Yilin Du_No.5

Q: How did the anthor address the question?

■Step1 Establishing MLLT1 and MLLT3 as essential AID cofactors Selective YEATS-domain inhibitors, CRISPR-Cas9 knockouts and conditional knockout mice were used to measure CSR, SHM, gene conversion, translocations and off-target mutagenesis across mouse, human and chicken B-cell systems. Combined MLLT1/MLLT3 loss phenocopied AID deficiency without materially changing cell growth, AID abundance or immunoglobulin germline transcription; the paralogues were redundant, with MLLT1 dominant in mouse B cells. ■Step2 Defining the chromatin state that specifies AID targets ChIP-seq mapped MLLT1, MLLT3, CDK9, AID and histone marks, while TT-seq and RNAPII ChIP-seq measured transcription. High MLLT1/MLLT3 occupancy precisely marked AID-mutated regions in mouse B cells and human lymphoma lines and correlated with patient mutation frequencies. Double knockout caused only modest transcriptional changes but abolished local AID enrichment despite preserved bulk chromatin association, supporting direct licensing rather than a transcriptional effect. ■Step3 Demonstrating condensate-based local licensing of AID Structure-function mapping, chromatin tethering, condensate assays, live-cell imaging and BioID showed that both readers recruit AID and that the MLLT1 intrinsically disordered region forms condensates that selectively concentrate AID; AID(R174E) failed to partition into them. AID-MLLT1 and, less efficiently, AID-MLLT3 fusions restored CSR and mutagenic activity in double-knockout cells, demonstrating sufficiency for the licensing step.

kawaokashinpei3

3 days ago

2 min read

Selected journal : Nature

Histone readers MLLT1 and MLLT3 concentrate AID to confer locus specificity

https://www.nature.com/articles/s41586-026-11087-1

What is the main question of the paper?

How do the histone readers MLLT1 and MLLT3 confer locus specificity by concentrating and licensing activation-induced cytidine deaminase (AID) at immunoglobulin and selected off-target regions?

How did the anthor address the question?

■Step1

Establishing MLLT1 and MLLT3 as essential AID cofactors

Selective YEATS-domain inhibitors, CRISPR-Cas9 knockouts and conditional knockout mice were used to measure CSR, SHM, gene conversion, translocations and off-target mutagenesis across mouse, human and chicken B-cell systems. Combined MLLT1/MLLT3 loss phenocopied AID deficiency without materially changing cell growth, AID abundance or immunoglobulin germline transcription; the paralogues were redundant, with MLLT1 dominant in mouse B cells.

■Step2

Defining the chromatin state that specifies AID targets

ChIP-seq mapped MLLT1, MLLT3, CDK9, AID and histone marks, while TT-seq and RNAPII ChIP-seq measured transcription. High MLLT1/MLLT3 occupancy precisely marked AID-mutated regions in mouse B cells and human lymphoma lines and correlated with patient mutation frequencies. Double knockout caused only modest transcriptional changes but abolished local AID enrichment despite preserved bulk chromatin association, supporting direct licensing rather than a transcriptional effect.

■Step3

Demonstrating condensate-based local licensing of AID

Structure-function mapping, chromatin tethering, condensate assays, live-cell imaging and BioID showed that both readers recruit AID and that the MLLT1 intrinsically disordered region forms condensates that selectively concentrate AID; AID(R174E) failed to partition into them. AID-MLLT1 and, less efficiently, AID-MLLT3 fusions restored CSR and mutagenic activity in double-knockout cells, demonstrating sufficiency for the licensing step.

What is the strength of the paper?

A major strength is the convergence of pharmacological, genetic, genomic, biochemical and imaging evidence across primary cells, multiple B-cell models and three vertebrate species. Nascent-transcription profiling, chromatin fractionation and DNA-break controls separate AID licensing from indirect transcriptional or repair effects. Structure-function mutants and AID-MLLT fusion rescues provide strong causal support for local concentration rather than mere correlation with active chromatin. Concordance with patient lymphoma mutation patterns strengthens the disease relevance.

Comment

What I found most interesting is that MLLT1 and MLLT3, known for regulating transcription, also determine where AID acts by concentrating it locally. Rather than stopping at correlation, the authors show that transcriptional changes cannot explain the effect and restore AID activity by directly linking it to MLLT1. This step-by-step approach makes the conclusion clear and convincing.

Comment by Tomoaki Shirakawa