Cultures and Characterization of hAECs
In order to establish the hAECs cultures, hAECs were isolated according to a modified protocol as previously described [28]. hAECs were derived from the epithelium layer of placenta and cultured in DMEM/F12 containing 10% FBS. In the cultures, hAECs displayed an epithelial morphology (Fig.S1A). Moreover, hAECs expressed high levels of Epcam, CD49f, CD29 and CD73, but much less levels of CD31, CD34, CD45 and human leukocyte antigen HLA-DR. The results indicated that hAECs isolated have a high purity and are useful for the following experiments.
hAECs Grafts Attenuate MPTP-Induced Neurodegeneration and Ameliorate Behavior Deficits of PD Mice
To determine if hAECs that we prepared attenuate MPTP-induced neurodegeneration and ameliorate behavioral deficits of PD mice, PD mice were induced by intraperitoneal injection of MPTP for consecutive six days. Then hAECs were transplanted into the unilateral striatum of PD mice at day 7 (Fig.1A). MPTP induced degeneration of nigrostriatal dopaminergic neurons, as evidenced by a significantly reduced number and neurite fragmentation of TH-positive cells in the SN and a lower density of the striatal axon fibers of mice (Fig.1, MPTP+PBS panel). MPTP treatment caused 35.6% ± 1.2% loss of TH-positive cells in the SN compared to that of normal CON group at 4 weeks post-grafting (Fig.1C). Interestingly, 4 weeks after post-grafting, a significantly higher number of nigral TH-positive cells were observed in the “MPTP+hAECs” group compared to that of “MPTP+PBS” group (Fig. 1B,1C). In addition, hAECs grafts alleviated MPTP-induced the neurite fragmentation of nigral dopaminergic neurons (Fig. 1D). Furthermore, the density of the striatal axon fibers in the PD mice was enhanced after hAECs transplantation (Fig. 1E, 1F). Remarkably, the hAECs grafts ameliorated MPTP-induced motor deficits, as measured by the rotarod test at 2 weeks after transplantation (Fig. 1G), but hAECs did not display an ameliorate effect at 1 week post-grafting (data not shown). Taken together, the results suggested that hAECs attenuated MPTP-induced motor deficits and promoted the survival of dopaminergic neurons and outgrowth of their neurites and projected striatal axon fibers.
Survival of Transplanted hAECs
To evaluate the survival of transplanted hAECs, they were prelabeled with PKH26 red fluorescent dyes and then transplanted into unilateral striatum of MPTP-induced PD mice. As shown in Fig. 2, the grafts could survive at least four weeks after transplantation, as demonstrated by human nuclei (hNuclei)-immunoreactive cells observed in the unilateral striatum of PD mice. Interestingly, the number of hNuclei positive cells in the striatum at 2 weeks post-grafting were higher than that at 4 weeks post-grafting (Fig 2), which indicated that hAECs were not overgrown. Surprisingly, double positive for hNuclei and TH double were rarely been observed (data not shown), suggesting that hAECs are difficult to differentiate into dopaminergic neurons 4 weeks post-grafting.
hAECs Promote Survival and Neurite Outgrowth of Mesencephalic Dopaminergic Neurons Lesioned by MPP+
To further illustrate the neuroprotective effects of hAECs on MPTP-induced PD mice , we treated primary mesencephalic neurons with MPP+ (20μM), an active toxic metabolite of MPTP, which is selectively taken up into dopaminergic neurons and induces toxicity to dopaminergic neurons [31]. After treatment with MPP+, damaged dopaminergic neurons were demonstrated by the reduced number and neurite fragmentation of TH positive neurons (Fig.3). Interestingly, when hAECs were co-cultured with lesioned mesencephalic neurons for 7 days, hAECs promote the survival and outgrowth of TH positive cells, as evidenced by the presence of more TH positive cells, longer neurites and more neurite branch points per TH positive neuron compared to those of the MPP+ group (Fig. 3A, 3B). On the other hand, hAECs were reported to produce a number of neurotrophic factors [20], which have neuroprotective effects on damaged neurons [22-24]. Thus, we wondered if the hAEC-CM also had neuroprotective effect on primary mesencephalic neurons lesioned by MPP+. As shown in Fig. 3C and 3D, hAEC-CM indeed displayed a pro-survival effect on lesioned dopaminergic neurons. In contrast, the control group (the CM of human foreskin fibroblasts, hEF-CM) did not show such an effect on dopaminergic neuron (data not shown).
Analysis of Paracrine Factors
To define what kinds of neurotrophic factors, growth factors and other factors beneficial to lesioned dopaminergic neurons in the hAEC-CM, a human cytokine antibody array 507 was performed with hAEC-CM and hEF-CM (as a control), respectively. Relative expression levels of 507 human target proteins can be simultaneously detected, including cytokines, chemokines, growth factors, angiogenic factors, soluble receptors, soluble adhesion molecules and other proteins in cell supernatants. We analyzed the human antibody array in both hAEC-CM and hEF-CM. The data were shown in the Table S1. The heatmap shows that the relative expression levels of most of secretory factors in the hAEC-CM were higher than those of hEF-CM (Fig.4A). Next, we further performed gene ontology (GO) function enrichment analysis of biological process and found that the biological processes related to cell migration and chemotaxis, positive regulation of locomotion, cellular component movement, response to external stimulus and tyrosine modification activity in the hAEC-CM were significantly upregulated compared to those of hEF-CM (Fig. 4B). These results indicated that hAECs secret some factors contributing to the migration of cells and the microenvironment at the lesioned site and the regulation of repair process. Furthermore, KEGG pathway enrichment analysis showed that several pathways including chemokine signaling, cytokine-cytokine receptor, JAK-STAT, PI3K-Akt signaling, MAPK signaling, Ras signaling, Rap1 signaling and TGF-beta signaling are activated in the hAEC-CM. Interestingly, these pathways are related to cell migration and homing, neuronal cell differentiation, the survival and growth of neurons, neurite outgrowth, anti-oxidative stress, anti-apoptotic and anti-inflammatory pathways[22], the data revealed their function on these biological processes (Fig. 4B).
