Microbial signals emanating from the bowels of mice somehow worsen tau pathology and neurodegeneration. This, according to a study that used microbe-nixing protocols to connect bacteria teeming in the gut, ApoE genotype, and tau-mediated neurodegeneration. Led by David Holtzman at Washington University in St. Louis, scientists found that tau transgenic mice raised in germ-free conditions harbored less tau pathology, and suffered less neurodegeneration, than their microbe-replete counterparts. Briefly killing off gut microbes with a round of antibiotics also fended off tau-mediated damage, but only in males, and more so in those expressing ApoE3 than ApoE4. The scientists tied short-chain fatty acids churned out by some gut microbes to cytokine production by innate immune cells that may stoke damaging glial responses. The findings, published in Science on January 13, support the idea that strong ties exist between the throngs of microbes living in the gut, and neuroinflammatory conditions in the brain.

  • Raising them sans microbes protects tau-transgenic mice.
  • Antibiotics only protect males, especially those expressing ApoE3.
  • Short-chain fatty acids produced by gut bacteria revved cytokines, riled glia in brain.

“Overall, these exciting re­sults highlight the potential to harness the gut microbiota to prevent or slow the pro­gression of AD and other tauopathies, and raise awareness about the potential long-term effects of early life diet,” wrote Tanya Jain and Yue-Ming Li of Memorial Sloan Kettering Cancer Center in New York in an editorial accompanying the paper in Science.

Besides tau, other neurodegenerative culprits, including Aβ and α-synuclein, are reportedly goaded by the diverse bacterial inhabitants of the gut (Dec 2016 news; Feb 2017 news; Apr 2020 conference news). Exactly how the bacteria manage to accelerate brain pathology is not entirely clear, however, studies suggest that microbial metabolites, including short-chain fatty acids, influence the state of peripheral immune cells as well as glia in the brain (Jun 2015 news).

First author Dong-Oh Seo and colleagues studied how these microbiome-microglia interactions might play out in the context of tauopathy. Previously, Holtzman’s group found that microglia were centrally involved in wreaking neuronal havoc in response to tau pathology, and that ApoE4 made things worse (Sep 2017 news). To find out if signals from the gut microbiome fit into this picture, the researchers raised P301S-tau/ApoE4 transgenic (TE4) mice, in germ-free conditions from birth. Remarkably, a life without gut microbes lessened the burden of tau tangles and protected the TE4 mice from neurodegeneration. Both males and females benefited, but not when the mice were fed a generous helping of gut bacteria—in the form of poop—from 40-week-old TE4 mice.

When the researchers intervened more transiently with the microbiome, by giving 2-week-old mice a week-long round of antibiotics, only male mice benefited, and P301S-tau/ApoE3 knock-in (TE3) males were more protected than TE4 males. At 40 weeks of age, they had much less tau pathology and neurodegeneration in their brains than untreated controls. Antibiotics did nothing to fend off tau pathology or neurodegeneration in female mice, regardless of ApoE genotype. Notably, female mouse microbiomes are different from male microbiomes. As reported previously, ApoE knockout, “TEKO” mice, were protected from both tau pathology and neurodegeneration.

Antibiotics Fend Off Tau. Hyperphosphorylated tau (brown) packed the hippocampi of male TE3 and TE4 mice (top). Treatment with antibiotics dramatically reduced tau burden in TE3, but less so in TE4 mice (bottom). Tau mice without ApoE have little pathology, with or without antibiotic treatment (right). [Courtesy of Seo et al., Science, 2023.]

Might the differential effect of the microbiome based on sex and ApoE genotype come down to glia? Using single-cell RNA sequencing, the researchers found that in response to tau pathology, astrocytes and microglia expanded and dramatically shifted their gene expression profiles. Treatment with antibiotics dampened this shift in male TE3 mice, but not in male TE4 mice, or in females of either genotype. The findings align with previous work from the Holtzman lab and other groups, which indicated that both female sex and ApoE4 genotype promote damaging responses in microglia (Oct 2019 newsJul 2019 conference news). “Perturbing the microbiome with antibiotics may be insufficient to overcome the effects of sex and ApoE genotype on microglia,” Holtzman said.

