Role of Krüppel-Like Factor 4 in the Maintenance of Chemoresistance of Anaplastic Thyroid Cancer

Abstract

Background:

Anaplastic thyroid cancer (ATC) has a very poor prognosis due to its aggressive nature and resistance to conventional treatment. Radiotherapy and chemotherapy are not fully effective because of the undifferentiated phenotype and enhanced drug resistance of ATC. The objective of this study was to evaluate the involvement of Krüppel-like factor 4 (KLF4), a stemness-associated transcription factor, in the undifferentiated phenotype and drug resistance of ATC.

Methods:

ATC cells were compared to papillary thyroid cancer cells in drug resistance and gene expression. The effects of KLF4 knockdown in ATC cells on in vitro and in vivo drug resistance were measured. The effects of KLF4 overexpression and knockdown on ABC transporter activity were determined.

Results:

ATC cells, such as HTH83, 8505C, and SW1736, exhibited higher resistance to the anticancer drug paclitaxel and higher expression of KLF4 than TPC-1 papillary thyroid cancer cells. Knockdown of KLF4 expression in ATC cells increased the expression of the thyroid-specific differentiation genes, such as thyrotropin receptor, thyroid peroxidase, thyroglobulin, and sodium–iodide symporter. Knockdown of KLF4 expression in ATC cells decreased the resistance to doxorubicin and paclitaxel, and reduced ABC transporter expression. Luciferase reporter assay results showed that KLF4 overexpression increased ABCG2 promoter activity, which was abolished by KLF4 knockdown. A tumorigenicity assay showed that the combination of paclitaxel treatment and KLF4 knockdown significantly decreased tumor mass originated from HTH83 cells in mice.

Conclusions:

ATC cells show high expression of KLF4, and KLF4 expression is necessary for maintaining the undifferentiated phenotype and drug resistance in vitro and in vivo. The present study identifies KLF4 as a potential therapeutic target for eliminating ATC cells.

Introduction

Thyroid cancer is the most common endocrine tumor. Most thyroid cancers are well-differentiated thyroid carcinomas. However, 2% of all thyroid cancer cases is undifferentiated/anaplastic thyroid cancer (ATC), an aggressive and lethal malignancy that occurs in one to two individuals per million every year in the United States. ATC patients have an average median survival of five months, and <20% live for a year after diagnosis (1,2). Current treatments for ATC are multimodal, which includes surgery, chemotherapy, and radiation treatment. Nevertheless, ATC is resistant to all conventional treatments, and the prognosis has remained unchanged, presenting a major diagnostic and therapeutic challenge (3). ATC cells usually lose the expression of thyroid-specific genes, including sodium–iodide symporter (NIS), thyroid peroxidase (TPO), thyrotropin receptor (TSHR), and thyroglobulin (TG), which results in a loss of the capacity for accumulating and organifying iodide, thus rendering radioactive iodine treatment impossible. Therefore, re-induction of a differentiated phenotype would be a means of treating ATC.

ATC exhibits resistance to conventional anticancer drugs, such as doxorubicin, cisplatin, and paclitaxel, possibly because the cells express high levels of multidrug resistance transporters of the ABC gene family, including ABCG2 (4). Intriguingly, doxorubicin failed to eradicate cancer stem cells (CSCs) derived from an ATC cell line, and ABCG2 was mainly responsible for drug resistance, as silencing of ABCG2 increased cell death in response to both cisplatin and doxorubicin (5). Therefore, downregulation of ABC transporters serves as a cornerstone in developing a cure for poorly differentiated ATC. However, the molecular mechanism involved in the elevated expression of ABC transporters in ATCs is still elusive.

The epithelial zinc-finger transcription factor Krüppel-like factor 4 (KLF4) is an important regulator of cellular proliferation and differentiation (6). KLF4 is highly expressed in embryonic stem cells, and the forced expression of KLF4 promoted self-renewal of stem cells (7,8). Overexpression of KLF4 with three other transcription factors, OCT4, c-MYC, and SOX2, was capable of reprogramming fibroblasts into induced pluripotent stem cells that are phenotypically similar to embryonic stem cells, suggesting that KLF4 is an important factor for the maintenance of stem cells (9,10). Recent studies have demonstrated that KLF4 can act as an oncogene or tumor suppressor, depending on tumor and tissue type (11). It has been reported that KLF4 regulates tumor-initiating cells or CSCs (12), and several studies have provided evidence that CSCs are involved in initiating and sustaining ATC growth (13 –17). Furthermore, overexpression of KLF4 has been reported in ATC specimens (5). However, the role of KLF4 in maintaining the undifferentiated status and drug resistance of ATC has not been investigated.

The present study examined whether KLF4 is involved in the regulation of the undifferentiated phenotype and drug resistance of ATC. The results suggest that KLF4 could be an important therapeutic target for treatment of ATC patients.

