Purpose: To recapitulate the generation of cancer stem cells in the context of an intact animal using a retroviral vector capable of in vivo delivery of oncogenes to primitive endothelial and hematopoietic stem cells.

Experimental Design: Targeting of these progenitors was achieved using transgenic mice in which the avian TVA retroviral receptor was placed under the control of the stem cell leukemia (scl/tal-1) gene promoter and SCL +19 enhancer.

Results: Injection of an avian retrovirus encoding polyoma middle T (PyMT), an oncogene that transforms endothelial cells, caused rapid lethality in all SCL-TVA mice but not in control TVA(−) littermates. The infected animals exhibited hemorrhagic foci in several organs. Histopathologic analysis confirmed the presence of hemangiomas and the endothelial origin of the PyMT-transformed cells. Surprisingly, the transformed endothelial cells contained readily detectable numbers of TVA(+) cells. By contrast, normal blood vessels had very few of these cells. The presence of TVA(+) cells in the lesions suggests that the cells originally infected by PyMT retained stem cell characteristics. Further analysis showed that the tumor cells exhibited activation of the phosphatidylinositol 3-kinase/Akt and S6/mammalian target of rapamycin pathways, suggesting a mechanism used by PyMT to transform endothelial progenitors in vivo.

Conclusions: We conclude that this experimental system can specifically deliver oncogenes to vascular endothelial progenitors in vivo and cause a fatal neoplastic disease. This animal model should allow the generation of endothelial cancer stem cells in the natural environment of an immunocompetent animal, thereby enabling the recapitulation of genetic alterations that are responsible for the initiation and progression of human malignancies of endothelial origin.

There is considerable evidence that the cells that undergo the first oncogenic alteration leading to neoplasms are stem cells, which then become a reservoir of cancer stem cells. The first indication for the existence of cancer stem cells came from the analysis of leukemias (1), but similar observations were subsequently made with breast cancer, gliomas, lung carcinomas, and other neoplasms (25). Most available therapies are designed to kill the prevalent, proliferating neoplastic population. However, these treatments may leave the reservoir of genetically altered stem cells largely unaffected. To devise new therapies to kill the cancer stem cells, it would be very desirable to recapitulate the sporadic genetic alterations that occur in stem cells in the context of an intact animal, as manipulation of stem cells outside of their natural microenvironment is likely to change their properties. To do this, it is necessary to design animal models in which cancer-causing genes can be introduced into stem cells in vivo. This requires molecular approaches for gene targeting of stem cells in their natural niche.

Genetic studies have uncovered many of the transcriptional regulatory mechanisms involved in lineage commitment of stem cells in the embryo and the adult, and this has made it possible to begin using molecular genetic approaches to identify and target adult stem cells in vivo. The stem cell leukemia (SCL) factor is a helix-loop-helix transcription factor that is essential for vascular endothelial and hematopoietic stem cell (HSC) development (69). Several enhancers within the scl/tal-1 gene have been implicated in regulation of cell type–specific expression (1012). One of these enhancers, called SCL +19, has been shown to direct reporter gene expression in HSC, endothelial progenitors, mast cells, megakaryocytes, and in the common progenitor that gives rise to these lineages, a cell known as the hemangioblast (8, 13). It was recently shown that the SCL +19 enhancer is also active in osteoblasts and vascular smooth muscle progenitors, leading to the suggestion that these lineages may arise from a common progenitor that also gives rise to endothelial cells and HSC (11). The SCL promoter and SCL +19 enhancer have been proven to be a very valuable research tool that has enabled the targeted expression of transgenes in HSC. For instance, transgenic mice expressing LacZ under the control of the SCL +19 enhancer have been used to enrich for long-term repopulating HSC (7).

Avian leukosis viruses are restricted in their ability to infect cells based on the presence or absence of the retroviral TVA receptor (14). This receptor is present in avian but not mammalian cells, and this is the basis of a tissue-specific gene delivery system in which transgenic mice are engineered to express the TVA receptor under the control of tissue-specific promoters. This experimental system has been used in combination with the replication-competent avian splice (RCAS) retroviral vector, to deliver oncogenes to specific cell types in mice, in vivo. This made it possible to induce cell type–specific neoplasms in TVA transgenic mice by infection with recombinant RCAS viruses (1518). Murphy et al. have reported previously on the creation of transgenic mice in which the TVA receptor of avian leukosis virus was under the control of the SCL promoter and the SCL +19 enhancer (19). This SCL-TVA/RCAS system was used to deliver genes to HSC using recombinant RCAS viruses in vitro. It was further shown that myb-encoding avian viruses were able to transform HSC infected in vitro, but a pathologic outcome from direct injection of oncogenic avian retroviruses into these mice was not reported at that time.

In this study, we show that SCL-TVA mice can be used to deliver genes to vascular endothelial progenitors in vivo and that this results in neoplastic lesions in vascular endothelium. Polyoma middle T (PyMT) is a membrane-associated oncoprotein that can transform endothelial cells and induce hemangiomas in chickens (20) and in mice (21, 22). Using SCL-TVA mice and recombinant RCAS/PyMT viruses, we were able to efficiently deliver PyMT to vascular endothelial progenitors and found that the infected mice rapidly succumbed to a fatal hemorrhagic disease due to hemangioma formation. The ability to target genes to vascular endothelial progenitors in vivo offers an invaluable experimental system to study the functional properties of these elusive cells. This animal model may allow identification of the signaling mechanisms responsible for the conversion of endothelial stem cells to hemangioma stem cells and point to key molecular targets for therapeutic reversal of this process.

