TBI is one of the leading causes of severe disability and mortality for all ages worldwide. Affected patients are often accompanied by resultant motor or cognitive dysfunction, leaving devastating effects on the ability to continue with a normal life. As a potential treatment strategy, stem cell therapy has received much attention over the years. Neural stem cells are perhaps the fundamental choice for transplantation, but the availability is somehow limited, and there are not always enough cells for therapy [26, 27]. Interestingly, mesenchymal stem cells from bone marrow, adipose tissue, and umbilical cords showed an attractive therapeutic effect by improving the impaired function in animal models [2830]. However, as one of the important components of MSCs, it remains unclear whether odontogenic stem cells could benefit TBI as well. Until now, with mesenchymal stem cell characteristics, odontogenic stem cells that have been well characterized are as follows: dental follicle progenitor cells (DFPCs), dental pulp stem cells (DPSCs), stem cells from apical papilla (SCAP), periodontal ligament stem cells (PDLSCs), and stem cells from exfoliated deciduous teeth (SHED) [31]. Among them, SHED strongly express NANOG, SOX-2, and OCT-4, all of which are embryonic stem cell markers, and have been reported to exhibit outstanding craniomaxillofacial tissue regeneration capacities [32]. The current study was designed as a first attempt to use SHED for the administration in TBI.

In our present in vitro study, the concentration of inflammatory factors TNF-α and IL-6 was evaluated with or without SHED treatment. It was found that SHED could significantly reduce the secretion of inflammatory factors by activated microglia. Previously, the administration of bone marrow mesenchymal stem cells in a TBI rat model also demonstrated the anti-inflammatory functions, which is in line with our results [33]. Microglia have been reported as the most important resident immune cells in the central nervous system [34]. After TBI, microglia were soon activated to release pro-inflammatory factors, for example, IL-6 and TNF-α, which may in turn exacerbate brain damage. The reduced release of these inflammatory factors by SHED treatment may be beneficial in TBI by inhibiting neuroinflammation. Nitrite is the end product of NO, which plays neurotoxic roles after TBI. Therefore, the decreased nitrite concentration in the SHED co-culture group may also account for the protective effects found in TBI.

The present study demonstrated that after co-culturing SHED with microglia for 48 h in the transwell system, the concentration of nitrite and inflammatory factors IL-6 and TNF-α all decreased compared with the single activated microglia group. As is reported previously, MSCs could secrete anti-inflammatory growth factors [35, 36]. While increasing data suggest that exosomes derived from mesenchymal stem cells was an also an important player in repairing tissue damage [37]. Therefore, we hypothesize that exosomes produced by the co-cultured SHED (SHED-Ex) may contribute the decreased secretion of inflammatory factors by microglia. To clarity if and to what extent SHED-Ex influence the co-cultured microglia, we designed the following experiments. We found that purified SHED-Ex was alone able to alter the polarization of microglia, and inhibit the inflammatory effects of M1 microglia. Therefore, we concluded that SHED-Ex, independent of the anti-inflammatory growth factors, could reduce inflammation, to an even larger extent than the reduction effect that was found in co-culture systems. It is reported that exosomes transport RNAs, proteins, and lipids to the targeted cells to reprogram cell behaviors [38]. Among them, noncoding RNAs, such as miRNAs or LncRNAs, have been reported to play crucial roles. Therefore, further studies are warranted to focus on specific noncoding RNAs that dominate the process. Once the specific molecule is revealed, the therapeutic efficiency will be enhanced by selective manipulation of the expression.

It has been demonstrated that microglia-mediated neuroinflammation plays a critical role in secondary brain injury in TBI [13]. In the present study, SHED-Ex could significantly reduce the pro-inflammatory microglia M1 phenotype cell markers; more importantly, it could do so in a dose-dependent manner. To further certify the role of SHED-Ex on microglial polarization, the mRNA levels of microglia M1/M2 phenotype markers were detected. As we hypothesized, after incubation with SHED-Ex for 48 h, a group of microglia was polarized from the pro-inflammatory phenotype into the anti-inflammatory phenotype, which led to a welcoming restoration after neuroinflammation. In this case, SHED-Ex were further investigated to detect the neuroprotective roles they may have played when administered to rats with TBI.

In the current study, 500 μg/ml SHED-Ex injected into rat brain rescued the cortical damage and improved the motor deficits resulting from TBI in rats. Actually, it was almost equivalent to the effect that occurred with a 105/3 μl SHED treatment for TBI, while 1000 μg/ml exosomes provided a better functional recovery. Therefore, the dose-response efficacy was determined. However, whether a higher dose could provide better efficacy is unknown. Additionally, CD68+ microglia were used to identify activated M1 microglia after TBI. Compared with the TBI group, 1000 μg/ml SHED-Ex could significantly suppress the CD68+ microglia, which suggested the potential therapeutic mechanisms of SHED-Ex transplantations. In addition, we cannot ignore the possibility of how SHED-Ex may also act, such as the benefits of direct or indirect neurovascular regeneration. Future investigations are therefore planned to determine whether neurovascular regeneration could also be manipulated by SHED-Ex.

We combined the above data to explain the therapeutic mechanism of SHED-Ex. At the very beginning of TBI, microglia become more polarized towards M1 activation states over M2. By releasing pro-inflammatory cytokines as well as free radicals, M1 microglia-mediated chronic neuroinflammation exacerbated neurological impairments. In contrast, when SHED-Ex were added, M2 microglia were robustly activated, and thus the neuroinflammation was suppressed by anti-inflammatory cytokines. Long-term functional recovery and reduced neurodegeneration are therefore developed (Fig. 

7

).

Fig. 7

Schematic of SHED-Ex repair of impaired CNS: SHED-Ex shifts microglia M1/M2 polarization. SHED-Ex contact with activated microglia to promote robust M2 polarization. As a result, anti-inflammatory cytokines were released to repair. CNS central nervous system, SHED-Ex stem cells from human exfoliated deciduous teeth-originated exosomes, TBI traumatic brain injury