In this study, we intended to precisely determine the optimal pO2 in a cell culture system that causes ASCs to release the secretome with the highest reparative and regenerative capacity in the liver. Hepatocytes cultured under 1% pO2 showed the highest expression of proliferation-associated markers such as IL-6, HGF, and VEGF. The AML12 cells cultured with the secretome of 1% pO2 showed the highest cell proliferation, followed by the cells cultured with the secretome of 21%, 10%, and 5% pO2, in that order. When the secretomes of various culture pO2 were injected into partially hepatectomized mice, the 1% pO2 secretome most significantly increased liver regeneration and reduced serum levels of proinflammatory mediators (IL-6 and TNF-α) and liver transaminases. In addition, analysis of the mouse liver specimens indicated that injection with 1% pO2 secretome maximized expression of the essential intermediates in the PIP3/Akt and IL-6/STAT3 signaling pathways, all of which are known to promote liver regeneration. Taken together, our experiment showed that, of the various culture pO2 (1%, 5%, 10%, and 21%), 1% pO2 was most advantageous for obtaining the secretome that maximizes liver regenerative and reparative potential.

Several lines of evidence indicated that stem cells proliferate better under hypoxic condition than under normoxia [2427]. For instance, Grayson et al. [25] showed that human MSCs under 2% pO2 exhibited 30-fold increase of proliferation for seven passages compared to MSCs under normoxia. Lennon et al. [26] also showed that culture of rat MSCs under 5% pO2 resulted in approximately 40% higher cell number at first passage than culture under normoxia. In addition, reduced pO2 was found to decrease the population doubling time of marrow-isolated adult multilineage inducible cells, with 3% pO2 showing the maximum effect [24]. Although the underlying mechanism is still unknown, it could be greatly attributed to the fact that the actual in-vivo pO2 is considerably low, ranging from 1% (deep zone of cartilage) to 12% (blood) [28]. Similar to other internal organs, the liver maintains low oxygen tension, ranging from 3% (perivenous area) to 7% (periportal area) [29]. Our study suggests that, of the various culture pO2 (1%, 5%, 10%, and 21%), 1% pO2 is most advantageous to stimulate ASC proliferation.

Until now, several explanations have been provided to clarify the reason why HP is beneficial to cell growth. Some authors insisted that HP extends the survival of cultured cells through a HIF-1α-mediated mechanism. They found that hypoxia activates a complex array of signaling pathways favoring the stabilization of HIF-1α, which otherwise would be degraded [13]. Stabilized HIF-1α binds and activates promoter regions of hypoxia-responsive genes, whose expression contributes to survivability of MSCs by promoting the expression of glucose-6-phosphate transporter. The target genes of HIF-1α also include those encoding pro-angiogenetic factors, such as IL-6 and VEGF. Moreover, hypoxia-induced activation of Akt/p38MAPK also leads to upregulation of anti-apoptotic proteins such as Bcl-2 and Bcl-xL [30]. The upregulated anti-apoptotic proteins in turn reduce apoptosis by decreasing the expression of pro-apoptotic proteins such as Bax and Caspase-3.

The fundamental concept of our study is that HP has the potential of upregulating secretome release from MSCs [3032]. The secretome elicited by HP is known to have immunomodulatory properties such as inhibition of the activities of peripheral blood mononuclear cells and promotion of inflammation via cytokines [30]. Ohnishi et al. [33] reported that culturing rat MSCs with 1% pO2 for 24 h increased the expression of a number of genes related to cell proliferation and survival, including VEGF-D, heparin-binding epidermal growth factor, matrix metalloproteinase-9, and placental growth factor. In another study, culturing human MSCs with 1% pO2 for 48 h increased the secretion of VEGF, decreased the levels of transforming growth factor-β1 and basic fibroblast growth factor (but insignificantly), and did not change the levels of IL-6, IL-8, and tissue inhibitor of metalloproteinases-1 [31]. Thus, it appears that HP can selectively increase, decrease, or have no impact on certain components of the secretome. In addition, HP can increase migration and homing abilities of MSCs by upregulating the expression of CX3XR1 and CXCR4, as well as cMet (receptor for hepatocyte growth factor) [3436].

