Membrane Proteins - Per Bullough Flashcards
What is the lattice of spots showing in Xray diffraction?
The magnified image of the protein crystal lattice
What does Braggs Law tell us
On a crystal surface, the angle of incidence will reflect with the same angle of scattering
What is the problem with X-ray?
No good lens can be created as wavelengths are too short = PHASE PROBLEM
Means we cannot tell relative positions of atom in molecule
How are crystallised proteins formed and prepared for X-ray Crystallography
Protein removed from membrane using detergent
Vapour diffusion by either hanging drop or sitting drop -> supersaturate protein solution until it precipitates out. Use a reservoir around the drop with higher concentration of precipitant solution. Causes water vapour on drop to evaporate leaving only crystallised protein.
Point where crystals form = supersaturation point
What does Braggs law explain?
If wavelengths are fired at a correct angle, where they are only one wavelength apart, their peaks and troughs will align causing reinforcement = very strong signal
What do the characteristics of diffraction spots tell us
Closer to outside = higher resolution information
Darker spot = more intense, so area of higher electron density
Regular arrangement shows it is a crystal, so crystallisation was successful
In general we are told: density of atoms and the density arrangement
What 2 methods can we use to solve the phase problem?
- Molecular replacement –> using a known, related crystal model’s phase information. Optimise orientations until they match
If no similar structure is available then use:
2. Heavy metal atom labelling –> isomorphous replacement. Bind heavy metal (U, Lb, Pl) to crystal and compare diffraction patterns. Use information of where heavy metal binds as anchor points
How and why do we find the correct detergent for a protein?
All slightly denature protein over time. Need it harsh to remove but not too harsh to denature soon
Need to find balance between one that’s best for crystallisation and also for purification -> most common is: DDM and 12M
How does the CMC come into this?
Conc at which micelles self-associate. Need to keep concentration above this to maintain protein solubility
What is good about detergent belts?
Mimic natural environment of protein better than micelles. Can improve crystallisation
How can crystal contacts be increased? why do this?
Proteins can only form contacts between parts not covered in detergent. Smaller proteins have bigger spaces. Bigger spaces make it harder to crystallize
Improved using Fab or Fv antibody fragments to help link proteins
Process of CryoEM:
1) Apply detergent-solubilised protein onto EM grid and blotted to leave a thin film of solution (by removing excess fluid).
2) Rapidly freeze sample by plunging into liquid ethane.
3)Transfer to electron microscope and fire electrons at the sample in a vaccuum.
What is an EM grid?
metal grid with film of carbon with many holes used to hold protein samples for CryoEM.
Why is rapid freezing essential in CryoEM?
Rapidly freezing the proteins in solution prevents the additional crystallisation of the water, ensuring vitrified ICE is formed (non-volatile)
Why is it essential that a sample used in electron microscopy is non-volatile?
So the solution doesn’t spontaneously form a gas, and interfere with the micrograph.
What are the advantages of electron microscopy of protein samples over Xray diffraction?
- Lenses can be used, and-so phase information can be deduced.
- The samples don’t need to be arranged in a lattice with random orientations preferred, crystallisation conditions therefore don’t need to be found.
- Higher Resolution - electrons moving at a shorter wavelength than the Xrays
- Scattering is 10,000 x greater than with X-rays, so much smaller samples need to be prepared (cost saving)
- Protein samples aren’t required to be in set orientations, unlike in X-ray diffraction, requiring manual rotation of crystal to give different orientations of the protein.
- Can be used to identify multiple protein conformations.
What are similarities and differences between X-ray and CryoEM?
Similarities:
- Multiple copies of protein needed
- Extraction of protein from membrane using detergent
- Want different orientations
-
Differences:
- CryoEM can use lens
- Needs smaller samples
- Uses EM images vs diffraction pattern
- Lower cost
- Higher resolution
- sample arranged in different orientations already
- Making 3D models: Single particle averaging vs ???
Does X-ray crystallography show any advantages over CryoEM?
Better at achieving very high resolution in atomic detail for smaller proteins (under a few hundred kDa)
CryoEM better at bigger proteins with more disordered assemblies
What are the issues with electron microscopy in the identification of protein structure?
Radiation damage caused by the electron ray, causes the protein domains to get damage and move, lowering the resolution of the image.
How has the issue of protein damage by electron beam been circumvented in electron microscopy?
- Sensitivity of electron detectors increased, and-so lower dose of electrons needed to capture data.
- Motion correction - techniques compensate for the movement of sample by tracking molecules
- Lower temperatures reduce molecule movement.
How are 3D models / images constructed from electron micrographs?
- Single particle averaging is used to remove background noise and identify the 2D orientations -> These are then aligned by orientation and classified by similarity, averages from each class are converted in a 3D shape according to their Euler angles -> 3D map created is a summation of contributions from each projection image.
What is Alpha Fold?
A program that uses a computational learning algorithm to predict 3D protein structures using 10^5 known structures from the Protein Data Bank (PDB).
The program produces a prediction with area colour coded based on confidence calculations.
How does Alpha Fold work?
