3,3'-Diindolylmethane stimulates exosomal Wnt11 autocrine signaling in human umbilical cord mesenchymal stem cells to enhance wound healing.

Human umbilical cord-derived mesenchymal stem cells (hucMSCs) are suggested as a promising therapeutic tool in regenerative medicine, however, their efficacy requires improvement. Small molecules and drugs come up to be a convenient strategy in regulating stem cells fate and function. Here, we evaluated 3,3'-diindolylmethane (DIM), a natural small-molecule compound involved in the repairing effects of hucMSCs on a deep second-degree burn injury rat model. HucMSCs primed with 50 μM of DIM exhibited desirable repairing effects compared with untreated hucMSCs. DIM enhanced the stemness of hucMSCs, which was related to the activation of Wnt/β-catenin signaling. β-catenin inhibition impaired the healing effects of DIM-primed hucMSCs (DIM-hucMSCs) in vivo. Moreover, we demonstrated that DIM upregulated Wnt11 expression in hucMSC-derived exosomes. Wnt11 knockdown inhibited β-catenin activation and stemness induction in DIM-hucMSCs and abrogated their therapeutic effects in vivo. Thus, our findings indicate that DIM promotes the stemness of hucMSCs through increased exosomal Wnt11 autocrine signaling, which provides a novel strategy for improving the therapeutic effects of hucMSCs on wound healing.

Introduction

The skin is the largest barrier in the human body against the external environment. Cutaneous wound healing is a critical process to restore skin integrity after an injury, which requires a cascade of biological interactions involving cell differentiation, migration, proliferation, and remodeling 1-3. Burn injury is a common cause of skin damage. Skin burn remains a major cause of high morbidity and mortality. A deep burn is unlikely to heal within 3 weeks and will heal with scarring 4. Although novel biomaterials and biomedical devices have been applied in treating skin burn, the cure rate of most clinical applications remains 50%-60%. Therefore, it is necessary to develop novel approaches to accelerate damage repair after burn injury.

Mesenchymal stem cells (MSCs) have considerable potential in regenerative medicine for their ability of renewal and differentiation into distinct cell types 5-9. MSCs can be isolated from the bone marrow, umbilical cord, adipose tissues, and other adult tissues 9-11. The easy availability of hucMSCs makes them a favourable candidate for injured tissue repair. hucMSCs are established to exhibit positive functional outcomes in various diseases 12. Our previous studies have revealed that hucMSCs attenuate renal and hepatic injury in animal model 13-17. In addition, hucMSC-derived exosomes (hucMSC-ex) displayed potent therapeutic effects on a deep second-degree burn injury rat model and can control stem cells expansion after a regenerative response to prevent tissue from overcrowding by modulating YAP to orchestrate controlled cutaneous regeneration 18, 19. Previous studies have reported that modified MSCs can play enhanced therapeutic roles in cardiac dysfunction and other diseases 20-23. Small-molecule compounds have been particularly used as tools to manipulate stem cell fate and function 24, 25. For modulating stem cell function, small-molecule compounds have several advantages, including the convenience in modifying their concentration, working duration, and rapid and reversible working effects 26.

DIM is a natural compound harvested from cruciferous vegetables, and it is the in vivo dimeric product of indole-3-carbinol (I3C), which belongs to the class of indole glucosinolate 27. Both I3C and DIM have been reported to exert an antitumour property against various human cancers in vitro and in vivo by regulating cancer cell apoptosis and proliferation, cell cycle, the epigenetic status, and the key signalling pathways in tumour progression 28-30. DIM has also been investigated for its potential in tissue injury repair, such as liver and acute lung injury 31, 32. We previously demonstrated that a low level of DIM enhances the stemness of gastric cancer cells to promote tumour progression 33. However, whether DIM can modify hucMSCs to improve their therapeutic effects on wound healing remains unknown.

In this study, we investigated the role of DIM in hucMSC modification for wound healing. We demonstrated that DIM induces exosomal Wnt11 autocrine signalling to activate β-catenin signalling and reinforces the stemness of hucMSCs. DIM-hucMSCs showed a desirable repairing ability in a deep second-degree burn injury rat model.

