bioRxiv2022
The benefits of dyes in nucleic acid assays above room temperature are limited by a nonlinear, highdimensional relationship between fluorescence and the biophysical and chemical processes occurring in solution. To overcome these limitations, we identify an experimental regime that eliminates bias and unnecessary complexities in this relationship, and develop an experimental-computational workflow to generate the property data required to describe the dependence of fluorescence on temperature and concentration. Specifically, we exploit the temperature-cycling capabilities of real-time PCR machine, as well as the utility of numerical optimization, to determine the binding strength and molar fluorescence of the SYTO-13 dye bound to double-stranded (DS) or single-stranded (SS) DNA at more than 60 temperatures. We find that the data analysis approach is robust; it can account for significant well-to-well and plate-to-plate variation. The weak binding strength of SYTO-13 relative to SYBR Green I is consistent with previous reports of its negligible influence on PCR and melting temperature. Discriminating between molar fluorescence and binding strength clarifies the mechanism for the larger fluorescence of a DS/dye solution than a SS/dye solution; in fact, the explanation is different at high temperature than at low temperature. The temperature-dependence of the binding strength allows for ascertainment of the enthalpic and entropic contributions to the free energy, as well as the sign of the differential heat capacity of binding. The temperature-dependence of the molar fluorescence allows for calculation of the brightness (quantum yield times molar extinction coefficient) of SYTO-13 bound to DS relative to SS. The more accurate and complete description of the relationship between solution behavior and fluorescence enabled by this work can lead to more accurate selection of dyes and quantification of nucleic acids. SIGNIFICANCEFluorescent dyes are often used to quantify nucleic acids. The accuracy and precision of quantification, however, is limited by a complex and high-dimensional relationship between fluorescence and solution behavior. This is especially true for assays above room temperature, where empirical approximations are often required in the absence of available property data. In this work, we present an experimental and computational workflow that can more accurately describe this relationship and more efficiently generate the temperature-dependent thermodynamic and photophysical properties required. This approach can improve quantification and selection of high-performing dyes for particular assays.