Fig 1.
Variation of NK cell responses to iRBCs among different individuals.
Human NK cells were purified (>95%) from fresh blood and co-cultured with iRBCs for 96 h. Parasitemia and the expression of various NK cell surface markers and intracellular proteins were assayed by flow cytometry. The level of cytokines in the culture supernatants was quantified by multiplex immunoassay at 48 h. (A) Representative CD56 vs. Hoechst staining profiles of iRBCs alone, iRBCs co-cultured with responder NK cells (R-NK) or with non-responder NK cells (NR-NK). The number represents the percentage of iRBCs (parasitemia). CD56-positive cells are NK cells. (B) Paired plots of parasitemia in the absence of NK cells (No NK) and with either R-NK or NR-NK. (C) Comparison of reduction in parasitemia in the presence of either R-NK (blue) or NR-NK (red) cells from different malaria-naïve individuals. (D) Percentage of live (DAPI-negative) NK cells after 96 hrs of co-culture. (E) Representative histograms comparing CD69, CD107 and CD25 expression on R-NK and NR-NK cells from one responder (blue trace) and one non-responder (red trace). (F-H) Paired plots showing changes in the percentage of NK cells positive for CD69 (F), CD107a (G) and CD25 (H) following co-culture with either RBC or iRBC. (I) Comparison of log2 fold change in expression levels of CD69, CD107 and CD25 between R-NK and NR-NK cells following co-culture with iRBC. (J) Representative histograms comparing granulysin, IFN-γ and perforin expression on R-NK and NR-NK cells from one responder and one non-responder. (K-M) Paired plots showing changes in the percentage of NK cells positive for granulysin (K), IFN-γ (L) and perforin (M) following co-culture with either RBC or iRBC. (N) Comparison of soluble mediators in the culture supernatants. (O) Comparison of reduction in parasitemia following co-culture of R-NK or NR-NK cells in the presence of different strains of parasites. R-NK and NR-NK cells were co-cultured with the indicated parasite strain for 96 h. Parasitemia was quantified by flow cytometry. (P) Comparison of reduction in parasitemia in the presence of NK cells from either uncomplicated malaria (UM) patients or severe malaria SM patients. NK cells were purified from frozen buffy coat and co-cultured with 3D7-infected RBCs for 96 h and parasitemia was quantified by flow cytometry. Each symbol in B-D, F-I, K-P represents a different individual. Joined lines show experimental pair. Error bars represent mean ± SD. * p<0.05, ** p<0.01, *** p<0.001, **** p<0.0001, ns: not significant.
Fig 2.
Transcriptional analysis of NK cells in response to RBCs and iRBCs.
NK cells from five responders and five non-responders were co-cultured with either iRBCs or RBCs for 96 h. NK cells were purified from the four cultures for RNA isolation, cDNA library construction and microarray analysis. (A) sPLS-DA supervised clustering of the indicated conditions. (B) Pairwise comparison of all groups. (C) Heatmap of expression levels of identified DEGs. Each row represents a gene and each column represents a sample of the indicated group. Hierarchical clustering of columns and rows were performed using Euclidean distance and represented as a dendrogram. The organization and length of the branches in the dendrogram reflect similarities in gene expression profiles. (D) Top 10 pathways obtained from functional network-based analysis of the DEGs in R-iRBC samples.
Fig 3.
Requirement of MDA5 in NK cell responses to iRBCs.
