The accuracy of reverse genetics systems for SARS-CoV-2: Circular polymerase extension reaction versus bacterial artificial chromosome.

Background: Reverse genetics systems to rescue viruses from modified DNA are useful tools to investigate the molecular mechanisms of viruses. The COVID-19 pandemic prompted the development of several reverse genetics systems for SARS-CoV-2. The circular polymerase extension reaction (CPER) method enables the rapid generation of recombinant SARS-CoV-2; however, such PCR-based approaches could introduce unwanted mutations due to PCR errors.

Methods: To compare the accuracy of CPER and a classic reverse genetics method using bacterial artificial chromosome (BAC), SARS-CoV-2 Wuhan/Hu-1/2019 was generated five times using BAC and five times using CPER. These 10 independent virus stocks were then deep sequencing, and the number of substitutions for which the frequency was greater than 10% was counted.

Results: No nucleotide substitutions with a frequency of greater than 10% were observed in all five independent virus stocks generated by the BAC method. In contrast, three to five unwanted nucleotide substitutions with a frequency of more than 10% were detected in four of the five virus stocks generated by the CPER. Furthermore, four substitutions with frequencies greater than 20% were generated in three virus stocks by using the CPER.

Conclusions: We found that the accuracy of the CPER method is lower than that of the BAC method. Our findings suggest care should be used when employing the CPER method.