News & Views | Published:


Gut microbes augment neurodegeneration

Nature volume 544, pages 304305 (20 April 2017) | Download Citation

Bacterial residents of the human body often provide beneficial effects, but some can be harmful. The action of gut bacteria has been found to be tightly linked to neurodegeneration in a mouse model of Parkinson's disease.

Trillions of resident bacteria, known as the microbiota, populate the bodies of their mammalian hosts in regions such as the nose, skin and gut1. This cohabitation can have mutual benefits. For example, the microbiota is required for some vitamin K synthesis, and bacterial fermentation of dietary fibre provides the host with energy in the form of, for example, short-chain fatty-acid molecules2. However, writing in Cell, Sampson et al.3 highlight a negative interaction, in which gut bacteria seem to affect disease severity in a mouse model of Parkinson's disease. Understanding the nature of this long-distance link between gut microbiota and neurodegeneration might illuminate the mechanisms responsible for this and other brain diseases.

Parkinson's disease is the second most common adult neurodegenerative disorder4, and symptoms include severely abnormal movement caused by defects in motor control. Less than 10% of cases of the disease are known to be caused by an inherited genetic mutation4, and most cases are of unknown cause. The main hallmarks5 of this incurable illness in the brain are the loss of midbrain neurons that release the neurotransmitter dopamine, and the presence of a misfolded, aggregated and neurotoxic form of the protein α-synuclein. There is firm evidence that microglia — the brain's main immune cells, a type of cell known as a phagocyte that can ingest material — are crucial for the course of neurodegeneration6. Techniques that allow specific targeting of the microglial-cell population7 have revealed that these cells can have a neurotoxic role in animal models of Parkinson's disease8.

Some reports9,10,11 have highlighted the strong communication between the gut and the brain. It has been proposed that the vagus nerve, which forms a neural connection between the gut and the brainstem, might be a site of origin for Parkinson's disease pathology before it spreads to the brain5, although the nature of the initial pathology at the cellular or molecular level is unknown. This hypothesis is supported by the discovery that people who have undergone surgery to remove part of their vagus nerve have a decreased risk of developing the disease12.

To investigate a possible role for the gut in Parkinson's disease, Sampson et al. used model mice that express high levels of human α-synuclein in brain neurons13. The authors raised these mice in a germ-free environment or raised them conventionally and removed their microbiota postnatally by antibiotic treatment. The absence of gut bacteria resulted in reduced severity of the animals' abnormal movements and lower levels of aggregated, misfolded α-synuclein compared with the model mice that retained their microbiota (Fig. 1).

Figure 1: A link between gut bacteria and Parkinson's disease.
Figure 1

In a mouse model of Parkinson's disease13, the animals express high levels of human α-synuclein protein in their brains, and have disease characteristics that include movement abnormalities, α-synuclein aggregation in neurons expressing the neurotransmitter dopamine, an immune response in the brain that includes the activation of microglial cells, and the production of potentially neurotoxic cytokine molecules. When Sampson et al.3 removed gut bacteria from the mice, the severity of the disease symptoms was reduced. The molecular and cellular connection between the gut and the brain that mediates the effect on symptoms is not fully understood (dotted line). Some evidence indicates that short-chain fatty acids (SCFAs) generated by gut bacteria might have a role. If the model mice for Parkinson's disease lacking gut bacteria received a transplant of faecal bacteria from individuals with the disease, the animals developed movement abnormalities that did not occur when faecal bacteria from healthy people were transplanted instead. Using wild-type mice for the same transplant experiments did not result in movement abnormalities in the animals.

The microbiota is known to affect microglia in the brains of wild-type mice12. Sampson and colleagues therefore tested whether the presence of the microbiota might modulate microglial-cell shape (as a way of monitoring the cells' activation status) or function in their mouse model. They found that the presence of the host microbiota resulted in an enhanced microglial immune response, known as activation, which included raised levels of the potentially neurotoxic cytokine proteins IL-6 and TNF-α.

Sampson and colleagues next investigated which bacterial products might modulate α-synuclein aggregation and the accompanying microglial activation. Short-chain fatty acids produced by bacteria are vital for immune-cell homeostasis in the brain11 and in other organs, including the colon14. The authors therefore fed the mice a mixture of acetate, propionate and butyrate, the three most abundant short-chain fatty acids in the gut14. This was sufficient to promote a microglia-mediated immune response, increase α-synuclein aggregation and cause movement abnormalities.

