Selection, drift, gene flow and breeding have extensively
shaped the genomic variability of domestic animals. In goat species, several mutations identified within the casein genes have been shown to
affect the level of gene expression of milk production traits. The four
casein genes – CSN1S1, CSN2, CSN1S2 and CSN3 – are organized in a cluster of
250 kb located in chromosome 6, and due to tight linkage, their genetic
variability is well depicted by haplotypes which are transmitted to the
progeny. Thirty single nucleotide polymorphisms (SNPs) located within the
casein gene cluster were used to characterize the haplotype variability of
six southern Italian goat breeds (Girgentana, Maltese, Rossa Mediterranea,
Argentata dell'Etna, Messinese, Capra dell'Aspromonte). A representative
sample of the Norwegian dairy goat breed (Norsk melkegeit) has been used as
an out-group to obtain a weighted measure of genetic diversity in the
metapopulation. A total of 54 haplotypes were detected among the seven
breeds: 26, 9, 8 and 11 haplotypes were found at CSN1S1, CSN2, CSN1S2 and
CSN3 respectively. The number of haplotypes per breed was 14 (Norwegian),
26 (Messinese), 27 (Rossa Mediterranea and Girgentana) and 31 (Maltese,
Argentata dell'Etna and Capra dell'Aspromonte). The Maltese breed showed the
highest number of private haplotypes, whereas the Norwegian goat recorded the
highest number of shared haplotypes. The linkage disequilibrium analysis
showed higher levels of association for the SNP pairs within casein loci
than SNP pairs between casein loci, likely reflecting low levels of
intra-genic recombination. The highest linkage disequilibrium values were
found in CSN1S1 and CSN2 genes in all the breeds, except for Argentata
dell'Etna and Rossa Mediterranea. The resolution of the haplotype diversity
at the casein cluster can be exploited both for selective and conservative
plans.
Introduction
The casein cluster is a genomic region of particular interest in dairy
ruminants. In goat species, the genetic variability of caseins arises from
several mutations that have been shown to affect gene expression level; the
quantitative alleles modify the amount of single caseins in individual milk
and, consequently, the technological and nutritional properties of goat milk
(Marletta et al., 2007). Due to the tight association among the four casein
genes, organized in a cluster of 250 kb, the study of the haplotype
diversity, instead of single locus genotyping, seems to be preferable in
view of planning selection strategies (Haplotype Assisted Selection) (Hayes
et al., 2006; Caroli et al., 2006) and investigating breed genetic
variability (Sacchi et al., 2005; Finocchiaro et al., 2008; Gigli et al.,
2008; Küpper et al., 2010). South Italy preserves a rich heritage of
dairy goat breeds: some of them originate from the Far and Middle East
(Porter, 1996); others are derived from indigenous goats or were locally
developed through different crossbreeding strategies. All these breeds have
low population sizes; some of them are threatened by extinction and genetic
erosion. In this regard, the molecular genetics can provide effective tools to effectively describe, monitor and manage genetic resources by handling variability within
and among breeds. A recent study (Criscione et al., 2016) dealt
with the genetic characterization of these goat genetic resources by means
of a set of microsatellites (STRs), assumed to be selectively neutral
markers and described the intra- and inter-breed diversity in view of
planning conservation priorities. However, the genomic regions harbouring
quantitative loci for milk, such as casein genes, are worth investigating
in dairy breeds, in order to better analyse, characterize and explore their
genetic diversity. The resolution of the haplotype diversity at the casein
cluster can be exploited both for selective and conservative plans. The aim
of this study was to investigate the haplotype diversity of the four casein
genes' cluster in six southern Italian goat breeds (Argentata dell'Etna, Capra
dell'Aspromonte, Girgentana, Maltese, Messinese, Rossa Mediterranea); a
representative sample of the Norwegian dairy goat breed (Norsk melkegeit)
was used as an out-group to obtain a weighted measure of genetic diversity in
the metapopulation. The haplotype variability, assessed by 30 single
nucleotide polymorphisms (SNPs), was used to estimate the genetic diversity within and between, as well as the genetic relationship among these dairy
goat breeds. The use of these data for a potential casein haplotype assisted
conservation plan was also briefly discussed.
