Interaction of human alpha-thrombin with organic ligands of ionic nature

Investigations results of human thrombin interaction with organic ligands of ion nature containing nonpolar groups are presented. It is shown that electrostatic interaction is the basic one under enzyme binding, while hydrophobic binding is only additional function in the reaction enzyme-ligand, this fact is confirmed by the absence of interaction between thrombin and rivanol which has a positive charge side by side with cumbrous hydrophobic group. New data are presented about the ligand specificity of binding sites of thrombin active centre. The importance of relative arrangement of hydrophobic ligand groups for interaction with enzyme is shown. It is supposed that thrombin binding with organic ligands occurs owing anionic site of beta-domain of active thrombin centre with the major aminoacids arginine and lysine (Lys 68, Arg 78, Arg 77, Arg 66 etc.). It is shown that the compounds containing negative group SO3 and have some cunbours hydrophobic groups interact more intensively with the enzyme. Thus, rosseline--with symmetrical hydrophobic nucleus (four benzene rings)--is the most efficient ligand for the binding with thrombin. The obtained investigation results evidence for bacteriostatical and stabilizing effect of low-molecular asobenzene ligands on rather labile thrombin molecules.     

Comparative study of the properties of serine proteases of lower and higher vertebrates

Results of the comparative study of trypsin- and chymotrypsin-like serine proteases from pyloric caeca of salmon fishes and trypsin and chymotrypsin of bulls are presented in the paper. The hydrolytic activity of salmon proteases with respect to methyl ethers of N-benzoyl-L-leucine is 2.4 times higher than that of bull chymotrypsin, but with respect to methyl esters of N-benzoyl-L-tyrosine and N-benzoyl-L-arginine the activity of salmon proteases is 6.5 and 80 times lower than that of bull chymotrypsin and trypsin. Salmon proteases in contrast to bull trypsin and chymotrypsin hydrolyze but slightly N-glutaryl-L-phenylalanine para-nitroanilide. It shown that fish proteases are not absolutely specific to synthetic substrates, which is a result of their less pronounced (than in case of bull trypsin and chymotrypsin) differences in structures of binding centres. The study of the salmon protease interaction with some immobilized ligands has confirmed the higher affinity of enzymes to reagents with two space-separated aromatic rings in their composition. It is supposed that salmon proteases interact with such reagents through two sites: hydrophobic "pockets" and probably additional binding site of the active centre. The salmon protease preparation demonstrates higher resistance to inactivating action of formaldehyde within the range of concentrations 2-16% than bull chymotrypsin does.     

Comparative characteristics of chymotrypsin covalently bound on chemically and enzymatically oxidized agaroses

Summary The enzymatically activated agaroses compared with chemically activated show 25 fold lower amount of generated aldehyde groups, 33 fold lower binding capacity for chymotrypsin, 3 fold lower proteolytic as well as amidolytic activity toward AntAlaAlaPheNA of the corresponding fixed enzyme. Trans-cinnamoylimidazole titration data demonstrate 100% active bound enzyme in the case of enzymatically oxidized agaroses and 57% for chemically oxidized. The enzymic activation offers a small number of sites for ligand attachment in a unique microenvironment. The chemical activation yields a suitable matrix for effective chymotrypsin immobilization.     

Kinetics of the inhibition by naphtholsulfonate compounds of AMP deaminase from chicken erythrocytes

The effect of a variety of naphthalene sulfonate compounds on the chicken erythrocyte AMP deaminase (AMP aminohydrolase, EC 3.5.4.6) reaction was analyzed kinetically. Of the naphthalene sulfonate derivatives tested, the compounds with hydroxyl, sulfonate and nitrogen groups such as amino, anilino or azo groups showed an inhibitory effect. The cooperative effect of AMP, analyzed in terms of Hill coefficient, was increased from about 2 to 4 and the maximal velocity was unchanged with the addition of these compounds, suggesting the ligands as an allosteric inhibitor of the enzyme. The inhibition of AMP deaminase by naphtholsulfonate compounds can be qualitatively and quantitatively accounted for by the Monod-Wyman-Changeux model. Theoretical curves yield a satisfactory fit of all experimental saturation and inhibition curves, assuming four binding sites for AMP and the inhibitor, and various KT(I) values. The structure-activity analysis of the interaction of the naphtholsulfonate compounds with AMP deaminase has demonstrated that the affinity of the enzyme for naphtholsulfonates as the inhibitors is correlated with electronic properties of the nitrogen atoms attached to naphthalene moiety: the delocalization of lone electron pair on nitrogen through naphtholsulfonate group makes the compound less basic, resulting in more tight binding of the ligand to the enzyme. Introduction of hydrophobic group to naphtholsulfonate moiety increases the binding affinity for the enzyme, and of the inhibition. These results suggest the location of hydrophobic regions as the allosteric inhibitory sites of the enzyme for the binding of naphtholsulfonate compounds.     

