The little elongation complex (ELL + ICE1 + ICE2 + ZC3H8) — the other place ELL works, and why ELL's pan-essentiality is probably not about the fusion
also called LEC · little elongation complex · ICE1/ICE2 complex · ELL-ICE1-ICE2-ZC3H8
ELL, one of the proteins in the machine the KMT2A fusion hijacks, also has a second job in a completely different complex that makes small nuclear RNAs. That second job is probably why every cell needs ELL, which means ELL's essentiality is not evidence that it matters to the fusion.
The same thing, for a scientist
The little elongation complex (LEC) comprises **ELL, ICE1, ICE2 and ZC3H8**, and drives RNA Pol II transcription of spliceosomal snRNA and snoRNA genes; the 7SK snRNP collaborates with it, and depletion of 7SK snRNA or LARP7 disrupts LEC integrity and reduces nascent snRNA/snoRNA synthesis. This entry exists because it reinterprets a number the graph already carries. ELL is DepMap common-essential (mean -1.30, 94% of lines depleted) while its SEC-slot paralogue ELL2 is not (-0.06, 1.0%), and the graph has treated that asymmetry as an open puzzle. LEC supplies the most likely resolution: **ELL has a second, SEC-independent job in snRNA gene transcription that ELL2 does not share**, and snRNA synthesis is required by every cell. If that is right, ELL's pan-essentiality is evidence about LEC rather than about the KMT2A fusion, and reasoning from 'ELL is essential' to 'the SEC ELL slot is a target' is invalid. The LEC subunits themselves show broad requirement and no KMT2A selectivity (ICE1 -0.773 mean, 83% depleted, difference -0.041; ICE2 -0.607, 54%, -0.080; ZC3H8 -0.775, 91%, **+0.126**).
This is a prediction, not a measurement. No one has seen this protein in this shape. The colours say how much the software trusts itself, and that is worth reading before anything else — the orange stretches are places the model is effectively saying this part has no fixed shape.
I have never read one of these pictures before
A protein is one long chain of amino acids that folds up. The ribbon traces that chain from one end to the other; it is not a wire, it is a simplification of where the backbone runs. Drag to rotate, scroll to zoom.
Domains are the stretches of chain that do a particular job, and they are what the coloured blocks above the picture mark out. Click one and the view flies to it. In this project the important question about any domain is usually whether the fusion still has it — the labels say lost where it does not.
Anything in yellow sticks is a bound small molecule, which is usually the drug.
What LEC is
The human 7SK snRNP — 7SK snRNA, MePCE and LARP7 — functions as a canonical transcription factor that, with the little elongation complex comprising ELL, ICE1, ICE2 and ZC3H8, promotes transcription of Pol II-specific spliceosomal snRNA and snoRNA genes 1. 7SK snRNA associates with a fraction of Pol II hyperphosphorylated at Ser5 and Ser7, the hallmark of polymerase engaged in snRNA synthesis, and ChIP and ChIRP found all 7SK snRNP components enriched on those genes. Depleting 7SK snRNA or LARP7 disrupts LEC integrity, inhibits Pol II recruitment to snRNA/snoRNA genes, and reduces nascent snRNA and snoRNA synthesis 1.
This is a complex the KMT2A fusion has nothing to do with. There is no KMT2A contact, no AFF1 contact, and no leukemia in the study. LEC is not part of the KMT2A::AFF1 interactome, and this entry is filed as a negative that changes the interpretation of a positive.
The reinterpretation, which is the reason to write this down
The graph's existing ELL2 entry poses the paralogue asymmetry as a live question: SEC contains one ELL-family protein and treats ELL, ELL2 and ELL3 as interchangeable, yet ELL is common-essential at -1.30 with 94% of lines depleted while ELL2 sits at -0.06 with 1.0%. Two explanations were offered there — ELL does something beyond the SEC slot that ELL2 cannot, or ELL2 is too narrowly expressed to score.
LEC is the concrete form of the first explanation. ELL, not ELL2, is the named LEC subunit 1. snRNA and snoRNA synthesis is required by every proliferating cell. A protein that is required for snRNA synthesis will be common-essential whatever its SEC biology, and its paralogue that is not in LEC will not be.
The consequence for this project is direct and negative: the observation 'ELL is pan-essential' carries no information about whether the SEC ELL slot matters to KMT2A::AFF1, because the essentiality is most simply explained by a complex the fusion never touches. Any argument that reasoned from ELL's essentiality toward the fusion needs to be withdrawn or re-grounded.
