structural biology Flashcards
(96 cards)
1
Q
cellular environment vs purified protein
A
purified protein gives more controlled environment, easier tot assay without background noise, this doesn’t fully reflect cellular environment, consequences on assays and structural approaches
2
Q
considerations when expressing a protein
A
how do we get sufficient quantities to study it, how much do we need, is it toxic to the cell, what does it do/regulate, is it in membrane or soluble, what post translational modifications may be required
3
Q
bacterial cell expression
A
advantages - cheap, easy to grow, easy to manipulate, fast doubling rate, genome is often well characterised and can be scaled up, disadvantages - hard to express non material proteins
4
Q
bacterial cell expression vectors
A
for bacterial cell expression important that the plasmid has the following - bacterial origin of replication, antibiotic resistance, transcriptional promoter (controls binding of RNA polymerase), multiple cloning sites, ribosome binding site (so ribosome can bind and transcription initiated), translation terminator
5
Q
bacterial lines
A
DH5 alpha - mutated in recA1 (inactivate homologous recombination), endA1 (inactivates endonuclease to prevent plasmid degradation and LacZM15 (blue white screening), BL21 - standard strain deficient in certain proteases snd can use the standard tac promoter system, BL21-(DE3) - uses T7 polymerase system giving a tight control of expression, BL21-Rosetta - also contains a pRARE plasmid to overcome problems with rare codons
6
Q
monitoring growth
A
in most instances the growth phase (exponential phase) used for target gene expression to maximise the amount of protein produced, in some cases stationary phase can be used `
7
Q
yeast cell expression
A
advantages - cheap easy to grow, easy to manipulate, genome is often all characterised and easy to scale up, eukaryotic cell, disadvantages - doesn’t to work for all eukaryotic proteins, can be hard to get protein out
8
Q
yeast cell expression vectors
A
for yeast cell expression important that the plasmid has the following - URA3/ampicillin antibiotic selection, multiple cloning site, under the Gal1 promoter, CYC1 TT terminator site, pUC1 ori origin of replication for bacteria
9
Q
HEK cell expression
A
advantages - post translational modifications, good human cell environment mimic, disadvantages - stable cell lines take a long time to generate and transient transfection costly and can be inefficient, the cells can be easily contaminated
10
Q
HEK cell expression vector
A
for HEK cell expression important that the plasmid has the following - multiple cloning site, antibiotic resistance gene, compatible promoter, optional tag (SV40) and bacterial origin of replication (f1 ori)
11
Q
insect cell expression
A
advantages - post translational modifications, can be scaled and the virus stocks can be down, disadvantages - can be costly and times consuming, can easily get infection of the growth
12
Q
insect cell expression - process
A
transformation of competent E coli with gene of interest, selection and expansion of positive clones, isolation of plasmid/expression vector, co-transfection of insect cell lines, production of high titer recombinant virus stock, infection of insect cells with high titer recombinant virus stock; isolation of proteins of interest
13
Q
cell free expression
A
advantages - can do membrane proteins and incorporate non natural amino acids and other labels, avoids toxicity, no proteases, disadvantages - very costly, not trivial to set up, can lack some post translational modifications - done as batch, continuous change or continuous flow
14
Q
native sources
A
advantages - natural environment and associated proteins, no artefacts of expression, for membrane proteins have natural lipid associated and post translational modifications are present, disadvantages - many proteins are in low abundance, no natural tags, often non optimal tissue for sample prep
15
Q
mechanical extraction
A
uses force to break open the cell and includes mortar and pestle, blender, bead beating, ultra sonication and homogenisation
16
Q
mechanical - bead beating
A
glass or ceramic beads used to crack open cell, common used for yeast
17
Q
mechanical - homogenisation
A
cells are lysed by forcing them through narrow space, uses shear force similar to bead beating
18
Q
mechanical - ultra sonication
A
induce vibration within titanium probe immersed in the solution, forms tiny bubbles and explode producing local shockwave (cavitation)
19
Q
non mechanical extraction - freeze/thawing
A
series of freeze thaw cycles ice crystals form which expand upon thawing and cause cells to rupture, why its hard to freeze fruit
20
Q
non mechanical extraction - microwave/thermolysis
A
