Preferential ligand binding to multi-state acceptor systems: the unexplored paradox of acceptor self-association that is ligand-mediated but detrimental to ligand binding

Consideration is given to the interactions of ligand with self-associating acceptor systems for which preferential ligand binding is an ambiguous term, in that the acceptor species with greater affinity for ligand possesses relatively fewer binding sites. A paradoxical situation wherein ligand-mediated self-association is seemingly detrimental to ligand binding is shown to be the predicted outcome for a transient range of ligand concentrations. This outcome reflects the existence of a critical point in the dependence of the extent of acceptor self-association upon ligand concentration that coincides with a cross-over point of ligand-binding curves for different, fixed total concentrations of acceptor. By classical differentiation methods the conditions for the existence of these critical points are established not only for two-state acceptor systems but also for three-state acceptor systems in which the ligand-binding form of monomer also undergoes reversible isomerization to an inactive state. Similar procedures are used to comment upon the forms of binding curves for the three-state acceptor systems, the Scatchard representations of which may exhibit as many as three critical points (two maxima and a minimum). This delineation of quantitative expressions for critical points and other distinctive features associated with the conflicting interplay of ligand-binding and self-association behaviour should provide a more definitive means of characterizing systems with one acceptor state the preferred binding form on affinity grounds but with the other the preferred state from the viewpoint of binding-site numbers.     

Tb3+ binding to bovine prothrombin and bovine prothrombin fragment 1

The binding of Tb3+ to bovine prothrombin and the amino-terminal 156 residues of prothrombin (F-1) was studied. On the basis of various Tb3+ emission properties, three classes of Tb3+-binding sites were described. The first class contained three high affinity sites in the F-1 region. These sites were filled noncooperatively and were saturated with Tb3+ before the other classes of sites started to fill. Ho3+ quenching of Tb3+ emission showed that these sites were in close proximity to one another (estimated distances 6-12 A). The second class of sites contained three lower affinity sites, also in the F-1 region. These sites bound Tb3+ in a stoichiometric manner and saturated prior to metal binding to the final class of sites. The number of protein ligands binding Tb3+ in the high affinity sites decreased as this second set of sites was filled. Ho3+ quenching of Tb3+ emission suggested that these sites were closely spaced and/or close to the first set of sites. The third class of sites contained 4-6 low affinity sites unique to prothrombin (not in the F-1 region). These sites were not studied extensively, but Tb3+ did not appear to bind stoichiometrically and did not saturate these sites in a manner similar to the other two classes of sites. The emission properties of Tb3+ bound to F-1 were different in KCl versus NaCl containing buffer while the emission properties of Tb3+ bound to prothrombin were not. Optimum conditions for studying lanthanide binding to F-1 (i.e. when Tb3+ bound to F-1 showed emission properties similar to Tb3+ bound to prothrombin) were when F-1 experiments were done at low F-1 concentrations in buffer containing 0.1 M KCl.     

The binding of an indefinitely associating ligand to acceptor: consideration of monovalent ligand species binding to a multivalent acceptor

Currently available binding theory is extended to incorporate the concept of indefinite self-association of the ligand. Binding equations are formulated in closed form for the case of the binding to a multivalent acceptor of a ligand capable of isodesmically indefinitely self-associating in a "head-to-tail" mode such that each ligand state bears one site capable of interacting with the acceptor. It is shown both mathematically and by way of numerical example that this system will give rise exclusively to binding curves convex to the r-axis in Scatchard format. Thus, the system provides another example of a binding mechanism capable of generating an apparent negatively co-operative binding response.     

Deglycosylated prothrombin fragment 1. Calcium binding, phospholipid interaction, and self-assocation

The carbohydrate portion of prothrombin fragment 1 has been removed by fluorolysis in anhydrous HF. The deglycosylated protein retains its calcium- and membrane-binding properties. The slow, calcium-dependent protein transition monitored by changes in intrinsic protein fluorescence remains intact for the aglycoprotein. Calcium-dependent protein-membrane binding is also observed and can be quantitatively reversed with EDTA. The major alteration resulting from carbohydrate removal is the degree of protein self-association. Both the normal and deglycosylated proteins undergo a rapid self-association which approaches a dimer in the presence of calcium. This self-association is independent of the slow change in intrinsic fluorescence. The deglycosylated protein then undergoes a secondary self-association with kinetics identical with the fluorescence change. This secondary self-association also occurs on the membrane surface. This suggests that the calcium-dependent conformational change exposes a site on the protein which functions in secondary self-association. The carbohydrate apparently masks this site in the native molecule.     