Materials and methodsSampling and SNPs analysis
A total of 207 goats and bucks representative of seven dairy goat populations
were sampled in Italy (31 Girgentana GIR, 30 Argentata dell'Etna ARG, 30 Maltese MAL, 22 Rossa Mediterranea ROS, 30 Messinese MES and 31 Capra
dell'Aspromonte ASP) and in Norway (33 Norwegian dairy goat NOR). All the
animals were reared under real commercial farm conditions. Authorized
personnel collected blood samples during the periodic veterinary control;
therefore, no pain, suffering, distress or lasting harm was caused to the
animals involved in the present study. The sampling was obtained from an
average of nine herds per population to avoid closely related individuals, also
according to the information obtained by farmers and genealogical data when
available. More information on breeds' origin, sampling regions, total
population size and the year of the studbook's establishment are reported in
Criscione et al. (2016). Blood samples were collected in 10 mL vacutainer
tubes (K3-EDTA). DNA was extracted from peripheral blood using the
Illustrablood genomic Prep Mini Spin kit (GE Healthcare, Little
Chalfont, UK). The whole sample was genotyped through a set of 30 SNP markers
located in the promoter, exonic and intronic regions of the four casein
genes (11 CSN1S1, 6 CSN2, 4 CSN1S2 and 9 CSN3). The genetic
characterization was performed using matrix-assisted laser
desorption–ionization time-of-flight mass spectroscopy (MALDI-TOF MS)
implemented in a MassARRAY System (Sequenom, San Diego, Ca, USA).
Statistical analysis
The software PHASE ver. 2.1, which implements a Bayesian statistical method,
was used to reconstruct haplotypes from the allelic frequencies within each
gene in each population (Stephens et al., 2001). The threshold frequency for
the determination of haplotypes was set to 1 %.
The level of linkage disequilibrium (LD) between all pairs of loci in each
breed was calculated through the software Haploview ver. 4.1 (Barrett et
al., 2005) using the r2 statistic (Hudson, 1985).
The deviation from Hardy–Weinberg equilibrium was investigated, for each
SNP in each population, by using the software Genepop ver. 4.2 (Raymond and
Rousset, 1995) according to the procedure of a probability test.
The frequencies of the inferred haplotypes of the four casein genes in each
goat breed were used to calculate the Nei's pairwise genetic distance (Ds) (Nei, 1972) by implementing the software Phylip 3.67 (Felsenstein, 2005). The Neighbour-Net algorithm was used to construct the phylogenetic network on Ds distances by running the software SplitsTree4 ver. 4.14.2 (Huson and Briant, 2006). The haplotypes inferred at the four casein genes in each goat breed, were used also to implement the principal component analysis (PCA) by means of the software Minitab ver. 16.1 (Minitab, Inc., 2009)
Results
The results of the SNP genotyping are shown in Table 1. Three SNPs were
found to be monomorphic in some breeds: SNP 13 in ROS, SNP 14 in GIR and NOR, SNP 19 in NOR. The distribution of the allelic frequencies was markedly
different between the Norwegian sample and the Italian breeds: in particular
20 of 30 SNPs in NOR had a frequency of the minor allele (MAF) below 20 %, whereas a maximum of 7 of 30 SNPs reported the same frequency threshold among the Italian breeds (ASP). The lowest frequency of the minor allele (MAF) was 0.017 at SNP 13 and 14 in MAL. The deletion (allele D) at
CSN1S1 exon 9 (SNP 8) was found rarely in NOR (0.076), while it showed a
frequency always above 30 % in all the Italian breeds and represented
the major allele in MAL and ROS.
SNP, gene, position, alleles (As), minor alleles (MAs)a, minor allele frequency (MAF) and p value of Hardy–Weinberg equilibrium test (HWpb) of the 30 casein SNP analysed in the seven goat breeds.
