Using CD Spectroscopy to Evaluate Protein Thermal Stability

Protocol · Far-UV circular dichroism · Ellipticity at 222 nm · Tm fitting · Reversibility

Published 2026 · Last updated 2026-09-28 · Hailu Zhou

Protein thermal stability is a key indicator of structural integrity and functional retention, and it is widely applied in basic research, protein engineering, and biopharmaceutical development. Circular dichroism (CD) spectroscopy is a sensitive, rapid, and low-sample-volume technique commonly used to monitor conformational changes in protein secondary structure. It is particularly suitable for evaluating thermal denaturation behaviour and thermal stability parameters such as the melting temperature (Tm). This protocol describes how to run the experiment, what the spectrum is actually reporting, and where the interpretation goes wrong.

The role of CD spectroscopy in protein thermal stability

CD spectroscopy measures the difference in absorption of left- and right-circularly polarised light by chiral molecules. For proteins, the secondary structures — α-helix, β-sheet, and disordered regions — generate characteristic signals in the far-UV region (190–250 nm). An α-helix typically shows two negative bands near 208 and 222 nm; a β-sheet shows a negative band around 216 nm; and disordered structure gives a weak negative signal near 195–200 nm. These signatures are the basis of everything that follows.

When increasing temperature induces thermal denaturation and disrupts the native ordered structure, the characteristic CD signals change markedly. By monitoring ellipticity at a chosen wavelength — most often 222 nm — as a function of temperature, one can track the unfolding process and calculate Tm to quantify thermal stability. The technique is label-free, requires only micrograms of protein, and reports directly on secondary-structure content, which is why it remains the workhorse method for conformational stability assessment alongside differential scanning fluorimetry and calorimetry.

Experimental design and operational key points

1. Protein sample preparation

Sample purity and homogeneity are critical. Proteins of >90% purity are recommended, and interference from aggregates or polymorphic species should be avoided. The buffer should be thermally stable and exhibit low UV absorbance; commonly used systems include PBS and HEPES, while temperature-sensitive buffers such as Tris should be avoided. Protein concentration is generally set to 0.1–0.5 mg/mL to ensure an adequate signal-to-noise ratio; far-UV measurements typically use a 0.1–1 mm path-length cuvette to keep absorbance within the linear range of the detector.

2. Wavelength selection

Thermal-denaturation experiments typically monitor 222 nm because it is most sensitive to changes in α-helical content. For proteins rich in β-sheet structure, the signal at 216 nm can also be informative. Acquiring a full far-UV CD spectrum (190–250 nm) before and after denaturation facilitates structural comparisons and helps assess the direction and extent of the unfolding process. Recording at a single wavelength throughout the ramp keeps acquisition fast; recording full spectra at intervals costs time but preserves more information about intermediate states.

3. Temperature range and heating rate

Temperature commonly starts at ~10 °C and is increased to 90–95 °C, adjusted according to the expected stability of the protein. A heating rate of 0.5–1 °C/min is recommended to maintain thermal equilibrium and minimise bias; faster ramps shift the apparent Tm upward through thermal lag. Recording a data point every 1–2 °C yields a smooth temperature–ellipticity curve. Equilibration at each point (typically a few seconds) improves reproducibility at the cost of total run time.

4. Reversibility test

In some studies, it is important to assess whether thermal denaturation is reversible. After the heating scan, slowly cool the sample back to the initial temperature and re-collect the CD spectrum for comparison with the preheating spectrum. If the structure is restored, the process exhibits a degree of reversibility, which is relevant for evaluating native proteins with recoverable conformations or certain engineered protein designs. Irreversibility is common when unfolding is coupled to aggregation; the re-scan then reports the aggregate state rather than the refolded protein.

Data interpretation and Tm calculation

CD thermal-denaturation data typically yield a sigmoidal (S-shaped) curve: initially the structure is stable and ellipticity remains constant; with increasing temperature the protein progressively unfolds and ellipticity decreases sharply; finally a plateau is reached, indicating the fully denatured state. The midpoint of that transition is the melting temperature.

