The Synaptic Complexity of a High-Integration Lobula Giant Neuron in Crabs

Arthropods are diverse, abundant, successful animals that exploit all available ecological niches. They sense the environment, move, interact with prey/predators/conspecifics, learn, and so forth using small brains with five orders of magnitude less neurons than mammals. Hence, these brains need to be efficient in information processing. One distinct aspect is the presence of large, easily identifiable single neurons that act as functional units for information processing integrating a high volume of information from different sources to guide behavior. To understand the synaptic organization behind these high-integration nodes research on suitable neurons is needed. The lobula giant neurons (LGs) found in the third optic neuropil, the lobula, of semiterrestrial crabs Neohelice granulata respond to moving stimuli, integrate information from both eyes, and show short- and long-term plasticity. They are thought to be key elements in the visuomotor transformation guiding escape responses to approaching objects. One subgroup, the MLG1 (monostratified LG type 1), is composed of 16 elements that have very wide main branches and a regular arrangement in a deep layer of the lobula which allows their identification even in unstained preparations. Here, we describe the types and abundance of synaptic contacts involving MLG1 profiles using transmission electron microscopy (TEM). We found an unexpected diversity of synaptic motifs and an apparent compartmentalization of the dendritic arbor in two domains where MLG1s act predominantly as presynaptic or postsynaptic, respectively. We propose that the variety of contact types found in the dendritic arbor of the MLG1s reflects the multiple circuits in which these cells are involved. Regarding the detection of approaching objects, the distinctive input contact motifs shared by lobula giant neurons in crabs and locusts suggest a similar organization of the collision-detecting pathways in both species.