Astrocytes make up almost half of a person’s brain cells and are increasingly recognized to play a role in neurodegenerative disease, yet surprisingly little is known about how these cells behave in the human cortex. Most data come from cultured cells or mouse studies. Now a new glimpse at mature human astrocytes suggests they possess unique properties. In the January 6 Neuron, researchers led by Ben Barres at Stanford University, Palo Alto, California, described a method for purifying astrocytes from adult human brain tissue. The researchers analyzed gene expression from these cells by microarray, turning up numerous differences from immature and cultured cells as well as from mature mouse astrocytes. The findings are freely available to researchers in a searchable online database, which includes expression data from other isolated human and mouse brain cell types.

Researchers called this database a valuable resource. “It is going to be quite useful for people to mine. I looked up a few of my favorite genes and learned some things very quickly,” said Kerry O’Banion at the University of Rochester, New York. Other commenters noted that the new findings were unsurprising but provide hard data that will help elucidate the differences between distinct types of astrocyte. “This is an important paper. It is probably the first proof that mouse and human astrocytes are different cell types,” said Maiken Nedergaard, also at Rochester.

Cultural Differences.

Human astrocytes cultured without serum (left) resemble those in the brain, with long thin processes, whereas in the presence of serum (right) they flatten out and look like fibroblasts. [Courtesy of Neuron, Zhang et al.]

One barrier to studying astrocytes has been the difficulty of isolating and culturing the mature cell from the brain. Most current protocols isolate precursor cells rather than mature astrocytes. These precursor cells proliferate in culture in the presence of serum, developing a flat, fibroblast-like appearance that bears little resemblance to astrocytes in the brain (see image above). Other isolation methods, such as fluorescence-activated cell sorting, are too harsh to produce viable cells, co-first author Ye Zhang told Alzforum.

To gently isolate mature astrocytes for culturing, Zhang and co-first author Steven Sloan developed an immunopanning protocol. In this procedure, dissociated cells from brain tissue were passed through several culture dishes coated with antibodies to specific cell-surface markers. The first several plates removed unwanted cell types, while the final plate contained anti-HepaCAM to capture astrocytes. After washing off contaminating cells, the authors detached the astrocytes with trypsin digestion.

Using this method, Zhang and colleagues purified astrocytes from healthy human temporal lobe tissue donated by 12 people undergoing neurological surgeries. The authors obtained about 200,000 mature, postmitotic cells per gram of brain tissue. About 95 percent of these cells expressed astrocyte markers. The authors first confirmed that these cells behaved like astrocytes in vitro, i.e., they were able to promote neuronal survival, stimulate synapse formation, and prune synapses. Then they isolated RNA from these cells, as well as from acutely purified human neurons, oligodendrocytes, microglia, and endothelial cells, to generate a database for comparing gene expression in specific cell types. The database includes previous findings from purified mouse brain cells as well (see Zhang et al., 2014). 

In particular, the authors wondered how similar mouse and human astrocyte gene expression would be. Previous studies by Nedergaard had highlighted morphological differences between these cell types, reporting that human astrocytes are three times larger and have 10 times more processes than those of rodents (see Oberheim et al., 2006; Oberheim et al., 2009). Supporting this, Zhang and colleagues found that while astrocytes from mouse and man expressed 90 percent of their genes in common, they had numerous significant differences as well. Only about half of the genes most highly expressed in mouse astrocytes compared to other mouse cell types were also expressed in human astrocytes, and only about a third of the genes most enriched in human astrocytes were also expressed in mouse. Genes unique to mouse cells included the platelet-activating factor Pla2g7 and the protease inhibitors Itih3 and Lxn, while those specific to human astrocytes included the Wnt inhibitor WIF1 and the enzymes GSTM2 and HSD17B6.

Mouse and human astrocytes also behaved differently in culture. The authors found that adult human astrocytes responded to extracellular glutamate by releasing calcium from internal stores, while those from adult mice remained quiescent under the same conditions.

