ALS - 运动神经元疾病：机制和新疗法的发展

My name's Jeff Rothstein. I'm a professor of neurology and neuroscience at Johns Hopkins University. So I'm both a neurologist and a neuroscientist.

And on the neurology side of my life I study and work clinically on a LS, which is a neuromuscular disease, atrophic lateral sclerosis. On the laboratory side, I try to understand why a LS occurs in terms of the pathophysiology. Why do motor neurons die?

Why do cells not work properly, including cells like astrocytes? And then at the very fundamental side, I work on synaptic biology of astrocytes And glutamate transporters. So motor neuron disease is a very general Term for a class of diseases that affect both upper or lower motor neurons.

In the United States, we often refer to this as a LS atrophic lateral sclerosis, but a LS is only one of a number of different motor neuron diseases. So as an example, if the disease only affects upper motor neurons, the cells in the cortex, the brain, then it's referred to as primary lateral sclerosis. And when the disease only affects lower motor neurons, those cells only in the spinal cord, it's considered spinal muscular atrophy.

These are all much more rare subtypes where the most common disease is A LS, which is a, again, a Disease of upper and lower motor neurons. At the pathological level, it's the Diseases I, I think I mentioned, the upper motor neuron pool. Those in the motor strip, the motor cortex, those motor neurons in the brainstem, what we refer to clinically as bulbar motor neurons and those motor neurons that lie within the ventral spinal cord.

The lower motor neurons in this disease, there's a progressive degeneration, classically thought of, of only motor neurons. That is only the large alpha motor neurons in the spinal cord and the cortical motor neurons. The bet cells unfortunately, is a bit of an overstatement, although the clinical phenotype of the disease comes from the loss of motor neurons.

In reality, we now know today that there's a disease of other cells. Inter neurons degenerate a LS.There's dysfunction of astrocytes both in cortex and, and spinal cord of a LS.So all, again, the clinical phenotype is a motor defect. The disease is really a disease of that's someone goes beyond the pure motor neuron pool.

The consequence of motor neuron degeneration is that once a muscle no longer receives, its innervation again referred to as denervation muscles begin to atrophy and develop wasting and weakness ensues and muscle twitches referred to as fasciculations. So the signs of lower motor neuron disease are really muscle wasting weakness in fasciculations. That's lower motor neuron.

And the signs of upper motor neuron is are changes in the circuits that lead to spasticity, muscle stiffness and hyperreflexia brisk reflexes, and then what's referred to As pathologic reflexes as well. There is no cure for A-L-S-A-L-S Is a member of the class of neurodegenerative diseases, which is an adult onset disease progresses typically, it begins typically in midlife and progresses slowly over about two to four years. It is a uniformly fatal disease.

There are essentially no exceptions, and the disease is fatal because, not because patients can't move their limbs, but because it affects respiratory motor neurons. And so the chest wall motor neurons and the diaphragm motor neurons degenerate and patients suffocate to death. A cure would be to really stop the disease, of course, and reverse the degeneration of motor neurons and in fact, to replace the motor neurons.

And that just does not exist today. There is one effective therapy for a LS and that is a drug called aole Aurel tech. It blocks a voltage activated sodium channels.

And for most of what we know about it, it blocks glutamate toxicity. At least three different international trials have shown that OLE is effective, it's slowing the disease. Patients essentially live longer while on liuzzo.

Again, the most recent study suggests when used early in the disease, patients can live up to a year longer. And again, most recent studies would suggest that if used early, there is a delay in symptoms. So it patients, the disease Just progresses out more slowly.

There is no one finding an a LS field That is more significant than another. Okay, the big advances first came in the early nineties when we discovered genes for the rare familial form of the disease. So a LS is largely sporadic.

It happens outta the blue. 5%of the cases inherited usually in autosomal dominant fashion in one fifth of those familial cases. We now know that mutations of superoxide dismutase and antioxidant enzyme present in all cells of our body causes a LS.So very tiny subset of a LS is caused by SOD one mutations.

