Dual Effects of Korean Red Ginseng on Astrocytes and Neural Stem Cells in Traumatic Brain Injury: The HO-1–Tom20 Axis as a Putative Target for Mitochondrial Function

Astrocytes display regenerative potential in pathophysiologic conditions. In our previous study, heme oxygenase-1 (HO-1) promoted astrocytic mitochondrial functions in mice via the peroxisome-proliferator-activating receptor-γ coactivator-1α (PGC-1α) pathway on administering Korean red ginseng extract (KRGE) after traumatic brain injury (TBI). In this study, KRGE promoted astrocytic mitochondrial functions, assessed with oxygen consumption and adenosine triphosphate (ATP) production, which could be regulated by the translocase of the outer membrane of mitochondria 20 (Tom20) pathway with a PGC-1α-independent pathway. The HO-1–Tom20 axis induced an increase in mitochondrial functions, detected with cytochrome c oxidase subunit 2 and cytochrome c. HO-1 crosstalk with nicotinamide phosphoribosyltransferase was concomitant with the upregulated nicotinamide adenine dinucleotide (NAD)/NADH ratio, thereby upregulating NAD-dependent class I sirtuins. In adult neural stem cells (NSCs), KRGE-treated, astrocyte-conditioned media increased oxygen consumption and Tom20 levels through astrocyte-derived HO-1. HO inactivation by Sn(IV) protoporphyrin IX dichloride in TBI mice administered KRGE decreased neuronal markers, together with Tom20. Thus, astrocytic HO-1 induced astrocytic mitochondrial functions. HO-1-related, astrocyte-derived factors may also induce neuronal differentiation and mitochondrial functions of adult NSCs after TBI. KRGE-mediated astrocytic HO-1 induction may have a key role in repairing neurovascular function post-TBI in peri-injured regions by boosting astrocytic and NSC mitochondrial functions.


Introduction
In central nervous system (CNS) injuries, Korean red ginseng extract (KRGE) and its components (e.g., ginsenoside) have favorable effects on neurovascular regeneration and anti-inflammation [1]. Several ginsenosides have improved behavior in animal models of neurological deficits [1]. Ginsenoside Rg1 is involved in neurotrophic factor-mediated adult hippocampal neurogenesis and exhibits antidepressant activity [2]. Ginsenoside Rb1 may be protective against traumatic brain injury (TBI) by enhancing the gap junction [3]. KRGE pretreatment protects against acute sensorimotor deficits and promotes its long-term recovery after ischemic stroke through the nuclear factor erythroid 2-related factor 2 (Nrf2) pathway [4].
consequently occluded for 4 h at 37 • C in an incubator. The media were replaced with differentiation media (StemCell) and astrocyte-conditioned media (ACM) in a 1:1 ratio. These cells were incubated at 37 • C in a normal oxygen (O 2 ) incubator (Thermo Fisher Scientific) for 4 days.

Oxygen-Glucose Deprivation and Astrocytes Media Preparation
Upon reaching 80% density, we incubated primary human brain astrocytes with 0% FBS in no-glucose DMEM media (Thermo Fisher Scientific). OGD was induced by perfusing 90% N 2 , 5% CO 2 , 5% H 2 -containing gas for 15 min in a hypoxia chamber (Billups-Rothenberg) with consequent occlusion for 8 h. After removing the dishes from the chamber, these cells' media were replaced into 0% FBS-containing DMEM (HyClone) with distilled water or 250 µg/mL KRGE for 24 h (OGD/R) at 37 • C in a normal O 2 incubator (Thermo Fisher Scientific). The astrocytes' media were collected and centrifuged at 1500 rpm for 5 min. The supernatant was kept in a −70 • C deep freezer for ACM.

Transfection in Astrocytes
Upon reaching 70% confluence, we transiently transfected the astrocytes with small interfering ribonucleic acids (siRNAs) for Nampt, SIRT1, SIRT2, SIRT3, Tom20, HO-1 (50 nM, Santa Cruz Biotechnology), or a negative control (50 nM, Thermo Fisher Scientific) by using RNAiMax (Thermo Fisher Scientific), based on the manufacturer's instructions. After approximately 14 h of recovery, the cells were incubated in OGD for 8 h and treated with or without KRGE for 24 h in 0% FBS with 1% penicillin-streptomycin-containing DMEM.

