DNA barcoding of Euphausiids and Chaetognaths off the east coast of Africa.

Abstract

DNA barcoding is a well-established method that has been used to augment traditional morphologybased methods to identify species. A 650 base pair region of the mitochondrial cytochrome oxidase I gene (COI) is used as the standard DNA barcode across many animal groups. The reliability of species delimitation by DNA barcoding can be biased by factors such as incomplete reference databases which link sequence data to taxonomically validated species; a lack of regional records for widespread species that may have below-species-level genetic structure; and not enough variable characters in the barcode region to confidently separate closely related species. Unfortunately, only a few studies investigate the accuracy of standard barcode markers, such as COI, in delimiting species within their chosen group of study. This study focuses on the DNA barcoding of Euphausiids and Chaetognaths. These two phyla form an important component of zooplankton. Zooplankton serve as critical links in marine food chains, contribute to nutrient recycling, and their presence and diversity act as bio-indicators of ecosystem health where changes in their populations can signal changes in environmental conditions. Identifying the species that make up zooplankton is crucial in diversity assessments, biomonitoring, and the detection of invasive species. The overall aim of this MSc study was to determine the utility of COI as a DNA barcode marker for Euphausiid and Chaetognath species. This was done by: (i) analysing all publicly available COI sequence data to evaluate whether there is a statistically significant separation between the maximum genetic differences observed within individuals belonging to the same species and the minimum genetic differences observed between different species. This separation is known as the “DNA barcode gap”. If the DNA barcode gap is present in the COI data for Euphausiids and Chaetognaths, then the COI region can be used to accurately delimit species. If there is overlap withinspecies and between-species genetic differences, then additional genetic markers or other information are needed to confidently separate species; (ii) generating new DNA barcodes for Euphausiids and Chaetognaths specimens collected from five different localities off the East Coast of South Africa in collaboration with an expert taxonomist to strengthen the current reference libraries for the two groups; (iii) evaluating the completeness of DNA reference libraries for Euphausiids and Chaetognaths by using data from recent metabarcoding studies to determine how many Euphausiid and Chaetognath sequences, amplified from zooplankton samples collected off the East Coast of South Africa, can be confidently assigned (>96% sequence similarity) to reference sequences available on BOLD and GenBank. These analyses are important as the exact number of species found in South African waters is still unclear. Chapter One provides an overall introduction to topics relevant to the dissertation. This chapter highlights the importance of monitoring the species that make up zooplankton. The chapter also addresses the taxonomic challenges associated with identifying species within Euphausiids and Chaetognaths. It introduces DNA barcoding as a vital tool to enhance the speed and accuracy of species identification within these two important zooplankton groups. The chapter concludes with an outline of the overall aims of the dissertation. Chapter Two aims to assess if the standard COI barcode region can statistically delimit Euphausiid and Chaetognath species. This was done by testing for the presence of the DNA barcode gap in the COI data currently available for Euphausiid and Chaetognath species. In this study, I extended this concept to also examine if sequence divergence differences consistently occur between different genera and families belonging to Euphausiids and Chaetognaths. The data analyzed in this chapter came primarily from BOLD and GenBank. Although analyses were conducted separately for Euphausiids and Chaetognaths, the results were very similar. In Chaetognaths, a clear DNA barcode gap was observed, indicating that sequences with more than 3% sequence divergence can be regarded as distinct species. Similarly, in Euphausiids, a clear DNA barcode gap was observed, indicating that sequences with more than 3 - 4% sequence divergence can be regarded as distinct species. Species pairs found in the overlap regions have low divergence values and are phenotypically very difficult to differentiate. Further statistical analyses using the Jeffries-Matusita test was conducted, confirming COI to be successful in separating different species and genera belonging to Chaetognaths and Euphausiids. In Chapter Three, the primary goal was to improve the records of Euphausiid and Chaetognath species from South Africa that are accessible in DNA reference libraries. The secondary goal was to test the utility and accuracy of these reference libraries using metabarcoding data generated in a previous zooplankton metabarcoding study. Euphausiid and Chaetognath specimens collected from five localities off the East Coast of South Africa were sorted and identified to the lowest taxonomic level using morphology-based keys under the supervision of a trained taxonomist. Up to five specimens of each species per locality were used for DNA barcoding. DNA barcodes (COI) were generated for 11 Euphausiid species belonging to four genera and one family, as well as for one species of Chaetognath from one genus and one family. These new records were added to BOLD. This study increased the number of barcodes on BOLD by 118 new records making up of two new DNA barcodes for Chaetognaths and 114 new DNA barcodes for Euphausiids, resulting in an increase in data available from South Africa for these taxa. New species records added to BOLD, include Euphausia recurva, Thysanoessa gregaria, and Sagitta enflata. To evaluate the impact of these new barcode records on species identification and determine the extensiveness of the DNA reference library, new sequences from the metabarcoding results were matched (>96% sequence similarity) in Euphausiids, at a 77% success rate, with 25 species and one genus identified and for Chaetognaths at a 64% success rate, with 12 species and one genus being identified. In Chapter Four I provide an overall summary of the work conducted towards the MSc. This study statistically tests if the COI sequence data can accurately separate species, genera and families of Euphausiids and Chaetognaths. Both Euphausiids and Chaetognaths play an essential role in marine food chains and are used as ocean health indicators, however, despite their importance, very few COI studies with species identification have been conducted. Euphausiids and Chaetognaths display cryptic morphology, and their taxonomy relies heavily on morphology-based methods of species delimitation, therefore molecular-based methods like DNA barcoding, has the ability to provide faster and more accurate species identification. For Chaetognaths, nuclear genes should be barcoded and combined with COI as they have unwonted patterns of mitochondrial evolution and display extraordinary heterogeneity. Hence COI together with nuclear genes will allow for a more valid and represented DNA barcode database for Chaetognaths. Through DNA barcoding, Euphausiids and Chaetognaths can be assessed for as biological indicators of the overall health of South African marine ecosystems.