Efficient engulfment of necroptotic and pyroptotic cells by nonprofessional and professional phagocytes

Dear Editor, Engulfment of dead cells is essential for tissue homeostasis and prevention of inflammation. Classical theory is that this process is performed largely by professional phagocytes such as macrophages and dendritic cells (DC). However, increasing data suggest a role of neighboring nonprofessional cells in this process. Uptake of apoptotic cells has been studied intensively, which involves recognition of apoptotic cells by receptors on engulfing cells that recognize the so-called ‘eat-me’ signals on the surface of the apoptotic cells. Clearance of necrotic cells is thought of high importance to tissue homeostasis as necrotic cells could leak intracellular materials that might evoke inflammation. There are studies of necrotic cell clearance, but it is limited especially for the recently uncovered programmed necrosis including necroptosis and pyroptosis. The ‘eat-me’ signal phosphatidylserine (PS) can be detected on necroptotic cells and was believed to drive recognition and phagocytosis by professional phagocytes. However, no direct evidence has been found to support that PS is required for the phagocytosis of necroptotic cells. Because of the interconversion between apoptosis and necroptosis or pyroptosis, mixed types of cell death often occurred in experiments, which could lead to misinterpretation of the results. In order to specifically induce a certain type of cell death, we used a dimerization/oligomerization system, which is based on antiestrogen 4-hydroxytamxifen (4-OHT)-induced homo-dimerization of hormonebinding domain’s G521R mutant (HBD*) of estrogen receptor, to artificially induce oligomerization of signaling molecules in cell death pathways in our previous studies. Inducible oligomerization of caspase-8, mixed lineage kinase domain-like N-terminal domain (MLKLND), and gasdermin D N-terminal domain (GSDMD-N) in given cells such as NIH3T3 was used here to specifically induce apoptosis, necroptosis, and pyroptosis, respectively (Supplementary Fig. S1a). We detected PS exposure in all three types of cell death (Supplementary Fig. S1b). The PS positive cells were committed to death since they could not regrow (Supplementary Fig. S1c). We analyzed their phagocytosis by their live peers (nonprofessional phagocytes) using NIH3T3 cells and used heat-killed NIH3T3 cells as controls. Whereas heat-killed cells were engulfed very inefficiently, we observed the uptake of not only apoptotic cells but also necroptotic and pyroptotic cells (Fig. 1a, b; Supplementary Fig. S2a, c). In contrast to the common assumption that apoptosis would undergo the process of phagocytosis easily, necroptotic cells and pyroptotic cells were engulfed much more efficiently than apoptotic cells by their live peers (Fig. 1b; Supplementary Fig. S2a). The engulfment of necroptotic and pyroptotic cells can take place among the same type of cells and also between different types of cells (Supplementary Fig. S3a, b). More phagocytosis of TNFinduced necroptotic cells was also observed than that of apoptotic cells (Supplementary Fig. S3c). We then compared phagocytosis of necrotic cells with that of apoptotic cells by bone marrow-derived macrophages (BMDM), peritoneal macrophages, and bone marrow-derived dendritic cells (BMDC), and found that necroptotic and


Phagocytosis assay
Cultured cells were labeled with indicated dye as mentioned in the main text. Live NIH3T3, L929, macrophages and dendritic cells were seeded in 48-well plate at a density of 1.5 x 10 5 per well, and 1.5 x 10 5 of indicated dying/dead cells were added to co-incubation for indicated time. Then cells were detached using EDTA or trypsin-EDTA, washed by PBS and collected to analyze. For professional phagocytes, cells were marked by anti-CD11b (macrophages) or anti-CD11c (dendritic cells) antibodies. Phagocytosis was measured by flow cytometry using Fortessa X-20 (BD biosciences). The percentage of live CellTrace + NIH3T3/L929 cells, CD11b + macrophages or CD11c + dendritic cells that had engulfed PKH + cells was calculated (PKH + CellTrace + / CellTrace + , PKH + CD11c + / CD11c + or PKH + CD11b + / CD11b + ). In some experiments, cells were plated on coverslips or glass-bottom dishes for co-incubation and images were generated using microscopy.

Microscopy
For fixed cell imaging, cells were plated on coverslips pre-coated with poly-L-lysine. In some experiments, cells were stained first with CFSE or PKH26/67 as indicated. After treatment, cells were washed with PBS followed by fixation with 4% paraformaldehyde in PBS. In some experiments, cells were then permeabilized with 0.2% Triton X-100 in PBS, blocked with 3% BSA in PBS and stained with anti-Lamp1 antibody and secondary antibody, or stained with Alexa Fluor 647 phalloidin. Cells were counterstained with Hoechst to visualize nuclei. Images were generated using Zeiss LSM 780 confocal microscope or Delta-Vision OMX v4 (GE Healthcare).
For time-lapse microscopy live cell imaging, PKH26-labeled L929 cells and lifeact-EGFPexpressing MLKL KO L929 cells were seeded in 35mm glass-bottom dishes in the ratio of 1:1. TZ was added to medium to induce necroptosis. Then imaging was carried out using Zeiss LSM 780 confocal microscope.

(e) Statistical analysis of data of source images in (d).
(f) PKH26-labeled necroptotic, apoptotic, pyroptotic and heat-killed cells treated as in Fig. 1a were co-cultured with CFSE-labeled BMDC in the ratio of 1:1 for 4 hrs. Representative confocal microscopy images are shown. Arrows indicate live cells that had engulfed dead cell(s). Scale bars, 10μm.
(g) Statistical analysis of data of source images in (f).
(h) PKH26-labeled necroptotic, apoptotic, pyroptotic and heat-killed cells treated as in Fig. 1a were co-cultured with peritoneal macrophage in the ratio of 1:1 for 4 hrs. Phagocytosis was analyzed using flow cytometry by measuring the percentage of CD11b + peritoneal macrophages that had engulfed PKH26 + NIH3T3 cells.
(i) PKH26-labeled necroptotic, apoptotic, pyroptotic and heat-killed cells treated as in Fig. 1a were co-cultured with BMDC in the ratio of 1:1 for 4 hrs. Phagocytosis was analyzed using flow cytometry to measure the percentage of CD11c + BMDC that had engulfed PKH26 + cells.
(j) PKH26-labeled necroptotic and apoptotic cells treated as in Fig. 1a were co-cultured with CFSE-labeled live NIH3T3 in the ratio of 1:1 and phagocytosis was analyzed by flow cytometry at indicated time points.
Supplementary Fig. S4   (g) Necroptotic or apoptotic NIH3T3 cells treated as in Fig. 1 were pretreated with BSA or Annexin V (100μg/ml) for 30 mins before co-culture with NIH3T3 (WT) or Tim4-overexpressing NIH3T3 (Tim4) for 4 hrs in the ratio of 1:1 (The final concentration of Annexin V and BSA was 50μg/ml).
Phagocytosis was analyzed by flow cytometry.