The Skaggs Institute
for Chemical Biology 2004

Scientific Report 2004

Fundamental Processes in Neural Development

G.M. Edelman, S. Aschrafi, A. Atkins, S. Chappell, J. Dresios, D.C.Y. Koh, G.W. Rogers, F. Smart, T. Stevens, M. Tsatmali, W. Zhou

Our overall goal is to describe the molecular mechanisms that regulate primary cellular processes during the development of the nervous system in vertebrates. Initially, we focused on the structure, function, and regulation of cell adhesion molecules. Although aspects of these studies are continuing, our efforts have expanded into other areas, including transcription, translation, neural differentiation, and regulation of synaptic plasticity. The training grant from the Skaggs Institute supports the work of several postdoctoral fellows. Their efforts in various projects with senior investigators are summarized here.

Although early studies by members of our laboratory led to a description of the first cell adhesion molecule, N-CAM, in the early 1980s, the mechanism by which N-CAM and most other cell adhesion molecules bind cells has proved elusive. Annette Atkins has used a model system with recombinant proteins attached to beads to provide new insights into N-CAM binding. The main adhesive interaction is mediated by the N-terminal 2 immunoglobulin-like domains, but 4 of the 5 remaining domains are necessary to provide the structural orientation for optimal binding. This finding not only adds new insights into how N-CAM functions in nerve and muscle but also most likely extends to a variety of other adhesion molecules.

The nervous system originates from an undifferentiated sheet of neuroepithelial cells that subsequently differentiate into all of the cells of the system. Neurons and glial cells have a common progenitor, and the mechanism by which these progenitors acquire their differentiated phenotypes is a subject of intense study because of the relevance for designing cellular therapies. Kathryn Crossin and colleagues previously showed that N-CAM plays a role in the maturation of progenitors into physiologically functional neurons. Marina Tsatmali has shown that as early neuronal cells differentiate, they express higher levels of reactive oxygen species than do the progenitors from which the cells originate. She is examining whether the production of reactive oxygen species is a causal factor in differentiation and which signal pathways the reactive oxygen species may influence to guide development.

Development requires multiple rounds of differential gene expression, which are regulated by DNA elements within the genes themselves and by protein factors that bind to the elements. Barx2 is such a factor that regulates genes for cell adhesion molecules and affects a wide variety of differentiation processes. Using chromatin immunoprecipitation and gene expression analysis, Tracy Stevens has identified direct targets of Barx2 in MCF7 breast cancer cells. She found that Barx2 and the estrogen receptor cooperatively regulate the expression of several genes. For example, Barx2 directly regulates the expression of 2 gene families that control the degradation of extracellular matrix: matrix metalloproteinases (MMPs) and tissue inhibitors of MMPs (TIMPs). A precise balance between MMPs and TIMPs is essential for normal mammary gland development, and disruption of this balance contributes to the progression of breast cancer. The finding that the same factors can differentially regulate both MMPs and TIMPs is an important step in understanding how this balance is maintained.

In examining the mRNAs expressed in response to neuronal adhesion, Vince Mauro noted that many of them had sequences that matched or were complementary to ribosomal RNA. By analyzing a subset of these sequences, his colleagues, Stephen Chappell, Wei Zhou, George Rogers, Jr., John Dresios, and Dora Koh, have begun to define characteristics of novel sequences within mRNAs that connect the translation machinery to the message in the absence of the traditional cap sequence. Dr. Chappell has shown that these so-called internal ribosome entry sites (IRESs) are composed of shorter modules that function in isolation. A variety of studies have indicated that some of these IRES elements can directly bind to ribosomes, the translation machinery of the cell, via interactions with ribosomal RNA, a finding buttressed by Dr. Zhou’s experiments in yeast, in which the complementary sequences in the ribosomal RNA were altered.

George Rogers has investigated the translation of the β -secretase mRNA, the overexpression of which is implicated in Alzheimer’s disease. This mRNA is translated by an unusual mechanism in which ribosomes are recruited at the 5´ end of the mRNA and then proceed to the initiation codon by bypassing many of the intervening nucleotides.

Translation of dendritic mRNAs is involved in many functions of the nervous system, and regulation of IRESs may play a role in the functioning of the synapse. Several changes in the efficacy of synaptic transmission occur in a brain region called the hippocampus, and translation of dendritically localized mRNAs is required for each change to persist beyond a few hours. Peter Vanderklish has been studying how translation is regulated by receptor-mediated signaling that occurs during the induction of changes in synaptic efficacy. He and Fiona Smart have obtained evidence that synaptic receptors that trigger different forms of changes in efficacy induce the translation of distinct sets of mRNAs according to the initiation mechanisms the associated signaling events prefer, that is, cap dependent or IRES dependent. In addition, Dr. Smart has shown that a function of local translation is to modify synaptic structure; this observation advances our understanding of synaptic dysfunction in a form of mental retardation known as fragile X syndrome.

The studies described complement work by Bruce Cunningham, Annette Atkins, and Armaz Aschrafi on the composition of neuronal mRNA transport granules. These researchers have shown that the RNA-binding motif protein 3, recently identified by Drs. Vanderklish and Smart as a component of dendritic granules, is associated with ribosomal subunits. Dr. Aschrafi has developed methods for isolating RNA granules, that is, RNA-protein complexes that include ribosomes. He is characterizing the components of these granules and the distribution of specific RNA-binding proteins. Drs. Aschrafi and Vanderklish are also characterizing granules in a mouse model of fragile X syndrome, which is caused by the lack of another mRNA-binding protein found in granules.

The aim of all these activities is the study of the molecular and cellular events that define and regulate the development of the nervous system. We are focusing on fundamental processes rather than on specific diseases. This strategy is based on the notion that understanding even a single primary process can provide the necessary framework for defining the mechanisms that underlie not just one but a variety of diseases.

Publications

Chappell, S.A., Edelman, G.M., Mauro, V.P. Biochemical and functional analysis of a 9-nt RNA sequence that affects translation efficiency in eukaryotic cells. Proc. Natl. Acad. Sci. U. S. A. 101:9590, 2004.

Makarenkova, H.P., Meech, R., Edelman, D.B., Jones, F.S. The homeobox transcription factor Barx2 is expressed during limb development, modulates chondrogenesis and is regulated by GDF5. Development, in press.

Mauro, V.P., Edelman, G.M., Zhou, W. Reevaluation of the conclusion that IRES-activity reported within the 5´ leader of the TIF4631 gene is due to promoter activity. RNA 10:895, 2004.

Rogers, G.W., Jr., Edelman, G.M., Mauro, V.P. Differential utilization of upstream AUGs in the β -secretase mRNA suggests that a shunting mechanism regulates translation. Proc. Natl. Acad. Sci. U. S. A. 101:2794, 2004.

Stevens, T.A., Iacovoni, J.S., Edelman, D.B., Meech, R. Identification of novel binding elements and gene targets for the homeodomain protein Barx2. J. Biol. Chem. 279:14520, 2004.