Introduction

The vertebrate Myb family of transcription factors consists of three members, A-Myb, B-Myb and c-Myb. A-Myb and c-Myb have specialized functions in spermatogenic cells and immature haematopoietic cells, respectively. These functions are mirrored by the phenotypes of knockout mice: disruption of c-myb causes defects of fetal hematopoiesis and embryonic death and A-myb knockout mice are infertile due to defective spermatogenesis (Mucenski et al., 1991; Toscani et al., 1997). By contrast, B-Myb appears to play a more general role in the cell cycle of all proliferating cells (for a review see Joaquin and Watson, 2003). In the mouse embryo, B-Myb is expressed in virtually all dividing tissues whereas nonproliferating cells lack B-Myb expression (Sitzmann et al., 1996). B-myb knockout mice show proliferation defects of the cells of the inner cell mass of the blastocyst causing early embryonic death (Tanaka et al., 1999). B-myb transcription peaks at the late G1/early S phase of the cell cycle (Lam and Watson, 1993); in addition, the activity of B-Myb is controlled by post-translational modification and interaction with other proteins. Notably, phosphorylation of B-Myb by cyclin A/Cdk2 stimulates the transactivation potential of B-Myb by relieving repressive effects exerted by the C-terminus of the protein (Lane et al., 1997; Ziebold et al., 1997). Several B-Myb target genes have been identified, such as the bcl-2, apolipoprotein J, cdc2 and cyclin A and B genes, suggesting a crucial role of B-Myb in cell cycle regulation and control of apoptosis (Grassilli et al., 1999; Cervellera et al., 2000; Zhu et al., 2004; Lang et al., 2005). In addition, there is evidence that B-Myb may exert some of its functions independently of its transcriptional activity. For example, B-Myb has been shown to overcome a G1 cell cycle block imposed by p107 apparently independent of the activation of specific target genes (Joaquin et al., 2002). However, although these and other findings collectively suggest an important role in the cell cycle no clear picture of the function of B-Myb has emerged yet.

The chicken DT40 cell line is an extremely useful model to study gene function by targeted gene disruption because homologous recombination occurs at very high frequencies in these cells (Buerstedde and Takeda, 1991). We have engineered DT40 cells to express endogenous B-myb in a doxycyclin-dependent manner. Surprisingly, we find that B-myb is not essential for cell proliferation per se; rather, our data suggest a novel function for B-myb in the response to DNA-damage.

Results

Targeted disruption of B-myb

To disrupt one copy of B-myb we transfected DT40 cells with a targeting construct, in which the promoter region and exons 1–6 were replaced by a histidinol resistance cassette. As shown in Figure 1a, homologous recombination of the targeting construct leads to the appearance of a novel 10 kb BglII restriction fragment. Southern blot analysis of one of several histidinol-resistant clones is shown in Figure 1c. To generate DT40 cells expressing B-myb conditionally, we then stably introduced a 'Tet-Off' Tet-VP16 expression vector into the heterozygous B-myb clone. Subsequently, a second targeting construct was transfected in which the B-myb promoter was replaced by the puromycin resistance gene and a promoter cassette containing the minimal CMV promoter and several copies of the Tet-operator (Figure 1b). Successful targeting of the second copy of B-myb results in the loss of a 10 kb EcoRI fragment (corresponding to the wild-type (wt) copy of B-myb) and the appearance of a novel 4.4 kb fragment. The analysis of a positive clone, designated as (-/^TRE), is shown in Figure 1c. The 7 kb fragment is derived from the B-myb copy already targeted in the first round of recombination (see Figure 1a). The 4 kb EcoRI band is due to hybridization of the probe with a downstream fragment. Of three independent B-myb (-/^TRE) clones two (clones 5 and 62) were studied further.

Figure 1.

