p53 Ser15 phosphorylation and histone modifications contribute to IR-induced miR-34a transcription in mammary epithelial cells

Abstract

Previous studies have demonstrated that miR-34a is a direct transcriptional target of tumor suppressor p53 and plays a crucial role in p53-mediated biological processes, such as cell cycle arrest, apoptosis and senescence. However, the role of p53 phosphorylation at Ser15 and histone modifications in ionizing radiation (IR)-induced miR-34a transcription in human mammary epithelial cells remains unknown. The present study showed that IR triggers miR-34a induction in rat mammary gland tissue and human mammary epithelial cells in a dose- and time-dependent fashion. Gene copy number and CpG methylation exhibit no effect on IR-inducible miR-34a expression, while the levels of phosphorylated p53 at Ser15 are markedly elevated in human mammary epithelial cells 96 h post-IR, which correlates with IR-inducible miR-34a transcription and the p38 MAPK pathway. Conversely, suppression of p38 MAPK with SB239063 inhibits IR-induced p53 phosphorylation at Ser15 and miR-34a expression in a dose-dependent manner. Our study found that wild-type p53 is enriched at miR-34a promoter, and luciferase activity of miR-34a promoter reporter is attenuated by either mutant p53 (Ser15Ala) or mutant miR-34a promoter. Furthermore, IR also triggers phosphorylation, tri-methylation and acetylation of histone H3 and acetylation of histone H4, which correlates with IR-inducible miR-34a transcription, while SAHA potentiates IR-inducible miR-34a expression. Moreover, acetyl-histone H3 is significantly enriched at miR-34a promoter in IR-exposed HMEC cells. Yet, we show that there is no correlation between IR-inducible miR-34a expression and IR-induced rapid and transient G₂/M arrest. In sum, our novel data for the first time demonstrate that IR-induced p53 Ser15 phosphorylation via p38 MAPK is essential for its functional regulation of IR-inducible miR-34a transcription in human mammary epithelial cells, and that histone modifications may also play a key role in IR-inducible miR-34a expression.

Introduction

Breast cancer is the most common malignancy in women worldwide and the second leading cause of cancer-related deaths among North American women.¹ The factors contributing to breast tumorigenesis are still unclear. However, ionizing radiation (IR) may be a high-risk factor for the development of breast cancer, since it has been shown to strongly induce breast cancer in exposed individuals.^2,3 Indeed, IR-induced breast cancer in the medically exposed population is a growing clinical problem,⁴ especially in young, pre-menopausal women. Furthermore, in vitro studies demonstrate that IR could alter tissue function by promoting neoplastic transformation of normal breast cells,^5,6 although the mechanisms involved in radiation responses of mammary gland tissue and in malignant transformation have not been elucidated.

Furthermore, there has been much debate recently about the benefits and risks of diagnostic mammography⁷ in the detection of breast cancer. Specifically, the risks of mammography-related IR exposure in conjunction with the risk of IR-induced carcinogenesis⁷ have been discussed. IR is the only genotoxic agent accepted as a breast-specific carcinogen.⁴ Glandular doses from mammographies are low, typically around 3 mGy of 26–30 kVp X-rays^7,8; however, the main concern is that these low-energy rays have been shown to be more hazardous per unit dose than the high-energy rays or γ-rays.⁷ The low-energy X-rays exhibit higher oncotransformation potential than the 200 kVp X-rays, suggesting that the low-energy X-rays used in mammography are considerably more biologically active than previously thought.^9-11 This finding is consistent with data on chromosome aberration induction by low-energy X-rays.¹² Furthermore, it is now well-accepted that even low doses of IR can cause oxidatively generated clustered DNA lesions.^13,14 These lesions may lead to radiation-induced mutagenesis, chromosomal instability and, further, to radiation carcinogenesis.¹⁴

Indeed, in vitro studies^4,15 have shown that IR can alter the function of tissue by promoting neoplastic transformation of normal breast cells,^5,6 but the specific mechanisms underlying these processes have not been elucidated. In the last decade, the influence of epigenetic changes on gene regulation has become a new and important focus of investigation. Three distinct mechanisms—small-interfering RNAs, DNA methylation and histone modifications—are believed to initiate and sustain epigenetic modifications.