Materials and Methods

Materials

Fetal bovine serum (FBS), trypsin, penicillin, streptomycin, 0.5% trypsin/EDTA solution, and Lipofectamine reagent were purchased from Thermo Fisher Scientific (Waltham, MA). Culture plates were purchased from Nunc (Roskilde, Denmark). Anti-NIS polyclonal antibody (SC-134515) and anti-TSHR monoclonal antibody (SC-515556) were purchased from Santa Cruz Biotechnology (Dallas, TX). An anti-KLF4 polyclonal antibody (AB151733) and a horseradish peroxidase-conjugated secondary antibody (AB6802) were purchased from Abcam (Cambridge, United Kingdom). An anti-GAPDH monoclonal antibody (MAB374) was purchased from EMD Millipore (Billerica, MA). Peroxidase-labeled secondary antibodies and the enhanced chemiluminescence Western blotting system were purchased from GE Healthcare Life Science (Pittsburgh, PA). TRI Reagent and other unlisted reagents were purchased from Sigma–Aldrich (St. Louis, MO).

Cell culture

The human thyroid follicular epithelial cell line Nthy-ori 3-1, which was immortalized by transfection with a plasmid encoding the SV40 large T antigene, was obtained from European Collection of Authenticated Cell Cultures (Salisbury, United Kingdom). 8505C cells were obtained from DSMZ GmbH (Braunschweig, Germany) and SW1736 cells were obtained from Cell Lines Service (www.clsgmbh.de). ATC cell lines were cultured in RPMI 1640 medium supplemented with 10% FBS, and TPC-1 and Nthy-ori 3-1 cells were cultured in Dulbecco's modified Eagle's medium (DMEM) with 10% FBS. Cells were maintained with culture medium supplemented with 10% FBS, 10 IU/mL of penicillin, and 100 μg/mL of streptomycin, and maintained at 37°C in 5% CO₂. For primary culture of normal thyroid epithelial cells, human thyroid tissue was obtained as surgical waste from patients undergoing thyroidectomies. Written informed consent was obtained from donors, and the protocol was approved by the Institutional Review Board of Pusan National University Hospital. The thyroid specimens were dissociated with 1.25 mg/mL of collagenase type I (Sigma–Aldrich) and 10 mg/mL of dispase I (Life Technologies, Carlsbad, CA) and washed three times, and follicles were seeded in T-25 culture plates cultured in RPMI 1640 medium with 10% FBS, 10 μg/mL of insulin (Sigma–Aldrich), 5 mIU/mL of thyrotropin (TSH; Sigma–Aldrich), 2.5 μg/mL of amphotericin B (Sigma–Aldrich), 10 IU/mL of penicillin, and 100 μg/mL of streptomycin at 37°C in 5% CO₂ for maintenance of follicular thyroid cells.

Western blotting

Cells were washed with ice-cold phosphate-buffered saline (PBS) and then lysed in lysis buffer (20 mM of Tris-HCl, 1 mM of EGTA, 1 mM of EDTA, 10 mM of NaCl, 0.1 mM of phenylmethyl sulfonyl fluoride, 1 mM of Na₃VO₄, 30 mM of sodium pyrophosphate, 25 mM of β-glycerol phosphate, 1% Triton X-100, pH 7.4). Cell lysates were separated by sodium dodecyl sulfate-polyacrylamide gel electrophoresis, transferred onto nitrocellulose membranes, and stained with 0.1% Ponceau S solution (Sigma–Aldrich). After blocking with 5% nonfat milk, the membranes were immunoblotted with antibodies against KLF4, NIS, THSR, or GAPDH (the gel loading control) for 2 h, and the bound antibodies were visualized with horseradish peroxidase-conjugated secondary antibodies using the enhanced chemiluminescence Western blotting system. Densitometry was performed with ImageJ software to quantify the protein band intensities, and relative protein band intensities were normalized to GAPDH.

Short hairpin RNA-mediated knockdown of KLF4

Lentiviral vectors (pLKO.1-puro) expressing KLF4-targeted short hairpin RNA (shRNA; TRCN0000005313) or non-target control shRNA (SHC002) were purchased from Sigma–Aldrich. For generation of lentiviral particles, HEK293FT cells were co-transfected with the plasmid and ViraPower Lentiviral packaging mix (pLP1, pLP2, pLP-VSV-G; Invitrogen) using Lipofectamine 2000 (Invitrogen). The culture supernatants containing virus particles were harvested at 48 h after transfection, filtered through 0.45 μm filters, and concentrated using a Lenti-X Concentrator (Clontech Laboratories, Inc., Mountain View, CA) for 24 h at 4°C. For lentiviral transduction, cells were treated with the shRNA-expressing lentivirus in the presence of 5 μg/mL of polybrene (Sigma–Aldrich). Twenty-four hours after infection, cells were cultured under puromycin (1 μg/mL) selection for one week and then passaged before use.

