Skip to content

NMR for Peptide Structure Determination

Nuclear magnetic resonance (NMR) spectroscopy provides atomic-resolution structural information for peptides in solution. Unlike X-ray crystallography, NMR determines structure under near-physiological conditions, revealing dynamic properties and solvent interactions.

NMR measures the magnetic properties of nuclei with non-zero spin quantum numbers:

NucleusSpinNatural AbundanceSensitivityApplication
¹H1/299.98%1.00Primary detection
¹³C1/21.11%0.016Backbone/side chain
¹⁵N1/20.37%0.001Backbone assignment
²H10.015%0.010Solvent suppression

The resonance frequency of a nucleus depends on its electronic environment:

  • Shielded: Higher electron density → lower frequency (upfield)
  • Deshielded: Lower electron density → higher frequency (downfield)
  • Chemical shift (δ): Referenced to TMS (0 ppm) or DSS (0 ppm)

A typical peptide ¹H NMR spectrum shows:

Region (ppm)Assignment
10–12Trp indole NH, Arg guanidinium
8.0–8.5Amide NH (α-helix: ~8.0; β-sheet: ~8.5)
7.0–7.5His C2-H, C4-H
6.5–7.5Tyr aromatic, Phe aromatic
5.5–6.0Trp aromatic
4.5–5.0α-H (α-helix: ~4.0; β-sheet: ~4.5)
3.5–4.5β-H, side chain CH₂
2.0–3.0Lys ε-CH₂, Arg δ-CH₂, Met CH₃
0.8–1.5Val, Leu, Ile methyl groups

The amide region (8.0–8.5 ppm) provides secondary structure information:

  • α-Helix: Narrow dispersion, NH at ~8.0 ppm
  • β-Sheet: Wide dispersion, NH at ~8.5 ppm
  • Random coil: Narrow dispersion, NH at ~8.3 ppm

TOCSY identifies amino acid spin systems through scalar (J) coupling:

Cross-peaks: Connect protons within the same amino acid residue

Amino AcidCharacteristic TOCSY Pattern
GlySingle cross-peak (α-H₂)
AlaDoublet at 1.4 ppm (β-CH₃)
ValTwo doublets at 0.9–1.0 ppm
Leu/IleTwo doublets at 0.8–0.9 ppm
LysThree cross-peaks (β, γ, δ, ε)
ArgThree cross-peaks (β, γ, δ)
ProCharacteristic α-H at 4.5 ppm

Experiment Parameters:

  • Mixing time: 60–80 ms (spin lock)
  • Sweep width: 12–14 ppm (¹H)
  • Data points: 2048 × 512
  • Solvent: D₂O or 90% H₂O/10% D₂O

NOESY (Nuclear Overhauser Effect Spectroscopy)

Section titled “NOESY (Nuclear Overhauser Effect Spectroscopy)”

NOESY identifies spatially close protons (<5 Å) through dipolar coupling:

Cross-peaks: Connect protons close in space (not necessarily bonded)

Sequential Assignment Strategy:

  1. Identify dαN(i, i+1) connectivities: α-H(i) → NH(i+1)
  2. Identify dNN(i, i+1) connectivities: NH(i) → NH(i+1)
  3. Build chain from N-terminus to C-terminus

Characteristic NOE Patterns:

StructureNOE Pattern
α-helixStrong dNN(i, i+1), weak dαN(i, i+1)
β-sheetStrong dαN(i, i+1), weak dNN(i, i+1)
TurndNN(i, i+1), dαN(i+2, i+3)

COSY identifies scalar-coupled protons (2–3 bonds):

  • α-H → β-H: Within each residue
  • β-H → γ-H: Side chain assignment
  • Less useful than TOCSY for peptides (limited to J-coupled networks)

HSQC (Heteronuclear Single Quantum Coherence)

Section titled “HSQC (Heteronuclear Single Quantum Coherence)”

HSQC correlates ¹H with directly bonded ¹³C or ¹⁵N:

¹H-¹⁵N HSQC:

  • Each amide N-H shows one cross-peak
  • Fingerprint of the peptide
  • Used for backbone assignment and monitoring folding

¹H-¹³C HSQC:

