Characterization of long-term mixed donor-donor chimerism after double cord blood transplantation.

Double cord blood transplantation (DCBT) with two matched or partially matched cord blood units has been implemented successfully to circumvent the limitations of graft cell dose associated with single CBT. After DCBT, sustained haematopoiesis is derived almost exclusively from only one of the donated units. None the less, we previously observed two of six evaluable DCBT patients still having mixed donor-donor chimerism at 28 and 45 months post-transplantation, respectively. In the present study we utilize flow cytometry techniques to perform the first thorough analysis of phenotype and functionality of cord blood units in patients with mixed donor-donor chimerism. Our results suggest that the two stable cord blood units are different phenotypically and functionally: one unit shows more naive T cells, lower T cell cytokine production and higher frequencies of natural killer cells, the other shows higher frequencies of well-differentiated and functional lymphocytes. Additionally, in comparison with control patients having a single prevailing cord blood unit, the patients with donor-donor chimerism exhibit less overall T cell cytokine production and a smaller fraction of memory T cells. Furthermore, our results indicate that human leucocyte antigen-C match of donor units may partly explain the development of a donor-donor mixed chimerism.

Herpes simplex virus infection downmodulates NKG2D ligand expression.

Herpes simplex virus (HSV) type 1 infection may cause orofacial infections in humans. The virus resides in a latent form in neural ganglia and occasionally reactivates and infects epithelial cells. Natural killer (NK) cells have been implicated in immune control of herpes virus infections, possibly by downmodulating major histocompatibility complex (MHC) class I and by other, as yet unidentified, mechanisms. Upon HSV-1 infection of cell lines, surface levels of NKG2D ligands MHC class I related proteins (MIC) A and UL16 binding protein 2 were downmodulated due to late viral gene product(s). As also MHC class I levels were reduced by HSV-1, NK cell recognition of HeLa cells was not affected by infection. Total cellular MICA contents remained unchanged, suggesting masking, internalization or intracellular retention of MICA as possible mechanisms of viral downregualtion of MICA surface levels. Furthermore, NK cells from patients with active HSV-1 infection had a tendency towards increased expression level of the activating receptor NKG2D. These data support a role for NKG2D-MICA interactions in immune responses to HSV-1 reactivation.

The expression of human natural killer cell receptors in early life.

Natural killer (NK) cells play an important role in tumour immunosurveillance and the early defence against viral infections. Recognition of altered cells (i.e. infected- or tumour-cells) is achieved through a multiple receptor recognition strategy which gives the NK cells inhibitory or activating signals depending on the ligands present on the target cell. NK cells originate from the bone marrow where they develop and proliferate. However, further maturation processes and homeostasis of NK cells in peripheral blood are not well understood. To determine the proportions of cells and the expression of NK cell receptors, mononuclear cells from children at three time points during early childhood were compared, i.e. cord blood (CB), 2 and 5 years of age. The proportion of NK cells was high in CB, but the interferon-gamma (IFN-gamma) production low compared to later in life. In contrast, the proportion of T cells was low in CB. This may indicate a deviation of the regulatory function of NK cells in CB compared to later in life, implying an importance of innate immunity in early life before the adaptive immune system matures. Additionally, we found that the proportion of LIR-1(+) NK cells increased with increasing age while CD94(+)NKG2C(-) (NKG2A(+)) NK cells and the level of expression of NKG2D, NKp30 and NKp46 decreased with age. These age related changes in NK cell populations defined by the expression of activating and inhibitory receptors may be the result of pathogen exposure and/or a continuation of the maturation process that begins in the bone marrow.

Proliferative and cytotoxic response of human natural killer cells exposed to transporter associated with antigen-processing-deficient cells.

