Natural killer cell antibody-dependent cellular cytotoxicity to Plasmodium falciparum is impacted by cellular phenotypes, erythrocyte polymorphisms, parasite diversity and intensity of transmission

Abstract

Objectives: Natural killer (NK) cells make important contributions to anti-malarial immunity through antibody-dependent cellular cytotoxicity (ADCC), but the role of different components of this pathway in promoting NK cell activation remains unclear.

Methods: We compared the functions and phenotypes of NK cells from malaria-exposed and malaria-naive donors, and then varied the erythrocyte genetic background, Plasmodium falciparum strain and opsonising plasma used in ADCC to observe their impacts on NK cell degranulation as measured by CD107a mobilisation.

Results: Natural killer cells from malaria-exposed adult Ugandan donors had enhanced ADCC, but an impaired pro-inflammatory response to cytokine stimulation, compared to NK cells obtained from malaria-naive adult North American donors. Cellular phenotypes from malaria-exposed donors reflected this specialisation for ADCC, with a compartment-wide downregulation of the Fc receptor γ-chain and enrichment of highly differentiated CD56^dim and CD56^neg populations. NK cell degranulation was enhanced in response to opsonised P. falciparum schizonts cultured in sickle cell heterozygous erythrocytes relative to wild-type erythrocytes, and when using opsonising plasma collected from donors living in a high transmission area compared to a lower transmission area despite similar levels of 3D7 schizont-specific IgG levels. However, degranulation was lowered in response to opsonised field isolate P. falciparum schizonts isolated from clinical malaria infections, compared to the 3D7 laboratory strain typically used in these assays.

Conclusion: This work highlights important host and parasite factors that contribute to ADCC efficacy that should be considered in the design of ADCC assays.

Keywords: antibody‐dependent cellular cytotoxicity; malaria; natural killer cells.

Figure 1

NK cells from malaria‐exposed Ugandan adults had ADCC activity superior to that of malaria‐naive North Americans. NK cell degranulation was quantified using the marker CD107a (LAMP‐1). PBMCs from North American (n = 10) and Ugandan (n = 10) donors were stimulated with iRBCs and uRBCs either alone or opsonised with serum at a ratio of 2 iRBCs to 1 PBMC (a). Comparison of North American and Ugandan NK cell degranulation in response to iRBCs opsonised in pooled Ugandan serum in all NK cells (b, c) and in individual NK cell subpopulations (d). P‐values were calculated using Wilcoxon rank sum tests (paired in b only). ADCC, antibody‐dependent cellular cytotoxicity; iRBCs, infected erythrocytes; NK, natural killer; PBMCs, peripheral blood mononuclear cells.

Figure 2

NK cells from malaria‐exposed donors had a poor inflammatory response to cytokine stimulation. PBMCs from North American (n = 10) and Ugandan (n = 8) donors were stimulated with a cocktail of IL‐12 (2.5 ng mL⁻¹), IL‐15 (10 ng mL⁻¹) and IL‐18 (0.25 μg mL⁻¹) or medium. NK cell activation was quantified by staining for intracellular IFNγ. Comparison of North American and Ugandan NK cell IFNγ production following stimulation in all NK cells (a, b) and within NK cell subpopulations (c). P‐values were calculated using Wilcoxon rank sum tests (paired in a only). NK, natural killer; PBMCs, peripheral blood mononuclear cells.

Figure 3

Endemic malaria exposure was associated with phenotypic maturity and extensive differentiation in NK cells. (a) Heat map showing the phenotypic characteristics of defined NK cell clusters identified from North American (n = 10) and Ugandan PBMC donors (n = 12). Marker expression was scaled between 0 and 1 by using lower and upper quantiles as boundaries prior to aggregating the data. Row and column clustering were performed using unscaled data. (b) Bar graphs representing the individual NK cell clusters that were found to be significantly enriched in Ugandan PBMC donors through differential abundance testing within CATALYST. P‐values were calculated using unpaired Wilcoxon rank sum tests. (c) UMAP visualisation of NK cells from North American (left) and Ugandan (right) donors, coloured by their corresponding cluster. (d) UMAP visualisations of NK cells from North American (top row) and Ugandan Expression data were scaled between 0 and 1 using lower and upper expression quantiles. NK, natural killer; PBMCs, peripheral blood mononuclear cells.

Figure 4

The intensity of malaria transmission at the site of opsonising plasma collection was positively correlated with NK cell degranulation during ADCC. Opsonising plasma was obtained from children living in three areas of Uganda with different local intensities of malaria transmission. (a) NK cells purified from Ugandan PBMC donors (n = 3, three independent experiments) were stimulated with iRBCs opsonised in pooled plasma from Jinja (aEIR = 3.8), Kanungu (aEIR = 32) and Tororo (aEIR = 301). P‐values were calculated using paired Wilcoxon rank sum tests. (b) Correlation of NK cell degranulation in response to iRBCs incubated in plasma from a subset of the individual donors used to create pools in a with their corresponding household‐level human biting rate values. For this scatterplot, Rho (ρ) and P‐values were calculated using Spearman's correlation. ADCC, antibody‐dependent cellular cytotoxicity; iRBCs, infected erythrocytes; NK, natural killer; PBMCs, peripheral blood mononuclear cells.

Figure 5

NK cell degranulation in response to opsonised iRBCs was enhanced by sickle cell trait. iRBCs were obtained from Pf cultures supplemented with blood with an HbAA or HbAS background. NK cells purified from Ugandan PBMC donors (n = 2, two independent experiments) were stimulated with iRBCs alone or opsonised in serum. P‐values were calculated using paired Wilcoxon rank sum tests. iRBCs, infected erythrocytes; NK, natural killer; PBMCs, peripheral blood mononuclear cells.