c-myc overexpression activates alternative pathways for intracellular proteolysis in lymphoma cells.

Burkitt's lymphoma (BL) is a highly malignant B-cell tumour characterized by chromosomal translocations that constitutively activate the c-myc oncogene. Here we show that BL cells are resistant to apoptosis and do not accumulate ubiquitin conjugates in response to otherwise toxic doses of inhibitors of the proteasome. Deubiquitinating enzymes and the cytosolic subtilisin-like protease tripeptidylpeptidase II are upregulated in BLs, and could be rapidly induced by the overexpression of c-myc in normal B cells carrying oestrogen-driven recombinant Epstein-Barr virus. Apoptosis was induced by inhibiting tripeptidylpeptidase II, suggesting that the activity of this protease may be required for the survival of BL cells. We thus show that there is a regulatory link between c-myc activation and changes in proteolysis that may affect malignant transformation.

Burkitt lymphoma in the mouse.

Chromosomal translocations juxtaposing the MYC protooncogene with regulatory sequences of immunoglobulin (Ig) H chain or kappa (Ig kappa) or lambda (Ig lambda) L chain genes and effecting deregulated expression of MYC are the hallmarks of human Burkitt lymphoma (BL). Here we report that lymphomas with striking similarities to BL develop in mice bearing a mutated human MYC gene controlled by a reconstructed Ig lambda locus encompassing all the elements required for establishment of locus control in vitro. Diffusely infiltrating lymphomas with a typical starry sky appearance occurred in multiple founders and an established line, indicating independence from positional effects. Monoclonal IgM(+)CD5(-)CD23(-) tumors developed from an initially polyclonal population of B cells. These results demonstrate that the phenotype of B lineage lymphomas induced by MYC dysregulation is highly dependent on cooperativity among the regulatory elements that govern expression of the protooncogene and provide a new system for studying the pathogenesis of BL.

Identification of a novel p53-dependent activation pathway of STAT1 by antitumour genotoxic agents.

Chemotherapeutic drugs such as fludarabine*, doxorubicin or cisplatin are very potent activators of the anti-oncogene p53. Convergent studies suggest that p53 and STAT1 (signal transducer and activator of transcription 1) cooperate in the induction of cell death. We show that these drugs are also activators of STAT1 in p53-expressing cells, but not in p53-null cells. STAT1 activation was obtained in the presence of both the secretion inhibitor brefeldine A and the inhibitor of RNA synthesis, actinomycin D. p53-dependent STAT1 activation was reversed by overexpression of MDM2 and siRNAs against p53. Genetic analysis of p53 showed that expression of transcriptionally inactive p53 punctual mutants markedly increased Y701-STAT1 phosphorylation, and suggests that the p53 DNA-binding domain was alternatively involved in STAT1 activation or p53 multimerization. Immunoprecipitation experiments showed that ataxia telangiectasia mutated, p53, STAT1 and c-Abl1 (Abelson murine leukaemia viral oncogene homologue 1) were associated together. Treatment of cells with the c-Abl1 tyrosine kinase inhibitor STI571 decreased STAT1 activation by genotoxic drugs. Finally, genotoxic agents sensitized cells in response to very low doses of both interferon alpha and gamma (IFNalpha and gamma). These results show that genotoxic drugs induce STAT1 activation, an effect that depends on p53 protein but not on p53 transcriptional activity, and point to a novel pathway of STAT1 activation by genotoxic drugs, with involvement of c-Abl1 tyrosine kinase in sensitizing cells to IFN response.

Individual tumors of multifocal EB virus-induced malignant lymphomas in tamarins arise from different B-cell clones.

Cotton-top tamarins were inoculated with sufficient Epstein-Barr virus to induce multiple tumors in each animal within 14 to 21 days. The tumors consisted of large-cell lymphomas that contained multiple copies of the Epstein-Barr virus genome and generated Epstein-Barr virus-carrying cell lines showing no detectable consistent chromosomal abnormality. Hybridization of tumor DNA with immunoglobulin gene probes revealed that each lymphoma was oligo- or monoclonal in origin and that individual tumors from the same animal arose from different B-cell clones. Thus the virus induced multiple transformation events in tamarins in vivo to cause malignant tumors resembling the Epstein-Barr virus-associated lymphomas of patients with organ transplants.

The nuclear form of phospholipid hydroperoxide glutathione peroxidase is a protein thiol peroxidase contributing to sperm chromatin stability.

