HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Regunathan, J. Articles by Malarkannan, S. Search for Related Content PubMed PubMed Citation Articles by Regunathan, J. Articles by Malarkannan, S.

Differential and Nonredundant Roles of Phospholipase C2 and Phospholipase C1 in the Terminal Maturation of NK Cells¹

Jeyarani Regunathan^*, Yuhong Chen^*, Snjezana Kutlesa^*, Xuezhi Dai^*,, Li Bai^*,, Renren Wen^*, Demin Wang^2,*, and Subramaniam Malarkannan^2,*,,¶

^* Blood Research Institute, Milwaukee, WI 53226; State Key Laboratory of Pharmaceutical Biotechnology, School of Life Science, Nanjing University, Nanjing, People’s Republic of China; Dali University, Dali, Yunnan, People’s Republic of China; Department of Microbiology and Molecular Genetics, Medical College of Wisconsin, Milwaukee, WI 53226; and ^¶ Department of Medicine, Medical College of Wisconsin, Milwaukee, WI 53226

	Abstract

Top Abstract Introduction Materials and Methods Results Discussion Disclosures References

 
NK cells play a central role in mediating innate immune responses. Activation of NK cells results in cytotoxicity, cytokine, and chemokine secretions. In this study, we show that in mice with targeted deletion of phospholipase C (PLC)2, one of the key signal transducers, there are profound effects on the development and terminal maturation of NK cells. Lack of PLC2 significantly impaired the ability of lineage-committed NK precursor cells to acquire subset-specific Ly49 receptors and thereby terminal maturation of NK cells. Overexpression of isozyme, PLC1, in PLC2-deficient NK cells resulted in the successful Ly49 acquisition and terminal maturation of the NK cells; however, it could only partially rescue NKG2D-mediated cytotoxicity with no cytokine production. Furthermore, PLC2-deficient NK cells failed to mediate antitumor cytotoxicity and inflammatory cytokine production, displaying a generalized hyporesponsiveness. Our results strongly demonstrate that PLC1 and PLC2 play nonredundant and obligatory roles in NK cell ontogeny and in its effector functions.

	Introduction

Top Abstract Introduction Materials and Methods Results Discussion Disclosures References

 
Natural killer cells constitute a major effector lymphocyte population of the innate immune response that is capable of generating a multitude of inflammatory cytokines and mediating cytotoxicity against tumor cells without prior sensitization (1, 2). A successful NK cell activation leads to multiple cellular events that govern the effector phase of the NK cells. These events include cell proliferation (3), degranulation of lytic compartments (4), cytotoxicity (5), and transcription of cytokine/chemokine-encoding genes (6).

One of the receptors that activate NK cells to mediate cytotoxicity and cytokine secretion is the C-type lectin, NKG2D, which recognizes different stress-induced nonclassical MHC class I molecules such as MICA, MICB, and UL-16-binding proteins in humans (7) and minor histocompatibility Ag 60 (H60)³ (8), Rae-1 (9), and Mult-1 (10) proteins in mice. NKG2D noncovalently associates with DAP10 or DAP12, which are phosphorylated by membrane-associated Src family protein tyrosine kinases (PTKs). Phosphorylations of tyrosines in the ITAM of the adaptor proteins generate the docking sites for the intracellular Syk family PTKs. Activated PTKs in turn phosphorylate multiple signaling molecules that initiate gene transcription and lytic granule release, leading to cytotoxicity and cytokine release (11, 12). Although these intracellular signaling events seem to be highly complex, distinct pathways associated with different NK cell effector functions are emerging. Recent studies demonstrated that the signaling events processed through NKG2D/DAP10 are distinct from that of NKG2D/DAP12 (12). Both the human and murine NK cells have been shown to mediate cytotoxicity in the absence of DAP12 or the associated PTKs, Syk and Zap70 (11, 12). However, apart from being able to mediate cytotoxicity, the NKG2D/DAP12-Syk pathway was exclusively responsible for cytokine secretion (12). Thus, NKG2D/DAP10 complex, which recruits PI3K through a YINM motif, was capable of triggering only cytotoxicity and not cytokine secretion. Although triggering through NKG2D/DAP10 and NKG2D/DAP12 can result in different effector functions, both the activation pathways recruit phospholipase C (PLC) as their major signal transducer (13). PLC play a vital role in the generation of secondary messenger molecules such as 1,2-diacylglycerol (DAG) and inositol phosphates through the hydrolysis of phosphatidylinositols (14). The DAG regulates the protein kinase (PKC) family members, and inositol 1,4,5-triphosphate (IP₃) mediates calcium (Ca²⁺) mobilization. The PLC family of enzymes has two isoforms, PLC1 and PLC2, which are 50% identical at the amino acid level; whereas the expression of PLC1 is ubiquitous, PLC2 is limited to hemopoietic tissues (15).

Despite our comprehensive understanding of receptor expression and effector functions of NK cells, the clear definition of developmental stages and the signaling events responsible are significantly lacking. Existence of common lymphoid progenitors (CLPs) that give rise to T, B, and NK cells in the bone marrow (BM) have been defined (16). Also, committed NK precursors (NKP) with specific cell surface markers have been identified in the fetal thymus and in adult BM, which exclusively mature into NK cells (17). This earliest developmental stage of NKP is defined by the expression of IL-2/IL-15 receptor common -chain (CD122) (18). Expression of CD122 in NKPs and their dependence on a common -chain and IL-15 generated from radio-resistant BM-derived stromal cells, demonstrate the vital role of IL-15-mediated activation during the development of committed NKPs (19). In the second stage, NKP cells acquire NK1.1, NKG2D, and the CD94/NKG2A/C/E in a sequential developmental process (20). The start of the expression of integrins such as CD49b, a pan NK cell marker that is identified as the integrin ₂ subunit associated with ₁ and the _v (CD51) are also part of this second stage of NK cell development (17, 20). Terminal maturation of NK cells is defined by the down-regulation of CD51, active proliferation, and a concomitant increase in the expression levels of both Mac 1 (CD11b) and CD43 receptors (21). At this stage, the immature NKP cells also start acquiring different Ly49 receptors, the expression of which defines a specific repertoire of NK cell subsets (17). The critical role of other factors such as IL-7, surface lymphotoxin (LT), and LT receptors in NK cell development is still being debated (22). Engagement of inhibitory Ly49 receptors with their respective MHC class I ligands on the target cells led to the activation of a negative signaling cascade with recruitment of protein tyrosine and lipid (SHIP) phosphatases to the ITIM in the cytoplasmic domains of Ly49 receptors (23).

