Building a brain is a complicated business. During the development of the human brain, billions of cells are born, orderly migrate to their appropriate locations, and synthesize connections with other cells near and far. Years ago, scientists discovered that, counter-intuitively, assembling a brain requires a great deal of disassembly. Only about half of the specialized brain cells (called neurons) generated during infancy survive to function in the adult brain. In addition, many of the connections between neurons are established only to disappear at a later stage of development. This disassembly of specific connections is critical for normal brain function as an adult, but why the brain evolved such a seemingly wasteful approach to self-assembly is a mystery. Researchers have a much better — although still emerging — understanding of how these connections disappear. Research published in the December issue of the journal Cell implicates some unexpected players in this process –immune molecules– which appear to flag some connections for destruction. These findings illuminate not only how the brain develops, but may also provide insight into how the nervous system degenerates in diseases such as glaucoma, Alzheimer’s disease, and Amyotrophic Lateral Sclerosis (ALS).

An unusual suspect

In the adult brain it is estimated that the average neuron makes approximately 1000 connections with other neurons, but some neurons may make as many of 200,000. These connections, called synapses, are critical because they enable the communication of information from one cell to another, and enable all our cognitive functions. Problematic synapses are thought to underlie many health conditions such as autism and schizophrenia. Researchers at Stanford University discovered that an immune protein, known to immunologists as C1q, is unexpectedly present in neurons of the developing mouse brain. The presence of this molecule in the brain was puzzling because previously immune proteins like C1q have only been associated with the brain in the context of disease, when the immune system is in overdrive. Further enhancing the mystery, the researchers noticed that not only was C1q present on neurons, its appearance coincided with the time during development when the most dramatic synapse elimination takes place. Based on this circumstantial evidence they hypothesized that C1q might shape synaptic architecture.

To ask whether C1q was driving synapse elimination, the researchers looked at synapses in mice engineered to be deficient in C1q. Studying well-defined connections between cells in the retina and neurons deep in the brain that receive visual information, they discovered an abnormal retention of synapses in C1q-deficient mice. Even though synapse elimination normally occurs shortly after birth in mice, adult mice lacking C1q still had too many synapses. Interestingly, the mice deficient in C1q were able to appropriately eliminate some synapses, indicating that C1q-mediated removal is not the only mechanism by which synapses are eliminated.

Cannibalism at the Synapse?

The researchers hypothesized that C1q flags “incorrect” or unnecessary synapses for destruction in a manner loosely analogous to C1q’s well-established function in the immune system. C1q, whose full name is sub-complement q of complement protein 1, plays an important role in the immune system because it initiates a cascade of events that culminates in the destruction of pathogens, such as invading bacteria. This cascade is known as the complement cascade and comprises a branch of the immune system known as innate immunity because it doesn’t require previous exposure to pathogens to work. C1q circulates in the blood like a sentry and when it detects a pathogen, such as a bacterial cell, it coats the surface of the pathogen, thereby flagging it for destruction. The actual destruction of the pathogen can occur by multiple grisly mechanisms: a Membrane Attack Complex may form which dissolves the invader from the inside out, or specialized immune cells may be recruited to engulf the invader like a cellular Pac-Man gobbling up an enemy. If an analogous process were to occur in the nervous system to eliminate synapses that would radically reshape the current notion of how the brain is sculpted.

To determine if C1q mediates synapse elimination by its known role in the complement cascade, or by some other novel mechanism, the researchers looked for evidence that the complement cascade was activated around synapses. They discovered that other proteins involved later in the cascade, such as C3, are also important for the correct elimination of synapses –suggesting that the complement cascade is indeed involved.

From Development to Disease

These intriguing findings lead the researchers to contemplate a tantalizing question: If C1q and the complement cascade eliminate synapses during normal development might they also be responsible for inappropriately eliminating synapses in diseased brains? It is widely believed that inappropriate synapse loss is one of the first manifestations of neurodegenerative disease. To address this possibility, the researchers looked in the brains of mice with a form of glaucoma. Glaucoma is a leading cause of blindness in the United States and usually occurs when the fluid pressure inside the eyes damages the optic nerve, a bundle of neuronal fibers that connects the back of the eye to neurons deep in the brain. C1q was nearly undetectable in the brains of normal adult mice, but in mice with early or moderate symptoms of glaucoma, C1q re-emerged, and looked strikingly similar to its pattern during development. This suggests that C1q is involved in the early stages of glaucoma, potentially targeting for destruction synapses that would be better left intact.

As with any good research study, this work raises more questions than it answers. How exactly does C1q eliminate synapses? How does it know which synapses to flag? Why doesn’t it flag all synapses? Might other immune molecules outside the complement cascade also play a role? Of course, the researchers are particularly interested in whether C1q plays a role in synapse loss in other nervous system diseases besides glaucoma. The jury is still out on that question, but if C1q were found to be reactivated during neurodegenerative diseases one could imagine ways to block the activity of C1q, potentially sparing the synapses, and presumably the devastating symptoms associated with these diseases.

–Kelly Dakin, Harvard Medical School

For More Information:

Science Daily’s coverage of this research:
< http://www.sciencedaily.com/releases/2007/12/071213121010.htm >

NIH summary of this research:
< http://www.nih.gov/news/pr/dec2007/nida-14.htm >

Primary Literature:

Stevens, B. et al. (2007) The classical complement cascade mediates CNS synapse elimination. Cell. 131(6): 1164-78