HHMI-NIH Research Scholars
Learn about the HHMI-NIH Research Scholars Program, also known as the Cloister Program. More

Janelia Farm Research Campus
Learn about the new HHMI research campus located in Virginia. More

Summary: David Agard's research is focused on elucidating the mechanism of microtubule nucleation, Hsp90 chaperone function, and the role of dynamics in enzyme function and folding.

Mechanisms of Microtubule Formation and the Role of γ-Tubulin Complexes as Nucleators
The spatial and temporal regulation of tubulin polymerization into microtubules (MTs) is a central question in cell biology. Our goal is to understand, in atomic detail, the molecular mechanisms underlying dynamic MT behavior and MT nucleation.

It is well established that GTP binding and hydrolysis play key roles in MT dynamics. The dominant model for the GTP-dependent assembly proposes that GDP-bound αβ-tubulin adopts an MT-incompatible curved conformation and that GTP binding to β-tubulin allosterically converts both subunits into a straight, MT-compatible conformation.

	Cellular proteins: folding and function... Hsp90 and protein folding... Simulation of alpha-lytic protease... Binding mode prediction for a flexible ligand...

As part of our efforts to understand the molecular basis of MT nucleation, we solved the atomic structures of human γ-tubulin complexed with GTP and GDP, providing the first GTP/GDP pair and the highest resolution for any tubulin. These structures provided two key insights: γ-tubulin forms MT-like lateral interactions independent of nucleotide, and γ-tubulin remains in a curved conformation independent of nucleotide, contrasting sharply with the prevailing allosteric hypothesis. Solution studies (conformation-specific ligand binding and small-angle x-ray scattering) on αβ-tubulin confirm that it too remains in a curved conformation independent of nucleotide.

As a major paradigm shift, we propose that the lattice and not the nucleotide is the allosteric effector. Instead, GTP plays a secondary role to tune the longitudinal affinity. Lattice metastability is determined not by GTP hydrolysis but by the mechanical spring constant for straightening. This new view has a dramatic impact on understanding MT formation.

By analogy with actin, MT assembly has been interpreted as a nucleation-polymerization mechanism where an unstable nucleus is formed during the lag phase. Disturbed by features of our kinetic data, implying that nonintegral numbers of tubulins form the nucleus in a nonintegral number of steps, we are developing a physically based quantitative model. Our modeling indicates that MT assembly is not a nucleation-polymerization mechanism, but instead resembles two-dimensional crystallization, and that polymer growth does not happen until thousands of tubulins have assembled.

Of key importance is the structural and functional analysis of γ-tubulin complexes, which act in vivo to nucleate MT growth. Although central to all MT nucleation, the γ-tubulin small-complex (γTuSC) is a surprisingly poor MT nucleator. We have determined the structure of the yeast γTuSC by single-particle EM (electron microscopy). A combination of in vivo FRET (fluorescence resonance energy transfer), gold labeling, and localizing YFP (yellow fluorescent protein)-tagged subunits has allowed us to assign the locations and orientations of all components within the complex. Remarkably, the two γ-tubulin heads are significantly separated, resulting in an MT-incompatible configuration and explaining the poor nucleating potential of γTuSC. Movement of a mobile arm is required to bring the γ-tubulins together, providing a template for MT growth.

Mechanism of Hsp90 Function
In eukaryotes, the ubiquitous Hsp90 molecular chaperone facilitates the folding and activation of a broad array of proteins important in cell signaling, proliferation, and survival. Unlike other molecular chaperones, Hsp90 preferentially stabilizes near-native-state structures, aiding the dynamic assembly and disassembly of signaling complexes. Hsp90 is thus an important therapeutic target. Our goal is to understand Hsp90 action and the structural basis for its requirement by substrate proteins.

Recently, we solved the first structure of an Hsp90 C-terminal dimerization domain, revealing an unexpected potential "client" protein-binding site, and subsequently solved the full-length Escherichia coli Hsp90 in both the apo and ADP states. Along with a concurrent structure of the ATP state from Laurence Pearl's lab (Institute of Cancer Research, London), our work profoundly affected models of the conformational cycle, suggesting where client proteins may bind and how nucleotide binding and hydrolysis propel the chaperone through conformational changes that lead to the release of client proteins.

Using small-angle x-ray scattering and newly developed molecular modeling methods, we determined the solution structure of apo HtpG and found that it is more extended than the crystal form. In addition to this novel conformation, we discovered that under physiologically relevant conditions, multiple conformations coexist in equilibrium: there is both a pH-dependent equilibrium (pK_a ~ 7.3) and an equilibrium with the ATP state. EM single-particle data indicate that the open-closed equilibrium is species specific. These remarkable conformational dynamics have important implications for the mechanism of Hsp90 action. First, they place limitations on the amount of energy available from ATP binding that could be utilized to drive substrate activation. Second, they suggest that ATP hydrolysis and subsequent release of ADP or phosphate could be the energetically most important step, much as in myosin, and could be coupled to client release.

The Function and Evolution of Kinetic Stability: A Study of α-Lytic Protease Folding
The central dogma in protein folding is that the native state of a protein is at the global free-energy minimum. This allows spontaneous folding to the active conformation. The family of extracellular bacterial proteases, typified by α-lytic protease (αLP), provides striking counterexamples in which the native state is substantially less stable than the fully unfolded molecule. Instead of being thermodynamically stabilized, αLP is trapped in its active conformation by a large energy barrier that effectively blocks unfolding (t_1/2 = 1.2 years). A further consequence is that the barrier to folding is even larger (t_1/2 = 1,800 years). αLP is synthesized with a 166-residue N-terminal pro region that solves the folding problem both by accelerating folding and by binding tightly to the native state to shift the thermodynamic equilibrium in favor of the active enzyme. Once the protease is folded, the pro region is destroyed by proteolysis. Our work suggests that this unusual folding mechanism provides an optimal solution to making proteins that are themselves resistant to proteolysis. Decreasing proteolysis requires that both the barrier height and the cooperativity be increased. Thus a protein's folding pathway can have a profound impact on the properties of the native state and not just dictate how that state is reached.

A combination of ultra-high-resolution x-ray crystallography (0.83 Å) and comparative structural and functional analysis with family members that survive extremes of pH and temperature is providing insights into the structure of the unfolding transition state. These studies indicate that a Phe in the αLP C-terminal domain is substantially bent and tie this directly to the barrier height. Other architectural features and electrostatic interactions also play key roles in determining the energetic landscape. These results have broad implications for the evolution of extremophiles.

Funding for the centrosome work was provided by the National Institutes of Health, a UC Discovery Grant contributed to the Hsp90 studies, and funds from the W.M. Keck Foundation contributed to our development efforts in light and electron microscopy.

Last updated: March 3, 2008