Activating transcription factor 4 (ATF4) is a transcription factor encoded by the ATF4 gene (Gene ID: 468) located on human chromosome 22q13.1.[1] Also known as CREB-2, it belongs to the basic leucine zipper (bZIP) superfamily of DNA-binding proteins and serves as a central regulator in cellular stress responses, particularly the integrated stress response (ISR) and the unfolded protein response (UPR).[1][2]ATF4 features a basic DNA-binding domain that recognizes cAMP-responsive elements (CREs) with the consensus sequence TGACGTCA, and a leucine zipper domain enabling homo- or heterodimerization with partners like C/EBPβ or CHOP (DDIT3).[3] Its activity is tightly controlled at the translational level through two upstream open reading frames (uORFs) in its 5' untranslated region, which allow selective translation during eIF2α phosphorylation by stress-activated kinases such as PERK, GCN2, PKR, or HRI.[2] Post-translational modifications, including phosphorylation at sites like S219 and ubiquitination, further modulate its stability and function.[2]In physiological contexts, ATF4 orchestrates adaptive gene expression programs essential for survival under stress, activating targets involved in amino acid metabolism (e.g., ASNS), redox homeostasis, autophagy, and ER homeostasis.[3][2] It plays critical roles in osteoblast differentiation and bone mineralization, as well as in neuronal synaptic plasticity and memory formation.[1][2]Dysregulated ATF4 contributes to various pathologies, acting as a pro-survival factor in cancers such as glioblastoma, where it promotes proliferation and angiogenesis, and in breast cancer, enhancing radioresistance.[3] In neurodegeneration, it influences proteostasis and has been linked to conditions like Alzheimer's disease through ISR pathway activation.[2] Additionally, ATF4 modulation affects metabolic disorders, including antipsychotic-induced weight gain,[4] and antiviral defenses during infections like those caused by HTLV-1.[1][2]
Gene
Genomic Location and Structure
The ATF4 gene is located on the long arm of human chromosome 22 at band q13.1, spanning positions 39,519,695 to 39,522,690 on the forward strand (GRCh38 assembly).[5] In the mouse, the orthologous Atf4 gene resides on chromosome 15 at positions 80,139,385 to 80,141,742 (GRCm39 assembly).The human ATF4 gene covers approximately 3 kb and comprises three exons interrupted by two introns, with the first exon containing the 5' untranslated region (UTR) and upstream open reading frames (uORFs) critical for its regulation.[1] There are two main transcript variants: NM_182810.3 (3 exons, uORF-regulated) and NM_001675.4 (2 exons, internal ribosome entry site-regulated), both encoding the same protein. The exon-intron boundaries follow the GT-AG rule, and the coding sequence is distributed such that exon 1 includes part of the 5' UTR and uORF1, exon 2 includes the remainder of the 5' UTR with uORF2 and the start codon followed by initial coding sequence, and exon 3 encodes the remainder of the protein-coding region including the basic leucine zipper (bZIP) domain.[1] The mouse Atf4 gene exhibits a similar organization, spanning about 2.4 kb with three main exons.The promoter region of the ATF4 gene, located upstream of exon 1, features multiple regulatory elements, including binding sites for CCAAT/enhancer-binding protein β (C/EBPβ) isoforms that mediate transcriptional repression under certain stress conditions.[6]Sequence conservation of the ATF4 gene is high across mammals, with coding regions showing over 90% identity between human and mouse orthologs, and orthologs present in more than 270 vertebrate species.[7] This conservation extends to non-coding elements like the 5' UTR uORFs, underscoring evolutionary pressures to maintain stress-responsive mechanisms. Within the ATF/CREB family, ATF4 diverged early from other members like ATF2 and ATF3, acquiring specific adaptations in its bZIP domain for heterodimerization and target gene specificity.[7]
