Exploring the Protein Funnel Energy Landscape for Folding

Exploring the Protein Funnel Energy Landscape for Folding and Function José N. Onuchic Workshop III - Structural Proteomics IPAM - UCLA Los Angeles, May 2004 Center for Theoretical Biological Physics UCSD ctbp. ucsd. edu

Energy Landscape Idea • Rough Landscape Random Heteropolymer Frustration!! Try to minimize frustration! Tf Stability Tg Roughness Stability Gap • Proteins Principle of Minimal Frustration (Bryngelson/Wolynes, Go) Realization of Minimal Frustration in Funnels

Go-model Tf/Tg=1. 6 Tf/Tg=1. 3 H. Nymeyer, N. Socci and J. Onuchic, PNAS 2000

C. Clementi, H. Nymeyer, J. Onuchic, JMB 2000 Go-like potential = only native interactions are taken into account (named after a paper by Taketomi, Ueda, Go 1975) 2 2

Analysis of two-state folders: Transition State structure for CI 2 and SH 3 Probability of contact formation at TS distal loop SH 3 Diverging turn CI 2 These descriptions are in good agreement with experimental results (Jackson & Fersht 1991, Grantcharova et al. 1998). C. Clementi, H. Nymeyer, J. Onuchic, JMB 2000

Analysis of proteins which fold through the formation of an intermediate: results for Barnase, Rnase. H, Che. Y Barnase Rnase H Che. Y A local minimum in the free energy profile between the unfolded and folded minima locates a folding intermediate state. C. Clementi, H. Nymeyer, J. Onuchic, JMB 2000

Che. Y In agreement with experimental results (Lopez-Hernandez & Serrano, 1996) we found that in the ``misfolded" structure, all the five a-helices are rather structured, while in the later occurring transition state ensemble, the helices 4 -5 are more unstructured.

How about water effects? r* Desolvation models: Hummer, G. , et al, PNAS 1997 M. Cheung, A. Garcia and J. Onuchic, PNAS 2002

Tf M. Cheung, A. Garcia and J. Onuchic, PNAS 2002

What are the folding routes for SH 3?

Why this kind of mutation? Alanine Q Valine Val la A Multiple -value analysis Energy Landscape Theory Pse Va udo l with Luis Serrano Q Th r Threonine

A kinetic analysis on SH 3 mutants using LJ and Desolvation potentials Using a desolvation potential, V 44 T and V 53 T show dramatic changes in folding time! 9 46 44 58 53 23

Although V 44 T and V 53 T show a dramatic decrease in folding rates, the nature of their kinetic traps are different. By going through alternate rounds of theoretical analysis and experimental development, we are revealing the mechanism of protein folding progressively!

RG Q J. Shea, J. Onuchic and C. Brooks III, 2002

Protein A - Distribution of F Values N(F) Helix I-III Turn/Helix I-II Turn II-III J. Shea, J. Onuchic and C. Brooks, PNAS 1999

Rmsd(A) vs. Native contacts: T~ T* Folded state divided into two basins: Folded solvated Folded desolvated with Angel Garcia

Replica Exchange Sample Trajectories with Angel Garcia

The Nature of the landscape Backbone Hydration G (minimal path) 5. 5 Free energy 5 4 3. 5 0 Helix formation 0. 2 0. 4 0. 6 0. 8 1 H (minimal path) 10 Enthalpy 5 0 -5 - 10 0 T=Tf 0. 2 0. 4 0. 6 0. 8 Q Energies in Kcal/mol

Collaborators Postdocs: Osamu Miyashita Koby Levy Antitsa Stoycheva John Finke Yoko Suzuki Shachi Gosavi Cecilia Clementi – Rice Joan-Emma Shea – UCSB Steve Plotkin – UBC Nick Socci – Albert Einstein Other Research Groups Peter Wolynes – UCSD Angel Garcia - LANL Pat Jennings – UCSD Charles Brooks – TSRI Zan Luthey-Schulten – UIUC Vitor Leite and Jorge Chahine - Brazil Yoshitaka Tanimura - Kyoto, Japan Yuko Okamoto – IMS, Japan Ulrich Hansmann – Michigan Tech Chigusa Kobayashi - Kobe, Japan Eric Nelson – Texas Students Leslie Chaves Sichun Yang Margaret Cheung - Maryland Hugh Nymeyer –LANL Peter Leopold – Funnel Idea -1992 Supported by the National Science Foundation and the Burroughs Wellcome Fund