NKp30 (NCR3) is a Pseudogene in 12 Inbred and Wild Mouse Strains, but an Expressed Gene in Mus caroli
http://www.100md.com
分子生物学进展 2005年第8期
* MRC Rosalind Franklin Centre for Genomics Research (formerly MRC UK HGMP Resource Centre), Wellcome Trust Genome Campus, Hinxton, Cambridge, United Kingdom; and Centro Nacional de Biotecnología–Consejo Superior de Investigaciones Científicas, Campus Cantoblanco, Madrid, Spain
E-mail: baguado@cnb.uam.es.
Abstract
Ancient duplications and rearrangements of protein-coding segments have resulted in complex gene family relationships. As a result, gene products may acquire new specificities, altered recognition properties, modified functions, and even loss of functionality. The natural cytotoxicity receptor (NCR) family are natural killer (NK)–activating receptors whose members are NKp46 (NCR1), NKp44 (NCR2), and NKp30 (NCR3). The NCR proteins are putative immunoglobulin superfamily members whose ligands are unknown. The NKp46 gene is present and expressed in human and mouse, NKp44 is only present and expressed in human, and NKp30 is present and expressed in human but is a nonexpressed pseudogene in mouse. By searching databases we have detected alternatively spliced forms of the three NCR members. In addition, we have shown by reverse transcription–polymerase chain reaction (RT-PCR) analysis that the human NKp30 gene presents differential expression patterns in tissues. However, no expressed sequence tags (ESTs) are detected for mouse NKp30, and the genomic sequence contains two premature stop codons, which would encode a severely truncated nonfunctional protein. We have sequenced genomic DNA from 13 mouse inbred and wild strains and discovered that NKp30 is a pseudogene in every mouse strain sequenced except Mus caroli where two single nucleotide polymorphisms (SNPs) abolished the premature stop codons. We observed that the laboratory-inbred strains are, for the exonic sequences, genetically identical, except Mus m. musculus C3H. The Mus musculus strains only have a few SNPs, but the rest of the Mus strains have accumulated gradually several SNPs, mainly in the functional immunoglobulin and intracellular domains. RT-PCR analysis performed on RNA from M. caroli tissue samples identified two transcripts, one of which would encode a putative soluble NKp30 protein, also detected in rat but not in human. We have observed that the intracellular domains of NKp30 (and NKp46) are not conserved among the different species, with the most striking difference when comparing human against mouse and rat. The NKp44 gene is only found in human and shows three different splice forms varying in their "stalk" and intracellular domains. Searching for NKp44 orthologs, we found similarity to ESTs from a novel rodent TREM family member, which we termed TREM6, and not to any possible NKp44 ortholog.
Key Words: NCR ? NKp30 ? splice forms ? pseudogene ? Mus caroli ? TREM6
Introduction
Pseudogenes are common and are encountered in a diverse range of life forms, but particularly in vertebrates. Genome complexity has evolved by the generation of gene families via gene duplication, and for reasons that remain contentious, some of these duplicated genes have become nonfunctional pseudogenes. Retrotransposed pseudogenes arise by a completely different mechanism and reflect a different aspect of genome evolution. Once established within a genome, pseudogenes evolve with time, although the mechanisms that control these changes remain very poorly understood (Mighell et al. 2000). Most pseudogenes have multiple features that confirm their nonfunctional status. However, there are genes that have many features of pseudogenes but are actually functional, and there are genes that are currently considered as pseudogenes that may potentially be functional. Accordingly, experimental design and interpretation across the whole field of molecular genetics must take pseudogenes into careful consideration (Mighell et al. 2000).
NKp30 (NCR3 or 1C7) is a natural killer (NK) cell receptor which is a functional gene in human (Pende et al. 1999) but a pseudogene in mouse (Xie et al. 2003). NK cells participate in the innate immune system of mammals. These NK cells lyse abnormal cells which contain downregulated cell surface major histocompatability complex (MHC) class I molecules. The NK-mediated cytotoxicity is regulated by signals from NK surface receptors, which are either activating, the ligands of which only a few are known, or inhibitory receptors, whose ligands are MHC class I molecules (Moretta et al. 2001, 2002a). The inhibitory receptor response is dominant over the activating receptor one and, consequently, only if a target cell lacks or has low expression of self–MHC class I proteins on its surface will the activating signals come into effect (Biassoni et al. 2000; Campbell and Colonna 2001; Moretta et al. 2002b, 2003; Lanier 2003; L. Moretta and A. Moretta 2003).
In humans and mice there are a number of distinct families of inhibitory and activating NK receptors, which reside in two main gene clusters on different chromosomes. On human 12p12-13 is the NK C-type lectin receptor gene cluster, with natural killer cell receptor P1 (Lanier, Chang, and Phillips 1994), CD69 (Schnittger et al. 1993), LOX1 (oxidized low-density lipoprotein receptor 1) (Yamanaka et al. 1998), CD94 (Chang et al. 1995), NKG2-D (Ho et al. 1998), and Ly49L (Westgaard et al. 1998) (which has a premature stop codon and is thought to be nonfunctional). In mouse, the orthologous chromosome 6 region has representatives of all of these NK receptors with the addition of 14 Ly49 family genes (Brown et al. 1997; Smith, Idris, and Yokoyama 2001). On human chromosome 19q13.42 is the leukocyte receptor cluster (LCR) (Wende, Volz, and Ziegler 2000), which has the following gene/families: killer inhibitory receptor (KIR) family (Wagtmann et al. 1997; Trowsdale et al. 2001), immunoglobulin (Ig)–like transcripts (Samaridis and Colonna 1997), leukocyte-associated inhibitory receptors (LAIRs) (Colonna et al. 1999), Fc alpha receptor (Borges et al. 1997; Torkar et al. 1998), killer activating protein (KAP10/DAP10) (Wilson, Lindquist, and Trowsdale 2000; Wu et al. 2000), killer activating receptor–associated protein (KARAP/DAP12) (Wilson, Lindquist, and Trowsdale 2000; Wu et al. 2000), and NKp46 (or NCR1) (Pessino et al. 1998; Biassoni et al. 1999). The orthologous mouse LCR region is on the syntenic chromosome 7 region and has a similar gene/family array except for the presence of paired Ig-like transcripts (Takai and Ono 2001) instead of LAIR and the lack of the Fc alpha receptor and, more noticeably, the KIR gene family homologues.
The human natural cytotoxicity receptor (NCR) family members NKp46 (NCR1) (Pessino et al. 1998; Biassoni et al. 1999), NKp44 (NCR2) (Vitale et al. 1998), and NKp30 (NCR3) (Pende et al. 1999) are located on human chromosome 6. The mouse counterpart of NKp46 is located on the syntenic region of mouse chromosome 7, but for NKp44, so far, no murine homologue has been identified anywhere in the mouse genome. NKp30 is located in the class III region of the human MHC (chromosome 6) (Neville and Campbell 1999), and the mouse homologue is on the syntenic chromosome 17 MHC region. However, the annotated NKp30 mouse genomic sequence clearly shows two premature stop codons, leading to a severely truncated nonfunctional protein (Sivakamasundari et al. 2000; Xie et al. 2003).
The NCRs have four main domains, a signal peptide, an extracellular domain, a transmembrane region, and an intracellular tail which does not contain any immunoreceptor tyrosine-based activation motifs (ITAM) (Lanier 1998, 2003; Moretta et al. 2000, 2001), required to transduce the activating signal. NKp46 is thought to be the main activating receptor expressed in both resting and activated NK cells (Sivori et al. 1999). It has two constant type 2 (C2-type Ig) domains in the extracellular region (Foster, Colonna, and Sun 2003), and recent work has shown that this region binds to particular viral glycoproteins (Mandelboim et al. 2001). Arginine, a positively charged amino acid present in the transmembrane domain of NKp46, is thought to stabilize its association with either CD3 delta or Fc epsilon R1 gamma, both of which have ITAMs present in their intracellular tails which are thought to transmit the activation signal (Pessino et al. 1998; Vitale et al. 1998). NKp44 (Cantoni et al. 1999) is selectively expressed on IL-2–activated NK cells and a subset of T-cell receptor-gamma/delta T cells. It has a single variable-type (V-type Ig) domain, and a lysine is present in its transmembrane domain which promotes association with the ITAM-bearing KARAP/DAP12 signal-transducing molecule (Cantoni et al. 1999). Finally, NKp30 has been described to be the major triggering receptor in the killing of certain tumors (Pende et al. 1999). It has a single Ig-type domain and an arginine present in its transmembrane domain which stabilizes its association with CD3 delta chains (Pende et al. 1999). The human NKp30 gene has six differentially spliced transcripts which differ in their Ig domains (V type or C type), and each of the extracellular domains can be linked to one of three different intracellular domains (Neville and Campbell 1999).
Ancient duplications and rearrangements of protein-coding segments have resulted in complex gene family relationships. Duplications can be tandem or dispersed and can involve entire coding regions or modules that correspond to folded protein domains. As a result, gene products may acquire new specificities, altered recognition properties, modified functions, and even loss of functionality (Henikoff et al. 1997). In this study we wanted to further investigate whether NKp30 is a pseudogene in different mouse strains and if any expressed NKp30 gene product could be detected. We also wanted to characterize the expression of the six different NKp30 splice forms in human tissues and perform computer-based analysis to study the presence of splice forms from all three NCR members in different species.
Materials and Methods
Single Nucleotide Polymorphism Characterization of Mouse Species
The genomic DNA from the mouse strains Mus m. domesticus (RBA/Dn), Mus m. CZECH II/Ei, Mus m. musculus C3H, Mus m. molossinus (MOLC/Rk), Mus m. castaneus (CASA/Rk), Mus hortulanus (PANCEBO), Mus spretus (spret/Ei), Mus caroli, and Mus pahari were obtained from The Jackson Laboratory (DNA Resource, Bar Harbour, Me.). DNA from M. m. musculus BALB/c and M. spretus (sega/Pas) were kindly donated by the Mouse Sequencing Group at the MRC Rosalind Franklin Centre for Genomics Research. Mouse tails for DNA extraction from M. m. musculus 129, C57BL/6, and BALB/c were kindly donated by Antonio Alcamí (Department of Medicine, Addenbrooke's Hospital, Cambridge, United Kingdom). The L929 cell line is derived from the connective tissue of M. m. musculus C3H.
Genomic DNA from three different mouse tails from M. m. musculus 129, C57BL/6, and BALB/c was prepared by the lysis method. Essentially, 500 μl of lysis buffer (100 mM TrisCl pH8.5, 5 mM ethylenediaminetetraacetic acid [EDTA], 0.2% sodium dodecyl sulfate [w/v], 200 mM NaCl) and 50 μg of proteinase K were added to 1.5 cm of mouse tail and rotated at 55°C overnight. Tubes were vortexed and spun at 14,000 rpm for 15 min, and the supernatant was removed. Isopropanol (350 μl) was added and tube were rotated to mix. DNA was spooled out, dipped into 70% ethanol, and then left to dry. DNA was left at 55°C for 2 h in 200 μl of TE (10 mM Tris pH 8.0, 1 mM EDTA) to dissolve and stored at –20°C. Genomic DNA from L929 cultured cells was prepared as described above, but using 107 cells.
The following primers (5'–3') were designed based on the published genomic DNA sequence from M. m. musculus 129 (accession number: AF109719) to cover all four predicted exons of NKp30. Primers for exon 1, forward (F)-ctatgatgcccaaagtccc and reverse (R)-ggagcatctaaattccacat were used for all strains except for M. m. musculus BALB/c and M. spretus, where (F)-tgaggtctgtccgcctcacc and (R)-cccactggagcatctaaat were used, and for M. pahari, where (F)-ctatgatgcccaaagtccc and (R)-ggagcatctaaattccacat were used. For exon 2, (F)-tggtcctgacagcgcttccc and (R)-acgtggagacggtggagagg were used for all strains except for M. m. musculus C3H, where (R)-cagccgagtcctgtctccc was used, for M. m. musculus BALB/c, where (F)-tgccatccgaggaaaaccg and (R)-atagatggtactgcccgtgg were used, and for M. pahari, where (F)-accgagtaagtgttctctgg and (R)-ccctgatcagcaactcacat were used. For exon 3, (F)-caagatgcagtgttactgg and (R)-agctttcctctcctcttgc were used for all strains. For exon 4, (F)-gcaagggagcagaaatggg and (R)-gatgatgtgaggaccaggg were used for all strains except for M. m. musculus C3H, M. molossinus, M. hortulanus, and M. caroli, where (R)-gccagtaaacagcttgtggg was used, and for M. pahari, where (F)-atgtgagttgctgatcaggg and (R)-ccctggtcctcacatcatc were used. Touchdown PCRs were performed using the following program: 94°C for 2 min, 94°C for 5 s, Tm + 5°C for 5 s, 72°C for 10 s (repeat nine times, each time decreasing Tm 0.5°C), then 94°C for 5 s, Tm for 5 s, 72°C for 10 s (repeat 24 times), then 72°C for 10 min, where the Tm of primer = (G + C x 4) + (A + T x 2) – 5. PCR products were directly sequenced or the bands extracted and cloned into pGEM. The Big Dye Terminator sequencing kit (ABI, Foster City, Calif.) was used, and the sequencing reactions were run on a 377 ABI Automatic fluorescent sequencer.
