CD3+/CD19+-depleted grafts in HLA-matched allogeneic peripheral blood stem cell transplantation lead to early NK cell cytolytic responses and reduced inhibitory activity of NKG2A | Leukemia CD3+/CD19+-depleted grafts in HLA-matched allogeneic peripheral blood stem cell transplantation lead to early NK cell cytolytic responses and reduced inhibitory activity of NKG2A AbstractNatural killer (NK) cells have an important function in the anti-tumor response early after stem cell transplantation (SCT). As part of a prospective randomized phase III study, directly comparing the use of CD3+/CD19+-depleted peripheral blood stem cell (PBSC) harvests with CD34+-selected PBSC harvests in allogeneic human leukocyte antigen-matched SCT, we here show that the use of CD3+/CD19+-depleted PBSC grafts leads to early NK cell repopulation and reconstitution of the CD56dim and CD56bright NK cell subsets, with concomitant high cytolytic capacity. In the CD34 group, this process took significantly longer. Moreover, in the CD3/19 group after reconstitution, a higher percentage of killer immunoglobulin-like receptor-positive NK cells was found. Although similar percentages of CD94-positive NK cells were found in both groups, in the CD34 group, almost all expressed the inhibitory CD94:NKG2A complex, whereas in the CD3/19 group, the inhibitory CD94:NKG2A and the activating CD94:NKG2C complex were equally distributed. This preferential development of NKG2C-expressing NK cells in the CD3/19 group was paralleled by a loss of NKG2A-mediated inhibition of NK cell degranulation. These results show that the use of CD3+/CD19+-depleted grafts facilitates strong NK cell cytolytic responses directly after SCT, and the rapid emergence of an NK cell receptor phenotype that is more prone to activation. IntroductionNatural killer (NK) cells are potent immune effector cells involved in the clearance of virus-infected and tumor cells. They are able to lyse infected or abnormal cells without prior sensitization or preactivation by antigen-presenting cells. NK cells survey potential target cells through a range of inhibitory and stimulatory receptors that detect the loss (or reduction) of expression of human leukocyte antigen (HLA) class I molecules or other nonclassical HLA class I-specific signals.1, 2 The balance between the inhibitory and stimulatory signals triggers and modulates the NK cell effector function and the response of NK cells toward other cells.The cytolytic function of NK cells is regulated by killer cell immunoglobulin-like receptors (KIR) including both inhibitory receptors (KIR-DL) and stimulatory receptors (KIR-DS). Whereas the ligands for the activating receptors is not yet unequivocally established, the ligands of KIR-DL receptors are specific for HLA class I molecules. KIR receptors are clonally distributed among NK cells within each individual.3 Next to the KIR, another important receptor in modulating NK cell responses is the CD94:NKG2 heterodimeric complex. This receptor is part of the C-type lectin family and its ligand is the HLA-E class I molecule.4 The CD94:NKG2 heterodimer can either have an inhibitory (NKG2A) or a stimulatory (NKG2C) function. Other NK cell receptors include the natural cytotoxicity receptors (NKp30, NKp44, and NKp46) and NKG2D (a C-type lectin homodimer). These receptors are capable of activating NK cells, whereby inducing NK cell-mediated cytolytic responses.5As NK cells are efficient effectors in eradicating tumor cells, they have an important function in the anti-tumor response after allogeneic stem cell transplantation (SCT).6, 7, 8, 9 In HLA-haploidentical SCT, Ruggeri et al.10 showed that donor alloreactive NK cells isolated from peripheral blood of the recipient were able to lyse tumor cells derived from the recipient, implying that within 1 month after SCT, NK cells may be able to provide some degree of immune reactivity by targeting residual tumor cells still present in the recipient. Indeed, fast recovery of NK cells and predicted alloreactivity toward host tumor cells has been associated with decreased relapse rates and better overall survival of the patient.11, 12, 13 Recently, however, Nguyen et al.14 showed that early reconstituting NK cells have an immature phenotype and that NK cell alloreactivity is impaired at an early stage after haploidentical SCT. The results of this study emphasize that proper reconstitution of the NK cell receptor repertoire is probably more important for the induction of alloreactive NK cell responses than just fast NK cell recovery. One way to overcome delayed immune reconstitution of NK cells after SCT is to change the selection procedure of the graft. Recent data have shown that graft selection for haploidentical SCT by depletion of CD3+ and CD19+ cells, instead of positive CD34+ cell selection, leads to faster engraftment and better overall immune reconstitution.15In our centre, a phase III clinical study started in January 2006, directly comparing the use of CD3+/CD19+-depleted peripheral blood stem cell (PBSC) harvests with commonly applied CD34+-selected PBSC harvests in allogeneic HLA-matched SCT. We investigated the immunological recovery and function of the NK cells after allogeneic SCT for a follow-up period of 1 year and compared the results between patients either having received a CD3+/CD19+-depleted graft or a CD34+-selected graft.Materials and methodsStudy populationIn a single centre randomized phase III clinical study at the Radboud University Nijmegen Medical Centre, patients with AML in first complete remission, ALL in first complete remission or high risk MDS (RAEB, RAEB-t) in complete remission, eligible for allogeneic peripheral blood stem cell transplantation (PSCT) received G-CSF-mobilized PBSCs of an HLA-identical sibling donor (SIB) or an HLA-A, -B, -C, -DR, -DQ-matched voluntary unrelated donor. In this study, patients were randomized to receive a PBSC graft processed either by positive CD34+ immunomagnetic selection or T (CD3+) and B-cell (CD19+) depletion by negative immunomagnetic selection using a CliniMACS cell selector device (Miltenyi Biotec, Bergisch Gladbach, Germany). All transplants contained at least 3 脳 106 CD34+ cells and 0.5 脳 106 T cells/kg body weight. The amount of T cells was achieved by adding back cells from the harvest. Between