temperature used to disrupt bonds within cell wall, can also denature the protein you’re interested in, thermophilic proteins perfect for this method
21
Q
non mechanical extraction - osmotic shock
A
use osmosis to increase size of cell till bursts, only used on animal cells and protozoa - don’t have cell wall
22
Q
non mechanical extraction - chemical methods
A
ether, benzene, surfactants etc used to solubilise cell membrane causing lysis, for bacteria EDTA used with chelate the Ca2+ and in turn destabilises the lipopolysaccharides leaving holes in the cell wall
23
Q
non mechanical extraction - enzymatic methods
A
use digestive enzymes to decompose the microbial cell wall, enzymes will depend on cell type as each have different walls/membranes, example is zymolase used for yeast to degrade tough cell wall
24
Q
tag purification - His tag
A
most common purification tag is His tag which is simply 6 to 10 His residues at the N or C terminus
25
tag purification - peptide/epitope tag
typically 8 to 12 amino acids in length and correspond to a specific immunoaffinity tag, one of most common is FLAG tag (typically Asp-Tyr-Lys-Asp-Asp-Asp-Lys)
26
tag purification - folded protein tag
include MBP and GST, columns can be used that are highly specific for these tags, one advantage is they can enhance target solubility
27
tag purification - precipitation/aggregation tags
these can be an effective an inexpensive way of purifying a protein and can have high expression yields with aggregation tags, doesn't not need any sort of purification column but the tags are often large and can require reflowing protocols (RTX tag undergoes reversible precipitation in response to Ca2+
28
tag purification - detection tags
often used for membrane proteins and/or proteins where cellular location is important, commonly a GRP or YFP tag is used which means you can use fluorescence to track the location, presence of the tag/fluorescence is good measure if amount of protein expressed
29
tag purification - solubility/folding tags
often used to increase solubility of the target as the tags are highly soluble, can be beneficial in refolding of target in some cases, their size can be problematic and decrease overall yield, SUMO (small ubiquitin related modifier) shown to enhance expression and folding of recombinant proteins in prokaryotic and eukaryotic hosts, specific SUMO protease means it can be easily cleaved
30
combining tags
different tags have different properties and therefore more than one tag might be useful, important consideration is ability to remove the tag as it could interfere with assays and cause false positives or negatives
31
size exclusion chromatography
separate proteins primarily based on size and shape, purification column contains a resin of beads which contains cavities, the smaller proteins reside longer in the matrix, larger one can pass through quicker (smaller ones get trapped in the cavities for a bit) and elute first
32
ion exchange chromatography
separates ions and polar molecules based on their affinity to the ion exchanger, proteins will bind to the oppositely charged insoluble stationary phase (resin beads) whilst passing through the column
33
centrifugations and glycerol gradients
can separate based on size shape and composition, old method but still effective for some protein systems especially larger protein complexes such as ribosomes ATP synthase etc
34
hydrophobic interaction chromatography
hydrophobic surfaces can attract each other, hydrophobic column can be used to separate proteins based on their hydrophobic nature, by changing buffer composition, in particular salt conc your an change surface characteristics and therefore tune when a protein stays attached to the column
35
lipid influence of structure and function
membrane not just there to support the protein but can also be important for stability and function, big problem when conducting drug screens snd doing structural biology as the proteins may not be representative of how they are in the cell
36
how do we study membrane proteins
many options for stabilising membrane proteins - detergents, nanodiscs, and amphipods , many of the extraction strategies involves detergents , these can remove the lipids and destabilise the protein, use of proteoliposomes and nanodiscs can restore some of these properties, the proteoliposome in particular is a good platform for maintaining bilayer
37
detergents
advantages - well established, larger number of different types so broad application, very effective at extracting membrane proteins and used in range of biophysical techniques, disadvantages - poorly reflect the bilayer of the membrane, viscous properties can slow down/inhibit movement of TM domain and perturb structure, expensive, do not maintain closely associated lipid environment and detergent micelles create problems in downstream biophysical analysis
38
amphipols