Decarboxylation of bovine prothrombin fragment 1 and prothrombin

Bovine prothrombin fragment 1 and prothrombin undergo decarboxylation of their gamma-carboxyglutamic acid residues when the lyophilized proteins are heated in vacuo at 110 degrees C for several hours. The fully decarboxylated fragment 1 product has lost its barium-binding ability as well as the calcium-binding function which causes fluorescence quenching in the presence of 2 mM Ca2+. There is no sign of secondary structure alteration in solution upon analysis by fluorescence emission and circular dichroic spectroscopy. A family of partially decarboxylated fragment 1 species generated by heating for shorter periods shows that the initial decrease in calcium-binding ability occurs almost twice as rapidly as the loss of gamma-carboxyglutamic acid. This is consistent with the idea that differential functions can be ascribed to the 10 gamma-carboxyglutamic acid residues in fragment 1, including both high- and low-affinity metal ion binding sites. Prothrombin itself also undergoes total decarboxylation without any apparent alteration in secondary structure. However, in this case the latent thrombin activity is progressively diminished during the heating process in terms of both clotting activity and hydrolysis of the amide substrate H-D-Phe-Pip-Arg-pNA. The present results indicate that in vitro decarboxylation of gamma-carboxyglutamic acid in dried proteins is useful for analyzing the detailed calcium-binding proteins of vitamin K dependent coagulation factors.     

pH dependence of magnesium ion binding to prothrombin fragment 1 and gamma-carboxyglutamic acid-containing peptides via 25Mg NMR.

The binding of magnesium ions to two tripeptides, $\text{L}$-Arg-$\text{D}$-Gla-$\text{D}$-Gla-OMe and Z-$\text{L}$-Arg(NO₂)-$\text{D}$-Gla-$\text{D}$-Gla-OMe, and to bovine prothrombin fragment 1 as a function of pH has been monitored by ²⁵Mg NMR spectroscopy. Binding to the tripeptide was dependent on peptide ionizations occurring at pH 4.6 – 4.8. The pH dependence of magnesium ion binding to fragment 1 reveals two inflection points 4.2 may be attributed to the deprotonation of the third side chain carboxylic acid group of the double γ-carboxyglutamic acid sequence. The origin of the increased binding of magnesium ions to fragment 1 at pH values above 7 is unknown.     

Preferential binding of a G-quadruplex ligand to human chromosome ends

The G-overhangs of telomeres are thought to adopt particular conformations, such as T-loops or G-quadruplexes. It has been suggested that G-quadruplex structures could be stabilized by specific ligands in a new approach to cancer treatment consisting in inhibition of telomerase, an enzyme involved in telomere maintenance and cell immortality. Although the formation of G-quadruplexes was demonstrated in vitro many years ago, it has not been definitively demonstrated in living human cells. We therefore investigated the chromosomal binding of a tritiated G-quadruplex ligand, ³H-360A (2,6-N,N′-methyl-quinolinio-3-yl)-pyridine dicarboxamide [methyl-³H]. We verified the in vitro selectivity of ³H-360A for G-quadruplex structures by equilibrium dialysis. We then showed by binding experiments with human genomic DNA that ³H-360A has a very potent selectivity toward G-quadruplex structures of the telomeric 3′-overhang. Finally, we performed autoradiography of metaphase spreads from cells cultured with ³H-360A. We found that ³H-360A was preferentially bound to chromosome terminal regions of both human normal (peripheral blood lymphocytes) and tumor cells (T98G and CEM1301). In conclusion, our results provide evidence that a specific G-quadruplex ligand interacts with the terminal ends of human chromosomes. They support the hypothesis that G-quadruplex ligands induce and/or stabilize G-quadruplex structures at telomeres of human cells.     

The binding of a ligand to an acceptor undergoing indefinite self-association

Explicit expressions are derived which describe the binding of a univalent ligand to equivalent and independent sites on each state of an acceptor undergoing indefinite self-association that is governed by an isodesmic equilibrium constant KI. From considerations of systems in which the same site-binding constant kA applies to all acceptor-ligand interactions, the general forms of binding curves and Scatchard plots are deduced for situations in which binding sites are either created or lost at each monomer-monomer interface. Greater generality is then introduced into the model by allowing ligand interactions with polymeric acceptor states to be governed by a site-binding constant kp that differs in magnitude from that for monomeric acceptor kA. Finally, experimental results with the glutamate dehydrogenase-GTP and lysozyme-saccharide systems are used to illustrate ways in which the present quantitative expressions may be applied to the characterization of inteactions between a ligand and an indefinitely self-associating acceptor.     

Metal binding sites of a gamma-carboxyglutamic acid-rich fragment of bovine prothrombin.

