science pharmaceutical expo biotech jobs
Originally published online as doi:10.1189/jlb.0306191 on September 11, 2006

Published online before print September 11, 2006
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Supplemental Data: jlb.0306191
Right arrow All Versions of this Article:
jlb.0306191v1
80/6/1529    most recent
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Wendt, K.
Right arrow Articles by Jacobs, R.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Wendt, K.
Right arrow Articles by Jacobs, R.
(Journal of Leukocyte Biology. 2006;80:1529-1541.)
© 2006 by Society for Leukocyte Biology

Gene and protein characteristics reflect functional diversity of CD56dim and CD56bright NK cells

Katy Wendt*,1, Esther Wilk*,1, Sabine Buyny*, Jan Buer{dagger}, Reinhold E. Schmidt* and Roland Jacobs*,2

* Department of Clinical Immunology, Hannover Medical School, Hannover, Germany; and
{dagger} Mucosal Immunity Group, Gesellschaft für Biotechnologische Forschung, Braunschweig, Germany

2 Correspondence: Dept. of Clinical Immunology, OE 6830, Hannover Medical School, Carl-Neuberg-Str. 1, Hannover D-30625, Germany. E-mail: jacobs.roland{at}mh-hannover.de


arrow
ABSTRACT
 
Recent findings underline the role of NK cell subsets in regulating adaptive immunity. To define characteristics of NK cell subpopulations, purified CD56dim and CD56bright NK cells were analyzed by using gene chip arrays covering more than 39,000 transcripts. Gene profiling revealed resting NK cells to differ in respect to 473 transcripts with 176 exclusively expressed in CD56dim and 130 solely in CD56bright NK cells. Results were compared with array analyses using mRNA obtained from activated CD56dim and CD56bright NK cells. In this approach, NK cell receptors, cytolytic molecules, adhesion structures, and chemokine ligands showed differential expression patterns in the two subpopulations. These data were validated using FACS, RT-qPCR, or cytokine bead array (CBA) techniques. Cytokines produced by CD56dim and CD56bright NK cells were determined using a protein array covering 79 different bioactive mediators. GDNF, IGFBP-1, EGF, and TIMP-2 were detected in both subsets. In contrast, IGFBP-3 and IGF-1 were mainly produced by CD56dim, while GM-CSF, TARC, and TGFβ3 were expressed by CD56bright NK cells. In summary, we report new characteristic features of CD56dim and CD56bright NK cells, further underscoring that they represent independent populations with functionally diverse capabilities. The information on NK cells generated in this study will help to define corresponding NK cell populations in other species that lack CD56 expression on NK cells, such as mice. This will subsequently lead to the establishment of suitable animal models for detailed analysis of NK cell populations in vivo.

Key Words: KIR • KLR • NCAM • IL-8


arrow
INTRODUCTION
 
NK cells are effector cells of the innate immune system recognizing stressed, transformed, or infected cells via a set of activating and inhibitory receptors. In addition to exerting cellular cytotoxicity, NK cells produce chemokines and cytokines, allowing them to also regulate immune functions. Recent data on crosstalk between NK cells, dendritic cells (DCs), and T cells indicate that NK cells can bridge innate and adaptive immunity [1 , 2 ].

Human NK cells are characterized phenotypically by expression of CD56 and lack of CD3. In addition, most NK cells coexpress Fc{gamma}RIII (CD16) [3 ]. Nevertheless, the NK cell compartment is not homogeneous but is rather composed of functionally distinct subsets [4 ]. Two subpopulations can be distinguished based upon CD56 cell-surface density. In peripheral blood, the majority (~90%) of human NK cells display low-density expression of CD56 (CD56dim) and high expression of CD16. The remaining 10% are termed CD56bright NK cells, as they express 5- to 10-fold higher levels of CD56. CD56bright NK cells present either no, or at most, dim expression of CD16 [5 ]. Functionally, CD56dim cells represent ‘classical NK cells’ with strong cytotoxic capacity [6 , 7 ]. In contrast, CD56bright NK cells are poor killers and characterized by enhanced cytokine production. CD56bright NK cells represent the main NK cell populations in lymph nodes and inflamed tissues [8 9 10 ]. In lymph nodes, NK cells have recently been shown to interact with T cells and DCs, which are in close proximity, thereby modifying immune responses [11 , 12 ]. CD56bright NK cells accumulate at inflammatory sites, including exudative pleural fluid, peritoneal fluid from patients with peritonitis, and synovial fluid or tissue from patients with inflammatory arthritis [10 ]. In human decidua, CD56bright NK cells represent the main lymphocyte population. But decidual CD56bright NK cells have been reported to express killer cell immunoglobulin-like receptors (KIRs) and exhibit some other unique properties, suggesting that they constitute an NK cell population, which is remarkably different from either subset of peripheral blood NK cells [13 , 14 ]. Decidual CD56bright NK cells are supposed to play an important role in decidualization and implantation [15 ].

NK cell function is regulated by the balance of activating and inhibitory signals mediated by respective receptors, most of which have been shown to be differentially expressed on NK cell subpopulations. NKp46, NKp44, and NKp30 comprise a family of activating NK cell receptors (NCR). A direct correlation exists between the surface density of NCR and the ability of NK cells to kill various tumors [16 ]. Cellular ligands of NCRs are still unknown, but NKp44 and NKp46 recognize hemagglutinin of influenza and sendai virus, while NKp30 binds to pp65 protein of human cytomegalovirus (HCMV) [17 18 19 ]. Recently, a novel NKp44 ligand (L) has been detected, which is induced after infection on CD4+ T cells by a motif of the HIV-1 envelope gp41 protein. NKp44L is suggested to play a key role in depletion of helper T cells by activated NK cells in HIV infection [20 ]. Another activating NK cell receptor is 2B4, whose ligand is CD48 [21 , 22 ]. Inhibitory receptors bind classical and nonclassical MHC-I and avert killing of MHC-I-bearing target cells. These receptors are crucial to maintain self-tolerance. Inhibition is mediated through one or more immunoreceptor tyrosine-based inhibitory motifs (ITIM) located in the cytoplasmic domain of KIRs and in killer cell lectin-like receptors (KLR) CD94/NKG2. In contrast, stimulatory receptors have short cytoplasmic domains without signal-transducing elements. Stimulatory receptors associate with adaptor molecules like DAP12 (KARAP), Fc{epsilon}RI{gamma}, and CD3{zeta}, which contain an immunoreceptor tyrosine-based activation motif (ITAM) or DAP10, which bears a YxxM motif [23 24 25 ]. The activating receptor NKG2D is present on all NK cells, whereas most receptors are expressed only on subsets of NK cells [26 ]. Another exceptional receptor is KIR2DL4, which is a member of the KIR family and displays structural features of both activating and inhibitory receptors [27 ]. Unlike other clonally distributed KIRs, KIR2DL4 is transcribed by all NK cells [28 , 29 ]. Signaling via this receptor induces IFN{gamma} production but not cytotoxicity. In contrast to other KIRs, KIR2DL4 is predominantly localized in endosomes and can pair with the Fc{epsilon}RI{gamma} chain [30 , 31 ].

Currently, the main differences of both NK cell subsets described so far, in addition to CD56 surface density include KIR expression, which is mostly restricted to CD56dim NK cells. This subset also expresses high amounts of cytolytic molecules and CD16 in high density. Together with the strong capacity to form conjugates with tumor- or virus-transformed cells, the cellular equipment enables CD56dim NK cells to efficiently lyse sensitive targets directly or by antibody-dependent cellular-cytotoxicity (ADCC). In contrast, CD56bright NK cells have only limited cytotoxic capacity. They largely lack KIRs but do express KLRs such as CD94 and CD161, and they are suspected to represent regulatory cells.

In this study, we analyzed highly purified CD56dim and CD56bright NK cells using gene array analyses covering up to 39,000 transcripts representing ~33,000 genes in order to further discriminate phenotypic and functional capacities of NK cell subpopulations. Subsequently, recorded data were evaluated and verified using various assay systems, including RT-qPCR, FACS analyses, cytokine arrays, and cytometric bead arrays for quantitative analysis of soluble factors. Our data add new information on the functional dissimilarity of CD56dim and CD56bright NK cells and underline the differential impact these two NK cell subpopulations have on immune responses.


arrow
MATERIAL AND METHODS
 
Micro array gene chip analysis
Affymetrix GeneChip® expression arrays U133A and U133B were performed by the array facility of GBF Braunschweig. Data were processed using Affymetrix software and analyzed on the basis of Affymetrix algorithms. One experiment was performed using sorted CD56dim and CD56bright NK cells from an individual. For the second array experiment, a pool of 10 donors of the respective NK cell populations was applied. The pools were prepared using equimolar amounts of RNA from the single donors. For the analysis of activated NK cells sorted cells of eight donors were stimulated with PMA/ionomycin for 48 h. RNA was then harvested, pooled in equimolar amounts, and analyzed accordingly using the U133A array.

