The present study investigated the transcriptional regulation of low-fidelity translesion DNA synthesis (TLS) polymerases in human esophageal carcinoma. Significantly higher mRNA expression of polymerase zeta (Polξ), RAD18, polymerase iota (Polι), and polymerase kappa (Polκ) was found in esophageal carcinomas. The increased expression of Polι in tumor samples was further confirmed by immunohistochemistry. The promoter of POLI that encodes Polι was found to be hypomethylated, although the overexpression of this gene was unlikely to be associated with methylation in tumors. We further identified Sp1 and Oct-1 binding sites present in the POLI promoter. We observed that the binding affinity of Sp1 to the POLI promoter was significantly increased in cancerous tissues and that Sp1 activated POLI gene transcription in cultured cell lines. The present study demonstrates overexpression of the TLS genes in esophageal carcinoma and identifies a key role for Sp1 in upregulating POLI gene expression. (Cancer Sci, doi: 10.1111/j.1349-7006.2012.02309.x, 2012)
Esophageal squamous cell carcinoma (ESCC) is the sixth most common cause of cancer-related deaths in modern society. Some of the etiological factors of ESCC, such as heavy alcohol drinking, cigarette smoking, micronutrient deficiency, and exposure to carcinogens, render cells prone to DNA damage. Mutations in specific genes could be central in tumorigenesis, such as a mutated p53 gene in ESCC.[4, 5] To maintain genome stability, cells have evolved mechanisms to repair the damaged DNA, but continuously occurring DNA lesions cannot be eliminated completely.
Error-free and/or error-prone translesion DNA synthesis (TLS) confers the majority of base substitutions when the replication complex is blocked in the DNA strand. The TLS process is undertaken by at least five accessory DNA polymerases. Based on structural homology, these polymerases are classified into the Y family (polymerase eta [Polη], polymerase iota [Polι], polymerase kappa [Polκ], and reversionless protein 1 [REV1]) or B family (polymerase zeta [Pol ξ]). Polymerase η is a homolog of yeast radiation-sensitive protein 30 (RAD30) and can efficiently bypass thymine–thymine cyclobutane pyrimidine dimers (T-T CPD), which is the most common DNA damage, induced by ultraviolet (UV) irradiation. Another homolog of yeast RAD30, namely Polι, is remarkably error prone when it replicates undamaged DNA in vitro. Polymerase ι can also facilitate mutagenic replication past a T-T CPD. In contrast, Polκ cannot bypass UV-induced photoproducts, although it can bypass certain bulky adducts with a moderate process. Unlike Polη, Polι and/or Polκ inserting a base across from a lesion, Polξ extends the mispair to form a template primer that can be further extended by Polδ. Human Polξ is composed of a catalytic subunit REV3L and the structural subunit REV7L. REV1L appears to insert dCMP into DNA opposite a guanine, an apurinic/apyrimidinic (AP) site, or a uracil. The low fidelity of these special polymerases in DNA replication compared with that of replicative polymerases is important for DNA damage tolerance. The RAD6–RAD18 complex, which adds ubiquitin to K164 of proliferating cell nuclear antigen (PCNA), regulates the balance of these special polymerases.
Unrestrained activity of error-prone polymerases would lead to widespread mutagenesis and affect genomic stability. Therefore, transcriptional regulation of the TLS polymerase genes may represent an underexplored etiology of ESCC. In the present study, we analyzed mRNA levels of REV1L, REV3L, REV7L, Polκ, RAD18, Polι and Polη in normal and cancerous esophageal tissues. We further investigated the mechanism responsible for the upregulation of Polι in cancerous esophageal tissues.
