Figure 1.
Expression of MMP-2 and -9, gelatinase secretion and invasion rate in different leukemic cells.
(A) qRT-PCR analysis of MMP-2 (A1) and MMP-9 (A2) gene transcription in SHI-1, HL-60 and U937 leukemic cells. (B) Gelatinase secretion from culture supernatants of three leukemic cells. (C) Cell invasion rates of SHI-1, HL-60 and U937 leukemic cells. Cell invasion rates were determined by the ratios relative to SHI-1 cells. Data represent the mean ± SD of triplicate measurement representative for three independent experiments relative to SHI-1 cells (set as 100%). Compare with SHI-1 cells *P<.001.
Figure 2.
Increased in vitro BBB permeability due to MMP-2 and -9 secreted by SHI-1, HL-60 and U937 cells.
(A) Morphological changes of BMVECs affected by SHI-1, HL-60 and U937 cells treated with or without GM6001. (B) Rates of leukemic cell invasion of co-culture with BMEVCs compared with those of monoculture (without BMEVCs) (P>.05) and those treated with GM6001 (*P<.001). Results are shown as the mean ± SD of three independent experiments. (C) Zymography gelatinase secretion in SHI-1, HL-60 and U937 cells treated with or without GM6001. Original magnification: (A)×600.
Figure 3.
Confocal imaging showed protein localization of ZO-1, claudin-5 and occludin in human BMEVCs.
(A) Confocal imaging of ZO-1, claudin-5 and occludin of human BMEVCs; (B, C, D) BMEVCs co-cultured with SHI-1, HL-60 and U937 cells with or without treatment by GM6001. (E) BMEVCs of co-cultured with SHI-1 cells after knock-down of the MMP-2/MMP-9 gene. Data in all panels are representative of at least 3 separate experiments on 3 distinct cultures. (Scale bars: 10 µm).
Figure 4.
Effects of MMP-2/MMP-9 knock-down on activity of gelatinase and rates of cell invasion by SHI-1 in the in vitro BBB model.
(A) Transcription of specific mRNAs quantified by qRT-PCR 48 h after siRNA transfection. SHI-1 cells transfected with siRNAs targeting the expression of MMP-2 (KD-MMP-2) or (B) MMP-9 (KD- MMP-9). Control SHI-1 cells transfected with non-target-directed siRNA (set as 100%). (C) Gelatinase secretion from SHI-1 cells transfected with siRNAs. (D) Rates of cell invasion of SHI-1 cells 48 h after transfection with siRNA in in vitro model of BBB. Data represent the mean ± SD of triplicate measurement representative for three independent experiments relative to SHI-1 cells (set as 100%). *P<.001.
Figure 5.
The SCI-nu/nu mice model of CNS leukemia.
(A) Clinical signs of CNS leukemia in mice. (A1, A2, arrow) The mouse developed paralysis in its left forelimb and cranial nerve with light loss. (A3, arrow) The mice developed paralysis in both fore limbs and hind limbs. (B, arrow) Neoplasms of different parts with or without intracranial hemorrhage in removed brains. (C) Infiltrations of SHI-1 examined with histopathology assay. (C1, C2, C3) Brain parenchyma pathobiology stained with H&E. (C4) Examination of bone marrow smears stained with Wright's-Giemsa. (D) Organs with infiltrations of SHI-1 examined with RT-PCR analysis of human MLL/AF6 fusion gene transcription in mice with or without GM6001 treatment. (D1) RT-PCR analysis of MLL/AF6 fusion gene of brains and bone marrow from SCI nu/nu-mice. Lanes 1–5: Human MLL/AF6 fusion genegene amplified in SHI-1 cell and in brain, bone marrow, stomach and liver of mice with CNS leukemia. lane 6–9: Human MLL/AF6 fusion gene amplified in stomach, liver, brain and bone marrow of mice with GM6001 treatment. lane 10: negative control. (D2) The number of organs with infiltrations of SHI-1 examined with RT-PCR analysis of human MLL/AF6 fusion gene transcription in mice with or without GM6001 treatment. Original magnification: (C1), ×40; (C2), ×200; (C3), ×1000; (C4), ×1000.
Figure 6.
Activities of MMP-2 and -9 secreted by SHI-1 with or without GM6001 treatment in CNS leukemia mice.
(A) RT-PCR analysis of human MMP-2 and -9 gene transcription in normal brain tissues, brain tissues infiltrated with SHI-1 cells with or without GM6001 treatment. (B, C and D) In situ zymographic analysis of MMP-2 and -9 activities mouse tissues. (B) Normal brain tissues; (C) Brain tissues infiltrated with SHI-1 cells; (D) Brain tissues after treatment with GM6001. Results shown are from individual animals and are representative of findings from 3 experiments with 6 animals per condition per experiment. (Scale bars: 10 µm).
Figure 7.
Increased BBB disruption, with a decrease in endothelial ZO-1, claudin-5 and occludin, avoided by GM6001.
(A) Confocal imaging of cerebral cortex from normal mice immunostained for ZO-1 localize at vessels, confirming the use of CD31. (B1, B2) Confocal imaging of TJ proteins ZO-1, claudin-5, occludin and fibrinogen in brain parenchyma in control and CNS leukemia mice. (C) Immunoblots of protein extracts from brains of control mice, 4 CNS leukemia mice with brain parenchyma and CNS leukemia mice treated with GM6001. In all figures, brain tissues of normal mice used as control. (D) Model of MMP-2 and -9 secreted by leukemic cells disrupting several key TJ proteins of BMVECs in BBB breakdown. Results shown are from individual animals and are representative of findings from 3 experiments with 4 animals per condition per experiment. (Scale bars: 10 µm).