DNA barcode como ferramenta na identificação de larvas de peixes da plataforma continental amazônica.

Abstract

The larval stage of fish life is characterized as the period of greatest vulnerability, during which the highest mortality rate occurs. Studies related to this stage are still very scarce; however, studying this life stage of these organisms is of great importance as it provides information about the biology and taxonomy of species, as well as helping in the identification of spawning areas and the monitoring of fish stocks. The morphological method, with the aid of identification guides, is the most commonly used method for identifying fish larvae. However, this methodology remains a challenge because there is a significant shortage of these guides, and most of them are local or regional and do not cover the northern coast of Brazil. Uncertain identification of fish larvae can lead to problems related to management strategies for commercially important species and, consequently, fishing activities. Therefore, many international researchers have started testing and using molecular analysis in some groups. Currently, the most widely used technique is DNA Barcoding, through the sequencing of the mitochondrial cytochrome c oxidase subunit I (COI) gene, which is the most commonly used molecular marker for fish and allows for more precise identification and differentiation between species. In Brazil, the use of this technique to identify marine fish larvae is still rare, and on the northern coast of Brazil, this technique has not yet been applied to fish larvae. This study aims to test the efficacy of the DNA Barcoding technique and evaluate the accuracy of morphological identification for fish larvae collected on the Amazon continental shelf. The study area focused on the Amazon continental shelf, and 12 sampling stations were analyzed. The larvae were collected between September and October 2021 and are part of the French-Brazilian cooperation project Amazon shelf mixing and its impacts in the ecosystems – AMAZOMIX. The samples were collected aboard the French research vessel Antares, using a Multinet Midi sampler (5 nets of 300 μm), collecting at different depths in the water column. After collection, the samples were fixed in 90% ethanol and refrigerated. A total of 343 fish larvae were identified using the morphological method, classified into 10 orders, 30 families, 29 genera, and 17 species. For DNA Barcoding analysis, 76 individuals from each taxonomic level were selected. A fragment of approximately 600bp from 54 larvae, encompassing 8 orders, 31 families, 16 genera, and 34 species, was subjected to DNA Barcoding analysis. There was a 27% discrepancy between the DNA Barcoding results and the morphological identification. These results indicate that DNA Barcoding, combined with morphological identification, can be a valuable tool for precise identification of fish larvae on the Amazon continental shelf.