**3. Discussion**

HSPCs with self-renewal and multipotent capacity offer a valuable source for cell-based therapy and regenerative medicine [36]. Many limitations facing during HSPC transplantations would have been overcome if ex vivo HSPC expansion and maintenance become possible. However, the characteristics of the HSPCs are often altered once they leave the hypoxic bone marrow niche, affecting the quality and quantity of the cultured HSPCs. Studies have shown that the fate of HSPCs is regulated by the microenvironment of the so-called "stem cell niches" that contain oxygen saturations of approximately 5% [7]. The increased oxygen tension in normoxic cultures causes the stem cells to lose their stemness [37]. Various attempts have been carried out to maintain the stemness of HSPCs and to overcome the limitations associated with ex vivo culturing. These include the utility of transcription factors, co-culturing with feeder cells [38], the addition of cytokine cocktails [39], and genetic modification [40]. However, these approaches have some drawbacks, especially when used in a clinical setting. A few previous studies have suggested that the use of antioxidants can ameliorate oxidative stress-mediated damage in cultured HSPCs [41]. In this study, we investigated whether PB-CD34<sup>+</sup> cells can be efficiently expanded ex vivo by utilizing Ech A, which has been demonstrated as an antioxidant in previous studies [32,34].

HSPCs remain quiescent in the osteoblastic niche—the lowest end of the oxygen gradient in the bone marrow. However, in the oxygen-rich vascular niche, stem cells can proliferate and differentiate closer to blood circulation, resulting in increased intracellular ROS levels. However, an extremely low or high level of ROS would cause impaired repopulation capacity or trigger exhaustion of HSPCs [9]. Physiologically, with the aid of intracellular antioxidant enzyme systems and endogenous antioxidants, HSPCs are able to cope with the damage caused by cumulative ROS [42]. However, the damage is overwhelmed when HSPCs are subjected to ex vivo expansion, wherein a sharp increase in ROS levels is often experienced. It is, therefore, critical to regulate the intracellular ROS levels during ex vivo culturing for the better expansion and maintenance of HSPCs. It has been previously demonstrated that a reduction in intracellular ROS levels by supplementing antioxidants or lowering oxygen tension in cell cultures could improve HSPC expansion, and their engraftment and hematopoietic reconstitution

abilities in non-obese diabetic/severe combined immunodeficiency (NOD/SCID) mice [43]. Consistent with this finding, PB-CD34<sup>+</sup> cells provoked aberrant ROS generation in the normoxic culture (Figure 1A). PB-CD34<sup>+</sup> cells exhibited higher ROS levels than BM-CD34<sup>+</sup> cells (Figure 1A). ROS is a critical mediator of HSPC quiescence with p38-MAPK. Higher ROS levels can activate the p38-MAPK pathway, which in turn can promote phosphorylation of p38-MAPK [44]. Our results also showed that PB-CD34<sup>+</sup> cells can upregulate phospho-p38-MAPK and phospho-JNK, likely due to high ROS levels (Figure 2A). p38-MAPK plays a role in hematopoiesis regulation, particularly in erythropoiesis and granule formation [45]. Recently, p38-MAPK was identified as an intrinsic modulator that can negatively regulate HSPC self-renewal [46]. Therefore, both ROS and p38-MAPK have been shown to play an important role in maintaining HSPC quiescence. Several studies have indicated that HSPCs lose their ability to regenerate due to elevated ROS levels and specific phosphorylation of p38-MAPK. Pharmacological inhibition of ROS or p38-MAPK activity can restore HSPC function in the ROShigh mouse population and rescue these mice from bone marrow damage [7]. Treatment of Ech A and NAC potently suppressed the activation of p38-MAPK and JNK, and the p38-MAPK inhibitor SB203580 and the JNK inhibitor SP600125 promoted CD34<sup>+</sup> cell expansion, potentially by inhibiting differentiation (Figure 2). It is noteworthy that the suppression of ROS by Ech A ameliorated the activation of p38-MAPK and JNK, suggesting that Ech A acts as an efficient antioxidant in the ex vivo culture of PB-CD34<sup>+</sup> cells, and subsequently promotes CD34<sup>+</sup> cell expansion.

