Molecular genetic approaches for identifying the basis of variation in resistance to tick infestation in cattle

In recent years there has been renewed interest in the adaptation of cattle to challenging environments, largely driven by advances in genomic methods. The current interest in tick resistance is understandable given the major production and welfare implications of tick infestation in tropical and subtropical zones where around 70% of beef cattle are located. Heritability for tick burden in cattle has been shown to range about 0.30, which is sufficient to result in the success of some programs of selection for tick resistance in cattle. Gene-expression studies strongly indicate that both immune and non-immune mechanisms are associated with tick resistance in cattle. Recent quantitative-trait mapping studies have identified chromosome segments and single nucleotide polymorphisms associated with tick burden, but no causal variant has been identified so far. Most of the genetic markers identified for tick burden explain a relatively small proportion of the variance, which is typical of markers for quantitative traits. This leads to the conclusion that panels of multiple markers for tick resistance rather than a single marker will most likely be developed, possibly involving specific panels for zebu or taurine breeds, which could be used for future selection and breeding programs in cattle.

The effect of measurement error of phenotypes on genome wide association studies

Background

There is an unspoken assumption that imprecision of measurement of phenotypes will not have large systematic effects on the location of significant associations in a genome wide association study (GWAS). In this report, the effects of two independent measurements of the same trait, subcutaneous fat thickness, were examined in GWAS of 940 individuals.

Results

The trait values obtained by two independent groups working to the same trait definition were correlated with r = 0.72. The allele effects obtained from the two analyses were only moderately correlated, with r = 0.53, and there was one significant (P < 0.0001) association in common to the two measurements. The correlation between allele effects was approximately equal to the square of the correlation between the trait measurements. An important quantitative trait locus (QTL) on BTA14 appeared to be shifted distally by 1 Mb along the chromosome. The divergence in GWAS was stronger with data coded into two discrete classes. Univariate trimming of the top and bottom 5% of data, a method used to control for erroneous trait values, decreased the similarity between the GWAS and increased the apparent shift of the QTL on BTA14. Stringent bivariate trimming of data, using only trait values that were similar to each other in the two data sets, substantially improved the correlation of trait values and allele effects in the GWAS, and showed evidence for two QTL on BTA14 separated by 1 Mb. Despite the reduction in sample size due to trimming, more SNP were significant. Using the mean of the two measurements of the trait was not as efficient as bivariate trimming.

Conclusions

It is recommended that trait values in GWAS experiments be examined for repeatability before the experiment is performed. For traits that do not have high repeatability (r < 0.95), two or more independent measurements of the same trait should be obtained for all samples, and individuals genotyped that have highly correlated trait measurements.

A genome-wide association study of meat and carcass traits in Australian cattle

Chromosomal regions containing DNA variation affecting the traits intramuscular fat percentage (IMF), meat tenderness measured as peak force to shear the LM (LLPF), and rump fat measured at the sacro-iliac crest in the chiller (CHILLP8) were identified using a set of 53,798 SNP genotyped on 940 taurine and indicine cattle sampled from a large progeny test experiment. Of these SNP, 87, 64, and 63 were significantly (P < 0.001) associated with the traits IMF, LLPF, and CHILLP8, respectively. A second, nonoverlapping sample of 1,338 taurine and indicine cattle from the same large progeny test experiment genotyped for 335 SNP, including as a positive control the calpastatin (CAST) c.2832A > G SNP, was used to confirm these locations. In total, 37 SNP were significantly (P < 0.05) associated with the same trait and with the same favorable homozygote in both data sets, representing 27 chromosomal regions. For the trait IMF, the effect of SNP in the confirmation data set was predicted from the discovery set by multiplying the estimated allele effect of each SNP in the discovery set by the number of copies of the reference allele of each SNP in the confirmation set. These weighted effects were then summed over all SNP to generate a molecular breeding value (MBV) for each animal in the confirmation data set. Using a bivariate analysis of MBV and IMF phenotypes of animals in the confirmation set, a panel of 14 SNP explained 5.6 and 15.6% of the phenotypic and genetic variance of IMF, respectively, in the confirmation data set. The amount of variation did not increase as more SNP were added to the MBV and instead decreased to 1.2 and 3.8% of the phenotypic and genetic variance of IMF, respectively, when 329 SNP were included in the analysis.

