Impact of mitotype diversity on metabarcoding biodiversity estimations in Insecta and Arachnida using different sample preparation strategies.

DNA barcoding and metabarcoding have been increasingly used in species delimitation and species diversity assessment, respectively, and the molecular markers used in animals are mainly derived from mitochondrial DNA. It is well known that the phenomenon of multiple mitochondrial haplotypes within the same specimen (hereafter referred to as "mitotype diversity") may have a negative impact on the proper assessment of biodiversity by metabarcoding. However, few studies have focused on the incidence of this phenomenon and its effects on metabarcoding results using different sample preparation strategies, such as mock community construction using pooled high-throughput sequencing (HTS) data, DNA-pooling and Tissue-pooling. In this study, we investigated mitotype diversity and its influence on metabarcoding based on 398 specimens from 66 species of Insecta and 82 specimens from 16 species of Arachnida by HTS of the mitochondrial cox1 gene fragment. The results revealed that mitotype diversity was common in the studied taxa and significantly increased the number of operational taxonomic units (OTUs) using the three sample preparation strategies. The results also showed that the bioinformatics pipeline based on authentic amplicon sequence variants was more reliable than the pipeline based on OTUs. Regarding the sample preparation strategies of DNA-pooling and Tissue-pooling commonly used in metabarcoding, our results revealed that their results of metabarcoding were quite similar, and the Tissue-pooling strategy was therefore preferred because of its simplicity. Our study calls for additional attention to the interference of mitotype diversity on the results of DNA metabarcoding in biodiversity assessment.