Materials and Methods
The mRNA levels of the TLS genes REV1L, Polξ (REV3L, REV7L), RAD18, Polη (POLH), Polι (POLI), and Polκ (POLK) were analyzed by real-time PCR. In total, 24 normal esophageal tissues and 60 human ESCC tissues were used. Cancerous tissues, as well as normal esophageal tissues, were obtained following surgical resections performed in 2008 at the Gastrointestinal Center, Jiangyin People's Hospital (Jiangyin, China). All patients provided signed informed consent for their tissues to be used for scientific research. Ethical approval for the study was obtained from the Jiangyin People's Hospital. The histological features of the specimens were evaluated by a senior pathologist according to the World Health Organization's (WHO) classification criteria (2000). Tissues were obtained before chemotherapy and radiation therapy and were frozen immediately before being stored at −80°C until use.
RNA extraction and reverse transcription–real-time PCR
Total RNA from esophageal tissues was extracted using Trizol (Invitrogen, Carlsbad, CA, USA) and reverse transcribed to cDNA using an oligo(dT)12 primer and Superscript II (Invitrogen). The SYBR green dye (Takara, Shiga, Japan) was used for the amplification of cDNA. The mRNA levels of REV1L, REV3L, REV7L, RAD18, Polη, Polι, and Polκ, as well as that of the internal standard glyceraldehyde 3-phosphate dehydrogenase (GAPDH), were measured by real-time quantitative PCR in triplicate using a Prism 7900 real-time PCR machine (Applied Biosystems, Foster City, CA, USA). The specific primers used for these genes are listed in Table S1, available as Supplementary Material to this paper.
All tissues were routinely fixed in 10% formalin, embedded in paraffin, and cut into 5-μm sections. Sections were deparaffinized in xylene and rehydrated in graded concentrations of ethyl alcohol. Antigen retrieval was performed by heating the sections in a 650 wattage microwave oven for 10 min with 0.01 mol/L citrate buffer (pH 6.0). Endogenous peroxidases and non-specific reactions were blocked with 3% hydrogen peroxide and 8% skim milk for 30 min, respectively. Sections were then incubated with an anti-Polι mAb (Santa Cruz Biotechnology, Santa Cruz, CA, USA) at a dilution of 1 : 200 for 2 h. Labeling for Polι was detected using biotinylated secondary antibodies, visualized with diaminobenzidine substrate (Sigma-Aldrich, St Louis, MO, USA), and counterstained with hematoxylin (Sigma-Aldrich). All steps were performed at room temperature. As a negative control, PBS was used in the absence of the primary antibody, confirming the specificity of this antibody. Immunostained slides were evaluated using light microscopy by two independent observers in a blinded manner.
Cell culture and demethylation experiment
The human esophageal cancer cell lines Eca-109 and TE-1 were maintained in DMEM supplemented with 10% FBS. Cells were grown in an incubator at 37°C with 5% CO2. To block DNA methylation or histone deacetylation, Eca-109 and TE-1 cells were treated with 10 μmol/L 5-aza-2′-deoxycytidine (5-Aza) or 2 μmol/L trichostatin A (TSA; Sigma-Aldrich) for 72 h. The cells were then collected for mRNA analysis using real-time PCR.
DNA extraction and bisulfite sequencing
Genomic DNA was extracted from tissue samples using the SDS and proteinase K methods and then subjected to sodium bisulfite treatment. We amplified and sequenced the proximal promoter DNA from −303 to −47 using the following primers: 5′-AATTTTAG-TTATTTGGGAGGTTGAG-3′ (forward) and 5′-ACCAACCTATTACCCAAAATAACAC-3′ (reverse). The PCR products were cloned into the T-easy vector (Tiangen, Beijing, China) and subjected to DNA sequencing.
Construction of promoter reporter plasmids
Fragments of the POLI promoter were amplified by PCR using the primers listed in Table S2. Each of the amplified fragments were inserted into the pGL3-basic vector (Promega, Madison, WI, USA). The insertion of the DNA fragments into the plasmid was confirmed by DNA sequencing.