Next, the researchers looked for signals from the microbiome that could explain the responses to antibiotics. Treatment with this cocktail in early life stripped the gut of microbes temporarily, and a different complement grew back into adulthood. Fewer Helicobacter, Ruminococcus, and Butyricicoccus populated the intestines at 40 weeks of age. Members of the latter two genera are known to churn out short-chain fatty acids (SCFAs). In line with this, concentrations of these fatty acids dropped in the gut in response to antibiotics, but only in males. Untreated females already have less SCFAs in their gut than do untreated males, perhaps explaining why the females did not respond to antibiotics. When the researchers added a cocktail of microbial SCFAs—acetate, butyrate, and proprionate—to the drinking water of the male TE4 mice that were raised in germ-free conditions, the mice developed substantial tau pathology and gliosis.

Glia do not express SCFA receptors, so why do the cells respond to the metabolites? The scientists hypothesized that SCFAs might activate glia indirectly, via peripheral immune cells that do express SCFA receptors, such as those that patrol the meninges. In support of this idea, they found that antibiotic-treated male mice had fewer meningeal γδ T cells and plasmacytoid dendritic cells. The latter produce most of the interferon in the body, Holtzman noted, and this inflammatory molecule is known to incite neuroinflammation in the brain. The findings hint that interferon cranked out by pDCs in the meninges could provide a pathogenic link between microbes in the gut and microglia in the brain.

Microbiome-Microglia Convergence. Gut microbes (bottom) release numerous metabolites, including short-chain fatty acids (SCFA). These interact with peripheral immune cells (center), which influence glial activity in the brain (top). [Courtesy of Seo et al., Science, 2023.]

Daniel Erny of the University of Freiburg in Germany noted that while SCFAs may indirectly transform microglia by activating peripheral immune cells, a direct pathway also exists. Previously, Erny reported that the acetate, a microbial SCFA, enters the CNS, where it is readily taken up by microglia (Erny et al., 2021). He found that while acetate worsened Aβ accumulation in 5xFAD mice, it was also essential for proper microglial maturation and homeostasis in healthy mice. Notably, SCFAs can be anti-inflammatory in other disease settings, such as multiple sclerosis (Melbye et al., 2019). “The role of SCFAs is health and disease-context dependent,” Erny said.

For Erny, the most important message from the paper is that the two hallmark pathologies of AD—amyloid plaque and now neurofibrillary tangles—are similarly exacerbated by the microbiome via microglia.

Steven Estus and Diana Zajac of the University of Kentucky in Lexington noted that SCFAs may also act independently of microglia in the brain by, for example, dampening levels of amyloid clearing enzymes (Harach et al., 2017).—Jessica Shugart


  1. In this heroic work, Seo et al. provide evidence for an ApoE-dependent and ApoE-isoform-dependent role of the gut microbiome in neuropathology in the P301S tau model, expressing human ApoE3 (TE3) or ApoE4 (TE4). Neuropathology in conventionally reared (Conv-R) TE4 mice, germ-free (GF) TE4 mice, and GF TE4 mice that received fecal implants at 12 weeks of age from sex-matched 40-week-old TR4 mice (Ex-GF), was analyzed at 40 weeks of age, when TE4 mice showed greater neuropathology than TE3 mice in earlier studies. Compared to E4 mice without P301S tau, ConV-R female and male TE4 mice showed hippocampal volume loss and enlargement of the lateral ventricle (LV) volume. GF TE4 female and male mice showed less hippocampal volume loss and less LV volume than ConV-R female and male TE4 mice, but less hippocampal volume and more LV volume than ConV-R and GF E4 mice.