SFKs in hematopoietic tissues can function as primary regulatory factors, as described in the first *p60-Src* gene perturbation experiment to confirm its role in osteopetrosis development [47]. Subsequent studies revealed SFK activities in B cells, bone marrow, obese cell lines, and Lyn-expressing HSPCs in all blood cell lines except T cells [19]. In some studies, Lyn was shown to play negative roles in monocyte production and plasma cell function, as revealed in *Lyn*-/- mice by M-phi tumorigenesis [48] and IgM hyperglobulinemia [20]. Although not widely studied, SFKs have also been suggested as important regulators of erythropoiesis. Avian Src was originally discovered as an oncogene that promotes sarcoma and erythroleukemia [49]. Interestingly, *Lyn*–/– mice among the SFK gene-deficient mice showed an age-dependent increase in the production of splenic erythroblasts [50]. In BM-derived cultures, early-stage *Lyn*–/– erythroblasts exhibited a reduced ability to expand and develop beyond the Kit+CD71<sup>+</sup> stage [51]. Therefore, SFKs, such as Lyn and Src, are major signaling mediators that modulate diverse stimuli to regulate differentiation, migration, proliferation, apoptosis, and metabolism. However, there is limited information regarding the precise role Lyn/Src and their underlying molecular mechanisms in the ex vivo expansion of PB-CD34<sup>+</sup> cells. PI3K/Akt signaling can be activated by the downregulation of PTEN by BCR–ABL2. PTEN is a lipid phosphatase that interferes with PI3K signaling by dephosphorylating phosphatidylinositol-3,4,5-trisphosphate. Class I PI3Ks consist of four different catalytic isoforms (p110α, p110β, p110γ, and p110δ) and two standard regulatory subunits (p85 and p101) [52]. PI3K plays an important role in HSPC maintenance and regulation of lineage development [53]. In this study, we found, for the first time to our knowledge, that Ech A activates Lyn, which in turn upregulates p110δ expression, and suppresses ROS production and p38-MAPK/JNK phosphorylation, resulting in enhanced ex vivo expansion of PB-CD34<sup>+</sup> cells (Figures 4 and 5). Furthermore, the addition of Ech A increased CFU/BFU-E producing cell numbers (Figure 6), which suggests that an appropriate use of Ech A is advantageous for CD34<sup>+</sup> cells to maintain self-renewal potential during ex vivo expansion.

Our findings, as summarized in Figure 7, demonstrate that Ech A can effectively inhibit ROS production in PB-CD34<sup>+</sup> cells. ROS-mediated p38-MAPK/JNK activation can reduce the number of CD34<sup>+</sup> cells, and decrease the self-renewal of PB-CD34<sup>+</sup> cells, which was reversed by Ech A treatment. Our results also demonstrate a novel Lyn-mediated p110δ expression by Ech A in PB-CD34<sup>+</sup> cells, although the precise molecular mechanism of Ech A-induced ex vivo expansion, especially how Lyn and p110δ are coordinated in PB-CD34<sup>+</sup> cells in response to Ech A, remains unclear. Taken together, Ech A was found to be an effective agent for promoting cell proliferation and maintaining the stemness

of HSPCs. Ech A is beneficial for CD34<sup>+</sup> cells to maintain their self-renewal potential and function during the ex vivo expansion, and possibly during in vivo expansion of HSPCs.

**Figure 7.** Schematic representation of Ech A and NAC in the ex vivo expansion of PB-CD34<sup>+</sup> cells. Mobilized PB-CD34<sup>+</sup> cells showed higher ROS levels than BM-CD34<sup>+</sup> cells. Ech A suppressed intracellular ROS production in PB-CD34<sup>+</sup> cells. Ech A also increased Lyn/Src phosphorylation and PI3K/Akt activation. Our results suggest that Ech A promotes ex vivo expansion of CD34<sup>+</sup> cells through Lyn/Src-mediated p110δ expression.
