key points Flashcards

1
Q

What problems can be explored in structural biology? (dntk)

A
  • Biomolecular self-assembly
  • Biomolecular dynamics
  • Structure-activity relationships
  • Biomolecular interactions
  • Biomolecular structure
  • Ligand binding/design
  • Physical chemistry of biomolecules (e.g. thermodynamic properties)
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
2
Q

What is a method we can use to help characterise biomolecular interactions?

A
  • Isotopic labelling can be used to enrich the system with 13C/15N.
  • As these are low abundance, they will stand out in analysis.
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
3
Q
  • What is the size limitation problem in solution NMR? How is solid state affected by this?(IMP)
A
  • Rotational correlation time (τc) scales inversely with MW
  • As MW ↑, T2 signal-to-noise ↓ and peaks broaden
  • This is due to T2 being so short that signals relax quicker than data can be acquired.
  • Solid state NMR does not suffer from this, however peaks are generally broader anyway
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
4
Q
  • What is the variable intensity problem? (IMP)
A
  • Intermediate exchange regime leads to variable peak intensities
  • meaning some peaks will be weak, others intense
  • 𝛿A = resonant frequency associated with state A
  • Larmor frequency difference (Δω) >> rate of exchange between A/B (kex) à see both states
  • Δω << (kex) forms single weighted average state
  • Δω ≈ (kex) results in broadening –> variable exchange
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
6
Q

What is the problem associated with proteins unfolding? What conditions can cause this to occur?

A
  • Aggregation on proteins can lead to broad peaks
  • Poor chemical shift dispersion of 1H peaks leading to severe overlap, and will appear all in the same environment
  • High pressure can cause proteins to unfold and chemical shift envelope to become more compact
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
7
Q

What is insensitivity in NMR? How can it be overcome?

A
  • 1 in 1E+06 signals are being measured of all excited nuclei, meaning experiments can take many days as this is very few signals (insensitive)
  • This can be overcome by running at higher temperatures, but biomolecule may not be able to retain its stable fold
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
8
Q

Why is the knowing stability of a protein fold important?

A

Conditions in which this takes place are important to know as will massively change structure of sample

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
9
Q

What is the problem with isotopic labelling?

A
  • Enriching our organic sample with low abundance 13C/15N isotopes is generally done by growing bacteria and enriching them.
  • These feedstocks can be very costly.
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
10
Q

Advantages of NMR

  • NMR provides atomic for molecules in solution with MW ≤ 1E+05 Da and can also mirror conditions. Doing this in solution is the best way to validate properties in a biological … ….
  • NMR can be used to access complexes in …-binding studies where are monitored
  • Accessible nuclei cover all biologically-relevant suspects: 1H, 15N, …, 31P
A

Advantages of NMR

  • NMR provides atomic resolution for molecules in solution with MW ≤ 1E+05 Da and can also mirror physiological conditions. Doing this in solution is the best way to validate properties in a biological cellular environment.
  • NMR can be used to access larger complexes in ligand-binding studies where ligands are monitored
  • Accessible isotopic nuclei cover all biologically-relevant suspects: 13C, 1H, 15N, 17O, 31P
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
11
Q

Why do we have a mixing period in 2/3D NMR.

A
  • In multi-dimensional NMR we want to know how nuclei interact with one another
  • Can be through scalar (bond) or dipolar (space) coupling
  • Mixing allows coherence between signals to transfer to find degree of each coupling.
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
12
Q

How are solvent peaks eliminated from FID acquisition?

A
  • Continuous wave decoupling destroys the coherence of large 1H peaks due to water
  • A high power 90O acquisition pulse if done quickly before water has a chance to relax back in to spectrum.
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
13
Q
  • A good 1D spectrum has good suppression of large solvent peaks. It can also tell us about the chemical shift dispersion in the spectrum. What is this and what else can a spectrum tell us?
A
  • Chemical shift dispersion describes the general spread of the chemical shifts for a given nucleus in the NMR spectrum. If this is very clustered, it would not be wise to proceed with further 2D/3D NMR, due to high overlap.
  • A good spectrum will also tell us the concentration of our sample of interest as well as its purity.
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
14
Q

What is the evolution delay, t1, in 2D NOESY?