Expression Patterns
ATF4 exhibits basal expression across most human tissues, with relatively low to moderate levels in the majority, but notably higher expression in the brain, liver, and pancreas.[8] In the brain, ATF4 RNA and protein are detected at elevated levels in neuronal tissues, while in the liver and pancreas, expression supports metabolic functions under normal conditions.[9] These patterns have been characterized using RNA sequencing data, showing transcripts per million (TPM) values that cluster ATF4 with genes involved in stress-responsive pathways, though specific TPM ranges vary by tissue subtype.[10]Under stress conditions, ATF4 expression is rapidly induced in various cell types. Endoplasmic reticulum (ER) stress triggers ATF4 upregulation through the unfolded protein response, particularly in osteoblasts and neurons, where it peaks within hours of exposure to agents like tunicamycin.[11] Nutrient deprivation, such as amino acid limitation, similarly induces ATF4 in multiple cell lines, including fibroblasts and hepatocytes, enhancing transcription of target genes like asparagine synthetase.[12] In hematopoietic stem cells, ER stress leads to high ATF4 induction promoting apoptosis, whereas progenitor cells show moderated levels.[11] These induction dynamics are often quantified via quantitative PCR (qPCR) for mRNA levels or Western blot for protein accumulation, as demonstrated in studies using thapsigargin-treated cells.[13]During development, ATF4 expression is upregulated in embryonic skeletal tissues. In mouse embryos, ATF4 is broadly expressed early but becomes restricted to chondrocytes and osteoblasts from embryonic day 14 onward, supporting differentiation and matrix production.[14] This pattern is evident in Western blots of nuclear extracts from skeletal elements, showing increased ATF4 protein in proliferating and differentiating osteoblasts.[15] Key studies have used in situ hybridization and qPCR to confirm ATF4's role in timely osteoblast onset during endochondral ossification.[16]
Protein
Primary Structure and Domains
The human ATF4 protein consists of 351 amino acids, with a calculated molecular weight of approximately 38.7 kDa.[17] This bZIP family transcription factor features a modular architecture, including an N-terminal transactivation domain (TAD) spanning residues 1–275, which is intrinsically disordered and facilitates recruitment of co-activators to promote gene expression.[18] The C-terminal region contains the basic leucine zipper (bZIP) domain, encompassing residues 276–351, which is responsible for DNA binding and protein dimerization.[19]Within the bZIP domain, the basic region (residues 280–301) is rich in positively charged amino acids that enable sequence-specific binding to DNA motifs such as the CRE (cAMP response element), while the leucine zipper (residues 302–341) mediates homo- or heterodimerization through hydrophobic interactions.[20] Structural studies reveal that the bZIP domain adopts an extended α-helical conformation upon DNA binding or dimerization, with the basic region forming a continuous helix that inserts into the major groove of DNA, and the zipper region consisting of two amphipathic α-helices coiled into a parallel dimer.[21] This secondary structure is critical for the stability and specificity of ATF4's interactions, as evidenced by crystallographic analysis of the ATF4-C/EBPβ heterodimer.[22]Sequence comparisons across species highlight high conservation of functional elements in ATF4, particularly in the bZIP domain. The human ATF4 shares approximately 87% amino acid identity with its mouse ortholog, with near-complete conservation (>95%) in the leucine zipper and basic regions, underscoring the evolutionary preservation of residues essential for DNA recognition and dimerization.[23] These conserved motifs ensure functional equivalence in stress response pathways between human and rodent models.