octyl groups form multiple hydrophobic interactions with TM domains, the carboxylic groups give the hydrophilicity/solubility, advantages - extensive hydrophobic interactions result in low Koff and small Kd, once added are not required in other buffers, no background micelles and a large library of polymers for different applications, disadvantages - nature of the polymer makes them sensitive to Ca2+ and Mg 2+ ions, viscous nature can slow down/ inhibit movement of the TM domain and do not maintain the closely associated lipid environment
39
nanodiscs
advantages - good mimic of bilayer environment, can supplement with different lipids, no background micelles, different size discs available, disadvantages - requires initial detergent extraction so many associated lipids can be striped away, housing in disc not trivial, discs can interfere with some biophysical techniques
40
SMA
directly extract from membrane and consists of polymer of repeating styrene and maleic acid, advantages - can extract membrane protein and closely associated lipids, large library of polymers available and increased stability and activity, disadvantages - polymer sensitive to Ca2+ and Mg2+ ions, size limitation in the extraction size and works best at 37 degrees
41
closed systems/liposomes
advantages - creates more bilayer environment making closed system for downstream analysis, can improve stability and the vesicle size and lipid composition can be tailored, disadvantages - lipid composition needs to be tailored, membrane curvature must be taken into account, protein orientation can change which affects assays, multi lamella structures can create problems
42
protein structure study techniques
crystallography, electron microscopy, NMR, atomic force microscopy, SAXS
43
x-ray crystallography
main technique for structural biology, key advantage is high resolution (typically <3A) that allows you to see individual atoms of both protein and where appropriate the bound inhibitor, first crystals are made (not trivial), expose these to x-rays and obtain deflation pattern, calculate electron density to give map of structure, computational refinement to obtain structure
44
crystallisation
very challenging, general rules - often have to screen temperature, pH, protein concentration and precipitant concentration, for any protein crystals to form must create contacts between proteins permitting regular crystal lattice to be produced (done by slowly increasing protein conc to put the proteins close together and encourage crystal formation, slats and other additives can help form close packing)
45
problems with membrane proteins
detergent micelle can create significant problems with shielding the membrane protein
46
chaperones
nano-bodies (from llamas) can lock protein and provide areas for crystals packing, not trivial as need to generate nano-bodies and no guarantee of success
47
more stable protein structures
fusing on additional proteins, akin to tags talked about, another option is to stabilise the protein, proteins are dynamic more rigid they are ore likely to crystallise, done through mutations, this is why many thermophilic bacteria are used, easy to purify and crystallise more readily as less mobile at room temp
48
lipidic cubic phase
can maintain more native like environment due to lipid bilayer, game changer for GPCR structural studies
49
robotics and high throughput
old days = crystal trials by hand, screen `200 conditions, now can screen ~1000 conditions with same amount of protein, quicker and more reproducible, machines available to visualise crystals, one challenge is salt also crystallises so hard to know is protein or salt
50
getting diffraction pattern from crystals
scoop up crystals and freeze in liquid nitrogen, crystal placed on beamline or in-house system, can be done with robotics, upon expose to x-rays the crystal will diffract, diffraction pattern determined by unit cell (smallest repeating unit in crystal), diffraction pattern will inform on content of the crystal and from here can determine structure, diffraction pattern shows intensities
51
diffraction to structure
diffraction pattern well understood, flows basic rules - Braggs law (we only have one part of this tho, intensities, we also need the phase info, to do this use either molecular replacement or heavy atom replacement)
52
power of fragment screening
advantage of crystallography and NMR is ability to fragment screen, fragments often poor binders (uM) but by selecting few fragments and joining can create potent inhibitors (rapid fragment screening - ~700 fragments screened a day)
53
new developments
brighter beams = smaller crystals used, pulse rate means you can start to trap distinct catalytic states, very expensive but changing the way structure and function understood
54
electron microscopes for structure discovery
