13 December 2008. Even in the face of mutated Cu/Zn superoxide dismutase (SOD1), which causes amyotrophic lateral sclerosis in people and in mice, sensitive motor neurons can get by with a little extra help from their antioxidant-pumping astrocyte friends. That’s the conclusion by scientists in the laboratory of Jeffrey Johnson at the University of Wisconsin-Madison, writing in the 10 December Journal of Neuroscience. They found that overexpressing an antioxidant booster called nuclear erythroid 2-related factor 2 (Nrf2) selectively in astrocytes protected motor neurons in culture, and extended the survival of mSOD1 mice by three weeks. The likely mechanism is that Nrf2 causes astrocytes to secrete more of the antioxidant glutathione, which motor neurons can scavenge for parts to then synthesize their own glutathione. The neurons can then protect themselves from oxidative damage.
Nrf2 is a transcription factor that, during times of oxidative stress, travels to the nucleus and activates genes by binding the antioxidant response element (ARE) located upstream of several genes for cell-protective proteins (reviewed in Lee and Johnson, 2004). Nrf2 has been shown to protect neurons from acute injury in culture (Shih et al., 2003; Kraft et al., 2004; Vargas et al., 2006) and in vivo (Calkins et al., 2005), but this is the first demonstration that increased Nrf2 activity can prevent or delay motor neuron degeneration in a chronic disease model. Johnson called the effect of extra Nrf2 “programmed cell life,” in contrast to the programmed cell death of apoptosis. By overexpressing Nrf2, he suggested, the equilibrium between life- and death-promoting signals shifts toward life.
Several SOD1 mutations lead to motor neuron degeneration and death. They account for 10 to 20 percent of familial ALS cases, and human SOD1 mutations cause similar symptoms in mouse models. Astrocytes containing mSOD1 are toxic to wild-type motor neurons (Nagai et al., 2007 and see ARF related news story). Johnson and colleagues investigated whether Nrf2 overexpression in mSOD1 astrocytes could stave off motor neuron death.
To amplify the Nrf2-ARE pathway in astrocytes, first author Marcelo Vargas and colleagues engineered mice that express Nrf2 under the astrocyte-specific promoter for glial fibrillary acidic protein (GFAP). Spinal cord astrocytes from those mice had a 2.5-fold increase in Nrf2 messenger RNA levels and were able to withstand higher concentrations of the toxin tert-butyl hydroperoxide than cells from non-transgenic animals. The transgenic cells upregulated production of ARE-influenced genes and made twice as much glutathione as did control cells.
More astrocyte glutathione leads, indirectly, to more motor neuron glutathione because neurons depend on astrocytes for their supply of its precursor peptides (reviewed in Dringen et al., 2000). The tripeptide glutathione is the product of two enzymes, glutamate-cysteine ligase and glutathione synthetase. Nrf2 upregulates both genes, amping up glutathione production. Astrocytes pump glutathione into the extracellular space via the multidrug resistance-associated protein 1 (Mrp1). For their part, motor neurons cannot take up glutathione from the extracellular space directly. Instead, they use membrane-bound enzymes to break down glutathione and then import the raw materials for their own glutathione synthesis.
Vargas tested whether the beefed-up glutathione supply chain could help motor neurons in culture, growing murine motor neurons together with astrocytes from mSOD1 mice that either carried or lacked the GFAP-Nrf2 construct. Single mutant mSOD1 astrocytes reduced the survival of cultured neurons, as expected, by 40 percent over three days. In co-cultures with the enhanced Nrf2 expression, the toxicity of mSOD1 astrocytes vanished. In addition, RNA-mediated silencing of the glutathione transporter gene Mrp1 removed the protective effect, confirming that Nrf2-expressing astrocytes help motor neurons by secreting glutathione.
The mice corroborated this in-vitro data. In GFAP-Nrf2/mSOD1 animals, the disease set in 17 days later than it did in single mutant mSOD1 animals. However, once symptoms began, disease proceeded similarly in both mouse lines; the double mutant mice lived an average of 20.5 days longer.
“It’s probably the cleanest evidence we have so far that astrocytes actually contribute to antioxidant protection of neurons,” said Raymond Swanson of the University of California, San Francisco, who was not involved with the study. Cell culture studies are artificial systems, Swanson said, but “They are looking at normal neuroanatomy in the animal.”
Even so, these animal studies leave open a few questions. It is possible, Swanson said, that instead of helping motor neurons to protect themselves, the added Nrf2 reduces the toxic effects that mSOD1 astrocytes have on neurons. Another explanation, Johnson said, is that another ARE gene is protective, but its effect is overshadowed by the increased glutathione levels. The Wisconsin group is working to prove that glutathione secretion is the root of the protective effect in mice whose astrocytes overexpress Nrf2. Direct administration of glutathione protects motor neurons in culture, Vargas said, but is toxic to mice, ruling out the most obvious experiment. Instead, the researchers are studying transgenic mice that do not increase glutathione production in response to Nrf2.
To be sure, this study does not discern the true killer of motor neurons in ALS. After a three-week reprieve, the mice still die. Instead, the GFAP-Nrf2 addition seems to augment the body’s own defenses by beefing up the antioxidant capacity of neurons. Astrocytes normally increase Nrf2 production at the onset of symptoms in mSOD1 mice (Vargas et al., 2005). “I think the body naturally tries to start this response, but it’s not strong enough and eventually gets overwhelmed,” Vargas said.
The scientists suggested that Nrf2 should be a target for ALS therapeutics, and the group is currently screening for drugs that activate Nrf2. Many do, Vargas said, but few, if any, cross the blood-brain barrier. Johnson and colleagues also suspect that Nrf2 could be protective in other neurodegenerative diseases, and are testing Nrf2’s effects in models of Alzheimer and Parkinson disease.
Enhanced Nrf2 activity could have side effects, though the transgenic mice showed none. “My guess is that you would never get cancer or autoimmune disease,” Johnson mused. His hunch that the GFAP-Nrf2 mice will age gracefully—retaining the youthful brain of a six-month-old into the ripe old mouse age of two years—is being put to the test in a longitudinal learning and memory study; alas, the answer will take those two years to come in.—Amber Dance.
Vargas MR, Johnson DA, Sirkis DW, Messing A, Johnson JA. Nrf2 activation in astrocytes protects against neurodegeneration in mouse models of familial amyotrophic lateral sclerosis. J. Neurosci. 2008 December 28(50):13574-13581.