Primed mesenchymal stem cells attenuate schistosomiasis fibrosis by enhancing macrophage subset switching and efferocytosis via Itgb2-Rac1 axis
Junxia Lei, Yaojia Ren, Zebin Chen, Wanxian Huang, Meiyining Xu, Zhongdao Wu
TL;DR
This study shows that primed mesenchymal stem cells reduce liver fibrosis in schistosomiasis by improving macrophage function through a specific molecular pathway.
Contribution
The study identifies the Itgb2-Rac1 axis as a novel mechanism through which primed MSCs enhance macrophage efferocytosis and subset switching to reduce fibrosis.
Findings
Primed MSCs promote conversion of macrophage subsets to pro-resolving types in schistosomiasis.
Primed MSCs enhance macrophage efferocytosis via the Itgb2-Rac1 axis.
Blocking Itgb2 or Rac1 reverses the anti-fibrotic effects of primed MSCs.
Abstract
The pathological hallmark of schistosomiasis is egg-induced granulomatous inflammation and the subsequent fibrosis in the liver; and mesenchymal stem cells have been explored for the treatment of schistosomiasis fibrosis. However, the underlying mechanisms are largely unknown. We demonstrated here that IFN-γ/ LPS-primed MSCs alleviate schistosomiasis fibrosis through promoting conversion of Ly6ChiCX3CR1lo to pro-resolving Ly6CloCX3CR1hi. Importantly, primed MSCs promoted macrophage efferocytosis in the infected mice, and in vitro experiments confirmed the direct role of primed MSC on enhancement of macrophage efferocytosis, as well as its conversion to pro-resolving type. Mechanistically, primed MSC promoted β2 integrins(Itgb2) expression within macrophages; and Itgb2 blockade not only inhibited GTPase Rac1 activity, a key regulator of actin filament organization during efferocytosis,…
Genes, proteins, chemicals, diseases, species, mutations and cell lines named across the full text — each resolved to its canonical identifier and authoritative record.
Click any figure to enlarge with its caption.
Figure 1
Figure 2
Figure 3
Figure 4
Figure 5
Figure 6
Figure 7
Figure 8
Figure 9- —National Natural Science Foundation of China (General Program; Grant number: 82372277)
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
Taxonomy
TopicsImmune cells in cancer · Parasites and Host Interactions · Phagocytosis and Immune Regulation
Introduction
Schistosomiasis is one of the world’s major public health problems in terms of morbidity and mortality [1]. Although praziquantel (PZQ) has been widely used to kill or damage the adult worms, the development of effective therapies for the regression of egg-induced granuloma and fibrosis remains a challenge [2]. Considering the similarities of immunopathological characteristics between human and mice schistosomiasis, murine schistosomiasis model has been extensively used for various aspects of human schistosomiasis [3]. With deposition of eggs in the liver 4-6 weeks after infection, a local Th2-driven granulomas inflammation is induced, a collection of inflammatory cells including eosinophils, neutrophils, lymphocytes and macrophages [1, 4–6], among which, upon infection-associated removal of KCs, recruited bone marrow monocyte-derived macrophages (MoMϕs) play a key role through connecting with inflammatory cells and hepatic stellate cells (HSCs), and as a result, with excessive accumulation of extracellular matrix (ECM), fibrosis develops nearly at 7-8 weeks postinfection [7, 8].
Based on cell surface glycoprotein Ly-6C expression, at least two subsets of MoMϕ are defined in mice, Ly6C^hi^ and Ly6C^lo^ subsets. It is generally considered that Ly6C^hi^ MoMϕs have a proinflammatory phenotype (similar to classical M1 phenotype), which can aggravate liver injury and fibrosis by releasing proinflammatory factors and fibrogenic cytokines, such as TNF-α, IL-1, CCL2, CCL5, TGF-β1, PDGF, while Ly6C^lo^ macrophages promote inflammation or fibrosis resolution by secreting anti-inflammatory factors or matrix metalloproteinases (MMPs) and inducing HSC apoptosis [9–11].
In murine schistosomasis model, bone marrow derived Ly6C^hi^ monocytes can be robustly recruited and responsible for Th2 associated host-protective macrophage accumulation [12, 13]; Additionally, it is reported that inducing pro-resolving phenotype (Ly6C^lo^ MoMϕs) through overexpression of miR-130a-3p in mice infected with S. japonicum promoted regression of fibrosis through decreasing tissue inhibitor of metalloproteinases (TIMP) 1 and increasing matrix metalloproteinase (MMP) 2 [14]. Thus, enhancement of phenotypic switching from Ly6C^hi^ to Ly6C^lo^ may confer anti-fibrotic benefits for liver schistosomiasis.
Recently, the accumulating evidence shows that MSCs has become a promising cyto-therapy for treating a variety of liver fibrosis [15–25]. There have been relatively consistent results showing that MSCs may alleviate hepatic fibrosis through promoting phenotypic conversion from Ly6C^hi^ toward pro-resolving Ly6C^lo^ macrophages in CCl4 [15, 23, 24] and MCD diet fibrosis model [25]; however, research on schistosomiasis fibrosis treatment with MSCs are still limited. A few reports have indicated that MSCs exhibited preliminary anti-fibrotic efficacy regarding the histopathological evaluation, yet it remains to be verified whether MSCs exert anti-fibrotic effects by regulating Ly6C^hi^ /Ly6C^lo^ phenotype conversion [19, 26–29]. More importantly, the significant pathological features dominated by the Th2 or M2 response in the schistosomiasis fibrosis stage renders this issue of particular interest [30, 31].
Our previous research have shown that toll like receptor (TLR)4 agonist and IFN-γprimed MSCs alleviated hepatic fibrosis through modulating Th1/Th2 responses [32], here we further investigate whether and how primed MSCs play anti-fibrotic role through regulation of Ly6C^hi^ /Ly6C^lo^ phenotype during chronic schistosomiasis, which will provide new evidence for MSC therapy of chronic inflammatory fibrosis.