The relative abundance of specific faecal gut microbes differs in individuals with Parkinson's disease compared with healthy controls15. Sampson and colleagues investigated whether this altered microbiota might contribute to the pathology of the disease. Faecal microbiota, either from people with Parkinson's disease or from healthy individuals, were transplanted into initially germ-free model mice. Model mice that received microbiota from people with Parkinson's showed an increase in movement abnormalities, higher levels of propionate and butyrate and lower levels of acetate in their faeces compared with model animals that received faecal microbiota from healthy individuals. Wild-type animals that received faecal microbiota from people with Parkinson's disease did not show movement abnormalities. Together, these results provide compelling evidence that gut microbes have a role in disease pathology.

Sampson and colleagues' findings raise several questions. How could short-chain fatty acids exert their effects on Parkinson's disease, given that the FFAR2 receptor protein for these molecules has not been observed to be expressed in the brain11? Does this mean that other cells are involved that express FFAR2 and produce metabolites that modify disease outcome, or can short-chain fatty acids influence brain cells without requiring the presence of FFAR2 receptors? Another question is whether the host microbiota shapes any other forms of neurodegeneration, as might be inferred from the previous finding that the amount of disease-associated amyloid protein is altered after antibiotic treatment in a mouse model of Alzheimer's disease16.

Other aspects of these results are worth considering. Absence of the host microbiota changes the permeability of the blood–brain barrier in mice and affects brain cells other than microglia9,10, which might account for the reduced Parkinson's symptoms observed by Sampson and colleagues. The synthesis of several neurotransmitters, including dopamine, is also affected by the absence of the host microbiota9. However, the authors did not investigate whether their model mice had altered numbers of dopamine-expressing neurons or dopamine levels.

Sampson et al. find that the gut microbiota is intimately involved in the pathogenesis of Parkinson's disease, and they suggest that this involvement is mediated by regulation of the activation status of microglia. Knowing how compounds generated by bacteria can shape the immune response in the brain and the development as well as the outcome of neurodegenerative disease will undoubtedly contribute to our understanding of Parkinson's disease, and perhaps open other therapeutic avenues for treating this devastating condition.



  1. 1.

    , & Cell 164, 337–340 (2016).

  2. 2.

    & Science 292, 1115–1118 (2001).

  3. 3.

    et al. Cell 167, 1469–1480 (2016).

  4. 4.

    & Lancet Neurol. 5, 525–535 (2006).

  5. 5.

    , , & J. Neural Transm. 110, 517–536 (2003).

  6. 6.

    & Nature Rev. Neurosci. 15, 300–312 (2014).

  7. 7.

    et al. Nature Neurosci. 16, 1618–1626 (2013).

  8. 8.

    et al. J. Exp. Med. 212, 469–480 (2015).

  9. 9.

    et al. Proc. Natl Acad. Sci. USA 108, 3047–3052 (2011).

  10. 10.

    et al. Sci. Transl. Med. 6, 263ra158 (2014).

  11. 11.

    et al. Nature Neurosci. 18, 965–977 (2015).

  12. 12.

    et al. Ann. Neurol. 78, 522–529 (2015).

  13. 13.

    et al. Neurotherapeutics 9, 297–314 (2012).

  14. 14.

    et al. Science 341, 569–573 (2013).

  15. 15.

    et al. Mov. Disord. 30, 350–358 (2015).

  16. 16.

    et al. Sci. Rep. 6, 30028 (2016).

Download references

Author information


  1. Daniel Erny and Marco Prinz are in the Faculty of Medicine, Institute of Neuropathology, University of Freiburg, D-79106 Freiburg, Germany.

    • Daniel Erny
    •  & Marco Prinz
  2. D.E. is also in the Berta-Ottenstein programme, University of Freiburg.

    • Daniel Erny
  3. M.P. is also in the BIOSS Centre for Biological Signalling Studies, University of Freiburg.

    • Marco Prinz


  1. Search for Daniel Erny in:

  2. Search for Marco Prinz in:

Corresponding author

Correspondence to Marco Prinz.

About this article

Publication history



Further reading


By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.

Newsletter Get the most important science stories of the day, free in your inbox. Sign up for Nature Briefing