Rac deletion in Nestin⁺ cells disturbs the vascular morphology in the medullary cavity. (A) Snapshot images of 3D reconstructed whole-mount sections showing BM microvasculature in the endosteal region of the femoral diaphysis from each genotype after staining for laminin (green, arteries and sinusoids), endoglin (red, sinusoids), and Sca-1 (blue, arteries) as detailed in Material and methods. (B) Sinusoidal volume and (C) arteriolar volume quantification of diaphysis from snapshot images of 3D-reconstructed whole-mount sections (as shown in [A]) was determined using Volocity 3D imaging software. Data represent mean ± standard error of the mean (SEM), N = 4 animals per group, *P < .05, 4-5 individual sections analyzed per mouse. (D) Changes in sinusoidal architecture as assessed LSC of whole longitudinal image of femoral cryosection stained with DAPI (nuclei) showing proximal metaphysis (PM), diaphysis (DIA), and distal metaphysis (DM). Representative LSC images (middle panels) of BM stained for endoglin (red), laminin (green), and DAPI (blue) of TG WT (top panel) and TG Rac1 ^Δ/ΔRac3^−/− mice (bottom panel). The middle panel shows high-resolution images of the indicated field. The right panel show images of contoured niche signals of indicated fields. Analysis of endoglin⁺ signal events as detailed in Material and methods, based on fluorescence intensity (as shown in supplemental Figure 2A-B) and detected by the iCys software (teal marks). (E) Quantitative analysis of the frequency of sinusoidal distribution by diameter of BM vessels of images in (D, middle panels) determined using iCys software, in a blinded manner. Data represent mean ± SEM, N = 3 animals per group, *P < .05, **P < .01. (F) Vascular sinusoidal niche area quantification of diaphysis from photographs in (D, right panel) was assessed using iCys software. Data represent mean ± SEM, N = 3 animals per group, *P < .05.