SEARCH

SEARCH BY CITATION

Keywords:

  • Cytokines;
  • cytotoxicity;
  • immunosuppression;
  • NK cells;
  • transplantation

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. Disclosure
  9. References
  10. Supporting Information

Cyclosporin A (CsA), rapamycin (Rapa) and mycophenolic acid (MPA) are frequently used for GVHD prophylaxis and treatment after allogeneic stem cell transplantation (SCT). As NK cells have received great interest for immunotherapeutic applications in SCT, we analyzed the effects of these drugs on human cytokine-stimulated NK cells in vitro. Growth-kinetics of CsA-treated cultures were marginally affected, whereas MPA and Rapa severely prevented the outgrowth of CD56bright NK cells. Single-cell analysis of NK cell receptors using 10-color flow cytometry, revealed that CsA-treated NK cells gained a similar expression profile as cytokine-stimulated control NK cells, mostly representing NKG2A+KIRNCR+ cells. In contrast, MPA and Rapa inhibited the acquisition of NKG2A and NCR expression and NK cells maintained an overall NKG2AKIR+NCR+/− phenotype. This was reflected in the cytolytic activity, as MPA- and Rapa-treated NK cells, in contrast to CsA-treated NK cells, lost their cytotoxicity against K562 target cells. Upon target encounter, IFN-γ production was not only impaired by MPA and Rapa, but also by CsA. Overall, these results demonstrate that CsA, MPA and Rapa each have distinct effects on NK cell phenotype and function, which may have important implications for NK cell function in vivo after transplantation.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. Disclosure
  9. References
  10. Supporting Information

Graft-versus-host disease (GVHD) is a major complication after allogeneic stem cell transplantation (SCT) resulting in high mortality rates. Various immunosuppressive drugs (ISD) are applied to efficiently prevent GVHD. These drugs mainly focus on targeting and suppressing unwanted adaptive immune responses. However, the effect on NK cell responses after SCT remains largely unknown.

NK cells have shown to possess immunotherapeutic activity in SCT. NK cells form the first line of defense in mediating immunity against microbial pathogens, directly through cytolysis of virus-infected cells and indirectly by the production of inflammatory cytokines such as IFN-γ (1,2). Furthermore, NK cells of donor origin are efficient effectors in eradicating tumor cells without inducing severe GVHD (3–5). In HLA-haploidentical SCT, Ruggeri et al. (6) showed that donor alloreactive NK cells isolated from peripheral blood of recipients were able to lyse tumor cells derived from the recipient. This implicates that NK cells may provide immune reactivity after SCT by targeting residual tumor cells. For HLA-matched SCT, we have previously shown that the use of NK cell-enriched stem cell grafts leads to early NK cell repopulation with concomitant high cytolytic capacity within the first months after SCT (7).

The cytolytic and cytokine-producing activities of NK cells are regulated through a range of inhibitory and stimulatory receptors. Killer cell immunoglobulin-like receptors (KIR), including inhibitory (KIR-DL) and stimulatory receptors (KIR-DS), specifically recognize HLA-A,-B and -C molecules (8). The CD94:NKG2 heterodimeric complex, with its inhibitory (NKG2A) and stimulatory (NKG2C) form, is part of the C-type lectin family and recognizes HLA-E class I molecules (9). Other stimulatory NK cell receptors include the natural cytotoxicity receptors (NCR; NKp30, NKp44 and NKp46), NKG2D (a C-type lectin homodimer) and 2B4 (CD244) (10). The balance between the inhibitory and stimulatory signals triggers and modulates NK cell effector function (11,12).

Although NK cells are exposed to ISD administered to the patient post-SCT, they have to exert optimal antitumor reactivity and produce cytokines. Among the various ISD, cyclosporin A (CsA), rapamycin (Rapa) and mycophenolic acid (MPA; the active metabolite of MMF) have been applied successfully for the prevention of GVHD (13–17). However, these drugs may influence NK cell reactivity, which could have implications for their use in clinical immunotherapeutic settings. Due to distinct experimental designs, previous studies on the effect of CsA on human and murine NK cells have shown contradictory results with respect to the functional activity of NK cells after treatment (18–23). In contrast to CsA, the effect of Rapa and MPA has been studied less extensively. A previous study on rat NK cells directly comparing CsA with Rapa, showed that Rapa, in contrast to CsA, significantly inhibited the growth and cytotoxicity of rat NK cells (22). Also, in a mouse SCT model, it has been shown that MMF does not inhibit graft-versus-leukemia responses or the activity of lymphokine-activated killer (LAK) cells, suggesting that MPA may not affect functional NK cell responses after SCT (24). Here, we tested the effects of CsA, Rapa and MPA on peripheral human NK cells and show that these drugs each have a distinct impact on NK cell phenotype and function, which may have important implications for NK cell function in vivo after transplantation.

Materials and Methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. Disclosure
  9. References
  10. Supporting Information

Cell isolation

Buffy coats from healthy human donors were purchased from Sanquin Blood Bank, Nijmegen, The Netherlands, upon written informed consent with regard to scientific use. PBMC were isolated by density gradient centrifugation (Lymphoprep; Nycomed Pharma, Roskilde, Denmark). NK cells were negatively selected (Miltenyi Biotec, Bergisch Gladbach, Germany) resulting in a purity of more than 95%.

