Establishment of a transgenic mouse to model ETV7 expressing human tumors
- 320 Downloads
The ETS transcription factor ETV7 has been characterized as a hematopoietic oncoprotein, which requires cooperating mutations for its leukemogenic activity. Although the ETV7 gene is highly conserved among vertebrates, part of the rodents, including Mus musculus, deleted the Etv7 gene locus. Many human hematopoietic malignancies upregulate ETV7 expression but contrary to ETV7’s role in oncogenesis, its physiological role in normal tissues is unknown. To determine the physiological function of ETV7 in vivo and determine its role in tumorigenesis in a mouse model, we have generated an ETV7 transgenic mouse that carries a single copy of human BAC DNA containing the ETV7 gene locus and its regulatory sequences. ETV7 heterozygous (ETV7Tg+/WT) mice were fertile, normal in size and born at a normal Mendelian frequency. They had a normal blood count, did not display any gross physical or behavioral abnormalities, and were not tumor-prone. The ETV7 expression pattern in hematopoietic cells of ETV7Tg+/WT mice is very similar to that in human hematopoietic cells. To examine the oncogenic potential of ETV7 in vivo, we crossed ETV7Tg+/WT mice with tumor-prone mouse models. ETV7 greatly accelerated loss of Pten (phosphatase and tensin homolog)-evoked leukemogenesis in PtenΔ/ΔETV7Tg+/WT mice after deletion of the conditional Pten allele. Consistent with this observation, ETV7 expression enhanced the colony-forming and self-renewal activities of primary myeloid Pten−/− cells. In this study we established a transgenic mouse in which we can more accurately model ETV7-associated human tumorigenesis in vivo.
KeywordsETV7 ETS transcription factor Transgenic mouse Tumor mouse model Leukemia
E26-transformation specific (ETS) transcription factors are involved in diverse biological processes including cellular proliferation, survival, differentiation, development, and transformation. We and others independently identified the ETS transcription factor ETV7, which is highly homologous to ETV6/TEL, a frequent target of chromosomal translocation in human leukemia (Fenrick et al. 2000; Poirel et al. 2000; Potter et al. 2000). Given that deletion or inactivation of ETV6 has been frequently observed in hematopoietic malignancies, ETV6 is also considered to be a tumor suppressor. In contrast, ETV7 is frequently upregulated in a variety of human cancers, including hematopoietic malignancies, in which ETV7 is overexpressed in 70% of myeloid and lymphoid leukemia. Previously we have shown that ectopic retroviral expression of ETV7 causes hematopoietic malignancies in the mouse (Cardone et al. 2005; Carella et al. 2006). More recently, we have demonstrated that morpholino knockdown of Etv7 in zebrafish leads to loss of hemoglobin-containing red blood cells by repression of the lanosterol synthase (lss) gene, indicating that in fish ETV7 is indispensable for normal red blood cell development (Quintana et al. 2014). However, the physiological and oncogenic roles of ETV7 in mammals in vivo remain to be investigated by using an appropriate mouse model.
Despite its high level of conservation among vertebrates, the Etv7 gene locus has been deleted in part of the rodents, including Mus musculus. To reverse this situation in the mouse, we have generated an ETV7 BAC transgenic mouse that carries a partial single copy of a human ETV7 BAC DNA. Like wild-type (WT) controls ETV7 heterozygous (ETV7Tg+/WT or ETV7Tg) mice develop normally, are not tumor-prone and have a normal lifespan. Importantly, the ETV7 expression pattern in hematopoietic cells of ETV7Tg+/WT mice was evaluated by qRT-PCR and was very similar to that in human hematopoietic cells, suggesting that our ETV7Tg+/WT mouse properly reflects the tissue-specific expression of human ETV7. Based on flow cytometric analysis with antibodies specific for lymphoid, myeloid, and erythroid cell types, the cellularity and distribution of hematopoietic cells in ETV7Tg BM, spleen, and thymus are similar to those in WT mice. Nonetheless, ETV7Tg BM cells proliferated faster in long-term culture, in which ETV7 enhanced proliferation of myeloid cells compared with that of control WT myeloid cells. To examine the oncogenic potential of ETV7 in vivo, we crossed ETV7Tg mice with an established leukemic mouse model. We found that ETV7 greatly accelerated PtenΔ/Δ leukemogenesis in Ptenfl/fl;Mx1-Cre;ETV7Tg+/WT mice. Thus, we created a valuable experimental animal model to investigate the mechanism of ETV7-associated human tumorigenesis in vivo. Moreover, our ETV7Tg mouse model, which faithfully recapitulates human tumors, might greatly facilitate the identification of therapeutic targets for ETV7-associated human cancer.