Materials and methods

All experimental protocols were approved by the Medical Ethics Committee of Jiangsu University (2012258).

Cell culture

HucMSCs were obtained from the affiliated hospital of Jiangsu University with the permission of mothers and were freshly processed within 6 h. HucMSCs were isolated as previously described34 and maintained in a low-glucose Dulbecco's modified Eagle medium (DMEM) containing 10% foetal bovine serum (FBS; Bovogen Biologicals, South America). HaCAT cells were purchased from American Type Culture Collection and cultured in a high-glucose DMEM with 10% FBS. The culture medium was changed every 2 days. Dermal fibroblasts (DFL) cells were isolated from rat back skin as previously described 18.The cells were passaged every 3 or 4 days, and those passaged at 3-6 were used for additional studies.

Preparation of DIM-hucMSCs and the conditioned medium

DIM was dissolved in dimethyl sulfoxide (DMSO) to prepare a 500 mM solution and diluted in DMEM to the working concentration. The cells treated with 0.1% DMSO serving as the control. HucMSCs were treated with different concentrations of DIM for 48 h and subsequently collected for following experiments. For preparing the conditioned medium (CdM), the culture supernatant of the treated hucMSCs were discarded, and a fresh medium was added for an additional 48 h.

Rat skin wound model and treatment

The deep second-degree burn injury rat model was established as previously described 18. After inducing burns at 80°C for 8 s, the wounds were administered immediate cell therapy by injecting 1 × 106 cells (with or without DIM) suspended in 200 μL of phosphate-buffered saline (PBS) at three sites. The rats in the control group were injected with 200 μL of PBS and were housed separately. Two weeks later, all the rats were sacrificed, and the injured area was analysed for further investigation.

Real-time reverse transcription polymerase chain reaction

The total RNA was isolated from hucMSCs and skin cells by using the Trizol reagent (Invitrogen). Two microgram aliquots of RNAs were synthesised according to the manufacturer's protocol (Vazyme). Real-time polymerase chain reactions (PCR) were performed using the QuantiTect SYBR Green PCR kit (Toyobo). The primer sequences are listed in Supplementary Table 1.

Western blot

The cells were lysed in a radioimmunoprecipitation assay buffer with proteinase inhibitors. An equal amount of protein samples was loaded and separated in a 12% sodium dodecyl sulfate polyacrylamide gel electrophoresis gel. Following electrophoresis, the proteins were transferred to a polyvinylidene fluoride membrane, subsequently blocked with 5% (w/v) nonfat milk, and incubated with primary antibodies at 4 °C overnight. The primary antibodies were anti-Sox2 (Millipore, USA); anti-Oct4 (SAB, USA); anti-SALL4 (Abnova, USA); anti-β-catenin (Cell Signaling Technology, USA); anti-CK19 and anti-PCNA (Bioworld, USA); anti-Wnt11 (Santa Cruz, USA); and anti-GAPDH (Kangcheng, China). After washing with Tris-buffered saline and Tween 20 for three times, the membranes were incubated with secondary antibodies (CWBIO, China) at 37°C for 1 h. The target proteins were subsequently visualised using enhanced chemiluminescence.

Immunohistochemistry

Skin tissues were obtained from sacrificed rats and processed into paraffin sections (5-mm thick). Immunohistochemical staining was performed as previously described 35. Furthermore, 1:50 and 1:100 dilutions were used for β-catenin primary antibody and proliferating cell nuclear antigen (PCNA), respectively. The secondary antibodies were purchased from Boster (Wuhan, China). The tissue sections were visualised with 3,3′-diaminobenzidine, counterstained with hematoxylin, and observed through high-power light microscopy (Nikon, Tokyo, Japan).

Osteogenic and adipogenic differentiation in vitro

HucMSCs and DIM-hucMSCs were seeded in 35-mm plates in an osteogenic differentiation medium (0.1 mM dexamethasone, 10 mM β-glycerophosphate, and 50 mg/L ascorbic acid) or adipogenic differentiation medium (Cyagen Biosciences, CA, USA) for 2 weeks, according to the manufacturer's instructions. After the induction, the osteogenic and adipogenic potential was evaluated through alizarin red and oil red O staining, respectively.