Results
IFN-γ + LPS-primed MSCs alleviated liver pathology
In our previous study, we showed that TLR4 agonist +IFN-γ primed MSCs (their identification results showed in Fig. 1s) alleviated liver pathology [32]. In this work, we confirmed the therapeutic potential of IFN-γ + LPS-primed MSCs in Schistosoma infection model again; at the end of the 2-week observation period after MSCs therapy, the primed MSCs group exhibited a liver with a grossly more reddish appearance and softer texture (Fig. 1A) and showed higher body weight as well, compared with none-treated control or unprimed MSCs group, respectively (Fig. 1B); and more significantly, the primed MSCs group showed alleviation in liver fibrosis at 1 week posttreatment, manifested by lower percentage of egg granulomas area and fibrosis area, compared with unprimed MSCs group, and even more clear improvement in liver fibrosis at 2 weeks posttreatment, compared with both non-treated control and unprimed MSCs group (Fig. 1C, D). Accordingly, collagen type III (COL-3) levels [33] in primed MSCs group significantly decreased both at 1 week and 2 weeks after treatment, compared with none-treated control or unprimed MSCs group, respectively, and similarly decreased for α-SMA (a key activation marker for HSC)at 1 week after treatment (Fig. 1E). Thus, only primed MSCs but not unprimed MSCs showed the therapeutic potential for schistosomiasis fibrosis.Fig. 1IFN-γ + LPS-primed MSCs alleviated liver pathology in mice infected with *Schistosoma japonicum.*Mice were infected with 16 ± 2 cercariae and treated with praziquantel (PZQ) at 6 weeks post-infection, followed by treatment with MSCs and then sacrificed at 1, 2 weeks after MSCs treatment, respectively. At 2 weeks after treatment, gross liver appearance was examined (A), and weekly body weight measurements were taken and compared (B). Histological change was evaluated. Representative granulomatous was showed by HE staining and by Masson staining (C). The percentage of egg granulomas area and fibrosis area were compared among different groups (D). Fibrosis-related genes Col3a1 and Acta2 were assessed by semi-quantitative PCR, normalized to GAPDH (E). Data are presented as the means ± SEM (n = 4–5). *P < 0.05, **P < 0.01, ****P < 0.0001 and n.s. not significant. Comparing each time among PBS, none primed MSCs and primed MSCs group.
Since the fibrosis development have been generally considered to be related to the phenotypes of macrophages; and two phenotypes of bone marrow/monocyte-derived macrophages (MoMϕs), either pro-inflammatory Ly6C^high^ or pro-restorative Ly6C^low^ MoMϕs have been described to play key role in the granulomatous inflammation of schistosomiasis [11, 34], we next focused on the alteration of phenotypic switching of Ly6C^high^/Ly6C^low^ over time following treatment with MSCs.
IFN-γ + LPS-primed MSCs led to lower Ly6ChiCX3CR1lo and higher Ly6CloCX3CR1hi levels in mice infected with Schistosoma japonicum
Two kinds of bone marrow-derived MoMϕs, named as Ly6C^hi^CX3CR1^lo^ and Ly6C^lo^CX3CR1^hi^ phenotypes in the liver were detected by FACS at 1 week and 2 weeks after MSCs treatment, respectively. First, non-MSCs control showed higher Ly6C^hi^CX3CR1^lo^ (compared with normal group) at the first observation timepoint, followed by comparable Ly6C^hi^CX3CR1^lo^ and significantly lower Ly6C^lo^CX3CR1^hi^ levels (compared with normal group) at the second observation timepoint (Fig. 2A, B). The primed MSCs group exhibited similar Ly6C^hi^CX3CR1^lo^ levels compared with non-MSCs control (despite a downward trend) but lower Ly6C^hi^CX3CR1^lo^ levels compared with unprimed MSCs group, and higher Ly6C^lo^CX3CR1^hi^ levels (compared with both non-MSCs control and unprimed MSCs group) one week after treatment, with higher Ly6C^lo^CX3CR1^hi^ persisting into week 2(compared with non-MSCs control). In contrast, the unprimed MSCs group showed persistently higher Ly6C^hi^CX3CR1^lo^ levels (compared with untreated control) both in weeks 1 and 2 after treatment, in spite of a higher Ly6C^lo^CX3CR1^hi^ levels (compared with untreated control) in week 2 (Fig. 2A, B). Thus, primed MSCs exhibited therapeutic effects, accompanied by a downward trend in Ly6C^hi^CX3CR1^lo^ and a persistent increase in Ly6C^lo^CX3CR1^hi^ levels during observation period, whereas non-primed MSCs showed no significant therapeutic effect, accompanied by a persistent increase in Ly6C^hi^CX3CR1^lo^, in spite of a higher Ly6C^lo^CX3CR1^hi^ levels at later stage of treatment.Fig. 2IFN-γ + LPS primed MSCs led to lower Ly6C^hi^CX3CR1^lo^ and higher Ly6C^lo^CX3CR1^hi^ levels in infected mice.Representative percentages of Ly6C^hi^CX3CR1^lo^ subset and Ly6C^lo^CX3CR1^hi^ subset as a proportion of MoMϕs (gated on CD11b^+^ F4/80^int^), analyzed by flow cytometry (A) and quantification of each subset as a proportion of MoMϕs (B) demonstrating the dynamics alteration of hepatic Ly6C^hi^ and Ly6C^lo^ in week1, 2 after MSCs treatment in infected mice, respectively. Data are presented as the means ± SEM, n = 4–5. *P < 0.05, **P < 0.01, ****P < 0.0001 and n.s. not significant. Comparing each time among PBS, none primed MSCs and primed MSCs group.