In vitro cell proliferation and chemoresistance assay

For measurement of cell proliferation and chemoresistance, cells were seeded in 24-well culture plates at 1 × 10⁴ cells/well in DMEM or RPMI 1640 medium supplemented with 10% FBS. After 24 h, cells were treated with DMEM or RPMI 1640 medium containing 5% FBS and various concentrations of paclitaxel or doxorubicin. Cell viability was determined by using the 3-(4, 5-dimethylthiazol-2-yl)-2, 5-diphenyltetrazoliumbromide (MTT) assay. Formazan granules generated by the cells were dissolved in 100 μL of dimethyl sulfoxide, and the absorbance at 570 nm was measured using a microplate spectrophotometer (TECAN, Männedorf, Switzerland).

Reverse transcription polymerase chain reaction and polymerase chain reaction–based short tandem repeat genotyping

Total cellular RNA was extracted using the TRI Reagent (Sigma–Aldrich). For reverse transcription polymerase chain reaction (RT-PCR) analysis, 2 μg aliquots of RNA were reverse transcribed to cDNA using 200 IU of HelixCript™ thermo reverse transcriptase (Nanohelix, Daejeon, South Korea) and 0.5 μg of oligo (dT) 18 primer. The cDNA in 20 μL of the reaction mixture was amplified with HelixAmp™ Ready-2 × -Go PCR MIX (Nanohelix) and 10 pmol of each forward and reverse primer. The primers used for specific amplification for RT-PCR are summarized in Supplementary Table S1 (Supplementary Data are available online at www.liebertpub.com/thy). PCR products were size fractionated on 1% ethidium bromide/agarose gels and quantified under UV transillumination.

Real-time quantitative PCR (qPCR) was performed in an ABI 7500 sequence detector (Applied Biosystems, Foster City, CA) using a SYBR Green Kit (Applied Biosystems). The reaction condition consisted of 20 μL of SYBR Green Mix mixture, cDNA, and 10 pmol of each primer. Annealing was performed at 60°C, and 40 reaction cycles were performed. mRNA levels were calculated by the threshold cycle (Ct) values and normalized to human GAPDH expression. The primers used for real-time PCR are summarized in Supplementary Table S2. The median Ct values are shown as a box plot in Supplementary Figure S1.

Cell authentication was confirmed by STR analysis performed by Cosmogenetech Co. Ltd. (Seoul, Republic of Korea; www.cosmogenetech.com; Supplementary Tables S3–S5).

Iodine uptake assay

Radioactive iodine uptake was determined according to a previously reported protocol with a slight modification (18). ATC cells infected with sh-control or sh-KLF4 lentiviruses were seeded in each well of a 12-well plate at 1 × 10⁵ cells/well. Na¹³¹I (1 μCi) and nonradioactive NaI (5 μM) were added to each well and incubated for 30 min. To avoid any nonspecific iodine uptake, parallel cells transfected with the control shRNA or KLF4-specific shRNA were preincubated with 100 μM of NaClO₄, a competitive NIS inhibitor, for 30 min before treatment with Na¹³¹I. After completion of the incubation time, the radioactive medium was aspirated, and the cells were washed with ice-cold PBS and lysed in 0.1 M of NaOH buffer at room temperature. The cell-associated radioactivity of the collected lysed cells was measured using a γ-counter. The specific iodine uptake value was obtained by subtracting the iodine uptake value in the presence of NaClO₄ from the values counted in the absence of NaClO₄. In parallel cell cultures, cells were incubated with the same conditions, harvested by trypsinization, and counted with a hemocytometer. Iodine uptake was normalized to cell numbers and expressed as cpm/10⁶ cells. To measure TSH-induced iodine uptake in normal thyrocytes, cells were cultured in RPMI 1640 medium supplemented with 10% FBS, 10 μg/mL of insulin, and 2.5 μg/mL of amphotericin B in the absence or presence of 5 mIU/mL of TSH. After 48 h of incubation, cells were seeded in a 12-well plate at 1 × 10⁵ cells/well, followed by measurement of iodine uptake.

Promoter luciferase assay

ABCG2 promoter constructs were kindly donated by Dr. Douglas D. Ross (University of Maryland School of Medicine, Baltimore, MD) (19). HTH83 cells and TPC-1 cells were seeded in 12-well culture plates at 3 × 10⁵ cells/well. Transfection experiments were performed 16 h after cell seeding using Lipofectamine/Lipofectamine plus (Life Technologies) according to the manufacturer's instructions. Transfection mixtures contained 400 ng of reporter construct (ABCG2 reporter constructs: pGL3-basic containing ABCG2 promoter [−1285] and deletion of KLF4 binding element ABCG2 promoter [−628]), 400 ng of nTAP-KLF4 expression plasmid, and 10 ng of internal control plasmid (pCMV-RL vector containing Renilla luciferase; Promega. Madison, WI). After 24 h, cells were harvested, lysed, and centrifuged at 2000 g for 3 min at 4°C, and the luciferase activity was determined using the dual-luciferase reporter assay system (Promega) on a VICTOR3 Multilabel Plate Reader (Perkin Elmer, Waltham, MA) after dilution to a linear range.