January 2006 and November 2007, 23 patients were included in this study. This study was approved by the medical ethics committee of the Radboud University Nijmegen Medical Centre and written informed consent was provided by patients and donors according to the Declaration of Helsinki. Table 1 outlines the demographics of the patients selected for this study. This study is registered at http://clinicaltrials.gov as T-cell and B-cell depletion in allogeneic PSCT (registration number: NCT00306332).Table 1 Patient characteristicsFull size tablePatient treatment proceduresAll patients transplanted with a graft from a sibling donor were conditioned with idarubicine (total dose 42鈥塵g/m2) i.v.,16 cyclophosphamide (120鈥塵g/kg body weight) i.v., and fractionated total body irradiation (TBI) given in two equal fractions on days 鈭? and 鈭? to a total dose of 9鈥塆y. In two patients, TBI was replaced by busulphan (3.2鈥塵g/kg body weight) intravenously. In patients transplanted with a graft from a voluntary unrelated donor, the conditioning regimen consisted of cyclophosphamide (120鈥塵g/kg body weight), anti-thymocyte globulin (2鈥塵g/kg body weight) every 24鈥塰 at four consecutive days (day 鈭? to 鈭?) i.v. and fractionated TBI given in two equal fractions on days 鈭? and 鈭? to a total dose of 9鈥塆y. The stem cells were infused 24鈥塰 after completion of TBI or 72鈥塰 after the last dose of busulphan. All patients were treated with cyclosporine A 3鈥塵g/kg/d i.v. from day 鈭? to +14, followed by 2鈥塵g/kg/d i.v. infusion until oral administration of cyclosporine A (6鈥塵g/kg/d) was possible. Cyclosporine A was gradually tapered off and discontinued 12 weeks post-grafting. Patients who did not develop acute graft versus host disease (GVHD) grade 2 and/or chronic GVHD, according to the criteria described by Glucksberg and Shulman, received pre-emptive donor lymphocyte infusion at a dose of 0.1 脳 108 CD3+ cells/kg body weight. All patients received oral selective gut decontamination (ciprofloxacin), as well as co-trimoxazole for pneumocystis carinii prophylaxis and oral acyclovir for prophylaxis of herpes infections.Collection and preparation of samplesBlood samples (50鈥塵l) were taken from patients before and at 0.5, 1, 2, 3, 6, 9, and 12 months after SCT. When possible, blood samples were collected from each donor before SCT. Peripheral blood mononuclear cells (PBMC) were isolated by density gradient centrifugation using Lymphoprep (Nycomed Pharma, Roskilde, Denmark). A fraction of the isolated PBMC was used to analyze the phenotype of the NK cells in detail. The remaining cells were cryopreserved in liquid nitrogen until further use. To evaluate immunological recovery, immunophenotyping was performed on fresh blood samples from patients.Phenotypic analysisThe phenotype of NK cells was analyzed by four-color flow cytometry measured on a Coulter Epics XL (Beckman Coulter, Miami, FL, USA). For cell surface staining, the after conjugated mAbs were used: CD45-PE, CD16-FITC (Dako, Glostrup, Denmark), CD3-ECD, CD56-PCy5, CD158a,h-PE, CD158b1b2,j-PE, NKG2A-PE, NKp30-PE, NKp44-PE, NK46-PE (Beckman Coulter), NKB1-PE (BD Biosciences, Erembodegem, Belgium), CD94-PE (Immunotech, Marseille, France), NKG2C-PE, and NKG2D-PE (R D Systems, Minneapolis, CA, USA). Isotype controls were used for marker settings. Analysis was performed using CXP analysis software (Beckman Coulter). All flow cytometric analyses were performed with a combination of CD3, CD16, and CD56鈥塵Abs combined to a fourth specificity to gate specifically on CD3鈭?/sup>CD56dimCD16bright and CD3鈭?/sup>CD56brightCD16low/neg NK cell populations and to distinguish them from NK-T and T cells. A dual platform technique was used to evaluate the immunological recovery of NK cells. With this technique, the absolute numbers of NK cells were calculated based on the flow cytometrically assessed percentage of CD45+CD56+CD16+/鈭?/sup>CD3鈭?/sup> cells and the white blood cell count from a hematology cell analyzer (ADVIA120 Hematology System, Bayer Diagnostics, Leverkusen, Germany).Functional analysisFor functional studies, cryopreserved PBMC were thawed and cultured overnight in the presence or absence of rhIL-2 (100鈥塙/ml; Chiron, Amsterdam, the Netherlands) and rhIL-15 (10鈥塶g/ml; BioSource International, Camarillo, CA, USA) in culture medium consisting of RPMI 1640 supplemented with pyruvate (0.02鈥塵M), penicillin (100鈥塙/ml), streptomycin (100鈥壩糶/ml), and 10% human pooled serum, in a 37鈥壜癈 95% humidity, 5% CO2 incubator.51Chromium-release assayPre-stimulated or unstimulated PBMC were used to examine the NK cell cytolytic activity using the major histocompatibility complex class I-deficient human erythroleukemia K562 cell line (ATCC; CCL-243) as target cells at PBMC:target ratios ranging from 80:1 to 0.625:1 in a standard 4-h 51Cr-release assay. All experiments were performed in triplicates and for each patient all samples were run together in one assay to avoid inter-assay variation. The percentage specific lysis was calculated as follows: % specific lysis=((experimental release鈥搒pontaneous release)/(maximum release鈥搒pontaneous release)) 脳 100%. To express NK cell cytolytic activity independent of an arbitrary selection of E:T ratio and to allow reliable comparisons between samples of patients tested at different moments in time, the specific lysis was converted into lytic units (LU). LU are defined as the number of effector cells contained in 106 effector cells required to lyse 20% of the target cells.17Redirection assaysRedirection assays were performed using the P815 cell line (ATCC; TIB-64) as target cells. P815 cells were incubated in culture medium with pre-stimulated PBMC at an E:T ratio of 1:2 in the presence of mAbs (10鈥壩糶/ml) specific for NKG2C, NKG2A, and NKp46 (R D Systems) or with IgG1, IgG2a, and IgG2b isotype controls (R D Systems) at 37鈥壜癈 95% humidity and 5% CO2. E:T ratios were based on the amount of live NK cells present in the PBMC fractions after overnight culture. To measure NK cell degranulation, cells were washed after 2鈥塰 of incubation, stained on ice with fluorochrome-conjugated anti-CD107a (BD Biosciences), anti-CD56, and anti-CD3鈥塵Abs and analyzed by flow cytometry. To calculate the specific degranulation, nonspecific degranulation induced by isotype controls was deducted from the