Since Ly6C^hi^ MoMϕ or Ly6C^lo^ MoMϕs have been described as pro-inflammatory or pro-resolving phenotype [6, 8], to further disclose their functional characteristics, representative pro-inflammatory or anti-inflammatory factors of macrophages were further detected in infected mice.
IFN-γ + LPS -primed MSCs induced lower levels of inflammatory factors and a moderate pro-resolving profile in the liver of infected mice
The key pro-inflammatory factors (TNF-α, IL-1β and iNOS) and inflammatory chemokines(CCL2, CX3CL1,CXCL1) were detected using the semi-quantitative PCR method, and so were anti-inflammatory factors (IL-10 and Arg-1). The primed MSCs group showed the lowest levels of TNF-α, IL-1βand iNOS among all the infected groups, as well as the lowest levels of chemokines (CCL2, CX3CL1,CXCL1) in week 1 after MSCs treatment, and continuously maintained lower levels of iNOS and CX3CL1 (compared with non-MSCs control) in week 2. By contrast, non-primed MSCs group showed either higher levels of IL-1βand iNOS or comparable levels of CCL2, CX3CL1 and CXCL1 to those in non-MSCs control in week 1 after MSCs treatment, followed by persistently higher levels of IL-1βand CCL2 (compared with non-MSC control)in week 2 (Fig. 3).Fig. 3IFN-γ + LPS-primed MSCs induced lower levels of inflammatory factors and a moderate pro-resolving profile in the liver of infected mice.Fifty to 100 mg of liver tissue was removed from infected mice at certain timepoints, and the expression of Nos2, Il1b, Tnf, Ccl2, Cx3cl1, Cxcl1, Arg1 and Il10 mRNA were assessed by semi-quantitative PCR analysis, normalized to GAPDH. Data are presented as the means ± SEM, n = 4-5. P < 0.001P* < 0.01, *P < 0.05.
As for anti-inflammatory or pro-resolving factors, consistent with most reports, control group exhibited pathological features dominated by a TH2 or M2 response, with persistently higher Arg-1 levels at both observation timepoints; primed MSCs group showed comparable Arg-1 levels to those in non-MSCs control in week 1 after MSCs treatment, followed by lower Arg-1 levels in week 2; by contrast, non-primed MSCs group showed persistently lower Arg-1 both in week1,2. Additionally, primed MSCs group moderately increasing levels of IL-10 among all the infected groups, whereas, non-primed MSCs group showed the highest IL-10 levels among all groups in week 1 (Fig. 3).
Collectively, primed MSC treatment led to reductions in pro-inflammatory cytokines, chemokines, and iNOS, while maintaining similarly high levels of Arg-1 as the control group at 1 week posttreatment, followed by lower levels of iNOS and CX3CL1 at 2 weeks; in contrast, treatment with non-primed MSCs increased pro-inflammatory cytokines and iNOS while decreasing Arg-1, followed by persistently higher levels of IL-1βand CCL2 at 2 weeks. This finding was just in accord with the results of macrophage subset analysis, in which primed MSCs induced a persistent increase in Ly6C^lo^CX3CR1^hi^ levels without increase of Ly6C^hi^CX3CR1^lo^, whereas non-primed MSCs led to a persistent increase in Ly6C^hi^CX3CR1^lo^, in spite of a higher Ly6C^lo^CX3CR1^hi^ levels at later stage of treatment.
To further confirm whether primed MSCs induced increase of Ly6C^lo^CX3CR1^hi^ levels through promoting Ly6C^hi^/Ly6C^lo^ conversion, we blocked recruitment of Ly6C^hi^ monocytes/macrophages using CCR2 antagonist before MSCs treatment, then identified Ly6C^lo^ MoMϕs percentage at 2 weeks posttreatment.
IFN-γ + LPS -primed MSCs ameliorate schistosomiasis fibrosis through promoting conversion of Ly6Chi/ Ly6Clo MoMϕs
We used RS102895 to block CCR2-dependent recruitment of Ly6C^hi^ monocytes/macrophages at the early stage of treatment with primed MSCs and further investigated subtypes switching of MoMϕs at 2 weeks posttreatment. As expected, the primed MSCs group exhibited higher levels of Ly6C^lo^ MoMϕs and the same level of Ly6C^hi^ MoMϕs, compared to the control group, while blocking CCR2 led to decrease of both Ly6C^hi^ MoMϕs and Ly6C^lo^ MoMϕs at 2 weeks posttreatment in infected mice (Fig. 4C, D), with a corresponding deteriorated liver pathology (Fig. 4A, B). This blocking assay confirmed that IFN-γ + LPS-primed MSCs enhanced Ly6C^lo^ subtypes dependent on recruitment of Ly6C^hi^ monocytes/macrophages at earlier stage post-treatment, and in other words, primed MSCs promoted conversion of Ly6C^hi^/ Ly6C^lo^ MoMϕs.Fig. 4RS102895 treatment reduced Ly6C^lo^ subtypes at 2 weeks posttreatment and consequently deteriorated liver pathology in infected mice.Histological change was evaluated at 2 weeks after treatment in infected mice, which were injected with PBS, primed MSCs or activated MSC + RS102895, respectively. Representative granulomatous was shown by HE staining and by Masson staining (A). The percentage of fibrosis area were measured by computer-assisted morphometric analysis (B). Representative percentages of Ly6C^hi^CX3CR1^lo^ subset and Ly6C^lo^CX3CR1^hi^ subset as a proportion of MoMϕs, analyzed by flow cytometry (C) and quantification of each subset as a proportion of MoMϕs (D). Data are presented as mean ± SEM; n = 3–4 mice for each group. (*P < 0.05, **P < 0.01 and n.s. not significant).
