCRs interact specifically with a cloned synthetic 24 base pair deoxyoligonucleotide containing the GCR binding sequence in the first intron of the human growth hormone (hGH) gene. This specific protein-DNA interaction is blocked by a single base pair change in the binding site. All three putative domains of the GCR molecule: the steroid binding, immunoreactive, and DNA binding have been conserved between two divergent species.     

The amino acid sequence of the sex steroid-binding protein of rabbit serum

The amino acid sequence of the sex steroid-binding protein (SBP or SHBG) of rabbit serum, specific for binding testosterone and 5 alpha-dihydrotestosterone, was determined using a complementary combination of mass spectrometric and Edman degradation techniques. The monomeric unit of the homodimeric protein is a single chain glycopeptide of 367 amino acid residues, with N-linked oligosaccharide side chains at Asn-345 and Asn-361 and disulfide bonds connecting Cys-158 to Cys-182 and Cys-327 to Cys-355. The polypeptide molecular weight of the monomer calculated from the sequence is 39,769. The molecular weight of the homodimer including 9% carbohydrate is 87,404. The sequence contains a relatively hydrophobic segment between Trp-241 and Leu-282, which includes many leucine residues in an alternating pattern. An amino acid sequence repeat is also located within that segment. Both of these patterns are present in human SBP and in the androgen-binding protein of rat epididymis. The sequence data indicate that the previously reported microheterogeneity of rabbit SBP in sodium dodecyl sulfate-polyacrylamide gel electrophoresis reflects variants generated by differential glycosylation of the monomer rather than different gene products. Seventy-nine percent of the amino acids of rabbit SBP are identical to those of human SBP; rabbit SBP thus joins human SBP and rat androgen-binding protein in one gene family that is distinct from the steroid hormone receptor superfamily. It appears that the problem of binding sex steroid hormones has been solved independently in two different gene families that contain completely different steroid-binding domains. Since the nonhomologous steroid-binding domains of both families of proteins recognize essentially the same steroid structure, it will be interesting to determine the structural basis of the two different protein designs that lead to similar steroid-binding specificity.     

Differences in steroid specificity for rat androgen binding protein and the cytoplasmic receptor.

Two proteins in the rat, androgen binding protein (ABP) and the cytoplasmic receptor (CR), have high affinity and limited capacity for binding androgens. To determine the structural requirements for binding with high affinity, each protein was partially purified and the ability of over 100 steroids to compete with [3H]dihydrotestosterone (17 beta-hydroxy-5 alpha-androstan-3-one) for binding sites was assessed. The results indicate marked differences in the steroid specificities of the two proteins. Some alterations of dihydrotestosterone at C-2 or C-2 and C-3 increase binding to ABP two to four-fold. Similarly, the affinity of 17 beta-hydroxy-7 alpha-methyl-4-estren-3-one for ABP increases two-fold when a double bond is created at C-14. Addition of a methyl group in the alpha position at C-7 or C-17, or an ethinyl group at C-17 cause little change in affinity; however, modifications at C-11 and C-17 beta, and deletion of the methyl group at C-10 significantly impair binding to ABP. Binding to the CR is maintained or increased by deletion of the methyl group at C-10. Binding is lessened by modifications at C-3 and C-17 beta. Most alterations at C-2, C-7, C-11, and C-17 alpha have only minor effects on binding to the CR. These studies should provide a molecular basis for predicting the effects of specific structural modifications. When some modifications at C-2 or C-2 and C-3 are combined with changes at C-17 beta, the resulting steroids retain very high affinity for ABP and very limited binding to the CR. Such steroids may provide a means for assessing the function of ABP.     

An androgen binding protein in the testis cytosol fraction of adult rats. Comparison with the androgen binding protein in the epididymis

