Introduction
The pathophysiological processes underlying neuropsychiatric disorders have been unknown; as a result, these disorders have lacked innovative medical therapies with new mechanisms of action. Schizophrenia, for example, is a severe, heritable brain illness affecting almost 1% of the adult population worldwide, at enormous cost to individuals, families, and public health. Many pharmaceutical companies have de-emphasized neuropsychiatric disorders as “too difficult” a challenge to warrant major investment. There is an enormous need for insights about the biological processes that underlie schizophrenia and other neuropsychiatric disorders, as such insights could inform the development of novel approaches to treatment including therapies that could alter target the disease mechanisms.
Recent insights point toward a potentially important role for immune molecules that shape the wiring of brain circuits. Large genome-wide association studies have pointed toward genomic loci that harbor risk variants, with the largest such effect involving schizophrenia’s association with common variants across the Major Histocompatibility Complex (MHC) locus on chromosome 6. We found that this genetic effect arises in substantial part from many structurally diverse alleles of the complement C4A and C4B genes, which associate with schizophrenia in proportion to the C4A expression each allele generates in the brain. We also found that C4 is present at synapses and that C4 shapes synaptic refinement in a mouse model of a postnatal activity- dependent critical period for synaptic rewiring.
These results may connect genetic risk to normal developmental processes of circuit maturation and synaptic pruning. In humans, such maturational processes continue into the third decade of life in distributed association regions of the cerebral cortex, such as the prefrontal cortex. In schizophrenia these same brain regions appear to become functionally impaired and exhibit reduced numbers of synapses. We propose that normal maturational processes can go awry in ways that lead to inappropriate or excessive elimination of synapses, and perhaps, to other kinds of pathophysiology in other adult-onset neuropsychiatric disorders such as bipolar disorder.
This model will require many more forms of evidence to fully evaluate. If correct, though, it would unite known pathologic features of schizophrenia, including reduced numbers of synapses in key cortical regions, loss of cortical gray matter without apparent cell death, and an age of onset that corresponds with developmentally timed waves of synaptic pruning in these regions. Such a model might also point toward novel therapeutic strategies, such as approaches to reduce complement expression or activity in the CNS, modulate the responsiveness of microglia, or dampen the intensity of the changes taking place during critical periods. Increasing appreciation of clinical and biological changes that often precede the development of schizophrenia suggest there may be opportunities for early intervention.
The goal of the work we envision for a Conte Center is to experimentally evaluate this model while also generating and sharing novel scientific resources – including animal models and single-cell expression data sets – that can be used to evaluate current and future hypotheses about schizophrenia-implicated genes, neural-immune interactions, and critical periods for synaptic refinement. We hope in this way to advance the search for molecular understanding and, eventually, novel therapies for schizophrenia. At the same time, we hope to enhance our understanding of brain development, the interacting influences of genes and environment on brain and behavior, and general principles that might be applicable to the mechanisms and pathways that go awry in other mental illnesses.
Project 1
In Conte Center Project 1, we will work to understand how cells of the central nervous system regulate the expression of complement and other immune proteins, and the extent to which this regulation is coupled to function and dysfunction at synapses. We will pursue these questions through experiments on clinical, post mortem, 
and biological samples. We will determine how the expression of C4A and C4B is regulated by allelic variation, cell type, and upstream biology. We will describe at single-cell resolution how diverse CNS cell types respond to synaptic dysfunction, utilizing the Drop-seq technology we developed to analyzed transcriptional responses in tens of thousands of individual cells. Finally, we will carefully evaluate cerebrospinal fluid as a potential reservoir of information about neural-immune interactions in the CNS. Through this work, we hope to better understand how neural-immune interactions play out through gene expression and may become visible in a clinical context.
Project 2
Project 2 will examine how human C4 allelic diversity and expression levels affect microglia-mediated synaptic pruning. This project develops and validates novel mouse genetic tools, including mice that carry human C4A and C4B genes, and mice overexpressing human C4A. Using mice expressing varying levels of human C4A or C4B protein, this work will ask whether C4A has a distinct function from C4B in synaptic pruning. We will also investigate how complement overexpression affects synaptic pruning by examining its roles both in the visual system, where excess C4A may promote excess opsonization of synapses, and in the periphery, where excess complement activity can stimulate changes in the cytokine environment and trigger neuroinflammation. This project will define for the first time whether the two isoforms of human C4 have different functions in the CNS and the impact of complement overexpression on brain development. The C4-humanized and C4A-overexpression mice generated in this project will also contribute to experiments in Projects 1 and 3.
Project 3
Project 3 will reveal the functional consequences of complement-cascade dysregulation, both over- and under-pruning, on circuit function and behavior. This project will study how cortical circuits develop and refine, and how perturbations in synaptic pruning through manipulating schizophrenia risk genes (C4A, CSMD1) impact large-scale cortical connectivity and behavior. We will investigate developmental synaptic pruning in mouse frontal association cortex (FAC) using anatomical, circuit-tracing, electrophysiology, and synaptic engulfment assays established in the Stevens and Sabatini laboratories. We aim to develop the first anatomical and functional maps of synaptic refinement in the frontal association cortex (FAC) of mice, a region involved in executive function and working memory. This work will draw upon mouse models generated in Project 2 and molecular markers for critical periods found in Project 1. Our work will provide new insights into how local and global perturbations of C4 and complement impact local and long-range cortical connectivity and behavior, and whether pruning and behavioral defects are rescued by inhibiting the classical complement cascade.