It remains to be investigated whether the aberrant methylation of DNA repair genes plays a pathogenic role in BRAF mutation-promoted tumorigenesis of papillary thyroid cancer (PTC).
It remains to be investigated whether the aberrant methylation of DNA repair genes plays a pathogenic role in BRAF mutation-promoted tumorigenesis of papillary thyroid cancer (PTC).
In the current study, the promoter methylation status of 23 DNA repair genes in relation to clinicopathologic characteristics and BRAF mutation was examined in PTC tumors using methylation-specific polymerase chain reaction.
Among the 38 PTC tumors examined, 3 of 23 DNA repair genes were hypermethylated, including the hMLH1 gene in 8 of 38 samples (21%), the PCNA gene in 5 of 38 samples (13%), and the OGG1 gene in 2 of 38 samples (5%). Methylation of these genes was also found in some thyroid cancer cell lines. Methylation of the hMLH1 gene in particular was found to be associated with lymph node metastasis of PTC (5 of 8 samples [63%] in the methylation group vs 3 of 30 samples [10%] in the nonmethylation group; P = .0049). Methylation of the hMLH1 gene was also found to be associated with the T1799A BRAF mutation in PTC (6 of 19 samples (32%) in the BRAF mutation-positive group vs 2 of 19 samples (11%) in the BRAF mutation-negative group; P = .042).
The data from the current study suggest that, as shown previously in colon cancer, aberrant methylation of the hMLH1 gene may play a role in BRAF mutation-promoted thyroid tumorigenesis. Cancer 2008. © 2008 American Cancer Society.
Thyroid cancer is the most common endocrine malignancy, with a rapidly increasing incidence reported in recent years.1–4 The most common type is papillary thyroid cancer (PTC), which accounts for >80% of all thyroid malignancies. Genetic and epigenetic alterations are the driving force behind thyroid tumorigenesis and progression. One of the major genetic alterations in thyroid cancer is the T1799A BRAF mutation, which occurs uniquely in PTCand PTC-derived anaplastic thyroid cancer, with a prevalence of approximately 45% in the former and 25% in the latter.5 Epigenetic alteration, particularly promoter DNA hypermethylation with the consequent gene silencing that commonly occurs in cancer,6 is also common in thyroid cancer.7 Aberrant methylation of several important genes has recently been shown to be associated with BRAF mutation in PTC cells, including the genes for tumor suppressor SLC5A8, the tissue inhibitor of metalloproteinase-3, death-associatedprotein kinase, the retinoic acid receptor β28, 9 and the thyroid-stimulating hormone receptor gene.10 These data suggest that, similar to genetic alterations, epigenetic alterations of major genes are also an important mechanism involved in thyroid tumorigenesis.
The BRAF mutation promotes tumorigenesis of PTC through aberrant activation of the RET/PTC → · · · → Ras → Raf → MEK → MAP kinase/ERK pathway (MAPK pathway). In comparison with other genetic alterations in this pathway, such as RET/PTC rearrangement (which is the second most common genetic alteration in PTC), BRAF mutation in particular promotes poorer clinicopathologic outcomes of PTC, such as increased extrathyroidal invasion, lymph node metastasis, advanced tumor stage, and tumor recurrence.11 This may be explained at least in part by the genetic instability that could preferentially be induced by BRAF mutation but not RET/PTC rearrangement.12 In colon cancer, BRAF mutation was also shown to be associated with genetic instability and, interestingly, the aberrant hypermethylation of certain DNA repair genes, such as the mismatch repair gene hMLH.13–16 DNA repair genes protect genetic integrity in normal cells by mismatch repair, nucleotide excision, base excision, and double-strand break repair.17–19 Aberrant methylation and hence silencing of the DNA repair genes could predispose cells to genetic instability. To our knowledge, it remains to be tested whether aberrant methylation of DNA repair genes is a mechanism in the BRAF mutation-promoted pathogenesis of thyroid tumors. In the current study, we examined the methylation status of a large number of DNA repair genes in human thyroid cancer cell lines and PTC tissues and analyzed its correlation with BRAF mutation and the clinicopathologic characteristics of PTC.
