Insights into Gene Expression Changes Impacting B-

http://www.100md.com 病菌学杂志 2006年第4期

     Laboratory of Experimental Hematology, Bordet Institute, 1000 Brussels, Belgium

    Institute of Pathological Physiology, Charles University, Prague, Czech Republic

    Etablissement Franais du Sang, 13009 Marseille, France

    Vaccine and Infectious Disease Organization, S7H5E3 Saskatoon, Canada

    ABSTRACT

    Large-animal models for leukemia have the potential to aid in the understanding of networks that contribute to oncogenesis. Infection of cattle and sheep with bovine leukemia virus (BLV), a complex retrovirus related to human T-cell leukemia virus type 1 (HTLV-1), is associated with the development of B-cell leukemia. Whereas the natural disease in cattle is characterized by a low tumor incidence, experimental infection of sheep leads to overt leukemia in the majority of infected animals, providing a model for studying the pathogenesis associated with BLV and HTLV-1. TaxBLV, the major oncoprotein, initiates a cascade of events leading toward malignancy, although the basis of transformation is not fully understood. We have taken a cross-species ovine-to-human microarray approach to identify TaxBLV-responsive transcriptional changes in two sets of cultured ovine B cells following retroviral vector-mediated delivery of TaxBLV. Using cDNA-spotted microarrays comprising 10,336 human genes/expressed sequence tags, we identified a cohort of differentially expressed genes, including genes related to apoptosis, DNA transcription, and repair; proto-oncogenes; cell cycle regulators; transcription factors; small Rho GTPases/GTPase-binding proteins; and previously reported TaxHTLV-1-responsive genes. Interestingly, genes known to be associated with human neoplasia, especially B-cell malignancies, were extensively represented. Others were novel or unexpected. The results suggest that TaxBLV deregulates a broad network of interrelated pathways rather than a single B-lineage-specific regulatory process. Although cross-species approaches do not permit a comprehensive analysis of gene expression patterns, they can provide initial clues for the functional roles of genes that participate in B-cell transformation and pinpoint molecular targets not identified using other methods in animal models.

    INTRODUCTION

    DNA microarray technology has facilitated the identification of a large number of genes involved in the complex deregulation of cell homeostasis taking place in cancer (25, 59). There exists, however, a significant limitation in the variety of organisms for which microarrays have been developed because of a lack of genomic sequence data. Although species-specific small-scale application-targeted arrays are useful for monitoring specific networks of genes (9, 66), they do not address the broad spectrum of genes involved in the complex deregulations taking place in cancer. Larger-scale microarrays for domestic species are currently under investigation, but these emerging tools suffer from a lack of annotated genes. A solution to this limitation is to use microarrays designed for one species to analyze RNA samples from closely related species. The assumption is that the conservation of gene sequences between species will be sufficient to generate a reasonable amount of good-quality data. While there have been relatively few previously published reports that described the use of microarrays for cross-species hybridization (8, 24, 26, 31, 47), this technique is potentially a powerful tool for understanding molecular mechanisms associated with cancer in model organisms such as sheep.

    The sheep has been of particular interest as a large-animal model for studying aspects of immunology and offers a number of experimental opportunities that are not available in murine systems. Furthermore, sheep develop B-cell leukemia following experimental transmission of bovine leukemia virus (BLV), a complex retrovirus that is structurally and functionally related to human T-cell leukemia virus type 1 (HTLV-1) (6, 42, 78). This virus-associated leukemia model in sheep has been extensively studied as an in vivo approach to understand the molecular basis of human leukemogenic processes.

    Transactivating oncoretroviruses (BLV and HTLV-1) induce tumors after long latency by using poorly understood mechanisms that involve Tax (6, 12, 45, 75). Both TaxHTLV-1 and TaxBLV were originally identified as transcriptional transactivators of viral expression through the indirect binding to Tax-responsive elements located in the viral long terminal repeat (4, 10, 76). Both proteins mediate the transformation of rat embryonic fibroblasts in cooperation with Ha-ras, and injection of these cells induces tumors in nude mice (57, 77). In addition, TaxHTLV-1 immortalizes human T lymphocytes (17), while TaxBLV was recently shown to disrupt the homeostatic control of ovine B cells (65). Altogether, Tax proteins are thought to be the major viral contributors to the leukemogenic processes leading to either T-cell leukemia (HTLV-1) or B-cell malignancy (BLV). Because Tax expression is not required to maintain the transformed phenotype (69, 71), it is believed to act at early stages in the multistep process leading to full malignancy. TaxHTLV-1 has been extensively studied and was found to affect a variety of cellular functions in T cells essentially through its ability to transcriptionally regulate cellular gene expression and functionally inactivate proteins involved in cell cycle progression and DNA repair (14, 36). Using a medium-scale microarray, genes related to apoptosis, the cell cycle, and DNA repair; signaling factors; immune modulators; cytokines; growth factors; and adhesion molecules were identified as TaxHTLV-1 responsive (50).

    We have recently provided evidence to support the oncogenic potential of TaxBLV in ovine B cells, and we suggest an important role for NF-B-dependent pathways in TaxBLV-associated B-cell leukemia (19, 65, 70). So far, however, little is known about the interactions of TaxBLV with cellular proteins, and there is no evidence for TaxBLV-mediated transcriptional changes associated with abnormal B-cell growth. Although there is compelling evidence that TaxBLV is an essential contributor in the initial steps of oncogenesis, it is not sufficient for transformation, and cumulative changes are necessary for the development of overt leukemia. In this study, we examined gene expression changes in response to TaxBLV in an attempt to identify genes that play a role in the molecular events leading to leukemia. We have taken a cross-species ovine-to-human gene-profiling approach using human cDNA microarrays to identify TaxBLV-associated transcriptional changes in two sets of ovine B-cell populations. The rationale behind this study based on cross-species hybridization was not to generate an exhaustive list of differentially expressed genes but rather to provide initial clues for the functional roles of genes that participate in B-cell transformation.

    MATERIALS AND METHODS

    Cell cultures. Ovine peripheral blood mononuclear cells (PBMC) were isolated from blood using standard Ficoll-Hypaque separation and maintained in six-well plates (BD Labware) (5 x 106 cells/well) with -irradiated murine CD154 (mCD154) L cells (0.4 x 106 cells/well) (a kind gift from Troy Randall) in AIM-V medium supplemented with 10% fetal bovine serum, 1 mM sodium pyruvate, 2 mM glutamine, nonessential amino acids, kanamycin (100 μg/ml) (Invitrogen), recombinant human interleukin-2 (IL-2) (20 U/ml), IL-4 (40 U/ml), IL-7 (20 U/ml), and IL-15 (20 U/ml) (Peprotech) (19). Clone2 and Clone2LTAXSN B-cell clones were established from ovine jejunal Peyer's patch (18) and maintained in similar conditions with 2% fetal bovine serum and IL-4 (20 U/ml). All cultures were transferred every 3 to 4 days to fresh medium, cytokines, and mCD154 L cells. The supernatant from PG13-derived retrovirus producer cell lines was used to transduce the SBL (sheep B-lymphocyte) cells as previously described (69). All cultures were incubated at 37°C in a 5% CO2 humidified atmosphere. Viable cell numbers were measured by trypan blue exclusion. Apoptotic cell death was assessed by a terminal deoxynucleotidyltransferase-mediated dUTP-biotin nick end-labeling assay according to the manufacturer's instructions (Roche), with the following modifications: cells were fixed in 2% paraformaldehyde and stored at –20°C in 70% ethanol before labeling (65).

