SET1/MLL complexes control transcription independently of H3K4me3

HYA Au; AT Szczurek; I de Krijger; N Kjelstrup-Osorio; AL Hughes; A Lastuvkova; M Vermeulen; RJ Klose

doi:10.64898/2025.12.08.692985

Reproducible

QED Impressions

This paper passed independent validation.

QED's automated review checked the evidence supporting each claim, the statistical methods, and whether the results are reproducible from the data provided.

It ranks in the top 1% of preprints assessed in Genomics this quarter.

Run full QED analysis

Read paper

Why this is interesting

Overturning the assumption that H3K4me3 drives gene transcription, this work shows SET1/MLL complexes control transcription burst size through physical occupancy—counteracting promoter-proximal termination—fundamentally reframing how these frequently mutated regulators operate and how to therapeutically target them.

Author perspective

What is the most important insight, discovery, or finding from your paper that you would like other scientists to know?

“

Our most important discovery is that Lysine-4 methyltransferases are primarily regulating gene expression independently of their histone modifying activity.

”

HYA Au

RJ Klose

QED claims analysis

Summary

SET1A/B and MLL1/2 complexes synergistically deposit H3K4me3 at gene promoters through a newly discovered SET1B-C complex, but control transcription primarily through H3K4me3-independent mechanisms by counteracting Restrictor-mediated premature termination to maintain transcriptional burst size.

See full analysis

Main Claim 1 –

SET1A/B and MLL1/2 Synergize to Deposit Promoter H3K4me3

Supporting evidence

SET1A/B Loss Modestly Reduces Downstream TSS H3K4me3
MLL1/2 Depletion Broadly Reduces H3K4me2/3 at TSSs
MLL3/4 Modestly Affect TSS H3K4me3 but Target Enhancers
H3K4me3 Partially Recovers After Depleting All Six Enzymes

Main Claim 2 –

C-Terminal SET1B Isoform Complex Contributes to H3K4me3

Supporting evidence

WRAD/CFP1 Complexes Persist After SET1/MLL Depletion
Internal SET1B Promoter Produces C-Terminal SET1B Isoform
SET1B-C Depletion Enhances H3K4me3 Loss at TSSs
Seven-Complex Depletion Phenocopies RBBP5 Loss of H3K4me3

Main Claim 3 –

SET1/MLL Complexes Regulate Transcription Independently of H3K4me3

Supporting evidence

KDM5 Inhibition Confirms Transcription Changes Are H3K4me3-Independent
SET1A/B/B-C Drive Transcriptional Control, Unlike MLL1/2
SET1A/B/B-C Drive Enhancer Transcription Independent of H3K4me3

Main Claim 4 –

SET1/MLL Sustain Burst Size by Preventing Promoter-Proximal Termination

Supporting evidence

MS2 Live Imaging Quantifies Real-Time Transcription Burst Parameters
Restrictor Loss Increases ON-Period Amplitude in SET1-Dependent Genes
Restrictor Co-depletion Rescues SET1/MLL Burst Defects

Abstract

Histone H3 lysine 4 trimethylation (H3K4me3) at gene promoters is thought to play a central role in gene transcription. H3K4 methylation is deposited by the SET1 (A/B) and MLL (1-4) multi-protein complexes, but discovering how these essential enzymes shape H3K4me3 has been extremely challenging due to their multiplicity. This has also made determining whether SET1/MLL complexes control transcription through H3K4me3, or non-catalytic activities, an impenetrable problem. Here, we overcome these challenges through leveraging genome-engineering and combinatorial SET1/MLL protein depletion, integrated with genomics, proteomics, and live-cell transcription imaging. We uncover a new SET1B complex and reveal that SET1 and MLL1/2 complexes synergise to define H3K4me3 at gene regulatory elements. Unexpectedly, by decoupling SET1/MLL complex occupancy at promoters from H3K4me3, we discover they primarily control transcription independently of H3K4me3 through counteracting promoter-proximal termination and supporting transcription burst size. These discoveries reveal a new H3K4me3-independent logic for SET1/MLL-dependent control of gene transcription.

How the QED Score works

QED Score is a validated AI-based quality metric that evaluates life science papers on two dimensions: originality and validity. Each manuscript is anonymized, broken down into its claims and supporting evidence, and examined by specialized AI agents for statistical rigor, consistency across figures, and fit with the existing literature. Scores are percentile ranks, so they're comparable across fields. In blinded tests, expert professors sided with QED Score over journal rank in 75% of decisive cases.

Read the methodology

Physiology

Animal Behavior and Cognition

Developmental Biology

Microbiology