Complex I inhibition triggers the mitochondrial integrated stress response through DELE1 and HRI, and that is what creates the MCL-1 dependency - a metabolic drug manufacturing an apoptotic vulnerability
also called DELE1-HRI-ATF4 axis · mitochondrial ISR and MCL-1 synthetic lethality · complex I plus MCL-1 inhibition
Damage a cell's mitochondria in a specific way and they send out a distress signal that changes which survival proteins the cell is leaning on. In leukaemia this makes the cell suddenly depend on MCL-1, so a drug that blocks MCL-1 - which does very little on its own - becomes lethal when given with a mitochondrial drug.
The same thing, for a scientist
The reported chain is: **electron transport chain complex I inhibition -> mitochondrial stress -> DELE1 -> HRI (EIF2AK1) -> eIF2-alpha phosphorylation -> ATF4 -> integrated stress response -> MCL-1 dependence**, with co-targeting of complex I and MCL-1 synergistically reducing viability in AML cell lines and patient-derived xenograft samples in vitro and significantly prolonging survival in PDX-bearing mice [E1]. This is the most mechanistically explicit link in this beat between a *metabolic* intervention and an *apoptotic* vulnerability, and it is the reason `gene-mcl1` belongs in a metabolism folder at all. It is also a partial answer to the observation in `gene-mcl1` that MCL-1 inhibitors do almost nothing as single agents in vivo: on this account they are not supposed to, because the dependency they exploit does not exist until something creates it. **The claim that this is KMT2A-specific is not established.** The paper names KMT2A-rearranged AML as the motivating high-risk subtype and as the population that might benefit; the abstract reports no KMT2A-stratified analysis and does not say how many models carried a KMT2A rearrangement.
Why this is a real mechanism and not a combination screen result
Two features separate it from an ordinary synergy finding.
- The intermediate is named and genetically tested. HRI activation by complex I inhibition is reported as dependent on DELE1 1. DELE1 is the mitochondrial protein that is cleaved on stress, released to the cytosol and binds HRI; a DELE1 dependency is a specific, falsifiable statement about the route, not a correlation.
- It explains an existing anomaly rather than adding a new one. MCL-1 inhibitors are potent in vitro and inert as single agents in vivo, in this graph's own infant KMT2A-rearranged models as well as in adult trials. A mechanism in which the dependency is induced predicts exactly that pattern.
The complex I inhibitor is the same class that is separately reported to work in KMT2A-rearranged AML, and separately reported to backfire in B-ALL
The complex I inhibitor with clinical-grade data in this disease is IACS-010759. A 2026 paediatric AML PDX platform paper reports that IACS-010759 plus venetoclax reduced AML progression in KMT2A-rearranged PDXs modelling both disease onset and relapse, and concludes that its results strongly support mitochondrial targeting in this genetic AML subtype 2. That is an independent, KMT2A-stratified, in vivo argument for the same axis by a different route (BCL-2 rather than MCL-1 as the apoptotic partner).
Against that sits the lymphoid inversion this graph already carries: in MLL-rearranged B-ALL, inhibiting oxidative phosphorylation blunts blast proliferation but promotes leukaemia-initiating cell emergence, while inhibiting hypoxia and glycolysis impairs them 3. So the metabolic move that creates a therapeutic vulnerability in the myeloid disease has been measured to expand the relapse-capable compartment in the lymphoid one.
Nobody has run complex I inhibition plus a BH3 mimetic in infant KMT2A::AFF1 with a leukaemia-initiating-cell readout. Until they do, the mechanism should be written as myeloid.
Identifier notes for the nodes in the chain
Resolved 2026-09-20, primary gene names read back, because two of the three are routinely written by an alias:
| protein in the paper | gene_primary | accession | length |
|---|---|---|---|
| DELE1 | DELE1 (synonyms DELE, KIAA0141) | Q14154 | 515 |
| HRI | EIF2AK1 (synonyms HRI, KIAA1369) | Q9BQI3 | 630 |
| MCL-1 | MCL1 (synonym BCL2L3) | Q07820 | 350 |
HRI is an alias, not a symbol, and a PubMed enumeration keyed to HRI will return heat, hazard-ratio and unrelated material; keyed to EIF2AK1 it will miss most of the ISR literature, which writes HRI. Both forms are needed. Neither DELE1 nor EIF2AK1 is given an entry of its own by this beat, because neither has any KMT2A-specific evidence: they are recorded here as components of a single reported chain, which is the disciplined alternative to two thin roster entries.