Genome-wide association studies for feedlot and growth traits in cattle

A genome wide-association study for production traits in cattle was carried out using genotype data from the 10K Affymetrix (Santa Clara, CA) and the 50K Illumina (San Diego, CA) SNP chips. The results for residual feed intake (RFI), BW, and hip height in 3 beef breed types (Bos indicus, Bos taurus, and B. indicus × B. taurus), and for stature in dairy cattle, are presented. The aims were to discover SNP associated with all traits studied, but especially RFI, and further to test the consistency of SNP effects across different cattle populations and breed types. The data were analyzed within data sets and within breed types by using a mixed model and fitting 1 SNP at a time. In each case, the number of significant SNP was more than expected by chance alone. A total of 75 SNP from the reference population with 50K chip data were significant (P < 0.001) for RFI, with a false discovery rate of 68%. These 75 SNP were mapped on 24 different BTA. Of the 75 SNP, the 9 most significant SNP were detected on BTA 3, 5, 7, and 8, with P ≤ 6.0 × 10(-5). In a population of Angus cattle divergently selected for high and low RFI and 10K chip data, 111 SNP were significantly (P < 0.001) associated with RFI, with a false discovery rate of 7%. Approximately 103 of these SNP were therefore likely to represent true positives. Because of the small number of SNP common to both the 10K and 50K SNP chips, only 27 SNP were significantly (P < 0.05) associated with RFI in the 2 populations. However, other chromosome regions were found that contained SNP significantly associated with RFI in both data sets, although no SNP within the region showed a consistent effect on RFI. The SNP effects were consistent between data sets only when estimated within the same breed type.

Bovine proteins containing poly-glutamine repeats are often polymorphic and enriched for components of transcriptional regulatory complexes

Background

About forty human diseases are caused by repeat instability mutations. A distinct subset of these diseases is the result of extreme expansions of polymorphic trinucleotide repeats; typically CAG repeats encoding poly-glutamine (poly-Q) tracts in proteins. Polymorphic repeat length variation is also apparent in human poly-Q encoding genes from normal individuals. As these coding sequence repeats are subject to selection in mammals, it has been suggested that normal variations in some of these typically highly conserved genes are implicated in morphological differences between species and phenotypic variations within species. At present, poly-Q encoding genes in non-human mammalian species are poorly documented, as are their functions and propensities for polymorphic variation.

Results

The current investigation identified 178 bovine poly-Q encoding genes (Q ≥ 5) and within this group, 26 genes with orthologs in both human and mouse that did not contain poly-Q repeats. The bovine poly-Q encoding genes typically had ubiquitous expression patterns although there was bias towards expression in epithelia, brain and testes. They were also characterised by unusually large sizes. Analysis of gene ontology terms revealed that the encoded proteins were strongly enriched for functions associated with transcriptional regulation and many contributed to physical interaction networks in the nucleus where they presumably act cooperatively in transcriptional regulatory complexes. In addition, the coding sequence CAG repeats in some bovine genes impacted mRNA splicing thereby generating unusual transcriptional diversity, which in at least one instance was tissue-specific. The poly-Q encoding genes were prioritised using multiple criteria for their likelihood of being polymorphic and then the highest ranking group was experimentally tested for polymorphic variation within a cattle diversity panel. Extensive and meiotically stable variation was identified.

Conclusions

Transcriptional diversity can potentially be generated in poly-Q encoding genes by the impact of CAG repeat tracts on mRNA alternative splicing. This effect, combined with the physical interactions of the encoded proteins in large transcriptional regulatory complexes suggests that polymorphic variations of proteins in these complexes have strong potential to affect phenotype.

DNA variation in the gene ELTD1 is associated with tick burden in cattle

Ticks and tick-born diseases are major constraints on cattle production in tropical and subtropical regions in the world. Previously, we identified single nucleotide polymorphisms (SNPs) associated with tick resistance on bovine chromosome 3 at approximately 70 Mb. In this study, we genotyped a dairy (n = 1133) and a beef (n = 774) sample to confirm the association of the intronic SNP rs29019303 and its gene (ELTD1) with tick burden. We genotyped 18 additional SNPs in a region of 181 kb and found that rs29019303 was significantly (P < 0.05) associated with tick burden in both samples with the same favourable allele. A second SNP in this same genomic region was also significantly associated with tick burden in each sample. The associations using haplotypes were stronger than for single markers, including a haplotype of nine tag SNPs that was highly significantly (P = 0.0008) associated with tick counts in the dairy animals. This haplotype and two others were significant after Bonferroni correction for multiple testing. The estimated size of the effects was close to 0.9% of the residual variance in both samples tested.

