New therapeutic opportunities from dissecting the pre-B leukemia bone marrow microenvironment

Abstract

The microenvironments of leukemia and cancer are critical for multiple stages of malignancies, and they are an attractive therapeutic target. While skeletal abnormalities are commonly seen in children with acute lymphoblastic leukemia (ALL) prior to initiating osteotoxic therapy, little is known about the alterations to the bone marrow microenvironment during leukemogenesis. Therefore, in this study, we focused on the development of precursor-B cell ALL (pre-B ALL) in an immunocompetent BCR-ABL1⁺ model. Here we show that hematopoiesis was perturbed, B lymphopoiesis was impaired, collagen production was reduced, and the number of osteoblastic cells was decreased in the bone marrow microenvironment. As previously found in children with ALL, the leukemia-bearing mice exhibited severe bone loss during leukemogenesis. Leukemia cells produced high levels of receptor activator of nuclear factor κB ligand (RANKL), sufficient to cause osteoclast-mediated bone resorption. In vivo administration of zoledronic acid rescued leukemia-induced bone loss, reduced disease burden and prolonged survival in leukemia-bearing mice. Taken together, we provide evidence that targeting leukemia-induced bone loss is a therapeutic strategy for pre-B ALL.

Fig. 1

Pre-B leukemia impairs hematopoiesis in the bone marrow. a Schematic representation of the experimental model. The normal bone marrow cells were transduced with MSCV-BCR-ABL1-IRES-mCherry retrovirus and transplanted into lethally irradiated Recipient 1. All subsequent transplantations were performed using non-irradiated recipients. b Kaplan–Meier survival curve of mice transplanted with 1000 BCR-ABL1 ALL cells (n = 24 mice). c Percentage of mCherry⁺ cells in bone marrow, spleen and blood during leukemia development. d Number of hematopoietic cells (CD45⁺mCherry⁻) in the bone marrow during leukemia development. e Percentage of B220⁺ cells in the CD45⁺mCherry⁻ fraction during leukemia development. f Percentage of CD11b⁺ cells in the CD45⁺mCherry^- fraction during leukemia development. g Percentage of NK1.1⁺ cells in the CD45⁺mCherry^- fraction during leukemia development. h Composition of B cell subpopulations during leukemia development (percentage in the B220⁺CD45⁺mCherry^- fraction). c-h Bone marrow cells were harvested from one femur and two tibias (Day 8, 10, 13, 15, and 16: n = 4 mice; Day 17: n = 3 mice; Day 20: n = 6 mice). i Representative images of immunostaining for collagen I (brown) from WT and mice at Day 8 and Day 20 post leukemia cell injection (scale bars, 50 μm) (left) and quantification of the percentage of collagen I per surface area (n = 3) (right). Throughout, *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001. Error bars represented mean ± SD

Fig. 2

Pre-B leukemia induces bone loss in vivo. a, b Bone marrow cells were harvested from one femur and two tibias after enzymatic dissociation (Day 8, 10, 13, 15, and 16: n = 4 mice; Day 17: n = 3 mice; Day 20: n = 6 mice). a Number of endothelial cells. b Number of osteoblastic cells. c Mean expression of osteocalcin (Bglap) in bone marrow CD45⁻ cells (WT), CD45⁺ (WT) and in CD45⁻ cells in leukemia-bearing mice at Day 16 post leukemia cell injection (n = 3). d 3D rendering of distal femur bone compartment (top panels: longitudinal sections; bottom panels: axial sections). Representative cortical and trabecular bones in the distal femurs of mice at stated days post leukemia cell injection (false colored images, n = 4 mice per time point). e–g Micro-CT analysis of femoral trabecular bone (n = 4 mice per time point). e Bone volume/total volume (BV/TV). f Trabecular spacing (Tb.Sp). g Trabecular number (Tb.N). h–i Micro-CT analysis of femoral cortical bone at trabecular zone (n = 4 mice per time point). h Cortical volume. i Cortical thickness. Throughout, *P < 0.05, ** P < 0.01, *** P < 0.001, **** P < 0.0001. Error bars represented mean ± SD

Fig. 3

Pre-B ALL increases the activity of osteoclasts and produces high levels of RANKL. a Representative images of paraffin sections of distal femur bone compartment stained for the osteoclastic marker TRAP at stated days post leukemia cell injection (n = 4 mice per time point, scale bars, 50 μm). b Quantification of TRAP⁺ osteoclasts in the distal femur bone compartment. c Serum levels of CTX in WT and leukemia mice (WT, Day 4, 8, and 10: n = 4 mice; Day 16: n = 3 mice). d Mean expression of Rankl and Csf1 in B220⁺ cells in WT mice relative to that in mCherry⁺ cells in mice at Day 10 and 16 post leukemia cell injection (WT, Day 10: n = 4 mice; Day 16: n = 3 mice). e Leukemia cells were cultured for 48 h and the level of RANKL in the supernatants was measured by ELISA (n = 3). f Representative western blot of RANKL from leukemia cell lysate (n = 3). Lane 1 is B220⁺ cell lysate and Lane 2 is leukemia cell lysate. β-actin was used as a loading control. g RANKL levels measured by ELISA in bone marrow supernatants from control and leukemia mice (n = 3). h Representative images of RAW 264.7 cells stained for TRAP after culture with 20 ng/mL RANKL or 1 × 10⁵ leukemia cells for 8 days (scale bars, 100 μm) (left) and quantification of the TRAP⁺ multinucleated osteoclasts (n = 3) (right). i Mean expression of Pax5 in pre-B cells (CD19⁺CD24⁺BP-1⁺IgM⁻) in WT mice relative to that in BCR-ABL1⁺ cells (Fig. 1a) and in mCherry⁺ cells in mice at Day 16 post leukemia cell injection (n = 3). Throughout, *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001. Error bars represented mean ± SD

Fig. 4

Zoledronic acid reduces leukemia burden and improves survival in mice. a Schematic diagram of zoledronic acid treatment schedule. b Leukemia cells were incubated with 0, 1 μM, or 10 μM of zoledronic acid for 48 h. Cell viability was evaluated by alamarBlue assay (n = 4). c Representative images of colony forming cells from leukemia cells treated with 0 or 10 μM of zoledronic acid for 7 days (scale bars, 500 μm) (left) and quantification of the number of colony forming cells (n = 3) (right). d Representative images of paraffin sections of distal femur bone compartment stained for the osteoclastic marker TRAP at Day 15 in mice treated with zoledronic acid or vehicle (scale bars, 10 μm). e Micro-CT analysis (left) and quantification (right) of femur trabecular bone at Day 15 in mice treated with zoledronic acid or vehicle (n = 8 mice/group). f Serum level of CTX at Day 15 in mice treated with either zoledronic acid or vehicle (n = 7 mice/group). g Percentages of mCherry⁺ leukemia cells at Day 15 in the bone marrow, spleen, and blood in mice treated with either zoledronic acid or vehicle (n = 8 mice/group). h Number of non-leukemia cells at Day 15 in the bone marrow of mice treated with either zoledronic acid or vehicle (n = 8 mice/group). i Kaplan–Meier survival curves of leukemia mice that were treated with either zoledronic acid or vehicle (n = 19 mice/group). Throughout, *P < 0.05, ***P < 0.001, ****P < 0.0001. Error bars represented mean ± SD