With approval of the hospital review board, conventional PTC tissues and clinicopathologic data were obtained from the files of patients who underwent total thyroidectomy at the No. 8 People's Hospital in Shenyang, China between 2001 and 2006. Thirty-eight cases of PTC were included. Clinicopathologic data, including patient age and sex, status of extrathyroidal invasion, lymph node metastasis, and tumor stage were collected retrospectively. Tumors were staged as per the staging system of the American Joint Committee on Cancer.
Five human thyroid cancer cell lines were used and were originally provided by the following investigators. The PTC cell line NPA, anaplastic thyroid cancer (ATC) cell lines DRO and ARO, and follicular thyroid cancer (FTC) cell line WRO were a gift from Dr. Guy J. F. Juillard (University of California at Los Angeles School of Medicine, Los Angeles, Calif) and the ATC cell line C643 was a gift from Dr. N. E. Heldin (University of Uppsala, Uppsala, Sweden). These cell lines were routinely cultured, as previously reported.10 ARO cells were treated with MEK inhibitor U0126 (Sigma Chemical Company, St. Louis, Mo) at 10 μM for 5 days in some experiments as indicated. The differentiated rat thyroid cell line PCCL3 with inducible expression of V600E BRAF was a gift from Dr. James A. Fagin (Memorial Sloan-Kettering Cancer Center, New York, NY). Cells were cultured as described previously.12 In some experiments, doxycycline (Sigma Chemical Company) at 1 μg/mL was used to treat cells for 6 days to induce BRAF V600E expression.
The histologic diagnosis of tumors was made and agreed upon by at least 2 experienced pathologists based on World Health Organization (WHO) criteria. Paraffin-embedded PTC samples were microdissected and DNA was isolated as per our standard protocols.20 Briefly, tissues dissected from paraffin-embedded specimens were first treated for 8 hours at room temperature with xylene to remove the paraffin. Thereafter, all the samples were subjected to digestion with 1% sodium dodecyl sulfate (SDS) and 0.5 mg/mL of proteinase K at 48°C for 48 hours. To facilitate the digestion, a mid-interval addition of a spiking dose of concentrated SDS-proteinase K was added to the sample tubes. DNA was subsequently isolated from the digested tissues by standard phenol-chloroform extraction and ethanol precipitation procedures. Cell line DNA was isolated by digesting cell pellets following a similar protocol.
DNA was treated with sodium bisulfite to convert cytosine to uracil, as described previously.9 Briefly, 2 μg of DNA in 20 μL of water containing 5 μg of salmon sperm DNA were denatured by incubation with 0.3 M NaOH at 50°C for 20 minutes, followed by incubation in 500 μL of a freshly prepared mixture containing 3 M sodium bisulfite and 10 mM hydroquinone at 70°C for 2 to 3 hours. DNA samples were purified with affinity chromatography columns (Wizard DNA Clean-Up System; Promega, Madison, Wis), treated with 0.3 M of NaOH for 10 minutes at room temperature, and precipitated with ethanol. Bisulfite-treated DNA was finally resuspended in 30 μL of water after vacuum drying for gene methylation analysis.
To our knowledge the T1799A transverse mutation in exon 15 of the BRAF gene is virtually the only BRAF mutation noted in PTC. Therefore we analyzed this mutation in the current study. The T1799A BRAF mutation was analyzed using genomic DNA by direct sequencing. For direct DNA sequencing, exon 15 of the BRAF gene was amplified by polymerase chain reaction (PCR), followed by Big Dye terminator cycle sequencing reaction and sequence reading on an ABI PRISM 3730 genetic analyzer (Applied Biosystems, Foster City, Calif). The PCR reaction protocol and primers for exon 15 of the BRAF gene used in this study were as described previously.20
Methylation-specific PCR (MSP) was performed on bisulfite-treated DNA in 20 μL of reaction mixture containing approximately 50 ng of bisulfite-treated DNA, 16.6 mM of ammonium sulfate, 67 mM of Tris (pH 8.8), 2 mM of MgCl2, 200 μM each of deoxynucleotide triphosphate (dATP, dCTP, dGTP, and dTTP), 200 nM of forward and reverse primers, and 0.5 U of platinum Taq DNA polymerase (Invitrogen Technologies, Inc. La Jolla, Calif). All the primers were synthesized using MSP primer sequences (Invitrogen Technologies, Inc), and PCR annealing temperatures are presented in Table 1. The MSP products were resolved with 2% agarose gel electrophoresis and visualized with ethidium bromide staining.
