Article
Reference
Peripheral blood CD56
brightNK cells respond to stem cell factor and adhere to its membrane-bound form after upregulation of c-kit
PRADIER, Amandine, et al.
PRADIER, Amandine, et al . Peripheral blood CD56
brightNK cells respond to stem cell factor and adhere to its membrane-bound form after upregulation of c-kit. European Journal of
Immunology , 2014, vol. 44, no. 2, p. 511-520
DOI : 10.1002/eji.201343868 PMID : 24150691
Available at:
http://archive-ouverte.unige.ch/unige:38276
Disclaimer: layout of this document may differ from the published version.
Eur. J. Immunol. 2013. 00: 1–10 DOI: 10.1002/eji.201343868 Innate immunity
1
Peripheral blood CD56 bright NK cells respond to stem cell factor and adhere to its membrane-bound form after upregulation of c-kit
Amandine Pradier
1, Severine Tabone-Eglinger
2, Vincent Huber
1, Carine Bosshard
1, Emmanuel Rigal
1, Bernhard Wehrle-Haller
2and Eddy Roosnek
11Division of Hematology, Department of Internal Medicine, Geneva University Hospitals and University of Geneva, Switzerland
2Department of Cell Physiology and Metabolism, Geneva Medical School, University of Geneva, Geneva, Switzerland
CD56brightNK cells express receptors for IL-2, IL-7, IL-15, and SCF. We found that human peripheral blood CD56bright NK cells responded to IL-2, IL-7, IL-15 by phosphorylating STAT-5, ERK, and Akt but did not respond to SCF. However, CD56bright NK cells in cul- ture upregulated c-kit transcription three to fourfold, which led to a steady increase in c-kit and a concomitant acquisition of responsiveness to SCF. After 44 h, CD56bright NK cells had upregulated c-kit approximately 20-fold and phosphorylated ERK and Akt in response to SCF concentrations well below levels present in plasma. CD56bright NK cells cultured in IL-15 maintained c-kit transcription/expression at ex vivo levels and did not become responsive to SCF. Furthermore, SCF-responsive, CD56brightc-kithighNK cells swiftly downregulated c-kit and stopped responding to SCF after IL-15 stimulation. How- ever, commitment of CD56bright NK cells to a c-kit-negative, SCF-unresponsive state did not occur, as after 5 days of culture, withdrawal of IL-15 restored c-kit to maximal levels and reestablished SCF-responsiveness. CD56bright NK cells that had upregulated c-kit firmly adhered to COS cells transfected with the membrane form of SCF. Furthermore, SCF signaling significantly increased the capacity of CD56bright NK cells to degranulate.
Collectively, our data suggest that c-kit on human CD56bright NK cells is a functional receptor that is downregulated in peripheral blood, possibly to render CD56brightNK cells unresponsive to the SCF therein.
Keywords: c-Kit rCytokines rHuman rInflammation rNK cells rSCF
Introduction
NK cells are involved in the early response to pathogens as well as in the recognition of autologous cells under stress resulting from infection or transformation. In humans, they are identified on the basis of expression of CD56 and absence of CD3. NK cells consist
Correspondence:Prof. Eddy Roosnek e-mail: eddy.roosnek@unige.ch
of at least two subpopulations [1, 2]. One population expresses relatively low levels of CD56 (CD56dim) and appears to be end- stage effector cells endowed with a strong cytolytic activity. In contrast, the population expressing a one-log higher level of CD56 (CD56bright) is less cytotoxic and has a higher capacity to produce cytokines [1–8].
CD56dim NK cells mainly produce cytokines when in contact with target cells [9, 10]. By contrast, CD56brightNK cells are more likely to be activated by IL-15 produced and presented in trans by activated monocytes or DC or by IL-2 produced by CD4-positive
T cells during the onset of the adaptive immune response. CD56dim NK cells represent the major subpopulation in peripheral blood (>90%) but CD56brightNK cells may outnumber CD56dimNK cells because they are the predominant population in secondary lym- phoid organs [7, 11].
CD56dimand CD56brightNK cells display different combinations of cell surface receptors. CD56dimNK cells express high levels of the Fcγ-receptor type III (CD16) and the chemokine receptors for IL-8 and fractalkine (CXCR1 and CX3CR1) to guide them into inflamed tissues [3, 12]. By contrast, CD56brightNK cells use CCR5 for entry into inflamed tissues and CCR7 to home into secondary lymphoid organs, where they can be found in close vicinity of DC [6,13,14].
Both subpopulations express the IL-2 receptor β-chain and the commonγ-chain that form heterodimers with low to intermediate affinity for IL-2 and IL-15 [15]. CD56bright NK cells also express low levels of the IL-2Rα-chain (CD25) conferring specificity and high affinity for IL-2 [16] as well as c-kit [3, 17–20], the type III tyrosine kinase receptor for SCF.
SCF is an essential growth factor for hematopoietic progenitor cells produced by stromal cells in bone marrow. Most hematopoi- etic lineages downregulate c-kit as they differentiate into more mature forms [21, 22], but mast cells, eosinophils and CD56bright NK cells retain considerable levels of c-kit. During inflamma- tion, fibroblasts and endothelial cells produce SCF [21, 22].
Mature mast cells and eosinophils respond to SCF with chemo- taxis [21,22], adhesion to fibronectin or endothelial cells [23–25]
and degranulation [26]. The response of CD56bright NK cells to SCF has remained somewhat elusive. It has been shown that SCF decreases apoptosis in CD56bright NK cells stimulated with IL-2, enhancing proliferation and IFN-γproduction in presence of IL-12 [17–20]. It is notable that these responses to SCF have been observed mainly in CD56brightNK cells cultured for several days.
