NeuronalConnectivity

Acapella® (PerkinElmer) script to analyse images of primary neuronal cultures, acquired on an Opera Phenix High Content Screening System (PerkinElmer) using a 40x water immersion lens (numerical aperture 1.1). Cultures are immunochemically labeled for a nuclear marker (DAPI), dendrite marker (MAP2) and a pre- and postsynaptic marker (e.g. synaptophysin and PSD-95). Images are read in per field of view. After maximum projection of the z-stacks obtained from the MAP2 and DAPI channel, the nuclei are detected using a manually assigned threshold. Dendrites are identified using a rough (user-defined threshold) and fine (user-defined threshold after Frangi filtering) segmentation. Neuronal nuclei are distinguished from non-neuronal based on a user-defined maximal projected area, minimal circularity and minimal occupancy in the dendrite mask. For both the dendrite network and nuclei a range of morphological and textural (object- and image-based) descriptors are extracted. Next, the dendrite mask and the neuronal nuclei mask are dilated and subtracted from each other to obtain a search region (i.e., dilated dendrites without neuronal nuclei) in which the pre- and postsynaptic spots are detected. The sharpest slice (based on the highest standard deviation of the intensity) from the presynaptic channel and its matched postsynaptic slice are used for spots detection. The spots are first enhanced using a difference of Gaussian filter with a user-defined kernel size, after which a user-defined threshold was applied to segment the spots. To minimize noise contributions, only spots larger than 4 pixels are retained for further analysis. The resolution of the microscope setup does not allow determining the exact location of individual markers within a synapse, but this is not the intention of the assay. Instead, the lower resolution is exploited to define synapses as those objects that demonstrate an overlap of minimum 1 pixel between the pre- and postsynaptic spots. In addition to this object-based colocalization, the Pearson correlation of the pre- and postsynaptic channel is calculated as an intensity-based colocalization metric. Next to morphological and textural descriptors, the density of pre- and postsynaptic spots and synapses are calculated.