Post-Translational Modifications
ATF4 undergoes several post-translational modifications that regulate its stability, transcriptional activity, and cellular localization, particularly in response to stress signals. These modifications include phosphorylation, ubiquitination, sumoylation, and acetylation, which collectively fine-tune ATF4's role as a transcription factor in the integrated stress response.Phosphorylation occurs at multiple sites on ATF4, influencing its activation and degradation. For instance, phosphorylation at Ser219 creates a recognition motif for the E3 ubiquitin ligase βTrCP, promoting ATF4 ubiquitination and subsequent proteasomal degradation under non-stress conditions.[11] Phosphorylation at Ser245 by ribosomal S6 kinase 2 (RSK2) enhances ATF4's transactivation activity, particularly in osteoblasts where it drives expression of genes like osteocalcin, and increases overall transcriptional potency.[11] Other sites, including Ser215 by casein kinase 2, further boost ATF4 activity during stress.[24] These modifications integrate with ERstress signaling by stabilizing ATF4 when phosphorylation is reduced, allowing its accumulation to activate stress-response genes.[2]Ubiquitination targets ATF4 for rapid proteasomal degradation in unstressed cells, maintaining low basal levels. This process is phosphorylation-dependent, with βTrCP recognizing the phospho-Ser219 motif to polyubiquitinate ATF4 at lysine residues such as Lys45, Lys53, Lys55, Lys75, Lys88, Lys92, Lys277, and Lys335.[25] Under non-stress conditions, this SCF^βTrCP-mediated ubiquitination ensures ATF4's short half-life of less than 1 hour, preventing ectopic activation. Mutagenesis studies confirm this: substitution of Ser219 with alanine (S219A) abolishes βTrCP binding, inhibits ubiquitination, and stabilizes ATF4, leading to enhanced transcriptional output and cellular resistance to stress-induced apoptosis.[11]Sumoylation at N-terminal lysines, including Lys45 and Lys53, modifies ATF4 and may compete with ubiquitination at shared sites, potentially influencing its stability and interactions, though the precise functional impacts remain under investigation.[24]Acetylation also regulates ATF4's stability and activity. The histone acetyltransferase p300 binds to ATF4, inhibiting its ubiquitination and thereby increasing protein half-life independently of its acetyltransferase activity; p300 also acetylates ATF4 at Lys311 and within residues 270-300 to enhance transcriptional efficacy.[26] Similarly, CREB-binding protein (CREBBP) acetylates ATF4 in the 270-300 region, enhancing its ability to activate target genes.[24] Mutagenesis of these lysine sites, such as K311R, reduces acetylation and impairs stress-induced gene expression, though stability in the presence of p300 remains unaffected.[11]
Regulation
Transcriptional Control
The transcriptional control of the ATF4 gene involves specific promoter elements that integrate stress signals to modulate its expression levels. The core promoter region spans approximately 2.5 kb upstream of the transcription start site and drives basal transcription, with stress-specific regulation occurring through binding sites for key transcription factors. Under endoplasmic reticulum (ER) stress induced by agents like thapsigargin, ATF4 transcription increases approximately 3- to 4-fold, as measured by quantitative RT-PCR and luciferase reporter assays in fibroblasts, contributing to amplification of the integrated stress response.[27]A prominent repressive mechanism operates during ultraviolet (UV) stress, where CCAAT/enhancer-binding protein β (C/EBPβ), particularly its liver inhibitory protein (LIP) isoform, binds to two critical C/EBP consensus sites in the ATF4 promoter at positions -950 to -935 bp (TAAATAGCAATCAAT) and -874 to -859 bp (TTGCAAATAATCACT). This binding, confirmed by chromatin immunoprecipitation (ChIP) assays, represses promoter activity, reducing ATF4 mRNA levels by about 3-fold within 6 hours post-UV irradiation (40 J/m²), thereby fine-tuning the stress response to prevent excessive activation. In contrast, activating transcription factor 6 (ATF6), a UPR sensor, indirectly supports ATF4 expression during ER stress by promoting overall UPR gene transcription, though direct binding to the ATF4 promoter has not been established.[6][6]Under oxidative stress, nuclear factor erythroid 2-related factor 2 (NRF2) directly transactivates ATF4 by binding to an antioxidant response element (ARE) within its promoter, as demonstrated in retinal pigment epithelial cells exposed to electrophiles or ER stressors. This NRF2-mediated activation enhances ATF4 mRNA expression to bolster antioxidant defenses, with ChIP assays verifying NRF2 occupancy at the ARE site. Similarly, in endothelial cells subjected to oxidative conditions, NRF2 binding to the ATF4 promoter upregulates its transcription, linking the oxidative stress response to ISR pathways.[28][29]Enhancer regions upstream of the ATF4 gene, including potential distal elements identified through genomic mapping, exhibit dynamic epigenetic modifications that distinguish basal from stress-induced states. In basal conditions, the ATF4 promoter displays moderate histone H3lysine 27 acetylation (H3K27ac) and open chromatin accessibility, supporting low-level expression. Upon stress activation, such as ER or oxidative insults, H3K27ac enrichment increases at these enhancers and promoter-proximal regions, facilitating transcription factor recruitment and RNA polymerase II pausing release, though genome-wide studies note variable changes without uniform upregulation across all ISR contexts. Histone H3lysine 4 monomethylation (H3K4me1), a hallmark of active enhancers, is present at select upstream sites under basal conditions and persists or intensifies during stress to maintain accessibility.