Cisplatin, which selectively binds to N7 atoms of purines to inhibit normal replication and transcription, is a widely applied chemotherapeutic drug in the treatment of cancer. Though direct identification of cisplatin lesions on DNA is of great significance, existing sequencing methods have issues such as complications of preamplification or enrichment-induced false-positive reports. Direct identification of cisplatin lesions by nanopore sequencing (NPS) is in principle feasible. However, relevant investigations have never been reported. By constructing model sequences (83 nucleotides in length) containing a sole cisplatin lesion, identification of corresponding lesions by NPS is achieved with <10 ng of input sequencing library. Moreover, characteristic high-frequency noises caused by cisplatin lesions are consistently observed during NPS, clearly identifiable in corresponding high-pass filtered traces. This feature is, however, never observed in any other combinations of natural DNA bases and could be taken as a reference to identify cisplatin lesions on DNA. Further investigations demonstrate that cisplatin stalls the replication of phi29 DNA polymerase, which appears as a ∼5 pA level fluctuation in the single-molecule resolution. These results have confirmed the feasibility of NPS to identify cisplatin lesions at the genomic level and may provide new insights into understanding the molecular mechanism of platinum-based drugs.
Keywords: MspA; cisplatin; high-pass filter; nanopore sequencing; replication stalling.