Previously Undiscovered Immune Cell Population Affects Learning and Memory

Venture into any immunology lab, and you’ll see at least one complicated poster mapping out the body’s immune cells. A quick look at one of these guides will tell you that there are a lot of immune cells with a lot of different functions. But a new study has suggested those posters might need to be redesigned. Not only are there immune cell populations hiding undiscovered in the body, but they might have impacts far beyond protection from intruding microorganisms and diseases.

The study, published in Science Immunology, suggests that, at least in mice, a previously undiscovered population of cells may impact learning and cognition in the early stages of life. Alongside other research, these findings point towards a growing understanding that immune cells may alter cognitive processes such as learning and sociality, despite the brain once being considered an “immunoprivileged” area that immune cells couldn’t affect.

Miguel Ribeiro, a researcher at the University of Lisbon, alongside an international team of colleagues investigated a population of immune cells, called γδ Th17 cells, that are resident in the outer layers of the brain, called the meninges. Removal of these cells in mice saw the mice suffer impacts to their short-term memory. The authors showed that injecting a neurotrophic molecule, induced by the γδ Th17 cells, rescued these cognitive deficits and suggested that the effects may be mediated by affecting synaptic plasticity, an important neurobiological mechanism that is key to learning.

Speaking to Technology Networks, study author Julie Ribot, a postdoctoral researcher at the University of Lisbon, explained that the structure and function of the lymphatic vessels that connect to the meninges had only been described a few years previously. Ribot’s initial goal was simply to work out whether γδ Th17 cells could make it past the brain’s tight immune defenses, collectively termed the blood brain barrier (BBB), to infiltrate the meninges. Th17 cells are usually found outside the brain, in the peripheral immune system. They are defined by their production of the immune cytokine IL-17, and are implicated in a host of neurological disorders such as the multiple sclerosis-mimicking model disease EAE.

Whilst many immune studies might involve patients or animals exposed to particular diseases or external insults, Ribot’s experiments looked at healthy mice, in the so-called “steady state”. “The cells are generated in the embryonic stage in the thymus, and at birth, migrate to the meninges, and colonize the tissue,” says Ribot. Her co-author Bruno Silva-Santos, a professor of medicine, explains that the cells’ numbers peak in the first week of life. “It’s interesting to think, if this is conserved in humans, that they could be an early contributor to learning and cognition,” says Silva-Santos.

Building chimeras

To study what happens when a particular protein or cell type is absent, researchers usually create a transgenic animal model, bred to be deficient in that cell type. But, Ribot explains, the field currently lacks a method of reliably targeting just γδ Th17 cells, let alone the population of those cells that inhabit the meninges. So, the team created chimera mice, which had their bone marrow destroyed, but were subsequently given adult bone marrow. As the γδ Th17 cells that the team wanted to target were uniquely produced from embryonic bone marrow, the chimeric knockout mice lacked only the meningeal cells of interest.

A safe place in the brain

The team tested their chimeric mice against control mice in a battery of memory tests. On long-term memory tests, such as a water maze that requires mice to remember where a platform is hidden over several days, there was no difference between groups. But when tested on their short-term memory, the chimeras performed worse.

Why would cells designed to boost memory be found in the outskirts of the brain, as opposed to in the thick of the action, in learning centers like the hippocampus? Silva-Santos suggests that the meninges act as a holding area, where the γδ Th17 cells can exert their pro-cognition effects without causing chaos within important brain structures. “This idea of an immune system in constant crosstalk with the brain and cognition; it seems the meninges were chosen as a safe place where this can happen. The cells infiltrate the meninges, coming from the lymphatic vessel, and produce soluble factors that can penetrate the parenchyma (the internal structures of the brain) without the cells themselves going there, and therefore without the danger of the cells attacking the cells of the brain.”     

A neurotrophic link

So how are these cells mediating an effect on memory? The team noticed that glial cells (support cells in the brain) supplemented with IL-17 produced higher amounts of brain derived neurotrophic factor (BDNF) a molecule with an important role in learning and memory. BDNF exerts its effects mainly through modulation of synaptic plasticity. Injections of BDNF into the chimeric mice rescued their short-term memory, which Ribot and Silva-Santos suggest implicates BDNF as the link between the immune cells and memory.

Why specifically short-term memory? Silva-Santos says that short-term memory may be dependent on the γδ Th17 cells’ presence in the brain. “For long-term memory, where you get training and basically, you get reinforcement of these experiences, you are recruiting immune cells from the periphery.” Once those other cells are in play, it’s not such an issue if meningeal IL-17 is lost.

The “Goldilocks” level of inflammation

There are plenty more steps of the process to detail. The mechanisms by which peripheral T cells might help long-term memory will need exploring, and a model that genetically abolishes γδ Th17 cells would provide additional evidence to back up the results from the chimeric mice.

But these are intriguing results that place additional responsibility on immune cells. Inflammation has been widely recognized as a response that accelerates disease processes, including neurodegenerative disease. How can we square this up with a potentially essential role for inflammation in learning? Ribot suggests that the solution might be a “Goldilocks” level of inflammation – not too much, not too little: “You really need a very specific amount of inflammation; too little is not good for learning in our knockout animals, because they have an impaired memory, but too much will also be very harmful because T cell infiltration will disrupt the BBB and form neurotoxic events.”

The team’s next steps are to investigate the role of their newfound cells in a diseased environment – and those immunology lab posters may have to be redesigned again soon.