Macrophage response to fibrin structure mediated by Tgm2-dependent mitochondrial mechanosensing

Abstract

Following an injury at the implantation position, blood-material interactions form a fibrin architecture, which serves as the initial activator of foreign body response (FBR). However, there is limited knowledge regarding how the topography of fibrin architectures regulates macrophage behavior in mitigating FBR. Mechanical cues of the microenvironment have been reported to shape immune cell functions. Here, we investigated macrophage mechanobiology at the organelle level by constructing heterogeneous fibrin networks. Based on findings in vivo, we demonstrated that adhesion-mediated differentiation of mitochondrial function modulated macrophage polarization. The finite activation of integrin signaling upregulated transglutaminase 2 (Tgm2) in a trans-manner, augments PGC1α-mediated mitochondrial biogenesis. Our study highlighted the previously overlooked spatial structures of host proteins adsorbed on material surfaces, advocating for a paradigm shift in material design strategies, from focusing solely on physical properties to considering the modification of host proteins.

Keywords: Biomaterial recognition; Foreign body response; Mechanotransduction; Mitochondrion; Transglutaminase.

Graphical abstract

Fig. 1

Macroporous hydrogels with varying protein absorption capacities resulted in distinct fibrin network structures. (A) Synthesis process of MSA-ZW and MSA-DA. (B) The fibrinogen content reflecting fibrinogen absorption MSA-ZW and MSA-DA (n = 3, two-tailed Student's t-test, ∗P < 0.05). (C) Schematic diagram of fibrin network formation. MSA-ZW and MAS-DA were incubated with fibrinogen and thrombin, leading to the cleavage and subsequent rearrangement of fibrinogen into fibrin networks. (D–F) Representative images of SEM images and quantification of fibrin network. (E) MSA-ZW and MSA-DA without fibrin network were labeled as “blank.” Scale bar, 50 μm. The presence of fibrin networks on MSA-ZW and MSA-DA was labeled as “+Fib.” Scale bar, 5 μm. (E) Quantification of the porosity (n = 3, two-tailed Student's t-test, ∗∗P < 0.01). (F) Quantification of the branch thicknesses (n = 3, two-tailed Student's t-test, ∗P < 0.05). (G) Compression Curve of macroporou hydrogels with or without fibrin network. (H) Elastic modulus of macroporou hydrogels with or without fibrin network (n = 3, two-tailed Student's t-test, ns, no significance).

Fig. 2

Foreign body response (FBR) was affected by distinct fibrin network structures in macroporous hydrogels.(A) Schematic diagram of implant surgery of MSA-ZW and MSA-DA implanted in the femoral defect and quadriceps muscle. The experimental timeline for surgery and detection. (B) Representative images of micro-CT images of implantation site of bone at 1, 2, and 4 weeks. The red circle indicated the primary defect range. Scale bar, 5 mm. (C) Quantification of femoral defect size (n = 3, two-tailed Student's t-test, ∗P < 0.05). (D) Representative images of hematoxylin and eosin (HE) staining of muscle tissue at 1, 2, and 4 weeks post-treatment are presented. The blue outline demarcates the residual material, whereas the orange line delineates the edge of the fibrous capsule. (E) Representative images of Masson's Trichrome staining of muscle at 1, 2, and 4 weeks. The black arrows indicated the collagen fibers.

Fig. 3

Macrophages responded differently at the material-tissue interface in vivo. (A) Schematic diagram of tissue flow cytometry. The tissues containing MSA-ZW and MSA-DA were digested into single cells. The cell clusters were labeled with fluorescent antibodies and the samples were loading to perform flow cytometry. The results were demonstrated with dot plot. (B–E) Flow cytometry results of tissue flow cytometry. (B) Representative dot plots of CD80⁺ CD206^- and CD80⁻CD206⁺ cells in femoral defect tissue by flow cytometry. (C) Representative dot plots of CD80⁺ CD206^- and CD80⁻CD206⁺ cells in muscle tissue by flow cytometry. (D) Quantification of CD80⁺ CD206^- and CD80⁻CD206⁺ cells in femoral defect tissue (n = 3, two-tailed Student's t-test, ∗P < 0.05, ∗∗∗P < 0.001). (E) Quantification of CD80⁺ CD206^- and CD80⁻CD206⁺ cells in muscle tissue (n = 3, two-tailed Student's t-test, ∗P < 0.05, ∗∗∗∗P < 0.0001). (. (F) Representative images of immunostaining of CD206 (green), F4/80 (red), DAPI (blue) in muscle and femoral defect. The yellow frame indicate the residual material. (G) Quantification of CD206 mean grey value. (Femoral defect: n = 10, two-tailed Student's t-test, ∗∗∗P < 0.001, muscle: n = 7, two-tailed Student's t-test, ∗P < 0.05). (H) Representative images of immunostaining of iNOS (green), F4/80 (red), DAPI (blue) in muscle and femoral defect. The yellow frame indicate the residual material (I) Quantification of iNOS mean grey value. (Femoral defect: n = 8, two-tailed Student's t-test, ∗∗∗P < 0.001, muscle: n = 6, two-tailed Student's t-test, ∗P < 0.05).

