The inhibitory NKR-P1B receptor identifies a subset of rat splenic NK cells that is low in Ly49 receptors but enriched for CD94/NKG2 receptors. We report in this study a novel NKR-P1Bbright NK subpopulation that is prevalent in peripheral blood, liver, and gut-associated lymphoid organs and scarce in the spleen, peripheral lymph nodes, bone marrow, and lungs. This NKR-P1Bbright NK subset displays an activated phenotype, expressing CD25, CD93, CX3CR1 and near absence of CD62-L, CD11b, and CD27. Functionally, NKR-P1Bbright NK cells are highly responsive in terms of IFN-γ production and exert potent cytolytic activity. They show little spontaneous proliferation, are reduced in numbers upon in vivo activation with polyinosinic:polycytidylic acid, and have poor survival in ex vivo cytokine cultures. Our findings suggest that NKR-P1Bbright NK cells are fully differentiated effector cells that rapidly die upon further activation. The identification of this novel rat NK cell subset may facilitate future translational research of the role of distinct NK cell subsets under normal physiological conditions and during ongoing immune responses.

Natural killer cells are important contributors to the early immune defense against infected or transformed cells. Their response is partly controlled by numerous NK cell receptors with both activating and inhibitory functions, including the human killer Ig-like receptors (KIRs), the rodent C-type lectin-like Ly49 receptors, and the CD94/NKG2 and NKR-P1 receptors (1). KIR, Ly49, and CD94/NKG2 receptors bind MHC class I molecules, and their expression is assumed to be a prerequisite for NK cells to acquire full functionality (2). NK cell receptors are gradually acquired during successive developmental stages from NK cell precursors primarily in the bone marrow but also in secondary lymphoid organs (2, 3). Acquisition of NKR-P1 is an early event marking NK-lineage commitment, followed by CD94/NKG2. Ly49 expression is thought to complete the final maturation step toward a full complement of NK cell receptors and effector functions (4, 5). In contrast to mature cells, immature NK cells are not fully capable of mediating lysis of target cells or of producing IFN-γ (2, 6).

The different stages of NK cell differentiation are distinguished by a combination of several markers, but expression of the integrin CD11b defines fully mature NK cells in both human and mouse (7). Mouse NK cells can be further subdivided based on expression of the TNF receptor superfamily member CD27. The most immature NK cells are CD11bCD27 double-negative, which transit through the CD11bCD27+ single-positive and CD11b+CD27+ double-positive stages before reaching the terminally differentiated CD11b+CD27 single-positive stage (810). Immature NK cells are most abundant within the bone marrow and lymph nodes, whereas the mature CD11b+ NK cells reside primarily in the spleen, peripheral blood, and lungs (7, 8, 11). A potential pitfall in defining subsets of differentiating NK cells is the upregulation or downregulation of several surface proteins upon activation of mature NK cells (12). It has been shown that within the pool of mature NK cells, resting cells may coexist with activated, effector, and memory-like cells, all with distinct phenotypes (13).

Human NK cells have traditionally been subdivided into two populations based on their surface density of CD56 (1416). A presumably fully differentiated CD56dim NK cell subset is associated with low CD27 expression and is highly cytotoxic in comparison with a minor subset of less mature CD56brightCD27+ cells (17, 18). The CD56bright NK cells are thought to be the predominant contributor of IFN-γ release upon cytokine stimulation, but CD56dim NK cells also produce high amounts of IFN-γ upon stimulation by cellular targets (19). CD56dim NK cells are further associated with expression of KIRs, whereas CD56bright NK cells preferentially express CD94/NKG2 receptors. A similar subdivision of the two receptor families has been observed in mice, where Ly49 receptors and NKG2 receptors are expressed by overlapping subsets of NK cells (4, 20).

In the rat, we have defined two major NK cell subsets in the spleen based on the complementary expression of either the Ly49s3 (Ly49 stimulatory receptor 3) receptor or the inhibitory NKR-P1B receptor [formerly NKR-P1C (21, 22)]. Ly49 expression is mainly confined to Ly49s3+ NK cells, which can hence be termed Ly49high, whereas NKG2 receptors are preferentially expressed by the Ly49low NKR-P1B+ NK cell subset (23). The inhibitory NKR-P1D receptor in C57BL/6 mice appears to be the functional homolog of the rat NKR-P1B receptor, and this receptor has been associated with a subset of NK cells with reduced expression of Ly49 receptors and elevated expression of NKG2 receptors (20). It should be noted that with the exception of NKR-P1F and -G, it is difficult to establish orthologous relationships between mouse and rat NKR-P1 receptors, presumably because of extensive homogenization of their ligand-binding parts (24, 25).

The separation of KIR/Ly49 and NKG2 into overlapping subsets thus appears to be a common phenomenon across the species. However, there is no clear separation of cytotoxicity and cytokine-producing functionalities between rodent NK cell subsets. In the rat, both NKR-P1B+ and Ly49s3+ NK cells produce comparable amounts of cytokines upon stimulation and are almost equally effective in lysing certain NK-sensitive tumor targets (21, 23). In the mouse, the Ly49low subset appears to be somewhat more cytotoxic and to produce more cytokines than the Ly49high subset (8, 20).

In this study, we have identified a novel NK cell subset predominant in the blood, liver, and gut-associated lymphoid organs that is characterized by enhanced expression levels of NKR-P1B. These NKR-P1Bbright NK cells display a unique activated phenotype and functional characteristics compatible with mature effector cells.

Eight- to twelve-week-old PVG-RT7b (PVG.7B) rats were used. The PVG.7B strain expresses the CD45 allotype (RT7.2) but is otherwise used interchangeably with the standard PVG strain (RT7.1). In some experiments, intra-MHC recombinant PVG.R23 or PVG-RT1u (PVG.1U) rats on the PVG strain background were used. These rat strains have been maintained at the Institute of Basic Medical Sciences (University of Oslo, Norway) for >20 generations. Rats were housed in compliance with guidelines set by the Experimental Animal Board under the Ministry of Agriculture of Norway and by The European Convention for the Protection of Vertebrate Animals Used for Experimental and Other Scientific Purposes. The laboratory animal facilities are subject to a routine health-monitoring program and tested for infectious organisms according to a modification of Federation of European Laboratory Animal Science Associations recommendations.

Abs against CD3 (G4.18–FITC or –PE), NKR-P1A (10/78–PE), CD11b (WT.5–biotin), and CD27 (LG.3A10–PE) and streptavidin–PerCP were obtained from BD Biosciences (Franklin Lakes, NJ). Polyclonal goat anti-CX3CR1 was from Santa Cruz Biotechnology (Santa Cruz, CA) and was conjugated to Alexa488 (Sigma-Aldrich, St. Louis, MO) in our laboratory. BioM2 (anti-FLAG; Sigma-Aldrich) was used as isotype control Ab (IgG1). Anti–c-Kit was from Neuromics (Minneapolis, MN) and detected by FITC anti-rabbit IgG (Jackson Immunoresearch Laboratories, Suffolk, U.K.). Anti-NKG2D was a kind gift from Dr. Sheri Krams (Stanford, CA) and was conjugated to Pacific blue (Invitrogen) in our laboratory. mAbs to NKR-P1A [3.2.3–biotin; from J.C. Hiserodt, Pittsburgh, PA (26)], NKR-P1B [STOK27–Alexa647 (21)], Ly49s3/i3/s4/i4 [DAR13–biotin or –Atto488 (27)], Ly49s5/i5 [Fly5–biotin (28)], Ly49i2 [STOK2–biotin (29)], KLRH1 [STOK9–biotin (30)], NKp46 [Wen23–biotin or –Pacific blue (31)], CD8 (OX8–biotin), CD25 (OX39–biotin), CD43 (W3/13–FITC), CD62-L (OX85–biotin), CD90 (OX7–biotin), and CD93 [Lov3–biotin (32)] were generated from hybridomas in our laboratory and conjugated according to standard protocols. Rat recombinant IL-2 was obtained from dialyzed cell culture supernatants of a CHO cell line stably transfected with a rat IL-2 expression construct (33), rat recombinant IL-12 was from BioSource (Invitrogen), rat recombinant IL-15 and mouse recombinant CXCL12 (SDF-1α) and CXCL10 (IP-10) from PeproTech (Rocky Hill, NJ), and rat recombinant IL-18 from R&D Systems (Minneapolis, MN).

