A Genetic Investigation of Island Jersey Cattle, the Foundation of the Jersey Breed: Comparing Population Structure and Selection to Guernsey, Holstein, and United States Jersey Cattle

Abstract

For two centuries, Jersey cattle were exported globally, adapting to varying climates and production systems, yet the founding population remained genetically isolated on the Island of Jersey. The Island of Jersey formally allowed the importation of pure Jersey cattle in 2008. This study characterized the genetic variation of 49 popular bulls from the Island of Jersey born from 1964 to 2004 and compared them to 47 non-Island Jersey bulls and cows, primarily from the United States In addition, 21 Guernsey cattle derived from the Island of Guernsey and 71 Holstein cattle served as reference populations for genetic comparison. Cattle were genotyped on the Illumina BovineHD Beadchip producing 777,962 SNPs spanning the genome. Principal component analysis revealed population stratification within breed reflective of individual animal's continental origin. When compared to Holstein and Guernsey, all Jersey clustered together by breed. The Jersey breed demonstrated increased inbreeding in comparison to Holstein or Guernsey with slightly higher estimates of inbreeding coefficients and identity-by-descent. The Island and United States Jersey have relatively similar, yet statistically different inbreeding estimates despite vastly different population sizes and gene flow. Signatures of selection within Island Jersey were identified using genome-wide homozygosity association and marker-based F_ST that provided population informative single-nucleotide polymorphism (SNPs). Biological significance of the homozygosity association results identified multiple genes on chromosomes 5, 24, and 27, involved in immune function and cellular processes. Overall, genomic variation was identified between the Island and non-Island Jersey cattle producing population informative SNPs and differing runs of homozygosity (ROH) over immune regulation and metabolic genes. Results on inbreeding measures and ROH may reflect varying effective population size or differential selection with grazing systems promoting natural selection for traits such as parasite resistance, whereas confinement systems demonstrate a more intensive artificial selection. More broadly, differences in breed formation, particularly between the two Channel Island breeds, likely contributed to the variation in ROH and inbreeding. This research provides a reference for the Jersey breed based on the genetic foundation of the Island cattle as compared to the intensively selected United States cattle, and identifies regions of the genome for future investigation of immune regulation and metabolic processes.

Keywords: FST; Jersey; Jersey Island; dairy cattle; runs of homozygosity; signatures of selection.

FIGURE 1

Principal Component Analysis showing clustering reflective of breed and sub-population. (A) Scatter plot of PCA with PC 1 on the x-axis, distinguishing Holstein (right-orange) and Jersey subpopulations (left-purple). PC 2 on the y-axis, distinguishes Guernsey from the other breeds. (B) Box-plots showing variation of principal components 1 (top), 2 (middle), and 3 (bottom) with individuals grouped by breed; Holstein and Guernsey or Jersey sub-populations; Danish (DNK), Island (ISL), New Zealand (NZL), and United States (USA). The ⁺ denote outlier samples in the dataset.

FIGURE 2

Principal Component Analysis of Jersey sub-populations based on country registration of animal. (A) Scatter plot with PC 1 on the x-axis, distinguishing Island (left-red) and United States (right-blue). PC 2 on the y-axis, distinguishes Danish. (B) Box-plots showing variation of principal components 1 (top), 2 (middle), and 3 (bottom) with individuals grouped by Jersey sub-populations; Canadian (CAN), Danish (DNK), Island (ISL), New Zealand (NZL), and United States (USA). The ⁺ denote outlier samples in the dataset.

FIGURE 3

Admixture analysis reflecting genetic clustering of breed; Holstein, Guernsey, and Jersey and geographic origins of Jersey subpopulations; Island and United States. Individual vertical bars along the x-axis represent individual cattle which are grouped by breed. Genetic clusters corresponding to breed or population are denoted as follows: Guernsey (light/aqua blue), Holstein (orange), Jersey (purple), Island Jersey (ISL-red), and United States Jersey (USA-blue). Jersey cattle originating from Canada (CAN), Denmark (DNK), and New Zealand (NZL) are denoted in the bottom legend but do not show a unique genetic cluster. The y-axis provides a measure of the percentage of each genetic population found within an individual. K represents the number of genetic populations used in each analysis with (A) showing K = 2, (B) showing K = 3, and (C) showing K = 4.

FIGURE 4

The average genetic admixture within each breed or Jersey subpopulation. Individual ADMIXTURE scores were averaged by breed or subpopulation designation. Genetic clusters corresponding to breed or population are denoted as follows: Guernsey (n = 21, light/aqua blue), Holstein (n = 65, orange), Jersey (n = 95, purple), Island Jersey (n = 49, ISL-red), and United States Jersey (n = 34, USA-blue). K represents the number of genetic populations used in each analysis with (A) showing K = 3 and (B) showing K = 4.

FIGURE 5

Signatures of selection are depicted using a Manhattan plot of marker-based F_ST values comparing Island Jersey and non-Island Jersey. The x-axis denotes the chromosome and the y-axis denotes the F_ST score for the SNP marker. The black horizontal line denotes five standard deviations above the mean marker F_ST value of 0.05.

FIGURE 6

The mean sum of Run of Homozygosity (ROH) per genotyped animal within each population. The x-axis denotes the ROH length category (Mb) and the y-axis denotes the mean sum of ROH. Population designation is based on breed and Jersey sub-populations (ISL- Jersey Island, USA- United States) with unique color bars representing each population. Eighty-three individuals represented the Jersey breed. Actual mean sum of ROH are shown below the bar graph to assist in visualization of ROH lengths.

FIGURE 7

Genome-wide homozygosity association analysis comparing 107 runs of homozygosity clusters between Island and non-Island Jersey. Each dot represents the first SNP within the ROH cluster analyzed and is plotted according to genome location and level of association. Horizontal black line signifies a false discovery rate <0.05.

FIGURE 8

Population divergence between Guernsey, Holstein, and Jersey sub-populations based on fixation index F_ST scores. F_ST scores range from 0, no genetic divergence, to 1 representing complete genetic isolation. The matrix style heatmap shows pairwise comparisons of each population and colors range from red = 0 to blue = 0.17 (highest F_ST score).

FIGURE 9

Average inbreeding estimations. Average inbreeding coefficient, f, and inbreeding estimate based on ROH measures, FROH. Population designation based on breed and pedigree for Jersey sub-populations. ISL, Jersey Island; USA, United States.

FIGURE 10

Variation in the average inbreeding coefficient, f, during each decade per population. Hardy-Weinberg Equilibrium is represented by a f value of zero (horizontal red line). Inbreeding coefficients below zero are representative of an excess of heterozygosity while those above zero are representative of an excess of homozygosity reflective of inbreeding. Breeds are represented by solid lines and Jersey sub-populations are represented by dotted lines. The absence of a line during a time period means that no individuals were analyzed from the said populations from that decade.