The misfolded enzyme that causes some kinds of amyotrophic lateral sclerosis poisons motor neurons by plugging up pores in their mitochondria, report researchers from the University of California, San Diego, in the August 26 issue of Neuron. Mutations in superoxide dismutase 1 (SOD1) are responsible for some cases of inherited ALS, and first author Adrian Israelson and colleagues from Don Cleveland’s laboratory used rats and mice carrying these mutations to show that the mutant protein shuts down the voltage-gated anion channel VDAC1, also called mitochondrial porin. With that portal locked, less ADP gets into the mitochondria, so they produce less ATP, and that energy loss could damage cells, the authors suggest.
The scientists knew mutant SOD1 interacted with the outer mitochondrial membrane (Vande Velde et al., 2008), so they sought specific proteins in that membrane that bind with the dismutase. “This becomes particularly relevant given the importance of mitochondria in both energy metabolism and apoptosis regulation,” wrote Mohanish Deshmukh of the University of North Carolina in Chapel Hill, who was not involved in the study, in an e-mail to ARF. VDAC1 transports nucleotides as well as ions across the mitochondrial membrane, and is also involved in regulating apoptosis. The interaction between mutant SOD1 and VDAC1 seems to happen specifically in the spinal cord, perhaps explaining why spinal motor neurons are particularly susceptible to ALS pathology.
Israelson and colleagues isolated mitochondria from transgenic rats carrying either wild-type human SOD1, or two mutant versions that cause ALS: G93A and H46R. When they immunoprecipitated SOD1 from those mitochondria, they found that VDAC1 came along with the mutant forms, but not the wild-type protein. Similarly, if they used an antibody specific for misfolded SOD1, they discovered it brought down VDAC1 from mitochondria isolated from the spinal cord. Therefore, something about malformed SOD1 makes it able to interact with VDAC1.
To see if this binding has any functional consequences, the researchers purified VDAC1 from the spinal cords of normal rats and reconstituted the protein in a lipid bilayer to create an in vitro version of the outer mitochondrial membrane. Then they added recombinant SOD1—wild-type or mutant—and measured the voltage across the membrane as the channels pumped ions. Wild-type SOD1 did not affect VDAC1 channel conductance, but the mutants diminished it. Mutant SOD1 also blocked transport across the channel. Mitochondria isolated from the spinal cords of rats carrying SOD1 mutations took up 25 percent less ADP than mitochondria from normal animals and, presumably, they also had less ATP output. “There is an interaction of mutant SOD1 with VDAC1, and this affects the conductance of these channels, leading to less energy in the mitochondria,” Israelson concluded.
Finally, the researchers studied the effects of the VDAC1-SOD1 interaction in the disease course of mutant mice. They crossed mice carrying human SOD1-G37R with mice carrying a null VDAC1 allele. The resulting double mutants, harboring SOD1 as well as one null copy of VDAC1, died nearly two months before their SOD1 single mutant counterparts. Therefore, the double whammy of low VDAC1 levels, and VDAC1 inhibition by SOD1, makes the disease worse than SOD1’s inhibition of the channel alone.
The authors suggest that by binding VDAC1, misfolded SOD1 stresses the cell’s energy system. This mitochondrial dysfunction could lead to formation of reactive oxygen species or otherwise stress the cell, making it more susceptible to further damage. In a commentary accompanying the paper, Virginia Le Verche and Serge Przedborski of Columbia University in New York called the idea that ALS could be a mitochondrial channelopathy an “exciting and novel hypothesis.” Israelson and colleagues also note that there are many ways neurons suffer damage in ALS (reviewed in Ilieva et al., 2009), and this is likely one of many mechanisms.
The research leaves many questions open. For one, the researchers found no SOD1-VDAC1 interaction in liver or brain mitochondria. “Exactly why the mutant, but not wild-type, SOD1 associates with VDAC1, and why this association seems selective to mitochondria from spinal cord are interesting directions that remain unclear,” Deshmukh wrote. One possibility, the study authors suggest, is that mitochondria in different tissues have different protein makeups. They note that hexokinase, which also interacts with VDAC1, is more prevalent in brain than spinal cord (Azoulay-Zohar et al., 2004), and could perhaps outcompete SOD1 for the porin.
The Cleveland lab has produced plenty of data suggesting the ALS is not simply a disease of motor neurons, but that glia (see ARF related news story on Lobsiger et al., 2007; Nagai et al., 2007; Di Giorgio et al., 2007), and even Schwann cells (see ARF related news story on Lobsiger et al., 2009), are involved. Since they prepared mitochondria from whole spinal cords, they cannot tell which cell types host the SOD1-VDAC1 interaction. “We think it is not specific for one type of cell,” Israelson said. They also cannot tell, from these experiments, whether VDAC1 might be involved in ALS not caused by SOD1 mutations. Nevertheless, if these results are confirmed, Le Verche and Przedborski note that VDAC1 could be a potential drug target in ALS. Glucose 6-phosphate opens VDAC1 channels (Azoulay-Zohar et al., 2004), for example, and may provide neuroprotection, they suggest. Sweet as it may sound, that theory needs rigorous testing in the laboratory before scientists think about trying it in the clinic.—Amber Dance