Interaction between chromium GTP and tubulin. Stereochemistry of GTP binding, GTP hydrolysis, and microtubule stabilization

Disruption of lipid-raft localized Gαs/tubulin complexes by antidepressants: a unique feature of HDAC6 inhibitors, SSRI and tricyclic compounds

Current antidepressant therapies meet with variable therapeutic success and there is increasing interest in therapeutic approaches not based on monoamine signaling. Histone deacetylase 6 (HDAC6), which also deacetylates α-tubulin shows altered expression in mood disorders and HDAC6 knockout mice mimic traditional antidepressant treatments. Nonetheless, a mechanistic understanding for HDAC6 inhibitors in the treatment of depression remains elusive. Previously, we have shown that sustained treatment of rats or glioma cells with several antidepressants translocates Gα_s from lipid rafts toward increased association with adenylyl cyclase (AC). Concomitant with this is a sustained increase in cAMP production. While Gα_s modifies microtubule dynamics, tubulin also acts as an anchor for Gα_s in lipid-rafts. Since HDAC-6 inhibitors potentiate α-tubulin acetylation, we hypothesize that acetylation of α-tubulin disrupts tubulin-Gα_s raft-anchoring, rendering Gα_s free to activate AC. To test this, C6 Glioma (C6) cells were treated with the HDAC-6 inhibitor, tubastatin-A. Chronic treatment with tubastatin-A not only increased α-tubulin acetylation but also translocated Gα_s from lipid-rafts, without changing total Gα_s. Reciprocally, depletion of α-tubulin acetyl-transferase-1 ablated this phenomenon. While escitalopram and imipramine also disrupt Gα_s/tubulin complexes and translocate Gα_s from rafts, they evoke no change in tubulin acetylation. Finally, two indicators of downstream cAMP signaling, cAMP response element binding protein phosphorylation (pCREB) and expression of brain-derived-neurotrophic-factor (BDNF) were both elevated by tubastatin-A. These findings suggest HDAC6 inhibitors show a cellular profile resembling traditional antidepressants, but have a distinct mode of action. They also reinforce the validity of antidepressant-induced Gα_s translocation from lipid-rafts as a biosignature for antidepressant response that may be useful in the development of new antidepressant compounds.

Detection of GTP-tubulin conformation in vivo reveals a role for GTP remnants in microtubule rescues

Microtubules display dynamic instability, with alternating phases of growth and shrinkage separated by catastrophe and rescue events. The guanosine triphosphate (GTP) cap at the growing end of microtubules, whose presence is essential to prevent microtubule catastrophes in vitro, has been difficult to observe in vivo. We selected a recombinant antibody that specifically recognizes GTP-bound tubulin in microtubules and found that GTP-tubulin was indeed present at the plus end of growing microtubules. Unexpectedly, GTP-tubulin remnants were also present in older parts of microtubules, which suggests that GTP hydrolysis is sometimes incomplete during polymerization. Observations in living cells suggested that these GTP remnants may be responsible for the rescue events in which microtubules recover from catastrophe.

Energetics of the cooperative assembly of cell division protein FtsZ and the nucleotide hydrolysis switch

FtsZ is the first protein recruited to the bacterial division site, where it forms the cytokinetic Z ring. We have determined the functional energetics of FtsZ assembly, employing FtsZ from the thermophilic Archaea Methanococcus jannaschii bound to GTP, GMPCPP, GDP, or GMPCP, under different solution conditions. FtsZ oligomerizes in a magnesium-insensitive manner. FtsZ cooperatively assembles with magnesium and GTP or GMPCPP into large polymers, following a nucleated condensation polymerization mechanism, under nucleotide hydrolyzing and non-hydrolyzing conditions. The effect of temperature on the critical concentration indicates polymer elongation with an apparent heat capacity change of -800 +/- 100 cal mol-1 K-1 and positive enthalpy and entropy changes, compatible with axial hydrophobic contacts of each FtsZ in the polymer, and predicts optimal polymer stability near 75 degrees C. Assembly entails the binding of one medium affinity magnesium ion and the uptake of one proton per FtsZ. Interestingly, GDP- or GMPCP-liganded FtsZ cooperatively form helically curved polymers, with an elongation only 1-2 kcal mol-1 more unfavorable than the straight polymers formed with nucleotide triphosphate, suggesting a physiological requirement for FtsZ polymerization inhibitors. This GTP hydrolysis switch should provide the basic properties for FtsZ polymer disassembly and its functional dynamics.