The LEC-specific subunits support this reading without proving it. ICE1 (2,266 aa) and ICE2 (982 aa) are themselves broadly required — -0.773 with 83.0% of lines depleted and -0.607 with 53.6% respectively — and ZC3H8 is required in 90.8% of lines at -0.775 2. None shows KMT2A selectivity; ZC3H8's difference is +0.126, the wrong sign. A complex whose dedicated subunits are broadly required in every lineage is doing housekeeping, which is what snRNA synthesis is.
What would settle it
The reinterpretation above is an inference, not a measurement. It would be settled by a separation-of-function experiment that nobody appears to have run: an ELL allele that retains LEC assembly but cannot enter SEC, and its converse. If the LEC-competent/SEC-incompetent allele rescues viability in an ELL-null background while the SEC-competent/LEC-incompetent allele does not, the pan-essentiality is LEC's and the SEC slot is free to be a target or not on its own evidence.
Until then this entry records the alternative explanation with its source, which is more than the graph had before.
| complex name | Little elongation complex (LEC) |
|---|---|
| uniprot | Q9Y2F5 |
| gene symbol | ICE1 |
| length AA | 2266 |
| protein name | Little elongation complex subunit 1 |
| alphafold ID | AF-Q9Y2F5-F1 |
| accession resolution | UniProt gene_exact queries with primary gene names read back: ICE1 -> Q9Y2F5 (2266 aa), ICE2 -> Q659A1 (982 aa), ZC3H8 -> Q8N5P1 (291 aa), ELL -> P55199. The top-level accession is ICE1, the largest dedicated subunit. |
| collaborating rnp | 7SK snRNP (7SK snRNA, MePCE, LARP7) |
| function | RNA Pol II transcription of spliceosomal snRNA and snoRNA genes |
| relationship to the fusion | CORRECTED cycle 113 - ONE contact now exists. IntAct IM-30059 / PMID 39251607 (Khoroshkin et al., Nature Communications 2024) records ZC3H8 as BAIT with KMT2A as PREY, MI:1314 proximity labelling, MI:1060 spoke expansion, in K562, miscore 0.27. Weak, single-method, spoke-expanded and in a fusion-free cell line, so it does not change this entry's conclusion - but the previous text, 'none - no KMT2A or AFF1 contact', is no longer accurate. Entered because ELL is shared with SEC and LEC membership is the most likely explanation of ELL's pan-essentiality. |
| shown for fusion | false |
| reinterprets | gene-ell2 (the ELL/ELL2 paralogue asymmetry) and protein-ell-eaf-module |
| druggability | Not a target. Housekeeping snRNA synthesis. |
subunits — the numbers behind the chart
[
{
"gene_symbol": "ELL",
"uniprot": "P55199",
"role": "shared with SEC; the overlap that makes this complex relevant",
"chronos_mean": -1.302,
"frac_depleted": 0.94
},
{
"gene_symbol": "ICE1",
"uniprot": "Q9Y2F5",
"length_aa": 2266,
"role": "LEC scaffold",
"chronos_mean": -0.773,
"frac_depleted": 0.8295,
"kmt2a_selectivity_difference": -0.041
},
{
"gene_symbol": "ICE2",
"uniprot": "Q659A1",
"length_aa": 982,
"role": "LEC subunit",
"chronos_mean": -0.6065,
"frac_depleted": 0.5364,
"kmt2a_selectivity_difference": -0.0795
},
{
"gene_symbol": "ZC3H8",
"uniprot": "Q8N5P1",
"length_aa": 291,
"role": "LEC subunit",
"chronos_mean": -0.7753,
"frac_depleted": 0.9081,
"kmt2a_selectivity_difference": 0.1256
}
]- If ELL is required for snRNA synthesis, any agent that removed ELL function would shut down splicing capacity in every tissue — an essential-function toxicity that a SEC-focused rationale would not have anticipated.
- A SEC-directed strategy that works by removing ELL from SEC could be rescued by the cell simply redistributing ELL from LEC, or could poison LEC as a side effect; neither has been modelled.
- Does an ELL allele competent for LEC but not SEC rescue viability in an ELL-null background? That single experiment would decide whether ELL's pan-essentiality is LEC's or SEC's, and it appears never to have been done.
- Is ELL2 or ELL3 ever found in LEC? If the LEC ELL slot is as paralogue-tolerant as the SEC slot is said to be, this entry's inference collapses.
- In a KMT2A::AFF1 B-ALL blast, what fraction of cellular ELL is in SEC and what fraction in LEC? Nobody has measured the partition in any cell, let alone this one.
