Effects of anion binding on the conformations of the two domains of ovotransferrin

A previous paper (Harris (1985) Biochemistry 24, 7412-7418) reported the occurrence of two classes of anion binding sites in transferrin. To evaluate the locations of the two anion binding sites in relation to the two major domains of transferrin we determined the binding constants of whole ovotransferrin and its two half-molecules by means of the difference UV spectroscopic technique. Anions induced strong negative absorbance at 245 nm in the order: citrate greater than phosphate greater than bicarbonate for whole ovotransferrin and the N-terminal half-molecule; and: phosphate greater than citrate greater than bicarbonate for the C-terminal half-molecule. The anion dissociation constants of the N-terminal half-molecule were consistent with lower dissociation constants, and those of the C-terminal half-molecule, with higher dissociation constants of whole ovotransferrin, indicating that the two classes of anion binding sites correspond to the binding sites in individual structural domains. Anion binding markedly protected the N-terminal half-molecule, but not the C-terminal half-molecule from digestion with trypsin and disulfide reduction with dithiothreitol. As to the far and near ultraviolet CD spectra data, however, there was no significant difference between in the presence and absence of an anion. Therefore, the binding of an anion would induce some conformational changes which were not reflected by the CD spectrum.     

Calorimetric studies of melanotransferrin (p97) and its interaction with iron

The mammalian molecule melanotransferrin (mTf), also called p97, is a member of the transferrin family of molecules. It exists in both secreted and glycosylphosphatidylinositol-anchored forms and is thought to play a role in angiogenesis and in transporting iron across the blood brain barrier. The binding affinity of iron to this molecule has not been formally established. Here, the binding of ferric ion (chelated with a 2-fold molar ratio of nitrilotriacetate) to mTf has been studied using isothermal titration calorimetry and differential scanning calorimetry. One iron-binding site was determined for mTf with similar binding characteristics to other transferrins. In the absence of bicarbonate, binding occurs quickly with an apparent association constant of 2.6 x 10(7) M(-1) at 25 degrees C. The presence of bicarbonate introduces kinetic effects that prevent direct determination of the apparent binding constant by isothermal titration calorimetry. Differential scanning calorimetry thermograms of mTf unfolding in the presence and absence of iron were therefore used to determine the apparent binding constant in the bicarbonate-containing system; at pH 7.5 and 25 degrees C, iron binding occurs in a 1:1 ratio with a K(app) of 4.4 x 10(17) M(-1). This affinity is intermediate between the high and low affinity lobes of transferrin and suggests that mTf is likely to play a significant role in iron transport where the high affinity lobe of transferrin is occupied or where transferrin is in proportionally low concentrations.     

The iron-binding properties of hen ovotransferrin.

1. The distribution of iron between the two iron-binding sites in partially saturated ovotransferrin was studied by labelling with 55Fe and 59Fe and by gel electrophoresis in a urea-containing buffer. 2. When iron is added in the form of chelate complexes at alkaline pH, binding occurs preferentially at the N-terminal binding site. In acid, binding occurs preferentially at the C-terminal site. 3. When simple iron donors (ferric and ferrous salts) are used the metal is distributed at random between the binding sites, as judged by the gel-electrophoresis method. The double-isotope method shows a preference of ferrous salts for the N-terminal site. 4. Quantitative treatment of the results of double-isotope labelling suggests that in the binding of iron to ovotransferrin at alkaline pH co-operative interactions between the sites occur. These interactions are apparently absent in the displacement of copper and in the binding of iron at acid pH.     

Physiological levels of binding and iron donation by complementary half-molecules of ovotransferrin to transferrin receptors of chick reticulocytes

