ATF2 is one of the most abundant cAMP response element (CRE)-binding proteins and is expressed in most cells and tissues. ATF2 mRNA is ubiquitously distributed in a variety of organs, such as the brain, liver, heart, lung, pancreas, and in several cell types, such as T cells, macrophages, blood mononuclear cells, and cancer cells, suggesting that ATF2 is important for signal transduction and cellular function in these organs. ATF2 can bind with high affinity to CRE (TGACGTCA) as a homodimer or as a heterodimer with c-Jun and can activate transcription from promoters containing CRE (Matsuda et al. 1991; Nomura et al. 1993). ATF2 acts as a stronger transactivator as a heterodimer with c-Jun than as a homodimer. It can also bind to activator protein 1 (AP-1) consensus sequences (5′-TGACTCA-3′). Numerous reports have shown that the DNA-binding domain containing a “B-ZIP” structure and the amino-terminal transcriptional activation domain containing a metal finger structure are both critically necessary for the transactivation activity of ATF2 (Matsuda et al. 1991).
ATF2 is mainly located in nucleus, cytoplasm, and mitochondrial outer membrane, where it plays different roles. In the nucleus, ATF2 contributes to global transcription and the DNA damage response, both of which are related to cell development, proliferation, and death. In the cytoplasm, ATF2 can interact with HK1- and VDAC1-containing complexes in the mitochondrial outer membrane, which can cause damage to the membrane, inducing mitochondrial leakage and promoting cell death.
The ATF2 subfamily contains two other members: CRE-BPa (also known as CREB5) and ATF7 (also known as ATF-a) (Matsuda et al. 1991; Nomura et al. 1993). All three proteins share the same transactivation domain, which consists of a metal finger element, stress-activated protein kinase (SAPK) phosphorylation sites, and a B-ZIP DNA-binding domain; this transactivation domain is widely expressed in various tissues and cells (Takeda et al. 1991). Interestingly, ATF2 can be phosphorylated by both p38 and JNK, while ATF7 is only phosphorylated by p38 (De Graeve et al. 1999). Moreover, both ATF2 and CRE-BPa can activate transcription from CRE promoters through interactions with co-activator CBP (Sano et al. 1998), but ATF7 represses transcription.
ATF2 Activates Signaling Pathways
In response to pathogen infection and psychological stress, toll-like receptors (TLRs), which play an important role in the innate immune response by recognizing surface patterns on microbial invaders, are stimulated. TLR signaling is initiated at intracellular Toll/interleukin-1 receptor (TIR) domains, such as MyD88, TIRAP, and TRIF (Yu et al. 2014). The induction of inflammatory cytokines requires MyD88, which recruits IL-1 receptor-associated kinase-4 (IRAK4) to the TLRs through interaction with the death domains of both molecules. IRAK1 is activated via phosphorylation and then associates with TRAF6, ultimately resulting in the activation of the mitogen-activated protein kinases (MAPKs) JNK, p38, and ERK (Blasius and Beutler 2010; Kawai and Akira 2010). ATF2 is then phosphorylated by these activated upstream signaling factors, p38 and JNK, at amino acids Thr69 and Thr71. The phosphorylated ATF2 may then form homodimers or heterodimers with other members of the ATF/CREB family and Fos/Jun family (De Cesare et al. 1995; Vlahopoulos et al. 2008) (Fig. 2).
Biological Functions of ATF2
A large number of reports indicate that ATF2 is critically involved in various biological activities. ATF2 complexes stimulate the transcription of a variety of genes implicated in inflammation, such as cell adhesion molecules (CAMs), proinflammatory cytokines, and chemokines. CAM proteins include integrins, cadherins, and selectins (E-selectin, P-selectin, and L-selectin). Selectins participate in the initial recruitment of leukocytes to the site of injury during inflammation. VCAM-1 may influence the development of atherosclerosis and rheumatoid arthritis. As reported in ATF2-deficient mice, E-selectin, P-selectin, and VCAM-1 expression in the lungs and kidneys following lipopolysaccharide stimulation was clearly reduced compared with control mice (Dinarello 2000). The proinflammatory cytokine tumor necrosis factor alpha (TNF-α), produced mainly by macrophages, lymphoid cells, mast cells, and adipose tissue, causes a variety of clinical inflammatory disorders, such as rheumatoid arthritis, psoriasis, refractory asthma, and inflammatory bowel disease. In ATF2-knockout mice, TNF-α expression was dramatically suppressed. Moreover, expression of interleukin (IL-)-1𝛽 and IL-6 was also significantly inhibited in ATF2-deficient mice (Reimold et al. 2001). Further, the soluble factor keratinocyte chemoattractants are the most highly inducible chemokines produced as a result of IL-1 and TNF-α expression. They are involved in chemotaxis, cell-mediated activation of neutrophils, and the neutrophil inflammatory response. Importantly, in ATF2-deficient mice, the expression of these keratinocyte chemoattractants was obviously suppressed (Reimold et al. 2001).
