An ex Vivo Model to Study v-Myb-Induced Leukemogenicity

Dvorakova et al.

Blood Cells, Molecules, and Diseases (2001) 27(2) Mar/Apr: 437– 445 doi:10.1006/bcmd.2001.0402, available online at http://www.idealibrary.com on

An ex Vivo Model to Study v-Myb-Induced Leukemogenicity Submitted 01/31/01 (Communicated by M. Lichtman, M.D., 02/08/01)

Marta Dvorakova,1 Jarmila Kralova,1 Vit Karafiat,1 Petr Bartunek,1 and Michal Dvorak1,2 ABSTRACT: The v-mybAMV oncogene transforms myelomonocytic cells in vitro and induces acute monoblastic leukemia in chickens. We analyzed the activity of the evolutionarily conserved PEST-like domain (P1 domain) for biochemical and biological activities of v-Myb in ex vivo cultures and in vivo. Deletion of the P1 domain did not affect v-Myb transcriptional activity, intracellular stability, or subcellular localization. However, it resulted in subtle yet important changes in biological activities. Although the mutant ⌬P1 v-Myb protein blocked the terminal differentiation of the monocyte/macrophage lineage as efficiently as the wild type (wt) in ex vivo cultures, it failed to induce the acute phase of monoblastic leukemia, with its fatal consequences, in vivo. Interestingly, in ⌬P1 v-myb-infected animals large numbers of monoblasts, comparable to those induced by wt v-myb, were present in the bone marrow but very few were found in the peripheral blood. The comparison of ex vivo wt- and ⌬P v-Myb bone marrow cells revealed several important features of v-Myb transformation: (i) the proliferation of transformed monoblasts is not an apparent consequence of the differentiation block with these processes being at least in part independent; (ii) the P1 domain is required for proliferation of v-Myb-mediated transformed monoblasts; (iii) the mechanism which renders transformed cells growth factor independent does not involve activation of an autocrine growth factor loop; and (iv) deletion of the P1 domain affects self-adhesion properties of v-myb-transformed monoblasts as well as their interaction with bone marrow stromal cells. These data indicate that the ⌬P1 v-myb mutant and ex vivo bone marrow cell cultures represent a valuable tool for studies on the mechanisms of leukemia formation. © 2001 Academic Press Key Words: v-Myb; oncoprotein; PEST domain; leucine zipper; monoblastic leukemia; ex vivo; bone marrow; proliferation; differentiation; cMGF.

INTRODUCTION

characteristics of transcription activators. They bind specifically to DNA and activate transcription of reporter genes that contain the Mybresponsive element within their promoter sequences (6, 7). The DNA binding and transcription activating properties of v-Myb are crucial for its transforming activity (8). The protein is believed to bind to the regulatory sequences of Myb target genes and to misregulate their transcription, thereby causing a distortion of the integrated signaling network of a myeloid cell. As a result of such misregulation, the differentiation program of myeloid cells is blocked at a stage resembling monoblasts. Moreover, transformed cells acquire additional properties that

The v-myb oncogene, carried by the genome of avian myeloblastosis virus, is a causative agent of acute monoblastic leukemia induced by the virus in susceptible chickens. Its ectopic expression in cultured bone marrow cells results in an outgrowth of transformed monoblasts. The v-myb oncogene represents a terminally truncated and point-mutated derivative of c-myb gene [for reviews, see (1– 4)]. These structural changes, namely the terminal truncations, resulted in the activation of the oncogenic potential of the c-myb proto-oncogene (5). The protein products of v- and c-myb genes possess

This paper summarizes a presentation made at the Second International Workshop on Myb Genes, held in Melbourne on November 22–24, 2000, sponsored in part by the Leukemia & Lymphoma Society (U.S.A.). 1 Institute of Molecular Genetics, Academy of Sciences of the Czech Republic, Flemingovon. 2, 166 37 Prague 6, Czech Republic. 2 Correspondence and reprint requests to: Michal Dvorak. 1079-9796/01 $35.00 Copyright © 2001 by Academic Press All rights of reproduction in any form reserved

