An enzyme called caspase 2 plays a key role in cognition by helping to remove unwanted connections, or synapses, between neurons in the brain, a new study from scientists at the Florida campus of Scripps Research suggests.
This “synapse pruning” process, together with the opposing process in which new synapses are formed, underlies everyday learning as well as the healthy maturation of the brain during childhood. The Scripps Research scientists used a variety of experimental approaches, including behavioral studies in genetically engineered mice, to show that synapse pruning depends heavily on caspase 2.
The discovery, published Aug. 9 in Nature Communications, is a big step towards a better understanding of how the brain flexibly rewires itself during development and learning. It also provides evidence for a possible new strategy—blocking caspase 2 activity—to treat Alzheimer’s disease and other memory disorders in which brain connections are lost.
“Our study reveals a critical role for caspase 2 in cognitive flexibility and its underlying molecular mechanisms,” says study senior author Baoji Xu, PhD,professor in the Department of Neuroscience at Scripps Research’s Florida campus.
Caspase enzymes are best known for their roles in apoptosis, a cellular self-destruct process that works, for example, to winnow the neuronal population during brain development. Xu and his colleagues wondered if caspases also are involved in the synapse pruning that occurs during development and learning.
In an initial set of tests, they blocked the activity of individual caspase enzymes in cultured neurons taken from a memory region in mouse brains. The aim was to see if that would lead to an increase in rootlike structures on the neurons called dendritic spines, where synapses form and receive signals from other neurons. Blocking one caspase enzyme, caspase 2, had a big effect, increasing the density of dendritic spines to a striking degree that was obvious in microscope images. The researchers observed these spines at short intervals over two hours to confirm that blocking the enzyme reduces the spine elimination rate without affecting the spine formation rate. When the scientists tried to induce spine shrinkage artificially, using a standard method, blocking caspase 2 activity prevented the shrinkage.
These tests showed that caspase 2 is highly important for the pruning of spines that accompanies synapse loss. The scientists in a further series of tests detailed the cascade of molecular signals that leads from caspase 2 activation to the weakening and loss of synapses and spines.
Versions of caspase 2 are found throughout the animal kingdom, which has long suggested to biologists that this enzyme has an important role conserved by evolution. That role, however, has remained murky for decades—in part because “knockout” mice bred without the caspase 2 gene do not show any obvious abnormalities. After their initial experiments, Xu and his team knew to look for more subtle abnormalities. They found, for example, that caspase 2 knockout mice, compared to their genetically intact littermates, have a similar increase in spine density on the same types of neurons evaluated in the previous experiments.
A battery of behavioral tests also revealed that caspase 2 knockout mice, while normal in most respects, show a reduced ability to forget things in certain contexts. For example, compared to their intact littermates they appear to have more intense and longer-lasting fear in standard fear-memory tests.
Apart from the light it sheds on a process of central importance to brain function, the study points to the possibility of a new way of slowing memory loss in conditions such as Alzheimer’s disease. Prior studies by Alzheimer’s researchers have shown that caspase 2 is produced at unusually high levels in the brains of people with this common form of dementia, while mice engineered to develop an Alzheimer’s-like condition are somehow protected from the loss of synapses, dendritic spines, and memory when caspase 2 activity is blocked.
“We suspect that the caspase 2 signaling pathway may be hijacked in Alzheimer’s to drive the synapse loss that correlates with memory impairment in this disease,” Xu says.
He and his colleagues now plan to study how caspase 2 normally is activated to prune neuronal connections during development and learning—and how it becomes abnormally activated in Alzheimer’s brains.
In addition to Xu, the study’s co-authors were Zhi-Xiang Xu, Ji-Wei Tan, Haifei Xu, Cassandra Hill, Olga Ostrovskaya, and Kirill Martemyanov, all of Scripps Research.
The research was funded by the National Institutes of Health (grants NS073930, DK103335, K105954, and MH105482).