S894 DEFECTIVE GLYCOLYSIS IN BONE MARROW ENDOTHELIAL CELLS OF PATIENTS WITH POOR GRAFT FUNCTION POST‐ALLOTRANSPLANT IDENTIFIES A POTENTIAL THERAPEUTIC STRATEGY

Abstract

Background:Poor graft function (PGF), defined as pancytopenia, is a serious complication after allogeneic hematopoietic stem cell transplantation (allo‐HSCT). Its pathobiology, found by our previous studies (BBMT 2013; BMT 2016; Blood 2016), involves the dysfunctional and reduced bone marrow(BM) endothelial cells(ECs), which is a key component of BM microenvironment to modulate the physiology and regeneration of hematopoietic stem cells (HSCs). Furthermore, the aberrant BM ECs derived from PGF patients could be attenuated by ROS scavenger, N‐acetyl‐L‐cysteine (NAC), in vitro (Blood 2016), implying the pivotal role of ROS in the impaired BM ECs of PGF patients. However, the mechanisms underlying the abnormal BM ECs of PGF, remains to be elucidated. Energy metabolism plays an instrumental role in maintaining EC function, and markedly perturbed of EC metabolism, results in high level of ROS and dysfunction of ECs, and contributes to many pathologies, like cancer and diabetes. However, little is known about the metabolism state and its role in BM ECs of PGF patients post‐allotransplant.Aims:To determine the metabolic state and its role in BM ECs of PGF patients post‐allotransplant. Moreover, to evaluate the therapeutical potential of anti‐metabolic drugs to the dysfunctional BM ECs derived from PGF patients.Methods:This prospective case‐control study enrolled 15 patients with PGF, 30 matched patients with good graft function (GGF), defined as persistent successful engraftment after allotransplant, and 15 healthy donors (HD). The mitochondrial mass, membrane potential and the protein expression level of metabolism enzymes, CPT1A and PFKFB3 in BM ECs, were detected by flow cytometry. In addition, the cultivated BM ECs were derived from BM mononuclear cells (BMMNCs), as previous reported. The glycolysis inhibitor 3‐PO was administrated to the 5‐day cultivated BM ECs until testing on day 7. The functions of BM ECs were evaluated by apoptosis, migration and tube formation assays. Glucose metabolism levels were measured by glucose consumption and lactate production assays.Results:In this study, elevated expression of the glycolytic activator PFKFB3 was observed in BM ECs of PGF patients, when compared with those of GGF patients and HD, but not the lipid metabolism enzyme CPT1A, mitochondrial mass or membrane potential. Moreover, glycolysis (PFKFB3) inhibitor 3‐PO treatment quantitatively and functionally improved BM ECs derived from patients with PGF in vitro. Mechanistically, we demonstrated that the aberrant glycolysis in BM ECs of PGF could be reduced by NAC treatment in vitro, while the glycolysis in BM ECs of GGF could be induced by hydrogen peroxide treatment in vitro, consistent with ROS‐induced dysfunction of BM ECs. Furthermore, Glycolysis inhibitor 3‐PO treatment attenuated the perturbed function and number of BM ECs derived from GGF patients treated with hydrogen peroxide.Summary/Conclusion:These findings reveal that hyper‐glycolysis is involved in the pathobiology of BM ECs of PGF patients, which is triggered by their high level of ROS. In turn, this metabolism alteration, mediated ROS‐induced dysfunction of BM ECs. What's more, the impaired BM ECs of PGF could be attenuated by glycolysis inhibitor 3‐PO in vitro. Given that several glycolysis enzyme PFKFB3 inhibitors entered Phase I clinical trials in patients with advanced solid malignancies. Our findings might merit further consideration of targeting BM ECs glycolysis as a promising therapeutic approach for PGF patients post‐allotransplant in the future.