Next, we ask how MSCs promote switching of Ly6C^hi^/Ly6C^lo^ subtypes in schistosomiasis model. It has been reported that cytokines in the microenvironment (such as IL-4 and IL-10) can promote macrophage phenotype switching [15, 35–37], but this is clearly not the case in our experiment, where the primed MSCs group showed positive therapeutic effects with moderate increase of IL-10, while non-primed MSCs group showed no anti-fibrotic efficacy despite inducing the highest IL-10 levels. Recently, more and more studies have noted that the synergy between macrophages and neutrophils affects post-liver injury repair; and the phagocytosis of dying neutrophils by macrophages, known as efferocytosis, served not only as clearance of the apoptotic cells [38, 39] but also promoted a pro-resolving phenotype (Ly6C^low^ MoMϕs) in liver injury model [40–42]. Because of the activation and recruitment of macrophages and neutrophils within the schistosomiasis granulomas, leading to amplified inflammatory response, we suspect there is a possibility that primed MSCs promote macrophage subtype switching through efferocytosis of apoptotic neutrophils. Therefore, the efferocytosis-mediated mechanism among different groups was investigated.
IFN-γ + LPS-primed MSCs promoted the phagocytosis of neutrophils by macrophages in infected mice
Since TAM proteins Mertk is one of the most universally expressed receptors in the phagocytic macrophages to mediate apoptotic neutrophils recognition [43], double staining of Mertk and neutrophils marker Ly6G are used to show engulfment of apoptotic neutrophils by macrophages. As our expectation, primed MSCs group showed significant higher engulfment ratio of neutrophils both at 1 week and 2 weeks posttreatment (compared with control group or non-primed MSCs group, respectively), which is exactly consistent with the higher Ly6C^lo^ macrophage in primed MSCs group (Fig. 5A–C). Thus, primed MSCs promote macrophage to phagocytose neutrophils in vivo, suggesting a correlation between macrophage subtype switching and efferocytosis. To further investigate the direct role of primed MSCs on efferocytosis and macrophage subtype conversion, in vitro co-culture of macrophages and apoptotic neutrophils was performed.Fig. 5IFN-γ + LPS -primed MSCs promoted macrophage to phagocytose neutrophils in the liver of infected mice.Mice were infected with 16 ± 2 cercariae and treated with praziquantel (PZQ) at 6 weeks post-infection, followed by treatment with MSCs. Efferocytosis was examined by dual immunofluorescence for MerTK (red) and Ly6G (green) in liver sections. Representative images of macrophage phagocytosis at 1 week (A), 2 weeks post-treatment (B), respectively (Arrows indicate phagocytosing macrophages with neutrophils inside). Scale bars, 100 µm and 400 µm, respectively. The number of efferocytosis cells per field (C). n = 4-5 per group. Data are presented as mean ± SEM (*P < 0.05, ***P < 0.001, ****P <0.0001).
IFN-γ + LPS-primed MSCs directly promoted macrophage efferocytosis, accompanied by decreased inflammatory cytokines and increased pro-resolving factors in vitro
To investigate the direct role of primed MSCs on engulfment of apoptotic neutrophils, we first incubated bone marrow derived macrophage with non-primed MSCs or primed MSCs, respectively, and then added Dil-labeled apoptotic neutrophils (induced by staurosporine). As expected, primed MSCs significantly increased macrophage phagocytic efficiency (compared with control group or non-activated MSC group, respectively) (Fig. 6A1, A2). Meanwhile, primed MSCs led to a decrease in pro-inflammatory cytokine mRNA Il1b, Tnf and Cx3cl1, and an increase in pro-resolving factors mRNA including Cx3cr1 (a key marker of pro-resolving macrophage), Il10 and Mmp2 in macrophages (Fig. 6B). In addition, protein detection in the co-culture system showed consistent results, lower TNF and higher IL10 level in the activated MSC group, compared with control group. However, non-activated MSC group showed comparable TNF levels to those of the control group, but higher IL-10 levels (Fig. 6C). It has been reported that the Ly6C^hi^/Ly6C^lo^ subset switch not only blocked the source of inflammatory factors, but also triggered pro-resolving factors IL-10 and MMPs production for inflammation or fibrosis resolution [15, 16]. Taken together, our results showed that IFN-γ + LPS-activated MSC directly promoted macrophage efferocytosis, as well as its subtypes switching. Next, the mechanisms underpinning efferocytosis were further investigated.Fig. 6IFN-γ + LPS -activated MSC promoted efferocytosis of macrophages, accompanied by decreased inflammatory cytokines and increased pro-resolving factors.Prior to the start of efferocytosis, MoMϕs from infected mice were first incubated with MSCs or primed MSCs, respectively, and then added Dil-labeled apoptotic neutrophils (induced by staurosporine). Macrophages and culture supernatant were collected after co-incubation for 1 hour or 12 hours, respectively, and the efferocytosis of macrophages was detected by immunofluorescence for MerTK and Dil-labeled apoptotic neutrophils and observed under a fluorescence microscope. Representative images of efferocytosis cells (A1), Scale bars, 100 µm, and the percentage of macrophages containing at least one ingested apoptotic neutrophil or multiple apoptotic neutrophils, presented as rate of efferocytosis or rate of multiple efferocytosis, respectively (A2); the mRNA levels of Il1b, Tnf, Cx3cl1, Cx3cr1, Il10 and Mmp2 were detected by qPCR (B); the protein levels of IL10 and TNF-ɑ in the culture supernatant were detected by ELISA(C). n = 3; Data are presented as mean ± SEM (*P < 0.05, **P < 0.01 and n.s. not significant).
