Genetic control of innate immune responses against ... - Nature

Genes and Immunity (2002) 3, 250–262  2002 Nature Publishing Group All rights reserved 1466-4879/02 $25.00 www.nature.com/gene

REVIEW

Genetic control of innate immune responses against cytomegalovirus: MCMV meets its match JR Webb, SH Lee and SM Vidal Department of Biochemistry, Microbiology and Immunology, University of Ottawa, Ottawa, Ontario, K1H 8M5, Canada

Cytomegalovirus (CMV) is a widespread pathogen that is responsible for severe disease in immunocompromised individuals and probably, associated with vascular disease in the general population. There is increasing evidence that cells of the innate immune system play a key role in controlling this important pathogen. This is particularly evident in the experimental murine CMV (MCMV) model of infection which has revealed an important role for natural killer (NK) cells in controlling early viral replication after infection with MCMV. In this model, different strains of inbred mice exhibit striking differences in their level of susceptibility to MCMV infection. Genetic studies, performed almost 10 years ago, revealed that this pattern of susceptibility/resistance can be attributed to a single genetic locus termed Cmv1 and recently several groups that have been working on the mapping and identification of Cmv1 have met with success. Interestingly, Cmv1 is allelic to a member of the Ly49 gene family, which encode activating or inhibitory transmembrane receptors present on the surface of NK cells. All Ly49 receptors characterized to date interact with MHC class I molecules on potential target cells, resulting in the accumulation of signals to the NK to either ‘kill’ or ‘ignore’ the cell based upon the repertoire of MHC class I molecules expressed. The identification of Cmv1 as Ly49H, a stimulatory member of the Ly49 family, adds an interesting twist to the Ly49 story. Although the ligand of Ly49H is not yet known, there is already compelling evidence that the ligand is upregulated on virally infected cells, resulting in specific activation of Ly49H-expressing NK cells. This review provides an historical perspective of the MCMV infection model from its inception to the discovery of the gene responsible for the phenotype and provides a basis for further experiments aimed at understanding the role of NK cells, in general, and Ly49H, in particular, in mediating resistance to cytomegalovirus. Genes and Immunity (2002) 3, 250–262. doi:10.1038/sj.gene.6363876 Keywords: murine cytomegalovirus; Ly49 receptors; natural killer cells

Cytomegalovirus infection in humans Human cytomegalovirus (HCMV) is a ␤-herpesvirus that infects more than 70% of the world population regardless of geographical location or gender.1,2 Following primary infection, HCMV establishes long-term latency with intermittent reactivation.1,3 Usually asymptomatic in immunocompetent individuals, HCMV infections may result in severe or even fatal disease in immunocompromised patients.4–6 In addition, HCMV is the leading cause of congenital viral infection in humans, with an incidence of 0.2–2.2% amongst live births, of which at least 10% will present with birth defects such as hearing loss or mental retardation.7,8 Interestingly, in the general population the clinical outcome of HCMV infection is non-random as polymorphisms in genes of pro-inflammatory cytokines (TNF-␣, IL-1RA) are in association with seropositivity.9 HCMV is also responsible for a substantial fraction of the morbidity and mortality that occurs following organ

Correspondence: SM Vidal, Faculty of Medicine, University of Ottawa, room 4207, 451 Smyth Road, Ottawa, Ontario, Canada, K1H 8M5. E-mail: svidal얀uottawa.ca This work was supported by research grants to SM Vidal from the Canadian Institute of Health Research (CIHR). SM Vidal is a scholar of CIHR. S-H Lee is supported by the CIHR Doctoral Scholarship. Received 22 January 2002; revised and accepted 22 February 2002

transplantation.10–14 In the transplant setting, symptomatic HCMV infection encompasses a range of clinical illnesses including interstitial pneumonia in 50–90% of bone marrow recipients and organ-specific disease in 20– 40% of solid organ transplant recipients.15–19 In transplant recipients susceptibility has been associated with specific genotypes, in particular with alleles of the human leucocyte antigen (HLA) locus, also indicating a genetic component of susceptibility.20 HCMV diseases, including retinitis, colitis, and encephalitis, occur in persons with AIDS and have been associated with decreased survival after diagnosis of HIV infection.21–23 Additionally, the impact of HCMV infection in the general population may have been underrated since clinical and experimental evidence is accumulating that there may be a link between HCMV infection and vascular disease.24–26 For example, HCMV seropositivity is a strong risk factor associated with restenosis following coronary atherectomy27,28 and transplant-associated atherosclerosis in heart transplant patients,29,30 whereas in animal models, murine CMV (MCMV) infection accelerates formation of atheromatous lesions in apoE−/− mice.31,32 Moreover, a related herpesvirus known as Marek’s disease virus (MDV) was found to cause the formation of atherosclerotic lesions in the aortas of chickens and led to accumulation of cholesterol in the aortic wall of MDV-infected chickens.33 There is currently no vaccine available against HCMV

Genetic control of innate immune responses against CMV JR Webb et al

and benefits of treatment with available drugs—foscarnet, gancyclovir and cidofovir—are undermined by toxicity and emergence of drug-resistant strains34–37 indicating the need for additional therapeutic approaches. The serious medical problems associated with HCMV infection have thus stimulated research aimed at understanding the complex interplay of viral and host functions that lead to pathogenesis. In this respect, understanding, at the molecular level, the early immune responses against HCMV infection could provide a rationale for developing alternative therapeutic strategies that either stimulate or exploit host resistance mechanisms.

Natural killer (NK) cells and innate immunity against HCMV Innate immunity is the first arm of defense against infection, providing a rapid (hours to days) and, most of the time, efficacious mechanism to eliminate pathogens. In particular, NK cells represent a component of the innate system that seem to be particularly important during infection with herpesviruses, including CMV.38 NK cells are so named because of their propensity to kill, without prior sensitization, certain neoplastic and virus-infected cells as well as normal allogenic bone marrow-derived cells.39 The crucial role of NK cells in controlling primary CMV infection and reactivation, in humans and in experimental infections in mice, has been recognized for more than 20 years.40 Data from naturally acquired infections indicate that low NK cell cytotoxic activity is linked with increased sensitivity to severe, disseminating, human herpesgroup viral infections, including herpes simplex virus (HSV), Epstein–Barr virus (EBV), and HCMV.38 In graft recipients, low NK cell activity is associated with HCMV reactivation and poor prognosis.41,42 The most direct evidence supporting a role for NK cells in defense against herpesviruses comes from a patient identified with a complete lack of NK cells as well as no spontaneous or IL-2-inducible NK cell cytotoxic function.43 Initially, this young patient presented with an overwhelming Herpes Zoster infection and developed, during the next 5 years, primary life-threatening HCMV infection and a severe HSV infection despite normal antibody and memory T cell functions. In addition, the identification of the molecular defect in X-linked lymphoproliferative (XLP) syndrome provided genetic evidence for the involvement of NK cells in viral defense. XLP patients are highly susceptible to EBV infection which results in fatal infectious mononucleosis, agammaglobulinemia, or B-cell lymphoma as a result of aberrant T cell and NK cell responses.44 The XLP phenotype is the result of a mutation in the gene encoding the SH2 domain protein-1A (SH2D1A),45 also known as SLAM-associated protein (SAP).46 SAP is exclusively expressed on T cells and NK cells where it controls activation pathways initiated by the signaling lymphocyte activation molecule (SLAM), expressed in T cells,46 or 2B4, an NK cell-specific protein implicated in the response to EBV.47–49 Together these data confirm the importance of NK cells during infection by CMV and other herpes viruses.

Mouse models of CMV infection Experimental models of infection of mice with MCMV have provided an excellent tool for the in vivo dissection

of the individual components of the host response that are important in mediating resistance to MCMV infection. Infection of mice with MCMV mimics virus-induced disease in humans in terms of presenting a large spectrum of clinical manifestations. For example, high viral titres in target organs are associated with pneumonitis,50 hepatitis51 and retinitis.52,53 Moreover high viral titres in the spleen or liver may lead to fatal infection.54 There are also models of congenital viral infection by MCMV.55,56 In addition, different strains of inbred mice exhibit striking differences in their level of susceptibility to MCMV infection, as measured by organ-specific viral replication in the first days after infection, disease severity and survival.57,58 These distinct patterns of susceptibility reveal genetic differences between strains that make the mouse an excellent model to examine the heterogeneous responses seen in humans and facilitate a genetic approach to the mapping of susceptibility traits. Furthermore, identification of the genetic determinants of MCMV-susceptibility in the mouse may aid in identification of key host response mechanisms and molecules required for control of viral infection in humans. Mouse studies have provided evidence supporting an important role for NK cells in combating viral disease, particularly during the very early stages of infection.59 Increased sensitivity to infection with MCMV60 and HSV61 was observed after in vivo depletion of NK cells with the monoclonal antibodies PK136 or anti-asialoGM1, which recognize surface antigens preferentially expressed on NK cells. Moreover, the identification of spontaneous mutations as well as the accessibility of the mouse genome to targeted mutagenesis helped to define key molecules participating in host defense against herpesviruses. As shown in Table 1, the vast majority of natural and induced mutations determining susceptibility to herpesvirus affect either the activation or function of NK cells and/or T cells. For example, the absence of chemokines normally expressed early during infection such as MIP1␣ (a mediator of inflammation) or of cytokines of the innate immune response such as IL-12 and IL-18 that stimulate IFN-␥ production by NK cells as well as T cells, result in high MCMV susceptibility.66,67,71 In addition, beige mice, which have a genetic disorder homologous to Chediak–Higashi syndrome (CHS) of man, exhibit defective NK activity despite the presence of normal numbers of NK cells and are highly susceptible to infection with MCMV.79 The beige mutation occurs within a gene encoding a protein called LYST, a regulatory protein required for appropriate vesicle trafficking in granule-containing cells including NK cells.80 Similarly, mice carrying targeted mutations within genes involved in the cytolytic function of NK cells including those encoding perforin64,69,73 and granzyme A/B64 also demonstrate high susceptibility to MCMV infection. Together these results indicate that both immunoregulatory and cytolytic function of T lymphocytes and NK cells are crucial for the control of MCMV infection. The distinct phenotypes of MCMV susceptibility exhibited by different inbred strains have stimulated investigation of the genetic components of resistance. Resistance of adult mice to organ-specific, viral replication early during infection or to acute lethal infection with MCMV is controlled, in part, by genes of the H2 complex that encode classical MHC class I molecules. Studies in H2 congenic mouse strains demonstrated that animals pos-