Even early signaling such as SCF-triggered phosphorylation of ERK through MAPK, the critical pathway to activate NK cells [27] has only been reported for NK-cell lines or for NK cells that had been stimulated with IL-2 [20]. Herein, we show that CD56bright NK cells in peripheral blood do not respond to SCF because they have downregulated c-kit. Furthermore, we provide data showing that the decrease of c-kit expression as well as the ensuing unrespon- siveness to SCF is readily reversed by withdrawing the stimuli at their origin.
Results
Early signaling in primary CD56brightNK cells stimulated by IL-2, IL-7, IL-15, or SCF
Commonγ-chain cytokines and SCF signal through MAPK-ERK, PI3K-Akt, and/or STAT-5 pathways. In NK cells, IL-2 and IL-15 have been shown to activate all three pathways [20, 27–30]. IL-7 signaling has not been studied in detail. SCF induces ERK phos- phorylation in hematopoietic progenitors [31], mast cells [22] as well as in c-kit positive endothelial cells [32] but apart from the observation that SCF enhances ERK phosphorylation in NK cells
stimulated with IL-2 [20], little is known of its impact on NK cells ex vivo. We found that the response of peripheral blood CD56bright NK cells to IL-15 was very similar to the response described for CD56dimNK cells or for NK cell lines (Fig. 1). Within minutes after stimulation, 100 pM-2 nM of IL-15 induced maximal phosphoryla- tion of STAT-5, ERK, and Akt while responses induced by concen- trations lower than 100 pM were suboptimal. IL-2 also signaled through each of the three pathways, but on a molecular basis, appeared to be approximately one log less efficient than IL-15 for CD56brightNK cells and two logs less efficient for CD56dimNK cells.
Furthermore, as reported for other lymphocytes, stimulation with IL-7 phosphorylated STAT-5, but not ERK [33, 34]. IL-7 induced phosphorylation of Akt varied from negative to borderline in dif- ferent experiments. CD56dimNK cells that do not express the IL-7 receptor [35] did not respond. We did not observe SCF-induced phosphorylation of ERK over a broad range of concentrations.
Phosphorylation of Akt was undetectable at SCF-concentrations lower than 5 nM and marginal or negative at higher concentra- tions. Hence, CD56bright NK cells in peripheral blood respond to commonγ-chain cytokines for which they express receptors but, in spite of expressing c-kit, do not respond to SCF.
CD56brightNK cells respond to SCF after separation from their cytokine milieu in vivo
CD56brightNK cells express c-kit and SCF has been shown to mod- ulate their response to cytokines in culture [17–19]. The latter is hard to reconcile without SCF-induced phosphorylation of ERK.
To test whether CD56bright NK cells had received signals in vivo interfering with SCF-induced phosphorylation of ERK, we moni- tored alterations of the response to SCF with time. We found that after 6–8 h of culture, CD56brightNK cells phosphorylated ERK and Akt (Fig. 2) but not STAT-5 (data not shown) in response to 10 nM of SCF. With time, sensitivity to SCF gradually increased. After 20 h, CD56brightNK cells phosphorylated ERK and Akt after stimu- lation by SCF-concentrations from 0.5–1 nM on (data not shown) while after 44 h, SCF concentrations as low as 2–5 pM induced half-maximal phosphorylation of ERK and Akt. By contrast, phos- phorylation of STAT-5, ERK, and Akt after stimulation with the commonγ-chain cytokines IL-7 and IL-15 was virtually identical (data not shown). Interestingly, CD56bright NK cells cultured in the presence of IL-7 became responsive to SCF in a similar man- ner but IL-15 blocked acquisition of responsiveness to SCF com- pletely. Hence, signaling through c-kit is controlled by cytokines and while circulating in blood and lymph nodes, CD56bright NK cells may receive signals that negatively regulate SCF signaling through c-kit.
CD56brightNK cells that respond to SCF have upregulated c-kit
Receptor downregulation is a common mechanism to attenu- ate cellular responses to their specific ligands. Proinflammatory
Eur. J. Immunol. 2013. 00: 1–10 Innate immunity
3
Figure 1. CD56bright NK cells in periph- eral blood respond to IL-2, IL-7, and IL-15 but not to SCF. STAT5-, ERK-, and Akt- phosphorylation is shown as assessed by FACS analysis of purified NK cells incu- bated for 5 (pSTAT5/pERK) or 15 (pAkt) min- utes at 37◦C in the absence of cytokines (Unst) or stimulated with IL-15 (5 nM), IL-2 (5 nM (IL-2H) or 50 pM (IL-2L)), IL-7 (5 nM), or SCF (5 nM). Dot plots were gated on lymphocytes by forward and side scatter and on CD56 whereas the dose–
response curves in the right panels show- ing STAT5-, ERK-, and Akt-phosphorylation in CD56bright or CD56dim NK cells stimu- lated with the indicated concentration of the respective cytokines were gated on either CD56bright(4 panels) or CD56dim(the middle panel marked CD56dim) lympho- cytes. The lower responses induced by suboptimal cytokine concentrations in the dose–response curves reflect lower phos- phorylation per cell rather than lower numbers of responding cells. MFIR: geomet- ric mean fluorescence intensity ratio (MFI stimulated sample/MFI unstimulated sam- ple). One representative experiment of at least 3 is shown.
signals such as IL-1β, TNF-α, IFN-γ, and GM-CSF downregulate c- kit on hematopoietic precursor cells and mast cells [21,22]. Hence, CD56brightNK cells ex vivo that express only low levels of c-kit may be unresponsive to SCF because they have downregulated c-kit in response to such stimuli in vivo. We have measured c-kit levels during the cultures depicted in Fig. 2 on cells that became respon- sive to SCF or not, and compared it with the levels of CD25 and CCR7 under the same conditions. We found that CD56bright NK cells cultured in the absence of cytokines (Fig. 3, left panel) or in the presence of IL-7 expressed approximately 20 times more c-kit than CD56bright ex vivo. By contrast, the level of c-kit on CD56brightNK cells cultured in the presence of IL-2, IL-15, or SCF had changed only little. Hence, responses to SCF are associated with a high surface expression of c-kit.