In transporter associated with antigen-processing (TAP)-deficient patients affected by a severe downmodulation of human leucocyte antigen class I (HLA-I) molecules, natural killer (NK) cells have an increased expression of the inhibitory receptor CD94/NKG2A. Focusing our attention on NK cells, we have investigated the phenotype, function and proliferative response of peripheral blood lymphocytes (PBLs) derived from healthy donors after coculturing with TAP (T2)- or HLA-I-deficient (721.221) cell lines and their related HLA-I-expressing transfectants (T3 and DT360, respectively). After 4 days, NK cells cocultured with T2 cells had a threefold increased CD94 expression compared to NK cells cocultured with T3. This increase was due to proliferation of the CD56brightCD94bright subset. In contrast, expression of other inhibitory receptors [killer cell immunoglobulin (Ig)-like receptors] was variable during time and was not related to HLA-I molecules expressed by stimulating cells. Similar results were obtained using HLA-I-deficient cells (721.221). The PBLs cocultured for 4 days with T2 cells displayed enhanced cytotoxic responses. The results suggest that CD56brightCD94bright NK cells are induced to proliferate and kill in response to a TAP-deficient environment. The changes seen in the NK-cell compartment were partially contributed by T lymphocytes present in the coculture. These data could explain the increased CD94 expression and autoimmune manifestations observed in TAP-deficient patients.

NK cells, MHC class I molecules and the missing self.

This article is based on a lecture presented at the Novartis Prize ceremony at the International Congress of Immunology in July 2001. It gives a personal and historical perspective on the research performed by the author and his colleagues during the development and pursuit of the model of 'missing-self recognition' for natural killer (NK) cells. This model is based on the idea that one important function of NK cells is to detect and eliminate cells because they fail to express normal self markers. Further mechanistic models predicted the existence of inhibitory major histocompatibility complex (MHC) class I specific receptors, later identified by the fellow Novartis laureates contributing in this issue. The article covers the first decade (1980-1990) of research on this concept. It discusses factors contributing to the formulation of a hypothesis, the use of predictions and experimental test models, the importance of international collaborations and reagent exchange, and several other aspects that allowed the progression of this research project. Finally, the perspective of today's knowledge is used to discuss some surprising findings where the missing-self hypothesis made the wrong predictions, or at least failed to make the correct ones.

The role of an exposed loop in the alpha2 domain in the mouse MHC class I H-2Dd molecule for recognition by the monoclonal antibody 34-5-8S and the NK-cell receptor Ly49A.

Natural killer (NK) cells express major histocompatibility complex (MHC) class I-specific inhibitory receptors. The region mediating the protective effect of the MHC class I molecule H-2Dd (Dd), recognized by the inhibitory receptor Ly49A, has been mapped to the alpha1/alpha2 domains. Here we have focused on an exposed loop in the N-terminal part of the alpha2 domain, which constitutes a major structural motif that differs between Dd (strong binding to Ly49A) and Db (weak binding to Ly49A at best). We mutated the residues 103, 104 and 107 in Dd to the corresponding amino acids in Db. The Dd mutant molecule retained the ability to be stabilized by a Dd-binding peptide. However, the mutation totally abolished the recognition by the conformational dependent monoclonal antibody (MoAb) 34-5-8S, known to inhibit the interaction between Dd and Ly49A. In addition, there was a marked impairment of the binding to Ly49A as evaluated by the ability of tetramers of the Dd mutant molecule to bind to Ly49A-transfected reporter cells and spleen cells. These results demonstrate that the introduced changes at positions 103, 104 and 107 directly or indirectly affect the epitopes for the MoAb 34-5-8S and the Ly49A receptor.

Human cytomegalovirus (HCMV)-infected endothelial cells and macrophages are less susceptible to natural killer lysis independent of the downregulation of classical HLA class I molecules or expression of the HCMV class I homologue, UL18.