Tissue Messenger RNA Extraction and Complementary DNA Preparation
Mouse liver, kidney, and brain organs from M. m. musculus 129 were kindly donated by Antonio Alcamí (Department of Medicine, Addenbrooke's Hospital) and from M. caroli were obtained from The Jackson Laboratory (DNA Resource). Approximately 0.08 g of frozen tissue was homogenized and poly(A)+ RNA prepared using the Quick Prep Micro Messenger RNA (mRNA) purification kit (Amersham Pharmacia Biotech, Buckingham, UK) following the manufacturer's instructions. Total RNA from human tissues was obtained from Stratagene (La Jolla, Calif.). Complementary DNA (cDNA) was synthesized using the Reverse Transcription System (Promega, Madison, Wisc.) with oligo dT primers following the manufacturer's instructions. Mouse poly(A)+ RNA (1 μg) and human total RNA (1 μg) were used to generate 20 μl of cDNA on which nested PCR was performed using gene-specific primers.
Mouse Reverse Transcription–Polymerase Chain Reaction Analysis
The following primers were used for the nested PCR covering the predicted exons 1–4 of NKp30 in M. m. musculus 129 and M. caroli: for the first round (F)-tgaggtctgtccgcctcacc and (R)-gtaaggacttattgttggc and for the second round (F)-atggccaaggtgctcctgg and (R)-agaatcacttctcagaggc. The following NKp46 primers were used as reverse transcription–polymerase chain reaction (RT-PCR) control for both strains: first round (F)-tggccactggtatgctgcc and (R)-attctgggttgtgtgatcc and second round (F)-gctgccaacactcactgcc and (R)-atcccagaaggcggagtcc. The PCRs were performed using the touchdown program described above. In the first round of PCR, 1 μl of cDNA was used, and the products were cleaned with Qiagen (West Sussex, UK) PCR purification kit and eluted into 30 μl. In the second round of PCR, a 1:10 (v/v) final dilution of the first round product was used, and 5 μl of the final PCR product analyzed on an agarose gel. All RT-PCR products were sequenced as described above.
Human RT-PCR Analysis
The following primers were used for the nested PCR covering exons 2–4 of human NKp30. With the splice forms NKp30c and NKp30f, for the first round (F)-atcctctgccttcctgccc and (R)-gaggactagggacatctggg were used and for the second round (F)-ggaatggaaccccagagtt for NKp30c, (F)-ttcaatgccagccaaggg for NKp30f, and (R)-tctggaatcatccctcggg for both forms. With splice forms NKp30a, NKp30b, NKp30d, and NKp30e, for the first round (F)-atcctctgccttcctgccc and (R)-atgacagtgttcagggaccc were used and for the second round (F)-ttcaatgccagccaaggg for NKp30b and NKp30d and (F)-ggaatggaaccccagagtt for NKp30a and NKp30e. The (R)-agcagatgtgctgagctcc primer was used for all forms. The PCRs were performed using the same conditions and touchdown PCR program described above.
Computer Analysis
Expressed sequence tag (EST) databases were searched using the National Center for Biotechnology Information Web site http://ncbi.nlm.nih.gov/BLAST/, where the Blast searches, protein BlastX, translated TBlastN, and translated TBlastX were used. The multiple sequence alignments of NKp30 and NKp46 were obtained using ClustalX, and the phylogenetic trees were obtained using ClustalX and Phylo_win, both programs can be found at the Web site http://www.rfcgr.mrc.ac.uk.
Results
Single Nucleotide Polymorphism Analysis of Mouse NKp30 in Different Mouse Strains
The annotated genomic sequence of mouse NKp30 (accession number AF109719) comes from the mouse strain M. m. musculus 129. Translation of this nucleotide sequence shows two premature stop codons at the beginning of exon 2, the first one only 6 amino acids into the extracellular Ig–like domain (which is predicted to be 115 amino acids long) and the second one 35 amino acids downstream from the first one, suggesting that NKp30 is a pseudogene in that particular mouse strain. To investigate whether NKp30 contained the same premature stop codons in other mouse strains, the NKp30 gene was sequenced in the laboratory mouse strains M. m. musculus (inbred strains 129, C57BL/6, BALB/c, and C3H), M. m. CZECH II/Ei, M. m. domesticus, M. m. castaneus, and M. m. molossinus and in the wild mouse species M. spretus (spret and sega), M. hortulanus, M. caroli, and M. pahari. In addition, the public annotated genomic sequence derived from M. m. musculus 129 was confirmed by sequencing the NKp30 gene in the cosmid 188A7 used for the sequencing project (kindly provided by Lee Rowen, Institute for Systems Biology, Seattle, Wash.).
Detailed sequence analysis of NKp30 genomic DNA, from all these mouse strains (table 1), showed that there are exonic single nucleotide polymorphisms (SNPs), compared to M. m. musculus 129, in all the strains except in the other laboratory-inbred M. m. musculus strains (C57BL/6 and BALB/c) which are, for the exonic sequences of NKp30, genetically identical. However, M. m. musculus C3H showed several SNPs and, interestingly, the cell line L929, derived from this mouse strain, shows exactly the same SNPs indicating that this gene has not accumulated mutations during the process of cell line establishment (table 1). The rest of the M. musculus strains analyzed showed only a few SNPs (table 1).
Table 1 SNP Analysis of NKp30 in 13 Mouse Strains. Missing Bases Are Represented by m, Inserted Bases by , and SNPs that Obliterate the Stop Codons by *
With the wild strains, it is interesting to note that the NKp30 sequences from the two M. spretus strains (spret and sega) differ very much in their SNPs, the most striking difference being in exon 4 (table 1). M. hortulanus has SNPs predominantly in exon 4. M. caroli showed 22 SNPs in total, and very surprisingly, 2 out of the 13 located in exon 2 change the premature stop codons TGA to TGG, therefore providing an open reading frame for a predicted NKp30 protein (table 1 and fig. 2). M. pahari also showed SNPs which change the premature stop codons to coding amino acids, but due to a base insertion at position 72217, the frame is changed and an additional stop codon is introduced in the middle of the Ig domain, leading to a nonfunctional truncated NKp30 protein. This indicates that the only mouse strain that could encode a functional NKp30 protein would be M. caroli.
FIG. 2.— ClustalX multiple sequence alignment of NKp30 in human (Hsapiens-a to -f), monkey (Mmulatta-c and -v), bull (Btaurus), rat (Rnorvegicus, full length and soluble form -s), and mouse (Mcaroli, predicted -p and soluble -s). Signal peptide is in italics, arrows denote the splice sites, and the transmembrane domain is underlined.
Furthermore, some SNPs are maintained in all the different strains, such as 68723 A-T (silent mutation) in exon 1 present in all strains except M. caroli and M. pahari, 72372 A-C (nonconserved Lys-Asn change) in exon 2 present in all strains, and 72715 C-A (conserved Leu-Ile change) in exon 3 present in all the strains except M. m. CZECH II/Ei and M. m. domesticus. Furthermore, in exon 4, 72921 T-A (nonconserved Ile-Asp) and 72927 A-C (nonconserved Asp-Ala) are present in most of the wild strains and 72974 T-C occurs in the six strains which have SNPs in this exon, although it is in the untranslated region (UTR is from 72956 onward).
Transcriptional Analysis of Mouse NKp30
To analyze whether mouse NKp30 was expressed at the RNA level, databases were initially searched for the presence of mouse ESTs using the human NKp30 cDNA sequence or the theoretical mouse NKp30 cDNA sequence, with a BlastN algorithm. In addition, the NKp30 human protein sequence or theoretical mouse NKp30 protein sequence was used with a TBlast algorithm to search EST databases. In any case, no mouse EST clones were found (data not shown). The same type of analysis was performed to search the general nonredundant database, and again no alignments with mouse NKp30 cDNA or protein sequences were found (data not shown). However, NKp30 ESTs and proteins from other mammalian species such as Rattus norvegicus, Macaca mulatta, and Bos taurus were detected (see below).
In addition, in order to experimentally analyze whether any NKp30 mRNA could be transcribed in M. caroli, RT-PCR analysis was performed on cDNA derived from liver, kidney, and brain tissue mRNA from M. caroli and from the laboratory strain M. m. musculus 129 as a negative control. As a positive control for the RT-PCR, NKp46 primers that would amplify RNA from the NK receptor NKp46 (Pessino et al. 1998) were used. PCR products of the correct expected size (747 bp) for the NKp46 transcripts were detected in all the tissues from both strains (fig. 1). As expected, no NKp30 transcripts were found in any tissues from M. m. musculus 129. However, in M. caroli, although the expected PCR product of 571 bp was absent, two PCR products were found, one in brain of 699 bp and the other in kidney of 682 bp (fig. 1). The 699-bp product was shown to cover exons 1–4 but with one new exon of 128 bp present between exons 1 and 2 (70966–71093, mouse BAC AF109719 [GenBank] ). When translated, the frame of this product is changed, and a premature stop codon is present five amino acids into the Ig domain, encoding a potentially severely truncated nonfunctional protein. In kidney, the 682-bp product also covers exons 1–4. However, in this case, the small intron of 111 bp between exons 2 and 3 is retained and interestingly, when translated, codes for a putative NKp30 protein containing a potential signal peptide, a V-type Ig domain and a tail of 93 amino acids, but does not contain a transmembrane domain. This indicates that M. caroli could potentially express a soluble form of NKp30 in kidney (fig. 2).
FIG. 1.— RT-PCR analysis of NKp30 and NKp46 in laboratory strain Mus m. musculus 129 (laboratory strain 129) and Mus caroli. The following tissues were analyzed: liver (L), kidney (K), and brain (B) and negative control (–). Arrows represent the transcripts detected and their size. The sizes of DNA markers are indicated in base pairs.
To investigate whether any other potential soluble forms were present in other species, apart from mouse, a detailed EST and cDNA database search was performed. Only a potential soluble form in rat was found (Backman-Petersson et al. 2003), which is generated in a similar way, by the retention of intron 2 in the mRNA, though in rat the sequence after the Ig domain is much shorter (fig. 2), consisting of only 20 amino acids, and with no homology to the equivalent region in mouse.
Transcriptional Analysis of Human NKp30
In humans, NKp30 (NCR3/1C7) shows six different splice forms named NKp30a (1C7a), NKp30b (1C7b), NKp30c (1C7c), NKp30d (1C7d), NKp30e (1C7e), and NKp30f (1C7f) (Neville and Campbell 1999) (fig. 3). Three of the isoforms encode a potential V-type Ig domain and the other three for a potential C-type Ig domain (figs. 2 and 3). These two different extracellular domains can be linked to three different intracellular domains (of 36, 25, and 12 amino acids) depending on which exon 4 they utilize (Neville and Campbell 1999).
FIG. 3.— Schematic diagram showing the different splice forms of the human NCR family. Ig denotes the immunoglobulin domain (V the variable and C the constant Ig-type domains), "stalk" the region between the Ig-like domain and the transmembrane domain, Tran the transmembrane domain, and Intra the intracellular domain.
A detailed RT-PCR transcriptional analysis of the six forms in different adult and fetal human tissues was carried out. Although RT-PCR is not a quantitative technique, the results are collated from five different experiments all of which gave similar results. The most ubiquitous and highly expressed form was NKp30c, except in fetal brain where it was absent (table 2). However, in contrast, NKp30f (containing the same intracellular tail as NKp30c but with a C-type Ig domain) was expressed in fetal brain but was not expressed in fetal liver and kidney, being the form with lower expression levels. Interestingly, NKp30a and NKp30b (which contain a V-type Ig domain, as does NKp30c, but different intracellular domain) were highly expressed in brain (table 2). Generally, it was noticeable that the more highly expressed forms were those with a V-type Ig domain (NKp30c, NKp30b, and NKp30a). When comparing fetal and adult tissues, the expression levels of the different splice forms were similar except for those containing C-type Ig domains (NKp30e, NKp30f, and NKp30d).
