Obatoclax overcomes resistance to cell death in aggressive thyroid carcinomas by countering Bcl2a1 and Mcl1 overexpression

in Endocrine-Related Cancer
View More View Less
  • 1 Department of Developmental and Molecular Biology, Departments of Medicine, Pediatrics and Committee on Genetics, Department of Pathology, Albert Einstein College of Medicine, Price Center for Genetic and Translational Medicine, 1301 Morris Park Avenue, Room 302, Bronx, New York 10461, USA

Poorly differentiated tumors of the thyroid gland (PDTC) are generally characterized by a poor prognosis due to their resistance to available therapeutic approaches. The relative rarity of these tumors is a major obstacle to our understanding of the molecular mechanisms leading to tumor aggressiveness and drug resistance, and consequently to the development of novel therapies. By simultaneously activating Kras and deleting p53 (Trp53) in thyroid follicular cells, we have generated a novel mouse model that develops papillary thyroid cancer invariably progressing to PDTC. In several cases, tumors further progress to anaplastic carcinomas. The poorly differentiated tumors are morphologically and functionally similar to their human counterparts and depend on MEK/ERK signaling for proliferation. Using primary carcinomas as well as carcinoma-derived cell lines, we also demonstrate that these tumors are intrinsically resistant to apoptosis due to high levels of expression of the Bcl2 family members, Bcl2a1 (Bcl2a1a) and Mcl1, and can be effectively targeted by Obatoclax, a small-molecule pan-inhibitor of the Bcl2 family. Furthermore, we show that Bcl2 family inhibition synergizes with MEK inhibition as well as with doxorubicin in inducing cell death. Thus, our studies in a novel, relevant mouse model have uncovered a promising druggable feature of aggressive thyroid cancers.

Abstract

Poorly differentiated tumors of the thyroid gland (PDTC) are generally characterized by a poor prognosis due to their resistance to available therapeutic approaches. The relative rarity of these tumors is a major obstacle to our understanding of the molecular mechanisms leading to tumor aggressiveness and drug resistance, and consequently to the development of novel therapies. By simultaneously activating Kras and deleting p53 (Trp53) in thyroid follicular cells, we have generated a novel mouse model that develops papillary thyroid cancer invariably progressing to PDTC. In several cases, tumors further progress to anaplastic carcinomas. The poorly differentiated tumors are morphologically and functionally similar to their human counterparts and depend on MEK/ERK signaling for proliferation. Using primary carcinomas as well as carcinoma-derived cell lines, we also demonstrate that these tumors are intrinsically resistant to apoptosis due to high levels of expression of the Bcl2 family members, Bcl2a1 (Bcl2a1a) and Mcl1, and can be effectively targeted by Obatoclax, a small-molecule pan-inhibitor of the Bcl2 family. Furthermore, we show that Bcl2 family inhibition synergizes with MEK inhibition as well as with doxorubicin in inducing cell death. Thus, our studies in a novel, relevant mouse model have uncovered a promising druggable feature of aggressive thyroid cancers.

Introduction

Thyroid cancer is the fifth most prevalent cancer in women, with over 45 000 new cases estimated for 2013 (Siegel et al. 2013). Although the vast majority of these tumors are effectively managed by surgical resection followed by radioactive iodine therapy (American Thyroid Association (ATA) Guidelines Taskforce on Thyroid Nodules and Differentiated Thyroid Cancer et al. 2009), a subset of lesions, which includes recurring well-differentiated, poorly differentiated, and anaplastic tumors, is refractory to current therapeutic approaches, behaves aggressively, is almost invariably fatal, and thus represents a critical clinical issue (Siironen et al. 2010, Ibrahimpasic et al. 2013).

Poorly differentiated thyroid tumors have been recently recognized as a defined, although quite heterogeneous, clinical entity, characterized by the presence of a solid, trabecular, or insular growth pattern, absence of conventional nuclear features of papillary carcinoma, and the presence of at least one of the following features: convoluted nuclei, high mitotic activity, or tumor necrosis (Volante et al. 2007).

Genetic lesions associated with aggressive poorly differentiated tumors include activation of RAS isoforms and inactivation of p53 (TRP53) (Nikiforov 2004) and, with a lower frequency, activation of BRAF or presence of the RET/PTC rearrangement (Ricarte-Filho et al. 2009).

Among the RAS isoforms, NRAS represents the most commonly mutated gene (Volante et al. 2009), although in a different series KRAS was found to be mutated in approximately 45% of all PDTCs analyzed (Garcia-Rostan et al. 2003).

Based on these data, we have generated a compound mouse model carrying a constitutively active allele of Kras (G12D) and a null p53 allele. The use of a human thyroid peroxidase promoter-driven Cre recombinase restricts the functionality of the engineered alleles to the thyroid epithelial cells. We show that double-mutant mice invariably develop PDTC in a background of papillary thyroid cancer and that these tumors phenocopy human PDTC and its aggressive behavior. Furthermore, we identify Bcl2a1 (Bcl2a1a) and Mcl1 as viable therapeutic targets for restoring the sensitivity of mutant cells to drug-induced cell death.

Materials and methods

Animals

The KrasG12D,p53lox/lox, and TPO-Cre strains have been described previously (Jonkers et al. 2001, Kusakabe et al. 2004, Tuveson et al. 2004). All strains were backcrossed in the 129Sv background for at least ten generations, and littermates were used as controls.

Hormone measurements

Blood was collected by cardiac puncture. Serum thyroid-stimulating hormone (TSH) was measured using a sensitive, heterologous, disequilibrium double-antibody precipitation RIA (Pohlenz et al. 1999), and results were expressed in mU/l. All samples were individually analyzed for each mouse. Total thyroxine (T4) concentrations were measured by a solid-phase RIA (Coat-a-Count; Diagnostic Products Corp., Los Angeles, CA, USA) adapted for mice. The values of the respective limits of assay sensitivities were assigned to samples with undetectable TSH or T4 concentrations.

Immunohistochemistry

Sections of 6 μm thickness were subjected to antigen retrieval, incubated with antibodies against thyroglobulin and Ki67 (Dako, Carpinteria, CA, USA), or vimentin, E-cadherin, and pAKT-S473 (Cell Signaling, Danvers, MA, USA), and counterstained with hematoxylin.

Establishment and maintenance of cell lines

Primary thyroid tumors were minced and resuspended in Ham's F12/10% FBS with 100 U/ml type I collagenase (Sigma) and 1 U/ml dispase (Roche). Enzymatic digestion was carried out for 60 min at 37 °C. After digestion, cells were seeded in Ham's F12 containing 40% Nu-Serum IV (Collaborative Biomedical, Bedford, MA, USA), gly–his–lys (10 ng/ml, Sigma), and somatostatin (10 ng/ml, Sigma) and allowed to spread and reach confluence before being passaged. After the fourth passage, tumor cells were adapted to grow in DMEM/10%FBS.

