The unique profile of cord blood natural killer cells balances incomplete maturation and effective killing function upon activation

Human Immunology 73 (2012) 248-257

Contents lists available at SciVerse ScienceDirect

The unique profile of cord blood natural killer cells balances incomplete maturation and effective killing function upon activation Martha Luevano a,†, Mehri Daryouzeh a,†, Rehab Alnabhan a, Sergio Querol a,b, Salim Khakoo c, Alejandro Madrigal a,*, Aurore Saudemont a a

Anthony Nolan Research Institute and UCL, Royal Free Campus, Pond Street, London NW3 2QG, UK Programa Concordia Banc de Sang i Teixits, Barcelona, Spain c Department of Medicine, Imperial College, London SW7 2AZ, UK b

A R T I C L E

I N F O

Article history: Received 13 May 2011 Accepted 21 December 2011 Available online 28 December 2011

Keywords: Natural killer cells Cord blood Immunotherapy Hematopoietic stem cell transplantation

A B S T R A C T

Cord blood (CB) is increasingly used as a source of stem cells for hematopoietic stem cell transplantation, and natural killer (NK) cells may be the effectors of the antileukemic response observed after CB transplantation. Here, we analyzed the phenotype and functions of CB NK cell subsets. We determined that the percentage of NK cells was higher in CB compared with peripheral blood (PB). Furthermore, there was a higher percentage of the CD56bright subset in CB. CB NK cells reached a late stage of differentiation, but exhibited higher expression of NKG2A and expressed fewer killer-cell immunoglobulin-like receptors, suggesting an incomplete maturation. CB NK cells highly expressed CXCR4, but did not express L-selectin, highlighting unique homing properties of CB NK cells. CB NK cells proliferated in response to interleukin-2 and degranulated in response to stimulation with tumor cells, but failed to lyse K562 cells in 51Cr-release assay. CB NK cells exhibited a lower interferon-␥ production in comparison with PB NK cells. Culture with IL-2 increased CB NK cell functions. Our study sheds light on CB NK cell properties and highlights the potential of CB as a source of NK cells for immunotherapy. 䉷 2012 American Society for Histocompatibility and Immunogenetics. Published by Elsevier Inc. All rights reserved.

1. Introduction Hematopoietic stem cell transplantation is most commonly indicated for leukemia, but is also used to treat different hematological and nonhematological malignancies, immunodeficiency, and autoimmunity [1]. Cord blood (CB) has been increasingly used as an alternative source of stem cells for hematopoietic stem cell transplantation in adults because it presents several advantages including faster availability, lower human leukocyte antigen matching requirement, lower incidence and severity of graft-versus-host disease [2– 4], and preservation of the graft versus leukemia effect described after allogeneic bone marrow transplantation. The use of intense immunosuppression after CB transplantation (CBT) in conjunction with the naivety of the T cells infused with the graft and the slow T cell reconstitution generate a long period of immunodeficiency after CBT. Natural killer (NK) cells constitute between 15 and 30% of CB lymphocytes [5,6], and after CBT, NK cell reconstitution occurs very early on, so NK cells constitute most of

* Corresponding author. E-mail address: [email protected] (A. Madrigal). † These authors contributed equally to this work.

the lymphocytes in circulation. The observation that they are capable of killing leukemia cells ex vivo [7] supports the hypothesis that NK cells are most likely to be a main effector of the graft-versusleukemia effect observed after CBT. It has been reported that the phenotype of CB NK cells demonstrates some similarities to peripheral blood (PB) NK cells [8]. CB CD56⫹ cells have been demonstrated to express CD94, some killercell immunoglobulin-like receptors (KIR), and activating receptors such as NKG2D and NKp46 [6,9]. Nonetheless, the current literature relating to the phenotype and functions of CB NK cells indicated some inconsistencies [8]. CB NK cells have been reported to be mature by some investigators [6,8,10] and immature by others [11]. In some circumstances, they have been demonstrated to be as cytolytic as PB NK cells [6,10] or deficient in killing target cells, potentially because of an immature phenotype [11], a high expression of NKG2A [9], or a low expression of adhesion molecules [12]. However, incubation with interleukin-2 (IL-2), IL-12, or IL-15 can increase their cytolytic activity [6,11]. In addition to the CD56dim and CD56bright subsets, CB has been demonstrated to be characterized by the presence of CD16⫹CD56⫺ cells, which are not typically present in healthy PB. Previous studies have reported that these cells are a unique cell subset present in CB [10,11,13] that exhibited reduced lytic activity and might be the precursor of mature NK cells. Tanaka et al. reported an increased

0198-8859/$36.00 䉷 2012 American Society for Histocompatibility and Immunogenetics. Published by Elsevier Inc. All rights reserved. doi:10.1016/j.humimm.2011.12.015