Rat brain hexokinase: the hydrophobic N-terminus of the mitochondrially bound enzyme is inserted in the lipid bilayer

Mitochondrially bound rat brain hexokinase was labeled with the photoactivatable reagent, 3-(trifluoromethyl)-3-(m-[125I]iodophenyl)diazirine. This highly hydrophobic reagent is strongly partitioned into the hydrophobic environment of the membrane core, and thus selectively labels segments of a protein that penetrate this region of the membrane. Labeling of hexokinase was shown to be restricted to the N-terminal region of the molecule. Approximately 80% of the radiolabel was removed by treatment of the enzyme with chymotrypsin, which preferentially cleaves a hydrophobic 9-residue sequence at the extreme N-terminus of the enzyme, and it is considered likely that the remaining 20% was associated with two additional hydrophobic residues, immediately adjacent to this segment but not susceptible to cleavage by chymotrypsin. Labeling of the enzyme was shown to be dependent on maintenance of the association with the membrane. These results are consistent with a model in which binding of hexokinase involves insertion of an 11-residue hydrophobic N-terminal "tail," possibly existing in alpha-helical secondary structure, into the hydrophobic core of the membrane.     

Site-site interactions in glycogen phosphorylase b probed by ligands specific for each site

Three ligand binding sites on glycogen phosphorylase b which were originally described by kinetic and physicochemical means, and more recently located and defined in molecular terms by X-ray crystallography, have been probed by ligands specific for each site. Kinetic analyses, supplemented by X-ray crystallographic binding studies, permit assignment of each ligand to a primary binding site, as well as determination of its dissociation constant and interaction with ligands binding to the other sites. 8-Anilino-1-naphthalenesulfonate binds most strongly to the activator site, in competition with adenosine 5'-phosphate, presumably because its sulfonate group interacts with several arginine residues, and binds only weakly to the hydrophobic inhibitor site, possibly because of charge repulsion. It is itself a weak activator and decreases binding affinities for compounds specific for the inhibitor site. Our results with 8-anilino-1-naphthalenesulfonate are not consistent with predictions of its expected behavior and suggest caution in the use of this reagent as an indicator of hydrophobicity. Our second major probe, caffeine, binds primarily to the inhibitor site, shows competitive inhibition with substrate binding to the catalytic site, and decreases the affinity for the activator at the activator site. The catalytic site was probed with two different types of ligand. Glucose, known to stabilize the inactive T conformation of the enzyme, competes with the substrate alpha-D-glucose 1-phosphate for the catalytic site and decreases the affinity of adenosine 5'-phosphate for the activator site. Glucose also improves the binding affinity of caffeine for the inhibitor site by 3-5-fold, both compounds synergistically stabilizing the inactive T conformation.(ABSTRACT TRUNCATED AT 250 WORDS)     

Interaction of pregna-D'-pentaranes--pentacyclic derivatives of progesterone--with gestagen-binding sites of uterine cytosol

The interaction of promegestone (R-5020), progesterone (P) and its derivatives having and additional carbocyclic D' (pregna-D'-pentrans) with progestin-binding cytosol system of the uterus was studied in different species (rabbits, rats, guinea-pigs and men). A comparative analysis of the competitive binding data for the mentioned compounds has shown interspecies differences in ligand specificity. Two types of binding sites for 3H-D'-pentran (in contrast to R-5020 and P) have been detected in rabbit uterus cytosol, both in intact and estrogenized animals. However, in rabbits, estrogenization altered the values of the apparent equilibrium constants and binding capacities. At the same time, the interaction of pentran with progestin-binding sites in guinea-pig and human uterus cytosol is nonspecific. It is suggested that the features of the interaction of 3H-D'-pentran with its binding sites in rabbit uterus cytosol may be determined by an increase in hydrophobic bond.     

