Original Manuscript

Leukemia (2005) 19, 1446–1451. doi:10.1038/sj.leu.2403839; published online 23 June 2005

Stem Cell Transplantation

A defined donor activating natural killer cell receptor genotype protects against leukemic relapse after related HLA-identical hematopoietic stem cell transplantation

S Verheyden1, R Schots2, W Duquet3 and C Demanet1

  1. 1HLA Laboratory, Academisch ziekenhuis–Vrije Universiteit Brussel (VUB), Brussels, Belgium
  2. 2Department of Medical Oncology and Hematology, Academisch ziekenhuis–Vrije Universiteit Brussel (VUB), Brussels, Belgium
  3. 3Department of Human Biometry and Biomechanics, Vrije Universiteit Brussel (VUB), Brussels, Belgium

Correspondence: Dr C Demanet, HLA Laboratory, Academisch Ziekenhuis–Vrije Universiteit Brussel, Laarbeeklaan 105, 1090 Brussels, Belgium. Fax: +32 2 477 67 27; E-mail: christian.demanet@az.vub.ac.be

Received 10 November 2004; Accepted 12 May 2005; Published online 23 June 2005.



Killer cell immunoglobulin-like receptors (KIRs) recognize different groups of Human Leukocyte Antigen (HLA) class I alleles and are expressed by natural killer (NK) cells and some T lymphocytes. NK cell cytotoxicity is triggered by failure to recognize the appropriate HLA class I ligand on target cells. Recently, it has been shown that HLA class I ligand incompatibility in the graft-versus-host (GvH) direction is associated with a better outcome in haploidentical hematopoietic stem cell transplantation (HSCT). Since KIR genotypes are very diverse in the population, we explored whether or not the donor KIR genotype could affect the graft-versus-leukemia (GvL) effect in the related HLA-identical HSCT setting. We determined the KIR and HLA genotypes of 65 HLA-identical patient–donor siblings. We found that the presence of two activating KIRs, 2DS1 and 2DS2, in the donor was significantly associated with a decreased leukemic relapse rate (P=0.03; OR=0.18; 95% CI: 0.037–0.88). Moreover, the probability of relapse at 5 years was significantly lower for patients who received a graft from a donor with the 2DS1(+)2DS2(+) genotype than for those who received a transplant from other donors (17 vs 63%, respectively; P=0.018). In conclusion, this study suggests that a joint effect of these two selected activating KIRs in the donor might confer some protection against leukemic relapse.


NK cells, KIR genes, leukemic relapse, graft-versus-leukemia effect, related HLA-identical hematopoietic stem cell transplantation



Natural killer (NK) cells are important effector cells in the innate host defense because of their capacity to lyse viral-infected and tumor cells spontaneously.1 In recent years, much progress has been made in the characterization of NK cell receptors. NK cells possess three main receptor families: natural cytotoxicity receptors (NCRs), C-type lectin receptors (CD94/NKG2) and killer cell immunoglobulin-like receptors (KIRs).2, 3, 4 The most polymorphic receptor gene family is the one encoding the KIRs. The KIRs belong to the immunoglobulin (Ig) superfamily that recognizes specific human leukocyte antigen (HLA) class I molecules and are found on the surface of NK cells and of some T lymphocyte subsets (www.ncbi.nlm.nih.gov/PROW). In humans, the majority of the HLA class I loci involved in the KIR interactions are determined by the recognition of HLA-C molecules. An allelic dimorphism in amino-acid residues 77 and 80 determines the specificity of KIRs for HLA-C allotypes.5, 6 KIR2DL1 recognize group 2 HLA-C molecules (HLA-C2) having asparagine at position 77 and lysine at position 80 (Cw*02, 0307, 0310, 0315, 04, 05, 06, 0707, 0709, 1204, 1205, 15, 1602, 17, 18), whereas KIR2DL2/2DL3 bind group 1 HLA-C molecules (HLA-C1), which include serine 77 and asparagine 80 (Cw*01, 03, 07, 08, 12, 13, 14, 1507, 16).5, 6 Not only does every HLA-C allotype interact with a KIR, but there also exists KIRs that interact with certain HLA-A and -B alleles (KIR3DL1 for HLA-Bw4 and KIR3DL2 for HLA- A3/11).7, 8 It has been suggested that the activating KIRs (2DS1, 2DS2 and 3DS1) recognize the same HLA class I ligands as their inhibitory counterparts, though with a lower affinity, because of the similarity in the extracellular domain.9, 10, 11, 12 The ligands for most activating KIRs have not yet been well defined, although non-HLA proteins, such as foreign antigens expressed on infected cells13, 14 or tumor-specific antigens15 may serve as candidate ligands.

