Michael Cleary (Stanford) & Akihiko Yokoyama (National Cancer Center, Tsuruoka) — the groups that discovered the menin requirement itself
also called Cleary Lab · Yokoyama Lab · Stanford Pathology · Tsuruoka Metabolomics Laboratory · AEP / MLL transcriptional activation system
In 2005 this pair showed that MLL fusion proteins must physically grip menin in order to cause leukemia at all. Every menin inhibitor in the clinic today exists because of that single result.
The same thing, for a scientist
Akihiko Yokoyama, working with Michael L. Cleary in the Department of Pathology at Stanford University School of Medicine, demonstrated that oncogenic MLL fusion proteins retain a high-affinity, amino-terminal, conserved menin-binding motif, that this interaction is required for the initiation of MLL-mediated leukemogenesis, and that menin is required for maintenance of MLL-associated but not other oncogene-induced myeloid transformation. They then showed menin acts as a molecular adaptor physically linking MLL to LEDGF/PSIP1, which is itself required for MLL-dependent transcription and transformation. Yokoyama now leads the Tsuruoka Metabolomics Laboratory at the National Cancer Center, Japan, where the programme has become the AEP / MLL transcriptional activation system model of aberrant self-renewal.
Why this node matters more than its citation count suggests1 deeper
The Iqbal Lab's central hypothesis is that obligate partner interfaces are druggable where the fusion itself is not. The 2005 Cell paper is the founding empirical statement of that hypothesis for one specific case 1. Reading it precisely matters, because the paper says three separable things, and the field routinely conflates them:
- MLL fusions retain the menin-binding motif (a structural claim).
- The interaction is required for initiation of leukemogenesis (a genetic claim about establishment).
- Menin is required for maintenance of MLL-associated, but not other, myeloid transformation (a genetic claim about an established tumour, and the one that makes a drug plausible).
Only the third supports treating a diagnosed patient. It is also the weakest of the three in terms of how many independent groups have reproduced it under drug-like partial-occupancy conditions rather than genetic ablation.
The 2008 follow-up added the piece the lab should care about most: menin is an adaptor, not an endpoint 2. It physically links MLL to LEDGF/PSIP1, and LEDGF is itself required for MLL-dependent transcription and transformation. That means there is a second interface in the same chain, and a subset of MEN1 mutations from MEN1 syndrome patients abrogate LEDGF binding while preserving MLL binding - a natural separation-of-function experiment already performed by human genetics.
Partnering angle, and the awkward truth about it
Both are senior, historically central, and not obviously in a position to need a small lab. Cleary's foundational work is two decades old; Yokoyama's current programme at Tsuruoka is a mechanism programme focused on the MLL/AEP transcriptional activation system and self-renewal 3, and he now co-writes the field's synthesis reviews with Patricia Ernst 4.
The honest assessment: this is intellectual-lineage territory rather than a partnership opportunity. What the lab should take from them is the precise formulation of the hypothesis and the separation-of-function logic, not an approach for reagents.
There is one real opening. The PSIP1/LEDGF arm of the chain is comparatively under-drugged relative to menin, and the naturally occurring MEN1 mutations that separate LEDGF binding from MLL binding 2 are an existing, free, human-genetic reagent for testing whether a second interface in the same complex is independently required. Nobody appears to have systematically revisited those variants in the menin-inhibitor era.
