MBM2, the second menin-binding motif of KMT2A (residues 23-40): 26-fold weaker than MBM1, never separately co-crystallised, and the only part of the menin grip a drug is not aimed at
also called MBM2 · menin-binding motif 2 · MLL 23-40
KMT2A holds menin mainly with one short motif, but it has a second, much weaker one just downstream. Every menin drug is built against the strong motif, and nobody has shown what the weak one does when the strong one is blocked.
The same thing, for a scientist
Grembecka et al. defined two menin-binding motifs in the KMT2A N-terminus: MBM1 (residues 4-15, core RWRFP 6-10) and MBM2 (residues 23-40). By fluorescence polarisation MBM2 binds menin at Kd 1,400 +/- 424 nM against MBM1's 53 +/- 4.2 nM - a 26-fold difference - and the full 46- and 160-residue fragments bind at 6.8 and 9.8 nM, i.e. roughly an order of magnitude tighter than MBM1 alone, which is the quantitative signature of a second contributing contact. Huang et al., describing the same region structurally, define a single 'MLL1 MBM' spanning residues 6-25 that binds at 82 nM by ITC, report the C-terminal pair Arg24/Arg25 as disordered in the crystal, and show that an R24E/R25E substitution costs 21-fold in affinity. So the two literatures disagree about whether this is one long motif or two, and the region beyond residue 23 is present in the 3U88 construct but not demonstrated to be ordered anywhere.
Two descriptions of the same stretch of chain
| source | what it calls the motif | span | affinity |
|---|---|---|---|
| Grembecka 2010 1 | MBM1 + MBM2, two motifs | 4-15 and 23-40 | 53 nM (FP) and 1,400 nM (FP) |
| Huang 2012 2 | one MLL1 MBM | 6-25 | 82 nM (ITC) |
They are not contradictory so much as differently bounded, and the arithmetic reconciles them: MBM1 alone is 53-72 nM, MLL 1-46 is 6.8 nM and MLL 1-160 is 9.8 nM 1. Something between residue 15 and residue 46 is worth roughly a log of affinity, and both papers put it at 23-25. Huang et al. measure that contribution directly - R24E/R25E costs 21-fold 2.
The practical consequence is that MBM1 affinity is not the affinity a drug has to beat. A clinical menin inhibitor competes with the intact N-terminal fragment of the fusion, which binds about eightfold tighter than the MBM1 peptide used in most competition assays.
Is MBM2 resolved in any structure?
The 3U88 MLL1 chain is an engineered 75-residue construct whose sequence, aligned to the fetched UniProt Q03164 sequence in this cycle, maps to 5-15 + 23-39 + 109-153 4. So the MBM2 span is present in the crystallised construct. That is not the same as being resolved: Huang et al. state that Arg24/Arg25 are disordered in the MBM complex and visible only in the equivalent JunD structure 2. No deposit in the PDB is annotated as a menin-MBM2 complex, and no separate MBM2 peptide co-crystal exists.
So MBM2 is the one arm of the menin clamp with a measured affinity and no structure. That is the reverse of the usual gap in this graph and it is why it deserves its own entry.
Why this matters for menin inhibitors
Every approved and experimental menin inhibitor is a ligand for the MBM1 pocket. Nothing published shows whether a saturating MBM1-site inhibitor also displaces the MBM2 contact, or whether MBM2 leaves a residual tether. The graph already records residual occupancy under saturating menin inhibition as an open experiment (exp-residual-occupancy-cutrun-saturating-menin); MBM2 is a concrete, structurally specified candidate mechanism for it.
It is also retained: residues 23-40 sit roughly 1,360 residues N-terminal to the breakpoint cluster region, so the fusion keeps MBM2 exactly as it keeps MBM1.