IFN-γ + LPS-primed MSCs promoted macrophage efferocytosis mediated by Itgb2 -Rac1 activated pathways
For triggering efferocytosis process, macrophages first need to find and recognize the apoptotic cells through interactions between their receptors and ligands exposed on apoptotic cells (such as phosphatidylserine, PS). Therefore, we first investigated the effects of MSCs on engulfment receptors that recognize PS [39, 44]. Our results showed that primed MSCs specially promoted Itgb2 expression, while non-primed MSCs reduced those levels (Fig. 7A); regarding other receptors, including TAM family Mertk, Axl and Gas6, BAI1, Timd4 and integrin Itgb5, there were no significant differences between control group and primed MSCs group (Fig. 2s).Fig. 7IFN-γ + LPS -primed MSCs promoted efferocytosis of macrophages via Itgb2 -Rac1 pathways.MoMϕs from infected mice were first incubated with MSCs or primed MSCs, respectively, and then added apoptotic neutrophils (or Dil-labeled apoptotic neutrophils). Macrophages were collected at different efferocytosis times, and the mRNA levels of Itgb2 and Rac1were detected by qPCR (A) after 12 h phagocytosis; prior to the start of phagocytosis, macrophages were incubated with CD18 neutralizing antibody(2.5 μg/ml) for 3 h to block Itgb2 signal pathway, and then the mRNA levels of Rac-1 were detected by qPCR after 12 h phagocytosis and activity of Rac1 in macrophages was determined by G-LISA after 1 h of phagocytosis (B); after blocking Itgb2 signal pathway, the efferocytosis efficiency of macrophages was detected by immunofluorescence for MerTK (green) and Dil-labeled apoptotic neutrophils (red), Representative images of efferocytosis cells (C1), Scale bars, 100 µm, and the percentage of macrophages engulfing apoptotic neutrophils (C2). n = 3; Data are presented as mean ± SEM (*P < 0.05, **P < 0.01 and n.s. not significant).
Following the ligand-receptor interaction, cytoskeletal rearrangement of macrophages can be triggered, a key step for the subsequent phagocytic activities, in which GTPase Rac1 plays an important role in the organization of actin filament networks and in membrane ruffling [39, 44]. Indeed, our results demonstrate that both mRNA levels of Rac1 and Rac1 activity of macrophages were significantly increased in the primed MSCs group, while decreased in the non-primed MSCs group (Fig. 7B); treatment of the primed MSCs group with Itgb2 blocking antibody significantly reduced both mRNA levels of Rac1 and Rac1 activity of macrophages (Fig. 7B). Furthermore, consistent with reduced Rac1 activity of macrophage in the blocking group, a significant decrease in the macrophage efferocytosis efficiency was observed in the blocking group as well (Fig. 7C1, C2).
Collectively, our results indicated that primed MSCs enhanced macrophage efferocytosis via Itgb2-Rac1 pathway, and combined with related reports that efferocytosis process can trigger macrophage switching from inflammatory subsets to anti-inflammatory or pre-resolving subsets [15, 16, 41], blocking experiment (blockade of Itgb2-Rac1 pathway) was performed to investigate whether primed MSCs promoted macrophage phenotypic switching via efferocytosis pathway.
Blocking key pathways of efferocytosis reversed the regulation of IFN-γ + LPS -primed MSCs on macrophage subset switching
Either CD18 blockade or inhibition Rac1 activation led to increased mRNA levels of Il1band Tnf, and decreased both Cx3cr1(key marker of reparative macrophages) and Mmp2/9 within macrophages. (Fig. 8). Thus, Itgb2-Rac1 pathway mediate effects of primed MSCs on macrophage function skewing. However, the regulation of IL-10 and TGF-βpresents a different scenario. Primed MSCs led to a decrease of Tgfb levels within macrophages compared with control group; CD18 blockade failed to increase Tgfb expression (in spite of an increasing trend), while Rac1 inhibitor NSC23766 could (Fig. 8). Most surprisingly, the neutralizing CD18 antibody failed to inhibit the effect of primed MSCs in promoting anti-inflammatory Il10 expression, while Rac1 inhibitor NSC23766 could (Fig. 8). Collectively, our data demonstrated that blocking key pathway Rac1 of efferocytosis reversed the regulation of IFN-γ + LPS -primed MSCs on macrophage subset switching, while Itgb2 blockade abolished most of the pro-reparative functional characteristics of macrophages, except for regulation of IL-10 expression by primed MSCs. The enhancement of IL-10 expression by primed MSCs may involve other pathways-triggered efferocytosis beyond Itgb2 activation.Fig. 8. Blocking key pathway of efferocytosis reversed the regulation of IFN-γ + LPS-primed MSCs on macrophage subset switching.MoMϕs from infected mice were first incubated with primed MSCs or without MSCs for 24 h (for blocking test, macrophages were still exposed to neutralizing antibody against CD18 or Rac1 inhibitor NSC23766 3 h prior to the start of efferocytosis), and then apoptotic neutrophils were added. Macrophages were collected after 12 h phagocytosis, and the mRNA levels of Rac1, IL-1β, Tnf-ɑ, TFG-β, Cx3cr1, Mmp2, Mmp9, and Il10 were detected by qPCR. n = 3; Data are presented as mean ± SEM (*P < 0.05, **P < 0.01, ***P < 0.001 and n.s. not significant).
Discussion
Macrophages are the key regulators of liver fibrosis and its resolution [34, 45, 46]; and an increasing number of studies have noted that recruited Ly6C^hi^ monocytes/macrophages and their descendants Ly6C^lo^ macrophage cooperate to influences the outcome of fibrosis in liver damage model [11, 40, 47, 48]. However, whether IFN-γ + LPS-primed MSCs alleviate schistosomiasis fibrosis through modulating Ly6C^hi^/Ly6C^lo^ subtypes conversion is waiting to be clarified. Therefore, based on our previous finding that toll like receptor (TLR)4 agonist and IFN-γ primed MSCs alleviated hepatic fibrosis through modulating Th1/Th2 responses [32], this study focuses on the regulation of primed MSCs on macrophage subsets, as well as the underpinning mechanisms, aiming to provide new clues for the precision application of MSCs in fibrosis treatment.
