Efferocytosis dysfunction in CXCL4-induced M4 macrophages: phenotypic insights in systemic sclerosis in vitro and in vivo

Abstract

Introduction: Systemic sclerosis (SSc) is an autoimmune disease characterized by antinuclear antibody production, which has been linked to an excess of apoptotic cells, normally eliminated by macrophages through efferocytosis. Additionally, circulating levels of CXCL4, a novel SSc biomarker, correlate with more severe fibrotic manifestations of the disease. Considering the defective efferocytosis of macrophages in SSc and the CXCL4-related M4 macrophage phenotype, we hypothesized that CXCL4 could be involved in the alteration of phagocytic functions of macrophages in SSc, including LC3-associated phagocytosis (LAP), another phagocytic process requiring autophagy proteins and contributing to immune silencing.

Methods: In this study, CXCL4 levels were measured by ELISA in vitro in the serum of SSc patients, and also in vivo in the serum and lungs of C57BL/6J SSc mice induced by intradermal injections of hypochloric acid (HOCl) or Bleomycin (BLM), with evaluation of M4 markers. Circulating monocytes from healthy donors were also differentiated in vitro into M4 monocytes-derived macrophages (MDMs) in the presence of recombinant CXCL4. In M4-MDMs, phagocytosis of fluorescent beads and expression level of efferocytic receptors were evaluated by flow cytometry in vitro, while efferocytosis of pHrodo-stained apoptotic Jurkat cells was evaluated by real-time fluorescence microscopy. LAP quantification was made by fluorescence microscopy in M4-MDMs exposed to IgG-coated beads as well as apoptotic Jurkat cells.

Results: Our results demonstrated that efferocytosis was significantly reduced in M0-MDMs from healthy donors exposed to the CXCL4-rich plasma of SSc patients. In vivo, CXCL4 expression was increased in the lungs of both SSc-mouse models, along with elevated M4 markers, while efferocytosis of BLM-mice alveolar macrophages was decreased. In vitro, M4-MDMs exhibited reduced efferocytosis compared to M0-MDMs, notably attributable to lower CD36 receptor expression and impaired phagocytosis capacities, despite enhanced LAP. Autophagic gene expression was increased both in vitro in SSc MDMs and in vivo in BLM mice, thus acting as a potential compensatory mechanism.

Discussion: Altogether, our results support the role of CXCL4 on the impaired efferocytosis capacities of human macrophages from SSc patients and in SSc mice.

Keywords: CXCL4; efferocytosis; fibrosis; macrophage; phagocytosis; systemic sclerosis.

Figure 1

Role of CXCL4-rich plasma on efferocytosis capacities of human monocyte-derived macrophages (MDMs) (A) Evaluation of efferocytosis capacities of CFSE⁺ apoptotic Jurkat cells by M0-MDMs from healthy donors (HD) and SSc patients by flow cytometry. The efferocytic index represents the mean ± SEM of HD (n=27) and patients with SSc (n=19). Student unpaired t-test, ***p< 0.001. (B) Evaluation of CXCL4 levels (µg/ml) in the serum of HD (n=42) and patients with SSc (n=150). Student unpaired t-test, *p< 0.05. (C) Evaluation of efferocytosis capacities of M0-MDMs exposed to HD or SSc plasma. The engulfment of CFSE⁺ apoptotic Jurkat cells by MDMs was expressed as a % of MDMs by flow cytometry after matching on age and sex for both HD and SSc. Data are expressed as mean ± SEM. ANOVA Newman-Keuls multiple comparison post hoc test, *p< 0.05; **p< 0.01.

Figure 2

SSc-ILD mouse models induced by either HOCl or BLM exhibit CXCL4-MDM markers, a phenotype associated with defective efferocytosis function in vivo. (A) Evaluation of CXCL4 secretion levels (pg/ml) by ELISA in the sera of HOCl and BLM mouse models in comparison with the saline (NaCl) control group. All experiments are the results of duplicate experiments and are expressed as means ± SEM (n = 5 mice per group). Student unpaired t-test, **p < 0.01. (B) Evaluation of M4 signature in the lung of HOCl and BLM mouse models. The mRNA relative expressions are expressed as means ± SEM with saline group (NaCl) arbitrarily set to 1 (n=4 to 9 mice). Student unpaired t-test, *p< 0.05; **p< 0.01; ***p<0.001. (C) Evaluation of S100A8 secretion levels (pg/ml) in lung extracts of BLM mice in comparison with the saline (NaCl) control group by ELISA. The results are expressed as means ± SEM (n = 5 mice per group). Student unpaired t-test, *p< 0.05. (D) Evaluation of efferocytosis capacities of murine alveolar macrophages (AM) in BLM mouse model. The percentage of AM (Gr1^Int and CD11b^Int) that have engulfed CFSE⁺ apoptotic Jurkat cells and the median fluorescence intensity (MFI) of CFSE+ AM were presented as the means ± SEM (n = 6 mice per group). Student unpaired t-test, ***p<0.001.