|Gene||Sense (5′ → 3′)||Antisense (5′ → 3′)||bp||Ta|
|Human DNA repair genes|
|BRCA1||U||TTGGTTTTTGTGGTAATGGA AAAGTGT||CAAAAAATCTCAACAAACTCA CACCA||86||58|
|Rat MLH1 gene|
Cells were lysed in the lysis buffer containing 150 mM of NaCl, 10 mM of Tris (pH 7.20), 1% SDS, 1% Triton X-100, 1% deoxycholate, 5 mM of ethylenediamine tetraacetic acid (EDTA), 2 mM of NaF, 1 mM of VaPO3, and a protease inhibitor cocktail (Sigma Chemical Company). Total cellular proteins were resolved on denaturing polyacrylamide gels, transferred to polyvinylidene difluoride membranes (Amersham Pharmacia Biotech, Piscataway, NJ), and blotted with specific primary antibodies against phosphorylated-ERK (p-ERK), total ERK (t-ERK), or β-actin (Santa Cruz Biotechnology, Santa Cruz, Calif). The antigen-antibody complexes were visualized using a horseradish-peroxidase–conjugated antibody (Santa Cruz Biotechnology) and a HyGLO horseradish-peroxidase detection kit (Denville Scientific, Metuchen, NJ). t-ERK or β-actin were used for quantity control of the proteins.
Categoric data were summarized using frequencies and percentages. Continuous variables, such as age and tumor size, were summarized using the means ± the standard deviation (SD). Categoric variables were compared using the chi-square test and continuous variables were compared using the Student t test. All reported P values were 2-sided. A P values <.05 was considered to be statistically significant. SPSS software (version 11.5; SPSS Inc, Chicago, Ill) was used for data analysis in the current study.
We analyzed the methylation status in the promoter areas of 23 genes in major DNA repair pathways, including mismatch repair (hMLH1, hMSH2, hMSH3, hMSH6, PMS1, and PMS2), nucleotide excision repair (PCNA, XPA, XPC, XPD, XPG, and ERCC1), base excision repair (XRCC1, OGG1, UNG, MPG, NTH1, MUTYH, and TDG), and double-strand-break repair pathway (BRCA1, XRCC2, XRCC3, and XRCC4) genes (Table 1). Among the 23 genes and 5 cell lines studied, 4 repair genes were found to be hypermethylated in some cell lines, including the hMLH1 gene in 1 of 5 cell lines, PCNA in 3 of 5 cell lines, PMS1 in 1 of 5 cell lines, and XPG in 1 of 5 cell lines, as were 3 of 23 genes in the NPA PTC cell line, 2 of 23 genes in the ARO ATC cell line, and 1 of 23 genes in the C643 ATC cell line (Fig. 1) (Table 2). No methylation was found in the remaining DNA repair genes in thyroid cancer cell lines.
Among the 38 PTC tumors and 23 DNA repair genes examined, 3 of 23 genes were found to be hypermethylated, including the hMLH1 gene in 8 of 38 tumors (21%), the PCNA gene in 5 of 38 tumors (13%), and the OGG1 gene in 2 of 38 tumors (5%), respectively (Fig. 1) (Table 2). No methylation was found in the remaining DNA repair genes in human PTC samples. Methylation of the 3 DNA repair genes was found to be collectively associated with lymph node metastasis (6 of 11 samples [55%] in the group with methylation vs 2 of 27 samples [7%] in the group without methylation; P = .0035). The most commonly methylated DNA repair gene in PTC was the hMLH1 gene, and this gene methylation alone was also found to be significantly associated with lymph node metastasis (5 of 8 samples [63%] in the group with methylation vs 3 of 30 samples [10%] in the group without methylation; P = .0049) (Table 3). Although no statistically significant association between this methylation and other clinicopathologic characteristics (including patient age, sex, tumor size, extrathyroidal invasion, and tumor stage) was noted, a tendency toward an association with some of these parameters could be observed (Tables 3). Methylation of the PCNA or OGG1 genes was not found to be associated with any of the clinicopathologic characteristics of the tumors, but a tendency toward an association was observed for some parameters (Table 3).
