J. Anim Sci. 2007. 85:577-582. doi:10.2527/jas.2006-217
© 2007 American Society of Animal Science
Haplotype analysis of the DQA genes in sheep: Evidence supporting recombination between the loci1
J. G. H. Hickford2,
H. Zhou and
Q. Fang
Gene-Marker Laboratory, Cell Biology Group, Agriculture and Life Sciences Division, PO Box 84, Lincoln University, Lincoln 7647, New Zealand
 |
Abstract
|
|---|
The ovine class II major histocompatibility complex mediates specific immune responses to exogenous antigens in sheep. A number of ovine class II loci have been identified, and most of them appear to be polymorphic. In this study we investigated the DQA1 locus of 520 sheep and the DQA2 locus of over 40,000 sheep, finding 12 sequences and 22 sequences, respectively, using DQA1- and DQA2-specific PCR primers. Among the DQA2 sequences, 2 groups of sequences can be found: those that share homology with the DQA2 sequences from closely related species and those that cluster with bovine DQA3 and DQA4 sequences and have been called DQA2-like in sheep. The occurrence of these DQA2-like sequences was once again confirmed to correspond with the absence of detectable DQA1 sequences, suggesting that they are found at the same location as DQA1. Within the sheep studied, 37 haplotypes could be detected, 23 being haplotypes of DQA1 and DQA2 sequences and with frequencies ranging from 0.38 to 9.27%, and 14 being haplotypes of DQA2 and DQA2-like sequences and with frequencies ranging from 0.03 to 14.53%. We discovered 12 DQA1-DQA2 combinations that were derived from 5 DQA1 alleles and 4 DQA2 alleles, and 8 DQA2-DQA2-like combinations from 5 DQA2 alleles and 2 DQA2-like sequences. The frequency of occurrence of recombined DQA1-DQA2 sequences and recombined DQA2-DQA2-like sequences is similar, once again suggesting the DQA2-like sequences are found at the DQA1 locus.
Key Words: DQA • haplotype • major histocompatibility complex • recombination • sheep
|
INTRODUCTION
|
|---|
The class II major histocompatibility complex (MHC) has been associated with specific immunity to exogenous antigens in vertebrates. However, the study of individual MHC genes and their role in disease resistance is complicated by the presence of multiple genes that are highly polymorphic and is hampered by our limited understanding of the organization of these genes.
Sheep have 1 DRA (Fabb et al., 1993
), 2 DRB (DRB1 and a pseudogene, DRB2; Dutia et al., 1994), and 1 or 2 DQA and DQB genes depending on the haplotype (Scott et al., 1987; van Oorschot et al., 1994; Hickford et al., 2000). The ovine DRA gene is not very polymorphic (Escayg et al., 1993). The polymorphic gene appears to be DRB1, with up to 106 sequences being reported (Konnai et al., 2003). Such a large number of alleles makes the typing of ovine DRB1 difficult. Although DQB is polymorphic in sheep (van Oorschot et al., 1994; Feichtlbauer-Huber et al., 2000), the absence of defining sequence characteristics in exon 2 makes locus assignment, based on sequence information, problematic (van Oorschot et al., 1994; Wright and Ballingall, 1994).
Variation in the ovine DQA genes is well characterized, with 14 DQA1, 18 DQA2, and 5 DQA2-like sequences being identified (Zhou and Hickford, 2001; Hickford et al., 2004; Zhou and Hickford, 2004). The ovine DQA1 and DQA2 genes are both expressed (Scott et al., 1991), and there are 2 distinct DQA haplotype configurations identified, one containing DQA1 and DQA2 genes and the other containing DQA2 and DQA2-like sequences. This retains the pattern of 2 DQA loci per haplotype (Hickford et al., 2000), although the organization of these DQA genes in the sheep genome is still poorly understood.
This paper describes the haplotypes of DQA1, DQA2, and DQA2-like sequences found in a large population of sheep and the identification of recombined ovine DQA haplotypes in the sheep population.
|
MATERIALS AND METHODS
|
|---|
The Lincoln University Animal Ethics Committee approved the collection of blood from animals used in this study.
Sheep and DNA Sources
Five hundred twenty sheep from 41 sires in New Zealand were MHC typed for the DQA1 and DQA2 genes, and an additional approximately 40,000 sheep, which spanned at least 16 historic breeds and a variety of composite breeds, were typed at the DQA2 gene only. Blood samples were collected on FTA Classic cards (Whatman BioScience, Middlesex, UK), and genomic DNA for PCR amplification was purified using a 2-step procedure (Zhou et al., 2006). Briefly, a 1.2-mm diam. blood disc was incubated with 200 µL of 20 mM NaOH solution for 30 min, and after the removal of the NaOH solution, the disc was washed in 200 µL of TE–1 buffer (10 mM Tris-HCl, 0.1 mM EDTA, pH 8.0).
