Intracellular Ca2+ signals that last more than a few minutes after the onset of stimulation depend critically on influx of extracellular Ca2+. Such Ca2+ influx can be triggered in many cell types by depletion of intracellular Ca2+ stores without detectable elevations of known messengers. The mechanism by which store depletion can control plasma membrane Ca2+ permeability remains controversial. Here we present evidence for a novel soluble mediator. Calcium depletion of a lymphocyte cell line caused the messenger to be released from intracellular organelles into the cytoplasm and to a much lesser extent into the extracellular medium. The messenger caused Ca2+ influx when applied to macrophages, astrocytoma cells, and fibroblasts and was therefore named CIF (for Ca(2+)-influx factor). CIF appears to have hydroxyls (or hydroxyl and amino groups) on adjacent carbons, a phosphate, and a M(r) under 500.

The structure of immunological synapses formed between murine naive T cells and mature dendritic cells has been subjected to a quantitative analysis. Immunofluorescence images of synapses formed in the absence of antigen show a diffuse synaptic accumulation of CD3 and LFA-1. In electron microscopy, these antigen-free synapses present a number of tight appositions (cleft size *15 nm), all along the synapse. These tight appositions cover a significantly larger surface fraction of antigen-dependent synapses. In immunofluorescence, antigen-dependent synapses show multiple patches of CD3 and LFA-1 with a variable overlap. A similar distribution is observed for PKCh and talin. A concentric organization characteristic of prototypical synapses is rarely observed, even when dendritic cells are paralyzed by cytoskeletal poisons. In T-DC synapses, the interaction surface is composed of several tens of submicronic contact spots, with no large-scale segregation of CD3 and LFA-1. As a comparison, in T-B synapses, a central cluster of CD3 is frequently observed by immunofluorescence, and electron microscopy reveals a central tight apposition. Our data show that it is inappropriate to consider the concentric structure as a "mature synapse" and multifocal structures as immature.