MEPCE - absent from this graph until now, and a deeper pan-dependency than BRD4
also called MEPCE · BCDIN3 · BCDIN3 domain containing · 7SK snRNA methylphosphate capping enzyme
MEPCE puts a protective cap on the 5' end of the 7SK RNA and then stays bound to hold the particle together. The string MEPCE appeared nowhere in this knowledge graph before this entry, even though cells need it about as much as they need BRD4.
The same thing, for a scientist
MEPCE (UniProt Q7L2J0, 689 aa; the gene was formerly BCDIN3) is the 7SK snRNA methylphosphate capping enzyme, identified as such from a proteomic survey of the human transcription machinery that placed it in an snRNP containing P-TEFb [E1]. It caps 7SK co-transcriptionally while LARP7-free, loses capping activity once LARP7 joins - probably because LARP7 occludes its catalytic centre - and then exerts a capping-independent function promoting the LARP7-7SK interaction, so MEPCE and LARP7 cooperatively stabilise 7SK [E2]. A de novo MEPCE nonsense variant in a boy with global developmental delay and seizures causes nonsense-mediated decay of MEPCE, downregulation of LARP7 and 7SK, upregulation of HEXIM1, reduced HEXIM1-cyclin T1 binding, hyperphosphorylation of the Pol II CTD and enhanced P-TEFb-dependent transcription - the reservoir disintegrating in a living human [E3]. **MEPCE was mentioned zero times anywhere in this graph before this entry**, and on this project's own scan it has mean Chronos -1.187 across 1,208 lines, deeper than BRD4's -1.015 [E4]. It is expressed in every KMT2A-rearranged panel line (median 64.2 nTPM) [E5].
A pan-essential gene hiding inside a negative regulator
| gene | mean Chronos, 1,208 lines | KMT2A-r myeloid | KMT2A-r lymphoid | other leukaemia | scan rank |
|---|---|---|---|---|---|
| CDK9 | -1.536 | -1.826 | -1.555 | -1.580 | 1,136 |
| MEPCE | -1.187 | -1.279 | -1.109 | -1.161 | 2,745 |
| BRD4 | -1.015 | - | - | - | 3,122 |
| LARP7 | -0.129 | -0.129 | -0.087 | -0.098 | 6,009 |
MEPCE sits with CDK9 and BRD4, not with LARP7 and HEXIM1 4. The obvious inference - that it is essential because the reservoir is essential - is wrong, because LARP7, its partner in the same core, is not. The likelier reading is that MEPCE has functions beyond 7SK; it is a methyltransferase-fold protein and the gene's former name, BCDIN3, comes from a domain family whose other member caps a different small RNA. This entry does not claim to know which function carries the essentiality, and the difference between the KMT2A-rearranged arm and other leukaemia lines is -0.06, which is nothing.
The finding for this project is narrower and solid: a gene that this graph never mentioned once, sitting in the middle of the mechanism it spends most of its pages on, is more essential than BRD4.
The human variant is the best functional evidence that the reservoir does something
Patient fibroblasts carrying a de novo MEPCE nonsense allele show the whole particle unravelling in sequence: MEPCE down by nonsense-mediated decay, then LARP7 and 7SK down, HEXIM1 up, less HEXIM1 bound to cyclin T1, a hyperphosphorylated Pol II CTD, and upregulation of ID2, ID3, MRPL11 and the U1, U2 and U4 snRNAs 3. Flavopiridol - a CDK9 inhibitor - and ectopic MEPCE both rescued the overexpression of six Pol II-sensitive genes, which closes the loop pharmacologically.
The phenotype is developmental delay and seizures. As with LARP7 and Alazami syndrome, losing the brake in a human does not produce leukaemia, which is a genuine constraint on any story in which an intact reservoir is what stands between a KMT2A fusion and unrestrained elongation.