The cytosol fraction of rat testicular homogenates contained a specific androgen binding protein (t-ABP), indistinguishable from, the androgen binding protean in. the epididymis(e-ABP) in terms of its chromatographic behaviour by gel filtration, sedimentation rate, mobility on polyacrylamide electrophoresis and steroid specificity(5α-diaydrotestosterone(DHT) > testosterone > estradiol-17β > parogesterone and estradiol-17α).The stability of t-ABP as well was similar to that of e-ABP. The aBP-DHT complexes were stable between pH 6.5–10, exhibiting the highest degree of binding between pH 8–9. The binding activity persisted after heating at 50°C for 30 min., whereas heating at 60°C for 30 min. completely destroyed the binding. DHT did not significantly increase the stability towards pH and temperature denaturation. Binding activity was not reduced by 1 mM p-chloro-mercuri-phenyl-sulphonate (PCMPS), whereas similar treatment with 1 nM N-ethyl-maleimide(NEM) or 1 mM Ellman's reagent reduced it by some 10–50 per cent. Exposure to 10 mM β-mercaptoethanol(βME) reduced the binding by 60–70 per cent. These studies strongly indicate that t-ABP and e-ABP are identical proteins.Hemicastration for 4 weeks eliminated the ABP from the epididymal cytosol fraction on the castrated side, indicating that this protein is produced by the testis.     

Functional analysis of the purified glucocorticoid receptor

Glucocorticoid-receptor complex (GR) has been purified from rat liver by differential affinity for DNA before and after activation, followed by ion-exchange chromatography. The purified GR has mol. wt 94,000 dalton. The protein contains three functional domains: (A) a steroid-binding domain; (B) a DNA-binding domain; and (C) a domain necessary for normal biological function. A second protein, with mol. wt 72,000 dalton, copurifies with the GR. This protein does not bind steroid, does not interact with antibodies raised against the GR and does not show the same susceptibility to limited proteolytic cleavage as the 94,000 dalton protein. Analysis of the specific interaction of the purified GR with the mouse mammary tumour virus gene, assayed by glycerol-gradient centrifugation, shows that one molecule of 94,000 dalton protein binds to each of the specific binding sites in the long terminal repeat region. Analysis of the fractions from the glycerol gradients show that the 72,000 dalton protein is associated to the binding species (94,000 dalton receptor protein) in about equimolar amounts. Analysis of the molybdate-stabilized non-activated receptor complex using monoclonal antibodies raised against the 94,000 dalton receptor protein indicates that the molybdate-stabilized complex is a hetero-oligomer. The hetero-oligomer consists of only one molecule of the 94,000 dalton receptor protein, in association with other non-steroid-binding proteins.     

The comparative specificity of 3 oestradiol-binding proteins. Rat alpha-foetoprotein, rat liver 17beta-hydroxy steroid dehydrogenase and anti-(oestradiol-6-carboxymethyloxime-bovine serum albumin) antiserum.

1. The specificity of 3 oestradiol-binding proteins was studied. Two of these proteins are naturally occurring (rat alpha-foetoprotein and rat liver microsomal 17beta-hydroxy steroid dehydrogenase) and the third is an artificially induced model, anti-(oestradiol-6-carboxymethyloxime-bovine serum albumin) gamma-globulins. 2. A specific binding procedure for each protein model permitted a determination of its affinity for oestradiol and for 30 other steroids. 3. The results obtained have brought to light the different areas of the steroid molecule that are important for its recognition by each of the three proteins. The two naturally occurring proteins (alpha-foetoprotein and 17beta-hydroxy steroid dehydrogenase) recognize the edge of the steroid defined by C-4, C-6, C-8 and C-15. On the other hand, the gamma-globulins recognize the opposite edge, i.e. that defined by C-2, C-10, C-11 and C-17. 4. Diethylstilboestrol, whose structure is analogous to that of a steroid, is only recognized by the two naturally occurring proteins.     

Corticosteroid binding globulin, testosterone-estradiol binding globulin, and androgen binding protein belong to protein families distinct from steroid receptors

The cDNA nucleotide sequences and the deduced amino acid sequences of human corticosteroid binding globulin (hCBG), human testosterone-estradiol binding globulin (hTeBG), and rat androgen binding protein (rABP) were determined. Studies of the steroid binding sites suggest they are toward the carboxy-terminus in hTeBG and rABP and more central in hCBG. hCBG has remarkable sequence homology with members of a superfamily whose functions have diverged; these include thyroxine-binding protein, serine protease inhibitors, egg white proteins, and angiotensinogen. hTeBG and rABP have a 68% amino acid sequence identity. Hybridization studies suggest that hTeBG is probably even more closely related, if not identical, to hABP. The carboxy-terminal sequences of hTeBG and rABP are also similar to that of protein S, a vitamin-K-dependent clotting factor. There were no nucleotide or amino acid sequence homologies between hCBG, hTeBG, or rABP and other steroid binding proteins such as steroid receptors, albumin, alpha-fetoprotein, and vitamin D binding protein. We conclude that the "extracellular steroid binding proteins" and steroid receptors do not appear to have descended from a common ancestor.     

