Angelica sinensis polysaccharides prevents hematopoietic regression in D-Galactose-Induced aging model via attenuation of oxidative stress in hematopoietic microenvironment

Extrinsic molecular mechanisms that regulate hematopoietic stem/progenitor cell (HSPC) aging are still poorly understood, and a potential protective medication needs to be explored. The senescent parameters of hematopoietic cells and bone marrow stromal cells (BMSCs) including cell cycle analysis, senescence-associated SA-β-gal staining and signals, hematopoietic factors and cellular junction were analyzed in femur and tibia of rats. Furthermore, Sca-1+ HSPCs and BMSCs co-culture system was established to evaluate the direct effects of BMSC feeder layer to HSPCs. Oxidative DNA damage indicators in Sca-1+ HSCs and senescence-associated secretory phenotype (SASP) of BMSCs, gap junction intercellular communication between BMSCs, osteogenesis/adipogenisis differentiation balance of BMSCs were detected. In the D-gal pre-administrated rats, ASP treatment rescued senescence of hematopoietic cells and BMSCs, reserved CFU-GEMM; also, ASP treatment attenuated stromal oxidative load, ameliorated SCF, CXCL12, and GM-CSF production, increased Connexin-43 (Cx43) expression. BMSCs and Sca-1+ HSPCs co-cultivation demonstrated that ASP treatment prevented oxidative DNA damage response in co-cultured Sca-1+ HSPCs induced by D-gal pre-administration of feeder layer and the underlying mechanism may be related to ASP ameliorating feeder layer dysfunction due to D-gal induced senescence via inhibiting secretion of IL-1, IL-6, TNF-α, and RANTES, enhancing Cx43-mediated intercellular communication, improving Runx2 expression whereas decreasing PPARγ expression in BMSCs. The antioxidant property of ASP may provide a stroma-mediated potential therapeutic strategy for HSPC aging.


Introduction
Not all types of somatic cells end up by apoptosis after Pengwei Jing and Xiaoying Song contributed equally to this work.  1 senescent hematopoietic stem cells (HSCs) accumulate, a long-term damage to the hematopoietic system will occur. HSC aging is becoming a concern; however, besides intrinsic factors, extrinsic molecular mechanisms that hematopoietic microenvironment (HM) control on HSC aging are still poorly understood.
Recent studies have suggested that the majority of HSCs are perivascular and are enriched in the highly vascular endosteal region which contains a complex network of non-hematopoietic stromal cells including perivascular mesenchymal progenitor cells, osteolineage cells, endothelial cells, CXCL12-abundant reticular cells, sympathetic nerve cells, and non-myelinating Schwann cells [3][4][5][6]. Niche cells regulate HSCs via production of an array of cytokines, extracellular matrix proteins and adhesion molecules. However, HM regresses in the function along with natural aging [7][8][9][10], also stromal damage including SIPS occurs in the condition of oxidative stress such as radiotherapy and chemotherapy [11,12]. Oxidative stress has been considered as a major causal factor of SIPS. Excessive intracellular ROS due to increased ROS production and/or declined free radical scavenging may trigger p53 signaling and cyclin-dependent kinase inhibitor p16, p19, p21, then cause cellular DNA damage, of which the most severe one is DNA double-strand breaks (DSB). Among hematopoietic cells, the primitive cell population is the most vulnerable to withstand ROS attack. Normally, HSPCs are in a hypoxic state partly because the neighborhood supporting stromal cells serve to reduce their ROS level. Multifaceted factors in bone marrow stromal microenvironment take a vital role in alleviating the oxidative burden of HSPCs, such as connexin-43 (Cx43)-mediated gap junction intercellular communication (GJIC) [13][14][15][16][17] and cytokines represented by SCF and membrane-bound CXCL12 [18][19][20]. However, as senescence-associated secreted phenotype (SASP), inflammatory profile in aging HM including interleukins, RAN-TES and TNFα may cause hyperoxia of HSPCs [21].
HM itself also alters substantially with oxidative stress, therefore modification of BMSC function might be pivotal for preventing hematopoietic regression. D-galactose (D-gal) is considered to be an ideal model for simulating oxidative stress induced aging. An excess of D-gal under the action of galactose oxidase generates aldohexose and hydrogen peroxide and promotes the generation of oxygenderived free radicals and the superoxide anion, which cause cellular aging [22]. In the current study, continuous exposure of rats to D-gal or exposure of BMSCs to D-gal were used to mimic oxidative stress in bone marrow microenvironment, furthermore BMSCs and HSPCs ex vivo co-culture system was established to investigate the effect of aging bone marrow stroma on hematopoietic cells. Herein we found that the aging of BMSCs accompanied by its functional alteration might be one potential cause of oxidative stress induced DNA damage of Sca-1 + HSPCs, thus impaired their proliferation and differentiation ability, also led to SIPS of hematopoietic cells. Angelica sinensis is a famous Chinese herbal medicine. Polysaccharides are the main effective components of Angelica sinensis with antioxidant, anti-tumor, hematopoietic regulation, immune regulation, and radiation protection and other important biological activities [23][24][25]. Literature reported that ASP exerts hematopoietic activity which promoted human CD34 + HSPCs proliferation and differentiation [26]; ASP has antioxidant property protecting endothelial progenitor cell, hepatocytes, peritoneal macrophages and nerve cells from oxidative damage [27,28]. Our recent studies have also shown that ASP has antiradiation or anti-aging effect on HSPCs [29]. Herein, the current study illustrated that the anti-aging protective effect of ASP on BMSCs indirectly ameliorated oxidative milieu of Sca-1 + HSPCs and reversed hematopoietic regression. Considering the hematopoietic activity and dual antioxidant regulation of Angelica polysaccharide on both hematopoietic stem cells and stromal cells, the study might lead to a new strategy for screening of ray preventive and chemotherapeutic therapeutic preventive agents.

