Findings

The autologous-allogeneic hematopoietic cell transplantation has been developed for decades [1,2], and numerous attempts have been made to expand the HSCs population in vitro and in vivo [3-5]. Agiopoietin-like proteins belong to a 7-member family of secreted glycoproteins that share sequence homology with angiopoietins, which are important modulators of angiogenesis [6]. Several Angptl family proteins promote the expansion of murine and human HSPCs in vitro and ex vivo [7-9]. Angptl7 in the ECM of the trabecular meshwork plays an important role in the deposition and organization of the matrix of the outflow tissue [10]. Recently, we found that ANGPTL7 stimulated the proliferation of human HSPCs ex vivo (Yiren Xiao, unpublished data). In addition, we uncovered that Angptl7, which was secreted by murine bone marrow SSEA4+ mesenchymal cells (Additional file 1: Figure S1), stimulated expansion of murine HSCs ex vivo (Additional file 2: Figure S2). However, whether Angptl7 is redundant and dispensable or not for repopulation of HSPCs in vivo remains unknown. In this study, we generated Angptl7 knockout mice and revealed that Angptl7 is essential for HSPC repopulation in a non-cell autonomous way.

To investigate whether the loss of Angptl7 affected HSPCs in vivo, we generated Angptl7-null mice by TALEN-mediated gene targeting (Additional file 1: Figure S1a-1c). The procedures of gene targeting were described in (Additional file 3: Supplementary methods) and (Additional file 4: Table S1). Loss of Angptl7 in Angptl7-null mice was confirmed in the bone marrow (BM) by RT-PCR and Western blotting (Additional file 5: Figure S3d-3e). The Angptl7-null mice did not display an overt phenotype. There were no significant differences in either birth weights or adult weights among Angptl7 −/−, Angptl7 +/−, and Angptl7 +/+ mice (Additional file 6: Table S2). In addition, we found no significant differences in staining profiles T cells, B cells, myeloid cells, and erythroid cells among the Angptl7 −/−, Angptl7 +/− and Angptl7 +/+ mice (Figure 1a).

Figure 1
figure 1

Reduced HSC in Angptl7 −/− mice. (a) Characterization of multiple hematopoietic lineages in BM. The major lineages of hematopoietic cells in BM in Angptl7 −/− Angptl7 +/− and Angptl7 +/+ mice were quantitated by flow cytometry analysis as erythroid cells (Ter119), myeloid cells (Ly6g and CD11b), T cells (CD3e) and B cells (CD19). Data represent the means +/− s.e.m (n = 5 for each group). (b, c) Comparison of the frequencies (b) and number (c) of Lin-Mac-1 + CD4- population in the fetal liver of Angptl7 −/− Angptl7 +/− and Angptl7 +/+ mice. Data represent the means +/− s.e.m (n = 10 for each group). *P < 0.05 versus Angptl7 −/− group for Angptl7 +/− and Angptl7 +/+. (d, e) Comparison of the frequencies (d) and number (e) of LSK population in the BM of Angptl7 −/− Angptl7 +/− and Angptl7 +/+ mice. Data represent the means +/− s.e.m (n = 6 for each group). *P < 0.05 versus Angptl7 −/− group for Angptl7 +/− and Angptl7 +/+ group. (f, g) Comparison of the frequencies (left) and number (right) of LT-HSCs in the BM of Angptl7 −/− Angptl7 +/− and Angptl7 +/+ mice. Data represent the means +/− s.e.m (n = 5 for each group). *P < 0.05 versus Angptl7 −/− group for Angptl7 +/− and Angptl7 +/+ group. (h) The colony numbers of CFU assays obtained from BM of Angptl7 −/− Angptl7 +/− and Angptl7 +/+ mice. Data represent the means +/− s.e.m (n = 3). *P < 0.05 versus bar 7 for bar 9.

Full size image

As HSCs undergo dramatic expansion in fetal liver during embryo development [11-13], we examined the fetal liver HSC compartment and found that the percentages and numbers of Lin-Mac-1 + CD4- population, which was highly enriched of HSCs in fetal liver [14,15], were significantly lower in the fetal livers of Angptl7 −/− mice than that in Angptl7 +/− and Angptl7 +/+ mice (Figure 1b-c). In addition, we found the percentages and numbers of Lin-Sca-1 + c-Kit + (LSK) population in the BM of adult Angptl7 −/− mice were significantly lower than that in Angptl7 +/− and Angptl7 +/+ mice (Figure 1d-e). Further analysis showed that Angptl7 −/− mice had fewer long-term HSCs (Lin−Sca-1+Kit+Flk2−CD34−) than Angptl7+/− and Angptl7 +/+ mice (Figure 1f-g). Cloning forming assay results showed that BM cells from Angptl7 −/− mice had fewer granulocyte/monocyte progenitors (CFU-GM), but similar numbers of erythroid precursors (CFU-E/BFU-E) and hematopoietic progenitors (CFU-GEMM) than WT BM (Figure 1h). Therefore, Angptl7-deficient BM had normal levels of terminally differentiated hematopoietic cells, but decreased myeloid progenitors. To investigate whether other angiopoietin-like proteins compensate the loss of Angptl7 in BM, we compared the expression levels of other angiopoietin-like proteins in Angptl7-deficient BM stromal cells to that in WT BM stromal cells and found that their expression levels did not significantly change due to loss of Angptl7 (Additional file 7: Figure S4).

