18 October 2013. Known genetic variants explain only 75 percent of familial amyotrophic lateral sclerosis cases, leaving more ALS genes to find. However, since the largest ALS families already have been scoured by gene-hunters, researchers now resort to studying smaller kindreds. In the October 10 American Journal of Human Genetics online, researchers from Japan reported that one such family led them to a new ALS gene candidate. Members of the family harbor mutations in a gene that goes by the mouthful v-erb-a erythroblastic leukemia viral oncogene homolog 4 (ErbB4). The small size of the family means the genetic statistics are less than a slam-dunk, but finding two unrelated cases of ALS with mutations in the same gene strengthens the connection, said senior author Shoji Tsuji of the University of Tokyo.
First author Yuji Takahashi, who has since moved to the National Center of Neurology and Psychiatry in Tokyo, and colleagues first heard about the kindred from the family's primary-care physicians. Three of the eight children in the family had ALS, and the father likely had it, too, although he was never diagnosed before he died at the age of 74. Tsuji and colleagues checked one son with ALS for known genetic causes, but finding none, they decided to look for novel mutations.
First, Tsuji and colleagues genotyped single-nucleotide polymorphisms in the mother, whom they deemed free of the disease gene, and all but one of the siblings (the one they did not test had never been diagnosed with ALS, but he had died so DNA was unavailable). Linkage analysis was difficult because the group was so small, and because any of the young, healthy siblings might be a carrier yet to develop symptoms. To determine linkage, the researchers had to assume 80 percent penetrance for any novel genetic variant—that is, they assumed at least one of the healthy siblings was actually a carrier. This analysis led them to seven genetic regions that associated with the disease with a logarithm of odds (LOD) score of 0.7 or higher. This is not a very high LOD score, noted John Landers of the University of Massachusetts Medical School in Worcester, who was not part of the study group. The National Human Genome Research Institute says a LOD score of three or higher typically indicates linkage. Tsuji conceded the statistics were weak, but said it was the best the researchers could expect when working with a small family.
Next, the researchers used whole-genome sequencing to look for novel variants in seven regions. This led them to three candidates that were predicted to change amino acid sequences. Of the three, only the ErbB4 variant, an arginine-for-glutamine swap at amino acid 927 (R927Q), was absent in 477 control samples collected in Japan, North America, and Australia.
While it is certainly plausible that the ErbB4 variant causesALS in this family, it remains plausible that something else is to blame, Landers said. He noted that the National Heart, Lung, and Blood Institute’s Exome Variant Server lists dozens of variants in ErbB4, and it is possible Tsuji and colleagues had discovered one that segregated with disease by chance, while missing out on the true ALS mutation.
Landers did agree that subsequent analyses boost the chance that ErbB4 is an ALS gene. Tsuji and colleagues checked ErbB4 in 364 people with familial ALS, and 818 with sporadic ALS, from the same geographic locales as the controls. The same R927Q variant appeared in a Canadian familial ALS case, but no DNA from that person’s family was available to check if the variant segregated with disease. In addition, a Japanese person with sporadic ALS carried an arginine-1275-tryptophan variant; this must have been a new mutation because it did not turn up in either parent. Though Tsuji conceded the linkage analysis could be stronger, he felt these additional cases confirmed the association between ErbB4 and ALS. The affected amino acids are conserved across species, indicating their importance for ErbB4 function.
Philip Wong of Johns Hopkins University in Baltimore, Maryland, who was not involved in the study, agreed that the small size of the family weakened the human genetics. What would really clinch the ALS connection, he suggested, would be a mouse model in which the ErbB4 mutations led to motor neuron disease. No such model exists, though a complete knockout of ErbB4 kills mouse embryos before birth, causing both cardiac and neural defects (Gassmann et al., 1995). Researchers also have made a conditional knockout mouse for ErbB4 in the nervous system, which exhibited weak motor defects (Golub et al., 2004 ).
The receptor is expressed in nervous system gray matter and spinal motor neurons. Neuregulins bind ErbB4, initiating autophosphorylation downstream signaling. By adding neuregulin to cultured monkey kidney cells, Tsuji and colleagues observed that the ALS-linked ErbB4 mutants exhibited less autophosphorylation than the wild-type receptor on a Western blot. However, this single experiment alone offers fairly minimal data, and the researchers did not quantitate the amount of phosphorylation, Landers noted. Neuregulin-ErbB4 signaling promotes myelination, which Wong speculated might explain the link between the receptor and ALS (see Alzforum News story on Roy et al., 2007 and Alzforum News story on Michailov et al., 2004).
Neuregulins are a substrate for BACE1, which cleaves the protein to release the functional signaling fragment. Might BACE inhibitors compromise motor neurons? Probably not, Wong said. For one, only neuregulin 1 type III is a BACE1 substrate, and other neuregulins might still activate ErbB4. In addition, BACE1 knockout animals do not exhibit motor neuron disease, he noted. “I personally do not think that BACE1 inhibitor will cause ALS-like effect,” Riqiang Yan of the Cleveland Clinic in Ohio wrote in an email to Alzforum.
ErbB4 variants also have been linked to schizophrenia ( Agim et al., 2013, Stefansson et al., 2002), and the receptor kinase is expressed at high levels in apoptotic neurons of people with Alzheimer’s (Woo et al., 2010). The cases in this study did not have schizophrenia or Alzheimer’s, displaying fairly typical ALS, Tsuji said.
Tsuji suggested that upregulating the neuregulin-ErbB4 pathway could be a therapeutic approach for ALS. Such a treatment might be applied broadly, even for people without ErbB4 mutations, he suggested, because people with sporadic ALS exhibit unusually low neuregulin 1 type III levels in their spinal motor neurons (Song et al., 2012).—Amber Dance.
Takahashi Y, Fukuda Y, Yoshimura J, Toyoda A, Kurppa K, Moritoyo H, Belzil VV, Dion PA, Higasa K, Doi K, Ishiura H, Mitsui J, Date H, Ahsan B, Matsukawa T, Ichikawa Y, Moritoyo T, Ikoma M, Hashimoto T, Kimura F, Murayama S, Onodera O, Nishizawa M, Yoshida M, Atsuta N, Sobue G, JaCALS, Fifita JA, Williams KL, Blair IP, Nicholson GA, Gonzalez-Perez P, Brown RH, Nomoto M, Elenius K, Rouleau GA, Fujiyama A, Morishita S, Goto J, Tsuji S. ERBB4 Mutations that Disrupt the Neuregulin-ErbB4 Pathway Cause Amyotrophic Lateral Sclerosis Type 19. Am J Hum Genet. 2013 Oct 8. [Epub ahead of print] Abstract