    RNA extraction and microarray probe preparation. RNA was extracted using TriPure (Roche) according to the manufacturer's instructions. Quality was assessed using the Agilent 2100 Bioanalyzer (Agilent Technology). Total RNA was linearly amplified according to a method described previously by Eberwine (11), with minor modifications. Briefly, RNA was reverse transcribed using a 63-nucleotide synthetic primer containing the T7 RNA polymerase binding site 5'-GGCCAGTGAATTGTAATACGACTCACTATAGGGAGGCGG(T)24-3'. Second-strand cDNA synthesis was obtained with RNase H, Escherichia coli DNA polymerase I, and E. coli DNA ligase (Invitrogen). cDNA was blunt ended with T4 DNA polymerase (Invitrogen). Double-stranded cDNA was transcribed using T7 polymerase (Ambion), yielding amplified antisense RNA which was purified using RNeasy Mini-Columns (QIAGEN). Total RNA from the Universal Human Reference (Stratagene) was amplified and used as a reference for microarray analysis. The cDNA microarray chips consisted of 10,336 total features and were manufactured at the Bordet Institute microarray facility using IMAGE clones. Three micrograms of RNA was reverse transcribed and labeled using cyanine 5-conjugated- or cyanine 3-conjugated dUTP. Hybridization was performed in 5x SSC (1x SSC is 0.15 M NaCl plus 0.015 M sodium citrate) and 25% formamide for 14 to 16 h at 42°C. Slides were washed, dried, and scanned using an Axon 4000a laser scanner. A detailed protocol for RNA amplification, cDNA probe labeling, and hybridization is available at http://nciarray.nci.nih.gov/reference/index.shtml. Genepix software (Axon Instruments) was used to analyze the raw data, which were then uploaded to a relational database maintained by the Bordet Institute.

    Data analysis. Images of scanned slides were inspected for artifacts, and aberrant spots and slide regions were flagged for exclusion from analysis. Log ratios for each spot were calculated as follows. In each channel, the signal was calculated as foreground median minus background median. If the signal was less than 300 in any single channel, the signal value in that channel was set to 300. If the signal was less than 300 in both channels, the spot was flagged as unreliable and not used in any further analysis. For all nonflagged spots, a log ratio was calculated as log2 [(red signal)/(green signal)]. The log ratios were then normalized within each array by subtracting from each array the median log ratio value across the spots on the array. The criteria for identifying genes with different levels of expression among the samples were based on the variance of their normalized log ratios across all experiments. The variance of the log ratio of each gene was compared with the median of all the variance, where the median is taken over all genes. A chi-squared statistic was computed in which the theoretical variance of the log ratios under true repetition was estimated by the median of the computed variances, and all genes with significance levels at a P value of >0.01 were filtered out.

    Northern blot analysis. Ten micrograms of total RNA was incubated for 10 min in a solution containing 6 M deionized glyoxal, 50% dimethyl sulfoxide, and 0.1 M phosphate buffer (pH 7.0) on ice followed by 3 min at 55°C. RNA was separated by electrophoresis through a 1% agarose gel containing 10 mM phosphate buffer, transferred onto nylon membranes (Amersham), and UV cross-linked. Hybridization was performed in a solution containing 50% deionized formamide and 5x SSC containing a 32P-labeled TAXBLV or glyceraldehyde-3-phosphate dehydrogenase (G3PDH) probe as previously described (69). Human cDNA probes (LENG5, FOSL2, CART1, methionine aminopeptidase 2 [METAP2], and butyrate response factor 2 [BRF2]) were obtained by T3/T7 PCR amplification of IMAGE clones (Invitrogen) selected according to GenBank accession number from the Invitrogen dbEST clone database. Signals were quantified by densitometric analysis using the Bio-Profil Bio-1D program, Windows Application V99.03.

    Quantitative RT-PCR. cDNA was synthesized by reverse transcription (RT) of total RNA from ovine B cells for 45 min at 42°C using Moloney murine leukemia virus reverse transcriptase (Promega). Real-time quantitative PCR was performed using SYBR green dye (SYBR green PCR master mix; Applied Biosystems) and either ovine (Mcl-1) or bovine (PTK2 and c-FOS) sequence-specific primers (Mcl-1 forward, ACGGCTTTCCAAGGCATG; Mcl-1 reverse, CCATCACTCGAGACAAAGACTTGA; PTK2 forward, CCAAATGGAGCCAGTGAACCT; PTK2 reverse, AAGCACGTGGCCTGCTATG; c-FOS forward, TCAATGACCCTGAGCCCAAG; c-FOS reverse, TCAGCCTTCAGCTCCATGCT). The amplification was performed in an ABI prism 7900 sequence detector system (Applied Biosystems) using 40 cycles of a two-step PCR (15 s at 95°C and 60 s at 60°C) after an initial activation step (95°C for 10 min). Melting curves from 60°C to 99°C were assessed to evaluate specificity. Serial dilutions of purified amplicons served to generate standard melting curves. Amplification of ovine -actin mRNA as an endogenous control was used to standardize the amount of sample added to the reactions (-actin forward, CCATCCTGCGTCTGGACCT; -actin reverse, CCACGTTCCGTGAGGATCTT).

    RESULTS

    Identification of cellular genes that undergo changes in response to TaxBLV: evaluation of an ovine-to-human microarray approach using a clonal B-cell population. TaxBLV plays a critical role in the early steps of the leukemogenic process, but so far, there is little information regarding cellular genes that may contribute to the deregulation of B-cell homeostasis. To identify TaxBLV-responsive genes, we used a cross-species microarray approach and well-characterized systems for the culture of primary ovine B cells. We have previously shown that B cells isolated from a variety of sheep lymphoid tissues including Peyer's patches and peripheral blood are permissive to BLV infection (19, 70). These culture systems, dependent on the costimulation with mCD154 and chain (c)-common cytokines, provide an excellent in vitro model for investigating cellular processes resulting from TaxBLV expression. It was deemed appropriate to first use a homogeneous B-cell population for designing optimal experimental conditions and evaluating the feasibility of the ovine-to-human microarray approach. We previously demonstrated the clear supportive role of TaxBLV on the growth of Clone2, a surface immunoglobulin M (IgM) (sIgM)-positive (sIgM+) jejunal Peyer's patch-derived B-cell clone dependent on mCD154 and IL-4 (18, 65). Clone2LTAXSN was derived following retroviral vector-mediated delivery of TaxBLV into Clone2. Both Clone2LTAXSN, which contains detectable TaxBLV protein, and the control parental Clone2 were further used for gene expression profiling.

    In the present study, we hybridized cDNA from ovine B-cell cultures to human cDNA-spotted high-density microarrays comprising 10,336 genes/expressed sequence tags (ESTs). Total RNA was extracted from Clone2LTAXSN and Clone2 cells at three time points posttransduction (days 125, 153, and 182) and treated as described in Materials and Methods. Amplified RNA was reverse transcribed and labeled with either cyanine 5-conjugated (Clone2LTAXSN) or cyanine 3-conjugated (Clone2) dUTP. Our experimental design thus included three replicate hybridizations with three independent RNA extractions referred to as hybridizations A, B, and C. The data were first processed as described in detail in Materials and Methods. Only the features that hybridized in at least two out of the three microarray experiments were processed for further analysis.

    Ovine cDNA from the B-cell clones hybridized to 6,484 (62.54%) of the human gene targets on a microarray containing 10,336 targets (Table 1). We considered genes that were up- or down-regulated by more than 1.5-fold to be differentially expressed. This cutoff was defined relatively arbitrarily, but as described below, it was independently validated using alternative techniques for measuring relative gene expression. Finally, across all experiments, we identified 263 genes that fulfilled the criterion of differential expression. A total of 211 genes (2.05%) were significantly up-regulated (median, >1.5), and 52 genes (0.50%) were down-regulated (median, <0.66) in Clone2LTAXSN compared to Clone2 cells, respectively. In the gene list presented here, we restricted the data set to a selection of 103 up-regulated (Table 2) and 23 down-regulated (Table 3) genes, according to existing ideas about how altered gene expression may have a functional contribution to the complex processes involved in leukemogenesis. However, to keep the benefits of an unbiased approach, it is important to provide access to the complete lists of altered genes. As our knowledge progresses, we or others might be able to make more sense of the comprehensive expression patterns, ultimately aiding in the understanding of networks that contribute to cancer. The complete microarray data set is available at http://www.bordet.be/servmed/martiat/avdb/data.htm. In Tables 2 and 3, genes are arranged according to the arithmetic means of the relative expression ratios of three independent experiments, experiments A, B, and C. We identified expression changes in genes from distinct groups, including proto-oncogenes; genes regulating apoptosis, DNA transcription, and repair; transcription factors; cell cycle regulators; kinases; phosphatases; and small Rho GTPases/GTPase-binding proteins. It is interesting that genes associated with various human malignancies, especially B-cell tumors, were largely represented among the differentially expressed genes. Other genes were novel or unexpected or had no established role in tumor development. We refer the reader to Discussion (see below) for further details regarding a selection of genes from these lists.