Lentiviral knockdown of Wnt11 in hucMSCs

A lentiviral expression vector containing the Wnt11 shRNA sequence (Sigma) was selected for targeting Wnt11; it was classified as Lenti-Wnt11 shRNA, and Lenti-GFP shRNA was used as the negative control vector. The Lenti-Wnt11 shRNA vectors were generated by ligating the vector Tet-pLKO-puro with Wnt11 shRNA oligonucleotides. The sequences of Wnt11 shRNA oligonucleotides were as follows: forward, 5′- CCGGGCAGTGCAACAAGACATCCAACTCGAGTTGGATGTCTTGTTGCACTGCTTTTTG-3′ and reverse, 5′-AATTCAAAAAGCAGTGCAACAAGACATCCAACTCGAGTTGGATGTCTTGTTGCACTGC-3′. The sequences of the control shRNA were as follows: forward, 5′-CCGGGCAAGCTGACCCTGAAGTTCATCTCGAGATGAACTTCAGGGTCACGTTGCTTTTTG-3′ and reverse, 5′-AATTCAAAAAGCAAGCTGACCCTGAAGTTCATCTCGAGATGAACTTCAGGGTCACGTTGC-3′. A recombinant lentivirus was produced by cotransfecting HEK293T cells with PLKO-GFP shRNA or PLKO-Wnt11 shRNA, PU1562, and PU1563 plasmids by using Lipofectamine 2000 (Invitrogen). The virus-containing supernatant was harvested at 48 h and 72 h post transfection. HucMSCs were transduced with the prepared lentivirus (Lenti-Wnt4 shRNA or Lenti-GFP shRNA) and selected with 1 μg/mL of puromycin (Invitrogen) for 15 days. shRNA expression was induced by adding 80 μg/mL of doxycycline. The efficiency of Wnt11 knockdown was evaluated through real-time quantitative reverse transcription (RT)-PCR and Western blot.

Luciferase reporter activity assay

The CdM of DMSO- and DIM-hucMSCs was prepared for the luciferase reporter activity assay. The TOP- or FOP-Flash luciferase reporter plasmid was cotransfected into HEK293T cells together with the Renilla luciferase gene, which is governed by the human β-actin promoter, to normalise the transfection efficiency. At 6 h post transfection, the prepared CdM was added and incubated for another 24 h. The cells were collected, and activities of both firefly and Renilla luciferases were quantified using the dual-luciferase reporter assay system (Promega). The Wnt reporter activity was determined using TOP-FOP luciferase.

Luminex assay

The concentrations of granulocyte-macrophage colony-stimulating factor (GM-CSF); interferon gamma; interleukin (IL)-6; IL-8; IL-10; IL-17a; interferon-inducible protein-10; monocyte chemotactic protein 1 (MCP-1); regulated on activation, normal T-cell expressed and secreted (RANTES); tumour necrosis factor-α; and vascular endothelial growth factor (VEGF) in hucMSCs were measured using Luminex kits (Millipore), according to the manufacturer's instructions.

Isolation and characterization of exosomes

Exosomes were extracted from the CdM of DMSO- and DIM-hucMSCs. Extraction and purification were conducted as previously described 33. The exosome-free CdM was collected and filtered through a 0.22-μm filter, and the exosomes were stored at -80 °C. The protein content of the exosomes was detected using the bicinchoninic acid protein assay kit (CWBIO, Shanghai, China). Furthermore, exosome morphology was identified through transmission electron microscopy (FEI Tecnai 12, Philips, Netherlands). The NanoSight LM10 system (NanoSight, Amesbury, UK) was used for exosome tracking, distribution, and particle number counting. The expression of CD9, CD63, and CD81, surface markers of exosomes, was determined through Western blot.