A
  • Evolution delay creates a second dimension via Fourier transform
  • Incrimented in time to capture different chemical shift information in each plane.
  • Collect # of planes with a different value for given incremented t1 delay.
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
15
Q

What is homonuclear NMR and what are some advantages and examples of types of molecules it can be used for

A
  • 1H detection in both dimensions
  • Circumvents need for expensive and time-consuming isotopic labelling.
  • Works with even only a few signals/1Hs
  • Can be used for
    • Peptides (short proteins)
    • Carbohydrates
    • DNA/RNA fragments
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
16
Q
  • … … (TOCSY) 2D spectrum yields correlation map for coupled 1Hs.
  • Coupling observed bonds of intervening 1H couples
  • Interrupted only by small/zero …-…coupling or atoms e.g carbonyl
A
  • Total correlation (TOCSY) 2D spectrum yields correlation map for scalar coupled 1Hs.
  • Coupling observed ~6 bonds of intervening 1H couples
  • Interrupted only by small/zero 1H-1H coupling or hetero atoms e.g carbonyl
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
17
Q

What are the limitations of TOCSY?

A
  • While TOCSY good for isolating spin systems due to coupling within an amino acid side chain.
  • Presence of carbonyl shuts of scalar coupling, halting the TOCSY chain
  • Can assign by residue type, but if > 1 of the same AA, cannot discriminate easily in sequential assignment of chain.
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
18
Q

What is spin locking?

A
  • A method of maintaining coherence and preventing dephasing during transfer of magnetisation through scalar coupling.
  • This is done by locking magnetisation along an axis using an RF (pulse) field, causing magnetisation to flip and reacquire coherence.
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
19
Q

The Nuclear Overhauser (NOE) effect is a form a … (…-…) relaxation, in which information is coupled between nuclei through (< Å)

A

The Nuclear Overhauser (NOE) effect is a form a dipolar (spin-lattice) relaxation, in which electronic information is coupled between nuclei through space (<5 Å)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
20
Q
  • Outline NOE’s origin
A
  • Perturbation of one nucleus (spin I) from equilibrium can cause changes in relaxation rate (R1) and population distribution of nearby spins (S)
  • Cross relaxation generates an increase or decrease in intensity on neighbouring spins (NOE).
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
22
Q
  • (IMP) Describe a 1H-1H NOESY pulse program
A
  • Three π/2/90o pulses
  • 1: Creates transverse magnetisation precessing during incremented evolution period, t1
  • 2: Creates longitudinal magnetization that mixes (population distribution changes via dipolar relaxation of excited spins) during mixing period, twhich varies with size
  • 3: Creates transverse magnetization from remaining longitudinal magnetization for detection.
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
23
Q

How does molecular size affect NOE?

A
  • Small molecules tumble rapidly (small τc) à +ve NOE
  • Vice versa with large/highly viscous molecules, giving – ve NOEs approaching 100%
  • Certain NOE’s may have zero intensity
  • Can be avoided by changing T or viscosity
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
24
Q

How can NOESY data be used in combination to TOCSY spectra

A
  • Spectra can be overlayed on to one another, the differences of which give an indication of 1Hs that are close in space but not in bonds
  • Namely, the coupling between residues in the fingerprint region, walking from one residue to another via NH-Ha dipolar coupling
25
Q

Give a solution to the size effect problem in NOESY.

A
  • Use ROESY (rotating frame Overhauser effect spectroscopy)
  • Operates in same plane (XY) and through space to give map of dipolar coupled 1Hs
  • Always gives a + ve value and leads to increase in signals regardless of size
26
Q

Why is NOESY useful?

A
  • Can determine 1Hs distant in sequence but close in space
  • This is key for 3D structure determination.
  • Can go past C=O in peptide and transmit information between AA sidechains
29
Q
  • Assigning CSI data
    • If Δ𝛿 > |0.1| residue assigned +1 or -1
    • If Δ𝛿 < |0.1| residue assigned value of 0
    • Stretch of 4 or more +ve/-ve values gets assigned as a helix/sheet
    • Mixed /0 assigned as random coil
    • 75% accuracy if only Hα used
    • >90% accuracy if 13C considered
A
32
Q

What is a key difference in the process of NOESY vs ROESY?

A
  • NOESY: spin-lattice relaxation (T1)
  • ROESY: spin-spin relaxation (T2)
33
Q
  • CSI or … … … method can be used to identify structure type e.g. α-helices, …-strands, … …
  • Attained from … … … of the backbone.
  • AA - 1Hα in α-helices shifted relative to RC values.
  • AA - 1Hα in β-sheets shifted relative to RC values.
A
  • CSI or chemical shift index method can be used to identify secondary structure type e.g. α-helices, β-strands, random coil.
  • Attained from sequential peak assignment of the backbone.
  • AA - 1Hα in α-helices shifted upfield relative to RC values.
  • AA - 1Hα in β-sheets shifted downfield relative to RC values.
34
Q

What is secondary chemical shift and what does it depend on?

