Deoxyribonucleic acid extraction for invertebrate metabarcoding from stream sediment and implication to biotic indices

BACKGROUND AND OBJECTIVES: Environmental deoxyribonucleic acid metabarcoding has become a powerful tool for assessing invertebrate biodiversity in stream sediments, providing a non-invasive approach for ecological monitoring. However, the precision of this method is largely contingent upon the nucleic acid extraction protocol utilized, as it can impact taxonomic recovery and the associated biotic indices. Previous studies have focused on nucleic acid yield and taxonomic composition, but limited research has evaluated the impact of extraction methods on ecological assessments. The objective of this study is to analyze four nucleic acid extraction methods to identify their effectiveness in optimizing metazoan recovery and bolstering the reliability of biotic indices generated from metabarcoding data. METHODS: Four nucleic acid extraction protocols were tested for invertebrate nucleic acid metabarcoding from stream sediment: a manual extraction and three Qiagen commercial kits (Powersoil, Blood and Tissue, and Plant Mini kit). Amplification of the extracted nucleic acid was performed utilizing cytochrome c oxidase I along with hypervariable markers 4 and 9 from the 18 Svedberg ribosomal ribonucleic acid, and subsequent sequencing was conducted using Illumina sequencing methods. The relative read abundance was analyzed to assess metazoan recovery and biotic index scores were calculated based metazoan community of respective extraction protocol. FINDINGS: Among the tested methods, PowerSoil kit consistently yielded the lowest metazoan relative read abundance (36.07 to 86.31 percent) from stream sediment and exhibited higher non-target amplification, including bacterial nucleic acid, due to its bead-beating step optimized for prokaryotic nucleic acid extraction. In contrast, different protocols exhibited a greater efficacy in recovering metazoans, achieving percentages between 63.83 and 98.58, and significantly limiting the co-amplification of bacterial DNA. This indicates a pronounced specificity for eukaryotic nucleic acid. Plant Mini kit performed unexpectedly well, aligning closely with the manual extraction in biotic indices, likely due to its effective inhibitor removal of amplification. Conversely, the blood and tissue kit demonstrated the most significant dissimilarity in biotic indices, likely resulting from inhibition that caused a decrease in the detection of sensitive taxa. Biotic indices derived from cytochrome c oxidase I and 18 Svedberg ribosomal ribonucleic acid hypervariable 9 markers were comparable, with cytochrome c oxidase I showing slightly higher scores due to its greater specificity in targeting key taxa used for index calculation. The 18 Svedberg ribosomal ribonucleic acid hypervariable 4 markers, resulted in diminished biotic index scores and was less successful in facilitating the recovery of metazoans, particularly among invertebrates that are vulnerable to ecological stressors. CONCLUSION: Overall, the choice of extraction method substantially influenced the detection of target metazoan taxa and the reliability of biotic indices, with the manual extraction and Plant Mini kit emerging as the most effective protocols. The findings illustrate the necessity of appropriate protocols selection in the metabarcoding of stream sediments, indicating that refined techniques for the effective recovery of metazoan nucleic acids are imperative for reliable ecological analysis.