degranulation given by specific triggering. IFN-纬 produced by NK cells was analyzed by intracellular staining after 5鈥塰 of incubation with P815 target cells at 37鈥壜癈 with 5鈥塶g/渭l of Brefeldin A (Sigma-Aldrich, Zwijndrecht, The Netherlands) for the last 4鈥塰. First, a surface staining was performed using anti-CD56 and anti-CD3鈥塵Abs. For intracellular staining, Fix and Fix/Perm buffer (eBioscience, San Diego, CA, USA) were used according to the manufacturer\'s instructions in combination with IFN-纬-PeCy7鈥塵Ab (eBioscience).Statistical analysisNonparametric tests were used to compare continuous variables between two groups. To test the differences between both patient groups or within one group between donors and recipients after SCT, a Mann鈥揥hitney U-test was performed. A Wilcoxon-signed rank test was used to compare the differences within each group between recipients before and after SCT. P 0.05 was considered significant.ResultsPatientsPatients eligible for inclusion in the study were randomized on a 1:1 basis for treatment with either a positive CD34+-selected graft (CD34 group) or a T- and B-cell-depleted graft (CD3+/CD19+ depletion; CD3/19 group). Between January 2006 and November 2007, 14 patients were included in the CD34 group and 9 patients were included in the CD3/19 group. Patient and donor characteristics are depicted in Table 1. The composition of the grafts selected by CD34+ selection and CD3+/CD19+ depletion are summarized in Table 2. After SCT, patients of both groups showed similar engraftment kinetics and there were no remarkable differences in the immunosuppressive regimens between the two groups (Table 3). There were no major differences in the occurrence of GVHD, fungal infections, and relapse rate. The incidence of bacterial infections was slightly higher in the CD34 group, possibly because of the fact that the CD34+-selected grafts contained less monocytes and granulocytes. In case of viral infections, we saw a tendency toward a higher infection rate in the CD3/19 group. This is due to a higher number of cytomegalovirus (CMV) reactivations that occurred within the CD3/19 group (4 patients versus 1 patient in the CD34 group).Table 2 Graft compositionFull size tableTable 3 Clinical follow-upFull size tableRepopulation of NK cells after SCTTo study NK cell repopulation and reconstitution of the NK cell receptor repertoire, peripheral blood samples were taken from each patient before SCT and at 0.5, 1, 2, 3, 6, and 12 months after SCT and were analyzed for the presence of NK cells by flow cytometric analysis. At each time point, the absolute numbers of NK cells were analyzed within whole blood samples (Figure 1a). In the CD34 group, at day 14 after SCT, only limited amounts of NK cells were present in the peripheral blood as compared with absolute NK cell numbers before SCT. The absolute numbers of NK cells reached pre-SCT levels 1 month after SCT. Thereafter, levels slowly increased and reached significant levels at 6 months after SCT (P 0.05, compared with numbers before SCT). In the CD3/19 group, absolute numbers of NK cells in the peripheral blood reached pre-SCT level already at 14 days after SCT. One month after SCT, the absolute NK cell numbers increased even further (P 0.05). Comparison of the CD34 group with the CD3/19 group shows that at day 14 after SCT significantly higher numbers of NK cells were present in the CD3/19 group (P 0.05).Figure 1Recovery of NK cells after SCT. (a) The recovery of NK cells (CD45+CD56+CD3鈭?/sup>) after SCT within the CD34 group and the CD3/19 group. NK cell numbers were analyzed within whole blood samples of recipients before SCT (R) and at several time points after SCT using flow cytometry. Within each group, differences between recipients before SCT and after SCT were analyzed using the Wilcoxon-signed rank test. The Mann鈥揥hitney U-test was used to test differences between both groups; *P 0.05, ***P 0.001. (b) Repopulation of NK cells and the different NK cell subsets, analyzed in freshly isolated PBMC fractions by flow cytometry, using CD3, CD45, CD16, and CD56鈥塵Abs. Displayed are the ratios of the CD56brightCD56dim NK cell subsets before SCT and at different time points after SCT. R, recipients before SCT; D, donors. Within each group, differences between donors and recipients after SCT were analyzed using the Mann鈥揥hitney U-test; **P 0.01, ***P 0.001.Full size imageThe repopulation of the CD56bright and CD56dim NK cell subsets was analyzed within isolated PBMC fractions by gating on the lymphocytes by CD45+ and a low side scatter and subsequently on CD56+ and CD3鈭?/sup> cells. In the CD34 group, the repopulation of the two NK cell subsets seemed to start with the development of the CD56bright NK cells. The balance between the CD56bright and CD56dim NK cell subsets was in favor of the CD56bright NK cells until 6 months after SCT (Figure 1b). This is comparable with the results described by earlier studies in which the development of NK cells after SCT was studied.18, 19 In contrast, in the CD3/19 group, both NK cell subsets were already present at 14 days after SCT (data not shown). Moreover, the outgrowth of the CD56bright NK cell subset was less profound as seen within the CD34 group and the balance between the CD56bright and CD56dim NK cell subsets was more comparable with the donor situation before SCT. In contrast to the different reconstitution kinetics of the NK cells, the reconstitution of both T and B cells was comparable between the CD34 and CD3/19 group during the first 3 months after SCT (Supplementary Figure S1).Reconstitution of the NK cell receptor repertoireAs the repopulation of the CD56dim and CD56bright NK cell subsets was much faster in the CD3/19 group, suggesting an earlier presence of mature NK cells, we investigated whether this phenomenon was also reflected in the reconstitution of the NK cell receptor repertoire after SCT.Comparison of KIR expression between the CD34 and the CD3/19 groupsIn both groups, reconstitution of KIR2DL/S1, KIR2DL/S2/3, and KIR3DL1 expression was monitored in the total NK cell population (CD45+CD3鈭?