Introduction: Fanconi anemia (FA) is the most common inherited bone marrow failure syndrome. FA patients develop bone marrow failure during the first decade of life due to attrition of hematopoietic stem cells (HSCs). FA is caused by autosomal recessive or X-linked mutations in one of nineteen FANC genes, the products of which cooperate in the FA/BRCA DNA repair pathway and regulate cellular resistance to genotoxic DNA cross-linking agents. Although its mechanism is unknown, bone marrow failure in FA may be the result, directly or indirectly, of hyperactivation of cell-autonomous or microenvironmental growth-suppressive pathways induced, in part, due to genotoxic stress. We have recently identified canonical transforming growth factor-β (TGF-β) pathway-mediated growth suppression of HSCs as a cause of bone marrow failure in FA (Zhang H et al, Cell Stem Cell, 2016). We have shown that TGF-β pathway inhibition rescues genotoxic stress, proliferation defects and engraftment defects of FA-deficient HSCs, and ameliorates bone marrow failure in FA mice. Previous studies have suggested that bone marrow stromal fibroblasts from human FA patients and FA pathway-deficient mouse models, like HSCs, are hypersensitive to genotoxic stress and have impaired growth. Here, we therefore investigated the possible suppressive function of the TGF-β pathway in bone marrow stromal cells derived from FA mice and patients with FA.

Methods: We established primary stromal cell lines from bone marrow of FA-deficient mice (Fancd2-/- mice) or wild-type sibling control mice. Primary bone marrow stromal cultures were also established from FA patients or normal healthy donors. The stromal cells were characterized and evaluated for growth kinetics, mitomycin C (MMC) sensitivity, chromosome breakage, inflammatory signals and response to the TGF-β inhibitors. CRISPR/Cas-9 technology was used to knockdown specific genes in stromal cells.

Results: As expected,the primary bone marrow stromal cells from Fancd2-/- mice exhibited classical FA phenotypes, including hypersensitivity to a DNA cross-linking agent, MMC, and increased MMC-induced chromosomal radials. Fancd2-/- stromal cells also demonstrated a growth defect characterized by an enrichment of cells in G1 and elevated p21 expression. Interestingly, the FA stromal cells derived from FA patients or from Fancd2-/- mice expressed constitutively elevated levels of phosphorylated (activated) ERK1/2 (pERK) compared to control cells. In order to determine whether the factor responsible for inducing ERK1/2 phosphorylation in the murine FA stromal cells was cell-intrinsic or cell-extrinsic, we examined the conditioned media from the stromal cells. Indeed the FA stromal cells secreted a high level of TGF-β cytokine responsible for increased pERK levels, and expressed a high level of secreted TGF-β mRNA. The high level of pERK indicated that the TGF-β non-canonical pathway was hyperactive in the FA stromal cells. Interestingly, CRISPR/Cas9-mediated knockdown of Tgfbr1 or inhibition of the TGF-β pathway by a treatment with a small molecule inhibitor of TGFβR1 or a neutralizing antibody against TGF-β in these cells reduced pERK levels, promoted DNA repair and rescued MMC sensitivity. In addition, a MEK inhibitor also significantly improved the clonogenic growth of Fancd2-/- stromal cells. However, CRISPR/Cas9-mediated knockdown of Smad3, a downstream target of the canonical TGF-β pathway, did not rescue the growth inhibition of FA stromal cells in MMC, further indicating that hyperactivation of the canonical pathway is less relevant to their growth defect. Collectively, these results demonstrated that the hyperactive TGF-β pathway increases phosphorylation of ERK1/2 in FA stromal cells through the non-canonical signaling pathway and impairs their growth after genotoxic stress.

Conclusions: The primary FA bone marrow stromal cells exhibit hyperactive non-canonical TGF-β pathway signaling and blocking this pathway improves their growth under genotoxic stress. The TGF-β signaling pathway-mediated growth suppression in bone marrow stromal cells may account, at least in part, for defective microenvironment, impaired HSC function and bone marrow failure in FA. This work suggests that the TGF-β signaling pathway may be a potential therapeutic target for the treatment of bone marrow failure in FA.


Shimamura:TransCellular Therapeutics: Other: Husband is founder. No revenue to date.; Novartis: Other: In discussion regarding possible clinical trial for aplastic anemia; Glaxo Smith Kline: Honoraria.

Author notes


Asterisk with author names denotes non-ASH members.