Fine mapping of a swine quantitative trait locus for number of vertebrae and analysis of an orphan nuclear receptor, germ cell nuclear factor (NR6A1)

Abstract

The number of vertebrae in pigs varies and is associated with meat productivity. Wild boars, which are ancestors of domestic pigs, have 19 vertebrae. In comparison, European commercial breeds have 21-23 vertebrae, probably owing to selective breeding for enlargement of body size. We previously identified two quantitative trait loci (QTL) for the number of vertebrae on Sus scrofa chromosomes (SSC) 1 and 7. These QTL explained an increase of more than two vertebrae. Here, we performed a map-based study to define the QTL region on SSC1. By using three F2 experimental families, we performed interval mapping and recombination analyses and defined the QTL within a 1.9-cM interval. Then we analyzed the linkage disequilibrium of microsatellite markers in this interval and found that 10 adjacent markers in a 300-kb region were almost fixed in European commercial breeds. Genetic variation of the markers was observed in Asian local breeds or wild boars. This region encoded an orphan nuclear receptor, germ cell nuclear factor (NR6A1, formerly known as GCNF), which contained an amino acid substitution (Pro192Leu) coincident with the QTL. This substitution altered the binding activity of NR6A1 to its corepressors, nuclear receptor-associated protein 80 (RAP80) and nuclear receptor corepressor 1 (NCOR1). In addition, somites of mouse embryos demonstrated expression of NR6A1 protein. Together, these results suggest that NR6A1 is a strong candidate for one of the QTL that influence number of vertebrae in pigs.

Figure 1.

Dissection of a QTL region on SSC1 for vertebral number in F₂ families. (A) Plots of F-ratio for interval mapping analyses in the Large White × Japanese wild boar population (red) and Jinhua × Duroc cross population (black). For the latter, the QTL effect on SSC7 was removed by incorporating the genotype of a marker in the QTL region into the model as a covariate. Bootstrap analysis (with 1000 repetitions) was performed to obtain the 95% confidence interval of the estimated QTL position on SSC1, and the results are shown in the inset bar graphs. Double-ended arrows indicate the 95% confidence interval in each population. (B) QTL type of an F₁ dam with a recombination in the QTL region. The dam was from a Meishan × Göttingen miniature cross. Meishan pigs were wt/wt, and the Göttingen miniature was heterozygous (wt/Ver) at the QTL. Solid circles indicate alleles from the parental chromosome with the Ver alleles. Clear circles indicate alleles from the parental chromosomes with the wt allele. Average number of vertebrae was compared between two groups of F₂ animals produced from the F₁ dam, categorized according to transmitted chromosomes (indicated by a or b). Correction for the QTL effect from the F₁ sire (Ver/wt) was also used. We assumed that the Ver allele increased the number of vertebrae by 0.57, an average effect in the experimental population in our previous study. (SE) Standard error.

Figure 2.

Analyses of genomic structure and genetic variation of the QTL region. (A) Gene map of a part of the human chromosome 9qter. (B) BAC contig for the QTL region. Vertical lines indicate positions of STS (broken) and microsatellite markers (solid). BAC clones (Suzuki et al. 2000) from which markers were developed and those in the minimum tiling path are shown. (C) Linkage map of microsatellite markers on SSC1qter. Arrowheads indicate positions of microsatellite markers, and clear arrowheads indicate those lacking genetic variation in European commercial breed pigs. (D) Defining the QTL region by the genetic variation of microsatellite markers. Sequencing analysis was performed for the region from SJ854 to SJ872. Exons of the genes found in this region are indicated, and arrows show the directions of gene transcription. We isolated 11 novel markers (underlined) and added them to the analysis of genetic variation. A reduction of genetic variation of microsatellite markers in European commercial breed pigs occurred between SJ641 and SJ820. The GenBank accession number for the genomic sequence in this region is AP009124. (E) Heterozygosities of markers in each breed. Breeds are: B, Berkshire; D, Duroc; H, Hampshire; L, Landrace; W, Large White; Y, Yorkshire; M, Meishan; J, Jinhua; and Wb, Japanese wild boar; and numbers of samples are shown in parentheses.

Figure 3.

(A) Comparison of amino acid sequences of Nr6a1 hinge domains of pig, human, and mouse. We identified an amino acid substitution (Pro192Leu; C → T at nucleotide 748 of AB248749) in pig NR6A1. Leucine occurred in alleles increasing vertebral number (Ver), whereas proline occurred in the wild-type pig allele (wt) as well as in human and mouse NR6A1/Nr6a1. (B) Effect of an amino acid substitution in pig Nr6a1 on its interaction with RAP80 and NCOR1. Interaction of pig NR6A1 and its corepressors RAP80 and NCOR1 was analyzed with a two-hybrid system in mammalian cells. CHO cells were transfected with pACT and pBIND plasmids containing the indicated DNA fragments. After 48 h, cells were collected and assayed for reporter activity. Relative luciferase activity was plotted and compared with the results of t-tests; error bars, 1 SD.

Figure 4.

Expression of Nr6a1 mRNA in embryonic day (ED)10.5 mouse embryos. Sense (A,B,D,F) and antisense (C,E,G) probe for Nr6a1 mRNA were hybridized to sections of ED10.5 mouse embryos. Faint signals for Nr6a1 mRNA were detected with antisense probe in the mandibular component of the first branchial arch (B), lung bud (D), and somites (F). No signals were detected in these tissues with sense probe (C,E,G). Lines with characters in the left image (A) indicate the positions of tissues in which signals were detected.

Figure 5.

Immunohistochemical study of Nr6a1 protein in embryonic day (ED)10.5 mouse embryos. Frozen sections of ED10.5 mouse embryos were fixed by ethanol and incubated either with rabbit anti-Nr6a1 antibodies (A,B,D) or with rabbit IgG isolated from preimmune sera (C,E). Signals were present in somites on both sides of the notochord. The vertical line in the sketch (F) indicates the location of the sections, and the photomicrographs at right (B,D) are magnified images of the boxed regions at left (A).