small molecule identification, large breadth of samples from small proteins to large cells (not reliant on crystals so more samples studied, no crystals needed so membrane proteins more possible study, allows for lipid interaction effect on activity and binding to be studied
55
electron microscopy process
purified protein assessed through negative stain, protein then frozen on grid and data collected on high end microscope, computers used to process data
56
grid preparations
grids ~0.5cm in diameter and contain either regular set of holes or random selection, protein then applied in buffer (~3uL) and blotted with filter paper to remove ~99% solution leaving thin layer, this then plunge frozen in liquid ethane to produce vitreous, a good grid has well distributed proteins (can be thin ice in centre of hole pushing particles to hole edge, affinity for carbon support, preferred orientation at air water interface
57
drift correction
depending on detector and mode used expose times vary between 1-60 secs, over this time there can be significant movement of the sample causing burred images, record movies rather than static single image
58
particle picking
once collected pictures in microscope need to pick particles, done by hand or computer program
59
classification
each particle has poor signal to noise so we need to average them to improve signal to noise, results in 2D classes that can show a lot of detail but need to workout how they form a 3D object
60
3D model generation
how is each 2D projection related to construct a 3D model, challenging as difference between views and/or conformational states can get blurred, can be done by common lines (dimention) can help assign angles
61
biased data
images are vary noisy, can result in problems with image bias and wrong structures
62
3D classification
once 3d model generated can classify based on structure, important as could get different states in one data set, aim is to use this to distinguish between bound and unbound states of protein, can look for conformational differences, can see additional interactions and look at mechanical cycling
63
tomography
allows us to look at protein in situ by directly looking at proteins in natural environment, instead of taking pictures of a protein we rotate the specimen in the microscope and take several views of the same specimen
64
advantages of EM
more native like states can be studied including proteins in situ, resolutions in cryo-EM structures have enabled visualisation of small molecules, lipids and antibodies, wide range of systems can be studied including membrane proteins, viruses, large protein complexes and filaments, different conformational states can be identified within data set, smaller amounts of protein (ug rather than mg) needed compared to crystallography or NMR, size and resolution limits constantly decreasing
65
disadvantages of EM
time scale form structure determination is days-weeks, which is low throughput for structure based drug discovery campaigns, grid preparation can be challenging with inconsistencies in ice thickness and particle distribution, although getting better resolutions attained typically lower than crystallography derived structures, instrumental and access to EM facilities can be challenging, computational infrastructure for image processing can be a significant investment
66
NMR
based on absorption of radio waves of certain nuclei in presence of strong magnetic field, some atoms nuclei have not spin (12C, 16O), some nuclei have spin (1H, 13C, 19F), the latter generate small magnetic field can be detected and measured (nuclear magnetic moment, in NMR and external strong magnetic field is generated (B0 in Tesla or Gauss), two nuclei most useful in NMR are 1H and 13C, both have +/- 1/2 spin, 1H at 99% natural abundance, 13C at 1.1% natural abundance
67
magnetic alignment
in absence of external field each nucleus is energetically degenerate, add strong external field (B0) and nuclear magnetic moment; aligns with low energy (align parallel with magnetic field) against high energy (align antiparallel to magnetic field), slight excess of nuclear magnetic moments aligned parallel to applied field
68
nuclear spin
spinning charge (nucleus of 1H and 13C) generates magnetic field, magnetic field generated by nucleus of spin +1/2 is opposite in direction from that generated by a nucleus -1/2 spin, distribution of nuclear spin is random in absence of external magnetic field
69
energy difference between nuclear spin states
no difference in absence of magnetic field, proportional to strength of external magnetic field
70
NMR magnets and spectrometer principles
current passing through coil induces a magnetic field, changing field in coil induces current, net magnetisation aligned along z-axis of magnetic field, B1 field (perpendicular to external applied magnetic field (B0)) is produced by small coil in NMR probe which is placed in bore of the large external magnet, net magnetisation aligned along x-axis of magnetic field after application of B1 field
