YB-1, an abundant core mRNA-binding protein, has the capacity to form an RNA nucleoprotein filament: a structural analysis.

The structural rearrangements accompanying mRNA during translation in mammalian cells remain poorly understood. Here, we discovered that YB-1 (YBX1), a major partner of mRNAs in the cytoplasm, forms a linear nucleoprotein filament with mRNA, when part of the YB-1 unstructured C-terminus has been truncated. YB-1 possesses a cold-shock domain (CSD), a remnant of bacterial cold shock proteins that have the ability to stimulate translation under the low temperatures through an RNA chaperone activity. The structure of the nucleoprotein filament indicates that the CSD of YB-1 preserved its chaperone activity also in eukaryotes and shows that mRNA is channeled between consecutive CSDs. The energy benefit needed for the formation of stable nucleoprotein filament relies on an electrostatic zipper mediated by positively charged amino acid residues in the YB-1 C-terminus. Thus, YB-1 displays a structural plasticity to unfold structured mRNAs into extended linear filaments. We anticipate that our findings will shed the light on the scanning of mRNAs by ribosomes during the initiation and elongation steps of mRNA translation.

The C terminus of tubulin, a versatile partner for cationic molecules: binding of Tau, polyamines, and calcium.

The C-terminal region of tubulin is involved in multiple aspects of the regulation of microtubule assembly. To elucidate the molecular mechanisms of this regulation, we study here, using different approaches, the interaction of Tau, spermine, and calcium, three representative partners of the tubulin C-terminal region, with a peptide composed of the last 42 residues of α1a-tubulin. The results show that their binding involves overlapping amino acid stretches in the C-terminal tubulin region: amino acid residues 421-441 for Tau, 430-432 and 444-451 for spermine, and 421-443 for calcium. Isothermal titration calorimetry, NMR, and cosedimentation experiments show that Tau and spermine have similar micromolar binding affinities, whereas their binding stoichiometry differs (C-terminal tubulin peptide/spermine stoichiometry 1:2, and C-terminal tubulin peptide/Tau stoichiometry 8:1). Interestingly, calcium, known as a negative regulator of microtubule assembly, can compete with the binding of Tau and spermine with the C-terminal domain of tubulin and with the positive effect of these two partners on microtubule assembly in vitro. This observation opens up the possibility that calcium may participate in the regulation of microtubule assembly in vivo through direct (still unknown) or indirect mechanism (displacement of microtubule partners). The functional importance of this part of tubulin was also underlined by the observation that an α-tubulin mutant deleted from the last 23 amino acid residues does not incorporate properly into the microtubule network of HeLa cells. Together, these results provide a structural basis for a better understanding of the complex interactions and putative competition of tubulin cationic partners with the C-terminal region of tubulin.

The state of the guanosine nucleotide allosterically affects the interfaces of tubulin in protofilament.

The dynamics of microtubules is essential for many microtubule-dependent cellular functions such as the mitosis. It has been recognized for a long time that GTP hydrolysis in αß-tubulin polymers plays a critical role in this dynamics. However, the effects of the changes in the nature of the guanosine nucleotide at the E-site in ß-tubulin on microtubule structure and stability are still not well understood. In the present work, we performed all-atom molecular dynamics simulations of a αßα-tubulin heterotrimer harboring a guanosine nucleotide in three different states at the E-site: GTP, GDP-Pi and GDP. We found that changes in the nucleotide state is associated with significant conformational variations at the α-tubulin N- and ß-tubulin M-loops which impact the interactions between tubulin protofilaments. The results also show that GTP hydrolysis reduces αß-tubulin interdimer contacts in favor of intradimer interface. From an atomistic point view, we propose a role for α-tubulin glutamate residue 254 in catalytic magnesium coordination and identified a water molecule in the nucleotide binding pocket which is most probably required for nucleotide hydrolysis. Finally, the results are discussed with reference to the role of taxol in microtubule stability and the recent tubulin-sT2R crystal structures.

The stathmin-derived I19L peptide interacts with FtsZ and alters its bundling.

FtsZ is a prokaryotic tubulin-like protein. Despite a low degree of sequence identity with tubulin, it presents the same folding pattern and some similar functions, notably in cell division. Indeed, FtsZ and tubulin polymerize to form bundles and microtubules, respectively, which are essential for cell cytokinesis. We previously demonstrated that peptides derived from the N-terminal stathmin domain interact with tubulin and impede microtubule formation. We demonstrated here that I19L, the most efficient of these peptides, also alters FtsZ bundling assembly in vitro. STD-NMR and TRNOESY experiments revealed that I19L interacts with FtsZ and folds upon its binding but in a way different from what we observed with tubulin. These NMR data were used in molecular modeling calculations to propose models of the I19L-FtsZ complex. Interestingly, two models, consistent with NMR data, show an interaction of I19L near the T7 loop or near the GTP binding site of FtsZ, explaining the modifications of the bundling assembly observed with this peptide. The fine analysis of the structural differences of the complexes of I19L with FtsZ and tubulin should help for the rational development of new specific antibiotic agents.