Since Angptl7 is secreted by stromal cells and binds to HSPCs [9], we speculated that Angptl7 may play a non-cell autonomous role in reconstitution of the hematopoietic system. To evaluate the hypothesis, we compared the extent of repopulation of HSCs in Angptl7 −/−, Angptl7 +/− and Angptl7 +/+ recipient mice. After WT donor BM cells were injected into lethally irradiated Angptl7 −/− Angptl7 +/− or Angptl7 +/+ recipients without competitors (Figure 2a), we found that the reconstitution efficiencies of HSCs were similar among Angptl7 −/−, Angptl7 +/− and WT recipient mice (Figure 2b), but the frequencies and numbers of LT-HSCs in Angptl7 −/− recipients were significantly lower than that in Angptl7 +/− and WT recipients, suggesting that Angptl7 supports the maintenance of the HSC pool in BM (Figure 2c-d). Continuously the BM cells from the primarily repopulated Angptl7 −/−, Angptl7 +/− and WT recipient mice were collected for secondary transplantation. We found that the repopulating activity of cells originating from the primary Angptl7 −/− recipients was significantly decreased compared with those from the primary Angptl7 +/− and WT recipients at different time points (Figure 2e). To test whether Angptl7 had a cell-intrinsic effect on HSCs, we transplanted Angptl7 −/−, Angptl7 +/− or WT BM cells into lethally irradiated WT recipients and measured the frequencies of donor cells in recipient mice (Figure 2f). We found the repopulation efficiencies, percentages, and numbers of donor Angptl7 −/−, Angptl7 +/− and WT LT-HSCs were similar in WT recipients (Figure 2g-i). We further transplanted the BM cells from the primary recipient mice to lethally irradiated WT mice. The repopulation efficiencies of secondary transplantation were similar among Angptl7 −/−, Angptl7 +/− and WT donors at indicated time points (Figure 2j). Overall, these results suggest that Angptl7 promoted HSC repopulation in a non-cell autonomous manner in vivo.

Figure 2
figure 2

Comparison of reconstitution capacity of HSC in Angptl7 −/− Angptl7 +/− and Angptl7 +/+ mice. (a) Experimental design for assessing the repopulation efficiency in Angptl7-null BM environment. (b) Comparison of the repopulation efficiency in BM of Angptl7 −/− Angptl7 +/− and Angptl7 +/+ recipients. Data represent the means +/− s.e.m. (n = 8 for each group). (c, d) Comparison of the frequencies (c) and number (d) of LT-HSCs in BM of Angptl7 −/− Angptl7 +/− and Angptl7 +/+ recipients. Data represent the means +/− s.e.m. (n = 8 for each group). *P ≤ 0.05 for bar 9 versus bar 7 and bar 8. (e) Comparison of the second transplantation repopulation efficiency in BM of Angptl7 −/− Angptl7 +/− and Angptl7 +/+ recipients as referred in (a). Data represent the means +/− s.e.m (n = 7 for each group). *P ≤ 0.05 for bar 6 versus bar 4 and bar 5, for bar 9 versus bar 7 and bar 8, for bar 12 versus bar 10 and bar 11. (f) Experimental design effect of endogenous deletion of Angptl7 in HSCs. Angptl7 −/− Angptl7 +/− or Angptl7 +/+ donor BM CD45.2 cells (5x105 cells) were transplanted into lethally irradiated WT CD45.1 recipient mice. (g) Comparison of the repopulation efficiency of donor BM Angptl7 −/− Angptl7 +/− and Angptl7 +/+cells. Data represent the means +/− s.e.m. (n = 8 for each group). (h, i) Comparison of the frequencies (h) and number (i) of donor Angptl7 −/− Angptl7 +/− and Angptl7 +/+ LT-HSCs in BM of WT recipients. Data represent the means +/− s.e.m. (n = 8 for each group). (j) Comparison of the repopulation efficiency of Angptl7 −/− Angptl7 +/− and Angptl7 +/+ donor BM cells in second recipients mice as described in (f). Data represent the means +/− s.e.m (n = 8 for each group).

Full size image

In this study, we demonstrated that deficiency of Angptl7 in the BM niche as a lack of exogenous Angptl7 stimulation impaired the reconstitution of HSCs in lethally irradiated mice, whereas endogenous deletion of Angptl7 in HSCs did not affect repopulation of HSCs in lethally irradiated mice. Thus, Angptl7 was indispensable for BM microenvironment to support HSC repopulation.