Mononuclear cells were prepared from peripheral blood (PBMC), spleen (depleted for Ig+ cells with sheep anti-rat IgG Dynabeads; Invitrogen Dynal, Oslo, Norway), cervical, inguinal, and mesenteric lymph nodes, bone marrow, liver, Peyer’s patches, lungs, thymus, and the peritoneal cavity. Lymphocytes from Peyer’s patches and lungs were obtained by shaking the tissue samples at 37°C for 20 min in PBS supplemented with 0.1% EDTA, 0.3 mg/ml DTT, and 5% FBS, crushing them through a cell strainer, and separation using Lymphoprep. Peritoneal cells were harvested by flushing the peritoneum with cold PBS. Liver NK cells were harvested by perfusion of the liver ex vivo by PBS containing 10 mM EDTA. In short, rats were anesthetized with Hypnorm/Dormicum, and the portal vein was cannulated with an i.v. catheter fixed with suture. Vena cava was cut open, and the liver was perfused with ∼10 ml PBS. Lymphocytes were eluted by injecting 40 ml PBS containing 10 mM EDTA. One million cells were labeled with different combinations of the Abs listed earlier and analyzed by four-color flow cytometry with a FACSCalibur or LSRII (BD Biosciences). NK cells in all tissues were defined and gated as NKR-P1A+CD3 unless stated otherwise, as shown in Supplemental Fig. 1A. For sorting of NK cell subsets, PBMCs were enriched for NK cells by negative selection using a mixture of Abs toward T cells (R73 and OX19), B cells (OX12 and OX33), monocytes (ED1), and macrophages/granulocytes (OX41) and Pan mouse IgG Dynabeads (Invitrogen Dynal). Cells were surface stained with appropriate Abs and sorted with a FACSAria (BD Biosciences). Large granular lymphocyte morphology was visually confirmed in microscope by cytospin from all subsets, and all subsets expressed similar levels of perforin and granzyme B (see Fig. 3).

FIGURE 3.

RT-PCR analysis of the expression of various markers by NKR-P1BbrightLy49s3, NKR-P1Bdim, and Ly49s3+ NK cells. NKR-P1BbrightLy49s3, NKR-P1Bdim, or Ly49s3+ single-positive PBMC NK cells were analyzed for relative expression levels of the indicated mRNAs by semiquantitative RT-PCR. The three NK cell subsets were obtained by FACS sorting. Tenfold dilutions of cDNA were analyzed, with CD45 as a cDNA-loading control.

FIGURE 3.

RT-PCR analysis of the expression of various markers by NKR-P1BbrightLy49s3, NKR-P1Bdim, and Ly49s3+ NK cells. NKR-P1BbrightLy49s3, NKR-P1Bdim, or Ly49s3+ single-positive PBMC NK cells were analyzed for relative expression levels of the indicated mRNAs by semiquantitative RT-PCR. The three NK cell subsets were obtained by FACS sorting. Tenfold dilutions of cDNA were analyzed, with CD45 as a cDNA-loading control.

Close modal

cDNA was generated from RNA extracted from sorted, single-positive NKR-P1Bbright(Ly49s3), NKR-P1Bdim(Ly49s3), and Ly49s3+(NKR-P1B) NK cells. Tenfold dilutions of cDNA, adjusted according to expression levels of CD45, were subjected to PCR with GoTaq polymerase (Promega, Madison, WI). Refer to Supplemental Table I for PCR primer sequences.

Stimulation of PBMCs by plate-bound Abs was performed by coating flat-bottom 96-well plates with 10 μg/ml F(ab′)2 goat anti-mouse IgG in PBS and adding freshly isolated PBMCs stained with anti–NKR-P1A. Cytokine stimulation, with either IL-2 alone or in combination with IL-12 (2 ng/ml) or IL-18 (25 ng/ml), was performed overnight in round-bottom 96-well plates at a cell concentration of 3 × 105/well. For the last 4 h of the stimulation, brefeldin A (Sigma-Aldrich) was added at 10 μg/ml. Cells were harvested and stained for four-color analysis with Abs against NKp46 (Wen23–biotin), Ly49s3 (DAR13–Atto488), and NKR-P1B (STOK27–A647). Afterward, the cells were fixed for 10 min in 2% paraformaldehyde, permeabilized with 0.5% saponin in PBS for 20 min, and stained with PE-conjugated anti-rat IFN-γ (BD Biosciences). Samples were analyzed on a FACSCalibur.

A standard 51Cr release assay was performed as previously described (34) with the NK cell-sensitive tumor cell line YAC-1, the mouse myeloma cell line NS0, or NS0 stably transfected with EMCV.Clr11-FLAG (22) as targets. Transfection of NS0 with 20 μg DNA was performed on 4 × 106 cells grown in antibiotic-free medium by using the BioRad GenePulser XCell (120 V, 950 μF, Ω = ∞). Cells were rested for 24 h in normal culture medium and then subjected to selection by 1 mg/ml of active G418 sulfate. FACS-sorted PBMC NK cell populations were rested for 1 h at 37°C prior to incubation with target cells at the indicated E:T ratios.

Chemotaxis was measured using Corning Transwell permeable polycarbonate inserts with 5-μm pores in a 24-well format (Sigma-Aldrich). Cells were washed in RPMI 1640 supplemented with 2% FBS and 25 mM HEPES and resuspended at 1 × 107 cells/ml. PBMCs (100 μl) were added to the insert in the absence or presence of either 100 ng/ml CXCL10 or CXCL12 in the lower compartment. Assays were performed in duplicate or triplicate at 37°C and 5% CO2 for 2 h. Cells that had passed through the filter into the lower chamber were collected and stained with the indicated Abs and analyzed by flow cytometry. Cell counts were normalized with an internal bead control (10 μl/sample of FITC-conjugated beads; Bangs Laboratories, Fishers, IN).

Rats were injected i.p. with 5 mg BrdU (BD Biosciences) diluted in PBS 3 h before rats were sacrificed. Isolated cells were stained with the indicated Abs, fixed, permeabilized, and DNAse-treated prior to staining for BrdU incorporation using the FITC BrdU Flow Kit (BD Biosciences) according to the manufacturer’s protocol. Cells were stained with anti-BrdU, 7-aminoactinomycin D, and the indicated Abs. For in vivo activation of lymphocytes, rats were stimulated with 1 mg polyinosinic:polycytidylic acid [poly(I:C)] (Sigma-Aldrich) in 0.5 ml 0.9% NaCl i.v. or i.p. 20–24 h before analysis. Control rats received 0.5 ml 0.9% NaCl only.

FACS-sorted, single-positive NKR-P1Bbright(Ly49s3), NKR-P1Bdim(Ly49s3), and Ly49s3+(NKR-P1B) NK cells from peripheral blood were labeled with 5 μM CFSE (Invitrogen, Molecular Probes) at 1 × 107 cells/ml in PBS, 2% FCS for 10 min at 37°C. Purity of sorted populations was tested on a FACSAria. Cells were washed and plated into 96-well plates at 1 × 105 cells/ml. IL-2 and IL-15 (10 ng/ml) were added to duplicate wells. Cells were harvested at indicated time points, stained, and analyzed by flow cytometry (FACSCalibur). Analysis of CD27 and CD11b expression after IL-2 stimulation was performed on sorted CD27+CD11b−/+, CD27CD11b, and CD27CD11b+ splenic NK cells. After sorting, cells were transferred to a 24-well culture plate and cultured in RPMI 1640 supplemented with 10% FBS, 2-mercaptoethanol, sodium pyruvate, antibiotics, and rat recombinant IL-2 and IL-15 (10 ng/ml). Cells were harvested at days 2 and 7 and restained with Abs before analysis by flow cytometry.

Experimental values were expressed as mean ± SEM. The statistical significance of differences between two mean values was evaluated by the two-tailed unpaired t test, where values of p < 0.05 were considered to be statistically significant.

We have previously identified and characterized two major subsets of splenic NK cells with the complementary expression of either Ly49s3 or NKR-P1B (21, 23). When we extended this analysis to peripheral blood, we could discriminate between distinct NK subpopulations based on the surface intensity of NKR-P1B; that is, NKR-P1Bdim and NKR-P1Bbright (Fig. 1). The NKR-P1BdimLy49s3 and Ly49s3+NKR-P1B NK cells corresponded with the previously identified Ly49low and Ly49high NK subsets in the spleen (21, 23), and are hereafter referred to as NKR-P1Bdim and Ly49s3+ cells, respectively. In addition, a novel NKR-P1Bbright subset could be subdivided into Ly49s3+ and Ly49s3 fractions. Thus, based on the expression profile of Ly49s3 and NKR-P1B, it is possible to discern four characteristic rat NK subsets in the peripheral blood (i.e., NKR-P1BbrightLy49s3, NKR-P1BbrightLy49s3+, NKR-P1Bdim, and Ly49s3+).

FIGURE 1.

A novel NKR-P1BbrightLy49s3+/− NK cell subset is prevalent in peripheral blood, gut-associated lymphoid organs, and in liver. The distribution of Ly49s3+ and NKR-P1B+ NK cell subsets was analyzed by four-color flow cytometry in the blood, liver, Peyer’s patches, mesenteric lymph nodes, thymus, spleen, cervical lymph nodes, inguinal lymph nodes, bone marrow, and lungs of 8- to 12-wk-old PVG.7B rats. NK cells were gated as NKR-P1A+CD3 cells (Supplemental Fig. 1). The plots and values shown are representative of at least three independent experiments. BM, bone marrow; CLN, cervical lymph nodes; ILN, inguinal lymph nodes; MLN, mesenteric lymph nodes; PP, Peyer's patches.

