Insights from the rescue and breeding management of Cuvier's gazelle (Gazella cuvieri) through whole-genome sequencing

Abstract

Captive breeding programmes represent the most intensive type of ex situ population management for threatened species. One example is the Cuvier's gazelle programme that started in 1975 with only four founding individuals, and after more than four decades of management in captivity, a reintroduction effort was undertaken in Tunisia in 2016, to establish a population in an area historically included within its range. Here, we aim to determine the genetic consequences of this reintroduction event by assessing the genetic diversity of the founder stock as well as of their descendants. We present the first whole-genome sequencing dataset of 30 Cuvier's gazelles including captive-bred animals, animals born in Tunisia after a reintroduction and individuals from a genetically unrelated Moroccan population. Our analyses revealed no difference between the founder and the offspring cohorts in genome-wide heterozygosity and inbreeding levels, and in the amount and length of runs of homozygosity. The captive but unmanaged Moroccan gazelles have the lowest genetic diversity of all genomes analysed. Our findings demonstrate that the Cuvier's gazelle captive breeding programme can serve as source populations for future reintroductions of this species. We believe that this study can serve as a starting point for global applications of genomics to the conservation plan of this species.

Keywords: Cuvier’s gazelle; captive breeding; genomics; inbreeding; reintroduction.

FIGURE 1

(a) Description of the dataset used in this study. Founder gazelles belong to two different captive populations, the SPA‐ALM from Estación Experimental de Zonas Áridas and the SPA‐CAN from Oasis Park Fuerteventura zoo. Founder gazelles gave rise to the first cohort born in Tunisia (F1). The second cohort (F2) originated from both the mating between only Founder gazelles or between a Founder gazelle and a female from F1. The Moroccan population is a semi‐captive but unmanaged and unrelated population of Cuvier’s gazelles. The airplane icon indicates the number of gazelles used as founder stock in the reintroduction to Tunisia and the DNA icon indicates the number of samples included in our dataset. Numbers without an icon in the offspring indicate the total number of gazelles born in each season. (b) Principal component analysis (PCA) of all gazelles sequenced in this study. SPA‐ALM gazelle in yellow, SPA‐CAN gazelles in red, TUN in blue and MOR in purple

FIGURE 2

(a) Genome‐wide heterozygosity (bp⁻¹) within groups (Founders, N = 13; Offspring F1, N = 8, Offspring F2, N = 4 and Moroccan, N = 5). ANOVA (p = 5.31 × 10⁻⁵) and Tukey HSD (p.adj_Founder‐F1 = 6.63 × 10⁻⁰³, p.adj_{Founder‐Moroccan} = 3.29 × 10⁻⁰², p.adj_F1‐F2 = 9.18 × 10⁻⁰⁴, p.adj_{F1‐Moroccan} = 4.47 × 10⁻⁰⁵). Boxplots show the median, the 25th and 75th percentiles, Tukey whiskers (median ± 1.5 times interquartile range). (b) Median proportion of the genome in RoH for each group with standard deviation represented as error bars. (Wilcoxon Test, p.adj_{Founder‐Moroccan} = 1.4 × 10⁻³, p.adj_{F1‐Moroccan} = 1.5 × 10⁻³). *p.adj < 0.05, **p.adj < 0.01, ***p.adj < 0.001, ****p.adj < 0.0001. In (a) and (b), Founder gazelles are coloured in orange, F1 in light blue, F2 in green and Moroccan gazelles in purple

FIGURE 3

(a) Correlation matrix of inbreeding coefficients. F_PED has N = 15, and the other estimates have N = 30. Colour intensity and circle size are proportional to the correlation coefficients. Values inside circles indicate correlation coefficients (see Table S2), ***p<0.001. (b) Inbreeding levels in each population estimated using the pedigree information (F_PED, N = 15) and genomic information (F_ROH1, F_ROH2, F_HOM, N = 30). Wilcoxon test with Benjamini–Hochberg (BH) correction for multiple testing (F_ROH1: p.adj_{Founder‐Moroccan} = 1.8 × 10⁻³, p.adj_{F1‐Moroccan} = 4.7 × 10⁻³; F_ROH2: p.adj_{Founder‐Moroccan} = 1.8 × 10⁻³, p.adj_{F1‐Moroccan} = 4.7 × 10⁻³, F_HOM: p.adj_{Founder‐Moroccan} = 7 × 10⁻³, p.adj_Founder‐F1 = 9.7 × 10⁻², p.adj_{F1‐Moroccan} = 4.7 × 10⁻², p.adj_F1‐F2 = 9.3 × 10⁻³). Boxplots show the median, the 25th and 75th percentiles, Tukey whiskers (median ± 1.5 times interquartile range). *p < 0.05, **p < 0.01, ***p < 0.001. Founder gazelles are coloured in orange, F1 in light blue, F2 in green and Moroccan gazelles in purple. (c) Kinship analysis of the Founder and Offspring gazelles. Kinship coefficient (Θ) of one individual with itself or a monozygotic twin equals 0.5. The lower the Θ value, the more distant is the familial relationship. Colour intensity and square size are proportional to the familial relationship estimated by NgsRelateV2