The selenoenzyme phospholipid hydroperoxide glutathione peroxidase (PHGPx) is regarded as the major molecular target of selenodeficiency in rodents, accounting for most of the histopathological and structural abnormalities of testicular tissue and male germ cells. PHGPx exists as a cytosolic form, mitochondrial form, and nuclear form (nPHGPx) predominantly expressed in late spermatids and spermatozoa. Here, we demonstrate that mice with a targeted deletion of the nPHGPx gene were, unlike mice with the full knockout (KO) of PHGPx, not only viable but also, surprisingly, fully fertile. While both morphological analysis of testis and epididymis and sperm parameter measurements did not show any apparent abnormality, toluidine blue and acridine orange stainings of spermatozoa indicated defective chromatin condensation in the KO sperm isolated from the caput epididymis. Furthermore, upon drying and hydrating, KO sperm exhibited a significant proportion of morphologically abnormal heads. Monobromobimane labeling and protein-free thiol titration revealed significantly less extensive oxidation in the cauda epididymis when compared to that in the wild type. We conclude that nPHGPx, by acting as a protein thiol peroxidase in vivo, contributes to the structural stability of sperm chromatin.

Control of cell growth by c-Myc in the absence of cell division.

The c-Myc protein (Myc) is a transcription factor, and deregulated expression of the c-myc gene (myc) is frequently found in tumours. In Burkitt's lymphoma (BL), myc is transcriptionally activated by chromosomal translocation. We have used a B-cell line called P493-6 that carries a conditional myc allele to elucidate the role of Myc in the proliferation of BL cells. Regulation of proliferation involves the coordination of cell growth (accumulation of cell mass) and cell division [1] [2] [3]. Here, we show that division of P493-6 cells was strictly dependent on the expression of the conditional myc allele and the presence of foetal calf serum (FCS). More importantly, cell growth was regulated by Myc without FCS: Myc alone induced an increase in cell size and positively regulated protein synthesis. An increase in protein synthesis is thought to be one of the causes of cell mass increase. Furthermore, Myc stimulated metabolic activities, as indicated by the acidification of culture medium and the activation of mitochondrial enzymes. Our results confirm the model that Myc is involved in the regulation of cell growth [4] and provide, for the first time, direct evidence that Myc induces cell growth, that is, an increase in cell size, uncoupled from cell division.

Target sequences for cis-acting regulation within the dual promoter of the human c-myc gene.

Recombinant plasmids of the human c-myc promoter-leader region and the bacterial chloramphenicol acetyltransferase (cat) gene were constructed. After transfection into different rodent and human cells, the 862-base-pair (bp) PvuII fragment carrying both c-myc promoters and 350 bp of the untranslated leader conferred 1/15 to 1/30 of the CAT activity mediated by the simian virus 40 promoter. The presence of additional sequences upstream of the PvuII fragment had an overall negative effect on c-myc promoter activity detectable by titration analysis with small amounts of transfected plasmid DNA. The analysis of numerous deletion constructs in the c-myc promoter-leader region as well as S1 mapping experiments demonstrated that the high CAT activity depended largely on the presence of the second promoter. By cotransfection of c-myc-cat constructs with plasmids carrying different parts of the c-myc promoter locus, targets for positively acting cellular factors were identified. Two positive regulatory elements were mapped within the 862-bp PvuII fragment. One was localized within the 248-bp PvuII-SmaI fragment -101 to -349 bp upstream of the first cap site and the other within the 142-pb XhoI-NaeI fragment of the first exon, comprising positions -95 to +47 relative to the second cap site. We conclude that the dual promotor of the human c-myc gene represents a strong eucaryotic promotor regulated by cooperation of positively and negatively acting cellular transcription factors.

Three breakpoints of variant t(2;8) translocations in Burkitt's lymphoma cells fall within a region 140 kilobases distal from c-myc.

The variant translocations t(2;8) in Burkitt's lymphoma cells join band q24 of chromosome 8, distal from c-myc, to the Igkappa locus, with considerable variation in the location of the breakpoints on chromosome 8. We report the cloning and molecular characterization of a chromosome 8 region, distal from the c-myc locus, which encompasses the breakpoints of the Burkitt's lymphoma cell lines BL64, BL21, and LY91 within 11 kilobase pairs, termed provisionally bvr-1 (Burkitt's variants' rearranging region 1). Using probes from the c-myc, the bvr-1, and the human pvt-1 loci obtained by chromosome walking coupled with pulsed-field gel electrophoresis, we have constructed a physical map of the region 3' of c-myc. We map bvr-1 and pvt-1 about 140 and 260 kilobase pairs, respectively, distal from c-myc.

