Is TIFAB a direct RelB target? Two papers in the same MLL-AF9 system answer differently
also called TIFAB RelB contradiction · NIK TIFAB direct target
Two studies of the same mouse leukaemia disagree about whether one protein switches another one off directly. A 2021 paper says RelB acts on TIFAB as a direct target; the earlier 2018 study from the same system tested which genes RelB physically sits on and reported TIFAB was not among them. Neither paper mentions the other on this point, and the disagreement matters because a whole chain of reasoning about how this pathway suppresses leukaemia hangs on that one link being direct.
The same thing, for a scientist
**PMID 34877491 (iScience 2021) states in its abstract that 'RelB suppresses its direct target, TIFAB'.** **PMID 29320732 (Cell Reports 2018), the companion study in the same MLL-AF9 NIK-stabilisation system, names Dnmt3a and Mef2c as the RelB-responsive genes and does not mention TIFAB at all**; this graph's `gene-tifab` E3 records that its chromatin immunoprecipitation found RelB at the Mef2c and Dnmt3a promoters **but not** at the Tifab promoter. The two cannot both be right as stated. Surfaced while rewriting counter-notes, not by the contradiction checker.
What each paper says
| PMID 29320732 (2018) | PMID 34877491 (2021) | |
|---|---|---|
| system | MLL-AF9, NIK stabilisation | MLL-AF9, RelB deletion |
| RelB-responsive genes named | Dnmt3a (up), Mef2c (down) | TIFAB |
| TIFAB in the abstract | not mentioned | 'RelB suppresses its direct target, TIFAB' |
| ChIP at the Tifab promoter | recorded in this graph as negative | directness asserted |
The 2021 claim is verbatim and unambiguous - I fetched and read the abstract back on 2026-09-20. The 2018 side is weaker as evidence here, because the negative ChIP result is recorded in this graph's gene-tifab E3 from that study's data and I did not re-verify it from the full text; the abstract simply does not mention TIFAB, which is consistent with the negative but does not establish it.
Neither paper cites the other on this point, so this is not a dispute the field has had - it is two claims sitting side by side.
Why it matters more than a promoter detail
The chain this graph carries runs: stabilising NIK drives non-canonical signalling → RelB → suppression of TIFAB → loss of HOXA9 support → MLL-AF9 leukaemia is suppressed. That chain is the basis for treating the non-canonical arm as therapeutically opposite to the canonical one, which is one of the sharper claims in the NF-kB beat.
If the RelB→TIFAB step is indirect, the chain still runs but stops being a mechanism - it becomes a correlation between two things that both move when NIK is stabilised, and any number of intermediates could sit between them. Directness is what makes it a target rather than a readout.
There is also a live reason to care beyond this beat: TIFAB has since been given a metabolic role in KMT2A::MLLT3 AML through HNF4A (PMID 39626355), so TIFAB is being built on from two directions at once while the wiring immediately upstream of it is unsettled.
What would settle it
Read both full texts side by side. Neither was read in full for this entry, and the 2018 ChIP figure is the specific thing to look at: which promoter regions were amplified, what the antibody was, and whether a Tifab amplicon was included at all. A promoter not assayed is not a promoter shown to be unbound, and that distinction decides whether this is a contradiction or a gap.
If both papers did assay it, the cheap discriminator is a RelB ChIP-seq in the same cells rather than two targeted ChIPs, because targeted ChIP at a handful of promoters is exactly the design in which a real site is missed by primer placement.
- The non-canonical arm is being discussed as therapeutically opposite to the canonical one, and this is one of the links that argument runs through. If the RelB-TIFAB step is indirect, the argument survives as a correlation but stops identifying a target.
- Nothing here is close to a patient. The caution is editorial: this graph should not promote 'direct target' from an abstract phrase into a mechanism, which is how a compression becomes a fact two entries later.
- Did the 2018 ChIP assay a Tifab amplicon at all? A promoter not assayed is not a promoter shown to be unbound, and that one fact decides whether this is a contradiction or a gap.
- Is there a RelB ChIP-seq in MLL-AF9 cells that would settle it without relying on primer placement?
- Does any of this hold in B-lineage KMT2A::AFF1, where the whole NF-kB beat is myeloid?