Epigenetic alterations and variations in gene expression have been associated with changes in the molecular functions of breast cells. Among those, DNA methylation occurs predominantly in the context of CG dinucleotides. It is crucially important for normal development, cell proliferation and proper maintenance of genome stability^16-18 and is associated with an inactive chromatin state and repressed gene expression activity.¹⁹ The regulatory potential of DNA methylation manifests in promoter regions that control the expression of adjacent genes.^16,20 DNA methylation changes are closely connected with alterations in the other components of chromatin structure, primarily histone modifications. Histone modifications encompass acetylation, methylation, phosphorylation, ubiquitination and sumoylation and are important for transcriptional regulation and genome stability.^21,22 For example, acetylation of several lysine residues within the N-terminal tails of H3 and H4 gives rise to an open or “on” chromatin state typically associated with DNA hypomethylation and expressed genes, while de-acetylation tends to coincide with hypermethylated DNA loci that are not expressed.²³ Histone methylation can lead to different transcriptional consequences, depending on the residue affected.^22,24

In addition, epigenetic control can be mediated by means of recently discovered small regulatory RNAs. Among those of special interest are microRNAs (miRNAs), which can inhibit the translation of a variety of proteins. MicroRNAs are abundant, small, single-stranded, noncoding RNAs that are potent regulators of gene expression.^25,26 To control the translation of their target mRNAs, miRNAs associate with the RNA-induced silencing complex (RISC) proteins and bind to the 3′-UTR of their cognate mRNAs, thus serving as translational suppressors that regulate the protein synthesis.^27,28 Regulatory microRNAs impact a wide variety of cellular processes, such as cellular differentiation, proliferation, apoptosis, genome stability and even predisposition to breast cancer.^29-31

Recent progress in cancer research has demonstrated that alterations in the cellular epigenome contribute significantly to the development of human cancers. Changes in DNA methylation, histone modifications and miRNAs are intricately linked to the initiation, promotion and progression of cancer.^32-35 Epigenetic changes may occur at very early stages of cancer predisposition and development.^36,37 Furthermore, exposure to carcinogenic agents in general, and radiation exposure in particular, has been shown to affect epigenetic changes.^38,39 Yet, very little is known about radiation-induced epigenetic changes in the mammary gland,⁴⁰ specifically the IR effects on mammary gland microRNAome.

Here we set out to analyze the effects of low, mammography-like doses of IR and high, therapy-like doses of IR on microRNAome of rat mammary gland and human mammary epithelial cells (HMEC) cells. We identified IR-induced miRNAs and further dissected in detail the mechanisms of IR-induced regulation of one of the significantly changed miRNAs, mir-34a. We found that IR potently induced p53 phosphorylation at Ser15 in mammary epithelial cells via the p38 MAPK pathway, and that p53 Ser15 phosphorylation contributed to the IR-inducible miR-34a transcription. The essential role of p53 Ser15 phosphorylation in miR-34a expression was further validated by luciferase assay and ChIP-PCR. Interestingly, IR-triggered histone modifications, including phosphorylation, methylation and acetylation of histone H3 and H4, also contributed significantly to IR-inducible miR-34a transcription. The findings highlight the crucial role of p53 Ser15 phosphorylation and histone modifications in IR-inducible miR-34a transcription.

Results

IR-induced microRNA expression in mammary gland cells

Our previous studies indicated that IR triggers a significant and sex-specific deregulation of the microRNAome and alters the levels of Dicer and components of the RISC complex in the somatic tissues of exposed mice.⁴¹ However, relatively little is known about the effects of IR exposure on the mammary gland. With this in mind, we set out to dissect the effects of IR on rat mammary gland tissue. To identify the microRNAs that are differentially expressed in mammary gland tissue in response to IR, 6-wk-old female Long-Evans rats were exposed to different doses of X-ray at different energy levels that resembled exposures in radiation therapy or mammography. MicroRNA profiling conducted at various times after IR exposure revealed interesting patterns. We found that 31 miRNAs (30 kVp/0.1 Gy) and 28 miRNAs (80 kVp/2.5 Gy) were differentially expressed.

Among these, the miR-34a was significantly induced by IR. It was significantly upregulated by the high, radiotherapy-like dose of 80 kVp/2.5 Gy and by a low, mammography-like dose of 30 kVp/0.1 Gy (Fig. 1A). This expression pattern was confirmed by quantitative, real-time PCR (qRT-PCR) (Fig. 1B). Furthermore, qRT-PCR using RNA samples from IR-exposed mammary gland tissue at different time-points (6 h and 2 wk after exposure) revealed both a time- and dose-dependent induction of miR-34a expression (Fig. 1C). Importantly, we found that IR also triggers a miR-34a induction in normal human mammary epithelial cells (HMEC, Fig. 1D).