Tm extraction. Tm is the central parameter for thermal stability, defined as the temperature at which half of the ordered structure is lost. Nonlinear fitting with a two-state transition model — which assumes a direct conversion between the native and denatured states — is commonly applied to thermal-denaturation curves. Such fitting provides an accurate Tm together with parameters describing transition cooperativity (e.g. the van 't Hoff enthalpy), thereby quantifying the unfolding process. For multi-domain proteins or proteins exhibiting multi-state unfolding, multiple Tm values may be involved; these should be inferred from the curve shape and the positions of peaks and inflection points, and multi-wavelength analysis can be used to improve resolving power.

Common experimental issues and optimisation

  1. Signal fluctuations or discontinuities. These may arise from buffer interference, protein aggregation, or bubbles. Degas the buffer, use suitable quartz cuvettes, and complete measurements soon after sample preparation.
  2. Poor repeatability of Tm. Verify the stability of the heating system and check for sample degradation or buffer failure. Test each sample at least three times to ensure repeatable Tm estimates.
  3. Incomplete unfolding or no apparent transition. The protein may be extremely stable, with a Tm beyond the measurement range, or highly flexible with weak thermal induction. In such cases, extend the temperature range, adjust the buffer environment, or consider combining thermal and chemical denaturation (e.g. urea-induced unfolding).
  4. High-tension (HT) voltage drift. If the photomultiplier HT signal rises sharply, the sample is absorbing too much light; reduce the path length or concentration, or narrow the wavelength window before trusting the data.

Reporting and interpretation

Report Tm as the mean ± standard deviation over at least three independent scans, and state the buffer, protein concentration, path length, wavelength monitored, heating rate, and fitting model used. Note whether unfolding was reversible and whether the HT voltage stayed within range. CD reports secondary structure, not tertiary detail: a preserved Tm with changed spectra can indicate a different folded state, and a shifted Tm in a mutant is evidence about stability only when the sample is confirmed to be homogeneous and correctly folded at the starting temperature. Interpret CD alongside orthogonal methods — DSF, DSC, or NMR — whenever a stability claim will carry weight.

References

  1. Greenfield NJ. Using circular dichroism spectra to estimate protein secondary structure. Nat Protoc. 2006;1(6):2876–2890. doi:10.1038/nprot.2006.202
  2. Greenfield NJ. Using circular dichroism collected as a function of temperature to determine the thermodynamics of protein unfolding and binding interactions. Nat Protoc. 2006;1(6):2527–2535. doi:10.1038/nprot.2006.204
  3. Kelly SM, Jess TJ, Price NC. How to study proteins by circular dichroism. Biochim Biophys Acta. 2005;1751(2):119–139. doi:10.1016/j.bbapap.2005.06.005
  4. Kelly SM, Price NC. The application of circular dichroism to studies of protein folding and unfolding. Biochim Biophys Acta. 1997;1338(2):161–185. doi:10.1016/s0167-4838(96)00190-2
  5. Benjwal S, Verma S, Röhm KH, Gursky O. Monitoring protein aggregation during thermal unfolding in circular dichroism experiments. Protein Sci. 2006;15(3):635–639. doi:10.1110/ps.051917406
  6. Schaub TE, Campbell SG, Whitten ST, et al. Thermal unfolding of the N-terminal region of p53 monitored by circular dichroism. Protein Sci. 2012;21(11):1682–1688. doi:10.1002/pro.2146
  7. Seelig J, Schönfeld HJ. Thermal protein unfolding by differential scanning calorimetry and circular dichroism spectroscopy: two-state model versus sequential unfolding. Q Rev Biophys. 2016;49:e9. doi:10.1017/s0033583516000044
  8. Shanmugam G, Polavarapu PL. Vibrational circular dichroism spectra of protein films: thermal denaturation and self-aggregation. Biophys Chem. 2004;111(1):73–77. doi:10.1016/j.bpc.2004.04.005