Biology subjects

Udupa, A.

Publications and source records attributed to Udupa, A..

3 recordsLinked to original sources

bioRxiv2026

Machine learning guided design of strong transcriptional activation domains

Artificial intelligence has accelerated the design of structured proteins, but it remains difficult to design functional intrinsically disordered proteins because of weak structural constraints and poorly defined sequence-to-function relationships1. Activation domains (ADs), transcription factor regions that recruit coactivators, are an excellent test case for the de novo design of intrinsically disordered sequences. ADs are one of the oldest groups of functional intrinsically disordered sequences2 and designed ADs have applications in synthetic biology, biotechnology, and agriculture. We computationally generated 500,000 candidate sequences that resembled known ADs and predicted AD function with extant machine learning models, including three neural networks. From high-scoring sequences, we selected 18,750 for experimental validation in yeast. Surprisingly, >90% of the 16,770 well-measured sequences function as ADs. Sequences predicted to be active by multiple models were stronger, motivating development of a fast and accurate consensus AD predictor, MADRIGAL. A number of our designed ADs retain functionality across fungi, plants, and human cells. This work illustrates a generalizable paradigm for incorporating sequence-to-function models into the design loop to efficiently generate functional intrinsically disordered sequences.

synthetic biology↗

bioRxiv2024

HP1 binding creates a local barrier against transcription activation and persists during chromatin decondensation

Mouse pericentric repeats form transcriptionally silent and compacted domains known as chromocenters. These prototypic heterochromatin compartments are marked by heterochromatin protein 1 (HP1). However, its contributions to chromocenter structure and function remain debated. We investigated the role of HP1 by recruiting the activators VP16, p65, and VPR to mouse fibroblast chromocenters and analyzed its silencing activity with a transcription reporter. Upon chromocenter decondensation and transcription activation, interactions of HP1 with chromatin and H3K9 trimethylation remained stable, suggesting stoichiometric binding rather than higher-order assembly. HP1-mediated repression required promoter-proximal binding and effectively suppressed VP16-triggered activation but not the stronger activation by VPR. These observations are explained by a 1D lattice binding model, which conceptualizes chromocenters as arrays of repeat units that can independently switch between silenced and activated states. Our findings provide a quantitative framework that explains how chromocenter organization responds to transcriptional activation while maintaining local heterochromatin features. GRAPHICAL ABSTRACT O_FIG O_LINKSMALLFIG WIDTH=189 HEIGHT=200 SRC="FIGDIR/small/627308v1_ufig1.gif" ALT="Figure 1"> View larger version (33K): org.highwire.dtl.DTLVardef@83816corg.highwire.dtl.DTLVardef@1a603c6org.highwire.dtl.DTLVardef@41d8b4org.highwire.dtl.DTLVardef@fa4d0d_HPS_FORMAT_FIGEXP M_FIG C_FIG HIGHLIGHTSO_LIHP1 represses transcription at mouse chromocenters and an ectopic reporter when competing with transcriptional activators VP16, p65 and VPR C_LIO_LIHP1 repression requires promoter-proximal binding and effectively counteracts the weak activator VP16 C_LIO_LIThe strong activator VPR overcomes HP1-mediated repression while HP1 remains bound to chromatin C_LIO_LIHP1 binding and H3K9me3 persist during chromocenter decondensation and transcriptional activation C_LIO_LIA 1D lattice binding model explains how independent repeat units transition between silenced and activated states without requiring phase separation C_LI

cell biology↗

bioRxiv2024

Nucleoporin Nup153 docks the splicing machinery to the nuclear pore for efficient mRNA processing

The nuclear pore complex (NPC), composed of proteins termed nucleoporins (Nups), intercalates the nuclear envelope, and is primarily involved in protein trafficking and mRNA export. At the nuclear basket, Nups have been associated with chromatin organization and postulated to function as transcriptional hubs, working in tandem with mRNA export machinery. However, little is known about the intermediate process of RNA splicing at the NPC. Here, we used BioID to screen for interactors of basket-Nups Nup153 and TPR and discovered the enrichment of splicing proteins across all spliceosome complexes (E, A, B, B*, P). The peripheral nature of the interaction between Nup153 and selected splicing components was confirmed by in-situ proximity ligation assay and STED microscopy. The presence of splicing components at the NPC, reduced upon splicing inhibition, is partly dependent on Nup153 and functionally correlated to the splicing of Nup153-bound genes. Assessed by DamID, Nup153-bound genes ([~]500) are characterized by multiple long introns with lower-than-average GC content. Positioned at the periphery but distinct from the neighbouring lamina-associated domain (LADs) in chromatin signatures and expression levels, these genes showed Nup153-dependent splicing defect, suggesting that splicing occurs at the NPC. Altogether, our data substantiates the gene gating theory bringing transcription and export, now accompanied by speckle-distant splicing events, at the level of the NPC.

cell biology↗