“The immune system is needed for repair and lifelong maintenance of the brain”


Two decades ago, it was anathema to go against the established idea that the nervous system doesn’t need the immune system at all for its protection or repair. But Michal Schwartz, who pioneered the whole new area of neuroimmunology, is confident that her latest work, which she presented a few weeks ago at the 5th Champalimaud Neuroscience Symposium in Lisbon, may soon radically transform the treatment of Alzheimer’s disease.

Michal Schwartz was born in Israel to a Holocaust-survivor father and a fifth-generation-Israeli mother. Her small, slim-boned frame, her big eyes, her oval face, her head of grayish-blond cascading curls, her soft voice, belie a singularly strong personality. Michal Schwartz doesn’t waste any words when talking about her professional and family life – she goes straight to the point. In fact, she is a determined, strong-headed woman. For years, she had to contend with the open opposition to her claims and results about the relationship between the brain and the immune system, not only from her own colleagues at the Weizmann Institute of Science, near Tel Aviv – where she still works today – but also from the scientific community at large. In spite of this hardship, as she told us with a faint smile, she survived – and today nobody questions the importance of her work anymore. On the contrary.

Why were you skeptical of the dogma that the brain and the immune system exist back-to-back, so to speak?

I wasn’t skeptical. What was happening was that, for many years, neuroimmunology was dominated by immunologists working on diseases where the immune system attacks the brain. As a result, the common wisdom was that immune cells were not allowed to enter the brain. And if they did enter, they caused pathology.
When I started working in this field I thought: why would a tissue as indispensable as the brain – we will never be able to replace the brain – have given up the opportunity to be assisted by the immune system? It doesn’t make sense; we have to revisit the whole relationship between the brain and the immune system. Maybe they have developed some kind of relationship that is different from what happens in any other tissue.
At the time, I really believed that the immune system was never used for brain repair, but I thought the rules might be different. So I started to revisit this idea and that’s how it happened. We were the first in the world to suggest and demonstrate that the immune system was actually needed for brain repair.

How did your peers first react to this “crazy” idea, to this unheard of notion? Did it take them a long time to accept it? Have they accepted it?

Nowadays, they have. But initially, it was so unbelievably difficult… there was a lot of opposition. Luckily, my first two papers that changed the dogma were published in Nature Medicine, but even so, they were totally ignored or cited in a negative way. This went on for a while. Many young scientists at that time built their own careers by publishing against me. But, you know, in retrospect, skepticism is good in science and since I was really the pioneer, it was my responsibility to provide more and more evidence to substantiate my new concept.

Why do you think nobody had looked into the relationship between the brain and the immune system before?

One reason is that the first observations of the immune system in the brain were made under pathological conditions, so it was very easy to believe that the diseases were caused by the immune system entering the brain. The other reason is that the brain is embedded in a barrier, the ‘blood-brain barrier’ and, at the time, it was thought that this barrier existed to protect the brain from the immune system. This is correct, but what we discovered was that there is another barrier, the blood-cerebrospinal fluid barrier (the cerebrospinal fluid, or CSF, is a liquid present in the brain and spine) which allows communication between the brain and the circulation.

When did you know for sure that you were right? Which  experiments proved this to you beyond a reasonable doubt?

It’s very difficult to identify the moment you realise you are right in science, because you are building a concept step by step. I knew that I was in the right direction, and every experiment I did added more pieces to the puzzle. You start to realize that you’re wrong when you do experiments and they don’t fit into your concept – and so you have to review and revise your working hypothesis. However, over the years, the more experiments we did, the more they supported our view and more pieces were added to the puzzle.
We reached a turning point when we published in 1998 and 1999, and a further big turning point happened in 2006, with a paper we published in Nature Neuroscience, where we demonstrated, for the first time, that the immune system is needed for repair and lifelong maintenance of the brain. This came as a big surprise, because in the healthy brain you don’t see lymphocytes (a type of immune cell).