Animal studies
All experiments were performed in accordance with institutional and national guidelines and were approved by the Animal ethics and Committee of Chongqing Medical University. Three-month-old male Sprague-Dawley rats were purchased from Laboratory Animal Center of Chongqing Medical University. Forty rats were randomly divided into 4 groups: control group; D-gal administration group; ASP treatment group; D-gal + ASP administration group. In the D-gal administration group, D-gal (120 mg/kg·day) was injected subcutaneously daily for 42 days. In the ASP treatment group, saline at the same amount of D-gal was injected subcutaneously for 42 days, and ASP (200 mg/kg·day) was injected intraperitoneally daily concomitantly for 35 days since day 8 of saline injection. In the D-gal + ASP administration group, ASP (200 mg/kg·day) was injected intraperitoneally daily concomitantly for 35 days since day 8 of D-gal injection. All the control mice were given the same amount of saline subcutaneously or intraperitoneally.

Isolation of bone marrow stromal cells
Bone marrow cells (BMCs) were collected from the aspirates of the femurs and tibias using a 21-gauge needle and syringe. Isolated cells were resuspended in DMEM/F12 supplemented with 10% FBS, penicillin (100U/ml) and streptomycin (100 µg/ml), plated in 25cm2 flasks, and incubated in 37℃, 5% CO2-humidified chamber. After 24 h, the supernatant and non-adherent cells were discarded, and the adherent cells were fed with fresh medium. After the cells reached 80-90% confluence, BMSCs were trypsinized then sub cultured at 1:2 split. After 3 passages the detached BMSC were collected for further assays. To assess senescence of BMSCs, cell cycle analysis, SA-β-gal staining, and western blot for senescent signals were performed. Also, ELISA was used to evaluate cytokine secretion of BMSCs, balance between intracellular oxidation and anti-oxidation was also analyzed.

Hematopoietic cell colony forming unit assay and flow cytometric analysis
BMCs were resuspended in IMEM and centrifuged through Histopaque 1083 (Sigma) to isolate bone marrow mononuclear cells (BM-MNCs). Isolated BM-MNCs were cultured with MethoCultTM GF M3434 methylcellulose medium in 24-well plates. Colonies of CFU-GEMM were scored in triplicate wells per group on the 12th day of the incubation. For flow cytometric analysis, BM-MNCs were cultured in long-term culture medium. 6 h later, non-adherent hematopoietic cells were collected and stained with anti-p-p38, anti-p21, anti-p16 antibodies.