A
  • Δ𝛿 = 𝛿obs – 𝛿RC
  • Depends on the secondary chemical structure
  • Hα secondary shift varies +/- 0.4 ppm
  • Cα ranges form + 2.6 (helix) -1.4 ppm (sheet)
  • Cβ ranges from – 0.4 (helix) -2.2 ppm (sheet)
35
Q

Why may experimentalists need heteronuclear 2D NMR to characterise a biomolecule fully

A
  • Above 3-4 kDa, homonuclear 1H NMR becomes very crowded, leading to large amount of overlap in spectra.
  • Instead a heteroatom X can be used as a second dimension
  • E.g. 13C or 15N
36
Q

Heteronuclear 2D NMR requires isotopic labelling, what are two methods in which this can be done?

A
  • Synthetic peptide preparation: labels added to specific AA’s during peptide synthesis, very expensive
  • Uniform labelling with bacteria: use feedstocks of E.coli to label proteins. Handles 15N well, but 13C gives a lower yield.
37
Q
  • In 1H-15N HSQC, every -bearing yields a peak (now labelled )
  • Every AA in backbone + additional AA’s with …-bearing
  • Polarisation transfer via (through ) from 1H spins (high γH) to spins with low γx
  • Enhances signal from … …nuclei.
A
  • In 1H-15N HSQC, every H-bearing N yields a peak (now labelled 15N)
  • Every AA in backbone + additional AA’s with N-bearing sidechains
  • Polarisation transfer via J-coupling (through bonds) from 1H spins (high γH) to spins with low γx
  • Enhances signal from low-sensitivity heteronuclei.
39
Q

When would 1H-13C HSQC be used?

A
  • For biomolecules with few N atoms, otherwise use 1H-15N
  • Isotopic enrichment with 13C can lead to lower yields.
40
Q

What are hybrid techniques?

A
  • Once a good HSQC is obtained, it can be used in combination with a homonuclear 2D method (e.g. TOCSY) as an additional filter
  • For example, combine with a 1H-1H TOCSY spectrum to see only peaks corresponding to that N chemical shift (CS), through addition of a 3rd dimension
  • HSQC can form hybrids before or after TOCSY/NOESY experiments.
41
Q

Describe an assignment strategy for a 1H/15N only system.

A
  • Acquire 1H-15N HSQC, HSQC-TOCSY, and HSQC-NOESY
  • In HSQC-TOCSY, obtain a 1H-1H TOCSY at each 15N shift of interest, using the strips of peaks gained from each different ppm to find patterns in data.
  • Use HSQC-NOESY (1H-1H NOESY at each 15N shift) to walk down protein backbone, collecting data at same ppm as previous to overlay differences (same as before but now with a nitrogen filter)
  • Plot each assignment back on original 1H-15N as you go.
42
Q
  • … … is sensitive to nuclear environment, for example a binding interaction will cause the protein site’s … … to change in response.
A
  • Chemical shift is sensitive to nuclear environment, for example a binding interaction will cause the protein binding site’s chemical shift to change in response.
43
Q

What sort of interactions cause a change in chemical shift (CS)?

A
  • Changes caused by covalent (bonds forming) and non-covalent (H-bonding, vdW) interactions are sensitive probes for folding, ligand binding.
44
Q

How would we interpret chemical shift perturbations in a spectrum?

A
  • As [ligand] increases, chemical shifts of nuclei at/near binding site affected
  • Given as a weighted change Δtotal
  • More smeared peak from blue to red show CS with increased concentration, indicating direct involvement with binding, or indirect conformational change due to binding.
45
Q

Discuss the drawbacks of using Chemical shift perturbation, and what types of interactions it is best suited for.

A
  • No direct way to know if interaction if due to direct binding or structural rearrangement upon complex formation
  • Spread of data gives an indication of which it is (tight –> binding site, diffuse –> conformational change)
  • This approach is best suited to weak interactions in low mM range showing clear fast exchange, where we know the old and new peak location and how they are related.
46
Q

What sort of chemical shift perturbations might we see in regions of our spectrum’s peaks in a ligand binding event?

A
  • No change (no binding)
  • Peak shift (binding, fast exchange)
  • Peak splitting (new conformers formed)
  • Peaks appear-disappear (binding; slow exchange)
49
Q

What is Deuterium exchange NMR and how can it be used in HSQC?