Table 2 Expression of Human NKp30 Splice Forms in Different Fetal and Adult Tissues. High Expression Is Indicated by +++, Intermediate by ++, Low by +, and Lack of Detectable Expression by –
Database searches were performed to detect the presence of alternative NKp30 transcripts in other species. In M. mulatta NKp30 forms were found with a V Ig–type (AAK63117 [GenBank] and a C Ig–type (AAK63118 [GenBank] domain which would correspond, in relation to the intracellular domain, to human NKp30c and NKp30f, respectively (fig. 2). A NKp30S (AAK63119 [GenBank] form was detected in M. mulatta, but this could correspond to a partial cDNA rather than to a soluble form of NKp30. However, in B. taurus (CB428714 [GenBank] and CB428376 [GenBank] ) and R. norvegicus (AJ430418 [GenBank] ) only the V-type Ig domain form was detected (fig. 2). Apart from human, it was not possible to detect different intracellular forms in any other species (data not shown).
Evolution of NKp30
To investigate the evolution and relationships of the NKp30 forms in different species, multiple sequence alignment (fig. 2) and phylogenetic trees (figs. 4 and 5) were performed, of all the NKp30 sequences analyzed. The multiple alignments clearly showed that the protein sequences are very conserved, except in exon 4 (intracellular domain) where each species showed a different amino acid sequence. The only conservation seen in exon 4 is between the human NKp30c (and NKp30f) form and monkey and bull (although the last two sequences are shorter), but the similarity is only for the first 11 amino acids of the tail and then the sequences diverge completely (fig. 2).
FIG. 4.— Phylogenetic analysis of the NKp30 gene sequences from different mouse strains, rat, monkey, and human. The annotation used for the mouse strains is as follows: M (Mus), mus (Musculus), m (musculus), C57B (C57BL/6), Domest (domesticus), CZEC (CZECH II/Ei), 129 (129), BALB (BALB/c), Castan (castaneus), Moliss (molossinus), C3H (C3H), spretusse (spretus, sega), spretussp (spretus, spret), Hortulanu (hortulanus), Caroli (caroli), and Pahari (pahari). The annotation used for rat, monkey, and human is as follows: RNorvegicu (Rattus norvegicus), MMulatta (Macaca mulatta), and HSapiens (Homo sapiens).
FIG. 5.— Phylogenetic analysis of the NKp30 protein sequences from rat (RNorvegicus), mouse soluble (Mcaroli-s), bull (BTaurus), monkey (MMulatta), and human (HSapiens).
The phylogenetic trees showed that at the nucleotide level (fig. 4) the species are separated into two major clusters, one containing the mouse strains and rat and the other containing monkey and human. All the mouse M. m. musculus (except C3H) were quite close to M. m. domesticus, M. m. CZECH II/Ei, and M. m. castaneus. In a small cluster lies M. m. molossinus, M. m. musculus C3H, and M. spretus (sega), while the other M. spretus (spret) is on a different branch close to M. hortulanus. Mus caroli and M. pahari are, as expected, on two separate branches. The human forms are grouped to each other in relation to their intracellular domain. When the analysis was performed at the protein level (fig. 5), mouse and rat are in one cluster (with the soluble forms from each species together), while the other mammalian sequences lie in another cluster. Furthermore, in this case, the human splice forms are grouped in relation to their Ig domains and not in relation to their intracellular domains.
Alternative Splice Forms of NKp46 and NKp44
To determine whether the other NCR family members, NKp46 and NKp44, also expressed alternative splice forms, database searches were performed. Human NKp46 showed four different alternative splice forms, two containing two V-type Ig domains (NKp46a and NKp46b) (AJ001383 [GenBank] and AJ006121 [GenBank] , respectively) and the other two containing only one V-type Ig domain (NKp46c and NKp46d) (AJ006122 [GenBank] and AJ006123 [GenBank] , respectively) (fig. 3). With each different Ig combination, one form lacks 17 amino acids in the region between the Ig-like domain and the transmembrane domain ("stalk"), the function of which is currently unknown. The intracellular part of the four NKp46 forms are identical. NKp44 presents only three splice forms, one with an intracellular region of 61 amino acids (NKp44) and the other two with a different intracellular region of 43 amino acids (NKp44RG1 and NKp44RG2). The difference between the last two is in the "stalk," where NKp44RG2 contains an extra 12 amino acids, the function of which is currently unknown (Allcock et al. 2003).
To investigate whether the splice forms of NKp46 were expressed in other species, databases were searched in a similar way to NKp30. NKp46 was found to be expressed in M. mulatta (AY035218 [GenBank] ) and M. fascicularis (AJ278288 [GenBank] ), M. musculus (AJ223765 [GenBank] ), R. norvegicus (NM057199), and B. taurus (BI682532 [GenBank] ), and it was possible to detect ESTs that code for partial cDNAs form Equus caballus (e.g., CD467744 [GenBank] ) and Sus scrofa (e.g., BF441695 [GenBank] ). The percentage of identity at the amino acid level varied between 89% when comparing R. norvegicus and M. musculus and 59% when comparing M. musculus and Macaca. It was not possible to detect any different splice forms, except in M. mulatta (AY035222 [GenBank] ) and E. caballus (CD467202 [GenBank] ), which both have one form that lacks 12 amino acids in the signal peptide and are named NKp46SD. In addition, forms in B. taurus (BM106550 [GenBank] , BF652037 [GenBank] ) that have 38 amino acids absent before the transmembrane domain were found, which could correspond to the isoforms NKp46b or NKp46d (although in human this is 17 amino acids). Interestingly, the intracellular domains of NKp46 are (as in NKp30) not highly conserved between human and mouse and rat (fig. 6).
FIG. 6.— ClustalX multiple sequence alignment of NKp46 in human (four differentially spliced forms, HSaNKp46a–d), monkey (MMulatta and MFascicularis), rat (RNorvergicus), mouse (MMusculus), and bull (BTaurus). Italics denotes the signal peptide, bold the Ig domains, and underlining the transmembrane domain.
When NKp44 was analyzed, there were no ESTs or cDNAs in the databases in other species apart from human. NKp44 is located on human 6p21.1 in the TREM cluster (Allcock et al. 2003). The mouse TREM cluster is located on the orthologous chromosome 17, but the NKp44 gene is not present in this region. Detailed analysis of that cluster highlighted the presence of four ESTs (BB617333 [GenBank] , BY217551 [GenBank] , BY234948 [GenBank] , and BG148075 [GenBank] ) and two mRNAs found between TREM3 and Tlt2 (which are located between TREM1 and TREM2) with 40% identity to human NKp44. The potential protein products of these ESTs contain a signal peptide, an Ig-like domain, a transmembrane region (containing a K residue), and an intracellular region, features in common with NCR genes and the rest of the TREM genes. Furthermore, in rat, the EST XM236902 and the mRNA XM346014 also map between TREM2 and TREM1, and the protein translation product is 70% identical to the mouse ESTs and mRNAs mentioned above. Consequently, this would represent a new rodent TREM member (which we have named TREM6) and not the NKp44 mouse/rat homologue as the percentage of identity/similarity is quite low. However, it is interesting to note that it was not possible to find the human homologue of TREM6. This, together with the fact that there are no NKp44 ESTs in any other species apart from human, could indicate that NKp44 is TREM6 (or TREM6 is NKp44) and they could derive from a common ancestor and have diverged during adaptive evolution of the rodent and human immune systems.
Discussion
Protein families can be used to understand many aspects of genomes, both their "live" and their "dead" parts (i.e., genes and pseudogenes). Pseudogenes in prokaryotes represent families that are in the process of being dispensed with, but there appears to be less pressure to delete pseudogenes in eukaryotes (Harrison and Gerstein 2002). Ig domains have a high degree of pseudogenicity, and pseudogenicity in eukaryotes appears to be linked to protein function that is needed for an environmental response. NKp30, an Ig superfamily member, is expressed in human and other mammals but is a pseudogene in mouse (except in M. caroli), and this could have a relevance to understanding the biological function of NKp30. The number of SNPs which have accumulated in mouse NKp30 is not very high, considering that it is a pseudogene and that the mouse genome has higher nucleotide substitution, insertion, and deletion rates than human (Zhang, Carriero, and Gerstein 2004). From this detailed mouse genome analysis of NKp30, the exon containing the least SNPs is exon 1 (signal peptide), followed by exon 3 (transmembrane domain), and the majority of these changes are silent or conservative mutations. In exon 4 (intracellular tail), the SNPs are nonconserved or in UTRs. Finally, in exon 2 (Ig domain), more nonconserved SNP changes have been accumulated, and this is where the premature stop codons are. Consequently, it appears that more mutations have accumulated in the domains which are directly involved in the biological function of the protein (exons 2 and 4).
It has been described that pseudogenes tend to be associated with lineage-specific (as opposed to highly conserved) families that have environmental response functions. This may be because, rather than being "dead," they form a reservoir of diverse "extra parts" that can be resurrected to help an organism adapt to its surroundings (Harrison and Gerstein 2002). Pseudogenes can be divided into two varieties, duplicated and processed. The latter involves reverse transcription from an mRNA intermediate resulting in a pseudogene that typically does not contain introns and sometimes has a recognizable 3' poly-adenine tail. The NKp30 pseudogene does not correspond to the processed type. However, it could have derived from the duplicated category, formed by duplication of an ancestral NCR gene which gave rise to the three NCR members. The appearance of the various NCR members from NKp46 to NKp30 and to NKp44 during phylogenesis has also previously been suggested (De Maria et al. 2001).
NKp46, NKp44, and NKp30 may have quickly evolved from the ancestor by rapid sequence diversification (no sequence similarity among them). In humans, the three members have been maintained as active genes/proteins. However, in mouse, NKp46 is also an active gene, but NKp30 is a pseudogene and NKp44 is lost, although a related member TREM6 (whose NK receptor function should be investigated) is present. It is interesting to note that NKp30 can be expressed in M. caroli, but only as a potential soluble protein, which can also be detected in rat, but not in any other mammal. Similarly, NKp44 is only found in human, but TREM6 is found in mouse and rat. This is interesting from a functional point of view, considering that in humans NKp46 is widely expressed in NK cells and is the main activating NK receptor but NKp30 has a more restricted expression pattern followed by NKp44, indicating specialization of the NCR-NK response in humans in relation to the one in mouse. It should also be noted that other NK receptors are not maintained in human and mouse. Ly49L, for example, is a pseudogene in human, but there are 14 genes in mouse. Similarly, the KIR family is only found in human and not in mouse. Strangely enough, the mouse Ly49 C-type lectin family and the human KIR Ig superfamily receptors are structurally unrelated (lectin or Ig-type domains, respectively), but both fulfill a similar biological role as they are receptors specific for various MHC class I molecules and they use similar mechanisms for signal transduction (Mager et al. 2001; McVicar and Burshtyn 2001).
Alternative splicing is emerging as a major mechanism of functional regulation in the human genome, and it appears to be of central importance for neuronal genes, immune-related genes, and other genes involved in "information-processing" functions. The sequencing of mammalian genomes has demonstrated the importance of alternative splicing. In humans, of the 245 genes present on chromosome 22, 59% are alternatively spliced, and the 544 genes on chromosome 19 result in 1,859 different messages (Stamm 2002). The comparison of ESTs with the human genome sequence indicates that 47% of human genes might be alternatively spliced. This mechanism contributes to the surprising finding that the approximately 30,000 human genes create a much larger proteome, estimated to be >90,000 proteins (Stamm 2002).
The use of alternative exons can be controlled by cells, in a developmental, tissue-specific, or pathology-specific manner. Cells use known phosphorylation pathways as well as relocalization and synthesis of splicing factors to change their alternative splicing patterns (Ladd and Cooper 2002; Stamm 2002). The importance of alternative splicing is further illustrated by the increasing number of human diseases that have been attributed to mis-splicing events. For example, the insulin receptor plays a vital role in mediating the actions of insulin. There are two insulin receptor isoforms which have different affinities for insulin. Several cancer cell types preferentially express one isoform, and activation of that isoform results in mitogenic effects and a potentiation of the cancer phenotype (Denley et al. 2003).
The presence of alternative splice forms of the human NKp30 receptor is very interesting from the functional and evolutionary points of view. The V-type Ig splice forms of NKp30 are generally highly expressed in human tissues, the splice forms are present only in primates (human and monkey), the spliced intracellular forms are present only in humans, and the soluble forms are found only in rodents. In particular, it is unusual that one gene can code for Ig domains of the V type and C type. This could be very relevant as these two forms could possibly bind different ligands or the same ligand with different affinities. Even more relevant could be the presence of different intracellular forms if they transduce different signals. There is a selective cross talk among NCRs in human NK cells, and engagement and clustering of one or another NCR results in the activation of an identical set of tyrosine kinases (Augugliaro et al. 2003). So, why do NKp30 and NKp44 genes code for differentially spliced intracellular forms? Are they involved in different signaling pathways? It is interesting to note that the intracellular forms of each NCR also differ among the different species, in particular from the point of view of evolution and the specialization of function. The role of the NCR isoforms, derived by alternative splicing, and in particular of NKp30 should be further investigated to analyze their potential biological function.