Cal62 cells were grown in RPMI/10%FBS. Identity was validated by amplifying and sequencing genomic fragments encompassing their known mutation (KRAS G12R).

Metaphase preparation and chromosome analysis

For chromosome preparation, primary cells at passage p1–p4 were plated in a 35 mm Petri dish and incubated with Colcemid (Sigma–Aldrich) at a concentration of 10 ng/ml overnight or for 3 h at 100 ng/ml. Chromosomes were extracted with standard hypotonic treatment (0.075 M KCl), dropped on microscope slides, and mounted with antifade containing DAPI (Invitrogen). Slides were imaged on a Zeiss Axiovert 200 Microscope using a DAPI filter (Chroma Technologies, Bellows Falls, VT, USA). Ten fields were selected randomly and 15–30 cells for each tumor line were subjected to chromosome counts and visually inspected for the presence of gross chromosome abnormalities and chromosome fragments.

Real-time PCR

Total RNA was extracted with Trizol and reverse transcribed using the Maxima First Strand cDNA Synthesis Kit (Thermo Scientific, Waltham, MA, USA). qRT-PCR was performed on a StepOne Plus apparatus using the Absolute Blue qPCR Rox Mix (Thermo) and TaqMan expression assays (Applied Biosystems) or the SYBR Green mix (Applied Biosystems) and custom-designed primers (sequences available upon request). Each sample was run in triplicate and 18S was used to control for input RNA. Data analysis was based on the threshold cycle (Ct) method, and experiments were repeated at least three times using at least two independent organ pools (at least five mice/pool).

Western blot analysis

Cells were homogenized on ice in RIPA buffer supplemented with Complete protease inhibitor tablets (Roche Diagnostics). Western blot analysis was performed on 20–40 μg proteins using phospho-antibodies obtained from Cell Signaling.

Antibody arrays

The Proteome Profiler Human Phospho-Kinase Antibody Array (#ARY003B R&D System, Minneapolis, MN, USA) was probed with extracts from control and GSK1120212-treated cells according to the manufacturer's instructions.

Drug treatments, cell viability, and cell proliferation analysis

GSK1120212 and Obatoclax (Selleck Chemicals, Houston, TX, USA) and doxorubicin (Sigma–Aldrich) were added 24 h after plating, in sextuplicate. After 48–72 h, viability was assessed using the Wst1 assay (Clontech) and IC50s were determined using the Prism software. Alternatively, cells were trypsinized and counted using a Z2 Coulter counter (Beckman Coulter, Indianapolis, IN, USA).

Synergy analysis

Statistical analysis of drug synergy was evaluated from the results of the Wst1 assays and calculated using the Chou–Talalay method (Chou & Talalay 1984) and Compusyn software (www.combosyn.com). To determine synergy between two drugs, the software uses a median-effect method that determines whether the drug combination produces greater effects together than expected from the summation of their individual effects. The combination index (CI) values were calculated for the different dose–effect plots (for each of the serial dilutions) based on the parameters derived from the median-effect plots of the individual drugs or drug combinations at the fixed ratios. The CI was calculated based on the assumption of mutually nonexclusive drug interactions. CI values significantly greater than 1 are antagonistic, not significantly different from one are additive, and values less than 1 are synergistic.

Cell-cycle DNA analysis

Cells were seeded in 10 cm plates, cultured overnight at 37 °C, and incubated for 48 h with GSK1120212 or Obatoclax. Cells were harvested by trypsin treatment and fixed in 75% ethanol on ice. After treatment with RNase, cells were stained with propidium iodide overnight at 4 °C, and the DNA content was measured using a Becton Dickinson LSRII System (BD Biosciences, Franklin lakes, NJ, USA).

Annexin V staining

Cells were seeded in 10 cm plates, cultured overnight at 37 °C, and incubated for 48 h with GSK1120212 or Obatoclax. Cell culture supernatant was collected and added to the cells harvested by trypsin treatment. Cells were stained with Annexin V FITC and PI (BD Pharmigen, Franklin Lakes, NJ, USA) for 15 min at room temperature in the dark. Samples were analyzed by flow cytometry within 1 h using a Becton Dickinson LSRII System (BD Biosciences).

Allograft generation and treatment

WT female 129Sv mice that were 8–10 weeks old received injections of 6×106 D445 cells. When tumors reached a size between 100 and 250 mm3, mice were randomized to placebo, GSK1120212 treatment (1 mg/kg), Obatoclax treatment (4 mg/kg), or combination treatment groups (n=7/group). GSK1120212 was administered via oral gavage, and Obatoclax via i.p. injection once every day, and the tumor volume was calculated from two-dimensional measurements using the following equation: tumor volume=(length×width2)×0.5.

Expression profiling analysis

The dataset GSE27155 was downloaded from the GEO repository and analyzed using GenePattern (Reich et al. 2006).

Statistical analysis

Experiments were carried out at least three times. Data were analyzed using the Prism software package. Differences with P values <0.05 were considered statistically significant.

Results

A novel mouse model of poorly differentiated thyroid cancer

Mice carrying a thyroid-specific deletion of p53 and expressing the oncogenic KrasG12D allele in the thyroid epithelial cells had a median survival of 8.8 months and developed with full penetrance neoplastic lesions starting at 5 months of age (Fig. 1A). Double-mutant mice exhibited a slight decrease in the serum levels of TSH, but no significant alterations in T4 levels (Table 1). Postmortem analysis of tumor-bearing mice revealed very enlarged glands (207.8±61 vs 18±0.5 mg in WT mice) often tightly adhering to nearby anatomical structures.

Figure 1
Figure 1

KrasG12D,p53thyr−/− mice develop papillary thyroid cancer progressing to poorly differentiated and anaplastic carcinomas. (A) Cumulative tumor incidence for mice of the indicated genotypes. (B) Low-magnification (40×) view of a representative tumor showing well-differentiated (PTC), poorly differentiated (PDTC), and anaplastic (ATC) components. The inset (200×) shows a nest of PDTC with adjacent tumor necrosis (asterisk). (C) Top left, PDTC lung metastasis with central tumor necrosis and presence of organoid growth (100×); top right, Ki67 staining showing elevated proliferative index; bottom, immunostaining for thyroglobulin in PTC and PDTC components. (D) Contiguous PTC and ATC (100×); inset: PTC papillae (asterisk), mitotic figure (arrowhead), and anaplastic cells (arrow) (200×). All tumors are from 40- to 50-week-old mice.

