Pulsed mTOR activation restores ATF4 competition with c‑Jun to destabilize senescence‑locked enhancer landscapes

Senescent cells maintain a stable SASP through early c‑Jun pioneer activity that remodels enhancers and later becomes resistant to JNK or mTOR inhibition [1][2]. We hypothesize that the irreversibility stems not from permanent chromatin marks but from a temporal mismatch: after the 72 h window, c‑Jun homodimers dominate AP‑1 complexes because defective ISR signaling prevents ATF4 synthesis, leaving no competitor for enhancer binding [3]. A brief, controlled pulse of mTOR activity—sufficient to restore ribosomal scanning and ATF4 translation without re‑engaging growth programs—should transiently increase ATF4‑containing AP‑1 dimers, allowing them to occupy c‑Jun‑primed enhancers and disrupt the positive feedback loop that sustains MKP5 suppression and oxidative stress [4][5].

Mechanistically, mTORC1 phosphorylates eIF2B and promotes eIF2α‑P dephosphorylation, relieving the translational block on ATF4 [6]. In senescence, chronic eIF2α‑P persists but ATF4 remains low due to impaired scanning. A short leucine or insulin pulse (15–30 min) would activate mTORC1, improve eIF2B activity, and permit a burst of ATF4 synthesis. ATF4 can then form heterodimers with c‑Jun or JunD, competing for the same AP‑1 sites established by pioneer c‑Jun. Because these enhancers are still accessible (they are not yet fully methylated), ATF4 binding would recruit co‑activators that favor antioxidant genes (e.g., GCLC, SOD2) over SASP chemokines, thereby reducing ROS, restoring MKP5 expression, and breaking the JNK‑MKP5 positive feedback loop.

Testable predictions:

1. In RAS‑induced senescent fibroblasts, a single mTOR pulse administered after 96 h of stress will increase ATF4 protein levels (Western blot) without raising p‑S6K beyond baseline, indicating transient mTORC1 activation.
2. ChIP‑qPCR will show increased ATF4 occupancy at c‑Jun‑bound enhancers of SASP promoters (IL6, CXCL1) and decreased c‑Jun homodimer binding at the same sites.
3. SASP mRNA and secreted protein levels will drop 48 h after the pulse, concomitant with reduced phospho‑JNK and elevated MKP5 phosphatase activity.
4. Continuous rapamycin treatment (24 h) will not produce these changes, confirming that the effect requires a pulse rather than sustained inhibition.
5. Re‑inducing oxidative stress after the pulse will fail to reestablish the SASP if ATF4 levels remain elevated, indicating a durable reset of the enhancer state.

Falsifiability: If mTOR pulses do not increase ATF4 translation, do not alter AP‑1 dimer composition at enhancers, or fail to reduce SASP despite successful mTORC1 activation, the hypothesis is refuted. Conversely, observing the predicted molecular and phenotypic changes would support the idea that senescence‑locked enhancer landscapes are reversible by temporally precise mTOR‑driven ATF4 competition, extending the civilization‑versus‑survival dial concept into a dynamic control strategy.