CHK2 Promotes Anoikis and is Associated with the Progression of Papillary Thyroid Cancer

Thyroid cancer, which is the most common endocrine malignancy, continues to be the most rapidly increasing cancer type [1]. Papillary thyroid cancer (PTC) is the most common malignant thyroid tumor, as it accounts for 75% to 85% of all thyroid cancers [2, 3]. The prognosis of treated PTC is generally excellent, and the 10-year overall survival rates exceed 90% [1], but the 5-year survival rate is only 50%, with repeated recurrence and/or distant metastases. Regional lymph node metastasis (LNM), which is common in PTC, is present in 20-90% of patients at the time of diagnosis [4, 5]. Emerging studies have shown that LNM is one of the most significant factors for locoregional recurrence (LRR), and patients with lymphatic metastasis are more likely to have poor survival rates [6-9]. In addition, the extent of LNM is also associated with the risk of recurrence [10], a finding that was included in the American Thyroid Association (ATA) risk of recurrence stratification system [11]. Compared with those who have fewer than five metastatic lymph nodes (LN mets), patients with five or more LN mets have a higher risk of recurrence [12]. Other factors are also associated with an unfavorable prognosis including age (>45 years), male sex, histology (tall cell variant histology), extrathyroidal extension (EXE), larger tumor size, stage III and IV disease, resistance to radioiodine therapy, distant metastasis, positive family history, and the BRAF^V600E mutation [6, 7, 9, 13-19]. Therefore, the underlying inherent molecular mechanism of the aggressiveness and progression of PTC has great significance in the ability to understand the biological basis for the improvement in clinical outcomes [20-22].

Checkpoint kinase 2 (CHEK2) is a tumor suppressor gene that encodes the serine/threonine protein kinase CHK2. The CHK2 kinase is a key mediator of the DNA damage checkpoint that responds to DNA double-strand breaks (DSBs), and it plays a crucial role in the maintenance of genomic integrity. Following genotoxic stress, CHK2 is activated upon its phosphorylation to p-CHK2, after which it phosphorylates proteins that are involved in the DNA damage response, including cdc25A, cdc25C, Mdmx, p53, BRCA1, PML, and E2F1, to induce cell cycle arrest in G2/M phase, apoptosis, and DNA repair [23].

Mutations in CHEK2 not only exist in subsets of sporadic cancers, but they also predispose patients to several types of familial carcinomas [24-26]. Individuals who carry certain CHEK2 mutations have a statistically significant increase in the incidence of prostate, breast, and papillary thyroid cancer, among others [26-32]. These mutations affect the stability or kinase activity of CHK2, which may contribute to the pathogenesis of tumors. In addition, CHK2 is constitutively activated in invasive urinary bladder carcinomas, and high expression of DNA repair pathways is associated with metastasis in patients with melanoma, which suggests that CHK2 can also mediate tumor progression [25, 33]. Although emerging studies have shown that patients who carry the CHEK2-associated variant have an increased risk of PTC [27, 28], the status and contribution of CHK2 to the tumorigenesis and progression of PTC are far from clear. In the present study, we investigated the CHK2 expression levels in PTC tissues and assessed the contribution of CHK2 to LNM. The role of CHK2 in the regulation of malignancy was also investigated.

In all, 100 patients who were diagnosed with PTC were selected from Peking Union Medical College Hospital (PUMCH), China, from January 2014 to June 2016. None of these patients received preoperative chemotherapy or radiotherapy. All enrolled patients underwent a thyroidectomy (total or lobectomy and isthmusectomy) with a level of more than 5 lymph nodes dissected for pathological diagnosis to document the true lymph node-negative cases that complemented the clinical assessment by ultrasound. All samples, including matched cancer tissues, tumor-adjacent thyroid tissues, and LN mets (for patients with LNM), were collected after written informed consent was provided by the recruited patients. All experiments involving human samples were approved by the Institutional Review Board of PUMCH. Moreover, this study was performed in accordance with the Declaration of Helsinki.