Citation: Endocrine-Related Cancer 21, 5; 10.1530/ERC-14-0268

Table 1

Hormonal status of control and mutant mice

WTKrasG12D,p53−/−
TSH (mU/l)336.5±31.5166.7±46.8
T4 (μg/dl)2.40±0.392.94±0.36

Histopathological analysis of the tumors developed by KrasG12D,p53thyr−/− mice indicated that they resemble follicular-cell-derived thyroid carcinomas in humans. Well-differentiated papillary thyroid carcinomas (PTCs) were most prevalent in younger mice and were characterized by papillae and follicle formations, the presence of colloid, low mitotic rate, and absent tumor necrosis (Fig. 1B). Poorly differentiated carcinomas (PDTCs) coexisted with PTCs, displayed organoid growth in the form of solid nests, papillae, or follicles, and exhibited tumor necrosis (often located in the center of the tumor nests) with increased mitotic activity (Fig. 1B). These tumors were often metastatic to the lungs (28% of cases) and displayed elevated staining for the proliferation marker Ki67 (Fig. 1C). Immunostaining for the differentiation marker thyroglobulin showed diffuse labeling in the PTC areas, while the stain labeled only focally the poorly differentiated component (Fig. 1C).

In 40% of mice (six out of 15), tumors contained areas of progression to anaplastic carcinoma, growing in sheets without any organoid formation, composed of markedly pleomorphic spindle or epithelioid cells with very high mitotic activity including atypical mitosis and large areas of tumor necrosis (Fig. 1D). Anaplastic carcinomas did not show any convincing specific staining for thyroglobulin (not shown).

To further characterize the progression from PTC to PDTC and anaplastic thyroid cancer (ATC), we performed immunohistochemistry for E-cadherin and vimentin, positive and negative markers of epithelial-to-mesenchymal transition (EMT) respectively. PDTC areas showed a moderate reduction in E-cadherin staining, and no increased vimentin expression, compared with the adjacent PTCs (Fig. 2). Conversely, ATCs showed complete loss of E-cadherin staining and dramatically increased vimentin expression, confirming that these more advanced tumors undergo EMT (Fig. 2). Furthermore, all three histological components (PTC, PDTC, and ATC) of these tumors showed no immunoreactivity for active AKT (pSer473), indicating that the PI3K pathway is not active in tumors driven by Kras activation and p53 loss (Fig. 2).

Figure 2
Figure 2

Immunohistochemical detection of E-cadherin, vimentin, and pAKT in PTC, PDTC, and ATC components of the same tumor. Note the progressive loss of E-cadherin, acquisition of vimentin, and absence of activated AKT.

Citation: Endocrine-Related Cancer 21, 5; 10.1530/ERC-14-0268

A frequent aspect of human PDTC and ATC is genomic instability, leading to aneuploidy (Wreesmann et al. 2002). We analyzed metaphase spreads from early-passage primary cultures of several tumors developed by KrasG12D,p53thyr−/− mice and found that these cultures contained a distinct aneuploid population (Supplementary Figure 1, see section on supplementary data given at the end of this article).

Human PDTC is generally characterized by preserved expression of the FOXE1, PAX8, and NKX2-1 transcription factors and the thyroid-specific genes TSHR, TPO, and TG. Expression of NIS (SLC5A5) is often reduced, leading to refractoriness to radioactive iodine therapy. Complete loss of thyrocyte differentiation is instead a hallmark of human ATC. We used real-time PCR to measure the expression levels of these genes associated with thyroid differentiation and function in freshly dissected glands and tumors. Expression of Foxe1, Pax8, and Nkx2-1, as well as the thyroid-specific genes Tshr, Tpo, and Tg, was generally not altered in single mutants, while the expression levels of Nis (Slc5a5) were increased in p53−/− glands and reduced in KrasG12D glands (Fig. 3A and B). PDTCs and ATCs retained expression of the thyroid-defining transcription factors and generally showed progressive (less expression with increasing ATC component) loss of Tpo, Tg, and Nis (Slc5a5) expression (Fig. 3A and B).

Figure 3
Figure 3

Expression of thyroid specification factors (A) and differentiation/functional (B) markers in KrasG12D,p53thyr−/− primary tumors. qPCR on RNA extracted from control and tumor tissues. *P<0.05. Tumors are arranged by increasing aggressiveness/dedifferentiation: PTC/PDTC (D316 and D420), and PDTC/ATC (D87 and D155).

Citation: Endocrine-Related Cancer 21, 5; 10.1530/ERC-14-0268

MEK inhibition arrests cell growth without inducing cell death

To aid in the molecular and preclinical characterization of the tumors developed by KrasG12D,p53thyr−/− mice, we have established several cell lines from lesions of different prevailing histologies: two of these lines, D316 (PDTC/ATC original tumor) and D445 (PTC/PDTC original tumor), have been used in the experiments described below.

The notion that activation of Kras is the driver oncogenic event for thyrocyte transformation in this model leads to the question of whether these tumors are sensitive to the inhibition of MEK1 and MEK2. We used two cell lines derived from our mouse model, as well as a human ATC cell line, Cal62, which harbors the same genetic alterations and thus represents an appropriate human counterpart to the mouse lines. These cell lines were treated with the specific MEK1/2 inhibitor GSK1120212 for 72 h to establish a dose–response curve. All three cell lines were effectively growth inhibited in the low nanomolar drug range, with IC50s between 2.5 and 4.5 nM (Fig. 4A).

Figure 4
Figure 4

MEK inhibition arrests but does not kill KrasG12D,p53thyr−/− cells. (A) Effect of GSK11201212 on the viability of mouse and human KrasG12D,p53thyr−/− tumor cells. (B) Flow cytometric detection of G1 arrest in MEK-inhibitor-treated cells (dark gray fill) compared with control cells (light gray fill). (C) Flow cytometric detection of apoptotic cells in control and inhibitor-treated D316 cells. Similar results were obtained for D445. (D) Western blot analysis of cells treated with GSK11201212 for the indicated times, showing complete and persistent MEK inhibition and absence of AKT feedback activation. (E) Representative phosphoarray analysis of control and MEK-inhibitor-treated Cal62 cells. Arrows point to pERK1/2. Similar results were obtained using D316 and D445 cells.

Citation: Endocrine-Related Cancer 21, 5; 10.1530/ERC-14-0268

Growth inhibition was associated with robust cell cycle arrest in the G1 phase in Cal62 cells (Fig. 4B). Although the presence of a significant tetraploid component in the mouse cell lines partially masked this effect (because the G1-arrested tetraploid cells overlap the diploid G2 peak), G1 arrest was also apparent in both D316 and D445 (Fig. 4B).

Importantly, both direct cell counting over several days (not shown) and Annexin V staining of GSK1120212-treated cells showed complete absence of any apoptotic response upon inhibition of the MEK/MAPK pathway in both mouse and human KrasG12D,p53−/− cells (Fig. 4C).

