Activated Allogeneic NK Cells as Suppressors of Alloreactive ...

Activated Allogeneic NK Cells as Suppressors of Alloreactive Responses Bo Hu,1 Yan He,1 Yan Wu,1 Guangming Bao,1 Haiyan Liu,1,2 Lisbeth A. Welniak,3 William J. Murphy3 Donor NK cells have been shown to be able to promote engraftment during allogeneic bone marrow transplantation. They could specifically suppress or delete host reactive cells, thereby facilitating engraftment of donor marrow. To further elucidate the mechanism, we showed that activated H2d ALAK cells (adherent lymphokine activated killer, IL-2 activated T cell-depleted bone marrow and spleen cells) from BALB/c mice significantly suppressed the proliferation of H2b splenocytes from C57BL/6 mice in mixed lymphocyte responses (MLR) stimulated with irradiated H2d splenocytes from BALB/c mice (P \ .01). The ability for H2b splenocytes to kill H2d tumor targets was also significantly inhibited by activated H2d ALAK cells (P \.01). The same number of H2b ALAK cells or H2d splenocytes did not show the same suppressive effect. These results suggested that activated H2d ALAK cells could specifically suppress the anti-H2d activity of the H2b splenocytes. Anti-tumor growth factor (TGF)b antibody blockade did not diminish this suppressive effect of ALAK cells, suggesting that this activity is not dependent on TGF-b secretion. ALAKs from gld (FasL mutant) mice suppressed the allo-responses as well as the wild-type ALAK cells. The ALAKs from pfp (perforin knockout) mice did not completely block the inhibitory effect, which suggested that the suppressive effect of the allogeneic ALAK cells could be partially caused by perforin-mediated killing. We further demonstrated that donor ALAK cells could promote engraftment by suppressing host alloreactive responses in a nonmyeloablative allogeneic BMT model. These studies suggest that activated donor NK cells specifically suppress the alloreactive cells and provide a promising way to promote donor engraftment without involving systemic and nonspecific suppression of the immune system. Biol Blood Marrow Transplant 16: 772-781 (2010) Ó 2010 American Society for Blood and Marrow Transplantation

KEY WORDS: Natural killer cells, Mixed lymphocyte responses, Allogeneic BMT

INTRODUCTION Allogeneic bone marrow transplantation (BMT) has revolutionized the treatment of leukemia, lymphoma, and other hematologic malignancies. During From the 1Laboratory of Cellular and Molecular Tumor Immunology, Cyrus Tang Hematology Center, Jiangsu Institute of Hematology, the First Affiliated Hospital, Soochow University, Suzhou, People’s Republic of China; 2Thrombosis and Hemostasis Key Lab of the Ministry of Health; Soochow University, Suzhou, People’s Republic of China; and 3Department of Microbiology and Immunology, University of Nevada, Reno, Nevada. Current address of William J. Murphy: Department of Dermatology, School of Medicine, University of California, Davis, CA. Financial disclosure: See Acknowledgments on page 781. Correspondence and reprint requests: Haiyan Liu, PhD, Laboratory of Cellular and Molecular Tumor Immunology, Cyrus Tang Hematology Center, Jiangsu Institute of Hematology, The First Affiliated Hospital, Soochow University, Suzhou, Jiangsu 215123, People’s Republic of China (e-mail: [email protected]). Received August 29, 2009; accepted February 23, 2010 Ó 2010 American Society for Blood and Marrow Transplantation 1083-8791/$36.00 doi:10.1016/j.bbmt.2010.02.023

772

BMT, a patient will receive a conditioning regimen that can either be myeloablative or of reduced intensity consisting of less cytoreductive conditioning but involves more extensive immunosuppression treatments [1,2]. However, despite immunosuppression, marrow rejection by host effector cells remains a significant concern. It would be extremely desirable to develop means to promote donor engraftment that do not involve extensive, systemic, and nonspecific suppression of the immune system. Natural killer (NK) cells represent a key component of the innate immune system and can mediate MHC unrestricted cytotoxicity against neoplastic and virally infected cells; they can also secrete numerous effector cytokines [3,4]. NK cells can mediate rejection of bone marrow but not solid-tissue allografts [5]. However, NK alloreactivity in the opposite direction, that is, donor-versus-recipient, have the potential to improve engraftment of mismatched transplants. In mice, the pretransplant infusion of donor alloreactive NK cells was successfully combined with reducedintensity conditioning (RIC) to achieve durable full donor engraftment [6]. It was also shown that donor