M. Luevano et al. / Human Immunology 73 (2012) 248-257

number of CD16⫹CD56⫺ NK cells in PB after CBT [12]. This NK cell subset was also reported in immunocompromised hosts with viral infections such as human immunodeficiency virus and hepatitis [14 –16]. All these studies focused on CD56⫹ or CD56⫺ cells, and the characteristics of CB CD56dim and CD56bright NK cells especially before infusion into patients must still be explored. Although a few studies focused on the functions and phenotype of CB CD56⫹ cells, none has addressed the question of their developmental stages and homing properties. In this study, we investigated the characteristics of CB CD56dim and CD56bright NK cells in term of expression of adhesion molecules, inhibitory and activating receptors, maturation markers and chemokine receptors, cytotoxic capacity, and interferon-␥ (IFN-␥) production in comparison with PB NK cells. Overall, our data indicate that CB NK cells are differentiated, but not fully mature, and have different homing properties than PB NK cells; our data also indicate that IL-2 treatment can complete the maturation of CB NK cells and therefore increase their effector functions. 2. Subjects and methods 2.1. Collection of blood samples Peripheral blood mononuclear cells were isolated from the blood of healthy volunteers upon informed consent. CB samples were obtained from the Anthony Nolan Cord Blood Bank (Nottingham, UK) or from the Programa Concordia Banc de Sang i Teixits (Barcelona, Spain). Samples were collected using routine banking procedures into a cord blood donation bag containing a citrate– phosphate– dextrose anticoagulant buffer and processed within 24 hours of collection. 2.2. Purification of CB and PB NK cells Mononuclear cells were isolated by density-gradient centrifugation using Ficoll–Paque Premium (GE Healthcare Bio-sciences, Uppsala, Sweden). Purified NK cells were obtained by negative selection using the NK cell isolation kit II (Miltenyi Biotec, Bergisch Gladbach, Germany) according to the manufacturer’s instructions. 2.3. Flow cytometry analysis The following monoclonal antibodies were purchased from BD Pharmingen (San Diego, CA): anti-CD3 (HIT3a and SK7), anti-CD14 (M5E2), anti-CD16 (NKp15), anti-CD19 (ID3), anti-CD34 (581), antiCD45 (HI30), anti-CD48 (BCM1), anti-CD56 (clone B159), antiCD62L (Dreg 56), CD94 (HP-3D9), CD107a (H4A3), anti-CD117 (104D2), anti-CD122 (Mik-B3), anti-CD127 (L112), anti-CD158a (HP-3E4), anti-CD158b (CH-L), anti-CD226 (DX11), anti-FAS-L (NOK-1), anti-LFA-1 (HI111), anti-NKp30 (p30-15), anti-NKp46 (9E2/Nkp46), and anti-TRAIL (RIK-2). Anti-CCR6 (53103), antiCXCR1 (42705), anti-CXCR4 (12G5), anti-CXCR7 (358426), and anti-NKG2C (34591) were purchased from R&D Systems (Minneapolis, MN); anti-NKG2A (z199) from Beckman Coulter (Marseille, France); and anti-NKG2D (BAR 221) from Miltenyi Biotec. AntiCD25 (BC96), anti-CD49d (9F10), and anti-integrin ␤7 (FIB504) were purchased from eBiosciences (San Diego, CA). For intracellular staining, 106 mononuclear cells were incubated at 37⬚C for 5 hours in medium with or without phorbol myristate acetate (PMA; 100 ng/mL) and ionomycin Ca2⫹ ionophore (1 ␮g/mL; all reagents were from Sigma Chemical Co.–Aldrich (St. Louis, MO). Two microliters of GolgiStop (BD Biosciences) was added during the last 4 hours of incubation. Thereafter, cells were surface stained with anti-CD3, anti-CD16, and anti-CD56 antibodies permeabilized using the Fix/ Perm cell permeabilization kit (BD Biosciences) for 5 minutes at room temperature and then stained for IFN-␥ (BD Biosciences). The cells were then washed and analyzed on a FACSCalibur instrument (BD Biosciences). When expression of perforin (dG9, BD Biosci-

249

ences) and granzyme B (GB11, BD Biosciences) was analyzed, the intracellular staining was performed directly without prior stimulation. 2.4. Proliferation analysis Purified NK cells were resuspended in phosphate-buffered saline (PBS) at 107 cells/mL and labeled with 2 ␮M carboxyfluorescein diacetate succinimidyl ester (Molecular Probes, Invitrogen, Carlsbad, CA) for 10 minutes in the dark at 37⬚C. Cells were then washed twice with complete culture medium and cultured for 7 days in the presence of 200 or 1,000 UI IL-2/mL. 2.5. Degranulation assay NK cells (2 ⫻ 105) were washed twice in PBS and added to 2 ⫻ 105 target cells in 200 ␮L complete medium and incubated for 2 hours at 37C⬚C. The cells were spun down and stained for CD56, CD3, and CD16 in PBS supplemented with 2% fetal bovine serum and 2 mM EDTA for 45 minutes on ice. The cells were then washed, resuspended in PBS containing 2% fetal bovine serum and 2 mM EDTA, and analyzed by flow cytometry. 2.6. Cytotoxicity assay For cytotoxicity assay, 51Cr-release assay was performed. K562 were used as target cells and pulsed with 100 ␮Ci Na2[51Cr]O4 for 45 minutes (PerkinElmer, Cambridge, UK). Freshly isolated NK cells or NK cells cultured for 7 days with 1,000 UI IL-2/mL were added to the target cells at effector-to-target ratios of 1:1, 5:1, and 10:1 in triplicate. 51Cr-release was assessed in the supernatant of each culture after 4 hours. The maximum release was determined by lysis of target cells with 2% Triton X-100. The percentage of specific lysis was calculated as (experimental release – spontaneous release)/(maximum release – spontaneous release) ⫻ 100. 2.7. Statistics Statistical comparisons were performed with GraphPad Prism software (GraphPad Software, San Diego, CA) using the nonparametric Mann–Whitney test or, when indicated, the unpaired t test or the F test to compare variances. Results are presented as means ⫾ SD. p values ⬍0.05 were considered statistically significant. 3. Results 3.1. CB NK cells are characterized by a heterogeneous CD56 expression The analysis of the percentage of CD56⫹CD3⫺ cells in CB and PB lymphocytes indicated that, as previously reported [5], CB contained more NK cells (18.21 ⫾ 1.146%) compared with adult PB (12.69 ⫾ 1.637%; p ⫽ 0.0161; Fig. 1A). The markers CD56 and CD16 define distinct PB NK cell subsets [17,18]; CD56bright NK cells are predominantly cytokine producers, whereas CD56dim NK cells mediate natural and antibody-dependent cellular cytotoxicity (ADCC). Expression of CD16 and CD56 was analyzed by flow cytometry in purified NK cells. The percentage of CD56bright NK cells was slightly higher in CB than in PB (p ⫽ 0.0135; Fig. 1B). CD56 expression was more heterogeneous in CB NK cells, and the mean fluorescence intensity (MFI) for CD56 was lower in CB NK cells than in PB NK cells (p ⬍ 0.0001; Fig. 1C). 3.2. CB NK cells are differentiated NK cells The stages of human NK cell development have recently been defined by Freud et al. [19,20], who proposed a developmental pathway from a CD34⫹ NK cell progenitor to a functionally mature NK cell defined by a different expression level of CD34, CD117, and CD94. To determine the developmental stage of CB NK cells, CB and PB NK cells were stained for CD16, CD56, CD94, and CD117. Based

