Share this article on:

A Novel Gene Pathway to Alzheimer's Discovered, Offering New Potential Therapeutic Target

Talan, Jamie

doi: 10.1097/01.NT.0000431664.17135.24
Back to Top | Article Outline




Investigators reported that CD33, a gene involved with innate immunity, has an active role in microglia in the brain and is key to sending the signal for microglia to clear away toxic amyloid-beta (Abeta). The findings suggest that it is possible to get rid of accumulating Abeta years after the disease has set its course.

A team of Harvard scientists has discovered a new pathway and a novel gene function critical to Alzheimer's disease (AD). The finding — that CD33, a gene involved with innate immunity, has an active role in microglia in the brain and is key to sending the signal for microglia to clear away toxic amyloid-beta (Abeta) — suggests that it is possible to get rid of accumulating Abeta years after the disease has set its course.

Rudolph E. Tanzi, PhD, and his colleagues identified CD33 in a large genome-wide association study (GWAS) in AD published in 2008 in the American Journal of Human Genetics. Four risk genes for late onset AD were added to the already well-known risk gene, apolipoprotein E4 (APOE4). The investigators have been working since to figure out CD33's involvement in AD. It was only known that it is a cell surface receptor that plays a role in innate immunity.



Dr. Tanzi and his colleagues say they now envision that small molecules developed to target CD33 in the brain could take the brake off microglia so it can engulf and chew up Abeta. Their study, published in the April 25 online edition of Neuron, details the potentially game-changing discovery.

CD33 is mainly in microglia and works as a switch to control how fast microglia can clean away toxic debris, which in AD means Abeta. Dr. Tanzi's group found that too much of the receptor can result in reduced clearance of Abeta and that process sets in motion a series of steps that trigger late onset AD.

“CD33 is the key to clearing away Abeta protein from the brain as we age,” said Dr. Tanzi, director of the Genetics and Aging Unit in the Massachusetts General Hospital department of neurology. The team is now working to identify or develop treatments that can safely inactivate CD33 on microglia that, Dr. Tanzi added, should stop the accumulation of the sticky protein that leads to AD.

There is evidence that Abeta starts accumulating in the brain ten to 15 years before the first clinical signs of disease.

Back to Top | Article Outline


The scientists conducted many different studies to make a case for the importance of this new pathway. First, they looked in brain tissue from people who died with and without AD and found AD patients had increased expression of CD33 on the surface of microglial cells. Abeta deposition increased as the numbers of CD33 receptors did. There was evidence from other groups that the minor allele of the CD33 gene variant called rs3865444 actually protects against AD. When they looked at the AD and non-AD brains of carriers of the protective allele, Dr. Tanzi's group found that those who carried it had decreased CD33 levels in their microglia and reductions in insoluble Abeta-42 levels.

Next, they looked at how CD33 affects the ability of microglial cells to clear Abeta in cell culture models. Using a variety of genetic variants of CD33, they showed that the more CD33 there was on the microglial cell surface, the less the cells ingested Abeta protein. When CD33 levels were lowered in the microglia, they engulfed much more Abeta protein and degraded it.

The investigators subsequently created a transgenic CD33 knockout mouse. As they expected, the animals had lower levels of Abeta. Without CD33 to put the brakes on microglia, the scavenger cells were faster at removing Abeta. The quintessential study was a double transgenic: the APP Swedish mouse model with the CD33 knockout. Again, the animals had less Abeta, and they also had much less plaque formation.

“We ameliorated AD pathology,” said Dr. Tanzi. “And we think that this means we can potentially manipulate microglia to clear away Abeta. If we can inactivate or turn down expression of CD33, this should allow microglia to get rid of Abeta. It's pretty exciting.”

This could prevent the development of clinical AD, he added. “Abeta accumulation drives tauopathy and the ensuing excess neuronal cell death eventually triggers inflammation. If we can get microglia to clear more Abeta, we avoid tauopathy and inflammation and, thus, cell death.”