Potential Roles of ATF2 in the Pathogenesis of Diseases
Numerous studies have demonstrated that the ATF2 protein plays a critical role in cell proliferation, apoptosis, inflammation, and cancer. Particularly, in vitro and in vivo studies in human and mouse cell lines, as well as in knockout mice, have revealed that the overactivation of ATF2 is associated with several inflammatory diseases, including obesity, hepatitis, inflammatory pain, and allergic asthma.
The white adipose tissue (WAT) that accumulates in obesity displays multiple markers of inflammation, with generation of reactive oxygen species (ROS) and progressive infiltration by macrophages. Increasing evidence indicates that this phenomenon is due to insulin resistance and adipokine dysregulation (Miyata et al. 2013). As examined in genetically obese (ob/ob) mice, both total and phosphorylated ATF2 are highly expressed in macrophages infiltrating the WAT. Additionally, in ATF2-mutant mice, WAT is significantly less abundant than in wild-type mice (Maekawa et al. 2010). Treatment of mice with inhibitors of p38-ATF2 signaling suppresses adipocyte differentiation and WAT accumulation. Additionally, this inhibition may counteract high-fat diet-induced obesity, insulin resistance, macrophage infiltration into WAT, and the associated increase in TNF-α expression (Maekawa et al. 2010). Therefore, the evidence for p38-linked ATF2 signaling as a regulator of obesity-related inflammation may lend insight into the pathological effects of overnutrition.
ATF2 is critically involved in the regulation of the most common etiology of liver inflammation. ATF2 is reported to suppress activity at the hepatitis B virus X promoter through competition for the AP-1 binding site and through formation of an ATF2-Jun heterodimer (Choi et al. 1997). In our previous research, mice who had hepatitis induced by treatment for 7 days with D-gal/LPS exhibited significantly higher levels of phosphorylated ATF2 expression compared to the control group (Shen et al. 2013). Upon treatment with ATF2 inhibitors, the symptoms of hepatitis in the mice regressed in parallel with a decrease in activated ATF2. In addition, knockdown of both ATF2 and ATF7 induced severe abnormalities in the developing liver and heart, resulting in embryonic death in mice and suggesting the potential role of ATF2 in liver inflammation (Breitwieser et al. 2007).
As reported by Fang group, in rats with chronic inflammatory pain induced by Complete Freund’s Adjuvant (CFA), cells expressing phospho (p)-ATF2-IR were found to be greatly accumulated in the spinal dorsal horn (Fang et al. 2013). Following treatment with medical therapy, ankle swelling and pain signs were decreased in parallel with a decline in the number of p-ATF2-IR-expressing cells. Therefore, these data seem to suggest an important role for ATF2 in the regulation of inflammatory pain.
Inflammation in allergen-induced asthma is mainly regulated by the release of eicosanoids (Bickford et al. 2012), which are bioactive lipids with both anti- and pro-inflammatory actions in pulmonary tissues. In a mouse model of allergic asthma, Aspergillus fumigatus induces cPLA2 (IVC PLA2 (phospholipase A2)) secretion in eosinophils and TNF-expression in lung epithelial cells through macrophage activation due to the recruitment of ATF2/JUN, RELA/RELA (p65/p65), and USF1/USF2 complexes to the PLA2G4C enhancer in lung epithelial cells in response to TNF stimulation (Bickford et al. 2012).
Brain Inflammatory Disorders
ATF2 has been shown to have a vital role in brain inflammation (Lin et al. 2014). Endothelin-1 (ET-1), which contributes to inflammatory responses, was increased in several brain inflammatory disorders through the upregulation of the cyclooxygenase-2 (COX-2)/prostaglandin E2 (PGE2) system. As clarified by Lin CC (Lin et al. 2014), in mouse brain microvascular endothelial cells, this inflammation was due to stimulation of intracellular Ca2+ release; subsequent phosphorylation of PKC-α, CaMKII, and the MAPKs ERK1,2, p38, MAPK, and JNK1,2; and finally to activation of ATF2/AP-1. These results demonstrate that Ca2+/PKC-α/CaMKII/MAPKs/ATF2/AP-1 signaling is essential for the ET-1-induced COX-2 upregulation that contributes to brain injury and inflammatory diseases.