250

M. Luevano et al. / Human Immunology 73 (2012) 248-257

difference was observed for NK cell stages 3 and 4 between CB and PB. Studies in vitro have reported that CD117 expression defines 2 differentiating NK cell subpopulations: CD56⫹CD117high and CD56⫹CD117low/⫺ [21]. The presence of these 2 subsets in CD56⫹ CB and PB NK cells, as well as the relative expression of CD94 in these 2 subsets, was further investigated. We determined that most NK cells were CD56⫹CD117low/⫺ in PB and CB (Fig. 2C) and that CD56⫹CD117⫹ cells were present in PB and in CB. No significant difference was observed for these 2 populations between CB and PB. For each subset, CD94 expression was then examined. We determined that PB CD56⫹CD117⫺ and CD56⫹CD117⫹ expressed more CD94 than their CB counterparts (Mann–Whitney p ⫽ 0.03, t test p ⫽ 0.0545; and Mann–Whitney p ⫽ 0.02, t test p ⫽ 0.049, respectively; Fig. 2C). Next, CD16 versus CD94 expression was assessed within isolated CB and PB NK cells because the progression from CD94⫹CD16⫺ to CD94⫹/⫺CD16⫹ might mark the terminal step of differentiation of human NK cells (Fig. 2D). PB CD56dim NK cells, which are CD94⫹/⫺CD16⫹, therefore represent stage 5 of human NK cell development and are fully mature NK cells. We observed that the CD94⫹CD16⫺ subset was higher in CB NK cells than in PB NK cells (Mann–Whitney p ⫽ 0.022, t test p ⬍ 0.001; Fig. 2E). The percentage of CD94⫹CD16⫹ cells as well as the percentage of CD94⫺CD16⫹ cells was lower in CB than in PB (Mann–Whitney p ⫽ 0.0152, t test p ⫽ 0.029; and Mann–Whitney p ⫽ 0.0087, t test p ⫽ 0.0347, respectively). Taken together, these results suggest that CB NK cells are mature and belong to developmental stages 4 and 5; however, it appears that fewer CB NK cells reach the final step of differentiation in comparison with PB NK cells. 3.3. CB NK cell maturation is incomplete

Fig. 1. Natural killer (NK) cell subsets in cord blood (CB). (A) Percentages of NK cells among lymphocytes in peripheral blood (PB; n ⫽ 13) and CB (n ⫽ 15). (B) Percentages of CD56dim (circles) and CD56bright (squares) in isolated NK cells in PB (white; n ⫽ 15) and CB (black; n ⫽ 25). (C) CD56 mean fluorescence intensity (MFI) in PB (white; n ⫽ 15) and CB (black; n ⫽ 25). *p ⬍ 0.05, ***p ⬍ 0.0001.

on the relative expression of these markers, 3 subsets can be identified: CD117⫹CD94⫺ stage 3 immature NK cells, a transitional CD117⫹CD94⫹ stage, and CD117⫺CD94⫹ stage 4 mature NK cells (Fig. 2A). Most PB and CB CD56⫹ cells were CD117⫺CD94⫹ stage 4 (Fig. 2B), whereas the percentage of cells belonging to the CD117⫹CD94⫺ stage 3 was low for both PB and CB. No significant

To be functional, NK cells must acquire a set of receptors that will complete their maturation and regulate their activity. Hence, we analyzed the expression of several inhibitory and activating receptors on PB and CB NK cell subsets. The percentage of CD56dim NK cells expressing the inhibitory KIR receptors CD158a and CD158b was lower in CB than in PB (CD158a, t test p ⫽ 0.0086 and F test p ⫽ 0.0122; CD158b, t test p ⫽ 0.0006 and F test p ⫽ 0.1242, respectively; Fig. 3A). PB CD56bright NK cells expressed a low level of these KIR receptors, and their expression was lower on CB CD56bright NK cells (CD158a, p ⫽ 0.0189 and F test p ⫽ 0.001; CD158b, p ⫽ 0.0189 and F test p ⫽ 0.001, respectively; Fig. 3B). The proportion of CD56dim and CD56bright NK cells expressing another inhibitory receptor, NKG2A, was higher in CB than in PB NK cells (CD56dim, p ⬍ 0.0001; CD56bright, p ⫽ 0.0013; Fig. 3A and B). We then investigated the expression of several activating receptors involved in the killing of tumor cells and infected cells. Most PB and CB CD56dim and CD56bright NK cells expressed NKG2D and NKp30 (Fig. 3C and D). The proportion of DNAM-1 or NKG2C expressing NK cells was lower in CB CD56dim NK cells (p ⫽ 0.0063 for DNAM-1 and p ⫽ 0.0003 for NKG2C) and CB CD56bright NK cells (p ⫽ 0.0207 for DNAM-1 and p ⫽ 0.0181 for NKG2C). NKp46 was expressed on PB and CB CD56dim NK cells, but we determined NKp46 expression to be lower on CB CD56bright than on PB CD56bright NK cells (p ⫽ 0.0019). All PB and CB CD56dim and CD56bright NK cells expressed 2B4 and CD48, the only known ligand for 2B4 [22] (data not shown); however, the MFI for 2B4 was significantly higher in CB NK cells than in PB NK cells (p ⬍ 0.0001 for CD56dim NK cells; p ⬍ 0.0001 for CD56bright NK cells; Fig. 3E), and the MFI for CD48 was also higher in CB CD56bright NK cells than in PB CD56bright NK cells (p ⫽ 0.0196; Fig. 3F). Thus, there are substantial differences between CB and PB NK cells, and the fact that CB NK cell subsets express lower level of KIRs as well as some activating receptors, but

