Transplantation of hESCs-Derived Neural Progenitor Cells Alleviates Secondary Damage of Thalamus After Focal Cerebral Infarction in Rats

Abstract

Human embryonic stem cells-derived neural progenitor cells (hESCs-NPCs) transplantation holds great potential to treat stroke. We previously reported that delayed secondary degeneration occurs in the ventroposterior nucleus (VPN) of ipsilateral thalamus after distal branch of middle cerebral artery occlusion (dMCAO) in adult male Sprague-Dawley (SD) rats. In this study, we investigate whether hESCs-NPCs would benefit the neural recovery of the secondary damage in the VPN after focal cerebral infarction. Permanent dMCAO was performed with electrocoagulation. Rats were randomized into Sham, dMCAO groups with or without hESCs-NPCs treatment. HESCs-NPCs were engrafted into the peri-infarct regions of rats at 48 h after dMCAO. The transplanted hESCs-NPCs survive and partially differentiate into mature neurons after dMCAO. Notably, hESCs-NPCs transplantation attenuated secondary damage of ipsilateral VPN and improved neurological functions of rats after dMCAO. Moreover, hESCs-NPCs transplantation significantly enhanced the expression of BDNF and TrkB and their interaction in ipsilateral VPN after dMCAO, which was reversed by the knockdown of TrkB. Transplantated hESCs-NPCs reconstituted thalamocortical connection and promoted the formation of synapses in ipsilateral VPN post-dMCAO. These results suggest that hESCs-NPCs transplantation attenuates secondary damage of ipsilateral thalamus after cortical infarction, possibly through activating BDNF/TrkB pathway, enhancing thalamocortical projection, and promoting synaptic formation. It provides a promising therapeutic strategy for secondary degeneration in the ipsilateral thalamus post-dMCAO.

Graphical abstract

Graphical Abstract.

Significance Statement.

HESCs-NPCs transplantation holds great potential to treat stroke. We investigate whether hESCs-NPCs would benefit neural recovery on the thalamic secondary damage after dMCAO. The transplanted hESCs-NPCs were able to attenuate secondary damage in the ipsilateral thalamus and improve the neurological functions of rats after dMCAO. These effects of hESCs-NPCs may be attributed to the activation of BDNF/TrkB pathway, reconstitution of cortico-thalamic connection, and promotion of synapse formation. It provides a promising therapeutic strategy for secondary damage of the ipsilateral thalamus after dMCAO.

Introduction

Stroke is the leading cause of mortality and disability in China.¹ It has been realized that cortical cerebral infarction leads to neuropathologic damages not only at the primary lesion site but also in nonischemic remote regions such as thalamus.² We previously reported that delayed secondary degeneration occurs in the ipsilateral ventroposterior nucleus (VPN) of thalamus after distal middle cerebral artery occlusion (dMCAO) in male rats.³ It becomes clear that focal infarction in the sensorimotor cortex leads to the disruption of cortex-thalamus connections and causes secondary damage in the thalamus.^4,5 This phenomenon was associated with the development of sensory disorder, cognitive impairment, and poor neurological outcome.^6-8 Although thrombolysis and mechanical thrombectomy in the acute stage of infarction have greatly improved the neurological outcomes of stroke patients,^9,10 the number of patients who can benefit from these treatments is limited. Additionally, these treatments have little effect on the restoration of damaged tissues.¹¹ Therefore, there is an urgent need for neuroprotection and neurorestoration against secondary degeneration after dMCAO.

Intriguingly, cell transplantation holds great potential to treat ischemic stroke.¹² Neural stem cells (NSCs) are multipotent and specifically differentiate into neural cell types, including neurons, astrocytes, and oligodendrocytes. Depending on the treatment protocol, NSCs are able to protect neural cells at risk, promote endogenous neurogenesis, foster synaptic remodeling, stimulate new vessel formation, and/or integrate into host neural circuits, which are associated with the improvements in cognitive and sensorimotor function.^13-15 Notably, transplantation of human embryonic stem cells-derived neural progenitor cells (hESCs-NPCs) in infarct cavities at 3 weeks after dMCAO in rats showed the potential of cell replacement and functional improvement at 9 weeks post-stroke.¹⁶ Moreover, NSCs can provide therapeutic gene products to modify the extracellular microenvironment and promote neuronal circuit plasticity.^17,18 Eckert et al reported that intracranial transplantation with human induced pluripotent stem cells-derived neural stem cells (hiPSCs-NSCs) at 24 h after MCAO/reperfusion increased brain-derived neurotrophic factor (BDNF) in the ipsilesional cortex and improved neurological functions of mice at 48 h post-stroke.¹⁹ It is well known that BDNF exerts its effects via binding with cell surface receptor tyrosine kinase receptor B (TrkB), which induces the activation of several signaling pathways such as the mitogen-activated protein kinase, the phospholipase C-gamma, and the phosphatidylinositol 3-kinase pathways, leading to cellular differentiation, survival, and synaptic plasticity.^20,21

To date, NSCs have been demonstrated to have a multimodal function in stroke treatment.²² However, there is no information on the therapeutic roles of hESCs-NPCs in the secondary degeneration of thalamic VPN after cortical infarction. Thus, the aim of this study is to investigate whether intracerebral transplantation of hESCs-NPCs attenuates the secondary degeneration of thalamic VPN following dMCAO. Further, we reveal that the neuroprotection of hESCs-NPCs was mediated through activating BDNF/TrkB pathway, enhancing thalamocortical projection, and promoting synaptic formation.

Materials and Methods

Human ESC Culture

The hESC lines H1 and green fluorescent protein (GFP) gene knock-in hESC lines H1-GFP were kindly provided by the Guangzhou Institutes of Biomedicine and Health, Chinese Academy of Sciences, China. For more information, see Supplementary experimental procedures.

Immunofluorescence

After fixing, cells were incubated with primary antibodies including octamer binding transcription factor (Oct) 4, Nanog, Nestin, paired box protein (Pax) 6, sry-related-high-mobility-group box (Sox) 2. After incubating overnight, the cells were incubated with the following secondary antibodies: Cy3-conjugated goat anti-rabbit immunoglobulin (Ig)G antibody and Cy3-conjugated goat anti-mouse IgG antibody. After that, slides were mounted with a mounting medium containing 4ʹ,6-diamidino-2-phenylindole (DAPI), and then analyzed with a confocal laser microscope.

