Original Paper

Oncogene (2005) 24, 3544–3553. doi:10.1038/sj.onc.1208399 Published online 31 January 2005

T-cell lymphomas mask slower developing B-lymphoid and myeloid tumours in transgenic mice with broad haemopoietic expression of MYC

Darrin P Smith1, Mary L Bath1, Alan W Harris1 and Suzanne Cory1

1The Walter and Eliza Hall Institute of Medical Research, 1G Royal Parade, Melbourne, Victoria 3050, Australia

Correspondence: S Cory, E-mail: cory@wehi.edu.au

Received 20 October 2004; Accepted 22 November 2004; Published online 31 January 2005.



Deregulation of MYC expression occurs in many haematological malignancies. Previous studies modelling MYC-induced lymphomagenesis in the mouse used transgenic vectors that directed MYC overexpression in a lineage-specific manner. Here, we describe a transgenic mouse strain in which constitutive MYC expression is driven broadly in haemopoiesis by a vector containing regulatory elements of the Vav gene. Healthy young VavP-MYC17 mice had multiple haemopoietic abnormalities, most notably increased size and numbers of B-lymphoid cells, monocytes and megakaryocytes. The mice rapidly developed tumours and, surprisingly, these were exclusively T-cell lymphomas, mostly of mature CD4+ CD8- T cells, a tumour type that is seldom seen in mouse models. To examine tumour development in the absence of the susceptible T cells, we bred VavP-MYC17 mice lacking the Rag1 recombinase. They survived longer and succumbed to tumours of several different haemopoietic cell types: pre-T cells, pro-B cells, macrophages and unusual progenitor cells. Thus, although T-lineage cells have the shortest latent period to transformation, the VavP-MYC17 transgene drives malignant transformation of multiple cell types and VavP-MYC17 mice provide a new model for tumours of multiple haemopoietic lineages.


MYC, transgenic, lymphoma, haemopoiesis, progenitor



The transcription factor Myc plays a central role in controlling cell size, proliferation and differentiation (reviewed by Grandori et al., 2000; Boxer and Dang, 2001). Expression of Myc is tightly cell-cycle regulated and, in response to cytokines or growth factors, Myc upregulation in G1 modulates the expression of an array of genes that promote cell growth and transition into S phase. Myc also sensitizes cells to diverse proapoptotic stimuli, a mechanism that is believed to limit the life of a cell with deregulated, potentially tumorigenic Myc expression (Evan and Littlewood, 1998).

The involvement of MYC in human cancer was first established in Burkitt lymphoma where its constitutive expression in B cells results from a reciprocal chromosome translocation linking the MYC gene to an immunoglobulin locus (reviewed by Cory, 1986). Subsequently, direct deregulation of MYC by translocation, mutation or amplification, or its overexpression as a result of mutations affecting upstream regulatory pathways, has been found in a variety of other malignancies. These include promyelocytic, B-cell acute lymphocytic, granulocytic, T-cell prolymphocytic and T-cell acute lymphoblastic leukaemias, plasma-cell neoplasms, large B-cell lymphoma (Marcu et al., 1992; Maljaie et al., 1995; Shou et al., 2000; Sanchez-Beato et al., 2003), and other cancers, such as lung, breast and colon carcinoma (Grandori et al., 2000; Boxer and Dang, 2001).

Transgenic tumour models have been used to investigate the mechanisms of tumorigenesis, to identify mutations that cooperate during tumour development and as tools to study responses to chemotherapy (Hann and Balmain, 2001). The Emu-Myc transgenic mouse (Adams et al., 1985; Harris et al., 1988), which models the Burkitt chromosome translocation by expressing Myc under the control of the Igh enhancer, has provided particular insight into Myc-induced tumorigenesis. In healthy young Emu-Myc mice, the proliferative effects of Myc drive the polyclonal expansion pre-B- and B-lymphocyte populations, but apoptosis restricts cell numbers as cytokine levels become limiting (Langdon et al., 1986; Adams et al., 1999). Progression to full malignancy is dependent on disabling such apoptosis. Within a few weeks or months after birth, the mice develop clonal pre-B-cell or B-cell tumours as the Myc-expressing cells acquire somatic mutations, some of which negate the proapoptotic function of Myc by inactivating the p53-p19Arf-Mdm2 axis (Eischen et al., 1999). Furthermore, lymphomagenesis in Emu-Myc mice is markedly accelerated by overexpression of the antiapoptotic Bcl2 oncogene (Strasser et al., 1990) or by inactivation of proapoptotic genes such as those encoding the tumour suppressors p53, p19Arf or the Bcl2 antagonist Bim (Eischen et al., 1999; Schmitt et al., 1999; Egle et al., 2004). Cooperating oncogenes have also been identified in retroviral insertional mutagenesis screens in Emu-Myc mice, including the antiapoptotic Bmi1 gene that represses p19Arf and p16Ink4a expression (Mikkers and Berns, 2003). Inactivation of the apoptotic programme controlled by p53 and p19Arf, or Bcl2 overexpression, compromises the response of Emu-Myc tumours to chemotherapy (Schmitt et al., 1999, 2000) and analysis of Emu-Myc tumours with defined antiapoptotic lesions is leading to an understanding of how tumour genotype can influence treatment outcome (Schmitt, 2003).

The VavP transgenic vector incorporates regulatory elements of the Vav gene and drives stable expression in every nucleated haemopoietic cell type analysed, including both differentiated cells and progenitor cells (Adams et al., 1999; Ogilvy et al., 1999a, 1999b). Transgenic mice in which MYC expression is driven by the VavP vector will provide resources to compare MYC targets and susceptibility to malignant transformation in the various haemopoietic lineages, and will generate new, multilineage models of MYC-induced leukaemogenesis. Here the complex pretumour phenotype of a VavP-MYC transgenic mouse line is described, and the multilineage tumorigenic impact of the MYC transgene examined.



