When Kevin Eggan of Harvard University decided to raise some C9ORF72-deficient mice down the road at MIT’s Broad Institute, he had little inkling of the can of worms—or microbes—that he was about to open. At Harvard, the C9-deficient mice spent short lives ravaged by a systemic autoimmunity but, lo and behold, in their environs at the Broad, the same strain of mice bustled about as long as controls. On May 13 in Nature, Eggan and colleagues tied this contrast to distinct gut microbiomes colonizing the animals in the two respective facilities. The Harvard-raised mice got better with antibiotics, or with fecal microbes from Broad-reared mice. The study helps explain why C9ORF72 knockouts engineered at different labs around the country have exhibited strangely inconsistent phenotypes, and raises questions about whether the gut microbiome has something to do with the incomplete penetrance of the C9ORF72 expansion in people.

  • In Harvard labs, C9 knockouts succumbed to an autoimmune disorder.
  • At the Broad Institute, they lived normal lifespans despite some inflammation.
  • Different gut microbiomes explained this variance.

The study implicates the microbiome as a potentially important, and previously underappreciated, modulator of ALS, wrote Eran Elinav of the Weizmann Institute of Science in Israel. “Understanding the molecular basis of host-microbiome interactions and their effects on ALS could lead to identification of new therapeutic targets of this deadly disease.”

“The variability of phenotype [of C9ORF72-deficient mice] has confused the field and has remained unresolved,” Aaron Gitler of Stanford University wrote to Alzforum. “In the current manuscript, the Eggan team seems to have solved this mystery and, in one fell swoop, also suggest a novel therapeutic strategy.”

Since tying hexanucleotide expansions in the first intron of the C9ORF72 gene to ALS/FTD in 2010, researchers have tried to work out the mechanisms involved (Sep 2010 news; Sep 2011 news). Three appear implicated: loss of function caused by hobbled expression of normal C9ORF72, gain of function related to extended repeat mRNAs, and polydipeptide repeats that arise from aberrant translation of the same. To focus on the loss of function, Eggan and several other groups generated knockouts. Across models and labs, these C9ORF72-deficient strains showed similar signs of age-related inflammation, such as swollen spleens, elevated pro-inflammatory cytokines, and an abundance of circulating immune cells. Beyond that, though, striking differences emerged. In Eggan’s mice, loss of one or both C9ORF72 alleles meant an early death, whereas another group saw that only in mice missing both alleles and another reported normal lifespan for both homo- and heterozygous knockouts (Jul 2016 news; Oct 2015 conference news; Mar 2016 news). Some researchers chalked up this variation to environmental differences, but what were they?

The Gut Factor. Variations in gut microflora influence inflammatory phenotypes in C9ORF72-deficient mice living in different facilities. [Courtesy of Fang and Hsiao, Nature News and Views, 2020.]

As it turned out, first author Aaron Burberry and colleagues decided to establish a C9ORF72-deficient mouse colony at the Broad Institute to take advantage of that facility’s set-up for behavioral studies. In a shocker, the mice at the Broad appeared normal from birth. Furthermore, they lived as long as control mice. The researchers established age-matched cohorts of mice at Harvard and the Broad to track their phenotypes. While the Harvard mice had sky-high levels of numerous pro-inflammatory cytokines in their blood, antibodies to self antigens, more neutrophils, fewer platelets, and enlarged spleens, all these phenotypes were much milder in the Broad mice, although the homozygous knockouts still had modestly enlarged spleens. By 37 weeks of age, the Harvard mice promptly fell off a rotating beam—a motor deficit that Eggan attributes to raging inflammation, as opposed to motor neuron degeneration. Broad mice stayed on the beam as long as control mice.

What about the Broad’s environment dampened these responses? Conditions at the two facilities—including diet and light cycle—were nearly identical. However, microbial screening reports were not. Compared with mouse poop collected at the Broad, droppings from mice at Harvard contained more immune-stimulating microbes, including murine norovirus, Helicobacter species, Pasteurella pneumotropica, and Tritrichomonas muris. Though elevated, levels of these microbes were still within a normal, non-pathogenic range.

