A new study raises the odds that a strategy already successful against some cancers may be deployed against Alzheimer’s. The research, which highlights the role of an immune system “checkpoint” molecule, showed improved cognition in tests with mice. It was published earlier this month in Nature.

In this edited conversation, the Gazette spoke with Vijay Kuchroo, the Samuel L. Wasserstrom Professor of Neurology at Harvard Medical School and Brigham and Women’s Hospital, and director of the Gene Lay Institute of Immunology and Inflammation of Brigham and Women’s Hospital, Massachusetts General Hospital, and Harvard Medical School.

Kuchroo, who was a senior author on the paper, outlined work that deleted the expression of a molecule called TIM-3, which blocks brain immune cells called microglia from attacking Alzheimer’s plaques, freeing the cells to clear plaques and restoring memory.

---

Your work was done in a model of late-onset Alzheimer’s disease. What proportion of cases is it?

Most cases of Alzheimer’s disease (AD), 90 percent to 95 percent, are late-onset. The molecule that we studied, called TIM-3, was linked by a genome-wide association study to late-onset Alzheimer’s and was found to be a genetic risk factor for the disease. There’s a polymorphism in the TIM-3 gene, HAVCR2, in patients with AD. TIM-3 is an inhibitory molecule utilized by the immune system to turn off the immune cells once activated. TIM-3 belongs to a group of inhibitory molecules called checkpoint molecules, which have been exploited for treatment of cancer.

Checkpoint molecules stop the body from attacking itself?

That’s one way to put it. If your immune system gets activated, the checkpoint molecules restrain the immune system from getting out of hand.

The best example is that every time you get an infection like the common cold, your lymph nodes get swollen because you make millions and millions of T cells to fight the virus. Once the infection goes away, checkpoint molecules come in to reduce the number of T cells to a normal level.

Cancers have exploited these checkpoint molecules for their own survival, and every time a T cell goes to attack a tumor cell, the tumor cell induces expression of checkpoint molecules so the T cells don’t attack the tumor cells. The T cells become dysfunctional or exhausted, and the tumor survives.

The new twist is that in Alzheimer’s disease, there is the accumulation of plaque in the brain that doesn’t get cleared by macrophage-like cells called microglia. The microglia show an increased expression of the checkpoint molecule TIM-3.

They’re basically the immune cells of the brain?

Microglia are the immune cells of the brain and have other important functions. During development, synapses are being formed, and synapses are how memory is stored. The problem is that even transient experiences make memories, so you want to get rid of some memories that are not being used again. So, the major job of microglia cells during development is to prune synapses that have not been used often enough in order to sharpen and sustain your memory.

After you’re born and have developed memories, you don’t want to lose them, so at about 28 to 40 days after birth in the mouse and a few months to few years in a human, there is a developmental mechanism by which microglia stop pruning to keep the memories that are made.

To stop the microglia from pruning, they increase expression of the checkpoint molecule TIM-3, and these microglia cells become homeostatic, they do not phagocytose anymore.

That’s good because you don’t want to prune your own memory, but it’s bad as you get older and accumulate gunk in the brain, which can’t be cleared. Who’s going to clean it up? Microglia cells have become homeostatic, and TIM-3 keeps them from engulfing the accumulated gunk, which results in the formation of plaques.

What’s the difference in TIM-3 in an older person who has Alzheimer’s disease, versus not?

There’s a polymorphism in the gene, and in Alzheimer’s patients with the polymorphism, TIM-3 is highly expressed on microglia, significantly more than those that don’t have the disease.

So that all that TIM-3 keeps the microglial cells at homeostasis and not attacking amyloid beta plaques even though they’re harming the brain?

Yes, microglia cells should be clearing amyloid plaque, but they don’t. We discovered this molecule on T cells in the immune system, but it is 100 times — in some cases 1,000 times — more expressed on microglia when they get activated.

So, the same molecule that’s shrinking the T cell population to normal size after infection is being used by microglia cells to stop them from excess pruning. But it’s also a liability, because it inhibits them from attacking plaques that accumulate in Alzheimer’s disease.

You tested this with lab mice who have the HAVCR2 gene — which makes TIM-3 — deleted?

Yes, these mice were made to test the role of TIM-3 in immune system autoimmunity and cancer.

We used the same mice. We genetically deleted the gene, and in these mice the microglia don’t express TIM-3 when the microglia get activated. That enhances clearance of the plaques and changes plaque behavior.

Toxic plaque has fingerlike projections that enter into the brain, but with the microglia nibbling on them, the plaques become compact. So, the deletion of TIM-3 in microglia not only reduces the number of plaques, it also changes the quality of the plaque. These mice actually get the cognition back. Not completely, but the cognitive behavior of these mice improves.

And when we talk about measuring cognitive behavior for mice, we’re talking about their ability to remember and navigate mazes?

That’s correct. When they have plaque burden in their brains, they don’t remember as much. They also have less fear. If you put them in an open space, normal mice will go to a corner, so they don’t wind up as prey. But if they have plaques, they sit there in the center of the maze and don’t hide. When you get rid of the plaques, memory comes back, and that response comes back, because an appropriate level of fear is important for survival.

What would a TIM-3 therapy for Alzheimer’s disease in humans look like?

Therapy would use an anti-TIM-3 antibody or a small molecule that can block the inhibitory function of TIM-3.

What’s the potential of this to make a difference against Alzheimer’s disease? After several failures of major drug trials, recently there have been some successes, though those showed just minor improvement.

Because amyloid beta is also in the endothelium in the blood vessels, a lot of antibody doesn’t go to the brain, it attacks the blood vessels, leading to strokes due to vascular damage, limiting the use of anti-amyloid antibodies in AD. Since TIM-3 has selective expression, existing anti-TIM-3 antibodies can be repurposed for treatment of AD.

How long did this work take?

Five years; each experiment takes about eight, nine months. I want to emphasize that this was in collaboration with a colleague here, Oleg Butovsky at the Ann Romney Center for Neurological Disease. There were about six people, three from my lab and three from his lab, who worked tirelessly to do these experiments.

What happens next?

We are trying to see whether human anti-TIM-3 can halt development of plaques in the brains of Alzheimer’s disease mouse models. We have a mouse model in which the human TIM-3 gene has been inserted, which will be very suitable for testing various candidate antibodies for human disease.

---

This research was funded in part by the National Institutes of Health.