Abstract
Duchenne muscle dystrophy (DMD), characterized by progressive loss of muscle architecture and function, is caused by lack of dystrophin expression in the sarcolemma of myofibers. Recurrent muscle damages in DMD patients and DMD mouse model, mdx, lead to chronic inflammation, which further exacerbate the muscle histopathology. It is critical to find a successful therapy that will improve the histopathology of muscles of DMD patients and restore skeletal muscle function. TIPE2 (tumor necrosis factor α-induced-protein 8-like 2), identified as a negative regulator of immune response, has been found to be expressed in various types of immune cells including macrophages. However, whether and how TIPE2 plays a role in the DMD-related inflammation remains unknown. In this study, we found the basal expression levels of TIPE2 in skeletal muscle from mdx mice are significantly lower than wild-type (WT) mice. To investigate the potential beneficial effect of TIPE2 in muscular dystrophy, we performed intramuscular injection of adeno-associated virus 9 carrying the TIPE2 gene in mdx mice. Our results indicate that the restoration of TIPE2 ameliorates muscular dystrophy phenotype through a reduction in inflammation and fibrosis. In addition, TIPE2 overexpression dramatically decreased the proliferation and migration rate of macrophages, as well as repressed the secretion of pro-inflammatory factors induced by tumor necrotic factor alpha. Taken together, our results indicate that a reduction of TIPE2 expression is observed in dystrophic skeletal muscle, when compared to WT and more importantly that TIPE2 gene delivery may provide as a novel anti-inflammatory therapy to alleviate the muscle weakness in DMD patients.
Introduction
Duchenne muscular dystrophy (DMD) patients, due to the lack of dystrophin and associated glycoprotein complex expression from the sarcolemmal membrane, rendered muscle fibers fragile and susceptible to damage during contraction which resulting in muscle weakness (1). In the process, the damaged muscle cells collapsed, died and subsequently led to immune cell infiltration, further contributing to muscle weakness and muscle cells necrosis. This results in macrophages infiltration to clean up the necrotic cells where inflammation factor tumor necrotic factor alpha (TNF-α) plays a key role in progression of DMD, which promotes muscle necrosis and eventually leads to muscle dysfunction. So it is important to establish anti-inflammatory therapeutic approaches for DMD patients; therefore, more potential therapeutic targets need to be explored in order to develop better treatments for DMD patients. Infiltration of multiple types of immune cells including macrophages, T cells and neutrophils were found in muscles of DMD patients and mdx mice, an animal model of DMD (2). It has been reported that macrophages comprise more than 35% of infiltrating mononuclear cells in DMD patients at age 2–8 years old. In mdx mice, macrophage is the predominant infiltrating immune cell at the age of 2–4 weeks old (2). The M1 subset of macrophages expresses high levels of pro-inflammatory cytokines such as monocyte chemoattractant protein-1 (MCP-1), IL-1β and TNF-α, and lyses the myofibers via production of nitric oxide synthase (NOS2) (3). M2 macrophages have increased synthesis of IL-10, an anti-inflammatory cytokine (4). During DMD-associated muscle histopathology, M1 macrophages infiltrate myofibers at the early stage, then convert to M2 phenotype to reduce inflammation and promote muscle regeneration (5).
Expression of TIPE2 in GM muscle from WT and mdx mice. (A) Immunohistochemical staining of TIPE2 in paraffin-embedded GM sections from WT mice and mdx mice (n = 4). Representative images are presented with arrowheads indicating positive TIPE2 staining in muscle. (B) Representative western blot and (C) densitometry analysis of TIPE2 expression in GM from WT and mdx mice (n = 3). Scale bar = 50 μm.
TNF-α, also known as the `master regulator’ of inflammatory cytokine synthesis, is involved in many inflammatory diseases (6). TNF-α is highly expressed in inflammatory cells and injured myotubes (7), also significantly increased on both mRNA and protein levels in muscles and serum from DMD patients and mdx mice (8). TNF-α takes into effect with interferon gamma (IFN-γ) to induce activation of macrophages (9). TNF-α has been found to prevent apoptosis of macrophages (10) and enhance macrophage migration (11). Furthermore, it facilitates recruitment of macrophages to injured central nervous system and peripheral nerve (12). Studies have shown that administration of TNF-α inhibitory drugs dramatically reduces inflammatory responses and muscle breakdown in dystrophic mdx mice (13,14).
Extensive research over the past several years has shown that chronic inflammation causes numerous human chronic diseases, including DMD (15). At the molecular level, inflammation is regulated by numerous molecules and factors, including cytokines interleukin (IL)-1, IL-2, IL-6, IL-12 and TNF-α. The current one paradigm for treatment of DMD is anti-inflammatory therapy. One commercial drug for DMD treatment is corticosteroids, which come with undesirable side effects and have limited therapeutic efficacy (16). Therefore, more potential therapeutic targets need to be explored in order to develop better anti-inflammatory treatments for DMD patients. Tumor necrosis factor alpha-induced protein 8-like 2 (TIPE2) belongs to the TNFAIP8 family, a group of proteins that share sequence homology and have been thought to regulate the immune system homeostasis (17). TIPE2 is expressed primarily in immune cells and lymphoid tissues. However, there is TIPE2 mRNA and protein detected in skeletal muscle (18). TIPE2 inhibits lipopolysaccharides (LPS)-triggered TNF-α production (19), as well as bacteria-induced expression of pro-inflammatory cytokines including IL-6 and TNF-α (20). TIPE2 null mice show multi-organ inflammation, splenomegaly and premature death (17). Macrophages isolated from TIPE2 knockout mice exhibit decreased migration and proliferation compared to wild-type (WT) macrophages (21). Additionally, previous studies indicate that TIPE2 skews macrophages to the M2 phenotype and inhibits macrophage differentiation into M1 phenotype (22). Whether TIPE2 is involved in the inflammation or muscle necrosis associated with DMD remains elusive.
