Scripps Research Logo

Discovering small molecule regulators of the proteostasis network

 

In order for cells to produce proteins, the cells have to maintain a balance between the competing and integrated biological pathways within cells that control biogenesis, translocation across membranes, folding vs. degradation, trafficking, etc. within and outside the cell. This balancing act is known as protein homeostasis, or proteostasis. Deficiencies in proteostasis lead to many metabolic, oncological, neurodegenerative, and cardiovascular diseases. Since the demand on cells to produce proteins changes with development, aging, and environmental stresses, mammals have evolved stress-responsive signaling pathways that adjusts the cells’ capacity to produce proteins to meet demand. We hypothesized that adapting proteostasis by using small molecule proteostasis regulators that activate a stress-responsive signaling pathway has significant promise to ameliorate the wide array of diseases caused by deficiencies in proteostasis.

A collaboration between the Wiseman and Kelly labs focuses on the Unfolded Protein Response (UPR) stress-responsive signaling pathway that modulates production of membrane proteins, secreted proteins, as well as proteins directed to certain organelles like the lysosome. The UPR comprises 3 signaling arms: the IRE1/XBP1s, PERK, and ATF6 arms that each end in the generation of a transcription factor. Activation of the IRE1/XBP1s and/or ATF6 arms leads to a transcriptional program resulting in the upregulation of chaperones and other proteostasis network components that can lead to enhanced folding or degradation of mutant proteins, without affecting the vast majority of the wild type proteome. Activation of the PERK arm results in attenuation of translation and, if activation is prolonged, induction of apoptosis. Thus we avoid activating this arm for therapeutic purposes.

In collaboration with the Wiseman Lab, we developed the HEK293DAX cell line, a stably transfected HEK293 cell line that can be induced to express XBP1s and/or ATF6, either separately or in combination. Tetracycline-inducible XBP1s can be positively regulated by the addition of doxycycline. ATF6 expression is positively regulated by the addition of trimethoprim, which stabilizes the destabilized DHFR domain that is fused to the ATF6 transcription factor. Using this chemical genetic activation of select arms of the UPR, we characterized 3 distinct proteostasis environments in the endoplasmic reticulum that are accessible by activating XBP1s and/or ATF6. Importantly, we showed that arm-selective UPR activation selectively reduces secretion of a disease-associated variant of transthyretin without affecting secretion of the wild-type protein. Strictly analogous effects were observed with amyloidogenic light chains, upon arm-selective UPR activation; the light chains were selectively degraded and not secreted.

Few compounds exist to selectively activate these UPR signaling pathways. Thus, we performed 2 separate high-throughput cell-based transcriptional reporter screens employing 650,000 candidate small molecules. Whole cell transcriptional and proteomic profiling was performed after appropriate counterscreens to establish first-in-class small molecule proteostasis regulators that selectively activate the IRE1/XBP1s or ATF6 signaling arms of the UPR.

We identified 79 compounds in 8 structural classes that preferentially activate the ATF6 UPR signaling arm. We are now using chemical proteomic and targeted RNAseq transcriptional approaches to define the mechanism of action for two of these structural classes found to function through covalent and non-covalent mechanisms, respectively. In addition, we are using medicinal chemistry in combination with biochemical and a variety of transcriptional profiling and proteomic approaches to establish next generation molecules with improved potency and selectivity for ATF6 activation in cells. We will scrutinize the hypothesis that highly prioritized compounds induce protective, ATF6-mediated remodeling of secretory proteostasis in young and old mice, in murine disease models, and in cells and cell-based pathology models to ameliorate a spectrum of proteinopathies.

Similarly, we identified 444 compounds in 10 structural categories that activate the IRE1/XBP1s arm of the UPR. We are investigating their mechanism of action, improving their selectivity and potency through medicinal chemistry efforts, and we are exploring the translational potential of highly prioritized compounds by scrutinizing compound-dependent IRE1/XBP1s activation in young and old mice and in disease-relevant cell culture and murine models.