Ginsenoside Rg1 can restore hematopoietic function by inhibiting Bax translocation-mediated mitochondrial apoptosis in aplastic anemia

The present study investigated, the anti-apoptotic activity of Ginsenoside Rg1 (Rg1) via inhibition of Bax translocation and the subsequent recovery of hematopoietic function. Mitochondrial apoptosis in bone marrow mononuclear cells (BMNCs) was observed in aplastic anemia (AA) patients. To establish a mouse model of AA, BALB/c mice were transplanted with lymph node cells from DBA/2 donor mice via vein injection after treatment with Co60 γ-radiation. After treatment with Rg1 for 14 days, the peripheral blood and Lin–Sca-1 + c-Kit + (LSK) cell counts of the treated group were increased compared with those of the untreated model mice. In in vivo and in vitro tests of LSKs, Rg1 was found to increase mitochondrial number and the ratio of Bcl-2/Bax and to decrease damage to the mitochondrial inner and outer membranes, the mitochondrial Bax level and the protein levels of mitochondrial apoptosis-related proteins AIF and Cyt-C by decreasing the ROS level. Rg1 also improved the concentration–time curve of MAO and COX and levels of ATP, ADP and AMP in an in vitro test. In addition, high levels of Bax mitochondrial translocation could be corrected by Rg1 treatment. Levels of markers of mitochondrial apoptosis in the Rg1-treated group were significantly better than those in the AA model group, implying that Rg1 might improve hematopoietic stem cells and thereby restore hematopoietic function in AA by suppressing the mitochondrial apoptosis mediated by Bax translocation.

www.nature.com/scientificreports/ cells (HSCs) via transplantation (HST). However, there are problems with these treatments, such as the limited efficacy of IST and the serious side effects of HST, such as graft versus host disease. Therefore, treatments aimed at restoring hematopoietic stem cell number and function a represent potential alternative approaches. Mitochondrial function is essential for cells, including HSCs, because mitochondria are the major site of adenosine-5′-triphosphate (ATP) production 1 . In addition to playing fundamental roles in energy production and metabolism, mitochondria exhibit other important functions, including the maintenance of calcium homeostasis, and regulation of cellular and intracellular signaling, inflammation, and apoptosis 2,3 . The multilineage differentiation and proliferation capacities of HSC are particularly vulnerable to inflammation and apoptosis 4,5 . Bax is a strong multidomain proapoptotic protein that resides in the cytoplasm as an inactive monomer in healthy cells 6 . Upon encountering apoptotic stimuli, Bax undergoes conformational activation, leading to its translocation to mitochondria. Studies have indicated that Bax translocation is a key mechanism of the apoptosis of human monocytes 7 .
Ginsenoside Rg1 (Rg1) is a steroidal saponin that is highly abundant in ginseng and one of its most important components 8 . Studies have shown that Rg1 can protect hematopoietic stem/progenitor cells (HSPCs) by attenuating oxidative stress 9 , improving the hematopoietic microenvironment 10 , protecting against X-ray irradiation-induced aging 11 or protecting against cyclophosphamide-induced myelosuppression in mice by recovering hematopoietic function 12 . Rg1 can alleviate oxidative stress and inflammation 13 , inhibit the excessive activation of Wnt/β-catenin signaling in aged HSPCs 9 , Rg1 protect mitochondrial function by inhibiting apoptosis through PI3K/Akt signaling 14 and improve mitochondrial activity 15 . However, whether Rg1 exerts anti-apoptosis apoptotic effects in AA by affecting mitochondrial pathway function remains unknown.
The purpose of this study was to investigate whether Rg1 can be used to effectively treat the hematopoietic stem cells (HSCs) of AA by suppressing the mitochondrial apoptosis induced by Bax translocation.
Patients and experimental protocol. Between September 2016 and February 2017, 8 patients (median age 43.6 years; range 32-64) were enrolled in this study. Eligible patients had a histologically confirmed diagnosis of AA. This study was approved by the Institutional Review Board of the Affiliated Hospital of Shandong University of Traditional Chinese Medicine, and written informed consent was obtained from all participants in accordance with the Declaration of Helsinki. Bone marrow mononuclear cells were obtained from patients for analyses of apoptosis, mitochondrial number, and Bcl-2/Bax mRNA level.
Animals and experimental protocol. Sixty healthy BALB/c male mice weighing 18-22 g and aged 6-8 weeks were provided by the Experimental Animal Center of Shandong University (China). The animals were housed in a warm, quiet environment with free access to food and water and acclimatized for one week before the experiments began. All animal procedures were performed with the approval of the Animal Ethics Committee of the Affiliated Hospital of Shandong University of Traditional Chinese Medicine. All experiments were performed in accordance with relevant guidelines and regulations.
The 60 mice were randomly divided into several groups: a normal control group, the model group and three treatment groups. The AA model was established as previously described 1,6 . Briefly, the mice were irradiated with 5.0 Gy Co60 γ-radiation, and 2 × 10 6 lymph node cells from DBA/2 donor mice were transplanted within 4 h after radiation.
The treatment groups were intraperitoneally injected with Rg1 (at 20, 40, and 80 mg/kg/day, for the low-dose group, medium-dose group and high-dose group, respectively, according to previous studies). The mice in the normal control and model groups were intraperitoneally injected with physiological saline (10 ml/kg/day). In addition, all the mice received a standard diet throughout the study. After treatment with Rg1 or physiological saline for 2 weeks, euthanasia was performed by cervical dislocation on day 14. Before euthanasia, blood was collected by puncturing the caudal vein, and after all of the animals were killed by cervical dislocation, the femur and spleen were removed immediately.
Blood samples were collected via the tail vein from the mice in all groups on day 14. Routine peripheral blood routine, counts of bone marrow mononuclear cells (BMNCs), burst-forming units-erythrocytes (BFU-E), and colony-forming units-erythrocytes (CFU-E) and bone marrow biopsy were conducted for verification of the model. Bax IHC was also performed. For the acute toxicity test, the maximum tolerated dose (500 mg/kg/ day) of Rg1 was administered for 2 days. Bone marrow and liver data are shown in Supplementary Fig. 1, which showed no toxicity of Rg1 in mice.  17 , for the in vivo experiments, bone marrow cells were obtained from the groups of animals on day 14. Lin-Sca-1 + c-Kit + (LSK) populations were sorted from bone marrow cells without RBCs by using a BD FACSAria II flow cytometer. We used FITC-labeled antibodies against lineage markers, including Mac-1, Gr-1, Ter119, CD4, CD8a, CD3, and B220 (BD Biosciences), and anti-c-Kit-APC and anti-Sca-1-PE antibodies. Sorted cells were immediately analyzed tested in IMDM (Thermo Fisher Scientific). For the in vitro experiment, bone marrow cells were obtained from all the groups of animals on day 7. After the LSK populations were sorted, LSK cells from the control and model groups were cultured in IMDMsupplemented with 10% FBS (Biological Industries, Beit-Haemek, Israel) and antibiotics (100 u/mL penicillin and streptomycin). LSK cells from the treatment groups were cultured in the same medium supplemented with FBS and antibiotics but also containing Rg1 (12.5, 25, or 50 μmol/L). These cells were cultured in a humidified incubator containing 5% CO 2 at 37 °C and then analyzed.

