This is Part 1 of a two-part series. See also Part 2.
18 November 2011. Researchers studying amyotrophic lateral sclerosis and other diseases related to RNA-binding proteins gathered in Arlington, Virginia, 10-11 November 2011, to revel in, amongst other things, widespread excitement about new genes recently identified in ALS. “The last two and a half months have been the most exciting time in the history of ALS,” said Don Cleveland of the University of California, San Diego, citing the discovery of ALS-linked variants in genes in ubiquilin 2 (see ARF related news story on Deng et al., 2011) and C9ORF72 (see ARF related news story on Renton et al., 2011 and Dejesus-Hernandez et al., 2011). This Society for Neuroscience satellite symposium titled RNA-Binding Proteins in Neurological Disease was convened by Paul Taylor of St. Jude Children’s Research Hospital in Memphis, Tennessee, and Fen-Biao Gao of the University of Massachusetts Medical School in Worcester.
Not surprisingly, the C9ORF72 work, presented at the meeting by Bryan Traynor of the National Institute on Aging in Bethesda, Maryland, and Rosa Rademakers of the Mayo Clinic in Jacksonville, Florida, made the biggest splash. The victory was all the sweeter because researchers had been hunting for the chromosome 9 mutation for five years (Morita et al., 2006). The repeat expansions in C9ORF72 found to cause disease probably come from a single founder, according to a paper published the week before the identification of the gene (Mok et al., 2011). This is remarkable, because this gene defect alone accounts for a sizable chunk of ALS cases. The anticipation continued with presentations by Aaron Gitler and James Shorter of the University of Pennsylvania in Philadelphia, who are sifting for new ALS gene candidates that are similar to already known genetic factors TAR DNA binding protein 43 (TDP-43; see ARF related news story on Sreedharan et al., 2008 and Gitcho et al., 2008) and Fused-in-Sarcoma (FUS; see ARF related news story on Kwiatkowski et al., 2009 and Vance et al., 2009).
Bob Brown of the University of Massachusetts Medical School in Worcester noted that there are now 22 genes or loci associated with familial as well as sporadic ALS. These account for approximately half of ALS cases with a clear inherited origin, as well as 4 to 5 percent of sporadic cases, Brown said. While superoxide dismutase 1 (SOD1) has long been the top gene—it explains one-fifth of familial ALS—TDP-43 and FUS appear to herald a new set of genes coding for RNA-binding proteins that relate to the disease. Genomewide association studies have yielded limited gene candidates (see ARF related news story on Chiò et al., 2009). In contrast, new hypothesis-driven approaches, in which scientists look for genes akin to those already implicated in ALS, are rapidly adding to the gene list. In this two-part series, Alzforum profiles the latest genes to capture attention for ALS and related diseases.
At the satellite meeting, Gitler, who works on yeast models of TDP-43 and FUS toxicity, and Shorter, who studies their propensity for aggregation (see ARF related news story on Sun et al., 2011), noted tantalizing similarities between the two disease proteins: Both bind RNA with prion-like sequences that form aggregates when expressed in yeast and in vitro (see ARF related news story on Johnson et al., 2008). Moreover, many other RNA-binding proteins share those "prion-esque" motifs. On a list of all known RNA-binding proteins with the most classical yeast prion-like regions, FUS is number 1 and TDP-43 falls in tenth place. Researchers have been wondering about numbers two through nine. Might they also be risk factors for ALS or other neurological diseases? Or as Shorter phrased it, “Are FUS and TDP-43 the tip of the iceberg?”
It is looking like they might be: Shorter reported on numbers 2 and 3 on that list: TAF15 (TAT box binding protein [TBP]-associated factor) and EWS (Ewing sarcoma protein), respectively. FUS, EWS, and TAF15 make up the FET family, with roles in RNA transcription, processing, and transport (Law et al., 2006; Tan and Manley, 2009; Kovar, 2011). Like TDP-43 and FUS, TAF15 and EWS aggregated in the cytoplasm of yeast, where they were toxic, and all four formed pore-shaped oligomers in vitro, Shorter reported. Perhaps, he suggested, all these RNA-binding proteins form self-templating aggregates based not on classical amyloid architecture, but on their prion domains. The infectious nature of prion particles could also explain why symptoms of ALS, in people, often move sequentially from one tissue to adjacent areas. “Is it possible,” Shorter asked, “that underlying this complexity is actually a very simple prion-based transfer mechanism?”
