Electrical Guidance of Human Stem Cells in the Rat Brain

Highlights

• Developed a technology and device delivering electric current to the brain in vivo

• Achieved stable delivery of currents to brain with monitoring and safety concerns

• Exhibited effective guidance of migration of transplanted human NSCs in live brain

• Demonstrated enhanced motility, survival, and differentiation of the guided hNSCs

Summary

Limited migration of neural stem cells in adult brain is a roadblock for the use of stem cell therapies to treat brain diseases and injuries. Here, we report a strategy that mobilizes and guides migration of stem cells in the brain in vivo. We developed a safe stimulation paradigm to deliver directional currents in the brain. Tracking cells expressing GFP demonstrated electrical mobilization and guidance of migration of human neural stem cells, even against co-existing intrinsic cues in the rostral migration stream. Transplanted cells were observed at 3 weeks and 4 months after stimulation in areas guided by the stimulation currents, and with indications of differentiation. Electrical stimulation thus may provide a potential approach to facilitate brain stem cell therapies.

Introduction

Neural stem cells/progenitor cells (NSCs) promise great hope for various neurological diseases. Researchers have demonstrated that NSCs are able to migrate and differentiate into adult rat brain and spinal cord (Flax et al., 1998, Snyder and Teng, 2012, Tabar et al., 2005). The human brain, however, poses a particular challenge for migrating NSCs or neuroblasts due to our larger brain size and the distances that cells must travel. Endogenous neuroblasts reside in the subventricular zone (SVZ) and hippocampus, deep in the brain. Neuroblasts from these niches have to migrate long distances to reach lesions in the cortex or other extra-hippocampal regions. Another hurdle is that transplanted NSCs have very poor motility due to suppression of migration by NSCs to each other and their progenies (Ladewig et al., 2014). Expanding the limits of migration of stem cells in brain is therefore a key step in stem cell therapies for the human brain (Trounson and McDonald, 2015). We aim to develop techniques that can mobilize and guide stem cells in the brain in vivo, which has not yet been achieved.

We chose the rostral migration stream (RMS) to develop our stimulation technique because this is one of the most active migratory paths in the brain, and its cellular and molecular mechanisms are well understood (Anton et al., 2004, Curtis et al., 2007, Mobley and McCarty, 2011, Sanai et al., 2011, Staquicini et al., 2009). Newly born neuroblasts and transplanted hNSCs placed at the SVZ normally migrate directionally downstream to the olfactory bulb (OB), guided by various cues, including multiple chemical gradients and flow of cerebral spinal fluid (Flax et al., 1998, Sawamoto et al., 2006, Snyder and Teng, 2012, Tabar et al., 2005, Wu et al., 1999). This model allows us to test whether our technique is able to guide hNSCs to travel upstream toward the lateral ventricle (LV) region on the ipsilateral side, against endogenous directional cues.

Electric fields (EFs) provide a powerful signal with which to stimulate and guide migration of many types of cells in vitro, including NSCs (Cao et al., 2013, Feng et al., 2012a, Li et al., 2008, Meng et al., 2011, Yao et al., 2008, Yao et al., 2009, Zhao et al., 2006). Normally, weak EFs in the brain may aid in guiding the migration of neuroblasts from the SVZ to the OB (Cao et al., 2013). We hypothesized that a stronger applied EF would provide a signal sufficient to guide some of the hNSCs injected in the middle of the RMS to move upstream toward the SVZ, against the intrinsic guidance mechanisms (Figure 1).

Results

The Overall Research Design

First, we transplant human neural stem cells (hNSCs) into the RMS (Figure 1A). The transplanted cells migrate to the OB, following the endogenous directional signal (Figure 1B). We then apply electric currents along the RMS with minimal effects on brain electrical activities and motor behavior (Figure 1C). If the EFs are applied against the endogenous direction of neuroblasts (i.e., downstream of the SVZ to the OB), and if the electrical guidance effect is strong enough, we should see transplanted cells being guided to migrate against the endogenous cues and upstream to the SVZ (Figure 1D).

