Many tumor cells predominantly generate energy through high rates of glycolysis. Pyruvate kinase M2 (PKM2) has been identified as being necessary for aerobic glycolysis.
The aim of this study was to investigate the expression pattern of PKM2 in papillary thyroid cancer (PTC) and the possible role of PKM2 in PTC.
The expression of PKM2 in human thyroid tissues was examined using immunocytochemistry. PKM2 expression in PTC correlated with clinicopathological features and the BRAF mutation. PTC cells were transfected with small interfering RNA against PKM2. PKM2 expression in cells was analyzed by Western blotting and real-time RT-PCR. Cell growth was evaluated both in vitro and in vivo. Lactate and ATP production and glucose consumption by cells were determined using commercial assay kits.
PKM2 was aberrantly overexpressed in PTC. This overexpression was associated with poor clinicopathological features including advanced tumor stages and lymph node metastasis. More intensive immunostaining of PKM2 was detected in PTCs harboring the BRAF mutation. Specific small interfering RNA against PKM2 in PTC cell lines retarded cell growth both in vitro and in xenograft mouse models. PKM2 knockdown also reduced lactate and ATP production and glucose consumption by PTC cells.
We conclude that overexpression of PKM2 provides a selective growth advantage for PTC cells through activation of glycolysis. Aberrant PKM2 overexpression may serve as a novel biomarker and a potential treatment target for PTC. The BRAF mutation may contribute to alterations in the expression pattern of glycolytic enzymes such as PKM2.
The high proliferative rate of tumor cells leads to metabolic needs distinct from those of normal cells. Before the discovery of the first oncogene, Otto Warburg and colleagues (1) demonstrated that many tumors were highly glycolytic compared with their normal counterparts. Even in the presence of oxygen, cancer cells predominantly use aerobic glycolysis with reduced mitochondrial oxidative phosphorylation for glucose metabolism. This phenomenon was subsequently termed the Warburg effect. The Warburg effect leads to the elevation in glucose consumption of cancer cells, which is now recognized to support the energetic requirements for proliferation, and provides catabolic intermediates to fuel lipid and nucleic acid biosynthesis (2, 3).
Several contributing factors have been proposed to explain the Warburg effect. It has been noted that aerobic glycolysis and the enhanced production of lactate in tumor cells, despite long-held beliefs, is not usually due to mitochondrial defects in oxidative phosphorylation (4). In 2008, Christofk et al (5) demonstrated that the glycolytic enzyme pyruvate kinase might play an important role in the origins of the Warburg effect. They found that expression of the M2 isoform of pyruvate kinase (PKM2) is necessary for aerobic glycolysis and that this metabolic phenotype provides a selective growth advantage for tumor cells in vivo (5). This finding is a milestone for cancer research, because it has provided insight into the cancer metabolic program. Recent studies have shown that PKM2, previously thought to be expressed primarily during development, is aberrantly overexpressed in various kinds of human cancers (5–12). This overexpression of PKM2 is involved in promoting proliferation and migration of some types of cancer cells (2, 5, 8).
Thyroid cancer is the most common endocrine malignancy and accounts for most endocrine cancer–related deaths each year. The incidence of thyroid cancer (mainly differentiated) is rising rapidly in many areas of the world, including China. This phenomenon is mainly due to an increase in the papillary histotype (13–18). Much effort has been made to understand the mechanisms behind the tumorigenesis of papillary thyroid cancer (PTC). A major effort is the targeting molecular alterations in signaling pathways, particularly the Ras → Raf → MAPK kinase → MAPK/ERK signaling pathway, which are commonly overactivated in PTC (19). Among these molecular alterations, the B-type Raf kinase (BRAF) mutation is regarded as the most prevalent and important genetic event. The BRAF mutation is PTC specific, occurring in about half of PTC cases on average (20) and is continually being detected at a high frequency (21). A number of studies have observed that this mutation is associated with the aggressiveness of PTC (20). There has been explosive progress in PTC research since the discovery of the BRAF mutation.
However, to date, we know little about the role of PKM2 in PTC. Given that PTC cells exhibit a growth advantage over noncancerous cells, we hypothesized that a PKM2-induced alteration in glucose metabolism enables PTC cells to sustain rapid production of energy for cell proliferation. To test this, we examined the expression pattern of PKM2 in PTC and the association between PKM2 expression and aggressive features of PTC as well as the BRAF mutation. In addition, we determined the involvement of PKM2 in regulating the glucose metabolic pathway in PTC cells.
Materials and Methods
Human thyroid tissues and metastatic lymph nodes tissues of PTC
Human thyroid tissues were initially retrieved from the archived formalin-fixed paraffin-embedded tissue blocks in the Department of Pathology, Shenyang Feng Tian Hospital (Shenyang, Liaoning, People's Republic of China). Samples include 80 classic PTCs (patient age, 17–72 years), 46 PTC-matched normal thyroid tissues (patient age, 17–69 years) and 40 benign nodules (patient age: 18–76 years). To assess the level of PKM2 expression in metastatic lesions of PTC, 20 metastatic lymph node blocks from 16 of the 80 patients with PTCs were also included in this study. Informed consent from individual patients was not necessary because all data were made anonymous.
