Application of CD27 as a marker for distinguishing human NK cell subsets

Transcription

1 International Immunology, Vol. 20, No. 4, pp doi: /intimm/dxn022 ª The Japanese Society for Immunology All rights reserved. For permissions, please Application of CD27 as a marker for distinguishing human NK cell subsets Anabel Silva 1, Daniel M. Andrews 1, Andrew G. Brooks 2, Mark J. Smyth 1 and Yoshihiro Hayakawa 1,3 1 Cancer Immunology Program, Trescowthick Laboratories, Peter MacCallum Cancer Centre, St Andrews Place, East Melbourne, Victoria 3002, Australia 2 Department of Microbiology and Immunology, University of Melbourne, Parkville, Victoria 3010, Australia 3 Present address: Pharmacology Department, Tsukuba Research Institute, Banyu Pharmaceutical Co. Ltd, 3 Okubo, Tsukuba, Ibaraki , Japan Keywords: human, NK cell, subset Abstract It has long been recognized that human NK cells can be divided into two phenotypically and functionally distinct subsets, based on their levels of expression of CD56. We recently found that CD27 distinguishes subsets of mature mouse NK cells. Here we report that CD27 can be used as a marker to discriminate human NK cell subsets. The majority of peripheral blood human NK cells were CD27 lo /CD56 dim NK cells, whereas the minor CD27 hi NK cell population correspondingly displayed a CD56 bright phenotype. Distinctions between CD27 lo and CD27 hi NK cells in their receptor expression and typical NK cell functions such as cytotoxicity and cytokine production can be easily delineated. Therefore, we propose the dual use of CD27 and CD56 as maturation/subset markers for human NK cells. The identification of CD27 subsets in both mice and humans will allow more accurate projections of the role of NK cell subsets in murine models of human pathologies where NK cells are involved. Introduction The concept of dividing human NK cells into functional subsets was originally proposed in the early 1980s (1), then revisited in late 1990s with the discovery that CD56 expression levels distinguished functionally distinct human NK cell subsets (2 5). Human CD56 bright NK cells are considered to be poorly cytotoxic, but potent cytokine producers, whereas CD56 dim NK cells display higher levels of cytotoxicity, but are poor cytokine producers. Although mouse NK cells resemble their human counterparts in many regards, including their cytotoxic ability and cytokine production, they do not express the murine homolog of CD56 and consequently the comparable subsets in the mouse have proven elusive. Therefore, it has been very difficult to relate our knowledge of mouse NK cell biology with human NK cell biology. Recently, we have identified two distinct populations of mouse NK cells based upon their cell surface expression of the tumor necrosis factor (TNF) receptor superfamily member CD27 (6). In mice, the mature Mac-1 hi NK cell pool can be divided into two functionally distinct CD27 hi and CD27 lo subsets as distinct from phenotypically immature Mac-1 lo NK cells (6). These mouse NK cell subsets show distinct character in their cytotoxicity, cytokine production, migratory capacity and tissue distribution. In this study, we report that CD27 can also be used as a marker to discriminate human NK cell subsets. The majority of human NK cells were CD27 lo /CD56 dim, whereas CD27 hi /CD56 bright NK cells were a minor population within peripheral blood. Similar to mouse NK cell subsets, there was a distinction between CD27 lo and CD27 hi NK cells in their receptor expression and NK cell functions such as cytotoxicity and cytokine production. Taken together, our study demonstrates that CD27 can be used in conjunction with CD56 to more accurately delineate human NK cell subsets. The identification of CD27 as a marker of NK subsets in humans should allow greater understanding of the relationship between mouse and human NK cells in pathologies. Materials and methods Human blood specimens PBMCs from healthy donors (Australian Red Cross Blood Service) were isolated by Ficoll Isopaque density gradient centrifugation (Amersham Bioscience) and then subsequently stained for flow cytometry analysis. For staining NK Correspondence to: Y. Hayakawa; yoshihiro_hayakawa@merck.com Received 1 October 2007, accepted 12 February 2008 Transmitting editor: K. Okumura Advance Access publication 7 March 2008

