Antibodies From Children With PANDAS Bind Specifically to CINs and Alter Their Activity

Interview with Christopher Pittenger, MD on the study: Antibodies From Children With PANDAS Bind Specifically to Striatal Cholinergic Interneurons and Alter Their Activity

ASPIRE sat down with Christopher Pittenger, MD, PhD, FAPA, FANA. He is an Associate Professor of Psychiatry; Assistant Chair for Translational Research, Psychiatry; Director, Yale OCD Research Clinic; Co-Director, Neuroscience Research Training Program. We discussed the results of the translational science study, Antibodies From Children With PANDAS Bind Specifically to Striatal Cholinergic Interneurons and Alter Their Activity.

Antibodies From Children With PANDAS Bind Specifically to Striatal Cholinergic Interneurons and Alter Their Activity Jian Xu, Rong-Jian Liu, Shaylyn Fahey, Luciana Frick, James Leckman, Flora Vaccarino, Ronald S. Duman, Kyle Williams, Susan Swedo, and Christopher Pittenger. Am Jrnl of Psychiatry 16 Jun 2020

Christopher Pittenger, MD, PhD, FAPA, FANA Full Biography

Gabriella: Can you provide us a little bit of background on this study?
Dr. Pittenger: We have been working for a few years on this study. This work was initially initiated by Kyle Williams when he was in my lab, doing his Ph.D. thesis some years ago. Then continued by a postdoc, Luciana Frick, and then another postdoc, Jian Xu; but it’s finally borne fruit.

Gabriella: In simple terms, what was the main question you were seeking to answer in this study, and how did you approach it?
Dr. Pittenger­: We posed the question, how can we better characterize the antibodies that cause the problems in PANDAS?

And we approached this in a very neutral way. Meaning, we know that there’s controversy around PANDAS. We know there have been findings where people have found antibody binding. But my read of the literature is that those have not proved to be terribly clinically useful and have not replicated consistently. That is obviously a matter of ongoing discussion about the Cunningham panel and the various groups that are looking at this. There certainly isn’t convergence or acceptance on the antibodies. So, we asked ourselves why that might be? Why is there not a consensus about antibodies and PANDAS?

Gabriella: Why isn’t there a consensus about the antibodies involved in PANDAS?
Dr. Pittenger: One possibility is there are no antibodies to be found. But we’re going to put that idea aside, for now.

To move forward, we said, we believe that there are antibodies but asked why they have been hard to characterize unambiguously? And there are a few reasons. One is, there may be a lot of different antibodies, and there may be different antibodies in different kids, so that might make them hard to find. And another reason might be that the techniques that people have used to find them just haven’t been the right techniques.

Gabriella: What were the techniques you chose to use, and why?
Dr. Pittenger: We, meaning Dr. Swedo, others, and myself, decided to look at a small number of kids and look at them very, very carefully. We chose not to look at 100 kids who are all different, as we knew we would not see anything consistent. Instead, we looked at five at the beginning. This first stage was done mostly by a postdoc in the lab at the time, Luciana Frick.

We chose not to use the old fashioned techniques in which people look at proteins and see which antibody sticks to which protein. This is usually done with denatured proteins, dissolved in soap, which causes them to lose a lot of their structure. So, if it works, it’s great, but there are lots of reasons it might not work. So, we went all the way to the end, and we did exactly the opposite. We looked at how the antibodies bind in the living brain. In this case, we used mouse brains. Instead of looking for the specific molecules, we looked for the cells the antibodies bind to, which means that even if antibodies from different kids bind to different molecules, we might still be able to find something consistent. And we did!

We found that the antibodies from these first five kids with PANDAS bind to specific cells in the striatum, a part of the brain that we already know is involved in OCD and Tourette syndrome. They’re called interneurons. There’s a small number of them, but they’re very important.

These are cells that we already know are involved in Tourette syndrome, from work in post-mortem brain by my colleague Flora Vaccarino here at Yale: adults with Tourette syndrome lack the same specific striatal interneurons. This suggests that problems with these cells may play a role in several conditions. And some previous work in my lab shows that if we muck with these cells experimentally, we get twitchy mice. So, we already know these cells are important in regulating behavior, and we already know they have something to do with Tourette syndrome. We don’t know about OCD. So finding the same cells as a target of antibodies in PANDAS was very exciting. We published that work two years ago now.

Gabriella: Nothing like a positive finding to get a PANS PANDAS advocate charged up. Okay, so that is a bit of the background; let’s talk about the study you just published. In layperson’s terms, what are “cholinergic interneurons (CINs) in the striatum?”
Dr. Pittenger: The striatum is an area of the brain, part of the basal ganglia; in humans, it’s divided into the caudate and the putamen. It’s been a target for research in OCD, Toruette’s, and PANDAS for years. Cholinergic interneurons (CINs) are one of several different types of interneurons in the striatum. They are critical regulators of brain activity. They modify the signaling of other cells nearby. The striatum is involved in voluntary motor control; it has a role in cognition and reward processes, and it is involved in OCD. So we wanted to look further into what affects these antibodies associated with PANDAS has on the CINs.