Cell scratch assay

HaCAT cells were seeded at the density of 2 × 105 cells/well in six-well plates and incubated at 37 °C in 5% CO2 for 24 h to create confluent monolayers. The monolayers were scratched with a sterile pipette tip. To measure cell mobility, we obtained images from five random fields at 24 h after scratching. The width of the original scratch was calculated using the NIH Image programme (http://rsb.info.nih.gov/nih-image/). The migration ratio was calculated as follows: (the width of the original scratch - the width of the actual scratch)/the width of the original scratch × 100.

Statistical analysis

All the data are shown as mean ± standard deviation. The statistically significant differences between groups were assessed using analysis of variance or t test by using Prism software (GraphPad, San Diego, USA). P < 0.05 was considered statistically significant.

Results

DIM-hucMSCs exerted an improved impairing effect on wound healing in a rat deep second-degree burn injury model

To determine whether DIM modulates the therapeutic effects of hucMSCs on wound healing, we established a second-degree injury rat model and evaluated the impairing ability of DIM-hucMSCs at 2 weeks after cell transplantation (Figure ). We observed that both DMSO- and DIM-hucMSCs favourably healed the wounds (complete re-epithelialization in all the six wounds in the DIM-hucMSC group; five of the six wounds in the DMSO-hucMSC group) compared with the control treatment (complete re-epithelialization in one of the six wounds). However, the scabs in the DIM-hucMSC group had sloughed, and the new skin tissues were exposed. Only a sloughing tendency appeared in the hucMSC group (Figure ). Hematoxylin and eosin staining confirmed that the DIM-hucMSC treatment led to a more favourable impairing effect on the epidermis, dermis, and hair follicle than did the hucMSC treatment (Figure ). PCNA immunochemical staining revealed an increased number of PCNA-positive cells in both DMSO- and DIM-hucMSC groups, and the DIM-hucMSC group had more PCNA-positive cells than did the DMSO-hucMSC group (Figure ). The histological scores of the wounds were also considerably higher in the DIM-hucMSC group than in the DMSO-hucMSC and control groups (Figure. 1e and Supplementary Table ). The relative expression of collagen I to collagen III reflects the scar formation potential. Real-time RT-PCR showed that collagen I:III expression was higher in the DMSO- and DIM-hucMSC groups than in the control group and was much higher in the DIM-hucMSC group (Figure ). Immunofluorescence staining showed that the expression of CK19, a marker of re-epithelialization, highly increased in the DIM-hucMSC group compared with that in the DMSO-hucMSC group (Figure ). β-Catenin activation is essential for wound healing 18. Therefore, we detected β-catenin expression in wounded tissues through immunohistochemical staining. We demonstrated a higher number of β-catenin-positive cells in the DIM-hucMSC group (Figure ). In addition, the results revealed that 50 μM of DIM did not increase the growth and the stemness marker expression level of skin cells in vitro (Figure ) and showed no significant repair effect in vivo (Figure ), which excluded the possibility that DIM directly affected skin tissue recovery. Altogether, DIM-hucMSCs showed a more efficient impairing ability than did hucMSCs in wound healing in vivo.

