One injection improved memory in old monkeys

Researchers at Yale and the University of California, San Francisco improved memory in aged rhesus monkeys with a single low dose of Klotho, and the effect lasted at least two weeks.

A gene named in mice

In 1997 Kuro-o and colleagues reported in Nature a new mouse gene they called klotho. Mice in which the gene was defective developed a syndrome that resembled human ageing: a short lifespan, infertility, arteriosclerosis, skin atrophy, osteoporosis and emphysema. The gene encoded a membrane protein with sequence similarity to beta-glucosidase enzymes, and the authors proposed that its product might be part of a signalling pathway that regulates ageing in vivo.

Eight years later Kurosu, Kuro-o and colleagues showed the converse in Science. Mice engineered to overexpress klotho lived longer than normal mice. The protein, they found, functions as a circulating hormone that binds a cell-surface receptor and represses intracellular signalling by insulin and insulin-like growth factor 1, a mechanism for extending lifespan that is conserved across species. A gene whose loss accelerated ageing and whose excess extended life had, by 2005, earned the label of a longevity factor.

A human variant

Humans carry the same gene. In 2002 Arking and colleagues identified an allele they called KL-VS, carrying six linked sequence variants, two of which change amino acids in the protein, F352V and C370S. People homozygous for KL-VS were underrepresented among the elderly in three separate populations, Bohemian Czechs, Baltimore Caucasians and Baltimore African-Americans, with a combined odds ratio of 2.59, and in cell assays the variants altered how much klotho was secreted.

In 2014 Dena Dubal, Mucke and colleagues connected the variant to the brain. Heterozygous carriers of KL-VS performed better on cognitive tests than non-carriers, and the allele raised klotho levels in serum. Transgenic mice with systemic klotho overexpression performed better than controls on multiple tests of learning and memory, showed enhanced long-term potentiation, a form of synaptic plasticity, and had more of the NMDA receptor subunit GluN2B at synapses; blocking GluN2B abolished the effect. The effects appeared in young mice too, and in humans did not track age, which suggested that klotho acted on cognition by a route separate from the ageing process itself.

One injection in mice

The question that followed was whether a single dose of the protein, rather than lifelong overexpression, could do the same. In 2017 Leon, Dubal and colleagues gave mice a fragment of the secreted alpha-klotho protein by peripheral injection. It enhanced cognition and neural resilience in young mice, ageing mice and a transgenic model of alpha-synuclein disease, despite the fact that the fragment does not cross the blood-brain barrier. Treatment induced cleavage of GluN2B and enhanced NMDA receptor-dependent synaptic plasticity, and GluN2B blockade abolished the effect. The protein was working on the brain from outside it.

Eighteen aged monkeys

Mice share about 70 percent of their phylogeny with humans. Rhesus macaques share about 93 percent, undergo age-related cognitive decline with synaptic changes but without large-scale neuronal loss, as humans do, and perform complex cognitive tasks that mice cannot. In 2023 Stacy Castner at Yale, Dubal at UCSF and colleagues reported in Nature Aging the first test of klotho in a nonhuman primate.

They first produced the rhesus form of the secreted klotho protein, 96 percent homologous to the human protein, and confirmed in mice that a dose of 10 micrograms per kilogram enhanced long-term potentiation and working memory four hours after injection, raising serum klotho about sixfold. In rhesus macaques they tested doses from 0.4 to 30 micrograms per kilogram and found that 10 micrograms per kilogram was the lowest dose that raised serum levels fivefold, into the range that had been effective in mice and the range seen in human cord blood.

The cognitive test was the spatial delayed response task: a monkey watches a food reward being hidden in one of several wells, a screen is lowered for a delay, and the monkey must then choose the correct well. The task is sensitive to ageing in both the hippocampus and prefrontal cortex, and because every trial is unique it does not improve with practice. Eighteen aged monkeys, 13 females and 5 males aged 15 to 28 with a mean of about 21.8 years, roughly equivalent to 65 in humans, were trained to a stable baseline, habituated to vehicle injections, and then given a single subcutaneous dose of vehicle or klotho.

Four hours after 10 micrograms per kilogram, performance on the high memory load version of the task, with up to nine wells, rose significantly above vehicle. The improvement was still present two weeks later. Performance on the normal memory load version also rose, across repeated tests in the first and second weeks. The effect was seen in both sexes.

Why the higher doses did nothing

In exploratory tests at 20 and 30 micrograms per kilogram, no significant improvement was observed on either version of the task. The higher doses did not impair performance; the changes were within 2 to 5 percent and not significant. This differed from mice, which continued to improve at much higher doses. The authors suggested that the greater structural and network complexity of the primate brain might narrow the window, and that lower, more physiological levels of the hormone may be what a therapeutic effect in humans requires.

The duration of the effect outlasted the hormone. Klotho's half-life was estimated at 29.5 hours in aged rhesus macaques, from eight samples in two monkeys, and about 7 minutes in rodents, yet the cognitive change persisted for two weeks in both species. Since peripherally injected klotho does not enter the brain, the authors wrote that the peripheral messengers that carry its signal across remain to be identified.

What the story does and does not show

The 2023 study tested 18 monkeys on one cognitive task, in a longitudinal design in which each animal served as its own control. The animals were not drug naive, though their baseline performance had been confirmed as stable. Two weeks is the longest interval measured; what happens at a month, or with repeated dosing, was not tested. The low dose was chosen for the primary analysis and the higher doses were exploratory, so the dose-response finding rests on fewer sessions. Nothing in the paper measured safety beyond cognition, and nothing was measured in a human.

What the sequence of studies does show is a single thread held across 26 years: a gene whose loss ages mice, whose excess extends their lives, whose human variant tracks with survival and with cognition, and whose protein, injected once at a low dose, improves memory in old mice and in old monkeys for at least two weeks. The last step, from primate to person, has not been taken. The authors' own closing claim is the limit of the evidence: that systemic low-dose klotho "may prove therapeutic in aging humans."

Old primates, one shot, better recall.