Cell transfection and luciferase assays
Cells were grown in a 24-well culture plate to 70–80% confluence and then transfected with the POLI–pGL3 reporter constructs using lipofectamine 2000 (Invitrogen) according to the manufacturer's instructions. For cotransfection, cells were transfected with 800 ng pGL3 reporter constructs with or without 200 ng cDNA vector coding for Sp1 or Oct1. In each transfection, 20 ng pRL-TK (Promega) was used to normalize the transfection efficiency. After 24-h incubation, the cells were lysed using passive lysis buffer and luciferase activity was measured. Promoter activities are expressed as the ratio of Firefly luciferase to Renilla luciferase activity.
Cultured cells and esophageal tissues were used for the ChIP assays. We used the EZ ChIP kit (Upstate Biotechnology, Lake Placid, NY, USA) and followed the manufacturer's instructions. The ChIP was conducted using antibodies against RNA polymerase II, Sp1, Oct-1, Pdx-1, and IgG. We designed five sets of primers targeting different regions in the POLI promoter (see Table S3). The results for the immunoprecipitated fragments were quantified and compared with the Ct values obtained for the input samples in each case. Results are expressed as a percentage of the input.
Data from at least three independent experiments are expressed as the mean ± SEM. Differences between two groups were analyzed using Student's t-test, whereas differences between more than two groups were analyzed using anova. Correlation analysis of mRNA expression was performed using Pearson's test. All tests were two-sided. Data were considered significant when P < 0.05.
Expression of TLS polymerases is upregulated in esophageal cancer tissues
To test the hypothesis that normal and cancerous esophageal tissues express a different subset of TLS genes, we assessed the mRNA levels of the genes involved in the TLS pathway using real-time PCR. The histopathological features of the 24 normal esophageal tissues and 60 ESCC samples are summarized in Table S4. Statistical analysis revealed significant differences between cancerous and normal tissues in the expression of the REV3L, REV7L, RAD18, Polκ, and Polι genes, but not in the expression of the REV1L and Polη genes (Fig. 1). The mRNA expression of the Polι gene was most significantly elevated in tumor tissues compared with normal controls (7.2-fold upregulation; P < 0.0001), followed by that of REV3L and REV7L. The increased expression of Polι in tumor samples was further confirmed by immunohistochemistry (Fig. 1c). These results demonstrate an increase in the transcription of the TLS pathway genes in esophageal carcinomas, suggesting a characteristic of this malignancy.
Hypomethylation of the POLI promoter in esophageal carcinomas
We first predicted the presence of cytosine–phosphorous–guanine (CpG) islands in the POLI promoter. The CpG Island Searcher program (http://www.uscnorris.com/cpgislands2/cpg.aspx, accessed 25 Apr 2010) helped us identify a CpG island near the transcription start site of POLI, indicating that the POLI gene may potentially be regulated through DNA methylation. To clarify the possible roles of this potential epigenetic regulation in POLI overexpression in esophageal carcinoma, sodium bisulfite sequencing was used to illustrate the methylation status in tissues. Our results revealed few methylated cytosines in CpG dinucleotides of the POLI promoter in esophageal cancer tissues as well as in their normal counterparts (Fig. 2a). To confirm this result, we performed methylation analysis in another five pairs of ESCC tissue samples. In line with previous results, no methylation was found among the five tumor samples and matched adjacent normal tissues (Fig. 2b), indicating that methylation regulation does not contribute to the increased expression of POLI in esophageal cancer tissues. We further treated two esophageal cancer cell lines (Eca-109 and TE-1) with 5-Aza, a methyltransferase inhibitor. Treatment with 5-Aza did not significantly affect POLI expression in these two cell lines (Fig. 2c), whereas 5-Aza significantly increased the relative expression of checkpoint kinase 2 (CHEK2) and microoRNA-132, as reported previously.[16, 19] Moreover, treatment of cells with TSA, a histone deacetylase inhibitor, increased the expression of protocadherin 17 (PCDH17), but not POLI (Fig. 2d). These results indicate that epigenetic regulations are unlikely to be involved in the increased transcription of POLI in ESCC.