251

Genes and Immunity


252

Table 1 Natural and induced mutations of genes affecting susceptibility against herpesvirus Gene name

Expression

Virus

References

Recombination-associated genes SCID RAG-2 −/−

T, B T, B

MCMV, HSV-1 MCMV

62, 63 64

Cytokine, chemokine-associated genes IL-2/115R␤ −/− IL-12 −/− IL-18 −/− IFN-␣/␤R −/− IFN-␥ −/− IFN-␥R −/− IRF-1 −/− Lymphotoxin (LT) ␣−/− MIP-1␣ (CCL3) −/−

T, NK B, Macrophage Macrophage NK, Dendritic Cell T, NK Macrophage T, NK T, B T, Monocyte, Mast cell

HSV-2 MCMV MCMV MCMV, MHV-68 MCMV, HSV-1 MCMV MCMV, MHV-68 HSV-1 MCMV

65 66, 67 66 68 63, 69 69 68, 69 70 71

Cytotoxicity-associated genes Beige (Lyst) Perforin −/− (PKO) Granzyme-A −/− Granzyme-B −/− NOS2 (iNOS) −/−

T, NK, Macrophage T, NK T, NK T, NK Macrophage

MCMV, HSV-1 MCMV, HSV-1 MCMV MCMV MCMV, HSV-1

Leukocyte receptor-associated genes Sh2d1a (SAP) −/− Ly49h (Klra8)

T, NK NK

EBV MCMV

sessing the H2k haplotype, particularly at the K/I-A region, are 10 times more resistant than mice of the H2b or H2d haplotype. Susceptibility is a dominant trait as revealed by Mendelian analysis.81–83 The role of the H2 molecules as part of the MCMV receptor was demonstrated by transfection of H2Kb or H2Dd into H2k macrophage and T cell lines and the subsequent potentiation of MCMV infection.84

The Cmv1 locus Despite the well-established role of the H2 complex in MCMV infection, a number of genetic studies have also demonstrated that non-H2 factors contribute to the pattern of resistance and susceptibility to MCMV, particularly during the early stages of infection and in distinct target organs. Specifically, enumeration of MCMV viral titers in the spleens and livers of different inbred mouse strains identified highly resistant and highly susceptible strains in terms of target organ replication.57 These nonH2 genes have been shown to primarily affect NK cell function,60 once again supporting an important role of this cell population in the control of MCMV infection. Genetic studies using these MCMV resistant and susceptible strains as parental strains to generate segregating F2 and backcross populations identified a single locus, Cmv1, which controls viral replication in an autosomal dominant fashion.85 Common inbred strains present two allelic forms for Cmv1: a dominant resistant allele, Cmv1r, expressed in mice of the C57BL/6J background and a recessive susceptible allele, Cmv1s, expressed in BALB/cJ, A/J and DBA/2J strains. Phenotypically, resistant mice display splenic viral titers 103 to 104-fold lower than those found in susceptible mice at day 3 post-infection with a sub-lethal dose of virus. Cmv1 also contributes to MCMV replication in the bone marrow and thymus and plays a role in determining the Genes and Immunity

63, 72 64, 69, 73 64 64 74 75, 76 77, 78

outcome of infection after injection with a lethal dose of virus.86 Studies in radiation chimeras following bone marrow transplantation together with in vivo depletion experiments indicated that the effect of Cmv1 is mediated primarily by NK cells.60 Mechanisms by which NK cells bearing the Cmv1r allele are able to control early MCMV replication are unknown, however, studies in mice bearing targeted disruptions at specific genes together with antibody depletion experiments indicate that the action of Cmv1 is comprised of a perforin-dependent and IFN␥-independent function.87,88 The chromosomal location of Cmv1 and identification of nearby loci was determined by assessment of the inheritance pattern from two sets of recombinant inbred (RI) parent strains, CXB and BXD.85 The inheritance pattern of Cmv1 from the RI strains was identical to that of a set of markers known to reside on a region of chromosome 6 called the NK complex (NKC).85 The NKC contains genes and gene families preferentially expressed in NK cells and since host NK cells are responsible for the control of early MCMV viral replication, the NKC region was considered to be fertile grounds for uncovering a potential candidate for Cmv1. In the mouse, the NKC is a 2 Mb genomic region that contains genes encoding a variety of surface receptors expressed on NK cells and some T cell populations (Figure 1a).89 Of particular interest, Ly49 receptors are encoded by a family of as many as 14 homologous genes that are expressed in overlapping subsets of NK cells and bind classical MHC class I molecules. CD94 is encoded by a single gene whereas NKG2 is comprised of a multigene family (NKG2A, C, D and E in the mouse) forming heterodimers that bind non-classical MHC class I molecule Qa-1 (HLA-E in human).94 The NKC also contains the Nkrp gene family (⬎3 members, including Nk1.1)95 and other single genes such as Shp1 (SH2-containing protein tyrosine phosphatase) that are also involved in NK-


253

Figure 1 Comparative maps of the natural killer complex (NKC) and the leukocyte receptor complex (LRC) in humans and mice. The NKC and the LRC are two major clusters of genes encoding activating or inhibitory receptors preferentially expressed on NK cells (see text). (a) Genetic organization of the NKC on human chromosome 12p13 and the syntenic region on mouse chromosome 6. Genes of the NKC code for C-type lectin receptors. Note the expansion of the genomic region in the vicinity of the LY49L pseudogene on the mouse chromosome domain harboring the Ly49 gene family. MAGOH (Mago nashi homologue),90 CSDA (cold shock domain protein A),91 T2R (taste receptor)92 and PRP (proline-rich protein)93 are tightly linked to the NKC although functionally unrelated. (b) Genetic organization of the LRC on human chromosome 19q13 and the syntenic region on mouse chromosome 7. Genes of the LRC encode for receptors of the immunoglobulin superfamily. Note the absence of the KIR domain on mouse chromosome 7.

cell activity.96 The NKC is conserved between mouse chromosome 6, rat chromosome 4 and human chromosome 12p. In contrast to the rodent species, the Ly49 gene cluster is absent in humans (a single copy of the Ly49L gene has been identified in the human genome, however, no functional mRNA has yet been detected) (Figure 1a).97 In humans, the family of killer cell immunoglobulinlike receptors (KIR) are considered the functional homologues of mouse Ly49 proteins.98 KIR receptors are specific for various MHC class I molecules, they are expressed in overlapping subsets of NK cells, and include inhibitory or activating members that signal through the same intracellular pathways as Ly49 molecules. KIR genes are localized to chromosome 19q13 in a region known as the leukocyte receptor cluster (LRC) (Figure 1b).98,99 This region also encodes other members of the immunoglobulin superfamily such as the immunoglobulin-like transcrips (ILTs) and the leukocyte-associated inhibitory receptor (LAIRs) that are expressed on an extended range of cell types, as well as NKp46, an activating NK receptor. Whereas Nkp46 and the gene cluster of paired-immunoglobulin like receptor family (Pir), likely orthologous of ILTs, are localized to a syntenic region on mouse chromosome 7, KIRs are completely absent from rodents (Figure 1b).100,101 A remarkable feature of the NKC is the presence of

numerous loci associated with immune function and infection susceptibility (Figure 1a). In addition to containing the Cmv1 gene which confers resistance to MCMV, the NKC region has been linked to other phenotypes dependent upon NK cell-mediated immunity such as Chok, a mouse locus controlling tumor killing by NK cells.102 Moreover, loci contributing to susceptibility to cutaneous leishmaniasis,103 insulin-dependent diabetes mellitus104 and ectromelia virus105 in mice have been also localized to this chromosomal region. In rats, Nka, an autosomal dominant locus that controls NK cell lysis of allogeneic lymphocytes106 and Oia2, a locus controlling susceptibility to oil-induced arthritis has been mapped to the rat NKC.107 Not surprisingly, the KIR in human has been associated with loci controlling innate or autoimmune immunity such as tumor killing108 and rheumatoid arthritis.109

Cloning of Cmv1 The identification of Cmv1 appeared as a central point to understanding not only the genetic control of host susceptibility to MCMV infection but also the relationship of Cmv1 to other loci mapped to the NKC considering coordinate function of genes of the NKC in NK cellmediated activities. Based upon the function and NKGenes and Immunity