DIM upregulated the stemness of hucMSCs

DIM has been reported to regulate stem cell function by inducing the expression of stemness transcription factors. To determine the mechanism underlying the improved therapeutic effects of DIM-hucMSCs on a deep second-degree burn injury, we detected the expression of stemness transcription factors though real-time RT-PCR and Western blot. We demonstrated that DIM can dose-dependently induce the expression of Oct4, Nanog, Sox2, and Sall4 in hucMSCs and that 50 μM of DIM exerts the strongest inducing effects on the expression of these factors (Figure ). Notably, we demonstrated that hucMSCs are more tolerant to DIM than gastric cancer cells. MTT assay was conducted to test the effect of DIM on the proliferation of gastric cancer cell lines SGC-7901, HGC-27, and hucMSCs for 48h (gastric cancer cell lines as control). The IC50 of DIM on three cell lines SGC-7901, HGC-27, hucMSC were calculated as 60.743μM, 42.812μM and 158.11μM respectively (Figure ). Furthermore, 50μM of DIM inhibited the expression of the stemness transcription factors in tumour cells, whereas it increased the expression of these factors in hucMSCs (Figure ). Immunofluorescent staining confirmed that 50μM of DIM can enhance the expression of Oct4 and Sall4 in hucMSCs (Figure ). DIM-hucMSCs proliferated faster and formed more colonies than did DMSO-hucMSCs, indicating an increased self-renewal and proliferation ability (Figure ). We next determined the differentiation potential of hucMSCs in each group and demonstrated an increased potential of DIM-hucMSCs to differentiate into adipocytes and osteoblasts in the appropriate CdM (Figure ). The expression of adiponectin and alkaline phosphatase (ALP), markers of adipocytes and osteoblasts, respectively, was consistently elevated in the DIM-hucMSC group (Figure ). The secretion of bioactive molecules is the main mechanism underlying the therapeutic effects of MSCs in tissue injury. We also evaluated the paracrine effects of DIM-hucMSCs and observed that the expression of multiple factors, including GM-CSF, IL-6, MCP-1, and VEGF, was considerably increased in DIM-hucMSCs (Figure ). In summary, DIM enhanced the stemness of hucMSCs.

DIM enhanced the stemness of hucMSCs through β-catenin activation

The Wnt/β-catenin signalling pathway is critically involved in regulating cell stemness36. We subsequently evaluated the activity of β-catenin in DIM-hucMSCs and observed that 50 μM of DIM markedly upregulated the TOP-flash reporter activity in hucMSCs (Figure ). DIM concentration-dependantly increased the β-catenin level in hucMSCs (Figure ). Moreover, immunofluorescent staining confirmed an increased expression and nuclear translocation of β-catenin in 50 μM DIM-treated hucMSCs (Figure ). To further verify the activation of Wnt/β-catenin signalling and its role in regulating the stemness of hucMSCs, we administered ICG001, a selective inhibitor of β-catenin-CREB-binding protein interaction, to block β-catenin activation, which did not affect the DIM-hucMSC cell viabilty under the condition of 50 μM of DIM, 48h treatment (Figure ). The inhibition of β-catenin reversed the DIM-induced enhancement of stemness. Moreover, 50 μM of DIM upregulated the expression of β-catenin and its downstream target genes (cyclin D3 and CD44), which were also abrogated by a simultaneous treatment with ICG001 (Figure ). We subsequently determined the adipogenic and osteogenic potential of hucMSCs in the presence or absence of ICG001. ICG001 significantly reduced the promoting role of DIM in adipogenesis and osteogenesis and expression of adiponectin and ALP in hucMSCs (Figure ). The colony formation assay revealed inhibitory effects of ICG001 on the DIM-induced proliferation of hucMSCs (Figure ). Moreover, ICG001 reversed the altered secretion of multiple factors in DIM-hucMSCs (Figure ). These results indicate that DIM enhanced the stemness of hucMSCs by activating Wnt/β-catenin signalling.

Inhibition of β-catenin activation reversed the improved repairing effects of DIM-hucMSCs on wound healing

To further clarify the role of DIM-induced β-catenin activation in the increased repairing effects of hucMSCs on wound healing, we injected DIM-hucMSCs with or without ICG001 into the injured skin tissues. As expected, DIM-hucMSCs exhibited a decent recovery (complete re-epithelialization in all the six wounds) at 2 weeks after transplantation, whereas the ICG001 co-injection group remained incompletely repaired with an inflammatory status (Figure ). The ICG001 co-injection group showed a relatively weak PCNA expression compared with the DIM-hucMSC group, indicating a reduced cell proliferation ability in the ICG001 co-injection group (Figure ). The histological scores in the ICG001 co-injection group were much lower than those in the DIM-hucMSC group and were comparable with those in the control group (Figure ). The relative expression of collagen I:III was decreased in the ICG001 co-injection group compared with the DIM-hucMSC group, indicating a reduced scar formation potential (Figure ). The ICG001 cotreatment also downregulated the CK19 expression in the injured skin tissues (Figure ). Furthermore, the ICG001 cotreatment reversed the increased β-catenin expression in DIM-hucMSCs (Figure ). Altogether, the improved repairing effects of DIM-hucMSCs on wound healing was reversed by β-catenin inhibition, indicating that DIM upregulated the hucMSC stemness to promote wound healing through β-catenin activation.

DIM induced β-catenin activation in hucMSCs through Wnt11

We subsequently determined the main factors responsible for DIM-induced β-catenin activation. We screened the expression of Wnt family members, including Wnt1, Wnt2, Wnt3, Wnt3a, Wnt4, Wnt5a, Wnt6, Wnt7b, Wnt10b, and Wnt11, in DIM-hucMSCs. The DIM treatment modified the expression of several Wnt molecules, of which Wnt11 showed the most significant increase. The expression of Wnt4, which markedly shares its sequence and activity with Wnt11 37, also slightly increased after the DIM treatment (Figure ). Moreover, DIM dose-dependantly increased the Wnt11 expression (Figure ). To substantiate the role of Wnt11 in DIM-hucMSCs, we knocked down Wnt11 by using shRNA and verified the efficiency by using both mRNA and protein levels (Figure ). Wnt11 knockdown led to a decreased expression of Oct4, Nanog, Sall4, and β-catenin, suggesting that Wnt11 acts as an upstream regulator of the canonical β-catenin signaling pathway (Figure ). Compared with the control knockdown cells (shGFP-hucMSCs), the Wnt11 knockdown hucMSCs (shWnt11-hucMSCs) treated with 50 μM of DIM for 48 h showed a reduced expression of the stemness transcription factors (Figure ). Wnt11 knockdown also impaired the enhanced colony formation (Figure ) and adipogenic differentiation ability of DIM-hucMSCs (Figure ). Furthermore, Wnt11 overexpression led to an enhanced β-catenin expression as well as the Wnt reporter activity (Figure ). Collectively, these results suggest that Wnt11 is critical for DIM-induced β-catenin activation and enhanced stemness of hucMSCs.

Wnt11 knockdown reversed the therapeutic effects of DIM-hucMSCs on wound healing

To assess the role of Wnt11 in the improved repairing effects of DIM- hucMSCs in vivo, we injected shWnt11-hucMSCs and shGFP-hucMSCs into the injured area in rats. The results revealed that Wnt11 knockdown led to a reduced effect on promoting re-epithelialization (Figure ). The number of PCNA-positive skin cells in the shWnt11-hucMSC group decreased compared with that in the shGFP group (Figure ). The wound scores in the shWnt11-hucMSC group were much lower than those the shWnt11-hucMSC group (Figure ). The collagen I:III expression was decreased (Figure ), and the CK19 expression was significantly reduced in the shWnt11-hucMSC group (Figure ). Immunohistochemical staining revealed that the DIM-induced increase of β-catenin expression was abrogated in the wounds in the shWnt11-hucMSC group (Figure ). In summary, Wnt11 knockdown reversed the improved therapeutic effects of DIM-hucMSCs on wound healing.

Wnt11 was transported by exosomes in an autocrine manner

We previously reported that exosomes mediated wound healing through Wnt4 transportation. Considering the low solubility of Wnt11 37, we determined whether DIM induced Wnt11 expression in exosomes. First, we analysed the CdM of DIM-hucMSCs and observed that it can upregulate the expression of the stemness transcription factors in hucMSCs (Figure ). The CdM of DIM-hucMSCs can also enhance the colony-forming ability of hucMSCs (Figure ). Enzyme-linked immunosorbent assay (ELISA) showed elevated Wnt11 expression in the CdM of DIM-hucMSCs (Figure ). We further extracted the exosomes from DMSO- and DIM-hucMSCs, which were termed as DMSO-ex and DIM-ex, respectively. The NanoSight LM10 system was used for exosome tracking, size detection, and particle number counting (Figure ). No difference was observed in the morphology or particle numbers between two types of exosomes after the DIM treatment. Both DMSO-ex and DIM-ex were 100-nm spherical vesicles; had a comparable particle number; and expressed CD9, CD63, and CD81 (Figure ). Exosomes derived from DIM-hucMSCs showed an increased expression of Wnt11, indicating that the DIM treatment affected the content of hucMSC-derived exosomes. We subsequently compared the expression of Wnt11 in exosomes and an exosome-free CdM (ex-free CdM) by using ELISA. DIM promoted Wnt11 secretion in exosomes. Moreover, exosomes showed a relatively higher level of Wnt11 expression than did ex-free CdM, indicating that Wnt11 was mainly present in exosomes rather than in ex-free CdM (Figure ). Compared with DMSO-ex, DIM-ex highly enhanced the colony formation and migration of skin cells. However, ex-free CdM from DIM-hucMSCs only showed slightly promoting roles (Figure ). These results suggest that DIM induce Wnt11 expression in exosomes in an autocrine manner.