RNA isolation, reverse transcription, and qPCR
Total RNA was isolated using RNeasy Micro and RNeasy Mini Kit (Qiagen, Hilden, Germany) with on-column DNase treatment. Quality of preparation was assessed using an Agilent Bioanalyzer, Pico Chip (Agilent Technologies, Palo Alto, CA). Reverse transcription was performed with Omniscript or Sensiscript RT Kit (Qiagen, Hamburg, Germany), according to the manufacturer’s instructions.

Two different methods were established for RT-qPCR (iCycler, Bio-Rad, Hercules, CA). To quantify PCR results 12.5 µl SYBRGreen Supermix (Bio-Rad) was added to primers (final concentration 250 nM) and equal amounts of cDNA in a total volume of 25 µl. Control cDNA was used to generate a standard curve. The following primers were used:

Actin (forward, fw) 5'-AGTACTCCGTGTGGATCGGC, (reverse, rev) 5'-GCTGATCCACATCTGGTGGA; INF{gamma} (fw) 5'-TTCGGTAACTGACTTGAATGT, (rev) 5'-ACCTCGAAACAGCATCTGA; TNF: (fw) 5'-ATCTTCTCGAACCCCGAGTGA, (rev) 5'-CGGTTCAGCCACTGGAGCT; NCAM (fw) 5'-GGAGGGGAACCAGGTGAACA, (rev) 5'-TGGTCGATGGATGGTGAAGAG; NKG2D (fw) 5'-CACGTCATTGTGGCCATTGT, (rev) 5'-AAGCACAGGCCAGCAAACTCT.

Instead of SYBRGreen in some cases FAM-TAMRA-labeled probes were generated and added to primer pairs. In this case 12.5 µl Taq Man Universal PCR Master Mix (Applied Biosystems, Foster City, CA) was added into a total volume of 25 µl with a concentration of 400 nM of each primer. Probes were applied at a final concentration of 200 nM. The following primers were used: Actin (fw) 5'-AGTACTCCGTGTGGATCGGC, (rev) 5'-GCTGATCCACATCTGGTGGA, (probe) 6-FAM-TCCATCCTGGCCTCGCTGTCCA-TAMRA; KIR2DL1 (fw) 5'-CGTGGGCGTGCCTGTC, (rev) 5'-AAGAACCCAACACACGCCAT, (probe) 6-FAM-CAGCACCATGTCGCTCTTGTTCGTCA-TAMRA; KIR2DS1 (fw) 5'-TGAACGTAGGCTCCCTGCA, (rev) 5'-CGGAAAGAGCCGAAGCATC, (probe) 6-FAM-CCAACTTTCCTCTGGGCCCTGCCACCCATGG-TAMRA; KIR2DL2 (fw) 5'-AGAGGCCCAAGACACCCC, (rev) 5'-CAAGGCCTGACTGTGGTGC, (probe) 6-FAM-AGATATCATCGTGTACGCGGAACTTCCAAATGCT-TAMRA; KIR2DS2 (fw) 5'-TGGTCAGATGTCAGGTTTGAGC, (rev) 5'-TGGCCTTGGAGACCCCAT, (probe) 6-FAM-AGGGGAAGTATAAGGACACTTTGCACCTCATTGGAGAGC-TAMRA; KIR2DL3 (fw) 5'-AACAGTGAACAGGGAGGACTCTG, (rev) 5'-AAGGGCGAGTGATTTTTCTCTG, (probe) 6-FAM-CTCAGGAGGTGACATATGCACAGTTGAATCA-TAMRA; KIR3DL1 (fw) 5'-TCCAAGGCCAATTTCTCCAT, (rev) 5'-GTGAGTAACAGAACCGTAGCATCTG, (probe) 6-FAM-TCCCATGATGCTTGCCCTTGCAG-TAMRA; KIR3DS1 (fw) 5'-CTAAGGACCCCTCACGCCTC, (rev) 5'-TCACTGGGAGCTGACAACTGA, (probe) 6-FAM-TTTCTCCATCGGTTCCATGATGCGTGCC-TAMRA; KIR2DL4 (fw) 5'-TCCCAGAGCTCCTTTGACAT, (rev) 5'-AGAAACAGGCAGTGGGTCAC, (probe) 6-FAM-CCATGGATCTCCCTACGAGTGGTT-TAMRA.

Cell sorting and FACS
White blood cells were flushed with PBS from leukocyte filters (Fresenius-Filter) used for preparation of erythrocyte concentrates for transfusion. The recovered white blood cells were separated by Ficoll-Hypaque gradient centrifugation. PBMC from the interphase were washed twice and depleted using a cocktail of monoclonal antibodies against CD3 (purified OKT3 hybridoma supernatant), CD14, CD19 (Immunotools, Frisoythe, Germany), and Dynabeads® M450 (Dynal, Hamburg, Germany). Finally, cells were stained with CD56PE and CD3FITC and sorted with FACStar (Becton Dickinson, San Diego, CA), according to expression density of CD56 antigen into CD3–CD56dim and CD3–CD56bright NK cells. Purity of the sorted NK cells was reanalyzed by flow cytometry (FACSCalibur, Becton Dickinson).

Intracellular staining
PBLs (3 x 106) were resuspended in RPMI 1640 supplemented with 10% FCS, 100 U/ml penicillin, 100 µg/ml streptomycin, 2 mM L-glutamine, 1 mM sodium pyruvate and were stimulated as indicated below. Brefeldin A (2 µg/ml) was added 1 h after starting stimulation, followed by an incubation of 4 or 24 h; 3 x 105 cells were used for each analysis. Surface markers CD56 and CD3 were stained before fixing the cells with PBS/4% paraformaldehyde. After 10 min, cells were washed and 50 µl of saponin buffer (PBS supplemented with 0.1% saponin and 0.01M HEPES), was pipetted before adding antibodies against intracellular proteins like granzyme K (Immunotools), IFN{gamma} (Caltag, Hamburg, Germany), TNF (PharMingen, Heidelberg, Germany), and IL-8 (R&D Systems, Wiesbaden, Germany).

Stimulation of cells
Cells were resuspended in RPMI 1640 supplemented with 10% FCS, 100 U/ml penicillin, 100 µg/ml streptomycin, 2 mM L-glutamine, and 1 mM sodium pyruvate at a concentration of 1x106 cells/ml (if not otherwise noted). 50 ng/ml PMA was used together with 500 ng/ml ionomycin. Sorted NK cells were incubated at 37°C in a humidified atmosphere with 5% CO2 as indicated. After stimulation, cells and supernatants were harvested.

Cytometric bead array (CBA)
Sorted CD56dim and CD56bright NK cells (1x106/ml) were stimulated for 24 h, as described above. Supernatants were harvested and stored at –20°C until cytokine analysis was performed. IL-1β, IL-2, IL-4, IL-6, IL-8, IL-10, IL-12p70, TNF, and IFN{gamma} were detected simultaneously using the human inflammation kit and Th1/Th2 cytokine cytometric bead array (CBA) kit 2 (PharMingen), according to the manufacturer’s instructions. Briefly, 50 µl of each sample was added to 50 µl of mixed capture beads and 50 µl of the respective PE conjugated detection reagent. Samples were incubated at room temperature for 3 h in the dark. After incubation, samples were washed once and resuspended in 300 µl of wash buffer before acquisition with the FACSCalibur. Data were analyzed using CBA software (PharMingen). Standard curves were generated for each cytokine using the mixed cytokine standard provided with the kit. The concentration for each cytokine in cell supernatants was determined by interpolation from the corresponding standard curve. The range of detection was 20–5000 pg/ml for each cytokine measured by CBA.

