GPCRs Flashcards
(39 cards)
Importance of understanding GPCRs
Ubiquitous in almost every cell
Understanding rhodopsin structure was important due to conserved structure in GPCRs allows us to develop improved ligands for others to treat disease. Ex. GPR84 has a fully conserved binding pocket across vertebrate orthologues. Can use a molecule that makes the same 5 H bonds as the natural agonist to activate. Drug tests on mice for human medicines will bind the same as the GPR84 in humans since conserved throughout evolution.
Heterotrimeric proteins: why not direct interaction of GPCR to enzyme, number, examples and function
Guanidine nucleotide binding proteins
-Allows one G alpha protein (~20 in humans, less than no. of GPCRs) to associate with multiple signaling pathways (effector for one GPCR can be ex. ion channel, adenylyl cyclase, IP3, etc; versatile).
Some have a widespread role (ex. only 2 Gs isoforms since many cells need regulation of cAMP), others are distinctive and only expressed in one cell subtype
-One activated GPCR activating multiple G proteins, which can activate multiple effector proteins allows for signal amplification
The most diverse of G alpha proteins share 50% amino acid identity with a very similar tertiary structure due to all binding GTP (when active) and GDP (when inactive), and working in the same general way, similar to how kinases all have a similar structure due to binding ATP.
Some G proteins are ubiquitous in almost every cell (ex. only 2 Gs isoforms since many cells need regulation of cAMP) wheras others expressed in specfic tissues (ex. Golf is a specialised G protein in nasal epithelium to couple G proteins that control sense of odour perception to adenylyl cyclase)
Gs stimulates adenylyl cyclase, increasing cAMP levels.
Gi inhibits adenylyl cyclase, thereby reducing cAMP levels.
Gt1 (transducin alpha-T1) and Gt2 are specialized G proteins expressed in photoreceptor cells of the eye, where they mediate vision.
Gq activates phospholipase C-β, leading to increased intracellular calcium. Specialized versions of Gq are expressed in specific tissues and cell types
First TM protein imaged, protein requirements for imaging and why said protein fit these. Information obtained from structure
Advances in biochemistry and related over last 20 years have changed how we think about how proteins and cell signalling works. Requirements for atomic protein imaging:
High purity
Stable, native conformation
Bovine rhodopsin is the most studied GPCR. The first transmembrane protein successfully purified (challenge to keep protein stable out native environment and detergent dissolves membrane) and imaged in 2000.
Rhodopsin is 50% of all protein in rod outer segment (couples to G alpha protein transducer (T1 for monochromatic recognition of light and T2 for colour vision)) so straightforward to obtain tissue and have sufficient protein from native source after purification (typically the amplification cascade in signalling means there’s low concentrations of receptor). Most GPCRs require 20,000-100,000 fold purification
7TM helices with kinks important in activation and deactivation
Covalently attached chromophore/ligand that isomerises when absorbs light on microsecond scale (very quickly): 11-cis retinal into all-trans retinal. This chemical change drives conformational change in protein
Evolution of GPCRs in biology (how they arose, no. of genes, why different no. in other organisms, how we know function)
Typically via gene duplication the genes diverse over time to produce different physiological functions in different tissues. Still conserved enough to respond to the same ligand.
Sequence alignments and grouping similarities can see those that respond to the same ligand.
Groups in phylogenetic tree that indicate convergent evolution shows responding to the same ligand and having 7TM helices were structures that occurred multiple times in evolution (as well as divergence), highlighting that those traits must give an advantage.
~800 genes encode GPCRs (3% of coding genes). ~400 olfactory (which is why non-humans have more GPCRs since competitive advantage to have good sense of smell. In humans have pseudogenes; non-functional). ~200 don’t know natural ligand that regulates (may aid treating disease to understand). Done by KO studies and observe change in function/phenotype or DNA fingerprinting to reflect differences. UK biobank allows people to ask if they have SNP associated with susceptibility to a disease.
