hADSCs transplantation improved motor performance and TH expression in the 6-OHDA-induced PD mice
Using the same method as previous described , 6-OHDA or vehicle (saline) was injected into the SNc in situ of mice brain to get the 6-OHDA-induced PD mice model, as well as control mice cohort. APO rotation experiment  was conducted after 1, 2, 3 weeks injection to select the PD mice. Mice with a rotation frequency above 7r/min were selected as PD mice and were used in the following experiments (video S1-S2). To evaluate the potential effects of hADSCs (Fig. 1S) on PD, hADSCs were transplanted in situ into the striatum of PD mice brain and the behavioral tests were performed at the 4th, 5th, and 6th weeks (Fig. 1a). We found that hADSCs grafted PD mice displayed faster moving velocity (F (2,12) = 28.11, p < 0.0001) and further distance (F (2,13) = 19.50, P = 0.0001), while less sedentary resting time (F (2,12) = 21.73, p = 0.0001) than the vehicle treated PD mice (Fig. 1b). Moreover, significant reduction of numbers in APO rotation experiments were observed in hADSCs grafted mice compared with the vehicle treated PD mice (F (2,12) = 130.3, P < 0.0001, Fig. 1c & video S3). These data indicated that hADSCs could improve the motor performances in the 6-OHDA-induced PD mice model.
To further investigate the effective treatment of hADSCs transplantation on 6-OHDA-induced PD mice, brain tissues were harvested at the end of the 6th week when the behavioral tests were completed. Then, western blot and immunohistochemistry were conducted on brain tissues and brain slices respectively. Western blot showed that hADSCs transplantation increased the expression of TH protein, a major marker of dopaminergic (DA) neurons, in the STR nucleus and the VTA + SNc (for STR: F (2,17) = 20.46, P < 0.0001; for VTA + SNc:F(2,6) = 16.09, P = 0.0039; Fig. 1d,e). Consistent with the observation in western blot, histological analyses showed that hADSCs treatment increased the percentage of TH-positive densitometry in STR on 6-OHDA-induced PD mice (6-OHDA-inuced PD mice without hADSCs transplantation: 8.7% ± 9.2%, with hADSCs transplantation: 48.15% ± 9.57%, F (2, 9) = 109.6, p < 0.0001, Fig. 1f). Subsequently, using densitometry analysis, we found that hADSCs transplantation could also increase the TH-positive fibers in VTA and SNc nucleus on 6-OHDA-induced PD mice (F (2, 6) = 73.98, p < 0.0001, Fig. 1g).
hADSCs protected the dopaminergic systems of 6-OHDA-damaged brain slices with an indirect manner
To further explore the origin of rejuvenated DA fibers around the grafts, we used Celltracker CM-Dil (located in the cell membrane and cytoplasm) to label hADSCs followed by the previous transplantation procedures. At 3rd week, brain tissues were harvested for sagittal immunofluorescence staining. We found that CM-Dil-labeled hADSCs were distributed in STR region of brain tissue, where these hADSCs were transplanted in situ of mice brain, while were not observed in the SNc region (Fig. 2a). These observations showing that grafted hADSCs didn’t transfer from STR to SNc region. Moreover, TH staining neurons showed no co-localization with CM-Dil-labeled hADSCs and revealed that hADSCs didn’t differentiate into DA neurons. These data showing that hADSCs protect DA neurons through an indirect manner, while not by transferring to SNc or differentiate into DA neurons.
Next, we evaluated the protective effects of hADSCs on DA neurons via hADSCs co-cultured with PD models-organic brain slices in vitro. Organotypic slice cultures were prepared according to the membrane interface method as previous described . In current study, we harvested the brains from young mice, and cut 350um thick sagittal black striatal brain slices with Tissue Culture Plate Insert and placed four slices on the membrane insert. Brain slices were allowed to recover for 7 days from cutting trauma (Fig.S2). On day 8, brain slices were exposed to combined 6-OHDA (600nM; Fig.S3) for 1h, followed by 1 x 105 hADSCs co-cultureing procedure with serum-free medium for 4 days (Fig. 2b). Subsequently, western blot was conducted to analysis the TH expression on brain slices of hADSCs co-cultured 6-OHDA damaged brain slices and we found that hADSCs effectively prevent the 6-OHDA induced TH annihilate (F (2, 9) = 69.38, p < 0.0001, Fig. 2c). Moreover, 6-OHDA damaged brain slices secrected extracelluar Lactate dehydrogenase (LDH) in culture medium was also tested with LDH assay, and we observed that hADSCs could markedly decrease the LDH release (F (6, 28) = 23.58, p < 0.0001, Fig. 2d). These data indicated that hADSCs could protect the dopaminergic neurons from damage which caused by 6-OHDA in an indirect manner.