An intact hydrophobic N-terminal sequence is critical for binding of rat brain hexokinase to mitochondria

The N-terminal sequence of rat brain hexokinase (ATP: D-hexose-6-phosphotransferase, EC 2.7.1.1) has been determined to be X-NH-Met-Ile-(Ala, Gln)-Ala-Leu-Leu-Ala-Tyr-, where X is a blocking group on the N-terminal methionine, probably an N-acetyl group. Modification of this hydrophobic N-terminal segment by endogenous proteases in crude brain extracts resulted in loss of the ability to bind to mitochondria, but had no effect on catalytic activity, resulting in the appearance of nonbindable enzyme reported by several previous investigators to be present in purified hexokinase preparations. Similar results can be obtained by deliberate limited digestion with chymotrypsin (cleavage points marked by arrows in sequence above). Both bindable and nonbindable enzyme, the latter generated either by endogenous proteases or with chymotrypsin, have an identical C-terminal dipeptide sequence, Ile-Ala. The great susceptibility of the N-terminus to proteolysis plus the marked effect that its proteolytic modification has on binding of hexokinase to anion exchange or hydrophobic (phenyl-Sepharose) matrices suggest that this N-terminal segment is prominently displayed at the enzyme surface. Epitopes recognized by two monoclonal antibodies which block binding of hexokinase to mitochondria (but have no effect on catalytic activity) have been mapped to a 10K fragment cleaved from the N-terminus by limited tryptic digestion. Thus the binding of hexokinase to mitochondria appears to occur via a "binding domain" constituting the N-terminal region of the molecule, with maintenance of an intact hydrophobic sequence at the extreme N-terminus being critical to this interaction. A resulting specific orientation of the molecule on the mitochondrial surface is considered to be a prerequisite for the observed coupling of hexokinase activity and mitochondrial oxidative phosphorylation.     

Thermodynamics of complex formation between bovine liver glutamate dehydrogenase and analogs of ADP.

A calorimetric study of the thermodynamic parameters for the binding of adenosine, AMP, ADP, and ATP to L-glutamate dehydrogenase shows that the variation of deltaG0 of binding is quite small and is correlated qualitatively both with the effectiveness of these ribonucleotides as activators of the L-glutamate dehydrogenase reaction and with size (for the first three). Much larger variations are observed for the deltaH0 of binding largely compensated by changes in deltaS0, with a zig-zag dependence on the number of phosphate groups. For comparison, the binding parameters are also obtained for the deoxyribose analogs of these compounds as well as cyclic adenosine 3':5'-monophosphate and purine riboside. Salt concentration and buffer composition were shown to affect mainly the entropy changes for ADP binding; and the deltaCp values for binding of AMP and ADP to the enzyme are quite small. It is suggested that the general area of the enzyme surface which includes the binding sites for ADP and its analogs contains a number of functional groups, each capable of an energetically small interaction with some group on one or more of the ligands, but so located that even the largest ligand cannot interact with all of them simultaneously. Each ligand minimizes the free energy of the system by selecting the best pattern of such individual interactions permitted by its geometry. Such a pattern of microheterogeneity of ligand-protein interactions may be a major source of the known specificity of binding in biological systems.     

Inactivation of Enzymes by Linoleic Acid Hydroperoxides and Linoleic Acid

The inactivation of the enzymes by linoleic acid hydroperoxides (LAHPO) was tested in connection with the toxicity of oxidized fat. At the same time, the inhibition of enzyme activities by linoleic acid was also tested. Ribonuclease (RNase), trypsin, chymotrypsin and pepsin which are considered to be simple proteins and not to be SH-enzymes were chosen as the enzymes. RNase was largely inhibited by LAHPO, but the other enzymes were inhibited by linoleic acid as well as LAHPO. The inhibition of each enzyme occurred at different pH. This fact may show that the inhibition occurs by binding of such hydrophobic compounds to the enzyme, and that the surface exposition of hydrophobic region may depend on the pH. Not only the reaction of some specific amino acid residue in the protein molecules with LAHPO, but also the binding of these hydrophobic compounds must be remembered in the mechanism of inhibition.     