254

specific expression, all members of the NKC were considered potential candidates for Cmv1. Therefore, an approach based on meiotic mapping and the definition of a minimal genetic interval for the localization of Cmv1 seemed the most rational approach to identify the locus. Fine genetic mapping helped to define a minimal genetic interval containing candidate genes for Cmv1 as well as to identify genes exterior to the interval that were unrelated to the susceptibility phenotype. In addition, cloning of the minimal genetic interval from large insert libraries provided a means to isolate all potential candidate genes. Lastly, the candidacy of individual genes was assessed by mutational analysis of the corresponding cDNAs obtained from Cmv1r and Cmv1s mouse strains. As the race to identify the gene within the NKC responsible for MCMV resistance/susceptibility ensued, several mapping panels were produced to help localize Cmv1. These included (BALB/cJ × C57BL/6J)F1 × BALB/cJ and (A/J × C57BL/6J)F1 × A/J110 informative back-cross mice as well as 18 lines of recombinant congenic strains at the NKC110 totaling more than 3500 meiosis. Segregation analysis of polymorphic markers, including microsatellite markers, gene sequences and anonymous probes, with respect to Cmv1 lead to the localization of Cmv1 to a 0.35cM interval in two independent genetic maps.110–112 The map generated by our group revealed gene order and intergene distances (in cM) in the vicinity of Cmv1 as: Nk1.1 – 0.3 – Cd94/Nkg2d – (0.05) – Ly49a-n/Cmv1 – (0.3) – Prp/Kap (Figure 2a). A high-resolution physical map of the Cmv1 genetic interval based on long-range restriction mapping by PFGE and the assembly of a contig of YAC and BAC clones translated the genetic interval into 1.6 Mb of genomic DNA containing a minimum of 18 transcription units including the 14 Ly49 gene cluster (Figure 2b, c).112 These results excluded Nk1.1, Cd94 and Nkg2d, while retaining the Ly49 gene cluster as candidates for Cmv1.110–112 Interestingly, a similar mapping approach being performed by a second group initially excluded Ly49 genes as potential candidates by placing Cmv1 distal to Ly49b gene, raising the possibility that one or more NKC genes could be involved in the resistance/susceptibility phenotype.115 Sequence comparison between Cmv1r and Cmv1s strains indicated allelic differences for transcripts analyzed in the Cmv1 interval116,117 (Figure 2b, Figure 3), precluding assessment of their candidacy based upon mutation analysis. In addition, the complex structural organization of the Ly49 gene family further complicated the analysis of the Cmv1 domain as many of the Ly49 genes have different copy numbers and genomic organization in inbred strains, making it difficult to distinguish allelic variants of the same or of distinct Ly49 gene copies.117,120 Thus as an alternative to fine mapping, haplotype analysis of a set of loci linked to Cmv1 was applied to a panel of 18 inbred strains characterized by their pattern of resistance or susceptibility to MCMV.118,119 A compilation of the results obtained for 27 independent markers evenly distributed over 1.8 Mb of genomic DNA demonstrated the presence of three major classes of independent haplotypes at this region (Figure 3). A class presented in the CMV-resistant strain C57BL/6J and its close relative C57BL/10J showed the same haplotype. The second class, presented in six susceptible inbred strains including FVB/NJ, carried haplotypes related to that of Cmv1r strains but presenting variations for loci

Genes and Immunity

Figure 2 Genetic mapping of the Cmv1 locus. (a) Composite genetic linkage map of mouse chromosome 6 in the vicinity of Cmv1. The order and distances of the loci was determined by pedigree analysis. The centromere is represented by a black circle. Recombination frequencies in centimorgans (cM) are shown to the right of the chromosome. Names of genes are shown in bold. Nkrp1, Cd94 and Nkg2d code for C-lectin type receptors. Prp: prolin-rich protein, Kap: Kidney androgen protein,113 Tel: translocation-ets-leukemia).114 (b) Blow-up of the minimal genetic interval between D6Ott8 and D6Ott115. This interval corresponds to 1.6 Mb of genomic DNA. Haplotypes for the resistant C57BL/6 strain and the susceptible BXD-8 and DBA/2 strains are shown using non-polymorphic (green rectangles) and polymorphic genetic markers (white rectangles, DBA/2 alleles; black rectangles, C57BL/6 alleles). (c) Transcriptional map showing candidate genes for Cmv1. Genes, other than members of the Ly49 family, presenting allelic polymorphisms between C57BL/6 and DBA/2 are indicated by an asterisk. The large red rectangle highlights the position of the genomic 23 kb deletion associated with MCMV susceptibility in the BXD-8 mouse.

distal to Ly49d. Genetic complementation of hybrids from an F1 cross between FVB/NJ and the strain C.B6-Ly49eD6Mit15 (congenic for a Cmv1r segment) showed that FVB/NJ, and probably other strains with related haplotypes, carry a susceptibility allele named Cmv1sFVB.118 The third class presented in nine strains including BALB/cJ, A/J, DBA/2J known to carry the susceptibility allele Cmv1s by genetic mapping (re-named Cmv1sBALB) carry a unique haplotype.118 Remarkably, the haplotype of Cmv1sBALB strains is distinct at all loci from those of either resistant, Cmv1r, or susceptible, Cmv1sFVB, mouse strains indicating an independent origin for Cmv1sBALB and Cmv1sFVB and raising the possibility of either genetic heterogeneity or allelic heterogeneity at Cmv1. Close inspection of the Cmv1sFVB haplotypes indicated allele sharing from proximal markers, ruling out Ly49e, Ly49f and Ly49d as Cmv1 candidates. However, the region in disequilibrium with susceptibility extended over distally to Ly49d. These results together with the localization of Cmv1 as


255

Figure 3 Composite haplotype map in the vicinity of Cmv1 for MCMV-resistant and MCMV-susceptible mouse strains. The top arrows indicate the reported genetic localization for Cmv1117,120 with respect to genetic markers used for haplotype analysis. Results from two different reports121,122 were compiled to construct the individual haplotypes. Markers indicated by an asterisk are localized between D6Ott115 and 242D11L2, however, the precise order has not been determined. A color code is used to identify different alleles with pink indicating MCMV-resistant C57BL/6J alleles and yellow indicating MCMV-susceptible AKR/J alleles. 0 indicates that there is no PCR product. ND indicates not determined. All markers have been previously reported.111,112,115,118,119

determined by the two previously described genetic maps112,115 made it difficult to rule out a possible contribution of several linked loci to the susceptibility/ resistance phenotype. Studies using the recombinant inbred strain BXD-8 were instrumental to pinpoint the genetic defect associated with MCMV susceptibility. Recombinant inbred strains were derived by brother–sister mating starting at the F2 generation originating from a cross between two standard inbred strains. As such, the BXD panel has been derived from progenitor strains C57BL/6 (Cmv1r) and DBA/2 (Cmv1s). BXD mice are homozygous for all loci, and their genome composition is on average 50% of C57BL/6 (B6) origin and 50% of DBA/2 origin.121 Each line in the BXD panel will have a unique chromosomal composition that is a mosaic of chromosomal segments inherited from one or the other parental strain. The BXD8 line was of particular interest since it was highly susceptible to MCMV infection but harbored a C57BL/6 haplotype at Cmv1 (Figure 2b).85 Linkage analysis using (BXD-8/Ty × C57BL/6)F2 populations together with genetic complementation experiments with congenic strains at Cmv1, indicated that the CMV susceptibility trait of BXD-8 was localized to the minimal Cmv1 interval.77 Close inspection of this interval with 45 STSs and polymorphic markers, together with pulse-field gel electrophoresis identified a 23 kb deletion,77 encompassing

Ly49h (Figure 2b, c). Moreover, analysis of the expression patterns of candidate NKC genes revealed that absence of the Ly49H activating receptor correlated with MCMV susceptibility, pointing to allelism between Ly49h and Cmv1. In addition, genotyping of mice with Ly49h derived markers together with RNA analysis (Vidal SM et al, personal communication) indicated the absence of this locus in all susceptible strains analysed (Figure 3). This was further corroborated by treatment of MCMVresistant C57BL/6 mice with an anti-Ly49H antibody which resulted in the acquisition of a susceptible phenotype.78,122

Overview of the Ly49 proteins As stated previously, the Ly49 gene family contains at least 14 members, however, only a handful have been extensively studied to date (predominantly Ly49A, C, D and G2). Sequence alignments indicate that all Ly49 proteins are highly related and all are likely to be comprised of disulfide linked type II membrane spanning homodimers with an extracellular domain comprised of a C-type lectin motif.123 Ly49 proteins are expressed predominantly on cells of the NK and NK T lineages although a small percentage of CD8 T cells also express Ly49 molecules.124 This restricted expression pattern is consistent with the demonstrated importance of NK cells during Genes and Immunity


