Exploiting orthologue diversity for systematic detection of gain-of-function phenotypes
© Martelli et al; licensee BioMed Central Ltd. 2008
Received: 28 February 2008
Accepted: 29 May 2008
Published: 29 May 2008
Systematic search for genes whose gain-of-function by exogenous expression confers an advantage in cell-based selective screenings is a powerful method for unbiased functional exploration of the genome, and has the potential to disclose new targets for cancer therapy. A major limit of this approach resides in the labor-intensive cloning of resistant cells, identification of the integrated genes and validation of their ability to confer a selective advantage. Moreover, the selection has to be drastic and genes conferring a limited advantage are typically missed.
We developed a new functional screening strategy based on transduction of mammalian cells of a given species with an expression library from another species, followed by one-shot quantitative tracing with DNA microarrays of all library-derived transcripts before and after selection. In this way, exogenous transcripts enriched after selection, and therefore likely to confer resistance, are readily detected. We transduced a retroviral cDNA expression library from mouse testis into human and canine cells, and optimized the use of commercial murine gene expression arrays for species-specific detection of library-derived transcripts. We then conducted a functional screening by growing library-transduced canine MDCK cells in suspension, to enrich for cDNAs conferring anchorage independence. Notably, these cells show partial resistance to loss of anchorage, and the selection can be of limited stringency, compromising approaches based on clonal selection or anyway requiring high stringency. Microarray analysis revealed reproducible enrichment after three weeks of growth on polyhema for seven genes, among which the Hras proto-oncogene and Sox5. When individually transduced into MDCK cells, Sox5 specifically promoted anchorage-independent growth, thereby confirming the validity and specificity of the approach.
The procedure described here brings substantial advantages to the field of expression cloning, being faster, more systematic and more sensitive. Indeed, this strategy allowed identification and validation of genes promoting anchorage-independent growth of epithelial cells under selection conditions not amenable to conventional expression cloning.
Functional screenings based on the gain-of-function approach proved extremely valuable in the identification of novel genes involved in key processes related to cancer onset and progression, such as neoplastic transformation, resistance to apoptosis, or escape from senescence [1–3]. Identification of a gene whose expression confers neoplastic properties to normal cells, or renders cancer cells resistant to death-promoting stimuli or drugs, directly defines that gene as a potential target for novel therapeutical strategies. Screenings of this type are usually performed in mammalian cells by transducing an expression library containing full length cDNAs into a given target cell line. Then a selective stress capable of strongly reducing cell viability and proliferative potential is applied. Only cells expressing exogenous cDNAs conferring resistance to the selection will grow and form resistant colonies . Then, a huge amount of work is typically required to identify the integrated cDNAs in the resistant colonies, and to verify that they effectively mediate the selective advantage. Moreover, the selection has to be drastic to avoid the emergence of spontaneously resistant colonies, which would dramatically increase the number of false hits. As a consequence, this approach does not allow identification of genes conferring a limited advantage per se and potentially synergizing with others. Different strategies have been developed to overcome at least in part these limitations, such as vector mobilization and re-screening , or gene capture by recombination . These approaches reduced the amounts of false hits to be analyzed, but were still quite labor intensive and required multiple screening cycles.
Commercial gene expression arrays are species-specific
Therefore, the commercial arrays analyzed can in principle be used for Xenoarray analysis, without the need of a specific design. To facilitate identification and filtering of probes giving cross-hybridization signals, and subsequent assessment of exogenous transcript enrichment after the selection, we implemented a software named Xenoarray Analysis Studio (XAS). XAS enables reading and displaying gene expression data from wild-type and transduced cells, before and after selection, filtering non-specific probes, and comparing selected and unselected cells to identify enriched exogenous transcripts.
Setup of Xenoarray analysis
To set-up the procedure, we transduced human HeLa cells with a retroviral mouse testis expression library at the estimated multiplicities of infection (MOIs) of ~1.25, ~2.5 and ~5 (for assessment of transduction efficiency, see Materials and Methods). The Scatter Plot in Figure 2B illustrates the results of gene expression profiling on control and transduced cells at MOI ~1.25, using Illumina Mouse_Ref-8_V1 arrays, and following the standard procedure with a higher amount of RNA (1 μg) and double reagents and reaction volume. Clearly the amount of exogenous transcripts specifically detected (less than 400) was not adequate for a systematic screening and required improvement. To increase detection of exogenous transcripts and decrease cross-hybridization with endogenous ones, we modified the procedure by introducing a vector-specific primer for reverse transcription (T7-pFB) and optimizing all the subsequent steps (detailed in Methods). The results obtained with this optimized protocol at the same MOI are shown in Figure 2C. The number of probes giving significant signal (detection > 0.99) in transduced cells was 1605. Conversely, only 285 probes gave significant signal in untransduced cells, indicating that over 1300 exogenous transcripts were specifically detected in transduced cells. The low background is due not only to the use of the T7-pFB primer, but also to the fact that non-specific reverse transcription of endogenous human RNAs by the pFB primer produces human cRNAs that do not hybridize efficiently to the murine array. Indeed, when the T7-pFB primer was used to reverse-transcribe a human RNA from wild-type cells and the corresponding cRNA was hybridized on human arrays, the background raised from less than 300 probes to over 1600 probes (data not shown).
