The Holy Grail - To show how a protein folds

Christian Anfinsen kicked it off with his work on Ribonuclease and by altering the chemistry of the di-sulphide bonds of the cysteine amino acid residues and addition of a denaturing agent like urea, he showed that the protein can unfold and refold back to its native configuration. What Professor Anfinsen has demonstrated was that the instruction for folding was encapsulated within the amino acid sequence.

Fast forward a few decades later, a voluminous amount of discovery was made as biochemists made inroads into understanding protein structures and elucidating their various functions. However, one thing still eludes scientists - to show fully how a protein with its primary structure will eventually arrive at its final functional conformation. Whilst the types of secondary structures formed, i.e. beta sheets, alpha helices, etc, are well-known, the process whereby the secondary structures combine and constitute the tertiary structure still remains a mystery.

Even at the level of the secondary structure, there is transition between beta sheet and alpha helices, especially in the case of amylogenic diseases (Kauffmann et al, 2001). Another group has shown the transition of poly-lysine from alpha helix to beta sheet structure at a certain temperature and that the longer polylysine chain, the lower the transition temperature (Dzwolak et al, 2004). Another group has shown the effect of pressure in inducing the transformation of alpha helices to beta sheets (Lin et al, 2002). The discovery of Prions, an infectious proteinacious particle, which netted Dr Stanley B. Prusiner the Nobel Prize for Physiology and Medicine, adds further complexity to the way we understand how proteins fold. The prion story is a little bit like Dr Jekyll and Mr Hyde. Our body has prion proteins and they are located at the surface of the cell. The good prions, PrPC, is correctly folded. The Mr Hydes, unfortunately are the misfolded prion proteins, and they are popularly denoted as PrPsc. PrPC can be misfolded and converted to PrPsc, in the presence of PrPsc. PrPsc is stable to heat and protease digestion (Safar et al, 1993). PrPC consists of mainly alpha helical secondary structures, whilst the misfolded PrPsc consists of mainly beta sheets. PrPsc is more energetically stable than PrPC (Eghiaian et al, 2004), i.e. PrPsc would be at a lower level in the energy funnel vis-a-vis PrPC.

What can we make out of Professor Christian Anfinsen's originial proposal that the instruction for protein folding is encapsulated within the amino acid sequence? The instruction for folding is contained within the amino acid sequence, BUT it is apparent that the folding process requires an interplay of not only the amino acid sequence BUT the external conditions like temperature and pressure, which have been instrumental in the transition of alpha helices to beta sheet, and also the presence of other proteins in the vicinity, as it was demonstrated in the example of prions that propagation of PrPsc can be achieved by misfolding of PrPC in the former's presence acting as a catalyst. Thus, if we consider the biophysical dynamics of protein folding, the final product could be the summation of amino acid sequence and conditions during foldings, and amino acid sequence sequence may not account for the end product alone.

At this current stage, Bioinformatics may be the tangible tool that we can use to understand protein folding with rapid progress in computing technology. One of the most popular algorithms used to solve enormously difficult problems in computer science is the ACO (Ant Colony Optimization) Algorithm. The algorithm is based on the basic behaviours of ants in the real world. Ants venture out of their nest to seek food, leaving behind a pheromone trail so that fellow ants can pick up the trail. Soon, there will be a trail of ants leading to the food source. The ant colony is capable of parallel processing, and can seek multiple food sources at one go by sending out ants in different directions.

The algorithm that I will be suggesting is a little bit different from the conventional ACO algorithm. Imagine the entire amino acid sequence as different types of ants being joined together. Every unique amino acid residue is treated like a unique ant. The type of pheromone the ant gives out is based on the properties of the amino acid residue represented by the ant. An ant representing a hydrophobic amino acid residue will secrete a "hydrophobic" pheromone that attracts other ants representing other hydrophobic amino acid residues. This concept of pheromone attraction is akin to hydrogen bonding, electrostatic forces, hydrophobic forces and van der Waals forces of attraction between amino acid residues within a protein structure. The body of each "ant" will be according to the property of the amino acid it represents, and the ants can "contort" their bodies according to the phi and psi angle specifications of the Ramachandran Plot.

The algorithm is as follows:
1) A unique ant will represent a unique amino acid
2) Ant will secrete a pheromone that can attract other ants representing amino acids with a similar property, ants representing other amino acids with different properties can be repelled by the pheromone
3) Ants are capable of responding to the nature of the environment, ants representing hydrophobic amino acids will avoid hydrophilic environment.
4) Ants can "contort" their bodies according to the phi and psi angle specifications of the Ramachandran Plot, with the exception of ants that represent proline. A turn is automatically generated for the proline residue.

Expectations and predictions of the algorithm: A study of how protein folds in four-dimension, and how folding can be affected by the presence of another protein since ants are capable of responding to the nature of the environment, and can react to the "ants" found in the impurity protein. Individual ants and their attracted partners can be customized to react accordingly to changes in environment like temperature and pressure. I postulate that this algorithm will be able to capture the misfolding process of PrPC in the presence of PrPsc.

Citations
1) Lin SY, Chu HL, Wei YS. Pressure-induced transformation of alpha-helix to beta-sheet in the secondary structures of amyloid beta (1-40) peptide exacerbated by temperature. J Biomol Struct Dyn. 2002 Feb;19(4):619-25.

2) Kauffmann E, Darnton NC, Austin RH, Batt C, Gerwert K. Lifetimes of intermediates in the beta -sheet to alpha -helix transition of beta -lactoglobulin by using a diffusional IR mixer. Proc Natl Acad Sci U S A. 2001 Jun 5;98(12):6646-9.

3) Dzwolak W, Muraki T, Kato M, Taniguchi Y. Chain-length dependence of alpha-helix to beta-sheet transition in polylysine: model of protein aggregation studied by temperature-tuned FTIR spectroscopy. Biopolymers. 2004 Mar;73(4):463-9.

4) Safar J, Roller PP, Gajdusek DC, Gibbs CJ Jr. Thermal stability and conformational transitions of scrapie amyloid (prion) protein correlate with infectivity. Protein Sci. 1993 Dec;2(12):2206-16.

5) Eghiaian F, Grosclaude J, Lesceu S, Debey P, Doublet B, Treguer E, Rezaei H, Knossow M. Insight into the PrPC-->PrPSc conversion from the structures of antibody-bound ovine prion scrapie-susceptibility variants. Proc Natl Acad Sci U S A. 2004 Jul 13;101(28):10254-9.

6) http://en.wikipedia.org/wiki/Ant_colony_optimization