Microsatellite genotype data for captive and wild Arabian leopards
Data files
May 02, 2024 version files 19 KB
-
Leopard_microsatelite_data_for_Dryad_HAH.xlsx
-
README.md
Abstract
Genetic diversity underpins evolutionary potential that is essential for the long-term viability of wildlife populations. Captive populations harbour genetic diversity potentially lost in the wild, which could be valuable for release programs and genetic rescue. The Critically Endangered Arabian leopard (Panthera pardus nimr) has disappeared from most of its former range across the Arabian Peninsula, with fewer than 120 individuals left in the wild, and an additional 64 leopards in captivity. We (i) examine genetic diversity in the wild and captive populations to identify global patterns of genetic diversity and structure; (ii) estimate the size of the remaining leopard population across the Dhofar mountains of Oman using spatially explicit capture-recapture models on DNA and camera trap data, and (iii) explore the impact of genetic rescue using three complementary computer modelling approaches. We estimated a population size of 51 (95% CI: 32–79) in the Dhofar mountains and found that 8 out of 25 microsatellite alleles present in eight loci in captive leopards were undetected in the wild. This includes two alleles present only in captive founders known to have been wild-sourced from Yemen, which suggests that this captive population represents an important source for genetic rescue. We then assessed the benefits of reintroducing novel genetic diversity into the wild population, as well as the risks of elevating the genetic load through the release of captive-bred individuals. Simulations indicate that genetic rescue can improve the long-term viability of the wild population by reducing its genetic load and realised load. The model also suggests that the genetic load has been partly purged in the captive population, potentially making it a valuable source population for genetic rescue. However, the greater loss of its genetic diversity could exacerbate genomic erosion of the wild population during a rescue program, and these risks and benefits should be carefully evaluated. The next step in the recovery plan of the Arabian leopard is to empirically validate these conclusions, implement and monitor a genomics-informed management plan, and optimise a strategy for genetic rescue as a tool to recover Arabia’s last big cat.
README: Microsatellite genotype data for captive and wild Arabian leopards
https://doi.org/10.5061/dryad.hdr7sqvrn
Microsatellite genotype data of wild and captive Arabian leopard individuals:
The microsatellite genotype data was produced using a set of microsatellite markers as per the publication (DOI:10.5061/dryad.hdr7sqvrn). A set of eight loci were used for the final genotyping after testing of a larger set of markers. The final set of eight were chosen based on producing clean PCR products, and clean peak profiles during the gene-scanning (using ABI 3730). The sampled individuals are either from the captive-breeding population of leopards (labelled ‘captive’ 17-53 in the data file), or are DNA samples from scat/faecal samples that were collected from across the different mountain ranges of Dohfar (Samham, Qara, Qamar, Nejd). This data comprises genetically different individuals based upon analyses as set out in DOI:10.5061/dryad.hdr7sqvrn. The wild samples are labelled ‘Scat XXX’. The genotype data set also includes a set of genotypes from museum skins (samples 1-14) collected either from leopards killed due to persecution (where the carcass was found) or from leopards that were wild-caught and sampled as part of telemetry studies; for additional location information on these and all other samples, please contact the first author, H. Al Hikmani).
Description of the data and file structure
In the data file, ID refers to the individual ID of the individual DNA samples genetically confirmed to be Arabian leopard. Some samples are from scat/faecal matter (labelled 'scat'), others are skin samples taken from museum specimens collected in the 1970s and 1980s (samples 1-14), and others are from blood samples from the captive population.
The data are arranged in a to column format, with two columns for each microsatellite marker. Each row corresponds to a sampled leopard individual. Cells containing ‘0’ indicates no genotype was scorable. The alleles refer to the fragment sizes of the amplified DNA fragments. The last eight listed samples (rows72-79, samples 44-53) are captive leopards that are known to have been wild-sourced from Yemen.