If bacteria were to blame for the ill fate of the Harvard C9-deficient mice, might antibiotics save them? Indeed, treatment of the Harvard C9 knockouts with broad-spectrum antibiotics throughout their lives prevented the emergence of all inflammatory phenotypes. Antibiotic treatment even reversed symptoms of inflammation, including the enlarged spleens, when given to older C9ORF72-deficient mice that were already sick, and improved their balance on the rotating beam. “I was blown away by this,” said Eggan, a self-described critic of rodent microbiome research.

Bestowing Harvard mice with poop from Broad mice via oral gavage also lessened all their autoimmune woes. Microbial profiling confirmed that their gut microbiomes had been transformed into a Broad mouse-like profile.

Lest this turn into petty Harvard-Broad rivalry, could microbial profiles explain the variability in C9-deficient phenotypes reported by other labs? To find out, the researchers put out a call for mouse droppings from other facilities. They profiled microbes from Johns Hopkins University, Baltimore, where C9 knockouts died young, and from the Jackson Labs, Bar Harbor, Maine, where they lived normal lifespans. Strikingly, they found similarities between microflora from the “pro-survival” Broad and Jackson facilities, while the composition of gut microbes from the two “pro-inflammatory” Harvard and Johns Hopkins environments clustered together, as well. The findings strengthened support for the idea that microbes in the gut toggle the severity of inflammation in C9-deficient mice. The researchers are continuing to sample gut microbiota from other facilities housing C9 mice.

Does this to-do about mouse poop have anything to do with ALS? The researchers assessed how gut microbiota influenced inflammation in the spinal cords of the mice, where infiltrating immune cells wreak havoc in people with this disease. In C9 knockouts raised at Harvard, the researchers found abundant myeloid cells infiltrating the spinal cord. Lifelong treatment with antibiotics prevented the invasion.

There were microglial effects also. In mice at Harvard, microglia in the spinal cord expressed high levels of lysosomal markers, in agreement with previous reports tying C9 deficiency to problems with endolysosomal trafficking (Ugolino et al., 2016; Feb 2018 news). Microglia also expressed activation markers. Antibiotic treatment assuaged microglial activation, but did not reduce lysosomal markers. The findings suggested that gut microbiota hold sway over some, but not all, consequences of C9 loss in microglia.

Eggan said the phenotypes in the various mice are a product of the net effect of C9 loss of function, which makes myeloid cells more sensitive to inflammatory stimuli, and the presence of certain immune-stimulating microbes, which kicks these cantankerous cells into attack mode.

Robert Baloh of Cedars-Sinai Medical Center in Los Angeles said the study provides clear evidence that the gut microbiome influences how severe inflammation triggered by C9 deficiency will get. In Baloh’s lab, similarly generated C9-deficient mice developed telltale signs of systemic inflammation, including an enlarged spleen and lymph nodes, but lived a normal lifespan (O’Rourke et al., 2016). Baloh views the various phenotypes reported by different labs as part of an inflammatory spectrum, ranging from systemic inflammation to autoimmune attack. “It’s really about the severity of the inflammatory response, rather than a bunch of totally different phenotypes,” he said.

Baloh noted that immune-related phenotypes are notoriously sensitive to environmental differences, including gut microflora. Florent Ginhoux of the Agency for Science, Technology and Research in Singapore agreed, noting that the cancer immunotherapy field, in particular, has been dogged by variable findings among facilities. He said that the microbiome should be taken into account when comparing study results.

C9-deficient mice do not exhibit motor symptoms of ALS, nor do they harbor RNA foci or dipeptide repeat inclusions found in people with hexanucleotide expansions in the gene. Still, Eggan said that under certain conditions, loss of C9ORF72 function can promote a powerful neuroinflammatory response, which is a key component of human ALS. He also noted that in the SOD1 mouse model of ALS, treatment with antibiotics exacerbated disease, suggesting that each pathogenic mutation has a unique relationship with gut microbiota (Blacher et al., 2019). For C9 carriers, differences in gut microbiota could help explain the wide variation in disease onset, or even incomplete penetrance, he said.