Although these studies have provided tremendous insights regarding the receptors, adapter molecules, and early signaling events of NK cell activation, the signal transducers that translate the activation stimuli into effector functions have not been clearly understood. PLC2 has a determinant role in the terminal maturation, differentiation, and function of B cells as a critical signal transducer (24). This is in contrast to T cells, where PLC1 is the major signal transducer and lack of PLC2 did not affect the normal development and functions of T cells (24). Although mature NK cells express detectable levels of PLC1 (25) and abundant quantities of PLC2 (13), the biochemical, functional, and nonredundant roles of these signal transducers in NK cell development and functions have not been determined. Recently, Caraux et al. (26) and Tassi et al. (27) have described the role of PLC2 in NK cell functions using PLC2 knockout animals; however, the relative contribution of PLC1 and PLC2 in NK cell development, maturation, and functions is still not understood.

In this study, using enzyme-deficient mice, we specifically investigated the role of PLC2 in the development and function of NK cells. Our results show that although PLC2 is a predominantly expressed signal transducer in NK cells, it is not an obligate signaling molecule during the early commitment phase of NK cells. Specifically, both the initiation of CD122 expression and the commitment of CD122⁺CD3^– NKPs into NK1.1⁺NKG2D⁺ cells are unhindered in the absence of PLC2. However, in PLC2^–/– mice, peripheral and BM-derived NK1.1⁺NKG2D⁺ numbers were greatly increased, which appears to remain halted at a preterminal maturation stage. We further show PLC2 plays a critical role in the terminal maturation of NK1.1⁺NKG2D⁺ NK cells by promoting the acquisition of subset-specifying Ly49 receptors and that the presence of PLC2 in NK cells is imperative for executing effector functions such as cytotoxicity and cytokine or chemokine productions. Finally, retroviral transfection of PLC1 into PLC2-deficient BM cells promoted a partial maturation of NK cells that resulted in the successful acquisition of Ly49 receptors. However, overexpression of PLC1 into PLC2-deficient NK cells could rescue only a limited level of cytotoxicity and PLC1-rescued NK cells completely failed to generate inflammatory cytokines. These results strongly indicate that the PLC isoforms perform nonredundant and obligatory functions during the development and activation of NK cells.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion Disclosures References

 
Mice and cell lines

Mice deficient for the PLC2 and JAK3 genes have been described through our earlier studies (24, 28). All mice used in this study were maintained in pathogen-free conditions at the Biological Resource Center at the Medical College of Wisconsin (Milwaukee, WI) and were used between 5 and 12 wk of age. All animal study protocols listed here are approved by the Biological Resource Center; WI.EL4, EL4 stably transfected with H60 (EL4^H60) RMA/S, and YAC-1 were described earlier (29); and all cells were grown at 37°C, 5% CO₂, in RPMI 1640 (Invitrogen Life Technologies) with L-glutamine, sodium pyruvate, 2-ME, antibiotics (penicillin/streptomycin), and 10% FCS.

NK cell preparation and cytotoxicity assays

NK cells from indicated mice strains were prepared from splenocytes using established methods (30). Briefly, single-cell suspensions from spleen were passed through nylon wool columns (Polysciences) for the depletion of adherent populations consisting of B cell and macrophages. Cells eluted from these columns were cultured with 1000 U of IL-2 (a gift from the National Cancer Institute-Biological Resources Branch, Preclinical Repository, Bethesda, MD). The purity of the NK cultures was checked, and preparations with >95% of NK1.1⁺ were used for the experiments. NK-mediated cytotoxicity was quantified using ⁵¹Cr-labeled target cells according to established protocols (31).

Flow cytometry and cell sorting

Hybridoma secreting anti-NK1.1 (PK136) was obtained from the American Type Culture Collection. anti-Ly49A (YE1/48), anti-Ly49C/I (SW5E6), anti-Ly49D (4E4), anti-Ly49G2 (4D11), anti-CD11b (M1/70), anti-CD43 (S7), anti-CD49b (DX5), anti-CD51 (H9.2B8), anti-CD69 (H1.2F3), anti-CD122, anti-NKG2D (A10), and anti-NKG2A/C/E (20d5) were obtained from either eBiosciences or BD Pharmingen. NK cells are stained with appropriate mAbs and subjected to cell sorting using FACSAria (BD Biosciences). FITC-conjugated rat anti-mouse Ig (eBiosciences) and PE-conjugated goat anti-human Fc-specific IgG (Jackson ImmunoResearch Laboratories) were used as secondary Abs. To avoid nonspecific binding through FcRs, NK cells were pretreated with Fc-Block (2.4G2, anti-CD16/CD32) mAb (BD Pharmingen), before adding specific mAb. Standard flow cytometry analysis was conducted in LSR-II using FACSDiva software (BD Biosciences).

Secretion and quantification of cytokines through ELISA and flow cytometry

IL-2-cultured, Fc-blocked NK cells were activated with titrated concentrations of plate-bound anti-NKG2D mAb, A10 (eBiosciences), or anti-NK1.1 mAb, PK136. Total concentrations of IFN- and GM-CSF were quantified in the supernatants of NK cells using ELISA kits from eBiosciences, following the manufacturer’s instructions. Standard curves generated using recombinant murine IFN- and GM-CSF were used to calculate the concentrations. MIP-1, MIP-1ß, and RANTES were quantified using Bioplex kit from BioRad following the manufacturer’s suggested protocol. Intracellular cytokine were quantified using established methodologies. Briefly, NK cells were activated with 5 µg/ml plate-bound anti-NKG2D (A10) or anti-NK1.1 (PK136) mAbs for 16 h. Activated NK cells were Fc-blocked, stained for NK1.1, fixed, permeabilized and quantified for intracellular IFN- using FITC-conjugated anti-IFN- mAb through flow cytometry. Standard flow cytometry analysis was conducted in LSR-II using FACSDiva software (BD Biosciences).

