Upregulation of the NKG2D Ligand ULBP2 by JC Polyomavirus Infection Promotes Immune Recognition by Natural Killer Cells

Abstract

Background: JC polyomavirus (JCPyV) causes progressive multifocal leukoencephalopathy (PML), a potentially fatal complication of severe immune suppression with no effective treatment. Natural killer (NK) cells play critical roles in defense against viral infections; however, NK-cell response to JCPyV infection remains unexplored.

Methods: NK- and T-cell responses against the JCPyV VP1 were compared using intracellular cytokine staining upon stimulation with peptide pools. A novel flow cytometry-based assay was developed to determine NK-cell killing efficiency of JCPyV-infected astrocyte-derived SVG-A cells. Blocking antibodies were used to evaluate the contribution of NK-cell receptors in immune recognition of JCPyV-infected cells.

Results: In about 40% of healthy donors, we detected robust CD107a upregulation and IFN-γ production by NK cells, extending beyond T-cell responses. Next, using the NK-cell-mediated killing assay, we showed that coculture of NK cells and JCPyV-infected SVG-A cells leads to a 60% reduction in infection, on average. JCPyV-infected cells had enhanced expression of ULBP2-a ligand for the activating NK-cell receptor NKG2D, and addition of NKG2D blocking antibodies decreased NK-cell degranulation.

Conclusions: NKG2D-mediated activation of NK cells plays a key role in controlling JCPyV replication and may be a promising immunotherapeutic target to boost NK-cell anti-JCPyV activity.

Keywords: JC polyomavirus; NK cell antiviral immunity; NKG2D; ULBP.

Figure 1.

NK- and T-cell responses to JCPyV VP1 in healthy donors by intracellular cytokine staining. Peripheral blood mononuclear cell from 24 healthy donors were incubated for 6 hours with 2 µg/mL of pools of peptides spanning JCPyV VP1 in the presence of conjugated anti-CD107a antibodies. BD GolgiStop and GolgiPlug were added for the last 2 hours of culture. The peptides were divided into 2 sequential pools as follows: VP1 1–93 + VP1 97–157 (n = 48) and VP1 161–253 + VP1 257–341 (n = 49). Dead cells were excluded using a viability dye. Dot plots represent proportions of CD107a⁺, IFN-γ⁺, or IL-2⁺ T or NK cells in response to JCPyV after subtracting proportions of unstimulated cells positive for each marker. Bars represent the median. * P < .05 compared to unstimulated. Abbreviations: IFN-γ, interferon-γ; IL-2, interleukin 2; JCPyV, JC polyomavirus; NK cell, natural killer cell.

Figure 2.

Development of a flow-cytometry-based assay to measure natural killer (NK)-cell cytotoxicity against JC polyomavirus (JCPyV)-infected cells. SVG-A cells were infected with M1-SVEΔ, which on average leads to 25% cells infected after 9–10 days of culture as assessed by the proportions of SVG-A cells positive for the early protein Large T (TAg) and/or the late capsid protein VP1 by intracellular flow cytometry staining. SVG-A cells were cocultured with purified NK cells from 16 healthy donors at 10:1 E:T ratio for 6 hours. A, Representative flow cytometry plots of uninfected and infected SVG-A cells stained with antibodies targeting TAg and VP1 after gating on live cells in the presence or absence of NK cells. B, Proportions of TAg⁺, VP1⁺, or Tag⁺VP1⁺ SVG-A cells after coculture with NK cells compared to those cultured without NK cells. C, Relative proportions of JCPyV-infected SVG-A cells in the presence of NK cells. D, Proportions of CD107a⁺ NK cells from 13 healthy donors cocultured with uninfected (NI) SVG-A cells or SVG-A cells infected with JCPyV Mad-1/SVEΔ (JCPyV) for 6 hours in the presence of conjugated anti-CD107a antibodies. **P < .01, **** P < .0001.

Figure 3.

The NKG2D ligand ULBP2 is upregulated in JCPyV-infected cells. 293T (A) and SVG-A (B) cells were infected with M1-SVEΔ or cultured without virus and after 9 days, cell surface expression of ligands for NKG2D was assessed by flow cytometry. *P < .05, **P < .01, *** P < .001, **** P < .0001. C, RNA was extracted from uninfected and infected SVG-A and 293T cells to quantify mRNA expression levels using primers specific for each ULBP by TaqMan qPCR. D, Volcano plot displaying fold-changes in transcripts levels for the indicated NKG2D ligands in HRPTE cells 9 days following infection with JCPyV Mad-1/SVEΔ compared to uninfected epithelial cells, using precomputed log₂ fold changes and P values available at https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc = GSE135833. Not significant defined as absolute value of log₂ fold change less than 1 and adjusted P value less than .05; significant defined as absolute value of log₂ fold change greater than 1 and adjusted P value less than .05. Abbreviations: JCPyV, JC polyomavirus; MFI, mean fluorescence intensity.

Figure 4.

NKG2D blockade impairs NK-cell response against JCPyV-infected SVG-A cells. A, Representative flow cytometry plots of CD107a⁺ NK cells following coculture with SVG-A cells infected with JCPyV Mad-1/SVE-Δ for 6 hours in the presence of isotype control (IgG1κ) or blocking antibodies against NKG2D. B, Compiled results for NK cells isolated from 7 healthy donors. *P < .05. Abbreviations: Ab, antibody; IgG, immunoglobulin G; JCPyV, JC polyomavirus; NK cell, natural killer cell.