To Track NSCs in the Brain, We First Developed an hNSC Line that Expresses EGFP

The previously described hNSCs from H9 (Feng et al., 2012a) were transduced with MNDU3-luciferase-PGK-EGFP, a lentiviral vector expressing EGFP. EGFP-positive cells enriched by cell sorting provided a consistent number of cells for transplantation (Figure S1). The transduced cells maintained markers for NSCs and allowed us to differentiate types of cells (Figures 2A–2D ). We tested whether expression of EGFP altered galvanotaxis. Applied EFs effectively mobilized and guided the migration of the hNSCs expressing EGFP (EGFP-hNSCs) in the same way as their parental cells, and that of neuroblasts from neonatal rat brain and from the SVZ of adult mice (Figures 2E–2H and S2; Movie S1) (Cao et al., 2013, Feng et al., 2012a, Li et al., 2008).

We Then Optimized the Electrical Stimulation Scheme to Effectively Guide Cells

Our setup had a unique modification of the classic galvanotaxis chamber with a very small conductive volume (∼20 μL) over a large surface area (400 mm²), ensuring minimal electric currents at physiological voltages (less than 1 mA). This design efficiently dissipates heat generated and minimizes changes in ions and perturbation of culture conditions. Cells exposed to a field of 100–200 mV/mm remained healthy and motile for several days (Song et al., 2007). With this design, however, it was not possible to deliver direct current (DC) EFs to the brain, because the large conductive volume reduces resistance and allows currents of hundreds and thousands times higher to pass through the tissues at similar voltage, inducing a significant Joule effect, changes in pH and ion concentrations, and electrode by-products. We developed optimal stimulation schemes using intermittent EFs (iEFs) that minimized detrimental effects while maintaining effective guidance for migration of hNSCs. iEFs with specific on and off ratios showed significant guidance effects on directional migration of hNSCs while maintaining cell viability after prolonged stimulation (Figures S3A and S3B; Movies S2 and S3), and induced negligible changes in temperature and pH in the culture chamber (Figures S3C–S3E3).

Electrode Pairs Were Used to Simultaneously Deliver and Monitor Stable Currents at the Same Time

Based on the preliminary test, we chose carbon electrodes to deliver current and silver/silver chloride (Ag/AgCl) electrodes to monitor the EF induced (Figures 3B and S4A). Two configurations of electrodes were used: two plus four (2 + 4) (Figure 3C), and two pairs (2 + 2) (Figure 3D). The 2 + 4 electrodes have more detailed measurement points. Paired electrodes were inserted and immobilized in the brain of anesthetized animals with a stereotactic frame. The electrical currents were delivered to the brain in animals under continuous anesthesia or in free-moving status after recovery from surgical anesthesia. After the rat head was immobilized in a stereotaxic stage, the electrodes were inserted into the brain from a window in the skull surgically made to expos the dura. The electrodes were fixed to a holder with a pre-determined depth and spacing that corresponded to the points along the RMS (Figures 3 and S4A). Histology sections confirmed that injection needle and electrodes were correctly placed and secured using stereotactic manipulation and electrode immobilization (Figures 4A, 4B, 4E, and 5A ). Continuous monitoring of the EFs and currents in the brain during stimulation allowed adjustment to ensure stable voltage and currents in the brain (Figures S4B–S4E). During continuous anesthesia, the electric currents were delivered to the brain along the RMS for up to 10 hr (Figure 3D). For animals that recovered from anesthesia and moved freely in the cage, we developed a portable miniaturized stimulator with a button battery and a programmable chip to administer stimulation. The battery and programmable chip were housed in a transparent plastic tube. After stereotaxic electrode implantation, the battery-chip tube configuration was secured to the skull with dental acrylic (Figure 3E).

Electroencephalographic and Motor Functions Suggested Good Tolerance of the Rats to Surgery and Electrical Stimulation

No seizures were observed in any animals during or after the electrical stimulation. We placed two electroencephalogram (EEG) recording electrodes in the left and right frontal bone next to the dura, using a modified method described previously (Feng et al., 2012b) (Figures S5A and S5B). Electrical stimulation used in our experiments had negligible effects on the EEG. Post-stimulation recording showed waveforms similar to those pre-stimulation (Figure S5E). Quantitative analysis confirmed complete overlap of the EEG recorded from rat before stimulation and after intermittent EF stimulation (Figure S5D). Frequency dependence of the power analysis based on the harvested EEGs showed no statistically significant changes in theta or beta waves, but slightly increased the power of low gamma wave during iEF stimulation (Figure S5E).

