Evolving a lingua franca and associated software infrastructure for computational systems biology: the Systems Biology Markup Language (SBML) project.

Biologists are increasingly recognising that computational modelling is crucial for making sense of the vast quantities of complex experimental data that are now being collected. The systems biology field needs agreed-upon information standards if models are to be shared, evaluated and developed cooperatively. Over the last four years, our team has been developing the Systems Biology Markup Language (SBML) in collaboration with an international community of modellers and software developers. SBML has become a de facto standard format for representing formal, quantitative and qualitative models at the level of biochemical reactions and regulatory networks. In this article, we summarise the current and upcoming versions of SBML and our efforts at developing software infrastructure for supporting and broadening its use. We also provide a brief overview of the many SBML-compatible software tools available today.

Replacement of tyrosine D with phenylalanine affects the normal proton transfer pathways for the reduction of P680+ in oxygen-evolving photosystem II particles from Chlamydomonas.

We have probed the electrostatics of P680(+) reduction in oxygenic photosynthesis using histidine-tagged and histidine-tagged Y(D)-less Photosystem II cores. We make two main observations: (i) that His-tagged Chlamydomonas cores show kinetics which are essentially identical to those of Photosystem II enriched thylakoid membranes from spinach; (ii) that the microsecond kinetics, previously shown to be proton/hydrogen transfer limited [Schilstra et al. (1998) Biochemistry 37, 3974-3981], are significantly different in Y(D)-less Chlamydomonas particles when compared with both the His-tagged Chlamydomonas particles and the spinach membranes. The oscillatory nature of the kinetics in both Chlamydomonas samples is normal, indicating that S-state cycling is unaffected by either the histidine-tagging or the replacement of tyrosine D with phenylalanine. We propose that the effects on the proton-coupled electron transfers of P680(+) reduction in the absence of Y(D) are likely to be due to pK shifts of residues in a hydrogen-bonded network of amino acids in the vicinity of Y(Z). Tyrosine D is 35 A from Y(Z) and yet has a significant influence on proton-coupled electron transfer events in the vicinity of Y(Z). This finding emphasizes the delicacy of the proton balance that Photosystem II has to achieve during the water splitting process.

Relationship between excitation energy transfer, trapping, and antenna size in photosystem II.

We present a systematic study of the effect of antenna size on energy transfer and trapping in photosystem II. Time-resolved fluorescence experiments have been used to probe a range of particles isolated from both higher plants and the cyanobacterium Synechocystis 6803. The isolated reaction center dynamics are represented by a quasi-phenomenological model that fits the extensive time-resolved data from photosystem II reaction centers and reaction center mutants. This representation of the photosystem II "trapping engine" is found to correctly predict the extent of, and time scale for, charge separation in a range of photosystem II particles of varying antenna size (8-250 chlorins). This work shows that the presence of the shallow trap and slow charge separation kinetics, observed in isolated D1/D2/cyt b559 reaction centers, are indeed retained in larger particles and that these properties are reflected in the trapping dynamics of all larger photosystem II preparations. A shallow equilibrium between the antennae and reaction center in photosystem II will certainly facilitate regulation via nonphotochemical quenching, and one possible interpretation of these findings is therefore that photosystem II is optimized for regulation rather than for efficiency.

Proton/hydrogen transfer affects the S-state-dependent microsecond phases of P680+ reduction during water splitting.

To investigate a possible coupling between P680+ reduction and hydrogen transfer, we studied the effects of H2O/D2O exchange on the P680+ reduction kinetics in the nano- and microsecond domains. We concentrated on studying the period-4 oscillatory (i.e., S-state-related) part of the reduction kinetics, by analyzing the differences between the P680+ reduction curves, rather than the full kinetics. Earlier observations that P680+ reduction kinetics have microsecond components were confirmed: the longest observable lifetime whose amplitude showed period-4 oscillations was 30 microseconds. We found that solvent isotope exchange left the nanosecond phases of the P680+ reduction unaltered. However, a significant effect on the oscillatory microsecond components was observed. We propose that, at least in the S0/S1 and S3/S0 transitions, hydrogen (proton) transfer provides an additional decrease in the free energy of the YZ+P680 state with respect to the YZP680+ state. This implies that relaxation of the state YZ+P680 is required for complete reduction of P680+ and for efficient water splitting. The kinetics of the P680+ reduction suggest that it is intraprotein proton/hydrogen rearrangement/transfer, rather than proton release to the bulk, which is occurring on the 1-30 microseconds time scale.