A
  • In a high pH solution of D2O, an undeuterated protein will undergo rapid H/D exchange.
  • Only H’s participating in H-bonds (amide hydrogen) and those in the centre of the fold (protected) will not exchange
  • Deuterium does not show up in proton spectra so in 1H-15N HSQC will see surface H’s peaks vanishing
  • Great chemical probe of exposed surface of protein to see what AA’s are available
50
Q
  • As protein size …, 1H-15N/1H-13C experiments become and difficult to interpret.
  • Can add additional …-… to further delineate residues based on chemical shift, giving resonant experiments.
  • E.g. a 3D HNCA experiment correlates H,N and Cα, where any α-carbons, will be directly above a corresponding HN shift
A
  • As protein size increases, 1H-15N/1H-15N hybrid experiments become crowded and difficult to interpret.
  • Can add additional 13C-filter to further delineate residues based on 13C chemical shift, giving triple resonant experiments.
  • E.g. a 3D HNCA experiment correlates H,N and Cα, where any α-carbons, will be directly above a corresponding HN shift
51
Q

Describe the CBCANH 3D correlation method

A
  • Cα are positive
  • Does not tell us which residue is i and which is i-1
52
Q

Describe the CBCA(CO)NH 3D correlation method

A
  • Always paired with CBCANH, to determine difference between residues i/i+1
  • Strips of these resolved peaks can be used to sequentially assign backbone (not identity of AA though)
  • Allows magnetisation only through C=O carbon
53
Q

Describe the HNCA/HN(CO)CA pair that could be run to reinforce speculations made so far.

A
  • HNCA: In each NH strip, 2 Cα peaks visible from residue i and i+1
  • HN(CO)CA: In each NH strip, 1 Cα visible from preceding residue (i-1)
54
Q

Describe the HNCO/HN(CA)CO pair that could be run to reinforce speculations made so far.

A
  • HNCO: In each NH strip, 1 CO visible from residue i-1
  • HN(CA)CO: In each NH strip, 2CO peaks visible from residue i and i-1
55
Q
  • What 2D experiments should be done to assign side chains of our protein first?
A
  • HCCH-TOCSY: At each 13C chemical shift, see 1H atoms attached to given C and all other Cs in side chain
  • HCCH-COSY: At each 13C chemical shift, see 1H atoms attached to given C and neighbouring Cs only.
  • Nitrogen methods not used as not present in most sidechains
56
Q

What additional experiments can be done to assign side chains of our protein?

A
  • Use CC(CO)NH and H(CCO)NH in combination with our HCCH-TOCSY/COSY
  • CC(CO)NH: in each NH strip, see all C nuclei in sidechain of residue i-1
  • H(CCO)NH: in each NH strip see, all H nuclei in sidechain of residue i-1
62
Q

What response is generated from plotting STD AF values vs [ligand] in STD NMR

A

Michaelis-Menten hyperbolic dose-response

64
Q

How does sensitivity vary throughout all the experiments described?

A
  • HNCA/HNCO have high sensitivity, despite less chemical information being given as a result of less filtering
  • HNCA gives poor resolution in 3D due to signal loss through added filter via relaxation effects.
  • Both have high signal:noise ratio, due to lots of acquisitions occurring throughout 3D data generation
  • 2D HSQC therefore takes a lot less time than these long 3D experiments.
65
Q

For proteins >25kDa, the methods discussed so far do not suffice. What is this due to and how can we get around it?

A
  • Broadening in peaks as we move to higher mass systems is the result of a shorter T2 associated with spin-spin relaxation, making resolution unmanageably poor to take in to 3D.
  • All Hs interacting due to spin-spin relaxation as well as the facilitation of spin relaxation of neighbouring C/N atoms is a problem that increases with molecular weight.
  • Circumvent this by deuterating our sample, reducing the proton density, slowing relaxation due to dipole-dipole interactions and CSA of quadrupolar deuterium
66
Q

Deuteration can be combined with TROSY to determine structure of large proteins, what is TROSY and how does it do this?

A
  • Transverse relaxation optimized spectroscopy selects the sharpest component where, DD and CSA relaxation mechanisms cancel
  • Less overlapping in this peak, allowing increased spectral resolution
  • Increases sensitivity
  • Extends mass range
67
Q

As well as uniform isotopic labelling, site-selective labelling can be employed, what is this? give a few examples

A
  • expensive/time consuming process
  • gives spectral simplification
  • good probe for structrual/dynamics studies
  • sequence-specific resonance assignments for peaks in crowded regions
  • amino acid selective labelling
  • methyl-specific protonation
68
Q

If a protein is unassignable, what is an alternative method that could be used to understand ligand binding dynamics

A
  • structural assignment not goal, so can use STD NMR to investigate binding constant of ligand binding
  • Titration series of increasing enzyme concentration done where change in intensity of ligand signals detected via NOE
  • Amplification factor calcuated at each [ligand] - describes how much peak impacted by ligand presence.