A number of reports have suggested that human cytomegalovirus (HCMV)-infected fibroblasts are resistant to natural killer (NK) lysis, and that the HCMV-encoded human leucocyte antigen (HLA) class I homologue UL18 may be responsible for this effect. While fibroblasts are easy to infect in vitro, their role in HCMV pathogenesis in vivo is unclear. Here, we have established systems to address NK recognition of infected endothelial cells and macrophages, two important HCMV cellular reservoirs in vivo. The HCMV-infected endothelial cells exhibited increased resistance to NK killing, and, in most experiments, infected macrophages demonstrated a decreased susceptibility to NK lysis. Infection with the mutant HCMV strain RV670, lacking the genes US1-9 and US11 that are responsible for downregulation of HLA class I molecules, also led to decreased NK susceptibility. Furthermore, reduced NK susceptibility was independent of the expression of the HLA class I homologue UL18, since cells infected with the UL18Delta HCMV strain were also less susceptible to NK killing. These results suggest that HCMV-induced resistance to NK cytotoxicity in endothelial cells and macrophages is independent of known pathways that interfere with the expression of cellular HLA class I A, B and C surface antigens and the HCMV encoded class I homologue UL18.

Acquisition of external major histocompatibility complex class I molecules by natural killer cells expressing inhibitory Ly49 receptors.

Murine natural killer (NK) cells express inhibitory Ly49 receptors specific for major histocompatibility complex (MHC) class I molecules. We report that during interactions with cells in the environment, NK cells acquired MHC class I ligands from surrounding cells in a Ly49-specific fashion and displayed them at the cell surface. Ligand acquisition sometimes reached 20% of the MHC class I expression on surrounding cells, involved transfer of the entire MHC class I protein to the NK cell, and was independent of whether or not the NK cell expressed the MHC class I ligand itself. We also present indirect evidence for spontaneous MHC class I acquisition in vivo, as well as describe an in vitro coculture system with transfected cells in which the same phenomenon occurred. Functional studies in the latter model showed that uptake of H-2D(d) by Ly49A+ NK cells was accompanied by a partial inactivation of cytotoxic activity in the NK cell, as tested against H-2D(d)-negative target cells. In addition, ligand acquisition did not abrogate the ability of Ly49A+ NK cells to receive inhibitory signals from external H-2D(d) molecules. This study is the first to describe ligand acquisition by NK cells, which parallels recently described phenomena in T and B cells.

Synergistic effect of IFN-gamma and human cytomegalovirus protein UL40 in the HLA-E-dependent protection from NK cell-mediated cytotoxicity.

Human CMV (HCMV) has evolved several strategies to evade the immune system of the infected host. Here, we investigated the role of the HCMV-encoded protein UL40 in the modulation of NK cell lysis. UL40 carries in its leader sequence a nonameric peptide similar to that found in many HLA class I molecules leader sequences. This peptide up-regulates the expression of HLA-E, the ligand for the NK cell inhibitory receptor CD94/NKG2A. The UL40-encoded HLA-E-binding peptide was present in all HCMV clinical (4636, 13B, 109B, 3C) and laboratory (AD169) strains analyzed. However, transfection of UL40 in different cell lines (293T, 721.221, K562) did not consistently confer protection from NK lysis (as measured using NKL and the newly generated NK line Nishi), despite a moderate up-regulation of HLA-E. Interestingly, combined transfection and treatment with IFN-gamma increased the inhibitory effect, via an HLA-E- and CD94/NKG2A-dependent mechanism. Although cells transfected with UL40 derived from either AD169 or 3C showed protection from NK cell lysis, infection of fibroblasts with the viruses resulted in a strong inhibition only with the clinical strain 3C. Our results suggest that UL40 and IFN-gamma-dependent up-regulation of HLA-E is only one possible mechanism to avoid NK cell recognition of HCMV infected cells.

MHC class I recognition by NK receptors in the Ly49 family is strongly influenced by the beta 2-microglobulin subunit.