It uses the sequence information to search related sequences, analysing amino acids changes and will try to see if there’s a correlation between one amino acid changing another, if so, they’re co-evolving and likely close in 3D space.
What are the limitations of Alpha fold?
- If the protein has few related sequences it’ll struggle to produce a structure.
- Can only estimate the structures of naturally evolving sequences, not as effective with mutations.
- Doesn’t necessarily do sensible chemistry (and-so don’t make sense)
- Limited use for large protein complexes
What is a good indicator for erroneous structures on alpha fold models?
- Low confidence values or long strings of unstructured regions.
How is Alphafold used in conjunction with CryoEM and X-ray?
AlphaFold can be used to help interpret the data, fitting the AlphaFold model to a low resolution CryoEM map or Xray diffraction model, to determine the details and functions of components of the structure. (By comparing the overlapping regions)
What are the two types of membrane bound receptors?
Ionotropic and Metabotropic
Features of ionotropic receptors:
- Ion channels themselves act as receptor
- Very rapid <50ms
- Only two operational states
- Multiple interchangeable subunits
Features of metabotropic receptors:
- Receptor is separate protein to ion channel
- Vary greatly in speed (~100ms to even minutes)
- Nuance to signals, with the ability to dampen or amplify
- Generally monomeric
What are Cys-loop Receptors?
Family of pentameric ligand-gated ion channels that act as neurotransmitter receptors.
Example of Cys-loop receptor:
Nicotinic acetylcholine receptor (nAChR) and glutamate gate chloride (GluCl) channels
What do nAChR channels discriminate ions by?
nAChR discriminate ions on the basis of charge, but not by size.
What is the function of nAChR?
To depolarise the plasma membrane upon binding to acetylcholine to release Ca2+ and cause muscle contraction.
What is acetylcholine an example of?
A neurotransmitter and small molecule ligand.
How was the structure of nAChR first identified?
Electron microscopy using thin crystals of the naturally abundant protein.
From what animal was the nAChR used as a structural model harvested from?
Torpedo electric organ.
What is each monomer within the pentameric assembly of nAChR made of?
4 transmembrane helices and some intracellular and extracellular domains
What is a feature of nAChR’s inner-vestibule to attract positive ions?
The inner-vestibule contains many electronegative ions.
What is the general shape of the pore of nAChR?
Two wide vestibule near the cytoplasmic and extracellular interface, with a more narrow central region within the transmembrane domain, containing the gate.
What is the general structure of nAChR?
It’s a heteropentamer, where each monomer contains 4 transmembrane helices, and extracellular beta sheet domains, these surround the vestibule which narrows into the Transmembrane domain with in-facing hydrophobic amino acids (acting as a gate), not letting the hydrate ions pass.
What is the action that causes the opening and closing of ionotropic channels? (nAChR)
The constriction and dilation of their gates.
Brief description of the opening/ closing of nAChR channels?
Acetylcholine binds at two faces at the extracellular interfaces (beta sheets) of each α subunit with other subunits, this induces a conformational change in the structure, communicated through the cys-loop, pushing on the M2 inner channel helices, opening up the gate (of hydrophobic leucine and valine) and allowing for the entry of the ions.
What soluble molecules can be used to aid crystallisation for structural analysis, used for the example of GluCl?
water soluble antibodies (used Fab to form a GluCl-Fab complex) and Ivermectin (ligand bound to induce active conformation)
What is the key difference between GluCl and nAChR channels?
-Different ions going through and work with different neurotransmitters
- GluCl channels bind to glutamate to let through anions and are inhibitory
-nAchR channels Bind to acetylcholine to let through cations (metal) and are excitatory
Function of C-loop in nAChR and GluCl binding domains and their key distinction?
C loops are near ligand binding domains, enclosing the ligand and within the binding pocket upon binding, and increasing its affinity. KEY DIFF is that nAChR takes place in 1 subunit whereas GluCl’s binding domain takes place over 2.
Function of putative transient binding site (GluCl):
5 electro positive pockets at the cytoplasmic mouth of the pore, attracts chlorides down through the channel.
Role of Picrotoxin:
Drug that binds to open channels, binding specifically to their inner helices.
What did the use of picrotoxin reveal about GluCl’s gating mechanism?
The open conformation had a diameter of 4.6 Angstroms compared to 1.8 Angstroms of Cl- -> suggests it lets through hydrated form of ion.
What are the two conformation states of GluCl found using bacterial homologues?
ELIC - Basal State
GLIC - Active state
How are the mechanisms of protein structures analysed and why?
Membrane proteins are difficult to work with to analyse structure, often clues to mechanism are found by investigating prokaryotic homologues, because they’re easier to handle.
What model is used to describe GluCl gating?
ELIC/GLIC model:
Superimpose the ELIC (red and closed) and GLIC (green and open) structures -> shows slight differences in the positions of beta sheets and the inner helices, specifically the most inner M2 and neighbouring M3. Shows movement of Cys-loop downwards, pushing helix outwards and twisting opens the channel from ELIC to GLIC.
How was the nAChR gating mechanism found?