Analysis of TIPE2 expression in GM muscle of mdx mice after intramuscular injection of AAV9. (A) Schematic representation of the recombinant AAV vector. The AAV-TIPE2 construct contains the mouse TIPE2 gene, P2A sequence followed by the ZsGreen gene. The expression of TIPE2 is driven by the CMV promoter. The AAV vectors (AAV9-ZsGreen and AAV9-TIPE2) were packaged into AAV9 capsids. ITR: inverted terminal repeat. (B) Expression of ZsGreen in AAV9-transduced GM sections was observed via fluorescence microscopy. Non-injected GM from mdx mice were used as controls. (C) Images of representative sections after IHC with antibody specific for TIPE2. (D) Representative western blot and densitometry analysis of TIPE2 expression in AAV9-injected GM. (E) Representative immunofluorescence images showing areas and cells with positive signal of F4/80 and GFP in AAV9-injected GM. **P < 0.01. Scale bars (B) and (E) = 100 μm; scale bar (C) = 50 μm.
In this study, we first examined the expression of TIPE2 in mdx mice muscles and found TIPE2 was downregulated in the diseased muscle. We constructed an adenovirus-associated viral (AAV) vector for TIPE2 overexpression and performed cellular transduction, intramuscular injection in mdx mice. Our results demonstrated that enhanced TIPE2 expression modulate inflammation and ameliorates the muscle histopathology in muscular dystrophy.
Results
TIPE2 levels are reduced in skeletal muscle of mdx mice
The basal levels of TIPE2 in WT and mdx mouse skeletal muscle were examined. Immunohistochemistry (IHC) was performed on gastrocnemius (GM) sections with anti-TIPE2 antibody. Our results revealed a decreased number of positive (TIPE2-stained) myofibers as well as weaker TIPE2-stained sections in the skeletal muscle of mdx mice, when compared to WT mice (Fig. 1A). The localization of TIPE2 was determined by co-staining TIPE2 with either a macrophage marker F4/80 or an α-bungarotoxin, a peptide that binds to nicotinic acetylcholine receptor (1–3) at neuromuscular junctions in skeletal muscles (23–25). These results indicate that TIPE2 localizes at neuromuscular junctions in WT GM, and that macrophages in mdx GM do not express TIPE2 (Supplementary Material, Figs 1 and 2). The expression of TIPE2 in the skeletal muscles of DMD patients and healthy individuals was also evaluated by immunofluorescence staining. The results showed that the level of TIPE2 expression decreased in the skeletal muscles of DMD patients when compared to the muscles of healthy individuals (Supplementary Material, Fig. 3). To further validate these results, western blots were conducted with protein lysates of GM from WT and mdx mice. After probing with anti-TIPE2 antibody, a reduction in TIPE2 expression was observed in mdx skeletal muscle when compared to WT mice (Fig. 1B). Densitometry analysis of the western blot results (Fig. 1C) indicate a reduction in TIPE2 expression (P = 0.05) in mdx mouse GM, when compared to WT mice.
Skeletal muscle pathology in GM muscle from mdx mice at 2 weeks post-intramuscular injection of AAV9. (A) H&E staining and Masson trichrome staining were performed on mouse GM sections. (B) The proportion of central nuclei was calculated to assess extent of muscle regeneration. (C) Blue staining area after trichrome staining was quantified to evaluate muscle fibrosis. (D) Necrotic fibers from muscles were visualized by positive IgG staining. Muscle sections were stained with an antibody specific for eMyHC as a marker for newly regenerated myofibers, and anti-laminin to visualize individual myofibers. Anti-F4/80 antibody was used to localize macrophages. (E) Quantification of IgG-positive myofibers. (F) Calculation of eMyHC-positive myofibers. (G) Quantification of F4/80-positive cell ratio. Values are represented as the mean + SD; n = 5. *P < 0.05; **P < 0.01. Scale bars (A) = 50 μm; (D) = 100 μm.
TIPE2 gene transfer in dystrophic muscle by AAV intramuscular injection
Based on the results demonstrating that skeletal muscle of mdx mice express lower levels of TIPE2 than WT mice, we designed and constructed an AAV vector which encoded for the mouse TIPE2 cDNA (Fig. 2A). Restoring expression of TIPE2 in the mdx mice was achieved by adeno-associated virus 9 (AAV9)–TIPE2 injected intramuscularly into the left hind limb GM, while AA9-ZsGreen was injected as a control into the right hind limb GM of the same mdx mouse. Each time four mice were used for the experiment and were sacrificed 2 weeks after AAV administration. The GM tissues were processed into frozen sections or paraffin sections. Frozen sections were observed under a fluorescence microscope and green fluorescence was observed in muscle injected with the AAV9 virus (Fig. 2B). Muscle lysate was collected, followed by western blotting to detect TIPE2 protein. The expression of TIPE2 was significantly higher in AAV9-TIPE2-injected muscle than AAV9-ZsGreen-injected muscle, which was quantified by densitometry analysis (Fig. 2C). IHC was performed on paraffin sections and TIPE2 staining revealed increased TIPE2 expression after AAV9-TIPE2 administration (Fig. 2D) when compared to AAV9-ZsGreen-injected skeletal muscle. Collectively, these data indicate that TIPE2 expression in mdx skeletal muscle was increased after intramuscular injection of AAV9-TIPE2 vector. GFP and F4/80 double positive cells were found in both AAV virus-injected groups (Fig. 2E).
TIPE2 overexpression ameliorates muscular dystrophy in mdx mice
The potential beneficial effect of TIPE2 overexpression in dystrophic muscle was evaluated by histopathological staining. Briefly, we examined the histopathology of GM tissue from mdx mice and observed that the number of centro-nucleated myofibers was dramatically diminished in AAV9-TIPE2-injected muscles when compared to non-injected mdx skeletal muscle. A significant reduction in overall fibrosis was also observed with the presence of increased TIPE2 expression in the mdx skeletal muscle (Fig. 3A–C). In addition, a lower percentage of mouse IgG-positive myofibers was observed in AAV9-TIPE2-injected muscle in comparison to the AAV9-ZsGreen-injected muscle indicating a reduction in the number of necrotic myofibers (Fig. 3D and E). Embryonic myosin heavy chain (eMyHC) was used to establish the level of muscle regeneration, and, consistent with the observation of central nuclei numbers, AAV9-TIPE2 injection significantly decreased the proportion of eMyHC-positive fibers (Fig. 3D and F). Quantitative real-time polymerase chain reaction (qRT-PCR) was performed to evaluate the mRNA expression of eMyHC 2 weeks after AAV injection in mdx muscle; the results showed lower eMyHC mRNA expression in AAV9-TIPE2-injected muscle compared to control group (Supplementary Material, Fig. 4). Because macrophages are important cellular components in inflammatory responses, muscle cryosections of mdx mice (with and without gene transfer) were stained with antibody specific for the macrophage marker F4/80. Muscle with TIPE2 overexpression showed a significant reduction in the number of F4/80-positive cells (Fig. 3D and G) when compared to non-injected muscle, suggesting a reduction in inflammation in the AAV-TIPE2-injected muscle.
Improved myofiber membrane integrity after delivery of AAV9-TIPE2 into mdx muscle
Myofiber membrane integrity damage occurs during dystrophic muscle degeneration and regeneration. Evans blue dye (EBD) leaks into myofibers with disrupted membranes and can be detected visually using red fluorescent light under a fluorescent microscope. However, it cannot leak into healthy myofibers, which have an intact membrane. To assess whether AAV9-TIPE2 treatment protected the integrity of myofiber membranes, an integrity assay was performed by tail vein injection of EBD after 2 weeks of treatment. Our results in the AAV9-TIPE2-treated group show that GFP-positive myofibers exhibited less EBD permeation (Fig. 4). In contrast, GFP-negative myofibers revealed a large amount of EBD uptake. This result demonstrates that myofiber membrane integrity is functionally improved by TIPE2 treatment in mdx mice.
TIPE2 improves muscle membrane integrity. AAV9-TIPE2-transduced myofibers showed positive GFP signal (green) and less EBD (red) leakage into myofibers compared to non-GFP-transduced myofibers, which take up EBD (red). Scale bar = 100 μm.
The effect of TIPE2 overexpression on inflammation-related genes in macrophages
Given the finding that the number of F4/80-positive cells was reduced following AAV9-TIPE2 injection, we explored the effects of AAV9-TIPE2 infection in RAW264.7 cells, a murine macrophage cell line. AAV9-TIPE2 vector-infected RAW264.7 cells were analyzed via fluorescent microscopy after incubation for 48 h and green fluorescence was observed in the transduced macrophages (Fig. 5A). The protein levels of TIPE2 in AAV9-ZsGreen and AAV9-TIPE2 vector-transduced macrophages were determined using western blot analysis. TIPE2 protein bands were detected on blots from extracts from both non-transduced and transduced cells with both the TIPE2 AAV and the empty vector (control). A stronger TIPE2 band was observed for the protein extract from cells infected with AAV9-TIPE2 compared to uninfected cells or AAV9-ZsGreen vector-infected cells (Fig. 5B). Macrophages were incubated with TNF-α (10 ng/mL) for 16 h, and qRT-PCR was performed to detect the expression of inflammatory genes MCP-1, IL-6, IL-1β and p65. Levels of mRNA expression of these genes in response to exogenous TNF-α addition were detected in RAW264.7 cells. The TIPE2-overexpressing cells exhibited a reduced upregulation of p65 with TNF-α incubation compared to the empty vector (control group). No significant differences for IL-1β or IL-6 mRNA expression levels were found between cells treated and untreated with TNF-α in the presence of TIPE2 overexpression. MCP-1 expression was greatly suppressed in TIPE2-overexpressing macrophages treated with TNF-α compared to the control cells (Fig. 5C). Cytokine and chemokine levels in macrophage culture medium were determined by cytokine array analysis as well (Fig. 5D). Increased secretion of multiple cytokines (i.e. IFN-γ, IL-7, IL-12p70, IL-16, IL-17 and IL-23) in both AAV vector-transduced cells with TNF-α treatment, with a reduced induction in TIPE2-overexpressing cells, were detected. Highly induced IL-1Rα levels were observed in AAV9-ZsGreen vector-transduced cells, but not in TIPE2-overexpressing cells, under TNF-α stimulation.
Expression of inflammatory genes in TNF-α-treated RAW264.7 cells transduced with AAV9-ZsGreen or AAV9-TIPE2. (A) Representative images showing ZsGreen fluorescence 2 days after viral transduction. (B) Western blot showing the protein levels of TIPE2 in non-transduced RAW264.7 cells and cells transduced with AAV9-ZsGreen or AAV9-TIPE2. (C) Levels of mRNA expression of inflammatory genes in AAV-infected RAW264.7 cells were examined by quantitative RT-PCR at 18 h after TNF-α treatment. (D) Relative expression levels of cytokines in culture medium were measured using a proteome cytokine array, and representative quantified data are shown as mean pixel density. Values are represented as the mean + SD; n = 3 per group for western blot and qRT-PCR analysis. For cytokine array, n = 2 per group. *P < 0.05; **P < 0.01 versus AAV9-ZsGreen in (B) densitometry analysis and versus untreated group in (C). NS: not significant. Scale bar = 50 μm.
TIPE2 overexpression reduces inflammation in mdx mice
To examine the effects of TIPE2 overexpression in the mdx skeletal muscle, we performed qRT-PCR to detect mRNA levels of inflammation-related genes (Fig. 6A). It was found that pro-inflammatory cytokines, including IL-1β, IL-6, MCP-1, IFN-γ and TNF-α were significantly elevated in GM from mdx mice injected with AAV9-ZsGreen or phosphate-buffered saline (PBS) (data not shown) compared with WT animals. However, injection of AAV9-TIPE2 significantly reduced expression of these pro-inflammatory cytokines. The high level of mRNA expression for NOS2 in mdx muscle was reduced after AAV9-TIPE2 injection (Fig. 6A). Expression of transforming growth factor-beta 1 (TGF-β1), a cytokine which plays a crucial role in inflammation and fibrosis, was lower in TIPE2-injected mdx muscle when compared to the control mdx muscle (Fig. 5A). The level of mRNA expression of the NFκB family member p65 was higher in mdx muscle injected with AAV9-ZsGreen, but the expression was reduced in AAV9-TIPE2-injected muscle (Fig. 6A). Protein levels of IL-1β, IL-6, MCP-1 and TNF-α were determined with immunoblotting (Fig. 6B), with more intense bands detected in mdx muscle injected with AAV9-ZsGreen compared to WT muscle. Nevertheless, TIPE2 overexpression in the GM dampened the induced expression of these pro-inflammatory cytokines. Densitometry analysis showed a significant increase in IL-1β, IL-6, MCP-1 and TNF-α expression levels in mdx muscle, when compared to mdx mice injected with AAV9-TIPE2 and normal WT muscle (Fig. 6B). Expression patterns of MCP-1, IL-6, NOS2 and TGF-β1 were visualized with immunofluorescence in cryosections from WT and mdx mouse GM (Fig. 6C). Compared with WT muscle, there was a larger proportion of MCP-1-positive myofibers, as well as stronger signal of each MCP-1-positive myofiber in AAV9-ZsGreen-injected mdx muscle. Significantly lower expression of MCP-1 was observed in mdx mice GM injected with AAV9-TIPE2 compared to AAV9-ZsGreen-injected mdx muscle (Fig. 6C). NOS2-positive cell numbers were higher in the AAV9-ZsGreen-injected group of mice, but barely detectable in AAV9-TIPE2-injected mdx muscle (Fig. 6A). Similarly, brighter fluorescence of IL-6 and TGF-β1 was observed in mdx GM injected with AAV9-ZsGreen vector, when compared to AAV9-TIPE2 overexpressing muscle. The expression of MCP-1, IL-6, NOS2 and TGF-β1 was decreased in mdx skeletal muscle after TIPE2 gene transfer (Fig. 6C). These data suggest that TIPE2 overexpression reduced inflammatory responses in the muscle from mdx mice.
