Introduction
A short peptide in a vial rarely forms a crystal, yet researchers still want to know what shape it takes in water. Nuclear magnetic resonance (NMR) spectroscopy is the main laboratory method for answering that question in solution. This article is for research and educational purposes only; it explains how the method works, not how any compound should be used.
In the pages below we walk through the workflow from sample tube to structural model: preparing the sample, identifying each amino acid's signals, measuring which atoms sit close together, and turning those measurements into an ensemble of conformers. We also cover where peptide NMR gets difficult, because small and flexible molecules often resist a single tidy answer.
Why solution NMR suits peptides
In short: NMR looks at peptides the way they exist in a liquid, flexible parts included. Most samples in protein NMR are examined in water, and because molecules in solution are flexible, a set of conformations may represent the data better than one structure, as the overview of protein NMR spectroscopy points out.
Solution state versus crystal
Crystallography needs an ordered crystal. Many short peptides will not form one, and even when they do, the crystal captures a single packed arrangement. NMR needs only a dissolved sample, which is why it is described in the literature as the method for structure determination in solution. The trade-off is that it yields less information per structure: NMR models have historically been of lower quality than X-ray models, partly for that reason.
How NMR fits beside other quality checks
Routine analytical work answers narrower questions. Peptide mapping confirms identity through fragment patterns, and specific optical rotation probes chirality in bulk. NMR is a different kind of tool. It connects individual atoms to their neighbors, both through chemical bonds and through space, so it can describe local geometry rather than only confirm a fingerprint.
Where NMR sits among structure methods
The Protein Data Bank notes that most structures come from X-ray diffraction, while about 7% are determined by protein NMR. That is a minority share, but a steady one, and it covers exactly the flexible and small molecules that crystallize poorly.
Sample preparation and solvent choice
In short: a small, clean, stable sample in the right solvent decides whether the spectra are usable. For proteins, samples are typically 300 to 600 µL of highly purified material at 0.1 to 3 mM in a thin-walled glass tube. A published peptide protocol in the Journal of Visualized Experiments describes a lighter-weight setup: about 1 to 2 mg of peptide in 450 to 500 µL of deuterated solvent, either 10% D₂O in water or d₆-DMSO for peptides that do not dissolve in water.
Solvent and acidity shape the result. They influence which conformations are populated and how quickly amide protons exchange with the solvent, the same reason pH and buffer choice matter for stability in solution generally. Researchers also pick a temperature that minimizes overlap among the amide signals and keeps lines narrow, and they record a quick 1D spectrum between longer experiments to confirm the sample has not changed.
Peptide studies in the literature run in a wide range of media. One heptapeptide study ran at 30 °C in deuterated DMSO at about 26 mM, while macrocyclic peptide work has used solvent mixtures chosen to mimic membrane-like, aqueous and binding-site environments. The solvent is part of the experiment's meaning, and a structure measured in one medium does not automatically carry over to another.