PAXIP1 (PTIP) - the subunit that tells an MLL3/4 complex apart from a menin complex, and the only one with a leukaemic-niche phenotype
also called PTIP · PAXIP1L · Ptip (mouse)
PTIP is the marker that distinguishes the two halves of this protein family: complexes that contain it never contain menin, and vice versa. In mice it turns out to be needed by the bone that houses leukaemia rather than by the leukaemia cell itself.
The same thing, for a scientist
PAXIP1/PTIP (UniProt Q6ZW49, 1,069 aa) is the BRCT-domain subunit unique to the MLL3/MLL4 COMPASS-like complexes; quantitative AP-MS of all six human SET1/MLL complexes assigns PTIP and PA1/PAGR1 to MLL3/4 and menin and PSIP1 to MLL1/2, with no overlap [E1]. **PTIP is therefore the cleanest available discriminator between the menin branch and the tumour-suppressor branch of this family**, and it is what a paper means when it frames a result at the level of the MLL3/4 complex. Its leukaemia phenotype is unusual and worth this graph's attention because it is **non-cell-autonomous**: PTIP is required to maintain the integrity of the bone marrow niche by promoting osteoclast differentiation through chromatin changes at *Pparg*, its deletion drastically reduces marrow HSCs and induces extramedullary haematopoiesis, and exposing AML cells to a PTIP-deficient marrow microenvironment reduces leukaemia-initiating cells and increases survival on transplantation [E2]. Separately, PTIP does substantial work **outside** the MLL3/4 complex: a PTIP-PA1 subcomplex promotes IgH class-switch transcription independently of MLL3/MLL4 and independently of PTIP's localisation to DNA damage [E3], so a PTIP result is not automatically an MLL3/4 result.
Why this entry exists in a KMT2A graph
Three reasons, in order of force.
1. It is the discriminator. This graph repeatedly has to decide whether a chromatin result belongs to the KMT2A-menin axis or to the MLL3/4-UTX axis, and the subunit compositions are the only non-negotiable difference. Menin and LEDGF sit on KMT2A and KMT2B; PTIP and PA1 sit on KMT2C and KMT2D; the quantitative measurement found no overlap in either direction 1. A claim that a menin inhibitor affects the MLL3/4 arm is therefore necessarily a claim about chromatin competition, never about a shared subunit.
2. The niche result is the kind this graph is short of. marrow-niche entries in this project argue that the leukaemic microenvironment is modifiable. 2 is a genetic demonstration that a chromatin regulator acting in the niche rather than in the blast changes leukaemia-initiating cell numbers and transplant survival. The mechanism is osteoclast differentiation via Pparg, which is neither H3K4 methylation of a leukaemia gene nor anything the fusion touches.
3. It warns against a complex-level attribution. 3 dissects PTIP's complexes in primary lymphocytes by quantitative proteomics and finds that its transcription and class-switch functions operate separately from its MLL3/MLL4 association and from its DNA-damage localisation, with a tandem BRCT domain and PA1 sufficient. So 'PTIP is a unique subunit of MLL3/4' and 'this PTIP phenotype is an MLL3/4 phenotype' are different statements and the second is often false.
Bounded absence
(PAXIP1 OR PTIP) AND (leukemia OR leukaemia) returns 8 records on 2026-09-20, all title-checked 4. Only 2 is a leukaemia study; the rest are class-switch recombination, DNA damage, UTX TPR mutations, TGF-beta transcription and PITX2. No record pairs PTIP with a KMT2A fusion. The complex forms required by rule 64 were run: ASCOM returns 3 and the MLL3/4-complex phrase query returns 37 dominated by plant biology; neither surfaces a PTIP-plus-KMT2A-fusion paper. The repository was grepped before recording this and no entry in this graph cites such an experiment.