IHC staining was performed on representative 4-um-thick formalin-fixed, paraffin-embedded tissue sections using standard techniques as previously described [34]. Two pathologists who specialize in PTC independently evaluated the staining intensity and the percentage of positive cells. For the staining assessment of CHK2, scoring systems were applied to rate the staining intensity (no staining: 0; weak staining: 1; intermediate staining: 2; and strong staining: 3) and to determine the percentage of positive cells (<5%: 0; 5-25%: 1; >25%-50%: 2; >50%-75%: 3; >75%: 4) [35]. For the staining assessment of p-CHK2, a similar scoring system was used to rate the staining intensity (no staining: 0; weak staining: 1; intermediate staining: 2; and strong staining: 3) and to determine the percentage of positive cells (<5%: 1; 5-25%: 2; >25%-50%: 3; >50%: 4) [36]. The final score was equal to the staining intensity multiplied by the percentage of positive cells. Based on the final scores, the stained tissues were divided into two categories: those with high expression and those with low expression. A high CHK2 expression level was defined if the final score was ≥4; otherwise, the CHK2 level was defined as low. For the p-CHK2 level, final scores equal to or greater than 2 were indicative of high expression, and scores <2 were indicative of low expression. Moreover, CHK2 and p-CHK2 staining was also classified into two categories: negative or positive expression. The expression was determined to be negative if no cells were stained; otherwise, the expression was considered to be positive.

The two human PTC cell lines BCPAP and KTC-1 were kindly provided by Stem Cell Bank, Chinese Academy of Sciences (Shanghai, China). Another human PTC cell line, TPC-1, was acquired from JENNIO Biological Technology (Guangzhou, China). These three cell lines were cultured in RPMI 1640 medium (containing non-essential amino acids for BCPAP and KTC-1) supplemented with 10% fetal bovine serum (FBS; HyClone, USA) at 37°C in a humidified chamber with 5% CO₂. For suspension cell culture conditions, cells were seeded in six-well plates coated with poly-HEMA (AMRESCO, USA).

Specific siRNA oligos for human CHEK2, p53, and PRAS40 as well as control oligos were synthesized by RiboBio Co. (Guangzhou, China). The siRNAs (50-100 nM) were transfected using Lipofectamine 3000 transfection reagent (Invitrogen) according to the manufacturer’s protocol. Control RNA (Ctrl-siN05815122147) was obtained from RiboBio Co. (Guangzhou, China).

The specific siRNA sequences were as follows:

CHEK2-specific siRNA-1, 5’ GUAAGAAAGUAGCCAUAAA dTdT 3’;

CHEK2-specific siRNA-2, 5’ GGAAUAAACGCCUGAAAGA dTdT 3’;

CHEK2-specific siRNA-3, 5’ GAACAUACAGCAAGAAACA dTdT 3’;

p53-specific siRNA, 5’-GGAAGACUCCAGUGGUAAUdTdT-3’

PRAS40-specific siRNA, 5-CCAGAAGCUGAAGCGGAAAUAUUGA-3

Total RNA was isolated from cells using TRIzol (Invitrogen, Carlsbad, CA, USA), and complementary DNA (cDNA) was synthesized from total RNA using a reverse transcription kit (TaKaRa, Japan) according to the manufacturer’s instructions. Real-time qRT-PCR was performed using SYBR Green Master Mix (TaKaRa, Japan) to quantify the mRNA levels. β-actin was used as a housekeeping gene.

The sequences of the CHEK2 primers used were as follows:

Forward Primer: 5’-TTATCTGCCTTAGTGGGTATCCA-3’;

Reverse Primer: 5’-CTGTCGTAAAACGTGCCTTTG-3.’

BCPAP cells that stably expressed either a non-targeting control (Control, Ctrl) or CHK2 were established. The full-length human CHK2 coding sequence was cloned into the GV141 vector, and the resulting vector was termed the CHK2-Overexpression (CHK2-OE) vector. BCPAP cells were seeded into 6-well plates at 50-60% confluence prior to their transfection with plasmids and were transfected with Lipofectamine 3000 transfection reagent according to the manufacturer’s protocol. Cells that stably overexpressed CHK2 were selected using G418 (Amresco, Solon, OH, USA) at a previously optimized concentration of 300 μg/ml. Clonal populations were isolated, expanded and maintained under G418 selection.

Total proteins were extracted from cells using RIPA buffer (Solarbio, Beijing, China). Forty micrograms of total proteins were separated using sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) and were transferred to polyvinylidene difluoride (PVDF) membranes (Millipore, Billerica, MA, USA). The membranes were incubated with primary antibodies overnight at 4°C and were then incubated with horseradish peroxidase-conjugated secondary antibodies at room temperature for 1 h. To assess the protein levels, the optical densities of the bands in each blot were analyzed using the ImageJ Analysis System (Bio-Rad Laboratories Inc., Hercules, CA, USA).

