• 2019-10
  • 2019-11
  • 2020-03
  • 2020-07
  • 2020-08
  • 2021-03
  • br Fig Identification of hsa


    Fig. 1. Identification of hsa_circ_0001461 [circFAT1(e2)] as a GC related circRNA. (A) Hierarchical clustering analysis of circRNAs that were differentially expressed in gastric tumor tissues and normal tissues in GSE100170 (left) and GSE83521 (right). (B) Overlap of dysregulated circRNAs in GSE100170 and GSE83521.
    (C) Schematic diagram of the genomic location of circFAT1(e2). (D) CircFAT1(e2) was validated by RT-PCR using divergent primers in cDNA and gDNA. (E) Sanger sequencing of circFAT1(e2) showed the back-splice junction. (F) RT-PCR analysis of circFAT1(e2) and linFAT1 in MGC-803 cells treated with actinomycin. (G) RT-PCR analysis of circFAT1(e2) and linFAT1 in MGC-803 cells treated with RNase R. The expression of circFAT1(e2) and linFAT1 were normalized to the value detected in the mock group.
    To further confirm the correlation between circFAT1(e2) and miR-548g, we established circFAT1(e2) overexpressed and blocked cell models by transfecting MGC-803 cells with circFAT1(e2) and si-circFAT1(e2), respectively. We designed three siRNAs against 
    Fig. 2. circFAT1(e2) was frequently downregulated in GC and predicted better prognosis. (A) Relative expression of circFAT1(e2) from 38 pairs of normal tissues and GC tissues measured by RT-PCR. (B) Relative expression of circFAT1(e2) from metastasis and non-metastasis GC tissues detected via RT-PCR. P < 0.05.
    that circFAT1(e2) could negatively regulate the expression of miR-548g in GC cells.
    3.5. miR-548g promoted GC cell tumorigenesis partly though RUNX1
    Bioinformatics analysis showed three Ac-DEVD-CHO for miR-548g in 
    the 3′-UTR of RUNX1 (Fig. 5A). We then evaluated the interaction between miR-548g and RUNX1 by dual-luciferase reporter assay (P < 0.05, Fig. 5B). In addition, Western blot assay showed that RUNX1 expression was significantly higher in circFAT1(e2) transfected MGC-803 and MKN-28 cells than in cells treated with NC (Fig. 5C). RUNX1 expression was obviously lower in si-circFAT1(e2) transfected
    Fig. 3. CircFAT1(e2) was an important tumor suppressor for GC cell survival and proliferation. (A) Schematic diagram of the cicrFAT1 overexpression vector.
    (B) The transfection efficiency of cicrFAT1 was evaluated by RT-PCR in MGC803 and MKN-45 GC cell lines, P < 0.05. (C and D) Growth curves of MGC-803 and MKN-28 cells transfected with circFAT1(e2) or negative control (NC), P < 0.05. A colony formation assay was performed to measure the cell proliferation of MGC-803 cells (E and F) and MKN-28 cells (G and H) transfected with circFAT1(e2) or NC, P < 0.05. (I-K) Hypodermic injection of MGC-803 cells transfected with circFAT1(e2) or NC into BALB/c nude mice established the xenograft model; tumor volume and weight were measured, P < 0.05. (L) Immunohistochemical analysis of Ki-67 expression in tissues from xenograft, P < 0.05.
    SGC-7901 and AGS cells than those cells treated with si-NC (Fig. 5D). RUNX2 and RUNX3 expression were not affected by circFAT1(e2) overexpression and knockdown in GC cells. Further exploration of RUNX1 expression revealed that miR-548g could reverse the upregu-lation of RUNX1 induced by circFAT1(e2) in MGC-803 and MKN-28 cells Ac-DEVD-CHO (Fig. 5E). Taken together, our findings demonstrated that cir-cFAT1(e2) could upregulate RUNX1 expression by miR-548g.
    Cell growth curves suggested that overexpression of miR-548g sig-nificantly promoted cell growth, and miR-548g could abolish the downregulation of cell growth induced by RUNX1 in MGC-803 and MKN-28 cells (P < 0.05, Fig. 5F and G). Colony formation assays showed a significantly higher number of colonies in miR-548g 
    overexpressed GC cells, and overexpression of miR-548g reversed the reduction of the number of colonies induced by RUNX1 (Fig. 5H and I). Moreover, we found that overexpression of miR-548g significantly in-creased the invasive ability of MGC-803 cells. However, overexpression of RUNX1 remarkably decreased the invasive ability of MGC-803 cells, and miR-548g could partly abolish the effects induced by over-expression of RUNX1 (P < 0.05, Figs. S4A and S4B). These results indicated that miR-548g could promote GC tumorigenesis through RUNX1.