The Journal of Immunology
Ancestral Organization of the MHC Revealed in the Amphibian Xenopus1 Yuko Ohta,2* Wilfried Goetz,* M. Zulfiquer Hossain,* Masaru Nonaka,† and Martin F. Flajnik* With the advent of the Xenopus tropicalis genome project, we analyzed scaffolds containing MHC genes. On eight scaffolds encompassing 3.65 Mbp, 122 MHC genes were found of which 110 genes were annotated. Expressed sequence tag database screening showed that most of these genes are expressed. In the extended class II and class III regions the genomic organization, excluding several block inversions, is remarkably similar to that of the human MHC. Genes in the human extended class I region are also well conserved in Xenopus, excluding the class I genes themselves. As expected from previous work on the Xenopus MHC, the single classical class I gene is tightly linked to immunoproteasome and transporter genes, defining the true class I region, present in all nonmammalian jawed vertebrates studied to date. Surprisingly, the immunoproteasome gene PSMB10 is found in the class III region rather than in the class I region, likely reflecting the ancestral condition. Xenopus DM␣, DM, and C2 genes were identified, which are not present or not clearly identifiable in the genomes of any teleosts. Of great interest are novel V-type Ig superfamily (Igsf) genes in the class III region, some of which have inhibitory motifs (ITIM) in their cytoplasmic domains. Our analysis indicates that the vertebrate MHC experienced a vigorous rearrangement in the bony fish and bird lineages, and a translocation and expansion of the class I genes in the mammalian lineage. Thus, the amphibian MHC is the most evolutionary conserved MHC so far analyzed. The Journal of Immunology, 2006, 176: 3674 –3685.
T
he MHC is the most gene-dense region in the human genome and plays an indispensable role in the adaptive immune system (1). Class I and class II Ag-presenting molecules present small peptides derived from pathogens to CD8⫹ and CD4⫹ T cells, respectively. In the class I system, endogenous peptides derived from intracellular pathogens are enzymatically cleaved into small peptides by the immunoproteasome containing the specialized -subunits PSMB8, PSMB9, and PSMB10, which upon infection replace the constitutive subunits, PSMB5, PSMB6, and PSMB7, respectively (2). Short peptides of 8 –11 aas are transported into endoplasmic reticulum by the TAP (TAP1 and TAP2) and then loaded onto class I molecules associated with tapasin (TAPBP). The resulting class I-peptide complexes move to the cell surface, where they are recognized by Ag-specific TCRs expressed by CD8⫹ T cells (3). Interestingly, in most mammals, the genes responsible for class I Ag processing are embedded in the class II region (e.g. PSMB8, PSMB9, TAP1, and TAP2) or in the extended class II region (e.g., TAPBP, class I transcription regulator, RXRB), whereas class I genes themselves are found in another region (4). In contrast, studies of nonmammalian vertebrates have shown that class I genes are tightly linked to class I-processing genes, sug-
*University of Maryland, Department of Microbiology and Immunology, 655 West Baltimore Street, BRB13-009, Baltimore, MD 21201; and †Department of Biological Sciences, Graduate School of Science, University ofTokyo, Hongo, Bunkyo-ku, Tokyo, Japan Received for publication November 1, 2005. Accepted for publication January 9, 2005. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. 1 This work was supported by National Institutes of Health Grant AI27877 (to Y.O., W.G., and M.F.F.) and Grant 15207019 from The Ministry of Education, Culture, Sports, Science, and Technology (to M.N.). 2 Address correspondence and reprint requests to Dr. Yuko Ohta, University of Maryland, Department of Microbiology and Immunology, 655 West Baltimore Street, BRB13-009, Baltimore, MD 21201. E-mail address:
[email protected]
Copyright © 2006 by The American Association of Immunologists, Inc.
gesting that this class I region is the primordial organization (5–7). In some nonmammalian species, there is only a single or few classical class I genes, perhaps due to a selection for coevolution with the Ag-processing genes. Thus, plasticity of class I genes in mammalian species is an evolutionarily derived characteristic (5, 7). Xenopus (especially Xenopus laevis and more recently Xenopus tropicalis) has been used historically for developmental studies (8). Regarding the MHC, this animal is the most comprehensively studied amphibian for characteristics of the adaptive immune system. Xenopus is a unique model because there are several polyploid species (2n–12n) within the genus that arose by recent genome-wide duplication (from 2 to 30 million years ago) (9). Because of its important phylogenetic position, and because it is a true diploid (genome size approximately half that of human), X. tropicalis has been selected as a model organism for a whole genome sequencing project (具www.jgi.doe.gov/xenopus典). BAC libraries have been constructed and available to the public for analysis and genetic manipulation. In addition, different sources of expressed sequences have been deposited into the expressed sequence tag (EST)3 databases for X. tropicalis and X. laevis, which facilitates gene annotation. In our previous studies of the Xenopus MHC in which we tediously cloned the genes orthologous to those of humans one by one, it was shown that synteny seemed to be stable between the two species separated by 350 million years (6, 10). This is in contrast to some other nonmammalian vertebrates in which the MHC genes are scattered over the genome, especially for class II and class III region genes (5, 7, 11–20). In this study, we took advantage of the genome project and the various EST databases and mined them for MHC genes. Our results reveal that the entire architecture of the Xenopus MHC is remarkably conserved when compared with human, and further show that the teleost and, to a 3 Abbreviations used in this paper: EST, expressed sequence tag; Igsf, Ig superfamily; BLAST, Basic local alignment search tool; ORF, open reading frame; TM, transmembrane; XMIV, Xenopus MHC-linked Ig superfamily.
0022-1767/06/$02.00
The Journal of Immunology lesser extent, bird MHCs are highly derived. In addition, analysis of the Xenopus MHC has revealed that some major immune genes seem to have emerged at the level of amphibians and has uncovered some new Ig superfamily (Igsf) genes that are activating or inhibitory receptor candidates, similar to those first discovered on NK cells (21–23).
Materials and Methods cDNA sequence database searches for MHC genes We obtained accession numbers for genes listed in the human MHC, excluding pseudogenes, from the Wellcome Trust Sanger Institute web site (具www.sanger.ac.uk典). Basic local alignment search tool (BLAST)p and tBLASTn were performed on the National Center for Bioinformatics Institute (NCBI) web site (具www.ncbi.nlm.nih.gov典) with either full-length amino acid sequences or domain-by-domain in the X. laevis, X. tropicalis, and/or EST_Others databases using the BLOSUM 45 matrix. Genes with E-values of ⬍0.05 were further confirmed by BLASTp or BLASTx searches in the vertebrate databases using the BLOSUM 45 matrix. When no positive result was obtained, we further searched Xenopus EST databases in the Wellcome Trust Sanger Institute using the BLOSUM 50 matrix.
Data-mining the X. tropicalis genome project We began this study with BLASTn searches of X. tropicalis version 3.0 (estimated genomic coverage of 7.4⫻) at the Department of Energy Joint Genome Institute (JGI; 具www.jgi.doe.gov/xenopus典) with MHC genes that were isolated over the past 10 years (class I (24), class II (25, 26), TAP1 (10), TAP2 (27), PSMB8 (28), PSMB9 (29), Ring3 (30), C4 (31), Factor B (32), HSP70 (33), and RXRB (34, 35)). In most cases, X. laevis genes were used for the searches because most genes were cloned from this species, and we were fortunate that usually there is enough sequence similarity in coding regions between X. laevis and X. tropicalis to permit isolation of the orthologues across species. Most scaffolds were large enough to contain multiple genes, and thus we used various MHC candidate genes found in the EST databases to screen other scaffolds containing the X. tropicalis orthologues of the human MHC genes. Individual scaffolds were then retrieved from the JGI browser window, and all “fgenesh” entries and EST hits were examined manually. To confirm the gene annotation, we searched all predicted genes by BLASTx in the NCBI vertebrate database, using the BLOSUM 45 matrix. In cases when we did not find Xenopus genes in the EST databases, we searched EST databases using reconstructed nucleotide sequences from the scaffolds. We tried to follow the nomenclature used in the map to the HUGO Gene Nomenclature Committee (36) and its database (37).
