| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Fox Chase Cancer Center, Division of Basic Science, Institute for Cancer Research, Philadelphia, PA 19111
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
production, or stimulate target cell cytotoxicity. We did not detect any tyrosine-dependent reduction or enhancement of NK cell activation through wild-type vs. Y238F mutant NKp44. Finally, the cytoplasmic tyrosine-based sequence did not provide a docking site for the AP-2 clathrin adaptor, nor did it potentiate receptor internalization. In summary, all activating properties and surface expression of NKp44 are mediated through its association with DAP12, and the putative ITIM in the NKp44 cytoplasmic domain does not appear to attenuate activating function. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
clones were found to express NKp44 (2), suggesting that a subset of T cells may also express this receptor. Anti-NKp44 mAb can reduce NK cell cytotoxicity toward certain tumor target cells, thereby indicating that these targets express an appropriate ligand for the receptor (1, 6). The identity of this ligand on tumors is currently unknown, although viral hemagglutinins have been reported to interact with sialic acid determinants on NKp44 and another NCR, NKp46, to promote cytolysis of influenza virus-infected target cells (7, 8). NKp44 is a type I transmembrane protein that contains a single V-type Ig-superfamily extracellular domain and is noncovalently associated in the plasma membrane with a disulfide-linked homodimer of the DNAX-activation protein of 12 kDa (DAP12; also known as killer cell-activating receptor-associated protein (KARAP)) (1, 2). DAP12 is a transmembrane accessory protein that contains an immunoreceptor tyrosine-based activation motif, which transduces intracellular activation signals. Despite its interesting characteristics, particularly its unique expression on only activated human NK cells, little is known regarding the signal transduction properties of NKp44, other than its apparent association with DAP12. Interestingly, the predicted protein sequence of the NKp44 intracytoplasmic domain (GenBank accession number AJ225109) contains a tyrosine-based sequence (EILY238HTVA), which matches the context of an immunoreceptor tyrosine-based inhibitory motif (ITIM; V/IxYxxL/V). The presence of an ITIM is uncharacteristic for an activating receptor, because ITIM sequences are critical negative signaling elements on many inhibitory receptors. Phosphorylation of an ITIM tyrosine serves as a docking site for Src homology 2 (SH2) domains of SH2 domain-containing protein tyrosine phosphatase (SHP)-1, SHP-2, or SH2 domain-containing 5'-inositol phosphatase (SHIP), which mediate inhibitory function. We hypothesized that the ITIM-like sequence of NKp44 could contribute inhibitory capacity that affects activation signaling through the receptor. We were particularly intrigued with this hypothesis upon finding the reports of two alternative NKp44 cDNAs (NKp44RG1 and NKp44RG2) in the GenBank/EMBL databases (accession numbers AJ010099, AJ010100, and AL136967.5; unpublished observations), which contain a frame shift that alters this cytoplasmic ITIM-like sequence, suggesting the existence of functionally divergent receptor isoforms. Importantly, the cytoplasmic tyrosine is situated near upstream acidic amino acids (glutamic acids at positions -3 and -5) and a hydrophobic residue at position +3 (valine), indicating that it can likely serve as an appropriate substrate for protein tyrosine kinases, particularly those of the Src family (9, 10).
