Modulation of recombinant human prostate-specific antigen: activation by Hofmeister salts and inhibition by azapeptides. Appendix: thermodynamic interpretation of the activation by concentrated salts.

Prostate specific antigen (PSA, also known as human kallikrein 3) is an important diagnostic indicator of prostatic disease. PSA exhibits low protease activity (>10(4)-fold less than chymotrypsin) under the usual in vitro assay conditions. In addition, PSA does not react readily with prototypical serine protease inactivators. We expressed human PSA (rh-PSA) in Escherichia coli and have demonstrated that rh-PSA has properties similar to those of native PSA isolated from human seminal fluid. Both PSA and rh-PSA are >10(3)-fold more active in the presence of 1.3 M Na(2)SO(4). This activation is anion-dependent, following the Hofmeister series when normality is considered: SO(4)(2)(-) approximately citrate > Ac(-) > Cl(-) > Br(-) > I(-). The nature of the cation has little effect on salt activation. The rate of inactivation of rh-PSA by DFP is 30-fold faster in the presence of 0.9 M Na(2)SO(4), and the rate of inactivation by Suc-Ala-Ala-Pro-Phe-CK is >20-fold faster under these conditions. Azapeptides containing Phe or Tyr at position P(1) also inactivate rh-PSA in the presence of high salt concentrations. These compounds represent the first described inhibitors designed to utilize the substrate binding subsites of PSA. CD spectroscopy demonstrates that the conformation of rh-PSA changes in the presence of high salt concentrations. Analytical ultracentifugation and dynamic light scattering indicate that PSA remains monomeric under high-salt conditions. Interestingly, human prostatic fluid contains as much as 150 micro mol citrate/g wet weight, which suggests that salt concentrations may regulate PSA activity in vivo.

In disperse solution, "osmotic stress" is a restricted case of preferential interactions.

In the practice of "osmotic stress," the effect of excluded cosolvents on a biochemical equilibrium is interpreted as the number of water molecules participating in the reaction. This action is attributed to lowering of solvent water activity by the cosolvent. This concept of osmotic stress in disperse solution is erroneous: (i) A cosolvent cannot be both excluded and inert, i.e., noninteracting, because exclusion requires a positive free energy change; (ii) a decrease in water activity alone by addition of solute cannot affect an equilibrium when the reacting surface is in contact with the solvent; and (iii) osmotic stress in disperse solution is a restricted case of preferential interactions; the reaction is driven by the free energy of cosolvent exclusion, and the derived number of water molecules is solely a measure of the mutual perturbations of the chemical potentials of the cosolvent and the protein.

Role of the colchicine ring A and its methoxy groups in the binding to tubulin and microtubule inhibition.

The roles of the methoxy substituents on ring A of two ring colchicine (COL) analogues were probed by the synthesis of a number of drugs and the examination of their effect on binding to tubulin, inhibition of microtubule assembly, and induction of GTPase activity. Selective elimination of ring A methoxy groups at positions 2, 3, and 4 weakened all three processes. The effects on binding and inhibition were independent of the nature of ring C (or C'). Specifically, excision of the 2- or 3-methoxy groups weakened binding by ca. 0.4 kcal mol-1, while that of the 4-methoxy group of ring A was weakened by 1.36 +/- 0.15 kcal mol-1. The effect on the inhibition of microtubule assembly, expressed as the equilibrium constant for the binding of the tubulin-drug complex to the end of a microtubule, was more complex and strongly dependent on the nature of ring C (or C'). This was attributed to the abilities of various groups on ring C' to overcome the wobbling in the tubulin-drug complex introduced by the weakening of the anchoring provided by ring A. It is concluded that ring A of COL is not germane to the mechanism of the inhibition of tubulin self-assembly. It serves only as a complex-stabilizing anchor. The control of this process resides in the interactions that key oxygen atoms of ring C of COL or C' of structural analogues establish with the protein. It is proposed that the 4-methoxy group of ring A serves as a key attachment point for immobilization of the drugs on the protein.

Linkages in tubulin-colchicine functions: the role of the ring C (C') oxygens and ring B in the controls.

