A new cell line for high throughput HIV-specific antibody-dependent cellular cytotoxicity (ADCC) and cell-to-cell virus transmission studies

Abstract

Several lines of evidence indicate that antibody-dependent cellular cytotoxicity (Wren et al., 2013) is important in the pathogenesis of HIV-1 infection. Namely, ADCC is induced during natural HIV-1 infection or in HIV-1 vaccine studies, the latter demonstrated by the RV144 vaccine trial. To expedite the assessment of ADCC in studies of HIV, we have developed a high throughput assay. We have optimized the rapid fluorometric antibody-mediated cytotoxicity assay (RFADCC) by transfecting the EGFP-CEM-NKr cell line to constitutively express SNAP-tagged CCR5. This cell line can then serve as a source of HIV-specific targets when coated with monomeric gp120, spinoculated with inactivated intact virions, infected by cell-free viral diffusion or infected by cell-to-cell transmission of virus. The optimized strategy has two significant advantages over the original RFADCC method: First, the preparation of detectable target cells is less labor intensive and faster as it does not rely on multiple staining and washing steps for target cells. Second, because the target cell markers GFP and SNAP are constitutively expressed, the assay provides highly reproducible data. These strengths make the optimized RFADCC assay suitable not only for studies of HIV-1 specific cytotoxicity but also for studies of cell-cell transmission of virus. In conclusion, this assay provides a new generation T cell line that can expedite large clinical studies as well as research studies in humans or non-human primates.

Keywords: EGFP-CEM-NKr-CCR5-SNAP; Flow-cytometry; HIV-1 virus; Monoclonal antibody; RFADCC.

Fig. 1

Generation and characterization of EGFP-CEM-NKr-CCR5-SNAP cells. Left panel. EGFP-CEM-NKr-CCR5-SNAP cells were generated by stable transfection of EGFP-CEM-NKr cells with Tag-lite pSNAP-CCR5 vector (pSNAP-CCR5, htfr). A–C. GFP and CCR5 expression in parental CEM-NKr-CCR5, EGFP-CEM-NKr and transfected EGFP-CEM-NKr-CCR5-SNAP cells. D–F. Labeling with surface SNAP tag-Alexa Fluor-647 dye vs. GFP expression in the three cell lines. Right panel. The levels of CCR5 and CD4 molecules expressed on EGFP-CEM-NKr-CCR5-SNAP cells in comparison to CEM NKr CCR5 were assessed by flow cytometry with Quantibrite PE calibration beads.

Fig. 2

Schematic representation of the new RFADCC assay outline conducted with EGFP-CEM-NKr-CCR5-SNAP cells. The optimized assay was modified from Gomez-Roman et al. (2006). A. For gp120-based ADCC assay, EGFP-CEM-NKr-CCR5-SNAP cells were stained with SNAP-Surface Alexa Fluor 647 for 20 min at 37 °C with or without monomeric HIV-1 gp120. B. For AT-2 inactivated virus-based ADCC assay EGFP-CEM-NKr-CCR5-SNAP cells were stained with Alexa Fluor 647-SNAP tag dye first and then spinoculated with the inactivated virus. C. For IMC-infected targets-based ADCC assay, EGFP-CEM-NKr-CCR5-SNAP were spinoculated with IMC, cultured for 5 days, washed twice and then stained with Alexa Fluor 647-SNAP tag dye. Sensitized cells were incubated with dilutions of antibodies for 15 min at room temperature (RT) and subsequently with PBMC as effector cells for 2 or 3 h at 37 °C. Cells were then washed and fixed in 1% paraformaldehyde. The readout is the loss of GFP, as a direct measure of the percentage of target cells cytotoxicity mediated by the mAbs. After coating with monomeric HIV-1 Bal gp120 and adding PBMC as effector cells, we measured the ADCC activity of a reference HIV-1 mAb, C11. The cytotoxicity readout in the original RFADCC was measured as loss of CSFE, while in the modified RFADCC it was the loss of GFP.