Culture conditions and ISD

Freshly isolated NK cells were cultured in the presence of rhIL-2 (100 U/mL; Chiron, Amsterdam, the Netherlands) and rhIL-15 (10 ng/mL; BioSource International, Camarillo, CA) in culture medium (RPMI 1640 medium supplemented with pyruvate (0.02 mM), glutamax (2 mM), penicillin (100 U/mL), streptomycin (100 μg/mL) and 10% human pooled serum (HPS)), in a 37°C, 95% humidity, 5% CO2 incubator. CsA (Novartis Pharma B.V., Arnhem, The Netherlands), MPA (Sigma-Aldrich, Zwijndrecht, The Netherlands) or Rapa (LC Laboratories, Woburn, MA) was added at final concentrations ranging from 0.01 to 10 μg/mL (CsA and MPA) or 0.1 to 100 ng/mL (Rapa) based on therapeutic serum levels handled at our medical centre and previously published data (25–28).

Flow cytometry

Cells were phenotypically analyzed using the FC500 and the Gallios™ Flow Cytometer (Beckman Coulter, Miami, FL). The following conjugated mAbs were used for four-color stainings (analyzed on the FC500): CD16-FITC (Dako, Glostrup, Denmark), CD69-PE, CD3-ECD and CD56-PC7 (Beckman Coulter). To exclude dead cells from analysis, 7-amino-actinomycin-D (7-AAD; Sigma-Aldrich) was added to cells prior to acquisition. For 10-color analyses on the Gallios™, the following conjugated mAbs were combined: CD16-FITC (Dako), CD159c-PE (NKG2C; R&D Systems, Minneapolis, CA), CD3-ECD, CD56-APC-A750, CD158b-PC7 (KIR2DL/S2/3), CD158e1-APC (KIR3DL1), CD158a-APC-A700 (KIR2DL/S1), CD159a-PB (NKG2A), CD45-PO, CD336-PE (NKp44), CD337-PC5.5 (NKp30), CD335-PC7 (NKp46), CD314-APC (NKG2D) and CD244-APC-A700 (2B4) (all provided by Beckman Coulter). The combinations were balanced in fluorochrome combinations to avoid antibody interactions, sterical hindrance and to detect also dimly expressing populations. Before 10-color analyses were performed, all conjugates were titrated and individually tested for sensitivity, resolution and compensation of spectral overlap. Isotype controls were used to define marker settings.

Proliferation assays

CFSE based proliferation analysis  Freshly isolated NK cells were labeled with 0.1 μM CFSE (Molecular Probe, Eugene, OR), aliquoted in CFSE labeling buffer (PBS containing 0.02% HPS), for 10 min at RT in the dark. The reaction was stopped by addition of equal volumes of cold HPS. Subsequently, the cells were washed three times with CFSE labeling buffer and resuspended in culture medium. The CFSE-labeled NK cells were cultured as described above and were analyzed using flow cytometry.

Quantitative flow cytometric analysis  Freshly isolated NK cells were incubated with CD16-FITC (Dako) and CD56-PC5 (Beckman Coulter) for 20 min at RT in the dark. CD56brightCD16+/− and CD56dimCD16+ cells were separated using high purity FACS (Altra flow cytometer, Beckman Coulter). After 5 days of culture, cells were stained for CD56 and CD16 and counted by flow cytometry using Flow-Count fluorospheres (Beckman Coulter). 7-AAD (Sigma-Aldrich) was used to exclude dead cells from analyses.

Cytotoxicity assay

HLA class-I deficient K562 target cells were labeled with 3.7 MBq (100 μCi) chromium-51 (51Cr; Perkin Elmer, Boston, MA) for 2 h in a 37°C, 95% humidity, 5% CO2 incubator. Labeled target cells were washed four times with culture medium and plated in triplicate in 96-well V-bottom plates (103 cells/well). NK cells were recovered from culture and viable NK cells were added in a 1:1 ratio and incubated for 4 h at 37°C, 95% humidity, 5% CO2. Supernatant was recovered and analyzed using a gamma counter. The percentage specific lysis was calculated as follows:% specific lysis = ((experimental release-spontaneous release)/(maximum release-spontaneous release))x100%. To assess the IFN-γ production of NK cells during target encounter, parallel incubations with unlabeled K562 target cells were performed.

ELISA

IFN-γ production of NK cells during incubation with K562 cells was measured by ELISA (Sanquin (CLB), Amsterdam, The Netherlands). Supernatants were collected after 4 h of incubation and stored at −80°C until further use. IFN-γ ELISA was performed according to the manufacturer's instructions.

Statistical analysis

Statistical analysis was performed using GraphPad Prism 4.0. For comparisons between cytokine-stimulated control NK cells and NK cells cultured in the presence of ISD, we used ANOVA analysis with Dunnett's multiple comparison test for post-testing. Paired t-tests were used to analyze the statistical difference between freshly isolated NK cells and cytokine-stimulated control NK cells. The Wilcoxon signed rank test was used for non-normally distributed data. p-values ≤0.05 were considered statistically significant.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. Disclosure
  9. References
  10. Supporting Information

Dose-dependent inhibition of NK cell proliferation

To study the effect of CsA, Rapa, and MPA on the proliferation capacity of isolated NK cells, CFSE-labeled NK cells were cultured in the presence of 100 U/mL IL-2 and 10 ng/mL IL-15 with increasing concentrations of the drugs. After a 5-day culture, CFSE dilution patterns showed a median proliferation rate of 44% (range 22–78%) of cytokine-stimulated control NK cells (Figure 1A), in five different division cycles (Figure 1B). Rapa and MPA significantly inhibited NK cell proliferation in a dose-dependent way. In contrast, CsA only showed a trend toward reduced proliferation.

image

Figure 1. Proliferation of NK cells at day 5 of culture after CsA, Rapa and MPA treatment. (A) The proliferation of NK cells isolated from 4 different donors with increasing amounts of CsA, Rapa and MPA. Results are shown in median with range. Differences between the cytokine-stimulated control and drug-treated NK cells were analyzed using ANOVA; *p < 0.05, **p < 0.01, ***p < 0.001. (B) CFSE division patterns of cytokine-stimulated control and CsA-, Rapa- and MPA-treated NK cells from one representative donor. Numbers in the upper right are the percentage of divided cells.

