Noggin promotes the formation of myotubes in DPSCs
The effective myogenic differentiation of stem cells is crucial for their repair of muscle injury. 5-Aza-2'-deoxycytidine (5-Aza) might be important for triggering the myogenic commitment of DPSCs. According to the results of Pisciotta et al., DPSCs cultured in the absence of preliminary 5-Aza treatment did not show any labeling for myogenic-specific markers, even when cultured in myogenic induction medium [8]. Therefore, we cultured and identified DPSCs (Fig. S1A-D) and found that 5-Aza induced myotube formation in DPSCs with an increase in myogenic markers (Fig. S2A-E). In the same field of vision, a relatively smaller number of myotubes that appeared to exhibit atrophy formed. However, 5-Aza-induced myogenic differentiation still showed low efficiency.
Noggin, a secreted BMP antagonist derived from the notochord and somite, is believed to improve myogenesis on epaxial somite [17]. To examine whether Noggin could affect the myogenic differentiation of DPSCs, we treated DPSCs with Noggin after 5-Aza induction (Fig. 1A). First, we found that Noggin had no effect on the cell proliferation of DPSCs (Fig. S3A-C) but increased myotube formation in the DPSCs (Fig. 1B-C). A more pronounced myotube morphology was observed after treatment with the Noggin protein for 21 days, and the number of multinuclear myotubes increased significantly (Fig. 1D). myosin heavy chain (MyHC) (Fig. 1E), such as MyHCIIA (Fig. 1F) and MyHCIIB (Fig. 1G) were significantly increased compared with those in the control group. This finding implies that Noggin might facilitate the formation of myotubes in DPSCs.
Noggin accelerates the progression of skeletal myogenic differentiation in DPSCs
Several myogenic genes, such as myogenic differentiation 1 (MyoD1), myogenic regulatory factor 4 (MRF4) and Desmin, have been proven to regulate myogenic processes, including myoblast differentiation into myocytes, fusion into myotubes and maintenance of the integrity of muscle cells [18, 19]. Therefore, we assessed the expression of MyoD1, MRF4 and Desmin on day 7, 14 and 21. The increased mRNA expression of MyoD1 (Fig. 2A) and Desmin (Fig. 2B) compared with that in the control group began on the seventh day. Noggin increased expression of MyoD1 and Desmin (Fig. 2A, B) but had no effect on the mRNA expression of MRF4 on day 7 (Fig. 2C). Immunofluorescence staining for MyoD1 (Fig. 2D) and Desmin (Fig. 2E) showed increased levels of MyoD1 (Fig. 2F) and an increased number of long, spindle-shaped myotube-like cells in Noggin-treated groups (Fig. 2G). On day 14, the increased protein expression of MyoD1 was also observed, indicating myoblast production [18]. In contrast, on day 21, MyoD1 expression decreased with disappeared trend (Fig. 2H, I). Our results were verified by an earlier study showing that secreted Noggin facilitated MyoD expression in embryonic tissues [20]. Moreover, the protein expression levels of Desmin (Fig. 2J) and MRF4 (Fig. 2K) were also significantly increased upon treatment with Noggin (Fig. 2H). In our study, Noggin has been shown to play an important regulatory role in accelerating the skeletal myogenic differentiation of DPSCs, which might be a novel factor in myogenesis.
The generation of satellite-like cells in DPSCs and their asymmetric self-renewal capacity
MPCs during development and SCs in adults are characterized by expression of paired box (Pax)3/7 [21]. SIX homeobox 1 (Six1) and EYA transcriptional coactivator and phosphatase 2 (Eya2) activate SCs (Pax7+) [22, 23], and the Six1-Eya2 complex functions in MPC specification by acting upstream of Pax3 and Myf5 expression and promoting myoblast differentiation [24, 25]. Thus, we measured the relative mRNA levels of Pax7, Pax3, Six1 and Eya2 onday1, 3, and 7 as well as the protein expression levels of Six1, Pax3, Pax7 and Eya2on day 7, 14 and 21.
