Ezrin drives adaptation of monocytes to the inflamed lung microenvironment

Abstract

Ezrin, an actin-binding protein, orchestrates the organization of the cortical cytoskeleton and plasma membrane during cell migration, adhesion, and proliferation. Its role in monocytes/macrophages (MΦs) is less understood. Here, we used a monocyte/MΦ-specific ezrin knock-out mouse model to investigate the contribution of ezrin to monocyte recruitment and adaptation to the lung extracellular matrix (ECM) in response to lipopolysaccharide (LPS). Our study revealed that LPS induces ezrin expression in monocytes/MΦs and is essential for monocytes to adhere to lung ECM, proliferate, and differentiate into tissue-resident interstitial MΦs. Mechanistically, the loss of ezrin in monocytes disrupts activation of focal adhesion kinase and AKT serine-threonine protein kinase signaling, essential for lung-recruited monocytes and monocyte-derived MΦs to adhere to the ECM, proliferate, and survive. In summary, our data show that ezrin plays a role beyond structural cellular support, influencing diverse monocytes/MΦ processes and signaling pathways during inflammation, facilitating their differentiation into tissue-resident macrophages.

Fig. 1. Ezrin is induced in monocytes and MΦs in response to LPS.

Quantitative PCR (qPCR) for ezrin in WT and Ezr-KO^m mouse bone marrow-derived macrophages (BMD-MΦs) (A) and primary bone marrow (BM) monocytes (B), untreated or treated with LPS for 6 h and 24 h. The relative mRNA expression of ezrin was normalized to STX5a and compared to WT untreated (Untr). Western blot (WB) and densitometric analysis for ezrin and moesin in mouse BMD-MΦs (C) and BM monocytes (D), untreated or treated with LPS for 6 h and 24 h. Protein fold increase was normalized to β-actin and relative to untreated cells. E Cartoon representation of the in vivo model of Pseudomonas aeruginosa—lipopolysaccharide (PA-LPS) nebulization. F qPCR of ezrin in murine WT and Ezr-KO^m monocyte-derived MΦs (moMs), interstitial MΦs (IMs) and alveolar MΦs (AMs), sorted from lung tissues of untreated or LPS treated mice. The relative ezrin mRNA expression was normalized to STX5a and the WT Untr., distinctly for each phenotype. Data are represented as mean ± SEM from three independent experiments with three or more mice per genotype. Statistical analysis was performed using one-way ANOVA or Tukey’s test for multiple comparisons between the genotype and treatment conditions. *p < 0.05, **p < 0.01 and ***p < 0.001. See also Supplementary Fig. S2.

Fig. 2. The loss of ezrin impacts the number of lung monocyte-derived IMs in response to LPS.

WT and Ezr-KO^m mice were nebulized with LPS (12.5 mg/5 ml for 15 min). The mice were sacrificed post-24 h of LPS and collected blood, BALF and lung tissue samples for flow cytometry and RNA analysis. A Numbers and percentages of lung moMs and IMs in lung tissue homogenates (inferior lobe) were quantified by sequential gating strategy from flow cytometry. Cell numbers were calculated from the percentage of viable cells multiplied by the total cell count in the inferior lung lobe. The gating strategy is described in Fig. S2C. B Flow cytometry-based dot plot representation of moMs, transitional IMs (tnsIMs) and mature IMs (matIMs) populations in WT, Ezr-KO^m and CCR2-KO mice, untreated or treated with LPS. C Numbers and percentages of moMs, tnsIMs and matIMs (quantification of (C)) in the lung (inferior lobe) tissue, untreated or treated with LPS. Quantification was performed as in (B). D Number (left) and percentage (right) of monocytes in the peripheral blood. Data are represented as mean ± SEM from three independent experiments with three or more mice per genotype. Each dot represents a biological replicate. Statistical analysis was performed using one-way ANOVA or Tukey’s test for multiple comparisons between the genotype and treatment conditions. *p < 0.05, **p < 0.01 and ns non significant. See also Supplementary Figs. S3–5.

Fig. 3. LPS-treated WT and Ezr-KO^m lung MΦs have an altered expression profile consistent with a lack of differentiation toward a tissue-resident macrophage phenotype.

A Multidimensional scaling (MDS) plot depicting the distinct clusters of moMs, tnsIMs and matIMs in WT and Ezr-KO^m mice based on RNA expression profiles. B Comparative expression of cellular markers and their transition gene expression profiles in moMs, tnsIMs and matIMs between WT and Ezr-KO^m mice treated with LPS. C Hierarchical clustered heatmap depicting RNA expression profiles of differentially expressed genes (DEGs) comparing moMs, tnsIMs and matIMs in WT and Ezr-KO^m mice treated with LPS. Each tile represents the triplicate average of normalized fragments per kilobase of transcript per million mapped (FPKM) reads and colored according to gene-specific z-scores. Data were generated from three mice per genotype. Data are represented as median from three biological replicates. Statistical analysis was performed using a t-test (non-parametric) between the subpopulations. *p < 0.05, **p < 0.01, ***p < 0.001 and ns non significant. See also Supplementary Fig. S6.