256

CMV infection, particularly during the very earliest stages of infection. It has been established for some time now that Ly49A, C, D and G2 proteins all share a ligand specificity for classical MHC class I molecules123 and this is consistent with the fact that all Ly49 proteins share a highly related extracellular domain. However, Ly49 proteins fall into two distinct categories based upon stimulatory or inhibitory signalling following ligand binding. These divergent functional activities are the consequence of the structural organization of their intracellular domains. Inhibitory members of the Ly49 family contain an intracellular immunoreceptor tyrosine-based inhibitory motif (ITIM) (VXYXXV/L) which becomes tyrosine phosphorylated upon ligand binding leading to recruitment of the SHP-1 phosphatase.125,126 The exact target of the SHP-1 phosphatase in NK signaling is not yet clear, however, it is likely that it interferes with the signaling activities of spatially proximal signaling molecules. As stated above, all known ligands of Ly49 proteins are MHC class I molecules however there are distinct binding specificities for each Ly49 molecule. For example the inhibitory molecule Ly49A binds to H2Db,d,k,p whereas Ly49C bind to H2Kb,d and H2Db,d,k.123 The Ly49G2 inhibitory receptor seems to be specific for H2Dd.127 Different subsets of NK cells appear to express overlapping sets of Ly49 molecules such that all cells express at least two different inhibitory receptors.123 The percentage of cells expressing a given Ly49 molecule varies between different strains of mice and appears to be regulated by the expression of the cognate MHC class I molecule in that strain.128,129 This process has been termed ‘receptor calibration’ and although the precise mechanism behind this regulated pattern of expression has been the subject of intense scrutiny we still know little about the process. Recently, the crystal structure of an inhibitory Ly49 molecule, Ly49A, complexed with its ligand H2Dd was reported.130 Interestingly, the Ly49A homodimer was found to simultaneously contact two MHC class I molecules, one at the ␣2 domain of the peptide-binding site and the other at a site overlapping the CD8 binding site. The authors suggest that Ly49A dimers may in fact bind to one MHC class I molecule on a putative target cell using the former site, and a second neighboring MHC class I molecule on the NK cell itself using the latter site. Indeed it has been shown that an MHC class I/peptide complex can be simultaneously bound by both a T cell receptor (TCR) and an inhibitory Ly49A molecule suggesting a potential role in the activation of TCR-expressing NK T cells.131 The second category of Ly49 molecules, the activating Ly49 molecules, have intracellular domains which are structurally distinct from those of the inhibitory Ly49 molecules. In contrast to the inhibitory Ly49 molecules, activating Ly49 receptors lack tyrosine residues in their intracellular domain and are thus incapable of signaling themselves. Rather, activating Ly49 receptors signal through an adapter molecule called DNAX activation protein 12 (DAP12) which associates with Ly49 through a non-covalent interaction using a charged residue in the membrane-spanning domain.132,133 DAP12 is a transmembrane protein with a very short extracellular domain containing a cysteine residue required for homodimer formation. The intracellular domain of DAP12 contains an immunoreceptor tyrosine-based activation motif (ITAM)

Genes and Immunity

which becomes phosphorylated upon ligand binding and recruits the cytoplasmic protein tyrosine kinases SYK and ZAP70. These in turn activate downstream signaling pathways that result in mobilization of intracellular calcium stores and activation of the cell’s cytotoxic activity. In addition to complexing with activating Ly49 molecules, DAP12 functions as an adapter molecule for a number of activating surface receptors including mouse and human CD94/NKG2C, activating human KIR proteins (see below), and the myeloid cell-specific factors MDL-1, TREM-1 and TREM-2.134 Consequently, mice deficient for DAP12 have diminished Ly49-mediated killing of certain tumors as well as other immune defects.134 Surprisingly, humans with DAP12 have neurological symptoms characterized by progressive presenile frontal lobe dementia135 indicating that DAP12 plays an important role in human brain function. However, no obvious immune defects have been reported associated with DAP12 deficiency in humans. Of the 14 Ly49 molecules encoded in the Ly49 locus of mouse chromosome 6, only Ly49D and Ly49H have been identified as activating receptors. Ly49D is known to recognize the MHC class I molecule H2Dd123 although the physiological significance of this recognition is not yet clear. Ligation of Ly49D with cross-linking antibodies or by H2Dd-expressing target cells results in both increased cytotoxicity and IFN-␥ production however both can be downregulated by a concurrent signal from the inhibitory receptor Ly49G2 (also H2Dd-specific).133 This latter finding suggests that signals from inhibitory Ly49 receptors are dominant over signals derived from activating Ly49 receptors. Although the shared presence of both positive and negative signaling receptors for the same ligand on a single NK cell presents a confusing picture in terms of NK function, recent evidence suggests that inhibitory or activation Ly49 receptors have different affinity for MHC class I and that this may be influenced by the peptide bound to MHC class I.136 In addition, transgenic transfer of Ly49dB6 into BALB/c NK cells was recently shown to confer cytotoxic activity against Chinese hamster ovary (CHO) cells, establishing that Ly49D is the Chok gene product,137 indicating that alternate Ly49D ligands may flag tumor cells for destruction by Ly49D-expressing NK cells. Recently, gene chip analyses have indicated that NK cells activated through cross-linking of Ly49D surface receptors upregulate a number of soluble mediators of inflammation, in particular the chemokines MIP1␣ and MIP1␤, suggesting that they may play a key role in initiating inflammation and recruitment of cells of the adaptive immune response.138 The identification of Ly49H as an important molecule during early MCMV infection provides tantalizing evidence that the activating molecules of the NK cell system may actually have a built in ability to discriminate between non-infected cells versus those infected by common or important pathogens. Approximately half of the NK cell population in adult C57BL/6 mice express the Ly49H molecule and mouse strains lacking Ly49H have a higher burden of MCMV in the spleen during the first days after infection.87 The Ly49H+ population is rapidly expanded after MCMV infection and evidence suggests that Ly49H+ positive cells at the site of infection are actively producing IFN-␥.122 Together these data suggest that Ly49H+ positive NK cells are receiving a stimulatory


signal from the infection and that the stimulatory signal is delivered via Ly49H.

Model for the role of Ly49H in MCMV resistance The regulation of the cytotoxic activity of NK cells by stimulatory and inhibitory receptors has been principally studied using tumor models where loss of MHC class I expression results in NK-mediated lysis, a process that fits with the ‘missing self hypothesis’ that was earlier proposed by Karre et al.139 Lysis of MHC class I-negative tumors is now known to be the result of a loss in the ability to ligate the inhibitory Ly49 receptors which are activated in the presence of MHC class I, thereby preventing NK-mediated lysis of normal cells (Figure 4a). In addition to tumor surveillance, NK cells also patrol for cells that are infected by virus because many viruses, including CMV, attempt to evade the scrutiny of the adaptive immune system by downregulating MHC class I expression, making them invisible to MHC class Irestricted CD8+ cytotoxic cells.141 However viral-

mediated down-regulation of MHC class I expression would be expected to interrupt the signaling via inhibitory receptors on NK cells, rendering infected cells susceptible to NK lysis. The apparent success of CMV to remain as a persistent latent infection for years despite the downregulation of MHC class I expression implies that it has acquired means of subverting the attack of both NK cells and T cells.142 Recent evidence indicated that MCMV deploys an MHC class I homologue that serves to evade the host immune response.140 Expression of this virally-encoded MHC class I homologue (m144) would serve the virus in two ways; first by interrupting the surveillance and destruction of the infected cell by virus-specific CD8 cytotoxic lymphocytes and second by engaging the MHC the class I specific inhibitory receptors of NK cells to avoid NK killing (Figure 4b). This is supported by in vivo experiments showing that deletion of the m144 gene results in reduced viral titers in Cmv1s MCMV-susceptible mice, suggesting that this MCH class I homologue interferes with NK cell cytotoxic activity, likely via an inhibitory NK receptor.140 Despite this proposed role of

257

Figure 4 Role of Ly49H in MCMV resistance. (a) The cytolytic activity of natural killer (NK) cells is regulated by signals received from stimulatory and inhibitory receptors present on the surface of the cell. Inhibitory receptors specific for classical MHC class I molecules prevent lysis of ‘normal’ cells by recruiting the SHP-1 phosphatase to intracellular immunoreceptor tyrosine-based inhibitory domains (ITIM) upon binding of MHC class I. (b) MCMV, as other viruses, attempts to evade the scrutiny of the adaptive immune system by downregulating MHC class I molecules on the surface of the infected cell. This mechanism, however, results in loss of signal from the NK cell inhibitory receptor and the infected cell thus becomes susceptible to NK-mediated killing. MCMV has one-upped the ante in this battle by encoding m144, an MHC class I homologue, that both avoids recognition by the adaptive immune system and tricks NK cells into thinking that the cell has ‘normal’ expression of MHC. Indeed loss of m144 results in reduced virulence,140 probably as a result of NKmediated killing. (c) We now know that MCMV-resistance in Cmv1r strains is mediated by the presence of the Ly49H stimulatory receptor on NK cells. Ly49H forms a complex with the signaling molecule DAP12 and thus ligand binding would be expected to result in NK stimulation via the immunoreceptor tyrosine-based activating domain (ITAM) of DAP12. The ligand of Ly49H is currently unknown, but mice that lack Ly49H on their NK cells are highly susceptible MCMV infection. Thus the ligand of Ly49H must be an endogenous or virally encoded molecule that is upregulated on MCMV-infected cells, rendering them susceptible to NK-mediated killing. As discussed in the text, it is possible that the MCMV-encoded MHC class I homologue m144 itself is the Ly49H-specific ligand. Genes and Immunity


258

m144, Cmv1r mice efficiently control viral replication. The recent identification of Ly49H as a critical molecule for viral resistance sheds new light on potential mechanisms of NK-mediated killing of infected cells. The precise role of Ly49H could be explained by two alternative models. First, ligation of Ly49H by an endogenous molecule that is specifically upregulated on virally infected cells would promote an activating signal capable of overriding the m144-elicited inhibitory signals. Alternatively Ly49H could itself recognize the m144 MHC class I homologue and deliver a positive signal that promotes target cell killing. This could be achieved either by absence of the m144 inhibitory receptor on Cmv1r strains or by competition between Ly49H and the inhibitory receptor for m144 binding. In this scenario the presence of Ly49H would tip the balance in favor of the host immune response by facilitating recognition of the infected cell by NK cells (see Figure 4c). Regardless of the mechanism, the identification of a ligand for Ly49H should be considered a high priority for it would likely provide critical information regarding the functional role of NK cells during MCMV infection. Considering that all other members of the Ly49 family with known ligand specificity recognize classical MHC class I molecules, these should be the first potential ligands to be ruled in or ruled out as Ly49Hspecific ligands.

