Well, I’ve always been interested in the mechanism of brain function, in how the brain works. I got interested in Alzheimer’s disease because, first of all, it’s an important disease for our patients and, second, at that time when I started working on it, there were a number of paths that were finally helping us to understand the actual mechanism of the disease. The disease was originally described in 1907, and for decades after that all we knew were the clinical manifestations of the disease and some of the microscopic abnormalities that were present in patients. It really wasn’t until the mid-1980s when a number of scientists started unraveling some of the molecular and cellular basis of this disease. I started in the late 1980s.

  What was your plan of attack for making sense of the mechanism?

Basically, what we tried to do was extend some of the observations made by other people in studying the composition of the senile plaque, which is one of the microscopic abnormalities of Alzheimer’s disease. Specifically, we tried to understand all the other molecules involved in making the plaque in addition to beta peptide, which is the primary constituent.

Let me start with the paper that preceded it ("Apolipoprotein-E—high-avidity binding to beta-amyloid and increased frequency of type-4 allele in late-onset familial Alzheimer-disease," PNAS 90[5]: 1977-81, 1 March 1993.). That’s second on the list of my highly cited papers and led us into that Science paper. It’s easier to understand this chronologically. What we set out to do was characterize the different proteins in the brain that interact with the beta peptide. That’s the basis of observations in that first PNAS paper. What we demonstrated was that a protein called apolipoprotein-E, which had been previously described in serum, was one of the major proteins binding to beta peptide. The first paper demonstrated that apolipoprotein-E was a very avid binder to beta peptide in a test-tube assay. We simply tried to identify different proteins in the spinal fluid that bound to the beta peptide.

  Was apolipoprotein-E a surprise?

It was. Apolipoprotein-E had been studied for many, many years for its role in cholesterol transport in the blood. Until we made this first observation, there was no strong evidence that apolipoprotein-E was involved in Alzheimer’s disease.

  Was there any evidence suggesting it might be when you made the discovery?

Two pieces of evidence: one is that other groups had shown that when peripheral nerve is destroyed the body produces increased amount of apolipoprotein-E in response. So there had been some speculation that apolipoprotein-E was involved in responses to peripheral nerve lesion. After we made our initial observation, we went back to the literature and found that another group had done immunohistochemistry in the brain showing that apolipoprotein-E was found in senile plaque.

  What is the relationship between apolipoprotein-E and the kind of cholesterol we're always hearing about with heart disease?

Basically, there’s a family of different lipoprotein particles that contain a lipid shell and cholesterol within. These contain different types of apolipoproteins. Apolipoprotein-E is one type. It binds to certain of these particles better than others. It depends on the type of apolipoprotein-E. There are primarily three different types of apolipoprotein-E. All are made by the same gene, but they have very subtle—one amino acid—differences in sequence.

  Is the amount of apolipoprotein-E related to cholesterol levels?

The type of apolipoprotein-E certainly is. Individuals who have genes making apolipoprotein-E type 2 have very high serum cholesterol. And there’s a subset with apolipoprotein-E type 4 who also have very high serum cholesterol.

  Can you affect the level of apolipoprotein-E with drugs or diet?

You cannot. Apolipoprotein-E is a protein that is difficult to regulate with drugs or with diet.

  What was it about the discovery of apolipoprotein-E and its binding affinity to beta peptide that gave your PNAS paper such impact?

Well, first of all it showed for the first time that apolipoprotein-E is bound to beta peptide. That was one observation. The other was that the senile plaque, also known as the neuritic plaque, in the brains of patients with Alzheimer’s disease also contains apolipoprotein-E. The third observation in the paper is that there are different alleles of apolipoprotein-E that are associated with different risks of Alzheimer’s disease. And we showed in that paper that patients with Alzheimer’s had a much higher probability of having apolipoprotein-E type 4 allele than control patients. So it demonstrated by several lines of evidence that apolipoprotein-E was important in Alzheimer’s disease.

  Now what about the Science paper that followed it?

The Science paper was a quantitative assessment of the impact of the different apolipoprotein-E gene alleles on the probability of developing Alzheimer’s disease. So in the first paper, the PNAS paper, we simply demonstrated that if you took a population of Alzheimer’s patients and compared it to a population of controls, then the Alzheimer’s patients had a higher frequency of apolipoprotein-E type 4 alleles. In the Science paper, we looked at a large population of individuals and asked what the probability was of developing Alzheimer’s as a function of the apolipoprotein-E genotype as well as the age of the patient.

  And what did you find?

We found that patients with apolipoprotein-E type 4 had a much higher risk not only of developing Alzheimer’s but also of developing Alzheimer’s disease at an earlier age. Individuals with one copy of the apolipoprotein-E type 4 allele have about a five-fold increased risk of developing Alzheimer’s disease. Those with two copies of the apolipoprotein-E type 4 allele have a 20-fold increase in risk of developing Alzheimer’s disease.

  Why did you publish the first paper in PNAS and the second paper in Science?

The first paper was rejected by a number of journals before it got accepted in PNAS.

  Why was it rejected?

I think there was, initially, a tremendous amount of skepticism about the basic conclusions since it was a novel observation.

  How has the research progressed since 1993?

The field of apolipoprotein-E and Alzheimer’s disease is basically focusing on the mechanism by which apolipoprotein-E is involved in the disease. The basic observation that apolipoprotein-E genotype impacts on Alzheimer’s disease has now been widely replicated. The major question right now is the mechanism by which apolipoprotein-E is involved in the process of Alzheimer’s disease or by which the brain is trying to respond to Alzheimer’s disease. This has taken a remarkably long time because it’s a tough problem.

  What makes it so tough?

Several things. One is that apolipoprotein-E is a difficult protein to work with. Two is that the function of apolipoprotein-E also depends on its binding to lipid particles. Three is that there is a whole family of different receptors for apolipoprotein-E in the brain. And four, and probably the most important, is it is difficult to know what assay needs to be done because we don’t really have a good animal model that fully replicates Alzheimer’s disease.

The fact is these observations were made in 1992 and published in 1993 and we—"we" being the whole field—still do not know how apolipoprotein-E is influencing Alzheimer’s disease, although there are a lot of different laboratories working on it and a number of different hypotheses.

  What do you consider the leading hypotheses?

I’ll toss out some ideas; there are experiments supporting each one of these. There may be a direct role of apolipoprotein-E in the processing of amyloid in the formation of the senile plaque; in the formation of the neuro-fibrillary tangle, which is one of the other lesions of Alzheimer’s; in recovery mechanisms, not only in Alzheimer’s but also in other acute neurological insults; and there is a wide spectrum of hypotheses that need to be tested.

  Are you optimistic that the answer will come soon?

Someone will find it, but I don’t know when.

  What are you working on now?

Well, it hasn’t been published yet, so why don’t we keep it really generic and just say we’re studying the role of lipoprotein particles in maintenance of cellular function.

  Was there an element of serendipity to your work on apolipoprotein-E?

The initial discovery that apolipoprotein-E binds to beta protein was itself serendipitous. We had no preconceived notions that that would be the case.

  Is there a take-away message that you would like to leave about your research?

One of the messages I would like to discuss, which we haven’t, is the fact we were able to make this observation and pursue it rapidly and aggressively because we were working in a clinical and laboratory environment. At Duke, we have large numbers of patients with Alzheimer’s disease and a lot of resources, and individuals who work collaboratively in advancing this whole hypothesis. That combination of clinic and laboratory is crucial to this kind of research.