Protein structure determination Flashcards
How does X-ray crystallography work?
-generated x-rays are fired at sample held in place
-x-rays scatter through sample, producing a scattering pattern
-must use crystals (unit cell made up of unit lattice and motif) -lots of identical copies
What information does a x-ray diffraction pattern give?
-types of atoms (amplitude of each spot)
-level of detail (position of spot)
NOT the relative positions of atoms (known as the phase problem!!) -But can calculate the phase :)
What is needed to crystallise a protein?
-use abundant supply of protein (natural or recombinant)
-variants (diff species, diff isoforms)
-purification protocol which is simple and quick
-purified protein
-homogenous sample
-sample conc 1-10mg/ml
-large quantities of protein (1-5mg)
How can conditions for screening for crystallisation be altered?
-using precipitating agents (eg. ammonium sulphate)
-vary pH
-vary temp
Find which conditions make protein supersaturated (aka a crystal structure)
Why is it difficult to use x-ray crystallography on membrane proteins?
-bilayer needs to be disrupted to purify protein
-once protein is extracted from bilayer, it will aggregate in water -so need to use detergents
What are the properties of detergents used for extracting membrane proteins?
-water soluble (more so than the lipids!)
-amphiphile (most have polar head group and long non-polar hydrocarbon chain)
How do detergents enable membrane protein extraction?
-at critical micellar concentration, detergents self-associate into micelles
-at high conc, the micelles will form with itself, lipids from bilayer or with membrane-proteins (what we want!)
N/B: at low conc, won’t form micelles, but get incorporated into membrane (not what we’re after!)
How does Cryo-EM work?
-membrane protein solution applied onto EM grid and blotted
-plunged into liquid ethene to freeze rapidly -maintains non-crystallised structure (water-like rather than ice-like), will be in random orientation
-e- microscopes used to take image
-backwards compute lots of images to reconstruct 3D struct
What recent developments in Cryo-EM have been made?
Since 2014
-new v. sensitive detectors
(-new computing software
-new ways of analysing noise)
-images have much better resolution
-can build atomic models
-enables multiple-conformation proteins to be studied
(great for membrane proteins!)