Formation of vascular networks in vivo with from LBW or CT-ECFCs neonates. A mixture of ECFCs and SMC (ratio 75:25) was resuspended in growth factor-reduced Matrigel and implanted on the groin area of 5-week nu/nu mice. Implants were harvested after 7 days and stained with hematoxylin and eosin or labeled by fluorescent staining as described in “Methods.” (A) Representative serial sections of Matrigel plugs filled with LBW-ECFCs (left panel) or CT-ECFCs (right panel): (i,v) Matrigel plugs recovered from nu/ν mice. H&E staining showing vessel-like structures in plugs filled with: (ii-iv) LBW-ECFCs and (vi-vii) NBW-ECFCs. Arrows indicate the presence of erythrocytes. (B) Quantification of large microvessel density per square millimiter. (C) Counts of erythrocyte-filled vessels per square millimiter. Results represent mean ± SEM from 9 fields in 2 separate sections. ***P < .001. (D) Representative photomicrographs of luminal microvessels in Matrigel plugs filled with: (i) LBW-ECFCs, (ii) CT-ECFCs, and (iii) unfilled Matrigel plug (left). Quantification of luminal structures per square millimiter. Results represent mean ± SEM from 9 fields in 3 separate sections. ***P < .005. (E) Immunofluorescent staining of: endothelial cells (i,v); with anti–human CD31 (green represents SMC) (ii,vi); with anti–human α-smooth muscle actin (red); merged image of CD31 and smooth muscle actin immunostaining (iii,vii); and 4,6-diamidino-2-phenylindole-stained nuclei (iv,viii; blue). Images were representative of implants into 2 mice (original magnification ×40). Scale bars represent 20 μm.