This study ensured the same positive effects of primed MSCs in improving fibrosis as our previous report [32], with a simultaneous increase in Ly6C^lo^CX3CR1^hi^ without increase of Ly6C^hi^CX3CR1^lo^ throughout the 2-week period after treatment. By contrast, non-primed MSC even aggravated fibrosis at 1 week post treatment with persistently higher levels of Ly6C^hi^CX3CR1^lo^ throughout the 2-week post-treatment period.
In accordance with the analysis of macrophage subtypes, activated MSC group exhibited a restorative profile manifested by reducing pro-inflammatory cytokines (such as iNOS, TNF-ɑ, L-1β, CCL2, CX3CL1 and CXCL1), and moderately increasing levels of IL-10 and maintaining similarly high levels of Arg-1 as the control group at acute stage, followed by lower levels of iNOS and CX3CL1 at 2 weeks posttreatment; whereas non-primed MSCs group exhibited the strongest inflammation state manifested by increasing pro-inflammatory cytokines iNOS, TNF-ɑ, L-1β, CCL2, CX3CL1 and CXCL1 and IL-10, and simultaneously maintaining the lowest levels of Arg-1 among different groups at early stage of treatment, followed by persistently higher levels of IL-1βand CCL2 at 2 weeks posttreatment. Taken together, it is suggested that primed MSCs exert the beneficial effect on schistosomiasis fibrosis through inducing pro-resolving Ly6C^lo^ subsets in a finely tuned manner. Furthermore, the finding that blocking recruitment of Ly6C^hi^ MoMϕ led to reduced percentage of Ly6C^lo^ MoMϕs, with a corresponding deteriorated liver pathology, confirmed that primed MSCs exhibited therapeutic effects through promoting conversion of Ly6C^hi^ to Ly6C^lo^ MoMϕs.
The main pathology of schistosomiasis is characterized by schistosome eggs induced Th2 or M2-rich granulomas, which marked by significant Arg1, IL-10 expression, etc [1, 4–6], though M2 macrophages is still considered to suppress Th2-dependent inflammation [49]. This raises confusion regarding the improvement of fibrosis through regulating macrophage subset conversion. In line with the definition of M2 polarization, this study demonstrated that primed MSCs induced an M2-dominant microenvironment at 1 week posttreatment, followed by the reduced Arg-1 level and lower inflammatory factors levels at 2 weeks. Therefore, primed MSCs did not induce a sustained M2 macrophage polarization state, but rather induced Ly6C^lo^ phenotypic switch to attenuate fibrosis during the 2-week observation period. Next, the mechanism underlying Ly6C^hi^ /Ly6C^lo^ subtypes conversion was further explored.
Paracrine mechanisms, such as IL-4, IL-13, IL-10, PGE2, TSG-6, etc [10, 15, 35–37], have been reported to mediate effects of MSCs on phenotypic switching of Ly6C^hi^/^lo^ macrophages. But this is clearly not the case in our experiment where the non-primed MSCs group showed no significant enhancement on Ly6C^lo^ subtypes switching at 1 week posttreatment, in spite of the highest levels of IL-10 among all groups. In addition, another study from our laboratory found that upregulation of host interleukin-10 (IL-10) expression counteract clearance of schistosomes by macrophages [50], thus the overall benefit of high levels of IL-10 in the treatment of schistosomiasis warrants prudent assessment. Recently, it is reported that the phagocytosis of dying neutrophils by macrophages, known as efferocytosis, not only prevents secondary lysis and spillage of noxious neutrophil substances, but also triggers an anti-inflammatory and pro-resolving response in macrophages [40–42, 51]. Meanwhile, considering the close interaction between macrophages and neutrophils within the schistosomiasis granulomas, we attempted to explore efferocytosis-mediated mechanisms.
First, we observed that significant higher engulfment ratio of neutrophils both at 1 week and 2 weeks posttreatment in primed MSCs group (compared with control group or non-primed MSCs group, respectively), which is exactly consistent with the higher Ly6C^lo^ macrophage in primed MSCs group (Figs. 5A–C). Furthermore, in vitro coculture experiment directly showed enhancement effects of primed MSCs on efferocytosis, accompanied by decreased inflammatory cytokines(IL-1β, Tnf-ɑ and CX3CL1 mRNA) and increased pro-resolving factors(CX3CR1,IL-10 and MMP2 [41]). Taken together, our results showed that primed MSCs promoted macrophage efferocytosis, as well as pro-inflammatory/pro-resolving subset switching in schistosomiasis mice.
There have been some similar reports in CCl4-induced fibrosis models, indicating that MSCs treatment promoted the macrophage transition from Ly6C^hi^ to Ly6C^lo^ population or induction of anti-inflammatory macrophages, while enhancing engulfment of apoptotic bodies [15] or the lysosomal function of macrophage [23]. Another report further disclosed that small extracellular vesicles (sEVs) derived from interferon-γ (IFN-γ) pre-conditioned MSCs exerted superior therapeutic effects to those of non-preconditioned MSCs therapy, which was related to induction of anti-inflammatory macrophages with increased phagocytic ability of bioparticles [24]. Here, we clearly showed that IFN-γ + LPS-primed MSCs promoted macrophage efferocytosis of neutrophils, as well as its subtype switching in schistosomiasis mice. Next, the potential targets of primed MSCs acting on during efferocytosis process were further identified.
We screened for engulfment receptors that can recognize PS (ligands exposed on apoptotic cells), including TAM family Mertk, Axl and Gas6, BAI1, Timd4 and integrin Itgb2,5 [39, 44]. Surprisingly, except that primed MSCs specially promoted levels of Itgb2 and the non-primed MSCs reduced Itgb2, no significant differences in other receptors were observed among different groups (Fig. 2s). Following engagement of eat-me signals with their receptors, downstream signaling pathways are activated, in which, GTPase Rac1 plays an important role in the organization of actin filament networks and in membrane ruffling to help macrophages complete the phagocytic program [39, 44]. Our results showed that primed MSCs specially promoted mRNA of Rac1, as well as Rac1protein activity at 1 hour after efferocytosis; Additionally, blocking Itgb2 ligation led to reduced Rac1 expression and Rac1 activity, as well as the decreased efferocytosis efficiency. Furthermore, both blocking Itgb2 ligation and inhibition Rac1 activation reverse promotion effects of primed MSCs in macrophage subtypes switching. Thus, our data suggest that primed MSCs promote Ly6C^hi^ /Ly6C^lo^ switching in a manner dependent on itgb2-Rac-1 mediated efferocytosis process.
