The authors have declared that no competing interests exist.
Conceived and designed the experiments: JYM TW JTG RTC CED. Performed the experiments: JYM TW JTG MAE. Analyzed the data: JYM JTG. Contributed reagents/materials/analysis tools: PDA. Wrote the paper: JYM CED.
We previously reported the development of a lethal myeloid sarcoma in a non-human primate model utilizing retroviral vectors to genetically modify hematopoietic stem and progenitor cells. This leukemia was characterized by insertion of the vector provirus into the
Recently we reported the development of an acute myeloid leukemia in a rhesus macaque transplanted with autologous CD34+ cells transduced with a murine stem cell virus-derived replication defective retroviral vector expressing only marker genes under control of the strong MCSV long terminal repeat (LTR). This animal had an unusual clonal reconstitution pattern the first year following transplant, with a single transduced myeloid progenitor cell clone accounting for up to 80% of the then normal myelopoiesis
BCL2A1, also known as Bfl-1, GRS, or BCL2L5, belongs to the BCL2 family of anti-apoptotic proteins. Murine BCL2A1 was originally identified as a gene product induced by the stimulation of primary bone marrow cells with GM-CSF
There is limited information linking BCL2A1 expression and human hematologic malignancies. In one small study
In this study we aimed to better understand the potential role of dysregulation of the
All mice were housed and handled in strict accordance with the recommendations in the “Guide for the Care and Use of Laboratory Animals” of the National Institutes of Health, Bethesda, MD. All animal experiments were carried out on National Heart Lung and Blood Institute Animal Care and Use Committee–approved protocol number H-1031R2.
Murine
LTRs are self inactivating (SIN), ψ (packaging sequence), RRE (Rev-Responsive Element), PPT (Polypurine Tract), SFFV (Spleen Focus-Forming Virus promoter), IRES (Internal Ribosome Entry Site), HA (Hemagglutinine antigen), GFP (Green Fluorescent protein), PRE (post-transcriptional regulatory element). (B,C) Expression was assessed by western blot using an anti-HA antibody to recognize HA-tagged BCL2A1, or anti-GFP, and anti GAPDH as a loading control. Western blots were carried out in 293T producer cell lines (B) and in transduced BAF3 and 32Dcl3 cells (C).
To produce lentiviral particles, 293T cells were seeded at 5 million cells per 10-centimeter dish. Twenty-four hours later, cells were co-transfected (Calcium Phosphate Transfection Kit, Sigma-Aldrich, Saint Louis, MO) with the following plasmids: pCDNA3.HIVgag/pol.4xCTE (12 ug), pMD2.G-VSV-G (1 ug), pRSV-REV (5 ug), and 10 ug of pRRL.PPT.SF.IRES.GFPpre/HA, pRRL.PPT.SF.IRES.GFPpre/HA-mBcl2a1a, or pRRL.PPT.SF.IRES.GFPpre/HA-hBCL2A1. Vector-supernatants were collected once a day for 3 days, filtered through 0.22-um filters, and concentrated by ultracentrifugation for 2 hr at 71 900 g and 4°C in a SW28 rotor (Beckman Coulter, Atlanta GA). Pelleted viral particles were resuspended in 1 mL of StemSpan® (StemCell Technologies Inc., Vancouver, Canada) media and stored at minus 80°C. Viral titers were determined on NIH 3T3 cells via standard assays