Biol Blood Marrow Transplant 16:772-781, 2010

NK cells can contribute to the antitumor effect [7,8]. Therefore, we hypothesize that transfer of donor NK cells may result in specific suppression or deletion of host alloreactive cells, thereby facilitating engraftment of donor marrow. In mice, we postulate that donor NK cells, being activated in vitro with IL-2, would promote engraftment by the mechanism in which host alloreactive cells upon interacting with the activated donor NK cells, would then be either deleted by donor NK killing or suppressed by tumor growth factor (TGF)-b, produced by the activated NK cells. It has been shown that NK cells could have veto activity against cytotoxic T cell precursors, which may suppress the alloreactivity from the host bone marrow [9]. ‘‘Veto activity’’ was defined in 1980 by Miller as the capacity to specifically suppress cytotoxic T cell (CTL) precursors directed against antigens of the veto cells themselves, but not against third party antigens [10-13]. It has been shown that the suppression by the veto activity is mediated by apoptosis [14,15]. Apoptosis can be induced by cytotoxic granule release (granzyme and perforin mediated killing), Fas-FasL mediated killing, or release of apoptosis inducing cytokines, such as TGF-b. NK cells are capable of inducing target cell apoptosis using all these primary killing mechanisms. It would be of interest to study whether donor NK cells might suppress the host alloreactive cells utilizing any of these mechanisms. In this article, we further investigated the affect of allogeneic NK cells as suppressors of allo-responses using an in vitro mixed lymphocyte response (MLR) model system. We then demonstrated the mechanisms NK cells using to suppress alloreactive CTL responses. Finally, using a nonmyeloablative allogeneic BMT animal model, we showed that donor NK cells can promote engraftment by suppressing host alloreactive responses. The results suggested that activated donor NK cells could specifically suppress the host reactive cells using multiple mechanisms, and they provide a promising way to promote donor engraftment during nonmyeloablative allogeneic BMT.

MATERIALS AND METHODS Animals Female BALB/c (H2d), C57BL/6 (B6, H2b), and C3H (H2k) mice were purchased from the Animal Production Area of the National Cancer Institute (Frederick, MD) and SLAC Animal Laboratory (Shanghai, China). gld mice and pfp mice were purchased from Jackson Laboratory (Bar Harbor, ME). Animals were kept in specific pathogen-free conditions. All animal protocols were approved at each of the 2 animal facilities (University of Nevada, Reno, and Soochow

Allogeneic NK Cells as Suppressors of Allo-responses

773

University). Mice were between 8 and 12 weeks of age at the start of the experiments. Cell Lines and Reagents P815 (H2d) murine lymphoblast-like mastocytoma cell line and EL4 (H2b) murine T cell lymphoma cell line were maintained in complete RPMI media. AntiThy1.2 mAb (30H12) and rabbit-complement were purchased from Cedarlane Laboratories (Burlington, Ontario, Canada). All the other antibodies were purchased from Becton-Dickinson and Company (BD, Mountain View, CA). CFSE was purchased from Molecular Probes (Eugene, OR). Cell Preparation Bone marrow cell (BMC) suspensions were prepared by gently releasing cells from the backbones, femurs, and tibiae into Dulbecco’s phosphatebuffered saline (PBS) solution with a mortar and pestle, filtering through a mesh filter to remove particulates, and washing the cell suspensions twice. Spleen cells were prepared by gently crushing the tissues to release the cells. Preparations were filtered to remove debris and washed twice in PBS before resuspending in RF10 complete media (RPMI 1640 supplemented with 10% FBS (Gemini Bio-Products, West Sacramento, CA); 100 U/mL penicillin/streptomycin, 2 mM L-glutamine, 10 mM HEPES [N-2-hydroxyethylpiperazine-N-2-ethanesulfonic acid], 1 mM nonessential amino acids, 1 mM sodium pyruvate, and 2.5 1025M2-ME). Cell counts were performed on a Coulter Z1 cell counter (Coulter Electronics, Hialeah, FL). Generation of ALAK Cells The ALAK cells were generated according to the method described before [16]. Briefly, splenocytes and BMCs from C57BL/6 or BALB/c mice were depleted of T cells by treatment with anti-Thy1.2 mAb (30H12) and rabbit-complement (Cedarlane Laboratories). T cell-depleted samples were cultured in RF10 complete medium (RPMI 1640 supplemented with 10% FBS (Gemini Bio-Products); 100 U/mL penicillin/streptomycin, 2 mM L-glutamine, 10 mM HEPES [N-2-hydroxyethylpiperazine-N-2-ethanesulfonic acid], 1 mM nonessential amino acids, 1 mM sodium pyruvate, and 2.5 1025 M2-ME containing 1000 IU/mL recombinant human IL-2 (rhIL-2; Developmental Therapeutics Program, NCI, Bethesda, MD) at 1 to 2 106 cells/mL for 5 to 7 days at 37 C, 5% CO2. At day 3 or 4, nonadherent cells were transferred to new flasks, and all cells were fed with 50% conditioned medium, 50% new RF10 complete medium, and rhIL-2. Only the adherent cells were used for subsequent experiments.

774

B. Hu et al.


(CFSE; Molecular Probes, Eugene, OR) and added into MLR reactions. After 3 days of culture, cells were stained with anti-H2Kb PE and washed in icecold PBS containing BSA (0.1%) and azide (0.01%) and analyzed on a BD FACScan or Calibur using CellQuest software (Becton-Dickinson).