Materials and methods
Generation of ETV7 BAC transgenic mice
Linearized RP11-918H23 BAC DNA (BACPAC Resources Center), containing the human Etv7 gene locus, was microinjected into the pronucleus of fertilized FVB mouse oocytes. Injected zygotes were transplanted into pseudo pregnant CD1 fosters. Tail biopsies of live born offspring were used to isolate genomic DNA for genotyping, using primers specific for exon 1 and 8 of human ETV7. Samples positive for both PCRs were subjected to PCR screening of the upstream and downstream sequences of ETV7 as well as the first and last exons of all open reading frames (ORFs) present within the RP11-918H23 BAC. When ETV7 was detected in tail biopsies, a fresh biopsy was obtained and subjected to fluorescent in situ hybridization (FISH) using a FITC labeled RP11-918H23 probe, to determine copy number and potential mosaicism of the founder mice. The FISH analysis was carried out by the Cytogenetic Core of St. Jude Children’s Research Hospital performed.
Cells (5 × 106) were taken up in TRIzol Reagent (Invitrogen) and incubated at room temperature for 10 min. Chloroform (Fisher-Scientific) was added to facilitate phase separation during centrifugation. 1 μg glycogen (Invitrogen) was added to the aqueous phase and the DNA was precipitated using 2-propanol (Fisher Scientific). RNA pellets were washed with 75% ethanol and dissolved in nuclease-free water (Ambion). The RNA was quantitated using a Nanodrop spectrophotometer (Thermo Scientific).
Quantitative reverse transcriptase PCR
Total RNA (5 μg) was pretreated with DNase (Invitrogen), followed by first strand cDNA synthesis, using Oligo-dT priming and the SuperScript III First Strand Synthesis System (Invitrogen). After first strand synthesis, samples were treated with RNase. Quantitative Real Time PCR amplification was performed with 1 μL cDNA, using TaqMan Gene Expression Master Mix (Applied Biosystems). The library of tissue-specific human cDNAs was purchased from Clontech. The TaqMan probe/primers set for human Etv7 was as described previously (Kawagoe et al. 2004). 20 μL reactions were loaded in a MicroAmp Optical 96-well reaction plate (Applied Biosystems) and amplification was performed and detected using the ABI Prism 7900HT Sequence Detection System (Applied Biosystems). Samples were amplified in parallel using human or murine HPRT as internal control. The sets of TaqMan probes and primers for human HPRT were as suggested by Applied Biosystems (4326321E). The murine HPRT TaqMan probe and primers are as follows: probe (5′-CGAGCAAGTCTTTCAGTCCTGTCCA-3′), forward (5′-ATTATGCCGAGGATTTGGAA-3′), and reverse (5′-CCCATCTCCTTCATGACATCT-3′). Standard curves were generated using 5 μL of serially diluted standards with a starting concentration of 2.40 × 109 copies. Human CD19+ (B-cells), CD3+ (T-cells), CD11b+CD15+ (Granulocytes), and CD11b+CD15+ (Monocytes) were sorted from human cord blood cells (St. Louis Cord Blood Bank) using a FACS Vantage-SE DiVa cell sorter (BD Biosciences), and the individual total RNA was purified as described above.
Tissue staining with ETV7 antibody
Murine normal tissues were obtained from humanely euthanized animals and fixed in 10% neutral buffered formalin. Tissues were paraffinized, embedded, and 5 μm thick sections were cut. For anti-ETV7 antibody staining, the sections were dried overnight, and baked at 65 °C for 30 min. The sections on slides were incubated in citrate buffer at pH 6.0 (Invitrogen) for 15 min at 100 °C. After antigen retrieval, endogenous peroxidase activity was blocked by incubating the slides in 3% peroxide (Sigma) in methanol for 5 min. All following steps were intermitted by washing in TBS with 0.5% Tween-20. Endogenous biotin was blocked using an avidin/biotin-blocking kit (Vector Labs) followed by a 30-min protein-blocking step with Serum-Free protein block (Invitrogen) at 37 °C. Sections were incubated with anti-ETV7 antibody overnight at 4 °C. For peptide competition, undiluted antibody was incubated with ETV7 peptide at room temperature (RT) for 30 min prior to its application to the slides. Biotinylated secondary antibody (Vector Labs) was used at 6 μg/mL for 30 min at RT, followed by streptavidin-HRP (DAKO) and DAB chromogen (DAKO), following the manufacturer’s protocols. Images were acquired using 200x or 400 × magnification on a Nikon E800 microscope in the Cell Imaging Core Facility of St. Jude Children’s Research Hospital. The ETV7 polyclonal antibody (Cardone et al. 2005) was raised against the ETV7-C-terminal peptide (DRIEFKDKRPEISP) and affinity purified using the same peptide coupled to an agarose column. The antibody was recovered using glycine elution.