Cell-Counting Kit-8 Assay for Proliferation and Viability
Upon reaching 80% confluency, human astrocytes in 12-well plates (Merck Sigma) were subjected to OGD for 8 h. The media were subsequently replaced with 0% FBScontaining DMEM (HyClone). Various concentrations of KRGE were added, and the cells were incubated for 23 h. Thereafter, 30 µL of a cell-counting kit-8 (CCK-8) agent (Dojindo, Fukuoka, Japan) was added to each well, and cells were incubated at 37 • C for 1 h. We used a plate reader (Epoch Microplate Spectrophotometer; BioTek, Santa Clara, CA, USA) to determine the absorbance at a wavelength of 450 nm. The background wave was determined at 630 nm and subtracted from the value measured at 450 nm.

Cytotoxicity Analysis by Cell Survival Ratio
A cytotoxicity assay was applied using a lactate dehydrogenase (LDH) assay kit (Merck Sigma). The astrocyte medium was subjected to centrifugation for 7 min at 7000 rpm and then shifted to 96-well plates (CM). The same volume (50 µL) of the medium from plates without cells was regarded as a blank CM. Astrocytes were washed with PBS and incubated with 500 µL of 5% triton X-100 (Sigma-Aldrich) in PBS at 37 • C for 20 min. The lysed cells were centrifuged for 5 min at 15,000 rpm. We transferred the supernatant (50 µL) to 96-well plates (WCL). The supernatant from plates without cells served as the blank WCL. A dye solution was mixed with the catalyst (45:1 ratio). The aforementioned mixed reagent was added to each of the assay wells on the top of the supernatant in fast sequence. The assay plates of total volume (100 µL) were incubated at room temperature with light protection for 20 min and read using a plate reader (Epoch Microplate Spectrophotometer; BioTek) at a wavelength of 490 nm. The values were introduced into the following equation: survival (%) = (WCL value-WCL blank)/([WCL value-WCL blank] + [CM value-CM blank]) × 100.

Intracellular Nicotinamide Adenine Dinucleotide/Nicotinamide Adenine Dinucleotide Hydrogen Assay
We quantified the intracellular NAD + and NAD hydrogen (NADH) levels by using an NAD/NADH quantification kit (BioVision, Milpitas, CA, USA), based on the manufacturer's instructions. Human astrocytes cultured in 60 mm dishes were extracted with 800 µL of the extraction buffer with two freeze/thaw cycles. To detect NADH, we heated 400 µL of the extracted samples to 60 • C for 30 min. Each 50 µL extracted sample with or without heating was added to a 96-well plate, followed by the addition of 50 µL of NAD + cycling mix and subsequent incubation for 30 min. We added 5 µL of the NADH developer into the mix and incubated it at room temperature for 30 min. The plates were read at 450 nm (Epoch Microplate Spectrophotometer, BioTek). The NAD + /NADH ratio in the normoxia/recovery (i.e., control) group was set at "1," and the values in other groups were adjusted to that of the control group.