Download figure to PowerPoint

Freshly isolated NK cells can generally be separated into two distinguishable subsets. The CD56brightCD16+/− subset comprises 5–10% of all peripheral blood NK cells, while the rest has a CD56dimCD16+ phenotype (29). In steady state, the CD56brightCD16+/− subset is described to be more immunoregulatory through cytokine production (e.g. IFN-γ, TNF-α), whereas the CD56dimCD16+ subset is more cytotoxic. After activation, both subsets are able to produce cytokines and possess cytolytic activity. We analyzed the proliferation capacity of both subsets under influence of the different ISD. The enriched CD56brightCD16+/− and CD56dimCD16+ subsets, having equal expression patterns of the IL-2 and IL-15 receptor (Figure S1, Table S1), were incubated separately in the presence or absence of CsA, Rapa and MPA (Figure 2). The cytokine-stimulated control CD56brightCD16+/− cells showed a significant increase in cell number after 5 days of culture as compared to the initial amount of cells (p = 0.014), whereas the amount of CD56dimCD16+/− cells decreased (p = 0.065) suggesting that the proliferation seen in the whole NK cell population (Figure 1) is only due to the proliferation of the CD56brightCD16+/− NK cell subset (Figure 2A). Treatment of either NK cell subset with ISD showed that Rapa and MPA impaired the growth of the CD56brightCD16+/− cells, whereas CsA only inhibited the growth of these cells at higher concentrations. In contrast to the CD56brightCD16+/− subset, Rapa and MPA did not affect the amount of CD56dimCD16+ cells. However, CsA showed a significant decrease at 1000 ng/mL. Thus, Rapa and MPA only influence the amount of NK cells within the CD56brightCD16+/− subset, whereas CsA is capable of affecting NK cell numbers of both subsets.

image

Figure 2. CD56brightCD16+/− and CD56dimCD16+ NK cell subsets at day 5 of culture after CsA, Rapa and MPA treatment. (A) Fold expansion of sorted CD56brightCD16+/− NK cells (upper panel) and sorted CD56dimCD16+ NK cells (lower panel) isolated from 4 different donors with increasing amounts of CsA, Rapa and MPA. Results are shown in median with range. (B) CD56/CD16 phenotype of CD56brightCD16+/− and CD56dimCD16+ NK cells before and after culture of one representative donor. The upper panels show the CD56/CD16 phenotype of both NK cell subsets before and after culture without ISD. The lower panels show the phenotype of the NK cell subsets after CsA, Rapa and MPA treatment. (C) The percentage of CD56++CD16+ cells within the sorted CD56bright subset before (start) and after culture in the absence (CTRL) or presence of CsA, Rapa and MPA. Results are shown in median with range. (D) The percentage of CD56++CD16+ cells within the sorted CD56dim subset before (start) and after culture in the absence (CTRL) or presence of CsA, Rapa and MPA. Results are shown in median with range. The difference cytokine-stimulated control NK cells before (Start) and after culture (CTRL) was analyzed using paired t tests; *p < 0.05, **p < 0.01, ***p < 0.001. Differences between the cytokine-stimulated control (CTRL) and drug-treated NK cells were analyzed using ANOVA; *p < 0.05, **p < 0.01, ***p < 0.001.

Download figure to PowerPoint

Rapa and MPA, but not CsA, affect the CD56bright/CD56dim balance

Interestingly, while maintaining their CD56bright phenotype, the sorted CD56brightCD16+/− subset increased the expression of CD16 during culture (p < 0.05), whereas the CD56dimCD16+/− cells elevated their CD56 expression (p < 0.05) and lowered their CD16 expression (Figure 2B (upper panels), C, D). In the CD56brightCD16+/− subset, Rapa (10 ng/mL, p < 0.05; 50 ng/mL, p < 0.01) and MPA (500 ng/mL, p < 0.001) dose-dependently inhibited the upregulation of CD16 expression, whereas CsA-treated CD56brightCD16+/− cells revealed a similar phenotype as compared to cytokine-stimulated control cells at day 5 of culture (Figure 2B (left panels), C, D). As cytokine-stimulated control CD56dimCD16+/− cells elevated their CD56 expression, especially Rapa inhibited the upregulation of CD56 with increasing concentrations (10 ng/mL and 50 ng/mL, p < 0.01; Figure 2B (right panels), C, D). This inhibition was also seen after treatment with CsA and MPA, although to a lesser extent.

