Fig 1.
Dose-dependent response to panobinostat or romidepsin in adherent TU-BcX-4IC cells.
TU-BcX-4IC cells were plated and treated with serial dilutions of panobinostat, romidepsin, or DMSO control for 72 hours. Cell viability and morphology were compared to DMSO control. Images were captured at 4X and 10X. (A) Panobinostat and romidepsin treatment to TU-BcX-4IC adherent cells demonstrate cytoxicity at 50 nM and 0.1 μM, respectively. Representative images are shown at 4X magnification. (B) Panobinostat and romidepsin treatment reverse the mesenchymal transition and promote epithelial-like characteristics. Representative images are higher magnifications of the samples in panel A; images are shown at 10X magnification. (C) Dose-response analysis of TU-BcX-4IC treated with panobinostat, romidepsin, or DMSO shows increased potency and efficacy of romidepsin compared to panobinostat. After crystal violet staining, plates were lysed and absorbance was measured (570 nm) and normalized to DMSO controls to quantify drug response. (D) TU-BcX-4IC cells stained with phalloidin and DAPI showed altered distribution of actin filaments in romidepsin, and panobinostat-treated cells. (E) Transwell migration of TU-BcX-4IC cells pre-treated for 72 hours with DMSO, romidepsin, or panobinostat (24 hours migration, 100 nM treatments). Error bars are represented as standard error of mean. ***p < 0.001. All scale bars represent 0.25 mm.
Fig 2.
Romidepsin cytotoxicity to TU-BcX-4ICT4 patient-derived organoids and in vivo effects on tumorigenesis and circulating tumor cells.
(A) PDX-Os were generated from TU-BcX-4ICT4 in 3D culture conditions and treated for 72 hours with romidepsin (100 nM) or DMSO control. PDX-Os were treated with a live (Calcein AM)/dead (EthD II) immunofluorescent staining kit to highlight live (green) or dead (red) cells. Representative images were captured at 40X magnification (brightfield images) or 50X magnification (immunofluorescent images). Scale bars represent 1.25 mm. (B) TU-BcX-4IC tumor pieces (passage 2; T2) were implanted bilaterally in the mammary fat pads of SCID/Beige mice treated with romidepsin or DMSO vehicle. Tumor volume was measured biweekly with calipers and after 20 days, tumors were excised and mice were sacrificed. (C) Twenty days after tumor excision, peripheral blood was collected to evaluate circulating tumor and stem cell populations, defined as HLA+CD44+CD24-. Percentage of circulating tumor cells were significantly reduced in mice treated with romidepsin compared to those treated with DMSO. Error bars represent standard error of mean. *p < 0.05.
Fig 3.
Romidepsin effect on metastasis in TU-BcX-4IC-implanted mice.
Lungs and livers from mice implanted with TU-BcX-4IC and treated with romidepsin or DMSO vehicle were harvested, formalin-fixed, paraffin-embedded, and H&E-stained to visualize and quantify metastases. Romidepsin treatment (A) significantly decreased the number of metastatic lesions per lung section and (B) overall decreased the area of lung metastases compared to DMSO vehicle control. (C) Representative images of H&E-stained lungs in both treatment groups. The black box on the low magnification images indicates the regions shown in the higher magnification views. In liver sections, romidepsin (D) dramatically reduced the number of metastatic lesions per liver section and (E) overall decreased the area of liver metastases compared to DMSO, but these findings were not statistically significant. (F) Representative images of H&E-stained livers with arrows indicating metastatic foci in both treatment groups. Error bars represent standard error of mean (S.E.M.). ***p < 0.0001; ns = not significant.
Fig 4.
RNA sequencing analysis of romidepsin-treated TU-BcX-4IC cells compared to DMSO control.
Romidepsin altered expressions of (A) EMT-related genes. Other cell signaling pathways that were affected include (B) cell cycle genes and (C) extracellular matrix-associated genes in TU-BcX-4IC cells. Data is shown as Log2 (fold change). mRNA expression of EMT genes (CDH1, VIM, CDH2, ZEB1, ZEB2) were analyzed using qRT-PCR.
Fig 5.
Effects of romidepsin on gene expression differs amongst treated cells, spheres, PDX-Os, and implanted tumors.
Romidepsin treatment in TU-BcX-4IC (A) cell line, (B) spheres, (C) PDX-Os, and (D) in vivo tumor pieces implanted in SCID/Beige mice. qRT-PCR analysis was repeated with genes affected by romidepsin treatment compared to DMSO control based on RNA sequencing analyses (PLK1, FOXM1, MKI67, CDKN1A, PLAU, FOS, FRA1). Again, romidepsin-treated TU-BcX-4IC (E) cells, (F) spheres, (G) PDX-Os, and (H) in vivo tumor pieces implanted in mice were used. Analyses were all normalized to β-actin and DMSO treated controls. Black bars represent DMSO; maroon bars represent romidepsin treatment (100 nM, 72 hours). *p < 0.05, **p < 0.01, ***p < 0.001.
Fig 6.
Romidepsin sensitizes TU-BcX-4IC cells to select oncology drugs.
TU-BcX-4IC cells were treated only with agents from the NCI-approved oncology set or pre-treated with romidepsin (50 nM, 72 hours) prior to the oncology agents. Crystal violet staining was performed to visualize differences in cell viability between the groups. (A) Romidepsin sensitized TU-BcX-4IC cells to HER-targeted inhibitors (afatinib, lapatinib), the ER inhibitor fulvestrant, MEK1/2 inhibitors (trametinib, cobimetinib), as well as other targeted inhibitors (dabrafenib, everolimus, ponatinib, crizotinib), and bleomycin. Representative images are shown at 40X magnification. Response of both treatment groups to fluorouracil is shown as an example of baseline response to romidepsin, as well an example of 50% response of TU-BcX-4IC cells. (B) TU-BcX-4IC cells were pre-treated with romidepsin (50 nM) for 24 hours before treatment with cobimetinib and trametinib (1 μM) for an additional 24 hours. (C) TU-BcX-4IC cells were co-treated with romidepsin (50 nM) and cobimetinib or trametinib (1 μM) for 72 hours. (D) Quantification of cells remaining when cells were treated with romidepsin alone or pre-treated with romidepsin for 24 hours followed by cobimetinib or trametinib. Notably, the abundance of cells in the DMSO group prevented quantification of this treatment control and is not included in the graph. All scale bars represent 0.25 mm. ***p < 0.001.