Figure 3

Phenotype and functions of CXCL4-induced M4-MDMs. (A) Simplified schema summarizing the steps to follow to obtain polarized human MDMs from healthy donors (HD). (B) Morphology of human MDMs (M0, M4, and GM) observed by light microscopy at x20 magnification. (C) Evaluation of mRNA expression by RT-qPCR in M4-MDMs relative to control resting M0-MDMs, arbitrarily set to 1 unit (dashed line). All experiments are the results of duplicate experiments conducted in MDMs from n = 4 to 6 independent HDs and are expressed as means ± SEM. Student paired t-test, *p< 0.05. (D) Evaluation of cytokine secretion levels (pg/ml) in the culture medium of M4-MDMs relative to control resting M0-MDMs by ELISA. All experiments are the results of duplicate experiments conducted in MDMs from n = 4 to 6 independent HDs and are expressed as means ± SEM. Student paired t-test, *p< 0.05. (F) Evaluation of phagocytic capacities of M4 by flow cytometry (see methods) relative to control resting M0-MDMs. Data are expressed as mean % of phagocytosis (ratio of median fluorescence intensity (MFI) at 37°C/MFI at 4°C) ± SEM from 4 independent HDs, ANOVA Dunnett’s post hoc test compared to resting M0-MDMs, * p< 0.05. (E) Evaluation of efferocytosis capacities of pHrodo⁺ apoptotic Jurkat cells by M4-MDMs compared to resting M0-MDMs pre-treated or not with CytoD. Data are expressed as mean pHrodo fluorescence intensity ± SEM of 4 independent experiences in duplicate, ANOVA Dunnett’s post hoc test compared to resting M0-MDMs. *p< 0.05; ***p< 0.001. (G) Expression of cell surface receptor involved in efferocytosis in M4-MDMs by flow cytometry relative to control resting M0-MDMs. Data are expressed as mean fluorescence intensity (MFI) relative to isotype control (ratio) ± SEM, from 5 independent HDs, Student paired t-test, *p< 0.05.

Figure 4

LAPosome quantification in polarized MDM from healthy donors (HD). (A, B) M0 and M4 were primarily exposed to either PKH67-stained (green) apoptotic Jurkat cells (PKH67+ apoJ) (white arrow, A), or IgG-coated beads (dotted arrows, B) for 45 min. Nuclei were then stained with DAPI (blue), and LC3 stained with rabbit anti-LC3B primary antibody targeted with goat anti-rabbit Alexa647 secondary antibody (red). LC3+ phagosomes (LAPosomes) are characterized by circular red fluorescence recruitment around phagosomes (yellow arrows). (C, D) Evaluation of phagocytosis capacities of M4-MDMs relative to M0-MDMs were obtained from 5 independent HDs, and not 4. (C) or IgG-coated beads (D) after 45 min of phagocytosis. Phagocytosis data represent the mean ± SEM of the percentage of MDMs that have phagocytosed either at least one PKH67+ apoptotic Jurkat cells (apoJ) (C), or at least one IgG-coated bead (D), from 4 independent HDs. Ten random quantifications per condition were performed manually and blindly by two independent investigators with the Fiji software and averaged for each condition. Student paired t-test, *p < 0.05. (E, F) Evaluation of LC3-associated phagocytosis (LAP) capacities of M4-MDMs relative to M0-MDMs. LAPosomes are identified as fluorescence recruitment of LC3 around phagosomes in cells containing either PKH67+ apoptotic Jurkat cells (apoJ) (E) or IgG-coated beads (F) after 45 min of phagocytosis, as described in (A, B), respectively. LAP data represent the mean ± SEM of the percentage of MDMs showing at least one LC3+ phagosome, among MDMs that have phagocytosed either PKH67+ apoJ (E) or IgG-coated beads (F), from 5 independent HDs. Ten random quantifications per condition were performed manually and blindly by two independent investigators with the Fiji software and averaged for each condition. Student paired t-test, *p < 0.05.

Figure 5

mRNA expression of genes involved in LAP. (A) mRNA expression in the lung of BLM-exposed mice (n=6 mice) compared to saline group (n=4 to 6). Student unpaired t-test, *p< 0.05; **p< 0.01; ***p<0.001. (B) mRNA expression in SSc-MDMs compared to HD-MDMs. All experiments are the results of duplicate experiments conducted in MDMs from independent HDs (n=8 to 14) and SSc patients (n=19 to 27). Student unpaired t-test, *p< 0.05. (C) mRNA expression in M4-MDM compared to M0-MDMs. All experiments are the results of duplicate experiments conducted in MDMs from n = 6 to 8 independent HDs and are expressed as means ± SEM. Student paired t-test, *p< 0.05.