For the first time, researchers investigating the role of superoxide dismutase 1 in amyotrophic lateral sclerosis have a rodent model that does not grossly overexpress the gene. Scientists have characterized mice with an ALS-linked mutation in their endogenous SOD1 gene—a substitution of glycine for the aspartic acid at position 83. According to a paper in the December 2 Human Molecular Genetics, mice homozygous for the mutation developed a disease akin to early stage ALS, with dying motor neurons and degenerating neuromuscular junctions. However, the animals never become paralyzed, a classic ALS hallmark. Though they do not model late-stage disease, these animals might provide clues to what kills motor neurons in ALS, suggested senior author Abraham Acevedo-Arozena of the Medical Research Council (MRC) in Harwell, U.K.

Several different mutations in SOD1 cause ALS. Scientists have created more than a dozen models by inserting different mutant versions of the human gene into the mouse genome. All have the limitation that they overexpress the transgene, some with as many as two dozen copies. If transcribed at such high levels, even wild-type SOD1 causes motor neuron disease (Graffmo et al., 2013). Studying these mice has yielded no treatments that pass muster in human trials, said Daryl Bosco of the University of Massachusetts Medical Center in Worcester. Another option would be welcome, said Bosco, who was not involved in the new study. Likewise, drugs developed using mouse models of Alzheimer's disease have done poorly in clinical trials, and researchers question the reliance on animals that overexpress disease-linked proteins. Knocking in genes may prove a better approach (see Apr 2014 Webinar).


In healthy neuromuscular junctions of 15-week-old SOD1-D83G mice (left), axons (green) terminate onto muscle endplates (red). However, by one year of age the axons have abandoned the muscle in many junctions (right). [Image courtesy of Abraham Acevedo-Arozena, Medical Research Council, Harwell, U.K.]

First author Peter Joyce of the MRC sought an alternative by hunting for SOD1 mutations in mice treated with a chemical mutagen (Quwailid et al., 2004). Researchers used N-ethyl-N-nitrosourea (ENU) to randomly generate genetic errors across the genome of 10,000 animals, then archived their DNA and sperm. Screening that DNA, Joyce found one sample with the SOD1-D83G substitution and used those sperm to generate his new line. Working with co-senior authors Linda Greensmith and Elizabeth Fisher of University College London, he and Acevedo-Arozena are mining the ENU archive for mutations in other ALS genes, such as FUS and TDP-43 (Ricketts et al., 2014).

This random mutagenesis technique means the SOD1-D83G sperm also contained other mutations that might contribute to or confound the phenotype. The researchers crossed the animals with the background C57BL/6 strain several times, eliminating half of those with each generation. However, whole-genome sequencing indicated that even so, several mutations remained in the animals Joyce used for his experiments. These should not affect the results, Acevedo-Arozena said, because most of the mutations are not genetically linked to the SOD locus, and were shuffled separately from SOD1-D83G when the animals bred. While the mice would be congenic for zero, one, or two copies of mutant SOD1, each animal would have its own hodgepodge of other mutations, hence these do not explain how all the SOD1-D83G animals had the same motor disease phenotype.

The D83G mutation alters SOD1’s zinc-binding site, destabilizing the protein. The researchers expected D83G to be pathogenic because it occured in a family afflicted with ALS. Most of the affected members died six to 12 months after diagnosis, compared to the average two or three years. In two relatives who underwent detailed clinical examination, the disease started with lower motor-neuron symptoms (Millecamps et al., 2010).

Joyce examined both heterozygous and homozygous SOD1-D83G mice. Despite some problems, such as trouble running in a wheel and elevated mitochondrial-membrane potential in their motor neurons, the heterozygotes moved fairly normally and lived a full lifespan. The homozygous animals were sicker. This differs from human ALS, where one SOD1 mutation causes disease. 

In the homozygous D83G mice, motor problems appeared early. Even at six weeks, their grip was weak. By five months, the tremors started, and over the ensuing months, their backs hunched. They dragged their hindquarters but never became paralyzed. U.K. guidelines required the scientists to euthanize animals once they lost 20 percent of their maximum body weight, which for the homozygotes was at about 16.5 months, compared to nearly two years for both the wild-type mice and SOD1-D83G heterozygotes. Joyce found the livers of the homozygotes riddled with tumors, indicating they probably would have died of liver cancer. These tumors likely resulted from the lack of normal SOD1, rather than the D83G substitution or other point mutations caused by ENU, since liver tumors plague SOD1 knockout mice as well (Elchuri et al., 2005). The liver in most SOD1-D83G heterozygotes appeared normal.

In keeping with the loss of muscular control, the scientists found pathological evidence for early stage motor neuron disease. Between six and 15 weeks, nearly a quarter of the lower-motor-neuron cell bodies in the lumbar spinal cord disappeared. Unlike in ALS, this neurodegeneration did not seem progressive, because no more motor neurons had died by one year. “We always assumed that we needed to overexpress SOD1 to get motor neuron degeneration, but that is not the case,” Acevedo-Arozena said.