Multilineage haemopoietic expression of transgenic MYC

Transgenic mice were generated by pronuclear microinjection of a human MYC cDNA into a Vav enhancer/promoter vector (Figure 1a). Of the five primary transgenic mice (see Materials and methods), two succumbed to very early onset lymphoma, and sperm from one of them was used for in vitro fertilization to generate the line described here, designated VavP-MYC17. As anticipated, the VavP-MYC17 transgene was expressed in both lymphoid and myeloid cells. Western blot analysis revealed overexpressed MYC protein in thymocytes, splenocytes and bone marrow cells from healthy young mice and in sorted populations of pre-B cells (B220+ IgM-), B cells (B220+ IgM+), monocytes (Mac1+ Gr1-) and granulocytes (Mac1+ Gr1+), the expression being lower in myeloid than in lymphoid cells (Figure 1b).

Figure 1.
Figure 1 - Unfortunately we are unable to provide accessible alternative text for this. If you require assistance to access this image, please contact help@nature.com or the author

Multilineage transgene expression in VavP-MYC17 mice promotes increased cell size. (a) The VavP-MYC transgene. The human MYC coding region was inserted into an expression vector containing regulatory elements from both the 5' untranslated region of the Vav gene (Vav 5') and intron 1 (Vav intron 1), a noncoding portion of Vav exon 1 (E1), a truncated SV40 late intron (int), and the SV40 late polyadenylation signal (pA). (b) Widespread haemopoietic expression of transgenic MYC protein. Western blots of lysates prepared from thymocytes, splenocytes, bone marrow cells and sorted pre-B cells (B220+ IgM-), B cells (B220+ IgM+), monocytes (Mac1+ Gr1-) and granulocytes (Mac1+ Gr1+) from 2-week-old VavP-MYC17 and B6 (wild type) littermates were probed with an anti-MYC antibody. The same filters probed with anti-beta-actin are shown as a loading control. (c) The size of multiple haemopoietic cell types was increased in VavP-MYC17 mice. Gated splenic B cells (B220+ IgM+), bone marrow pre-B cells (B220+ IgM-), thymic pre-T (CD4+ CD8+) and CD4+ CD8- T cells, blood monocytes (Gr1- Mac1+) and bone marrow granulocytes (Gr1+ Mac1+) from 2-week-old VavP-MYC17 mice (solid lines) and wild-type littermate controls (broken lines) were compared by forward light scatter analysis, using flow cytometry. Representative examples from greater than or equal to3 mice examined are shown. (d) Megakaryocytes (arrowheads) in the spleen of 2-week-old VavP-MYC17 mice were larger, more abundant, contained a more complex nucleus and were frequently mitotic (see cell indicated by *). H and E-stained sections, bar=20 mum

Full figure and legend (272K)

MYC expression can increase cell size, due to both intrinsic cell growth and entry into cycle (Iritani and Eisenman, 1999). Increased cell size, indicative of functional transgene expression, was evident in several types of cells in healthy young VavP-MYC17 mice. These included the four major subpopulations of thymocytes (CD4+ CD8-, CD4- CD8+, CD4+ CD8+, CD4- CD8-), peripheral T cells, bone marrow pre-B cells, splenic B cells and blood monocytes (examples are shown in Figure 1c), and megakaryocytes (Figure 1d). The increased size of pre-B cells was not due to selective proliferation of the minor, CD43+ fraction of normally large cells as this population was increased in number only slightly (1.7-fold). No consistent increase in the size of granulocytes (Figure 1c) or nucleated erythroid cells (Ter119+) (not shown) was seen.

Pleiotropic consequences for haemopoietic homeostasis

No malignant cells were detectable in 2-week-old transgenic mice by transplantation tests (see Materials and methods). Nevertheless, their blood leucocyte profile was clearly abnormal. Nucleated cells had increased approximately 10-fold, due to large increases in B-lineage cells (both pre-B and B cells) and myeloid cells (both monocytes and granulocytes), and a small increase in T cells (Figure 2a, left panel; Table 1). By 4 weeks, however, the blood leucocyte count (and nucleated cell composition) had returned almost to normal (Figure 2a, right panel).

Figure 2.
Figure 2 - Unfortunately we are unable to provide accessible alternative text for this. If you require assistance to access this image, please contact help@nature.com or the author

Haemopoiesis is greatly perturbed in healthy young VavP-MYC17 mice. (a) Transient elevation of blood leucocyte counts. B-lymphoid and myeloid cells were greatly increased in 13- to 17-day-old (2 weeks) VavP-MYC17 mice, but almost normal in 23- to 29-day-old (4 weeks) mice. Total nucleated counts were determined with a Coulter counter and the number of B-lymphoid (B220+), T-lymphoid (Thy1+) and myeloid (Mac1+) cells by flow cytometry. Error bars indicate s.e.m. (b) Flow cytometry of B-lymphoid and monocyte populations in haemopoietic tissues. In all, 2-week-old VavP-MYC17 mice have a much greater ratio of pre-B (B220+ IgM-) to B (B220+ IgM+) cells in the spleen and an expanded monocyte (Mac1+ Gr1-) population in the bone marrow. The percentage of cells in each quadrant is indicated

Full figure and legend (101K)

Perturbed homeostasis was also evident within the haemopoietic tissues of the 2-week-old transgenic mice (Figure 2b; Table 1). The spleen was three times larger than normal, due to substantial increases in both B-lymphoid and erythroid cells. Strikingly, pre-B cells constituted approximately 40% of the splenic B-lineage cells. In the bone marrow, the most conspicuous change was the substantial increase in monocytes, apparently at the expense of erythropoiesis. The thymus appeared normal, both in cellularity and cell composition. Quantification of megakaryocytes in histological sections (20–35 high-power fields counted) from 2-week-old VavP-MYC17 mice and normal littermates revealed that these cells were much more abundant in transgenic spleen (7.1plusminus0.6 per field vs 3.6plusminus0.4, P< 0.001; overall increase was approximately sixfold due to the increased spleen size) and liver (1.0plusminus0.2 per field vs 0.02plusminus0.02, P<0.001), but not bone marrow (3.6plusminus0.3 per field vs 4.2plusminus0.3, P=0.131).

