Conservation and Variation in Human and Common Chimpanzee CD94 and NKG2 Genes

This information is current as Benny P. Shum, Laura R. Flodin, David G. Muir, Raja of October 2, 2021. Rajalingam, Salim I. Khakoo, Sophia Cleland, Lisbeth A. Guethlein, Markus Uhrberg and Peter Parham J Immunol 2002; 168:240-252; ; doi: 10.4049/jimmunol.168.1.240 http://www.jimmunol.org/content/168/1/240 Downloaded from

References This article cites 66 articles, 16 of which you can access for free at: http://www.jimmunol.org/content/168/1/240.full#ref-list-1 http://www.jimmunol.org/

Why The JI? Submit online.

• Rapid Reviews! 30 days* from submission to initial decision

• No Triage! Every submission reviewed by practicing scientists

• Fast Publication! 4 weeks from acceptance to publication by guest on October 2, 2021 *average

Subscription Information about subscribing to The Journal of Immunology is online at: http://jimmunol.org/subscription Permissions Submit copyright permission requests at: http://www.aai.org/About/Publications/JI/copyright.html Email Alerts Receive free email-alerts when new articles cite this article. Sign up at: http://jimmunol.org/alerts

The Journal of Immunology is published twice each month by The American Association of Immunologists, Inc., 1451 Rockville Pike, Suite 650, Rockville, MD 20852 Copyright © 2002 by The American Association of Immunologists All rights reserved. Print ISSN: 0022-1767 Online ISSN: 1550-6606. Conservation and Variation in Human and Common Chimpanzee CD94 and NKG2 Genes1

Benny P. Shum, Laura R. Flodin, David G. Muir, Raja Rajalingam, Salim I. Khakoo,2 Sophia Cleland, Lisbeth A. Guethlein, Markus Uhrberg,3 and Peter Parham4

To assess polymorphism and variation in human and chimpanzee NK complex genes, we determined the coding-region sequences for CD94 and NKG2A, C, D, E, and F from several human (Homo sapiens) donors and common chimpanzees (Pan troglodytes). CD94 is highly conserved, while the NKG2 genes exhibit some polymorphism. For all the genes, alternative mRNA splicing variants were frequent among the clones obtained by RT-PCR. Alternative splicing acts similarly in human and chimpanzee to produce the CD94B variant from the CD94 gene and the NKG2B variant from the NKG2A gene. Whereas single chimpanzee orthologs for CD94, NKG2A, NKG2E, and NKG2F were identified, two chimpanzee paralogs of the human NKG2C gene were defined. The chimpanzee Pt-NKG2CI gene encodes a protein similar to human NKG2C, whereas in the chimpanzee Pt-NKG2CII gene the Downloaded from translation frame changes near the beginning of the carbohydrate recognition domain, causing premature termination. Analysis of a panel of chimpanzee NK cell clones showed that Pt-NKG2CI and Pt-NKG2CII are independently and clonally expressed. Pt-NKG2CI and Pt-NKG2CII are equally diverged from human NKG2C, indicating that they arose by gene duplication subsequent to the divergence of chimpanzee and human ancestors. Genomic DNA from 80 individuals representing six primate species were typed for the presence of CD94 and NKG2. Each species gave distinctive typing patterns, with NKG2A and CD94 being most

conserved. Seven different NK complex genotypes within the panel of 48 common chimpanzees were due to differences in Pt-NKG2C http://www.jimmunol.org/ and Pt-NKG2D genes. The Journal of Immunology, 2002, 168: 240Ð252.

uman NK cell receptors specific for HLA class I molecules CD94:NKG2C heterodimer is a receptor with ligand specificity are encoded in two genetic complexes, both situated on similar to CD94:NKG2A, but which generates activating signals different chromosomes from the MHC on human chromo- within NK cells (22, 23). Human NKG2D is a homodimeric acti- H 5 some 6 (1, 2). The killer cell Ig-like receptors (KIRs) are encoded on vating receptor that interacts with MHC class I-like chain A (24, the leukocyte receptor complex or cluster (LRC) on human chromo- 25), a MHC class I-like protein with considerable polymorphism, some 19q13.4 (3Ð6) and the lectin-like CD94 and NKG2 receptors and with unique long binding proteins (ULBPs), newly identified are encoded within the NK complex (NKC) on human chromosome molecules distantly related to MHC class I (26). In NK cells, ac- by guest on October 2, 2021 12p12.3-p13.1 (7Ð12). In human, one CD94 and five NKG2 (A/B, C, tivating signals are conveyed via positively charged amino acid D, E/H, F) genes are located within the NKC (7Ð12). residues in the transmembrane domains of activating receptors The CD94 and NKG2A polypeptides form a heterodimeric re- (NKG2C, D) that interact with signal-transducing adaptor mole- ceptor (13, 14) that recognizes the MHC nonclassical class I cules such as DAP-10 and DAP-12 (27, 28). The exact roles of HLA-E ligand complexed with peptides derived from the leader human NKG2E and NKG2F are uncertain, but both encode a pos- sequences of certain classical HLA-A, B, C, and nonclassical itively charged lysine residue in the transmembrane domain, as HLA-G H chains (15Ð19). Ligand recognition transduces inhibi- does NKG2C. tory signals in NK cells via the immunomodulatory tyrosine-based The KIR genes of the human LRC encode receptors that are inhibition motifs on the cytoplasmic tail of NKG2A (20Ð22). The specific for polymorphic determinants of HLA-A, B, and C, and also for HLA-G (1, 29). Previous work from our laboratory re- vealed diversity of KIR haplotype in the human population, with Departments of Structural Biology and Microbiology and Immunology, Stanford Uni- versity School of Medicine, Stanford, CA 94305 respect to the number and content of KIR genes (30) and also Received for publication May 4, 2001. Accepted for publication October 22, 2001. allelic polymorphism at individual genes (31Ð34). Furthermore, The costs of publication of this article were defrayed in part by the payment of page comparison of human and chimpanzee KIR revealed many species- charges. This article must therefore be hereby marked advertisement in accordance specific features to the KIR gene systems (35, 36). Given the di- with 18 U.S.C. Section 1734 solely to indicate this fact. versity and rapid evolution of the KIR genes it became of particular 1 This work was supported by National Institutes of Health Grants AI17892 and interest to investigate intraspecies and interspecies diversity of AI31168 (to P.P.). S.I.K. was a fellow of the Cancer Research Institute of New York. M.U. was supported by the Ministerium fu¬r Schule und Weiterdildung, Wissenschaft NKC genes that encode NK cell receptors for MHC class I mol- und Forschung des Landes Nordrhein-Westfalen. ecules. Therefore, we have investigated the diversity and variation 2 Current address: Department of Cell and Molecular Medicine, Southampton General in the CD94 and NKG2 genes of humans and apes. Hospital, Southampton, U.K. 3 Current address: Institute for Transplantation Diagnostics and Cell Therapeutics, Materials and Methods Heinrich Heine University Medical Center, Dusseldorf, Germany. Source materials from humans and apes 4 Address correspondence to Dr. Peter Parham, Department of Structural Biology, Stan- ford University School of Medicine, 299 Campus Drive West, Sherman Fairchild Build- Healthy human donors recruited from our laboratory represent a variety of ing, D-157, Stanford, CA 94305-5126. E-mail address: [email protected] ethnic and racial backgrounds, as shown in Fig. 1. Tissue samples from 5 ␤ ␤ non-human primates were purchased from the Yerkes Regional Primate Abbreviations used in this paper: KIR, killer cell Ig-like receptor; 2m, 2-micro- globulin; CRD, carbohydrate recognition domain; LRC, leukocyte receptor complex Center at Emory University (Atlanta, GA) and from the Laboratory for or cluster; NKC, NK complex; ULBP, unique long binding protein. Experimental Medicine and Surgery in Primates (New York University

Copyright © 2002 by The American Association of Immunologists 0022-1767/02/$02.00 The Journal of Immunology 241

Medical Center, Tuxedo, NY). Blood was drawn from animal subjects as cDNA analysis of CD94 and NKG2 genes part of routine health examinations. Total cytoplasmic RNA was prepared from PMBC isolated from human and chimpanzee donors. In addition, Total cytoplasmic RNA was isolated with the reagent RNAzol following RNA was extracted from NK cell clones NK1.3 and M1.1 established from manufacturer’s protocol (Tel-Test, Friendswood, TX). First-strand cDNA chimpanzees Cathy and Mason, respectively (35); a splenic sample from was prepared from total RNA using murine Moloney leukemia virus-re- the chimpanzee Edwina was also used as source material. Genomic DNA verse transcriptase (Life Technologies, Rockville, MD) by standard pro- of higher primates used for PCR typing analysis was isolated from previ- cedures (38). ously established EBV-transformed B cell lines (36, 37). These include cDNA was amplified by PCR in 1ϫ AmpliTaq buffer (Applied Biosys- samples from the family Hominidae: humans (Homo sapiens, n ϭ 11), tems, Foster City, CA), 0.25 mM of each dNTP, 0.8 pM each of a sense common chimpanzees (Pan troglodytes, n ϭ 48), pygmy chimpanzees or (forward) and antisense (reverse) primer, and1UofAmpliTaq DNA poly- bonobos (P. paniscus, n ϭ 11), gorillas (Gorilla gorilla, n ϭ 2), and or- merase (Applied Biosystems). The following PCR conditions were used: angutans (Pongo pygmaeus, n ϭ 5); and from the family Hylobatidae: initial denaturation at 96¡C for 1 min; then 30 cycles of denaturation at common gibbons (Hylobates lar, n ϭ 3). 94¡C for 30 s, primer annealing at 52Ð60¡C for 30 s, and extension at

Table I. Oligonucleotide primers

Name Orientation Sequence (5Ј33Ј)