We thus tested whether MEK inhibition might result in the activation of feedback loops restoring MAPK phosphorylation or activating compensatory PI3K-mediated pathways in order to support survival in the face of MEK inhibition. Time course analysis of mouse and human cells treated with 100 nM GSK1120212 for up to 48 h did not reveal any restoration of phosphorylated ERK1/2, and no activation of AKT was detectable at any time point (Fig. 4D). We also used an antibody array to interrogate on a larger scale the changes in the phosphokinome of both D316 and Cal62 cells upon GSK1120212 treatment. Besides the expected inactivation of ERK1/2, no significant and consistent GSK1120212-induced changes were detected in either cell lines (Fig. 4E and data not shown). These data strongly indicate that resistance to apoptosis upon inhibition of MEK/MAPK is an intrinsic trait of KrasG12D,p53−/− cells and not an adaptive response to MEK inhibition.

Krasmut thyroid tumors overexpress Bcl2a1 and Mcl1

We proposed the hypothesis that the intrinsic resistance to cell death might be associated with altered levels of pro- and anti-apoptotic members of the Bcl2 family. Thus, we measured the expression levels of eight pro-apoptotic and five anti-apoptotic genes in normal thyroid and primary thyroid tumors from KrasG12D, p53thyr−/− mice. As shown in Fig. 5, thyroid tumors with varying predominance of the three histological components (PTC, PDTC, and ATC) displayed significantly (at least twofold) elevated mRNA levels for Bax, Bak (Bak1), Bid, Bim (Bcl2l11), Noxa (Pmaip1), and Puma (Bbc3), indicating that these tumors are in fact primed to undergo apoptosis. The same trend was observed at the protein level (Fig. 5C). Strikingly, at the same time, these tumors also showed substantial upregulation, both at the RNA and protein levels, of two potent anti-apoptotic genes, Bcl2a1 and Mcl1.

Figure 5
Figure 5

Dysregulation of pro-apoptotic (A) and anti-apoptotic (B) Bcl2 family members in KrasG12D,p53thyr−/− primary tumors. qPCR was performed on RNA extracted from tumors with different prevailing histological components. (C) Western blot analysis of the expression of selected pro- and anti-apoptotic proteins in the same tumors used for the RNA analysis above.

Citation: Endocrine-Related Cancer 21, 5; 10.1530/ERC-14-0268

A Bcl2 family inhibitor restores cell death response in Krasmut tumors

To test whether the high levels of Bcl2a1 and Mcl1 protect KrasG12D,p53−/− cells from apoptosis, we used two BH3 mimetic inhibitors, ABT-263 (Tse et al. 2008) and Obatoclax (Nguyen et al. 2007). ABT-263 targets Bcl2, Bclxl, and Bclw, but not Bcl2a1 or Mcl1, while Obatoclax targets all anti-apoptotic members of the Bcl2 family. ABT-263 was modestly effective in reducing cell viability, while Obatoclax efficiently inhibited cell proliferation in all the three KrasG12D,p53−/− cell lines tested (Fig. 6A), with IC50s in the low nanomolar range. While at lower doses, Obatoclax induced, as shown previously in other cells (Urtishak et al. 2013), an accumulation of cells in S-phase without evidence of cell death (Fig. 6B and C), treatment with 500 nM Obatoclax for 48 h induced massive cell death (Fig. 6C). Thus, KrasG12D,p53−/− cells are effectively killed by BH3 mimetics that specifically bind to and inhibit Bcl2a1 and Mcl1.

Figure 6
Figure 6

Obatoclax induces cell death in KrasG12D,p53thyr−/− cells and synergizes with MEK inhibition. (A) Effect of ABT-263 and Obatoclax on the viability of tumor cells. (B) Flow cytometric cell cycle analysis of D316 cells treated with 100 and 500 nM Obatoclax. (C) Annexin V staining showing extensive cell death upon treatment of D316 cells with 500 nM Obatoclax. Experiments were replicated with similar results in D445 cells. (D) Dose–response curves showing the cooperative effect of a GSK11201212/Obatoclax combination on the viability of the mouse cell line D445. Drugs were combined at a fixed ratio of 1:5 (GSK/Oba). CIED95: combination index at a drug dose causing 95% reduction in viability. Experiments were replicated with similar results in D316 cells. (E and F) qPCR analysis of the expression of pro- and anti-apoptotic Bcl2 family members in D316 cells upon 6 h treatment with Obatoclax, GSK1120212, and both in combination. Experiments were replicated with similar results in D445 cells.

Citation: Endocrine-Related Cancer 21, 5; 10.1530/ERC-14-0268

Then, we tested whether Obatoclax would synergize with inhibition of the RAS/MEK/ERK signaling cascade, which drives these tumors. We compared the single-agent dose–response curves in the mouse cell lines with the curve obtained using the two drugs combined at a fixed ratio based on the most effective single-dose concentration for each drug. The combination treatment was more effective than each single agent over a wide range of concentrations (Fig. 6D). In order to establish whether the combined effects of GSK1120212 and Obatoclax were synergistic rather than additive, we calculated the CI according to the well-established Chou and Talalay median-effect method (Chou & Talalay 1984). The CI for the ED95 indicated clearly synergistic effects (CI=0.47).

In order to define the molecular basis for the synergy between MEK1/2 and Bcl2 family inhibition, we determined the changes in expression levels of pro- and anti-apoptotic molecules upon exposure to the inhibitors. GSK1120212 and Obatoclax treatment, alone and in combination, increased the levels of several pro-apoptotic Bcl2 family members (Bax, Noxa, and Bim), thus predisposing the tumor cells to death (Fig. 6E). At the same time, GSK1120212 strongly reduced the expression of Bcl2a1, alone and in combination with Obatoclax (Fig. 6F). On the other hand, Obatoclax, as a single agent, increased the expression of most anti-apoptotic Bcl2 family members, probably through a compensatory feedback mechanism. Thus, the synergy between the two compounds is probably obtained through simultaneous upregulation of pro-apoptotic genes, and downregulation and inhibition of anti-apoptotic genes.

The ability of Obatoclax to induce cell death in therapy-resistant cells prompted us to test whether inhibition of Bcl2a1 and Mcl1 would sensitize these cells to doxorubicin treatment, which is a current, although poorly effective, standard in advanced thyroid cancer chemotherapy (Shimaoka et al. 1985). Indeed, both mouse and human KrasG12D,p53−/− cell lines exhibited significantly increased cell death when treated with a combination of GSK1120212 and Obatoclax, than when treated with each compound alone (Fig. 7).

Figure 7
Figure 7

Obatoclax sensitizes KrasG12D,p53thyr−/− cells to doxorubicin treatment. The graph shows the number of remaining cells 2 days after treatment with each drug and both in combination, relative to the number of cells plated at the beginning of the experiment.