| partner a | gene-men1 |
|---|---|
| partner b | gene-kmt2a |
| motif b | MBM2, KMT2A residues 23-40 (Grembecka parsing); the same chain region is covered by residues 6-25 in the Huang single-MBM parsing |
| motif a | the same central cavity of the menin TPR core that accepts MBM1 and JunD |
| kd NM | 1400 |
| fate in fusion | retained - residues 23-40 lie about 1,360 residues N-terminal to the breakpoint cluster region |
| ligand note | all three are MBM1-pocket ligands; none is reported to engage or displace MBM2 |
| druggability | not separately assessed; MBM2 has an affinity but no structure, which is the inverse of the usual gap in this graph |
| resolution note | PDB entity sequences fetched from data.rcsb.org 2026-09-20 and aligned to the UniProt Q03164 sequence fetched the same day. |
interface residues b
[ "KMT2A R24", "KMT2A R25" ]
kd measurements — the numbers behind the chart
[
{
"ligand": "KMT2A 23-40 (MBM2)",
"kd_nM": 1400,
"error_nM": 424,
"method": "fluorescence polarisation",
"source_pmid": "20961854"
},
{
"ligand": "KMT2A 4-15 (MBM1, comparator)",
"kd_nM": 53,
"error_nM": 4.2,
"method": "fluorescence polarisation",
"source_pmid": "20961854"
},
{
"ligand": "KMT2A 1-46",
"kd_nM": 6.8,
"error_nM": 1.7,
"method": "ITC",
"source_pmid": "20961854"
},
{
"ligand": "KMT2A 1-160",
"kd_nM": 9.8,
"error_nM": 2.5,
"method": "ITC",
"source_pmid": "20961854"
},
{
"ligand": "KMT2A 6-25 (single-MBM parsing)",
"kd_nM": 82,
"method": "ITC",
"source_pmid": "22327296"
},
{
"ligand": "KMT2B/MLL2 16-35",
"kd_nM": 120,
"method": "ITC",
"source_pmid": "22327296"
}
]structures
[ "3U88" ]
structures absent
[ "no PDB deposit is ANNOTATED as a menin-MBM2 complex - which is a metadata statement, not a structural one, and it misled this entry", "no separate MBM2 peptide co-crystal exists" ]
structure details
[
{
"pdb": "3U88",
"title_fetched": "Crystal structure of human menin in complex with MLL1 and LEDGF",
"method": "X-ray",
"resolution_A": 3,
"kmt2a_construct_maps_to": "5-15 + 23-39 + 109-153",
"fetched": "2026-09-20"
}
]ligands
[ "drug-revumenib", "drug-ziftomenib", "drug-bleximenib" ]
menin mbm1 kd conditions
{
"_why": "Added by king sweep 2026-09-21 (Q82 corrections-owed #3). This entry quotes the menin-MBM1 53 nM figure. That number is NOT comparable to the KMT2A CXXC-DNA affinities without stating its buffer, and the omission is how a 1,300-fold 'discrepancy' was manufactured.",
"value": "53 nM",
"motif": "KMT2A MBM1, residues 4-15, FITC-labelled peptide at 15 nM",
"method": "fluorescence polarisation",
"buffer": "50 mM Tris, pH 7.5, 50 mM NaCl, 1 mM DTT",
"nacl_mM": 50,
"companion_itc": "72 nM by ITC, 50 mM phosphate pH 7.5, 50 mM NaCl, 1 mM beta-mercaptoethanol, 27 C",
"source_pmid": "20961854",
"source_quote": "\"in the FP buffer (50 mM Tris, pH 7.5, 50 mM NaCl, 1 mM DTT) used for binding experiments (Kd measurements)\" -- Grembecka 2010 JBC, Methods, fetched and read from PMC3003368 on 2026-09-21",
"do_not_compare_naively": {
"against": "KMT2A CXXC-DNA Kd 44 uM (PMID 20010842, Cierpicki 2010)",
"because": "that measurement was made at 300 mM NaCl -- verbatim from its Methods, \"50 mM phosphate buffer (pH 7.1), 300 mM NaCl, 2 mM DTT at 25 C\" (PMC2908503, read 2026-09-21) -- SIX-FOLD more salt, on a salt-sensitive protein-DNA contact, against a hydrophobic peptide pocket that is not.",
"best_matched_pair": "44 uM at 300 mM NaCl vs the 5.7 uM tagless CXXC measurement at 150 mM NaCl gives 108-fold against menin, not ~1,000-fold",
"see": "ana-the-menin-affinity-comparison-was-made-at-one-sixth-the-salt"
},
"added": "2026-09-21"
}cxxc dna kd conditions
{
"_why": "Added by king sweep 2026-09-21 (cycle 46). This entry quotes the KMT2A CXXC-DNA 44 uM. That number is a property of five choices, not of the molecule, and quoting it bare is how this graph manufactured a 1,300-fold 'discrepancy' with the menin contact.",
