AnalysismoderateQ145-irx1-in-hoxa-negative-infants

In the HOXA-dead infant states IRX1 carries H3K4me3 and ENL while HOXA9/HOXA10 carry neither

In two infant leukaemia samples the HOXA genes are completely shut — no active mark, no machinery. In the very same cells IRX1 has both. The cells did not lose the programme; they moved it.

The same thing, for a scientist

In p318 (the AML arising from p279's lineage switch) and p454 (an independent infant B-ALL), HOXA9 and HOXA10 carry H3K4me3 at **0.004-0.068x** the mean promoter and ENL at **0.000-0.109x** — a locus that is not merely unoccupied by the fusion but epigenetically dead. In the same nuclei the IRX1 promoter carries **H3K4me3 at 0.705-0.969x** and **ENL at 4.4-6.5x** [E1]. This matters because it distinguishes the two readings of q142's HOXA result: the fusion is not simply absent from HOXA in a cell that has shut everything down; the cell is running a different homeobox locus with the elongation machinery on it.

Go deeper

The numbers

Within-sample share; H3K4me3 R1/R2 are true replicates of one antibody (Active Motif 39159), ENL R1/R2 likewise (CST D9M4B). Every number is a WITHIN-SAMPLE share: an interval's mean coverage divided by that same library's mean promoter coverage, so 1.0 = an average promoter in that library. GSE252378 has no spike-in and no drug arm, so no between-sample absolute statement is licensed; the only cross-sample quantity used is a ratio of two loci measured inside one library, where the unknown per-library scale cancels.

sample	track	IRX1	HOXA9	HOXA10	MEIS1 (positive)	GAPDH
p279	H3K4me3 R1/R2	1.42 / 1.45	1.10 / 0.86	0.92 / 0.93	3.5 / 3.0	3.5 / 3.7
p318	H3K4me3 R1/R2	0.76 / 0.71	0.004 / 0.008	0.034 / 0.009	3.4 / 4.2	4.1 / 3.5
p454	H3K4me3 R1/R2	0.86 / 0.97	0.008 / 0.005	0.044 / 0.068	3.9 / 3.9	2.8 / 2.9
p279	ENL R1/R2	12.2 / 10.3	9.3 / 2.1	2.2 / 1.8	57 / 77	6.1 / 4.7
p318	ENL R1/R2	4.4 / 2.7	0.000 / 0.000	0.000 / 0.000	41 / 48	12 / 18
p454	ENL R1/R2	6.2 / 6.5	0.078 / 0.109	0.000 / 0.000	75 / 78	2.0 / 2.7

The H3K4me3 ratio IRX1:HOXA9 is ~100-190x in p318 and p454, and ~1.3x in p279.

Why H3K4me3 is the load-bearing track here, not ENL

ENL is a cofactor and the deposit paper itself shows ENL persisting at oncoprotein-target loci in a lineage-switching sample after the oncoprotein and menin have fallen to undetectable levels 2. So high ENL at IRX1 is consistent with the fusion being there and also with the fusion having been there. H3K4me3 is the cleaner statement: it says the IRX1 promoter is in an active chromatin state while HOXA9/HOXA10 are not, independently of what is bound.

IRX1's H3K4me3 is ~0.7-0.97x, i.e. about an average promoter — not a hot one. This is a modest mark, and the contrast with HOXA is carried mostly by how completely dead HOXA is.

Facts and figures

mark	H3K4me3
cofactor	ENL (MLLT1)

samples

[
  "p318",
  "p454"
]

h3k4me3 irx1

[
  0.759,
  0.705,
  0.858,
  0.969
]

h3k4me3 hoxa9

[
  0.004,
  0.008,
  0.008,
  0.005
]

enl irx1

[
  4.416,
  2.672,
  6.213,
  6.543
]

enl hoxa9

[
  0,
  0,
  0.078,
  0.109
]

How this could bite us back

If HOXA is epigenetically dead rather than merely unbound, re-activating it is not a therapeutic route in these patients and any strategy premised on differentiating them through HOXA is misdirected.

Still open

Is the IRX1 promoter H3K4me3 deposited by the fusion's own retained KMT2A N-terminus, by wild-type KMT2A, or by another SET-domain enzyme? KMT2A-C is near zero at IRX1, which argues against wild-type KMT2A, but that is an absence in a sparse track.
Does H3K27me3 occupy HOXA9/HOXA10 in these samples? No H3K27me3 track exists in GSE252378, so 'dead' cannot currently be resolved into 'polycomb-repressed' versus 'simply never opened'.

Every source behind this entry (2)

E1moderate confidence 3 of 5: One strong study, or several smaller consistent ones.dataset
In p318 and p454, H3K4me3 at HOXA9/HOXA10 is 0.004-0.068x mean promoter and at IRX1 is 0.705-0.969x, with ENL at 0.000-0.109x and 4.4-6.5x respectively.
q145 re-analysis of GSE252378: IRX1 and HOXA oncoprotein occupancy across seven KMT2A::AFF1 patients (analyses/q145-irx1-in-hoxa-negative-infants) · Iqbal Lab oncofusion knowledge graph, beat q145 · internal analysis over deposited GSE252378 bigWigs 2026
What this evidence cannot show
Two sample records, one of which (p318) is a post-chemotherapy relapse AML and the other an untreated B-ALL, so the pair is not a replicate of anything. Activity tracks do not exist for the other five patients. Every number is a WITHIN-SAMPLE share: an interval's mean coverage divided by that same library's mean promoter coverage, so 1.0 = an average promoter in that library. **GSE252378 has no spike-in and no drug arm**, so no between-sample absolute statement is licensed; the only cross-sample quantity used is a ratio of two loci measured inside one library, where the unknown per-library scale cancels.
Evidence that argues against this (1)
- One deposit, one lab, one assay, one genome build, and activity tracks in only three of the eight sample records. A second cohort profiled by a different group could reverse any of this. Janssens DH et al. 2024
E2strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper
In a lineage-switching sample treated with revumenib, ENL persists on numerous oncoprotein-target loci although the oncoprotein and menin fall to undetectable levels.
MLL oncoprotein levels influence leukemia lineage identities · Janssens DH, Duran M, Otto DJ, Wu W, Xu Y, Kirkey D, Mullighan CG, Yi JS, Meshinchi S, Sarthy JF, Ahmad K, Henikoff S · Nat Commun 2024
PMID 39472576 DOI
What this evidence cannot show
That observation is from the KMT2A::ENL menin-inhibitor pair in the same deposit, not from the KMT2A::AFF1 samples analysed here, so its transfer to this context is an inference about the assay and not a measurement in these patients.
Evidence that argues against this (1)
- If ENL persistence is specific to menin-inhibitor treatment, it may not apply to these untreated KMT2A::AFF1 samples, and ENL would then be a better fusion proxy here than this caveat allows. Janssens DH et al. 2024

Connected to

supports →IRX1 — Iroquois homeobox 1, the gene that names the HOXA-negative half of infant t(4;11) ALL
IRX1 carries H3K4me3 and ENL where HOXA carries neither
related to →MLLT1 / ENL — the reader that beat its own paralogue, with a degrader already in hand
ENL at IRX1 at 4.4-6.5x while HOXA9 reads 0.000-0.109x
supports →iALL-IRX1 — the HOXA-negative, IRX1-high state of infant KMT2A::AFF1 ALL
the chromatin signature of the state
related to →DOT1L — the target where excellent chemistry met an indirect interaction
DOT1L measured at IRX1 alongside ENL, and is much weaker (0.53-4.61x)