The murine myelomonocytic WEHI-3 (ATCC, #TIB-68, Manassas, VA) cell line was grown in IMDM supplemented with 10% FCS, 2.5 mM β2-mercapto-ethanol, glutamine, penicillin and streptomycin. Murine NIH3T3 (ATCC, #CRL-1658) and human 293T (ATCC, #CRL-11268) fibroblastic cell lines were grown in I10 media (IMDM supplemented with 10% FCS, glutamine, penicillin and streptomycin). The murine pro-B BaF3 (Leibniz-Institut DSMZ, #ACC300, Braunschweig, Germany) and murine myeloblast-like 32Dcl3 (ATCC, #CRL-11346) cell lines were grown in RPMI1640X supplemented with 10% WEHI-3 conditioned media as a source of IL-3, 10% fetal calf serum (FCS Heat-inactivated, Sigma-Aldrich), glutamine, penicillin and streptomycin (Gibco, Life Technologies, Grand Island, NY). The human erythroblastoid UT7/Epo-S1 cell line was grown in I10 media containing 2 U/mL of human erythropoietin (Amgen, Thousand Oaks, CA). Fifty thousand BaF3, 32Dcl3, or UT7/Epo-S1 cells
Murine bone marrow cells were obtained by flushing bone marrow from humeri, femurs, and tibias of C57BL/6 Ly5.1 mice (C57BL/6 Ly5.1 (Pep3b) Stock 002014, Jackson Laboratories, Bar Harbor, ME). Red cells were lysed with ACK buffer (Quality Biological Inc., Gaithersburg, MD). Nucleated cells were purified using the MACS mouse lineage depletion kit (Miltenyi Biotec, Auburn, CA), and the resulting lineage negative progenitor cells were stimulated for two days at a starting density of 5×105 cells/mL in StemSpan® media supplemented with 10 ng/mL murine IL-3 (R&D Systems, Minneapolis, MN), 100 ng/mL murine IL-11 (R&D Systems), 50 ng/mL murine stem cell factor (Research Diagnostic Inc., Concord, MA), and 100 ng/mL human Flt-3 Ligand (R&D Systems). Each recipient mouse was transplanted with the equivalent of 5×105 stimulated cells that were transduced twice with lentiviral supernatants at a MOI of 5 on retronectin-coated plates in the presence of protamine sulfate 4 ug/mL.
Transduced lineage negative bone marrow cells were resuspended in StemSpan® media. Following 900 rads total body irradiation each recipient C57BL/6 Ly5.2 mouse (Stock 000664, Jackson Laboratories) received transduced cells resuspended in 500 uL media via tail vein injection.
To perform secondary transplants, bone marrow cells were collected by flushing humeri, femurs, and tibias of primary mice, and red cells were lysed using ACK buffer. Using a ratio of 1 primary mouse for 3 secondary mice, primary bone marrow cells were reinfused by tail vein injection into sub-lethally irradiated secondary mice C57BL/6 (Ly5.2) (900 rads total body irradiation).
Blood was collected monthly via retro-orbital bleeding, and used for a complete blood count (Hemavet 950FS, Drew Scientific, Waterbury, CT), blood smear (hematoxylin and eosin (H&E) stained), and flow cytometric analysis.
Organs including heart, lung, kidney, spleen, liver, salivary glands, lymph nodes, any obvious tumor masses, and sternum from pre-morbid mice were collected, fixed in 10% formalin (Fischer Scientific, Thermo Fisher Scientific Inc., Pittsburgh, PA), and embedded into paraffin blocks for subsequent sectioning and H&E staining (Histosev, Gaithersburg, MD) or immunohistochemistry. Single cell bone marrow, lymph node, and spleen suspensions were obtained by flushing tissues with media. Cytospins were prepared from 105 cells, using 300 rpm for 5 minutes (CytoSpin 4 Cytocentrifuge, Thermo Scientific, Thermo Fisher Scientific Inc.). Cell suspensions were analyzed by flow cytometry. Genomic DNA was isolated from cells or tissues using Qiagen DNeasy Tissue and Blood kit according to manufacturer’s recommendation.