Gitler agreed. “These results are tremendously exciting and surprising, and might help to explain the variable phenotype of C9ORF72 mutations in humans,” he wrote. “They could suggest that antibiotics could facilitate therapeutic efforts to treating (or preventing) disease caused by C9ORF72 mutations.”

Stanley Appel of Houston Methodist Neurological Institute in Texas said the paper beautifully documents how gut microbiota influence systemic inflammatory responses that exacerbate neurodegenerative disease. However, he questioned how the findings relate to ALS, given the lack of motor neuron degeneration in the C9-deficient mice. He emphasized that ALS starts with insults in motor neurons, which in turn instigate damaging neuroinflammation. Therefore, he said that while gut microbiota may aggravate disease, they are unlikely to dictate onset or penetrance among C9 carriers.

Eggan and other researchers are investigating these questions with a large, NIH-funded, natural history study of C9ORF72 carriers. They will ask whether gut microbiomes differentiate between carriers who develop the disease early, late, or not at all.

These are obviously highly intriguing findings with profound implications for the treatment of ALS and FTLD based on modifying the gut microbiome,” wrote Nanda Kumar, Deepak Kumar, and Rudolph Tanzi of Massachusetts General Hospital in Boston. They noted that the findings could have relevance for other neurodegenerative diseases including AD and PD, where gut microbiota have also been implicated (Dec 2016 news; Feb 2017 newsApr 2020 conference news). They believe future studies should consider how antibiotic treatments impact not only bacteria, but also fungi and other microbial inhabitants in the gut.

In their editorial accompanying the paper, Ping Fang and Elaine Y. Hsiao of the University of California, Los Angeles, wrote that more research is needed to zero in on the particular microbes and microbial functions that sway inflammation outside of the CNS, and to assess whether this peripheral inflammation influences the degeneration of motor neurons. “Unravelling [these mechanisms] would advance our understanding of the interactions between environmental factors and genetic risk factors in ALS, and might lead to new targets for clinical intervention.”—Jessica Shugart


  1. This new manuscript from Kevin Eggan and colleagues is surprising and remarkably important, with direct implications for therapies that are currently in development for ALS and FTD. Mutations in the C9ORF72 gene are the most common cause of ALS. This discovery has completely revolutionized the ALS research field. Intense efforts are now focused on understanding the mechanism(s) by which C9ORF72 mutations cause disease and developing therapeutic strategies to treat patients harboring C9ORF72 mutations.

    The mutation is a GGGGCC hexanucleotide repeat expansion. This is thought to contribute to ALS through a combination of reduced C9ORF72 expression and the production of toxic RNAs or dipeptide repeat proteins from the repetitive RNA. Unlike other, more fully penetrant, ALS disease genes (e.g., SOD1), the penetrance of C9ORF72 mutations is variable. Some individuals with these mutations develop ALS, some develop frontotemporal dementia, and others are unaffected. Adding complexity, it has emerged that individuals with C9ORF72 mutations have increased susceptibility to autoimmune disorders, like lupus and multiple sclerosis.

    To explore the normal function of C9ORF72, and to define the potential role of reduced C9ORF72 function in disease, in the last few years researchers in the ALS field (including the Eggan team) have generated several independent lines of C9ORF72 +/- and -/- knockout mice. Perplexingly, these mice, some of which were generated using the same targeted alleles on the same genetic background, presented with variable phenotypes depending on the group. For example, the Baloh team’s knockout at Cedar’s Sinai in Los Angeles presented with splenomegaly, autoimmunity, and no shortened lifespan, whereas the Eggan knockout mouse at Harvard in Boston presented with the splenomegaly, autoimmunity, but with a shortened lifespan. This variability of phenotype has confused the field and has remained unresolved.

    In the current manuscript, the Eggan team seems to have solved this mystery and, in one fell swoop, also suggests a novel therapeutic strategy.