X. laevis cDNA library screening We isolated two genes that have important roles in the mammalian immune system. Probes were made from an EST entry for the partial DM gene (BX845472) by PCR at nucleotide positions 63–331, from a X. laevis cDNA library made from mixture of spleen and intestine mRNA. The C2 probe was made by PCR using primers taken from EST entry (BX853282) corresponding to nucleotide positions 42– 462, from a X. laevis cDNA library made from mixture of liver, spleen, and thymus mRNA (10). The PCR amplicons were cloned into the TA cloning vector (Invitrogen Life Technologies) and sequenced. Both library screenings and washings were conducted under high stringency conditions (38). Positive clones were isolated and sequenced in their entirety. The sequences are deposited to GenBank, and accession numbers are given as DQ268506 for X. laevis DM and DQ268507 for X. laevis C2.
Phylogenetic trees The deduced DM␣ (EST clone, AAH61681) and DM amino acid sequences were aligned using Clustal X, and Neighbor-Joining bootstrapping trees (1000 trial runs) were made and viewed in the TreeView 1.6.6 program (39). The deduced X. laevis and X. tropicalis (reconstructed from scaffold) C2 amino acid sequences were also aligned with factor B and C2 of tetrapod species, bony and cartilaginous fish Bf/C2, whose assignment to Bf or C2 is not clear, and lamprey and invertebrate Bf/C2 are considered to represent the preduplication Bf/C2 state (40, 41). For both trees gaps were included, and multiple substitutions were not taken into account.
3675 with HindIII or SacI, and fragmented DNA was separated on an agarose gel and blotted onto membranes. The DNA amount was increased proportionally to the ploidy level. The gene-specific Ig-domain probe (EST entry CN328971; nt 300 –587) was made by using PCR from cDNA library made from X. laevis spleen and intestine, and the sequence was confirmed. Primers used for amplification were as follows: 5⬘-AAA GTG GAA CAG CCT GAG CG-3⬘ and 5⬘-CAT CAC ATG CAC AAT GGT TCC-3⬘. Hybridization was performed under low stringency conditions (30% formamide; 6 ⫻ SSC) at 42°C for overnight, and washed in 2 ⫻ SSC, 1% SDS at room temperature, followed by 2 ⫻ SSC, 0.1% SDS at 55°C (38). The same blot was later washed under high stringency conditions (0.2 ⫻ SSC, 0.1% SDS at 65°C) to eliminate low-homology signals.
Results Database mining The chicken DM␣1 and 1-encoding exons (obtained from AL023516) were used to search databases for the Xenopus DM genes; the deduced amino acid sequences of these regions of the bird sequences were found to be more specific for DM compared with their ␣2 or 2 Igsf domains, which more readily selected classical class II sequences in BLAST searches. We and others (S. Beck, personal communication) have done exhaustive searches in the EST and genomic databases for teleost DM genes and could not identify them, suggesting either that teleosts have lost the DM genes or they arose in the tetrapod lineage after its divergence from bony fish. The Xenopus sequences were used in a phylogenetic analysis, and the trees solidify the hypothesis that the DM class II genes are as old as classical class II␣ and class II (Ref. 42 and Fig. 1). Thus, we think it is more likely that these genes have been lost in teleosts and will be found in the cartilaginous fish. From the initial EST searches with human MHC genes (fulllength amino acid sequences), we found most of the Xenopus housekeeping genes with significant E-values. Each gene was further confirmed by BLASTx for their orthology. Using these genes found in the EST databases, we then BLAST-searched the X. tropicalis scaffolds version 3.0 (典www.jgi.doe.gov/xenopus具). All scaffolds so-identified were then inspected for open reading frames (ORF), which were manually verified and then used to rescreen the GenBank database (see percentage of identities in Table I). During this process, we identified other genes on the scaffolds that were then used to screen the EST databases. A total of 122 ORFs were found on the eight genomic scaffolds (Tables I and II) encompassing 3.65 Mbp of which 110 genes were annotated that showed significant similarity to genes in the databases. Twelve genes had no database match (denoted as ORF). At least one gene on each X. tropicalis scaffold shown in Fig. 2 has been rigorously analyzed for MHC linkage previously (24 –33) (Y. Ohta and M. F. Flajnik, unpublished data for FABGL, DAXX, and FLOT1), and thus we are certain that all of these scaffolds are in the Xenopus MHC. We could not decide by phylogenetic analyses whether DDAH and NOTCH were orthologues of the human MHC-encoded DDAH2 and NOTCH4 or their paralogues found on human chromosomes 1, 9, and 19 (5). Their location within the MHC makes it likely that they are the orthologues of the MHC-encoded human genes. Conversely, several genes were found in the Xenopus MHC that are present on different chromosomes in the human, sometimes in paralogous regions (Table I and red numbered loci in Fig. 2). This is likely due to differential silencing of genes after divergence from the common ancestor. These subjects are further described and discussed in more detail below.
Southern blotting
Extended class II region
Genomic DNA from different Xenopus species (2n–12n), or from siblings in a family ( f/g ⫻ f/r) with known MHC haplotypes (27), was digested
All 15 functional genes in the human extended class II region and 3 of 4 genes flanking this region were found in ⬃415- and 200-kb
3676
SYNTENIC INTEGRITY OF MHC THROUGH EVOLUTION regions of two X. tropicalis scaffolds, respectively; the gene density is ⬃24 kb/gene and ⬃40 kb/gene, respectively. The genes between RXRB and PHF1 are inverted but in the same order compared with the human MHC (4) (Fig. 2), implying an en bloc inversion. Because all genes in this inverted region are found on a single scaffold, it is unlikely to be an assembly artifact. By BLAST searching the end of scaffold 917 for contiguous scaffolds, we were able to connect scaffolds 917 and 726, covering over 1 Mbp genomic region linking the extended class II region to the class II␣ gene. In a later version of the genome assembly (version 4.0 and 4.1), these scaffolds are indeed connected (scaffold 396; see Table III). In summary, this region is remarkably well conserved between Xenopus and human. Class II region and the specialized nonmammalian class I region
FIGURE 1. Phylogenetic analysis of vertebrate MHC molecules demonstrates an ancient origin of DM. The extracellular domains for the fulllength EST for DM␣ (GenBank accession no. AAH61681; A) and our full-length clone for DM (ABB85336; B) were used in the construction of the trees. Genetic distance is shown as a bar on the bottom. Accession nos. for each sequences are as follows: human DM␣ (AAH11447), mouse DM␣ (NP_034516), rat DM␣ (CAA89831), cow DM␣ (BAA11171), chicken DM␣ (CAA18966), quail DM␣ (BAC82512), Xenopus DM␣ (AAH61681), human DP␣ (NP_291032), human DQ␣ (NP_002113), human DR␣ (NP_061984), mouse IA␣ (AAB81529), rat II␣ (AAB35454), Xenopus DBAf2 (AAH57744), Xenopus DAAf1 (AAL58430), Caiman II␣ (AAF99282), catfish II␣ (AAD39871), zebrafish II␣ (NP_001007049), nurse shark DAA (AAA49310), human DM (NP_002109), mouse DM2 (A55242), rat DM (CAA89832), cow DM (BAA11172), rabbit DM (AAB53264), chicken DM1 (CAA18968), chicken DM2 (CAA18967), quail DM1 (BAC82514), quail DM2 (BAC82513), human DO (AAA59717), mouse A2 (AAA51637), chicken II (AAS00716), quail II (BAC82510), Xenopus II (BAA02845), medaka II (BAA94279), catfish II (AAB67871), carp II (CAA64709), trout II (AAD53026), shark II (L20274), human A1 (NP_002107), mouse H2-K (AAA80451), rat RT1␣ (XP_579224), chicken B-F (CAA18972), Xenopus I␣ (AAA16064), nurse shark UAA (AAC60347).