Figure 1. IR induces miR-34a expression in rat mammary gland and HMEC cells in a dose- and time-dependent manner. (A) Total RNA isolated from mammary gland tissue of 6-wk-old female Long-Evans rats exposed to either 30 kVp/0.1 Gy or 80 kVp/2.5 Gy X-ray or left sham-treated as controls 96 h post-irradiation was subjected to microRNA microarray, as described in “Materials and Methods.” (B and C) Total RNA was isolated from mammary gland tissue of 6-wk-old female Long-Evans rats, as described in (A), and the levels of rno-miR-34a were examined by real-time RT-PCR, as described in “Materials and Methods.” (D) Total RNA was isolated from HMEC cells exposed to either 30 kVp/0.1 Gy or 80 kVp/2.5 Gy X-ray or left unexposed as control 96 h post-irradiation; the level of hsa-miR-34a was determined by real-time RT-PCR, as described in “Material and Methods.” (E and F) Genomic DNAs were extracted from IR-exposed HMEC cells; miR-34a gene copy number and promoter CpG island methylation were determined, as described in “Materials and Methods.” The asterisk indicates p < 0.05.

IR-inducible miR-34a expression is not due to changes in gene copy number and CpG methylation

Having seen significant changes in the IR-induced induction of miR-34a in rat and human cells, we further proceeded to dissect the mechanisms behind this phenomenon. Considering the role of gene copy number and CpG island methylation in the control of gene expression, we then determined the changes in miR-34a gene copy number and promoter methylation in HMEC cells upon IR exposure. Interestingly, our results showed that IR had no effect on the copy number of miR-34 gene (Fig. 1E), and that CpG methylation of miR-34a promoter is not a principal factor in the regulation of IR-inducible miR-34 expression (Fig. 1F). These findings suggested that some other mechanisms may be implicated in the control of IR-inducible miR-34a transcription.

p53 Ser15 phosphorylation contributes to IR-inducible miR-34a transcription via the p38 MAPK pathway

Since previous evidence demonstrated that miR-34a was transcriptionally regulated by p53,^29,42,43 we then used HMEC as a model to gain valuable insights into the role of p53 in IR-inducible miR-34a expression. As expected, p53 expression was elevated in HMEC cells 96 h post-irradiation (Fig. 2A). Interestingly, the levels of phospho-p53 (Ser15) were also significantly increased in HMEC cells 96 h post-IR (Fig. 2A), which correlated well with IR-induced miR-34a expression (Fig. 2B), thus suggesting that p53 phosphorylation at Ser15 may play a key part in IR-inducible miR-34a transcription. Although p53 has been shown to upregulate p21 in vitro in response to DNA-damaging agents,⁴⁴ p53 Ser15 phosphorylation here does not contribute much to p21 expression in response to IR of either 30 kVp/0.1 Gy or 80 kVp/2.5 Gy (Fig. 2A).

Figure 2. IR-induced p53 phosphorylation at Ser15 plays a crucial role in IR-inducible miR-34a expression via the p38 MAPK pathway. (A) Whole cellular lysate was prepared from IR-exposed HMEC cells at the indicated time-points post-irradiation; western blot analysis was performed using antibodies to p-p53, p53, p21 and actin. (B) Total RNA was isolated from HMEC cells exposed to IR at the indicated time-points; the levels of hsa-miR-34a were examined by real-time RT-PCR, as described in “Material and Methods.” (C) Whole cellular lysate was prepared as described in (A); western blot analysis was performed using antibodies to p-ERK1/2 or ERK1/2, p-p38 or p38, p-RSK2 or RSK2, p-MSK1 or MSK1, and actin. (D) Whole cellular lysate was prepared from HMEC cells exposed to IR in combination with different concentrations of SB239063 96 h post-irradiation; western blot analysis was performed using antibodies to p-p53, p53, p-p38, p38 and actin. (E) Total RNA was isolated from HMEC cells exposed to IR (upper penal, 30 kVp/0.1 Gy; lower penal, 80 kVp/2.5 Gy) in combination with different concentrations of SB239063 96 h post-irradiation; quantitative real-time RT-PCR was performed using hsa-miR-34a primers. (F) HEK293 cells grown to 90% confluency were cotransfected with either wild-type miR-34a promoter reporter or mutant miR-34a promoter reporter and either pEGFP-C1/WT-p53 or pEGFP-C1/MT-p53; 24 h after transfection, the cell lysate was subjected to luciferase assay, as described in “Materials and Methods.” (G) HMEC cells grown to 90% confluency were transfected with either pEGFP-C1/WT-p53 or pEGFP-C1/MT-p53; 96 h after transfection, real-time ChIP-PCR was performed, as described in “Materials and Methods.” The asterisk indicates p < 0.05.