Variations in gene number, gene content and allelic polymorphism within individual KIR genes account for the high diversity in human KIR haplotypes among individuals.16, 17, 18 Moreover, KIRs are clonally expressed in a stochastic manner in the peripheral blood, although each clone will express at least one inhibitory receptor, by which they are even able to sense single HLA class I allelic losses on target cells.19 One of the mechanisms by which NK cells are activated is via disruption of the dominant inhibitory signal by downregulation of a specific HLA class I molecule. This frequently occurs in certain viral-infected and tumor cells to evade T-cell recognition.19, 20, 21, 22, 23 This 'missing-self' principle was exploited in related haploidentical24, 25 and unrelated HLA mismatched HSCT.26, 27 In these transplantation settings, both the recipient and donor are mismatched for one or more HLA class C (-B) alleles and this may result in recipients lacking the appropriate HLA class I ligands recognized by the donor inhibitory KIRs. This could favor donor NK cell activation with a potent antileukemic effect.24, 25, 26, 27

Most patients receive their graft from an HLA-identical donor. Therefore, we analyzed whether or not certain donor KIR genotypes could be associated with a graft-versus-leukemia (GvL)-effect after related HLA-identical HSCT. Since KIR and HLA haplotypes segregate independently, the KIR type of the related HLA-identical donor will mostly (75%) differ from that of the patient. The HLA-identical transplant situation permits one to investigate the influence of KIR genes on transplant outcome without the confounding effects of HLA disparity.


Patients and methods

Patient variables and definitions

The KIR genotype was studied in HLA-identical patient–donor siblings. All family members (father, mother, siblings) were typed for HLA-A and -B by serology and DRB1 by molecular techniques. Patient and sibling donor were finally typed with molecular techniques for HLA-A, -B, -C, DRB1, DQB1 and DPB1 (Biotest Seralc, Kortenberg and Innolipa®, Innogenetics, Ghent, Belgium).

A total of 65 consecutive patients with leukemia who received a first transplant at our institution between 1991 and 2002 were included in the study. Two patients who rejected their graft were excluded.

The median age of the patients was 34 years (range, 5–53). Diagnosis at the time of transplantation was 22 patients with acute myeloid leukemia (AML), of whom 15 were in first, five in second and two in third complete remission; 27 patients with chronic myeloid leukemia (CML), of whom 24 were in first, one in second chronic phase and two in accelerated phase; 16 patients with acute lymphoid leukemia (ALL), of whom 10 were in first and six in second complete remission. Patients with acute leukemia in first complete remission and CML in first chronic phase were considered as standard risk patients, representing 75% of cases.

All patients were given myeloablative therapy, which included total body irradiation (except for two patients) and chemotherapy. The source of stem cells was bone marrow in 47 (72%) and peripheral blood in 18 (28%) cases. All patients received either standard cyclosporin in combination with methotrexate (n=31) or T-cell depletion (n=34) as graft-versus-host disease (GvHD) prophylaxis. The mean number of T-cells infused after T-cell depletion was 4.53 times 105/kg (range 0.1–14) as compared to 2.08 times 107/kg (range 1.08–4.47) in non-T-cell-depleted grafts, corresponding to a plusminus2 log depletion.

Acute and chronic GVHD were diagnosed according to standard criteria. Only patients requiring systemic glucocorticosteroid therapy were considered as having clinically significant chronic GvHD. Engraftment was defined as the achievement of durable trilinear hematopoietic recovery. Relapse was defined as recurrence of disease after transplantation, as based on cytomorphological or histological criteria. Transplant-related mortality (TRM) was defined as death due to any reason without evidence of relapse. Overall survival (OS) was determined from the time between transplantation and death due to any cause.