Sharing/ Access Information
For additional location information on these and all other samples, please contact the first author, H. Al Hikmani
Methods
We identified a set of 65 published polymorphic markers (Menotti-Raymond et al., 1999; Uphyrkina et al., 2001; Williamson et al., 2002; Mondol et al., 2009) and tested their amplification success and extent of polymorphism in Arabian leopards using DNA from three scat samples genetically confirmed to be from leopards in Dhofar. Thirty-five markers successfully amplified during initial PCR trials and were then included in the design of seven multiplex sets (Supplementary Information, Table S2). Multiplex PCRs were performed using fluoro-labelled forward primers to genotype all genetically confirmed leopard samples. Felid-specific PCR primers designed to amplify the amelogenin and zinc-finger regions of y‐chromosome were used to assign individual sex (Pilgrim et al., 2005). All PCR products were genotyped using an Applied Biosystems 3730 DNA Analyser and ROX 500 ROX™ size-standard (DBS Genomics, Durham UK). We used Genemapper v3.7 (Applied Biosystems, UK) to identify and score the alleles. See Supplementary Information for full details on assessment of genotypes.
We performed PCR amplification in reaction volumes of 10 µl containing 5 µl MyTaq HS Red Mix (Bioline), 1.6 µl dH2O, 0.2 µl (0.2 µM) of each forward and reverse primer, 1µl BSA (0.01 μg/μL) (Bovine Serum Albumin, New England Biolabs Inc.) and 2 µl of DNA. PCR cycling conditions consisted of an initial hot start of 95°C for 8 min followed by 35 cycles of 94°C for 30 s, 50°C for 1 m and 72°C for 1 s, and a final incubation period of 10 min at 72°C. To reduce the risk of contamination, PCR preparation and DNA extraction were performed in separate laboratories at the Durrell Institute of Conservation and Ecology (DICE), University of Kent. All PCRs included a negative control. PCR products were initially visualised on 2% agarose gels using electrophoresis to check for amplification and to monitor for signs of contamination in negative controls. PCR products that indicated DNA originating from leopard scat were then purified and sequenced using a 3730X analyser (Macrogen, Amsterdam, Netherlands). The resulting forward and reverse sequences were edited and aligned using Jalview v2 (Waterhouse et al., 2009). Individual consensus sequences were then searched for using BLAST (NCBI); samples from which sequences aligned with the single Arabian leopard NADH5 sequence (GenBank accession: AY035279) were assumed to be leopard; non-leopard DNA samples were excluded from further analyses.
Of 35 markers applied to the DNA sample set, eight produced suitably scoreable genotypes and were observed to be polymorphic. Ten loci - of which seven are known to be polymorphic in other leopard subspecies (Uphyrkina et., 2001; Mondol et al., 2009; Sugimoto et al., 2014) - were found to be monomorphic in Arabian leopard samples (Supplementary Information, Table S7). Remaining loci either did not amplify consistently (four loci) or failed to amplify (13 loci). No evidence of false, null alleles or scoring errors were found, but we observed a small rate of dropout from scat samples: mean dropout rate 0.048 (range = 0.00–0.08).
PCR conditions used for PCR sexing loci
PCR reactions (10 µl volume) contained 5µl Qiagen multiplex PCR buffer mix (Qiagen Inc.), 0.2 µl (0.2 µM) fluoro-labelled forward primer (Eurofins Genomics), 0.2 µl (0.2 µM) reverse primer, 0.5 µl (0.005 μg/μL) BSA, 0.5 µl PCR anti-inhibitor and 3 µl of template DNA. The PCR cycling conditions for all multiplexes consisted of an initial denaturation of 95°C for 15 min, 45 cycles of denaturation (94°C for 30 s), annealing (Ta ranges from 54°C to 58°C for 90 s), extension (72°C for 90 s), and a final extension of 10 min at 72°C.
Assessment of genotypes
To reduce errors associated with degraded DNA we genotyped each sample at least three times and determined consensus genotypes using the multiple-tubes approach (Taberlet et al., 2002). Samples that amplified fewer than four loci were discarded from further analysis. We accepted a genotype to be true if repeated genotypes matched 100% across all loci at least twice, otherwise the sample was excluded. Allelic dropout and false alleles were identified using GIMLET v1.3.3 (Valiere, 2002). Scoring errors and null alleles were identified using Microchecker (Van Oosterhout et al., 2004). Deviation from Hardy-Weinberg equilibrium (HWE) and linkage disequilibrium (LD) was assessed using Genepop 4.7 (Raymond & Rousset, 1995) with sequential Bonferroni correction (Rice, 1989) applied for multiple LD tests.