We previously mapped class II␣, class II, Ring3 (BRD2), proteasome PSMB8 and PSMB9, and transporter TAP1 (ABCB2) and TAP2 (ABCB3) genes to the MHC by segregation analyses in X. laevis families (24 –33). We now report the order of these genes in the class II region and the primordial class I region (Fig. 2). In addition, the nonclassical class II molecules, DM␣ and DM, were mapped into the class II region. The class II region (five genes, from classical class II␣ to DM) encompasses ⬃217 kb on scaffold 1109, whereas the class I region (five genes, from class Ia to TAP2) is ⬃274 kb on scaffold 1316. From Southern blotting analysis, two class II␣ and class II genes were found in X. tropicalis (L. Du Pasquier, personal communication). Class II␣ genes are split onto two scaffolds (exons 1 and 2 on 917 and 3 and 4 on 1109); however, it is likely that the presence of the two tandemly duplicated highly homologous genes obstructed a correct sequence assembly. So far, only one class II gene was found on the scaffolds. However, the distance between class II and class II␣ on the scaffold is ⬃244 kb, seemingly too large compared with intergenic distances in other MHC regions. There are many repetitive elements and fragments of retrotransposons in this area, including 2 contigs that match perfectly to Magnetococcus sp. MC-1 sequences (AAAN03000014). Thus, this region seems to have been contaminated with sequences from other species (even in the version 4.0 scaffold), and thus we must wait to clarify the sequence and distance between the class II loci. PSMB9 is also split between two scaffolds (1109 and 1316); however, because there is only a single locus from Southern blotting analysis (29), these scaffolds are within an intron length of each other. A BTNL-II gene (butyrophilin-like MHC class II-associated), located at the border of the mammalian class II and class III regions (43), was found neither in the EST databases nor the genomic scaffolds. However, other BTN genes are found in the human class I region. The BTN genes are Igsf members (44) that display notable sequence similarity to other MHC genes, particularly to human MOG and bird B-G (13) (see below). Class III region Forty-five human genes listed in the human class III region were found on five scaffolds spanning ⬃2 Mbp (Fig. 2), suggesting that like the extended class II region, the class III region is old and extremely well conserved. Because the NOTCH gene is in scaffold 1316, where the majority of genes are in the class I region, the class III region is contiguous with the class I region. Like the extended class II region, there seems to have been at least three en bloc inversions between C4 and PPT2, STK19 and C6orf29, and HSP70 and CSNK2B. There are also potential translocations (e.g., ATP6V1G2, BAT1, and NEU1). PCR was used to identify the gap between scaffolds 895 and 1207. A 0.8-kb fragment was sequenced
The Journal of Immunology
3677
Table I. List of genes, scaffold numbers, and database accession no. found in the databases
Gene
Scaffold
Region flanking the extended class II subregion ZBTB9 726 C6orf82 726 PHF1 726 Extended class II subregion KIFC 726 DAXX 726 ZNF297 726 TAPBP 726 RGL2 726 HKE2 726 C6org11 726 B3GALT4 726 RPS18 726 VPS52 726 RING1 726 HSD17B8 726 SLC39A7 726 RXRB Col11A2 Classical class II subregion DO BRD2 DM␣ DM Class II␣ Class II PSMB9 TAP1 PSMB8 TAP2 C6orf10 Class III subregion NOTCH4 GPSM3 PBX2
726 917 917 No hit 1109 1109 1109 917 1109 917 1109 1316 1316 1316 1316 No hit 1316 1026 1026
RNF5 AGPAT1
1026 1026
EGFL8 PPT2 C6orf31 KFBPL CREBL1
1026 1207 1026 1026 1026
TNXB CYP21A2 C4 STK19
1026 1026 1026 895
DOM3Z SKIV2L RDBP BF
895 1207 1207 1207
C2
895 1207
ZBTB12 BAT8 C6orf29
1207 1207 1207
NEU1
895
HSP70
547
X. tropicalisa
X. laevisa
Percentage of Identities/aab
No hit No hit No hit
BC087446 (F) BC078516 (F) AF130453 (F)
54/196 (Full) 73/109 (Full) 54/571 (Full)
CR848323 (F) BX759743 No hit No hit AL594530 AL633707 BX737105 No hit AL960564 BX758938 AL628421 CR760242 (F) BX721639 AL968958 AL868786
U82809 (F) BC079997 BJ042776 CF283998 No hit BC084766 (F) BC077337 (F) BP688393 BC068873 (F) BJ057872 BC081039 (F) No Hit BQ733936
AAB40402 48/229 (Gen) 51/538 (Gen) 31/193 (Part) 41/734 (Gen) 75/156 (Full) 60/551 (Full) 38/186 (Gen) 98/131 (Full) 72/621 (Gen) Q66J69 67/251 (Full) 73/161 (Gen)
BC073179
AAH73179
No hit
CB942366 BF845697
67/1476 (Gen)
No hit BX750317 No hit No hit CR760040 CF524557 BC087775 NM_001003660
No hit U51449 BC061681 (F) DQ268506 (F) AF454378 (F)
NA AAB18943 38/229 (Full) 31/241 (Full) AAL58434
D13684 (F) D87687 (F)
BAA02841 BAA19759
No hit AB033151 No hit No hit
AY204552 (F) D44540 (F) AY204554 (F) No hit
AAP36718 BAA07945 AAP36720 NA
No hit No hit AL679273 BX733561 AL848378 AL959082 BX758119 CR761323 (F) No Hit BX730761 BX741700 AL878690 AL872993 No hit No hit No hit AL866392 AL680203 CR761194 AL648835 BX705721 AL658338 BQ519832 BX711550 BX739487 CV811291 BX772914 AL882026 AL656289 AL867952 BX730249 BX736306 CX414897
No hit BP703429 BC071048 (F)
38/1426 (Gen) 53/62 (Gen) 84/357 (Full)
CA793548 (F) BC081085 (F)
80/120 (Full) 71/265 (Full)
No hit BC059297 (F) BC074389 No hit No hit
42/281 66/283 64/225 40/259 53/402
BX846582 BC079793 (F) D78003 (F) BC087397 (F)
40/1103 (Gen) 45/460 (Full) BAA11188 45/266 (Full)
BC088900 (F) CV077949 No hit D49373 (F)
53/392 (Full) 56/341 (Gen) 78/135 (Gen) BAA08371
DQ268507
39/752 (Full)
BC072114 (F) BX854837 BC073678 (F)
58/468 (Full) 69/558 (Gen) 67/703 (Full)
BC074166 (F)
P12890
X01102 (F)
(Full) (Full) (Gen) (Gen) (Gen)
P02827 (Table continues)
3678
SYNTENIC INTEGRITY OF MHC THROUGH EVOLUTION Table I. Continued Percentage of Identities/aab
Scaffold
X. tropicalisa
X. laevisa
LSM2 VARS2
547 547
BP708188 (F) BC084762 (F)
AAH90606 66/1211 (Full)
C6orf26
547
AW644535
36/149 (Part)
MSH5 CLIC1 DDAH2 LY6G6C LY6 BAT5 CSNK2B BAT4 APOM BAT3
547 547 547 547 547 547 547 895 895 895
BP691385 AY277695 (F) BC078574 (F) No hit No hit BC077872 (F) BC083993 (F) BJ045925 BC078609 BC060479 (F)
57/512 (Gen) AAH59765 55/278 (Full) 31/100 (Full) 30/123 (Full) AAH75571 93/234 (Full) 39/361 (Gen) 45/188 (Gen) 51/1185 (Full)
BAT2
895
BX753796 (F) AL967562 BX730148 CF224182 CF224183 No hit BC059765 (F) BC075381 (F) BX687777 (F) AL958403 (F) BC075571 (F) BC077003 (F) BG487159 No hit BX764202 BG348638 BG486156 AL955051 AL966216 BX698872 AL633000 AL792183 BX732414 BX716126 BX771103 No hit No hit AL661130 No Hit BC061280 (F)
BC051009 (F)
37/2232 (Full)
BQ398178 BU905532 No hit BJ614008 CD254193 BC084004 (F)
34/152 (Gen) 35/207 (Gen) 38/159 (Gen) 34/392 (Gen) 68/117 (Gen) AAH61280
No hit BX769629 AL646254 No hit AL866886 CR760749 BC074549 (F) AL969933 BX752964 BX779269 No hit BX738291 AL632457 BX696867 BC087624 (F) No hit BX740904 AL633731 AL871337 AL629506 AL628261 BX44344 No hit BX727166 BX731212 No hit No hit
No hit AW782810 BF613095 BC060755 (F) AF545659 (F)
NA 40/332 (Gen) 32/712 (Gen) 60/930 (Full) AAN86277
BC049004 (F) BP681667
AAH75549 44/246 (Gen)
BC073486 (F) BJ636557 BP709014
53/253 (Full) 48/145 (Gen) 89/529 (Gen)
No hit BC089285 (F) BP705701 BC074405 (F)
45/204 (Full) 55/429 (Full) 55/201 (Gen) Q6GLQ4
BC081034 (F) BJ614782 No hit
87/600 (Full) 61/618 (Gen) 78/629 (Gen)
No hit No hit
49/291 (Gen) 48/309 (Gen)
Gene
LTB 895 TNF 895 LTA 895 NFKBIL1 895 ATP6V1G2 895 BAT1 895 Classical and extended class I subregion POU5F1 No hit TCF19 547 C6orf18 547 DDR1 547 FLOT1 547 TUBB MDC1
547 547
NRM KIAA1949 DHX16
547 547 547
C6orf136 C6orf134 MRPS18B PPP1R10
547 547 547 547
ABCF1 GNL1 GABBR1
547 547 726
Olfactory receptor Olfactory receptor
726 726
Gene
Scaffold
Human chromosome no.