All organisms have GPCRs, highlighting importance (except parasites)
Types of drug types based on activation
Full agonist: molecules that enrich the active state (R*)
In physiology, anything with less efficacy than a full agonist (partial agonist, antagonist, partial inverse agonist, inverse agonist) will partially block it’s effect
Full inverse agonist: molecules that enrich the inactive state (R)
Antagonist: molecule that doesn’t change balance between R and R*
All receptors (even rhodopsin which has extremely little) constitutive activity. Like how enzymes increase the rate of a reaction that would occur without it’s intervention, receptors can still become activated without their ligand. Different receptors have different levels of constitutive activity. All have it though since their activation must be energetically favourable to occur with ligand, and so must be able to activate alone.
Information obtained through phylogenetic tree of GPCRs
Through informatics, aligning proteins and grouping by similarity, identified number of receptors that respond to a ligand
GPCRs frizzled, glutamate, or adhesion/secretin have no similarity to other GPCRs (other than having 7TM helices) that can all be grouped together (opiates, cannabinoid, etc)
This indicates they did not arise from gene divergence of a gene gaining mutations, instead convergent evolution.
Highlights use of GPCRs
Direct Binding Experiments (method, information obtained)
- Use increasing concentrations of radiolabelled drug in the presence or
absence of an excess of a second drug which is known to compete with the
radioligand for the receptor - Provides information on:
- The total number of receptors present
- The affinity of the radiolabelled ligand for the receptor
- In a similar fashion to linear transformation of the Michaelis-Menten
equation to Lineweaver-Burk or Eadie-Hofstee plots, the same can be done
with binding data. The most common form is the Scatchard plot.
Criteria for specific, physiologically relevant binding
-Saturable (dose dependent so can regulate)
-Stereoselective (receptors are often chiral so will only bind one enantiomer)
-Suitable tissue profile (only picked up in target tissue; confirm expression of the target receptor in a tissue via qrt-PCR or protein levels)
-Binding should be competed for by pharmacological doses of
receptor selective drugs
-Binding should not be displaced by drugs of other classes
-Selective for the one receptor
-High affinity, low Kd (low dosage required)
Initial studies on 2nd messengers showed and next steps
Addition of glucagon raised cAMP levels (measured with radioimmunoassay in which [3H]cAMP incubated with Ab and competitive antigen, and radioactivity measured and concentration calculated with calibration curve) in hepatocytes
Not obvious from these studies that GTP was required since ATP used was purified imperfectly from tissues and contaminated with sufficient GTP. Once chemically synthesised ATP was available in became clear that GTP
also had to be added for function
Conclusion: there is a step in the cascade from glucagon activating the GPCR to 2nd messenger cAMP production via AC that requires GTP
What was the molecular nature of this ‘transducer’?
Experiment perform to prove drug target for a GPCR was a protein + next steps
Through covalent labelling
Example: Identification of muscarinic acetylcholine receptor
Modified antagonist propylbenzilylcholine to react via covalent bonds (propylbenzilylcholine mustard), comprised of radioactive hydrogens ([3H] propylbenzilylcholine mustard)
Knew muscarinic acetylcholine receptor is present in brain so incubated animal brain with antagonist, solbilised, denatured and added to SDS-PAGE
Added excess of another antagonist atropine. On SDS Page gel imaged with autoradiography, saw signal band without atropine but none with atropine, suggesting both antagonists bound the same receptor
Experiment performed to purify a GPCR + next steps
Example: Purification of the delta-opioid receptor
Used neuroblastoma cell line grown in lab and knew delta-opioid receptor was present that has a fentanyl ligand
Modified fentanyl with radioligand and highly reactive chemical group to covalently bind to what fentanyl binds to ([3H] fentanyl isothiocyanate)
Solubilised membrane with detergent
Ran on SDS-PAGE which showed lots of bands (all proteins in cell solute) and autoradiograph showed multiple bands but less (only the few proteins bound to fentanyl since it has multiple targets)
Applied solution column of wheat germ agglutinin - a lectin so selective for carbohydrates (PTM for all PM; N-linked glycosylation) to purify fentanyl (with bound receptor). Saw only one band in autoradiograph so successfully purified out other proteins that bind fentanyl. SDPAGE showed fewer but lots of bands since purification isn’t perfect
Applied solution to another column of antibody to the ligand on sepharose beads.