Pentraxin 3 was a potential hADSCs secrected protective molecule on DA neurons in PD models.
As hADSCs showed protective effects on PD models both in-vitro and in-vivo through an indirect way, we hypothesized that hADSCs may protect the 6-OHDA damaged brain slices through secrecting extracullar cytokines. To further explore the extracellular mechanism of how hADSCs protected the DA neurons on 6-OHDA-treated brain slices, ater 4 days co-cultivation of hADSCs and 6-OHDA-treated brain slice (Fig. 3a), we collected the hADSCs for RNA-seq high-throughput sequencing and RT-qPCR. Hierarchical clustering heat map showed RNA-seq high-throughput sequencing results of distinct gene expression patterns in hADSCs and hADSCs which co-cultured with brain slices (p < 0.05, Fig. 3b). Co-cultureing with 6-OHDA-treated brain slices lead to 1577 mRNAs appeared differential expression including 276 upregulated and 1301 downregulated (Table. S1). Meanwhile, we collected the conditioned media from hADSCs single culturing group and hADSCs cultured with 6-OHDA-treated brain slice group, as well as 6-OHDA-treated brain slice single culturing group. These conditioned media were then applied to human cytokine array and label-free quantitative proteomics. Human protein cytokine arrays showed that cytokines appeared to a difference between hADSCs cultured conditioned media, 6-OHDA-treated brain slices cultured conditional media and the conditioned media of hADSCs co-cultured with 6-OHDA-treated brain slices (Fig. 3c, Table. S2). In label-free quantitative proteomics, a total of 153 proteins with significant differences were identified (Table. S3). Screening of differentially expressed proteins according to the standard of expression fold change of more than 2.0 times (up-regulation greater than 2.0-fold or down-regulation less than 0.5-fold), a total of 153 differential proteins were selected as target cytokines (Fig. 3d). The results of RNA-seq high-throughput sequencing and Human protein cytokine arrays as well as label-free quantitative proteomics were subsequently subjected to Venn intersection and we found that PTX3 was the exclusive protein which showed remarkable difference in all three tests and analysis (Fig. 3e). The following RT-qPCR showed that the mRNA level of PTX3 in the hADSCs co-cultured 6-OHDA-treated brain slices group was higher than the hADSCs single culturing group (p < 0.05, Fig. 3f). Therefore, these data suggested that PTX3 which secreted by hADSCs may play an important role in the hADSCs transplantation therapy on 6-OHDA-induced PD mice model.
PTX3 treatment mimicked the effect of hADSCs transplantation to improve the motor performances on 6-OHDA-induced PD mice
Successful PD mice model were selected as previous descried and were used in following experiments (Fig. 4a). To evaluate the potential effects of PTX3 on Parkinson’s disease, 1 x 105 hADSCs (n = 7), 1 x 105 si-PTX3 (n = 7, hADSCs with PTX3 knockdown), equal volumes of rhPTX3 (4ul, 0.50 mg/ml, n = 7, Fig. S4) or vehicle (saline) (4ul, n = 7) were injected in situ into the right striatum of PD mice, open field experiment (Fig. 4b) showed that hADSCs, si-PTX3 and rhPTX3(2000ng/ml, Fig. S5) treatment improved the motor performances on 6-OHDA-induced PD mice. Both hADSCs, si-PTX3 and rhPTX3 treatment groups showed further moving distance (F (4, 13) = 12.71, p = 0.0002, Fig. 4c), faster moving velocity (F (4, 13) = 12.70, P = 0.0002, Fig. 4d) and less resting time (F (4, 14) = 25.25, p < 0.0001, Fig. 4e) when compared with vehicle treatment group on 6-OHDA-induced PD mice. Consisted with what were observed in open field experiments, APO rotation test showed that both si-PTX3 and rhPTX3 could decrease the rotation frequency on PD mice. Importantly, APO rotation test also showed that si-PTX3 blunted the behavioral improvement effects of hADSCs on PD mice. Moreover, rhPTX3 administration decreased rotations showing a effect similar to those in the hADSCs grafted mice (F (4, 16) = 37.37, p < 0.0001, Fig. 4f, vedio 4S-5S). These data indicated that the transplantation of hADSCs had a neuroprotective effect in the 6-OHDA-induced mouse model, part of the effect was attributed to PTX3 secreted by hADSCs.
