ISTS42 Program/Agenda

The current biodiversity crisis demands urgent conservation actions to halt and reverse the situation. In this context, conservation genomics can assist biodiversity conservation and management with reference genomes being key centrepieces in such analyses. Consequently, many initiatives are in progress to provide reference genomes with the ultimate goal of sequencing all earth biodiversity, and marine turtles are no exception, as reference genomes of several species are already available. However, no single genome can represent the diversity of a species, as structural and sequence polymorphisms may differentially drive the future conservation actions of populations. We aimed to go beyond the reference genome of the loggerhead turtle by adding nuclear and mitochondrial genomes from individuals sampled worldwide. We performed whole genome sequencing (> 30 Gb output) on 12 individuals from four different Regional Management Units (RMU): Pacific (Mexico: 2), western Atlantic (Mexico: 2; Florida: 1), eastern Atlantic (Boavista: 2) and Mediterranean (Libya: 1; Greece: 2; Turkey: 2). We assembled the 12 mitogenomes (coverage > 900X) and combined our results with the full mitogenomes available from GeneBank to produce a final set of 20, covering the four RMUs. The structure of the mitogenomes was similar across all individuals. The phylogeographic analysis supported a scenario of an initial haplogroup split between the Atlantic and Pacific after the closure of the Panama Isthmus and a secondary introduction to the Atlantic from the Pacific that reached the Mediterranean and western Atlantic. The genotyping against the reference genome (GCA_023653815.1) produced a set of more than 12.7 million nuclear SNPs across all individuals (coverage >11.5X), with the number of SNPs per chromosome highly correlated to chromosome size (R²=0.99). All the individuals were clustered by RMU on multidimensional scaling, the Pacific being the most different to the remaining RMUs on the first axis. The second axis separated the Mediterranean from the Atlantic RMUs, the third axis separated the Mediterranean populations and the fourth axis separated the two Atlantic RMUs. The genetic distances, measured as Fst, showed the same pattern, with a high differentiation of the Pacific RMU and the two Atlantic RMUs being less differentiated. When performing pairwise Fst comparisons by chromosome, the most extreme values were always concentrated on the macrochromosomes, suggesting different evolutionary patterns across chromosomes. Thus, macrochromosomes would host highly conserved regions but also highly differentiated regions across RMUs that might drive local adaptation. An analysis of Runs of Homozygosity (ROHs) showed that the degree of inbreeding was uneven across RMUs. The individuals from the Mediterranean had genomes with longer ROHs and a higher portion of the genome in homozygosity, indicating higher inbreeding. Moreover, the ROHs distribution across the chromosomes was not uniform, as macrochromosomes had a higher proportion of their content in homozygosis than microchromosomes. In summary, combining the reference genome with whole genome sequencing data worldwide can provide novel insights into marine turtle conservation genomics. Future efforts to increase sample size and to incorporate more RMUs are advisable to expand our comprehension of marine turtle genomes for conservation.

We are facing the sixth mass extinction, and marine turtles are particularly vulnerable to global warming, especially in light of their temperature-dependent sex determination and sea-level rise. The loggerhead turtle (Caretta caretta) nesting sites in the Mediterranean sea are predominantly concentrated in the eastern basin, where the temperature is expected to increase higher than the global average in the upcoming decades in this area. Consequently, evaluating the conservation status of the populations in the eastern Mediterranean, whether they are localy adapted and their structure, is crucial for the species' survival in this area. Genomics is a key tool allowing genome-wide knowledge of both population structure and local adaptation. The recent publication of the reference genome of the loggerhead turtle enables unprecedented fine-level resolution genomic studies in this species. Here, we provide a large-scale genomic analysis of loggerhead turtles inhabiting the Mediterranean. To do so, we sequenced 243 individuals from 11 nesting populations using 2b-RAD technology and genotyped them using the recently assembled reference genome. We assessed the population structure and local adaptation by combining genomic information with environmental (salinity and temperature), behavioural (hatchling dispersal patterns and adult foraging strategies) and reproductive (clutch sizes) population-specific published data. We found genetic differentiation among most Mediterranean rookeries clustered in three main groups: nord-eastern (Greece), sud-eastern (Libya) and east-eastern (Turkey, Syria, Lebanon, and Israel), which was supported by both DPAC and hierarchical discriminant analysis using all loci. To scan for local adaptation in the Mediterranean populations we used a combination of Outlier and Redundancy Analyses. We identified a highly differentiated set of 326 candidate loci for local adaptation and genomic signatures significantly associated with environmental, behavioural and reproductive population parameters. Of note, candidate loci were enriched in genes, as 56.43 % of the candidate SNPs were located either in exonic (3.46% of them) or intronic (96.54% of them) regions. Interestingly, more than 50% of the candidate genes were specifically found in a single outlier or RDA analysis, suggesting that different gene networks are driving the adaptation to the different phenotypic and environmental variables tested. On the contrary, only three were shared among all analyses, indicating that testing multiple potential drivers is key for a comprehensive understanding of local adaptation. We functionally characterised the candidate loci for local adaptation using a Gene Ontology analysis and found that the candidate genes are involved in a myriad of basic biological cell functions, including immunity, cytokinesis or protein synthesis among others. Finally, we integrated genetic structure and connectivity in a conservation prioritisation analysis to identify high-priority areas for protection. We conclude that there is a need to protect most populations to maintain healthy levels of genetic diversity. Our study provides a baseline for future genomic studies and establishes conservation priorities of the main nesting sites of the loggerhead turtle in the Mediterranean while providing target genes for future research directed to the study of the genomic basis of adaptation.

