BU Today

Science & Tech

The Gene Hunter

Searching for the disease’s genetic roots

Alzheimer’s disease is an epidemic. It attacks the brain’s nerve cells, causing memory loss, behavioral changes, confusion, and deterioration of language skills. It affects more than 5 million Americans 65 and older, and that is expected to increase to 13.8 million by 2050 unless science finds a treatment. Alzheimer’s and other forms of dementia are projected to cost the nation $236 billion this year and the figure could reach $1 trillion by 2050, according to the Alzheimer’s Association.

At Boston University, dozens of researchers are looking for tests that could lead to early diagnosis and interventions to prevent or delay the disease, running clinical trials that may result in treatments and an eventual cure, working to understand genetic risk factors, and studying Alzheimer’s impact on caregivers.

The Boston University Alzheimer’s Disease Center, established in 1996, is one of 31 such centers nationwide funded by the National Institutes of Health and dedicated to conducting research into the disease, enhancing clinical care, and providing education.

In this special report, BU Today examines the work of five BU researchers.


In 1984, Lindsay Farrer was a young medical geneticist finishing his PhD at the Indiana University School of Medicine, studying families afflicted with Huntington’s disease, a devastating inherited disorder that causes nerve cells in the brain to waste away, leading to uncontrolled movements, dementia, and eventually, death.

One day, he walked into his advisor’s office and saw a huge family genealogy, or pedigree, spread across the desk. “Is it a new Huntington’s pedigree?” he asked. “No,” replied his advisor. “It’s Alzheimer’s disease.”

Farrer had never heard of Alzheimer’s, and his advisor had scarce information to share: “I only know a little bit about it,” he said, “but we’re going to find out.”

Three decades later, Alzheimer’s has moved from the ranks of obscure diseases to the forefront of the national conversation. And Farrer, BU Distinguished Professor of Genetics, chief of biomedical genetics, and a BU Alzheimer’s Disease Center faculty member, is a leader in studying the genetics of the disease. Today, he says, genetic research offers the best hope for understanding and treating Alzheimer’s.

“Already we’re seeing, from the genetics alone, multiple biological pathways that are somehow involved with the disease,” says Farrer. “I expect that reliable biomarkers for predicting Alzheimer’s and effective treatment are going to emerge from some of these gene targets. They are going to yield fruit.”

All the blueprints for building our bodies, from belly button to brain stem, lie in our genes. A gene is a stretch of DNA that carries the instructions for making a single protein. For example, a gene called APP spells out the code for amyloid beta precursor protein (also called APP), a key player in Alzheimer’s disease. Many genes have variants—sections of code that vary from person to person, giving one woman blue eyes, for instance, and another brown.

Already we’re seeing, from the genetics alone, multiple biological pathways that are somehow involved with the disease,” says Farrer.

In the early days of medical genetics, back when Farrer was starting out, scientists targeted only the simplest genetic diseases, like Huntington’s, caused by mutations in a single gene. They tackled these diseases by collecting detailed patient information, drawing extended family pedigrees, profiling family members for genetic variants, and statistically analyzing the data for clues as to what variants are linked to, or track with, the unknown disease gene. When scientists had a sense of where a gene might be, they painstakingly “walked” the chromosome, starting at the linked genetic variant and tagging a tiny, known sequence with fluorescent dye, then decoding, bit by tiny bit, ever-longer stretches of DNA until they found something out of the ordinary.

This method, called linkage analysis, although slow and labor-intensive, led scientists in the early 1990s to discover the genes for Huntington’s disease, cystic fibrosis, and the BRCA1 form of breast cancer. Then Farrer and other scientists aimed for something tougher: the genes associated with early-onset Alzheimer’s, a rare form of the disease passed from parent to child, which accounts for only about 0.1 percent of cases. People with this form of the disease usually develop symptoms in their 40s and 50s. In 1991, scientists in London were the first to find genetic variants associated with early-onset Alzheimer’s—mutations in the APP gene that cause the body to dramatically increase production of beta-amyloid, the sticky protein fragments that form plaques in the brains of Alzheimer’s patients. Soon afterward, Farrer and colleague Peter St. George-Hyslop at the Cambridge Institute for Medical Research in England, discovered two genes, presenilin 1 and presenilin 2, which code for proteins involved in processing APP. Almost all cases of familial early-onset Alzheimer’s are caused by variants of these two genes and a third, APP.

The discovery of these three genes emboldened researchers, including Farrer, to aim for something even tougher: late-onset Alzheimer’s, the more common form of the disease, typically occurring in people after age 65. A team of Duke researchers led by Allen Roses scored an early victory in 1993, using linkage analysis to find that variants in a gene called APOE substantially increase or decrease a person’s risk of getting late-onset Alzheimer’s.

“That was such a revolutionary finding,” says Farrer. “At the time, the known primary role of apolipoprotein E—the protein encoded by APOE—was to transport cholesterol. Who would have imagined that Alzheimer’s disease has anything to do with cholesterol? Moreover, this discovery goes beyond Alzheimer’s disease. Geneticists thought, well, gee, we can do this for diabetes, we can do this for cancer, we can do this for all kinds of things. It turns out, however, that was not the case.”