Linkages between structural components of colchicine (COL) and its biphenyl analogues (allocolchicine, ALLO, and its analogues) in the binding to tubulin and its functional consequences were scrutinized. Three ring ALLO analogues with the carbomethoxyl in position 4' of ring C' replaced by a carbomethyl (KAC) and methoxy (MAC) groups were synthesized. The binding properties and consequences of binding (microtubule inhibition, abnormal polymerization, and induction of GTPase activity) were compared within the series of three ring and two ring compounds, as well as between pairs consisting of a two ring and a three ring compound with identical groups in position 4'. Binding measurements showed that the binding of KAC to the COL binding site proceeded with similar chemical characteristics as that of its two ring analogue (TKB), but with the kinetic characteristics of ALLO. The binding constant of KAC was found to be 1.9 x 10(6) M-1 and that of MAC was 4.6 x 10(5) M-1. The binding strength of the three ring analogues in descending order was KAC > ALLO > MAC, with increments similar to the biphenyl compounds, TKB > TCB > TMB. The difference in binding affinities between the pairs of three ring and two ring molecules was invariant (delta delta G degree = -1.3 +/- 0.2 kcal/mol-1), showing that in all cases ring B makes only an entropic contribution by suppressing free rotation about the biaryl bond. In the case of microtubule inhibition, all three ring compounds inhibited strongly with similar potencies, even though the spread in inhibition strength between the corresponding two ring molecules was > 3.3 kcal mol-1 of free energy. This difference was interpreted in terms of the ability of the various molecules to maintain tubulin in the proper conformation for binding in abnormal geometry to the growth end of a microtubule. This ability attains a maximal plateau value for three ring compounds, independently of the oxygen-containing group in ring C' (or C) and is maintained for the methyl ketone whether in a two or three ring compound. The induction of the GTPase activity was found to follow in general the binding affinity, with the exception that molecules that contained a methyl ketone were stronger GTPase inducers than expected from their alignment according to binding affinity. The finding that the binding of tropolone methyl ether (ring C of COL) induced a GTPase activity shows that ring C contains the ability to induce both substoichiometric microtubule inhibition and GTPase activity. Rings A and B act only as anchors in the binding, with ring A making an energetic contribution, while the effect of ring B is only entropic. It was concluded that both microtubule assembly inhibition and induction of GTPase activity were modulated by the same postbinding conformational change in tubulin. The difference between the strengths of these activities induced by ligands reflects the difference between a narrow allosteric effect between two well-defined sites in the case of GTPase activity and a broad effect aimed at the multiple sites involved in the incorporation of a tubulin protomer into the microtubule structure. Thus, there seems to be a loose thermodynamic linkage between binding and GTPase activity, while there is none between binding and microtubule inhibition, the two phenomena being linked only kinetically.

The thermodynamic mechanism of protein stabilization by trehalose.

The stabilization of ribonuclease A by alpha-alpha-trehalose was studied by preferential interaction and thermal unfolding. The protein is stabilized by trehalose at pH 2.8 and pH 5.5. Wyman linkage analysis showed increased exclusion of trehalose from the protein domain on denaturation. Preferential interaction measurements were carried out at 52 degrees C at pH 5.5 and 2.8, where the protein is native and unfolded, respectively, and at 20 degrees C where the protein is native at both pH values. At the low temperature, the interaction was preferential exclusion. At 52 degrees C the interaction was that of preferential binding, greater to the native than the unfolded protein, the increment on denaturation being identical to that deduced from the Wyman analysis. The stabilizing effect of trehalose can be fully accounted by the change in transfer free energy on unfolding. The temperature dependence of the preferential interactions of 0.5 M trehalose with ribonuclease A showed that it is the smaller preferential binding to the unfolded protein than to the native one which gives rise to the stabilization. A thermodynamic analysis of the data led to approximate values of the transfer enthalpies and transfer entropies for the trehalose-ribonuclease A system.

Mechanism of the stabilization of ribonuclease A by sorbitol: preferential hydration is greater for the denatured then for the native protein.