To evaluate whether a higher MOI could increase detection of library-derived genes, Xenoarray analysis was performed also on the populations transduced at MOI ~2.5 and ~5, using the standard or the T7-pFB primer for cDNA synthesis. The analysis confirmed that use of the T7-pFB primer increased detection of library-derived transcripts also at higher MOIs (Figure 2D). Notably, the number of library-derived transcripts reached a plateau at MOI ~2.5 for both primers. Specific detection of almost 1700 murine transcripts at this MOI further confirmed the efficiency of retroviral expression libraries as a tool for functional screenings . In view of the fact that, by average, 9000–10000 transcripts are typically detected in a microarray experiment with Illumina beadchips, we estimated that Xenoarray analysis detects the 15–20% most represented exogenous cDNAs. Rarer cDNAs are not detected in the initial transduced population, but may become detectable if enriched by the selection procedure. To estimate the enrichment rate required to render a rare cDNA detectable after selection, we exploited a key feature of gene profiling based on Serial Analysis of Gene Expression (SAGE), i.e. the absolute abundance of each transcripts measured as parts per million (PPM) . In particular, we analyzed the distribution of the abundance of transcripts available from a published SAGE analysis on mouse testis . We observed that, like a typical microarray experiment, SAGE identified around 10000 transcripts, and the 1700 most abundant ones had a representation of 76 PPM or higher, while the remaining transcripts had a representation of 8 to 76 PPM [see Additional file 1]. Based on this analysis, a 10-fold enrichment of a rare transcript should be enough to render it detectable by Xenoarray analysis.
Xenoarray analysis identifies genes promoting anchorage-independent growth
The four populations were also assayed for proliferation in adherence conditions and in soft agar (Figure 3B, C). The LIB-SEL population displayed the highest growth rate in adherence, and formed much larger colonies in soft agar compared to LIB-UNS cells (p < 0.0025). A slightly increased growth was also observed for GFP-SEL compared to GFP-UNS cells (p < 0.021), possibly highlighting insertional mutagenesis events which frequently occur with retroviral vectors . However, such events cannot explain the greater increase observed in the LIB-SEL population (LIB-SEL vs GFP-SEL: p < 0.02), which therefore is likely to derive from expression of advantageous exogenous transcripts. To identify these transcripts, we performed xenoarray analysis comparing LIB-SEL vs. LIB-UNS cells (Figure 3D), and observed a significant number of transcripts detected at higher levels in selected cells.
Interestingly, no library-derived transcripts are less abundant after selection. This indicates a limited stringency of the selection, allowing even rare populations to remain represented. In fact, we noticed that after an initial crisis in the first two days of culture in suspension, MDCK cells adapted to the condition by forming aggregates and reducing the proliferation rate, without massive cell death. More drastic selection procedures, such as crossing a transwell membrane and "diving" to the bottom of the well, induced loss of the majority of the library-derived transcripts (data not shown). To validate reproducibility of the selection within the same transduced population, a second selection had been conducted in parallel on the library-transduced cells described above ("Selection B"). Similarly, also the LIB-UNS population was split in two for replicate analysis (LIB-UNS A and LIB-UNS B). Also in Selection B, Xenoarray analysis highlighted enriched transcripts [see Additional file 2]. To verify if the same transcripts were enriched in both selections, we calculated the log2 ratio between the signals in selected and unselected cells for each transcript in each selection, and compared the results of the two selections. Over 70 percent of the genes enriched more than 2-fold in selection A were also enriched over 2-fold in selection B. To validate the 2-fold threshold for significant enrichment, we compared Xenoarray data obtained from the two LIB-UNS populations. Only two probes gave a fold-change higher than 2 (2.08 and 2.04) in this control comparison.
Library-derived genes enriched in MDCK cells by two independent selections for anchorage independent growth.
Illumina Probe ID
Fold Enrichment Selection A
Fold Enrichment Selection B
To verify the specificity of the selection hits, we carried out two alternative selection procedures, one based on serum withdrawal, and therefore not involving anchorage independence, and the other based on the crossing of a transwell membrane and "diving" to the bottom of the plate, more directly linked to anchorage independence. Then we compared the enrichment of the 34 probes emerged from polyhema selection with the enrichments observed in these two selections [see Additional file 3]. While no correlation was observed between polyhema selection and serum withdrawal (Pearson = -0.32), a striking concordance was observed between polyhema and transwell "diving" selections (Pearson = 0.94). These results confirm that cDNA enrichment is not just the result of a general tendency of a subpopulation of cells to overgrow, but rather is specifically driven by the type of selection.