For his part, Gitler presented data published in the November 7 Proceedings of the National Academy of Sciences USA, which also points to TAF15 as an ALS gene. Julien Couthouis, Michael Hart, and Shorter, all at the University of Pennsylvania, were co-first authors. The group started out with the knowledge that there are 213 human RNA-binding proteins that contain an RNA recognition motif, or RRM, homologous to those in TDP-43 and FUS. Gitler obtained clones for 133 of the genes and gave them to a cadre of high school students he had recruited for the summer. The students inserted these genes into yeast expression vectors, fusing them to the gene for yellow fluorescent protein.
Then, the team put each gene through its paces, looking for those that would behave like TDP-43 and FUS. Using fluorescence microscopy, the researchers determined that 80 of their candidates localized to the cytoplasm, as do TDP-43 and FUS. Of that 80, 38 were toxic to yeast. Thirteen of those 38 proteins, including TDP-43, FUS, and TAF15, contain prion-like domains.
Gitler and colleagues then focused on that part of the TAF15 gene homologous to the two FUS regions where many disease-linked mutations concentrate: the arginine-glycine-glycine (RGG) domain required for aggregation, and the carboxyl-terminal proline-tyrosine-rich motif required for nuclear localization. They sequenced TAF15 DNA from 735 people with ALS and 1,328 healthy controls. The DNA samples came from mostly Caucasian libraries at the Coriell Institute for Medical Research in Camden, New Jersey; the University of Pennsylvania; and the Mayo Clinic in Jacksonville, Florida. The team discovered three mutations (glycine-391-glutamate; arginine-408-cysteine; and glycine-473-glutamate) that were present only in cases, not controls. They identified another variant (methionine-368-threonine) in a separate, Swedish cohort, and a fifth (glycine-452-glutamate) in Australian patients. The work jibes with two further TAF15 mutations (alanine-31-threonine and arginine-395-glutamine) that Brown reported at the meeting (Ticozzi et al., 2011). Brown also mentioned discovering potential ALS mutations in ubiquilin 1, a ubiquilin 2 homolog.
Furthermore, Gitler and colleagues transfected TAF15 into rat embryonic spinal cord neurons and discovered that the ALS-linked mutations promoted the formation of TAF15 cytoplasmic inclusions. TAF15 also aggregated in Shorter’s in-vitro assays, à la TDP-43 and FUS, with disease-linked mutations promoting aggregation. And in collaboration with Nancy Bonini, also at the University of Pennsylvania, the team determined that TAF15 upregulation caused neurodegeneration and death in fruit flies. Finally, working with UPenn colleagues Virginia Lee and John Trojanowski, the team showed that TAF15 is mislocalized to cytoplasmic inclusions in postmortem spinal cord tissue from people who had sporadic ALS. These inclusions were distinct from those that contain TDP-43. Mutations in RNA-binding proteins may work to dysregulate RNA metabolism and cause disease, Gitler suggested. Whether they do so individually or in concert is not clear yet. Gitler invited other researchers to investigate the RNA-binding proteins on his list that contain prion domains.
Gregory Petsko of Brandeis University in Waltham, Massachusetts, who attended the meeting and also works with yeast models of TDP-43 and FUS toxicity (see ARF related news story on Ju et al., 2011), commended the Gitler-Shorter approach to ALS gene discovery: “I think that RNA-binding proteins with prion-like domains are going to be potential genetic modifiers or causes of ALS. This makes a tremendous amount of sense,” Petsko told ARF.
While researchers such as Gitler are starting with a gene list, others are starting with ALS-linked protein aggregates to work backward to the genes involved in disease. Their hits include new RNA-binding proteins and players related to protein degradation. For that, read upcoming Part 2 of this report.—Amber Dance.
Couthouis J, Hart MP, Shorter J, DeJesus-Hernandez M, Erion R, Oristano R, Liu AX, Ramos D, Jethava N, Hosagadi D, Epstein J, Chiang A, Diaz Z, Nakaya T, Ibrahim F, Kim HJ, Solski AJ, Williams KL, Mojsilovic-Petrovic J, Ingre C, Boylan K, Graff-Radford NR, Dickson DW, Clay-Falcone D, Elman L, McCluskey L, Greene R, Kalb RG, Lee VM, Trojanowski JQ, Ludolph A, Robberecht W, Andersen PM, Nicholson GA, Blair IP, King OD, Bonini NM, Van Deerlin V, Rademakers R, Mouelatos Z, Gitler AD. A yeast functional screen predicts new candidate ALS disease genes. Proc Natl Acad Sci U S A. 2011 Nov 7. Abstract
This is Part 1 of a two-part series. See also Part 2.