| uniprot | Q7L2J0 |
|---|---|
| length AA | 689 |
| gene symbol | MEPCE |
| uniprot verified | Q7L2J0 = MEPCE_HUMAN, primary gene name MEPCE (synonym BCDIN3), 689 aa, 7SK snRNA methylphosphate capping enzyme, Homo sapiens; UniProt REST fetched and read back 2026-09-20 |
| role | gamma-monomethyl phosphate cap on the 7SK 5' end; capping-independent stabilisation of the LARP7-7SK interaction inside the particle |
| human lof phenotype | de novo nonsense variant: global developmental delay and seizures, with 7SK snRNP disintegration (PMID 31467394) |
| breakpoint side | not applicable - no documented KMT2A contact |
| graph gap note | Before this entry the string MEPCE appeared zero times in any node of this knowledge graph, and BCDIN3 likewise. |
essentiality
{
"chronos_mean": -1.1866,
"n_lines": 1208,
"chronos_mean_kmt2ar_myeloid": -1.279,
"chronos_mean_kmt2ar_lymphoid": -1.1088,
"chronos_mean_other_leukemia": -1.1611,
"scan_rank": 2745,
"scan_n_genes": 17728,
"difference": -0.0605,
"interpretation": "pan-essential, deeper than BRD4; no selectivity"
}expression hpa celline nTPM — the numbers behind the chart
{
"SEM": 61,
"RS4;11": 337,
"MV4-11": 82.3,
"THP-1": 61.9,
"NOMO-1": 64.2,
"MOLM-13": 68.4,
"OCI-AML-2": 46,
"median": 64.2
}related entries
[ "protein-7sk-snrnp", "gene-larp7", "noncoding-rn7sk" ]
depmap kmt2a selectivity
{
"chronos_mean_kmt2a_rearranged": -1.2185,
"chronos_mean_other_leukaemia": -1.158,
"difference": -0.0605,
"pan_essentiality_all_lines": -1.1866,
"selectivity_rank": 2745,
"rank_denominator": 17728,
"provenance": "Joined 2026-09-21 from analyses/depmap-kmt2a-selective-scan-full (generated 2026-09-19T10:12:06Z), the project's own genome-wide scan over 11 KMT2A-rearranged and 43 other leukaemia DepMap lines. NOT re-fetched and NOT independent of that scan: every caveat it carries applies here, including that the KMT2A-rearranged arm is a small number of lines drawn from nine individuals, that the rank denominator is 17,728 genes, and that a difference is not a dependency. Added because only 26 of 308 gene entries carried this number, which made any KMT2A-selectivity question answerable for 8% of the corpus.",
"how_to_read": "A negative difference means MORE required in KMT2A-rearranged lines than in other leukaemia lines. A rank near 17,728 is ANTI-selective. Neither is evidence of a fusion-driven dependency on its own: check pan_essentiality_all_lines before calling anything selective, and note that a paralogue-redundant gene can score near zero for reasons that have nothing to do with this disease - see ana-the-paralogue-redundancy-excuse-is-not-tested-and-my-proxy-decided-the-answer."
}- MEPCE is depleted more deeply than BRD4 across 1,208 lines, so it is a pan-essential gene and not a therapeutic target in this disease.
- The only human with a characterised MEPCE loss-of-function allele has developmental delay and seizures, which is what partially dismantling the reservoir costs in a person.
- Which MEPCE function carries its essentiality - 7SK capping, snRNP stabilisation, or something unrelated to 7SK? A separation-of-function allele would settle it and none has been reported.
- Does MEPCE have substrates besides 7SK in human cells? Its former name places it in a domain family whose other member caps a different small RNA.
- Why is MEPCE pan-essential when LARP7, its partner in the same stable core, is not?
- E1established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.paper
BCDIN3 is the 7SK snRNA methylphosphate capping enzyme, present in an snRNP complex containing both RNA-processing and transcription factors including the elongation factor P-TEFb.
Systematic analysis of the protein interaction network for the human transcription machinery reveals the identity of the 7SK capping enzyme. · Jeronimo C, Forget D, Bouchard A, Li Q, Chua G, Poitras C, Therien C, Bergeron D, Bourassa S, Greenblatt J, Chabot B, Poirier GG, et al · Mol Cell 2007What this evidence cannot show
Identified within a large affinity-purification/mass-spectrometry survey of 32 tagged polypeptides; the enzymatic assignment rests on complex membership plus activity rather than on a structure.
Evidence that argues against this (1)
- A dedicated follow-up shows that inside the assembled snRNP MEPCE has no capping activity at all, so the name describes a reaction that happens before assembly and not the function the protein performs in the particle. Xue Y et al. 2010
- E2established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.paper
MePCE caps 7SK while LARP7-free and probably co-transcriptionally; within the 7SK snRNP it loses capping activity, probably through occlusion of its catalytic centre by LARP7, and then promotes the LARP7-7SK interaction so that MePCE and LARP7 act cooperatively to stabilise 7SK and maintain snRNP integrity.