In vivo chemoresistance assay in a xenograft tumor model

All animal studies adhered to protocols approved by the Pusan National University Institutional Animal Care and Use Committee. To assess the role of KLF4 in the drug resistance of ATC in vivo, HTH83 cells were infected with lentivirus encoding control shRNA or KLF4-specific shRNA. Infected cells were injected into alternate thigh muscles of eight-week-old male BALB/c nude mice (1 × 10⁶ cells/injection). When tumor volumes reached approximately 150–300 mm³, mice were randomly assigned to two treatment groups: the paclitaxel-treated group or the untreated control group. Paclitaxel (5 mg/kg) was intraperitoneally injected twice a week. Tumor burden was measured twice a week by palpation and calipers, and tumor volumes were calculated by the following formula: (length × width × height)/2. All mice were sacrificed by anesthetic overdose on day 33, and weights of xenograft tumors were quantified. Occasional tumors containing cysts filled with fluid were excluded from volume measurement.

Immunofluorescence staining

For immunofluorescence staining, tumor specimens were removed on day 33 after injection of HTH83 cells, fixed with formalin, and embedded in paraffin. Tissue sections of 6 μm thickness were prepared from the paraffin-embedded specimens at 150 μm intervals, stained with the indicated primary antibodies (1:100) for 24 h, and then incubated with Alexa Fluor 488- and Alexa Fluor 568-conjugated secondary antibodies (1:1000; Life Technologies) for 2 h. The immunolabeled specimens were washed and mounted in Vectashield medium (Vector Laboratories, Burlingame, CA) with 4′,6-diamidino-2-phenylindole for visualization of nuclei. The stained sections were visualized using laser scanning confocal microscopy (Olympus FluoView FV1000).

Statistical analysis

Results of multiple observations are presented as mean ±standard deviation. Multiple group means were compared by one- or two-way analyses of variance, followed by Scheffe's post hoc test.

Results

ATC cells show high drug resistance and high expression of KLF4

To evaluate the drug resistance of ATC, the effects of paclitaxel on the viability of ATC cell lines 8505C, HTH83, and SW1736, as well as the papillary thyroid cancer cell line TPC-1 and the normal thyroid cell line Nthy-ori 3-1, were determined by MTT assay. As shown in Figure 1A, paclitaxel treatment drastically inhibited cell viability of TPC-1 and Nthy-ori 3-1 cells at ≥0.1 μM concentration. Compared to paclitaxel-sensitive properties of Nthy-ori 3-1 and TPC-1 cells, three ATC cell lines, including 8505C, HTH83, and SW1736 cells, displayed a significant resistance to paclitaxel.

FIG. 1.

Drug resistance and Krüppel-like factor 4 (KLF4) expression in thyroid cells, thyroid cancer cells, and anaplastic thyroid cancer (ATC) cells. (A) Drug resistance determined by 3-(4, 5-dimethylthiazol-2-yl)-2, 5-diphenyltetrazoliumbromide (MTT) assay. Cell viability was measured after treating ATC cells (8505C, HTH83, and SW1736), papillary thyroid cancer cells (TPC-1), and normal thyroid cells (Nthy-ori 3-1) with the indicated concentrations of paclitaxel for 48 h (n = 4; *p < 0.05 vs. mock-treated control; data presented as mean ± standard deviation [SD]). (B) Western blotting analysis of KLF4 expression (upper panel). 8505C, HTH83, SW1736, TPC-1, and Nthy-ori 3-1 cell lysates were subjected to Western blotting analysis with anti-KLF4 and anti-GAPDH antibodies. Quantification of KLF4 expression (lower panel). The band intensities of KLF4 and GAPDH were quantified and the relative intensities of KLF4 relative GAPDH are presented as mean ± SD (n = 4; *p < 0.05 vs. TPC-1).

To evaluate the role of stemness factor KLF4 in ATC, the expression levels of KLF4 in three ATC cell lines, Nthy-ori 3-1, and TPC-1 cells were compared using Western blotting. As shown in Figure 1B, KLF4 expression was significantly higher in the paclitaxel-resistant ATC cell lines (8505C, HTH83, and SW1736) than the paclitaxel-sensitive TPC-1 and Nthy-ori 3-1 cells. Among ATC cell lines, HTH83 showed the most potent resistance against paclitaxel, along with high level of KLF4 expression. Thus, HTH83 was chosen for assessing the possible role of KLF4 in maintaining drug resistance and the undifferentiated phenotype of ATC.

KLF4 knockdown induces differentiation of ATC cells

To evaluate the role of KLF4 in maintaining the undifferentiated phenotype of ATC, KLF4 expression in ATC cells was knocked down by infection with lentivirus encoding a KLF4-specific shRNA. The effect of silencing of KLF4 gene expression on the expression levels of thyroid-specific genes was examined by qPCR and Western blotting. The KLF4-specific shRNA potently reduced the expression level of KLF4 compared to control shRNA, whereas KLF4 silencing significantly increased the protein levels of NIS and TSHR in ATC cells. KLF4 expression in normal thyrocytes was quite low in contrast to high expressions of NIS and TSHR (Fig. 2A). To confirm the results, mRNA levels of thyroid-specific genes were quantified by qPCR analysis in thyrocytes from normal thyroid tissue and ATC cells with or without KLF4 silencing. All three ATC cell lines exhibited low expression levels of thyroid-specific genes, including NIS, TPO, TG, and TSHR, compared to normal thyrocytes, and knockdown of KLF4 significantly increased the expression levels of all four thyroid-specific genes in ATC cells (Fig. 2B). In addition, KLF4 knockdown decreased CD133 expression in 8505C and SW1736 cells (Supplementary Fig. S2). These results suggest that KLF4 plays an important role in maintaining the undifferentiated phenotype of ATC.