X. tropicalisa
Genes not found in the human MHC, but in the Xenopus scaffoldsa KIAA1720 726 1 AL774574 408 ribosomal protein S5 726 X BX718023 CR848217 RASA1 726 5 No Hit HGMA1L1 726 X BX741519 Thymopoietin 917 12 No hit MAP1 1109 14 CX40492 Carnitine O-acetyltransferase 1026 9 BC063356 (F)
X. laevisa
Percentage of Identities/aab
BC084800 (F) 62/398 (Full) BC054263 (F) 96/203 (Full) CD253980 BX849417 No hit BC073201 BC072849 (F)
70/708 (Gen) 43/91 (Gen) 31/107 (Gen) 26/361 (Gen) 50/587 (Full) (Table continues)
The Journal of Immunology
3679
Table I. Continued
Gene
CENPA PSMB10 PDE4DIP Similar to CTGF NPDC1-like RGS3
Scaffold
Human chromosome no.
1026 1026 1207
2 16 1
547 547 895
6 (not MHC) 9 9
Scaffold
X. tropicalisa
AL863713 No hit AL633679 BX700846 No hit No hit No hit
Percentage of Identities/aab
X. laevisa
BC092389 (F) BC056039 (F) No Hit
67/90 (Full) 62/276 (Full) 22/350 (Gen)
BG234029 No hit CA788699
40/132 (Gen) 30/120 (Gen) 36/176 (Gen)
X. tropicalis
X. laevis
Unknown genes (ORF) found in the Xenopus scaffolds 726 AL635951 726 No hit 726 BX751117 1109 No hit 1109 1316 547 547 547 895 895 895
No hit No hit No hit CD099547 BP700625 No hit No hit No hit No hit No hit No hit CB942034 No hit
BX688173 AL963961 BX781893 BX717682 No hit AL868311 No hit No hit
a
Percentage of Full-length DNA sequences are noted as (F). Percentage of amino acid identities are shown in this Table over the matching region. We used the longest and more reliable sequences, either partial cDNA, or genomic sequences retrieved from scaffolds, or full-length cDNA sequences when available. The query sequences used for this column were as follows: full-length cDNA sequences (Full), partial-length cDNA sequences (Part), retrieved genomic sequences (Gen). Accession no. are shown when the sequences were annotated. NA, Not applicable. b
from both ends, confirming that the gap between scaffolds 895 and 1207 is short and contains a single intron of a factor B gene. In the MHC of all teleosts so far studied, the three (or more) immunoproteasome genes are tightly linked to class I genes (Refs. 18, 20, 45– 47; also see Fig. 5), and thus we were surprised to find the third immunoproteasome gene PSMB10 in the class III region. Because teleost MHC genes are found in many linkage groups and spread onto different chromosomes, PSMB10 consequently may have remained in the class I region in bony fish as a result of translocation of ancestral class III region genes out of the MHC (12, 15, 48) and coevolution via “functional clustering” of immunoproteasome and class I genes (49 –51). In contrast, because most genes have maintained their ancient synteny in the Xenopus MHC, it is likely that early in evolution PSMB10 indeed was located in the class III region. Alternatively, PSMB10 translocated out of the class I region in Xenopus, and its present location is a derived
characteristic. We await studies of the elasmobranch class I region to elucidate the original location of PSMB10. We found a gene similar to the complement C2 gene in the scaffold 1207 near the factor B gene. Phylogenetic analysis strongly supports that the gene is more similar to C2 than to factor B (Fig. 3). In teleost fish, the Bf/C2 genes are often duplicated, and one of these genes encodes a protein that, like C2, functions in the classical complement pathway (48, 52–54). However, upon phylogenetic tree analysis, most of the teleost genes form a monophyletic cluster independent of tetrapod Bf and C2 clusters. Thus, whereas it is still not clear whether the Bf/C2 gene duplication and functional differentiation predated the emergence of teleosts, our data clearly demonstrate that the Bf/C2 duplication predated the appearance of amphibians. The three TNF members encoded in the human class III region (LTA, LTB, and TNF) are found in the
Table II. Genes found in the Xenopus MHC
Category
Ag processing/presentation Inflammation Leukocyte maturation Complement Immune regulation Stress response Ig superfamily Olfactory receptors Nonimmune genes Genes not found in the human MHC Unknown genes Total number of genes
Number of Genes
11 5 3 3 3 1 6 2 64 13 12 122
Genes
Class I, class II␣, class II, DM␣, DM, PSMB8, PSMB9, PSMB10, TAP1, TAP2, TAPBP ABCF1, DAXX, LTA, LTB, TNF␣ DDAH1/2, LY6, LY6G6C BF, C2, C4 NFKBIL1, RXRB, FKBPL HSP70 XMIV PSMB10 and others ORF
3680
SYNTENIC INTEGRITY OF MHC THROUGH EVOLUTION Table III. Update on scaffold assembly Scaffolds
Version 3.0 726 (656,544 bp) 917 (462,912 bp) 1109 (310,645 bp) 1316 (205,920 bp) 1026 (374,626 bp) 1207 (255,515 bp) 1207 (255,515 bp) 895 (482,803 bp) 547 (899,709 bp)
Version 4.0 396 (1,113,890 bp) 396 (1,113,890 bp) 895 (310,644 bp) 1038 (205,919 bp) 744 (484,856 bp) 744 (484,856 bp) 1175 (141,455 bp) 752 (471,867 bp) 488 (901,819 bp)
Xenopus MHC, suggesting that the synteny is old, and that nonMHC linkage of the teleost TNF family members is a derived feature (55, 56), like most of the other class III genes in this phylogenetic group. Of greatest interest to us, there is a cluster of related genes residing on the edge of the two scaffolds 547 and 895 (Fig. 2 and Table II). From the order, positions, and similarity of these genes, we predict that scaffolds 547 and 895 are tightly linked; however, BLAST searches using end sequences failed to unite them. The conserved cysteines (C), tryptophan (W), and spacing between the deduced amino acid sequences mark these genes as Igsf members, and the GXG motif in the G-strand suggests that these sequences could be V domains in the Ag receptor family (Fig. 4A), thus we named them XMIV (Xenopus MHC-linked Ig superfamily V genes). In fact, BLAST searches usually selected TCR or IgNAR V domains (58) as the most similar sequences, although with low sequence identity (data not shown). The putative expressed sequences were pieced together from the genomic exons. Some of the XMIV genes seem to contain cytoplasmic tails followed by one or two ITIMs (shaded in Fig. 4A) (22, 23), suggesting these XMIV molecules are inhibitory receptors. Other genes have positively charged amino acids in their transmembrane (TM) regions, suggesting that they could interact with ITAM-containing adaptors (21); however, these lysine (K) and/or arginine (R) residues are usually located centrally in the TM in activating receptors. One to four additional cysteine residues found in the IgSF domains in the XMIV may form intra- and/or inter-chain disulfide bridges (boxed in Fig. 4A). Unfortunately, our EST database searches resulted in only one full-length entry (CN328971) for these XMIV genes, in X. laevis (Fig. 4A). To confirm that CN328971 is indeed the X. laevis counterpart of the X. tropicalis genes, we performed Southern blotting on a family with known MHC haplotypes (27). In all 20 siblings, restriction fragment length polymorphism for CN328971 matched perfectly to the known MHC haplotypes in the family (Fig. 4B). In the human MHC, NKp30 is found in the location where XMIV genes are present in the Xenopus MHC (see NCR3 in Fig. 2). However, we found multiple genes that showed significant similarity to human NKp30 in other scaffolds, and these XMIV genes are not similar to NKp30 (amino acid identity ⬍22–26%), as
FIGURE 2. Comparison of the X. tropicalis MHC to the human MHC (modified from Ref. 4) reveals extraordinary conservation. Partial human MHC genes are listed as the template in the center and the Xenopus MHC
scaffolds on both sides. Subregions of the human MHC are color-coded: area flanking to the extended class II region, gray; extended class II region, pink; class II region, blue; class III region, peach; class I region, green; extended class I region, yellow. Gene symbols were followed as assigned by the HUGO Gene Nomenclature Committee and ImMunoGeneTics/HLA Sequence Databases. Xenopus genes found outside the human MHC are marked as green boxes, with the human chromosome numbers listed as red superscripts. Igsf genes unique to the Xenopus MHC on scaffolds 547 and 895 are shown in red boxes. Transcriptional orientations are indicated as gradients and arrows on the right bottom.
