Final Report Summary - HYDROXYMETHYLATION (Characterization of the role of DNA methylation, hydroxymethylation and TET proteins in progesterone-mediated signaling in breast cancer cells.)
The steroid hormone progesterone plays critical roles in a variety of physiological functions including normal breast development during puberty and in preparation for lactation. Furthermore, its synthetic analogs, progestins, are frequently prescribed as contraceptives or to alleviate menopausal symptoms. However, progesterone and progestins are also involved in the etiology and progression of hormone-dependent cancers such as breast and endometrial cancer. Among different types of cancers, breast and endometrial cancers are the most common cause of cancer death in women worldwide. The knowledge of molecular mechanisms involved in the progesterone-mediated gene regulation could enable the identification of new targets for the diagnosis, prevention and treatment of hormone-dependent cancers.
Progesterone and progestins regulate (activate or repress) transcription of thousands genes by modulating their epigenetic modifications. Epigenetic modifications refer to DNA and histone chemical modifications that are heritable between cell divisions in order to maintain cellular identity. In the last decade it was largely demonstrated the critical role of histone modifications during progesterone-mediated gene regulation, however, whether the DNA modifications play a role in this context has never been investigated. Until few years ago, the term DNA modification just refered to the process where a methyl group is added to the cytosine to form the 5-methylcytosine (5mC). However, it was recently demonstrated that the ten eleven-translocation proteins (TET1,2,3) can demethylate the DNA through three consecutive oxidation reactions sintesyzing new DNA modifications: 5-hydroxymethylcytosine (5hmC), 5-formylcytosine (5fC), and 5-carboxylcytosine (5-CaC). In contrast to the 5fC and 5CaC, the 5hmC as well as the 5mC play critical role in the regulation of gene expression. TET proteins mutations occur in a significant proportion of patients with different types of cancers, whether they play a role in hormone-dependent cancers has never been investigated.
In this project we demonstrate that TET2 is significantly overexpressed in breast cancers and the breast cancer patients with higher TET2 levels have a poor clinical outcome. Moreover by genome-wide approaches we found that TET2 contributes to the regulation of a subset of progesterone-responsive genes in breast cancer cells and progesterone-treatment only slightly affect the DNA methylation as well as the hydroxymethylation patterns of breast cancer cells. These data demonstrate that TET2 regulate progesterone-mediated gene transcription most probably without modifying their methylation/hydroxymethylation patterns.
TET2 is overexpressed in breast cancers and is associated with poor clinical outcome.
Mutations of TET proteins are frequently observed in different types of cancer. In order to investigate whether the TET proteins could play a role in breast cancer, we compared the TET (1-3) expression levels between breast cancers versus normal breasts using the gene expression data from five different studies (references, oncomine databases). All the breast cancers samples showed comparable TET1 and TET3 levels with normal breasts, in contrast, TET2 transcription levels was significantly overexpressed in breast cancers (fold change >1.5 p-value <0.01) (fig. 1a). Subsequently, in order to check whether the clinical outcome of the breast cancer patients could dependent on TET2 levels, based on TET2 expression levels, we classified the breast cancers samples in three different classes: patients with the highest (fold change>1.5) intermediate (fold change<1.5 and >-1.5) and lowest (fold change <-1.5) TET2 levels and compared their clinical outcome by checking the patients survival percentage during the time. We found that the survival percentage of patients with the highest TET2 expression was significantly lower than the breast cancers with intermediate and lowest TET2 expression levels (fig. 1b). The difference of the clinical outcome between the patients with the different TET2 levels is stronger when we compared the survival percentage of the patients with the highest TET2 levels (top 10%) with the lowest TET2 expression (bottom 10%) suggesting that TET2 is involved in the breast cancer progression (fig. 1b)
TET2 regulates the transcription of a large subset of progestin-regulates genes.
