Published online before print May 22, 2003
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
* School of Cell and Molecular Biosciences, The Medical School, Newcastle, United Kingdom; and
Department of Cytokine Biology, ZymoGenetics Inc., Seattle, Washington
Correspondence: Dr. Colin G. Brooks, School of Cell and Molecular Biosciences, The Medical School, Newcastle NE2 4HH, UK. E-mail: colin.brooks{at}newcastle.ac.uk
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
chains, but we show here that immature and mature mouse NK cells express IL-2R chain mRNA and that low levels of IL-2R chains can be detected on the surface of immature and mature NK cells provided they are cultured in the absence of IL-2. Despite their potential expression of high-affinity IL-2 receptors, immature NK cells only proliferate if IL-2 is present at extremely high concentrations. Surprisingly, IL-15 can also only support the growth of immature NK cells at high, presumably nonphysiological concentrations. Although NK cells express mRNA for the high-affinity IL-15R chain, they also express a variety of alternately spliced transcripts whose protein products could potentially disrupt signaling through IL-15 receptors. The requirement for high concentrations of IL-2 and IL-15 suggests that if these cytokines play any role in the proliferative expansion of NK cells in vivo, they act indirectly via other cells or in cooperation with other factors. In support of the latter possibility, we report that the recently described cytokine IL-21 can markedly enhance the proliferation of immature (and mature) NK cells in the presence of doses of IL-2 and IL-15 that by themselves have little growth-promoting activity.
Key Words: rodent • cellular proliferation • differentiation
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
c have moderate numbers of B cells and T cells but no detectable NK cells [1
, 2
]. c Forms part of the receptors for several cytokines, including interleukin (IL)-2, IL-4, IL-7, IL-9, IL-15 [3
], the recently discovered IL-21 [4
], and possibly other as-yet unknown cytokines and signaling molecules, implying that one or more of these c-containing receptors plays a critical role in NK cell development. The ability of IL-2 to promote the growth and activation of mature NK cells makes IL-2 an attractive candidate. Indeed, numerous in vitro studies have shown that IL-2 can promote the development of NK cells from immature cells obtained from human [5
, 6
] and mouse [7
8
9
] bone marrow (BM) and from human [10
] and mouse [11
, 12
] fetal liver and thymus. Furthermore, IL-2-deficient humans [13
] and mice [14
] have markedly reduced levels of NK cells and NK cell functions, and IL-2R [15
]- and IL-2Rß [16
]-deficient mice have greatly reduced numbers of NK cells. However, IL-2 and IL-2R deficiency causes a general disturbance of lymphocyte homeostasis that could indirectly affect NK cell numbers and activity. In addition, it is now known that the IL-2Rß chain participates in the formation of receptors for another cytokine IL-15. In conjunction with c, IL-2Rß forms part of a low-affinity receptor that binds IL-15 with an association constant estimated at between 270 pM and 2.5 nM [17
18
19
20
]. A high-affinity receptor, involving the participation of the IL-15R chain, has also been identified, which binds IL-15 with an affinity between 12 and 200 pM [17
18
19
20
21
22
]. Evidence that IL-15 might be involved in NK cell development in vivo was initially suggested by the finding that mice lacking the transcription factor interferon (IFN)-regulatory factor 1, which is required for IL-15 production, had a deficiency of NK cells that could be overcome, at least in vitro, by soluble IL-15 [23
]. More directly, IL-15 [24
]- and IL-15R [25
]-deficient mice were subsequently shown to lack readily detectable NK cells. One explanation for these findings is that IL-15 directly promotes the growth and differentiation of NK cell progenitors via high-affinity IL-15 receptors expressed on NK cells. Several studies have shown that IL-15 can indeed promote the development of NK cells from human [26
] and mouse [27
28
29
] progenitors in vitro. However, in the absence of suitable antibodies against the IL-15R chain, there is currently no direct evidence that NK cells or their progenitors express IL-15R at the cell surface. A prediction of this hypothesis would be that the growth and differentiation of immature NK cells would occur at doses of IL-15 sufficient to saturate high-affinity IL-15 receptors. In the present study we report that although immature mouse NK cells can indeed proliferate vigorously and differentiate in response to soluble IL-15, they do so only at doses much higher that those that would be required to saturate high-affinity receptors. Adult mouse NK cells also cannot proliferate or be activated by low doses of soluble IL-15 alone, nor surprisingly can mouse T cells. These findings raise profound questions concerning the exact role of IL-15 in promoting NK cell development and T cell homeostasis. One possibility is that efficient interaction of IL-15 with NK cells requires the participation of other cell-bound or soluble factors. In support of this, we report here that low doses of IL-21 enhance the responsiveness of immature and mature NK cells to suboptimal doses of IL-15 and IL-2. |
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Culture media and reagents
Cells were cultured in a 10% CO2 atmosphere at 37°C in Dulbecco’s modified Eagle’s medium (52100-039; Life Technologies, Paisley, UK) made up in highly purified water and supplemented with 2x nonessential amino acids, 5 x 10-5 M 2-mercaptoethanol, and 10% fetal bovine serum (FBS; F-7524; Sigma, Poole, UK). Mouse recombinant (mr)IL-4 was obtained as the supernatant of x6310 cells transfected with mIL-4 cDNA [30
], kindly provided by Professor F. Melchers (Basel Institute for Immunology, Switzerland). Unitage was determined by titration on CTLL2 cells [31
]. Human (h)rIL-2 was obtained from Cetus (Emeryville, CA). mrIL-2, mrIL-15, and hrIL-15 were obtained from Peprotech (Rocky Hill, NJ). mrIL-21 was prepared as described previously [32
].