Fluorescent immunohistochemistry was conducted. Antibodies used in these studies included human nuclei antigen (hNA), Nestin, doublecortin (DCX), neuron-specific nuclear-binding protein (NeuN), Synapsin I, human cytoplasmic marker Stem121, glial fibrillary acidic protein (GFAP), ionized calcium-binding adaptor molecule-1 (Iba-1), oligodendrocyte lineage transcription factor (Olig) 2, microtubule-associated protein (MAP)-2, TrkB, anti-nuclear-associated antigen Ki-67 (Ki-67), and the secondary antibody. Fluorescent images were obtained with a confocal laser microscope.

A list of antibodies and detailed protocols for immunohistochemistry are provided in Supplementary experimental procedures.

Flow Cytometry Analysis

Accutase was used to digest sample cells into single cells. Subsequently, these single cells were fixed and then incubated with the corresponding antibodies including Oct4, stage-specific embryonic antigen (SSEA) 4. A list of antibodies and detailed protocols for flow cytometry analysis are provided in Supplementary experimental procedures.

Reverse Transcription Quantitative Polymerase Chain Reaction (RT-qPCR)

Total RNA was extracted from cells with TRIzol and reverse transcribed with a HiScript II 1st Strand cDNA Synthesis Kit. RT-qPCR was performed according to the standard protocol. For more information, see Supplementary experimental procedures.

Teratoma Formation

The teratoma formation experiments were approved by the Ethical Committee on Animal Experiments at Guangzhou Institutes of Biomedicine and Health, Chinese Academy of Sciences (NO. 2010012). The experiment was performed as previously described.²³ Detailed protocols are provided in Supplementary experimental procedures.

Karyotype Analysis

Karyotype analysis was performed as described in our previous study.²³ For more information, see Supplementary experimental procedures.

Neural Progenitor Cells Induction and Maintenance

Human ESCs were induced into primitive neuroepithelial cells in a monolayer culture system using the dual small body size mothers against decapentaplegic (SMAD) inhibition protocol.²⁴ Detailed methods are presented in Supplementary experimental procedures.

Animals

All procedures in Sprague-Dawley (SD) rats were performed in accordance with Animal Research: Reporting In Vivo Experiments (ARRIVE) guidelines and were approved and monitored by the Ethical Committee on Animal Experiments from Guangzhou Institutes of Biomedicine and Health, Chinese Academy of Sciences (NO. 2010012) and Guangzhou Medical University (NO. A2020-019). Detailed procedures used in the study can be found in Supplementary experimental procedures.

In this study, 236 rats were used for the experiments (Supplementary Table S2). Two rats during the surgical procedure and 3 after surgery died in the dMCAO groups. Three rats in the vehicle group and 1 in the hESCs-NPCs group died during the surgical procedure. In addition, 2 rats died after injection of biotinylated dextran amine (BDA), and 7 rats (3 in the dMCAO with vehicle group, 2 in the dMCAO with shTrkB group, and 2 in the dMCAO with hESCs-NPCs group) were excluded because neither neurologic deficit nor cortical infarction after dMCAO was observed.

Distal Middle Cerebral Artery Occlusion and Cell Transplantation

Permanent dMCAO was performed using an electrocoagulation methodology described previously.^3,25 Sham operation (Sham) was performed with the same surgical procedures except for electrocoagulation of the distal MCA. Then, cell transplantation was performed 2 days after dMCAO. Intracortical transplantation of a total of 3 × 10⁵ hESCs-NPCs was performed stereotaxically at 3 sites close to the ischemic injury. Three microliters of vehicle (DMEM/F12) were injected in to Sham or dMCAO rats. In this study, rats were assigned randomly to 4 groups, including Sham + vehicle, Sham + hESCs-NPCs, dMCAO + vehicle, and dMCAO+hESCs-NPCs. Detailed protocols are provided in Supplementary experimental procedures.

Assessment of Cerebral Cortical Expansion

As previously described,²⁶ the assessment of cerebral cortical expansion was performed at 4 weeks after dMCAO. Detailed protocols are provided in Supplementary experimental procedures.

Histology

Animals were perfused at 4 weeks after dMCAO, and the brain tissues were processed. As previously described, an immunostained assay of NeuN was used for infarct volume evaluation.²⁷ Detailed protocols for histology are provided in Supplementary experimental procedures.

Immunohistochemistry

Single-labeled immunohistochemistry assay was performed using the avidin-biotin-peroxidase complex (ABC) technique. The antibodies used include NeuN and GFAP, Iba-1, Olig2, TrkB, and IgG. A list of antibodies and detailed protocols for immunofluorescence and immunohistochemistry are provided in Supplementary experimental procedures.

Western Blotting and Co-Immunoprecipitation

Animals were perfused at 4 weeks after dMCAO and the brain tissues were processed. Western blotting and immunoprecipitation procedures were performed as previously described.²⁸ The antibodies used include TrkB, BDNF, synapsin-I, postsynaptic density protein (PSD) 95, GADPH. A list of antibodies and detailed protocols are provided in Supplementary experimental procedures.

Lentiviral Construction Preparation and Administration

Three interfering short hairpin RNA sequences targeted rat TrkB (GenBank accession number NM_001163168.3) and a negative control (NC) sequence (CON323) were designed by Genechem (Shanghai, China). The sequences for Ntrk2 shRNA #1-#3 are GCTTATGCTTGCTGGTCTTGG, GCTTAAAGTTTGTGGCTTACA, and GGTTGGAACCTAACAGCATTG, respectively. The NC scrambled shRNA sequence is TTCTCCGAACGTGTCACGT. lentivirus-mediated TrkB shRNA (shTrkB) or shNC was injected stereotactically into ipsilateral VPN. Detailed protocols are provided in Supplementary experimental procedures.

Biotinylated Dextran Amine (BDA) Injections

The BDA anterograde tracing was performed as previously described.²⁹ Detailed protocols are provided in Supplementary experimental procedures.

Transmission Electron Microscopy (TEM)

Rats were perfused transcardially with 4% paraformaldehyde, containing 0.2% glutaraldehyde, and then brains were removed and cut on a vibratome. Ultrathin sections were mounted on grids, examined, and photographed using TEM (Tecnai G2 Spirit, FEI). Detailed protocols are provided in Supplementary experimental procedures.