Expression of inflammation-related genes in GM muscle following intramuscular AAV9 injection. (A) qRT-PCR analysis of inflammatory genes in the GM of WT and mdx mice injected with AAV9 vector, with GAPDH as the reference gene. (B) Representative western blot and (C) densitometry analysis of protein levels of inflammation-associated genes in mouse GM. (D) Immunofluorescence with anti-MCP-1, anti-IL-6, anti-NOS2 and anti-TGF-β1 antibody in mouse GM from WT and mdx mice injected with AAV9-ZsGreen or AAV9-TIPE2. Values are represented as the mean + SD; n = 4 per group. *P < 0.05; **P < 0.01; ***P < 0.001 versus WT mice when not specified. #P < 0.05; ###P < 0.001 versus mdx + AAV9-ZsGreen mice. NS: not significant. Scale bars = 100 μm.
TIPE2 overexpression inhibits proliferation and migration of macrophages
The proliferation rate of RAW264.7 cells was examined using the MTT assay. Macrophages transduced with AAV9-TIPE2 exhibited a lower proliferation rate than untransduced cells or AAV9-ZsGreen-transduced cells. There was no significant difference in proliferation rates between transduced and non-transduced cells with AAV9-ZsGreen (Fig. 7A). Scratch wound healing assay results showed that RAW264.7 cells overexpressing TIPE2 had decreased migratory abilities compared to non-transduced cells or AAV9-ZsGreen-transduced cells (Fig. 7B). AAV9-TIPE2, but not AAV9-ZsGreen, significantly inhibits the migration of macrophages after MCP-1 exposure compared to non-transduced cells or AAV9-ZsGreen-trasnduced cells (Fig. 7C and D).
TIPE2 overexpression regulates proliferation and migration of RAW264.7 macrophages. (A) Cell proliferation was evaluated using the MTT assay in non-transduced macrophages and cells transduced with AAV9 vector. Migratory capabilities of RAW264.7 cells were assessed by (B) a transwell migration assay conducted 24 h after adding MCP-1 (100 ng/mL) to culture medium and (C) a wound healing scratch assay performed 24 h after the cultured macrophage monolayer was scratched with a sterile pipette tip. Quantifications of the wound healing scratch assay were expressed as migration rates, normalized to the uninfected group (D). Quantification of the transwell migration assay was expressed as cell number in each field. At least six random fields were taken for each group. Values were represented as the mean + SD; *P < 0.05. NS: not significant. Scale bars = 100 μm.
Discussion
Chronic inflammation is an important characteristic of DMD pathology; therefore, inflammation suppression serves as a potential approach to alleviate muscular dystrophy. In this study, we discovered that TIPE2 negatively regulates inflammation and reduces fibrosis in skeletal muscles of mdx mice. This is the first report showing a potential therapeutic approach to alleviate muscle weakness in DMD through TIPE2 gene transfer.
The TNFAIP8 family of proteins is comprised of four members: TNFAIP8, TIPE1, TIPE2 and TIPE3, all which are involved in inflammatory diseases and carcinogenesis (26). TIPE2-/- mice exhibit worsened atherosclerosis with increased responses by macrophages to oxidized low density lipoprotein (27) and enhanced neointima formation (28). Exacerbated production of pro-inflammatory cytokines in response to autoimmune hepatitis is observed in TIPE2-deficient mice (29). TIPE2 overexpression dramatically suppresses asthma-related inflammation by inhibition of the Wnt/β-catenin pathway (30). In our current study, we demonstrated that mdx mice had lower expression of TIPE2 in the GM compared to WT mice. Considering that TIPE2 is identified as a negative regulator of inflammation, we hypothesized that the decreased level of TIPE2 contributes to enhanced inflammatory responses in mdx mice. Hence, we constructed an AAV vector that contains mouse TIPE2 cDNA and utilized AAV9, which was found to have high efficiency for gene transfer in skeletal muscle (31) (Fig. 2B). We found that AAV9-mediated overexpression of TIPE2 in the GM remarkably alleviated inflammation and fibrosis in mdx mice (Fig. 3A).
The expressions of inflammation-related cytokines/chemokines, such as IL-16 and IL-17 in macrophages, were dampened by TIPE2 overexpression. Originally described as a chemoattractant factor, IL-16 is a pro-inflammatory cytokine released by various types of cells, including lymphocytes. It promotes the production of other pro-inflammatory factors, such as IL-1β, IL-6 and TNF-α from monocytes (32). A previous investigation revealed that IL-16 activates the SAPK signaling pathway in macrophages (33). Mdx mice have elevated levels of IL-16 in skeletal muscles (34), but few studies have explored the effect of IL-16 on pathogenesis in muscular dystrophy. In our present study, we found IL-16 secretion suppressed by TIPE2 overexpression (Fig. 5), which has never been reported. IL-17 also acts as a pro-inflammatory cytokine, the expression of which is markedly higher in DMD patients than healthy controls (35). Although not required for macrophage activation in vitro (36), increased levels of IL-17 correlates with an increase in the M1/M2 ratio of macrophages at the lesion site of bisphosphonate-related osteonecrosis of the jaw in mice and humans (37). Whereas, the mechanism underlying the regulation of IL-17 by TIPE2 overexpression needs further investigation.