Cells
Adult NK cells were purified from C57BL/6 spleens as described previously [33
]. Following expansion in IL-2, >98% of cells were NK1.1+CD3-. Fetal thymocytes were prepared from the day 14 embryos of timed-mated C57BL/6 mice (day of vaginal plug=0), cultured for 2–3 days in medium containing 10 U/ml IL-4 and 10 ng/ml phorbol 12-myristate 13-acetate (PMA; P8139; Sigma), and were then transferred to medium containing the test cytokines. Clones were obtained by limiting dilution in 96-well plates at the time of transfer to IL-2. The established NK cell clones 1608b and I2/22 had been maintained in continuous culture for >1 year in medium containing 20 nM IL-2. Approximately 1 month before their use in experiments, sublines of these were set up and maintained in parallel in 200 pM IL-2.
Mouse CD4 and CD8 T cells were obtained by first depleting CD8 T cells with the monoclonal antibody (mAb) 3.168 (kindly provided by Professor F. Fitch, University of Chicago, IL) or depleting CD4 T cells with the mAb RL172.4 (kindly provided by Professor H. R. MacDonald, Ludwig Institute, Epalinges, Switzerland) and normal rabbit serum. Aliquots of 1 x 106 cells were then cultured in 24-well plates in medium containing 2 µg/ml concanavalin A (Con A; Sigma) for 1 day and washed, and aliquots of 
2 x 105 cells were set up in 24-well plates containing 1 ml/well fresh medium with appropriate cytokines. At the end of the culture period, the CD4 T cell population typically comprised >85% CD4+CD3+ cells, and the CD8 T cell population typically comprised >95% CD8+CD3+ cells. CTLL2 cells [31
] were grown continuously in medium containing 60 pM IL-2.
Human peripheral blood mononuclear cells (PBMC) were prepared by density-step separation on Histopaque (Sigma). The growth of NK cells was measured by incubating aliquots of 1 x 106 whole PBMC in 24-well plates containing 1 ml/well medium with appropriate cytokines. Cultures were refed and/or subcultured twice per week for 14 days, and then the numbers of cells were determined and their composition analyzed by immunofluorescence staining as described below. Typically, 80% of cells were CD56/CD16+CD3-. The growth of T cells was measured by incubating aliquots of whole PBMC in 24-well plates containing 1 ml/well medium with 5 µg/ml Con A. Cells were then washed and replated at 2 x 105/well in medium containing appropriate cytokines. After 7 days, cell numbers were determined, and their composition was analyzed. Typically, >90% of cells were CD56/CD16- CD3+.
Cell growth assay
At the end of the culture period, the concentration of cells in cultures was determined by adding a known number of FlowCount beads (Beckman-Coulter, Miami, FL), running the mixture through a FACScan (Becton Dickinson, San Jose, CA), and determining the ratio of beads to viable cells using forward- and side-scatter characteristics. In the case of human cells, the number of NK cells and T cells was calculated by multiplying the total number of cells by the proportion that was NK or T cells as determined by immunofluorescence. All cultures were set up in triplicates.
Immunofluorescence and flow cytometry
Aliquots of 2 x 105 cells were incubated at room temperature with appropriate combinations of reagents in Hanks’ balanced saline solution (61200-093; Life Technologies) containing no bicarbonate and supplemented with 2% FBS and 0.2% sodium azide, except for staining with Qa1 tetramers that require incubation at 37°C [34
]. Staining was analyzed on a FACScan using forward- and side-scatter to gate on single viable cells. Compensation for spectral overlap of dyes was set by running mixtures of unstained cells and cells stained with each fluorochrome singly. To permit comparison between the levels of staining in different experiments, the same reagent stocks and FACScan acquisition parameters were used throughout. Consistency was confirmed by the finding that the median fluorescence level of control beads (FluoroSpheres; Dako, Glostrup, Denmark) did not vary by more than 10% between experiments. Data were collected using Lysis II software (Becton Dickinson), converted to a PC format using Lifutil, analyzed using FCS Express V2 software, and compiled in Microsoft Excel.