- Does 7SK sequestration of P-TEFb couple the two complexes, so that perturbing SEC changes snRNA output and vice versa?
- E1established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.paper
The human 7SK snRNP, in collaboration with the little elongation complex comprising ELL, Ice1, Ice2 and ZC3H8, promotes RNA polymerase II transcription of spliceosomal snRNA and snoRNA genes; 7SK snRNA associates with Pol II hyperphosphorylated at Ser5 and Ser7; and depletion of 7SK snRNA or Larp7 disrupts LEC integrity, inhibits Pol II recruitment to those genes and reduces nascent snRNA and snoRNA synthesis.
The 7SK snRNP associates with the little elongation complex to promote snRNA gene expression. · Egloff S, Vitali P, Tellier M, Raffel R, Murphy S, Kiss T · The EMBO Journal 2017What this evidence cannot show
Abstract read in full; not full text. HeLa cells; no leukemia, no KMT2A, no fusion protein. The paper establishes LEC composition and function; it does not test whether ELL's essentiality is attributable to LEC rather than SEC, which is this entry's inference and is labelled as such.
Evidence that argues against this (1)
- The alternative explanation for the ELL/ELL2 asymmetry has not been excluded. ELL2 is expressed narrowly - high in liver and bone marrow, near-absent from classical monocytes and naive B cells - and a gene expressed in few of the screened lines scores as non-essential regardless of its biology. If that is the whole story, LEC membership is incidental and this entry's central inference is wrong. The two explanations are not mutually exclusive and neither has been tested directly. Human Protein Atlas; Broad Institute DepMap; Open Targets Platform 2026
- E2strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.dataset
ICE1 has DepMap mean gene effect -0.773 across 1,208 lines with 83.0% of lines depleted and a KMT2A-rearranged minus other-leukemia difference of -0.041; ICE2 has mean -0.6065 with 53.6% depleted and a difference of -0.080; ZC3H8 has mean -0.7753 with 90.8% depleted and a difference of +0.126.
DepMap CRISPR (Chronos) gene effect via the Open Targets Platform GraphQL API, with KMT2A-rearrangement status from Cellosaurus; per-gene recomputation for this beat · Broad Institute DepMap; Cellosaurus; Open Targets Platform; analysis run by this project · Database and derived analysis 2026What this evidence cannot show
DepMap release version not stated by the API. Eleven KMT2A-rearranged lines. That LEC subunits are broadly required is consistent with the housekeeping interpretation but does not demonstrate that ELL's essentiality comes from LEC rather than from something else.
Evidence that argues against this (1)
- Pan-essentiality is the finding here and it argues against the target rather than for it. A gene depleted in more than 90% of all cell lines is required by the patient's normal marrow as much as by the leukaemia, so its appearance in a KMT2A complex is a statement about the complex and not a therapeutic opening. ICE1 at 83% of lines depleted with a KMT2A-selectivity difference of −0.041 is a broadly required gene with no measurable selectivity. Oncofusion knowledge graph (internal analysis) 2026
Qualification: the 2.0 A OCEL-AFF4 structure settles the fold and the interface but not the bar's endpoints. The crystallised construct (5JW9) is O00472 519-640, wider than UniProt's 526-636 at both ends, and the databases spread the N-terminal boundary from 526 to 536. Only OCEL is a UniProt feature at all; ELL-N and the winged-helix module come from Pfam/InterPro models.
ELL is a LEC subunit as well as a SEC subunit — the overlap that explains its pan-essentiality
- refutes →ELL2 — a SEC subunit that almost no cell needs, next to a paralogue almost every cell needs
offers a mechanistic answer to the ELL/ELL2 paralogue asymmetry that entry left open
the second complex ELL-family factors work in, outside SEC
- related to →KMT2A::AFF1 (MLL-AF4), t(4;11)(q21;q23)
NOT an interactome member: no KMT2A or AFF1 contact; entered to reinterpret ELL's essentiality
LEC scaffold; required for the INITIATION phase of snRNA gene transcription
fourth LEC subunit; no phase of snRNA transcription assigned to it in the primary literature
Sits on the same elongation axis the graph covers through LEC, DSIF, NELF and SEC, and is the chromatin-chaperone arm none of those entries covers.
the fourth LEC subunit (ELL, ICE1, ICE2, ZC3H8) and the only one that had no entry of its own
- ← related toSEC-L2 — the AFF2-scaffolded super-elongation-like complex: P-TEFb plus ENL and AF9, and NO ELL
the ELL module absent from SEC-L2 has its own AFF-independent complex