The effects of extracellular matrix glycosylation inhibition on osteosarcoma invasiveness / Sarmila Hanim Mustafa

Osteosarcoma (OS) is a rare malignant bone cancer affecting children and young
adults. To date, OS still has a poor prognosis despite the use of advanced multimodal
therapy in its treatment. Recurrence cases has also shown to be increased in metastatic
patients especially to the lung. A possible cause for this poor prognosis could be
because of lack of understanding in the underlying mechanism of OS progression and
metastasis. Previously, loss of Wnt/β-catenin pathway activity during osteoblast
differentiation was suggested to contribute to the OS development. In addition, a
metastatic condition of OS was shown to be established through the adhesion of OS
cells to blood vessel mediated by α4β1 integrin. It has been known that aberrant
glycosylation is associated with metastasis in various cancer by modulating the cells
invasiveness and their dissemination through the extracellular matrix (ECM) layer.
However, the effect of aberrant glycosylation, mainly during N-glycan processing in
OS has yet to be elucidated. Therefore, the objective of this study is to investigate the
effects of aberrant glycosylation towards OS invasiveness via the cell-ECM
interaction. In this study, the inhibition of glycosylation was carried out by the
treatment of OS cell line (MG-63) with 1-deoxynojirimycin (1-DNJ); an α-
glucosidase-I/II inhibitor, and 1-deoxymannojirimycin (1-DMJ); an α-mannosidase I
inhibitor. Then, its effects on glycosylation pattern, invasion ability, protein
expression of α4β1 integrin and the expression level of ECM degradative enzymes
genes; MMP-2, MMP-9, TIMP-1, TIMP-2 and β-catenin genes were determined. In
this study, the normal osteoblast cell (hFOB1.19) was also treated with 1-DNJ and 1-
DMJ and both samples were subjected to the same method as MG-63 cells except for
cell invasion assay. The results showed that the inhibition of N-glycosylation by 1-
DMJ enhanced the invasion of MG-63 cells through the ECM layer significantly by
3.87 fold (p<0.05). It was also found that 1-DMJ significantly (p<0.05) regulates the
expression of α4β1 integrin protein in hFOB1.19 cells. qRT-PCR analysis showed that
1-DMJ downregulate the expression MMP-9 gene (2.42 fold ± 0.284; p<0.05) in MG-
63 cells. On the other hand, TIMP-1 (1.54 fold ± 0.035), TIMP-2 (1.22 fold ± 0.041)
and β-catenin (1.24 fold ± 0.07) genes were significantly (p<0.05) upregulated in MG-
63 cells when compared to control. However, in hFOB1.19 cells treated with 1-DMJ,
MMP-2 (48.32 fold ± 0.16), MMP-9 (4.38 fold ± 0.389), and TIMP-1 (1.45 fold ±
0.064) genes were significantly (p<0.05) downregulated when compared to control.
Whereas, the expression of TIMP-2 (1.2 fold ± 0.03), and β-catenin gene (2.22 fold ±
0.232) in hFOB1.19 cells were significantly (p<0.05) upregulated. 1-DNJ was shown
to downregulate the expression of MMP-2 in both cells; MG-63 (1.29 fold ± 0.067)
and hFOB1.19 (55.25 fold ± 0.097) significantly (p<0.05) when compared to control.
However, the expression of β-catenin gene in 1-DNJ treated hFOB1.19 cells was
significantly (p<0.05) increased by 2.4 fold ± 0.103 when compared to control. This
study suggested that 1-DMJ may promote the invasion of MG-63 cells through the
ECM layer via the regulation of MMP-9, TIMP-1, and TIMP-2 genes. Whereas the
overexpression of β-catenin genes in 1-DMJ and 1-DNJ treated hFOB1.19 cells might
give an insight on the epithelial-to-mesenchymal transition (EMT) of normal
osteoblast cells. This study might provide more information on the underlying
mechanism of OS therefore contributes to OS prognosis improvement.