The purity and phenotype of mouse NK cells were determined using the following reagents: fluorescein isothiocyanate (FITC) PK136 anti-NK1.1 (BD PharMingen, San Diego, CA); FITC 18d3 anti-CD94 (kindly provided by Professor D. Raulet, University of California, Berkeley); biotinylated Qa1b–Qdm tetramers, refolded using human ß2-microglobulin (National Institutes of Health Tetramer Facility, Atlanta, GA) and assembled with Red670-streptavidin (InVitrogen, Carlsbad, CA); KT3 anti-CD3 (kindly provided by Professor E. Simpson, Imperial College, London, UK); A1 anti-Ly49A (kindly provided by Professor J. Allison, University of California, Berkeley); 5E6 anti-Ly49C/I (BD PharMingen); 4D11 anti-Ly49G (kindly provided by Dr. L. Mason, National Cancer Institute, Frederick, MD); 4D12 anti-Ly49C/E (kindly provided by Dr. G. Leclercq, University of Ghent, Belgium); or 10A7 anti-NKRP1 (kindly provided by Prof. V. Kumar, University of Chicago, IL), followed by Alexa Fluor 488-conjugated goat anti-rat immunoglobulin G (IgG) or goat anti-mouse IgG (Molecular Probes, Eugene, OR) as appropriate. IL-2 receptors on mouse NK cells were identified by staining with biotinylated 7D4 anti-IL-2R (BD PharMingen) followed by Alexa Fluor 647-streptavidin (Molecular Probes) or TMß1 anti-IL-2Rß (kindly provided by Dr. T. Tanaka, University of Tokyo, Japan) + Alexa Fluor 488 goat anti-rat IgG. Human NK and T cells were identified by staining with Simultest NK reagent (Becton Dickinson), which contains a cocktail of phycoerythrin (PE) anti-CD56, PE anti-CD16, and FITC anti-CD3 mAb.
Cytotoxicity assays
These were performed in a standard manner by incubating serial dilutions of effector cells for 4 h in V-bottomed microtest plates with 5000 51Cr-labeled YAC-1 or blast target cells. The latter were prepared from frozen CD8-depleted spleen cells of C57BL/6 mice and mice homozygous for the ß2m knockout mutation on a C57 background, kindly provided by Professor E. Jenkinson (University Birmingham, UK). Thawed spleen cells were cultured for 2–3 days in medium containing 2 µg/ml Con A and 200 pM IL-2.
Reverse transcriptase-polymerase chain reaction (RT-PCR)
RNA was prepared using RNAzol (CS-104, Biogenesis, Poole, UK), according to the manufacturer’s instructions. cDNA was prepared by incubating 20 µl denatured (70°C/10 min) RNA at 150 µg/ml with 500 U/ml RNasin (Promega, Madison, WI), 0.5 mM dNTPs, 5 µM oligo dT, and 2500 U/ml MMV H- RT (Promega), according to the manufacturer’s instructions. For PCR, 20 µl reactions containing 20 U/ml Taq polymerase (Bioline, Randolf, MA) in the manufacturer’s buffer, 2 mM dNTPs, 3 mM Mg, and cDNA corresponding to known numbers of cells were incubated with the following primer pairs for 1 min at 95°C followed by 40 cycles of 95°C/30 s, 58°C/30 s, and 72°C/60 s: IL-2R, forward ATGTGCCAGGAAGATGG, reverse CTAGATGGTTCTTCTGCTC; IL-2Rß, forward GGTTGGCGTAGGGTAAAGAC, reverse AGGGGACAGGCGAGGAGAGC; IL-2R, forward CTCCTACTCTGCCCCTTCCA, reverse TCCATTTACTCCACTGTTGA; IL-15R isoform 1, 5'-untranslated region (UTR), forward CTTGCGTCCCGTTGGGTC; IL-15R isoforms 1 and 2 internal, forward TCTCCCCACAGTTCCAAAAT; IL-15R isoform 2, 5'-UTR, forward GAAAAGGGAGATCGCCGGCTT; IL-15R isoforms 1 and 2, 3'-UTR, reverse GGCACCCAGGCTCAGTAAAA. Aliquots of PCR reactions were run on 1% agarose gels containing ethidium bromide. PCR products were purified from gels using Qiagen (Hilden, Germany) gel extraction kits and cloned into pCR4-TA (InVitrogen), according to the manufacturers’ instructions. Plasmids were prepared from individual colonies according to standard methods, purified on Minelute columns (Qiagen), and sequenced using M13 forward and reverse primers.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
20 nM, suggesting that IL-2 may be acting as an inefficient surrogate of some other cytokine, which more efficiently supports the expansion of NK precursors in vivo. One cytokine often considered as a candidate for such a role is IL-15. In several studies [26
27
28
29
], IL-15 has been found to support the development of immature NK cells in vitro, but no detailed quantitation of this phenomenon has been reported previously. As shown in Figure 1A
, although m- and hIL-15 can support the rapid proliferation of immature mouse NK cells in vitro, like IL-2 they do so only at nanomolar concentrations (Fig. 1A)
. Nanomolar concentrations of IL-15 are also required to support vigorous proliferation of adult splenic NK cells (Fig. 1B)
and normal mouse CD4 and CD8 T cells (Fig. 1C
and 1D) , despite the fact that the latter cells respond to picomolar concentrations of IL-2 (50% maximal proliferation at 20–200 pM). By contrast, all four cytokines induce proliferation of the mouse T cell line CTLL2 at low concentrations, 50% maximal proliferation occurring with 2 pM hIL-2, mIL-2, and hIL-15 and with 200 pM mIL-15 (Fig. 1E)
. Interestingly, although freshly derived immature and mature NK cells proliferate only with high concentrations of IL-2 and IL-15, established mouse NK cell lines such as 1608b and I2/22 can respond well to low concentrations of IL-2, especially if maintained for some time in a low dose of IL-2 (Fig. 1F
1G
1H
1I)
. However, the heightened responsiveness to IL-2 of established NK lines is not accompanied by an equally heightened responsiveness to IL-15. In contrast to the situation in the mouse, freshly prepared human NK cells responded relatively efficiently to hIL-2, 50% maximal proliferation occurring at doses of IL-2 (50–100 pM), similar to those required for the proliferation of human T cells (Fig. 1J
and 1K)
. In addition, human NK cells and T cells also responded comparatively efficiently to IL-15, although the amounts of IL-15 required were noticeably higher than the amounts of IL-2.
