Abstract
Myelotoxic injury unlocks the vigorous power of hematopoietic stem cells (HSCs) to replenish the hematopoietic system, making quiescent HSCs enter the cell cycle. Microscopically, it is well known that adipose tissue replaces cellular components in bone marrow (BM) after myeloablation by chemotherapeutic agents or irradiation. In a steady-state hematopoiesis, both HSC-intrinsic and -extrinsic mechanisms enforce quiescence of HSCs, and the interaction between HSCs and BM microenvironment has been drawing much attention in maintenance of the quiescence. In this meaning, it is supposed that the drastic change in BM microenvironment by myeloablation might trigger and promote the cell cycle entry of HSCs. We have previously reported that adiponectin, adipocyte-derived anti-diabetic hormone, indirectly enhances proliferation of murine immature myeloid progenitors upon granulocyte-colony stimulating factor treatment (emergency granulopoiesis) in Socs3-Stat3 dependent fashion by suppressing TNF-α production from macrophages (ASH meeting 2013, Abstract 221), however, its direct effect against HSCs in vivo is needed to be elucidated. Additionally, we have shown that both genetic loss and high-fat diet-induced reduction of adiponectin have no impact on steady-state hematopoiesis. Considering BM adipose tissue is an endocrine organ and adipocytes are the major cellular component in ablated marrow, we hypothesized that adiponectin derived from adipocytes might be implicated in HSC activation and subsequent hematopoietic recovery.
Adiponectin-null (adipo-/-) mice showed significantly delayed hematopoietic recovery after 5-fluorouracil (5-FU) administration. In 5-FU-treated BM, adipo-/- SLAM-HSCs (CD150+ CD48- Lin- Sca-1+) and CD34- SLAM-HSCs were more quiescent than adipo+/+ counterparts. Adipo-/- mice survived longer than adipo+/+ control mice after serial 5-FU treatment. Taken into account our previous data showing that impaired emergency granulopoiesis of adipo-/- mice is Socs3-Stat3 dependent and Socs3 haploinsufficiency ameliorated the defect, we further investigated whether activation of adipo-/- HSCs on 5-FU treatment was potentiated by genetic loss of Socs3. But Socs3 haploinsufficiency had no capacity to revert impaired activation of adipo-/- HSCs, suggesting some mechanisms other than that of impaired emergency granulopoiesis in adipo-/- mice.
Strikingly, adipo-/- HSCs were shown to be defective in mTORC1 activation, phosphorylation of S6 and mitochondrial activity after 5-FU treatment. In vivo rapamycin treatment cancelled the effect of adiponectin upon HSC activation by 5-FU, suggesting that adiponectin enhances HSC activation through mTORC1-dependent mechanism. Physiological isoform of adiponectin (full-length adiponectin) enhanced not only 5-FU-induced mTORC1 activation in vivo but also cytokine-induced activation in vitro, shortened the time to first division, without affecting subsequent proliferation of HSCs, in contrast to the previous report using non-physiological isoform (globular adiponectin) in vitro.
The concentration of adiponectin in BM had a 4-fold increase after 5-FU treatment while the level in plasma remained unchanged. In a steady state, adiponectin level in adipocyte-rich tibia is higher than femur, suggesting local production of adiponectin constitutes a significant portion of BM adiponectin. Every BM cell components examined expressed adiponectin mRNA, and adipocytes had the highest. As 5-FU treatment had little effect on adiponectin expression in adipocytes, it was suggested that increased adipocytes in BM contributed to increased adiponectin upon myelotoxic injury. Furthermore, reciprocal transplants with adipo+/+ and adipo-/- mice demonstrated that adiponectin from BM environment of recipient mice plays a major role in the activation of HSC after in vivo 5-FU treatment and in vitro cytokine stimulation.
These data reveal that adiponectin, produced mainly from increased adipocytes after myelotoxic injury, positively regulates HSC activation and subsequent hematopoietic recovery. Our data also highlight adipocytes as an essential source of adiponectin to ensure the proliferative burst of hematopoietic cells in myeloablated marrow. Adiponectin treatment could be clinically applied to relieve myelosuppression by chemotherapy.
No relevant conflicts of interest to declare.
Author notes
Asterisk with author names denotes non-ASH members.
This feature is available to Subscribers Only
Sign In or Create an Account Close Modal