Progesterone-binding globulin interaction with its steroid ligands: study of the protein binding site topography using spin labeled steroids and electron spin resonance spectroscopy

The binding site topography of progesterone-binding globulin (PBG) purified from pregnant guinea pig serum was examined using synthesized spin-labeled ligands and electron spin resonance (ESR) spectroscopy. A series of deoxycorticosterone-nitroxide (DOC-NO) derivatives were prepared, bearing the free radical on the side chain at increasing distance (d) from the steroid nucleus. The ability of the spin-labeled steroids to specifically bind to PBG was assessed by measurement of their relative binding affinity as compared to progesterone. ESR spectra of the bound steroid nitroxide radical were used to calculate the rotational correlation times tau c for the nitroxides as a function of their distance d to the protein-bound steroid nucleus. The data showed that the side chain nitroxide exhibited an unrestrained rotation in a water-like environment when d reached about 18 A. This would correspond to a PBG steroid binding site depth of about 28 A and suggests that the bound steroid in the PBG site is oriented with the side chain at C-17 directed toward the outside of the protein binding crevice.     

The role of asparagine-linked oligosaccharides in the subunit structure, steroid binding, and secretion of androgen-binding protein.

Testicular androgen-binding protein (ABP) and liver sex hormone-binding globulin are encoded by the same gene. These proteins have the same primary amino acid sequences, but they differ in attached oligosaccharides; the differences are presumably due to cell-specific glycosylation mechanisms. To investigate the role of oligosaccharides in ABP/sex hormone-binding globulin subunit structure, secretion, and steroid binding, mutant rat ABP proteins were constructed that eliminated one or both of the two potential sites of asparagine (Asn)-linked glycosylation. Immunoblot analysis of wild type recombinant ABP yielded the typical heterogeneous banding pattern. Secreted ABP was composed of two protomers of M(r) 46,000 and M(r) 43,000, while cellular ABP yielded three mol wt species (M(r) 43,000, 41,000, and 39,000). Substitution of the Asn residue in either consensus sequence for Asn-linked glycosylation with an Ile residue resulted in increased mobility of the immunoreactive ABP species. These changes are consistent with the loss of an Asn-linked oligosaccharide. Substitution of both Asn residues yielded a single immunoreactive species in the medium and cell extracts that migrated as a M(r) 39,000 protein. These results demonstrate that the mol wt heterogeneity of ABP is due to differential Asn-linked glycosylation of both potential sites. All three mutant forms of ABP were secreted by the COS cells. However, the amount of immunoreactive ABP and [3H]5 alpha-dihydrotestosterone binding in the medium was lower than wild type (100%) in one of the single mutants (65%) and in the double mutant (29%). Unlike the glycosylation mutants, alteration of other residues, not involved in glycosylation, yielded cellular ABP and no detectable medium ABP.(ABSTRACT TRUNCATED AT 250 WORDS)     

Hydrophobic cluster analysis (HCA) of the hormone-binding domain of receptor proteins

A new technique of protein sequence analysis, namely, Hydrophobic Cluster Analysis (HCA), has been used to align and compare the sequences of proteins belonging to the receptor superfamily (steroid, thyroid hormone and retinoic acid receptors) and serpin superfamily (corticosteroid binding globulin (CBG) and alpha 1-antitrypsin (alpha 1-AT]. By matching up clusters of hydrophobic amino-acids that oftenmost correspond to identifiable secondary structures (alpha-helices, beta-strands etc.), it has been possible to deduce the following information on the secondary structures of these proteins: CBG is structurally related to alpha 1-AT (HCA score greater than 80%), the structures of the hormone-binding domains of the steroid receptors that bind 3-keto-delta 4-steroids are closely interrelated (greater than 80%) but less closely related to that of the estrogen receptor (ER) (approximately 75%), vitamin D, retinoic acid and thyroid hormone receptors are structurally closely related (greater than or equal to 80%). Their secondary structures are, however, also related to that of the steroid receptors (approximately 70%), and a high degree of analogy exists between the structures of serpins and of the hormone-binding domains of members of the steroid superfamily (60-70%). HCA has clearly shown that a previous local sequence alignment of the estrogen receptor with other steroid receptors and cytochromes P450 has to be reconsidered. The published consensus steroid binding sequence previously identified in cytochromes is in fact 80 amino-acids upstream from its previously defined position. Other regions of contiguous sequence identity have also been identified which may be involved in the hydrophobic core of the protein or in steroid binding. Their positions have been indicated using the crystal structure of alpha 1-AT as a model.     