Immunohistochemistry (IHC)
For IHC, 4-μm sections were cut from paraffin-embedded tissue blocks. We used a rabbit monoclonal antibody against PKM2 (Cell Signaling Technology, Inc) together with a highly sensitive and specific polymer detection system using horseradish peroxidase (Santa Cruz Biotechnology). The immunostaining signal was developed using a permanent brown chromogenic substrate system (Sigma-Aldrich). Finally, nuclei were counterstained with hematoxylin for 5 minutes.
The staining of PKM2 was assessed using the Remmele immunoreactive score (IRS) (22) by multiplying the level of staining intensity (0–3 points: absent, weakly, intermediate, and strong) with the percentage of positive tumor cells (0–4 points: cutoffs: 0%, <10%, 11%–50%, 51%–80%, and >80%). The staining intensity was evaluated according to the following scale: negative (0 points), weakly positive (1 points), and positive (≥2 points).
Primary cell culture
PTC tissue was obtained as surgical waste from 12 patients undergoing thyroidectomies for the treatment, stored in sterilized precooled PBS, and transported to the laboratory at 4°C. The informed consent was obtained from all the 12 participants involved. The tissue was trimmed of connective tissue, finely minced, and then suspended in PBS. The suspension was filtered through a 100-μm cell strainer (BD Biosciences, San Jose, California), resuspended in PBS with 0.5% dispase and 0.2% collagenase (Roche Applied Science), and incubated for 30 min at 37°C with a magnetic stirrer. The digested mixture was again filtered through a 100-μm cell strainer, and the supernatant was collected. The supernatant was centrifuged at 1000 × g for 10 minutes. Cells were suspended in RPMI 1640 medium and seeded onto 6-cm plates overnight at 37°C. These were then washed to remove any remaining blood cells, the medium was changed, and the cells were grown at 37°C in RPMI 1640 medium with 10% fetal bovine serum (Invitrogen).
DNA isolation
Paraffin-embedded thyroid samples from patients were microdissected, and DNA was isolated as described previously (23). In brief, tissues dissected from paraffin-embedded specimens were treated for 8 hours at room temperature with xylene, followed by digestion with 1% sodium dodecyl sulfate and 0.5 mg/mL proteinase K at 48°C for 48 hours. To facilitate the digestion, a midinterval addition of a spiking aliquot of concentrated sodium dodecyl sulfate–proteinase K was added to the samples. DNA was subsequently isolated from the digested tissues by standard phenol-chloroform extraction and ethanol precipitation procedures.
Detection of BRAF mutation
The BRAF mutation was detected in thyroid samples. Because the T1799A transversion mutation is virtually the only BRAF mutation that has been described in PTC with a high prevalence in previous studies (20), we examined this particular mutation in tissues in the present study. The BRAF mutation was analyzed using genomic DNA by direct sequencing. For direct DNA sequencing, exon 15 of the BRAF gene was amplified by PCR, followed by a Big Dye terminator cycle sequencing reaction and sequence reading on an ABI PRISM 3730 genetic analyzer (Applied Biosystems). The PCR protocol and primers for exon 15 of the BRAF gene were as described previously (23).
Human thyroid cell lines
The PTC cell line BCPAP was purchased from The DSMZ-Deutsche Sammlung von Mikroorganismen und Zellkulturen GmbH (German Collection of Microorganisms and Cell Cultures; Braunschweig, Germany). The PTC cell line K1 and a noncancerous thyroid cell line Nthy-ori 3–1 were purchased from The Health Protection Agency Culture Collections (Salisbury, United Kingdom). The Nthy-ori 3–1 is a normal thyrocyte cell line that has been immortalized using the simian virus 40 large T antigen. These cells were grown at 37°C in RPMI 1640 medium with 10% fetal bovine serum (Invitrogen).
Generation of small interfering (si) RNA knockdown stable cell lines
siRNA-mediated knockdown was performed with pRNAT-U6.1/Neo (SD1211; GeneScript, Inc) siRNA expression vector. siRNA expression plasmids were constructed according to the manufacturer's instructions. siRNA primers were designed by siRNA Target Finder and siRNA Construct Builder (GeneScript, Inc). The nucleotide sequence of the inserted PKM2 siRNA was 5′-GGATCCCGTGTGACGAG AACATCCTGTTCAAGAGACAGGATGTTCTCGTCACACTTTTTTCCAAAAGCTT. The negative control vector pRNA-U6.1/Neo/CTL (SD1801; GeneScript, Inc) was used as a control. Purified plasmid DNA was confirmed by DNA sequencing and transfected into the immortalized thyrocytes using Lipofectamine 2000 (Invitrogen) transfection reagents according to the manufacturer's protocol. Stable cell clones were obtained with neomycin selection and analyzed by immunoblotting.
