Amplification-Free Quantification of Endogenous Mitochondrial DNA Copy Number Using Solid-State Nanopores

Abstract

Mitochondrial DNA (mtDNA) quantification is crucial in understanding mitochondrial dysfunction, which is linked to a variety of diseases, including cancer and neurodegenerative disorders. Traditional methods often rely on amplification-based techniques, which can introduce bias and lack the precision needed for clinical diagnostics. Solid-state nanopores, an emerging biosensing platform, have the advantage of offering single-molecule and label-free approaches by enabling the direct counting of DNA molecules without amplification. The ion-current signatures obtained from each DNA molecule contain rich information on the molecules’ lengths and origin. In this study, we present an amplification-free method for mtDNA quantification using solid-state nanopores and machine learning. Intriguingly, we find that native (unamplified) mtDNA translocations harbor structurally distinctive features that can be exploited to specifically detect and quantify mtDNA copies over the background of genomic DNA fragments. By combining selective degradation of linear genomic DNA (gDNA) via exonuclease V with a support vector machine (SVM)-based model, we isolate and quantify mtDNA directly from biological samples. We validate our method using plasmids or isolated mtDNAs by spiking in predetermined quantities. We then quantify endogenous mtDNAs in a cancer cell line and in blood cells and compare our results with qPCR-based quantification of the mtDNA/nuclear DNA ratios. To elucidate the source of the ion-current signatures from the native mtDNA molecules, we perform synchronous electro-optical sensing of mtDNAs during passage through the nanopore after NHS ester reaction with fluorophore compounds. Our results show correlated electro-optical events, indicating that the mtDNA is complexed with packaging proteins. Our assay is robust, with a high classification accuracy and is capable of detecting mtDNA at picomolar levels, making it suitable for low-abundance samples. This technique requires minimal sample preparation and eliminates the need for amplification or purification steps. The developed approach has significant potential for point-of-care applications, offering a low-cost and scalable solution for accurate mtDNA quantification in clinical settings.