[30][31]Feedback loops in ATF4 transcriptional regulation primarily occur indirectly through downstream effectors. For instance, ATF4 induces CHOP expression, which in turn can modulate ISR components, but direct auto-regulation of the ATF4 promoter by ATF4 itself has not been observed in reporter assays under stress. ChIP-seq studies have mapped transcription factor binding across the ATF4 locus, identifying enriched motifs for C/EBP family members and NRF2 at promoter and enhancer regions in stressed cells, such as fibroblasts and epithelial lines, highlighting context-dependent regulation. These findings, derived from seminal works using luciferase reporters, ChIP, and sequencing, underscore the promoter's role as a stress-responsive hub without evidence for p53-mediated repression.[27][6][30]
Translational Control
The translation of ATF4 mRNA is tightly regulated by two upstream open reading frames (uORFs), uORF1 and uORF2, located in its 5' untranslated region (UTR), which suppress protein synthesis under normal cellular conditions.[32] uORF1, encoding a short peptide of three amino acids, permits efficient ribosomal reinitiation, while uORF2, encoding a longer 59-amino-acid peptide that overlaps the ATF4 coding sequence, predominantly captures reinitiating ribosomes, thereby inhibiting translation of the main ATF4 open reading frame (ORF).[33] This uORF-mediated repression ensures low basal ATF4 expression in unstressed cells.[32]During endoplasmic reticulum (ER) stress, the mechanism shifts to allow selective ATF4 translation through delayed ribosomal reinitiation and bypassing of uORF2, triggered by phosphorylation of eukaryotic initiation factor 2 alpha (eIF2α).[33] Phosphorylated eIF2α reduces ternary complex levels (eIF2-GTP-Met-tRNAi), slowing the recruitment of ribosomes to uORF1 and extending the time available for them to scan past uORF2 to reach the ATF4 ORF.[32] This ribosomal bypassing is facilitated by the positive role of uORF1 in promoting reinitiation at the ATF4 ORF under low eIF2-GTP conditions, integrating ATF4 translation into the unfolded protein response (UPR) pathway.[33]Quantitative models of this process indicate that ER stress induces a approximately 20-fold increase in ATF4 translation efficiency compared to basal levels, reflecting the enhanced ribosomal flux to the main ORF.[32] Experimental validation through luciferase reporter assays fused to the ATF4 5' UTR has demonstrated that mutating uORF2 alone results in constitutive high expression, while stress-induced increases are abolished in cells expressing non-phosphorylatable eIF2α (S51A mutant), confirming the eIF2α-dependent mechanism.[33] Furthermore, PERK knockout studies show that loss of this eIF2α kinase prevents ATF4 induction during ER stress, underscoring PERK's essential role in uORF-mediated translational control.[32]
Biological Functions
Role in Unfolded Protein Response
ATF4 plays a central role in the PERK branch of the unfolded protein response (UPR), a cellular mechanism activated by endoplasmic reticulum (ER) stress to restore proteostasis. Upon accumulation of unfolded proteins in the ER, the kinase PERK oligomerizes and autophosphorylates, leading to phosphorylation of eukaryotic initiation factor 2 alpha (eIF2α) at serine 51. This phosphorylation generally attenuates global protein translation but selectively enhances the translation of ATF4 mRNA through a mechanism involving upstream open reading frames (uORFs) that allow ribosomal reinitiation at the ATF4 coding sequence under stress conditions.[33] Once translated, ATF4 translocates to the nucleus, where it binds to amino acid response elements (AAREs) or C/EBP-ATF response elements (CAREs) as a homodimer or heterodimer to activate transcription of UPR target genes.In the adaptive phase of the UPR, ATF4 promotes ER homeostasis by upregulating genes involved in protein folding and degradation. For instance, ATF4 directly activates the promoter of the chaperone BiP (also known as GRP78) via an upstream ATF/CRE site, independent of classical ER stress elements, thereby increasing BiP levels to facilitate proper protein folding and alleviate ER stress.[34] Additionally, ATF4 induces expression of ER-associated degradation (ERAD) components, such as Herp (homocysteine-inducible ER protein), which coordinates retrotranslocation of misfolded proteins from the ER to the cytosol for proteasomal degradation, preventing their accumulation.[35] These actions represent key elements of the ATF4-specific branch in the UPR pathway diagram, where PERK-eIF2α signaling converges on ATF4 to branch toward pro-survival adaptations before potentially shifting to pro-apoptotic outcomes.Under prolonged or severe ER stress, ATF4 contributes to the transition to apoptosis by forming heterodimers with C/EBP homologous protein (CHOP, also known as GADD153). ATF4 induces CHOP transcription by binding to CARE sites in the CHOP promoter, and the resulting ATF4-CHOP heterodimers then activate pro-apoptotic genes while repressing anti-apoptotic factors, such as Bcl-2, thereby promoting cell death to eliminate irreparably stressed cells. This heterodimerization is a critical node in the UPR pathway, illustrating how the ATF4 arm integrates adaptive and terminal responses to ER stress.