71
NMR signal
electromagnetic radiation is absorbed when the energy of photon corresponds to difference in energy between two spin states
72
NMR signal sensitivity
relative sensitivity of 1H,13C and other nuclei NMR spectra depend on - gyromagnetic ratio (g), natural abundance of the isotope - gamma = intrinsic property of nucleus cannot be changed, (gH/gC)^3 for 13C is 64x so 1H is ~64x as sensitive as 13C, consider the natural abundance of 13C is 1.1% and relative sensitivity increases to ~6400x, a higher magnetic field will increase sensitivity but comes at a very high cost, sensitivity can be increased by an increase in sample concentration
73
protons in a protein
different protons in a protein are surrounded by molecular fields of variable strengths, takes a stronger or weaker B0 to overcome these molecular fields, this was different protons are resolved at different positions in NMR spectrum
74
nuclear shielding
external magnetic field affects motion of electrons in molecules, inducing magnetic field within molecule, direction of induced magnetic field is opposite to that of applied field, induced field shields the nuclei from the applied field, stronger external field needed for energy difference between spin states to match energy of radiation
75
chemical shift (1H)
measure of degree to which nucleus in molecule;e is shielded, protons in different environments are shielded to greater or lesser degrees, they have different chemical shifts
76
electron paramagnetic resonance - EPR vs NMR
detect unpaired electrons in presence of external magnetic field (B0) (unlike NMR which detects protons), larger charge/mass ration than NMR (magnetic moment at least 680 times greater), higher frequencies, shorter time scales (1ps....1ms), higher sensitivity (`10^11 spins at 2uL at 100nmol/L), longer distances (0.5...8nm)
77
different topology of typical NMR and EPR sin systems
NMR - everybody talks to everybody; lots of information but hard to disentangle, EPR - at least one partner in a talk is an electron spin, often there is only one electron spin
78
the Zeeman effect
EPR - different spin states carry different energy
79
EPR - unpaired electron - spin probes and labels
molecules under investigation like proteins don't carry unpaired electrons so attach stable free radicals to site of interest in a complex, spin probes - stable paramagnetic compounds are added to a system, spin labels - stable paramagnetic compounds covalently bound to molecule, introduced on specific site of interest by site directed mutagenesis, san investigate properties of local environment of the sites within biomolecules, nitroxides as stable free radicals - most common class of soil probes and labels, stable for years at ambient temperature in solids, thermally stable up to ~140 degrees, sensitive strong acids and reducing conditions (TEMPO derivatives, PEROXYL derivatives, imidazoline nitroxide, DOXYL derivatives, MTSSL)
80
high resolution solution NMR of proteins
protons have property called spin angular momentum, behave like small bar magnets and align with or against a magnetic field, these small magnets itinerant with each other, magnetisation can be transferred between 1H, 13C and 15N to establish links and tell us chemical shifts, J-couplings (through bonds), dipolar couplings (through space), with NMR we observe protons (1H), differs from x-ray diffraction where structure determined from electron density from electron rich atoms (C,N,O), protein solubilised in buffer, we can assign proton resonances to individual amino acids (proton resonances often resolved by differences in chemical shifts, measure intra-residue and inter-residue proton to proton distances through dipolar couplings, measure torsion angles through J-couplings, use distance and torsion angles to determine secondary and tertiary stricture,13C and 15N also have spin angular momentum and 'interact' with 1H
81
protein sample to structure determination
sample preparation, resonance assignment, collection of structural restraints, structure calculation, further NMR experiments; dynamic study, interaction study
82
NMR protein sample
isotope labelling - unlabelled (peptides), 15N labelled (small proteins < 10 kDa), 15N & 13C labelled (larger proteins up to 30-40 kDa), 15N & 13C & 2H labelled (large proteins >10 kDa), protein production - E coli, quite a lot, very pure and stable (500 uL of 0.5 mM solution = 5 mg per sample), 13C labelling costly (1000s per sample), preferable low salt, low pH, no additives
83
effect of lipid environment/membrane in NMR spectra
can do these in lipid native environments - important for membrane proteins
84
