[intro music]
Host – Dan Keller
Hello, and welcome to Episode Forty-Four of Multiple Sclerosis Discovery, the podcast of the MS Discovery Forum. I’m your host, Dan Keller.
This week’s podcast features an interview with Dr. Monika Bradl, who discusses animal models of neuromyelitis optica, NMO. But first, here are some of the new items on the MS Discovery Forum.
According to our curated list of the latest scientific articles related to MS, 69 such articles were published last week. To see the list, go to msdiscovery.org and click on Papers. We selected two of those papers as Editors’ Picks. One – on the use of MRI in NMO –included no fewer than 48 co-authors, a veritable Who’s Who of prominent MS researchers. The other editor’s pick, which had “only” 36 co-authors, was a large study providing strong evidence that disease-modifying treatment reduces disability worsening events in clinically isolated syndrome and early MS.
.
[transition music]
Now to the interview. Dr. Monika Bradl is an associate professor in the center for brain research at the Medical University of Vienna, Austria. I talked to her in her office about her work with animal models of neuromyelitis optica to probe what occurs in the early stages of the disease. She first describes why animal models are important.
Interviewee – Monika Bradl
NMO is a very rare disease, and so you have the problem that you get only very little pathological material, and so when you want to know what's going on at the very beginning of the disease you have to use animal models. And so our pioneer work there in the NMO field was to find out whether the antibody that characterizes about 80% of NMO patients, that's an antibody directed against aquaporin-4, water channel astrocytes, is pathogenic or not. And so what we did is that we injected this antibody then in experimental animals. When we had the animals without any additional CNS inflammation going on, they remained completely fine, and that was at that time a bit of a debate because people thought that the antibodies could enter the central nervous system anyhow.
But then it turned out that this failure of the antibody to reach the uninflamed brain had also predecessor in humans. There they had an NMO patient in Japan who was diagnosed with NMO, and when they found that he has pathogenic antibodies, they were first afraid because this patient was blood donor with the Japanese Red Cross, and so at that time then they stored serum samples of all the blood donors for quite some time, and they found out that this person had pathogenic antibodies already for more than 10 years without showing any signs of disease.
And so this was then the human patient correlated to what we found in our NMO animals, and what we then also saw is immediately when we make our CNS inflammation with CNS-specific T cells which break open the blood-brain barrier, then the antibody gets access to the central nervous system, binds to the astrocytes, and then induces two different type of killing procedures. So the astrocytes are then killed with either the help of complement or with the help of a mechanism that's called antibody-dependent cytotoxicity; so both of these mechanisms are then responsible for tissue destruction.
Interviewer – Dan Keller
Getting back to the Japan patient, did they also follow the recipients of that person's blood? It seems like this patient did not have that second hit which would allow the antibody to cause problems, but giving it passively to someone who already had the first hit might cause a problem. Did they look at the recipients?
Dr. Bradl
I'm sure they did, but there are no records about it to my knowledge. They might have been published in Japanese in some of these Japanese journals, but not in the international journals. But I'm quite sure that there was no immediate transfer of the disease with these antibodies because that would have made headlines. So one can conclude from that that this must have been harmless.
MSDF
And what animal models are you using?
Dr. Bradl
We are using rat models, but there are other groups that are working in mice. We use Lewis rats and we think they are great because the rat complement works with the human antibodies, so it provides the help. And we have an NMs strain the Lewis rat which is extremely susceptible to all different types of autoimmune diseases, and so therefore we like rats and their CNS is larger and nicer. But people who work with mice, they also have advantages because they can use the entire transgenic zoo of knockouts or gene-mutated animals, and with this they can learn more about the contribution of individual molecules to the disease process.
MSDF
Now that you bring up the mice, are some mice more susceptible based on MHC than others; are some resistant?
Dr. Bradl
There you have to consider one peculiarity of the mouse system. If you use mice, then you have the wrong complement system. So no matter what kind of inbred strain you use, you have to transfer human complement along with the human antibodies to get an effect, plus people who use the mouse model directly inject complement and antibodies into that brain to circumvent the blood-brain barrier. And when they do that, the MHC type of the particular mouse strain doesn't play a role.
MSDF
Is this using only passively transferred antibody, or do you try to raise antibodies by injecting antigen or modifying antigen?
Dr. Bradl
Yup. We desperately try to do so, but I have to say that this was not a real success story. So we first tried, as many other people did, to use just convention and normal aquaporin-4 as it is normally produced, or longer fragments of this, but obviously this does not work. And we now know that the antibody recognizes its target only if the aquaporin-4 is correctly folded within the same membrane. And only if this is the case, then there are three extra cellular loops which are available for antibody binding, and these three loops must be properly oriented and strictly optimally aligned in order for the antibody to bind. And this can only be hardly mimicked in the animal model just by immunization.
We then tried also to immunize with membranes of aquaporin-4 transfected cells, and there we got a little of antibody titer, but when we used these antibodies to stain tissue in order to find out whether they are good one, we saw much more staining than we would have liked, and so that means that the membranes are probably contain some antigens which were then, after immunization, targets of antibody responses. So this was so far in our hands a failure. And as far as I know, we are not the only ones that suffer from that. So there is currently, unfortunately, no model which works after immunization with aquaporin-4.