The effect of interactions of sorbitol with ribonuclease A (RNase A) and the resulting stabilization of structure was examined in parallel thermal unfolding and preferential binding studies with the application of multicomponent thermodynamic theory. The protein was stabilized by sorbitol both at pH 2.0 and pH 5.5 as the transition temperature, Tm, was increased. The enthalpy of the thermal denaturation had a small dependence on sorbitol concentration, which was reflected in the values of the standard free energy change of denaturation, delta delta G(o) = delta G(o) (sorbitol) - delta G(o)(water). Measurements of preferential interactions at 48 degrees C at pH 5.5, where protein is native, and pH 2.0 where it is denatured, showed that sorbitol is preferentially excluded from the denatured protein up to 40%, but becomes preferentially bound to native protein above 20% sorbitol. The chemical potential change on transferring the denatured RNase A from water to sorbitol solution is larger than that for the native protein, delta mu(2D) > delta mu(2N), which is consistent with the effect of sorbitol on the free energy change of denaturation. The conformity of these results to the thermodynamic expression of the effect of a co-solvent on denaturation, delta G(o)(W) + delta mu(D)(2)delta G(o)(S) + delta mu(2D), indicates that the stabilization of the protein by sorbitol can be fully accounted for by weak thermodynamic interactions at the protein surface that involve water reversible co-solvent exchange at thermodynamically non-neutral sites. The protein structure stabilizing action of sorbitol is driven by stronger exclusion from the unfolded protein than from the native structure.

Temperature dependence of the preferential interactions of ribonuclease A in aqueous co-solvent systems: thermodynamic analysis.

The temperature dependence of preferential solvent interactions with ribonuclease A in aqueous solutions of 30% sorbitol, 0.6 M MgCl2, and 0.6 M MgSO4 at low pH (1.5 and 2.0) and high pH (5.5) has been investigated. This protein was stabilized by all three co-solvents, more so at low pH than high pH (expect 0.6 M MgCl2 at pH 5.5). The preferential hydration of protein in all three co-solvents was high at temperatures below 30 degrees C and decreased with a further increase in temperature (for 0.6 M MgCl2 at pH 5.5, this was not significant), indicating a greater thermodynamic instability at low temperature than at high temperature. The preferential hydration of denatured protein (low pH, high temperature) was always greater than that of native protein (high pH, high temperature). In 30% sorbitol, the interaction passed to preferential binding at 45% for native ribonuclease A and at 55 degrees C for the denatured protein. Availability of the temperature dependence of the variation with sorbitol concentration of the chemical potential of the protein, (delta mu(2)/delta m3)T,p,m2, permitted calculation of the corresponding enthalpy and entropy parameters. Combination with available data on sorbitol concentration dependence of this interaction parameter gave (approximate) values of the transfer enthalpy, delta H2,tr, and transfer entropy delta S2,tr. Transfer of ribonuclease A from water into 30% sorbitol is characterized by positive values of the transfer free energy, transfer enthalpy, transfer entropy, and transfer heat capacity. On denaturation, the transfer enthalpy becomes more positive. This increment, however, is small relative to both the enthalpy of unfolding in water and to the transfer enthalpy of the native protein from water a 30% sorbitol solution.

Structural analysis of the substoichiometric and stoichiometric microtuble-inhibiting biphenyl analogues of colchicine.

The structures of the colchicine (COL) analogues, 2,3,4-trimethoxy-4'-acetyl-1,1'-biphenyl (TKB)and 2,3,4,4'-tetramethoxy-1,1'-biphenyl (TMB), were solved by X-ray diffraction. Their comparison with the structure of colchicine indicated the ability of both compounds to enter into a colchicine binding pocket. Comparison of TKB with 2,3,4-trimethoxy-4'-carbomethoxy-1,1'-biphenyl (TCB) showed that the methyl group of the carbomethoxy group in position 4' of TCB protrudes beyond the (C=O)-CH3 group in the same position in TKB. Superposition of both structures on the van der Waals surface of COL clearly demonstrates that TKB can fully fit within that domain, while the CH3 group of TCB protrudes beyond the COL contour. This is proposed to be the source of the inability of TCB to inhibit microtubule assembly substoichiometrically, while TKB is a very strong inhibitor. While the same steric hindrance to entering into the COL site on tubulin must exist in allocolchicine (ALLO), in its case, this is overcome by the rigidity of the three-ring structure which abolishes the loss on binding of the entropy of free rotation between the two rings of the biphenyl TCB.

Stoichiometric and substoichiometric inhibition of tubulin self-assembly by colchicine analogues.