Fig. 4

Macrophages exhibited distinct responses within the fibrin architectures generated by MSA-ZW and MSA-DAin vitro.(A) Schematic diagram of RAW264.7 culture strategies: The disinfection of macroporous hydrogels (MSA-ZW and MSA-DA) was achieved using 75 % ethanol and UV light. Subsequently, the macroporous hydrogels were incubated with a solution containing 4 mg/mL Fibrinogen and 5U/mL Thrombin at 37 °C for 5 min. RAW264.7 cells of M0 phenotype were then seeded onto the macroporous hydrogels coated with fibrin networks. The subsequent assessment involved determining the percentage of CD206⁺ phenotype macrophages (M2) and CD80⁺ phenotype macrophages (M1). To induce polarization, macrophages (M0) were treated with LPS, transforming them into macrophages (M1). These macrophages (M1) were subsequently seeded onto the macroporous hydrogels with fibrin networks, and the CD80⁺ phenotype percentage of macrophages (M1) was measured. (G–H) RAW264.7 cells were seeded on MSA-ZW, MSA-DA, and cell culture well plate (ctrl). (B) Representative flow cytometry analysis and quantification of CD206+ phenotype percentage after seeding macrophages (M0) for 24h (n = 3, one‐way ANOVA with Tukey's post‐test, ∗∗∗∗P < 0.0001). (C) Representative flow cytometry analysis and quantification of CD80⁺ phenotype percentage (n = 3, one‐way ANOVA with Tukey's post‐test, ∗P < 0.05, ∗∗P < 0.01). (D–F) RAW264.7 cells were pretreated with LPS before seeding on the MSA-ZW and MSA-DA. The measurements were taken after culturing for 24 h. (I) Representative images of immunostaining of iNOS (red) and DAPI (blue). (J) Representative images of SEM images. (K) Representative images of immunostaining of phalloidin (red) and DAPI(blue).

Fig. 5

The adhesive response of macrophages influenced mitochondrial function and TGM2 expression. RAW264.7 cells were pretreated with LPS before seeding on the MSA-ZW and MSA-DA. The measurements were taken after culturing for 24 h. (A) Gene expression analysis of integrin-associated genes CD11b and CD18 (n = 4, two-tailed Student's t-test, ∗P < 0.05, ∗∗∗∗P < 0.0001). (B) Protein expression measured by western blot and quantification of p-FAK, FAK, and CD18 (n = 4–5, two-tailed Student's t-test, ∗P < 0.05, ∗∗P < 0.01, ∗∗∗∗P < 0.0001, ns, no significant difference). (C) Representative images and quantification of mitoSOX (n = 5, two-tailed Student's t-test, ∗P < 0.05). (D) Representative images and quantification of TMRM (n = 8, two-tailed Student's t-test, ∗P < 0.05). (E) Representative histogram and quantification of DCFH-DA (n = 3, two-tailed Student's t-test, ∗P < 0.05). (F) Representative histogram and quantification of TMRM (n = 3, two-tailed Student's t-test, ∗∗P < 0.01). (G) ATP content (n = 4, two-tailed Student's t-test, ∗∗P < 0.01). (H-J) RNA-seq analysis of LPS-pretreated RAM264.7 cells cultured on MSA-ZW and MSA-DA for 24h. (H) The volcano plot illustrates the differentially expressed genes, with Tgm2 identified as one of them. (I) Heat map of gene expression. (J) Biological process (BP), molecular function (MF), and cellular component (CC) pathways of gene ontology (GO) enrichment analysis.