Ovine DQA1 and DQA2 Typing
The ovine DQA1 and DQA2 genes were typed separately using PCR-single-strand conformational polymorphism (PCR-SSCP) technique, as described in Zhou and Hickford (2004) and Hickford et al. (2004). Briefly, the polymorphic exon 2 of the DQA1 or DQA2 gene was amplified using DQA1- or DQA2-specific primers (Table 1). The DQA2-specific primers amplify DQA2 and DQA2-like sequences. Amplification was carried out in a 20-µL reaction containing genomic DNA on a 1.2-mm diam. disc of FTA, 0.25 µM of each primer, 150 µM of dNTP (Eppendorf, Hamburg, Germany), 0.5 U of Taq DNA polymerase (Qiagen, Hilden, Germany), and 1x the reaction buffer supplied (containing 1.5 mM MgCl2). After initial denaturation at 94°C for 2 min, 35 cycles of 94°C for 30 s, 61°C (for DQA1) or 62°C (for DQA2) for 30 s, and 72°C for 30 s were utilized, followed by a final 5-min extension step at 72°C. Amplicons were subjected to SSCP analysis using 16% (for DQA1) or 14% (for DQA2) acrylamide:bisacrylamide (37.5:1; Bio-Rad, Hercules, CA) gels. Amplicons representative of the known DQA1 or DQA2 sequences were also included in each polyacrylamide gel, and their banding patterns were used as standards for determining the alleles present in individual sheep.
Haplotype Determination
The PCR-SSCP method does not determine specific haplotypes but instead reveals what alleles of DQA1, DQA2, or DQA2-like sequences a sheep carries. To determine DQA haplotypes, sires, dams, and some siblings, if available, were genotyped, and based on the coinheritance of sequences, haplotypes were indirectly determined. The presence of a DQA1-null allele was ascertained by its association with a DQA2 allele and DQA2-like sequence, as described in Hickford et al. (2000).
|
RESULTS
|
|---|
Recombined Haplotypes of the DQA1 and DQA2 Loci
Twenty-three DQA1-DQA2 haplotypes composed of 12 sequences and 9 sequences at the DQA1 and DQA2 loci, respectively, were identified in the sheep investigated. The frequency of these haplotypes ranged from 0.38 to 9.27%, totaling 64.51% in sheep studied (Table 2). Of these haplotypes, 12 contained combinations of sequences between 5 DQA1 alleles (*0101, *0901, *0103, *0501, and *0601) and 4 DQA2 alleles (*0301, *0602, *0601, and *1101). A summary of these haplotypes is shown in Table 2, and the linkage of these recombined DQA1 and DQA2 sequences is illustrated in Figure 1.
Recombined Haplotypes of DQA2 Alleles and DQA2-like Sequences
Fourteen DQA2-DQA2-like haplotypes were revealed, containing 6 clearly identified DQA2 alleles and 5 clearly identified DQA2-like sequences. These had an overall frequency of 34.87%. All of these haplotypes were present in all breeds, with the exception of haplotypes *0701–*1601 and *0702–*1601 (see Table 3). The haplotype *08012–*0201 suggests the presence of additional DQA2 and DQA2-like sequences, although these cannot be definitively assigned to a locus based on the sequence data available. Of these, 10 haplotypes were found to contain combinations of 5 DQA2 alleles (*0101, *0102, *0401, *0701, and *0702) and 2 DQA2-like sequences (*1401 and *1601; Table 3, Figure 2). Recombined haplotypes were not detected between the other DQA2 alleles and DQA2-like sequences.