How did you prove that the immune system was also critical for maintaining brain health?

In fact, what we demonstrated in 2006 was that, unexpectedly, healthy brain plasticity, neurogenesis and cognition are tightly dependent on the immune system. That was the beginning of a new era in our research: we set out to investigate how the immune system is able to act on the healthy brain if immune cells are excluded from entering it.
However, the biggest turning point happened in 2013, when we discovered that there is an interface through which the brain and the immune system actually communicate.
In fact, as I said before, this interface is not a barrier between the brain and the blood but between the CSF and the blood. It was a very exciting instant in our careers when we discovered that monocytes (another type of immune cells) can get through this barrier.

Gateway to the brain

Most people have heard about the blood-brain barrier, which keeps intruders – bacteria, many drugs and also the immune system – out of the brain. But what you are saying is that the interface you discovered is actually permeable to certain immune cells, more like a gateway than a real barrier. How does it work?

The blood-brain barrier is made very tightly connected cells that line every blood vessel in the brain. It’s really, really a barrier. But the blood-CSF barrier cells are not tightly connected.
When you sustain a nervous system injury, molecules associated with the pathology come from the brain or the spinal cord through the CSF and approach this barrier, which senses them. The barrier is then activated and this can allow monocytes or T-lymphocytes present in the blood that express certain specific molecules to cross the barrier, get into the CSF and patrol the brain.

Where in the brain is this interface situated?

In the choroid plexus, a layer of cells lining vessels in the ventricles of the brain, where the CSF is produced.

Does the fact that immune cells cross this barrier mean that they actually get into the brain and interact directly with neurons?

That’s a good point. In the healthy brain, they don’t get into the brain. The T-cells, for example, are sitting at the barrier and they enter into the meninges and circulate in the CSF. They affect the brain from afar by releasing molecules that the brain needs and by activating the epithelial (barrier) cells to produce molecules that the brain needs. When you have a pathology like Alzheimer’s or an injury, this leads to signalling from inside the tissue itself that brings immune cells to the site of the pathology. But in the healthy brain they don’t enter.

So these immune cells do not reside in the brain, they reside on the other side of the barrier and they control,  as if by remote control, what’s going on in the brain and if everything’s all right?


You coined your new theory ‘protective autoimmunity’. It was a pretty provocative name, your colleagues must have taken it as a contradiction in terms…

Yes, but I think it’s becoming more and more accepted. First, all of us have these autoimmune T-cells and we don’t get autoimmune diseases. So we need them. Secondly, in cancer, it’s now clear that you need T-cells that recognize the tumor (we might object that the tumor is non-self, but even so, it’s expressed in the body). So it’s a matter of time until we find out that autoimmunity is not a synonym for autoimmune disease, and that it only develops into autoimmune disease when it’s no longer under control.

You also showed that when immune cells do enter the brain, they are first “trained” to fight against whatever is injuring neurons. Is this like the acquired immunity the rest of our body develops when we’re vaccinated against a disease? Does this acquired neuroimmunity also have a ‘memory’, do these cells remember what to do next time the brain has the same problem?

In principle, we believe that this is true, but we don’t have proof for that. We did prove the opposite: that if you vaccinate animals with a brain self-antigen (a molecule produced by the brain’s own tissue that elicits antibody production in abnormal situations), you increase their resilience to brain injury – the recovery from injury is better, the recovery from stressful conditions is better. In other words, if you artificially create memory T-cells to self-antigens, you cope better with brain injury or pathology.
So in a way we proved it, but we don’t as yet have strong proof that when we injure or we accumulate injurious conditions, we are accumulating memory T-cells. I believe that we are, but we still don’t have strong enough evidence.

Fighting Alzheimer’s with autoimmunity?

Patients with Alzheimer’s have been treated with anti-inflammatory drugs. Do you think these treatments could have favored the progression of the disease? Are they still being used?

