Fucose-deficient hematopoietic stem cells have decreased self-renewal and aberrant marrow niche occupancy

Abstract

Background: Modification of Notch receptors by O-linked fucose and its further elongation by the Fringe family of glycosyltransferase has been shown to be important for Notch signaling activation. Our recent studies disclose a myeloproliferative phenotype, hematopoietic stem cell (HSC) dysfunction, and abnormal Notch signaling in mice deficient in FX, which is required for fucosylation of a number of proteins including Notch. The purpose of this study was to assess the self-renewal and stem cell niche features of fucose-deficient HSCs.

Study design and methods: Homeostasis and maintenance of HSCs derived from FX(-/-) mice were studied by serial bone marrow transplantation, homing assay, and cell cycle analysis. Two-photon intravital microscopy was performed to visualize and compare the in vivo marrow niche occupancy by fucose-deficient and wild-type (WT) HSCs.

Results: Marrow progenitors from FX(-/-) mice had mild homing defects that could be partially prevented by exogenous fucose supplementation. Fucose-deficient HSCs from FX(-/-) mice displayed decreased self-renewal capability compared with the WT controls. This is accompanied with their increased cell cycling activity and suppressed Notch ligand binding. When tracked in vivo by two-photon intravital imaging, the fucose-deficient HSCs were found localized farther from the endosteum of the calvarium marrow than the WT HSCs.

Conclusions: The current reported aberrant niche occupancy by HSCs from FX(-/-) mice, in the context of a faulty blood lineage homeostasis and HSC dysfunction in mice expressing Notch receptors deficient in O-fucosylation, suggests that fucosylation-modified Notch receptor may represent a novel extrinsic regulator for HSC engraftment and HSC niche maintenance.

Figure 1. Mild homing defects associated with FX^-/- marrow progenitors are corrected by exogenous fucose

Lethally irradiated WT recipients were injected with bone marrow cells derived from WT mice, mice raised on standard chow (FX^-/-, - fucose), or mice raised on fucose-supplemented chow (FX^-/-, + fucose). (A) CFUs were determined from the recipient bone marrow 16 h after injection, and expressed as percentage recovery of infused CFU-Cs (n=5 per group). (B) Proportions of different lineages of CFUs were determined from the recovered marrow progenitors, including Burst-forming unit-erythroid (BFU-E), myeloid lineage colonies of granulocyte/macrophage/granulocyte macrophage (CFU-G/M/GM), and multi-potential progenitors CFU-GEMM (CFU-granulocyte, erythroid, macrophage, megakaryocyte).

Figure 2. FX^-/- HSCs have decreased self-renewal in serial trasnplantation, are less quiescent, and have increased cell cycling activity

Flow cytometry analysis of LSK cells in the primary recipients 4 months after transplantation (A) and secondary recipients 3 months after transplantation (B). N=5 for each group, *p < 0.05, ** p< 0.01 compared with primary (A) or secondary (B) transplant WT recipients receiving WT donor cells. (C) Shown is a representative distribution of G₀ versus G₁ in the LSKs defining an increased cycling fraction in FX^-/- mice. Mouse bone marrow cells were stained with lineage antibodies, APC-anti-ckit, PE-anti-Sca-1, Pyronin Y (RNA dye), and Hoechst 33342 (DNA dye). LSK cells were gated by means of a stringent parameter. Cells residing in G₀ appear at the bottom of the G₀/G₁ peak, and G₁ cells are the upper part as indicated. (D) Mice were injected intraperitoneally with one dose of Brdu and fed with water containing Brdu for 3 days. The percent of LSKs in S-G₂/M phase of the cell cycle was analyzed by anti-Brdu and 7-amino-actinomycin D (7-AAD). Data shown in (C) and (D) is one representative of 3 similar experiments.

Figure 3. FX^-/- HSCs have decreased Notch ligand binding but similar physical relationship to the vasculature compared to WT HSCs after homing in mouse calvarium bone marrow

(A) Marrow LSK cells from WT-NTg or FX^-/--NTg mice without fucose supplementation were examined for GFP expression. GFP+ cells were gated and analyzed for their binding with vector control or recombinant Notch ligand Dll4 using PE-anti-hIgG Fc by flow cytometric analysis. Data shown is one representative of five similar experiments as indicated by the mean fluorescent intensity (MFI) after binding of LSKs with vector or Dll4. (B) Shown are 2 representative images of each taken by 2-photon intravital microscopy of engrafted GFP+ marrow progenitors isolated from either WT-NTg (left 2 panels) or FX^-/--NTg (- fucose) (right 2 panels) mouse donors, respectively. Calvarium blood vessels were highlighted by TRITC-Dextran.

Figure 4. Marrow HSCs without fucose are localized more distant to the endosteum in the calvarium

(A) A representative image of 3 similar experiments showing HSCs (GFP+, green) localized in the marrow after transplantation. The shortest three-dimensional distance (μm) between HSC and the endosteum (blue) was determined and marked with white lines and numbers. (B) GFP+ cells from fucose-replete mice are closer to the endosteum than GFP+ cells from fucose-depleted mice.