The Journal of Immunology
3681 EST databases (BC073304, BC074259) when screening with the human gene (E-value of ⬍e⫺39). Xenopus C9orf58 gene is on scaffold 191 and linked to other genes encoded on human chromosome 9. Thus, owing to differential silencing after the en bloc duplications early in vertebrate evolution (59), Xenopus AIF1 was shut down in the MHC, whereas functional human AIF1 and C9orf58 genes are on chromosomes 6 and 9, respectively.
Genes in the mammalian class I region
FIGURE 3. Phylogenetic tree of Factor B and C2 sequences identifies Xenopus C2. The tree was constructed by the Neighbor-Joining Method based on an alignment of amino acid sequences using Clustal X version 1.81. Numbers indicate the bootstrap values, supporting the depicted partitioning from 1000 trials. Genetic distance is shown as a bar on the bottom. Abbreviations and accession nos. for each sequence are as follows: Hosa Bf, Homo sapiens Bf (P00751); Hosa C2 (AAB97607); Mumu Bf, Mus musculus Bf (NM_008198); Mumu C2 (NM_013484); Xela BfA, X. laevis BfA (BAA06179); Xela BfB (BAA08371); Xela C2, X. laevis C2 (ABB85337); Omny Bf-1, Oncorhynchus mykiss Bf-1 (AAC83699); Omny Bf-2 (AAC83698); Orla Bf/C2, Oryzias latipes Bf/C2 (BAA12207); Dare Bf, Danio rerio Bf (NP_571413); Cyca Bf/C2B, Cyprinus carpio Bf/C2B (BAA34707); Cyca Bf/C2-A3 (BAB32650); Trsc Bf, Triakis scyllium Bf (BAB63203); Leja Bf, Lethenteron japonicum Bf (I50807); and Stpu Bf, Strongylocentrotus purpuratus Bf (AAC79682).
mentioned above. It is possible that the scaffold containing Xenopus NKp30 (data not shown) could be between scaffolds 547 and 895, although it seems unlikely because this would be too large of a disruption (⬃2 Mb) in the midst of the other densely packed class III genes. Furthermore, searches of the X. tropicalis genome with Igsf domains only selected these MHC scaffolds, suggesting that all XMIV members are in the MHC. All of the Xenopus species (2n–12n) have multiple copies of XMIV genes, with no obvious increase in gene numbers in the higher-order polyploids (Fig. 4C), like for many other immune genes (Ref. 58 and unpublished data). The following class III genes were not detected on the scaffolds or in the EST databases: AGER, C6orf48, C6orf27, C6orf25, LY6G6E, LY6G6D, LY6G5C, LY6G5B, C6orf47, and LST1. The LY6 family members in human seem to be derived from recent duplications and thus would not be expected to be found in Xenopus. The AIF1 paralogue, C9orf58, was found in the X. laevis
The human class I region designation cannot be applied to nonmammalian species, because class I genes are embedded within the class II region closely linked to the immunoproteasome and transporter genes (Figs. 2 and 5). Despite the absence of class I genes in this region, we found 15 Xenopus genes orthologous to the human genes, including TUBB and FLOT1, which are also located in the teleost MHC (linked to the teleost class I region) (Fig. 5). Thus, the architecture/framework of the extant mammalian class I region pre-existed 450 million years ago and appears stable over evolutionary time; class I genes were translocated from the true class I region and expanded in the modern class I region in the mammalian lineage, as previously proposed (6). GABBR1 and two olfactory genes on scaffold 726 are found outside of the extended class II region, suggesting a reorganization of the genes either in an ancestor of Xenopus or human. No MOG-containing Igsf domains were found in scaffolds 726 and 547, consistent with the fact that we did not find any other Igsf-containing human homologues such as AGER, C6orf25, or BTN in any region of the MHC. However, when we extended our analysis to the Xenopus nonclassical class I (XNC) genes (60), which are located at the telomere of the same arm of the chromosome as MHC (which is near the centromere), a cluster of BTN genes was indeed identified, near to the XNC genes (data not shown). These data demonstrate that the class I-BTN association is old.
Categories of genes in the Xenopus MHC Next, we classified genes found in the Xenopus MHC by their functions (Table II). We found genes belonging to each category as detailed in the human MHC such as those involved in the following: Ag processing for class I and class II molecules, inflammation, leukocyte maturation, complement, immune regulation, Igsf, and heat shock protein (HSP). Again, the overall MHC architecture is well conserved. In the human MHC, ⬃28% of the expressed transcripts are potentially associated with immunity (4). In the Xenopus MHC, 32 genes (26.2%) fall into this category, also quite similar to that of human. Thirteen genes were found in Xenopus that are not in the human MHC, five of which are encoded on MHC paralogous regions: KIAA1720 and ODE4DIP on human chromosome 1, and RGS3, Carnitine acetyltransferase, and NPDC1-like on human chromosome 9. As described above for AIF1, the likely explanation for this finding is differential silencing (in these cases) on the MHC paralogous regions in Xenopus. Furthermore, as mentioned above PSMB10 is found in the Xenopus MHC, whereas it is located in humans on chromosome 16. Previous work in teleosts suggested that PSMB10 was originally located in the MHC class I region, and subsequently translocated onto a separate chromosome in the mammalian genome (51). Interestingly, we found carnitine acetyltransferase in the vicinity of the constitutive proteasome subunit and direct homologue of PSMB10 and PSMB7, further supporting the idea that PSMB8, -9, and -10 arose from duplication of PSMB5, -6, and -7, with subsequent
3682
FIGURE 4. A, Deduced amino acid alignment of XMIV genes found in the X. tropicalis scaffolds 547 and 895, and an entry found in the X. laevis EST database CN328971. Sequence numbers correlate with the scaffold and location. ORFs for XMIV1 and -4 contain two Igsf domains and are designated as ⫺1 and ⫺2. Evolutionary conserved amino acids, ITIM, and positively charged residues in the TM and cytoplasmic (CYT) regions are highlighted in gray, and extra cycteins are boxed. The X. tropicalis sequences were pieced together from exons on the two scaffolds. B, Linkage of the X. laevis gene (CN328971) to the Xenopus MHC. A X. laevis family (f/g ⫻ f/r) with known MHC haplotypes (f, g, r) was used for the linkage analysis. The father is indicated as P, and siblings are indicated with numbers. The previously determined MHC haplotypes are shown above the blot (27). Haplotype-specific bands are shown as arrows on the left. C, Novel MHC-linked XMIV members are present in other polyploid Xenopus species. Southern blotting with the Igsf domain of CN328971 (X. laevis) probe was performed under low stringency conditions. Ploidy levels are noted underneath the blot.
SYNTENIC INTEGRITY OF MHC THROUGH EVOLUTION
The Journal of Immunology
FIGURE 5. Genomic organization in the Xenopus class I region compared with the medaka (47) and the chicken (13) MHCs. Transcriptional orientations are indicated on each side of the center bars (arrows at bottom right). Boxes noting class Ia genes and pseudogenes are p and f, respectively.
translocation of PSMB5 and -6 (51). Synteny of PSMB10 and carnitine acetyltransferase in the Xenopus MHC suggests that differential silencing resulted from the presence of PSMB7-carnitine acetyltransferase in the primordial MHC. Two olfactory genes were found, but it is not clear whether these genes are orthologous to those in the human MHC. There are 12 unknown genes of which nine genes are found in the EST database, suggesting functional genes.