  • Early-onset, or “familial,” Alzheimer’s is rare, making up only about 0.1 percent of cases. It strikes between the ages of 30 and 60 and is caused by three known genes: presenilin 1, presenilin 2, and APP.

  • Late-onset Alzheimer’s is far more common, usually affecting people over age 65. No single gene causes the disease, but more than 20 genetic variants have been shown to increase or decrease susceptibility.

Late-onset Alzheimer’s proved a nuanced nemesis. The genes associated with the disease don’t follow a familiar, predictable pattern from parent to child. Rather, scientists see scattered relatives with the disease—maybe a handful of cousins or a smattering of siblings—or even no affected relatives. This unpredictable pattern of inheritance means that no single gene causes the disease, as in Huntington’s, but rather, many genes are involved, each conferring some degree of risk.

For a decade, the field stumbled through what Farrer calls the dark era of Alzheimer’s genetics. “Between the years 1997 and 2007, there were more than 1,000 papers reporting associations in about 400 genes, but none of them except APOE turned out to be correct,” he says, noting that many scientists at the time based their discoveries on statistical analyses in a relatively small samples of patients. “I was known by some people in Alzheimer’s research as Dr. No,” he says. “People would publish and then we would report that we don’t see any evidence, generally in a much larger sample. So the field got very frustrated.”

The Human Genome Project spawned the technology and infrastructure that finally pushed the field forward, allowing scientists to hunt for genes through genome-wide association studies, or GWAS. The new technology used for GWAS permits scientists to examine the genomes of thousands—even tens of thousands—of Alzheimer’s patients and age-matched people who do not have the disease and examine millions of variants, arrayed all across the genome, at a reasonable cost and in a relatively short time. GWAS is science on a massive scale, and Farrer has emerged as a leader.

“For a long time, everyone wanted to be the one to say, ‘I discovered this gene.’ But it’s not possible to do it alone anymore,” says longtime Farrer collaborator Carmela Abraham, a MED professor of biochemistry and pharmacology and experimental therapeutics. “These variations individually have a small influence on disease risk, and we need a lot of research subjects to be able to find them. Bringing scientists together and forgetting about your own ego—that’s the most important thing.”

Farrer is one of nine principal investigators on the ongoing National Institute on Aging (NIA) Alzheimer’s Disease Sequencing Project, and one of five leaders of the Alzheimer’s Disease Genetics Consortium, which is also funded by the NIA. By pooling and analyzing data from many smaller studies, these groups have found genetic variants associated with late-onset Alzheimer’s at more than 20 distinct locations of the genome, in genes that affect a host of things, from inflammation to lipid metabolism. Each discovery offers a clue to the still-incomplete big picture.

Ultimately, treatment is not going to be one size fits all,” says Farrer. “There are people who have a set of risk factors—some genetic, some other—that may contribute to disease.”

“With Alzheimer’s we don’t have a full understanding of the underlying process. The fundamental first step, the underlying pathology—that’s not known,” says Richard Sherva, a MED research assistant professor of biomedical genetics, who works with Farrer. “These association studies are a good way to get at that.”

Farrer also looks for Alzheimer’s genes with a more traditional approach. For example, he and his colleagues thought that a family of proteins, including SORL1, which transport other proteins around the cell, might play a role in Alzheimer’s. Over five years, he and collaborators from the University of Toronto and Columbia University studied DNA samples from more than 6,000 people in four distinct ethnic and racial groups: Caucasians, African Americans, Caribbean-Hispanics, and Israeli-Arabs. They found that certain variants of the SORL1 gene are more common among people with late-onset Alzheimer’s. Furthermore, the genetic signature appears different among ethnic groups; in this case, the Caucasians and African Americans showed diverse variants. Published in Nature Genetics in 2007, SORL1 was only the second gene, after APOE, to be linked to late-onset Alzheimer’s, and the first to demonstrate racial differences in disease risk variants in the same gene.

“The racial background turns out to be really important in the development of the disease,” Abraham says. She and Farrer are investigating a protein called AKAP9, which has variants discovered by Farrer and associated with late-onset Alzheimer’s, that appear only in African Americans. Such variants are critical, because African Americans between the ages of 65 and 74 are more than three times as likely to develop Alzheimer’s as Caucasians, and those between the ages of 75 and 84 are nearly twice as likely. “On a biochemical level, we need to know: what is this doing in the brain?” she says. “It’s only by understanding the proteins involved in the biochemical pathways that we can design rational drugs.”

“Ultimately, treatment is not going to be one size fits all,” Farrer says. “There are people who have a set of risk factors—some genetic, some other—that may contribute to disease; if you are able to identify them, this will inform the development of new treatments and then doctors can apply the particular treatment that is best suited for the individual person. This is a glimpse of personalized medicine in the future.”

November has been designated as National Alzheimer’s Disease Awareness Month.

Read the other stories in our “Unraveling Alzheimer’s Disease” series here.

+ Comments
Barbara Moran, Senior Science Writer
Barbara Moran

Barbara Moran can be reached at bmoran@bu.edu.

Post Your Comment

(never shown)