Generation of DT40 B-myb (-/^TRE) cells. (a, b) Disruption of B-myb (a) and insertion of a tetracycline-regulatable promoter (b). The B-myb locus, the targeting constructs and the targeted locus are depicted at the top, middle and bottom, respectively. Exons and the selection cassettes are indicated by black and white boxes. The B-myb promoter and the tet-inducible promoter are marked by black and white arrows, respectively. The probe region is indicated by a white bar and the fragments detected by the probe are shown as black bars. Relevant restriction sites: E (EcoRI), Bg (BglII), Ag (AgeI), X (XhoI) and N (NotI). (c) Southern blot of DT40 wt, () and (-/^TRE) cells. BglII- or EcoRI-digested genomic DNA was hybridized with the probe shown in (a) and (b). (d) RNA from wt DT40 cells and two B-myb (-/^TRE) clones grown in the absence or presence of 1 g/ml doxycyclin for three days was analysed by Northern blotting with probes specific for B-myb and S17. (e) Cell lysates from wt DT40 cells and two B-myb (-/^TRE) clones grown for three days in the absence or presence of different amounts of doxycyclin were analysed by Western blotting using B-Myb specific antibodies. Numbers at the bottom indicate the concentration (g/ml) of doxycyclin

Full figure and legend (174K)

To confirm doxycyclin-dependent downregulation of B-myb expression RNA isolated from B-myb (-/^TRE) clones and wt cells was analysed by Northern blotting. Figure 1d shows that the amount of B-myb mRNA in wt cells was not affected by doxycyclin whereas it was detected in the (-/^TRE) clones only in the absence of doxycyclin. Western blotting showed that B-Myb was only detectable in the (-/^TRE) clones in the absence of doxycyclin (Figure 1e). By contrast, wt cells showed no effect of doxycyclin on the amount of B-Myb. We used a concentration of 1 g/ml of doxycyclin for most subsequent experiments; however, it is evident that even lower concentrations effectively downregulate B-Myb expression. We conclude from the data shown in Figure 1 that we have successfully inactivated one copy of the endogenous B-myb and converted the other copy to a conditional form.

Growth characteristics and gene expression of B-myb (-/^TRE) cells

We next studied the growth of B-myb (-/^TRE) cells. Surprisingly, there was no significant effect of doxycyclin on the growth rate of either wt or B-myb (-/^TRE) cells (Figure 2a–c). We also determined the cell cycle distribution of the different cell populations by flow cytometry. The resulting patterns (Figure 3) looked very similar in each case. Taken together, our data suggest that the absence of B-Myb in DT40 cells does not lead to overt distortion of the cell cycle.

Figure 2.

Growth characteristics and cell cycle distribution of DT40 B-myb (+/+) and (-/^TRE) cells. (a–c) The growth of wt DT40 cells (a) and two B-myb (-/^TRE) clones (b, c) was followed in the absence (white squares) or presence (black circles) of 1 g/ml doxycyclin. Cumulative cell numbers were plotted against the time in culture. (d) Northern blot analysis of B-myb (-/^TRE) cells grown for 10 weeks in the presence or absence of doxycyclin. RNA isolated from B-myb (-/^TRE) clone 62 grown for 10 weeks in the presence or absence of 1 g/ml doxycyclin was analysed by Northern blotting with probes specific for B-myb, A-myb, c-myb and S17

Full figure and legend (94K)

Figure 3.

Wt DT40 cells and two B-myb (-/^TRE) clones grown in the presence of doxycylin for 3 days or in the absence of doxycyclin were stained with propidium iodide and analysed by flow cytometry. The fluorescence intensity is plotted against the cell number. Peaks of G1- and G2-phase cells are marked

Full figure and legend (64K)

Since these results were unexpected, we wondered whether B-Myb expression remains undetectable after prolonged growth in the presence of doxycyclin or whether the levels of A-Myb or c-Myb increase over time to compensate for the lack of B-Myb. To address this issue, we cultivated B-myb (-/^TRE) for 10 weeks either in the presence or absence of doxycyclin. As doxycylin has a short half-life in tissue culture, fresh doxycyclin was added on a regular basis to keep the concentration around 1 g/ml at all times. The cells were then analysed by Northern blotting (Figure 2d). As shown in Figure 2d, even after continuous cultivation of the cells in the presence of doxycyclin there was no obvious upregulation of B-myb mRNA. Furthermore, the levels of A-myb and c-myb mRNA were affected only slightly.