The mechanism of the stoichiometric and substoichiometric inhibitions of tubulin self-assembly by several structural analogues of colchicine (COL) was investigated. The inhibition data were analyzed in terms of a simple model that takes into consideration Kg, the normal microtubule growth constant, equal to Cr-1 (Cr is the critical concentration for microtubule formation), and Kb, the binding constant of the drug to tubulin. In this manner, the value of the microtubule inhibition constant (Ki), which is the binding constant of the tubulin-drug complex to the end of a growing microtubule (which stops the microtubule growth), was determined. The results of the analysis of microtubule inhibition by the various colchicine analogues show that all the inhibitions can be expressed reasonably by this model. The strongest inhibitors found were colchicine (COL), allocolchicine (ALLO), and the biphenyl keto analogue 2,3,4-trimethoxy-4'-acetyl-1,1'-biphenyl (TKB), which had essentially identical values of Ki = (2.1 +/- 0.3) x 10(6) M(-1). MTC, the two-ring analogue of colchicine, was weaker (Ki = 5.6 x 10(5) M(-1). A most striking result was that tropolone methyl ether (TME), which is ring C of COL, and which binds very weakly to tubulin (Kb = 3.5 x 10(2) M(-1)), is a substoichiometric inhibitor. Its Ki value of 8.7 x 10(5) M(-1) makes it identical in strength to MTC, suggesting that ring A makes little or no contribution to the induction of assembly inhibition. The three biphenyls, which bind to tubulin with similar affinity, spanned the spectrum from strong substoichiometric inhibition (TKB) to stoichiometric inhibition for 2,3,4-trimethoxy-4'-carbomethoxy-1,1'-biphenyl (TCB) and an intermediate mode for the methoxy derivative 2,3,4,4'-tetramethoxy-1,1'-biphenyl (TMB). The extent of tubulin bound to drugs at 50% inhibition (r) was ca. 2% for TKB, ALLO, and COL, i.e. one liganded tubulin for every 40-50 molecules of free protein (substoichiometric). This ratio was 1:1.5 for TCB (stoichiometric) and 1:6 for TMB (intermediate). For TME, which is a single ring compound, it was 1:25. The progression of the stoichiometries varied directly with Ki and was totally unrelated to the values of Kb, which indicated the control of the stoichiometry by Ki and the close thermodynamic linkage between r and Ki. Comparison of the inhibitory capabilities of the various drugs identified the need for strong substoichiometric inhibition of a carbonyl group on ring C or C'. Furthermore, this group must be properly oriented by interaction with the protein or by the structural rigidity imparted by ring B, as in ALLO. The simple linked equilibrium model developed in this paper permits the alignment of drugs along a continuum that ranges from stoichiometric to strong substoichiometric modes of microtubule inhibition. Furthermore, it shows that the previously identified two classes are the two ends of a monotonously progressing spectrum described by a single mechanism of action.

On the role of surface tension in the stabilization of globular proteins.

The stabilization of proteins by a variety of co-solvents can be related to their property of increasing the surface tension of water. It is demonstrated that, during the thermal unfolding of proteins, this increase of the surface tension can be overcome by the increase in the temperature of the solution at the midpoint of the transition, Tm, and the weak binding of co-solvent molecules. Three such co-solvents were studied: trehalose, lysine hydrochloride (LysHCl), and arginine hydrochloride (ArgHCl). Trehalose and LysHCl increase the midpoint of Tm. The increase of the surface tension by addition of trehalose is completely compensated by its decrease due to the increase in Tm. However, for LysHCl, the increase of the surface tension by the co-solvent is partly reduced by its binding to the protein. For trehalose, preferential interaction measurements with RNaseA demonstrate that it is totally excluded from the protein. In contrast, LysHCl gives evidence of binding to RNaseA. ArgHCl also increases the surface tension of water. Nevertheless, Tm of RNaseA decreases on addition of ArgHCl to the solution. Preferential interaction measurements showed very small values of preferential hydration of the native protein, indicating extensive binding of ArgHCl to the protein. During unfolding, the amount of additional ArgHCl binding is sufficiently large to counteract the surface tension effects, and the protein is destabilized. Therefore, although surface tension appears to be a critical factor in the stabilization of proteins, its increase by co-solvent does not ensure increased stabilization. The binding of ligands can reduce significantly, or even overwhelm, its effects.

Contribution of the surface free energy perturbation to protein-solvent interactions.

