This review gathers evidence from yeasts, worms, flies and mice that autophagy promotes longevity. The strongest support is genetic: knocking down autophagy genes abolishes lifespan extension in every long-lived mutant tested, with little effect in normal invertebrates, and overexpressing single genes such as Atg8a, Atg1 or Atg5 extends lifespan in flies and mice (Table 1).
Alternative hypothesis: Autophagy is needed for animals to tolerate the conditions that extend lifespan, rather than being what slows ageing. Dietary restriction and germline loss change nutrient and lipid handling, and autophagy is a survival response to nutrient deprivation, so its loss could shorten life through a cost specific to that state. For the overexpression lines, the review notes that neuronal Atg1 lowers brain insulin-like peptides, so part of the gain may come through hormonal signalling rather than enhanced clearance.
Ideally, the knockdown survival data would be modelled for mortality: if autophagy slows ageing, its loss should steepen the age-dependent rise in mortality, whereas a condition-specific cost would be expected to raise mortality at all ages without changing that slope. For the overexpression lines, blocking lysosomal degradation would show whether the extension requires cargo breakdown or persists through signalling.