Mechanism of lipoxygenase inactivation by the linoleic acid analogue octadeca-9,12-diynoic acid.

During the irreversible inactivation of soybean Fe(III)-lipoxygenase [Fe(III)-LOX] by octadeca-9,12-diynoic acid (ODYA), significant quantities of 11-oxooctadeca-9,12 diynoic acid (11-oxo-ODYA) are formed [Nieuwenhuizen, W. F., et al. (1995) Biochemistry 34, 10538-10545]. To elucidate the inactivation mechanism, a quantitative study into the relationship between the inactivation and 11-oxo-ODYA formation was carried out. The following observations were made (1) LOX (0.84 microM) was completely inactivated by 10 to 80 microM ODYA. However, at ODYA concentrations greater than 100 microM, LOX was only partially inactivated, and there was no inactivation at all at ODYA concentrations above 750 microM. The average number of turnovers in which 11-oxo-ODYA was formed increased from 1.2 to 12 when the ODYA concentration increased from 1 to 50 microM and then decreased again to 1.2 at 1000 microM ODYA. (2) The enzyme that was not irreversibly inactivated by ODYA was in the Fe(III) form at ODYA concentrations below 10 microM but in the Fe(II) form at ODYA concentrations greater than 100 microM. (3) In the presence of 750 microM ODYA and 25 microM 13(S)-hydroperoxy-9Z,11E-octadecadienoic acid, all of the enzyme was inactivated. On the basis of these results, it is proposed that the dioxygenation product of ODYA is 11-hydroperoxyoctadeca-9,12-diynoic acid (11-HP-ODYA), which can convert Fe(II)-LOX into its Fe(III) form. However 11-HP-ODYA is converted into 11-oxo-ODYA, which cannot perform the oxidation. It is proposed that the inactivating agent is either 11-HP-ODYA or the 11-peroxy-octadeca-9,12-diynoic acid radical (11-peroxy-ODYA radical), formed from the ODYA radical and O2. The oxidation of Fe(II)-LOX into its Fe(III) form as well as the inactivation of Fe(III)-LOX is competitively inhibited by ODYA

Fe(III)-lipoxygenase converts its suicide-type inhibitor octadeca-9,12-diynoic acid into 11-oxooctadeca-9,12-diynoic acid.

Triple bond analogues of polyunsaturated fatty acids irreversibly inactivate lipoxygenases. During the inactivation the inhibitors are converted enzymatically [Kühn, H., et al. (1984) Eur. J. Biochem. 139, 577-583]. Since the converted inhibitor molecules may hold important information about the inactivation mechanism, we have determined the structure of the product that is formed during the irreversible inactivation of soybean lipoxygenase-1 by octadeca-9,12-diynoic acid (ODYA), the triple bond analogue of linoleic acid. This product is formed only in the presence of Fe(III)-lipoxygenase-1 and O2. It was purified by C18 solid phase extraction and reversed phase HPLC and was identified with UV, IR, and NMR spectroscopic and mass spectrometric techniques as the novel lipoxygenase product, 11-oxooctadeca-9,12-diynoic acid (11-oxo-ODYA). It is estimated that each lipoxygenase molecule produces 8-10 11-oxo-ODYA molecules before it is inactivated. Furthermore, we have shown that in a secondary reaction 3-4 molecules of 11-oxo-ODYA are covalently attached per lipoxygenase molecule, most likely, to solvent-exposed amino groups. This leads to the formation of a N-penten-4-yn-3-one chromophore, RC(NHX)=CHC(O)C=CR1, in which X stands for the protein and R or R1 for CH3(CH2)4- or -(CH2)7COOH, respectively. Fe(II)- and Fe(III)-lipoxygenase remain active upon reaction with purified 11-oxo-ODYA. It is concluded that (a) several enzymatic turnovers are required for the complete inactivation of lipoxygenase by ODYA and (b) covalent attachment of 11-oxo-ODYA occurs outside the active site and is not the cause of the inactivation.