Cytokine array
Sorted NK cells from two groups of eight and seven different donors were stimulated as mentioned above. After 48 h, supernatants were collected (100 µl each), pooled and applied to the antibody-coated membrane, as recommended by the manufacturer (Raybiotech, Norcross, GA). Biotinylated antibody was added followed by HRP-conjugated streptavidin. After treatment with detection buffers, the membrane was placed on a Biomax MR film (Kodak, Rochester, NY) and exposed for 10 min. The spots on the developed film were analyzed using the free software ScionImage v 4.02 (Scion Corp, Frederick, MD). The protein array used in this study was basically not established for quantitative analyses. To quantify the clear differences that were observed in diameter and density of the dots, we applied estimations according to densiometric techniques, generally used in blot analyses. A histogram was automatically calculated according to density and diameter of the single dots by the software. Area of the histogram was used for approximate quantification of the corresponding proteins. The dots were normalized based on the mean of the six positive controls, as recommended by the manufacturer. The value of each cytokine was divided by the mean of the corresponding positive controls and termed relative protein level.


arrow
RESULTS
 
Expression patterns of CD56dim and CD56bright NK cells
All experiments were performed using highly purified NK cells. After enrichment by depletion with Dynabeads® M450 and a cocktail of monoclonal antibodies against CD3, CD14, and CD19, cells were sorted into CD56dim and CD56bright NK cell subpopulations with a purity of more than 99% for each population. Sorted fractions were reanalyzed on a FACSCalibur cytometer with the gates set on all recovered cells excluding debris and erythrocytes. Samples with contamination of the respective other subpopulation were discarded. To evaluate possible biases of the gene array analysis due to a possible uniqueness of the chosen donor or the peculiarity of an RNA pool, both sources were used for gene profiling. Expression data were compared by plotting respective values of the individual vs. the pool. A high correlation between both analyses of the CD56dim (r2=0.9632) and the CD56bright (r2=0.9662) NK cell populations proved the equivalence of both RNA sources, but for further analyses, we only considered data of pooled samples (Fig. 1 ). To validate results of Affymetrix GeneChips, we compared analyses with data previously generated in our lab. In addition, verification by FACS or RT-qPCR was performed for several molecules, which appeared to be differentially regulated in both NK cell subsets. Absence of specific transcripts of T and B cell or monocyte origin confirmed the purity of samples (data not shown). Focusing on NK cells only, analysis of GeneChips U133A and U133B using pooled RNA from 10 healthy donors revealed that CD56dim and CD56bright NK cells significantly differ in the expression of 473 transcripts. One-hundred seventy-six of these genes were exclusively active in CD56dim and 130 only in CD56bright NK cells.


Figure 1
View larger version (20K):
[in this window]
[in a new window]

  		Figure 1. Correlation of expression data of different gene chip arrays. RNA from either a single individual (x-axis) or a pool of 10 donors mixed in equimolar amounts (y-axis) was analyzed. Expression data from both sources are shown for comparison of CD56dim (A) and CD56bright (B) NK cells.

To investigate whether both subsets differ in their cellular response profile after stimulation, sorted CD56dim and CD56bright NK cells were activated 48 h with PMA/ionomycin, which was chosen as a strong and unspecific stimulus. This treatment is hereinafter referred to as activation or stimulation and the cells accordingly as activated or stimulated NK cells. Isolated mRNA of these cells was then analyzed on a GeneChip® U133A. The array revealed that activated CD56dim and CD56bright NK cells significantly differed in the expression of 391 transcripts, with 47 genes exclusively active in CD56dim NK cells and 76 genes solely active in CD56bright NK cells.

Expression data of all arrays analyzed in this study are provided as supporting information stored in EMBO-EBI ArrayExpress database (URL: http://www.ebi.ac.uk/aerep/entry?) under accession no.: E-MEXP-380.

All of the following data obtained from the gene array analyses were verified by control array and/or a different experimental approach (RT-qPCR or on protein level, respectively).

In accordance with the differential surface expression of the sorted cell populations, higher levels of CD56 mRNA were confirmed with RT-qPCR in CD56bright NK cells (data not shown). CD56dim cells exhibited CD16, whereas CD16 mRNA was absent in CD56bright NK cells. However, a low density of CD16 protein was detectable in a few donors of the sample pool, possibly due to down-regulation via degradation of CD16 specific RNA. KIR expression in NK cell populations was analyzed by RT-qPCR. The genetic approach allows a more detailed analysis of KIR expression, since this technique can distinguish between activating and inhibiting KIR isoforms. As described previously for expression on the protein level, we found a predominant gene expression of KIRs in CD56dim cells (Fig. 2A ) [5 ]. The only exception is KIR2DL4, which is known to be transcribed by all NK cells [28 , 29 ]. However, expression of this molecule in peripheral blood is mainly restricted to CD56bright NK cells [32 ]. Our RT-qPCR approach using primers comprising all KIR2DL4 (9a and 10a alleles) variants revealed higher mRNA content for this molecule in CD56bright NK cells. After activation, KIR expression patterns remained stable, although these receptors were generally down-regulated (data not shown). In contrast, most KLRs were expressed in comparable patterns on CD56bright and CD56dim NK cells (Fig. 2B) . Receptors of the family including NKG2A (KLRC1), NKG2C (KLRC2), and NKG2E (KLRC3) were present at similar levels in both subpopulations. KLRC1-3 form dimeric receptor complexes with CD94. In contrast, higher amounts of CD94 mRNA were detected in CD56dim cells, probably indicating a significant expression of CD94 homodimers in this population. After activation, message for CD94 was decreased by ~71% in CD56dim (intensity signals: 1720 vs. 503) and 47% in CD56bright (intensity signals: 624 vs. 332) NK cells as compared with resting NK cell subpopulation (data not shown). This finding suggests a differential regulation of CD94 on the transcriptional level in both populations. Expression levels of CD160 (BY55), CD244 (2B4), and CD161 were higher in CD56dim but were also detectable in CD56bright NK cells. Among adaptor proteins mediating activation of NK cells like CD3{zeta}, DAP10, DAP12, and Fc{epsilon}RI{gamma}, only CD3{zeta} was expressed significantly stronger in CD56dim NK cells, which might be due to its role as signaling adaptor molecule of CD16, which is expressed at a much higher level in CD56dim NK cells. Most lytic molecules like perforin, granzyme A, B, M, and CTLA1 were transcribed at significantly higher levels in CD56dim NK cells. However, granzyme K exhibited a stronger expression in resting CD56bright NK cells, which was also confirmed on the protein level (Fig. 3A and 3B ). Interestingly, the pattern of lytic molecules expressed by both NK cell subsets changed after stimulation (Fig. 3C) . In this regard, expression of perforin and granzyme A was not affected by stimulation and remained similar to the results in resting NK cells with significantly higher levels in CD56dim NK cells. Granzyme B was up-regulated in both subpopulations, whereas CTLA1, granzyme M and K were decreased upon stimulation.


Figure 2
View larger version (20K):
[in this window]
[in a new window]

  		Figure 2. Differential expression of NK cell receptors in CD56dim and CD56bright NK cells. Amounts of NK cell receptor encoding mRNA in sorted CD56dim (open bars) and CD56bright NK cells (solid bars) are depicted. (A) Gene expression of KIRs was investigated by RT-qPCR using primers described in Materials and Methods. y-axis indicates the multiple of the higher signal relating to the respective lower. (B) Array data of killer cell lectin-like receptors (KLRs) are presented. Significant differences are indicated according to the Affymetrix algorithm (***P≤0.001; **P≤0.01).


Figure 3
View larger version (16K):
[in this window]
[in a new window]

  		Figure 3. Gene expression of cytolytic molecules. FACS analysis depicts intracellular granzyme K expression in resting CD3–CD56+ lymphocytes (A). Array data of resting (B) and activated (C) CD56dim (open bars) and CD56bright NK cells (solid bars) illustrate differential equipment with molecules needed for cytolytic function. Levels of significance are indicated according to the Affymetrix algorithm (***P≤0.001).

CD56dim and CD56bright NK cells substantially differ in the cellular equipment required for migration. Adhesion molecules CD44 and L-selectin (CD62L) are present in both NK subpopulations, but mRNA for both molecules was significantly increased in CD56bright NK cells (Fig. 4A ). In contrast to CD44, CD62L was further increased after stimulation. This finding was confirmed on the protein level by FACS analysis (data not shown). RNAs encoding activated leukocyte cell adhesion molecule (ALCAM, CD166) or ICAM2 (CD102) could only be detected after stimulation and were expressed significantly higher in CD56dim NK cells (Fig. 4B) . On the other hand, ICAM1 (CD54) was only expressed in resting CD56dim NK cells and was down-regulated upon activation. In contrast, ICAM3 (CD50) was stably expressed in resting and activated cells. Expression of CD31 was higher in resting CD56dim NK cells but was preferentially induced in CD56bright NK cells after stimulation.


Figure 4
View larger version (17K):
[in this window]
[in a new window]

  		Figure 4. CD56dim and CD56bright NK cells differ in expression patterns of molecules needed for population-specific functions. Comparison of molecules involved in adhesion, migration, and cell-cell crosstalk are depicted in resting (A) and activated cells (B). Array data of highly purified CD56dim (open bars) and CD56bright NK cells (solid bars) are shown. Bars are arbitrarily set to one either for the population with the lower expression level or the one that exclusively expresses the respective message. If present in both, the higher level is given as multiple of the corresponding lower value. Levels of significance are indicated according to the Affymetrix algorithm (***P≤0.001; **P≤0.01).

Directed migration of lymphocytes is maintained by soluble factors that act via suitable receptors. Comparison of molecules from both NK cell populations revealed that chemokine receptors CCR7 and CXCR3 were exclusively expressed by CD56bright NK cells, indicating a possible chemotactic response toward corresponding ligands of this particular NK cell subset (Fig. 4) .

Once the cells are attracted to the site of immune response, their functional impact may further be perpetuated by activation and/or proliferation of the cells. Therefore, we also focused on molecules that could mediate such signals to NK cells. Particularly, CD56bright NK cells were found to express receptors that are involved in induction of cell proliferation or activation. Namely, these are IL-7R and IL-12Rβ (Fig. 4) .