    Important criteria for evaluating any microarray system include the reproducibility of the data, the specificity of detection, and the validity of the results that identify differences in gene expression. Reproducibility was evident when the distributions of data from one replicate to another were compared, as indicated by scatter plots of data generated from replicate hybridizations with independent ovine RNA extractions (data not shown). Furthermore, cDNA from ovine B-cell clones consistently hybridized to human targets known to be expressed in B cells (Ig or Ig-related genes CD79, CD83, REA, BTG1, and early B-cell factor [EBF]) as well as to previously well-documented TaxHTLV-1-responsive genes (activating transcription factor 4 [ATF4], FOS, FOSL2, tumor necrosis factor superfamily member 4 [TNFSF4], and BRF2), thereby demonstrating the specificity of the system. To obtain independent confirmation of the microarray results, both Northern blot analysis and quantitative RT-PCR (qRT-PCR) were performed. Transcription levels of five putative differentially expressed genes (BRF2, LENG5, METAP2, FOSL2, and CART1) with relative expression ratios of >1.5 were examined by Northern blot with human cDNA probes. Analysis of Clone2LTAXSN and Clone2 RNA used for microarray hybridizations A and B confirmed the predicted gene expression patterns (Fig. 1, lanes A and B). Sequence-specific primers were designed for c-FOS, Mcl-1, and PTK2, three overexpressed genes for which the ovine or bovine sequences are known. Using qRT-PCR with these species-specific primers, overexpression was confirmed for all three of the genes examined and ranged between a mean of 1.59- and 1.90-fold (Table 4). In conclusion, there was a good agreement between the microarray and both the Northern blot and qRT-PCR data, thereby validating the microarray results.

    Identification of TaxBLV-responsive genes in peripheral blood-derived ovine B cells. Gene expression profiling of Clone2LTAXSN B cells using ovine-to-human cross-species microarrays revealed genes that were regulated as a result of TaxBLV expression. However, although the parental Clone2 B cells were initially derived using mCD154 and IL-2, IL-4, IL-7, and IL-15, these cells developed a relative cytokine independence over time (IL-2, IL-7, and IL-15) and increased resistance to cell death with a requirement for only IL-4. This probably reflects the selection of one specific B cell during the cloning process and suggests that physiological changes have occurred in these cells. Thus, although microarray analysis of this B-cell clone was of interest for validating the cross-species hybridization approach, the analysis of Clone2 runs the risk of identifying differences in expression that are not connected to the gene of interest but occur independently of oncogenic processes. To reduce the number of changes that could account for clonal selection, we examined ovine B cells isolated from blood. These B cells were derived from PBMC cultured in the presence of mCD154 and c-common cytokines IL-2, IL-4, IL-7, and IL-15 as described in detail in Materials and Methods. After a 2-week costimulation period, the cultures consisted exclusively of sIgM+ B cells, as indicated by flow cytometry phenotypic analysis (data not shown), and were referred to as SBL (sheep B lymphocytes). The Clone2 and polyclonal SBL populations probably share many biological similarities in terms of B-cell development stage. The continued expression of sIgM in both populations suggests that their development may be arrested prior to the memory B-cell stage. The SBL, however, represent a more heterogenous sIgM+ B-cell population that has retained its dependence on both mCD154 (T-cell-dependent development) and the c-common cytokines for continued growth. Thus, the major difference is that SBL cells require a broader range of cytokines, which may reflect the diversity of this B-cell population.

    For the transfer of tax cDNA into SBL cells, we designed a vector derived from pSF (3), referred to as pSFTaxCMVGFP. In this vector, the BLV tax cDNA is under the control of a hybrid spleen focus-forming virus/murine embryonic stem cell virus promoter demonstrated to be active in a wide range of hematopoietic cells at different stages of differentiation. The enhanced version of the green fluorescent protein (GFP) is under the control of the cytomegalovirus promoter. The pSFiECMVGFP control vector lacks tax. In contrast to the pLTaxSN/pNUNL vectors used for the transduction of Clone2 (65), this type of vector offers the possibility of selecting GFP-positive (GFP+) cells by flow cytometry and monitoring GFP+ cells over time in mixed cultures consisting of both transduced and untransduced cells. The vectors were first evaluated in Clone2 cells, given that TaxBLV expression is known to enhance their viability. Transfer of pSFTaxCMVGFP, but not pSFiECMVGFP, confers a growth advantage to Clone2 cells, confirming that TaxBLV and not the vector itself is involved in the altered phenotype (Fig. 2). The BLV tax cDNA was then delivered into SBL cells using supernatant-mediated retroviral vector transduction, and these cultures were further maintained for 7 days in the presence of mCD154 and IL-2, IL-4, IL-7, and IL-15. The pSFTaxCMVGFP- and pSFiECMVGFP-transduced B cells, referred to as SBLTCE and SBLiE, respectively, were then examined for GFP expression by flow cytometry. The GFP+ cells, representing approximately 5% of the total cell population, were sorted (Fig. 3a), and TaxBLV expression was evaluated by Northern analysis of RNA extracted from the purified B-cell populations (Fig. 3b). We then examined the growth characteristics of these cells and found that the levels of apoptotic cell death were decreased in SBLTCE compared to SBLiE under normal costimulated culture conditions (Fig. 4a) as well as in the absence of c-common cytokines both with and without mCD154, an essential factor required to support the continuous growth of B cells (Fig. 4b). In addition, we found that SBLTCE cells displayed increased viable cell numbers (Fig. 4c), providing clear evidence that TaxBLV alters the growth of peripheral blood-derived B cells.

    Total RNA was extracted from both the SBLTCE and SBLiE cells at three time points posttransduction (days 128, 157, and 185) and used in ovine-to-human microarrays. The experimental design was similar to that utilized for Clone2 cells. The three replicate hybridizations with three independent RNA extractions are referred hereafter as K, L, and M. Raw data and preliminary results were processed as described above in detail for Clone2LTAXSN/Clone2. From a total of 10,336 spotted genes/ESTs, 6,360 (61.34%) hybridized in at least two out of the three microarray experiments (Table 1). From these, 292 genes (2.82%) were significantly up-regulated (relative expression ratio median, >1.5) and 149 genes (1.44%) were down-regulated (relative expression ratio median, <0.66) in SBLTCE cells compared to SBLiE cells, respectively. In the gene lists shown here, we restricted the displayed data set to a cohort of 107 and 38 genes that were significantly up- or down-regulated, respectively, in the SBLTCE cells compared to the SBLiE cells (Tables 5 and 6). These genes were selected similarly to Clone2LTAXSN/Clone2-associated genes, according to the current knowledge regarding the potential role they might play in the disruption of B-cell homeostasis. The complete microarray data set is also available at http://www.bordet.be/servmed/martiat/avdb/data.htm. Similarly to our observations with Clone2, we found differentially expressed genes in distinct functional categories, suggesting that many aspects of cell physiology are altered in response to TaxBLV. Details related to these genes are provided in Discussion (see below). The ovine-to-human microarray approach was validated in PBMC-derived B cells using the criteria described above for Clone2. We found that data from replicate hybridizations were equally distributed and thus highly reproducible. In addition, targets expected to be expressed in B cells were consistently detected, while known TaxHTLV-1-responsive genes were found to be differentially expressed, suggesting the specificity of detection of the method. Finally, overexpression was independently confirmed both by Northern blot analysis using human cDNA probes for BRF2, LENG5, METAP2, FOSL2, and CART1 (Fig. 1, lane K) and by qRT-PCR with ovine or bovine sequence-specific primers for c-FOS, Mcl-1, and PTK2 (Table 4). The qRT-PCR severalfold changes were consistently higher than the microarray ratios, but given that species-specific primers were used, they should be a better reflection of the relative gene expression levels. Altogether, these results validate the gene expression data and provide a rationale for the microarray cutoffs. Our observations suggest that gene expression profiling of PBMC-derived ovine B cells may provide insights into the mechanisms by which complex retroviruses induce malignancy.