Discussion Genes involved in immune responses evolve rapidly, likely to combat rapidly evolving or emerging pathogens (61, 62). This rapid evolution of immune genes can be obstructive when pursuing orthologous genes in divergent species; BLAST searches of the genome using amino acid sequences from other species often resulted in “no hit.” Sometimes, even using the relatively closely related X. laevis did not result in identification of the X. tropicalis sequences. By contrast, searches using genes in large families such as TAP1 and TAP2 resulted in too many hits because the conserved ATP-binding domain selects other ABC transporter genes as well. Similar problems were observed with the paralogous genes. Vertebrate genome analysis had revealed that there are three other gene complexes similar to the MHC (63). We were usually successful in identifying which of the paralogues was orthologous to the human MHC counterpart, but sometimes it was not clear. For example, we could not identify Xenopus DDAH as either DDAH2 (human MHC) or DDAH1 (chromosome 9), and thus named the Xenopus clone DDAH1/2 in Fig. 2. This was true of NOTCH as well, where different exons seemed to be orthologous to the different NOTCH paralogues; therefore, we could not confidently identify the MHC-encoded gene as NOTCH4. These phenomena
3683 can be explained as differential subfunctionalization or neofunctionalization of the genes between human and Xenopus over evolutionary time. For these reasons, each scaffold was carefully examined by eye, and the final decision was made in most cases when the linkage was conserved in the vicinity of the gene in question. For example, TNXB is in MHC, and TNC is in paralogous regions of chromosome 9; searching the genomic database with TNC resulted in selection of a scaffold that does not contain genes orthologous to MHC, whereas we found C4 and PSMB10 genes closely linked to TNXB. Because of the highly conserved synteny and relatively large scaffolds, we were able to distinguish MHC scaffolds from paralogous scaffolds. An additional point to be emphasized is that the scaffolds have been assembled automatically, and although the standard of the assembly is high, the assembly is incomplete and perhaps incorrect in some places. From our previous family analyses, we identified two class II␣ and two class II genes in X. tropicalis (personal observation), but our searches did not select other scaffolds. In addition, this region contains highly repetitive and transposable elements (and, unfortunately, an artifact), making assembly difficult. Previously, we found multiple MHC-linked HSP70 genes (unpublished data), but we only found one in these scaffolds, suggesting that the sequence may not be entirely accurate in these regions. It is possible, or even likely, that regions containing some repeats are biased or difficult to sequence and/or assemble. During the writing of this manuscript new versions of the genome assembly were released (version 4.0 and 4.1, coverage 7.65 genome equivalents). The sizes and general organization are almost identical with the previous version except that two scaffolds were connected (Table III). From our previous work using recombinant X. laevis, we ordered the nonmammalian class I region as class II, TAP/LMP, class I/C4 (6, 10). However, as shown in Fig. 2, the order of the genes in the X. tropicalis scaffolds is class II, class I, TAP/LMP, C4. Again, this could either be due to an assembly error in X. tropicalis or alternatively to genomic re-organization that happened during tetraploidization. The frog used for the scaffold assembly was heterozygous for the two class I region lineages that are found in all of the Xenopus species (10, 64). We have found that the lineages of class I/PSMB8/TAP1/TAP2 are always found within a set in wild-caught animals (10), suggesting that there is a block in recombination between genes in these lineages, perhaps because of major sequence modifications in noncoding regions, recently shown to be true in medaka (65). Unfortunately, the assembly was complete for only one of the lineages (lineage A) in this particular region, so we will have to wait to test this hypothesis. There are two large clusters of histone and tRNA genes in the human extended class I subregion. It is proposed that the MHC may have hitchhiked with these clusters (or vice versa) to maximize transcriptional activity (4). Unfortunately, our analysis of Xenopus MHC did not include this extended class I region to examine whether the cluster of histone and tRNA genes is an evolutionary conserved feature of MHC. We await future versions of the genome project to examine this question. In the vicinity of the human MHC, there are 34 olfactory-receptor loci, 14 of which are potentially functional. Sperm-expressed olfactory-receptor genes may be functionally involved in the selection of spermatozoa by the female (sperm receptor selection hypothesis), as well as many other functions (66). Thus far, we have found two olfactory receptors in the MHC and XNC scaffolds (data not shown). Therefore, it will be interesting to determine
3684 whether there are larger clusters of olfactory genes near the Xenopus MHC genes and determine their tissue and ontogenic distribution. The chicken MHC contains putative NK receptors in the C-type lectin family that are most related to genes in the mammalian NK cell complex (NKC) (13, 67). This finding not only demonstrates a common evolutionary origin of the NK cell complex and MHC, but also suggests coevolution of the linked NK cell and class I receptors (13, 68). Similarly, the discovery of XMIV genes most related to Ag receptor genes in the Xenopus MHC suggests that they could coevolve with polymorphic class I and/or class II molecules. In addition, based on the fact that class I, class II, and Ag receptor genes all have the specialized Igsf C1-type domain (69, 70), it seems likely that the ancestral Ag receptor genes were genetically linked to the MHC-presenting genes. It is possible that the Igsf genes encoded in the Xenopus MHC are relics of such original Ag receptors. As described throughout the text, close linkage of class I-presenting genes and processing genes is found in all nonmammalian vertebrates (Fig. 5). Interestingly, similar to what has been described recently in the chicken (71), the Xenopus TAP genes TAP1/ TAP2 are in opposite transcriptional orientations and may use a bidirectional promoter. However, the MHCs in birds and teleost fish have been extensively modified (5, 7, 12, 14 –16). A few of the genes in the extended class II and class I regions have maintained their synteny in bony fish (47, 72), but by and large their MHC genes have been scattered throughout the genome. The teleosts studied to date have small genomes, and their MHC genes may be indicative of general modifications that have taken place for other syntenic groups (15). This may be true of the chicken and quail as well, in which the MHC is encoded on a microchromosome where genes and intergenic distances have been greatly shortened compared with most vertebrates (13, 19). Birds have seemingly dispensed altogether with immunoproteasome genes and perhaps other housekeeping genes. In contrast, this current study (and our previous predictions) has shown that Xenopus is much similar to human and allows for an understanding of the common MHC ancestor, at least at the level of amphibian emergence. By contrast, the mammalian MHC has a peculiar organization in which the class I genes are not closely linked to TAP and PSMB genes. The loss of this linkage seems to be accompanied by the loss of the genetic dimorphism of the linked class I/PSMB/TAP genes observed in amphibian, teleosts, and cartilaginous fish (10, 64, 65, 73). Thus, the amphibian seems to be the best model to study evolution of the vertebrate MHC.
Acknowledgments We thank Louis Du Pasquier for discussions about the data.
Disclosures The authors have no financial conflict of interest.
References 1. Kumanovics, A., T. Takada, and K. F. Lindahl. 2003. Genomic organization of the mammalian MHC. Annu. Rev. Immunol. 21: 629 – 657. 2. Driscoll, J., M. G. Brown, D. Finley, and J. J. Monaco. 1993. MHC-linked LMP gene products specifically alter peptidase activities of the proteasome. Nature 365: 262–264. 3. Antoniou, A. N., S. J. Powis, and T. Elliott. 2003. Assembly and export of MHC class I peptide ligands. Curr. Opin. Immunol. 15: 75– 81. 4. Horton, R., L. Wilming, V. Rand, R. C. Lovering, E. A. Bruford, V. K. Khodiyar, M. J. Lush, S. Povey, C. C. Talbot, Jr., M. W. Wright, et al. 2004. Gene map of the extended human MHC. Nat. Rev. Genet. 5: 889 – 899. 5. Flajnik, M. F., and M. Kasahara. 2001. Comparative genomics of the MHC: glimpses into the evolution of the adaptive immune system. Immunity 15: 351–362.