Understanding the current population genetics and connectivity of sea turtles and the recent development is crucial for effective conservation management of the species. Five sea turtle species, green turtle (Chelonia mydas), loggerhead turtle (Caretta caretta), hawksbill turtle (Eretmochelys imbricata), olive ridley turtle (Lepidochelys olivacea) and leatherback turtle (Dermochelys coriacea), are recorded in the East Asia Region situated in the western side of the North Pacific Ocean. We compiled information from 35 published genetic studies on the five sea turtle species, with a focus on green turtle and loggerhead turtle, which are the most studied species (in 30 studies) in view of their commonness and occurrence of nesting populations. This literature review provided an overview of the key methods and findings of these previous studies, addressing two main objectives on genetic structure of the rookeries and their differences compared to other populations, and connectivity of the rookeries and foraging aggregations. By identifying information gaps and conservation needs, we discussed future developments for sea turtle genetic studies and conservation implications in the region.

The hawksbill turtle (Eretmochelys imbricata) is globally threatened and listed as Critically Endangered in the IUCN Red List, partly due to the loss of nesting and feeding habitats, and illegal wildlife trade. Establishing a regional reference of spatial genetic variation is crucial for monitoring changes in genetic diversity necessary for adaptive potential, and tracing geographic origins of unknown individuals, such as those from the wildlife trade. In Thailand, however, the genetic database of wild hawksbill turtle populations remains scant, hindering effective conservation strategies to promote population viability and species recovery. In this study, we aimed to (1) assess the number of mitochondrial DNA (mtDNA) haplotypes (Nh), haplotype diversity (h) and nucleotide diversity (pi) of hawksbill turtles from the head-starting facility at Koh Mannai, along the eastern coast of the Gulf of Thailand using partial 776bp Control Region sequences, and (2) infer phylogenetic relationships and phylogeographic partitioning between eastern Thailand populations and other range countries. Based on 22 blood samples of hatchlings collected from Koh Kram and raised at Koh Mannai, 19 samples (86.4%) were successfully sequenced based on the partial hypervariable portion (776bp) of the Control Region. Of the 19 samples and seven variable sites, a high level of mtDNA diversity was detected from the eastern Thailand populations (Nh = 6, h = 0.789 (± 0.073), pi = 0.00303 (± 0.00037)). Phylogenetic analyses strongly supported that all six haplotypes were clustered within the previously characterized Indo-Pacific III Clade, including haplotypes from Peninsular Malaysia and East Malaysia. Of the total six haplotypes, only the EiIP-151 haplotype is considered to be a newly characterized haplotype, while the remaining ones were shared with populations from Peninsular Malaysia and East Malaysia. Phylogenetic placement provides insight into the maintenance of dispersal and genetic connectivity between eastern Thailand populations with Malaysian populations. This genetic database will help guide management actions for hawksbill turtle conservation.