Transduced BaF3, 32Dcl3, and UT7/Epo-S1 cells were sorted for GFP expression using a BD FACSAria cell sorter from BD Bioscience (San Jose, CA). Peripheral blood, bone marrow, and spleen cells were resuspended in FACS buffer (2.7 mM Potassium chloride, 1.5 mM Sodium phosphate dibasic heptahydrate, 1.5 mM Potassium phosphate monobasic, Sodium chloride 137 mM, Sodium azide 7.7 mM, 1% (w/v) BSA in water) and incubated with the following cocktail of antibodies from BD Pharmingen™ (BD Bioscience, San Diego, CA): T-cells-anti-mouse CD3e (Hamster, APC), granulocytes and NK cells-anti-mouse CD11b (Rat, PE-Cy7), B-cells-anti-mouse CD45R/B220 (Rat, APC-Cy7). Engraftment of donor cells was monitored via anti-mouse CD45.1 (Mouse, R-PE) and anti-mouse CD45.2 (Mouse, PerCP-Cy5.5). After washing with FACS buffer, cells were analyzed with a BD™ LsrII flow cytometry system from BD Bioscience. Data were analyzed using FlowJo from Treestar Inc., Ashland, OR).
Cell cycle status was assessed using the NuCycl™ PI Kit from Exalpha (Watertown, MA) according to the manufacturer’s recommendation. Apoptosis was assessed using the Annexin V-PE apoptosis detection kit I from (BD Pharmingen™, BD Biosciences). Analysis of cell cycle and apoptosis was performed on a BD™ LsrII flow cytometer. Data were processed using ModFit™ software from Verity Software House (Topsham, ME).
Immunohistochemical staining was performed on formalin-fixed, paraffin-embedded tissue sections from the spleen, liver, sternal bone marrow, affected lymph nodes, and tumor masses. Cut tissue sections were deparaffinized, and endogenous peroxidase was inactivated. Antibodies included rat anti-mouse B220 (clone RA3-6B2, BD Pharmingen™, BD Biosciences), rabbit polyclonal anti-mouse CD3 (Dako, Glostrup, Denmark), rat anti-mouse CD34 (Clone MEC 14.7, Abcam, Cambridge, MA), rat anti-mouse CD68 (Clone FA-11, Abcam), mouse anti-mouse Lysozyme (clone BGN/06/961, Abcam), rat anti-mouse Mac2 (Clone M3/38, Cedarlane Laboratories Limited, Burlington, NC), rabbit anti-human myeloperoxidase (MPO) with cross-reactivity to mouse (Dako), peanut agglutinin (Vector Laboratories, Burlingame, CA), and rabbit anti-human TdT with cross-reactivity to mouse (Supertechs Inc., Rockville MD). For CD3, CD34, lysozyme, Mac2 and MPO staining, antigen retrieval was performed using either the Bond Epitope Retrieval Solution 1 (ER1) or the Bond Epitope Retrieval Solution 2 (ER2) (Leica Biosystems, Newcastle Upon Tyne, UK) at 99–100°C for 20–30 min. Enzymatic retrieval was performed for CD68. TdT was kept in a pressure cooker in citrate buffer with pH = 6 for 10 minutes. Subsequently, CD3, MPO, and TdT sections were incubated with the primary antibody for 30 minutes, followed by incubation with enVision + System HRP labeled polymer anti-rabbit secondary antibody (Dako) for 25 minutes. Antibodies for B220, CD34 and Mac2 were incubated with the primary antibody for 30 minutes, followed by a secondary biotinylated polyclonal goat anti-rat antibody (BD Pharmingen™, BD Biosciences) for 25 minutes and Streptavidin-HRP (Dako) for 25 minutes. Lysozyme was stained using the Animal Research Kit Peroxidase for mouse primary antibodies (Dako), as described by the manufacturer. Sections underwent colorimetric development with diaminobenzidine (DAB, Leica Biosystems), were counterstained with hematoxylin, dehydrated using graded alcohols and mounted in Cytoseal™ XYL (Richard-Allan Scientific, Kalamazoo, MI).