VavP-MYC17 mice develop early onset T-cell lymphomas

All of the VavP-MYC17 mice monitored (23/23) succumbed to tumours between 7 and 10 weeks of age, and all of those analysed (22/22) were T-cell lymphomas: they expressed the T-lineage markers Thy1, TCRbeta and CD3, and lacked B-lymphoid (CD19), erythroid (Ter119) and myeloid (Mac1) markers (Figure 3a and c). The most frequently occurring lymphoma had the phenotype of a mature CD4+ CD8- T cell (12/22). Also seen were tumours of pre-T-cell (CD4+ CD8+) (4/22) and 'mixed T-cell' (3/22) origin; the latter contained both CD4+ CD8- and CD4+ CD8+ cells (Figure 3b and c). The variable latent period before tumour onset implied that each tumour had acquired somatic, cooperating, oncogenic mutations. The tumours of uniform phenotype presumably represented single transformed clones, whereas the 'mixed T-cell' tumours may have originated as separate clones, expected given the high rate of tumorigenesis in these mice.

Figure 3.
Figure 3 - Unfortunately we are unable to provide accessible alternative text for this. If you require assistance to access this image, please contact help@nature.com or the author

T-cell lymphomas in VavP-MYC17 mice. (a) Flow cytometric analysis of a typical (Thy1+ TCRbeta+ TCRitalic gamma/delta- CD19-) VavP-MYC17 tumour. The percentage of cells in each quadrant is indicated. (b) CD4/CD8 expression varied on T-cell lymphomas. Flow cytometric analysis is shown for a typical CD4+ CD8- tumour (left panel) and a less common CD4+ CD8+ tumour (right panel). (c) Summary of the expression of surface markers on the 22 VavP-MYC17 T lymphomas analysed, determined by flow cytometry. (d) Some T-cell lymphomas contained cells expressing low levels of B220. B220 expression on a typical B220lo T lymphoma (red solid line) is compared to that on gated Thy1+ wild-type peripheral T cells (grey shaded) and gated B220+ wild-type peripheral B cells (black solid line). (e) A CD4+ CD8- T-cell lymphoma with immunoblastoid appearance. Note the large nuclei, frequently containing a large, intensely stained nucleolus, and a tessellated appearance. H and E-stained section, bar=20 mum. (f) Another lymphoma deposit in the same mouse as in (e) showing classical lymphoblastic appearance, with smaller nuclei and indistinct cell borders. H and E-stained section, bar=20 mum

Full figure and legend (274K)

None of the tumours expressed the early activation markers CD25 or CD69, but all expressed the late activation/memory marker CD44 (Figure 3c). Six also contained a significant number of cells (>30%) expressing low levels of B220, which can be a late T-cell activation marker in the mouse (Figure 3c and d).

Histological analysis indicated that the tumours had disseminated throughout the haemopoietic system, with frequent heavy invasion of the spleen (mean weight 1.0plusminus0.1 vs 0.1plusminus0.0 g in age-matched controls; n=28), thymus, bone marrow, lymph nodes and one or more Peyer's patches. Many sick mice also presented with leukaemia (mean leucocyte count 37plusminus6 times 106 vs 9plusminus1 times 106 per ml in age-matched controls; n=32). The invaded bone marrow usually showed the 'starry sky' appearance (Morse et al., 2002), thought to result from the engulfment of apoptotic tumour cells by macrophages. The liver, kidney and lungs commonly contained extensive tumour deposits, but the heart was only occasionally involved. Tumour cell morphology ranged, in a spectrum, from classic lymphoblastic (Morse et al., 2002) to immunoblastoid (Figure 3e and f), the latter with a large, vesicular nucleus containing a very large, strongly stained nucleolus, and moderately stained cytoplasm with readily visible cell borders. The frequency of mitotic figures (12plusminus1 mitoses per high-power field; greater than or equal to6 fields counted for tumours from 10 mice) was indicative of a high neoplastic grade. All eight tumours tested were readily transplantable.

Foetal liver reconstitution

In view of the multilineage abnormalities preceding tumorigenesis, it was surprising to find that the VavP-MYC17 mice developed only T-lineage lymphomas. Seeking to reveal a broader susceptibility to tumorigenesis, we undertook foetal liver reconstitution of lethally irradiated mice, during which myeloid precedes lymphoid regeneration. Liver cell suspensions from five E14 VavP-MYC17 foetuses were each injected into three lethally irradiated B6 mice. All 15 reconstituted mice rapidly developed tumours (median 14 weeks after transplantation) and, somewhat surprisingly, all 14 analysed were T-cell lymphomas (not shown). Two (derived from different foetal livers) were Thy1+ TCRitalic gamma/delta+ tumours, which may indicate that, during immune reconstitution, italic gamma/delta T cells were a proportionally greater target for transformation. The other tumours were of the types seen in VavP-MYC17 mice (three CD4+ CD8-, six CD4+ CD8+, one CD4- CD8+, one CD4- CD8- and one mixed CD4+ CD8-/CD4+ CD8+), but differed by the predominance of pre-T lymphomas.

Rag1-/- VavP-MYC17 mice develop a range of tumour types

As an alternative approach to determining whether other haemopoietic cell types in VavP-MYC17 mice were at risk of malignant transformation, we bred them with Rag1-deficient mice. The Rag1 recombinase gene is required for Tcr and Ig gene rearrangement and in its absence T-cell development is arrested at the pro-T3 stage (DN3 or pre-T1 in other nomenclatures) and B-cell development at the late pro-B stage (pre-B1 or Hardy fraction C) (Mombaerts et al., 1992b; Spanopoulou et al., 1994; Hardy and Hayakawa, 2001; Ceredig and Rolink, 2002; Marsden and Strasser, 2003). While the Rag1-/- VavP-MYC17 mice lived substantially longer than VavP-MYC17 mice, they all succumbed to tumours (median survival 13 weeks vs 8 weeks) (Figure 4). Cell surface analysis by flow cytometry identified four distinct tumour types in the 14 mice analysed: pro-B-cell, progenitor-cell, pre-T-cell and macrophage tumours.