 View larger version (44K): [in a new window] |
Figure 1. Growth of NK cells and T cells in IL-2 and IL-15. (A–E) Mouse immature and mature NK cells, CD4 and CD8 T cells, and CTLL2 cells were cultured for 3–5 days in various concentrations of m- or hIL-2 and -IL-15. (F–I) The long-term mouse NK cell lines 1608b and I2/22 were maintained in parallel in 20 nM or 0.2 nM IL-2 for 1 month and were then tested in a 3-day assay for growth in various concentrations of hIL-2 and mIL-15. (J) Human PMBC were cultured for 14 days in various concentrations of hIL-2 or hIL-15, and then the growth of NK cells was calculated from the total cell numbers per culture and the percentage of these that were CD16/CD56+ CD3-. (K) PBMC were activated for 1 day with 5 µg/ml Con A, washed, and cultured in hIL-2 or mIL-15 for 7 days, and then the growth of T cells was calculated from the total cell numbers per culture and the percentage of these that were CD16/CD56- CD3+. The data shown are representative of at least three experiments with each cell type.
|
20 nM IL-2 or IL-15 (Fig. 2
).
 View larger version (21K): [in a new window] |
Figure 2. Induction of cytotoxic activity by IL-2 and IL-15. Triplicate cultures containing 0.5 million spleen cells were incubated with 0.2 nM ( ), 2 nM (•), or 20 nM ( ) hIL-2 or mIL-15. Three days later, cells were washed and incubated at various dilutions with YAC-1 targets. The effector:target (E:T) ratios shown are based on the initial number of spleen cells. Fresh, uncultured spleen cells gave 5% cytotoxicity at an E:T ratio of 100:1, and spleen cells cultured for 3 days without any cytokine showed no detectable cytotoxicity (data not shown).
|
 View larger version (42K): [in a new window] |
Figure 3. IL-15 induces the differentiation of NK cells in an identical manner to IL-2. (A) Expression of NK1.1 and Qa1 receptors on day-14 fetal thymocytes that had been cultured for 2 days in IL-4 and PMA and subsequently for 1 or 3 days in 20 nM hIL-2 or mIL-15. (B) Expression of the Ly49 molecules recognized by the mAb 4D12, of CD94 molecules, and of the NKRP1 molecules recognized by the mAb 10A7 following 5 days and 20 days of culture in 20 nM IL-2 or IL-15.
|
c mRNA, they also express readily detectable levels of IL-15R mRNA and IL-2R mRNA transcripts. IL-15R and IL-2R transcripts were present at all stages during the in vitro development of NK cells from immature progenitors (Fig. 4B)
. Using primers that bind to the 5' and 3' UTRs of IL-15R transcripts, complex banding patterns were obtained with RNA from NK cells and T cells, indicating extensive, alternative splicing of IL-15R primary transcripts (Fig. 4A)
. Amongst the novel transcripts found in clones of immature NK cells was one in which the use of a cryptic splice site in exon 4 would potentially generate a form of the IL-15R chain (termed isoform 1A), lacking the first 33 amino acids of the Pro/Thr-rich membrane proximal domain (Fig. 4C)
. Another transcript was found that lacked the whole of exon 4 and would potentially generate a protein (termed isoform 1B) having no Pro/Thr-rich domain. A third transcript lacked exons 3 and 4 and would potentially generate a protein (termed isoform 1C) whose extracellular region comprised only the N-terminal Sushi domain. In addition, NK cells (and T cells) contained transcripts of isoform 2 of the IL-15R chain (GenBank NM133836) whose presence in lymphoid cells has not been described previously. Comparison with the recently obtained genomic sequence (GenBank AL831794.6) indicates that this isoform arises from splicing events that introduce an alternate first exon, which we have termed exon -1, located 250 bases upstream of exon 1. This exon lacks an initiator ATG codon and would potentially generate a greatly foreshortened IL-15R chain lacking the normal leader sequence, the Sushi domain, and linker domain (Fig. 4C)
. Isoform 2 was also subject to extensive, alternate splicing, resulting in the loss of one or more internal exons (Fig. 4A
; sequences deposited in GenBank).
