Navigating the genetic map: the path towards Zincore

Cells use the same DNA but express very different genes depending on their cellular state and cell type. A key part of controlling which genes get activated involves proteins called transcription factors, which bind DNA, and coregulators, which are recruited by transcription factors to help switch genes on or off. To find novel protein complexes, in this thesis we analyzed a collection of 30 genome-wide genetic screens covering diverse cellular phenotypes simultaneously. The rationale was that genes affecting many different processes in a correlated way are likely part of the same regulatory complex. This approach identified two poorly characterized genes, QRICH1 and SEPHS1, whose loss had strikingly similar effects across many cellular readouts. Biochemical experiments confirmed that the two proteins physically interact and form a complex, which was named Zincore. Further characterization of this complex showed that Zincore functions as a transcriptional coregulator that activates the expression of a large set of genes. In contrast to most coregulators, which bind to a separate activation domain on a transcription factor, Zincore binds directly to the DNA-binding domain of zinc finger (ZNF) transcription factors. In addition to binding to the ZNF proteins, Zincore physically stabilizes their contact with the DNA and in the absence of Zincore, ZNF proteins like ZFP91 bind their target sites much less efficiently. A cryo-EM structure confirmed this "locking" mechanism in atomic detail, and showed that a specific arginine clamp in SEPHS1 grips the conserved zinc finger fold. Notably, this fold is shared across the ~800 ZNF proteins encoded in the human genome, raising the possibility that Zincore acts as a broadly dedicated coregulator for this entire protein family. In this thesis we also describe that Zincore activity is downregulated in cancer and is potentially regulated at the level of nuclear localization, suggesting it is connected to signaling pathways that control differentiation. Furthermore, heterozygous mutations in QRICH1 or SEPHS1 in human patients have been linked to neurodevelopmental syndromes, and QRICH1 or SEPHS1 loss in mice causes embryonic lethality, indicating this complex is important during development.