    These effects showed anatomical specificity, because no difference in neuropathology in the entorhinal-piriform cortex was seen. The protective effects seen for hippocampal volume and hippocampal tau phosphorylation in GF TE4 mice were not seen in Ex-GF TE4 mice, while the protective effects seen for LV volume seemed partially retained, especially in males. The protective effects in GF TE4 mice were associated with reduced immunoreactivity for markers of astrocytes and activated microglia and with large and more branched astrocytes, an astrocytic phenotype already seen at 12 weeks of age and preceding severe tau pathology. Consistent with these changes, differential gene expression at 40 weeks of age was seen comparing ConV-R and GF-E4 mice but not comparing ConV-R and Ex-GF TE4 mice.

    Antibiotic treatment from postnatal 16 through 22 increased hippocampal volume at 40 weeks in TE3 male, but not female, mice. There was a trend toward a treatment effect in TE4 male, but not female, mice but it did not reach significance. Antibiotic treatment improved nest building in TE3 and TE4 male, but not female, mice, and in both genotypes nest building was positively correlated with hippocampal volume. Effects of antibiotic treatment on LV volume were also seen only in TE3 and TE4 males but not females. Consistent with this pattern, hippocampal tau phosphorylation was significantly reduced only in TE3 male mice following antibiotic treatment.

    Single-nucleus RNA sequencing revealed that specific neuronal populations were reduced and the microglial population increased in the presence of tau pathology in the hippocampi of ConV-R TE3, TE4, and TE mice without ApoE (TEKO) male mice and these effects were reduced with antibiotic treatment. The antibiotic treatment was associated with an increased cecal/body weight ratio and an altered gut microbiome, as analyzed by 16S ribosomal RNA. The effects of the gut microbiome on glial function might involve microbiota-generated short-chain fatty acids (SCFAs) and a peripheral cytokine response. Bacteria generating SCFAs acetate, propionate, and butyrate were reduced by antibiotic treatment in TE3 and TE4 males. In addition, administration of SCFAs in the drinking water to 10- or 31-week-old TE4 GF mice for five or four weeks, respectively, altered alveolar macrophage gene expression and increased gliosis and tau phosphorylation in the hippocampi of the older TE4 GF mice.

    The results of the current study are consistent with the fact that the human gut microbiome diversifies with age, reflects healthy versus unhealthy aging, associates with a healthy lipid profile, predicts survival (Wilmanski et al., 2021), and that alterations in microbiome composition link to Alzheimer’s disease (AD) and impact AD-associated behaviors and brain pathologies (Kundu et al., 2022; Kundu et al., 2021; Marizzoni et al., 2020). A relationship between the gut microbiome and the brain is not limited to AD but is also seen in other neurodegenerative conditions. Taxa that comprise the gut microbiome link to Parkinson’s disease (PD) as well (Fonseca et al., 2019; Elfil et al., 2020; Keshvarzian et al., 2020; Koutzoumis et al., 2020). Consistent with these human data, gut microbiota regulate motor impairments and neuroinflammation in a PD mouse model overexpressing α-synuclein (Sampson et al., 2016) and we showed that in wild-type mice and mice lacking the metabotropic receptor 8 (mGlu8), the neurotoxin 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP) and genotype affected the diversity of the gut microbiome and that there were significant associations between microbiome α-diversity and sensorimotor performance and between microbiome composition and fear learning (Torres et al., 2018). Interestingly, MPTP affected the swim speeds of E4, but not E3, mice in the water maze (Torres et al., 2020), consistent with an ApoE isoform-dependent response using treatments that affect the gut microbiome.

    In the current study, based on the age difference between the donor and recipient mice, and that only one genotype was used as donor, the role of age and/or the genotype of the donor in the effects of the fecal implant are unclear. Our prior work supports the notion that genotype impacts the microbiome’s contribution to cognition: Using 6-month-old knock-in (KI) mice expressing human amyloid precursor protein with dominant mutations (AppNL-F or AppNL-G-F), we showed that the microbiome’s composition as well as its association with mouse cognition differs based on the APP genotype and DNA methylation of the APOE gene in the hippocampus (Kundu et al., 2021). In addition, transplanting microbiomes from AppNL-G-F and AppNL-G-F mice also expressing E4 (AppNL-G-F/E4) into germ-free WT mice revealed donor genotype-dependent differences in recipient mouse behavioral and cognitive performance Insoluble cortical Ab40 levels were detected in AppNL-G-F and AppNL-G-F/E4 recipient mice. Although tau pathology is more associated with cognitive impairments than Aβ pathology, recipients of AppNL-G-F donor mice had cortical insoluble Aβ40 levels that positively correlated with activity levels on the first and second day of open field testing. This relationship was genotype-dependent and not seen in recipients of AppNL-G-F/E4 donor mice. In addition to genotype, the relationship between the gut microbiome and behavioral and cognitive measures are also modulated by sex (Raber et al., 2020). This is consistent with the sex-dependent effects of antibiotic treatments on neuropathology seen in the current study.