FIGURE 1.

A novel NKR-P1BbrightLy49s3+/− NK cell subset is prevalent in peripheral blood, gut-associated lymphoid organs, and in liver. The distribution of Ly49s3+ and NKR-P1B+ NK cell subsets was analyzed by four-color flow cytometry in the blood, liver, Peyer’s patches, mesenteric lymph nodes, thymus, spleen, cervical lymph nodes, inguinal lymph nodes, bone marrow, and lungs of 8- to 12-wk-old PVG.7B rats. NK cells were gated as NKR-P1A+CD3 cells (Supplemental Fig. 1). The plots and values shown are representative of at least three independent experiments. BM, bone marrow; CLN, cervical lymph nodes; ILN, inguinal lymph nodes; MLN, mesenteric lymph nodes; PP, Peyer's patches.

Close modal

A more extensive analysis of NK cell receptor coexpression showed that a considerable fraction of the NKR-P1BbrightLy49s3+ cells coexpressed Ly49i2 and Ly49s5/i5, in contrast to the NKR-P1BbrightLy49s3 cells (Fig. 2A). The KLRH1 receptor, which is encoded by a gene in the proximal end of the Ly49 cluster in the NK gene complex (30), was also mainly expressed by the NKR-P1BbrightLy49s3+ cells (Fig. 2A). These data indicate that a mechanism for clustering of Ly49 receptors operates within Ly49s3-expressing cells independent of their expression of the NKR-P1B receptor.

FIGURE 2.

Expression of NK cell receptors and other markers by NKR-P1BbrightLy49s3+/− NK cells, which display a unique CD25+CD93brightCX3CR1+CD117lowCD62L phenotype. Expression of typical NK cell receptors (A) and other markers (B) by NKR-P1BbrightLy49s3, NKR-P1BbrightLy49s3+, NKR-P1Bdim, and Ly49s3+ subsets of PBMC NK cells (NKR-P1Abright) was analyzed by four-color flow cytometry. Filled histograms: stainings with isotype-matched control Abs. Results are representative of two to three independent experiments.

FIGURE 2.

Expression of NK cell receptors and other markers by NKR-P1BbrightLy49s3+/− NK cells, which display a unique CD25+CD93brightCX3CR1+CD117lowCD62L phenotype. Expression of typical NK cell receptors (A) and other markers (B) by NKR-P1BbrightLy49s3, NKR-P1BbrightLy49s3+, NKR-P1Bdim, and Ly49s3+ subsets of PBMC NK cells (NKR-P1Abright) was analyzed by four-color flow cytometry. Filled histograms: stainings with isotype-matched control Abs. Results are representative of two to three independent experiments.

Close modal

A more comprehensive analysis of phenotypic differences between the different subsets was performed by flow cytometry or semiquantitative PCR (NKR-P1BbrightLy49s3+ cells were not included for the latter). All NK cell subsets uniformly expressed NKp46 and NKG2D receptors (Fig. 2A) and comparable levels of CD16 (Fig. 3). Also, the majority of the NKR-P1Bbright(Ly49s3+/−) NK cells expressed CD8α but not CD90, CD11c, or MHC class II (Fig. 2B and data not shown). In agreement with our previous observations, NKG2A was mainly expressed by the NKR-P1B+ cells, with only small differences between NKR-P1Bdim or NKR-P1BbrightLy49s3 NK cells (Fig. 3).

Expression level of CD117 (c-Kit) was reduced for the NKR-P1BbrightLy49s3+/− subset compared with the NKR-P1Bdim or Ly49s3+ subsets, indicating that the former are more differentiated than the latter (Fig. 2B). The NKR-P1BbrightLy49s3+/− NK cells also completely lacked CD62L, expressed lower levels of KLRG1 and CD127, and higher levels of CD25 and CD93 (Figs. 2B, 3), in contrast to both NKR-P1Bdim and Ly49s3+ NK cells. Based on these findings, we concluded that NKR-P1BbrightLy49s3+/− NK cells are phenotypically distinct, displaying a characteristic CD62LCD25+CD93bright profile. Such cells were observed not only in PVG strain rats but also in AO, which is closely related to PVG in the NK gene complex (35).

We next performed an analysis of the distribution of NKR-P1Bbright NK cells in different organs. In agreement with previous results (21), both Ly49s3+ and NKR-P1Bdim NK cell subsets were present in all organs tested, albeit at different ratios and numbers. A marked selectivity of NKR-P1Bbright NK cells was observed with primary localization to mesenteric lymph nodes, Peyer’s patches, and liver, in addition to peripheral blood, and near absence from bone marrow and cervical and inguinal lymph nodes (Fig. 1). Notably, the subset of NKR-P1BbrightLy49s3+ NK cells was nearly absent in the gut but constituted a significant fraction of blood and liver NK cells. The NKR-P1Bbright NK cells observed in the gut and liver, as well as in the spleen and lungs, shared the characteristic CD25+CD93brightCD62L phenotype displayed by NKR-P1Bbright NK cells in the blood (Fig. 4), indicating that these populations are related. Of note, NKR-P1Bbright NK cells in Peyer’s patches and lungs expressed particularly high levels of CD25. The majority of thymic NK cells (∼80%) were NKR-P1B+Ly49s3 (Fig. 1), in combination with expression of a CD25+CD93bright, but CD62L+ phenotype (Fig. 4). Thymic NK cells thus partially resemble NKR-P1Bbright NK cells.

FIGURE 4.

CD25+CD93brightCD62L NKR-P1Bbright NK cells are prevalent in blood, liver, and gut-associated lymphatic organs and represent a minor subset in peripheral lymph nodes, spleen, lungs, and bone marrow. The expression of CD62L, CD25, and CD93 was analyzed on NKR-P1B+ NK cells from blood, liver, Peyer’s patches, mesenteric lymph nodes, thymus, spleen, cervical lymph nodes, inguinal lymph nodes, bone marrow, and lungs of 8- to 12-wk-old PVG.7B rats by four-color flow cytometry. NK cells were gated as NKR-P1A+CD3 cells. Results are representative of two to three independent experiments. BM, bone marrow; CLN, cervical lymph nodes; ILN, inguinal lymph nodes; MLN, mesenteric lymph nodes; PP, Peyer's patches.

FIGURE 4.

CD25+CD93brightCD62L NKR-P1Bbright NK cells are prevalent in blood, liver, and gut-associated lymphatic organs and represent a minor subset in peripheral lymph nodes, spleen, lungs, and bone marrow. The expression of CD62L, CD25, and CD93 was analyzed on NKR-P1B+ NK cells from blood, liver, Peyer’s patches, mesenteric lymph nodes, thymus, spleen, cervical lymph nodes, inguinal lymph nodes, bone marrow, and lungs of 8- to 12-wk-old PVG.7B rats by four-color flow cytometry. NK cells were gated as NKR-P1A+CD3 cells. Results are representative of two to three independent experiments. BM, bone marrow; CLN, cervical lymph nodes; ILN, inguinal lymph nodes; MLN, mesenteric lymph nodes; PP, Peyer's patches.

Close modal

The tissue-specific compartmentalization of NKR-P1BbrightLy49s3+/− NK cells suggested that these cells express a deviant pattern of chemokine responsiveness. We found that NKR-P1BbrightLy49s3+/− NK cells selectively expressed the fractalkine receptor CX3CR1 (Fig. 2B). With regard to other chemokine receptors, RT-PCR analyses suggested a tendency of higher expression levels of CXCR6 and CCR5 and lower levels of CCR7 and CCR9 by NKR-P1Bbright NK cells in comparison with NKR-P1Bdim and Ly49s3+ cells (Fig. 3). Although we detected similar levels of CXCR4 in all NK cell subsets, blood NKR-P1BbrightLy49s3+/− NK cells responded poorly to the CXCR4 ligand CXCL12 in an in vitro Transwell chemotaxis assay compared with NKR-P1Bdim and Ly49s3+ NK cells (Fig. 5). All three subsets showed comparable responses to the CXCR3 ligand CXCL10 (Fig. 5). We were unfortunately not able to assess responsiveness toward CX3CR1, as rat lymphocytes failed to respond to the commercially available CX3CL1 chemokines. In summary, the observed differences in expression patterns of chemokine receptors could direct specific tissue distribution of NKR-P1Bbright NK cells.

FIGURE 5.

Distinct chemotactic responsiveness of NKR-P1BbrightLy49s3+/− NK cells. Chemotactic responsiveness of NKR-P1BbrightLy49s3+/−, NKR-P1Bdim, or Ly49s3+ NK cells was determined by a Transwell chemotaxis assay against either CXCL10 or CXCL12 (both 100 ng/ml). Bulk PBMCs were added to the upper chamber of the Transwells. Migrating cells were harvested from the lower chamber and enumerated by four-color flow cytometry. Values represent the mean chemotactic index (cells migrated toward the indicated chemokine/cells migrated toward medium alone) of at least three independent experiments ±SEM.

