The Skaggs Institute
for Chemical Biology

Scientific Report 2008

Studies of Macromolecular Recognition by Multidimensional Nuclear Magnetic Resonance

P.E. Wright, H.J. Dyson, M. Martinez-Yamout, M. Arai, S.-H. Bae, D. Boehr, B. Buck-Koehntop, P. Deka, D. Felitsky, J. Ferreon, P. Haberz, C.W. Lee, D. Meinhold, S.-J. Park, M. Landes, E. Manlapaz

Specific interactions between molecules are of fundamental importance in all biological processes. An understanding of how biological macromolecules such as proteins and nucleic acids recognize each other is essential for understanding the fundamental molecular events of life. Knowledge of the 3-dimensional structures of biological macromolecules is key to understanding their interactions and functions and also forms the basis for rational design of new drugs. A particularly powerful method for mapping the 3-dimensional structures and interactions of biological macromolecules in solution is multidimensional nuclear magnetic resonance (NMR) spectroscopy. We are using this method to study a number of protein-protein and protein–nucleic acid interactions of fundamental importance in health and disease.

Transcriptional regulation in eukaryotes relies on protein-protein interactions between DNA-bound factors and coactivators that, in turn, interact with the basal transcription machinery. A major effort in our laboratory is focused on elucidating the structural principles that determine specificity of key protein-protein interactions involved in regulation of gene expression. The transcriptional coactivator CREB-binding protein (CBP) and its ortholog p300 play a central role in cell growth, differentiation, and development in higher eukaryotes. CBP and p300 mediate interactions between a number of gene regulatory proteins and viral proteins, including proteins from several tumor viruses and hepatitis B virus. Understanding the molecular mechanisms by which CBP recognizes its various target proteins is of fundamental biomedical importance. CBP has been implicated in diverse human diseases such as leukemia, cancer, and mental retardation and is a novel target for therapeutic intervention.

We have initiated a major program to determine the structure of CBP and p300 and map their functional interactions with other components of the transcriptional machinery. Our research reveals that many regions of these coactivators are intrinsically disordered, as are many of the transcriptional regulatory proteins with which they interact. Indeed, our results have indicated that coupled folding and binding processes play a major role in transcriptional regulation.

We have performed NMR relaxation experiments to elucidate the mechanism of coupled folding and binding processes and to identify "hot spots" in protein-protein interfaces that could potentially be targeted by small-molecule inhibitors. We initially used these methods to investigate the interactions involved in the regulation of hypoxia, namely binding of the α-subunit of the hypoxia-inducible transcription factor (HIF-1 α) to the TAZ1 zinc finger motif of CBP/p300. We have now extended these relaxation measurements to the complex formed between the activation domain of the p160 nuclear receptor coactivator ACTR and the nuclear coactivator binding domain of CBP. Both proteins are intrinsically disordered and fold synergistically upon binding. Although the free proteins are highly flexible, the complex has the motional characteristics of a globular protein domain, with no significant residual flexibility that might compensate for the loss of entropy incurred upon formation of a complex.

Some years ago, we determined the 3-dimensional structure of the phosphorylated kinase inducible activation domain (pKID) of the transcription factor CREB bound to its target domain (the KIX domain) in CBP. The structure provides a starting point for design of small molecules that can inhibit the CREB-KIX interactions, an important goal in development of novel therapeutics for treatment of diabetes. We have developed a new method, using R₂ relaxation dispersion experiments and NMR titrations, to investigate the pathway by which intrinsically disordered proteins fold into ordered structures upon binding to their biological targets. We have used this method to study the mechanism of pKID binding to KIX.

The pKID first forms an ensemble of transient encounter complexes at multiple sites on the surface of KIX and then folds via a pathway involving a partially structured intermediate. Folding of the pKID helices occurs on the surface of KIX; the mechanism of recognition involves an induced protein folding event, rather than selection of a small population of prefolded helical structures from the solution conformational ensemble.

We have also used the method to study mechanisms of binding of the hydroxylated HIF-1 α transactivation domain to the TAZ1 domain of CBP and have commenced studies of the binding of the proto-oncogene cMyb to the KIX domain of CBP. This research is leading to a new understanding of the molecular mechanisms by which intrinsically disordered proteins perform their diverse biological functions. In the course of these studies, we have developed novel methods for measuring the affinities with which intrinsically disordered proteins bind to their targets (Fig. 1).