Relevance to HCMV infection NK cells are also thought to play a critical role in HCMV infections.59 Although they lack Ly49 molecules, human NK cells have the KIR system which is able to detect the downregulation of MHC class I molecules on pathogeninfected cells. Inhibitory KIR have intracellular ITIM motifs and activating KIR associate with ITAM-containing DAP12 adapter molecules. However the extracellular domains of the KIR proteins are structurally distinct from those of Ly49, being comprised of immunoglobulin (Ig) domains rather than C-type lectin domains. Like Ly49 proteins, the KIR proteins recognize classical MHC class I molecules.94 Thus downregulation of MHC class I molecules by HCMV would leave cells susceptible to lysis by NK cells in the absence of a KIR inhibitory ligand. As was observed with the m144 protein of MCMV, HCMV also encodes a MHC class I homologue called UL18.143 UL18 associates with microglobulin and although it was hypothesized to represent a ligand for inhibitory receptors on NK cells,144 subsequent analyses showed that it actually augmented NK-mediated lysis.145 UL18 is now known to bind to an inhibitory receptor called leukocyte Ig-like receptor (LIR), however LIR appears to be expressed predominantly on monocytes and macrophages rather than NK cells thus its role in NK surveillance is not yet clear.146 Analogous to the emerging role of the activating receptor Ly49H in MCMV infection, activating receptors of human NK cells seem to play a critical role in recognition of HCMV infected cells. The activating receptor NKG2D in particular seems to be important for control of viral replication as well as for control for tumorigenic precursors.147 Like Ly49, NKG2D is a type II disulfide-linked dimer with a lectin-like extracellular domain. However unlike the activating Ly49 molecules NKG2D signals through the adaptor molecule DAP10 rather than DAP12. DAP10 does not contain a conventional ITAM motif but Genes and Immunity

instead activates the phosphatidylinositol 3-kinase (PI3K) pathway.148 NKG2D is expressed in both human and murine NK cells as well as most CD8+ T cells and ␥␦ T cells.149 The ligands for human NKG2D include the highly polymorphic MHC class I-related molecules MIC A and MIC B which seem to be upregulated on tumor cells or otherwise stressed cells.150 NKG2D also binds to the ULBP family of MHC class I-related molecules151 and the ULBP and MIC proteins provide a positive signal to NK cells that can overcome even the inhibitory signals provided by KIR molecules. Interestingly, another HCMV encoded protein, UL16, seems to block the binding of MIC B and ULBP 1 and 2 to NKG2D on human NK cells151 thus providing another potential escape mechanism for the virus. Although homologues of MIC or ULBP have not been identified in the mouse, two novel ligands of murine NKG2D (H-60 and Rae1␤) have recently been identified.152,153 H-60 and Rae1␤ are distantly related to MHC class I molecules and, like the human MIC proteins, are found to be upregulated on a number of murine tumors. Given the apparent propensity of CMV to counteract any strategy used by the host to thwart infection it seems likely that MCMV-encoded inhibitors of H-60 and Rae1␤ are also likely to be found in the near future.

Concluding remarks In conclusion, the identification of the activating Ly49H receptor as an important molecule for the activation of NK cells during the early stages of MCMV infection in the mouse constitutes an important step forward in understanding the function of these enigmatic cells. This finding provides additional evidence that NK cells have a capacity to discriminate between infected and uninfected cells despite the lack of variant receptors seen on cells of the adaptive immune system. The consequence of this recognition is removal of the suspect cell via NKmediated lysis thus the activities of the NK cell population must be kept closely harnessed. This fact helps to explain the presence of both activating and inhibitory receptors on NK cells, however one intriguing aspect regarding the repertoire of NK cell surface molecules is the apparent rapidity of their evolution. Indeed, although they are functionally analogous, the Ly49 gene family of mice and the KIR gene family of humans appear to have evolved independently as evidenced by their highly divergent ligand-binding domains. This divergence may be related to differences in MHC molecules which are common ligands of Ly49 and KIR proteins and which are themselves highly polymorphic. However, the MHC molecules of mouse and human are highly similar from a structural standpoint thus it is surprising that NK cells have evolved MHC class I-binding proteins that utilize completely distinct binding motifs. Furthermore the presence of at least 14 Ly49 family members in the mouse genome suggests that this locus has rapidly expanded since the divergence of mouse and humans. Thus although Ly49H appears to play an important role in MCMV infection, it is likely that other members of the Ly49 gene complex play distinct roles during infection by other viruses or other infectious diseases. For example, the importance of the Cmv1/Ly49H during MCMV infection, together with the role of Chok/Ly49D in tumor killing expands the functional diversity of Ly49 genes in NK


cells. These results suggest that Ly49 paralogous genes are also likely candidates for phenotypically defined loci in the NKC such as Rmp1, Scl and Idd6. The proposal that such functions would be strictly dependent on activating members of the Ly49 family supports a direct role of these molecules in recognition of infection. Similarly, it is possible that variation in KIR family members may contribute to susceptibility to HCMV and other related NK cell-mediated pathologies in humans.

18 19 20

21

Acknowledgements The authors thank Erwin Schurr from the McGill Center of Host Resistance (Montreal, Canada) for critical reading of this manuscript.

22

23

References

24

1 Britt W, Alford CA. Cytomegalovirus. In: Fields BN (eds). Fields Virology. Lippincott-Raven: New York, 1996, pp 2493–2523. 2 Rawlinson WD. Broadsheet. Number 50: diagnosis of human cytomegalovirus infection and disease. Pathology 1999; 31: 109–115. 3 Soderberg-Naucler C, Nelson JY. Human cytomegalovirus latency and reactivation – a delicate balance between the virus and its host’s immune system. Intervirology 1999; 42: 314–321. 4 Reusser P. Current concepts and challenges in the prevention and treatment of viral infections in immunocompromised cancer patients. Support Care Cancer 1998; 6: 39–45. 5 Reusser P, Attenhofer R, Hebart H, Helg C, Chapuis B, Einsele H. Cytomegalovirus-specific T-cell immunity in recipients of autologous peripheral blood stem cell or bone marrow transplants. Blood 1997; 89: 3873–3879. 6 Soderberg-Naucler C, Emery VC. Viral infections and their impact on chronic renal allograft dysfunction. Transplantation 2001; 71: SS24–SS30. 7 Trincado DE, Scott GM, White PA, Hunt C, Rasmussen L, Rawlinson WD. Human cytomegalovirus strains associated with congenital and perinatal infections. J Med Virol 2000; 61: 481–487. 8 Demmler GJ. Infectious Diseases Society of America and Centers for Disease Control. Summary of a workshop on surveillance for congenital cytomegalovirus disease. Rev Infect Dis 1991; 13: 315–329. 9 Hurme M, Helminen M. Resistance to human cytomegalovirus infection may be influenced by genetic polymorphisms of the tumour necrosis factor-alpha and interleukin-1 receptor antagonist genes. Scand J Infect Dis 1998; 30: 447–449. 10 Kaufman DB, Leventhal JR, Gallon LG et al. Risk factors and impact of cytomegalovirus disease in simultaneous pancreaskidney transplantation. Transplantation 2001; 72: 1940–1945. 11 Singh N, Wagener MM, Gayowski T. Seasonal pattern of early mortality and infectious complications in liver transplant recipients. Liver Transpl 2001; 7: 884–889. 12 Nachbaur D, Bonatti H, Oberaigner W et al. Survival after bone marrow transplantation from cytomegalovirus seropositive sibling donors. Lancet 2001; 358: 1157–1159. 13 Farmer DG, McDiarmid SV, Yersiz H et al. Outcome after intestinal transplantation: results from one center’s 9-year experience; discussion 1031–2. Arch Surg 2001; 136: 1027–1031. 14 Rubin RH. Importance of CMV in the transplant population. Transpl Infect Dis 1999; 1 (Suppl 1): 3–7. 15 de Medeiros CR, Moreira VA, Pasquini R. Cytomegalovirus as a cause of very late interstitial pneumonia after bone marrow transplantation. Bone Marrow Transplant 2000; 26: 443–444. 16 Sparrelid E, Emanuel D, Fehniger T, Andersson U, Andersson J. Interstitial pneumonitis in bone marrow transplant recipients is associated with local production of TH2-type cytokines and lack of T cell-mediated cytotoxicity. Transplantation 1997; 63: 1782–1789. 17 Stratta RJ. Clinical patterns and treatment of cytomegalovirus