NK cell recognition of targets is strongly affected by MHC class I specific receptors. The recently published structure of the inhibitory receptor Ly49A in complex with H-2Dd revealed two distinct sites of interaction in the crystal. One of these involves the alpha1, alpha2, alpha3, and beta2-microglobulin (beta2m) domains of the MHC class I complex. The data from the structure, together with discrepancies in earlier studies using MHC class I tetramers, prompted us to study the role of the beta2m subunit in MHC class I-Ly49 interactions. Here we provide, to our knowledge, the first direct evidence that residues in the beta2m subunit affect binding of MHC class I molecules to Ly49 receptors. A change from murine beta2m to human beta2m in three different MHC class I molecules, H-2Db, H-2Kb, and H-2Dd, resulted in a loss of binding to the receptors Ly49A and Ly49C. Analysis of the amino acids involved in the binding of Ly49A to H-2Dd in the published crystal structure, and differing between the mouse and the human beta2m, suggests the cluster formed by residues Lys3, Thr4, Thr28, and Gln29, as a potentially important domain for the Ly49A-H-2Dd interaction. Another possibility is that the change of beta2m indirectly affects the conformation of distal parts of the MHC class I molecule, including the alpha1 and alpha2 domains of the heavy chain.

Purified MHC class I molecules inhibit activated NK cells in a cell-free system in vitro.

Natural killer cells have been shown to interact with MHC class I molecules via inhibitory receptors. However, it is not known whether the inhibition induced by MHC class I molecules requires other NK cell-target cell interactions. Thus, we examined whether purified MHC class I molecules alone were able to inhibit NK cell function. Purified H-2K(b) and H-2D(b) molecules inhibited the release of IFN-gamma from spleen (H-2(b))-derived lymphokine-activated killer (LAK) cell cultures stimulated by anti-NK1.1 antibody in a concentration-dependent manner. LAK cells generated from newborn mice that express low levels of MHC class I binding Ly49 inhibitory receptors were significantly less sensitive to inhibition by H-2K(b) compared to LAK cells from adult mice. Furthermore, LAK cells generated from spleen cells of Ly49C-transgenic mice were significantly more sensitive to inhibition by H-2K(b) compared to non-transgenic littermates. Taken together, the data indicate that MHC class I induced inhibition of NK cell mediated effector functions, as assessed by IFN-gamma release after NK1.1 triggering, does not require additional cell surface molecules other than MHC class I.

Human immunodeficiency virus type 1 Nef epitopes recognized in HLA-A2 transgenic mice in response to DNA and peptide immunization.

We investigated the immune response against a human immunodeficiency virus type 1 (HIV-1) nef DNA sequence administered epidermally in mice transgenic for the human major histocompatibility complex (MHC) class I molecule HLA-A201. Ten potential HLA-A2 binding 9-mer Nef peptides were identified by a computer-based search algorithm. By a cell surface MHC class I stabilization assay, four peptides were scored as good binders, whereas two peptides bound weakly to HLA-A2. After DNA immunization, cytotoxic T lymphocyte (CTL) responses were predominantly directed against the Nef 44-52, 81-89, and 85-93 peptides. Interestingly, the 44-52 epitope resides outside the regions of Nef where previously described CTL epitopes are clustered. Dominance among Nef-derived peptides did not strictly correlate with HLA-A2 binding, in that only one of the high-affinity binding peptides was targeted in the CTL response. The 44-52, 85-93, and 139-147 peptides also generated specific CTLs in response to peptide immunization. T helper cell proliferation was detected after stimulation with 20-mer peptides in vitro. Three Nef regions (16-35, 106-125, and 166-185) dominated the T helper cell proliferation. The implications of these results for the development of DNA-based vaccines against HIV is discussed.

Inhibition of human NK cell-mediated killing by CD1 molecules.

It is now well established that NK cells recognize classical and nonclassical MHC class I molecules and that such recognition typically results in the inhibition of target cell lysis. Given the known structural similarities between MHC class I and non-MHC-encoded CD1 molecules, we investigated the possibility that human CD1a, -b, and -c proteins might also function as specific target structures for NK cell receptors. Here we report that expression of CD1a, -b, or -c can partially inhibits target cell lysis by freshly isolated human NK cells and cultured NK lines. The inhibitory effects of CD1 molecules on NK cell could be shown upon expression of individual CD1 proteins in transfected NK-sensitive target cells, and these effects could be reversed by incubation of the target cells with mAbs specific for the expressed form of CD1. Inhibitory effects of CD1 expression on NK-mediated lysis could also be shown for cultured human dendritic cells, which represent a cell type that prominently expresses the various CD1 proteins in vivo. In addition, the bacterial glycolipid Ags known to be bound and presented by CD1 proteins could significantly augment the observed inhibitory effects on target cell lysis by NK cells.

