BenchMasteryMolecular biology techniques & virtual lab

NMR spectroscopy

The ¹H–¹⁵N HSQC shows one peak for each backbone amide, at the frequencies of its proton and nitrogen. It is the first spectrum of any protein NMR project and the readout of most titration and screening experiments.

chemical shiftsSimulator readyCore simulatorEngine: Binding, biophysics & structure data
Choose a sub-technique

Our ruleSimulations prepare you for the bench but don't replace hands-on practice.

Sub-technique you chose

HSQC

The ¹H–¹⁵N HSQC shows one peak for each backbone amide, at the frequencies of its proton and nitrogen. It is the first spectrum of any protein NMR project and the readout of most titration and screening experiments.

How it differs from the standard method

  • Needs only a ¹⁵N-labelled sample at 50 µM–1 mM; minutes to an hour on a cryoprobe.
  • Dispersion from about 6.5 to 10 ppm in ¹H says the protein is folded.
  • Chemical-shift perturbations during a ligand titration map the binding site.
In this lesson · about 2 min read
  1. Listening to nuclei
  2. Why size matters
  3. Labelling makes protein NMR possible
  4. How it fails
  5. Key terms
  6. When to use it
  7. Sub-techniques

1Listening to nuclei

Some atomic nuclei, including ¹H, ¹⁵N and ¹³C, behave like tiny magnets. In a strong magnetic field they line up, and a radio-frequency pulse tips them over so that they precess and induce a signal in a coil.

HSQC

Fingerprint

Correlates each amide ¹H with its directly bonded ¹⁵N: one peak per residue (except proline). Well spread peaks mean a folded protein.

triple resonance

Assignment

Experiments such as HNCA and HNCACB link each amide to its own and its neighbour's carbons, walking along the backbone.

NOESY

Distances

Cross-peaks between protons closer than about 5–6 Å, through space, with intensity falling as the sixth power of distance.

2Why size matters

The signal decays as the molecule tumbles. Small proteins tumble fast and give sharp lines; larger ones tumble slowly, their transverse relaxation (R2) speeds up, the lines broaden and the signal fades during the transfer delays of every pulse sequence.

3Labelling makes protein NMR possible

Natural nitrogen is 99.6% ¹⁴N, which is useless for these experiments, and natural carbon only 1.1% ¹³C. Proteins are therefore expressed in minimal medium in which the only nitrogen source is ¹⁵NH₄Cl and, for assignment and structure work, the only carbon source is ¹³C-glucose. Deuterated media (D₂O, deuterated glucose) remove most protons to slow relaxation in large proteins.

4How it fails

  • Missing peaks from loops at high pH or temperature: amide protons exchange with water faster than the experiment can follow.
  • Broad, weak peaks: the protein is too large, aggregating, or the field was not locked and shimmed.
  • All peaks bunched between 7.8 and 8.5 ppm: the protein is unfolded.
  • More peaks than residues: more than one conformation, a mixture of species, or a degraded sample.
  • A compact structure that violates its own restraints: spin diffusion at long NOESY mixing times, or mis-assigned NOEs.

Key terms

Practise these as flashcards →
Chemical shift (ppm)

Resonance frequency relative to a reference, in parts per million of the spectrometer frequency; reports the local chemical environment.

HSQC

Heteronuclear single-quantum coherence: a 2D spectrum correlating each proton with the heteroatom it is bonded to.

τc

Rotational correlation time: how long a molecule takes to tumble. It sets the linewidth.

R2 and linewidth

Transverse relaxation rate; the linewidth in Hz is about R2/π.

NOE

Nuclear Overhauser effect: transfer of magnetisation between protons through space, proportional to r⁻⁶.

Mixing time

The delay in a NOESY during which NOEs build up; too long and magnetisation relays through intermediate protons (spin diffusion).

TROSY

An experiment that selects the slowest-relaxing component of the signal, which together with deuteration makes large proteins accessible.

Lock

The spectrometer's feedback on the deuterium signal of D₂O that keeps the magnetic field stable during long experiments.

When to use it

  • Small proteins or domains (up to about 30 kDa) that do not crystallise, or whose dynamics matter as much as the structure.
  • Mapping where a ligand or partner binds by following which HSQC peaks shift during a titration.
  • Checking quickly whether a construct is folded, and whether mutants keep the same fold.
  • Studying intrinsically disordered regions, which no crystal or map can show.

Compared with alternatives

X-ray crystallography →

Higher resolution for proteins of any size if they crystallise, but a static picture without dynamics.

Cryo-EM →

The method of choice for large complexes; little help below about 50 kDa.

Circular dichroism →

Seconds of measurement to see whether a protein has secondary structure, but no residue-level information.

Sub-techniques

Choosing one here switches the whole page, including the protocol and the lab, to it.

The ¹H–¹⁵N HSQC shows one peak for each backbone amide, at the frequencies of its proton and nitrogen. It is the first spectrum of any protein NMR project and the readout of most titration and screening experiments.

  • Needs only a ¹⁵N-labelled sample at 50 µM–1 mM; minutes to an hour on a cryoprobe.
  • Dispersion from about 6.5 to 10 ppm in ¹H says the protein is folded.
  • Chemical-shift perturbations during a ligand titration map the binding site.