Mixed Lymphocyte Response To prepare responder C57BL/6 spleen cells, spleen cells were made into single-cell suspension and red blood cells were lysed. Stimulator cells were prepared from single-cell suspensions of spleens from BALB/c or C3H mice and irradiated (30 Gy). Responders and stimulators were cultured at a final concentration of 5 106/mL, pulsed with tritiated thymidine (1 mCi/well) (Amersham Life Sciences, Buckinghamshire, UK) 16 to 18 hours prior to harvesting and counted on a b-plate reader (Packard Instruments, Meriden, CT). Four individual wells were analyzed per data point.

Animal Model of Nonmyeloablative BMT Female C57BL/6 recipient mice received nonmyeloablative doses (800–850 cGy) of total body irradiation (TBI) from a 137Cesium source. Irradiation was followed by the infusion of 1.0 or 1.5 106 BALB/c BMCs i.v. with or without ALAK cells from BALB/c mice (5 106 cells i.v.). Mice received gentamycin sulfate oral suspension (Ryen Pharma Co., China) in their drinking water beginning 7 days before the transplant. Mice were monitored and weighed weekly. The experiments were performed 3 times with 5 to 7 mice per group. Two months after BMT, flow cytometry analysis was performed to assess the donor engraftment by labeling the splenocytes with anti-H2kd FITC and anti-H2kb PE antibodies (Becton-Dickinson). The alloreactivity against donor type antigens and killing capacity to donor type tumor targets were also measured by MLR or cytotoxicity assay.

Cytotoxicity Assay Alloreactive cytotoxic T cell activity was determined by using a chromium-release assay with responding cells after a 72-hour MLR reaction or splenocytes 2 months after BMT. P815 or EL4 cells were labeled with 51 Cr (300 mCi) at 37 C for 1 hour. They were then washed and incubated at 104/well at different effector: target ratios for 4 hours at 37 C. Supernatants were then harvested for gamma-counting. Maximum release (M) was determined by adding Triton X-100 (1%) to the targets. Spontaneous release (S) was determined from targets cultured in the absence of effector cells. The level of specific chromium release was calculated as: % specific release 5 (counts – S)/(M – S), where counts are the experimental chromium release in the presence of effector cells.

RESULT

CFSE Labeling and Flow Cytometry

H2d ALAK Cells Can Significantly Suppress the Proliferation and the Killing Capacity of H2b Splenocytes in Allogeneic Primary MLR

In vitro activated NK cells were labeled with 1-2.5 mM carboxyfluorescein diacetate succinimidyl ester

To determine if allogeneic ALAK cells can suppress the host allo-responses, we set up an in vitro

B

Balb/C ALAK

2 5 0 00

Balb/C Splenocytes

2 0 0 00

CPM

1 5 0 00 1 0 0 00 5 0 00

Percent Killing

Balb/C ALAK

60

50

50

40 30 20 10

0 10

E5

5 2. 5x

0E x1

1. 25

10

E5

0

1.2x10E5 2.5x10E5 5x10E5

40

Media control

30 20 10

0

0. 6x

Balb/C Splenocytes

60

Percent Killing

A

0 5 0 :1

1 7 :1

E:T Ratio

6 :1

5 0 :1

1 7 :1

6:1

E:T Ratio

ALAK cell number Figure 1. H2d ALAK cells from BALB/c mice significantly suppress the (A) proliferation and (B) killing capacity of H2b splenocytes from C57BL/6 mice in allogeneic MLR stimulated with irradiated H2d splenocytes from BALB/C mice. Allogeneic MLR was set up with 5 105 splenocytes from both C57BL/6 and BALB/c mice. The ALAK cell numbers were as indicated in the figure. The same number of BALB/c splenocytes was also added as controls. *Indicates P \.01.The data shown are the representative of four experiments.


A

H2d ALAK

9 00 00

H2b ALAK

H2d

20

H2b anti-H2d response ALAK H2b ALAK 20

6 00 00 5 00 00 4 00 00

10 5

4X10E5

15

Percent Killing

Percent Killing

7 00 00

1x10E5 2X10E5

15

8 00 00

CPM

B

H2b anti-H2d response

10 00 00

775


Media control

10 5 0

0

3 00 00 2 00 00

-5

E5

5 0 :1

4x 10

E5 2x 10

E5 1x 10

0

-5 16 : 1

5 0 :1

5 :1

1 6 :1

5 :1

E:T Ratio

E:T Ratio

ALAK cell number H2b anti-H2d response

H2b anti-H2k response 150000

100000

100000

50000

50000

0

0

H2b anti-H2d response Media Control H2d ALAK H2k ALAK

Media Control 25 ALAK 1.25x10E5

CPM

150000

D

Percent Killing

C

20 15 10 5 0

H2d ALAK

H2k ALAK

-5

H2k ALAK

5 0:1

1 6:1

5:1

Figure 2. Only the ALAK cells of stimulator origin can effectively suppress the allo-responses in the allogeneic MLR. The same number of H2d ALAK cells and H2b ALAK cells were added to the H2b anti-H2d allogeneic MLR same as in Figure 1. The (A) proliferation and (B) killing capacity of H2b splenocytes from C57BL/6 mice were measured at the end of the 3-day culture. (C) 1.25 105 H2d or H2k ALAK cells were added to H2b anti-H2d MLR. The same number of H2k ALAK cells was also added to H2b anti-H2k MLR. The proliferation of H2b splenocytes was measured at the end of the 3-day culture. (D) The killing capacity of H2b splenocytes was also measured at the end of the 3-day culture. *Indicates P\.01. The data shown are the representative of 3 experiments.