ARG ASP GIR MAL MES ROS NOR SNPGenePositionAMAMAFHWpMAMAFHWpMAMAFHWpMAMAFHWpMAMAFHWpMAMAFHWpMAMAFHWp1CSN1S1PromoterA/GA0.4670.029G0.4680.075G0.3710.006G0.3830.007A0.3830.055G0.4320.001A0.1971.0002CSN1S1PromoterA/GA0.4331.000A0.3870.444A0.5000.029G0.3830.703A0.4500.481G0.4090.071A0.1971.0003CSN1S1PromoterA/GG0.4171.000G0.3871.000G0.3550.018G0.3000.688A0.4670.724G0.2951.000A0.1971.0004CSN1S1PromoterC/TC0.4171.000C0.3871.000C0.3550.018C0.3000.688T0.4670.724C0.2951.000T0.1971.0005CSN1S1PromoterA/GG0.4171.000G0.3871.000G0.3550.018G0.3000.688A0.4670.724G0.2951.000A0.1971.0006CSN1S1Exon 4C/TT0.4171.000T0.3871.000T0.4840.070T0.3170.674C0.4670.724T0.3180.620C0.1971.0007CSN1S1Exon 4C/GC0.4330.718C0.3871.000C0.5000.029C0.3170.674G0.4670.724C0.3180.620G0.1971.0008CSN1S1Exon 9cC/DD0.4001.000D0.3391.000D0.4190.277C0.4170.706D0.3170.674C0.4320.419D0.0761.0009CSN1S1Intron 8A/GA0.4330.718A0.4030.711A0.4840.070A0.3331.000A0.4830.714A0.3180.620G0.1971.00010CSN1S1Exon 10C/GC0.4330.718C0.4030.711C0.4840.070C0.3331.000C0.4830.714C0.3180.620G0.1971.00011CSN1S1Exon 17C/TT0.1830.553T0.2900.073T0.0320.016T0.0831.000T0.2000.307T0.0911.000T0.1211.00012CSN2Exon 7C/TC0.1670.568C0.2900.378C0.0320.016C0.0671.000C0.1501.000C0.0911.000C0.0911.00013CSN2PromoterA/GG0.0671.000G0.0811.000G0.0320.016G0.017–G0.0501.000A0.000–G0.0761.00014CSN2PromoterA/GA0.0501.000A0.0321.000G0.000–A0.017–A0.0331.000A0.0451.000G0.000–15CSN2PromoterA/GG0.4500.152G0.3060.217G0.4840.070A0.4000.447G0.3500.418A0.4320.419G0.0761.00016CSN2PromoterA/TA0.4830.160A0.3391.000T0.3230.034T0.3500.418A0.3830.703T0.3410.338A0.0761.00017CSN2PromoterC/TC0.4500.476T0.3710.449C0.2900.071C0.3330.682T0.4001.000C0.3410.338T0.0761.00018CSN1S2Exon 3A/GA0.3830.009A0.3551.000A0.2740.170A0.5000.074A0.3000.688A0.4550.684A0.0911.00019CSN1S2Exon 16C/GC0.4330.010C0.4030.007G0.4840.279G0.3000.217G0.3170.014G0.3860.653C0.000–20CSN1S2Exon 16C/TT0.0671.000T0.0811.000T0.0481.000T0.0831.000T0.1670.563T0.0681.000T0.4240.28621CSN1S2Exon 16A/TT0.1671.000T0.2420.327T0.0651.000T0.2331.000T0.3330.423T0.2270.538T0.4240.28622CSN3PromoterA/GA0.2500.646A0.1940.307A0.2420.042A0.3001.000A0.3000.688A0.2500.568A0.3331.00023CSN3PromoterA/TT0.2500.646T0.2421.000T0.2740.170T0.3001.000T0.3170.674T0.2500.568T0.3790.72224CSN3PromoterA/TA0.2500.646A0.1940.307A0.2260.145A0.3001.000A0.3000.688A0.2500.568A0.3331.00025CSN3PromoterC/GC0.2500.646C0.2421.000C0.2740.170C0.3001.000C0.3170.674C0.2500.568C0.3790.72226CSN3PromoterG/TG0.2500.646G0.1940.307G0.2260.145G0.3001.000G0.3000.688G0.2500.568G0.3331.00027CSN3PromoterG/TG0.2000.559G0.1771.000G0.2420.042G0.2670.383G0.2831.000G0.2270.271G0.3331.00028CSN3PromoterC/TT0.4330.069T0.4350.157C0.2100.010C0.4500.157C0.3830.055C0.4320.103C0.4240.28629CSN3PromoterG/TT0.4330.718T0.4191.000T0.1450.093T0.3000.378T0.2330.632T0.2730.623T0.4240.28630CSN3PromoterA/GG0.2671.000G0.2740.179A0.4350.469G0.3501.000G0.4000.264G0.4321.000G0.1970.307
a The allele with the lowest frequency. b Significant p values are in bold. c Allele D at CSN1S1 exon 9 stands for allele deletion.