Immunoprecipitations and Western blotting

NK cells were washed and lysed with buffer containing 10 mM Tris-HCl, 50 mM NaCl, 5 mM EDTA, 50 mM NaF, 30 mM Na₄P₂O₇, 1 mM PMSF, 5 µg/ml aprotinin, 10 µg/ml leupeptin, and 1% Triton X-100. Soluble proteins were resolved using 8% SDS-PAGE gels, transferred to nitrocellulose membranes, and probed with antisera specific for PLC1 or PLC2 (Santa Cruz Biotechnology). Western blotting for phosphorylated PLC1 or PLC2 proteins was performed as previously described (32). In brief, 10 x 10⁶ purified NK cells were stimulated with soluble anti-NKG2D (20 µg/ml) followed by cross-linking with 10 µg of the appropriate secondary IgG Ab (Jackson ImmunoResearch Laboratories) per ml. After incubation at 37°C for 5 min, the cells were washed in PBS containing 10 mM Na₃VO₄, lysed in ice-cold lysis buffer, and immunoprecipitated with either anti-PLC 1 or anti-PLC2 Ab (32). The precipitates were subjected to 8% SDS-PAGE and transferred to nitrocellulose membranes. Filters were incubated with the indicated Abs. The protein-Ab complexes were detected using the ECL detection system.

Retroviral transduction and BM transplantation

Retroviral transductions and BM transplantation were performed as described previously (33). Briefly, the rat PLC1 or rat PLC2 gene was cloned into a bicistronic retrovirus murine stem cell virus promoter-internal ribosome entry site (IRES)-GFP vector. The expression of the cloned gene and GFP is under the control of murine stem cell virus promoter, and the GFP was used as a marker for the identification of transduced cells. Conditioned media containing high-titer, amphotropic retrovirus particles were derived by cotransfection of 293T cells with the retrovirus vector expressing the cloned gene and GFP and a helper plasmid pEQPAM3 that contains the required gag, pol, and env retroviral genes driven by a Moloney leukemia virus long terminal repeat. This medium was filtered and used to transduce ecotropic packaging cells (GP + E86) with 6 µg/ml polybrene (Sigma-Aldrich) for a total of six times over 3 days. Cells exhibiting the highest levels of GFP expression were sorted under sterile conditions and subsequently expanded as virus-producing cells. Murine BM cells were transduced by retrovirus as follows: PLC2-deficient mice (8–12 wk old) were injected i.p. with 150 mg/kg 5-fluorouracil 48 h before BM harvest. BM was isolated from both hind limbs and prestimulated with 20 ng/ml murine IL-3, 50 ng/ml human IL-6, and 50 ng/ml rat stem cell factor for 48 h. Cells were then cocultured on irradiated ecotropic producer cells (GP+E86) in the presence of IL-3, IL-6, stem cell factor, and polybrene (6 µg/ml). After 48 h, 2–5 x 10⁶ nucleated BM cells were injected i.p. into sublethally irradiated JAK3-deficient recipient mice (300 rad); 4 wk later, the development of NK cells was analyzed through FACS and functional assays.

	Results

Top Abstract Introduction Materials and Methods Results Discussion Disclosures References

 
Lack of PLC2^–/– affects the development of NK cells

To determine the role of PLC2 on NK cell development, we first analyzed various tissues from wild-type (WT) littermate controls and PLC2^–/– mice for cells expressing early NK cell marker, NK1.1 through flow cytometry. Fig. 1A demonstrates the status of NK cell numbers in spleen, BM, thymus, and liver, where there was a considerable increase in the percentages of the NK1.1⁺CD3^– population (NK cells) exclusively in PLC2^–/– mice compared with their WT controls. Thymi from these mice lacked sufficient number of NK1.1⁺CD3^– cells for analysis, although there was a substantial decrease in the NK1.1⁺CD3⁺ NKT cells in the case of PLC2^–/– mice compared with the WT littermates (Fig. 1A). The increases in the percentages were not due to a systemic effect on all lymphocyte populations, given that there were no statistical differences in the number of NK1.1^–CD3⁺ T cell populations (data not shown). Furthermore, in contrast to NK or T cells, the number of NK1.1⁺CD3⁺ NKT cells population drastically decreased in spleen, thymus, and liver (Fig. 1A). Collectively, these results indicate a consistent increase in the population of NK1.1⁺CD3^– NK cells in PLC2^–/– mice. The significant increases in the NK1.1⁺CD3^– cells were also reflected in the absolute numbers in the spleen (Fig. 1B), BM, and liver (data not shown). On the basis of these observations, we conclude that the absolute numbers NK1.1⁺CD3^– cells in PLC2^–/– mice were higher (at least 2 fold) compared with that of WT controls. To further validate our observations with PLC2^–/– animals, we also used another pan NK cell marker, CD49b, and obtained similar results indicating a specific increase in the number of NK cells (data not shown). Although for reasons presently not known, the concomitant decrease in the number of CD3⁺NK1.1⁺ cells independently validates our hypotheses that the PLC2 plays selective and cell type-specific functions at different of their developmental stages.

View larger version (60K): [in this window] [in a new window]   FIGURE 1. Peripheral deficiency and developmental impairment of NK cells in PLC2–/– mice. A, Single-cell suspensions prepared from the indicated organs of WT and PLC2–/– mice were stained for NK1.1 and CD3. The numbers of NK and NKT cells are shown as percentage of total lymphocytes. Data are representative of at least four independent experiments. B, Absolute numbers of total splenocytes and CD3–NK1.1+ cells. Data are the averages of cell numbers obtained from eight WT and PLC2–/– each. C, Expression of early developmental markers CD122, NKG2D, NKG2A/C/E, CD51, CD43, CD11b, CD49b, and CD69 were analyzed with their respective mAbs exclusively in the NK1.1+ NK cell population from the spleen (as indicated in A). D, Expression of specific Ly49 receptors on the NK1.1+ NK cell population from the spleen. Percentages of positive populations are indicated in each panel of C and D using defined gates. Data are representative of five separate experiments.

Other critical developmental markers such as CD51, CD49b, CD11b (integrins), and CD43 (leukosialin) have been shown to be sequentially acquired by the developing NK cells (20). Expression of CD49b is initiated in the committed CD122⁺NK1.1⁺ NKPs (36). However, levels of CD49b are known to be decreased in the activated NK cells, including lymphokine-activated killer cells. In the case of PLC2^–/–-derived fresh NK cells, the levels of CD49b were largely unaltered (Fig. 1C). In contrast, PLC2-deficient NK cells have decreased levels of CD43 that are reported to be up-regulated during NK cell development (37). Levels of other developmental markers such as _v (CD51) are reported to be expressed at a higher level in the committed NKP but down-regulated upon NK cell maturation (20). In PLC2-deficient NK cells, the levels of CD51 and the activation stage marker CD69 were unchanged (Fig. 1C). These results demonstrate an increase in NK1.1⁺CD3^– NKPs and a partial halt in the NK cell developmental process in PLC2^–/– mice.