The metal binding sites of a gamma-carboxyglutamic acid-rich fragment derived from bovine prothrombin were examined using paramagnetic lanthanide ions to evaluate the role of gamma-carboxyglutamic acid resideus in metal binding. A gamma-carboxyglutamic acid-rich peptide, fragment 12-44, was isolated from a tryptic digest of prothrombin. Using 153Gd(III), fragment 12-44 was found to contain one high affinity metal binding site (KD = 0.55 microM) and four to six lower affinity metal binding sites (KD approximately 4 to 8 microM). The S-carboxymethyl derivative of fragment 12-44, in which the disulfide bond in fragment 12-44 was reduced and alkylated, contained no high affinity metal binding site and four or five lower affinity sites (KD = 8 microM). The effects of paramagnetic lanthanide ions on fragment 12-44 and its S-carboxymethyl derivative were studied by natural abundance 13C NMR spectroscopy. The 13C NMR spectrum of fragment 12-44 was recorded at 67.88 MHz and the resonances were assigned by comparison to the chemical shift of carbon resonances of amino acids and peptides previously studied. The proximity between bound metal ions and carbon atoms in fragment 12-44 was estimated using Gd(III), based upon the strategy that the magnitude of the change in the transverse relaxation rate of resonances of carbon nuclei induced by bound metal ions is related in part to the interatomic distances between bound metal and carbon nuclei. Titration of fragment 12-44 with Gd(III) resulted in the selective broadening of the gamma-carboxyl carbon, C gamma, C beta, and C alpha resonances of gamma-carboxyglutamic acid, and the C epsilon of the arginines. S-Carboxymethyl fragment 12-44, which lacked the high affinity metal binding site, showed markedly decreased perturbation of the C epsilon of the arginine residues upon titration with Gd(III). These studies indicate that gamma-carboxyglutamic acid residues in prothrombin fragment 12-44 participate in metal liganding. A high affinity metal binding site in fragment 12-44 is in close proximity of Arg 16 and Arg 25 and is stabilized by the disulfide bond. On the basis of these data, a model of the metal binding sites is proposed in which the high affinity site is composed of two gamma-carboxyglutamic acid residues which participate in intramolecular metal-dependent bridging of two regions of the polypeptide chain. The lower affinity metal binding sites, formed by single or paired adjacent gamma-carboxyglutamic acid residues, then may participate in intermolecular metal-dependent protein . protein or protein . membrane complex formation.     

Binding equations for interacting systems comprising multivalent acceptor and bivalent ligand: application to antigen-antibody systems

Consideration is given to the reversible interaction of a bivalent ligand, B, with a multivalent acceptor, A (possessing f reactive sites) which leads to the formation of a series of complexes, A_iB_j, comprising networks of alternating acceptor and ligand molecules. A binding equation is derived on the basis of a site association constant, k, defined in terms of reacted site probability functions. This equation, which relates the binding function, r (the moles of ligand bound per mole of acceptor) to the concentration of unbound ligand, m_b, is used to show that plots of r vs. 2km_B constructed with fixed but different values of $\text{k}\text{m}{}_{\mathrm{A}}$ intersect at the point ($\text{m}{}_{\mathrm{B}}\text{=}\text{1}\text{2k}\text{, r =}\text{f}\text{2}$) where the extent of reaction and the concentrations of those complexes for which $\text{j}\text{i}\text{=}\text{f}\text{2}$ attain maximal values. Corresponding Scatchard plots are shown by numerical example to be non-linear, their second derivative being positive for all r. It follows that such deviations from linearity cannot be taken alone as evidence for site heterogeneity in cross-linking systems. The binding equation obtained directly is shown to be identical with that obtained with f = 2 by summation procedures involving the general expression for concentrations of complexes, m_AiBj, formulated in terms of appropriate statistical factors. In this way, previous findings on precipitation and gel formation in cross-linking systems are correlated with the present development of binding theory.     

Thyroid hormone receptor dimerization function maps to a conserved subregion of the ligand binding domain.

Thyroid hormone receptors (TRs) bind as dimers to specific DNA response elements. We have used a genetic approach to identify amino acid sequences required for dimerization of the TR beta isoform. Bacteria expressing a chimeric repressor composed of the DNA binding domain of the bacteriophage lambda cl repressor fused to the TR beta ligand binding domain are immune to lambda infection as a consequence of homodimerization activity provided by the receptor sequences. The phenotypes of deletions and point mutations of the TR beta sequences map dimerization activity to a subregion of the ligand binding domain that is highly conserved among all members of the nuclear hormone receptor superfamily. These results confirm and extend previous findings indicating that this subregion plays an important role in the dimerization of TR beta and other superfamily members.     