94-CyF1 Sense TGG CAG TGT TTA AGA CCA CT

94-E6R1 Antisense TTC ACA GGA TTC ATC TAA AGC Downloaded from 94g-E5F1 Sense TTT TAT GAG CTC CAG TCA ACA 94g-E6R1 Antisense CCA TTT GGA TTA TAC GCT ATG A-E2F1 Sense TCA AAA AGC TTC TCA GGA TTT A-E5R1 Antisense TGA GGA TGG TGA AAT GAT GGA Ag-E1F1a Sense GAA CAG GAA ATA ACC TAT GC Ag-E2R1 Antisense AGG CCA TTA AGA TAA GAC AG ␤ m-F1 Sense TGT GTC TGG GTT TCA TCC ATC

2 http://www.jimmunol.org/ ␤ 2m-R1 Antisense CTA TCT TGG GCT GTG ACA AAG CD94-E4F2 Sense CCG GTG CAA CTG TTA CTT CA CD94-E5R2 Antisense GGT GTG CTC CTC ACT GTA AGA CD94-PCRF Sense CCT TCT CTA CTT CGC TCT TG CD94-PCRR Antisense TAA TGT TGG GTC TTT ACT CTC Cg-E1F1 Sense CTT CAA AAT CCT TCC Cg-E4R1 Antisense GGA GTT CTT CGA AGT cG2C1-F1 Sense GTG GAT CTT CAA TAA TAT ATC cG2C1-R1 Antisense CAT CAC GAC ACA AAG AAA CA cG2C2-F1 Sense GTG GAT CTT CAA TGA TGT ATT

cG2C2-R1 Antisense CAT CAC GAC ACA AAG TAA CG by guest on October 2, 2021 CI-E1F1 Sense GAA CAG GAA ATA TTC CAA GTA CI-E6R1 Antisense AGC TTA TGC TTA CAA TGA TAT CII-E1F1 Sense GAA CAG GAA ATA TTC CAA GTG CII-E6R1 Antisense AGC TTA TGC TTA CAA TAA TAC D-CyF1 Sense AGT GAT TTT TCA ACA CGA TGG D-E10R1 Antisense CTT TAA AGC TCG AGG CAT AGA Dfor1 Sense AGC AAT ATG GAG TGC TGT ATT DFORa Sense TAT CCA CAA GAA TCA AGA TCT Dg-E4F1 Sense GTG GAT TCG TGG TCG GAG GTC Dg-E5R1 Antisense TTG CTT TTG CCA TCG TGT TGA DREV Antisense GTC CTG TGG AGT CTA AGA AAT Drev1 Antisense TTT TAG GAC ATG GGC CAC AGT A E-E7R1 Antisense ATG ATG AAA CCC CGT CTA ATG E-R1 Antisense GGG CGG TGG CTT GTG TCA GTA E-R4 Antisense GAG GCC AAG ATG TGT TGA TT ECF-E1F1 Sense CGA ACA GGA AAT ATT CCA AGT ECF-F1 Sense TTA TCA TAG AGC ACA GTC CCT ECF-F4 Sense CAG TCC CTC ACA TCA CAC AGC hNKG2AC-R1 Antisense GGC AAC CAC TAT TCT ACT TTC hNKG2A-PCRF Sense CTG GGG ACA GAA GAG TAC AGT hNKG2ABC-R3 Antisense TAT AGA AAG CAG ACT GGA GTT hNKG2AC-PR1 Antisense GCC CGA CAC AAA TGC TAG GAT NKG2ABC-F2 Sense GTG GCC ATT GTC CTG AGG AGT NKG2ABC-F3 Sense ATC AGC CCA GTG TGG ATC TTC NKG2ABC-PCRF Sense ATC ACA CAG CTG CAG AGA NKG2ABC-PCRR Antisense TGA TGT TTA GTA TAT TCG CA NKG2ABC-R2 Antisense ATA AAA TGT ATC TGA TGC ACT NKG2ABC-R3 Antisense TAT AGA AAG CAG ACC GGA GTT NKG2C-F1 Sense ATC ACA CAG CTG CAG AG NKG2C-PCRF Sense AGC ACA GTC CCT CAC ATC ACA NKG2C1-R1 Antisense CAC AAA GAA ACA TTC TAT CAA NKG2CN-R1 Antisense TGA GCA AAA TGA GCC CA NKG2F-F1 Sense CTG CAG TTT GCC TAT ACC AG NKG2F-for Sense GCA TTG TCC TGA TGG CCA CTG T NKG2F-R1 Antisense ACA TTT ATA GAA GCA TGG GTT NKG2F-rev Antisense ACA GTG GCC ATC AGG ACA ATG C 242 HUMAN AND CHIMPANZEE CD94 AND NKG2 GENE VARIATION

72¡C for 1 min; followed by 72¡C for 10 min to promote complete Table II. CD94 and NKG2 sequences extension. Primers were designed from available human cDNA sequences in the Sequence Gene Symbola Accession No. GenBank database. Initially, certain primer sets were designed to amplify more than one NKG2 gene; gene-specific primer sets were used subse- CD94 KLRD1 NM_002262 quently. The following oligonucleotide primers were used for our cDNA Pt-CD94 AF259054 amplifications: CD94-PCRF and CD94-PCRR (for CD94 and Pt-CD94), NKG2D01 D12S2489E NM_007360 NKG2ABC-PCRF and NKG2ABC-PCRR (for NKG2A/B,C,andPt- NKG2D02 D12S2489E AF260135 NKG2A/B, C); hNKG2A-PCRF and hNKG2AC-R1 (for NKG2A/B and NKG2D03 D12S2489E AF260136 Pt-NKG2A/B); NKG2C-PCRF and hNKG2AC-PR1 (for NKG2C); Pt-NKG2D AF259063 NKG2C-F1 and NKG2C1-R1 (for Pt-NKG2CI); NKG2C-F1 and NKG2A KLRC1 NM_002259 NKG2CN-R1 (for Pt-NKG2CII); DFORa and DREV (for NKG2D and Pt- Pt-NKG2A01 AF259055 NKG2D); ECF-F1 and E-R1 (for NKG2E); ECF-F4 and E-R4 (for Pt- Pt-NKG2A02 AF259056 NKG2E); ECF-E1F1 and E-E7R1 (for Pt-NKG2E partial); and NKG2F-F1 Pt-NKG2A03 AF350005 and NKG2F-R1 (for NKG2F and Pt-NKG2F) (Table I). NKG2C01 KLRC2 Y13055 PCR-amplified cDNA was cloned into the pCR-TOPO plasmid vector NKG2C02 KLRC2 AF260134 (Invitrogen, San Diego, CA) and multiple clones from each amplification Pt-NKG2CI01 AF259057 were selected for analysis. Nucleotide sequences of plasmid DNA insert Pt-NKG2CI02 AF259058 were determined on both strands using the Big-Dye Terminator Cycle Se- Pt-NKG2CI03 AF259059 quencing kit (Applied Biosystems) and a 377 Automated DNA sequencer Pt-NKG2CI04 AF259060 (Applied Biosystems). In addition to the universal T3 and T7 oligonucle- Pt-NKG2CII01 AF259061 otide primers, the following primers were used to sequence completely the Pt-NKG2CII02 AF259062 amplified cDNA: for NKG2A/B, NKG2ABC-F2 and hNKG2ABC-R3; for NKG2E01 KLRC3 NM_002261 Downloaded from Pt-NKG2A/B, NKG2ABC-F2 and NKG2ABC-R3; for NKG2C and Pt- NKG2E02 KLRC3 AF350016 NKG2CI/CII, NKG2ABC-F2, NKG2ABC-R2, NKG2ABC-F3, and NKG2E03 KLRC3 AF350017 NKG2ABC-R3; for NKG2D and Pt-NKG2D, Dfor1 and Drev1; and for Pt-NKG2E01 AF350006 NKG2F and Pt-NKG2F, NKG2F-forward and NKG2F-reverse (Table I). Pt-NKG2E02 AF350007 DNA sequences were assembled and analyzed using the computer pro- Pt-NKG2E03 AF350008 gram AutoAssembler (version 2.1; Applied Biosystems) and the Wisconsin Pt-NKG2E04 AF350009 Package sequence analysis software (version 10.1; Genetics Computer Pt-NKG2E05 AF350010 Group, Madison, WI). NKG2F01 KLRC4 U96845 http://www.jimmunol.org/ Nucleotide sequences NKG2F02 KLRC4 NM_013431 NKG2F03 KLRC4 AF350018 All common chimpanzee CD94 and NKG2 genes were assigned the prefix NKG2F04 KLRC4 AF350019 “Pt- ” for P. troglodytes. New nucleotide sequences determined here have Pt-NKG2F01 AF350011 been deposited into the GenBank database under the accession numbers Pt-NKG2F02 AF350012 AF259061Ð63, AF260135Ð6, and AF350005Ð19 (Table II). Sequences Pt-NKG2F03 AF350013 previously reported by us have the accession numbers AF259054Ð60 and Pt-NKG2F04 AF350014 AF260134 (35) (Table II). Sequences isolated from this study and reported Pt-NKG2F05 AF350015 by others are also shown in Table II; the “NM” prefix in the accession a Human Gene Mapping Workshop approved symbols.