It is notable that after 44 h of culture, we recovered fewer CD56brightNK cells from cultures without cytokines than from cul- tures containing IL-7. This was not unexpected because CD56bright NK cells undergo apoptosis after 2 days in culture without growth factors [18]. However, during the first 44 h of culture, we never
observed differences in c-kit upregulation or responses to SCF of CD56bright NK cells cultured in medium without cytokines or in medium containing IL-7.
The change in receptor density on unstimulated CD56brightNK cells was highly specific for c-kit. CD25 was upregulated only after stimulation with IL-15, whereas CCR7 was downregulated only after stimulation with IL-2 or IL-15. Receptor expression was mainly regulated through transcription (Fig. 3, right panel) but the presence of the ligand-limited cell surface expression. CD56bright NK cells ex vivo and CD56bright NK cells stimulated with IL-2 or IL-15 transcribed c-kit at an almost identical level, which was sig- nificantly lower than that of unstimulated cells. Hence, transcrip- tion levels concurred accurately with the respective levels of c-kit at the cell surface. CD56brightNK cells stimulated with SCF tran- scribed c-kit at a similar level as unstimulated cells but expressed only low levels of c-kit, most likely because c-kit is rapidly internal- ized and degraded upon binding of its ligand [22,36]. The level of CD25-transcription increased sharply after stimulation with IL-2 or IL-15 but as has been reported for T cells [37], only cells
Figure 2. CD56bright NK cells acquire responsiveness to SCF with time. ERK- and Akt-phosphorylation in purified NK cells ex vivo stimulated with SCF (100 pM or 10 nM, 5 min at 37◦C) (upper panels) or cultured in the absence of cytokines for 8 (middle panels) or 44 h (lower panels). The dose–
response curves show SCF-induced ERK- and Akt-phosphorylation in CD56brightNK cells cultured for 44 h without cytokines or in the presence of IL-7 (5 nM) or IL-15 (5 nM).
FACS gates as in Fig. 1. One representative experiment of 3 is shown.
stimulated with IL-15 expressed high levels of CD25. Stimulation with IL-2 or IL-15 increased transcription of CCR7 during the first hours (not shown) but significantly decreased transcription there- after, which led to the disappearance of CCR7 from the membrane during the second day of culture.
c-Kit downregulation by IL-15 and the ensuing SCF unresponsiveness is instantaneous but reversible
The fact that CD56bright NK cells cultured in the presence of IL- 15 do not upregulate c-kit (Fig. 3) nor become responsive to
SCF (Fig. 2) suggests that IL-15 is able to regulate responses to SCF. However, because CD56bright NK cells cultured in IL-15 enter into cell cycle and start to express markers that are consid- ered to be features of CD56dim NK cells [7, 11], one could also argue that the IL-15 induced low levels of c-kit and unresponsive- ness to SCF are simply hallmarks of cell division or of the start of a possible maturation into c-kit-negative CD56dimNK cells. To investigate this further, we cultured CD56brightNK cells for 44 h to induce maximal c-kit expression and stimulated with IL-7 or IL-15. CD56brightNK cells stimulated with IL-7 did not downreg- ulate c-kit and remained fully responsive to SCF (Fig. 4A). By contrast, IL-15 swiftly downregulated c-kit to a level comparable
Figure 3. Impact of IL-15, IL-7, IL-2, and SCF on the expression of c-kit, CD25 and CCR7 on CD56brightNK cells. Expression of c-kit, CD25 and CCR7 (left panels) on CD56brightNK cells, either ex vivo or cultured for 44 h in the absence of cytokines (Unst) or in the presence of IL-15 (5 nM), IL-2 (5 nM), IL-7 (5 nM), or SCF (5 nM). FACS gates as in Fig. 1. One representative experiment of 4 is shown. The right panels show the fold increase in expres- sion of c-kit, CD25, and CCR7, relative to the house keeping gene TBP, under the same conditions. Data are shown as mean+SD of three samples pooled from at least three independent exper- iments. Responses of cultured cells were compared with that of ex vivo CD56brightNK cells using one-way ANOVA followed by Tukey’s (***p<0.001, **p<0.01, *p<0.05).
Eur. J. Immunol. 2013. 00: 1–10 Innate immunity
5
Figure 4. IL-15-induced downregulation of c-kit and SCF- responsiveness is instantaneous but reversible for at least 5 days. (A) c-kit expression and (B) SCF-induced ERK phosphorylation in CD56bright NK cells cultured for 44 h in the absence of cytokines followed by stimulation with IL-7 or IL-15 for 1–4 h. (C) c-Kit expression and (D) SCF-induced ERK phosphorylation in purified CD56bright NK cells cultured in IL-15 for 6 days; cultured in IL-15 for 5 days, washed and subsequently cultured in IL-7 for 1 day; or cultured in IL-15 for 4 days, washed and subsequently cultured in IL-7 for 2 days. Analysis of SCF-induced ERK phosphorylation was performed as in the previous figures. After withdrawal of IL-15, IL-7 was added to improve cell viability. Data are shown as mean+SD of one sample from at least three independent experiments. Comparison of responses with the response of unstimulated cells (A, B) or with cells cultured in IL-15 for 6 days (C, D) was done as described in Fig. 3. (***p<0.001, **p<0.01,
*p<0.05).
with that of CD56brightNK cells cultured for 2 days in IL-15 in the experiments shown in Fig. 3. This was owed to a prompt six to sevenfold decrease of transcription that had returned to ex vivo levels after 4 h (data not shown). In addition, 1 h of stimulation with IL-15 was sufficient to decrease SCF-induced phosphoryla- tion of ERK significantly (Fig. 4B). Interestingly, even after 5 days in culture, withdrawal of IL-15 not only restored c-kit to maxi- mal levels but also reestablished responsiveness to SCF (Fig. 4C, D). Hence, IL-15 induced downregulation of c-kit and its ensuing
Figure 5. SCF-signaling increases K562-induced CD107 expression in CD56brightNK cells that have upregulated c-kit. CD56brightNK cells ex vivo (left) or CD56bright/CD56dimcells (middle and right) previously cul- tured for 44 h under the conditions in Fig. 2 were stimulated overnight with SCF (5 nM) and assessed for expression of CD107 by FACS analysis.