M. Luevano et al. / Human Immunology 73 (2012) 248-257

A

PB

251

B

CB

CD117

100

0.98

1.22

2.97

80

1.36

60

54.3

40

58.29

20 0

CD94

Stage 3

PB

C

Stage 4 Stage 4 (transitional)

CB

79.21

78.1

6.89

CD56

5.81

CD117

71

98.5

51.4

71.2

CD94 PB 104

104

103

103

102

101

100 100

101

102

CD16

E

CB

CD94

CD94

D

103

104

*

*

*

CD94+CD16+

CD94-CD16+

102

101

100 100

102

101

103

104

CD16 CD94+CD16-

Fig. 2. Developmental stages of cord blood (CB) natural killer (NK) cells. (A) Representative analysis of CD117 and CD94 expression in isolated peripheral blood (PB) and CB NK cells. (B) Percentages of developmental stages 3 and 4 defined by CD117 and CD94 expression in isolated PB (white; n ⫽ 9) and CB (black; n ⫽ 9) NK cells. (C) Representative analysis of CD56 and CD117 expression in purified NK cells isolated from PB (n ⫽ 9) and CB (n ⫽ 9). For each population, CD94 expression was then examined. (D) Representative analysis of CD94 and CD16 expression in PB and CB NK cells. (E) Percentages of different maturation stages defined by CD16 and CD94 expression in purified NK cells isolated from PB (white; n ⫽ 6) and CB (black; n ⫽ 6). *p ⬍ 0.05.

highly expressed NKG2A, is consistent with an incomplete maturation of CB NK cell subsets. 3.4. CB NK cell homing signature differs from that of PB NK cells Because mature and immature lymphocytes may traffic to distinct sites, we examined the expression on PB and CB NK cell

subsets of receptors involved in extravasation (CD49d) [23], homing to the bone marrow (CXCR4 and CXCR7) [24 –26], trafficking to lymph nodes (L-selectin) [27], gut (␣4␤7 integrin) [29,30], or inflammatory sites (CXCR1) [31] (Fig. 4). CXCR1 was more expressed on PB CD56dim than on CB CD56dim NK cells (p ⫽ 0.047), whereas CXCR1 expression did not differ between PB and CB CD56bright NK

252

M. Luevano et al. / Human Immunology 73 (2012) 248-257

Fig. 3. Expression of activating and inhibitory receptors by cord blood (CB) natural killer (NK) cell subsets. The expression of the inhibitory receptors CD158a, CD158b, and NKG2A by peripheral blood (PB) and CB CD56dim (A) and CD56bright (B) NK cells was analyzed by flow cytometry. Expression of the activating receptors DNAM-1, NKG2C, NKG2D, NKp30, and NKp46 by PB and CB CD56dim (C) and CD56bright (D). PB samples are represented in white and CB samples in black. The mean fluorescence intensity for 2B4 (E) and for CD48 (F) was evaluated for CD56dim (gray) and CD56bright (white) isolated from PB and CB. *p ⬍ 0.05.

cells. Interestingly, the bone marrow homing receptor CXCR4 was upregulated on both CB CD56dim NK cells and CD56bright NK cells compared with PB NK cells (p ⫽ 0.0197 and p ⫽ 0.0376, respectively). No difference in expression of CXCR7, another bone marrow homing receptor, between PB and CB CD56dim NK cells and PB and CB CD56bright NK cells was observed. We determined that all PB and CB NK cells expressed CD49d but did not express ␤7integrin. The vast majority of PB CD56bright NK cells and a smaller population of PB CD56dim NK cells expressed L-selectin, whereas neither CB CD56bright NK cells (p ⬍ 0.0001) nor CD56dim expressed a high level of L-selectin (p ⬍ 0.0001). Taken together, these results suggest that CB NK cells are more likely to home to the bone marrow rather than lymph nodes. 3.5. CB NK cells can degranulate and produce IFN-␥ in the presence of tumor cells NK cell expansion can be achieved in vitro and in vivo by exposure to IL-2. To analyze the capacity of CB NK cells to respond and

proliferate in response to IL-2, freshly isolated CB or PB NK cells were labeled with carboxyfluorescein diacetate succinimidyl ester and cultured in the presence of different concentrations of IL-2. PB NK cells were capable of responding to 200 UI IL-2, whereas CB NK cells exhibited low proliferation and/or underwent apoptosis at that concentration (data not shown). Therefore, we assessed the response of CB NK cells to a higher concentration of IL-2. When 1,000 UI IL-2 was used, we observed that PB NK cells responded to IL-2 and completely divided after 2 days of treatment (Fig. 5A). As illustrated in Fig. 5A, CB NK cells could respond to IL-2. Most CB NK cells proliferated after 2 days of exposure to 1,000 UI IL-2; however, they went through much fewer divisions than PB NK cells and did not completely divide after 7 days of culture with IL-2. The higher IL-2 requirement, as well as the lower cell cycles observed when CB NK cells are stimulated with high-concentration IL-2, suggests a certain degree of immaturity of CB NK cells. Because PB and CB NK cells exhibited different proliferative response after exposure to IL-2, we next evaluated the expression