FIG. 2.

Effect of KLF4 knockdown on differentiation of ATC cells. (A) Induced expression of thyroid markers in response to KLF4 knockdown. KLF4 expression in ATC cells was knocked down by transfection of KLF4-targeted shRNA. Western blotting analysis results with the indicated antibodies are shown. NT, normal thyrocytes. (B) KLF4 knockdown in ATC cells and quantitative polymerase chain reaction (PCR) analysis of thyroid-specific markers. ATC cells with or without KLF4 knockdown and normal thyrocytes were subjected to real-time PCR analysis with the indicated probes. Data are presented as mean ± SD (n = 4; *p < 0.05). (C) KLF4 knockdown in ATC cells and iodine uptake analysis. ATC cells with or without KLF4 knockdown were incubated with radioactive iodine, and the iodine uptake by ATC cells were quantified (left panel). Iodine uptake by normal thyrocytes with and without thyrotropin stimulation was quantified after incubation with radioactive iodine (right panel). Data are presented as mean ± SD (n = 3; *p < 0.05).

In order to explore whether the differentiation of ATCs induced by KLF4 silencing can affect iodine uptake, iodine uptake was measured in the three ATC cell lines. Consistent with the result that KLF4 silencing increased the expression levels of thyroid-specific genes, knockdown of KLF4 expression led to elevated uptake of iodine in three ATC cell lines, including 8505C, HTH83, and SW1736, and normal thyrocytes showed an elevation of iodine uptake after TSH stimulation (Fig. 2C). KLF4 knockdown showed little effect on the proliferation of 8505C, HTH83, and SW1736 cells when followed for 48 h after introduction of shRNA (Supplementary Fig. S3), which indicates that the increased iodine uptake in the KLF4-silenced ATC cells was not due to a change in cell proliferation. These results support the results that silencing of KLF4 expression promotes the differentiation of ATCs.

KLF4 knockdown decreases drug resistance of ATC cells

To examine whether KLF4 contributes to the anticancer drug resistance of ATC, the effects of KLF4 knockdown on resistance of ATC cells to the conventional anticancer drugs doxorubicin and paclitaxel were evaluated. KLF4 knockdown resulted in increased sensitivity to doxorubicin and paclitaxel compared to control shRNA treatment (Fig. 3A). To elucidate the molecular mechanism of KLF4-dependent acquisition of drug resistance in ATC, the effects of KLF4 knockdown on the expression levels of multidrug transporters were further examined. Consistently, KLF4 knockdown reduced the mRNA levels of ABC transporters ABCG2, ABCB1, and ABCC6 in ATC cells (Fig. 3B). These results suggest that KLF4 is required for the expression of ABC transporters in ATC cells, thereby conferring anticancer drug resistance to ATC cells.

FIG. 3.

Effects of KLF4 knockdown on drug resistance and ABC transporter expression in ATC cells. (A) KLF4 knockdown in ATC cells and drug resistance. ATC cells with or without KLF4 knockdown were treated with the indicated concentrations of doxorubicin and paclitaxel and the viabilities of cells were measured by MTT assay at 48 h after drug treatment. Data are presented as mean ± SD (n = 9; *p < 0.05). (B) KLF4 knockdown in ATC cells and reverse transcription RT-PCR analysis. ATC cells with or without KLF4 knockdown were subjected to RT-PCR analysis with indicated probes. Representative data from three independent experiments are shown.

KLF4 expression increases ABCG2 promoter activity

To assess whether KLF4 directly regulates the transcription of the ABCG2 transporter gene, two luciferase reporter constructs of the ABCG2 promoter were generated—one containing two KLF4 binding sites (−1285 bp), and the other containing one KLF4 binding site (−628 bp)—and promoter activation by KLF4 was compared (Fig. 4A). Co-transfection of the KLF4 gene together with the ABCG2-1285 luciferase reporter construct or ABCG2-628 luciferase reporter construct markedly upregulated luciferase activity compared to the transfection control in HTH83 cells (Fig. 4B). The KLF4-dependent luciferase activity of cells expressing the ABCG2-1285 construct was much higher than that of cells expressing the ABCG2-628 luciferase reporter construct. Conversely, shRNA-mediated knockdown of KLF4 expression abrogated the upregulation of ABCG2 promoter activity induced by KLF4 overexpression. Not only HTH83 cells but also TPC-1 papillary thyroid cancer cells showed an activation of ABCG2 promoters in response to KLF4 overexpression (Fig. 4C). These results suggest that KLF4 regulates ABCG2 expression by controlling the promoter activity of the ABCG2 gene.