Mononuclear cells were freshly harvested from bone marrow (BM), thymus, and spleen of 8–12 week-old mice and immediately stained with the antibodies of interest; B220-eFluor780, CD43-PE, IgM-PE-Cy7, IgD-APC, Mac1-Alexa700, and Gr1-APC-Cy7 for bone marrow cells, CD4-PerCP-Cy5.5, CD8-Alexa700, CD25-APC, CD44-PE-Cy7, CD3-PE, cKit-APC-eFluor780, and Lin+ cocktail (B220, Mac1, and Gr1)-FITC for thymocytes, and B220-eFluor605, CD3-APC, Mac1-Alexa700, CD4-PE, CD8-PE-Cy7, and Gr1-APC-Cy7 for splenocytes as shown in Fig. 4. Single-cell suspensions were incubated on ice for 30 min in staining medium (SM; PBS with 5% FBS), containing 100 mg/mL human gamma globulin solution to block non-specific staining. After washing, the cells were incubated on ice for 15 min in SM containing fluorochrome-conjugated antibodies. For detection of cells undergoing apoptosis, samples were incubated at room temperature for 15 min in Annexin V binding buffer (10 mM HEPES, 0.9% NaCl, 2.5 mM CaCl2, and 0.1% BSA) containing Annexin V-FITC antibody. DAPI was used as a dead cell marker. All FCM analyses were carried out using a BD LSR II flow cytometer (BD Biosciences).
Pten fl/fl, Pten fl/fl ;Mx1-Cre and Pten fl/fl ;Mx1-Cre;ETV7Tg +/WT mice
Animals were housed in the St Jude Animal Resources Center with access to sterilized food and water ad libitum, and all experiments were approved by the Institutional Animal Care and Use Committee of St. Jude Children’s Research Hospital. B6.129S4fltm1Hwu/J (Ptenfl/fl) mice were kindly provided by Dr. Suzanne Baker. ETV7Tg+/WT on the 129/CL57/B6 (129SvEv;C57Bl/6 mixed genetic background) were generated by backcrossing ETV7Tg+/WT FVB mice onto wild-type 129/CL57/B6 mice for more than 10 generations prior to crossing them with Ptenfl/fl;Mx1-Cre. 4–6-Week-old Ptenfl/fl;Mx1-Cre and Ptenfl/fl;Mx1-Cre;ETV7Tg+/WT mice were injected intraperitoneally with seven doses of polyinosine–polycytidine (pIpC) (25 μg/g) every other day for 14 days to induce Cre expression as described previously (Yilmaz et al. 2006). Five days after pIpC injection, total BM was harvested and analyzed by colony-forming cell assays. The pIpC treated mice were observed daily and moribund mice were euthanized by CO2 inhalation. For pathologic diagnosis, the spleen, thymus, and sternum of sick mice were fixed in 10% neutral-buffered formalin and embedded in paraffin. All sections were stained with hematoxylin and eosin (H&E), and immunostained with anti-CD3 (Santa Cruz), anti-CD45R/B220 (BD Biosciences), and anti-myeloperoxidase (MPO; Dako) antibodies.
Colony-forming cell (CFC) assay
BM cells were plated in methylcellulose-based media supplemented with 10 μg/mL insulin, 200 ng/mL human transferrin, 50 ng/mL mSCF, 10 ng/mL mIL-3, 10 ng/mL hIL-6, and 3U/mL erythropoietin (MethoCult M3434, StemCell Technologies) at a density of 10,000 cells per dish. The colonies were counted and pooled 10–14 days later, and re-plated into a secondary methylcellulose culture (MC2). This procedure was repeated for 4 rounds (MC4).