In this report, we examined whether specific sequence elements on NKp44 contribute to its activating function. Our results demonstrated that NKp44 surface expression requires physical association with DAP12, which is dependent upon the integrity of a transmembrane lysine on NKp44. We also tested whether the putative ITIM sequence in NKp44 exhibits inhibitory capacity and/or influences the activation signals derived through the receptor. Our experiments indicate that, in isolation, the tyrosine-based sequence from the NKp44 cytoplasmic domain does not possess the capacity to contribute inhibitory function. Accordingly, the tyrosine-phosphorylated cytoplasmic domain did not recruit SHP-1, SHP-2 or SHIP phosphatases. In addition, phenylalanine mutation of the ITIM-like tyrosine in full-length NKp44 did not influence numerous activating functions of the receptor. We also found that the cytoplasmic tyrosine motif does not influence NKp44 internalization nor bind the AP-2 clathrin adaptor. In conclusion, our data indicate that all known activation through NKp44 can be attributed to the noncovalent association with the DAP12 homodimer.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
NK-92 cells were grown in 
-MEM supplemented with IL-2 as previously described (11). Phoenix-Amphotropic retroviral packaging cells were kindly provided by Garry Nolan (Stanford University, Stanford, CA) and grown as described (11). Jurkat cells were grown in RPMI 1640 medium supplemented with 10% FBS. DX9 (anti-KIR3DL1), 3.43.13 (anti-NKp44), and B159 (anti-CD56) mAbs were purified with protein G from hybridomas obtained from Drs. L. Lanier (University of California, San Francisco, CA), M. Colonna (Washington University, St. Louis), and B. Perussia (Thomas Jefferson University, Philadelphia, PA), respectively. PE-conjugated DX9 was purchased from BD PharMingen (San Diego, CA). The anti-FLAG mAb, M2, was purchased from Sigma-Aldrich (St. Louis, MO). Rabbit anti-DAP12 Ab was produced toward a recombinant form of the human DAP12 cytoplasmic domain and purified on protein A by Exalpha Biologicals (Boston, MA). PE-conjugated goat anti-mouse 
light chain Ab was obtained from Southern Biotechnology Associates (Birmingham, AL).
Cloning of cDNA constructs
cDNA constructs were prepared by PCR from a NK cell cDNA library using the synthetic oligonucleotides shown in Table I to generate wild-type, chimerized, or mutated NKp44 and KIR3DL1 receptor cDNAs with engineered endonuclease restriction sites, as marked in Fig. 2A. Full-length or truncated NKp44 constructs were engineered with the NKp44 leader, the FLAG-epitope tag sequence, and the NKp44 protein sequence. Transition amino acid sequences between leader, FLAG, and amino terminus of full-length NKp44 is F–6PGSQADYKDDDDKLQSKAQV6 (FLAG sequence underlined, outside residues numbered (2), and italic leucine encoded by HindIII restriction site). The cDNA for 3DL1.p44 chimeric receptors consisted of the KIR3DL1 sequence until methionine-373 in the cytoplasmic domain, a BspHI site (as described in Ref. 11), and the cytoplasmic domain of NKp44 from leucine-226 (Fig. 2A). Transition between KIR3DL1 and NKp44 is K367KNAAVMNLRSL208 (KIR3DL1 in bold, NKp44 underlined, asparagine encoded by BspHI site in italics, and representative amino acids in each protein numbered). Tyrosine-238 in the cytoplasmic domain of NKp44 was mutated to phenylalanine (Y238F constructs; numbering as in Ref. 2). The cDNA constructs were ligated into the pBMN-IRES-EGFP plasmid (from G. Nolan) or a modified version lacking the internal ribosome entry site (IRES) and enhanced green fluorescent protein (EGFP) sequences (pBMN-NoEGFP) as previously described (11). The KIR3DL1 construct was previously described (11). The cDNAs of DAP12/KARAP, DNAX-activation protein of 10 kDa (DAP10; also known known as PI-3 kinase-associated protein of 10 kDa (KAP10)), and Fc
RI accessory proteins were cloned into pBMN-IRES-EGFP using the oligos listed in Table I. DAP12/KARAP and FcRI were cloned from a cDNA library, and DAP10/KAP10 was cloned from a His-tagged cDNA construct in the pJFE14 vector that was kindly provided by John Trowsdale and Julie Fairbanks (Cambridge University, Cambridge, U.K.). All PCR were performed using Platinum Pfx DNA polymerase (Life Technologies, Rockville, MD), and the integrity of all constructs was confirmed by automated sequencing in the Fox Chase Cancer Center (FCCC) DNA Sequencing Facility.
