Carrying a Neanderthal Gene Comes With Extra Muscle

Every day for five days, somebody sat down and counted cells by hand. Three counts per flask, averaged, and done blind, so whoever was holding the slide had no idea which flask was which. Inside them were mouse blood cells rigged to starve unless growth hormone reached them, and the only thing separating one flask from the next was a single protein threaded through the cell membrane: in some, the growth hormone receptor that most people alive today carry; in others, the version Neanderthals carried. By day five, the Neanderthal flasks held 39 percent more cells.

That receptor is not a reconstruction sitting in a museum case. It is in living bodies right now, most of them in South and East Asia, inherited from a single extended pulse of interbreeding thought to have happened around 47,000 years ago.

Growth hormone leaves the pituitary, a pea-sized gland at the base of the brain, and travels the bloodstream looking for cells that will listen. Listening means having the right receptor on the surface: the hormone clamps on, the receptor’s inner tail rearranges, a kinase parked underneath starts tacking phosphate groups onto that tail, and a transcription factor docks there, gets switched on, and carries the message to the nucleus. Neanderthals had a version of that receptor with two amino acid substitutions absent from the modern human reference genome, plus a deletion that swallowed the whole of the gene’s third exon. Which was worth a look mainly because of what Neanderthals looked like: heavy bones, big muscle attachments, brow ridges, barrel chests, and famously stubby tooth roots, a combination distinctive enough that early researchers reckoned they belonged on a branch of their own.

One Amino Acid Does All the Work

Two changes and a deletion, though, is three things, and three things can be pulled apart. So the team, led by Hugo Zeberg at the Karolinska Institutet and Philipp Kanis at the Max Planck Institute for Evolutionary Anthropology, built every combination of them: eight receptor variants in all, each dropped into the same mouse cell line, each grown out and counted, 187 experiments deep.

Only one of the three mattered. Swapping a proline for a threonine at position 561, deep inside the receptor’s inner tail, made the cells grow 6 percent faster on its own and accounted for the entire difference; the other substitution and the deletion did nothing measurable, and there was no sign of the changes working on each other. Cells carrying threonine at 561 also held 22 percent more of the activated transcription factor, which is what you would expect if the receptor were simply shouting louder.

What makes the paper unusual is that the dish and the population data point the same way, which is not how these things usually go. “What struck me most was that two very different types of evidence told the same story,” says Kanis, the study’s first author. “Cells grown in the laboratory responded more strongly, while data from more than a million people showed signs of the same effect in the body. It is rare to see laboratory and population evidence fit together so clearly.”

Nothing Shows Up Until After Childhood

The body data come from 1.1 million adults across five population biobanks (Britain, Finland, Japan, Korea and Taiwan). Carriers weigh more, by 285 grams per copy of the variant, which sounds like nothing until you ask where it sits. Almost all of it is muscle, 271 grams of it, split between the arms and legs and the fat-free mass of the trunk, and carriers run slightly leaner on body fat rather than heavier. In the Taiwanese data the gluteal region is a fraction wider around with no matching rise in fat percentage, which is the sort of detail that turns a weight association into a muscle one. They are taller too, by 0.30 centimeters per copy, a figure the authors go out of their way to call minor and almost certainly drowned out by everything else that decides how tall a person gets.

None of it shows up in children. In 6,602 children from a Bradford birth cohort (about half of the families there are of South Asian heritage, which matters, because that is where the variant is common), measured at four to five, seven to eight and ten to eleven, it tracked with nothing at all. One possible reading is that the difference only kicks in around puberty, when growth hormone levels run about three times higher than in childhood. Miriam Berreiter, a co-author who does CrossFit, said she was delighted to see Neanderthal genetics and muscle mass in the same study, then added the caveat: “But even a Neandertal growth hormone receptor is no substitute for training.”

Two wrinkles, though. The stretch of Neanderthal DNA carrying the receptor gene runs on past its far end and takes in two neighbors, CCDC152 and SELENOP, so the authors cannot rule out that some of what turns up in carriers is coming from those instead; and, more awkwardly, the Neanderthal version of the gene is expressed at lower levels than the modern human one in people today, which pushes against the very effect the cell work demonstrates.

What the Jaw and the Tooth Roots Kept

Still, the jaw. Pooling four published studies, the team found carriers have a mandibular ramus, the rising rear branch of the lower jaw, about 3.6 millimeters shorter per copy, and a scan of 110 Japanese volunteers agreed. Two Indian studies point to a raised risk of overbite, and two more, both small enough that the authors ask for bigger ones, hint at shorter tooth roots. A trait so bound up with Neanderthals that the word for it, taurodontism, was coined while anatomists picked over their remains. “But this is only one of many genetic influences on growth and body shape,” says Zeberg, the study’s senior author. “It cannot explain the Neandertal body type on its own, and it certainly does not determine a person’s overall appearance.”

How common is it, though? The Max Planck press release says the variant is carried by up to 24 percent of people in parts of South Asia, against around 0.5 percent of Europeans, but the paper gives those as allele frequencies, the share of chromosomes rather than the share of people, and since everyone carries two copies of the gene, the proportion holding at least one Neanderthal copy runs higher than either number suggests. Which leaves the obvious experiment undone: nobody has yet watched this receptor work inside a human muscle cell, or a bone, or a growth plate. Only in a dish, in mouse cells. And a dish is not a shoulder.

Frequently Asked Questions