Progesterone and its analogs progestins are involved in the progression of hormone-dependent cancers including breast cancer. Since our data suggest that TET2 can be involved in the same process, we investigate whether TET2 partecipate in progestin-dependent transcriptional responses of breast cancer cells. For this purpose, we first analyzed the transcription levels of TET genes in T47D breast cancer cells before and after progestin-induction in order to understand which is the highest TET expressed gene in breast cancer cells. The transcription analysis was performed by reverse-transcrinptases quantitative PCR (RT-qPCR) using specific primers to amplify TET1, TET2 and TET3 genes. The results showed that TET2 is the highest expressed TET genes and while progestin-induction does not affect TET2 and TET3 transcriptional levels, it strongly reduces TET1 gene expression (fig. 2 a,b). To confirm that TET1 is directly regulated by progesterone, we analyzed the progesterone receptor (PR) binding over the TET1 gene (promoter and gene body) prior and after 1 hour of progestin induction. The binding analysis was performed by chromatin immunoprecipitation using a polyclonal PR or negative control antibody. The immunoprecipitated DNA was amplified by qPCR using specific primers for different regions of TET1 gene. The PR binding strongly increased after hormone-treatment within the first intron of the gene demonstrating that, in contrast to TET2 and TET3 gene, TET1 is a direct progesterone target gene (sup. fig. 1).
To explore whether TET2 regulates progesterone-responsive genes we reduced TET2 levels by short interfering RNA approach (siRNA). Upon knock down confirmation at protein and RNA levels (fig. 3 a, b), to define if gene expression changes exist, and are thus, TET2 dependent, we have analysed by RNA-sequencing the transcriptome of untreated and progestin-treated (6h) cells of siTET2 knocked-down and scrambled (sicontrol) transfected samples. We found that TET2 depletion profoundly altered progesterone-responsive genes expression at basal level (hormone-independent) and after progestin-treatment. Indeed, comparing the transcriptome profiles between untreated siTET2 and sicontrol transfected cells (basal expression), we revealed that 525 genes were upregulated (TET2 negatively regulated-genes; > 2 fold, P< 0.001) and 514 genes were downregulated (TET2 positively regulated-genes; <2 fold, P< 0.001) in TET2 depleted cells (fig. 3c). In contrast, progestin-stimulated gene expression profiles (hormone-dependent) demonstrated that 314 genes were upregulated (TET2 negatively regulated-genes; > 2 fold, P< 0.001) and 188 genes were downregulated (TET2 positively regulated-genes; <2 fold, P< 0.001) in TET2 knock-down compared to sicontrol transfected cells (fig.3c). Molecular gene ontology analysis, moreover, revealed that TET2-dependent progestin-responsive genes encode mainly transcription factors and they can be involved in different cell growth pathways including FGF, CDC42, EPHA2 and Erb signalings (data not shown) suggesting that TET2 could be required for the progestin-induced cell growth. To explore whether TET2 acts at the level of chromatin, we are currently testing the genome-wide TET2 binding by chomatin-immunoprecipitation combined with high-throughput sequencing assay.
Progesterone-treatment slightly affect the genome-wide DNA methylation and hydroxymethylation pattern of breast cancer cells.
The Ten eleven-translocation protein 2 can regulate gene expression demethylating the DNA through oxidation reactions. During this process they catalyze the 5-hydroxymethylcitosine as intermediate. Since our data demostrate that TET2 cooperates in the progesterone-mediated gene regulation in order to explore whether progestin modify the DNA methylation (5mC) and hydroxymethylation (5hmC) patterns of breast cancer cells, we treated T47D cells with progestin R5020 for 6 hours or 24 hours and analyzed the methylome and hydroxymethylome prior and after progestin-induction by MeDIP-sequencing method. The comparison between the 5mC and 5hmC profiles between untreated and progestin-treated cells revealed that the hormone-treatment slightly affect the genome-wide DNA methylation anf hydroxymethylation patterns. Just few hundred of genomic regions change their 5mC or 5hmC patterns after hormone treatment and, apart two (ZNF488 and PDE4D genes), all of them are localized in intragenic regions (Table 1). The mC and hmC profiles of the TET2-dependent regulated genes, moreover, are not affected after progestin treatment suggesting that TET2 can regulate the expression of progesterone-responsive genes independently from its enzymatic activity. To test this possibility we are currently analyzing the DNA methylation and hydroxymethylation patterns upon TET2 depletion.
Conclusion and impact.
TET2 protein convert 5-methylcitosine into 5-hydroxymethylcytosine (5hmC) through oxidation reaction enabling the DNA demethylation process. The loss-of-function mutations of this protein occur in a significant proportion of patients with myeloid malignancies, however, whether it plays a role in solid cancers has never been investigated. In the current study, we first compared the TET (1-3) gene expression levels in breast cancers versus normal breast by a metagene analysis and found that in contrast to TET1 and TET3, TET2 is significantly overexpressed in breast cancers and the breast cancer patients with higher TET2 levels have a poor clinical outcome. TET2 depletion from breast cancer cells, moreover, profoundly altered the expression of a large subset of progesterone-responsive genes. Finally progesterone-treatment did not change the methylation and hydroxymethylation patterns of TET2-dependent progestin-responsive genes suggesting that TET2 can regulate progesterone-mediated gene expression independently from its demethylases activity. Our results strongly suggest that TET2 could contribute to the progression of hormone-dependent breast cancer.