Two fragments, each corresponding to approximately half of the ovotransferrin (OTf) molecule and containing an iron-binding site were produced by digestion with affinity bound trypsin and were purified by isoelectric focusing and gel filtration chromatography. The immunologically distinct "half-molecules" individually have little ability to bind to transferrin receptors on chick embryo red blood cells or to donate iron to them. Combining them, however, leads to both binding and iron donation approaching that found for holo-OTf. Furthermore, similar amounts of radiolabeled iron can be extracted into the putative heme fraction from Fe2OTf and from the various combined half-molecules. These findings conflict with those reported by Keung and Azari ( (1982) J. Biol. Chem. 257, 1184-1188) for subtilisin-derived half-molecules of OTf examined in a similar system. They found that each half-molecule appeared to bind at a level of approximately one-third that of Fe2OTf and that the half-molecules competed with each other for binding sites. In contrast, our equilibrium binding studies, in the presence of 2,4-dinitrophenol to prevent iron removal, led to the determination of 4.79 X 10(4) binding sites/cell for Fe2OTf, 4.44 X 10(4) for the NH2-terminal half-molecules in the presence of excess COOH-terminal half-molecules and 4.17 X 10(4) for COOH-terminal half-molecules in the presence of NH2-terminal half-molecules; apparent binding constants were estimated to be 3.29 X 10(6), 1.19 X 10(6), and 0.67 X 10(6) M-1 for these same samples. Problems associated with equilibrium binding studies in which a narrow range of concentrations of ligand is used and/or iron is being removed are discussed. Labeled combined half-molecules were half as effective as labeled Fe2OTf in competition with unlabeled Fe2OTf. These findings are consistent with the lower apparent binding constant found in the equilibrium binding studies. Equimolar apo-OTf had no effect on binding of either Fe2OTf or the combined half-molecules. It seems apparent from our studies that the NH2- and COOH-terminal half-molecules each contain a recognition region both of which are necessary for binding to the transferrin receptor and iron donation to the chick embryo red blood cell.     

Site selectivity in the binding of inorganic anions to serum transferrin

Equilibrium constants for the sequential binding of two anions at the specific metal-binding sites of apotransferrin have been measured by difference ultraviolet spectroscopy in 0.1 M N-2-hydroxyethylpiperazine-N'-2-ethanesulfonic acid (Hepes) at pH 7.4 and 25 degrees C. Log K1 values for phosphate, phosphite, sulfate, and arsenate fall in the narrow range of 3.5-4.0, while the log K1 for bicarbonate is 2.73. No binding is observed for nitrate, perchlorate, or borate. A dinegative charge appears to be the most important criterion for anion binding. Equilibrium constants have also been measured for binding of anions to both forms of mono(ferric)transferrin. There appears to be a very small site selectivity (0.2 to 0.4 log units) for phosphate, arsenate, and phosphite that favors binding to the N-terminal site, but there is no detectable selectivity for binding of sulfate or bicarbonate. Comparison of the binding affinities and anion selectivity with literature data on anion-binding to protonated macrocyles and cryptates strongly supports the existence of specific anion-binding sites on the protein. Binding constants were also measured in 0.01 M Hepes. The anionic sulfonate group of the buffer appears to have a small effect on anion binding.     

Perturbed angular correlation studies of the metal-binding sites in ovotransferrin and its C- and N-terminal halves

The perturbed angular correlation (PAC) technique has been applied to study the electric quadrupole interaction of ¹⁸¹Hf nuclei at the binding sites of ovotransferrin (OTF) molecules. Two specific electric field gradients were observed. Their relative intensities depend on the pH value and the temperature of the samples, whereas the electric quadrupole interaction parameters themselves remain unaffected. In order to compare the binding sites in OTF, experiments with N- and C-terminal half-molecules were performed. Both specific configurations are observed at the N-terminal and at the C-terminal binding site with similar quadrupole parameters as for the intact protein. Remarkably, the stability of the hafnium binding to the C-terminal fragment appears to be reduced as compared with the N-terminal half and the intact protein.     

Reversible association of half-molecules of ovotransferrin in solution. Basis of co-operative binding to reticulocytes.

In the present paper, gel-filtration studies of diferric-ovotransferrin (Fe2OTf), the individual half-molecules of ovotransferrin (OTf) and equimolar mixtures of half-molecules have been interpreted according to the Gilbert theory as developed by Ackers & Thompson [(1965) Proc. Natl. Acad. Sci. U.S.A. 53, 342-349]. The data indicate that the half-molecules associate reversibly in solution and allow determination of a dissociation constant, Kd' = 8.0 (+/- 2.7) microM. Equilibrium binding studies have been performed using NH4Cl to block removal of iron from equimolar differentially iodine-labelled half-molecules (125I and 131I), in order to evaluate the binding of each to chick-embryo red blood cells under identical conditions. The amount of associated half-molecules over a range of concentrations has been calculated using the constant derived from the gel-filtration experiments described above. A computerized non-linear least-squares regression analysis of the data leads to determination of Kd* (the apparent dissociation constant for the interaction between OTf or half-molecules and the transferrin (Tf) receptors of chick-embryo red blood cells) and Bmax (binding at infinite free-ligand concentration) for the half-molecules similar to those found for Fe2OTf. Recent reports confirm that the two iron-binding domains of both OTf and human lactotransferrin associate non-covalently in solution. Our work shows that the isolated half-molecules of OTf are able to reassociate in solution and that this reassociation has functional significance by allowing the complex to be recognized by the Tf receptor.     