Neuroinflammation commonly occurs in various neurological diseases. In a study by Kumar et al., ATF2 was shown to be essential in the regulation of neuroinflammation (Kumar et al. 2015). Knockdown of ATF2 inhibited the production of proinflammatory cytokines, particularly IL-6 and TNF-α, in microglia. Interestingly, miR-26a directly targeted ATF2 and caused its downregulation. Taken together, these findings indicate that miR-26a/ATF2 could be potential therapeutic molecular targets for neurological inflammation.
Moreover, in Alzheimer’s, Parkinson’s, and Huntington’s diseases, ATF2 is downregulated in the hippocampus and caudate nucleus (Huang et al. 2016; Pearson et al. 2005), implying that it may be essential for regulation of neuronal viability and normal neurological function. Moreover, ATF2 may be an active component in autoimmune disease, vascular homeostasis, and angiogenesis (Liao et al. 2010; Licht et al. 2006).
ATF2 in Cancer
ATF2 has been investigated as a potential carcinogenic biomarker of certain types of cancers. A large body of evidence has demonstrated that ATF2 is markedly overexpressed in human non-small cell lung carcinoma (NSCLC) cells and tissues (You et al. 2016). ATF2 was closely associated with adverse clinical characteristics such as TNM stage, tumor size, and metastasis due to its role in regulating cell proliferation. A Kaplan-Meier analysis indicated that patients with high levels of ATF2 and p-ATF2 had a significantly shorter overall survival compared with patients exhibiting low expression levels. Interestingly, as microRNAs play a critical role in cancer development and progression, miR-204 was shown to suppress the development of NSCLC by inversely affecting endogenous ATF2 expression at both the mRNA and protein levels in vitro (Zhang et al. 2016). Moreover, ATF2 elicits oncogenic activities in melanoma, and its inhibition attenuates melanoma development through mediation of pigmentation, immune infiltration, and metastatic propensity (Claps et al. 2016). ATF2 also plays a significant role in renal cell carcinoma (RCC) (Wu et al. 2016). ATF2 knockdown in RCC cells reduced proliferative and metastatic potential by regulating the transcription of downstream targets, including CyclinB1, CyclinD1, Snail, and Vimentin. This indicates that ATF2 expression could be closely correlated with aggressive clinicopathological characteristics and predicts the poor prognosis of RCC patients. In addition, ATF2 was involved in regulating liver cancer cell migration targeting miR-451 (Lv et al. 2014), while ATF2 expression suppressed tumor formation in an orthotopic model of liver cancer and cellular transformation in vitro (Gozdecka et al. 2014). Targeting the miR-622/ATF2 axis was found to be a novel therapeutic approach for blocking glioma invasion (Zhang et al. 2015). As reported by Zhao Y. et al., JNK2/ATF2 signaling was critically involved in human gastric cancer development (Zhao et al. 2014). ATF2 has been implicated as a tumor suppressor in breast cancer (Rudraraju et al. 2014), while ATF2-targeting miR-10b promoted breast cancer cell motility and invasiveness (Ibrahim et al. 2012). In addition, ATF2 also plays a vital role in other series of cancer types, such as head and neck carcinoma (Duffey et al. 2011), mesenchymal tumors (Endo et al. 2014), lung cancer (Arora et al. 2011), pancreatic cancer (Xu et al. 2012), prostate cancer (Nair et al. 2010), and skin cancer (Bhoumik et al. 2008). Taken together, these results indicate that ATF2 is a critical regulator of the occurrence and development of various types of cancers.
The ATF2 subfamily is a family of transcription factors that share a bZIP domain. Increasing evidence has demonstrated that ATF2 is critically involved in the regulation of a variety of signaling pathways, induced by factors such as growth factors, inflammatory factors, environmental stress, and radiation. This involvement indicates its significant role in mediating cellular functioning, including cell proliferation, apoptosis, and invasion. This rapidly expanding body of data on ATF2 as a proinflammatory, regulatory protein and cancer mediator calls for the investigation of ATF2 as a molecular target for treating these kinds of diseases. This review highlights the evidence that ATF2 most likely plays an essential role in the pathogenesis of inflammation. A critical question is whether this evidence, derived from cell culture and mouse models, can be transformed to benefit human health; its clinical utility and therapeutic index in humans have yet to be determined. Upon this determination, the identification of specific and effective inhibitors that target ATF2 could be designed and synthesized based on structural and functional domains of ATF2 to augment treatments for human inflammatory and cancer diseases.
- Fang JQ, Du JY, et al. Intervention of electroacupuncture on spinal p38 MAPK/ATF-2/VR-1 pathway in treating inflammatory pain induced by CFA in rats. Mol Pain. 2013;9:13.Google Scholar