Progesterone and progestins regulate (activate or repress) transcription of thousands genes by modulating their epigenetic modifications. Epigenetic modifications refer to DNA and histone chemical modifications that are heritable between cell divisions in order to maintain cellular identity. In the last decade it was largely demonstrated the critical role of histone modifications during progesterone-mediated gene regulation, however, whether the DNA modifications play a role in this context has never been investigated. Until few years ago, the term DNA modification just refered to the process where a methyl group is added to the cytosine to form the 5-methylcytosine (5mC). However, it was recently demonstrated that the ten eleven-translocation proteins (TET1,2,3) can demethylate the DNA through three consecutive oxidation reactions sintesyzing new DNA modifications: 5-hydroxymethylcytosine (5hmC), 5-formylcytosine (5fC), and 5-carboxylcytosine (5-CaC). In contrast to the 5fC and 5CaC, the 5hmC as well as the 5mC play critical role in the regulation of gene expression. TET proteins mutations occur in a significant proportion of patients with different types of cancers, whether they play a role in hormone-dependent cancers has never been investigated.
In this project we demonstrate that TET2 is significantly overexpressed in breast cancers and the breast cancer patients with higher TET2 levels have a poor clinical outcome. Moreover by genome-wide approaches we found that TET2 contributes to the regulation of a subset of progesterone-responsive genes in breast cancer cells and progesterone-treatment only slightly affect the DNA methylation as well as the hydroxymethylation patterns of breast cancer cells. These data demonstrate that TET2 regulate progesterone-mediated gene transcription most probably without modifying their methylation/hydroxymethylation patterns.
TET2 is overexpressed in breast cancers and is associated with poor clinical outcome.
Mutations of TET proteins are frequently observed in different types of cancer. In order to investigate whether the TET proteins could play a role in breast cancer, we compared the TET (1-3) expression levels between breast cancers versus normal breasts using the gene expression data from five different studies (references, oncomine databases). All the breast cancers samples showed comparable TET1 and TET3 levels with normal breasts, in contrast, TET2 transcription levels was significantly overexpressed in breast cancers (fold change >1.5 p-value <0.01) (fig. 1a). Subsequently, in order to check whether the clinical outcome of the breast cancer patients could dependent on TET2 levels, based on TET2 expression levels, we classified the breast cancers samples in three different classes: patients with the highest (fold change>1.5) intermediate (fold change<1.5 and >-1.5) and lowest (fold change <-1.5) TET2 levels and compared their clinical outcome by checking the patients survival percentage during the time. We found that the survival percentage of patients with the highest TET2 expression was significantly lower than the breast cancers with intermediate and lowest TET2 expression levels (fig. 1b). The difference of the clinical outcome between the patients with the different TET2 levels is stronger when we compared the survival percentage of the patients with the highest TET2 levels (top 10%) with the lowest TET2 expression (bottom 10%) suggesting that TET2 is involved in the breast cancer progression (fig. 1b)
TET2 regulates the transcription of a large subset of progestin-regulates genes.
Progesterone and its analogs progestins are involved in the progression of hormone-dependent cancers including breast cancer. Since our data suggest that TET2 can be involved in the same process, we investigate whether TET2 partecipate in progestin-dependent transcriptional responses of breast cancer cells. For this purpose, we first analyzed the transcription levels of TET genes in T47D breast cancer cells before and after progestin-induction in order to understand which is the highest TET expressed gene in breast cancer cells. The transcription analysis was performed by reverse-transcrinptases quantitative PCR (RT-qPCR) using specific primers to amplify TET1, TET2 and TET3 genes. The results showed that TET2 is the highest expressed TET genes and while progestin-induction does not affect TET2 and TET3 transcriptional levels, it strongly reduces TET1 gene expression (fig. 2 a,b). To confirm that TET1 is directly regulated by progesterone, we analyzed the progesterone receptor (PR) binding over the TET1 gene (promoter and gene body) prior and after 1 hour of progestin induction. The binding analysis was performed by chromatin immunoprecipitation using a polyclonal PR or negative control antibody. The immunoprecipitated DNA was amplified by qPCR using specific primers for different regions of TET1 gene. The PR binding strongly increased after hormone-treatment within the first intron of the gene demonstrating that, in contrast to TET2 and TET3 gene, TET1 is a direct progesterone target gene (sup. fig. 1).
