NMR Techniques in Protein Analysis

Nuclear Magnetic Resonance (NMR) spectroscopy is a powerful technique for studying the structure and dynamics of proteins.
NMR peak positions are highly sensitive to:
- Chemical modifications
- Conformational changes
- Interactions with other molecules

Nuclei closest to a site of change exhibit the greatest shift in chemical shift (measured in ppm).
In cases of global changes, such as protein unfolding, most NMR signals typically undergo alterations.
Challenges in 1D 1H NMR spectra:
- Crowded spectra: High number of signals makes it difficult to map changes directly to specific resonances.
- A need for spectral resolution requires a separate technique to manage peak overlap, which can be achieved by using multiple dimensions.

Utilizing isotopes 15N and 13C alongside 1H allows for:
- Less crowded 2D and 3D NMR data sets to be acquired.
- Spread of peaks across two or more dimensions decreasing overlap among peaks.
Isotopic enrichment: 15N and/or 13C labeling of proteins is necessitated for effective study.

Proteins can be enriched in 15N and/or 13C by recombinant expression under conditions where only 15N and 13C sources are available.
Commonly performed in E. coli under minimal medium conditions:
- 15N ammonium sulfate: $(15NH4)2SO_4$ at 0.6g/l.
- 13C glucose: $C6H{12}O_6$ at 2g/l.
Includes components in the phosphate buffer:
- Sodium sulfate, sodium chloride, magnesium sulfate, calcium chloride, and trace amounts of:
- Manganese chloride, iron chloride, zinc chloride, copper chloride, cobalt chloride, boric acid, biotin, thiamine.
Typical enrichment levels of 95-98% verified by mass spectrometry, without chemical differences between labeled and natural proteins.
E. coli utilizes 13C glucose $ (C6H{12}O6) $ and $(15NH4)2SO4 $ as the sole carbon and nitrogen sources.

2D Heteronuclear NMR spectrum: The most commonly used is HSQC (Heteronuclear Single Quantum Correlation).
HSQC correlates 1H with its directly bonded 15N or 13C atom, with a single peak representing each residue except certain cases (e.g., proline).

Example protein: ERp27 (14 kDa).
- NMR experiments:
- 1D 1H NMR
- 2D 1H, 15N-HSQC
Amide (NH) region: Frequencies separated based on the resonance frequency of the 15N nucleus; peak intensities displayed as contours.
Typically only the left-hand side of the spectrum is shown since aliphatic 1H does not correlate to 15N.

Example sequences from a 20-residue peptide:
- 1GFTTGRRGDL
- 11ATIHGMNRPF
Key observations:
- Each NH correlates to a single peak in the spectrum.
- Sidechains such as Asn (NH2) and Arg (NH) show corresponding peaks as well.
- The N-terminal NH3+ often does not yield a peak.
- No peak observed for proline due to lack of backbone NH.
Data points indicating chemical shifts in ppm along with specific residues:
- 5G, 15G, 8G, 12T, 4T, 3T, 17N, 16M, 9D, 6R, 14H, 7R,10L, 13I, 11A, 2F, 20F, 18R.

Utilization of 1H, 15N-HSQC NMR spectra to observe chemical changes under varying conditions (e.g., oxidative vs. reducing environment):
- Differential peaks assigned to residues help in identifying chemical changes.
Catalytic site investigation in Proteins Disulphide Isomerase (PDI):
- Contains two cysteine residues that can exist as free thiols or in a disulphide bond state.

Overlay of HSQC spectra illustrates peak shifts:
- Red represents reduced conditions, blue for oxidised.
Peak characteristics observed:
- Unaffected peaks
- Slight shifts in peaks
- Significant shifts observed in:
- 33A, 36C, 35W, 26Y, 90N, and 32Y.

Chemical shift change measured in ppm is defined as:
- $ext{Chemical shift change} = ext{sqrt}\bigg((\bigtriangleup 1H)^2 + \bigg( rac{\bigtriangleup 15N}{8} \bigg)^2 \bigg)$
- The dimension of 15N is scaled down by a factor of 1/8 to correct for the larger ppm range.
Largest observed chemical shifts correlated to the catalytic site within the domain.
Two cysteine residues present at this site, discerned through examining both oxidised and reduced forms.

Analysis using 1H, 15N-HSQC NMR spectra to identify sites of ligand binding in the b' domain of PDI by increasing peptide concentration.
Structural analysis showing regions of significant chemical shift changes highlighted in yellow and red.
Binding site dislocation when a peptide ligand interacts with the protein:
- The binding site occupies the linker region between the b' and a' domains, indicated in grey.

A molecule in solution undergoes tumbling, measured by:
- Correlation time (τm): The time for the molecule to tumble through 57.3° (1 radian).
NMR peak width is directly proportional to τm, influenced by:
- Volume of a sphere ($V$)
- Viscosity of the fluid ($ ext{η}$)
- Boltzmann constant ($K_b$)
- Temperature in Kelvin ($T$)
Peak widths broaden when:
- Higher molecular weight ($MW
  earrow$)
- Viscosity increases ($ ext{η}
  earrow$)
- Temperature decreases ($T
  earrow$)
Note: Narrowest peaks obtained from small proteins in low viscosity buffers at high temperatures.

NMR spectra exhibit sensitivity to chemical and structural changes in proteins.
The crowded nature of 1D 1H NMR spectra can obscure individual resonances due to excessive peaks (over 800 peaks).
Heteronuclear NMR techniques provide simplified spectra with fewer overlaps.
In 1H, 15N-HSQC, each residue shows as a single peak, providing clearer insights (exceptions: N-terminus and proline).
2D HSQC spectra facilitate tracking and mapping of protein changes.
Increasing molecular weight broadens peaks, correlating with slower molecular tumbling.
Higher temperatures can enhance tumbling but also risk denaturation and instability.