Behavioral Assessment

Behavioral assessment was performed by an investigator who was blinded to the experimental groups. In this study, behavioral approaches were used to evaluate neurological, sensorimotor, and cognitive functions of rats, including Bederson scores, beam-walking test, adhesive removal test, and Morris water maze (MWM). For more information, see Supplementary experimental procedures.

Data Analyses

Statistical Package for Social Sciences Software for Windows, version 25.0 (SPSS, Inc., Chicago, IL, US, RRID: SCR_016479) was used for statistical analyses. When the data were normally distributed, one-way ANOVA and t-test were applied. Otherwise, nonparametric tests were used. For more information, see Supplementary experimental procedures.

Results

Generation of NPCs From hES Cells In Vitro

First, we ascertained that NPCs were derived from hESCs. As shown in Supplementary Fig. S1A, H1-GFP displayed typical morphologies of hESCs and expressed GFP. GFP⁺ cells colocalized with OCT4⁺ and NANOG⁺ cells in H1-GFP (Supplementary Fig. S1B). Further, we confirmed that the pluripotent protein expressions of H1-GFP were identified as OCT4 and SSEA4, which were in line with those of cell lines H1 (Supplementary Fig. S1C–S1F). H1-GFP had a normal karyotype (Supplementary Fig. S1G). To examine the differentiation potential of cells, H1-GFP was injected into non-obese diabetic severe combined immune deficiency (NOD-SCID) mice. It was demonstrated that the teratoma contained tissues derived from 3 germ layers (Supplementary Fig. S1H). To identify hESCs-NPCs, using RT-qPCR we found that the mRNA expressions of pluripotent genes OCT4 and NANOG were fully suppressed, whereas the mRNA expressions of hESCs-NPCs genes NESTIN, PAX6, SOX 1, and SOX 2 were obvious. Among them, the upregulation of SOX2 was the most significant (Supplementary Fig. S1I). In addition, it was shown that GFP⁺ cells colocalized with Nestin⁺, Pax 6⁺, and Sox 2⁺ cells in hESCs-NPCs (Supplementary Fig. S1J).

Neuronal Differentiation and Migration of Transplanted hESCs-NPCs in Ischemic Brain

The schematic diagram in Fig. 1A represented that NPCs derived from differentiation of hESCs at day 40 in vitro were dissociated and transplanted into the peri-infarct regions of rats at 48 h after dMCAO. Grafted hESCs-NPCs were identified by GFP. According to the injection coordinates described in Supplementary experimental procedures, hESCs-NPCs were transplanted into the cortex of both Sham and dMCAO groups. At 4 weeks after dMCAO, the ischemic lesion was restricted to the cortex, mostly in the somatosensory cortex and partially including the motor area. As expected, transplanted cells were located adjacent to the edge of the infarct cortex (Fig. 1B). Further we detected the differentiation of grafts injected in site 1 (Supplementary Fig. S2A–S2B), site 2 (Fig. 1C–1G), and site 3 (Supplementary Fig. S2C–S2D). Most of hNA⁺ and Stem121⁺ cells were Nestin⁺ (Fig. 1C, a–e; Supplementary Fig. S2A, a–d, S2C, a–d) and DCX⁺ (Fig. 1C, f–j; Supplementary Fig. S2A, e–h, S2C, e–h), and parts of them were NeuN⁺ (Fig. 1C, k–o; Supplementary Fig. S2A, i–l and SC, i–l), indicating grafted hESCs-NPCs predominantly exhibited neural progenitors, immature neurons and some of cells displayed mature neurons in Sham rats. Similarly, hNA⁺ and Stem121⁺ cells colocalized with Nestin⁺ (Fig. 1D, a-e; Supplementary Fig. S2B, a–d and S2D, a–d), DCX⁺ (Supplementary Fig. 1D, f–j; Supplementary Fig. S2B, e–h and S2D, e–h), and NeuN⁺ (Fig. 1D, k–o; Supplementary Fig. S2B, i–l and S2D, i–l) in dMCAO rats with hESCs-NPCs transplantation. However, compared with Sham + hESCs-NPCs group, the percentages of Nestin⁺/hNA⁺, DCX⁺/hNA⁺, and NeuN⁺/hNA⁺ cells in transplanted cells were significantly reduced in dMCAO+hESCs-NPCs rats (Fig. 1E–1G). In addition, we observed whether hESCs-NPCs differentiate into other cell types, such as glial cells. It was displayed that in ipsilateral cortex Stem121⁺ cells did not colocalized with GFAP⁺, Iba-1⁺, and Olig2⁺ cells in both Sham and dMCAO rats (Supplementary Fig. S3D–S3E).

Figure 1.

Neuronal differentiation of transplanted hESCs-NPCs in ischemic brain. (A) Experimental scheme of hESCs-NPCs transplantation experiments in ischemic brain. (B) Immunofluorescent image showed that transplanted cells were located adjacent to the edge of infarct cortex at 4 weeks after dMCAO. Scale bar: 2 mm. (C, D) Representative photomicrographs showed that the co-localization of GFP (green), hNA (red) and Nestin/DCX/NeuN (purple) in ipsilateral cortex of injection site 2 from Sham and dMCAO animals with hESCs-NPCs transplantation. Scale bar: 75 μm (a, f, k), 25 μm (b-e, g-j, l-o). (E-G) Quantitative analyses of Nestin⁺/hNA⁺/GFP⁺ (E), DCX⁺/hNA⁺/GFP⁺ (F), and NeuN⁺/hNA⁺/GFP⁺ (G) cells in ipsilateral cortex. Each bar represents the mean ± SD. ^*P < .05 versus Sham group with hESCs-NPCs treatment (n = 4 in each group). Sham: sham operation; dMCAO: distal middle cerebral artery occlusion; hESCs: human embryonic stem cells; NPCs: neural progenitor cells; hESCs-NPCs: human embryonic stem cells-derived neural progenitor cells; GFP: green fluorescent protein; NeuN: neuronal nuclei; hNA: human nuclei antigen; DCX: doublecortin; site 2: 2.8 mm posterior to bregma, 1.5 mm lateral to bregma, 2.2 mm below the dura; h: hour; w: week; R: right; L: left.

