Labeling and confocal imaging of neurons in thick invertebrate tissue samples

Neuroscience researchers have long sought methods to describe the neural connectivity of the circuits responsible for specific behaviors. One major obstacle is scale: Neural spines can be <1 μm in diameter, but axons can range from millimeters to centimeters (or larger) in length, making tissue imaging and neuron reconstruction a challenging task. New tissue-clearing agents and long-working-distance objectives offer improved imaging conditions, and here we present a complete protocol for invertebrate tissue that uses these advances. In this protocol, tissue-processing steps previously published in separate articles are combined with recent advances in confocal imaging to visualize invertebrate tissue samples that are >500 μm thick and contain dye-filled neurons. The steps describe dye filling, fixing, antibody labeling, clearing, whole tissue mounting, and confocal imaging with matched refractive indexes. Thus, manual sectioning or "flipping" the tissue to image the whole volume is not required. With matched refractive indexes, loss of resolution and signal is avoided. Tissue volumes are imaged in one stack and nonlinear deformations caused by tissue flipping are prevented. We apply the protocol to whole dragonfly thoracic ganglia (2 × 1 × 0.6 mm) and cephalopod skin samples (20 × 2 × 0.6 mm) with minimal tissue deformation. The resulting images will be used to develop a three-dimensional connectivity atlas of dragonfly ganglia and cephalopod skin innervation. This protocol can be applied to other invertebrate species, and has the advantage that it avoids problems with antigen specificity.