To gain further insight into the regulation of IR-induced miR-34a transcription, we set out to identify the signaling pathway(s) that may contribute to p53 Ser15 phosphorylation. We first determined the pathways activated in HMEC cells in response to IR. Western blot analysis revealed that IR triggers activation of ERK1/2 and p38 pathways in HMEC cells (Fig. 2C). This activation is especially evident 96 h post-irradiation and agrees well with the mode of signaling transduction pathway regulation following initial exposure to IR proposed previously by Valerie et al.⁴⁵

To confirm that the aforementioned pathway(s) indeed contributed to IR-inducible phosphorylation of p53 at Ser15, we logically examined the effect of p38 and ERK1/2 inhibitors on IR-induced p53 phosphorylation. Western blot analysis revealed that ERK1/2 inhibitor AZD6244 had no effect on IR-induced p53 phosphorylation at Ser15 (Fig. S1); p38 MAPK inhibitor SB239063, however, suppressed IR-inducible p53 phosphorylation at Ser15 in a dose-dependent manner (Fig. 2D), and IR-inducible miR-34a expression was also attenuated by SB239063 (Fig. 2E), which correlated with the levels of phospho-p53 (Ser15).

To learn more about the role of p53 phosphorylation in miR-34a transcription, we generated a reporter construct harboring a miR-34a promoter and a luciferase reporter gene. Luciferase activity of wild-type miR-34a promoter reporter was increased by wild-type p53, which was attenuated by mutant p53 (Ser15Ala), whereas the luciferase activity of mutant miR-34a promoter reporter was decreased (Fig. 2F). Additionally, real-time ChIP PCR revealed an enrichment of wild-type p53 at miR-34a promoter (Fig. 2G). These results led us to believe that p53 Ser15 phosphorylation may play a crucial role in IR-inducible miR-34a transcription in HMEC cells via activation of the p38 pathway.

Histone modifications contribute to IR-inducible miR-34a expression

A large body of research has provided evidence of the key role of histone modifications in the transcriptional control of gene expression. However, nothing is known about how histone modifications regulate miR-34 expression. Therefore, we next determined the contribution of histone modifications to IR-inducible miR-34 transcription. Western blot analysis showed a transient elevation in the levels of phospho-histone H3 (Ser10), tri-methyl-histone H3 (Lys4) and acetyl-histone H3 (Lys18) 12–24 h after irradiation (Fig. 3A). This pattern correlated well with IR-induced transient activation of the ERK1/2 pathway (Fig. 2A). Interestingly, 96 h post-irradiation, the levels of phospho-histone H3 (Ser10), tri-methyl-histone H3 (Lys4) and acetyl-histone H3 (Lys18) were potently re-elevated; likewise, the level of acetyl-histone H4 (Lys16) was also higher (Fig. 3A). These patterns too were in good agreement with IR-inducible miR-34a expression (Fig. 2B).

Figure 3. Phosphorylation, methylation and acetylation of histone H3 and H4 contribute to IR-inducible miR-34a transcription. (A) Whole cellular lysate was prepared from IR-exposed HMEC cells at the indicated time-points post-irradiation; western blot analysis was performed using antibodies to phospho-histone H3 (Ser10), acetyl-histone H3 (Lys18), tri-methyl-histone H3 (Lys4), acetyl-histone H4 (Lys16), histone H3, histone H4 and actin. (B) Whole cellular lysate was prepared from IR-exposed HMEC cells in combination with different concentrations of SAHA 96 h post-irradiation; western blot analysis was performed using antibodies to acetyl-histone H3 (Lys18), acetyl-histone H4 (Lys16), histone H3, histone H4 and actin. (C) Total RNA was isolated from IR-exposed HMEC cells in combination with different concentrations of SAHA 96 h post-irradiation; quantitative real-time RT-PCR was performed using hsa-miR-34a primers. (D and E) HMEC cells grown to 50–60% confluency were exposed to either 30 kVp/0.1 Gy or 80 kVp/2.5 Gy X-ray; 96 h post-irradiation, cells were harvested; real-time ChIP-PCR and conventional ChIP-PCR were performed, as described in “Materials and Methods.” The asterisk indicates p < 0.05.