Terminal maturation of PLC2^–/– NK cells is impaired

Next, to assess the role of PLC2 in the differentiation of NKPs into mature NK subsets, we studied the expression patterns of subset-defining Ly49 receptors (38, 39). Recent studies have also indicated that the acquisition of different Ly49 receptors by the respective NK subsets to be the final stage of NK cell maturation process (17, 20). Thus, the maturity of the NK cells can be determined through their expression levels of Ly49 receptors (40). Results presented in Fig. 1D show a considerable decrease in the expression levels of Ly49 receptors in fresh NK1.1⁺CD3^– NK cells. Among Ly49 receptors, expression levels of Ly49A and Ly49G were most affected in PLC2^–/– relative to WT NK cells (Fig. 1D). Thus, PLC2 plays a vital role in the terminal maturation of the NK cells. Together, our results strongly demonstrate that: 1) the commitment of CLPs to become CD122⁺ and their expression of NKG2D, NKG2A/C/E and NK1.1 do not depend on PLC2; 2) however, the terminal maturation in terms of acquiring Ly49 receptors is entirely dependent on the presence and functions of PLC2.

An increase in the CD122⁺ immature NKPs is responsible for a higher NK cell number in PLC2^–/– mice

Next we analyzed the number of CD122⁺CD3^–NK1.1^–CD49b^– immature NKPs in the BM of PLC2^–/– animals. Immature NKPs are the earliest known NKPs to become NK cell lineage in the BM and are defined by the expression of CD122 (IL-2 and IL-15 receptor -chain) and by the absence of mature NK cell markers such as NK1.1, CD49b, and Ly49 receptors (19). Defects in the NK cell development in the BM of PLC2^–/– mice were evident for the following reasons. First, there was a consistent increase (2-fold) in the population of CD122⁺CD3^–NK1.1^–CD49b^– immature NKPs in the BM of PLC2^–/– relative to WT mice (Fig. 2, A–C). Interestingly, the second stage CD122⁺NK1.1⁺CD49b^– cell population, which is an indicator of immature NK phenotype, was increased considerably (Fig. 2B). Alternatively, the expression of NKG2A/C/E is least affected in the CD122⁺NK1.1⁺CD3^– population (Fig. 2D). Because the positivity for CD49b marks a third stage of NK cell maturation, we further analyzed the CD122⁺NKG2D⁺CD49b⁺ population (Fig. 2E). Consistent with our observations, there was an increase in the immature NKPs. Also there was an increase in the CD122⁺NKG2D⁺CD49b^– NK cells in the case of PLC2-deficient mice compared with that of WT. These results indicate that, although the commitment of early NKPs into the CD122⁺NK1.1⁺NKG2A/C/D/E⁺CD49b⁺ developmental stage does not depend on the functions of PLC2, this signaling molecule plays a critical role in the terminal maturation of the committed NKPs. Also, the PLC2^–/–-deficient NK cells possess the other isozyme, PLC1, in normal quantities. The role of PLC1 in the development of NKPs is yet to be determined. Our results presented here are parallel to B cell defects in the case of PLC2 and Btk-deficient mice where the functional defects were observed only after pre-B cell commitment and development (33).

View larger version (40K): [in this window] [in a new window]   FIGURE 2. Ex vivo analyses of early developmental markers in the BM of PLC2-deficient NKPs. A, Percentages of committed CD122+NK1.1–CD49b– progenitor cells in the BM of WT and PLC2–/– mice were analyzed by gating exclusively CD3–CD122+ cells. B–D, Gated CD3–CD122+ fresh BM cells were further stained for CD11b (B), CD49b (C), and NKG2A/C/E along with NK1.1 to exclude committed NKPs thereby to quantify the CD3–NK1.1–CD122+ T/NK progenitor cells. E, Gated CD3–CD122+ fresh BM cells were stained for CD49b and NKG2D to exclude committed NKPs and to further confirm the altered percentages of T/NK progenitor cells between the WT and PLC2–/– mice. Data in A–E are representative of three independent experiments. F, Rate of NK cell proliferation in PLC2–/– mice. Flow cytometric analyses of CD3–NK1.1+, IL-2-activated NK cells that are labeled with BrdU-FITC and 7-aminoactinomycin (7-AAD). Gates depicting NK cells in sub-G0, G0-G1, and S-G2-M phase were indicated. G, Rate of PLC2-deficient NK cell proliferations is comparable with that of WT. Age- and sex-matched WT and PLC2–/– mice were injected with BrdU i.p. Four days later, BM cells were prepared and analyzed for the incorporation of BrdU using a FITC-conjugated anti-BrdU Ab. Data are representative from an experiment of two animals each.

IL-2 and IL-15 cytokines fail to rescue the developmental defects in PLC2^–/– NK cells

We next examined whether cytokine-mediated activation could drive PLC2-deficient NK cells to mature in vitro. NK1.1⁺CD3^– cells were generated from the spleen and activated with IL-2. Fig. 3 demonstrates that the levels of CD11b after IL-2 activation remained higher and were not reduced in the case of PLC2^–/– compared with that of WT. The majority of the developing NK1.1⁺CD3^– cells have been shown to express the integrin _v (CD51) in the BM (34). However, the levels of CD51 have been reported to be down-regulated during NK cell maturation (20). Our results show that although the majority of the NK cells were positive, the levels of CD51 remained significantly low on per cell basis in PLC2-deficient NK cells compared with WT (Fig. 3A). Because the level of CD51 markers on the unstimulated, fresh NK1.1⁺CD3^– cells from PLC2^–/– animals were comparable with the WT subjects and the reduction is seen only in the case of IL-2-activated NK cells, we conclude that in the absence of PLC2 the NK cells fail to increase the levels of CD51. The inability of PLC2^–/–-derived NK cells to augment CD51 integrins upon IL-2 activation further confirms an early developmental defect of NK cells. Also, no differences were observed in the level of either CD43 or CD122, confirming that the commitment of NKPs into NK lineage occurs at normal phase. Significantly, IL-2 activation did not increase the levels of all the Ly49 receptors (Ly49A, C, I, D, G) tested which remained unchanged in PLC2-deficient NK cells (Fig. 3B). Staining for all these Ly49 receptors at the same time indicated tha only 23% of the NK1.1⁺ NK cells expressed one or more Ly49 receptors compared with 75% of NK1.1⁺ NK cells from the WT mice. Addition of IL-15, another key cytokine that has been shown to play an important role in the terminal maturation of NK cells (42) along with IL-2, did not alter the outcome of these results (data not shown). Therefore, activation of PLC2-deficient NK cells with cytokines could not drive PLC2-deficient NK cells to mature.