Transgenic SCL-TVA mice. The transgenic SCL-TVA/C57BL6 mice used in these studies have been described previously (19). SCL-TVA mice carry a 9-kb transgene (p6E5/TVA/3′E) in which a 950-bp TVA cDNA was cloned downstream of p6E5, a 2.8-kb murine scl promoter element, and upstream of the 5.5-kb murine 3′ enhancer from the murine scl locus (Fig. 1). Control and transgenic TVA mice were maintained at the University of Maryland Animal Facility, and all of the procedures described here have been approved by the University of Maryland Animal Care and Use Committee. Genotyping for identification of TVA(+) mice was carried out by PCR amplification of tail DNA using specific primers as we have described (17).

Fig. 1.

SCL-TVA transgene used to create SCL-TVA mice. The genomic structure of the scl gene shown here was adapted from ref. 19. The 950-bp TVA gene is flanked upstream by the 2.8-kb 6E5 scl promoter and downstream by the 5.5-kb scl 3′ enhancer. Numbered boxes, scl gene exons.

Fig. 1.

SCL-TVA transgene used to create SCL-TVA mice. The genomic structure of the scl gene shown here was adapted from ref. 19. The 950-bp TVA gene is flanked upstream by the 2.8-kb 6E5 scl promoter and downstream by the 5.5-kb scl 3′ enhancer. Numbered boxes, scl gene exons.

Close modal

Plasmid DNA. RCAS(BP), abbreviated as RCAS, is a subgroup A avian leukosis virus–derived cloning vector (23). RCAS and RCAS-PyMT were described previously (17).

Cell and virus production. The immortalized chicken fibroblast cell line DF-1 has been described previously (24). These cells were grown in DMEM/10% fetal bovine serum.

For virus production, RCAS plasmids were transfected by lipofection into DF-1 cells. Viral supernatants were filtered and concentrated by ultracentrifugation, resuspended in 1:100 of the original volume, and stored at -70°C until use. To determine RCAS virus titers, we used NIH3T3 cells engineered to express TVA retroviral receptors as the indicator cell line (17). Viral titers were determined by immunofluorescence staining of cells infected with serial dilutions of the virus as described previously (17, 25).

In vivo infection. Eight- to 12-week-old SCL-TVA mice were injected i.v. through the tail vein with 150 mg/kg 5-fluorouracil. Four days later, the mice were injected i.v. with the indicated retroviral stocks (107 IU/mL, 300 μL/mouse). The injected mice were analyzed as described in the text.

Antibodies. The rabbit polyclonal antibodies to TVA were obtained from A. Leavitt (University of California-San Francisco; ref. 26). Rat monoclonal antibodies to PyMT were obtained from CRT. Rabbit polyclonal antibodies to KDR were obtained from Rolf A. Brekken (University of Texas Southwestern Medical Center; ref. 27). Rabbit polyclonal antibodies to von Willebrand factor were purchased from DakoCytomation. Goat anti-SCL polyclonal antibody E-14 was purchased from Santa Cruz Biotechnology. Rabbit polyclonal antibody to CD133 (ab19898) was obtained from Abcam. Anti-phospho-Akt (pAkt; Ser473) rabbit monoclonal antibody (736E11), anti-pAkt (Thr308) rabbit monoclonal antibody (244F9H2), and anti-phospho-S6 (Ser235/Ser236) rabbit monoclonal antibody (91B2) were obtained from Cell Signaling Technology.

Histopathology. Liver, spleen, lung, kidney, ovaries, testicles, and other tissues were fixed with 4% paraformaldehyde and embedded in paraffin, and 5 μm sections were stained with H&E for diagnostic purposes.

Immunohistochemistry. The tissues slides were processed for immunohistochemistry as we have described previously (17). Tissue sections were incubated with the primary antibody and then incubated with the biotinylated secondary antibody (Vector Laboratories) followed by the avidin-biotin complex method (Vector Stain Elite, ABC kit; Vector Laboratories). The slides were developed in 3,3′-diaminobenzidine (Sigma FASTDAB tablet; Sigma), and the tissues were counterstained with Mayer's hematoxylin. The slides were scanned in a T3 Aperio Scanscope (Aperio Technologies), and images of specific areas were captured using the Aperio Imagescope software.

In vivo infection of SCL-TVA mice with RCAS-PyMT causes rapid lethality. SCL-TVA mice were used previously in conjunction with RCAS retroviral vectors to deliver genes to HSC. Because the regulatory elements used to engineer SCL-TVA mice can direct gene expression in endothelial progenitors, we investigated whether this experimental system can also be used to deliver oncogenes to these cells. Therefore, we injected SCL-TVA(+) mice with a RCAS retroviral vector that encodes PyMT. This oncogene was chosen because although it can efficiently transform endothelial cells, transformation of hematopoietic cells by PyMT has not been observed. SCL-TVA/C57BL mice were first injected with 5-fluorouracil to mobilize the stem cell compartment and 4 days later with a dose of 107 RCAS-PyMT virus particles per mouse. All of the SCL-TVA(+) mice became moribund or died within 2 months, whereas all of the control TVA(−) littermates showed no signs of disease after a year of observation (Fig. 2). This was true in three independent experiments. These results showed that RCAS-PyMT virus caused a rapid and fatal disease in SCL-TVA transgenic mice but not in control mice, indicating a tight control of in vivo infectivity by the presence or absence of the SCL-TVA transgene.