FIGURE 5.

Distinct chemotactic responsiveness of NKR-P1BbrightLy49s3+/− NK cells. Chemotactic responsiveness of NKR-P1BbrightLy49s3+/−, NKR-P1Bdim, or Ly49s3+ NK cells was determined by a Transwell chemotaxis assay against either CXCL10 or CXCL12 (both 100 ng/ml). Bulk PBMCs were added to the upper chamber of the Transwells. Migrating cells were harvested from the lower chamber and enumerated by four-color flow cytometry. Values represent the mean chemotactic index (cells migrated toward the indicated chemokine/cells migrated toward medium alone) of at least three independent experiments ±SEM.

Close modal

Functional differences in IFN-γ production between NKR-P1BbrightLy49s3+/− cells versus NKR-P1Bdim or Ly49s3+ blood NK cells (NKp46+) in response to both Ab and cytokine stimulation were assessed by flow cytometry. PBMCs were stimulated in vitro in the presence of either plate-bound anti–NKR-P1A or isotype control Abs or with IL-2 alone or in combinations with IL-12 or IL-18 for 18 h. Only the NKR-P1Bbright NK cells responded to NKR-P1A ligation, irrespective of their Ly49s3 coexpression (Fig. 6A). Such a functional skewing of the IFN-γ response was also indicated in response to IL-12 or IL-18, where we observed more IFN-γ–positive cells among the two NKR-P1Bbright populations (Ly49s3+ and Ly49s3) compared with the Ly49s3+ or NKR-P1Bdim subsets (Fig. 6A). It was also clear that Ly49s3+ cells produced more IFN-γ compared with NKR-P1Bdim cells, in line with our previous observation using IL-2–activated NK cells (21). In contrast to the other NK cell subsets, NKR-P1BbrightLy49s3+/− NK cells also produced IFN-γ in response to IL-2 alone, indicating that these cells are particularly responsive. Also, analyzing the mean fluorescence intensity of IFN-γ–producing cells, the NKR-P1Bbright cells (both Ly49s3+ and Ly49s3 fractions) produced 2- to 3-fold more IFN-γ in response to IL-18 than the other NK subsets, whereas there were only minor differences between the subsets in response to the other stimuli (Supplemental Fig. 2). A double-negative subset of blood NK cells lacking expression of both Ly49s3 and NKR-P1B receptors appeared less responsive in terms of IFN-γ production (Fig. 6A), corroborating recent data obtained with NK subsets from the spleen (23).

FIGURE 6.

NKR-P1BbrightLy49s3+/− NK cells mediate a particularly strong IFN-γ response upon cytokine stimulation and have full cytotoxic capacity. (A) PBMCs were stimulated for 18 h in vitro by the indicated plate-bound Abs or cytokine(s), and the proportion of IFN-γ–positive cells was assessed by four-color flow cytometry, gating on NKR-P1BbrightLy49s3, NKR-P1BbrightLy49s3+, NKR-P1Bdim, or Ly49s3+ single-positive or Ly49s3NKR-P1B double-negative NKp46+ NK cells. The data are representative of three independent experiments. (B and C) The cytolytic activity of purified NKR-P1BbrightLy49s3, Ly49s3+, NKR-P1Bdim single-positive or NKR-P1BLy49s3 PBMC NK cells (NKR-P1A+CD3) was determined against the NK-sensitive tumor target cell lines YAC-1, NS0, and NS0 transfected with Clr11 in a standard 4-h 51Cr release assay. Data represent the mean values of triplicates ±SEM from one representative experiment.

FIGURE 6.

NKR-P1BbrightLy49s3+/− NK cells mediate a particularly strong IFN-γ response upon cytokine stimulation and have full cytotoxic capacity. (A) PBMCs were stimulated for 18 h in vitro by the indicated plate-bound Abs or cytokine(s), and the proportion of IFN-γ–positive cells was assessed by four-color flow cytometry, gating on NKR-P1BbrightLy49s3, NKR-P1BbrightLy49s3+, NKR-P1Bdim, or Ly49s3+ single-positive or Ly49s3NKR-P1B double-negative NKp46+ NK cells. The data are representative of three independent experiments. (B and C) The cytolytic activity of purified NKR-P1BbrightLy49s3, Ly49s3+, NKR-P1Bdim single-positive or NKR-P1BLy49s3 PBMC NK cells (NKR-P1A+CD3) was determined against the NK-sensitive tumor target cell lines YAC-1, NS0, and NS0 transfected with Clr11 in a standard 4-h 51Cr release assay. Data represent the mean values of triplicates ±SEM from one representative experiment.

Close modal

We compared cytolytic capacities of freshly isolated NKR-P1BbrightLy49s3, NKR-P1Bdim, Ly49s3+, and Ly49s3NKR-P1B NK cells from PBMCs after FACS sorting. Their capacity to kill YAC-1 tumor targets was tested in a standard 51Cr release assay. We repeatedly observed that NKR-P1BbrightLy49s3 NK cells killed YAC-1 cells with slightly higher efficiency than either NKR-P1Bdim or Ly49s3+ NK cells (Fig. 6B, two representative experiments).

To ascertain that the bright NKR-P1B staining pattern was due to expression of the NKR-P1B receptor and not to mAb STOK27-mediated cross-reaction with some other surface marker, we tested whether NKR-P1BbrightLy49s3 NK cells responded to its ligand Clr11, using NS0 cells stably transfected with Clr11. Whereas NKR-P1Bbright NK cells, as well as NKR-P1Bdim and Ly49s3+ cells, lysed untransfected NS0 cells at approximately the same levels, both NKR-P1Bbright and NKR-P1Bdim NK cells showed reduced killing of NS0.Clr11 transfectant targets (Fig. 6C). The opposite was observed for Ly49s3+ cells, which showed somewhat increased killing of NS0.Clr11 targets. This could be due to the engagement of the activating NKR-P1A receptor by Clr11, which has previously been shown to promote natural cytotoxicity (22). It should be noted that Ly49s3NKR-P1B double-negative NK cells were inefficient killers against both YAC-1 and NS0 compared with the other three subsets, and the presence of Clr11 did not induce killing despite expression of NKR-P1A by these cells (Fig. 6B, 6C), which strengthens the view that they are generally hyporeactive.

The high capacity of NKR-P1BbrightLy49s3+/− NK cells for cytotoxicity and cytokine production and their low c-Kit expression suggest they are a subset of mature NK cells. To test this assumption, we stained NK cells with CD11b and CD27 Abs, which are used to differentiate between maturation stages of NK cells in man and mouse. Blood NK cells clearly separated into three major groups that were either CD11bCD27, CD11bCD27+, or CD11b+CD27, and a minor CD11b+CD27+ fraction (Fig. 7A). The profile was different in the spleen, with a considerably larger proportion of CD27+ cells (CD11b−/+) and much fewer CD11bCD27 cells (Fig. 7A). CD11b+CD27 NK cells are assumed to represent the most mature NK cells (8, 10). In accordance with this, the CD11b+CD27 cells expressed higher levels of the maturation marker CD43 compared with the other fractions (Fig. 7B).

FIGURE 7.

NKR-P1Bbright NK cells lack expression of CD11b and CD27. (A) Distribution of CD27 and CD11b by NK cells (NKp46+) from blood and spleen by four-color flow cytometry. (B) PBMC NK cells were subgated into CD11bCD27, CD11bCD27+, CD11b+CD27+, or CD11b+CD27 cells and analyzed for the expression of CD43 by four-color flow cytometry. (C) Expression of CD11b and CD27 by NKR-P1BbrightLy49s3+/−, NKR-P1Bdim, and Ly49s3+ NKp46+ PBMC NK cells as determined by five-color flow cytometry. Numbers represent percentages of cells within each quadrant. Data shown are representative of three independent experiments.

FIGURE 7.

NKR-P1Bbright NK cells lack expression of CD11b and CD27. (A) Distribution of CD27 and CD11b by NK cells (NKp46+) from blood and spleen by four-color flow cytometry. (B) PBMC NK cells were subgated into CD11bCD27, CD11bCD27+, CD11b+CD27+, or CD11b+CD27 cells and analyzed for the expression of CD43 by four-color flow cytometry. (C) Expression of CD11b and CD27 by NKR-P1BbrightLy49s3+/−, NKR-P1Bdim, and Ly49s3+ NKp46+ PBMC NK cells as determined by five-color flow cytometry. Numbers represent percentages of cells within each quadrant. Data shown are representative of three independent experiments.

Close modal

We found that the majority of NKR-P1Bdim and Ly49s3+ NK cells in the blood were CD11b+CD27 (Fig. 7C), whereas the same subsets in the spleen were mainly CD27+CD11b+/− (data not shown). Surprisingly, the majority of NKR-P1BbrightLy49s3+/− NK cells lacked expression of both CD11b and CD27, which explained why the fraction of CD11bCD27 NK cells was much larger in the blood than in spleen (Fig. 7A), as the NKR-P1BbrightLy49s3+/− cells compose a much larger proportion of blood NK cells.