unpaired electrons in life sciences; unravelling biomolecule structure and function with EPR
oligomerisation (a-/symmetry), conformation in solution or lipids, protein folding, lipid localisation, metal localisation/detection, DNA binding, ligand binding/localisation, structural changes, radical detection, metal detection, side chain mobility, solvent accessibility
85
PELDOR (DEER) spectroscopy advantages over SDSL
condition flexibility (pH, salt, buffer, lipid membranes) one label does it all, small label (specificity, accuracy, accessibility, non-invasive), sub-angstrom (A) resolution, immediate results with no computational requirements, alternative labelling strategies, martial detection, ligand or protein labelling, don't need large amount of protein, same samples can be used with multiple EPR methods, protein folding (state and oligomerisation)
86
EPR sample requirements
30-35 uL, frozen (50-80 K), 10-100 uM of spins, no protein size consideration (as long as distances you want to measure in protein complex are between 2nm and 16nm)
87
molecular dynamics - what is it
way we can computationally study dynamics of protein or protein system and mimic what atoms do in real life, energy function allows us to calculate force experienced by atom and allows us to calculate its time dependant behaviour, newton's laws are used to tell us how the forces affect the moons of atoms and computers calculate their motion, number of length and time scales we can work with from quantum (individual atoms/ electrons) to continuum (look at the protein as more if a material), useful for larger protein complexes where they contain lots of atoms and the information we are interested in is in the ms time scale
88
molecular dynamics - applications
running molecular dynamic simulations can allow us to assess factors such as; conformational changes, protein folding, protein stability, transport and catalysis, drug design,
89
how big can we go
satellite tobacco virus with up to 1 million atoms over 50 ns, even at such short time-frame possible to see in absence of RNA capsid collapses, provides insights into virus assemble and infection
90
Atomistic approach
basic underpinning of molecular dynamic simulations is straightforward and we use newtons second law (force=mass x acceleration) to model the atoms, from here we can build up much more complex systems, E-bond-stetch represents interaction between between atomic pairs where atoms are separated by one covalent bond, this is a trade off between the ideal bond length and the force constants (the bond wish to be ideal but often a force is trying to push or pull the atom), E-angle-bend is associated with the alteration of bond angles from the ideal value and is a trade off/balancing of forces, E-rotate-along-bend models the presence of steric barriers between atoms, these three added together gets E-bonded
91
constraints
bond length and angle is constrained and can only deviate from ideal value through energetic cost, these rules allow us to model the structure and mean that the atoms can be modelled with certain criteria
92
modelling inter and intra bonds
can look at different ways atoms interact this can be through rigid connects or flexible springs, characteristic of bond type can determine nature of the bond, the electrostatic interaction between a pair of atoms is represented by the coulomb potential with charges, Van Der Waals also play a role, must consider all these when putting model together
93
modelling diferent contributions
can build a picture of which atoms are directly linked together by bonds and which have other forces such as Van Der Waals and electrostatics, by doing this each atom can contribute in different ways to simulation, we also need to consider the protein environment (water/membrane), this can play key role is stability/function of protein so just be modelled, challenging as don't always know what outside environment is composed of, when you consider the amount of data points atomistic approach becomes difficult especially for larger proteins or large length scales
94
coarse graining
has to be a trade off between level of complexity we can study and time time frame, more data we sample the more computational resources required, reducing complexity means we get less "detail", hybrid models can combine these elements to ensure that longer times frames or bigger systems can be studied
95
pocket flexibility/plasticity
when designing a new lead compound you must consider the flexibility within the pocket
96
modelling clustering
MD simulations are not just about the individual protein but how do a variety of proteins behave in the membrane, do they cluster and interact? by using MD we can start with proteins equally spaced then see how long it takes snd that organisation the proteins make in the membrane, can provide information of packing within membrane
