Lost from the KMT2A half: PHD2 (aa 1479-1533), an E3 ubiquitin ligase that limits KMT2A's own activity
also called MLL PHD2 E3 ligase · CDC34 MLL · MLL self-ubiquitination · PHD1 PHD2
One of KMT2A's own finger-shaped modules is an enzyme that tags KMT2A itself for destruction, keeping its levels and its activity down. The fusion protein does not have it, so that self-limiting loop is gone.
The same thing, for a scientist
The second PHD finger of MLL1 (UniProt Q03164, aa 1479-1533) is an E3 ubiquitin ligase in the presence of the E2 conjugating enzyme CDC34; the activity is conserved in MLL4's second PHD finger but not in MLL2 or MLL3. Mutating PHD2 **stabilises** MLL1 and **increases** both its transactivation ability and its recruitment to target loci, so PHD2 negatively regulates MLL1 [E1]. All four PHD fingers are 'invariably deleted in oncogenic MLL1 fusion proteins'. This is a second, mechanistically independent brake in the same lost region as the PHD3/CyP33 switch, and it is the one the graph has not recorded: PHD3 controls *what KMT2A does* (activate or repress); PHD2 controls *how much KMT2A there is and how much target it reaches*. The fusion sheds both in one break, which is the strongest structural argument for a 'loss of restraint' reading - and also the place where that reading is most at risk of being asserted rather than tested, because no add-back of PHD2 into a fusion has been done.
What the experiment showed1 deeper
Wang et al. found 1:
- PHD2 of MLL1 is an E3 ubiquitin ligase in the presence of the E2 enzyme CDC34.
- The activity is conserved in MLL4's PHD2 but absent from MLL2 and MLL3 - so it is a property of a subset of the family, not a generic PHD-finger feature.
- Mutating PHD2 leads to MLL1 stabilisation, increased transactivation ability, and increased recruitment of MLL1 to target gene loci.
The direction is what matters. PHD2 mutation makes MLL1 more abundant, more active and more present at its targets. It is a self-imposed ceiling, and the fusion protein has no ceiling of this kind.
Why this rescues PHD1/PHD2 from being an afterthought
Chen et al.'s add-back result is routinely summarised as 'PHD3 matters, PHD1 and PHD2 do not', because inserting PHD1+PHD2 into MLL-ENL did not block immortalisation [see ana-phd3-addback-replication-audit]. That is a statement about a serial replating assay, not about PHD2's biochemistry.
PHD2's documented function is to limit MLL1 abundance and target occupancy 1. A replating assay over a few weeks is close to the worst available readout for a change in protein-level set-point, and the insertion construct was tested in a fusion whose stability is, per Hsieh et al., conferred by the partner anyway [see subtraction-kmt2a-fyrn-fyrc-heterodimer]. So 'PHD1+PHD2 had no effect' should be read as 'this assay did not detect an effect', not as 'PHD2 loss is inconsequential'.
This matters practically because the graph's proposed proximity-based modality is built on PHD3. If PHD2 is a second, independent brake, then a proximity agent that restores CyP33/HDAC1 recruitment addresses one of two lost restraints.
The boundary problem, stated honestly
UniProt places PHD1 at 1431-1482 and PHD2 at 1479-1533 2 - overlapping by four residues. That overlap is a curation artefact of profile-based assignment, and it means no experiment can cleanly delete PHD2 without touching PHD1. Wang et al. used point mutation rather than deletion, which sidesteps the problem for their purpose but leaves the graph unable to give a clean 'PHD2 lost, PHD1 retained' coordinate story for any breakpoint.
The practical consequence: PHD1 gets no separate entry in this beat. There is no primary functional study assigning MLL1 PHD1 an independent activity that I could verify, and inventing one to fill a slot in a domain ledger is exactly the failure mode this beat is supposed to correct. PHD1 is lost and its function is unknown. That is the entry.
| module | PHD2 E3 ubiquitin ligase |
|---|---|
| uniprot | Q03164 |
| start | 1479 |
| end | 1533 |
| fate in fusion | lost |
| e2 partner | CDC34 |
| wild type function | negative regulator: PHD2 mutation stabilises MLL1 and increases transactivation and target recruitment |
| paralogue specificity | conserved in MLL4 PHD2; absent from MLL2 and MLL3 |
| demonstrated functional consequence of loss to the fusion | none; no PHD2 add-back into a fusion has been reported, and the one PHD1+PHD2 insertion experiment used a replating readout poorly matched to a stability phenotype |
| loss class | lost, brake documented in wild-type context only |
neighbours — the numbers behind the chart
[
{
"name": "PHD1",
"start": 1431,
"end": 1482,
"note": "overlaps PHD2 by four residues in UniProt annotation; no verified independent function"
},
{
"name": "PHD4",
"start": 1931,
"end": 1978,
"note": "annotated by UniProt; usually omitted from the graph's 'three PHD fingers' phrasing"
},
{
"name": "pre-PHD C2HC",
"start": 1870,
"end": 1910,
"source": "UniProt Q03164"
}
]- Restoring a degradation-promoting module to the fusion sounds attractive but would also, if delivered by any agent engaging shared N-terminal sequence, destabilise wild-type KMT2A - a gene under extreme human loss-of-function constraint.