Fig. 6

Tgm2 built connections between fibrin architecture and mitochondria.(A) The protein expression of FAK, pFAK, Tgm2, and β-actin when LPS-pretreated RAW264.7 cells were cultured with blank control (Blank), Tgm2 inhibitor (500 μM Cystamine), and pFAK inhibitor (1 mM Defactinib) for 24h. (B) Quantification of protein expression levels of pFAK/FAK and Tgm2 (n = 3, two-tailed Student's t-test, ∗P < 0.05, ∗∗P < 0.01). (C) Schematic diagram illustrated the relationship between FAK, pFAK, and Tgm2: Upon formation of a dense fibrin network with thick fibers by MSA-DA, the level of pFAK was upregulated, and the Tgm2 decreased. Conversely, when a coarse fibrin network with thin fibers was formed by MSA-ZW, the pFAK level decreased, while the Tgm2 level was upregulated. (D–F) LPS-pretreated RAM264.7 cells were cultured on MSA-ZW, and MSA-ZW with addition of Cystamine (MSA-ZW + Cystamine) for 24h. (D) Representative images and quantification of mitoSOX (n = 5, two-tailed Student's t-test, ∗∗∗P < 0.001). (E) Representative images and quantification of TMRM (n = 8, two-tailed Student's t-test, ∗∗∗P < 0.001). (F) ATP content (n = 3, two-tailed Student's t-test, ∗∗P < 0.01). (G–I) RAW264.7 cells were transfected with non-silencing siRNA as negative control (siNC) and siRNA silencing Tgm2 (siTgm2). (G) Representative Tgm2 protein expression of siNC and siTgm2. (H) Gene expression heat map of RNA-seq analysis of LPS-pretreated RAM264.7 cells cultured on MSA-ZW for 24h. The items related to mitochondria were highlighted in red. (I) GO pathway enrichment.

Fig. 7

The knock-down of Tgm2 suppressed PGC1α-mediated mitochondrial biogenesis. RAW264.7 cells were transfected with siNC and siTgm2 and then LPS-pretreated siNC and siTgm2 were seeded on MSA-ZW. (A) Representative images of staining of mitoSOX (red) and DAPI (blue). (B) Quantification of mitoSOX mean grey value (n = 3 individual samples, two-tailed Student's t-test, ∗P < 0.05). (C) Representative flow cytometry analysis and quantification the media fluorescence index (MFI) of DCFH-DA (n = 5, two-tailed Student's t-test, ∗∗P < 0.01). (D) Representative images of staining of TMRM (red) and DAPI (blue). (E) Quantification of TMRM mean grey value (n = 3 individual sample, two-tailed Student's t-test, ∗P < 0.05). (F) Representative flow cytometry analysis and quantification the MFI of TMRM (n = 3, two-tailed Student's t-test, ∗P < 0.05). (G) ATP content (n = 6, two-tailed Student's t-test, ∗∗P < 0.01). (H-I) Seahorse XF mito stress test assay. (H) The oxygen consumption rate (OCR) of RAW264.7 (n = 5). (I) Analysis of the basal respiration, ATP production, maximal respiration, and spare respiratory capacity (n = 5, two-tailed Student's t-test, ∗∗∗P < 0.001, ∗∗∗∗P < 0.0001). (J) Representative images of staining of Mito-Tracker deep red (green) and DAPI (blue) show the mitochondrial network morphology. The yellow arrows indicate the aggregated mitochondrial clusters. (K) Relative mtDNA levels show the duplication of mtDNA (n = 3, two-tailed Student's t-test, ∗P < 0.05). (M) The protein expression of PGC1α and TFAM. (N) Quantification of protein expression of PGC1α and TFAM (n = 3–4, two-tailed Student's t-test, ∗∗P < 0.001, ∗∗∗P < 0.0001).

Scheme 1

Constructing heterogeneous fibrin architectures deciphers macrophage mitochondrial mechanosensing to thwart the FBR. Fibrin architectures formed on macroporous hydrogels define the mechanical microenvironments of macrophages and modulate FBR. Differential activation of integrins, induced by the conformational variations in fibrin networks, regulates macrophage mitochondrial mechanosensing in a Tgm2-dependent manner.