9 January 2013. As if the genetic basis for amyotrophic lateral sclerosis was not complicated enough, attendees at the International Symposium on ALS/MND, 5-7 December 2012 in Chicago, Illinois, heard another wrinkle to the story. Michael van Es, of the University Medical Center in Utrecht, the Netherlands, proposed that in certain familial cases, not one but two major ALS risk genes contribute to disease. After screening families for known mutations, Van Es and his colleagues discovered far more double mutants than they would expect by chance alone. Mutations in one gene, a second, or both might explain why carriers in the same family often present with different phenotypes, or escape disease altogether, suggested van Es. The mechanism behind the dual-gene pathology remains unclear, and some scientists even doubt that mutant pairs work together.

Geneticists have typically explained the variable age at onset, speed of progression, and symptoms of people with the same ALS mutation by hypothesizing that other factors in their genomes, perhaps many variants, influence the disease. The Utrecht group’s “oligogenic” hypothesis is not that. It blames ALS phenotypes on two genetic co-stars, instead of one headliner and a supporting cast of bit players (Van Blitterswijk et al., 2012).

Van Es’ colleague Marka van Blitterswijk headed the study before moving to Rosa Rademakers' lab at the Mayo Clinic in Jacksonville, Florida. In an interview with Alzforum, she said the geneticists happened upon people carrying two ALS risk alleles when they set out to perform exome and whole genome sequencing. They checked their samples for known ALS mutations first, and discovered that five of their 97 families carried pairs of mutations. The angiogenin variant K17I showed up with FUS-R521C or TARDBP-N352S, a variant of the TDP-43 gene. C9ORF72 expansions paired with TARDBP-N352S, SOD1-D90A, or FUS-Q210H. In a further study published last month, Van Blitterswijk reported on a C9ORF72 expansion/VAPB-V234I duo (Van Blitterswijk et al., 2012).

In healthy control genomes, variants of known ALS genes only appeared in 0.5 percent of samples, and never more than one in a given person. The researchers calculated that the frequency of double mutations in families with ALS was well above the rate that should occur by chance.

However, the oligogenic interpretation is complicated by the fact that not all of the implicated variants are confirmed as pathogenic mutations. For example, FUS-Q210H and angiogenin-K17I also occurred individually in control subjects. Thus, they might not be causative mutations, but could still represent risk factors or disease modifiers, Van Blitterswijk suggested in an e-mail to Alzforum. Researchers debate the role of angiogenin as an ALS risk factor (see ARF related news story on Greenway et al., 2006; Corrado et al., 2007; Kirby et al., 2012), but several of the other mutations crop up in previously reported ALS cases. This includes FUS-R521C (Sproviero et al., 2011) and TARDBP-N352S, which was fairly common among Van Blitterswijk’s pedigrees (Kühnlein et al., 2008; Kamada et al., 2009). C9ORF72 expansions, recently discovered to be a long-sought risk factor, are prevalent in cases as well (see ARF related news story on Renton et al., 2011, and Dejesus-Hernandez et al., 2011). And SOD1-D90A is among the most frequent disease-causing SOD1 mutations globally (Rabe et al., 2010; Giannini et al., 2010; Eisen et al., 2008).

Other researchers have also reported cases with double mutations (Luigetti et al., 2011; Chiò et al., 2012; Lattante et al., 2012; Millecamps et al., 2010). Van Es said he is unsure how common pathogenic mutation pairs are overall. He hypothesized that the additive effects of two genetic risk alleles cause full-blown ALS in some families, while a single mutant allele may lead to different ALS or FTLD phenotypes, which may manifest as later onset of disease or slower progression. “It is more the sum of mutations that determines what the phenotype looks like, rather than individual mutations causing different phenotypes,” Van Es said. Importantly, that does not mean all inherited ALS cases result from a genetic double whammy, just that it is possible in some families, Van Blitterswijk noted.

Van Blitterswijk has taken the work a step further in her new position. Rademakers' group studies frontotemporal dementia (FTD), which shares several genes and features with ALS. As outlined in a poster at the Chicago conference, Van Blitterswijk examined DNA from people with known mutations in ALS and FTD genes, such as tau and progranulin, for the presence of the newly discovered C9ORF72 expansion. Again, she observed an unexpectedly high rate of double mutants—out of 218 families, three had a C9ORF72 expansion plus progranulin variants, and one had the C9ORF72 variant plus a tau mutation. The progranulin mutations included two frameshifts and an R493X substitution; the tau was P301L. All of these are linked to FTD beyond a doubt, Van Blitterswijk wrote (Huey et al., 2006; Spina et al., 2007; Pickering-Brown et al., 2008; Nasreddine et al., 1999). “This confirms the oligogenic pathogenesis and shows that it is not just specific for ALS; it can be detected in FTD patients as well,” Van Blitterswijk said.

Reaction from scientists has been mixed. Not all have embraced the hypothesis. Rademakers told Alzforum that she was initially skeptical, but now that she has seen how frequently mutations co-occur, she agrees that the pairs constitute a real phenomenon.

Guy Rouleau, of McGill University in Montréal, Canada, sees it differently. He argued that one of the two mutations is not a pathogenic mutation, but simply a benign variant. Just because the polymorphisms are present does not mean they contribute to disease, he told Alzforum, noting that every person’s genome is littered with variants that do not cause problems.

The presence of two mutations in so many ALS and FTD families has major implications for both researchers and the families, the presenters said. Scientists often eliminate carriers of known genes from new genetic analyses, but that would be a mistake if those carriers have more important mutations lurking in their genomes that researchers want to find. “I would advise we include [samples with known mutations] in any future screening to find evidence for double mutations,” Van Es said at the meeting.

Double mutations will complicate genetic counseling and risk prediction for families that harbor risk alleles. “I really do not know how we are going to work around this in clinical practice,” Van Es told Alzforum. “We do not know what to tell them anymore.” Already, Van Blitterswijk said she knows of one family that was seeing a genetic counselor, thinking they had only the angiogenin K17I variant to contend with (Van Es et al., 2009). Further testing showed that they also carried the TARDBP-N352S mutation.—Amber Dance.