Citation: Endocrine-Related Cancer 21, 5; 10.1530/ERC-14-0268

Finally, we used an immunocompetent in vivo system, i.e. 129Sv mice bearing D445 allograft tumors, to test the efficacy of Obatoclax as a sensitizer to cell death. Strikingly, the combination of Obatoclax (4 mg/kg) and GSK1120212 (1 mg/kg) resulted in a significant degree of tumor control compared with each drug alone (Fig. 8).

Figure 8
Figure 8

Obatoclax and GSK1120212 activity in a D445 allograft tumor model. Tumor volume was determined every 3 days. Asterisks indicate significant differences between combination and single treatments (P<0.05).

Citation: Endocrine-Related Cancer 21, 5; 10.1530/ERC-14-0268

Dysregulation of Bcl2 family members is a feature of human aggressive thyroid carcinomas

We used a large expression-profiling dataset (99 samples) available from the GEO repository (Giordano et al. 2005) to test whether the dysregulation of the expression levels of Bcl2 family members observed in our mouse model recapitulates specific human thyroid cancer subtypes. Strikingly, overexpression of MCL1, BCLXL (BCL2L1), and BID, and downregulation of BCL2, BAD, and BIK (as observed in the PTCs, PDTCs, and ATCs developed by KrasG12D,p53−/− mice) were able to separate classical papillary, tall-cell-variant papillary, and anaplastic carcinomas from all the other subtypes present in the dataset. Additional upregulation of BCL2A1 and NOXA identified a smaller subset of thyroid carcinomas that included all the anaplastic tumors present in the dataset. These results strongly indicate that dysregulation of a set of Bcl2 family members, including MCL1 and BCL2A1, identifies thyroid carcinomas with particularly aggressive features (Fig. 9).

Figure 9
Figure 9

Heat map showing that different subtypes of human thyroid cancer cluster together based on the expression of BCL2 family members. Genes in green are downregulated and genes in red are upregulated in the mouse model. The red bar below the heat map indicates tumors overexpressing MCL1, and the blue bar indicates tumors overexpressing both MCL1 and BCL2A1, as in the mouse model.

Citation: Endocrine-Related Cancer 21, 5; 10.1530/ERC-14-0268

Discussion

As for other tumor types, genetically defined murine models of thyroid cancer are invaluable tools to dissect the molecular pathways and events leading to neoplastic transformation and tumor progression, and to test novel, targeted therapeutic approaches in a physiological, immunocompetent system. The past few years have thus seen the generation and exploitation of highly relevant mouse models for follicular (Suzuki et al. 2002, Antico-Arciuch et al. 2010, Pringle et al. 2014), papillary (Jhiang et al. 1996, Knauf et al. 2005, Chakravarty et al. 2011), and ATC (Antico Arciuch et al. 2011, Charles et al. 2014, McFadden et al. 2014). Areas of poor differentiation were observed in advanced papillary tumors developed by BRAFV600E mice (Knauf et al. 2011). Herein, we report the generation and characterization of a RAS-driven model that shows progression from papillary lesions to PDTC and, in several cases, to ATC. This model faithfully recapitulates human PDTC, both in terms of morphological and immunohistochemical features, and in terms of genomic instability and tumor aggressiveness. As such, this model represents a valuable tool for identifying key signaling nodes that may allow therapeutic exploitation.

As expected for a model driven by activation of the RAS/MEK/ERK cascade, a MEK inhibitor was able to interfere with tumor cell proliferation. Interestingly, such inhibition was not accompanied by any of the compensatory feedback responses that have been observed in other systems and that can reactivate signaling and proliferation, in particular in KRAS mutant tumors (Yoon et al. 2010, Turke et al. 2012, Carlino et al. 2014). However, despite prolonged inhibition of MAPK activity and arrest in G1 phase, tumor cells derived from our mouse model did not undergo cell death, indicating that this system is inherently resistant to apoptosis, similar to the situation observed in human aggressive thyroid tumors, for which cytotoxic therapy often fails. Our data show that the inability to undergo apoptosis, in spite of high levels of expression of pro-apoptotic effectors (Bak, Bid, Noxa, and Puma), is at least in part associated with the elevated expression levels of two anti-apoptotic members of the Bcl2 family, Bcl2a1 and Mcl1. BCL2A1 has been recently described as an oncogene responsible for hematopoietic stem cell transformation (Metais et al. 2012) and for resistance to BRAF inhibition in melanoma (Haq et al. 2013). The role of MCL1 in cancer is well established (Ertel et al. 2013), and elevated levels of MCL1 have been reported to correlate with dedifferentiation in thyroid cancer (Branet et al. 1996). Accordingly, the pan-Bcl2 family inhibitor Obatoclax, but not the more restricted inhibitor ABT-263, which does not target Bcl2a1 and Mcl1, was able to induce tumor cell death even as a single agent, and to cooperate with MEK inhibition both in cultured cells and, in vivo, in tumor cell allografts, as well as with doxorubicin treatment. Thus, targeted inhibition of anti-apoptotic proteins appears to be a valid strategy to sensitize otherwise resistant thyroid cancer cells to cell death induced by inhibition of the driver pathway or cytotoxic chemotherapy. In view of the possible ability of Obatoclax to induce different modes of cell death, including apoptosis, necroptosis, and autophagic death (Heidari et al. 2010, Basit et al. 2013, Urtishak et al. 2013, Yu & Liu 2013), future studies are required to dissect in depth the specific route through which RAS-mutant thyroid cells undergo cell death upon inhibition of anti-apoptotic Bcl2 family members, so that additional fine tuning of the death-inducing system can be achieved for optimal therapeutic responses.

Finally, we have shown that human thyroid cancer subtypes cluster in two main groups based on the expression of BCL2 family members: follicular, follicular variant, and oncocytic carcinomas express high levels of BCL2 and BCLW (BCL2L2), and thus may benefit from the sensitizing action of inhibitors such as ABT-263. Classic papillary, as well as the more aggressive tall-cell and anaplastic carcinomas, overexpress MCL1, BCLXL, and BCL2A1, and thus are more sensitive to the inhibitory activity of Obatoclax and its derivatives.

Thus, using a novel mouse model of progression from PTC to PDTC and ATC, as well as a genetically corresponding human cell line, we have uncovered a novel strategy to target thyroid cancer cells for death.

Supplementary data

This is linked to the online version of the paper at http://dx.doi.org/10.1530/ERC-14-0268.

Declaration of interest

The authors declare that there is no conflict of interest that could be perceived as prejudicing the impartiality of the research reported.

Funding

This work was supported by the Albert Einstein Cancer Center Core Grant, and by National Institutes of Health (NIH) grants CA128943 and CA167839 to A Di Cristofano. S Refetoff was supported by grant R37DK15070 from the NIH and the Seymour Abrams fund. A Di Cristofano was a recipient of the Irma T. Hirschl Career Scientist Award.