Lack of expression of CD11b and CD27 could be taken as evidence that NKR-P1Bbright NK cells are immature (7, 10). However, NKR-P1BbrightLy49s3+/− NK cells are clearly highly functional with respect to both cytotoxicity and cytokine production, suggesting that this is not the case. Because CD11b, CD27, and CD43 can all be downregulated as a result of cellular activation (3638), we hypothesized that this could be the case for rat NK cells. To test this, combinations of CD27+ and CD11b+ NK cells were purified from spleen cells by FACS sorting and cultured in IL-2 for up to 7 d. CD11bCD27 cells were included as a negative control. We observed an almost complete loss of CD11b expression and downregulation of CD27 on a majority of the cells, whereas CD11bCD27 NK cells maintained their negative phenotype (Fig. 8). These data show that rat NK cells lose expression of CD11b and CD27 as a consequence of cellular activation in vitro. If recapitulated in vivo, this could explain the observed double-negative phenotype of NKR-P1BbrightLy49s3+/− NK cells.

FIGURE 8.

Rat NK cells lose CD11b and CD27 upon activation with IL-2 in vitro. CD27+ and CD11b+ single-positive and CD27CD11b double-negative splenic NK cells were sorted by FACS and cultured with IL-2. Shown is the modulation of expression levels of CD11b and CD27 analyzed by four-color flow cytometry at the indicated time points of culture.

FIGURE 8.

Rat NK cells lose CD11b and CD27 upon activation with IL-2 in vitro. CD27+ and CD11b+ single-positive and CD27CD11b double-negative splenic NK cells were sorted by FACS and cultured with IL-2. Shown is the modulation of expression levels of CD11b and CD27 analyzed by four-color flow cytometry at the indicated time points of culture.

Close modal

Proliferative capacities are reduced as cells differentiate, and we hypothesized that NKR-P1Bbright NK cells were fully differentiated cells with a limited ability to proliferate. We purified NKR-P1BbrightLy49s3+/−, NKR-P1Bdim, and Ly49s3+ blood NK cells and tested their in vitro proliferative capacities in response to IL-2. Cells were stained with CFSE to monitor cell divisions, and IL-15 was added to increase cell survival. We found that <5% of the NKR-P1BbrightLy49s3+/− NK cells were alive after 4 d compared with ∼50% survival for the other two populations (Fig. 9A). Massive cell death within the NKR-P1BbrightLy49s3+/− population was observed already after 24 h of culture (data not shown). Analyzing the CFSE-dilution profile at day 4, we observed that the majority of the remaining, live NKR-P1Bbright NK cells had undergone only one cell division, indicating an overall poor capacity to proliferate. In contrast, NKR-P1Bdim and in particular Ly49s3+ cells had undergone several cell divisions at this time point.

FIGURE 9.

Low proliferative response and survival of NKR-P1BbrightLy49s3+/− NK cells in response to activation in vitro and in vivo. (A) FACS-sorted, CFSE-labeled NKR-P1BbrightLy49s3+/− or NKR-P1Bdim and Ly49s3+ single-positive NK cells were cultured for 4 d in the presence of IL-2 and IL-15. The CFSE dilution profiles were analyzed by flow cytometry. Dead cells were excluded from the analysis by propidium iodide. Data are representative of two independent experiments. (B) Steady-state proliferation of blood NK cells was analyzed by an acute 3-h in vivo BrdU labeling assay. The proportion of BrdU-positive cells was assessed by flow cytometric analysis, gating on NKR-P1BbrightLy49s3+/− or NKR-P1Bdim and Ly49s3+ single-positive NK cells. Data are representative of three independent experiments. (C) Twenty-hour in vivo poly(I:C) stimulation i.p. reduces the proportion of NKR-P1BbrightLy49s3+/− NK cells in the peritoneum and blood. (D) Analysis of BrdU labeling in poly(I:C)-stimulated rats [poly (I:C) given 21 h i.v. in advance of BrdU] by flow cytometry. The indicated gate includes S-phase cells by means of 7-aminoactinomycin D (7AAD) and BrdU staining. The data in (C) and (D) are representative of three experiments.

FIGURE 9.

Low proliferative response and survival of NKR-P1BbrightLy49s3+/− NK cells in response to activation in vitro and in vivo. (A) FACS-sorted, CFSE-labeled NKR-P1BbrightLy49s3+/− or NKR-P1Bdim and Ly49s3+ single-positive NK cells were cultured for 4 d in the presence of IL-2 and IL-15. The CFSE dilution profiles were analyzed by flow cytometry. Dead cells were excluded from the analysis by propidium iodide. Data are representative of two independent experiments. (B) Steady-state proliferation of blood NK cells was analyzed by an acute 3-h in vivo BrdU labeling assay. The proportion of BrdU-positive cells was assessed by flow cytometric analysis, gating on NKR-P1BbrightLy49s3+/− or NKR-P1Bdim and Ly49s3+ single-positive NK cells. Data are representative of three independent experiments. (C) Twenty-hour in vivo poly(I:C) stimulation i.p. reduces the proportion of NKR-P1BbrightLy49s3+/− NK cells in the peritoneum and blood. (D) Analysis of BrdU labeling in poly(I:C)-stimulated rats [poly (I:C) given 21 h i.v. in advance of BrdU] by flow cytometry. The indicated gate includes S-phase cells by means of 7-aminoactinomycin D (7AAD) and BrdU staining. The data in (C) and (D) are representative of three experiments.

Close modal

We next investigated the basal proliferation rate in vivo by an acute BrdU incorporation assay in adult rats, which allows the detection of S-phase cells. BrdU was injected i.p. and blood harvested 3 h later. We consistently detected lower BrdU incorporation by NKR-P1BbrightLy49s3+/− NK cells compared with NKR-P1Bdim or Ly49s3+ NK cells, where the Ly49s3+ subset appeared to have the highest spontaneous proliferation rate in vivo (Fig. 9B). To address proliferation of NKR-P1Bbright NK cells upon in vivo activation, rats were stimulated by poly(I:C) i.p. 24 h prior to analysis. The unprimed peritoneal cavity harbors NK cells that were predominantly NKR-P1B+, with similarities to blood NKR-P1BbrightLy49s3+/− NK cells (Fig. 9C, Supplemental Fig. 3). After stimulation with poly(I:C) i.p., the fraction of the NKR-P1BbrightLy49s3+/− NK cells was somewhat reduced (Fig. 9C). There was also a simultaneous reduction in the proportion of NKR-P1BbrightLy49s3+/− NK cells in the blood, which was also observed in response to poly(I:C) given i.v. (Fig. 9C, Supplemental Fig. 2). We did detect a higher fraction of the NKR-P1BbrightLy49s3+/− NK cell subset in the S-phase after i.v. poly(I:C) treatment in comparison with NKR-P1Bdim NK cells (Fig. 9D), but with a net reduction in relative numbers in all organs tested. Taken together, these data suggest that NKR-P1Bbright cells rapidly die upon stimulation both in vitro and in vivo.

Based on the above data, we hypothesized that NKR-P1BbrightLy49s3+/− NK cells are fully differentiated cells on the brink of exhaustion and that we should not see any further phenotypic changes with respect to surface expression of either NKR-P1B or Ly49s3 after in vitro culture in IL-2 and IL-15. NKR-P1BbrightLy49s3, NKR-P1Bdim, or Ly49s3+ cells were FACS sorted and cultured for 7 d in the presence of IL-2 and IL-15. Upon termination of the culture, we found that the initially cultured NKR-P1Bbright NK cells were still NKR-P1Bbright (Fig. 10). Also, the initially cultured NKR-P1Bdim or Ly49s3+ cells remained stable, with the exception of the appearance of some double-positive cells within both groups. These data also suggest that cytokine activation in vitro is not enough to induce a general high expression level of NKR-P1B by NK cells.

FIGURE 10.

NKR-P1Bbright NK cells remain phenotypically stable in cytokine cultures. NKR-P1BbrightLy49s3, NKR-P1Bdim, or Ly49s3+ single-positive blood NK cells (A) were sorted by flow cytometry (FACSAria), cultured for 7 d in the presence of IL-2 and IL-15, and analyzed for changes in the expression of Ly49s3 or NKR-P1B by flow cytometry (FACSCalibur) (B). The numbers represent percentages of cells, and the data are representative of four independent experiments.

FIGURE 10.

NKR-P1Bbright NK cells remain phenotypically stable in cytokine cultures. NKR-P1BbrightLy49s3, NKR-P1Bdim, or Ly49s3+ single-positive blood NK cells (A) were sorted by flow cytometry (FACSAria), cultured for 7 d in the presence of IL-2 and IL-15, and analyzed for changes in the expression of Ly49s3 or NKR-P1B by flow cytometry (FACSCalibur) (B). The numbers represent percentages of cells, and the data are representative of four independent experiments.