In the present study, we noticed that MCP-1 in RAW264.7 macrophages was induced by incubation with TNF-α. Murao et al. showed similar stimulation in endothelial cells, mediated by the Akt/PKB pathway (38). MCP-1 is an essential chemokine that regulates monocyte/macrophage migration and infiltration. MCP-1 knockout mice have reduced neuro-inflammatory responses at the site of endotoxin insult (37–39). The level of MCP-1 appears to be increased in DMD patients and mdx mice (40). Consistently, we found the mRNA and protein levels of MCP-1 remarkably increased in muscle from mdx mice compared to WT controls. However, this upregulation of MCP-1 was diminished by TIPE2 overexpression (Fig. 5C). It has been reported that TIPE2-/- mice have significantly higher expression of MCP-1 in brain following cerebral ischemia than WT mice (41). Ldlr-/- mice show dramatically higher levels of serum MCP-1 after transplantation with TIPE2-/- bone marrow cells (27). Our results indicate that TIPE2 decreased MCP-1 expression in mdx mice, which may be involved in suppression of immune cell infiltration in dystrophic skeletal muscles.
Besides the regulation of the function of immune cells and inflammation-related genes, TIPE2 negatively regulates TGFβ-1-induced proliferation and activation of quiescent hepatic satellite cells (42), a major cell player in tissue fibrosis. We observed reduced levels of TGFβ-1 and fibrosis in muscle of mdx mice injected with AAV9-TIPE2 (Fig. 3A and Fig. 6A, C). Yet it is still unclear whether TIPE2 ameliorates muscle fibrosis via direct modulation of TGFβ-1 signaling or through attenuated inflammatory responses in mdx mice, considering that macrophages are the major source of TGFβ-1 production in tissue undergoing fibrosis (43). Based on the results from immunofluorescence staining showing decreased IgG + necrotic myofibers in AAV9-TIPE2-treated mdx muscles, we speculated that TIPE2 protects the membrane integrity of myofibers in mdx mice. The skeletal muscles of mdx mice tend to lose sarcolemmal integrity, allowing the influx of calcium into myofibers. We therefore performed a membrane integrity assay and found that elevated TIPE2 expression in mdx mice stabilized membrane integrity in muscle fibers, as evidenced by a reduction in EBD permeation in these TIPE2-expressing myofibers compared to non-transduced myofibers. Our findings from membrane integrity using EBD injection indicate that AAV-TIPE2 treatment can protect the myofiber membrane, suggesting that the anti-inflammatory effect by increasing TIPE2 expression in muscle can be an additional strategy for restoring muscle function.
We noticed that increased mRNA levels of p65 in response to TNF-α stimuli were suppressed by TIPE2 overexpression. Also, reduced transcriptional expression of p65 was observed in muscle injected with the AAV9-TIPE2 vector in mdx mice (Fig. 5A). As a subunit of the NF-κB transcription factor complex, p65 is responsible for initiating transcription activity of NF-κB (44). The NF-κB pathway is a critical mediator of inflammation and is activated in DMD patients as well as mdx mice (45,46). NF-κB signaling can be induced by mechanical stress in myofibers in a time-dependent manner (47). Bioavailable NF-κB inhibitors have been proven to improve the phenotype of mdx mice, with regards to inflammation, fibrosis and muscle function (48). TNF-α treatment leads to phosphorylation of IκBα, an inhibitor of kappa B, resulting in enhanced mRNA expression of the p65 subunit (49) and nuclear translocation of NF-κB (50). An interesting study demonstrated the inhibitory action of TIPE2 on TNF-α-induced metastasis in carcinoma cells via inhibiting NF-κB activation (19). Moreover, Ono et al. (51) have reported that TIPE2-overexpressing macrophages exhibit lower NF-κB activity under stimuli with LPS, through the binding ability of TIPE2 to TGF-β-activated kinase 1 (TAK1). Intriguingly, the level of TAK1 protein was found to be significantly higher in mdx muscles and is believed to regulate muscle mass and regeneration (52). Hence, TAK1 signaling may potentially contribute to the reduced p65 mRNA levels we observed in TIPE2-overexpressing skeletal muscles of mdx mice.
Primer sequences
| Gene | Sequence | Amplicon size (bp) |
|---|---|---|
| P65 | F:5′- CACCAAGGATCCACCTCACC-3′ R:5′- AATGGCTTGCTCCAGGTCTC-3′ | 160 |
| IL-6 | F:5′-GGGACTGATGCTGGTGACAA-3′ R:5′- ACAGGTCTGTTGGGAGTGGT-3′ | 90 |
| IL-1β | F:5′-GCCACCTTTTGACAGTGATGAG-3′ R:5′- AATGGCTTGCTCCAGGTCTC-3′ | 95 |
| MCP-1 | F:5′- CTGCATCTGCCCTAAGGTCT-3′ R:5′- ACTGTCACACTGGTCACTCC -3′ | 127 |