Subsequently, we mainly focused on the neurotrophic factors, growth factors, cell adhesion molecules and anti-inflammatory factors as these four types of factors have been implicated previously involved in PD[22] [32, 33]. As shown in Table S2, indeed, the levels of these factors in hAEC-CM were significantly higher compared to those of hEF-CM and were displayed as a heatmap (Fig.4C). In particular, BDNF, Ciliary neurotrophic factor (CNTF), oncostatin M (OSM), neuronal cell adhesion molecule (NrCAM) and anti-inflammatory factors IL-1ra, a natural inhibitor of IL-1, were much high (see Table S2, fluorescence signal intensity at least 10000), implicating their importance in the neuroprotection.
Antibody Neutralizing Experiments
Next, to determine whether these above-mentioned factors are important for the survival of dopaminergic neurons and the outgrowth of their neuritis, we performed antibody neutralization experiments using blocking antibodies against BDNF, CNTF, OSM, granulocyte-macrophage colony stimulating factor (GM-CSF), neuronal cell adhesion molecule (NrCAM), a natural inhibitor of IL-1 (IL-1ra), IL10 and IL13, respectively. First, primary mesencephalic neurons were lesioned by MPP+, then the lesioned dopaminergic neurons were cultured in hAEC-CM containing a corresponding antibody for 4 days. Lastly, the mesencephalic neurons were immunoassayed with a TH antibody. As shown in Fig. 4D, the number of survival dopaminergic neurons and the outgrowth of their neurite were significantly reduced in the present of neutralizing antibodies against BDNF, CNTF, GM-CSF, OSM and NrCAM compared to that of IgG group. Quantification revealed a lower percentage of the dopaminergic neurons bearing neurites, and significantly shorter neurites length and fewer branching points per dopaminergic neuron in the present of neutralizing antibodies against these factors compared to those of IgG group (Fig.4E). In addition, the survival of lesioned dopaminergic neurons was evidently reduced in the present of neutralizing antibodies against anti-inflammatory factors IL-1raand IL13, but not with anti-inflammatory IL-10. In particular, the branching points per dopaminergic neuron was also much lower compared to that of IgG group in the present of neutralizing antibodie against IL-1ra(Fig.4D and 4E). Collectively, hAECs have neuroprotective effects on PD mice attributed to the paracrine factors including neurotrophic factors, growth factors, NrCAM, anti-inflammatory factors.
hAECs Inhibits Microglial Activation and Neuroinflammation of MPTP-induced PD mice
Neuroinflammatory processes including activated microglial and increased concentration of pro-inflammatory factors play an important role in the progression of PD [33, 34]. Here, activated microglia displaying an amoeboid-cell body were evident, and elevated concentration of IL-1b and TGFa was also observed in MPTP-induced PD mice (Fig. 5). Interestingly, 4 weeks after transplantation of hAECs into the PD mice, hAECs attenuated the microglial activation in the SN of PD mice, as validated by the morphology change of microglial cells from a larger and amoeboid cell body with a few ramified patterns to a round cell body with many ramified patterns similar to that in the CON group (Fig. 5A). Besides, the levels of pro-inflammatory cytokines IL-1b and TNFa in the serum of PD mice were also decreased in the MPTP+hAECs group than those in the MTPT+PBS group (Fig. 5B). Consistently, Fig. 4D, 4E also showed that the majority of anti-inflammatory cytokines such as IL-1ra, IL-10, IL-13 in the hAEC-CM were richer and promoted the dopaminergic neuronal survival. Taken together, the data revealed that hAECs play an anti-inflammatory role by inhibiting microglial activation and secreting some anti-inflammatory factors.
hAECs Attenuates MPTP-induced Oxidative Stress
Recent studies suggest that oxidative stress contribute to the loss of dopaminergic neurons in the pathogenesis of PD[35]. MPTP has been showed to cause the production of excess levels of reactive oxygen species (ROS) in a mouse models[36]. To measure the MPTP-elicited ROS level in the SN of PD mice, we co-stained brain slides with TH antibody and oxidant-sensing fluorescence probe H2-DCFDA. Fig.6A shows MPTP induced the production of a great deal of ROS, as evidenced by a high green fluorescence density in the SN of the mice of MTPT group compared to those of normal group (no MTPT).Interestingly, hAECs significantly reduced MPTP-elicited oxidative stress in the SN (TH positive area) of PD mice based on low green fluorescence density in MPTP+hAECs group compared to that of MPTP group (Fig.6A). Quantification revealed that the relative percentages of green fluorescence density in SN of PD mice in MPTP+hAECs group compared to that of MTPT group were evidently reduced (20.34%±5.85%) (Fig. 6B). Thus, the results indicated that hAECs grafts strikingly reduced MPTP-induced oxidative stress.