RNA extraction and real-time quantitative RT-PCR analysis
Total RNA was isolated from cells using TRIzol reagent according to the instructions of the manufacturer (Invitrogen), and 2 μg of RNA was processed directly to cDNA using a reverse transcription kit (Promega), according to the manufacturer's instructions. Amplification reactions were performed in a 15-μl volume with 0.2 μl of SYBR Green (Bio-Rad). These reactions were performed in triplicate using a BioRad iCycler (Bio-Rad). β-Actin was used as an internal control. The primers used were as follows: PKM2, 5′-CCATTACCAGCGACCCCACAG-3′ (forward) and 5′-GGGCACGTGGGCGGTATCT-3′ (reverse); and β-Actin, 5′-GATCATTGCTCCTCCTGAGC-3′ (forward) and 5′-AC TCCTGCTTGCTGATCCAC-3′ (reverse). The specificity of real-time quantitative PCR was verified by melting curve analysis and agarose gel electrophoresis.
Western blotting
Cells were washed twice with ice-cold PBS and solubilized in 1% Triton lysis buffer (1% Triton X-100, 50 mM Tris-Cl [pH 7.4], 150 mM NaCl, 10 mM EDTA, 100 mM NaF, 1 mM Na3VO4, 1 mM phenylmethylsulfonyl fluoride, and 2 μg/mL aprotinin) on ice and then was quantified using the Lowry method. Cellular proteins (50 μg) were subjected to 10% SDS-PAGE and transferred onto nitrocellulose membranes (Immobilon-P; Millipore). The membranes were probed with anti-PKM2 antibody. Antigen-antibody complexes were visualized using horseradish peroxidase–conjugated anti-rabbit (Santa Cruz Biotechnology) IgG antibody and the ECL Western Blotting Analysis System (Pierce).
3-(4,5-Dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) and colony formation assay
For cell growth analysis, an equal number of cells were seeded in 48-well plates. The total number of cells was measured every day by the MTT assay according to the manufacturer's protocol (Roche Applied Science). For colony formation analysis, a total of 500 cells/well were seeded into 6-well plate in triplicate. Cells were cultured for 14 days, when most colonies contained more than 50 cells. Cells were washed with PBS twice, and 1 mL of paraformaldehyde was added in each well and incubated for 10 minutes for fixation. Then cells were stained with 500 μL of Giemsa for 20 minutes. The number of colonies in each group was counted.
Xenografts
The BCPAP and K1 xenografts were produced on 6-week-old female athymic nude nu/nu mice (Vital River Laboratories) by sc injection of 2 × 106 cells in 0.2 mL of PBS into the flank of each mouse (10 mice/group). Tumor formation was assessed every 5 days. Tumor volume was calculated according to the formula V = (a × b2)/2 (where a is the largest superficial diameter and b is the smallest superficial diameter).
Measurement of lactate concentration, glucose consumption, and ATP production
After 48 hours of incubation, lactate, glucose, and ATP levels in the culture media were measured using commercially available assay kits (purchased from BioVision, Sigma-Aldrich, and Roche Applied Science, respectively). Results were corrected for the final cell count number.
Statistical analysis
For the IHC data obtained from human thyroid tissues, we used the χ2 test, ANOVA, or t test to compare experimental groups. A value of P < .05 was considered statistically significant.
These experiments in vitro were performed at least 2 or 3 times. Most of the measurements were conducted in triplicate and some in duplicate. Data were compared using the t test. We used the t test to compare data from in vivo studies as well. Significance was also defined as P < .05. Unless indicated, data shown in the figures are representatives.
Ethics
This study was approved by the ethics committees of the First Hospital of China Medical University. All of the animal study procedures were in accordance with the National Institutes of Health Guide for the Care and Use of Laboratory Animals and were performed according to the institutional ethical guidelines.
Results
PKM2 expression in human thyroid tissues and PTC metastatic lymph node tissues
To explore whether PKM2 expression is aberrantly up-regulated in PTC, we first examined PKM2 protein levels in human thyroid tissues. In the 166 human thyroid samples, positive staining of PKM2 was detected in 91% (73 of 80) of classic PTCs, none of the 46 PTC-matched normal thyroid samples and 10% (4 of 40) of benign nodules (P < .001). A weak positive staining was observed in 8% (6 of 80) of classic PTCs, 2% (1 of 46) of normal tissues, and 10% (4 of 40) of benign nodules (P < .001). The mean IRS was significantly higher in PTCs than in normal thyroid samples and benign nodules (P < .001) (Table 1 and Figure 1).