25 26

27

28

29

30

31

32

33

34 35

36 37 38

39 40 41

42

infection after solid-organ transplantation. Transplant Proc 1993; 25: 15–21. Tolkoff-Rubin NE, Rubin RH. Viral infections in organ transplantation. Transplant Proc 1998; 30: 2060–2063. Snydman DR. Infection in solid organ transplantation. Transpl Infect Dis 1999; 1: 21–28. Yamada S, Takatsuka H, Takemoto Y et al. Association of cytomegalovirus interstitial pneumonitis with HLA-type following allogeneic bone marrow transplantation. Bone Marrow Transplant 2000; 25: 861–865. Salmon-Ceron D. Cytomegalovirus infection: the point in 2001. HIV Med 2001; 2: 255–259. Murphy EL, Collier AC, Kalish LA et al. Highly active antiretroviral therapy decreases mortality and morbidity in patients with advanced HIV disease. Ann Intern Med 2001; 135: 17–26. Welch K, Kissinger P, Bessinger R, Dascomb K, Morse A, Gleckler E. The clinical profile of end-stage AIDS. AIDS Patient Care STDS 1998; 12: 125–129. Vercellotti GM. Overview of infections and cardiovascular diseases. J Allergy Clin Immunol 2001; 108: S117–S120. Levi M. CMV endothelitis as a factor in the pathogenesis of atherosclerosis. Cardiovasc Res 2001; 50: 432–433. Valantine HA. Role of CMV in transplant coronary artery disease and survival after heart transplantation. Transpl Infect Dis 1999; 1 (Suppl 1): 25–30. Zhou YF, Leon MB, Waclawiw MA et al. Association between prior cytomegalovirus infection and the risk of restenosis after coronary atherectomy. N Engl J Med 1996; 335: 624–630. Neumann FJ, Kastrati A, Miethke T et al. Previous cytomegalovirus infection and restenosis after coronary stent placement. Circulation 2001; 104: 1135–1139. Koskinen PK, Kallio EA, Tikkanen JM, Sihvola RK, Hayry PJ, Lemstrom KB. Cytomegalovirus infection and cardiac allograft vasculopathy. Transpl Infect Dis 1999; 1: 115–126. Epstein SE, Speir E, Zhou YF, Guetta E, Leon M, Finkel T. The role of infection in restenosis and atherosclerosis: focus on cytomegalovirus. Lancet 1996; 348 (Suppl 1): s13–s17. Hsich E, Zhou YF, Paigen B, Johnson TM, Burnett MS, Epstein SE. Cytomegalovirus infection increases development of atherosclerosis in Apolipoprotein-E knockout mice. Atherosclerosis 2001; 156: 23–28. Burnett MS, Gaydos CA, Madico GE et al. Atherosclerosis in apoE knockout mice infected with multiple pathogens. J Infect Dis 2001; 183: 226–231. Fabricant CG, Fabricant J. Atherosclerosis induced by infection with Marek’s disease herpesvirus in chickens. Am Heart J 1999; 138: S465–S468. Villarreal EC. Current and potential therapies for the treatment of herpesvirus infections. Prog Drug Res 2001; 56: 185–228. Zaia JA, Sissons JG, Riddell S et al. Status of cytomegalovirus prevention and treatment in 2000. Hematology (Am Soc Hematol Educ Program) 2000; 1: 339–355. Chou S. Antiviral drug resistance in human cytomegalovirus. Transpl Infect Dis 1999; 1: 105–114. Emery VC. Progress in understanding cytomegalovirus drug resistance. J Clin Virol 2001; 21: 223–228. Biron CA, Nguyen KB, Pien GC, Cousens LP, Salazar-Mather TP. Natural killer cells in antiviral defense: function and regulation by innate cytokines. Annu Rev Immunol 1999; 17: 189–220. Trinchieri G. Biology of natural killer cells. Adv Immunol 1989; 47: 187–376. Tay CH, Szomolanyi-Tsuda E, Welsh RM. Control of infections by NK cells. Curr Top Microbiol Immunol 1998; 230: 193–220. Venema H, van den Berg AP, van Zanten C, van Son WJ, van der GM, The TH. Natural killer cell responses in renal transplant patients with cytomegalovirus infection. J Med Virol 1994; 42: 188–192. Slavin M, Dobbs S, Crawford S, Bowden R. Interleukin-2, interferon-gamma and natural killer cell activity in bronchoalveolar lavage fluid from marrow transplant recipients with cytomegalovirus pneumonia. Bone Marrow Transplant 1993; 11: 113–118.

259

Genes and Immunity


260

43 Biron CA, Byron KS, Sullivan JL. Severe herpesvirus infections in an adolescent without natural killer cells. N Engl J Med 1989; 320: 1731–1735. 44 Sullivan JL, Woda BA. X-linked lymphoproliferative syndrome. Immunodefic Rev 1989; 1: 325–347. 45 Coffey AJ, Brooksbank RA, Brandau O et al. Host response to EBV infection in X-linked lymphoproliferative disease results from mutations in an SH2-domain encoding gene. Nat Genet 1998; 20: 129–135. 46 Sayos J, Wu C, Morra M et al. The X-linked lymphoproliferativedisease gene product SAP regulates signals induced through the co-receptor SLAM. Nature 1998; 395: 462–469. 47 Tangye SG, Lazetic S, Woollatt E, Sutherland GR, Lanier LL, Phillips JH. Cutting edge: human 2B4, an activating NK cell receptor, recruits the protein tyrosine phosphatase SHP-2 and the adaptor signaling protein SAP. J Immunol 1999; 162: 6981– 6985. 48 Parolini S, Bottino C, Falco M et al. X-linked lymphoproliferative disease. 2B4 molecules displaying inhibitory rather than activating function are responsible for the inability of natural killer cells to kill Epstein–Barr virus-infected cells. J Exp Med 2000; 192: 337–346. 49 Sayos J, Nguyen KB, Wu C et al. Potential pathways for regulation of NK and T cell responses: differential X-linked lymphoproliferative syndrome gene product SAP interactions with SLAM and 2B4. Int Immunol 2000; 12: 1749–1757. 50 Shanley JD. Host genetic factors influence murine cytomegalovirus lung infection and interstitial pneumonitis. J Gen Virol 1984; 65: 2121–2128. 51 Trgovcich J, Stimac D, Polic B et al. Immune responses and cytokine induction in the development of severe hepatitis during acute infections with murine cytomegalovirus. Arch Virol 2000; 145: 2601–2618. 52 Hayashi K, Suwa Y, Shimomura Y, Ohashi Y. Pathogenesis of ocular cytomegalovirus infection in the immunocompromised host. J Med Virol 1995; 47: 364–369. 53 Hayashi K, Kurihara I, Uchida Y. Studies of ocular murine cytomegalovirus infection. Invest Ophthalmol Vis Sci 1985; 26: 486– 493. 54 Shanley JD, Biczak L, Forman SJ. Acute murine cytomegalovirus infection induces lethal hepatitis. J Infect Dis 1993; 167: 264–269. 55 Shellam GR, Flexman JP. Genetically determined resistance to murine cytomegalovirus and herpes simplex virus in newborn mice. J Virol 1986; 58: 152–156. 56 Fitzgerald NA, Shellam GR. Host genetic influences on fetal susceptibility to murine cytomegalovirus after maternal or fetal infection. J Infect Dis 1991; 163: 276–281. 57 Allan JE, Shellam GR. Genetic control of murine cytomegalovirus infection: virus titres in resistant and susceptible strains of mice. Arch Virol 1984; 81: 139–150. 58 Shanley JD. Host genetic factors influence murine cytomegalovirus lung infection and interstitial pneumonitis. J Gen Virol 1984; 65: 2121–2128. 59 Biron CA, Brossay L. NK cells and NKT cells in innate defense against viral infections. Curr Opin Immunol 2001; 13: 458–464. 60 Scalzo AA, Fitzgerald NA, Wallace CR et al. The effect of the Cmv-1 resistance gene, which is linked to the natural killer cell gene complex, is mediated by natural killer cells. J Immunol 1992; 149: 581–589. 61 Bukowski JF, Welsh RM. The role of natural killer cells and interferon in resistance to acute infection of mice with herpes simplex virus type 1. J Immunol 1986; 136: 3481–3485. 62 Welsh RM, Brubaker JO, Vargas-Cortes M, O’Donnell CL. Natural killer (NK) cell response to virus infections in mice with severe combined immunodeficiency. The stimulation of NK cells and the NK cell-dependent control of virus infections occur independently of T and B cell function. J Exp Med 1991; 173: 1053–1063. 63 Ellison AR, Yang L, Voytek C, Margolis TP. Establishment of latent herpes simplex virus type 1 infection in resistant, sensi-