437

Blood Cells, Molecules, and Diseases (2001) 27(2) Mar/Apr: 437– 445 doi:10.1006/bcmd.2001.0402, available online at http://www.idealibrary.com on

Dvorakova et al.

mia, and modified the adhesion properties of transformed monoblasts.

allow them to escape physiological control mechanisms. There are two protein domains absolutely required for the transforming activity of v-Myb. The N-terminal DNA-binding domain (9) is responsible for the specific association of Myb proteins with the PyAAC(T/G)G consensus motif (10). This domain also harbors the main signal for nuclear localization (11). The acidic transactivation domain (6) is localized within the central part of the protein. Deletions of or mutations within these domains result in the loss of either DNA-binding, nuclear translocation or transcription activating abilities of the oncoprotein and completely eliminate the transforming potential of v-Myb. The role of other v-Myb regions such as the leucine zipper, has been analyzed recently (12). It was shown that mutations of leucines 3 and 4 of the leucine-zipper motif to alanines totally inhibit the leukemogenicity of v-myb. The in vitro proliferation of monoblasts transformed by the L3,4A v-Myb mutant was shown to be partially dependent on exogenous growth factors and was temperature sensitive. It was also demonstrated that the FAETL region, which is a part of the leucine repeat, is essential for both transcriptional activation and in vitro transformation by v-Myb (13). Here we analyzed the significance of the evolutionarily conserved v-Myb region localized Nterminally to the leucine zipper motif. This Pro-, Glu-, and Ser-rich sequence represents the Cterminus of the v-Myb phosphopeptide (14, 15) and coincides with the potential PEST domain HGCLPEESASPAR (16). According to the PEST hypothesis, P,E,S,T-rich sequences are likely to serve for an important regulatory function by providing a signal for rapid protein degradation (16). In c-Myb however, this domain had no effect on the protein turnover (17). We also found that the deletion of the EESASPARCM sequence in vMyb did not affect the intracellular stability of v-Myb. Moreover, the deletion had no obvious effect on the transcriptional activity and intracellular distribution of v-Myb. However, it changed the v-Myb-induced proliferation of monoblasts, abolished the acute phase of monoblastic leuke-

MATERIALS AND METHODS Leukemogenesis Assay Brown Leghorn C/E gs⫺ chickens from leukosis-free flocks were infected with virus by injecting 0.5 ml of virus-containing tissue culture supernatants into vena saphena as described (12). Animals were inspected for leukemia formation by analyzing peripheral blood smears and micro hematocrits at regular time intervals. Cell Culture Ex vivo cultures of monoblasts and stromal cells were prepared from the bone marrow of 20-day-old chicks infected with myb retroviruses. The cells were grown in 5% CO2 atmosphere at 41°C in the complete media: Dulbecco’s modified Eagle’s medium (DMEM) supplemented with 8% fetal calf serum (FCS, Sebak, Switzerland) and 2% chicken serum (Sigma), 20 mM Hepes, pH 7.3, 100 units/ml Penicillin/Streptomycin (GibcoBRL) and further additives (12). During frequent initial passages, monoblasts were separated from stromal cells. Chicken embryo fibroblasts (CEF) and BM2 cells were grown at 37°C in the same medium. For lipopolysaccharide (LPS) induction, HD11 cells were seeded in serum-free complete medium with insulin-transferrin-sodium selenite media supplement (5 ␮g/ml each, Sigma). LPS (Sigma) was added at the final concentration 5 ␮g/ml for 3 h (18). Primary spleen cells from 3-week-old chicks were cultivated in the complete medium in the presence of 10 ␮g/ml concanavalin A (Sigma) for 4 days. Cell Proliferation Assay Cell proliferation was measured as the rate of DNA synthesis by [3H]thymidine incorporation. Monoblasts (4 ⫻ 104 cells/100 ␮l) were grown in 96-well plates in the assay medium (complete medium containing only 1.8% FCS and 0.2% chicken serum) at 41°C for 36 h. In assays where wt and ⌬P1 v-myb cells were cocultivated, 2 ⫻ 438

Dvorakova et al.

