The most common genetic cause of two neurodegenerative disorders, amyotrophic lateral sclerosis (ALS) and frontotemporal dementia (FTD), is the mutation of a gene called C9orf721. The mutation involves expansion of a section of DNA in which six bases — four guanine bases followed by two cytosine — are repeated many times. Much research has focused on the hypothesis that this alteration produces rogue RNAs or proteins that are toxic to cells. However, the mutation also lowers the amount of normal C9orf72 protein, which might contribute to disease. Writing in Nature, McCauley et al.2 report a mechanism by which reduced C9orf72 function causes defects in the immune system. Their findings might explain why people with ALS and FTD who harbour the C9orf72 mutation seem to be susceptible to autoimmune disorders3.
There are several mouse strains that lack the C9orf72 protein. These mice do not develop ALS- or FTD-like traits, but instead exhibit enlarged spleens and lymph nodes, as well as other defects related to the immune system4–7. McCauley et al. hypothesized that these immune defects might be caused by loss of C9orf72 function in a subset of immune cells called myeloid cells. They confirmed this idea by generating a strain of mouse in which C9orf72 is deleted in myeloid cells, but remains intact in other cells. These mice exhibited similar defects to those of the original C9orf72-deficient animals.
McCauley and colleagues went on to ask exactly how loss of C9orf72 caused these immune defects. The authors measured RNA levels from immune cells of mice that completely lacked C9orf72. They found a striking increase in messenger RNAs encoding both type I interferon-β protein and interferon-stimulated genes (expression of which is activated by interferon proteins) in the mutant cells, compared with controls. Type I interferons are normally secreted in the body to help ward off infection by modifying the activity of the immune system. These findings therefore pinpoint the specific immune defect caused by loss of C9orf72 — an increased type I interferon response.
Next, McCauley et al. cultured immune cells from the bone marrow of wild-type and C9orf72-mutant mice and stimulated them with several known activators of the type I interferon response. Almost all of the molecules they tested elicited similar responses in both mutant and control cells. But one, cyclic GAMP (cGAMP), induced a hyperactive type I interferon response in the mutant cells. cGAMP is an activator of the cGAS–STING signalling pathway, which senses double-stranded DNA in the cell cytoplasm as part of the immune system’s first line of defence against viral infection8.
Why would loss of C9orf72 function result in increased STING protein activity? STING is normally degraded by a cellular process called autophagy9, and the regulation of this process is one of the functions of C9orf7210. The researchers therefore reasoned that STING degradation might be impaired in the C9orf72-mutant immune cells (Fig. 1). In support of this idea, they showed that the breakdown of STING was delayed in the bone-marrow cells of C9orf72 mutants, compared with those from wild-type animals.
These findings raise the possibility that reducing STING activity could eradicate the defects associated with C9orf72 deficiency. McCauley et al. inhibited STING activity in mice and cells lacking C9orf72, using drugs or by introducing a mutation in the gene that encodes STING. These approaches restored normal type I interferon responses and spleen sizes in the mutant mice. McCauley and colleagues’ work therefore identifies STING as a potential therapeutic target.
Finally, the authors applied their findings to ALS in humans. When they performed RNA sequencing on cells from people with ALS who harbour C9orf72 mutations, they again found an enhanced type I interferon signature. The researchers also showed that this response is, at least partly, mediated by STING, because treatment with a chemical STING inhibitor suppressed this enhanced interferon response. These findings might help to explain why C9orf72 mutations are over-represented in the few people who have a combination of ALS and the autoimmune disorder multiple sclerosis, compared with the population of people with ALS as a whole11.
McCauley and colleagues’ findings also raise the exciting possibility that pharmacologically inhibiting STING could decrease the risk of autoimmunity in people who have ALS. Indeed, small-molecule inhibitors of STING have been developed, and seem safe and effective in preclinical models of autoinflammatory disease12.
With many studies suggesting that toxic RNAs or proteins are produced by the C9orf72 repeat expansion, considerable work has focused on reducing the levels of these products. For instance, antisense oligonucleotides (short nucleic acids that bind to RNAs containing the C9orf72 repeat and trigger their degradation) have shown promising results in preclinical studies7,13. A clinical trial of one such molecule is under way in people who have ALS (see go.nature.com/3g0cpat). If the trial is successful, it will be exciting to see whether boosting levels of normal C9orf72 or mitigating the effects of reduced C9orf72 function — perhaps by targeting the STING pathway — will have added benefits.
Nature 585, 34-35 (2020)