- E1established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.paper
A 2021 study of RelB deletion in MLL-AF9 AML states that deletion of RelB prevents the inhibitory effect of NIK stabilisation and that, mechanistically, RelB suppresses its direct target TIFAB, which is upregulated in human AML and correlates negatively with patient survival; forced expression of TIFAB reverses the NIK-induced impairment of AML.
TIFAB accelerates MLL-AF9-Induced acute myeloid leukemia through upregulation of HOXA9. · Zhao J, Xiu Y, ... Zhao C · iScience 2021What this evidence cannot show
Abstract fetched and read back on 2026-09-20; full text not read. Established as what the paper states - the phrase 'its direct target' is verbatim. **It is a mouse MLL-AF9 myeloid system, not the index infant B-lineage KMT2A::AFF1 disease.** 'Direct target' in an abstract is a compression, and the underlying evidence for directness was not inspected here.
Evidence that argues against this (1)
- The companion study in the same system names Dnmt3a and Mef2c as the RelB-responsive genes and does not mention TIFAB, and this graph records its ChIP as negative at the Tifab promoter. Until both full texts are read side by side, 'direct target' should be carried as the 2021 authors' claim rather than as an established property of the wiring. Xiu Y et al. 2018
- E2established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.paper
The 2018 companion study in the same MLL-AF9 NIK-stabilisation system reports that non-canonical signalling upregulates Dnmt3a and downregulates Mef2c, which suppress and promote AML development respectively, and its abstract does not mention TIFAB.
Stabilization of NF-kappaB-Inducing Kinase Suppresses MLL-AF9-Induced Acute Myeloid Leukemia · Xiu Y, ..., Zhao C · Cell Reports 2018What this evidence cannot show
Abstract fetched and read back on 2026-09-20. **Established only as a statement about the abstract's content.** The stronger claim carried elsewhere in this graph - that its ChIP found RelB at the Mef2c and Dnmt3a promoters but NOT at the Tifab promoter - comes from `gene-tifab` E3 and **was not re-verified from the full text for this entry**. Absence from an abstract is not a negative result, and a promoter not assayed is not a promoter shown to be unbound.
Evidence that argues against this (1)
- A 2018 paper cannot be faulted for not discussing a link proposed in 2021, so the silence is weak evidence of disagreement on its own. The contradiction is real only if the 2018 ChIP actually assayed the Tifab promoter and found nothing - which is exactly the thing this entry says has not been checked. Zhao J et al. 2021
- E3moderate confidence 3 of 5: One strong study, or several smaller consistent ones.paper
TIFAB has separately been assigned a metabolic role in KMT2A::MLLT3-induced AML acting through HNF4A, with Tifab deletion impairing leukaemia stem/progenitor enrichment.
TIFAB modulates metabolic pathways in KMT2A::MLLT3-induced AML through HNF4A. · Wang Y, Xiu Y, ... Zhao C · Blood Advances 2025What this evidence cannot show
Carried from this graph's `gene-tifab` E2, abstract-level. Moderate: it establishes requirement in that model rather than that HNF4A is the route. Included here because it shows TIFAB is being built on from two directions while the step immediately upstream of it is unsettled - **it does not bear on whether RelB binds the Tifab promoter.**
Evidence that argues against this (1)
- A gene can be genuinely required in a model while the specific upstream and downstream links proposed for it are both wrong; convergent interest is not convergent evidence, and citing a second role does not strengthen the first. Xiu Y et al. 2018
THE CASE THAT MOTIVATED THE SCREEN, and the one it could not have found: two entries, two papers, no shared PMID. It was caught by rewriting counter-notes by hand.
ITS TWO EVIDENCE ITEMS DISAGREE: E2 carries RelB suppressing TIFAB as a direct target while E3 records the companion study's ChIP as negative at the Tifab promoter.
THE DIRECTNESS CLAIM IS UNSETTLED: 'RelB suppresses its direct target TIFAB' is the 2021 abstract's wording, and the 2018 companion study names Dnmt3a and Mef2c instead.
A LINK IN THE CHAIN THIS ENTRY DEPENDS ON: if RelB to TIFAB is indirect, the non-canonical arm's suppression of MLL-AF9 remains real but stops naming a target.
BOTH SIDES OF THIS CONTRADICTION ARE MYELOID MLL-AF9, which is the lineage split that entry documents.