One immunoprecipitation that decides whether 14-3-3 touches KMT2A at all
also called 14-3-3 bridge test · AFF depletion pulldown
Pull down 14-3-3 from leukaemia cells and see whether KMT2A comes with it. Then remove the partner protein AF4 and do it again. If KMT2A stops coming, it was never holding on itself.
The same thing, for a scientist
A three-arm co-immunoprecipitation in a t(4;11) line (SEM or RS4;11) and a KMT2A-wild-type control: (1) anti-YWHAQ and anti-YWHAG IP, blot for KMT2A, KMT2A fusion, AFF1 and AFF4; (2) the same after AFF1+AFF4 double knockdown; (3) the same after lambda-phosphatase treatment of the lysate. **Arm 2 discriminates the bridged model from a direct contact; arm 3 discriminates a phospho-dependent contact from a co-purification artefact.** A fourth arm, AFF1 S588A knock-in, converts a positive into a residue-level result. Total cost is one antibody set and two lentiviral knockdowns; the readout is a western blot. This is the cheapest decisive experiment anywhere in this beat, and the question it settles - whether the retained KMT2A fragment has a phospho-gated non-menin partner - is the one the beat exists for.
Design
| arm | condition | KMT2A recovered? | interpretation |
|---|---|---|---|
| 1 | control | yes | reproduces the interactome observation in a leukaemic cell |
| 1 | control | no | the interactome recovery does not hold in this cell type; stop here |
| 2 | AFF1+AFF4 knockdown | lost | bridged - no 14-3-3 handle on KMT2A |
| 2 | AFF1+AFF4 knockdown | retained | direct - a KMT2A site exists and is unmapped |
| 3 | lambda-phosphatase | lost | phospho-dependent, consistent with a genuine 14-3-3 contact |
| 3 | lambda-phosphatase | retained | likely a co-purification artefact |
| 4 | AFF1 S588A | reduced | the bridge is that residue |
Controls that matter: IgG IP; blot the reciprocal direction (anti-KMT2A IP, blot 14-3-3); confirm AFF knockdown does not destabilise KMT2A itself, because a loss of recovery from a loss of bait is not a loss of interaction.
Why it is worth doing before anything else in this beat
Every other lead here needs a bigger experiment. The CK2-LEDGF phospho-switch needs CUT&RUN under inhibitor at two doses; the SCF(Skp2) question needs turnover measurements across three fusions; the MEF2C S222 biomarker needs patient samples.
This one needs a lysate and antibodies that already exist, and it can return any of four informative answers including the one that closes the question. If arm 2 shows bridging, the graph can retire the 14-3-3 lead and the cycle-28 menin-only reading of the retained fragment survives on this axis - which is a real result, not a null.
| design | Co-immunoprecipitation in SEM or RS4;11 (t(4;11)) plus a KMT2A-wild-type leukaemia control; anti-YWHAQ and anti-YWHAG IP with IgG control and reciprocal anti-KMT2A IP; arms are untreated, AFF1+AFF4 double knockdown, lambda-phosphatase-treated lysate, and AFF1 S588A knock-in |
|---|---|
| readout | Western blot for KMT2A (N-terminal antibody, which detects both wild-type and fusion), fusion-specific blot, AFF1, AFF4, 14-3-3 paralogues |
| controls | IgG IP; reciprocal IP; verification that AFF knockdown does not reduce KMT2A abundance; phosphatase activity control on a known phospho-target |
| cost tier | low |
| timeline | 6-8 weeks including knockdown line generation |
tests hypotheses
[ "hyp-kmt2a-14-3-3-association-is-bridged-by-aff-family" ]
- A negative in arm 1 would not clear the question: co-immunoprecipitation is less sensitive than the proximity methods that generated the original observation, so failing to recover KMT2A would be ambiguous between no interaction and an interaction too transient for IP.
- AFF1 and AFF4 knockdown perturbs transcription globally, so a loss of recovery could reflect a changed cell state rather than a lost bridge. The S588A arm is the one that avoids this and it is also the slowest to build.
- Which KMT2A antibody detects the retained fragment reliably in a fusion-expressing line? A C-terminal antibody would miss the fusion entirely and produce a false bridged-model result.
- Is AFF1+AFF4 double knockdown survivable long enough to run the IP? Both are elongation scaffolds and the double may be lethal, in which case a degron system is needed.
- Should the phosphatase arm use a phosphatase-inhibitor-supplemented control lysate as well, to test the converse?
- E1strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper
KMT2A is recovered as prey of six 14-3-3 family members in a curated interactome of all seven human paralogues, the study that assigns the family a chaperone-like role in which loss of 14-3-3 binding causes clients to coalesce into foci.
A central chaperone-like role for 14-3-3 proteins in human cells. · Segal D, et al. · Mol Cell 2023What this evidence cannot show
Abstract read 2026-09-19; this beat did not re-derive the KMT2A rows from the deposition and inherits that reading from the graph's proximity entry. Interactome recovery is not evidence of a direct contact, which is the premise of the experiment rather than a weakness of it.
Evidence that argues against this (1)
- If the family's principal role is chaperone-like prevention of client aggregation, a large and promiscuous client list is expected, and KMT2A's presence on it carries less information about a specific regulatory contact than a targeted interaction study would. Segal D et al. 2023
- E2moderate confidence 3 of 5: One strong study, or several smaller consistent ones.paper
14-3-3 theta binds AF4 at Ser588 and forms a complex with MLL-AF4 in t(4;11)-positive RS4;11 and MV4-11 cells, which are the cell lines this experiment would use.
Crosstalk between 14-3-3θ and AF4 enhances MLL-AF4 activity and promotes leukemia cell proliferation. · Fioretti T, et al. · Cell Oncol (Dordr) 2019What this evidence cannot show
Abstract-level. The paper does not report an S588A binding test, which is why arm 4 is part of this design rather than a citation.
Evidence that argues against this (1)
- The same paper reports that 14-3-3 theta knockdown impairs chromatin recruitment of AF4 but not of MLL-AF4, so the fusion's relationship to 14-3-3 theta may be indirect even on the partner side, and arm 2 could return an uninterpretable intermediate result. Fioretti T et al. 2019
AFF1/AFF4 depletion arm discriminates bridged from direct
run in SEM or RS4;11