References

  • American Thyroid Association (ATA) Guidelines Taskforce on Thyroid Nodules and Differentiated Thyroid Cancer , Cooper DS, Doherty GM, Haugen BR, Kloos RT, Lee SL, Mandel SJ, Mazzaferri EL, McIver B & Pacini F 2009 Revised American Thyroid Association management guidelines for patients with thyroid nodules and differentiated thyroid cancer. Thyroid 19 11671214. (doi:10.1089/thy.2009.0110).

    • Search Google Scholar
    • Export Citation
  • Antico Arciuch VG, Dima M, Liao XH, Refetoff S & Di Cristofano A 2010 Cross-talk between PI3K and estrogen in the mouse thyroid predisposes to the development of follicular carcinomas with a higher incidence in females. Oncogene 29 56785686. (doi:10.1038/onc.2010.308).

    • Search Google Scholar
    • Export Citation
  • Antico Arciuch VG, Russo MA, Dima M, Kang KS, Dasrath F, Liao XH, Refetoff S, Montagna C & Di Cristofano A 2011 Thyrocyte-specific inactivation of p53 and Pten results in anaplastic thyroid carcinomas faithfully recapitulating human tumors. Oncotarget 2 11091126.

    • Search Google Scholar
    • Export Citation
  • Basit F, Cristofanon S & Fulda S 2013 Obatoclax (GX15-070) triggers necroptosis by promoting the assembly of the necrosome on autophagosomal membranes. Cell Death and Differentiation 20 11611173. (doi:10.1038/cdd.2013.45).

    • Search Google Scholar
    • Export Citation
  • Branet F, Brousset P, Krajewski S, Schlaifer D, Selves J, Reed JC & Caron P 1996 Expression of the cell death-inducing gene bax in carcinomas developed from the follicular cells of the thyroid gland. Journal of Clinical Endocrinology and Metabolism 81 27262730. (doi:10.1210/jcem.81.7.8675602).

    • Search Google Scholar
    • Export Citation
  • Carlino MS, Todd JR, Gowrishankar K, Mijatov B, Pupo GM, Fung C, Snoyman S, Hersey P, Long GV & Kefford RF 2014 Differential activity of MEK and ERK inhibitors in BRAF inhibitor resistant melanoma. Molecular Oncology 8 544554. (doi:10.1016/j.molonc.2014.01.003).

    • Search Google Scholar
    • Export Citation
  • Chakravarty D, Santos E, Ryder M, Knauf JA, Liao XH, West BL, Bollag G, Kolesnick R, Thin TH & Rosen N 2011 Small-molecule MAPK inhibitors restore radioiodine incorporation in mouse thyroid cancers with conditional BRAF activation. Journal of Clinical Investigation 121 47004711. (doi:10.1172/JCI46382).

    • Search Google Scholar
    • Export Citation
  • Charles RP, Silva J, Iezza G, Phillips WA & McMahon M 2014 Activating BRAF and PIK3CA mutations cooperate to promote anaplastic thyroid carcinogenesis. Molecular Cancer Research 12 979986. (doi:10.1158/1541-7786.MCR-14-0158-T).

    • Search Google Scholar
    • Export Citation
  • Chou TC & Talalay P 1984 Quantitative analysis of dose–effect relationships: the combined effects of multiple drugs or enzyme inhibitors. Advances in Enzyme Regulation 22 2755. (doi:10.1016/0065-2571(84)90007-4).

    • Search Google Scholar
    • Export Citation
  • Ertel F, Nguyen M, Roulston A & Shore GC 2013 Programming cancer cells for high expression levels of Mcl1. EMBO Reports 14 328336. (doi:10.1038/embor.2013.20).

    • Search Google Scholar
    • Export Citation
  • Garcia-Rostan G, Zhao H, Camp RL, Pollan M, Herrero A, Pardo J, Wu R, Carcangiu ML, Costa J & Tallini G 2003 ras mutations are associated with aggressive tumor phenotypes and poor prognosis in thyroid cancer. Journal of Clinical Oncology 21 32263235. (doi:10.1200/JCO.2003.10.130).

    • Search Google Scholar
    • Export Citation
  • Giordano TJ, Kuick R, Thomas DG, Misek DE, Vinco M, Sanders D, Zhu Z, Ciampi R, Roh M & Shedden K 2005 Molecular classification of papillary thyroid carcinoma: distinct BRAF, RAS, and RET/PTC mutation-specific gene expression profiles discovered by DNA microarray analysis. Oncogene 24 66466656. (doi:10.1038/sj.onc.1208822).

    • Search Google Scholar
    • Export Citation
  • Haq R, Yokoyama S, Hawryluk EB, Jonsson GB, Frederick DT, McHenry K, Porter D, Tran TN, Love KT & Langer R 2013 BCL2A1 is a lineage-specific antiapoptotic melanoma oncogene that confers resistance to BRAF inhibition. PNAS 110 43214326. (doi:10.1073/pnas.1205575110).

    • Search Google Scholar
    • Export Citation
  • Heidari N, Hicks MA & Harada H 2010 GX15-070 (obatoclax) overcomes glucocorticoid resistance in acute lymphoblastic leukemia through induction of apoptosis and autophagy. Cell Death & Disease 1 e76. (doi:10.1038/cddis.2010.53).

    • Search Google Scholar
    • Export Citation
  • Ibrahimpasic T, Ghossein RM, Carlson DL, Chernichenko N, Nixon I, Palmer FL, Lee NY, Shaha AR, Patel SG & Tuttle RM 2013 Poorly differentiated thyroid carcinoma presenting with gross extrathyroidal extension: 1986–2009 Memorial Sloan-Kettering Cancer Center experience. Thyroid 23 9971002. (doi:10.1089/thy.2012.0403).

    • Search Google Scholar
    • Export Citation
  • Jhiang SM, Sagartz JE, Tong Q, Parker-Thornburg J, Capen CC, Cho JY, Xing S & Ledent C 1996 Targeted expression of the ret/PTC1 oncogene induces papillary thyroid carcinomas. Endocrinology 137 375378. (doi:10.1210/endo.137.1.8536638).

    • Search Google Scholar
    • Export Citation
  • Jonkers J, Meuwissen R, van der Gulden H, Peterse H, van der Valk M & Berns A 2001 Synergistic tumor suppressor activity of BRCA2 and p53 in a conditional mouse model for breast cancer. Nature Genetics 29 418425. (doi:10.1038/ng747).

    • Search Google Scholar
    • Export Citation
  • Knauf JA, Ma X, Smith EP, Zhang L, Mitsutake N, Liao XH, Refetoff S, Nikiforov YE & Fagin JA 2005 Targeted expression of BRAFV600E in thyroid cells of transgenic mice results in papillary thyroid cancers that undergo dedifferentiation. Cancer Research 65 42384245. (doi:10.1158/0008-5472.CAN-05-0047).

