Weindruch · The Journal of nutrition 1986 · Controlled animal feeding study · n=?

The retardation of aging in mice by dietary restriction: longevity, cancer, immunity and lifetime energy intake.

Cited 1006 times in the scientific literature.

Level 5 - mechanism / opinion, no new human data

Animal research (non-clinical study design)

PubMed 3958810 · doi:10.1093/jn/116.4.641 · record verified 2026-08-30

What was done

Female mice from a long-lived strain were assigned post-weaning to one of six dietary regimens: 1) ad libitum nonpurified diet; 2) 85 kcal/week of a purified diet (~25% restriction); 3) 50 kcal/week of an essential nutrient-enriched purified diet (~55% restriction); 4) the same 50 kcal/week diet initiated before weaning; 5) 50 kcal/week with progressive lifetime protein reduction; or 6) 40 kcal/week of the enriched diet (~65% restriction). Investigators tracked lifespan, tumor patterns, age-related T-lymphocyte proliferation declines, body weight, and lifetime caloric intake per gram of body mass.

What was found

Mice receiving 50–65% restriction (groups 3–6) showed mean and maximal lifespans 35% to 65% greater than ad libitum controls (group 1) and 20% to 40% greater than 25% restricted mice (group 2). Mice in group 6 lived longest, with the longest-lived 10% averaging 53.0 months. Group 6 also displayed the most pronounced preservation of T-lymphocyte proliferation and reductions in tumors. Across groups 3–6, mice consumed ~30% more lifetime calories per gram of body mass than group 2 mice. Adult body weight positively correlated with longevity within groups 3–5.

Why it matters

This study provides landmark evidence in rodents that graded dietary restriction without malnutrition extends maximum lifespan in a dose-responsive manner up to ~65% restriction, concurrently delaying oncogenesis and immune senescence.

Limits

The study was conducted entirely in female mice of a single long-lived strain, limiting direct generalizability to other strains, males, and humans. Total sample sizes, variance measures, and exact quantitative outcomes for tumor rates and immune assays are omitted from the abstract.

Cited by