Connecting the Drops: Computational Dissection of Intrinsic Disorder and Phase-Separation Propensity in the U2OS Tunneling Nanotube Proteome

Abstract Tunneling nanotubes (TNTs) are actin-supported open membrane conduits that connect remote cells and carry vesicles, organelles, RNAs, and protein aggregates. A recent proteomic study defined the TNT proteome (TNTome) of U2OS human osteosarcoma cells as a set of 1,177 proteins and identified the tetraspanins CD9 and CD81 as non-redundant TNT regulators. Here we performed a proteome-wide computational analysis of the intrinsic disorder and liquid–liquid phase-separation (LLPS) propensity of the TNTome using RIDAO, CH-CDF analysis, D 2 P 2 , FuzDrop, the ELM resource, STRING, AlphaFold, and AFflecto. Of 1,176 proteins with complete predictions, 255 (21.7%) were highly disordered and 443 (37.7%) mostly ordered, with a mean disorder of 27.6 ± 27.2%. Among 253 highly disordered proteins scored by FuzDrop, 219 (86.6%) were predicted LLPS drivers; these drivers formed a dense STRING network (209 proteins, 2,200 edges, 21 hubs led by HNRNPA1, HNRNPC, HNRNPA2B1, SRSF1, NPM1, and FUS) that was strongly enriched in RNA-binding, mRNA-processing and ribonucleoprotein-complex functions. A bilateral deep dive into the 15 most disordered and 15 most ordered proteins, together with a catalogue of 1,647 short linear motifs, suggested a two-tier architecture, in which disordered, phase-separation-competent scaffolds coexist with ordered structural and enzymatic modules, including CD9 and CD81. We emphasize that all disorder predispositions and LLPS assignments are predictive and that, because the TNTome is cancer-derived and no non-cancer TNT proteome exists, the disorder/LLPS enrichment cannot yet be ascribed to TNT biology versus the malignant state. These intrinsically disordered proteins (IDPs) and intrinsically disordered regions (IDRs) and their predicted condensate behavior represent testable hypotheses for the molecular organization of TNTs.

Authors

Institutions

Publication Details

Journal
Research Square
Published
2026-11-16
DOI
https://doi.org/10.21203/rs.3.rs-11420044/v1
Primary Topic
Protein Structure and Dynamics
Type
preprint
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
preprint

Connecting the Drops: Computational Dissection of Intrinsic Disorder and Phase-Separation Propensity in the U2OS Tunneling Nanotube Proteome

Mohamed A. Khalaf, Vladimir Nikolaevich Uversky, Arjan van der Vaart
1 links
Research Square
Protein Structure and Dynamics
preprint

Connecting the Drops: Computational Dissection of Intrinsic Disorder and Phase-Separation Propensity in the U2OS Tunneling Nanotube Proteome

Mohamed A. Khalaf, Vladimir Nikolaevich Uversky, Arjan van der Vaart
preprint en

Abstract

Abstract Tunneling nanotubes (TNTs) are actin-supported open membrane conduits that connect remote cells and carry vesicles, organelles, RNAs, and protein aggregates. A recent proteomic study defined the TNT proteome (TNTome) of U2OS human osteosarcoma cells as a set of 1,177 proteins and identified the tetraspanins CD9 and CD81 as non-redundant TNT regulators. Here we performed a proteome-wide computational analysis of the intrinsic disorder and liquid–liquid phase-separation (LLPS) propensity of the TNTome using RIDAO, CH-CDF analysis, D 2 P 2 , FuzDrop, the ELM resource, STRING, AlphaFold, and AFflecto. Of 1,176 proteins with complete predictions, 255 (21.7%) were highly disordered and 443 (37.7%) mostly ordered, with a mean disorder of 27.6 ± 27.2%. Among 253 highly disordered proteins scored by FuzDrop, 219 (86.6%) were predicted LLPS drivers; these drivers formed a dense STRING network (209 proteins, 2,200 edges, 21 hubs led by HNRNPA1, HNRNPC, HNRNPA2B1, SRSF1, NPM1, and FUS) that was strongly enriched in RNA-binding, mRNA-processing and ribonucleoprotein-complex functions. A bilateral deep dive into the 15 most disordered and 15 most ordered proteins, together with a catalogue of 1,647 short linear motifs, suggested a two-tier architecture, in which disordered, phase-separation-competent scaffolds coexist with ordered structural and enzymatic modules, including CD9 and CD81. We emphasize that all disorder predispositions and LLPS assignments are predictive and that, because the TNTome is cancer-derived and no non-cancer TNT proteome exists, the disorder/LLPS enrichment cannot yet be ascribed to TNT biology versus the malignant state. These intrinsically disordered proteins (IDPs) and intrinsically disordered regions (IDRs) and their predicted condensate behavior represent testable hypotheses for the molecular organization of TNTs.

Research Square
University of South Florida (US)
Good health and well-being
Protein Structure and Dynamics
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon