Stem cells are a research hotspot in tissue engineering in recent years. Stem cells usually come from bone marrow[25].Because of their self-renewal properties, plasticity of differentiation potential, immunosuppressive and anti-inflammatory functions, stem cells have gradually become the key to cell therapy and are applied in regenerative medicine. However, the difficulty of obtaining stem cells limits their research applications. Therefore, it is essential to seek efficient and large number of stem cell sources.
In recent years, the first batch of PDLSCs were isolated from in vitro tissues by cloning technology[26],PDLSCs have many characteristics, such as three-way differentiation ability, immunomodulatory function, anti-apoptotic effect, etc., which make them become a potential cell source for tissue engineering rapidly. Studies have found that PDLSCs have strong survival, proliferation and differentiation abilities in vitro, and can also stimulate their regenerative potential after transplantation in vivo[13], Therefore, the use of PDLSCs seems to be a promising approach. It was found that the grafted PDLSCs can produce connective tissue like structures, bone and collagen fibers in the periodontal ligament without adverse effects[27], Park et al. transplanted human PDLSCs into the back of immunodeficient mice to form complexes with the most commonly used calcium phosphate scaffolds HA and beta-tricalcium phosphate (HA/TCP) and found that they expressed periodontal tissue-like structures in the subcutaneous tissues[28], Feng et al. reported that autologous human PDLSCs could achieve stable regeneration effect in vivo[29], All these experiments showed that PDLSCs as alternative therapy can be used to repair defects with stable effect and few ethical complications. More importantly, PDLSCs are easy to obtain, and can be collected through minimally invasive separation of periodontal tissue in clinical flap surgery, root surface leveling surgery and standard dental cleaning, or can be obtained as clinical waste. The process of collecting PDLSCs from the clinic is minimally invasive or even non-invasive, and the small or even non-wound surface allows the donor to heal without scar. The unique advantages of ethics may be an important reason for selecting PDLSCs for clinical application in the future.
MF plays a critical role in both the proliferative and remodeling stages of wounds. MF regenerates connective tissue by secreting and organizing extracellular matrix components, including different types of collagen, proteoglycans, and signaling molecules[30],Therefore, the origin, activity, apoptosis and function of MF may affect the wound healing process[31].Whether hPDLSCs can perform myofibroblast differentiation is the basis of whether they can be applied in wound healing and repair. In previous studies, α-SMA is the most commonly used molecular marker of MF, but this molecular marker is also highly expressed in other cells, such as smooth muscle cells. Considering this, COLIA1 was also added in this study as an additional differentiation indicator to demonstrate the transformation of hPDLSCs into myofibroblasts. In addition to hPDLSCs, the increase of COL1A1 can also be observed when other cells differentiate into myoblasts, which has a positive significance for the formation of extracellular matrix in wound healing[32].
Generally, the recognized cell algebra used for experiments is P3-7. Based on previous experimental experience, this study extracted hPDLSCs and cultured them to the third passage before they were applied to experiments, and P7-11 was used to show them more clearly. When viewed under a microscope, the cells change from a typical fibrous spindle-shaped shape to a larger, flat, broad shape, and expressed muscle fibroblast markers and showed high contractile activity, suggesting that PDLSCS could conduct myofibroblast differentiation, which was similar to the experimental results of Kengo Iwasaki et al. [24].Periodontal membrane stem cells changed from fusiform to MF-like stress-stretching morphology accompanied by increased levels of markers related to myofibroblasts differentiation. Our experiment showed that the cell morphology changed significantly from the 9th passage, which was also similar to previous research results[24]. With the increase of cell algebra in H-DMEM medium, the expression level of α-SMA and COLIA1 was increased, indicating that hPDLSCs had a trend of myofibroblast differentiation during the in vitro expansion process.