To investigate gene regulation in both NK cell populations and to address the question whether CD56bright NK cells might be closely related to activated CD56dim NK cells, we analyzed gene profiles of both NK cell subpopulations after stimulation with PMA/ionomycin. First, we selected the 487 genes that were at least threefold up-regulated in CD56dim NK upon activation (Fig. 5A ). Details on the 487 selected genes are listed in an Excel file, which is provided as supporting material (Supplemental Table 1). In parallel, expression signals of the same genes in resting CD56bright NK cells were compared with resting CD56dim NK cells. If CD56bright NK cells represent an activated stage of CD56dim NK cells, most of the genes which were up-regulated in activated CD56dim would also be expected to be increased in resting CD56bright when compared with resting CD56dim NK cells. As shown in Fig. 5 , expression patterns of these genes clearly differ between resting CD56bright and resting CD56dim NK cells (Fig. 5A and 5B) . Although, expression of 117 genes was at least twofold (59 genes >3-fold) higher in resting CD56bright than in resting CD56dim NK cells, 159 genes were concomitantly down-regulated, arguing rather against a close relation of resting CD56bright and activated CD56dim NK cells.


Figure 5
View larger version (53K):
[in this window]
[in a new window]

  		Figure 5. Regulation of gene expression in resting and activated NK cell subpopulations. Sorted CD56dim and CD56bright NK cells were activated with PMA/ionomycin and subjected to gene array analysis. Up (>1)- and down-regulation (<1) of genes in the population given on the top of each diagram is shown as compared with the population indicated on the x-axis. To enhance readability of the graphs, data points were connected with a line. (A) 487 genes, which were at least threefold up-regulated in CD56dim NK cells upon stimulation are depicted. (B) The same genes as plotted in (A) were compared with the expression of resting CD56bright NK cells in order to evaluate a possible relationship between activated CD56dim and resting CD56bright NK cells. (C) Regulation of the particular 487 genes was also analyzed in activated CD56bright NK cells to assess whether CD56bright NK cells can be further stimulated. (D) All 1719 genes, which were at least threefold increased in CD56bright NK cells upon stimulation, are shown.

In addition, most of these particular genes were considerably up-regulated in activated CD56bright NK cells in relation to CD56dim NK cells, indicating that resting CD56bright NK cells can be activated (Fig. 5C) . This becomes even more apparent when all genes, which were at least threefold up-regulated in CD56bright NK cells after activation, were included is this analysis (Fig. 5D) . The array revealed 1719 genes meeting this criterion, including genes involved in apoptosis (27 pro- and 11 anti-apoptotic), proliferation (10 pro- and anti-proliferative), IL-15, IL-16, and IL-2Ra (CD25). The difference of gene expression patterns exerted by activated CD56dim vs. resting CD56bright NK cells and the strong response of CD56bright NK cells toward a stimulus indicates that CD56bright NK cells represent an independent cell population that is neither terminally differentiated nor functionally exhausted. Details on the 1719 selected genes are listed in an Excel file, which is provided as supporting material (Supplemental Table 2).

Cytokine and chemokine production
In addition to their cytotoxic capacity, NK cells are able to produce a variety of soluble factors. To further evaluate the specific capability of both NK cell subpopulations, we analyzed sorted NK cells prior to and after stimulation with PMA/ionomycin. Our array experiments confirmed the preferential production of TNF and IFN{gamma} by CD56bright NK cells, which has been reported previously [5 ]. However, quantification by using the CBA after 24 h activation revealed that concentrations of IFN{gamma} (3100 pg/ml vs. 2900 pg/ml) and TNF (500 pg/ml vs. 380 pg/ml) did not significantly differ between supernatants of activated CD56dim and CD56bright NK cells. CBA data were supported by results from the protein array after 48 h stimulation (data not shown). The discrepancy between our experiments and the literature might be due to the power or duration of the stimulus. To address this question, we determined mRNA levels of IFN{gamma} and TNF at four time points (0, 2, 4, and 24 h) after stimulation with PMA/ionomycin. Kinetic experiments revealed considerably increased mRNA expression for both factors in CD56bright NK cells as compared with CD56dim NK cells within 4 h, and we determined a faster induction of TNF, which reached maximal RNA level after 2 h, as compared with 4 h for IFN{gamma}. After 24 h, both RNAs were decreased to a quite similar level, indicating a differential regulation of each cytokine (Fig. 6 ). In addition, soluble mediators have short half-lives (on the order of minutes to hours) and are rapidly degraded [33 ]. Furthermore, it cannot be excluded that one or the other cytokine, which is induced in CD56dim or CD56bright NK cells by activation can instantly be consumed and subsequently affect the production of other factors in a loop-back mechanism in the respective population. Thus, one or a combination of the different mechanisms might be responsible for net production of similar IFN{gamma} and TNF levels, as assessed in the above experiments. Other cytokines, which can concomitantly be detected by the CBA (IL-1β, IL-2, IL-4, IL-6, IL-10, and IL-12p70) remained below sensitivity levels (20 pg/ml).


Figure 6
View larger version (14K):
[in this window]
[in a new window]

  		Figure 6. Kinetics of cytokine induction in NK cell subpopulations. Sorted CD56dim (—{circ}—) and CD56bright (—{blacksquare}—) NK cells were stimulated with PMA/ionomycin. At the indicated time points, induction of IFN{gamma} (A) and TNF (B) mRNA was determined by RT-qPCR using primers as described in Material and Methods.

IL-8 production by NK cells
Three different GeneChip arrays revealed a significantly higher expression of IL-8 in CD56dim as compared with CD56bright NK cells. Message for this chemokine was already present in resting cells, indicating a constitutive expression in both NK cell populations. After activation, the signal was drastically increased in both populations without changing the relation between CD56dim and CD56bright NK cells (Fig. 7A ). Supernatants were harvested from sorted CD56dim and CD56bright NK cells after stimulation with PMA/ionomycin for 24 h. Quantitative analyses of the supernatants using cytometric bead arrays confirmed stronger secretion of IL-8 by stimulated CD56dim NK cells as compared with CD56bright NK cells on the protein level, although the difference failed to reach a level of significance (Fig. 7B) . An increase of IL-8 after activation was also detected by intracellular staining. By using this technique, 10% of each NK cell subpopulation could clearly be identified as a source for IL-8 production (Fig. 7C) .


Figure 7
View larger version (18K):
[in this window]
[in a new window]

  		Figure 7. IL-8 production by NK cells. (A) IL-8 encoding mRNA as determined by three different gene chip arrays is shown. Levels of significance are indicated according to the Affymetrix algorithm (***P≤0.001). Data were confirmed on the protein level using two different methods. (B) Supernatants from CD56dim (open bars) and CD56bright NK cells (solid bars) treated with PMA/ionomycin for 24 h, were harvested, and IL-8 was measured by cytometric bead array. (C) Intracellular IL-8 was determined by FACS analysis in both NK cell subsets. ***P ≤ 0.001.

Variety of soluble factors produced by NK cells
To analyze the variety of factors which can be produced by each NK cell population on the protein level, we performed protein arrays covering ~80 factors with supernatants from PMA/ionomycin-activated NK cells. To assess whether quantities of mRNA encoding soluble factors correlate with proteins released by activated NK cells, we used the same experimental samples as sources for both kinds of substances. Supernatants were collected and applied to the protein array (Fig. 8 ). Cells from the same experiments were harvested and subjected to corresponding gene array analyses of activated NK cells. The protein array gives an impression of differential capabilities of both NK populations in respect to secreting soluble factors. In the particular experiment depicted in Fig. 8 A and B (e.g., VEGF, PIGF, and TGFβ2 seem to be exclusively produced by CD56bright NK cells and eotaxin-2, and eotaxin-3 by CD56dim NK cells, respectively). Normalized protein array data disproved exclusive, but confirmed that VEGF, PIGF, and TGFβ2 were preferentially secreted by CD56bright NK cells, while factors like eotaxin-2 and eotaxin-3 were mainly produced by CD56dim NK cells. The protein array revealed more soluble mediators produced by NK cell subpopulations. Similar amounts of GDNF, IGFBP1, TIMP-2, and EGF were secreted by both subsets (Fig. 8D) . IGF-1, eotaxin-3, angiogenin, and IGFBP-3 were preferentially produced by CD56dim (Fig. 8E) , while TARC, PARC, TGFβ3, and IL-15 were predominant in supernatants of CD56bright NK cells (Fig. 8F) .


