Human CD16 as a lysis receptor mediating direct natural killer cell cytotoxicity

Abstract

In addition to their role in peptide antigen presentation, class I MHC proteins also play a critical role in inhibiting natural killer (NK) cytotoxicity through interaction with NK inhibitory receptors. Thus, NK cells are cytotoxic to virus-infected and tumor cells that have lost class I MHC protein expression. However, the nature of the receptors involved in the triggering of lysis of target cells is poorly understood. CD16 (Fcgamma receptor III) has been described as a receptor expressed on NK cells that facilitates antibody-dependent cellular cytotoxicity (ADCC) by binding to the Fc portion of various antibodies. However, we show here that CD16 has a broader function and is directly involved in the lysis of some virus-infected cells and tumor cells, independent of antibody binding. The presence of a putative CD16 ligand on appropriate target cells has also been demonstrated by the use of a CD16-Ig fusion protein.

Figure 1

Relationship between CD16 expression on NK cell lines and lysis of various target cells. (A) FACS analysis of NK lines (NK 62, NK 70, and NK m-1) stained with FITC-labeled anti-CD16 mAb 3G8 (dotted line), overlaid on a control FITC-labeled IgG1 mAb (bold line). FL1-H, fluorescence (FITC). (B) Killing of different target cells by various NK cell lines. E:T ratio, ratio of effector to target cells.

Figure 2

F(ab′)₂ of anti-CD16 mAb 3G8 blocks lysis of some target cells. NK clone 721 (CD56⁺, CD16⁺, NKIR2⁺ NKIR1⁻, NKB1⁻, CD94⁺) was incubated with 40 μg/ml of the F(ab′)₂ fragment of either the anti-CD16 mAb 3G8 or the anti-CD99 mAb 12E7 for 1 hr on ice. The NK clone 721 incubated with no mAb was used as a control. Cells were washed and incubated with the ³⁵S-labeled target cells at an E:T ratio of 5:1 in a 5-hr killing assay. The figure shows one representative experiment of three performed. Similar results were obtained with various E:T ratios and with 10 other NK clones and 10 other NK lines from different donors.

Figure 3

Blocking the lysis of 1106 mel target cells is proportional to the amount of mAb added. NK clone 10 (CD56⁺, CD16⁺, NKIR2⁺, NKIR1⁻, NKB1⁻, CD94⁺) was incubated with various concentrations of F(ab′)₂ fragment of either the anti-CD16 mAb 3G8 or the anti-CD99 mAb 12E7 for 1 hr on ice. The NK clone 10 incubated with no mAb was used as a control. Cells were washed and incubated with the ³⁵S-labeled target cells at an E:T ratio of 5:1 in a 5-hr killing assay. The figure shows one representative experiment of three performed.

Figure 4

Killing pattern of the NK92 tumor clone. The NK92 tumor line (CD56⁺, CD16⁻, NKIR2⁻, NKIR1⁻, NKB1⁻, CD94⁻) was incubated with various ³⁵S-labeled target cells for 5 hr at different E:T ratios. The figure shows one representative experiment of three performed.

Figure 5

Staining of various target cells with the CD16-Ig fusion protein. Various target cells were incubated with 50 μg/ml of either the CD16-Ig (thick line) or the CD99-Ig (dotted line) fusion proteins for 1 hr on ice and analyzed by flow cytometry. Controls were the same cells incubated with the secondary mAb alone (thin line). The figure shows one representative experiment of five performed.