In vivo tumor challenge with Necl-2⁺ EL-4 cells. (A-B) EL-4 cells transfected with mouse Necl-2 or EL-4 cells were labeled with 2 different concentrations of CFSE and mixed at a 1:1 ratio. Necl-2 expression was assessed by labeling with CRTAM-hFc, which binds Necl-2 but not CD96, DNAM, Nectin-1, Nectin-2, Necl-1, Necl-4, or Necl-5. (C) The cell mixture was injected intraperitoneally into C57BL/6 mice that had been previously treated with poly-I:C 24 hours earlier. Forty-eight hours after tumor challenge, cells were recovered by peritoneal lavage and examined by flow cytometry. Approximately half of the CFSE low population (EL-4–Necl-2) was rejected. One representative profile of 12 mice is shown. (D) Summary of EL-4–Necl-2/EL-4 ratios in 12 non–NK cell–depleted and 5 NK cell–depleted mice. The slight decrease in fluorescence observed in both CFSE peaks after intraperitoneal injection is due to a fluorescence quenching that occurs within the first 4 hours after labeling all cells tested. Poly-I:C injection did not induce up-regulation of CRTAM on mouse NK cells, as assessed by Necl-2–hFc binding to NK cells obtained by peritoneal lavage (data not shown). However, it significantly accelerated tumor rejection, possibly by inducing recruitment of NK cells to the peritoneal cavity (data not shown). Indeed, rejection of cells injected intraperitoneally occurred as late as 72 of 96 hours after injection in the absence of poly-I:C treatment. Horizontal bars indicate the mean ratio.