    Intersect of TaxBLV-responsive differentially expressed genes between Peyer's patch B-cell clones and PBMC-derived B cells. We compared the data sets derived from Clone2LTAXSN/Clone2 cells with those from SBLTCE/SBLiE cells to identify genes differentially expressed in both systems. We detected 80 genes with significant altered expression in both Clone2LTAXSN and SBLTCE cells, emphasizing their putative role in TaxBLV-associated deregulation. Sixty-five genes were commonly up-regulated, while only 15 genes were scored as down-regulated. Tables 7 and 8 display a list of 59 up-regulated and 11 down-regulated genes, respectively. Thus, overlapping yet distinct gene expression patterns are associated with TaxBLV-mediated phenotypic changes in Peyer's patch- and PBMC-derived ovine B cells (Fig. 5). We then compared the distribution of data from one cell type to the other and found that TaxBLV induced similar changes of overall expression profiles in both ovine B-cell populations (Fig. 6). In sharp contrast, we identified only one single gene (HAT1 [histone acetyltransferase 1]) that was significantly up-regulated in one B-cell population (SBL) while it was down-regulated in the other (Clone2). Genes found to be differentially expressed in both B-cell populations likely play significant roles in signaling networks in which Tax participates. In this group (Tables 7 and 8), we once more identified genes from various functional categories expected to play a role in leukemogenesis, genes previously found to have altered expression in well-documented human cancers, as well as the TaxHTLV-1-responsive genes ATF4, FOS, and BRF2.

    DISCUSSION

    Large-animal models for cancer have the potential to contribute to the understanding of molecular pathways involved in oncogenesis. BLV-associated B-cell leukemia in sheep provides a unique model for investigating the molecular basis of human leukemogenic processes including HTLV-1-associated T-cell malignancies. TaxBLV likely has a multitude of functions. The ability of this viral transactivator to both transcriptionally regulate cellular gene expression and directly interact with cellular proteins provides the basis for leukemogenesis. Chromosomal aberrations, genome instability, and the control of protein activity by TaxBLV definitely play a major role in the development and progression of these multistep tumors. However, although the regulation of host cell mRNA levels is only one aspect of the biological control by TaxBLV, the mechanisms by which TaxBLV modulates host cell transcription are of considerable interest. In this study, we examined TaxBLV-mediated gene expression changes in ovine B cells, the targets of BLV, in an attempt to identify genes that play a role in the cascade of events leading to B-cell leukemia. We have taken a cross-species ovine-to-human gene-profiling approach using human cDNA microarrays to identify transcriptional changes in ovine B cells that express TaxBLV. We first examined hybridization characteristics and data variability from sheep-to-human microarrays using Clone2LTAXSN, a TaxBLV-positive clonal B-cell population characterized by abnormal growth (65). Using human cDNA microarrays (10,336 genes/ESTs), more than 62% of human probes were effective at ovine transcript detection. Approximately 2% (211 genes) were significantly up-regulated, and 0.5% (52 genes) were down-regulated. Besides genes with high levels of differential expression, those showing low differences were of particular interest, given that small changes in gene expression are expected to be sufficient to disrupt cell homeostasis. Moreover, there is no universal standard to accurately measure genes expressed at a low level or genes with low severalfold changes. In our study, we considered expression ratios of >1.5 and <0.66 to be significant. This arbitrary cutoff was independently validated using two alternative methods for measuring gene expression. Low ratios, however, still make it challenging to separate causal from coincidental changes in gene expression, especially in cross-species techniques, and this requires a thorough validation of data for each individual gene under investigation.

    Using these microarrays, we evaluated gene expression profiles of previously described B-cell clones (Clone2LTAXSN/Clone2) as well as PBMC-derived SBL cells. Both Clone2 and SBL are characterized by the surface expression of IgM, a consistent finding in leukemic B cells isolated from BLV-infected sheep. Those B-cell populations probably represent a stage in T-cell-dependent B-cell development that may occur within germinal centers, but the continued expression of sIgM would indicate that they are not centroblasts and are probably not memory B cells, since there is no evidence for isotype switching. The Clone2 and polyclonal SBL populations thus share many biological similarities in terms of B-cell development stage, but Clone2 developed a relative cytokine independence over time, probably reflecting the selection of one specific B cell during the cloning process. However, these B cells remain dependent upon both mCD154 stimulation and the presence of IL-4 to proliferate and remain viable, since removal of these stimuli results in apoptosis. Thus, these are not transformed B cells. The SBL represent a more heterogenous sIgM-positive B-cell population that has retained its dependence on both CD154 (T-cell-dependent development) and the spectrum of c-common cytokines (IL-2, IL-4, IL-7, and IL-15) for continued growth. Thus, the major difference is that SBL cells require a broader range of cytokines, which may reflect the diversity of this B-cell population and may be associated with a decreased risk of identifying changes attributable to an individual cell type. Data are presented from SBL cells isolated from the blood of only one sheep because we observed very little variation between microarray data sets for separate blood B-cell cultures from different sheep (data not shown). This suggests that animal-to-animal variation has little impact on baseline gene expression patterns in B-cell populations generated by costimulation with CD154 and cytokines. These stimuli appear to impose a strong selection on the surviving B cells. Ultimately, to avoid the limitations of the present system for exploring genes involved in oncogenesis, it will be important to analyze Tax-induced gene expression changes in B cells directly ex vivo, but this will require vectors that are capable of delivering a transgene into resting cells. The biological relevance of these two B-cell populations for studying human B-cell malignancies is that they both represent T-cell-activated B-cell populations and may be most comparable to human follicular B cells. Thus, these sheep B-cell lines may be most relevant for understanding human follicular lymphomas.

    We have previously shown that TaxBLV has a clear supportive role in the growth of these B cells, with an impact on B-cell proliferation, survival, and cell cycle phase distribution (65). In both B-cell populations, we identified a significant number of genes that underwent changes in response to TaxBLV (263 genes in Clone2LTAXSN and 441 genes in SBLTCE), with an intersect of approximately 25%. Although the samples for comparison were as closely matched as possible, analysis of gene expression data generated long gene lists, suggesting that TaxBLV expression in cultured B cells leads to direct or indirect changes in mRNA levels of hundreds of genes. Faced with this mass of data, the temptation was to search for genes that conform to existing ideas about how B cells acquire abnormal phenotypes. A selection of those genes is displayed in Tables 2, 3, and 5 to 8, but only a limited number are discussed here. The benefits of an unbiased approach, however, is lost if exploration is limited to our current framework of understanding. Therefore, in this study, although we do not provide clues as to which of these changes may be important, we provide the complete data set of differentially expressed genes. It is thus important to underline that we do not exclude a critical functional role for some of the genes that are not displayed in the shortened lists.

    The first striking observation was the strong agreement between our results and those for genes previously reported as being TaxHTLV-1 responsive or TaxHTLV-1 interacting in HTLV-1-infected T-cell cultures and human T-cell lines that express TaxHTLV-1 (ATF4, FOS, FOSL2, TNFSF4, and BRF2), confirming the specificity of our approach. We identified two members of the bZip transcription factors (FOS and FOSL2), ATF4, that forms dimers with bZip proteins including Fos (22), and TNFSF4 or TaxHTLV1 transcriptionally activated glycoprotein 1. Furthermore, given that TaxHTLV1 is capable of modulating the expression of BRF1 through a cis-acting replication element (38), it is tempting to speculate that BRF2, scored as up-regulated in ovine B cells, may have a similar function.

    There was a good agreement between the microarray results and both the Northern blot and qRT-PCR experiments, two methods designed to validate our data. Human cDNAs were used for probing the Northern blots, while ovine and bovine sequence-specific primers were utilized in qRT-PCR assays. In many cases, the exact severalfold changes were higher than the microarray ratios. This is commonly observed, and it is now widely accepted that those methods are a better reflection of the gene expression levels in each sample. Northern blot and qRT-PCR data thus provide a rationale for the 1.5-fold cutoff used in the microarray analysis.