The differential immunosuppressive effect on the proliferation and phenotype of the separately cultured NK cell subsets was also reflected on the phenotype in cultures of the whole NK cell population. During culture, the balance between the CD56bright and CD56dim subsets shifted toward a more CD56bright phenotype during culture (Figure 3A and B). As CsA showed to have a marginal effect on the growth of CD56brightCD16+/− cells and did not strongly affect the changing phenotype of the NK cell subsets (Figure 2), the whole NK cell population showed the same shift toward a CD56bright phenotype as compared to cytokine-stimulated control cells (Figure 3A and B). In contrast, mainly due to their strong inhibitory effect on the proliferation of the CD56bright subset, the majority of the NK cell population after Rapa and MPA treatment maintained a CD56dim phenotype. Thus, under immunosuppressive treatment, Rapa and MPA, but not CsA, inhibit the outgrowth of CD56bright NK cells and thereby affect the CD56bright/CD56dim distribution of the whole NK cell population.

image

Figure 3. CD56brightCD16+/− and CD56dimCD16+ distribution within the NK cell population before and after culture with CsA, Rapa and MPA. (A) Shown are two representative donors before (start) and after culture in the absence (CTRL) or presence of CsA (1000 ng/mL), Rapa (50 ng/mL) and MPA (500 ng/mL). Gated are the CD56brightCD16+/- and CD56dimCD16+ NK cells. (B) The percentage of CD56brightCD16+/− (left panel) and CD56dimCD16+ (right panel) NK cells within the NK cell population. Shown are the results for 7 different donors before and after culture. The difference between cytokine-stimulated control NK cells before (S) and after culture (CTRL) was analyzed using paired t tests; ***p < 0.001. Differences between the cytokine-stimulated control (CTRL) and drug-treated NK cells were analyzed using ANOVA; **p < 0.01.

Download figure to PowerPoint

Differential phenotypical changes of cultured NK cells under immunosuppressive treatment

Using 10-color flow cytometry, we analyzed the effect of the different ISD on the phenotype of cultured NK cells. After a 5-day culture, cytokine-stimulated control NK cells showed a significant increase in the percentage of activating receptors (NKG2D, NCR) as compared to freshly isolated NK cells (Table 1). In addition, the percentage of KIR+ cells was reduced, paralleled by a significant increase in the percentage of NKG2A+ cells. Thus, in a cytokine-driven NK cell culture, the NK cell population appears to reach a more activated state and shifts the inhibitory control by the KIR repertoire toward control by the NKG2A receptor complex.

Table 1.  NK cell phenotype after culture
 Isolation %CTRLCsARapaMPA
%p1%2p%p2%2p
  1. NK cells were isolated and cultured for 5 days in the absence (CTRL) or presence of immunosuppressive drugs (CsA; 1000 ng/mL, Rapa; 50 ng/mL, MPA; 500 ng/mL). Isolated NK cells and (un)treated NK cells were analyzed by flow cytometric analysis using anti-KIR, -NKG2A, -NKG2C, -NKG2D, -CD244 and -NCR mAbs. Medians (with ranges) are shown (n = 7). ns indicates no significant difference.

  2. 1The difference between cytokine-stimulated control NK cells before (Isolation) and after culture (CTRL) was analyzed using the Wilcoxon signed rank test.

  3. 2Differences between the cytokine-stimulated control (CTRL) and drug-treated NK cells were analyzed using ANOVA.

KIR2DL/S121 (13–28)17 (8–24)ns16 (5–21)ns20 (9–31)<0.0125 (16–36)<0.01
KIR2DL/S2/340 (16–62)31 (14–43)ns24 (7–44)<0.0530 (15–50)ns48 (19–61)<0.01
KIR3DL115 (0.0–28) 5 (0.1–16)<0.01 3 (0.3–15)ns 3 (0.3–16)ns 4 (0.4–21)ns
NKG2A44 (16–58)84 (68–94)<0.00172 (45–89)<0.0569 (44–81)<0.0150 (34–74)<0.01
NKG2C14 (4–50)13 (7–34)ns14 (7–31)ns13 (6–27)ns20 (6–38)ns
NKG2D59 (46–81)95 (44–98)<0.0594 (56–99)ns84 (19–97)<0.0577 (14–92)<0.01
2B4 (CD244)99 (96–100)96 (88–98)<0.0597 (86–98)ns86 (78–95)<0.0188 (78–96)<0.01
NKp3051 (29–57)83 (28–94)ns93 (69–99)<0.0574 (17–95)ns55 (11–88)<0.05
NKp441.1 (0.3–2)64 (37–96)<0.00165 (29–95)ns33 (10–87)<0.0119 (5–59)<0.01
NKp4669 (36–78)90 (56–94)<0.0581 (50–88)ns85 (56–93)ns85 (48–92)ns

Cytokine-stimulated NK cells treated with CsA only showed minor changes in the percentages of NK cell receptor-positive cells as compared to cytokine-stimulated control NK cells (Table 1). With respect to KIR expression, the percentage of KIR2DL/S2/3+ NK cells was significantly lower (p < 0.05) and there was significant less increase of the percentage of NKG2A+ NK cells (p < 0.05). With regard to NCR expression, the percentage of NKp30+ NK cells was significantly higher as compared to cytokine-stimulated control NK cells (p < 0.05). In contrast to the minor changes observed for CsA treatment, MPA treatment had a major impact on the NK cell receptor repertoire, i.e. the percentage of KIR+ NK cells remained significantly higher (p < 0.01 for KIR2DL/S1 and KIR2DL/S2/3) and the percentage of NKG2A+ NK cells was significantly lower (p < 0.01) as compared to cytokine-stimulated control NK cells. Furthermore, the KIR/NKG2A distribution within the NK cells much resembled the situation of freshly isolated NK cells. Nevertheless, MPA allowed for the activation of NK cells since the percentage of NK cells positive for NCR and NKG2D increased during NK cell culture. For all NK cell receptors analyzed, Rapa showed to have intermediate effects on the NK cell receptor repertoire as compared to CsA and MPA. In summary, MPA, and Rapa to a lesser extent, inhibited reformation of the NK cell receptor repertoire during culture, whereas CsA only had mild effects on the NK cell receptor repertoire.