M. Luevano et al. / Human Immunology 73 (2012) 248-257

253

Fig. 4. Expression of chemokine receptors and integrins by cord blood (CB) natural killer (NK) cell subsets. The expression of the chemokine receptors CXCR1, CXCR4, and CXCR7 was analyzed by flow cytometry in peripheral blood (PB) and CB CD56dim NK cells (A) and CD56bright (B) (N ⫽ 4 –7 for PB and CB); similarly, the expression of CD49d, integrin ␤7, and L-selectin was assessed in PB and CB CD56dim NK cells (C) and CD56bright (D) (N ⫽ 8 –9 for PB and N ⫽ 6 –11 for CB). PB samples are represented in white and CB samples in black. *p ⬍ 0.05, ***p ⬍ 0.0001.

of CD25 (IL-2R␣) and CD122 (IL-2R␤) in PB and CB NK cell subsets. As illustrated in Fig. 5B, CB CD56dim NK cells expressed a lower level of CD25 than PB CD56dim NK cells (p ⬍ 0.0001); all PB and CB CD56bright NK cells expressed CD25 and were CD122⫹. Previous studies report CB NK cells to be poor killers or as cytolytic as PB NK cells. To shed light on the capacity of CB NK cells to kill targets, we evaluated NK cell degranulation by analysis of CD107a expression at the cell surface of PB and CB NK cell subsets after PMA/ionomycin stimulation or incubation with K562 in a 1:1 ratio. In response to PMA/ionomycin stimulation, we observed that both CB CD56bright and CD56dim NK cells degranulated less efficiently than PB NK cells (p ⫽ 0.011 and p ⫽ 0.033, respectively; Fig. 6A). When NK cells were challenged with K562, the potential of CD56bright and CD56dim to degranulate was similar between CB and PB. After 7 days in culture with 1,000 UI IL-2, CD56 expression on PB and CB NK cells was uniform, and no NK cell subsets could be distinguished. Therefore, CD107a expression was then only investigated in CD56⫹ NK cells. Degranulation of CB and PB NK cells was identical when NK cells were stimulated with PMA/ionomycin or K562 (Fig. 6B). Although CB NK cells exhibited less degranulation when stimulated with PMA/ionomycin, they were capable of degranulating in the presence of tumor cells and after stimulation with IL-2 as efficiently as PB NK cells. Killing of K562 by purified PB and CB NK cells at days 0 and 7 of culture with IL-2 was then assessed by 51Cr-release assay. PB NK cells were able to lyse K562, whereas CB NK cells were not able to kill K562 cells at all (Fig. 6C). Culture with IL-2 enhanced killing of K562 by PB and CB NK cells, but no significant difference of killing of K562 was observed between IL-2-activated PB and CB NK cells (Fig. 6D). CB NK cells were able to mediate ADCC, but no significant difference was observed between PB and CB NK cells (Fig. 6E).

PB NK cells can kill targets via the granzyme/perforin pathway or via activation of cell death receptors. To evaluate whether these molecules were expressed in CB NK cells, we then assessed FAS-L and TRAIL cell surface expression as well as granzyme B and perforin intracellular expression in PB and CB NK cell subsets by flow cytometry. CB CD56dim NK cells had a lower expression of granzyme B (p ⫽ 0.0211) as well as perforin (p ⫽ 0.0144) and FAS-L (p ⫽ 0.0048) than PB CD56dim NK cells (Table 1). CD56bright NK cells expressed FAS-L and a low level of granzyme B and perforin, and no difference was observed between PB and CB. Both PB and CB NK cell subsets expressed TRAIL to a similar level (Table 1). After 7 days in culture with 1,000 UI IL-2, no difference in expression of granzyme B, perforin, FAS-L, and TRAIL was observed between PB and CB NK cells (Table 1). These results suggest that CB CD56dim NK cells express molecules associated with cytotoxicity at a lower level than PB NK cells; however, the expression level of these molecules was equivalent in CB and PB NK cells after stimulation with IL-2. Finally, we assessed the capacity of CB NK cells to secrete IFN-␥ in response to different stimuli. CB CD56dim NK cells produce significantly less IFN-␥ than PB CD56dim cells in response to PMA/ ionomycin (p ⫽ 0.0244; Fig. 7A). There was also a nonsignificant trend for CB CD56bright NK cells to produce less IFN-␥ than PB CD56bright NK cells, but no difference was observed between PB and CB in numbers of IFN-␥-producing cells when coincubated with K562 and after 7 days of culture with IL-2 (Fig. 7B), suggesting that IL-2 stimulation increases the ability of CB NK cells to produce IFN-␥. 4. Discussion The purpose of this study was to compare the phenotype and functions of CB NK cell subsets with that of their counterparts in PB.

254

M. Luevano et al. / Human Immunology 73 (2012) 248-257

Fig. 5. Proliferation of freshly isolated cord blood (CB) natural killer (NK) cells in response to interleukin-2 (IL-2). (A) Freshly isolated NK cells from peripheral blood (PB; n ⫽ 4) or CB (n ⫽ 4) were labeled with carboxyfluorescein diacetate succinimidyl ester (CFSE) and cultured in the presence of 1,000 UI IL-2 for 7 days and compared with the CFSE staining from day 0 (open histogram). Representative histograms of 1 experiment for different time point are presented. The expression of CD25 and CD122 by PB and CB CD56dim (B) and CD56bright NK cells (C) was analyzed by flow cytometry. PB samples are represented in white (n ⫽ 5) and CB (n ⫽ 8) samples in black. ***p ⬍ 0.0001.