Effect of SNP origin on analyses of genetic diversity in cattle

The methods of single nucleotide polymorphism (SNP) identification can lead to ascertainment bias, which will affect population genetic analyses based on those data. In livestock species, the methods of SNP identification through genome sequencing are likely to suffer from this ascertainment bias. In the present study, a subset of data from the Bovine HapMap Project was re-analysed to quantify the effects of ascertainment bias on a range of common analyses and statistics. Data from 189 animals of the zebu breeds Brahman, Nelore and Gir, taurine beef Angus, Limousin and Hereford and taurine dairy Holstein, Jersey and Brown Swiss were analysed. There were 141 SNPs each of Angus, Brahman and Holstein origin, giving a total of 423 SNPs organised in 141 triplets. Each triplet consisted of one SNP of each breed, separated on average by 0.75 Mb within each triplet and where triplets were separated by 14.96 Mb to ensure that each triplet was unaffected by linkage disequilibrium. The minor allele frequency distribution, estimates of the F-statistic, FST, the partitioning of variance and population substructure were relatively unaffected by breed of origin of the SNPs. Estimates of heterozygosity were significantly affected by breed of origin of the SNPs. The clustering of animals of closely related breeds varied in the principal component analyses (PCA). However, in the PCA the effect of breed of origin of 141 SNPs was similar to the effect of using different panels of 141 SNPs of all three breeds, so the differences found in the PCA may not be all due to bias by the origin of the SNPs. Based on these results, analyses that depend on FST, including signatures of selection, gene flow and effective population size are unlikely to be strongly affected by SNP origin. Analyses that partition genetic variance and some analyses of population substructure will also be largely unaffected. However, analyses that are dependent on locus heterozygosity, which can be used for studying population bottlenecks, or those that study selection using extended haplotype homozygosity may be significantly affected by breed of origin of the SNPs.

A genome-wide association study of tick burden and milk composition in cattle

To study the genetic basis of tick burden and milk production and their interrelationship, we collected a sample of 1961 cattle with multiple tick counts from northern Australia of which 973 had dairy production data in the Australian Dairy Herd Information Service database. We calculated heritabilities, genetic and phenotypic correlations for these traits and showed a negative relationship between tick counts and milk and milk component yield. Tests of polymorphisms of four genes associated with milk yield, ABCG2, DGAT1, GHR and PRLR, showed no statistically significant effect on tick burden but highly significant associations to milk component yield in these data and we confirmed separate effects for GHR and PRLR on bovine chromosome 20. To begin to identify some of the molecular genetic bases for these traits, we genotyped a sample of 189 of these cattle for 7397 single nucleotide polymorphisms in a genome-wide association study. Although the allele effects for adjusted milk fat and protein yield were highly correlated (r = 0.66), the correlations of allele effects of these milk component yields and tick burden were small (|r| ≤ 0.10). These results agree in general with the phenotypic correlations between tick counts and milk component yield and suggest that selection on markers for tick burden or milk component yield may have no undesirable effect on the other trait.

A splice site single nucleotide polymorphism of the fatty acid binding protein 4 gene appears to be associated with intramuscular fat deposition in longissimus muscle in Australian cattle

Fatty acid binding protein 4 (FABP4) is a candidate gene affecting fatness traits of mammals. However, its association with fatness traits in cattle and other livestock species is not consistent from one study to another. Here, we sequenced the coding sequence of FABP4 looking for non-synonymous variants. We identified a splice site mutation between the third exon and the third intron of bovine FABP4. We genotyped this SNP, FABP4:g.2502C>G, in 1409 cattle with intramuscular fat measurements from seven breeds. The average allele frequency of the C allele was 0.66 with a range of 0.45 to 0.85. A regression on the number of G alleles shows a statistically significant effect of alpha = 0.11, P = 0.044. This appears to confirm an association between IMF and variation at FABP4, with an effect of 0.3% of the variation in our sample when using this SNP.

Genome-wide survey of SNP variation uncovers the genetic structure of cattle breeds

The imprints of domestication and breed development on the genomes of livestock likely differ from those of companion animals. A deep draft sequence assembly of shotgun reads from a single Hereford female and comparative sequences sampled from six additional breeds were used to develop probes to interrogate 37,470 single-nucleotide polymorphisms (SNPs) in 497 cattle from 19 geographically and biologically diverse breeds. These data show that cattle have undergone a rapid recent decrease in effective population size from a very large ancestral population, possibly due to bottlenecks associated with domestication, selection, and breed formation. Domestication and artificial selection appear to have left detectable signatures of selection within the cattle genome, yet the current levels of diversity within breeds are at least as great as exists within humans.

Genome wide signatures of positive selection: the comparison of independent samples and the identification of regions associated to traits

Background

The goal of genome wide analyses of polymorphisms is to achieve a better understanding of the link between genotype and phenotype. Part of that goal is to understand the selective forces that have operated on a population.