Each line represents an individual experiment.
unresponsiveness to SCF is instantaneous but remains reversible for at least 5 days.
SCF-signaling of CD56brightNK cells with upregulated c-kit increases their capacity to degranulate
Phosphorylation of ERK and Akt does not necessarily mean that SCF also activates signaling pathways downstream. To determine the impact of SCF on CD56brightNK cells, we incubated purified NK cells overnight with SCF and measured degranulation after stim- ulation with K562 target cells by the percentage of cells express- ing CD107a with FACS gates on viable CD56brightor CD56dimNK cells that served as control. Very few NK cells (3.1±2.4%,n= 20) degranulated without being stimulated with K562 (data not shown). Overnight incubation with SCF did not change the per- centage of CD56bright NK cells ex vivo degranulating in response to K562 (20.1±17.9 versus 22.0±16.8, see left panel of Fig. 5 for five independent experiments). By contrast, SCF stimulation significantly increased the percentage of degranulating CD56bright NK cells that had upregulated c-kit during the preceding 44 h cul- ture (21.2±11.0 versus 35.9±13.9 (p<0.004), middle panel, n=8), whereas the percentage of degranulating CD56dimNK cells cultured under the same conditions did not change (31.3±17.1 versus 33.1±15.7, right panel,n=6). Hence, CD56brightNK cells respond to SCF by increasing their capacity to degranulate but only after they have upregulated c-kit.
CD56brightNK cells that have upregulated c-kit adhere to COS cells transfected with mSCF
SCF induces mast cell adhesion to fibronectin and endothelial cells [23–25]. To study whether CD56brightNK cells behaved simi- larly, we compared adherence to COS cells transfected with mem- brane bound SCF (mSCF) of CD56brightNK cells ex vivo to that of
CD56brightthat had upregulated c-kit after 44 h of culture (Fig. 6).
The number of CD56brightNK cells ex vivo adhering to mSCF-COS (1–2 CD56bright/COS) did not exceed background as defined by the number of CD56bright NK cells expressing high levels of c-kit adhering to mSCF-negative COS. By contrast, three times as many CD56brightNK cells adhered to mSCF-COS (5–6 CD56bright/COS) after upregulation of c-kit during 44 h of culture. Hence, besides increasing sensitivity to SCF signaling, upregulation of c-kit also increases adherence to cells expressing mSCF and it is possible that this contributes to the retention of CD56bright NK cells in inflammatory sites.
Discussion
NK cells have been named after their capacity to kill tumor cell lines without need for previous priming. To date, we begin to appreciate numerous other functions they may exert. This holds in particular for CD56bright NK cells, a subpopulation of NK cells that express a unique repertoire of cytokine, chemokine and adhe- sion receptors [3, 12]. Many of these receptors are informative on CD56brightNK cells homing patterns and on how these cells may drive T cell polarization or assist in the control of viral infections.
CD56brightNK cells express CCR7 and L-selectin that guides them to secondary lymphoid organs where they can be found in T cell area in close vicinity of DC [8, 14, 38, 39]. Furthermore, they express receptors for IL-12 and IL-15 [40,41] so that they are activated by DC that have sensed danger signals as well as receptors for IL-2 and IL-4 to interact with a concomitant T-cell response [2, 6, 16].
Once activated, they leave the secondary lymphoid organs and are guided by CCR5 to tissues where they contribute to the inflamma- tory response [8, 42–44].
CD56bright NK cells also express c-kit [3, 17–20]. Mast cell precursors in blood use c-kit to adhere to endothelial cells that express mSCF under inflammatory conditions. In addition, c-kit- mediated signaling contributes to their maturation and survival after entry into inflamed tissues [21, 22, 24, 25]. It is conceivable that CD56bright NK cells follow similar routes but after its dis- covery more than 20 years ago [17], the interest in c-kit on NK cells has somewhat faded not only because the effects of SCF on CD56bright NK cells appeared to be only minor [17–20] but also because CD56bright NK cells are considered by many as a sort of precursor of CD56dimNK cells [7,11,45,46], which would demote c-kit to a receptor on its way out. Recently, we have reported that after hematopoietic stem cell transplantation, CD56brightNK cells are mostly c-kit negative but swiftly upregulate c-kit during cul- ture. Furthermore, we noticed that c-kit expression on CD56bright NK cells in transplanted patients and in control individuals are modulated by cytokines [47]. In this report, we show underlying mechanisms and functional consequences. We believe that these findings allow speculating on features of CD56brightNK cells com- plementary to those inferred from their repertoire of chemokine and adhesion receptors.
We found that CD56brightNK cells in peripheral blood express only low levels of c-kit and do not respond to SCF. The latter
Figure 6. CD56brightNK cells with upregulated c-kit adhere to COS cells transfected with mSCF. Adherence of either (A) purified CD56brightNK cells ex vivo or (B) purified CD56brightNK cells expressing high levels of c-kit after 44 h in culture to mSCF-GFP-expressing COS cells was exam- ined by phase-contrast microscopy (magnification 63). The inserts show immunofluorescence in the same field, depicting the presence mSCF- GFP in the transfected cells. (C) The mean number of CD56brightcells adhered to COS cells counted per field is shown, which corresponds approximately to the number of CD56bright adhered to 1.3 COS cells.