However, losing even that many motor neurons would not likely cause the noticeable disabilities that emerged over the next 37 weeks, he said. Rather, he suspects the movement problems resulted from denervation of the neuromuscular junctions. At 15 weeks, these connections between nerve and muscle in the front of the hindlimbs looked fine. Then, over time, the motor neuron axons that innervate the junctions pulled away, so that by one year, 15 percent of junctions were denervated (see image above). During this same time period, the animals developed their motor symptoms, with tremors starting at 20 to 22 weeks and sagging hindquarters at 32 to 39 weeks.

How did the mutant SOD1 do this? In people, mutant SOD1aggregates to form protein inclusions, but Joyce and colleagues found neither insoluble SOD1 nor aggregates in the spinal cords of their animals. In fact, the SOD1-D83G mice contained only 12 percent of the normal amount of SOD1. This may be because the enzyme was unstable without zinc. Acevedo-Arozena suspects that the denervation resulted from the lack of the normal dismutase, rather than accumulation of the mutant, because SOD1 knockouts have this denervation, as well (Flood et al., 1999). However, the knockouts have no motor-neuron death, hence he surmised that this aspect of the SOD1-D83G mouse must result from the small amount of mutant SOD1 present in the animals.

Since the neurodegeneration in the mice appears to arrest in an early phase, Acevedo-Arozena hopes to use the model to understand why only some motor neurons are susceptible. That might provide ideas for potential treatments, he said. Bosco also thought this model might be useful to test therapeutics using endpoints other than simple survival. The animals will be available from Jackson Laboratory in Bar Harbor, Maine, Acevedo-Arozena said.—Amber Dance


No Available Comments

Make a Comment

To make a comment you must login or register.


Webinar Citations

  1. Good-Bye Overexpression, Hello APP Knock-in. A Better Model?

Paper Citations

  1. . Expression of wild-type human superoxide dismutase-1 in mice causes amyotrophic lateral sclerosis. Hum Mol Genet. 2013 Jan 1;22(1):51-60. PubMed.
  2. . A gene-driven ENU-based approach to generating an allelic series in any gene. Mamm Genome. 2004 Aug;15(8):585-91. PubMed.
  3. . A nonsense mutation in mouse Tardbp affects TDP43 alternative splicing activity and causes limb-clasping and body tone defects. PLoS One. 2014;9(1):e85962. Epub 2014 Jan 21 PubMed.
  4. . SOD1, ANG, VAPB, TARDBP, and FUS mutations in familial amyotrophic lateral sclerosis: genotype-phenotype correlations. J Med Genet. 2010 Aug;47(8):554-60. Epub 2010 Jun 24 PubMed.
  5. . CuZnSOD deficiency leads to persistent and widespread oxidative damage and hepatocarcinogenesis later in life. Oncogene. 2005 Jan 13;24(3):367-80. PubMed.
  6. . Hindlimb motor neurons require Cu/Zn superoxide dismutase for maintenance of neuromuscular junctions. Am J Pathol. 1999 Aug;155(2):663-72. PubMed.

External Citations

  1. SOD1 
  2. FUS 
  3. TDP-43

Further Reading


  1. . Design, power, and interpretation of studies in the standard murine model of ALS. Amyotroph Lateral Scler. 2008;9(1):4-15. PubMed.
  2. . Genome-wide association analysis reveals a SOD1 mutation in canine degenerative myelopathy that resembles amyotrophic lateral sclerosis. Proc Natl Acad Sci U S A. 2009 Feb 1;106(8):2794-9.
  3. . C57BL/6J congenic Prp-TDP43A315T mice develop progressive neurodegeneration in the myenteric plexus of the colon without exhibiting key features of ALS. Brain Res. 2014 Oct 10;1584:59-72. Epub 2013 Oct 18 PubMed.
  4. . Disulphide-reduced superoxide dismutase-1 in CNS of transgenic amyotrophic lateral sclerosis models. Brain. 2006 Feb;129(Pt 2):451-64. PubMed.
  5. . Motor neuron degeneration in mice that express a human Cu,Zn superoxide dismutase mutation. Science. 1994 Jun 17;264(5166):1772-5. PubMed.
  6. . A comprehensive assessment of the SOD1G93A low-copy transgenic mouse, which models human amyotrophic lateral sclerosis. Dis Model Mech. 2011 Sep-Oct;4(5):686-700. PubMed.

Primary Papers

  1. . A novel SOD1-ALS mutation separates central and peripheral effects of mutant SOD1 toxicity. Hum Mol Genet. 2015 Apr 1;24(7):1883-97. Epub 2014 Dec 2 PubMed.