A sizable number (13) of SNPs were in Hardy–Weinberg disequilibrium in GIR
(Table 1), while a maximum of three SNPs were in HW disequilibrium in the other
Italian breeds, and there were none in NOR. A total of 54 haplotypes were inferred
within the four casein genes (Table 2): 26 at CSN1S1, 9 at CSN2, 8 at CSN1S2
and 11 at CSN3. In each breed, the first two haplotypes always accounted for
more than 50 % of the total frequency, ranging from 55 % to 84 % in
the Italian breeds, except for MES and ROS at CSN1S2. NOR breed showed the
highest frequency of the first two haplotypes at CSN1S1, CSN2 and CSN1S2,
always above 90 %. Private haplotypes had a frequency always below 5 % except for haplotype a16 in GIR (f=0.13). The breeds showing the
highest number of haplotypes (31) were ARG, ASP and MAL (Table 3); the
latter had also the highest number of private haplotypes. In contrast, NOR
had the least number of haplotypes (14) together with the highest percentage
of shared haplotypes.
Inferred haplotypes per casein gene of the seven goat breed analysed (ARG, ASP, GIR, MAL, MES, ROS and NOR), haplotype code (Hc)* and frequency (f) ≥ 1 %.
ARG ASP GIR MAL MES ROS NOR HaplotypeHcfHaplotypeHcfHaplotypeHcfHaplotypeHcfHaplotypeHcfHaplotypeHcfHaplotypeHcfCSN1S1GGGCGTCCACCa10.33GGGCGTCCACCa10.33GGGCGTCCACCa10.32AAATACGDGGCa20.49GGGCGTCCACCa10.45AAATACGDGGCa20.35GGGCGTCCACCa10.80AAATACGDGGCa20.24AAATACGDGGCa20.22AAATACGDGGCa20.31GGGCGTCCACCa10.26AAATACGDGGCa20.20GGGCGTCCACCa10.29AAATACGCGGTa50.12GAATACGDGGCa30.09AGATACGCGGTa60.22AGATATCCACCa160.13GGATACGCGGTa70.04GAATACGDGGCa30.10GAATACGDGGCa30.17AAATACGDGGCa20.08GGGCGTCDACCa40.07GAATACGDGGCa30.07AAATACGCGGCa170.10GAATACGDGGCa30.03AAATACGCGGTa50.09GGATACGCGGTa70.06AAATACGCGGTa50.07AAATACGCGGTa50.04GGGCGTCDACCa40.03AGATACGDGGCa140.02GGGCGTCCGGTa120.05GGATACGDGGCa210.04AGATACGCGGTa60.06GGGCGTCDACCa40.03AAATACGCGGTa50.03AAATACGCGGCa170.02AAATACGCGGCa170.03GGATATCCACCa260.02GGATACGCGGTa70.04GGGCGTCCGGTa120.02GAATACGDGGCa30.03AGATACGCGGTa60.01GGGCGTCCACTa250.03AAATACGCGGTa50.02AGATACGCGGCa80.03GAATACGCACCa130.02AGATACGDGGCa140.02AAATACGCGGTa50.01GAATACGCGGTa200.02AAATACGCGGCa170.01GGGCGCCCACCa90.02AGATACGDGGCa140.02AAATACCDGGCa180.01AGGCGTCCACCa190.01GAATACGCGGCa100.01AAATACGCACCa150.02GAATACGCGGTa200.01AAATATCCACCa110.01GGATACGDGGCa210.01AAGCGTCCACCa220.01GAATACGDACCa230.01GAGCGTCCACCa240.01CSN2TAGGATb10.44TAGATCb20.34TAGGATb10.48TAGGATb10.58TAGATCb20.46TAGGATb10.56TAGATCb20.83TAGATCb20.30TAGGATb10.30TAGATCb20.26TAGATCb20.26TAGGATb10.34TAGATCb20.26CGGATCb40.09TAGATTb30.07CAGATCb60.18TAGAATb70.19TAGAATb70.06CAGATCb60.07TAGAATb70.09TAGGATb10.08CGGATCb40.07CGGATCb40.08TAGATTb30.03CAGATCb60.03CGGATCb40.05CAAATCb50.04CAAATCb50.05CAAATCb50.03CGGATCb40.03CAAATCb50.02TAGAATb70.03CAGATCb60.04CAGATCb60.04TAGAATb70.03TAGGTTb90.02CAAATCb50.02CAGGATb80.01TAGAATb70.03TAGATTb30.03CAGGATb80.01CSN1S2GGCAc10.37GGCAc10.33GGCAc10.35