Dr. Pittenger: This study, Antibodies From Children With PANDAS Bind Specifically to Striatal Cholinergic Interneurons and Alter Their Activity, is the follow-up to the study we published a couple of years ago. We replicated those initial findings about interneurons. In the first study, we only used five kids. But five kids is not a large enough study to really crack the larger problems. So with Sue’s help, we looked at three different groups of kids with PANDAS, a total of 27, compared to 23 kids without. This work was spearheaded by Jian Xu, a current postdoc in the lab; it was really a heroic amount of work.

Gabriella: Did you use the same technique as the first study?
Dr. Pittenger: Now that we knew what we were looking for, we used a simpler technique, which allowed us to do things more quickly. We ran a bunch of controls.

Gabriella: What did you find?
Dr. Pittenger: So let’s start with what we know or think we know. PANDAS may result from induced autoimmunity against brain antigens. Our previous study suggested that PANDAS patients may have a high level of IgG antibodies concentrated in the striatum that may bind/attach to cholinergic interneurons (CINs) in the striatum. We replicated that here: we found higher binding to CINs by antibodies from the kids with PANDAS. And, importantly, we did not find higher binding to other interneurons in the striatum, or to a bunch of other cell types we examined. So the effect seems to be pretty specific.

If you mess up the activity of the CINs, it would disrupt the signaling of other nearby cells. In this new work, we found that when the antibody attaches to these CINs, their activity is reduced. We already know that CIN deficiency has also been associated with mice with repetitive behavioral pathology and with humans with tics like those with Tourette Syndrome. So it makes sense that an antibody binding to these cells and disrupting their function could lead to similar effects.

11 of these kids were treated with IVIG and got better. (We specifically chose kids who got better with IVIG treatment, from a study that Sue Swedo, Kyle Williams, and others conducted a few years back.). When we looked at the after treatment with IVIG, the IgG binding to CINs was reduced, and this reduction correlated with symptom improvement. Meaning that when they’re treated with IVIG and they get better, the binding to CINs goes down, and so does the effect on their activity. So we think it probably means something.

Since these antibodies can make those cells function suddenly less well, this is potentially an explanation for how they cause disease. Simply put, you get antibodies, they stick to these cells that reduce their function that causes the whole system to go a little haywire, and you get a twitchy mouse – and perhaps a twitchy kid.

Gabriella: What is translational science? And how does this study relate to Dr. Dritan Agalliu and his lab’s work at Columbia on the Blood-Brain Barrier? *1

Dr. Pittenger: Translational science is using observations in patients to guide studies in the lab, and then using the results found in the laboratory to guide the development of diagnostic tests, new therapeutics, medical procedures, or behavioral interventions. The study done here is maybe a first step in that direction. We don’t have anything like a diagnostic test or a new therapy yet. But we have a new, very specific idea about where to focus efforts in that direction going forward.

We’re asking different questions than Dritan; the two bodies of work complement each other. We’re asking what these antibodies can bind to in the brain, and what the consequences are. But in order for this to happen, the antibodies must get there in the first place. That’s what Dritan and his lab are focused on. I hope that, over time, these two lines of research will come together to give us a more complete understanding of what’s going on in kids with PANDAS and PANS.

Gabriella: Taking your research and the research done by Dr. Dritan Agalliu, let’s break down what the research suggests might happen to a patient who develops PANDAS and how your research could possibly inform future diagnostic methods and treatment. Let me know if this correct.

  1. Healthy Patient is exposed to GAS/strep infections.
  2. The body mounts an immune response against the strep by producing antibodies. Some patients, due to an unlucky combination of genetic susceptibility, the specific strain of strep they are infected with, and perhaps other factors, produce antibodies against the strep that also attack the body’s own cells, specifically in the Basal Ganglia, in the case of PANDAS.
  3. Repeated exposure to GAS results in the production of proinflammatory TH17 and TH1 lymphocytes (types of white blood cell, part of the immune system). TH17 cells may travel along the olfactory nerves to the brain. They release proinflammatory cytokines (chemical messengers that help regulate immune response) that cause microglia (mediate immune responses in the central nervous system) to release other inflammatory cytokines.
  4. Cytokines from Microglia and TH17 lymphocytes prompt the breakdown of the blood-brain barrier (BBB ), which is made up of endothelial cells. The BBB is supposed to prevent specific molecules in the blood from entering the brain. So when it breaks down, large molecules like antibodies can more easily get into the brain and cause inflammation.
  5. Antibodies to the GAS in the blood enter the brain via the weakened BBB. Some of them bind to cholinergic interneurons (CINs), impairing their function.
  6. Reduced activity in the CINs disrupts the signaling of other nearby cells and thus the normal function of the striatum.
  7. This disruption leads to the onset of symptoms.
  8. With treatment, we can seek to remove both the source of infection and the pathogenic antibodies. When it works, treatment with IVIG may reduce or eliminate the antibodies that bind to the CINs, allowing them to function properly.
  9. If all goes well, symptoms improve.