About one to 2%of all cases. Those mutations are essentially not present in this sporadic population since those original genetic very important genetic discoveries, we now know of other genes that contribute to the familial form of a LS, but those genes are even more rare than the SOV one form. So the bulk of a LS essentially almost 98%of it.

We really don't know what causes the disease. Newer and very important advances have come from doing risk gene analysis, which have just been, are just being done as, as we film now by multiple labs around the world with the most the the first published studies coming out of NIH. And it's our first attempt to find out why are sporadic patients at risk for developing a disease.

So familial gene was an important observation because it allowed us to build tools. A mouse model is an example. In fact, now multiple mouse models that contain the gene that overexpress the gene, develop a disease that looks a lot like the human disease and have been used both to understand how that mu mutation might cause the disease.

And in parallel to find drugs that might be useful in the disease. Unfortunately, the mouse, as good as it is for helping us discover pathways, has not been very predictive for discovering new drugs. Although we continue to use it, we all continue to believe it could be useful for drug discovery.

To date is largely failed At helping us in that regard. Well, so the other part of a LS is what goes Wrong once some toxin begins the insult as an example, we know that, so D one mutations start familial A LSA subset of patients, but we don't know what propagates the disease. Think of it as a DO series of dominoes.

We know the first domino of fall is SOD one mutations, but all the ensuing steps that follow which really propagate the disease as dominoes move along, we have only begun to pick at those various steps because if you intervene in those steps, you might change the course of the disease, essentially stopping another domino from falling. So my lab played a role because we discovered the glutamate, the role of glutamate as a toxin in a LS, actually more specifically that astrocytes don't work properly. In the early nineties, we learned that there was too much glutamate in a LS patients, and that's actually the mechanism in part of ole the protective mechanism.

And we went on to discover that that defect was the cells around motor neurons called astrocytes. They're support cells. So about by 1992 to 95, we had learned that the support cells were major contributors.

That was then followed by elegant work by other laboratories confirming an observation and extending it. That formed a basis essentially for drug discovery, which my lab is heavily engaged in, and to understand why astrocytes don't work properly in the disease. Nevertheless, a LS is really a multifactorial process by, the reason that's actually quite important is by the time you act, you see a patient in your clinic.

That first domino, that first step that happened a long time ago, you'll learn you're in the stage where all the other subsequent events are occurring that contribute to the disease. So what used to be considered ah, epiphenomenon secondary events, we now better realize that those are the events that we have to intervene in as clinicians. So if I take my science hat off, I put my clinician hat on.

I know at those secondary events of the steps that I need to change to try to stop the, Essentially stem, the progression of the disease. There are many technical challenges in a LS one, unlike diseases. Let's use an example of Parkinson's where it's a small group of cells in one part of the brain.

A LS involve cells throughout your entire neuro axis, the entire spinal cord, a good port part of the, the, the rerum. So it's difficult to unify a single defect into all those reasons, yet we know that mutations certainly do that. If we jump ahead to think about replacing cells in a LS, which someday will we certainly seriously consider, it also is a technical challenge.

It's not like Parkinson's where we might just put a few dopaminergic neurons back into the central nervous system. We have to replace cells throughout the entire spinal cord and motor strip, which is technically quite challenging. And on top of that, if we think, again, at the level of even at stem cells replacing the largest cell in your body, motor neurons are two to four feet long.

They're huge cells. How do you replace those cells? Profit and reestablish connections.

That's the problems, if you will, on stem cells that we face at least motor neuron stem cells. And the other problem is that, as I mentioned earlier, if a LS is in fact multifactorial, let's just make up those five different processes that go wrong. By the time you see a patient, how do you sort out how to treat each of those processes?

Do You need to treat them all at once? So the technical challenge in a LS In part are prioritization. Is there any one pathway that's more important than the other?

Today we have no way of sorting that out. Are there is a glutamate pathway, a mitochondrial pathway, and axon transport pathway? These are just some of the things that are being bantered about in 2007 as targets for both defects and therapy.