ACCAc30.49GGCAc10.22AGCAc20.25GCCAc50.50AGCAc20.22AGCAc20.26GCCAc50.32GCCAc50.16ACCAc30.21ACCAc30.20GCTTc60.40ACCAc30.16GCCTc40.15AGCAc20.14GCCTc40.15GCCTc40.16GCCTc40.18ACCAc30.07GCCTc40.11ACCAc30.10ACCAc30.12GGCAc10.10GCCAc50.16GCCAc50.16ACTTc80.02GCCAc50.08GCCAc50.06GCTTc60.05GCTTc60.08GCTTc60.16GGCAc10.16GCTTc60.05GCTTc60.06GCCTc40.01AGCAc20.08GCTTc60.04GGCTc70.04CSN3GATGTTCTAd10.43GATGTTCTAd10.41GATGTTTGGd30.52GATGTTCTAd10.30GATGTTTGGd30.30GATGTTTGGd30.31GATGTTCTAd10.42ATACGGTGAd20.20GATGTTTGGd30.19ATACGGTGAd20.22ATACGGTGAd20.27ATACGGTGAd20.28GATGTTCTAd10.27ATACGGTGAd20.33GATGTTTGGd30.18ATACGGTGAd20.18GATGTTCTAd10.14GATGTTTGGd30.25GATGTTCTAd10.23ATACGGTGAd20.20GATGTTTGGd30.20GATGTTCGAd40.10GATGTTCGAd40.09GATGTTCGGd60.03GATGTTCGAd40.08GATGTTCGGd60.10GATGTTCGAd40.09GTTCTTTGAd70.05ATACGTTGGd50.05GATGTTCGGd60.06GATGTTCGAd40.03GATGTTCGGd60.07GATGTTCGAd40.05GATGTTCGGd60.07GATGTTCGGd60.03GTTCTTTGAd70.05GTTCTTTGGd80.03ATACGTTGGd50.03ATACGTTGAd100.02ATACGTTGGd50.03ATACGTTGGd50.02ATTCTGTGAd90.01ATACGGTGGd110.02
* The codes of private haplotypes are highlighted in bold type.
Number of inferred haplotypes (f≥1 %) per breed and gene, percentage of private haplotypes (% PH), and percentage of haplotypes shared among all seven goat breeds (% SH7).
In Table 4 are reported the results of LD analysis as average values within
each locus and across loci in each breed. In all the breeds, the intra-locus
LD level highlighted the highest values at CSN1S1 and then at CSN3, CSN2 and CSN1S2, in order of decreasing SNP number. NOR showed the
highest level of LD within each of the casein genes, with a very high value
at CSN1S1. Across the genes, very high LD values were found between CSN1S1
and CSN2, in all the breeds except for ARG and ROS.
Average linkage disequilibrium within and between casein loci per breed.
The matrix of pairwise Ds genetic distance (Table 5) reported NOR and MES as the most and the least distinctive breed respectively. ASP and ARG
highlighted the lowest pairwise value of the matrix, while NOR and MAL were the highest in the data set. Among the Italian goats, GIR and MAL were the most
genetically distant breeds, even if the pairwise values of Ds distance among the Italian goat populations were mostly comparable. The representation of the Ds genetic distance using the Neighbour-Net algorithm (Fig. 1) clearly shows the dichotomy between the Norwegian goats and the Italian breeds, in which the three most selected breeds (GIR, ROS and MAL) grouped in a cluster separated from the mountain populations (MES, ASP and ARG). The haplotype frequencies, condensed in the principal component analysis (PCA), showed a spatial distribution of goat breeds (Fig. 2) in which the NOR was clearly separated from the Italian goats, the most selected Sicilian breeds (GIR, MAL and ROS) were clustered closely and were separated from MES and from the ARG-ASP group.