Discussion

Burn injury treatments have always been a intractable medical problem. The therapeutic methods so far are function-limited. For instance, biofilms easily lead to a wound infection and initiates inflammatory response. Skin grafts can reduce deaths from infection. However, such injuries are long-lasting and many patients suffer from chronic pain for a long time 39, 40. Cell-based therapies and tissue regeneration are new approach to overcome the present limitations of burn wound healing. Functionally and phenotypically identical cells are generated to replace cells lost to disease and injury would have better specificity compared to conventional therapeutics by producing a supply of immunologically tolerant cells or tissue, leading to a permanent recovery for damaged tissues 26.

MSCs are excellent candidated in cell-based therapy for their promoting roles in tissue repair, particularly in skin burn injury, by increasing angiogenesis, extracellular matrix (ECM) production and remodeling, differentiation of MSCs into endothelial cells (EC), pericytes (PC), fibroblasts and keratinocytes, and also the secretion of exosomes 41. The maintenance of stemness is also related to the rapidly growing, multidirectional differentiation, and paracrine abilities of MSCs, which allows them to facilitate wound healing. The secretion of a broad range of bioactive molecules is now considered to be the main mechanism by which MSCs achieve their therapeutic effect. The loss of stemness, such as ageing, may cause a declined paracrine of VEGF, placental growth factor, and hepatocyte growth factor in stem or progenitor cells and an attenuation of their regenerative potential in cardiovascular diseases and osteoarthritis 42-44. In the present study, we demonstrated that DIM could improve the stemness of hucMSCs and enhance their proliferation, differentiation, and paracrine abilities. The blockade with ICG001 reduced the secretion of the paracrine factors in DIM-hucMSCs, suggesting that DIM enhances the stemness of hucMSCs to make them more powerful and helpful MSCs in improving their repairing effciency.

As for the DIM treated MSCs, there is an increasing interest exists about the role of small molecule drugs in modulating the stem cell fate and function. Chemical approaches will play leading roles inguiding therapeutic developments in regenerative medicine 26. Enabling and improving the generation of cells as well as enhancing their functions by chemical approaches will undoubtedly expand cell-based therapies to address many medical needs.Small molecule drugs are more convenient for providing a rapid and reversible effects by altering their working concentration, duration time, and compositions 26. Small molecule compounds have been widely used in the generation of induced pluripotent stem cells (iPSCs) 36, 45, which has shown predominance in regenerative medicine, such as Parkinson disease and sickle cell anaemia 46, 47. The anticancer drug OHTM could replace some of the Yamanaka factors to generate iPSCs, resulting in safer therapies 48. Moreover, human-induced pluripotent stem cell-derived MSCs have shown a more satisfactory alleviation of heart failure, less cardiomyocyte apoptosis, and fibrosis than naive MSCs in anthracycline-induced cardiomyopathy 20. In addition, the pharmacological priming with ISX1, a 3,5-disubstituted isoxazoles, could enhance the myocardiocyte differentiation ability of adipose-derived stem cells and the persistence of exogenously transplanted cells to alleviate the impaired cardiac function48. The pretreatment of adipose-derived stem cells with curcumin before transplantation also facilitates myocardial recovery through antiapoptosis and angiogenesis 49, 50. These studies indicate that stem cells can be modified to achieve more desirable repairing effects, and small molecule drugs have proven to be potential candidates for stem cell manipulation in regenerative medicine.