The concentration dependence of active K+ transport in the turkey erythrocyte. Hill analysis and evidence for positive cooperativity between ion binding sites

A mathematical model is presented which describes the theoretical relationship between ligand concentration and physiological response for systems in which the response is dependent upon simultaneous occupancy of two receptor ligand-binding sites. The treatment considers both the possibility of intrinsic differences between the binding sites with regard to ligand affinity, as well as the possibility of mutually induced changes in affinity resulting from allosteric interactions. Unlike the Monod-Wyman-Changeux formulation for allosteric enzymes, the general model put forward here makes double occupancy an absolute requirement for enzymatic function. It is shown that such a model leads to the prediction of a curvilinear Hill plot from which one can obtain an explicit estimate of the degree of allosteric interaction between the two ligand binding sites as well as the Gibbs standard free energy change for the overall binding reaction. It is then shown that, in the specific instance of Na, K-ATPase-mediated K+ transport by the turkey erythrocyte, the configuration of the Hill curve describing the rate of ouabain-sensitive K+ transport as a function of external K+ concentration conforms closely to that predicted by the model described above. The results are of particular interest because they indicate a strongly cooperative interaction between the two K+ binding sites on the transport protein such that occupancy of one site results in an enhancement of the affinity of the other site for K+ by a minimum of 15- to 20-fold. Finally, we consider in detail a model of the Monod-Wyman- Changeux type in which, by contrast, both singly and doubly occupied forms of the enzyme are assumed to be catalytically active, and which we analogously extend to allow for the possibility of asymmetry between the two ligand binding sites. Although it is shown that the two models can not be differentiated from each other in the present experimental system, they yield virtually identical estimates for the degree of positive cooperativity between the two K+ binding sites.     

Structural consequences of accommodation of four non-cognate amino acid residues in the S1 pocket of bovine trypsin and chymotrypsin

Crystal structures of P1 Gly, Val, Leu and Phe bovine pancreatic trypsin inhibitor (BPTI) variants in complex with two serine proteinases, bovine trypsin and chymotrypsin, have been determined. The association constants for the four mutants with the two enzymes show that the enlargement of the volume of the P1 residue is accompanied by an increase of the binding energy, which is more pronounced for bovine chymotrypsin. Since the conformation of the P1 side-chains in the two S1 pockets is very similar, we suggest that the difference in DeltaG values between the enzymes must arise from the more polar environment of the S1 site of trypsin. This results mainly from the substitutions of Met192 and Ser189 observed in chymotrypsin with Gln192 and Asp189 present in trypsin. The more polar interior of the S1 site of trypsin is reflected by a much higher order of the solvent network in the empty pocket of the enzyme, as is observed in the complexes of the two enzymes with the P1 Gly BPTI variant. The more optimal binding of the large hydrophobic P1 residues by chymotrypsin is also reflected by shrinkage of the S1 pocket upon the accommodation of the cognate residues of this enzyme. Conversely, the S1 pocket of trypsin expands upon binding of such side-chains, possibly to avoid interaction with the polar residues of the walls. Further differentiation between the two enzymes is achieved by small differences in the shape of the S1 sites, resulting in an unequal steric hindrance of some of the side-chains, as observed for the gamma-branched P1 Leu variant of BPTI, which is much more favored by bovine chymotrypsin than trypsin. Analysis of the discrimination of beta-branched residues by trypsin and chymotrypsin is based on the complexes with the P1 Val BPTI variant. Steric repulsion of the P1 Val residue by the walls of the S1 pocket of both enzymes prevents the P1 Val side-chain from adopting the most optimal chi1 value.     

Exploring the Binding Nature of Pyrrolidine Pocket-Dependent Interactions in the Polo-Box Domain of Polo-Like Kinase 1

Background

Over the years, a great deal of effort has been focused on the design and synthesis of potent, linear peptide inhibitors targeting the polo-like kinase 1 (Plk1), which is critically involved in multiple mitotic processes and has been established as an adverse prognostic marker for tumor patients. Plk1 localizes to its intracellular anchoring sites via its polo-box domain, and inhibiting the Plk1 polo-box domain has been considered as an approach to circumvent the specificity problems associated with inhibiting the conserved adenosine triphosphate-binding pocket. The polo-box domain consists of two different binding regions, such as the unique, broader pyrrolidine-binding pocket and the conserved, narrow, Tyr-rich hydrophobic channel, among the three Plk polo-box domains (Plks 1–3), respectively. Therefore, the studies that provide insights into the binding nature of the unique, broader pyrrolidine-binding pocket might lead to the development of selective Plk1-inhibitory compounds.