Genes and Immunity

64

65

66

67

68

69

70

71

72

73

74

75

76

77

78

79

80

81

82

83

tive, and immunodeficient mouse strains. Virology 2000; 268: 17–28. Riera L, Gariglio M, Valente G et al. Murine cytomegalovirus replication in salivary glands is controlled by both perforin and granzymes during acute infection. Eur J Immunol 2000; 30: 1350–1355. Tsunobuchi H, Nishimura H, Goshima F et al. A protective role of interleukin-15 in a mouse model for systemic infection with herpes simplex virus. Virology 2000; 275: 57–66. Pien GC, Satoskar AR, Takeda K, Akira S, Biron CA. Cutting edge: selective IL-18 requirements for induction of compartmental IFN-gamma responses during viral infection. J Immunol 2000; 165: 4787–4791. Carr JA, Rogerson JA, Mulqueen MJ, Roberts NA, Nash AA. The role of endogenous interleukin-12 in resistance to murine cytomegalovirus (MCMV) infection and a novel action for endogenous IL-12 p40. J Interferon Cytokine Res 1999; 19: 1145– 1152. Dutia BM, Allen DJ, Dyson H, Nash AA. Type 1 interferons and IRF-1 play a critical role in the control of a gammaherpesvirus infection. Virology 1999; 261: 173–179. Fernandez JA, Rodrigues EG, Tsuji M. Multifactorial protective mechanisms to limit viral replication in the lung of mice during primary murine cytomegalovirus infection. Viral Immunol 2000; 13: 287–295. Kumaraguru U, Davis IA, Deshpande S, Tevethia SS, Rouse BT. Lymphotoxin alpha −/− mice develop functionally impaired CD8+ T cell responses and fail to contain virus infection of the central nervous system. J Immunol 2001; 166: 1066–1074. Salazar-Mather TP, Orange JS, Biron CA. Early murine cytomegalovirus (MCMV) infection induces liver natural killer (NK) cell inflammation and protection through macrophage inflammatory protein 1 alpha (MIP-1 alpha)-dependent pathways. J Exp Med 1998; 187: 1–14. Rager-Zisman B, Quan PC, Rosner M, Moller JR, Bloom BR. Role of NK cells in protection of mice against herpes simplex virus1 infection. J Immunol 1987; 138: 884–888. Ghiasi H, Cai S, Perng G, Nesburn AB, Weschsler SL. Perforin pathway is essential for protection of mice against lethal ocular HSV-1 challenge but not corneal scarring. Virus Res 1999; 65: 97–101. MacLean A, Wei XQ, Huang FP, AI Alem UA, Chan WL, Liew FY. Mice lacking inducible nitric-oxide synthase are more susceptible to herpes simplex virus infection despite enhanced Th1 cell responses. J Gen Virol 1998; 79: 825–830. Morra M, Howie D, Grande MS et al. X-linked lymphoproliferative disease: a progressive immunodeficiency. Annu Rev Immunol 2001; 19: 657–682. Wu C, Nguyen KB, Pien GC et al. SAP controls T cell responses to virus and terminal differentiation of TH2 cells. Nat Immunol 2001; 2: 410–414. Lee SH, Girard S, Macina D et al. Susceptibility to mouse cytomegalovirus is associated with deletion of an activating natural killer cell receptor of the C-type lectin superfamily. Nat Genet 2001; 28: 42–45. Brown MG, Dokun AO, Heusel JW et al. Vital involvement of a natural killer cell activation receptor in resistance to viral infection. Science 2001; 292: 934–937. Shellam GR, Allan JE, Papadimitriou JM, Bancroft GJ. Increased susceptibility to cytomegalvirus infection in beige mutant mice. Proc Natl Acad Sci USA 1981; 78: 5104–5108. Barbosa MD, Nguyen QA, Tchernev VT et al. Identification of the homologous beige and Chediak-Higashi syndrome genes. Nature 1996; 382: 262–265. Chalmer JE, Mackenzie JS, Stanley NF. Resistance to murine cytomegalovirus linked to the major histocompatibility complex of the mouse. J Gen Virol 1977; 37: 107–114. Grundy JE, Mackenzie JS, Stanley NF. Influence of H-2 and nonH-2 genes on resistance to murine cytomegalovirus infection. Infect Immun 1981; 32: 277–286. Price P, Winter JG, Nikoletti S, Hudson JB, Shellam GR. Func-


tional changes in murine macrophages infected with cytomegalvirus relate to H-2-determined sensitivity to infection. J Virol 1987; 61: 3602–3606. 84 Wykes MN, Shellam GR, McCluskey J, Kast WM, Dallas PB, Price P. Murine cytomegalovirus interacts with major histocompatibility complex class I molecules to establish cellular infection. J Virol 1993; 67: 4182–4189. 85 Scalzo AA, Fitzgerald NA, Simmons A, La Vista AB, Shellam GR. Cmv-1, a genetic locus that controls murine cytomegalovirus replication in the spleen. J Exp Med 1990; 171: 1469–1483. 86 Price P, Olver SD, Gibbons AE, Teo HK, Shellam GR. Characterization of thymic involution induced by murine cytomegalovirus infection. Immunol Cell Biol 1993; 71: 155–165. 87 Dokun AO, Kim S, Smith HR, Kang HS, Chu DT, Yokoyama WM. Specific and nonspecific NK cell activation during virus infection. Nat Immunol 2001; 2: 951–956. 88 Tay CH, Welsh RM. Distinct organ-dependent mechanisms for the control of murine cytomegalovirus infection by natural killer cells. J Virol 1997; 71: 267–275. 89 Brown MG, Fulmek S, Matsumoto K et al. A 2-Mb YAC contig and physical map of the natural killer gene complex on mouse chromosome 6. Genomics 1997; 42: 16–25. 90 Newmark PA. Boswell RE. The mago nashi locus encodes an essential product required for germ plasm assembly in Drosophila. Development 1994; 120: 1303–1313. 91 Coles LS, Diamond P, Occhiodoro F, Vadas MA, Shannon MF. Cold shock domain proteins repress transcription from the GMCSF promoter. Nucleic Acids Res 1996; 24: 2311–2317. 92 Adler E, Hoon MA, Mueller KL, Chandrashekar J, Ryba NJ, Zuker CS. A novel family of mammalian taste receptors. Cell 2000; 100: 693–702. 93 Maeda N, Kim HS, Azen EA, Smithies O. Differential RNA splicing and post-translational cleavages in the human salivary proline-rich protein gene system. J Biol Chem 1985; 260: 11123– 11130. 94 Raulet DH, Vance RE, McMahon CW. Regulation of the natural killer cell receptor repertoire. Annu Rev Immunol 2001; 19: 291–330. 95 Ryan JC, Turck J, Niemi EC, Yokoyama WM, Seaman WE. Molecular cloning of the NK1. 1 antigen, a member of the NKR-P1 family of natural killer cell activation molecules. J Immunol 1992; 149: 1631–1635. 96 Yi T, Gilbert DJ, Jenkins NA, Copeland NG, Ihle JN. Assignment of a novel protein tyrosine phosphatase gene (Hcph) to mouse chromosome 6. Genomics 1992; 14: 793–795. 97 Westgaard IH, Berg SF, Orstavik S, Fossum S, Dissen E. Identification of a human member of the Ly-49 multigene family. Eur J Immunol 1998; 28: 1839–1846. 98 Barten R, Torkar M, Haude A, Trowsdale J, Wilson MJ. Divergent and convergent evolution of NK-cell receptors. Trends Immunol 2001; 22: 52–57. 99 Wilson MJ, Torkar M, Haude A et al. Plasticity in the organization and sequences of human KIR/ILT gene families. Proc Natl Acad Sci USA 2000; 97: 4778–4783. 100 Biassoni R, Pessino A, Bottino C, Pende D, Moretta L, Moretta A. The murine homologue of the human NKp46, a triggering receptor involved in the induction of natural cytotoxicity. Eur J Immunol 1999; 29: 1014–1020. 101 Kubagawa H, Cooper MD, Chen CC et al. Paired immunoglobulin-like receptors of activating and inhibitory types. Curr Top Microbiol Immunol 1999; 244: 137–149. 102 Idris AH, Iizuka K, Smith HR, Scalzo AA, Yokoyama WM. Genetic control of natural killing and in vivo tumor elimination by the Chok locus. J Exp Med 1998; 188: 2243–2256. 103 Beebe AM, Mauze S, Schork NJ, Coffman RL. Serial backcross mapping of multiple loci associated with resistance to Leishmania major in mice. Immunity 1997; 6: 551–557. 104 Melanitou E, Joly F, Lathrop M, Boitard C, Avner P. Evidence for the presence of insulin-dependent diabetes-associated alleles on the distal part of mouse chromosome 6. Genome Res 1998; 8: 608–620.

