CREB3L2-ATF4 Heterodimerization Defines a Transcriptional hub of Alzheimer's Disease Gene Expression Linked to Neuropathology

Cláudio Gouveia Roque, PhD		Ulrich Hengst, PhD

Ever since the emergence of genome-wide profiling methodologies, gene expression changes have been one of the most researched molecular features of Alzheimer’s disease. These system-level studies are highly informative and can be used to inform unbiasedly on specific aspects of disease pathogenesis. A potential drawback is the complexity of these readouts, as it is not uncommon for thousands of genes to be highlighted in comparisons between affected and healthy individuals. An open question in the field has been whether these changes are causative agents of disease progression or a mere bystander response to the pathophysiological process accompanying Alzheimer’s. By initially focusing on a single transcriptional pathway regulated by β-amyloid, an early trigger of Alzheimer’s decline, our latest work provides strong evidence for the importance of gene expression alterations in the pathological cascade of this neurodegenerative disorder.

Figure 6. CREB3L2-ATF4 interacts with β-amyloid and tau neuropathologies. (G) The findings support a model whereby CREB3L2-ATF4 promotes AD-linked gene expression changes and contributes to characteristic features of AD pathophysiology, including retromer dysfunction, altered β-amyloid metabolism, and tau hyperphosphorylation.

The findings, published recently in Science Advances, were driven by a well-known but typically disregarded aspect of transcription factor biology: their dependency on dimerization for binding to DNA. Building on previous research by our group, we first identified CREB3L2 as a new binding partner of ATF4, a central effector of cellular responses to stress. Interestingly, while CREB3L2-ATF4 dimers can be found in neurons under normal conditions, the interaction between these two transcription factors is substantially potentiated in neurons challenged by β-amyloid. Subsequent efforts focused on elucidating the transcriptional response downstream of CREB3L2-ATF4, which included the development of a novel, broadly accessible genomic technique for studying transcription factor dimers.

Crossing this dataset with brain gene expression profiles of Alzheimer’s disease sufferers led to the identification of several relevant pathways linked to pathogenesis, including retromer dysfunction and tau regulation. Further experiments demonstrated that the transcriptional response activated by CREB3L2-ATF4 is sufficient to trigger tau hyperphosphorylation in neurons. In other words, the work now online provides direct support for a transcriptional link between the two hallmark lesions associated with Alzheimer’s disease, β-amyloid and tau. Finally, by characterizing CREB3L2-ATF4 in detail, our lab hopes to use this knowledge to guide the identification of a promising candidate drug, dovitinib, which can normalize gene expression downstream of β-amyloid. Unlike available anti-amyloid therapies, dovitinib is predicted to act downstream of β-amyloid and may one day be used as a complementary approach to mitigate the harmful effects mediated by this disease insult.

Cláudio Gouveia Roque, PhD
Associate Research Scientis in the Taub Institute
cag2230@cumc.columbia.edu

Ulrich Hengst, PhD
Associate Professor of Pathology and Cell Biology & Taub Insitute
uh2112@cumc.columbia.edu