FIG. 4.

Effect of KLF4 overexpression on ABCG2 promoter activity. (A) Schematic presentation of ABCG2 promoters inserted into the reporter constructs. The ABCG2 promoter −1285 contains two KLF4 binding sites (−1285 bp and −628 bp), and the ABCG2 promoter −628 construct contains one KLF4 binding site (−628bp). (B) KLF4-dependent activation of ABCG2 promoter in HTH83 cells. HTH83 cells were transfected with −628 or −1285 ABCG2 promoter reporter construct in combination with KLF4 overexpression (nTAP-KLF4) and KLF4 knockdown (sh-KLF4). Luciferase activities were normalized to transfection overexpression control (nTAP). Data are presented as mean ± SD (n = 6; *p < 0.05). (C) KLF4-dependent activation of ABCG2 promoter in TPC-1 cells. TPC-1 cells were transfected with −628 or −1285 ABCG2 promoter reporter construct in combination with KLF4 overexpression (nTAP-KLF4). Luciferase activities were normalized to transfection overexpression control (nTAP). Data are presented as mean ± SD (n = 6; *p < 0.05).

KLF4 knockdown and paclitaxel treatment decreases ATC-derived tumor size

To assess whether KLF4 knockdown affects the drug resistance of ATC in vivo, KLF4-silenced HTH83 cells and control HTH83 cells were transplanted into contralateral thigh muscles of nude mice. Paclitaxel was administered twice a week when tumor growth reached an appropriate volume, and the effects of paclitaxel administration on tumor growth and tumor weight were explored. Repeated administration of paclitaxel dramatically reduced the size of tumors derived from KLF4-knockdown HTH83 cells, whereas tumors from control virus-infected HTH83 cells did not respond to paclitaxel treatment (Fig. 5A and B). Moreover, paclitaxel administration significantly reduced the weight of xenograft tumors derived from KLF4-knockdown HTH83 cells compared to that of control xenograft (Fig. 5C). Some xenograft mice had cystic tumors (Supplementary Table S6), and the cystic tumors were excluded from calculation of tumor volume and weight. Consistent with in vitro results, immunofluorescence analysis showed that KLF4 knockdown decreased ABCG2 expression in tumors (Fig. 5D). These results suggest that knockdown of KLF4 increases sensitivity to paclitaxel by reducing the expression of ABC transporters. Collectively, these results suggest KLF4 can be a potential therapeutic target for eliminating ATC.

FIG. 5.

ATC-derived tumor xenografts with or without KLF4 knockdown. (A) Representative picture of nude mice injected with HTH83 cells with or without KLF4 knockdown at contralateral sites. Paclitaxel (PTX; 5 mg/kg) was injected into mice on day 16 after HTH83 cell injection. The image was taken on day 33 after injection of HTH83 cells (left panel), and representative images of tumors collected were taken on day 33 after injection of HTH83 cells (right panel). (B) Tumor volumes at the indicated time points after injecting HTH83 cells with or without KLF4 knockdown in combination with PTX treatment (n = 8; *p < 0.05; data presented as mean ± SD). Arrows indicate the time points of intraperitoneal PTX injection (5 mg/kg). (C) Tumor weight on day 33 after injection of HTH83 cells with or without KLF4 knockdown in combination with PTX treatment (n = 8; *p < 0.05; data presented as mean ± SD). (D) KLF4 and ABCG2 expression in xenograft tumors. Representative immunofluorescence data of xenograft tumor tissues derived from HTH83 cells with or without KLF4 knockdown on day 33 with the indicated antibodies are shown. Scale bar = 50 μm.

Discussion

The presence of CSCs may account for the high degree of dedifferentiation, sustained proliferation, and resistance to chemotherapy of ATC (20,21). Increasing evidence suggests that ATC is initiated by dedifferentiation of papillary thyroid cancer cells or overwhelming growth of pluripotent CSCs embedded in benign or malignant thyroid follicular cells (21). CSC markers such as CD133 and nestin were more highly expressed in ATC cells than in papillary thyroid cancer cells (22). In the analysis of patient-derived samples, microRNA expression in neoplastic human thyroid tissue samples showed a downregulation of miR-25, which targets KLF4 (23,24). In the analysis of formalin-fixed, paraffin-embedded ATC specimens, KLF4 showed variable hyperexpression (5). Though not conclusive, these results support the relevance of investigating the role of KLF4 in ATC.