    Interestingly, we identified several genes differentially expressed in both the B-cell clone and the PBMC-derived B cells, emphasizing their contribution to the TaxBLV-associated abnormal B-cell growth. The observation, however, that a significant proportion of altered genes do not overlap between Clone2LTAXSN and SBLTCE may appear surprising. One possible explanation is that differences in expression patterns found in B-cell clones are partly linked to changes that occur downstream from the early TaxBLV-associated events. Furthermore, given the greater diversity of SBL cells, the finding that genes are differentially expressed in SBL cells while they are not in B-cell clones does not exclude a functional role of these genes in leukemogenic processes. We thus considered that the genes differentially expressed in both B-cell populations as well as those identified in a single B-cell population were of interest.

    In the vast majority of genes, increases in expression levels did not exceed two- to threefold, with the exception of CART1 (ninefold), fibrinogen-like protein 2 (FGL-2) (fivefold), and fibrillin 1 (FBN-1) (fivefold). CART1 encodes a paired-like DNA-binding homeoprotein with transactivating transcriptional activity, but its target gene remains unknown (13). p300/CREB binding protein (CBP) acts as a coactivator to CART1 through direct interaction and lysine acetylation, enhancing its transactivation capacity (29). Given that genes encoding proteins with HAT activity, including p300/CBP, were up-regulated in our study (discussed below), it is tempting to speculate that CART1 overexpression is a secondary event and not a direct consequence of TaxBLV expression. Both FGL-2, a prothrombinase, and FBN-1, involved in the structure of connective tissue, were found to be overexpressed by fivefold. To our knowledge, aberrant expression of neither of these genes has been correlated with pathways known to be associated with deregulated cell growth. Keeping in mind that the interpretation of data is complicated by changes that may occur either downstream from or independently of the stimulus under consideration, our results do not allow primary and secondary changes in gene expression to be distinguished, and further investigation is necessary to draw conclusions from these data.

    In examining the expression profiles, several alterations were found in genes involved in DNA transcription/repair and genes encoding or interacting with proteins possessing HAT activity (p300, p400, GPS2, EBF, CART1, and ATF4). Gps2 modulates bovine papillomavirus type 1 E2 expression through recruiting p300 and directly interacts with TaxHTLV1 (54), known to target p300/CBP, leading to transcriptional deregulation of several cellular genes (16). EBF activates transcription factors through p300/CBP (83), and PA2G4, a transcriptional repressor, acts via recruiting histone deacetylases (82). CART1, increased by ninefold and discussed above, recruits p300/CBP to a CART-specific subnuclear compartment, thereby modulating transcription from different promoters. Furthermore, we identified transcriptional alterations of members of the chromodomain helicase DNA-binding protein (CHD) and SWI/SNF chromatin-remodeling complex gene families (CHD2, CHD4, CHD6, SWI/SWF-related regulator of chromatin 2 [SMARCA2], and SMARCA5), genes encoding DNA-binding proteins (TCF4, SREBF2, TRIP-BR2, PC4, and BHLHB2), and genes involved in DNA damage/repair (DDB2, TOP1, APEX1, and RAD51). Suppression of DNA repair likely increases the incidence of genomic mutations and may contribute to B-cell transformation. A striking example supporting the impact of DNA damage/repair imbalance in TaxBLV-mediated leukemogenesis is the up-regulation of topoisomerase 1 (TOP1) in both B-cell populations. TOP1 is involved in multiple DNA transactions, including DNA relaxation (56), and is the fusion partner of the nucleoporin gene NUP98 in acute myeloid leukemia-associated rearrangements. NUP98-TOP1 blocks differentiation of hematopoietic precursors and induces a lethal leukemia related to the DNA-binding properties and the capacity to induce DNA lesions mediated by excess TOP1. TOP1 is the target of novel anticancer drugs, the camptothecins (21).

    Among the cohort of genes with altered expression, we identified genes associated with NF-B activation and p53 suppression, including MDM2 (20), SSRP1 (81), Trip-Br2 (28), and many others. Interestingly, PPM1D overexpression abrogates the p53 suppressor activity through the inactivation of p38-mitogen-activated protein kinase (MAPK), a product of MAPK14, down-regulated in our study (5). Translocated in liposarcoma (FUS/TLS) interacts with NF-B-p65, and its fusion to different transcription factors is associated with various malignancies, including acute myeloid leukemia (44, 67). Our observations thus support previous findings that suggested that NF-B activation and p53 inhibition contribute to the oncogenic effects of TaxHTLV1/TaxBLV (14, 65, 80). We also observed altered expression of numerous small Rho GTPases/GTPase-binding proteins. These small G proteins contribute to various regulatory processes and play a critical role in mechanisms associated with cancer, including cell survival, invasion, and angiogenesis (23, 39, 60). Members of this family were shown to interact with TaxHTLV1 (79). We identified Ras homolog gene family members Q, B, and H (ARHQ, ARHB, and ARHH), members 1 and 4 of the ARF family of GTP-binding proteins (ARL1 and ARL4), members of Ras oncogene family (RAB6A and RAB5B), GTP-binding protein cell division cycle 42 (CDC42), and other GTPase-binding proteins (ARL6IP, SEC61B, RACGAP1, BCAR3, RAB5EP, DAAM2, C8FW, RASAL1, RGR, DOCK2, and CHN1). Our findings thus suggest that TaxBLV may exert its largely pleiotropic effects partly through the deregulation of a broad spectrum of small Rho GTPase/cytoskeleton-related genes. Furthermore, a large number of well-known oncogenes and diverse genes known to be associated with human malignancies were deregulated, including EMP3, PDE3B, LRMP, LASP1, L7a, PTK2, MEIS1, BCAS2, LMO4, OS4, and VEGF. Although the majority were found in hematological disorders, especially B-cell malignancies (PDE3B [35, 37, 48], LRMP [1], LASP1 [7, 63], MEIS1 [15, 43, 55], and VEGF [61, 74]), others were identified in hepatocarcinoma, breast cancer, and colorectal and prostate cancers (EMP3 [41], BCAS2 [40, 72], L7a [34, 68, 73, 85], OS4 [64], and PTK2 [51]).

    Interestingly, transcription patterns of antiapoptotic genes were changed. MCL1 is overexpressed in various B-cell malignancies and is found to be associated with drug resistance in B-lineage acute lymphoblastic leukemia, leading to poor response to treatment (27, 33, 53, 58). MCL1-transgenic mice develop chronic clonal B-cell lymphoma closely matching BLV-induced leukemia (84). High levels of METAP2 were detected in B-cell lymphoma subtypes derived from germinal center B cells, and METAP2 expression correlated with that of BCL6, a contributor to B-cell lymphomagenesis (2, 32, 46) up-regulated in response to TaxBLV. Furthermore, METAP2 represents a validated target for inhibition of neovascularization, suggesting that it may play a complex role in tumor progression. Finally, decreases as well as increases in cellular gene expression are important regulatory events. We have identified 52 and 149 down-regulated genes in Clone2LTAXSN and SBLTCE cells, respectively (Table 1), but only 15 genes were commonly repressed. Interestingly, among these genes, we identified the proto-oncogene ARHH, the inactivation of which promotes transformation in germinal center-derived diffuse large B-cell lymphomas (52). Also striking was the down-regulation of STMN1 in Clone2LTAXSN and DOC1 in SBLTCE. STMN1, a gene regulating cell division, is decreased in hepatoblastomas and neuroblastomas (30, 49), while DOC1 is critical for allowing cells to bypass senescence (62).

    Altogether, our results identify for the first time a cohort of TaxBLV-associated transcriptionally regulated genes. The good agreement between our findings and previous observations in both human B-cell tumors and HTLV-1-associated T-cell malignancies validates cross-species chip approaches for further elucidation of molecular pathways associated with oncogenesis. The genes identified as either up- or down-regulated likely play important roles in the signaling network in which TaxBLV participates. Analysis of these genes, in terms of known function, allows several important conclusions. There is no single regulatory process that is solely targeted by TaxBLV. Rather, a network of interrelated pathways is deregulated, leading to the disruption of growth control and favoring B-cell survival. Interestingly, given that we identified changes previously associated with HTLV-1, a T-cell-tropic oncoretrovirus, as well as genes linked to non-virus-associated human B-cell malignancies, our results suggest the implication of broad mechanisms that are independent of both lineage and etiology. The power of microarray analysis is that it reveals the complexity of biological events. Cell transformation is a process that affects many different aspects of cell physiology. The complexity of cell signaling may make it difficult to identify the critical or limiting events that occur during transformation. TaxBLV-responsive genes might not be unique contributors to the malignant phenotype, and additional events are required for progression to full-blown leukemia. Analysis directed towards gene expression profiling of B-cell tumors isolated from leukemic animals will permit the elucidation of molecular pathways that promote leukemia progression. Although cross-species approaches do not permit a comprehensive analysis of gene expression patterns, we provide insights into cellular events impacting B-cell transformation and pinpoint molecular targets not identified using other methods in animal models.

