Angry Astrocytes kill nerves and probably contribute to progressive MS

Liddelow SA, Guttenplan KA, Clarke LE, Bennett FC, Bohlen CJ, Schirmer L, Bennett ML, Münch AE, Chung WS, Peterson TC, Wilton DK, Frouin A, Napier BA, Panicker N, Kumar M, Buckwalter MS, Rowitch DH, Dawson VL, Dawson TM, Stevens B, Barres BA.Neurotoxic reactive astrocytes are induced by activated microglia.Nature. 2017. doi: 10.1038/nature21029.

Reactive astrocytes are strongly induced by central nervous system (CNS) injury and disease, but their role is poorly understood. Here we show that a subtype of reactive astrocytes, which we termed A1, is induced by classically activated neuroinflammatory microglia. We show that activated microglia induce A1 astrocytes by secreting Il-1α, TNF and C1q, and that these cytokines together are necessary and sufficient to induce A1 astrocytes. A1 astrocytes lose the ability to promote neuronal survival, outgrowth, synaptogenesis and phagocytosis, and induce the death of neurons and oligodendrocytes. Death of axotomized CNS neurons in vivo is prevented when the formation of A1 astrocytes is blocked. Finally, we show that A1 astrocytes are abundant in various human neurodegenerative diseases including Alzheimer’s, Huntington’s and Parkinson’s disease, amyotrophic lateral sclerosis and multiple sclerosis. Taken together these findings help to explain why CNS neurons die after axotomy, strongly suggest that A1 astrocytes contribute to the death of neurons and oligodendrocytes in neurodegenerative disorders, and provide opportunities for the development of new treatments for these diseases.

  • Inflammation activates microglia
  • Activated microglia produced a gliotic astrocyte to be produced
  • These gliotic A1 astrocytes block oligodendrocyte formation
  • These gliotic A1 astrocytes are neurotoxic and make toxic elements to kill nerves
  • Blocking A1 astrocytes may slow nerve loss
  • Progressive MS is likely to involve glial inflammation 
Astrocytes are an underated cell, but we know they support the function of the brain and control the tightness of the blood brain barrier, they control neurotransmitter levels, support the energy and growth factor requirements of nerves and other cells and cause problems, when they are targeted for damage in neuromyelitis optica. 

They are also the major scar forming cell in the demyelinated lesion and so contribute to lack of remyelination. When astrocytes are activated some people call them “reactive astrocytes” other people including me would call them “gliotic” where they put out more proteins including glial fibrillary acidi proteiin ( a hall mark of astrocytes) and many years ago I showed that they contain lots of IL-6 at least in the mouse. In this study term an reactive astrocyte subset and call it A1. However this should not be confused if you read about type I and type II astrocyttes.

This study shows that activated microglial produce some inflammatory mediators and this helps create the astrocytes. These are a complement component and interleukin-1 and tumor necrosis factor, which are well known pro-inflammatory mediators produced by inflammatory such as microglia and other cells….the other cells may relate to why blocking TNF can lead to demyelinating disease and MS worsening.

A1 astrocytes, which are present in diseases associated with neurodegeneration, lose the ability to promote neuronal survival, nerve outgrowth, synaptogenesis and phagocytosis, and induce the death of neurons and oligodendrocytes. 

A1 astrocytes loose the ability to promote nerve survival and induce the death of nerves and oligodendrocytes. So this allows you to have nerve damage without needing a T or a B cell anywhere near the system. Importantly if you cut/transect the axon to simulate damage then the death of axotomized CNS neurons is prevented when the formation of A1 astrocytes is blocked,

Therefore, targeting this process may slow the processes of neurodegeneration, but is the the T and B cell theory intact.

Sure it it is because you have to ask what makes them microglia angry and what causes the transections (cutting of the axon). We know that active inflammation causes 11,000 transections/mm2. 

What made the microglial activated then this is where T and B cells will come in, they trigger/create the environment to activate microglial and make astrocytes gliotic, 

However, this is where there are aspects of progression that do not respond to certain anti-inflammatory drugs. That astrocytes contribute to nerve survival is well known and they do such things as control the lactate that impinges on energy levels of the nerve and its survival

We have been saying for some time that you have to control the glial inflammation to control progression. There is a toxic factor. What is it? This is a new part of a jigsaw, but this does not change my world view. 

If it thrusts the astrocyte into the limelight… great…it is an underestimated cell.

This aspect that was targeted in the Progressive Alliance by a group in Harvard. We know that astrocytes control the glutamate-glutamine cycle and we know how important this may be in glutamate excitotixicity killing nerves. In this case they produce toxic factors that did not support nerve survival and also they block the proliferation of oligodedrocytes and so would inhibit remyelination. It has been known for some time that astrocytes block remyelination

In this study they found that caspase (cell death effectors) inhibitors blocked death caused  by the astrocyte derived factors. We showed this a few years ago to block  nerve loss. 

