Goal: Determine atomic structure of proteins.
Applications:
Drug design (active site info).
Understanding binding mechanisms.
High-resolution structures.
Can trap intermediates (via flash freezing).
Suitable for small & large proteins.
Rapid (esp. with synchrotron sources).
Typically done at −173 °C (not physiological).
Static, average structure.
Hydrogens often not visible.
Membrane proteins are hard to crystallize.
Good for small, dynamic proteins.
Difficult sample prep.
Uses radio waves.
Great for large complexes.
Low protein concentrations required.
Needs cryogenic samples.
Crystal: Ordered 3D array of molecules.
Unit cell: Smallest repeating unit.
3 axes (a, b, c) and 3 angles (α, β, γ).
Bragg’s Law: Constructive interference when path difference = nλ.
Crystals amplify X-ray signal due to uniform atomic planes.
Supersaturate protein solution.
Add precipitant → nucleation → crystal growth.
pH, buffer, salt type/concentration, temperature, protein concentration.
Hanging drop: Good visibility/control.
Sitting drop: Automation/high throughput.
Vapour diffusion: Equilibration increases protein/precipitant concentration slowly.
No universal rules—trial and error.
Requires pure, concentrated, stable protein.
Not all proteins crystallize.
Flash freeze with cryoprotectant (e.g., 25% glycerol).
Reduces radiation damage.
Traps enzyme intermediates and oxidation-sensitive states.
Laboratory: Fixed wavelength (1.54 Å), low brightness, slow (1–2 days).
Synchrotron: Tunable (0.6–1.6 Å), fast (minutes), higher brightness.
SFX (Serial Femtosecond Crystallography):
Streams microcrystals.
Captures ultrafast events.
X-rays diffract off electron clouds.
Data recorded as reflections (spots) on detector.
Rotate crystal → build 3D dataset.
Pattern is in reciprocal space (closer spots = wider atomic spacing).
Measured in Ångströms (1 Å = 0.1 nm).
Lower Å = higher resolution.
X-rays detect electrons (hydrogens poorly visualized).
X-ray measures amplitude, not phase.
Phase must be estimated to build model.
Molecular Replacement: Needs model with >30% identity.
Experimental Phasing:
MAD (Multi-wavelength Anomalous Diffraction)
SAD (Single-wavelength)
MIR (Multiple Isomorphous Replacement)
Fit atomic model into electron density map.
Validation:
Bond lengths/angles.
Ramachandran plot (φ, ψ angles of amino acids).
Side-chain rotamers.
Difference maps.
B-factor (temperature factor) checks.
Spike protein bound to ACE2 receptor.
Stabilized by 4 disulfide bonds (oxidized cysteines).
Crystallography helped determine structure for vaccine design.