Results

In this study we compared the signals of selection, identified through population divergence in the Bovine HapMap project, to those found in an independent sample of cattle from Australia. Evidence for population differentiation across the genome, as measured by F_ST, was highly correlated in the two data sets. Nevertheless, 40% of the variance in F_ST between the two studies was attributed to the differences in breed composition. Seventy six percent of the variance in F_ST was attributed to differences in SNP composition and density when the same breeds were compared. The difference between F_ST of adjacent loci increased rapidly with the increase in distance between SNP, reaching an asymptote after 20 kb. Using 129 SNP that have highly divergent F_ST values in both data sets, we identified 12 regions that had additive effects on the traits residual feed intake, beef yield or intramuscular fatness measured in the Australian sample. Four of these regions had effects on more than one trait. One of these regions includes the R3HDM1 gene, which is under selection in European humans.

Conclusion

Firstly, many different populations will be necessary for a full description of selective signatures across the genome, not just a small set of highly divergent populations. Secondly, it is necessary to use the same SNP when comparing the signatures of selection from one study to another. Thirdly, useful signatures of selection can be obtained where many of the groups have only minor genetic differences and may not be clearly separated in a principal component analysis. Fourthly, combining analyses of genome wide selection signatures and genome wide associations to traits helps to define the trait under selection or the population group in which the QTL is likely to be segregating. Finally, the F_ST difference between adjacent loci suggests that 150,000 evenly spaced SNP will be required to study selective signatures in all parts of the bovine genome.

Bovine and ovine DNA microsatellites from the EMBL and GENBANK databases

Bovine and ovine microsatellite sequences were extracted from the EMBL and GENBANK databases. When analysed for number of alleles and degree of heterozygosity in the CSIRO cattle reference families, allele numbers range from 1 to 14 with heterozygosities, in the polymorphic systems ranging from 15.8% to 100%. Six (46%) of the 13 bovine systems tested gave specific and polymorphic products in sheep. Similarly 2 of the 4 ovine systems gave specific and polymorphic products in cattle. These data define 11 bovine and 8 ovine microsatellite systems which are associated with known genes and are thus useful for comparative mapping studies.

Variation at CPE but not CEBPA appears to be associated with intramuscular fat deposition in the longissimus muscle of cattle

An important step in the localisation of quantitative trait loci is the confirmation of trait-marker associations in independent studies. In this report, we test three single nucleotide polymorphisms (SNP) of two genes for associations to intramuscular fat (IMF) measurements in cattle. We genotyped SNP of carboxypeptidase E (CPE) and ccaat/enhancer binding protein, α (CEBPA) in a sample of a total of 813 cattle of taurine, composite and indicine breeds. All three polymorphisms showed significant differences between breeds, with the widest range found in CEBPA:g.271A > C where the A allele frequency ranged from P = 0.07 in Brahman to 0.88 in Shorthorn. The taurine breeds showed high linkage disequilibrium between the pair of CPE SNP, with all four breeds showing r2 = 1.0. The Brahman and Santa Gertrudis showed r2 ≤ 0.17. Both CPE:g.445C > T and CPE:g.601C > T SNP showed significant allele substitution effects to IMF in animals of taurine ancestry, with an allele substitution effect of α = 0.22, P = 0.020 for CPE:g.445C > T, explaining 0.4% of the phenotypic variance.

Variation at the Calpain 3 gene is associated with meat tenderness in zebu and composite breeds of cattle

Background

Quantitative Trait Loci (QTL) affecting meat tenderness have been reported on Bovine chromosome 10. Here we examine variation at the Calpain 3 (CAPN3) gene in cattle, a gene located within the confidence interval of the QTL, and which is a positional candidate gene based on the biochemical activity of the protein.

Results

We identified single nucleotide polymorphisms (SNP) in the genomic sequence of the CAPN3 gene and tested three of these in a sample of 2189 cattle. Of the three SNP genotyped, the CAPN3:c.1538+225G>T had the largest significant additive effect, with an allele substitution effect in the Brahman of α = -0.144 kg, SE = 0.060, P = 0.016, and the polymorphism explained 1.7% of the residual phenotypic variance in that sample of the breed. Significant haplotype substitution effects were found for all three breeds, the Brahman, the Belmont Red, and the Santa Gertrudis. For the common haplotype, the haplotype substitution effect in the Brahman was α = 0.169 kg, SE = 0.056, P = 0.003. The effect of this gene was compared to Calpastatin in the same sample. The SNP show negligible frequencies in taurine breeds and low to moderate minor allele frequencies in zebu or composite animals.

Conclusion

These associations confirm the location of a QTL for meat tenderness in this region of bovine chromosome 10. SNP in or near this gene may be responsible for part of the overall difference between taurine and zebu breeds in meat tenderness, and the greater variability in meat tenderness found in zebu and composite breeds. The evidence provided so far suggests that none of these tested SNP are causative mutations.