Sca-1 + HSPCs and BMSCs co-culture
BMSCs and hematopoietic cells were isolated as above from the femur and tibia of normal C57BL/6 mice. BMSC feeder layers were prepared as followed. BMSCs on the third passage were divided into 4 groups (control, D-gal, ASP, and D-gal + ASP). The control group was cultured as routine; the D-gal feeder layer was administrated with 30 g/L D-gal for 96 h; the ASP feeder layer was administrated with 0.1 g/L ASP for 96 h; the D-gal + ASP feeder layer was pre-administrated with 30 g/L D-gal for 48 h, then concomitantly treated with 0.1 g/L ASP for 48 h. Four groups of BMSCs were utilized for further analysis. Bone marrow-derived Sca-1 + HSPCs were isolated as previously described [30] using positive selection of magnetic activated cell sorting (MACS) technique with EasySepTM Mouse SCA1 Positive Selection Kit. The positive rate is (83.32 ± 2.57) %. In the co-culture system, Sca-1 + HSPCs at 1 × 105 ~ 1 × 106 cells per ml were plated directly on prepared BMSC layers. After 48 h, Sca-1 + HSPC were used for subsequent experimental measurement.

Cell cycle analysis
1 × 106 cells in each group were collected and analyzed according to our previously reported procedures [31].

Senescence-associated β-galactosidase (SA-β-gal) staining
SA-β-gal staining kit was used according to the manufacturer's instructions and our previously reported procedures [29]. A minimum of 1000 cells in 10 random fields were counted.
Data were normalized to an endogenous reference gene GAPDH expression. The PCR primers used are provided in the table as follow:

easurement of the level of 8-OH-dG and γH2AX staining in Sca-1 + HSPC
Sca-1 + HSPC co-cultured with BMSC layer of each group. The levels of 8-OH-dG in different groups were detected by an ELISA kit following the manufacturer's instruction. The levels of γH2AX were determined by flow cytometric analysis.

Statistical Analysis
Statistical analyses were carried out using SPSS 19.0 software. Data are presented as the mean ± SD. Two-way ANOVA was used for comparison of mean values across the groups and multiple comparisons were made by LSD test. Differences were considered significant at P<0.05.

Determination of cytokines by ELISA
BMSCs were collected. Hematopoietic cytokines (SCF and GM-CSF) and inflammatory cytokines (IL-1, IL-6, TNFα, and RANTES) in each group were measured by ELISA according to the kit manufacturer's instructions.

Western blot analysis
BMSCs and hematopoietic cells were collected respectively, and fifty microgram of protein samples were loaded for SDS-PAGE and transferred onto 0.2 μm PVDF membrane, which were incubated with the primary antibodies against p53, p21, p16, p-p38. β-actin was used as the internal control protein. After appropriate HRP-conjugated secondary antibody incubation in TBST, protein bands were detected using an ECL detection system. The semi-quantificaton analysis was performed using Quantify One software (BioRad, Hercules, CA, USA).

Detection of oxidative indicators
BMSCs or Sca-1 + HSPCs were loaded with 5µM of 2',7'-dichlorodihydrofluorescein diacetate (DCF-DA). After 30 min incubation at 37℃, the ROS level was determined and indicated as the mean fluorescence intensity (MFI) of DCF. SOD activity and MDA content in the supernatant of BMSCs were measured using ELISA assay kits.

Analysis of cellular gap junction Cx43 in BMSCs
Cx43 protein expression and its mediated intercellular communication function in BMSCs prepared for co-culture system were detected by western blot and SLDT (scrapeloading and dye transfer) assay respectively. Lucifer Yellow as a small molecule which can pass through intercellular gap junction was used in SLDT according to previous protocol [31]. The mean cumulative fluorescence intensity values were analyzed using Ipp6.0 to evaluate the function of GJIC.

Real-time reverse transcription-PCR (RT-PCR)
Total RNA was isolated from cells using TRIzol reagent (Invitrogen, USA) following the manufacturer's protocol. First-strand cDNA was synthesized from 2 µg of total RNA using Taqman RT reagents (Applied Biosystems, USA). The cDNAs were determined by quantitative real-time PCR using SYBR Green Supermix (BioRad). All samples were analyzed in triplicate using iCycler Real-Time Detection System (cfx96, BioRad). The changes in mRNA expression were calculated by the comparative cycle threshold method.
The reservoir and vitality of early hematopoietic progenitor cells are evaluated by CFU-GEMM frequency. As shown in Fig. 1a, compared with that of the control group, the CFU-GEMM frequency decreased in the D-gal administration group, however, ASP treatment alone notably Representative flow cytometric images of p-p38, p21, and p16 in hematopoietic cells are presented and the results are presented as means ± SD (n = 6). * p < 0.01 senescence via targeting downstream p16-Rb and its interacting p53-p21 pathway. The results showed that ASP treatment attenuated D-gal-induced increases of p-p38, p16 and p21 protein expression (Fig. 1c). Those evidences above suggested that D-gal induced aging probably resulted in HPC proliferation and differentiation impairment, whereas ASP prevented cellular aging and restored HPC function.