Genetically discrete, demographically independent populations serve as a framework for structuring management efforts in wild and imperiled species. In species with broad, complex dispersal or recently established populations, evidence of demographic independence in neutral genetic markers can be weak or contradictory, and it can often be beneficial to seek additional evidence. Genome-wide single nucleotide polymorphisms (SNPs) have been shown to reliably detect weak population structure in these species, and may produce informative loci for genetic stock inference (GSI). Sea turtles comprise seven species with well-known broad, sex-biased dispersal and complex population structure, for which informative loci for accurate GSI do not always exist. We therefore generated genome-wide SNP data for four genetically discrete management units (MUs) of loggerhead turtles in the Gulf of Mexico, to determine if SNPs would reflect demographic independence between MUs and could elucidate informative loci for GSI. Analyses of 30,776 SNP-containing loci across 69 individuals from the Northern Gulf of Mexico (n=18), Central Western Florida (n=18), Southwestern Florida (n=15) and the Dry Tortugas (n=18) showed weak but significant structuring (F_ST=0.001) between MUs. We subsequently identified 238 outlier loci for GSI, which were able to re-assign individuals to MUs with 100% success. Analyses of genome-wide SNPs provided deeper insights into the eco-evolutionary dynamics governing demographically independent populations in sea turtles and demonstrated the potential of SNP-containing loci to serve as informative markers for accurate GSI in species with broad dispersal and shallow phylogeographic histories. Future work will include expanding sample sizes at MUs to validate loci for GSI, expanding genome-wide SNP data generation to other loggerhead MUs, and developing genotyping in sequencing by-thousands panels (allowing for parallel sequencing of hundreds to thousands of targeted, informative loci) for GSI in loggerheads in the Northwest Atlantic Discrete Population Segment.

Sex ratios are one of the determinants of population size. If the operational sex ratio is extremely skewed to one sex, the population will not grow due to its small effective population size. In sea turtles whose sexes are determined by incubation temperatures, low temperatures produce males whereas high temperatures produce females. At rookeries exposed to high incubation temperatures, primary sex ratios are thus skewed to females. However, because high incubation temperatures also affect body size and activity levels of hatchlings negatively, female offspring are not considered to be able to survive better than male offspring, leading to reversed (i.e., male biased) sex ratios in adults. Here, we examined (1) whether the operational sex ratio is balanced, and (2) whether sex ratios change with ontogeny, at Yakushima Island, Japan, the largest loggerhead turtle (Caretta caretta) rookery in the North Pacific, which is not exposed to high incubation temperatures. Incubation durations were recorded for 107 nests during five years (2014–2016, 2020, and 2021), and they were converted into mean sand temperatures during incubation. A regression equation between air temperature, rainfall, and sand temperature were constructed with a generalized linear model (R² = 0.940). Using this equation, mean sand temperatures experienced by clutches laid during six bins of the nesting season were estimated for 48 years (1975–2022). Primary sex ratios for the six bins were estimated from the mean sand temperatures, assuming a pivotal temperature of 29.7℃. By multiplying a primary sex ratio and nest frequency for each bin and summing them, an annual primary sex ratio was calculated. In 2020 and 2021, tissues for genetic analysis were collected from 28 females and their 548 offspring. Five microsatellite loci were analyzed, and the number of sires were calculated using two programs (GERUD and COLONY). The mean (±SD) primary sex ratio (%female) for 48 years was 27.8 ± 14.0. The operational sex ratio was 40.9 in GERUD and 25.2 in COLONY, with the mean being 33.1. The lack of an extreme bias in the operational sex ratio means that the effective population size of loggerhead turtles that breed at Yakushima Island is large. A small difference between primary and operational sex ratios suggests that the survival rate of female offspring produced at Yakushima Island is similar to that of male offspring. These suggest that the population size of sea turtles may be primarily determined by the incubation environment that affects hatchling characteristics.

Global environmental change poses a significant threat to turtle populations worldwide. Understanding sex ratios and sex determination in turtles is therefore critical for effective conservation and management efforts. However, traditional sex determination methods in turtles tend to be invasive or require head starting neonates. There is hence a pressing need for accurate, non-lethal techniques to determine sex in hatchling. DNA methylation, the addition of a methyl group on DNA nucleotide, has been associated with gene regulation including of sex-related genes. Here, we developed a non-lethal sex determination method focusing on DNA methylation from blood samples of hatchling loggerhead turtles. This method was applied to a field split-clutch design experiment where nests were exposed to either male producing depth (lower temperatures) or female-producing depth (higher temperature). We identified several sex-specific differentially methylated genomic regions, which allow for highly accurate molecular sexing of turtles from blood samples. In this presentation we will also show the feasibility of the sequencing approach in the field. Overall, these findings indicate that DNA methylation analysis of blood samples provides a straightforward, non-lethal method to evaluate sex ratios in sea turtle populations. Widespread application of this technique will substantially improve knowledge of turtle primary sex ratios, enabling better targeted conservation strategies. Portable sequencing technology makes in-field molecular sexing possible, giving researchers an invaluable new tool to understand and protect turtle populations in a rapidly changing global environment.