Data are shown as mean±SD of at least 13 fields from each of the three independent experiments. Adherence of purified CD56brightNK cells expressing high levels of c-kit to COS-mSCF (right bar) was com- pared with adherence of CD56brightNK cells ex vivo to COS-mSCF (left bar) or with adherence of CD56brightNK cells expressing high levels of c- kit to untransfected COS cells (middle bar) as described in the previous figures (**p<0.01).
Eur. J. Immunol. 2013. 00: 1–10 Innate immunity
7
makes sense because responses to SCF present in blood should be avoided. We also show that downregulation of c-kit ren- ders CD56brightNK cells nonresponsive to SCF. Although it took 6–8 h of culture to upregulate c-kit to a level that CD56bright NK cells responded to SCF in vitro, this process could well be faster after cells have extravasated and are stimulated by mSCF on cells simultaneously presenting ligands for other recep- tors. Indeed, thresholds are likely to be different in vivo when CD56bright NK cells receive other signals that might be acces- sory. Hence, once extravasated, upregulation of c-kit may not only render CD56bright NK cells sensitive to SCF-induced sur- vival signals but also firmly anchor them to cells expressing mSCF.
We used IL-15 to show that CD56bright NK cells shut down responsiveness to SCF immediately after being stimulated. We would have obtained similar results with IL-2 but we have no indication that one of these cytokines is the cause of the down- regulation of c-kit on peripheral blood CD56bright NK cells in vivo. SCF also downregulates c-kit (Fig. 3) but its concentration in plasma (242.1 ± 23.0 pg/mL) [48] is too low by 2 logs to have an impact (data not shown). Downregulation may have occurred in LN on encounter with DC presenting IL-15 or in response to other cytokines with similar effects. We did observe that the upregulation of c-kit on CD56brightNK cells and its ensu- ing increase of responsiveness to SCF in cultures of peripheral blood mononuclear cells were less pronounced than in cultures of purified NK cells. Although it is tempting to speculate that the membrane bound IL-15 on monocytes present in the same cul- ture prevented much of the upregulation, we have not been able to pinpoint the cause exactly and many other signals may also contribute.
Our crucial finding is the fact that peripheral blood CD56bright NK cells do not respond to SCF does not mean that will not do so later on. Furthermore, we show that commitment to a c-kit- negative NK cell is not readily obtained by stimulating CD56bright NK cells with IL-15 because even after one week, withdrawal almost immediately restores c-kit to a level that is one log higher than found on CD56bright NK cells in peripheral blood. We also tested complexes of IL-15 and its soluble receptor α-chain that signals human NK cells in a similar way as IL-15 presented in trans [30, 46], but did not find indications that CD56brightNK cells would become more committed to a c-kit-negative cell than after stimulation with IL-15 alone (data not shown).
Collectively, our data suggest that peripheral blood CD56bright NK cells are a distinct subpopulation of NK cells that migrates to inflammatory sites and becomes effector cells after stimula- tion by antigen-presenting cells [8, 14, 42, 43], endothelial cells, or platelets [44, 49]. Although during this process, CD56brightNK cells may acquire features of CD56dimNK cells [7, 11], we do not believe that this is a major pathway for the generation of CD56dim and that it may be time to fine-tune the current paradigm of periph- eral blood CD56bright NK cells being precursors of CD56dim NK cells.
Materials and methods
NK cell isolation and culture
Purified NK cells or purified CD56brightNK cells were obtained from buffy coats from healthy blood donors by standard gradient Ficoll density centrifugation and MACS using the human NK Cell Isola- tion Kit or the CD56+CD16−human NK Cell Isolation Kit, respec- tively (Miltenyi Biotec, Bergisch Gladbach, Germany), according to manufacturer’s instructions. Purities were always>90 for NK cells and >75% for CD56bright NK cells. Blood donors (age 45
±16, 55% male, 45% female) from our blood transfusion cen- ter had given consent and the study had been approved by our institutional ethical committee. Purified NK cells were cultured in AIMV medium (Life Technology, Zug, Switzerland) containing 2% Hepes and incubated at 37◦C in the presence or absence of IL-2, IL-7, IL-15, or SCF (Miltenyi Biotec) for the indicated time.
Because CD56-upregulation by CD56dim NK cells from the third day of culture on would render discrimination of CD56brightfrom CD56dim by FACS less precise, cultures longer than 44 h were always performed with purified CD56brightNK cells.
FACS analysis and Phosflow staining
After staining with fluorochrome-conjugated antibodies, FACS analysis was performed using a Gallios (Beckman Coulter, Nyon, Switzerland) and FlowJo software. Sytox Blue (Invitrogen, Basel, Switzerland) gating was used to exclude dead cells. The following antibodies were used: PE labeled anti-CD56 (AF12–7H3) from Mil- tenyi, AlexaFluor488 labeled anti-pERK1/2 (20A), AlexaFluor647 labeled anti-pSTAT5 (47/Stat5pY694), and AlexaFluor647 labeled anti-pAkt (M89–61) were from BD biosciences, Allschwil, Switzerland, allophycocyanin labeled anti-CD117 (104D2), PE labeled anti-CD25 (BC96), BrillantViolet421 labeled anti-CD56 (HCD56) from Biolegend, Luzern, Switzerland and FITC labeled anti-CCR7 (150503) from R&D Systems, Abingdon, UK. For Phosflow-staining, purified NK cells or purified CD56brightNK cells were first incubated in medium with cytokines (see above) for 0–20 min at 37◦C. Equal volumes of pre-warmed BD Cytofix fixation buffer (BD Biosciences) was then added to the cell sus- pension and fixation was carried out at 37◦C for 15 min. Cells were then washed with PBS containing 0.1% BSA, incubated at 4◦C for 30 min in 200μL of ice-cold BD Phosflow Perm Buffer III (BD Biosciences) and washed before staining with antibod- ies at room temperature. Different phosphorylation states were expressed as geometric mean fluorescence intensity ratios calcu- lated by dividing the MFI of the stimulated sample by the MFI of the corresponding unstimulated sample.