Immunocytochemistry with Mitotracker Staining
Intracellular mitochondria activity was obtained from fluorescence imaging using 1 µM of Mitotracker (Thermo Fisher Scientific, Waltham, MA, USA), a mitochondrial membrane potential-sensitive dye. Astrocytes plated on 18 mm round coverslips in 12-well plates and were cultured to 80% confluency. The cells were subjected to distilled water or 500 µg/mL KRGE for 23.5 h, then 1 µM of Mitotracker was added for 30 min. Following washing with PBS, the cells were fixed in 4% paraformaldehyde for 10 min. They were washed with 0.1% → 0.2% → 0.1% (vol/vol) Tween 20 in PBS, each for 10 min. The blocking step, using 3% BSA prepared in PBST (i.e., PBS containing 0.1% Triton X-100), was introduced for 1 h at room temperature, followed by overnight incubation with rabbit anti-Tom20 antibody (1:3000, Abcam) prepared in PBST (PBS containing 0.1% Triton X-100) at 4 • C. After washing with 0.1% → 0.2% → 0.1% (vol/vol) Tween 20 in PBS, each for 10 min, the cells were incubated with FITC-conjugated donkey anti-rabbit IgG (1:300, Jackson ImmunoResearch) for 1 h at room temperature. We consequently placed a round cover glass over each well by using a mounting agent (Fluoro-Gel II with DAPI; Electron Microscopy Sciences). The fluorescent images were obtained using a microscope (Eclipse Ti2-U; Nikon, Tokyo, Japan).

O 2 Consumption
O 2 consumption in live astrocytes or adult NSCs was detected by using the O 2 Consumption Rate Assay Kit (Cayman, Ann Arbor, MI, USA). In astrocytes, cells with 70% confluent seeded on a 60 mm dish (Merch Sigma) were transfected with the negative control (si-NC) or Tom20 siRNA (si-Tom20) for 6 h. Cells were detached using trypsin and transferred into a 96-well black polystyrene microplate (Merch Sigma). The cells were subjected to OGD/R with 250 µg/mL KRGE or with distilled water for 24 h in serum-free DMEM media. An O 2 sensor probe was added to each well. In adult NSCs, cells with 30-40% density in a laminin-coated, 96-well black polystyrene microplate (Merch Sigma) were subjected to 4 h of hypoxia, followed by 4 days of recovery. During recovery, the media were replaced with ACM and differentiation media (1:1 ratio), and NSCs were incubated for 4 days. Similar procedures in 96-well plates without a cell were conducted to evaluate the blank value. To detect the O 2 consumption rate, an O 2 sensor probe and mineral oil were added to the wells. Plates were introduced into the detector (Synergy H1, Hybrid Multi-Mode reader; BioTek) to obtain the absorbance value by using a filter combination and the emission and excitation wavelengths of 650 nm and 380 nm, respectively, at 37 • C for 75 min with 5 min interval. The value of absorbance in wells with cells and without cells at 15 min after O 2 consumption rate detection in the hypoxia/recovery group (normoxia/recovery ACM-treated control) was set at "1," and values in the other groups were adjusted to that of the control group.

Adenosine Triphosphate Levles
Intracellular adenosine triphosphate (ATP) levels were measured using an ATP colorimetric assay kit (BioVision). We transfected 70% of the confluent astrocytes with the Tom20 siRNA and exposed them to OGD/R with 250 µg/mL KRGE or with distilled water in serum-free DMEM media. They were lysed in the ATP assay buffer and centrifuged at 15,000 rpm for 5 min at 4 • C. The collected supernatant was combined with the same volume of the reaction mixture reagent (50 µL). The plates were incubated at room temperature for 30 min while being protected from light. We measured the absorbance at 570 nm by using a reader (Epoch Microplate Spectrophotometer; BioTek). The protein content in the lysed cells was quantified using bicinchoninic acid (Thermo Fisher Scientific) and measured at 562 nm using an Epoch Microplate Spectrophotometer (BioTek). The ATP levels/protein amount (ATP/protein amount) in the control group were subsequently set to 1, whereas the levels in the other groups were adjusted to that of the control group.

Data Analysis
The ImageJ (http://rsb.info.nih.gov/ij/ (accessed on 1 July 2021)) program was applied to detect the intensity of the protein band obtained from the Western blot experiments and immunofluorescence from immunocytochemistry. Values were analyzed using Prism 6 (GraphPad, San Diego, CA, USA). We conducted multiple comparisons using the one-way analysis of discrepancy and Tukey's test (data are presented as the mean ± the standard deviation (SD)). Values of p < 0.05 were statistically significant (* p < 0.05, ** p < 0.01, and *** p < 0.001).