 View larger version (40K): [in a new window] |
Figure 4. Expression of mRNA for IL-15 and IL-2 receptor chains in immature and mature NK cells. (A) cDNA from equal numbers of cells of each type was amplified using internal primers for IL-2/IL-15Rß and c or "full-length" primers for IL-15R isoform (Iso)1, IL-15R isoform 2, and IL-2R. RAGko, RAG2 knockout. (B) cDNA from equal numbers of cells at different stages of NK development was amplified with internal primers for IL-15R, IL-2Rß, and c and with full-length primers for IL-2R. FTC, feral thymocytes. (C) Diagrammatic representation of alternate splicing of IL-15R transcripts. The upper part of each diagram shows the exonic structure of the transcripts detected in this study, and open triangles show the position of the primers and bent arrows, the translational start sites. The lower part of each diagram shows the putative polypeptide with domains designated according to Giri et al. [22
]: line, leader sequence; upward diagonal shading, "Sushi domain"; checkered shading, linker domain; downward diagonal shading, Pro/Thr-rich domain; solid shading, transmembrane domain; gray shading, cytoplasmic domain. The isoform 1 sequences have been deposited in Genbank under accession numbers AY219715–717, and the isoform 2 sequences under accession numbers AY221616–619.
|
chains are expressed on the surface of NK cells and if so, in what form cannot currently be answered, as no suitable antibodies are available. However, although IL-2R chains were not detectable on freshly derived immature or mature NK cells grown in IL-2, using a sensitive-staining technique IL-2R chains could be detected on cells grown in IL-15, albeit at levels less than one-tenth of those for CD4 and CD8 T cells and less than one-hundredth of those for CTLL2 cells (Fig. 5
). By contrast, all these cells expressed similar quantities of IL-2/IL-15Rß chains. Established NK cell lines grown in 200 pM IL-2 expressed much higher levels of IL-2R chains than did freshly derived NK cells, even when the latter were grown in IL-15.
 View larger version (40K): [in a new window] |
Figure 5. Expression of IL-2R and IL-2/IL-15Rß chains on immature and mature NK cells. The following cells were stained with medium (dotted lines) and anti-IL-2R or anti-IL-2/IL-15Rß chain mAb (solid lines), followed by the appropriate second layer reagent as described in Materials and Methods. (A) Immature NK cells that had been grown for 3 days in IL-4 + PMA and subsequently for 5 days in 20 nM hIL-2. (B) The same cells as in A but grown for 5 days in 20 nM mIL-15. (C) Purified adult splenic NK cells grown for 5 days in 20 nM IL-2. (D) The same cells grown in 20 nM IL-15. (E) The established NK clone 1608b grown in 200 pM IL-2. (F) The established NK clone I2/22 grown in 200 pM IL-2. (G) CD4-depleted spleen cells grown for 5 days in 200 pM IL-2 (98% CD3+, 85% CD4+, 5% CD8+). (H) CD8-depleted spleen cells grown for 5 days in 200 pM IL-2 (100% CD3+, 0% CD4+, 100% CD8+). (I) CTLL2 cells grown continuously in 60 pM IL-2.
|
, a series of short-term reculture experiments was performed. When immature NK cells that had been grown in 20 nM IL-15 for 5 days were washed and reincubated in 20 nM IL-2, the proportion of cells showing high-surface staining for IL-2R chains declined from 26% to 7% within 3 h (Fig. 6A
). No such decline occurred when cells were incubated in medium or in 20 nM IL-15, but a similar decline occurred when they were incubated in a mixture of IL-2 and IL-15, demonstrating that the effect of IL-2 was dominant. None of these changes in IL-2R expression occurred when the incubations were conducted at ice temperature (data not shown). This, together with the observation that incubation of CTLL2 cells with 20 nM IL-2 at 37°C did not affect staining with anti-IL-2R mAb (not shown) argue strongly that the effect of IL-2 was to down-regulate the expression of IL-2R chains on the surface of NK cells rather than to merely block the binding of mAb. A prediction of this hypothesis would be that if immature NK cells grown in IL-2 were washed and incubated in medium alone or IL-15, there would be a rapid increase in IL-2R staining. Such an increase was indeed observed (Fig. 6B)
. Importantly, 200 pM IL-2 was as effective as 20 nM IL-2 in down-regulating surface IL-2R chain expression (Fig. 6A
and 6B)
, indicating that most of the IL-2R chains expressed on the surface of NK cells are integrated into high-affinity receptor complexes that are rapidly internalized upon binding IL-2. This raised the possibility that low concentrations of IL-2 might enhance the growth of NK cells in the presence of IL-15. However, no such synergy between IL-2 and IL-15 could be detected (Fig. 7
).