Close modal

NK cells are far more heterogeneous than previously appreciated with the identification of novel NK subsets with unique combinations of phenotypes and functionalities in various tissues and with varying degrees of plasticity in response to changing milieus. The distinction between subsets of mature NK cells and of intermediate stages of developing NK cells may not be straightforward. It has been shown that distinctions can be made between subsets of resting, effector, or memory cells among the pool of mature NK cells. It is thus important to gain more phenotypic and functional information to understand how subsets of NK cells participate in innate and adaptive immune responses.

We report here the identification of a novel NKR-P1BbrightLy49s3+/− NK cell subset in the blood of rats with potent cytokine-producing and cytotoxic capacities. The cells are uniformly NKp46+ and have the morphology of large granular lymphocytes similar to NKR-P1Bdim and Ly49s3+ NK cells. The cells are further distinguished by limited expression of CD62L, CD11b, and CD27, relatively low levels of CD117, CD127, and KLRG1, and enhanced expression of CD25, CD93, and CX3CR1. Although the lack of CD11b and CD27 could classify NKR-P1BbrightLy49s3+/− NK cells as immature based on results in human and mouse models (7, 8, 10), our data collectively point to these cells as fully mature NK cells that have been activated at some point to become a type of effector cells: i) We show that both CD11b and CD27 are downregulated on NK cells in response to activation, in concurrence with other reports showing downregulation of CD11b either in response to IL-2 in vitro or upon chronic viral infections in vivo (36, 37). In this context, it is interesting to note that the CD11b and CD27 profile of splenic NK cells in wild mice, which is exposed to a much more challenging microbiota than that in the laboratory mouse, resembles that observed in this study for the rat (39). ii) NKR-P1BbrightLy49s3+/− NK cells are fully functional or even more reactive than NKR-P1Bdim or Ly49s3+ NK cells in terms of IFN-γ production and cytotoxicity, which are not attributes of immature NK cells. iii) NKR-P1BbrightLy49s3+/− NK cells express high levels of the classical activation marker CD25 and display a profile of homing markers commonly associated with activated lymphocytes such as lack of CD62L, reduced responsiveness to the CXCR4 chemokine CXCL12, and high expression of the inflammatory chemokine receptor CX3CR1 (16, 40, 41). iv) NKR-P1BbrightLy49s3+/− NK cells are slightly larger in the forward and light scatter profile in the flow cytometer compared with both Ly49s3+ and NKR-P1Bdim NK cells (data not shown), which is again indicative of cellular activation. Of note, the functional features of NKR-P1BbrightLy49s3+/− NK cells were generally monitored 24 h after their initial isolation from the blood, and the large proportion of cells dying in response to the stimulation were excluded from these analyses. We did not, however, detect basal IFN-γ production in unstimulated NKR-P1BbrightLy49s3+/− NK cells, which indicates that stimulation or restimulation is necessary to detect IFN-γ in these cells (data not shown).

Notably, the phenotype of NKR-P1BbrightLy49s3+/− NK cells shares some striking similarities with that of a subset of short-lived effector CD8+ T cells expressing limited levels of CD62L, CCR7, and CD27, but high levels of KLRG1, and which differentiates from resting, naive T cells that express CD62L and CCR7, but limited KLRG1 (42). A similar differentiation program was recently suggested to operate for NK cells (13). NKR-P1BbrightLy49s3+/− NK cells not only lack CD62L and CD27 but also express less CCR7 transcripts than the other two subsets of NK cells, although they express reduced levels of KLRG1. Moreover, most effector CD8+ T cells have limited capacity for further proliferation and differentiation toward central memory T cells and rather undergo cell death upon further stimulation (42). We repeatedly observed that the majority of the NKR-P1BbrightLy49s3+/− cells died within 24–48 h of culture, after 1 or 2 rounds of divisions, and that the fraction of NKR-P1BbrightLy49s3+/− NK cells in blood was reduced upon in vivo poly(I:C) activation. Also, the reduced proliferative ability of NKR-P1BbrightLy49s3+/− NK cells compared with Ly49s3+ or NKR-P1Bdim NK cells indicates that NKR-P1Bbright NK cells might have reached an endpoint in the differentiation program.

In the mouse, memory NK cells are suggested to bear a KLRG1+CXCR6+ phenotype (13, 43). The putative human counterpart has not yet been established, but recent articles have described the further subdivision of human CD56dim NK cells into subsets of either CD56dimCD57+ and CD56dimCD57 cells or CD56dimCD62L+ and CD56dimCD62L cells (4446). The CD56dimCD57+ and the CD56dimCD62L subsets both appear to represent terminally differentiated cells, characterized by lower proliferative capacity, poor responses to cytokine stimulation, but with intact ability to kill MHC-deficient target cells. The CD56dimCD57+ NK cells also lack CD62L (44) but express CX3CR1 (47), and the CD56dimCD62L subset also expresses less CD27 (46). Again, these are phenotypes with similarities to NKR-P1BbrightLy49s3+/− NK cells, although NKR-P1BbrightLy49s3+/− NK cells mediate brisk effector functions upon both cytokine stimulation and target cell stimulation. Because of the unavailability of suitable Abs, we have not tested the expression of CD57 by rat NKR-P1BbrightLy49s3+/− NK cells, but they do express CXCR6 transcripts. Whether a fraction of NKR-P1BbrightLy49s3+/− NK cells could represent putative rat memory NK cells is unknown at the moment. Evidence has been provided that rat NK cells can show memory-like responses in vivo as a response to repeated allo-immunizations (48).

Apart from the blood, NKR-P1BbrightLy49s3+/− NK cells were predominant in the gut lymphatic tissues, liver, and the peritoneal cavity. Thus, NKR-P1BbrightLy49s3+/− NK cells seem to be largely excluded from the bone marrow and secondary non-gut lymphatic organs, which is also concurrent with their low expression of the lymph node homing molecule CD62L. Of note, we have not investigated NK cells from other mucosal lymphatic tissues, such as tonsils, but there is a small subset of NKR-P1BbrightLy49s3+/− NK cells among lung NK cells. Notably, studies in the mouse have shown that there exist for T cells separate mucosal and peripheral lymph node circulation pathways dictated by a selectivity in expression of chemokine receptors and adhesion molecules. Similar mechanisms likely operate for NK cells, and although speculative at this time point, NKR-P1BbrightLy49s3+/− NK cells could represent cells circulating between gut and liver, both tissues where cells are likely to encounter activation signals.

At this point, we can only speculate as to where and how NKR-P1BbrightLy49s3+/− NK cells are generated. We find it unlikely that NKR-P1BbrightLy49s3+/− NK cells are generated in the peripheral blood, and at least a fraction of blood NKR-P1BbrightLy49s3 NK cells could represent circulating gut or mucosal NK cells. We also find it unlikely that the thymus is an essential contributor of NKR-P1BbrightLy49s3+/− NK cells, as they are scarce in the thymus (>0.01% of the thymocytes) and are present in large numbers in congenitally athymic nude rats (M. Inngjerdingen and L. Kveberg, unpublished observation). The NKR-P1Bbright NK cells expressed reduced levels of CD127, which is associated with thymic NK cells in the mouse (49). However, this does not exclude that thymic NK cells may be related to or contribute to the populations of NKR-P1BbrightLy49s3+/− NK cells found in gut and liver. The nature of the precursor for NKR-P1BbrightLy49s3+/− NK cells is also uncertain, as we could not induce the NKR-P1BbrightLy49s3+/− phenotype from purified populations of NKR-P1Bdim or Ly49s3+ NK cells. However, the activated/effector phenotype certainly suggests they have encountered activation signals, which may be provided at anatomical sites where such activation is likely to occur, like the gut and liver. As to the nature of the activation signal that may induce the NKR-P1BbrightLy49s3+/− phenotype, it is known that combinations of cytokines such as IL-12, IL-15, or IL-18 may lead to the downregulation of CD11b, CD43, and CXCR4 and the upregulation of CD25 on mature bulk mouse splenic NK cells (38). This phenotype is very similar to that of NKR-P1BbrightLy49s3+/− cells. As NKR-P1BbrightLy49s3+/− NK cells show marked tissue localization, we believe it is likely that a combination of both soluble and cell contact-dependent factors is necessary for their generation.

An interesting aspect is the high expression levels of NKR-P1B by NKR-P1Bbright NK cells. It is tempting to speculate that high expression levels of the inhibitory NKR-P1B receptor are useful in situations where NK cells may need to be tolerogenic, as in the gut environment, or to dampen an immune response by activated NK cells in vivo. Also, the expression level of NKR-P1B is likely influenced by the presence of ligands at different anatomical sites (35). NKR-P1B reporter cells respond to leukocytes from diverse lymphoid tissues (22), strongly suggesting that the Clr11 ligand is broadly expressed among hematopoietic cells in line with its mouse homolog, Clrb (50). Our data indicate that cytokine stimulation alone is not enough to induce the high levels of NKR-P1B expression observed on NKR-P1BbrightLy49s3+/− NK cells, which suggests that other soluble signals and/or cellular interactions may be required.