PTIP is short and word-like as an alias, so the 8 records were title-checked rather than counted - the opposite failure from an under-counted retired symbol, and one this project has been caught by (RD for NELFE returned 115 where the true number was 2).
| uniprot | Q6ZW49 |
|---|---|
| length AA | 1069 |
| gene primary verified | PAXIP1 |
| accession resolution | gene_exact:PAXIP1 AND organism_id:9606 AND reviewed:true returned exactly one row; gene_primary equals PAXIP1; length 1069 asserted |
| complex branch | MLL3/MLL4 COMPASS-like, specific subunit; mutually exclusive with the menin/LEDGF branch |
| discriminator role | presence of PTIP/PA1 versus menin/LEDGF is the measured subunit-level difference between the KMT2C/KMT2D and KMT2A/KMT2B branches |
| leukaemia phenotype | non-cell-autonomous - PTIP deletion degrades the marrow niche via impaired osteoclast differentiation (Pparg), reducing leukaemia-initiating cells and improving transplant survival |
| mll34 independent functions | PTIP-PA1 subcomplex drives IgH class-switch transcription independently of MLL3/MLL4 and of DNA-damage localisation |
| fusion association evidence | none published |
pubmed counts 2026 09 20
{
"(PAXIP1 OR PTIP) AND (leukemia OR leukaemia)": 8
}aliases searched
[ "PAXIP1", "PTIP", "PAXIP1L" ]
depmap kmt2a selectivity
{
"chronos_mean_kmt2a_rearranged": -0.5644,
"chronos_mean_other_leukaemia": -0.4405,
"difference": -0.1239,
"pan_essentiality_all_lines": -0.2264,
"selectivity_rank": 731,
"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."
}- The only anti-leukaemic effect reported for PTIP loss comes with drastic HSC depletion and extramedullary haematopoiesis. Damaging the niche to starve the leukaemia starves the graft too, and in an infant with marrow-replacing disease that is not a window, it is the same injury.
- A PTIP result is not an MLL3/4 result. PTIP works outside the complex, so any inference from PTIP pharmacology or genetics to the tumour-suppressor branch of this family needs a separation-of-function allele that does not exist.
- There is no PTIP chemical matter at all. This is a BRCT-domain scaffolding protein with no reported ligand, so the discriminator role is diagnostic, not therapeutic.
- Does the niche result transfer to a KMT2A-rearranged leukaemia? The AML model used is not stated in the abstract and is not KMT2A-rearranged. Infant t(4;11) blasts are exquisitely niche-dependent in this graph's own entries, so this is the single most valuable untested transfer in the entry.
- Is PTIP measurable as a marker of which COMPASS branch a chromatin result belongs to? If the discriminator argument is right, a PTIP co-IP or CUT&RUN would classify an ambiguous result in one experiment - and nobody uses it that way.
- How much of the PTIP leukaemia phenotype is MLL3/4 and how much is the PTIP-PA1 subcomplex? The CSR work says the two are separable; the niche work does not test the separation.
- E1strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper
Quantitative label-free AP-MS across the six human SET1/MLL complexes assigned PTIP (PAXIP1) and PA1 as the specific subunits of the MLL3/4 complexes and menin and PSIP1 as the specific subunits of MLL1/2, and found that menin and Psip1 do not interact with the Set1a/b or Mll3/4 methyltransferases.
Quantitative dissection and stoichiometry determination of the human SET1/MLL histone methyltransferase complexes. · van Nuland R, Smits AH, Pallaki P, Jansen PW, Vermeulen M, Timmers HT · Mol Cell Biol 2013What this evidence cannot show
Full text read via the PMC HTML route (PMC3647974) on 2026-09-20 after routes 2 and 3 failed. HeLa, GFP-tagged overexpressed baits, and the MLL3/4 baits are isoform-3 fragment accessions. A single cell type: subunit exchange between branches under a different signalling state would not be visible here.
Evidence that argues against this (1)
- PTIP's own complexes are more heterogeneous than a clean branch assignment implies: quantitative proteomics in primary lymphocytes shows a PTIP-PA1 subcomplex that functions independently of MLL3/MLL4, so 'PTIP means MLL3/4' is a good heuristic and not a fact about every PTIP molecule. Starnes LM et al. 2016
- E2strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper
PTIP is required to maintain the integrity of the bone marrow niche by promoting osteoclast differentiation, directly promoting chromatin changes required for expression of Pparg; PTIP deletion drastically reduces HSCs in the marrow and induces extramedullary haematopoiesis, and exposure of acute myeloid leukaemia cells to a PTIP-deficient bone marrow microenvironment reduces leukaemia-initiating cells and increases survival upon transplantation.