Reproducible enrichment of exogenous cDNAs by parallel selection in the same transduced population does not rule out two main possible artefacts: (i) enrichment may derive from deregulation of an endogenous gene by insertional mutagenesis; (ii) a very small subpopulation of resistant cells may exist from the beginning, and a cDNA transduced in these cells may get reproducibly enriched by carryover.
Library-derived genes enriched in MDCK cells by two independent infections and selections for anchorage independent growth.
Illumina Probe ID
Fold Enrichment Infection 1
Fold Enrichment Infection 2
Repetition of the transduction and selection procedure led to a limited validation of the hits identified in the first screening. This result corroborates the idea that process streamlining allowed by the xenoarray approach should be exploited for performing multiple independent screenings, thereby allowing identification of more consistent hits. However, it should be noted that genes enriched in only one transduction-selection experiment are not necessarily false hits. Indeed, the complexity of the integrated cDNA repertoire can vary across independent transduction experiments. This is particularly true for rare transcripts, which may not be represented in one of the two transduced populations, or may integrate in unfavourable regions of the host genome. Moreover, rare transcripts need to be highly enriched to emerge from the microarray background, and even in that case the differential signal may not necessarily be greater than 2-fold.
Validation of the ability of SOX5 to promote anchorage-independent growth
After confirming expression of the exogenous protein by Western Blot (Figure 4B), SOX5- or GFP-transduced populations were assayed for proliferation on plastic substrate, polyhema or soft agar (Figure 4C, D). In all three assays, increased growth was observed for SOX5-transduced cells compared to the GFP controls. Interestingly, the growth advantage was more evident when cells were cultured in the absence of anchorage (1.8-, 2.3- and 2.4- fold increase, respectively for growth on plastic, polyhema and soft agar). These data confirm that Xenoarray-based screenings identify hits with specific biological properties defined by the type of selection.
DNA microarrays have greatly advanced our ability to identify genes whose expression is associated with particular phenotypes or biological processes [28, 29]. Nevertheless, they have been falling short in determining cause-effect relationships between genes and phenotypes, especially in the field of cancer research. Recent and more direct approaches employed microarrays for "barcode" screenings with shRNA vectors, to identify genes whose loss may render cancer cells resistant to selective stresses [8, 9]. However, such genes are not immediately exploitable as therapeutical targets because their blockade is beneficial to cancer cells. Rather, they may provide the rationale for identifying other druggable genes whose loss of function is selectively toxic for cancer cells. The potential usefulness of gain-of-function-based approaches in cancer research is further confirmed by recent evidence of the involvement of the PI3K pathway in the resistance of human breast cancer to the HER2-blocking antibody Trastuzumab. In particular, Berns and colleagues identified PTEN in an shRNA-based barcode screening for resistance to Trastuzumab, and then found that the acquisition of resistance by shRNA-driven loss of PTEN is mirrored by a much stronger in vitro resistance phenotype driven by overexpression or mutation of PIK3CA, i.e. by a gain-of-function approach .
All screenings based on selection find a major challenge in the need of avoiding false hits. Being substantially streamlined, the screening described here can be easily repeated multiple times, which increases the rate of true positive hits. It should be also noticed that most of the significant enrichments observed occurred for cDNAs that were not detected in transduced cells before selection. This further confirms that the range of functional exploration extends well beyond the 1700 cDNAs already detectable in unselected cells. The xenoarray approach does not directly address a key bottleneck for expression library-based screens, i.e. their reliance on high quality, representative cDNA collections and the ability to efficiently introduce these genes into mammalian cells. However, xenoarray analysis comparing transduced and untransduced cells can be used for the optimization of library construction and transduction, allowing a one-shot measurement of the complexity of the transduced library. A specific caveat to be considered is that the xenoarray approach assumes functional conservation between orthologous proteins. However, functional conservation across mammalian species is generally the rule rather than the exception, and single hit functional conservation can be further checked using orthologous proteins databases such as P-POD . Moreover, hit validation would typically imply transduction of the target cells with an ORF derived from their species of origin, or RNAi-based targeting of the endogenous gene, which would further support the biological relevance of the finding.