Corticosteroid-binding globulin

Among the numerous corticosteroid-binding proteins that have been demonstrated to exist throughout the vertebrates, the corticosteroid-binding globulins (CBG or transcortin) of man and guinea pig have been most thoroughly investigated. Both are glycoproteins that contain one binding site for corticosteroid hormones and other steroids. The binding forces are predominantly of a hydrophilic nature, in addition to hydrophobic interaction. Human CBG forms a reversible dimer that maintains full binding affinity and capacity. The monomeric form is composed of two electrophoretic variants that appear indistinguishable in their chemical, immunological, and steroid binding properties. Thermodynamic analysis indicates different binding mechanisms for human and cavian CBG at the lower and higher temperature range. Entrance of hydroxy groups into the steroid molecule increases the affinity to CBG; the contributions to the free energy of complex formation, resulting from individual substitutions or other alterations in the steroid molecule, are additive. Determination of binding specificity leads to a conceptual image of the steroid binding site with respect to the nature of the various binding domains.     

Plectin and IFAP-300K are homologous proteins binding to microtubule-associated proteins 1 and 2 and to the 240-kilodalton subunit of spectrin

Structural and functional characteristics of plectin from intermediate filament preparations of rat glioma C6 cells were compared to those of the intermediate filament-associated protein of Mr = 300,000 (IFAP-300K) of baby hamster kidney cells (Yang, H.-S., Lieska, N., Goldman, A.E., and Goldman, R.D. (1985) J. Cell Biol. 100, 620-631). After radiolabeling and proteolytic digestion under varied conditions, both proteins yielded nearly identical peptide maps. Immunological cross-reactivity, co-migration on one- and two-dimensional high-resolution gels, chromatofocusing, and amino acid analysis demonstrated structural homology as well. In vivo labeling with 32Pi showed that plectin was the target for cAMP-independent protein kinases which phosphorylated 18-kDa domains at the end(s) of the molecule. Previously reported phosphorylation sites for cAMP-dependent and a newly identified site for Ca2+/calmodulin-dependent protein kinases were located on different domains. In solid-phase binding assays, plectin bound to vimentin, microtubule-associated proteins 1 and 2, the 240-kDa chain of brain fodrin, and alpha-spectrin from human erythrocytes. Similar characteristics were revealed for corresponding 300-kDa components of various other cell lines, supporting the concept that plectin is a general cytoskeletal cross-linking element, probably of multiple function.     

Progesterone-binding site of adult male rat liver microsomes

We investigated in greater detail the relationship between steroid structures and their binding affinities to the microsomal progesterone-binding site of the adult male rat liver. Only six steroids of the 100 compounds tested, namely, progesterone, 3 alpha-hydroxy-5 alpha-pregnan-20-one, 3 alpha,11 beta-dihydroxy-5 alpha-pregnan-20-one, 5 alpha-pregnane-3,20-dione, pregnenolone, and pregnenolone-3-sulfate, had high binding affinities to this site. Thus, we were able to draw a conclusion on the steroid structure which should be required for the best interaction with this site. That is, regardless of the whole molecule's lipophilicity, such a steroid should possess not only a planar A-B ring configuration but also the same side chains as progesterone with C-17 beta-acetyl and non-hydroxyl groups. The exception to this are hydrophilic C-3 groups, which may somewhat increase binding affinities in some cases. We compared the steroid specificity of this binding site with those of various other progesterone-binding components. As a result, this site appears to be a novel type of progesterone binder. We, furthermore, examined the relationship between this microsomal progesterone-binding site and the microsomal progesterone-metabolizing activity. The results, although preliminary, suggest that this binding site does not participate universally in the progesterone-metabolizing processes.     