To determine the effects of electrode implantation and stimulation on animal behavior, we recorded normal movement of the animals after recovery from anesthesia and for more than 3 weeks (Figure S5F). We evaluated the effects on locomotor function (Rotarod) and fine motor coordination function (horizontal ladder walk) (Hamm et al., 1994, Metz and Whishaw, 2002, Metz and Whishaw, 2009). The Rotarod performance of the cell transplantation group alone and the cell transplantation plus electrodes groups were essentially unchanged from baseline on all post-surgery test days. There appeared to be a weak trend for the cell transplantation with stimulation group to have a shorter duration on the Rotarod at day 1 compared with baseline, but differences were not statistically significant (p = 0.33, t test). The horizontal ladder walk performance of the cell transplantation group alone and the cell transplantation plus electrode groups were essentially unchanged from baseline on all post-surgery test days. There appeared to be a weak trend for the cell transplantation with stimulation group to have a greater number of foot slips on the ladder walk on day 1 compared with baseline, but differences were not statistically significant (p = 0.32, t test). These tests suggest that the surgery and stimulation did not induce significant alterations of motor function (Hamm et al., 1994, Metz and Whishaw, 2009, Stout et al., 2013).

We Stimulated and Guided Migration of hNSCs Transplanted into the Middle of the RMS

With no electrical stimulation (no electrode insertion control group, and electrode insertion with no electrical stimulation sham group), hNSCs transplanted into the middle of the RMS migrated uniformly toward the OB (Figures 4A and 4E), consistent with previous reports of migration of hNSCs toward the OB when transplanted in the SVZ (Flax et al., 1998, Tabar et al., 2005). These results confirmed the robust intrinsic guidance signals toward the OB. When an electrical stimulation was applied toward the LV (upstream of the RMS), significant numbers of cells were found near the ipsilateral LV region (Figures 4B and 4B″), which was not observed in any of the control and sham brains (Figures 4A″ and 4E″). Serial sections of the brain demonstrated consistent migration of hNSCs upstream toward the ipsilateral SVZ (Figure 4D). Co-staining with human SOX2 antibodies confirmed the undifferentiated status of the migrating hNSCs (Figure 4C). To exclude the possibility of electrode insertion as a cause of inducing migration of hNSCs toward the LV direction, we positioned electrodes the same way without switching on the field (sham group). No EGFP signals were observed around the electrode near the LV region (Figures 4E and 4E″). In brain sections with the injection sites with electrode positions visible, we counted the EGFP-positive cells and measured the distance between the transplant site and cells furthest in the OB direction, and in the LV direction. Following electrical stimulation, NSCs were observed upstream near the LV region. No NSCs were observed in the LV region in control group and sham group (Figure 4F). Maximal migration distance of hNSCs upstream to the LV region was found to be the same as that to the OB when the electrical stimulation was applied to the direction of LV (Figure 4G).

The Guided Migration and Enhanced Cell Motility in the Brain Persisted Long after Stimulation

To determine longer-term cell survival, we examined serial sagittal sections of the rat brains 3 weeks and 4 months after electrical stimulation. Detailed examination of the serial brain sections revealed that NSCs migrated from the injection site to the LV region and to the contralateral hemisphere (Figure 5B). EGFP signals were detected further away from the transplant site even beyond the electrode in the posterior region of the SVZ, and in the corpus callosum of the contralateral hemisphere, which were not found in the control and sham brains (Figures 5 and S6). The migration of transplanted cells beyond the attractive electrode to the LV region and contralateral side suggests that the mobilizing effect of the electrical stimulation on hNSCs persisted even after the stimulation.

Using the transplant site as the middle point, we counted EGFP dots detected in the OB direction and that in the LV direction on the ipsilateral side, the sum being 100% (Figures 5B and 5C). Three weeks later, predominant signals of EGFP were in the LV region. Compared with the results immediately after electrical stimulation, more hNSCs were detected in the LV region than in the OB direction (compare Figure 5A with Figure 4A). Consistently, brains 4 months after stimulation showed an even greater proportion of hNSCs in the LV region (Figure 5C). The electrical stimulation not only guided cells to the SVZ, but the cells that traveled to in the LV region appeared to survive and maintain EGFP expression longer than the cells without stimulation.