Involvement in Amino Acid Metabolism and Oxidative Stress
ATF4 plays a central role in the integrated stress response (ISR), where it is selectively translated in response to amino acid deprivation through phosphorylation of eukaryotic initiation factor 2 alpha (eIF2α), enabling cells to adapt by reprogramming gene expression to restore nutrient homeostasis.[36] Under conditions of amino acid limitation, ATF4 binds to amino acid response elements (AAREs) in the promoters of target genes, inducing the expression of amino acid transporters such as SLC7A11 (also known as xCT), which facilitates cystine uptake to support glutathione synthesis and mitigate oxidative damage.[37] Additionally, ATF4 upregulates biosynthetic enzymes, including asparagine synthetase (ASNS), to replenish non-essential amino acids and sustain protein synthesis during stress.[38] ATF4 also induces autophagy genes, promoting autophagic flux as an adaptive mechanism to recycle cellular components under nutrient scarcity.[39]In the context of oxidative stress, ATF4 coordinates an antioxidant defense program by activating genes involved in redox balance, such as those in the glutathione pathway, allowing cells to counteract reactive oxygen species (ROS) accumulation.[36] For instance, through its regulation of SLC7A11/xCT, ATF4 enhances cystine import, which is rate-limiting for glutathione production, thereby protecting against ROS-induced damage in various cell types.[40] This transcriptional activity integrates with the broader ISR to promote cellular adaptation, balancing metabolic demands with survival under nutrient scarcity and oxidative challenges.[41]Evidence from ATF4 knockout models underscores its essential function in these processes; Atf4^{-/-} embryonic fibroblasts exhibit impaired induction of amino acid importer genes and reduced glutathione biosynthesis, leading to heightened sensitivity to oxidative stress and defective metabolic adaptation.[36] In vivo, ATF4-deficient mice display perinatal lethality associated with metabolic defects, including failures in amino acid transport and utilization in developing tissues like the lens and skeleton, highlighting ATF4's non-redundant role in stress-induced metabolic resilience.
Molecular Interactions
Protein-Protein Interactions
ATF4, a member of the basic leucine zipper (bZIP) family of transcription factors, primarily engages in protein-protein interactions through its conserved bZIP domain, which facilitates dimerization with other bZIP proteins to modulate transcriptional activity. These heterodimers enable ATF4 to co-activate or repress target genes in response to cellular stress, such as endoplasmic reticulum (ER) stress or amino acid deprivation.[42]A key interaction occurs between ATF4 and CHOP (also known as DDIT3), where the two proteins form heterodimers that negatively regulate stress-induced gene expression, including the repression of asparagine synthetase (ASNS) during amino acid limitation and ER stress. This partnership promotes pro-apoptotic signaling in stressed cells, as evidenced by yeast two-hybrid screening that identified CHOP as an ATF4 binding partner, followed by co-immunoprecipitation (co-IP) confirmation of their in vivo association. The functional impact includes reduced ASNS promoter activity upon CHOP overexpression, highlighting CHOP's role in fine-tuning ATF4's transactivation potential.[43][44]ATF4 also heterodimerizes with C/EBPβ (CEBPB), a interaction mediated by their bZIP domains, which enhances ATF4's role in processes like skeletal muscle atrophy and osteoblast differentiation. Crystal structures of the ATF4-C/EBPβ bZIP heterodimer reveal a stable coiled-coil interface that directs binding to specific DNA motifs, and co-IP studies have validated this partnership in cellular contexts. This complex influences gene expression outcomes related to cellular adaptation under stress.[45]Heterodimerization with ATF3 further expands ATF4's regulatory scope, particularly in the integrated stress response, where the pair upregulates pro-apoptotic genes and modulates stress adaptation. This interaction, observed through binding assays and chromatin immunoprecipitation, allows ATF3 to compete or cooperate with ATF4 at response elements, altering transcriptional outputs in stressed cells.