Analysis of apoptosis.
Cell apoptosis of BMNCs from humans or LSKs from mice cultured for 48 h was quantified using an Annexin V-FITC kit (BD Biosciences, San Jose, CA, USA) according to the manufacturer's instructions and analyzed by flow cytometry.
Mitochondrial number assay of human BMNCs. Mitochondrial number in human BMNC cells was measured by using MitoTracker Green FM (Molecular Probes) as described by Mancini 18 . Cells were collected by trypsinization, suspended in PBS and fixed with a fixative containing 2% glutaraldehyde and 2% formaldehyde in PBS for MitoTracker Green FM staining. After being washed and suspended in PBS, the cells (3 × 10 5 cells/mL) were stained with 75 nM MitoTracker Green FM for 30 min at room temperature in the dark, and then subjected to flow cytometric analysis.

Measurement of mRNA.
Genomic DNA for the analysis of mtDNA content was isolated from human BMNC cells or mouse LSK cells using a TaKaRa MiniBEST Universal Genomic DNA Extraction kit (TaKaRa Bio Inc.). The relative mtDNA copy number was determined by qPCR with primers for the mitochondrial 16S rRNA gene and the nuclear Actin gene as previously described 19 . All PCRs were performed in triplicate. The primers used to amplify 16S rRNA were 5′-GGT GCA GCC GCT ATT AAA GG-3′ (16S rRNA, forward) and 5′-ATC ATT TAC GGG GGA AGG CG-3′ (16S rRNA, reverse).