2 626 Human CD27 NK cell subsets cells, PBMCs were incubated with a saturating amount of mabs and flow cytometric analysis was performed with an LSR II instrument (BD Bioscience). Reagents Antibodies to CD3 (SK7), CD161 (DX12), CD56 (MEM188 or B159), CD158a (HP-3E4), CD158b (CH-L), CD27 (LG.3A10), LIAR-1 (DX26), NKG2A (131411), NKG2C (134591), CD94 (HP-3D9), NKG2D (1D11), NKp46 (9E2/NKP46), CXCR3 (49801), CCR7 (150503) and CD62L (DREG-56) were purchased from BD PharMingen (San Diego, CA, USA), R&D Systems (Minneapolis, MN, USA) and ebioscience (San Diego, CA, USA). Cytotoxicity assay Cytotoxic activity was assessed against K562 or P815 cells expressing MICA (P815 MICA) target cells by a standard 51 Cr-release assay. Effector cells were isolated from PBMCs by auto-macs (NK Isolation Kit, Miltenyi Biotec) and subsequently enriched by cell sorting [purified CD27 hi and CD27 lo NK cells (NK cells defined as CD3 CD161 + CD56 + ) at >90% purity]. Target cells were labeled with 100 lci ml 1 of Na 51 2 CrO 4 for 60 min at 37 C and labeled target cells were incubated in a total volume of 200 ll with effector cells in 96-well U-bottom plates. The plates were centrifuged before incubation, and after 4 h the supernatant was harvested and counted in a gamma counter. Cytokine assay For in vitro NK cell culture, NK cells were isolated from PBMCs by auto-macs (NK Isolation Kit, Miltenyi Biotec) and enriched by cell sorting [purified CD27 hi and CD27 lo NK cells (NK cells defined as CD3 CD161 + CD56 + ) at >90% purity]. Cells were stimulated with phorbol myristate acetate (PMA) (100 ng ml 1, Sigma) and ionomycin (100 ng ml 1, Sigma), and the cell-free supernatants were harvested after 20 h of incubation and subjected to BD Cytometric Bead Array (Human T h 1/T h 2 Cytokine Kit-II, BD Bioscience) according to the manufacturer s instructions. Results Dissection of human NK cells using CD27 First, we investigated the application of CD27 as a novel mature subset marker by examining its expression in the human NK cell lineage. Mature blood NK cells were delineated as CD3 CD161 + CD56 + cells within the PBMCs. By electronically gating on mature CD3 CD161 + CD56 + NK cells, we detected in a number of healthy donors two distinct CD27 hi and CD27 lo sub-populations (Fig. 1). The majority of bloodcirculating NK cells were CD27 lo NK cells and these cells preferentially expressed lower levels of CD56 (CD56 dim ). In contrast, there were two types of CD27 hi NK cells that expressed the CD56 marker at either higher (CD56 bright )or lower (CD56 dim ) levels; however, CD56 bright NK cells tended to be a major population among CD27 hi NK cells. These observations clearly indicate that dual staining of CD27 with CD56 acts as a more precise method to distinguish mature NK cell subsets in humans, in comparison to classical studies using CD56 alone. Surface phenotype of NK cell subsets distinguished by CD27 Since it had been previously suggested that NK cell receptor (NKR) expression was tightly linked to the differentiation or maturation of NK cells (7 9), we examined the expression of NKR on human NK cell subsets distinguished by CD27 (Fig. 2). The expression of inhibitory NKRs for self-mhc molecules [killer-cell Ig-like receptors (KIRs); CD158a, CD158b] was more frequent on CD27 lo NK cells, while non-mhc inhibitory receptor LIAR-1 expression was observed at similar level on both subsets. Although there were no clear differences in the expression of NKG2A or NKG2C, CD94 (which forms heterodimer receptor with either NKG2A or NKG2C) was highly expressed in CD27 hi populations. Among the activating receptors, we observed the higher expression of NKp46 on CD27 hi NK cells, whereas the expression of NKG2D was similar between the subsets. It has been recognized that NK cell behavior can