Thus, we hypothesized that histone modifications may play a role in IR-inducible miR-34 expression. To test this hypothesis, HMEC cells were exposed to different doses/energies of X-rays in combination with different concentrations of histone deacetylase inhibitor, suberoylanilide hydroxamic acid (SAHA). Western blot and real-time RT-PCR analysis indicated that SAHA caused a significant elevation in the levels of acetyl-histone H3 (Lys18) and acetyl-histone H4 (Lys16), which further enhanced IR-induced miR-34a expression in a dose-dependent manner. Interestingly, SAHA alone could also induce miR-34a expression (Fig. 3B and C). Furthermore, ChIP-PCR and real-time ChIP revealed that acetyl-histone H3 (Lys18) was remarkably enriched at miR-34a promoter in HMEC cells in response to different doses/energies of X-rays (Fig. 3D and E). These results suggested that phospho-histone H3, tri-methyl-histone H3, acetyl-histone H3 and acetyl-histone H4 contributed significantly to miR-34a expression in HMEC cells’ response to IR.

miR-34a does not correlate to IR-induced transient G₂/M arrest

It has been demonstrated that IR could induce cell cycle arrest.^46,47 Considering the well-documented role of miR-34a in cell cycle control,⁴³ we wondered whether it too could mediate IR-induced cell cycle arrest. The cell cycle analysis showed a rapid and transient G₂/M arrest in HMEC cells exposed to both low-dose/low-energy and high-dose/high-energy X-rays (Fig. 4A and B), which was consistent with previous reports.^46,47 However, the IR-induced G₂/M arrest was not correlated with IR-induced miR-34 expression (Fig. 2B), suggesting that miR-34a may not play a role in IR-induced G₂/M arrest in HMEC cells. Although IR causes a transient cell cycle arrest, MTT assay showed that it had little effect on HMEC proliferation (Fig. S2).

Figure 4. X-ray causes a transient G₂/M arrest in HMEC cell cycle. (A) HMEC cells grown to 50–60% confluency were exposed to either 30 kVp/0.1 Gy or 80 kVp/2.5 Gy X-ray, cells were harvested at the indicated time-points post-irradiation, and the cell cycle analysis was performed using a BD FACSCanto II Flow Cytometer. (B) G₂/M phase cells were statistically analyzed and presented as percentages.

Discussion

MicroRNA lin-4 and let-7 were first discovered in the developmental timing in Caenorhabditis elegans in 2001^48-50; since then, more than 700 microRNAs have been identified in humans, and their functions have been extensively studied. As a novel class of small non-coding RNAs, microRNAs have been shown to play a key role in many biological and pathological processes via targeting 3′-UTR of mRNAs. However, the mechanisms involved in the transcriptional control of microRNAs remain to be elucidated. In the present study, we used female Long-Evans rats and human mammary epithelial cells (HMEC) as model systems to analyze the effects of X-ray irradiation on miRNA expression and to dissect the mechanism of IR-induced transcriptional control.

We have shown, for the first time, that p53 Ser15 phosphorylation and histone modifications contribute to miR-34a transcription in mammary epithelial cells in response to IR. Our microRNA microarray and qRT-PCR data indicated that IR-induced miR-34a expression in IR-exposed rat mammary gland tissue in a dose- and time-dependent fashion, which is consistent with the previous report in which the inducible miR-34 expression was revealed in IR-exposed spleens.⁴³ We also found miR-34 induction in human mammary epithelial cells exposed to IR.

The microRNA miR-34a belongs to a family of evolutionarily conserved microRNAs, which includes two other members, miR-34b and miR-34c, with single, recognizable orthologs in several invertebrate species.⁴³ It was originally identified as a potential tumor suppressor thanks to its induction of apoptosis in neuroblastoma cells.⁵¹ Shortly after this report, several laboratories almost simultaneously reported that miR-34a is a direct transcriptional target of tumor suppressor p53.^{29,42,43,52,53} As an important component of the p53 network, miR-34a plays a crucial role in p53-mediated biological processes, such as cell cycle arrest, apoptosis and senescence, by directly silencing target genes associated with cell cycle control and proliferation, such as cyclin E2, CDK4/6, MET, Notch and E2F.^43,54,55 Recently, several lines of evidence have indicated a p53-independent, ELK1-mediated upregulation of miR-34a in oncogene-induced senescence.⁵⁶ Furthermore, in addition to p53, CpG methylation of miR-34a promoter may also play a role in transactivation of miR-34a.^57,58 However, the transcriptional control of miR-34a is not yet completely understood.

Furthermore, to date, there is no evidence of miR-34a mutation. However, in the future it would be reasonable to look at the miR-34a mutation in cancer and other diseases, since this mutation may result in loss-of-function and/or gain-of-function of tumor suppressor miR-34a. Although there is a redundancy among miR-34 family members, for example, miR-34a, miR-34b and miR-34c overlap in their 5′-terminus seed-site sequence (5′-GGC AGU GU-3′),⁵⁹ implicating the similarity in their targets, they are all transcriptional targets of p53.

Alterations in gene copy number and CpG methylation have a well-defined role in gene expression. Inactivation of miR-34a by aberrant CpG methylation has been indicated in human malignancies.^57,60 However, in our study, both copy number and CpG methylation had little or no effect on IR-induced miR-34a expression, implicating the involvement of other mechanism(s) in IR-inducible miR-34a transcription in HMEC cells.