NK cell degranulation (CD107a expression) was measured as published by Alter et al. [50] by stimulating NK cells that had been cultured for 20 h±SCF for 4 hours at 37◦C with K562 cells
at an E:T ratio of 1:1 in the presence of an APC labeled anti- CD107a antibody (H4A3 BD Biosciences). After 1 hour of culture 0.5μL/mL of Golgistop (BD Biosciences) was added. Thereafter, NK cells were washed and stained with BrillantViolet421 labeled anti-CD56 for 30 min at 4◦C for FACS analysis with gates on viable cells using Sytox Blue.
RNA isolation and RT-PCR
RNA isolation from cells cultured under the conditions described above was performed with the RNeasy Mini kit (Qiagen, Mag- den, Switzerland), according to the manufacturer’s instructions.
Complementary strand synthesis was performed with 0.1μg of isolated total RNA using 0.5μg random primers, 3mM MgCl2, 0.5mM dNTP, 20 U RNasin Ribonuclease inhibitor and 160U of ImProm II reverse transcriptase (Promega, D¨ubendorf, Switzer- land). Reversed transcription was carried out at 25◦C for 5 min, 42◦C for 60 min followed by 70◦C for 15 min. PCR was performed using QuantiTect SYBR Green PCR Kit and QuantiTect Primer Assays for c-kit, CCR7, CD25, and Tata Box-binding protein (TBP, Qiagen) at 95◦C for 10 min, followed by 40 cycles at 95◦C for 15 s and 60◦C for 1 min. The 2–CT method was used to analyze real-time PCR data. The expression of each gene was normalized against the housekeeping gene TBP.
NK adhesion to mSCF-GFP transfected COS-7 cells
Twenty-four hours after seeding on glass coverslips, COS-7 cells (American Type Culture Collection) were transiently transfected using Xtrem-9 (Roche, Indianapolis, IN, USA) with constructs con- taining the sequence of the membrane bound form of human SCF (mSCF) coupled to the sequence of Green Fluorescence Protein (mSCF-GFP) [51]. CD56bright NK cells were added (900 cells/mm2) and fixed after 1 h for 15 min in 4% paraformalde- hyde/PBS. After washing, phase contrast and GFP fluorescence images (filter set XF116; Omega, Brattelboro, VT) were taken with a PlanNeofluar 63×NA 1.4 oil immersion objective (Zeiss, Zurich, Switzerland) on a Zeiss-Axiovert100TV microscope (Zeiss) equipped with a 10-bit CCD camera (Hamamatsu Photonics, Tokyo, Japan), controlled by the Openlab software (PerkinElmer, Waltham, MA).
Statistics
We used GraphPad Prism software for statistical analysis com- puting statistical significance between groups by one-way ANOVA followed by a Tukey test to compare means of several experiments ort-tests when appropriate.p-values<0.05 were considered sig- nificant and are shown as *** whenp<0.001, ** whenp<0.01 and * whenp<0.05.
Acknowledgments: ER is supported by grants of the Swiss National Science Foundation and of the Swiss Cancer Research and by the “Dr. Henri Dubois-Ferri`ere-Dinu Lipatti” Foundation.
Conflict of interest: The authors declare no financial or commer- cial conflict of interest.
References
1 Lanier, L. L.,Le, A. M.,Civin, C. I.,Loken, M. R. and Phillips, J. H., The relationship of CD16 (Leu-11) and Leu-19 (NKH-1) antigen expression on human peripheral blood NK cells and cytotoxic T lymphocytes.J.
Immunol.1986.136: 4480–4486.
2 Nagler, A.,Lanier, L. L.,Cwirla, S. and Phillips, J. H., Comparative studies of human FcRIII-positive and negative natural killer cells.J. Immunol.
1989.143: 3183–3191.
3 Cooper, M. A.,Fehniger, T. A. and Caligiuri, M. A., The biology of human natural killer-cell subsets.Trends Immunol.2001.22: 633–640.
4 Cooper, M. A.,Fehniger, T. A.,Turner, S. C.,Chen, K. S.,Ghaheri, B.
A.,Ghayur, T.,Carson, W. E. et al., Human natural killer cells: a unique innate immunoregulatory role for the CD56(bright) subset.Blood2001.97:
3146–3151.
5 Colucci, F.,Caligiuri, M. A. and Di Santo, J. P., What does it take to make a natural killer?Nat. Rev. Immunol.2003.3: 413–425.
6 Fehniger, T. A.,Cooper, M. A.,Nuovo, G. J.,Cella, M.,Facchetti, F., Colonna, M. and Caligiuri, M. A., 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.Blood2003.
101: 3052–3057.
7 Ferlazzo, G.,Thomas, D.,Lin, S. L.,Goodman, K.,Morandi, B.,Muller, W. A.,Moretta, A. et al., The abundant NK cells in human secondary lymphoid tissues require activation to express killer cell Ig-like receptors and become cytolytic.J. Immunol.2004.172: 1455–1462.
8 Strowig, T.,Brilot, F. and Munz, C., Noncytotoxic functions of NK cells:
direct pathogen restriction and assistance to adaptive immunity. J.
Immunol.2008.180: 7785–7791.