Looking at Acetylcholine receptor in Freeze trapping and CyroEM, the same protein can be used in a sample and mimic the activity at the synapse. In an experiment Ach can be sprayed at the sample, (with ferritin -> dense enough to be seen electron micrograph -> proves the Ach has hit the sample) -> a moment later the sample is frozen to trap the sample in the open state
What is a nano-disk?
What does the GlyR Cl- channel tell us?
Discovered 3 states:
- Open = inner helices pulled apart, allowing hydrated Cl- in (visualised with picrotoxin)
- Closed = No hydrated Cl- can enter, narrowest part closed
- Desensitised = Glycine still bound, but hydrated Cl- blocked but lower down –> way of turning off signal if on too long
Describe the broad mechanism of Rhodopsin
Photon hits Rhodopsin, causing a conformational change of Retinal from 11-cis-retinal to all-trans. This pushes against Rhodopsin causing it to become excited (R+) and move through transition states until it reaches Meta-II. At Meta-II it exposes binding site for Transducin.
Transducin associates, exchanging GDP for GTP and going on to cause drop in cGMP. Ion channels close and creates hyperpolarised membrane = NT release
How is amplification achieved?
1 photon can activate >500 transducin molecules that can close 1000s channels.
What are the 3 main structures in Rhodopsin that undergo changes
Pivot on TM6 helix
TM5 helix
Ionic Lock
How do these change and what do they cause?
- TM6 –> Pushed by retinal isomerisation and straightening to bend outwards, opening up space for transducin
- TM5 –> Extends out, alongside TM6, to further open binding site
- Ionic Lock –> Forms between TM3 and TM6 to keep TM6 in place (stabilising inactive state). Centred around interactions with Arg135 and other residues. Isomerisation breaks lock
What 2 ways were used to trap intermediates for structural studies?
- Removing retinal and adding fragment of Transducin’s Ga
- Making mutant with constitutive activity (E.g. M257Y mutant -> makes retinal stay in all-trans conformation)
What main similarities does Rhodopsin show with other GPCRs
- Same general spacing of helices
- Water molecules fit in same place
- Ligands bind roughly same place
- Both use ionic lock
What is the role of b2AR
Binds ligand (e.g. adrenaline) that stimulates g-protein to activate Adenylyl Cyclase, decreasing cAMP, closing cAMP-dependant Ca2+ channels and hyperpolarising membrane
What different response levels can they give - what causes this
Can give varying response levels due to different agonists
-Full agonist E.g. salbutamol
- Partial agonist
- Neutral antagonist (keeps at basal activity
- Inverse antagonist (lowers response) E.g. beta blockers
What was the first crystal structure of b2AR
b2AR bound to Carazolol (inverse agonist) stabilised and solubilised by T4 phage Lysozyme (T4L)
What did this show about ligand binding in b2AR?
Carazolol has a hydrophobic ring structure and a polar tail. These interacted perfectly with the side chains of the b2AR binding site
- showed binding site was perfectly designed for tight binding with low dissociation
- greedy drug companies loved this
Also showed hydrophobic sandwich formed around ligand = clamping interactions to keep tight
Compare binding with b2AR and Rhodopsin
Rhodopsin has a lid over the top of the molecule, as its ligand (retinal) is already bound
bAR2 doesn’t have this as it needs ligand to enter
BUT both binding sites are in roughly the same place. And both use same mechanisms: TM6, TM5, Ionic Lock
Why do different ligands enact different change in b2ADs
Full agonists interact with Ser215
Antagonists do not interact with Ser215
This is due to their differences in binding with the receptor
Ser215 is important as it interacts with TM6 and TM5
Why were GPCRs being looked at for drug design?
Looking for a non-opioid analgesic that didn’t cause sedation
Sedation caused by drugs having off-target effects
What did they do?
Looking at A2AAR but based most research on A2BAR as it has its structure resolved and is similar to A2AAR
1. Computationally tested 301 million molecules from data bank against ligand binding site (DOCK 3.7)
2. Gave them a crude enthalpic score based on: Vander Waals + electrostatic interactions and de-solving ligand energy score
3. Filtered scores based on being above this score threshold, and those that presented novel chemotypes and structures
What did they find?
48 potential found from the 31 million. These had biochemistry performed on them -> 65% had hit rates on A2BAR and then 35% had hit rates on A2AAR
How did they optimise these drugs?
Needed tighter binding (in low nM not uM) so made tweaks to structures to do this.
Why is membrane fusion needed?
Membrane repair
Viral entry
Exocytosis
Endosome/Lysosome fusion
What is needed to mediate this formation - why are they needed?
Fusion proteins
Process cannot happen spontaneously as it is very energetically taxing
Describe the key example
Haemagglutinin (HA) fusion protein from influenza virus mediating viral fusion
Describe the key steps of membrane fusion and the barriers they present
- Membrane reorganising to clear patches for contact. Requires H2O removal which is energetically difficult, and bringing membranes together has to stop natural membrane ‘wiggling’
- Transient hemifusion stalk forms which matures into a hemifusion diaphragm. Energetically difficult as lipids are exposed to outside environment, and needs to bend membrane
- Pore opens forming aqueous channel. Fusion proteins stabilise this formation
Eventually have energy payback
Humps formed for each step make this process able to happen