Changes in the iron coordination sphere of Fe(II) lipoxygenase-1 from soybeans upon binding of linoleate or oleate.

Fe K-edge X-ray absorption spectra of the non-heme iron constituent of lipoxygenase-1 from soybeans were obtained. The spectrum of 2.5 mM Fe(II) lipoxygenase, mixed with 1.2 M linoleate in the absence of O2, was compared to the spectrum of the native (i.e. untreated) enzyme. In the lipoxygenase-linoleate complex, an edge shift to lower energy was observed. This indicated that the iron-ligand distances in this complex are slightly longer than those in the untreated enzyme species. The extended X-ray absorption fine structure spectrum of Fe(II) lipoxygenase, prepared by anaerobic reduction of 2.5 mM Fe(II) lipoxygenase with 1.2 M linoleate, was very similar to the spectrum of the anaerobic lipoxygenase-linoleate complex. We conclude that the conformational differences between the iron coordination spheres of native and cycled Fe(II) lipoxygenase must be ascribed to the presence of linoleate, and not to changes in the enzyme that occur only after one cycle of oxidation and reduction. Furthermore, spectra of 2.5 mM Fe(II) lipoxygenase mixed with 1.2 M oleate, either in the absence or in the presence of O2, were also identical to the spectrum of the Fe(II) lipoxygenase-linoleate complex. This finding is in agreement with our observation that oleate is a competitive inhibitor of the lipoxygenase reaction. Moreover, the similarity of the lipoxygenase-oleate complexes in the presence and absence of O2 excludes the possibility that O2 binding to the iron cofactor is induced upon binding of a fatty acid to lipoxygenase.

Effect of nonionic detergents on lipoxygenase catalysis.

In many studies on lipoxygenase catalysis, nonionic detergents are used to obtain an optically transparent solution of the fatty acid substrate. In order to resolve some controversies that exist with regard to the interpretation of kinetic data obtained with solutions containing nonionic detergents, a systematic investigation was undertaken into the effects of Lubrol, Tween-20 and Triton X-100 (0-0.8 g/L) on the kinetics of linoleate (2.5-110 microM) dioxygenation, catalyzed by lipoxygenase-1 or lipoxygenase-2 from soybean, at pH 9 or 10, at 25 degrees C. Under most conditions, it was found that the detergents slowed down the reaction. However, at high linoleate concentrations, where substrate inhibition of lipoxygenase is significant, small amounts of detergent increased the dioxygenation rate. In a quantitative analysis of the results, a kinetic model in which the incorporation of linoleate in the detergent micelles is formulated as a simple reversible equilibrium, and in which both lipoxygenase-1 and -2 interact with free linoleate, but not with linoleate incorporated in the micelles, appeared to be sufficient to predict experimental results over a wide range of experimental conditions. According to this model, the changes in the dioxygenation kinetics caused by the presence of nonionic detergents are similar (but not equal) to those caused by competitive inhibitors. The conclusions that monomeric, nonmicellar linoleate is the preferred substrate for lipoxygenase and that the observed inhibition and stimulation are solely due to changes in the effective linoleate concentration strongly corroborate the earlier observations by Galpin and Allen [Biochim. Biophys. Acta 488 (1977), 392-401].

The dioxygenation rate in lipoxygenase catalysis is determined by the amount of iron (III) lipoxygenase in solution.