Figure 4.
Figure 4 - Unfortunately we are unable to provide accessible alternative text for this. If you require assistance to access this image, please contact help@nature.com or the author

Tumorigenesis in Rag1-/- VavP-MYC17 mice is delayed. VavP-MYC17 mice (n=23) were all sick within 10 weeks, with a median survival of 8 weeks, but Rag1-/- VavP-MYC17 mice (n=18) showed a median survival of 13 weeks. Mice were monitored for tumours and killed when sick. The percentage of mice surviving at weekly intervals is shown

Full figure and legend (9K)

Pro-B-cell tumours

Six mice developed pro-B lymphomas, which bore the B-lineage markers B220 and CD19, and lacked T-lineage (Thy1, CD3), myeloid (Gr1, Mac1) and erythroid (Ter119) markers (Table 2). As they expressed HSA, BP-1, AA4.1, and CD43, but not CD25, they appear to have been derived from late pro-B cells. At autopsy, the mice had somewhat increased blood leucocyte counts (5.3plusminus1.9 times 106 per ml vs 1.4plusminus0.1 times 106 per ml in healthy young Rag1-/- VavP-MYC17 mice) and typically had large tumour deposits in the thymus and lymph nodes, but with variable involvement of the spleen and bone marrow. Invasion of nonhaemopoietic tissues (including the liver, kidney, lung and heart) was generally limited. Histologically, the tumour cells were predominantly lymphoblastic (not shown).

Novel Thy1+B220+ tumours

Six mice developed unusual tumours that expressed high levels of both B220 and Thy1 and were also positive for the myeloid markers Gr1 and F4/80 (Table 2). This pattern of cell surface markers is suggestive of a multipotential progenitor. The presentation of these tumours was quite distinct from that of the pro-B lymphomas. The spleen was invariably greatly enlarged (mean weight 1.3plusminus0.1 g), blood leucocyte counts were high (17.4plusminus7.3 times 106 per ml), very large tumour deposits were generally seen in the lymph nodes and bone marrow, and involvement of the thymus was variable. Tumour infiltration was common in the liver and lungs and, to a lesser extent, in the kidney and heart. The tumours had immunoblastoid morphology (not shown).

Pre-T-cell lymphomas

Two Rag1-/- VavP-MYC17 mice developed Thy1+ T-lineage lymphomas. Surprisingly, both tumours had the phenotype of pre-T cells (CD4+ CD8+ CD25- CD2+) (Figure 5a; Table 2) rather than pro-T3 cells (Thy1+ CD4-/lo CD8- CD25+ CD2- CD44-), the earlier stage at which T-cell development is arrested in Rag1-/- mice. The tumours also expressed CD44, a marker not normally present on pre-T cells, but expression of CD44 can reflect proliferation (Min et al., 2003) as well as developmental stage. They were confined to the thymus, with only minimal dissemination to the blood or tissues. Both were lymphoblastic in morphology (not shown).

Figure 5.
Figure 5 - Unfortunately we are unable to provide accessible alternative text for this. If you require assistance to access this image, please contact help@nature.com or the author

T-cell lymphomas and T-cell development in Rag1-/- VavP-MYC17 mice. (a) A typical Rag1-/- VavP-MYC17 T-cell lymphoma with the surface phenotype (Thy1+ CD4+ CD8+ CD25-) of pre-T cells. (b) A novel T-cell population in the pretumour Rag1-/- VavP-MYC17 thymus. Pre-T cells predominated in the Rag1+/- thymus (left panels) and pro-T3 cells (Thy1+ CD4-/lo CD8- CD25+) in the Rag1-/- thymus (middle panels). The thymus of age-matched (5-week-old) healthy Rag1-/- VavP-MYC17 mice (right panels) contained a novel population of Thy1hi CD4hi CD8- CD25lo cells (indicated by the box) (28plusminus5% of thymocytes; n=4) in addition to pro-T3 cells. The low cell content typical of the Rag1-/- thymus (1.4plusminus0.2 times 107 vs 1.4plusminus1 times 108 in Rag1+/- thymus) was not restored in the Rag1-/- VavP-MYC17 thymus (1.4plusminus0.2 times 107)

Full figure and legend (65K)

The pre-T-cell phenotype of these tumours implied that they had by-passed the beta-selection checkpoint that arrests Rag1-/- thymocyte development at the pro-T3 stage. To investigate whether the differentiation block had been overcome prior to tumour onset, we analysed healthy 5-week-old mice (Figure 5b). The thymus in Rag1-/- VavP-MYC17 mice was as reduced in cell content as that of Rag1-/- littermates, and the pro-T3 cell population was prominent, as expected. Significantly, there were no cells with the surface phenotype of pre-T cells. However, over 25% of cells in the Rag1-/- VavP-MYC17 thymus were a novel population not seen in the Rag1-/- thymus. These cells expressed higher levels of Thy1 and CD4 than pre-T cells, and lower levels of CD25 than pro-T3 cells, and their surface phenotype (Thy1hi CD4hi CD8- CD25lo CD2+ CD44+) was distinct from any presently defined stage of T-cell development.

Cutaneous macrophage tumours

The five longest surviving Rag1-/- VavP-MYC17 mice developed cutaneous tumours that coexisted with progenitor cell tumours (three mice) and pre-T-cell lymphomas (two mice). The three tumours analysed expressed F4/80 and Mac1, but were negative for the granulocyte marker Gr1 (and other lineage markers including B220, CD19, Thy1, CD3 and Ter119) (Figure 6a; Table 2), implying that they were of monocyte/macrophage origin. Histologically, the tumours had the morphology of histiocytic sarcoma (Figure 6b) and each was confined to a single site except in one mouse where some lymph nodes were also involved. The tumours were locally invasive, investing the fat and muscle of the hypodermis and heavily populating the dermis. Of the four tumours tested, three were transplantable, spreading throughout the haemopoietic system and other organs such as the liver and kidney.