also be regulated by chemokine/adhesion molecules as well as by other immune cells. Therefore, we examined the expression of chemokine receptors and lymphocyte homing receptor CD62L on human NK cell subsets as determined by CD27 expression. Although we did not detect distinct CXCR3 and CCR7 expression on human CD27 NK cell subsets (Fig. 2), the CD27 hi NK cell subset displayed higher CD62L expression. Taken together, human NK cell subsets, as determined by CD27 expression, displayed large diversity in their cell surface expression of receptors (summarized in Table 1) that regulate their prototypic effector functions and migratory/adhesive capacity. Functionally distinct CD27 hi and CD27 lo NK cell subsets We next determined the functional differences between human CD27 hi and CD27 lo NK cell subsets. We first examined NK cell cytotoxicity against K562 or P815-MICA target cells. Although there are individual variations between the donors in terms of the efficiency of their cytotoxicity, the CD27 lo subset tended to display higher cytotoxicity than CD27 hi NK cells against K562 cells (Fig. 3). In addition to the classical K562 human NK cell target, we determined NKG2D-specific cytotoxic capacity of CD27 NK cell subsets by using Fig. 1. CD27 dissects mature human NK cells into two subsets. Cells isolated from human peripheral blood samples were stained for CD161, CD3, CD56 and CD27. The dot plots shown are representative donor s profiles of 23 donors for the expression of CD56 and CD27 on electronically gated CD161 + CD3 cells. Numbers represent percentage of population in each region.

3 Human CD27 NK cell subsets 627 P815 MICA, which is a ligand for NKG2D. Distinct from cytotoxicity against K562 target cells, the NKG2D-specific cytotoxic capacity of CD27 hi NK cells was comparable or even higher (Fig. 3, Donor 1) compared with CD27 lo NK cells. While NKG2D expression was similar between the populations (Fig. 2), CD27 hi NK cells showed a far lower expression of perforin compared with CD27 lo subset (Fig. 3). In addition to cytotoxic capacity, early cytokine production by NK cells is considered to be important for effective innate immune responses and is further involved in the induction of subsequent adaptive immune responses. Therefore, we examined the potential of human CD27 NK cell subsets to produce inflammatory cytokines in vitro. Among the cytokines we tested (IL-2, IL-4, IL-6, IL-10, IFN-c and TNF-a) produced after the stimulation with PMA/ionomycin, CD27 hi NK cells produced higher amounts of IFN-c and TNF-a compared with the CD27 lo subset (Fig. 4). These results clearly indicate that CD27 hi NK cells possessed a greater ability to produce inflammatory cytokines than CD27 lo NK cells. Discussion In this study, we have verified that TNF receptor family CD27 can be applied as a marker to dissect human NK cells into CD27 hi and CD27 lo subsets with distinct cell surface phenotypes and functions. Human NK cell subsets determined by CD27 showed an overlapping phenotype with previously described CD56 bright and CD56 dim NK cell sub-populations in their cell surface characteristics and functions such as cytotoxicity and cytokine production. Our current understanding of NK cell development has primarily been derived from findings in mice. Since definition of the NK cell lineage between species has been difficult to determine owing to the differences in antigen expression, it Table 1. Summary of quadrant status of NK cell surface phenotype Quadrant Upper left Upper right Lower left Lower right CD158a CD158b LIAR NKG2A NKG2C CD NKG2D NKp CXCR CCR CD62L Data represent mean (%) 6 standard deviation of four donors shown in Fig. 1. Fig. 2. Cell surface phenotype of human NK cell subsets dissected by CD27. Cells isolated from human peripheral blood samples were stained with antibodies against indicated molecules together with CD161, CD3, CD56 and CD27. The dot plots shown are representative donor s profiles of 23 donors for the expression of CD27 and the indicated molecules electronically gated CD161 + CD56 + CD3 cells.

