**2. Electrical stimulation for cerebral infarction**

After the onset of cerebral infarction, more endgenous stem cells are produced in the SVZ and these endogenous stem cells migrate and differentiate into neurons in order to replace the area of infarction. Most of these newly produced endogenous stem cells cannot survive, however, and die within a few weeks [8]. That is why we expected that enhancing the survival of these newly produced endogenous stem cells or enhancing the production of endogenous stem cells by electrical stimulation could result in an improved neurorescue effect.

The invasiveness of the electrical stimulation of the brain varies with the position of the electrode, e.g., brain parenchyma, brain surface (subdural), epidural space, or skin. In the DBS for Parkinson's disease, electrodes are inserted in deep brain areas such as the subthalamic nucleus. Although DBS is considered to be one of the gold standards of therapy for Parkinson's disease, less invasive and more effective stimulation techniques are urgently required. Compared to the use of exogenous stem cells such as stem cell transplantation, the enhance‐ ment of endogenous stem cells is expected to be less invasive. We believe that continuous epidural stimulation is the best choice as it is less invasive and can be expected to produce a similar neuroprotective effect as DBS.

To examine the neurorescue effect of electrical stimulation for cerebral infarction, we applied epidural electrical stimulation to an acute-stage cerebral infarction model of rats. To identify the best stimulation parameter, we stimulated the rats with different frequencies and intensi‐ ties (see Table). We found that continous electrical stimulation with low frequency produced a better neurorescue effect compared to that with high frequency. Moreover, this lowfrequency continous epidural electrical stimulation had a better neurorescue effect through an increase in trophic factors such as BDNF, GDNF, and VEGF (Fig.1), suppressed inflamatory response (Fig. 2), enhanced angiogenesis (Fig.2) and played anti-apoptotic effect with the upregulation of phosphorylated Akt (Fig.3). LY294002 administration into the lateral ventricle suppressed the neurorescue effect of electrical stimulation (Fig.4), which showed that this antiapoptotic effect was exerted through the PI3K-Akt signaling pathway [9].

ment showed that it took several weeks to obtain a sufficient number of autologous adult neural stem cells [5]. Many researchers have pointed out that the effect of these stem cell transplantations is derived not from cell replacement, which is the original purpose of stem cell transplantation, but mainly from the trophic effect brought about by transplanted stem cell secretions. Generally speaking, compared to exogenous stem cell transplantation, the enhancement of endogenous stem cells is expected to be less invasive. Previous reports showed that an enriched environment and running enhanced endogenous stem cell generation [6] and that trophic factors produced by these enhanced endogenous stem cells had a neuroprotective effect on these neurological disorders. Deep brain stimulation (DBS) is one of the gold standards of treatment for Parkinson's disease (PD) in current clinical pratice. Recent reports have showed that DBS can enhance endogenous stem cells in a PD model of rats [7]. Although phase 3 or 4 clinical trials with DBS for depression and obesity and epidural electrical stimu‐ lation for pain are currently ongoing, they are not being performed for regenerative medicine (http://clinicaltrials.gov). Based on this report, we examined whether electrical stimulation has

After the onset of cerebral infarction, more endgenous stem cells are produced in the SVZ and these endogenous stem cells migrate and differentiate into neurons in order to replace the area of infarction. Most of these newly produced endogenous stem cells cannot survive, however, and die within a few weeks [8]. That is why we expected that enhancing the survival of these newly produced endogenous stem cells or enhancing the production of endogenous stem cells

The invasiveness of the electrical stimulation of the brain varies with the position of the electrode, e.g., brain parenchyma, brain surface (subdural), epidural space, or skin. In the DBS for Parkinson's disease, electrodes are inserted in deep brain areas such as the subthalamic nucleus. Although DBS is considered to be one of the gold standards of therapy for Parkinson's disease, less invasive and more effective stimulation techniques are urgently required. Compared to the use of exogenous stem cells such as stem cell transplantation, the enhance‐ ment of endogenous stem cells is expected to be less invasive. We believe that continuous epidural stimulation is the best choice as it is less invasive and can be expected to produce a

To examine the neurorescue effect of electrical stimulation for cerebral infarction, we applied epidural electrical stimulation to an acute-stage cerebral infarction model of rats. To identify the best stimulation parameter, we stimulated the rats with different frequencies and intensi‐ ties (see Table). We found that continous electrical stimulation with low frequency produced a better neurorescue effect compared to that with high frequency. Moreover, this lowfrequency continous epidural electrical stimulation had a better neurorescue effect through an increase in trophic factors such as BDNF, GDNF, and VEGF (Fig.1), suppressed inflamatory response (Fig. 2), enhanced angiogenesis (Fig.2) and played anti-apoptotic effect with the

therapeutic potential for CNS diseases by activating endogenous stem cells.

by electrical stimulation could result in an improved neurorescue effect.

**2. Electrical stimulation for cerebral infarction**

746 Regenerative Medicine and Tissue Engineering

similar neuroprotective effect as DBS.