Nedergaard noted that the differences in astrocyte gene expression between species contrast with the close similarities seen between mouse and human neurons. “Neuronal genes are very preserved during evolution,” she said, adding that human astrocytes seem to have evolved more rapidly. “It may be that human brains are so capable because our astrocytes can process more information,” she speculated. Nedergaard had previously reported that transplantation of human astrocytes into mouse brains improved the animals’ ability to learn and remember (see Han et al., 2013). 

Even among human astrocytes, different cell types exist. Zhang and colleagues isolated astrocytes from six human fetal brains. These cells proliferated in culture and expressed a distinct suite of genes from mature astrocytes. The immature cells, which the authors dubbed astrocyte precursor cells, expressed more cell cycle and proliferative genes, while mature astrocytes expressed more genes involved in cell signaling, adhesion, and synaptic transmission, for example a subunit of the GABA-A receptor. Altogether, about 2,000 genes varied between precursor cells and mature astrocytes.

How does astrocyte expression change during disease states? To examine this, the authors isolated cells from glioblastomas, and found that astrocytes from the tumor core of adults expressed a similar set of genes as astrocyte precursor cells. On the other hand, astrocytes isolated from epileptic tissue displayed an altogether different set of changes. It involved a boost in reactive genes, which also rise in astrocytes during infections or oxygen deprivation. Notably, astrocytes cultured in serum likewise expressed these reactive genes, highlighting their differences from mature resting cells.

In future work, Zhang plans to isolate astrocytes from human brains with conditions such as Alzheimer’s disease. In some instances, related brain diseases are diagnosed with the help of biopsies, for example, inflammatory cerebral amyloid angiopathy (Jan 2016 news); such biopsy tissue could become a source for disease-associated mature human astrocytes.

Zhang pointed out that human transcriptome data of another brain disease, autism, showed that only half of disease-associated genes were enriched in neurons, while the other half popped up mostly in astrocytes and microglia. As in AD, most autism research has focused on neurons. “These data suggest that we are probably missing half of the picture. We cannot understand the brain unless we understand glia,” Zhang said.

Other researchers agreed. “With so many therapeutics failing in clinical trials after performing brilliantly in mouse models, this study makes a strong case for understanding the complexities and intricacies of human astrocytes,” wrote Melanie Das at Cedars-Sinai Medical Center, Los Angeles (see full comment below).—Madolyn Bowman Rogers


  1. This study by Zhang et al. developed a novel technique to isolate and culture human astrocytes in serum-free conditions, thus recapitulating the resting, non-reactive state normally observed in healthy astrocytes in vivo. The study then went on to show how the technique can be applied to model and study fetal and mature astrocytes in health and disease.

    Fortunately for mouse users everywhere, the findings in this study demonstrate that human astrocytes are effectively very similar in their functional capacity when compared to mouse astrocytes. However, the authors also illustrated that human astrocytes still possess unique qualities at the genomic and functional levels. Exploring the differences between rodent and human astrocytes can yield interesting insights into the differences between our species such as in cognitive function, aging, and disease.

    Another important application of this model is that it can be a useful way to shed light on the fundamental processes behind astrocyte maturation in humans. The initial excitement of using induced pluripotent stem cells (iPSCs) to model diseases has since been quelled by the fact that we still lack the ability to drive cells to a mature and/or aged state in vitro. A better understanding of human astrocyte development is crucial to be able to eventually mature and age iPSC-derived astrocytes.

    While the widespread usage of this in vitro model of primary human astrocytes may be hindered due to the limited availability of fresh human tissue, the methodologies and data presented in this study will be useful in validating findings from mouse models and creating better models of mature human astrocytes. With so many therapeutics failing in clinical trial after performing brilliantly in mouse models, this study makes a strong case for understanding the complexities and intricacies of human astrocytes. 