Frozen spleen fragments were ground in a total volume of 750 uL of RIPA buffer (Sigma-Aldrich) supplemented with a cocktail of protease inhibitors (Complete Mini EDTA-Free and PhosSTOP, Roche, Indianapolis, IN). After clarification by centrifugation for 20 min at 20 800 g and 4°C, protein concentration was determined using the Protein Assay reagent (Bio-Rad, Hercules, CA). Proteins from 14 ug of lysate were separated by electrophoresis on 4–15% acrylamide Ready gel (Biorad) in Tris/Glycine/SDS buffer (Bio-Rad). Protein standards BenchMark™ and MagicMark (Invitrogen™, Life Technologies) were used as recommended by manufacturer. Proteins were transferred to nitrocellulose membranes (Nitrocellulose Membrane Filter Paper Sandwich, Invitrogen™, Life Technologies) for 1 hr at 90 V in 10% methanol Tris/Glycine buffer (Bio-Rad). Membranes were rinsed in water, and blocked in 5% milk (Bio-Rad)/0.05% Tween20 (Bio-Rad)/PBS, pH 7.4 for 1 hr at room temperature. Primary mouse monoclonal antibodies including anti-HA (Clone 16B12, Covance, Berkeley, CA), anti-GAPDH (Clone 6C5, Ambion, Austin, TX), or anti-GFP (Clone 7.1 and 13.1, Roche) were diluted in 5% milk/0.05% Tween/PBS and incubated at 4°C overnight. Membranes were washed twice in 0.05% Tween/PBS for 10 min at room temperature. Secondary anti-mouse immunoglobulins/HRP polyclonal antibody (Dako) conjugated to HRP was diluted in 5% milk/0.05% Tween/PBS and incubated at room temperature for 45 min. After washing three times with 0.05% Tween/PBS and once with PBS, the membrane was incubated with electrochemiluminescent substrate (LumiGLOR, Cell Signaling Technology, Danvers, MA). Membranes were then exposed to a Luminescent Image Analyser (LAS-400, Fujifilm, Tokyo, Japan).
Total RNA was isolated using RNeasy mini kit according to manufacturer’s instruction (Qiagen). DNAse I (RNAse-free DNAse set, Qiagen) was added to remove any residual contaminating genomic DNA. Real-time PCR was carried out on 75 ng of total RNA per sample using the TaqMan® One-Step RT-PCR Master Mix Reagents Kit (Applied Biosystems, Foster City, CA) according to manufacturer’s instruction. Primers and probes designed to detect total murine
An adaptation of the PCR-based method reported by Schlissel and colleagues was used
Genomic DNA was digested overnight with HindIII/Sac1 (TCR) or PciI (GFP) and separated on a 0.8% agarose (Ultrapure Agarose, Invitrogen™, Life Technologies) gel and transferred overnight to a nylon membrane (Amersham Hybond™-XL, GE Healthcare, Buckinghamshire, UK). The TCRβ probe was obtained by digesting pBR322/TCRβ2 by BglII and the GFP probe was obtained by PCR performed on pRRL.PPT.SF.IRES.GFPpre/HA plasmid with SB-GFP-FOR
Kaplan–Meier survival curves (overall survival and disease-free survival) were established using Prism 4 from GraphPad Software (La Jolla, CA). This software was also used for other statistical analyses. One-way ANOVA was used to compare more than 2 data sets. Two-way ANOVA was used to compare more than 2 data sets over time. Student t-test was used to compare 2 data sets.
We chose to utilize lentiviral vectors to over-express BCL2A1, given their more robust expression levels, higher titers, and marked decrease in insertional mutagenesis as compared to standard retrovirus vectors
BaF3 and 32Dcl3 cells are dependent on IL-3, and UT7/Epo-S1 cells are dependent on erythropoietin. When these cells are deprived of cytokines they undergo rapid apoptosis. The over-expression of murine HA-BCL2A1a or human HA-BCL2A1 did not alter the growth of BaF3 cells in the presence of IL-3 (
(A) Growth curves for BAF3 cells grown with or without murine IL-3. Viable cell numbers are shown over time. (B) Cell cycle analysis on BAF3 cells grown without IL-3. (C) Apoptosis assessed via Annexin V-PE staining in the absence of IL-3. All experiments were repeated 3 times and carried out in triplicates. Figure shows results obtained for one representative experiment. Data averages plus or minus standard deviations were plotted. HA-mBCL2A1a and HA-hBCL2A1 are respectively murine and human HA-BCL2A1.
