The Protein Folding Problem — Why It Took 50 Years to Solve

The central question

How does a linear chain of amino acids fold into a precise three-dimensional structure in milliseconds? This question — the protein folding problem — was one of the grand challenges of biology for over 50 years. Proteins don't try every possible conformation; they can't. The solution must be encoded in the physics and chemistry of the amino acid sequence itself.

Anfinsen's thermodynamic hypothesis (1961)

Christian Anfinsen performed the experiment that launched the field. He denatured (unfolded) the enzyme ribonuclease with urea and a reducing agent, destroying its 3D structure and breaking its disulfide bonds. When he removed the denaturant, the protein refolded to its original, functional state — spontaneously and correctly. This proved that the amino acid sequence alone contains all the information needed to determine the 3D structure. Anfinsen won the 1972 Nobel Prize for this work.

Levinthal's paradox (1969)

Cyrus Levinthal pointed out a devastating mathematical problem. If a 100-residue protein sampled just 3 conformations per residue, it would need to explore 3^100 (approximately 5 x 10^47) possible states. Even sampling one conformation per picosecond, this would take longer than the age of the universe. Yet real proteins fold in milliseconds to seconds. Clearly, proteins don't search randomly — they follow some kind of directed pathway.

The energy funnel model

The modern view of protein folding is the energy landscape or "folding funnel" model. Imagine the protein's possible conformations as a hilly landscape, with the height representing free energy. The native (folded) state sits at the bottom of a funnel. From any starting point, the protein rolls downhill toward this minimum, guided by the progressive formation of favorable contacts. There are many different paths down the funnel, not a single pathway.

The funnel isn't perfectly smooth — there are local minima (energy traps) that can cause temporary misfolding. Molecular chaperones are proteins that help other proteins escape these traps and find the correct fold.

Protein misfolding and disease

When folding goes wrong, the consequences can be severe. Misfolded proteins can aggregate into amyloid fibrils — long, insoluble fibers that accumulate in tissues and cause disease. Alzheimer's disease involves amyloid-beta and tau protein aggregation. Parkinson's involves alpha-synuclein. Prion diseases like Creutzfeldt-Jakob disease involve a self-propagating misfolded form of the prion protein.

The computational challenge: CASP

In 1994, John Moult founded the Critical Assessment of protein Structure Prediction (CASP) competition. Every two years, research groups would predict the structures of proteins whose structures had been solved experimentally but not yet published. Progress was painfully slow for two decades — the best methods could handle only simple folds.

Then in 2018, DeepMind entered CASP13 with AlphaFold and immediately outperformed the field. In 2020, AlphaFold2 at CASP14 essentially solved the problem, with predictions nearly indistinguishable from experimental structures. The era of computational structural biology had arrived.