KIR genotyping

The sequence specific primer-polymerase chain reaction (PCR-SSP) technique was performed for the amplification of five groups of inhibitory (KIR2DL1-3 and KIR3DL1-2) and six groups of activating KIR genes (KIR2DS1-5 and KIR3DS1). The PCR-SSP assays were based on conditions and primers that have been previously reported.28 In addition, KIR2DS4-specific primers were used for the amplification of 221/199 bp region in exon 5 of the 2DS4 gene.29

Statistical analysis

The associations between baseline characteristics as well as KIR genes with transplant outcome variables were analyzed by Fisher's exact test (two-tailed). Probability of relapse, OS, TRM, acute and chronic GVHD were calculated by using the method of Kaplan–Meier and compared with the log-rank test. Multiple stepwise logistic regression analysis (SPSS software version 12) was performed to test the influence of the different variables (baseline characteristics and KIR genes) on relapse, which was the primary endpoint of this study. Significance was established at Pless than or equal to0.05.



Clinical outcome

The median follow-up time for the 65 patients included in the study was 20 months (range: 1–134 months). Patients who survived more than 1 month were considered to be appropriate for analyzing the following transplant outcome variables. Only two patients died within 2 months after HSCT.

At the time of analysis, 25 patients (38.5%) relapsed and 13 patients (20%) died from a transplant-related event. Acute and chronic GvHD were observed in 18 (27.7%) and 10 (16.9%) patients, respectively.

Patient and transplant characteristics

We investigated the influence of different baseline variables on the end points of relapse, TRM, acute and chronic GvHD (Table 1). None of these factors regarding patient age, gender, risk category, diagnosis, disease stage, source of stem cells, GvHD prophylaxis and identical donor and recipient KIR genotypes were found to have a significant influence on relapse, TRM or acute GVHD. However, chronic GvHD was significantly more frequent in transplantations with no T-cell depletion (P<0.05) (Table 1).


Univariate comparison of 11 individual KIR frequencies between 65 patients and their HLA-identical related donors was conducted before transplantation. No significant differences in the 11 KIR frequencies were observed between patients and donors (data not shown). We further examined if either the presence or absence of any individual KIR gene of the donor had an influence on relapse. Therefore, the frequencies of donor KIR genes in patients who relapsed (n=25) were compared with those who did not relapse (n=40) (Figure 1). Nearly all the inhibitory KIR genes (2DL1, 2DL3, 3DL1 and 3DL2) were present in the majority of the donors. Most variations between the two groups were observed in the number of activating receptors. We found higher frequencies of all donor activating KIR genes in the non-relapsing group compared to the relapsing group, but no significant differences in the frequencies of individual KIR genes were found. There was no significant correlation between the total number of activating receptors possessed by a donor and relapse rate (P=0.147), suggesting that the type of KIR and not the number of KIRs is more important.

Figure 1.
Figure 1 - Unfortunately we are unable to provide accessible alternative text for this. If you require assistance to access this image, please contact help@nature.com or the author

Frequencies of donor KIR genes in relapsing and non-relapsing patients. The PCR-SSP technique was performed for the amplification of five groups of inhibitory KIRs (KIR2DL1-3 and KIR3DL1-2) and six groups of activating KIRs (KIR2DS1-5 and KIR3DS1). The frequency of 2DS4 included only individuals who possessed at least one copy of the full-length form of the KIR2DS4 gene. The frequencies of donor KIR genes (%) were evaluated in patients who did not relapse (n=40) and patients who relapsed (n=25).

Full figure and legend (17K)

We showed that the combined presence of two activating KIRs, 2DS1 and 2DS2, in donors was significantly associated with a decreased incidence of relapse (P=0.03; OR=0.18; 95% CI: 0.037–0.88). Only two of the 15 patients who were given HSCT from a 2DS1(+)2DS2(+) donor, relapsed. No other possible combination with two KIR genes showed a significant association with relapse rate (data not shown). Moreover, in the multivariate analysis, only this combined presence of the activating 2DS1 and 2DS2 in the donor (P=0.016) and a younger patient age (P=0.046) were significantly associated with protection against relapse. The Kaplan–Meyer estimates for relapse at 5 years were 17 and 61.3% for patients who received a graft from a 2DS1(+)2DS2(+) donor and for those who received a transplant from donors who were positive for only one of these receptors or negative for both KIR2DS1 and KIR2DS2, respectively (P=0.018) (Figure 2a). The 5-year OS also tended to be better (59.2 and 35.5% in the 2DS1(+)2DS2(+) and other donors, respectively, P=0.109) (Figure 2b).