MLR system mimicking the allo-responses during the allogeneic bone marrow transplantation. Allogeneic NK cells of the stimulator origin were added into the MLR reactions. After a 3-day culture, the proliferation capacity of H2b splenocytes (responders) from C57BL/6 mice was measured with 3H-thymidine assay. H2d ALAK cells significantly suppressed the proliferation of H2b splenocytes from C57BL/6 mice in allogeneic MLR stimulated with irradiated H2d splenocytes from BALB/c mice (Figure 1A). The extent of the suppressive effect was dependent on the number of allogeneic ALAK cells added. The same number of BALB/c splenocytes was also added as controls, and they did not show any suppressive effect, which suggested that the suppressive effect was caused by NK cell activities, not just the increased number of cells of H2d origin. The allogeneic activities of the H2b responders were also measured by their killing capacity of H2d tumor targets (P815). The H2b responders were harvested after a 3-day MLR reaction mentioned above and cocultured with 51Cr-labeled tumor targets at different effector:target ratios for 4 hours. The killing capacity was measured by 51Cr release in the

supernatants. Adding H2d ALAK cells during the MLR reaction significantly suppressed the allogeneic killing capacity of the H2b responders (Figure 1B). The suppressive effect was also dose dependent. The same number of BALB/c splenocytes did not show any suppression but slightly increased the killing activity of the responders (Figure 1B). To determine whether the suppressive effect of the allogeneic ALAK cells is dependent on their alloreactivity, the same number of H2d ALAK cells (allogeneic) and H2b ALAK cells (syngeneic) were added to the allogeneic MLR. The proliferation and killing capacity of H2b responders were measured at the end of the 3-day culture using the method mentioned above (Figure 2). Comparing to the allogeneic ALAK cells, the syngeneic ALAK cells had no suppressive effect on the allogeneic responses. The third-party H2k ALAK cells from C3H mice did not suppress the anti-H2d MLR either, whereas they could suppress the anti-H2k responses of the H2b responders (Figure 2C). The H2k ALAK cells could slightly suppress the killing capacity of H2b responders in an anti-H2d MLR only at a 50:1 ratio, but the suppressive

776

B. Hu et al.


B

A

C

Figure 3. The suppressive effect of allogeneic ALAK cells was not dependent on TGF-b secretion. Anti-TGF-b and the isotype control antibodies (35 mg/mL) were added to the allogeneic MLR described in Figure 1. (A) The levels of TGF-beta secreted in the MLR reaction with or without anti-TGF-beta antibody were measured by ELISA. ND indicates not detected. (B) The proliferation of H2b splenocytes from C57BL/6 mice stimulated with irradiated H2d splenocytes were measured at the end of the 3-day culture by 3H-thymidine incorporation assay. (C) The killing capacity of H2b splenocytes killing H2d tumor cells (P815) were measured at the end of the 3-day culture by cytotoxicity assay. *Indicates P\.01. #Indicates P\.05. The data shown are the representative of 3 experiments.

effect was a lot less comparing to H2d ALAK cells. Therefore, only the ALAK cells of stimulator (donor) origin could efficiently suppress the allogeneic responses of the responders (host). The Suppressive Effect of Allogeneic ALAK Cells Is Not Dependent on TGF-b Secretion To determine whether the suppressive effects of the allogeneic ALAK cells were caused by the release of TGF-b, anti-TGF-b blocking antibody and the isotype control were added to the allogeneic MLR described before (Figure 3). The production of TGF-b could be detected in the assay and adding allogeneic ALAK cells caused a slight increase, whereas the anti-TGF-b blocking antibody significantly reduced it to the level that could not be detected (Figure 3A). The proliferation and killing capacity of alloreactive cells were measured at the end of the 3-day culture (Figure 3B and C). The suppressive effects of the allogeneic ALAK cells were not affected by blocking the TGF-b activities, which suggested that the suppressive effect of allogeneic NK cells was not mediated by the TGF-b secretion.