T cell tolerance based on avidity thresholds rather than complete deletion allows maintenance of maximal repertoire diversity.

Given the flexible nature of TCR specificity, deletion or permanent disabling of all T cells with the capacity to recognize self peptides would severely limit the diversity of the repertoire and the capacity to recognize foreign Ags. To address this, we have investigated the patterns of CD8+ CTL reactivity to a naturally H-2Kb-presented self peptide derived from the elongation factor 1alpha (EF1alpha). EF1alpha occurs as two differentially expressed isoforms differing at one position of the relevant peptide. Low avidity CTLs could be raised against both variants of the EF1alpha peptide. These CTLs required 100-fold more peptide-H-2Kb complexes on the target cell compared with CTLs against a viral peptide, and did not recognize the naturally expressed levels of EF1alpha peptides. Thus, low avidity T cells specific for these self peptides escape tolerance by deletion, despite expression of both EF1alpha isoforms in dendritic cells known to mediate negative selection in the thymus. The low avidity in CTL recognition of these peptides correlated with low TCR affinity. However, self peptide-specific CTLs expressed elevated levels of CD8. Furthermore, CTLs generated against altered self peptide variants displayed intermediate avidity, indicating cross-reactivity in induction of tolerance. We interpret these data, together with results previously published by others, in an avidity pit model based on avidity thresholds for maintenance of both maximal diversity and optimal self tolerance in the CD8+ T cell repertoire.

Innate and adaptive immunity to tumors: IL-12 is required for optimal responses.

We have investigated the importance of endogenously produced IL-12 in innate and adaptive immunity to tumor transplants. The immunogenic lymphoma RMA and its TAP-deficient variant RMA-S were tested for rejection responses by normal and IL-12-deficient mice. IL-12 was crucial for the immunity induced by one immunization with irradiated RMA cells, as well as for in vivo priming of a CTL response in mixed lymphocyte tumor cultures against this MHC class I-expressing tumor. The defective in vivo response could be overcome by multiple immunizations. In further studies of in vitro CTL responses, we found that IL-12 production from either the antigen-pulsed dendritic cells or from host cells was necessary to obtain strong CTL responses. In the complete absence of IL-12, no or only very weak responses could be detected. NK cell-mediated innate resistance, as assessed in non-immunized mice inoculated with a threshold dose of RMA-S cells, also required IL-12. However, NK cells with reduced activity were present in IL-12-deficient mice and contributed to innate resistance, as demonstrated with lower cell dose challenges. In conclusion, IL-12 is required for optimal adaptive and innate responses against tumors.

Human cytomegalovirus strain-dependent changes in NK cell recognition of infected fibroblasts.

NK cells play a key role in the control of CMV infection in mice, but the mechanism by which NK cells can recognize and kill CMV-infected cells is unclear. In this study, the modulation of NK cell susceptibility of human CMV (hCMV)-infected cells was examined. We used a human lung and a human foreskin fibroblast cell line infected with clinical isolates (4636, 13B, or 109B) or with laboratory strains (AD169, Towne). The results indicate that all three hCMV clinical isolates confer a strong NK resistance, whereas only marginal or variable effects in the NK recognition were found when the laboratory strains were used. The same results were obtained regardless of the conditions of infection, effector cell activation status, cell culture conditions, and/or donor-target cell combinations. The NK cell inhibition did not correlate with HLA class I expression levels on the surface of the target cell and was independent of the leukocyte Ig-like receptor-1, as evaluated in Ab blocking experiments. No relevant changes were detected in the adhesion molecules ICAM-I and LFA-3 expressed on the cell surface of cells infected with hCMV clinical and laboratory strains. We conclude that hCMV possesses other mechanisms, related neither to target cell expression of HLA-I or adhesion molecules nor to NK cell expression of leukocyte Ig-like receptor-1, that confer resistance to NK cell recognition. Such mechanisms may be lost during in vitro passage of the virus. These results emphasize the differences between clinical hCMV isolates compared with laboratory strains.