ETV7 expression in normal human tissues
Generation of ETV7 BAC transgenic mice
ETV7Tg mice have a normal phenotype and lifespan
To examine the expression pattern of ETV7 in the hematopoietic system, we compared the expression level of ETV7 in hematopoietic tissues of ETV7Tg mice and humans. Murine or human B-cells, T-cells, and myeloid cells were isolated by fluorescence-activated cell sorting (FACS) from ETV7Tg splenocytes or human cord blood, respectively, and RNA of these cells was subjected to qRT-PCR analysis. As in humans, ETV7 expression was low in both B- and T-cells compared with that in myeloid cells of ETV7Tg mice (Fig. 3d). This result indicated that the ETV7 expression pattern in hematopoietic cells of ETV7Tg mimics that in humans.
ETV7Tg mice express ETV7 in hematopoietic tissues
ETV7 expression does not discernably alter hematopoietic tissues in ETV7Tg mice
ETV7 enhanced cell proliferation of myeloid cells in vitro
ETV7 accelerates development of PTENΔ/Δ T-cell lymphoblastic leukemia in mice
ETV7 has been implicated in both human and mouse hematopoietic malignancy. Elevated ETV7 expression was observed in 70% of pediatric ALL/AML patients and overexpression of ETV7 inhibited monocytic differentiation in human cell lines (Kawagoe et al. 2004). In the mouse, retroviral transduction of ETV7 in BM caused myeloproliferative disease and in combination with Myc over expression accelerated B-cell lymphoma development (Cardone et al. 2005; Carella et al. 2006). However, these studies were not intended to identify the role of endogenous ETV7 in normal development and tumorigenesis because the oncogenic capabilities of ETV7 became apparent upon forced overexpression of ETV7 in murine BM and transformed human cell lines. Indeed, qRT-PCR showed that the expression level of ETV7 driven by the MSCV promoter in BM was over 1,000,000-fold higher than that in ETV7Tg BM (data not shown). It is curious that in spite of its high level of conservation among vertebrates the Etv7 gene locus was lost in part of the rodents including the mouse. This has limited the study of both its physiological and tumorigenic function in a mammalian model in vivo. In this study, we first established a unique transgenic mouse model, which carries a single copy of the human ETV7 gene and its regulatory elements (10 kb upstream sequences, 33.7 kb representing ETV7 and 26 kb of downstream sequences). Unfortunately, we managed to only obtain a single transgenic line as the other two chimeric transgenic founders never gave germline transmission of the transgene. Given that all three founders were chimeric, we suspect that after oocyte injection the BAC might be toxic during the transient phase of gene expression before integration into the genome, resulting in only chimeric offspring. Therefore, a limitation of this study is that our data and conclusions are derived from a single transgenic mouse line.
ETV7 expression in hematopoietic tissues in ETV7Tg mice was similar to that in humans, and the hematopoietic system developed normally without any skewing of cell types, or symptoms of disease. However, once our ETV7Tg mice were crossed with oncogenic Ptenfl/fl;Mx1-Cre mice, ETV7 accelerated PTENΔ/Δ leukemia. This result demonstrates the potential of the ETV7Tg mouse to serve as a more appropriate surrogate to study ETV7-positive human malignancies.
Recently we reported a potential in vivo function of ETV7 using zebrafish (Quintana et al. 2014). ETV7 knock-down in these animals disrupted red blood cell development indirectly through inhibition of the cholesterol synthesis pathway. Based on these results we speculated that appropriately regulated but nonetheless ectopically expressed human ETV7 in the mouse might affect red blood cell development or perturb other developmental aspects of hematopoiesis. However, we could not find any alteration in adult hematopoiesis or an altered distribution of hematopoietic cells in ETV7Tg mice. This could be the result of an as yet unidentified compensatory mechanism, or alternatively, low level of strictly regulated ETV7 expression might be insufficient to induce a discernable phenotypic change. Human hematopoietic cells are able to express ETV7 without detrimental effects, in contrast to the increased expression found in a large proportion of their malignant counterpart, further underwriting the potential impact of altered expression levels of ETV7. To establish the biological significance of the ETV7Tg mice for the analysis of ETV7 in normal ontogenesis and tumorigenesis, it is crucial to isolate the ETV7-interacting factors and identify its direct or indirect transcriptional targets. Moreover, besides hematopoietic tissues, it will be interesting to investigate the physiological or tumorigenic role of ETV7 in other tissues such as colon and small intestine that showed relative high expression of ETV7 in both normal human tissues and the ETV7Tg mouse.