|
|
Retrovirus was prepared as previously described (11). Briefly, the Phoenix packaging line was transfected with pBMN vectors, and culture supernatant was harvested 2 days later. NK-92 cells were incubated with viral supernatant for 8 h with Lipofectamine Plus (Life Technologies), and cells expressing EGFP and/or surface receptor were sorted 3 days later on a FACSVantageSE (BD Biosciences, Mountain View, CA) in the FCCC Cell Sorting Facility. Surface expression of transduced receptors was stable for at least 2 months. To cotransduce NKp44 and accessory proteins into Jurkat cells, the Jurkat parent was transduced with retrovirus containing cDNA for either DAP12/KARAP, DAP10/KAP10, or FcRI in pBMN-IRES-EGFP. Transduced cells were sorted for EGFP 5 days later and subsequently transduced with retrovirus containing the FLAG-NKp44 cDNA in the pBMN-NoEGFP vector. The entire cell population was analyzed 4 days later by staining for surface expression with anti-FLAG mAb.
Redirected cytotoxicity assay
Receptors transduced into NK-92 cells were tested for impacts on killing in a redirected cytotoxicity assay using the FcR+ P815 murine mastocytoma target cell line as previously described (11). NK-92 cells were either resting (4 days after IL-2 addition) or stimulated (1 day after IL-2) to test activating or inhibitory properties of the full-length or chimeric receptors, respectively. Cytotoxicity was measured in triplicate determinations using a standard 4-h 51Cr release assay with or without Abs (5 µg/ml) to engage the KIR3DL1 extracellular domain (DX9), FLAG-tagged NKp44 (M2), or the control receptor CD56 (B159).
Cell surface activation marker expression assay
Six-well plates were coated with mAbs (10 µg/ml HBSS overnight at 4°C or 2 h at 37°C) to engage receptors, washed, and seeded with 2 x 106 resting NK-92 cells (day 4 after IL-2) in 5 ml of complete medium. Cells were stimulated for 18 h before harvest by harsh pipetting, washed on ice, and labeled with PE-conjugated Abs to CD69, CD25, or CD95L (BD PharMingen) in the presence of azide. Cells were analyzed on a FACScan analyzer (BD Biosciences).
IFN- assay
Receptor-transduced NK-92 cells (2 x 106) were stimulated with Abs coated on six-well plates (5 ml/well) as described above for 18 h. Supernatants were harvested and assayed for IFN- production using an ELISA kit (BD PharMingen).
Metabolic labeling, immunoprecipitation, and immunoblotting
For metabolic labeling, NK-92 cells (160 million) were washed twice in HBSS and cultured for 75 min in 5 ml of cysteine/methionine-free DMEM (Sigma-Aldrich) containing 5 mCi [35S]cysteine/methionine (Promix; Amersham Pharmacia Biotech, Piscataway, NJ). Labeled cells were washed once with HBSS. Some cell samples were treated for 10 min with pervanadate (100 µM sodium orthovanadate plus 10 mM H2O2). Cells were lysed for 30 min on ice in 1 ml of buffer containing 1% of either Triton X-100, digitonin, Brij 58 (polyoxyethylene 20 cetyl ether) (all from Sigma-Aldrich), sodium n-dodecyl-
-D-maltoside, or n-octyl--D-glucopyranoside (octyl glucoside) (both from Calbiochem, La Jolla, CA) as previously described (11). Detergent-insoluble debris was removed from lysates by centrifugation at 20,800 x g for 15 min at 4°C. Lysates were then precleared twice for 30 min each at 4°C with protein G-coupled agarose (Upstate Biotechnology, Lake Placid, NY). FLAG-tagged NKp44 was immunoprecipitated for 90 min at 4°C with 30 µl of M2 mAb-Sepharose beads (1 mg mAb/ml cyanogen bromide-conjugated Sepharose 4 Fast Flow; Amersham Pharmacia Biotech), separated on SDS-PAGE, transferred to polyvinylidene difluoride (PVDF) membrane (Bio-Rad, Hercules, CA), and probed with rabbit anti-DAP12 Ab (0.25 µg/ml), rabbit anti-SHP-1 Ab (1 µg/ml; Upstate Biotechnology), rabbit anti-SHP-2 Ab (1 µg/ml; Santa Cruz Biotechnology, Santa Cruz, CA), or rabbit anti-SHIP (1 µg/ml; Santa Cruz Biotechnology) plus HRP-conjugated protein G (Calbiochem), anti-FLAG mAb (M2) plus HRP-conjugated donkey anti-mouse IgG (Jackson ImmunoResearch, West Grove, PA), or HRP-conjugated anti-phosphotyrosine mAb (4G10; Upstate Biotechnology). Immunoblotted proteins were visualized by chemiluminescence using the ECL detection reagents (Amersham, Arlington Heights, IL).
