18 - So what about non-destructive COI metabarcoding?
Non destructive COI metabarcoding refers to using a sample preparation method where the individual or bulk community is not homogenized prior to DNA-extraction or direct-PCR.
The first paper I'm aware of that talks about using non-destructive sampling of sample preservative for COI barcoding was done by doing direct-PCR on evaporated mescal (similar to tequila) to successfully detect the caterpillar (agave butterfly larvae) without having to homogenize the entire the sample first (Shokralla et al., 2010 Biotechniques 48: 233). This advance showed that COI barcoding from preservative ethanol was possible, leaving an intact voucher specimen for future work.
This was followed up with a paper that extended this non-destructive technique to analyze freshwater benthic macroinvertebrate community from live sorted bulk samples preserved in ethanol (Hajibabaei et al., 2012 BMC Ecology 12: 28). Mixed community COI metabarcoding from preservative ethanol compared favorably to single-specimen tissue extractions and detected all taxa that were present from individuals that comprised more than 1% of the sample. One advantage stressed in this paper is that direct PCR from dried ethanol alleviates a DNA extraction bottleneck, streamlining sample processing in large-scale studies. Of course, this also leaves voucher specimens intact for future work.
Recently, a flurry of new work has re-iterated how COI metabarcoding could benefit from non-destructive sampling techniques such as soaking specimens in buffer and removing them prior to DNA extraction or using direct-PCR on preservative ethanol:
Porter & Hajibabaei, 2018 Over 2.5 million COI sequences in GenBank and growing. bioRxiv, doi: https://doi.org/10.1101/353904
Zizka et al., 2018 DNA metabarcoding from sample fixative as a quick and voucher preserving biodiversity assessment method. bioRxiv, doi: https://doi.org/10.1101/287276
Disadvantages:
Compared with conventional bulk sample processing, COI metabarcoding from preservative ethanol may recover less overall biodiversity (Zizka et al., 2018).
It has been noted that heavily sclerotized specimens and specimens retracted into stone cases were less likely to be detected when specimens were incubated in DNA extraction buffer then removed prior to DNA extraction (Carew et al., 2018).
Advantages:
Generally, voucher samples can be retained for further work such as morphological characterization or stable isotope analysis when direct PCR on preservative ethanol was performed (Shokralla et al., 2010; Hajibabaei et al., 2012).
When the specimen-sorting step is eliminated and preservative ethanol is sequenced, this further streamlines sample processing (Zizka et al., 2018).
When specimens were incubated in DNA extraction buffer and removed prior to DNA extraction and metabarcoding, results are similar to tissue metabarcoding or the metabarcoding from dissected individuals (Carew et al., 2018).
COI metabarcoding using non-destructive methods could turn the table on the traditional paradigm where metabarcoders are consumers of taxonomic information in reference databases compiled by taxonomists. Taxonomists could instead mine COI metabarcode datasets in species-discovery mode. Any novel COI metabarcodes that are discovered bioinformatically could theoretically be verified by going back to intact voucher specimens (Porter and Hajibabaei, 2018).
There is definitely more work to be done to optimize detection rates using non-destructive methods. As it stands, COI metabarcoding from unsorted bulk specimens from preservative ethanol shows promise as a streamlined approach for quick biodiversity surveys. For detailed richness estimates, however, more conventional COI metabarcoding approaches may be more appropriate.