Dr. Pittinger: That’s a plausible series of steps. I have to caution that a lot of them aren’t proven to actually happen in this way, in this order, in patients – this is more a set of hypotheses that we’re trying to test than a road map to what is happening in each and every patient. I’m sure that some of these ideas will prove not to be exactly right and will evolve as we do more experiments, and I’m sure that patients will turn out to differ from one another in complicated ways. But this is a reasonable outline of our current thinking.

Gabriella: You looked at a very strict PANDAS criteria; do you think you would get similar results in a PANS patient triggered by a sinus infection or mycoplasma pneumonia?
Dr. Pittenger: This is a key next question. I expect there to be overlapping mechanisms, but I doubt that every kid with PANS is going to have exactly the same things going on. We started with strictly defined PANDAS to reduce the variability, figuring that that would make it easier to find something meaningful. Now that we have a specific hypothesis to test, we’re going to look at PANS more broadly. I’m working on that grant right now!

Gabriella: Did you find a biomarker for PANDAS?
Dr. Pittenger: It is a little strong to say this is a biomarker for PANDAS.  What we found isn’t clear cut enough to be used as a diagnostic test, and we don’t see it in every single patient. I think of it more of an important step forward in understanding the pathophysiology. I’d call it strong evidence for striatal CINs as a critical cellular target that may contribute to pathophysiology in children with rapid-onset OCD. Turning this into a clinically useful biomarker, or developing new treatments based on it, is a long-term goal.

Gabriella: How many times was this study replicated, and why is that important? Will it help squash the controversy over whether PANDAS is a real disorder?
Dr. Pittenger: I think that clear biological data like this help bolster that case that PANDAS is a real diagnostic entity. Over time, as we look at more heterogeneous groups of patients, this may help clarify exactly what the appropriate boundaries of the diagnosis is. We took great pains in this study to make everything as rigorous as we possibly could, and to replicate everything. I’ve had three different people do these experiments using two different techniques across three different cohorts of patients. In all, we see the same results consistently. This study does not have a large cohort, but it is a significant increase over the preliminary studies; the early study looks at five kids, this looks at 27 PANDAS kids and 23 control kids. So replicating this study three times with this size cohort is significant.

Gabriella: Do you anticipate that there will be a way to test humans for this? Would it be a spinal tap or a blood test? What impact do these results have on treatment possibilities
Dr. Pittenger: Not yet, hopefully, in the future. A critical next step is to find out exactly what the antibodies in these kids are binding to – not just the cells, but the molecules. We’re working on that. Testing for antibodies to specific proteins is much easier and much easier to turn into a clinically useful assay than the binding to cells we’ve examined in this study.

Gabriella: What are the next steps?
Dr. Pittenger: I approach this as a careful scientific skeptic. I don’t have a long history, clinically, in PANDAS. I am a psychiatrist, and I mostly treat adults with OCD. But I’ve come into this more as a skeptical neurobiologist. I think we’ve got a finding here that that has some legs.

What we want to do now is scale up more. So we scaled up from five patients to 26. I want to scale up to 200. I want to look at kids who have PANS but not PANDAS. I want to look at kids with Tourette’s or pediatric OCD, or perhaps adult OCD, who don’t have PANDAS. We want to see if this is specific to PANDAS, or is something that’s seen more generally? Either answer to that question is interesting. So a first next step is to look at a lot more clinical samples. The second step, as I said before, is to try to find the specific antibody. We know these antibodies are there because we see their effect. We know the cell they are sticking to, but we don’t know the molecule they stick to. I suspect that there won’t be just a single answer to this question – that there will be a group of antibody targets that can lead to PANDAS. If we can find those, then we’ll be significantly closer to something that could have a real clinical impact.

Gabriella: Thank you so much for taking the time to talk about this study. It is so helpful to have you clarify the study’s key points, the history of the preliminary studies, and what this means for the PANDAS community. As you said, you think these findings have some legs. We look forward to you proceeding further into this area of study. 
Dr. Pittenger: You are most welcome. Onward and upward!


Related Articles:


Platt MP, Agalliu D, Cutforth T. Hello from the Other Side: How Autoantibodies Circumvent the Blood-Brain Barrier in Autoimmune Encephalitis. Front Immunol. 2017;8:442. Published 2017 Apr 21. doi:10.3389/fimmu.2017.00442

Cutforth T., DeMille MMC., Agalliu I. and Agalliu D. (2016). CNS autoimmune disease after Streptococcus pyogenes infections: animal models, cellular mechanisms and genetic factors. Future Neurology: 10.2217/fnl.16.4.

Dileepan T, Smith ED, Knowland D, Hsu M, Platt M, Eddy-Bittner P, Cohen B, Southern P, Latimer E, Harley E, Agalliu D* & Cleary PP* (2016). Group A streptococcus intranasal infection promotes CNS infiltration by streptococcal-specific Th17 cells. Journal of Clinical Investigation 126: 303-317. *equal authorship

Leave a comment