222 ‒ How nutrition impacts longevity | Matt Kaeberlein, Ph.D.

KEYPOINTS

• 01

  Human nutrition science is structurally hard to do well.

  Kaeberlein and Attia note that because people age so slowly, longitudinal studies of today's 70-year-olds rely on data and methods from 40 years ago; much nutritional epidemiology is decades old. Both have grown less dogmatic: healthy people eat very different diets, from protein-hungry ancestral patterns to ketogenic diets, often sharing little besides leafy vegetables.

• 02

  Calorie restriction is the oldest, most robust longevity intervention in animals.

  Experiments date to the early-to-mid 1930s. In mice, CR extends lifespan substantially — the largest effect Kaeberlein has seen is about 65% restriction producing about a 65% lifespan increase in Weindruch and Walford's graded study. CR mice also have less age-related disease and preserved function in heart and immune system, which he finds more compelling than epigenetic clocks.

• 03

  CR has real costs to muscle and strength.

  Calorie-restricted mice lose body weight and grip strength — normalized grip strength drops on the order of 30–35%. Attia argues few humans would view losing 30% of weight and 30% of strength as good; Kaeberlein agrees the human trade-off is uncertain.

• 04

  The monkey studies disagreed.

  The Wisconsin rhesus study (2009) reported CR animals outlived controls with fewer age-related diseases; the NIA study (2012) found no lifespan difference. Wisconsin controls ate a diet with about 28.5% of calories from sucrose, a key reason for the divergence; a later reconciliation paper tried to explain it. The 30–40-year experiments can't realistically be repeated. Kaeberlein reads the overall monkey data as suggestive that CR improves health in primates, with the lifespan question unresolved.

• 05

  Much of the 'fasting' benefit may be calorie restriction — and vice versa.

  Lab CR mice eat their ration quickly, so they also fast most of the day. Studies that fed mice continuously found about two-thirds of CR's benefit came from fewer calories and one-third from the fasting period. Kaeberlein's 2021 Science review argued much intermittent-fasting and time-restricted-feeding literature conflates the two.

• 06

  Starting CR late helps far less; rapamycin works late.

  Most evidence suggests CR started past about 15 months in mice (like a 40–50-year-old human) yields little benefit. In a 2019 study of 800 female mice switched between diets, late CR had small effects. By contrast, the NIA Interventions Testing Program — rapamycin nominated by Dave Sharp — found robust lifespan extension starting at 20 months, about a 60–65-year-old human, published 2009.

• 07

  mTOR inhibition didn't cause the expected muscle loss.

  Because mTOR drives muscle protein synthesis, researchers expected rapamycin and CR to accelerate sarcopenia. Lab animals didn't show that, though Kaeberlein cautions against extrapolating.

• 08

  Protein restriction extends lifespan in mice, but humans are another matter.

  Restricting total protein or specific amino acids — methionine, tryptophan, branched-chain amino acids — extends mouse lifespan, possibly via different mechanisms, including FGF21 released on low-protein diets. Steve Simpson's nutritional-geometry work explored these ratios.

• 09

  Protein needs probably rise with age.

  In the Levine/Longo epidemiological study, lower protein was associated with lower mortality before 65 but higher mortality after. Because death rates climb nonlinearly after 65, Attia's own modelling suggested high protein's net mortality benefit begins around 50. Attia targets about 1 g per pound of body weight — over twice the 0.8 g/kg RDA — in roughly four 40 g meals.

• 10

  IGF-1 is probably not very informative after about 50.

  An IGF-1 receptor antibody in old mice (Nir Barzilai and Yousin Suh) gave roughly a 14–15% median lifespan extension. But Attia argues that a fit person with slightly higher IGF-1 and a younger immune system likely has lower total cancer risk; lowering IGF-1 from high levels may cut cancer risk without the reverse being symmetric.

• 11

  Function and exercise matter most.

  So-so epidemiology suggests 30–50% of exercise's full benefit comes from going from nothing to about 15 MET-hours per week. Attia emphasizes that frailty dominates life after 75: living to 100 cancer-free is pointless if you're an invalid for 25 years.