Immunoreactive Staining of PKM2 In Human Thyroid Tissues
| Tissue type | No. | Staining of PKM2, n (%) | IRS (mean ± SD) | ||
|---|---|---|---|---|---|
| Negative | Weakly Positive | Positive | |||
| Normal | 46 | 45 (98) | 1 (2) | 0 (0) | 0.1 ± 0.0 |
| Benign nodule | 40 | 32 (80) | 4 (10) | 4 (10) | 0.6 ± 0.2 |
| PTC | 80 | 1 (1) | 6 (8) | 73 (91)a | 7.4 ± 4.0a |
| Tissue type | No. | Staining of PKM2, n (%) | IRS (mean ± SD) | ||
|---|---|---|---|---|---|
| Negative | Weakly Positive | Positive | |||
| Normal | 46 | 45 (98) | 1 (2) | 0 (0) | 0.1 ± 0.0 |
| Benign nodule | 40 | 32 (80) | 4 (10) | 4 (10) | 0.6 ± 0.2 |
| PTC | 80 | 1 (1) | 6 (8) | 73 (91)a | 7.4 ± 4.0a |
Compared with normal and goiter tissues, P < .001.
Immunoreactive Staining of PKM2 In Human Thyroid Tissues
| Tissue type | No. | Staining of PKM2, n (%) | IRS (mean ± SD) | ||
|---|---|---|---|---|---|
| Negative | Weakly Positive | Positive | |||
| Normal | 46 | 45 (98) | 1 (2) | 0 (0) | 0.1 ± 0.0 |
| Benign nodule | 40 | 32 (80) | 4 (10) | 4 (10) | 0.6 ± 0.2 |
| PTC | 80 | 1 (1) | 6 (8) | 73 (91)a | 7.4 ± 4.0a |
| Tissue type | No. | Staining of PKM2, n (%) | IRS (mean ± SD) | ||
|---|---|---|---|---|---|
| Negative | Weakly Positive | Positive | |||
| Normal | 46 | 45 (98) | 1 (2) | 0 (0) | 0.1 ± 0.0 |
| Benign nodule | 40 | 32 (80) | 4 (10) | 4 (10) | 0.6 ± 0.2 |
| PTC | 80 | 1 (1) | 6 (8) | 73 (91)a | 7.4 ± 4.0a |
Compared with normal and goiter tissues, P < .001.
PKM2 immunohistochemistry (×200) of human thyroid tissues and metastatic lymph nodes tissues of PTC. A, PTC: PKM2-positive group, IRS = 12. B, PTC: PKM2–weakly positive group, IRS = 3. C, Normal thyroid tissue: PKM2-negative group, IRS = 0. D, Metastatic lymph node tissue of PTC: PKM2-positive, IRS = 6.
To assess the level of PKM2 expression in metastatic lesions of PTC, 20 metastatic lymph nodes blocks from 16 patients with PTCs were examined. All of these 16 patients had positive staining of PKM2 in thyroid tumors. Of the metastatic nodes, 70% (14 of 20) and 30% (6 of 20) exhibited positive staining and weak positive staining of PKM2, respectively (Figure 1). The IRS in metastatic nodes correlated with that in the corresponding thyroid tumors (r = 0.76, P = .000).
The T1799A BRAF mutation in PTCs
We isolated DNA from paraffin-embedded thyroid samples and performed PCR amplification followed by direct DNA sequencing to detect the T1799A BRAF mutation in this set of thyroid tissue samples. Not surprisingly, the BRAF mutation was not detected in normal thyroid tissues and benign nodules, whereas 46 of 80 PTC samples harbored the T1799A BRAF mutation. Thus, the overall prevalence of the mutation was 58%. We then analyzed the relationship between the BRAF mutation and clinicopathologic features of the tumors. We did not see an age or sex preference of the BRAF mutation in these PTCs. Unlike many other studies, a significant association between lymph nodes metastasis or advanced tumor stages (III/IV) and the BRAF mutation status was not detected in the present study. One possible reason is the relatively small number of cases studied.
Correlation of PKM2 expression in human PTC with clinical aggressiveness and the presence of a BRAF mutation
Clinicopathologic characteristics of the PTC samples, including patient age, sex, lymph node metastasis, and TNM (tumor, node, metastasis) stage, were recorded and examined for a possible correlation with PKM2 expression. Positive staining of PKM2 was seen in all PTCs with lymph nodes metastasis or advanced tumor stages (III/IV). The IRS values significantly correlated with the aggressive features of PTC, including lymph nodes metastasis and advanced tumor stages (III/IV) (Table 2).