The discovery fits nicely with another newly identified late AD risk gene, TREM2 (triggering receptor expressed on myeloid cells 2). “It's the yin to CD33's yang,” said Dr. Tanzi. TREM2 works on inflammatory pathways and activation of TREM2 can block microglia from going into an inflammatory state. The researchers think that CD33 works with TREM2 in microglia to regulate inflammation in the brain.

Dr. Tanzi and his colleagues are now conducting studies to see how manipulating both genes could alter the course of AD — at the level of Abeta clearance and on inflammation.

The researchers must strike a complex balance to succeed. While microglia clear cellular debris from the brain and secrete neuroprotective factors, too much immune activity can cause microglia to go into inflammatory overdrive and shoot off free radical species and cytokines that can kill neurons “by friendly fire,” said Dr. Tanzi.

In theory, CD33 inhibition could be used to treat other neurological diseases that involve abnormal protein aggregation. “This model could work in virtually all neurodegenerative diseases with misfolded proteins,” said Dr. Tanzi. “If you can stimulate microglia via CD33 to clear abnormally aggregated protein you can potentially save neurons and simultaneously stave off inflammation.”

Back to Top | Article Outline


When Robert Vassar, PhD, a professor of cell and molecular biology at the Feinberg School of Medicine at Northwestern University, saw the data, he couldn't see a downside to the model. He still doesn't. “This is another avenue to eliminate or reduce levels of Abeta peptide in the brain. It could definitely work to prevent or slow AD. It is a tremendous discovery.

“Increasing phagocytosis would be a good thing,” he added. “They have shown that CD33 puts the brakes on phagocytosis.” Of course, he said, there is a lot of work to be done in finding just the right molecule to inhibit CD33. “Every drug has a therapeutic window, a sweet spot. You don't want to over-inhibit CD33.”

Dr. Vassar's lab at Amgen, Inc., discovered beta-secretase in 1999. It is one of two important enzymes that cleave the amyloid precursor protein and trigger the release of Abeta from cells. There are dozens of compounds in development and testing that target beta-secretase as a potential treatment for AD.

“It's a convincing story,” said Thomas Lehner, PhD, MPH, director of the Office of Genomics Research at the National Institute of Mental Health, a co-funder of the Harvard study. “What I find most intriguing is that people have known about APOE4 and it has not yet evolved into a target for treatment. In identifying and then understanding CD33's risk in AD, these researchers were able to figure out how it worked and showed how they can target CD33 and clear away amyloid-beta. They showed us a new pathway and a statistical signal, and used cutting-edge science to show a biological connection between the gene and a phenotype. This could really be a stepping stone for the development of therapeutic drugs for AD.”

Sam Gandy, MD, PhD, professor of neurology and psychiatry and Mount Sinai chair in Alzheimer's disease research and director of cognitive health at Mount Sinai Medical Center in New York, said that the field has been focused on genes that regulate how Abeta is initially produced and “there has been a suspicion that we were missing part of the picture — a defect in the clearance of Abeta. Genetics point you to things that are important in the disease pathway and we were hoping that we would find something that would regulate this clearance.

“We know that amyloid is being deposited 20 years before symptoms,” he added. “Until now, we didn't have a way to stimulate the garbage disposal system to clear out the amyloid-beta deposits. This puts us on a new track. It might be the best way to clear out the plaques.” The way he sees it, a person might have amyloid build-up beginning at 50 and take a pill to clear out the toxic protein that might stall the process for another decade or two.

Back to Top | Article Outline


•. Griciuc A, Serrano-Pozo A, Tanzi RE, et al. Alzheimer's disease risk gene CD33 inhibits microglial uptake of amyloid beta. Neuron 2013. E-pub 2013 Apr. 25.
    •. Bertram L, Lange C, Ionita I, et al. Genome-wide association analysis reveals putative Alzheimer's disease susceptibility loci in addition to APOE. Am J Hum Genet 2008;83:623–632.
      •. Neurology Today archive on Alzheimer's disease:
        ©2013 American Academy of Neurology