Microglia and Complement Prune Synapses

A developing retinal neuron makes C1q, the first protein of an immune cascade that marks bacteria for destruction.

Complement is a set of proteins best known from the blood. They bind microbes, dead cells and clumps of antibody, and they coat them so that phagocytes, the cells that eat other cells, will swallow them. Finding complement in a young neuron looked like a category mistake. It turned out to be a clue to one of the main ways a developing brain removes connections it has made too many of. This lesson follows that clue from a gene list to the inside of a microglial lysosome.

Engulfment and phagocytosis: the brain keeps its own eaters

The brain builds more synapses than it keeps, and what is removed has to go somewhere. Two kinds of glia do much of the eating. Microglia are the brain’s resident immune cells and its main phagocytes. Their origin, outside the neural tube, belongs to that lesson. They carry many of the sensors of an immune cell, including Toll-like receptors that switch on NF-κB, which that lesson covers. Astrocytes, the star-shaped glia that wrap synapses, eat too: in the mouse visual thalamus they engulf synapses through the receptors MEGF10 and MERTK.

So a synapse is not only a meeting of two neurons. The classical picture is bipartite, a presynaptic terminal and a postsynaptic cell. Adding the astrocyte that enwraps it gives the tripartite synapse. Adding the microglial process that touches it and can remove it gives what Chris calls the “quadripartite” synapse.

A phagocyte has two jobs before it eats: it has to get close, and it has to choose. That is the logic of the whole lesson. Some signals say “find me” and draw the eater in. Others say “eat me” and mark the target. The best-studied system for both is the dorsal lateral geniculate nucleus (dLGN), the relay in the thalamus where retinal axons end. In a newborn mouse, inputs from the two eyes overlap there. Over roughly the first postnatal week and a half they sort into separate territories, a process driven partly by spontaneous retinal waves (see that lesson). Inputs that lose are removed.

An immune cascade at a synapse

In the immune system, the classical complement pathway starts when C1q binds a target. C1q is shaped like a bunch of six tulips, each head able to grip a surface. Binding activates two proteases in its complex, which cut C4 and C2. The pieces assemble into an enzyme, a C3 convertase, that cuts C3 into a small fragment, C3a, and a large one, C3b. Cutting exposes a reactive bond in C3b that fixes it to the nearest surface. C3b is then trimmed to iC3b. A phagocyte carrying complement receptor 3 binds iC3b and engulfs whatever it coats. Complement receptor 3 is an integrin, the CD11b and CD18 pair, of the adhesion-receptor family taught in that lesson.

Beth Stevens, working in Ben Barres’s laboratory at Stanford, found that purified mouse retinal ganglion cells raised their expression of C1q when grown with astrocytes. In the young mouse, she and her colleagues found C1q in the retina and dLGN early in postnatal life; neuronal C1q is normally turned down in the adult. In the dLGN, C1q and C3 lay at many small synaptic contacts, which the team took to be immature or disappearing ones.

The test was genetic. A knockout mouse lacks a working copy of one gene in every cell. In mice lacking C1q, the two eyes’ territories in the dLGN failed to refine normally and still overlapped more than in normal littermates. Recordings from single relay neurons told the same story in another way: most normal relay cells had pruned down to few inputs, while most C1q-knockout cells still received several. Mice lacking C3 showed a similar overlap. The authors proposed that unwanted synapses are tagged by complement for elimination.

Notice what the knockouts did not show. They showed that complement is needed for normal refinement in this circuit. They did not show which synapse got a tag, what read it, or which cell carried the input away.

Caught in the act: microglia take the quieter eye

Dorothy Schafer, in the Stevens laboratory with Barres and others, went after the missing step in 2012. The team injected red and blue tracers into the two eyes of mice at P4 and looked in the dLGN a day later, in mice whose microglia were marked green. Retinal material sat inside microglial lysosomes. A dye that glows only in acid, and a genetic label of retinal neurons, gave the same result. Electron microscopy and staining for VGlut2, a protein of retinal terminals, showed that the engulfed material was mainly presynaptic.