ASP attenuated oxidative stress in bone marrow hematopoietic microenvironment in D-gal preadministrated rat
In the D-gal administrated rats, ROS level of BMSCs rose up to 150% of that in the control rats, SOD activity markedly decreased whereas MDA level elevated significantly compared with the control group. However, in the D-gal + ASP group, ASP evidently restored SOD activity and retracted MDA level ( Table 1). As expected, it was also found that compared with the control rats, D-gal significantly enhanced p53, p21 and p16 protein expression, thus held BMSC growth arrest at G1 phase and increased cellular aging population. However, ASP inhibited D-gal-induced stress signals thus rejuvenated cell cycle progression and decreased senescent BMSC frequency (Fig. 2a ~ 2c). Collectively, these results focused the highlights that D-gal simultaneously caused oxidative damage to the stromal cells in which aging hematopoietic cells located; on the opposite ASP also exhibited anti-oxidant property to protect stromal cells from SIPS.

ASP ameliorated hematopoietic microenvironment in D-gal pre-administrated rat
Cx43 and CXCL12 histologically concentrate at the endosteal and perivascular regions. Expression of Cx43 and CXCL12 in the D-gal administration rats was significantly lower than that in the control group. ASP treatment increased Cx43 and CXCL12 expression compared to the control, moreover ASP partially rescued D-gal induced decreased expression ( Fig. 3a and b). ELISA results showed that ASP treatment reversed the D-gal decreased SCF and GM-CSF levels in the BMSCs (Fig. 3c). It was speculated that in oxidative stress setting, aging HM altered signals between niche cells and hematopoietic cells which were unfavorable to scavenge intracellular ROS in hematopoietic cells and maintain progenitor cells, however, ASP could ameliorate the function of niche cells to sustain and support HPCs.
increased the CFU-GEMM frequency. Also, ASP treatment reversed D-gal-induced CFU-GEMM loss. To explore the mechanism of HPC population loss, Senescence-associated SA-β-gal staining was performed to evaluate hematopoietic cells. The ratio of positive cells in D-gal-administrated rats markedly rose compared with the control, whereas after ASP treatment, the percentage of aging cells was reduced (Fig. 1b). Activated phosphorylated p38 mediates HSPC  in HM may elicit oxidative load to HSPCs and the antiaging effect of ASP on hematopoietic cells may partly relate to alleviating oxidative stress in HM, we established BMSCs and Sca-1 + HSPCs co-cultivation system in vitro and observed the direct effects of different feeder layers on co-cultured Sca-1 + HSPCs. It was found that ASP treatment to D-gal pre-administrated BMSC layer reversed aberrant intracellular ROS accumulation in Sca-1 + HSPCs, thus prevented DNA damage in Sca-1 + HSPCs and weakened those senescence-associated signals in hematopoietic cells (Fig. 4a and c). Taken together, it hinted that oxidative stress from aging HM could lead to oxidative load and subsequent DNA damage on HSPCs. Activated DDR signaling pathways may result in HSPC senescence, which may be one of the mechanisms of hematopoietic dysfunction. Whereas ASP may protect HSPCs from senescence via alleviating oxidative stress in HM.

ASP ameliorated dysfunction of BMSC feeder layer induced by D-gal
The multiple underlying mechanisms how BMSC feeder layers may influence co-cultured Sca-1 + HSPCs were further explored. It was found that ASP may inhibit BMSC feeder layer pro-inflammatory cytokine production including IL-1β, IL-6, TNF-α, and RANTES (Fig. 5a). Also, ASP treatment significantly rescued a decreased Cx43 protein expression and retrieved D-gal-weakened intercellular transfer ability by BMSC feeder layers (Fig. 5b and c). qRT-PCR results showed ASP may promote differentiation potential shift from adipogenisis to osteogenisis under the condition of oxidative stress as ASP increased Runx2 expression and decreases PPARγ expression in the D-gal-pretreated BMSC layer (Fig. 5d). Collectively, ASP may alter hematopoietic stromal microenvironment via multiple mechanisms, subsequently support HSPCs.

