Ligand-dependent kinase activity of MERTK drives efferocytosis in human iPSC-derived macrophages

Abstract

Removal of apoptotic cells by phagocytes (also called efferocytosis) is a crucial process for tissue homeostasis. Professional phagocytes express a plethora of surface receptors enabling them to sense and engulf apoptotic cells, thus avoiding persistence of dead cells and cellular debris and their consequent effects. Dysregulation of efferocytosis is thought to lead to secondary necrosis and associated inflammation and immune activation. Efferocytosis in primarily murine macrophages and dendritic cells has been shown to require TAM RTKs, with MERTK and AXL being critical for clearance of apoptotic cells. The functional role of human orthologs, especially the exact contribution of each individual receptor is less well studied. Here we show that human macrophages differentiated in vitro from iPSC-derived precursor cells express both AXL and MERTK and engulf apoptotic cells. TAM RTK agonism by the natural ligand growth-arrest specific 6 (GAS6) significantly enhanced such efferocytosis. Using a newly-developed mouse model of kinase-dead MERTK, we demonstrate that MERTK kinase activity is essential for efferocytosis in peritoneal macrophages in vivo. Moreover, human iPSC-derived macrophages treated in vitro with blocking antibodies or small molecule inhibitors recapitulated this observation. Hence, our results highlight a conserved MERTK function between mice and humans, and the critical role of its kinase activity in homeostatic efferocytosis.

Fig. 1. Human iPSC-derived macrophages express AXL and MERTK.

A–H Macrophage progenitor cells were generated from iPSCs and cultured in vitro using either M-CSF or GM-CSF. Macrophages were stimulated with Poly I:C or dexamethasone or left untreated. After 24 h AXL and MERTK expression was determined by flow cytometry or by qRT-PCR. A–D Representative histograms showing expression of AXL and MERTK on M-CSF differentiated macrophages upon stimulation using indicated conditions. Bar graphs show geometrical mean fluorescence intensity of AXL and MERTK or mRNA expression levels of Axl and Mertk. Shown is mean with SD and individual samples. E–H Representative histograms showing expression of AXL and MERTK on GM-CSF differentiated macrophages upon stimulation with indicated conditions. Bar graphs show geometrical mean fluorescence intensity of AXL and MERTK or mRNA expression levels of Axl and Mertk. Shown is mean with SD and individual samples. I, J Absolute flow cytometric quantification of AXL and MERTK cell surface expression on iPSC- and monocyte-derived macrophages. Antibodies per cell (ABC) have been determined using a bead based assay. All experiments have been performed with three (flow cytometry) or four (qRT-PCR) technical replicates. Data were representative of at least three individual experiments.

Fig. 2. GAS6 enhances efferocytosis in vitro in human iPSC-derived macrophages.

A Flow cytometric analysis of efferocytosis in vitro using iPSC-derived macrophages. Apoptotic cells (ACs) were generated as described and live cells (LC) were used as control. ACs were labeled with indicated concentrations of GAS6. For LCs, GAS6 was added soluble. Macrophages and ACs were cultured for 2 h at a ratio of 1:6 and analyzed afterwards for pHrodo signal. B Frequency of pHrodo⁺ macrophages. Bar graphs show mean with SD and individual samples. C High content imaging to determine frequency of efferocytic macrophages. Macrophages were stained with Calcein-AM prior to the assay. ACs were labeled with indicated concentrations of GAS6 and added to macrophages for 2 h. Afterwards, efferocytic macrophages were quantified using the Operetta CLS High-Content Analysis system. D Frequency of pHrodo⁺ macrophages. Bar graphs show mean with SD. E Flow cytometric analysis of efferocytosis in vitro using monocyte-derived macrophages. F Frequency of pHrodo⁺ macrophages. Bar graphs show mean with SD and individual samples. All experiments have been performed with three (flow cytometry) or 32 (imaging) technical replicates. Data were representative of at least three individual experiments.

Fig. 3. GAS6 augments homeostatic efferocytosis in murine peritoneal macrophages.