Figure 2.
Figure 2 - Unfortunately we are unable to provide accessible alternative text for this. If you require assistance to access this image, please contact help@nature.com or the author

Long-term outcome of patients receiving a graft from donors who are positive for both KIR2DS1 and KIR2DS2 vs donors who are positive for only one of these receptors or negative for both KIR2DS1 and KIR2DS2: significantly lower probability of relapse and a superior OS in patients who were given HSCT from donors with a 2DS1(+)2DS2(+) KIR genotype. (a) Kaplan–Meier plot to estimate the probability of relapse. (b) Kaplan–Meier plot to estimate the probability of OS.

Full figure and legend (37K)

Based on recent clinical studies that showed that NK cell activity would be more pronounced in patients with myeloid leukemia, relapse rates were analyzed separately in myeloid (AML+CML) and lymphoid (ALL) leukemia. We observed that the relapse rate in myeloid patients who received a graft from a 2DS1(+)2DS2(+) donor (2 of 12 patients relapsed) was significantly lower than in comparison with the myeloid patients who received a transplant from other donors (19 of 37 patients relapsed) (P=0.0471). None of the lymphoid patients receiving a graft from a 2DS1(+)2DS2(+) donor relapsed (0 of 3 patients relapsed in the 2DS1(+)2DS2(+) group vs 4 of 13 patients with transplants from other donors).

Additionally, we analyzed if there was a potential influence of the different HLA-C groups (C1C1, C1C2 and C2C2) only or in combination with the activating KIRs on relapse rate, but no such influences were seen (data not shown).

Finally, there were no significant differences in the patient and transplant characteristics between patients receiving a graft from a 2DS1(+)2DS2(+) donor and those receiving a graft from donors who were positive for only one of these receptors or negative for both KIR2DS1 and KIR2DS2 (data not shown).

Transplant-related mortality and GVHD

TRM, acute or chronic GvHD rates were not significantly associated with the presence of any particular KIR gene or the combined presence of KIR2DS1 and KIR2DS2 in the donors (data not shown). We also did not find any significant influence of the total number of donor activating KIRs on these end points.

However, the incidence of acute GvHD was decreased in patients who were heterozygous for HLA-C1 and C2 alleles. Only three of the 24 patients developed acute GvHD compared with 15 of 41 patients who were homozygous for C1 or C2 alleles (P=0.046; OR=0.25; 95% CI: 0.063–0.97).



In this study, we analyzed whether or not, certain KIRs in the donor could be associated with a GvL-effect following a related HLA-identical HSCT. In this situation, the influence of KIR genes on transplant outcome could be investigated in a direct way without the confounding effects of HLA disparity. The first interesting finding was that the frequencies of all activating KIRs of the donor were higher in the non-relapsing than in the relapsing patient group. However, there was no significant association between the total number of activating receptor possessed by a donor and relapse rate (P=0.147). A more detailed analysis revealed that the combined presence of two activating KIRs, 2DS1 and 2DS2, in the donor was significantly associated with a decreased relapse rate. Only two of the 15 patients (13.3%) who received HSCT from a 2DS1(+)2DS2(+) donor relapsed compared to 46% of the control group. These data suggest that the presence of these two activating KIRs, 2DS1 and 2DS2 is important in the protection against leukemic relapse as there was no correlation between the total number of activating KIRs possessed by the donor or any other KIR gene pair and relapse rate. The 5-year OS also tended to be superior in the 2DS1(+)2DS2(+) group, but did not reach statistical significance.

