Bone marrow microenvironmental defects in jam-b deficient mice. (A) Representative flow cytometry profiles used to define the bone marrow stromal subsets are shown (EC indicates endothelial cell; OB, osteoblast; and MSC, mesenchymal stem cell). Histograms in the left panel show the expression profiles of JAM-B on the indicated subsets. Percentages are indicated on the gates. Bones from 6 mice were pooled and 1 representative experiment of 3 is shown. (B) Confocal images of 20-μm-thick femoral sections from control mice stained with the indicated markers are shown. High expression of JAM-B is found in close association with CD31 positive structures. (C) Confocal images showing a Lin^NegCD48^Neg CD41^NegCD150^Pos cell (HSC, single red cell, arrow) in contact with JAM-B^Pos stromal structure (blue). (D) Representative flow cytometry profiles showing JAM-B and Sca-1 expression by bone marrow stromal cells defined as CD45^NegLin^Neg cells in wild-type littermate and jam-b deficient mice. (E) Representative flow cytometry profiles showing the bone marrow stromal composition of jam-b deficient mice. The percentages of ECs, OBs, and MSCs are shown. There is a 2-fold reduction in the proportions of ECs, OBs, and MSCs compared with control mice shown in panel A. Bones from 6 mice were pooled and one representative experiment is shown. Results from 2 additional independent experiments are provided in supplemental Figure 6. (F) Confocal images of 20-μm-thick femoral sections from jam-b deficient mice stained with the indicated markers are shown. No signal is observed for JAM-B and decreased signal for CD31 is found compared with wild type sections shown in panel B. The graph in the left panel show a quantification of the CD31 staining observed on 8 femoral sections of control littermate (white bar) and jam-b deficient mice (black bar).