The Increased Number of H2d ALAK Cells Can Cause the Diminishing of the H2b CTLs in the Allogeneic MLR The allogeneic NK cells could suppress the function of the alloreactive CTLs by direct killing. It would be of interest to determine whether adding allogeneic ALAK cells could cause the diminishing of the alloreactive CTLs in the allogeneic MLR. The CFSElabeled or unlabeled H2d ALAK cells were added to the allogeneic MLR described before. It was shown that the capacity of H2d ALAK cells to inhibit H2b splenocytes killing H2d tumor target was not affected by CFSE labeling (Figure 4A). Therefore, CFSE labeling could be used to identify the ALAK cells from the alloreactive CTLs besides using MHC class I molecules as markers in the flow cytometry assay. The number of CFSE2 H2b1PI2 cells was calculated as the indication of live H2b CTLs left in the MLR culture, and it was significantly decreased by adding CFSElabeled H2d ALAK cells (Figure 4B). However, adding the CFSE-labeled H2d splenocytes did not significantly change the number of CFSE2 H2b1PI2 cells


A


CFSE labeled

Unlabeled

30

1.25x10E5

30

2.5x10E5

25

Percent Killing

25

20 15

15

10

10 5

0

0

50

Number of H2b+CFSE- Cells (x103)

5.0x10E5

Media Control

20

5

B

777

1

16

1

5

1

50

60

60

50

50

40

40

30

30

20

20

10

10

1

16

1

5

1

0

0 0

1.25x10E5

2.5x10E5

5x10E5

0

1.25x10E5

d

H2 ALAK cell number d

2.5x10E5

5x10E5

d

H2 splenocyte number b

Figure 4. The increased number of H2 ALAK cells could cause the diminishing of the H2 CTLs in the allogeneic MLR. The CFSE-labeled or unlabeled H2d ALAK cells were added to the allogeneic MLR described in Figure 1. (A) The capacity of H2d ALAK cells to inhibit H2b splenocytes killing H2d tumor target was not affected by CFSE labeling. (B) The number of live H2b CTLs left in the culture was significantly decreased by adding H2d ALAK cells. *Indicates P \.01. The data shown are the representative of 3 experiments.

in the culture. The data indicated that the mechanism for allogeneic NK cells to suppress the allo-responses could be direct killing of the alloreactive CTLs. The Suppressive Effect of Allogeneic ALAK Cells Is Partially Dependent on Perforin-Mediated Killing Mechanism To further study the killing mechanisms allogeneic NK cells might be using to suppress the alloreactive CTLs, the ALAK cells from wild-type C57BL/6 mice, gld mice (FasL mutant), and pfp mice (perforin KO, both on C57BL/6 background) were used in the MLR assay (Figure 5). The proliferation and killing capacity of the alloreactive responders were measured at the end of the 3-day culture. The ALAK cells from the gld or pfp mice did not affect the proliferation suppression of the alloreactive cells by the allogeneic ALAK cells (Figure 5A). However, the ALAK cells from pfp mice did not suppress the killing capacity of the alloreactive CTLs as well as the ALAK cells from the wild-type mice or the gld mice (P \ .05, Figure 5B). The pfp ALAK cells were less effective than gld and wild-type ALAK cells at both 50:1 and

16:1 E:T ratio with addition of 2.5 105 or 1.25 105 ALAK cells (P \ .05). It suggested that perforinmediated killing mechanism could be involved in the direct killing of alloreactive CTLs by allogeneic NK cells, but not the only mechanism during the process. Donor ALAK Cells Promote Donor Engraftment in Nonmyeloablative Allogeneic BMT To demonstrate that donor ALAK cells could promote donor engraftment during nonmyeloablative allogeneic BMT by suppressing the host alloreactive responses, the ALAK cells from BALB/c mice were infused into host C57BL/6 mice during nonmyeloablative allogeneic BMT. Flow analysis demonstrated that donor ALAK cell infusion (H2d NK, middle panel) promoted donor engraftment dramatically from 0.26% to 50.60% (Figure 6A). Syngeneic ALAK cells (H2b NK, right panel) only slightly increased donor engraftment to 2.01%. A summary of the results from each experiment was shown in Figure 6B. With 5 to 7 mice per group, all 3 experiments showed similar results that donor ALAK cell

778

B. Hu et al.


110000

A

WT

90000

GLD

CPM

70000

Perforin KO

50000 30000 10000 -10000 0 1. 2

B

WT ALAK

E5

1 . 5x

0E

5

1 5x

0E

5

2 ALAK cell number Gld ALAK

NK 5x 10E5

Pfp ALAK

NK 2. 5x 10E5

65

Percent Killing

5

0 x1

NK 1.25x10E5

55

Media Control

45

65

65 55

55 45

45 35 25 15

35 25 15

35 25 15 5

5 -5

5 -5 50:1

16:1

5:1

-5 50:1

E:T Ratio

16:1

5:1

E:T Ratio

50:1

16: 1

5: 1

E:T Ratio

Figure 5. The suppressive effect of allogeneic ALAK cells was partially dependent on perforin-mediated killing mechanism. The ALAK cells from wild type C57BL/6 mice, gld mice, and pfp mice (both on C57BL/6 background) were used in the MLR assay. The (A) proliferation and (B) killing capacity of H2d splenocytes from BALB/c mice were measured at the end of the 3-day culture. *Indicates P \.01. The data shown are the representative of 2 experiments.