[46][42]In addition to bZIP partners, ATF4 interacts with co-activators such as CBP and p300, which acetylate ATF4 to stabilize the protein and enhance its transcriptional activity by inhibiting ubiquitination and promoting chromatin remodeling. Co-IP experiments have confirmed these associations, demonstrating p300's binding to ATF4's transactivation domain during stress conditions. These interactions contribute to broader geneactivation without directly altering dimerization specificity.[47]ATF4 engages with upstream kinases like PERK in the ER stress signaling pathway, where PERK phosphorylates eIF2α to preferentially translate ATF4 mRNA, thereby relaying signals for stress adaptation; while direct binding evidence is limited, functional co-regulation has been established through pathway inhibition studies. Yeast two-hybrid and co-IP approaches across these interactions underscore their reliability, with heterodimers often leading to context-dependent transcriptional co-activation or repression.[42][44]
Target Genes and Pathways
ATF4, a basic leucine zipper (bZIP) transcription factor, primarily binds to cyclic AMP response element (CRE) and ATF sites within the promoters of target genes to regulate their expression in response to cellular stress.[48] Notable direct targets include ASNS, encoding asparagine synthetase, which is activated by ATF4 binding to the nutrient-sensing response element-1 (NSRE-1) in its promoter during amino acid deprivation.[12] Similarly, ATF4 directly binds to the promoter of PPP1R15A (encoding GADD34), a gene involved in dephosphorylating eIF2α, as demonstrated by chromatin immunoprecipitation (ChIP) assays under endoplasmic reticulum (ER) stress conditions.[49]Genome-wide ChIP-seq studies have identified approximately 200–300 direct ATF4 target genes under stress conditions, with binding often occurring at C/EBP-ATF response elements (CARE) in promoter regions.[50] For instance, one systematic compilation lists 234 ATF4 targets, enriched in stress-responsive categories, while ChIP-seq in stressed cells reveals hundreds of occupied sites priming genes for activation.[50][30] These targets span multiple functional groups, reflecting ATF4's role in coordinating adaptive responses.ATF4 influences several key pathways through its targets. In the unfolded protein response (UPR), it regulates genes such as XBP1, enhancing ER homeostasis by promoting XBP1 mRNA expression independently of ATF6.[51] Within the integrated stress response (ISR), ATF4 upregulates DDIT4 (encoding REDD1), which inhibits mTORC1 to conserve resources during nutrient scarcity.[52] In angiogenesis, ATF4 drives VEGFA expression via direct transcriptional activation, particularly under hypoxia or amino acid restriction, facilitating vascular remodeling.[53][54]ATF4 exhibits a dual role as both an activator and repressor, context-dependent on dimerization partners; for example, ATF4-CHOP heterodimers repress certain pro-survival genes while activating pro-apoptotic ones like PUMA.[55] This versatility allows ATF4 to fine-tune gene expression across stress pathways.[2]
Clinical Significance
Implications in Cancer
ATF4 plays a dual role in cancer progression, acting as both a promoter and suppressor depending on the tumor context and cellular environment. In many solid tumors, ATF4 activation under endoplasmic reticulum (ER) stress enhances cancer cell survival, proliferation, and dissemination by regulating genes involved in amino acid transport, redox balance, and vascularization. This pro-oncogenic function is particularly evident in aggressive malignancies where ATF4 drives adaptive responses that confer resistance to therapies and environmental stresses.[56]In glioblastoma, ATF4 promotes tumor angiogenesis and metastasis by transcriptionally upregulating SLC7A11 (also known as xCT), a cystine-glutamate antiporter that supports glutathione synthesis and ferroptosis resistance, thereby facilitating vascular remodeling and glioma cell invasion. Similarly, in breast cancer, ATF4 enhances metastasis to distant sites, such as the lungs, through induction of vascular endothelial growth factor (VEGF) expression, which stimulates neovascularization and tumor dissemination in triple-negative subtypes. In colorectal cancer, ATF4 contributes to metastatic potential by activating glutamine metabolism genes like GPT2, enabling nutrient adaptation and progression in hypoxic tumor microenvironments. These mechanisms highlight ATF4's role in fostering a permissive niche for tumor spread across these cancer types.