Endogenous oncogenic Nras mutation promotes aberrant GM-CSF signaling in granulocytic/monocytic precursors in a murine model of chronic myelomonocytic leukemia

Abstract

Oncogenic NRAS mutations are frequently identified in myeloid diseases involving monocyte lineage. However, its role in the genesis of these diseases remains elusive. We report a mouse bone marrow transplantation model harboring an oncogenic G12D mutation in the Nras locus. Approximately 95% of recipient mice develop a myeloproliferative disease resembling the myeloproliferative variant of chronic myelomonocytic leukemia (CMML), with a prolonged latency and acquisition of multiple genetic alterations, including uniparental disomy of oncogenic Nras allele. Based on single-cell profiling of phospho-proteins, a novel population of CMML cells is identified to display aberrant granulocyte-macrophage colony stimulating factor (GM-CSF) signaling in both the extracellular signal-regulated kinase (ERK) 1/2 and signal transducer and activator of transcription 5 (Stat5) pathways. This abnormal signaling is acquired during CMML development. Further study suggests that aberrant Ras/ERK signaling leads to expansion of granulocytic/monocytic precursors, which are highly responsive to GM-CSF. Hyperactivation of Stat5 in CMML cells is mainly through expansion of these precursors rather than up-regulation of surface expression of GM-CSF receptors. Our results provide insights into the aberrant cytokine signaling in oncogenic NRAS-associated myeloid diseases.

Figure 1

Somatic expression of oncogenic Nras at its endogenous level does not lead to acute development of myeloproliferative disease. Six-week-old control and Nras G12D mice were injected with pI-pC as described in “Mice.” One week after last injection, different tissues were isolated and analyzed. (A) Schematic illustration of floxed and activated oncogenic Nras alleles. (B) Genotyping analysis of genomic DNA to detect wild-type allele, LSL allele, and recombined LSL allele (1 LoxP allele). (C) Total RNA was extracted from bone marrow cells. Direct sequencing of RT-PCR–amplified Nras gene using a reverse primer to confirm the sequences at the codon 12 (underlined in red). Arrows indicate the wild-type and mutated nucleotide at the codon 12. (D) Levels of Nras-GTP, the active form of Nras, were analyzed in lysates extracted from bone marrow cells by affinity purification (AP) of lysates using a glutathione S-transferase fusion with the Ras binding domain of Raf (Raf RBD) immobilized on agarose beads followed by Western blot analysis using an antibody against Nras. The top panel illustrates the total input levels of Nras proteins. (E) Peripheral blood samples were collected from Nras G12D mice and control mice. Debris and unlysed red blood cells (low forward scatter) and dead cells (propidium iodide positive) were excluded from analysis. The percentages of T cells (Thy1.2), B cells (CD19), and myeloid cells (Mac1 and Gr1) are indicated on each plot.

Figure 2

All recipient mice transplanted with whole bone marrow cells expressing oncogenic Nras died after a prolonged latency. Lethally irradiated mice were transplanted with 2.5 × 10⁵ control bone marrow cells (n = 7) or bone marrow cells expressing oncogenic Nras (n = 26) along with same number of competitor cells. (A) Dynamic contribution of donor-derived white blood cells (CD45.2⁺) in peripheral blood of recipient mice at different time points posttransplant. (B) Kaplan-Meier comparative survival analysis of reconstituted mice. Cumulative survival was plotted against days after transplantation.

Figure 3

The majority of recipient mice transplanted with whole bone marrow cells expressing oncogenic Nras developed a CMML-like disease. (A) Dynamic percentages of donor-derived monocytes and neutrophils in peripheral blood of diseased mice transplanted with Nras G12D cells or control mice transplanted with control cells. (B) Splenomegaly and hepatomegaly in diseased mice. (Top) Splenomegaly in a representative recipient mouse that developed a CMML-like disease. (Bottom) Results are presented as averages of spleen weights or liver weights + SD. (C) Representative histologic hematoxylin and eosin–stained sections from spleen showed an extensive infiltration of myelomonocytic cells and extramedullary hematopoiesis in recipient mice transplanted with bone marrow cells expressing oncogenic Nras. (D) Flow cytometric analysis of bone marrow cells and splenocytes using myeloid specific markers (Mac1 and Gr1). The percentages of monocytic (top left) and granulocytic (top right) lineages of cells are indicated on the plots. Representative data from 8 CMML mice are shown. (E) Quantification of monocytes (W1) and neutrophils (W2) in peripheral blood were based on their surface expression of Mac1 and Gr1. The right panels show May-Grunwald Giemsa–stained cytospin preparations of cells sorted from W1 and W2 regions to confirm the identities of cells.