Further support that certain activating KIRs can be implicated in the immune response to diseases came from a recent study, which demonstrated that the activating KIR3DS1 in combination with HLA-Bw4 was associated with delayed progression to AIDS in individuals infected with the human immunodeficiency virus type I.30 Along the same line, another study reported that the activating KIR2DS1 and/or KIR2DS2 receptors in the absence of specific HLA-C alleles, were associated with psoriatic arthritis.31

It has been shown that HLA-C molecules bind KIR2D receptors and, therefore, could play a major role in the regulation of NK cell activity.5, 6 We could not find a significant effect of the HLA-C groups or activating KIR-HLA-C ligand interactions on relapse rate in this study. However, we cannot exclude that other non-HLA ligands can serve as candidate ligands, because not much is known about the activating KIR ligands except for the binding of HLA-C (HLA-B), which is much weaker than the binding of these molecules by inhibitory KIRs. In addition, it has been reported that leukemic cells can have downregulated expression of HLA class I molecules, which could decrease the threshold for NK cell activation by diminishing the interactions with their relevant inhibitory KIRs.21, 22, 23 A balance between the downregulation of HLA molecules and a joint effect of the two appropriate KIRs, may perhaps account for the mechanism by which residual leukemic cells are eliminated. A similar study determined the impact of donor KIR and recipient HLA-C types on outcome following related HLA-identical HSCT.32 The study demonstrated a poor survival rate in patients with myeloid leukemia who did not carry a group 1 HLA-C and whose donors were also positive for the activating KIR2DS2. No significant effect of activating KIR/HLA-C ligand combinations on relapse rate was found, but no effort was made to correlate the KIR genotype by itself to relapse.

As Ruggeri et al24, 25 demonstrated that in haploidentical HSCT more AML than ALL cells were killed in vitro by KIR-HLA mismatched NK clones, we have reanalyzed our data according to lymphoid and myeloid leukemia. A significant lower relapse rate was noted in myeloid patients who received a graft from 2DS1(+)2DS2(+) donors compared to all other donors. Less relapse was also observed in lymphoid leukemia patients but, unfortunately, the group was too small for statistical analysis.

Varying results have been reported about the beneficial effect of KIR ligand incompatibility seen in patients with AML following unrelated27, 33, 35 and related34 HSCT. Differences in patient characteristics and transplantation procedures may account for discrepancies in the results from these various studies. It has been demonstrated that disease status and the number of T cells in the graft are critical variables that can affect transplantation outcome. There was no significant difference with respect to disease status between patients receiving a graft from 2DS1(+)2DS2(+) and other donors. In contrast to other studies,27, 32, 34, 35 most of our patients (75%) were in first complete remission or in first chronic phase and it is apparent that the NK cell-mediated effect in these patients is more beneficial than in patients in the advanced disease phase. Furthermore, one of the two relapsed patients in the 2DS1(+)2DS2(+) group was in an advanced disease stage at transplant.

Several studies showed that the number of T cells in the graft could influence NK cell alloreactivity caused by an HLA mismatch.27, 33, 34, 35 Moreover, haploidentical transplants24, 25 require a high CD34+ cell dose and an extensive T-cell depletion (104 T cells/kg); which are typically associated with a rapid recovery of NK cells. In our study, 31 HLA-identical transplants were not T cell depleted and contained an average of 2.08 times 107 T cells/kg, which is much higher than in the study of Ruggeri et al.24, 25 Additionally, the remaining 34 transplants that were T cell depleted resulted in an average number of T cells of 4.53 times 105/kg, corresponding to plusminus2 log depletion. Consequently, the T-cell depletion in our HLA identical transplants were less vigorous compared to the 4-log T-cell depletion applied in HLA incompatible transplants. Furthermore, Giebel et al27 applied a uniform GvHD prophylaxis and were able to demonstrate that the type of immunosuppressive treatment could influence the beneficial effects of the alloreactive NK cells. In contrast, in our study, the GvHD prophylaxis involved either post-transplant immune suppression (cyclosporin and MTX) or T-cell depletion. We could not, however, exclude the possibility that the use of these two GvHD prophylactic approaches may have had a different effect on the schedule of NK cell recovery. Nevertheless, there was no significant difference regarding GvHD prophylaxis between the 2DS1(+) 2DS2(+) and the control group. Furthermore, the presence of 2DS1(+)2DS2(+) in the T-cell depleted group (2 of 10 patients with a 2DS1(+)2DS2(+) donor relapsed vs 14 of 24 patients with other donors) and in the non-T-cell depleted group (0 of five patients with a 2DS1(+)2DS2(+) donor relapsed vs 9 of 17 patients with other donors) was still associated with a decreased relapse rate, although not significant because of the low numbers. These data suggest that the effect of 2DS1(+)2DS2(+) on relapse rate is independent of the GvHD prophylaxis. Finally, a multivariate analysis that considered all variables influencing relapse, showed that KIR2DS1 and KIR2DS2 in the donor and a younger patient age were the only independent factors predicting a decreased risk of relapse.

