What question does this answer? Did water-solubility engineering cost us folding stability? Tm is the hard number. The QTY code systematically replaces hydrophobic residues — leucine, isoleucine, valine, phenylalanine — with threonine, glutamine and tyrosine so that transmembrane proteins can live in water without detergent. Those substitutions improve solubility by construction, but a hydrophobic core is also what holds a fold together. It is entirely possible to gain solubility and pay for it in stability. Circular dichroism tells us about secondary-structure content; far-UV CD also gives a Tm, but nanoDSF gives a fast, label-free, high-throughput stability measurement on very little material, and it is the measurement we use to quantify the trade-off.
The principle: intrinsic fluorescence as a thermometer
nanoDSF requires no dye. It exploits the protein's own intrinsic tryptophan and tyrosine fluorescence and tracks how it changes while the sample is heated. In a folded protein, a tryptophan buried in the hydrophobic core is shielded from water and its fluorescence is relatively blue-shifted and intense. As the protein unfolds, that tryptophan becomes solvent-exposed; the emission spectrum shifts and changes in intensity because its microenvironment has changed. The instrument monitors two emission wavelengths and reports their ratio. This ratiometric readout cancels concentration and path-length variation, so the signal is robust. Plot the ratio against temperature and you get a sigmoidal denaturation curve; the inflection point is the melting temperature Tm, the temperature at which half the protein is unfolded. A sharp, cooperative transition is the signature of a well-defined folded structure.
Parameters and the reason for each
| Parameter | Value | Reason |
|---|---|---|
| Protein concentration | 5 µM | Enough intrinsic fluorescence signal from Trp/Tyr without aggregating; consistent across all variants for comparability |
| Buffer | PBS + 5 mM DTT | Matches the pipeline buffer; DTT keeps the 9–14 cysteines reduced so we measure unfolding, not disulfide cross-linking |
| Sample volume | 10 µL per capillary | Minimal consumption of purified protein — a key practical advantage |
| Capillaries | UV capillaries | Standard capillaries fluoresce in the UV and would swamp the Trp signal |
| Instrument | Prometheus NT.48 | Capillary-based nanoDSF with parallel channels for replicate measurements |
| Temperature ramp | 20 → 90 °C at 1 °C/min | Slow enough for equilibration at each point (near-equilibrium unfolding) and fast enough to be practical; 90 °C exceeds the expected Tm so the transition completes |
The negative control: prove the signal is real unfolding
A fluorescence ratio that changes with temperature is not automatically evidence of a folded protein — any environment shift moves a spectrum. So we include a control that can fail the measurement. A sample of the same protein is pre-heated to 90 °C for 15 min to fully denature it, then re-run. If the untreated protein gives a cooperative transition with a defined Tm, while the pre-denatured sample gives a flat, featureless trace with no transition, we have shown that the inflection in the intact sample genuinely comes from the cooperative unfolding of a folded structure rather than from an arbitrary signal drift. This mirrors the specificity logic used at every other stage of the pipeline: design the experiment so the artefact would look different from the result.
The DTT echo
It is worth noticing what the buffer choice repeats. The sample is still in PBS + 5 mM DTT, the same reducing condition used for purification and MST. This is not incidental. Because the cysteine counts across the variants span 12, 9, 10 and 14, a non-reducing buffer would allow disulfide-mediated aggregation whose melting profile would be meaningless. Keeping DTT present means the whole quantitative suite — binding and stability — is measured under one consistent reducing condition, so numbers from different assays are comparable.
Comparing variants and reading ΔTm
Tm is only interpretable comparatively. We measure the same construct alongside the native protein and alongside sibling QTY variants, all at 5 µM in the same buffer with the same ramp. Then ΔTm = Tm(variant) − Tm(reference) becomes the quantity of interest: a small negative ΔTm means the solubility substitutions cost little stability; a large negative shift means the fold was compromised. Replicates give the confidence interval, and a shift smaller than the replicate scatter is not a result. Reporting ΔTm rather than a bare Tm is what lets us answer the actual question: whether the water-soluble variant is still a properly folded protein, not merely a soluble one.
Multi-domain caveat
For single-domain proteins nanoDSF gives one clean transition. For multi-domain proteins and antibodies, unfolding is sequential and the thermogram shows multiple transitions; a single reported Tm is then ambiguous because it belongs to whichever domain unfolds at that temperature. The correct practice is to report each transition separately and assign the reported Tm to the domain it belongs to. Where a construct contains more than one folded unit, identify all transitions before quoting a number.
Troubleshooting and pitfalls
- No transition. A flat trace usually means the protein is already unfolded or aggregated before heating. Check the sample by SEC and confirm the negative control behaves as expected.
- Broad, low-amplitude transition. Often aggregation during heating or a poorly folded population; compare with a fresh, well-behaved batch and lower the concentration if aggregation is suspected.
- Irreversible aggregation and precipitation. Once denatured, many proteins aggregate; the post-transition baseline may rise or scatter. Interpret the transition itself, not the high-temperature tail.
- Bubbles or inconsistent fill. 10 µL capillaries are sensitive to trapped air, which distorts the fluorescence baseline. Fill carefully and inspect each capillary.
- Wrong capillary type. Ordinary capillaries fluoresce under UV and mask the intrinsic signal; always use UV-transparent capillaries.
- Avoidable disulfides. Omitting DTT from the buffer changes the thermogram through cross-linking rather than unfolding; keep the reducing condition constant.
- Over-interpreting a single number. A Tm without a reference, a replicate and an assigned domain is not a result; report ΔTm with its scatter.
Together with MST and far-UV CD, nanoDSF completes the functional picture: the variant binds its native ligand like the native receptor, and its Tm tells us how much folding stability the solubility engineering cost. Both numbers are needed — a stable but non-binding protein, or a binding but unfolded one, would each be a failure that the other assay alone could hide.
References
- Martin A, Maschberger M, Breitsprecher D. Thermal unfolding of antibodies with the Prometheus NT.48. BioTechniques. 2015;59(4):251. doi:10.2144/000114347
- Zhang S, et al. QTY code enables design of detergent-free chemokine receptors that retain ligand-binding activities. Proc Natl Acad Sci USA. 2018;115(37):E8652–E8659. doi:10.1073/pnas.1811031115
- Greenfield NJ. Using circular dichroism spectra to estimate protein secondary structure. Nat Protoc. 2006;1(6):2876–2890. doi:10.1038/nprot.2006.202
- 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
- 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
- Pace CN, Vajdos F, Fee L, Grimsley G, Gray T. How to measure and predict the molar absorption coefficient of a protein. Protein Sci. 1995;4(11):2411–2423. doi:10.1002/pro.5560041120
- Folta-Stogniew E. Oligomeric states of proteins determined by size-exclusion chromatography coupled with light scattering, absorbance, and refractive index detectors. Methods Mol Biol. 2006;328:97–112. doi:10.1385/1-59745-026-x:97