Our data show that one of the identified murine genes, SOX5, actually promotes anchorage-independent growth of MDCK epithelial cells also when exogenously expressed in its human version, and further support a possible role for this gene in tumor onset and progression. The third-best hit of the screening, Hras1, is a well-known proto-oncogene, and its ability to confer anchorage independence to normal cells has been well documented . The most frequent mechanism of activation of RAS-family genes is point mutation . However, increased expression of normal RAS proto-oncogenes due to gene amplification has also been reported to occur in human cancer . Therefore, the finding of this gene as a hit can be considered as a positive control of the screening efficiency. Existing knowledge about the other screening hits is compatible with their potential role in anchorage-independent growth as well. Hnrpc encodes an mRNA-binding protein found to stabilize the mRNA and to increase expression of the urokinase receptor , a well-known player in cancer onset and progression . The Akap4 gene is expressed only in the postmeiotic phase of spermatogenesis and its protein product anchors cAMP-dependent protein kinase A in a restricted region of the mammalian sperm flagellum . Its overexpression in MDCK is therefore likely to drive PKA activation and relocalization to the cytoskeleton, thereby affecting cell motility and adhesion. However, due to its extremely restricted expression, this gene is unlikely to physiologically promote anchorage independence in epithelial cells. Fbxo6b is a member of the E3 glycoprotein-specific ubiquitin ligase family, playing a role in endoplasmic reticulum-associated degradation . Its overexpression may therefore promote anchorage-independent growth by modifying the expression pattern of transmembrane glycoproteins. Given the low stringency of the polyhema growth selection on MDCK cells, which did not reduce the repertoire of library-derived transcripts detected by the xenoarray after the selection, it is likely that many of the above-described hits provide only a limited advantage, and could not be detected by a classical approach based on more drastic selection strategies. In this view, it will be interesting to assess their possible reciprocal cooperation.
The procedure described here employs full-length cDNA expression libraries derived from a given tissue or cell line, which brings advantages and disadvantages, compared to arrayed collections of open reading frames (ORFs). The main disadvantage of using libraries is that some genes may fail to be represented and others may be overrepresented. In this view, Xenoarray analysis could also be applied to ORF collections, which would allow a much tighter control on the composition and relative abundance of the cDNAs used for the screening, as well as focused exploration of gene subsets. In this case, however, it should be noted that the majority of the probes of commercial expression arrays fall outside of the transcripts' ORFs. Therefore, custom species-specific expression arrays should be designed to cover areas of non-homology within the ORF regions. As an advantage, libraries provide a more comprehensive repertoire of all the transcripts and of their isoforms expressed in the tissue of origin, thereby being more explorative. Moreover, when expression libraries are derived from cancerous tissues or cells, this approach can be combined with resequencing of the hits to highlight mutated genes, potentially exploitable as therapeutical targets. It therefore provides a powerful tool for the dissection of the mechanisms of cancer onset and progression and of resistance to anti-neoplastic treatments.
The Xenoarray technology described here provides a new, efficient approach to expression cloning and functional genomics. It takes full advantage of genome-wide expression profiling to identify genes that confer resistance to a specific selective stress, thereby establishing a cause-effect relationship. Being more sensitive and systematic, the procedure does not require extreme selection stringency and isolation of resistant individual clones. In this way, also genes conferring a partial advantage can be identified and further explored for their possible reciprocal cooperation.
Cell Culture, Reagents and viral transduction
Madin-Darby canine kidney (MDCK) cells and HeLa cells were from ATCC. They were cultured in Dulbecco's modified Eagle's medium (DMEM) (Gibco) supplemented with 10% FBS (Sigma) in a humidified atmosphere of 5% CO2. The mouse testis retroviral expression library, packaged in the VSV envelope was purchased from Stratagene (ViraPort, Cat n. 972300). To titer the library viral supernatant, we used a GFP retroviral supernatant provided by the manufacturer at the same titer, seeding 5*104 cells onto 35 mm tissue culture plates. The following day, 1 ml of dilutions from 10-1 to 10-5 of the pFB-hrGFP retroviral supernatant in growth medium supplemented with 10 μg/ml DEAE-dextran (Amersham Bioscence) were added to each well. After 3 hours, an additional 1 ml of growth medium was added to each well. GFP expression analysis was performed after 48 hours by flow cytometry: cells were trypsinized, diluted in a 1% paraformalhdeide-2% FBS solution and analyzed on a FACS Calibur flow cytometer (Becton Dickinson). The titer of the library, expressed as GFP transduction units (TU) per ml, was calculated as 105 (the number of infected cells) times the fraction of green cells times the dilution factor. HeLa transduction experiments were performed by plating 5*104 cells in 35 mm wells. After one day, 1 ml of library supernatant dilutions prepared as above were added. Medium with no virus was added to generate an uninfected control. The plates were returned to 37° for 3 hours, then 1 ml of growth medium was added to each well. After 48 hours infected and control cells were expanded and used for microarray analysis. MDCK transduction was performed as above, except that 5*105 cells were infected with about 1*106 TU of viral supernatant in 60 mm dishes (MOI = 2). The pFB-hrGFP retroviral supernatant was used at the same MOI as a control. After 48 hours library- and GFP-infected cells were expanded for the functional screening.
