Our studies focus on the cellular and molecular mechanisms underlying the specification of neuronal identity, growth cone motility and navigation, target selection and the generation of neuronal arbors. We work with identified neurons in the medicinal leech, Hirudo medicinalis, a system that has several advantages for our studies. First, we can study a particular neuron at all stages of development, in the animal; second, we can visualize many cells in the intact animal as they differentiate, using time-lapse 2-photon confocal imaging, and operate on them with a laser microbeam to ask very specific questions; and third, we can inject macromolecules into embryonic or adult neurons, which means that we can transform a single cell in a wild-type environment.

In current cellular studies, we are examining the formation of peripheral arbors by sensory and motor projections in the animal's bodywall and in specific target tissues, like the heart tubes. Time-lapse observations in situ show that arbors are highly dynamic, with higher-order branches undergoing repeated cycles of extension and retraction. We have also identified several interactions (e.g., between homologous neurons, between pioneers and their followers, between axons and non-neuronal substrates, between different branches of the same neuron, between neurons and their targets) that play roles in defining the size and shape of the arbor. One experiment underway explores how a cell recognizes itself by examining what happens after a branch has been severed by laser photoablation - the result is that the cell stops recognizing the distal stump as self, suggesting that surface markers are not responsible for self recognition. To get some incling of how back-branching might be induced by interactions with substrates or targets, in another set of experiments we are beginning to map the disposition and dynamics of cytoskeletal proteins at branch points forming behind growth cones.

Among current molecular studies, we are continuing the characterization of a family of transcription factors that are thought to be responsible for the specification of neuronal phenotype. These are genes of the Hox/HOM homeobox family, which we were the first to clone and characterize in the leech (e.g., Lox1, Lox2, Lox3, Lox4, Lox6, Lox 15). They are expressed in overlapping sets in the leech nervous system. We have identified several neurons that express them, and have found in some cases that changes or differences in expression correlate with phenotypic differences among segmental homologues. These differences include changes in arborization patterns and peripheral targets, tying these molecular studies to the cellular studies summarized above. On-going efforts to affect expression will test whether these correlations signify causal relationships. An initial result of expressing Lox1 ectopically by injecting its mRNA in adult neurons is a reproducible change in the electrical properties of the injected cells.

In a complementary set of molecular studies, we have begun to examine the functions of membrane-bound and secreted recognition/adhesion factors that are thought to be involved in the cell interactions we have documented to occur during arbor formation and target selection. We have recently cloned the leech homologues, HmLAR1 and HmLAR2, of the LAR family of receptor phosphatases and have determined that they are expressed by identified neurons (in processes and growth cones) as well as some muscle and other types of cells. Blocking function by injecting into embryos antibodies to the extracellular domain or recombinantly expressed extracellular domains, or using RNAi in tissues or individual cells to knock down expression of these receptors, shows that HmLAR1, which is expressed by the heart tubes, is necessary for the innervation of the heart, and that HmLAR2 is involved in the regulation of axonal pathfinding growth cone stability and process outgrowth. In addition, we have characterized a leech netrin, which is also expressed by specific central neurons as well as ventral (but not dorsal) longitudinal muscles. Initial results of injecting leech netrin antibodies into live embryos suggest that afferent projections may use a netrin gradient to grow towards the central nervous. We are presently beginning to examine the formation of peripheral neuronal arbors in the presence of these blocking antibodies.

Research in this laboratory is supported by grants from the National Science Foundation and the National Institutes of Health.