**Table 1.** The table presents Limb Placement Test (LTP) scores (3 days and 1 week after MCAO) and the percentages of infarct volumes measured 3 days post-MCAO relative to the intact side of rats receiving electrical stimulation with several parameters. Initially, 0, 2, 10, and 50 Hz were used with 100 μA current. Next, exploration using 200 μA was performed. Based on this experiment, we decided that 2 Hz 100 μA stimulaton was the optimal therapeutic condition. This table was reproduced and/or modified from the original article [9].

**Figure 1.** (A) The schematic diagram shows the regions in the cortex and striatum of rats with electric stimulation. Black box, cortex; white box, striatum. Each pair of brain tissues in the cortex and striatum per hemisphere was punched. (B) Upregulation of neurotrophic factors such as BDNF, GDNF, and VEGF was observed in the stroke rats re‐ ceiving electric stimulation. (C) Quantification. \**p* < 0.05 versus cortex of the control rats. \*\**p* < 0.05 versus striatum of the control rats. This figure was reproduced and/or modified from the original article [9].

**Figure 2.** Laminin staining (A–D and I) and Iba-1 staining (E–H and J) show angiogenic and anti-inflammatory effects in the stimulation cortex were exerted by electric stimulation. A and E are the intact cortex, B and F are nonstimulated cortex, and C, D, G, and H are the ischemic cortex (D and H are with high magnification). \**p* < 0.05 versus cortex of the control rats (I and J). Scale bar: 100 μm (A–C, E–G), 25 μm (D, H). This figure was reproduced and/or modified from the original article [9].

**Figure 3.** Anti-apoptotic effects of electric stimulation were exerted through phosphorylated Akt. (A-C): TUNEL stain‐ ing revealed the anti-apoptotic effect of electrical stimulation (B) compared to those in the control group (A). Scale bar: 100 μm. (C): Quantitative analyses. \**p* < 0.05 versus cortex of the control rats. (D–F): Phosphorylated Akt staining revealed a surge of stained cells in the ischemic cortex with electric stimulation (E) compared with the control group (D). (F) Quantitative analyses. \**p* < 0.05 versus cortex of the control rats. \*\**p* < 0.05 versus striatum of the control rats. This figure was reproduced and/or modified from the original article [9].

**Figure 4.** LY294002 blocked the neuroprotective effects of electric stimulation. The infarct volumes of rats receiving electric stimulation (B) were significantly decreased compared with those of control rats (A) and ischemic rats that re‐ ceived LY294002 alone (C). In contrast, the neuroprotective effects of electric stimulation were blocked by intraven‐ tricular administration of LY294002 (D). Quantitative analysies of infarct volumes is shown in E. \**p* < 0.05 versus all other groups. This figure was reproduced and/or modified from the original article [9].

**Figure 2.** Laminin staining (A–D and I) and Iba-1 staining (E–H and J) show angiogenic and anti-inflammatory effects in the stimulation cortex were exerted by electric stimulation. A and E are the intact cortex, B and F are nonstimulated cortex, and C, D, G, and H are the ischemic cortex (D and H are with high magnification). \**p* < 0.05 versus cortex of the control rats (I and J). Scale bar: 100 μm (A–C, E–G), 25 μm (D, H). This figure was reproduced and/or modified from the

**Figure 3.** Anti-apoptotic effects of electric stimulation were exerted through phosphorylated Akt. (A-C): TUNEL stain‐ ing revealed the anti-apoptotic effect of electrical stimulation (B) compared to those in the control group (A). Scale bar: 100 μm. (C): Quantitative analyses. \**p* < 0.05 versus cortex of the control rats. (D–F): Phosphorylated Akt staining revealed a surge of stained cells in the ischemic cortex with electric stimulation (E) compared with the control group (D). (F) Quantitative analyses. \**p* < 0.05 versus cortex of the control rats. \*\**p* < 0.05 versus striatum of the control rats.

This figure was reproduced and/or modified from the original article [9].

original article [9].

748 Regenerative Medicine and Tissue Engineering

As the next step, we applied epidural electrical stimulation to a chronic-phase cerebral infarction model of rats. However, we could not obtain the same neuroprotective effect. This is why we needed to change the method. According to the clinical application of DBS for Parkinson's disease, we decided to insert an electrode in the infarct striatum although this method is more invasive than epidural stimulation. Before giving striatum electrical stimula‐ tion to the chronic-phase ischemia model of rats, we applied electrical stimulation to normal rats and found that low-frequency continuous striatal electrical stimulation upregulates GDNF and vascular endothelial growth factor (VEGF) and enhances the neuronal differnetiation of endogenous stem cells and angiogenesis. We then applied striatal stimulation to the chronicphase ischemia model of rats. We conducted magnetic resonance imaging 28 days after ischemic induction and inserted electrodes into the striatum based on these MR images followed by electrical stimualation for a week; rats were sacrificed on day 60. In this experi‐ ment, we found that striatum stimulation exerted behavioral improvement and MR images taken after striatal stimulation showed a reduced infarct volume compared with those taken before stimulation. Moreover, striatal stimulation during the chronic-phase ischemia model of rats resulted in enhanced migration of neural progenitor cells from SVZ to ischemic penumbra, enhanced neuronal differentiation, and suppressed microglial activation in the ischemic penumbra [10].