The dioxygenation rate in reactions catalyzed by lipoxygenase-1 from soybeans has been measured as a function of the enzyme present in the Fe(III) form with rapid kinetic techniques. The experiments were carried out at pH 10, 25 degree C. The product concentration and the fraction of iron (III) lipoxygenase were monitored by measuring the absorbance at 243 nm and the tryptophan fluorescence at 330 nm (excitation at 287 nm), respectively. In reactions started with 1.3 microM iron (II) lipoxygenase and 9 microM linoleate, the initial rate, r(init) (estimated from the increase in absorbance over the initial 0.02 s of the reaction), is very small (4 s-1). In contrast, when the reactions are started with 1.3 microM (III) lipoxygenase, r(init) is large (150 s-1). In reactions started with mixtures of iron(II) and iron(III) lipoxygenase, r(init) is linearly related to the initial concentration of the Fe (III) enzyme form. Redistributions of the Fe(II) and Fe(III) enzyme forms during the reaction with 12 nM enzyme and 10, 50, or 100 microM linoleate appear to be directly reflected in changes in the dioxygenation rate. The observations provide solid evidence for the hypothesis that only iron (III) lipoxygenase can catalyze the hydrogen abstraction step in the dioxygenation reaction, and thus can be regarded as the active enzyme species. The observed dynamics are accurately predicted by a nonallosteric, two-step model for lipoxygenase catalysis [Schilstra et al. (1992) Biochemistry 31, 7692-7699].

Effects of the tubulin-colchicine complex on microtubule dynamic instability.

The effects of the tubulin-colchicine complex (Tu-Col) on the dynamic behavior of microtubules have been examined under steady-state conditions in vitro. The addition of Tu-Col to tubulin microtubules at steady state results in only partial microtubule disassembly. Nevertheless, both the rate and the extent of tubulin exchange into microtubules are markedly suppressed by concentrations of Tu-Col which are low relative to the total amount of free tubulin. In addition, the time-dependent changes in microtubule length distribution, characteristic of dynamic instability, are suppressed by the addition of Tu-Col. Examination by video-enhanced dark-field microscopy of individual microtubules in the presence of Tu-Col shows that the principal effect of this complex is to reduce the growth rate at both ends of the microtubule. We have used computer simulation to rationalize the action of Tu-Col in terms of its effects on the experimentally observable parameters, namely, the rates of growth and shortening and the mean lifetimes of growth and shortening, which provide an empirical description of the dynamic behavior of microtubules. The results have been interpreted within the framework of the lateral cap formulation for microtubule dynamic instability [Martin, S. R., Schilstra, M. J., & Bayley, P. M. (1993) Biophys. J. 65, 578-596]. The simplest model mechanism requires only that Tu-Col binds to the microtubule end and inhibits further addition reactions in either the 5-start or the 8-start direction of the microtubule lattice. Monte Carlo simulations show that Tu-Col can, in this way, cause major suppression of the dynamic transitions of microtubules without inducing bulk microtubule disassembly. This type of mechanism could be important for the regulation of microtubule dynamics in vivo.

Dynamic instability of microtubules: Monte Carlo simulation and application to different types of microtubule lattice.

Dynamic instability is the term used to describe the transition of an individual microtubule, apparently at random, between extended periods of slow growth and brief periods of rapid shortening. The typical sawtooth growth and shortening transition behavior has been successfully simulated numerically for the 13-protofilament microtubule A-lattice by a lateral cap model (Bayley, P. M., M. J. Schilstra, and S. R. Martin. 1990. J. Cell Sci. 95:33-48). This kinetic model is now extended systematically to other related lattice geometries, namely the 13-protofilament B-lattice and the 14-protofilament A-lattice, which contain structural "seams". The treatment requires the assignment of the free energies of specific protein-protein interactions in terms of the basic microtubule lattice. It is seen that dynamic instability is not restricted to the helically symmetric 13-protofilament A-lattice but is potentially a feature of all A- and B-lattices, irrespective of protofilament number. The advantages of this general energetic approach are that it allows a consistent treatment to be made for both ends of any microtubule lattice. Important features are the predominance of longitudinal interactions between tubulin molecules within the same protofilament and the implication of a relatively favorable interaction of tubulin-GDP with the growing microtubule end. For the three lattices specifically considered, the treatment predicts the dependence of the transition behavior upon tubulin concentration as a cooperative process, in good agreement with recent experimental observations. The model rationalizes the dynamic properties in terms of a metastable microtubule lattice of tubulin-GDP, stabilized by the kinetic process of tubulin-GTP addition. It provides a quantitative basis for the consideration of in vitro microtubule behaviour under both steady-state and non-steady-state conditions, for comparison with experimental data on the dilution-induced disassembly of microtubules. Similarly, the effects of small tubulin-binding molecules such as GDP and nonhydrolyzable GTP analogues are readily treated. An extension of the model allows a detailed quantitative examination of possible modes of substoichiometric action of a number of antimitotic drugs relevant to cancer chemotherapy.