- If the fusion's stability is set by the partner rather than by KMT2A determinants, then re-installing a KMT2A-side brake may simply be overridden, and the negative 2008 replating result would then be the right answer for the wrong reason.
- Insert PHD2 alone into KMT2A::AFF1 and measure fusion protein half-life and chromatin occupancy rather than colony counts. The 2008 negative result used the wrong readout for this brake, and the right one is cheap.
- Does the retained wild-type KMT2A allele in a KMT2A-rearranged cell still self-limit through PHD2, and does that set a ceiling on wild-type KMT2A that the fusion escapes? If so, the fusion/wild-type stoichiometry at shared target loci is a regulated quantity nobody has measured.
- Is CDC34 or a related E2 a dependency in KMT2A-rearranged cells? If PHD2's E3 activity is lost from the oncogene but retained on the wild-type allele, CDC34 perturbation would act asymmetrically on the two alleles.
- E1strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper
MLL1 contains four highly conserved PHD fingers which are invariably deleted in oncogenic MLL1 fusion proteins; the second PHD finger (PHD2) of MLL1 is an E3 ubiquitin ligase in the presence of the E2-conjugating enzyme CDC34; this activity is conserved in the second PHD finger of MLL4 but not MLL2 or MLL3; and mutation of PHD2 leads to MLL1 stabilisation as well as increased transactivation ability and increased MLL1 recruitment to target gene loci, indicating that PHD2 negatively regulates MLL1 activity.
A subset of mixed lineage leukemia proteins has plant homeodomain (PHD)-mediated E3 ligase activity · Wang J, Muntean AG, Wu L, Hess JL · Journal of Biological Chemistry 287(52):43410-43416 2012What this evidence cannot show
ABSTRACT ONLY. In-vitro ubiquitination assays plus HEK293 cell experiments with mutant constructs; no haematopoietic model, no leukemia endpoint, and no experiment in which PHD2 is restored to a fusion protein. Whether the E3 activity acts in cis on MLL1 itself or in trans on another substrate is not resolved by the abstract, though the stabilisation phenotype is consistent with self-ubiquitination.
Evidence that argues against this (1)
- The one experiment that inserted PHD1+PHD2 into an MLL fusion found no effect on serial-replating immortalisation, in contrast to PHD3. That is a genuine negative result and the main reason the field treats PHD2 loss as unimportant - though a replating assay is poorly suited to detecting a change in protein-level set-point. Chen J et al. 2008
- E2established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.database
UniProt Q03164 annotates PHD-type zinc finger 1 at residues 1431-1482, PHD-type 2 at 1479-1533, PHD-type 3 at 1566-1627, a C2HC pre-PHD-type finger at 1870-1910 and PHD-type 4 at 1931-1978; PHD1 and PHD2 as annotated overlap by four residues.
UniProtKB entry Q03164 (KMT2A_HUMAN) - zinc finger feature table, retrieved via the UniProt REST API · UniProt Consortium · UniProtKB 2026What this evidence cannot show
Accession verified by reading back the primary gene name (KMT2A) and length (3,969 aa). The PHD1/PHD2 overlap is an artefact of profile-based boundary assignment rather than a biological statement; it should not be read as two fingers sharing residues. Note also that UniProt annotates a FOURTH PHD finger at 1931-1978 plus a pre-PHD C2HC finger, which the graph's common phrase 'the three PHD fingers' omits.
Evidence that argues against this (1)
- A UniProt feature table is a curated model rather than a measurement. Profile-based domain calls for KMT2A differ between UniProt, Pfam, SMART, InterPro and CATH, and the sibling atlas beat was corrected last cycle for quoting one source's range as though it were observed. Quoting coordinates without the spread states a choice as a fact. Journal of Biological Chemistry 2012
the same lost cluster carries both the reader function and the E3 activity
[references] PHD2 loss reclassified: disabling it stabilises MLL1 and raises its occupancy
- supports →Hypothesis: what the KMT2A half loses is control of its own abundance, not control of its activity
[supported_by] PHD2 mutation stabilises MLL1 and raises transactivation and recruitment
- part of →KMT2A / MLL1 — the gene that breaks
PHD2 aa 1479-1533, an E3 ligase that limits KMT2A abundance and occupancy; lost in every fusion
- related to →Lost from the KMT2A half: PHD3 and the bromodomain are one cassette, not two modules (aa 1566-1765)
a second, mechanistically independent brake lost in the same break: PHD3 sets direction, PHD2 sets amount
the PHD1+PHD2 negative arm of the 2008 add-back used a readout unsuited to a stability phenotype
- related to →Lost from the KMT2A half: the FYRN-FYRC clamp (2018-2074 / 3666-3747) that holds wild-type KMT2A together
two independent stability determinants lost together; the fusion retains neither
row: PHD2 1479-1533
two independent brakes in the lost region is the strongest point in the hypothesis's favour