Blood Cells, Molecules, and Diseases (2001) 27(2) Mar/Apr: 437– 445 doi:10.1006/bcmd.2001.0402, available online at http://www.idealibrary.com on

104 cells of each type were mixed per well. Before harvesting, 0.8 ␮Ci of [3H]thymidine (specific activity 29 Ci/mmol; Amersham) was added for 2 h. Average values from triplicate samples were determined. Conditioned media were prepared by growing 3 ⫻ 106 cells/ml in the assay media for 24 h. Preparation of recombinant cMGF was described earlier (12).

cDNA sequence (20). The mutant gene was inserted into the pNeoAMV vector (21). RNase Protection Assay Total RNA was prepared by the standard procedure (22). For RNase protection assays, the 306-bp cDNA fragment of chicken GBX2 (nucleotides 905–1211; GenBank Accession No. AF022151) and the 400-bp cDNA fragment of chicken cMGF (nucleotides 121–521; GeneBank Accession No. X14477) were subcloned into pGEM3Z. The constructs were linearized, transcribed by SP6 polymerase (Ambion) in the presence of [␣-32P]GTP, and radioactive probes were used in RNase protection assays as described previously (23). The chicken actin cDNA fragment was used as a control.

Pulse-Chase Analysis Transformed monoblasts (3 ⫻ 106/ml) were incubated in methionine-free DMEM supplemented with 8% dialyzed FCS for 20 min and labeled for 15 min with 500 ␮Ci/ml [35S]methionine. Half of the cells were washed in PBS and lysed in 0.5% deoxycholate; 0.5% TX-100; 25 mM Tris–FCl, pH 8.1; 50 mM NaCl and 0.1% methionine; 0.1 mg/ml PMSF; 1 ␮g/ml leupeptin and 1 ␮g/ml Trasylol. The remaining cells were washed in isotope free medium, suspended in the complete medium, incubated for 60 min, washed, and lysed. After centrifugation (5 min, 15k) lysates were precleared with anti-rabbit IgG agarose. Immunoprecipitations were carried out at 4°C using rabbit polyclonal antibodies specific for chicken Myb. Anti-rabbit IgG coupled to agarose beads (Sigma) was used as a secondary antibody and immunoprecipitates were analyzed on 10% gel using SDS–PAGE followed by autoradiography.

Transient Transactivation Assay Chicken embryo fibroblasts, CEF (6 ⫻ 105 cells), were seeded onto 60-mm dishes, transfected with pNeov-myb, pNeo⌬P1v-myb, or pNeo empty vector (control), 3xMRE-CAT reporter vector and pRSV ␤-Gal vector. The cells were processed as described earlier (12). The CAT activity was determined by the ELISA (Boehringer Mannheim). The transcriptional activity of wt v-Myb was assumed to be 100%. RESULTS AND DISCUSSION

Subcellular Fractionation of Monoblasts and Western Blot Analysis

PEST-like Sequence Conservation and Role in Leukemogenesis

Ex vivo monoblasts were fractionated using the Triton X-100/ammonium sulfate procedure (19). The protein content in each fraction was measured, the same aliquots were separated on a 10% SDS–PAGE gel and blotted onto ECL membranes (Amersham). Blots were processed as described (12). The 2.32 anti-Myb monoclonal antibody served as the primary antibody.

Two structural elements can be recognized within the C-terminal part of AMV v-Myb, the leucine zipper motif and the PEST-like region. The PEST-like motif, HGCLPEESASPAR, maps to the C-terminus of the v-Myb phosphoprotein directly preceding the downstream leucine-zipper sequence (Fig. 1A). Comparison of various vertebrate myb genes revealed the absolute evolutionary conservation of the LPEESASPAR element (Fig. 1B), which in addition contains a putative p34/cdc2 consensus phosphorylation site. To study the role of the PEST-like motif in the context of v-Myb oncoprotein, the P1 mutant

Plasmid Construction To generate ⌬P1 v-myb, the sequence encoding amino acids 301–311 was deleted. The numbering of amino acids is based on the v-myb 439

Blood Cells, Molecules, and Diseases (2001) 27(2) Mar/Apr: 437– 445 doi:10.1006/bcmd.2001.0402, available online at http://www.idealibrary.com on

Dvorakova et al.