Histone methylation regulator PTIP is required to maintain normal and leukemic bone marrow niches. · Das P, Veazey KJ, Van HT, et al. · Proc Natl Acad Sci U S A 2018What this evidence cannot show
Abstract fetched and read 2026-09-20; full text (PMC6205459) not obtained in this beat, so the AML model used, the magnitude of the LIC reduction and the survival numbers are not recorded here. Mouse genetics with whole-compartment PTIP deletion - the niche defect and the HSC loss are not separable from each other, so 'the leukaemia does worse in a PTIP-deficient niche' is confounded by a niche that supports normal haematopoiesis badly too. **No KMT2A rearrangement anywhere in the study.** A therapeutic reading would require damaging the niche in a child, which is not a plausible intervention.
Evidence that argues against this (1)
- A niche in which leukaemia-initiating cells fall is also a niche in which HSCs are drastically reduced and haematopoiesis moves extramedullary, so the anti-leukaemic effect and the marrow-failure effect are the same lesion. Nothing in the study separates a therapeutic window from marrow destruction. Starnes LM et al. 2016
- E3strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper
Dissection of PTIP's minimal structural requirements and protein complexes by quantitative proteomics in primary lymphocytes showed that PTIP functions in transcription and class switch recombination separately from its association with the MLL3/MLL4 complex and separately from its localisation to sites of DNA damage, with a tandem BRCT domain sufficient for class switching and PA1 as its main functional partner.
A PTIP-PA1 subcomplex promotes transcription for IgH class switching independently from the associated MLL3/MLL4 methyltransferase complex. · Starnes LM, Su D, Pikkupeura LM, et al. · Genes Dev 2016What this evidence cannot show
Abstract fetched and read 2026-09-20; full text (PMC4719306) not obtained. Primary murine B lymphocytes undergoing class switch recombination - a specialised programmed-DNA-damage context, not a leukaemic one. The separation of functions is established for CSR and transcription at the Igh locus and does not establish that PTIP's other functions are MLL3/4-independent.
Evidence that argues against this (1)
- The same subunit assignment that makes PTIP a discriminator is a population statement: PTIP is found in MLL3/4 complexes and MLL3/4 complexes contain PTIP, and a subcomplex that lacks the methyltransferase does not overturn that. The risk is in the reverse inference, from a PTIP phenotype to an MLL3/4 mechanism. van Nuland R et al. 2013
- E4moderate confidence 3 of 5: One strong study, or several smaller consistent ones.dataset
PubMed run 2026-09-20: `(PAXIP1 OR PTIP) AND (leukemia OR leukaemia)` returns 8 records, all title-checked, of which exactly one is a leukaemia study and none pairs PTIP with a KMT2A fusion protein; UniProt reviewed human record Q6ZW49 gives gene_primary PAXIP1, length 1,069 aa, synonyms PAXIP1L and PTIP.
NCBI PubMed E-utilities esearch with querytranslation inspected, and UniProtKB reviewed human record Q6ZW49 (PAXIP1); queries run 2026-09-20 by this project · NCBI PubMed; UniProt Consortium; queries run by this project · Database 2026What this evidence cannot show
PTIP is a short word-like alias and inflates counts in other fields, so titles were read rather than the raw count taken; the reported 8 is the raw count and all 8 were inspected. The complex forms ASCOM and the MLL3/4-complex phrase were also run per rule 64. A paper framing a result at 'the MLL3/4 complex' without naming PTIP would not be retrieved by the symbol query, and the phrase query is heavily contaminated by plant literature.
Evidence that argues against this (1)
- The complex-form route is the documented way a subunit paper hides from a symbol query, and this project has recorded a wrong absence produced exactly that way. A count of 8 on a gene whose principal identity is 'a subunit of a complex' should be treated as a lower bound. Kato H et al. 2020
PTIP is a specific subunit of the MLL3/MLL4 complexes and absent from the menin branch
PTIP/PA1 mark the KMT2C/KMT2D branch; menin/LEDGF mark the KMT2A/KMT2B branch
8 records title-checked; the complex forms ASCOM and 'MLL3/4 complex' run per rule 64