AUTHOR=Lee Benjamin B. , Schott Eric J. , Behringer Donald C. , Bojko Jamie , Kough Andrew , Plough Louis V. 

TITLE=Rapid Genetic Identification of the Blue Crab Callinectes sapidus and Other Callinectes spp. Using Restriction Enzyme Digestion and High Resolution Melt (HRM) Assays

JOURNAL=Frontiers in Marine Science

VOLUME=7

YEAR=2020

URL=https://www.frontiersin.org/journals/marine-science/articles/10.3389/fmars.2020.00633

DOI=10.3389/fmars.2020.00633

ISSN=2296-7745

ABSTRACT=<p>The blue crab <italic>Callinectes sapidus</italic> is one of the most widely studied marine crustaceans due to its high economic value and ecological significance. Despite extensive research on the blue crab in North America, many questions remain about the distribution and abundance of the species in the subtropics and tropics. In many places, <italic>C. sapidus</italic> is sympatric with morphologically similar <italic>Callinectes</italic> spp., which has implications for seafood mislabeling. To enable rapid identification of the species, we designed and tested two PCR-based assays targeting the 12S rRNA mitochondrial gene. The first assay discriminates <italic>C. sapidus</italic> from other <italic>Callinectes</italic> spp. via post-PCR restriction digestion (PCR-RFLP) and the second assay discriminates among multiple <italic>Callinectes</italic> spp. through High Resolution Melting (HRM) analysis and supervised machine learning analyses. A total of 58 DNA samples from five <italic>Callinectes</italic> spp. (validated via 12S gene sequencing) were used for assay testing. The PCR RFLP assay was 100% accurate identifying <italic>C. sapidus</italic> from other <italic>Callinectes</italic> spp. HRM analysis of amplicons showed good discrimination among species, with distinct clusters formed between species with higher sequence homology. Linear discriminant analysis (LDA) classification of HRM curves was quite successful given the small dataset available, producing ∼90–91% mean accuracy in classification over all species with 100-fold cross validation. Much of the error came from misclassifications between <italic>C. similis</italic> and <italic>C. danae</italic>, which are ∼99% similar in sequence for the amplicon; collapsing them into a single class increased overall classification success to 94%. Error also arose from <italic>C. bocourti</italic> classifications, which had a reference set containing only three samples. Classification accuracy of <italic>C. sapidus</italic> alone via HRM was 97.5%. Overall, these assays show great promise as rapid and inexpensive methods to identify <italic>Callinectes</italic> spp. and have application for both ecological research and seafood identification or labeling.</p>