The foundations of modern neuroscience rest in the classical histological studies of Ramón y Cajal and other early anatomists. These studies established that there are at least several hundreds of distinct classes of neurons and glia in the mammalian brain. We now know that these cell types are arranged into complex circuits, and that these circuits control specific aspects of cognition and behavior.
Although it was evident from these pioneering studies that the morphology of specific neuronal cell types varies dramatically and hence must reflect their distinct functions in the nervous system, the ability to perform detailed studies to characterize specific cell types and understand the biochemical basis of this diversity eluded scientists for nearly a century. To address this problem, our laboratory has developed a suite of novel approaches based on the manipulation of bacterial artificial chromosomes (BACs) for the investigation of genes, cells, and circuits in vivo. We are interested in employing these methodologies to investigate molecular mechanisms that contribute to the histological and functional complexities of the mammalian brain, and to understand how these mechanisms become dysfunctional in disease.
Genetic Dissection of CNS Cell Types and Circuits
To identify the cellular and molecular mechanisms that are essential for the functions of the mammalian brain—and to understand how these processes go awry in disease—requires the ability to reproducibly genetically target defined central nervous system (CNS) cell types. This can be achieved using large DNA constructs that carry all of the regulatory information required for accurate expression in vivo (e.g., BACs). Our laboratory invented DNA engineering by homologous recombination in Escherichia coli (a process now commonly referred to as "recombineering") and demonstrated that engineered BAC transgenes can be accurately expressed in vivo. Thus, BAC transgenic technology provides a genetic solution for detailed studies of the properties of the CNS cell types first visualized by Ramón y Cajal a century ago.
As a first step toward characterization of all cell types in the mouse central nervous system, our laboratory, in collaboration with Mary Hatten (Rockefeller University), launched the Gene Expression Nervous System Atlas (GENSAT) project, supported by the National Institute of Neurological Disorders and Stroke (NINDS GENSAT is a large-scale screen using BAC transgenic mice to create an atlas of CNS gene expression at the cellular level. It provides detailed anatomical data on cell types targeted in more than 1,200 BAC transgenic mouse lines and provides a library of verified BAC vectors and transgenic mouse lines to the community that offer experimental access to CNS regions, cell classes, and pathways. GENSAT serves as the foundation for many of the studies currently pursued in our laboratory, including those focusing on biochemical and epigenetic mechanisms of neuronal function, because it allows investigation of the detailed properties of specific cell types in vivo.
Translational Profiling of CNS Cell Types in Health and Disease
One of the central conundrums of neuroscience is the observation that despite the distinct clinical presentations of human neurological and psychiatric disorders, the vast majority of genes contributing to these disorders are expressed widely or ubiquitously in the brain. We believe the specific consequences of these genetic events must reflect their differential impact on the finely tuned biochemical pathways controlling the activities of neurons and circuits most impacted in a given disorder. To investigate this issue, our laboratory, in collaboration with the laboratory of Paul Greengard (Rockefeller University), developed the translating ribosome affinity purification (TRAP) approach.
The TRAP approach rests on two fundamental facts: all proteins in all cell types are produced from mRNA by ribosomes through the process of translation, and exogenous proteins can be expressed in cells of interest using either BAC transgenic techniques (see above) or gene targeting. The TRAP approach takes advantage of these facts to identify translating mRNAs from genetically targeted cell types. It employs an affinity tag (EGFP, enhanced green fluorescent protein) fused to a ribosomal protein to allow isolation of bound mRNAs from a targeted cell type, without requiring isolation of that cell type from the tissue of interest.
Our laboratory is using bacTRAP transgenic mice and TRAP profiling to determine the molecular constitutions of a wide variety of classically defined and newly discovered cell types in the mouse brain. We are focusing primarily on molecular definition of neuronal populations occurring in the cerebral cortex and those present in circuits most impacted in common neurological and psychiatric disorders. These studies have revealed that an unprecedented level of molecular complexity is present in specific CNS cell types and that the response of closely related cell types to genetic or pharmacological perturbations can vary substantially. We are currently using TRAP "molecular phenotyping" in mouse models of autism-spectrum disorders, amyotrophic lateral sclerosis, addiction, and depression to identify cell types and biochemical pathways whose altered activity contributes to the pathophysiology of these disorders.
Epigenetic Regulation of the Neuronal Genome: The Role of 5-Hydroxymethylcytosine in Neurons
From the earliest days of light microscopy, it has been apparent that the structure of the nucleus varies tremendously between cell types. Ramón y Cajal first noted these differences in neuronal nuclei, providing illustrations that neuronal nuclear size and shape vary dramatically and identifying a nuclear structure capping the nucleolus that still bears his name (the Cajal body). With the advent of electron microscopy, examination of neuronal nuclei reached a new level of sophistication, resulting in the observation that large neuronal nuclei contain very little heterochromatin.
Over the past several decades, a strong connection between the presence of 5-methylcytosine (mC), chromatin organization, and gene expression has been established. While investigating the relationship between nuclear structure and the content of mC in cerebellar Purkinje and granule genomic DNA, we discovered that 5-hydroxymethylcytosine (hmC), is present in the mammalian genome and that it is specifically enriched in neurons. At the same time, the presence of low levels of hmC in mouse embryonic stem cells and the discovery of an enzyme that could convert mC to hmC in mammalian DNA were reported. Together, these two studies identified a novel epigenetic mark on the mammalian genome that had not previously been observed in metazoans. The discovery that hmC is specifically enriched in neuronal genomes has opened a new field of investigation into the significance of this finding for neuronal function. We are now addressing its potential impact on nuclear structure and gene expression, its significance for epigenetic mechanisms of neurological and psychiatric disease, and the contributions of hmC to CNS development.
Biochemical Mechanisms of Neuronal Function
The specialized functions of neurons rely on complex biochemical mechanisms that often involve large macromolecular assemblies. To understand the roles of genes identified in a variety of early screens conducted in our laboratory, we must characterize their encoded products in their native biochemical context. The strategy we have used to accomplish this is to use BAC transgenic mice to express chosen fusion proteins in cell types of interest, to biochemically prepare crude fractions that are enriched for the complexes of interest, to affinity purify from that mix those complexes incorporating the proteins being studied, and to use mass spectroscopy to identify interacting partners (in collaboration with Brian Chait, Rockefeller University). Using this approach, we have begun to characterize the biochemical properties of specific synapse types in the mammalian brain and to identify complexes containing Beclin1 that are involved in autophagy and endocytosis. In each of these cases, new cell-type-specific proteins regulating these processes have been discovered, and novel insights into cell biological mechanisms regulating neuronal function have been obtained. It is evident from these studies that in vivo biochemical profiling provides an important avenue toward deciphering the complex phenotypes encountered using conventional genetic perturbations to understand the gene's function.
Grants from the National Institutes of Health and the Simons Foundation provided partial support for these projects.
As of March 3, 2016