Although ETV7Tg mice showed no obvious phenotype in hematopoietic tissues in vivo, ETV7Tg myeloid cells grew faster than control cells in in vitro culture. This was the result of accelerated cell cycle traverse rather than inhibition of apoptosis (Fig. 6b, d, e). Since ETV7Tg mice need to acquire other genetic and/or epigenetic mutations for transformation, the enhanced cell proliferation is likely to be an essential step in the accumulation of genetic mutations, perhaps as a result of increased replicative stress. ETV7Tg primitive myeloid progenitors showed reduced colony-forming activity during serial methylcellulose assays (Fig. 6a, MC3 and MC4), suggesting that the colony-forming activity of progenitors and/or the self-renewal activity of hematopoietic stem cells (HSC) is impaired in ETV7Tg BM. The ETS transcription factor TEL/ETV6, a frequent target of chromosomal translocation in human leukemia (Bohlander 2005), is known to be a selective and essential regulator of HSC survival in mice (Hock et al. 2004). Given that ETV6 and ETV7 have opposite biological functions and can physically interact via their PNT domains (Fenrick et al. 2000; Kawagoe et al. 2004; Potter et al. 2000; Rompaey et al. 2000), it is conceivable that ETV7 interferes with ETV6’s function in HSC maintenance. However, during constitutively activated PI3K/Akt signaling due to loss of Pten, ETV7 enhanced myeloid colony-formation compared with PtenΔ/Δ control cells, suggesting that ETV7 works as a positive regulator in absence of the Pten tumor suppressor.
In addition to PTENfl/fl mice, we also crossed ETV7Tg mice with Arf−/−(Arf−/−) and Ink4aArf−/−(Ink4aArf−/−) mice to generate Arf−/−ETV7Tg+/WT (Arf−/−ETV7Tg) and Ink4aArf−/−ETV7Tg+/WT (Ink4aArf−/−ETV7Tg) double mutant mice, respectively. Intriguingly, ETV7 slightly shortened the tumor onset of Arf−/− mice but Ink4aArf−/− and Ink4aArf−/−ETV7Tg+/WT mice showed completely overlapping survival curves (Supplemental Fig. 1). All mice developed a variety of tumors as described previously (Kamijo et al. 1997; Serrano et al. 1996) and ETV7 did not affect the tumor spectrum of Arf−/− and Ink4aArf−/− mice. Nonetheless, 1 out of 26 Arf−/−ETV7Tg mice succumbed to myeloid leukemia and hemangiosarcoma, which has never been reported in Arf−/− control mice. These results suggest that ETV7 expression is functionally comparable with reduced pRb cell cycle control, which agrees with the observation that ETV7 accelerates cell cycle traverse in myeloid cells.
In conclusion, we provide evidence that our ETV7 BAC transgenic mouse model developed normally and did not show any apparent phenotype but showed increased tumor incidence when crossed onto a tumor-prone mouse background. The ETV7Tg mouse is therefore a more faithful cancer animal model to investigate ETV7-associated human tumors. We have started crossing the ETV7Tg mice with various other cancer mouse models to determine if ETV7 also accelerates tumorigenesis in those models in vivo. We believe that our ETV7Tg mouse not only enables us to more faithfully model human ETV7-associated tumorigenesis in vivo but also provides a preclinical model with which to test efficacy of future drugs directed against ETV7-positive human cancers.
We thank Drs. Richard Ashmun and Scott Perry for FCM analysis. We thank the St. Louis Cord Blood Bank for the supply of human cord blood cells. This work was supported in part by Grant RO1-72996, the Cancer Center Core Grant CA021765, and the American Lebanese Syrian Associated Charities (ALSAC).
- Kawagoe H, Potter M, Ellis J, Grosveld GC (2004) TEL2, an ETS factor expressed in human leukemia, regulates monocytic differentiation of U937 cells and blocks the inhibitory effect of TEL1 on ras-induced cellular transformation. Can Res 64:6091–6100. https://doi.org/10.1158/0008-5472.CAN-04-0839 CrossRefGoogle Scholar
- Potter MD, Buijs A, Kreider B, van Rompaey L, Grosveld GC (2000) Identification and characterization of a new human ETS-family transcription factor, TEL2, that is expressed in hematopoietic tissues and can associate with TEL1/ETV6. Blood 95:3341–3348Google Scholar
Open AccessThis article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made.