Figure 6.
Figure 6 - Unfortunately we are unable to provide accessible alternative text for this. If you require assistance to access this image, please contact help@nature.com or the author

Cutaneous macrophage tumours in Rag1-/- VavP-MYC17 mice. (a) Flow cytometric analysis of a typical cutaneous macrophage (Mac1+ F4/80+ Gr1-) Rag1-/- VavP-MYC17 tumour. The percentage of cells in each quadrant is indicated. (b) High-power views of two cutaneous macrophage tumours. They differ somewhat in morphology, but both comprise masses of large, mitotically active cells with bulky, variably vacuolated cytoplasm and a vesicular, sometimes pleomorphic nucleus, characteristics of histiocytic sarcoma. H and E-stained section, bar=20 mum

Full figure and legend (225K)



In this study, we set out to explore the impact of deregulated MYC expression on different haemopoietic lineages. We have previously established that the VavP transgenic vector used here enforces expression in multiple haemopoietic cell types, including progenitor cells (Ogilvy et al., 1999a) and, as expected, exogenous MYC protein was detectable in T- and B-lymphoid cells, granulocytes and monocytes of VavP-MYC17 mice (Figure 1b). Furthermore, cells of multiple haemopoietic lineages were larger than normal (Figure 1c, d), consistent with MYC-promoted cell growth and/or increased cycling.

Transplantation tests detected no tumour cells in 2-week-old VavP-MYC17 mice, but their haemopoietic system was markedly perturbed, most notably in the B-lymphoid, myeloid and megakaryocyte compartments (Figures 1d, 2a and b; Table 1). In the bone marrow, there was a marked increase in monocytes at the expense of erythroid cells. Indeed, erythropoiesis was largely displaced to the spleen, which also contained excessive numbers of B-lymphoid cells (both pre-B cells and mature B cells) and megakaryocytes. In addition, there was a 10-fold increase in blood leucocytes, primarily B-lymphoid cells and monocytes, although levels were almost normal by 4 weeks, perhaps reflective of more effective homeostatic control mechanisms by this age.

The VavP-MYC17 mice were highly prone to tumour development and succumbed before 10 weeks of age (Figure 4). In view of the modest pretumour changes in the T-lymphoid compared to other lineages, we were surprised to find that all the tumours were T-cell lymphomas (Figure 3a and c). Furthermore, in contrast to the CD4+ CD8+ pre-T-cell lymphomas typically seen in other Myc transgenic mice that develop T-lineage tumours, such as those expressing a CD2 promoter-driven (Stewart et al., 1993) or a tetracycline-inducible transgene (Felsher and Bishop, 1999), most tumours in VavP-MYC17 mice were CD4+ CD8- T-cell lymphomas.

The VavP-MYC17 T lymphomas expressed CD44 but were negative for other activation markers (CD25, CD69), and so the CD44 expression probably reflects multiple rounds of cell division (Min et al., 2003). Intriguingly, the B-cell marker B220 was detected at low levels on a proportion of cells in some tumours (6/22) (Figure 3c and d). Low B220 expression on T cells normally coincides with the approach of post-activation apoptosis (Renno et al., 1998). This feature of VavP-MYC17 T-cell lymphomas may indicate, therefore, that acquisition of malignancy has involved the disabling of an apoptotic programme that normally eliminates T cells after multiple rounds of division.

Mouse models of mature CD4+ CD8- T-cell lymphoma are rare and the VavP-MYC17 mice may prove to be a useful tool. While their short lifespan presents breeding difficulties, these can be overcome by in vitro fertilization. Moreover, the tumours are readily transplantable and, like B-cell tumours transplanted from Emu-Myc mice (Schmitt et al., 1999, 2000), could be used to test for synergistic mutations and to study chemotherapy responses. In humans, T-cell chronic lymphocytic leukaemia and T-cell prolymphocytic leukaemia are largely diseases of CD4+ CD8- T cells (Harris et al., 1994) and share further surface characteristics (TCRbeta+ CD3+ CD5+ CD2+ CD25-) with the VavP-MYC17 tumours (Figure 3c). These leukaemias are associated with chromosomal rearrangements activating the expression of TCL1 and MTCP1 (Pekarsky et al., 2001) and, furthermore, trisomy 8q associated with increased MYC expression occurs frequently in T-cell prolymphocytic leukaemia (Maljaie et al., 1995). Transgenic mice overexpressing TCL1 or MTCP1 in T cells usually developed late-onset CD4- CD8+ T-cell tumours, however, rather than the CD4+ CD8- neoplasms they aimed to model (Pekarsky et al., 2001).

In the Rag1-deficient background, in which the T-cell populations most at risk of malignant transformation by the VavP-MYC17 transgene are absent, there was a dramatic shift to the development of multilineage tumours (Table 2). Most were pro-B-cell lymphomas (43%) or unusual B220+ Thy1+ tumours (43%) rather than T-cell lymphomas (14%). In addition, many mice (36%) developed macrophage tumours along with the other tumour types. The latency of all these tumours was longer than that of the T-cell lymphomas in a Rag1-proficient background (Figure 4), indicating that, although multiple cell types are susceptible, T-cell lymphomas predominate overwhelmingly because they have the shortest latency. The differential transformation rate may reflect a combination of factors including the relative level of MYC overexpression in the various cell types, the size of the cell population at risk, the number of synergistic somatic mutations required for tumorigenesis and the ease with which they are acquired, and the proliferation rate of malignant clones.