    • Search Google Scholar
    • Export Citation
  • Knauf JA, Sartor MA, Medvedovic M, Lundsmith E, Ryder M, Salzano M, Nikiforov YE, Giordano TJ, Ghossein RA & Fagin JA 2011 Progression of BRAF-induced thyroid cancer is associated with epithelial–mesenchymal transition requiring concomitant MAP kinase and TGFβ signaling. Oncogene 30 31533162. (doi:10.1038/onc.2011.44).

    • Search Google Scholar
    • Export Citation
  • Kusakabe T, Kawaguchi A, Kawaguchi R, Feigenbaum L & Kimura S 2004 Thyrocyte-specific expression of Cre recombinase in transgenic mice. Genesis 39 212216. (doi:10.1002/gene.20043).

    • Search Google Scholar
    • Export Citation
  • McFadden DG, Vernon A, Santiago PM, Martinez-McFaline R, Bhutkar A, Crowley DM, McMahon M, Sadow PM & Jacks T 2014 p53 constrains progression to anaplastic thyroid carcinoma in a Braf-mutant mouse model of papillary thyroid cancer. PNAS 111 E1600E1609. (doi:10.1073/pnas.1404357111).

    • Search Google Scholar
    • Export Citation
  • Metais JY, Winkler T, Geyer JT, Calado RT, Aplan PD, Eckhaus MA & Dunbar CE 2012 BCL2A1a over-expression in murine hematopoietic stem and progenitor cells decreases apoptosis and results in hematopoietic transformation. PLoS ONE 7 e48267. (doi:10.1371/journal.pone.0048267).

    • Search Google Scholar
    • Export Citation
  • Nguyen M, Marcellus RC, Roulston A, Watson M, Serfass L, Murthy Madiraju SR, Goulet D, Viallet J, Belec L & Billot X 2007 Small molecule obatoclax (GX15-070) antagonizes MCL-1 and overcomes MCL-1-mediated resistance to apoptosis. PNAS 104 1951219517. (doi:10.1073/pnas.0709443104).

    • Search Google Scholar
    • Export Citation
  • Nikiforov YE 2004 Genetic alterations involved in the transition from well-differentiated to poorly differentiated and anaplastic thyroid carcinomas. Endocrine Pathology 15 319327. (doi:10.1385/EP:15:4:319).

    • Search Google Scholar
    • Export Citation
  • Pohlenz J, Maqueem A, Cua K, Weiss RE, Van Sande J & Refetoff S 1999 Improved radioimmunoassay for measurement of mouse thyrotropin in serum: strain differences in thyrotropin concentration and thyrotroph sensitivity to thyroid hormone. Thyroid 9 12651271. (doi:10.1089/thy.1999.9.1265).

    • Search Google Scholar
    • Export Citation
  • Pringle DR, Vasko VV, Yu L, Manchanda PK, Lee AA, Zhang X, Kirschner JM, Parlow AF, Saji M & Jarjoura D 2014 Follicular thyroid cancers demonstrate dual activation of PKA and mTOR as modeled by thyroid-specific deletion of Prkar1a and Pten in mice. Journal of Clinical Endocrinology and Metabolism 99 E804E812. (doi:10.1210/jc.2013-3101).

    • Search Google Scholar
    • Export Citation
  • Reich M, Liefeld T, Gould J, Lerner J, Tamayo P & Mesirov JP 2006 GenePattern 2.0. Nature Genetics 38 500501. (doi:10.1038/ng0506-500).

  • Ricarte-Filho JC, Ryder M, Chitale DA, Rivera M, Heguy A, Ladanyi M, Janakiraman M, Solit D, Knauf JA & Tuttle RM 2009 Mutational profile of advanced primary and metastatic radioactive iodine-refractory thyroid cancers reveals distinct pathogenetic roles for BRAF, PIK3CA, and AKT1. Cancer Research 69 48854893. (doi:10.1158/0008-5472.CAN-09-0727).

    • Search Google Scholar
    • Export Citation
  • Shimaoka K, Schoenfeld DA, DeWys WD, Creech RH & DeConti R 1985 A randomized trial of doxorubicin versus doxorubicin plus cisplatin in patients with advanced thyroid carcinoma. Cancer 56 21552160. (doi:10.1002/1097-0142(19851101)56:9<2155::AID-CNCR2820560903>3.0.CO;2-E).

    • Search Google Scholar
    • Export Citation
  • Siegel R, Naishadham D & Jemal A 2013 Cancer statistics, 2013. CA: A Cancer Journal for Clinicians 63 1130. (doi:10.3322/caac.21166).

  • Siironen P, Hagstrom J, Maenpaa HO, Louhimo J, Heikkila A, Heiskanen I, Arola J & Haglund C 2010 Anaplastic and poorly differentiated thyroid carcinoma: therapeutic strategies and treatment outcome of 52 consecutive patients. Oncology 79 400408. (doi:10.1159/000322640).

    • Search Google Scholar
    • Export Citation
  • Suzuki H, Willingham MC & Cheng SY 2002 Mice with a mutation in the thyroid hormone receptor β gene spontaneously develop thyroid carcinoma: a mouse model of thyroid carcinogenesis. Thyroid 12 963969. (doi:10.1089/105072502320908295).

    • Search Google Scholar
    • Export Citation
  • Tse C, Shoemaker AR, Adickes J, Anderson MG, Chen J, Jin S, Johnson EF, Marsh KC, Mitten MJ & Nimmer P 2008 ABT-263: a potent and orally bioavailable Bcl-2 family inhibitor. Cancer Research 68 34213428. (doi:10.1158/0008-5472.CAN-07-5836).

    • Search Google Scholar
    • Export Citation
  • Turke AB, Song Y, Costa C, Cook R, Arteaga CL, Asara JM & Engelman JA 2012 MEK inhibition leads to PI3K/AKT activation by relieving a negative feedback on ERBB receptors. Cancer Research 72 32283237. (doi:10.1158/0008-5472.CAN-11-3747).

    • Search Google Scholar
    • Export Citation
  • Tuveson DA, Shaw AT, Willis NA, Silver DP, Jackson EL, Chang S, Mercer KL, Grochow R, Hock H & Crowley D 2004 Endogenous oncogenic K-ras G12D stimulates proliferation and widespread neoplastic and developmental defects. Cancer Cell 5 375387. (doi:10.1016/S1535-6108(04)00085-6).

    • Search Google Scholar
    • Export Citation
  • Urtishak KA, Edwards AY, Wang LS, Hudome A, Robinson BW, Barrett JS, Cao K, Cory L, Moore JS & Bantly AD 2013 Potent obatoclax cytotoxicity and activation of triple death mode killing across infant acute lymphoblastic leukemia. Blood 121 26892703. (doi:10.1182/blood-2012-04-425033).

