PF150 ADAM28 OVEREXPRESSION IN ACUTE LYMPHOBLASTIC LEUKEMIA PROMOTE ITS INVASION IN CENTRAL NEURAL SYSTEM VIA CROSS TALKING WITH CXCR4 PATHWAY

Background: ADAM28 has been shown to relate with tumor proliferation, metastasis and prognosis. Our previous work showed that overexpression of ADAM28 in bone marrow of acute lymphoblastic leukemia (ALL) patients is related with poor prognosis. Central neural system (CNS) involvement is still a challenge in treatment of ALL and the mechanism is unknown. Our study revealed that the overexpression of ADAM28 might be one of the indicator and key mechanism in the occurrence of CNS leukemia. Aims: To explore the effect and mechanism of ADAM28 expression in ALL blast cell on its CNS metastasis. Methods: Bone marrow samples were collected from 50 ALL patients and 10 healthy donors. The expression of ADAM28 in BM were detected through flow cytometry. The tumor xenograft mouse model was established via the femoral vein injection of primary ALL cells or Nalm‐6 cell line after a sublethally irradiation and intraperitoneal injection with CD122 in male 8‐week‐old NOD/SCID mice. CNS symptoms such as seizures and paraplegia were observed after the injection of ALL cells. The mice were sacrificed on days 1, 3, 5, and 7 after injection respectively, and the brain, liver, spleen, kidney and BM were collected, samples were then embedded in paraffin. Sections were mounted on slides, deparaffinized and stained with hematoxylin and eosin (HE) or anti‐CD45 antibody. Knock‐down and overexpression of ADAM28 in primary ALL cells and cell line were achieved by transfection of ADAM28‐shRNA or ADAM28 plasmid. Ly294002 as ADAM28 inhibitor and CXCR4 Inhibitory antibody were added in the in vitro culture of primary ALL cells and cell lines. Transwell chambers were used to verify the migration and invasion ability of ALL cells. Western blot (WB) and RT‐PCR was performed to verify the expression level of ADAM28, CXCR4, IGFBP3 in order to analysis the cross talk between the two pathways. Results: The expression of ADAM28 in BM of ALL patients with CNSL were significantly higher than that in ALL patients without CNSL. In the xenograft mouse model, mice injected with ALL cells in which ADAM28 was overexpressed showed a significant higher incidence rate of seizures or paraplegia. The HE stain results showed that the involvement of brain and meninges in ADAM28 overexpress group was significantly higher than control, while there was no difference in liver, spleen and kidney invasion in the two groups. Clonal tracking experiments results that most of the ALL cells observed in brain and meninges of mice injected the mixture of ADAM28‐overexpression‐GFP ALL cells and empty plasmid ALL cells were GFP positive, which demonstrated that the invasion of ALL cells was increased when ADAM28 overexpressing. In vitro, transwell chamber results showed that the migration ability of ALL cells with high ADAM28 level was significantly higher than that of controls, and the migration ability was decreased when ADAM28 was knocked down or treated with ADAM28 inhibitor. Additional, when ADAM28 was knocked down, the expression of CXCR4 in mRNA level and protein level were decreased, and the ADAM28 expression level was decreased when decreased when treated with CXCR4 inhibitory antibodies, which suggested that there might be a cross talk between ADAM28 and CXCR4 signaling pathway. Summary/Conclusion: The incidence of CNSL in ADAM28 overexpressing ALL patients was increased. The invasion into CNS of ALL cells was increased when ADAM28 was overexpressed in xenograft mouse model. ADAM28 might promote the migration and invasion ability via cross talking with CXCR4 pathway.