Differential distribution of KIR/NKG2A and NCR upon immunosuppressive treatment

By using KIR and NKG2A/C monoclonal antibodies (mAbs) in one 10-color panel and NKG2D, 2B4, NCR mAbs in a second 10-color panel, we could analyze the expression of multiple combinations of different NK cell receptors on a single-cell level.

As there were no significant changes in NKG2C expression (Table 1), we focused our analysis on NK cell subsets positive for KIR and/or NKG2A. This analysis showed that freshly isolated NK cells largely consist of subsets that either express NKG2A without KIR (median:26%) or KIR without NKG2A (median:42%) (Figure 4Ai). Only a small proportion of cells expressed both receptors (median:10%) and the remaining cells had no expression of NKG2A and KIR. Culturing NK cells with IL-2 and IL-15 significantly changed the distribution of these subsets as almost all NK cells expressed NKG2A either in combination with KIR expression (median:25%) or without KIR expression (median:58%). The percentages of NKG2A negative NK cell subsets were significantly decreased after culture. As compared to cytokine-stimulated control cells, CsA-treated cells showed a similar distribution pattern of NKG2A+ and/or KIR+ NK cells after culture and most of the cells were NKG2A+KIR (median:58%) (Figure 4Aii). However, there were significant less NKG2A+KIR+ NK cells and significant more NKG2AKIR NK cells present within the whole NK cell population. In contrast to CsA treatment, MPA-treated, and to a lesser extent Rapa-treated, NK cells had a similar distribution of NKG2A and KIR as freshly isolated NK cells and in this respect significantly differed from cytokine-stimulated control cells.

image

Figure 4. KIR/NKG2A and NCR distribution within the NK cell population. Shown are the results for 7 different donors in median with range. (A) The distribution of KIR and NKG2A expression within the NK cell population from (i) freshly isolated NK cells (start) versus cytokine-stimulated control NK cells (CTRL) cultured for 5 days and (ii) the KIR and NKG2A expression after 5 days of culture in the absence (CTRL) versus presence of CsA (1000 ng/mL), Rapa (50 ng/mL) and MPA (500 ng/mL). (B) The expression of the different NCR (NKp30, NKp44, NKp46) within the NK cell population: —= NCR negative; +–= single NCR+ (cells expressing only one of the NCR); ++- = double NCR+ (cells expressing two different NCR); +++= triple NCR+ (cells expressing all three NCR). Shown are the distribution of NCR expression within the NK cell population from (i) freshly isolated NK cells (start) and cytokine-stimulated control NK cells (CTRL) cultured for 5 days and (ii) the NCR expression after 5 days of culture in the absence (CTRL) versus presence of CsA (1000 ng/mL), Rapa (50 ng/mL) and MPA (500 ng/mL). The difference between cytokine-stimulated control NK cells before and after culture (CTRL) was analyzed using the Wilcoxon signed rank test; *p < 0.05, **p < 0.01. Differences between the cytokine-stimulated control (CTRL) and drug-treated NK cells were analyzed using ANOVA; *p < 0.05, **p < 0.01.

Download figure to PowerPoint

Further analysis of the stimulatory NCR receptors (NKp30, NKp44, and NKp46) and their distribution within the NK cell population, showed that freshly isolated NK cells either contained single (one NCR; median:37%) or double (combination of two different NCR; median:35%) NCR+ NK cells or did not express any NCR (median:21%) (Figure 4Bi). After culture, the NK cell population showed a more activated phenotype as they primarily consisted of triple NCR+ NK cells (expression of all three NCR; median:56%), only 2% remained NCR. CsA treatment had no effect on the acquisition of NCR by NK cells and showed the same distribution pattern as cytokine-stimulated control NK cells (Figure 4Bii). As MPA-treated NK cells were hampered in their upregulation of NKp30 and NKp44 (Table 1), most NK cells were single (median:28%) or double (median:39%) NCR+ and only a median of 27% expressed all three NCR (Figure 4Bii). As compared to MPA-treated NK cells, Rapa inhibited NCR upregulation to a lesser extent and revealed a median expression distribution of 20% single, 31% double, and 28% triple NCR+ NK cells. Thus, while KIR/NKG2A data suggest that the NK cells maintain their original status, NCR data indicate that the cells are able to upregulate NCR expression, albeit it significantly lower than the cytokine-stimulated control NK cells.

In summary, MPA and Rapa inhibited the NK cell receptor repertoire from shifting toward an overall NKG2A+KIRNCR+ phenotype and maintained an overall NKG2AKIR+NCR+/− phenotype, whereas CsA-treated NK cells were hardly different from cytokine-stimulated control NK cells. Detailed analysis on KIR/NKG2A and NCR modulation within sorted and separately cultured CD56brightCD16+/− and CD56dimCD16+ subsets, revealed that both subsets are subjected to change in phenotype after culture (Figure S2, S3). Thus, the net results for the whole NK cell population are based on the KIR/NKG2A and NCR modulation of both NK cell subsets together with the proliferation of the CD56bright subset.