MSDF
Where do you go from here?
Dr. Bradl
Well, we are currently modifying our animal models to the extent that we study much more the T cell responses, and we also try to modify the B cell site, but this is a bit of a, let's call it easy way modification. Because we learned along the way that when we have a very, very, very good NMO IgG from a patient, we can work with very low antibody titers, and so that gives us a very nice animal model. And we also know that there are some NMO IgGs which make high titers in the patients but which are relatively lousy in animal models. So we learned from this that we just select and search for the best animal IgG for the model to transfer this; that's the B cell side. And on the T cell side, you'll find T cells in NMO lesions, but people had a hard time to get aquaporin-4 specific T cells.
So it was not quite clear whether one needs aquaporin-4 specific T cells at all for the formation of lesions, or whether any other activated T cell that recognizes different proteins in the CNS could do the job as well. So over the last few months, we now were really able to produce really highly pathogenic aquaporin-4 specific T cells which do the job and which guide lesions to sites where they are also seen in NMO patients. And so with this we were now able to really advance our model much, much more than we had done before.
MSDF
So these T cells you've generated, and these are directly cytotoxic?
Dr. Bradl
We are not dealing with CD8-positive or cytotoxic T cells, we are dealing with helper T cells. And these helper T cells, we know that they exist because the pathogenic antibodies of the patients have a phenotype that needs T cell help in their formation. But it was all the time unclear whether the T cells only help in antibody formation, or whether they also help in localizing lesions to the correct places. And now we have really for the first time the impression that we have a cell line that does exactly this.
MSDF
How do you translate what you're finding out in the animal models to the clinical situation? Is it developed enough now that you can make correlates?
Dr. Bradl
Well, that's a good point. I mean, when you look, for example, at our T cell work, then we observed in our animals that there are a large number of epitopes available for antigen recognition by T cells in the rat. And then it turned out that people observed the same thing in mice, and now we know it's also the same thing in humans. And then when you have so many different epitopes or so many different parts of a protein that can be recognized by the immune system, then you have to figure out whether all of them could give rise to pathogenic T cells or not.
And in the Lewis rat, for example, one knows that on myelin basic protein, there are two adjacent peptides which can induce very nice T cell responses, but only one T cell response is pathogenic and the other harmless. And so we initially were facing the same problem with our Lewis rats and the many different epitopes on aquaporin-4, and there we found out that in principle we can also rise T cell responses against many of these epitopes, but we have to use an enormous amount of T cells to get lesions in the CNS.
But with our new T cell line, now we know that we only have to use very few cells to get the lesion, so they are the dominant pathogenic T cells. And it's quite nice that in NMO patients with a very peculiar MHC phenotype – that's an MHC phenotype that's mostly seen Brazilian NMO patients – they recognize dominantly an epitope that's very close to ours, and they termed this also immunodominant epitope. And it could be that it's pathogenic as well, but there is not yet any proof for that in humans.
MSDF
Looking at aquaporin-4 as a target in NMO, do these cells just use it as a target to destroy the cell that it's on, or does it result in a pathologic process by inhibiting the action of the channel?
Dr. Bradl
There are reports about knockout animals where there is no aquaporin-4 available, also on astrocytes in the CNS. And these animals are apparently healthy under normal conditions, but they show a disease phenotype under conditions where there is tissue swelling going on; for example, under ischemia, and so they cannot cope with that properly. So that means the complete absence of this channel is also bad. Then there are currently two different groups of thinking in the scientific community. There are reports that antibodies can bind to aquaporin-4 and inhibit water flow through this channel, but there are other groups that could not reproduce it. And at the moment it could just be a matter of different antibody preparations or different test systems or different species, so this issue is not 100% solved yet.
MSDF
Anything we've missed or interesting to add on the topic?
Dr. Bradl
I think the only thing one can say is that since NMO is such an extremely rare disease and since this makes it necessary that people all over the world cooperate with each other, that leads to an enormously research-friendly atmosphere and an enormous willingness of the people to cooperate with each other, and so on all different types of subjects.
MSDF
How many patients are there?
Dr. Bradl
Well, when you look here in Austria, we have about 8 million inhabitants; there are 8,000 MS patients and approximately 80 NMO patients. And this frequency is more or less encountered throughout the world; it's a very rare disease.
MSDF
Very good, thank you.
[transition music]
Thank you for listening to Episode Forty-four of Multiple Sclerosis Discovery. This podcast was produced by the MS Discovery Forum, MSDF, the premier source of independent news and information on MS research. MSDF’s executive editor is Robert Finn. Msdiscovery.org is part of the non-profit Accelerated Cure Project for Multiple Sclerosis. Robert McBurney is our President and CEO, and Hollie Schmidt is vice president of scientific operations.
Msdiscovery.org aims to focus attention on what is known and not yet known about the causes of MS and related conditions, their pathological mechanisms, and potential ways to intervene. By communicating this information in a way that builds bridges among different disciplines, we hope to open new routes toward significant clinical advances.
We’re interested in your opinions. Please join the discussion on one of our online forums or send comments, criticisms, and suggestions to editor@msdiscovery.org.
[outro music]