Yeast two-hybrid system
The yeast two-hybrid vectors pPC62 and pPC86 were obtained from P. Chevray and D. Nathans (Johns Hopkins University, Baltimore, MD) (12), and use of the yeast two-hybrid system was described previously (13, 14). Receptor cytoplasmic domain bait constructs for human KIR2DL1 (aa 265–348), human ICAM3 (CD50; aa 482–518), and human NKp44 (aa 220–276) were generated by PCR and ligated into a modified version of the pPC86 plasmid (named pPC87) in which the Gal4 DNA binding domain fusion cassette from pPC62 was introduced to replace the transcriptional activation cassette. In this way, pPC87 produced a Gal4 DNA binding fusion protein in a plasmid under TRP1 selection, which is compatible with the pACT vector encoding the Gal4 transcriptional activation domain fused with aa 3–436 of the µ2 (AP50) adaptor protein under LEU2 selection (construct kindly provided by Chander Raman, University of Alabama, Birmingham, AL (15)). A construct encoding the Gal4 DNA binding domain fused to the cytoplasmic domain of murine CD5 (aa 378–471) in pGBT9 plasmid (TRP1 selection; Clontech, Palo Alto, CA) was kindly provided by C. Raman (15). The reporter yeast strain was AH109 (BD Biosciences), which contained two integrated growth selection reporters (His3 and Ade2). Both reporters contain upstream LexA operators that are each believed to bind two LexA hybrid proteins. Yeast culture conditions, transfection protocol, and chemiluminescence assay for -galactosidase were performed as previously described (14).
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The DAP12/KARAP accessory protein was previously reported to co-immunoprecipitate with NKp44 from digitonin lysates of NK cells (1, 2). We tested this association in NK-92 cells that had been metabolically labeled with [35S]cysteine/methionine and lysed with digitonin (1, 2). The receptor complex was immunoprecipitated with a mouse anti-NKp44 mAb, 3.43.13 (kindly provided by Marco Colonna, Washington University, St. Louis). We confirmed that the mAb can immunoprecipitate NKp44 in association with DAP12, but our experiments showed only weak co-immunopreciptation of DAP12 from digitonin lysates as assessed by immunoblotting (Fig. 1). The results prompted us to further examine the capacities of other mild nonionic detergents to optimally release NKp44 and retain the DAP12 association. When directly compared, dodecyl maltoside proved to be a significantly better detergent than digitonin for this purpose (Fig. 1). In contrast, octyl glucoside only weakly solubilized the receptor and Brij 58 did not detectably release NKp44 from the plasma membrane (Fig. 1). Therefore, we confirmed that DAP12 is associated with NKp44 in NK cells, and we conclude that dodecyl maltoside is a substantially better detergent for solublizing the intact receptor complex from NK cells.