ROS staining.
Cells were collected at the selected time point, the medium was discarded, and the cells were washed three times with 1 mL of PBS buffer. One milliliter of the fluorescent probe DCFH-DA was diluted to 1:1000 added to the cells, which were then incubated at 37 °C for 30 min. The cells were then washed three times with 1 mL of PBS buffer, counterstained with DAPI, and washed another 3 times. A confocal laser scanning microscope was used to image the cells and determine the distribution and expression of green fluorescence [indicative of reactive oxygen species (ROS)].
Mitochondrial analyses by TEM morphometry. Mitochondrial structure was analyzed by electron microscopy. Then, images were obtained using a digital video camera (JVC, ky-F30B3-CCD, Japan), and the images were transferred to a computer. Then, the number, perimeter and area of each mitochondrial crosssection were calculated (Shen and Shen 1991). The measurements from all images were based on the same reference area and obtained by the image analysis system (Kontron Ibas 2.0 Germany). The quantitative density parameters [volume density (Vv) and numerical density (Nv)] and average surface area (S) were calculated based on the method of Ref. 21 .

Mitochondrial lysis time curves in vitro.
After mitochondria were extracted from LSK cells incubated in medium with or without Rg1, MAO and COX concentrations were detected as biomarkers of mitochondrial membrane lysis. The concentration of MAO indicates the concentration of mitochondria and was determined using 200 U/mL mitochondrial suspensions from serum cultures within 12 h of harvest at different time points (separated by 1-hintervals). The concentration of COX in the medium was also determined. The peak time analysis of the mitochondrial membrane and matrix-specific enzyme concentration-time curves revealed the trend in mitochondrial membrane lysis after treatment with Rg1.

Determination of ATP, ADP and AMP concentrations in vitro.
Perchloric acid and high-performance liquid chromatography (HPLC) were used to determine the ATP, ADP and AMP levels 22 . LSK cells were lysed in 50% perchloric acid and then centrifuged (at 4 °C and 12,000 rpm for 30 min). As soon as the supernatant was transferred to a new centrifuge tube, 2.5 M K 2 CO 3 was added, and the solution was centrifuged under the same conditions. Then, the supernatant was immediately analyzed by HPLC under the following detection conditions: HYPERSIL C18 5u analytical column; 250 mm column length; 4.6 mm column diameter;

Colocalization analysis of mitochondria with Bax in vitro. The colocalization of mitochondria with
Bax was observed by confocal fluorescence microscopy 23 . To explore the connection between mitochondria Statistical analysis. Statistical analysis was performed using a computer software SPSS Version19.0 (SPSS Inc., Chicago, IL, USA). The data are expressed as the means ± SDs. One-way ANOVA was used to test the significance of differences among groups, followed by Scheffe's modified F-test for multiple comparisons. A value of P < 0.05 was considered statistically significant.
Ethics approval and consent to participate. All experiments were conducted in compliance with the ARRIVE guidelines. We confirm that animal care and experimental procedures were carried out in accordance with the guidelines of the Animal Ethics Committee of the Affiliated Hospital of Shandong University of Traditional Chinese Medicine. The reference number is 2015-5-07.

Results
The mitochondrial apoptosis in BMNCs. There were significantly higher levels of apoptosis in BMNCs from AA patients than in those from normal individuals according to the flow cytometry data (Fig. 1A). As presented in Fig. 1B, the mitochondrial number determined by using molecular probes was significantly lower in the BMNCs from AA patients' than in those from normal individuals (P < 0.01). We also analyzed the number of mitochondrial DNA copies by real-time PCR. There was greater loss of mtDNA in AA patients than in control subjects (Fig. 1C). To verify this result, we investigated the levels of mitochondrial apoptosis-associated markers. The relative level of Bax of AA patients was significantly increased compared with that of normal individuals, whereas no marked decrease in Bcl-2 levels were observed in AA patients (Fig. 1D).
Effect of Rg1 treatment on the recovery of hemopoietic function in AA mice. Using a fully automated blood cell analyzer, we found that the numbers of peripheral blood cells and BMNCs and the area of hematopoietic tissue in bone marrow biopsies from the AA mice were notably decreased compared to those from the control mice (P < 0.01), indicating the successful establishment of the mouse model of AA (Fig. 2). Treatment of these mice with medium-or high-dose Rg1 for 14 h increased survival time (P < 0.01; Fig. 2B), and WBC, RBC, PLT and BMNC counts as well as weight in the treated mice were increased compared with those of the model control group (P < 0.01; Fig. 2C-E).

Effect of Rg1 treatment on BFU-E and CFU-E in AA mice. The counts of BFU-E and CFU-E in the
bone marrow of AA mice were found to be significantly lower than those in the bone marrow of normal control mice (P < 0.01). Furthermore, microscopy revealed depressed growth and proliferation of erythroid progenitor cells in the AA mice. After treatment with medium-dose Rg1 (40 mg/kg/day), the BFU-E and CFU-E counts were restored to 66.8% (17.5/26.2) and 77.25% (59.1/76.5), respectively, of the normal levels (Fig. 3A).