With the 10 haplotypes containing combinations of 5 DQA2 alleles and 2 DQA2-like sequences, 2 of them (*0101–*1401 and *0102–*1601) were much more common in the sheep studied with a frequency of 14.53 and 5.48%, respectively (Table 3
).
|
DISCUSSION
|
|---|
The identification of recombined haplotypes of the DQA1 and DQA2 loci, or the DQA2 alleles and DQA2-like sequences in sheep, suggests that recombination has occurred within the ovine DQA region. This is the first report of this in sheep, but it is not without precedence. For example, recombination between the MHC class II genes has been reported in humans (Carrington, 1999), mice (Yauk et al., 2003), and cattle (Andersson et al., 1988). In humans, 2 recombination hotspots have been located between DQB3 and DQB1/DQA1 and between DPB1 and RING3 (Cullen et al., 2002), and recombination between the DQA1 and DQB1 loci, separated by 11.13 kb, has been recently reported in a human family (Tiercy and Villard, 2004). The ovine DQA1 and DQA2 loci are separated by 33 kb (Wright and Ballingall, 1994), whereas the physical distance between the DQA2 locus and the locus at which the DQA2-like sequences are found is unknown. However, recombination between sheep DQA loci is supported by reports of recombination in the human homologues (Carrington, 1999; Tiercy and Villard, 2004).
Equally, whereas haplotyping was based on sequence analysis of the exon 2 region, the possibility of exon-shuffling, which has been observed with MHC (Doxiadis et al., 2006) and non-MHC genes (Long et al., 1996), cannot be ruled out. Recombination within the DQA genes is also a possibility. The presence of a recombination hotspot in the second intron of the mouse MHC Eß gene (the human DR homologue; Yauk et al., 2003), also raises the alternative possibility of intra-gene recombination, but further investigation of the other exons of the DQA1 and DQA2 genes, and exons associated with the DQA2-like sequences, is needed to confirm this. This would necessitate the development of a sensitive typing system for those gene regions.
The failure to detect recombination in the ovine DQ region in a previous study (Escayg et al., 1996) may be for 2 reasons: the use of RFLP-Southern hybridization methods and the analysis of fewer sheep. The RFLP-Southern hybridization is a low-resolution typing technique, and alleles or sequences that differ outside of restriction enzyme recognition sites cannot be easily resolved and hence will be mistyped as a single allele or sequence. In contrast, the application of PCR-SSCP, a technique with potential to detect single nucleotide variation in small amplimers (Zhou et al., 2005), provides a more precise typing mechanism for the ovine DQA genes. The larger numbers of DQA1 and DQA2 alleles detected by PCR-SSCP (Zhou and Hickford, 2004; Hickford et al., 2004), compared with RFLP-Southern hybridization analysis (Escayg et al., 1996), support this notion. Equally, the previous study RFLP typed only 76 sheep at DQA1 and 50 sheep at DQA2 (Escayg et al., 1996), and hence the recombinations seen here at relatively low frequencies may not have been found in the population.
This study provided evidence of historical recombination events but no conclusion on when these have occurred could be drawn. This may be for several reasons: 1) Only 1 or 2 generations were investigated. 2) Many of sheep did not have pedigree records. In many cases, only 1 parent (sire) was available. 3) The DQA1 and DQA2 loci are very close together, and the frequency of recombination is therefore probably very low. If there were no interference, then the physical distance of 33 kb between the known DQA1 and DQA2 loci would correspond to 1 recombination per 3,000 chromosomes (assuming 1 Mb = 1% recombination frequency). Hence the number of new recombination events expected in the 520 sheep of known parentage that were analyzed (1,040 chromosomes) would be approximately 0.35.
Because only a relatively small number of sheep were investigated at the DQA1 and DQA2 loci in this study, the DQA1-DQA2 haplotype frequencies cited may not precisely represent the New Zealand sheep population as a whole, and it is accordingly very difficult to identify the original or unrecombined haplotypes based on the frequency information. However, if it is presumed that 11 of the 23 total DQA1-DQA2 haplotypes are original or unrecombined and that of the 12 potentially recombined haplotypes, 5 are probably unrecombined, then there would be 7 recombined DQA1-DQA2 haplotypes. This would mean (12 – 5)/23 = 30% of DQA1-DQA2 haplotypes are recombined haplotypes.
The absence of 2 DQA2-DQA2-like haplotypes (*0701–*1601 and *0702–*1601) in Corriedale and Romney breeds may be due to the small numbers of sheep investigated and reflect the low frequencies of these 2 haplotypes in the sheep population. Given that Corriedale sheep are originally one-quarter Merino (which have these haplotypes), this argument would appear to be correct.