The 'missing-self' principle exploited in related haploidentical24, 25 and unrelated HLA mismatched HSCT27, 33, 35 was based only on the HLA genotyping and not on the KIR genotype of the donor. We went a step further and studied the HLA genotype as well as the KIR genotype of the recipients and donors, all in the quest to possibly understand in more detail the mechanism of NK cell activity. In the future, we should also analyze the expression of these KIRs, because the presence of certain KIR variants should not be taken as evidence of KIR expression.36 Unfortunately, there are not yet antibodies available that are able to discriminate the inhibitory from the activating KIRs and therefore, the KIR phenotype cannot be analyzed accurately.

In conclusion, we have found that the combined presence of two donor activating KIRs, 2DS1 and 2DS2, was the most significant predictor for protection against leukemic relapse in our related HLA-identical HSCT population. We are currently investigating whether or not the donor 2DS1(+)2DS2(+) genotype also prevents against relapse in unrelated HLA-identical HSCT and further prospective studies involving larger numbers of transplantation studies have to be done to confirm our results.



  1. Miller JS. The biology of natural killer cells in cancer, infection, and pregnancy. Exp Hematol 2001; 29: 1157−1168. | Article | PubMed | ISI | ChemPort |
  2. Biassoni R, Cantoni C, Pende D, Sivori S, Parolini S & Vitale M et al.. Human natural killer cell receptors and co-receptors. Immunol Rev 2001; 181: 203−214. | Article | PubMed | ISI | ChemPort |
  3. Farag SS, Fehniger T, Ruggeri L, Velardi A & Caligiuri MA. Natural killer cell receptors: new biology and insights into the graft-versus-leukemia effect. Blood 2002; 100: 1935−1947. | Article | PubMed | ISI | ChemPort |
  4. Natarajan K, Dimasi N, Wang J, Mariuzza R & Margulies DH. Structure and function of natural killer cell receptors: Multiple molecular solutions to self, nonself discrimination. Annu Rev Immunol 2002; 20: 853−885. | Article | PubMed | ISI | ChemPort |
  5. Colonna M, Brooks EG, Falco M, Ferrara GB & Strominger JL. Generation of allospecific natural killer cells by stimulation across a polymorphism of HLA-C. Science 1993; 260: 1121−1124. | PubMed | ISI | ChemPort |
  6. Moretta A, Vitale M, Bottino C, Orengo AM, Morelli L & Augugliaro R et al.. P58 molecules as putative receptors for major histocompatibility complex (MHC) class I molecules in human natural killer (NK) cells: anti-P58 antibodies reconstitute lysis of MHC class I-protected cells in NK clones displaying different specificities. J Exp Med 1993; 178: 597−604. | Article | PubMed | ISI | ChemPort |
  7. Gumperz JE, Litwin V, Phillips JH, Lanier LL & Parham P. The Bw4 public epitope of HLA-B molecules confers reactivity with NK cell clones that express NKB1, a putative HLA receptor. J Exp Med 1995; 181: 1133−1144. | Article | PubMed | ISI | ChemPort |
  8. Döhring C, Scheidegger D, Samaridis J, Cella M & Colonna M. A human killer inhibitory receptor specific for HLA-A. J Immunol 1996; 156: 3098−3101. | PubMed | ISI | ChemPort |
  9. Valés-Gomez M, Reyburn HT, Erskine RA & Strominger J. Differential binding to HLA-C of p50-activating and p58-inhibitory natural killer cell receptors. Proc Natl Acad Sci USA 1998; 95: 14326−14331. | Article | PubMed | ISI | ChemPort |
  10. Moretta A, Sivori S, Vitale M, Pende D, Morelli L & Augugliaro R et al.. Existence of both inhibitory (p58) and activating (p50) receptors for HLA-C molecules in human natural killer cells. J Exp Med 1995; 182: 875−884. | Article | PubMed | ISI | ChemPort |
