Microscopy images of the effects of PDZ-RhoGEF manipulation on dendritic spines and videos of effects on PDZ-RhoGEF on mouse behavioral phenotypes

PDZ-RhoGEF is a RhoA-specific guanine nucleotide exchange factor (GEF) for the RhoA small GTPase. The purpose of these experiments is to determine the effects of PDZ-RhoGEF overexpression and knockdown on mouse behavioral phenotypes. Relevant behaviors including assessments of spatial working memory, recognition memory, social behavior, and stress coping. Herpes simplex viral vectors (HSV) were used to overexpress PDZ-RhoGEF. Knockdown of PDZ-RhoGEF was accomplished via packaging of short hairpin RNA (shRNA) into adeno-associated viral vectors (AAV) against Arhgef11 (the gene that encodes the PDZ-RhoGEF protein). Viral vectors were surgically infused directly into the medial prefrontal cortex (mPFC) of young adult male mice. Using the same HSV approach, the effects of PDZ-RhoGEF overexpression on the density and morphology of dendritic spines was also analyzed. Finally, using structured illumination microscopy (SIM), the spatial location of PDZ-RhoGEF and that of one of its direct interacting proteins, DISC1, was examined within individual dendritic spines.

Methods
Object in place behavior videos: Mice were infused with HSV-PDZ-RhoGEF-green fluorescence protein (HSV-PRG-GFP) or HSV-GFP into the medial prefrontal cortex of mice. 5 days post-infusion, mice were assessed in an object-in-place recognition task. During trial 1, mice were allowed to freely explore 4 unique, non-identical objects in an open arena. Mice were then returned to their home cages for 7 minutes and then placed back into the open field arena (trial 2) with the same objects as before, however, the location of two of the objects were swapped in location. This results in two objects whose location is fixed between trials and two objects whose location is altered between trials. The amount of time mice spent directly investigating the swapped and non-swapped objects was done manually using a scientific stopwatch with 0.01 second accuracy. For the knockdown of PDZ-RhoGEF, these same procedures were followed except mice were infused with AAV-Arhgef11 shRNA-GFP or AAV-scrambled control shRNA-GFP into the medial prefrontal cortex and assessed four weeks post-infusion. The videos of this behavior have been uploaded as part of this data set. The raw data obtained by scoring the behavior from these videos was manually logged into spreadsheets that are included as supplementary data in the source publication.

Y-maze spontaneous alternation behavior videos: Mice were pre-tested in the Y-maze five days prior to viral infusion. Mice were allowed free exploration of the three arms of the Y-maze. Five days following the pre-test, mice were infused with either HSV-PDZ-RhoGEF-GFP or HSV-GFP into the medial prefrontal cortex. Five days post-surgery, mice were tested again in the Y-maze (post-test). The arm entries of each mouse were manually scored and the spontaneous alternation percentage calculated as the number of arm triplet entries divided by the number of total possible arm triplet entries. The videos of this behavior have been uploaded as part of this data set. The raw data obtained by scoring the behavior from these videos was manually logged into spreadsheets that are included as supplementary data in the source publication.

Social approach behavior videos: Mice were infused with HSV-PDZ-RhoGEF-GFP or HSV-GFP into the medial prefrontal cortex. Five days post-infusion, mice were placed in the center of a three chamber social apparatus. Each end chamber of the apparatus contained a mesh cylinder. The cylinder in one end chamber was empty and that of the other end chamber contained an aged and sex matched stimulus mouse. The amount of time mice spent directly investigating each cylinder was determined manually using a scientific stopwatch. The videos of this behavior have been uploaded as part of this data set. The raw data obtained by scoring the behavior from these videos was manually logged into spreadsheets that are included as supplementary data in the source publication.

Forced swim test behavior videos: Mice were infused with HSV-PDZ-RhoGEF-GFP or HSV-GFP into the medial prefrontal cortex. Five days post-infusion, mice were individually placed into a beaker of room temperature water. The amount of time mice spent actively swimming while in contact with the water was manually scored using a scientific stopwatch. The videos of this behavior have been uploaded as part of this data set. The raw data obtained by scoring the behavior from these videos was manually logged into spreadsheets that are included as supplementary data in the source publication.

The effects of PDZ-RhoGEF on dendritic spine density and morphology: Mice were infused with HSV-PDZ-RhoGEF-GFP or HSV-GFP into the medial prefrontal cortex. Five days post-infusion, mice were transcardially perfused and the GFP signal from the viral label enhanced using immunohistochemistry. Brains were coronally sectioned at 100um. Neurons were imaged using a Keyence BZ-X700E scanning microscope. Within neurons, we imaged secondary dendrites from the apical tree. Imaging was done using at 100x using a 0.1 µm step size for Z-stack collection. Images for analyses, which are included in this dataset, were collapsed and deconvolved using BZ analyzer software. Dendritic spine analysis, which is included in this dataset, was done using NeuronStudio software.

The localization of endogenous PDZ-RhoGEF within dendritic spines: Day in vitro (DIV) 21-23 (DIV 21-23) cultured neurons derived from embryonic day 18 Sprague-Dawley rat were transfected with GFP and fixed 2 days later (3.7% formaldehyde), and subsequently immunostained for GFP and endogenous PDZ-RhoGEF. Neurons were imaged using a Nikon Structured Illumination microscope using a 100× 1.49 NA oil objective with a laser excitation of 488nm (for GFP), 561nm (for PDZ-RhoGEF). Z stack (z = 0.12μm) images of secondary dendrites on pyramidal neurons were processed and analyzed using Nikon Elements software. Imaging and reconstruction parameters were determined with the support of the Biochemistry Optical Core at the University of Wisconsin-Madison. Acquisition was set to 5MHz at 14bit with EM gain and no binning. Auto exposure was kept consistent between images and within 250-300ms, the EM gain multiplier was kept below 300, and conversion gain was held at 1x. Reconstruction parameters (Illumination Modulation Contrast, High Resolution, Noise Suppression and Out of Focus Blur Suppression) were kept consistent across all images. Dendrites and spines were segmented in the channel of the cell fill (GFP-488nm) using the Imaris surface tool and manually corrected when needed (i.e. splitting two overlapping spines). Object-to object stats were applied, and local background subtraction was used. Thresholding values were kept consistent between images. Spine segmentation was done by manually separating each spine from the dendrite surface using the cut surface tool. This segmentation was duplicated, and the head and neck were manually split in the duplicated segmentation using the same method. Spine classification was done using spine morphology, namely head volume and neck length. Spines with no discernable neck region were assigned as stubby spines, spines with a head volume of 0.4µm3 or larger as mushroom spines, and those with a head volume below this value classified as thin.

Spatial overlap between endogenous PDZ-RhoGEF and overexpressed DISC1: The same procedures were followed as in the above section described for PDZ-RhoGEF. The only difference is that cultured neurons were transfected not only with GFP, but also with DISC1. Immunohistochemisty was then used to label the GFP, endogenous PDZ-RhoGEF, and overexpressed DISC1. The same imaging procedures were used as in the above sections for endogenous PDZ-RhoGEF, with laser excitations of 488nm (for GFP), 561nm (for PDZ-RhoGEF), and 640nm (for DISC1).