 View larger version (35K): [in a new window] |
Figure 6. IL-2 modulates the expression of IL-2R chains on immature NK cells. (A) Immature NK cells that had been grown for 3 days in IL-4 + PMA and subsequently for 5 days in 20 nM mIL-15 were washed and placed in medium alone or medium containing 20 nM hIL-2, 200 pM hIL-2, 20 nM mIL-15, or 20 nM hIL-2 + 20 nM mIL-15 and were stained with medium (dotted lines) or anti-IL-2R mAb (solid lines) followed by second-layer reagent immediately or following 3 h incubation at 37°C. The percentage of cells expressing high levels of IL-2R chains is indicated. (B) As in A, but cells were precultured for 5 days in 20 nM hIL-2.
|
 View larger version (19K): [in a new window] |
Figure 7. Lack of synergism between IL-2 and IL-15 in promoting the growth of immature NK cells. Following exposure to IL-4 + PMA for 2 days, immature NK cells were incubated for 3 days in medium containing 0, 0.2, 2, or 20 nM hIL-2 in the presence of 0, 0.2, 2, or 20 nM mIL-15.
|
/ß, IFN-, tumor necrosis factor , and transforming growth factor-ß, were tested, and although several of these cytokines could inhibit the growth of NK cells in the presence of IL-2 or IL-15, none could enhance it. By contrast, the recently discovered cytokine IL-21, which can promote the growth of human NK cells from immature progenitors [32
], displayed a remarkable, biphasic effect on the growth of mouse NK cells: At high doses (2 nM), it generally inhibited growth, whereas at low doses (approximately 20 pM), it could strongly enhance the growth of immature NK cells (Fig. 8A
) and mature splenic NK cells (data not shown), and the enhancing effect was much more pronounced at limiting doses of IL-2 and IL-15. IL-21 by itself had no ability to induce the growth of immature or mature NK cells. IL-21 also enhanced the cytolytic activity of immature and mature NK cells cultured in IL-2 but had no effect on the acquisition of mature NK cell markers by developing NK cells (data not shown). In particular, cells grown in the presence of IL-21 acquired NK1.1, CD94, and the Ly49 molecules recognized by the 4D12 mAb in a similar manner to that shown in Figure 3
but did not acquire Ly49A, C, I, or G. In contrast to its effects on NK cells, IL-21 had little effect on the growth of CD4 or CD8 T cell blasts at any dose of IL-2 tested (Fig. 8B
and 8C)
.
 View larger version (30K): [in a new window] |
Figure 8. IL-21 can enhance the growth of NK cells in the presence of limiting doses of IL-2 or IL-15. (A) Following exposure to IL-4 + PMA for 2 days, immature NK cells were incubated for 3 days in medium containing various concentrations of hIL-2 or mIL-15 together with titrated doses of mIL-21. (B) CD4-depleted spleen cells were incubated for 1 day with Con A, then washed, and cultured for 3 days in medium containing various concentrations of hIL-2 together with titrated doses of mIL-21. (C) CD8-depleted spleen cells were incubated for 1 day with Con A, then washed, and cultured for 3 days in medium containing various concentrations of hIL-2 together with titrated doses of mIL-21.
|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
, it is not possible to determine directly whether normal NK cells and T cells express IL-15R chains at the cell surface. RT-PCR followed by cloning and sequencing demonstrated unambiguously that highly purified mouse NK cells and T cells contained full-length (isoform 1) transcripts of IL-15R chains. However, these procedures also revealed the presence of a series of alternately spliced IL-15R transcripts. Three of these transcripts (isoforms 1A, 1B, and 1C) potentially encode truncated forms of the IL-15R chain lacking parts or all of the membrane-proximal, extracellular domains but retaining the N-terminal Sushi domain that binds IL-15 [38
]. NK cells also expressed IL-15R transcripts that potentially encode a form of the IL-15R chain, termed isoform 2, which lacks the N-terminal 140 amino acids of isoform 1. Isoform 2 transcripts were also subject to extensive alternate splicing. Thus, even if NK cells and T cells express IL-15R chains on their surfaces, they may also express additional forms of the IL-15R polypeptide that alter or interfere with signal transduction.
The situation regarding IL-2 and IL-15 dosimetry in man is less clear-cut. Human NK cells and T cells have been reported to express high-affinity receptors for IL-15 with association constants of 12–58 pM [17
, 18
, 20
, 21
] and low-affinity receptors with association constants of 0.5–2.5 nM [17
, 18
, 20
]. Our studies showed that human NK cells and T cells responded to IL-15 with 50% maximal responses at 0.2–2 nM, in good agreement with other reports [17
18
19
20
21
]. These doses are much lower than those required to support the proliferation of mouse NK cells and T cells but exceed those that would be needed to occupy most high-affinity IL-15 receptors. By contrast, the doses of IL-2, which were required for 50% maximal proliferation of human NK and T cells (50–100 pM), corresponded closely to the measured association constants of high-affinity IL-2 receptors on NK cells [18
] and T cells [20
, 21
]. Furthermore, the proliferation of not only T cells but also NK cells to low doses of IL-2 can be inhibited by antibodies to the IL-2R chain [18
, 39
]. However, the ability of fresh human NK cells to proliferate to low doses of IL-2 is entirely accounted for by the existence of a small (2–5%) subpopulation of NK cells which expresses high levels of the IL-2R chain [40
, 41
]. This subpopulation of NK cells is phenotypically and functionally distinct from the major population of NK cells, being CD56hiCD16-, having relatively low cytotoxic activity, and accounting for nearly all cytokine secretion [40
41
42
]. It also contains all of the cells that respond to IL-15 [18
]. The failure of fresh mouse NK cells to proliferate to low doses of IL-2 (and IL-15) implies that no corresponding subset of NK cells exists in this species. Early reports that mouse NK cells could respond to low doses of IL-2 (e.g., refs. [43
, 44
]) can now be dismissed as artifacts caused by impurities in the IL-2 preparations used and in the cell populations studied, exacerbated by the propensity of IL-2-activated mouse T cells to express asialoGM1 [45
], NK1.1 [46
], and lytic activity against YAC cells [47
].