We assessed the cell cycle status of transduced cells grown in the presence or absence of IL-3. No obvious difference in cell cycle characteristics was seen in the presence of IL-3 (data not shown). However, 1 day after IL-3 deprivation a sub-G1 population appeared. As shown in
Cells over-expressing murine BCL2A1a or human BCL2A1 were protected against apoptosis compared to MOCK or control GFP vector-transduced cells following IL-3 deprivation. Two days following removal of IL-3, almost all MOCK and vector control cells were apoptotic, compared to 50% and 20% for murine HA-BCL2A1a and human HA-BCL2A1 (p<0.0001) (
We studied the impact of over-expression of murine BCL2A1a on the behavior of primary hematopoietic stem and progenitor cells using a murine congenic transplantation model. Bone marrow cells from Ly5.1 C57BL/6 mice were transduced with lentiviruses expressing GFP alone or GFP/HA-BCL2A1a prior to reinfusion into sublethally-irradiated Ly5.2 C57BL/6 recipient mice. Five mice received untransduced donor cells, 15 mice received GFP-vector transduced cells, and 15 received GFP-murine BCL2A1a-vector transduced donor cells. From the donor cells kept in vitro we determined the fold increase expression at the mRNA level of
(A) Ratio of mRNA expression of total
Up to six months following transplantation, all mice remained healthy and with normal blood counts. However, following this time point, mice in the BCL2A1a cohort began to show symptoms of a systemic disease with weight loss, hunching, and lack of grooming, and consequently had to be euthanized. Peripheral blood counts were still in the normal range at that time (Supplemental
(A) Micrographs obtained from H&E stained slides of sternal sections or cytospins from bone marrow cells. Primary BCL2A1a mice #25 and #33 as well as secondary mouse #24-12 are represented. Sternal sections were appreciated at 20X magnification whereas blastic cells from cytospins were appreciated at 100X magnification. (B) Example of immunohistochemical staining of lymph nodes from primary BCL2A1a mouse (#33). Lymph node sections were stained with antibodies directed against antigens including B220 for B-cells, CD3 for T-cells, TdT for immature leukemic cells, lysozyme for myeloid cells, CD68 for monocytes or histiocytes. Sections were also stained with Hematoxylin and Eosin (H&E).
Kaplan-Meier overall survival and disease-free survival curves for primary transplanted mice and secondary transplanted mice (vector control or BCL2A1a): Overall survival includes deaths due to hematologic tumors and all other causes (radiation toxicity and related premature aging), with censoring of primary animals only when euthanized and utilized for secondary transplants. Overall survival curves shown in (A), (C), and (E) respectively correspond to primary mice, first set of secondary mice, and second set of secondary mice. The disease-free survival curves show development of hematologic disease (leukemia/lymphoma) as assessed by autopsy and histopathology analyses of all organs. Animals that died of non-hematologic causes were censored at their time of death. Disease-free survival curves shown in (B), (D), and (F) respectively correspond to primary mice, first set of secondary mice, and second set of secondary mice. The number of mice constituting a group is indicated between parenthesis, and median survivals are indicated in days. n.s. = not significant.