"value": "44 uM",
"construct": "isolated CXXC domain, KMT2A 1147-1203, untagged, 15N-labelled",
"ligand": "DNA pal oligonucleotide, one CpG element",
"method": "NMR titration",
"buffer": "50 mM phosphate, pH 7.1, 300 mM NaCl, 2 mM DTT, 25 C",
"nacl_mM": 300,
"source_pmid": "20010842",
"source_quote": "\"Determination of the dissociation constants by NMR for the CXXC domain (MLL 1147-1203) and mutants was performed by titration of 15N-labeled protein with concentrated solutions of DNA pal oligonucleotide. Experiments were carried out in 50 mM phosphate buffer (pH 7.1), 300 mM NaCl, 2 mM DTT at 25 C.\" -- Cierpicki 2010 NSMB, Methods, fetched and read from PMC2908503 on 2026-09-21. The result sentence reads \"The wild type protein yielded a Kd of 44 uM\".",
"this_is_the_high_salt_end": "300 mM NaCl is roughly twice physiological ionic strength and SIX TIMES the 50 mM the menin-MBM1 53 nM was measured in. The same domain measured tagless at 150 mM NaCl gives 5.7 uM -- 7.7-fold tighter from salt alone.",
"do_not_compare_naively": {
"against": "menin-MBM1 53 nM (PMID 20961854), measured at 50 mM NaCl by FP",
"best_matched_pair": "108-fold, not ~1,000-fold",
"see": "ana-the-menin-affinity-comparison-was-made-at-one-sixth-the-salt"
},
"added": "2026-09-21"
}CORRECTION cycle 63 MBM2 IS RESOLVED IN 3U88
{
"what_this_entry_said": "that Arg24/Arg25 are reported disordered in the menin-MBM crystal, and by implication that MBM2 has never been seen",
"what_is_true": "that is a statement about 3U85, where the KMT2A peptide is modelled at 5-13 only. In 3U88 the KMT2A chains M and N are modelled at 6-15, 23-39 and 107-135 - so residues 23-39, which include MBM2, ARE resolved.",
"the_contacts": "verified by the king from 3U88 coordinates: KMT2A Arg24 is 3.2 A from menin Asp370 in chain pair M/A and 2.9 A in N/B, and 3.0 and 3.5 A from menin Glu366.",
"why_it_matters": "Glu366 and Asp370 are the same residues that line the drug pocket and that UniProt curates as reducing KMT2A binding when mutated. So MBM2's anchor and the inhibitor site are the same acidic patch, which is a structural fact this entry was recording as absent.",
"reported_by": "Q34; the modelled ranges and every distance re-computed by the king"
}- Residual tethering: a second, weaker menin contact is the simplest structural explanation for incomplete fusion eviction under a saturating MBM1-site inhibitor, and it cannot be mutated away by the tumour because it is not the drug's binding site.
- Potency mis-specification: competition assays that use an MBM1 peptide report an IC50 against a ligand roughly eightfold weaker than the physiological fragment, so cellular potency can lag biochemical potency for a purely thermodynamic reason.
- Selectivity: KMT2B carries an almost identical motif binding menin at 120 nM, so any MBM-directed agent engages the paralogue too.
- Delivery: MBM2 is unstructured, so it is a poor small-molecule target in its own right and would need a different modality entirely.
- Does a saturating clinical menin inhibitor displace the MBM2 contact, or does MBM2 leave a residual tether that would show up as incomplete eviction in CUT&RUN?
- Is the 8-fold gap between MBM1 peptide affinity (53-72 nM) and MLL 1-46 affinity (6.8 nM) entirely attributable to residues 23-40, or is some of it avidity from the intrinsically disordered linker?
- ANSWERED IN CYCLE 63: yes. 3U88 models KMT2A 23-39, so MBM2 is ordered in that deposit; the disorder statement belonged to 3U85, whose peptide is modelled at 5-13 only. What remains open is the original second half - whether menin inhibitor potency against the intact fragment is the only meaningful measure.