We examined differentiation markers in hNSCs 3 weeks after transplantation and stimulation. No glial fibrillary acidic protein (GFAP) and neuron-specific class III β-tubulin (TUJ1) staining was found to co-localize with EGFP signals (Figure 6A). Four months after transplantation and stimulation, sparse EGFP signals could be shown to co-localize with antibodies against NEUN, GFAP, IBA1, and myelin basic protein (MBP), suggesting differentiation of some hNSCs to neuron, astrocytes, microglia, or oligodendrocytes (in CA1/CA2 region of hippocampus, in OB, in hippocampus, and in RMS, respectively.) (Figure 6B). These results are consistent with previous reports that transplanted hNSCs were able to differentiate in rodent brains (Flax et al., 1998, Tabar et al., 2005).

Discussion

One unmet need in brain regenerative medicine is to effectively and safely mobilize and guide NSCs to migrate to the appropriate brain lesion sites for repair. Researchers have demonstrated the rich availability of NSCs both endogenously and through transplantation (Arvidsson et al., 2002, Bond et al., 2015, Gage, 2000). Inefficient migration, however, is one of the barriers to effective clinical use (Kornblum, 2007). For efficient treatment, transplanted stem cells must establish functional connections with the host cells to repair damage and restore function (Karp and Leng Teo, 2009, Khaldoyanidi, 2008, Laflamme and Murry, 2005). In most cases, very few stem cells are able to migrate to injured or diseased regions and integrate structurally and functionally in the well-differentiated host tissues (Rakic, 2004).

Our results demonstrated the feasibility of the application of electrical stimulation directly in the brain in a manner that stimulates and guides the migration of NSCs. Our optimized stimulation strategy and electrode system is able to deliver and monitor stable stimulation to the brain with minimized detrimental effects to a tolerable level, supported by assessment of EEG and motor function. In the RMS, where intrinsic signals normally guide NSCs toward the OB, the applied electrical signals were able to override the effect of intrinsic signals and guided some NSCs toward the SVZ. Significantly, electrical stimulation mobilized transplanted hNSCs to migrate beyond the attractive electrode to the SVZ and to a limited extent to the corpus callosum. hNSCs were found in the SVZ 3 weeks and 4 months after transplantation, long after EGFP signals disappeared in the OB. The hNSCs appeared to colonize some sites in the SVZ. Does this indicate that transplanted and electrically stimulated cells were able to replenish the stem cell reservoirs? We believe that this is a very important question to be answered in the future. A few transplanted cells appeared to start to express nerve cell markers, suggesting potential differentiation of those cells. The short-term guidance effects, and potential long-term motility increase and differentiation, suggest a useful aspect of electrical stimulation in brain function regulation through stem cells. We therefore provide proof-of-concept results to use electrical stimulation to guide and mobilize NSCs in the brain in vivo. One should note with caution that the RMS is a specific permissive area for neuronal migration in the adult brain. The microenvironment in injured brain is significantly different from that within the RMS. Future investigation is therefore essential to test subsequent hypotheses about the use of electrical stimulation to mobilize and guide migration of NSCs in diseased/injured brains.

Some neuroblasts migrate long distances from their place of origin to the resident destination throughout development and into adulthood (Altman, 1969, Alvarez-Buylla and Lim, 2004, Lois and Alvarez-Buylla, 1994, Luskin, 1993). Experiments using rodent models have produced significant insights into mechanisms that regulate NSC migration and show that chemical gradients are important guides in NSC migration (Aguirre et al., 2010, Duan et al., 2007, Famulski et al., 2010, Gaiano, 2008, Ishizuka et al., 2011, Li et al., 1999, McKay, 1997, Molnar and Clowry, 2012, Wang et al., 2011, Wu et al., 1999). Although powerful, chemical gradients are difficult to control in vivo. Application of EFs has flexibility of varying strength, time, and direction across long distances. The electric signal can be immediately switched on and off, and strength adjusted if so desired. Electrical guidance therefore may provide a useful approach in human brain, which is significantly larger—over a thousand times larger—than that of mouse or rat. EFs may be able to help to unify multiple cues in guiding cells as is currently being used in the treatment of epithelial migration in wound healing (Zhao et al., 2006). Further testing and optimization of electrical stimulation is needed in larger-brained animals and eventually in human brains. Development of stimulation technology in primates will lead to technology with promising clinical applications in human patients. Recent developments in deep brain stimulation technology and in vivo wearable electrode arrays suggest promising tools for regulation of function and structure of migrating hNSCs (Santhanam et al., 2006, Viventi et al., 2011).