A Gating strategy to identify murine large (LPM) and small (SPM) peritoneal macrophages. Cells were gated as CD19⁻, Ly6G⁻ live cells. B Representative flow cytometric analysis for expression of AXL and MERTK on LPM and SPM. C Frequency of AXL or MERTK single and double positive LPM and SPM. Bar graphs show mean with SD and individual samples. D Experimental set-up to study efferocytosis in the murine peritoneum. Apoptotic cells (AC) were generated by treating murine thymocytes with dexamethasone. Afterwards, cells were labeled with pHrodo-Red and washed with EDTA/BSA and PBS. To set-up the model, initially different numbers of ACs were injected intraperitoneally. Peritoneal exudate cells (PECs) were analyzed at different time points by flow cytometry. Created with BioRender.com E Representative flow cytometric analysis of PECs for expression of F4/80 and pHrodo-Red signal. F Representative histograms for pHrodo-Red upon injection of different numbers of ACs and PECs isolated after 4 h (Gated on CD19⁻, F4/80⁺ live cells). Statistical analysis of pHrodo⁺ cells. Bar graphs show mean with SD and individual samples. G Representative histograms for pHrodo-Red upon injection of 2.5 × 10⁶ ACs and PECs were isolated at indicated time points (Gated on CD19⁻, Ly6G⁻, and F4/80⁺ live cells). Statistical analysis of pHrodo⁺ cells. Bar graphs show mean with SD and individual samples. H Representative histograms for pHrodo-Red upon injection of 2.5 × 10⁶ ACs with or without 50 nM GAS6. PECs were isolated at indicated time points (Gated on CD19⁻, Ly6G⁻, and F4/80⁺ live cells). Bar graphs show mean with SD and individual samples. Data were representative of two individual experiments (n = 3–5).

Fig. 4. MERTK kinase activity is essential for clearance of apoptotic cells in vivo.

A (top) targeted codon in Mertk for the generation of Mertk^K614M/K614M is indicated in blue. A KpnI restriction enzyme site was engineered to aid PCR-based genotyping. (bottom) Amino acid sequence of the targeted region, indicating the introduced mutation at Lysine (K) 614 for Methionine (M) 614. B Diagnostic PCR results, after DNA amplification and digestion with KpnI. The WT allele yields one 548 bp product and the mutant allele yields a 184 and a 364 bp band. C Representative histograms showing expression of MERTK in bone marrow-derived macrophages from mice with the indicated genotypes. Bar graphs show geometrical mean fluorescence of MERTK. Shown is mean with SEM and individual samples. D Representative flow cytometric analysis of efferocytosis with bone marrow-derived macrophages from mice with the indicated genotypes. Macrophages and ACs were co-incubated for 1 h in the presence of serum at a ratio of 1:6 and analyzed afterwards for pHrodo signal. Bar graphs show mean with SEM and individual samples. E Experimental set-up to study contribution of AXL and MERTK for efferocytosis in the murine peritoneum. ACs (2.5 × 10⁶) were injected intraperitoneally into mice of indicated genotype. After 4 h PECs were analyzed via flow cytometry. Created with BioRender.com. F Representative histograms for pHrodo-Red upon injection of ACs into mice of indicated genotype (Gated on CD19⁻, Ly6G⁻, and F4/80⁺ live cells). Frequency of pHrodo⁺ cells. Bar graphs show mean with SD and individual samples. Data were representative of two individual experiments (n = 3–5).

Fig. 5. MERTK kinase activity is critical for efferocytosis in human iPSC-derived macrophages in vitro.

A Flow cytometric analysis of efferocytosis in vitro using iPSC-derived macrophages. Macrophages were stimulated with Poly (I:C) or dexamethasone for 24 h prior to the assay. ACs were generated as described and labeled with 50 nM GAS6. Prior to addition of apoptotic cells, macrophages were incubated with indicated antibodies for 30 min. B Frequency of pHrodo⁺ macrophages. Bar graphs show mean with SD and individual samples. C Flow cytometric analysis of efferocytosis in vitro using iPSC-derived macrophages. Macrophages were stimulated with Poly (I:C) or dexamethasone for 24 h prior to the assay. ACs were generated as described and labeled with 50 nM GAS6. Prior to addition of apoptotic cells, macrophages were incubated with indicated inhibitors for 30 min. D Frequency of pHrodo⁺ macrophages. Bar graphs show mean with SD and individual samples. Data were representative of at least three individual experiments with technical replicates.