/sup>CD56+) during the follow-up at different time points after SCT.Within the CD34 group, the percentages of KIR2DL/S1+ NK cells never reached donor levels throughout the entire follow-up. At 1 year after follow-up, on average only 10% of the NK cells expressed KIR2DL/S1 compared with 30% of the NK cells in the donor. Moreover, the percentage of KIR2DL/S2/3-expressing NK cells was strongly reduced at 3 and 6 months, but did reach donor levels at 1 year post-SCT (Figure 2). The reconstitution of KIR3DL1+ NK cells already reached donor level within 1 month after SCT. Notably, during the remaining follow-up, the percentage of KIR3DL1+ NK cells was decreased and did not regain donor levels during the rest of the follow-up. Compared with KIR2DL/S, the reconstitution of KIR3DL1 seemed to show an opposite reconstitution pattern over time. This phenomenon has earlier been described after HLA-mismatched/haploidentical SCT.20Figure 2Reconstitution of KIR expression by NK cells. The reconstitution of KIR2DL/S1 (CD158a,h), KIR2DL/S2/3 (CD158b1b2,j), and KIR3DL1 (NKB1) expression by CD45+CD56+CD3鈭?/sup> NK cells was analyzed by flow cytometry in freshly isolated PBMC fractions, at different time points after SCT. Displayed are the percentage KIR-positive cells in recipients before SCT (R), in donors (D) and in recipients at 1, 3, 6, and 12 months after SCT. Data are shown for the CD34 group and the CD3/19 group. Within each group, differences between donors and recipients after SCT were analyzed using the Mann鈥揥hitney U-test; *P 0.05, **P 0.01, ***P 0.0001.Full size imageIn the CD3/19 group, KIR reconstitution was faster compared with the CD34 group. Although in the CD34 group full reconstitution took a year, as has also been described by others,21 the NK cells in the CD3/19 group reached normal expression frequencies directly after SCT. Here, the amount of KIR2DL/S1+ NK cells was only significantly decreased up to 1 month after SCT compared with donor level and within 1 year after SCT, expression levels were fully restored. In addition, the amount of KIR2DL/S2/3+ and KIR3DL1+ NK cells were already at donor level at 14 days after SCT (data not shown). Remarkably, during the remaining time of the follow-up, the number of KIR2DL/S2/3+ NK cells was enhanced from a median level of 19% at 2 months after SCT to 30% at 1 year after SCT. Similar to the CD34 group, reconstitution of the KIR3DL1+ NK cells followed the same pattern over time and did not significantly decrease compared with donor level. Thus, as the reconstitution of the number of KIR2DL/S+ NK cells was rather slow in the CD34 group and was significantly lower within the first months after SCT, frequencies of KIR2DL/S expression in the CD3/19 group were not significantly different from donor level and reconstituted faster to full donor level. In summary, the NK cells quickly recovered in the CD3/CD19 group and maintained similar percentages of KIR-bearing NK cells as compared with the donor before SCT.Alternative reconstitution of CD94:NKG2A/C expressionDuring NK cell development, KIR expression is preceded by expression of CD94:NKG2A.22 In the first months after SCT, NK cell subsets were shown to be characterized by a high frequency of CD94:NKG2A expression and low expression of KIR.19, 20, 21 This suggests that at early stages after SCT, the function of NK cells is primarily regulated through the interaction of CD94:NKG2 heterodimeric complexes with HLA-E molecules. In both groups, the frequency of CD94-expressing NK cells was upregulated during the first 3 months after SCT, after which the percentage of CD94-expressing NK cells slowly decreased toward donor level (Figure 3). In the CD34 group, the amount of CD94+ NK cells was higher during the first months after SCT. However, frequencies of CD94-expressing NK cells between the two groups were similar from 3 months to 1 year after SCT. Remarkably, the NKG2A expression levels in the CD3/19 group were significantly decreased from a median percentage of 76% at 1 month to 36% at 1 year after SCT (P 0.05) whereas at the same time, the frequency of NKG2C-expressing NK cells was significantly increased from a median of 8鈥?3%, respectively (P 0.01) (Figure 4a). This phenomenon seemed to be specific for the CD3/19 group, as this was not seen within the CD34 group (Figure 4b).Figure 3Reconstitution of CD94 expression by NK cells. CD94 expression by the CD45+CD56+CD3鈭?/sup> NK cell population in freshly isolated PBMC fractions of donors (D) and of recipients before (R) and at 1, 3, 6, and 12 months after SCT, was analyzed using flow cytometry. Data are shown for the CD34 group and the CD3/19 group. Within each group, differences between donors and recipients after SCT were analyzed using the Mann鈥揥hitney U-test; **P 0.01, ***P 0.0001.Full size imageFigure 4Reconstitution of NKG2A and NKG2C expression by NK cells after SCT. Expression of the inhibitory NKG2A and the activating NKG2C receptors by CD45+CD56+CD3鈭?/sup> NK cells was analyzed by flow cytometry in freshly isolated PBMC fractions from recipients of both CD3/19 and CD34 groups. (a) Percentages of NKG2A- and NKG2C-expressing NK cells in recipients before SCT (R), in donors (D) and in recipients at 1, 3, 6, and 12 months after SCT in both groups. Differences between groups were analyzed using the Mann鈥揥hitney U-test; *P 0.05, **P 0.01. (b) A representative example of NKG2A and NKG2C expression on NK cells in individual recipients of either group over time. Histogram analysis shows the percentage of receptor expressing NK cells at different time points after SCT.Full size imageWe observed no differences in the reconstitution of the other activating NK cell receptors (Figure 5). Although the frequencies of NKp30 expression seemed to drop slightly in the CD3/19 group as compared with the CD34 group, percentages of NKG2D and natural cytotoxicity receptor expression were similar in both groups after SCT. In addition, frequencies of NKG2D-, NKp30-, and NKp44-expressing NK cells did not significantly change compared with donor levels. In both groups, the amount of NKp46 expression was significantly higher directly after SCT and remained above donor levels throughout the rest of the follow-up.Figure 5Reconstitution of activating NK cell receptors NKG2D, NKp30, NKp44, and NKp46. Percentages of NKG2D-, NKp30-, NKp44-, and NKp46-expressing CD45+CD56+CD3鈭?