| group | Michael L. Cleary (Stanford); Akihiko Yokoyama (independent since; now National Cancer Center Japan) |
|---|---|
| institution | Department of Pathology, Stanford University School of Medicine, Stanford CA, USA (2005-2008 papers); Tsuruoka Metabolomics Laboratory, National Cancer Center, Tsuruoka, Yamagata, Japan (Yokoyama, current) |
| focus | Discovery of the menin requirement for MLL-fusion leukemogenesis; menin as a molecular adaptor to LEDGF/PSIP1; MLL/AEP transcriptional activation system and aberrant self-renewal |
| partnering angle | Intellectual lineage rather than a reagent partnership. The exploitable opening is the PSIP1/LEDGF arm: the 2008 paper identified naturally occurring MEN1 variants that separate LEDGF binding from MLL binding, a free human-genetic separation-of-function reagent that nobody appears to have revisited in the menin-inhibitor era. Yokoyama, as an active mid-career PI in Japan, is far more approachable than the Stanford legacy line. |
| relationship | foundational, not competing; low-probability but high-prestige collaboration |
key papers — the numbers behind the chart
[
{
"title": "The menin tumor suppressor protein is an essential oncogenic cofactor for MLL-associated leukemogenesis",
"year": 2005,
"venue": "Cell",
"pmid": "16239140",
"doi": "10.1016/j.cell.2005.09.025"
},
{
"title": "Menin critically links MLL proteins with LEDGF on cancer-associated target genes.",
"year": 2008,
"venue": "Cancer Cell",
"pmid": "18598942",
"doi": "10.1016/j.ccr.2008.05.003"
},
{
"title": "Leukemogenesis via aberrant self-renewal by the MLL/AEP-mediated transcriptional activation system",
"year": 2021,
"venue": "Cancer Science",
"pmid": "34251718",
"doi": "10.1111/cas.15054"
},
{
"title": "KMT2A-rearranged leukemia: from mechanism to drug development",
"year": 2025,
"venue": "Experimental Hematology",
"pmid": "40921261",
"doi": "10.1016/j.exphem.2025.105247"
}
]leverage they hold
[ "definitional authority on the menin requirement", "the LEDGF/PSIP1 adaptor model and its separation-of-function variants", "the AEP transcriptional-activation framework" ]
- The founding result is genetic ablation, not pharmacological inhibition; a drug that occupies the pocket partially may never reproduce the knockout phenotype, and the gap between the two is where clinical disappointment lives.
- If menin's essential role is as an adaptor, then any drug blocking one arm of the adaptor may be bypassed by the other, and the MEN1 resistance mutations that preserve MLL1 binding are the clinical proof of principle for exactly that.
- LEDGF/PSIP1 is a general chromatin tether with functions well beyond MLL, so targeting it trades selectivity for durability.
- The 2005 maintenance result rests on acute genetic ablation. Has anyone shown that partial, reversible, drug-like occupancy of the menin pocket produces the same maintenance phenotype in KMT2A::AFF1 infant disease, rather than in the AML models the original work used?
- Do the MEN1 variants that separate LEDGF binding from MLL binding alter menin-inhibitor sensitivity? This is a cheap, informative experiment that uses reagents described in 2008 and still appears unexamined.
- If menin is an adaptor rather than an endpoint, is the PSIP1/LEDGF interface a better target than the menin pocket, given that resistance mutations in MEN1 spare the MLL1 interaction?
- E1established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.paper
Oncogenic MLL fusion proteins stably associate with menin through a high-affinity, amino-terminal, conserved binding motif; this interaction is required for initiation of MLL-mediated leukemogenesis, and menin is essential for maintenance of MLL-associated but not other oncogene-induced myeloid transformation, with acute genetic ablation of menin reversing aberrant Hox gene expression.
The menin tumor suppressor protein is an essential oncogenic cofactor for MLL-associated leukemogenesis · Yokoyama A, Somervaille TCP, Smith KS, Rozenblatt-Rosen O, Meyerson M, Cleary ML · Cell 2005What this evidence cannot show
Murine transformation models with complete genetic ablation of menin, which is a much more severe perturbation than partial pharmacological occupancy; the maintenance claim is the therapeutically relevant one and rests on acute genetic deletion rather than on drug-like inhibition. The specificity control ('not other oncogene-induced transformation') covers a limited set of comparator oncogenes.
Evidence that argues against this (1)
- Genetic essentiality and pharmacological dependency are not the same claim, and the clinical record separates them. Menin ablation reverses the MLL-fusion program in models, but disrupting the menin-MLL interface with a drug produces CR+CRh in only about a quarter of patients (pooled CR+CRh 28.5% across 14 studies and 784 treated adults), and PRIMARY, pre-existing resistance to menin inhibitors is demonstrable in MLL-fusion AML models and is overcome only by adding KAT6/KAT7 catalytic inhibition, with KAT7 remaining a dependency in both genetic and non-genetic acquired resistance. The 2005 result establishes that the interaction is required for leukemogenesis; it does not establish that blocking it is sufficient to collapse an established leukemia, and that gap is where the field's clinical disappointment lives. Gordon SJV et al. 2025
- E2strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper
Menin functions as a molecular adaptor physically linking MLL to LEDGF/PSIP1; LEDGF is required for both MLL-dependent transcription and leukemic transformation, and a subset of menin mutations found in MEN1 syndrome patients abrogate the LEDGF interaction while preserving the MLL interaction yet still compromise MLL/menin-dependent function.