| TNFα | F:5′-AGCCGATGGGTTGTACCTTG-3′ R:5′- ATAGCAAATCGGCTGACGGT-3′ | 99 |
| IFNγ | F:5′- TGAGTATTGCCAAGTTTG-3′ R:5′- CTTATTGGGACAATCTCTTCC-3′ | 159 |
| NOS2 | F:5′- CATCAACCAGTATTATGGCTC-3′ R:5′- TTTCCTTTGTTACAGCTTCC-3′ | 80 |
| TGF-β1 | F:5′- CTGCTGACCCCCACTGATAC-3′ R:5′- GCCCTGTATTCCGTCTCCTT-3′ | 93 |
| eMyHC | F:5′- GAACGGGCTGATATCGCAGA-3′ R: 5′-TCCTCGCTTTCATGGACCAC-3′ | 95 |
| GAPDH | F:5′- GCACAGTCAAGGCCGAGAAT-3′ R:5′- GCCTTCTCCATGGTGGTGAA-3′ | 151 |
| Gene | Sequence | Amplicon size (bp) |
|---|---|---|
| P65 | F:5′- CACCAAGGATCCACCTCACC-3′ R:5′- AATGGCTTGCTCCAGGTCTC-3′ | 160 |
| IL-6 | F:5′-GGGACTGATGCTGGTGACAA-3′ R:5′- ACAGGTCTGTTGGGAGTGGT-3′ | 90 |
| IL-1β | F:5′-GCCACCTTTTGACAGTGATGAG-3′ R:5′- AATGGCTTGCTCCAGGTCTC-3′ | 95 |
| MCP-1 | F:5′- CTGCATCTGCCCTAAGGTCT-3′ R:5′- ACTGTCACACTGGTCACTCC -3′ | 127 |
| TNFα | F:5′-AGCCGATGGGTTGTACCTTG-3′ R:5′- ATAGCAAATCGGCTGACGGT-3′ | 99 |
| IFNγ | F:5′- TGAGTATTGCCAAGTTTG-3′ R:5′- CTTATTGGGACAATCTCTTCC-3′ | 159 |
| NOS2 | F:5′- CATCAACCAGTATTATGGCTC-3′ R:5′- TTTCCTTTGTTACAGCTTCC-3′ | 80 |
| TGF-β1 | F:5′- CTGCTGACCCCCACTGATAC-3′ R:5′- GCCCTGTATTCCGTCTCCTT-3′ | 93 |
| eMyHC | F:5′- GAACGGGCTGATATCGCAGA-3′ R: 5′-TCCTCGCTTTCATGGACCAC-3′ | 95 |
| GAPDH | F:5′- GCACAGTCAAGGCCGAGAAT-3′ R:5′- GCCTTCTCCATGGTGGTGAA-3′ | 151 |
Primer sequences
| Gene | Sequence | Amplicon size (bp) |
|---|---|---|
| P65 | F:5′- CACCAAGGATCCACCTCACC-3′ R:5′- AATGGCTTGCTCCAGGTCTC-3′ | 160 |
| IL-6 | F:5′-GGGACTGATGCTGGTGACAA-3′ R:5′- ACAGGTCTGTTGGGAGTGGT-3′ | 90 |
| IL-1β | F:5′-GCCACCTTTTGACAGTGATGAG-3′ R:5′- AATGGCTTGCTCCAGGTCTC-3′ | 95 |
| MCP-1 | F:5′- CTGCATCTGCCCTAAGGTCT-3′ R:5′- ACTGTCACACTGGTCACTCC -3′ | 127 |
| TNFα | F:5′-AGCCGATGGGTTGTACCTTG-3′ R:5′- ATAGCAAATCGGCTGACGGT-3′ | 99 |
| IFNγ | F:5′- TGAGTATTGCCAAGTTTG-3′ R:5′- CTTATTGGGACAATCTCTTCC-3′ | 159 |
| NOS2 | F:5′- CATCAACCAGTATTATGGCTC-3′ R:5′- TTTCCTTTGTTACAGCTTCC-3′ | 80 |
| TGF-β1 | F:5′- CTGCTGACCCCCACTGATAC-3′ R:5′- GCCCTGTATTCCGTCTCCTT-3′ | 93 |
| eMyHC | F:5′- GAACGGGCTGATATCGCAGA-3′ R: 5′-TCCTCGCTTTCATGGACCAC-3′ | 95 |
| GAPDH | F:5′- GCACAGTCAAGGCCGAGAAT-3′ R:5′- GCCTTCTCCATGGTGGTGAA-3′ | 151 |
| Gene | Sequence | Amplicon size (bp) |
|---|---|---|
| P65 | F:5′- CACCAAGGATCCACCTCACC-3′ R:5′- AATGGCTTGCTCCAGGTCTC-3′ | 160 |
| IL-6 | F:5′-GGGACTGATGCTGGTGACAA-3′ R:5′- ACAGGTCTGTTGGGAGTGGT-3′ | 90 |
| IL-1β | F:5′-GCCACCTTTTGACAGTGATGAG-3′ R:5′- AATGGCTTGCTCCAGGTCTC-3′ | 95 |
| MCP-1 | F:5′- CTGCATCTGCCCTAAGGTCT-3′ R:5′- ACTGTCACACTGGTCACTCC -3′ | 127 |
| TNFα | F:5′-AGCCGATGGGTTGTACCTTG-3′ R:5′- ATAGCAAATCGGCTGACGGT-3′ | 99 |
| IFNγ | F:5′- TGAGTATTGCCAAGTTTG-3′ R:5′- CTTATTGGGACAATCTCTTCC-3′ | 159 |
| NOS2 | F:5′- CATCAACCAGTATTATGGCTC-3′ R:5′- TTTCCTTTGTTACAGCTTCC-3′ | 80 |
| TGF-β1 | F:5′- CTGCTGACCCCCACTGATAC-3′ R:5′- GCCCTGTATTCCGTCTCCTT-3′ | 93 |
| eMyHC | F:5′- GAACGGGCTGATATCGCAGA-3′ R: 5′-TCCTCGCTTTCATGGACCAC-3′ | 95 |
| GAPDH | F:5′- GCACAGTCAAGGCCGAGAAT-3′ R:5′- GCCTTCTCCATGGTGGTGAA-3′ | 151 |
Antibodies used
| Target | Brand | Cat. No. | Application (dilution) | ||
|---|---|---|---|---|---|
| WB | IF | IHC | |||
| TIPE2 | Proteintach | 15940-1-AP | 1:2000 | 1:50 | 1:200 |
| IL-1β | R&D Systems | AF-401-SP | 1:5000 | ||
| TNF-α | R&D Systems | AF-410-SP | 1:2000 | ||
| MCP-1 | Santa Cruz | sc-52701 | 1:200 | 1:50 | |
| NOS2 | Abcam | ab15323 | 1:100 | ||
| Laminin | Abcam | Ab11575 | 1:500 | ||
| TGF-β1 | Abcam | ab92486 | 1:100 | ||
| IL-6 | Novus Biologicals | 600-1131 | 1:1000 | 1:200 | |
| F4/80 | Bio-Rad | MCA497R | 1:100 | ||
| eMyHC | Devel Studies | F1.625 | 1:50 | ||
| GAPDH | GeneTex | GTX100118 | 1:30000 | ||
| Target | Brand | Cat. No. | Application (dilution) | ||
|---|---|---|---|---|---|
| WB | IF | IHC | |||
| TIPE2 | Proteintach | 15940-1-AP | 1:2000 | 1:50 | 1:200 |
| IL-1β | R&D Systems | AF-401-SP | 1:5000 | ||
| TNF-α | R&D Systems | AF-410-SP | 1:2000 | ||
| MCP-1 | Santa Cruz | sc-52701 | 1:200 | 1:50 | |
| NOS2 | Abcam | ab15323 | 1:100 | ||
| Laminin | Abcam | Ab11575 | 1:500 | ||
| TGF-β1 | Abcam | ab92486 | 1:100 | ||
| IL-6 | Novus Biologicals | 600-1131 | 1:1000 | 1:200 | |
| F4/80 | Bio-Rad | MCA497R | 1:100 | ||
| eMyHC | Devel Studies | F1.625 | 1:50 | ||
| GAPDH | GeneTex | GTX100118 | 1:30000 | ||
Antibodies used
| Target | Brand | Cat. No. | Application (dilution) | ||
|---|---|---|---|---|---|
| WB | IF | IHC | |||
| TIPE2 | Proteintach | 15940-1-AP | 1:2000 | 1:50 | 1:200 |
| IL-1β | R&D Systems | AF-401-SP | 1:5000 | ||
| TNF-α | R&D Systems | AF-410-SP | 1:2000 | ||
| MCP-1 | Santa Cruz | sc-52701 | 1:200 | 1:50 | |
| NOS2 | Abcam | ab15323 | 1:100 | ||
| Laminin | Abcam | Ab11575 | 1:500 | ||
| TGF-β1 | Abcam | ab92486 | 1:100 | ||