    Especially as beneficial interventions in AD will likely need to start relatively early and be safe and affordable, manipulation of the gut microbiome is an attractive therapeutic strategy to consider and future studies are warranted to further assess this strategy. However, as the current and earlier studies indicate, genotype and sex of the recipient, and in case of fecal transplants of the donor as well, need to be carefully considered.


    . Gut microbiome pattern reflects healthy ageing and predicts survival in humans. Nat Metab. 2021 Feb;3(2):274-286. Epub 2021 Feb 18 PubMed.

    . Fecal Implants From App NL-G-F and App NL-G-F/E4 Donor Mice Sufficient to Induce Behavioral Phenotypes in Germ-Free Mice. Front Behav Neurosci. 2022;16:791128. Epub 2022 Feb 8 PubMed.

    . Integrated analysis of behavioral, epigenetic, and gut microbiome analyses in AppNL-G-F, AppNL-F, and wild type mice. Sci Rep. 2021 Feb 25;11(1):4678. PubMed.

    . Short-Chain Fatty Acids and Lipopolysaccharide as Mediators Between Gut Dysbiosis and Amyloid Pathology in Alzheimer's Disease. J Alzheimers Dis. 2020;78(2):683-697. PubMed.

    . The Gut and Parkinson's Disease-A Bidirectional Pathway. Front Neurol. 2019;10:574. Epub 2019 Jun 4 PubMed.

    . Implications of the Gut Microbiome in Parkinson's Disease. Mov Disord. 2020 Jun;35(6):921-933. Epub 2020 Feb 24 PubMed.

    . The gut microbiome in Parkinson's disease: A culprit or a bystander?. Prog Brain Res. 2020;252:357-450. Epub 2020 Mar 5 PubMed.

    . Alterations of the gut microbiota with antibiotics protects dopamine neuron loss and improve motor deficits in a pharmacological rodent model of Parkinson's disease. Exp Neurol. 2020 Mar;325:113159. Epub 2019 Dec 13 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.

    . Effects of Sub-Chronic MPTP Exposure on Behavioral and Cognitive Performance and the Microbiome of Wild-Type and mGlu8 Knockout Female and Male Mice. Front Behav Neurosci. 2018;12:140. Epub 2018 Jul 18 PubMed.

    . ApoE isoform-specific differences in behavior and cognition associated with subchronic MPTP exposure. Learn Mem. 2020 Sep;27(9):372-379. Print 2020 Sep PubMed.

    . Effects of Six Sequential Charged Particle Beams on Behavioral and Cognitive Performance in B6D2F1 Female and Male Mice. Front Physiol. 2020;11:959. Epub 2020 Aug 28 PubMed.

  2. Seo et al. add to our understanding of the microbiome’s impact on murine AD models by showing that neurodegeneration is reduced in gnotobiotic P301S mice. Several lines of evidence suggest microglia are key to this neuroprotection because (i) neurodegeneration is reduced in P301S mice lacking TREM2, a microglial activator (Leyns et al., 2017), (ii) microglial function is reduced with gnotobiotic conditions, an effect reversed by short-chain fatty acid (SCFA) supplementation (Erny et al., 2015), and (iii) Seo et al. report that SCFA supplementation reversed the gnotobiotic neuroprotection in the P301S model.