To investigate migration capability of transplanted hESCs-NPCs, we examined the GFP⁺/hNA⁺/NeuN⁺ cells in cortical regions, hippocampus, and thalamus, respectively. As is shown in the schematic diagram (Fig. 2A, 2B), GFP⁺/hNA⁺/NeuN⁺ cells are mainly located in the infarct core and peri-infarct regions (Fig. 2A, a2–b2). However, no GFP⁺/hNA⁺/NeuN⁺ cells were observed to migrate to the ipsilateral hippocampus (Fig. 2A, c2–e2) and ipsilateral thalamus (Fig. 2A, f2). In the contralateral brain, there were no GFP⁺/hNA⁺/NeuN⁺cells (Fig. 2B). Moreover, there were no GFP⁺ signals in the corpus callosum in both Sham and dMCAO groups (Supplementary Fig. S3G). Meanwhile, there were no GFP⁺/Stem121⁺/GFAP⁺, GFP⁺/Stem121⁺/Iba-1⁺, and GFP⁺/Stem121⁺/Olig2⁺ cells in the corpus callosum in dMCAO rats (Supplementary Fig. S3F).

Figure 2.

Migration of transplanted hESCs-NPCs and endogenous neurogenesis in ischemic brain. (A, B) Migration of transplanted hESCs-NPCs in ischemic brain. Schematic diagram of transplanted cells (a1, marked with green), cortical infarction (b1, marked with gray) and ipsilateral VPN (f1, marked with pink). Location of illustrated areas includes ipsilateral regions (a1-f1) and contralateral regions (a1ʹ-f1ʹ). The overlapped images showed that GFP⁺(green)/hNA⁺(red)/NeuN⁺(purple) cells mainly located in the infarct core and peri-infarct regions, rather than in the ipsilateral hippocampus, including CA1, CA3, and DG areas, ipsilateral thalamus and contralateral corresponding regions of the brain. Scale bar: 50 μm. (C) Endogenous neurogenesis in SVZ and SGZ in dMCAO groups with or without hESCs-NPCs transplantation. Representative fluorescent-images of Ki67 (green) and DAPI (blue) in SVZ (a-c) and SGZ (d-f). Ki67⁺ cells in SGZ were marked with red arrows. Scale bar: 50 μm. (D) Quantitative analyses of Ki67⁺ cells in SVZ and SGZ. Each bar represents the mean ± SD (n =4 in each group). Sham: sham operation; dMCAO: distal middle cerebral artery occlusion; hESCs-NPCs: human embryonic stem cells-derived neural progenitor cells; GFP: green fluorescent protein; NeuN: neuronal nuclei; hNA: human nuclei antigen; Ki-67: Anti-nuclear-associated antigen Ki-67; CA1 and 3: cornu ammonis 1 and 3; DG: dentategyrus; SVZ: subependymal ventricular zone; SGZ: subgranular zone; DAPI: 4ʹ,6-diamidino-2-phenylindole; ip: ipsilateral; con: contralateral; vehicle: DMEM/F12.

To assess the effects of transplanted hESCs-NPCs on neurogenesis after dMCAO, we observed the distribution of Ki-67⁺ cells in the subependymal ventricular zone (SVZ) and subgranular zone (SGZ). As shown in Fig. 2C, Ki-67⁺ cells were distributed in SVZ and SGZ both in Sham and dMCAO groups with or without hESCs-NPCs transplantation. However, there were no significant differences among these groups in the number of Ki-67⁺cell in SVZ and SGZ (Fig. 2D).

Transplanted hESCs-NPCs Ameliorate Secondary Damage in the ipsilateral VPN After dMCAO

To ascertain whether transplanted hESCs-NPCs attenuate cortex injury, cortical width index and relative infarct volume were measured at 4 weeks after dMCAO. As revealed by Fig. 3A, 3B, the cortical width index was significantly increased in the dMCAO group with hESCs-NPCs treatment compared to vehicle-treated rats. However, no significant difference in relative infarct volume was observed between hESCs-NPCs and vehicle groups after dMCAO (Fig. 3C, 3D). Further, we investigated the neuroprotection offered by transplanted hESCs-NPCs in the ipsilateral VPN after dMCAO. It was obvious that transplanted hESCs-NPCs significantly alleviated neuronal loss and gliosis in ipsilateral VPN at 4 weeks after dMCAO (Fig. 3E–3H).

Figure 3.

Transplanted hESCs-NPCs ameliorate secondary damage of ipsilateral VPN after focal cortical infarction. (A) Ex vivo images of cortical-width index with or without hESCs-NPCs transplantation after dMCAO in rats; x and y mark the maximum width from the midline to the edge of the infarcted and noninfarcted hemispheres, respectively. Dashed line shows a mirror image of the edge of contralateral hemisphere. Scale bar: 5 mm. (B) Quantitative analyses of cortical-width index. Each bar represents the mean ± SD. ^*P < .05 versus Sham group with or without hESCs-NPCs treatment and ^#P < .05 versus dMCAO + vehicle group (n = 6 in each group). (C) Immunostained sections of NeuN in cortex at 4 weeks after dMCAO with or without hESCs-NPCs transplantation. Scale bar: 2 mm. (D) Quantitative analyses of relative infarct volume. Each bar represents the mean ± SD. ^*P < .05 versus Sham group with or without hESCs-NPCs treatment (n = 6 in each group). (E) Representative microphotographs of immunohistochemistry for NeuN, GFAP, and Iba-1 in VPN after dMCAO with or without hESCs-NPCs transplantation. Scale bar: 250 μm (a, e, i, m) and 25 μm (b-d, f-h, j-l, n-p). (F-H) Quantitative analyses of NeuN⁺ cell, GFAP density, and Iba-1⁺ cell in VPN, respectively. Each bar represents the mean ± SD. ^*P < .05 versus Sham group with or without hESCs-NPCs treatment, ^#P < .05 versus dMCAO + vehicle group (n = 6 in each group). Sham: sham operation; dMCAO: distal middle cerebral artery occlusion; hESCs-NPCs: human embryonic stem cells-derived neural progenitor cells; NeuN: neuronal nuclei; GFAP: glial fibrillary acidic protein; Iba-1: ionized calcium-binding adaptor molecule 1; VPN: ventroposterior nucleus; ip: ipsilateral; con: contralateral; R: right; L: left; +, positive; vehicle: DMEM/F12.