Fig. 2.

Survival of SCL-TVA mice infected with RCAS-PyMT virus. Kaplan-Meier plot showing survival of SCL-TVA (n = 13) versus control (n = 7) mice injected with RCAS-PyMT virus. TVA(+), SCL-TVA mice; TVA(−), control littermates.

Fig. 2.

Survival of SCL-TVA mice infected with RCAS-PyMT virus. Kaplan-Meier plot showing survival of SCL-TVA (n = 13) versus control (n = 7) mice injected with RCAS-PyMT virus. TVA(+), SCL-TVA mice; TVA(−), control littermates.

Close modal

PyMT causes hemorrhagic disease in infected SCL-TVA mice. To characterize the disease induced by PyMT, we carried out histopathologic analysis of infected animals. SCL-TVA/C57BL mice injected with RCAS-PyMT were subjected to necropsy between 5 and 9 weeks after injection. Examination at necropsy showed that injected SCL-TVA(+) mice died of internal hemorrhages. The infected TVA(+) mice had a hematocrit of 10% to 20% compared with 40% to 50% for control TVA(−) animals euthanized at the same time. In the most extreme cases, the liver was pale and the spleen was enlarged, with some animals showing four to five times the normal spleen size (Fig. 3A). Spleen sections showed the presence of erythroid precursors and megakaryocytes, evidence of extramedullary hematopoiesis to compensate for the hemorrhagic blood loss. SCL-TVA(+) mice had retroperitoneal hemangiomas and hemangiomas in the mesentery, intestine, liver, ovary, uterus, seminal vesicles, and testicles (Fig. 3B and C). Analysis of bone marrow and peripheral blood showed the presence of increased numbers of RBC progenitors but no evidence of leukemia or lymphoma.

Fig. 3.

Top, A to C, infection of SCL-TVA mice with PyMT virus causes hemorrhagic disease. A, abdominal area of PyMT-infected SCL-TVA mouse showing hemorrhagic seminal vesicle (black arrow) and vascular lesions in testicular area (empty arrow) and in the mesentery (white arrows). B, hemangiomatous lesions in ovaries (white arrows) and uterus (empty arrow). C, fragment of liver showing multiple hemorrhagic lesions. Bottom, D to F, representative H&E stain of vascular lesions found in testicular area of SCL-TVA mice injected with RCAS-PyMT virus. D, at low magnification (×1.4), the tumor is composed of irregular cystic cavities full of blood (white arrow). Black arrow, remnant testicular tissue. E and F, at higher magnification, the cavities are lined by flattened endothelial cells (black arrows in E and F). Solid sheets of the endothelial cells can also be seen with amorphous fibrin-like extracellular material (white arrow in E, ×200). Stellate-type cells can also be found as part of the proliferating transformed endothelium (white arrow in F, ×200).

Fig. 3.

Top, A to C, infection of SCL-TVA mice with PyMT virus causes hemorrhagic disease. A, abdominal area of PyMT-infected SCL-TVA mouse showing hemorrhagic seminal vesicle (black arrow) and vascular lesions in testicular area (empty arrow) and in the mesentery (white arrows). B, hemangiomatous lesions in ovaries (white arrows) and uterus (empty arrow). C, fragment of liver showing multiple hemorrhagic lesions. Bottom, D to F, representative H&E stain of vascular lesions found in testicular area of SCL-TVA mice injected with RCAS-PyMT virus. D, at low magnification (×1.4), the tumor is composed of irregular cystic cavities full of blood (white arrow). Black arrow, remnant testicular tissue. E and F, at higher magnification, the cavities are lined by flattened endothelial cells (black arrows in E and F). Solid sheets of the endothelial cells can also be seen with amorphous fibrin-like extracellular material (white arrow in E, ×200). Stellate-type cells can also be found as part of the proliferating transformed endothelium (white arrow in F, ×200).

Close modal

These results showed that in vivo delivery of PyMT causes a fatal disease characterized by extensive hemorrhagic loss, most likely a consequence of transformation of vascular endothelial cells by PyMT.

PyMT transforms vascular endothelial cells of SCL-TVA mice. To further characterize the histopathologic changes induced by PyMT infection, tissues obtained at necropsy from PyMT-infected SCL-TVA mice were stained with H&E and with antibodies to endothelial cell markers. PyMT caused benign angiomas and cavernous hemangiomas (Fig. 3D-F). The tumor shown in Fig. 3D is composed of irregular cystic cavities. At higher magnification, solid sheets of the endothelial-like cells are evident (Fig. 3E), and stellate-type cells can also be found as part of the proliferating set (Fig. 3F). The cells lining the cavities were positive for factor VIII (Fig. 4A), showing their endothelial origin. The abnormal cells lining cystic cavities were also strongly positive for PyMT (Fig. 4B), whereas no PyMT was detectable in normal vasculature (data not shown). Control TVA(−) mice injected with the same virus exhibited a normal vascular architecture in all organs examined, and the same was true for SCL-TVA mice infected with empty RCAS vector (data not shown).

Fig. 4.

Factor VIII and PyMT expression in cells lining the cystic cavities of infected SCL-TVA mice. A, staining with antibody to von Willebrand factor. The cells lining the cavities are positive for von Willebrand factor (inset, arrowhead, ×200), showing their endothelial origin. White arrow, a remaining testicular structure (×14). B, staining with anti-PyMT antibody. The neoplastic cells show moderate to strong positive staining (white arrow, ×110). Inset, arrowhead, a group of PyMT-positive cells (×200).