However, the frequent presumption that mouse NK cells do not express IL-2R chains is also incorrect. As shown here, homogeneously pure populations of mouse NK cells clearly express IL-2R mRNA, and long-term lines of mouse NK cells expressed easily detectable levels of IL-2R chains on the cell surface. Freshly derived immature and mature NK cells cultured in IL-2 showed no detectable surface expression of IL-2R chains, but when cultured in IL-15 or even in medium alone, but not when cultured in a mixture of IL-2 and IL-15, low levels of surface IL-2R chains were clearly present and in contrast to the situation in man, were expressed on most or all NK cells. Collectively, these results indicate that the lack of detectable IL-2R chains on the surface of NK cells grown in IL-2 is a result of the continuous down-regulation of IL-2R chains by exogenous IL-2. The speed with which IL-2R chains appeared when NK cells were cultured in IL-2-free medium (close to maximal expression within 3 h) suggests that their expression is controlled at least in part at a post-transcriptional level, a view supported by a recent study of the long-term human NK cell line YT, where it was found that high concentrations of IL-2 and IL-15 could promote the synthesis of intracellular IL-2R chains, but surface expression was only detected when cells were cultured in IL-15 [48
]. Surprisingly, we found that surface IL-2R expression could be efficiently down-regulated at picomolar doses of IL-2, implying that most or all of the IL-2R chains on NK cells are associated with high-affinity receptors. This is in line with previous studies that showed that the IL-2-driven proliferation of immature and mature NK cells in the presence of limiting concentrations of IL-2 could be inhibited by mAb to the IL-2R chain [11
].
The concentrations of IL-2 and IL-15 required for efficient proliferation and differentiation of immature NK cells in vitro are much higher than those likely to be generated in vivo. This is especially the case for IL-15, whose efficiency of production is severely constrained by a series of post-transcriptional regulatory events that include multiple upstream AUG codons, alternate splicing, and inefficient translocation and processing in the endoplasmic reticulum [49
50
51
]. How can these observations be reconciled with the finding that IL-15-/-and IL-15R-/- mice are grossly deficient in NK cells? First, it should be noted that there is in fact no evidence that IL-15 directly promotes the growth and differentiation of immature NK cells in vivo. The recent finding that the introduction of a bcl2 transgene into IL-2/15Rß knockout mice causes the restoration of normal numbers of NK cells and that these NK cells have normal expression of Ly49 molecules but lack cytolytic activity [52
] strongly suggests that neither IL-15 nor IL-2 is required for the proliferative expansion of immature NK cells and that the principal roles of IL-15 and/or IL-2 are to promote the survival of an early NK progenitor and the development or maintenance of cytotoxic activity in NK cells. Recent studies have also revealed that although IL-15R chains are required for homeostatic proliferation of CD8 T cells, there is no requirement for these to be expressed on the CD8 T cells themselves [53
]. Second, IL-15 may not act as a soluble mediator. Provocative studies by Waldmann and colleagues [54
] indicate that minute quantities of IL-15 transported from intracellular stores by the IL-15R chain can be expressed at cell surfaces and stimulate "in trans" the proliferation of cells bearing IL-2/15Rß/c receptors. An implicit conclusion from these studies is that IL-15R chains would not need to be expressed on immature NK cells for these cells to respond efficiently to IL-15. Low concentrations of IL-2 may also be able to efficiently stimulate IL-2R-negative cells in trans [55
]. Third, IL-15 may act indirectly by inducing the production of soluble or cell-bound stimulatory factors from "stromal" cells present at the sites of NK cell development. Lastly, IL-15 at physiological concentrations may, on its own, be incapable of triggering NK cells. In view of the finding that mouse NK cells can clearly express IL-2R chains, one possibility would be that IL-15 and IL-2 act synergistically. However, although IL-15 and IL-2 can act synergistically to initiate the growth of human NK cells [56
], in the present study, we could find no evidence of a synergistic interaction between IL-2 and IL-15 for the growth of mouse NK cells. By contrast, another c cytokine, IL-21, markedly enhanced the growth of NK cells in the presence of low concentrations of IL-15 or IL-2. This effect was seen only with low concentrations of IL-21; high concentrations of IL-21 profoundly inhibited the growth of NK cells, in agreement with a recent report [57
]. The effects of IL-21 were specific for NK cells in that IL-21 failed to enhance the growth of T cell blasts at any dose tested and also, only minimally inhibited the growth of T cells at high doses. This contrasts with reports that IL-21 enhances the growth of fresh mouse T cells when added together with anti-CD3 mAb or allogeneic cells [32
, 57
], suggesting that IL-21 exerts its primary effect on T cells when present at the time of activation. The ability of IL-21 to enhance the growth of immature mouse NK cells is in line with similar observations in man [32
]. However, recent studies have shown that NK cells develop normally in mice lacking the only known receptor for IL-21 [57
], indicating that IL-21 is not essential for NK cell development. Taken together, the results reported here and elsewhere suggest that systemic concentrations of IL-21 emanating from sites of T cell activation may enhance the production of NK cells from immature progenitors in the BM and stimulate the proliferation of mature NK cells at distal sites, whereas at sites of inflammation where the concentrations of IL-21 are higher, the further proliferation of infiltrating NK cells would be blocked.
