Comprehensive analysis of MHC class I genes from the U-, S-, and Z-lineages in Atlantic salmon

Abstract

Background: We have previously sequenced more than 500 kb of the duplicated MHC class I regions in Atlantic salmon. In the IA region we identified the loci for the MHC class I gene Sasa-UBA in addition to a soluble MHC class I molecule, Sasa-ULA. A pseudolocus for Sasa-UCA was identified in the nonclassical IB region. Both regions contained genes for antigen presentation, as wells as orthologues to other genes residing in the human MHC region.

Results: The genomic localisation of two MHC class I lineages (Z and S) has been resolved. 7 BACs were sequenced using a combination of standard Sanger and 454 sequencing. The new sequence data extended the IA region with 150 kb identifying the location of one Z-lineage locus, ZAA. The IB region was extended with 350 kb including three new Z-lineage loci, ZBA, ZCA and ZDA in addition to a UGA locus. An allelic version of the IB region contained a functional UDA locus in addition to the UCA pseudolocus. Additionally a BAC harbouring two MHC class I genes (UHA) was placed on linkage group 14, while a BAC containing the S-lineage locus SAA (previously known as UAA) was placed on LG10. Gene expression studies showed limited expression range for all class I genes with exception of UBA being dominantly expressed in gut, spleen and gills, and ZAA with high expression in blood.

Conclusion: Here we describe the genomic organization of MHC class I loci from the U-, Z-, and S-lineages in Atlantic salmon. Nine of the described class I genes are located in the extension of the duplicated IA and IB regions, while three class I genes are found on two separate linkage groups. The gene organization of the two regions indicates that the IB region is evolving at a different pace than the IA region. Expression profiling, polymorphic content, peptide binding properties and phylogenetic relationship show that Atlantic salmon has only one MHC class Ia gene (UBA), in addition to a multitude of nonclassical MHC class I genes from the U-, S- and Z-lineages.

Figure 1

Gene organization of Atlantic salmon MHC class I regions. LG15 is the Atlantic salmon MHC IA region with the previous sequenced BACs 30C23, 868O01, 92I04, 714P22 and 523M19 extended with 129P21. LG3-1 is the Atlantic salmon MHC IB region with the previous sequenced BACs 8I14, 424M17, 15L20 and 189M18 extended with 439H13 and 68O19, and LG3-2 is the allelic BAC 439J08. LG10 and LG14 containing the BACs 114L13 and LG14 respectively. Genbank accession numbers are indicated after the BAC clone name. Locus designation is based on sequence identity to matching ESTs and human nomenclature is used. The regions are drawn to scale. Color code: red is MHC class I genes, yellow is TAP genes, green is proteasome genes, blue is human extended MHC class II region genes, purple is human class I region genes, grey is non-human class I region genes, pseudogenes are striped.

Figure 2

Exon and intron boundaries for Atlantic salmon MHC class I genes from the U-, Z- and S- lineages. The exons are boxed with the sizes in bp above sequence and intron sizes below. Green exons indicate genes that do not contain typical transmembrane and cytoplasmic domains, yellow exon in ZCA indicate merged exons of α3 domain and TM and CYT.

Figure 3

Unrooted Phylogenetic tree of teleost MHC class I sequences. Phylogenetic tree analysis performed using PhyML 3.0 based on JTT model of evolution for full-length amino acid sequences. Consensus trees were based on 100 bootstrap replications and reported with the bootstrap support values (in percent) indicated at the respective nodes. Sequence references are as follows:Auha-UA [Genbank: AAD37813], Dare-UBA*01 [Genbank: CAA86732], Dare-UAA*01 [Genbank: CAA86731], Dare-UDA*01 [Genbank: AAF20178], Dare-UEA [Genbank: AAH53140], Dare-ZE*0201 [Genbank: CAD12790], Dare-L [Genbank: CAD56801], Furu-UBA [Genbank: AAC41236], Gaac-UAA*01 [Genbank: ABN14358], Gaac-UBA*01, Genbank: ABN14357], Icpu-SAA [Genbank: CK423282], Icpu-UAA [Genbank: AAD08650], Icpu-UBA [Genbank: AAD08648], Icpu-UCA [Genbank: AAD08647], Onmy-SAA [Genbank: AF091779], Onmy-UBA [Genbank: AF287483], Onmy-UCA [Genbank: BAD89552], Onmy-UDA [Genbank: AY523666], Onmy-UEA [Genbank: BAD89553], Onmy-UGA [Genbank: AAP04358 ], Onmy-LAA [Genbank: ABI21842 ], Onmy-LBA [Genbank: ABI21844], Orla-UAA*0101 [Genbank: BAD93265], Orla-UBA*0201 [Genbank: BAB83850], Orla-UCA*0101 [Genbank: BAB63957], Orla-UDA*0201 [Genbank: BAB83843], Orla-UEA*0201 [Genbank: BAB83837], Paol-UA1 [Genbank: BAD13367], Paol-ZE [Genbank: BAD13366], Pore-UA [Genbank: CAA90791], Sasa-LBA [Genbank: DY733800 and GO062643], Sasa-UBA*0301 [Genbank: AAN75116], Sasa-UBA*1001 [Genbank: AAN75118 ], [Genbank: ABQ59666], Sasa-ZAA*0101 [Genbank: DQ099914], Teni-UA [Genbank: CR724171], Teni-ZE [Genbank: CAF90807], HLA-A2 [Genbank: AAA76608]. Sasa-UDA, Sasa-UGA, Sasa-UHA1, Sasa-UHA2, Sasa-ULA, Sasa-SAA, Sasa-ZAA*0201, Sasa-ZBA, Sasa-ZCA are described in this paper.

Figure 4

Expression analysis of MHC class I genes in Atlantic salmon. Relative expression of Sasa-UBA, Sasa-UDA, Sasa-UGA, Sasa-UHA1, Sasa-UHA2, Sasa-ULA, Sasa-ZAA, Sasa-ZBA, Sasa-ZCA and Sasa-SAA in various tissues of Atlantic salmon using EF1A as reference gene. The Relative expression values for UBA in hindgut and spleen has been truncated and their values are indicated above their respective bars.

Figure 5

Alignment of Atlantic salmon MHC class I sequences. Comparison of Atlantic Salmon UBA*0301 [Genbank: AAN75116], UBA*1001 [Genbank: AAN75118], ULA [Genbank: ABQ59666], UGA, UDA, UHA1, UHA2, ZAA, ZBA, ZCA, SAA (described in this paper), LBA [Genbank: DY733800 and GO062643] and HLA-A2 [Genbank: AAA76608]. Dots indicate identities, dashes indicate gaps or missing sequence information. Cysteine residues involved in putative Ig fold are marked ♦ above sequences. Anchor residues known to bind the ends of peptide are marked # based on mammalian positions. Unique N-linked glycosylation sites are marked ! and CD8 binding site are boxed based on the acidic stretch from mammalian. b defines Z, S, L and UHA residues potentially involved in interactions with a different beta2-microglobulin. Individual domains/exon for LBA are based on the Sasa- UBA*0301 sequence.