Of the 10 DQA2-DQA2-like haplotypes that containing combinations of 5 DQA2 alleles and 2 DQA2-like sequences, 2 haplotypes (*0101–*1401 and *0102–*1601, for DQA2 and DQA2-like, respectively) made up an approximate 57% of all DQA2-DQA2-like haplotypes found. They therefore probably represent the original or unrecombined haplotypes at these loci. Although it is difficult to identify other potential unrecombined DQA2-DQA2-like haplotypes, it is possible that there are another 3 original or unrecombined haplotypes. This gives approximately (10 – 5)/14 = 36% of DQA2-DQA2-like haplotypes being recombined haplotypes. This figure is similar to that observed for DQA1-DQA2 haplotypes, suggesting that DQA2-like sequences occur at the DQA1 locus, or at a similar distance in the opposite direction from the DQA2 gene. However, a much lower number of sheep were typed for DQA1 and DQA2 than were typed for DQA2 and DQA2-like, and as more sheep are typed, more recombined haplotypes of the DQA1 and DQA2 loci may be found.
Recombination within the ovine DQA region appears to have only occurred for some DQA1 and DQA2 alleles because only selected recombined haplotypes were identified. This may be due to a lower number of sheep being both DQA1 and DQA2 typed, but this phenomenon of a discrete number of recombined haplotypes was also observed for the DQA2 alleles and DQA2-like sequences, where genotyping of more than 40,000 sheep revealed recombinant haplotypes for some, but not all DQA2 alleles and DQA2-like sequences. This suggests that recombination within the ovine DQA region may occur in a haplotype-dependent manner. Haplotype-specific recombination has previously been reported for the mouse MHC (Shiroishi et al., 1995), and it is also suggested to occur in humans (Thomsen et al., 1994), but it has been difficult to address this issue given the constraints in studying large human populations (Carrington, 1999).
One possible explanation for haplotype-dependent recombination is structural constraints such as variable haplotype lengths and different gene organizations, which will inhibit the proper alignment of some homologous chromosomes, which would be necessary for recombination (Carrington, 1999). Another explanation is that certain haplotypes of DQA are selected because of their complementary benefit to the host. Pressure to retain 2 loci of DQA per haplotype may reflect the necessity to preserve certain combinations of alleles at these loci, which together result in successful elimination of common pathogens (Carrington, 1999).
It is difficult to determine how long ago the recombination events described are likely to have arisen, and whether they are restricted to any breed, as in New Zealand with possibly the exception of some parts of the Merino industry, the majority of sheep are cross-bred, and hence breed purity cannot be determined. Accordingly the origin of recombinant haplotypes cannot be defined. Breed histories are unreliable because there is no absolute guarantee of breed purity.
|
IMPLICATIONS
|
|---|
The high number of combinations within the ovine DQA region, especially between DQA2 and DQA2-like sequences, underlines the importance of interlocus recombination in shaping class II major histocompatibility complex haplotypic diversity. This diversity will need to be taken into account when investigating disease association with individual class II major histocompatibility complex genes.
|
Footnotes
|
|---|
1 We thank Y.-S. Lin for the technical assistance. 
2 Corresponding author: hickford{at}lincoln.ac.nz
Received for publication April 6, 2006.
Accepted for publication November 3, 2006.
|
LITERATURE CITED
|
|---|
Andersson, L., A. Lunden, S. Sigurdardottir, C. J. Davies, and L. Rask. 1988. Linkage relationships in the bovine MHC region. High recombination frequency between class II subregions. Immunogenet. 27:273–280.[CrossRef][Medline]
Carrington, M. 1999. Recombination within the human MHC. Immunol. Rev. 167:245–256.[CrossRef][Medline]
Cullen, M., S. P. Perfetto, W. Klitz, G. Nelson, and M. Carrington. 2002. High-resolution patterns of meiotic recombination across the human major histocompatibility complex. Am. J. Hum. Genet. 71:759–776.[CrossRef][Medline]
