Fig 2: Effect of inhibition of MELK signaling on tumor growth and lung metastasis in TNBC xenograft mouse models. A, MELK expression in Cas9-p15 (P15) and MELK KO MDA-MB-231 (C3 and C28) cells. α-Tubulin was used as a loading control. B–F, MELK KO significantly suppressed lung metastases in an MDA-MB-231 xenograft mouse model. Female athymic nude mice were injected intravenously with luciferase-transfected Cas9-p15 or MELK KO MDA-MB-231-Luc-GFP stable cells. Metastatic tumors were measured weekly for 7 weeks by whole-body luciferase imaging using an IVIS 100 Imaging System. Shown are mouse whole-body luciferase images (B), photon counts per area (C), lung weight per mouse measured at week 7 following cell inoculation (D), mouse OS over a period of 80 days following cell inoculation (E), and IHC staining for hematoxylin and eosin (H&E), MELK, proliferation (Ki67), and mesenchymal markers (vimentin and fibronectin; F) in mice inoculated with Cas9-p15 or MELK KO MDA-MB-231 cells. Images were taken under 200 × magnification. G, MELK-In-17 significantly suppressed tumor growth in a 4T1 xenograft mouse model (P < 0.05). Murine 4T1 TNBC cells were injected into the mammary fat pads of female BALB/c mice. When tumor size reached 75–150 mm3, grouped mice were treated with vehicle or MELK-In-17 at 5 or 10 mg/kg via intraperitoneal injections daily for 25 days. In C–G, data are presented as mean ± SD. *, P < 0.05; **, P < 0.01.

Fig 3: Differential gene expression profile, consolidated gene set enrichment, and network analyses. A, Differences in gene expression are shown in volcano plot format, where the X-axis denotes the log2-transformed fold-change and the Y-axis denotes the –log10-transformed P value. The log2 fold-change was calculated as Cas9-p15 control cell fold-change/MELK KO MDA-MB-231 cell fold-change, resulting in positive values for genes overexpressed in the WT condition and negative values for genes overexpressed in the KO condition. The nominal P value threshold at 5% is indicated by a dashed blue horizontal line. The 10 most differentially expressed genes by P value are labeled using their gene symbol. B, Differences in the hallmark enrichment pattern are shown in a modified volcano plot format, where the X-axis denotes the normalized enrichment score and the Y-axis denotes the –log10-transformed P value. The normalized enrichment score was calculated as Cas9-p15 control cell score/MELK KO MDA-MB-231 cell score, resulting in positive values for hallmarks enriched in the WT condition and negative values for hallmarks enriched in the KO condition, as indicated at the top of the plot. The nominal P value threshold at 5% is indicated by a dashed blue horizontal line. Significant hallmarks are labeled using their gene set name. C, Pathway enrichment profile of the MELK-centered PPI network. In the left part of the plot, significantly enriched pathways, listed along the Y-axis, are represented in dot plot format; the X-axis represents the –log10-transformed P value; and the size of the dots reflects the number of pathway genes contained in the network. To the right of the dot plot, overlaps between enriched pathways and three gene sets defined after analyzing the MELK-centered protein–protein interaction network are indicated in heat map format. The gene sets of interest are indicated in the X-axis underneath the heat map, where dark blue denotes the presence of pathway genes and light blue denotes the absence of pathway genes in these gene sets.

Fig 4: Effect of MELK knockdown or overexpression on proliferation, motility, EMT, and CSC properties in TNBC cells. Effect of MELK knockdown or KO (A) or overexpression (B) on proliferation and anchorage-independent growth of TNBC cells. HCC70 and BT-549 cells (A, bottom left) were treated with two different siMELK sequences for 48 hours. MELK expression was then determined using Western blot analysis (top), and cell viability was determined using trypan blue exclusion assay (bottom left). In MELK KO MDA-MB-231 cells (A, bottom right), anchorage-independent growth was determined using soft-agar assay at week 3 following incubation. β-Actin was used as a loading control for Western blot analysis. SUM149 and SUM159 cells (B) were transfected with plasmids encoding WT MELK gene or KD MELK gene and cultured for 48 hours. MELK expression was then determined using Western blot analysis (top), and cell growth was determined using anchorage-independent growth assay (bottom). β-Actin was used as a loading control for Western blot analysis. Effect of MELK knockdown (C) or overexpression (D) on migration and invasion of TNBC cells. Cells were transfected with two different siMELK sequences for 72 hours or plasmids encoding WT MELK gene or KD MELK gene for 48 hours and then assayed for migration and invasion. E, Effect of MELK knockdown on expression of EMT epithelial and mesenchymal markers in TNBC cells. Cells were treated with two different siMELK sequences for 48 hours, and then MELK expression was determined using Western blot analysis. α-Tubulin was used as a loading control. Effect of knockdown or KO (F) or overexpression (G) of MELK on mammosphere formation in TNBC cells. Cells were transfected with two different siMELK sequences or plasmids encoding WT MELK gene or KD MELK gene and 48 hours later were seeded for mammosphere formation. MELK KO clones were cultured for 7 days and then analyzed for mammosphere formation. H, Effect of MELK knockdown on CD44+/CD24− and ALDH1+ subpopulations in TNBC cells. Cells were treated with two different siMELK sequences and 48 hours later subjected to flow cytometry analysis. In A and F, P15 indicates Cas9-p15 control cells; C3, MELK KO C3 clones; and C28, MELK KO C28 clones. In A–D, G, and F, data are presented as mean ± standard deviation. *, P < 0.05; **, P < 0.01; ***, P < 0.001; ****, P < 0.0001.

Fig 5: Inhibitory effect of MELK-In-17 on anchorage-independent growth, motility, CSC properties, and EMT transition in TNBC cells. A–D, MELK-In-17 inhibited anchorage-independent growth (A), suppressed migration and invasion (B), reduced mammosphere formation (C), and reduced CD44+/CD24− and ALDH1+ CSC subpopulations (D) in TNBC cells. In A, the anchorage-independent growth was determined using a soft-agar assay at week 3 following treatment. In B, cells were pretreated with MELK-In-17 for 2 hours and then assayed for migration and invasion in the presence of MELK-In-17. In C, cells were seeded for mammosphere formation and the next day treated with MELK-In-17. In D, cells were treated with MELK-In-17 for 48 hours and then subjected to flow cytometry analysis. E, Western blot analysis of the effects of treatment with MELK-In-17 on the expression of EMT markers. α-Tubulin was used as a loading control. Band intensity of proteins is normalized to that of α-Tubulin. In A–C, data are presented as mean ± SD. *, P < 0.05; **, P < 0.01; ***, P < 0.001; ****, P < 0.0001.