To identify gene expression changes induced in B-myb (-/^TRE) cells by doxycyclin, we analysed mRNA from these cells using chicken cDNA microarrays that represent approximately 2200 individual genes (Neiman et al., 2001). cDNA probes labelled with Cy3 and Cy5 were prepared using RNA from B-myb (-/^TRE) cells grown for 3 days with or without doxycyclin and hybridized in appropriate combinations to separate microarrays to generate dye-reversed replicate data sets. The position of each dot in the diagram shown in Figure 4 reflects the difference in expression of the corresponding gene in the absence or presence of B-Myb as determined by dye-reversed hybridization experiments. Surprisingly, there were no reproducible gene expression changes greater than twofold (such changes would appear as dots in areas 'a' and 'c'). Some of the genes close to areas 'a' and 'c' were analysed individually by Northern blotting; however, no significant effects of B-Myb on their expression were observed (data not shown). The positions of several putative B-Myb target genes are indicated in Figure 4. None of these genes, including bcl-2, apoliprotein J and cyclin A and B, was affected significantly by B-Myb.

Figure 4.

cDNA microarray analysis of B-myb (-/^TRE) cells. RNA isolated from B-myb (-/^TRE) clone 62 grown for 3 days in the presence or absence of doxycyclin was labelled with Cy3 or Cy5, respectively, mixed and hybridized to chicken cDNA microarrays. Each dot represents a cDNA spot of the array. Its position along the vertical axis reflects the fold induction (values larger than 1) or repression (values smaller than 1) in B-myb (-/^TRE) cells grown without doxycyclin versus the same cells grown with doxycyclin. The position of each dot along the horizontal axis reflects the fold induction or repression in a dye-reversed experiment. Area 'b' contains genes whose expression is reproducibly induced or repressed twofold or less by B-Myb. Genes reproducibly induced or repressed twofold or more by B-Myb would fall into the areas 'a' or 'c', respectively. Dots outside of the marked areas correspond to genes whose expression differs by more than twofold in only one of the hybridization experiments. The positions of some genes are marked as follows: 1: apo J; 2: bcl-2; 3: cyclin A; 4: cyclin B

Full figure and legend (25K)

DT40 cells lacking B-Myb show increased sensitivity to DNA-damage

Since there were no obvious effects of doxycyclin on B-myb (-/^TRE) cells under normal growth conditions, we considered the possibility that B-Myb may be required only under certain circumstances. In particular, we were interested to know if B-Myb plays a role in the response to DNA-damage. DT40 wt and B-myb (-/^TRE) cells kept in the presence or absence of doxycyclin were irradiated with UV-light. Identical numbers of cells were then cultivated with or without doxycyclin and the number of viable cells was determined three days later. Figure 5a shows that B-myb (-/^TRE) cells were more sensitive to UV-irradiation in the presence than in the absence of doxycyclin while wt cells were not affected by doxycyclin. This suggests that B-Myb may play a role in the response to UV-induced DNA-damage. Figure 5e shows a representative growth curve of DT40 wt and B-myb (-/^TRE) cells following UV irradiation. The increase of the cell numbers in all cases excludes the possibility that the B-myb (-/^TRE) cells simply die faster in the presence of doxycyclin than in its absence and demonstrates that there is a difference in survival of these cells in the presence or absence of doxycyclin.

Figure 5.