Gene duplication followed by adaptive evolution is one of the primary forces for the emergence of new gene function. Here we present more evidence by using comparative genomics analysis on protein families, to indicate that the specialization of function in evolution can be due to the appearance of family members, with various splice forms for each member.
Acknowledgements
Special thanks go to Lee Rowen (Institute for Systems Biology, Seattle, Wash.) for the mouse genomic nucleotide sequence information (strain M. m. musculus 129). We thank several members of the Functional Genomics Group for general advice. This research was funded by the United Kingdom Medical Research Council. B.A. holds a Ramón y Cajal fellowship funded by the Spanish Ministerio de Educación y Ciencia and the Consejo Superior de Investigaciones Científicas since September 2003.
References
Allcock, R. J., A. D. Barrow, S. Forbes, S. Beck, and J. Trowsdale. 2003. The human TREM gene cluster at 6p21.1 encodes both activating and inhibitory single IgV domain receptors and includes NKp44. Eur. J. Immunol. 33:567–577.
Augugliaro, R., S. Parolini, R. Castriconi, E. Marcenaro, C. Cantoni, M. Nanni, L. Moretta, A. Moretta, and C. Bottino. 2003. Selective cross-talk among natural cytotoxicity receptors in human natural killer cells. Eur. J. Immunol. 33:1235–1241.
Backman-Petersson, E., J. R. Miller, M. Hollyoake, B. Aguado, and G. W. Butcher. 2003. Molecular characterization of the novel rat NK receptor 1C7. Eur. J. Immunol. 33:342–351.
Biassoni, R., C. Cantoni, M. Falco, D. Pende, R. Millo, L. Moretta, C. Bottino, and A. Moretta. 2000. Human natural killer cell activating receptors. Mol. Immunol. 37:1015–1024.
Biassoni, R., A. Pessino, C. Bottino, D. Pende, L. Moretta, and A. Moretta. 1999. The murine homologue of the human NKp46, a triggering receptor involved in the induction of natural cytotoxicity. Eur. J. Immunol. 29:1014–1020.
Borges, L., M. L. Hsu, N. Fanger, M. Kubin, and D. Cosman. 1997. A family of human lymphoid and myeloid Ig-like receptors, some of which bind to MHC class I molecules. J. Immunol. 159:5192–5196.
Brown, M. G., S. Fulmek, K. Matsumoto, R. Cho, P. A. Lyons, E. R. Levy, A. A. Scalzo, and W. M. Yokoyama. 1997. A 2-Mb YAC contig and physical map of the natural killer gene complex on mouse chromosome 6. Genomics 42:16–25.
Campbell, K. S., and M. Colonna. 2001. Human natural killer cell receptors and signal transduction. Int. Rev. Immunol. 20:333–370.
Cantoni, C., C. Bottino, M. Vitale, A. Pessino, R. Augugliaro, A. Malaspina, S. Parolini, L. Moretta, A. Moretta, and R. Biassoni. 1999. NKp44, a triggering receptor involved in tumor cell lysis by activated human natural killer cells, is a novel member of the immunoglobulin superfamily. J. Exp. Med. 189:787–796.
Chang, C., A. Rodriguez, M. Carretero, M. Lopez-Botet, J. H. Phillips, and L. L. Lanier. 1995. Molecular characterization of human CD94: a type II membrane glycoprotein related to the C-type lectin superfamily. Eur. J. Immunol. 25:2433–2437.
Colonna, M., H. Nakajima, F. Navarro, and M. Lopez-Botet. 1999. A novel family of Ig-like receptors for HLA class I molecules that modulate function of lymphoid and myeloid cells. J. Leukoc. Biol. 66:375–381.
De Maria, A., R. Biassoni, M. Fogli et al. (12 co-authors). 2001. Identification, molecular cloning and functional characterization of NKp46 and NKp30 natural cytotoxicity receptors in Macaca fascicularis NK cells. Eur. J. Immunol. 31:3546–3556.
Denley, A., J. C. Wallace, L. J. Cosgrove, and B. E. Forbes. 2003. The insulin receptor isoform exon 11- (IR-A) in cancer and other diseases: a review. Horm. Metab. Res. 35:778–785.
Foster, C. E., M. Colonna, and P. D. Sun. 2003. Crystal structure of the human natural killer (NK) cell activating receptor NKp46 reveals structural relationship to other leukocyte receptor complex immunoreceptors. J. Biol. Chem. 278:46081–46086.
Harrison, P. M., and M. Gerstein. 2002. Studying genomes through the aeons: protein families, pseudogenes and proteome evolution. J. Mol. Biol. 318:1155–1174.
Henikoff, S., E. A. Greene, S. Pietrokovski, P. Bork, T. K. Attwood, and L. Hood. 1997. Gene families: the taxonomy of protein paralogs and chimeras. Science 278:609–614.
Ho, E. L., J. W. Heusel, M. G. Brown, K. Matsumoto, A. A. Scalzo, and W. M. Yokoyama. 1998. Murine Nkg2d and Cd94 are clustered within the natural killer complex and are expressed independently in natural killer cells. Proc. Natl. Acad. Sci. USA 95:6320–6325.
Ladd, A. N., and T. A. Cooper. 2002. Finding signals that regulate alternative splicing in the post-genomic era. Genome Biol. 3:reviews0008.
Lanier, L. L. 1998. NK cell receptors. Annu. Rev. Immunol. 16:359–393.
———. 2003. Natural killer cell receptor signaling. Curr. Opin. Immunol. 15:308–314.
Lanier, L. L., C. Chang, and J. H. Phillips. 1994. Human NKR-P1A. A disulfide-linked homodimer of the C-type lectin superfamily expressed by a subset of NK and T lymphocytes. J. Immunol. 153:2417–2428.
Mager, D. L., K. L. McQueen, V. Wee, and J. D. Freeman. 2001. Evolution of natural killer cell receptors: coexistence of functional Ly49 and KIR genes in baboons. Curr. Biol. 11:626–630.
Mandelboim, O., N. Lieberman, M. Lev, L. Paul, T. I. Arnon, Y. Bushkin, D. M. Davis, J. L. Strominger, J. W. Yewdell, and A. Porgador. 2001. Recognition of haemagglutinins on virus-infected cells by NKp46 activates lysis by human NK cells. Nature 409:1055–1060.
McVicar, D. W., and D. N. Burshtyn. 2001. Intracellular signaling by the killer immunoglobulin-like receptors and Ly49. Sci. STKE 75:RE1.
Mighell, A. J., N. R. Smith, P. A. Robinson, and A. F. Markham. 2000. Vertebrate pseudogenes. FEBS Lett. 468:109–114.
Moretta, A., R. Biassoni, C. Bottino, M. C. Mingari, and L. Moretta. 2000. Natural cytotoxicity receptors that trigger human NK-cell-mediated cytolysis. Immunol. Today 21:228–234.
Moretta, A., C. Bottino, M. C. Mingari, R. Biassoni, and L. Moretta. 2002a. What is a natural killer cell? Nat. Immunol. 3:6–8.
Moretta, A., C. Bottino, M. Vitale, D. Pende, C. Cantoni, M. C. Mingari, R. Biassoni, and L. Moretta. 2001. Activating receptors and coreceptors involved in human natural killer cell-mediated cytolysis. Annu. Rev. Immunol. 19:197–223.
Moretta, L., C. Bottino, G. Ferlazzo, D. Pende, G. Melioli, M. C. Mingari, and A. Moretta. 2003. Surface receptors and functional interactions of human natural killer cells: from bench to the clinic. Cell. Mol. Life Sci. 60:2139–2146.
Moretta, L., C. Bottino, D. Pende, M. C. Mingari, R. Biassoni, and A. Moretta. 2002b. Human natural killer cells: their origin, receptors and function. Eur. J. Immunol. 32:1205–1211.
Moretta, L., and A. Moretta. 2003. Unravelling natural killer cell function: triggering and inhibitory human NK receptors. EMBO J. 18:18.
Neville, M. J., and R. D. Campbell. 1999. A new member of the Ig superfamily and a V-ATPase G subunit are among the predicted products of novel genes close to the TNF locus in the human MHC. J. Immunol. 162:4745–4754.
Pende, D., S. Parolini, A. Pessino et al. (13 co-authors). 1999. Identification and molecular characterization of NKp30, a novel triggering receptor involved in natural cytotoxicity mediated by human natural killer cells. J. Exp. Med. 190:1505–1516.
Pessino, A., S. Sivori, C. Bottino, A. Malaspina, L. Morelli, L. Moretta, R. Biassoni, and A. Moretta. 1998. Molecular cloning of NKp46: a novel member of the immunoglobulin superfamily involved in triggering of natural cytotoxicity. J. Exp. Med. 188:953–960.
Samaridis, J., and M. Colonna. 1997. Cloning of novel immunoglobulin superfamily receptors expressed on human myeloid and lymphoid cells: structural evidence for new stimulatory and inhibitory pathways. Eur. J. Immunol. 27:660–665.
Schnittger, S., J. Hamann, C. Dannenberg, H. Fiebig, M. Strauss, and C. Fonatsch. 1993. Regional sublocalization of the human CD69 gene to chromosome bands 12p12.3-p13.2, the predicted region of the human natural killer cell gene complex. Eur. J. Immunol. 23:2711–2713.
Sivakamasundari, R., A. Raghunathan, C. Y. Zhang, R. R. Chowdhury, and S. M. Weissman. 2000. Expression and cellular localization of the protein encoded by the 1C7 gene: a recently described component of the MHC. Immunogenetics 51:723–732.
Sivori, S., D. Pende, C. Bottino, E. Marcenaro, A. Pessino, R. Biassoni, L. Moretta, and A. Moretta. 1999. NKp46 is the major triggering receptor involved in the natural cytotoxicity of fresh or cultured human NK cells. Correlation between surface density of NKp46 and natural cytotoxicity against autologous, allogeneic or xenogeneic target cells. Eur. J. Immunol. 29:1656–1666.
Smith, H. R., A. H. Idris, and W. M. Yokoyama. 2001. Murine natural killer cell activation receptors. Immunol. Rev. 181:115–125.
Stamm, S. 2002. Signals and their transduction pathways regulating alternative splicing: a new dimension of the human genome. Hum. Mol. Genet. 11:2409–2416.
Takai, T., and M. Ono. 2001. Activating and inhibitory nature of the murine paired immunoglobulin-like receptor family. Immunol. Rev. 181:215–222.
Torkar, M., Z. Norgate, M. Colonna, J. Trowsdale, and M. J. Wilson. 1998. Isotypic variation of novel immunoglobulin-like transcript/killer cell inhibitory receptor loci in the leukocyte receptor complex. Eur. J. Immunol. 28:3959–3967.
Trowsdale, J., R. Barten, A. Haude, C. A. Stewart, S. Beck, and M. J. Wilson. 2001. The genomic context of natural killer receptor extended gene families. Immunol. Rev. 181:20–38.
Vitale, M., C. Bottino, S. Sivori, L. Sanseverino, R. Castriconi, E. Marcenaro, R. Augugliaro, L. Moretta, and A. Moretta. 1998. NKp44, a novel triggering surface molecule specifically expressed by activated natural killer cells, is involved in non-major histocompatibility complex-restricted tumor cell lysis. J. Exp. Med. 187:2065–2072.
Wagtmann, N., S. Rojo, E. Eichler, H. Mohrenweiser, and E. O. Long. 1997. A new human gene complex encoding the killer cell inhibitory receptors and related monocyte/macrophage receptors. Curr. Biol. 7:615–618.
Wende, H., A. Volz, and A. Ziegler. 2000. Extensive gene duplications and a large inversion characterize the human leukocyte receptor cluster. Immunogenetics 51:703–713.
Westgaard, I. H., S. F. Berg, S. Orstavik, S. Fossum, and E. Dissen. 1998. Identification of a human member of the Ly-49 multigene family. Eur. J. Immunol. 28:1839–1846.
Wilson, M. J., J. A. Lindquist, and J. Trowsdale. 2000. DAP12 and KAP10 (DAP10)-novel transmembrane adapter proteins of the CD3zeta family. Immunol. Res. 22:21–42.
Wu, J., H. Cherwinski, T. Spies, J. H. Phillips, and L. L. Lanier. 2000. DAP10 and DAP12 form distinct, but functionally cooperative, receptor complexes in natural killer cells. J. Exp. Med. 192:1059–1068.
Xie, T., L. Rowen, B. Aguado, M. E. Ahearn, A. Madan, S. Qin, R. D. Campbell, and L. Hood. 2003. Analysis of the gene-dense major histocompatibility complex class III region and its comparison to mouse. Genome Res. 13:2621–2636.
Yamanaka, S., X. Y. Zhang, K. Miura, S. Kim, and H. Iwao. 1998. The human gene encoding the lectin-type oxidized LDL receptor (OLR1) is a novel member of the natural killer gene complex with a unique expression profile. Genomics 54:191–199.