4 628 Human CD27 NK cell subsets Fig. 3. Cytotoxicity of human NK cell subsets dissected by CD27. CD27 hi and CD27 lo NK cells were sorted from human peripheral blood samples. Cytotoxicity was determined against K562 (A) or P815 MICA (B) target cells in standard 4-h 51 Cr-release assay. The results are shown as mean 6 standard deviation and represent two experiments. (C) Human peripheral blood samples were stained for intracellular perforin together with CD161, CD3, CD56 and CD27. The dot plots shown are each donor s profiles for the expression of intracellular perforin and CD27 electronically gated CD161 + CD56 + CD3 cells and represent three experiments. Fig. 4. Cytokine production of NK cell subsets dissected by CD27. Sorted NK cells (Donor 1, ; Donor 2, ) were stimulated with PMA and ionomycin for 20 h and cell-free culture supernatants were subjected to BD Cytometric Bead Array assay. The representative results from three experiments are shown. has remained a challenge to define the precise pathway of NK cell development in humans. Both the mouse in vivo and human in vitro studies have indicated that NK cell functional maturity, such as cytotoxicity and cytokine production, seems to be acquired at late stage of their development (10, 11). Further, human NK cell differentiation has been known to associate with the acquisition of series of NK cell receptors. In humans, CD161 and NKp46 are known to be the first NK cell receptors, among other receptors, followed by CD94, and lastly KIRs (5, 11 13). Recent studies have further identified the intermediates of the human NK cell lineage within human secondary lymphoid tissues. These classifications can be made on the basis of their expression of CD56, CD117, CD94 and CD161 (14). By analyzing the intermediate stage of NK cell maturation, it has been proved that the potential of lineage differentiation, functional capacity and the expression of transcription factors can be distinct at different stages of human NK cell differentiation (14, 15). In this study, we have classified mature NK cells as CD3 CD161 + CD56 + reflecting the observation that CD3 CD161 +

5 cells may include immature types of human NK cells. Then we further dissected mature human NK cells into two subsets by the co-expression of high levels of CD27 and CD56. Although the majority of CD27 hi and CD27 lo NK cells overlap with CD56 bright and CD56 dim NK cells, respectively, we observed a minor population of CD27 hi NK cells bearing the CD56 dim phenotype or CD27 lo NK cells bearing the CD56 bright phenotype. These results may suggest that a combination of the relatively ignored CD27 marker with the conventionally employed CD56 human NK cell subset marker will further dissect NK cells into functional sub-populations. Considering the expression profile of NKR, other differentiation markers and functional maturity, it is clear that the CD27 lo NK cell subset, similar to mouse CD27 lo NK cells, is likely at a more mature stage in human NK cell differentiation (6, 16). Further study would be required to determine the precise mechanism by which human NK cell maturation can be regulated by cytokines and cell cell interactions. Early studies of the resting CD56 dim human NK subset revealed that these cells were more cytotoxic than CD56 bright NK cells (17), although this difference was compensated by in vitro activation (18 20). Consistent with differences in their cytotoxic potential, the CD56 dim subset has higher levels of cytotoxic granule expression than CD56 bright cells (17). Moreover, human NK cell subsets also have distinct expression of NKRs that might also account for the differences in their cytotoxic capacity (21, 22). By comparing human NK cell and mouse NK cell subsets, the CD27 hi mouse NK subset shows a similar NKR expression pattern to the CD56 bright human subset; however, by contrast, the CD27 hi mouse subset is highly cytotoxic as opposed to the CD56 bright human subset. One caveat is