number denotes GenBank reference sequences. Numerous alternatively by guest on October 2, 2021 spliced variants were also identified in our analysis; these have not been deposited into the GenBank database but are available from B. P. Shum by request. Analysis of the intron 4/exon 5 boundary of Pt-CD94 Pt-NKG2A, A-E2F1 and A-E5R1; for NKG2C and Pt-NKG2CI, CI-E1F1 and CI-E6R1; for Pt-NKG2CII, CII-E1F1 and CII-E6R1; and for NKG2D To examine whether alternative splicing could generate the chimpanzee and Pt-NKG2D, D-CyF1 and D-E10R1 (Table I). Selected amplification CD94 and CD94B variants, which differ by insertion of three nucleotides products were sequenced to confirm specificity of amplifications. (encoding a glutamine) between exon 4 and exon 5 in the cDNA, we determined the DNA sequence at the intron 4/exon 5 boundary of the Genomic DNA typing Pt-CD94 gene. Genomic DNA from the common chimpanzee Cathy was amplified by PCR with oligonucleotide primers CD94-E4F2 and CD94- The presence of CD94 and NKG2 genes was typed by PCR amplification E5R2 (Table I). The 50-␮l reaction contained 1ϫ AmpliTaq buffer (Ap- of genomic DNA isolated from a panel of higher primates using gene- plied Biosystems), 0.25 mM of each dNTP, 0.1 ␮M of each primer, 2.5 U specific oligonucleotide primers. Because these amplification primers were of AmpliTaq DNA polymerase (Applied Biosystems), and 200 ng of based on human and chimpanzee sequences, negative results in other spe- genomic DNA. PCR was performed with an initial denaturation at 96¡C for cies should not be interpreted to mean the absence of a gene. All typing 2 min; 30 cycles of denaturation at 94¡C for 30 s, annealing at 60¡C for experiments were repeated at least once for confirmation. The genomic 30 s, extension at 72¡C for 2 min; and a final extension of 72¡C for 10 min. typing primers used were as follows: for CD94 and Pt-CD94, 94g-E5F1 The PCR product (ϳ1969 bp) was cloned into the pCR-TOPO vector (In- and 94g-E6R1; for NKG2A and Pt-NKG2A, Ag-E1F1a and Ag-E2R1; for vitrogen). Fifteen plasmid clones were isolated and their nucleotide se- NKG2C, Pt-NKG2CI, and CII, Cg-E1F1 and Cg-E4R1; for Pt-NKG2CI, quence was determined using T3 and T7 universal primers. cG2C1-F1 and cG2C1-R1; for Pt-NKG2CII, cG2C2-F1 and cG2C2-R1; for ␤ NKG2D and Pt-NKG2D, Dg-E4F1 and Dg-E5R1; and for 2-microglobu- Transcription of the common chimpanzee Pt-NKG2CI and CII genes ␤ ␤ ␤ lin ( 2m) control, 2m-F1 and 2m-R1 (Table I). ␮ ϫ To examine the transcription of Pt-NKG2CI and Pt-NKG2CII genes in NK PCRs were performed in a volume of 25 l containing 1 AmpliTaq ␮ cells, we amplified cDNA by RT-PCR with gene-specific primers from buffer (Applied Biosystems), 0.1 mM of each dNTP, 0.4 M of each individual NK cell clones; these NK clones established from the common primer,2UofAmpliTaq DNA polymerase (Applied Biosystems), and ϳ chimpanzee Cathy have been described previously (35). Each cDNA typ- 100 ng of genomic DNA. The following conditions were used for am- ing reaction of 25-␮l volume contained 1ϫ AmpliTaq buffer (Applied plification: initial denaturation at 96¡C for 2 min; 10 cycles of a first PCR Biosystems), 0.25 mM of each dNTP, 0.8 ␮M of each primer, and1Uof (denaturation at 94¡C for 15 s, annealing at 65¡C for 30 s, extension at AmpliTaq DNA polymerase (Applied Biosystems). PCRs were performed 72¡C for 1.5 min); 27 cycles of a second PCR (denaturation at 94¡C for with an initial denaturation at 96¡C for 1 min; followed by 35 cycles of 15 s, annealing at 60¡C for 30 s, extension at 72¡C for 1 min); and a final denaturation at 94¡C for 15 s, annealing at 52¡C for 15 s, and extension at extension at 72¡C for 10 min. The amplified products were separated by 72¡C for 30 s; then a final extension at 72¡C for 10 min. Amplified product electrophoresis on a 1.5% agarose gel in 0.5ϫ Tris buffer with EDTA was visualized under UV light after separation by electrophoresis on a buffer and were visualized by ethidium bromide staining afterward. In- 1.2% agarose E-gel following manufacturer’s instructions (Invitrogen). cluded as positive controls for specificity were ϳ1 pg of plasmids con- Gene-specific oligonucleotide primers used for cDNA typing were as fol- taining cDNAs of each human and chimpanzee gene. Selected amplifica- lows: for CD94 and Pt-CD94, 94-CyF1 and 94-E6R1; for NKG2A and tion products were sequenced to confirm specificity. The Journal of Immunology 243

Phylogenetic analysis CD94 Sequences were aligned with the Pileup program of the Wisconsin Package Data from six humans and four chimpanzees reveals no CD94 (version 10.1; Genetics Computer Group) and manually adjusted after- polymorphism in either species (Fig. 1). Human and chimpanzee ward. Phylogenetic analysis was performed using the PAUP* 4.0 software CD94 differ at only two nucleotide positions in the coding se- package (Phylogenetic Analysis Using Parsimony (*and Other Methods), version 4.0b4a; Sinauer Associates, Sutherland, MA). The neighbor-join- quence, one of which is nonsynonymous and the other synony- ing method was used to construct dendrograms. Confidence of individual mous (Fig. 2). nodes was evaluated by 1000 bootstrap replications, and majority-rule con- CD94B is a previously reported cDNA variant of human CD94 sensus trees were generated. (GenBank accession no. AJ000001) that differs from CD94 by the coding of an additional glutamine residue (position 183 in Fig. 3). Results We found variants of both types in both human and chimpanzee Allelic and splice variations of human and common chimpanzee (Figs. 1, 3, and 4). Comparison of the two cDNA sequences with NKC genes the CD94 gene sequence indicates that the difference between To investigate the polymorphism and interspecies divergence of CD94 and CD94B is due to alternative patterns of mRNA splicing CD94 and NKG2 genes, we determined their coding region cDNA at the junction between intron 4 and exon 5 (Fig. 4B). To deter- sequences from four common chimpanzees and from healthy hu- mine whether the common chimpanzee Pt-CD94 and Pt-CD94B man donors. The panel of 14 human donors represented major variants can be generated in the same manner, we amplified, ethnic groups, but not all of the genes were characterized for all cloned, and sequenced a genomic DNA segment spanning the donors (Fig. 1). No allelic polymorphism was found for CD94 in boundary of intron 4 and exon 5 from chimpanzee Cathy, who either human or common chimpanzee, whereas polymorphism was expressed both Pt-CD94 and Pt-CD94B (Fig. 1B). Fifteen clones Downloaded from seen for all NKG2 genes in either one or both species. Many al- were analyzed and they all yielded a sequence that was identical to ternatively spliced variants emerged from our analysis, including that of the human gene for 40 nt flanking each side of the intron/ the previously described CD94/B and NKG2A/B (GenBank acces- exon boundary. Thus the mechanism of alternative splicing by sion no. AJ000001) (7). Alternatively spliced variants of several which the two CD94 variants are produced appears the same in types were identified: ones lacking exons, ones including introns, chimpanzee and human.

ones generated from cryptic splice sites, and one having combi- Clones for five additional alternatively spliced forms of Pt- http://www.jimmunol.org/ nations of the above. Assignments for cryptic splice sites were CD94 were obtained (Fig. 4). Four of these variants were spliced made from comparison of the variant cDNA sequence with the like CD94B but had additional changes. One lacked exon 2 (vari- sequence of full-length cDNA and/or corresponding gene se- ant 1), one lacked exon 4 (variant 2), one lacked both exons 2 and quences. The presence of highly conserved “GT” or “AG” dinucle- 4 (variant 3), and one lacked the 3Ј half of exon 4 (variant 4). otides at most of the alternatively spliced junctions revealed the Variant 5 was similarly missing the 3Ј half of exon 4, but it is use of cryptic splice sites (data not shown) (39). spliced like CD94; this truncation of exon 4 was likely due to a by guest on October 2, 2021

FIGURE 1. Individuals studied and the CD94 and NKG2 cDNA defined. For each human donor (A) and each common chimpanzee (B) the numbers of cDNA clones characterized for each allele defined are listed. In addition, the number of clones characterized the CD94B splice variant of CD94 and the NKG2B splice variant of NKG2A are shown. All alleles were defined by the sequences of more than one cDNA clone. 244 HUMAN AND CHIMPANZEE CD94 AND NKG2 GENE VARIATION

cryptic 5Ј donor splice site within the exon. An alternatively spliced form of human CD94 lacking exon 2 has also been de- scribed previously (40). Only variant 1 maintains the proper frame of translation throughout, while variants 2Ð5 all have a frameshift and premature termination within exon 5. In summary, CD94 is encoded by a nonpolymorphic gene that is highly conserved in human and chimpanzee, having a nucleotide sequence similarity of 99.6%. Alternatively spliced CD94 variants were found in both species.

NKG2D Three NKG2D alleles were characterized from the analysis of 14 human donors (Fig. 1). Each allele was identified from more than one donor and at least five donors were NKG2D heterozygotes (Fig. 1A); NKG2D01 corresponds to the sequence originally de- scribed by Houchins et al. (7). The three NKG2D alleles are very similar, differing by substitutions at only two nucleotide positions, of which one is nonsynonymous and the other synonymous (Figs.

