Fig. 2: Repression of circMAN1A2(2,3,4,5) inhibits colorectal cancer cell proliferation and tumor progression.

a, b CircMAN1A2(2,3,4,5) was mainly localized in the cytoplasm of HT29 cells, as revealed by smFISH staining of circMAN1A2(2,3,4,5) (white), MAN1A2 (red) and nuclei (blue) (a), and fractionation of the cytoplasmic and nuclear RNA assay (b; n = 3 independent experiments) with GAPDH and MALAT1 as cytoplasmic and nuclear RNA markers. c Copy numbers of circMAN1A2(2,3,4,5) in HT29 and DLD-1. d Designed shRNA-c1 and shRNA-c2 specifically targeting the BSJ site of circMAN1A2(2,3,4,5), both spanning the 5′ splice site of exon5 and the 3′ splice site of exon2. nt, nucleotide. e, f KD efficiency of circMAN1A2(2,3,4,5) in HT29 and DLD-1 cells, as detected by RT-qPCR and normalized to ACTB (e) and Northern Blot (NB) (f) with violet arrow represents circMAN1A2(2,3,4,5). g, h KD of circMAN1A2(2,3,4,5) inhibited cell proliferation in HT29 and DLD-1 cells, as revealed by MTT cell proliferation assay (g; n = 5 wells of cells examined over 3 independent experiments) and colony formation assay (h) with representative images of cell colonies (left panel) and statistics of colony numbers (right panel). i–k Tumor growth curves from nude mice injected subcutaneously with scramble and circMAN1A2(2,3,4,5)-KD HT29 (top panel) and DLD-1 cells (bottom panel) were shown in (i); the images of tumors were shown in (j); statistics of tumor weight were shown in (k). Tumor size was measured every 4 days for 24 days and calculated as width² × length × 0.5. n = 5 biologically independent samples. l, m The image of tumors from NSG mice injected cecum with scramble and circMAN1A2(2,3,4,5)-KD HT29 (top panel) and DLD-1 cells (bottom panel) were shown in (l); statistics of tumor weight were shown in (m). n = 5 biologically independent samples. RNA levels were detected by RT-qPCR (b). Images are representative of three independent experiments (a, f, h). Data are presented as mean ± s.d. (b, c, e, g–i, k, m), with biologically individual data points shown (c, e, h, k, m). P values were determined by ordinary two-way ANOVA with Dunnett’s post-hoc comparisons versus scramble control at each row factor level (adjusted P values) (g, i) and ordinary one-way ANOVA test with Dunnett’s post-hoc comparisons versus scramble control (FDR-adjusted P values) (h, k, m). Source data are provided as a Source Data file.