[3][57][56]Conversely, ATF4 exhibits context-dependent tumor-suppressive effects in prostate cancer, where its overexpression inhibits tumor growth by downregulating androgen receptor (AR) expression via the PERK/eIF2α/ATF4 axis, thereby inducing apoptosis and sensitizing cells to ER stress. This suppressive action is linked to activation of pro-apoptotic pathways, such as CHOP-mediated signaling, which counters proliferation in castration-resistant prostate cancer models. Such duality underscores ATF4's reliance on upstream regulators and downstream targets for its oncogenic or tumoricidal outcomes.[58][59]Therapeutic strategies targeting ATF4 have emerged for ER stress-dependent tumors, with inhibitors of the upstream PERK pathway showing promise in preclinical and early clinical settings. For instance, the selective PERK inhibitor HC-5404 is in phase I clinical trials for advanced solid tumors, including renal cell carcinoma, where it reduces ATF4 activation to exacerbate proteotoxic stress and enhance sensitivity to vascular endothelial growth factorreceptor tyrosine kinase inhibitors. These approaches exploit ATF4's pro-survival functions in the unfolded protein response to selectively eliminate cancer cells reliant on ER homeostasis.[60]
Roles in Neurodegenerative and Metabolic Diseases
ATF4 upregulation in response to endoplasmic reticulum (ER) stress contributes to neuronal apoptosis in Alzheimer's disease by inducing the transcription factor CHOP, which in turn activates pro-apoptotic genes such as PUMA, promoting cell death in affected neurons.[61] In prolonged ER stress conditions, ATF4 shifts neuronal responses from adaptive survival to a pro-apoptotic state, exacerbating neurodegeneration in Alzheimer's models.[62] Similarly, in Parkinson's disease, ATF4 promotes dopaminergic neuronal death through CHOP-dependent pathways, as it is required for the transcriptional induction of pro-apoptotic factors like CHOP, Trib3, and PUMA in response to neurotoxins such as MPP+ and pathogenic α-synuclein aggregates, although its role may be context-dependent with protective effects observed in some models.[63][64][65] This pro-apoptotic role of ATF4 is evident in cellular models where its activation enhances vulnerability to oxidative and proteotoxic stresses characteristic of Parkinson's pathology.[64]In metabolic diseases, ATF4 protects against β-cell loss in diabetes by regulating amino acid metabolism and maintaining proteostasis during endoplasmic reticulum (ER) stress, with ATF4 deficiency leading to exacerbated hyperglycemia, reduced insulin production, and increased β-cell apoptosis in mouse models.[66] In non-alcoholic fatty liver disease (NAFLD), ATF4 drives hepatic steatosis by regulating lipogenesis; its activation in response to nutrient excess, such as high-carbohydrate diets, upregulates lipid synthesis pathways, leading to fat accumulation in hepatocytes.[67] Conversely, ATF4 deficiency protects against diet-induced steatosis, highlighting its role in promoting lipid dysregulation in metabolic liver disorders.[68]ATF4 also facilitates skeletal muscle atrophy in cachexia through heterodimerization with C/EBPβ, a transcriptional regulator that enhances expression of atrophy-related genes, resulting in protein degradation and muscle wasting.[69] This ATF4-C/EBPβ complex is sufficient to induce atrophic effects in muscle fibers under catabolic conditions, independent of other stress pathways.[70]Studies using ATF4 knockout (ATF4-/-) mice demonstrate protection against neurodegeneration; for instance, these mice exhibit reduced neuronal apoptosis and preserved dopaminergic function in Parkinson's models exposed to neurotoxins, indicating ATF4's essential role in stress-induced cell death.[63] In glaucoma models, ATF4 deletion in retinalganglion cells rescues neurodegeneration by blocking the ER stress-mediated apoptotic axis, further supporting ATF4 inhibition as a neuroprotective strategy.[71]