Effect of DNA damage on wt and B-myb (-/^TRE) DT40 cells. (a–d) Cultures of DT40 wt and B-myb (-/^TRE) clone 62 cells were kept with or without 1 g/ml doxycyclin for 2 days and were then treated as indicated at the top of each panel. UV irradiation was performed at 25–50 J/m² and, caffeine and MMS were used at 1 mM and 20 M, respectively. Etoposide was used at concentrations between 10 and 100 nM with essentially identical results. The columns show the number of viable cells after 3 days in the culture. Cell numbers in cultures containing doxycyclin are normalized relative to the cell number of the same type of cells in the cultures without doxycyclin. (e) The growth of wt DT40 wt (triangles) and B-myb (-/^TRE) clone 62 cells (circles) was followed in the absence (white symbols) or presence (black symbols) of 1 g/ml doxycyclin. The cells were kept in the presence or absence of doxycyclin for 48 h, then identical numbers of cells were UV-irradiated (day 0) and cultivated further. The number of viable cells was determined over a period of 7 days and plotted against the time in culture. (f) Cultures of DT40 wt and B-myb (-/^TRE) clone 62 cells kept in the absence or presence of 1 g/ml doxycyclin were analysed for the presence of apoptotic cells using annexin V staining. Cells were analysed without UV-irradiation (left panel) or after UV-irradiation (100 J/m²) and further cultivation for 8 h before analysis (right panel). The columns show the percentage of annexin-V positive cells in each cell population

Full figure and legend (100K)

We also measured the number of cells that undergo apoptosis after UV-irradiation. wt and B-myb (-/^TRE) cells grown in the presence or absence of doxycyclin were UV-irradiated and cultivated for 8 h. Apoptotic cells were then quantified by annexin V staining using a flow cytometer. As control, cells were also analysed without UV-irradiation. Figure 5f shows that small numbers of apoptotic cells were detected in all the nonirradiated cultures. There was only a slight effect of doxycyclin on wt cells and no significant effect on B-myb (-/^TRE) cells, demonstrating that the absence of B-Myb does not cause apoptosis per se. At 8 h after UV-irradiation the number of apoptotic B-myb (-/^TRE) cells was significantly higher in the presence than in the absence of doxycyclin while doxycyclin had only little effect on wt cells. Thus, DT40 cells undergo UV-induced apoptosis more readily when B-Myb is absent, supporting the notion that B-Myb plays a role in the response to UV-induced DNA-damage.

Eukaryotic cells trigger DNA-damage responses primarily by activating the ATM- and ATR protein kinases (Nyberg et al., 2002). To investigate if B-Myb acts downstream of ATM or ATR, we UV-irradiated B-myb (-/^TRE) and wt cells in the presence of caffeine, which abrogates ATM- and ATR-dependent DNA-damage responses by inhibiting the activities of these protein kinases. Figure 5b shows that caffeine did not abolish the increased sensitivity to UV-irradiation of B-myb (-/^TRE) cells in the presence of doxycyclin (as would be expected if B-Myb is a downstream effector of ATM or ATR), but rather exaggerated the effect of UV-irradation on B-myb (-/^TRE) cells. Thus, B-Myb is unlikely to act as part of an ATM or ATR damage response.

Finally, we were interested to know whether the absence of B-Myb also affects the response to other agents causing DNA-damage. Methyl methanesulfonate (MMS), a DNA alkylating agent, had a similar effect on B-myb (-/^TRE) cells as UV-irradiation (Figure 5c). Enhanced sensitivity towards DNA damage was, however, not generally observed in cells lacking B-Myb. Etoposide, which induces DNA breaks via topoisomerase II inhibition, had similar effects in DT40 wt and B-myb (-/^TRE) cells (Figure 5d).

Discussion

B-Myb is a cell cycle-regulated transcription factor, which is thought to play a key role in the cell cycle of vertebrate cells. Despite recent progress in understanding how B-Myb expression and activity is regulated and the identification of several putative B-Myb target genes the role of B-Myb in proliferating cells is not yet well understood. To explore the function of B-Myb, we have used the chicken DT40 cell line as a model to generate cells in which the level of endogenous B-Myb can be regulated by doxycyclin. Surprisingly, the absence of B-Myb does not affect the proliferation of DT40 cells or the expression of a set of 2200 different chicken genes, including the putative B-Myb target genes, bcl-2, apo J, and the cyclin A and B genes. Our results therefore clearly demonstrate that B-Myb, contrary to the current view, is not an essential component of the cell cycle machinery and does not contribute to the expression of these genes, at least in DT40 cells.