RBP-L, a transcription factor related to RBP-Jkappa.

RBP-Jkappa is a sequence-specific DNA binding protein which plays a central role in signalling downstream of the Notch receptor by physically interacting with its intracellular region. Although at least four Notch genes exist in mammals, it is unknown whether each Notch requires a specific downstream signalling molecule. Here we report isolation and characterization of a mouse RBP-Jkappa-related gene named RBP-L that is expressed almost exclusively in lung, in contrast to the ubiquitous expression of RBP-Jkappa. For simplicity, we propose to call RBP-Jkappa RBP-J. The RBP-L protein bound to a DNA sequence almost identical to that of RBP-J. Surprisingly, RBP-L did not interact with any of the known four mouse Notch proteins. Although we found that RBP-L and EBNA-2 cooperated in transcriptional activation, they did not show significantly strong protein-protein interaction that can be detected by several in vivo and in vitro assays. This is again in contrast to physical association of RBP-J with EBNA-2. Several models to explain functional interaction between RBP-L and EBNA-2 are discussed.

The transcriptional program of a human B cell line in response to Myc.

The proto-oncogene c-myc (myc) encodes a transcription factor (Myc) that promotes growth, proliferation and apoptosis. Myc has been suggested to induce these effects by induction/repression of downstream genes. Here we report the identification of potential Myc target genes in a human B cell line that grows and proliferates depending on conditional myc expression. Oligonucleotide microarrays were applied to identify downstream genes of Myc at the level of cytoplasmic mRNA. In addition, we identified potential Myc target genes in nuclear run-on experiments by changes in their transcription rate. The identified genes belong to gene classes whose products are involved in amino acid/protein synthesis, lipid metabolism, protein turnover/folding, nucleotide/DNA synthesis, transport, nucleolus function/RNA binding, transcription and splicing, oxidative stress and signal transduction. The identified targets support our current view that myc acts as a master gene for growth control and increases transcription of a large variety of genes.

Attachment of antinuclear antibodies to nasopharyngeal carcinoma or other cells during preparation of biopsy imprints.

The presence of Epstein-Barr virus (EBV) genomes in nasopharyngeal and other carcinomas or Burkitt's and other B-cell lymphomas can be established by the demonstration of viral nucleic acid sequences in DNA extracts from biopsy specimens, the detection of EBV-associated nuclear antigen (EBNA) in biopsy imprints, and the inhibition of leukocyte migration by tumor extracts. Of these techniques, the detection of EBNA-positive tumor cells can be performed most readily in the laboratory. This report shows that a patient's antibodies to nuclear antigens can gain access to cell nuclei during the preparation of imprints. If the antibodies are directed against EBNA, nuclear immunofluorescence is elicited solely in the tumor cells when only complement (C') and fluorescein-labeled antibodies to C' are applied to the imprints without prior exposure to anti-EBNA-positive sera. If nonspecific antinuclear antibodies (ANA) are involved, the nuclear immunofluorescence seen in the EBNA-specific and control assays is not limited to the tumor cells but extends to any normal cells that may be present in the imprints. Furthermore, nuclear fluorescence is elicited when solely an anti-human IgG conjugate is applied because ANA is measurable by indirect immunofluorescence, whereas detection of EBNA requires augmentation of the antigen-antibody complexes by C', which differentiates further between EBNA-specific and nonspecific staining. Attachment of antibodies to nuclei can be avoided by minimizing the deposit of blood during imprint preparation and by rapid drying of the imprints. Similar results are obtained experimentally when smears of lymphoblasts are made in the presence of anti-EBNA or ANA.

Association of elevated mutagenesis in the spleen with genetic susceptibility to induced plasmacytoma development in mice.

Using the phage lambdaLIZ-based transgenic in vivo mutagenesis assay, mean mutant rates were determined in the spleen of mice exposed to sustained oxidative stress and were found to be increased approximately 3-fold in plasmacytoma-susceptible BALB/c and C.D2-Idh1-Pep3 mice, but not in plasmacytoma-resistant DBA/2N mice. This finding suggests a correlation between the genetic susceptibility to inflammation-induced peritoneal plasmacytomagenesis and the phenotype of increased mutagenesis in lymphoid tissues, raising the possibility that plasmacytoma resistance genes may inhibit tumor development by minimizing oxidative mutagenesis in B cells.