In our study, SIRT1 expression was highest in the mouse liver in which the secretome of the lowest (1%) culture pO2 was infused. SIRT1 deacetylates a number of essential transcriptional regulatory proteins, and thereby regulates a variety of physiological processes, including stress responses, metabolism, apoptosis, calorie restriction, and aging [37, 38]. In the liver, SIRT1 functions as an essential regulator of metabolic processes, such as gluconeogenesis, fatty acid beta-oxidation, mitochondrial activity, and cholesterol flux, all of which occur in response to an intracellular rise in the NAD+/NADH ratio when energy supplies are low [39]. Accumulating evidence indicates that SIRT1 is considerably involved in the liver regeneration [4043]. Bellet et al. [43] showed that the higher expression of SIRT1 directly facilitates liver regeneration by promoting G1/S transition and fatty acid beta-oxidation which is essential for liver regeneration.

Liver regeneration is essentially achieved by the combination of two distinct factors: cell proliferation (increase in the number of hepatocytes) and hypertrophy (increase in the size of hepatocytes). It has been identified that cell proliferation is mediated by the IL-6/STAT3 pathway, and hypertrophy is mediated by the PI3-K/PDK/Akt pathway [

44

]. We previously showed that the secretome obtained from HP promotes liver regeneration by persistent and uninhibited expression of STAT3 in the liver which is caused by decreased expression of SOCS3 [

10

]. In this study, apart from the HP effects on the IL-6/STAT3 pathway, we investigated the HP effects on the PI3-K/PDK/Akt pathway. The group of 1% pO

2

secretome showed the highest expression of SIRT1 in the liver specimens. This could be explained by a previous study which demonstrated that acute hypoxia increases the SIRT1 expression in a HIF-dependent manner [

45

]. The group of 1% pO

2

secretome also showed the highest expression of Akt in the liver specimens. Recent advancements in cell biology have also identified sirtuins as major regulators of Akt activation [

46

]. Of the sirtuins, SIRT1 was discovered to directly deacetylate Akt, thereby enabling its binding to phosphatidylinositol (3,4,5)-triphosphate (PI3P). The binding between Akt and PI3P results in a conformational change that exposes the kinase domain of Akt for phosphorylation and activation by 3-phosphoinositide-dependent protein kinase 1 (PDK1). Taken altogether, we think that the 1% pO

2

secretome induces the highest liver regenerative potential through the activation of both the IL-6/STAT3 and the PI3K/PDK/Akt pathways (Fig. 

6

).

Fig. 6

Proposed mechanism of hypoxic-conditioned secretome effects on liver regeneration. Liver regeneration is essentially achieved by the combination of two distinct factors: cell proliferation (increase in the number of hepatocytes) and hypertrophy (increase in the size of hepatocytes). Cell proliferation is mediated by the IL-6/STAT3 pathway, and hypertrophy is mediated by the PI3K/PDK/Akt pathway. We provide two independent mechanisms by which hypoxic-conditioned secretome improves liver regeneration. First, hypoxic-preconditioned secretome promotes liver cell proliferation by persistent and uninhibited expression of STAT3 in the liver which is caused by decreased expression of SOCS3. Second, hypoxic-preconditioned secretome promotes liver cell hypertrophy by upregulating Akt expression which is activated by higher expression of SIRT1. IL interleukin, PDK1 3-phosphoinositide-dependent protein kinase 1, PI3K phosphoinositide 3-kinase, STAT3 signal transducers and activators of transcription 3, SIRT1 sirtuin 1, SOCS3 suppressor of cytokine signaling 3

In our study, hepatocytes responded more effectively with hypoxic secretome than the kidney cells, thereby resulting in higher expression of the proliferation maker (PCNA) and the intermediates. These differences may be due to the specific characteristics of hepatocytes. Unlike other cells, hepatocytes are capable of responding positively to damage through matrix turnover or remodeling, rather than simply being the targets of insults or bystanders in the process of injury. The ability to cope with such insults is evidenced by improved regenerative potential of hepatocytes. In transplantation experiments using fumarylacetoacetate hydrolase (FAH)-deficient mice, hepatocytes have proved to have the capability of regenerating more than 70 times [47]. Thus, although hepatocytes appear to be quiescent in normal liver, they exhibit enormous proliferation potential when they are stimulated. We believe that such characterization of hepatocytes is the reason why hypoxic secretome treatment was more effective in hepatocytes than in kidney cells.