View larger version (33K): [in this window] [in a new window]   FIGURE 3. Phenotypic characterization of IL-2-activated NK cells. A, Expression of developmental and maturation markers in lymphokine-activated NK cells. NK cells were generated as described in Materials and Methods with 1000 U of IL-2/ml. On day 6, cells were stained with anti-CD3 and anti-NK1.1 mAbs, and CD3–NK1.1+ cells were gated and analyzed further for the expression of indicated developmental markers. Data are representative of three independent experiments. B, Expression of different Ly49 receptors was analyzed on CD3–NK1.1+ IL-2-activated NK cells from WT or PLC2–/– mice. Data are the averages of percentages from NK cells obtained from eight WT and PLC2–/– mice each.

Role of PLC2 in the downstream activation of NKG2D receptor

NKG2D is one of the important receptors that govern NK cell functions. Therefore, we next analyzed whether PLC2 is involved in NKG2D-mediated activation in NK cells from the WT. Purified NK cells from WT mice were stimulated with anti-NKG2D mAb, A10, followed by cross-linking with secondary anti-hamster IgG-mediated mAbs (43). Fig. 4A demonstrates that engagement of NKG2D induced a considerable level of phosphorylation of PLC2, an indication of PLC2 activation. Thus, NKG2D receptor activates PLC2 in NK cells.

View larger version (28K): [in this window] [in a new window]   FIGURE 4. Expression and specific activation of PLC2 in NK cells. IL-2-activated NK cells from WT mice were harvested on day 6 and activated with the anti-NKG2D mAb A10 or an isotype-matched control mAb as described in the Materials and Methods. PLC2 or PLC1 proteins was immunoprecipitated from the activated NK cells, resolved by SDS-PAGE, transferred to membrane, and then probed with the mAb 4G10 to determine the level of tyrosine phosphorylation or polyclonal Ab specific for PLC2 (5 x 106 NK cells/lane; A) or PLC1 (10 x 106 NK cells/lane; B) to quantify the total immunoprecipitated protein. C, Quantification of PLC1 (5 x 106 NK cells/lane) and PLC2 (5 x 106 NK cells/lane) proteins in WT and PLC2–/– mice. IL-2-activated NK cells from WT and PLC2–/– mice were harvested on day 6 and analyzed for the levels of PLC isoforms. Total cell lysates from NK cells were resolved on SDS-PAGE, transferred to membrane, and probed with polyclonal Abs specific for either PLC1 or PLC2.

Tumor recognition including NKG2D-mediated cytotoxicity is defective in PLC2^–/– NK cells

NKG2D is expressed by the majority of NK cells (44, 45, 46). Earlier studies illustrated that ectopic expression of NKG2D ligands, H60 molecules, on autologous tumor cells can render them susceptible to NK-mediated cytotoxicity through NKG2D receptor (8, 9). Although the impaired Ly49 receptor expression demonstrates incomplete maturation of PLC2-deficient NK cells, the expression levels of NKG2D receptor was unaffected (Fig. 1C). To determine whether PLC2 plays a role in NKG2D-mediated NK functions, we examined the abilities of PLC2-deficient NK cells to lyse target tumor cells that are stably transfected with H60 encoding cDNA in conventional ⁵¹Cr release assays (47). Nontransfected, parental EL4 cells were used as negative controls. NK cells derived from WT mice mediated cytotoxicity against EL4^H60 targets but not against parental EL4 cells; however, PLC2-deficient NK cells were unable to recognize and lyse the EL4^H60 targets (Fig. 5A). Furthermore, intracellular staining for perforin molecules demonstrated a comparable level of this important lytic agent present in both PLC2^–/– and WT-derived NK cells (data not shown). Thus, the NKG2D-mediated cytotoxicity of NK cells requires the PLC2-dependent activation pathway.

View larger version (37K): [in this window] [in a new window]   FIGURE 5. Functional defects of PLC2-deficient, IL-2-activated NK cells. A, NK cells from PLC2–/– mice have substantially decreased potential to mediate cytotoxicity compared with the WT counterparts. IL-2-activated NK cells were tested against 51Cr-labeled EL4, EL4H60, YAC-1, or MHC class I– RMA/S cells at the indicated E:T ratios. After 4 h at 37°C, supernatants were assayed for the levels of 51Cr release. Averages of percentages with SD from each triplicate values were calculated for NK cells from each animal. NK cells from a total of six to eight animals for each category were used. Open and closed circles represent the NK cells from WT and PLC2–/– mice, respectively. B, NK cells from PLC2–/– mice failed to generate cytokines in response to anti-NKG2D and anti-NK1.1 mAb-mediated activation. ELISA plates (Immunolon; Nunc) were coated with indicated concentrations of anti-NKG2D or anti-NK1.1 mAbs. NK cells (105/well) derived from WT or PLC2–/– mice were added to the mAb-coated ELISA plates and incubated for 18–20 h before harvesting the culture supernatants. Culture supernatants were assayed for NKG2D or NK1.1 receptor-mediated IFN- or GM-CSF productions. C, Failure to generate cytokines in PLC2–/– mice-derived NK cells is not due to an inability to secrete. ELISA plates were coated with 5 µg/ml anti-NKG2D or anti-NK1.1 mAb. NK cells (2 x 105) from WT or PLC2–/– mice were added to the mAb-coated plates and incubated for 12–14 h. Activated NK cells were stained with anti-NK1.1 mAb, fixed, permeabilized, and stained with FITC-conjugated anti-IFN- mAb, and the percentage of IFN-+ cells was enumerated through flow cytometry. Data are representative of three independent experiments. D, NK cells from PLC2–/– mice failed to generate chemokines in response to anti-NKG2D and anti-NK1.1 mAb-mediated activation. ELISA plates (Immunolon) were coated with 5 µg/ml anti-NKG2D or anti-NK1.1 mAbs. NK cells (2 x 105/well) derived from WT or PLC2–/– mice were added to the mAb-coated plates and incubated for 18–20 h before harvesting the culture supernatants. Culture supernatants were assayed for NKG2D or NK1.1 receptor-mediated MIP-1, MIP-1, or RANTES productions using a BioRad Bioplex cytokine assay kit.