We further characterized the blast cells infiltrating tissues in BCL2A1a mice, because they were not consistently positive for standard lineage markers used for flow cytometric characterization of murine hematopoietic cells in peripheral blood and bone marrow (Supplemental
We analyzed the tumor cells at the molecular level. The possibility of T cell lineage of the blast cells was ruled out via Southern blot, which demonstrated germline configuration of the TCR genes (
(A) T-cell receptor β gene rearrangement. Genomic DNA isolated from bone marrow of three primary mice diagnosed with leukemia/lymphoma (#25, 32, and 33) and expressing HA-tagged BCL2A1a were subjected to a Southern blot analysis using a Cβ2 probe able to recognize the Cβ1 and Cβ2 region of the T-cell receptor β gene. Germline genomic DNA should produce bands of 8.9 and 2.9 Kb when digested with HindIII, and a band of 11.1 Kb when digested with SacI. NIH3T3 cells (N) were used as a control. Ladder size is given in base pairs. (B, C) Immunoglobulin gene rearrangements. PCRs to assess V to DJ rearrangements were performed with primers identifying the degenerated heavy variable region (VH558 and VH7183) as forward primers and J3 (B) or J4 (C) primers recognizing J3 or J4 genes. PCRs were performed on genomic DNA isolated from the bone marrow of control mice (MOCK #3 and vector #7), three primary mice diagnosed with leukemia/lymphoma (#25, 32, and 33), and one secondary mouse also diagnosed with leukemia (#24-12 expressing HA-tagged BCL2A1a). Genomic DNA obtained from BaF3 cells (B) and regular bone marrow from C57BL/6 mouse (M) were used as controls as well as a no-template control with water (W). Ladder size is given in base pairs, and are designated by a white dot. White arrows indicate the J gene involved in the rearrangement. The complete panel of PCRs for V to DJ and V to J rearrangements as well as germline configurations are shown in Supplemental
Southern blot performed with a GFP probe on whole bone marrow cells showed that primary BCL2A1a mice diagnosed with the lymphoid disease (#25, 26, 28, 32, and 33) did not share the same profile of integration, suggesting that different infiltrating clones were responsible for the disease (
In order to further study the behavior of BCL2A1a-transduced cells, we performed secondary transplants consisting of transplanting marrow from primary mice engrafted 6 months or more with transduced cells into secondary recipients. We chose a ratio of 1 primary for 3 secondary mice, and reinfused the cells in this experiment into non-irradiated secondary mice. All mice receiving cells from BCL2A1a primary mice died within a month post transplant, with a disease phenotype similar to the one described above for primary mice (
In our second attempt secondary mice were sublethally irradiated at 900 rads. We observed a significant difference in the overall survival between BCL2A1a and vector control mice (
Southern blot with GFP probe showed different profiles of integration for BCL2A1a secondary mice from the first cohort (#29-16, 35-18, 35-19) and the second cohort (23-13) suggesting different clones responsible for the disease (
We recently reported an adverse event in a monkey treated with genetically modified stem cells expressing GFP as a marker from a MSCV-derived retrovirus
To address whether BCL2A1 over-expression can promote development of leukemia/lymphoma, in the absence of a specific cooperating oncogene such as MYC, we used a murine bone marrow transplantation model. We used a lentiviral vector expressing HA tagged BCL2A1a along with a GFP marker to over-express this protein in hematopoietic stem and progenitor cells that were then reinfused into congenic mice. While any integrating vector can potentially induce leukemia by insertional mutagenesis alone, the risk is relatively low using lentiviral as compared to standard murine retroviral vectors, and comparison to the control GFP vector allows analysis of the impact of BCL2A1a over-expression independent of any vector effects
Our previous study
In conclusion, we determined that the over-expression of BCL2A1a in murine hematopoietic stem and progenitor cells can prevent apoptosis, enhance engraftment, and promote eventual outgrowth of fully transformed leukemic cells, primarily of the B lymphoid lineage. Our studies suggest that further investigation of BCL2A1 as a participant in hematologic transformation is indicated, and it adds
(TIF)
(TIF)
(TIF)
(TIF)
(XLSX)
We are thankful to the Flow Cytometry Core Facility at NHLBI. We thank A. Biancotto, CHI/NIH for helping with flow cytometry analyses and we thank S. Gough NCI/NIH for the help with the IgH rearrangement PCRs.