BDNF/TrkB Pathway Mediates the Neuroprotection of hESCs-NPCs Against Secondary Damage in the Ipsilateral VPN After dMCAO

To determine whether hESCs-NPCs alleviate secondary damage of thalamus via the activation of BDNF/TrkB pathway, we first examined the expressions of BDNF and TrkB. As demonstrated in Fig. 4A–4D, in contrast with the Sham group with or without hESCs-NPCs, the expressions of BDNF and TrkB in infarct cortex and ipsilateral VPN decreased at 4 weeks after dMCAO, which were upregulated by hESCs-NPCs treatment. Further, we examined the interaction between BDNF and TrkB. An obvious increase in the BDNF-TrkB interaction was observed in infarct cortex and ipsilateral VPN in hESCs-NPCs groups when compared with dMCAO groups (Fig. 4E–4F). Similar results in the expressions of TrkB were found (Fig. 4G, 4H). Most TrkB⁺ cells from Sham rats had round nuclei with a granular appearance (Fig. 4G). Immunofluorescent analysis revealed that the majority of cells were NeuN and MAP2 positive. Only minorities were GFAP and Iba-1 positive (Fig. 4I, a–h), suggesting that TrkB predominantly localized in neurons in Sham rats with or without hESCs-NPCs. Alternatively, at 4 weeks after dMCAO, TrkB staining mainly existed in GFAP⁺ and Iba-1⁺ cells (Fig. 4I, i–l). Notably, in comparison with dMCAO with vehicle group, more TrkB⁺ cells colocalized with NeuN and MAP2, parts of cells with GFAP and Iba-1 in dMCAO with hESCs-NPCs group (Fig. 4I, i–p).

Figure 4.

BDNF/TrkB pathway partially mediates the neuroprotective effect of hESCs-NPCs on secondary damage of thalamus after focal cortical infarction. (A-D) Representative immunoblots of BDNF and TrkB in cortex and VPN after dMCAO with or without hESCs-NPCs transplantation. Quantitative analyses of BDNF protein (A and C) and TrkB protein (B and D). Each bar represents the mean ± SD. ^*P < .05 versus Sham group with or without hESCs-NPCs treatment, ^#P < .05 versus dMCAO + vehicle group (n = 3 in each group). (E-F) Immunoprecipitation of TrkB was measured in the dMCAO group with or without hESCs-NPCs treatment. Quantitative analyses of BDNF protein in cortex and VPN. Each bar represents the mean ± SD. ^*P < .05 versus Sham group with or without hESCs-NPCs treatment, ^#P < .05 versus dMCAO +vehicle group (n = 3 in each group). (G) Representative images showed TrkB immunopositive cells in VPN of dMCAO rats treated with or without hESCs-NPCs. Scale bar: 250 μm (a, c, e, g) and 25 μm (b, d, f, h). (H) Quantitative analyses of TrkB immunopositive cell in VPN. Each bar represents the mean ± SD. ^*P < .05 versus Sham group with or without hESCs-NPCs treatment, ^#P < .05 versus dMCAO + vehicle group (n = 6 in each group). (I) Representative photomicrographs showed that the co-localization of TrkB (green), NeuN/MAP-2/GFAP/Iba-1 (red), and DAPI (blue) in ipsilateral VPN at 4 weeks after dMCAO with or without hESCs-NPCs treatment. Scale bar: 25 μm (a-p). Sham: sham operation; dMCAO: distal middle cerebral artery occlusion; hESCs-NPCs: human embryonic stem cells-derived neural progenitor cells; BDNF: brain-derived neurotrophic factor; TrkB: tyrosine kinase receptor B; GAPDH: glyceraldehyde 3-phosphate dehydrogenase; VPN: ventroposterior nucleus; GFP: green fluorescent protein; NeuN: neuronal nuclei; MAP2: microtubule-associated protein 2; GFAP: glial fibrillary acidic protein; Iba-1: ionized calcium-binding adaptor molecule 1; DAPI: 4, 6-diamidino-2-phenylindole; IP: immunoprecipitation; IB: immunoblot; ip: ipsilateral; con: contralateral; w: week; vehicle: DMEM/F12.

To further confirm the upregulation of BDNF/TrkB pathway was involved in the alleviation of secondary damage of thalamus after dMCAO mediated by hESCs-NPCs, shTrkB was administrated. Two weeks later, rats were subjected to Sham or dMCAO with or without hESCs-NPCs treatment (Fig. 5A). As shown in Fig. 5B, the neurons can be transfected by shTrkB in ipsilateral VPN. The expression of TrkB significantly decreased after shTrkB (#1- #3) administration at the dosage of 16.0 × 10⁸ TU/mL in Sham rats. However, there were no significant inhibitory effects of shTrkB (#1-#3) at 2 dosages (4.0 and 8.0 × 10⁸ TU/mL) on the expression of TrkB (Supplementary Fig. S3A–S3C). Thus, the dosage of 16.0 × 10⁸ TU/mL shTrkB #1 was selected in the following experiments. In hESCs-NPCs transplanted group at 4 weeks after dMCAO, TrkB expression in ipsilateral VPN was inhibited with shTrkB #1 treatment (Fig. 5C), the interaction of BDNF-TrkB was interrupted (Fig. 5D–5E). Accordingly, the effects of shTrkB #1 administration on the neuroprotection after hESCs-NPCs transplantation were examined. As expected, the knockdown of TrkB inhibited the expression of TrkB in the dMCAO group with hESCs-NPCs transplantation, decreased NeuN positive cells, increased the density of GFAP and Iba-1 positive cells of ipsilateral VPN, indicating that shTrkB can abolish the protective effects of hESCs-NPCs transplantation on secondary damage of VPN via inhibiting the activation of TrkB after dMCAO (Fig. 5F–5I).

Figure 5.