Genomic analysis of probes cross-hybridization
Probes cross-hybridization analysis was performed by blasting all the probes from the Illumina Mouse-6_V1 chip against the transcriptome databases from Ensembl (release 91. The interrogated databases were the mm_cdna35 and est_mus for mouse, hs_cdna36 and est_hum for human. Dog transcripts were obtained from cf_cdna_broadd1 and est_mam, which required filtering to exclude non-canine transcripts. The blastjob run was launched according to the following parameters:
$ blastall -p blastn -i fasta -d "database" -v 100 -b 100 -o out -a 2 -W 7 -m 7,
where the W parameter sets the alignment seed to 7 bases. The blast output data were parsed with a Perl script to display the number of identical nucleotides between each probe and its best hit.
Anchorage-independent growth selection
Polyhema-coated 100 mm Petri dishes were prepared by applying 4 ml of a 12 mg/ml solution of poly-hydroxy-ethyl-methacrylate (polyhema; Sigma) in ethanol, drying under tissue culture hood, repeating the application once and incubating the plates overnight at 37°C. For the selection, after trypsinization, 1.5 × 106 cells were plated onto polyhema plates and cultured for one week, removing cell debris by spinning the suspension at low speed (400 rpm) and resuspending the pellet in fresh medium every 2–3 days. Cells were then allowed to recover on normal dishes for 24 hours, after which the selection was repeated for a total of 3 cycles. Selected cells were expanded on regular plates for one week before being used for microarray analysis and functional assays.
Adherent and suspension growth assays
For cell viability assays, 103 cells of each cell line were seeded in triplicate in 96-well plates, one for each time of the growth curve assay, both on plastic and on Polyhema coated plates. The day after, a tetrazolium salt-based reagent (CellTiter96 Aqueous One Solution, Promega) was added to each well according to the instructions provided by the manufacturer. After an incubation of 2 h, absorbance was read at 490 nm on a DTX 880 plate reader (Beckman Coulter, Milan, Italy). For the soft agar growth assay, 104 cells were resuspended in 1 ml of 0.5% top agar (SeaPlaque Agarose, Cambrex, UK) in growth medium and seeded in 6-well plates previously coated with 2 ml of 1% basal agar in growth medium. The assay was performed in duplicate. After 2 weeks, phase-contrast pictures were captured and analyzed with a BD Pathway Workstation.
RNA extraction and processing for microarray analysis
RNA was extracted using the TRIzol reagent (Invitrogen), according to the manufacturer's protocol, and then further purified using the RNeasy Mini kit from Qiagen. The quantification and quality analysis of RNA was performed on a Bioanalyzer 2100 (Agilent). Synthesis of cDNA and biotinylated cRNA was performed using the Illumina TotalPrep RNA Amplification Kit (Ambion Cat. n. IL1791), according to the manufacturer's protocol, with the following variations to optimize xenoarray analysis: (i) standard cDNA synthesis with the T7-dT(24) primer: 1 μg of total RNA was used, with the doubling of the reaction volume and of all reagents; (ii) library-specific cDNA synthesis: 20 μg of total RNA were used with 4 pm of a pFB-specific primer (T7-pFB, sequence: GGCCAGTGAATTGTAATACGACTCACTATAGGGAGGCGGCGAACCCCAGAGTCCCGCTCA, HPLC-purified from Sigma) with standard reaction conditions. The T7-pFB primer contains the T7 promoter (for cRNA synthesis) followed by a vector-specific sequence, which is present in the 3' region of all transcripts derived from the library. Quality assessment and quantification of cRNAs were performed on Bioanalyzer 2100.
Hybridization of HeLa-derived cRNAs was carried out for 18 hours on Illumina Mouse_Ref-8_V1 arrays (Cat. N. BD-26-201) according to the manufacturer's protocol, using 15 μg of T7-dT(24)-derived cRNA or 1.5 μg of T7-pFB-derived cRNA. These arrays contain circa 24000 probes exploring the transcripts contained in the Refseq database, and therefore more reliable. Hybridization of MDCK-derived cRNAs was carried out on Illumina Mouse-6_V1 arrays (Cat. N. BD-26-101), using 1.5 μg of T7-pFB-derived cRNA. These arrays contain all the RefSeq probes present in the Mouse_Ref_8_V1 arrays, plus additional 24 k probes exploring less characterized transcripts, additional Unigene clusters and singletones, to allow a more complete coverage of the transcriptome. Array washing was performed using Illumina High-stringency wash buffer for 30 min at 55°C, and followed by staining and scanning according to standard Illumina protocols. Probe intensity and detection data were obtained using the Illumina BeadStudio software, and further processed with the XAS software. Microarray data are available at GEO, dataset GSE11721.