9 Fauriat, C.,Long, E. O.,Ljunggren, H. G. and Bryceson, Y. T., Regula- tion of human NK-cell cytokine and chemokine production by target cell recognition.Blood2010.115: 2167–2176.
10De Maria, A.,Bozzano, F.,Cantoni, C. and Moretta, L., Revisiting human natural killer cell subset function revealed cytolytic CD56(dim)CD16+NK cells as rapid producers of abundant IFN-gamma on activation.Proc. Natl.
Acad. Sci. USA2011.108: 728–732.
11Romagnani, C.,Juelke, K.,Falco, M.,Morandi, B.,D’Agostino, A.,Costa, R.,Ratto, G. et al., CD56brightCD16-killer Ig-like receptor-NK cells dis- play longer telomeres and acquire features of CD56dim NK cells upon activation.J. Immunol.2007.178: 4947–4955.
12Campbell, J. J.,Qin, S.,Unutmaz, D.,Soler, D.,Murphy, K. E.,Hodge, M. R., Wu, L. et al., Unique subpopulations of CD56+NK and NK-T peripheral blood lymphocytes identified by chemokine receptor expression reper- toire.J. Immunol.2001.166: 6477–6482.
13Ferlazzo, G.,Pack, M.,Thomas, D.,Paludan, C.,Schmid, D.,Strowig, T., Bougras, G. et al., Distinct roles of IL-12 and IL-15 in human natural killer
Eur. J. Immunol. 2013. 00: 1–10 Innate immunity
9
cell activation by dendritic cells from secondary lymphoid organs.Proc.
Natl. Acad. Sci. USA2004.101: 16606–16611.
14Barreira da Silva, R. and Munz, C., Natural killer cell activation by den- dritic cells: balancing inhibitory and activating signals.Cell Mol. Life Sci.
2011.68: 3505–3518.
15Becknell, B. and Caligiuri, M. A., Interleukin-2, interleukin-15, and their roles in human natural killer cells.Adv. Immunol2005.86: 209–239.
16Nagler, 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.1990.171: 1527–1533.
17Matos, M. E.,Schnier, G. S.,Beecher, M. S.,Ashman, L. K.,William, D. E.
and Caligiuri, M. A., Expression of a functional c-kit receptor on a subset of natural killer cells.J. Exp. Med.1993.178: 1079–1084.
18Carson, W. E.,Haldar, S.,Baiocchi, R. A.,Croce, C. M. and Caligiuri, M. A., The c-kit ligand suppresses apoptosis of human natural killer cells through the upregulation of bcl-2.Proc. Natl. Acad. Sci. USA1994.91:
7553–7557.
19Carson, W. E.,Fehniger, T. A. and Caligiuri, M. A., CD56bright natural killer cell subsets: characterization of distinct functional responses to interleukin-2 and the c-kit ligand.Eur. J. Immunol.1997.27: 354–360.
20Benson, D. M., Jr.,Yu, J.,Becknell, B.,Wei, M.,Freud, A. G.,Ferketich, A. K.,Trotta, R. et al., Stem cell factor and interleukin-2/15 combine to enhance MAPK-mediated proliferation of human natural killer cells.
Blood2009.113: 2706–2714.
21Ashman, L. K., The biology of stem cell factor and its receptor C-kit.Int.
J. Biochem. Cell Biol.1999.31: 1037–1051.
22Reber, L.,Da Silva, C. A. and Frossard, N., Stem cell factor and its receptor c-Kit as targets for inflammatory diseases.Eur. J. Pharmacol.2006.533:
327–340.
23Kinashi, T. and Springer, T. A., Steel factor and c-kit regulate cell-matrix adhesion.Blood1994.83: 1033–1038.
24Lorentz, A.,Schuppan, D.,Gebert, A.,Manns, M. P. and Bischoff, S.
C., Regulatory effects of stem cell factor and interleukin-4 on adhe- sion of human mast cells to extracellular matrix proteins.Blood2002.99:
966–972.
25Mierke, C. T.,Ballmaier, M.,Werner, U.,Manns, M. P.,Welte, K. and Bischoff, S. C., Human endothelial cells regulate survival and prolifera- tion of human mast cells.J. Exp. Med.2000.192: 801–811.
26Lukacs, N. W.,Kunkel, S. L.,Strieter, R. M.,Evanoff, H. L.,Kunkel, R. G., Key, M. L. and Taub, D. D., The role of stem cell factor (c-kit ligand) and inflammatory cytokines in pulmonary mast cell activation.Blood1996.
87: 2262–2268.
27Yu, T. K.,Caudell, E. G.,Smid, C. and Grimm, E. A., IL-2 activation of NK cells: involvement of MKK1/2/ERK but not p38 kinase pathway.J. Immunol.
2000.164: 6244–6251.
28Yu, C. R.,Young, H. A. and Ortaldo, J. R., Characterization of cytokine differential induction of STAT complexes in primary human T and NK cells.J. Leukoc. Biol.1998.64: 245–258.
29Jiang, K.,Zhong, B.,Ritchey, C.,Gilvary, D. L.,Hong-Geller, E.,Wei, S.
and Djeu, J. Y., Regulation of Akt-dependent cell survival by Syk and Rac.
Blood2003.101: 236–244.
30Pillet, A. H.,Bugault, F.,Theze, J.,Chakrabarti, L. A. and Rose, T., A programmed switch from IL-15- to IL-2-dependent activation in human NK cells.J. Immunol.2009.182: 6267–6277.
31Kapur, R.,Chandra, S.,Cooper, R.,McCarthy, J. and Williams, D. A., Role of p38 and ERK MAP kinase in proliferation of erythroid progenitors in response to stimulation by soluble and membrane isoforms of stem cell factor.Blood2002.100: 1287–1293.