The B220+ Thy1+ tumours in Rag1-/- VavP-MYC17 mice (Table 2) are presumably derived from a rare progenitor cell although, to our knowledge, no normal counterpart of this cell type has yet been described and we have been unable to detect such a population in the spleen, thymus or bone marrow of healthy young VavP-MYC17 or Rag1-/- VavP-MYC17 mice. While Emu-BCL2-Myc bi-transgenic mice develop B220+ Thy1lo progenitor cell tumours (Strasser et al., 1996), these differ from those described here in the expression of several surface markers.

The histiocytic monocyte/macrophage tumours that developed in Rag1-/- VavP-MYC17 mice (Figure 6a and b; Table 2) were cutaneous, with limited involvement of other tissues, suggesting that they arose by mutation of tissue-resident macrophages. However, on transplantation in syngeneic recipients, tumours spread throughout the haemopoietic system and other organs (including the liver and kidney) (not shown). Of note, the skin is a site for the rare human monocytic neoplasms – extramedullary myeloid cell tumour (monocytic subtype) and true histiocytic lymphoma (Elghetany, 1997).

Two of the tumours that developed in Rag1-/- VavP-MYC17 mice were pre-T-cell (CD4+ CD8+) lymphomas (Figure 5a; Table 2). This phenotype was unexpected because T-cell development in Rag1-/- mice is blocked at the earlier pro-T3 stage by the beta-selection checkpoint, which deletes cells lacking TCRbeta by apoptosis (Marsden and Strasser, 2003). Certain genetic changes (both proliferative and anti-apoptotic) can force Rag1-/- thymocytes through this checkpoint and partially restore thymus cellularity (Michie and Zuniga-Pflucker, 2002). Since the thymus in healthy young Rag1-/- VavP-MYC17 mice was as small as that in nontransgenic Rag1-/- littermates and contained no pre-T cells (Figure 5b), MYC overexpression per se does not overcome the beta-selection block. Instead this checkpoint seems to have been by-passed during tumour development, perhaps by inactivation of the apoptotic machinery controlling this process.

Unexpectedly, in addition to pro-T3 cells, the premalignant thymus of 5-week-old Rag1-/- VavP-MYC17 mice contained a significant number of cells having a surface phenotype (Thy1hi CD4hi CD8- CD25lo CD2+ CD44+) distinct from any previously described population of thymocytes (Figure 5b). These cells may represent a MYC-driven expansion of an unrecognized intermediate stage in T-cell development. Cells of this type were not evident in the thymus of young VavP-MYC17 mice, but would only be expected to constitute at most 3% of the thymocytes. It is unclear whether the pre-T-cell tumours that arose in Rag1-/- VavP-MYC17 mice derived from this unusual population of cells or from the pro-T3 population.

In summary, the VavP-MYC17 transgene perturbs the growth and proliferation of multiple lineages of haemopoietic cells. VavP-MYC17 mice should therefore be a useful resource for comparative studies of MYC targets in different haemopoietic cell types. In a wild-type background, the transgene promoted the rapid development of mature CD4+ CD8- T-cell lymphomas, which are rarely seen in other mouse models, and in the Rag1-deficient background macrophage, pro-B-cell and progenitor cell tumours developed. These mice therefore represent a new tool to study mechanisms of leukaemogenesis, synergism between genetic lesions, chemotherapeutic efficacy and chemoprevention in several haemopoietic tumour types. In the Rag1-deficient background, cell types that may be novel intermediates during haemopoietic development may also have been revealed. Crosses of VavP-MYC17 mice with other mutant strains (Mombaerts et al., 1992a; Nehls et al., 1994; von Freeden-Jeffry et al., 1995) deficient in T cells might yield further novel progenitor cell populations and tumour models.


Materials and methods

Transgene construction and generation of VavP-MYC mice

A human MYC cDNA, encompassing the coding region but lacking any 5' or 3' untranslated sequence (GenBank Accession number V00568; nucleotides 559 to 1878), was PCR amplified with primers that included SfiI (5') and NotI (3') linkers. The SfiI to NotI fragment was then cloned into the Vav enhancer/promoter vector (VavP) (Ogilvy et al., 1999a) (Figure 1a) and sequenced. The VavP-MYC construct was separated from the vector by HindIII digestion and injected into a pronucleus of C57BL/6J (B6) mouse zygotes, which were then transferred into pseudopregnant recipients. Transgenic offspring were identified by PCR of tail biopsy DNA using primers to the vector SV40 polyadenylation signal (gccgcagacatgataagatacatt and tcggctcgcgaggttttac). Five primary transgenic mice resulted. One died early of lymphoma without issue. Another also died early of lymphoma, but was used for in vitro fertilization to produce the VavP-MYC17 line described here, which expressed MYC at high levels. Two other primary mice produced descendant lines (to be described elsewhere) that expressed MYC at low levels and developed T lymphomas and/or macrophage tumours with longer latency. A fifth animal produced a line that expressed no detectable MYC.


All mice were of the B6 background and bred in the animal facilities of this institute. As female mice of the VavP-MYC17 transgenic line described here succumbed to lymphomas before completion of pregnancy or nursing, it was propagated by breeding transgenic males with normal B6 mice, either by natural mating or, when larger numbers were required, by in vitro fertilization (Bath, 2003). Transgenic progeny was identified by PCR amplification of blood DNA and increased size of blood leucocyte nuclei.

Introduction of the VavP-MYC17 transgene into Rag1-/- mice (Spanopoulou et al., 1994) was performed by in vitro fertilization; eggs from approximately 4-week-old VavP-MYC17 mice were fertilized with B6.Rag1-/- sperm, and then eggs from approximately 4-week-old VavP-MYC17 Rag1+/- offspring were fertilized with Rag1-/- sperm. The VavP-MYC17 transgene was detected by PCR, and Rag1-/- mice were identified by their low blood leucocyte count.

To confirm the pretumour status of young transgenic mice, approximately 3 times 106 viable cells dispersed from the thymus, spleen and bone marrow of 13- to 17-day-old mice were injected intraperitoneally into B6 mice (two mice per tissue) and the recipients monitored for tumour growth for at least 90 days.