Fig. 1. Global fit of chemical-shift titration data to obtain accurate dissociation constants.

CBP and p300 contain several zinc-binding domains (ZZ domain, PHD motif, TAZ1 and TAZ2 domains) that mediate critical interactions with numerous transcriptional regulators. We have determined the structures of each of these domains during recent years. Our current efforts are focused on structural analysis of the complexes formed between the TAZ1 and TAZ2 domains and the activation domains of the numerous transcription factors with which the TAZ1 and TAZ2 domains interact. We have determined the structures of the complexes formed between the TAZ domains and the activation domains of the signal transducer and activator of transcription (STAT) family of transcriptional regulators. These interactions play a key role in cytokine-dependent signal transduction. Structures have been determined for the complex of TAZ1 with the STAT2 activation domain and for TAZ2 bound to STAT1 (Fig. 2). The STAT1 and STAT2 activation domains are intrinsically disordered and fold upon binding to the TAZ motifs, burying a large surface area and forming a hydrophobic intermolecular core. The different structural features of the TAZ1 and TAZ2 scaffolds dictate the conformation and sites of binding of the STAT2 and STAT1 motifs.

Fig. 2. Structures of the TAZ1-STAT2 complex (A) and the TAZ2-STAT1 complex (B). The protein backbones of the STAT activation domains are shown as pink ribbons; the backbones of the TAZ1 and TAZ2 domains, as blue and green ribbons, respectively.

CBP and p300 play a critical role in the regulation of the tumor suppressor p53. They interact directly with p53 and are required for p53-mediated transcriptional activation. They also function to regulate p53 stability. We have used NMR spectroscopy and isothermal titration calorimetry to investigate the binding interactions between the transcriptional activation domain of p53 and its target domains in CBP/p300. We found that the p53 activation domain can bind simultaneously to CBP/p300 and the ubiquitin ligase HDM2, which regulates p53 stability, to form a ternary complex. Phosphorylation of the p53 activation domain inhibits binding of HDM2 and enhances binding to CBP/p300, thereby stabilizing p53 and activating transcription of p53-regulated genes. Our findings provide novel insights into the mechanism of p53 regulation in response to DNA damage and genotoxic stress. In addition, we have determined the structures of the complexes formed between the KIX domain of CBP and the p53 activation domain and between the TAZ2 domain of CBP and the adenoviral oncoprotein E1A.

Finally, we have made major advances in understanding the mechanism by which the zinc finger protein muscleblind recognizes both pathogenic double-stranded repeat RNA sequences and single-stranded regulatory RNA elements. Sequestration of muscleblind by CUG- and CCUG-repeat RNA disrupts alternate RNA splicing and is the underlying molecular cause of myotonic dystrophy, the most common form of adult-onset muscular dystrophy. We have determined the structure of the first 2 zinc fingers of muscleblind, which fold into a unique globular structure (Fig. 3), and we have mapped their interactions with single-stranded RNA. We have identified the specific RNA sequence required for high-affinity binding and are currently working on the structure of the RNA complex.

Fig. 3. Ribbon representation of the structure of muscleblind zinc fingers.

Publications

Boehr, D.D., Dyson, H.J., Wright, P.E. Conformational relaxation following hydride transfer plays a limiting role in dihydrofolate reductase catalysis. Biochemistry 47:9227, 2008.

Boehr, D.D., Wright, P.E. How do proteins interact? Science 320:1429, 2008.

Ebert, M.-O., Bae, S.-H. Dyson, H.J., Wright, P.E. NMR relaxation study of the complex formed between CBP and the activation domain of the nuclear hormone receptor coactivator ACTR. Biochemistry 47:1299, 2008.

Felitsky, D.J., Lietzow, M.A., Dyson, H.J., Wright, P.E. Modeling transient collapsed states of an unfolded protein to provide insights into early folding events. Proc. Natl Acad. Sci. U. S. A. 105:6278, 2008.

Sugase, K., Landes, M.A., Wright, P.E., Martinez-Yamout, M.A. Overexpression of post-translationally modified peptides in Escherichia coli by co-expression with modifying enzymes. Protein Expr. Purif. 57:108, 2008.

Sugase, K., Lansing, J.C., Dyson, H.J., Wright, P.E. Tailoring relaxation dispersion experiments for fast-associating protein complexes. J. Am. Chem. Soc. 129:13406, 2007.