Xenoarray data analysis
The Xenoarray analysis pipeline was implemented in the XAS software (C++ with QT library from Trolltech ) and subdivided in 3 main tasks: (i) Removal of probes cross-hybridizing to endogenous transcripts. For this task, the "Detection" value provided by the Illumina Beadstudio software was used. This value ranges from 0 (non detected) to 1 (detected), with 0.99 or higher as the usual thresholds for significant detection. All probes with a Detection value higher than the threshold set by the user in untransduced cells are excluded from the subsequent steps. (ii) Data preprocessing to improve reproducibility, and reduce technical noise. The data can be either rank invariant normalized or scaled according to the tenth percentile of the average distribution of all the chips. A background reduction is then performed by subtracting 2/3 of the minimum signal value. (iii) Identification of probes giving higher signal after selection. The software compares Xenoarray data from transduced cells before and after the selection, and calculates for each probe the log2 of the ratio between the signals before and after selection. If multiple selection experiments are conducted, the software can compare them to identify reproducibly enriched exogenous cDNAs. The software and user guide are available on Sourceforge .
Analysis of Sox5 expression by PCR and western blot
For PCR, cDNA was synthesized using the RT high capacity cDNA kit (Applied Biosystems), and the mouse-specific Sox-5 primers: 5'-GATATTGGGATCTCGCTGGA-3'; 5'- AAGTACTGCCGCATTTCCTG-3', and the following cycle program: 5' at 94°, followed by 25 cycles (2' at 94°- 30" at 55°- 30" at 72°), and 10' at 72°. For Western blot, anti-human Sox5 antibody was purchased from Abcam (Cambridge, UK). Total cellular proteins were extracted by solubilizing the cells in boiling Laemmli buffer followed by sonication. 100 μg of lysates were run on SDS-polyacrylamide gels and transferred onto nitrocellulose membranes (Hybond; GE Healthcare). Nitrocellulose-bound antibodies were detected by the ECL system (GE Healthcare).
Sox5 cDNA cloning and expression
The full length human SOX5 cDNA was purchased from RZPD German Resource Center for Genome Research and the full coding sequence was cloned into the pFB retroviral vector (Stratagene). Retroviral supernatant was produced by transfecting pFB-SOX5 or pFB-hrGFP with the pVPack-GP (gag-pol-expressing vector; Stratagene) and the pVPack-VSV-G (env-expressing vector (Stratagene) in 293T packaging cells. The pFB-hrGFP retroviral supernatant was used as control of infection efficiency. MDCK transduction with the supernatants was performed as above.
Xenoarray Analysis Studio
Multiplicity of Infection
Serial Analysis of Gene Expression
Part per Million
Madin-Darby Canine Kidney
Green Fluorescent Protein
Princeton Protein Orthology Database
Open Reading Frame.
We thank Tommaso Renzulli for support with image analysis and Barbara Martinoglio for technical assistance. We also thank Antonella Cignetto and Michela Bruno for secretarial assistance. This research was supported by grants from AIRC, the EC (contract n. 503438 "TRANSFOG"), CNR-MIUR, FIRB-MIUR, Regione Piemonte and the Foundations CRT and 'Compagnia di San Paolo'.
- Aruffo A: Expression cloning systems. Curr Opin Biotechnol. 1991, 2: 735-741. 10.1016/0958-1669(91)90044-6.PubMedView ArticleGoogle Scholar
- Kitamura T, Koshinoa Y, Shibata F, Okia T, Nakajima H, Nosakab T, Kumagai H: Retrovirus-mediated gene transfer and expression cloning: Powerful tools in functional genomics. Exp Hematol. 2003, 31: 1007-1014.PubMedView ArticleGoogle Scholar
- Nakayama N, Yokota T, Arai K: Use of mammalian cell expression cloning systems to identify genes for cytokines, receptors, and regulatory proteins. Curr Opin Biotechnol. 1992, 3: 497-505. 10.1016/0958-1669(92)90077-V.PubMedView ArticleGoogle Scholar