|
RI accessory proteins. As shown in Table II, co-expression of DAP12 supported surface expression of NKp44, while DAP10 or FcRI did not. Parental Jurkat cells endogenously express TCR-
(data not shown), which indicates that TCR- is also incapable of promoting surface expression of NKp44. We further mutated the transmembrane lysine residue in NKp44 (K183) to alanine to test whether it is required for DAP12 association and receptor surface expression. As shown in Fig. 3B, the K183A mutant was only marginally expressed on the surface of NK-92 cells, despite co-expression of DAP12. Furthermore, the NKp44.K183A mutant did not associate with DAP12 in dodecyl maltoside lysates, while the endogenous NKp44 (subsequently immunoprecipitated from the same lysates with the anti-NKp44 mAb, 3.43.13) was strongly associated with the accessory protein, as shown in Fig. 3C. Therefore, DAP12 association is required for cell surface expression of the receptor, and association is dependent upon integrity of lysine-183 of NKp44.
|
|
We next addressed whether the ITIM-like sequence in the NKp44 cytoplasmic domain influences activating function of the receptor complex. To test the inhibitory capacity of the cytoplasmic domain, we generated a chimeric receptor in which the cytoplasmic domain, including the ITIM-like sequence of NKp44, was directly fused to the extracellular and transmembrane domains of KIR3DL1 to create a chimeric receptor, which was designated 3DL1.p44.wt (schematized in Fig. 2A). As a negative control, a 3DL1.p44 chimera was also produced in which the ITIM-like tyrosine of NKp44 (tyrosine-238) was mutated to phenylalanine (designated 3DL1.p44.Y238F). We have previously validated this chimeric killer cell Ig-like receptor (KIR) strategy as a powerful technique to demonstrate the strong inhibitory properties of the cytoplasmic domain of KIR2DL4 (11). The 3DL1.p44 chimeric receptors were expressed at similar levels on the surface of transduced NK-92 cells, as assessed by anti-KIR mAb staining (Fig. 2B).
As in our previous studies (11, 16), the inhibitory capacities of the 3DL1.p44 chimeric receptors were tested on IL-2-stimulated NK-92 cells in "redirected" cytotoxicity assays toward the FcR+ P815 target cell, using the DX9 mAb as a surrogate ligand to specifically engage the chimeric receptors. DX9-mediated engagement of the wild-type KIR3DL1 receptor in this assay induced potent inhibition of cytotoxicity (Fig. 4A). In contrast, engagement of either 3DL1.p44 chimeric receptor did not influence the cytolytic response when compared with control cultures treated with anti-CD56 mAb or lacking added Ab (Fig. 4A). Inhibition was not evident even when NK-92 cells were tested under resting conditions (4 days after low-dose IL-2; Fig. 4A). These results indicated that the tyrosine-based motif on NKp44 is not functioning as an ITIM.
|
62 kDa (Fig. 4B). In contrast, the Y238F mutant chimeric receptor was not tyrosine-phosphorylated (Fig. 4B). Together, these results demonstrate that the ITIM-like sequence of NKp44 can be tyrosine-phosphorylated in the context of the chimeric receptor. We could not, however, detect co-immunoprecipitation of SHP-1, SHP-2, or SHIP with tyrosine-phosphorylated 3DL1.p44.wt (Fig. 4B; SHIP data not shown). On the contrary, our analyses have routinely shown that inhibitory KIR, such as 3DL1, strongly bind SHP-1 and/or SHP-2 under these conditions (11, 16). Therefore, the cytoplasmic tyrosine from NKp44 can serve as a protein tyrosine kinase substrate, but this phosphorylated tyrosine cannot serve as a docking site for SH2-containing phosphatases that are known effectors of negative signaling. The tyrosine-based sequence in NKp44 does not affect activation properties of the receptor
Next, we tested whether the tyrosine-based sequence on NKp44 influences the activating properties of the full-length receptor. To study its influence, we mutated the ITIM-like tyrosine (Y238) to phenylalanine on the FLAG-epitope tagged NKp44 (to generate NKp44.Y238F) as shown in Fig. 2A. NKp44.wt and NKp44.Y238F were expressed at similar levels in NK-92 cells by retroviral transduction (Fig. 2B) and compared for functional activation capacities.