The isolation and characterization of specific 3-methylcholanthrene-binding proteins from rat liver cytosol

The major proteins to which 3-methylcholanthrene specifically binds have been purified over 480-fold with a 45% yield compared to a rat liver 100,000g supernate. The procedure involved a batch ion-exchange technique together with hydrophobic gel filtration and chromatofocusing chromatography. The multiple, specific 3-methylcholanthrene-binding proteins obtained from this protocol had apparent isoelectric points of pH 6.3, 6.0, 5.7, and 5.5 on elution from a chromatofocusing column. They all shared a common sedimentation coefficient as determined by sucrose gradient analysis of 4.4 S. Gel filtration on Sephadex G-75 gave a common Stokes radius of 27 A. An analysis of these chromatofocusing peaks by sodium dodecyl sulfate-polyacrylamide gel electrophoresis showed those which eluted at pH 6.3 and 6.0 to contain two major protein bands of Mr 32,000 and 34,000, together with several contaminating proteins. In contrast, the peaks from chromatofocusing which eluted at pH 5.7 and 5.5 contained three major proteins of Mr 40,000, 25,000, and 14,000. The specific binding capability of these chromatofocusing peaks was found to be unstable to temperatures of -30 degrees C and below. Competition studies showed that these proteins were not steroid receptors, and that only polycyclic aromatic hydrocarbons which could induce cytochrome P-450c were able to displace 3-methylcholanthrene from the binding site. A marked preference was noted for polycyclic aromatic hydrocarbons with four to five benzene rings arranged in a nonlinear fashion, suggesting the stereochemical requirements of the protein binding site. The stability of the noncovalent interaction between the proteins and 3-methylcholanthrene was in the range of pH 7 to 9.     

Genetic mapping of the gene for androgen-binding protein/sex hormone-binding globulin to mouse chromosome 11

Southern blot analysis of DNAs from Chinese hamster x mouse and rat x mouse somatic cell hybrids showed that the mouse gene encoding androgen-binding protein/sex hormone-binding globulin (ABP-SHBG) is on Chromosome 11. Progeny from an intersubspecies backcross were analyzed to position this locus, termed Shbg, between Il-3 and Int-4 in the middle of this chromosome. Shbg is thus closely linked to several neurological mutations, one of which, Tr, is also associated with male sterility. The recent finding that ABP-SHBG is found throughout the rat brain raises the possibility that one of these mutations may be due to a defect in Shbg.     

Isolation and characterization of a 14-kDa albumin-binding fragment of streptococcal protein G

Protein G, a streptococcal cell wall protein, has separate binding sites for human albumin and IgG. Streptococci expressing protein G were treated with the bacteriolytic agent mutanolysin. Several IgG- and human serum albumin (HSA)-binding peptides were identified in the material thus solubilized and one of these, a 14-kDa peptide, was found to bind HSA but not IgG in Western blot experiments. This molecule was purified by affinity chromatography on Sepharose coupled with HSA followed by gel filtration on Sepharose 6B and a final affinity chromatography on IgG-Sepharose, by which low Mr W(15 to 20 kDa)IgG-binding peptides were removed. In different binding experiments the purified 14-kDa peptide bound exclusively HSA and the equilibrium constant between the peptide and HSA was determined to be 3.4 X 10(8) M-1. The relation between the 14-kDa molecule and protein G was studied by analyzing the N-terminal amino acid sequence of the peptide and comparing it with the previously determined protein G sequence. The 40 N-terminal amino acids were found to be identical with an amino acid sequence starting at position 62 in the protein G molecule. These and previous data enabled us to locate the albumin binding to the repetitively arranged domains in the N-terminal half of the protein G molecule.     

The cDNA-deduced primary structure of human sex hormone-binding globulin and location of its steroid-binding domain

We have sequenced a cDNA for sex hormone-binding globulin (SHBG) isolated from a phage lambda gt11 human liver cDNA library. The library was screened with a radiolabeled rat androgen-binding protein (ABP) cDNA, and the abundance of SHBG cDNAs was 1 in 750,000 plaques examined. The largest human SHBG cDNA (1194 base-pairs) contained a reading frame for 381 amino acids. This comprised 8 amino acids of a signal peptide followed by 373 residues starting with the known NH2-terminal sequence of human SHBG, and ending with a termination codon. The predicted polypeptide Mr of SHBG is 40,509, and sites of attachment of one O-linked (residue 7) and two N-linked oligosaccharide (residues 351 and 367) chains were identified. Purified SHBG was photoaffinity-labeled with delta 6-[3H]testosterone and cleaved with trypsin. The labeled tryptic fragment was isolated by reverse-phase HPLC, and its NH2-terminal sequence was determined. The results suggest that a portion of the steroid-binding domain of SHBG is located between residue 296 and the 35 predominantly hydrophilic residues at the C-terminus of the protein.