Negative cooperativity of chicken ovotransferrin on Al(III)-binding

Chicken ovotransferrin, an iron binding protein, has two metal binding sites (amino (N) and carboxy (C) terminal sites). It binds Cu(II), Al(III), Co(II), and other metals, as well as Fe(III). In this study, the selectivity and cooperativity of the N and C sites on Al(III), Co(II), and Tb(III) binding were investigated. Metals were classified into two groups according to their site preference. Co(II) and Al(III) bound to the N site more preferably than to the C site, whereas Tb(III) bound to the C site more preferably. On Fe(III) binding, the binding constant of Fe(III) becomes larger when the other site is already occupied. Thus, positive cooperativity is seen. In the present study, the binding cooperativities of Co(II), Tb(III), and Al(III) as to the N and C sites were investigated. On Co(II) and Tb(III) binding, no cooperativity was observed, as in the case of Cu(II) [Yamamura, T. et al. (1985) in Proteins of Iron Storage and Transport (Spik, G., Montreuil, J., Crichton, R.R., & Mazurier, J., eds.) pp. 53-56, Elsevier Science Publ. B.V., Amsterdam]. In contrast, negative cooperativity was observed on Al(III) binding. Based on a model proposed by Yamamura et al. [Yamamura, T. et al. (1985) ibid.], the ratio of the binding constants, KC/KN, and the stacking coefficient, Kst, were estimated. KC/KN is 2.2 +/- 0.4 for the Tb(III) ion, 0.5 +/- 0.1 for the Co(II) ion, and 0.12 +/- 0.02 for the Al(III) ion. Kst (= 1 in a non-cooperative case) is 0.98 +/- 0.02 for the Tb(III) ion, 1.03 +/- 0.02 for the Co(II) ion, and 0.55 +/- 0.22 for the Al(III) ion.     

Estimation of the ferrous-transferrin binding constants based on thermodynamic studies of nickel(II)-transferrin

The equilibrium constants for the binding of Ni2+ to human serum transferrin in 0.01 M hepes containing 5 mM sodium bicarbonate at 25 degrees C and pH 7.4 have been measured. The effective binding constants are log K1 = 4.10 +/- 0.15 and log K2 = 3.23 +/- 0.31 for the reactions Ni2+ + apoTr (K1) in equilibrium Ni2+-Tr. Ni2+ + Ni2+-Tr (K2) in equilibrium Ni2+-Tr-Ni2+ where the explicit terms for bicarbonate and hydrogen ion have been incorporated into the effective binding constants. Titration of both forms of mono(ferric)transferrin indicates that unlike other metal ions, Ni2+ binds preferentially to the N-terminal binding site, but that the site preference is rather small. A linear-free-energy relationship (LFER) for the complexation of Ni2+ and Fe2+ has been prepared. This LFER has been used to estimate effective binding constants of log K1 = 3.2 and log K2 = 2.5 for the ferrous-transferrin complex. These ferrous constants have been combined with the literature binding constants for ferric-transferrin to estimate formal reduction potentials of -340 mV vs. NHE for the C-terminal site and -280 mV for the N-terminal site.     

1H NMR study of effects of synergistic anion and metal ion binding on pH titration of the histidinyl side-chain residues of the half-molecules of ovotransferrin

Separation of ovotransferrin into C-terminal (OTf/2C) and N-terminal (OTf/2N) half-molecules has made possible the resolution of all expected histidinyl C(2)H resonances by proton nuclear magnetic resonance at 250 MHz. The chemical shift of many of the resonances decreases with increasing pH, allowing construction of titration curves, whereas a few resonances fail to titrate. On formation of the GaIIIOTf/2(C2O4) ternary complexes, two of the low-field C(2)H resonances in each half-molecule fail to titrate. This behavior implicates the imidazole groups giving rise to these resonances as ligands to the bound metal ion. A third C(2)H resonance in each half-molecule undergoes a marked reduction in pK'a on formation of the ternary complex. The imidazole group displaying this resonance is implicated in a proton-relay scheme involved in binding the synergistic anion, oxalate, and a water of hydration on the bound metal ion. The titration curves for the various imidazole resonances have been fit to a four-parameter equation involving estimation of the pK'a, the limiting chemical shift values, and a Hill constant n. Hill constants of less than 1 can be rationalized by correcting the titration curve for the charge Z on the protein as a function of pH and the work function w. The titration curve for the imidazole group in OTf/2C involved in the proton-relay scheme shows a value for n greater than 1, which suggests positive cooperativity in the titration of this residue. The basis for this behavior cannot be rationalized at this time.(ABSTRACT TRUNCATED AT 250 WORDS)     