Cytolytic activity and IFN-γ production after immunosuppressive treatment

As the balance between the inhibitory and stimulatory receptors of the NK cell receptor repertoire was differentially affected by CsA, Rapa and MPA, we assessed the overall cytotoxicity against MHC class I-negative K562 target cells in a 1:1 ratio. In parallel, we analyzed the IFN-γ production after the encounter with K562 target cells. To analyze the overall activation of NK cells before target encounter, we analyzed the expression of CD69, an early lymphocyte activation marker that plays a role in various NK cell functions (30).

Upon NK cell isolation from healthy donor PBMC fractions, only few NK cells express CD69 (data not shown). After culture, Rapa showed a dose-dependent inhibition of CD69 expression with a significant lower percentage of CD69+ cells at 100 ng/mL. Although not significant, CsA- and MPA-treated NK cells showed a trend toward a dose-dependent inhibition in the median percentage of CD69 expression (Figure 5A).

image

Figure 5. Activation, cytolytic activity and IFN-γ production after culture with CsA, Rapa and MPA. After culture in the absence or presence of CsA, Rapa and MPA, the expression of CD69, the cytolytic NK cell activity and the IFN-γ production was analyzed. (A) Expression of the early activation marker CD69 at day 5 of culture after CsA, Rapa and MPA treatment. Shown are the results for 5 different donors in median with range. (B) The cytolytic activity (specific lysis) was measured in a 4 h 51Cr-release assay containing NK cells in a 1:1 ratio with K562 target cells. Shown are the results for 6 different donors in median with range. (C) Supernatant of parallel 4 h co-cultures, containing NK cells from the same donors (n = 6) and K562 target cells in a 1:1 ratio, was analyzed in ELISA assays for the production of IFN-γ. Differences between the cytokine-stimulated control (0 ng/mL) and drug-treated NK cells were analyzed using ANOVA; *p < 0.05, **p < 0.01.

Download figure to PowerPoint

After a 5-day culture, NK cells showed to be significantly impaired in their cytolytic activity after Rapa and MPA treatment, but not after CsA treatment (Figure 5B). However, as CsA did not affect the cytolytic activity of the NK cells, IFN-γ production was severely impaired (Figure 5C). Thus, upon target encounter, Rapa- and MPA-treated NK cells, but not CsA-treated NK cells, are impaired in their cytolytic activity. IFN-γ production was affected by all three ISD.

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. Disclosure
  9. References
  10. Supporting Information

Within the first months after SCT, NK cells contribute to graft-versus-leukemia (GVL) responses without the induction of GVHD and form a first line of defense against infectious agents. This makes NK cells attractive candidates for immunotherapeutic purposes in SCT. The presence of NK cells in the graft proved to be beneficial for clinical outcome (31,32). In these cases, NK cells need to be fully functional in the early phase after SCT, despite high level of immunosuppressive treatment. Therefore, it is important to study the influence of ISD on NK cell activity in order to select the drug with less influence on NK cell function. In this study, we show that CsA, Rapa and MPA have diverse effects on NK cell phenotype and function, which may have important implications for their use in GVHD prophylaxis.

We show that CsA had minor effects on NK cell proliferation, whereas Rapa and MPA severely inhibited the outgrowth of CD56brightCD16+/− NK cells. As we showed that CsA-treated NK cells had a similar CD56bright/CD56dim distribution as cytokine-stimulated control NK cells, Wang et al. (21) revealed that CsA could even increase the percentage of CD56bright NK cells as compared to cytokine-stimulated control cells. This difference might be explained by our distinct gating strategy, as we, in contrast to Wang et al. (21), also included the CD56brightCD16+ cells within our CD56bright subset, in consensus with gating strategies described by Cooper et al. (29). When excluding the CD16+ cells from the CD56bright subset, we also noticed a tendency toward a higher percentage of CD56brightCD16 NK cells after CsA (Figure 3A).

Treatment of NK cells with CsA had a minimal effect on the changes within the NK cell receptor distribution during culture, resembling the NK cell receptor repertoire of cytokine-stimulated control NK cells. However, MPA- and Rapa-treated NK cells were inhibited in their shift toward a NKG2A+KIRNCR+ phenotype and largely maintained a NKG2AKIR+NCR+/− phenotype, also observed in freshly isolated NK cells. However, in contrast to freshly isolated or cytokine activated NK cells the MPA or Rapa treated cells were significantly inhibited in their cytolytic capacity. Thus, MPA and Rapa are able to inhibit shifts in the NK cell receptor repertoire and also inhibit cytolytic capacity, whereas CsA does not interfere with changes in the NK cell receptor repertoire within an in vitro culture system and leaves the cytolytic capacity intact. Recently, we also demonstrated that CsA does not affect NK cell receptor repertoire shifts and function in vivo, as we showed, in a randomized phase III study, that transplantation with CD3+/CD19+ cell depleted grafts, in contrast to CD34+ enriched grafts, resulted in the development of a different NK cell population which was more prone to activation due to equal distribution of NKG2A and NKG2C (7). Both patient groups were treated with the same immunosuppressive regimen, using CsA as a first-line agent, as part of their GVHD prophylaxis. Thus, MPA and Rapa, but not CsA, are likely to inhibit the outgrowth of functional NK cells and therefore could have a negative effect on NK cell mediated GVL responses in vivo after SCT.

