Scientists have worked out how a meal eaten in infancy can still be shaping an animal weeks later, and part of the answer is that a good deal of the animal is still the meal.
A team led by Fumiaki Obata, a biologist at the RIKEN Center for Biosystems Dynamics Research in Kobe, Japan, reported in Nature that adult fruit flies carry a chemical record of what they ate as larvae, and that the record helps set how long they live.
Laboratory fruit flies are raised on yeast and sugar, with the yeast supplying protein. "We decreased only the yeast concentration in the diet, from eight percent to either one or two percent," Obata said. The change was made roughly halfway through the larval stage; once the flies emerged from their pupae they went back to standard food.
The protein-restricted flies, male and female alike, outlived their well-fed siblings. They were also paler and lighter, sometimes by 28 per cent, and the females laid fewer eggs. "It's very common, actually. Reproduction and lifespan are always in a trade-off relationship," Obata said.
To find where the record was kept, the team fed larvae a synthetic diet in which two amino acids, lysine and arginine, were built from heavier isotopes of carbon and nitrogen. Adults then received food tagged with lighter isotopes, and mass spectrometry weighed protein fragments taken from the flies' heads to establish which meal each fragment came from.
Three days into adulthood, nearly 64 per cent of the proteins in a fly's head were still made from amino acids eaten by the larva. By day six the figure had fallen to about 46 per cent.
Prominent among them were ribosomal proteins, the machinery that builds all the others. Flies restricted as larvae had fewer of them and made new proteins more slowly through their first week.
The carrier turned out to be larval serum protein 2, or Lsp2, which larvae stockpile as a pantry for metamorphosis, when they stop eating. Silencing Lsp2 in larvae produced the same outcome as the low-protein diet: fewer ribosomal proteins, slower production, longer lives.
Humans have no direct equivalent of Lsp2. Obata suggests the nearest counterparts are albumin and the globulins, which together make up around 90 per cent of the protein in our blood, and says the idea would have to be tested in mice first.
The isotope figures deserve to be read literally, because there is no other way to read them. At three days old, an adult fruit fly's head is 64 per cent childhood by weight of protein. At six days it is 46 per cent.
Somewhere between the two, the animal stops being mostly a maggot.
This newspaper proposes that the crossing point be recognised. A fly leaves its pupa an adult in the anatomical sense, but at that moment most of its head remains the property of a previous life stage, assembled out of yeast eaten before it could fly.
Day six is therefore the first day on which a fruit fly is principally itself. The Department of Unnecessary Research has proposed it as the age of majority for Drosophila melanogaster, on the grounds that a simple majority is precisely what it is.
The flies fed less as larvae live longest, which the study attributes to slower protein production. There is a second reading, and it is the one this newspaper prefers: they had less childhood to carry.