KRGE Induces HO-1-Mediated Tom20 in a PGC-1α-Independent Manner in Astrocytes
To determine the relationships among mitochondrial function-related proteins such as HO-1, Tom20, and PGC-1α, we transfected astrocytes with specific siRNA. Astrocytes subjected to KRGE in OGD/R elevated the expression of HO-1, Tom20, and PGC-1α. Moreover, these effects were effectively reduced by HO-1 knockdown using siRNA for HO-1 (si-HO-1) (Figure 2a). siRNA for PGC-1α (si-PGC-1α) did not reduce the expression of Tom20 and HO-1 (Figure 2b), whereas siRNA for the translocase of the outer membrane of mitochondria (si-Tom20) did not decrease the protein levels of PGC-1α and HO-1 (Figure 2c). Therefore, HO-1 regulated both Tom20 and PGC-1α based on different pathways. Furthermore, we examined the function of KRGE-mediated Tom20 by examining the components of mitochondrial electron transport chain (i.e., cytochrome c oxidase subunit 2 (MTCO2) and cytochrome c). Diminished Tom20 expression decreased MTCO2 and cytochrome c in OGD/R-conditioned astrocytes (Figure 2d). KRGE upregulated astrocytic mitochondrial functions partly through the HO-1-Tom20 pathway in a PGC-1α-independent manner.

KRGE Induces Mitochondrial Membrane Potential and ATP Production via Tom20
We examined the roles of Tom20 in the energy production of astrocytes exposed to KRGE in OGD/R conditions. The treatment of human astrocytes with si-Tom20 significantly reduced the intensity of KRGE-induced Tom20 in OGD/R conditions, assessed with immunocytochemistry (Figure 3a, upper). The density and shape of nuclei were nearly the same in all groups and showed a similar cell condition (Supplementary Figure S2). si-Tom20 reduced the immunoreactivity of Mitotracker, a mitochondria membrane potential marker (Figure 3a, lower), which was critically increased by KRGE (Figure 3b,c). An interesting finding was that KRGE increased O 2 consumption and ATP production via a Tom20-dependent pathway (Figure 3d,e). Thus, astrocyte-derived Tom20 may have a key role in mitochondrial activity and energy production in KRGE-treated OGD/R conditions. Tom20 reduced the immunoreactivity of Mitotracker, a mitochondria membrane potential marker (Figure 3a, lower), which was critically increased by KRGE (Figure 3b,c). An interesting finding was that KRGE increased O2 consumption and ATP production via a Tom20-dependent pathway (Figure 3d,e). Thus, astrocyte-derived Tom20 may have a key role in mitochondrial activity and energy production in KRGE-treated OGD/R conditions. (a-c) Astrocytes on 12-well plates are transfected with small interfering ribonucleic acid (siRNA) for the negative control (si-NC) or with siRNA for Tom20 (si-Tom20), followed by oxygen-glucose deprivation/recovery (OGD/R) with or without KRGE. Normoxia followed by recovery (Nor/R) is the control for OGD/R. (a) Representative image of Tom20 (green), Mitotracker (red; Thermo Fisher Scientific, Waltham, MA, USA), and DAPI staining (blue) in human astrocytes (n = 3 per group; the scale bar = 25 μm). (b,c) Relative fluorescent intensity of randomized cells, measured using ImageJ (GraphPad, San Diego, CA, USA). (d) Live oxygen (O2) consumption after probe treatment is measured over 75 min, followed by the quantification of the relative O2 consumption for 15 min (n = 6 independent experiments). (e) Relative ATP level/mg protein is detected and quantified (n = 4 independent experiments). * p < 0.05, ** p < 0.01, and *** p < 0.001.