The inhibition of TrkB counteracts the neuroprotection induced by hESCs-NPCs transplantation in ipsilateral VPN after dMCAO. (A) Schematic depicting the design of the experiments. (B) Fluorescent image from coronal section of ipsilateral VPN following the injection of TrkB lentiviral vectors in Sham animal. Scale bar: 50 μm. (C) Representative immunoblots of TrkB in ipsilateral VPN of Sham, dMCAO, and hESCs-NPCs groups with or without shTrkB #1 treatment and quantitative analyses of TrkB in ipsilateral VPN. (D-E) Immunoprecipitation and Western blotting showed the effects of hESCs-NPCs transplantation on the interaction between BDNF and TrkB in ipsilateral VPN after dMCAO with or without shTrkB #1 treatment. Each bar represents the mean ± SD. ^*P < .05 versus Sham + shNC group, ^#P < .05 versus dMCAO + shNC group and ^&P < .05 versus hESCs-NPCs +shNC group (n = 4 in each group). (F) Representative images of immunohistochemistry for NeuN, GFAP, and Iba-1 in ipsilateral VPN of Sham, dMCAO, and hESCs-NPCs groups with or without shTrkB #1 treatment. Scale bar: 250 μm (a, e, i, m, q, u) and 25 μm (b-d, f-h, j-l, n-p, r-t, v-x). (G-I) Quantitative analyses of NeuN⁺ cell, GFAP density, and Iba-1⁺ cell in VPN, respectively. Each bar represents the mean ± SD. ^*P < .05 versus Sham + shNC group, ^#P < .05 versus dMCAO + shNC group and ^&P < .05 versus hESCs-NPCs + shNC group (n = 6 in each group). Sham: sham operation; dMCAO: distal middle cerebral artery occlusion; hESCs-NPCs: human embryonic stem cells-derived neural progenitor cells; GFP: green fluorescent protein; BDNF: brain-derived neurotrophic factor; TrkB: tyrosine kinase receptor B; GAPDH: glyceraldehyde 3-phosphate dehydrogenase; VPN: ventroposterior nucleus; NeuN: neuronal nuclei; GFAP: glial fibrillary acidic protein; Iba-1: ionized calcium-binding adaptor molecule 1; IP: immunoprecipitation; IB: immunoblot; ip: ipsilateral; con: contralateral; sh: shRNA; NC: negative control; w: week.

Transplanted hESCs-NPCs Reconstitute Thalamocortical Connections After dMCAO

To visualize the rewiring of the thalamocortical connections, the anterograde axonal tracer BDA was injected into the ipsilateral VPN after dMCAO with or without hESCs-NPCs transplantation (Fig. 6A). In Sham rats, a lot of BDA⁺ staining was visible in the ipsilateral VPN, corpus callosum, and cortex (Fig. 6B, a–c). On the contrary, BDA⁺ staining was mainly distributed in the ipsilateral VPN, only a few were found in the ipsilateral corpus callosum and cortex at 4 weeks after dMCAO (Fig. 6B, d–f, and C). Importantly, compared with vehicle groups, hESCs-NPCs treatment distinctly increased the number of BDA⁺ projections in the ipsilateral corpus callosum and density of BDA⁺ projections in the ipsilateral cortex (Fig. 6Be–f, h-i, 6C). These results suggest that the transplantation of hESCs-NPCs potentially reconstituted cortico-thalamic connections after dMCAO.

Figure 6.

Transplanted hESCs-NPCs reconstitute thalamocortical connections after focal cortical infarction. (A) Schematic diagram of brain section showed the location of cortical infarction (black area) and the ipsilateral thalamus (pink area) where BDA was delivered. (B) Representative confocal images of BDA staining in ipsilateral cortex, corpus callosum, and VPN at 4 weeks after dMCAO with or without hESCs-NPCs treatment. Scale bar: 50 μm (a-i). (C) Quantitative analyses of BDA⁺ projections in corpus callosum and density of BDA⁺ in cortex. Each bar represents the mean ± SD. ^*P < .05 versus Sham + vehicle group, ^#P < .05 versus dMCAO + vehicle group (n = 4 in each group). (D) Representative fluorescent images of Synapsin-I (red), MAP-2 (green), and DAPI (blue) in ipsilateral VPN at 4 weeks after dMCAO with or without hESCs-NPCs treatment. Synapse formation was measured by co-labeling of Synapsin-I and MAP-2. Scale bar: 25 μm (a-l). (E) Quantitative analyses of synapsin-I⁺-MAP-2⁺ puncta in ipsilateral VPN. Each bar represents the mean ± SD. ^*#P < .05 versus Sham + vehicle group, ^#P < .05 versus dMCAO + vehicle group (n = 6 both in Sham + vehicle and dMCAO + vehicle groups, n = 4 in dMCAO + hESCs-NPCs group). (F) The alterations of synaptic structure and quantification of synapses in ipsilateral VPN after dMCAO with or without hESCs-NPCs treatment. Arrowhead represented synaptic structure (a, c, e). Yellow arrowhead (a, c, e) indicated the magnified regions displayed in the right panel (b, d, f) and yellow asterisk represented synaptic structure (b, d, f). Each bar represents the mean ± SD. ^*P < .05 versus Sham + vehicle group, ^#P < .05 versus dMCAO + vehicle group (n = 6 in each group). (H-I) Representative immunoblots and quantitative analyses of synapsin-I and PSD95 in VPN after dMCAO with or without hESCs-NPCs transplantation. Each bar represents the mean ± SD. ^*P < .05 versus Sham group with or without hESCs-NPCs treatment, ^#P < .05 versus dMCAO + vehicle group (n = 4 in each group). Sham: sham operation; dMCAO: distal middle cerebral artery occlusion; hESCs-NPCs: human embryonic stem cells-derived neural progenitor cells; BDA: biotinylated dextran amine; MAP2: microtubule-associated protein 2; DAPI: 4ʹ, 6-diamidino-2-phenylindole; PSD95: postsynaptic density 95; GAPDH: glyceraldehyde 3-phosphate dehydrogenase; VPN: ventroposterior nucleus; ip: ipsilateral; con: contralateral; vehicle: DMEM/F12.

Transplanted hESCs-NPCs Promote Synapse Formation in the Ipsilateral VPN After dMCAO

To determine whether hESCs-NPCs promote synapse formation of thalamus, we examined the synaptic quantity and structure by immunohistochemical assay and TEM. As shown in Fig. 6D–6E, most of synapsin-I⁺ puncta were colocalized with MAP-2 in the Sham group, and only a few were colocalized with MAP-2 in dMCAO with vehicle group. In contrast, engrafted hESCs-NPCs increased the number of synapsin-I⁺ puncta co-labeled with MAP-2.