Surface tension measurements were carried out at 20 degrees C by a capillary drop-weight method on aqueous solutions of sodium glutamate (NaGlu), lysine hydrochloride (LysHCl), potassium aspartate (KAsp), arginine hydrochloride (ArgHCl), lysylglutamate (LysGlu), argininylglutamate (ArgGlu), guanidinium sulfate, trehalose, trimethylamine N-oxide (TMAO), dimethyl sulfoxide, 2-methyl-2,4-pentanediol (hexylene glycol), and poly(ethylene glycol)s of molecular weights 200, 400, 600, and 1000. All of the salts and the sugar increased the surface tension of water, while the last four compounds decreased it, with 2-methyl-2,4-pentanediol lowering it most effectively and TMAO being the least effective. The preferential hydration of bovine serum albumin (BSA) and lysozyme was measured in KAsp, ArgHCl, LysGlu, and ArgGlu. The high values of preferential hydration found in all cases, except for BSA in ArgHCl, suggest that they should stabilize protein structure, as had been found for lysine hydrochloride and monosodium glutamate [Arakawa, T., & Timasheff, S. N. (1984) J. Biol. Chem. 259, 4979-4986]. A correlation was found for both BSA and lysozyme in KAsp, NaGlu, LysHCl, ArgGlu, and LysGlu between the surface tension effect and the observed preferential interactions, indicating that the change in the surface free energy of the protein-containing cavity due to the surface tension increase for water by these amino acid salts contributes dominantly to the observed increase in the chemical potential of the protein by their addition. The lack of a correlation observed for BSA, but not lysozyme, in ArgHCl at low concentrations where preferential binding is close to zero suggests, however, that the surface tension effect is not the sole factor involved in the protein-solvent interactions in these amino acid salts. Binding of ArgHCl to BSA, probably through hydrogen bonds between the Arg guanidinium group and peptide bonds, was proposed to occur, the affinity of Arg+ being reduced by electrostatic repulsion when proteins carry a net positive charge, such as is the case with lysozyme. Since the four organic solvent additives also lead to protein preferential hydration, no correlation exists between their preferential interactions and the surface free energy perturbation. Therefore, in their case, the preferential hydration must be ascribed to other factors that overcome the preferential binding expected from the Gibbs adsorption isotherm. The surface tension results, however, are consistent with the binding of the organic solvents to proteins through hydrophobic interactions, explaining, at least in part, the observed concentration dependence of the interactions.

Why do some organisms use a urea-methylamine mixture as osmolyte? Thermodynamic compensation of urea and trimethylamine N-oxide interactions with protein.

Many organisms accumulate low molecular weight substances known as osmolytes when they experience environmental water stress. The main classes of osmolytes are sugars, polyhydric alcohols, amino acids and their derivatives, and methylamines, and all are known to be protein stabilizers. However, marine cartilaginous fishes and the coelacanth use, as osmolytes, a combination of urea and methylamines, i.e., a denaturant and a stabilizer, in a 2:1 molar ratio. Preferential binding and thermal denaturation measurements in the presence of each cosolvent separately and in their mixtures have been carried out using ribonuclease T1 (RNase T1) as the protein. At a 2:1 molar ratio of urea and trimethylamine N-oxide (TMAO), the effects of the two cosolvents on the transition temperature (Tm) were found to be essentially the algebraic sum of their effects when used individually. Preferential interaction measurements of urea, TMAO and urea in its 2:1 molar ratio mixture with TMAO, have shown that the presence of TMAO has no effect on the interaction of urea with the protein in either the native or the unfolded (reduced carboxymethylated RNase T1) state. The preferential interaction of TMAO in the presence of urea could not be measured for technical reasons. Calculations of transfer free energy in the two end states of the denaturation reaction have shown that 2 M urea destabilizes RNase T1 by 3.8 +/- 0.3 kcal/mol whether 1 M TMAO is present or not. The contribution of 1 M TMAO to stabilization is calculated to be 3.1 kcal/mol in the presence of 2 M urea and is measured to be 2.7 kcal/mol in its absence.

Energy transfer studies of the distances between the colchicine, ruthenium red, and bisANS binding sites on calf brain tubulin.