| IL-6 | Novus Biologicals | 600-1131 | 1:1000 | 1:200 | |
| F4/80 | Bio-Rad | MCA497R | 1:100 | ||
| eMyHC | Devel Studies | F1.625 | 1:50 | ||
| GAPDH | GeneTex | GTX100118 | 1:30000 | ||
| Target | Brand | Cat. No. | Application (dilution) | ||
|---|---|---|---|---|---|
| WB | IF | IHC | |||
| TIPE2 | Proteintach | 15940-1-AP | 1:2000 | 1:50 | 1:200 |
| IL-1β | R&D Systems | AF-401-SP | 1:5000 | ||
| TNF-α | R&D Systems | AF-410-SP | 1:2000 | ||
| MCP-1 | Santa Cruz | sc-52701 | 1:200 | 1:50 | |
| NOS2 | Abcam | ab15323 | 1:100 | ||
| Laminin | Abcam | Ab11575 | 1:500 | ||
| TGF-β1 | Abcam | ab92486 | 1:100 | ||
| IL-6 | Novus Biologicals | 600-1131 | 1:1000 | 1:200 | |
| F4/80 | Bio-Rad | MCA497R | 1:100 | ||
| eMyHC | Devel Studies | F1.625 | 1:50 | ||
| GAPDH | GeneTex | GTX100118 | 1:30000 | ||
In mdx mice, inflammatory infiltrates happen as early as 2 weeks of age and increase significantly during 4 weeks to 8 weeks of age, with macrophages being the dominant infiltrating cell type (2). A previous study showed that bone marrow-derived macrophages isolated from TIPE2-/- mice exhibited higher proliferation and migration rates than macrophages obtained from WT mice (21). Consistent with previous findings, we found overexpression of TIPE2 inhibited the proliferation and migration of RAW264.7 macrophages (Fig. 7), which might be the potential cellular mechanism underlying the reduced infiltration of macrophages in mdx muscle injected with AAV9-TIPE2 vector.
The present work shows that TIPE2 overexpression alleviates inflammation and fibrosis in mdx mice, mediated by AAV9 viral vector gene transfer in skeletal muscle. The anti-inflammatory effect of TIPE2 is partially attributed to the inhibition of proliferation and migration of macrophages. Future studies are necessary to evaluate the effects of TIPE2 on muscle mass and function, as well as to determine the specific signaling pathways involved in the attenuation of pathogenesis in muscular dystrophy.
Materials and Methods
Animals
C57BL/10J and mdx mice were purchased from the Jackson Laboratory (Bar Harbor, ME, USA). Experimental protocols were approved by the institutional Animal Welfare Committee at the University of Texas Health Science Center at Houston. Mice at 6–7 weeks old were used in all the experiments described in this study. AAV vector (1010 vg of AAV9-ZsGreen or AAV9-TIPE2) was injected intramuscularly into the GM muscle (10) of mdx mice. At 2 weeks postinjection, mice were euthanized and the GM was excised for analysis.
Construction of the AAV-TIPE2 vector
Plasmid vectors used in this study include pAAV-human cytomegalovirus (CMV)-ZsGreen and a modified vector in which a NheI-AscI-P2A-AgeI linker was inserted between CMV and ZsGreen in pAAV-CMV-ZsGreen to obtain the pAAV-CMV-P2A-ZsGreen vector. The mouse TIPE2 open reading frame (ORF) was amplified from mouse blood cells by RT-PCR with the forward primer 5′-TCTAGAGCCACCATGGAGTCCTTCAGCTCAAAG-3′ and reverse primer 5′-GGCGCGCCTCAGAGCTTGCCCTCGTCCAGCA-3′. The XbaI and AscI digesting sites were used to assemble the TIPE2 ORF into the pAAV-CMV-P2A-ZsGreen vector under the CMV promoter.
AAV production and titration
Viral production and purification of AAV9 were carried out with triple plasmid co-transfection in HEK293 cells and aqua phase partitioning as previously described (53). The number of vector genomes (vg) was determined using AAV titration kit (Takara, Mountain View, CA). The titers for AAV9-ZsGreen and AAV9-TIPE2 were 4.7 × 1012 and 1 × 1012 GCP/ml, respectively.
Cell culture
The mouse macrophage cell line RAW264.7 was obtained from American Type Culture Collection and maintained in DMEM (Invitrogen, Carlsbad, CA) supplemented with FBS. For viral infection, 1 × 1010 vg AAV9 was added onto RAW264.7 cells cultured in T-75 flasks. For TNF-α treatment, cells with or without AAV9 infection were seeded in a 12-well plate, then incubated with TNF-α (10 ng/mL) for 18 h.
Mouse cytokine array
RAW264.7 cells were infected with AAV9-ZsGreen or AAV9-TIPE2 and cultured with TNF-α (10 ng/mL) for 48 h. Then culture medium was collected and centrifuged briefly to remove cell debris. Supernatant was used for cytokine arrays using a Proteome Profiler Mouse Cytokine Array Kit (R&D Systems, Minneapolis, MN).
MTT cell proliferation assay
The MTT cell proliferation assay was carried out using a Vybrant™ MTT Cell Proliferation Assay Kit (Invitrogen, Carlsbad, CA). RAW264.7 cells were seeded at 8000 cells per well in a 96-well plate and cultured for 48 h. MTT stock solution and SDS–HCl solution were added according to the manufacturer’s instructions.