In previous studies, miR-34a has been well demonstrated as a transcriptional target of p53.^{29,42,43,52,53} The role of p53 Ser15 phosphorylation in IR-induced miR-34a transcription, however, has been unexplored until the present study. Our findings revealed that in, addition to p53, the levels of phosphorylated p53 at Ser15 were elevated in HMEC cells in response to IR (Fig. 2A). Interestingly, this result is consistent with a previous report, in which p53 Ser15 phosphorylation in a p53-wild type human lung carcinoma cell line was induced 2 h after exposure to as little as 2 Gy of IR.⁶¹ Our observed p53 prosphorylation correlated with the IR-induced miR-34a expression (Fig. 2B). We also showed that p38 MAPK, not ERK1/2, contributes to IR-induced p53 phosphorylation at Ser15 (Fig. 2C–E). Most importantly, the wild-type p53 physically interacted with miR-34a gene promoter (Fig. 2G) and caused a dramatic increase in luciferase activity of wild-type miR-34a promoter reporter, which is attenuated by mutations of either p53 at Ser15 (Ser15Ala) or p53-binding site at miR-34a promoter (Fig. 2F). Taken together, these results indicate that p53 Ser15 phosphorylation via the activation of p38 MAPK may be a principal factor in IR-inducible miR-34a transcription in HMEC cells.

Yet, p53 may not be the only transcription factor governing miR-34a expression. p53 may directly, or indirectly through target miRNAs, control the expression of other transcription factors. These transcription factors may bind to miR-34a promoter and control miR-34a transcription. Also, p53 may lose its function due to mutation, leading to a downregulation of its target transcription factors, resulting in a reduction of luciferase activity in both wild-type and mutant miR-34a promoter reporters.

Chromatin structure and remodeling can have a profound influence on gene transcription. To our knowledge, this influence on IR-inducible miR-34a expression has been largely unexplored. Histone modifications can lead to recruitment of protein complexes that regulate transcription. Our findings indicated that IR triggers transient induction in phospho-histone H3, tri-methyl-histone H3, acetyl-histone H3 and acetyl-histone H4. These IR-induced modifications of histone H3 and H4 were associated with the levels of phospho-p53, phospho-ERK1/2, phospho-p38 and miR-34a expression in HMEC cells in response to IR. The IR-activated ERK1/2 and p38 MAPK pathways may contribute to the phosphorylation of histone H3 at Ser10, since it has been demonstrated that UV-B-activated RAS-MAPK and p38 MAPK pathways induce phosphorylation of histone H3 at serine 10 and 28, leading to transcription initiation of immediate early genes.⁶²

Interestingly, in our study, SAHA itself caused a dramatic increase in acetyl-histone H3 and H4 and miR-34a levels; it also strongly potentiated IR-inducible miR-34a expression. Moreover, the acetyl-histone H3 (Lys18) was significantly enriched at miR-34a promoter in IR-exposed HMEC cells (Fig. 3D and E) by real-time ChIP-PCR, the low-dose/low-energy X-ray in particular, further confirming the role of acetyl-histone H3 in IR-inducible miR-34a transcription. A possible explanation for this is that IR-activated p53 (phosphorylation at Ser15) binds to miR-34a promoter and recruits p300, resulting in acetylation of histone H3 and H4 and transcription initiation, because p53 has been indicated to bind strongly to the CR2 domain of both CBP and coactivator p300.⁶³

Although evidence has demonstrated a crucial role of miR-34a in cell cycle arrest,^42,43 unfortunately, in our case, it did not contribute to IR-induced rapid and transient G₂/M arrest in HMEC cells (Figs. 2B and 4). The IR-induced acute and transient G₂/M arrest may represent a cellular recovery mechanism, although it appears to be not mediated by miR-34a, since it has been known for some years that IR-induced DNA damage initiates cellular recovery mechanisms, including activation of DNA damage response pathways, cell cycle arrest and apoptosis.⁶⁴

Taken together, our findings demonstrate that IR-induced phosphorylation of p53 at Ser15 via p38 MAPK pathway is essential for the functional regulation of IR-inducible miR-34a transcription in human mammary epithelial cells. Histone modifications may also play a key role in IR-inducible miR-34a expression, but miR-34a does not contribute to IR-induced acute and transient G₂/M arrest.

