Sequence analysis of a pea comb locus on chicken chromosome 1.

To facilitate gene identification, this study aimed to narrow the scope of the genome region affecting chicken comb type by using two bird populations. First, an F2 resource population was generated by crossing Japanese game fowl (Shamo; pea comb, P/p and P/P) with White Plymouth Rock (single comb, p/p). Comb types of the 240 F2 offspring produced by an F1 intercross between eight males and 57 females were segregated at a ratio of 3:1 (pea:single). The pea comb locus was mapped to a chromosomal region on Gallus gallus chromosome 1 that was flanked by microsatellite markers MCW0112, MCW0019 and ABR521. The second population (five-generation, n=1300 animals) was derived from a cross between Shamo and Rhode Island Red (single comb, p/p) that had been genotyped for additional polymorphic single nucleotide polymorphisms and microsatellite markers within this region through development of chicken draft sequences. To close some gaps in these draft sequences, we constructed a bacterial artificial chromosome contig and sequenced it using the shotgun sequencing technique. Chickens selected from pedigrees in these populations were grouped by inheritance of a P or p haplotype at the locus constructed by the additional markers. Finally, this locus was fine-mapped to roughly 60 kb based on the association of haplotypes and comb types. Chicken genome sequences suggest that the most likely polymorphism responsible for the pea comb locus is a duplicated sequence and that the sex determining region Y-box 5 gene, one predicted gene and one expressed sequence tag in a critical region may be associated with the duplicated sequence.

Mapping of the recessive white locus and analysis of the tyrosinase gene in chickens.

An F(2) chicken population of 265 individuals, obtained from an intercross between the Japanese Game (colored plumage) and the White Plymouth Rock (the recessive white) and genotyped for microsatellite markers, was used for determining the locus of the gene responsible for the recessive white plumage phenotype in chickens. Two hundred twenty-five markers were mapped in 28 linkage groups. Linkage analysis revealed that the recessive white gene was mapped to chromosome 1. Detailed analysis using additional markers uncovered a significant linkage between 2 new markers, mapped to the flanking region of the tyrosinase gene, which is associated with skin and plumage color. The sequence of the tyrosinase gene was investigated in recessive white chickens and colored chickens. There were no obvious differences in the tyrosinase gene exons between the recessive white chicken and the colored chicken. However, sequence analysis of tyrosinase intron 4 in the recessive white chicken revealed a presence of an insertion of an avian retroviral sequence. The White Plymouth Rock and the F(2) generation with white plumage were identified as homozygous carriers of the retroviral sequence. Expression of the normal transcript containing exon 5 was substantially decreased in the recessive white chicken compared with the colored chicken. Some abnormal tyrosinase gene transcripts were expressed in the skin of the White Plymouth Rock: reverse transcription PCR products amplified from exon 3 to intron 4 and from retroviral sequence 3' long terminal repeat to exon 5. Based on these results, it was confirmed that an avian retroviral sequence insertion in the tyrosinase gene was the cause of recessive white phenotype in chickens.

Effect of genotyped cows in the reference population on the genomic evaluation of Holstein cattle.

This study evaluated the dependence of reliability and prediction bias on the prediction method, the contribution of including animals (bulls or cows), and the genetic relatedness, when including genotyped cows in the progeny-tested bull reference population. We performed genomic evaluation using a Japanese Holstein population, and assessed the accuracy of genomic enhanced breeding value (GEBV) for three production traits and 13 linear conformation traits. A total of 4564 animals for production traits and 4172 animals for conformation traits were genotyped using Illumina BovineSNP50 array. Single- and multi-step methods were compared for predicting GEBV in genotyped bull-only and genotyped bull-cow reference populations. No large differences in realized reliability and regression coefficient were found between the two reference populations; however, a slight difference was found between the two methods for production traits. The accuracy of GEBV determined by single-step method increased slightly when genotyped cows were included in the bull reference population, but decreased slightly by multi-step method. A validation study was used to evaluate the accuracy of GEBV when 800 additional genotyped bulls (POPbull) or cows (POPcow) were included in the base reference population composed of 2000 genotyped bulls. The realized reliabilities of POPbull were higher than those of POPcow for all traits. For the gain of realized reliability over the base reference population, the average ratios of POPbull gain to POPcow gain for production traits and conformation traits were 2.6 and 7.2, respectively, and the ratios depended on heritabilities of the traits. For regression coefficient, no large differences were found between the results for POPbull and POPcow. Another validation study was performed to investigate the effect of genetic relatedness between cows and bulls in the reference and test populations. The effect of genetic relationship among bulls in the reference population was also assessed. The results showed that it is important to account for relatedness among bulls in the reference population. Our studies indicate that the prediction method, the contribution ratio of including animals, and genetic relatedness could affect the prediction accuracy in genomic evaluation of Holstein cattle, when including genotyped cows in the reference population.

Bovine quantitative trait loci analysis for growth, carcass, and meat quality traits in an F2 population from a cross between Japanese Black and Limousin.

A genome-wide scan for QTL affecting economically important traits in beef production was performed using an F(2) resource family from a Japanese Black x Limousin cross, where 186 F(2) animals were measured for growth, carcass, and meat-quality traits. All family members were genotyped for 313 informative microsatellite markers that spanned 2,382 cM of bovine autosomes. The centromeric region of BTA2 contained significant QTL (i.e., exceeding the genome-wide 5% threshold) for 5 carcass grading traits [LM area, beef marbling standards (BMS) number, luster, quality grade, and firmness), 8 computer image analysis (CIA) traits [LM lean area, ratio of fat area (RFA) to LM area, LM area, RFA to musculus (M.) trapezius area, M. trapezius lean area, M. semispinalis lean area, RFA to M. semispinalis area, and RFA to M. semispinalis capitis area], and 5 meat quality traits (contents of CP, crude fat, moisture, C16:1, and C18:2 of LM). A significant QTL for withers height was detected at 80.3 cM on BTA5. We detected significant QTL for the C14:0 content in backfat and C14:0 and C14:1 content in intermuscular fat around the 62.3 to 71.0 cM region on BTA19 and for C14:0, C14:1, C18:1, and C16:0 content and ratio of total unsaturated fatty acid content to total SFA content in intramuscular fat at 2 different regions on BTA19 (41.1 cM for C14:1 and 62.3 cM for the other 4 traits). Overall, we identified 9 significant QTL regions controlling 27 traits with genome-wide significance of 5%; of these, 22 traits exceeded the 1% genome-wide threshold. Some of the QTL affecting meat quality traits detected in this study might be the same QTL as previously reported. The QTL we identified need to be validated in commercial Japanese Black cattle populations.