The mRNA expression of Pax7 began to increase on day 1, increased in a concentration-dependent manner until peaking on day 3, and remained higher level compared to control on day 7 (Fig. 3A). Consistent with the expression of Pax7, the mRNA expression of Pax3 was increased on day 3, and remained increased on day 7 (Fig. 3B), and the mRNA expression of Six1 (Fig. 3C) and Eya2 (Fig. 3D) was increased on day 7. Western blotting (Fig. 3E) and immunofluorescence staining (Fig. 3F) for Pax7 showed that Noggin upregulated the protein expression of Pax7 on day 7, 14 and 21 (Fig. 3G, H). The protein expression levels of Pax3 (Fig. 3I), Six1 (Fig. 3J) and Eya2 (Fig. 3K) were also increased in the Noggin-treated groups on day 14 and remained high on day 21 (Fig. 3E). Previous studies have concluded that Pax3-mediated myogenesis requires an environment in which Six1 synergizes with Eya2 to activate the expression of MyoD [24, 26]. Our results showed that Noggin promoted the expression of Six1 and Eya2, suggesting that these Pax3/7 + cells were under such an environment for subsequent myogenic differentiation.
To address the functional significance of Six1/Eya2 expression in satellite-like cells, we compared our sequential expression data with that from other studies. Chang et al. introduced Pax7 + satellite-like cells from mES cells and found an earlier appearance of Pax3 expression on day 3, followed by Pax3/Pax7 expression on day 10, with the expression of Pax3 stronger than that of Pax7 [27]. In our study, we found that with Noggin treatment, the protein expression of Pax7 was at a high level on day 7 and Pax3/Six1/Eya2 expressions were increased on day 14 and 21 (Fig. 3E, H-K). One possible explanation for this difference is that hDPSCs went through an initial wave of differentiation into myogenic precursor cells that expressed Pax7 to promote myotube formation, followed by a second wave of differentiation into cells expressing Pax3/Six1/Eya2. This explanation implies that Noggin promotes the myogenic process in hDPSCs through upregulating the expression of members of the Pax3/Six1/Eya2 axis in addition to Pax7.
Noggin facilitates the skeletal myogenic differentiation of DPSCs via Smad/Pax7 pathway
As described before, Noggin is an antagonist of BMP. Cao et al. reported that BMP-4 appeared to inhibit myogenic differentiation of bone marrow-derived mesenchymal stromal cells by suppressing the transcriptional activity of myogenic factors [28]. To determine whether Noggin regulates the myogenic differentiation of DPSCs by regulating BMP signaling, we first simulated 3D protein structures using SWISS-MODEL and visualized them with PyMOL software, which showed that Noggin competitively inhibits the binding of BMP4 to BMP-receptor I A (BMPRIA) (Fig. 4A). Then, we blocked the effect of Noggin by adding the BMP protein. The protein levels of Pax7, Pax3, Eya2, MyoD1, and Desmin were observed (Fig. 4B). Compared with Noggin treatment (200N), the expression decreased in Noggin + BMP treatment. Among them, Pax3 and Pax7 was most obvious when compared with 5-Aza or control groups.
To further explore downstream regulation of BMP signaling by Noggin, the phosphorylation levels of members of the BMP/p-Smad pathway were also detected (Fig. 4C-E). The protein levels of p-Smad 1/5/9 were increased after activation of the BMP pathway by adding BMP4 (Fig. 4C). Noggin persistent eliminated the phosphorylation levels at 3h, 6h, and 9h even subsequently stimulated with BMP4 (Fig. 4C-E). The downstream effectors of the BMP/p-Smad pathway, such as inhibitor of DNA binding 1 (ID1) and msh homeobox 1 (MSX1) were also downregulated by Noggin treatment (Fig. 4F-J). Consistent with our results, studies have also shown that ID1 and MSX1 can restrict myogenic gene expression, and impeded myoblast differentiation [29, 30]. Therefore, our results indicated that Noggin facilitates the skeletal myogenic differentiation of DPSCs via Smad/Pax7 pathway.