Methodology/Principal Findings

In an attempt to retain the monospecificity by targeting the unique, broader pyrrolidine-binding pocket, here, for the first time, a systematic approach was undertaken to examine the structure-activity relationship of N-terminal-truncated PLHSpTM derivatives, to apply a site-directed ligand approach using bulky aromatic and non-aromatic systems, and to characterize the binding nature of these analogues using X-ray crystallographic studies. We have identified a new mode of binding interactions, having improved binding affinity and retaining the Plk1 polo-box domain specificity, at the pyrrolidine-binding pocket. Furthermore, our data revealed that the pyrrolidine-binding pocket was very specific to recognize a short and bulky hydrophobic ligand like adamantane, whereas the Tyr-rich hydrophobic channel was specific with lengthy and small hydrophobic groups.

Conclusion/Significance

The progress made using our site-directed ligands validated this approach to specifically direct the ligand into the unique pyrrolidine-binding region, and it extends the applicability of the strategy for discovering selective protein-protein interaction inhibitors.     

Solution structure of a repeated unit of the ABA-1 nematode polyprotein allergen of Ascaris reveals a novel fold and two discrete lipid-binding sites

Background

Nematode polyprotein allergens (NPAs) are an unusual class of lipid-binding proteins found only in nematodes. They are synthesized as large, tandemly repetitive polyproteins that are post-translationally cleaved into multiple copies of small lipid binding proteins with virtually identical fatty acid and retinol (Vitamin A)-binding characteristics. They are probably central to transport and distribution of small hydrophobic compounds between the tissues of nematodes, and may play key roles in nutrient scavenging, immunomodulation, and IgE antibody-based responses in infection. In some species the repeating units are diverse in amino acid sequence, but, in ascarid and filarial nematodes, many of the units are identical or near-identical. ABA-1A is the most common repeating unit of the NPA of Ascaris suum, and is closely similar to that of Ascaris lumbricoides, the large intestinal roundworm of humans. Immune responses to NPAs have been associated with naturally-acquired resistance to infection in humans, and the immune repertoire to them is under strict genetic control.

Methodology/Principal Findings

The solution structure of ABA-1A was determined by protein nuclear magnetic resonance spectroscopy. The protein adopts a novel seven-helical fold comprising a long central helix that participates in two hollow four-helical bundles on either side. Discrete hydrophobic ligand-binding pockets are found in the N-terminal and C-terminal bundles, and the amino acid sidechains affected by ligand (fatty acid) binding were identified. Recombinant ABA-1A contains tightly-bound ligand(s) of bacterial culture origin in one of its binding sites.

Conclusions/Significance

This is the first mature, post-translationally processed, unit of a naturally-occurring tandemly-repetitive polyprotein to be structurally characterized from any source, and it belongs to a new structural class. NPAs have no counterparts in vertebrates, so represent potential targets for drug or immunological intervention. The nature of the (as yet) unidentified bacterial ligand(s) may be pertinent to this, as will our characterization of the unusual binding sites.     

Kinetics and specificity of serine proteases in peptide synthesis catalyzed in organic solvents

Initial rates of peptide-bond synthesis catalyzed by poly(ethylene glycol)-modified chymotrypsin in benzene were determined using high-performance liquid chromatography. Enzymatic synthesis of N-benzoyl-L-tyrosyl-L-phenylalanine amide from N-benzoyl-L-tyrosine ethyl ester and L-phenylalanine amide was found to obey Michaelis-Menten kinetics an to be consistent with a ping-pong mechanism modified by a hydrolytic branch. The catalytic activity of modified chymotrypsin was dependent on both water concentration and type of organic solvent, the highest synthesis rate being obtained in toluene. Since the chymotrypsin specificity in the organic phase was actually altered, the enzyme's apparent kinetic parameters were determined for different substrates and compared to those obtained with other serine proteases in benzene. Both N-benzoyl-L-tyrosine ethyl ester and N-alpha-benzoyl-L-lysine methyl ester were comparable acyl donors in benzene and the (kcat/Km)app value of modified chymotrypsin was only 10-fold smaller than that obtained with poly(ethylene glycol)-modified trypsin in the synthesis of N-alpha-benzoyl-L-lysyl-L-phenylalanine amide. The change in chymotrypsin specificity was also confirmed through the binding of trypsin inhibitors in benzene. The overall results suggest that hydrophobic bonding between the enzyme and its substrate should not be taken into account during catalysis in the organic phase. In general, if hydrophobic interactions are involved in the binding of substrates to the active site in aqueous media, the replacement of water by hydrophobic solvents will induce some change in enzyme specificity. Moreover, secondary residues of enzyme-binding sites may also exert a significant influence on specificity since, as observed in this study, chymotrypsin exhibited high affinity for cationic substrates and cationic inhibitors as well in apolar solvents.     