Scratch wound stimulation
Confluent RAW264.7 cell monolayers in a 6-well plate were scratched with a sterile pipette tip and rinsed with PBS to remove cell debris, then cultured for up to 24 h. Images of the closing `scratch’ wound were taken by a microscope and the area of the wound was quantified with ImageJ.
Transwell cell migration assay
RAW264.7 cells resuspended in culture medium was seeded in each upper chamber of well inserts having 8 μm (pore size) membranes in a 12-well plate. MCP-1 (100 ng/mL) was used to stimulate cell migration and added to the culture medium in the lower chamber of each well. After 24 h of incubation, migrated cells were fixed with 70% ethanol and stained with 0.2% crystal violet. Microscopic images were captured and the numbers of migrated cells per well were counted using ImageJ.
Histological analysis
Muscle samples were frozen in 2-methylbutane precooled with liquid nitrogen, then embedded in optimal cutting temperature compound and stored at −80°C. For the evaluation of TIPE2 expression in DMD patients, we used frozen muscle sections from the laboratory of Dr Jacques P. Tremblay (Laval University, Canada). Healthy human muscle samples were obtained from the National Disease Research Interchange (NDRI; Philadelphia, PA). TIPE2 expression in human muscle samples were stained with TIPE2 antibody. To assess TIPE2 expression in mice, the GMs were fixed with formalin and embedded in paraffin. Sections were cut at 6 μm thickness, deparaffinized, rehydrated, followed by antigen retrieval and then incubated with TIPE2-specific antibody (Proteintech, Wuhan, China) at 4°C overnight. After secondary antibody incubation, a substrate-chromogen mix was used to visualize immunogen. To evaluate muscle pathology, the slides were stained with hematoxylin and eosin (H&E) or with a Masson trichrome reagent. The percentages of centrally nucleated fibers were calculated manually. The percentages of blue areas in Masson trichrome-stained slides were assessed using Fiji software (National Institutes of Health). For IHC, the sections were fixed with 4% paraformaldehyde, rinsed thoroughly with PBS and incubated with 5% FBS/BSA diluted in PBS. Muscle sections were stained using a mouse on mouse (MOM) kit (Vector Labs, Burlingame, CA) along with antibodies specific for F4/80 (Bio-Rad, Hercules, CA), eMyHC (Developmental Studies Hybridoma Bank, Iowa City, IA), MCP-1 (Santa Cruz Biotechnology, Santa Cruz, CA), IL-6 (Novus Biologicals, Centennial, CO), NOS2 (Abcam, Cambridge, UK), laminin (Abcam, Cambridge, UK) and TGF-β1 (Abcam, Cambridge, UK), and followed by incubation with Alexa Fluor® 647-conjugated anti-rabbit IgG, or Alexa Fluor® 594-conjugated anti-rat/mouse/rabbit IgG (Invitrogen, Carlsbad, CA). Endogenous IgG in the tissue was visualized using direct staining with Alexa Fluor® 594-conjugated anti-mouse IgG. α-Bungarotoxin, Alexa Fluor™ 488 conjugate (Invitrogen, Carlsbad, CA) was reconstituted in PBS and added to slides at 10 μg/mL. For the quantification of eMyHC-positive myofibers, at least six random 20× magnification images per sample were taken and blindly counted. The percentage of eMyHC-positive myofibers was calculated as the number of eMyHC-positive myofibers divided by the total number of muscle fibers per image.
Membrane integrity assay
For the determination of sarcolemma integrity in vivo, EBD was dissolved in PBS (10 mg/ml) and sterilized by passing through a 0.2 μm pore size membrane filter. EBD was intravenously injected into mice at a dose of 0.1 mg/g body weight 2 weeks after AAV TIPE2 treatment. The mice were then sacrificed 16 h later, and cryosections of GM tissue from five mice per group were directly imaged and analyzed.
qRT-PCR
Total RNA was extracted from cells or liquid nitrogen-treated GM tissue with TRIzol reagent (Thermo Fisher Scientific, Waltham, MA). The first-strand cDNA synthesis was performed using iScript™ Reverse Transcription Supermix (Bio-Rad, Hercules, CA). qRT-PCR for specific genes was performed using a CFX PCR detection system (Bio-Rad, Hercules, CA) and glyceraldehyde 3-phosphate dehydrogenase (GAPDH) as an internal control. Relative expression was calculated using the 2-DDCt method (54). Primer sequences are listed in Table 1.
Western blotting
Proteins were extracted from cells or tissues using RIPA buffer (Life Technology, Carlsbad, CA). Proteins were separated on precast polyacrylamide gels (Bio-Rad, Hercules, CA), and transferred to PVDF membranes (Millipore, Burlington, MA). Membranes were blocked with 5% BSA in TBS-T (Tris-buffered saline, 0.1% Tween 20) at room temperature for 1 h. To detect protein expression, antibody specific for TIPE2 (Proteintech, Wuhan, China), IL-1β (R&D Systems, Minneapolis, MN), IL-6 (Novus Biologicals, Centennial, CO), MCP-1 (Santa Cruz Biotechnology, Santa Cruz, CA) and TNF-α (R&D Systems, Minneapolis, MN) were used. Catalog numbers and dilutions of antibodies used for western blotting and immunofluorescence are listed in Table 2. Membranes were then incubated with horseradish peroxidase-conjugated secondary antibody and signals were developed using enhanced chemiluminescence substrate (Thermo Fisher Scientific, Waltham, MA).
Statistical Analysis
Data were analyzed to determine means and standard deviations (mean ± SD). The Student’s t-test was applied to perform statistical comparisons between two groups. A one-way analysis of variance was conducted to assess differences between multiple groups. A P < 0.05 was considered statistically significant.
Acknowledgements
We thank Jeannie Zhong for assistance with imaging. We thank Dr Yan Cui for helping with animal care and Dr Zhenhan Deng for sample collection. We appreciate Dr Jacques P. Tremblay (Laval University, Canada) for sharing muscle samples from DMD patients. We are grateful for scientific editing of the manuscript by Dr Mary Hall.
Conflict of Interest statement. None declared.
Funding
Department of Orthopedic Surgery, McGovern Medical School at the University of Texas Health Science Center at Houston (UTHealth) start-up funds (to P.G., to J.H.).