FIG. 1. Schematic representation of the structure of c-Myb and v-Myb proteins and the evolutionary conservation of the primary sequence of the PEST-like domain region. (A) DBD, DNA-binding domain; TA, transactivation domain; PEST, PEST-like element; LZ, leucine-zipper motif; NR, negative regulatory domain. (B) Alignment of bovine (BO), human (HU), murine (MU), chicken (CH), AMV, and Xenopus (XE) myb sequences spanning the PEST-like element.

⌬P1 mutant in several similar experiments, while wt v-myb infection lead to almost 100% mortality. Next, peripheral blood smears and bone marrow cytospins were analyzed in chicks 20 days after infection with wt- and ⌬P1 v-myb viruses. As illustrated in Figs. 3A and 3D, both the peripheral blood and the bone marrow of wt v-myb-infected animals contained large numbers of monoblasts. In ⌬P1 vmyb-infected animals, numerous monoblasts were detectable only in the bone marrow (Fig. 3E), whereas very few could be observed in the periphery (Fig. 3B). Figures 3C and 3F show cells from the periphery and bone marrow of an uninfected animal, respectively. The defect in leukemia formation could be due to a low number of target cells for the ⌬P1 v-Myb mutant. Therefore, colony formation assays in semisolid media were performed. Bone marrow cells were infected in the tissue culture with the same aliquots of wt- and ⌬P1 v-myb viral stocks that were used for infection of experimental animals. Twenty-four hours later the cells were seeded into semisolid medium and colonies were counted 3 weeks later. The number of colonies formed in wtand ⌬P1 v-myb-infected cultures were always very similar (data not shown).

lacking the EESASPARCM peptide was inserted into the retroviral vector. To compare the biological activity of the wt- and ⌬P1 v-myb genes, one-day-old chicks were infected with respective retroviruses and their blood was analyzed at regular intervals. As shown in Fig. 2 all the chicks infected with the wt oncogene displayed a rapidly increasing percentage of monoblasts in the peripheral blood starting on day 17 postinfection. In ⌬P1 v-myb-infected chicks, some immature myeloid and erythroid cells appeared in the periphery approximately 30 days postinfection. Although in some animals blast counts rose up to 20%, they usually returned to normal values after a few days. Up to 95% of animals survived infection with the

In Vitro Growth of Leukemic Monoblasts FIG. 2. Development of leukemia in chicks infected on day 1 by wt- or ⌬P1 v-myb retroviruses. Results of one representative experiment are shown.

To study the properties of leukemic monoblasts, ex vivo cultures were established with bone 440

Dvorakova et al.

Blood Cells, Molecules, and Diseases (2001) 27(2) Mar/Apr: 437– 445 doi:10.1006/bcmd.2001.0402, available online at http://www.idealibrary.com on

FIG. 3. Morphology of peripheral blood (A–C) and bone marrow (D–F) cells from chicks 20 days postinfection with wt(A, D) and ⌬P1 (B, E) v-myb retroviruses. Similar samples from an uninfected animal (C, F) were used as a control.

