Figure 1.
Growth delay in Plasmodium falciparum development after NK cell contact.
A: Representative figure of 3D7-iRBC before the start of co-cultures. Parasites were synchronized by magnetic cell sorting columns for late stages. B: Representative figure of 3D7-iRBC after 24 h of incubation. The parasites have developed into normal ring-stage forms that are expected for this time point. Infected RBC were either co-cultured 3∶1 with autologous NK cells (C), NK cells+TKD (pre-stimulated for 5 days before co-culture with 2 µg/ml TKD peptide) (D), PBMCs (E), or PBMCs stimulated with TKD peptide (F). Normal parasite development of a control culture of parasites without leukocytes was observed in parallel. A blood smear was prepared before and after 24 hours of incubation and stained with 10% Giemsa. Experiments were repeated with RBC from 3 different donors.
Figure 2.
Flow cytometry analysis of erythrocytes for possible NK cell ligands.
0.5×106 i/uRBC or Hela cells were stained with anti-hMICA/B-PE (A), anti-hHLA-E-PE (B) or the respective isotype control. As a control, 0.5×106 iRBC, uRBC or NK92 cells were also stained with anti-hHLA-E-PE (B) or a PE-isotype control. The presence of Hsp70 on uRBC (C) and iRBC (D, E) was determined with anti-Hsp70 mAb cmHsp70.1-FITC compared to FITC isotype. Parasite DNA was stained with Hoechst dye (schizonts, D) that is detected in the Pacific blue channel or Hydroethidine (rings, E) that is detected in the PE channel. Experiments were repeated 3 times.
Figure 3.
Presence of Hsp70 in the membrane of iRBC or senescent uRBC.
Cytosolic and membrane protein extracts were prepared from iRBC and uRBC and submitted to SDS-PAGE. Western blots were incubated with anti-Hsp70 antibody or anti-β-Actin to control protein loading. Experiments were repeated 3 times. A: representative immunoblot of 3–4 week old erythrocytes B: representative immunoblot of fresh blood erythrocyte extracts.
Figure 4.
Surface expression analysis of possible recognition receptors of iRBC on NK92 cells.
NK92 cells were analyzed by flow cytometry for surface expression of NKG2C and CD94. 0.5×106 NK92 cells were stained with anti-CD56-FITC, anti-CD3-PE and anti-NKG2C-APC (A) or anti-CD94-FITC (B) after 24 h incubation in growth medium (GM), with 1.5×106 iRBC, 1.5×106 uRBC, or IL-12/-18 (IL) and analyzed by flow cytometry; a total of 10,000 events was counted for each sample. CD56+/CD3− cells were gated according to FSC/SSC properties in the unstained autofluorescence control and compared to the respective isotype controls. Experiments were repeated three times.
Figure 5.
Transcriptional and translational changes of GzmA, GzmB and Perforin in NK92 cells after 24 hours stimulation.
NK92 cells were left untreated (ns) or cultured with IL-12/-18 (IL), IFN-α, uRBC or iRBC (1∶3) for 24 h. A: Changes on transcriptional level of stimulated cells in comparison to untreated cells were analyzed for GzmA, GzmB and Perforin. β-Actin expression served as house-keeping gene normalizer. Data are represented as mean ± SD (student's t test * p<0.05; ** p<0.01) and are representative of three independent experiments each performed in duplicate. B: After 24 h, 1×107 NK92 cells were lysed, incubated 30 min on ice and subsequently centrifuged at 13000× g for 15 min at 4°C. 10 µg of total supernatant protein were separated by 7.5% SDS-PAGE and blotted onto a nitrocellulose membrane. After blocking, membranes were incubated for 1 h with anti-β-Actin (lane 1), anti-hGzmA (lane 2), anti-hGzmB (lane 3), or anti-hPrf (lane 4).
Figure 6.
Granzyme B-Elispot of NK92 cells (A) or isolated NK cells (B,C) after 24 hours of co-culture with i/uRBC.
Some cultures were pre-activated 5 days with TKD peptide or the scrambled NGL peptide and/or pre-incubated with blocking Hsp70 antibody (cmHsp70.2) or a blocking antibody IgM-isotype control for 20 minutes before the start of the experiment. After stimulation, 2000 NK cells were cultured either alone, with iRBC or uRBC (1∶3 or 10∶1) on a GzmB antibody-coated 96-well plate. GzmB-releasing cells were counted after 24 hours of incubation (* p≤0.05, ** p≤0.01, *** p≤0.001, student's t test, n = 6).
Figure 7.
Eryptosis of iRBC after co-culture with PBMCs or purified NK cells.
iRBC were co-cultured 3∶1 for 24 hours with NK cells, NK cells+TKD (5 day stimulation), PBMC, PBMC+TKD (5 day stimulation) of the same donor. 0.5×106 erythrocytes were washed in 5 mM Ringer solution. Afterwards, cells were stained for 15 minutes with Annexin-V (1∶500) and propidium iodide (1∶50). Eryptotic cells were determined as Annexin-V-positive (AV+) and propidium iodide-negative (PI−). A: Baseline levels of eryptotic uRBC and iRBC at the start of the experiment and after 24 h of culture in growth medium without leukocytes. RBC were stained with AV+PI or left unstained. Gating was done based on FSC/SSC properties and the unstained control. AV was measured in the FL1 channel and PI in the FL2 channel. B: Percentage of AV+/PI−-iRBC after co-culture in growth medium (GM) or with different effector cells in relation to starting parasitemia (** p≤0.01, student's t test, n = 6). C: Normalized FSC of iRBC after co-culture in growth medium (GM) or with different effector cells (* p≤0.05, student's t test, n = 6). FSC values of co-cultured iRBC were normalized to untreated iRBC (GM). D: PI+ necrotic iRBC after culture in growth medium (GM) or with various effector cells. Numbers of necrotic cells were normalized to untreated iRBC (GM).
Figure 8.
Hypothetical model of NK cell response to iRBC and senescent uRBC.
If erythrocytes become senescent (A) or are infected with Plasmodium falciparum (B,C) host-Hsp70 will be recruited to the cell membrane. ?1 NK cells recognize Hsp70-exposure by yet unknown receptors, possibly CD94. Recognition of Hsp70 leads to GzmB release. ?2 GzmB will enter the target cell with either assistance of Hsp70, an unknown receptor or become endocytosed. ?3 Once inside the iRBC, GzmB induces eryptosis. Pre-stimulation with TKD peptide enhances both GzmB release and the amount of eryptotic iRBC (C).