Fluorescence energy transfer experiments were performed in order to measure the spatial separation between the colchine and Ruthenium Red binding sites, the high-affinity bisANS and Ruthenium Red sites, and the allocolchicine and high-affinity bisANS sites on calf brain tubulin. Energy transfer was observed between both colchicine and allocolchicine and Ruthenium Red, resulting in a distance of 40-45 A between these sites on the tubulin molecule. No detectable energy transfer could be observed when allocolchicine was used as fluorescence donor and bisANS as acceptor or when bisANS was used as donor and Ruthenium Red as acceptor. This indicates that the distance of separation between the allocolchicine and bisANS sites is greater than 50 A, while that between the bisANS and Ruthenium Red sites is greater than 72 A. On the basis of these and previous distance measurements (Ward & Timasheff, 1988), two triangles of binding sites have been defined (colchicine-bisANS-E-site and colchicine-bisANS-Ruthenium Red). Since the dihedral angle between them is not known, a schematic model has been drawn with all the sites located in a single plane.(ABSTRACT TRUNCATED AT 250 WORDS)

Cooperative multiple binding of bisANS and daunomycin to tubulin.

The binding of daunomycin and bisANS to tubulin was studied by direct equilibrium techniques. Both ligands generated abnormal Scatchard plots. Their concave-downward nature indicated positive cooperativity. The data conform to tubulin possessing ca. 35 daunomycin binding sites with a binding constant of 570-1430 M-1. The binding of bisANS is characterized by 1 strong binding site (KA = 4.5 x 10(5) M-1) and 40-50 lower affinity sites. Hill plots of both showed low degrees of cooperativity (m = 1.8 for daunomycin and 2.3 for bisANS). A detailed analysis was carried out of the cooperativity of binding of daunomycin to tubulin. Concentration differences spectra and sedimentation velocity analysis of daunomycin showed that this molecule undergoes self-association in the drug concentration range used in the binding study. The low level of polymerization (approximately tetramer), however, indicated that this could not be the source of the observed cooperativity between 35 molecules. Both the shape and concentration dependence of the daunomycin concentration difference spectra were strikingly similar to those generated on the binding of daunomycin to tubulin, which indicates the stacking of daunomycin in both cases. The observed Scatchard plot of the binding was found to be consistent with a process that involves in part ligand-ligand interactions when complexed to tubulin. Examination of the binding of bisANS in the presence of daunomycin revealed a strong increase of bisANS binding to tubulin, which suggests a loosening of tubulin structure with the exposure of new sites as these ligands bind. The mutual interaction between the two ligands in dilute solution was demonstrated by difference spectroscopy.(ABSTRACT TRUNCATED AT 250 WORDS)

The colchicine-induced GTPase activity of tubulin: state of the product. Activation by microtubule-promoting cosolvents.

Colchicine induces a weak assembly-independent GTPase activity in calf brain tubulin [David-Pfeuty, T., Erickson, H. P., & Pantaloni, D. (1977) Proc. Natl. Acad. Sci. U.S.A. 74, 5372-5376; Andreu, J. M., & Timasheff, S. N. (1981) Arch. Biochem. Biophys. 211, 151-157]. Kinetic analysis shows a turnover number of 2 x 10(-4) s-1 in 0.01 M sodium phosphate and 4 mM MgCl2, pH 7.0, with an apparent Km for GTP of 10 microM. This activity, which requires Mg2+ ions and attains a plateau at 4 mM MgCl2, is independent of pH over the pH range of 6.6-7.4. This GTPase activity was induced by all colchicine analogues that contain rings A and C (or C'), the strength varying in a manner parallel to the free energy of binding of the ligand. The specific GTPase activity was found to be independent of the tubulin-colchicine complex concentration over the range of 2-20 microM. Sedimentation velocity examination of the product of the reaction showed that GDP-tubulin-colchicine generated by hydrolysis of the E-site GTP was indistinguishable from that produced by nucleotide exchange at the site, the protein assuming the "curved" conformation in both cases. Steady-state kinetic analysis in the presence of high concentrations of microtubule-inducing cosolvent additives revealed an increase in kcat/Km of up to 1 order of magnitude that followed the order poly(ethylene glycol) 6000 (PEG-6000 > PEG-1000 = 2-methyl-2,4- pentanediol > sucrose > L-glutamate > glycerol = PEG-200 > betaine, with no apparent change in Km.(ABSTRACT TRUNCATED AT 250 WORDS)

Cosolvent modulation of the tubulin-colchicine GTPase-activating conformational change: strength of the enzymatic activity.