Effect of Rg1 treatment on mitochondrial quantity of LSKs. After treatment with 40 mg/kg/day
Rg1 or saline solution for 14 days, LSK cells were obtained from mouse bone marrow for mitochondrial apoptosis tests. LSK level significantly recovered in the treatment group (Fig. 3B), and the data were similar to the in vitro data (Fig. 4A). Rg1 also increased mitochondrial number (Fig. 3C) and alleviated the apoptosis level (Fig. 3D) compared with the number and level observed in the model group (P < 0.01). Transmission electron microscopy of the mouse bone marrow showed that the size and shape of the mitochondria were normal in the control group. In contrast, the mitochondria in the model control group exhibited enlarged globular structures and the disruption or disappearance of cristae (Fig. 3D). Mitochondrial number was significantly improved after treatment with Rg1, indicating that Rg1 recovered mitochondrial number in AA mice.

Effect of Rg1 treatment on mitochondrial apoptosis in LSKs. The levels of apoptosis and related
proteins were investigated in LSKs. The flow cytometry data showed that Rg1 could alleviate the abnormal increased in apoptosis observed in the model control group (Fig. 3E). The mechanism was as follows: Rg1 decreased the levels of mitochondrial apoptosis markers such as Bax, Cyt-C, Apaf-1 and AIF and cleaved caspase 3 (Fig. 3G), by inhibiting the production of ROS (Fig. 3F). Importantly, there was no significant decrease in Bak   www.nature.com/scientificreports/ level after Rg1 treatment (Fig. 3G). The Bax translocation ratio is important in the induction of mitochondrial apoptosis, and Rg1 can inhibit this induction (Fig. 3H). The Bax IHC data supported these results (Fig. 3I).

Effect of Rg1 treatment on the mitochondrial lysis time curve.
In the control group, the COX level peaked after 3.5 h. The MAO level in the culture did not change significantly over time, but the COX level gradually increased with time until peaking and then gradually declining. Thus, the complete cleavage of mitochondrial contents released by cells required approximately 3.5 h to peak. The COX peaks in the AA group appeared at 1.5 h and 1.375 U/L. The COX peaks in the treatment group appeared at 5.5 h and 1.341 U/L. At 1.5 h, the COX level in the AA group was significantly higher than that in the treatment group (P < 0.05).
Therefore, the complete cleavage of mitochondrial contents in the serum was significantly delayed after the addition of Rg1 (P < 0.05), and the peak was slightly lower than that in the AA group (Fig. 4B,C).
Influence of Rg1 treatment on the recovery of energy generation in vitro. The ATP/ADP level indicates the ability of HSCs to generate energy. The ATP/ADP level in the AA group and treatment group decreased gradually over time, but was higher in the treatment group than in the AA group at 24 h and 48 h (P < 0.05; Fig. 4D,E). In contrast to the changes observed in ATP level, the AMP level in model and treatment groups was significantly elevated relative to the control level and increased markedly in the AA group. At 48 h, the AMP content in the treatment group was lower than that in the AA group (P < 0.05; Fig. 4F). These data showed that Rg1 could ameliorate the dysregulation of energy generation in AA.

Influence of Rg1 treatment on Bax translocation-related mitochondrial apoptosis.
The levels of protein markers of mitochondrial apoptosis were detected to determine the effects of Rg1 in vitro. The quantitative protein expression data were normalized to β-actin expression and are shown as a percentage of the expression in the control group and as the means ± standard deviations (n = 3). The data demonstrated that Rg1treatment decreased the ROS level (Fig. 4G) and rectified the abnormal protein levels of Bcl-2/Bax by decreasing the levels of Bax, Cyt-C, Apaf-1, AIF and cleaved caspase 3 at 48 h (P < 0.01) (Fig. 4H). However, these levels were not restored to normal levels either in vivo or in vitro (P < 0.01). The Bax translocation level was then assessed by using Western blotting, and mitochondrial colocalization with Bax was investigated. Rg1 decreased the ratio of Bax mitochondrial translocation (Fig. 4H) in vivo and alleviated the colocalization of Bax with mitochondria (Fig. 4J). Rg1 also restored mitochondrial number and structure, as revealed by PCR and TEM analyses (Fig. 4I,K).