Clearly, MPA and Rapa showed a dose-dependent inhibition in NK cell cytotoxicity against K562 target cells, whereas CsA did not affect cytotoxicity. In addition, each of the ISD severely inhibited the IFN-γ production upon target encounter. As NK cells are a major source of IFN-γ and IFN-γ plays a major role in facilitating GVL effects and preventing GVHD (33), this could have clinical implications for patients after SCT. A recent study, however, showed that CsA-treated NK cells sustain their IFN-γ production after IL-12 and IL-18 stimulation (21). This suggests that CsA selectively inhibits IFN-γ production upon target encounter, but sustains the IFN-γ production capacity after IL-12 and IL-18 stimulation. Fauriat et al. (34) recently described that upon target encounter the CD56dim subset, and not the CD56bright subset, is responsible for IFN-γ production, whereas the CD56bright subset is the major source of IFN-γ during IL-12 and IL-18 stimulation. Thus, our data suggest that CsA, Rapa, and MPA treatment clearly inhibit the CD56dim cells from producing IFN-γ, leaving the CD56bright cells as the only remaining NK cells left for IFN-γ production. As MPA, and Rapa to a lesser extent, prevent the outgrowth of CD56bright cells, this could have important implications for the overall IFN-γ production and therefore the IFN-γ-mediated GVL effects after SCT. In contrast, as CsA does not inhibit the outgrowth of CD56bright NK cells, the overall IFN-γ production after SCT might be less affected and IFN-γ-mediated GVL effects driven by cytokines should remain intact. As the immunotherapeutic applications for NK cells have received huge interests in the field of SCT, it is important to gain insight in the effects of the limiting conditions in the different settings of transplantation in which NK cells need to be able to function. Immunosuppression is one of the conditions that may influence NK cell functionality. In this study, we have demonstrated that CsA, in contrast to Rapa and MPA has the least effect on NK cell phenotype and function in vitro. Thus, this study clearly suggests that the choice of immunosuppressive treatment might affect the outcome of NK cell therapy in vivo after transplantation. Additional studies on NK cell phenotype and function of patients after SCT using different immunosuppressive strategies are warranted to survey the in vivo effect of the different immunosuppressive regimens in more detail.

Acknowledgments

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. Disclosure
  9. References
  10. Supporting Information

Technical support on the Gallios™ and specified conjugated mAbs were kindly provided by Beckman Coulter. The authors thank Rob Woestenenk (Dept. of Laboratory Medicine – Laboratory of Hematology) for cell sorting. We also thank Talia Latuhihin for her technical support in the early stages of this work.

Authorship: D.E. designed the research, performed experiments, analyzed data and wrote the paper. A.M. designed the research, analyzed data and wrote the paper. B.C. performed experiments and analyzed data. F.P. designed the research and gave technical support. I.J. designed the research, analyzed data and wrote the paper.

Disclosure

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. Disclosure
  9. References
  10. Supporting Information