Both lipid environment and pH are critical for determining physiological solution structure of 3-D-conserved epitopes of the HIV-1 gp41-MPER peptide P1.

In terms of background, the solution structure of monomeric peptide P1 (residues 649-683), located in the conserved membrane proximal region (MPER) of HIV-1 envelope glycoprotein gp41, is first reported here in dodecylphosphocholine (DPC) micelles. P1 is the minimal MPER region that permits interaction with the mucosal galactosyl ceramide HIV-receptor; it also contains epitopes recognized by major gp41-specific, broadly neutralizing immunoglobulin Gs (IgGs), 2F5 and 4E10, determinant in HIV fusion/infection. Our principal findings were as follows: the structural stability of P1 is pH dependent, as the alpha-helix comprising Q653 I682, present at pH 3.3, is destabilized at higher pH values. At pH 6, the E-rich N-terminal half of P1 (residues 650-666), partially overlapping the 2F5-specific epitope, becomes fully disordered, while the W-rich C-terminal half conserves two shorter helices (W666-W670 and W672-W680), separated by a well-defined bend overlapped by the 4E10-specific epitope. The two IgGs bind to P1 on DPC micelles with binding parameters (K(d)) in the nanomolar range. Next, P1 was derivatized with phosphatidylethanolamine at its C terminal and inserted into liposomes of varied lipid composition, thereby enabling P1 to move laterally. Alternatively, an infectious virus-binding assay was established. The K(d) of both 2F5 and 4E10 IgGs measured on viral liposome and virus are similar and much lower than for the binding of the free peptide. In conclusion, P1, in a lipid environment, is an optimized MPER-derived peptide suitable for designing an immunogen inducing broadly neutralizing antibodies to HIV.

Benomyl and colchicine synergistically inhibit cell proliferation and mitosis: evidence of distinct binding sites for these agents in tubulin.

Benomyl, a tubulin-targeted antimitotic antifungal agent, belongs to the benzimidazole group of compounds, which are known to inhibit the binding of colchicine to tubulin. Therefore, benomyl was thought to bind at or near the colchicine-binding site on tubulin. However, recent mutational studies in yeast and fluorescence studies involving competitive binding of benomyl and colchicine on goat brain tubulin suggested that benomyl may bind to tubulin at a site distinct from the colchicine-binding site. We set out to examine whether colchicine and benomyl bind to tubulin at distinct sites using a human cervical cancer (HeLa) cell line with the thinking that these agents should exert either additive or synergistic activity on cell proliferation if their binding sites on tubulin are different. We found that benomyl and colchicine synergistically inhibited the proliferation of HeLa cells and blocked their cell cycle progression at mitosis. The synergistic activity of benomyl and colchicine was also apparent from their strong depolymerizing effects on both the spindle and interphase microtubules when used in combinations, providing further evidence that these agents bind to tubulin at different sites. Using NMR spectroscopy, we finally demonstrated that benomyl and colchicine bind to tubulin at different sites and that the binding of colchicine seems to positively influence the binding of benomyl to tubulin and vice versa. Further, an analysis of the saturation transfer difference NMR data yielded an interesting insight into the colchicine-tubulin interaction. The data presented in this study provided a mechanistic understanding of the synergistic effects of benomyl and colchicine on HeLa cell proliferation.

YB-1 promotes microtubule assembly in vitro through interaction with tubulin and microtubules.

BACKGROUND: YB-1 is a major regulator of gene expression in eukaryotic cells. In addition to its role in transcription, YB-1 plays a key role in translation and stabilization of mRNAs. RESULTS: We show here that YB-1 interacts with tubulin and microtubules and stimulates microtubule assembly in vitro. High resolution imaging via electron and atomic force microscopy revealed that microtubules assembled in the presence of YB-1 exhibited a normal single wall ultrastructure and indicated that YB-1 most probably coats the outer microtubule wall. Furthermore, we found that YB-1 also promotes the assembly of MAPs-tubulin and subtilisin-treated tubulin. Finally, we demonstrated that tubulin interferes with RNA:YB-1 complexes. CONCLUSION: These results suggest that YB-1 may regulate microtubule assembly in vivo and that its interaction with tubulin may contribute to the control of mRNA translation.

Cucurbitacin I elicits the formation of actin/phospho-myosin II co-aggregates by stimulation of the RhoA/ROCK pathway and inhibition of LIM-kinase.

Cucurbitacins are cytotoxic triterpenoid sterols isolated from plants. One of their earliest cellular effect is the aggregation of actin associated with blockage of cell migration and division that eventually lead to apoptosis. We unravel here that cucurbitacin I actually induces the co-aggregation of actin with phospho-myosin II. This co-aggregation most probably results from the stimulation of the Rho/ROCK pathway and the direct inhibition of the LIMKinase. We further provide data that suggest that the formation of these co-aggregates is independent of a putative pro-oxidant status of cucurbitacin I. The results help to understand the impact of cucurbitacins on signal transduction and actin dynamics and open novel perspectives to use it as drug candidates for cancer research.

Probing interactions of tubulin with small molecules, peptides, and protein fragments by solution nuclear magnetic resonance.