  • Correlates ¹H with directly bonded ¹³C
  • Side chain assignment
  • Methyl groups appear in aliphatic region
  1. Identify amino acid types: TOCSY
  2. Sequential assignment: NOESY (dαN, dNN connectivities)
  3. Side chain assignment: TOCSY + COSY + HSQC
  4. Stereospecific assignment: NOE patterns, J-coupling
ResidueTOCSY Signature
Glyα-H₂ at ~3.9 ppm
Alaβ-CH₃ doublet at 1.4 ppm
Valβ-H doublet at 2.1 ppm; γ-CH₃ doublets
Leu/Ileβ-H multiplet; δ-CH₃ doublets
Proα-H at 4.5 ppm; no amide NH
Lysε-CH₂ at 3.0 ppm
Argδ-CH₂ at 3.2 ppm
  1. Start at N-terminus (dαN(1,2))
  2. Walk through sequence using dαN(i, i+1)
  3. Verify with dNN(i, i+1) and dβN(i, i+1)
  4. Use prolines as break points (no amide NH)
  1. From α-H assignment, use TOCSY to find side chain protons
  2. Use COSY to connect within spin systems
  3. Use ¹³C HSQC for carbon chemical shifts

Compare observed α-H shifts to random coil values:

DeviationInterpretation
Δδ > 0.1 ppm upfieldα-helix
Δδ < −0.1 ppm downfieldβ-sheet
Δδ

From NOESY cross-peak intensities:

NOE IntensityDistance Range (Å)
Strong1.8–2.5
Medium1.8–3.5
Weak1.8–5.0

Distance Geometry:

  • Convert NOE distances to 3D coordinates
  • Algorithms: DGSA, DGS, AMBER

Simulated Annealing:

  • Minimize energy with NOE restraints
  • Programs: X-PLOR, CNS, CYANA, ARIA

Molecular Dynamics:

  • Refine structures with explicit solvent
  • Programs: GROMACS, NAMD, AMBER
MetricGoodAcceptable
RMSD (backbone)<0.5 Å<1.0 Å
RMSD (all heavy)<1.0 Å<1.5 Å
NOE violations<0.1 Å<0.3 Å
Ramachandran favored>90%>80%
Ramachandran allowed>98%>95%
PROCHECK score>−0.5>−1.0
ParameterRequirement
Concentration0.5–2 mM (optimal)
Volume300–600 µL
SolventD₂O or 90% H₂O/10% D₂O
pH4–7 (exchange-dependent)
Temperature25°C (standard)
Salt50–100 mM NaCl
Isotopic labeling¹⁵N, ¹³C for large peptides (>15 aa)
  • Amide exchange: Broadens NH peaks at high pH
  • Optimal pH: 4–5 for most peptides
  • D₂O exchange: Progressive loss of NH signals
  • H₂O/D₂O: Use 90% H₂O/10% D₂O to preserve NH signals
TemperatureEffect
5°CSlower exchange, sharper peaks
25°CStandard
37°CPhysiological, faster exchange
50°CReduced viscosity, broader lines
  • Chemical shift analysis: CSI method
  • J-coupling: ³J_{HNα} values (helix: ~4 Hz; sheet: ~9 Hz)
  • NOE patterns: dαN vs dNN connectivity patterns
  • Hydrogen bonding: H/D exchange experiments
  • Relaxation (T₁, T₂): Backbone dynamics
  • Exchange rates: Conformational flexibility
  • μs-ms dynamics: Conformational exchange (CPMG, R₁ρ)
  • ps-ns dynamics: Internal motions (Model-free analysis)
  • Chemical shift perturbation: Map binding interface
  • Saturation transfer difference (STD): Ligand binding
  • WaterLOGSY: Ligand screening
  • Relaxation dispersion: Binding kinetics
  • Size limit: >50 aa requires isotopic labeling and TROSY
  • Conformational exchange: Broadens lines, obscures peaks
  • Overlap: Sequence-specific assignment challenging
  • Time: Full structure determination requires weeks
  • Cost: NMR spectrometer time is expensive
  1. Wüthrich K. “NMR of proteins and nucleic acids.” Wiley 1986.
  2. Cavanagh J, et al. “Protein NMR Spectroscopy.” Academic Press 2007.
  3. Williamson MP. “Using chemical shift perturbation to characterise ligand binding.” Prog Nucl Magn Reson Spectrosc 2013;73:1-16.
  4. Marion D. “Introduction to biological NMR spectroscopy.” J Biomol NMR 2013;55:303-310.
  5. Pervushin K, et al. “TROSY: a new approach for high-resolution protein NMR.” J Biomol NMR 1998;12:345-353.