Homology modeling and molecular dynamics simulation of human prothrombin fragment 1.

The crystallographic structure of bovine prothrombin fragment 1 bound with calcium ions was used to construct the corresponding human prothrombin structure (hf1/Ca). The model structure was refined by molecular dynamics to estimate the average solution structure. Accommodation of long-range ionic forces was essential to reach a stable solution structure. The gamma-carboxyglutamic acid (Gla) domain and the kringle domain of hf1/Ca independently equilibrated. Likewise, the hydrogen bond network and the calcium ion coordinations were well preserved. A discussion of the phospholipid binding of the vitamin K-dependent coagulation proteins in the context of the structure and mutational data of the Gla domain is presented.     

Surface binding kinetics of prothrombin fragment 1 on planar membranes measured by total internal reflection fluorescence microscopy.

Total internal reflection fluorescence microscopy (TIRFM) has been employed to investigate the Ca(2+)-dependent membrane-binding characteristics of fluorescein-labeled bovine prothrombin-fragment 1 (F-BF1). Light scattering measurements demonstrated that F-BF1 bound to small unilamellar phosphatidylserine/phosphatidylcholine (25/75, mol/mol) vesicles with an apparent dissociation constant (1.5 +/- 0.2 microM) similar to that of unlabeled protein (1.1 +/- 0.1 microM). Negatively charged supported planar membranes were constructed by fusing small unilamellar vesicles at quartz surfaces. TIRFM measurements under equilibrium conditions showed that F-BF1 bound to planar membranes with an apparent dissociation constant (0.9 +/- 0.2 microM) approximately equal to that on vesicles. Total internal reflection/fluorescence photobleaching recovery (TIR/FPR) curves for F-BF1 on 25 mol% PS planar surfaces were diffusion-influenced at F-BF1 solution concentrations less than or equal to 5 microM. Fluorescence recovery rates from samples of high F-BF1 concentrations were slowed by increasing the solution viscosity with glycerol, thus providing further support for a diffusion-limited effect at low F-BF1 concentrations. Analysis of the reaction-limited fluorescence recovery curves at F-BF1 solution concentrations greater than or equal to 10 microM gave average association and dissociation kinetic rates of approximately 10(5) M-1 s-1 and approximately 0.1 s-1, respectively. Kinetic association rates increased significantly with increasing PS, whereas kinetic dissociation rates increased only slightly. Fluorescence recovery curves were nonmonoexponential; possible mechanisms for this behavior are described.     

Three-dimensional structure of the kringle sequence: structure of prothrombin fragment 1

The three-dimensional structure of bovine prothrombin fragment 1 has been solved at 2.8-A resolution. The electron density clearly reveals four disulfide bridges along with more than 80% of the side chains completely in density, which correspond faithfully to the kringle sequence, its preceding 30 residues, and the dodecapeptide carboxy terminal; the polysaccharide and the first 35 residues of the amino terminal of fragment 1 are disordered or about 40% of the structure. The folding of the kringle sequence is based upon close disulfide van der Waals contacts between Cys-87-Cys-127 and Cys-115-Cys-139 (4.1 A between midpoints of the bridges), two antiparallel strands of highly conserved (113-118, 124-129) beta-structure, and the stacking of some conserved aromatic residues, all near the center of the folded structure. Moreover, the overall folding appears to be duplicated as a pair of stacked duplex loops with an antiparallel open loop. The overall shape of the kringle structure approximates an eccentric oblate ellipsoid of dimensions 11 X 28 X 30 A. The residues immediately preceding the kringle are dominated by alpha-helical structure (Phe-41-Cys-48; Leu-56-Glu-63). Residues Phe-41-Trp-42 and Tyr-45, which are conserved in factor IX, factor X, protein C, and protein Z, form another aromatic stacked cluster while the Cys-48-Cys-61 disulfide loop corresponds to the well-known alpha/beta structural unit. The dodecapeptide carboxy-terminal interkringle chain extends along the periphery of the kringle in its plane and forms a beta-structure with the kringle-closing Ser-140-Val-143 tetrapeptide.     

