Calcium wave fronts that cross gap junctions may signal neuronal death during development.

Embryonic anterior pagoda (AP) neurons in the leech interact with their segmental homologs in adjacent ganglia through transient axons that overlap during a critical period of development and then retract. However, when an AP neuron is ablated mechanically or by irradiation during this period, an adjacent homolog responds by reinitiating growth of its overlapped axon and thereby taking over vacated territory (Gao and Macagno, 1987b; Gao, 1989). The death of an AP cell is therefore communicated to its homolog, but the mechanism underlying this signaling is presently unknown. Since it was recently found that AP homologs are electrically and dye coupled through their transient axons (Wolszon et al., 1994), we investigated the possibility that gap junctions may mediate the cell death signal that could occur between developing neurons. Among several candidate intercellular signals, we began by studying calcium dynamics in embryonic AP cells, in situ, since calcium is known to cross gap junctions and is implicated in cell death in many systems. We found that elements that usually increase [Ca2+]i in adult neurons, such as releasable internal stores or voltage-dependent calcium channels, were not present at the critical period. Instead, mechanisms that reduce free calcium, such as buffering and pumping, were the most robust. When a large, focal calcium rise was produced in an AP axon by making a lesion with a UV microbeam (leading to eventual death of these neurons), calcium did not rise quickly throughout the cell, but rather moved in a slow (0.05-0.25 micron/sec) wave front away from the lesion site, into other processes of the damaged cell. Furthermore, when a calcium wave front reached the growth cone of a transient axon, it crossed at the gap junctions into the coupled axon of the neighboring AP neuron, but went no further. Since it is known that an AP responds to a neighbor's death by reinitiating growth only in that axon that contacts the dying cell (Gao and Macagno, 1987b; Gao, 1989), these observations are consistent with calcium playing a role in the signaling of cell death to homologs that are coupled to a dying cell.