Elevated mutant frequencies in lymphoid tissues persist throughout plasmacytoma development in BALB/c.lambdaLIZ mice.

Using the phage lambdaLIZ-based transgenic in vivo mutagenesis assay, the mean mutant frequencies in the target gene, lacI, were found to be significantly increased in lymphoid tissues of congenic BALB/c.lambdaLIZ N5 mice in the terminal stage of a plasmacytoma induction experiment, 213-280 days after the first i.p. injection of the plasmacytomagenic agent pristane (2,6,10,14-tetramethylpentadecane). In plasmacytoma-bearing mice (n = 7), mutant frequencies in the spleens and mesenteric lymph nodes were elevated 2.46-fold and 5.35-fold, respectively, when compared with age-matched controls. In plasmacytoma-negative mice (n = 11), mutant frequencies were increased 2.30-fold (spleens) and 3.48-fold (mesenteric nodes). These results, interpreted in conjunction with our previous findings (K. Felix et al., Cancer Res., 58: 1616-1619, 1998) of approximately 3-fold elevations in pristane-induced splenic mutagenesis on day 42 postpristane, indicate that increased mutant levels in lymphoid tissues persist throughout plasmacytomagenesis in genetically susceptible BALB/c mice.

Mapping chromosomal breakpoints of Burkitt's t(8;14) translocations far upstream of c-myc.

To analyze the region upstream of c-myc, a number of novel probes were established. These were generated by chromosomal walking starting from the breakpoint of the chromosomal translocation of the B-cell line 380 and by cloning the breakpoint of the translocation of the Burkitt lymphoma cell line IARC/BL72. Using the newly isolated probes a detailed physical map of 500 kilobases of the region upstream of c-myc was established applying pulsed-field gel electrophoresis. The chromosomal breakpoint of IARC/BL72 cells was mapped to a site 55 kilobases 5' of c-myc. A region 20 kilobases in length and containing the breakpoints of 380, EW36, P3HR-1, and Daudi cells was identified 170-190 kilobases upstream of c-myc. In addition the HPV18 integration site in HeLa cells was located between 340 and 500 kilobases 5' of c-myc. The probes were used to define the c-myc amplification units in Colo320-HSR and HL60 cells as well as in four cases of small cell lung cancer. Evidence is provided that the amplicon of HL60 cells is discontinuously organized.

Two antisense promoters in the immunoglobulin mu-switch region drive expression of c-myc in the Burkitt's lymphoma cell line BL67.

In the Burkitt's lymphoma (BL) cell line BL67 the first exon of the c-myc gene is fused to the mu-switch region of the immunoglobulin heavy-chain gene (IgH). BL67 cells express IgH/c-myc hybrid RNAs which are initiated in the immunoglobulin locus, transcribed across the chromosomal breakpoint into the first exon of c-myc and spliced using the physiological splice donor and acceptor sites of the c-myc gene. We have isolated cDNAs of these hybrid RNAs and characterized the start points in the Ig heavy-chain gene. Two promoters were identified in the mu-switch region of BL67 cells which give rise to antisense transcription of the mu-gene. These promoters are also active in other BL cell lines, in B cells without Ig translocation and in a T-cell line. Both promoters co-localize with DNAase I-hypersensitive sites, HNF and HSW, in the mu-switch region. The structures of IgH/c-myc hybrid RNAs and of the corresponding promoters are described.

The block of elongation in c-myc exon 1 is abolished in Burkitt's lymphoma cell lines with variant translocation.

Dimethylsulfoxide (DMSO) induces a block of c-myc RNA elongation in the human B cell line BJAB. In Burkitt's lymphoma cell lines with variant translocations, which are characterized by mutations in and around the first c-myc exon, DMSO is not capable of inducing the RNA elongation block. The action of DMSO is, however, not restricted to regulation at the level of RNA elongation. In the cell line BL2 with a t(8;22) translocation c-myc steady-state RNA decreased about 20 fold 1 and 2 h after DMSO treatment, followed by an increase to approximately initial levels after 4 h. During the time course of the experiment the usage of the dual promoter P1/P2 shifted from the ratio 3:1 in untreated cells to the ratio of 1:5 in BL2 cells treated with DMSO for 4 h. This promoter shift is presumably regulated at the transcriptional level. In BJAB cells an isolated intragenic transcription was detected at the boundary of intron 1 and exon 2. This transcription appeared 2 to 4 h after addition of DMSO when expression of the c-myc gene was downregulated by blocking RNA elongation at the end of exon 1.