Zhang, Z., N. Carriero, and M. Gerstein. 2004. Comparative analysis of processed pseudogenes in the mouse and human genomes. Trends Genet. 20:62–67.(Martine Hollyoake*, R. Du)
E-mail: baguado@cnb.uam.es.
Abstract
Ancient duplications and rearrangements of protein-coding segments have resulted in complex gene family relationships. As a result, gene products may acquire new specificities, altered recognition properties, modified functions, and even loss of functionality. The natural cytotoxicity receptor (NCR) family are natural killer (NK)–activating receptors whose members are NKp46 (NCR1), NKp44 (NCR2), and NKp30 (NCR3). The NCR proteins are putative immunoglobulin superfamily members whose ligands are unknown. The NKp46 gene is present and expressed in human and mouse, NKp44 is only present and expressed in human, and NKp30 is present and expressed in human but is a nonexpressed pseudogene in mouse. By searching databases we have detected alternatively spliced forms of the three NCR members. In addition, we have shown by reverse transcription–polymerase chain reaction (RT-PCR) analysis that the human NKp30 gene presents differential expression patterns in tissues. However, no expressed sequence tags (ESTs) are detected for mouse NKp30, and the genomic sequence contains two premature stop codons, which would encode a severely truncated nonfunctional protein. We have sequenced genomic DNA from 13 mouse inbred and wild strains and discovered that NKp30 is a pseudogene in every mouse strain sequenced except Mus caroli where two single nucleotide polymorphisms (SNPs) abolished the premature stop codons. We observed that the laboratory-inbred strains are, for the exonic sequences, genetically identical, except Mus m. musculus C3H. The Mus musculus strains only have a few SNPs, but the rest of the Mus strains have accumulated gradually several SNPs, mainly in the functional immunoglobulin and intracellular domains. RT-PCR analysis performed on RNA from M. caroli tissue samples identified two transcripts, one of which would encode a putative soluble NKp30 protein, also detected in rat but not in human. We have observed that the intracellular domains of NKp30 (and NKp46) are not conserved among the different species, with the most striking difference when comparing human against mouse and rat. The NKp44 gene is only found in human and shows three different splice forms varying in their "stalk" and intracellular domains. Searching for NKp44 orthologs, we found similarity to ESTs from a novel rodent TREM family member, which we termed TREM6, and not to any possible NKp44 ortholog.
Key Words: NCR ? NKp30 ? splice forms ? pseudogene ? Mus caroli ? TREM6
Introduction
Pseudogenes are common and are encountered in a diverse range of life forms, but particularly in vertebrates. Genome complexity has evolved by the generation of gene families via gene duplication, and for reasons that remain contentious, some of these duplicated genes have become nonfunctional pseudogenes. Retrotransposed pseudogenes arise by a completely different mechanism and reflect a different aspect of genome evolution. Once established within a genome, pseudogenes evolve with time, although the mechanisms that control these changes remain very poorly understood (Mighell et al. 2000). Most pseudogenes have multiple features that confirm their nonfunctional status. However, there are genes that have many features of pseudogenes but are actually functional, and there are genes that are currently considered as pseudogenes that may potentially be functional. Accordingly, experimental design and interpretation across the whole field of molecular genetics must take pseudogenes into careful consideration (Mighell et al. 2000).
NKp30 (NCR3 or 1C7) is a natural killer (NK) cell receptor which is a functional gene in human (Pende et al. 1999) but a pseudogene in mouse (Xie et al. 2003). NK cells participate in the innate immune system of mammals. These NK cells lyse abnormal cells which contain downregulated cell surface major histocompatability complex (MHC) class I molecules. The NK-mediated cytotoxicity is regulated by signals from NK surface receptors, which are either activating, the ligands of which only a few are known, or inhibitory receptors, whose ligands are MHC class I molecules (Moretta et al. 2001, 2002a). The inhibitory receptor response is dominant over the activating receptor one and, consequently, only if a target cell lacks or has low expression of self–MHC class I proteins on its surface will the activating signals come into effect (Biassoni et al. 2000; Campbell and Colonna 2001; Moretta et al. 2002b, 2003; Lanier 2003; L. Moretta and A. Moretta 2003).
In humans and mice there are a number of distinct families of inhibitory and activating NK receptors, which reside in two main gene clusters on different chromosomes. On human 12p12-13 is the NK C-type lectin receptor gene cluster, with natural killer cell receptor P1 (Lanier, Chang, and Phillips 1994), CD69 (Schnittger et al. 1993), LOX1 (oxidized low-density lipoprotein receptor 1) (Yamanaka et al. 1998), CD94 (Chang et al. 1995), NKG2-D (Ho et al. 1998), and Ly49L (Westgaard et al. 1998) (which has a premature stop codon and is thought to be nonfunctional). In mouse, the orthologous chromosome 6 region has representatives of all of these NK receptors with the addition of 14 Ly49 family genes (Brown et al. 1997; Smith, Idris, and Yokoyama 2001). On human chromosome 19q13.42 is the leukocyte receptor cluster (LCR) (Wende, Volz, and Ziegler 2000), which has the following gene/families: killer inhibitory receptor (KIR) family (Wagtmann et al. 1997; Trowsdale et al. 2001), immunoglobulin (Ig)–like transcripts (Samaridis and Colonna 1997), leukocyte-associated inhibitory receptors (LAIRs) (Colonna et al. 1999), Fc alpha receptor (Borges et al. 1997; Torkar et al. 1998), killer activating protein (KAP10/DAP10) (Wilson, Lindquist, and Trowsdale 2000; Wu et al. 2000), killer activating receptor–associated protein (KARAP/DAP12) (Wilson, Lindquist, and Trowsdale 2000; Wu et al. 2000), and NKp46 (or NCR1) (Pessino et al. 1998; Biassoni et al. 1999). The orthologous mouse LCR region is on the syntenic chromosome 7 region and has a similar gene/family array except for the presence of paired Ig-like transcripts (Takai and Ono 2001) instead of LAIR and the lack of the Fc alpha receptor and, more noticeably, the KIR gene family homologues.
The human natural cytotoxicity receptor (NCR) family members NKp46 (NCR1) (Pessino et al. 1998; Biassoni et al. 1999), NKp44 (NCR2) (Vitale et al. 1998), and NKp30 (NCR3) (Pende et al. 1999) are located on human chromosome 6. The mouse counterpart of NKp46 is located on the syntenic region of mouse chromosome 7, but for NKp44, so far, no murine homologue has been identified anywhere in the mouse genome. NKp30 is located in the class III region of the human MHC (chromosome 6) (Neville and Campbell 1999), and the mouse homologue is on the syntenic chromosome 17 MHC region. However, the annotated NKp30 mouse genomic sequence clearly shows two premature stop codons, leading to a severely truncated nonfunctional protein (Sivakamasundari et al. 2000; Xie et al. 2003).
The NCRs have four main domains, a signal peptide, an extracellular domain, a transmembrane region, and an intracellular tail which does not contain any immunoreceptor tyrosine-based activation motifs (ITAM) (Lanier 1998, 2003; Moretta et al. 2000, 2001), required to transduce the activating signal. NKp46 is thought to be the main activating receptor expressed in both resting and activated NK cells (Sivori et al. 1999). It has two constant type 2 (C2-type Ig) domains in the extracellular region (Foster, Colonna, and Sun 2003), and recent work has shown that this region binds to particular viral glycoproteins (Mandelboim et al. 2001). Arginine, a positively charged amino acid present in the transmembrane domain of NKp46, is thought to stabilize its association with either CD3 delta or Fc epsilon R1 gamma, both of which have ITAMs present in their intracellular tails which are thought to transmit the activation signal (Pessino et al. 1998; Vitale et al. 1998). NKp44 (Cantoni et al. 1999) is selectively expressed on IL-2–activated NK cells and a subset of T-cell receptor-gamma/delta T cells. It has a single variable-type (V-type Ig) domain, and a lysine is present in its transmembrane domain which promotes association with the ITAM-bearing KARAP/DAP12 signal-transducing molecule (Cantoni et al. 1999). Finally, NKp30 has been described to be the major triggering receptor in the killing of certain tumors (Pende et al. 1999). It has a single Ig-type domain and an arginine present in its transmembrane domain which stabilizes its association with CD3 delta chains (Pende et al. 1999). The human NKp30 gene has six differentially spliced transcripts which differ in their Ig domains (V type or C type), and each of the extracellular domains can be linked to one of three different intracellular domains (Neville and Campbell 1999).
Ancient duplications and rearrangements of protein-coding segments have resulted in complex gene family relationships. Duplications can be tandem or dispersed and can involve entire coding regions or modules that correspond to folded protein domains. As a result, gene products may acquire new specificities, altered recognition properties, modified functions, and even loss of functionality (Henikoff et al. 1997). In this study we wanted to further investigate whether NKp30 is a pseudogene in different mouse strains and if any expressed NKp30 gene product could be detected. We also wanted to characterize the expression of the six different NKp30 splice forms in human tissues and perform computer-based analysis to study the presence of splice forms from all three NCR members in different species.
Materials and Methods
Single Nucleotide Polymorphism Characterization of Mouse Species
The genomic DNA from the mouse strains Mus m. domesticus (RBA/Dn), Mus m. CZECH II/Ei, Mus m. musculus C3H, Mus m. molossinus (MOLC/Rk), Mus m. castaneus (CASA/Rk), Mus hortulanus (PANCEBO), Mus spretus (spret/Ei), Mus caroli, and Mus pahari were obtained from The Jackson Laboratory (DNA Resource, Bar Harbour, Me.). DNA from M. m. musculus BALB/c and M. spretus (sega/Pas) were kindly donated by the Mouse Sequencing Group at the MRC Rosalind Franklin Centre for Genomics Research. Mouse tails for DNA extraction from M. m. musculus 129, C57BL/6, and BALB/c were kindly donated by Antonio Alcamí (Department of Medicine, Addenbrooke's Hospital, Cambridge, United Kingdom). The L929 cell line is derived from the connective tissue of M. m. musculus C3H.
Genomic DNA from three different mouse tails from M. m. musculus 129, C57BL/6, and BALB/c was prepared by the lysis method. Essentially, 500 μl of lysis buffer (100 mM TrisCl pH8.5, 5 mM ethylenediaminetetraacetic acid [EDTA], 0.2% sodium dodecyl sulfate [w/v], 200 mM NaCl) and 50 μg of proteinase K were added to 1.5 cm of mouse tail and rotated at 55°C overnight. Tubes were vortexed and spun at 14,000 rpm for 15 min, and the supernatant was removed. Isopropanol (350 μl) was added and tube were rotated to mix. DNA was spooled out, dipped into 70% ethanol, and then left to dry. DNA was left at 55°C for 2 h in 200 μl of TE (10 mM Tris pH 8.0, 1 mM EDTA) to dissolve and stored at –20°C. Genomic DNA from L929 cultured cells was prepared as described above, but using 107 cells.
The following primers (5'–3') were designed based on the published genomic DNA sequence from M. m. musculus 129 (accession number: AF109719) to cover all four predicted exons of NKp30. Primers for exon 1, forward (F)-ctatgatgcccaaagtccc and reverse (R)-ggagcatctaaattccacat were used for all strains except for M. m. musculus BALB/c and M. spretus, where (F)-tgaggtctgtccgcctcacc and (R)-cccactggagcatctaaat were used, and for M. pahari, where (F)-ctatgatgcccaaagtccc and (R)-ggagcatctaaattccacat were used. For exon 2, (F)-tggtcctgacagcgcttccc and (R)-acgtggagacggtggagagg were used for all strains except for M. m. musculus C3H, where (R)-cagccgagtcctgtctccc was used, for M. m. musculus BALB/c, where (F)-tgccatccgaggaaaaccg and (R)-atagatggtactgcccgtgg were used, and for M. pahari, where (F)-accgagtaagtgttctctgg and (R)-ccctgatcagcaactcacat were used. For exon 3, (F)-caagatgcagtgttactgg and (R)-agctttcctctcctcttgc were used for all strains. For exon 4, (F)-gcaagggagcagaaatggg and (R)-gatgatgtgaggaccaggg were used for all strains except for M. m. musculus C3H, M. molossinus, M. hortulanus, and M. caroli, where (R)-gccagtaaacagcttgtggg was used, and for M. pahari, where (F)-atgtgagttgctgatcaggg and (R)-ccctggtcctcacatcatc were used. Touchdown PCRs were performed using the following program: 94°C for 2 min, 94°C for 5 s, Tm + 5°C for 5 s, 72°C for 10 s (repeat nine times, each time decreasing Tm 0.5°C), then 94°C for 5 s, Tm for 5 s, 72°C for 10 s (repeat 24 times), then 72°C for 10 min, where the Tm of primer = (G + C x 4) + (A + T x 2) – 5. PCR products were directly sequenced or the bands extracted and cloned into pGEM. The Big Dye Terminator sequencing kit (ABI, Foster City, Calif.) was used, and the sequencing reactions were run on a 377 ABI Automatic fluorescent sequencer.
