High-throughput drug library screening identifies colchicine as a thyroid cancer inhibitor

Abstract

We employed a high-throughput drug library screening platform to identify novel agents affecting thyroid cancer cells. We used human thyroid cancer cell lines to screen a collection of approximately 5200 small molecules with biological and/or pharmacologial properties. Parallel primary screens yielded a number of hits differentially active between thyroid and melanoma cells. Amongst compounds specifically targeting thyroid cancer cells, colchicine emerged as an effective candidate. Colchicine inhibited cell growth which correlated with G2 cell cycle arrest and apoptosis. These effects were hampered through inhibition of MEK1/2 and JNK. In contrast, inhibition of p38-MAPK had little effect, and AKT had no impact on colchicine action. Systemic colchicine inhibited thyroid cancer progression in xenografted mice. These findings demonstrate that our screening platform is an effective vehicle for drug reposition and show that colchicine warrants further attention in well-defined clinical niches such as thyroid cancer.

Keywords: BRAF; BRAF resistance; colchicine; high throughput drug screening; thyroid cancer.

Figure 1. High-throughput drug library screening

BRAF-mutant thyroid cancer 8505C and KTC-1 cells and melanoma cells (Malme-3M) were used as biological sources to screen, under high-throughput conditions, a multi-vendor collection of ∼5200 chemical entities with known pharmacological and biological properties. (A) Top left panel shows Z-score values obtained from parallel screens of cells. Bottom left panel displays a comparative chart filled with control-based normalized data resulting from the screen of thyroid cells. Samples with less than or equal to 40% viability are highlighted. Control-based normalization is calculated as follows: 100 (signal-negative control)/(positive control-negative control). PLX4720 is a BRAF inhibitor. A heat map of drug effects on cell viability is shown on the right. Red indicates highest and blue reflects lowest cell viability. AV-412 is an EGF-R inhibitor, and SB590885 is a BRAF inhibitor. (B) The dose-dependent effect of the BRAF inhibitor PLX4032 and (C) The EGF-R inhibitor AV-412 on cell viability after 48 hrs incubation. Results are expressed as a percentage in relation to positive control (100%) treated with vehicle alone, and reported as the means ± SD of three independent experiments performed with four replicates. *p < 0.05; **p < 0.01 comparing melanoma with thyroid cancer cells at each indicated dose.

Figure 2. Validation of colchicine as an inhibitor of thyroid cancer cells

(A) BRAF-mutant thyroid (8505C and KTC-1) and melanoma (Malme-3M) cells were treated in the presence of increasing doses of colchicine for 48 hrs and assessed for cell viability. *p < 0.05; **p < 0.01 comparing melanoma with thyroid cancer cells at each indicated dose. (B) Cell density was also monitored in two additional thyroid cancer cell lines that are BRAF-WT (WRO and TPC-1). Values are means ± SD of three independent experiments. **p < 0.01 comparing colchicine with DMSO control at the same time point. (C) Cell cycle analysis was monitored by flow cytometry using propidium iodide (PI) dye staining. After 24 hrs of serum starvation, cells were treated with vehicle (DMSO) or colchicine at different doses and times as indicated. Cell cycle profile is estimated by gating histograms generated with the FL2-area variable. The percentage of cells is shown as the mean ± SD of three independent experiments immediately below. *p < 0.05; **p < 0.01 comparing indicated dose of colchicine with DMSO control at the same time point and cell cycle phase.

Figure 3. Impact of colchicine on thyroid cancer cell apoptosis

(A) After 24 hrs serum starvation, 8505C and WRO cells were incubated with vehicle (DMSO) or colchicine as shown. The apoptotic cell population was detected by Annexin V-FITC and PI staining using flow cytometry. The percentage of apoptotic cells is shown as the mean ± SD of three independent experiments immediately below. (B) 8505C and WRO cells were treated with or without 0.1 μM of colchicine and incubated for different times as shown prior to Western blotting. (C) 8505C and WRO cells were treated for 72 hrs with the indicated doses of colchicine prior to Western blotting. *p < 0.05; **p < 0.01 comparing indicated dose of colchicine with DMSO control at the same time point.

Figure 4. Involvement of MAPK in colchicine-induced thyroid cancer cell apoptosis

Western blotting of 8505C and WRO cells to assess PARP cleavage under different treatment conditions. After 24 hrs serum starvation, cells were incubated with vehicle (DMSO), 15 μM of the MEK/ERK1/2 inhibitor U0126 (A), 50 μM of the p38 inhibitor SB203580 (B), 25 μM of the JNK inhibitor SP600125 (C), and 50 μM of the PI3K inhibitor LY294002 (D) in the presence or absence of 0.1 μM of colchicine for 72 hrs. (E) Cells were treated as shown in panels A–D and subjected to flow cytometry analysis for evaluation of Annexin V-FITC and PI staining. Results were interpreted as follows: (Annexin V⁻, PI⁻ : viable cells); (Annexin V⁺, PI⁻ : cells undergoing early apoptosis); (Annexin V⁺, PI⁺: cells in late apoptosis); and (Annexin V⁻, PI⁺: necrotic cells). The percentage of apoptotic cells is shown as the mean ± SD of three independent experiments. (F) Effects of MAPK and PI3K inhibitors on cell viability under the treatment conditions shown. Cell viability was monitored by the Alamar Blue assay. Data are shown as the mean ± SD of three experiments. *p < 0.05; **p < 0.01 comparing colchicine with combination treatments of indicated inhibitors.

Figure 5. Colchicine–resistant thyroid cancer cells (8505C) show reduced MEK/ERK1/2 and JNK activity

(A) Parental 8505C cells and two resistant clones 8505C–R2 and 8505C–R4 were subjected to serum starvation for 24 hrs. Cells were then treated with vehicle (DMSO) or colchicine (0.1 μM and 1 μM) for 72 hrs, and subjected to Western blotting for assessment of MDR-1 expression, MAPK activation, and PARP cleavage. (B) Cell viability by Alamar Blue staining of parental 8505C cells and two resistant clones 8505C–R2 and 8505C–R4 was performed in the presence of increasing concentrations of colchicine incubated for 72 hrs as indicated. The percentage of viable cells is shown as the mean ± SD of three independent experiments. **p < 0.01 comparing parental cells with resistant clones at each indicated dose.

Figure 6. Colchicine activity in thyroid cancer mouse xenografts

Five × 10⁶ thyroid cancer 8505C and WRO cells were introduced subcutaneously into female SCID mice. Once a tumor was palpable, animals were treated for two weeks with the indicated doses of colchicine or DMSO vehicle control. Results shown include the following evaluations: Tumor volumes (A, F), tumor weight (B, G), body weight (E, J), TUNEL staining of xenografted tissue to assess apoptosis (C, H), and PHH3 staining to visualize mitotic phase (D, I). Values represent the means ± SD with 5 animals in each experimental group. *p < 0.05; **p < 0.01 comparing colchicine with DMSO control at each indicated dose.