Neighbour-Net obtained from Ds distance among the seven goat breeds.
Score plot of the principal component analysis on inferred haplotypes among the seven goat breeds.
Pairwise Nei's genetic distance (Ds) between the seven goat breeds.
The demographic history of a breed, including gene flow, drift and
bottleneck events, shapes the distribution of nucleotide polymorphisms and
the extent of genomic linkage disequilibrium (Nordborg and Tavare, 2002);
therefore, the estimate of the haplotype variation and the LD in a casein cluster
are very informative in estimating the effects of the selection, migrations
or admixture in dairy goat populations.
Among the six Italian breeds, a rather high number of haplotypes per
population was observed (from 26 in MES to 31 in ARG, ASP and MAL) with a
frequency distribution and a number of private haplotypes, which denoted a
high rate of within-breed variability and, at least partly, the absence of
systematic breeding strategies. In contrast, the NOR breed showed the least
within-breed genetic variability; the low values of the MAF for most of the
30 SNPs, the low number of haplotypes per locus and the unique private
haplotype c8 (CSN1S2 locus) described a very low variability in the casein
cluster as well as strong distinctiveness. These findings could be linked to
the selection management and the low number of founders of the Norwegian
goat breed (Hayes et al., 2006). In fact, the planned mating system that
uses buck circles on the national territory has undergone a downsizing
following the implementation of the national programme for the eradication of
caprine arthritis encephalitis virus since 2001 (Ådnøy, 2014); the
number of available Norwegian bucks has decreased, and this has probably
reduced genetic diversity. Nevertheless, the marked homogeneity of the
casein cluster highlighted here does not seem to have led to a severe
inbreeding rate, as shown by the population study through anonymous neutral
marker STRs (Criscione et al., 2016).
In the Italian breeds the percentage of shared haplotypes (never
>42.3 %) was quite low in comparison with NOR (78.6 %). These
data are partially unexpected taking into account the gene flow that can
potentially occur between Italian populations reared in the same area, in
extensive systems and in the absence of structured selective breeding programmes, but they are in agreement with the pronounced degree of biodiversity and
morphological differentiation described in these local breeds. In
particular, MAL, with its high number of haplotypes and private
combinations, seems to retain a great amount of heterogeneity mainly located
in the CSN1S1 locus. It is also worth noting the presence of a peculiar
allelic distribution on SNP 8, in which the deletion (exon 9) represents the
major allele in MAL and ROS goats. This mutation has previously been
described as common in some Mediterranean breeds, including GIR and MAL
(Sacchi et al., 2005; Gigli et al., 2008; Finocchiaro et al., 2008).
The level of linkage disequilibrium for pairs of markers within each casein
locus was higher than for pairs of markers from different loci, as reported
by Hayes et al. (2006) and Finocchiaro et al. (2008). The highest LD values
were found in CSN1S1 and CSN2 genes in all breeds, except for ARG and ROS
(Table 4).
The genetic relationship among the breeds, described according to Ds
distance, showed the phylogenetic distinctness of NOR from the Italian goat
breeds. The set of SNPs used for the genetic characterization separated
clearly the most selected dairy breeds GIR, MAL and ROS from the unselected
populations (MESS, ASP and ARG) reared in mountainous regions and generally
less productive. The spatial distribution of the seven goat breeds,
represented through the score plot of the first two components of the PCA, was highly comparable to that of the Neighbour-Net built on the Ds distance. Overall, 79 % of the variance showed two main results, as expected: the high distance between the NOR and the Italian goats as well as the clustering of the Sicilian breeds, which have been historically subjected to selection for dairy purposes (GIR, MAL and ROS). This last outcome slightly
differs from the previous study, carried out by means of 20 microsatellites,
which highlighted a clear admixture between ROS and the genetic pool of the mountainous population (Criscione et al., 2016), and it is due primarily to
the use of different unlinked and non-neutral markers, which showed a
different pattern of evolution.