- E1established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.paper
KMT2A binds menin through two motifs, MBM1 (4-15) and MBM2 (23-40); MBM2 binds at Kd 1,400 nM by fluorescence polarisation against MBM1's 53 nM, while MLL 1-46 and MLL 1-160 bind at 6.8 and 9.8 nM by ITC.
Molecular basis of the mixed lineage leukemia-menin interaction: implications for targeting mixed lineage leukemias. · Grembecka J, Belcher AM, Hartley T, Cierpicki T · J Biol Chem 2010What this evidence cannot show
FP and ITC on different constructs in different buffers; the MBM2 number is FP-only with a 30% standard error and no calorimetric confirmation. All measurements use isolated KMT2A fragments and bacterially expressed menin, not a fusion protein. Full text reached by the PMC HTML route (46,394 characters) after fullTextXML returned HTTP 500.
Evidence that argues against this (1)
- Huang et al. treat the same region as a single motif spanning 6-25 rather than two motifs and do not report an independent MBM2 affinity, so the two-motif description is one group's parsing rather than a settled fact. Huang J et al. 2012
- E2established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.paper
MLL1 residues 6-25 are necessary and sufficient for menin binding at Kd 82 nM by ITC; Arg24/Arg25 are disordered in the crystal yet an R24E/R25E substitution reduces affinity 21-fold.
The same pocket in menin binds both MLL and JUND but has opposite effects on transcription. · Huang J, Gurung B, Wan B, et al. · Nature 2012What this evidence cannot show
A disordered-but-functionally-important pair is exactly the case where a crystal structure under-reports an interface; the 21-fold figure is from a charge-reversal double mutant, which is a harsher perturbation than alanine substitution.
Evidence that argues against this (1)
- Grembecka et al. measure the 23-40 fragment as an independently binding motif with its own dissociation constant, which a 6-25 boundary cannot accommodate. Grembecka J et al. 2010
- E3established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.database
PDB 3U85 is a 3.0 A crystal structure of menin with a 21-residue MLL1 peptide beginning SRWRFPARPGT.
PDB 3U85: Crystal structure of human menin in complex with MLL1 · Huang J, Gurung B, Wan B, et al. · RCSB Protein Data Bank 2012What this evidence cannot show
Title, method, resolution and entity sequences fetched from data.rcsb.org on 2026-09-20. The deposited peptide covers MBM1 and the polyglycine linker; it does not extend through MBM2.
Evidence that argues against this (1)
- The 1.55 A menin-MLL peptide structure 4GQ6 uses a still shorter 12-residue peptide, so the higher-resolution deposit contains even less of the region in question - resolution and coverage move in opposite directions here. Huang J et al. 2012
- E4strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.database
The MLL1 chain deposited in 3U88 is an engineered construct whose sequence maps to Q03164 residues 5-15, 23-39 and 109-153, so it contains the MBM2 span.
PDB 3U88: Crystal structure of human menin in complex with MLL1 and LEDGF · Huang J, Gurung B, Wan B, et al. · RCSB Protein Data Bank 2012What this evidence cannot show
This is an alignment of the fetched RCSB entity sequence to the fetched UniProt Q03164 sequence performed in this cycle; presence in the construct does not establish presence in the electron density, and the deposit's own residue numbering is not a continuous natural fragment.
Evidence that argues against this (1)
- Murai et al. state that this crystal structure contains no MLL contacts upstream of residue 135 on the LEDGF side; a construct that splices three separated segments must be read segment by segment, not as a fragment. Murai MJ et al. 2014
MBM2 binds the same menin cavity at 1,400 nM
KMT2A residues 23-40
the weak second motif of the same grip, 26-fold lower affinity
- supports →Experiment: map what stays on chromatin when the menin inhibitor has done everything it can do
a concrete structural candidate for residual occupancy under saturating inhibitor
retained in the index fusion
JunD's equivalent C-terminal basic pair is ordered where KMT2A's Arg24/Arg25 are not
MBM2 residues contact menin in both copies; KMT2B identity across MBM1 and MBM2
supplies the menin counter-surface for MBM2 and shows KMT2A 23-39 is ordered in 3U88
refutes the field stating that MBM2 is unresolved: KMT2A 23-39 is modelled in 3U88 and R24 contacts menin D370/E366
same record set; the menin-MLL contact is the one load-bearing claim the census failed to find a weakness in