I doubt that they are. There were several clinical trials with Alzheimer’s patients where steroidal and non-steroidal anti-inflammatory drugs were tested. None of them were actually effective, and I think there was even some indication of adverse effects. Nowadays, however, the physicians are not recommending any anti-inflammatory drug.

What about immunosuppressive drugs? Wouldn’t this be a bad strategy, given that the immune system is critical for brain repair?

Same thing here: physicians don’t prescribe them anymore.

But the fact is that the wrong treatments were used against Alzheimer’s for a long time.

That’s because when inflammation was observed in Alzheimer’s patients’ brains, the common dogma lead people to believe that the immune system was causing it. But now, based on our understanding, we know you have to fight inflammation not by suppressing, but by activating the immune system, in order to allow immune cells to get into the brain to tackle it. So we certainly need to resolve the inflammation, but the way to do it is exactly the opposite of what was previously thought.

You published a paper in the beginning of 2016 that shows that boosting the immune system of mice with Alzheimer’s not only slows disease progression, but even reverses both the neural damage and the cognitive decline in these animals. Is this the equivalent of immunotherapy for cancer patients?

Yes, in many ways it is, and when we discovered this, it was the most exciting moment in my career. Prior to that, however, we had published three articles, from 2013 to 2016, all of which demonstrated that, in order to maintain the communication between the brain and the immune system, you need to activate the choroid plexus epithelium (the blood-CSF barrier).
First, we found that to keep this gateway active you need interferon-gamma, which is a protein produced by the T-cells in the gateway. We further showed that, in the aging brain or in age-related dementia and Alzheimer’s, this gateway was not operating optimally. Then, we found that in these pathologies there is a deficiency in the production of interferon-gamma by the T-cells in the barrier.
Finally, in 2015, we showed that if we reduced the immune suppression, we augmented the level of interferon-gamma, thereby activating the gateway.
Based on all this, the way was paved for us to go and check whether a treatment called immunotherapy – very successful against cancer tumors by targeting the immune system rather than the tumor – would also work in the brain.
Cancer immunotherapy revitalizes the immune system, and it’s the immune system that then fights the tumor. So we thought we were in a similar situation.
We used the antibody that is used in tumor treatment, and we realized that we augmented the immune system response and then drove a cascade of events that starts in the periphery and brings immune cells to the brain. With this, there was really a complete reversal of the disease: the animals recovered cognition, and many of the characteristics of the pathology were eliminated.

How does the antibody work?

By blocking an ‘inhibitory immune checkpoint’. In animals and humans, the immune system is kept under tight control: on one end, it is restrained to prevent autoimmune disease, and on the other it is stimulated to allow it to work when you need it without delay.
The inhibitory immune checkpoints are molecules which keep the immune system from overshooting. So, in order to remove this restraint and boost the immune system, you need to administer antibodies that are either directed against these inhibitory molecules themselves or against their receptors on the immune cells, the two parts of this communication which restrains the immune system.
We used these anti-inhibitory immune checkpoint antibodies, specifically, an antibody directed against a protein, Programmed cell Death 1 (PD-1), which has shown a very successful effect against some tumors. We are not planning to use the exact same protocol as in cancer immunotherapy, because the mechanism we discovered in Alzheimer’s is different, involving different types of immune cells. Nevertheless, all the required immune cells in the case of Alzheimer’s expressed the receptor PD-1, and our strategy was so successful!
In essence, our idea is the same: to use antibodies that will reduce the restraint on the immune system, and the beauty of it is that the treatment is not targeting any pathology in the brain. It’s directed to the immune system and the immune system does the job.

How do you administer this drug? Directly into the blood?

Yes, in the circulation. You don’t need to treat the brain, that’s the beauty of it.

Could this potentially have huge implications for the treatment of Alzheimer’s? What about other degenerative brain diseases and acute injuries of the central nervous system?