  11. Biassoni R, Cantoni C, Falco M, Verdiani S, Bottino C & Vitale M et al.. The human leucocyte antigen (HLA)-C-specific 'activating' or 'inhibitory' natural killer cell receptors display highly homologous extracellular domains but differ in their transmembrane and intracytoplasmic portions. J Exp Med 1996; 183: 645−650. | Article | PubMed | ISI | ChemPort |
  12. Valés-Gomez M, Reyburn HT, Mandelboim M & Strominger J. Kinetics of interaction of HLA-ligands with natural killer cell inhibitory receptors. Immunity 1998; 9: 337−344. | Article | PubMed | ISI | ChemPort |
  13. Smith HR, Heusel JW, Mehta IK, Kim S, Dorner BG & Naidenko OV et al.. Recognition of a virus-encoded ligand by a natural killer cell activation receptor. Proc Natl Acad Sci USA 2002; 99: 8826−8831. | PubMed | ChemPort |
  14. Arase H, Mocarski ES, Campbell AE, Hill AB & Lanier LL. Direct recognition of cytomegalovirus by activating and inhibitory NK cell receptors. Science 2002; 296: 1323−1326. | Article | PubMed | ISI | ChemPort |
  15. Katz G, Gazit R, Arnon TI, Gonen-Gross T, Tarcic G & Markel G et al.. MHC class I-independent recognition of NK-activating receptor KIR2DS4. J Immunol 2004; 173: 1819−1825. | PubMed | ChemPort |
  16. Uhrberg M, Valiante NM & Shum BP. Human diversity in killer cell inhibitory receptor genes. Immunity 1997; 7: 753−763. | Article | PubMed | ISI | ChemPort |
  17. Witt CS, Dewing C, Sayer DC, Uhrberg M, Parham P & Christiansen FT. Population frequencies and putative haplotypes of the killer cell immunoglobuline-like receptor sequences and evidence for recombination. Transplantation 1999; 68: 1784−1789. | Article | PubMed | ChemPort |
  18. Shilling HG, Guethlein LA, Cheng NW, Gardiner CM, Rodriguez R & Parham P. Allelic polymorphism synergizes with variable gene content to individualize human KIR genotype. J Immunol 2002; 168: 2307−2315. | PubMed | ISI | ChemPort |
  19. Ljunggren HG & Karre K. In search of the missing self: MHC molecules and NK cell recognition. Immunol Today 1990; 11: 237−244. | Article | PubMed | ISI | ChemPort |
  20. Algarra I, Cabrera T & Garrido F. The HLA crossroad in tumor immunology. Hum Immunol 2000; 61: 65−73. | Article | PubMed | ISI | ChemPort |
  21. Elkins WL, Pickard A & Pierson GR. Deficient expression of class-I HLA in some cases of acute leukemia. Cancer Immunol Immunother 1984; 18: 91−100. | Article | PubMed | ChemPort |
  22. Brouwer RE, van der Heiden P, Schreuder GM, Mulder A, Datema G & Anholts JD et al.. Loss or downregulation of HLA class I expression at the allelic level in acute leukemia is infrequent but functionally relevant, and can be restored by interferon. Hum Immunol 2002; 63: 200−210. | Article | PubMed | ChemPort |
  23. Demanet C, Mulder A, Deneys V, Worsham MJ, Claas FH & Ferrone S. Down-regulation of HLA-A and HLA-Bw6, but not HLA-Bw4, allospecificities in leukemic cells: an escape mechanism from CTL and NK attack? Blood 2004; 103: 3122−3130. | Article | PubMed | ChemPort |
  24. Ruggeri L, Capanni M, Martelli MF & Velardi A. Cellular therapy: exploiting NK cell alloreactivity in transplantation. Curr Opin Immunol 2001; 8: 355−359. | ChemPort |
  25. Ruggeri L, Capanni M, Urbani E, Perruccio K, Shlomchik WD & Tosti A et al.. Effectiveness of donor natural killer cell alloreactivity in mismatched hematopoietic transplants. Science 2002; 295: 2097−2101. | Article | PubMed | ISI | ChemPort |