    ACKNOWLEDGMENTS

    This work was supported by the Medic Foundation, the Fonds National de la Recherche Scientifique, TELEVIE, the International Brachet Foundation, and the Fondation Bekales. P.K. has fellowships from the Lady Tata Memorial Trust and NATO, and M.S. and M.M. are supported by TELEVIE. A.V. is "Collaborateur Scientifique F.N.R.S.-Televie."

    We thank L. Vanderweerden for website development and Troy Randall for the generous gift of mCD154 L cells.

    P.K. and M.S. contributed equally to this work.

    REFERENCES

    Akasaka, T., I. S. Lossos, and R. Levy. 2003. BCL6 gene translocation in follicular lymphoma: a harbinger of eventual transformation to diffuse aggressive lymphoma. Blood 102:1443-1448.

    Albagli-Curiel, O. 2003. Ambivalent role of BCL6 in cell survival and transformation. Oncogene 22:507-516.

    Baum, C., S. Hegewisch-Becker, H. G. Eckert, C. Stocking, and W. Ostertag. 1995. Novel retroviral vectors for efficient expression of the multidrug resistance (mdr-1) gene in early hematopoietic cells. J. Virol. 69:7541-7547.

    Brady, J., K. T. Jeang, J. Duvall, and G. Khoury. 1987. Identification of p40x-responsive regulatory sequences within the human T-cell leukemia virus type I long terminal repeat. J. Virol. 61:2175-2181.

    Bulavin, D. V., O. N. Demidov, S. Saito, P. Kauraniemi, C. Phillips, S. A. Amundson, C. Ambrosino, G. Sauter, A. R. Nebreda, C. W. Anderson, A. Kallioniemi, A. J. Fornace, Jr., and E. Appella. 2002. Amplification of PPM1D in human tumors abrogates p53 tumor-suppressor activity. Nat. Genet. 31:210-215.

    Burny, A., L. Willems, I. Callebaut, E. Adam, I. Cludts, I. F. Dequiedt, L. Droogmans, C. Grimonpont, P. Kerkhofs, M. Mammerickx, D. Portetelle, A. Van den Broeke, and R. Kettman. 1994. Bovine leukemia virus: biology and mode of transformation, p. 313-334. In A. C. Minson, J. C. Neil, and M. A. McRae (ed.), Viruses and cancer. Cambridge University Press, Cambridge, United Kingdom.

    Butt, E., S. Gambaryan, N. Gottfert, A. Galler, K. Marcus, and H. E. Meyer. 2003. Actin binding of human LIM and SH3 protein is regulated by cGMP- and cAMP-dependent protein kinase phosphorylation on serine 146. J. Biol. Chem. 278:15601-15607.

    Chismar, J. D., T. Mondala, H. S. Fox, E. Roberts, D. Langford, E. Masliah, D. R. Salomon, and S. R. Head. 2002. Analysis of result variability from high-density oligonucleotide arrays comparing same-species and cross-species hybridizations. BioTechniques 33:516-518, 520, 522.

    Coussens, P. M., and W. Nobis. 2002. Bioinformatics and high throughput approach to create genomic resources for the study of bovine immunobiology. Vet. Immunol. Immunopathol. 86:229-244.

    Derse, D. 1987. Bovine leukemia virus transcription is controlled by a virus-encoded trans-acting factor and by cis-acting response elements. J. Virol. 61:2462-2471.

    Eberwine, J. 1996. Amplification of mRNA populations using aRNA generated from immobilized oligo(dT)-T7 primed cDNA. BioTechniques 20:584-591.

    Franchini, G. 1995. Molecular mechanisms of human T-cell leukemia/lymphotropic virus type I infection. Blood 86:3619-3639.

    Furukawa, K., T. Iioka, M. Morishita, A. Yamaguchi, H. Shindo, H. Namba, S. Yamashita, and T. Tsukazaki. 2002. Functional domains of paired-like homeoprotein Cart1 and the relationship between dimerization and transcription activity. Genes Cells 7:1135-1147.

    Gatza, M. L., J. C. Watt, and S. J. Marriott. 2003. Cellular transformation by the HTLV-I Tax protein, a jack-of-all-trades. Oncogene 22:5141-5149.

    Geerts, D., N. Schilderink, G. Jorritsma, and R. Versteeg. 2003. The role of the MEIS homeobox genes in neuroblastoma. Cancer Lett. 197:87-92.

    Goodman, R. H., and S. Smolik. 2000. CBP/p300 in cell growth, transformation, and development. Genes Dev. 14:1553-1577.

    Grassmann, R., S. Berchtold, I. Radant, M. Alt, B. Fleckenstein, J. G. Sodroski, W. A. Haseltine, and U. Ramstedt. 1992. Role of human T-cell leukemia virus type 1 X region proteins in immortalization of primary human lymphocytes in culture. J. Virol. 66:4570-4575.

    Griebel, P., T. Beskorwayne, D. L. Godson, Y. Popowych, and W. Hein. 2000. Cloning non-transformed sheep B cells. J. Immunol. Methods 237:19-28.

    Griebel, P., T. Beskorwayne, A. Van den Broeke, and G. Ferrari. 1999. CD40 signaling induces B cell responsiveness to multiple members of the gamma chain-common cytokine family. Int. Immunol. 11:1139-1147.

    Gu, L., H. W. Findley, and M. Zhou. 2002. MDM2 induces NF-kappaB/p65 expression transcriptionally through Sp1-binding sites: a novel, p53-independent role of MDM2 in doxorubicin resistance in acute lymphoblastic leukemia. Blood 99:3367-3375.

    Gurevich, R. M., P. D. Aplan, and R. K. Humphries. 2004. NUP98-topoisomerase I acute myeloid leukemia-associated fusion gene has potent leukemogenic activities independent of an engineered catalytic site mutation. Blood 104:1127-1136.

    Hai, T., and T. Curran. 1991. Cross-family dimerization of transcription factors Fos/Jun and ATF/CREB alters DNA binding specificity. Proc. Natl. Acad. Sci. USA 88:3720-3724.

    Hakoshima, T., T. Shimizu, and R. Maesaki. 2003. Structural basis of the Rho GTPase signaling. J. Biochem. (Tokyo) 134:327-331.

    Hawkins, R. D., S. Bashiardes, C. A. Helms, L. Hu, N. L. Saccone, M. E. Warchol, and M. Lovett. 2003. Gene expression differences in quiescent versus regenerating hair cells of avian sensory epithelia: implications for human hearing and balance disorders. Hum. Mol. Genet. 12:1261-1272.

    Hermeking, H. 2003. Serial analysis of gene expression and cancer. Curr. Opin. Oncol. 15:44-49.

    Higgins, M. A., B. R. Berridge, B. J. Mills, A. E. Schultze, H. Gao, G. H. Searfoss, T. K. Baker, and T. P. Ryan. 2003. Gene expression analysis of the acute phase response using a canine microarray. Toxicol. Sci. 74:470-484.

    Holleman, A., M. H. Cheok, M. L. den Boer, W. Yang, A. J. Veerman, K. M. Kazemier, D. Pei, C. Cheng, C. H. Pui, M. V. Relling, G. E. Janka-Schaub, R. Pieters, and W. E. Evans. 2004. Gene-expression patterns in drug-resistant acute lymphoblastic leukemia cells and response to treatment. N. Engl. J. Med. 351:533-542.

    Hsu, S. I., C. M. Yang, K. G. Sim, D. M. Hentschel, E. O'Leary, and J. V. Bonventre. 2001. TRIP-Br: a novel family of PHD zinc finger- and bromodomain-interacting proteins that regulate the transcriptional activity of E2F-1/DP-1. EMBO J. 20:2273-2285.