We determined that the major NK cell subsets, based on the relative expression of CD16 and CD56, were similar between CB and PB, although CB contains a slightly higher percentage of CD56bright NK cells than PB. CB has been demonstrated to contain a high percentage of NK cells [32] as well as CD16⫹CD56⫺ cells. The latest cells have been demonstrated to exhibit a low level of cytotoxicity and to be possible precursors of mature NK cells [33]. However, our analysis indicated that CD56 expression on CB NK cells is heterogeneous and lower on CB NK cells than on PB NK cells, which makes the identification of CD56⫹ and CD56⫺ cells difficult and may lead to the inclusion of CD56low cells in the CD16⫹CD56⫺ subset and, by consequence, may lead to a higher percentage of this subset. Based on the proposed model by Freud et al. [19,20], we demonstrated that most CB NK cells are CD94⫹CD16⫹CD117⫺ and therefore belong to a late stage of development and could be considered mature NK cells. To better define their degree of maturation, the repertoire of CB NK cells was analyzed and compared with that of PB NK cells. We demonstrated that CB NK cell subsets express fewer KIRs, NKG2C,

NKp46, and DNAM-1 compared with PB NK cells and are characterized by a high expression of NKG2A. We concluded that although CB NK cells belong to late stages of NK cell development, their maturation seems to be incomplete. Our results also suggest that CB NK cells might be disarmed [34,35] or unlicensed [36]. It has been demonstrated that NK cells could acquire a licensed phenotype when transferred into a major histocompatibility complex (MHC) class I– sufficient environment [37], exposed to cytokines such as IL-18, IL-12, or PMA/ionomycin [38,39] or to an altered MHC environment [40]. Therefore, it is possible that CB NK cells could become licensed and fully functional in patients after CBT. All CB NK cells express the receptor 2B4 and its ligand CD48. During NK cell development, 2B4 can display an inhibitory function, which controls the autoaggression mediated by immature NK cells that have acquired cytotoxicity activity through earlier acquisition of natural cytotoxicity receptors [41,42]. A higher expression of 2B4 and its ligand may control CB NK cell cytotoxicity along with other inhibitory receptors.

M. Luevano et al. / Human Immunology 73 (2012) 248-257

255

Fig. 6. Cytolytic activity of cord blood (CB) natural killer (NK) cells. (A) CD107a expression on the surface of peripheral blood (PB) and CB CD56dim and CD56bright NK cells was analyzed by flow cytometry with or without stimulus (phorbol myristate acetate [PMA]/ionomycin or incubation with K562) when freshly isolated (n ⫽ 7 for PB and CB) or (B) after 7 days in culture with 1,000 UI interleukin-2 (IL-2; n ⫽ 6 for PB and n ⫽ 5 for CB). Lysis of K562 by freshly isolated (C) or IL-2-activated PB and CB NK cells (D) and antibody-dependent cellular cytotoxicity (ADCC) killing by PB and CB NK cells (E) was assessed in a standard 51Cr-release assay (n ⫽ 4 for PB and CB). E:T, effector-to-target ratio. NS, non-stimulated.

We observed a very low killing of K562 cells by CB NK cells. This could be explained by the low expression of molecules involved in cytotoxicity such as granzyme B and perforin by CB NK cells. Even if CB NK cells were capable of degranulating as efficiently as PB NK cells in response to stimulation with K562 cells, this would result in poorer killing of the cells. These results are in accordance with

some of the previous studies indicating that CB NK cells have a lower cytotoxic function against typically used cell lines such as K562 and Daudi [6,43]. However, we observed that CB NK cells are as responsive as PB NK cells after exposure to IL-2. CB NK cells were able to degranulate and kill K562 as efficiently as PB NK cells and expressed similar levels of granzyme B and perforin as PB NK cells.

Table 1 Intracellular expression of perforin and granzyme B (%) and surface expression of FAS-L and TRAIL (%) analyzed by flow cytometry using isolated peripheral blood (PB) and cord blood (CB) CD56dim and CD56bright natural killer cells at days 0 and 7 of culture with 1,000 UI IL-2 (n ⫽ 6 for PB and n ⫽ 9 for CB) PB

Perforin Granzyme B FAS-L TRAIL

CB

PB

CD56dim

CD56bright

CD56dim

CD56bright

CD56⫹

CB CD56⫹

78.66 ⫾ 22.5 84.23 ⫾ 12.6 71.27 ⫾ 18.1 35.33 ⫾ 11.4

49.52 ⫾ 38.6 19.74 ⫾ 9.2 41.6 ⫾ 20.57 42.31 ⫾ 24.6

30.29 ⫾ 36.9 54.36 ⫾ 25.1 33.73 ⫾ 19.7 38.93 ⫾ 7.76

23.74 ⫾ 35.6 20.26 ⫾ 11.7 51.18 ⫾ 22.1 42.44 ⫾ 22.4

36.58 ⫾ 18 84.28 ⫾ 17.7 37.99 ⫾ 4.29 39.64 ⫾ 5.63

29.87 ⫾ 23.4 74.3 ⫾ 15.99 40.34 ⫾ 23.6 41.56 ⫾ 14.7

256

M. Luevano et al. / Human Immunology 73 (2012) 248-257

Fig. 7. (A) Percentages of interferon-␥ (IFN-␥)-producing cells in peripheral blood (PB; n ⫽ 5) and cord blood (CB; n ⫽ 5) CD56bright and CD56dim natural killer (NK) cells at day 0 and (B) day 7 of culture with 1,000 UI interleukin-2 (PB n ⫽ 6 and CB n ⫽ 8) with or without stimulus (phorbol myristate acetate [PMA]/ionomycin or incubation with K562). *p ⬍ 0.05. NS, non-stimulated.