Fig. 4.

Factor VIII and PyMT expression in cells lining the cystic cavities of infected SCL-TVA mice. A, staining with antibody to von Willebrand factor. The cells lining the cavities are positive for von Willebrand factor (inset, arrowhead, ×200), showing their endothelial origin. White arrow, a remaining testicular structure (×14). B, staining with anti-PyMT antibody. The neoplastic cells show moderate to strong positive staining (white arrow, ×110). Inset, arrowhead, a group of PyMT-positive cells (×200).

Close modal

We conclude that infection of SCL-TVA mice with PyMT-encoding retrovirus results in the transformation of vascular endothelial cells. The hemangiomas observed were benign, indicating that additional genetic alterations may be required for progression to a full neoplastic phenotype. These results clearly show that SCL-TVA mice allow in vivo delivery of oncogenes to cells of endothelial origin and that this leads to pathologic abnormalities caused by hemangioma formation.

Target of RCAS-PyMT infection in SCL-TVA mice are vascular endothelial progenitors. A major question raised by the phenotype we observed is the identity of the cells that were originally infected by the PyMT virus. To begin addressing this question, we stained tissues of infected SCL-TVA mice with antibodies to KDR/VEGFR2, TVA, SCL, and CD133/prominin. A high percentage of the transformed endothelial cells showed strong reactivity for KDR, which is considered a marker of proliferating endothelial cells (Fig. 5A; ref. 27). As shown in Fig. 5B, among the differentiated endothelial cells in normal vessels, there were very few, if any, TVA(+) cells. However, to our surprise, there were readily detectable TVA(+) cells in the transformed endothelium. Because the tissue specificity of the SCL enhancer restricts expression of TVA to early hematopoietic and endothelial progenitors, this population of TVA(+) cells is likely to have retained characteristics of stem cells and may represent hemangioma stem cells. Furthermore, the proliferating transformed endothelium expressed endogenous SCL, as determined by direct staining with anti-SCL antibodies (Fig. 4D), and these focal areas also showed immunoreactivity for CD133, which is a marker of vascular endothelial progenitors and other stem cells (Fig. 5C; ref. 28). Based on these results, we tentatively conclude that the primary target of the PyMT virus may have been TVA(+) endothelial progenitors and that these cells were capable of undergoing partial differentiation to give rise to the benign hemangiomas we observed. Further analysis will be required to ascertain the precise origin, phenotype, and functional properties of these cells.

Fig. 5.

PyMT-induced lesions contain clusters of TVA-expressing cells. A, staining with antibody to KDR/VEGFR2 in vascular lesions and normal vasculature. KDR is expressed in a high percentage of the endothelial cells; most show strong reactivity (black arrow, ×110; black arrowhead in inset, ×200), with a few isolated cells remaining negative (white arrowhead). B, staining with anti-TVA antibody. White arrows, blood-filled cavities with negative staining. Even at this low magnification (×115), positive staining is readily detected in the proliferating cells in the lesion (black arrow). At higher magnification (top inset, ×200), positive staining for TVA is evident in the endothelial proliferating cells (black arrowhead), whereas a few cells remain unreactive (white arrowhead). Nonneoplastic vessels in the vicinity of the tumor show only weak and focal staining or remain negative (black arrowheads, bottom inset). C, CD133 staining. Arrow, a group of strongly positive transformed endothelial cells in the hemangiomatous lesion (×110). Inset, a normal vessel where endothelial cells are negative for CD133. Arrow, nucleus of a normal endothelial cell (×200). D, staining with anti-SCL antibody. Arrows, positive cells within the proliferating endothelium in vascular lesion (×110). Inset, a normal venous structure that is close to the lesion. The normal endothelium (arrow) remains negative (×200).

Fig. 5.

PyMT-induced lesions contain clusters of TVA-expressing cells. A, staining with antibody to KDR/VEGFR2 in vascular lesions and normal vasculature. KDR is expressed in a high percentage of the endothelial cells; most show strong reactivity (black arrow, ×110; black arrowhead in inset, ×200), with a few isolated cells remaining negative (white arrowhead). B, staining with anti-TVA antibody. White arrows, blood-filled cavities with negative staining. Even at this low magnification (×115), positive staining is readily detected in the proliferating cells in the lesion (black arrow). At higher magnification (top inset, ×200), positive staining for TVA is evident in the endothelial proliferating cells (black arrowhead), whereas a few cells remain unreactive (white arrowhead). Nonneoplastic vessels in the vicinity of the tumor show only weak and focal staining or remain negative (black arrowheads, bottom inset). C, CD133 staining. Arrow, a group of strongly positive transformed endothelial cells in the hemangiomatous lesion (×110). Inset, a normal vessel where endothelial cells are negative for CD133. Arrow, nucleus of a normal endothelial cell (×200). D, staining with anti-SCL antibody. Arrows, positive cells within the proliferating endothelium in vascular lesion (×110). Inset, a normal venous structure that is close to the lesion. The normal endothelium (arrow) remains negative (×200).