marrow cells from chicks infected with wt- and ⌬P v-myb. The proliferation rates of transformed cells were initially very similar (Fig. 4); only after 14 days of culture did ⌬P1 v-myb cells display longer doubling times than wt v-myb cells. To show that ⌬P1 v-myb gene is stable and does not undergo any rearrangement during frequent reinfections in the experimental animal, in one experiment, RNA was isolated from ⌬P1 v-myb-transformed cells and the 3⬘ part of v-myb cDNA was amplified by the RT-PCR procedure. DNA se-

quencing revealed no deviations from the original vector construct (data not shown). Biochemical Properties of ⌬P1 vMyb in Leukemic Monoblasts The above-mentioned results demonstrate that monoblasts transformed with ⌬P1 v-myb do not differ from wt-transformed cells in vitro. However, they appear to be handicapped in vivo, particularly in the peripheral blood. To identify the cause of ⌬P1 v-Myb biological deficiency, several properties of ⌬P1 v-Myb were compared to those of the wt oncoprotein. The pulse-chase experiment revealed that the metabolic turnover of both proteins was very similar (Fig. 5A). In addition, no apparent difference was observed in their subcellular localization (Fig. 5B), including their association with DNA sensitive to DNasel treatment. Transient transfection experiments in fibroblasts showed that the ⌬P1 v-Myb reproducibly activated transcription of 3xMRE-CAT reporter 1.5 to 2 times more than the wt v-Myb (Fig. 5C).

FIG. 4. Proliferation of bone marrow monoblasts transformed by wt- and ⌬P1 v-myb in ex vivo cultures. Data from one representative experiment are shown. 441

Blood Cells, Molecules, and Diseases (2001) 27(2) Mar/Apr: 437– 445 doi:10.1006/bcmd.2001.0402, available online at http://www.idealibrary.com on

Dvorakova et al.

FIG. 5. (A) Intracellular stability of wt- and ⌬P1 v-Myb proteins. SDS–PAGE of immunoprecipitates from cell lysates at the end of pulse with [35S]methionine (lanes 2 and 4) and after the 60-min chase (lanes 3 and 5). Lane 1 contains precipitate from wt v-myb cells by preimmune antiserum. (B) Distribution of wt- (odd lanes) and ⌬P1 (even lanes) v-Myb proteins in subcellular fractions of ex vivo monoblasts. Proteins from different fractions (specified above the picture) were analyzed by the Western blot procedure. (C) Transcriptional activity of wt- and ⌬P1 v-Myb proteins measured in CAT assays. The empty (myb-less) vector was used in control experiments.

Thus, none of these v-Myb properties were deleteriously affected by the P1 deletion. Growth Factor Dependency of ⌬P1 Monoblasts However, a substantial difference between wt and ⌬P1 v-myb-transformed ex vivo bone marrow cells was found in standard proliferation assays performed in low serum. While in 10% serum no significant difference in thymidine incorporation was observed (Fig. 6A), in 2% serum, the DNA synthesis in wt- and ⌬P1 cells was reduced by 50% and by more than 90%, respectively (Fig. 6B). Addition of 10 ng/ml of cMGF induced thymidine incorporation to the level comparable with that achieved in 10% serum (Fig. 6C). The growth arrest of ⌬P1 cells was not accompanied by differentiation. Therefore, the P1 deletion is not equivalent to inactivation of v-Myb in conditional mutants (24, 25). It rather appears that P1 deletion impaired only the proliferation-inducing activity of the oncoprotein while the differentiation block was unaffected. Since it is possible to partially reactivate proliferation of ⌬P1 v-myb monoblasts after several days of serum starvation, this system can be used for the dissection and studies of pathways controlling terminal differen-

FIG. 6. Proliferation assays of ex vivo bone marrow cells transformed by wt- and ⌬P1 v-myb in 10% serum (A) and in 2% serum (B–E). cMGF was used at 10 ng/ml (C). (D) Cultures were supplemented by 40% media conditioned by wt- (wtCM) or ⌬P1 (⌬P1CM) ex vivo monoblasts, or by BM2 (BM2CM) cells. (E) Cocultivation experiment. 442

Dvorakova et al.