Interestingly, we also demonstrated that CB NK cells could mediate ADCC to a similar level as PB NK cells. CB NK cells produced less IFN-␥ when stimulated with PMA/ ionomycin but not when stimulated with K562 cells in comparison with PB NK cells. When CB NK cells were cultured with IL-2, an increased IFN-␥ production was observed that reached a level equivalent to the production observed with PB NK cells. Because NK cells are key producers of IFN-␥ during the first stage of the antiviral immune response, these results might suggest that the low IFN-␥ production of CB NK cells might also contribute to the increased number of infections observed after CBT. To shed light on the migration properties of CB NK cells, the expression of several chemokine receptors and adhesion molecules involved in NK cell migration and trafficking was analyzed. CB NK cells were observed to express a high level of CXCR4 compared with PB NK cells, suggesting that CB NK cells could have a greater potential to home to the bone marrow than PB NK cells. Furthermore, L-selectin expression was also assessed and no expression was observed on CB NK cells, confirming CB NK cell immaturity and suggesting that CB NK cells might be unable to traffic to secondary lymphoid tissues. These results suggest that CB NK cell homing properties differ from that of PB NK cells; however, it should be kept in mind that CB NK cells might change their migration pattern under the influence of a new environment. In summary, we demonstrated that although CB NK cells might not be fully mature, they do express some of the receptors involved in cytotoxicity. CB NK cell degranulation and IFN-␥ production were decreased after PMA/ionomcyin stimulation. CB NK cells were able to degranulate in the presence of K562, but were not able to kill K562 cells. It should be noted that culture with IL-2 increased their effector functions to levels observed for IL-2-activated PB NK cells.

Our results suggest that CB NK cells could be important key players in the outcome of CBT and also underline the fact that they could also be considered a new source of NK cells for immunotherapy28. Acknowledgments This work was funded by Anthony Nolan and by a Marie Curie fellowship. We thank Michelle Escobedo and Sameer Tulpule for critical reading of the manuscript, James Devitt and Richard Duggleby for their helpful flow cytometry technical expertise, and Hazel Forde for technical help. References [1] Copelan EA. Hematopoietic stem-cell transplantation. N Engl J Med 2006;354: 1813–26. [2] Shlomchik WD. Graft-versus-host disease. Nat Rev Immunol 2007;7:340 –52. [3] Rocha V, Wagner JE Jr, Sobocinski KA, Klein JP, Zhang MJ, Horowitz MM, et al. Graft-versus-host disease in children who have received a cord-blood or bone marrow transplant from an HLA-identical sibling. Eurocord and International Bone Marrow Transplant Registry Working Committee on Alternative Donor and Stem Cell Sources. N Engl J Med 2000;342:1846 –54. [4] Barker JN, Wagner JE. Umbilical cord blood transplantation: current practice and future innovations. Crit Rev Oncol/Hematol 2003;48:35– 43. [5] Kotylo PK, Baenzinger J, Yoder MC, Engle WA, Bolinger CD. Rapid analysis of lymphocyte subsets in cord blood. Am J Clin Pathol 1990;93:263– 6. [6] Dalle JH, Menezes J, Wagner E, Blagdon M, Champagne J, Champagne MA, et al. Characterization of cord blood natural killer cells: implications for transplantation and neonatal infections. Pediatr Res 2005;57:649 –55. [7] Beziat V, Nguyen S, Lapusan S, Hervier B, Dhedin N, Bories D, et al. Fully functional NK cells after unrelated cord blood transplantation. Leukemia 2009; 23:721– 8. [8] Verneris MR, Miller JS. The phenotypic and functional characteristics of umbilical cord blood and peripheral blood natural killer cells. Br J Haematol 2009; 147:185–91. [9] Wang Y, Xu H, Zheng X, Wei H, Sun R, Tian Z. High expression of NKG2A/CD94 and low expression of granzyme B are associated with reduced cord blood NK cell activity. Cell Mol Immunol 2007;4:377– 82.

M. Luevano et al. / Human Immunology 73 (2012) 248-257

[10] Phillips JH, Hori T, Nagler A, Bhat N, Spits H, Lanier LL. Ontogeny of human natural killer (NK) cells: fetal NK cells mediate cytolytic function and express cytoplasmic CD3 epsilon,delta proteins. J Exp Med 1992;175:1055– 65. [11] Gaddy J, Risdon G, Broxmeyer HE. Cord blood natural killer cells are functionally and phenotypically immature but readily respond to interleukin-2 and interleukin-12. J Interferon Cytokine Res 1995;15:527–36. [12] Tanaka H, Kai S, Yamaguchi M, Misawa M, Fujimori Y, Yamamoto M, et al. Analysis of natural killer (NK) cell activity and adhesion molecules on NK cells from umbilical cord blood. Eur J Haematol 2003;71:29 –38. [13] Bradstock KF, Luxford C, Grimsley PG. Functional and phenotypic assessment of neonatal human leucocytes expressing natural killer cell-associated antigens. Immunol Cell Biol 1993;71:535– 42. [14] Gonzalez VD, Falconer K, MichaÌlsson J, Moll M, Reichard O, Alaeus A, et al. Expansion of CD56⫺ NK cells in chronic HCV/HIV-1 co-infection: reversion by antiviral treatment with pegylated IFNalpha and ribavirin. Clin Immunol 2008; 128:46 –56. [15] Mavilio D, Lombardo G, Benjamin J, Kim D, Follman D, Marcenaro E, et al. Characterization of CD56⫺/CD16⫹ natural killer (NK) cells: a highly dysfunctional NK subset expanded in HIV-infected viremic individuals. Proc Natl Acad Sci U S A 2005;102:2886 –91. [16] Zarife MA, Reis EA, Carmo TM, Lopes GB, BrandÄo EC, Silva HR, et al. Increased frequency of CD56Bright NK-cells, CD3-CD16⫹CD56⫺ NK-cells and activated CD4⫹T-cells or B-cells in parallel with CD4⫹CDC25High T-cells control potentially viremia in blood donors with HCV. J Med Virol 2009;81:49 –59. [17] Cooper MA, Fehniger TA, Caligiuri MA. The biology of human natural killer-cell subsets. Trends Immunol 2001;22:633– 40. [18] Cooper MA, Fehniger TA, Turner SC, Chen KS, Ghaheri BA, Ghayur T, et al. Human natural killer cells: a unique innate immunoregulatory role for the CD56(bright) subset. Blood 2001;97:3146 –51. [19] Freud AG, Caligiuri MA. Human natural killer cell development. Immunol Rev 2006;214:56 –72. [20] Freud 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– 43. [21] Grzywacz B, Kataria N, Sikora M, Oostendorp RA, Dzierzak EA, Blazar BR, et al. Coordinated acquisition of inhibitory and activating receptors and functional properties by developing human natural killer cells. Blood 2006;108:3824 –33. [22] Brown MH, Boles K, van der Merwe PA, Kumar V, Mathew PA, Barclay AN. 2B4, the natural killer and T cell immunoglobulin superfamily surface protein, is a ligand for CD48. J Exp Med 1998;188:2083–90. [23] Berlin C, Bargatze R, Campbell JJ, von Andrian UH, Szabo MC, Hasslen SR, et al. Alpha 4 integrins mediate lymphocyte attachment and rolling under physiologic flow. Cell 1995;80:413–22. [24] Beider K, Nagler A, Wald O, Franitza S, Dagan-Berger M, Wald H, et al. Involvement of CXCR4 and IL-2 in the homing and retention of human NK and NK T cells to the bone marrow and spleen of NOD/SCID mice. Blood 2003;102: 1951– 8. [25] Peled A, Petit I, Kollet O, Magid M, Ponomaryov T, Byk T, et al. Dependence of human stem cell engraftment and repopulation of NOD/SCID mice on CXCR4. Science 1999;283:845– 8. [26] Hartmann TN, Grabovsky V, Pasvolsky R, Shulman Z, Buss EC, Spiegel A, et al. A crosstalk between intracellular CXCR7 and CXCR4 involved in rapid CXCL12-