Neuroprotection in a novel mouse model of multiple sclerosis.
Lidster K, Jackson SJ, Ahmed Z, Munro P, Coffey P, Giovannoni G, Baker MD, Baker D. PLoS One. 2013;8(11):e79188.

However, we could not get the company interested producing the treatment agent to try the approach in MS, and we never got any funding to develop this, so it is interesting that pharmacological inhibitors are becoming available at least for experimental use…However, they may have all sorts of problems if used globally. 

Likewise there is perhaps issues with the statement that there are FDA approved inhibitors of the molecules. Whilst these A1 cells are  formed by TNF, blocking this in MS is a no-no. It makes MS worse or causes demyelinating disease.

Interleukin-1, blockade of this may be interesting but as it is a globally used cytokine involved in the generation of all sorts of cells, it will not come without side effects.

Is this enough to treat progression?

Comabella M, Julià E, Tintoré M, Brieva L, Téllez N, Río J, López C, Rovira A, Montalban X. Induction of serum soluble tumor necrosis factor receptor II (sTNF-RII) and interleukin-1 receptor antagonist (IL-1ra) by interferon beta-1b in patients with progressive multiple sclerosis. J Neurol. 2008; 255:1136-41.

Soluble receptor blocks TNF and IL-1Ra blocks IL-1, but is beta interferon great for progression…not really.

However, the plus side it gives a focus on astrocytes..on the bad side we will now see a sea of uninterpretable acute EAE experiments with astrocyte conditional knockout mice that are not particularly directly relevant to astrocyte activity in progressive MS:-)

About the author



  • Is this post too long?

    Part of the function of the blog is to support our own work and it is so it is important to mention our output.

    • No, post is not too long 😉
      Thanks for link to your work from four years ago, I see you had funding from the MS Society, do they often support your research?

    • I would not be here if it was not for the MS Society. The MS Society supported me throughout my time as a young scientist and I can't thank them and their supporters enough.

      I started research in MS because I got the Angela Limerick MS Fellowship aimed at bringing people into MS research. This was an endowed position created by the Angela Limerick Estate, which was administered via the MS Society.

    • Hey you two mice that's really encouraging to hear. Will now support MS Society with increased enthusiasm. I wish they'd make clear in their accounts how funds are spent. I am guilty of being sceptical of this c £28m charity and have generally favoured the much smaller MS Trust.
      This changes everything.

    • The MSTrust would not fund basic research and focu what they support and they are great MS nurse champions

      The MS Society publishes stuff they fund but their research budget is not great and smaller proportion than when it was based in Fulham. However, now there are many more staff in the Society and their remit is much broader.

      Currently the NMSS in the US is funding some of our work

    • Eventually found it, deep in the notes to the accounts, in 2015 MSS awarded £4m in research grants, spent £10.5m on staff, £600k for executive directors and £129k for CEO. Interestingly they also have £18m of fixed assets, not entirely sure what they're saving for we need research today not in years to come 😉

    • The research grants however are not necessarily open competition, e.g. they used to fund brain bank, MRI scanner at Queen Square, clinical trials unit activities, myelin repair centres. I believe the funding for scanner terminated and we were told that we have to pay for bank bank tissue so I suspect the support is limited but they fund trials and this eats all their cash

      The MS society also have different branches and I suspect some money remains in the branches and is not available for the central Society in London to use.

  • Furthermore it takes space to put the work in context especially as the report implies that there is a quick use to target these cells eg using TNF antibodies. This in my opinion is unlikely. Hence showing examples where the logic could fall down

  • Next the post was made in response to a direct request to comment on "the ground breaking research". Check out yesterday's comments if you don't believe this

  • This publication was dense, put the neurons to work.

    So are A1 astrocytes responsible for weakening BBB and allowing hemoglobin to enter the CNS? The order of "events" of the EM could be: the microglia is activated, which would eventually generate A1 astrocytes that would lead to the death of the oligodendrocytes, nerves, neurons, etc. With the death of the oligodendrocyte fat molecules would be dumped into the CNS, which would activate T and B cells and recruit them into the CNS, leading to further inflammation and damage?

    • These astrocytes are not the ones controlling BBB function.
      We can make many scenarios you idea could be one way

  • Thanks for this review MD, much appreciated. I did not know your team worked on this as well-respect. This is published in one of the most prestigious medical journals-Nature and its importance in terms of understanding progressive MS should not be understated.

    If you figured out what the APC in the activated microglia was, would you likely stop MS progression completely? If you halted "angry" microglia and therefore its 3 cytokines you listed would it prevent progression? Is this how Ibudilast is proposed to work? If you halt A1 or gliotic astrocytes would you also halt the disease?

By MouseDoctor



Recent Posts

Recent Comments