Amino-terminal and carboxyl-terminal half-molecules of ovotransferrin: preparation by a novel procedure and their interactions

The amino- (N-) and carboxyl- (C-)terminal half-molecules of ovotransferrin were prepared by a novel procedure. The trypsin-nicked ovotransferrin (Ikeda et al. (1985) FEBS Lett. 182, 305-309), in which the two half-molecules interact non-covalently forming a stable dimer, was purified by gel filtration and anion-exchange column chromatography. By subsequent cation-exchange chromatography, the nicked form was distinctly separated into an equivalent amount of the N-terminal and C-terminal half-molecules. Analyses of the N-terminal and C-terminal sequences indicated that the N-terminal and C-terminal half-molecules comprised the alignments of residues 1-332 and 342-686 of ovotransferrin, respectively. Anion-exchange chromatography, gel filtration chromatography, and non-denaturing polyacrylamide gel electrophoresis revealed that the isolated half-molecules had the ability to re-associate in solution. The contents of alpha-helix and beta-sheet of the two half-molecules, as determined by circular dichroism (CD) spectra, were very similar to those of intact ovotransferrin. No prominent alteration in the secondary structure of the two half-molecules was induced by the re-association.     

The interaction of auramine O with calmodulin: location of the binding site on the connecting strand.

The cationic dye auramine O forms a fluorescent complex with Ca(2+)-liganded calmodulin. One moderately strong binding site is present, as well as one or more weaker sites. The binding site for auramine O is different from those for toluidinyl-naphthalene sulfonate. The dependence of binding upon electrolyte concentration suggests a substantial electrostatic component of the free energy of binding. The splitting of the bond between residues 77 and 78 by trypsin digestion abolishes auramine O binding; the N- and C-terminal half-molecules have virtually no binding capacity. This suggests that the primary binding site is located near the midpoint of the connecting strand and includes elements of both half-molecules. Thrombin digestion, which splits calmodulin between residues 106 and 107, also substantially reduces auramine O binding; this may be interpreted in terms of the stabilization of the structure of the connecting strand by interaction with residues within binding domain IV. The binding affinity at pH 5.0, where the helical organization of the connecting strand may be intact, is greater than at neutral pH.     

Alternative structural state of transferrin. The crystallographic analysis of iron-loaded but domain-opened ovotransferrin N-lobe

Transferrins bind Fe3+ very tightly in a closed interdomain cleft by the coordination of four protein ligands (Asp60, Tyr92, Tyr191, and His250 in ovotransferrin N-lobe) and of a synergistic anion, physiologically bidentate CO32-. Upon Fe3+ uptake, transferrins undergo a large scale conformational transition: the apo structure with an opening of the interdomain cleft is transformed into the closed holo structure, implying initial Fe3+ binding in the open form. To solve the Fe3+-loaded, domain-opened structure, an ovotransferrin N-lobe crystal that had been grown as the apo form was soaked with Fe3+-nitrilotriacetate, and its structure was solved at 2.1 A resolution. The Fe3+-soaked form showed almost exactly the same overall open structure as the iron-free apo form. The electron density map unequivocally proved the presence of an iron atom with the coordination by the two protein ligands of Tyr92-OH and Tyr191-OH. Other Fe3+ coordination sites are occupied by a nitrilotriacetate anion, which is stabilized through the hydrogen bonds with the peptide NH groups of Ser122, Ala123, and Gly124 and a side chain group of Thr117. There is, however, no clear interaction between the nitrilotriacetate anion and the synergistic anion binding site, Arg121.     

