Studies on human thyroxine-binding globulin. 8. Isoelectric focusing evidence for microheterogeneity of thyroxine-binding globulin

Highly purified thyroxine-binding globulin from pooled human serum homogeneous by conventional criteria, subjected to isoelectric focusing in polyacrylamide gels in a pH gradient from 3–6, produced a pattern of at least nine stainable protein bands. All of these bands appeared to bind thyroxine. Completely desialylated thyroxine-binding globulin subjected to isoelectric focusing produced the same number and pattern of bands located at a different area in the pH gradient. Thyroxine-binding globulin purified from the serum of a single donor was subjected to isoelectric focusing. This thyroxine-binding globulin had the same pattern of protein bands with the exception that one of the major bands seen in the thyroxine-binding globulin from pooled serum was absent. Several possible explanations for these phenomena are discussed.     

Thyroxine-protein interactions. Interaction of thyroxine and triiodothyronine with human thyroxine-binding globulin.

The effect of temperature on the binding of thyroxine and triiodothyronine to thyroxine-binding globulin has been studied by equilibrium dialysis. Inclusion of ovalbumin in the dialysis mixture stabilized thyroxine-binding globulin against losses in binding activity which had been found to occur during equilibrium dialysis. Ovalbumin by itself bound the thyroid hormones very weakly and its binding could be neglected when analyzing the experimental results. At pH 7.4 and 37 degrees in 0.06 M potassium phosphate/0.7 mM EDTA buffer, thyroxine was bound to thyroxine-binding globulin at a single binding site with apparent association constants: at 5 degrees, K = 4.73 +/- 0.38 X 10(10) M-1; at 25 degrees, K = 1.55 +/- 0.17 X 10(10) M-1; and at 37 degrees, K = 9.08 +/- 0.62 X 10(9) M-1. Scatchard plots of the binding data for triiodothyronine indicated that the binding of this compound to thyroxine-binding globulin was more complex than that found for thyroxine. The data for triiodothyronine binding could be fitted by asuming the existence of two different classes of binding sites. At 5 degrees and pH 7.4 nonlinear regression analysis of the data yielded the values n1 = 1.04 +/- 0.10, K1 = 3.35 +/- 0.63 X 10(9) M-1 and n2 = 1.40 +/- 0.08, K2 = 0.69 +/- 0.20 X 10(8) M-1. At 25 degrees, the values for the binding constants were n1 = 1.04 +/- 0.38, K1 = 6.5 +/- 2.8 X 10(8) M-1 and n2 = 0.77 +/- 0.22, K2 = 0.43 +/- 0.62 X 10(8) M-1. At 37 degrees where less curvature was observed, the estimated binding constants were n1 = 1.02 +/- 0.06, K1 = 4.32 +/- 0.59 X 10(8) M-1 and n2K2 = 0.056 +/- 0.012 X 10(8) M-1. When n1 was fixed at 1, the resulting values obtained for the other three binding constants were at 25 degrees, K1 = 6.12 +/- 0.35 X 10(8) M-1, n2 = 0.72 +/- 0.18, K2 = 0.73 +/- 0.36 X 10(8) M-1; and at 37 degrees K1 = 3.80 +/- 0.22 X 10(8) M-1, n2 = 0.44 +/- 0.22, and K2 = 0.43 +/- 0.38 X 10(8) M-1. The thermodynamic values for thyroxine binding to thyroxine-binding globulin at 37 degrees and pH 7.4 were deltaG0 = -14.1 kcal/mole, deltaH0 = -8.96 kcal/mole, and deltaS0 = +16.7 cal degree-1 mole-1. For triiodothyronine at 37 degrees, the thermodynamic values for binding at the primary binding site were deltaG0 = -12.3 kcal/mole, deltaH0 = -11.9 kcal/mole, and deltaS0 = +1.4 cal degree-1 mole-1. Measurement of the pH dependence of binding indicated that both thyroxine and triiodothyronine were bound maximally in the region of physiological pH, pH 6.8 to 7.7.     

Effect of deglycosylation on the binding and immunoreactivity of human thyroxine-binding globulin.

