Amira software

To create the charts and movies, we used the Amira software (version 2020.1, FEI, Merignac Cedex, France) and the Imaris software (version 9.7.2, bitplane, Switzerland) to visualize the dataset. Amira software was used to import the preprocessed dataset using a desktop graphical workstation (7920 with one Intel Xeon Gold 6226R CPU, 512 GB memory and an Nvidia GeForce RTX 3090 graphics card, Dell Inc., Round Rock, Texas, USA). We used the OTSU thresholding technique in conjunction with manually corrected parameters on the red channel cytoarchitecture pictures to reconstruct the bronchus. Amira was used to convert the TIFF data format to the native LDA type so that the TB-sized data could be processed on a single workstation. The extraction of the data in the range of interest, sampling or interpolation, reslicing of the images, determination of the maximum intensity projection, volume and surface rendering, and creation of movies using Amira's main module were all steps in the visualization process [30 (link), 31 (link)].

For the histological analysis, mice (n = 3) were anaesthetized with 3% isoflurane. After the pedal reflex was lost, mice were sacrificed by cervical dislocation. The aortas were dissected from the animals and fixed in 10% buffered formalin for 48 hours. Later they were immersed in 70% ethanol overnight at 4°C, and then embedded in paraffin [11 (link)]. Tissue specimens were transversely sectioned at 5 μm and deparaffinized in xylene and rehydrated in graded ethanol. Elastic and collagen fibers of aortas were stained by use of Accustain® Elastic Stain kit (Sigma-Aldrich, St. Louis, MO) according to the manufacturer’s instructions and as described previously [11 (link)]. The number and percentage area of elastic fibers in the aorta were determined by using Amira software (FEI, France), a commercial three-dimensional (3D) image processing and analysis program. By tracing the elastic fibers for the whole transverse section, pixel size was counted and then converted to the actual size. MATLAB (Mathworks, Natick, MA) was used for calculating the area of elastic fibers of the aorta.

Average sized 5- and 11-month-old sh3pxd2b^+/+ and sh3pxd2b^Δ/Δ fish were euthanised by Tricaine overdose, fixed in 4% PFA as described above, rinsed in dH₂O, and dehydrated through graded water into ethanol (50–75–100% (w/v)) over the course of 3 days at 4°C, after which pure ethanol was refreshed once and fish stored at 4°C until imaging. Microcomputed tomography (µCT) images were acquired as described before (De Vos et al., 2018 (link)). Tissue density was calculated from a Hounsfield unit corrected standard curve. Skeletal 3D renders were generated with AMIRA software (FEI, Mérignac Cedex, France) with constant window settings.

To improve the signal-to-noise ratio of the fMOST imaging data obtained from both the GFP and PI channels, image preprocessing procedures were performed that included image stitching, brightness adjustment, and noise filtering. The preprocessed imaging data were saved for both the PI and GFP channels. We transformed the preprocessed data into cuboid data via TDat 2017 software⁵⁹ . After importing the data into Amira software (version 6.1.1, FEI), we viewed the image stacks and traced the neurite skeletons in an interactive manner using the filament editor^{17 (link)}. Two experienced annotators traced each neuron independently and then compared their reconstructions to produce a final consensus. To make direct comparisons between the morphologies of L2/3 long-range axonal projections labelled by 2-SPARSE and those labelled by the conventional binary AAV expression system, we randomly selected axon branches in contralateral cortices as starting points and performed manual neuronal tracing in retrograde directions, that is, from the contralateral hemispheres to the somata in the injection site, following the same tracing procedures for both groups. For neuronal reconstruction in all other cases, tracing was performed in anterograde directions (from the somata in the injection site to diverse target regions). All traced points for reconstructing neurons were saved in SWC format.

DyLight594-lycopersicon esculentum lectin (1 mg/ml, 0.2 ml per mouse; Vector Labs, United States) was administered via the tail vein 15 days after MCAO. The dye was allowed to circulate for 10 min. Mice were anesthetized and then transcardially perfused with normal saline and 4% PFA in turn. The brains were removed and placed in 4% PFA for 24 h. After embedding in resin (Gong et al., 2016 (link); Gang et al., 2017 (link)), the whole brain specimens were imaged continuously using the fMOST system (BioMapping 5000; OEBio Inc., Wuhan, China). All the image tiles were obtained with a time-delay integration line scan charge-coupled device (Zhou YF. et al., 2019 (link)), and the images had a voxel size of 0.35 μm × 0.35 μm × 1 μm. Afterwards, the original images were preprocessed (Gong et al., 2016 (link)), and visualization and 3D data block contour rendering were performed using Amira software (version 5.2.2; FEI, France). The blood vessels of each data block (size of 400 μm × 400 μm × 400 μm) were tracked semi-automatically, and the VLD and FVV were calculated.

Mice were fixed by cardiac perfusion with 4% paraformaldehyde in 0.1 M phosphate buffer (pH 7.4). Cerebellar sections were post fixed with 2% paraformaldehyde and 2% glutaraldehyde in 0.1 M phosphate buffer (pH 7.4) overnight followed by post fixation with 1% OsO₄, dehydration with a graded series of ethanol and embedding in Epon812 (Oken Shoji). Ultrathin sections were cut with an ultramicrotome UC6 (Leica Microsystems), stained with uranyl acetate and lead citrate and examined with a transmission EM HT7700 (Hitachi). For volume EM, images were taken with a focused ion beam scanning EM (FIB-SEM) Helios Nanolab 660 (FEI Company). Epon blocks were milled by a focused ion beam and a serial block-face for every 20 nm was imaged using a backscattered electron detector (MD detector) at an acceleration voltage of 2.0 kV and a current of 0.4 nA. 3D reconstruction was performed using Amira software (FEI Company).