/sup> NK cells were analyzed in freshly isolated PBMC fractions for both the CD34 group and the CD3/19 group at 1, 3, 6, and 12 months after SCT. Within each group, differences between donors and recipients after SCT were analyzed using the Mann鈥揥hitney U-test; **P 0.01, ***P 0.0001.Full size imageCytolytic activity against HLA class I-negative target cellsWe observed that, whereas in the CD34 group only few NK cells were present at day 14 after SCT and full repopulation in the CD34 group took place from 1 to 2 months after SCT, in the CD3/19 group the repopulation of NK cells had already taken place just after SCT. Moreover, we observed that the ratio of CD56bright versus CD56dim NK cells in the CD3/19 group after SCT was distinct from that found in the CD34 group. Thus, we set out to investigate whether the repopulation profiles and phenotypic differences were paralleled by a distinct functional activity of the cell populations. To assess the cytolytic potential of the repopulated NK cells after SCT, we performed cytolytic assays using the K562 cell line as a target. Before the assay, NK cells were incubated overnight in the absence or presence of IL-2 and IL-15. Unstimulated NK cells derived from the CD34 and CD3/19 group before SCT showed equal cytolytic activity (Figure 6a). But, after SCT, the NK cells in the CD3/19 group not only seemed to show a faster reconstitution of the different NK cell subsets and the NK cell receptor repertoire, at 14 days after SCT, they also showed stronger cytolytic capacity. Moreover, at this time point, the cytolytic activity was even higher than before SCT (P 0.05) and this was maintained throughout the rest of the follow-up. Parallel to the duration of NK cell repopulation, in the CD34 group, enhanced cytolytic activity levels were only reached at 1 to 2 months after SCT. Although enhanced, the trend in cytolytic activity was similar for the pre-stimulated NK cells (Figure 6b). This implies that the use of CD3+/19+-depleted grafts for SCT may lead to an earlier and longer time frame of NK cell cytolytic activity after SCT.Figure 6NK cell cytolytic activity against K562 target cells. Thawed PBMC fractions were cultured overnight in the absence (a) or presence (b) of 100鈥塙/ml IL-2 and 10鈥塶g/ml IL15 and subsequently incubated with 51Cr-labeled K562 target cells in a standard 4-h 51Cr-release assay. Displayed are the lytic units (LU) within each PBMC fraction of the CD34 group and the CD3/19 group before SCT (R) and at 0.5, 1, 2, 3, 6, and 12 months after SCT. LU comparisons between recipients before SCT and after SCT were analyzed using the Wilcoxon-signed rank test; *P 0.05, **P 0.01. Differences between groups were analyzed using the Mann鈥揥hitney U-test; *P 0.05.Full size imageDiminished NK cell inhibition through NKG2A triggering in the CD3/19 groupEarlier studies showed that high expression levels of CD94:NKG2A after SCT may lead to impaired cytolytic NK cell function.14 As the NKG2A:NKG2C ratio in the CD3/19 group was in favor of the stimulatory NKG2C receptor, we set out to investigate whether this distinct phenotype was associated with enhanced functional activity. To this end, we set up a redirection assay in which NK cells were triggered by P815 cells labeled with NKG2C- and NKp46-specific antibodies in the absence or presence of NKG2A-specific antibodies. Typically, triggering of NK cells with NKp46 and NKG2C antibodies leads to NK cell degranulation, whereas NKG2A triggering leads to inhibition of NK cell degranulation. At 1 month after SCT, the CD3/19 and CD34 groups showed similar NKG2A-mediated inhibition of NK cell degranulation (Figure 7), which is in line with the similar percentage of NKG2A-expressing NK cells in both groups (Figure 5). Notably, from 1 month post-SCT onward, only in the CD3/19 group, the inhibitory effect of NKG2A triggering was diminished, corresponding with an increased frequency of CD94:NKG2C+ NK cells and a decreased percentage of CD94:NKG2A+ NK cells. In the CD34 group, the inhibitory effect of NKG2A triggering was maintained. In addition, the percentage of degranulating NK cells as induced by NKG2C and NKp46 triggering tended to be higher in the CD3/19 group as compared with the CD34 group (3 months: median of 16% versus 3%; at 6 months: median of 9% versus 3%). The level of degranulation, induced by NKG2C and NKp46 triggering, was also reflected in the IFN-纬 production (data not shown), as the NK cells within the CD3/19 group produced higher levels of IFN-纬.Figure 7Redirected degranulation of NK cells by specific triggering of the activating NKG2C and NKp46 receptors in the presence or absence of NKG2A specific antibodies. PBMC fractions of the CD34 group (n=6) and the CD3/19 group (n=5) were stimulated overnight with IL-2 (100鈥塙/ml) and IL-15 (10鈥塶g/ml) and subsequently incubated for 2鈥塰 at an NK:target cell ratio of 1:2 in a redirection assay using P815 target cells and NKG2A, NKG2C, NKp46, or isotype control mAbs. Specific degranulation was determined within the CD56+CD3鈭?/sup> NK cell population as the percentage of CD107+CD56+CD3鈭?/sup> NK cells with specific mAbs minus the percentage of CD107+CD56+CD3鈭?/sup> NK cells in the presence of isotype control mAbs. Negative values are displayed as equal to 0. Specific degranulation was determined for recipients at 1, 3, and 6 months after SCT. Significance of inhibition through NKG2A was analyzed using the Wilcoxon-signed rank test; NS=not significant, *P 0.05.Full size imageIn conclusion, we show that the use of CD3+/19+-depleted grafts for allogeneic HLA-matched SCT leads to a preferential development of NKG2C-expressing NK cells, paralleled by a loss of NKG2A-mediated inhibition of NK cell degranulation at later stages after SCT.DiscussionImmediately after SCT, alloreactive NK cells have been shown to be beneficial not only for boosting the anti-tumor response, but also for the prevention of GVHD as well as infections. In these cases, full functional activity of NK cells already in the early phase after SCT is essential, and, therefore, the presence of NK cells in the graft seems to be beneficial for transplant outcome.15, 23 In this study, we show that patients that received a CD3+/CD19+-depleted graft, which contains NK cells in substantial numbers, exhibited a faster recovery of a functional NK cell receptor repertoire as compared with patients that received a conventional CD34+-selected graft, whereas both groups were treated