Kinetic analysis of the induction period in lipoxygenase catalysis.

The dioxygenation of 50 microM linoleate at 0.1 microM (13S)-hydroperoxylinoleate, 240 microM O2, pH 10, and 25 degrees C, catalyzed by varying amounts of soybean lipoxygenase-1, was studied with rapid kinetic techniques. The aim was to assess the effect of transient redistributions of the Fe(II) and Fe(III) enzyme forms on the shape of the reaction progress curves. Reactions initiated with iron(II) lipoxygenase show an initial increase in rate, the "kinetic lag phase" or "induction period". The duration of this induction period varies from approximately 1 s at [lipoxygenase] > 20 nM to 5 s at [lipoxygenase] = 3 nM. At [lipoxygenase] < 2 nM, the duration of the induction period in these curves is inversely proportional to [lipoxygenase]. The integrated steady-state rate equation for the single fatty acid binding site model of lipoxygenase catalysis [Schilstra et al. (1992) Biochemistry 31, 7692-7699] also shows an induction period whose duration is inversely proportional to [lipoxygenase]. These observations, in combination with non-steady-state numerical simulations, lead to the conclusion that, at [lipoxygenase] < 2 nM, pre-steady-state redistributions of enzyme intermediates occur fast with respect to the rate at which the concentrations of substrates and products change. At higher lipoxygenase concentrations, the pre-steady-state redistributions contribute significantly to the induction period. From a nonlinear least-squares fit to the steady-state rate equation of data obtained at lipoxygenase concentrations of 0.5 and 1 nM, it was calculated that 1% of the linoleate radicals that are formed after hydrogen abstraction dissociate from the active site before enzymic oxygen insertion has occurred.

Effect of lipid hydroperoxide on lipoxygenase kinetics.

In order to investigate the activation of lipoxygenase and to clarify the role of the oxygenation product hydroperoxide in this process, the effect of 13-hydroperoxylinoleic acid (P, 0-35 microM) on linoleic acid (S, 1-80 microM) oxygenation catalysis by 12 nM lipoxygenase-1 from soybean was studied at pH 10, 25 degrees C, and 240 microM O2 with rapid kinetic techniques. The following observations were made: (1) Iron(II) and iron(III) lipoxygenases are kinetically different: reactions started with the Fe(II) enzyme form show a lag phase, whereas iron(III) lipoxygenase induces an initial burst. (2) Oxidation of the enzyme alone is not sufficient to abolish the lag phase: at [S] greater than 50 microM, the initial burst in the iron(III) lipoxygenase curves is still followed by a lag. The lag phase disappears completely only in the presence of micromolar quantities of P. (3) The approximate dissociation constants for S and P are 15 and 24 microM, respectively, 1 order of magnitude smaller than the corresponding values in the absence of oxygen. The observed kinetics are predicted by numerical integration of the rate equations of a model based on the single lipid binding site mechanism for the anaerobic lipoxygenase reaction [Ludwig et al. (1987) Eur. J. Biochem. 168, 325-337; Verhagen et al. (1978) Biochim. Biophys. Acta 529, 369-379]. A quasi-steady-state approximation of the model suggests that a high [S]/[P] the fraction of active iron(III) lipoxygenase is small and that, therefore, a lag phase is intrinsic to the mechanism.