Fig. 3

ADCC assay and mAbs surface staining conducted with cells coated with HIV-1 Bal gp120. A. Comparison of RFADCC assay layout with CEM NKr CCR5 vs. EGFP-CEM-NKr-CCR5-SNAP cells. CEM NKr CCR5 were stained with PKH-26 and CSFE (Gomez-Roman et al., 2006), while EGFP-CEM-NKr-CCR5-SNAP cells were stained with Alexa Fluor 647-SNAP tag dye. Killing by C11 mAb (1 μg/ml) is determined as loss of CSFE in PKH26-positive CEM NKr CCR5 cells or loss of GFP in SNAP-positive EGFP-CEM-NKr-CCR5-SNAP cells. B. Cytotoxicity curves for gp120-coated EGFP-CEM-NKr-CCR5-SNAP. The ADCC results are representative of three independent assays and the bars indicate the range of the values of cytotoxicity of duplicate samples. The binding of 1 μg/ml Alexa-Fluor 647-conjugated mAbs C11 (C and F), N5-i5 (D and G) or N12-i2 (E and H) was compared in gp120-coated-CEM NKr CCR5 vs. EGFP-CEM-NKr-CCR5-SNAP cells.

Fig. 4

ADCC assay using cells spinoculated with AT-2 inactivated HIV-1 BaL virions. Left panel. EGFP-CEM-NKr-CCR5-SNAP cells were labeled with Alexa Fluor 647-SNAP tag and then spinoculated with HIV-1 Bal AT-2 virus at 2000 rpm for 2 h at 12 °C. After two washes, cells were incubated with dilutions of mAbs (C11, N5-i5, N12-i2 or Synagis) for 15 min at RT, then PBMC were added to the reaction for 3 h at 37 °C. At the end of the incubation, the samples were washed with PBS, fixed with 1% paraformaldehyde and analyzed by flow cytometry. The ADCC results are representative of three independent assays and the bars indicate the range of the values of cytotoxicity of duplicate samples. Right panel. The efficiency of the spinoculation was evaluated by cell surface staining with 2 μg/ml Alexa Fluor-594-conjugated mAbs C11 (Panel A), N5-i5 (Panel B) or N12-i2 (panel C).

Fig. 5

ADCC assay using cells infected with HIV-1 BaL IMC. Left panel. EGFP-CEM-NKr-CCR5-SNAP cells were spinoculated with HIV-1 Bal molecular clone at 2000 rpm for 2 h at 12 °C. After 5 days of co-culture with the virus, cells were washed twice, labeled with Alexa Fluor 647-SNAP tag dye and incubated with dilutions of mAbs (C11, N5-i5, N10-U1 or Synagis) for 15 min at RT, then PBMC were added to the reaction for 3 h at 37 °C. At the end of the incubation, the samples were washed with PBS, fixed with 1% paraformaldehyde and analyzed by flow cytometry. The ADCC data represent the typical results obtained in three independent experiments. Upper right panel. The efficiency of the infection was evaluated by staining of the cells with live/dead (not shown), CD4-APC and p24-PE. Lower right panel. Binding of infected EGFP-CEM-NKr-CCR5-SNAP cells with 5 μg/ml Alexa Fluor-647-conjugated mAbs C11 (panel A), N5-i5 (panel B) or N10U1 (panel C).

Fig. 6

EGFP-CEM-NKr-CCR5-SNAP cells: New tool to study the cell–cell virus spread. A. Schematic representation of the experimental design (modified from Zhong et al., 2013). CEM NKr CCR5 cells, infected with HIV-1 Bal molecular clone, were used at the peak of infection as virus donors. EGFP-CEM-NKr-CCR5-SNAP cells were used as virus acceptor targets of the infection transmitted cell to cell. Donors and targets were plated together to allow the spread of the virus, and the readout of the newly infected targets was the count of GFP⁺/P24⁺ cells. Background of cell-free virus spread was determined with donors and targets separated in Transwell chambers and with targets subjected to supernatants harvested from infected donor cells. B. and C. The count of GFP⁺/p24⁺/CD4⁺ or ⁻ cells was monitored for 7 days from the beginning of the co-culture by flow cytometry.