Discussion
Increasing evidences have revealed that microenvironment plays a critical role in protecting HSPCs from cytotoxic agents and metabolic by-products, and in maintaining dormancy. Perturbation of niche or stromal cells may cause malfunctions in HSPCs, hence lead to hematological diseases. Most recently, it was found that hematopoietic reconstitution was compromised in aging BM due to ROS invasion of stromal cells, which, however, was partly reversed by a polyphenolic antioxidant curcumin [33]. In the current study, we have shown that ASP, polysaccharides extractive taken from traditional Chinese medicine, attenuated ROS

ASP-treated BMSC feeder layer protected cocultured Sca-1 + HSPCs from oxidative DNA damage
It is known that the more primitive hematopoietic precursors are, the more vulnerable to ROS attack. For DNA-damage response (DDR), γH2AX foci is a classical indicator of double-stranded DNA cleavage [32] and 8-hydroxy-2′deoxyguanosine (8-OH-dG) is an oxidative adduct. DDR machinery may induce senescence via typical DDR pathways including ATM/Chk2 sensitized p53-p21-p16 and p38-p53/p16 signaling. To determine the oxidative stress that in vivo, ASP administration promoted CFU-GEMM formation and protected BM hematopoietic cells from D-gal induced senescence via down-regulation of stress-activated signals including p38 MAPK, p21 and p16 proteins. Meanwhile, ASP was found to exert the role of ameliorating the aging niche: rejuvenated BMSCs; prevented BMSCs from ROS accumulation and p53, p21, and p16 activation. Also, ASP functioned as a potent quencher of cellular ROS in BM niche cells as improving ROS scavenger SOD level whereas reducing lipid peroxidation. However, it is not known that whether aging BM niche cells induce agingrelated changes in HSPCs. We hypothesized that under the load in stromal cells, which consequently prevented HSPCs from SIPS due to oxidative DNA damage.
There has been growing proof that increased levels of ROS with age renders oxidative stress and lead to dysregulation of tissues, especially BM. Interestingly, steadily intracellular ROS accumulation is significantly more evident in the stromal compartment. The reason that ROS level varies between HSCs and stromal cells is that HSCs are metabolically less active with various of robust ROS-detoxifying systems, whereas high proliferative stromal cells consume more O2 hence increase ROS content and are self-exposed to oxidative stress [34]. Herein, the current study showed BMSCs share the hypoxic niche with the hematopoietic progenitor cells that they support. Stromal cells participate in the regulation of intracellular ROS of primitive hematopoietic cells, to adjust and balance ROS content in primitive cells via intercellular communication, adhesion, and cytokine milieu. Cx43, postulated to be a self-renewal gene, maintains hematopoietic precursors in hematopoiesis. In stress circumstance, it can protect HSPCs via transferring the intracellular ROS of HSPCs to HM. CXCL12 facilitates HSPC maintenance and low ROS level in HM elicits CXCL12 presentation on the membrane of BMSCs [18]. Hematopoietic growth factors SCF can also regulate intracellular ROS level in HSPCs and GM-CSF is vital condition of oxidative stress, ASP rejuvenated BMSCs to provide enough support to HSPCs thus prevented HSPCs from aging. Therefore, BMSCs and hematopoietic cells co-culture system was established. Sca-1 + HSPCs were co-cultivated with D-gal-treated or -untreated stromal cells for 48 h. D-gal-induced aging feeder layer provoked DDR machinery in Sca-1 + cells; whereas ASP-treated feeder layer attenuates ROS-mediated cellular DNA damage in co-cultivated Sca-1 + HSPCs. How the interplay with the stromal niche controls HSPC function including the aging process remains to be elucidated. Hence, we further focused on the differences between the aging niche and ROS load-attenuated niche. (b) Connexin-43 protein expression in stromal feeder layer was detected via Western blot assay and the results were normalized to an internal control β-actin. The relative protein expression of connexin-43 is presented as means ± SD (n = 3). (c) The function of gap junction channels was analyzed by SLDT assay using Lucifer Yellow, a low-molecularweight water-soluble fluorescent dye which can pass through gap junctions. Lucifer Yellow transmission across BMSCs of the feeder layer was shown with inverted fluorescence microscopy (Scale bar = 50 μm). The capacity of intercellular transfer ability was represented by a dye-coupling mean cumulative fluorescence intensity value and presented as means ± SD in triplicate independent experiments. (d) A representative analysis of PPARγ and Runx2 mRNA expression detected by qRT-PCR is presented as means ± SD of triplicate. # p < 0.05 down-regulation of lymphoid-specification genes including Ikaros and Gata3 [32,43]. Functional redundancy among stromal cells to maintain hypoxia in HSPCs remains not clear, anyway, the excessive oxidative load in the niche provoked oxidative stress in the co-cultivated HSPCs.
It is encouraging that with hematopoietic stimulating property and ROS scavenging property, ASP exerted protective effects on HSPCs from DDR induced premature senescence. The underlying mechanism is that ASP reduced ROS load in stromal cells, minimized stress-responsive signaling, attenuated ROS-mediated cellular aging; ameliorate hematopoietic milieu thus rescued HSPCs from aging and dysfunction due to a damaged bone marrow niche. Generally, as major constituents of root extraction of Chinese Angelica Sinensis, ASP underlines the niche-mediated hematopoietic protective function to some extent, and it might lead to new strategies for screening of hematopoietic protective agents.