|Methylation +||Methylation −||P||Methylation +||Methylation −||P||Methylation +||Methylation −||P|
|Age at diagnosis, y||46.5 ± 11.4||45.1 ± 9.6||.73||48.6 ± 7.8||44.9 ± 10.2||.44||52.0 ± 9.9||45.0 ± 9.9||.34|
|Male gender||0 (0%)||5 (17%)||.56||0 (0%)||5 (15%)||1||0 (0%)||5 (14%)||1|
|Tumor size, cm||3.0 ± 1.4||3.2 ± 1.4||.78||3.2 ± 1.3||3.1 ± 1.4||.97||3.2 ± 1.4||2.5 ± 1.0||.51|
|Extrathyroidal invasion||1 (13%)||2 (7%)||.52||1 (20%)||2 (6%)||.35||0 (0%)||3 (8%)||1|
|Lymph node metastasis||5 (63%)||3 (10%)||.0049||2 (40%)||6 (18%)||.28||1 (50%)||7 (19%)||.38|
|I/II||4 (50%)||24 (80%)||.20||3 (60%)||25 (76%)||.59||1 (50%)||27 (75%)||.46|
|III/IV||4 (50%)||6 (20%)||.20||2 (40%)||8 (24%)||.59||1 (50%)||9 (25%)||.46|
We found the T1799A BRAF mutation in 19 of 38 PTC samples (50%). We did not observe a statistically significant association between this mutation and any of the common clinicopathologic characteristics of PTC, although a strong tendency toward an association with some of these clinicopathologic parameters was noted (Table 4). These data were inconsistent with previous results,11 a finding that was likely due to the relatively small number of tumor samples included in the current study. In contrast, we observed a close association between hMLH1 gene methylation and BRAF mutation (6 of 19 samples [32%] in the BRAF mutation-positive group vs 2 of 19 samples [11%] in the BRAF mutation-negative group; P = .042). There was a statistically insignificant trend toward an association between methylation of the combined 3 DNA repair genes and a BRAF mutation (7 of 19 samples [37%] in the BRAF mutation-positive group vs 4 of 19 samples [21%] in the mutation-negative group; P = .47).
|T1799A BRAF mutation||Wild-type BRAF||P|
|Age at diagnosis, y||45.0 ± 10.1||45.8 ± 9.9||.81|
|Male gender||4 (21%)||1 (5%)||.34|
|Tumor size, cm||3.2 ± 1.3||2.9 ± 1.2||.43|
|Extrathyroidal invasion||2 (11%)||1 (5%)||1|
|Lymph node metastasis||5 (26%)||3 (16%)||.69|
|I/II||13 (68%)||15 (79%)||.71|
|III/IV||6 (32%)||4 (21%)||.71|
|DNA repair gene methylation|
|hMLH1||6 (32%)||2 (11%)||.042|
|PCNA||2 (11%)||3 (16%)||1|
|OGG1||0 (0%)||2 (11%)||.49|
In an attempt to explore the functional correlation between MAP kinase signaling and MLH1 gene methylation, we tested the effects of alterations in MAP kinase signaling on the methylation status of the MLH1 gene in thyroid cancer cell lines. We first used the human thyroid cancer cell line ARO, which naturally harbored the BRAF mutation with aberrantly activated MAP kinase signaling. Treating the ARO cells with U0126 virtually abolished the MAP kinase pathway signaling, as reflected by the complete loss of p-ERK (Fig. 2A). No change in the naturally existing methylation of the hMLH1 gene occurred after 5 days of such treatment with U0126 in these cells. We next used a rat thyroid cancer cell line, PCCL3, that had doxycycline-inducible expression of transfected BRAF V600E. As shown in Figure 2B, forced expression of the mutant BRAF with a dramatic increase in p-ERK for 6 days failed to induce methylation of the MLH1 gene in PCCL3 cells. These results suggest that acute changes in the MAP kinase pathway signaling have no direct effect on MLH1 gene methylation.
Although DNA methylation has been studied in many genes in thyroid cancer, to our knowledge, the epigenetic alteration of DNA repair genes has not been investigated to date in thyroid cancer. Given the recent demonstration of the hypermethylation of some DNA repair genes, particularly the hMLH1gene, and its association with BRAF mutation in colon cancer, it is highly tempting to ask whether this significant epigenetic alteration also occurs in PTC, which has a high prevalence of BRAF mutations. To our knowledge, the current study is the first to investigate the methylation status of the promoters of a large number of DNA repair genes in PTC and its correlation with BRAF mutation and clinicopathologic characteristics of the tumor.
Among the 23 DNA repair genes analyzed, only the hMLH1, PCNA, and OGG1 genes were found to be methylated in PTC tumors, suggesting that gene methylation of DNA repair systems is not widespread but instead occurs selectively within a few genes involved in PTC tumorigenesis. It is interesting to note that methylation of the 3 genes, either collectively or hMLH1 alone, was found to be highly associated with lymph node metastasis, suggesting an important role of the epigenetic alteration of these genes in the development of disease progression and metastasis in patients with PTC. These data also suggest that the methylation of these DNA repair genes could be a good prognostic molecular marker for lymph node metastasis in PTC. It remains to be determined how hMLH1 gene methylation could be associated with lymph node metastasis in PTC. It is possible that aberrant methylation and hence silencing of this mismatch DNA repair gene may cause a defect in the DNA repair system within the cell, resulting in genetic alterations, such as functionally important mutations in PTC. Protein products of such mutations could be important tumor-promoting molecules involved in the invasion and metastasis of PTC, promoting lymph node metastasis. The BRAF mutation did not demonstrate a statistically significant association with lymph node metastasis, most likely because of the small number of samples (ie, there were only 19 BRAF mutation-positive and 19 BRAF mutation-negative samples in the current study). Even with this small number of samples, hMLH1 methylation demonstrated a strong association with lymph node metastasis, suggesting that this epigenetic alteration may be a stronger predictor than BRAF mutation for PTC metastasis.
It is interesting to note that in the current study, we also found that methylation of the hMLH1 gene was closely associated with BRAF mutation in PTC. Remarkably, these results closely mimic the close association between hMLH1 gene methylation and BRAF mutation observed in colon cancer,13–16 suggesting a common role of this epigenetic alteration in BRAF mutation-promoted tumorigenesis of colon and thyroid cancers. Because the hMLH1 gene is a mismatch repair gene, it might be logical to speculate about a defective function of this gene, such as that caused by aberrant methylation as noted in PTC (the current study) and colon cancer,13–16 in the development of the BRAF mutation. However, this is apparently not the case as suggested by the infrequent occurrence of BRAF mutations in colon cancer patients with a defective hMLH1 gene due to germline mutations.13 Instead, aberrant methylation of the hMLH1gene, and perhaps the other 3 DNA repair genes as well, may be a consequence of the BRAF mutation. Although the results of the current study demonstrated that acute activation of the MAP kinase signaling did not cause MLH1 methylation in the cell lines studied, it remains to be investigated whether this epigenetic alteration is a consequence of the long-term and chronic impact of MAP kinase pathway signaling aberrantly activated by the BRAF mutation in tumors.
In the current study, we investigated the methylation status of a large number of DNA repair genes in PTC. We found frequent hypermethylation of the mismatch repair gene hMLH1 and its close association with lymph node metastasis and, as noted in colon cancer, BRAF mutation in PTC. These data provide a second example of a cancer in which aberrant hMLH1 gene methylation may play a role in BRAF mutation-promoted tumorigenesis.
We thank Dr. N. E. Heldin (University of Uppsala, Uppsala, Sweden) for kindly providing us with the C643 cells, Dr. James A. Fagin (Memorial Sloan-Kettering Cancer Center, New York, NY) for the PCCL3 cells, and Dr. Guy J. F. Juillard (University of California at Los Angeles School of Medicine, Los Angeles, CA) for providing the remainder of the cells.