Cis-acting regulatory elements in the POLI promoter
To determine whether transcriptional regulation of the POLI gene is responsible for its overexpression in esophageal carcinomas, the presence of cis-acting regions in the POLI promoter was explored. Promoter fragments ranging from −1631, −794, −541, −367 to −10 (relative to the transcription start site) were cloned into the pGL3 vector upstream of a luciferase reporter gene and assessed for their transcriptional activity in Eca-109 and TE-1 cells (Fig. 3a). The luciferase activity pattern driven by these promoter fragments was very similar in both cell lines. The promoter region between −367 and −794 did not significantly enhance luciferase activity (Fig. 3a). The −1631 fragment, which carries the longest cloned promoter, significantly increased luciferase activity, especially in TE-1 cells. These findings suggest that the promoter region between −1631 and −794 may contain cis-activating elements critical for transcriptional regulation of this gene.
Activation by Sp1 of POLI transcription
The bioinformatics tools TFsearch (http://www.cbrc.jp/research/db/TFSEARCH.html, accessed 5 May 2010) and Alibaba (http://www.gene-regulation.com/pub/programs/alibaba2/index.html, accessed 5 May 2010) were used to predict common transcription factors in the POLI promoter. Results revealed that there were several presumed binding sites for the transcription factors Sp1 and Oct-1 (Fig. 3a), indicating the potential requirement of these trans-acting factors for POLI transcription. We then investigated whether overexpression of Sp1 and Oct-1 can regulate the expression of POLI. Compared with the control vector, transfection of the vector expressing Sp1 significantly elevated the endogenous expression of POLI in TE-1 cells. In contrast, transfection of the vector expressing Oct-1 had little effect on POLI mRNA expression (Fig. 3b). Reporter gene assays were performed to further confirm the importance of Sp1 in upregulating POLI expression. The reporter construct containing the −1631 fragment of the POLI promoter was cotransfected with the vectors encoding Sp1 or Oct-1 into TE-1 cells. Overexpression of Sp1 increased the reporter activity up to fivefold, whereas overexpression of Oct-1 showed weaker transactivating ability (Fig. 3c). These results demonstrate the transactivating role of Sp1 in POLI gene transcription.
Correlation between endogenous POLI expression and Sp1
We further investigated whether the expression of POLI was correlated with that of Sp1 in all tissue samples studied. Using quantitative real-time PCR, we found that mRNA levels of endogenous POLI were significantly correlated with those of Sp1 (P = 0.001; r = 0.532; Fig. 4a), but not Oct-1 (P = 0.882; r = 0.06; Fig. 4b).
Increased Sp1 binding to the POLI proximal promoter in cancer tissues
Using ChIP assays, we explored the region-specific accumulation of transcription factors and RNA polymerase II during POLI activation. Five regions of the POLI promoter locus were probed to detect the in vivo binding for RNA polymerase II, Sp1, Oct-1, and the negative control Pdx-1. In both Eca-109 and TE-1 cells, there was a marked enrichment of binding affinity for Sp1 and RNA polymerase II at Region C (region labeled “C” in Fig. 3a; −1291 to −1081 relative to the transcription start site), followed by Region E, which was consistent with the results of our luciferase assays (Fig. 5a). Occupancy by Oct-1 was observed in Regions B and D in Eca-109 cells and only in Region D in TE-1 cells. In addition, we used quantitative ChIP to detect the relative binding amount of Sp1 at the five regions in normal and cancerous esophageal tissues. We found that, in vivo, Sp1 binding was increased fourfold in Region C in esophageal cancer tissues relative to normal tissues. There was no significant difference in Oct-1 binding between normal and tumor tissues. Therefore, Sp1 binding between −1291 and −1081 may confer overexpression of POLI in esophageal cancer tissues (Fig. 5b).
