Search bioRxiv⌕ Search

bioRxiv · 10.1101/2025.09.24.678339

Genetic landscape of DNMT3A R882H reveals mechanism of aberrant oligomerization

Abstract

DNA methyltransferase 3A (DNMT3A) is a de novo DNA methyltransferase that is recurrently mutated in hematological malignancies and developmental disorders. The most prevalent mutation, R882H, compromises DNMT3A activity in a dominant-negative manner, but its precise biochemical mechanism has been debated. Here, we use paired deep mutational scanning of the wild-type and R882H-mutant proteins to systematically identify mutations on a massively parallel scale that modify DNMT3A activity by suppressing, phenocopying, or selectively rescuing the dominant-negative effect of R882H. By leveraging the mutational depth and unbiased nature of the paired genetic landscapes, we uncover two distinct mechanisms that can rescue DNMT3AR882H activity, providing novel insights into the function of the R882 hotspot. First, by analyzing the effects of combinatorial mutations in the target recognition domain (TRD), we reveal that its crosstalk with the ADD regulatory domain modulates DNMT3A DNA binding and enzymatic activity, partially compensating for R882H-induced loss-of-function. Second, pairwise analysis of variant effects across the two genetic backgrounds supports the notion that R882H promotes aberrant macro-oligomerization of DNMT3A via its central dimerization interface, which accounts for its dominant-negative effect. Critically, we show that the R882 position exhibits a distinct dominant-negative signature in the genetic landscape, where positively charged residues at this position safeguard against aberrant macro-oligomerization. By performing hydrogen-deuterium exchange mass spectrometry (HDX-MS) on DNMT3A mutants, we show that R882H dramatically alters the protein dynamics of DNMT3A, rigidifying its central dimerization interface to promote oligomerization. Our data support a new model in which R882H removes the critical function of R882, where the arginine attenuates the pre-organization of the interface and subsequent oligomerization at a supramolecular assembly hotspot1. Altogether, we map the genetic landscape underpinning the DNMT3AR882H hotspot mutation, illuminating the unexpected molecular mechanism of macro-oligomerization that drives its dominant-negative effect.

Explore related subjects

Keep this discovery

Explore connections, maps & timelines

BibTeXRIS

Garcia, E. M., Lavidor, O. R., Raval, S., Lue, N. Z., Liang, J. K., McFadden, M., Hwang, D. D., Zepeda, M. U., Chen, L. W., Carr, S. A., Papanastasiou, M., Liau, B. B.. 2025-09-25. Genetic landscape of DNMT3A R882H reveals mechanism of aberrant oligomerization. https://doi.org/10.1101/2025.09.24.678339

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related preprints

MgATP/MgADP-dependent conformational dynamics and intrinsically disordered regions of vascular KATP channels revealed by cryoEM

Vascular smooth muscle KATP channels, composed of the pore-forming Kir6.1 and regulatory SUR2B subunits, control vascular tone, dysfunction of which causes systemic disease. Vascular KATP is regulated by Mg-nucleotides, but the underlying structural mechanism has remained elusive. Here, we determined cryoEM structures of these channels in the presence of MgATP and MgADP. Two key structures captured, one showing the SUR2B-nucleotide binding domains (NBDs) separated and one showing the SUR2B-NBDs dimerized, reveal conformation-specific organization of intrinsically disordered regions (IDRs) found in both Kir6.1 and SUR2B. In the NBD-separated conformation, the Kir6.1-N terminal IDR (KNt) sits within the central cleft of the ABC-core of SUR2B. In the NBD-dimerized conformation, KNt is excluded from the central cleft and instead forms contacts with an ED domain comprising 15 consecutive glutamate and aspartate residues within a SUR2B IDR, the N1-T2 linker connecting NBD1 (N1) to transmembrane domain 2 (T2). Moreover, within the N1-T2 linker a regulatory helix seen between the two NBDs in the NBD-separated conformation moves to outside the dimerized NBDs, interacting with the C-terminal residues unique to SUR2B, in the NBD-dimerized conformation. MD simulations further reveal that transient but frequent interactions mediated by the IDRs may facilitate Mg-nucleotide dependent conformational switch in vascular KATP channels.

biochemistry↗

Probing the sequence variability tolerance in a de novo α-helical barrel biocatalyst

De novo-designed enzymes have recently achieved high catalytic activity and stereoselectivity while demonstrating exceptional thermostability in entirely novel protein scaffolds. Among these, -helical barrel protein scaffolds are attractive structures for biocatalysis due to their structural simplicity, high thermostability, and rationalizable sequence patterning. However, enabling major structural reengineering of these scaffolds while maintaining the structure, stability and catalytic activity while also improving soluble protein production remain major challenges and pose the fundamental question how engineerable a de novo backbone-sequence pair is. Here, we combine deep learning based and classic computational protein design to modify and optimize de novo -helical barrel biocatalysts. Using the previously reported six-helical barrel 6H5L as a model scaffold, AlphaFold2-guided RosettaRemodel enabled the design of a truncated variant, whose crystal structure closely matches the design model. Additional sequence-redesign using ProteinMPNN generated a variant with a tenfold increase of soluble protein yield in Escherichia coli. Biochemical, biophysical, and structural analyses showed that both variants retained the overall barrel architecture, high thermal stability, and catalytic activity for both purified protein and whole-cell systems. Detailed kinetic analysis on the variants showed both variation in kcat and Km, reflecting changes in catalytic turnover and substrate binding. Together, these approaches provide new insights and possibilities for the further engineering of functional de novo -helical barrels, their ability to withstand dramatically large sequence changes and their broader application in biocatalysis and biotechnology.

biochemistry↗

Cytokine-induced nuclear translocation of STAT1 via a non-transferable NLS

The targeting function of nuclear localization signals (NLSs) is generally considered independent of a protein's native sequence or fold and is readily transferable to heterologous cargos. Contrary to this paradigm, rapid nuclear translocation of phosphorylated STAT1 (pSTAT1) following cytokine stimulation requires importin {beta}, Ran-GTP, and the importin 5 isoform, which recognizes a non-transferable NLS. Here, we present cryo-EM structures of pSTAT1 bound to importin 5, revealing an asymmetric 2:1 complex that diverges from canonical NLS-mediated cargo recognition. Importin 5 occupies the DNA-binding groove of the pSTAT1 dimer, with a single STAT1 N-terminal domain positioning the C-terminal Armadillo repeats 9-10 (S1B domain) orthogonal to the DNA-binding interface. This interface is also targeted by the Ebola virus protein VP24, an antagonist of interferon signaling. We further show that Ran-GTP alone is insufficient to trigger nuclear release of pSTAT1, which additionally requires the exportin CAS. A cryo-EM reconstruction of the CAS-Ran-GTP-5 complex, supported by in vitro competition assays, demonstrates that CAS and pSTAT1 are mutually exclusive ligands for importin 5. Together, these findings define the molecular choreography of cytokine-induced STAT1 nuclear translocation and release, establishing a general paradigm for STAT family signaling.

biochemistry↗