More than half of all human proteins are glycosylated, and the physiological significance of glycosylation is exemplified by the numerous instances in which variable glycosylation compromises protein function and causes developmental defects and disease. Despite this, the factors that control which glycans are assembled on proteins are not well understood. Polysialylation is a striking example of a protein specific modification that can dramatically change protein function. Polysialic acid (polySia) is best known for its ability to block neural cell adhesion molecule (NCAM)-dependent cell adhesion and signaling, and for its roles in cell migration, axon guidance, synaptic plasticity, an nervous system development. PolySia is also upregulated on damaged peripheral neurons and facilitates their regeneration, and on the surface of several different types of cancers where it promotes their growth and invasiveness. Remarkably, polySia is found on only five proteins in addition to the polysialyltransferases (polySTs) that modify their own N-glycans. The recent identification of two of these polyST substrates, SynCAM 1, a synaptic adhesion molecule, and neuropilin-2 (NRP-2), a semaphorin and VEGF co-receptor, suggests that the roles of polySia may be more extensive than previously thought, and raises the question of how the polySTs recognize and modify these distinct substrates. Our long-term objectives are to determine the mechanism of protein specific polysialylation, what factors regulate the polymerization of polySia chains on specific substrates, and how polySia modulates the functions of the proteins it modifies. In this proposal we will test the hypothesis that the polySTs recognize common amino acid and structural features of their substrates and that this interaction allows an initial polymerization of the polySia chain on a substrate's glycans, and that this is followed by a polyST-polySia chain interaction that promotes further chain elongation. To do this we will evaluate the domain and sequence requirements for polyST recognition and polysialylation of NCAM, SynCAM 1 and NRP-2, and determine whether residues in a conserved polyST polybasic region mediate substrate protein and/or polySia chain interaction to promote protein specific polySia chain polymerization. We will also test the hypothesis that changes in the length of the stalk regions of SynCAM 1 and NRP-2, generated by alternative splicing, alter their alignment with membrane- associated polySTs and control the polysialylation of these proteins in a cell- and tissue-specific manner. We anticipate that these studies will allow us to identify points in the polysialylation process that are subject to physiological regulation, and that will b amenable to experimental and therapeutic manipulations to control substrate polysialylation and function during development, repair, and disease.