On this basis, we found that PDLSCs using α-MEM medium had slightly better differentiation ability than H-DMEM medium, but the difference was not significant (P < 0.05). The comparison of medium showed that α-MEM medium supplemented various amino acids, ascorbic acid, biotin, ribonucleoside, deoxyribonucleoside and vitamin B12. These components are important factors in cell survival clonal growth, energy metabolism, and cell protection[33], which may explain some of the results. Similarly, Im-Hee Jung et al. compared the effects of α-MEM and DMEM, the two most widely used culture-medium at that time, on the proliferation ability of hPDLSCs and the maintenance of stem cell characteristics, and the results showed that: hPDLSCs cultured in α-MEM medium proliferated more than those cultured in DMEM medium and showed higher ALP activity and more mineralized nodules, but there was no significant difference between the levels of stem cell surface markers[34].In view of the fact that this experiment was only to explore the ability of PDLSCs on wound healing, and only the ability of single myofibroblast differentiation will be discussed later, and H-DMEM medium is more suitable for the survival of PDLSCs in a high glucose environment, we discussed the use of H-DMEM medium for further discussion to verify whether PDLSCS can resist AGEs injury. It is worth mentioning that if the multi-differentiation potential of PDLSCs is needed for wound repair, a slightly better α-MEM medium should be selected to obtain better fiber and bone tissue repair.
In the process of body damage, various factors in the microenvironment interact with each other. The high sugar environment in diabetic patients seriously affects the process of body damage repair. The changes of ROS and hypoxia in adjacent cells during high glucose levels are the main reasons for the changes in stem cell signaling. The production and persistence of ROS will severely inhibit antioxidant enzymes and non-enzymatic antioxidants in various tissues, further aggravate oxidative stress[35], affect the proliferation of stem cells, and thus slow down tissue repair and healing, and damage the innate tissue repair mechanism and regeneration[36].Previous studies have shown that diabetes can cause dysfunction of mesenchymal stem cells in vivo by affecting cell vitality, proliferative ability, dryness and pluripotency[37][38],Harsh hyperglycemia and inflammatory environment will damage the survival and secretion of mesenchymal stem cells[39], which cause certain difficulties in treatment. Therefore, it is necessary to explore whether the microenvironment of AGEs affects the differentiation ability of hPDLSCs, especially myofibroblast differentiation, which is related to whether hPDLSCs can play a role in diabetic wound healing.
In our experiment, we found that compared with the control group, the cell morphology of the age-induced group showed significant changes, such as cell hypertrophy, significantly reduced fibrous structure, and significantly reduced cell number. The proliferation ability of PDLSCs gradually decreased with the increase of the concentration of AGEs after treatment with different concentrations of AGEs, suggesting that the proliferation ability of stem cells was greatly affected by high sugar. In order to further verify the effects of AGEs on PDLSCs, real-time quantitative PCR and western blot were used to detect the expressions of α-SMA, COL1 and vimentin, which are related to myofibroblast differentiation ability. The results showed that for PDLSCs, The expression of α-SMA and vimentin in myofibroblasts decreased after AGEs injury. After the intervention of AGEs for 4d and 8d, we found that the expression of COLIA1 in hPDLSCs decreased, indicating that AGEs caused certain damage to the collagen synthesis ability of hPDLSCs, which indicated the effectiveness of the model. The activation of classical Wnt/ 𝛽-catenin pathway is one of the mechanisms by which AGEs lead to the decrease of COLIA1[40].On this basis, we detected the expression level of α-SMA, a marker related to myofibroblast differentiation, and WB results showed that the protein expression level of α-SMA was slightly decreased for 4 days and 8 days after treatment of AGEs, but the difference was not statistically significant. These results indicate that hPDLSCs may have a certain ability to resist the damage of AGEs, and their ability to differentiate myofibroblasts is not significantly inhibited, which further indicates that although AGEs can reduce the level of COLIA1 in hPDLSCs, hPDLSCs still have the potential to be applied in diabetic wound healing. Therefore, we can conclude that hPDLSCs have a certain ability to tolerate the damage of AGEs. Although AGEs inhibit the collagen synthesis ability of hPDLSCs, they have no significant effect on the myofibroblast differentiation ability of hPDLSCs.