(A) Heatmap of DEGs involved in RLR signaling. Each column represents an individual of the indicated group. (B-C) NK cells were co-cultured with iRBCs for 96 h, and intracellular staining of MDA5 was analyzed by flow cytometry. Shown are representative histogram (B) and MFI (C) of MDA5 staining of R-NK and NR-NK cells. (D) Western blot of MDA5 and GAPDH levels in R-NK cells transduced with lentivirus expressing CRISPR/Cas9 and a gRNA targeting either MDA5 (gMDA5) or a scramble sequence (gScrble). (E-F) Transduced NK cells were co-cultured with iRBCs for 48 h and parasitemia was quantified. Representative Hoechst vs CD56 staining profiles of iRBCs alone, iRBCs co-cultured with R-NK cells with or without MDA5 knockdown (E) and comparison of reduction in parasitemia (F). Numbers in E indicate parasitemia. (G-H) R-NK cells were co-cultured with either lipofectamine or varying concentrations of lipofectamine-formulated (Lf)-poly I:C for 48 h. Surface expression of CD69 was determined by flow cytometry (G), and IFN-γ secretion was determined by immunoplex assay (H). (I) iRBCs were cultured alone or with either R-NK or NR-NK cells in the presence of 1000μg/ml lipofectamine-formulated poly I:C for 96 h. Parasitemia was quantified by flow cytometry. Representative Hoechst vs CD56 staining profiles is shown. (J) R-NK and NR-NK cells were co-cultured with iRBCs under the indicated conditions for 96 h, and reduction in parasitemia was quantified by flow cytometry. Poly I:C was used at 100 μg/ml (poly I:C-100). poly I:C (Lf) was used at 100, 10 and 1 μg/ml. (K) NR-NK were co-cultured with iRBCs with or without 5’pppRNA (Lf) (10 μg/ml), and reduction in parasitemia was quantified by flow cytometry. Each symbol represents a different individual. Error bars represent mean ± SD. *p<0.05, **p<0.01, ***p<0.001, ****p<0.0001, ns: not significant.
Fig 4.
TBK1/IKKε mediates NK cell responses to iRBCs.
R-NK cells were co-cultured with iRBCs in the presence or absence of TBK1/IKKε inhibitor Bx795 (10 nM) for 96 h and parasitemia was quantified by flow cytometry. Representative Hoechst vs CD56 staining profiles of indicated cultures. Numbers indicate parasitemia (A). Comparison of reduction in parasitemia (B). Representative histograms of CD69 expression (C) and comparison of CD69 MFI (D) on NK cells from the indicated co-cultures at 96 h. (E) Quantification of the percentage of CD25-positive NK cells after co-culture for 96 h. (F) IFN-γ levels in the supernatants of indicated co-cultures after 96 h. Each symbol represents an individual. Error bars represent mean ± SD. *p<0.05, ****p<0.0001.
Fig 5.
Microvesicles derived from iRBCs prime NK cells.
(A-B) iMVs (1 μg/ml) were either labeled or not labeled with PKH26 and then cultured with R-NK cells for 24 h. Shown are representative PKH26 vs CD56 staining profiles. Numbers indicate the percentage of positive cells in the gated region (A). Comparison of percentages of PKH26-positive NK cells in cultures with unlabeled or labeled iMV (B). (C) Confocal microscopy of PKH67-labelled R-NK cells after 24 h incubation with PKH26-labeled iMVs. Shows are representative images of an NK cell with an internalized iMV. Z-stack of 1 μm slices were captured and 3D volume rendering was performed to demonstrate the location of the internalized iMV (white arrow). (D) Gel electrophoresis of cDNA derived from capped and uncapped RNAs of MVs and iMVs. Size markers are labeled on the left. No RNA: no RNA was added during the reverse transcription. (E-F) Representative CD69 histograms of R-NK cells cultured with RBC (NK+RBC), iRBCs (NK+iRBCs), MVs at 10 μg/ml (NK+MV10), iMVs at 1 (NK+iMV1) and 10 μg/ml (NK+iMV10) (E) and comparison of CD69 MFI on R-NK cells from different donors in the indicated culture conditions (F). (G) Comparison of CD69 MFI on R-NK cells cultured with iMVs (at 10 μg/ml) in the presence of DMSO (control) or Bx795 (10 nM). (H) IFN-γ levels in the culture supernatants of R-NK cells cultured with RBCs, iRBCs, and iMVs at 100 μg/ml for 48 h. (I-J) R-NK cells were cultured in the presence of MVs or iMVs at 10 μg/ml for 96 h, and LFA-1 activation was assessed using m24 antibody. Shown are representative LFA-1 (m24) vs CD56 staining profiles of NK cells (I), and comparison of percentages of LFA-1 (m24)-positive NK cells (J) in the indicated conditions. (K) Comparison of parasitemia reduction when NR-NK cells were co-cultured with iRBCs without iMVs, with 10 μg/ml iMVs (iMV10) or 100 μg/ml iMVs (iMV100) for 96 h. Each symbol represents a different individual. Error bars represent mean ± SD. *p<0.05, **p<0.01, ***p<0.001, ****p<0.0001, ns: not significant.