Noggin-pretreated DPSCs combined Matrigel can effectively repair muscle injury in VML
To test the utility of DPSCs in repairing muscle injury, we established a mouse VML injury model that accounted for a loss of approximately 40% of the tibialis anterior muscle (Fig. 5A-C). In the untreated VML-injured group, the volumetric defect remained (Fig. 5D, top panel, large dotted blackline). Shrinkage resulting from the defect area occurred. The deformity of the muscle defect showed the deposition of a thin layer of disorganized, collagenous scar tissue (Fig. 5D top panel, small dotted purple line). The injury defect exceeded the threshold and could not be restored by the endogenous regenerative potential of the skeletal muscle. This indicated the success of VML modeling. Quarta et al. reported a similar mouse model of VML that resulted in the irrecoverable loss of muscle function and structure. Consistent with our model, peripheral fibrotic scarring was observed in place of the excised muscle and also extended into the belly of muscle [31].
Hydrogel biomaterials are widely used in skeletal muscle regeneration as they provide a structural framework for the delivery of cells or growth factors to damaged muscle [32]. Therefore, we used the thermosensitive hydrogel Matrigel to provide a scaffold for stem cell transplantation (Fig. 5D). We observed that the Matrigel group exhibited a reduced shrinkage area and decreased muscle fibrosis (Fig. 5D, second panel, and Fig. 5D). Sicariet al. reported histological changes in VML muscles treated with a biologic scaffold material at different time points and showed that the defect area was infiltrated with cells during repair [33]. We also found some mononuclear cell infiltrate in the defect (Fig. 5D, second panel, small dotted red line).
To examine whether Noggin-pretreated DPSCs could better repair muscle injury than the control conditions, we transplanted DPSCs bioconstructs into the defects of VML muscles (Fig. 5A-C). Quarta et al. treated VML muscles with hydrogel Matrigel reconstituted muscle stem cells, which resulted in muscle tissue formation and fibrotic infiltration [31]. We implanted Noggin-pretreated or untreated DPSCs reconstituted in Matrigel into the defects. Morphometric analysis of muscle cross-sections revealed that, relative to the untreated DPSCs groups (Fig. 5D, third panel), Noggin-pretreated DPSCs groups showed decreased size of defect and scar tissue (Fig. 5D, bottom panel, and Fig. 5E, F). In contrast to the above two groups, which showed irreversible and robust fibrotic scars, DPSC-treated defects consisted of markedly reduced fibrotic tissue surrounded by cells of varying morphologies, including fibrotic, inflammatory, and vessel-like cells, indicating the process of tissue repair (Fig. 5D, third panel, small dotted red line). In contrast, Noggin-pretreated DPSCs might have accelerated this process and facilitated the improved formation of muscle tissue, leaving little cell infiltration (Fig. 5D, bottom panel, small dotted red line).
Noggin-pretreated DPSCs can benefit to muscle satellite cell population and promote myogenic repair
To explore the contribution of grafted cells to muscle injury, immunostainings of muscle cross-sections were performed by Pax7 (satellite cell marker), MyoD (activated satellite cells/myoblasts), human Nucleoli (hNu) and human LaminA/C (specific antibody to track human cells), and Laminin (to identify position within the sarcolemma) (Fig. 5G-N). Pax7/MyoD co-staining revealed that stem cell transplantation increased proportion of activated satellite cells when compared with sham or Matrigel groups (Fig. 5G, H). We also observed that hNu was integrated into the nucleus of regenerated tissue and was located on the Laminin-stained muscle sarcolemma (Fig. 5I, white arrow), that was more readily discovered in Noggin-treated DPSCs groups than DPSCs groups. An increased number of hLaminA/C+/Pax7+ (Fig. 5J, left two columns, white arrow) and hNu+/MyoD+ (Fig. 5J, right two columns, purple arrow) cells were also discovered in Noggin-treated DPSCs groups when compare with DPSCs groups, representing increased donor-derived satellite cells (Fig. 5K-N). These results suggested that Noggin-treated DPSCs had partly benefit to satellite cell population. This benefit might offer support to muscle regeneration in VML injury.