Correlation of PKM2 Expression in Human PTC With Clinical Aggressiveness and the BRAF Mutation
| Clinical Features | No. | Staining of PKM2, n (%) | IRS (mean ± SD) | P Valuea | ||
|---|---|---|---|---|---|---|
| Negative | Weakly Positive | Positive | ||||
| Age | .08 | |||||
| <45 y | 41 | 1 (2) | 4 (10) | 36 (88) | 6.6 ± 3.6 | |
| ≥45 y | 39 | 0 (0) | 2 (8) | 37 (95) | 8.1 ± 4.0 | |
| Sex | .14 | |||||
| Male | 20 | 0 (0) | 3 (15) | 17 (85) | 6.3 ± 3.7 | |
| Female | 60 | 1 (2) | 3 (5) | 56 (93) | 7.7 ± 3.8 | |
| Multifocality | .71 | |||||
| No | 63 | 1 (2) | 4 (6) | 58 (92) | 7.5 ± 3.9 | |
| Yes | 17 | 0 (0) | 2 (12) | 15 (88) | 7.9 ± 3.7 | |
| Lymph node metastasis | .003 | |||||
| No | 41 | 1 (2) | 6 (15) | 34 (83) | 6.1 ± 3.8 | |
| Yes | 39 | 0 (0) | 0 (0) | 39 (100) | 8.4 ± 3.5 | |
| Stages | .01 | |||||
| I/II | 62 | 1 (2) | 6 (10) | 55 (88) | 6.7 ± 3.7 | |
| III/IV | 18 | 0 (0) | 0 (0) | 18 (100) | 9.4 ± 3.8 | |
| BRAF gene | <.001 | |||||
| Wild type | 34 | 1 (3) | 5 (15) | 28 (82) | 4.5 ± 2.5 | |
| T1799A mutation | 46 | 0 (0) | 1 (2) | 45 (98) | 9.5 ± 3.2 | |
| Clinical Features | No. | Staining of PKM2, n (%) | IRS (mean ± SD) | P Valuea | ||
|---|---|---|---|---|---|---|
| Negative | Weakly Positive | Positive | ||||
| Age | .08 | |||||
| <45 y | 41 | 1 (2) | 4 (10) | 36 (88) | 6.6 ± 3.6 | |
| ≥45 y | 39 | 0 (0) | 2 (8) | 37 (95) | 8.1 ± 4.0 | |
| Sex | .14 | |||||
| Male | 20 | 0 (0) | 3 (15) | 17 (85) | 6.3 ± 3.7 | |
| Female | 60 | 1 (2) | 3 (5) | 56 (93) | 7.7 ± 3.8 | |
| Multifocality | .71 | |||||
| No | 63 | 1 (2) | 4 (6) | 58 (92) | 7.5 ± 3.9 | |
| Yes | 17 | 0 (0) | 2 (12) | 15 (88) | 7.9 ± 3.7 | |
| Lymph node metastasis | .003 | |||||
| No | 41 | 1 (2) | 6 (15) | 34 (83) | 6.1 ± 3.8 | |
| Yes | 39 | 0 (0) | 0 (0) | 39 (100) | 8.4 ± 3.5 | |
| Stages | .01 | |||||
| I/II | 62 | 1 (2) | 6 (10) | 55 (88) | 6.7 ± 3.7 | |
| III/IV | 18 | 0 (0) | 0 (0) | 18 (100) | 9.4 ± 3.8 | |
| BRAF gene | <.001 | |||||
| Wild type | 34 | 1 (3) | 5 (15) | 28 (82) | 4.5 ± 2.5 | |
| T1799A mutation | 46 | 0 (0) | 1 (2) | 45 (98) | 9.5 ± 3.2 | |
Comparisons between the IRS values in two groups.
Correlation of PKM2 Expression in Human PTC With Clinical Aggressiveness and the BRAF Mutation
| Clinical Features | No. | Staining of PKM2, n (%) | IRS (mean ± SD) | P Valuea | ||
|---|---|---|---|---|---|---|
| Negative | Weakly Positive | Positive | ||||
| Age | .08 | |||||
| <45 y | 41 | 1 (2) | 4 (10) | 36 (88) | 6.6 ± 3.6 | |
| ≥45 y | 39 | 0 (0) | 2 (8) | 37 (95) | 8.1 ± 4.0 | |
| Sex | .14 | |||||
| Male | 20 | 0 (0) | 3 (15) | 17 (85) | 6.3 ± 3.7 | |
| Female | 60 | 1 (2) | 3 (5) | 56 (93) | 7.7 ± 3.8 | |
| Multifocality | .71 | |||||
| No | 63 | 1 (2) | 4 (6) | 58 (92) | 7.5 ± 3.9 | |
| Yes | 17 | 0 (0) | 2 (12) | 15 (88) | 7.9 ± 3.7 | |
| Lymph node metastasis | .003 | |||||
| No | 41 | 1 (2) | 6 (15) | 34 (83) | 6.1 ± 3.8 | |
| Yes | 39 | 0 (0) | 0 (0) | 39 (100) | 8.4 ± 3.5 | |
| Stages | .01 | |||||
| I/II | 62 | 1 (2) | 6 (10) | 55 (88) | 6.7 ± 3.7 | |
| III/IV | 18 | 0 (0) | 0 (0) | 18 (100) | 9.4 ± 3.8 | |
| BRAF gene | <.001 | |||||
| Wild type | 34 | 1 (3) | 5 (15) | 28 (82) | 4.5 ± 2.5 | |
| T1799A mutation | 46 | 0 (0) | 1 (2) | 45 (98) | 9.5 ± 3.2 | |
| Clinical Features | No. | Staining of PKM2, n (%) | IRS (mean ± SD) | P Valuea | ||
|---|---|---|---|---|---|---|
| Negative | Weakly Positive | Positive | ||||
| Age | .08 | |||||
| <45 y | 41 | 1 (2) | 4 (10) | 36 (88) | 6.6 ± 3.6 | |
| ≥45 y | 39 | 0 (0) | 2 (8) | 37 (95) | 8.1 ± 4.0 | |
| Sex | .14 | |||||
| Male | 20 | 0 (0) | 3 (15) | 17 (85) | 6.3 ± 3.7 | |
| Female | 60 | 1 (2) | 3 (5) | 56 (93) | 7.7 ± 3.8 | |
| Multifocality | .71 | |||||