NK cells also mediate lysis of tumor cells that have lost or reduced their expression levels of MHC class I in a non-NKG2D receptor-dependent manner (48). RMA/S (H-2^b) cells have been widely used as a MHC class I^– target cells to test this natural cytotoxicity of NK cells (38). In our assays, PLC2-deficient NK cells were unable to lyse these MHC class I^– target cells, thereby demonstrating their inability to recognize missing-self (Fig. 5A). Although the specific receptor-ligand interaction that is responsible the lysis of target cells during the missing-self recognition is yet to be defined, these results demonstrate that this specific activation involves PLC2-mediated signaling events. Thus, our results show a global hyporesponsiveness of PLC2-deficient NK cells to various activation stimuli.

NK cells from PLC2^–/– mice fail to generate cytokines and chemokines

In addition to cytotoxicity, NKG2D also mediates production of inflammatory cytokines, including IFN- and GM-CSF, and chemokines in NK cells (49). To determine the role of PLC2 in the NKG2D-mediated production of cytokines and chemokines, we first examined their secretion by WT and PLC2-deficient NK cells following engagement of NKG2D with immobilized anti-NKG2D mAbs. Anti-NK1.1 mAbs were used as positive cytokine production controls in WT NK cells (50). As shown in Fig. 5B, in the case of NK cells derived from the WT, significant quantities of IFN- (25–30 ng/ml) and GM-CSF (3000 pg/ml) were generated with anti-NKG2D or anti-NK1.1 mAbs in a concentration dependent manner. To the contrary, the PLC2-deficient NK cells were severely impaired in their ability to produce either IFN- or GM-CSF (Fig. 5B). As well, the isotype matched control Abs had no or negligible effect on the cytokine secretion by the NK cells (data not shown). These results demonstrate that PLC2 plays a critical role in the NKG2D as well as NK1.1-mediated cytokine production.

The reason for the inability of PLC2^–/– NK cells to secrete cytokines upon anti-NKG2D or anti-NK1.1 mAbs-mediated activation could be due to defects either in the production or in secretion of these cytokines. Previous studies indeed have shown that PLC2 deficiency affect cytokine secretion in monocytes (32). To differentiate between these two possibilities, we measured cytokine production by intracellular staining in PLC2-deficient NK cells following either anti-NKG2D or anti-NK1.1 mAbs stimulation. Consistent with cytokine secretion results, the cytokine specific intracellular staining revealed the inability of the NK cells from PLC2^–/– to generate these cytokines (Fig. 5C). Similar to that of cytokines, PLC2^–/–-derived NK cells also failed to generate chemokines such as MIP-1, MIP-1, and RANTES (Fig. 5D). These results demonstrate that PLC2 plays an important role in effecting not only the cytotoxicity but also the cytokine and chemokine production of NK cells.

Both PKC and IP₃ cascades govern NK activation and functions

Although the role of PLC2 in effecting NK cell cytotoxicity and cytokine secretions was proven, the specific downstream pathway(s) has not been defined. Receptor-mediated phosphorylation of PLC results in the activation of DAG and IP₃. To understand which of these pathways was responsible for cytokine syntheses and cytotoxic granule release, we activated the NK cells from PLC2^–/– animals with PMA and ionomycin. Activation of cells with these pharmacological agents can bypass the requirements of PLC and can directly result in the activation of PKC and induction of calcium flux. Our results presented in Fig. 6A demonstrate that the NK cells from PLC2^–/– mice were able to mediate cytotoxicity against the EL4^H60 cells when stimulated with both PMA and ionomycin. However, PMA or ionomycin alone could not restore the ability of PLC2-deficient NK cells to mediate cytotoxicity or cytokine secretions (Fig. 6A and data not shown). Irrespective of the concentration of specific stimulators, the NK cells from the WT mice mediated effective cytotoxicity against EL4^H60. This is due to the fact that they already have normal levels of PLC2 and are thereby able to mediate lysis even in the absence of these pharmacological agents (Fig. 6A). Intracellular staining of PLC2^–/– NK cells also has demonstrated a comparable level of perforin available in both PLC2^–/– and WT-derived NK cells (data not shown), indicating that the generation of lytic granules was not affected in the absence of PLC2. Similar analyses for the generation of IFN- demonstrated comparable levels of cytokine production between the PLC2^–/– and WT-derived NK cells (Fig. 6B). Thus, our data demonstrate that the functional defects that we observed in the case of PLC2-deficient NK cells were due to the lack of stimulation of both DAG/PKC and IP₃/calcineurin-mediated activation cascades.

View larger version (26K): [in this window] [in a new window]   FIGURE 6. Activation with PMA and ionomycin can bypass the NK cell defects generated by the lack of PLC2. A, Cytotoxicity of NK cells mediated through NKG2D receptor can be restored with PMA and ionomycin in PLC2–/– mice. IL-2-activated NK cells were tested against 51Cr-labeled EL4 or EL4H60 at E:T 20:1 after activation with the indicated concentrations of PMA and ionomycin. NK cells were incubated with PMA, ionomycin, or both for 10 min, washed, and added with the indicated 51Cr-loaded target cells. After 4 h at 37°C, supernatants were assayed for the levels of 51Cr release. Averages of percentages with SD from each triplicate values were calculated and presented. B, IL-2-activated NK cells from WT and PLC2–/– mice were incubated with PMA, ionomycin, or both at the indicated concentrations for 10 h, and the culture supernatants were collected and assayed for the quantity of IFN- by ELISA.

Functions of PLC2 in NK cell activation are nonredundant and cell intrinsic

The microenvironment in the BM is speculated to play a major role in the development of the NK cells (51). Committed NKPs have been preferentially identified in the BM and in thymus (17, 52). Cytokines such as GM-CSF, IL-7, IL-15, and IL-21 generated in the microenvironment play a critical role in the initiation and sustenance of NK cell development through IL-15R, IL-2/IL-15R (CD122), and c-Kit (CD117) receptors (37). Furthermore, the cell surface-bound LT on the developing NKPs that binds to the LT receptor on BM stromal cells have been specifically shown to play a role in the terminal maturation of NKPs to express Ly49 receptors (22). Therefore, we examined whether defects in PLC2-deficient NK cells are due to hemopoietic cell intrinsic deficiencies or an impaired microenvironment of developing NKPs, such as stromal cells in the BM. In addition, NK cells express both isoforms of PLC, PLC1 and PLC2. Endogenous PLC1 fails to compensate for PLC2 deficiency in NK cell maturation and functions; one explanation could be its inherent low level of expression. To determine whether PLC2 indeed plays a unique and nonredundant role in NK cells, we overexpressed PLC1 and analyzed its ability to restore the development and function of PLC2-deficient NK cells.