Materials and methods

Animal exposure

Six-wk-old female Long-Evans rats were randomly assigned to different treatment groups and were exposed to X-ray irradiation. The doses and energy levels were as follows: Group 1, 30 kVp X-ray, 0.1 Gy (low-dose/low-energy, cumulative dose from multiple mammography screenings); Group 2, 80 kVp X-ray, 2.5 Gy (high-dose/high-energy); Group 3, sham-treated controls. These doses were given as a single exposure.

Ten rats per group were sacrificed at 6 h, 96 h, 4 wk and 24 wk after irradiation to analyze early and delayed/persistent effects of IR exposure. Mammary gland tissues were collected post-mortem, frozen immediately and stored at −80°C or fixed and embedded in paraffin. Handling and care of animals were in accordance with the recommendations of the Canadian Council for Animal Care and Use. The procedures were approved by the University of Lethbridge Animal Welfare Committee.

Cell culture

Human mammary epithelial cells (HMEC) purchased from Invitrogen were cultured in HuMEC Basal Serum-Free Medium (Invitrogen) containing HuMEC Supplement (Invitrogen), 100 IU/ml penicillin and 100 mg/ml streptomucin at 37°C in a humidified atmosphere of 5% CO₂. HEK293 cells were grown in DMEM/High Glucose (Thermo Scientific Limited) containing 10% FBS, 100 IU/ml penicillin and 100 mg/ml streptomucin at 37°C in a humidified atmosphere of 5% CO₂. Confleunce was detected microscopically.

Wild-type and mutant p53 expression constructs

Total RNA isolated from HMEC cells was subjected to RT-PCR using SuperScript II reverse transcriptase (Invitrogen) and Pfu DNA polymerase (Fermentas) per the manufacturers’ instruction. The following primer pairs were used: p53 F: 5′-TTA AGC TTC GAT GGA GGA GCC G-3′, p53 R: 5′-AAG GAT CCG TCT GAG TCA GGC-3′. After A-tailing, the p53 fragment was cloned into pGEM T-easy vector (Promega), and the sequence identity was confirmed by automated DNA sequencing. The p53 insert was then released by Hind III and BamH I, and subcloned into pEGFP-C1 vector (Clontech). Point mutation of p53 at Ser15 was induced by QuikChange Multi Site Directed Mutagenesis kit (Agilent Technologies Canada Inc) per the manufacturer’s instruction. The following primer pairs were used in p53 mutagenesis: A43G_G44C sense: 5′-GCG TCG AGC CCC CTC TGG CTC AGG AAA CAT TTT CAG-3′, A43G_G44C antisense: 5′-CTG AAA ATG TTT CCT GAG CCA GAG GGG GCT CGA CGC-3′. WT-miR-34a promoter reporter (EST-Prom-Luc with wild-type p53 binding site) and MT-miR-34a promoter reporter (EST-Prom-M-Luc with mutant p53 binding site) were gifts from Prof. Moshe Oren (Department of Molecular Cell Biology, The Weizmann Institute of Science); pRL-TK was purchased from Promega.

microRNA profiling

Total RNA was isolated from mammary gland tissue of different groups of IR-exposed rats using TRIzol reagent (Invitrogen) according to the manufacturer’s instruction. The microRNA profiling, clustering and data analysis were performed by LC Sciences.

microRNA real-time PCR

Total RNA isolated from IR-exposed rat mammary gland tissue and HMEC cells were subjected to real-time RT-PCR using a primer set of either rno-miR-34a or has-miR-34a (SABiosciences) per the manufacturer’s instruction; human RNU6-2 served as a loading control.

miR-34a gene copy number and promoter CpG methylation analysis

Genomic DNAs extracted from IR-exposed HMEC cells using a kit for purification of total DNA from animal blood or cells (QIAGEN) were subjected to real-time PCR using SsoFast EvaGreen Spermix (Bio-Rad) with the following primers: 34aCopyNo-F: 5′-GGC CAG CTG TGA GTG TTT CT-3′, 34aCopyNo-R: 5′-CAA CGT GCA GCA CTT CTA GG-3′; RPP38-F: 5′-TGG TTG TGA AGA CGT CGT TGA-3′, RPP38-R: 5′-TGC ATA TCC TCG CTC TCC AGA-3′. The copy number level relative to the internal control (RNase P/RPP38) was calculated by the comparative threshold cycle (CT) method (Shitashige et al., 2007). Methylation analysis of miR-34a promoter CpG was performed using Methyl-Profiler^™ DNAMethylation Enzyme kit (QIAGEN) and Methyl-Profiler^™ DNA Methylation qPCR Primer Assays (QIAGEN) as described by the manufacturer.