Tissue Messenger RNA Extraction and Complementary DNA Preparation
Mouse liver, kidney, and brain organs from M. m. musculus 129 were kindly donated by Antonio Alcamí (Department of Medicine, Addenbrooke's Hospital) and from M. caroli were obtained from The Jackson Laboratory (DNA Resource). Approximately 0.08 g of frozen tissue was homogenized and poly(A)+ RNA prepared using the Quick Prep Micro Messenger RNA (mRNA) purification kit (Amersham Pharmacia Biotech, Buckingham, UK) following the manufacturer's instructions. Total RNA from human tissues was obtained from Stratagene (La Jolla, Calif.). Complementary DNA (cDNA) was synthesized using the Reverse Transcription System (Promega, Madison, Wisc.) with oligo dT primers following the manufacturer's instructions. Mouse poly(A)+ RNA (1 μg) and human total RNA (1 μg) were used to generate 20 μl of cDNA on which nested PCR was performed using gene-specific primers.
Mouse Reverse Transcription–Polymerase Chain Reaction Analysis
The following primers were used for the nested PCR covering the predicted exons 1–4 of NKp30 in M. m. musculus 129 and M. caroli: for the first round (F)-tgaggtctgtccgcctcacc and (R)-gtaaggacttattgttggc and for the second round (F)-atggccaaggtgctcctgg and (R)-agaatcacttctcagaggc. The following NKp46 primers were used as reverse transcription–polymerase chain reaction (RT-PCR) control for both strains: first round (F)-tggccactggtatgctgcc and (R)-attctgggttgtgtgatcc and second round (F)-gctgccaacactcactgcc and (R)-atcccagaaggcggagtcc. The PCRs were performed using the touchdown program described above. In the first round of PCR, 1 μl of cDNA was used, and the products were cleaned with Qiagen (West Sussex, UK) PCR purification kit and eluted into 30 μl. In the second round of PCR, a 1:10 (v/v) final dilution of the first round product was used, and 5 μl of the final PCR product analyzed on an agarose gel. All RT-PCR products were sequenced as described above.
Human RT-PCR Analysis
The following primers were used for the nested PCR covering exons 2–4 of human NKp30. With the splice forms NKp30c and NKp30f, for the first round (F)-atcctctgccttcctgccc and (R)-gaggactagggacatctggg were used and for the second round (F)-ggaatggaaccccagagtt for NKp30c, (F)-ttcaatgccagccaaggg for NKp30f, and (R)-tctggaatcatccctcggg for both forms. With splice forms NKp30a, NKp30b, NKp30d, and NKp30e, for the first round (F)-atcctctgccttcctgccc and (R)-atgacagtgttcagggaccc were used and for the second round (F)-ttcaatgccagccaaggg for NKp30b and NKp30d and (F)-ggaatggaaccccagagtt for NKp30a and NKp30e. The (R)-agcagatgtgctgagctcc primer was used for all forms. The PCRs were performed using the same conditions and touchdown PCR program described above.
Computer Analysis
Expressed sequence tag (EST) databases were searched using the National Center for Biotechnology Information Web site http://ncbi.nlm.nih.gov/BLAST/, where the Blast searches, protein BlastX, translated TBlastN, and translated TBlastX were used. The multiple sequence alignments of NKp30 and NKp46 were obtained using ClustalX, and the phylogenetic trees were obtained using ClustalX and Phylo_win, both programs can be found at the Web site http://www.rfcgr.mrc.ac.uk.
Results
Single Nucleotide Polymorphism Analysis of Mouse NKp30 in Different Mouse Strains
The annotated genomic sequence of mouse NKp30 (accession number AF109719) comes from the mouse strain M. m. musculus 129. Translation of this nucleotide sequence shows two premature stop codons at the beginning of exon 2, the first one only 6 amino acids into the extracellular Ig–like domain (which is predicted to be 115 amino acids long) and the second one 35 amino acids downstream from the first one, suggesting that NKp30 is a pseudogene in that particular mouse strain. To investigate whether NKp30 contained the same premature stop codons in other mouse strains, the NKp30 gene was sequenced in the laboratory mouse strains M. m. musculus (inbred strains 129, C57BL/6, BALB/c, and C3H), M. m. CZECH II/Ei, M. m. domesticus, M. m. castaneus, and M. m. molossinus and in the wild mouse species M. spretus (spret and sega), M. hortulanus, M. caroli, and M. pahari. In addition, the public annotated genomic sequence derived from M. m. musculus 129 was confirmed by sequencing the NKp30 gene in the cosmid 188A7 used for the sequencing project (kindly provided by Lee Rowen, Institute for Systems Biology, Seattle, Wash.).
Detailed sequence analysis of NKp30 genomic DNA, from all these mouse strains (table 1), showed that there are exonic single nucleotide polymorphisms (SNPs), compared to M. m. musculus 129, in all the strains except in the other laboratory-inbred M. m. musculus strains (C57BL/6 and BALB/c) which are, for the exonic sequences of NKp30, genetically identical. However, M. m. musculus C3H showed several SNPs and, interestingly, the cell line L929, derived from this mouse strain, shows exactly the same SNPs indicating that this gene has not accumulated mutations during the process of cell line establishment (table 1). The rest of the M. musculus strains analyzed showed only a few SNPs (table 1).
Table 1 SNP Analysis of NKp30 in 13 Mouse Strains. Missing Bases Are Represented by m, Inserted Bases by , and SNPs that Obliterate the Stop Codons by *
With the wild strains, it is interesting to note that the NKp30 sequences from the two M. spretus strains (spret and sega) differ very much in their SNPs, the most striking difference being in exon 4 (table 1). M. hortulanus has SNPs predominantly in exon 4. M. caroli showed 22 SNPs in total, and very surprisingly, 2 out of the 13 located in exon 2 change the premature stop codons TGA to TGG, therefore providing an open reading frame for a predicted NKp30 protein (table 1 and fig. 2). M. pahari also showed SNPs which change the premature stop codons to coding amino acids, but due to a base insertion at position 72217, the frame is changed and an additional stop codon is introduced in the middle of the Ig domain, leading to a nonfunctional truncated NKp30 protein. This indicates that the only mouse strain that could encode a functional NKp30 protein would be M. caroli.
FIG. 2.— ClustalX multiple sequence alignment of NKp30 in human (Hsapiens-a to -f), monkey (Mmulatta-c and -v), bull (Btaurus), rat (Rnorvegicus, full length and soluble form -s), and mouse (Mcaroli, predicted -p and soluble -s). Signal peptide is in italics, arrows denote the splice sites, and the transmembrane domain is underlined.
Furthermore, some SNPs are maintained in all the different strains, such as 68723 A-T (silent mutation) in exon 1 present in all strains except M. caroli and M. pahari, 72372 A-C (nonconserved Lys-Asn change) in exon 2 present in all strains, and 72715 C-A (conserved Leu-Ile change) in exon 3 present in all the strains except M. m. CZECH II/Ei and M. m. domesticus. Furthermore, in exon 4, 72921 T-A (nonconserved Ile-Asp) and 72927 A-C (nonconserved Asp-Ala) are present in most of the wild strains and 72974 T-C occurs in the six strains which have SNPs in this exon, although it is in the untranslated region (UTR is from 72956 onward).
Transcriptional Analysis of Mouse NKp30
To analyze whether mouse NKp30 was expressed at the RNA level, databases were initially searched for the presence of mouse ESTs using the human NKp30 cDNA sequence or the theoretical mouse NKp30 cDNA sequence, with a BlastN algorithm. In addition, the NKp30 human protein sequence or theoretical mouse NKp30 protein sequence was used with a TBlast algorithm to search EST databases. In any case, no mouse EST clones were found (data not shown). The same type of analysis was performed to search the general nonredundant database, and again no alignments with mouse NKp30 cDNA or protein sequences were found (data not shown). However, NKp30 ESTs and proteins from other mammalian species such as Rattus norvegicus, Macaca mulatta, and Bos taurus were detected (see below).
In addition, in order to experimentally analyze whether any NKp30 mRNA could be transcribed in M. caroli, RT-PCR analysis was performed on cDNA derived from liver, kidney, and brain tissue mRNA from M. caroli and from the laboratory strain M. m. musculus 129 as a negative control. As a positive control for the RT-PCR, NKp46 primers that would amplify RNA from the NK receptor NKp46 (Pessino et al. 1998) were used. PCR products of the correct expected size (747 bp) for the NKp46 transcripts were detected in all the tissues from both strains (fig. 1). As expected, no NKp30 transcripts were found in any tissues from M. m. musculus 129. However, in M. caroli, although the expected PCR product of 571 bp was absent, two PCR products were found, one in brain of 699 bp and the other in kidney of 682 bp (fig. 1). The 699-bp product was shown to cover exons 1–4 but with one new exon of 128 bp present between exons 1 and 2 (70966–71093, mouse BAC AF109719 [GenBank] ). When translated, the frame of this product is changed, and a premature stop codon is present five amino acids into the Ig domain, encoding a potentially severely truncated nonfunctional protein. In kidney, the 682-bp product also covers exons 1–4. However, in this case, the small intron of 111 bp between exons 2 and 3 is retained and interestingly, when translated, codes for a putative NKp30 protein containing a potential signal peptide, a V-type Ig domain and a tail of 93 amino acids, but does not contain a transmembrane domain. This indicates that M. caroli could potentially express a soluble form of NKp30 in kidney (fig. 2).
FIG. 1.— RT-PCR analysis of NKp30 and NKp46 in laboratory strain Mus m. musculus 129 (laboratory strain 129) and Mus caroli. The following tissues were analyzed: liver (L), kidney (K), and brain (B) and negative control (–). Arrows represent the transcripts detected and their size. The sizes of DNA markers are indicated in base pairs.
To investigate whether any other potential soluble forms were present in other species, apart from mouse, a detailed EST and cDNA database search was performed. Only a potential soluble form in rat was found (Backman-Petersson et al. 2003), which is generated in a similar way, by the retention of intron 2 in the mRNA, though in rat the sequence after the Ig domain is much shorter (fig. 2), consisting of only 20 amino acids, and with no homology to the equivalent region in mouse.
Transcriptional Analysis of Human NKp30
In humans, NKp30 (NCR3/1C7) shows six different splice forms named NKp30a (1C7a), NKp30b (1C7b), NKp30c (1C7c), NKp30d (1C7d), NKp30e (1C7e), and NKp30f (1C7f) (Neville and Campbell 1999) (fig. 3). Three of the isoforms encode a potential V-type Ig domain and the other three for a potential C-type Ig domain (figs. 2 and 3). These two different extracellular domains can be linked to three different intracellular domains (of 36, 25, and 12 amino acids) depending on which exon 4 they utilize (Neville and Campbell 1999).
FIG. 3.— Schematic diagram showing the different splice forms of the human NCR family. Ig denotes the immunoglobulin domain (V the variable and C the constant Ig-type domains), "stalk" the region between the Ig-like domain and the transmembrane domain, Tran the transmembrane domain, and Intra the intracellular domain.
A detailed RT-PCR transcriptional analysis of the six forms in different adult and fetal human tissues was carried out. Although RT-PCR is not a quantitative technique, the results are collated from five different experiments all of which gave similar results. The most ubiquitous and highly expressed form was NKp30c, except in fetal brain where it was absent (table 2). However, in contrast, NKp30f (containing the same intracellular tail as NKp30c but with a C-type Ig domain) was expressed in fetal brain but was not expressed in fetal liver and kidney, being the form with lower expression levels. Interestingly, NKp30a and NKp30b (which contain a V-type Ig domain, as does NKp30c, but different intracellular domain) were highly expressed in brain (table 2). Generally, it was noticeable that the more highly expressed forms were those with a V-type Ig domain (NKp30c, NKp30b, and NKp30a). When comparing fetal and adult tissues, the expression levels of the different splice forms were similar except for those containing C-type Ig domains (NKp30e, NKp30f, and NKp30d).
Table 2 Expression of Human NKp30 Splice Forms in Different Fetal and Adult Tissues. High Expression Is Indicated by +++, Intermediate by ++, Low by +, and Lack of Detectable Expression by –
Database searches were performed to detect the presence of alternative NKp30 transcripts in other species. In M. mulatta NKp30 forms were found with a V Ig–type (AAK63117 [GenBank] and a C Ig–type (AAK63118 [GenBank] domain which would correspond, in relation to the intracellular domain, to human NKp30c and NKp30f, respectively (fig. 2). A NKp30S (AAK63119 [GenBank] form was detected in M. mulatta, but this could correspond to a partial cDNA rather than to a soluble form of NKp30. However, in B. taurus (CB428714 [GenBank] and CB428376 [GenBank] ) and R. norvegicus (AJ430418 [GenBank] ) only the V-type Ig domain form was detected (fig. 2). Apart from human, it was not possible to detect different intracellular forms in any other species (data not shown).
