Methods
St. John is part of the US Virgin Islands which includes St. John, St. Thomas, St. Croix, and Water Island. St. Thomas is the nearest island which also contains deer (Figure 1). Virgin Islands National (VINP) park lies on the island of St. John and comprises 60% of the landmass of the island. VINP protects one of the largest and most mature tracts of secondary dry forest in the eastern Caribbean (Ray et al. 1998). The island vegetation is largely represented by low-to-mid elevation dry scrub forest on soils with fairly low soil nutrient content (Oswalt et al. 2006) and is considered marginal habitat for deer. A severe drought was present on St. John and the surrounding region that lasted for the duration of 2015 and caused water, food, and environmental stress to the St. John deer population (Nelson et al. 2017). During the drought, deer showed signs of stress such as highly elevated tick and mange levels, muscular atrophy, poor coat quality, weight loss, lethargy, reduced reproduction, and death (Reuter and Nelson 2018).
DNA samples were collected in two ways, either by collecting hair samples from deer while they were anesthetized, or by collecting DNA from fecal samples. To collect hair samples, adult deer were anesthetized using butorphanol, azaperone, and medetomidine (BAM, Wildlife Pharmaceuticals, Windsor, Colorado, USA). Only adult does and bucks were immobilized for this project, no pregnant deer or fawns were used. Vitals monitored included heart rate, respiratory rate, mucous membrane color, body temperature, time to recumbency, and recovery. Following hair collection and after examination, the anesthesia was reversed with 2–3 mL of atipamezole (25 mg/mL) and 0.5 mL of naltrexone (50 mg/mL, Wildlife Pharmaceuticals). Deer recovered to standing with full stability within five minutes (Reuter and Nelson 2018). Hair samples were individually labelled and placed in coin envelopes in frozen storage until analysis.
Fecal DNA samples were collected only from freshly deposited fecal samples with the deer in view. Several toothpicks were rubbed gently over the surface of the fecal sample for each sample collected. The samples were allowed to dry and placed in a coin envelopes in frozen storage until analysis. Research on live animals followed ASM guidelines (Sikes et al. 2016) and was completed under Scientific Research and Collection permit VIIS-2016-SCI-0026 for the USVI National Park and the University of Colorado Boulder and the National Park Service Institutional Animal Care and Use Committee (1602.01-15Mar2016).
DNA was extracted from both the hair samples using QIAGEN DNeasy Blood and Tissue kits and following QIAGEN’s tissue protocol. All hair samples yielded at least 10 guard hair roots (Paetkau 2003). A standard set of 21 microsatellite markers that are used for parentage certification in game farming applications, and which were found originally in populations of mainland deer from North Carolina, Wisconsin, and Texas, were amplified for the 80 adult deer sampled on St. John (Wildlife Genetics International, Inc.).
Individuals with >10 genotyped microsatellites (N = 42) were analyzed using GENEPOP (Ver. 4.2) (Raymond and Rousset 1995; Rousset 2008) to determine number of alleles per locus, observed heterozygosity, and inbreeding coefficient (FIS) for comparison to non-island populations of white-tailed deer.