    Iioka, T., K. Furukawa, A. Yamaguchi, H. Shindo, S. Yamashita, and T. Tsukazaki. 2003. P300/CBP acts as a coactivator to cartilage homeoprotein-1 (Cart1), paired-like homeoprotein, through acetylation of the conserved lysine residue adjacent to the homeodomain. J. Bone Miner. Res. 18:1419-1429.

    Janoueix-Lerosey, I., E. Novikov, M. Monteiro, N. Gruel, G. Schleiermacher, B. Loriod, C. Nguyen, and O. Delattre. 2004. Gene expression profiling of 1p35-36 genes in neuroblastoma. Oncogene 23:5912-5922.

    Ji, W., W. Zhou, K. Gregg, N. Yu, S. Davis, and S. Davis. 2004. A method for cross-species gene expression analysis with high-density oligonucleotide arrays. Nucleic Acids Res. 32:e93.

    Kanno, T., H. Endo, K. Takeuchi, Y. Morishita, M. Fukayama, and S. Mori. 2002. High expression of methionine aminopeptidase type 2 in germinal center B cells and their neoplastic counterparts. Lab. Investig. 82:893-901.

    Khoury, J. D., L. J. Medeiros, G. Z. Rassidakis, T. J. McDonnell, L. V. Abruzzo, and R. Lai. 2003. Expression of Mcl-1 in mantle cell lymphoma is associated with high-grade morphology, a high proliferative state, and p53 overexpression. J. Pathol. 199:90-97.

    Kroes, R. A., A. Jastrow, M. G. McLone, H. Yamamoto, P. Colley, D. S. Kersey, V. W. Yong, E. Mkrdichian, L. Cerullo, J. Leestma, and J. R. Moskal. 2000. The identification of novel therapeutic targets for the treatment of malignant brain tumors. Cancer Lett. 156:191-198.

    Lawrence, B., A. Perez-Atayde, M. K. Hibbard, B. P. Rubin, P. Dal Cin, J. L. Pinkus, G. S. Pinkus, S. Xiao, E. S. Yi, C. D. Fletcher, and J. A. Fletcher. 2000. TPM3-ALK and TPM4-ALK oncogenes in inflammatory myofibroblastic tumors. Am. J. Pathol. 157:377-384.

    Lemoine, F. J., D. R. Wycuff, and S. J. Marriott. 2001. Transcriptional activity of HTLV-I Tax influences the expression of marker genes associated with cellular transformation. Dis. Markers 17:129-137.

    Lerner, A., D. H. Kim, and R. Lee. 2000. The cAMP signaling pathway as a therapeutic target in lymphoid malignancies. Leuk. Lymphoma 37:39-51.

    Li, B., T. Fink, P. Ebbesen, X. D. Liu, and V. Zachar. 2003. Expression of butyrate response factor 1 in HTLV-1-transformed cells and its transactivation by Tax protein. Arch. Virol. 148:1787-1804.

    Li, X., and B. Lim. 2003. RhoGTPases and their role in cancer. Oncol. Res. 13:323-331.

    Maass, N., F. Rosel, C. Schem, J. Hitomi, W. Jonat, and K. Nagasaki. 2002. Amplification of the BCAS2 gene at chromosome 1p13.3-21 in human primary breast cancer. Cancer Lett. 185:219-223.

    Mackay, A., C. Jones, T. Dexter, R. L. Silva, K. Bulmer, A. Jones, P. Simpson, R. A. Harris, P. S. Jat, A. M. Neville, L. F. Reis, S. R. Lakhani, and M. J. O'Hare. 2003. cDNA microarray analysis of genes associated with ERBB2 (HER2/neu) overexpression in human mammary luminal epithelial cells. Oncogene 22:2680-2688.

    Mammerickx, M., R. Palm, D. Portetelle, and A. Burny. 1988. Experimental transmission of enzootic bovine leukosis to sheep: latency period of the tumoral disease. Leukemia 2:103-107.

    Martin, M. E., T. A. Milne, S. Bloyer, K. Galoian, W. Shen, D. Gibbs, H. W. Brock, R. Slany, and J. L. Hess. 2003. Dimerization of MLL fusion proteins immortalizes hematopoietic cells. Cancer Cell 4:197-207.

    Martini, A., R. La Starza, H. Janssen, C. Bilhou-Nabera, A. Corveleyn, R. Somers, A. Aventin, R. Foa, A. Hagemeijer, C. Mecucci, and P. Marynen. 2002. Recurrent rearrangement of the Ewing's sarcoma gene, EWSR1, or its homologue, TAF15, with the transcription factor CIZ/NMP4 in acute leukemia. Cancer Res. 62:5408-5412.

    Matsuoka, M. 2003. Human T-cell leukemia virus type I and adult T-cell leukemia. Oncogene 22:5131-5140.

    Michaud, J., H. S. Scott, and R. Escher. 2003. AML1 interconnected pathways of leukemogenesis. Cancer Investig. 21:105-136.

    Moody, D. E., Z. Zou, and L. McIntyre. 2002. Cross-species hybridisation of pig RNA to human nylon microarrays. BMC Genomics 3:27.

    Moon, E., R. Lee, R. Near, L. Weintraub, S. Wolda, and A. Lerner. 2002. Inhibition of PDE3B augments PDE4 inhibitor-induced apoptosis in a subset of patients with chronic lymphocytic leukemia. Clin. Cancer Res. 8:589-595.

    Nagata, T., Y. Takahashi, Y. Ishii, S. Asai, Y. Nishida, A. Murata, T. Koshinaga, M. Fukuzawa, M. Hamazaki, K. Asami, E. Ito, H. Ikeda, H. Takamatsu, K. Koike, A. Kikuta, M. Kuroiwa, A. Watanabe, Y. Kosaka, H. Fujita, M. Miyake, and H. Mugishima. 2003. Transcriptional profiling in hepatoblastomas using high-density oligonucleotide DNA array. Cancer Genet. Cytogenet. 145:152-160.

    Ng, P. W., H. Iha, Y. Iwanaga, M. Bittner, Y. Chen, Y. Jiang, G. Gooden, J. M. Trent, P. Meltzer, K. T. Jeang, and S. L. Zeichner. 2001. Genome-wide expression changes induced by HTLV-1 Tax: evidence for MLK-3 mixed lineage kinase involvement in Tax-mediated NF-kappaB activation. Oncogene 20:4484-4496.

    Okamoto, H., K. Yasui, C. Zhao, S. Arii, and J. Inazawa. 2003. PTK2 and EIF3S3 genes may be amplification targets at 8q23-q24 and are associated with large hepatocellular carcinomas. Hepatology 38:1242-1249.

    Pasqualucci, L., P. Neumeister, T. Goossens, G. Nanjangud, R. S. Chaganti, R. Kuppers, and R. Dalla-Favera. 2001. Hypermutation of multiple proto-oncogenes in B-cell diffuse large-cell lymphomas. Nature 412:341-346.

    Pedersen, I. M., S. Kitada, L. M. Leoni, J. M. Zapata, J. G. Karras, N. Tsukada, T. J. Kipps, Y. S. Choi, F. Bennett, and J. C. Reed. 2002. Protection of CLL B cells by a follicular dendritic cell line is dependent on induction of Mcl-1. Blood 100:1795-1801.

    Peng, Y. C., D. E. Breiding, F. Sverdrup, J. Richard, and E. J. Androphy. 2000. AMF-1/Gps2 binds p300 and enhances its interaction with papillomavirus E2 proteins. J. Virol. 74:5872-5879.

    Poppe, B., J. Vandesompele, C. Schoch, C. Lindvall, K. Mrozek, C. D. Bloomfield, H. B. Beverloo, L. Michaux, N. Dastugue, C. Herens, N. Yigit, A. De Paepe, A. Hagemeijer, and F. Speleman. 2004. Expression analyses identify MLL as a prominent target of 11q23 amplification and support an etiologic role for MLL gain of function in myeloid malignancies. Blood 103:229-235.

    Pourquier, P., and Y. Pommier. 2001. Topoisomerase I-mediated DNA damage. Adv. Cancer Res. 80:189-216.