under the same immunosuppressive regimen as part of their GVHD prophylaxis. Furthermore, transplantation with a CD3+/CD19+-depleted graft resulted in the development of a functionally different NK cell population that was more prone to activation through the NKG2C:CD94 receptor complex and less sensitive to inhibition through the NKG2A:CD94 receptor complex.Several studies have shown that the beneficial effect of NK cell activity after SCT is dependent on the fast reconstitution of a cytolytic NK cell repertoire that is regulated by KIR24 and that a slower reconstitution toward a KIR-dependent repertoire is associated with poor clinical outcome.14, 20, 21, 25 In the group receiving a CD3+/CD19+-depleted graft, we observed functional NK cells already at 14 days after SCT, with KIR-expressing NK cell numbers similar to those found in the donor before SCT. After 14 days, the cytolytic activity and number of KIR-expressing cells increased even further and remained high during follow-up. In contrast, in patients that received a CD34+-selected graft KIR reconstitution took much longer, whereby during the first 6 months after SCT, the repertoire was skewed toward NKG2A-expressing cells rather than KIR-expressing cells, a phenomenon also described by others.21So it would seem that the conditions created after SCT using a CD3+/CD19+-depleted graft support the development of a KIR-based repertoire. The clinical relevance of such a finding is exemplified by the fact that in the clinical studies that reported a function for NK cells in the anti-tumor response after SCT, the nature of the interaction between KIR and HLA class I seemed decisive for the outcome.8, 9, 25, 26, 27 The most straightforward example hereof is the haploidentical SCT setting, whereby donor NK cells expressing KIR specific for self-HLA class I ligands sense the absence of this ligand in the recipient. This results in cytolytic responses against the recipient tumor cells and reduced risk of relapse (鈥榤issing self鈥?.8, 28, 29, 30 A similar phenomenon can also be observed in HLA-matched SCT for AML, when the donor expresses a KIR for which the ligand is missing in the patient (鈥榤issing ligand鈥?.27 Although it is appealing to incorporate the receptor-ligand model into the criteria for donor selection in HLA-matched SCT, recent findings in the haploidentical setting showed that this model is more complex as not only inhibitory KIR, but also certain activating KIR have an important function in the process of tumor cell recognition and eradication.31The high NK cell numbers observed in the CD3/19 group may be the result of NK cells present in the graft. Indeed, it has been shown that alloreactive NK cells can expand and persist for at least 28 days.32 Alternatively and not mutually exclusive, the high numbers may be a result of de novo generation from CD34+ progenitor cells, as the dose of CD34+ cells was much higher in the CD3/19 group as compared with the CD34 group.Freud and Caligiuri 33 have proposed a model for NK cell development based on different maturation stages of NK cells in secondary lymphoid organs. According to this model, NK cells go through four maturation stages before they become mature peripheral NK cells (stage V). In the last stage (IV) before reaching maturity, the NK cells have high expression of CD56 and express NKG2A. In stage V, the NK cells have a lower level of CD56 expression and are KIR and/or NKG2A positive.34 This fits the observations made after SCT using a CD34+-selected graft. We and others18, 21 observed that most of the NK cells found in peripheral blood in the first 6 months after SCT resemble those in stage IV (CD56bright and NKG2A+), and that gradually over time, more NK cells reach stage V. Clinical data suggest that a sustained high number of NK cells resembling stage IV (CD56bright and NKG2A+) is associated with poor clinical outcome.14, 20Interestingly, the formation of the NK cell receptor repertoire in case of CD3+/CD19+-depleted grafts is clearly different. Not only is there a faster shift to a KIR-based repertoire, but the reduction in NKG2A-expressing cells seems to coincide with the occurrence of a population expressing the activating NKG2C receptor. At 1 year after SCT, as much as 50% of the NK cells express NKG2C. Normally, in peripheral blood only few mature NK cells express NKG2C. Moreover, only low numbers of stage IV NK cells express NKG2C.33 The occurrence of the NKG2C-expressing NK cell population may have important consequences for the anti-tumor response after SCT. The stimulatory receptor NKG2C, such as NKG2A, binds to HLA-E, although the affinity of NKG2C for HLA-E seems to be lower.35 The NK cells in the CD3/19 group were more prone to activation by NKG2C and NKp46 triggering in terms of cytolytic activity and IFN-纬 production. Furthermore, the inhibitory effect of NKG2A triggering on the whole population was lost at later stages after SCT, coinciding with the occurrence of NKG2C-expressing NK cells. In contrast, NK cells in the CD34 group maintained a high frequency of NKG2A-expressing cells and NKG2A triggering completely inhibited cytolytic NK cell activity. A similar observation was made for NK cells after CD34+ haploidentical SCT.14 Moreover, in this patient group, practically all NK cells expressed NKG2A during the first 4 months after SCT and expression of HLA-E on the tumor cells completely blocked the cytolytic activity of NKG2A+ NK cells. This may suggest that a shift in balance from a CD94:NKG2A+ to a more CD94:NKG2C+ phenotype in the CD3/19 group, and a subsequent overruling of NKG2A-mediated inhibition, may lead to a stronger anti-tumor response and a better clinical outcome.Although the two patient groups are yet too small to compare clinical outcomes, we did observe that human CMV reactivation was more common in the CD3/19 group (4 out of 9 patients) than in the CD34 group (1 out of 14 patients). In earlier studies, it was shown that CMV has an impact on the expression of NK cell receptors on NK cells and CD8+ T cells irreversibly resulting in increased CD94:NKG2C expression levels and subsequently the loss of CD94:NKG2A expression.36, 37, 38 However, when we excluded the CMV-positive patients from our analyses, the change from a overall NKG2A+ to a more NKG2C+ phenotype was also seen in