Differential effects on T cell and NK cell development by tissue-specific expression of H-2D(d) transgene.

The effect of tissue-specific expression of the MHC class I molecule H-2D(d) on T cell and NK cell specificity was studied in transgenic mice expressing the H-2D(d) gene under the control of the mouse metallothionein-I promoter. MTD mice expressed high amounts of H-2D(d) in the liver, intestine and testis, but only minute amounts in the thymus, spleen and kidney. Zinc administration resulted in a 1.5- and 8.5-fold increase in H-2D(d) expression in the liver and the intestine, respectively, but did not affect expression in the other organs tested. T cell tolerance developed towards H-2D(d) in MTD mice, even in the absence of zinc. In contrast, NK cell-mediated natural resistance against lymphoma grafts was not seen in MTD mice, despite zinc administration. NK cells in MTD mice also failed to develop self tolerance to H-2D(d). The lack of functional effects did not result from inability of NK cells in MTD mice to interact with H-2D(d), as down-regulation of Ly49A receptor expression was observed on liver NK cells in MTD mice. Our data reveal a difference between T cells and NK cells in their requirements for MHC class I molecules in specificity development.

Recruitment and activation of natural killer (NK) cells in vivo determined by the target cell phenotype. An adaptive component of NK cell-mediated responses.

Natural killer (NK) cells can spontaneously lyse certain virally infected and transformed cells. However, early in immune responses NK cells are further activated and recruited to tissue sites where they perform effector functions. This process is dependent on cytokines, but it is unclear if it is regulated by NK cell recognition of susceptible target cells. We show here that infiltration of activated NK cells into the peritoneal cavity in response to tumor cells is controlled by the tumor major histocompatibility complex (MHC) class I phenotype. Tumor cells lacking appropriate MHC class I expression induced NK cell infiltration, cytotoxic activation, and induction of transcription of interferon gamma in NK cells. The induction of these responses was inhibited by restoration of tumor cell MHC class I expression. The NK cells responding to MHC class I-deficient tumor cells were approximately 10 times as active as endogenous NK cells on a per cell basis. Although these effector cells showed a typical NK specificity in that they preferentially killed MHC class I-deficient cells, this specificity was even more distinct during induction of the intraperitoneal response. Observations are discussed in relation to a possible adaptive component of the NK response, i.e., recruitment/activation in response to challenges that only NK cells are able to neutralize.

Visualization of inhibitory Ly49 receptor specificity with soluble major histocompatibility complex class I tetramers.

Murine natural killer (NK) cells are inhibited from killing their targets by the interaction between inhibitory, C-type lectin like Ly49 receptors and major histocompatibility complex (MHC) class I molecules. The receptors have overlapping specificity, and it has been difficult to analyze specific aspects of the interaction between different Ly49 receptors and their respective ligands. We have addressed this problem using tetramers of bacterially expressed, non-glycosylated, MHC class I molecules refolded with different peptides. Our results indicate that this technology is useful for analysis of Ly49 receptor specificity as well as for monitoring of NK cell subsets, with the following major conclusions emerging from this study: (1) tetramers of H-2D(d) bound the Ly49A receptor; the MHC associated glycan, previously suggested to be involved in recognition by this receptor, is thus not required for Ly49A receptor binding; (2) in support and extension of a recent report indicating peptide selectivity in the recognition of H-2K(b) by Ly49C(+) cells, H-2K(b) tetramer binding to Ly49C receptors was strongly influenced by the peptide presented by the MHC class I molecule; (3) tetramer binding allowed visualization of interactions that have not previously been detected in functional studies, such as the recognition of H-2D(b) by Ly49A and Ly49C.