The effect of solution composition on microtubule dynamic instability.

The exchange of tubulin dimer into steady-state microtubules was studied over a range of solution conditions, in order to assess the effects of various common buffer components on the dynamic instability of microtubules. In comparison with standard buffer conditions (100 mM-Pipes buffer, pH 6.5, containing 0.1 mM-EGTA, 1.8 mM-MgC12 and 1 M-glycerol), the rate and extent of exchange, and thus of dynamic instability, are suppressed by increasing the concentration of glycerol above 2 M. Exchange is enhanced by the addition of further Mg2+ (up to 17 mM) or by the addition of Ca2+ (up to 0.4 mM). Phosphate ion (150 mM) has relatively little effect on the dynamic behaviour of microtubules, as judged by the exchange method. The findings are interpreted within the framework of the Lateral Cap model for microtubule dynamic instability, in terms of the effects of these changes on the intrinsic rate constants of the system. By contrast, the extent of tubulin exchange depends selectively on the value of the dissociation rate constant for tubulin-GDP. A decrease in the extent of exchange, and hence in dynamic activity, is associated with a decreased value for this rate constant, and vice versa. The results also show good agreement of predictions of the model in treating the observed variations in the dynamic properties of individual microtubules, induced by different solution conditions.

Opposite-end behaviour of dynamic microtubules.

Microtubules are dynamic polar structures with different kinetic properties at the two ends. The inherent asymmetry of the microtubule lattice determines that the relationship between the addition reaction of tubulin-GTP and the associated hydrolysis of a tubulin-GTP on the polymer is different at the two ends of the microtubule. We present a unified treatment for both ends of the microtubule, using the principles of the Lateral Cap formulation for microtubule dynamic instability. This shows that the two ends can exhibit significantly different dynamic properties in terms of amplitudes and lifetimes of growth and shrinking, depending on the relative importance of longitudinal and lateral contacts in the coupling of tubulin-GTP hydrolysis. These predictions are readily amenable to experimental verification. This modelling suggests that fine details of the subunit-subunit interactions at the microtubule end can determine the characteristic differences in kinetic behaviour of the opposite ends of dynamic microtubules. Variation of these interactions would provide a potentially sensitive general mechanism for the control of such dynamics, both in vitro and in vivo.

Microtubule dynamic instability: numerical simulation of microtubule transition properties using a Lateral Cap model.

We present a numerical formulation for the dynamic instability of microtubules involving the stabilisation of growing microtubules by a single layer of tubulin-GTP, with GTP hydrolysis effectively coupled to tubulin-GTP addition. This Lateral Cap model provides a readily visualised, working mechanism for the co-existence and interconversion of growing and shrinking microtubules. This class of model is specified in terms of a hydrolysis rule, whereby the addition of tubulin-GTP causes hydrolysis of GTP on a previously terminal tubulin-GTP molecule as it becomes incorporated into the microtubule lattice. A specific formulation is illustrated, though this is not unique. A limited set of parameters defines the kinetics and affinity for tubulin-GTP at the binding sites at a given end of the microtubule. The rate constants are a function of the nucleotide composition of the binding site, principally comprising the two tubulin molecules, which interact laterally and longitudinally with the incoming tubulin-GTP molecule. The Lateral Cap formulation demonstrates that a single terminal layer of tubulin-GTP is sufficient to reproduce the apparently complex behaviour of a dynamic population of microtubules. It differs significantly from the fluctuating tubulin-GTP cap model of Chen and Hill (1985). It gives a molecular description to the switching of individual microtubules between growing and shrinking states in terms of the composition of the multi-start terminal layer of the microtubule, and provides a general mechanism for the differential kinetic behaviour at opposite ends of dynamic microtubules. It reproduces the essential features of microtubule length excursions, and predicts detailed characteristics of microtubule dynamics, including the basis of the apparently cooperative nature of the transition behaviour as a function of the concentration of tubulin-GTP. It is readily amenable to further experimental test and refinement.

A simple formulation of microtubule dynamics: quantitative implications of the dynamic instability of microtubule populations in vivo and in vitro.