Under TEM, micrographs of synapses of VPN in the Sham group displayed a typical synaptic structure with clustering of synaptic vesicles close to the presynaptic membrane, a synaptic cleft containing an intermediate layer of dense material, and a postsynaptic membrane with postsynaptic densities (Fig. 6F, a–b). However, in the dMCAO group with vehicle, the presynaptic and postsynaptic structures were destructed (Fig. 6F, c–d) and the number of synapses in the ipsilateral VPN decreased (Fig. 6G). Interestingly, the transplanted hESCs-NPCs maintained synapses relatively intact (Fig. 6F, e–f) and apparently increased the number of synapses at 4 weeks after dMCAO (Fig. 6G). Also, hESCs-NPCs treatment in dMCAO rats increased the levels of synapsin I and PSD95 in the ipsilateral VPN after dMCAO (Fig. 6H–6I).

Transplanted hESCs-NPCs Improve Neurological Functions After dMCAO

To assess long-term functional deficits and possible recovery with or without transplanted hESCs-NPCs after dMCAO in rats, neurobehavioral changes in rats were observed. Compared with vehicle-treated rats, rats treated with hESCs-NPCs exhibited higher Bederson scores and beam-walking test scores at 14 days after dMCAO (Fig. 7A, 7B). In the adhesive removal test, the mean time to remove the adhesive from the forepaws was significantly shorter in the hESCs-NPCs group than that in the vehicle group after 14 days of dMCAO (Fig. 7C).

Figure 7.

Transplanted hESCs-NPCs improve neurological functions after focal cortical infarction. (A-C) Neurological functions were measured on day 1 before dMCAO and on day 1, 3, 7, 14, 21, and 28 after dMCAO with or without hESCs-NPCs transplantation (n = 12 in each group). Quantitative analyses of Bederson’s scores (A), beam-walking scores (B), and asymmetry scores in adhesive removal test (C). Cognitive function was tested by MWM (n = 10 in each group). (D) Representative path tracings in each quadrant during the training period (learning) and probe trial (memory). Quantitative analyses of the path length (E), escape latency (F), and swimming speed (G) in Sham and dMCAO group with or without hESCs-NPCs treatment during the training period. (H) Quantitative analyses of the percentage of time spent in the target quadrant to total time (120 s) at day 28 after dMCAO. Data are expressed as the mean ± SD. ^*P < .05 versus Sham group with or without hESCs-NPCs treatment, ^#P < .05 versus dMCAO + vehicle group. (n = 12 in each group for Bederson’s neurological test, beam-walking test, and adhesive removal test; n = 10 in each group for MWM). Sham: sham operation; dMCAO: distal middle cerebral artery occlusion; hESCs-NPCs: human embryonic stem cells-derived neural progenitor cells; MWM: morris water maze; s: second; cm: centimeter; d: day; vehicle: DMEM/F12.

The cognitive function of rats was assessed by MWM. After 5 days of spatial orientation training, all experimental rats improved their ability to locate the platform. The path length and escape latency in dMCAO rats were longer than those in the Sham group with vehicle or hESCs-NPCs treatment. Notably, transplanted hESCs-NPCs improved spatial learning ability from the 3rd to 5th day of training when compared to dMCAO rats (Fig. 7D–7F). Swimming velocity did not significantly differ among Sham and dMCAO groups with vehicle or hESCs-NPCs treatment (Fig. 7G). In the probe phase, the retention of long-term spatial memory was analyzed (Fig. 7H). With hESCs-NPCs treatment, the rats improved their long-term spatial memory after dMCAO, when compared to dMCAO with vehicle rats. These data suggest that transplanted hESCs-NPCs could improve the neurological functions of rats after dMCAO.

Discussion

In this study, we confirmed that the transplanted hESCs-NPCs in peri-infarct regions could survive and partially differentiateinto mature neurons in adult rats after dMCAO. Transplantation with hESCs-NPCs significantly alleviated secondary damage of the ipsilateral thalamus following focal cerebral ischemia. The neuroprotective effects induced by hESCs-NPCs on the secondary damage of thalamus post-cortical infarction were mediated by activating the BDNF/TrkB pathway. Transplantation with hESCs-NPCs reconstituted thalamocortical connection promoted the formation of synapses, and improved neurological functional outcome after cortical infarction.

NPCs used for this study were derived from hESCs. Since hESCs-NPCs could express the markers of NPC, including Nestin, Pax6, Sox1, and Sox2, they were considered to be generated successfully from hESCs. It is known that transplanted NPCs survive and differentiate in the rodent brain after cerebral ischemia, improving functional outcomes.^30,31 After implantation of exogenous human fetal NSCs into cortical infarct border at 7 days after dMCAO in mice, the engrafted cells differentiate into mature neurons at 4 weeks, showing that cerebral cortical expansion was increased and sensorimotor functions were improved.²⁶ Our results also confirmed that transplanted hESCs-NPCs can survive and differentiate into mature neurons in the ischemic region. Although there were no significant differences among Sham, dMCAO, and dMCAO with hESCs-NPCs groups in the number of Ki-67⁺ cells in SVZ and SGZ at 4 weeks after dMCAO, the transplantation with hESCs-NPCs significantly alleviated secondary damage of ipsilateral thalamus and improved neurological functions. These findings indicated that the reduction of secondary thalamic injury by hESCs-NPCs transplantation was not mainly due to endogenous neurogenesis after cortical infarction. In addition, a study by Wen et al showed that the improvement of neurological functions positively correlated with cortical expansion after cerebral infarction.³² Wang et al found that the increase in cortical width was due to a higher number of grafted cells in peri-infarct area at 5 weeks after dMCAO.²⁶ Similarly, our study showed that the transplantation of hESCs-NPCs increased cortical width after cortical infarction, which may be attributed to the higher number of cells in peri-infarct area. The association of neurological functional improvement and cortical expansion offered by transplanted hESCs-NPCs remains unclear and needs further study.