KRGE Induces the HO-1-Nampt Circuit, Thereby Leading to an Elevated NAD + /NADH Ratio in Astrocytes
Astrocyte treatment with two combinatory HO metabolites-namely CO and bilirubin-increased Nampt levels [20]. Intracellular Nampt exerts an enzymatic activity that is responsible for salvaging the pathways of NAD + synthesis, thereby activating NAD + -dependent protein deacetylases [10,20]. Thus, we determined the relationship between HO-1 and Nampt in the KRGE-treated OGD/R conditions. Astrocytes subjected to KRGE in OGD/R conditions revealed upregulated Nampt, which was significantly reduced by si-HO-1 (Figure 4a) or the HO inhibitor SnPP (Figure 4b). KRGE-induced HO-1 expression

KRGE Induces the HO-1-Nampt Circuit, Thereby Leading to an Elevated NAD + /NADH Ratio in Astrocytes
Astrocyte treatment with two combinatory HO metabolites-namely CO and bilirubinincreased Nampt levels [20]. Intracellular Nampt exerts an enzymatic activity that is responsible for salvaging the pathways of NAD + synthesis, thereby activating NAD + -dependent protein deacetylases [10,20]. Thus, we determined the relationship between HO-1 and Nampt in the KRGE-treated OGD/R conditions. Astrocytes subjected to KRGE in OGD/R conditions revealed upregulated Nampt, which was significantly reduced by si-HO-1 (Figure 4a) or the HO inhibitor SnPP (Figure 4b). KRGE-induced HO-1 expression in OGD/R conditions was interestingly markedly blocked by si-Nampt (Figure 4c). Nampt crosstalk with HO-1 consequently leads to the upregulation of Tom20 protein (Figure 4b,c).
We subsequently examined the NAD + /NADH ratio after astrocyte transfection with or without si-Nampt in the KRGE-treated OGD/R conditions. KRGE promoted the intracellular NAD + /NADH ratio, which was blocked by si-Nampt (Figure 4d). To evaluate NAD + -dependent class I SIRTs, we used nicotinamide (NAM) as an inhibitor for class I SIRTs [21]. The treatment of 5 mM NAM with KRGE during recovery decreased the protein levels of class I SIRTs (i.e., SIRT1, SIRT2, and SIRT3) without altering the HO-1 expression (Figure 4e and Supplementary Figure S3). The shorter size of the SIRT2 and SIRT3 bands may imply protein import into the mitochondria [11,22]. Similar to NAM, si-Nampt decreased the protein levels of SIRT1, concomitant with the shorter size of SIRT2 and SIRT3 (Figure 4f). Hence, KRGE may induce the HO-1-Nampt circuit and lead to the upregulation of SIRT1 levels and the cleavage form of SIRT2 and SIRT3.

Astrocytic HO Enhances Crosstalk between Astrocytes and Adult NSCs
Energy-producing Tom20 was induced in the astrocytes of TBI plus KRGE mouse brains and was inhibited by SnPP-mediated HO inhibition, followed by TBI plus KRGE (Figure 6a). KRGE-administered TBI mouse brains revealed closely localized GFAP-positive astrocytes and Nestin-positive NSCs (Figure 6b). Thus, we examined the impact of energetic astrocytes by KRGE administration on neighboring cells such as NSCs. KRGE- The indicated antibodies are detected in brain sections (approximately bregma −1 to −2) obtained from mice subjected to TBI, followed by KRGE treatment with or without SnPP, assessed with Western blotting (n = 3 or 4 per group). * p < 0.05, ** p < 0.01, and *** p < 0.001.