AGEs deposition in high glucose environment causes difficulty in wound healing in diabetic patients, and the difficulty in healing and repair of injury limits the clinical treatment. Common treatment strategies for diabetic wounds include debridement, infection model, lower limb revascularization, decompression, and ultrasound[41].However, currently available treatment methods are still limited in terms of therapeutic effect, especially for oral inflammation healing in diabetic patients, which is more difficult and complicated. Therefore, we begin to pay attention to periodontal ligament stem cells that have not yet become therapeutic cells, and pay attention to the application of such stem cell therapy in chronic wound healing, in order to provide theoretical basis for oral clinical diagnosis and treatment. Studies have found that animals with diabetes show higher levels of inflammation, more apoptotic cells, and less fibroblast proliferation[42],Epithelial and connective tissue repair is delayed by increased apoptosis and decreased cell proliferation[43],Therefore, the periodontal ligament stem cells selected in this study are worthy of further study on their various changes and tolerance to high glucose environment in vivo. Our in vitro cell experiments confirmed that the proliferation of hPDLSCs was inhibited under the intervention of high glucose medium combined with AGEs, but the myofibroblast differentiation ability of hPDLSCs was not significantly inhibited, suggesting that hPDLSCs may play a role in diabetic wound healing. In order to clarify this possibility and provide experimental basis for clinical stem cell application, we constructed a diabetic mice skin wound model, injected hPDLSCs into the wound, and finally observed the wound healing through a series of experiments including HE staining, Masson staining and immunohistochemical staining.
The skin wound model of diabetic mice was established, and the cultured hPDLSCs were injected into the wound bed. The effectiveness of hPDLSCs transplantation was evaluated by the wound closure area, inflammatory infiltration, collagen deposition and arrangement. It should be noted that through the previous study, we found that the survival rate of P7 cells was higher than that of P11 cells under the intervention of hPDLSCs in H-DMEM medium at the same concentration of AGEs. This indicated that although P11 had higher myofibroblast differentiation ability compared with P7, P11 had lower survival ability in the diabetic microenvironment. Therefore, P7 cells were selected for subsequent animal experiments.
Our results showed that a circular wound was established on the skin of diabetic mice after successful modeling, and the wound was obvious on the 3rd day after surgery, with no redness and swelling of the surrounding mucosa. It is worth noting that the appendage (hair follicle, etc.) of the dermis after diabetic wound healing was more difficult than that of normal mice. However, according to the comparison of the width of the wound bed in each figure, it was found that the width of the wound bed in the hPDLSCs treatment group was significantly smaller than that in the vehicles group and the model group, indicating that hPDLSCs have the potential to repair the dermis of the wound and thus achieve the reduction of the wound. From the perspective of wound closure speed, the healing speed of the hPDLSCs group was significantly faster than that of the model group. HE results showed that 14 days after surgery, compared with the hPDLSCs group, there was still more inflammatory cell infiltration in the model group, which suggested that the inflammatory period in the wound healing process of the model group might be prolonged, that is, the proliferation period entered at a relatively late time, which seemed to be related to the result of epidermis and dermis separation. At Day 21, more fibroblasts parallel to the epidermis were observed in the dermis of hPDLSCs compared with those in the model group. We suggest that this may be related to the paracrine mechanism of hPDLSCS[44],After injection into the wound, hPDLSCs can release cytokines such as TGF-β1 that stimulate fibroblast proliferation[45].Masson staining showed that the collagen density in the hPDLSCs group was higher than that in the model group at Day 14, and the collagen arrangement was more orderly, with abundant blood vessels in the base. This result indicated that, although there was no significant difference in the wound size between the model group and the hPDLSCs group at Day14 from the surface, the effect of hPDLSCs was superior to that of the model group in the healing of the dermis and basal tissues.
It is worth noting that our experiment found that the α-SMA content was the highest in streptozolin-induced diabetic mice on Day14. At the 21th day after surgery, immunohistochemistry showed that the expression of α-SMA in the hPDLSCs group was decreased, while it was still increased in the model group. This suggests a delay in myofibroblast differentiation during the early healing stage of diabetes, which is similar to the conclusions of Retamal et al[46].These findings suggest that the mechanism of promoting healing may be related to the emergence of α-SMA, and also indicate that MF does not persist and highly express α-SMA during the whole process of wound healing. MF is highly expressed in α-SMA, but its differentiation is delayed in diabetic wounds. With the passage of healing time, α-SMA is still highly expressed at a later time point. In normal wounds, MF appears early in healing and is reduced during tissue remodeling, but MF activity should be reduced or eliminated by apoptosis during late healing[47].