We first compared the abilities of these FLAG-tagged forms of NKp44 to stimulate the up-regulation of several cell surface activation markers in resting NK-92 cells in response to engagement with plate-bound anti-FLAG mAb. The activation markers CD25 (IL-2 receptor -chain), CD69, and CD95L (FasL) were all increased to similar extents by engagement of either NKp44.wt or NKp44.Y238F (Fig. 5A). In contrast, engagement of the 3DL1.p44.wt chimera did not increase activation marker expression (Fig. 5A), which demonstrates that the tyrosine-based motif does not contribute direct activating capacity to receptor function. To confirm that the tyrosine-based sequence does not influence activation potential under weak stimulating conditions, we performed a dose-response comparison. As shown in Fig. 5B, similar CD69 up-regulatory capacities were also observed when both full-length NKp44 receptors were engaged with low doses of anti-FLAG mAb. Thus, the tyrosine-based sequence in NKp44 did not exhibit intrinsic activating properties, nor did it inhibit or potentiate the capacity of the receptor to trigger up-regulation of activation markers on NK-92 cells.
|
release when NK-92 cells expressed either of the receptors (Fig. 6B). In conclusion, our results indicate that the tyrosine-based sequence in the cytoplasmic domain of NKp44 does not influence activating function of the full-length receptor in assays of activation marker induction, cytotoxicity, or IFN- release.
|
Because signaling function of the ITIM-like sequence was not evident, we reasoned that the sequence could alternatively contribute to the regulation of NKp44 surface expression through physical association with the AP-2 clathrin adaptor complex, which targets cell surface receptors for endocytosis through clathrin-coated vesicles (17). The µ2 subunit of AP-2 is known to associate with Yxx
sequences ( = large hydrophobic or aromatic residue), which matches the ITIM-like sequence of NKp44 (18, 19, 20). Therefore, we used the yeast two-hybrid system to test whether the ITIM-like sequence of NKp44 acts as a tyrosine-based internalization motif through association with the µ2 subunit (also known as AP50) of the AP-2 complex. As shown in Fig. 7, we did not detect any µ2 association with the cytoplasmic domains of NKp44, KIR2DL1, or ICAM3, while interaction with the cytoplasmic domain of CD5 was readily detectable, as previously reported (15). In accordance with this result, we did not detect differential internalization rates of NKp44.wt, NKp44.Y238F, and NKp44.trunc receptors in response to engagement with anti-FLAG mAb and secondary cross-linking Abs in NK-92 cells (data not shown). The truncated receptor (NKp44.trunc) was prepared, because AP-2 is also known to interact with phenylalanine motifs (Fxx) (21). Nonetheless, even the receptor lacking the majority of its cytoplasmic domain did not exhibit altered rates of internalization, recycling, or functional response (cytotoxicity or IFN- production). Taken together, our experiments indicate that the ITIM-like motif of NKp44 does not contribute to receptor signaling or internalization functions of the NKp44/DAP12 receptor complex.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
production. In summary, the noncovalently associated DAP12 appears to contribute all activating properties to NKp44. It is indeed possible that the tyrosine-based sequence of NKp44 provides a functional role or roles that were not identified in our experiments. However, the lack of detectable recruitment of inhibitory phosphatases, despite strong tyrosine phosphorylation, and the lack of negative impacts upon several prominent functional responses triggered by NKp44, suggests that this role is subtle at best. Importantly, we also did not observe any evidence of a role for the tyrosine-based sequence in promoting activation through NKp44. Interestingly, we have not been able to convincingly demonstrate tyrosine phosphorylation of NKp44 upon treatment of NK-92 cells or primary NK cells with pervanadate. The fact that we observed tyrosine phosphorylation of the cytoplasmic domain in the context of the 3DL1.p44.wt chimeric receptor indicates that the tyrosine motif has the potential to serve as a kinase substrate as we had predicted, but the conditions that might result in such a phosphorylation event in the NKp44/DAP12 receptor complex are unclear. It is possible that the tyrosine is inaccessible in the receptor complex, due to physical obstruction by the DAP12 accessory protein.