2 and 3). A single Pt-NKG2D sequence was encountered in the Downloaded from analysis of the four chimpanzees, and it differs from the human NKG2D alleles by 7- to 8-nt substitutions which produce 1- to 2-aa differences (Figs. 1B, 2, and 3). Several alternatively spliced NKG2D variants were character- ized, two of which (variants 1 and 2) were found in both human and common chimpanzee (Fig. 4). Variants were found that in- http://www.jimmunol.org/ cluded 106 bp of the 3Ј end of intron 3, intron 6, 21 bp at the 3Ј end of intron 8, or 23 bp within intron 9, or were missing exon 8. The additional intron sequence in variant 1 is in the 5Ј untranslated region (3Ј end of intron 3) and should not affect the protein prod- uct. The intron 6 inclusion (in variants 2, 3, and 4) results in a stop codon almost immediately after the start of the intron 6 sequence and the encoded proteins therefore lack an extracellular domain. Variant 5 has seven additional amino acid residues in the carbo-

hydrate recognition domain (CRD) due to inclusion in the mRNA by guest on October 2, 2021 21 bp from the 3Ј end of intron 8. The additional 23 bp of intron 9 in variant 6 introduces a stop codon that prevents the translation of exon 10. In summary, we have defined three very similar human NKG2D alleles and a single chimpanzee Pt-NKG2D allele. The sequences have a nucleotide identity of ϳ98.9%, demonstrating that NKG2D, like CD94, is a highly conserved gene. A number of alternatively spliced NKG2D variants were identified in both human and chimpanzee.

NKG2A Whereas a single NKG2A sequence was identified from analysis of six human donors, three NKG2A alleles were found in the four chimpanzees (Fig. 1). The human sequence was identical to that previously reported (7, 41). The three Pt-NKG2A alleles differ at four nucleotide substitutions of which two are nonsynonymous and two are synonymous (Figs. 1Ð3). Human and chimpanzee NKG2A differ by 7Ð10 nt substitutions in the coding region and the encoded proteins differ by 4Ð6 amino acid substitutions (Figs. 2 and 3). FIGURE 2. Nucleotide and amino acid differences in human and com- Previous analysis indicated that NKG2A and NKG2B are alter- mon chimpanzee CD94 and NKG2 sequences. Listed are the positions of natively spliced forms of the same gene, with NKG2A including all nucleotide sequence difference within the set of human and chimpanzee exons in the message, whereas NKG2B message lacks exon 4 (Fig. alleles for each gene. Numbering starts with the initiating codon “ATG” of the cDNA. Each sequence is compared with the left-most human sequence; identities are indicated by a dash and deletions by “del.” In the column for CD94; AJ001687, AJ001688, and AJ001689 for NKG2D; AF023840 for headed “Amino Acid Substitutions,” both the residue in the reference se- NKG2A; AJ001684 for NKG2C; AJ001685 for NKG2E; and AJ001686 quence and other human and chimpanzee allotypes are indicated where for NKG2F). Protein domains are designated according to annotations in Gen- there are differences. Substitution to a termination codon is indicated by an Bank reference sequences (NM 002259Ð62, 007360, 013431). CY, Cyto- asterisk. Species-specific nucleotides are shaded gray. The exon numbering plasmic domain; TM, transmembrane domain; S, extracellular stalk. The derives from the following sequences in the GenBank database (AJ000673 Pt-NKG2CII sequences are not included. The Journal of Immunology 245

4 and Refs. 7 and 41). Clones corresponding to NKG2B were ob- which the Pt-NKG2CI and CII genes arose by gene duplication tained from all six human donors analyzed and from one of the within the chimpanzee lineage after its split from the human four chimpanzees (Fig. 1). An additional splice variant found in lineage. chimpanzee Pt-NKG2A03 (variant 1) lacking exons 4 and 6 causes Alternatively spliced variants of both Pt-NKG2CI and CII were a frameshift and premature termination in exon 7. encountered (Fig. 4). Both Pt-NKG2CI variant 1 and CII variant 1 In summary, NKG2A is a highly conserved gene with ϳ98.8% were missing the same 48- to 49-bp region at the 5Ј end of exon 4 nucleotide sequence identity in human and chimpanzee and share likely due to use of a cryptic splice site. Variant 2 from the Pt- a pattern of alternative mRNA splicing to give the NKG2B form. NKG2CI02 allele might be generated by cryptic splice sites in Human NKG2A appears to have no polymorphism, whereas chim- exons 2 and 4. Pt-NKG2CII variant 2 was missing exon 5. All of panzee Pt-NKG2A has modest polymorphism. these variants contain frameshifts that cause termination at the same place as in Pt-NKG2CII and (Pt-)NKG2F. NKG2C Transcription of the Pt-NKG2CI and CII genes in a panel of NK Analysis of nine human donors identified two NKG2C alleles, both cell clones was examined. Seventeen NK cell clones derived from of which were carried by several individuals, two of whom were the PBMC of chimpanzee Cathy (35) were typed by RT-PCR for shown to be NKG2C heterozygotes (Fig. 1A). The NKG2C01 allele the presence of Pt-NKG2CI and Pt-NKG2CII mRNA (Fig. 6A). has a sequence identical to that deposited by Cantoni et al. (23) Six of the clones (35%) typed positively for Pt-NKG2CI and three into GenBank (accession no. Y13055). NKG2C02 differs from it (18%) typed positively for Pt-NKG2CII. One of the NK cell clones by two nonsynonymous substitutions at nucleotide positions 5 and typed positively for both Pt-NKG2CI and Pt-NKG2CII. The neg- 305 (Figs. 2 and 3). The NKG2C sequence described by Houchins ative typing reactions were not due to poor quality mRNA, as all Downloaded from et al. (7) (GenBank accession no. NM_002260) was not encoun- the NK cell clones typed positively for expression of Pt-NKG2D. tered in our study and differs from NKG2C01 by one nonsynony- In conclusion, there is differential expression of the two Pt- mous nucleotide substitution. NKG2C genes in chimpanzee NK cell populations. Six different Pt-NKG2C sequences were seen in the cDNA In comparison to CD94, NKG2D, and NKGA,wefind NKG2C clones isolated from the four common chimpanzees (Fig. 1B). Four to be both polymorphic within chimpanzees and humans and more of the sequences encode proteins that are similar in length and divergent between the species (ϳ3.2Ð3.3% nucleotide differences). sequence to human NKG2C (Pt-NKG2CI01–04; Figs. 2 and 3). Of Within-species variation is more striking for chimpanzee than for http://www.jimmunol.org/ these the Pt-NKG2CI01, 02, and 04 proteins have two extra amino human, the former species having two distinctive, transcribed acid residues in the carboxyl terminus when compared with the NKG2C genes. human NKG2C; Pt-NKG2CI02 is five residues shorter due to an additional 21-bp deletion in exon 1 that maintains the proper read- NKG2E ing frame (Figs. 2 and 3). The other two chimpanzee sequences Human NKG2E is closely related to NKG2C, but its mRNA splices (Pt-NKG2CII01 and 02) share a nucleotide deletion at position 358 differentially into an additional exon 7 not used by any other NKG2 that alters the reading frame near the start of the region encoding gene (Fig. 4). An alternatively spliced NKG2H form (more similar

the CRD. Starting with position 140 (peptide alignment in Fig. 3) in its splicing pattern to NKG2C) has also been described (Figs. 3 by guest on October 2, 2021 the proteins encoded by Pt-NKG2CII01 and 02 become totally and 4 and Ref. 42). The oligonucleotide primers we used for am- different from those encoded by Pt-NKG2CI and NKG2C, and they plification were designed to detect the NKG2E form, but not the terminate 35 amino acid residues following the frameshift. Thus NKG2H form. Three NKG2E alleles were identified from nine hu- the predicted Pt-NKG2CII proteins have cytoplasmic, transmem- man donors analyzed, with at least two donors being heterozygotes brane, and extracellular stalk regions like NKG2C, but the CRD is (Fig. 1A). The human NKG2E alleles vary at only two nucleotide replaced by a shorter region of unknown structure. Pt-NKG2CII positions, of which one is nonsynonymous and the other synony- terminates at the same position as human and chimpanzee NKG2F, mous (Figs. 2 and 3). NKG2E01 is identical to the GenBank ref- but the two genes have different frameshift mechanisms (Fig. 3). erence sequence NM 002261 (43). In comparison to the CD94, NKG2D, and NKG2A genes, The oligonucleotide primers used to amplify human NKG2E NKG2C was found to be most diverged (Fig. 2). Human NKG2C failed to amplify from chimpanzee cDNA. Successful amplifica- and the chimpanzee alleles that encode similar protein products tion of chimpanzee Pt-NKG2E was eventually achieved with two (Pt-NKG2CI) differ by 22Ð24 nucleotide substitutions other primer sets, one that amplified a segment of 552 nt corre- (96.6Ð96.9% sequence identities) that produce 14Ð17 amino acid sponding to about three quarters of the coding region, and another differences (Figs. 3 and 5A). Furthermore, that three different Pt- that amplified the entire coding region. Clones corresponding to NKG2C sequences were isolated from two chimpanzees (Edwina Pt-NKG2E were obtained from three of the four chimpanzees and Elwood; Fig. 1B) indicates that there can be two Pt-NKG2C tested. Most of the clones corresponded to alternatively spliced genes in common chimpanzee genomes. In addition, clones cor- variants or partial Pt-NKG2E sequences; only one full-length clone responding to both Pt-NKG2CI and CII sequences were isolated was obtained. The divergence between the human and chimpanzee from all four chimpanzees examined. Pairwise comparison of the NKG2E nucleotide sequences (ϳ3.4%) is comparable to that seen coding region (Fig. 5A) and the 3Ј untranslated region (Fig. 5B)of for NKG2C. The nucleotide sequences of Pt-NKG2E clones ana- six Pt-NKG2C sequences clearly shows that they divide into two lyzed provided evidence for five alleles, distinguished by the pat- groups: one comprising the Pt-NKG2CI sequences, the other com- terns of nucleotide substitution at six positions. Four of these po- prising the Pt-NKG2CII sequences. Comparison between the cod- sitions involve nonsynonymous substitutions and two involve ing regions of the two common chimpanzee genes yields differ- synonymous substitutions. All three chimpanzees from whom Pt- ences of 11Ð16 nucleotide substitutions, not including gaps (Fig. NKG2E sequences were obtained appeared to be heterozygotes; 5A). The range of nucleotide comparisons within each group repeated attempts to amplify Pt-NKG2E from chimpanzee Elwood (NKG2C, Pt-NK2CI, and CII) is lower and does not overlap with failed. the range of differences for intergroup comparisons (Fig. 5). That Two alternatively spliced variants were identified from the hu- both chimpanzee Pt-NKG2C genes are equally divergent in nucle- man NKG2E (Fig. 4, variants 1 and 2) which were distinct from the otide sequence from human NKG2C is consistent with a history in eight variants of Pt-NKG2E (variants 3Ð10), but all contain a 246 HUMAN AND CHIMPANZEE CD94 AND NKG2 GENE VARIATION Downloaded from http://www.jimmunol.org/ by guest on October 2, 2021