Is one pathway more important in terms of drug development? Probably not, at least as we understand it today. The technical challenges I mentioned is this is a D diverse diffuse disease throughout multiple parts of the nervous system.

And it's a spreading disease. It starts very focally and spreads to other parts of the body. In terms of the neuro axis, how do we understand that spreading process and does it impede how we develop research tools?

Are the familial models that we've so relied on have been so useful in fact going to be useful for the common sporadic disease? We don't know that answer yet, and we just assume that it's the best starting point to work with, but maybe it won't be someday. It it.

However, it remains our main focus for developing understandings of the disease and tools. Another of course, a major impediment to all medical research today is of course funding. How do we make sure we bring the funds to bear onto a terrible disease like a LS?

And it's always not always funding that's important. It's, it's bringing really bright young and senior investigators the experience of multiple different scientists to engage, work together to help Tackle the disease. There are multiple reasons why we do not Translate our laboratory work to the clinic efficiently.

Some of those are inherent characteristics of how many scientists work. We work in our individual pos that are individual labs. We wait until someone else publishes something.

We don't tend to collaborate a lot as a community. That's not unique to a LS.But that does impair rapid transition. It doesn't stop research from occurring on.

It just impairs rapid transition. The translation often has been handled in the past and even now by pharma, big, small and small pharma. A LS is not a major target for drug development.

There's not a big market for ls. It's an orphan disease. And so they don't pay a lot of attention and they don't put their own resources to bear to help us.

We often don't realize that mu much of drug development really occurs with the research within drug companies. We often provide the initial observations about a defect in the disease. They take those defects and find drugs that work in those defects.

We're not facile at doing that. We don't typically do high throughput drug screening. We don't typically develop assays that are used by drug companies.

And so the translation of what we learn at the basic science level pathway defects is not always well handled by us as academics for drug discovery and drug companies, since they don't come to help us and don't work in that area. There's this enormous gap between fundamental discovery and drug discovery. So one question is, can we facilitate that process?

Can we do it better? And some groups have begun to do that. We started a group called the Packard Center now about seven years ago, which was designed to facilitate, designed to fill the gaps that any academic really understands about disease research, bring people together, mandate collaboration, and find ways to bring bio biopharmaceutical companies into the mix so that as we discover things, we can hand it off to biotech companies or pharma and they can help us work with drug discovery And Packard's really about that.

So there's several approaches I think that will, That will help. A LS one of course is stem cells. Although stem cells has an enormous amount of hype, the base realities, once we learn how to use the cells, identify which of the right cells for a LS as an example, motor neurons might be an obvious example for a LS patients.

Astrocytes, we know a lot about astrocyte dysfunction in a ls. So minimum those two cells look to be good targets. But it's, and that's important.

That's no different than saying, Hmm, you have an infection. What's the right antibiotic? Can penicillin vancomycin, which is the right stem cell ils, which is your right therapy.

The issues then are, and and not thought of by many are the practical issues of drug development. How many cells do you use? Do you have to reinject cells?

How do you engraft them into patients? How do you protect the cells from staying alive? How do you verify that each batch of cells are the same from one week to the next when we get to the point of therapy?

So those are practical issues that actually are being dealt with by a number of labs. Now, how do we use astrocytes? Do we modify them to enhance them just like we modify antibiotics to make them more potent?

How do we deliver the cells properly? Do we have devices that can inject cells into the spinal cord? There are labs working on that.

Now, how do we follow the cells? Can we track them if they migrate? How do we enhance motor neurons to reconnect the muscle?

Motor neurons grow very slow. And one would estimate that if I put a mo new motor neuron into an a LS patient today, it could take about three years to reconnect to their feet. Can we do that faster?

So those kinds of approaches are being considered in in, in a LS therapeutics. Of course, when one brings up stem cells, there's another side of stem cells, which is very practical. Can we just use stem cells as a tool for drug discovery?

If human astrocytes become defective in a LS, can we use them to discover drugs? Can we actually make a LS astrocytes to better understand if the drug would work in that astrocyte or not? So there are approaches that are being done right here now to address Those issues.