The bicarbonate-dependence of zinc(II)-transferrin binding

The binding of zinc(II) to human serum transferrin has been studied as a function of the solution concentration of sodium bicarbonate in 100 mM, pH 7.4 hepes buffer at 25 degrees C. The apparent molar absorptivity of the zinc-transferrin complex has been determined from the initial slopes of titration curves of delta epsilon versus the ratio of [Zn]/[Tf]. This absorptivity represents the difference between the positive absorbance of the ternary Zn-HCO3-Tf species in the sample cuvette and the negative absorbance of binary HCO3-Tf species in the reference cuvette. Higher concentrations of bicarbonate increase the degree of saturation of apo-Tf with bicarbonate and thus increase the apparent absorptivity of the zinc-Tf complex. Titrations of apo- and monoferric transferrins with bicarbonate indicate that there is little, if any, difference in the bicarbonate binding constants of the two specific transferrin binding sites. An equilibrium constant of log K = 2.49 has been used to calculate the degree of saturation of the C-terminal binding site with bicarbonate. The zinc-binding affinity of this site depends linearly on this degree of saturation. The scatter in the zinc-binding constants of the weaker N-terminal site precludes a similar analysis of the bicarbonate-dependence of binding at this site. The results strongly support the previous proposal that binding of the synergistic bicarbonate anion is responsible for the uv absorption observed upon addition of bicarbonate to apoTf.     

Competitive binding of iron by transferrins from different vertebrates.

1. A competitive dialysis technique has been used to study the relative affinities of the two iron-binding sites on transferrin molecules and the relative binding strengths of transferrins isolated from plasma of different species. 2. The comparisons were extended to include desialylated human transferrin, ovotransferrin, and a cyanogen bromide fragment of the latter. 3. Although the results of bilateral experiments could generally be accounted for in terms of the theory of independent sites, there were some exceptions, and cyclic comparisons were inconsistent. 4. All the comparisons made were compatible with a model in which site-interaction occurred, but it was not possible to decide whether the sites were intrinsically identical or not. For most species this corresponded to positive cooperativity, but for rabbit it was negative. 5. The average affinity of transferrin for iron depended on species, but the variation was never more than about one order of magnitude. 6. No effect on the binding constants for human transferrin could be detected when the sialic acid residues were removed. 7. The fragment of ovotransferrin competed fairly effectively with the native molecule for iron, although the average relative affinity was only about 1:15. 8. The relative binding of iron by ovotranferrin and human transferrin was affected little when bicarbonate anion was replaced by oxalate, although the ratio of the two binding constants for ovotranferrin increased.     

Studies on the binding of copper to dopamine beta-monooxygenase and other proteins using the Cu2+ ion-selective electrode

The binding of Cu2+ to native and copper-free dopamine beta-monooxygenase has been investigated by potentiometric titrations using a Cu2+-selective electrode. Stoichiometric formation constants have been determined from regression analysis of the resulting titration curves. The results establish a stoichiometry of four high-affinity binding sites for Cu/+ (log Kf approximately 11) per enzyme tetramer, and more binding sites of lower affinity (log Kf approximately 5-7). The data for binding of the first four Cu2+ to the enzyme tetramer indicate interactions in the binding to the sites. Bovine serum albumin, metal-free carbonic anhydrase, and ovotransferrin have also been titrated with Cu2+, and the formation constants of both high-affinity binding sites and other sites have been determined. The stoichiometry of one high-affinity binding site of Cu2+ for carbonic anhydrase (log Kf approximately 10-12) and two sites for ovotransferrin (log Kf approximately 11) agree with the reported metal binding properties of these proteins. The number of high-affinity binding sites for bovine serum albumin was pH dependent.     

Calcium binding to thermitase. Crystallographic studies of thermitase at 0, 5, and 100 mM calcium

The three-dimensional crystal structure of thermitase complexed with eglin-c in the presence of 100 mM calcium has been determined and refined at 2.0-A resolution to a R-factor of 16.8%. This crystal structure is compared with previously determined structures of thermitase at 0 and 5 mM calcium concentration. In the presence of 100 mM calcium all three calcium binding sites in thermitase are fully occupied. At 100 mM CaCl2 the "weak" calcium binding is occupied by a calcium ion, which is chelated by three protein ligands and four water molecules in a pentagonal bipyramid geometry. Thermitase has, apparently, a monovalent and divalent cation binding position at 2.5-A distance from each other at this site. At low calcium concentrations the monovalent-ion position is occupied by a sodium or potassium ion. The "medium strength" binding site shows in the presence of 100 mM CaCl2 a square antiprism arrangement with eight ligands, of which seven are donated by the protein. At low calcium concentrations we observe a distorted pentagonal bipyramid coordination at this site. The largest difference between these two conformations is observed for ligand Asp-60, which has two conformations with 0.8-A difference in C alpha positions. The "strong" calcium binding site has a pentagonal bipyramid coordination and is fully occupied in all three structures. Structural changes on binding calcium to the weak and "medium strength" calcium binding sites of thermitase are limited to the direct surroundings of these sites. Thermitase resembles in this respect subtilisin BPN' and does not exhibit long-range shifts as have been reported for proteinase K.     