FIGURE 3. Predicted amino acid sequences of human and common chimpanzee CD94 and NKG2 polypeptides. Sequences are compared with the consensus and identities are indicated by dashes. Numbering refers only to the consensus sequence and not to any individual protein. Protein domains are designated according to annotations in GenBank reference sequences (NM_002259Ð62, 007360, 013431). Gaps were introduced to optimize alignment and are indicated with periods. NKG2H (NM_007333), an alternatively spliced form of the human NKG2E, is also included for comparison. The cytoplasmic, transmembrane, and stalk regions of NKG2D are probably not homologous with their counterparts in other NKG2 molecules, nor is the The Journal of Immunology 247 change in the normal reading frame resulting in premature termi- (Pt-)NKG2C, one targeted at Pt-NKG2CI, and one targeted at nation. Whereas all variants involved “conventional” splicing dif- Pt-NKG2CII. ␤ ferences within the exons and introns of Pt-NKG2E, variant 10 All 80 samples typed positively for 2m and NKG2A, and all but (Fig. 4) consists of four exons from Pt-NKG2CII02 (though miss- the three gibbons typed positively for CD94. The 11 humans gave ing the 5Ј third of exon 4), followed by 50 bp identical to a part of identical typing reactions in which the only negatives were for the the NKG2E promoter (also identical to part of the NKG2C pro- two Pt-NKG2C genes. In contrast, a heterogeneity of typing pat- moter), and then continues with exons 3 through 6 of Pt-NKG2E03 tern was seen within the panel of 48 common chimpanzees. The (but missing the 3Ј part of exon 4). This suggests that the dupli- most frequent pattern, seen in 30 individuals, comprised positive cated Pt-NKG2CII gene is directly upstream of Pt-NKG2E. typing for all the genes tested. Six other patterns, all seen in at least Of the NKC genes compared in chimpanzee and human, NKG2E two individuals, were distinguished by negative reactions for and NKG2C show the highest interspecies divergence (ϳ3.2Ð3.4% NKG2C and NKG2D genes. Most divergent from the dominant difference in nucleotide sequence) and they are both polymorphic pattern 1 was pattern 2, in which typing for NKG2D and all three within each species. Alternatively spliced variants of NKG2E were NKG2C reactions was negative (Fig. 6B). This pattern of reactivity frequently encountered, all of them containing frameshifts that was also given by the two gorillas tested. cause premature termination. The 11 pygmy chimpanzees tested gave an identical pattern which distinguished them from all individuals of the other five NKG2F species. This pattern was comprised of positive reactions for all the Four NKG2F alleles were obtained from the analysis of five human genes tested, with the exception of Pt-NKG2CI. Species-specific patterns of typing were also seen for the five orangutans and three

donors and five alleles from the four chimpanzees (Fig. 1), result- Downloaded from ing in three NKG2F and five Pt-NKG2F allotypes (Figs. 2 and 3). gibbons tested. Whereas the gibbons were negative for NKG2C The human NKG2F01 allele is identical to the original sequence and NKG2D (as well as CD94), the orangutans were negative for from Plougastel and Trowsdale (44) (GenBank accession no. NKG2C but positive for NKG2D. U96845) and NKG2F02 is identical to the GenBank reference se- It must be emphasized that negative typing reactions do not quence NM 013431 originally described by Glienke et al. (10). show that a gene is absent, merely that at least one of the target At least two humans and three chimpanzees appear to be NKG2F sequences used for PCR is not present. Thus the results of the http://www.jimmunol.org/ heterozygotes (Fig. 1). Clones corresponding to three NKG2F al- typing provide a measure of divergence from the human and chim- leles were obtained from Elwood, the chimpanzee from whom panzee sequences used to design the typing system. Most striking NKG2E could not be amplified. The alleles Pt-NKG2F04 and 05 is the phylogenetic conservation of NKG2A and CD94 compared differ from the others by a 3-bp deletion in the 5Ј untranslated with NKG2C and NKG2D, a result that is consistent with and region and a 4-bp deletion in the 3Ј untranslated region (data not complementary to the comparison of cDNA sequences in human shown). and common chimpanzee. Also impressive is the variability of the Human and chimpanzee NKG2F have ϳ98.6% sequence simi- NKC genes within the common chimpanzee. The actual heteroge- larity, differing by 5- to 9-bp differences between the two species. neity and polymorphism of the chimpanzee NKC is probably much

Striking is that only 2 of the 16 polymorphic positions are species greater than is apparent from this analysis because only individuals by guest on October 2, 2021 specific (Fig. 2). who are negative on both haplotypes will give the negative reac- Three alternatively spliced variants of Pt-NKG2F01 were en- tions needed to be distinguished from pattern 1. countered, all involving exon 4 and the use of cryptic splice sites. Variant 1 lacked 183 bp within exon 4, but the deletion was to the Discussion 3Ј side of the NKG2F stop codon. Variant 2 lacked 68 bp at the 5Ј Upon phylogenetic analysis, the six related human NKC genes we end of exon 4, which changes the reading frame to the one used by studied subdivide into three lineages. CD94 and NKG2D comprise NKG2A, C, and E in this exon. Variant 3 lacked the same internal two of the lineages; the third lineage includes NKG2A, NKG2C, exon 4 sequence as variant 1 but also lacked 49 bp from the 5Ј end NKG2E, and NKG2F (Fig. 7A). Functions for the CD94, NKG2A, of the exon and is prematurely terminated. NKG2C, and NKG2D polypeptides have been identified, but none In summary, the NKG2F gene has a pattern of variation that yet for NKG2E and NKG2F. Our assessment of polymorphism in distinguishes it from the other NKC genes examined. It combines the human genes indicates that all of the genes are relatively con- relative conservation between the two species with greater poly- served with allotypes differing by no more than two amino acid morphism within each species. substitutions (Figs. 2 and 3). No polymorphism in human CD94 and NKG2A was found, in- Haplotypic variation in the chimpanzee NKC dicating that the heterodimeric, inhibitory receptor formed by these Based upon the sequences of the human and chimpanzee CD94 two polypeptides with specificity for HLA-E (13Ð19, 21, 22) is and NKG2A, C, and D genes, a system of PCR sequence-specific largely conserved within the human population. CD94 also forms primers typing of genomic DNA was developed and used to ana- an activating receptor for HLA-E by association with NKG2C (22, lyze samples obtained from 11 human donors and 69 apes repre- 23). We identified two NKG2C allotypes differing at two amino ␤ senting six taxonomic species in total. Typing for the 2m gene acid residues. Both allotypes were well represented in the sample was included as a positive control (Fig. 6B). Typing for NKG2C population, demonstrating modest polymorphism in this receptor, involved three reactions: one aimed at a generic typing for perhaps at a level comparable to that of the oligomorphic HLA-E

short cytoplasmic tail of CD94. Different translational frames are also shown for the carboxyl-terminal ends of Pt-NKG2CII, (Pt-)NKG2E, (Pt-)NKG2F, and NKG2H. Elements of secondary structure in the CRD of CD94 were based on the structure determined by Boyington et al. (65): loops are designated by L plus a number, ␤ strands by ␤ plus a number, and an ␣ helix as ␣1. Shaded positions show the differences between the alternatively spliced forms of the CD94 gene product (CD94 and CD94B), interspecies variations, and allelic polymorphisms in NKG2 loci. A full-length cDNA clone was not obtained for Pt-NKG2E05 and the sequence shown is for a splice variant lacking exon 4. Boxed are the immunoreceptor tyrosine-based inhibitory motifs (V/IxYxxL) in the cytoplasmic region and positively charged residues in the transmembrane region. 248 HUMAN AND CHIMPANZEE CD94 AND NKG2 GENE VARIATION

ligand (45). Two NKG2D allotypes that differ at a single residue were identified and both were well represented in the sample pop- ulation. The ligand for NKG2D is the polymorphic MHC class I-like polypeptide MIC (24, 25, 45, 46) and the divergent class I-related ULBPs (47), which therefore has potential to create func- tional heterogeneity in the ligand-receptor interaction. In parallel with the analysis of human diversity in NKC genes we performed a similar analysis in common chimpanzee. For each of the human genes we identified chimpanzee counterparts that are Ͼ96% identical in nucleotide sequence. Thus, as a group these NKC lectin-like receptors of NK cells are more conserved than the KIR genes encoding Ig-like receptors (Fig. 8). With the exception of NKG2C, each of the human NKC genes corresponds to a single chimpanzee NKC gene which we interpret to be its ortholog. In contrast, the single human NKG2C gene is equally related to two chimpanzee paralogs, one of which (Pt-NKG2CI) encodes a pro- tein like human NKG2C, whereas the other (Pt-NKG2CII) has a frameshift that produces a shorter protein which lacks an extracel- lular CRD. Both Pt-NKG2C genes were demonstrated to be inde- Downloaded from pendently and clonally expressed in chimpanzee NK cell clones (Fig. 6A). More polymorphism was observed in the NKC genes of four chimpanzees than was observed in the larger panel of human donors. However, the numbers of allotypes were few, as were the number of amino acid residues that distinguished them. Thus in both human and chimpanzee the NKC genes are conserved. Genes related to human and chimpanzee CD94 and NKG2 have http://www.jimmunol.org/ been described for rhesus monkey (48, 49), mouse (50Ð55), and rat (56Ð58). Phylogenetic comparison of the genes from the five spe- cies gives trees with major branches corresponding to the three lineages: CD94, NKG2D, and a third lineage consisting of human NKG2A, C, E, F, and related genes from other species (Fig. 7). All three major branches bifurcate into sublineages of rodent and pri- mate genes (Fig. 7A). Whereas the trees support a model in which