Figure 8
View larger version (53K):
[in this window]
[in a new window]

  		Figure 8. Protein array analysis of soluble factors in supernatants of activated CD56dim and CD56bright NK cells. (A) Developed membrane of a cytokine/chemokine array analyzing 79 different factors as assessed from activated CD56dim (A) and CD56bright NK cells (B). (C) Legend of the chemokine array explaining location of controls and probes on the membrane. D–F: Quantified results arranged into three groups: secreted to a similar extent by both subsets (D), preferentially produced by CD56dim (open bars, E), and predominantly secreted by CD56bright NK cells (solid bars, F). Means of two independent experiments are shown (D–F). For the two experiments, we used a pool of 7 and 8 different donors, respectively.

Gene arrays revealed significantly higher mRNA levels of MIP1{alpha}, MIP1β, and RANTES in resting and activated CD56dim NK cells (Table 1 ). But MIP1β and RANTES were detected in supernatants of both subpopulations. Similarly, mRNAs for MCP-2 and MCP-3 were only found in activated CD56dim NK cells, but proteins were present in supernatants of both subsets. While equal amounts of MCP-3 and GCP-2 were secreted by CD56dim and CD56bright NK cells, the former produced more MCP-2. I-309 (CCL1), and GM-CSF transcripts were absent in resting NK cells, but, after activation, CCL1 was expressed preferentially, and GM-CSF was expressed exclusively in CD56bright NK cells. Furthermore, mRNAs of some members of the TNF superfamily were up-regulated in this subset after stimulation, e.g., OX-40 Ligand (TNFSF4), TRAIL (TNFSF10), and RANKL (TNFSF11) (Table 1) .


View this table:
[in this window]
[in a new window]

  		Table 1. Factors Produced by Resting and Activated NK Cell Populations

As mentioned above, protein and mRNA levels were analyzed in parallel from the same experiments giving us the opportunity to compare both metabolites representing different steps of the same pathway. For some of the proteins, array analyses corresponded to mRNA levels (e.g., CCL1, 2, 4, 5, 8, IL-16) whereas others did not (e. g. CCL7, CXCL10). Discrepancies between amounts of proteins and the encoding RNAs are not necessarily contradicting. It has to be considered that maturation, stability, and degradation of RNA definitely affect the levels of detectable RNA after 48 h of stimulation. On the other hand, protein secretion could be posttranslationally regulated and free protein concentrations can be secondarily reduced by degradation due to short half-lives or consumption by the cells. This is particularly true for the relatively long incubation time of 48 h and might explain the lack of CXCL1 protein despite the presence of specific RNA. However, detailed investigation of posttranscriptional regulation and kinetics was beyond the scope of this study. Importantly, the differences between CD56dim and CD56bright NK cells in terms of producing bioactive factors and utilization of obviously diverging control mechanisms emphasize the functional diversity of both NK cell subpopulations.


arrow
DISCUSSION
 
Human NK cells are defined as CD3–CD56+ lymphocytes. Although the function of CD56 on lymphocytes is not fully understood, this molecule is expressed in higher density on a subset of NK cells. Because of this phenotype, they are termed CD56bright NK cells. It has already been shown that these cells differ from ‘classical’ CD56dim NK cells not only phenotypically but also functionally [4 5 6 7 , 34 , 35 ]. To characterize both NK cell subpopulations in more detail, we performed gene chip arrays covering ~39,000 transcripts. To evaluate any bias due to a possible singularity of an individual donor or artifacts of pooled RNA, we used both sources for array analysis. Results of both arrays correlated strongly and data obtained previously on the protein level corresponded quite well, indicating the reliability of the array approach. In the end, we used only arrays of pooled samples for further analyses.

Focusing on NK cells only, we detected 473 differentially expressed transcripts with 176 exclusively expressed in CD56dim and 130 in CD56bright NK cells. In accordance with the differential surface expression of the sorted cell populations, higher levels of CD56 mRNA were determined in CD56bright NK cells. In contrast, CD16 mRNA was only detected in CD56dim, although in a few individuals low surface expression of CD16 was observed in CD56bright NK cells. This finding does not necessarily contradict our array data, since lack of CD16 mRNA might be caused by posttranscriptional regulation. This is a common regulatory mechanism that affects RNA stability and might account for cell-specific degradation of CD16 mRNA below the detection level of our method [36 ]. Confirming the phenotypical data we previously reported, message for KIRs excluding KIR2DL4 was only present in CD56dim NK cells [5 ]. KIR2DL4 is unique among the KIRs defined so far. Two alleles exist that encode the transmembrane domain of KIR2DL4. Only the 10a allele encodes a full-length receptor that is stably expressed on the surface, whereas the 9a allele results in a truncated cytoplasmic domain [31 , 32 ]. KIR2DL4 can associate with the Fc{epsilon}RI{gamma} chain and mediate activating signals but also has inhibitory potential via an ITIM in its cytoplasmic tail [30 , 37 ]. Unlike other KIRs, 2DL4 promotes cytokine production, but not cytotoxicity, in resting blood NK cells. The lack of almost all KIRs indicates that CD56bright NK cells might mainly be regulated via KLRs, another family of NK cell receptors, which they abundantly express. Whether KIRs with identical extracellular structure have inhibitory or activating potential is determined by the presence of either ITIMs or ITAMs in intracellular residues of the receptors. For this reason, some functionally different KIRs cannot be distinguished by antibodies. Therefore we confirmed all results from the gene arrays for KIRs and also for KLRs by RT-qPCR, indicating that CD56bright NK cells might preferentially be regulated via lectin-like receptors. KLRC1-3 are known to form dimeric molecules with CD94 [38 ]. A significantly higher expression of CD94 in resting CD56dim NK cells but similar levels of the possible dimerization partners (KLRC1-3) in both populations, as assessed in this study, is not necessarily conflicting. There could be a preferential expression of CD94 homodimers on CD56dim NK cells. In contrast to CD94 heterodimers, homodimeric CD94 has been reported to represent a nonfunctional receptor [39 ]. On the other hand, mRNA expression may be differentially regulated in both NK cell subpopulations. This is supported by our data analyzing activated cells since we determined a much stronger down-regulation of CD94 (71%) in CD56dim NK cells as compared with CD56bright NK cells (47%).

CD56dim NK cells have been shown to be main mediators of natural cytotoxicity. This is stressed by the finding of abundant amounts of cytolytic substances like perforin and granzyme A in CD56dim NK cells and the strong capacity of these cells to bind target cells. The array gene profile completely corroborates previous studies and further increases the known number of cytotoxic substances mainly expressed by ‘classical’ NK cells [5 ]. Namely, granzyme B, CTLA1, and granzyme M were found to be expressed at a significantly higher level in CD56dim NK cells as compared with the CD56bright cells. The only exception is granzyme K, which has been shown to substitute for lack of granzyme A function in a knockout model [40 ]. This granzyme is preferentially present in CD56bright NK cells. In human lymph nodes, CD56bright NK cells have been postulated to eliminate immature DCs [12 , 41 , 42 ]. Granzyme K might play a crucial role in this context and may contribute to the low basic cytotoxic capacity observed in CD56bright NK cells. After activation, granzyme K is down-regulated but granzyme B transcription is strongly induced in these cells. This might contribute to enhanced but still moderate cytotoxic function of preactivated CD56bright NK cells, as described previously [7 ]. In addition to granzyme K, CTLA1 and granzyme M are down-regulated in both activated NK populations. This finding needs to be further investigated. It might represent a regulatory mechanism for natural cytotoxicity, possibly based on differential susceptibility of particular target cells toward respective lytic molecules.

Since CD56bright NK cells comprise the main NK population in human lymph nodes, the origin of CD56bright NK cells in this lymphoid tissue is a major point of discussion. Two reasonable explanations for the presence of CD56bright NK cells in lymph nodes are: 1) specialized NK cells are equipped to migrate from the bone marrow to the lymph nodes or 2) immature lymphoid cells reach the lymphoid tissue and differentiate on site into CD56bright NK cells. Recently, the development of CD56bright NK cells from immature CD34+ cells in lymph nodes has been described [43 ]. On the other hand, CD56bright NK cells are usually present in peripheral blood and strong proliferation during reconstitution after hematopoietic stem cell transplantation has been reported [7 ]. The present study revealed that resting and activated CD56bright NK cells express specific molecules needed to migrate into lymph nodes, such as CD62L and CCR7. This argues rather for migration of specialized NK cell subsets into lymphoid tissue as it has been similarly described for T cell populations [44 ].