We have found no evidence of a role for the tyrosine-based cytoplasmic sequence in promoting constitutive or ligation-induced internalization of NKp44 (data not shown), suggesting that it does not couple the receptor to clathrin adaptor complexes to mediate endocytosis through clathrin-coated pits. Furthermore, our yeast two-hybrid experiment demonstrated that the µ2 subunit of the AP-2 clathrin adaptor did not interact with the NKp44 cytoplasmic domain. The µ2 subunit is known to bind to Yxx motifs in cytoplasmic domains of transmembrane proteins, such as CD5 (15, 18, 19, 20), which served as a positive control in our experiments.
It is important to note that we did not clone any NKp44 cDNAs with alternative cytoplasmic domains. We originally embarked upon these experiments based in part upon GenBank entries that described cDNAs with cytoplasmic frame-shifted sequences that encode proteins lacking the ITIM-like sequence (NKp44RG1 and NKp44RG2). Although we cannot rule out the existence of similar frame-shifted gene products in the human population, our data provide evidence that they would not appear to disrupt or enhance NKp44 function.
It has previously been proposed that ITIM-like sequences that were noted within the cytoplasmic domains of numerous activating receptors may directly impact upon their functions (23). Our work suggests that at least some of these sequences are not functional ITIMs. In addition, despite a great deal of analysis through mutations and sequence comparisons over the years, the consensus ITIM sequence remains loosely defined, and the specific primary sequence elements that constitute docking sites for distinct phosphatases (SHP-1, SHP-2, SHIP-1, and SHIP-2) continue to remain elusive. Another NK cell-activating receptor, KIR2DL4, also contains a cytoplasmic ITIM with proven inhibitory potential in combination with a basic transmembrane residue that enables activating function (11, 24). Interestingly, that ITIM does not seem to suppress or enhance NK cell activation through KIR2DL4, despite its strong inhibitory potential (24, 27). Alternatively, SHP-2 recruitment to ITIM-like sequences on several other activating receptors has been shown to potentiate activating function by SHP-2 serving as an adaptor molecule, as opposed to attenuation of the signaling via protein tyrosine phosphatase activity (25, 26). Therefore, ITIM or ITIM-like sequences on activating receptors must be analyzed on an individual basis to directly assess their impacts on function. In the case of NKp44, our evidence in this report indicates that the tyrosine-based cytoplasmic sequence is not an ITIM and does not attenuate or enhance activating function.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Address correspondence and reprint requests to Dr. Kerry S. Campbell, Fox Chase Cancer Center, Institute for Cancer Research, 333 Cottman Avenue, Philadelphia, PA 19111-2497. E-mail address: Kerry.Campbell{at}fccc.edu
3 Current address: Institute of Neuropathology, University Hospital of Zürich, Schmelybergstrasse 12, CH-8091 Zürich, Switzerland.
4 Abbreviations used in this paper: NCR, natural cytotoxicity receptor; KIR, killer cell Ig-like receptor; wt, wild type; SH2, Src homology 2; SHP, SH2 domain-containing protein tyrosine phosphatase; SHIP, SH2-containing 5'-inositol phosphatase; ITIM, immunoreceptor tyrosine-based inhibitory motif; PVDF, polyvinylidene difluoride; IRES, internal ribosome entry site; EGFP, enhanced green fluorescent protein.