infusion dramatically promoted donor engraftment during nonmyeloablative allogeneic BMT. We further assessed the host alloreactive responses after BMT by studying the proliferative responses to donor type antigens and the killing capacity to donor type tumor targets (Figure 7B and C). Two months after BMT and ALAK cell infusion, the host derived CD41 and CD81 T cells were measured by flow cytometry (Figure 7A). Most of the T cells were host derived (H2b1) without donor ALAK cell infusion. Host T cell percentage was reduced to around 60% when donor ALAK cells were infused (H2d NK). Donorderived ALAK cell infusion during BMT significantly suppressed the proliferation and IL-2 secretion in response to alloantigen and the killing capacity of the donor type tumor targets were also significantly reduced. The alloreactivity to donor type antigen was reduced to 25% to 30% with donor ALAK cell infusion (H2d NK) compared to BMT alone and syngeneic ALAK cell infusion (H2b NK). With about 60% host T cells remaining, the 25% to 30% alloreactivity was not only because of less host T cells present, but also the reduced alloreactivity of each cells. Therefore, donor ALAK cells could promote engraftment by

suppressing host alloreactive responses in a nonmyeloablative BMT model.

DISCUSSION In this article, we demonstrated that allogeneic ALAK cells could suppress the host allo-responses in an in vitro MLR system. The suppressive effects are not only manifested by reduced proliferation, but diminished killing capacity of the allo-responders. These data reconcile with the early observations of engraftment promotions by donor NK cells in an allogeneic BMT. They provide insights to promoting engraftment in not only nonmyeloablative BMT, but also solid organ transplantation. The results further elucidated the mechanism donor NK cells might be using to suppress the host allo-responses. The inhibitory effect of TGF-b on cytotoxic T cells is well demonstrated before [17]. Our data showing that anti-TGF-b antibody treatment did not affect the suppression of allo-responses by allogeneic NK cells suggest that TGF-b may not be the soluble factor released by NK cells to mediate their suppressive


BMT+H2dNK

BMT

779

BMT+H2bNK

H2Kb PE

A


H2Kd FITC Summary of all the experiments

B Exp No.

Group

No. of Animals

% Engraftment

Exp.1

BMT

5

0.55 0.35

H2 NK+BMT

7

49.73

2.91

H2bN K+BMT

5

1.78

0.28

BMT

5

0.51 0.28

H2 NK+BMT

6

54.60 4.05

H2bNK+BMT

6

1.73

d

Exp.2 d

Exp.3

0.33

BMT

5

0.21 0.02

H2d NK+BMT

6

53.76 2.60

H2b N K+BMT

6

1.66 0.32

Figure 6. Donor ALAK cells promoted donor engraftment in nonmyeloblative allogeneic BMT. The ALAK cells from BALB/c mice (H2d NK) were infused into host C57BL/6 mice during nonmyeloblative allogeneic BMT. The ALAK cells from syngeneic C57BL/6 mice (H2b NK) were also infused as control. (A) Flow analysis of the percent of donor engraftment 2 months after BMT. (B) Summary and statistical analysis of all the experiments.

effects. There might also be direct killing of the alloresponders by allogeneic NK cells. Therefore, we examined the suppressive effects of the NK cells from gld and pfp mice. The allogeneic ALAK cells from gld mice suppressed the allo-responses as well as the wild-type ALAK cells, whereas the ALAK cells from pfp mice had reduced suppressive effects compared with the wild-type ALAK cells. Therefore, the data suggest that the allogeneic NK cells might be using perforin-mediated killing mechanism to suppress the allo-responses, but it is not the only mechanism for their suppressive functions. It has been well demonstrated that allogeneic NK cells could potentially enhance the engraftment of the allografts [7,18,19]. This could be because of the veto effect of donor NK cells toward the host alloreactive CTLs [9]. Although the veto effects on CTLs could also be generated by other immune cells, activated NK cells have been shown to be the dominant veto cells shown by a TCR transgenic 2C mouse model [20]. Activated CD41 T cells may also exert veto effect on CTLs, but activated B cells, nonactivated CD41, or NK cells are completely devoid of veto activities. In a recent study, transferring of

allogeneic cytokine-induced killer (CIK) cells resulted in minimal graft-versus-host disease (GVHD) with retention of antitumor activity [21]. In that case, CIK cells may exert cytotoxic activities on both syngeneic and allogeneic targets. In the current studies, we examined the NK cell activities using the ex vivo IL-2activated ALAK cells. They are over 98% NK cells and should perform as in vitro activated NK cells. We showed that the donor NK cells could suppress the host allo-responses. In the in vitro experiments, third-party ALAK cells could not suppress the allogeneic MLR as well as the ALAK cells of stimulator origin, suggesting that the alloreactive CTLs could specifically recognize the ALAK cells of stimulator origin, and then be eliminated by the ALAK cells through veto mechanisms. ALAK cells could also be activated by the NKG2D ligands expressed on the alloreactive CTLs as well as the loss of inhibitory signals by the absence of self MHC class I molecules on the alloreactive CTLs. The fact that third-party ALAK cells could also suppress the killing capacity of the responders suggests that multiple mechanisms might be involved (ie, NKG2D). More experiments are needed to demonstrate whether this suppressive