Most of the genes that we found differentially expressed were confirmed by two other studies analyzing a lower number of genes. Koopman et al. [13 ] compared peripheral blood CD56dim, CD56bright, and decidual CD56bright NK cells by using the Affymetrix HGU95Av2 array covering ~10,000 genes. They demonstrated that decidual NK cells are remarkably different from both peripheral NK cell subpopulations, and they discussed that peripheral CD56bright NK cells are far more similar to the peripheral CD56dim NK cell subset than they are to the decidual CD56bright NK cells. Hanna et al. analyzed ~20,000 genes of peripheral blood CD16+CD56dim, CD16–CD56bright, and activated CD16+CD56dim NK cells by using the Amersham CodeLink Human 20K I Bioarray [45 ]. We used the U133A and U133B chips from Affymetrix, which allowed us to analyze ~33,000 genes of resting peripheral blood CD56dim and CD56bright NK cells. For analysis of activated cells, the U133A chip, which allows analysis of ~22,000, was used. In contrast to the other studies, our NK cell populations were sorted only according to differential CD56 density, thus we included the minor NK cell populations displaying a CD56dimCD16– and CD56brightCD16+ phenotype, respectively. Although, three different chips were applied and cell sources varied a bit according to the preparation regimen, all three studies reveal similar gene profiling results (e.g., distribution of KIRs, cytolytic molecules, chemokine receptors, and adhesion molecules) confirming the reliability of gene arrays.

To investigate gene regulation in both NK cell populations and whether CD56bright NK cells might display an activated stage of CD56dim NK cells, we analyzed gene profiles of both NK cell subpopulations after stimulation with PMA/ionomycin. This stimulus was chosen as a strong polyclonal activator for all NK cells. More physiological stimuli like IL-2 or cross-linking of CD16 were not feasible due to differential expression of receptors involved [6 , 7 , 46 ]. The exemplary analysis of genes, which were at least threefold up-regulated in activated CD56dim NK cells revealed that most of these genes were down-regulated or not even expressed in resting CD56bright NK cells. On the other hand, most of these particular genes were considerably up-regulated in activated CD56bright NK cells, indicating that resting CD56bright NK cells are not closely related to activated CD56dim NK cells. This is in line with the study of Hanna et al., who also analyzed the gene expression profile of in vitro activated CD56dimCD16+ NK cells, however using a different experimental regimen. They stimulated PBMC on an irradiated feeder layer with PHA and IL-2 for 14 days prior to sorting of CD56dim NK cells [45 ]. Comparing resting CD56bright and activated CD56bright in respect to the genes that were at least threefold up-regulated in CD56dim NK cells revealed for most genes a strong up-regulation in CD56bright NK cells after activation. This further underscores that 1) resting CD56bright NK cells are considerably different from activated CD56dim NK cells and 2) resting CD56bright NK cells can effectively be activated. The positive response of resting CD56bright NK cells toward activating stimuli hints that the cells are not terminally matured or bound to undergo apoptosis but can further develop and regulate immune responses probably mainly by producing cytokines. Activated CD56bright NK cells infiltrating skin lesions of patients have recently been shown to accelerate the course of psoriasis by producing TNF and IFN{gamma} [47 ].

Our data clearly show on the gene and protein levels that NK cell populations can respond differently to activation. Receptor ligands XCL1, CCL1, CXCL9, TNFSF4 (OX-40L), and TNFSF14 (LIGHT) were induced in both NK cell subsets upon stimulation. In contrast, CCL4, CSF2 (GM-CSF), and TNFSF11 (RANKL) were up-regulated solely in CD56bright NK cells and XCL1 (lymphotactin), CCL7 (MCP3), CCL8 (MCP2), CXCL1 (GRO{alpha}), CXCL2 (GROβ), CXCL6 (GCP2), and TNFSF3 (lymphotoxin B) were up-regulated only in CD56dim NK cells. Several of the ligands were expressed constitutively in one or the other NK cell population. Together with this finding, the differential production of cytokines/chemokines like MCP2, TGFβ3, TNF{alpha}, IFN{gamma}, and IL-8 by either resting or activated NK cell subpopulations further emphasizes that the human NK cell compartment is composed of functionally distinct subsets. The production of IL-8 is particularly interesting in the sense that this chemokine can potently attract and activate neutrophils. CD56dim NK cells turned out to be better producers of this mediator, which makes sense physiologically, since CD56dim NK cells represent the majority in peripheral blood and can therefore quickly reach every part of the body in order to substantially initiate or perpetuate immune functions. A comparable interplay of NK cells with monocytes in inflammatory sites has been described [10 ]. This means that NK cells do not only interact with cells of the adaptive immune system e.g., via IFN{gamma}, but are also able to regulate other innate immune cells like monocytes and granulocytes. Via secretion of respective cytokines or chemokines, NK cells can differentially attract leukocytes of all lineages to inflammatory sites and activate or cooperate with the cells recruited this way. VEGF and PIGF, which are mainly produced by CD56bright NK cells, have been shown to increase vascular permeability and attract mononuclear phagocytes [48 ]. VEGF-mediated migration can be inhibited by TIMP-2, which is released by activated NK cells [49 ]. Thus NK cells can probably control transmigration of cells by regulating permeability of vessel walls. TGFβ2 is a versatile cytokine, which can potently down-regulate IL-2-induced T and NK cell responses, but also acts as chemoattractant for PMNC [50 , 51 ]. Eotaxins can recruit eosinophils, basophils, and T cells [52 , 53 ]. GDNF has been shown to reduce the amount of TNF secreted by activated PBMC without altering its transcription [54 ]. There was a discrepancy in our experiments between protein and corresponding RNA levels for a few molecules e.g., VEGF or MCP-2. This could probably be explained by differential kinetics of RNA expression for respective mediators, as we have shown for IFN{gamma} and TNF and/or by the well-known instability of cytokine and chemokine mRNA, which is caused by AU-rich elements in their nucleotide sequences [55 ]. Furthermore, some of the induced cytokines can themselves affect the cells in an autocrine fashion, which leads to secondary regulation of secreted factors [56 ]. Among the strongly up-regulated genes in activated CD56bright NK cells, we detected IL-15, IL-16, and IL-2R{alpha} (CD25), suggesting an autocrine pathway for perpetuating activation of CD56bright NK cells. IL-15 and IL-2 are important cytokines promoting NK cell proliferation and their functional properties [57 ]. IL-16 has been shown to induce IL-2R{alpha}, which is needed for assembling the high-affinity IL-2R({alpha},β,{gamma}) [58 ]. These interdependencies make it difficult, in particular, to quantify factors, but it also illustrates the variety of factors that can be produced by NK cells. The multifaceted repertoire of mediators suggests a broad target cell spectrum and that NK cells can, for instance, initialize and perpetuate inflammation by recruiting various white blood cells and supporting their transmigration on site.

In the future, the huge numbers of different factors produced by each population might serve as markers for identification of additional subpopulations based on their specific functional capabilities. Because of the lack of CD56 in mice, NK cell populations resembling human CD56dim and CD56bright NK cells have not been characterized yet. In this study, we aimed for a more detailed definition of functional and phenotypic features of NK cell subpopulations in humans, also with the view to find appropriate markers to define respective murine NK cell subpopulations. Our study revealed that CCR7, CXCR3, IL-7R, IL-12Rβ2, CD44, and CD62L are probable promising candidates to identify CD56bright NK cell corellates. Despite the well-described diverse characteristics of MHC-I-specific NK cell receptors (KIRs vs. Ly49) and their differential regulation as compared with humans, suitable mouse models would allow analysis of kinetics and functional capabilities of NK cell subpopulations in vivo [59 60 61 ].


arrow
ACKNOWLEDGEMENTS
 
This study was supported by the priority program SPP1110 of the Deutsche Forschungsgemeinschaft (grant JA1058) and Graduate Program of Lower Saxony. We thank Dr. Tanja Momot for providing primers and probes for KIRs and Dr. Michael Morgan for thoroughly reading and improving the manuscript.


arrow
FOOTNOTES
 
1 These authors contributed equally to this work. Back

Received March 13, 2006; revised July 17, 2006; accepted July 17, 2006.