Calcium binding to calmodulin. Cooperativity of the calcium-binding sites

The effects of Mg2+ ion, pH, and KCl concentration on Ca2+ binding to calmodulin were studied by using a Ca2+ ion-sensitive electrode. The Ca2+ ion affinity of calmodulin increased with increasing pH or decreasing KCl concentration. Cooperativity between the Ca2+-binding sites was observed, and increased with decreasing pH or increasing KCl concentration. Free Ca2+ ion concentration was decreased by adding MgCl2 ion at low Mg2+ concentration and increased at higher concentrations in the presence of small amounts of Ca2+ ion. The decrease of free Ca2+ ion concentration by Mg2+ ion strongly suggests cooperativity between the Ca2+-binding sites, and it is difficult to explain the decrease in terms of the ordered binding models previously proposed. These results can be explained by a simple model which has four equivalent binding sites that bind Ca2+ and Mg2+ competitively, and showing cooperativity when either Ca2+ or Mg2+ is bound. Mg2+ ion binding to calmodulin was measured in the presence or absence of Ca2+ to confirm the validity of this model, and no Mg2+-specific site was observed.     

Homo- and heterodimer formation with prothrombin and prothrombin fragment 1 in the presence of calcium ions

The purpose of the current study is to present further evidence for prothrombin self-association as assessed by chemical crosslinking. When the self-association (evaluated by covalent crosslinking with dithiobis(succinimidylpropionate) of prothrombin or fragment 1 was evaluated at the same molar concentration of protein, similar rates of dimer formation were observed for either protein. When prothrombin and fragment 1 were incubated together with the crosslinking reagent and calcium ions, a heterodimer consisting of prothrombin and fragment 1 was observed in addition to prothrombin dimer and fragment 1 dimer. Similar experiments with prethrombin 1 showed neither significant self-association nor effect on prothrombin self-association. Comparison of the formation of prothrombin fragment 1 heterodimer formation with the effect of fragment 1 on prothrombin activation by factor Xa suggests that the anticoagulant activity of fragment 1 is not solely a result of the formation of a heterodimer between prothrombin and fragment 1.     

The two-domain NK1 fragment of plasminogen: folding, ligand binding, and thermal stability profile

The two-domain fragment N+K1 (rNK1) [Glu(1)-Glu(163)] of human plasminogen was expressed in E. coli as a hexahistidine-tagged fusion protein and chromatographically purified. The (1)H NMR spectrum supports proper folding of the K1 component within the refolded rNK1 construct (rNK1/K1). The functional properties of rNK1/K1 were investigated via intrinsic fluorescence titration with kringle-specific omega-aminocarboxylic acid ligands. The affinities closely match those previously measured for the isolated K1, which indicates that the N-domain does not significantly affect the interaction of ligands with the lysine binding site of K1. Far-UV CD spectra recorded for the N-domain suggest conformational plasticity and flexibility for the module. Two classes of spectra, referred to as types A and B, were identified with the type A spectrum reflecting a higher secondary structure content than that estimated for the type B spectrum. Subtracting the CD spectrum of rK1 from that of rNK1 yields a spectrum (Delta) which reflects the conformation of the N-domain within the rNK1 construct (rNK1/N). Delta resembles the type A spectrum, suggesting that rNK1/N adopts a relatively more ordered conformation, stabilized by the adjacent rNK1/K1 domain. In contrast, thermal unfolding curves determined via CD indicate that the rNK1/N slightly lowers the melting temperature (T(m)) of rNK1/K1. Independence of the two domains within rNK1 was tested by monitoring the thermal unfolding of rNK1/K1 when in the presence of the kringle-specific ligand AMCHA, which left the rNK1/N T(m) essentially unaffected, while increasing that of the rNK1/K1 by approximately 10 degrees C.     

Structure of bovine prothrombin fragment 1 refined at 2.25 A resolution.

The structure of bovine prothrombin fragment 1 has been refined at 2.25 A resolution using high resolution measurements made with the synchrotron beam at CHESS. The synchrotron data were collected photographically by oscillation methods (R-merge = 0.08). These were combined with lower order diffractometer data for refinement purposes. The structure was refined using restrained least-squares methods with the program PROLSQ to a crystallographic R-value of 0.175. The structure includes 105 water molecules with occupancies of greater than 0.6. The first 35 residues (Ala1-Leu35) of the N-terminal gamma-carboxy glutamic acid-domain (Ala1-Cys48) of fragment 1 are disordered as are two carbohydrate chains of Mr approximately 5000; the latter two combine to render 40% of the structure disordered. The folding of the kringle of fragment 1 is related to the close intramolecular contact between the inner loop disulfide groups. Half of the conserved sequence of the kringle forms an inner core surrounding these disulfide groups. The remainder of the sequence conservation is associated with the many turns of the main chain. The Pro95 residue of the kringle has a cis conformation and Tyr74 is ordered in fragment 1, although nuclear magnetic resonance studies indicate that the comparable residue of plasminogen kringle 4 has two positions. Surface accessibility calculations indicate that none of the disulfide groups of fragment 1 is accessible to solvent.