    Germ-free conditions have also been reported to decrease amyloid burden in APPPS1 mice (Harach et al., 2017). Although this effect was also reversed by SCFA treatment (Colombo et al., 2021), the mechanism for gnotobiotic reduction of amyloid burden is less clear. In particular, Harach et al. noted that germ-free conditions resulted in an increase in amyloid clearing enzymes, and that this increase was blunted by SCFA. Hence, SCFA may act in both microglial-dependent and -independent pathways to alter disease pathology. The effects of SCFA supplementation appear most robust in gnotobiotic conditions, noting that inconsistent results have been obtained in SCFA-treated APP mice with a conventional microbiome (Colombo et al., 2021; Fernando et al., 2020; Zajac et al., 2022).


    . TREM2 deficiency attenuates neuroinflammation and protects against neurodegeneration in a mouse model of tauopathy. Proc Natl Acad Sci U S A. 2017 Oct 24;114(43):11524-11529. Epub 2017 Oct 9 PubMed.

    . Host microbiota constantly control maturation and function of microglia in the CNS. Nat Neurosci. 2015 Jun 1; PubMed.

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

    . Microbiota-derived short chain fatty acids modulate microglia and promote Aβ plaque deposition. Elife. 2021 Apr 13;10 PubMed.

    . Sodium Butyrate Reduces Brain Amyloid-β Levels and Improves Cognitive Memory Performance in an Alzheimer's Disease Transgenic Mouse Model at an Early Disease Stage. J Alzheimers Dis. 2020;74(1):91-99. PubMed.

    . Exogenous Short Chain Fatty Acid Effects in APP/PS1 Mice. Front Neurosci. 2022;16:873549. Epub 2022 Jul 4 PubMed.

  3. In this paper Dong-oh Seo and colleagues have investigated the impact of ApoE isoforms on tau-mediated neurodegeneration. The authors show that mice raised in germ-free conditions, or treated with antibiotics, are protected against tau pathology and neurodegeneration, and that these benefits were more pronounced in ApoE-3-expressing male mice. Intriguingly, short-chain fatty acids produced by the gut microbes may have played a role in promoting this harmful phenotype.

    This is a very exciting and thought-provoking work from the Holtzman group which adds further support to the role of the APOE genotype on gut-brain axis modulation, and suggests that the gut microbiome is worth further investigation as a potential target to mitigate the deleterious impact of the APOE genotype on cognitive decline and the prevention of AD.

Make a Comment

To make a comment you must login or register.


News Citations

  1. Do Microbes in the Gut Trigger Parkinson’s Disease?
  2. Microbes in the Gut Egg on Aβ Pathology in Mice
  3. ‘Working from Home’: Do Gut Microbes Hold Sway Over Glia, Aβ?
  4. To Be Hale and Hearty, Brain Microglia Need a Healthy Gut
  5. ApoE4 Makes All Things Tau Worse, From Beginning to End
  6. In Tauopathy, ApoE Destroys Neurons Via Microglia
  7. Down to Sex? Boy and Girl Microglia Respond Differently

Mutations Citations

  1. APOE C130R (ApoE4)

Research Models Citations

  1. Tau P301S (Line PS19)
  2. APOE4 Knock-In, floxed (CureAlz)
  3. APOE3 Knock-In, floxed (CureAlz)
  4. 5xFAD (C57BL6)

Paper Citations

  1. . Microbiota-derived acetate enables the metabolic fitness of the brain innate immune system during health and disease. Cell Metab. 2021 Nov 2;33(11):2260-2276.e7. PubMed.
  2. . Short-chain fatty acids and gut microbiota in multiple sclerosis. Acta Neurol Scand. 2019 Mar;139(3):208-219. Epub 2018 Dec 3 PubMed.
  3. . Reduction of Abeta amyloid pathology in APPPS1 transgenic mice in the absence of gut microbiota. Sci Rep. 2017 Feb 8;7:41802. PubMed.

Further Reading

No Available Further Reading

Primary Papers

  1. . ApoE isoform- and microbiota-dependent progression of neurodegeneration in a mouse model of tauopathy. Science. 2023 Jan 13;379(6628):eadd1236. PubMed.