- Simonsen H, Lodish HF: Cloning by function: expression cloning in mammalian cells. Trends Pharmacol. 1994, 15: 437-441. 10.1016/0165-6147(94)90052-3.View ArticleGoogle Scholar
- Peeper DS, Shvarts A, Brummelkamp T, Douma S, Koh EY, Daley GQ, Bernards R: A functional screen identifies hDRIL1 as an oncogene that rescues RAS-induced senescence. Nat Cell Biol. 2002, 4: 148-153. 10.1038/ncb742.PubMedView ArticleGoogle Scholar
- Hannon GJ, Sun P, Carnero A, Xie LW, Maestro R, Conklin DS, Beach D: MaRX: An Approach to Genetics in Mammalian Cells. Science. 1999, 283: 1129-1130. 10.1126/science.283.5405.1129.PubMedView ArticleGoogle Scholar
- Shoemaker DD, Lashkari DA, Morris D, Mittmann M, Davis RV: Quantitative phenotypic analysis of yeast deletion mutants using a highly parallel molecular bar-coding strategy. Nature Genetics. 1996, 14: 450-456. 10.1038/ng1296-450.PubMedView ArticleGoogle Scholar
- Berns K, Hijmans EM, Mullenders J, Brummelkamp TR, Velds A, Heimerikx M, Kerkhoven RM, Madiredjo M, Nijkamp W, Weigelt B, Agami R, Ge W, Cavet G, Linsley PS, Beijersbergen RL, Bernards R: A large-scale RNAi screen in human cells identifies new components of the p53 pathway. Nature. 2004, 428: 431-437. 10.1038/nature02371.PubMedView ArticleGoogle Scholar
- Silva JM, Li MZ, Chang K, Ge W, Golding MC, Rickles RJ, Siolas D, Hu G, Paddison PJ, Schlabach MR, Sheth N, Bradshaw J, Burchard J, Kulkarni A, Cavet G, Sachidanandam R, McCombie WR, Cleary MA, Elledge SJ, Hannon GJ: Second-generation shRNA libraries covering the mouse and human genomes. Nat Genet. 2005, 11: 1281-1288. 10.1038/nm1205-1281.View ArticleGoogle Scholar
- Mouse Genome Sequencing Consortium, Waterston RH, Lindblad-Toh K, Birney E, Rogers J, Abril JF, Agarwal P, Agarwala R, Ainscough R, Alexandersson M, An P, Antonarakis SE, Attwood J, Baertsch R, Bailey J, Barlow K, Beck S, Berry E, Birren B, Bloom T, Bork P, Botcherby M, Bray N, Brent MR, Brown DG, Brown SD, Bult C, Burton J, Butler J, Campbell RD, Carninci P: Initial sequencing and comparative analysis of the mouse genome. Nature. 2002, 420: 520-562. 10.1038/nature01262.View ArticleGoogle Scholar
- Velculescu VE, Zhang L, Vogelstein B, Kinzler KW: Serial analysis of gene expression. Science. 1995, 270: 484-487. 10.1126/science.270.5235.484.PubMedView ArticleGoogle Scholar
- Siddiqui AS, Khattra J, Delaney AD, Zhao Y, Astell C, Asano J, Babakaiff R, Barber S, Beland J, Bohacec S, Brown-John M, Chand S, Charest D, Charters AM, Cullum R, Dhalla N, Featherstone R, Gerhard DS, Hoffman B, Holt RA, Hou J, Kuo BY, Lee LL, Lee S, Leung D, Ma K, Matsuo C, Mayo M, McDonald H, Prabhu AL: A mouse atlas of gene expression: large-scale digital gene-expression profiles from precisely defined developing C57BL/6J mouse tissues and cells. Proc Natl Acad Sci USA. 2005, 102: 18485-18490. 10.1073/pnas.0509455102.PubMedPubMed CentralView ArticleGoogle Scholar
- Paul D: Growth control in HeLa cells by serum and anchorage. Exp Cell Res. 1978, 114: 435-438. 10.1016/0014-4827(78)90503-7.PubMedView ArticleGoogle Scholar
- Macville M, Schrock E, Padilla-Nash H, Keck C, Ghadimi BM, Zimonjic D, Popescu N, Ried T: Comprehensive and definitive molecular cytogenetic characterization of HeLa cells by spectral karyotyping. Cancer Res. 1999, 59 (1): 141-150.PubMedGoogle Scholar
- Frisch SM, Francis H: Disruption of epithelial cell-matrix interactions induces apoptosis. J Cell Biol. 1994, 124: 619-626. 10.1083/jcb.124.4.619.PubMedView ArticleGoogle Scholar
- Mantovani J, Holic N, Martinez K, Danos O, Perea J: A high throughput method for genome-wide analysis of retroviral integration. Nucleic Acids Res. 2006, 34: e134-10.1093/nar/gkl716.PubMedPubMed CentralView ArticleGoogle Scholar
- Schubbert S, Shannon K, Bollag G: Hyperactive Ras in developmental disorders and cancer. Nat Rev Cancer. 2007, 7: 295-308. 10.1038/nrc2109.PubMedView ArticleGoogle Scholar
- Singh B, Arlinghaus RB: The mos proto-oncogene product: its role in oocyte maturation, metaphase arrest, and neoplastic transformation. Mol Carcinog. 1992, 3: 182-189. 10.1002/mc.2940060303.View ArticleGoogle Scholar