Blood Cells, Molecules, and Diseases (2001) 27(2) Mar/Apr: 437– 445 doi:10.1006/bcmd.2001.0402, available online at http://www.idealibrary.com on

tiation and proliferation of monocyte/macrophage cells. As a first step in this direction, the serum independent proliferation of wt v-myb transformed monoblasts was analyzed. To find out whether this is due to autocrine stimulation induced by wt v-Myb, experiments with conditioned media (CM) were performed. Surprisingly, wtCM had no effect on DNA synthesis in ⌬P1 v-myb monoblasts. In contrast, the media conditioned by BM2 cell line activated proliferation of ⌬P1 cells substantially (Fig. 6D). The lack of activation by wtCM might be due to the extreme instability of the growth factor or to a strict association with the surface of factor-producing cells. To allow the direct contact, wt and ⌬P1 v-myb cells were cocultivated. Figure 6E shows that the cocultivation does not activate proliferation of ⌬P1 v-myb cells either, as thymidine incorporation values in mixed cultures always represented the arithmetic sum of respective thymidine incorporation values achieved by cells grown separately. These data show that in primary leukemic cells, v-Myb allows growth factor independence by a mechanism different from the autocrine stimulation by cMGF suggested by others (26). Therefore, expression of GBX2 and cMGF in v-mybtransformed cells was examined. The RNAse protection assays reproducibly showed that primary ex vivo monoblasts transformed by wt v-Myb did not express GBX2 nor cMGF mRNAs (Fig. 7, lanes wt). However, low levels of GBX2 and cMGF mRNAs were detected in growth factor dependent ⌬P1 v-myb cells (Fig. 7, lanes ⌬P1). Rather high expression of both genes was observed in the BM2 cell line (Fig. 7, lanes BM2) suggesting that BM2 cells are likely to produce cMGF. This correlates well with the stimulatory effect of BM2-conditioned media on the proliferation of ⌬P1 v-myb cells (Fig. 6D). This is also in agreement with earlier studies demonstrating the presence of cMGF in BM2 cell-conditioned media (27). On the other hand, the primary monoblasts transformed by wt AMV v-myb neither synthesize cMGF nor do they seem to liberate any soluble growth factor that would activate their proliferation in an autocrine manner. These results support the conclusion that v-Myb generates an intracellular signal required for in vitro and in

FIG. 7. Determination of cMGF (A) and GBX2 (B) mRNAs by the RNase protection assay in primary chicken splenocytes activated by concanavalin A (spleen ⫹ con A); BM2 cell line (BM2); wt- and ⌬P1 ex vivo bone marrow monoblasts (wt, ⌬P1); HD11 macrophage cell line (HD11) and in HD11 cells activated by lipopolysaccharide (HD11 ⫹ LPS).

vivo proliferation of v-Myb-transformed myeloid cells. This signal, mediated by the leucine zipper region (LZR, composed of P1 and LZ sequences), appears to substitute for activity of extracellular growth factors like cMGF. Indeed, cMGF was found to rescue the proliferative potential of the v-Myb LZ mutant in tissue culture (12). Most importantly, cMGF ectopically coexpressed with nonleukemic v-Myb lacking LZR sequences rescued its leukemic potential in vivo (28). Cell Adhesion To further analyze the effect of P1 deletion, the adhesion properties of wt and ⌬P1 v-myb ex vivo monoblasts were analyzed in pilot experiments. It was observed that their self-adherence differs and strongly depends on the culture con443

Blood Cells, Molecules, and Diseases (2001) 27(2) Mar/Apr: 437– 445 doi:10.1006/bcmd.2001.0402, available online at http://www.idealibrary.com on

Dvorakova et al.

FIG. 8. Interaction of transformed monoblasts with stromal cells in ex vivo bone marrow cultures from chicks infected with wt- and ⌬P1 v-myb viruses.