In conclusion, we have identified and characterized a novel subset of NK cells, which appears to be fully differentiated and characterized by high surface expression levels of the inhibitory NKR-P1B receptor in combination with a characteristic CD25+CX3CR1+CD93brightCD11bCD27CD62L phenotype. The NKR-P1BbrightLy49s3+/− NK cells may arise in tissues particularly enriched for activation signals, such as the gut and liver, but with the inhibitory NKR-P1B receptor playing a tolerogenic role. Future studies are required to determine how NKR-P1BbrightLy49s3+/− NK cells are related to the two other main NK cell subsets, NKR-P1Bdim or Ly49s3+ NK cells, and how NKR-P1BbrightLy49s3+/− NK cells selectively home to or arise in their target tissues.

We thank Stine Martinsen for technical assistance and Yan Zhang at the Flow Cytometry Core Facility at Oslo University Hospital and University of Oslo for cell sorting.

This work was supported by the Norwegian Cancer Society, the Norwegian Research Council, and the South-Eastern Norway Regional Health Authority.

The online version of this article contains supplemental material.

Abbreviations used in this article:
KIR

killer Ig-like receptor

poly(I:C)

polyinosinic:polycytidylic acid.

1
Lanier
L. L.
2005
.
NK cell recognition.
Annu. Rev. Immunol.
23
:
225
274
.
2
Vosshenrich
C. A.
,
Samson-Villéger
S. I.
,
Di Santo
J. P.
.
2005
.
Distinguishing features of developing natural killer cells.
Curr. Opin. Immunol.
17
:
151
158
.
3
Freud
A. G.
,
Caligiuri
M. A.
.
2006
.
Human natural killer cell development.
Immunol. Rev.
214
:
56
72
.
4
Fernandez
N. C.
,
Treiner
E.
,
Vance
R. E.
,
Jamieson
A. M.
,
Lemieux
S.
,
Raulet
D. H.
.
2005
.
A subset of natural killer cells achieves self-tolerance without expressing inhibitory receptors specific for self-MHC molecules.
Blood
105
:
4416
4423
.
5
Kim
S.
,
Poursine-Laurent
J.
,
Truscott
S. M.
,
Lybarger
L.
,
Song
Y. J.
,
Yang
L.
,
French
A. R.
,
Sunwoo
J. B.
,
Lemieux
S.
,
Hansen
T. H.
,
Yokoyama
W. M.
.
2005
.
Licensing of natural killer cells by host major histocompatibility complex class I molecules.
Nature
436
:
709
713
.
6
Freud
A. G.
,
Yokohama
A.
,
Becknell
B.
,
Lee
M. T.
,
Mao
H. C.
,
Ferketich
A. K.
,
Caligiuri
M. A.
.
2006
.
Evidence for discrete stages of human natural killer cell differentiation in vivo.
J. Exp. Med.
203
:
1033
1043
.
7
Kim
S.
,
Iizuka
K.
,
Kang
H. S.
,
Dokun
A.
,
French
A. R.
,
Greco
S.
,
Yokoyama
W. M.
.
2002
.
In vivo developmental stages in murine natural killer cell maturation.
Nat. Immunol.
3
:
523
528
.
8
Hayakawa
Y.
,
Smyth
M. J.
.
2006
.
CD27 dissects mature NK cells into two subsets with distinct responsiveness and migratory capacity.
J. Immunol.
176
:
1517
1524
.
9
Huntington
N. D.
,
Tabarias
H.
,
Fairfax
K.
,
Brady
J.
,
Hayakawa
Y.
,
Degli-Esposti
M. A.
,
Smyth
M. J.
,
Tarlinton
D. M.
,
Nutt
S. L.
.
2007
.
NK cell maturation and peripheral homeostasis is associated with KLRG1 up-regulation.
J. Immunol.
178
:
4764
4770
.
10
Chiossone
L.
,
Chaix
J.
,
Fuseri
N.
,
Roth
C.
,
Vivier
E.
,
Walzer
T.
.
2009
.
Maturation of mouse NK cells is a 4-stage developmental program.
Blood
113
:
5488
5496
.
11
Andrews
D. M.
,
Smyth
M. J.
.
2010
.
A potential role for RAG-1 in NK cell development revealed by analysis of NK cells during ontogeny.
Immunol. Cell Biol.
88
:
107
116
.
12
Inngjerdingen
M.
,
Kveberg
L.
,
Naper
C.
,
Vaage
J. T.
.
2011
.
Natural killer cell subsets in man and rodents.
Tissue Antigens
78
:
81
88
.
13
Sun
J. C.
,
Lopez-Verges
S.
,
Kim
C. C.
,
DeRisi
J. L.
,
Lanier
L. L.
.
2011
.
NK cells and immune “memory”.
J. Immunol.
186
:
1891
1897
.
14
Lanier
L. L.
,
Le
A. M.
,
Civin
C. I.
,
Loken
M. R.
,
Phillips
J. H.
.
1986
.
The relationship of CD16 (Leu-11) and Leu-19 (NKH-1) antigen expression on human peripheral blood NK cells and cytotoxic T lymphocytes.
J. Immunol.
136
:
4480
4486
.
15
Sedlmayr
P.
,
Schallhammer
L.
,
Hammer
A.
,
Wilders-Truschnig
M.
,
Wintersteiger
R.
,
Dohr
G.
.
1996
.
Differential phenotypic properties of human peripheral blood CD56dim+ and CD56bright+ natural killer cell subpopulations.
Int. Arch. Allergy Immunol.
110
:
308
313
.
16
Frey
M.
,
Packianathan
N. B.
,
Fehniger
T. A.
,
Ross
M. E.
,
Wang
W. C.
,
Stewart
C. C.
,
Caligiuri
M. A.
,
Evans
S. S.
.
1998
.
Differential expression and function of L-selectin on CD56bright and CD56dim natural killer cell subsets.
J. Immunol.
161
:
400
408
.
17
Vossen
M. T.
,
Matmati
M.
,
Hertoghs
K. M.
,
Baars
P. A.
,
Gent
M. R.
,
Leclercq
G.
,
Hamann
J.
,
Kuijpers
T. W.
,
van Lier
R. A.
.
2008
.
CD27 defines phenotypically and functionally different human NK cell subsets.
J. Immunol.
180
:
3739
3745
.
18
Silva
A.
,
Andrews
D. M.
,
Brooks
A. G.
,
Smyth
M. J.
,
Hayakawa
Y.
.
2008
.
Application of CD27 as a marker for distinguishing human NK cell subsets.
Int. Immunol.
20
:
625
630
.
19
Fauriat
C.
,
Long
E. O.
,
Ljunggren
H. G.
,
Bryceson
Y. T.
.
2010
.
Regulation of human NK-cell cytokine and chemokine production by target cell recognition.
Blood
115
:
2167
2176
.
20
Aust
J. G.
,
Gays
F.
,
Mickiewicz
K. M.
,
Buchanan
E.
,
Brooks
C. G.
.
2009
.
The expression and function of the NKRP1 receptor family in C57BL/6 mice.
J. Immunol.
183
:
106
116
.
21
Kveberg
L.
,
Bäck
C. J.
,
Dai
K. Z.
,
Inngjerdingen
M.
,
Rolstad
B.
,
Ryan
J. C.
,
Vaage
J. T.
,
Naper
C.
.
2006
.
The novel inhibitory NKR-P1C receptor and Ly49s3 identify two complementary, functionally distinct NK cell subsets in rats.
J. Immunol.
176
:
4133
4140
.
22
Kveberg
L.
,
Dai
K. Z.
,
Westgaard
I. H.
,
Daws
M. R.
,
Fossum
S.
,
Naper
C.
,
Vaage
J. T.
.
2009
.
Two major groups of rat NKR-P1 receptors can be distinguished based on chromosomal localization, phylogenetic analysis and Clr ligand binding.
Eur. J. Immunol.
39
:
541
551
.
23
Kveberg
L.
,
Jiménez-Royo
P.
,
Naper
C.
,
Rolstad
B.
,
Butcher
G. W.
,
Vaage
J. T.
,
Inngjerdingen
M.
.
2010
.
Two complementary rat NK cell subsets, Ly49s3+ and NKR-P1B+, differ in phenotypic characteristics and responsiveness to cytokines.
J. Leukoc. Biol.
88
:
87
93
.
24
Kveberg
L.
,
Dai
K. Z.
,
Inngjerdingen
M.
,
Brooks
C. G.
,
Fossum
S.
,
Vaage
J. T.
.
2011
.
