Old mice lived longer on a drug from Easter Island soil

Three laboratories of the NIA Interventions Testing Program gave rapamycin to mice already old and saw them live up to 14 percent longer, the first drug to extend a mammal's lifespan when started late.

A soil sample from Rapa Nui

Rapamycin was not discovered as a drug for ageing. It was isolated as an antifungal. In 1975 Claude Vézina, Surendra Sehgal and colleagues reported in the Journal of Antibiotics that a streptomycete isolated from a soil sample collected on Easter Island, Rapa Nui, inhibited the growth of Candida albicans and two dermatophyte fungi. The organism was identified as Streptomyces hygroscopicus. The active compound was extracted from the mycelium, crystallised, and named rapamycin after the island. Its minimum inhibitory concentration against ten strains of Candida ranged from 0.02 to 0.2 micrograms per millilitre. It had no activity against bacteria, and its acute toxicity in mice was low.

The compound was later approved for clinical use, and its molecular target was identified as a kinase now called mTOR, the mechanistic target of rapamycin. mTOR integrates signals about nutrients and growth factors and sets the cell's rate of protein synthesis and growth. By the early 2000s, inhibiting the TOR pathway by mutation or by drug had been shown to extend lifespan in yeast, nematode worms and fruit flies. Whether the same was true in a mammal was unknown.

A programme built to be disbelieved

The question was taken up by the Interventions Testing Program of the National Institute on Aging, a consortium set up to test candidate anti-ageing compounds under conditions designed to remove the usual reasons for doubting a lifespan result. Every compound is tested simultaneously at three independent sites: the Jackson Laboratory in Maine, the University of Michigan and the University of Texas Health Science Center at San Antonio. The mice are genetically heterogeneous, bred from a cross of four inbred strains so that each animal is genetically unique yet a full sibling of every other, and no finding can be attributed to a quirk of one genotype. Group sizes give 80 percent power to detect a 10 percent change in mean lifespan even if one site's data were lost.

Rapamycin presented a practical problem: the molecule is unstable in food, so an encapsulated form was developed and incorporated into mouse chow for distribution to all three sites. Treatment in the main cohort began when the mice were 600 days old. A 600-day-old mouse, the authors noted, is in terms of percentage of maximal lifespan roughly the equivalent of a 60-year-old person.

14 percent, at three sites

The result was published in Nature in 2009 by David Harrison, Randy Strong, Richard Miller and colleagues. The dataset was closed on February 1, 2009, with 38 of 1,901 mice still alive. At all three sites, in both sexes, mice fed encapsulated rapamycin from 600 days of age lived longer than controls, with a log-rank p value below 0.0001 for the pooled data.

The headline figure is the change in maximal lifespan, defined as the age at which 90 percent of each group had died. Pooled across sites, rapamycin increased that age by 14 percent in females and 9 percent in males. The site-by-site figures were 17, 14 and 8 percent for females at the Jackson Laboratory, Michigan and Texas, and 10, 4 and 16 percent for males. Expressed as mean lifespan, the gains were 13 percent for females and 9 percent for males. Expressed as remaining life expectancy at 600 days, the age of first exposure, they were 38 percent for females and 28 percent for males. Summed across the three sites, 4.8 percent of female control mice were alive at the age when 90 percent of the joint cohort had died, compared with 21.5 percent of rapamycin-treated females.

Blood levels of rapamycin in the treated animals were 60 to 70 nanograms per millilitre, similar in both sexes, and the drug reduced phosphorylation of ribosomal protein S6, a readout of mTOR activity, four- to five-fold. Pathology at death did not differ between treated and control mice: rapamycin did not change what the animals died of, only when.

The paper described these as the first results to show a role for mTOR signalling in mammalian lifespan and the first pharmacological extension of lifespan in both sexes of a mammal.

Starting earlier, and giving more

The 2009 paper included an interim analysis of a second cohort fed rapamycin from 270 days of age, which was also living longer. That cohort was published in full in 2011 in the Journals of Gerontology. Fed from nine months, rapamycin extended median survival by an average of 10 percent in males and 18 percent in females at each of the three sites, and increased maximum lifespan. Two other compounds tested in parallel, resveratrol at two doses and simvastatin at two doses, had no significant effect on survival in either sex.

In 2014 the same consortium reported in Aging Cell that at a dose three times higher than in the earlier studies, rapamycin increased median lifespan by 23 percent in males and 26 percent in females, with maximal longevity increased in both sexes. At every dose tested, females gained more than males, which the authors attributed in part to higher blood levels of the drug in females. The metabolic and gene expression profile of rapamycin-treated mice was distinct from that of mice on dietary restriction, the other intervention known to extend mouse lifespan, indicating two different mechanisms rather than one.

Three months is enough to measure

The question of how long treatment must last was addressed in 2016 by Alessandro Bitto, Matt Kaeberlein and colleagues in eLife. A three-month course of rapamycin given to middle-aged mice was sufficient to increase remaining life expectancy by up to 60 percent and to improve measures of healthspan. The transient treatment also remodelled the gut microbiome, sharply increasing segmented filamentous bacteria in the small intestine. The same study identified a dose in females that did not extend lifespan at all but shifted the pattern of cancers at death toward aggressive blood cancers and away from solid tumours, a reminder that the drug's effects are not uniformly benign.

The first human data

The 2009 paper's closing suggestion was that interventions targeting mTOR might be relevant to age-related disease in people. The first controlled human test came in 2014 from Joan Mannick and colleagues, published in Science Translational Medicine. Elderly volunteers were given RAD001, everolimus, a rapamycin analogue, for six weeks before an influenza vaccine. At doses that were relatively well tolerated, the drug improved the antibody response to the vaccine by about 20 percent and reduced the proportion of CD4 and CD8 T cells expressing PD-1, an inhibitory receptor whose expression rises with age. The authors framed the result as the first evidence that mTOR inhibition might affect a consequence of ageing in humans. It was a six-week study of a vaccine response, not a study of lifespan.

What the story does and does not show

The 2009 study showed that rapamycin fed to genetically heterogeneous mice from 600 days of age increased the age at 90 percent mortality by 14 percent in females and 9 percent in males, at three independent sites, in a cohort of 1,901 animals. The 2011 and 2014 studies showed larger gains when treatment started earlier or used a higher dose. The 2016 study showed that three months of treatment in middle age had a lasting effect.

The story does not show that rapamycin extends human life. Every lifespan result is in mice, and the one human trial fetched for this story measured a vaccine response over six weeks with a related compound. The 2016 study found a dose that changed the kind of cancer mice died of without lengthening their lives. The 14 percent figure applies to maximal lifespan in females; the gain in mean lifespan was 13 percent for females and 9 percent for males, and at one site the male gain was 4 percent. Rapamycin did not alter the diseases the mice died from.

What the sequence does show is that a compound isolated from Easter Island soil as an antifungal, given to mice at the equivalent of age 60, lengthened their lives at three laboratories at once, and that the finding has been replicated and extended by the same consortium for fifteen years. Whether it can be made to count for people is the question the mice raised and have not answered.

The first drug to lengthen a mammal's life, started old.