To address the above questions, we used a previously developed retrovirus-mediated gene transfer with BM reconstitution strategy (33) First, PLC2-deficient BM cells were infected in vitro with a retrovirus encoding PLC1, an IRES, and GFP. Next, the retrovirally transduced BM cells were transplanted into sublethally irradiated JAK3-deficient mice, which inherently lack T, B, and NK cells (28, 53). These retrovirally transduced BM cells repopulated the different lymphocyte populations, including NK cells, in the recipient JAK3-deficient mice. Virus-transduced cells were identified by virtue of the GFP gene. A retrovirus encoding PLC2 and GFP (PLC2-IRES-GFP) was used as a positive control. A retrovirus encoding GFP alone was used as a negative control as well as to examine whether the defects of PLC2^–/– NK cells are hemopoietically intrinsic.

First, the GFP⁺ NK cells from the JAK3^–/– recipients transplanted with PLC2^–/– BM that were transduced with PLC1-IRES-GFP virus exhibited higher levels of PLC1 protein expression than the WT NK cells (Fig. 7A). Furthermore, the GFP⁺ NK cells from the recipient transplanted with PLC2^–/– BM that were transduced with PLC2-IRES-GFP virus exhibited normal levels of PLC2 protein expression (Fig. 7A). Next, the retroviral vector-alone-transduced BM cells give rise to a number of GFP⁺ and GFP^– cells similar to that of the PLC1 or PLC2 rescued cells (Fig. 7, B–D, leftmost panels, and data not shown) demonstrating comparable levels of viral transductions.

View larger version (39K): [in this window] [in a new window]   FIGURE 7. PLC1 and PLC2 differentially restores the expression of developmental and maturation markers of NK cells. A, Levels of PLC1 and PLC2 proteins were quantified in retrovirally tranduced NK cells. IL-2-activated NK cells generated the spleen of PLC2–/– mice that were transplanted with retrovirally transduced with either IRES-GFP-PLC1 or with IRES-GFP-PLC2 were harvested on day 6 and analyzed for the levels of PLC isoforms. Total cell lysates (4 µg) from NK cells derived from the recipient mice were resolved on SDS-PAGE, transferred to membrane, and probed with polyclonal Abs specific for PLC1, PLC2, or anti-actin-specific mAbs. As a positive control, we transduced the IRES-GFP retroviruses into the BM cells of WT and transplanted them into PLC2–/– mice. PLC1 and PLC2 from NK cells in these combinations were indicative of the naturally expressed levels and were used to compare with other rescued NK cells. Phenotypic characterization of NK cells that underwent enforced expression of vector alone (B), PLC1 (C), or PLC2 (D). PLC2-deficient mice were reconstituted with BM-derived from PLC2–/– mice that are transduced either IRES-GFP retroviruses (B: Vector); with IRES-GFP-PLC1 (C: PLC1); or with IRES-GFP-PLC2 (D: PLC2). After 2 months of reconstitution, animals were sacrificed, and BM cells of the recipient animals were used to generate IL-2-activated NK cells. On day 6, cells were sorted based on the GFP expression and further cultured for 24 h. GFP+ and GFP– cells were gated for NK1.1 expression and further analyzed for the expression levels of different developmental markers such as integrins CD11b, CD49b, and CD51; CD43; inhibitory receptors Ly49A and Ly49C/I.

Our data show that, whereas the levels of CD43 were unchanged irrespective of the viral constructs transduced, the levels of CD51 were highest exclusively with PLC2 reconstitution (Fig. 7C). Also, the levels of CD49b were down-regulated only in the case of NK cells that were transduced with PLC2. Neither the GFP^–NK1.1⁺ NK cells from the same animals, nor NK cells transduced with PLC1, nor the vector alone-transduced NK cells failed to show such a CD49b down-modulation upon IL-2 activation (Fig. 7, B and C). These data noticeably demonstrate that the requirement for PLC2 and a substitution of PLC2 with PLC1 cannot compensate for the successful phenotypic maturation of NK cells. Interestingly, analyses of Ly49 receptors demonstrate that both PLC2 and PLC1 could rescue the expression of these receptors. NK cells from PLC2^–/– animals do express a small but detectable level of PLC1. Although, this level PLC1 was evidently insufficient for the complete maturation process of NK cells, an overexpression of PLC1 helped to rescue some of these phenotypic defects. Roles of PLC1 in the early commitment phase of NKPs are exciting novel possibilities; however, the exact functions have yet to be defined. Therefore, we propose that the expression of receptors such as NKG2D, NK1.1, NKG2A/C/E, and CD122 may be dependent on the expression and function of PLC1 but not PLC2. Nevertheless, the later NK cell developmental processes are virtually dependent on the presence of PLC2.

As expected, virally transduced, GFP⁺, PLC2-overexpressing NK cells regained their ability to mediate the cytotoxicity of El4^H60 target cells through NKG2D receptor. Fig. 8A demonstrates that the percentage of cytotoxicity mediated by the PLC2-reconstituted NK cells increased significantly up to 60% (at an E:T ratio of 40:1), indicating that restoration of PLC2 expression could correct the functional defects. Moreover, surprisingly, NK cells that were reconstituted with the PLC1 could also partially regain their ability to mediate cytotoxicity (40% at an E:T ratio of 40:1). These results demonstrate that although the comparatively low but normal levels of PLC1 in PLC2-deficient NK cells were insufficient to act as signal transducer, the overexpression of the PLC1 could reconstitute some functions of PLC2. The two isoforms of PLC conforms around 50% amino acid homology; thus, it is biochemically possible that some functions of PLC2 can be performed by the overexpression of PLC1. However, the specific protein domain(s) and their secondary messengers that allow the PLC1 to perform PLC2-mediated function have yet to be defined.