The locus of action of cosolvent additives in the activation of the tubulin-colchicine GTPase was investigated. The GDP off rates were slowed down by the cosolvents in a manner that parallels their specific viscosities, indicating that diffusion-controlled release of GDP may be rate-limiting under the conditions of these studies. Yet, the net effect of cosolvents was to increase the overall rate of GTP hydrolysis. Pre-steady-state kinetics of liganded tubulin in the presence of 1%, w/v, poly(ethylene glycol) 6000 (PEG-6000) exhibited a burst of inorganic phosphate release indicating that the cosolvents act at an early step in the process. A similar conclusion was drawn from measurements of the activation energy (Ea) of the reaction, which showed that 3.4 M glycerol decreased the value of Ea to 10.6 kcal mol-1 from 17.3 kcal mol-1 in its absence. The observed difference in apparent binding free energies of the colchicine analogues allo-colchicine (ALLO) and 2-methoxy-5-(2,3,4-trimethoxyphenyl)-2,4,6-cycloheptatrien-1-one (MTC, or des-ring B colchicine), when measured by fluorescence and enzyme activity titrations, identified the presence of a GTPase-activating protein conformational transition subsequent to the physicochemical binding of the ligands. The decrease of the apparent binding constant measured by enzyme activity in dilute buffer relative to that measured by fluorescence [for ALLO, Kb(fluor) = 1.46 x 10(6) M-1; Kb(enz act) = 1.1 x 10(5) M-1] yielded the value of the enzyme-activating conformational transition constant, K3 = 0.08.(ABSTRACT TRUNCATED AT 250 WORDS)

Effect of colchicine analogues on the dissociation of alpha beta tubulin into subunits: the locus of colchicine binding.

A combination of ligand binding and sedimentation equilibrium studies was used to characterize the thermodynamic linkages between alpha beta tubulin association, nucleotide binding, and the interaction of colchicine analogues with dimeric and dissociated tubulins. The strength of binding of allocolchicine to the tubulin dimer was identical (8 x 10(5) M-1) whether the exchangeable nucleotide site (E site) was occupied by GTP or GDP. This drug bound to dimeric (alpha beta) tubulin and to one of the monomeric subunits, and the binding affinity for the dissociated state was linked to occupancy of the exchangeable nucleotide site. When the exchangeable site was occupied by GTP, the drug bound with very similar affinities to the dimeric and dissociated states of the protein. For tubulin-GDP, the binding of the drug to the dissociated state was significantly weaker (6.3 x 10(4) M-1) than to the dimeric state, suggesting the existence of an E-site-related conformational change in the dissociated state. Podophyllotoxin, which contains the A-ring portion of colchicine, bound with equal affinity to the dimeric and dissociated forms of both tubulin-GTP and tubulin-GDP, indicating that it is the C-ring portion of colchicine that is linked to the E-site-related conformational change.(ABSTRACT TRUNCATED AT 250 WORDS)

Linkages between the dissociation of alpha beta tubulin into subunits and ligand binding: the ground state of tubulin is the GDP conformation.

The effects of ligands on the dissociation of the alpha beta tubulin dimer into the two subunits were investigated using calf brain tubulin. Sedimentation equilibrium studies showed a number of linkages. In the absence of magnesium in the medium, tubulin-GTP, tubulin-GDP, and tubulin with the exchangeable site unoccupied associate with essentially the same strength (K alpha beta = 1 x 10(7) M-1). This indicates that the ground state of tubulin (i.e., in the absence of magnesium) is not affected by occupancy of the exchangeable nucleotide binding site (E site). The alpha beta association is enhanced by magnesium ions. The association of tubulin with GDP in the E site is linked to the uptake of twice as many magnesium ions as that of tubulin with GTP in the E site. This suggests that magnesium binding is linked to an E-site-related conformational change. Consideration of the linkages between the binding of magnesium ions, E-site occupancy, and tubulin conformation in terms of the model [Howard, W. D., & Timasheff, S. N. (1986) Biochemistry 25, 8292-8300] in which the tubulin alpha beta dimer exists in an equilibrium between two conformations, a microtubule-forming ("straight") state favored by GTP and a double-ring-forming ("curved") state favored by GDP, leads to the conclusion that the ground state of tubulin is the ring-forming or "curved" conformation. Thus, in the absence of magnesium, the tubulin heterodimer exists in the ring-forming conformation, whether the E site is occupied by GTP or GDP.(ABSTRACT TRUNCATED AT 250 WORDS)