| genetic test in the chain | HRI activation by complex I inhibition is reported as DELE1-dependent |
|---|---|
| direction of causation | a metabolic intervention MANUFACTURES an apoptotic dependency that did not previously exist |
| explains | why MCL-1 inhibitors are potent in vitro and inert as single agents in vivo |
| kmt2a status | KMT2A-MOTIVATED, not KMT2A-demonstrated: the source abstract names KMT2A-r AML as the target population but reports no KMT2A-stratified analysis |
| independent support for the complex i arm | IACS-010759 plus venetoclax in KMT2A-rearranged paediatric AML PDXs (PMID 41576348) |
| lineage inversion | OXPHOS inhibition promotes leukaemia-initiating cell emergence in MLL-rearranged B-ALL (PMID 35476984) |
| alias note | HRI is an alias of EIF2AK1; a literature enumeration needs both forms, and bare 'HRI' returns heat, hazard-ratio and unrelated material |
| no separate entries for | DELE1 and EIF2AK1 - neither has KMT2A-specific evidence, so they are recorded as chain components rather than given thin roster entries |
chain
[ "complex I inhibition", "mitochondrial stress", "DELE1 (Q14154, 515 aa)", "HRI / EIF2AK1 (Q9BQI3, 630 aa)", "eIF2-alpha phosphorylation", "ATF4", "integrated stress response", "MCL-1 (Q07820, 350 aa) dependence" ]
- Complex I is required by every oxidative tissue. The clinically characterised agent in this class produced dose-limiting neurotoxicity and lactic acidosis in adults, and the paediatric exposure needed for this mechanism has not been established.
- The mechanism's second drug is an MCL-1 inhibitor, a class with published cardiac troponin elevations in two first-in-human studies and a low response rate. Combining two agents with unresolved on-target toxicity does not obviously widen a therapeutic index.
- If the mechanism generalises, it means the leukaemia is not MCL-1-dependent until you make it so. Give the complex I inhibitor without the MCL-1 inhibitor - for example if the second drug is held for toxicity - and you have induced a survival dependency and then left it unexploited.
- In the lymphoid disease the same first move has been measured to promote emergence of leukaemia-initiating cells. Applying a myeloid-derived mitochondrial strategy to infant B-lineage KMT2A-rearranged ALL could reduce blast counts while increasing relapse risk, and blast count is what a clinician would be watching.
- Does complex I inhibition create MCL-1 dependence in infant KMT2A::AFF1 B-lineage cells, or does it expand the leukaemia-initiating compartment as OXPHOS inhibition does in MLL-rearranged B-ALL? One experiment with both a viability and a limiting-dilution readout settles the most important open question in this beat.
- Is the DELE1-HRI route necessary, or merely present? A DELE1 knockout combined with an MCL-1 inhibitor and a complex I inhibitor would show whether removing the messenger removes the synergy.
- Which complex I inhibitor, and at what exposure? IACS-010759's clinical dose-limiting toxicity is the practical obstacle to this entire axis, and the mechanism paper does not name its compound in the abstract.
- Does ATF4 activation itself drive the MCL-1 dependence, or is MCL-1 dependence a parallel consequence of mitochondrial stress? ATF4 has documented effects on both pro- and anti-apoptotic BCL-2 family members.
- E1strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper
Inhibition of electron transport chain complex I is a synthetic lethal partner for MCL-1 inhibition: co-targeting complex I and MCL-1 synergistically reduces viability in AML cell lines and patient-derived xenograft samples in vitro and significantly prolongs survival in mice bearing PDX AML; mechanistically, complex I inhibition induces activation of the integrated stress response, resulting in ATF4 activation downstream of the eIF2-alpha kinase HRI, and HRI activation via complex I inhibition is dependent on the mitochondrial stress messenger DELE1. MCL-1 is stated to promote survival and confer therapeutic resistance in AML, particularly in high-risk subtypes harbouring KMT2A rearrangements.
