Mice and pigs with brain injuries did better on movement and memory tests after daily shin squeezes

The experiment I find most persuasive is the knockout. Osteocytes, cells buried deep in bone, carry ion channels called Piezo1 that open when they sense force ^[11]. Bai's team removed Piezo1 from the osteocytes of a separate group of brain-injured mice, and in those animals bone compression had no effect on motor or memory function ^[11]. Those mice were compressed too, so the handling and loading procedure by itself cannot account for the gains in normal mice ^[4].

The idea came from earlier work running in the other direction. Xiaochun Bai at Southern Medical University in Guangzhou and colleagues had found that hundreds of bone-released molecules change with weight-bearing activity such as walking or running. They had also found that brain injuries speed the healing of broken bones, and the new study asked whether loading bone could in turn help an injured brain ^[6].

The serum experiment is the step toward a drug. Serum from treated mice carried raised levels of BDNF, which supports the survival and growth of neurons; PF4, which is involved in inflammation; and dopamine ^[12]. Injecting that serum into untreated mice seemed to pass on the brain benefits with no compression at all ^[12]. The thing this doesn't tell you is which of the three molecules carries the effect, or whether something else in the serum does. I think that experiment matters more for an eventual treatment than the squeeze itself, provided the effect can be pinned to a molecule someone can manufacture.

Durability is mixed. Treated mice came down a pole about as fast as uninjured mice at both one and eight weeks after injury ^[9]. In a water-maze memory test, they crossed the hidden platform nearly five times as often as untreated mice at one week, and about 1.5 times as often at eight ^[10]. A ratio like that can shrink because treated mice slip or because untreated mice recover on their own, and the ratio alone cannot separate the two. The New Scientist account does not give group sizes.

Dose is the harder problem for people. The mice took about 15 times their body weight, applied lengthwise from knee to ankle for 2.5 minutes a day over five days ^[8], a total of 12.5 minutes of loading ^[2]. Human shin bones usually tolerate up to about five times body weight without breaking, according to the researchers ^[15], so the mouse load is three times that ceiling ^[1]. The pigs, male adolescents whose brains are more like ours, got twice their body weight ^[13], or 40 percent of the human figure ^[3]. They lost fewer neurons and less brain tissue, had less inflammation and performed better cognitively than untreated pigs, with the same three molecules raised in their serum ^[13]. Neither species showed visible damage or bruising ^[14].

The clinical gap is large. Each year, emergency departments in England and Wales see more than 1 million people with a recent head injury, about 40,000 of whom have a traumatic brain injury ^[4]. No approved drug restores lost brain function after such an injury ^[5]. "This is big news to me, as it reveals a previously underappreciated bone-brain communication system, and could potentially lead to new therapeutic approaches in the future," said Kristian Franze of the Max Planck Center for Physics and Medicine, who was not involved in the work ^[3]. Medha Pathak at the University of California, Irvine, wants the study repeated in females, since the mice and the pigs were all male ^[7][13]. "Given that oestrogen regulates how bone responds to mechanical load, we need to know whether the effect extends to females," she said ^[16].