The present study demonstrates that KLF4 is highly expressed in ATC cell lines compared to a control papillary thyroid cancer cell line. Knockdown of KLF4 expression led to increased expression of the thyroid-specific genes NIS, TG, TPO, and TSHR, suggesting that KLF4 plays an important role in maintaining the undifferentiated status of ATC cells and that downregulation of KLF4 promotes differentiation. In addition to a function in self-renewal of ES cells, KLF4 has also been shown to regulate cell proliferation and differentiation. Inhibition of KLF4 expression reduced the expression of colon CSC marker genes in colon CSC-enriched spheroid cells (25). In addition, KLF4 was upregulated in breast CSC-like cells, and downregulation of KLF4 decreased tumor formation in vivo (26). Moreover, forced expression of KLF4 inhibited ES cells from differentiating into erythroid progenitors and increased their ability to generate secondary embryoid bodies, suggesting a role in maintaining the self-renewal capacity of ES cells (27). These results suggest that KLF4 plays a key role in the regulation of the undifferentiated phenotype and the proliferative characteristics of ATC cells. However, the expression levels of thyroid-specific genes could not be fully restored by silencing of KLF4 expression in ATC cell lines. Consistently, the increase in iodine uptake induced by KLF4 silencing was relatively marginal compared to normal thyrocytes. These results suggest that the undifferentiated phenotypes of ATCs may be regulated not only by KLF4 but also other factors, although the molecular identity of these factors needs to be further clarified.

ABCG2, a member of the human ATP-binding cassette transporter family, is overexpressed in many cancer types and plays a critical role in maintaining chemoresistance (28,29). A recent study showed that the failure of doxorubicin to eradicate ATC cells was mainly due to the resistance of CSCs to the chemotherapeutic drug and the expression of ABCG2 in HTH47R ATC cells (4). Moreover, ABCG2 has been shown to confer resistance against various chemotherapeutic drugs to prostatic cancer cells (30). Increased expression of ABCG2 together with stem-cell markers, including SOX2, OCT4, and KLF4, was suggested to enhance chemoresistance and promote recurrence of prostatic cancer (31). The present study demonstrates that silencing of KLF4 expression abrogated the expression of ABCG2 as well as two other ABC transporters, ABCB1 and ABCC6, in HTH83 ATC cells. Ectopic overexpression of KLF4 potentiated activation of the ABCG2 promoter through activation of a KLF4-responsive element. Moreover, KLF4 knockdown enhanced the sensitivity of HTH83 cells to several anticancer drugs in vitro and to paclitaxel in an in vivo xenograft transplantation model. These results suggest that KLF4 plays a pivotal role in the development of drug resistance in ATC by upregulating expression of multidrug resistance proteins.

The present study demonstrates that a high level of KLF4 expression in ATC cells is necessary for maintaining the undifferentiated phenotype and the high drug resistance of ATC cells. These results may provide a novel opportunity for developing therapeutics to eradicate ATC and restore patients' health.

Footnotes

Acknowledgments

This research was supported by grants (NRF-2015R1A5A2009656; NRF-2015R1B1A1A01060977) of the National Research Foundation of Korea funded by the Ministry of Education, Science and Technology and Cancer Control Ministry for Health Welfare and Family Affairs of Korea (0920050).

Author Disclosure Statement

The authors declare no conflict of interest.

References

Kebebew

, Greenspan

, Clark

, Woeber

, McMillan

. 2005. Anaplastic thyroid carcinoma. Treatment outcome and prognostic factors. Cancer, 103:1330–1335.

Smallridge

, Copland

. 2010. Anaplastic thyroid carcinoma: pathogenesis and emerging therapies. Clin Oncol, 22:486–497.

Ain

. 1999. Anaplastic thyroid carcinoma: a therapeutic challenge. Semin Surg Oncol, 16:64–69.

Zheng

, Cui

, Xu

, Brabant

, Derwahl

. 2010. Doxorubicin fails to eradicate cancer stem cells derived from anaplastic thyroid carcinoma cells: characterization of resistant cells. Int J Oncol, 37:307–315.

Carina

, Zito

, Pizzolanti

, Richiusa

, Criscimanna

, Rodolico

, Tomasello

, Pitrone

, Arancio

, Giordano

. 2013. Multiple pluripotent stem cell markers in human anaplastic thyroid cancer: the putative upstream role of SOX2. Thyroid, 23:829–837.

Shields

, Christy

, Yang

. 1996. Identification and characterization of a gene encoding a gut-enriched Kruppel-like factor expressed during growth arrest. J Biol Chem, 271:20009–20017.

Zhang

, Andrianakos

, Yang

, Liu

, Lu

. 2010. Kruppel-like factor 4 (Klf4) prevents embryonic stem (ES) cell differentiation by regulating Nanog gene expression. J Biol Chem, 285:9180–9189.

, McClintick

, Zhong

, Edenberg

, Yoder

, Chan

. 2005. Murine embryonic stem cell differentiation is promoted by SOCS-3 and inhibited by the zinc finger transcription factor Klf4. Blood, 105:635–637.

Takahashi

, Yamanaka

. 2006. Induction of pluripotent stem cells from mouse embryonic and adult fibroblast cultures by defined factors. Cell, 126:663–676.

10.

Wernig

, Meissner

, Foreman

, Brambrink

, Ku

, Hochedlinger

, Bernstein

, Jaenisch

. 2007. In vitro reprogramming of fibroblasts into a pluripotent ES-cell-like state. Nature, 448:318–324.

11.