that it is very difficult to determine the cytotoxic capacity of human NK cells against MHC-matched target cells simply because there is usually a lack of autologous tumor target cells against which to test. Therefore, results suggesting that the CD56 bright NK cell subset is a less-cytotoxic type of NK cell need to be carefully interpreted. In contrast to cytotoxic potential, human CD56 bright NK cells have been considered to be effective producers of inflammatory cytokines, such as IFN-c. In this study, we demonstrated that human CD56 bright /CD27 hi NK cells produce much higher levels of IFN-c and TNF-a compared with the CD56 lo /CD27 lo subset. Therefore, the distinct cytokine production capacity of CD27 hi NK cell subset suggests their role, providing an early source of IFN-c or other NK cellderived cytokines during the innate immune response. While little information is available about the tissue distribution of human NK cell subsets, it is known that a greater proportion of the CD56 bright subset is found in human lymph nodes than in peripheral blood (10, 23). In mouse, the tissue distribution of NK cell subsets in various lymphoid and non-lymphoid organs including BM, spleen, LN, peripheral blood, liver and lung is quite distinct (6, 16, 24, 25). It has been shown that cytokine production by NK cells supports T h 1-mediated immunity in secondary lymphoid organs (26), indicating that NK cells provide a direct link between innate and adaptive immunity during T cell priming in the lymph nodes. Given that the CD27 hi subset showed similarities to the CD56 bright NK cell subset [i.e. the predominant resident mature NK cell population within secondary lymphoid organs in both human Human CD27 NK cell subsets 629 and mouse and possessing greater cytokine production potential (6, 10, 14)], these subsets may play a distinct role in priming adaptive immunity. It has been also demonstrated that CD27 directly interacts with its ligand CD70 to activate human NK cells (27, 28), and therefore the CD27 expression level may have some bearing in NK cell responsiveness in the immune responses involving CD70 interactions. Nevertheless, by discovering that CD27 is a common maturation marker for human and mouse NK cell subsets, our study provides a very important platform on which to study mature NK cells in immune responses in experimental animals. It is very likely that this may provide a rapid conduit of information from the mouse that can be potentially applied in the study of human NK cells following vaccination and in a variety of human infectious diseases, autoimmunity and cancer. Acknowledgements We gratefully thank Kazuyoshi Takeda and Hilary Warren for their very helpful discussions. We wish to thank Mark Shannon, Ralph Rossi, Daud Duhur and the staff of the Pathology Department at Peter MacCallum Cancer Centre for their generous support. Abbreviations KIR killer-cell Ig-like receptor NKR NK cell receptor PMA phorbol myristate acetate P815 MICA P815 cells expressing MICA TNF tumor necrosis factor MICA MHC class I chain-related protein A BM Bone marrow LN Lymph node References 1 Lanier, L. L., Le, A. M., Phillips, J. H., Warner, N. L. and Babcock, G. F Subpopulations of human natural killer cells defined by expression of the Leu-7 (HNK-1) and Leu-11 (NK-15) antigens. J. Immunol. 131: Sedlmayr, P., Schallhammer, L., Hammer, A., Wilders-Truschnig, M., Wintersteiger, R. and Dohr, G Differential phenotypic properties of human peripheral blood CD56dim+ and CD56bright+ natural killer cell subpopulations. Int. Arch. Allergy Immunol. 110: Frey, M., Packianathan, N. B., Fehniger, T. A. et al Differential expression and function of L-selectin on CD56bright and CD56dim natural killer cell subsets. J. Immunol. 161: Cooper, M. A., Fehniger, T. A. and Caligiuri, M. A The biology of human natural killer-cell subsets. Trends Immunol. 22: Colucci, F., Caligiuri, M. A. and Di Santo, J. P What does it take to make a natural killer? Nat. Rev. Immunol. 