SYNTENIC INTEGRITY OF MHC THROUGH EVOLUTION 6. Nonaka, M., C. Namikawa, Y. Kato, M. Sasaki, L. Salter-Cid, and M. F. Flajnik. 1997. Major histocompatibility complex gene mapping in the amphibian Xenopus implies a primordial organization. Proc. Natl. Acad. Sci. USA 94: 5789 –5791. 7. Kelley, J., L. Walter, and J. Trowsdale. 2005. Comparative genomics of major histocompatibility complexes. Immunogenetics 56: 683– 695. 8. Vogel, G. 1999. Frog is a prince of a new model organism. Science 285: 25. 9. Kobel, H. R., and L. Du Pasquier. 1986. Genetics of polyploid Xenopus. Trends Genet. 12: 310 –314. 10. Ohta, Y., S. J. Powis, R. L. Lohr, M. Nonaka, L. Du Pasquier, and M. F. Flajnik. 2003. Two highly divergent ancient allelic lineages of the transporter associated with antigen processing (TAP) gene in Xenopus: further evidence for co-evolution among MHC class I region genes. Eur. J. Immunol. 33: 3017–3027. 11. Koch, C. 1986. A genetic polymorphism of the complement component factor B in chickens not linked to the major histocompatibility complex (MHC). Immunogenetics 23: 364 –367. 12. Graser, R., V. Vincek, K. Takami, and J. Klein. 1998. Analysis of zebrafish Mhc using BAC clones. Immunogenetics 47: 318 –325. 13. Kaufman, J., S. Milne, T. W. Gobel, B. A. Walker, J. P. Jacob, C. Auffray, R. Zoorob, and S. Beck. 1999. The chicken B locus is a minimal essential major histocompatibility complex. Nature 401: 923–925. 14. Sato, A., F. Figueroa, B. W. Murray, E. Malaga-Trillo, Z. Zaleska-Rutczynska, H. Sultmann, S. Toyosawa, C. Wedekind, N. Steck, and J. Klein. 2000. Nonlinkage of major histocompatibility complex class I and class II loci in bony fishes. Immunogenetics 51: 108 –116. 15. Naruse, K., S. Fukamachi, H. Mitani, M. Kondo, T. Matsuoka, S. Kondo, N. Hanamura, Y. Morita, K. Hasegawa, R. Nishigaki, et al. 2000. A detailed linkage map of medaka. Oryzias latipes: comparative genomics and genome evolution. Genetics 154: 1773–1784. 16. Juul-Madsen, H. R., T. S. Dalgaard, B. Guldbrandtsen, and J. Salomonsen. 2000. A polymorphic major histocompatibility complex class II-like locus maps outside of both the chicken B-system and Rfp-Y-system. Eur. J. Immunogenetics 27: 63–71. 17. Kuroda, N., F. Figueroa, C. O’hUigin, and J. Klein. 2002. Evidence that the separation of Mhc class II from class I loci in the zebrafish, Danio rerio, occurred by translocation. Immunogenetics 54: 418 – 430. 18. Phillips, R. B., A. Zimmerman, M. A. Noakes, Y. Palti, M. R. Morasch, L. Eiben, S. S. Ristow, G. H. Thorgaard, and J. D. Hansen. 2003. Physical and genetic mapping of the rainbow trout major histocompatibility regions: evidence for duplication of the class I region. Immunogenetics 55: 561–569. 19. Shiina, T., S. Shimizu, K. Hosomichi, S. Kohara, S. Watanabe, K. Hanzawa, S. Beck, J. K. Kulski, and H. Inoko. 2004. Comparative genomic analysis of two avian (quail and chicken) MHC regions. J. Immunol. 172: 6751– 6763. 20. Shiina, T., J. M. Dijkstra, S. Shimizu, A. Watanabe, K. Yanagiya, I. Kiryu, A. Fujiwara, C. Nishida-Umehara, Y. Kaba, I. Hirono, et al. 2005. Interchromosomal duplication of major histocompatibility complex class I regions in rainbow trout (Oncorhynchus mykiss), a species with a presumably recent tetraploid ancestry. Immunogenetics 56: 878 – 893. 21. Reth, M. 1989. Antigen receptor tail clue. Nature 338: 383–384. 22. Long, E. O., D. N. Burshtyn, W. P. Clark, M. Peruzzi, S. Rajagopalan, S. Rojo, N. Wagtmann, and C. C. Winter. 1997. Killer cell inhibitory receptors: diversity, specificity, and function. Immunol. Rev. 155: 135–144. 23. Vivier, E., and M. Daeron. 1997. Immunoreceptor tyrosine-based inhibition motifs. Immunol. Today 18: 286 –291. 24. Shum, B. P., D. Avila, L. Du Pasquier, M. Kasahara, and M. F. Flajnik. 1993. Isolation of a classical MHC class I cDNA from an amphibian: evidence for only one class I locus in the Xenopus MHC. J. Immunol. 151: 5376 –5386. 25. Sato, K., M. F. Flajnik, L. Du Pasquier, M. Katagiri, and M. Kasahara. 1993. Evolution of the MHC: isolation of class II -chain cDNA clones from the amphibian Xenopus laevis. J. Immunol. 150: 2831–2843. 26. Liu, Y., M. Kasahara, L. L. Rumfelt, and M. F. Flajnik. 2002. Xenopus class II A genes: studies of genetics, polymorphism, and expression. Dev. Comp. Immunol. 26: 735–750. 27. Ohta, Y., S. J. Powis, W. J. Coadwell, D. E. Haliniewski, Y. Liu, H. Li, and M. F. Flajnik. 1999. Identification and genetic mapping of Xenopus TAP2 genes. Immunogenetics 49: 171–182. 28. Namikawa, C., L. Salter-Cid, M. F. Flajnik, Y. Kato, M. Nonaka, and M. Sasaki. 1995. Isolation of Xenopus LMP-7 homologues: striking allelic diversity and linkage to MHC. J. Immunol. 155: 1964 –1971. 29. Nonaka, M., C. Namikawa-Yamada, M. Sasaki, L. Salter-Cid, and M. F. Flajnik. 1997. Evolution of proteasome subunits delta and LMP2: complementary DNA cloning and linkage analysis with MHC in lower vertebrates. J. Immunol. 159: 734 –740. 30. Salter-Cid, L., L. Du Pasquier, and M. Flajnik. 1996. RING3 is linked to the Xenopus major histocompatibility complex. Immunogenetics 44: 397–399. 31. Mo, R., Y. Kato, M. Nonaka, K. Nakayama, and M. Takahashi. 1996. Fourth component of Xenopus laevis complement: cDNA cloning and linkage analysis of the frog MHC. Immunogenetics 43: 360 –369. 32. Kato, Y., L. Salter-Cid, M. F. Flajnik, M. Kasahara, C. Namikawa, M. Sasaki, and M. Nonaka. 1994. Isolation of the Xenopus complement factor B complementary DNA and linkage of the gene to the frog MHC. J. Immunol. 153: 4546 – 4554. 33. Salter-Cid, L., M. Kasahara, and M. F. Flajnik. 1994. Hsp70 genes are linked to the Xenopus major histocompatibility complex. Immunogenetics 39: 1–7. 34. Marklew, S., D. P. Smith, C. S. Mason, and R. W. Old. 1994. Isolation of a novel RXR from Xenopus that most closely resembles mammalian RXR  and is expressed throughout early development Biochim. Biophys. Acta 1218: 267–272.