Cell cycle analysis

HMEC cells grown to 50–60% confluency were exposed to either 30 kVp/0.1 Gy or 80 kVp/2.5 Gy X-ray and left unexposed as control; 6 h, 12 h, 24 h and 96 h after exposure, the cells were harvested for cell cycle analysis, performed using BD FACSCanto II Flow Cytometer (BD Biosciences).

Western blot analysis

Western immunoblotting was conducted as previously described. Detailed protocol is provided in Supplementary Materials and Methods.

MTT assay

HMEC cells grown to 50–60% confluency were exposed to either 30 kVp/0.1 Gy or 80 kVp/2.5 Gy X-ray; 24 h post irradiation, 3.0 × 10³ cells were plated in 96-well plates. 3-(4,5-Dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide assays were performed using the Cell Proliferation Kit I (Roche Diagnostics GmbH) as described by the manufacturer. The spectrophotometric absorbance of samples was measured at 595 nm using a microtiter plate reader (FLUOstar Omega, BMG LABTECH).

Inhibition of p38 and ERK1/2 MAPKs

HMEC cells grown to 50–60% confluency (as determined by microscopy analysis) were exposed to either 30 kVp/0.1 Gy or 80 kVp/2.5 Gy or left unexposed as a control; 1 h after irradiation, the cells were treated with either SB239063 (inhibitor of p38 MAPK, Sigma-Aldrich Canada Ltd) or AZD6244 (ERK1/2 MAPK inhibitor, Selleck) at the indicated final concentrations and incubated at 37°C in a humidified atmosphere of 5% CO₂ for 95 h.

Inhibition of HDACs

HMEC cells grown to 50–60% confluency were exposed to either 30 kVp/0.1 Gy or 80 kVp/2.5 Gy or left unexposed as a control; 72 h after irradiation, the cells were treated with suberoylanilide hydroxamic acid (SAHA, Selleck) at the indicated final concentrations and incubated at 37°C in a humidified atmosphere of 5% CO₂ for 24 h.

Transient transfection and luciferase assay

HEK293 cells grown to 90% confluence in 6-well plate in antibiotic-free DMEM/High Glucose (Thermo Scientific Ltd) containing 10% FBS were transiently transfected with 0.5 μg reporter plasmid (either WT-miR-34a promoter or MT-miR-34a promoter), 1 μg pEGFP-C1/WT-p53 or pEGFP-C1/MT-p53 plasmid, and 5 ng pRL-TK plasmid using Lipofectamine 2000 (Invitrogen) per manufacturer’s instruction. Cells were harvested 24 h after transfection, the relative luciferase activity was measured by Dual-Luciferase Reporter Assay System (Promega) using a luminometer (FLUOstar Omega, BMG LABTECH) and with Firefly luciferase data normalized to Renilla.

Transient transfection and ChIP

HMEC cells grown to 90% confluence in a 100 mm dish in an antibiotic-free HMEC Basal Serum-Free Medium (Invitrogen) containing HMEC Supplement (Invitrogen, Carlsbad, CA, USA) were transfected with 6 μg either pEGFP-C1/WT-p53 or pEGFP-C1/MT-p53 plasmid using Lipofectamine 2000 (Invitrogen) as manufacturer’s instruction. Quantitative ChIP assays were performed as detailed elsewhere 96 h after transfection (Nelson et al., 2006; Pimanda et al., 2006). For further details, see Supplementary Materials and Methods.

Statistical analysis

The Student's t-test was used for statistical significance of differences in miR-34a expression, luciferase activity, enrichment of p53 and acetyl histone H3 at miR-34a promoter between groups. p < 0.05 was considered to be significant.

Glossary

Abbreviations:

CDK

cyclin-dependent kinases

ChIP-PCR

chromatin immunoprecipitation-polymerase chain reaction

DMEM

Dulbecco's modified Eagle medium

DNA

deoxyribonucleic acid

EGFP

enhanced green fluorescent protein

EST

expressed sequence tag

Gy

Gray

HDAC

histone deacetylases

HMEC

human mammary epithelial cell

IR

ionizing radiation

IU

international unit

kVp

peak kilovoltage

MAPK

mitogen-activated protein kinases

mGy

microGray

MT

mutant

MTT

3-[4,5-dimethylthiazol-2-yl]-2,5 diphenyl tetrazolium bromide

qRT-PCR

quantitative real-time polymerase chain reaction ERK1/2

RISC

RNA-induced silencing complex

RNA

ribonucleic acid

SAHA

suberoylanilide hydroxamic acid

UTR

untranslated region

UV

ultraviolet

WT

wild type

Disclosure of Potential Conflicts of Interest

No potential conflicts of interest were disclosed.

Supplemental Materials

Supplemental materials may be found here: 
www.landesbioscience.com/journals/cc/article/25135