Evans

, Liu

. 2008. Roles of Krupel-like factor 4 in normal homeostasis, cancer and stem cells. Acta Biochim Biophys Sin, 40:554–564.

12.

Tetreault

, Yang

, Katz

. 2013. Kruppel-like factors in cancer. Nat Rev Cancer, 13:701–713.

13.

Lin

. 2011. Thyroid cancer stem cells. Nat Rev Endocrinol, 7:609–616.

14.

Mitsutake

, Iwao

, Nagai

, Namba

, Ohtsuru

, Saenko

, Yamashita

. 2007. Characterization of side population in thyroid cancer cell lines: cancer stem-like cells are enriched partly but not exclusively. Endocrinology, 148:1797–1803.

15.

Todaro

, Iovino

, Eterno

, Cammareri

, Gambara

, Espina

, Gulotta

, Dieli

, Giordano

, De Maria

, Stassi

. 2010. Tumorigenic and metastatic activity of human thyroid cancer stem cells. Cancer Res, 70:8874–8885.

16.

Thomas

, Friedman

, Lin

. 2008. Thyroid stem cells: lessons from normal development and thyroid cancer. Endocr Relat Cancer, 15:51–58.

17.

Zhang

, Zuo

, Ozaki

, Nakagomi

, Kakudo

. 2006. Cancer stem cell hypothesis in thyroid cancer. Pathol Int, 56:485–489.

18.

Weiss

, Philp

, Grollman

. 1984. Iodide transport in a continuous line of cultured cells from rat thyroid. Endocrinology, 114:1090–1098.

19.

Bailey-Dell

, Hassel

, Doyle

, Ross

. 2001. Promoter characterization and genomic organization of the human breast cancer resistance protein (ATP-binding cassette transporter G2) gene. Biochim Biophys Acta, 1520:234–241.

20.

Takano

, Amino

. 2005. Fetal cell carcinogenesis: a new hypothesis for better understanding of thyroid carcinoma. Thyroid, 15:432–438.

21.

Zito

, Richiusa

, Bommarito

, Carissimi

, Russo

, Coppola

, Zerilli

, Rodolico

, Criscimanna

, Amato

, Pizzolanti

, Galluzzo

, Giordano

. 2008. In vitro identification and characterization of CD133(pos) cancer stem-like cells in anaplastic thyroid carcinoma cell lines. PloS One, 3:e3544.

22.

Jung

, Han

, Seol

, Park

, Koh

, Lee

, Kim

, Choi

, Myung

. 2015. Expression of cancer stem cell markers and epithelial-mesenchymal transition-related factors in anaplastic thyroid carcinoma. Int J Clin Exp Pathol, 8:560–568.

23.

Fuziwara

, Kimura

. 2014. MicroRNA deregulation in anaplastic thyroid cancer biology. Int J Endocrinol, 2014:743450.

24.

Esposito

, Tornincasa

, Pallante

, Federico

, Borbone

, Pierantoni

, Fusco

. 2012. Down-regulation of the miR-25 and miR-30d contributes to the development of anaplastic thyroid carcinoma targeting the polycomb protein EZH2. J Clin Endocrinol Metab, 97:E710–718.

25.

Leng

, Tao

, Xia

, Tan

, Yue

, Chen

, Xi

, Li

, Zheng

. 2013. Kruppel-like factor 4 acts as an oncogene in colon cancer stem cell-enriched spheroid cells. PloS One, 8:e56082.

26.

, Li

, Chen

, Fu

, Ray

, Huang

, Zheng

, Ai

. 2011. Kruppel-like factor 4 (KLF4) is required for maintenance of breast cancer stem cells and for cell migration and invasion. Oncogene, 30:2161–2172.

27.

Zheng

, Pritchard

, Yang

, Bennett

, Liu

, Ai

. 2009. KLF4 gene expression is inhibited by the notch signaling pathway that controls goblet cell differentiation in mouse gastrointestinal tract. Am J Physiol Gastrointest Liver Physiol, 296:G490–498.

28.

Hirschmann-Jax

, Foster

, Wulf

, Nuchtern

, Jax

, Gobel

, Goodell

, Brenner

. 2004. A distinct “side population” of cells with high drug efflux capacity in human tumor cells. Proc Natl Acad Sci U S A, 101:14228–14233.

29.

Doyle

, Yang

, Abruzzo

, Krogmann

, Gao

, Rishi

, Ross

. 1998. A multidrug resistance transporter from human MCF-7 breast cancer cells. Proc Natl Acad Sci U S A, 95:15665–15670.

30.

Gangavarapu

, Azabdaftari

, Morrison

, Miller

, Foster

, Huss

. 2013. Aldehyde dehydrogenase and ATP binding cassette transporter G2 (ABCG2) functional assays isolate different populations of prostate stem cells where ABCG2 function selects for cells with increased stem cell activity. Stem Cell Res Ther, 4:132.

31.

Guzel

, Karatas

, Duz

, Solak

, Ittmann

, Ozen

. 2014. Differential expression of stem cell markers and ABCG2 in recurrent prostate cancer. Prostate, 74:1498–1505.