A long non-coding RNA binds DOT1L's catalytic domain and turns its activity up - and blocking it makes the fusion protein itself get eaten
also called LAMP5-AS1 · LAMP5 antisense 1 · lncRNA-DOT1L-LAMP5 axis
A long RNA molecule called LAMP5-AS1 sticks directly to DOT1L's business end and makes it work harder, which raises the H3K79 tag and keeps leukaemia stem-cell genes on. Patients with more of this RNA do worse, and removing it makes the leukaemia's own fusion protein get broken down by the cell's recycling system.
The same thing, for a scientist
LAMP5-AS1 is a long non-coding RNA that is **specifically overexpressed in MLL-rearranged leukaemia** (58 MLL versus 163 MLL-wild-type patients, p<0.001) and whose higher expression predicts **reduced 5-year leukaemia-free survival** (p<0.01) [E1]. Mechanistically it **binds the lysine-rich region of the DOT1L catalytic domain directly** - shown by RNA EMSA, histone methyltransferase assay, RNA pull-down and FISH - and facilitates DOT1L's methyltransferase activity, raising **global H3K79me2 and me3** and the HOXA-cluster self-renewal programme; its suppression reduced colony formation and increased differentiation of primary MLL leukaemia CD34+ cells. A follow-up from the same group defines a second arm: LAMP5-AS1 recruits DOT1L to the *LAMP5* locus to activate *LAMP5*, acts as an **autophagic suppressor**, and blockade of the axis **enhances autophagy and promotes degradation of MLL fusion proteins**, with knockdown inducing apoptosis in lines and primary cells and extending mouse survival [E2]. **Two things make this worth an entry.** It is a **direct activator of DOT1L catalysis** and therefore a target that lowers the mark without touching the enzyme's scaffolding role - a modality that sits between catalytic inhibition and degradation. And it is the only route in this beat that degrades the **fusion protein** rather than one of its cofactors.
A third way to lower the mark
This beat's modality table has had two entries: block the active site (drug-pinometostat) or remove the protein (drug-dot1l808). LAMP5-AS1 adds a third: remove an activator of catalysis.
It has a theoretical advantage over both. Knocking down LAMP5-AS1 lowers DOT1L's activity without removing DOT1L, so the scaffolding functions that adult marrow appears to need are left in place, and the RNA is leukaemia-restricted in expression rather than ubiquitous - which is the source of selectivity that neither a catalytic inhibitor nor a degrader has.
The practical objections are large and should be stated. Antisense oligonucleotide or siRNA delivery to leukaemic blasts and particularly to quiescent leukaemia-initiating cells is an unsolved problem; the effect size of an activator knockdown is bounded by how much of DOT1L's activity the activator contributes; and this is one laboratory's finding across both papers 12, with no independent replication found in this beat's reading.
The degradation arm, and why it is a different claim
The 2024 paper's headline is not about H3K79 methylation at all: blockade of the LAMP5-AS1-LAMP5 axis enhances autophagy and represses MLL fusion proteins by promoting their degradation 2.
If that holds, it is the most direct anti-fusion mechanism in this beat - every other approach here targets a cofactor and leaves the oncoprotein in place. It would also be orthogonal to menin inhibition, which displaces the fusion from chromatin but does not remove it from the cell.
The caution is that the causal chain is long: LAMP5-AS1 → DOT1L recruitment to the LAMP5 locus → LAMP5 expression → autophagic suppression → MLL fusion protein stability. Each step is separately supported in the abstract and the chain as a whole is a single group's model. A graph entry should record the chain and its length rather than the headline.
| mechanism class | non-coding RNA regulation of enzyme activity |
|---|---|
| rna | LAMP5-AS1 (LAMP5 antisense 1) |
| target protein | gene-dot1l |
| binding site | lysine-rich region of the DOT1L catalytic domain |
| effect on mark | raises global H3K79me2 and H3K79me3 |
| downstream programme | HOXA-cluster self-renewal genes; LAMP5 expression |
| second arm | autophagic suppression protecting MLL fusion proteins from degradation; blockade enhances autophagy and degrades the fusion |
| modality class | remove an activator of catalysis - between catalytic inhibition and degradation |
| replication status | both papers from one laboratory; no independent replication found in this beat's reading |
clinical association
{
"cohort_mll": 58,
"cohort_mll_wt": 163,
"overexpression_p": "<0.001",
"leukaemia_free_survival_p": "<0.01",
"adjustment": "none stated"
}methods used
[ "RNA EMSA", "histone methyltransferase assay", "RNA pull-down", "RNA FISH", "ChIRP", "ChIP", "LC3B puncta / TEM / mRFP-GFP-LC3 flux" ]
- Lowering DOT1L's catalytic output is lowering the function adult marrow requires, so the modality's safety rests entirely on the RNA being absent from normal progenitors rather than on sparing any DOT1L function.
- Enhancing autophagy to degrade the fusion oncoprotein cuts both ways: autophagy is a documented survival pathway for leukaemia cells under therapeutic stress, so the same manipulation could protect cells in combination settings.
- Delivering an oligonucleotide to quiescent leukaemia-initiating cells in marrow is an unsolved problem, and those are the cells that matter for relapse.
- Has any independent group replicated the LAMP5-AS1-DOT1L interaction or the survival association? Both papers come from one laboratory and the clinical claim rests on a single unadjusted 58-patient cohort.
- How restricted is LAMP5-AS1 expression really - is it absent from normal haematopoietic progenitors, or merely lower? The entire selectivity argument for this modality depends on the answer.