View larger version (23K): [in this window] [in a new window]   FIGURE 8. Cytotoxicity and cytokine secretions of NK cells can be successfully mediated only in the presence of PLC2. A, Ability of PLC-reconstituted NK cells to mediate cytotoxicity through NKG2D receptor. PLC2–/– mice were transplanted with BM cells (from PLC2–/–) that were transduced with vector-IRES-GFP; IRES-GFP-PLC1 or with IRES-GFP-PLC2. Two months later, IL-2-activated NK cells were prepared from the splenocytes of the reconstituted animals. Transduced and nontransduced NK cells were sorted by virtue of presence or absence of GFP expression. Sorted NK cells were tested against 51Cr-labeled EL4 or EL4H60 target cells at the indicated E:T ratios. After 4 h at 37°C, supernatants were assayed for the levels of 51Cr release. Averages of percentages with SD from each triplicate values were calculated for NK cells that are generated and pooled from four animals in each category. B, Ability of PLC-reconstituted NK cells to generate cytokines in response to anti-NKG2D mAb-mediated activations. ELISA plates (Immunolon) were coated with 5 µg/ml anti-NKG2D mAb A10. NK cells (105/well) derived from reconstituted mice were added to the mAb-coated ELISA plates and incubated for 18–20 h before harvesting the culture supernatants. Culture supernatants were assayed for NKG2D-mediated IFN- production. C, PMA and ionomycin were used with the reconstituted NK cells to enumerate their ability in generating IFN-. NK cells were incubated with PMA (100 µg/ml) and ionomycin (5 ng/ml) for 10 h, and the culture supernatants were collected and assayed for the quantity of IFN- by ELISA.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion Disclosures References

 
Despite substantial progress in understanding the functions of NK cells, less is known about the development of these cells. Recent studies start to define developmental stages of NK cells with specific cell surface markers and functional characteristics (17, 19, 20, 21, 54, 55). Although recent studies have implicated PLC2 in NK cell functions (11, 26, 27), our current study for the first time compares the nonredundant roles of PLC1 and PLC2 in NK cell development and functions.

One of the key observations is that there is significant increase in the numbers of CD3^–NK1.1⁺ NK cells in PLC2^–/– mice. Our observations are different from that of Caraux et al. (26), where a transfer of PLC2-deficient BM cells into Rag2^–/–c^–/– did not result in an overall increase in the CD3^–NK1.1⁺ cells. One possibility for this discrepancy could be the host animal backgrounds. Similar cell transfer experiments involving PLC2-BM cells into JAK3^–/– mice also indicated an increase in the CD3^–NK1.1⁺ cells compared with that of WT-derived cell transfers (S. Malarkannan, unpublished observations). Therefore, we predict that although the increase in the CD3^–NK1.1⁺ cells in PLC2^–/– mice could be cell intrinsic, there may be other unknown environmental factors that could regulate the total CD3^–NK1.1⁺ cell numbers.

On the basis of these results, we conclude that although the PLC2 is predominantly expressed, it is not an obligate signaling molecule during the early developmental phase of NK cells. In particular, both the initiation of CD122 expression and the commitment of CD122⁺CD3^– NKPs into NK1.1⁺NKG2D⁺ cells are unhindered in the absence of PLC2. Nonetheless, NK1.1⁺NKG2D⁺ NK cells in PLC2^–/– mice appear to remain halted at a preterminal maturation stage, which fail to down-modulate CD11b and to augment CD51 integrin expression upon IL-2 activation. Importantly, a lack of PLC2 demonstrates that it plays a critical role in the terminal maturation of NK cells by promoting the acquisition of Ly49 receptors. Consistent with this notion, BM-derived NK cells from PLC2^–/– mice have increased proliferation rates compared with those from WT mice. Previous studies from Yokoyama’s laboratories have demonstrated that the heightened proliferation of NKPs at the preterminal stage (CD11b^lowCD49b^highCD43^lowCD51^high) is an inherent phenomenon of these immature cells (20). NK cells at this specific developmental stage start to acquire different Ly49 receptors and CD94/NKG2 complexes, and their ability to proliferate was also reported to be decreased upon further maturation (20). Therefore, our study indicates that the development of PLC2^–/– NK cells halts at a postexpansion and prematuration stage. The exact mechanism by which lack of PLC2 affects acquisition of the maturation markers and impairs maturation of NK cells is not clear. However, it is possible that PLC2 deficiency could result in impaired activation of a number of transcription factors. Similar to the present study, an increase in the CD122⁺NK1.1⁺TCR^– cells have been observed exclusively in the BM of transcription factor T-bet-deficient mice (40). However, in contrast to T-bet-deficient mice, PLC2^–/– mice contained increased CD122⁺NK1.1⁺CD3^– cells in the spleen, BM, and liver. The relationship between the activation PLC2 and T-bet is not known and is worth investigating. In addition, lack of transcription factor GATA-3 results in the reduced expression of Ly49A, C/I, and D receptors (40). However, the expression of Ly49G2 and CD94NKG2A were relatively increased in GATA-3-deficient mice (56). GATA-3 is predicted to control the timing or accessibility of distinct Ly49 receptor genes and thereby their gene transcription (56). Thus, it is quite possible that the GATA-3 could be one possible target transcription factor downstream of the PLC2 signaling cascade. Moreover, lack of SHIP phosphatase also results in the altered levels of several Ly49 and CD94 molecules (57). In the adult SHIP^–/– mice, the expressions of Ly49A, Ly49C were overrepresented, whereas Ly49D, Ly49G2, Ly49I, and CD94 were underrepresented (57). Lastly, PLC2 deficiency might also affect activation of other specific transcription factors, including IFN-regulatory factor-1 (58), IFN-regulatory factor-2 (59), ID2 (60), Ets-1 (61), MEF (62), and PU-1 (63) that are involved in NK cell development. Investigation of a role of PLC2 in the activation of these transcription factors may shed some light on the mechanism by which it mediates maturation of NK cells.

The other important finding is that PLC2 deficiency disrupts all effector functions of NK cells that we have examined. Our results also demonstrate that engagement of NKG2D significantly induces activation of PLC2. In agreement with this finding, PLC2 deficiency completely disrupts NKG2D-mediated cytotoxicity and cytokine production. Furthermore, activation through NK1.1 receptors is dependent on PLC2 since the PLC2-deficient NK cells fail to generate cytokines/chemokines in response to anti-NK1.1 mAb. Taken together, these findings reveal PLC2 as a master signal transducer for most, if not all, of the activation receptors in NK cells. The fact that neither PMA nor ionomycin alone fails to rescue functional defects of PLC2-deficient NK cells, whereas PMA plus ionomycin successfully restore effector functions of mutant NK cells, demonstrate that both DAG/PKC and IP₃/Ca²⁺ pathways are required for the cytotoxicity and cytokine production.

One possible explanation for the impaired cytotoxicity of PLC2-deficient NK cells could be due to insufficient expression of Ly49 receptors. Low or nonexpression of Ly49 receptors hav