No overproportional release of CFU-Cs from the BM into the PB. (A) Single-cell suspensions were made from BM and spleen, and erythrocyte-depleted blood cells (PB) from control and β1β7 mutant BM chimeras were prepared at indicated times after the gene deletion. BM cells (180 000), splenocytes (3 600 000), and PB (250 μL) were seeded into MethoCult GF M3534 medium and counted 7 days later. Total numbers of colonies per femur, spleen, and mL PB are shown. Error bars show the standard deviation, star indicates significant difference (P < .05) (n (control BM chimera)/(β1β7 mutant BM chimera): 2 months 5/5, 10 months 3/3). (B) Single-cell suspensions from BM of nonchimeric control (β1^fl/flMx-cre^–) and β1^fl/flMx-cre⁺ mutant mice were plated on tissue-culture plastic dishes. After 24 hours, nonadherent cells were removed and adherent cells detached. Adherent and nonadherent cells were then stained for Ly-5.2 and Ter119 and tested for β-galactosidase activity by an FDG assay as described in “Materials and methods.” Since loss of β1 integrin results in expression of the β-galactosidase reporter, 12 high β-galactosidase activity indicates deletion of the β1 gene. Representative histogram overlays show the β-galactosidase activity on hematopoietic (Ly-5.2⁺ or Ter119⁺) and nonhematopoietic (Ly-5.2^–, Ter119^–) cells of control (filled line) and mutant mice (line). The marked region on the overlay indicates cells with high β-galactosidase activity. PolyIC injection induced an efficient deletion of the β1 integrin gene on hematopoietic cells; about 86% of the (Ly-5.2⁺ or Ter119⁺) cells of the mutant mice showed high green fluorescence, compared with only 6% of the corresponding cells of the control mice. Also, among the nonhematopoietic BM cells (Ly-5.2^–, Ter119^–) the percentage of cells with high β-galactosidase activity increased from less than 5% in control to more than 42% in mutant, clearly indicating the presence of β1 integrin–deficient nonhematopoietic cells in the BM of mutant mice.