Conclusion
The results suggest that ASP may reduce ROS load in stromal cells, alleviate oxidative stress, attenuate stress induced premature aging; subsequently facilitate stromal function including keeping osteogenesis/acidogenesis balance, boosting intercellular communication between stromal cells and hematopoietic cells, and altering cytokine milieu, thus protect HSPCs from DDR induced aging. ASP is a promising medication to protect HSPCs from aging via alleviating oxidative stress and ameliorating the function of hematopoietic microenvironment.
for CFU-GEMM formation. However, the BMSCs and niche themselves might be altered by ROS accumulation in hematopoietic microenvironment. Oxidative stress will drive stromal progenitor cells differentiation, and different levels of ROS dictate osteogenisis versus adipogenisis differentiation. As known, osteoblastic differentiation facilitates hematopoiesis. Increases in adipocytes in BM related to repression of growth and differentiation of HSPCs partly due to the reduced production of GM-CSF and G-CSF. Mice which were genetically incapable of adipocyte formation demonstrated an increase in HSC engraftment after irradiation [35]. Adipogenisis differentiation is regulated by transcription factors such as CREB and C/EBPb, which act together with PPARγ, a downstream signal of ROS. Additionally, hemeoxygenase-1 (HO-1) activity involves reduction of ROS levels in stromal progenitor cells by promoting osteoblast differentiation [36,37]. In the current study, after ectogenic oxidant D-gal treatment, high level of ROS caused senescence of stromal cells. Also, the expression of the osteolineage related transcription factor Runx2 in stromal cells decreased, however, the adipogenic lineage related transcription factor PPARγ was up-regulated, suggesting gradual potential shift from osteogenesis to adipogenesis differentiation in stromal progenitor cells. The result was in accordance with the documents that aging BM is accompanied with a prominent increase in adipocytes, and aging alters cellular composition of stroma with a decrease in the frequency of stromal cells, osteoblasts, endothelial cells, MSCs, and CAR + cells [38,39].
Moreover, it was reported that the niche microenvironment was altered under the condition of oxidative stress. First, ROS may damage perivascular niche and osteoblastic niche from which most SCF and CXCL12 originate, therefore the fundamental hematopoietic factors are reduced [38,40]. In this study, after D-gal treatment, high level of ROS inhibited SCF secretion and CXCL12 presentation on the stromal cell membrane. It was speculated that decreased SCF and CXCL12 may subsequently translocate HSCs from the BM endosteal area to areas around the sinuses which promote ROS accumulation in HSCs [41,42]. Second, D-gal induced oxidative load altered intercellular gap junction Cx43, which is a major ROS scavenger transferring from HSCs to stromal cells during genotoxic stress, and also declined the ability of GJIC. Third, during injury process of aging, BMSCs produced various pro-inflammatory cytokines including IL-1, IL-6 and TNF-α. The over-production of pro-inflammatory cytokines is one mechanism of activation of the oxidative DNA damage checkpoint in HSPCs [21]. Among the pro-inflammatory cytokines, RANTES promotes myeloid-biased lineage differentiation via upregulation of pro-myeloid transcription factors like Gata2 and inhibits early lymphopoiesis and T-cell development via