the CD94 genes and the NKG2D genes in the various species are by guest on October 2, 2021 orthologs, that is not the case for the third lineage, in which several genes from each species are represented. Even within the rodent sublineage orthologous relationships are uncertain because further bifurcation segregates the rat and mouse genes. Vance et al. (53) proposed two models for the evolution of the genes named NKG2A, C, and E in humans and mice: in the first model they are not orthologs and derived from lineage-specific pathways of gene duplication, deletion, and divergence; in the sec- ond model they are orthologs but have acquired strong species- specific characteristics by concerted evolution through gene con- versions. In either case the process led to order-specific and species-specific divergence. Distinguishing the two alternatives is difficult, but the latter model was favored by Vance et al. (53) for the following reasons. First, they considered it unlikely to be co- incidental for both activating and inhibitory genes to be inherited in both mammalian lineages; second, the overall gene orders of NKG2A, C, and E are preserved in the human and mouse NKC (12, FIGURE 4. Alternatively spliced variants of CD94 and NKG2 cDNAs. 53); and third, the 5Ј ends of NKG2A (inhibitory form) and A, Alternatively spliced variants identified in this study. Genomic struc- NKG2C (activating form) are divergent (52, 53), indicating that the tures of the common chimpanzee genes are predicted from those of the two genes have been separated for a long time. Providing some human genes. Exons are depicted as numbered boxes and introns are not contrast to this view is our comparison of human and chimpanzee drawn to scale. Exon boxes are shaded to reflect homology between genes, NKC genes showing how genes within the third lineage can evolve with filled boxes indicating high homology to NKG2A and gray boxes by species-specific duplications over short periods of evolutionary indicating low or no homology. Open boxes represent untranslated regions time. Thus the two chimpanzee Pt-NKG2C genes are both paralogs of cDNA and dashed boxes represent pseudo exons in human genes. Below of the human NKG2C and neither is an ortholog. Moreover, the the gene structures are representations of the various cDNAs isolated from this study; also included is the human NKG2H coding cDNA (42). B,A apparent haplotypic diversity of the Pt-NKG2C genes seen in com- schematic showing how the CD94 and CD94B variants are generated using mon chimpanzees indicates that this subregion of the NKC has two alternative “AG” acceptor splice sites at the 3Ј end of intron 4. As a repeatedly been a target for evolutionary change. Thus a model in consequence, CD94B has an additional residue, a glutamine (Q), which is which the human and mouse NKG2A, C, and E genes are paralogs not present in CD94. rather than orthologs cannot be ruled out. The Journal of Immunology 249

FIGURE 5. Common chimpanzee Pt-NKG2CI and CII are equally diverged from human NKG2C. Numbers of nucleotide differences between sequence pairs are shown above the diagonal, while percentages of nucleotide differences are shown below. In parentheses are the maximum lengths of the regions compared; insertions and deletions were excluded from analysis. In the comparison in A, sequences downstream of the premature termination codon of Pt-NKG2CII alleles, corresponding to the coding regions of NKG2C and Pt-NKG2CI, were included. In the comparison of the 3Ј untranslated region in B, Pt-NKG2CII02 was not included because only 190 bp of 3Ј untranslated region was available due to the positioning of the specific oligonucleotide primer used to amplify this allele. The human sequences we determined were similarly truncated in the 3Ј untranslated region, so the NKG2C01 sequence used is from a previously reported genomic sequence (GenBank accession no. AJ001684). Downloaded from Further evidence for species-specific evolution within the Although the CD94 and NKG2 genes of the NKC are generally NKC is seen in a tree made from just the primate NKG2A, C, E, older, less variable, and slower evolving than the KIR gene family F, and related sequences (Fig. 7B). This tree contains three of the LRC (Fig. 8), phylogenetic comparisons suggest that they distinguishable groups of sequences, corresponding to the have nontrivially diverged during the course of mammalian evo- NKG2A sequences, the NKG2F sequences, and the combination lution. That is clearly the case for the ligands recognized by these of NKG2C and NKG2E sequences. The two rhesus monkey lectin-like receptors. Thus the Qa-1 class I ligand for mouse CD94: http://www.jimmunol.org/ (Macaca mulatta) sequences in the NKG2A group (48) are con- NKG2A and CD94:NKG2C receptors does not in its sequence siderably diverged, both from human and chimpanzee NKG2A have particular affinity with the HLA-E class I ligand for human and from themselves (11 amino acid differences). Either there is CD94:NKG2A and CD94:NKG2C receptors (52, 53, 60, 61). Even a polymorphism greater than is seen in human and chimpanzee more different are the divergent MHC class I-related ligands for at a single Mm-NKG2A locus or there are two Mm-NKG2A NKG2D, which comprise MIC (24, 25) and ULBPs (47) in hu- genes. It is also intriguing that the rhesus monkey MHC-E is mans, and the retinoic acid early inducible RAE-1 and H60 in mice oligomorphic or polymorphic (59). Similarly, within the (62, 63). NKG2F group there are two divergent rhesus monkey NKG2F- Comparison of the human and chimpanzee NKC genes indicates like sequences (48) of which one is closer to human and chim- that they have evolved under different selection pressures: CD94 is by guest on October 2, 2021 panzee NKG2F. Within the NKG2C group, the eight Mm- most conserved, followed by NKG2D, NKG2A, and NKG2F, with NKG2C sequences (48) form a species-specific subgroup that NKG2C and NKG2E being most divergent (Fig. 8) (35). The hu- does not interleaf with the human and chimpanzee NKG2C se- man and common chimpanzee CD94 have only 0.4% nucleotide quences. All these data point to the NKC of the rhesus monkey difference, well below the average ϳ1.24% variation as suggested having significant differences in gene content and haplotypic for intergenic regions of human and chimpanzee genes (64). variation from the NKCs of human and chimpanzee. NKG2A and NKG2D also appear to be under purifying selection.

FIGURE 6. Typing for the expression and distribution of NKC genes. A, The results of RT-PCR typing for the transcription of Pt-NKG2CI and Pt-NKG2CII genes in NK cell clones isolated from the common chimpanzee Cathy (35); typing for NKG2D transcription was included as control. B, ␤ Genomic DNA typing of different ape species for NKG2A, C, D, and CD94 genes; typing for 2m was included as positive control. In parentheses are the numbers of individuals included for analysis. Only 7 of the 11 pygmy chimpanzee were typed for NKG2D. The percentage of positive typing results within each group is shown. 250 HUMAN AND CHIMPANZEE CD94 AND NKG2 GENE VARIATION Downloaded from http://www.jimmunol.org/

FIGURE 7. Phylogenetic relationships of CD94 and NKG2 in primates and rodents. Phylogenetic analyses were performed for selected CD94 and NKG2 cDNA sequences from human (bold type), common chimpanzee (Pt-), rhesus macaque (Mm-) (48), mouse (Mus-) (50Ð53) and rat (Rat-) (55Ð57) using the neighbor-joining method. Shown are consensus phylogenetic trees (dendrograms) allowing polytomy. Nodes were calculated from 1000 bootstrap Ͼ replications and corresponding bootstrap values ( 70) were noted on branches. The tree in A includes representative CD94 and NKG2 sequences from by guest on October 2, 2021 primates and rodents. The tree was made from sequences of exon 5 of NKG2A and from sequences of homologous exons in CD94 and other NKG2 genes; these exons are those most highly conserved during gene divergence (Fig. 4). A 55% majority-rule consensus tree is shown. Entire coding-region cDNA sequences were used in a separate analysis (data not shown) and did not alter the overall topology. The tree in B includes NKG2A, C, E, F sequences within the primate lineage. The sequences compared were exons 2Ð6ofNKG2A and the corresponding sequences in other genes. To include as much informative data as possible for this analysis, the corresponding pseudo exon 5 in the human NKG2F was included; no pseudo exon sequence was available for the common chimpanzee gene and only one Pt-NKG2F allele is used. A 50% majority-rule consensus tree is shown.