The metal-binding properties of ovotransferrin. An investigation of cobalt(II) derivatives

Cobalt(II) ovotransferrin bicarbonate and oxalate ternary complexes were prepared and investigated in the pH range 7-10.5. Cobalt(II) provides an excellent and unique spectroscopic probe to monitor subtle structural differences in solution between the two sites of ovotransferrin and to investigate the structural dependence on pH. CD spectroscopy on one side and 1H NMR spectroscopy of isotropically shifted signals on the other are extremely sensitive techniques and are particularly suited for high spin cobalt(II)-containing compounds. In the case of the oxalate derivative the metal-binding ability of the protein is different at the two binding sites and is pH dependent; the CD spectra reveal two different sites, one of which is clearly pH dependent with a pKa of 9.5. On the contrary the bicarbonate analogue does not show any spectral difference between the two sites; both of them change with pH, the pKa being again 9.5. 1H NMR spectra of the oxalate derivatives at pH 7-8 reveal the presence of conformers, the distribution of which depends on the H2O/D2O ratio. Such conformers are not revealed in the bicarbonate system; at pH around 10 the NMR spectra of both systems show inequivalence between the two sites and/or the presence of different conformers for each site. Such differences are discussed in terms of the possible implications in mechanism and function. The overall spectral data are consistent with the donor groups being two histidines, two tyrosines, the synergistic anion, and possibly a solvent molecule.     

Formation of monoferric ovotransferrins in the presence of chelates.

1. When ovotransferrin is partially saturated with iron, endotherms for apo-ovotransferrin, two monoferric ovotransferrins and Fe2-ovotransferrin are observed by differential scanning calorimetry. The relative sizes of the endotherms are changed in the presence of the iron-chelating agents nitrilotriacetic acid and ATP. 2. When iron is added as Fe(III)-nitrilotriacetate, at Fe-nitrilotriacetate: ovotransferrin ratios less than unity, the endotherm for Fe2-ovotransferrin is essentially absent. At Fe-nitrilotriacetate: ovotransferrin ratios of unity the only species present in solution in appreciable concentration as evidenced by their differential-scanning-calorimetry endotherms, are two monoferric ovotransferrins in approximately equal amounts. At Fe-nitrilotriacetate: ovotransferrin ratios greater than unity, the apo-ovotransferrin endotherm is absent, and the endotherms for the two monoferric ovotransferrins decrease in size as the endotherm for Fe2-ovotransferrin increases. 3. In the presence of nitrilotriacetate, binding of iron to the two sites of ovotransferrin is highly anti-co-operative, but essentially indiscriminate. When monoferric ovotransferrin is formed from apo-ovotransferrin, binding at one site is slightly favoured compared with binding at the other site, but once iron has been bound at either site, the binding affinity for iron at the unoccupied site is much decreased.     

Independent binding sites in mouse 5.8S ribosomal ribonucleic acid for 28S ribosomal ribonucleic acid

Limited digestion of mouse 5.8S ribosomal RNA (rRNA) with RNase T2 generates 5'- and 3'-terminal "half-molecules". These fragments are capable of independently and specifically binding to 28S rRNA, so there exist at least two contacts in the 5.8S rRNA for the 28S rRNA. The dissociation constants for the 5.8S/28S, 5' 5.8S fragment/28S, and 3' 5.8S fragment/28S complexes are 9 x 10(-8) M, 6 x 10(-8) M, and 13 x 10(-8) M, respectively. Thus, each of the fragment binding sites contributes about equally to the overall binding energy of the 5.8S/28S rRNA complex, and the binding sites act independently, rather than cooperatively. The dissociation constants suggest that the 5.8S rRNA termini from short, irregular helices with 28S rRNA. Thermal denaturation data on complexes containing 28S rRNA and each of the half-molecules of 5.8S rRNA indicate that the 5'-terminal binding site(s) exist(s) in a single conformation while the 3'-terminal site exhibits two conformational alternatives. The functional significance of the different conformational states is presently indeterminate, but the possibility they may represent alternative forms of a conformational switch operative during ribosome function is discussed.