Engulfment had a clock. It was high at P5, when segregation is under way, and much lower by P9. It also had a receptor. Microglia lacking complement receptor 3, or in mice lacking C3, took up roughly half as much. Those mice later had more overlap between the eye territories and more retinal synapses about a month after birth. About half the uptake remained without the receptor, so other routes contribute.

The sharpest experiment asked whether engulfment follows activity. Tetrodotoxin (TTX) blocks sodium channels and silences neurons; forskolin raises cAMP and increases their activity. With one eye silenced, microglia held more material from that eye. With one eye boosted, they held more from the untreated eye. The comparison was between whole eyes, not between identified synapses, and it did not reveal how activity becomes a tag. Still, it points one way: in this circuit, microglia preferentially remove less active inputs.

engulfment index = (volume of retinal label inside a microglial cell) / (volume of that microglial cell) × 100

Design problem

You inject different tracers into the two eyes of P4 mice. In one group you silence one eye with TTX; in another you boost activity in one eye with forskolin. A day later you measure the engulfment index for each eye’s material in dLGN microglia. Predict the result for each group, and say what control you need to make the comparison fair.

”Find me” signals and “eat me” signals

Find-me signals act at a distance. ATP released from damaged cells draws microglial processes toward the damage. Fractalkine, also called CX3CL1, is a membrane protein on neurons that can be cut free; its receptor, CX3CR1, is on microglia. In 2011 Rosa Paolicelli, in Cornelius Gross’s laboratory, reported that mice lacking CX3CR1 had fewer microglia and more dendritic spines in the developing hippocampus, with signs of immature connections. The excess was transient, so the result is best read as delayed pruning in that region. Later studies in other circuits have given mixed answers about whether CX3CR1 is required.

Eat-me signals sit on the target. Complement tags are one kind. Phosphatidylserine (PS) is another. PS is a membrane lipid normally kept on the inner, cytoplasmic face of the membrane (see that lesson). A dying cell moves it to the outer face, where phagocytes recognize it. In 2020, a study jointly led by Nicole Scott-Hewitt and Fabio Perrucci, from Stevens’s group and Michela Matteoli’s group, injected a PS-binding probe into the brain ventricles of 4-day-old mice. In the dLGN, PS appeared on the presynaptic side of a small fraction of synapses, and that fraction fell between P5 and P10. In the hippocampus the signal peaked later. Mice lacking caspase-3, an enzyme of apoptosis, showed no significant difference in the PS signal. So a synapse can display a death signal locally while its neuron lives. Chris calls this the differential use of PS and complement: two kinds of tag, possibly read by different receptors, possibly working together.

Two other forms of removal belong elsewhere. Axons that prune themselves using caspases and p75 are in that lesson, and the loss of extra inputs at the neuromuscular junction is in that lesson.

When pruning goes wrong: links, not causes

If pruning shapes circuits, too much or too little might matter for disease. Changes in synapse number have been proposed to contribute to autism and to schizophrenia. These are hypotheses, supported by associations, and the disorders themselves are covered in that lesson.

The strongest single example is genetic. In 2016, Aswin Sekar, Steven McCarroll and colleagues, including the Stevens laboratory, studied C4, the complement protein cut right after C1q acts. The human gene comes in two forms, C4A and C4B, in varying copy numbers. Variants predicted to raise C4A expression in the brain were associated with a higher risk of schizophrenia. In mice, loss of C4 reduced pruning in the dLGN. Together these suggest that more complement activity might mean more pruning, which fits the long-standing observation of fewer dendritic spines in some schizophrenia brains. Each copy of C4A adds only a small increase in risk, and the mouse result comes from the visual thalamus, not the human cortex.

Keep the scope in view. The engulfment and knockout results come from mouse dLGN in the first postnatal week and a half. Microglia prune in other regions too, but which signals they use varies by circuit and age, and later imaging suggests they often take small bites of presynaptic membrane rather than whole synapses.