Close modal

Targeted expression of PyMT in vascular endothelial progenitors activates pathways involved in cell survival and proliferation. To delineate the mechanisms by which PyMT may have transformed endothelial cells in vivo, we examined several downstream signaling pathways. As shown in Fig. 4, PyMT was expressed in the majority of the endothelial cells present in the lesions. This suggests that PyMT transformed cells by acting directly on intracellular targets rather than paracrine mechanisms. This is in contrast to cell transformation by v-GPCR, an oncogene of herpesvirus-8 that transforms endothelial cells by paracrine mechanisms (17, 29). Because PyMT has been shown to activate phosphatidylinositol 3-kinase (PI3-K) in vascular tumors (30), we investigated whether the PI3-K/Akt survival pathway was activated in the PyMT-induced vascular lesions of SCL-TVA mice. To address this question, we carried out immunohistochemical analysis using antibodies to pAkt Thr308 and pAkt Ser473. Figure 6A and B show that both phosphorylated forms of Akt were expressed in the transformed endothelial cells. Strong pAkt Thr308 staining was seen in the cytoplasm of proliferating endothelial cells in the lesions, and the same was true for pAkt Ser473 immunoreactivity. Thus, PyMT transformation of endothelial cells involved activation of the PI3-K pathway. The mammalian target of rapamycin (mTOR) pathway has been established as a major effector of Akt in the transformation of endothelial cells (31, 32). Therefore, we examined the state of activation of mTOR in the affected tissues by assessing the level of phosphorylation of S6 ribosomal protein, which is a substrate of p70 S6 kinase. Immunohistochemical analysis showed strong immunoreactivity for phospho-S6 in most proliferating cells (Fig. 6C). This suggests that mTOR and its downstream effectors played a role in PyMT transformation of endothelial cells in vivo.

Fig. 6.

PyMT-infected endothelial cells in tumor areas of SCL-TVA mice have elevated levels of phospho-Akt and phospho-S6. A, staining with anti-pAkt Thr308 antibodies in vascular lesions and normal vasculature. Arrow, cells that are positive for pAkt Thr308 (×110). Top inset, arrowhead, pAkt Thr308-positive cells (×200); bottom inset, normal endothelium showing low immunoreactivity for pAkt Thr308. B, staining with anti- pAkt Ser473 antibodies in lesions. Top inset, arrowhead, strong pAkt Ser473 staining in vessel structures; bottom inset, arrowhead, mild staining in normal endothelial cells of unaffected vasculature. C, tissues from affected and normal areas were stained with anti-phospho-S6 antibodies. Arrow, proliferating cells with strong immunoreactivity for pS6 (×110). Top inset, arrowhead, pS6-positive cells in vascular structure (×200); bottom inset, normal endothelium showing low pS6 immunoreactivity.

Fig. 6.

PyMT-infected endothelial cells in tumor areas of SCL-TVA mice have elevated levels of phospho-Akt and phospho-S6. A, staining with anti-pAkt Thr308 antibodies in vascular lesions and normal vasculature. Arrow, cells that are positive for pAkt Thr308 (×110). Top inset, arrowhead, pAkt Thr308-positive cells (×200); bottom inset, normal endothelium showing low immunoreactivity for pAkt Thr308. B, staining with anti- pAkt Ser473 antibodies in lesions. Top inset, arrowhead, strong pAkt Ser473 staining in vessel structures; bottom inset, arrowhead, mild staining in normal endothelial cells of unaffected vasculature. C, tissues from affected and normal areas were stained with anti-phospho-S6 antibodies. Arrow, proliferating cells with strong immunoreactivity for pS6 (×110). Top inset, arrowhead, pS6-positive cells in vascular structure (×200); bottom inset, normal endothelium showing low pS6 immunoreactivity.

Close modal

In this study, we show that the SCL-TVA/RCAS system can be used for the targeted delivery of genes to vascular endothelial cells in vivo and that introduction of the PyMT oncogene leads to tumor formation within 2 months of infection. We also identified the PI3-K and mTOR pathways as major downstream targets of PyMT-dependent transformation in endothelial progenitors.

Previous studies using SCL-TVA mice to assess the possibility of delivering genes to early hematopoietic progenitors showed that these cells could be specifically targeted with RCAS viruses (19). It was shown that infection of bone marrow or fetal liver cells from SCL-TVA mice with RCAS-GFP resulted in selective infection of CD34+/c-kit+/Sca-1+ cells, indicating that cells with a HSC phenotype could be readily targeted using this system (19). When these GFP-expressing cells were used to reconstitute lethally irradiated mice, spleen colonies had a majority of GFP(+) cells, suggesting that the transplanted cells had the functional properties of HSC. Furthermore, injection of RCAS-GFP virus-producing DF-1 cells into SCL-TVA mice resulted in the recovery of a small percentage of GFP(+) cells with characteristics of early progenitors 6 months after injection into recipients (19). It was also shown that infection of TVA(+) bone marrow or fetal liver cells with avian replication-defective E26 and AMV resulted in the transformation of HSC in vitro. These studies showed the ability of the RCAS/TVA system to deliver genes to HSC. We now show that this experimental system also has applications for specific gene delivery to vascular endothelial progenitors in vivo.

The recombinant RCAS viruses we used in our studies were replication competent in avian but not mammalian cells, as the latter do not allow production of the gag, pol, and env genes required for replication (18). Therefore, the cells that were originally infected were transformed by PyMT but produced no new virus particles, and there was no viremia. Thus, the rapid tumor development we observed was likely due to the oncogenic effects of PyMT and not the result of promoter insertion mechanisms (33).