Antibodies against CHK2 (cat. #6334), phospho-CHK2 (cat. #2197), PRAS40 (cat. #2691S), phospho-PRAS40 (cat. #2997S), AKT (cat. #4691), phospho-AKT (cat. #4060), GSK-3β (cat. #12456), phospho-GSK-3β (cat. #9323), Erk1/2 (cat. #4695), phospho-Erk1/2 (cat. #4370), p38 MAPK (cat. #8690), phospho-p38 MAPK (cat. #4511), p53 (cat. #2527), CD44 (cat. #3570), Caspase-3 (cat. #9662), PARP (cat. #9532) and GAPDH (cat. #5174) were all purchased from Cell Signaling Technology (Danvers, MA, USA).

PTC cells were transfected in six-well plates (5×10⁵ cells/well) for 24 h, then trypsinized and reseeded in 96-well plates (2000 cells/well). Cell counting kit (CCK-8) reagent (10 μl/well) was added at 0, 24, 48, and 72 h; the cells were then incubated for another 2.5 h at 37°C. The optical density (OD) was measured at a wavelength of 450 nm (OD450) by a microplate reader (Wellscan MK3, Thermo/Labsystems, Finland).

Invasion and migration were measured using Transwell assays. Invasion was analyzed in Transwell membranes coated with Matrigel, whereas cell migration was analyzed in Transwell membranes without Matrigel. Transfected cells were seeded in medium without FBS in the upper chamber of the Transwell system (polycarbonate membrane, 6.5 mm in diameter, 8-μm pore size; Corning Costar, USA). Medium supplemented with 20% FBS was added to the bottom chamber. After a 24-h incubation, migrating or invading cells were fixed in formalin and stained with hematoxylin and eosin (Zhong Shan Golden Bridge Company, Beijing, China). Three parallel cultures were measured to determine the cell invasion and migration ratios of five randomly selected microscopic fields at a 200× magnification.

Cell apoptosis was measured according to Annexin V/PI-positive staining. Briefly, the cells were harvested, resuspended in binding buffer, stained using an Annexin V-FITC Kit (BD Biosciences, Franklin Lakes, NJ, USA) according to the manufacturer’s instructions and analyzed by flow cytometry (FACScan; BD Biosciences, USA).

PTC cells were collected and evaluated for ALDH activity using an ALDEFLOUR assay kit according to the manufacturer’s instructions (StemCell Technologies, Vancouver, BC, Canada). The analysis of the ALDEFLOUR assay was performed by flow cytometry (FACScan; BD Biosciences, USA).

We suspended cells in 0.3% agar diluted in complete medium (DMEM with 20% FBS) and cultured them on a bottom layer of 0.6% agar in complete medium in 6-well dishes. We then added additional medium and cultured the cells for 10-14 days before colonies consisting of at least five cells from 10× fields or whole wells were counted. We replaced the medium every other day. We embedded the transfected cells in soft agar 48 h after transfection.

Statistical analyses and the generation of graphs were performed using the SPSS 13.0 software (SPSS Inc., Chicago, IL, USA) and GraphPad Prism 5 Software (GraphPad, San Diego, CA, USA), respectively. Measurement data are presented as the means ± standard deviation (SD) and were compared by Student’s t test. Categorical data were compared by Pearson χ² test or Fisher’s exact test. A p value <0.05 was considered statistically significant.

To investigate the role of CHK2 in tumor malignancy, we first evaluated the expression levels of CHK2 and p-CHK2 in cancer tissues and matched tumor-adjacent tissues from 100 patients by IHC staining. With respect to the tumor tissues, 86 showed positive expression of CHK2 (86%), and 14 showed negative expression (14%). However, in the matched tumor-adjacent tissues, only 12 showed positive expression of CHK2 (12%), whereas 88 showed negative expression (88%, p=0.000, Fig. 1). Regarding the expression of p-CHK2, 61 tumor tissues exhibited positive expression (61%), and only 5 matched tumor-adjacent tissues exhibited positive expression (5%, p=0.000, Fig. 1). These results showed that both CHK2 and p-CHK2 were significantly overexpressed in human PTC cancer tissues compared with tumor-adjacent tissues. We next compared CHK2 (and p-CHK2) expression in matched pairs of primary tumors and LN mets. CHK2 was highly expressed in 51 of 80 primary tumors (64%) and in 37 of 80 matched LN mets (46%, p=0.039, Fig. 1). Similarly, p-CHK2 was highly expressed in 54 of 80 primary tumors (67%) and in 41 of 80 matched LN mets (51%, p=0.036, Fig. 1). These data showed that the levels of both CHK2 and p-CHK2 were significantly decreased in LN mets compared with primary tumor tissues.