Secondary alteration in areas functionally related to the lesion site after stroke in patients and animal models greatly contributes to long-term functional impairments.^33,34 After dMCAO, neuronal death, axonal degeneration, gliosis, and marked atrophy have been found in the ipsilateral thalamus.³⁵ Our results showed that intracranial transplantation of hESCs-NPCs reduced neuronal loss and gliosis of ipsilateral VPN, implying a substantial postischemic repair potential of grafted hESCs-NPCs therapy. Interestingly, transplanted hESCs-NPCs cannot migrate toward the area outside the infarct core, such as VPN. Limited cell replacement has been observed in several studies. It is believed that NSCs differentiate into functional neurons require a considerable amount of time.^36,37 There is still not enough evidence to show that cell replacement is vital for NSCs-mediated recovery. For instance, Eckert et al demonstrated that hiPSCs-NSCs could survive up to 30 days after intracranial transplantation at 24 h post-stroke, but the vast majority of donor cells remained as nestin-positive NSCs. Only a small percentage of exogenous hiPSCs-NSCs co-expressed the neuronal marker TuJ-1.¹⁹ Consistent with another observation that interventions led to recovery but had no effect on tissue outcomes,³⁸ hESCs-NPCs treatment did not reduce infarct volume at 4 weeks of dMCAO, but it could improve the neurological functions of rats. The gradual sensorimotor improvement over several weeks after cell transplantation suggests that the hESCs-NPCs have a true recovery-enhancing effect after stroke.

To date, the mechanisms of recovery mediated by transplanted cells after ischemic stroke have not been completely elucidated. Previous studies have shown that the stimulation of neurogenesis in SVZ and SGZ by NSC transplantation is associated with functional improvement after MCAO in rodents.^15,39 However, our results showed that no significant changes were detected in the number of proliferating Ki-67⁺ cells in SVZ at 28 days after dMCAO with or without hESCs-NPCs treatment, which was consistent with those of others.^40,41 The study from Cuartero et al⁴² found that the increase of proliferation in SGZ starts approximately 1 week after stroke, peaks at days 10-14, and returns to basal levels within 4-5 weeks after onset of MCAO. Therefore, in this study, SVZ or SGZ-derived neurogenesis may play a nonprimary role in contributing to the recovery after stroke with hESCs-NPCs treatment. It is possible that hESCs-NPCs promote neurological recovery through other indirect mechanisms.

Recent studies have shown that transplanted NSCs could exert a bystander effect by preventing tissue damage, interfering with the pathogenic process, or rescuing endogenous neural cells.^17,43 NSC-secreted BDNF is a main neurotrophin that promotes neurogenesis and improves neurological function after stroke. The activation of BDNF/TrkB signaling pathway plays a crucial role in the regulation of brain plasticity.^44,45 For example, the downregulation of TrkB increased infarct size and aggravated neurological outcomes in adult male mice of permanent focal ischemia.⁴⁶ Our results revealed that the level of TrkB in the ipsilateral VPN was decreased at 4 weeks after dMCAO. However, there was no more reduction of TrkB level in the shTrkB-treated dMCAO group, indicating that when the expression of TrkB in the ipsilateral VPN was at lower level after dMCAO, there were no additive effects of shTrkB treatment and cerebral ischemia on the inhibition of TrkB expression. Huang et al reported that intracranial transplantation of hNSC following MCAO/reperfusion in mice significantly increased BDNF expression in the hippocampus and slowed the progression of behavioral dysfunction, suggesting that BDNF upregulation offers neuroprotection against cerebral ischemia.⁴⁷ Our results also revealed that transplanted hESCs-NPCs activated the BDNF/TrkB pathway, thereby facilitating neuronal survival in ipsilateral VPN after cortical infarction. When TrkB was inhibited by shTrkB, the neuroprotection mediated by hESCs-NPCs was counteracted, implying these effects induced by hESCs-NPCs transplantation was indirect.

Clinically patients with preserved thalamocortical connection integrity have significantly higher scores in motor function than those with disrupted thalamocortical connection at 6 months post-stroke.⁴⁸ Axonal sprouting occurs with new projections after stroke to target denervating area.⁴⁹ In rodent and primate models of ischemic cortical injury, such sprouting has been observed locally around the infarct area.⁵⁰ Our study demonstrated that hESCs-NPCs transplantation promoted axonal sprouting in the ipsilateral cortex and corpus callosum after cortical infarction. This is consistent with a previous report from Daadi et al, where hESCs-NPCs were used in hypoxic-ischemic brain injury of immunosuppressed animals.³¹ Human NPCs transplantation can enhance stroke-induced axons sprouting of cortico-thalamic tract.^30,37 In this study the disruption of cortico-thalamic connections induced by dMCAO occurred, but this neuronal circuit deficit was significantly ameliorated by hESCs-NPCs treatment, indicating that cell-based rehabilitation strategy contributes to the restoration of damaged neuronal connectivity. BDNF is known to be an activity-dependent neurotrophic factor that promotes axonal sprouting and extension.⁵¹ HiPSCs-NPCs stimulated by optogenetic activation can upregulate presynaptic and postsynaptic markers, synapsin-1, and PSD95.³⁰ In line with that study, hESCs-NPCs treatment upregulated the expressions of BDNF, synapsin-I, and PSD95 in the ipsilateral VPN after dMCAO, suggesting the formation and stabilization of new synaptic contacts through beneficial paracrine and synaptogenesis effects. This was further supported by immunofluorescence and TEM analysis which show that hESCs-NPCs treatment promoted the formation of synapses in ipsilateral VPN after dMCAO.

Conclusion

For the first time, we explored the roles and potential mechanisms involved in the neuroprotective effects offered by the transplanted hESC-NPCs on secondary damage of the ipsilateral thalamus after cortical infarction in rats. The transplanted hESCs-NPCs were able to attenuate secondary damage in the ipsilateral thalamus and improve the neurological functions of rats after dMCAO. These effects of hESCs-NPCs may be attributed to the activation of BDNF/TrkB pathway, reconstitution of cortico-thalamic connection, and promotion of synapse formation. It provides a promising therapeutic strategy for secondary degeneration in the ipsilateral thalamus post-dMCAO.

Funding

This work is supported by the Technology Planning Project of Guangdong Province, China (Grant No. 2015B020228003), the National Natural Science Foundation of China (Grant No. 81971233), the Natural Science Foundation of Guangdong Province, China (Grant No. 2021A1515011269) and the Key Medical Disciplines and Specialties Program of Guangzhou (2021-2023).

Conflict of Interest

The authors declared no potential conflicts of interest.

Author Contributions

E.X., G.P., and K.L. conceived the study. E.X. and K.L. designed the experiments and assembled all the figures. K.L., L.P., and Q.X. performed the experiments with the assistance of XL.Z., H.L., W.S., XF.Z., W.H., and T.Z. This article was written by K.L. and E.X., G.P., and L.Z. assisted with the revision of manuscript. All authors read and approved the final version of this article.

Data Availability

All data included in this study are available upon request by contact with the corresponding author.