arrow
REFERENCES
 
    1
  1. Raulet, D. H. (2004) Interplay of natural killer cells and their receptors with the adaptive immune response Nat. Immunol. 5,996-1002[CrossRef][Medline]
    2. 2
  2. Assarsson, E., Kambayashi, T., Schatzle, J. D., Cramer, S. O., von Bonin, A., Jensen, P. E., Ljunggren, H. G., Chambers, B. J. (2004) NK cells stimulate proliferation of T and NK cells through 2B4/CD48 interactions J. Immunol. 173,174-180[Abstract/Free Full Text]
    3. 3
  3. Cooper, M. A., Fehniger, T. A., Caligiuri, M. A. (2001) The biology of human natural killer-cell subsets Trends Immunol. 22,633-640[CrossRef][Medline]
    4. 4
  4. Jacobs, R., Weber, K., Wendt, K., Heiken, H., Schmidt, R. E. (2004) Altered coexpression of lectin-like receptors CD94 and CD161 on NK and T cells in HIV patients J. Clin. Immunol. 24,281-286[CrossRef][Medline]
    5. 5
  5. Jacobs, R., Hintzen, G., Kemper, A., Beul, K., Behrens, G., Kempf, S., Sykora, K. W., Schmidt, R. E. (2001) CD56bright cells differ in their KIR repertoire and function from CD56dim NK cells Eur. J. Immunol. 31,3121-3126[CrossRef][Medline]
    6. 6
  6. Nagler, A., Lanier, L. L., Cwirla, S., Phillips, J. H. (1989) Comparative studies of human FcRIII-positive and negative natural killer cells J. Immunol. 143,3183-3191[Abstract]
    7. 7
  7. Jacobs, R., Stoll, M., Stratmann, G., Leo, R., Link, H., Schmidt, R. E. (1992) CD16- CD56+ natural killer cells after bone marrow transplantation Blood 79,3239-3244[Abstract/Free Full Text]
    8. 8
  8. Ferlazzo, G., Pack, M., Thomas, D., Paludan, C., Schmid, D., Strowig, T., Bougras, G., Muller, W. A., Moretta, L., Munz, C. (2004) Distinct roles of IL-12 and IL-15 in human natural killer cell activation by dendritic cells from secondary lymphoid organs Proc. Natl. Acad. Sci. USA 101,16606-16611[Abstract/Free Full Text]
    9. 9
  9. Fehniger, T. A., Cooper, M. A., Nuovo, G. J., Cella, M., Facchetti, F., Colonna, M., Caligiuri, M. A. (2003) 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 101,3052-3057[Abstract/Free Full Text]
    10. 10
  10. Dalbeth, N., Gundle, R., Davies, R. J., Lee, Y. C., McMichael, A. J., Callan, M. F. (2004) CD56bright NK cells are enriched at inflammatory sites and can engage with monocytes in a reciprocal program of activation J. Immunol. 173,6418-6426[Abstract/Free Full Text]
    11. 11
  11. Hayakawa, Y., Screpanti, V., Yagita, H., Grandien, A., Ljunggren, H. G., Smyth, M. J., Chambers, B. J. (2004) NK cell TRAIL eliminates immature dendritic cells in vivo and limits dendritic cell vaccination efficacy J. Immunol. 172,123-129[Abstract/Free Full Text]
    12. 12
  12. Della Chiesa, M., Vitale, M., Carlomagno, S., Ferlazzo, G., Moretta, L., Moretta, A. (2003) The natural killer cell-mediated killing of autologous dendritic cells is confined to a cell subset expressing CD94/NKG2A, but lacking inhibitory killer Ig-like receptors Eur. J. Immunol. 33,1657-1666[CrossRef][Medline]
    13. 13
  13. Koopman, L. A., Kopcow, H. D., Rybalov, B., Boyson, J. E., Orange, J. S., Schatz, F., Masch, R., Lockwood, C. J., Schachter, A. D., Park, P. J., et al (2003) Human decidual natural killer cells are a unique NK cell subset with immunomodulatory potential J. Exp. Med. 198,1201-1212[Abstract/Free Full Text]
    14. 14
  14. Lukassen, H. G., Joosten, I., van Cranenbroek, B., van Lierop, M. J., Bulten, J., Braat, D. D., van der Meer, A. (2004) Hormonal stimulation for IVF treatment positively affects the CD56bright/CD56dim NK cell ratio of the endometrium during the window of implantation Mol. Hum. Reprod. 10,513-520[Abstract/Free Full Text]
    15. 15
  15. King, A. (2000) Uterine leukocytes and decidualization Hum. Reprod. Update 6,28-36[Abstract/Free Full Text]
    16. 16
  16. Moretta, A., Bottino, C., Vitale, M., Pende, D., Cantoni, C., Mingari, M. C., Biassoni, R., Moretta, L. (2001) Activating receptors and coreceptors involved in human natural killer cell-mediated cytolysis Annu. Rev. Immunol. 19,197-223[CrossRef][Medline]
    17. 17
  17. Arnon, T. I., Achdout, H., Lieberman, N., Gazit, R., Gonen-Gross, T., Katz, G., Bar-Ilan, A., Bloushtain, N., Lev, M., Joseph, A., et al (2004) The mechanisms controlling the recognition of tumor- and virus-infected cells by NKp46 Blood 103,664-672[Abstract/Free Full Text]
    18. 18
  18. Mandelboim, O., Lieberman, N., Lev, M., Paul, L., Arnon, T. I., Bushkin, Y., Davis, D. M., Strominger, J. L., Yewdell, J. W., Porgador, A. (2001) Recognition of haemagglutinins on virus-infected cells by NKp46 activates lysis by human NK cells Nature 409,1055-1060[CrossRef][Medline]
    19. 19
  19. Arnon, T. I., Achdout, H., Levi, O., Markel, G., Saleh, N., Katz, G., Gazit, R., Gonen-Gross, T., Hanna, J., Nahari, E., et al (2005) Inhibition of the NKp30 activating receptor by pp65 of human cytomegalovirus Nat. Immunol. 6,515-523[CrossRef][Medline]
    20. 20
  20. Vieillard, V., Strominger, J. L., Debre, P. (2005) NK cytotoxicity against CD4+ T cells during HIV-1 infection: a gp41 peptide induces the expression of an NKp44 ligand Proc. Natl. Acad. Sci. USA 102,10981-10986[Abstract/Free Full Text]
    21. 21
  21. Mathew, P. A., Garni-Wagner, B. A., Land, K., Takashima, A., Stoneman, E., Bennett, M., Kumar, V. (1993) Cloning and characterization of the 2B4 gene encoding a molecule associated with non-MHC-restricted killing mediated by activated natural killer cells and T cells J. Immunol. 151,5328-5337[Abstract]
    22. 22
  22. Nakajima, H., Colonna, M. (2000) 2B4: an NK cell activating receptor with unique specificity and signal transduction mechanism Hum. Immunol. 61,39-43[CrossRef][Medline]
    23. 23
  23. Vivier, E., Morin, P., O’Brien, C., Druker, B., Schlossman, S. F., Anderson, P. (1991) Tyrosine phosphorylation of the Fc gamma RIII(CD16): zeta complex in human natural killer cells. Induction by antibody-dependent cytotoxicity but not by natural killing J. Immunol. 146,206-210[Abstract]
    24. 24
  24. Vivier, E., Morin, P. M., O’Brien, C., Schlossman, S. F., Anderson, P. (1991) CD2 is functionally linked to the zeta-natural killer receptor complex Eur. J. Immunol. 21,1077-1080[Medline]
    25. 25
  25. Tomasello, E., Vivier, E. (2005) KARAP/DAP12/TYROBP: three names and a multiplicity of biological functions Eur. J. Immunol. 35,1670-1677[CrossRef][Medline]
    26. 26
  26. Bauer, S., Groh, V., Wu, J., Steinle, A., Phillips, J. H., Lanier, L. L., Spies, T. (1999) Activation of NK cells and T cells by NKG2D, a receptor for stress-inducible MICA Science 285,727-729[Abstract/Free Full Text]
    27. 27
  27. Selvakumar, A., Steffens, U., Dupont, B. (1996) NK cell receptor gene of the KIR family with two IG domains but highest homology to KIR receptors with three IG domains Tissue Antigens 48,285-294[Medline]
    28. 28
  28. Valiante, N. M., Uhrberg, M., Shilling, H. G., Lienert-Weidenbach, K., Arnett, K. L., D’Andrea, A., Phillips, J. H., Lanier, L. L., Parham, P. (1997) Functionally and structurally distinct NK cell receptor repertoires in the peripheral blood of two human donors Immunity 7,739-751[CrossRef][Medline]
    29. 29
  29. Rajagopalan, S., Long, E. O. (1999) A human histocompatibility leukocyte antigen (HLA)-G-specific receptor expressed on all natural killer cells J. Exp. Med. 189,1093-1100[Abstract/Free Full Text]
    30. 30
  30. Kikuchi-Maki, A., Catina, T. L., Campbell, K. S. (2005) Cutting edge: KIR2DL4 transduces signals into human NK cells through association with the Fc receptor gamma protein J. Immunol. 174,3859-3863[Abstract/Free Full Text]
    31. 31
  31. Rajagopalan, S., Bryceson, Y. T., Kuppusamy, S. P., Geraghty, D. E., van der Meer, A., Joosten, I., Long, E. O. (2006) Activation of NK cells by an endocytosed receptor for soluble HLA-G PLoS Biol. 4,e9[CrossRef][Medline]
    32. 32
  32. Goodridge, J. P., Witt, C. S., Christiansen, F. T., Warren, H. S. (2003) KIR2DL4 (CD158d) genotype influences expression and function in NK cells J. Immunol. 171,1768-1774[Abstract/Free Full Text]