Phylogenetic and functional conservation of the NKR-P1F and NKR-P1G receptors in rat and mouse.
Immunogenetics
63
:
429
436
.
25
Fossum
S.
,
Saether
P. C.
,
Vaage
J. T.
,
Daws
M. R.
,
Dissen
E.
.
2011
.
Paired opposing leukocyte receptors recognizing rapidly evolving ligands are subject to homogenization of their ligand binding domains.
Immunogenetics
63
:
809
820
.
26
Chambers
W. H.
,
Vujanovic
N. L.
,
DeLeo
A. B.
,
Olszowy
M. W.
,
Herberman
R. B.
,
Hiserodt
J. C.
.
1989
.
Monoclonal antibody to a triggering structure expressed on rat natural killer cells and adherent lymphokine-activated killer cells.
J. Exp. Med.
169
:
1373
1389
.
27
Naper
C.
,
Hayashi
S.
,
Kveberg
L.
,
Niemi
E. C.
,
Lanier
L. L.
,
Vaage
J. T.
,
Ryan
J. C.
.
2002
.
Ly-49s3 is a promiscuous activating rat NK cell receptor for nonclassical MHC class I-encoded target ligands.
J. Immunol.
169
:
22
30
.
28
Naper
C.
,
Dai
K. Z.
,
Kveberg
L.
,
Rolstad
B.
,
Niemi
E. C.
,
Vaage
J. T.
,
Ryan
J. C.
.
2005
.
Two structurally related rat Ly49 receptors with opposing functions (Ly49 stimulatory receptor 5 and Ly49 inhibitory receptor 5) recognize nonclassical MHC class Ib-encoded target ligands.
J. Immunol.
174
:
2702
2711
.
29
Naper
C.
,
Ryan
J. C.
,
Nakamura
M. C.
,
Lambracht
D.
,
Rolstad
B.
,
Vaage
J. T.
.
1998
.
Identification of an inhibitory MHC receptor on alloreactive rat natural killer cells.
J. Immunol.
160
:
219
224
.
30
Naper
C.
,
Hayashi
S.
,
Løvik
G.
,
Kveberg
L.
,
Niemi
E. C.
,
Rolstad
B.
,
Dissen
E.
,
Ryan
J. C.
,
Vaage
J. T.
.
2002
.
Characterization of a novel killer cell lectin-like receptor (KLRH1) expressed by alloreactive rat NK cells.
J. Immunol.
168
:
5147
5154
.
31
Westgaard
I. H.
,
Berg
S. F.
,
Vaage
J. T.
,
Wang
L. L.
,
Yokoyama
W. M.
,
Dissen
E.
,
Fossum
S.
.
2004
.
Rat NKp46 activates natural killer cell cytotoxicity and is associated with FcepsilonRIgamma and CD3zeta.
J. Leukoc. Biol.
76
:
1200
1206
.
32
Løvik
G.
,
Vaage
J. T.
,
Dissen
E.
,
Szpirer
C.
,
Ryan
J. C.
,
Rolstad
B.
.
2000
.
Characterization and molecular cloning of rat C1qRp, a receptor on NK cells.
Eur. J. Immunol.
30
:
3355
3362
.
33
McKnight
A. J.
,
Classon
B. J.
.
1992
.
Biochemical and immunological properties of rat recombinant interleukin-2 and interleukin-4.
Immunology
75
:
286
292
.
34
Vaage
J. T.
,
Naper
C.
,
Løvik
G.
,
Lambracht
D.
,
Rehm
A.
,
Hedrich
H. J.
,
Wonigeit
K.
,
Rolstad
B.
.
1994
.
Control of rat natural killer cell-mediated allorecognition by a major histocompatibility complex region encoding nonclassical class I antigens.
J. Exp. Med.
180
:
641
651
.
35
Kveberg
L.
,
Dai
K. Z.
,
Dissen
E.
,
Ryan
J. C.
,
Rolstad
B.
,
Vaage
J. T.
,
Naper
C.
.
2006
.
Strain-dependent expression of four structurally related rat Ly49 receptors; correlation with NK gene complex haplotype and NK alloreactivity.
Immunogenetics
58
:
905
916
.
36
Mäenpää
A.
,
Jääskeläinen
J.
,
Carpén
O.
,
Patarroyo
M.
,
Timonen
T.
.
1993
.
Expression of integrins and other adhesion molecules on NK cells; impact of IL-2 on short- and long-term cultures.
Int. J. Cancer
53
:
850
855
.
37
Lima
M.
,
Almeida
J.
,
dos Anjos Teixeira
M.
,
Queirós
M. L.
,
Justiça
B.
,
Orfão
A.
.
2002
.
The “ex vivo” patterns of CD2/CD7, CD57/CD11c, CD38/CD11b, CD45RA/CD45RO, and CD11a/HLA-DR expression identify acute/early and chronic/late NK-cell activation states.
Blood Cells Mol. Dis.
28
:
181
190
.
38
Brady
J.
,
Carotta
S.
,
Thong
R. P.
,
Chan
C. J.
,
Hayakawa
Y.
,
Smyth
M. J.
,
Nutt
S. L.
.
2010
.
The interactions of multiple cytokines control NK cell maturation.
J. Immunol.
185
:
6679
6688
.
39
Boysen
P.
,
Eide
D. M.
,
Storset
A. K.
.
2011
.
Natural killer cells in free-living Mus musculus have a primed phenotype.
Mol. Ecol.
20
:
5103
5110
.
40
Inngjerdingen
M.
,
Damaj
B.
,
Maghazachi
A. A.
.
2001
.
Expression and regulation of chemokine receptors in human natural killer cells.
Blood
97
:
367
375
.
41
Barlic
J.
,
Sechler
J. M.
,
Murphy
P. M.
.
2003
.
IL-15 and IL-2 oppositely regulate expression of the chemokine receptor CX3CR1.
Blood
102
:
3494
3503
.
42
Sheridan
B. S.
,
Lefrançois
L.
.
2011
.
Regional and mucosal memory T cells.
Nat. Immunol.
12
:
485
491
.
43
Paust
S.
,
von Andrian
U. H.
.
2011
.
Natural killer cell memory.
Nat. Immunol.
12
:
500
508
.
44
Björkström
N. K.
,
Riese
P.
,
Heuts
F.
,
Andersson
S.
,
Fauriat
C.
,
Ivarsson
M. A.
,
Björklund
A. T.
,
Flodström-Tullberg
M.
,
Michaëlsson
J.
,
Rottenberg
M. E.
, et al
.
2010
.
Expression patterns of NKG2A, KIR, and CD57 define a process of CD56dim NK-cell differentiation uncoupled from NK-cell education.
Blood
116
:
3853
3864
.
45
Lopez-Vergès
S.
,
Milush
J. M.
,
Pandey
S.
,
York
V. A.
,
Arakawa-Hoyt
J.
,
Pircher
H.
,
Norris
P. J.
,
Nixon
D. F.
,
Lanier
L. L.
.
2010
.
CD57 defines a functionally distinct population of mature NK cells in the human CD56dimCD16+ NK-cell subset.
Blood
116
:
3865
3874
.
46
Juelke
K.
,
Killig
M.
,
Luetke-Eversloh
M.
,
Parente
E.
,
Gruen
J.
,
Morandi
B.
,
Ferlazzo
G.
,
Thiel
A.
,
Schmitt-Knosalla
I.
,
Romagnani
C.
.
2010
.
CD62L expression identifies a unique subset of polyfunctional CD56dim NK cells.
Blood
116
:
1299
1307
.
47
Hamann
I.
,
Unterwalder
N.
,
Cardona
A. E.
,
Meisel
C.
,
Zipp
F.
,
Ransohoff
R. M.
,
Infante-Duarte
C.
.
2011
.
Analyses of phenotypic and functional characteristics of CX3CR1-expressing natural killer cells.
Immunology
133
:
62
73
.
48
Rolstad
B.
,
Naper
C.
,
Løvik
G.
,
Vaage
J. T.
,
Ryan
J. C.
,
Bäckman-Petersson
E.
,
Kirsch
R. D.
,
Butcher
G. W.
.
2001
.
Rat natural killer cell receptor systems and recognition of MHC class I molecules.
Immunol. Rev.
181
:
149
157
.
49
Vosshenrich
C. A.
,
García-Ojeda
M. E.
,
Samson-Villéger
S. I.
,
Pasqualetto
V.
,
Enault
L.
,
Richard-Le Goff
O.
,
Corcuff
E.
,
Guy-Grand
D.
,
Rocha
B.
,
Cumano
A.
, et al
.
2006
.
A thymic pathway of mouse natural killer cell development characterized by expression of GATA-3 and CD127.
Nat. Immunol.
7
:
1217
1224
.
50
Carlyle
J. R.
,
Jamieson
A. M.
,
Gasser
S.
,
Clingan
C. S.
,
Arase
H.
,
Raulet
D. H.
.
2004
.
Missing self-recognition of Ocil/Clr-b by inhibitory NKR-P1 natural killer cell receptors.
Proc. Natl. Acad. Sci. USA
101
:
3527
3532
.

The authors have no financial conflicts of interest.