Comparison of destructive and non-destructive DNA extraction methods for the metabarcoding of arthropod bulk samples

DNA metabarcoding is routinely used for biodiversity assessment, especially targeting highly diverse groups for which limited taxonomic expertise is available. Various protocols are currently in use, although standardization is key for its application in large-scale monitoring projects. DNA metabarcoding from arthropod bulk samples can either be conducted destructively from sample tissue, or non-destructively from sample fixative or lysis buffer. Non-destructive methods are highly desirable for the preservation of sample integrity but have yet to be experimentally evaluated in more detail. Here, we compare diversity estimates obtained from 14 Malaise trap samples processed consecutively with three non-destructive and one destructive approach from homogenized tissue. In the non-destructive approaches, DNA was extracted using fixative ethanol and two lysis-based approaches. With non-destructive approaches, highest species number was detected by extraction from fabricate lysis buffer, which composition showed a high overlap to species list assessed from grinded ground tissue. A significantly divergent community was detected from preservative ethanol-based extraction. No consistent trend was found in species richness with increasing incubation time in lysis buffer. These results indicate that non-destructive DNA extraction from incubation in lysis buffer could provide a comparable alternative to destructive approaches with the added advantage of preserving the specimens for post-metabarcoding taxonomic work.