    33. 33
  33. Friberg, D., Bryant, J., Shannon, W., Whiteside, T. L. (1994) In vitro cytokine production by normal human peripheral blood mononuclear cells as a measure of immunocompetence or the state of activation Clin. Diagn. Lab. Immunol. 1,261-268[Medline]
    34. 34
  34. Caligiuri, M. A., Zmuidzinas, A., Manley, T. J., Levine, H., Smith, K. A., Ritz, J. (1990) 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. 171,1509-1526[Abstract/Free Full Text]
    35. 35
  35. Jacobs, R., Heiken, H., Schmidt, R. E. (2005) Mutual interference of HIV and natural killer cell-mediated immune response Mol. Immunol. 42,239-249[CrossRef][Medline]
    36. 36
  36. Young, H. A., Ghosh, P. (1997) Molecular regulation of cytokine gene expression: interferon-gamma as a model system Prog. Nucleic Acid Res. Mol. Biol. 56,109-127[Medline]
    37. 37
  37. Faure, M., Long, E. O. (2002) KIR2DL4 (CD158d), an NK cell-activating receptor with inhibitory potential J. Immunol. 168,6208-6214[Abstract/Free Full Text]
    38. 38
  38. Lazetic, S., Chang, C., Houchins, J. P., Lanier, L. L., Phillips, J. H. (1996) Human natural killer cell receptors involved in MHC class I recognition are disulfide-linked heterodimers of CD94 and NKG2 subunits J. Immunol. 157,4741-4745[Abstract]
    39. 39
  39. Jabri, B., Selby, J. M., Negulescu, H., Lee, L., Roberts, A. I., Beavis, A., Lopez-Botet, M., Ebert, E. C., Winchester, R. J. (2002) TCR specificity dictates CD94/NKG2A expression by human CTL Immunity 17,487-499[CrossRef][Medline]
    40. 40
  40. Shresta, S., Goda, P., Wesselschmidt, R., Ley, T. J. (1997) Residual cytotoxicity and granzyme K expression in granzyme A-deficient cytotoxic lymphocytes J. Biol. Chem. 272,20236-20244[Abstract/Free Full Text]
    41. 41
  41. Ferlazzo, G., Tsang, M. L., Moretta, L., Melioli, G., Steinman, R. M., Munz, C. (2002) Human dendritic cells activate resting natural killer (NK) cells and are recognized via the NKp30 receptor by activated NK cells J. Exp. Med. 195,343-351[Abstract/Free Full Text]
    42. 42
  42. Ferlazzo, G., Munz, C. (2004) NK cell compartments and their activation by dendritic cells J. Immunol. 172,1333-1339[Free Full Text]
    43. 43
  43. Freud, A. G., Becknell, B., Roychowdhury, S., Mao, H. C., Ferketich, A. K., Nuovo, G. J., Hughes, T. L., Marburger, T. B., Sung, J., Baiocchi, R. A., et al (2005) A human CD34(+) subset resides in lymph nodes and differentiates into CD56bright natural killer cells Immunity 22,295-304[CrossRef][Medline]
    44. 44
  44. Sallusto, F., Lenig, D., Forster, R., Lipp, M., Lanzavecchia, A. (1999) Two subsets of memory T lymphocytes with distinct homing potentials and effector functions Nature 401,708-712[CrossRef][Medline]
    45. 45
  45. Hanna, J., Bechtel, P., Zhai, Y., Youssef, F., McLachlan, K., Mandelboim, O. (2004) Novel insights on human NK cells’ immunological modalities revealed by gene expression profiling J. Immunol. 173,6547-6563[Abstract/Free Full Text]
    46. 46
  46. Nagler, A., Lanier, L. L., Phillips, J. H. (1990) Constitutive expression of high affinity interleukin 2 receptors on human CD16-natural killer cells in vivo J. Exp. Med. 171,1527-1533[Abstract/Free Full Text]
    47. 47
  47. Ottaviani, C., Nasorri, F., Bedini, C., de Pita, O., Girolomoni, G., Cavani, A. (2006) CD56brightCD16(-) NK cells accumulate in psoriatic skin in response to CXCL10 and CCL5 and exacerbate skin inflammation Eur. J. Immunol. 36,118-128[CrossRef][Medline]
    48. 48
  48. Shen, H., Clauss, M., Ryan, J., Schmidt, A. M., Tijburg, P., Borden, L., Connolly, D., Stern, D., Kao, J. (1993) Characterization of vascular permeability factor/vascular endothelial growth factor receptors on mononuclear phagocytes Blood 81,2767-2773[Abstract/Free Full Text]
    49. 49
  49. Oh, J., Seo, D. W., Diaz, T., Wei, B., Ward, Y., Ray, J. M., Morioka, Y., Shi, S., Kitayama, H., Takahashi, C., et al (2004) Tissue inhibitors of metalloproteinase 2 inhibits endothelial cell migration through increased expression of RECK Cancer Res 64,9062-9069[Abstract/Free Full Text]
    50. 50
  50. Espevik, T., Figari, I. S., Ranges, G. E., Palladino, M. A., Jr (1988) Transforming growth factor-beta 1 (TGF-beta 1) and recombinant human tumor necrosis factor-alpha reciprocally regulate the generation of lymphokine-activated killer cell activity. Comparison between natural porcine platelet-derived TGF-beta 1 and TGF-beta 2, and recombinant human TGF-beta 1 J. Immunol. 140,2312-2316[Abstract]
    51. 51
  51. Parekh, T., Saxena, B., Reibman, J., Cronstein, B. N., Gold, L. I. (1994) Neutrophil chemotaxis in response to TGF-beta isoforms (TGF-beta 1, TGF-beta 2, TGF-beta 3) is mediated by fibronectin J. Immunol. 152,2456-2466[Abstract]
    52. 52
  52. Park, Y. R., Choi, S. C., Lee, S. T., Kim, K. S., Chae, S. C., Chung, H. T. (2005) The association of eotaxin-2 and eotaxin-3 gene polymorphisms in a Korean population with ulcerative colitis Exp. Mol. Med. 37,553-558[Medline]
    53. 53
  53. Yamada, H., Hirai, K., Miyamasu, M., Iikura, M., Misaki, Y., Shoji, S., Takaishi, T., Kasahara, T., Morita, Y., Ito, K. (1997) Eotaxin is a potent chemotaxin for human basophils Biochem. Biophys. Res. Commun. 231,365-368[CrossRef][Medline]
    54. 54
  54. Vargas-Leal, V., Bruno, R., Derfuss, T., Krumbholz, M., Hohlfeld, R., Meinl, E. (2005) Expression and function of glial cell line-derived neurotrophic factor family ligands and their receptors on human immune cells J. Immunol. 175,2301-2308[Abstract/Free Full Text]
    55. 55
  55. Fan, J., Heller, N. M., Gorospe, M., Atasoy, U., Stellato, C. (2005) The role of post-transcriptional regulation in chemokine gene expression in inflammation and allergy Eur. Respir. J. 26,933-947[Abstract/Free Full Text]
    56. 56
  56. Campbell, J. D., Cook, G., Robertson, S. E., Fraser, A., Boyd, K. S., Gracie, J. A., Franklin, I. M. (2001) Suppression of IL-2-induced T cell proliferation and phosphorylation of STAT3 and STAT5 by tumor-derived TGF beta is reversed by IL-15 J. Immunol. 167,553-561[Abstract/Free Full Text]
    57. 57
  57. Farag, S. S., Caligiuri, M. A. (2006) Human natural killer cell development and biology Blood Rev. 20,123-137[CrossRef][Medline]
    58. 58
  58. Cruikshank, W. W., Lim, K., Theodore, A. C., Cook, J., Fine, G., Weller, P. F., Center, D. M. (1996) IL-16 inhibition of CD3-dependent lymphocyte activation and proliferation J. Immunol. 157,5240-5248[Abstract]
    59. 59
  59. Shilling, H. G., McQueen, K. L., Cheng, N. W., Shizuru, J. A., Negrin, R. S., Parham, P. (2003) Reconstitution of NK cell receptor repertoire following HLA-matched hematopoietic cell transplantation Blood 101,3730-3740[Abstract/Free Full Text]
    60. 60
  60. Korten, S., Wilk, E., Gessner, J. E., Meyer, D., Schmidt, R. E. (1999) Altered donor and recipient Ly49+ NK cell subsets in allogeneic H-2d -> H-2b and H-2b -> H-2d bone marrow chimeras J. Immunol. 163,5896-5905[Abstract/Free Full Text]
    61. 61
  61. Wilk, E., Korten, S., Zielinska-Skowronek, M., Schmidt, R. E., Meyer-Olson, D. (2002) Downregulation of Ly49 D expression and its coexpression with Ly49 G2 and Ly49 A after allogeneic bone marrow transplantation Immunobiology 205,267-281[CrossRef][Medline]



This article has been cited by other articles:


Home page
 Brief Funct Genomic ProteomicHome page
G. B. Scott, J. L. Meade, and G. P. Cook
Profiling killers; unravelling the pathways of human natural killer cell function
Brief Funct Genomic Proteomic, January 21, 2008; (2008) elm037v1.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Supplemental Data: jlb.0306191
Right arrow All Versions of this Article:
jlb.0306191v1
80/6/1529    most recent
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Wendt, K.
Right arrow Articles by Jacobs, R.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Wendt, K.
Right arrow Articles by Jacobs, R.