Scientist have identified a molecular switch that appears to control the dual functions of transcription factor ATF2, which can play an oncogenic role in melanoma but a tumor suppressor role in nonmalignant skin cancers. A team at Sanford-Burnham Medical Research Institute, Yale University, the University of California, San Diego, and Ajou University School of Medicine in Korea has found that ATF2’s function in cells subjected to genotoxic stress depends on whether the transcription factor is held in the nucleus or allowed to translocate into the cytoplasm. This trafficking is controlled by protein kinase Cε (PKCε).
Describing their work in cultured cells, Sanford-Burnham’s Ze’ev A. Ronai , Ph.D., and colleagues subsequently confirmed that in primary human melanoma, high levels of PKCε are associated the most with aggressive tumors and poorer patient survival. Reporting in Cell, the team claims its findings could provide the foundation for the development of prognostic tests and new approaches to rendering melanoma less resistant to therapy. Their published paper is titled “PKCε Promotes Oncogenic Functions of ATF2 in the Nucleus while Blocking Its Apoptotic Function at Mitochondria.”
Activating transcription factor 2 (ATF2) plays a key role in regulating cellular growth, development, and response to stress, the latter being mediated through its impact on gene expression programs implicated in cell cycle control, cytokine expression, and cell death. In addition to its transcriptional role, however, ATF2 also plays a role in cells’ response to DNA damage response, Dr. Ronai’s team reports.
All these functions are mediated by ATF2 located in the nucleus but, the researchers note, “growing evidence points to cytoplasmic localization of ATF2, although its function there remains elusive.” Although the role of cytoplasmic ATF2 isn’t completely clear, the transcription factor appears to be able to exert both oncogenic and tumor-suppressor effects depending on its localization. In melanoma, for example, nuclear ATF2 is required for tumor development and is associated with metastasis and poor prognosis, whereas cytoplasmic ATF2 is associated with nonmalignant skin cancers and better prognosis.
The Sanford-Burnham team’s work was designed to unravel this apparently dual role of ATF2. Prior research had found that squamous cell carcinoma (SSC) accumulates ATF2 in the cytosol, so the team started their work in an SSC cell line. They initially confirmed that whereas nonstressed cells exhibited predominantly nuclear ATF2, when the cells were put under genotoxic stress through treatment with etoposide (ETO), ultraviolet C, or ionizing radiation, ATF2 trafficked to the cytoplasm and associated with proteins at the mitochondrial outer membrane (MOM).
This mitochondrial localization of ATF2 following genotoxic stimuli was similarly exhibited in other cell types including normal human fibroblasts (HSF), primary human keratinocytes (NHEK), and melanocytes (HEM) as well as other SCC cell lines. In contrast, while some melanoma cell lines exhibited partial ATF2 accumulation at mitochondria in response to genotoxic stimuli, in a number of melanoma lines the transcription factor simply didn’t translocate to mitochondria.
Subsequent work using the SSC cells demonstrated that localization of ATF2 at the MOM directly blocked the formation of HK1/VDAC1 complexes that normally form in response to various forms of stress and disrupted MOM integrity, leading to mitochondrial leakage as a precursor to cell death.
Studies also confirmed that the ability of AFT2 to move from the nucleus to the MOM and its transcriptional activity, were both regulated by protein kinase Cε (PKCε) through phosphorylation of ATF2 at residue Thr52. The pivotal role of PKCε in allowing AFT2 trafficking out of the nucleus was demonstrated using a translocation inhibitor (PKCε-i) or siRNA-mediated knockdown of PKCε. In these cells ATF2 translocated to the mitochondria even in the absence of ETO. In contrast, overexpression of a constitutively active form of PKCε blocked ATF2 accumulation at mitochondria following ETO treatment.
The relevance of Thr52 phosphorylation was evaluated in cells expressing either a mutant form of ATF2 with a nonphosphorylatable alanine at position 52 (ATF2T52A) or a mutant glutamic acid phosphomimic (ATF2T52E). Cells carrying the mutant glutamic acid phosphomimic ATF2 exhibited constitutive nuclear localization of the transcription factor, whereas the ATF2T52A mutant localized to the cytoplasm and mitochondria, even in the absence of genotoxic stress.
Thus, in the SCC cells, genotoxic stress reduces PKCε’s hold on ATF2 and allows its nuclear export and localization at the mitochondria, whereas high levels of PKCε block ATF2 nuclear export and function at the mitochondria, the researchers state. Notably, chemically inhibiting PKCε or using an siRNA to knock out the kinase resensitized a melanoma cell line to genotoxic stress-induced cell death.
To see how relevant these findings were to melanoma patients, the researchers looked at PKCε expression and activity in a panel of melanoma cell lines and a large cohort of melanoma tissue arrays (TMA). They first confirmed that overall, melanoma cell lines exhibited higher PKCε expression than primary keratinocytes, melanocytes, or SCC lines and that Thr52 phosphorylation was lowest in the primary keratinocytes and melanocytes and highest in the melanoma cell lines.
When they looked at the relationship between PKCε expression levels, patient survival, and other clinical and pathological factors in the TMAs, they found that PKCε expression was higher in metastatic than primary specimens, and in primary specimens PKCε expression was higher in thicker or ulcerated lesions. Interestingly, PKCε expression in primary tumors, though not metastatic tumors, was strongly associated with decreased patient survival.
“Thus, PKCε dictates the tumor suppressor or oncogenic activities of ATF2 by directly affecting its nuclear or cytosolic localization,” the Sanford-Burnham team concludes. “We show that the ability of ATF2 to abrogate the HK1/VDAC complex is perturbed in melanomas, which offers insight into the mechanisms underlying the notorious resistance of melanoma to therapy.”
Dr. Ronai’s lab is now working in collaboration with Sanford-Burnham's Conrad Prebys Center for Chemical Genomics to identify small molecule drug candidates that could help release ATF2 from PKCε's control, so the protein can perform its key role in ensuring apoptosis is triggered when needed. "This work has clear potential for translation from a basic laboratory discovery to a melanoma therapy," said Michael Jackson, Ph.D., vice president of drug discovery and development at Sanford-Burnham. "We are excited to begin the screening process to identify a new class of drugs to treat cancer."