Transplantation tests on tumours were performed by injecting approximately 2 times 106 viable cells from dispersed tumour tissue both subcutaneously and intraperitoneally into B6 mice and monitoring recipients until they became sick or for at least 90 days.

For haemopoietic reconstitution experiments, cell suspensions were prepared from E14 foetal liver of VavP-MYC17 mice, and approximately 2 times 106 viable cells were injected into the tail vein of lethally irradiated (2 times 5.5 Gy spaced by 2 h) B6 mice.

For histological analysis, soft tissues were fixed in Bouin's, and bones in 10% buffered formalin. After standard processing, 2-mum tissue sections were stained with haematoxylin and eosin (H and E).

Nucleated cell counts and nuclear size distributions were determined with a Coulter Z2 particle count and size analyzer (Beckman Coulter).

Flow cytometry

Cell suspensions were surface stained with fluorescein (FITC)- or biotin-conjugated antibodies in balanced salt solution/2% foetal calf serum/1% rat serum. Biotinylated antibodies were detected with R-phycoerythrin (PE)–streptavidin (Caltag). Approximately 5000 viable (unstained by propidium iodide) cells were analysed in a FACScan II flow cytometer (Becton Dickinson). Monoclonal antibodies (conjugated in our laboratory or, where indicated, obtained commercially) used were: RA3-6B2 anti-B220; RM2-1 anti-CD2; KT-3 anti-CD3; AA4.1 and BP-1 anti-early B-lineage markers; H129.19 anti-CD4; 53-7.3 anti-CD5; YTS 169 anti-CD8; N418 anti-CD11c; ID3 anti-CD19; PC61 anti-CD25; 57 anti-CD43; IM7 anti-CD44; H1.2F3 anti-CD69; Ter119 anti-erythroid marker; M1/69 anti-HSA; RB6-8C5 anti-Gr1; 5.1 anti-IgM; MI/70 anti-Mac1; F4/80 anti-macrophage marker; M5/114.15.2 anti-MHC class II (I-A/I-E); DX5 anti-pan NK marker (Caltag); E13 161-7, anti-Sca-1; H57-597, anti-TCRbeta; GL3, anti-TCRitalic gamma/delta; 30-H12 anti-Thy1 (Pharmingen). Cell populations were isolated using a MoFlo cell sorter (Dako-Cytomation).

Western blots

Cell extracts were prepared by lysis of tissue suspensions and sorted cell populations in RIPA buffer (50 mM Tris-Cl pH 8.0, 150 mM NaCl, 0.1% SDS, 1% Nonidet P-40, 0.5% sodium deoxycholate) containing proteinase inhibitors (1.25 mg/ml Pefabloc SC, 2.5 mug/ml leupeptin, 2.5 mug/ml aprotinin, 250 mug/ml soybean trypsin inhibitor, 5 mug/ml E64, 10 mug/ml pepstatin A, 1 mM phenylmethylsulphonylfluoride). Aliquots (40 mug of protein; Biorad protein assay) were fractionated by 12% SDS–polyacrylamide gel electrophoresis, and the proteins transferred to nitrocellulose membranes and stained with rabbit antibody to MYC (#06-340; Upstate), a mouse monoclonal antibody to beta-actin (AC-15; Abcam) and HRP-conjugated secondary antibodies (Pierce). Blots were developed with ECL reagents (Amersham) and visualized by autoradiography.



