“Almost every synapse in the brain is completely enveloped by astrocytes,” said Rothstein, reasoning that interaction between neurons and astrocytes is unlikely to go one way only. Rothstein, first author Yongjie Yang, also of Johns Hopkins, and colleagues used EAAT2 expression to track what astrocytes were doing in the presence or absence of neurons. EAAT2’s presence in a synapse, to remove excess glutamate, is critical: “If it’s not there, synapses just melt,” Rothstein said.

To evaluate the interaction between neurons and astrocytes, Yang and his co-authors used a two-compartment microfluidics chamber (Park et al., 2006). They grew neurons in one side and astrocytes on the other; narrow connecting channels allowed axons to cross to the glia’s side. The scientists used astrocytes isolated from mice expressing GFP under control of the GLT1/EAAT2 promoter (Regan et al., 2007) to track the control of GLT1 expression in the astrocytes. They found that both secreted factors—likely including glutamate itself—as well as axon-to-astrocyte contact, were important to turn on the GLT1-GFP. The exact nature of the cell-cell contact is unclear, but Rothstein suspects that membrane proteins interact. “There is this intimate relationship between an axon and an astrocyte,” Rothstein said.

Next, Yang and colleagues sought to define the factor that controls GLT1/EAAT2 expression in response to neuronal contact with the astrocytes. They cloned fragments of the human EAAT2 promoter with a luciferase reporter and determined that deletion or mutation of a short sequence nearly 700 base pairs upstream of the gene’s start sharply reduced promoter activity. Adding a snippet of DNA with that sequence to mouse cortex nuclear extract, the researchers purified the transcription factor κB-motif binding phosphoprotein (KBBP). In mice, they found that the KBBP-binding sequence was important for activation of a DsRed reporter driven by the EAAT2 promoter, and that KBBP was upregulated at the same time as GLT1-GFP.

Next, the scientists considered how the relationship between axons and astrocytes might be broken. When the axons of mice were damaged by transection, ricin, or the overexpressed superoxide dismutase 1 (SOD1) that causes ALS-like symptoms, both KBBP and GLT1 levels decreased in the nearby astrocytes. The result makes sense: If the neuron is gone, there is no need for the astrocyte to clear away glutamate. But this event in a single synapse could spread, Rothstein warned, because every astrocyte interacts with a great many synapses. “That astrocyte complex is affecting another neuron somewhere else,” he said. “Astrocytes propagate their injury.”

The information from the current study could help discover drugs that modify glutamate uptake, said Robert Edwards of the University of California, San Francisco, who wrote an accompanying commentary in Neuron. “Maybe this will give you more powerful ways to prevent the downregulation of the transporters,” he suggested.

Trying to alter glutamate cycling is hardly a new idea. The only FDA-approved drug for ALS, riluzole, acts on glutamate transport (Frizzo et al., 2004). Rothstein previously reported that the antibiotic β-lactam boosts GLT1/EAAT2 levels in animal models (see ARF related news story and Rothstein et al., 2005). In a more recent study, joint first authors William Boston-Howes and Eric Williams, principal investigator Davide Trotti, and colleagues at Thomas Jefferson University in Philadelphia, Pennsylvania, reported that the anti-inflammatory drug nordihydroguaiaretic acid (NDGA) increased glutamate uptake in cell culture and in normal mice. But when they administered the drug to mice carrying human mutant SOD1, a common model for ALS, they found that the initial uptick in glutamate recycling dropped after 10 days.

These researchers also noted that mSOD1 animals express higher-than-normal levels of the multidrug transporter P-glycoprotein. This transporter pumps toxins out of the cell. It has been known in cancer studies to also export therapeutic drugs, and, being expressed in capillary endothelial cells of the blood-brain barrier, is a well-known foe of AD drug developers, as well. P-glycoprotein might export ALS medications, too, Maragakis wrote in his current commentary. Many therapeutics that appear to have potential to treat ALS fail in human trials; it seems possible that they are unable to help the cell before multidrug transporters expel them. “Rethinking traditional approaches to drug therapies in ALS may be on the horizon,” Maragakis wrote. He suggested scientists should look for excess P-glycoprotein in autopsy tissue from people who died of ALS.

Trotti noted that multidrug transporters have not received much attention in ALS drug testing studies before. “This is the first attempt to understand why so many drug trials have failed for ALS,” he said. Trotti hopes that by blocking P-glycoprotein activity, he could give NDGA a better shot at boosting glutamate transport and reducing symptoms in mSOD1 mice. In AD, experimental inhibitors of the drug target BACE are being tested with such pump blockers, too (Hussain et al., 2007). However, Trotti is quick to note that P-glycoprotein blockers do not yet belong in the ALS clinic. “This concept still has to be proven,” he told ARF. Blocking multidrug transporters would prevent the cell from expelling toxins, Trotti noted, so could have serious side effects. If he does show that P-glycoprotein or other transporters mediate drug efflux, then that could have implications for future drug trials. “When you are designing a therapeutic trial for ALS, you might have to deal with proteins that are acting against your strategy,” he said.—Amber Dance.

References:
Yang Y, Gozen O, Watkins A, Lorenzini I, Lepore A, Gao Y, Vidensky S, Brennan J, Poulsen D, Won Park J, Li Jeon N, Robinson MB, Rothstein JD. Presynaptic regulation of astroglial excitatory neurotransmitter transporter GLT1. Neuron 1009 Mar 26;61(6):880-94. Abstract

Edwards R. What the neuron tells glia. Neuron 2009 Mar 26;61(6):811-2. Abstract. Abstract

Maragakis NJ. Rethinking a drug treatment failure on a traditional ALS target. Exp Neurol. 2009 Apr;216(2):254-7. Abstract. Abstract

Boston-Howes W, Williams EO, Bogush A, Scolere M, Pasinelli P, Trotti D. Nordihydroguaiaretic acid increases glutamate uptake in vitro and in vivo: therapeutic implications for amyotrophic lateral sclerosis. Exp Neurol. 2008 Sep;213(1):229-37. Abstract