African painted dogs (Lycaon pictus, APD) are highly endangered, with fewer than 7,000 remaining in nature. Captive breeding programs can preserve a genetically diverse population and provide a source of individuals for re-introductions. However, most programs are initiated from few founders and suffer from low genetic diversity and inbreeding. The aims of this study were to use molecular markers to assess genetic variation, inbreeding, and relatedness among APDs in the North American captive population, to use these data to realign studbook records, and to compare these data to wild populations and to the European captive population to facilitate development of a global management plan. We sequenced mitochondrial and major histocompatibility (MHC) class II loci, and genotyped 14 microsatellite loci from 109 APDs from 34 institutions in North America. We identified three likely studbook errors and resolved ten cases of uncertain paternity. Overall, microsatellite heterozygosity was higher than reported in Europe, but effective population size estimates were lower. Mitochondrial sequence variation was extremely limited, and there were fewer MHC haplotypes than in Europe or the wild. Although the population did not show evidence of significant inbreeding overall, several individuals shared high relatedness values, which should be incorporated into future breeding programs.

Samples from 109 African painted dogs (61 males: 48 females) were obtained from 34 captive breeding institutions in North America. Blood was collected on FTA Elute collection cards (GE Healthcare, Pittsburgh PA, USA) according to the manufacturer’s instructions, and FTA cards were stored and shipped at room temperature. DNA was extracted from the FTA Elute cards according to the manufacturer’s instructions, eluted into 180 ml 1X Tris-EDTA, and stored at either 4˚C or -20˚C. We included negative controls with DNA extractions to monitor for contamination.

To assess neutral genetic diversity, we genotyped each individual at 14 physically unlinked microsatellite loci, which we amplified in five multiplexes: (1) Pez12, FH2054, FH2611, (2) Pez8, FH2010, (3) FH3399, FH3965, (4) FH2785, FH2658, Pez15, and (5) CXX155, CXX250, CXX263 and CXX366 (Ostrander et al. 1993, 1995, Francisco et al. 1996, Neff et al. 1999, Breen et al. 2001, Guyon et al. 2003, Eichmann et al. 2006). Amplification conditions have been described in detail previously (Miller-Butterworth et al. 2019). Fragment analysis was performed using an Applied Biosystems genetic analyzer (model 3730 XL; Waltham, MA) at the Penn State Genomics Core Facility, University Park, PA. One negative control (deionized water) and two previously genotyped samples were included on every plate to monitor for contamination and to ensure consistency across electrophoretic runs.

Allele sizes were called using GeneMarker software (SoftGenetics, State College, PA, USA), based on a known DNA size standard (GeneScan 500 LIZ Dye Size Standard; ThermoFisher Scientific).  Each locus was amplified independently two to four times to confirm each genotype, and putative homozygotes were accepted only if the same single allele was observed in at least three independent PCRs. Thereafter, a consensus multilocus genotype was generated for each individual.

Breen et al. 2001. Genome research 11:1784-1795.

Eichmann et al. 2006. International Congress Series 1288:813-815.

Francisco et al. 1996.  Mammalian Genome 7:359-362.

Guyon et al. 2003. Proceedings of the National Academy of Sciences 100:5296-5301.

Miller-Butterworth et al. 2019. Mammal Research:1-5.

Neff et al. 1999. Genetics 151:803-820.

Ostrander et al. 1995. Mammalian Genome 6:192-195.

Ostrander et al. 1993. Genomics 16:207-213.

The first column lists the ID of the dog. The first row gives the names of the 14 loci. Allele sizes (in base pairs) are listed for the 14 loci, two columns per locus. Unknown genotypes (missing data) are listed as zeros.