We next assessed the correlation between CHK2 (and p-CHK2) levels and the clinicopathological parameters of PTC (Table 1). CHK2 expression levels were negatively correlated with multifocality (p=0.014). Specifically, patients with unifocal PTC were more likely to have high CHK2 expression compared with those with multifocal tumors. No correlation was observed between the levels of CHK2 and other clinicopathological parameters. The high expression of p-CHK2 was correlated with male gender (p=0.000), a known risk factor for more aggressive tumors, and classical histology (p=0.047), which is a more aggressive histopathologic subtype than the follicular variant of PTC. Additionally, p-CHK2 expression levels were positively correlated with LNM (p=0.031) but were negatively correlated with advanced N stage (p=0.014). This indicated that patients with high expression of p-CHK2 possibly have LNM that are limited to the central region, but that those with low expression of p-CHK2 may have LNM beyond the central region.

To investigate the biological function of CHK2, we detected its expression in PTC cells and then examined the effect of the transient silencing of CHK2 after the transfection of CHK2-specific small interfering RNAs (Fig. 2A, 2B). Clonal populations of BCPAP cells that stably over-expressed CHK2 were also established (Fig. 2C). Using CCK-8 and Transwell assays, we found that the regulation of CHK2 had no significant effect on proliferation, invasion and migration of PTC cells (Fig. 2D, 2E). We next evaluated PTC cells for the expression of the CSC markers CD44 and ALDH activity by Western blotting and the ALDEFLOUR assay, respectively. The results showed that the downregulation of CHK2 reduced the CD44 expression level, whereas the upregulation of CHK2 increased the CD44 expression level in BCPAP cells (Fig. 2F). Using the ALDEFLOUR assay, ALDH activity was significantly decreased and increased after the downregulation and upregulation of CHK2, respectively, in BCPAP cells (Fig. 2G, 2H).

Growth under anchorage-independent conditions is a hallmark of tumor cell transformation and has been suggested to play a role in metastasis [37-40]. To discern the potential role of CHK2 to account for the regulation of metastatic potential, we first investigated whether the regulation of CHK2 expression in PTC cell lines could affect their viability under suspension culture conditions. Annexin V/PI-positive staining showed that the downregulation of CHK2 reduced the level of apoptosis of detached TPC-1 and BCPAP cells (Fig. 3A, 3B); by contrast, the forced upregulation of CHK2 increased the level of apoptosis of detached BCPAP cells (Fig. 3C). Furthermore, the silencing of CHK2 decreased the levels of the apoptosis markers cleaved-caspase3 and cleaved-PARP, while the forced upregulation of CHK2 increased their levels (Fig. 3D).

We next tested whether the regulation of CHK2 expression could affect the apoptosis of attached PTC cells; however, no significant correlation was found (Fig. 3E). Moreover, the clonogenicity assay showed that the forced downregulation of CHK2 provided an anchorage-independent growth advantage in soft agar for TPC-1 and BCPAP cells, whereas the upregulation of CHK2 decreased this advantage for BCPAP cells (Fig. 3F). These data indicated that CHK2 is associated with the survival of detached tumor cells.

To investigate the mechanism of anoikis in PTC cells, we tested the activation levels of key proteins in the cell survival pathway. In TPC-1, BCPAP and KTC-1 cells, detachment from the extracellular matrix (ECM) was associated with increased ERK1/2 and p38 MAPK activation, whereas the level of AKT and GSK-3β activation was decreased (Fig. 4A). Evidence from previous studies has supported the idea that activation of the anti-apoptotic P13K-AKT signaling pathway plays an important role in anoikis of PTC cells [41]. In addition, recent studies have reported that PRAS40 acts as a downstream regulator of the P13K-AKT pathway and that the Thr246 phosphorylation site can be induced by Akt [42]. Here, we observed that detachment reduced the activation of PRAS40, which reminded us that the level of phosphorylation at the Thr246 site was also dependent on Akt in PTC cells (Fig. 4A). Moreover, we observed that the detachment of the cells from the ECM triggered CHK2 downregulation (Fig. 4B).

To investigate the potential underlying mechanism of CHK2-mediated anoikis, we regulated CHK2 expression to assess the activation levels of the kinases mentioned above in the setting of PTC cell detachment from the ECM. Surprisingly, we found that the downregulation of CHK2 increased PRAS40 activation in PTC cells but had no effect on the activation of other kinases (Fig. 4C). To further investigate the role of PRAS40 in anoikis of PTC cells, we blocked PRAS40 expression by siRNA (Fig. 4D). The results showed that the forced inhibition of PRAS40 expression in PTC cells increased anoikis (Fig. 4E, 4F). Collectively, these data indicated that CHK2 mediates anoikis of PTC cells via the regulation of PRAS40 activation.