Refined structure of alpha beta-tubulin at 3.5 A resolution

We present a refined model of the alpha beta-tubulin dimer to 3.5 A resolution. An improved experimental density for the zinc-induced tubulin sheets was obtained by adding 114 electron diffraction patterns at 40-60 degrees tilt and increasing the completeness of structure factor amplitudes to 84.7 %. The refined structure was obtained using maximum-likelihood including phase information from experimental images, and simulated annealing Cartesian refinement to an R-factor of 23.2 and free R-factor of 29.7. The current model includes residues alpha:2-34, alpha:61-439, beta:2-437, one molecule of GTP, one of GDP, and one of taxol, as well as one magnesium ion at the non-exchangeable nucleotide site, and one putative zinc ion near the M-loop in the alpha-tubulin subunit. The acidic C-terminal tails could not be traced accurately, neither could the N-terminal loop including residues 35-60 in the alpha-subunit. There are no major changes in the overall fold of tubulin with respect to the previous structure, testifying to the quality of the initial experimental phases. The overall geometry of the model is, however, greatly improved, and the position of side-chains, especially those of exposed polar/charged groups, is much better defined. Three short protein sequence frame shifts were detected with respect to the non-refined structure. In light of the new model we discuss details of the tubulin structure such as nucleotide and taxol binding sites, lateral contacts in zinc-sheets, and the significance of the location of highly conserved residues.

Interaction of myosin with F-actin: time-dependent changes at the interface are not slow

The kinetics of formation of the actin-myosin complex have been reinvestigated on the minute and second time scales in sedimentation and chemical cross-linking experiments. With the sedimentation method, we found that the binding of the skeletal muscle myosin motor domain (S1) to actin filament always saturates at one S1 bound to one actin monomer (or two S1 per actin dimer), whether S1 was added slowly (17 min between additions) or rapidly (10 s between additions) to an excess of F-actin. The carbodiimide (1-ethyl-3-(3-dimethylaminopropyl) carbodiimide, EDC)-induced cross-linking of the actin-S1 complex was performed on the subsecond time scale by a new approach that combines a two-step cross-linking protocol with the rapid flow-quench technique. The results showed that the time courses of S1 cross-linking to either of the two actin monomers are identical: they are not dependent on the actin/S1 ratio in the 0.3-20-s time range. The overall data rule out a mechanism by which myosin rolls from one to the other actin monomer on the second or minute time scales. Rather, they suggest that more subtle changes occur at the actomyosin interface during the ATP cycle.

Nutrient augmentation of Ca2+-dependent and Ca2+-independent pathways in stimulus-coupling to insulin secretion can be distinguished by their guanosine triphosphate requirements: studies on rat pancreatic islets

To delineate the underlying mechanisms by which glucose augments both Ca2+-dependent and Ca2+-independent insulin release, the latter induced by the simultaneous activation of protein kinases A and C, we examined the effects of GTP depletion by mycophenolic acid (MPA), an inhibitor of GTP synthesis, on the augmentation of insulin release from rat pancreatic islets. MPA treatment reduced GTP content by 30-40% and completely abolished glucose-induced augmentation of Ca2+-independent insulin release. Thus, this pathway is extremely sensitive to a decrease in cellular GTP content. Complete inhibition was also observed in islets treated with MPA plus adenine, to maintain ATP levels, under which conditions GTP is selectively depleted. Provision of guanine, which increases the activity of a salvage pathway for GTP synthesis and normalizes GTP content, completely reversed the inhibitory effect of MPA. Neither glucose utilization nor glucose oxidation was affected by MPA. The augmentation of Ca2+-independent insulin release by several other metabolizable nutrients including alpha-ketoisocaproic acid (KIC) was also inhibited by MPA. In sharp contrast, augmentation of Ca2+-dependent insulin release by KIC was resistant to GTP depletion, indicating that nutrient-induced augmentation of the Ca2+-dependent- and Ca2+-independent secretory pathways can be differentiated by GTP dependency. We interpret these data in accord with current knowledge concerning the two known stimuli for exocytosis, Ca2+ and GTP (independently of Ca2+). We propose that both Ca2+-dependent and Ca2+-independent augmentation occurs via one metabolic pathway acting upon Ca2+- and upon GTP-stimulated exocytosis. Activation of PKA and PKC stimulates the GTP-sensitive exocytosis.

Control of the structural stability of the tubulin dimer by one high affinity bound magnesium ion at nucleotide N-site

Tubulin liganded with GTP at the N-site in the alpha-subunit and with GDP at the E-site in the beta-subunit (GDP-tubulin) reversibly binds one high affinity Mg2+ cation (Kb = 1.1 x 10(7) M-1), whereas tubulin liganded with GTP at both subunits (GTP-tubulin) binds one more high affinity Mg2+. The two cation binding loci are identified as nucleotide sites N and E, respectively. Mg2+ at the N-site controls the stability and structure of the alphabeta-tubulin dimer. Mg2+ dissociation is followed by the slow release of bound nucleotide and functional inactivation. Mg2+ bound to the N-site significantly increases the thermal stability of the GDP-tubulin dimer (by 10 degrees C and approximately 50 kcal mol-1 of experimental enthalpy change). However, the thermal stability of Mg2+-liganded GDP- and GTP-tubulin is the same. Mg2+ binding to the N-site is linked to the alphabeta-dimer formation. The binding of Mg2+ to the alpha-subunit communicates a marked enhancement of fluorescence to a colchicine analogue bound to the beta-subunit. Colchicine, in turn, thermally stabilizes Mg2+-depleted tubulin. The tubulin properties described would be simply explained if the N-site and the colchicine site are at the alpha-beta dimerization interface. It follows that the E-site would be at the beta-end of the tubulin dimer, consistent with the known functional role of the E nucleotide gamma-phosphate and coordinated cation controlling microtubule stability.