30% purity of the opioid receptor was achieved.
Next steps:
Chopped up, sequenced, and designed primers based on cDNA library to be able to clone protein. However, sequencing doesn’t tell you which codons are used to encode that amino acid (except tryptophan which has has one codon that encodes for it since it was the last produced in evolution)
Experiment performed to understand beta-2 adrenoreceptor + next steps
Took lung from guinea pig since they knew β-adrenoreceptors would be present there as drugs to treat asthma target lung and are selective for that receptor
Solubilised membrane (since it’s an intrinsic membrane protein) and did tested multiple detergents
Use affinity chromatography of sepharose column with aprenolol (treats heart failure), a high-affinity ligand for the receptor that doesn’t bind covalently (reversible attachment)
Purified 100,000 fold (typically need 20,000 - 100,000 fold for GPCRs) a single polypeptide of 64,000
Next steps: characterise and answer if the polypeptide is sufficient to produce functional characteristics of the receptor (are there other interacting proteins necessary)
Chopped up, sequenced, and obtain aa information to design primers based on cDNA library to be able to clone protein.
However, sequencing doesn’t tell you which codons are used to encode that amino acid (except tryptophan which has has one codon that encodes for it since it was the last produced in evolution)
Predicated MW based on amino acid sequence to be more than 64,000 (must have PTM). Using enzymes that cleave off carbohydrates measured MW to be 64,000 on gel
Discovered 7 segments rich in hydrophobic residues (20-24aa to pass PM), inferring these regions passed the PM. Since there were multiple proteins with this feature although in different tissues (muscarinic adrenoreceptor in muscle, and rhodopsin in eye), must be a protein family
Example of GPCR with same ligand but different effects
Acetylcholine in brain (research for dementia), heart (parasympathetic slows heart rate), iris in eye, etc
Is this multiple genes that respond to acetylcholine, or the same one expressed in these tissues
Now know it’s different acetylcholine receptors
Drug Pirenzepine discovered that has different affinity to block muscarinic receptors in brain and heart, indicating almost certainly it’s a different protein
How did we move from isolating individual GPCRs ‘one at a time’ to
identifying the full family in humans?
Linda Buck won Nobel prize for paper published in 1991
Noticed TMD3 and TMD6 tended to have high sequence similarity (now know highly conserved due to binding )
Intracellular loop 3 (links TMD5 and TMD6) varies in length
Developed primers for TMD3 going 3’ to 5’ and an opposite primer for TMD6 from 5’ to 3’ to amplify region inbetween that includes intermolecular loop 3
Run on agarose gel and obtained lots of bands of different sizes; fragment library of different lengths of 3rd loop, and so different GPCRs
Cut out band and sequenced. Compare with database to identify if it’s a GPCR that has been cloned before and matches known sequence. If no match then indicates it’s a novel GPCR.
Identified existence of over 400 novel GPCRs of odorant sensors in olfactory neurons of our nose
Example of molecular basis of selective binding of a GPCR
GPCRs that bind catecholamine ligands
All have key residues conserved in the same position
Amine head group forms salt bridge with aspartic acid in TMD3
Hydroxyl groups in catechol ring make H bonds with pair of serines three amino acids away in primary sequence and so so adjacent in alpha helix (~3.6 aa per turn) in TMD5
Benzene core makes pi-pi stacking interactions with phenylalanine in TMDVI
Structural alterations
associated with GPCR activation, experiments that show
Protein imaging based approach:
Bottom of TMD6 has glutamate
DRY domain in TMD3 (sometimes D is E) is the most conserved residues in family
Close in tertiary structure in inactive receptor state and interact, forming ionic lock
TMD6 moves outwards when ionic lock is broken, swinging out like a gate allowing space for G protein helix to enter bottom of receptor and induce next step in signal cascade (ex. initial experiments in x-ray crystallography imaged inactive rhodopsin to active opsin,TMD6 moves 12A when activated by isolating protein in dark/red light, and beta2 adrenoreceptor).