Transcription of c-myc in human mononuclear cells is regulated by an elongation block.

Human mononuclear cells (MNC) isolated from tonsils express very low amounts of c-myc RNA. High expression of the c-myc gene is achieved by stimulation of MNC with pokeweed mitogen (PWM). Nuclear run-on transcription experiments revealed activation of the c-myc gene on the transcriptional level. In unstimulated cells a high transcriptional activity is observed on exon 1, but not on exon 2 and 3. After PWM stimulation, however, a high transcription rate is detected on the entire c-myc gene. Therefore, we conclude that c-myc gene expression in MNC is regulated by an RNA elongation block within the first exon. The analysis of the c-myc chromatin structure revealed that DNAaseI hypersensitive site II upstream of the gene is present in unstimulated as well as stimulated MNC, indicating that accessibility of this site is an essential but not sufficient requirement for c-myc gene expression.

c-myc expression is activated by the immunoglobulin kappa-enhancers from a distance of at least 30 kb but not by elements located within 50 kb of the unaltered c-myc locus in vivo.

50 kb of contiguous DNA sequences covering the human c-myc coding region and approximately 20 kb of flanking upstream and downstream sequences were cloned onto a prokaryotic F-factor derived plasmid, which also contains a selectable marker and the plasmid origin of DNA replication oriP of Epstein Barr virus (EBV). Since these plasmids replicate extrachromosomally after stable transfection into EBV-positive B-cell lines, the gene regulation of c-myc can be analysed independent from chromosomal integration positions. Despite the presence of all known c-myc regulatory elements on these constructs, expression from the stably transfected c-myc gene was barely detectable in either cell line. Hypermethylation of these plasmids could be excluded as a mechanism for the lack of gene expression. Insertion of the immunoglobulin kappa-intron and 3' enhancers, however, activated c-myc transcription, when placed adjacent to or separated from the c-myc promoters by as far as 30 kb. These results indicate that transcription of c-myc in vivo requires additional and still unidentified control elements located outside this 50 kb fragment, and experimentally demonstrate long range enhancer function in vivo.

The intron enhancer of the immunoglobulin kappa gene activates c-myc but does not induce the Burkitt-specific promoter shift.

In Burkitt's lymphoma cells the c-myc gene locus is consistently fused to the constant region of one of the immunoglobulin genes by chromosomal translocation. The translocated c-myc gene is transcriptionally activated and preferentially transcribed from the P1 promoter whenever the exon-intron structure of c-myc remains intact. In order to define elements involved in this promoter shift we have cloned the translocated c-myc allele from Burkitt's lymphoma cell line BL60, which is characterized by several point mutations. The mutated c-myc allele of BL60 was stably introduced into baby hamster kidney and Burkitt's lymphoma cells. S1 nuclease and RNAase protection mapping experiments demonstrated that the mutated c-myc allele was expressed at a low level and with a normal promoter usage (P2 greater than P1) in Burkitt's lymphoma and baby hamster kidney cells. Furthermore, we have studied the expression of a construct consisting of the mutated c-myc allele, part of the bvr1 (Burkitt's variant rearranging region 1) locus, the human immunoglobulin kappa constant region, and the kappa intron enhancer after stable transfection into Burkitt's lymphoma cells. Although c-myc expression was about fivefold increased, the transcripts still initiated predominantly at promoter P2. This indicates that 5 kb of the constant kappa light-chain locus including the kappa intron enhancer is not sufficient to induce the Burkitt's lymphoma-specific promoter shift.

Expression of P0- and P3-RNA from the normal and translocated c-myc allele in Burkitt's lymphoma cells.

We have studied the allele specific expression of c-myc P0- and P3-RNA in Burkitt's lymphoma (BL) cells. The steady state levels of P0-RNA show considerable variations in BL cells. Expression of P0-RNA was found to be restricted to the translocated allele, but could be induced by TPA from the normal allele. P0-transcription was particularly sensitive to inhibitors of protein synthesis compared to expression of P1-, P2- and P3-RNA. Transcription of P3-RNA is initiated in the first intron of the c-myc gene and has previously been described to be specific for translocated c-myc alleles in BL cells broken within exon 1 or intron 1. Here we show that P3-RNA is also expressed from an unrearranged c-myc gene. In the BL cell line Raji, substantial amounts of c-myc RNA are derived from the P3-promoter of the normal allele. This indicates that repression of the normal allele in BL cells does not include the P3-promoter. The potential coding capacity of P3-RNA is discussed.