Peptide: MHC-based DNA vaccination strategy to activate natural killer cells by targeting killer cell immunoglobulin-like receptors

Abstract

Background: Natural killer (NK) cells are increasingly being recognized as agents for cancer immunotherapy. The killer cell immunoglobulin-like receptors (KIRs) are expressed by NK cells and are immunogenetic determinants of the outcome of cancer. In particular, KIR2DS2 is associated with protective responses to several cancers and also direct recognition of cancer targets in vitro. Due to the high homology between activating and inhibitory KIR genes to date, it has been challenging to target individual KIR for therapeutic benefit.

Methods: A novel KIR2DS2-targeting therapeutic peptide:MHC DNA vaccine was designed and used to immunize mice transgenic for KIR genes (KIR-Tg). NK cells were isolated from the livers and spleens of vaccinated mice and then analyzed for activation by flow cytometry, RNA profiling and cytotoxicity assays. In vivo assays of NK cell function using a syngeneic cancer model (B16 melanoma) and an adoptive transfer model for human hepatocellular carcinoma (Huh7) were performed.

Results: Injecting KIR-Tg mice with the vaccine construct activated NK cells in both liver and spleens of mice, with preferential activation of KIR2DS2-positive NK cells. KIR-specific activation was most marked on the CD11b+CD27+ mature subset of NK cells. RNA profiling indicated that the DNA vaccine upregulated genes associated with cellular metabolism and downregulated genes related to histone H3 methylation, which are associated with immune cell maturation and NK cell function. Vaccination led to canonical and cross-reactive peptide:MHC-specific NK cell responses. In vivo, DNA vaccination led to enhanced antitumor responses against B16F10 melanoma cells and also enhanced responses against a tumor model expressing the KIR2DS2 ligand HLA-C*0102.

Conclusion: We show the feasibility of a peptide-based KIR-targeting vaccine strategy to activate NK cells and hence generate functional antitumor responses. This approach does not require detailed knowledge of the tumor peptidomes nor HLA matching with the patient. It therefore offers a novel opportunity for targeting NK cells for cancer immunotherapy.

Keywords: immunity; immunogenicity; innate; killer cells; natural; vaccine.

Figure 1

A peptide:MHC DNA vaccine that targets KIR2DS2 activates NK cells. (A) The conformation of the constructs used in this study to inoculate the mice. (B) Gating strategy for KLRG1 on KIR2DS2+ NK cells derived from the spleens (top panels) and livers (lower panels) of KIR-Tg mice. KIR2DS2+ NK cells were identified using the antibody 1F12 and the numbers indicate the percentage positive cells in the gate. (C) The frequency of KLRG1 expression on KIR2DS2+ NK cells in the spleen and livers of KIR-Tg mice vaccinated with DNA plasmids containing HLA-C*0102 (C*0102, light gray bars), HLA-C*0102 plus IVDLMCHATAAA (C*0102-AAA, gray bars), HLA-C*0102-LNPSVAATL (C*0102-LNP, dark gray bars), HLA-C*0102-IVDLMCHATF (C*0102-IVDL, black bars) and compared with PBS alone control mice (white bars). (D, E) Comparison of KLRG1 frequencies on KIR2DS2+ (filled circles) or KIR2DS2− (open circles) CD3-NK1.1+NK cells in the spleens (D) and livers (E) following vaccination. All analyses were performed 1 week following the second vaccination. Comparisons between two groups were made by paired t-test (two groups) (D and E) and two‐way analysis of variance with Dunnett’s test for multiple comparisons to compare individual groups (1C) (*p<0.05, **p<0.01, ****p<0.001). KIR, killer cell immunoglobulin-like receptor; NK, natural killer.

Figure 2

Vaccination activates mature and terminally differentiated NK cells. KIR-Tg mice were injected with two doses of the indicated DNA construct 1 week apart and then assessed for expression of KLRG1 on CD11b, CD27 NK cell subsets from the spleens and livers 1 week after the final injection. (A–D) KLRG1 on splenic CD11b+CD27+ (A) and CD11b+CD27− (B) NK cells and on hepatic CD11b+CD27+ (C) and CD11b+CD27− (D) NK cells following vaccination. N=7–14 mice per group. (E) CD69 expression on KIR2DS2-positive CD11b CD27 NK cell subsets (n=3 per group). (F, G) Comparison of KLRG1 expression on KIR2DS2+ and KIR2DS2− splenic NK cells in the CD11b+CD27+ (F) and CD11b+CD27− (G) subpopulations. N=7–14 mice per group. (H) KIR-Tg mice crossed with C57BL/6 mice were injected with two doses of the indicated DNA construct 1 week apart and then assessed for expression of KLRG1 on total NK cells, and CD11b, CD27, NK cell 2DS2+ and 2DS2− subsets from their spleens. N=4–6 mice per group. Comparisons between two groups were made by paired t-test (two groups) (E–G) and two‐way analysis of variance with Dunnett’s test for multiple comparisons to compare individual groups (A–D and H) (*p<0.05, **p<0.01, ***p<0.005, ****p<0.001). KIR, killer cell immunoglobulin-like receptor; NK, natural killer.