How can our data be reconciled with the previously published observation that B-Myb deficient mice are very early embryonic lethals (Tanaka et al., 1999)? On one hand, since the molecular basis of embryonic lethality in these mice has not been clarified, it remains possible that B-Myb plays a crucial role early in embryonic development, which is unrelated to cell proliferation. It is also possible that the requirement of B-Myb for cell proliferation differs between early embryonic and more differentiated cells. In this respect, it is interesting to note that embryonic stem cell lines express B-Myb at much higher levels than other proliferating cells (Kamano et al., 1995). On the other hand, we cannot completely exclude the possibility that the Tet-responsive promoter shows some leakiness in the absence of Tet-VP16 and that the B-myb (-/^TRE) cells described here therefore have a residual level of B-Myb expression, which is below the limit of detection but which, nevertheless, affects the behaviour of the cells. We also cannot rigorously exclude the possibility that other Myb family members, such as c-Myb or A-Myb, can compensate the loss of B-Myb to some extent, thus providing a possible explanation for the absence of overt effects of B-Myb on cell proliferation and gene expression. Such an explanation is, however, difficult to reconcile with the completely divergent phenotypes of the different myb knockout mice or the different biochemical properties of A-Myb, B-Myb and c-Myb. For example, c-Myb, unlike B-Myb, is not regulated by cell cycle-dependent phosphorylation, making it unlikely that c-Myb can functionally replace B-Myb. In fact, some of the alleged B-Myb target genes, such as the cyclin A and B genes, have been shown not to be targets of c-Myb (Zhu et al., 2004). A further argument for a unique function of B-Myb is provided by recent work on Drosophila melanogaster. Drosophila has a single myb gene (Dm-myb) that has been implicated in cell proliferation and in controlling the activity of certain origins of DNA replication (Katzen et al., 1998; Beall et al., 2002; Fitzpatrick et al., 2002). Recent work has shown that B-myb, but not A-myb or c-myb, is able to complement a defective Dm-myb gene (Davidson et al., 2004), suggesting a unique and probably ancestral function for B-myb, which cannot be complemented by other vertebrate myb family members.

We have made the intriguing observation that cells lacking B-Myb survive DNA-damage induced by UV-irradiation and alkylating agents less well than cells expressing B-Myb. Our work, therefore, provides the first direct evidence suggesting a role for B-Myb in the response to DNA-damage rather than in cell proliferation per se.

The DT40 cell line is remarkable for its persistent high frequency of immunoglobulin gene conversion, which is typical of immature B-cells in chickens but not mice or humans. It is therefore possible that the role of B-Myb DNA-damage repair might be more extensive than is evident from our data because certain aspects of DNA-damage repair might be masked in these cells due to the hyperreactivity of homologous recombination.

It is interesting to note that Drosophila Dm-myb mutants accumulate chromosomal abnormalities (Manak et al., 2002), implicating Dm-myb in the maintenance of genomic stability. Furthermore, one of the proteins interacting with B-Myb, poly-(ADP-ribose) polymerase (PARP) (Cervellera and Sala, 2000), has also been implicated in maintaining genomic stability (Huber et al., 2004), thus providing another potential link between B-Myb and DNA-damage response. How B-Myb acts in response to DNA damage remains to be explored. In light of the apparent functional similarities between B-Myb and Dm-Myb it will be interesting to see whether B-Myb only modulates transcriptional processes or whether it might also have nontranscriptional functions involved in the control of DNA replication processes.

