Depletion of wild-type target enhances the hybridization-based sensitivity of low-abundant mutation detection by reference capture probes

Nucleic acids duplex formation via hybridization is a crucial reaction in many processes and application across different disciplines. In life sciences the detection of mutations is an important application for which hybridization is used, e.g. in diagnostics via single-nucleotide variants (SNVs). This paper deals with the physicochemical aspects of hybridization-based detection of low-abundance mutations, which is challenging due to unavoidable competitive hybridization of high-abundant wild type sequence with the low-abundant variants. We apply two experimental methods based on theoretical hybridization models to show how sensing of DNA mutation can be significantly improved. This is implemented on two SNV biomarkers for which we first select a reference capture probe. This is a probe designed to match neither the wild type nor the SNV sequence, but to have an equal affinity to the wild-type as the SNV-matching probe. This allows the mutation-specific signal to be expressed as a ratiometric quantity, leading to increased assay robustness. Secondly, we selectively deplete the wild-type species by introducing an excess of wild-type-specific capture probes, and account for these depletion effects in the theoretical model. We demonstrate the detection of 0.05% mutant species in a wild-type background, which is an improvement of an order of magnitude in the limit of detection in comparison with the nodepletion case. This sensitivity is comparable with digital PCR results, showing performance suitable for e.g. clinical applications in liquid biopsy context. The principles of this work apply to a wide range of hybridizationbased DNA biosensing technologies, irrespective of the underlying transducer principle. This work was supported by the Research Foundation Flanders (FWO) and the Flemish Institute for Technological Research VITO nv (grant numbers 1S69320N, 1159719N ) and by the EU H2020 M3DLoC project (grant number 760662).