ditions (data not shown). Moreover, preliminary observations suggest that ⌬P1 cells in ex vivo cultures strongly adhere to fibroblast-like stromal cells, whereas wt v-Myb monoblasts display a rather weak adherence (Fig. 8). The adhesion properties of ⌬P1 v-Myb monoblasts could be one of the reasons why these cells do not massively infiltrate the peripheral blood. Thus, the ⌬P1 vMyb mutant cells in ex vivo cultures might also facilitate analysis of the mechanisms that control mobilization of immature myeloid cells and their transport into the peripheral blood. This study, as well as the accompanying work (12, Karafiat et al., submitted) shows that the leucine zipper region (LZR) represents a regulatory domain with a specific impact on v-Mybinduced phenotype. This region is crucial for the growth factor independent proliferation and leukemic properties of v-myb-transformed myeloid cells (12, this article). However, it does not seem to participate in the differentiation block caused by v-Myb in monoblasts. Moreover, as has been found recently, the LZR biases the development of a common myeloid progenitor in favor of the monocyte/macrophage lineage and thus appears to be involved in the regulation of lineage commitment of progenitor cells (Karafiat et al., submitted). Thus, the LZR in v-Myb strongly supports development and proliferation of immature monocyte/macrophage cells. Its role in c-Myb biology remains to be determined. Further exploi-

tation of the experimental system based on primary chicken cells (ex vivo bone marrow or blastoderm) infected with specific LZR mutants of v-Myb and c-Myb could help in elucidating the molecular basis of cell cycle activation and lineage commitment regulation in hematopoietic cells. It may also facilitate studies on the mechanisms that control mobilization of blast-like cells from the bone marrow into the periphery. Experiments along these lines are in progress. ACKNOWLEDGMENTS This work was funded by grants from the Grant Agency of the Czech Republic (301/98/K042 and 204/ 00/0554), by Grant A5052805 from the Grant Agency of the Academy of Sciences of the Czech Republic to M.D., and by a grant from the Howard Hughes Medical Institute (HHMI, 75195-540401) to M.D. M. Dvorak is an International Research Scholar of the Howard Hughes Medical Institute.

REFERENCES 1.

2.

444

Graf, T. (1992) Myb: A transcriptional activator linking proliferation and differentiation in hematopoietic cells. Curr. Opin. Genet. Dev. 2, 249 –255. Introna, M., Luchetti, M., Castellano, M., Arsura, M., and Golay, J. (1994) The myb oncogene family of transcription factors: Potent regulators of hematopoietic cell proliferation and differentiation. Semin. Cancer. Biol. 5, 113–124.

Dvorakova et al.

3. 4. 5.

6.

7.

8.

9.

10.

11.

12.

13.

14.

15.

16.

Blood Cells, Molecules, and Diseases (2001) 27(2) Mar/Apr: 437– 445 doi:10.1006/bcmd.2001.0402, available online at http://www.idealibrary.com on

Ness, S. A. (1996) The Myb oncoprotein: Regulating a regulator. Biochim. Biophys. Acta 1288, F123–F139. Lipsick, J. S., and Wang, D. M. (1999) Transformation by v-Myb. Oncogene 18, 3047–3055. Grasser, F. A., Graf, T., and Lipsick, J. S. (1991) Protein truncation is required for the activation of the c-myb proto-oncogene. Mol. Cell. Biol. 11, 3987– 3996. Weston, K., and Bishop, J. M. (1989) Transcriptional activation by the v-myb oncogene and its cellular progenitor, c-myb. Cell 58, 85–93. Sakura, H., Kanei-Ishii, C., Nagase, T., Nakagoshi, H., Gonda, T. J., and Ishii, S. (1989) Delineation of three functional domains of the transcriptional activator encoded by the c-myb protooncogene. Proc. Natl. Acad. Sci. USA 86, 5758 –5762. Lane, T., Ibanez, C., Garcia, A., Graf, T., and Lipsick, J. (1990) Transformation by v-myb correlates with trans-activation of gene expression. Mol. Cell. Biol. 10, 2591–2598. Klempnauer, K. H., and Sippel, A. E. (1987) The highly conserved amino-terminal region of the protein encoded by the v-myb oncogene functions as a DNAbinding domain. EMBO J. 6, 2719 –2725. Biedenkapp, H., Borgmeyer, U., Sippel, A. E., and Klempnauer, K. H. (1988) Viral myb oncogene encodes a sequence-specific DNA-binding activity. Nature 335, 835– 837. Ibanez, C. E., and Lipsick, J. S. (1988) Structural and functional domains of the myb oncogene: Requirements for nuclear transport, myeloid transformation, and colony formation. J. Virol. 62, 1981–1988. Bartunek, P., Karafiat, V., Dvorakova, M., Zahorova, V., Mandikova, S., Zenke, M., and Dvorak, M. (1997) The Myb leucine zipper is essential for leukemogenicity of the v-Myb protein. Oncogene 15, 2939 – 2949. Fu, S. L., and Lipsick, J. S. (1996) FAETL motif required for leukemic transformation by v-Myb. J. Virol. 70, 5600 –5610. Bading, H., Beutler, C., and Moelling, K. (1989) Mapping of a small phosphopeptide at the carboxyterminus of the viral myb protein by monoclonal antibodies. Oncogene 4, 33–38. Boyle, W. J., Smeal, T., Defize, L. H., Angel, P., Woodgett, J. R., Karin, M., and Hunter, T. (1991) Activation of protein kinase C decreases phosphorylation of c-Jun at sites that negatively regulate its DNA-binding activity. Cell 64, 573–584. Rogers, S., Wells, R., and Rechsteiner, M. (1986) Amino acid sequences common to rapidly degraded