  26. Bishara A, Amar A, Brautbar C, Condiotti R, Lazarovitz V & Nagler A. The putative role of HLA-C recognition in graft versus host disease (GVHD) and graft rejection after unrelated bone marrow transplantation (BMT). Exp Hematol 1995; 23: 1667−1675. | PubMed | ChemPort |
  27. Giebel S, Locatelli F, Lamparelli T, Velardi A, Davies S & Frumento G et al.. Survival advantage with KIR ligand incompatibility in hematopoietic stem cell transplantation from unrelated donors. Blood 2003; 102: 814−819. | Article | PubMed | ChemPort |
  28. Verheyden S, Bernier M & Demanet C. Identification of natural killer cell receptor phenotypes associated with leukemia. Leukemia 2004; 18: 2002−2007. | Article | PubMed | ChemPort |
  29. Maxwell LD, Williams F, Gilmore P, Meenagh A & Middleton D. Investigation of killer cell immunoglobulin-like receptor gene diversity: II. KIR2DS4. Hum Immunol 2004; 65: 613−621. | Article | PubMed | ChemPort |
  30. Martin MP, Gao X, Lee JH, Nelson GW, Detels R & Goedert JJ et al.. Epistatic interaction between KIR3DS1 and HLA-B delays the progression to AIDS. Nat Genet 2002; 31: 429−434. | Article | PubMed | ISI | ChemPort |
  31. Nelson GW, Martin MP, Gladman D, Wade J, Trowsdale J & Carrington M. Cutting edge: heterozygote advantage in autoimmune disease: hierarchy of protection/susceptibility conferred by HLA and killer Ig-like receptor combinations in psoriatic arthritis. J Immunol 2004; 173: 4273−4276. | PubMed | ChemPort |
  32. Cook MA, Milligan DW, Fegan CD, Darbyshire PJ, Mahendra P & Craddock CF et al.. The impact of donor KIR and patient HLA-C genotypes on outcome following HLA-identical sibling hematopoietic stem cell transplantation for myeloid leukemia. Blood 2004; 103: 1521−1526. | Article | PubMed | ChemPort |
  33. Davies SM, Ruggieri L, DeFor T, Wagner JE, Weisdorf DJ & Miller JS et al.. Evaluation of KIR ligand incompatibility in mismatched unrelated donor hematopoietic transplants. Blood 2002; 100: 3825−3827. | Article | PubMed | ISI | ChemPort |
  34. Bishara A, De Santis D, Witt CC, Brautbar C, Christiansen FT & Or R et al.. The beneficial role of inhibitory KIR genes of HLA class I NK epitopes in haploidentically mismatched stem cell allografts may be masked by residual donor-alloreactive T cells causing GVHD. Tissue Antigens 2004; 63: 204−211. | Article | PubMed | ChemPort |
  35. Bornhauser M, Schwerdtfeger R, Martin H, Frank KH, Theuser C & Ehninger G. Role of KIR ligand incompatibility in hematopoietic stem cell transplantation using unrelated donors. Blood 2004; 103: 2860−2862. | Article | PubMed |
  36. Gardiner CM, Guethlein LA, Shilling HG, Pando M, Carr WH & Rajalingam R et al.. Different NK cell surface phenotypes defined by the DX9 antibody are due to KIR3DL1 gene polymorphism. J Immunol 2001; 166: 2992−3001. | PubMed | ISI | ChemPort |


We thank A Willekens and her nursing staff from the HSC Transplant Intensive Care Unit for the excellent care of the patients and clinical data collection. We also thank B Guns for editing the manuscript. We are grateful to Dr Y Van Riet and Dr S Sumbwanyambe for a critical review of the manuscript. This work was supported by a grant from the Scientific Fund W Gepts AZ-VUB.



These links to content published by NPG are automatically generated


Immunology The footprint of a killer

Nature News and Views (01 Jun 2000)

Antiviral synergism

Nature Genetics News and Views (01 Jun 2005)

'Unleashed' natural killers hinder HIV

Nature Genetics News and Views (01 Jun 2007)