A capping-independent function of MePCE in stabilizing 7SK snRNA and facilitating the assembly of 7SK snRNP. · Xue Y, Yang Z, Chen R, Zhou Q · Nucleic Acids Res 2010What this evidence cannot show
Biochemistry and co-IP in human cell lines. The occlusion mechanism is stated as 'probably' by the authors and was not shown structurally.
Evidence that argues against this (1)
- A human MEPCE nonsense variant reduces LARP7 and 7SK as predicted but also raises HEXIM1, an outcome the cooperative-stabilisation model alone does not anticipate, so the system compensates in ways cell-line biochemistry did not show. Schneeberger PE et al. 2019
- E3established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.paper
A de novo MEPCE nonsense variant (c.1552C>T, p.Arg518*) in a boy with global developmental delay and seizures causes nonsense-mediated decay of MEPCE mRNA, downregulation of LARP7 and 7SK snRNA, upregulation of HEXIM1, reduced HEXIM1 binding to cyclin T1, hyperphosphorylation of the Pol II C-terminal domain and upregulation of ID2, ID3, MRPL11 and snRNAs U1, U2 and U4; flavopiridol and ectopic MEPCE both rescued the overexpression of six Pol II-sensitive genes.
de novo MEPCE nonsense variant associated with a neurodevelopmental disorder causes disintegration of 7SK snRNP and enhanced RNA polymerase II activation. · Schneeberger PE, Bierhals T, Neu A, Hempel M, Kutsche K · Sci Rep 2019What this evidence cannot show
A single patient, assayed in fibroblasts. n=1, so the genotype-phenotype link is a report rather than a statistical association, and fibroblasts are not blood cells.
Evidence that argues against this (1)
- The same paper's rescue with flavopiridol shows the transcriptional phenotype is P-TEFb-dependent, but flavopiridol is a promiscuous CDK inhibitor, so the rescue does not isolate CDK9. Schneeberger PE et al. 2019
- E4strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.database
MEPCE has mean Chronos -1.187 across 1,208 DepMap lines - deeper than BRD4's -1.015 - with -1.279 in the KMT2A-rearranged myeloid half and -1.109 in the lymphoid half against -1.161 in other leukaemia lines, ranking 2,745th of 17,728 on index-minus-comparator difference.
DepMap CRISPR (Chronos) gene effect via the Open Targets Platform GraphQL API, re-queried for this beat on 2026-09-20 and split by lineage inside the KMT2A-rearranged arm; genome-scale ranks read from this repository's analyses/depmap-kmt2a-selective-scan-full/results.json (17,728 genes, 1,208 lines) · Broad Institute DepMap; Open Targets Platform; Oncofusion KG internal scan · Databases 2026What this evidence cannot show
Six myeloid and four lymphoid lines. The index-minus-comparator difference is -0.06, which is inside noise, and at this depth of depletion the measurement is floor-compressed: this is a pan-essentiality claim and explicitly not a selectivity claim.
Evidence that argues against this (1)
- LARP7, MEPCE's partner in the same stable core, is not essential at all (-0.129), so the essentiality cannot be attributed to the 7SK snRNP without further evidence, and no such evidence is cited here. Krueger BJ et al. 2008
- E5established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.database
MEPCE is detected in all seven KMT2A-rearranged panel lines the Human Protein Atlas carries, median 64.2 nTPM.
Human Protein Atlas cell-line RNA nTPM, rna_celline.tsv (24,315,372 rows, 1,206 cell lines), downloaded and filtered for this beat on 2026-09-20 · Human Protein Atlas · Database 2026What this evidence cannot show
Bulk poly-A cell-line RNA. RS4;11 returns 337.0 nTPM against 46-82 in the other six and is a high outlier for several unrelated genes in the same pull, so the median is quoted rather than the mean.
Evidence that argues against this (1)
- Expression cannot distinguish the capping function from the capping-independent one, and the essentiality may belong to neither. Xue Y et al. 2010
stable core component; caps 7SK before assembly and stabilises the particle afterwards
gamma-monomethyl phosphate cap on the 5' end
- co-occurs with →LARP7 - the clamp on the 3' end of 7SK, whose human loss-of-function phenotype is dwarfism and intellectual disability rather than cancer
cooperate to stabilise 7SK; LARP7 occludes MEPCE's catalytic centre in the assembled particle
supplies the epigenetics record gene-mepce lacks: an RNA methyltransferase, not a histone one