Received March 10, 2003; accepted April 2, 2003.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
 |
|
 |
|
  A. Thedrez, C. Harly, A. Morice, S. Salot, M. Bonneville, and E. Scotet IL-21-Mediated Potentiation of Antitumor Cytolytic and Proinflammatory Responses of Human V{gamma}9V{delta}2 T Cells for Adoptive Immunotherapy J. Immunol., March 15, 2009; 182(6): 3423 - 3431. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 W. J. Leonard, R. Zeng, and R. Spolski Interleukin 21: a cytokine/cytokine receptor system that has come of age J. Leukoc. Biol., August 1, 2008; 84(2): 348 - 356. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Onoda, M. Rahman, H. Nara, A. Araki, K. Makabe, K. Tsumoto, I. Kumagai, T. Kudo, N. Ishii, N. Tanaka, et al. Human CD4+ central and effector memory T cells produce IL-21: effect on cytokine-driven proliferation of CD4+ T cell subsets Int. Immunol., October 1, 2007; 19(10): 1191 - 1199. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B. R. Barker, J. G. Parvani, D. Meyer, A. S. Hey, K. Skak, and N. L. Letvin IL-21 Induces Apoptosis of Antigen-Specific CD8+ T Lymphocytes J. Immunol., September 15, 2007; 179(6): 3596 - 3603. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. Zeng, R. Spolski, E. Casas, W. Zhu, D. E. Levy, and W. J. Leonard The molecular basis of IL-21-mediated proliferation Blood, May 15, 2007; 109(10): 4135 - 4142. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Cappuccio, M. Elishmereni, and Z. Agur Cancer immunotherapy by interleukin-21: potential treatment strategies evaluated in a mathematical model. Cancer Res., July 15, 2006; 66(14): 7293 - 7300. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Sun, H. Wei, R. Sun, W. Xiao, Y. Yang, and Z. Tian Human Interleukin-15 Improves Engraftment of Human T Cells in NOD-SCID Mice Clin. Vaccine Immunol., February 1, 2006; 13(2): 227 - 234. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Caraux, N. Kim, S. E. Bell, S. Zompi, T. Ranson, S. Lesjean-Pottier, M. E. Garcia-Ojeda, M. Turner, and F. Colucci Phospholipase C-{gamma}2 is essential for NK cell cytotoxicity and innate immunity to malignant and virally infected cells Blood, February 1, 2006; 107(3): 994 - 1002. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. A. Perez, L. G. Mahaira, P. A. Sotiropoulou, A. D. Gritzapis, E. G. Iliopoulou, D. K. Niarchos, N. T. Cacoullos, Y. G. Kavalakis, A. I. Antsaklis, N. N. Sotiriadou, et al. Effect of IL-21 on NK cells derived from different umbilical cord blood populations Int. Immunol., January 1, 2006; 18(1): 49 - 58. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
F. Gays, K. Martin, R. Kenefeck, J. G. Aust, and C. G. Brooks Multiple Cytokines Regulate the NK Gene Complex-Encoded Receptor Repertoire of Mature NK Cells and T Cells J. Immunol., September 1, 2005; 175(5): 2938 - 2947. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. Takaki, Y. Hayakawa, A. Nelson, P. V. Sivakumar, S. Hughes, M. J. Smyth, and L. L. Lanier IL-21 Enhances Tumor Rejection through a NKG2D-Dependent Mechanism J. Immunol., August 15, 2005; 175(4): 2167 - 2173. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. L. Vollmer, R. Liu, M. Price, S. Rhodes, A. La Cava, and F.-D. Shi Differential Effects of IL-21 during Initiation and Progression of Autoimmunity against Neuroantigen J. Immunol., March 1, 2005; 174(5): 2696 - 2701. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. Jin, R. Carrio, A. Yu, and T. R. Malek Distinct Activation Signals Determine whether IL-21 Induces B Cell Costimulation, Growth Arrest, or Bim-Dependent Apoptosis J. Immunol., July 1, 2004; 173(1): 657 - 665. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
