Their study, published in the March 15 issue of The Journal of Neuroscience, describes for the first time a chain of molecular events that is activated in the brain's primary memory center, called the hippocampus, when estrogen bathes neurons (nerve cells).
The study details how these nerve cells "grow in complexity" when exposed to estrogen, increasing connections among nerve cells in an area of the brain needed to store new memories, retrieve older ones and even recall location of an object or event in space.
A second study, published in the same journal by Weill Cornell Medical College scientists, led by Teresa Milner, Ph.D., in collaboration with Rockefeller University investigators, finds the same results in animal tissue experiments. Both the first study, at the test tube level, and the Milner tissue study were conducted simultaneously but independently, and serve as sort of "blind controls" in support of each other.
"We found a novel way in which estrogen affects neuronal structural remodeling in the hippocampus," says paper co-author Bruce S. McEwen, Ph.D., Professor and head of the Harold and Margaret Milliken Hatch Laboratory of Neuroendocrinology at Rockefeller University.
"It shows us that estrogen plays an unsuspected role in primary biological processes involved in strengthening normal learning and memory function," says McEwen.
"We observed the neuronal structural remodeling at the subcellular level through electron microscopy," notes Milner, professor of neuroscience in the Division of Neurobiology at Weill Cornell. "We were able to visualize precise changes in protein distribution at the actual dendritic spines of the neurons."
Findings from several previous studies have been mixed about whether estrogen replacement therapy bolsters brain functioning in postmenopausal women. McEwen says that the new study suggests some form of postmenopausal estrogen replacement may indeed be both helpful and neuroprotective.
"Even without estrogen, there are still plenty of synaptic connections in the hippocampus," McEwen says. "The study suggests that without estrogen, the connections that are there don't work as efficiently in storing and recalling certain types of memories, such as word lists, or remembering where something is in space," he said. "The hope is that an estrogen mimic could be developed that protects women not just against memory loss, but Alzheimer's disease, the consequences of stroke and other brain disorders."
"Hormones like estrogens, which circulate in the bloodstream, are a major part of the communication system in the brain," McEwen concludes.
At its most fundamental, the study solves several neurobiological mysteries that were seemingly unrelated, says co-author Keith T. Akama, Ph.D., a postdoctoral researcher in McEwen's laboratory. It answers the question of why estrogen receptors, through which the hormone stimulates the cell, are located on the outer reaches of the neurons-- an observation that many researchers have been trying to explain --and it uncovers a functional role for protein synthesis that occurs far away from the cell body near the synapses. Protein synthesis is thought to be important in learning and memory.
"This marries two schools of neurobiology together," Akama says. "Much was known within those two separate investigative paths, but this experiment connected the dots," he says.
Estrogen at synapses
In earlier research with rodents, Milner, together with the McEwen laboratory, had demonstrated for the first time that estrogen receptors occur at the edges of nerve cells in the CA1 region of the hippocampus, far away from the cell's nucleus where most estrogen receptors are traditionally found. These receptors at the edges of the cells are found on structures known as "dendritic spines" -- the part of the cells that receives signals from other neurons in the central nervous system.
From a nerve cell, long tentacles called dendrites branch many times, extending out to reach other neurons. Dendritic branch points are covered with the nub- or bump-like spines, which are often the sites of synapses, junctions between neurons that pass chemical messages. When the spines are activated, they grow, or mature, into mushroom caps in order to make connections with the next neuron.
McEwen and many other scientists believe "plasticity," or the constant structural reshaping of synapses in forming new dendritic spines, encodes processes necessary to promote learning and memory. These spines are diminished in the aged brain and are atrophied in Alzheimer's disease. Furthermore, McEwen also believes that the formation of new spines may be a major way by which the brain protects itself from damage such as trauma and stroke. Plasticity also allows the brain to relearn skills that may have been lost to injury -- such as by stroke -- by rewiring important functions via alternate nervous system pathways.
Previous studies in mammals by McEwen and others showed that low estrogen levels reduced the animals' performance on learning and memory tests, but estrogen treatment reversed this negative effect, thus providing a link between estrogen and activity in the hippocampus. Human studies have also shown that the ability of women to remember word lists and other experimental tasks varies during a normal monthly estrous cycle, which is characterized by the ebb and flow of estrogen.
Further animal investigation revealed a dramatic decrease in dendritic spine density in rats whose ovaries were removed and thus were relatively low in blood estrogen levels; however, administering estrogen to the animals increased their spine formation. McEwen and his colleagues also found that the density of synapses and synaptic spines fluctuates during an animal's estrous cycle, increasing in response to estrogen.
This new study is the first to shed light on the precise molecular pathway by which estrogen increased the "plasticity" of neuronal spines. In the study, McEwen and Akama explored the question of how estrogen receptors influence growth of dendritic spines.
"We know how estrogen works genomically inside the cell's nucleus, how it turns on gene transcription, producing proteins, which are then shipped to where they are needed," says Akama. "But it is a long way from the nucleus to the synaptic ends of the neuron, where changes occur very rapidly, so estrogen has also found a way to work at edges of the nerve cell. We wanted to find out how."
Messenger RNA hanging out
The researchers reasoned that an increase in the number of spines requires the translation, or synthesis, of new proteins, and they chose to investigate a key protein that has been found near estrogen receptors at the spine with undefined regulation of new protein synthesis.
That protein, postsynaptic density-95 (PSD-95), is a structural protein that researchers believe plays a critical role in building a synapse and maintaining plasticity.
"A lot of researchers have looked at PSD-95, but it was not known to play any role with estrogen," says Akama. "Furthermore, no one knew how a neuron regulates PSD-95 production."
Through a series of test tube experiments, Akama and McEwen were able to delineate the molecular mechanisms by which estrogen might directly orchestrate such spine formation and development of synapses.
They found that in a neuronal cell line, estrogen binding to its receptor led to a series of signaling switches that resulted in PSD-95 protein translation. These switches involves rapid activation of an enzyme called Akt, a common intermediate in signaling pathways, which subsequently disinhibits 4E-BP1 (eukaryotic initiation factor-4E binding protein 1) to allow new protein synthesis.
"PSD-95 mRNA is hanging out near the spines and was not being translated because it had a big, inhibiting protein complex bound to it," says Akama. "Phosphorylation of 4E-BP1 disrupts this binding and when estrogen stimulates this release of 4E-BP1, new PSD-95 proteins were rapidly synthesized. More PSD-95 protein translated immediately at the spine increases spine maturation and synaptic formation. All this action is occurring far away from the nucleus, way off in the dendrite, without estrogen traveling back and forth down to the nucleus.
"In addition to the genomic mechanisms initiated within the nucleus, we have shown another way that estrogen can regulate dendritic function, and this gives us hope that selective agents can be developed that work through these signal pathways."
The study was funded by grants from the National Institutes of Health and the Ares-Serono Foundation.
Founded by John D. Rockefeller in 1901, The Rockefeller University was this nation's first biomedical research university. Today it is internationally renowned for research and graduate education in the biomedical sciences, chemistry, bioinformatics and physics. A total of 22 scientists associated with the university have received the Nobel Prize in medicine and physiology or chemistry, 18 Rockefeller scientists have received Lasker Awards, five have been named MacArthur Fellows, and 11 have garnered the National Medal of Science. More than a third of the current faculty are elected members of the National Academy of Sciences.