The Journal of Immunology 35. Sharpe, C. R., and K. Goldstone. 1997. Retinoid receptors promote primary neurogenesis in Xenopus. Development 124: 515–523. 36. Povey, S., R. Lovering, E. Bruford, M. Wright, M. Lush, and H. Wain. 2001. The HUGO Gene Nomenclature Committee (HGNC). Hum. Genet. 109: 678 – 680. 37. Wain, H. M., M. J. Lush, F. Ducluzeau, V. K. Khodiyar, and S. Povey. 2004. Genew: the Human Gene Nomenclature Database, 2004 updates. Nucleic Acids Res. 32: D255–D257. 38. Bartl, S., M. A. Baish, M. F. Flajnik, and Y. Ohta. 1997. Identification of class I genes in cartilaginous fish, the most ancient group of vertebrates displaying an adaptive immune response. J. Immunol. 159: 6097– 6104. 39. Page, R. D. 1998. GeneTree: comparing gene and species phylogenies using reconciled trees. Bioinformatics 14: 819 – 820. 40. Nonaka, M., M. Takahashi, and M. Sasaki. 1994. Molecular cloning of a lamprey homologue of the mammalian MHC class III gene, complement factor B. J. Immunol. 152: 2263–2269. 41. Smith, L. C., L. A. Clow, and D. P. Terwilliger. 2001. The ancestral complement system in sea urchins. Immunol. Rev. 180: 16 –34. 42. Flajnik, M. F., K. Miller, and L. Du Pasquier. 2003. Evolution of the immune system. In Fundamental Immunology, 5th ed. W. E. Paul, ed. Lippincott Williams & Wilkins, Philadelphia, pp. 519 –570. 43. Stammers, M., L. Rowen, D. Rhodes, J. Trowsdale, and S. Beck. 2000. BTL-II: a polymorphic locus with homology to the butyrophilin gene family, located at the border of the major histocompatibility complex class II and class III regions in human and mouse. Immunogenetics 51: 373–382. 44. Tazi-Ahnini, R., J. Henry, C. Offer, C. Bouissou-Bouchouata, I. H. Mather, and P. Pontarotti. 1997. Cloning, localization, and structure of new members of the butyrophilin gene family in the juxta-telomeric region of the major histocompatibility complex. Immunogenetics 47: 55– 63. 45. Michalova´, V., B. W. Murray, H. Su¨ltmann, and J. Klein. 2000. A contig map of the Mhc class I genomic region in the zebrafish reveals ancient synteny. J. Immunol. 164: 5296 –5305. 46. Clark, M. S., L. Shaw, A. Kelly, P. Snell, and G. Elgar. 2001. Characterization of the MHC class I region of the Japanese pufferfish (Fugu rubripes). Immunogenetics 52: 174 –185. 47. Matsuo, M. Y., S. Asakawa, N. Shimizu, H. Kimura, and M. Nonaka. 2002. Nucleotide sequence of the MHC class I genomic region of a teleost, the medaka (Oryzias latipes). Immunogenetics 53: 930 –940. 48. Sultmann, H., A. Sato, B. W. Murray, N. Takezaki, R. Geisler, G. J. Rauch, and J. Klein. 2000. Conservation of Mhc class III region synteny between zebrafish and human as determined by radiation hybrid mapping. J. Immunol. 165: 6984 – 6993. 49. Hughes, A. L. 1997. Evolution of the proteasome components. Immunogenetics 46: 82–92. 50. Murray, B. W., H. Sultmann, and J. Klein. 1999. Analysis of a 26-kb region linked to the Mhc in zebrafish: genomic organization of the proteasome component /transporter associated with antigen processing-2 gene cluster and identification of five new proteasome  subunit genes. J. Immunol. 163: 2657–2666. 51. Clark, M. S., P. Pontarotti, A. Gilles, A. Kelly, and G. Elgar. 2000. Identification and characterization of a  proteasome subunit cluster in the Japanese pufferfish (Fugu rubripes). J. Immunol. 165: 4446 – 4452. 52. Sunyer, J. O., I. Zarkadis, M. R. Sarrias, J. D. Hansen, and J. D. Lambris. 1998. Cloning, structure, and function of two rainbow trout Bf molecules. J. Immunol. 161: 4106 – 4114. 53. Nakao, M., Y. Fushitani, K. Fujiki, M. Nonaka, and T. Yano. 1998. Two diverged complement factor B/C2-like cDNA sequences from a teleost, the common carp (Cyprinus carpio). J. Immunol. 161: 4811– 4818. 54. Nakao, M., M. Matsumoto, M. Nakazawa, K. Fujiki, and T. Yano. 2002. Diversity of complement factor B/C2 in the common carp (Cyprinus carpio): three
3685
55.
56.
57.
58.
59. 60.
61. 62. 63. 64.
65.
66.
67.
68. 69. 70.
71.
72.
73.
isotypes of B/C2-A expressed in different tissues. Dev. Comp. Immunol. 26: 533–541. Savan, R., T. Kono, D. Igawa, and M. Sakai. 2005. A novel tumor necrosis factor (TNF) gene present in tandem with the TNF-␣ gene on the same chromosome in teleosts. Immunogenetics 57: 140 –150. Kono, T., J. Zou, S. Bird, R. Savan, M. Sakai, and C. J. Secombes. 2005. Identification and expression analysis of lymphotoxin- like homologues in rainbow trout Oncorhynchus mykiss. Mol. Immunol. In press. Greenberg, A. S., D. Avila, M. Hughes, A. Hughes, E. C. McKinney, and M. F. Flajnik. 1995. A new antigen receptor gene family that undergoes rearrangement and extensive somatic diversification in sharks. Nature 374: 168 –173. Courtet, M., M. Flajnik, and L. Du Pasquier. 2001. Major histocompatibility complex and immunoglobulin loci visualized by in situ hybridization on Xenopus chromosomes. Dev. Comp. Immunol. 25: 149 –157. Holland, P. W., J. Garcia-Fernandez, N. A. Williams, and A. Sidow. 1994. Gene duplications and the origins of vertebrate development. Dev. Suppl. 125–133. Flajnik, M. F., M. Kasahara, B. P. Shum, L. Salter-Cid, E. Taylor, and L. Du Pasquier. 1993. A novel type of class I gene organization in vertebrates: a large family of non-MHC-linked class I genes is expressed at the RNA level in the amphibian Xenopus. EMBO J. 12: 4385– 4396. Hughes, A. L., and M. Yeager. 1997. Molecular evolution of the vertebrate immune system. Bioessays 19: 777–786. Trowsdale, J., and P. Parham. 2004. Mini-review: defense strategies and immunity-related genes. Eur. J. Immunol. 34: 7–17. Kasahara, M. 1998. What do the paralogous regions in the genome tell us about the origin of the adaptive immune system? Immunol. Rev. 166: 159 –175. Flajnik, M. F., Y. Ohta, A. S. Greenberg, L. Salter-Cid, A. Carrizosa, L. Du Pasquier, and M. Kasahara. 1999. Two ancient allelic lineages at the single classical class I locus in the Xenopus MHC. J. Immunol. 163: 3826 –3833. Tsukamoto, K., S. Hayashi, M. Y. Matsuo, M. I. Nonaka, M. Kondo, A. Shima, S. Asakawa, N. Shimizu, and M. Nonaka. 2005. Unprecedented intraspecific diversity of the MHC class I region of a teleost medaka, Oryzias latipes. Immunogenetics 57: 420 – 431. Ziegler, A., G. Dohr, and B. Uchanska-Ziegler. 2002. Possible roles for products of polymorphic MHC and linked olfactory receptor genes during selection processes in reproduction. Am. J. Reprod. Immunol. 48: 34 – 42. Rogers, S. L., T. W. Gobel, B. C. Viertlboeck, S. Milne, S. Beck, and J. Kaufman. 2005. Characterization of the chicken C-type lectin-like receptors B-NK and B-lec suggests that the NK complex and the MHC share a common ancestral region. J. Immunol. 174: 3475–3483. Parham, P. 1999. Immunogenetics. Soaring costs in defence. Nature 401: 870 – 871. Williams, A. F., and A. N. Barclay. 1988. The immunoglobulin superfamily: domains for cell surface recognition. Annu. Rev. Immunol. 6: 381– 405. Du Pasquier, L., I. Zucchetti, and R. De Santis. 2004. Immunoglobulin superfamily receptors in protochordates: before RAG time. Immunol. Rev. 198: 233–248. Walker, B. A., A. van Hateren, S. Milne, S. Beck, and J. Kaufman. 2005. Chicken TAP genes differ from their human orthologues in locus organisation, size, sequence features and polymorphism. Immunogenetics 57: 232–247. Sambrook, J. G., R. Russell, Y. Umrania, Y. J. Edwards, R. D. Campbell, G. Elgar, and M. S. Clark. 2002. Fugu orthologues of human major histocompatibility complex genes: a genome survey. Immunogenetics 54: 367–380. Ohta, Y., E. C. McKinney, M. F. Criscitiello, and M. F. Flajnik. 2002. Proteasome, transporter associated with antigen processing, and class I genes in the nurse shark Ginglymostoma cirratum: evidence for a stable class I region and MHC haplotype lineages. J. Immunol. 168: 771–781.