Doxiadis, G. G., M. K. van der Wiel, H. P. Brok, N. G. de Groot, N. Otting, B. A. ’t Hart, J. J. van Rood, and R. E. Bontrop. 2006. Reactivation by exon shuffling of a conserved HLA-DR3-like pseudogene segment in a New World primate species. Proc. Natl. Acad. Sci. USA 103:5864–5868.[Abstract/Free Full Text]
Dutia, B. M., I. McConnell, K. T. Ballingall, P. Keating, and J. Hopkins. 1994. Evidence for the expression of two distinct MHC class II DR beta like molecules in the sheep. Anim. Genet. 25:235–241.[Medline]
Escayg, A. P., J. G. Hickford, G. W. Montgomery, K. G. Dodds, and D. W. Bullock. 1996. Polymorphism at the ovine major histocompatibility complex class II loci. Anim. Genet. 27:305–312.[Medline]
Escayg, A. P., G. W. Montgomery, J. G. Hickford, and D. W. Bullock. 1993. A BglII RFLP at the ovine MHC class II DRA locus. Anim. Genet. 24:217.[Medline]
Fabb, S. A., J. F. Maddox, K. J. Gogolin-Ewens, L. Baker, M. J. Wu, and M. R. Brandon. 1993. Isolation, characterization and evolution of ovine major histocompatibility complex class II DRA and DQA genes. Anim. Genet. 24:249–255.[Medline]
Feichtlbauer-Huber, P., M. J. Stear, R. Fries, and J. Buitkamp. 2000. Reference-strand-mediated conformation analysis of MHC alleles: A new method for high-resolution typing of the Ovar-DQB genes. Immunogenetics 51:65–68.[CrossRef][Medline]
Hickford, J. G. H., H. J. Ridgway, and A. P. Escayg. 2000. Evolution of the ovine MHC DQA region. Anim. Genet. 31:200–205.[CrossRef][Medline]
Hickford, J. G., H. Zhou, S. Slow, and Q. Fang. 2004. Diversity of the ovine DQA2 gene. J. Anim. Sci. 82:1553–1563.[Abstract/Free Full Text]
Konnai, S., Y. Nagaoka, S. Takeshima, M. Onuma, and Y. Aida. 2003. Sequences and diversity of 17 new Ovar-DRB1 alleles from three breeds of sheep. Eur. J. Immunogenet. 30:275–282.[CrossRef][Medline]
Long, M., S. J. de Souza, C. Rosenberg, and W. Gilbert. 1996. Exon shuffling and the origin of the mitochondrial targeting function in plant cytochrome c1 precursor. Proc. Natl. Acad. Sci. USA 93:7727–7731.[Abstract/Free Full Text]
Scott, P. C., C. L. Choi, and M. R. Brandon. 1987. Genetic organization of the ovine MHC class II region. Immunogenet. 25:116–122.[CrossRef][Medline]
Scott, P. C., K. J. Gogolin-Ewens, T. E. Adams, and M. R. Brandon. 1991. Necleotide sequence, polymorphism, and evolution of ovine MHC class II DQA genes. Immunogenet. 34:69–79.[Medline]
Shiroishi, T., T. Koide, M. Yoshino, T. Sagai, and K. Moriwaki. 1995. Hotspots of homologous recombination in mouse meiosis. Adv. Biophys. 31:119–132.[CrossRef][Medline]
Thomsen, M., M. Neugebauer, J. Arnaud, N. Borot, A. Sevin, M. Baur, and A. Cambon-Thomsen. 1994. Recombination fractions in the HLA system based on the data set ‘provinces Francaises’: Indications of haplotype-specific recombination rates. Eur. J. Immunogenet. 21:33–43.[Medline]
Tiercy, J. M., and J. Villard. 2004. A new DR7-DQ8 haplotype resulting from a recombination between the DQA1 and DQB1 loci in a leukemic patient of Caucasoid origin. Immunogenetics 56:301–303.[Medline]
van Oorschot, R. A., J. F. Maddox, L. J. Adams, and S. A. Fabb. 1994. Characterization and evolution of ovine MHC class II DQB sequence polymorphism. Anim. Genet. 25:417–424.[Medline]
Wright, H., and K. T. Ballingall. 1994. Mapping and characterization of the DQ subregion of the ovine MHC. Anim. Genet. 25:243–249.[Medline]
Yauk, C. L., P. R. Bois, and A. J. Jeffreys. 2003. High-resolution sperm typing of meiotic recombination in the mouse MHC Ebeta gene. EMBO J. 22:1389–1397.[CrossRef][Medline]
Zhou, H., and J. G. Hickford. 2001. Rapid communication: Three new allelic sequences at the ovine MHC class II DQA1 locus. J. Anim. Sci. 79:779–780.[Free Full Text]
Zhou, H., and J. G. Hickford. 2004. Allelic polymorphism in the ovine DQA1 gene. J. Anim. Sci. 82:8–16.[Abstract/Free Full Text]
Zhou, H., J. G. Hickford, and Q. Fang. 2005. Technical note: Determination of alleles of the ovine PRNP gene using PCR-single-strand conformational polymorphism analysis. J. Anim. Sci. 83:745–749.[Abstract/Free Full Text]
Zhou, H., J. G. Hickford, and Q. Fang. 2006. A two-step procedure for extracting genomic DNA from dried blood spots on filter paper for PCR amplification. Anal. Biochem. 354:159–161.[CrossRef][Medline]
Top
Abstract
INTRODUCTION