Previous studies have shown that CHK2 can induce cells death via the phosphorylation of the pro-apoptotic protein kinase p53 or via p53-independent mechanisms [43, 44]. To investigate whether p53 is involved in anoikis of PTC cells, we silenced p53 by siRNA. The results showed that the forced inhibition of p53 expression did not rescue BCPAP cells from anoikis, which indicates that CHK2 triggers anoikis in a p53-independent manner (Fig. 5).

Genomic instability is a major hallmark of cancer cells and is crucial for tumorigenesis. The DNA damage-induced cell-cycle checkpoints play a central role in the maintenance of genomic integrity, and defects in this checkpoint system can result in genomic instability and cancer predisposition. CHK2 is a central effector of the cell’s response to DNA damage, and the variants of CHK2 that occur in subsets of diverse sporadic malignancies predispose individuals to several types of hereditary carcinomas [24, 25, 28].

Our current study was the first to report the role of CHK2 in the progression and regulation of the malignancy of human papillary thyroid cancer. Our IHC analysis revealed that both CHK2 and p-CHK2 proteins were significantly elevated in most of the human PTC cancer samples examined compared with tumor-adjacent thyroid tissues. Similarly, previous studies have shown that the CHK2 expression level is also increased in the majority of human DLBCL [36] and urinary bladder cancer cases [25] and that more than 50% of the CHK2 protein is phosphorylated at Thr68 in surgically resected lung and breast cancer specimens from untreated patients [45]. The mechanism for elevated CHK2 protein levels in PTC remains poorly understood. It is also unclear why CHK2 is ‘constitutively’ activated even with no exposure to external DNA damage in a high proportion of tumors; however, our study revealed that high expression of p-CHK2 was associated with tumor aggressiveness, including male gender, advanced N stage, classical histology and LNM potential. We speculate that activated CHK2 may be useful in the selection of variant cancer cells with a defect in the DNA damage response and induced tumor progression. It has been reported that CHK2 is highly expressed in both human embryonic stem (ES) cells and induced pluripotent stem (iPS) cells [23, 46]. Our results showed that CHK2 upregulated CSC markers in BCPAP cells. This result demonstrated that tumors with high CHK2 expression could also promote progression via the upregulation of CSC markers. However, we presented unexpected evidence that both the CHK2 and p-CHK2 levels were reduced in LN mets in contrast to matched primary tumors. These data indicated that lower CHK2 expression may be associated with increased tumor cell survival in the process of metastasis.

Next, we investigated the role of CHK2 in tumor malignancy. Interestingly, our data suggested that the inhibition CHK2 expression reduced detachment-induced apoptosis but did not influence the ability of cells to migrate and invade, which illustrates that CHK2 could inhibit tumor progression and metastatic potential by enhancing anoikis.

As a pro-apoptotic kinase, CHK2 has never been reported to mediate anoikis in tumor cells and is thought to induce cell death mainly under the conditions of genomic instability [23, 47]. Our study, for the first time, identified CHK2 as a novel mediator of anoikis in PTC cells, as has already been reported in intestinal epithelial cells [48]. Anoikis resistance is a hallmark of malignant cells and is believed to play a role in the survival of circulating tumor cells. Tumor cells could develop a mechanism to escape anoikis, which could then contribute to their metastatic behavior [40]. Our results showed that the anti-apoptotic signaling mediators Erk1/2 and p38 MAPK were constitutively activated in PTC cells that were detached from the ECM, while the levels of AKT and GSK-3β activation were decreased. However, the regulation of CHK2 had no effect on the activation of these kinases. We found that the level of p-PRAS40, a downstream regulator of the P13K-AKT pathway, was reduced when cells were detached from the ECM. The silencing of CHK2 was associated with increased p-PRAS in PTC cells following their detachment from the ECM. We speculated that the inhibition of CHK2 increased p-PRAS40 and prevented anoikis in PTC cells. Thus, our study identified the CHK2/PRAS40 pathway as a novel regulator of anoikis in PTC cells. Additionally, the results showed that the forced inhibition of p53 expression did not rescue BCPAP cells from anoikis, which indicates that CHK2 triggers this process in a p53-independent manner.

The current study demonstrated that CHK2 plays a role in the modulation of the progression and metastasis of PTC cells through the mediation of anoikis via the regulation of PRAS40. The expression levels of CHK2 and p-CHK2 were significantly reduced in LN mets, which indicates that CHK2 expression levels may be negatively associated with the survival of circulating tumor cells and metastatic behavior.

This work was supported by the grants from the CAMS Innovation Fund for Medical Sciences (CIFMS, 2016-I2M-3-005).

The authors declare no conflicts of interest.