Instead, given the remarkable genetic and geographical separation of NOR
from the Italian breeds, the high level of shared
haplotypes (78 %) between the Norwegian sample and the Italian breeds seems quite surprising. Due
to the great geographical distance and the reproductive isolation, an
explanation is not straightforward. Our findings confirm the low haplotypic
variability at the casein cluster detected by Finocchiaro et al. (2008) in a
different sample of GIR and NOR goats and the general assessment that goat
populations show a low genetic differentiation rate (Luikart et al., 2001;
Canon et al., 2006; Nicoloso et al., 2015). The severe shortage of
haplotypes in the NOR breed also reflects the high geographical distance
from the centre of goat domestication.
In the light of the casein haplotype diversity and according to the
relatively rich patrimony of rare private haplotypes, all the Italian breeds
are worth safeguarding. The endangered GIR, ASP and especially MAL breeds
seem to acquire conservation priorities in the context of the southern
Italian dairy goat, thanks to their degree of distinction and the private
allelic combinations they have maintained.
Conclusion
Understanding the extent, distribution and origin of current genetic
diversity in livestock populations requires diverse sources of information, including molecular data from different classes of markers. This study
analysed an important region in the goat genome: the casein cluster located in chromosome 6. The results presented here have been compared with those
obtained from the characterization of the same sampling by selectively
neutral marker STRs. The characterization by means of this set of 30 SNPs
highlighted a significant genetic diversity and differentiation among the
six dairy goat breeds reared in southern Italy. All the Italian breeds, but
mainly Maltese and to some extent Girgentana and Capra dell'Aspromonte,
showed a relevant rate of genetic distinctness, especially when compared
with the Norwegian breed. All these Italian breeds are reared in traditional management systems and in marginal areas, and some of them are
endangered. Our findings could help interpret the evolutionary history of
these breeds and represent a potential tool for the development of future
management, conservation and breeding strategies.
Data availability
The data set is available upon request from the
corresponding author.
Author contributions
AC, DM and SB designed the study and collected the blood samples. AC performed the laboratory experiments. AC and ST analysed the data and did the statistical analysis. All the authors wrote and revised the paper.
Competing interests
The authors declare that they have no conflict of
interest.
Acknowledgements
We thank Sigbjørn Lien for the kind hospitality in the CIGENE (Centre for Integrative Genetics, 1434 Ås, Norway), for the scientific skills made available and for covering the costs of genetic characterization analyses.
Financial support
This research has been supported by the University of Catania (grant no. Piano della Ricerca 2016–2018, linea di intervento 2 “dotazione ordinaria – prog. N. 4 Sist. Agroal. – genetica e alimentazione”).
Review statement
This paper was edited by Steffen Maak and reviewed by two anonymous referees.
ReferencesÅdnøy, T.: The dairy goat industry in Norway: Challenges in a historical perspective, Small Ruminant Res., 122, 4–9,
10.1016/j.smallrumres.2014.07.011, 2014.Barrett, J. C., Fry, B., Maller, J., and Daly, M. J.: Haploview: analysis and visualization of LD and haplotype maps, Bioinformatics, 21, 263–265, 10.1093/bioinformatics/bth457, 2005.Canon, J., Garcia, D., Garcia-Atance, M. A., Obexer-Ruff, G., Lenstra, J. A.,
Ajmone-Marsan, P., and Dunner, S., The ECONOGENE Consortium.: Geographical
partitioning of goat diversity in Europe and the Middle East, Anim. Genet.,
37, 327–334, 10.1111/j.1365-2052.2006.01461.x, 2006.Caroli, A., Chiatti, F., Chessa, S., Rignanese, D., Bolla, P., and Pagnacco, G.: Focusing on the goat casein complex, J. Dairy Sci., 89, 3178–3187,
10.3168/jds.S0022-0302(06)72592-9, 2006.Criscione, A., Bordonaro, S., Moltisanti, V., and Marletta, D.: Differentiation of South Italian goat breeds in the focus of biodiversity
conservation, Small Ruminant Res., 145, 12–19,
10.1016/j.smallrumres.2016.09.011, 2016.