About acute injuries, I can’t say, because the therapeutical dose is different. However, for Alzheimer’s, the implications are immediate and we believe that within a short time, we’ll be able to get it to the clinic. It’s beyond my control, because it requires financial support and assistance from the industry, but this is our vision and our expectation. It’s very important to say that we tested the treatment at various stages of the disease, and it was effective at every stage – early, middle and very severe. So we’ll move on to Alzheimer’s patients.
With respect to other neurodegenerative diseases, I believe that the concept will be applicable. Whether we will use the same immune checkpoint or others, we don’t know yet, we are currently testing this.

But in fact, these molecules are already used in cancer, so couldn’t they be used also in this case?

We will not use the same antibody approved for cancer.

When do you think this will get into the clinic?

We’re planning to get it into the clinic in one-and-a-half years.

Autism, schizophrenia & Co.

In another paper in 2016, you showed that alterations of the immune system in both a pregnant woman and her foetus, following the mother’s infection by a herpes virus (cytomegalovirus) could increase the risk of the child later developing autism or schizophrenia. Could this also open totally new avenues for the prevention/treatment of these psychiatric conditions?

First of all, the work was done in tight collaboration with an outstanding young scientist, Kuti Baruch, also from the Weizmann Institute, so it is not my work only. The novelty here is not that the virus causes autism and schizophrenia; this has been known for decades. The novelty is that we discovered that the virus affects the brain’s immune system at an early stage and interferes with normal development.
This has opened a new understanding in the field of autism and schizophrenia that we hope will be transposable to the clinic. We are currently working to understand the signalling that mothers are producing as a result of the viral infection which affects the foetus. If we discover this signaling, we will know it affects the foetus for these diseases emerging later on in life, and we will be able to perturbate it.

So, to sum up, what you are saying is that the health and integrity (and also the plasticity) of the brain depend on the immune system. Does this mean we can fight cognitive decline just by keeping our immune system healthy?

In theory, just by keeping our immune system healthy we can postpone cognitive decline. So I believe that this is definitely helpful. How to keep the immune system healthy is another question, and there are many genetic, nutritional and other aspects which are beyond my research.

Strong-headed woman

Tell us a little about how you came to immunology.

All my life, I’ve had a passion for science, but it went up and down, based on the topic I selected for my research. But as soon as I selected this topic, I knew it was right for me.

How many hours do you work daily? If you could, would you be in the lab 24/7?

Yes I would, but I have four kids. My philosophy of life is to try to do the best, and I decided, because I’m a mother, to do the science and to be very little involved in any other issue beside science, because of my kids.
So I used to work every day, from early morning till early afternoon, around 5p.m., then go to my family and then go back to work every night. This was my style for many, many years. But it’s not simple with children and they claim that they paid a penalty for it.

Has the passion you have for your work made up for all the opposition you had to endure?

Yes. But it was very difficult because there was opposition abroad and a lot of opposition inside the institute. It was also very difficult to get grants. Definitely, it didn’t help my health, but I survived.

You must be a very strong-headed woman to have taken on this battle, which was so difficult in the beginning, and to go on fighting…

I adapted to my own career this very nice remark made by Abraham Lincoln, during the revolution that he was leading, the social revolution. He said that when he headed this revolution he had two options: either to stop doing it and try to convince all his opponents that he was right or to keep on doing the work. He further said: if I try to convince my opponents, it will take too much of my time; and if I am right, that won’t be necessary, while if I am wrong, it won’t help to convince them. So that’s what I did, I decided to keep on doing the work. It’s not easy, and if you are a woman, it’s even more difficult, and I’m not sure it would be easier today.



Ana Gerschenfeld works as a Science Writer at the Science Communication Office at the Champalimaud Neuroscience Programme



Edited by: Catarina Ramos (Science Communication office), Ivo Marcelo (section editor),
Clara Howcroft Ferreira (editor-in-chief)
Photo Credit: Francisco Romero (PhD student)
Video Credit: João Camilo (Science Communication office)


Loading Likes...