patients that did not suffer from CMV reactivation (data not shown). The alternative reconstitution of the NK cell receptor repertoire, characterized by the change in balance of CD94:NKG2A+ NK cells to more CD94:NKG2C+ NK cells, and its impact on clinical outcomes after SCT is therefore a subject for further study.In summary, in this study we show that the use of selectively depleted grafts (CD3+/CD19+ cell depletion) as compared with CD34+-selected grafts in HLA-matched SCT leads to faster reconstitution of the KIR repertoire resulting in a longer time frame after SCT for NK cell cytolytic activity. Moreover, at a later stage after SCT, there was a large population of NKG2C-expressing cells and NK cells seemed to be more prone to activation. Extended studies are underway to reveal whether these findings have an effect on clinical outcome.Conflict of interestThe authors declare no conflict of interest. References1Farag SS, Caligiuri MA . Human natural killer cell development and biology. Blood Rev 2006; 20: 123鈥?37.CAS聽 Article聽Google Scholar聽 2Papamichail M, Perez SA, Gritzapis AD, Baxevanis CN . Natural killer lymphocytes: biology, development, and function. Cancer Immunol Immunother 2004; 53: 176鈥?86.Article聽Google Scholar聽 3Parham P . MHC class I molecules and KIRs in human history, health and survival. Nat Rev Immunol 2005; 5: 201鈥?14.CAS聽 Article聽Google Scholar聽 4Braud VM, Allan DS, O鈥機allaghan CA, Soderstrom K, D鈥橝ndrea A, Ogg GS et al. HLA-E binds to natural killer cell receptors CD94/NKG2A, B and C. Nature 1998; 391: 795鈥?99.CAS聽 Article聽Google Scholar聽 5Moretta A, Bottino C, Vitale M, Pende D, Cantoni C, Mingari MC et al. Activating receptors and coreceptors involved in human natural killer cell-mediated cytolysis. Annu Rev Immunol 2001; 19: 197鈥?23.CAS聽 Article聽Google Scholar聽 6Cook 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鈥?526.CAS聽 Article聽Google Scholar聽 7Giebel S, Nowak I, Wojnar J, Markiewicz M, Dziaczkowska J, Wylezol I et al. Impact of activating killer immunoglobulin-like receptor genotype on outcome of unrelated donor-hematopoietic cell transplantation. Transplant Proc 2006; 38: 287鈥?91.CAS聽 Article聽Google Scholar聽 8Ruggeri 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鈥?100.CAS聽 Article聽Google Scholar聽 9van der Meer A, Schaap NPM, Schattenberg AVMB, van Cranenbroek B, Tijssen HJ, Joosten I . KIR2DS5 is associated with leukemia free survival after HLA identical stem cell transplantation in chronic myeloid leukemia patients. Mol Immunol 2008; 45: 3631鈥?638.CAS聽 Article聽Google Scholar聽 10Ruggeri L, Capanni M, Casucci M, Volpi I, Tosti A, Perruccio K et al. Role of natural killer cell alloreactivity in HLA-mismatched hematopoietic stem cell transplantation. Blood 1999; 94: 333鈥?39.CAS聽Google Scholar聽 11Kim DH, Won DI, Lee NY, Sohn SK, Suh JS, Lee KB . Non-CD34(+) cells, especially CD8(+) cytotoxic T cells and CD56(+) natural killer cells, rather than CD34 cells, predict early engraftment and better transplantation outcomes in patients with hematologic malignancies after allogeneic peripheral stem cell transplantation. Biol Blood Marrow Transplant 2006; 12: 719鈥?28.CAS聽 Article聽Google Scholar聽 12Savani BN, Mielke S, Adams S, Uribe M, Rezvani K, Yong AS et al. Rapid natural killer cell recovery determines outcome after T-cell-depleted HLA-identical stem cell transplantation in patients with myeloid leukemias but not with acute lymphoblastic leukemia. Leukemia 2007; 10: 2145鈥?152.Article聽Google Scholar聽 13Ruggeri L, Mancusi A, Capanni M, Urbani E, Carotti A, Aloisi T et al. Donor natural killer cell allorecognition of missing self in haploidentical hematopoietic transplantation for acute myeloid leukemia: challenging its predictive value. Blood 2007; 110: 433鈥?40.CAS聽 Article聽Google Scholar聽 14Nguyen S, Dhedin N, Vernant JP, Kuentz M, Al Jijakli A, Rouas-Freiss N et al. NK cell reconstitution after haploidentical hematopoietic stem cell transplants: immaturity of NK cells and inhibitory effect of NKG2A override GvL effect. Blood 2005; 105: 4135鈥?142.CAS聽 Article聽Google Scholar聽 15Bethge WA, Haegele M, Faul C, Lang P, Schumm M, Bornhauser M et al. Haploidentical allogeneic hematopoietic cell transplantation in adults with reduced-intensity conditioning and CD3/CD19 depletion: fast engraftment and low toxicity. Exp Hematol 2006; 34: 1746鈥?752.CAS聽 Article聽Google Scholar聽 16Schaap N, Schattenberg A, Bar B, Preijers F, Geurts van KA, van der MR et al. Outcome of transplantation for standard-risk leukaemia with grafts depleted of lymphocytes after conditioning with an intensified regimen. Br J Haematol 1997; 98: 750鈥?59.CAS聽 Article聽Google Scholar聽 17Bryant J, Day R, Whiteside TL, Herberman RB . Calculation of lytic units for the expression of cell-mediated cytotoxicity. J Immunol Methods 1992; 146: 91鈥?03.CAS聽 Article聽Google Scholar聽 18Dulphy N, Haas P, Busson M, Belhadj S, Peffault de Latour R, Robin M et al. An unusual CD56brightCD16low NK cell subset dominates the early posttransplant period following HLA-matched hematopoietic stem cell transplantation. J Immunol 2008; 181: 2227鈥?237.CAS聽 Article聽Google Scholar聽 19Vago L, Forno B, Sormani MP, Crocchiolo R, Zino E, Di Terlizzi S et al. Temporal, quantitative and functional characteristics of single-KIR positive alloreactive NK cell recovery account for impaired graft versus leukemia activity after haploidentical HSCT. Blood 2008; 112: 3488鈥?499.CAS聽 Article聽Google Scholar聽 20Zhao XY, Huang XJ, Liu KY, Xu LP, Liu DH . Reconstitution of natural killer cell receptor repertoires after unmanipulated HLA-mismatched/haploidentical blood and marrow transplantation: analyses of CD94:NKG2A and killer immunoglobulin-like receptor expression and their associations with clinical outcome. Biol Blood Marrow Transplant 2007; 13: 734鈥?44.CAS聽 Article聽Google Scholar聽 21Shilling HG, McQueen KL, Cheng NW, Shizuru JA, Negrin RS, Parham P . Reconstitution of NK cell receptor repertoire following HLA-matched hematopoietic cell transplantation. Blood 2003; 101: 3730鈥?740.CAS聽 Article聽Google Scholar聽 22Vitale C, Chiossone L, Morreale G, Lanino E, Cottalasso F, Moretti S et al. Analysis of the activating receptors and cytolytic function of human natural killer cells undergoing in vivo differentiation after allogeneic bone marrow transplantation. Eur J Immunol 