SCL-TVA mice injected with RCAS-PyMT developed hemangiomas affecting several visceral organs. There were retroperitoneal hemangiomas and cavernous hemangiomas in mesentery, intestine, liver, ovary, uterus, seminal vesicles, and testicles. The disease was fatal due to hemorrhagic blood loss, with histologic evidence of extramedullary hematopoiesis in the spleen to compensate for the blood loss. In some cases, the spleen was considerably enlarged, whereas the liver was pale. Analysis of bone marrow and peripheral blood from affected mice showed the presence of increased numbers of RBC progenitors, but there was no evidence of leukemia or lymphoma. Considering that the cells originally infected with PyMT were likely to be early progenitors, which are present in very small numbers, the celerity of the disease points to a very efficient gene delivery system. This experimental system may be useful to model human pediatric hemangiomas, which are believed to be derived from embryonic hemanangioblasts (34, 35). In fact, a high percentage of cells in human hemangioblastomas express SCL (34), which is consistent with the expression of SCL and TVA we observed in SCL-TVA hemangiomas. Furthermore, some of the lesions developed in SCL-TVA mice resemble those seen in Kasabach-Meritt syndrome, first described in an infant with a vascular anomaly that included microangiopathic hemolytic anemia, a consumptive coagulopathy, thrombocytopenia, and an enlarging vascular lesion (36, 37). Our findings suggest that some pediatric and adult human hemangiomas may result from genetic alterations in the hemangioblast stem cell compartment, and our animal model may help to test new therapies for these abnormalities and identify new angiogenesis inhibitors.

Many studies have shown that PyMT can transform epithelial and endothelial cells leading to hemangiosarcomas, mammary tumors, and other neoplasms of epithelial origin (17, 20, 22, 38). However, leukemia or lymphoma are not a reported consequence of PyMT expression in animal models. Because of the known specificity of the SCL promoter and 3′ enhancer used to drive TVA expression (1012), PyMT infection was largely circumscribed to hematopoietic and endothelial cell progenitors. Immunohistochemistry analysis showed that the cells transformed by PyMT were of endothelial origin. Because in SCL-TVA mice HSC also express TVA (19), and these cells were likely to have also been infected by the RCAS-PyMT virus, our data strongly suggest that HSC may not be very susceptible to transformation by PyMT. Although HSC and angioblasts are believed to be derived from a common progenitor, our observations suggest that the networks that regulate cell proliferation in these two lineages are different. It should be noted that when Bcr-Abl was placed under control of the same SCL promoter and 3′ enhancer used in our studies, the outcome of Bcr-Abl expression in an inducible transgenic mouse model was HSC transformation, leading to a disease very similar to CML (39). Bcr-Abl is known to cause leukemia and lymphoma, but hemangioma formation has not been reported (40). This is also in agreement with the idea that HSC and endothelial progenitors have different susceptibilities to transforming genetic alterations, and that this is the result of differences in signaling pathways that regulate survival and proliferation in these cell types.

Many of the cells in the PyMT-induced lesions expressed factor VIII and KDR/VEGFR2, which are markers of endothelial cells. Most surprisingly, these lesions contained readily detectable SCL(+) and TVA(+) cells, in contrast to normal vascular endothelium, where very little, if any, SCL and TVA immunostaining was detectable. The hemangioma lesions were also positive for the stem cell marker CD133, suggesting that they retained characteristics of immature endothelial cells. It is tempting to speculate that the TVA(+) cells in the PyMT lesions represent cancer stem cells, and that some of these PyMT-transformed angioblasts were able to at least partially differentiate into abnormal endothelial cells. The possibility that this group of TVA(+) cells are hemangioma stem cells is one of the most interesting observations of our study, and this question will be the subject of future investigation.

Montaner et al. have used mice in which the endothelial-specific tie2 promoter was used to drive TVA expression (17). In this experimental system, RCAS-PyMT infection caused lethality due to hemorrhage and hemangiomas within days compared with weeks in our SCL-TVA model. This may be due to the fact that tie2 is expressed in most endothelial cells, offering a larger pool of infectable cells. The difference in pathogenicity observed may also be related to different susceptibility of transformation depending on the state of maturation of the endothelial cells at the time of PyMT infection.

Our immunohistochemistry analysis showed that most of the endothelial cells in the hemangiomas expressed PyMT, indicating that tumor formation was probably the result of the direct action of this oncogene on intracellular targets. To delineate the possible mechanism by which PyMT caused primary hemangiomas, we analyzed pathways known to be important in endothelial cell transformation. Previous studies have implicated PI3-K/Akt in the mechanism of transformation by PyMT in vitro and in vivo (41, 42). Immunohistochemistry analysis of PyMT hemangiomas showed that Akt was highly phosphorylated on Ser473 and Thr308. This suggests that the PI3-K/Akt pathway may be involved in establishment and maintenance of tumors that originate from endothelial precursors. S6 ribosomal protein, a substrate of S6 kinase, was highly phosphorylated in hemangiomas, implicating the mTOR pathway in PyMT-induced tumorigenesis. This is a major effector of Akt-dependent tumorigenesis whose inhibition by rapamycin can reverse hemangiomas caused by v-GPCR expression in the in vivo tie2-TVA model of Kaposi's sarcoma (43). In our SCL-TVA model, the target of PyMT transformation is likely to be an earlier endothelial progenitor than in the tie2-TVA model, and it would be interesting to test if mTOR inhibitors might also reverse the phenotype of uncontrolled angiogenesis that we observed in SCL-TVA mice. Due to the similarity of the lesions in SCL-TVA mice and those observed in Kasabach-Meritt syndrome, our experimental system might also be useful as a model for testing new therapies aimed at treatment of pediatric and adult hemangiomas.