- Is LAMP5-AS1 expression associated with fusion partner? If it tracks AF9/ENL-partnered disease and not KMT2A::AFF1, the prognostic association may be a partner effect.
- E1moderate confidence 3 of 5: One strong study, or several smaller consistent ones.paper
The lncRNA LAMP5-AS1 is specifically overexpressed in MLL-rearranged leukaemia patients (n=58) compared with MLL-wild-type leukaemia (n=163, p<0.001), and higher expression is associated with reduced 5-year leukaemia-free survival (p<0.01); LAMP5-AS1 facilitates DOT1L's methyltransferase activity by directly binding the lysine-rich region of its catalytic domain, promoting global H3K79 di- and trimethylation and upregulating HOXA-cluster self-renewal genes; LAMP5-AS1 suppression significantly reduced colony formation and increased differentiation of primary MLL leukaemia CD34+ cells.
The lncRNA LAMP5-AS1 drives leukemia cell stemness by directly modulating DOT1L methyltransferase activity in MLL leukemia. · Wang WT, Chen TQ, Zeng ZC, et al. · J Hematol Oncol 2020What this evidence cannot show
Structured abstract read from the eutils record; full text not read. The survival association comes from a retrospective cohort of 58 MLL patients with no stated adjustment for fusion partner, age, lineage or treatment era, so it is a univariate prognostic association rather than an independent prognostic factor. RNA-protein binding is shown by EMSA and pull-down, which establish association more securely than direct high-affinity binding, and no affinity is reported. Single laboratory.
Evidence that argues against this (1)
- A lncRNA that raises GLOBAL H3K79me2/3 is acting on the endogenous DOT1L activity as much as on the fusion-directed one, which is the unit normal haematopoiesis uses. The selectivity argument for targeting it rests entirely on the RNA's restricted expression, not on any locus selectivity of its effect. Gilan O et al. 2023
- E2moderate confidence 3 of 5: One strong study, or several smaller consistent ones.paper
LAMP5-AS1 recruits DOT1L to the LAMP5 locus to directly activate LAMP5 expression and acts as a novel autophagic suppressor safeguarding MLL fusion proteins from autophagic degradation; knocking down LAMP5-AS1 induces apoptosis in MLL leukaemia cell lines and primary cells and extends mouse survival in vivo, and blockade of the LAMP5-AS1-LAMP5 axis represses MLL fusion proteins by enhancing their degradation.
Blockade of the lncRNA-DOT1L-LAMP5 axis enhances autophagy and promotes degradation of MLL fusion proteins. · Chen TQ, Huang HJ, Zhu SX, et al. · Exp Hematol Oncol 2024What this evidence cannot show
Structured abstract read from the eutils record; full text not read. Same laboratory as E1, so the two papers are not independent. The causal chain has four steps - LAMP5-AS1 to DOT1L recruitment to LAMP5 expression to autophagic suppression to fusion-protein stability - each supported separately and none demonstrated end to end by a single manipulation. Autophagy readouts (LC3B puncta, TEM autolysosomes, mRFP-GFP-LC3 flux) are standard but indirect. The MLL fusions used are not named in the abstract. **Resolved:** the H3K79 arm is **8 MOLM-13 samples, one replicate per condition**. MOLM-13 carries KMT2A::MLLT3, not the index fusion, and single-replicate ChIP arms cannot support a quantitative claim. GSE150483. Resolved 2026-09-19 by the Q18b unstated-systems sweep from the study's deposition, not from its text; bounded as a statement about the deposited data rather than about the whole study.
Evidence that argues against this (2)
- Enhancing autophagy to degrade a fusion oncoprotein is a mechanism with an obvious double edge: autophagy is also a survival pathway under therapeutic stress in acute leukaemia, and the same manipulation could protect cells in a different context. The direction of the net effect is likely to be context-dependent rather than a property of the axis. Wang WT et al. 2020
- One replicate per condition in a ChIP experiment gives no estimate of variance, so any difference reported between conditions is unquantified. Combined with the fusion being MLLT3 rather than AFF1, this supports the existence of an effect and not its size. NCBI Gene Expression Omnibus 2026
- E3strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper
Adult mouse haematopoiesis requires DOT1L's catalytic activity - an enzymatically inactive DOT1L mutant fails to rescue sustained haematopoiesis - while the AF9 protein-protein interaction is dispensable there.
Elucidating the Importance of DOT1L Recruitment in MLL-AF9 Leukemia and Hematopoiesis. · Grigsby SM, Friedman A, Chase J, et al. · Cancers (Basel) 2021What this evidence cannot show
Abstract read from the eutils record; full text not read. Cited here to qualify this entry's own selectivity argument: an activator knockdown that lowers DOT1L catalytic output is acting on the function normal marrow needs, so the modality's advantage over a catalytic inhibitor is not that it spares catalysis but that the activator's expression is restricted.
Evidence that argues against this (1)
- If LAMP5-AS1 expression is genuinely restricted to MLL-rearranged cells, then lowering catalytic output only where the RNA is present sidesteps the marrow requirement entirely, and the qualification in this item does not apply. Whether the restriction is absolute or relative is not established by the source cohort of 58 patients. Wang WT et al. 2020
lncRNA binds the lysine-rich region of the DOT1L catalytic domain and raises activity
raises global H3K79me2 and me3
upregulates the HOXA-cluster self-renewal programme
a third modality: remove an activator of catalysis rather than the enzyme or its recruitment
the H3K79 arm is eight MOLM-13 samples, one replicate per condition