The error-prone propensity of TLS polymerases has been predisposed to cancers, yet the expression pattern of the TLS polymerase genes in cancer tissues remains contentious. Overexpression of TLS polymerases has been reported in various tumor types. The expression of Polκ is increased in non-small cell lung cancer tissues, but decreased in colorectal carcinoma. Another study has reported downregulation of mRNA expression of Polκ, Polη, Polι, and Polξ in human lung, stomach, and colorectal cancers. These results suggest that TLS polymerases may have different expression profiles in different types of cancer. In the present study, we found that mRNA expression of Polξ, Polκ, RAD18, and Polι was significantly upregulated in esophageal tumor samples. The increased expression of these enzymes may relieve the blockage of DNA adducts, resulting in the resistance of esophageal cancer cells to exogenous and endogenous mutagenic insults and the accumulation of DNA mutations.
Of these TLS polymerases, the overexpression of Polι in esophageal cancer was most marked. Polymerase ι, the product of the RAD30B gene, has been documented to be upregulated in several cancers,[25, 26] but its expression pattern has not been reported in esophageal cancer tissues. Polymerase ι could be the most error-generating DNA polymerase, commonly misincorporating G opposite a template T in an undamaged DNA strand. An early study demonstrated that overexpression of Polι in breast cancer cells was involved in the generation of spontaneous and translesion mutations during DNA replication. Therefore, overexpression of Polι in esophageal cancer may be correlated with an increased mutation rate and genomic instability, which are central to tumorigenesis.
Primary control of gene expression occurs at the level of initiation of transcription. Gene expression is initiated by the binding of transcription factors to their promoters. In addition, DNA methylation, an epigenetic event, acts as a general mechanism to silence gene transcription. The present study analyzed the role of trans-acting transcription factors as well as DNA methylation in regulating POLI expression in esophageal carcinoma. We found that the transcription factor Sp1, but not Oct-1, was critical for upregulation of POLI gene transcription, which is consistent with previous findings in human FL cells. Although several Oct-1 binding sites were found in the POLI promoter, the binding affinity of Oct-1 to the promoter was much weaker compared with that of Sp1. The binding affinity of Sp1 to the POLI promoter was also much higher in cancerous esophageal tissues than that in normal tissues. A correlation between the expression of Sp1 and POLI further supports the conclusion that Sp1 is critical in upregulating POLI expression in esophageal cancer tissues. Sp1 is a ubiquitously expressed transcription factor[29, 30] and can be an integral player in DNA methylation via interactions with DNA methyltransferase 1 (Dnmt1). However, in the present study, Sp1 is less likely to induce significant DNA methylation of the POLI promoter in esophageal cancer tissues because the POLI gene is highly expressed in these tissues and demethylation did not alter POLI expression levels in an esophageal cancer line.
The expression of mutagenic Polι is normally under tight control and overexpression of Polι may increase DNA mutations in eukaryotic cells. Accumulation of mutations and chromosome abnormalities in a long-term process contribute significantly to cancer initiation. Whether overexpression of Polι contributes to tumorigenesis of esophageal cancer merits further investigation.
In summary, we have demonstrated that mRNA levels of Polξ, RAD18, Polι, and Polκ are higher in esophageal carcinomas than in normal esophageal tissues. Increased binding of Sp1 to the POLI proximal promoter in esophageal cancer cells most likely contributes to the enhanced expression of the Pol ι gene. Overexpression of Polι may represent a potential diagnostic biomarker, as well as a therapeutic target, for esophageal carcinomas.
This work is supported by the National Natural Science Foundation of China (81102078 and 81172597), the Key Programs of Natural Science Foundation of Jiangsu Educational Committee (11KJA310001), the Research Foundation of Health Agency of Jiangsu Province (H200837), the Priority Academic Program Development of Jiangsu Higher Education Institutions (PAPD) and the Suzhou Administration of Science & Technology (SYS201046 and SZS201004).