Interaction of tubulin with guanosine 5'-O-(1-thiotriphosphate) diastereoisomers: specificity of the alpha-phosphate binding region

The exchangeable nucleotide-binding site of tubulin has been studied using diastereoisomers A (Sp) and B (Rp) of guanosine 5'-O-(1-thiotriphosphate) (GTP alpha S) in which the phosphorus atom to which sulfur is attached is chiral. GTP alpha S(A) (10 microM) nucleated assembly of purified tubulin (20 microM) into microtubules in buffer containing 0.1 M 2-(N-morpholino)ethanesulfonic acid with 3 mM Mg2+ and 1 mM EGTA, pH 6.6 at 37 degrees C. With 0.2 mM GTP alpha S(A), the critical concentration (Cc; minimum protein concentration required for assembly) was 8 microM tubulin. Neither 0.2 mM GTP nor GTP alpha S(B) promoted microtubule assembly in buffer with 0.5-6.75 mM Mg2+ and 20-70 microM tubulin. The Cc values for GTP alpha S-(A)-induced assembly of tubulin in buffer with 30% glycerol and of microtubule protein (tubulin and microtubule-associated proteins) in buffer were lower than for GTP. GTP alpha S(A)-induced microtubules were more stable to the cold and to Ca2+. GTP alpha S(A) and GTP but not GTP alpha S(B) bound tightly to tubulin at 4 degrees C. Although GTP alpha S(B) did not nucleate assembly, it did bind to tubulin since it was incorporated into the growing microtubule. Both isomers were hydrolyzed in the microtubules. These studies show that GTP alpha S(A) promotes tubulin assembly better than GTP and GTP alpha S(B) and that there is stereoselectivity at the alpha-phosphate binding region of tubulin. The stereoselectivity may be due to different MgGTP alpha S(A) and -(B) interactions with tubulin.

Thermodynamics of tubulin polymerization into zinc sheets: assembly is not regulated by GTP hydrolysis

The thermodynamics of tubulin assembly into Zn sheets have been studied, with special emphasis on the role of bound nucleotide and of GTP hydrolysis in polymerization. In contrast to microtubules, Zn sheets could be assembled from GDP-tubulin as well as from GTP-tubulin. Accordingly, no appreciable destabilization of the Zn sheets was observed following GTP hydrolysis and P(i) release, indicating that the binding of Zn2+ to tubulin has abolished the regulatory switch role played by GTP hydrolysis in tubulin assembly. As a consequence, the critical concentration for assembly of Zn sheets did not increase with tubulin concentration, a feature characteristic of microtubule assembly. Zn sheets do not bind P(i) analogs, indicating that the gamma-phosphate binding locus of the E-site of tubulin is occluded following GTP hydrolysis in these GDP-tubulin polymers. Nonlinear van't Hoff plots were obtained for assembly of Zn sheets in the presence of either GTP or GDP, consistent with a change in heat capacity. Enthalpy, entropy, and heat capacity changes had values similar to those reported for assembly of microtubules or polymerization of tubulin-colchicine, indicating that hydrophobic tubulin-tubulin interactions are of comparable size in these different polymers.

Should the tubulins be members of the GTPase superfamily?

The beta-subunit of the alpha/beta tubulin heterodimer resembles other members of the GTPase superfamily in that: it binds GTP, the GTP is hydrolysed to GDP on microtubule assembly and this induces a conformational change; it exhibits a similar nucleotide stereospecificity; aluminium and beryllium fluorides inhibit this hydrolysis-dependent conformational change; and beta-tubulin contains peptides which are similar to the consensus motifs characteristic of the GTPase superfamily proteins. By contrast, UV photo-cross-linking and other binding studies have identified peptides which may contribute to the GTP-binding site but which are absent from the GTPase superfamily proteins. We suggest that beta-tubulin has a 'dual personality', with the characteristics of the GTP-binding site depending upon the precise conformation of the protein and upon whether the experimental assays probe nucleotide binding or the hydrolytic mechanism. We suggest that the hydrolytic mechanism of beta-tubulin resembles that of the other members of the GTPase superfamily, although the differences within the consensus motifs dictate that the architecture of the GTP pocket cannot be identical.