    The authors start by re-deriving their original knockout mouse, but this time in the Broad Institute, in a different mouse facility. They were surprised to find that the same line of C9ORF72 knockouts, now at the Broad, did not have the shortened lifespan like they did at Harvard. This launched a mission to figure out what it was in the Harvard facility (or not in the Broad facility) that was causing the difference in lifespan. Even though both facilities were employing high standards for animal care and cleanliness, the authors noted that the Harvard facility had a higher level of some bacteria and viruses than the Broad one. They hypothesized that perhaps the reduced expression of C9ORF72 in mouse could make them more sensitive to bacterial and viral infections, and subsequent pro-inflammatory phenotypes, which could explain the differential phenotypes.

    To test this hypothesis, they treated the Harvard housed +/- and -/- mice with broad spectrum antibiotics. Remarkably, this was sufficient to ameliorate phenotypes in the Harvard mice! Thus, the presence of a pro-inflammatory microbial burden contributes to reduced lifespan in the C9ORF72 KO mouse, while a C9ORF72 KO strain re-derived and bred in aseptic environment has a normal lifespan. 

    There are several extremely compelling experimental results in this paper. Among them are that the authors were able to obtain samples from colonies in other institutions that either had early mortality in the knockout mouse (C9Harvard, Johns Hopkins) or normal survival (C9Broad, Jackson labs). After additional microbial profiling of these new samples, the authors found differential expression of bacterial species along the lines of early mortality/pro-inflammatory cohorts and pro-survival cohorts. These results strengthen the authors’ claims that the gut microflora can influence the survival of C9 knockout mouse, as the findings are no longer restricted to the authors’ colonies. 

    The other key experiment showed that fecal transplantation from the pro-survival Broad colony to the knockout C9Harvard mice can dampen the pro-inflammatory immune reaction in the knockout mice, compared to fecal transplantation from the Harvard colony to the knockout C9Harvard mice. This again supports the authors’ central claim that the environment and gut microbial context can influence the pathological inflammatory response in the C9KO mice, and complements the broad-spectrum antibiotic experiments.

    Of course, the reverse experiment (fecal transplantation from Harvard to the C9Broad mice) and survival data would be desirable and perhaps an area for future consideration. The paper also leaves open the question about the potential for anti-inflammatory, pro-survival microbial species present in the Broad/Jackson cohorts.

    Collectively, these results are tremendously exciting and surprising. They might help to explain the variable phenotype of C9ORF72 mutations in humans, and perhaps suggest that antibiotics could facilitate therapeutic efforts to treating (or preventing) disease caused by C9ORF72 mutations.

  2. Using a mutant C9ORF72 mouse model, Burberry et al. report on the impact of the gut microbiome on disease progression in ALS. They show that fecal transplant using gut microflora from mice with loss of both (-/-) alleles of C9ORF72 raised in a protective environment (Broad Institute, MIT) ameliorated disease pathology when introduced into mutant mice raised in an inflammation-prone environment (Harvard University). Apart from having a shorter lifespan, C9ORF72 mutant mice raised at the Harvard facility showed increased levels of pro-inflammatory signals and autoantibodies, higher neutrophil infiltration, and lower platelet counts, as compared to mice raised at the Broad Institute.

    Previous studies have reported profound effects of the gut microbiome on Alzheimer’s disease (AD) and Parkinson disease (PD) phenotypes (Minter et al., 2017Sampson et al., 2016; Harach et al., 2017). Here, the authors found different microbial compositions under different environmental conditions that profoundly influence the C9ORF72 mutant phenotype and survival rate, as well as CNS recruitment of peripheral myeloid progenitors.

    These are obviously highly intriguing findings with profound implications for the treatment of ALS and FTLD based on modifying the gut microbiome. However, with regard to the finding that broad-spectrum antibiotics significantly reduced inflammatory and autoimmune phenotypes in the mutant mice, it will be important in future studies to also compare the gut mycobiome profiles of these two groups of mice.

    The gut environment is a symbiotic ecosystem that requires cooperative interaction among all microbial groups. Prolonged antibiotic use in mice would likely promote fungal overgrowth, which can be harmful or beneficial depending on the competing species. While the mycobiome remains one of the most understudied areas of the gut microbiome, their synergistic coexistence in the gut ecosystem, and recent reports of their link to other neurodegenerative disease such as AD and PD, underscore their importance in promoting overall gut health. 

    Therefore, correlation of mycobiome modulation pre- and post- antibiotic administration, and post-FMT will be very interesting to investigate in future studies.