Some reports have disclosed the significance of LFA-1(αLβ2) or itgb2 (CD18)) within human monocyte derived macrophage in the efferocytosis and its association with inflammation trajectory [52, 53]; a diminished efferocytosis was observed in CD18−/− mouse model, which drive inflammatory bone loss of periodontitis and related upregulation of interleukin (IL)-23 and IL-17 [53]. More interestingly, it has been indicated that CD18 controls the proliferation and maturation of monocytes in schistosomiasis mice; and low CD18 expression affects the accumulation of Ly6C^high^ CD11b^+^ monocytes in the liver and their subsequent polarization to pro-resolving macrophages, thus interfere tissue repair during chronic schistosomiasis [54]. Our results provide application potential for this finding, suggesting that MSC therapy targeting to enhance Itgb2-Rac1 activation in a meticulous manner may be beneficial for schistosomiasis fibrosis treatment. Anyway, we still propose that other crosstalk signaling pathways may jointly contribute to efferocytosis process. In-depth mechanistic research and its applied validation studies in this field are indispensable.
Conclusion
In conclusion, our data demonstrated IFN-γ + LPS-primed MSCs promoted macrophage efferocytosis of neutrophil, as well as macrophage subset switching to participate in fibrosis resolution; and more importantly, macrophage efferocytosis is closely related to Itgb2-Rac1 pathway (Fig. 9). Although the mechanisms by which primed MSCs modulate Itgb2 expression remain to be elucidated, it offers novel insights into the therapeutic targets of MSC-based anti-fibrotic therapy.Fig. 9. Schematic of the mechanisms by which IFN-γ + LPS-primed MSCs regulate macrophage phenotypic switching through efferocytosis pathway.
Materials and methods
Host animals
BALB/c male mice, 6–8 weeks old, were obtained from the Laboratory Animal Center of Guangdong province (Guangzhou, China) and housed in the Biosafety Level-2 (BSL-2) laboratory of Sun Yat-sen University at 5 mice per cage on a 12-h light/dark cycle with 21–26 °C room temperature, 40–70% humidity, and ad libitum access to food and water. Experiments were performed as a non-blinded approach, and all animal work was performed according to the recommendations of the Laboratory Animal Welfare and Ethics Committee (LAWEC) of China and approved by the LAWEC of South China University of Technology.
Isolation and culture of mouse MSCs
Mouse MSCs were isolated from the bone marrow of 6-week-old C57BL/6 mice according to the improved low-density culture method reported previously [32]. In brief, femurs and tibiae were removed and bone marrow cavity was flushed with L-Dulbecco’s modified essential medium (DMEM) complete medium, and individual cells were obtained. After red blood cells were removed by ammonium chloride lysis, the remaining cells were cultured at the low density of 5ⅹ10^4^ cells/cm^2^. Through continuous passage, nonadherent cells were gradually removed and the remaining adherent cells at 3 passages were collected and identified. Those positive for Sca-1and CD105, but negative for CD45, CD11b and MHC-Ⅰ, Ⅱ, were used for in vitro and in vivo experiment.
Liver fibrosis models induced by schistosomiasis
BALB/c male mice were divided randomly into 4 groups: normal group (uninfected), infected control group, MSC group, and IFN-γ plus LPS pretreated MSC group using a random number table. The number of animals in each group was indicated in the figure legend, which was determined by reference to our study experience and related researches [32, 55, 56]. Mice were percutaneously infected with 16 ± 2 pairs cercaria collected from S. japonicum-infected snails (bought from the Jiangsu Institute of Parasitic Diseases, Wuxi, China), followed by 5 days-long treatment with praziquantel (PZQ), once per day, oral gavage(i.g.) of 150 mg/kg, at 6 weeks post-infection, and then mice received 1ⅹ 10^5^MSC via the tail vein or pretreated MSC suspended in 0.2 ml of saline. Animals were euthanized by cervical dislocation under anesthesia at 1 and 2 weeks after MSC treatment (or at the corresponding observation period in non-MSC control group). Thus, the effects of MSC treatment for the hepatic fibrosis induced by schistosomiasis are investigated.
When inhibition of monocyte recruitment was required, mice received intraperitoneal (i.p.) injection of 500 µLCCR2 antagonist (RS102895, KKL-med, USA) (15 mg/kg/day) The injections were performed three times a day for 1 week, prior to MSC treatment.
Hepatic histologic evaluation
Liver samples were obtained at 1 week, 2 weeks after MSC treatment, respectively. The remaining right liver lobe of each mouse (others were used for FACS analysis and RNA extraction, respectively) was fixed in 4% paraformaldehyde, embedded in paraffin, and cut into about 5-μm sections. The sections were stained with hematoxylin and eosin (H&E) and Masson to show granuloma formation and the deposition of collagen fibers in liver. Representative sections from each animal were scanned under ×200 magnifications with a compound microscope, and the degree of liver fibrosis was quantified by calculating the percentage of fibrosis area using AxioVision Rel 4.7 (Carl Zeiss GmbH, Jena, Germany).
