What the DNA of a 194-year-old tortoise hints about a very long life

Jonathan is an Aldabra giant tortoise living on the island of St. Helena in the South Atlantic. The researchers put his age at an estimated 194 years and describe him as the oldest known land animal. An earlier study they cite estimated that Aldabra tortoises live about 80 years on average, with an upper limit of about 94. Jonathan has outlived that by roughly a century.

A team led from the Kallel Foundation in Nashville, working with Vanderbilt University, the St. Helena Government and groups in the US and Europe, read Jonathan's genes and the chemical marks on his DNA to look for clues. Keep in mind what kind of study this is: one very old animal, compared with a handful of others.

Some background. Our genes are written in DNA, but cells also put small chemical tags called methyl groups on the DNA. This DNA methylation doesn't change the genetic code; it helps decide which genes are switched on or off. The pattern of tags shifts with age in a fairly regular way, which is why scientists can build epigenetic clocks: formulas that read the tags and estimate how old a body is. This study did not use a ready-made clock. Instead it looked at how much methylation there was and how orderly it was.

What they did

St. Helena's officials did not allow a blood draw from Jonathan, to avoid any risk of infection. So the team collected cells by scraping the inside of his mouth, under the tongue, plus saliva. Attempts to get long, intact DNA from these samples failed, so they read the DNA in short pieces and lined it up against a full reference genome made from the blood of Tank, a 36-year-old Aldabra tortoise kept in captivity.

They then did two things:

• Genes. They compared Jonathan's genes with those of other tortoises (including Lonesome George, a Galapagos tortoise whose genome was published earlier), humans and other species, looking for gene variants found only in Jonathan and for genes that showed signs of having changed under evolutionary pressure.
• Methylation. They mapped the methyl tags across the whole genome in mouth-scrape DNA from Jonathan and four other Aldabra tortoises: a 5-year-old, a 12-year-old, and two adult males whose exact ages are unknown but who were already adults in 1969.

What they found

• Less methylation overall. Compared with the 5-year-old, Jonathan had lower methylation across the genome. Of about 67,000 regions that differed between the two, 91% had fewer tags in Jonathan. The team notes this matches what is seen in human centenarians compared with newborns.
• More disorder, except in gene switches. With age, the tags tend to become more random, a kind of "noise" the researchers measure as methylation entropy. Jonathan's DNA was more disordered than the younger tortoises' almost everywhere. The exception was promoters, the stretches of DNA just in front of a gene that act like its on-switch: there, all five tortoises had low disorder. Across the five animals, average disorder rose closely with age.
• A group of "young-looking" switches. In 2,895 promoters with enough data, the team found 272 where Jonathan, like the 5- and 12-year-olds, had low disorder, while the two older adults had more. Many of these genes work in mitochondria, the cell's power plants (35 genes), in handling RNA, the working copies of genes (25), or in repairing DNA and chromosome ends (15).
• Gene variants in aging pathways. The team found 287 genes with protein changes seen only in Jonathan among the genomes compared. Computer models flagged 41 as possibly important, including genes for DNA repair and for protecting chromosome ends. Jonathan also had 10 copies of a gene for perforin, a protein immune cells use to kill damaged cells; Lonesome George had 7.

What to keep in mind

• One animal. Everything centers on Jonathan, with only four other tortoises for the methylation comparison, two of them of unknown exact age.
• No proof of cause. The authors write that their data are "insufficient to support a direct causal inference". Their idea, that keeping low disorder in certain gene switches helps the cell's power supply, RNA handling and DNA repair stay efficient, is presented as a "hypothetical model".
• Limited samples. Without blood, the team could not use the best sequencing methods, could not measure gene activity (RNA), could not confirm Jonathan's unique variants, and could not test whether disorder in a switch actually makes a gene's activity noisier. Mouth scrapes also gave thin coverage: they could study only about 10% of the tortoise's promoters.
• Predictions, not tests. Whether the gene variants change how proteins work was predicted by computer models, not checked in the lab.
• Tortoises, not people. More animals are needed to know whether these patterns are special to Aldabra tortoises or part of aging in general.

Why it matters

Most aging research looks at the average animal. Jonathan is an extreme outlier, and the team notes that no earlier study had asked whether giant tortoises show the usual age-related changes in DNA methylation. He does, in many ways, but some key gene switches seem to have stayed orderly. That points to a concrete question for future work: does keeping those switches tidy actually help cells and bodies last longer? Answering it will take more animals, better samples and experiments, not one remarkable tortoise.