    . Interactions between the microbiota, immune and nervous systems in health and disease. Nat Neurosci. 2017 Feb;20(2):145-155. Epub 2017 Jan 16 PubMed.

    . Antibiotic-induced perturbations in microbial diversity during post-natal development alters amyloid pathology in an aged APPSWE /PS1ΔE9 murine model of Alzheimer's disease. Sci Rep. 2017 Sep 5;7(1):10411. PubMed.

    . Gut Microbiota Regulate Motor Deficits and Neuroinflammation in a Model of Parkinson's Disease. Cell. 2016 Dec 1;167(6):1469-1480.e12. PubMed.

    . Reduction of Abeta amyloid pathology in APPPS1 transgenic mice in the absence of gut microbiota. Sci Rep. 2017 Feb 8;7:41802. PubMed.

    . Potential roles of gut microbiome and metabolites in modulating ALS in mice. Nature. 2019 Aug;572(7770):474-480. Epub 2019 Jul 22 PubMed.

  3. We have read with great interest the important new paper from Burberry and colleagues demonstrating the influence of gut bacteria on the ability of the C9ORF72 hexanucleotide repeat to suppress inflammation in the brain (Burberry et al., 2020). We agree that it will be of value to also evaluate the contribution of fungi, as suggested by  Shanmugam, Kumar and Tanzi.

    In addition, it is important to consider where the interaction between the microbiota and the host is taking place. The oral cavity is home to a vibrant microbiome, which has profound influences on the immune and nervous systems. It has been proposed that the site at which this interaction occurs influences the initial phenotype of the neurodegenerative disorder, e.g. oral microbes may lead to bulbar ALS, and colonic microbes may lead to spinal onset (Friedland and Chapman, 2017). In order to understand the variable phenotypes of the C9ORF72 related disorders we need to comprehensively consider the microbial agents involved, as well as where the microbial-host interaction is taking place.


    . C9orf72 suppresses systemic and neural inflammation induced by gut bacteria. Nature. 2020 Jun;582(7810):89-94. Epub 2020 May 13 PubMed.

    . The role of microbial amyloid in neurodegeneration. PLoS Pathog. 2017 Dec;13(12):e1006654. Epub 2017 Dec 21 PubMed.

Make a Comment

To make a comment you must login or register.


News Citations

  1. ALS GWAS Confirm Chromosome 9 Risk Factor—But What Is It?
  2. Corrupt Code: DNA Repeats Are Common Cause for ALS and FTD
  3. Paper Alert: Autoimmunity in Another C9ORF72-Deficient Mouse Strain
  4. C9ORF72 Mice Point to Gain of Toxic Function in ALS, FTD
  5. C9ORF72 Knockout Causes Inflammation, not Neurodegeneration
  6. Lack of C9ORF72 Protein Renders Neurons More Vulnerable to Degeneration
  7. Do Microbes in the Gut Trigger Parkinson’s Disease?
  8. Microbes in the Gut Egg on Aβ Pathology in Mice
  9. ‘Working from Home’: Do Gut Microbes Hold Sway Over Glia, Aβ?

Paper Citations

  1. . Loss of C9orf72 Enhances Autophagic Activity via Deregulated mTOR and TFEB Signaling. PLoS Genet. 2016 Nov;12(11):e1006443. Epub 2016 Nov 22 PubMed.
  2. . C9orf72 is required for proper macrophage and microglial function in mice. Science. 2016 Mar 18;351(6279):1324-9. PubMed.
  3. . Potential roles of gut microbiome and metabolites in modulating ALS in mice. Nature. 2019 Aug;572(7770):474-480. Epub 2019 Jul 22 PubMed.

Further Reading


  1. . Lack of C9ORF72 coding mutations supports a gain of function for repeat expansions in amyotrophic lateral sclerosis. Neurobiol Aging. 2013 Sep;34(9):2234.e13-9. PubMed.

Primary Papers

  1. . C9orf72 suppresses systemic and neural inflammation induced by gut bacteria. Nature. 2020 Jun;582(7810):89-94. Epub 2020 May 13 PubMed.