The authors of this manuscript have no conflicts of interest to disclose as described by the American Journal of Transplantation.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. Disclosure
  9. References
  10. Supporting Information
  • 1
    Robertson MJ, Ritz J. Biology and clinical relevance of human natural killer cells. Blood 1990; 76: 24212438.
  • 2
    Trinchieri G. Biology of natural killer cells. Adv Immunol 1989; 47: 187376.
  • 3
    Cook MA, Milligan DW, Fegan CD 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: 15211526.
  • 4
    Giebel S, Nowak I, Wojnar J et al. Impact of activating killer immunoglobulin-like receptor genotype on outcome of unrelated donor-hematopoietic cell transplantation. Transplant Proc 2006; 38: 287291.
  • 5
    Ruggeri L, Capanni M, Urbani E et al. Effectiveness of donor natural killer cell alloreactivity in mismatched hematopoietic transplants. Science 2002; 295: 20972100.
  • 6
    Ruggeri L, Capanni M, Casucci M et al. Role of natural killer cell alloreactivity in HLA-mismatched hematopoietic stem cell transplantation. Blood 1999; 94: 333339.
  • 7
    Eissens DN, Schaap NP, Preijers FW 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 2010; 24: 583591.
  • 8
    Parham P. MHC class I molecules and KIRs in human history, health and survival. Nat Rev Immunol 2005; 5: 201214.
  • 9
    Braud VM, Allan DS, O’Callaghan CA et al. HLA-E binds to natural killer cell receptors CD94/NKG2A, B and C. Nature 1998; 391: 795799.
  • 10
    Moretta A, Bottino C, Vitale M et al. Activating receptors and coreceptors involved in human natural killer cell-mediated cytolysis. Annu Rev Immunol 2001; 19: 197223.
  • 11
    Farag SS, Caligiuri MA. Human natural killer cell development and biology. Blood Rev 2006; 20: 123137.
  • 12
    Papamichail M, Perez SA, Gritzapis AD, Baxevanis CN. Natural killer lymphocytes: Biology, development, and function. Cancer Immunol Immunother 2004; 53: 176186.
  • 13
    Cutler C, Li S, Ho VT et al. Extended follow-up of methotrexate-free immunosuppression using sirolimus and tacrolimus in related and unrelated donor peripheral blood stem cell transplantation. Blood 2007; 109: 3108114.
  • 14
    Haentzschel I, Freiberg-Richter J, Platzbecker U et al. Targeting mycophenolate mofetil for graft-versus-host disease prophylaxis after allogeneic blood stem cell transplantation. Bone Marrow Transplant 2008; 42: 113120.
  • 15
    Schleuning M, Judith D, Jedlickova Z et al. Calcineurin inhibitor-free GVHD prophylaxis with sirolimus, mycophenolate mofetil and ATG in Allo-SCT for leukemia patients with high relapse risk: An observational cohort study. Bone Marrow Transplant 2008; 43: 717723.
  • 16
    Vogelsang GB, Arai S. Mycophenolate mofetil for the prevention and treatment of graft-versus-host disease following stem cell transplantation: Preliminary findings. Bone Marrow Transplant 2001; 27: 12551262.
  • 17
    Neumann F, Graef T, Tapprich C et al. Cyclosporine A and Mycophenolate Mofetil vs Cyclosporine A and Methotrexate for graft-versus-host disease prophylaxis after stem cell transplantation from HLA-identical siblings. Bone Marrow Transplant 2005; 35: 10891093.
  • 18
    Introna M, Allavena P, Spreafico F, Mantovani A. Inhibition of human natural killer activity by cyclosporin A. Transplantation 1981; 31: 113116.
  • 19
    Kosugi A, Shearer GM. Effect of cyclosporin A on lymphopoiesis. III. Augmentation of the generation of natural killer cells in bone marrow transplanted mice treated with cyclosporin A. J Immunol 1991; 146: 14161421.
  • 20
    Petersson E, Qi Z, Ekberg H, Ostraat O, Dohlsten M, Hedlund G. Activation of alloreactive natural killer cells is resistant to cyclosporine. Transplantation 1997; 63: 11381144.
  • 21
    Wang H, Grzywacz B, Sukovich D et al. The unexpected effect of cyclosporin A on CD56+CD16- and CD56+CD16+ natural killer cell subpopulations. Blood 2007; 110: 15301539.
  • 22
    Wai LE, Fujiki M, Takeda S, Martinez OM, Krams SM. Rapamycin, but not cyclosporine or FK506, alters natural killer cell function. Transplantation 2008; 85: 145149.
  • 23
    Lin SJ, Kuo ML. Effect of cyclosporin A on interleukin-15-activated umbilical cord blood natural killer cell function. Cytotherapy 2008; 10: 397405.
  • 24
    Shapira MY, Hirshfeld E, Weiss L et al. Mycophenolate mofetil does not suppress the graft-versus-leukemia effect or the activity of lymphokine-activated killer (LAK) cells in a murine model. Cancer Immunol Immunother 2005; 54: 383388.
  • 25
    Bornhauser M, Schuler U, Porksen G et al. Mycophenolate mofetil and cyclosporine as graft-versus-host disease prophylaxis after allogeneic blood stem cell transplantation. Transplantation 1999; 67: 499504.
  • 26
    Cutler C, Kim HT, Hochberg E et al. Sirolimus and tacrolimus without methotrexate as graft-versus-host disease prophylaxis after matched related donor peripheral blood stem cell transplantation. Biol Blood Marrow Transplant 2004; 10: 328336.
  • 27
    Jenke A, Renner U, Richte M et al. Pharmacokinetics of intravenous mycophenolate mofetil after allogeneic blood stem cell transplantation. Clin Transplant 2001; 15: 176184.
  • 28
    Kagawa Y, Sawada J, Yamada S et al. Relationship between development of nephrotoxicity and blood concentration of cyclosporine A in bone-marrow transplanted recipients who received the continuous intravenous infusion. Biol Pharm Bull 2003; 26: 11151119.
  • 29
    Cooper MA, Fehniger TA, Caligiuri MA. The biology of human natural killer cell subsets. Trends Immunol 2001; 22: 633640.
  • 30
    Borrego F, Robertson MJ, Ritz J, Pena J, Solana R. CD69 is a stimulatory receptor for natural killer cell and its cytotoxic effect is blocked by CD94 inhibitory receptor. Immunology 1999; 97: 159165.
  • 31
    Bethge WA, Haegele M, Faul C 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: 17461752.
  • 32
    Kim DH, Sohn SK, Lee NY 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: 299308.
  • 33
    Yang YG, Wang H, Asavaroengchai W, Dey BR. Role of Interferon-gamma in GVHD and GVL. Cell Mol Immunol 2005; 2: 323329.
  • 34
    Fauriat C, Long EO, Ljunggren HG, Bryceson YT. Regulation of human NK cell cytokine and chemokine productionby target cell recognition. Blood 2010; 115: 21672176.

Supporting Information

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. Disclosure
  9. References
  10. Supporting Information

Figure S1: Expression of IL2Ra (CD25) and IL2Rb (CD122) on sorted CD56brightCD16+/- and CD56dimCD16+ NK cell subsets.

Figure S2: KIR/NKG2A and NCR distribution within the CD56brightCD16+/- NK cell subset.

Figure S3: KIR/NKG2A and NCR distribution within the CD56dimCD16+ NK cell subset.

Table S1: Expression of IL2Rα (CD25) and IL2Rβ (CD122) on sorted CD56brightCD16+/- and CD56dimCD16+ NK cells.

Please note: Wiley-Blackwell are not responsible for the content or functionality of any supporting materials supplied by the authors. Any queries (other than missing material) should be directed to the corresponding author for the article.

FilenameFormatSizeDescription
AJT_3242_sm_FigureS1.tif274KSupporting info item
AJT_3242_sm_FigureS2.tif776KSupporting info item
AJT_3242_sm_FigureS3.tif777KSupporting info item
AJT_3242_sm_TableS1.doc29KSupporting info item

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.