Materials and methods

Targeting constructs

The B-myb targeting vector pBMKO-his-rev was generated by cloning a 2.5 kb EcoRI/BglII fragment from the B-myb upstream region and a 0.6 kb BglII/NotI fragment from the intron between exons 6 and 7 between the EcoRI and NotI sites of pbluescript. The histidinol resistance cassette was then inserted at the BglII site between the two genomic fragments. Homologous recombination of the targeting construct with B-myb will delete the B-myb promoter and exons 1–6 of the gene. To create a tetracycline-inducible copy of B-myb the short arm of the construct described above was replaced by an AgeI/NotI fragment spanning exons 4–6, the AgeI site being located in exon 4. Upstream of the AgeI site we inserted a XhoI/AgeI fragment that was derived from a plasmid containing the chicken B-myb coding region under the control of a tetracycline-responsive minimal CMV promoter (pTRE-B-Myb). Finally, the histidinol-resistance cassette was replaced by a puromycin-resistance cassette. The resulting targeting construct (pBMKO-TRE-B-myb-puro-rev) will replace the B-myb promoter and exon 1–3 by the tetracycline-responsive promoter and B-myb cDNA sequences to restore the complete B-myb coding region.

Growth curves and cell viability measurements

Growth curves were determined by seeding cells at the desired concentration and counting them at the indicated time points with a haemacytometer. Viable cells were identified by Trypan blue exclusion.

DNA-transfection and blotting

DT40 cells were grown, transfected and analysed by Southern blotting as described (Appl and Klempnauer, 2002). Polyadenylated RNA was prepared using oligo-dT cellulose (Klempnauer and Bishop, 1983) and chicken B-myb, A-myb and c-myb mRNA, and B-Myb protein was detected as described (Foos et al., 1992, 1994).

Induction of DNA-damage

UV-irradiation was performed as described (Appl and Klempnauer, 2002) at 25–50 J/m². Methyl methanesulfonate (MMS) was used at a final concentration of 20 M. Etoposide was used at concentrations between 10 and 100 nM. Cells were cultivated in the presence of the drugs for prolonged times (usually 2 days) until they were split. Caffeine was used and 1 mM final concentration. Cells to be subjected to DNA-damage in the presence of doxycyclin were cultivated for 48–72 h in the presence of doxycyclin before initiating DNA-damage.

Flow cytometry

Cell cycle analysis was performed as described (Appl and Klempnauer, 2002), using a FACScan flow cytometer (Becton Dickinson). Apoptotic cells were quantified with a flow cytometer after staining with annexin V-FITC as described previously (Lang et al., 2002). Briefly, cells were labelled after UV-irradiation with annexin V-FITC (BD Pharmingen) in staining buffer (PBS containing Ca²⁺ and Mg²⁺, 0.05% FCS; 0.5 mg/ml NaN₃) for 20 min on ice. Cells were then washed with PBS and subjected to flow cytometry using a FACScan flow cytometer. There was also a low number (2–4%) of necrotic cells, as determined by staining with propidium iodide, in all samples.

Microarray hybridizations

Microarray hybridizations were performed as described (Neiman et al., 2001).