    • Search Google Scholar
    • Export Citation
  • Volante M, Collini P, Nikiforov YE, Sakamoto A, Kakudo K, Katoh R, Lloyd RV, LiVolsi VA, Papotti M & Sobrinho-Simoes M 2007 Poorly differentiated thyroid carcinoma: the Turin proposal for the use of uniform diagnostic criteria and an algorithmic diagnostic approach. American Journal of Surgical Pathology 31 12561264. (doi:10.1097/PAS.0b013e3180309e6a).

    • Search Google Scholar
    • Export Citation
  • Volante M, Rapa I, Gandhi M, Bussolati G, Giachino D, Papotti M & Nikiforov YE 2009 RAS mutations are the predominant molecular alteration in poorly differentiated thyroid carcinomas and bear prognostic impact. Journal of Clinical Endocrinology and Metabolism 94 47354741. (doi:10.1210/jc.2009-1233).

    • Search Google Scholar
    • Export Citation
  • Wreesmann VB, Ghossein RA, Patel SG, Harris CP, Schnaser EA, Shaha AR, Tuttle RM, Shah JP, Rao PH & Singh B 2002 Genome-wide appraisal of thyroid cancer progression. American Journal of Pathology 161 15491556. (doi:10.1016/S0002-9440(10)64433-1).

    • Search Google Scholar
    • Export Citation
  • Yoon YK, Kim HP, Han SW, Oh do Y, Im SA, Bang YJ & Kim TY 2010 KRAS mutant lung cancer cells are differentially responsive to MEK inhibitor due to AKT or STAT3 activation: implication for combinatorial approach. Molecular Carcinogenesis 49 353362. (doi:10.1002/mc.20607).

    • Search Google Scholar
    • Export Citation
  • Yu L & Liu S 2013 Autophagy contributes to modulating the cytotoxicities of Bcl-2 homology domain-3 mimetics. Seminars in Cancer Biology 23 553560. (doi:10.1016/j.semcancer.2013.08.008).

    • Search Google Scholar
    • Export Citation

If the inline PDF is not rendering correctly, you can download the PDF file here.

 

Society for Endocrinology

Sept 2018 onwards Past Year Past 30 Days
Abstract Views 0 0 0
Full Text Views 875 86 5
PDF Downloads 269 50 4
  • View in gallery

    KrasG12D,p53thyr−/− mice develop papillary thyroid cancer progressing to poorly differentiated and anaplastic carcinomas. (A) Cumulative tumor incidence for mice of the indicated genotypes. (B) Low-magnification (40×) view of a representative tumor showing well-differentiated (PTC), poorly differentiated (PDTC), and anaplastic (ATC) components. The inset (200×) shows a nest of PDTC with adjacent tumor necrosis (asterisk). (C) Top left, PDTC lung metastasis with central tumor necrosis and presence of organoid growth (100×); top right, Ki67 staining showing elevated proliferative index; bottom, immunostaining for thyroglobulin in PTC and PDTC components. (D) Contiguous PTC and ATC (100×); inset: PTC papillae (asterisk), mitotic figure (arrowhead), and anaplastic cells (arrow) (200×). All tumors are from 40- to 50-week-old mice.

  • View in gallery

    Immunohistochemical detection of E-cadherin, vimentin, and pAKT in PTC, PDTC, and ATC components of the same tumor. Note the progressive loss of E-cadherin, acquisition of vimentin, and absence of activated AKT.

  • View in gallery

    Expression of thyroid specification factors (A) and differentiation/functional (B) markers in KrasG12D,p53thyr−/− primary tumors. qPCR on RNA extracted from control and tumor tissues. *P<0.05. Tumors are arranged by increasing aggressiveness/dedifferentiation: PTC/PDTC (D316 and D420), and PDTC/ATC (D87 and D155).

  • View in gallery

    MEK inhibition arrests but does not kill KrasG12D,p53thyr−/− cells. (A) Effect of GSK11201212 on the viability of mouse and human KrasG12D,p53thyr−/− tumor cells. (B) Flow cytometric detection of G1 arrest in MEK-inhibitor-treated cells (dark gray fill) compared with control cells (light gray fill). (C) Flow cytometric detection of apoptotic cells in control and inhibitor-treated D316 cells. Similar results were obtained for D445. (D) Western blot analysis of cells treated with GSK11201212 for the indicated times, showing complete and persistent MEK inhibition and absence of AKT feedback activation. (E) Representative phosphoarray analysis of control and MEK-inhibitor-treated Cal62 cells. Arrows point to pERK1/2. Similar results were obtained using D316 and D445 cells.

  • View in gallery

    Dysregulation of pro-apoptotic (A) and anti-apoptotic (B) Bcl2 family members in KrasG12D,p53thyr−/− primary tumors. qPCR was performed on RNA extracted from tumors with different prevailing histological components. (C) Western blot analysis of the expression of selected pro- and anti-apoptotic proteins in the same tumors used for the RNA analysis above.

  • View in gallery

    Obatoclax induces cell death in KrasG12D,p53thyr−/− cells and synergizes with MEK inhibition. (A) Effect of ABT-263 and Obatoclax on the viability of tumor cells. (B) Flow cytometric cell cycle analysis of D316 cells treated with 100 and 500 nM Obatoclax. (C) Annexin V staining showing extensive cell death upon treatment of D316 cells with 500 nM Obatoclax. Experiments were replicated with similar results in D445 cells. (D) Dose–response curves showing the cooperative effect of a GSK11201212/Obatoclax combination on the viability of the mouse cell line D445. Drugs were combined at a fixed ratio of 1:5 (GSK/Oba). CIED95: combination index at a drug dose causing 95% reduction in viability. Experiments were replicated with similar results in D316 cells. (E and F) qPCR analysis of the expression of pro- and anti-apoptotic Bcl2 family members in D316 cells upon 6 h treatment with Obatoclax, GSK1120212, and both in combination. Experiments were replicated with similar results in D445 cells.

  • View in gallery

    Obatoclax sensitizes KrasG12D,p53thyr−/− cells to doxorubicin treatment. The graph shows the number of remaining cells 2 days after treatment with each drug and both in combination, relative to the number of cells plated at the beginning of the experiment.

  • View in gallery

    Obatoclax and GSK1120212 activity in a D445 allograft tumor model. Tumor volume was determined every 3 days. Asterisks indicate significant differences between combination and single treatments (P<0.05).

  • View in gallery

    Heat map showing that different subtypes of human thyroid cancer cluster together based on the expression of BCL2 family members. Genes in green are downregulated and genes in red are upregulated in the mouse model. The red bar below the heat map indicates tumors overexpressing MCL1, and the blue bar indicates tumors overexpressing both MCL1 and BCL2A1, as in the mouse model.