105 Delano ML, Brownstein DG. Innate resistance to lethal mousepox is genetically linked to the NK gene complex on chromosome 6 and correlates with early restriction of virus replication by cells with an NK phenotype. J Virol 1995; 69: 5875–5877. 106 Dissen E, Ryan JC, Seaman WE, Fossum S. An autosomal dominant locus, Nka, mapping to the Ly-49 region of a rat natural killer (NK) gene complex, controls NK cell lysis of allogeneic lymphocytes. J Exp Med 1996; 183: 2197–2207. 107 Jansson AM, Jacobsson L, Luthman H, Lorentzen JC. Susceptibility to oil-induced arthritis linked to Oia2 on chromosome 4 in a DA(DA × PVG.1AV1) backcross. Transplant Proc 1999; 31: 1597–1599. 108 Ikeda H, Lethe B, Lehmann F et al. Characterization of an antigen that is recognized on a melanoma showing partial HLA loss by CTL expressing an NK inhibitory receptor. Immunity 1997; 6: 199–208. 109 Yen JH, Moore BE, Nakajima T et al. Major histocompatibility complex class I-recognizing receptors are disease risk genes in rheumatoid arthritis. J Exp Med 2001; 193: 1159–1167. 110 Depatie C, Muise E, Lepage P, Gross P, Vidal SM. High-resolution linkage map in the proximity of the host resistance locus Cmv1. Genomics 1997; 39: 154–163. 111 Depatie C, Chalifour A, Pare C, Lee SH, Vidal SM, Lemieux S. Assessment of Cmv1 candidates by genetic mapping and in vivo antibody depletion of NK cell subsets. Int Immunol 1999; 11: 1541–1551. 112 Depatie C, Lee SH, Stafford A et al. Sequence-ready BAC contig, physical, and transcriptional map of a 2-Mb region overlapping the mouse chromosome 6 host-resistance locus Cmv1. Genomics 2000; 66: 161–174. 113 Melanitou E, Simon-Chazottes D, Guenet JL, Rougeon F. The gene coding for the kidney androgen-regulated protein (Kap), maps between the Gapd and Kras-2 genes on mouse chromosome 6. Mamm Genome 1991; 1: 191–192. 114 Golub TR, Barker GF, Lovett M, Gilliland DG. Fusion of PDGF receptor beta to a novel ets-like gene, tel, in chronic myelomonocytic leukemia with t(5;12) chromosomal translocation. Cell 1994; 77: 307–316. 115 Brown MG, Zhang J, Du Y, Stoll J, Yokoyama WM, Scalzo AA. Localization on a physical map of the NKC-linked Cmv1 locus between Ly49b and the Prp gene cluster on mouse chromosome 6. J Immunol 1999; 163: 1991–1999. 116 Brennan J, Lemieux S, Freeman JD, Mager DL, Takei F. Heterogeneity among Ly-49C natural killer (NK) cells: characterization of highly related receptors with differing functions and expression patterns. J Exp Med 1996; 184: 2085–2090. 117 Makrigiannis AP, Anderson SK. Ly49 gene expression in different inbred mouse strains. Immunol Res 2000; 21: 39–47. 118 Lee SH, Gitas J, Zafer A et al. Haplotype mapping indicates two independent origins for the Cmv1s susceptibility allele to cytomegalovirus infection and refines its localization within the Ly49 cluster. Immunogenetics 2001; 53: 501–505. 119 Brown MG, Scalzo AA, Stone LR et al. Natural killer gene complex (Nkc) allelic variability in inbred mice: evidence for Nkc haplotypes. Immunogenetics 2001; 53: 584–591. 120 Makrigiannis AP, Pau AT, Saleh A, Winkler-Pickett R, Ortaldo JR, Anderson SK. Class I mhc-binding characteristics of the 129/j ly49 repertoire. J Immunol 2001; 166: 5034–5043. 121 Taylor BA, Wnek C, Kotlus BS, Roemer N, Mac Taggart T, Phillips SJ. Genotyping new BXD recombinant inbred mouse strains and comparison of BXD and consensus maps. Mamm Genome 1999; 10: 335–348. 122 Daniels KA, Devora G, Lai WC, O’Donnell CL, Bennett M, Welsh RM. Murine cytomegalovirus is regulated by a discrete subset of natural killer cells reactive with monoclonal antibody to Ly49H. J Exp Med 2001; 194: 29–44. 123 Anderson SK, Ortaldo JR, McVicar DW. The ever-expanding Ly49 gene family: repertoire and signaling. Immunol Rev 2001; 181: 79–89. 124 Ortaldo JR, Winkler-Pickett R, Mason AT, Mason LH. The Ly-

261

Genes and Immunity


262 125

126

127

128

129

130

131

132

133

134

135

136 137

138

139

49 family: regulation of cytotoxicity and cytokine production in murine CD3+ cells. J Immunol 1998; 160: 1158–1165. Mason LH, Gosselin P, Anderson SK, Fogler WE, Ortaldo JR, McVicar DW. Differential tyrosine phosphorylation of inhibitory versus activating Ly-49 receptor proteins and their recruitment of SHP-1 phosphatase. J Immunol 1997; 159: 4187–4196. Nakamura MC, Niemi EC, Fisher MJ, Shultz LD, Seaman WE, Ryan JC. Mouse Ly-49A interrupts early signaling events in natural killer cell cytotoxicity and functionally associates with the SHP-1 tyrosine phosphatase. J Exp Med 1997; 185: 673–684. Mason LH, Ortaldo JR, Young HA, Kumar V, Bennett M, Anderson SK. Cloning and functional characteristics of murine large granular lymphocyte-1: a member of the Ly-49 gene family (Ly-49G2). J Exp Med 1995; 182: 293–303. Kase A, Johansson MH, Olsson-Alheim MY, Karre K, Hoglund P. External and internal calibration of the MHC class I-specific receptor Ly49A on murine natural killer cells. J Immunol 1998; 161: 6133–6138. Andersson M, Freland S, Johansson MH et al. MHC class I mosaic mice reveal insights into control of Ly49C inhibitory receptor expression in NK cells. J Immunol 1998; 161: 6475–6479. Tormo J, Natarajan K, Margulies DH, Mariuzza RA. Crystal structure of a lectin-like natural killer cell receptor bound to its MHC class I ligand. Nature 1999; 402: 623–631. Natarajan K, Boyd LF, Schuck P, Yokoyama WM, Eliat D, Margulies DH. Interaction of the NK cell inhibitory receptor Ly49A with H-2Dd: identification of a site distinct from the TCR site. Immunity 1999; 11: 591–601. Lanier LL, Corliss BC, Wu J, Leong C, Phillips JH. Immunoreceptor DAP12 bearing a tyrosine-based activation motif is involved in activating NK cells. Nature 1998; 391: 703–707. Mason LH, Willette-Brown J, Mason AT, McVicar D, Ortaldo JR. Interaction of Ly-49D+NK cells with H-2Dd target cells leads to Dap-12 phosphorylation and IFN-gamma secretion. J Immunol 2000; 164: 603–611. Lanier LL, Bakker AB. The ITAM-bearing transmembrane adaptor DAP12 in lymphoid and myeloid cell function. Immunol Today 2000; 21: 611–614. Paloneva J, Kestila M, Wu J et al. Loss-of-function mutations in TYROBP (DAP12) result in a presenile dementia with bone cysts. Nat Genet 2000; 25: 357–361. Nakamura MC, Seaman WE. Ligand interactions by activating and inhibitory Ly-49 receptors. Immunol Rev 2001; 181: 138–148. Idris AH, Smith HR, Mason LH, Ortaldo JR, Scalzo AA, Yokoyama WM. The natural killer gene complex genetic locus Chok encodes Ly-49D, a target recognition receptor that activates natural killing. Proc Natl Acad Sci USA 1999; 96: 6330–6335. Ortaldo JR, Bere EW, Hodge D, Young HA. Activating Ly-49 NK receptors: central role in cytokine and chemokine production. J Immunol 2001; 166: 4994–4999. Karre K, Ljunggren HG, Piontek G, Kiessling R. Selective rejec-

Genes and Immunity

140

141 142 143

144

145

146

147 148

149

150

151

152

153

tion of H-2-deficient lymphoma variants suggests alternative immune defence strategy. Nature 1986; 319: 675–678. Farrell HE, Vally H, Lynch DM et al. Inhibition of natural killer cells by a cytomegalovirus MHC class I homologue in vivo. Nature 1997; 386: 510–514. Cerwenka A, Lanier LL. Natural killer cells, viruses and cancer. Nat Rev 2001; 1: 41–49. Alcami A, Koszinowski UH. Viral mechanisms of immune evasion. Immunol Today 2000; 21: 447–455. Beck S, Barrell BG. Human cytomegalovirus encodes a glycoprotein homologous to MHC class-I antigens. Nature 1988; 331: 269–272. Fahnestock ML, Johnson JL, Feldman RM, Neveu JM, Lane WS, Bjorkman PJ. The MHC class I homolog encoded by human cytomegalovirus binds endogenous peptides. Immunity 1995; 3: 583– 590. Leong CC, Chapman TL, Bjorkman PJ et al. Modulation of natural killer cell cytotoxicity in human cytomegalovirus infection: the role of endogenous class I major histocompatibility complex and a viral class I homolog. J Exp Med 1998; 187: 1681–1687. Cosman D, Fanger N, Borges L et al. A novel immunoglobulin superfamily receptor for cellular and viral MHC class I molecules. Immunity 1997; 7: 273–282. Yokoyama WM. Now you see it, now you don’t! Nat Immunol 2000; 1: 95–97. Wu J, Song Y, Bakker AB et al. An activating immunoreceptor complex formed by NKG2D and DAP10. Science 1999; 285: 730–732. Groh V, Rhinehart R, Randolph-Habecker J, Topp MS, Riddell SR, Spies T. Costimulation of CD8alphabeta T cells by NKG2D via engagement by MIC induced on virus-infected cells. Nat Immunol 2001; 2: 255–260. Bauer S, Groh V, Wu J et al. Activation of NK cells and T cells by NKG2D, a receptor for stress-inducible MICA. Science 1999; 285: 727–729. Cosman D, Mullberg J, Sutherland CL et al. ULBPs, novel MHC class I-related molecules, bind to CMV glycoprotein UL16 and stimulate NK cytotoxicity through the NKG2D receptor. Immunity 2001; 14: 123–133. Cerwenka A, Bakker AB, McClanahan T et al. Retinoic acid early inducible genes define a ligand family for the activating NKG2D receptor in mice. Immunity 2000; 12: 721–727. Diefenbach A, Jamieson AM, Liu SD, Shastri N, Raulet DH. Ligands for the murine NKG2D receptor: expression by tumor cells and activation of NK cells and macrophages. Nat Immunol 2000; 1: 119–126.

Note added in proof A recent online publication (Science, 11 April, 2002) reported that the ligand of Ly49H corresponds to the MHC-like protein m157 encoded by MCMV.