Astrocytic HO Enhances Crosstalk between Astrocytes and Adult NSCs
Energy-producing Tom20 was induced in the astrocytes of TBI plus KRGE mouse brains and was inhibited by SnPP-mediated HO inhibition, followed by TBI plus KRGE (Figure 6a). KRGE-administered TBI mouse brains revealed closely localized GFAP-positive astrocytes and Nestin-positive NSCs (Figure 6b). Thus, we examined the impact of energetic astrocytes by KRGE administration on neighboring cells such as NSCs. KRGE-treated ACM from the OGD/R conditions were transferred into adult NSCs in the hypoxia/recovery (H/R) phase for 4 days. The aforementioned ACM increased NeuN, the mature neuronal marker protein, in a concentration-dependent manner (Supplementary Figure S4). However, the direct treatment of adult NSCs with various concentrations of KRGE did not induce NeuN expression in adult NSCs (Supplementary Figure S4. We subsequently collected ACM from KRGE with or without SnPP during OGD/R conditions and transferred them into adult NSCs for 4 days. KRGE-treated ACM promoted the protein expression of mature neuronal markers such as NeuN and GAP43, compared to the non-KRGE-treated ACM (Figure 6c), which was blocked by ACM from the cotreatment of SnPP with KRGE in adult NSCs (Figure 6c). Of note, the direct treatment of SnPP with KRGE-treated ACM into NSCs did not significantly block the neuronal differentiation of NSCs (Figure 6c). Astrocytederived factors may collectively induce the mature neuronal differentiation of adult NSCs with KRGE treatment in OGD/R conditions, possibly through an HO-dependent pathway.  Figure S4). However, the direct treatment of adult NSCs with various concentrations of KRGE did not induce NeuN expression in adult NSCs (Supplementary Figure S4. We subsequently collected ACM from KRGE with or without SnPP during OGD/R conditions and transferred them into adult NSCs for 4 days. KRGE-treated ACM promoted the protein expression of mature neuronal markers such as NeuN and GAP43, compared to the non-KRGE-treated ACM (Figure 6c), which was blocked by ACM from the cotreatment of SnPP with KRGE in adult NSCs (Figure 6c). Of note, the direct treatment of SnPP with KRGE-treated ACM into NSCs did not significantly block the neuronal differentiation of NSCs (Figure 6c). Astrocyte-derived factors may collectively induce the mature neuronal differentiation of adult NSCs with KRGE treatment in OGD/R conditions, possibly through an HO-dependent pathway.

KRGE Induces NSCs' Mitochondrial Functions through Astrocytic HO-1
We determined the mitochondrial functions in adult NSCs, which could be required for neuronal regeneration in peri-injured TBI brains. Tom20 expression was used for the mitochondrial component protein. We observed Tom20 induction in Nestin-positive NSCs upon administering KRGE in mice after TBI ( Figure 7a); therefore, KRGE enhanced mitochondrial proteins such as Tom20 in NSCs and in astrocytes. The direct treatment of Figure 6. Astrocytic heme oxygenase (HO) enhances crosstalk between astrocytes and adult neural stem cells (NSCs). (a) Representative image of translocase of outer membrane of mitochondria 20 (Tom20, green) and glial fibrillary acidic protein (GFAP, red) in a mouse brain subjected to traumatic brain injury (TBI), followed by Korean red ginseng extract (KRGE) treatment with or without Sn(IV) protoporphyrin IX dichloride (SnPP, n = 3 per group. 4 ,6-Diamidino-2-phenylindole (DAPI, blue) is used for nucleus detection (scale bar = 20 µm). (b) Representative image of GFAP (green) and Nestin (red) in a mouse brain subjected to TBI, followed by KRGE treatment (the scale bar = 20 µm). (c) Adult rat NSCs were incubated under hypoxia for 4 h. The cells were recovered using astrocyte-conditioned media (ACM) obtained from KRGE with or without SnPP. After 4 days of hypoxia recovery (H/R), the NSCs' lysates underwent Western blotting. The indicated protein levels are detected in five independent experiments. * p < 0.05 and ** p < 0.01.