Disadvantages: static image of snapshot of a dynamic process
Fluorescence quenching based approach:
Modified Bimane so the closer to tryptophan, to more it quenches tryptophan signal (best fluorescent signal)
Added bimane to TMD6 (since moves a lot). Observed large change in fluorescent signal from inactive to active state.
Disadvanges: Needed to purify protein to perform
Hydrogen-deuterium exchange mass spectrometry (HDX-MS):
Used in early 2010s to study conformational changes in GPCR activation. Amide H on protein backbone can exchange with deuterons (2H) in D2O containing buffer, and the exchange rate increases with increased solvent accessibility and decreased H bonding.
With mass spectrometry, and rate of increase in mass over time for each peptide segment indicates flexibility and solvent exposure.
Performed on beta2 adrenoreceptor with isoproternol agonist and showed an increased in flexibility in ICL3 (for G protein binding) and TM6, and antagonists and inverse agonists caused more rigid structure.
Disadvantages: Information is low res.(5-15 aa peptides) on relative dynamics not specific aa movements, so coupled with imaging techniques and molecular dynamic simulations
Molecular dynamic simulations:
Computational technique to model movement of each aa over microsecond time scales
Uses data from experiments and mathmatical models based on Newton’s laws of motion to calculate movement under influence of physical forces (van der waals, etc)
Disadvantages: lots of computer power required
Use of GPCRs in disease succeptibility
Splice and polymorphic variation (SNPs) can introduce further diversity
potentially modifying regulation and function
Some are associated with certain diseases. GPCRs are very prevalent in body (3% of coding genes) so high chance they contain SNPs
Ex. GPR65
I231L increases risk for Inflammatory Bowel Disease
Mutations investigating activity of GPCRs
-Beta2-adrenoreceptor:
Mutated aa close to junction where TMD6 becomes the 3rd intracellular loop.
Modified beta2-adrenoreceptor with changes to look like alpha1-adrenoreceptor.
Thought since beta2 modifies cAMP but alpha1 calcium, thought they’d obtain receptor that responded to signals similar to alpha1, but with effects of beta2
Obtained constitutively active beta2-adrenoreceptor with higher agonist activity
R* agonists bind receptor more tightly than R, since it promotes binding of G protein to receptor
-Alpha2-adrenoreceptor:
Different amino acid substitutions of T348 changed balance of R to R* induced by agonist
WT had the lowest constitutive activity
-Alpha1-adrenoreceptor:
Changed A293 to all amino acids.
Alanine had the lowest basal activity
Cholera toxin: mode of action in body, mechanism, activation
Cholera bacteria enters endothelial gut lining (sheds regularly so bacteria is excreted) and produces toxin
Causes efflux of water by modifying ion channel
Symptoms is watery diarrhea and persist if contaminated water is continually ingested
Activation:
Persistently increases cAMP levels in intestinal cells
ADP-ribosyltransferase: Covalently modifyes Galphas (takes NAD and adds ADP ribose)
Inhibits GTP hydrolysis, locking Galphas in active GTP bound state
Adenylyl cyclase continually activated and increases cAMP
Method to identify components of GPCR signalling and what they found
S49 lymphoma cells WT and UV treated (allow cells to grow to clone out DNA damage), both have isoproterenol added.
In WT, cAMP levels increase (due to GPCR activation) leading to cell apoptosis
In UV treated cells survive, indicating the receptor is no longer expressed due to DNA damage to components of signalling: can’t produce cAMP or respond to it).
Multiple lines made (cyc-, unc, etc) since UV may have damaged multiple proteins so to validate results.