2004; 34: 455鈥?60.CAS聽 Article聽Google Scholar聽 23Kim DH, Sohn SK, Lee NY, Baek JH, Kim JG, Won DI et al. Transplantation with higher dose of natural killer cells associated with better outcomes in terms of non-relapse mortality and infectious events after allogeneic peripheral blood stem cell transplantation from HLA-matched sibling donors. Eur J Haematol 2005; 75: 299鈥?08.Article聽Google Scholar聽 24Yu J, Venstrom JM, Liu XR, Pring J, Hasan RS, O鈥橰eilly RJ et al. Breaking tolerance to self, circulating natural killer cells expressing inhibitory KIR for non-self HLA exhibit effector function after T cell-depleted allogeneic hematopoietic cell transplantation. Blood 2009; 113: 3875鈥?884.CAS聽 Article聽Google Scholar聽 25Cooley S, McCullar V, Wangen R, Bergemann TL, Spellman S, Weisdorf DJ et al. KIR reconstitution is altered by T cells in the graft and correlates with clinical outcomes after unrelated donor transplantation. Blood 2005; 106: 4370鈥?376.CAS聽 Article聽Google Scholar聽 26Giebel 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鈥?19.CAS聽 Article聽Google Scholar聽 27Hsu KC, Keever-Taylor CA, Wilton A, Pinto C, Heller G, Arkun K et al. Improved outcome in HLA-identical sibling hematopoietic stem-cell transplantation for acute myelogenous leukemia predicted by KIR and HLA genotypes. Blood 2005; 105: 4878鈥?884.CAS聽 Article聽Google Scholar聽 28Farag SS, Fehniger TA, Ruggeri L, Velardi A, Caligiuri MA . Natural killer cell receptors: new biology and insights into the graft- versus-leukemia effect. Blood 2002; 100: 1935鈥?947.CAS聽 Article聽Google Scholar聽 29Karre K . Immunology. A perfect mismatch. Science 2002; 295: 2029鈥?031.CAS聽 Article聽Google Scholar聽 30Velardi A, Ruggeri L, Moretta A, Moretta L . NK cells: a lesson from mismatched hematopoietic transplantation. Trends Immunol 2002; 23: 438鈥?44.CAS聽 Article聽Google Scholar聽 31Pende D, Marcenaro S, Falco M, Martini S, Bernardo ME, Montagna D et al. Anti-leukemia activity of alloreactive NK cells in KIR ligand-mismatched haploidentical HSCT for pediatric patients: evaluation of the functional role of activating KIR and redefinition of inhibitory KIR specificity. Blood 2009; 113: 3119鈥?129.CAS聽 Article聽Google Scholar聽 32Miller JS, Soignier Y, Panoskaltsis-Mortari A, McNearney SA, Yun GH, Fautsch SK et al. Successful adoptive transfer and in vivo expansion of human haploidentical NK cells in patients with cancer. Blood 2005; 105: 3051鈥?057.CAS聽 Article聽Google Scholar聽 33Freud AG, Caligiuri MA . Human natural killer cell development. Immunol Rev 2006; 214: 56鈥?2.CAS聽 Article聽Google Scholar聽 34Freud AG, Yokohama A, Becknell B, Lee MT, Mao HC, Ferketich AK et al. Evidence for discrete stages of human natural killer cell differentiation in vivo. J Exp Med 2006; 203: 1033鈥?043.CAS聽 Article聽Google Scholar聽 35Kaiser BK, Pizarro JC, Kerns J, Strong RK . Structural basis for NKG2A/CD94 recognition of HLA-E. Proc Natl Acad Sci USA 2008; 105: 6696鈥?701.CAS聽 Article聽Google Scholar聽 36Guma M, Angulo A, Vilches C, Gomez-Lozano N, Malats N, Lopez-Botet M . Imprint of human cytomegalovirus infection on the NK cell receptor repertoire. Blood 2004; 104: 3664鈥?671.CAS聽 Article聽Google Scholar聽 37Guma M, Budt M, Saez A, Brckalo T, Hengel H, Angulo A et al. Expansion of CD94/NKG2C+ NK cells in response to human cytomegalovirus-infected fibroblasts. Blood 2006; 107: 3624鈥?631.CAS聽 Article聽Google Scholar聽 38van Stijn A, Rowshani AT, Yong SL, Baas F, Roosnek E, ten Berge IJM et al. Human cytomegalovirus infection induces a rapid and sustained change in the expression of NK cell receptors on CD8+ T cells. J Immunol 2008; 180: 4550鈥?560.CAS聽 Article聽Google Scholar聽 Download referencesAcknowledgementsWe thank the Immunophenotyping unit and the Stem Cell Laboratory of the Laboratory of Hematology (Department of Laboratory Medicine) at the RUNMC for the immunophenotyping of the whole blood samples, and processing of the stem cell transplants and preparing the isolated PBMC fractions, respectively, during this study.Author informationAffiliationsDepartment of Laboratory Medicine, Laboratory of Medical Immunology, Radboud University Nijmegen Medical Centre, Nijmegen, The NetherlandsD N Eissens,聽B van Cranenbroek,聽I Joosten聽 聽A van der MeerDepartment of Hematology, Radboud University Nijmegen Medical Centre, Nijmegen, The NetherlandsN P M Schaap聽 聽A V M SchattenbergDepartment of Laboratory Medicine, Laboratory of Hematology, Radboud University Nijmegen Medical Centre, Nijmegen, The NetherlandsF W M B Preijers聽 聽H DolstraAuthorsD N EissensView author publicationsYou can also search for this author in PubMed聽Google ScholarN P M SchaapView author publicationsYou can also search for this author in PubMed聽Google ScholarF W M B PreijersView author publicationsYou can also search for this author in PubMed聽Google ScholarH DolstraView author publicationsYou can also search for this author in PubMed聽Google ScholarB van CranenbroekView author publicationsYou can also search for this author in PubMed聽Google ScholarA V M SchattenbergView author publicationsYou can also search for this author in PubMed聽Google ScholarI JoostenView author publicationsYou can also search for this author in PubMed聽Google ScholarA van der MeerView author publicationsYou can also search for this author in PubMed聽Google ScholarCorresponding authorCorrespondence to I Joosten.Additional informationSupplementary Information accompanies the paper on the Leukemia website (http://www.nature.com/leu)Supplementary information Supplementary Figure S1 (JPG 561 kb) Supplementary Figure Legend (DOC 23 kb)Rights and permissionsReprints and PermissionsAbout this articleCite this articleEissens, D., Schaap, N., Preijers, F. et al. CD3+/CD19+-depleted grafts in HLA-matched allogeneic peripheral blood stem cell transplantation lead to early NK cell cytolytic responses and reduced inhibitory activity of NKG2A. Leukemia 24, 583鈥?91 (2010). https://doi.org/10.1038/leu.2009.269Download citationReceived: 16 July 2009Revised: 09 October 2009Accepted: 04 November 2009Published: 24 December 2009Issue Date: March 2010DOI: https://doi.org/10.1038/leu.2009.269KeywordsNK cellsallogeneic stem cell transplantationNK cell reconstitutionNKG2Acytotoxicity B Federmann, M H盲gele, M Pfeiffer, S Wirths, M Schumm, C Faul, W Vogel, R Handgretinger, L Kanz W A Bethge Leukemia (2011)