For release: Tuesday, March 11, 2008
As the cells that generate the brain's electrical signals, neurons tend to grab the limelight when it comes to studies of brain function. Until recently, brain cells called glia have been mostly ignored, and their roles remain poorly understood, despite the fact that they outnumber neurons by about 10 to 1.
With growing evidence that glia regulate such complex functions as brain inflammation, cerebral blood flow and even circadian rhythms, interest in them is on the rise – and so is the need for better tools to characterize them.
In a study in the Journal of Neuroscience*, researchers examined patterns of gene activity in neurons and two kinds of glia – astrocytes and oligodendrocytes – isolated from the mouse brain. They found a gene whose activity can be used as a powerful identifier, or marker, for astrocytes. Moreover, although astrocytes were already known to influence neuronal function, the new study suggests they do so in unexpected ways.
"If we're going to understand how neurons and glia are talking to each other, we have to understand the division of labor between them," says Ben Barres, M.D., Ph.D., the study's senior investigator and a neurobiologist at Stanford University in California.
Some roles of glia are already understood. Oligodendrocytes, for example, form layers of a fatty material called myelin, which insulates axons – the slender cables that neurons use to connect with each other.
The functions of astrocytes, which were named for their star-like shape, are more mysterious and potentially vast. Astrocytes appear to serve metabolic functions, including taking up and storing glucose, which suggests that they might provide an energy reserve for glucose-starved neurons. They make contact with blood vessels in the brain, and there is evidence that they can adjust cerebral blood flow to match the brain's energy demands. They also produce signals that can stimulate or suppress inflammation. All of this means that astrocytes could be uniquely equipped to protect neurons from conditions such as stroke or traumatic brain injury.
But testing hypotheses about astrocyte function has been difficult because researchers have lacked reliable methods to identify astrocytes and grow them in cultures free of other cell types, says Dr. Barres. Cells derived from the developing brain and cautiously dubbed astroglia have been the best available approximation, but they appear to behave differently than mature astrocytes.
Dr. Barres' team was able to harvest astrocytes, oligodendrocytes and neurons from both the developing and mature mouse brain. They then used gene chip technology to measure the activity, or expression, of more than 20,000 genes in the three cell types. They found about 100 genes whose expression in one cell type was enriched at least 15-fold higher than in the other two. The work was supported by the National Institute of Neurological Disorders and Stroke (NINDS) and the National Eye Institute (NEI). John Cahoy, a graduate student in Dr. Barres' lab who did most of the experiments, was supported by a training grant from the National Institute of General Medical Sciences (NIGMS).
Most of the genes enriched in neurons and oligodendrocytes were predictable, says Dr. Barres. For example, genes active at axons and synapses (the connections between neurons) dominated the expression profile of neurons, while genes involved in myelin production were highly expressed in oligodendrocytes.
In contrast, many of the astrocyte-enriched genes have unknown functions, or functions that haven't previously been associated with astrocytes. That suggests the cells may have hidden talents, says Dr. Barres. For example, astrocytes apparently express many genes known to be involved in phagocytosis – a gruesome process in which one cell engulfs another cell, in whole or in part. Dr. Barres hypothesizes that during brain development, phagocytosis by glia might help cut back excessive axon sprouting by neurons, an idea supported by research on fruit flies.
Dr. Barres and his group confirmed that astrocytes highly express genes involved in glucose metabolism, but they also found a surprising enrichment of genes involved in other metabolic pathways. One of these genes, called Aldh1L1, is involved in the metabolism of folate (vitamin B9).
Aldh1L1 turns out to be a useful astrocyte marker. Using a fluorescently tagged antibody that recognizes the Aldh1L1 protein, Dr. Barres' group was able to visualize the star-like shape of astrocytes within the mouse brain, including the cells' tiny, higher-order branches. Meanwhile, they found that antibodies to glial fibrillary acidic protein – until now considered the best marker for astrocytes – lit up only the largest branches of some, but not all, astrocytes.
The researchers also used a mouse line produced by the Gene Expression Nervous System Atlas (GENSAT) to validate the use of Aldh1L1 as a marker. GENSAT, supported by NINDS and the NIH Neuroscience Blueprint, is a project to map gene expression in the mouse nervous system. Toward that end, the project has generated mice that carry artificial chromosomes in which the protein-coding parts of genes like Aldh1L1 have been replaced with the gene for jellyfish green fluorescent protein (GFP). Peering into the brains of Aldh1L1-GFP mice, Dr. Barres' group observed green-glowing astrocytes whose distribution and appearance matched what they had seen using Aldh1L1 antibodies.
With an improved astrocyte marker and a catalog of genes enriched in astrocytes, future studies should be able to tackle unresolved questions about astrocyte development and function, Dr. Barres says. The role of astrocytes in disease and the significance of an injury response called reactive astrocytosis – where astrocytes enlarge, undergo shape changes and produce pro-inflammatory signals – will be easier to address too, he says.
To support those efforts, Dr. Barres is feeding his data back into GENSAT and into other resources available to the neuroscience community. Prior to publication of the study, he provided information about the most highly enriched genes found in astrocytes (and neurons and oligodendrocytes) to the NIH program officers who manage GENSAT and Neuromab, a project to develop highly specific antibodies against proteins found in the brain.
Neuromab is now making an improved antibody for detecting Aldh1L1. Meanwhile, GENSAT is making a mouse line in which the expression of Cre recombinase, a DNA-modifying enzyme, is placed under the control of the Aldh1L1 gene. This line will allow researchers to manipulate the expression of almost any gene within astrocytes in the living mouse brain – a technique that could reveal, for example, whether phagocytosis by astrocytes is important for normal brain development.
*Cahoy et al. "A Transcriptome Database for Astrocytes, Neurons, and Oligodendrocytes: A New Resource for Understanding Brain Development and Function." Journal of Neuroscience, January 2, 2008, Vol. 28(1), pp. 264-278.
-By Daniel Stimson, Ph.D.
Last Modified August 23, 2010