Thyroxine-binding globulin (TBG), prepared from human serum by an improved purification method, was treated with a mixture of neuraminidase, beta-galactosidase, alpha-mannosidase, and beta-N-aectylglucosaminidase, which resulted in the removal of approximately 86% of saccharides. Purification by thyroxine-Sepharose affinity chromatography gave a homogeneous protein as shown by equilibrium sedimentation and sodium dodecylsulfate-polyacrylamide gel electrophoresis. Amino acid and NH2-terminal sequence analysis indicated that the protein moiety was intact. Deglycosylation had no effect on the stoichiometry of the binding of L-thyroxine as shown by tryptophanyl fluorescence quenching and equilibrium dialysis at pH 8.6 and 25 degrees C. However, the affinity constant for L-thyroxine was reduced from 1.6 X 10(9) M-1 to 0.58 X 10(9) M-1. Analysis of radioimmunoassay data revealed that deglycosylation resulted in a slight decrease of the affinity constant for anti-TBG antibody from 3.9 X 10(10) M-1 to 1.8 X 10(10) M-1. These results suggest that the polypeptide moiety, rather than the heterosaccharides, contains the antigenic determinants. Removal of the majority of the heterosaccharides of TBG has only a minor effect on its immunoreactivity and on the binding of thyroid hormone.     

Comparative characteristics of molecular forms of human thyroxine-binding globulin

Two molecular variants, of thyroxin-binding globulin (TBG), TBG-1 and TBG-2, were obtained from human retroplacental blood by fractionation of pure TBG on concanavalin A-Sepharose. It was found that both variants are immunologically identical, have similar molecular weights, amino acid composition and spectral properties, and possess the same affinity for thyroid hormones. However, TBG-1 and TBG-2 differed in charge upon isoelectrofocusing and had different monosaccharide composition. The existence of two molecular variants of TBG in pregnancy is probably due to the peculiarities of the polypeptide chain glycosylation during TBG biosynthesis.     

Characterization of the thyroxine-binding site of thyroxine-binding globulin by site-directed mutagenesis.

The principal transport protein for T4 in human blood, thyroxine-binding globulin (TBG), binds T4 with an exceptionally high affinity (Ka = 10(10) M(-1)). Its homology to the superfamily of the serpins has recently been used in the design of chimeric proteins, providing experimental evidence that an eight-stranded beta-barrel domain encompasses the ligand-binding site. We have now characterized the T4 binding site by site-directed mutagenesis. Sequence alignment of TBG from several species revealed a phylogenetically highly conserved stretch of amino acids comprising strands 2B and 3B of the beta-barrel motif. Mutations within this region (Val228Glu, Cys234Trp, Thr235Trp, Thr235Gln, Lys253Ala, and Lys253Asp), designed to impose steric hindrance or restriction of its mobility, had no significant influence on T4 binding. However, binding affinity was 20-fold reduced by introduction of an N-linked glycosylation site at the turn between strands 2B and 3B (Leu246Thr) without compromising the proper folding of this mutant as assessed by immunological methods. In most other serpins, this glycosylation site is highly conserved and has been shown to be crucial for cortisol binding of corticosteroid-binding globulin, the only other member of the serpins with a transport function. The ligand-binding site could thus be located to a highly aromatic environment deep within the beta-barrel. The importance of the binding site's aromatic character was investigated by exchanging phenylalanines with alanines. Indeed, these experiments revealed that substitution of Phe249 in the middle of strand 3B completely abolished T4 binding, while the substitution of several other phenylalanines had no effect.     

Pregnancy-associated molecular variants of human serum transcortin and thyroxine-binding globulin

Affinity chromatography on immobilised concanavalin A revealed that transcortin and thyroxine-binding globulin isolated from human postpartum serum contained ～10% of molecular variants that did not occur in these glycoproteins isolated from normal donor serum (both male and female). The chromatographic behaviour of the pregnancy-associated glycoprotein variants, their monosaccharide compositions, and the results of methylation analysis indicated that these variants contained only triantennary oligosaccharide chains of the N-acetyl-lactosamine type.     

Characterization of human thyroxine-binding globulin. Evidence for a single polypeptide chain.

Thyroxine-binding globulin (TBG) was purified from fresh human plasma by affinity, anion exchange, and gel filtration chromatography. The protein gave a single band in overloaded analytical disc gel electrophoresis. The molecular weight was 54,000 and E1%/1 cm at 280 nm, corrected for thyroxine (T4) absorbance, was 6.17. Six preparations of TBG contained from 0.09 to 0.64 mol of T4/mol; the TBG used in this study contained 0.19 mol of T4 and was able to bind an additional 0.85 mol. The carbohydrate composition was determined and accounted for 23% of the molecular weight. Four lines of chemical and physical evidence failed to demonstrate subunits. These included quantitative COOH-terminal amino acid analysis, peptide mapping and amino acid composition, treatment with sodium dodecyl sulfate, and denaturation of the reduced, alkylated protein with guanidine. From these data, we conclude that TBG is a single polypeptide chain.     