For CD94 and NKG2A, the identical inhibitory MHC class I spec- (68) have recently shown that a gene related to human Ly49L is ificity of the human and chimpanzee CD94:NKG2A receptors sup- present in common chimpanzee as well as in other primates ports this mode of selection (35). The human and common chim- (gorilla, orangutan, gibbon, baboon, and African green monkey) panzee NKG2C and NKG2E genes differ by 3.2Ð3.4% in and other mammalian orders. In contrast, genes related to the nucleotide sequences, values higher than expected for the diver- expanded rodent Ly49 family were not detectable by Southern gence of human and chimpanzee genes under neutral selection, and blot analysis in non-rodent species. Partial sequence analysis similar to those KIR genes which are orthologous in chimpanzee showed that common chimpanzee Ly49L has the same inacti- and human (Fig. 8). Most nucleotide substitutions that distinguish vating mutations as the human gene, whereas that was not so for NKG2C and Pt-NKG2CI change residues in the CRD loops (Figs. the other primates. The extent to which these differences are 2 and 3) predicted to interact with ligand (65). These differences fixed in humans, chimpanzees, and other species was not ad- might affect the ligand-binding specificity and be a consequence of dressed, as analysis was confined to individuals of a species. selection upon activating receptors by pathogens. Such selection For baboon, a complete cDNA sequence for Ly49L was deter- might also be the cause of the greater haplotypic diversity of mined (ϳ95% nucleotide identity with the human Ly49L in the NKG2C compared with other NKC genes, as seen both within spe- cies (as in the common chimpanzee) and between species. coding region) and it was shown to be expressed in baboon The rodent NKC contains a diverse family of Ly49 genes that lymphocytes along with transcripts from several KIR genes. encode lectin-like receptors specific for polymorphic determi- These observations raise the possibility that Ly49L encodes a nants of classical MHC class I molecules (66). In humans the functional receptor in some primate species and that its inacti- Ly49 family is represented by a single nonfunctional gene, vation in humans and chimpanzee may be exceptional. Also Ly49L (67), while KIR provide functions analogous to those of worthy of consideration is that inactivation of the Ly49L gene Ly49 in mice. In our investigation we did not examine the struc- may not be a property of all human and chimpanzee NKC ture and diversity of the Ly49L gene. However, Mager et al. haplotypes. The Journal of Immunology 251

FIGURE 8. Polymorphism and variation among human and common chimpanzee CD94, NKG2, and KIR. The within-species and between-species variation in coding-region nucleotide sequences are shown as percentage differences. Within a species, the average pairwise differences between alleles are shown. A dash indicates that only one allele has been described. Be- tween-species variation is the average of all possible pairwise comparisons between alleles of the two spe- cies. Orthologous genes are on the same line. The ma- jority of human and chimpanzee KIR lineage III genes are not orthologous (35); they were considered to-

gether and compared only as species-specific groups Downloaded from (shaded). The exon sequences encoding the D0 do- main in chimpanzee KIR3D of this group were not included in the interspecies comparisons because their human counterpart lack this domain (35). http://www.jimmunol.org/

Acknowledgments 12. Sobanov, Y., J. Glienke, C. Brostjan, H. Lehrach, F. Francis, and E. Hofer. 1999. by guest on October 2, 2021 Linkage of the NKG2 and CD94 receptor genes to D12S77 in the human natural We thank the staff at Yerkes Regional Primate Research Center of Emory killer gene complex. Immunogenetics 49:99. University (especially Rickie Bass) and the Laboratory for Experimental 13. Phillips, J. H., C. Chang, J. Mattson, J. E. Gumperz, P. Parham, and L. L. Lanier. Medicine and Surgery in Primates for assistance in obtaining chimpanzee 1996. CD94 and a novel associated protein (94AP) form a NK cell receptor tissue samples. We also thank Dr. Erin Adams for generating the chim- involved in the recognition of HLA-A, HLA-B, and HLA-C allotypes. Immunity 5:163. panzee B cell lines and Drs. Stewart Cooper and Ann Erickson for the RNA 14. Lazetic, S., C. Chang, J. P. Houchins, L. L. Lanier, and J. H. Phillips. 1996. sample from the chimpanzee Todd. Human NK cell receptors involved in MHC class I recognition are disulfide- linked heterodimers of CD94 and NKG2 subunits. J. Immunol. 157:4741. 15. Braud, V. M., D. S. Allan, C. A. O’Callaghan, K. Soderstrom, A. D’Andrea, References G. S. Ogg, S. Lazetic, N. T. Young, J. I. Bell, J. H. Phillips, et al. 1998. HLA-E 1. Lanier, L. L. 1998. NK cell receptors. Annu. Rev. Immunol. 16:359. binds to natural killer cell receptors CD94/NKG2A, B and C. Nature 391:795. 2. Klein, J. 1986. Natural History of the Major Histocompatibility Complex. Wiley, 16. Borrego, F., M. Ulbrecht, E. H. Weiss, J. E. Coligan, and A. G. Brooks. 1998. New York. Recognition of human histocompatibility leukocyte antigen (HLA)-E complexed 3. Suto, Y., K. Maenaka, T. Yabe, M. Hirai, K. Tokunaga, K. Tadok, and T. Juji. with HLA class I signal sequence-derived peptides by CD94/NKG2 confers pro- 1996. Chromosomal localization of the human natural killer cell class I receptor tection from natural killer cell-mediated lysis. J. Exp. Med. 187:813. family genes to 19q13.4 by fluorescence in situ hybridization. Genomics 35:270. 17. Lee, N., M. Llano, M. Carretero, A. Ishitani, F. Navarro, M. Lopez-Botet, and 4. Wagtmann, N., S. Rojo, E. Eichler, H. Mohrenweiser, and E. O. Long. 1997. A D. E. Geraghty. 1998. HLA-E is a major ligand for the natural killer inhibitory new human gene complex encoding the killer cell inhibitory receptors and related receptor CD94/NKG2A. Proc. Natl. Acad. Sci. USA 95:5199. monocyte/macrophage receptors. Curr. Biol. 7:615. 18. Llano, M., N. Lee, F. Navarro, P. Garcia, J. P. Albar, D. E. Geraghty, and 5. Wende, H., M. Colonna, A. Ziegler, and A. Volz. 1999. Organization of the M. Lopez-Botet. 1998. HLA-E-bound peptides influence recognition by inhibi- leukocyte receptor cluster (LRC) on human chromosome 19q13.4. Mamm. Ge- tory and triggering CD94/NKG2 receptors: preferential response to an HLA-G- nome 10:154. derived nonamer. Eur. J. Immunol. 28:2854. 6. Wilson, M. J., M. Torkar, A. Haude, S. Milne, T. Jones, D. Sheer, S. Beck, and J. Trowsdale. 2000. Plasticity in the organization and sequences of human KIR/ 19. Brooks, A. G., F. Borrego, P. E. Posch, A. Patamawenu, C. J. Scorzelli, ILT gene families. Proc. Natl. Acad. Sci. USA 97:4778. M. Ulbrecht, E. H. Weiss, and J. E. Coligan. 1999. Specific recognition of 7. Houchins, J. P., T. Yabe, C. McSherry, and F. H. Bach. 1991. DNA sequence HLA-E, but not classical, HLA class I molecules by soluble CD94/NKG2A and analysis of NKG2, a family of related cDNA clones encoding type II integral NK cells. J. Immunol. 162:305. membrane proteins on human natural killer cells. J. Exp. Med. 173:1017. 20. Vivier, E., and M. Daeron. 1997. Immunoreceptor tyrosine-based inhibition mo- 8. Yabe, T., C. McSherry, F. H. Bach, P. Fisch, R. P. Schall, P. M. Sondel, and tifs. Immunol. Today 18:286. J. P. Houchins. 1993. A multigene family on human chromosome 12 encodes 21. Carretero, M., C. Cantoni, T. Bellon, C. Bottino, R. Biassoni, A. Rodriguez, natural killer-cell lectins. Immunogenetics 37:455. J. J. Perez-Villar, L. Moretta, A. Moretta, and M. Lopez-Botet. 1997. The CD94 9. Renedo, M., I. Arce, A. Rodriguez, M. Carretero, L. L. Lanier, M. Lopez-Botet, and NKG2-A C-type lectins covalently assemble to form a natural killer cell and E. Fernandez-Ruiz. 1997. The human natural killer gene complex is located inhibitory receptor for HLA class I molecules. Eur. J. Immunol. 27:563. on chromosome 12p12Ðp13. Immunogenetics 46:307. 22. Houchins, J. P., L. L. Lanier, E. C. Niemi, J. H. Phillips, and J. C. Ryan. 1997. 10. Glienke, J., Y. Sobanov, C. Brostjan, C. Steffens, C. Nguyen, H. Lehrach, NK cell cytolytic activity is inhibited by NKG2-A and activated by NKG2-C. E. Hofer, and F. Francis. 1998. The genomic organization of NKG2C, E, F, and J. Immunol. 158:3603. D receptor genes in the human natural killer gene complex. Immunogenetics 23. Cantoni, C., R. Biassoni, D. Pende, S. Sivori, L. Accame, L. Pareti, 48:163. G. Semenzato, L. Moretta, A. Moretta, and C. Bottino. 1998. The activating form 11. Plougastel, B., and J. Trowsdale. 1998. Sequence analysis of a 62-kb region of CD94 receptor complex: CD94 covalently associates with the Kp39 protein overlapping the human KLRC cluster of genes. Genomics 49:193. that represents the product of the NKG2-C gene. Eur. J. Immunol. 28:327. 252 HUMAN AND CHIMPANZEE CD94 AND NKG2 GENE VARIATION

