Li-Huei Tsai, a professor of neuroscience in M.I.T.'s department of Brain and Cognitive Sciences and the study's senior author, notes that for decades neuroscientists have known that living conditions with lots of novel stimuli improve rodent memory. Using a flexible mouse model, her team set out to determine whether anything could be done to restore memory "after a significant number of neurons are lost in the brain."
Tsai's group manipulated the gene for the protein p25, which has been implicated in several neurodegenerative diseases, so that it could be easily switched on and off. When the protein was on, the mice typically accrued nerve and brain damage and eventually forgot information they had recently learned. Tsai says the mouse model allowed researchers to differentiate between learning impairment and long-term memory loss.
"In most [Alzheimer's] studies, people just look at learning impairment," she says. "In our models, we can really control the timing in which neurodegeneration will happen" to observe loss of memory as well as learning ability.
To establish whether environmental changes could improve learning ability the team put mice with significant neuronal damage in one of two types of cages. One was a standard, single mouse cage with bedding, food and water, and the other was a much larger design in which several mice were grouped together. The latter group were provided with a set of props changed daily, such as running wheels, tunnels and umbrella-shaped toys of different colors. Sure enough, mice that lived in stimulating environments performed better on learning tasks involving associations and spatial reasoning.
The researchers then trained mice to do several activities and respond to stimuli, such as a shock when they were in certain cages. A mouse repeatedly put in a shock box would typically freeze as soon as it was placed inside—a signal that it was afraid and, therefore, remembered being zapped in this particular cage.
After four weeks of such training, researchers reactivated the p25 gene in the mice for six weeks and then tested them to see if they recalled their shocking experiences. They did not.
They then segregated the mice again: One set spent the next four weeks in an enriched environment, the others lived in unadorned, standard cages. When they were returned to the shock cage after the four-week period, the nonenriched mice still had no idea what might be in store, whereas the stimulated rodents became paralyzed with fear, indicating they did.
"That result directly suggests that the memories are not really lost in the first place," says Tsai. "More likely, those long-term memories become inaccessible after a significant number of neurons are lost." When researchers dissected the brains of the environmentally enriched mice (as well as those who also suffered neurodegeneration, but had not lived in plush environs), they saw the same extent of damage done to hippocampus and cortex—two brain regions intimately linked to learning and memory.
"While environmental enrichment didn't seem to significantly influence the number of neurons [that were damaged]," explains Tsai, "it does induce the growth of dendrites and high numbers of synapses," both of which are neuronal structures that underlie how brain cells communicate with one another. In other words, Tsai says, the memories are not actually lost, rather neuronal damage breaks many of the retrieval pathways in the brain.
David Sweatt, chairman of neurobiology department at the University of Alabama at Birmingham, says that the new paper provides evidence for a phenomenon many neuroscientists believed to be impossible: lost memories could be accessed once again. "It's a proof of principle that memories that were unavailable for recall might be recaptured and made available for recall," he explains. "The fact that it can happen at all is pretty profound."
Tsai's team also probed the biochemical mechanism by which environmental enrichment worked its magic, tracing it to proteins called histones, the spools that DNA winds around to create complexes called chromatin in the nuclei of cells. A process called acetylation, where a chemical group is added to histones, causes the protein structures to unravel, exposing parts of DNA, which may have only been accessible during development. This action can result in the activation of genes that had been switched off, and could be the cause of the recovery of synaptic strength. When the researchers administered drugs called histone deacetylase inhibitors (HDACs)—which promote acetylation—to mice, they had the same effect as environmental stimulation.
Tsai says her team wants to probe this biological mechanism further to find the actual targets of HDACs. "We want to know which are the most important genes that are responsible for the improved learning and recovery of memories," she adds. This way, more specific therapeutics can be tailored to activate the correct proteins to counter the effects of neuronal loss. "I really hope,'' she says, "that our studies will one day translate into helping human patients with dementia."