3: Hayakawa, Y. and Smyth, M. J CD27 dissects mature NK cells into two subsets with distinct responsiveness and migratory capacity. J. Immunol. 176: Dorfman, J. R. and Raulet, D. H Acquisition of Ly49 receptor expression by developing natural killer cells. J. Exp. Med. 187: Roth, C., Carlyle, J. R., Takizawa, H. and Raulet, D. H Clonal acquisition of inhibitory Ly49 receptors on developing NK cells is successively restricted and regulated by stromal class I MHC. Immunity 13: Raulet, D. H., Vance, R. E. and McMahon, C. W Regulation of the natural killer cell receptor repertoire. Annu. Rev. Immunol. 19: Freud, A. G., Becknell, B., Roychowdhury, S. et al A human CD34(+) subset resides in lymph nodes and differentiates into CD56bright natural killer cells. Immunity 22: Zamai, L., Ahmad, M., Bennett, I. M., Azzoni, L., Alnemri, E. S. and Perussia, B Natural killer (NK) cell-mediated cytotoxicity:

6 630 Human CD27 NK cell subsets differential use of TRAIL and Fas ligand by immature and mature primary human NK cells. J. Exp. Med. 188: Yokoyama, W. M., Kim, S. and French, A. R The dynamic life of natural killer cells. Annu. Rev. Immunol. 22: Mrozek, E., Anderson, P. and Caligiuri, M. A Role of interleukin-15 in the development of human CD56+ natural killer cells from CD34+ hematopoietic progenitor cells. Blood 87: Freud, A. G., Yokohama, A., Becknell, B. et al Evidence for discrete stages of human natural killer cell differentiation in vivo. J. Exp. Med. 203: Grzywacz, B., Kataria, N., Sikora, M. et al Coordinated acquisition of inhibitory and activating receptors and functional properties by developing human natural killer cells. Blood 108: Hayakawa, Y., Huntington, N. D., Nutt, S. L. and Smyth, M. J Functional subsets of mouse natural killer cells. Immunol. Rev. 214: Nagler, A., Lanier, L. L., Cwirla, S. and Phillips, J. H Comparative studies of human FcRIII-positive and negative natural killer cells. J. Immunol. 143: Caligiuri,M.A.,Zmuidzinas,A.,Manley,T.J.,Levine,H.,Smith,K.A. and Ritz, J Functional consequences of interleukin 2 receptor expression on resting human lymphocytes. Identification of a novel natural killer cell subset with high affinity receptors. J. Exp. Med. 171: Nagler, A., Lanier, L. L. and Phillips, J. H Constitutive expression of high affinity interleukin 2 receptors on human CD16- natural killer cells in vivo. J. Exp. Med. 171: Robertson, M. J., Soiffer, R. J., Wolf, S. F. et al Response of human natural killer (NK) cells to NK cell stimulatory factor (NKSF): cytolytic activity and proliferation of NK cells are differentially regulated by NKSF. J. Exp. Med. 175: Colonna, M., Navarro, F., Bellon, T. et al A common inhibitory receptor for major histocompatibility complex class I molecules on human lymphoid and myelomonocytic cells. J. Exp. Med. 186: Voss, S. D., Daley, J., Ritz, J. and Robertson, M. J Participation of the CD94 receptor complex in costimulation of human natural killer cells. J. Immunol. 160: Fehniger, T. A., Cooper, M. A., Nuovo, G. J. et al CD56bright natural killer cells are present in human lymph nodes and are activated by T cell-derived IL-2: a potential new link between adaptive and innate immunity. Blood 101: Kim, S., Iizuka, K., Kang, H. S. et al In vivo developmental stages in murine natural killer cell maturation. Nat. Immunol. 3: Takeda, K., Cretney, E., Hayakawa, Y. et al TRAIL identifies immature natural killer cells in newborn mice and adult mouse liver. Blood 105: Martin-Fontecha, A., Thomsen, L. L., Brett, S. et al Induced recruitment of NK cells to lymph nodes provides IFN-gamma for T(H)1 priming. Nat. Immunol. 5: Sugita, K., Robertson, M. J., Torimoto, Y., Ritz, J., Schlossman, S. F. and Morimoto, C Participation of the CD27 antigen in the regulation of IL-2-activated human natural killer cells. J. Immunol. 149: Yang, F. C., Agematsu, K., Nakazawa, T. et al CD27/CD70 interaction directly induces natural killer cell killing activity. Immunology 88:289.