The protein amyloid-beta (Aβ) was long considered the primary “bad guy” of Alzheimer’s disease, accumulating into plaques and causing plenty of trouble for nearby cells. These days, though, we know that Aβ doesn’t act alone. One of its possible co-conspirators? A malfunctioning brain immune system.

To safeguard the brain from invading microbes, passage from the bloodstream into the brain is tightly restricted; even cells of the immune system don’t generally go there. Instead, the brain employs a local system that serves as both police force and sanitation department: the microglial cells. These cells rove throughout the brain maintaining order. They respond to threats, manage inflammation, and, when necessary, clear out toxic materials. Recent work suggests that microglia also play an important role in the brain’s response to, and destruction of, Aβ. Long-term exposure to Aβ, by contrast, makes microglia less effective. But what mediates this microglial response, and is there any way to harness it for treatment?

Enter Dr. Jenny Johansson and colleagues of Stanford University, authors of a study published this January in the Journal of Clinical Investigation. The researchers were inspired by population studies showing that NSAIDs – a class of anti-inflammatory drugs that includes ibuprofen and aspirin – prevent development of Alzheimer’s disease. Still other studies indicate that NSAIDs block production of the protein prostaglandin, which can interact with microglia through a receptor protein called EP2. EP2 sits on the outside surface of microglial cells, waiting for prostaglandin to bind to it and thereby initiate a signaling cascade – imagine prostaglandin as a snowball that starts an avalanche. The authors hypothesized that EP2 signaling suppresses normal microglial responses to threats like Aβ, and that if one could turn EP2 off, the microglia would be protective again.
As it turns out, when we’re older, our microglia don’t work as well. When the researchers introduced Aβ into cells from both old and young mice, the EP2 response was more robust in the older mice, and associated with an inflammatory response. The older mice also produced fewer of the proteins that break down Aβ. In other words, the older a mouse was, the less capable its microglia were of managing threats and the more violently they responded – not so good for the brain.

And what if one gets rid of EP2? The researchers looked next at mice genetically engineered to develop symptoms of Alzheimer’s disease; when these mice lacked EP2 from birth, the plaques in their brains were smaller, and around each plaque were more microglia than in mice with EP2. Johansson et al. also examined the acute response to Aβ by injecting the protein directly into the brain of a non-Alzheimer’s mouse. In normal mice, this procedure increased inflammatory signaling. When the researchers repeated the procedure in mice lacking EP2, they found that not only were many of those inflammatory signals reduced, but that several factors that promote cell survival after brain injury were increased. EP2 deletion didn’t just block inflammation: it supported a more constructive cellular response to Aβ.

The kicker, of course, is that it didn’t matter when the mice lost EP2, nor how they were exposed to Aβ: EP2 deletion improved the mice’s performance in memory tests on which EP2-positive mice exposed to Aβ normally do poorly. Now that’s a powerful little receptor protein! Could modifying EP2 be a quick fix for Alzheimer’s disease? It seems reasonable to suppose, doesn’t it?

If you’re wondering why this magic anti-inflammatory bullet hasn’t cured Alzheimer’s yet, well, you’re right to be skeptical. Despite the abundance of evidence from epidemiological studies suggesting that NSAIDs prevent Alzheimer’s disease, not one clinical trial has borne out the promise of such treatments. And there have been multiple trials.

Though the Stanford findings are novel and exciting, a notorious challenge of Alzheimer’s research is that drugs that succeed in mice often fail in human trials, in part because we can’t treat humans in time. Current studies suggest that, by the time someone’s decline merits a trip to the neurologist, disease progression has been underway for decades. While many scientists are working to develop reliable pre-clinical markers, this skewed treatment window may help explain why epidemiological studies suggest benefits where clinical trials find none. Perhaps at-risk subjects in these studies are, quite by accident, pre-treating themselves well before disease onset. And if that’s true, then once we optimize an early-detection system, perhaps anti-inflammatory drugs deserve another chance in the clinic.

Until then, an aspirin a day probably won’t keep the doctor away – at least not in Alzheimer’s disease.

Further reading:
Various groups have published several detailed reviews