  2. Glial cells constitute more than 50 percent of the cells in the human brain. Astrocytes, which make up the largest glial population, are crucial to the regulation of synaptic connectivity and plasticity, brain tissue repair, formation of the blood-brain barrier (BBB), and regulation of cerebral blood flow (CBF). Although transcriptional profiling to detect astrocyte functions has been done on numerous occasions in multiple animal models, how relevant these data are to human astrocytes remains largely unknown.

    Zhang and colleagues developed a new immunopanning and culture approach to generate populations of fetal and adult human astrocytes that contain faithful characteristics of astrocytes in vivo. They provided a functional and molecular comparison of human and murine astrocytes in vitro using standard assays in order to detect potential differences and/or similarities between species. They first reported that primary human astrocytes are larger and contain more processes than primary rat astrocytes, and also implied different signaling mechanisms. Interestingly, human astrocytes respond differently to neurotransmitters such as glutamate, suggesting that calcium responses might be fundamentally different in human than in rodent astrocytes. That should also differentially affect brain functions including modification of cerebral blood flow (CBF) and blood-brain barrier (BBB) permeability modification.

    Even though about 90 percent of the genes are similar between species, human astrocytes have hundreds of unique genes when compared to mice. Here, the authors used an RNA-seq protocol to understand the genes expressed uniquely in human astrocytes, and in other brain cell types as well, e.g., neurons, endothelial cells, oligodendrocytres, and microglia. They applied this method in a variety of conditions including development (i.e., fetal brains) and diseases (i.e., brain cancer or epilepsy surgeries). Zhang and coworkers showed that astrocytes derived from the tumor core were more transcriptionally similar to fetal immature astrocytes, whereas astrocytes obtained from the tumor periphery were more similar to adult mature astrocytes. It is known from animal studies that astrocytes become highly reactive in neurodegenerative diseases and change their morphology and phenotype accordingly. Further studies screening drugs that can promote or knock down the expression of specific genes are needed to increase knowledge about astrocyte functions. Thus, Zhang et al. concluded their thought-provoking study saying that the observed functional and morphological differences between human and murine astrocytes might be explained by differences in gene expression.

    Finally, another interesting discovery is that astrocytes exist in at least two distinct developmental stages, i.e., progenitor and mature cells. These results enable new perspectives and future directions to better understand astrocyte developmental origins, genetic regulation, and terminal differentiation. These recent findings will be useful to continue seeking new treatment approaches, especially in neurodevelopmental disorders, as well as autoimmune and neurodegenerative diseases.

    The identification of novel astrocyte-encoded genes will certainly allow for discovering new roles for astrocytes during brain development, normal brain aging, and under pathological conditions. These new insights herald the concept that astrocytes represent a diverse population of genetically tractable cells that mediate various roles in health and disease, and that might provide new opportunities to treat central nervous system disorders.

Make a Comment

To make a comment you must login or register.


News Citations

  1. Diagnosis for Inflammation in Cerebral Amyloid Angiopathy

Paper Citations

  1. . An RNA-sequencing transcriptome and splicing database of glia, neurons, and vascular cells of the cerebral cortex. J Neurosci. 2014 Sep 3;34(36):11929-47. PubMed.
  2. . Astrocytic complexity distinguishes the human brain. Trends Neurosci. 2006 Oct;29(10):547-53. Epub 2006 Aug 30 PubMed.
  3. . Uniquely hominid features of adult human astrocytes. J Neurosci. 2009 Mar 11;29(10):3276-87. PubMed.
  4. . Forebrain engraftment by human glial progenitor cells enhances synaptic plasticity and learning in adult mice. Cell Stem Cell. 2013 Mar 7;12(3):342-53. PubMed.

External Citations

  1. database

Further Reading

Primary Papers

  1. . Purification and Characterization of Progenitor and Mature Human Astrocytes Reveals Transcriptional and Functional Differences with Mouse. Neuron. 2016 Jan 6;89(1):37-53. Epub 2015 Dec 10 PubMed.