References

1. Appl H & Klempnauer K-H. (2002) Oncogene 21: 3076–3081. | Article | PubMed | ISI | ChemPort |
2. Beall EL, Manak JR, Zhou S, Bell M, Lipsick JS & Botchan MR. (2002) Nature 420: 833–837. | Article | PubMed | ISI | ChemPort |
3. Buerstedde JM & Takeda S. (1991) Cell 67: 179–188. | Article | PubMed | ISI | ChemPort |
4. Cervellera M, Raschella G, Santilli G, Tanno B, Ventura A, Mancini C, Sevignani C, Calabretta B & Sala A. (2000) J. Biol. Chem. 275: 21055–21060. | Article | PubMed | ISI | ChemPort |
5. Cervellera MN & Sala A. (2000) J. Biol. Chem. 275: 10692–10696. | Article | PubMed | ISI | ChemPort |
6. Davidson C, Tirouvanziam R, Herzenberg L & Lipsick J. (2004) Genetics 169: 215–229. | Article | PubMed | ISI | ChemPort |
7. Fitzpatrick CA, Sharkov NV, Ramsay G & Katzen AL. (2002) Development 129: 4497–4507. | PubMed | ISI | ChemPort |
8. Foos G, Grimm S & Klempnauer K-H. (1992) EMBO J. 11: 4619–4629. | PubMed | ISI | ChemPort |
9. Foos G, Grimm S & Klempnauer K-H. (1994) Oncogene 9: 2481–2488. | PubMed | ISI | ChemPort |
10. Grassilli E, Salomoni P, Perrotti D, Franceschi C & Calabretta B. (1999) Cancer Res. 59: 2451–2456. | PubMed | ISI | ChemPort |
11. Huber A, Bai P, de Murcia JM & de Murcia G. (2004) DNA Repair 3: 1103–1108. | Article | PubMed | ISI | ChemPort |
12. Joaquin M, Bessa M, Saville MK & Watson RJ. (2002) Oncogene 21: 7923–7932. | Article | PubMed | ISI | ChemPort |
13. Joaquin M & Watson RJ. (2003) Cell Mol. Life Sci. 60: 2389–2401. | Article | PubMed | ISI | ChemPort |
14. Kamano H, Burk B, Noben-Trauth K & Klempnauer K-H. (1995) Oncogene 11: 2575–2582. | PubMed | ISI | ChemPort |
15. Katzen AL, Jackson J, Harmon BP, Fung SM, Ramsay G & Bishop JM. (1998) Genes Dev. 12: 831–843. | PubMed | ISI | ChemPort |
16. Klempnauer K-H & Bishop JM. (1983) J. Virol. 4: 565–572.
17. Lam EW-F & Watson RJ. (1993) EMBO J. 12: 2705–2713. | PubMed | ISI | ChemPort |
18. Lane S, Farlie P & Watson R. (1997) Oncogene 14: 2445–2453. | Article | PubMed | ISI | ChemPort |
19. Lang D, Dohle F, Terstesse M, Bangen P, August C, Pauels HG & Heidenreich S. (2002) J. Immunol. 168: 6152–6158. | PubMed | ISI | ChemPort |
20. Lang G, Gombert WM & Gould HJ. (2005) Immunology 114: 25–36. | Article | PubMed | ISI | ChemPort |
21. Manak JR, Mitiku N & Lipsick JS. (2002) Proc. Natl. Acad. Sci. USA 99: 7438–7443. | Article | PubMed | ChemPort |
22. Mucenski ML, McLain K, Kier AB, Swerdlow SH, Schreiner CM, Miller TA, Pietryga DW, Scott WJ, Jr & Potter SS. (1991) Cell 65: 677–689. | Article | PubMed | ISI | ChemPort |
23. Neiman PE, Ruddell A, Jasoni C, Loring G, Thomas SJ, Brandvold KA, Lee R, Burnside J & Delrow J. (2001) Proc. Natl. Acad. Sci. USA 98: 6378–6383. | Article | PubMed | ChemPort |
24. Nyberg KA, Michelson RJ, Putnam CW & Weinert TA. (2002) Annu. Rev. Genet. 36: 617–656. | Article | PubMed | ISI | ChemPort |
25. Sitzmann J, Noben-Trauth K, Kamano H & Klempnauer K-H. (1996) Oncogene 12: 1889–1894. | PubMed | ISI | ChemPort |
26. Tanaka Y, Patestos NP, Maekawa T & Ishii S. (1999) J. Biol. Chem. 274: 28067–28070. | Article | PubMed | ISI | ChemPort |
27. Toscani A, Mettus RV, Coupland R, Simpkins H, Litvin J, Orth J, Hatton KS & Reddy EP. (1997) Nature 386: 713–717. | Article | PubMed | ISI | ChemPort |
28. Zhu W, Giangrande PH & Nevins JR. (2004) EMBO J. 23: 4615–4626. | Article | PubMed | ISI | ChemPort |
29. Ziebold U, Bartsch O, Marais R, Ferrari S & Klempnauer K-H. (1997) Curr. Biol. 7: 253–260. | Article | PubMed | ISI | ChemPort |

Acknowledgements

We thank D Wenning for excellent technical assistance and J Delrow and the staff of the microarray facility of the Fred Hutchinson Cancer Research Center for performing the microarray experiments. This work was supported by the DFG (KL461/9-2).