The description of the molecular mechanisms of interaction between tubulin or microtubules and partners at atomic scale is expected to have critical impacts on the understanding of basic physiological processes. This information will also help the design of future drug candidates that may be used to fight various pathologies such as cancer or neurological diseases. For these reasons, this aspect of tubulin research has been tackled since the seventies using many different methods and at different scales. NMR appears as a unique approach to provide, with atomic resolution, the solution structure and dynamical properties of tubulin/microtubule partners in free and bound states. Though tubulin is not directly amenable to solution NMR, the NMR ligand-based experiments allow one to obtain valuable data on the molecular mechanisms that sustain structure-function relationship, in particular atomic details on the partner binding site. We will first describe herein some basic principles of solution NMR spectroscopy that should not be missed for a comprehensive reading of NMR reports. A series of results will then be presented to illustrate the wealth and variety of NMR experiments and how this approach enlightens tubulin/microtubules interaction with partners.

The PN2-3 domain of centrosomal P4.1-associated protein implements a novel mechanism for tubulin sequestration.

Microtubules are cytoskeletal components involved in multiple cell functions such as mitosis, motility, or intracellular traffic. In vivo, these polymers made of alphabeta-tubulin nucleate mostly from the centrosome to establish the interphasic microtubule network or, during mitosis, the mitotic spindle. Centrosomal P4.1-associated protein (CPAP; also named CENPJ) is a centrosomal protein involved in the assembly of centrioles and important for the centrosome function. This protein contains a microtubule-destabilizing region referred to as PN2-3. Here we decrypt the microtubule destabilization activity of PN2-3 at the molecular level and show that it results from the sequestration of tubulin by PN2-3 in a non-polymerizable 1:1 complex. We also map the tubulin/PN2-3 interaction both on the PN2-3 sequence and on the tubulin surface. NMR and CD data on free PN2-3 in solution show that this is an intrinsically unstructured protein that comprises a 23-amino acid residue alpha-helix. This helix is embedded in a 76-residue region that interacts strongly with tubulin. The interference of PN2-3 with well characterized tubulin properties, namely GTPase activity, nucleotide exchange, vinblastine-induced self-assembly, and stathmin family protein binding, highlights the beta subunit surface located at the intermolecular longitudinal interface when tubulin is embedded in a microtubule as a tubulin/PN2-3 interaction area. These findings characterize the PN2-3 fragment of CPAP as a protein with an unprecedented tubulin sequestering mechanism distinct from that of stathmin family proteins.

NMR assignment of PN2-3, a tubulin interaction subdomain of the CPAP protein.

We report the NMR assignment of the PN2-3 subdomain of the CPAP protein. It has been previously shown that this motif interacts with tubulin, inhibits microtubule nucleation from the centrosome and depolymerizes taxol-stabilized microtubules.

NMR solution structure of neurotensin in membrane-mimetic environments: molecular basis for neurotensin receptor recognition.

Neurotensin (NT) is a 13-residue neuropeptide that exerts multiple biological functions in the central and peripheral nervous system. Little is known about the structure of this neuropeptide, and what is known only concerns its C-terminal part. We determined here for the first time the structure of the full-length NT in membrane-mimicking environments by means of classical proton-proton distance constraints derived from solution-state NMR spectroscopy. NT was found to have a structure at both its N and C termini, whereas the central region of NT remains highly flexible. In TFE and HFIP solutions, the NT C-terminus presents an extended slightly incurved structure, whereas in DPC it has a beta turn. The N-terminal region of NT possesses great adaptability and accessibility to the microenvironment in the three media studied. Altogether, our work demonstrates a structure of NT fully compatible with its NTR-bound state.

N-terminal stathmin-like peptides bind tubulin and impede microtubule assembly.

Microtubules are major cytoskeletal components involved in numerous cellular functions such as mitosis, cell motility, or intracellular traffic. These cylindrical polymers of alphabeta-tubulin assemble in a closely regulated dynamic manner. We have shown that the stathmin family proteins sequester tubulin in a nonpolymerizable ternary complex, through their stathmin-like domains (SLD) and thus contribute to the regulation of microtubule dynamics. We demonstrate here that short peptides derived from the N-terminal part of SLDs impede tubulin polymerization with various efficiencies and that phosphorylation of the most potent of these peptides reduces its efficiency as in full-length stathmin. To understand the mechanism of action of these peptides, we undertook a NMR-based structural analysis of the peptide-tubulin interaction with the most efficient peptide (I19L). Our results show that, while disordered when free in solution, I19L folds into a beta-hairpin upon binding to tubulin. We further identified, by means of saturation transfer difference NMR, hydrophobic residues located on the beta2-strand of I19L that are involved in its tubulin binding. These structural data were used together with tubulin atomic coordinates from the tubulin/RB3-SLD crystal structure to model the I19L/tubulin interaction. The model agrees with I19L acting through an autonomous tubulin capping capability to impede tubulin polymerization and provides information to help understand the variation of efficiency against tubulin polymerization among the peptides tested. Altogether these results enlighten the mechanism of tubulin sequestration by SLDs, while they pave the way for the development of protein-based compounds aimed at interfering with tubulin polymerization.