    Pozzatti, R., J. Vogel, and G. Jay. 1990. The human T-lymphotropic virus type I tax gene can cooperate with the ras oncogene to induce neoplastic transformation of cells. Mol. Cell. Biol. 10:413-417.

    Rassidakis, G. Z., R. Lai, T. J. McDonnell, F. Cabanillas, A. H. Sarris, and L. J. Medeiros. 2002. Overexpression of Mcl-1 in anaplastic large cell lymphoma cell lines and tumors. Am. J. Pathol. 160:2309-2310.

    Russo, G., C. Zegar, and A. Giordano. 2003. Advantages and limitations of microarray technology in human cancer. Oncogene 22:6497-6507.

    Sahai, E., and C. J. Marshall. 2002. RHO-GTPases and cancer. Nat. Rev. Cancer 2:133-142.

    Schuch, G., M. Machluf, G. Bartsch, Jr., M. Nomi, H. Richard, A. Atala, and S. Soker. 2002. In vivo administration of vascular endothelial growth factor (VEGF) and its antagonist, soluble neuropilin-1, predicts a role of VEGF in the progression of acute myeloid leukemia in vivo. Blood 100:4622-4628.

    Schwarze, S. R., S. E. DePrimo, L. M. Grabert, V. X. Fu, J. D. Brooks, and D. F. Jarrard. 2002. Novel pathways associated with bypassing cellular senescence in human prostate epithelial cells. J. Biol. Chem. 277:14877-14883.

    Strehl, S., A. Borkhardt, R. Slany, U. E. Fuchs, M. Konig, and O. A. Haas. 2003. The human LASP1 gene is fused to MLL in an acute myeloid leukemia with t(11;17)(q23;q21). Oncogene 22:157-160.

    Su, Y. A., M. M. Lee, C. M. Hutter, and P. S. Meltzer. 1997. Characterization of a highly conserved gene (OS4) amplified with CDK4 in human sarcomas. Oncogene 15:1289-1294.

    Szynal, M., Y. Cleuter, T. Beskorwayne, C. Bagnis, C. Van Lint, P. Kerkhofs, A. Burny, P. Martiat, P. Griebel, and A. Van den Broeke. 2003. Disruption of B-cell homeostatic control mediated by the BLV-Tax oncoprotein: association with the upregulation of Bcl-2 and signaling through NF-kappaB. Oncogene 22:4531-4542.

    Tao, W., B. Mallard, N. Karrow, and B. Bridle. 2004. Construction and application of a bovine immune-endocrine cDNA microarray. Vet. Immunol. Immunopathol. 101:1-17.

    Uranishi, H., T. Tetsuka, M. Yamashita, K. Asamitsu, M. Shimizu, M. Itoh, and T. Okamoto. 2001. Involvement of the pro-oncoprotein TLS (translocated in liposarcoma) in nuclear factor-kappa B p65-mediated transcription as a coactivator. J. Biol. Chem. 276:13395-13401.

    Vaarala, M. H., K. S. Porvari, A. P. Kyllonen, M. V. Mustonen, O. Lukkarinen, and P. T. Vihko. 1998. Several genes encoding ribosomal proteins are over-expressed in prostate-cancer cell lines: confirmation of L7a and L37 over-expression in prostate-cancer tissue samples. Int. J. Cancer 78:27-32.

    Van den Broeke, A., C. Bagnis, M. Ciesiolka, Y. Cleuter, H. Gelderblom, P. Kerkhofs, P. Griebel, P. Mannoni, and A. Burny. 1999. In vivo rescue of a silent tax-deficient bovine leukemia virus from a tumor-derived ovine B-cell line by recombination with a retrovirally transduced wild-type tax gene. J. Virol. 73:1054-1065.

    Van den Broeke, A., Y. Cleuter, T. Beskorwayne, P. Kerkhofs, M. Szynal, C. Bagnis, A. Burny, and P. Griebel. 2001. CD154 costimulated ovine primary B cells, a cell culture system that supports productive infection by bovine leukemia virus. J. Virol. 75:1095-1103.

    Van den Broeke, A., Y. Cleuter, G. Chen, D. Portetelle, M. Mammerickx, D. Zagury, M. Fouchard, L. Coulombel, R. Kettmann, and A. Burny. 1989. Even transcriptionally competent proviruses are silent in bovine leukemia virus induced tumor cells. Proc. Natl. Acad. Sci. USA 85:9263-9267.

    Visvader, J. E., D. Venter, K. Hahm, M. Santamaria, E. Y. Sum, L. O'Reilly, D. White, R. Williams, J. Armes, and G. J. Lindeman. 2001. The LIM domain gene LMO4 inhibits differentiation of mammary epithelial cells in vitro and is overexpressed in breast cancer. Proc. Natl. Acad. Sci. USA 98:14452-14457.

    Wang, Y., D. Cheong, S. Chan, and S. C. Hooi. 2000. Ribosomal protein L7a gene is up-regulated but not fused to the tyrosine kinase receptor as chimeric trk oncogene in human colorectal carcinoma. Int. J. Oncol. 16:757-762.

    Weber, P. J., C. P. Eckhard, S. Gonser, H. Otto, G. Folkers, and A. G. Beck-Sickinger. 1999. On the role of thymopoietins in cell proliferation. Immunochemical evidence for new members of the human thymopoietin family. Biol. Chem. 380:653-660.

    Willems, L., A. Burny, D. Collete, O. Dangoisse, F. Dequiedt, J. S. Gatot, P. Kerkhofs, L. Lefebvre, C. Merezak, T. Peremans, D. Portetelle, J. C. Twizere, and R. Kettmann. 2000. Genetic determinants of bovine leukemia virus pathogenesis. AIDS Res. Hum. Retrovir. 16:1787-1795.

    Willems, L., A. Gegonne, G. Chen, A. Burny, R. Kettmann, and J. Ghysdael. 1987. The bovine leukemia virus p34 is a transactivator protein. EMBO J. 6:3385-3389.

    Willems, L., H. Heremans, G. Chen, D. Portetelle, A. Billiau, A. Burny, and R. Kettmann. 1990. Cooperation between bovine leukaemia virus transactivator protein and Ha-ras oncogene product in cellular transformation. EMBO J. 9:1577-1581.

    Willems, L., R. Kettmann, F. Dequiedt, D. Portetelle, V. Voneche, I. Cornil, P. Kerkhofs, A. Burny, and M. Mammerickx. 1993. In vivo infection of sheep by bovine leukemia virus mutants. J. Virol. 67:4078-4085.

    Wu, K., M. E. Bottazzi, C. de la Fuente, L. Deng, S. D. Gitlin, A. Maddukuri, S. Dadgar, H. Li, A. Vertes, A. Pumfery, and F. Kashanchi. 2004. Protein profile of tax-associated complexes. J. Biol. Chem. 279:495-508.

    Yoshida, M. 2001. Multiple viral strategies of HTLV-1 for dysregulation of cell growth control. Annu. Rev. Immunol. 19:475-496.

    Zeng, S. X., M. S. Dai, D. M. Keller, and H. Lu. 2002. SSRP1 functions as a co-activator of the transcriptional activator p63. EMBO J. 21:5487-5497.

    Zhang, Y., N. Woodford, X. Xia, and A. W. Hamburger. 2003. Repression of E2F1-mediated transcription by the ErbB3 binding protein Ebp1 involves histone deacetylases. Nucleic Acids Res. 31:2168-2177.

    Zhao, F., R. McCarrick-Walmsley, P. Akerblad, M. Sigvardsson, and T. Kadesch. 2003. Inhibition of p300/CBP by early B-cell factor. Mol. Cell. Biol. 23:3837-3846.

    Zhou, P., N. B. Levy, H. Xie, L. Qian, C. Y. Lee, R. D. Gascoyne, and R. W. Craig. 2001. MCL1 transgenic mice exhibit a high incidence of B-cell lymphoma manifested as a spectrum of histologic subtypes. Blood 97:3902-3909.

    Zhu, Y., H. Lin, Z. Li, M. Wang, and J. Luo. 2001. Modulation of expression of ribosomal protein L7a (rpL7a) by ethanol in human breast cancer cells. Breast Cancer Res. Treat. 69:29-38.(Pavel Klener, Maud Szynal)

    http://www.100md.com/html/DirDu/2006/10/03/20/21/33.htm