Termed cyc- cells for cyclic AMP deficient. Adenylyl cyclase originally thought to be the mutated protein
S49 cyc- cells were charecterised and showed the mutant protein linked the beta-adrenergic receptor to AC:
Receptor binding assay (radiolabelled agonist propranolol and measured binding affinity) showed the receptor was present and could bind ligand properly (not the mutant)
Adding a direct AC activator (forskolin) led to cell death (cAMP increase) in S49 cells, indicating the mutant protein in the GPCR signalling pathway acted upstream
Radioligand binding assay: incubate cell with radiolabelled non-hydrolysable GTP analogue [32S]GTPgammaS or [3H]GDP to measure guanidine nucleotide shift (change in affinity of GTP/GDP for G alpha once the GPCR is activated). No increase in GTP binding or GDP disassociation after receptor stimulation,
Cholera toxin:
Lack of incorporation of [32P]ADP ribose by cholera toxin to S49 cyc- cells and didn’t increase cAMP
In WT S49, using [32P]NAD+ with cholera toxin results in incorporation of [32P]ADP-ribose into a 45kDa polypeptide (G alpha s) to a key arginine
Indicates mutant in a protein that must link GPCR to AC
Found the specific mutation in cyc- cells was in promoter
How to define molecular nature of Gs
Functional reconstitution assay
Membranes from S49 cyc- cells obtained (contains AC and beta-adrenergic receptor)
Add column fractions from detergent-extracted mouse liver membranes (relevant protein must be present since previous studies done with hepatocytes)
Optical density measured at 280nm, of the amount of protein present to monitor
If Gs is present in a fraction this should now allow isoproterenol stimulation of cAMP production
Further purification steps, keeping the fractions that reconstitutes the pathway
Ideally reach a single homogeneous polypeptide that displays the relevant activity
Found that no matter the fractionation scheme used, the active fraction always contained a 45 kDa protein (corresponds to protein cholera toxin modified), but also another protein at 35kDa and 8kDa
Presumably a hetero-trimeric complex
Pertussis toxin: mode of action in body, mechanism, activation
Produced by bacteria that causes whooping cough
ADP-ribosyltransferase that in pancreatic islets with pertussis toxin, cAMP increases
Using [32P]NAD+ plus Pertussis Toxin
results in incorporation of [32P]ADP-ribose into a 41kDa polypeptide to a key cysteine residue
Toxin acts to prevent receptor interaction with Gi alpha protein by inhibiting a Gi mediated decrease of cAMP
Mechanism of heterotrimeric G protein regulation
In the resting/inactive state, the G proteins are in a heterotrimeric complex with the G alpha subunit (guanine nucleotide binding protein) is bound to GDP
Receptor activation causes the receptor to form a complex with the G-proteins and promote the off-rate, allowing disassociation of GDP
In a healthy cell, the high energy charge (sufficient levels of ATP and GTP for energy, and GDP and ADP low) allows GTP concentrations to be sufficiently high it takes the place of GDP, promoting the disassociation of G alpha from the complex to activate downstream signals
To desensitise the signal, G alpha has GTPase activity to hydrolyse the terminal gamma phosphate of GTP to have GDP in the binding pocket. The G alpha protein recombines with G beta and gamma
Terms:
The receptor is a guanine exchange factor (GEF) - promotes the exchange of GDP for GTP
Regulators of G protein signalling (RGS) proteins - controls rate of GTPase rate
GTPase activating protein (GAPs) enhance the rate of GTPase activity
Toxins effects on G proteins
Cholera toxin - ADP-ribosylation on a key arginine near the its GTPase domain blocks the GTPase. The Gs remains in active state, continually activating AC
Pertussis toxin - ADP-ribosylation on a key cysteine residue near the GTP binding site of Gi prevents release of GDP by preventing the interaction with the receptor. Locks in inactive state
Rare mutation that alters arginine cholera toxin modifies
Pituitary tumours occurs from increased cAMP production
Mutation