Received for publication May 29, 2003. Accepted for publication November 4, 2003.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
production. J. Immunol. 171:3415.This article has been cited by other articles:
|
|
 |
|
  S. M. Shahjahan Miah, T. L. Hughes, and K. S. Campbell KIR2DL4 Differentially Signals Downstream Functions in Human NK Cells through Distinct Structural Modules J. Immunol., March 1, 2008; 180(5): 2922 - 2932. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. W. Ho, O. Hershkovitz, M. Peiris, A. Zilka, A. Bar-Ilan, B. Nal, K. Chu, M. Kudelko, Y. W. Kam, H. Achdout, et al. H5-Type Influenza Virus Hemagglutinin Is Functionally Recognized by the Natural Killer-Activating Receptor NKp44 J. Virol., February 15, 2008; 82(4): 2028 - 2032. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. J. Wilson, R. M. Presti, I. Tassi, E. T. Overton, M. Cella, and M. Colonna FcRL6, a new ITIM-bearing receptor on cytolytic cells, is broadly expressed by lymphocytes following HIV-1 infection Blood, May 1, 2007; 109(9): 3786 - 3793. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Martinez-Barriocanal and J. Sayos Molecular and Functional Characterization of CD300b, a New Activating Immunoglobulin Receptor Able to Transduce Signals through Two Different Pathways. J. Immunol., September 1, 2006; 177(5): 2819 - 2830. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. A. Hamerman and L. L. Lanier Inhibition of Immune Responses by ITAM-Bearing Receptors Sci. Signal., January 31, 2006; 2006(320): re1 - re1. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Fuchs, M. Cella, T. Kondo, and M. Colonna Paradoxic inhibition of human natural interferon-producing cells by the activating receptor NKp44 Blood, September 15, 2005; 106(6): 2076 - 2082. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B. C. Viertlboeck, F. A. Habermann, R. Schmitt, M. A. M. Groenen, L. Du Pasquier, and T. W. Gobel The Chicken Leukocyte Receptor Complex: A Highly Diverse Multigene Family Encoding at Least Six Structurally Distinct Receptor Types J. Immunol., July 1, 2005; 175(1): 385 - 393. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Kikuchi-Maki, T. L. Catina, and K. S. Campbell Cutting Edge: KIR2DL4 Transduces Signals into Human NK Cells through Association with the Fc Receptor {gamma} Protein J. Immunol., April 1, 2005; 174(7): 3859 - 3863. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Hanna, P. Bechtel, Y. Zhai, F. Youssef, K. McLachlan, and O. Mandelboim Novel Insights on Human NK Cells' Immunological Modalities Revealed by Gene Expression Profiling J. Immunol., December 1, 2004; 173(11): 6547 - 6563. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
) during cytotoxicity of the Fc
) or with anti-CD56 mAb (B159; •). NK-92 cells were stimulated with IL-2 on the day before assay to enhance their cytolytic capacity toward P815. The lower panel is an assay of KIR3DL1.p44.wt-transduced cells stimulated 4 days after low-dose IL-2 stimulation (half of normal amount). Percent of maximal cytotoxicity is indicated. B, The cytoplasmic tyrosine of NKp44 can be tyrosine-phosphorylated, but does not recruit SHP-1 or SHP-2 phosphatases. NK-92 cells transduced with KIR3DL1.p44.wt or KIR3DL1.p44.Y238F were either untreated (-) or pervanadate-treated (+) for 10 min before lysis in 1% Triton X-100 buffer and immunoprecipitation with anti-KIR3DL1 mAb (DX9). Immunoprecipitates were separated on 10% SDS-PAGE, transferred to PVDF membranes, and probed with anti-phosphotyrosine (4G10-HRP), anti-SHP-1, anti-SHP-2, or anti-SHIP Abs plus protein G-HRP secondary. SHIP was never observed to bind phosphorylated KIR3DL1. p44.wt (data not shown).
) or anti-CD56 mAb (
). Supernatants were collected and assayed for IFN-