NSAIDs Could Hold Key to Blocking Alzheimer’s-Related Protein

Researchers at the University of California, San Diego (UCSD), and the Mayo Clinic in Jacksonville, Fla., have discovered that certain nonsteroidal anti-inflammatory drugs (NSAIDs) directly block the production of amyloid β-42, a peptide essential to the development of the brain plaques that are the hallmark of Alzheimer’s disease (AD). The finding could lead to the development of new targeted drugs that block the protein without the side effects commonly associated with NSAIDs.

In a report in the November 8 Nature, lead investigators Todd Golde, M.D., Ph.D., a consultant and assistant professor of pharmacology at Mayo, and Edward Koo, M.D., an associate professor of neurosciences at UCSD, detail findings from a long-standing collaboration between their two teams. More than two years ago, the groups began looking at how amyloid β is produced in the brain.

Prior research indicated that the plaque-building, amyloid β peptide was the end product of cleavage of amyloid precursor protein (APP) by the recently identified enzyme, gamma secretase. The two teams started looking for ways to modulate activity of gamma secretase in hopes of being able to down regulate the production of the disease-causing form of amyloid β.

Epidemiological data had documented that there appeared to be a reduced prevalence of AD in patients who had been long-term users of NSAIDs. Researchers had speculated that the effect might be due to the anti-inflammatory effects of the drugs through their inhibition of the enzyme cyclooxygenase (COX). It was thought that the highly damaging inflammatory process, if it were occurring in a brain in the early stages of AD, could speed up or even cause the plaque-building process and that by blocking the inflammation in patients taking NSAIDs, the development of AD might be slowed or possibly blocked.

Golde and Koo wondered whether NSAIDs could be altering the actual processing of APP and, therefore, the production of amyloid β. In particular, they wanted to know if NSAIDs could alter the production of one specific peptide cleaved from APP, the amyloid β-42 peptide (Aβ42—an amyloid β chain that is 42 amino acids in length).

Aβ42 has been shown in previous work to be the main culprit in the development of brain plaques, whereas other common but shorter chains (38 or 40 amino acids in length) do not appear to cause pathology.

In the current research the two teams examined whether NSAIDs could alter levels of amyloid β (both the 40 and 42 amino acid chains) in cultured mouse, hamster, and human cells, as well as in mice genetically altered to produce amyloid plaques. The work was funded by grants from the National Institutes of Health and several independent foundations.

“Lo and behold,” Golde told Psychiatric News, “several of the nonsteroidals that we tested lowered Aβ42 selectively.” The team screened all NSAIDs currently approved by the FDA in the cell cultures. Only a few of them actively reduced Aβ42 in culture. In particular the group found that ibuprofen, indomethacin, and sulindac preferentially reduced the 42 amino-acid chain, while other NSAIDs had little or no effect on amyloid metabolism. Each drug active in cell culture is now being tested in animal models.

‘Batting 100 Percent’

“So far,” Golde said, “we are batting 100 percent—every time it lowers Aβ42 in cell culture, it lowers it in animals.”

Interestingly, the researchers found that total amyloid β production was not altered by the NSAIDs. While Aβ42 levels decreased, there appeared to be an offsetting increase in Aβ38 production. The significance of the increase in the 38 amino-acid chain is not yet fully understood, Golde told Psychiatric News. The researchers do not believe, however, that the 38 amino-acid chain contributes to the buildup of brain plaques.

What is significant, Golde emphasized, is that amyloid β production is simply being modulated by the NSAIDs, not actually inhibited.

“What is important here is that in this pathway, the NSAIDs are shifting cleavage [from the 42 to the 38 amino-acid chain] and not really altering total production of amyloid β. All of this leads to the likelihood that these particular NSAIDs are acting to shift cleavage by modulating gamma secretase activity,” he explained. Actual inhibition of the gamma secretase, he said, could have far-reaching implications, although the full picture of the enzyme’s functions is not yet understood.

The UCSD and Mayo groups then looked to see whether the changes in Aβ42 were tied to NSAID inhibition of COX. They measured amyloid β levels in cells exposed to the NSAIDs that were genetically altered to remove the COX enzyme. Again the researchers saw preferential reductions of Aβ42 in the absence of any possible COX interaction.

The team is continuing the line of research beyond that reported in Nature and believes that it is likely that by understanding the true target of the drugs, they will be able to define a novel site for drug targeting. This could lead to the development of drugs that modulate the production of Aβ42 in such a way that the course of AD is altered. The two teams already have potential drug compounds that have no COX activity but do lower Aβ42.

“So we think that we’ll be able to evaluate those as a potential therapy for AD and that we’ll avoid the COX mediated toxicity,” Golde said.

Caution on NSAID Use

The last thing the team would want, he emphasized, is to deliver the message that people need to go out and start taking high doses of ibuprofen.

“In order to see this effect, you are probably near or above the maximum dose of ibuprofen you could safely take,” Golde explained. At such doses, the incidence of gastrointestinal breakdown and kidney damage would be high.

Golde added, “The real question will be whether an antiamyloid therapy will be efficacious once an individual already has AD, or does it have to be instituted well before any clinical symptoms develop?” He noted that most researchers believe there is a long lag phase between deposition of amyloid into plaques in the brain and the appearance of clinical symptoms. “Antiamyloid therapy,” Golde said, “much like anticholesterol therapy, is much more likely to be effective as a preventive strategy than as a therapeutic strategy, although it certainly wouldn’t hurt as a therapeutic strategy as well.”

An abstract of the report, “Some NSAIDs Lower Amyloidogenic Aβ42 Independently of Cyclooxygenase Activity,” can be accessed on the Web at www.nature.com by searching on “Golde.” ▪