[27]

[28]

[29]

[30]

[31] [32]

[33]

[34] [35]

[36] [37]

[38]

[39] [40]

[41]

[42]

[43]

257

triggered integrin activation but not in chemokine-triggered motility of human T lymphocytes and CD34⫹ cells. J Leukoc Biol 2008;84:1130 – 40. Frey M, Packianathan NB, Fehniger TA, Ross ME, Wang WC, Stewart CC, et al. Differential expression and function of L-selectin on CD56bright and CD56dim natural killer cell subsets. J Immunol 1998;161:400 – 8. Palmer LA, Sale GE, Balogun JI, Li D, Jones D, Molldrem JJ, et al. Chemokine receptor CCR5 mediates alloimmune responses in graft-versus-host disease. Biol Blood Marrow Transplant 2010;16:311–9. Hamann A, Andrew DP, Jablonski-Westrich D, Holzmann B, Butcher EC. Role of alpha 4-integrins in lymphocyte homing to mucosal tissues in vivo. J Immunol 1994;152:3282–93. Kellersmann R, Lazarovits A, Grant D, Garcia B, Chan B, Kellersmann A, et al. Monoclonal antibody against beta7 integrins, but not beta7 deficiency, attenuates intestinal allograft rejection in mice. Transplantation 2002;74:1327–34. Maghazachi AA. Role of chemokines in the biology of natural killer cells. Curr Top Microbiol Immunol 2010;341:37–58. Theilgaard-M×nch K, Raaschou-Jensen K, Palm H, Schjødt K, Heilmann C, Vindeløv L, et al. Flow cytometric assessment of lymphocyte subsets, lymphoid progenitors, and hematopoietic stem cells in allogeneic stem cell grafts. Bone Marrow Transplant 2001;28:1073– 82. Gaddy J, Broxmeyer HE. Cord blood CD16⫹56- cells with low lytic activity are possible precursors of mature natural killer cells. Cell Immunol 1997; 180:132– 42. Gasser S, Raulet DH. Activation and self-tolerance of natural killer cells. Immunol Rev 2006;214:130 – 42. Joncker NT, Raulet DH. Regulation of NK cell responsiveness to achieve selftolerance and maximal responses to diseased target cells. Immunol Rev 2008; 224:85–97. Jonsson AH, Yokohama WM. Natural killer cell tolerance licensing and other mechanisms. Adv Immunol 2009;101:27–79. Elliott JM, Wahle JA, Yokoyama WM. MHC class I-deficient natural killer cells acquire a licensed phenotype after transfer into an MHC class I-sufficient environment. J Exp Med 2010;207:2073–9. Kim S, Poursine-Laurent J, Truscott SM, Lybarger L, Song YJ, Yang L, et al. Licensing of natural killer cells by host major histocompatibility complex class I molecules. Nature 2005;436:709 –13. Yokoyama WM, Kim S. Licensing of natural killer cells by self-major histocompatibility complex class I. Immunol Rev 2006;214:143–54. Joncker NT, Shifrin N, Delebecque F, Raulet DH. Mature natural killer cells reset their responsiveness when exposed to an altered MHC environment. J Exp Med 2010;207:2065–72. Moretta L, Bottino C, Pende D, Mingari MC, Biassoni R, Moretta A. Human natural killer cells: their origin, receptors and function. Eur J Immunol 2002; 32:1205–11. Vacca P, Pietra G, Falco M, Romeo E, Bottino C, Bellora F, et al. Analysis of natural killer cells isolated from human decidua: evidence that 2B4 (CD244) functions as an inhibitory receptor and blocks NK-cell function. Blood 2006; 108:4078 – 85. Qian JX, Lee SM, Suen Y, Knoppel E, van de Ven C, Cairo MS. Decreased interleukin-15 from activated cord versus adult peripheral blood mononuclear cells and the effect of interleukin-15 in upregulating antitumor immune activity and cytokine production in cord blood. Blood 1997;90:3106 –17.