Mitochondrial integrated stress response activation creates a therapeutic vulnerability to MCL-1 inhibition in acute myeloid leukemia. · Brakefield-Laird L, Budhraja A, Hall PM, ..., Opferman JT · Cell Death Dis 2026What this evidence cannot show
Abstract read this session; full text not retrieved. **This is KMT2A-motivated, not KMT2A-demonstrated.** The abstract names KMT2A-rearranged AML as the high-risk subtype of interest and as the population that could benefit, but reports no KMT2A-stratified result and does not state how many cell lines or PDX samples carry a KMT2A rearrangement. The dispatch brief for this beat asked explicitly whether this paper is KMT2A-specific: on the abstract, the answer is no. Myeloid only; no lymphoid, infant or KMT2A::AFF1 data. Which complex I inhibitor was used is not named in the abstract, which matters because the clinically characterised agent in this class (IACS-010759) has a documented dose-limiting neurotoxicity and lactic acidosis profile.
Evidence that argues against this (1)
- In MLL-rearranged B-ALL, inhibiting oxidative phosphorylation blunts blast proliferation while promoting the emergence of leukaemia-initiating cells, whereas inhibiting hypoxia and glycolysis impairs them and gives therapeutic benefit in xenografts. The manoeuvre that creates this therapeutic window in myeloid disease has been measured as expanding the relapse-capable compartment in the lymphoid disease. Morris V et al. 2022
- E2strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper
In a platform of 26 patient-derived xenografts of paediatric AML representing 14 high-risk genetic subtypes, druggable variants and aberrantly activated pathways were selected in KMT2A-rearranged AML and screened in a three-dimensional coculture model; the combination of IACS-010759, a mitochondrial complex I inhibitor, with venetoclax reduced AML progression in KMT2A-rearranged PDXs modelling both disease onset and relapse, and adding a stromal-targeting drug to that combination slowed progression in a resistant model; the authors conclude their results strongly support mitochondrial targeting in this genetic AML subtype.
Advancing precision therapy in pediatric acute myeloid leukemia through PDX models and mitochondrial targeting. · Da Ros A, Peloso A, Longo G, ..., Pigazzi M · Blood Adv 2026What this evidence cannot show
Abstract read this session; full text not retrieved, so the number of KMT2A-rearranged PDX models in the combination experiment, the effect sizes and the survival statistics were not checked. Paediatric **AML** only. The apoptotic partner here is venetoclax (BCL-2), not an MCL-1 inhibitor, so this supports the complex I arm of the mechanism rather than the MCL-1 arm. IACS-010759 has been through clinical evaluation in AML and solid tumours with dose-limiting toxicity; a preclinical PDX benefit does not resolve that.
Evidence that argues against this (1)
- Complex I inhibition and BCL-2 inhibition both converge on mitochondrial apoptosis, so a benefit from their combination is compatible with simple additivity on one axis rather than with the specific DELE1-HRI-ATF4 priming mechanism; the two studies support the same clinical move for potentially different reasons. Brakefield-Laird L et al. 2026
- E3strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper
Single-cell transcriptomics and quantitative xenotransplantation in MLL-rearranged B-ALL found that while inhibiting oxidative phosphorylation blunts blast proliferation, this intervention promotes leukaemia-initiating cell emergence, whereas inhibiting hypoxia and glycolysis impairs MLL-rearranged B-ALL leukaemia-initiating cells and provides therapeutic benefit in xenotransplantation systems.
Hypoxic, glycolytic metabolism is a vulnerability of B-acute lymphoblastic leukemia-initiating cells. · Morris V, Wang D, Li Z, ..., Rowe RG · Cell Rep 2022What this evidence cannot show
Abstract read this session; identifiers re-fetched from PubMed rather than copied from the sibling entries in this graph that already cite this paper. Xenotransplantation into immunodeficient mice measures engraftment capacity, which is the best available proxy for leukaemia-initiating capacity and is not the same thing. The OXPHOS inhibitor used is not named in the abstract, so a class statement rests on the authors' framing. MLL-rearranged B-ALL generally; infant-specific stratification is not reported in the abstract.
Evidence that argues against this (1)
- A paediatric AML PDX study reaches the opposite operational conclusion for the myeloid disease, reporting that complex I inhibition combined with venetoclax reduces KMT2A-rearranged AML progression at both onset and relapse. The contradiction is between lineages rather than between studies, and no experiment has yet run both readouts in one system. Da Ros A et al. 2026
complex I inhibition creates the MCL-1 dependency via DELE1-HRI-ATF4
complex I inhibition manufactures the MCL-1 dependency via DELE1-HRI-ATF4
mechanistic account of why complex I blockade helps in KMT2A-r AML
the mechanism requires an OXPHOS-dependent starting state
KMT2A-motivated, not KMT2A-demonstrated: no stratified analysis in the source