Discussion
Mitochondrial dysfunction and a decrease in mitochondrial number may result in a reduction in mtDNA. As evidence of acquired mtDNA deletions in the hematopoietic compartment has been found, mtDNA mutations and severe pancytopenia or reticulocytopenia are believed to be closely related 24 . Events such as mtDNA damage and abnormal mtRNA transcription, protein synthesis, and mitochondrial function can lead to mitochondrial injury, and both mtDNA mutation and mtDNA copy number reduction are causes of disease 25,26 . If the damaged mitochondria are cleared, erythrocyte maturation and homeostasis can be accelerated 27 . Studies have clarified that mitochondrial apoptosis is a key factor in AA. ROS-dependent pathway is always an important pathway which leads to the apoptosis death via mediating Bax translocation 28 or JNK-p38 29 and other pathways. Increased ROS defects contribute to severe combined anemia and thrombocytopenia 30 . By inhibiting abnormal mitochondrial oxidative phosphorylation, rapamycin can ameliorate the phenotype of the immune-mediated AA model 31 . If mitochondrial dysfunction is decreased, the genomic and functional integrity in the hematopoietic system can be protected 32 . Furthermore, in the AA rat model, the reversal of abnormal levels of mitochondrial DNA content and function, can restore the characteristics of healthy rats 33 .
Rg1 has been used treat anemia and bone marrow damage for many years. It can alleviate hematopoietic homeostasis defects 34 , delay hematopoietic stem/progenitor cell senescence by influencing the SIRT6/NF-κB signaling axis 35 , and protect against HSC aging by regulating the SIRT1-FOXO3 and SIRT3-SOD2 signaling pathways 36 or by regulating bone marrow stromal cells 37 to recover hematopoietic function. The anti-apoptosis, effects of Rg1 have been proven in several studies, with Rg1 being capable of protecting cardiomyocytes 38 and lung epithelial cells 39 .In addition, in hematology studie, Rg1 has been shown to inhibit mitochondrial dysfunction 40 and, delay senescence in BMNCs 37 . Rg1 can also inhibit apoptosis by decreasing Bax level and restore an abnormal Bcl-2/Bax ratio to normal levels 36,41,42 . Therefore, we investigated whether Rg1 can mitigate AA by preventing mitochondrial apoptosis in hematopoietic cells.
In our study, significant apoptosis and a decrease in mitochondrial number were found in AA patients due to an abnormal Bcl-2/Bax ratio (Fig. 1). We established a mouse model of AA by applying a combination of 60Co γ-radiation and transplantation with lymph node cells from DBA/2 donor mice. The AA mice showed statistically significant reductions in the measures of peripheral blood leukocytes, hemoglobin and PLTs and severe reductions in the numbers of BMNCs (Fig. 2B-D) and marrow-committed progenitor cells (Fig. 3A), which are clinical characteristics of AA. Furthermore, unbalanced Bcl-2/Bax ratio-induced mitochondrial apoptosis was verified in the mouse model (Fig. 3G,I). AA mice treated with Rg1 showed an increase in the number of BMNCs (Fig. 2C), and with the increases in BFU-E and CFU-E counts (Fig. 3A), LSK cell levels increased (Figs. 3B, 4A) compared with those in the AA group. Additionally, the increase in mitochondrial number (Figs. 3C, 4I) and the decreases in mitochondrial apoptosis level in vivo and in vitro verified the positive effect of Rg1 in AA mice (Figs. 3E-I, 4H). Similar to Meng's data 43 , Rg1 is proved to decrease the ROS level (Fig. 3F, 4G) which is a main reason for preventing apoptosis. Moreover, the anti-apoptosis effect of Rg1 was found to be mediated by its inhibition of Bax www.nature.com/scientificreports/ translocation, as evidenced by the decrease in the level of colocalization of Bax with mitochondria following Rg1 without a significant decrease in Bak level. Furthermore, we found that Rg1 treatment corrected the abnormal mitochondrial lysis time curve and energy level of LSK cells in vitro (Fig. 4B-F). To further explore the mechanism, we analyzed the Bax ratio data of mitochondria and total and found that Rg1 inhibited Bax assembly from the cytoplasm in mitochondria (Figs. 3H, 4H). The colocalization of Bax with mitochondria was inhibited by Rg1 (Fig. 4J), resulting in apoptosis resistance (Fig. 3E). These results showed that Rg1 recovered hematopoietic function by promoting BMNC proliferation, increasing mitochondrial number, stabilizing the mitochondrial membrane and restoring the energy supply by inhibiting Bax translocation-induced mitochondrial apoptosis, and that Rg1 treatment prolonged the survival of AA mice.
In brief, AA model mice exhibit severe issues with mitochondrial apoptosis that can be ameliorated by treatment with Rg1. Thus, therapeutic targets that can maintain mtDNA integrity and copy number may be considered crucial players in the treatment of AA.

Data availability
The datasets supporting the conclusions of this article are included within the article. The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.