A simple formulation of microtubule dynamic instability is presented, which is based on the experimental observations by T. Horio and H. Hotani of coexisting, interconverting growing and shrinking microtubules. Employing only three independent, experimentally determined parameters for a given microtubule end, this treatment accounts quantitatively for the principal features of the observed dynamic behaviour of steady-state tubulin microtubules in vitro. Experimental data are readily reproduced for microtubule length redistribution, and for the kinetics of tubulin exchange processes, including pulse-chase properties. The relative importance of dynamic incorporation and that due to treadmilling are assessed. Dynamic incorporation is found to dominate the overall exchange properties; polarized incorporation due to treadmilling generally becomes significant only when the dynamics are largely suppressed. This treatment also permits simulation of certain cellular phenomena, showing how microtubule renucleation can control microtubule growth, by means of changes in microtubule number concentration in a system at constant microtubule mass. A relatively simple extension of the formulation accounts quantitatively for non-steady-state microtubule properties, e.g. dilution-induced rapid disassembly and the oscillatory mode of microtubule assembly. The principles relating dynamic instability and oscillatory behaviour are clearly indicated. Possible mechanisms of the switching of microtubules are briefly discussed.

The effect of podophyllotoxin on microtubule dynamics.

We have investigated the effects of podophyllotoxin on the dynamic properties of microtubules assembled from pure tubulin dimer. Excess podophyllotoxin causes the complete disassembly of microtubules, through formation of a tubulin-GTP-podophyllotoxin ternary complex with a dissociation rate constant of 160 s-1 at 37 degrees C, similar to that found upon extensive isothermal dilution in this buffer system. Addition of substoichiometric concentrations of podophyllotoxin causes partial disassembly of microtubules through production of an equivalent amount of the ternary complex. Microtubule length measurements and incorporation of [3H]GTP-tubulin dimer show that podophyllotoxin can suppress the dynamic instability of tubulin dimer microtubules and that it acts substoichiometrically in so doing. We interpret the action of substoichiometric podophyllotoxin on microtubule ends in terms of effects on interconversion of growing and shrinking microtubules in a dynamic system in which tubulin-GTP-podophyllotoxin is kinetically analogous to tubulin-GTP in addition and to tubulin-GDP in dissociation. The ability to suppress dynamic instability may be one way in which drugs such as podophyllotoxin, acting at relatively low concentrations, are able to arrest cell growth and development in a selective way, without necessarily affecting the integrity of the major part of the cytoskeletal microtubule network.

Dynamic properties of microtubules at steady state of polymerisation.

Microtubules of tubulin dimer, polymerised in vitro to steady-state are shown to incorporate tubulin rapidly and extensively. The method involves adding [3H]-GTP to microtubules at steady state, and analysing for non-exchangeable [3H]-GDP in the presence of a GTP regenerating system. The rate and extent of this exchange process is dependent on the length distribution of the microtubules, and is notably faster with sheared microtubules. We simulate all these features of the exchange kinetics, together with the length redistributions occurring at steady state of polymerisation, using a simple model based on a limited number of kinetic parameters deriving from the measurements of microtubule dynamics by Horio and Hotani. The observed exchange kinetics thus provide a direct experimental criterion of 'dynamic instability' of microtubules at steady state of polymerisation.

On the relationship between nucleotide hydrolysis and microtubule assembly: studies with a GTP-regenerating system.

The assembly of pure tubulin dimer has been studied in two buffer systems (containing low and high glycerol/Mg), using a regeneration system protocol to assess the amount of GDP-tubulin in the assembling polymer. For both assembly systems studied, the GDP content is effectively stoichiometric with tubulin throughout assembly. This indicates a high degree of coupling between assembly and GTP-hydrolysis, giving a hydrolysis rate at least 10-fold faster than previously deduced. The steady state GTP hydrolysis rate is quantitatively consistent with this finding. We conclude that the extent of any GTP-tubulin cap is below the detectable limit, both during elongation and at steady state.