Fig. 4. MetaCore enrichment pathway analysis of DEGs in LPS-treated WT and ezrin-KO moMs, tnsMs and matMs.

A–C MetaCore enrichment pathway analysis and volcano plots of differentially expressed genes (DEGs) in Ezr-KO^m vs WT moMs (A), tnsIMs (B) and matIMs (C) populations treated with LPS. Dot plots show upregulated or downregulated pathways in Ezr-KO^m cell phenotypes and their associated genes. D Bioplanet pathway analysis of differentially expressed genes in Ezr-KO^m vs WT moMs, tnsIMs and matIMs populations treated with LPS. Node size: odds ratio associated with enrichment; p-value for Fisher’s exact test. Data were generated from three mice per genotype. See also Supplementary Fig. S6.

Fig. 5. Monocyte/MΦs lacking ezrin exhibit proliferation defects in response to LPS.

A Schematic of in vivo proliferation assay model. WT and Ezr-KO^m mice were administered with BrdU via intraperitoneal injection starting 24 h before the LPS nebulization. BrdU concentration was maintained in their drinking water throughout the experiment. B Flow cytometry analysis of moMs, tnsIMs and matIMs populations, co-stained with anti-BrdU (FITC) and anti-Ki67 (PE) antibodies, in WT and Ezr-KO^m mice treated with LPS. Bar graphs represent their corresponding percentages of BrdU⁺ Ki67⁺ cells. Data are represented as mean ± SEM from two independent experiments with three or more mice per genotype per experiment. Each dot represents a biological replicate. Statistical analysis was performed using one-way ANOVA or Tukey’s test for multiple comparisons between the genotype and treatment conditions. *p < 0.05, **p < 0.01 and ns non significant. See also Supplementary Fig. S7.

Fig. 6. Ezrin is required for efficient monocyte/MΦ filipodia formation and cell spreading during activation with LPS.

A Primary BM monocytes were isolated from murine WT and Ezr-KO^m and treated with LPS for 6 h on a collagen-coated surface. Representative immunofluorescence (IF) showing murine WT and Ezr-KO^m primary BM monocytes, untreated (A, E) or treated with LPS for 6 h (B–D and F–H). DAPI represents nucleus, Green represents actin and Red represents Ezrin staining. Scale bar = 2 μm. B Quantification of the area of the cells and plasma membrane (PM) F-actin intensity in murine WT and Ezr-KO^m, treated with LPS. C Representative SEM images of murine WT and Ezr-KO^m primary BM monocytes, cultured on 3D type I collagen hydrogel and untreated or treated with LPS. Scale bar = 5 μm. D, E Quantification of the volume (D) and area (E) of primary BM monocytes in both genotypes, of images. Data are represented as mean ± SEM from three independent experiments. Each dot represents an individual cell. Statistical analysis was performed using one-way ANOVA or Tukey’s test for multiple comparisons between the genotype and treatment conditions. **p < 0.01 and ****p < 0.0001. See also Supplementary Fig. S8.

Fig. 7. Ezrin is required for FAK and AKT signaling in monocytes leading to cell adhesion to ECM and survival in response to LPS.

A Schematic representation of ezrin signaling. Ezrin links plasma membrane (PM) and filamentous actin (F-actin) and adheres to the extracellular matrix (e.g., collagen) via integrin α/β. The activation of focal adhesion kinase leads to efficient cell spreading and adhesion, and PI3K/AKT signaling activation is required for cell survival. B Quantification of crystal violet absorbance emitted from adherent primary BM monocytes of murine WT and Ezr-KO^m untreated or treated with LPS (6 h). Absorbance values are relative to WT Untr. C Representative WB of phospho-FAK (pFAK, tyrosine 397), total FAK, phospho-AKT (pAKT, serine 473) and total AKT and densitometric analysis of pFAK and pAKT in primary BM monocytes of murine WT and Ezr-KO^m untreated or treated with LPS (6 h). Bar graphs represent pFAK/FAK and pAKT/AKT ratios. Protein fold increase was normalized to β-actin and relative to untreated cells. D Quantification of the percentage of dead cells in primary BM monocytes of murine WT and Ezr-KO^m, post-exposure to LPS. See also Fig. S8C. E Relative fluorescence units measuring caspase3/7 activity in primary BM monocytes of murine WT and Ezr-KO^m, untreated or treated with LPS at different time points. Data are represented as mean ± SEM from three independent experiments. Statistical analysis was performed using one-way ANOVA or Tukey’s test for multiple comparisons between the genotype and treatment conditions. *p < 0.05, **p < 0.01 and ****p < 0.001. See also Supplementary Fig. S9.