Scientists racing to solve the puzzles of neurodegenerative diseases and devise effective treatments are picking up the pace of research discovery. Still, obstacles remain in sorting out the complexities.
I had assumed that somehow we were further along in fighting Parkinson’s disease than in combating Alzheimer’s. After all, L-dopa, a treatment that can vanquish or greatly diminish the most famous symptoms of Parkinson’s – tremor and rigidity – has been available for decades, while researchers are still struggling to turn back Alzheimer’s to the same degree.
But L-dopa and similarly acting drugs are only a temporary fix, at best. These drugs treat symptoms, and only some of the symptoms. And they do not stop the cruel progression of the disease. In terms of identifying clues as to how a neurodegenerative disease arises, Alzheimer’s researchers actually began making progress earlier.
Robert Nussbaum, MD, is a physician who specializes in caring for adults with hereditary disorders, but he also has a passion for research. A decade ago, a colleague encouraged him to direct his genetics expertise to the study of Parkinson’s disease. Nussbaum became a leader of the first team to identify a genetic cause for an inherited form of Parkinson’s. That discovery has opened up the field of Parkinson’s research in the same way that similar discoveries earlier had propelled Alzheimer’s research forward.
Now Nussbaum, chief of Medical Genetics at UCSF and a member of the UCSF Institute for Human Genetics, is using what he has learned about genes, proteins and Parkinson’s to search for environmental factors that trigger the disease in individuals with a genetic susceptibility. He shares his thoughts on this week’s UCSF Science Café.
- Fast-Changing Field of Medical Genetics Embraces Personalized Medicine
- UCSF Today, Oct. 21, 2008
- Parkinson’s Disease Genetics Expert Wins Research Grant
- UCSF News Release, June 29, 2006
I'm here today on the UCSF Science Café with Robert Nussbaum, UCSF Holly Smith Distinguished Professor in Science and Medicine.
Bob, can you tell me how you came to be doing what you're doing today? That is to say, studying Parkinson's, but also studying genetics.
I began in genetics about 30 years ago, and I was motivated by the fact that you could combine the clinical care of families with genetic diseases with a basic science study of what these diseases were, and that one could use genetics to identify genes that were altered in these families, and in that way, both help the families by being able to do carrier detection; being able to counsel the families; being able to help them identify who in the family was at risk for these diseases; but also by finding the genes involved - they gave us a lot of insight into the pathways, the particular steps that led to the diseases. So genetics is a wonderful way of combining clinical medicine and the care of families with basic science research.
The Parkinson's disease interest really comes from an invitation that I received sometime in the mid 1990s from the head of the neurology institute, Dr. Zach Hall, formerly of UCSF, and then of the regenerative medicine program out here in California. So I was in the intramural program at NIH at the time, studying human genetics, and Zach called me up and said that they were having a very large national meeting in Washington to discuss the question of what's happening in Parkinson's disease.
And this was motivated to some extent by the neurology institute's interest in pushing the research agenda forward, but also because patient advocacy groups and other individuals with a personal interest in Parkinson's disease really felt like the field had stagnated - that since the discovery of L-dopa as a treatment for some of the motor symptoms of Parkinson's disease, there really hadn't been a lot of progress over the next couple of decades. And what Zach asked me to do was to apply the principles and methods of human genetics to Parkinson's disease.
Parkinson's disease had been known by the medical community. Sort of the wisdom of the medical community was that Parkinson's disease was not a genetic disease, and in fact, if you found families where there appeared to be inherited forms of Parkinson's disease, by definition, it wasn't "real Parkinson's disease."
There is a well-known and accepted approach to many diseases that relies upon identifying rare families in which the disorder is being inherited. Because it's being inherited, because it's in the family, the tools of the Human Genome Project allow us to quickly identify the genes involved. Once we know what the genes are, we can then look at the pathways of the proteins that those genes encode, and ask a simple question. Are the same genes and the same pathways involved in the apparently nongenetic forms as the genetic forms?
And, of course, probably the most spectacular success of that approach was to try to understand hypercholesterolemia, which affects 5 percent of the US population, by studying the rarer families - the one in 500 to one in 1,000 families - where there was a defect in the receptor for cholesterol, in particular for LDL. And in these families with familial hypercholesterolemia, you could identify the gene - it turned out to be the LDL receptor - and that led to an investigation of the entire cholesterol pathway and ultimately led to the identification of statin drugs, which are now useful, not just in the rare familial forms of hypercholesterolemia, but in all forms.
So my hope was, if we could identify some of the genes that are responsible for the rare forms of familial Parkinson's disease, it would illuminate the pathways and give us tools to try to understand sporadic Parkinson's disease. So that was the proposal I made at that meeting, and during the coffee break after I gave my talk, a gentleman came up to talk to me and he turned out to be a neurologist from New Jersey, whose name was Roger Duvoisin. And Roger had been studying a large family with inherited Parkinson's disease which looked for all the world like Parkinson's in terms of its pathology and its clinical appearance, etc., with probably two minor differences.