| No | 63 | 1 (2) | 4 (6) | 58 (92) | 7.5 ± 3.9 | |
| Yes | 17 | 0 (0) | 2 (12) | 15 (88) | 7.9 ± 3.7 | |
| Lymph node metastasis | .003 | |||||
| No | 41 | 1 (2) | 6 (15) | 34 (83) | 6.1 ± 3.8 | |
| Yes | 39 | 0 (0) | 0 (0) | 39 (100) | 8.4 ± 3.5 | |
| Stages | .01 | |||||
| I/II | 62 | 1 (2) | 6 (10) | 55 (88) | 6.7 ± 3.7 | |
| III/IV | 18 | 0 (0) | 0 (0) | 18 (100) | 9.4 ± 3.8 | |
| BRAF gene | <.001 | |||||
| Wild type | 34 | 1 (3) | 5 (15) | 28 (82) | 4.5 ± 2.5 | |
| T1799A mutation | 46 | 0 (0) | 1 (2) | 45 (98) | 9.5 ± 3.2 | |
Comparisons between the IRS values in two groups.
We also determined the association between the PKM2 expression and the T1799A BRAF mutation. The IRS scores in PTCs harboring this BRAF mutation were significantly higher than in PTCs with the wild-type BRAF gene (9.5 ± 3.2 vs 4.5 ± 2.5, P < .001), showing a clear association between PKM2 protein overexpression in IHC and the occurrence of the T1799A BRAF mutation (Table 2).
PKM2 expression in human thyroid cell lines and primary cultured human PTC cells
We then examined PKM2 expression in 3 human thyroid cell lines, including 2 PTC cell lines, BCPAP and K1, as well as a noncancerous thyroid cell line Nthy-ori 3–1. Aberrant overexpression of PKM2 protein and mRNA were detected in both PTC cell lines (Figure 2).
PKM2 expression in human thyroid cells. A and B, Aberrant overexpression of PKM2 protein (A) and mRNA (B) was detected in PTC cell lines BCPAP and K1 compared to the Nthy-ori 3-1 cells (*, P < .05), C, Overexpression of PKM2 protein was also detected in primary cultured human PTC cells from 12 patients. D and E, Significant knockdown of PKM2 expression was achieved by siRNA in PTC cell lines BCPAP and K1 (*, P, < .05). siRNA-mediated knockdown was conducted with pRNAT-U6.1/Neo siRNA expression vector. The negative control vector pRNA-U6.1/Neo/CTL was used as a control. BRAF(+), with T1799A BRAF mutation; BRAF(−), with wild-type BRAF.
Primary cultured PTC cells from 12 cases were examined for PKM2 expression as well. Overexpression of PKM2 protein was detected in all cases, which was more evident in PTCs harboring the BRAF mutation than in PTCs with the wild-type BRAF gene (Figure 2). However, statistical analysis was not performed because of the small number of cases included.
Knockdown of PKM2 by siRNA in human PTC cell lines BCPAP and K1
Given that PKM2 is preferably expressed in PTC cells, we assessed the importance of this observation on cell proliferation via siRNA knockdown of PKM2. We expressed siRNA from the PKM2 sequence in the PTC cell lines BCPAP and K1. Cell clones stably expressing PKM2 siRNA were generated and confirmed for PKM2 knockdown by immunoblotting and real-time PCR. As shown in Figure 2, expression of PKM2 protein assayed by immunoblotting was hardly detectable, and the PKM2 mRNA level was substantially reduced by 83% and 78% in the PKM2-siRNA expressing BCPAP and K1 cells, respectively. PKM2 expression remained unaltered in both PTC cell lines expressing a control siRNA.
Inhibition of cell growth by siRNA knockdown of PKM2 in human PTC cell lines BCPAP and K1
To assess tumor cell growth, we used MTT and colony formation assays. As shown in Figure 3, cells expressing PKM2 siRNA cells exhibited less proliferation than the parental cell line and cells expressing the control siRNA. The growth of BCPAP and K1 cells expressing PKM2 siRNA decreased by 51% and 43%, respectively, after 48 hours of incubation compared with that of the control cells (P < .05) (Figure 3A). Consistently, PKM2 knockdown in these cells prevented colony formation, indicating that anchorage-dependent growth was restored (Figure 3B).