Genetic studies on human thyroxine-binding globulin (TBG)

Summary An enzyme immunoassay technique combined with Western blotting is described to demonstrate thyroxine-binding globulin (TBG) by isoelectric focusing in thin-layer polyacrylamide gels with 8mol/l urea. Quantitative evaluation was by laser densitometry. No genetic charge variants of TBG were encountered in a sample of 840 unrelated individuals from southwestern Germany. There was no correlation between structural and quantitative variations in the TBG protein. Results from a family with quantitative TBG deficiency strongly support the postulated X-linked mode of inheritance. The method described can be considered as an additional diagnostic tool in thyroid evaluation.     

Comparison of human thyroxine-binding globulin purification by affinity chromatography procedures

Three procedures for the isolation of thyroxine-binding globulin from human serum, using affinity chromatography on triiodothyronine (T3) linked to Sepharose (A), thyroxine (T4) linked to Sepharose (B) or T3 linked to epoxy-Sepharose (C) as the first purification step, were compared. With the use of additional purification steps, the three procedures yielded pure thyroxine-binding globulin without desialylation. With procedure A, the initial binding of T4-binding globulin to T3-Sepharose was very low, yielding a poor final recovery (17%). Procedure B gave the highest yield (35%) after a three-step purification, with a low T4 content (0.15-0.30 mol/mol). Procedure C also gave a high yield (28%) after only two purification steps, with a T4 content greater than 0.7 mol/mol. The microheterogeneity of T4-binding globulin obtained with these three procedures was demonstrated by isoelectric focusing: five major bands were observed between pH 4.1 and 4.6, and intermediate faint bands (often doublets) in the same pH range. However, with procedures A and C, the most acidic bands (pH 4.10-4.20) were always absent. Thyroxine-binding globulin was preincubated with radioactively labelled T3 or T4 and the hormone-protein complex was analyzed by isoelectric focusing. The binding of T3--compared to that of T4--was reduced in the most acidic protein subspecies. These results suggest differences in the thyroid hormone binding properties of the various subspecies of human T4-binding globulin.     

Preparation of glycopeptides and oligosaccharides from thyroxine-binding globulin.

Over 99% of thyroxine (T4), the major form of thyroid hormone in plasma, is bound to the plasma glycoprotein thyroxine-binding globulin (TBG). The carbohydrate composition of TBG (14.6% by weight) consists of mannose, galactose, N-acetylglucosamine, and N-acetylneuraminic acid in the molar ratios of 11:9:16:10 per mol of glycoprotein. No fucose or N-acetylgalactosamine were detected. Amino acid analyses were performed. Glycopeptides, prepared by exhaustive pronase treatment of the glycoprotein, were separated by gel filtration and ion exchange chromatography. All glycopeptides contained the four sugars present in the native glycoprotein. One-fourth of the glycopeptide fraction was resolved into a discrete component, glycopeptide I. The remaining glycopeptides were a mixture termed glycopeptides II and III. Glycopeptides II and III were resolved into two discrete carbohydrate units, termed oligosaccharides A and B, by alkaline-borohydride treatment and DEAE-cellulose chromatography. We propose that TBG contains four oligosaccharide chains as calculated from the molecular weights of the glycopeptides and from compositional data assuming 1 asparagine residue/glycopeptide. The carbohydrate structures of the glycopeptides and relative affinities of TBG, glycopeptides and oligosaccharides for hepatocyte plasma membrane binding are presented in the accompanying paper (Zinn, A.B., Marshall, J.S., and Carlson, D.M. (1978) J. Biol. Chem. 253, 6768-6773.     

Binding of thyroid hormones and their analogues to thyroxine-binding globulin in human serum.