24. Bauer, S., V. Groh, J. Wu, A. Steinle, J. H. Phillips, L. L. Lanier, and T. Spies. 46. Fodil, N., L. Laloux, V. Wanner, P. Pellet, G. Hauptmann, N. Mizuki, H. Inoko, 1999. Activation of NK cells and T cells by NKG2D, a receptor for stress- T. Spies, I. Theodorou, and S. Bahram. 1996. Allelic repertoire of the human inducible MICA. Science 285:727. MHC class I MICA gene. Immunogenetics 44:351. 25. Groh, V., R. Rhinehart, J. Randolph-Habecker, M. S. Topp, S. R. Riddell, and 47. Cosman, D., N. Fanger, L. Borges, M. Kubin, W. Chin, L. Peterson, and T. Spies. 2001. Costimulation of CD8␣␤ T cells by NKG2D via engagement by M. L. Hsu. 1997. A novel immunoglobulin superfamily receptor for cellular and MIC induced on virus-infected cells. Nat. Immunol. 2:255. viral MHC class I molecules. Immunity 7:273. 26. Cosman, D., J. Mullberg, C. L. Sutherland, W. Chin, R. Armitage, W. Fanslow, 48. LaBonte, M. L., D. B. Levy, and N. L. Letvin. 2000. Characterization of rhesus M. Kubin, and N. J. Chalupny. 2001. ULBPs, novel MHC class I-related mole- monkey CD94/NKG2 family members and identification of novel transmem- cules, bind to CMV glycoprotein UL16 and stimulate NK cytotoxicity through brane-deleted forms of NKG2-A, B, C, and D. Immunogenetics 51:496. the NKG2D receptor. Immunity 14:123. 49. Kravitz, R. H., R. L. Grendell, I. I. Slukvin, and T. G. Golos. 2001. Selective 27. Wu, J., Y. Song, A. B. Bakker, S. Bauer, T. Spies, L. L. Lanier, and J. H. Phillips. expression of NKG2-A and NKG2-C mRNAs and novel alternative splicing of 5Ј 1999. An activating immunoreceptor complex formed by NKG2D and DAP10. exons in rhesus monkey decidua. Immunogenetics 53:69. Science 285:730. 50. Vance, R. E., D. M. Tanamachi, T. Hanke, and D. H. Raulet. 1997. Cloning of 28. Lanier, L. L., B. Corliss, J. Wu, and J. H. Phillips. 1998. Association of DAP12 a mouse homolog of CD94 extends the family of C-type lectins on murine natural with activating CD94/NKG2C NK cell receptors. Immunity 8:693. killer cells. Eur. J. Immunol. 27:3236. 29. Long, E. O., and S. Rajagopalan. 2000. HLA class I recognition by killer cell 51. Ho, E. L., J. W. Heusel, M. G. Brown, K. Matsumoto, A. A. Scalzo, and Ig-like receptors. Semin. Immunol. 12:101. W. M. Yokoyama. 1998. Murine Nkg2d and Cd94 are clustered within the natural 30. Uhrberg, M., N. M. Valiante, B. P. Shum, H. G. Shilling, K. Lienert-Weidenbach, killer complex and are expressed independently in natural killer cells. Proc. Natl. B. Corliss, D. Tyan, L. L. Lanier, and P. Parham. 1997. Human diversity in killer Acad. Sci. USA 95:6320. cell inhibitory receptor genes. Immunity 7:753. 52. Vance, R. E., J. R. Kraft, J. D. Altman, P. E. Jensen, and D. H. Raulet. 1998. 31. Valiante, N. M., M. Uhrberg, H. G. Shilling, K. Lienert-Weidenbach, Mouse CD94/NKG2A is a natural killer cell receptor for the nonclassical major K. L. Arnett, A. D’Andrea, J. H. Phillips, L. L. Lanier, and P. Parham. 1997. histocompatibility complex (MHC) class I molecule Qa-1(b). J. Exp. Med. 188: Functionally and structurally distinct NK cell receptor repertoires in the periph- 1841. eral blood of two human donors. Immunity 7:739. 53. Vance, R. E., A. M. Jamieson, and D. H. Raulet. 1999. Recognition of the class 32. Shilling, H. G., K. Lienert-Weidenbach, N. M. Valiante, M. Uhrberg, and Ib molecule Qa-1(b) by putative activating receptors CD94/NKG2C and CD94/ P. Parham. 1998. Evidence for recombination as a mechanism for KIR diversi- NKG2E on mouse natural killer cells. J. Exp. Med. 190:1801. Downloaded from fication. Immunogenetics 48:413. 54. Silver, E. T., J. C. Lau, and K. P. Kane. 1999. Molecular cloning of mouse 33. Gardiner, C. M., L. A. Guethlein, H. G. Shilling, M. Pando, W. H. Carr, NKG2A and C. Immunogenetics 49:727. R. Rajalingam, C. Vilches, and P. Parham. 2001. Different NK cell surface phe- 55. Lohwasser, S., P. Hande, D. L. Mager, and F. Takei. 1999. Cloning of murine notypes defined by the DX9 antibody are due to KIR3DL1 gene polymorphism. NKG2A, B and C: second family of C-type lectin receptors on murine NK cells. J. Immunol. 166:2992. Eur. J. Immunol. 29:755. 34. Rajalingam, R., C. M. Gardiner, F. Canavez, C. Vilches, and P. Parham. 2001. 56. Dissen, E., S. F. Berg, I. H. Westgaard, and S. Fossum. 1997. Molecular char- Identification of seventeen novel KIR variants: fourteen of them from two non- acterization of a gene in the rat homologous to human CD94. Eur. J. Immunol. Caucasian donors. Tissue Antigens 57:22. 27:2080.

35. Khakoo, S. I., R. Rajalingam, B. P. Shum, K. Weidenbach, L. Flodin, D. G. Muir, 57. Berg, S. F., E. Dissen, I. H. Westgaard, and S. Fossum. 1998. Two genes in the http://www.jimmunol.org/ F. Canavez, S. L. Cooper, N. M. Valiante, L. L. Lanier, and P. Parham. 2000. rat homologous to human NKG2. Eur. J. Immunol. 28:444. Rapid evolution of NK cell receptor systems demonstrated by comparison of 58. Berg, S. F., E. Dissen, I. H. Westgaard, and S. Fossum. 1998. Molecular char- chimpanzees and humans. Immunity 12:687. acterization of rat NKR-P2, a lectin-like receptor expressed by NK cells and 36. Rajalingam, R., M. Hong, E. J. Adams, B. P. Shum, L. A. Guethlein, and resting T cells. Int. Immunol. 10:379. P. Parham. 2001. Short KIR haplotypes in pygmy chimpanzee (bonobo) resemble 59. Boyson, J. E., S. N. McAdam, A. Gallimore, T. G. Golos, X. Liu, F. M. Gotch, the conserved framework of diverse human KIR haplotypes. J. Exp. Med. 193: A. L. Hughes, and D. I. Watkins. 1995. The MHC E locus in macaques is poly- 135. morphic and is conserved between macaques and humans. Immunogenetics 41: 37. Lawlor, D. A., E. Warren, F. E. Ward, and P. Parham. 1990. Comparison of class 59. I MHC alleles in humans and apes. Immunol. Rev. 113:147. 60. Salcedo, M., P. Bousso, H. G. Ljunggren, P. Kourilsky, and J. P. Abastado. 1998. 38. Sambrook, J., E. F. Fritsch, and T. Maniatis. 1989. Molecular Cloning: A Lab- The Qa-1b molecule binds to a large subpopulation of murine NK cells. Eur. oratory Manual. Cold Spring Harbor Lab. Press, Cold Spring Harbor, NY. J. Immunol. 28:4356. 39. Shapiro, M. B., and P. Senapathy. 1987. RNA splice junctions of different classes 61. Kurepa, Z., C. A. Hasemann, and J. Forman. 1998. Qa-1b binds conserved class by guest on October 2, 2021 of eukaryotes: sequence statistics and functional implications in gene expression. I leader peptides derived from several mammalian species. J. Exp. Med. 188:973. Nucleic Acids Res. 15:7155. 62. Cerwenka, A., A. B. Bakker, T. McClanahan, J. Wagner, J. Wu, J. H. Phillips, 40. Furukawa, H., T. Yabe, K. Watanabe, R. Miyamoto, T. Akaza, K. Tadokoro, and L. L. Lanier. 2000. Retinoic acid early inducible genes define a ligand family S. Tohma, T. Inoue, K. Yamamoto, and T. Juji. 1998. An alternatively spliced for the activating NKG2D receptor in mice. Immunity 12:721. form of the human CD94 gene. Immunogenetics 48:87. 63. Diefenbach, A., A. M. Jamieson, S. D. Liu, N. Shastri, and D. H. Raulet. 2000. 41. Plougastel, B., T. Jones, and J. Trowsdale. 1996. Genomic structure, chromosome Ligands for the murine NKG2D receptor: expression by tumor cells and activa- location, and alternative splicing of the human NKG2A gene. Immunogenetics tion of NK cells and macrophages. Nat. Immunol. 1:119. 44:286. 64. Chen, F. C., and W. H. Li. 2001. Genomic divergences between humans and 42. Bellon, T., A. B. Heredia, M. Llano, A. Minguela, A. Rodriguez, M. Lopez-Botet, other hominoids and the effective population size of the common ancestor of and P. Aparicio. 1999. Triggering of effector functions on a CD8ϩ T cell clone humans and chimpanzees. Am. J. Hum. Genet. 68:444. upon the aggregation of an activatory CD94/kp39 heterodimer. J. Immunol. 162: 65. Boyington, J. C., A. N. Riaz, A. Patamawenu, J. E. Coligan, A. G. Brooks, and 3996. P. D. Sun. 1999. Structure of CD94 reveals a novel C-type lectin fold: implica- 43. Adamkiewicz, T. V., C. McSherry, F. H. Bach, and J. P. Houchins. 1994. Natural tions for the NK cell-associated CD94/NKG2 receptors. Immunity 10:75. killer lectin-like receptors have divergent carboxy-termini, distinct from C-type 66. Takei, F., J. Brennan, and D. L. Mager. 1997. The Ly-49 family: genes, proteins lectins. Immunogenetics 39:218. and recognition of class I MHC. Immunol. Rev. 155:67. 44. Plougastel, B., and J. Trowsdale. 1997. Cloning of NKG2-F, a new member of 67. Westgaard, I. H., S. F. Berg, S. Orstavik, S. Fossum, and E. Dissen. 1998. Iden- the NKG2 family of human natural killer cell receptor genes. Eur. J. Immunol. tification of a human member of the Ly-49 multigene family. Eur. J. Immunol. 27:2835. 28:1839. 45. Marsh, S. G., J. G. Bodmer, E. D. Albert, W. F. Bodmer, R. E. Bontrop, 68. Mager, D. L., K. L. McQueen, V. Wee, and J. D. Freeman. 2001. Evolution of B. Dupont, H. A. Erlich, J. A. Hansen, B. Mach, W. R. Mayr, et al. 2001. natural killer cell receptors. coexistence of functional Ly49 and KIR genes in Nomenclature for factors of the HLA system, 2000. Tissue Antigens 57:236. baboons. Curr. Biol. 11:626.