In conclusion, in this article, we describe the use of the SCL-TVA experimental system to model tumor formation by direct injection of an avian retrovirus encoding an oncogene into SCL-TVA mice. This animal model was originally designed to target HSC. Here, we show that it can efficiently deliver genes to endothelial progenitors capable of differentiation into vascular endothelial cells. The potential of this system for delivering genes to hemangioblasts represents an invaluable tool to elucidate the critical steps that specify lineage commitment to the hematopoietic and endothelial pathways, and will facilitate the study of angiogenesis from its earliest endothelial progenitors in an in vivo setting. It may also provide a preclinical model for antiangiogenic therapy.

The authors state that there are no potential conflicts of interest.

Grant support: Department of Defense grant W81XWH-04-1-0801 and University of Maryland Greenebaum Cancer Center Pilot Award (R.A. Feldman).

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

Note: Current address for N. Takebe: Division of Cancer Treatment and Diagnosis, National Cancer Institute, Rockville, MD.

We thank A. Leavitt and Rolf A. Brekken for generously providing us with purified rabbit IgG against TVA and KDR, respectively.

1
Reya T, Morrison SJ, Clarke MF, Weissman IL. Stem cells, cancer, and cancer stem cells.
Nature
2001
;
414
:
105
–11.
2
Bruggeman SW, Hulsman D, Tanger E, et al. Bmi1 controls tumor development in an ink4a/arf-independent manner in a mouse model for glioma.
Cancer Cell
2007
;
2
:
328
–41.
3
Dalerba P, Dylla SJ, Park IK, et al. Phenotypic characterization of human colorectal cancer stem cells.
Proc Natl Acad Sci U S A
2007
;
104
:
10158
–63.
4
Dick JE. Breast cancer stem cells revealed.
Proc Natl Acad Sci U S A
2003
;
100
:
3547
–9.
5
Kim CF, Jackson EL, Woolfenden AE, et al. Identification of bronchioalveolar stem cells in normal lung and lung cancer.
Cell
2005
;
121
:
823
–35.
6
Begley CG, Green AR. The SCL gene: from case report to critical hematopoietic regulator.
Blood
1999
;
93
:
2760
–70.
7
Chen CZ, Li M, de Graaf D, et al. Identification of endoglin as a functional marker that defines long-term repopulating hematopoietic stem cells.
Proc Natl Acad Sci U S A
2002
;
99
:
15468
–73.
8
D'Souza SL, Elefanty AG, Keller G. SCL/Tal-1 is essential for hematopoietic commitment of the hemangioblast but not for its development.
Blood
2005
;
105
:
3862
–70.
9
Sanchez MJ, Bockamp EO, Miller J, Gambardella L, Green AR. Selective rescue of early haematopoietic progenitors in Scl(-/-) mice by expressing Scl under the control of a stem cell enhancer.
Development
2001
;
128
:
4815
–27.
10
Gottgens B, Broccardo C, Sanchez MJ, et al. The SCL +18/19 stem cell enhancer is not required for hematopoiesis: identification of a 5′ bifunctional hematopoietic-endothelial enhancer bound by Fli-1 and Elf-1.
Mol Cell Biol
2004
;
24
:
1870
–83.
11
Pimanda JE, Silberstein L, Dominici M, et al. Transcriptional link between blood and bone: the stem cell leukemia gene and its +19 stem cell enhancer are active in bone cells.
Mol Cell Biol
2006
;
26
:
2615
–25.
12
Silberstein L, Sanchez MJ, Socolovsky M, et al. Transgenic analysis of the stem cell leukemia +19 stem cell enhancer in adult and embryonic hematopoietic and endothelial cells.
Stem Cells
2005
;
23
:
1378
–88.
13
Choi K, Kennedy M, Kazarov A, Papadimitriou JC, Keller G. A common precursor for hematopoietic and endothelial cells.
Development
1998
;
125
:
725
–32.
14
Federspiel MJ, Bates P, Young JA, Varmus HE, Hughes SH. A system for tissue-specific gene targeting: transgenic mice susceptible to subgroup A avian leukosis virus-based retroviral vectors.
Proc Natl Acad Sci U S A
1994
;
91
:
11241
–5.
15
Fisher GH, Orsulic S, Holland E, et al. Development of a flexible and specific gene delivery system for production of murine tumor models.
Oncogene
1999
;
18
:
5253
–60.
16
Holland EC, Varmus HE. Basic fibroblast growth factor induces cell migration and proliferation after glia-specific gene transfer in mice.
Proc Natl Acad Sci U S A
1998
;
95
:
1218
–23.
17
Montaner S, Sodhi A, Molinolo A, et al. Endothelial infection with KSHV genes in vivo reveals that vGPCR initiates Kaposi's sarcomagenesis and can promote the tumorigenic potential of viral latent genes.
Cancer Cell
2003
;
3
:
23
–36.
18
Orsulic S, Li Y, Soslow RA, Vitale-Cross LA, Gutkind JS, Varmus HE. Induction of ovarian cancer by defined multiple genetic changes in a mouse model system.
Cancer Cell
2002
;
1
:
53
–62.
19