Evolution of NKp30
To investigate the evolution and relationships of the NKp30 forms in different species, multiple sequence alignment (fig. 2) and phylogenetic trees (figs. 4 and 5) were performed, of all the NKp30 sequences analyzed. The multiple alignments clearly showed that the protein sequences are very conserved, except in exon 4 (intracellular domain) where each species showed a different amino acid sequence. The only conservation seen in exon 4 is between the human NKp30c (and NKp30f) form and monkey and bull (although the last two sequences are shorter), but the similarity is only for the first 11 amino acids of the tail and then the sequences diverge completely (fig. 2).
FIG. 4.— Phylogenetic analysis of the NKp30 gene sequences from different mouse strains, rat, monkey, and human. The annotation used for the mouse strains is as follows: M (Mus), mus (Musculus), m (musculus), C57B (C57BL/6), Domest (domesticus), CZEC (CZECH II/Ei), 129 (129), BALB (BALB/c), Castan (castaneus), Moliss (molossinus), C3H (C3H), spretusse (spretus, sega), spretussp (spretus, spret), Hortulanu (hortulanus), Caroli (caroli), and Pahari (pahari). The annotation used for rat, monkey, and human is as follows: RNorvegicu (Rattus norvegicus), MMulatta (Macaca mulatta), and HSapiens (Homo sapiens).
FIG. 5.— Phylogenetic analysis of the NKp30 protein sequences from rat (RNorvegicus), mouse soluble (Mcaroli-s), bull (BTaurus), monkey (MMulatta), and human (HSapiens).
The phylogenetic trees showed that at the nucleotide level (fig. 4) the species are separated into two major clusters, one containing the mouse strains and rat and the other containing monkey and human. All the mouse M. m. musculus (except C3H) were quite close to M. m. domesticus, M. m. CZECH II/Ei, and M. m. castaneus. In a small cluster lies M. m. molossinus, M. m. musculus C3H, and M. spretus (sega), while the other M. spretus (spret) is on a different branch close to M. hortulanus. Mus caroli and M. pahari are, as expected, on two separate branches. The human forms are grouped to each other in relation to their intracellular domain. When the analysis was performed at the protein level (fig. 5), mouse and rat are in one cluster (with the soluble forms from each species together), while the other mammalian sequences lie in another cluster. Furthermore, in this case, the human splice forms are grouped in relation to their Ig domains and not in relation to their intracellular domains.
Alternative Splice Forms of NKp46 and NKp44
To determine whether the other NCR family members, NKp46 and NKp44, also expressed alternative splice forms, database searches were performed. Human NKp46 showed four different alternative splice forms, two containing two V-type Ig domains (NKp46a and NKp46b) (AJ001383 [GenBank] and AJ006121 [GenBank] , respectively) and the other two containing only one V-type Ig domain (NKp46c and NKp46d) (AJ006122 [GenBank] and AJ006123 [GenBank] , respectively) (fig. 3). With each different Ig combination, one form lacks 17 amino acids in the region between the Ig-like domain and the transmembrane domain ("stalk"), the function of which is currently unknown. The intracellular part of the four NKp46 forms are identical. NKp44 presents only three splice forms, one with an intracellular region of 61 amino acids (NKp44) and the other two with a different intracellular region of 43 amino acids (NKp44RG1 and NKp44RG2). The difference between the last two is in the "stalk," where NKp44RG2 contains an extra 12 amino acids, the function of which is currently unknown (Allcock et al. 2003).
To investigate whether the splice forms of NKp46 were expressed in other species, databases were searched in a similar way to NKp30. NKp46 was found to be expressed in M. mulatta (AY035218 [GenBank] ) and M. fascicularis (AJ278288 [GenBank] ), M. musculus (AJ223765 [GenBank] ), R. norvegicus (NM057199), and B. taurus (BI682532 [GenBank] ), and it was possible to detect ESTs that code for partial cDNAs form Equus caballus (e.g., CD467744 [GenBank] ) and Sus scrofa (e.g., BF441695 [GenBank] ). The percentage of identity at the amino acid level varied between 89% when comparing R. norvegicus and M. musculus and 59% when comparing M. musculus and Macaca. It was not possible to detect any different splice forms, except in M. mulatta (AY035222 [GenBank] ) and E. caballus (CD467202 [GenBank] ), which both have one form that lacks 12 amino acids in the signal peptide and are named NKp46SD. In addition, forms in B. taurus (BM106550 [GenBank] , BF652037 [GenBank] ) that have 38 amino acids absent before the transmembrane domain were found, which could correspond to the isoforms NKp46b or NKp46d (although in human this is 17 amino acids). Interestingly, the intracellular domains of NKp46 are (as in NKp30) not highly conserved between human and mouse and rat (fig. 6).
FIG. 6.— ClustalX multiple sequence alignment of NKp46 in human (four differentially spliced forms, HSaNKp46a–d), monkey (MMulatta and MFascicularis), rat (RNorvergicus), mouse (MMusculus), and bull (BTaurus). Italics denotes the signal peptide, bold the Ig domains, and underlining the transmembrane domain.
When NKp44 was analyzed, there were no ESTs or cDNAs in the databases in other species apart from human. NKp44 is located on human 6p21.1 in the TREM cluster (Allcock et al. 2003). The mouse TREM cluster is located on the orthologous chromosome 17, but the NKp44 gene is not present in this region. Detailed analysis of that cluster highlighted the presence of four ESTs (BB617333 [GenBank] , BY217551 [GenBank] , BY234948 [GenBank] , and BG148075 [GenBank] ) and two mRNAs found between TREM3 and Tlt2 (which are located between TREM1 and TREM2) with 40% identity to human NKp44. The potential protein products of these ESTs contain a signal peptide, an Ig-like domain, a transmembrane region (containing a K residue), and an intracellular region, features in common with NCR genes and the rest of the TREM genes. Furthermore, in rat, the EST XM236902 and the mRNA XM346014 also map between TREM2 and TREM1, and the protein translation product is 70% identical to the mouse ESTs and mRNAs mentioned above. Consequently, this would represent a new rodent TREM member (which we have named TREM6) and not the NKp44 mouse/rat homologue as the percentage of identity/similarity is quite low. However, it is interesting to note that it was not possible to find the human homologue of TREM6. This, together with the fact that there are no NKp44 ESTs in any other species apart from human, could indicate that NKp44 is TREM6 (or TREM6 is NKp44) and they could derive from a common ancestor and have diverged during adaptive evolution of the rodent and human immune systems.
Discussion
Protein families can be used to understand many aspects of genomes, both their "live" and their "dead" parts (i.e., genes and pseudogenes). Pseudogenes in prokaryotes represent families that are in the process of being dispensed with, but there appears to be less pressure to delete pseudogenes in eukaryotes (Harrison and Gerstein 2002). Ig domains have a high degree of pseudogenicity, and pseudogenicity in eukaryotes appears to be linked to protein function that is needed for an environmental response. NKp30, an Ig superfamily member, is expressed in human and other mammals but is a pseudogene in mouse (except in M. caroli), and this could have a relevance to understanding the biological function of NKp30. The number of SNPs which have accumulated in mouse NKp30 is not very high, considering that it is a pseudogene and that the mouse genome has higher nucleotide substitution, insertion, and deletion rates than human (Zhang, Carriero, and Gerstein 2004). From this detailed mouse genome analysis of NKp30, the exon containing the least SNPs is exon 1 (signal peptide), followed by exon 3 (transmembrane domain), and the majority of these changes are silent or conservative mutations. In exon 4 (intracellular tail), the SNPs are nonconserved or in UTRs. Finally, in exon 2 (Ig domain), more nonconserved SNP changes have been accumulated, and this is where the premature stop codons are. Consequently, it appears that more mutations have accumulated in the domains which are directly involved in the biological function of the protein (exons 2 and 4).
It has been described that pseudogenes tend to be associated with lineage-specific (as opposed to highly conserved) families that have environmental response functions. This may be because, rather than being "dead," they form a reservoir of diverse "extra parts" that can be resurrected to help an organism adapt to its surroundings (Harrison and Gerstein 2002). Pseudogenes can be divided into two varieties, duplicated and processed. The latter involves reverse transcription from an mRNA intermediate resulting in a pseudogene that typically does not contain introns and sometimes has a recognizable 3' poly-adenine tail. The NKp30 pseudogene does not correspond to the processed type. However, it could have derived from the duplicated category, formed by duplication of an ancestral NCR gene which gave rise to the three NCR members. The appearance of the various NCR members from NKp46 to NKp30 and to NKp44 during phylogenesis has also previously been suggested (De Maria et al. 2001).
NKp46, NKp44, and NKp30 may have quickly evolved from the ancestor by rapid sequence diversification (no sequence similarity among them). In humans, the three members have been maintained as active genes/proteins. However, in mouse, NKp46 is also an active gene, but NKp30 is a pseudogene and NKp44 is lost, although a related member TREM6 (whose NK receptor function should be investigated) is present. It is interesting to note that NKp30 can be expressed in M. caroli, but only as a potential soluble protein, which can also be detected in rat, but not in any other mammal. Similarly, NKp44 is only found in human, but TREM6 is found in mouse and rat. This is interesting from a functional point of view, considering that in humans NKp46 is widely expressed in NK cells and is the main activating NK receptor but NKp30 has a more restricted expression pattern followed by NKp44, indicating specialization of the NCR-NK response in humans in relation to the one in mouse. It should also be noted that other NK receptors are not maintained in human and mouse. Ly49L, for example, is a pseudogene in human, but there are 14 genes in mouse. Similarly, the KIR family is only found in human and not in mouse. Strangely enough, the mouse Ly49 C-type lectin family and the human KIR Ig superfamily receptors are structurally unrelated (lectin or Ig-type domains, respectively), but both fulfill a similar biological role as they are receptors specific for various MHC class I molecules and they use similar mechanisms for signal transduction (Mager et al. 2001; McVicar and Burshtyn 2001).
Alternative splicing is emerging as a major mechanism of functional regulation in the human genome, and it appears to be of central importance for neuronal genes, immune-related genes, and other genes involved in "information-processing" functions. The sequencing of mammalian genomes has demonstrated the importance of alternative splicing. In humans, of the 245 genes present on chromosome 22, 59% are alternatively spliced, and the 544 genes on chromosome 19 result in 1,859 different messages (Stamm 2002). The comparison of ESTs with the human genome sequence indicates that 47% of human genes might be alternatively spliced. This mechanism contributes to the surprising finding that the approximately 30,000 human genes create a much larger proteome, estimated to be >90,000 proteins (Stamm 2002).
The use of alternative exons can be controlled by cells, in a developmental, tissue-specific, or pathology-specific manner. Cells use known phosphorylation pathways as well as relocalization and synthesis of splicing factors to change their alternative splicing patterns (Ladd and Cooper 2002; Stamm 2002). The importance of alternative splicing is further illustrated by the increasing number of human diseases that have been attributed to mis-splicing events. For example, the insulin receptor plays a vital role in mediating the actions of insulin. There are two insulin receptor isoforms which have different affinities for insulin. Several cancer cell types preferentially express one isoform, and activation of that isoform results in mitogenic effects and a potentiation of the cancer phenotype (Denley et al. 2003).
The presence of alternative splice forms of the human NKp30 receptor is very interesting from the functional and evolutionary points of view. The V-type Ig splice forms of NKp30 are generally highly expressed in human tissues, the splice forms are present only in primates (human and monkey), the spliced intracellular forms are present only in humans, and the soluble forms are found only in rodents. In particular, it is unusual that one gene can code for Ig domains of the V type and C type. This could be very relevant as these two forms could possibly bind different ligands or the same ligand with different affinities. Even more relevant could be the presence of different intracellular forms if they transduce different signals. There is a selective cross talk among NCRs in human NK cells, and engagement and clustering of one or another NCR results in the activation of an identical set of tyrosine kinases (Augugliaro et al. 2003). So, why do NKp30 and NKp44 genes code for differentially spliced intracellular forms? Are they involved in different signaling pathways? It is interesting to note that the intracellular forms of each NCR also differ among the different species, in particular from the point of view of evolution and the specialization of function. The role of the NCR isoforms, derived by alternative splicing, and in particular of NKp30 should be further investigated to analyze their potential biological function.
Gene duplication followed by adaptive evolution is one of the primary forces for the emergence of new gene function. Here we present more evidence by using comparative genomics analysis on protein families, to indicate that the specialization of function in evolution can be due to the appearance of family members, with various splice forms for each member.