The present study was undertaken to study the binding of several thyroid hormones and structurally related compounds to human serum thyroxine-binding alpha-globulin (TBG). The source of TBG was normal human serum diluted 1:100 in 0.035 M barbital buffer, pH 7.4. In the binding assays, 125I-thyroxine, unlabeled thyroxine, and diluted serum were incubated for 20 h at 37 degrees in Plexiglas equilibrium dialysis units. Two orders of binding sites were discerned: a high affinity, low capacity binding site with an affinity constant of approximately 2.5 X 10(9) M-1, and a low affinity, very high capacity binding site with an affinity constant of less than 10(6) M-1. Studies with purified TBG, serum deficient in TBG, and purified human serum albumin indicated that the high affinity site represented binding to TBG and the low affinity site represented binging to albumin. The ability of several groups of thyroid hormone analogues to bind to TBG was then investigated. As a result of these studies, the following structural features of thyroid hormones were found to be important for optimal binding activity: (a) the L-alanine side chain conformation, (b) the presence of a 4'-hydroxyl group, (c) the presence of two substituents in the inner and outer rings (positions 3, 5, 3', and 5'), and (d) the presence of either bromines or iodines in the inner ring and iodines in the outer ring. Of lesser importance was the presence of an oxygen atom in the ether position.     

Sex hormone-binding globulin and thyroxine-binding globulin levels in premature thelarche

Premature thelarche is defined as the isolated development of breast tissue in girls less than 8 years of age. Although breast development is an estrogen-dependent process, these girls do not have elevated serum estrogen levels, and the hormonal basis for their condition is unclear. We studied the levels of two estrogen-dependent transport proteins, sex hormone-binding globulin (SHBG) and thyroxine-binding globulin (TBG), in order to determine if there was evidence for a more subtle estrogen effect in girls with premature thelarche. SHBG levels in girls with premature thelarche were not significantly different from those of prepubertal girls of the same ages and were significantly lower than those in girls undergoing pubertal development at the appropriate age (P less than 0.05) and in normal women (P less than 0.001). There was no statistically significant difference in TBG levels between the girls with premature thelarche and prepubertal controls. There was also no significant difference in TBG levels between prepubertal girls and girls in early puberty. In contrast, women had TBG levels that were significantly lower than those in all girls studied. We conclude that the estrogen exposure (whether endogenous or exogenous) of girls with premature thelarche is less than that of girls in early true puberty and similar to that of other prepubertal girls. Further, changes in serum TBG are not as sensitive an indicator of estrogen effect as is breast development or changes in SHBG. This study also suggests that large amounts of exogenous estrogens are not an element in the development of premature thelarche.     

New aspects of isolation and purification of thyroxine-binding globulin from human biological fluids

Based on the new data concerning the multicomponent system of thyroxine-binding proteins in human plasma, some methodological aspects of isolation and purification of thyroxine-binding globulin (TBG) were examined, and a simple two-step procedure for TBG purification was developed. Normal human blood serum, retroplacental serum and amniotic fluid were used as TBG sources. The procedure includes affinity chromatography and adsorption chromatography on a hydroxyapatite column. A biospecific adsorbent was synthesized by stepwise binding of epichlorohydrin and thyroxine to Sepharose. The yield of pure TBG varied from 60 to 80%, depending on the TBG source used. The properties of TBG preparations from retroplacental serum and amniotic fluid were identical; both preparations contained a pregnancy-associated molecular variant, TBG-1. Two novel serum thyroxine-binding proteins were detected, isolated and partly characterized.     

Concentration of thyroxine-binding globulin: value of direct assay.

The concentration of thyroxine-binding globulin (TBG) in the serum can now be measured by direct assays that are simple and inexpensive. Comparison of a direct measurement of TBG concentration with a widely used indirect method (Thyopac-3) showed that the indirect method was inaccurate when TBG concentrations were high. This will result in an increase in the derived free thyroxine index (FTI), so that euthyroid patients with a raised TBG concentration may be at risk of being labelled thyrotoxic. Correction of serum total thyroxine (T4) concentration according to the actual TBG concentration (T4:TBG ratio) provided a better correlation with thyroid state than FTI.     

Partial amino acid sequence of human thyroxine-binding globulin. Further evidence for a single polypeptide chain.

The NH₂-terminal amino acid of highly purified thyroxine-binding globulin has been identified by dansyl chloride, cyanate and Edman degradation methods. All three gave alanine as the only amino terminal residue. Carbamylation and Edman degradation of the denatured protein yielded 0.86 and 0.98 – 1.05 mole of alanine per mole of protein, respectively. These data further indicate that thyroxine-binding globulin is composed of a single polypeptide chain. Automated Edman degradation gave the partial sequence as: Ala-Ser-Pro-Glu-Gly-Lys-Val-Thr-Ala-Asp-Ser-Ser-Ser-Gln-(Pro)-X-Ala-(Ser)-Leu-Tyr- A computer search revealed no homology of the NH₂-terminal segment of thyroxine-binding globulin with human prealbumin. The NH₂-terminal portion of prealbumin contains part of the thyroxine binding site.