Figure 4

Myeloid progenitors from CMML mice show abnormal growth pattern in semisolid culture. (A) A quantity of 5 × 10⁴ bone marrow cells isolated from CMML mice and control mice were plated in duplicate in semisolid medium with or without GM-CSF. The data are presented as average percentages (from multiple CMML-like and control mice) of maximum number of colonies formed in culture with 0.2 ng/mL of GM-CSF. Photomicrographs (original magnification ×20) showed control and CMML myeloid progenitor colonies grown in different concentrations of GM-CSF. Student t test was performed. Error bars show SDs. Asterisks indicate P < .002. (B) Analysis of myeloid progenitors in bone marrow cells isolated from CMML and control mice. Percentages of myeloid progenitors (IL-7Rα^– Lin^– Sca1^– c-Kit⁺ cells), common myeloid progenitors (CMPs; FcγR^lo/− CD34⁺), GMPs (FcγR^hi CD34⁺), and megakaryocyte erythroid progenitors (MEPs; FcγR^loCD34^–) relative to total bone marrow cells are indicated. (C-D) Quantitative analysis of myeloid progenitor compartment in control mice (n = 6) and CMML mice (n = 11). Results are presented as averages + SD. Student t test was performed. Asterisks indicate P < .05.

Figure 5

c-Kit⁻ [Lin Sca-1 IL7Rα]^−/low cells from CMML mice show hyperactivation of the ERK and Stat5 pathways. Total bone marrow cells isolated from CMML mice or control mice were serum- and cytokine-starved for 1 hour and stimulated with various concentrations of GM-CSF (0, 0.16, 0.32, 2, and 10 ng/mL) at 37°C for 10 minutes. Levels of phosphorylated ERK1/2 were measured using phospho-specific flow cytometry. Nonneutrophil bone marrow cells were gated for data analysis. Representative gating strategy and plots of p-ERK1/2 (A) and p-Stat5 (B) are shown. Myeloid progenitors are enriched in c-Kit⁺ [Lin Sca-1 IL7Rα]^−/low cells as described in Figure 4 (G1 cells), whereas G2 cells are defined as c-Kit⁻ [Lin Sca-1 IL7Rα]^−/low cells. To quantify the activation of ERK1/2 and Stat5 in CMML cells and control cells, median intensities of p-ERK1/2 and p-Stat5 at different GM-CSF concentrations are compared with their respective control cells at 0 ng/mL, which is arbitrarily set at 1. Representative results of 3 experiments of more than 10 independent experiments are shown.

Figure 6

A fraction of CMML mice acquire uniparental disomy of oncogenic Nras allele. (A) Western blot analysis of total Nras expression levels in bone marrow cells from representative animals. Note: Nras expression is completely absent in LSL/LSL mice due to the stop cassette, whereas Nras expression in LSL/+ mice is half of that of wild-type (WT) mice. (B) Total RNA and genomic DNA were extracted from bone marrow cells. Pyrosequencing of RT-PCR amplified products and direct sequencing of genomic PCR products at the G12 codon in representative animals are shown. The wild-type and mutated nucleotides are highlighted in yellow.

Figure 7

Granulocytic/monocytic precursor cells are expanded in c-Kit⁻ [Lin Sca-1 IL7Rα]^−/low compartment in CMML-like mice. Total bone marrow cells were isolated from control and CMML-like mice and simultaneously stained for myeloid progenitors as well as for granulocytic/monocytic precursors. (A) Regions I-V are defined by characteristic staining pattern of cells, including ER-MP12⁺ ER-MP20⁻, ER-MP12⁺ ER-MP20^mid, ER-MP12⁻ ER-MP20^mid, ER-MP12⁻ ER-MP20^hi, and ER-MP12⁺ ER-MP20^hi, respectively. The density plots and percentages of regions I-V cells in c-Kit⁻ [Lin Sca-1 IL7Rα]^−/low compartment (B) are shown from 1 representative control and 1 CMML-like mouse. (C) Sorted cells from region III were subjected to phospho-flow analysis in the ERK and Stat5 pathways as described in Figure 5.