DAP10 associates with Ly49 receptors but contributes minimally to their expression and function in vivo

Abstract

NK cells recognize target cells through activating receptors, many of which rely on the transmembrane adaptors DAP10, DAP12 and FcR-gamma to deliver intracellular signals. Because these adaptors initiate distinct signaling pathways, they dictate the type of response mediated by receptor engagement. DAP10, for example, primarily triggers cytotoxicity, whereas DAP12 induces both cytotoxicity and IFN-gamma secretion. In mice, NKG2D signals through both DAP10 and DAP12, which broadens and modulates the type of response engendered by encounter with ligand. Although initial studies indicated that Ly49H and Ly49D recruit only DAP12, a recent report suggested that they also associate with DAP10. We asked whether this association occurs and is functionally significant under physiologic conditions. Our data demonstrate that DAP10 does associate with Ly49H and Ly49D in primary NK cells. While this association contributes slightly to cell surface expression of both receptors, it has no significant impact on Ly49H-mediated control of murine cytomegalovirus infection. Thus, while many activating NK-cell receptors are promiscuous in terms of adaptor association, our data indicate that the functional consequences of such promiscuity may vary widely and may not be evident in all cases.

Fig. 1. DAP10 contributes to Ly49H and Ly49D expression

(A) NK cells from WT, DAP10-/-, DAP12-/- and DAP10/DAP12-/- mice were purified from spleen and stained ex vivo or after culture in IL-2 (LAK) with anti-Ly49D and anti-Ly49H antibodies. Ly49D and Ly49H expression on gated NK1.1⁺ CD3^- cells is shown. Results are representative of at least 3 independent experiments. (B) 15×10⁶ IL-2 cultured NK cells were lysed and immunoprecipitated with anti-Ly49H or anti-NK1.1 as a control. Immunoprecipitates were analyzed by immunoblotting with anti-DAP10 and anti-DAP12 antibodies. The right panels show immunoblots of whole cell lysates with anti-DAP10 and anti-DAP12 antibodies to control for the amount of protein.

Fig. 2. Ly49H- and Ly49D-mediated cytotoxicity and IFN-γ production in vitro are DAP10-independent

(A-C) WT, DAP10-/-, DAP12-/- and DAP10/DAP12-/- NK cells cultured in IL-2 for 5-7 days were challenged with parental Baf/3 cells, Baf/3 cells transfected with m157 (A) or Hm1-C4 in the presence or absence of an anti-Ly49D antibody (B), or CHO cells in the presence or absence of an anti-Ly49D antibody (C). Cytotoxicity was measured by standard chromium release assay. Results are representative of at least 3 independent experiments. (D) WT and DAP12-/- NK cells were cultured in IL-2 for 3 days and tested for cytolytic activity against Baf/3 cells transfected with m157 or CHO cells in the presence or absence of an anti-Ly49D or anti-Ly49H antibody. Results are representative of at least 3 independent experiments. (E) DAP10-/-, DAP12-/-, DAP10/DAP12-/- and WT NK cells cultured in IL-2 for 5-7 days were stimulated with plastic-coated anti-Ly49D, -Ly49H, -NKG2D and - NK1.1 antibodies. After 16 hours of stimulation, cell culture supernatants were assayed by cytometric beads array for IFN-γ production. Data presented are mean values ± SD from 5 independent experiments.

Fig. 3. Control of MCMV infection is DAP10-independent

(A) Age-matched WT, DAP10-/-, DAP12-/- and DAP10/DAP12-/- mice were infected i.p. with 5 × 10⁴ PFU of MCMV. Four days post infection, mice were sacrificed and viral loads determined in the spleen (A) and liver (B) by plaque assay. Data presented are mean values ± SD from 3 independent experiments. Although DAP12-/- mice had higher viral titers in the spleen than WT mice, the difference only borders on statistical significance. In contrast, the difference between viral titers in the spleens of DAP10/DAP12-/- and WT mice is highly statistically significant.