DIM is widely reported for its anti-tumour activity. The role of DIM in stem cell manipulation has not been characterised. 50 μM of DIM can inhibit the growth and expression of stemness transcription factors of gastric cancer cells. However, it promotes the growth of hucMSCs and increases the expression of stemness transcription factors, suggesting that stem cells are more tolerant to anticancer drugs, making them potential candidates for pharmacological manipulation. Studies on the crossover of MSCs with chemicals, such as DIM, would supply a potential therapeutic strategy and benefit the regenerative medicine 26.

Wnt proteins are divided into two classes according to their biological activities. The Wnt1/Wg class includes Wnt1, Wnt3, Wnt3a, and Wnt8, which represent the canonical Wnts activating the Wnt/β-catenin pathway; the Wnt5a class includes Wnt4, Wnt5a, Wnt5b, and Wnt11 that represent the noncanonical Wnts activating the Wnt/calcium and Wnt/JNK (PCP) pathways 37, 45. Kestler et al. demonstrate that both canonical and noncanonical Wnts have the potential to activate all the pathways 51. Wnt11 is associated with the canonical Wnt receptor LRP6 to activate the β-catenin pathway. Furthermore, the Wnt11 promoter contains two conserved TCF/LEF binding sites, suggesting that β-catenin signals can directly regulate Wnt11 transcription 52, 53. The depletion of Wnt11 reduces the activity of β-catenin 54. In this study, we demonstrated that DIM induced Wnt11 expression to activate Wnt/β-catenin signalling, leading to enhanced stemness in hucMSCs. Our findings are in accordance with the crosstalk between canonical and noncanonical Wnt/β-catenin pathways. In addition, we demonstrated that Wnt11 knockdown can reverse the increased repairing effects of hucMSCs through DIM, indicating that Wnt11 is crucial for DIM-induced promoting role of hucMSCs in wound healing.

Exosomes are bioactive vesicles released by multiple cell types into the extracellular space. Exosomes act as transporters in cell-cell communication to deliver the message from their original cells to the recipient cells. The contents in exosomes include lipids, nucleic acids (DNA, mRNA, miRNA, and noncoding RNAs) and active proteins 55. Wnt11 is a low soluble molecule, which requires post-translational modifications. Wnt is transferred by binding to the exosomal protein and exosomes are suggested as carriers for Wnt secretion and extracellular traveling 56, 57. Wnts cannot be detected in exosome-free conditioned media of either mammalian or Drosophila cells 57-60. Moreover, Wnt11 has been reported to participate in interfollicular epidermal stem cell self-renewal in an autocrine manner 36, 61. In this study, we provided evidence that exosomes act as crucial transporters of Wnt11 in hucMSCs, and DIM can increase the secretion of exosomal Wnt11 to activate the Wnt/β-catenin pathway in an autocrine manner. Furthermore, increasing evidence of the therapeutic potential of exosomes in several diseases has emerged, including kidney disease, cardiac disease, skin injury. With respect to skin wound healing, HucMSC‐exosome mediated ‐Wnt4 signaling is required for cutaneous wound healing 18. Guo et al. also reported that Exosomes derived from platelet-rich plasma promote the re-epithelization of chronic cutaneous wounds via activation of YAP in a diabetic rat model 62. These findings strongly support the therapeutic potential of MSC-derived exosomes. To enhance the therapeutic effects of MSC-derived exosomes, it will be possible to modify the exosomes. Whether our DIM-hucMSC derived exosomes would have such better repairing effect and the potential mechanisms are intresting research deserve further investigations.

In summary, our findings indicate that DIM, a natural small molecule compound, can enhance the stemness of hucMSCs through exosomal Wnt11 autocrine to mediate Wnt/β-catenin activation and eventually improve the repairing ability of hucMSCs in wound healing.

Supplementary figures and tables.

Click here for additional data file.