Inhibition of cell growth by siRNA knockdown of PKM2 in human PTC cell lines BCPAP and K1. A, Using the MTT, we observed that the growth of BCPAP and K1 cells with siRNA knockdown of PKM2 was inhibited by 51% and 43% after 48 hours of incubation, respectively, compared with that in the control cells (P < .05). B, Colony formation of BCPAP and K1 cell lines was reduced significantly by siRNA knockdown of PKM2. C and D, The in vivo tumorigenic capacity of BCPAP and K1 cell lines was reduced significantly by siRNA knockdown of PKM2. E, Tumors were dissected and measured at 25 days (BCPAP) and 35 days (K1) after injection. For all matched xenografts, PKM2-siPKM2 tumors were significantly smaller than the controls (P < .05).
Inhibition of cell growth by siRNA knockdown of PKM2 in xenograft mouse models
The in vivo tumorigenic capacity of BCPAP and K1 PKM2-siPKM2 cells and control cells was assessed by sc injection of cells in athymic nude mice. Tumors were allowed to grow, and mice were killed 25 (BCPAP) and 35 (K1) days after injection. Six of 40 mice did not form tumors. For all matched xenografts, PKM2-siPKM2 tumors were significantly smaller than those of the controls (Figure 3, C–E).
siRNA knockdown of PKM2 reduced lactate and ATP production and glucose consumption in human PTC cell lines BCPAP and K1
Because PKM2 is a key enzyme for aerobic glycolysis, alteration of its expression in PTC cells should affect glucose metabolism. To test this hypothesis, we compared lactate and ATP production and glucose consumption between PKM2 siRNA knockdown cells and control cells after a 48-hour incubation. When PKM2 was knocked down by siRNA, the ability of the cell to produce lactate and ATP was dramatically decreased in both BCPAP cells (by 67% and 48% at 48 hours, respectively) and K1 cells (by 50% and 44% at 48 hours, respectively). Meanwhile, PKM2 knockdown led to significantly reduced glucose consumption in BCPAP cells (by 38%) and K1 cells (by 37%) after 48 hours of incubation (Figure 4).
Reduced lactate and ATP production and glucose consumption by siRNA knockdown of PKM2 in human PTC cell lines BCPAP and K1. After 48 hours of incubation, PKM2 knockdown led to significantly reduced lactate and ATP production, as well as glucose consumption, in BCPAP and K1 cells compared with that in control cells (P < .05).
Discussion
Although many efforts have been made to uncover the mechanisms underlying PTC tumorigenesis, most of these studies have focused on the abnormal activation of the MAPK pathway. In this study, we tested the hypothesis that PKM2, a key enzyme that catalyzes the rate-limiting final step of glycolysis and promotes glucose uptake and lactate production, is overexpressed in PTC and plays an important role in the development of PTC. This finding identifies a new molecular player in the development of PTC.
This study on human thyroid samples revealed a highly specific aberrant overexpression of PKM2 based on IHC results of positive rate and IRS score in PTCs compared with those in normal thyroid tissues and benign nodules. Previous studies have shown PKM2 up-regulation in a number of malignancies originating from other organs, including lung, stomach, colon, and brain (5–12). Taking these results together with our findings, we surmise that the aberrant up-regulation of PKM2 is not specific to one certain type of cancer. Aberrant overexpression of PKM2 indicates that alterations in glucose metabolism in cells may be involved in the pathogenesis of PTC. A recent study has found that the glucose transporter 1 (GLUT1) is overexpressed in nearly 70% samples of PTC (24). Given that GLUT1 facilitates the extensive glucose uptake of cancer cells, this study provides another hint that the glucose metabolism pathway is of importance in the carcinogenesis of PTC. Moreover, the specifically higher expression of PKM2 in PTC than in normal cells suggests a potential for PKM2 as a biomarker for diagnosing indeterminate thyroid nodules. However, additional studies using follicular thyroid tumors are needed before a final conclusion can be drawn.
We also demonstrated that PKM2 overexpression is associated with the aggressive clinicopathological characteristics of PTC, including lymph node metastasis and advanced cancer stages. These aggressive features are major risk factors for recurrence, loss of response to treatment, and disease-specific death. Unfortunately, we did not obtain long-term follow-up data from patients with PTCs whose postoperative thyroid tissues were used in this study. Therefore, we could not reveal the direct linkage between PKM2 expression and disease prognosis. Nevertheless, PKM2 overexpression represents an adverse prognostic factor and a novel indicator of the progression and aggressiveness of PTC. In fact, the correlation of PKM2 expression with a shorter overall survival has been proven in some other types of malignancy, such as lung cancer and signet ring cell gastric cancer (6, 10). Another interesting issue worthy of being investigated is the correlation of expression grade of PKM2 with fluorine-18 fluorodeoxyglucose uptake on a positron emission tomogram. However, we were not able to explore this correlation due to lack of preoperative fluorine-18 fluorodeoxyglucose-positron emission tomography imaging.