Felsenstein, J.: PHYLIP (Phylogeny Inference Package) version 3.6,
Distributed by the author, Department of Genome Sciences, University of
Washington, Seattle, 2005.
Finocchiaro, R., Hayes, B. J., Siwek, M., Spelman, R. J., van Kaam, J. B. C. H. M., Ådnøy, T., and Portolano, B.: Comparison of casein haplotypes
between two geographically distant European dairy goat breeds, J. Anim. Breed. Genet., 125, 68–72, 2008.Gigli, I., Maizon, D. O., Riggio, V., Sardina, M. T., and Portolano, B.:
Casein haplotype variability in Sicilian dairy goat breeds, J. Dairy Sci.,
91, 3687–3692, 10.3168/jds.2008-1067, 2008.Hayes, B., Hagesæther, N., Ådnøy, T., Pellerud, G., Berg, P. R.,
and Lien, S.: Effects on production traits of haplotypes among casein genes
in Norwegian goats and evidence for a site of preferential recombination,
Genetics, 174, 455–464, 10.1534/genetics.106.058966, 2006.Hudson, R. R.: The sampling distribution of linkage disequilibrium under an
infinite allele model without selection, Genetics, 109, 611–631, 1985.
Huson, D. H. and Bryant, D.: Application of Phylogenetic Networks in
Evolutionary Studies, Mol. Biol. Evol., 23, 254–267, 10.1093/molbev/msj030, 2006.Küpper, J., Chessa, S., Rignanese, D., Caroli, A., and Erhardt, G.:
Divergence at the casein haplotypes in dairy and meat goat breeds, J. Dairy
Res., 77, 56–62, 10.1017/S0022029909990343, 2010.Luikart, G., Gielly, L., Excoffier, L., Vigne, J. D., Bouvet, J., and Taberlet, P.: Multiple maternal origins and weak phylogeographic structure
in domestic goats, P. Natl Acad. Sci. USA, 98, 5927–5932,
10.1073/pnas.091591198, 2001.Marletta, D., Criscione, A., Bordonaro, S., Guastella, A. M., and D'Urso, G.: Casein polymorphism in goat's milk, Le Lait., 87, 491–504,
10.1051/lait:2007034, 2007.Minitab, Inc.: Minitab Statistical Software, Release 16 for Windows, State
College, Pennsylvania, Minitab® is a registered trademark of Minitab, Inc., available at: http://www.minitab.com/en-us/ (last access: July 2019), 2009.
Nei, M.: Genetic distance between populations, Am. Nat., 106, 283–292, 1972.Nicoloso, L., Bomba, L., Colli, L., Negrini, R., Milanesi, M., Mazza, R., Sechi, T., Frattini, S., Talenti, A., Coizet, B., Chessa, S., Marletta, D.,
D'Andrea, M., Bordonaro, S., Ptak, G., Carta, A., Pagnacco, G., Valentini, A., Pilla, F., Ajmone-Marsan, P., and Crepaldi, P.: Genetic diversity of
Italian goat breeds assessed with a medium-density SNP chip, Genet. Sel. Evol., 47, 1–10, 10.1186/s12711-015-0140-6, 2015.Nordborg, M. and Tavare, S.: Linkage disequilibrium: what history has to tell us, Trends Genet., 18, 83–90, 10.1016/S0168-9525(02)02557-X, 2002.
Porter, V.: Goats of the World, Farming Press, Ipswich, UK, 1996.Raymond, M. and Rousset, F.: GENEPOP Version 1.2: population genetics software for exact tests and ecumenicism, J. Hered., 86, 248–249,
10.1093/oxfordjournals.jhered.a111573, 1995.Sacchi, P., Chessa, S., Budelli, E., Bolla, P., Ceriotti, G., Soglia, D., Rasero, R., Cauvin, E., and Caroli, A.: Casein haplotype structure in five
Italian goat breeds, J. Dairy Sci., 88, 1561–1568,
10.3168/jds.S0022-0302(05)72825-3, 2005.Stephens, M., Smith, N. J., and Donnelly, P.: A new statistical method for
haplotype reconstruction from population data, Am. J. Hum. Genet., 68, 978–989, 10.1086/319501, 2001.