One was there was a relatively earlier onset. Instead of occurring in the 60s and 70's, the Parkinson's disease was occurring in the 40s and 50s in this family. And second of all, the family probably had a little bit more of the rigidity and the dystonia that you see in Parkinson's, and a little less of the tremor.
And we established a collaboration. What I basically told him was, "If you can provide us with a sufficient number of DNA samples from this family, we can find the gene responsible," and we did that in collaboration with my colleague at NIH at the time, Dr. Mihael Polymeropoulos. We first localized the gene to chromosome 4, and then identified the gene, and it turned out to be the gene for alpha-synuclein. So that was the first familial form of Parkinson's disease for which a gene was known.
In sporadic Parkinson's disease, patients' brains contain aggregates of protein. They're called Lewy bodies. Well, once alpha-synuclein was identified as being a gene in which mutations can cause familial Parkinson's, antibodies against alpha-synuclein were used, and it turned out that the brains of patients with sporadic Parkinson's disease had widespread abnormalities in alpha-synuclein protein deposition. The Lewy bodies turned out to be essentially alpha-synuclein bodies - aggregates of alpha-synuclein.
So that's how, from my point of view, human genetics and Parkinson's disease were married, and that's what I've been working on for the last 10 years.
Before you started studying Parkinson's, there was great dissatisfaction with that status. Is that right? Is that because that didn't really represent all of what was going on in Parkinson's, or because the treatment was ineffective, or because of a little of both?
The discovery of the loss of subtantia nigra neurons, and those are the neurons that are sitting in a particular region of the midbrain and send their extensions, their axons, up to the striatum and deliver dopamine there. There's no question that the death of neurons is responsible for an important subset of the clinical syndrome called Parkinson's disease.
Are those all the motor defects?
It's primarily the motor defects. It does not affect some of the other very serious problems in Parkinson's, including dementia, loss of intellectual function, depression and sympathetic abnormalities such as abnormalities in heart rate, in the ability to maintain blood pressure, and also an almost universal finding in Parkinson's patients, and that is severe constipation. Those do not respond at all to L-dopa, and the reason for that is because I think our view of Parkinson's disease has changed over the last four decades from being a disease of the dopaminergic neurons of the substantia nigra to a systemic disease. Not just a generalized brain disease, it's a systemic disease that affects both the brain and the peripheral nervous system.
At the time you got into the field then, there was an awareness that we needed to know more about these other effects, and there still, I imagine, was no understanding of the cause or a way to halt the progression of the disease. Is that -?
I mean, I think that's a really important point because the treatment with L-dopa or other sorts of interventions that increase the availability of dopamine are really symptomatic relief. They don't treat the underlying problem. They don't stop the neurodegeneration. They don't reverse any of the neuronal death. They don't rescue neurons. It simply replaces what's not being delivered.
And as I said, you already have widespread damage before you even get the obvious symptoms. Many patients will go for years with, for example, some decrease in their sense of smell and constipation, and perhaps even depression, as the only indicators of the fact that they are in the process of developing Parkinson's disease. And only later, when the motor symptoms become obvious, is the diagnosis made.
And is that still the state of the art in terms of diagnosis?
Yes, it is.
So you identified a protein that appears causative, one would conclude, from the study of the family in which you identified it - alpha-synuclein. Is that it?
And you also found that it was deposited in patients with sporadic forms of the disease, so that I presume offered an opportunity to find out, at least through animal models, if not in humans, the role of that protein. What have you been able to do since you did identify the protein as being present in these aggregates that form in the people with the disease?
So first of all, in addition to finding that there were mutations in alpha-synuclein, I also was involved in a large study of a family with early-onset Parkinson's disease and very severe dementia. This is an American family in Iowa that had been identified originally and studied by John Hardy and Andy Singleton and Matt Farrer, and what we found was that in this family, the alpha-synuclein itself was not abnormal.
The gene had been triplicated, so there were four copies of the gene instead of two in an individual, and so people were simply making too much. The protein's normal; there's just too much of it. And this clearly demonstrated that alpha-synuclein needed to be tightly regulated in order to prevent Parkinson's disease from occurring.
Well, what happens if there are minor genetic variations in and around the alpha-synuclein gene which either increase or decrease the expression of alpha-synuclein by 10 percent or 20 percent? Would 10 or 20 percent be enough to increase people's susceptibility to developing Parkinson's disease? I think of alpha-synuclein variation now as being a potential target for studies on susceptibility. And in my lab, what we did was to ask, "Okay, what variant that's associated with an increase risk of Parkinson's disease, what does that variant do to the expression of alpha-synuclein?"