Acknowledgements
Special thanks go to Lee Rowen (Institute for Systems Biology, Seattle, Wash.) for the mouse genomic nucleotide sequence information (strain M. m. musculus 129). We thank several members of the Functional Genomics Group for general advice. This research was funded by the United Kingdom Medical Research Council. B.A. holds a Ramón y Cajal fellowship funded by the Spanish Ministerio de Educación y Ciencia and the Consejo Superior de Investigaciones Científicas since September 2003.
References
Allcock, R. J., A. D. Barrow, S. Forbes, S. Beck, and J. Trowsdale. 2003. The human TREM gene cluster at 6p21.1 encodes both activating and inhibitory single IgV domain receptors and includes NKp44. Eur. J. Immunol. 33:567–577.
Augugliaro, R., S. Parolini, R. Castriconi, E. Marcenaro, C. Cantoni, M. Nanni, L. Moretta, A. Moretta, and C. Bottino. 2003. Selective cross-talk among natural cytotoxicity receptors in human natural killer cells. Eur. J. Immunol. 33:1235–1241.
Backman-Petersson, E., J. R. Miller, M. Hollyoake, B. Aguado, and G. W. Butcher. 2003. Molecular characterization of the novel rat NK receptor 1C7. Eur. J. Immunol. 33:342–351.
Biassoni, R., C. Cantoni, M. Falco, D. Pende, R. Millo, L. Moretta, C. Bottino, and A. Moretta. 2000. Human natural killer cell activating receptors. Mol. Immunol. 37:1015–1024.
Biassoni, R., A. Pessino, C. Bottino, D. Pende, L. Moretta, and A. Moretta. 1999. The murine homologue of the human NKp46, a triggering receptor involved in the induction of natural cytotoxicity. Eur. J. Immunol. 29:1014–1020.
Borges, L., M. L. Hsu, N. Fanger, M. Kubin, and D. Cosman. 1997. A family of human lymphoid and myeloid Ig-like receptors, some of which bind to MHC class I molecules. J. Immunol. 159:5192–5196.
Brown, M. G., S. Fulmek, K. Matsumoto, R. Cho, P. A. Lyons, E. R. Levy, A. A. Scalzo, and W. M. Yokoyama. 1997. A 2-Mb YAC contig and physical map of the natural killer gene complex on mouse chromosome 6. Genomics 42:16–25.
Campbell, K. S., and M. Colonna. 2001. Human natural killer cell receptors and signal transduction. Int. Rev. Immunol. 20:333–370.
Cantoni, C., C. Bottino, M. Vitale, A. Pessino, R. Augugliaro, A. Malaspina, S. Parolini, L. Moretta, A. Moretta, and R. Biassoni. 1999. NKp44, a triggering receptor involved in tumor cell lysis by activated human natural killer cells, is a novel member of the immunoglobulin superfamily. J. Exp. Med. 189:787–796.
Chang, C., A. Rodriguez, M. Carretero, M. Lopez-Botet, J. H. Phillips, and L. L. Lanier. 1995. Molecular characterization of human CD94: a type II membrane glycoprotein related to the C-type lectin superfamily. Eur. J. Immunol. 25:2433–2437.
Colonna, M., H. Nakajima, F. Navarro, and M. Lopez-Botet. 1999. A novel family of Ig-like receptors for HLA class I molecules that modulate function of lymphoid and myeloid cells. J. Leukoc. Biol. 66:375–381.
De Maria, A., R. Biassoni, M. Fogli et al. (12 co-authors). 2001. Identification, molecular cloning and functional characterization of NKp46 and NKp30 natural cytotoxicity receptors in Macaca fascicularis NK cells. Eur. J. Immunol. 31:3546–3556.
Denley, A., J. C. Wallace, L. J. Cosgrove, and B. E. Forbes. 2003. The insulin receptor isoform exon 11- (IR-A) in cancer and other diseases: a review. Horm. Metab. Res. 35:778–785.
Foster, C. E., M. Colonna, and P. D. Sun. 2003. Crystal structure of the human natural killer (NK) cell activating receptor NKp46 reveals structural relationship to other leukocyte receptor complex immunoreceptors. J. Biol. Chem. 278:46081–46086.
Harrison, P. M., and M. Gerstein. 2002. Studying genomes through the aeons: protein families, pseudogenes and proteome evolution. J. Mol. Biol. 318:1155–1174.
Henikoff, S., E. A. Greene, S. Pietrokovski, P. Bork, T. K. Attwood, and L. Hood. 1997. Gene families: the taxonomy of protein paralogs and chimeras. Science 278:609–614.
Ho, E. L., J. W. Heusel, M. G. Brown, K. Matsumoto, A. A. Scalzo, and W. M. Yokoyama. 1998. Murine Nkg2d and Cd94 are clustered within the natural killer complex and are expressed independently in natural killer cells. Proc. Natl. Acad. Sci. USA 95:6320–6325.
Ladd, A. N., and T. A. Cooper. 2002. Finding signals that regulate alternative splicing in the post-genomic era. Genome Biol. 3:reviews0008.
Lanier, L. L. 1998. NK cell receptors. Annu. Rev. Immunol. 16:359–393.
———. 2003. Natural killer cell receptor signaling. Curr. Opin. Immunol. 15:308–314.
Lanier, L. L., C. Chang, and J. H. Phillips. 1994. Human NKR-P1A. A disulfide-linked homodimer of the C-type lectin superfamily expressed by a subset of NK and T lymphocytes. J. Immunol. 153:2417–2428.
Mager, D. L., K. L. McQueen, V. Wee, and J. D. Freeman. 2001. Evolution of natural killer cell receptors: coexistence of functional Ly49 and KIR genes in baboons. Curr. Biol. 11:626–630.
Mandelboim, O., N. Lieberman, M. Lev, L. Paul, T. I. Arnon, Y. Bushkin, D. M. Davis, J. L. Strominger, J. W. Yewdell, and A. Porgador. 2001. Recognition of haemagglutinins on virus-infected cells by NKp46 activates lysis by human NK cells. Nature 409:1055–1060.
McVicar, D. W., and D. N. Burshtyn. 2001. Intracellular signaling by the killer immunoglobulin-like receptors and Ly49. Sci. STKE 75:RE1.
Mighell, A. J., N. R. Smith, P. A. Robinson, and A. F. Markham. 2000. Vertebrate pseudogenes. FEBS Lett. 468:109–114.
Moretta, A., R. Biassoni, C. Bottino, M. C. Mingari, and L. Moretta. 2000. Natural cytotoxicity receptors that trigger human NK-cell-mediated cytolysis. Immunol. Today 21:228–234.
Moretta, A., C. Bottino, M. C. Mingari, R. Biassoni, and L. Moretta. 2002a. What is a natural killer cell? Nat. Immunol. 3:6–8.
Moretta, A., C. Bottino, M. Vitale, D. Pende, C. Cantoni, M. C. Mingari, R. Biassoni, and L. Moretta. 2001. Activating receptors and coreceptors involved in human natural killer cell-mediated cytolysis. Annu. Rev. Immunol. 19:197–223.
Moretta, L., C. Bottino, G. Ferlazzo, D. Pende, G. Melioli, M. C. Mingari, and A. Moretta. 2003. Surface receptors and functional interactions of human natural killer cells: from bench to the clinic. Cell. Mol. Life Sci. 60:2139–2146.
Moretta, L., C. Bottino, D. Pende, M. C. Mingari, R. Biassoni, and A. Moretta. 2002b. Human natural killer cells: their origin, receptors and function. Eur. J. Immunol. 32:1205–1211.
Moretta, L., and A. Moretta. 2003. Unravelling natural killer cell function: triggering and inhibitory human NK receptors. EMBO J. 18:18.
Neville, M. J., and R. D. Campbell. 1999. A new member of the Ig superfamily and a V-ATPase G subunit are among the predicted products of novel genes close to the TNF locus in the human MHC. J. Immunol. 162:4745–4754.
Pende, D., S. Parolini, A. Pessino et al. (13 co-authors). 1999. Identification and molecular characterization of NKp30, a novel triggering receptor involved in natural cytotoxicity mediated by human natural killer cells. J. Exp. Med. 190:1505–1516.
Pessino, A., S. Sivori, C. Bottino, A. Malaspina, L. Morelli, L. Moretta, R. Biassoni, and A. Moretta. 1998. Molecular cloning of NKp46: a novel member of the immunoglobulin superfamily involved in triggering of natural cytotoxicity. J. Exp. Med. 188:953–960.
Samaridis, J., and M. Colonna. 1997. Cloning of novel immunoglobulin superfamily receptors expressed on human myeloid and lymphoid cells: structural evidence for new stimulatory and inhibitory pathways. Eur. J. Immunol. 27:660–665.
Schnittger, S., J. Hamann, C. Dannenberg, H. Fiebig, M. Strauss, and C. Fonatsch. 1993. Regional sublocalization of the human CD69 gene to chromosome bands 12p12.3-p13.2, the predicted region of the human natural killer cell gene complex. Eur. J. Immunol. 23:2711–2713.
Sivakamasundari, R., A. Raghunathan, C. Y. Zhang, R. R. Chowdhury, and S. M. Weissman. 2000. Expression and cellular localization of the protein encoded by the 1C7 gene: a recently described component of the MHC. Immunogenetics 51:723–732.
Sivori, S., D. Pende, C. Bottino, E. Marcenaro, A. Pessino, R. Biassoni, L. Moretta, and A. Moretta. 1999. NKp46 is the major triggering receptor involved in the natural cytotoxicity of fresh or cultured human NK cells. Correlation between surface density of NKp46 and natural cytotoxicity against autologous, allogeneic or xenogeneic target cells. Eur. J. Immunol. 29:1656–1666.
Smith, H. R., A. H. Idris, and W. M. Yokoyama. 2001. Murine natural killer cell activation receptors. Immunol. Rev. 181:115–125.
Stamm, S. 2002. Signals and their transduction pathways regulating alternative splicing: a new dimension of the human genome. Hum. Mol. Genet. 11:2409–2416.
Takai, T., and M. Ono. 2001. Activating and inhibitory nature of the murine paired immunoglobulin-like receptor family. Immunol. Rev. 181:215–222.
Torkar, M., Z. Norgate, M. Colonna, J. Trowsdale, and M. J. Wilson. 1998. Isotypic variation of novel immunoglobulin-like transcript/killer cell inhibitory receptor loci in the leukocyte receptor complex. Eur. J. Immunol. 28:3959–3967.
Trowsdale, J., R. Barten, A. Haude, C. A. Stewart, S. Beck, and M. J. Wilson. 2001. The genomic context of natural killer receptor extended gene families. Immunol. Rev. 181:20–38.
Vitale, M., C. Bottino, S. Sivori, L. Sanseverino, R. Castriconi, E. Marcenaro, R. Augugliaro, L. Moretta, and A. Moretta. 1998. NKp44, a novel triggering surface molecule specifically expressed by activated natural killer cells, is involved in non-major histocompatibility complex-restricted tumor cell lysis. J. Exp. Med. 187:2065–2072.
Wagtmann, N., S. Rojo, E. Eichler, H. Mohrenweiser, and E. O. Long. 1997. A new human gene complex encoding the killer cell inhibitory receptors and related monocyte/macrophage receptors. Curr. Biol. 7:615–618.
Wende, H., A. Volz, and A. Ziegler. 2000. Extensive gene duplications and a large inversion characterize the human leukocyte receptor cluster. Immunogenetics 51:703–713.
Westgaard, I. H., S. F. Berg, S. Orstavik, S. Fossum, and E. Dissen. 1998. Identification of a human member of the Ly-49 multigene family. Eur. J. Immunol. 28:1839–1846.
Wilson, M. J., J. A. Lindquist, and J. Trowsdale. 2000. DAP12 and KAP10 (DAP10)-novel transmembrane adapter proteins of the CD3zeta family. Immunol. Res. 22:21–42.
Wu, J., H. Cherwinski, T. Spies, J. H. Phillips, and L. L. Lanier. 2000. DAP10 and DAP12 form distinct, but functionally cooperative, receptor complexes in natural killer cells. J. Exp. Med. 192:1059–1068.
Xie, T., L. Rowen, B. Aguado, M. E. Ahearn, A. Madan, S. Qin, R. D. Campbell, and L. Hood. 2003. Analysis of the gene-dense major histocompatibility complex class III region and its comparison to mouse. Genome Res. 13:2621–2636.
Yamanaka, S., X. Y. Zhang, K. Miura, S. Kim, and H. Iwao. 1998. The human gene encoding the lectin-type oxidized LDL receptor (OLR1) is a novel member of the natural killer gene complex with a unique expression profile. Genomics 54:191–199.
Zhang, Z., N. Carriero, and M. Gerstein. 2004. Comparative analysis of processed pseudogenes in the mouse and human genomes. Trends Genet. 20:62–67.(Martine Hollyoake*, R. Du)