KRGE Induces NSCs' Mitochondrial Functions through Astrocytic HO-1
We determined the mitochondrial functions in adult NSCs, which could be required for neuronal regeneration in peri-injured TBI brains. Tom20 expression was used for the mitochondrial component protein. We observed Tom20 induction in Nestin-positive NSCs upon administering KRGE in mice after TBI ( Figure 7a); therefore, KRGE enhanced mitochondrial proteins such as Tom20 in NSCs and in astrocytes. The direct treatment of NSCs with KRGE did not induce Tom20 protein expression (Figure 7b). In contrast, treatment with KRGE-treated ACM significantly upregulated Tom20 expression (Figure 7c). Furthermore, we investigated whether transient HO-1 induction in astrocytes would elevate the neuronal differentiation and Tom20 expression in adult NSCs. The si-HO-1-KRGEtreated ACM significantly reduced the neuronal markers (i.e., Nestin, NeuN, and GAP43), as well as Tom20, compared to the si-control-KRGE-treated ACM on ACM transfer into adult NSCs for 4 days (Figure 7c). In addition, KRGE-treated ACM increased NSC proliferation, detected with the CKK-8 assay (Figure 7d) and O 2 consumption (Figure 7e). The aforementioned effects were inhibited by the combination of si-HO-1-KRGE-treated ACM (Figure 7d,e). However, we could not identify any cell cytotoxicity by si-HO-1 ACM, assessed by the LDH assay (Figure 7f). The KRGE-mediated transfer of astrocyte-derived factors may enhance neuronal differentiation after TBI. NSCs with KRGE did not induce Tom20 protein expression (Figure 7b). In contrast, treatment with KRGE-treated ACM significantly upregulated Tom20 expression (Figure 7c). Furthermore, we investigated whether transient HO-1 induction in astrocytes would elevate the neuronal differentiation and Tom20 expression in adult NSCs. The si-HO-1-KRGE-treated ACM significantly reduced the neuronal markers (i.e., Nestin, NeuN, and GAP43), as well as Tom20, compared to the si-control-KRGE-treated ACM on ACM transfer into adult NSCs for 4 days (Figure 7c). In addition, KRGE-treated ACM increased NSC proliferation, detected with the CKK-8 assay (Figure 7d) and O2 consumption (Figure 7e). The aforementioned effects were inhibited by the combination of si-HO-1-KRGE-treated ACM (Figure 7d,e). However, we could not identify any cell cytotoxicity by si-HO-1 ACM, assessed by the LDH assay (Figure 7f). The KRGE-mediated transfer of astrocyte-derived factors may enhance neuronal differentiation after TBI.

KRGE Induces Markers for Regeneration through HO Activation following TBI
The in vivo TBI model was injected with SnPP, followed by KRGE administration for 3 days. SnPP treatment diminished the expression of Tom20 and NeuN, particularly in the cornu ammonis 2 (CA2) region of the hippocampus, which increased after KRGE administration (Figure 8a). A recent report [23] reveals that mitochondrial pathways and respiration are overrepresented in CA2 cell bodies and dendrites. Tom20 surrounded the

KRGE Induces Markers for Regeneration through HO Activation following TBI
The in vivo TBI model was injected with SnPP, followed by KRGE administration for 3 days. SnPP treatment diminished the expression of Tom20 and NeuN, particularly in the cornu ammonis 2 (CA2) region of the hippocampus, which increased after KRGE administration (Figure 8a). A recent report [23] reveals that mitochondrial pathways and respiration are overrepresented in CA2 cell bodies and dendrites. Tom20 surrounded the NeuN in the CA2 region of the KRGE intake mouse brains. The aforementioned effect was markedly blocked by 50 µmol/kg and 75 µmol/kg SnPP (Figure 8a). We observed greater reductions in Nestin, NeuN, and GAP43 by 75 µmol/kg SnPP than by 50 µmol/kg SnPP after subjecting KRGE-administered TBI brains (approximately bregma −1 to −2) to Western blotting (Figure 8b). Therefore, KRGE-induced HO activity can promote neuronal repair with elevated mitochondrial functions, followed by TBI.

Discussion
KRGE exerts beneficial effects on neuroinflammatory diseases in the CNS [1]. The transient induction of astrocytic HO-1 by KRGE after TBI may promote mitochondrial functions [6]. However, researchers have not yet established the molecular mechanisms by which KRGE induces astrocytic HO-1 in TBI. Li et al. [24] reported that quercetin induces Nrf2 nuclear translocation and enhances mitochondrial functions such as mitochondria membrane potential and ATP production in TBI. The Nrf2-HO-1 pathway exerts regenerative effects by modulating inflammation and vascular remodeling such as vasculogenesis and angiogenesis [7,19]. Our findings demonstrated that Nrf2 can act as an HO-1 regulator in TBI brains and astrocytes.