Menin critically links MLL proteins with LEDGF on cancer-associated target genes. · Yokoyama A, Cleary ML · Cancer Cell 2008What this evidence cannot show
The separation-of-function MEN1 variants were tested in experimental systems rather than in patient leukaemia; whether the same variants would confer or prevent menin-inhibitor response in a human leukaemia has not been tested. LEDGF/PSIP1 is a broadly used chromatin tether (it is the HIV-1 integration cofactor), so its requirement is less selective than menin's. No study contradicting the menin-LEDGF adaptor model was located; the closest independent test, a conditional Psip1/LEDGF knockout, supports rather than disputes it by showing LEDGF/p75 is dispensable for normal haematopoiesis but essential for MLL-rearranged leukemogenesis (PMID 29084774, Blood 2018). The claim therefore stands on two concordant reports from partly overlapping intellectual lineages rather than on an adversarial test, and the specific sub-claim - that MEN1-syndrome mutations can abrogate the LEDGF interaction while preserving the MLL interaction - has not been re-examined structurally since 2008.
Evidence that argues against this (1)
- Menin linking MLL to LEDGF is the mechanistic foundation of this project's central hypothesis, and the entry files it as a lab credential. The substantive limitation is that LEDGF is required for MLL-dependent transcription generally, not only for the fusion, so the interface is shared with the wild-type protein — which is the therapeutic-window problem this graph keeps returning to. Cancer Cell 2008
- E3moderate confidence 3 of 5: One strong study, or several smaller consistent ones.review
Yokoyama's current programme, based at the Tsuruoka Metabolomics Laboratory of the National Cancer Center in Japan, frames MLL leukaemogenesis as aberrant self-renewal conferred on non-stem hematopoietic progenitors by constitutive activation of CpG-rich promoters through an MLL/AEP-mediated transcriptional activation system.
Leukemogenesis via aberrant self-renewal by the MLL/AEP-mediated transcriptional activation system · Yokoyama A · Cancer Science 2021What this evidence cannot show
A single-author review presenting the group's model; the AEP-centric framing is one of several competing accounts of how MLL fusions activate transcription and should be cited as a position rather than a settled fact.
Evidence that argues against this (1)
- A current-programme description sourced to one review goes stale silently and cannot characterise an active research direction. The AEP/self-renewal framing is also one model among several here: this graph separately holds a forced-oligomerisation account that explains partner classes AEP cannot. Cancer Science 2021
- E4moderate confidence 3 of 5: One strong study, or several smaller consistent ones.review
Yokoyama and Patricia Ernst (University of Colorado) jointly authored the field's current mechanism-to-drug synthesis, stating that a uniform molecular dependency of MLL fusion oncoproteins underlies the therapeutic opportunity.
KMT2A-rearranged leukemia: from mechanism to drug development · Ernst P, ... Yokoyama A · Experimental Hematology 2025What this evidence cannot show
A review; useful as a map of where the field believes it stands in 2025 and as evidence that these two groups now write together, but not a source of primary data.
Evidence that argues against this (1)
- A jointly authored mechanism-to-drug synthesis is a review, and reviews in this graph are the second-worst-opposed citation class at 48% unopposed. It is a good guide to consensus and a poor source for any specific claim, which should be taken from the primary work it cites. Experimental Hematology 2025
identified menin as the oncogenic cofactor and LEDGF as its partner
discovered the high-affinity N-terminal menin-binding motif retained by MLL fusions and its requirement for leukemogenesis
showed menin is a molecular adaptor linking MLL to LEDGF/PSIP1, with separation-of-function MEN1 variants
founding demonstration that menin is an essential oncogenic cofactor for MLL-associated leukemogenesis
co-author with Yokoyama on the field's current mechanism-to-drug synthesis