  1. Adams JM, Harris AW, Pinkert CA, Corcoran LM, Alexander WS, Cory S, Palmiter RD and Brinster RL. (1985). Nature, 318, 533–538. | Article | PubMed | ISI | ChemPort |
  2. Adams JM, Harris AW, Strasser A, Ogilvy S and Cory S. (1999). Oncogene, 18, 5268–5277. | Article | PubMed | ISI | ChemPort |
  3. Bath ML. (2003). Biol. Reprod., 68, 19–23. | Article | PubMed | ChemPort |
  4. Boxer LM and Dang CV. (2001). Oncogene, 20, 5595–5610. | Article | PubMed | ISI | ChemPort |
  5. Ceredig R and Rolink T. (2002). Nat. Rev. Immunol., 2, 888–897. | Article | PubMed | ISI | ChemPort |
  6. Cory S. (1986). Adv. Cancer Res., 47, 189–234. | PubMed | ISI | ChemPort |
  7. Egle A, Harris AW, Bouillet P and Cory S. (2004). Proc. Natl. Acad. Sci. USA, 101, 6164–6169. | Article | PubMed | ChemPort |
  8. Eischen CM, Weber JD, Roussel MF, Sherr CJ and Cleveland JL. (1999). Genes Dev., 13, 2658–2669. | Article | PubMed | ISI | ChemPort |
  9. Elghetany MT. (1997). Leukemia, 11, 762–764. | Article | PubMed | ChemPort |
  10. Evan G and Littlewood T. (1998). Science, 281, 1317–1321. | Article | PubMed | ISI | ChemPort |
  11. Felsher DW and Bishop JM. (1999). Mol. Cell, 4, 199–207. | Article | PubMed | ISI | ChemPort |
  12. Grandori C, Cowley SM, James LP and Eisenman RN. (2000). Ann. Rev. Cell Dev. Biol., 16, 653–699. | Article | ChemPort |
  13. Hann B and Balmain A. (2001). Curr. Opin. Cell Biol., 13, 778–784. | Article | PubMed | ISI | ChemPort |
  14. Hardy RR and Hayakawa K. (2001). Ann. Rev. Immunol., 19, 595–621. | Article | ChemPort |
  15. Harris AW, Pinkert CA, Crawford M, Langdon WY, Brinster RL and Adams JM. (1988). J. Exp. Med., 167, 353–371. | Article | PubMed | ISI | ChemPort |
  16. Harris NL, Jaffe ES, Stein H, Banks PM, Chan JK, Cleary ML, Delsol G, De Wolf-Peeters C, Falini B and Gatter KC. (1994). Blood, 84, 1361–1392. | PubMed | ISI | ChemPort |
  17. Iritani BM and Eisenman RN. (1999). Proc. Natl. Acad. Sci. USA, 96, 13180–13185. | Article | PubMed | ChemPort |
  18. Langdon WY, Harris AW, Cory S and Adams JM. (1986). Cell, 47, 11–18. | Article | PubMed | ISI | ChemPort |
  19. Maljaie SH, Brito-Babapulle V, Matutes E, Hiorns LR, De Schouwer PJ and Catovsky D. (1995). Leukemia, 9, 1694–1699. | PubMed | ChemPort |
  20. Marcu KB, Bossone SA and Patel AJ. (1992). Ann. Rev. Biochem., 61, 809–860. | Article | PubMed | ISI | ChemPort |
  21. Marsden V and Strasser A. (2003). Ann. Rev. Immunol., 21, 71–105. | Article | ChemPort |
  22. Michie AM and Zuniga-Pflucker JC. (2002). Semin. Immunol., 14, 311–323. | Article | PubMed | ISI | ChemPort |
  23. Mikkers H and Berns A. (2003). Adv. Cancer Res., 88, 53–99. | Article | PubMed | ISI | ChemPort |
  24. Min B, McHugh R, Sempowski GD, Mackall C, Foucras G and Paul WE. (2003). Immunity, 18, 131–140. | Article | PubMed | ISI | ChemPort |
  25. Mombaerts P, Clark AR, Rudnicki MA, Iacomini J, Itohara S, Lafaille JJ, Wang L, Ichikawa Y, Jaenisch R, Hooper ML and Tonegawa S. (1992a). Nature, 360, 225–231. | Article | PubMed | ISI | ChemPort |
  26. Mombaerts P, Iacomini J, Johnson RS, Herrup K, Tonegawa S and Papaioannou VE. (1992b). Cell, 68, 869–877. | Article | PubMed | ISI | ChemPort |
  27. Morse III HC, Anver MR, Fredrickson TN, Haines DC, Harris AW, Harris NL, Jaffe ES, Kogan SC, MacLennan IC, Pattengale PK and Ward JM. (2002). Blood, 100, 246–258. | Article | PubMed | ISI | ChemPort |
  28. Nehls M, Pfeifer D, Schorpp M, Hedrich H and Boehm T. (1994). Nature, 372, 103–107. | Article | PubMed | ISI | ChemPort |
  29. Ogilvy S, Metcalf D, Gibson L, Bath ML, Harris AW and Adams JM. (1999a). Blood, 94, 1855–1863. | PubMed | ISI | ChemPort |
  30. Ogilvy S, Metcalf D, Print CG, Bath ML, Harris AW and Adams JM. (1999b). Proc. Natl. Acad. Sci. USA, 96, 14943–14948. | Article | PubMed | ChemPort |
  31. Pekarsky Y, Hallas C and Croce CM. (2001). JAMA, 286, 2308–2314. | Article | PubMed | ISI | ChemPort |
  32. Renno T, Attinger A, Rimoldi D, Hahne M, Tschopp J and MacDonald HR. (1998). Eur. J. Immunol., 28, 540–547. | Article | PubMed | ChemPort |
  33. Sanchez-Beato M, Sanchez-Aguilera A and Piris MA. (2003). Blood, 101, 1220–1235. | Article | PubMed | ISI | ChemPort |
  34. Schmitt CA. (2003). Nat. Rev. Cancer, 3, 286–295. | Article | PubMed | ISI | ChemPort |
  35. Schmitt CA, McCurrach ME, de Stanchina E, Wallace-Brodeur RR and Lowe SW. (1999). Genes Dev., 13, 2670–2677. | Article | PubMed | ISI | ChemPort |
  36. Schmitt CA, Rosenthal CT and Lowe SW. (2000). Nat. Med., 6, 1029–1035. | Article | PubMed | ISI | ChemPort |
  37. Shou Y, Martelli ML, Gabrea A, Qi Y, Brents LA, Roschke A, Dewald G, Kirsch IR, Bergsagel PL and Kuehl WM. (2000). Proc. Natl. Acad. Sci. USA, 97, 228–233. | Article | PubMed | ChemPort |
  38. Spanopoulou E, Roman CAJ, Corcoran LM, Schlissel MS, Silver DP, Nemazee D, Nussenzweig MC, Shinton SA, Hardy RA and Baltimore D. (1994). Genes Dev., 8, 1030–1042. | PubMed | ISI | ChemPort |
  39. Stewart M, Cameron E, Campbell M, McFarlane R, Toth S, Lang K, Onions D and Neil JC. (1993). Int. J. Cancer, 53, 1023–1030. | PubMed | ChemPort |
  40. Strasser A, Elefanty AG, Harris AW and Cory S. (1996). EMBO J., 15, 3823–3834. | PubMed | ISI | ChemPort |
  41. Strasser A, Harris AW, Bath ML and Cory S. (1990). Nature, 348, 331–333. | Article | PubMed | ISI | ChemPort |
  42. von Freeden-Jeffry U, Vieira P, Lucian LA, McNeil T, Burdach SEG and Murray R. (1995). J. Exp. Med., 181, 1519–1526. | Article | PubMed | ChemPort |


We thank J Adams for helpful discussions, A Strasser for his gift of antibodies, A Wiegmans for technical assistance, C Young for cell sorting, S Mihajlovic for histological processing, and K Birchall and A Naughton for animal husbandry. This work was supported by the National Health and Medical Research Council (NHMRC, Canberra, Program Grant 257502), the US National Cancer Institute (Grant CA43540) and the Leukemia and Lymphoma Society (New York, Specialized Center of Research Grant 7015).



These links to content published by NPG are automatically generated


Immunology Blood lines redrawn

Nature News and Views (10 Apr 2008)

Immunology Wavering on commitment

Nature News and Views (07 Oct 1999)