PTX3 treatment mimics the effects of hADSCs transplantation to improve TH expression in vivo and in vitro
By the end of behavioral test, mice were anesthetized and fixed with formaldehyde perfusion, and then the brain tissues of VTA + SNc and STR regions were harvested to immunofluorescence staining for TH. Mice with 6-OHDA administration in the right side showed that there was a remarkable loss of dopamine-containing STR and VTA + SNc neurons at the right side, while 3 weeks after hADSCs were injected into 6-OHDA-induced mice, TH-positive neurons increased significantly. Moreover, rhPTX3 (2000ng/ml, Fig. S5) administration increased TH-positive cells showing a effect similar to those in the hADSCs grafted mice. What’s more, si-PTX3 were injected into 6-OHDA-induced mice, the density of TH positive cells wwas less pronounced than that in the hADSCs group (Fig. 5a-d). These results indicated that PTX3 could mimic the effects of hADSCs on protecting dopaminergic neurons from 6-OHDA neurotoxicity in mice.
PD brain slices are prepared as described above,1 x 105 hADSCs (n = 7), 1 x 105 si-PTX3 (n = 7), equal volumes of rhPTX3 (1000ng, n = 7, Fig. S6) or vehicle(saline) (n = 7) were co-culturing procedure with serum-free medium for 4 days. Then, we performed brain slice immunofluorescence, western blot and LDH test. The data indicated that the same tendency wwas observed in immunofluorescent staining, western blot and LDH test in vitro (F (4, 20) = 46.81, P < 0.0001, Fig. 5e-h, Immunofluorescence video-S). The above results indicated that PTX3 secreted by hADSCs has a protective effect on DA neurons.
PTX3 protected DA neuron through inhibition of apoptotic death-inducing signal complex
Next, we sought to investigate the underlying mechanism of PTX3-mediated neuroprotection effects in parkinson's disease. As apoptosis is one of the most important pathogenesis of PD, we hypothesised that PTX3 may protect DA neurons by inhibiting the apoptotic pathway. To further confirm this, another individual experiments were conducted and the VTA + SNc regions of brains from saline treated with vehicle mice, saline treated in 6-OHDA-induced PD mice, hADSCs transplanted in 6-OHDA-induced PD mice, siPTX3 transplanted in 6-OHDA-induced PD mice and rhPTX3 treated in 6-OHDA-induced PD mice, were harvested for Fluoro-Jade C staining(for the severity of neuronal degeneration), Terminal deoxynucleotidyl transferase dUTP nick end labeling (TUNEL) and TH dual immunofluorescence staining(for apoptosis). We found that the number of Fluoro-Jade C positive cells in VTA + SNc regions of 6-OHDA-induced PD mice injected saline was higher than 6-OHDA-induced PD mice treated with hADSCs, but its’ effect was in part reversed by si-PTX3. More interestingly, there was a similar effect in rhPTX3 injected PD mice to hADSCs-treated PD mice (Fig. 6a). Finally, the same trend could be also obtained in the results of TH and TUNEL co-staining (Fig. 6b).
Since PTX3 is also called tumor necrosis factor (TNF)-inducible gene 14 protein (TSG-14). The TNF family is closely related to exogenous apoptosis. Therefore, we speculated whether PTX3 exerted DA neuron protection by inhibiting the exogenous apoptosis pathway. RT-qPCR and western blot were used to test the changes of exogenous apoptosis related genes and proteins with saline transplanted vehicle mice, saline transplanted 6-OHDA-induced PD mice, hADSCs transplanted 6-OHDA-induced PD mice, rhPTX3 treated 6-OHDA-induced PD mice. RT-qPCR analysis showed that 6-OHDA treatment increased the mRNA level of caspase3 (F (4, 17) = 13.47, P < 0.0001), caspase8 (F (4, 18) = 17.52, P < 0.0001), FADD (F (4, 18) = 20.51, P = 0.0001) and TRADD (F (4, 16) = 32.71, P < 0.0001), where were decreased in hADSCs-treated mice, siPTX3-treated mice, rhPTX3-treated mice. Howerver, si-PTX3-treated could decreased the effect partly, while rhPTX3-injected group having similar inhibitory apoptosic effect as hADSCs-treated group (Fig. 6c). Western blots showed lower expression of apoptotic genes levels in hADSCs-injected mice than in 6-OHDA-injected mice, but these genes were not lower in the si-PTX3 group than in the hADSCs group, which had the lowest among the 3 groups. Although rhPTX3 group was also lower than 6-OHDA group, while rhPTX3 group, while the decrease in rhPTX3 group was not as obvious as that in hADSCs group (Fig. 6d).
Finally, after 4 days co-culturvation, collecting brain slices, q-PCR and western blot were also conduceted, and we could find the trend was the same as in vivo model (Fig. 6e,f). Since DISC (death-inducing signal complex) composed of TRADD/FADD/caspase8 induced a downstream cascade reaction. Thus, the above data indicated that PTX3 could protect dopaminergic neurons by inhibiting exogenous apoptosis similar to hADSCs.