Figure 3

Transcriptomic analysis of NK cells following DNA vaccination. (A) Principal component analysis (PCA) of whole NK cell transcriptomes from C*0102-IVDL and C*0102-AAA vaccinated mice taken 1 week after the second vaccination. KIR2DS2-negative NK cells from both groups are shown in the left panel and KIR2DS2-positive NK cells in the right panel. Counts were normalized and filtered using EdgeR. The first two components of the PCA are shown. (B) Heatmap of the top 100 differentially expressed genes derived from the comparison of KIR2DS2-positive NK cells from C*0102-IVDL and C*0102-AAA vaccinated mice. (C, D) EGSEA analysis of C5 gene ontology (C) and C2 canonical pathway (D) signatures comparing KIR2DS2+ and KIR2DS2− NK cell populations in both C*0102-IVDL and control-vaccinated mice. Effect significances were calculated individually for each arm of the study and the plots indicate the overall effects of vaccination on KIR2DS2+ NK cells in the C*0102-IVDL vaccinated mice (‘positive’) compared with the other three groups (‘negative’). The color denotes the direction of the change and the size of the bar represents the -Log10(FDR). All categories shown were significant at FDR<0.05. (E) Western blot analysis of histone 3 marks (H3K4me2 and H3K27me3) on purified NK cells from the spleens of vaccinated mice 1 week after the second vaccination with either C*0102-AAA (AAA) or C*0102-IVDL (IVDL). NK, natural killer.

Figure 4

DNA vaccination induces functional NK cell responses. (A) KIR-Tg mice were injected subcutaneously with B16F10 melanoma cells on day 0 and then vaccinated intramuscularly with C*0102-IVDL (squares) or C*0102-AAA (triangles) on days 0 and 7 or untreated (circles) and tumor volume measured (n=4 mice per group: one of two independent experiments). (B, C, D) Mice were injected intramuscularly with C*0102-IVDL (black bars) or C*0102-AAA (gray bars) on days 0 and day 7 and then NK cells purified from the spleens on day 14 for in vitro assays of activation. (B) Degranulation of KIR2DS2+ and KIR2DS2- KIR-Tg NK cells to B16F10 melanoma cells (n=4 mice per group). (C) Degranulation of NK cells from KIR-Tg mice to human 721.221 cells expressing HLA-C*0102 alone (221 C*0102) or HLA-C*0102 in combination with the peptide: LNPSVAATL (221 C*0102-LNP) or IVDLMCHATF (221 C*0102-IVDL) and (D) IFNγ expression following incubation with 221 C*0102 and 221 C*0102-LNP cells (n=7–8 mice per group). (E) Killing of Huh7-C*0102 hepatoma cells by NKL cells either untransfected (NKL) or transfected with KIR2DS2 (NKL-2DS2) or the inhibitory receptor KIR2DL2 (NKL-2DL2) at the indicated effector to target (E:T) ratios. Cytotoxicity was determined by flow cytometry using the LIVE/DEAD stain. Shown are the results of two independent experiments performed in triplicate. (F) NK cells from KIR-Tg mice vaccinated either with C*0102-IVDL or C*0102-AAA as a peptide control were adoptively transferred into NSG mice inoculated subcutaneously with Huh7-C*0102 hepatoma cells and tumor volume was measured (n=4 mice per group, one of two independent experiments). The control mice (n=3) did not receive any NK cells. Comparisons were by Student’s t-test (two groups) (B, D) or two‐way analysis of variance with Dunnett’s test for multiple comparisons to compare individual groups (A, C, E and F). For all plots *p<0.05, **p<0.01, ***p<0.005, ****p<0.0001. KIR, killer cell immunoglobulin-like receptor; NK, natural killer.