Circular Dichroism of Peptides: What a CD Spectrum Can Tell a Lab
Most tests on a peptide certificate describe composition: the mass of the molecule, the order of its residues, the share of the sample that is the target. None of them says anything about shape. Circular dichroism fills that gap. For a peptide, a circular dichroism spectrum reports the average secondary structure the chain adopts in a particular solution: helix, sheet, turn or no fixed structure at all. It is a specialist method rather than a routine release test, but QC teams working with longer or disulfide-constrained sequences, or comparing structural behaviour across lots, benefit from knowing what it can and cannot show.
How the measurement works
Light can be circularly polarised in either a left-handed or a right-handed sense. Chiral molecules absorb these two forms by slightly different amounts. A CD instrument alternates between them and records the difference in absorbance across a range of wavelengths, usually expressed as ellipticity.
Peptides give a signal because every residue except glycine carries a chiral alpha carbon, and the backbone amide groups are arranged in a repeating pattern whose geometry depends on how the chain is folded. Change the fold and the pattern of unequal absorption changes with it. That is what makes the spectrum a readout of secondary structure. It also explains why a racemic mixture shows nothing: equal amounts of mirror-image forms cancel each other, which links the technique to questions of chiral purity.
Far UV versus near UV
Two wavelength regions are used, and they report different things:
| Region | Approximate range | What absorbs | What it reports |
|---|---|---|---|
| Far UV | about 190 to 250 nm | Backbone amide bonds | Secondary structure (helix, sheet, disorder) |
| Near UV | about 250 to 320 nm | Aromatic side chains (Trp, Tyr, Phe) and disulfide bonds | The local environment of those groups, i.e. tertiary packing |
Almost all peptide work uses the far UV. Near-UV signals are much weaker, need more material and only mean something when the chain packs into a defined three-dimensional arrangement, which most short peptides never do.
Recognising the classic signatures
Three reference patterns account for most far-UV spectra:
- Alpha helix: a strong positive band near 190 to 192 nm and two negative minima of similar depth near 208 and 222 nm. The 222 nm minimum is the one most often monitored because it is least affected by other structure types.
- Beta sheet: one broad negative band around 216 to 218 nm and a positive band near 195 nm, generally weaker and more variable than a helix pattern.
- Disordered (random coil): a pronounced negative band close to 198 nm, often with little else beyond a faint feature near 220 nm.
Measured spectra are usually blends. Deconvolution software fits the curve as a weighted combination of reference spectra and returns percentages of each structure type. Treat those percentages as model outputs, not direct measurements. Different mixtures of structures can produce curves that fit almost equally well, and the answer depends on the reference set chosen.
Getting a clean spectrum: sample requirements
The far UV is a crowded region. Many common buffer components, and oxygen in the light path, absorb strongly there, so a careless sample produces noise instead of data. A lab preparing samples for CD should plan for the following:
- Choose a transparent buffer. Chloride absorbs heavily below about 200 nm. Dilute phosphate or fluoride salts are common replacements.
- Consider the counter-ion. Trifluoroacetate, the usual counter-ion from peptide purification, also absorbs in the far UV and can interfere with measurement of the very peptide it accompanies. The trade-offs of salt exchange are set out in counter-ions and salt forms.
- Know the concentration precisely. Results are normalised per residue (mean residue ellipticity), so any concentration error flows straight into the reported structure content. An independent determination such as A280 with an extinction coefficient belongs in the measurement record.
- Use a short path length. Cells of 1 mm or less keep total absorbance manageable in the far UV.
- Keep the solution clear. Particles bend light away from the detector and warp the baseline, so a cloudy sample does not give a structural result; see peptide aggregation in solution.
- Purge the instrument with nitrogen so that oxygen does not absorb the shortest wavelengths.
- Record a buffer blank under identical conditions and subtract it.
Thermal melts and the “apparent” melting temperature
Following a single wavelength, typically 222 nm, while the temperature is raised in steps produces a melt curve. As a helix unfolds, the negative signal weakens. Where the curve is halfway between its folded and unfolded levels gives what is called the apparent melting temperature.
The word “apparent” carries weight. The value belongs to that peptide at that concentration in that buffer, not to the molecule in general. It has thermodynamic meaning only if cooling brings the original spectrum back. Many peptides aggregate as they are heated, giving an irreversible curve whose midpoint marks the onset of aggregation rather than unfolding. Running a return scan after cooling is the simplest way to tell the two apart.
Circular dichroism of peptides: the limits worth knowing
- It averages the whole sample. A result of “30% helix” could mean 30% of chains fully helical, every chain helical over 30% of its length, or any mix between. The spectrum cannot distinguish these.
- It has no positional information. CD cannot say which residues are structured. That requires NMR or crystallography.
- It is neither an identity nor a purity test. A deletion sequence can give the same spectrum as the full-length peptide, and a sample with plenty of unstructured impurities can still produce a tidy curve.
- Short peptides are usually disordered in water. A helix needs several turns before it is stable, and many short chains simply do not hold a shape in plain buffer. A disordered spectrum from such a peptide is expected, not a defect.
- Structure can be induced. Solvents such as trifluoroethanol, or membrane-mimicking detergents, can make a disordered chain look helical. That result describes the peptide’s tendency under those conditions and must always be reported with them.
When a bulk buyer might ask for CD data
For a typical short linear peptide, CD adds little to a reversed-phase chromatogram and a mass spectrum, which is why it rarely appears on routine certificates. It becomes informative for longer chains and disulfide-constrained sequences, where the right mass and the right connectivity still leave the fold open, a situation discussed in where peptide analysis becomes protein analysis.
It can also be useful for comparison. A lab that receives the same structured peptide in several lots can record a far-UV spectrum for each under fixed conditions and file it with the lot record. Overlaid spectra give a quick, independent check on structural consistency that complements the composition tests, in the spirit of the independent methods approach.
Bulk Peptides products are for in-vitro research and analytical use only and are not intended for human or veterinary use.