To explore whether TET2 regulates progesterone-responsive genes we reduced TET2 levels by short interfering RNA approach (siRNA). Upon knock down confirmation at protein and RNA levels (fig. 3 a, b), to define if gene expression changes exist, and are thus, TET2 dependent, we have analysed by RNA-sequencing the transcriptome of untreated and progestin-treated (6h) cells of siTET2 knocked-down and scrambled (sicontrol) transfected samples. We found that TET2 depletion profoundly altered progesterone-responsive genes expression at basal level (hormone-independent) and after progestin-treatment. Indeed, comparing the transcriptome profiles between untreated siTET2 and sicontrol transfected cells (basal expression), we revealed that 525 genes were upregulated (TET2 negatively regulated-genes; > 2 fold, P< 0.001) and 514 genes were downregulated (TET2 positively regulated-genes; <2 fold, P< 0.001) in TET2 depleted cells (fig. 3c). In contrast, progestin-stimulated gene expression profiles (hormone-dependent) demonstrated that 314 genes were upregulated (TET2 negatively regulated-genes; > 2 fold, P< 0.001) and 188 genes were downregulated (TET2 positively regulated-genes; <2 fold, P< 0.001) in TET2 knock-down compared to sicontrol transfected cells (fig.3c). Molecular gene ontology analysis, moreover, revealed that TET2-dependent progestin-responsive genes encode mainly transcription factors and they can be involved in different cell growth pathways including FGF, CDC42, EPHA2 and Erb signalings (data not shown) suggesting that TET2 could be required for the progestin-induced cell growth. To explore whether TET2 acts at the level of chromatin, we are currently testing the genome-wide TET2 binding by chomatin-immunoprecipitation combined with high-throughput sequencing assay.
Progesterone-treatment slightly affect the genome-wide DNA methylation and hydroxymethylation pattern of breast cancer cells.
The Ten eleven-translocation protein 2 can regulate gene expression demethylating the DNA through oxidation reactions. During this process they catalyze the 5-hydroxymethylcitosine as intermediate. Since our data demostrate that TET2 cooperates in the progesterone-mediated gene regulation in order to explore whether progestin modify the DNA methylation (5mC) and hydroxymethylation (5hmC) patterns of breast cancer cells, we treated T47D cells with progestin R5020 for 6 hours or 24 hours and analyzed the methylome and hydroxymethylome prior and after progestin-induction by MeDIP-sequencing method. The comparison between the 5mC and 5hmC profiles between untreated and progestin-treated cells revealed that the hormone-treatment slightly affect the genome-wide DNA methylation anf hydroxymethylation patterns. Just few hundred of genomic regions change their 5mC or 5hmC patterns after hormone treatment and, apart two (ZNF488 and PDE4D genes), all of them are localized in intragenic regions (Table 1). The mC and hmC profiles of the TET2-dependent regulated genes, moreover, are not affected after progestin treatment suggesting that TET2 can regulate the expression of progesterone-responsive genes independently from its enzymatic activity. To test this possibility we are currently analyzing the DNA methylation and hydroxymethylation patterns upon TET2 depletion.
Conclusion and impact.
TET2 protein convert 5-methylcitosine into 5-hydroxymethylcytosine (5hmC) through oxidation reaction enabling the DNA demethylation process. The loss-of-function mutations of this protein occur in a significant proportion of patients with myeloid malignancies, however, whether it plays a role in solid cancers has never been investigated. In the current study, we first compared the TET (1-3) gene expression levels in breast cancers versus normal breast by a metagene analysis and found that in contrast to TET1 and TET3, TET2 is significantly overexpressed in breast cancers and the breast cancer patients with higher TET2 levels have a poor clinical outcome. TET2 depletion from breast cancer cells, moreover, profoundly altered the expression of a large subset of progesterone-responsive genes. Finally progesterone-treatment did not change the methylation and hydroxymethylation patterns of TET2-dependent progestin-responsive genes suggesting that TET2 can regulate progesterone-mediated gene expression independently from its demethylases activity. Our results strongly suggest that TET2 could contribute to the progression of hormone-dependent breast cancer.