Mapping of MN1 sequences necessary for myeloid transformation

Abstract

The MN1 oncogene is deregulated in human acute myeloid leukemia and its overexpression induces proliferation and represses myeloid differentiation of primitive human and mouse hematopoietic cells, leading to myeloid leukemia in mouse models. To delineate the sequences within MN1 necessary for MN1-induced leukemia, we tested the transforming capacity of in-frame deletion mutants, using retroviral transduction of mouse bone marrow. We found that integrity of the regions between amino acids 12 to 458 and 1119 to 1273 are required for MN1's in vivo transforming activity, generating myeloid leukemia with some mutants also producing T-cell lympho-leukemia and megakaryocytic leukemia. Although both full length MN1 and a mutant that lacks the residues between 12-228 (Δ12-228 mutant) repressed myeloid differentiation and increased myeloproliferative activity in vitro, the mutant lost its transforming activity in vivo. Both MN1 and Δ12-228 increased the frequency of common myeloid progentiors (CMP) in vitro and microarray comparisons of purified MN1-CMP and Δ12-228-CMP cells showed many differentially expressed genes including Hoxa9, Meis1, Myb, Runx2, Cebpa, Cebpb and Cebpd. This collection of immediate MN1-responsive candidate genes distinguishes the leukemic activity from the in vitro myeloproliferative capacity of this oncoprotein.

Figure 1. Effects of MN1-deletion mutants on myeloid differentiation of U937 cells.

(A) Schematic representation of MN1 and MN1-deletion mutant proteins. (B) Western blot analysis of GFP-sorted U937 stable cell lines overexpressing the indicated proteins. Figure shows immunoblotting results of anti-MN1 antibodies recognizing either the N-terminus (top left panel) or C-terminus of the protein (top right panel) and anti-human ACTIN antibody (bottom right panel). The commassie-blue stained image of the PAGE gel (bottom left panel; CB) and a pan-actin blot (bottom right panel; ACTIN) are shown as loading controls. (C) Immuno-cytochemical analysis of the above cells using the same MN1-antibodies (red). Blue shows nuclear staining with 4,6-diamidino-2-phenylindole (DAPI). Images were captured with an Olympus BX-50 microscope (equipped with UPlanFL 40×/0.75 numerical apertures with a SPOT camera and SPOT Advanced Imaging software of Diagnostic Instruments. The original magnification was x400. (D, E) FACS analysis showing CD11b expression of indicated U937 stable cell lines 3 days after treatment with vehicle, vitamin D₃ (Vit-D3) or ATRA (mean ± SEM of duplicates; *P<.05, ***P<.001, ^#not significant).

Figure 2. Deletion of aa residues between 18–458 or 570–1273 of MN1 restores MN1-impaired myeloid differentiation of mouse HSPC in vitro.

(A) FACS analysis showing the distribution of MEP, CMP and GMP populations in lineage depleted mouse bone marrow HSPC that were transduced with retroviruses expressing GFP alone (mock), MN1, or the indicated MN1-deletion mutants. Freshly isolated bone marrow cells were used to set up the gates. Figure shows one representative results from 3 independent experiments. (B) Mouse HSPCs transduced with indicated retroviruses were subjected to a second lineage depletion 96 hrs after transduction (Day-0) and cells were cultured for an additional 2 days (Day-2). FACS analysis shows Mac1 and Gr1 expression of unsorted cells at Day-0 and Day-2 of culture. (C) FACS analysis showing the Sca1 and c-Kit expression of the GFP (+) cells in panel B.

Figure 3. Amino acid residues 12–228 of MN1 are not required for its proliferative and self-renewing activity in vitro.

(A) Periodic GFP analysis was performed at indicated times of liquid cultures using FACS. GFP percentage at Day-0 of analysis (4 days after the last transduction) in each sample was set to 1 and the results of all indicated time points were calculated relative to Day-0. Graph shows one representative analysis of three independent experiments. (B) Methylcellulose assays showing colony numbers after serial replating (1 to 5) of linage-depleted mouse HSPCs transduced with the indicated retroviruses (mean ± SEM of duplicates). (C) May-Grunwald/Giemsa stained images of cytospins of the cells in panel B (after the first methylcellulose culture). (D, E) Differential-count of cytospins is shown in panel C. A total of 100 cells were counted in each sample. The maturation pool includes neutrophils and metamyelocytes whereas the proliferation pool comprises myelocytes and promyelocytes. The images in panels C and D were captured with an Olympus BX41 microscope, equipped with SPOT Insight Color Mosaic 2 MP camera and SPOT imaging software (original magnification x1000).

Figure 4. Deletions in MN1 that interfere with its leukemogenic activity.

(A) Comparison of in vivo leukemogenic activity of MN1 and Δ1260–1320 transduced bone marrow after transplantation in 10 lethally irradiated mice each. (B) FACS GFP analysis of the peripheral blood (PB) of mice that received transplants of BM expressing MN1 or the indicated MN1 mutants at the indicated time points after transplantation. WBC: White blood cells. (C) Kaplan-Meier survival plots of mice transplanted with bone marrow cells transduced with each indicated retrovirus.

Figure 5. Kaplan-Meier survival curves of mice receiving MN1 or mutant MN1-transduced bone marrow.

(A) Schematic representations of MN1 deletion mutants. Mutants depicted in blue do cause leukemia, while mutants in red do not. Leukemia types generated by each mutant are indicated in the Figure. Stippled lines indicate the size of MN1 deletions. (B) Comparison of Kaplan-Meier survival curves of mice receiving transplants of bone marrow transduced with the indicated MN1 deletion mutants. (C) Histological analysis of a bone marrow sample of a representative leukemic mouse in each indicated group. Images were taken at 400X using an Olympus BX41 microscope, equipped with an Insight 2 Mega-pixel Color Mosaic camera, employing SPOT advanced imaging software. H&E = hematoxilin & eosin staining, GFP = green fluorescent protein staining, MPO = myeloperoxidase staining.

Figure 6. Expression profile analysis of MN1, 12–228 and 18–458 cells.

(A, B, C) Pair wise comparison of CMP/GFP⁺ cells transduced with MN1, Δ12–228, Δ18–458 retrovirus with CMP/GFP⁺ cells transduced with empty vector (mock). (D) Venn diagram showing the number of probe sets that are differentially expressed in the pair wise comparison in A, B and C. (E) Hierarchical clustering of 502 transcripts with >2-fold differential expression in MN1 CMP/GFP⁺ cells and two bulk leukemic MN1-bone marrow (Leuk #1, #2) as compared with CMP/GFP⁺ cells transduced with empty vector (GFP). Heat map indicates expression relative to the mean in standard deviation units.

Figure 7. FACS analysis of leukemic bone marrow samples of MN1-mice.

(A) FACS analysis of bone marrow samples from a wild type (wt-bone marrow) and two moribund MN1-mice to determine the size of the CMP, MEP and GMP populations. GFP⁻ and GFP⁺ gates were used for wild type or MN1-bone marrow cells, respectively. (B) Giemsa staining of leukemic bone marrow cells of the same mice as in A showing their morphology. Arrows indicate different myeloid cell types.