Recent advances in cardiac glycoside-Na+,K+-ATPase interaction.

Na+,K+-ATPase has been purified from lamb kidney and consists of two polypeptide peaks on polyacrylamide gel electrophoresis with an enzyme activity of 1,000 mumole Pi/mg pro per hr. A scheme depicting the interaction of cardiac glycoside with the enzyme and ligand effects on binding has been constructed. Under all ligand conditions, ouabain binding tends to reach the same maximum if sufficient ouabain is present. Initial rates vary with ligand conditions. Using a chase method, the rate of dissociation of the glycoside from the enzyme is not influenced by the ligands present, although with separation of the enzyme-glycoside complex from the binding medium, differences are noted. The effect of ouabain on Na binding demonstrated two classes of sites, KD = 0.2 mM and KD = 18 mM. Denaturation decreased the high affinity sites. There was also a good correlation between ouabain binding and inhibition of Na binding. Clearly, ligands are critical in regulating cardiac glycoside interaction with the enzyme.     

Field‐based comparison of ligand and coactivator binding sites of nuclear receptors

A structure‐based comparison of the ligand‐binding domains of 35 nuclear receptors from five different subfamilies is presented. Their ligand and coactivator binding sites are characterized using knowledge‐based contact preference fields for hydrophobic and hydrophilic interactions implemented in the MOE modeling environment. Additionally, for polar knowledge‐based field points the preference for negative or positive electrostatic interactions is estimated using the Poisson‐Boltzmann equation. These molecular‐interaction fields are used to cluster the nuclear receptor family based on similarities of their binding sites. By analyzing the similarities and differences of hydrophobic and polar fields in binding pockets of related receptors it is possible to identify conserved interactions in ligand and coactivator binding pockets, which support e.g. design of specific ligands during lead optimization or virtual screening as docking filter. Examples of remarkable similarities between ligand binding sites of members from phylogenetically different nuclear receptor families (RXR, RAR, HNF4, NR5) and differences between closely related subtypes (LXR, RAR, TR) are discussed in more detail. Significant similarities and differences of coactivator binding sites are shown for NR3Cs, LXRs and PPARs. © 2009 Wiley Periodicals, Inc. Biopolymers 91: 884–894, 2009. This article was originally published online as an accepted preprint. The “Published Online” date corresponds to the preprint version. You can request a copy of the preprint by emailing the Biopolymers editorial office at biopolymers@wiley.com     

A model for the behaviour of phosphorylase b. The generation of different binding sites via intermediate enzymatic states.

The model given in this paper can be applied to enzymatic systems which have more than two conformational states in equilibrium and which clearly exhibit heterogeneity in the binding of one ligand. The model we propose makes possible quantitative interpretation of our experimental results and of those of many other workers as well. In some cases calorimetric, dialysis and kinetic magnitudes, when plotted against ligand concentration, give multiregional or "stepwise" curves. We suggest that such a behaviour arises because total occupation of one class of binding sites completely moves the enzyme towards a different conformational state in which the affinity for the ligand is greatly increased by the formation of a new class of binding sites. Our calorimetric results for the interaction between some nucleotides and phosphorylase b closely conform to our model.     

Localization of the high affinity calcium-binding site on tubulin molecule

Tubulin is a calcium-binding protein. Two different modes of interaction of calcium with tubulin have been described: a high affinity interaction to one or two binding sites and lower affinity interactions to several other binding sites. In the present study, we have used limited proteolysis of tubulin with trypsin, chymotrypsin, and subtilisin to localize the high affinity calcium-binding sites. Our results indicate that two sites are located in the carboxyl-terminal region of both tubulin subunits, and that tubulin deprived of its carboxyl-terminal region is able to polymerize in the presence of 0.5 mM calcium.