17.

18.

19.

20.

21.

22.

23.

24.

25.

26.

27.

28.

445

proteins: The PEST hypothesis. Science 234, 364 – 368. Bies, J., Nazarov, V., and Wolff, L. (1999) Identification of protein instability determinants in the carboxy-terminal region of c-Myb removed as a result of retroviral integration in murine monocytic leukemias. J. Virol. 73, 2038 –2044. Leutz, A., Beug, H., Walter, C., and Graf, T. (1988) Hematopoietic growth factor glycosylation. Multiple forms of chicken myelomonocytic growth factor. J. Biol. Chem. 263, 3905–3911. Fey, E. G., Krochmalnic, G., and Penman, S. (1986) The non-chromatin substructures of the nucleus: The ribonucleoprotein (RNP)-containing and RNP-depleted matrices analyzed by sequential fractionation and resinless section electron microscopy. J. Cell Biol. 102, 1654 –1665. Bartunek, P., Kozmik, Z., and Dvorak, M. (1992) Cloning and nucleotide sequence of the 5⬘ part of v-myb cDNA. Virology 190, 882– 883. Lipsick, J. S., Ibanez, C. E., and Baluda, M. A. (1986) Expression of molecular clones of v-myb in avian and mammalian cells independently of transformation. J. Virol. 59, 267–275. Chomczynski, P., and Sacchi, N. (1987) Single-step method of RNA isolation by acid guanidinium thiocyanate–phenol– chloroform extraction. Anal. Biochem. 162, 156 –159. Kaspar, P., Dvorakova, M., Kralova, J., Pajer, P., Kozmik, Z., and Dvorak, M. (1999) Myb-interacting protein, ATBF1, represses transcriptional activity of Myb oncoprotein. J. Biol. Chem. 274, 14422–14428. Burk, O., and Klempnauer, K. H. (1991) Estrogendependent alterations in differentiation state of myeloid cells caused by a v-myb/estrogen receptor fusion protein. EMBO J. 10, 3713–3719. Engelke, U., Wang, D. M., and Lipsick, J. S. (1997) Cells transformed by a v-Myb– estrogen receptor fusion differentiate into multinucleated giant cells. J. Virol. 71, 3760 –3766. Kowenz-Leutz, E., Herr, P., Niss, K., and Leutz, A. (1997) The homeobox gene GBX2, a target of the myb oncogene, mediates autocrine growth and monocyte differentiation. Cell 91, 185–195. Dini, P. W., and Lipsick, J. S. (1993) Oncogenic truncation of the first repeat of c-Myb decreases DNA binding in vitro and in vivo. Mol. Cell. Biol. 13, 7334 –7348. Metz, T., Graf, T., and Leutz, A. (1991) Activation of cMGF expression is a critical step in avian myeloid leukemogenesis. EMBO J. 10, 837– 844.