The CAM can be roughly classified into 5 morphological subtypes.

A. A 3D reconstruction of the 53 CAMs, color-coded by subtype, is overlaid on a grayscale representation of the 516 reconstructed cells, illustrating their extensive ramification within the SFL tract area (highlighted in green). CAMs display a broad spectrum of morphological subtypes, bifurcating into two primary classes: horizontal cells, encompassing the flag (Fl, in blue) and putative long-range (Lr, in orange) cells, and vertical cells, which include the narrow dendritic (Nw, in red), very narrow dendritic (Vw, in green), and thorny (Th, in pink) cells. Each subtype is spotlighted within a dashed square panel. Notably, several cells remain unclassified (Un, in gray) due to their limited presence within the volume or insufficient observations for categorization. B. Morphological differences among CAMs correlate with variations in synaptic input proportions, with the analysis specifically restricted to inputs within the SFL tract for normalization purposes. All Fisher exact tests with Bonferroni correction, based on the count of SFL inputs per subtype, yielded significant results. A 3D model for each subtype is presented, with postsynaptic input sites (puncta) color-coded based on their presynaptic partner. The postsynaptic site density within the SFL tract area varies significantly among subtypes (Kruskal-Wallis test, p<0.001). A permutation test indicated that flag and thorny cells possess notably higher synaptic densities compared to narrow and very narrow dendritic field cells. Levels of statistical significance are denoted by asterisks: *** p<0.0001, ** p<0.001, * p<0.005.

The 5 morphological subtypes of CAMs display different backbone diameters, suggesting variations in their electrical properties.

A. Reconstructions of CAMs, color-coded by subtype, are overlaid on the SFL tract (green). CAMs exhibit a wide array of morphological subtypes. These bifurcate into two primary classes: horizontal cells, which include the flag (Fl, in blue) and putative long-range (Lr, in orange) cells, and vertical cells, which comprise the narrow dendritic (Nw, in red), very narrow dendritic (Vw, in green), and thorny (Th, in pink) cells. The neuritic backbone significantly varies among subtypes (Kruskal-Wallis test, p<0.001). A permutation test with Bonferroni correction indicated pairwise differences. Levels of statistical significance are denoted by asterisks: *** p<0.0001, ** p<0.001, * p<0.005.

Topological organization of the VL input layer demonstrates distinct synaptic compartments formed by clustering varicosities of similar nature.

A. 3D reconstruction of adjacent varicosities, color-coded by type, within an approximate 15,000 μm3 volume of the SFL tract, revealing three distinct clusters. B. 246 large SFL boutons depicted in varying shades of green. C. 495 small SFL boutons illustrated in different shades of blue. D. A combined representation of 149 CIN (in pink), NM (in brown), and uncharacterized boutons (labeled as UN). E. Low-resolution electron micrograph (EM) of the VL section used for the reconstruction. Highlighted squares (labeled Cube 1-3) pinpoint the locations of the three 125μm3 cubes subjected to saturated reconstruction and synaptic identification. F. Volumetric analysis unveils diverse compositions based on the proportion of neuronal elements present. G, H, and I. Each panel showcases an EM highlighting the predominant synapses of the respective cube (Scale bar = 500nm), a 3D reconstruction of the cube color-coded by cell type, and the connectivity diagram within that cube. Light gray and transparent elements signify minor connections and less prevalent neuronal elements within the cube. Collectively, the three cubes elucidate three unique patterns of synaptic connection areas.

Adult Neurogenic Niche in the VL: Cellular Composition and Organization.

A. Reconstructions of putative neurogenic cells (n=24) in the inner cortex, highlighting neural progenitor cells (purple) and immature neurons (gray), representing a continuum of differentiation. B. SAMs reconstructions (n=43) color-coded by synaptic output density (syn out per 100um normalized within the SFL tract) positioned above the neurogenic niche and adjacent to radial glia. The bottom-left plot shows the significant Pearson correlation between cell body position and synaptic output density, indicating SAM maturation. C. Schematic representation of the VL’s adult neurogenic niche organization and composition.

Synaptic and Cellular Blueprint of the Vertical Lobe.

This schematic captures the intricate organization of the Vertical Lobe (VL) along with its wiring diagram (modified from Bidel, Meirovitch et al., 2023). Two primary afferents, the Superior Frontal Lobe Axons (SFL) and the Complex Input Neurons (CIN), convey secondary visual cues and potential “pain” signals, respectively. They innervate two distinct amacrine interneuron populations within the VL’s input layer, the SFL tract. This layer shows a meticulous organization, with like boutons clustering together, forming distinct synaptic glomeruli clusters. The Simple Amacrines (SAMs), constituting the majority at around 23 million, stand out with their singular input and simplicity in morphology and connectivity. In contrast, the Complex Amacrines (CAMs) represent only 400 000 cells of the cortex, and exhibit a rich morphological diversity, with various subtypes each having unique input densities and fractions. A potential neurogenic niche in the inner cortex suggests the capability for continuous integration of new amacrines throughout the octopus’s lifespan. These interneurons converge onto the Large Neuron Processes (LN), the sole VL output, within the neuropile. The circuit is further enhanced by a vast modulatory network, including the Ascending Fibers (AF) and the Neuromodulatory Fibers (NM).