Murphy GJ, Gottgens B, Vegiopoulos A, et al. Manipulation of mouse hematopoietic progenitors by specific retroviral infection.
J Biol Chem
2003
;
278
:
43556
–63.
20
Kornbluth S, Cross FR, Harbison M, Hanafusa H. Transformation of chicken embryo fibroblasts and tumor induction by the middle T antigen of polyomavirus carried in an avian retroviral vector.
Mol Cell Biol
1986
;
6
:
1545
–51.
21
Bautch VL, Toda S, Hassell JA, Hanahan D. Endothelial cell tumors develop in transgenic mice carrying polyoma virus middle T oncogene.
Cell
1987
;
51
:
529
–37.
22
Williams RL, Courtneidge SA, Wagner EF. Embryonic lethalities and endothelial tumors in chimeric mice expressing polyoma virus middle T oncogene.
Cell
1988
;
52
:
121
–31.
23
Federspiel MJ, Hughes SH. Retroviral gene delivery.
Methods Cell Biol
1997
;
52
:
179
–214.
24
Schaefer-Klein J, Givol I, Barsov EV, et al. The EV-O-derived cell line DF-1 supports the efficient replication of avian leukosis-sarcoma viruses and vectors.
Virology
1998
;
248
:
305
–11.
25
Kim J, Feldman RA. Activated Fes protein tyrosine kinase induces terminal macrophage differentiation of myeloid progenitors (U937 cells) and activation of the transcription factor PU.1.
Mol Cell Biol
2002
;
22
:
1903
–18.
26
Murphy GJ, Leavitt AD. A model for studying megakaryocyte development and biology.
Proc Natl Acad Sci U S A
1999
;
96
:
3065
–70.
27
Feng D, Nagy JA, Brekken RA, et al. Ultrastructural localization of the vascular permeability factor/vascular endothelial growth factor (VPF/VEGF) receptor-2 (FLK-1, KDR) in normal mouse kidney and in the hyperpermeable vessels induced by VPF/VEGF-expressing tumors and adenoviral vectors.
J Histochem Cytochem
2000
;
48
:
545
–56.
28
Salven P, Mustjoki S, Alitalo R, Alitalo K, Rafii S. VEGFR-3 and CD133 identify a population of CD34+ lymphatic/vascular endothelial precursor cells.
Blood
2003
;
101
:
168
–72.
29
Pati S, Cavrois M, Guo HG, et al. Activation of NFkB by the human herpesvirus-8 chemokine receptor ORF74: evidence for a paracrine model of Kaposi's sarcoma pathogenesis.
J Virol
2001
;
75
:
8660
–73.
30
Dong QG, Graziani A, Garlanda C, et al. Anti-tumor activity of cytokines against opportunistic vascular tumors in mice.
Int J Cancer
1996
;
65
:
700
–8.
31
Manning BD, Cantley LC. United at last: the tuberous sclerosis complex gene products connect the phosphoinositide 3-kinase/Akt pathway to mammalian target of rapamycin (mTOR) signalling.
Biochem Soc Trans
2003
;
31
:
573
–8.
32
Richardson CJ, Schalm SS, Blenis J. PI3-kinase and TOR: PIKTORing cell growth.
Semin Cell Dev Biol
2004
;
15
:
147
–59.
33
Hayward WS, Neel BG, Astrin SM. Activation of a cellular onc gene by promoter insertion in ALV-induced lymphoid leukosis.
Nature
1981
;
290
:
475
–80.
34
Glasker S, Li J, Xia JB, et al. Hemangioblastomas share protein expression with embryonal hemangioblast progenitor cell.
Cancer Res
2006
;
66
:
4167
–72.
35
Regnier S, Dupin N, Le Danff C, Wassef M, Enjolras O, Aractingi S. Endothelial cells in infantile haemangiomas originate from the child and not from the mother (a fluorescence in situ hybridization-based study).
Br J Dermatol
2007
;
157
:
158
–60.
36
Kasabach HH, Merrit KK. Capillary hemangioma with extensive purpura.
Am J Dis Child
1940
;
59
:
1063
–70.
37
Verheul HM, Panigrahy D, Flynn E, Pinedo HM, D'Amato RJ. Treatment of the Kasabach-Merritt syndrome with pegylated recombinant human megakaryocyte growth and development factor in mice: elevated platelet counts, prolonged survival, and tumor growth inhibition.
Pediatr Res
1999
;
46
:
562
–5.
38
Cuevas BD, Winter-Vann AM, Johnson NL, Johnson GL. MEKK1 controls matrix degradation and tumor cell dissemination during metastasis of polyoma middle-T driven mammary cancer.
Oncogene
2006
;
25
:
4998
–5010.
39
Koschmieder S, Gottgens B, Zhang P, et al. Inducible chronic phase of myeloid leukemia with expansion of hematopoietic stem cells in a transgenic model of BCR-ABL leukemogenesis.
Blood
2005
;
105
:
324
–34.
40
Tefferi A, Gilliland DG. Oncogenes in myeloproliferative disorders.
Cell Cycle
2007
;
6
:
550
–66.
41
Fukui Y, Kornbluth S, Jong SM, Wang LH, Hanafusa H. Phosphatidylinositol kinase type I activity associates with various oncogene products.
Oncogene Res
1989
;
4
:
283
–92.
42
Whitman M, Kaplan DR, Schaffhausen B, Cantley L, Roberts TM. Association of phosphatidylinositol kinase activity with polyoma middle-T competent for transformation.
Nature
1985
;
315
:
239
–42.
43
Sodhi A, Chaisuparat R, Hu J, et al. The TSC2/mTOR pathway drives endothelial cell transformation induced by the Kaposi's sarcoma-associated herpesvirus G protein-coupled receptor.
Cancer Cell
2006
;
10
:
133
–43.