And what was shown is that it increases the expression, and so we're starting to complete the circle here. Mutation or overexpression of alpha-synuclein can essentially cause the disease in families. Two, is that there are variants that increase the susceptibility to Parkinson's - not in the setting of a family, but in sporadic Parkinson's disease. And three, those variants are capable of increasing the expression - not by 50 percent, but more like 10 or 20 percent, so -
Can you talk a little bit about, historically, the role of mouse models and understanding this disease? What was wrong with the other mouse models, and what are your hopes for the model you've created?
This is a fundamental issue that comes from all animal model research, is that there really is no animal model that we use, probably with the exception of nonhuman primates, that have anywhere near the life span that humans have. And so what people frequently do is try to push the lesion to try to speed up the process because mice only live two years. But by pushing it, that doesn't necessarily mean that the process is exactly the same.
The animal models that we've developed here recently in my lab have been based on a different philosophy - a different idea - and that is that we should try to mimic as closely as possible the normal expression of alpha-synuclein, but use the mutant form - the same mutant forms that cause the disease in people - and use those animal models not to model the Parkinson's disease process per se, but to develop animal models that would be in essence sensitized models on which one can now test the environmental factors that almost certainly contribute to the development of Parkinson's one way or the other.
Although our animals do not show any neurodegeneration in substantia nigra, they are showing significant constipation and abnormalities in the enteric nervous system. So I think we're starting to mimic the early signs of Parkinson's disease, and now we have to find out what environmental trigger will push those animals over the edge and reproduce the neurodegeneration that occurs in people.
And at those early stages, are you not seeing the formation of clumps of alpha-synuclein?
That's right, we're not seeing aggregates, and so I think that what this is telling us is that in the early stages of this disease, there is dysfunction of the neurons without aggregation of the protein and without cell death, which in many ways is good news, 'cause what that means then is that it's not irreversible.
You are planning to use the same sorts of toxins that have been used in earlier mouse models and at lower doses?
No. Actually I'm not a great believer in the hypothesis that chemicals in the environment are what's responsible. I think it's a much more widespread environmental exposure, and I am very much informed by the careful studies that have been done by a German neuropathologist by the name of Heiko Braak who's in Frankfurt, Germany, whom I've gotten to know through a formal postdoctoral fellow of mine.
And Braak has made a very interesting - I think critical - observation, and that is he believes Parkinson's disease first begins in the brain stem, not in the midbrain where the substantia nigra is, and that it begins in the region of the brain where the nucleus of the vagus nerve - motor nerve number 10, which is a parasympathetic nerve that innervates the heart and innervates the enteric nervous system in the gut - and that's one of the areas where you see the earliest signs of change.
What Braak has hypothesized is that Parkinson's disease may actually develop from the exposure of the nervous system to some kind of a toxin or infectious agent that's entering through the gut. He mentions this because, in addition to the vagal nerve abnormalities, it's now quite clear that in very early stages of Parkinson's disease, you already start to see alpha-synuclein aggregates in the nerves that line the gut - the so-called Auerbach plexus or myenteric plexus. This is a network of small neurons that innervate the entire gut and are involved with the nerve impulses that allow peristalsis and normal movement of food and fecal material through the enteric nervous system.
Is that known from humans as well as animals?
Yes. It's been documented in humans.
And so what Braak is suggesting is that perhaps there's a viral agent which finds its way up to the brain through the vagus nerve number 10. And we know there are viruses that are capable of retrograde transport along axons, that that same viral agent infects other neurons in the lining of the gut and cause the alpha-synuclein to aggregate. That's one possibility.
Another possibility is perhaps it's chronic inflammation and that there's a level of inflammation that is a result of the continual exposure to bacteria in the intestine, and that this low level of inflammation is capable of triggering in susceptible individuals a neurodegenerative response. And there's now good evidence in animal models from a group in Dallas, Texas, as well as a group in Philadelphia, that chronic inflammation delivered in the peritoneal cavity, or directly in the brain, can cause neurodegeneration.
Do you have a way of inducing this immune response, then, in the mice that you - ?
Yes, and that's the plan. The plan here with my mice is to use them as a sensitized model and to produce intestinal inflammation. Our first step is going to be intraperitoneal injection of very low doses of lipopolysaccharide cell wall constituents from gram-negative bacteria.
For the UCSF Department of Public Affairs, I'm Jeff Norris. Production support for this podcast has been provided by Anthony Taliaferro. Related links and an archive of earlier conversations are available at www.ucsf.edu/sciencecafe. UCSF is a leading university dedicated to promoting health worldwide through advanced biomedical research, graduate-level education in the life sciences and health professions, and excellence in patient care.