The activating mutation of the BRAF gene is of particular importance in PTC (25). This mutation is regarded as a major cause of aberrant activation of the MAPK pathway in PTC (26–28). Moreover, it has been shown to down-regulate the expression of major tumor suppressor and thyroid iodide-metabolizing genes, to up-regulate cancer-promoting molecules, such as vascular endothelial growth factor, matrix metalloproteinases, nuclear transcription factor κB, and c-Met, and to correlate with the aberrant methylation of various genes (29, 30). Therefore, this mutation directly promotes tumorigenesis and the metastatic potential of PTC. This study provides the first evidence of a correlation between the BRAF mutation and PKM2 expression. Combined with the findings of Grabellus et al (24) showing that BRAF mutation associates with GLUT1 overexpression in PTC, we propose that one mechanism through which the BRAF mutation causes PTC growth and progression involves the glucose metabolic pathway. This enriches what we know regarding the molecular contribution of the BRAF mutation to PTC progression. It may take time for the BRAF mutation to affect the glucose metabolic pathway, because our data suggested that acute changes in the MAPK pathway signaling had no direct effect on PKM2 expression (see Supplemental Results and Supplemental Figure 1 published on The Endocrine Society's Journals Online web site at http://jcem.endojournals.org).
Subsequent experiments in our study confirmed the up-regulation of PKM2 expression in PTC cells and demonstrated that PKM2 was necessary for maintaining rapid cell proliferation and altering glucose metabolism. This result is consistent with previous findings in other types of cancer. However, much needs to be done before we completely understand the function of PKM2 in PTC. As indicated in recent studies, PKM2 also has numerous nonmetabolic functions: it allows cancer cells to mount an antioxidant response and thereby support cell survival under acute oxidative stress (31); it induces epidermal growth factor receptor–dependent β-catenin transactivation, which leads to cell proliferation and tumorigenesis (32); it phosphorylates histone H3 and promotes gene transcription and tumorigenesis (33); and it affects cell survival by regulating Bcl-xL at the transcriptional level (34). Although these nonmetabolic functions may not universally apply to all kinds of cancer including PTC, the function of PKM2 in PTC must be very complex and needs to be further elucidated. In addition, the mechanisms of the regulation of PKM2 expression specifically in PTC should be studied.
Finally, given our findings that aberrant overexpression of PKM2 is common in PTC and that silencing of PKM2 can inhibit tumor cell proliferation via inhibition of glycolysis signaling, targeting of glycolysis through PKM2 is an attractive means for treatment of PTC, particularly for those patients who do not respond to traditional therapeutic interventions. This consideration has been especially encouraged by recent studies, in which orlistat, cyclosporin A, resveratrol, and shikonin were found to be capable of inhibiting cancer cell growth via down-regulation of PKM2 expression (35–38). Obviously, more studies are required to fully investigate this finding.
In conclusion, we have shown in this study that PKM2 is aberrantly up-regulated in PTC. PKM2 expression was associated with clinical aggressiveness and the presence of the BRAF mutation. Inhibition of PTC cell growth by silencing PKM2 expression was observed in vitro and in vivo. Down-regulation of PKM2 by siRNA also decreased lactate and ATP production and glucose consumption in vitro. The results of this study suggest that overexpression of PKM2 provides a selective growth advantage for PTC cells through activation of aerobic glycolysis and that aberrant PKM2 expression may serve as a useful biomarker as well as a potential treatment target for PTC. In addition, the BRAF mutation may lead to tumor aggressiveness via, at least in part, changing the expression of enzymes involved in glycolysis such as PKM2.
Acknowledgments
We thank pathologists Yan Wang, Yong Zhang, and Shuli Liu, for their generous help in characterizing and preparing tumor samples.
This work was supported by the National Natural Science Foundation, Beijing, China (Grants 30801120 and 81102471>), the Education Department Foundation of Liaoning Province, Shenyang, China (Grant 2008T204), and the Fund for Scientific Research of The First Hospital of China Medical University (Grant FSFH1202).
Authors' statement: The authors hereby confirm that neither the manuscript nor any part of it, except for abstracts of less than 400 words, has been published or is being considered for publication elsewhere. By signing this letter each of us acknowledges that he or she participated sufficiently in the work to take public responsibility for its content.
Disclosure Summary: The authors have nothing to disclose.
Abbreviations
- GLUT1
glucose transporter 1
- IHC
immunohistochemistry
- IRS
immunoreactive score
- MTT
3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide
- PKM2
pyruvate kinase M2
- PTC
papillary thyroid cancer
- si
small interfering.
References