32Matsui, J.,Wakabayashi, T.,Asada, M.,Yoshimatsu, K. and Okada, M., Stem cell factor/c-kit signaling promotes the survival, migration, and capillary tube formation of human umbilical vein endothelial cells.J.
Biol. Chem.2004.279: 18600–18607.
33Boothby, M.,Mora, A. L. and Stephenson, L. M., Lymphokine-dependent proliferation of T-lymphoid cells: regulated responsiveness and role in vivo.Crit. Rev. Immunol.2001.21: 487–522.
34Wofford, J. A.,Wieman, H. L.,Jacobs, S. R.,Zhao, Y. and Rathmell, J. C., IL-7 promotes Glut1 trafficking and glucose uptake via STAT5-mediated activation of Akt to support T-cell survival.Blood2008.111: 2101–2111.
35Michaud, A.,Dardari, R.,Charrier, E.,Cordeiro, P.,Herblot, S. and Duval, M., IL-7 enhances survival of human CD56bright NK cells.J. Immunother.
2010.33: 382–390.
36Babina, M.,Rex, C.,Guhl, S.,Thienemann, F.,Artuc, M.,Henz, B. M.
and Zuberbier, T., Baseline and stimulated turnover of cell surface c-Kit expression in different types of human mast cells.Exp. Dermatol.2006.
15: 530–537.
37Alileche, A.,Goldman, C. K. and Waldmann, T. A., Differential effects of IL-2 and IL-15 on expression of IL-2 receptor alpha.Biochem. Biophys. Res.
Commun.2001.285: 1302–1308.
38Chiesa, M. D.,Vitale, M.,Carlomagno, S.,Ferlazzo, G.,Moretta, L. and Moretta, A., The natural killer cell-mediated killing of autologous den- dritic cells is confined to a cell subset expressing CD94/NKG2A, but lack- ing inhibitory killer Ig-like receptors.Eur. J. Immunol.2003.33: 1657–1666.
39Ferlazzo, G. and Munz, C., NK cell compartments and their activation by dendritic cells.J. Immunol.2004.172: 1333–1339.
40Carson, W. E.,Giri, J. G.,Lindemann, M. J.,Linett, M. L.,Ahdieh, M., Paxton, R.,Anderson, D. et al., Interleukin (IL) 15 is a novel cytokine that activates human natural killer cells via components of the IL-2 receptor.
J. Exp. Med.1994.180: 1395–1403.
41Wang, X.,Lupardus, P.,Laporte, S. L. and Garcia, K. C., Structural biology of shared cytokine receptors.Annu. Rev. Immunol.2009.27: 29–60.
42Dalbeth, N.,Gundle, R.,Davies, R. J.,Lee, Y. C.,McMichael, A. J. and Callan, M. F., CD56bright NK cells are enriched at inflammatory sites and can engage with monocytes in a reciprocal program of activation.J.
Immunol.2004.173: 6418–6426.
43Ottaviani, C.,Nasorri, F.,Bedini, C.,de Pita, O.,Girolomoni, G. and Cavani, A., CD56brightCD16(-) NK cells accumulate in psoriatic skin in response to CXCL10 and CCL5 and exacerbate skin inflammation.Eur. J. Immunol.
2006.36: 118–128.
44Ansari, M.,Vukicevic, M.,Rougemont, A. L.,Moll, S.,Parvex, P.,Gumy- Pause, F.,Chalandon, Y. et al., Do NK cells contribute to the patho- physiology of transplant-associated thrombotic microangiopathy?Am.
J. Transplant.2011.11: 1748–1752.
45Freud, A. G., Yokohama, A., Becknell, B., Lee, M. T., Mao, H. C., Ferketich, A. K. and Caligiuri, M. A., Evidence for discrete stages of human natural killer cell differentiation in vivo.J. Exp. Med.2006.203:
1033–1043.
46Huntington, N. D.,Legrand, N.,Alves, N. L.,Jaron, B.,Weijer, K.,Plet, A.,Corcuff, E. et al., IL-15 trans-presentation promotes human NK cell development and differentiation in vivo.J. Exp. Med.2008.206: 25–34.
47Vukicevic, M.,Chalandon, Y.,Helg, C.,Matthes, T.,Dantin, C.,Huard, B., Chizzolini, C. et al., CD56-bright NK cells after hematopoietic stem cell transplantation are activated mature NK cells that expand in patients with low numbers of T cells.Eur. J. Immunol.2010.40: 3246–3254.
48Chklovskaia, E.,Nowbakht, P.,Nissen, C.,Gratwohl, A.,Bargetzi, M. and Wodnar-Filipowicz, A., Reconstitution of dendritic and natural killer cell subsets after allogeneic stem cell transplantation: effects of endogenous flt3 ligand.Blood2004.103: 3860.
49Jin, Y., Damaj, B. B. and Maghazachi, A. A., Human resting CD16-, CD16+and IL-2-, IL-12-, IL-15- or IFN-alpha-activated natu- ral killer cells differentially respond to sphingosylphosphorylcholine, lysophosphatidylcholine and platelet-activating factor.Eur. J. Immunol.
2005.35: 2699–2708.
50Alter, G.,Malenfant, J. M. and Altfeld, M., CD107a as a functional marker for the identification of natural killer cell activity.J. Immunol. Methods 2004.294: 15–22.
51Wehrle-Haller, B. and Imhof, B. A., Stem cell factor presentation to c-Kit.
Identification of a basolateral targeting domain.J. Biol. Chem.2001.276:
12667–12674.
Full correspondence: Prof. Eddy Roosnek PhD, Division of Hematology, Department of Internal Medicine, Geneva University Hospitals and University of Geneva, Switzerland
Fax:+41-22-372-72-88 e-mail: eddy.roosnek@unige.ch
Received: 3/7/2013 Revised: 2/9/2013 Accepted: 14/10/2013
Accepted article online: 22/10/2013
