Introduction: Bone marrow progenitor cells from Fanconi Anemia (FA) patients have a hyperactive TGF-β signaling pathway which may explain hematopoietic stem cell depletion leading to bone marrow failure (Zhang, et al, Cell Stem Cell, 18:668-681, 2016). To determine whether blockade of TGF-B signaling by knockout of Smad3 normalized hematopoiesis in Fancd2-/- mice, we bred two double knockout mouse strains. The first (DKO#1) was derived by crossing C57BL/6 Fancd2 +/- mice with 129/Sv Smad3 +/- mice. The second (DKO#2) was derived by mating 129/Sv Fancd2 +/- mice with the same 129/Sv Smad3 +/- mice.

Materials and Methods: Long term bone marrow cultures (LTBMCs) were established from DKO#1, DKO#2, parental mouse strains and F1 (129/Sv X C57Bl/6) control mice for DKO#1, 129/Sv Smad3 -/-, and C57BL/6 Smad3 -/- mice . The cultures were scored for weekly numbers of cobblestone islands as an indicator of stem cells in the adherent layer, weekly production of nonadherent cells, and cells producing Day 7 and Day 14 CFU-GEMM .Bone marrow stromal cell lines were derived from the adherent layer of LTBMCs and radiosensitivity measured in clonogenic radiation survival curves. Cells were irradiated to doses of 0 to 8 Gy, plated in 4 well linbro plates, incubated for 7 days at 37oC, stained with crystal violet and colonies of greater than 50 cells counted. Western analysis of the cell lines quantitated expression levels of proteins involved in the TGF-β signaling pathway, and DNA double strand break repair by homologous recombination and nonhomologous end-joining.

Results: Both DKO1 and DKO2 LTBMCs showed decreased duration and magnitude of hematopoiesis, thus being similar to that observed with both C57BL/6 Fancd2-/- and 129/Sv Fancd2-/- mouse marrow cultures. Both were significantly lower than that for control mouse or Smad3-/- (129/Sv) marrow cultures. As expected, radiation survival curves showed that both C57Bl/6 Fancd2-/- and 129/Sv Fancd2-/- marrow stromal cell lines were radiosensitive compared to control cell lines including: F1 control, C57BL/6 control, and 129/Sv control (Berhane, et al, Rad Res. 181:76-89, 2014, Berhane et.al, Rad Res 182:35-49, 2014). In contrast marrow stromal cell lines from Smad3-/- (129/Sv) marrow cultures were radioresistant (Epperly, et al, Rad Res 165:671-677, 2006). Fresh marrow CFU-GEMM from both DKO#1 and DKO#2 mice showed resistance to abrogation of colony formation by increasing concentrations of TGF-B, (similar to the 129/Sv Smad3-/- cell line). In contrast, cell lines from all controls and both Fancd2-/- mice showed clear TGF-B mediated inhibition of hemopoietic colony formation. In striking contrast to the above similarities between DKO#1 and DKO#2 mice, marrow stromal cell lines from DKO#1 were radiosensitive (like theirFancd2-/- parent) while those from DKO#2 were radioresistant (like their Smad3-/- parent). Thus, DKO#1 retained the C57Bl/6 Fancd2-/- genotype cell line radiosensitivity: Do of 1.41 ± 0.03 Gy and 1.45 ± 0.05 Gy respectively, and were more radiosensitive than the control F1 bone marrow stromal cell line (p = 0.0230 and 0.0418, respectively) and DKO#2 retained the 129/Sv Smad3-/- cell line genotype radioresistance (Do = 2.15 ± 0.13 Gy) compared to control 129/Sv stromal cells (Do = 1.86 ± 0.04, p = 0.0054). Western analysis revealed that p21 was elevated in DKO#2 but not DKO#1 marrow stromal cell lines.

Conclusions: While marrow from both DKO#1 and DKO#2 mice showed resistance to TGF-B signaling consistent with their smad3-/- genotype, only DKO#1 stromal cells retained the radiosensitivity of their Fancd2-/- genotype. Reduced p21 in irradiated DKO#1 marrow stromal cells may have allowed procession through G to S phase causing reduced time for DNA repair and radiosensitivity. The irradiated DKO#2 cells may have been blocked by p21 from passing through the G1 checkpoint and may have allowed DNA strand break repair and radioresistance. These DKO mice and derived cell lines should be valuable for analysis of the interaction of TGF-B signaling and FA pathways.

Supported by NIAID/NIH U19-AI068021


No relevant conflicts of interest to declare.

Author notes


Asterisk with author names denotes non-ASH members.