Flow cytometry detection
Percentages of Ly6C^high^CX3CR1^low^ and Ly6C^low^CX3CR1^high^ macrophages in the liver of infected mice at 1 week and 2 weeks after treatment with MSC, respectively, were detected by flow cytometry. Liver mononuclear cells were isolated from the livers of mice as previously described [32]. Briefly, mouse livers were minced through a 100 μm cell strainer (BD Falcon) and re-suspended in PBS (pH 7.2) and centrifuged at 50 × g for 5 min. Supernatants containing liver mononuclear cells were collected and washed in PBS. The cell pellets were re-suspended in 35% Percoll solution (Sigma-Aldrich) and gently overlaid onto 70% Percoll. After centrifuged (600 × g) for 30 min at room temperature, the inter-phase containing liver mononuclear cells were collected and washed twice in PBS. These cells were subsequently used for surface staining.
The following antibodies were used for flow cytometry analysis: FITC-anti-mouse CD11b (11-0112-41, Thermofisher, San Diego, CA, USA), PE-cyanine 7-anti-mouse F4/80 (25-4801-82, Thermofisher), Percp-cyanine5.5-anti-mouse Ly-6C (128012, BioLegend, San Diego, CA, USA), PE-anti-mouse CD192(CCR2, 160105, BioLegend), APC/Cyanine7-anti-mouse CX3CR1(149047, BioLegend). Stained cells were detected by flow cytometry (CytoFLEX S, Beckman, USA), and the data were analyzed using FlowJo software (TreeStar, Ashland, OR, USA).
Real-time polymerase chain reaction
Total RNA was extracted with the TRIzol reagent (Invitrogen, CA, USA), and reverse transcription (RT) was performed using Evo M-MLV RT master mix (Accurate biology, China). The cDNA thus obtained was subjected to real-time PCR with the SYBR Green reagent (Roche Life Science, Switzerland) using Primers specific for the following genes listed in Table 1s. The housekeeping gene GADPH was used as an internal control and Quantitation of relative mRNA expression was calculated using the 2^−ΔΔCt^ method
Liver immunofluorescence staining
About 4-mm-thick paraffin sections were dewaxed and incubated, according to the manufacturer’s instructions. The sections banded with MERTK (EPR17534-139, Abcam, USA; diluted at 1:1000) and Ly6G (EPR22909-135, Abcam, USA; diluted at 1:1000)) antibodies at 4 °C overnight and then were incubated with the corresponding secondary antibodies for 45 min. Nuclei were counterstained with 4’, 6-diamidino-2-phenylindole (DAPI, 564907, BD Pharmingen, USA; diluted at 1:1000). Observation and photo-taking were conducted by a confocal laser-scanning microscope (LSM700, Carl Zeiss, Germany)
Generation of bone marrow-derived macrophages (BMDMs) and preparation
CD45^+^CD11b^+^ monocytes were isolated from bone marrow cells by FACS sorting (using BD FACSAria II); and the enriched monocyte fraction was cultured in medium (Dulbecco’s modified Eagle’s medium/F12 media; Thermo Fisher Scientific, Waltham, MA) containing 10 ng/ml CSF-1 (Recombinant Murine M-CSF; Peprotech Inc., Rocky Hill, NJ) at 37 °C in 5% CO_2_. After 7 days-incubation, BMDMs were collected and identified. These cells were subsequently incubated without or with unpretreated MSC, or TLR4 + IFN-γ-treated MSC for 24 h at 37 °C in 5% CO_2_, and then used for coculture with apoptotic neutrophils.
Generation of apoptotic neutrophils
Bone marrow neutrophils were using Histopaque 11191 and 10771 (Sigma-Aldrich) as previously described [56]. Isolated neutrophils were stimulated with staurosporine (1 μM, 8 h) for induction of apoptosis and labeled with 2 μM Celltracker CM-DiI following the manufacturer’s instructions (Yeasen Biotechnology, Shanghai). The percentage of apoptosis was detected by FACS (>90%).
Assessment of efferocytosis in vitro
In vitro efferocytosis assay was performed as previously described [57], by co-culturing BMDMs with apoptotic neutrophils labeled with CM-DiI in a proportion of three apoptotic neutrophils: one macrophage. After co-incubation for 1 hour, cells were washed and fixed with 4% PFA and further stained for MERTK (EPR17534-139, Abcam, USA; diluted at 1:500). The efferocytosis was examined with a confocal laser-scanning microscope (LSM700, Carl Zeiss, Germany), and efferocytosis index was determined on 300 cells as the percentage of BMDMs containing at least one ingested apoptotic neutrophil. Digital images were processed using ImageJ software (National Institutes of Health, Bethesda, MD, USA).
ELISA
The levels of cytokine in the culture supernatants were determined according to the instruction of the ELISA Kits (eBioscience). In brief, mouse TNF and IL-10 were detected by biotinylated monoclonal antibodies, which were evidenced by avidin-conjugated horseradish peroxidase followed by incubation with TMB substrate. OD values at 450 nm were recorded using MK3 microplate reader.
Macrophage Rac1 activity G-LISA assays
BMDMs were cocultured with apoptotic neutrophils in 24-well plate, as described above. After co-coculture for indicated times, BMDMs were collected, and active Rac1 was measured using the Rac1 G-LISA Activation Assay kit (BK126, Cytoskeleton) according to the manufacturer’s instructions. For blocking test, before adding apoptotic neutrophils, BMDMs were first incubated with neutralizing anti-CD18 antibody (2.5ug/ml, eBioscience™) for 3 h. Background control values from cells pre-treated with Rac1 inhibitor NSC23766 (100 µm, Biofount) were subtracted from sample values to get Rac1 activity values.
Statistical analysis
All graphs and statistical analyses were performed using GraphPad Prism 8.0 (GraphPad Software Inc., San Diego, CA, USA). All data are shown as mean ± SEM. Multiple comparisons were analyzed by two-way ANOVA with Dunnett’s posttest for comparisons of three or more groups at different time points after treatment, or One-way ANOVA with Tukey’s multiple comparisons for comparisons of three groups at certain time point. All data were assessed for Gaussian distribution using the D’Agostino-Pearson normality test before analysis with these parametric tests).
Supplementary information
supplementary material