Discussion
KRGE exerts beneficial effects on neuroinflammatory diseases in the CNS [1]. The transient induction of astrocytic HO-1 by KRGE after TBI may promote mitochondrial functions [6]. However, researchers have not yet established the molecular mechanisms by which KRGE induces astrocytic HO-1 in TBI. Li et al. [24] reported that quercetin induces Nrf2 nuclear translocation and enhances mitochondrial functions such as mitochondria membrane potential and ATP production in TBI. The Nrf2-HO-1 pathway exerts regenerative effects by modulating inflammation and vascular remodeling such as vasculogenesis and angiogenesis [7,19]. Our findings demonstrated that Nrf2 can act as an HO-1 regulator in TBI brains and astrocytes.
The Nrf2-HO-1 signaling cascade simultaneously activated PGC-1α and Tom20 on treating astrocytes with KRGE. The interaction between Nrf2 and PGC-1α signaling pathways is involved in mitochondrial biogenic activity [25]. PGC-1α is a master regulator of mitochondria biogenesis [13,[26][27][28]. However, our results implied that KRGE induced the PGC-1α-independent and Tom20-dependent pathways, which act as other important regulation signals for O 2 consumption and energy production in astrocytes and TBI.
KRGE induced the Nampt-HO-1 circuit in astrocytes. The combination of two HO metabolites-namely CO and bilirubin-upregulates Nampt expression [20]. In this study, we found that the upstream and circuit of HO-1 is Nampt. Therefore, intracellular NAD + levels can affect mitochondrial functions through the positive circuit of Nampt-HO-1 and the consequent class I SIRTs activation.
To the best of our knowledge, this study is the first to identify the role of KRGE-induced Tom20 in astrocytic energy-producing steps by increasing the O 2 consumption rate through HO-1-Nampt-class I SIRT (e.g., SIRT1, SIRT2, and SIRT3) pathways. NAM diminished the protein expression of KRGE-mediated SIRT1, SIRT2, and SIRT3; therefore, the inhibition of NAD + -dependent protein deacetylases reduced their expression as well. KRGE-induced Tom20 also regulates the mitochondrial membrane potential, thus establishing an active mitochondria-existing environment. We did not explore the mechanisms underlying the effect of SIRTs on Tom20 expression. We intend to determine the relationship between Nampt-mediated SIRTs activation and Tom20 in KRGE-treated astrocytes in a future study.
Astrocytes have emerged as active players in brain energy delivery by coordinately communicating with the neuronal system [29][30][31], thereby leading to neurovascular repair after CNS damage [16,30]. The transfer of ACM into NSCs regulates NSC proliferation and differentiation [32]. Researchers have not investigated the interaction between astrocytes and neuronal system in TBI through astrocytic HO-1. We hypothesized that secreting factor(s) originating from astrocytic HO-1 in the presence of KRGE may facilitate neuronal repair post-TBI. The inhibition of or reduction in astrocytic HO-1 affects the differentiation and proliferation of adult NSC. In addition, TBI brains with KRGE administration showed NeuN expression in the CA2 region, whereas this effect was blocked by SnPP cotreatment. Therefore, KRGE may accelerate astrocyte-NSCs communication, partly through astrocytic HO-1.

Conclusions
In this study, KRGE-treated astrocytes upregulated the proliferation and neuronal differentiation of adult NSCs, supposedly through astrocyte-neuronal system cooperation. Our results suggested that KRGE likely strengthens astrocyte mitochondrial function by HO-1 induction. In addition, HO-1-induced astrocytes contribute to the neuronal differentiation of adult NSCs. The HO-1-Tom20 axis may act on mitochondrial functions in intracellular and intercellular pathways. Taken together, the KRGE-induced HO-1 pathway may contribute to mitochondrial activity and energy production in astrocytes, and thereby lead to the potential improvement of neurovascular repair post-TBI.

Data Availability Statement:
The data presented in this study are contained within the article. Original data will be made available on request from the corresponding author.