780

B. Hu et al.


Figure 7. Donor ALAK cells could suppress host alloreactive responses in nonmyeloablative allogeneic BMT. The ALAK cells from BALB/c mice (H2d NK) were infused into host C57BL/6 mice during nonmyeloablative allogeneic BMT. The ALAK cells from syngeneic C57BL/6 mice (H2b NK) were also infused as control. (A) Percent of host derived CD41 and CD81 T cells (H2b T cells) out of total T cells 2 months after BMT. The host (B) proliferative responses and IL-2 production responding to donor type antigens (irradiated BALB/c splenocytes) and (C) the killing capacity to donor type tumor targets (P815 cells) were measured 2 months after BMT. Base control was the readout without alloantigen stimulation. *Indicates P \.01. The data shown are the representative of 3 experiments.

effect is because of the veto activity of the donor NK cells in vivo. Although TGF-b was shown not a critical cytokine for the suppressive effect of the allogeneic NK cells, there might be other cytokines that could engage suppressive effect on host allo-activities, such as IL-10. It has been suggested that a subset of IL-10-secreting NK cells can suppress the antigen-specific T cell responses [22]. Its suppressive effect is dependent on IL-10 secretion. Whether allogeneic NK cells are also predominantly IL-10-secreting cells and use the same mechanism to suppress allo-responses need further investigation. NK cells can use a series of proteins to induce the death of the target cells. Perforin is a critical component in the cytotoxic granules to initiate apoptosis pathway [16]. The interaction between Fas ligand expressed by killer cells and the Fas receptor on the target cell membrane was initially regarded as the main mechanism of granule-independent killing mechanism by CTL and NK cells [16,23]. In the current studies, we investigated whether allogeneic NK cells could be using one of these two killing mechanisms to suppress host alloreactive CTLs. The results indicated that granule-mediated killing might be part of the mechanisms for allo-response

suppression. However, apoptosis of target cells could also be induced by several other members of the tumor necrosis factor (TNF) ligand superfamily [24-26], such as TNF-a, TRAIL, and TWEAK. TRAIL is widely expressed on the surface of activated T and B cells, NK cells, dendritic cells, and monocytes [27-29]. It interacts with death receptor DR4 and DR5 [24,30,31] and cause TRAIL-induced apoptosis. Therefore, the granule-mediated killing pathway may not be the only mechanism for allogeneic NK cell suppression. Our data also suggested that other suppressive mechanisms may exist during the inhibition of allo-responses. The relevance of these mechanisms in the nonmyeloablative BMT model would need to be examined with further experiments using functionally defective NK cells. In conclusion, we demonstrate that activated allogeneic NK cells can specifically suppress the alloreactive cells, inhibiting both their proliferation and killing capacities. Granule-mediated killing might be part of the suppressive mechanism, whereas the Fas-FasL pathway and TGF-b signaling are not involved. These specific suppressive effects of allogeneic NK cells provide a promising way to promote donor engraftment without involving systemic and nonspecific suppression of the immune system.



ACKNOWLEDGMENTS Financial disclosure: This work has been supported in part by the PhD Program Funds from the Chinese Ministry of Education, the startup funds, and the Key Project in Science & Technology Cultivation Program from Soochow University.

15.

16.

17.

REFERENCES 1. Blume KG, Forman SJ, Appelbaum FR. Hematopoietic Cell Transplantation, 3rd ed. Malden, MA: Blackwell Science; 2004. 2. Copelan EA. Hematopoietic stem cell transplantation. N Engl J Med. 2006;354:1813-1826. 3. Yuan D, Wilder J, Dang T, Bennett M, Kumar V. Activation of B lymphocytes by NK cells. Int Immunol. 1992;4:1373-1380. 4. Ortaldo JR, Mason LH, Gregorio TA, Stoll J, WinklerPickett RT. The Ly-49 family: regulation of cytokine production in murine NK cells. J Leukoc Biol. 1997;62:381-388. 5. Murphy WJ, Kumar V, Bennett M. Acute rejection of murine bone marrow allografts by natural killer cells and T cells. Differences in kinetics and target antigens recognized. J Exp Med. 1987;166:1499-1509. 6. Ruggeri L, Capanni M, Urbani E, et al. Effectiveness of donor natural killer cell alloreactivity in mismatched hematopoietic transplants. Science. 2002;295:2097-2100. 7. Murphy WJ, Bennett M, Kumar V, Longo DL. Donor-type activated natural killer cells promote marrow engraftment and B cell development during allogeneic bone marrow transplantation. J Immunol. 1992;148:2953-2960. 8. Asai O, Longo DL, Tian ZG, et al. Suppression of graft-versushost disease and amplification of graft-versus-tumor effects by activated natural killer cells after allogeneic bone marrow transplantation. J Clin Invest. 1998;101:1835-1842. 9. Azuma E, Yamamoto H, Kaplan J. Use of lymphokine-activated killer cells to prevent bone marrow graft rejection and lethal graft-vs-host disease. J Immunol. 1989;143:1524-1529. 10. Miller RG. An immunological suppressor cell inactivating cytotoxic T-lymphocyte precursor cells recognizing it. Nature. 1980; 287:544-546. 11. Miller RG. Lymphoid suppressor cells. In: Sercarz EE, Cunningham AJ, editors. Strategies of Immune Regulation. New York: Academic Press; 1980 p. 507. 12. Miller RG, Derry H. A cell population in nu/nu spleen can prevent generation of cytotoxic lymphocytes by normal spleen cells against self antigens of the nu/nu spleen. J Immunol. 1979;122: 1502-1509. 13. Muraoka S, Miller RG. Cells in bone marrow and in T cell colonies grown from bone marrow can suppress generation of cytotoxic T lymphocytes directed against their self antigens. J Exp Med. 1980;152:54-71. 14. Reich-Zeliger S, Gan J, Bachar-Lustig E, Reisner Y. Tolerance induction by veto CTLs in the TCR transgenic 2C mouse

18.

19.

20.

21.

22.

23.

24. 25. 26.

27.

28.

29.

30.

31.

781

model. II. Deletion of effector cells by Fas-Fas ligand apoptosis. J Immunol. 2004;173:6660-6666. Hiruma K, Nakamura H, Henkart PA, Gress RE. Clonal deletion of postthymic T cells: veto cells kill precursor cytotoxic T lymphocytes. J Exp Med. 1992;175:863-868. Ka¨gi D, Vignaux F, Ledermann B, et al. Fas and perforin pathways as major mechanisms of T cell-mediated cytotoxicity. Science. 1994;265:528-530. Pardoux C, Asselin-Paturel C, Chehimi J, Gay F, MamiChouaib F, Chouaib S. Functional interaction between TGFbeta and IL-12 in human primary allogeneic cytotoxicity and proliferative response. J Immunol. 1997;158:136-143. Velardi A, Ruggeri L, Alessandro, Moretta, Moretta L. NK cells: a lesson from mismatched hematopoietic transplantation. Trends Immunol. 2002;23:438-444. Ruggeri L, Capanni M, Casucci M, et al. Role of natural killer cell alloreactivity in HLA-mismatched hematopoietic stem cell transplantation. Blood. 1999;94:333-339. Reich-Zeliger S, Bachar-Lustig E, Gan J, Reisner Y. Tolerance induction by veto CTLs in the TCR transgenic 2C mouse model. I. Relative reactivity of different veto cells. J Immunol. 2004;173:6654-6659. Nishimura R, Baker J, Beilhack A, et al. In vivo trafficking and survival of cytokine-induced killer cells resulting in minimal GVHD with retention of antitumor activity. Blood. 2008;112: 2563-2574. Deniz G, Erten G, Ku¨cu¨ksezer UC, et al. Regulatory NK cells suppress antigen-specific T cell responses. J Immunol. 2008; 180:850-857. Lowin B, Hahne M, Mattmann C, Tschopp J. Cytolytic T-cell cytotoxicity is mediated through perforin and Fas lytic pathways. Nature. 1994;370:650-652. Nagata S. Apoptosis by death factor. Cell. 1997;88:355-365. Lincz LF. Deciphering the apoptotic pathway: all roads lead to death. Immunol Cell Biol. 1998;76:1-19. Griffith TS, Lynch DH. TRAIL: a molecule with multiple receptors and control mechanisms. Curr Opin Immunol. 1998;10: 559-563. Mariani SM, Krammer PH. Surface expression of TRAIL/Apo2 ligand in activated mouse T and B cells. Eur J Immunol. 1998; 28:1492-1498. Zamai L, Ahmad M, Bennett IM, Azzoni L, Alnemri ES, Perussia B. Natural killer (NK) cell-mediated cytotoxicity: differential use of TRAIL and Fas ligand by immature and mature primary human NK cells. J Exp Med. 1998;188:2375-2380. Griffith TS, Wiley SR, Kubin MZ, Sedger LM, Maliszewski CR, Fanger NA. Monocyte-mediated tumoricidal activity via the tumor necrosis factor-related cytokine, TRAIL. J Exp Med. 1999;189:1343-1354. Kwon B, Youn BS, Kwon BS. Functions of newly identified members of the tumor necrosis factor receptor/ligand superfamilies in lymphocytes. Curr Opin Immunol. 1999;11: 340-345. Golstein P. Cell death: TRAIL and its receptors. Curr Biol. 1997;7:R750-R753.