- Dufault VM, Oestreich AJ, Vroman BT, Karnitz LM: Identification and characterization of RAD9B, a paralog of the RAD9 checkpoint gene. Genomics. 2003, 6: 644-651. 10.1016/S0888-7543(03)00200-3.View ArticleGoogle Scholar
- Hiraoka Y, Ogawa M, Sakai Y, Kido S, Aiso S: The mouse Sox5 gene encodes a protein containing the leucine zipper and the Q box. Biochim Biophys Acta. 1998, 1399 (1): 40-46.PubMedView ArticleGoogle Scholar
- Wunderle VM, Critcher R, Ashworth A, Goodfellow PN: Cloning and characterization of SOX5, a new member of the human SOX gene family. Genomics. 1996, 2: 354-358. 10.1006/geno.1996.0474.View ArticleGoogle Scholar
- Smits P, Li P, Mandel J, Zhang Z, Deng JM, Behringer RR, de Crombrugghe B, Lefebvre V: The transcription factors L-Sox5 and Sox6 are essential for cartilage formation. Dev Cell. 2001, 2: 277-290. 10.1016/S1534-5807(01)00003-X.View ArticleGoogle Scholar
- Ueda R, Yoshida K, Kawakami Y, Kawase T, Toda M: Expression of a transcriptional factor, SOX6, in human gliomas. Brain Tumor Pathol. 2004, 1: 35-38. 10.1007/BF02482175.View ArticleGoogle Scholar
- Ueda R, Yoshida K, Kawase T, Kawakami Y, Toda M: Preferential expression and frequent IgG responses of a tumor antigen, SOX5, in glioma patients. Int J Cancer. 2007, 8: 1704-1711. 10.1002/ijc.22472.View ArticleGoogle Scholar
- Iguchi H, Urashima Y, Inagaki Y, Ikeda Y, Okamura M, Tanaka T, Uchida A, Yamamoto TT, Kodama T, Sakai J: SOX6 suppresses cyclin D1 promoter activity by interacting with beta-catenin and histone deacetylase 1, and its down-regulation induces pancreatic beta-cell proliferation. J Biol Chem. 2007, 26: 19052-19061. 10.1074/jbc.M700460200.View ArticleGoogle Scholar
- Zafarana G, Gillis AJ, van Gurp RJ, Olsson PG, Elstrodt F, Stoop H, Millán JL, Oosterhuis JW, Looijenga LH: Coamplification of DAD-R, SOX5, and EKI1 in human testicular seminomas, with specific overexpression of DAD-R, correlates with reduced levels of apoptosis and earlier clinical manifestation. Cancer Res. 2002, 62 (6): 1822-1831.PubMedGoogle Scholar
- Storlazzi CT, Albano F, Lo Consolo C, Dogliosi C, Guastadisegni MC, Impera L, Lonoce A, Funes S, Macrì E, Iuzzolino P, Panagopoulos I, Specchia G, Rocchi M: Upregulation of the SOX5 by promoter swapping with the P2RY8 gene in primary splenic follicular lymphoma. Leukemia. 2007, 21 (10): 2221-2225. 10.1038/sj.leu.2404784.PubMedView ArticleGoogle Scholar
- Schena M, Shalon D, Davis RW, Brown PO: Quantitative monitoring of gene expression patterns with a complementary DNA microarray. Science. 1995, 270: 467-470. 10.1126/science.270.5235.467.PubMedView ArticleGoogle Scholar
- Lockhart DJ, Dong H, Byrne MC, Follettie MT, Gallo MV, Chee MS, Mittmann M, Wang C, Kobayashi M, Horton H, Brown EL: Expression monitoring by hybridization to high-density oligonucleotide arrays. Nature Biotechnol. 1996, 14: 1675-1678. 10.1038/nbt1296-1675.View ArticleGoogle Scholar
- Berns K, Horlings HM, Hennessy BT, Madiredjo M, Hijmans EM, Beelen K, Linn SC, Gonzalez-Angulo AM, Stemke-Hale K, Hauptmann M, Beijersbergen RL, Mills GB, Vijver van de MJ, Bernards R: A Functional Genetic Approach Identifies the PI3K Pathway as a Major Determinant of Trastuzumab Resistance in Breast Cancer. Cancer Cell. 2007, 12: 395-402. 10.1016/j.ccr.2007.08.030.PubMedView ArticleGoogle Scholar
- Heinicke S, Livstone MS, Lu C, Oughtred R, Kang F, Angiuoli SV, White O, Botstein D, Dolinski K: The Princeton Protein Orthology Database (P-POD): a comparative genomics analysis tool for biologists. PLoS ONE. 2007, 2: 766-10.1371/journal.pone.0000766.View ArticleGoogle Scholar
- Hurlin PJ, Fry DG, Maher VM, McCormick JJ: Morphological transformation, focus formation, and anchorage independence induced in diploid human fibroblasts by expression of a transfected H-ras oncogene. Cancer Res. 1987, 47: 5752-5757.PubMedGoogle Scholar
- Bos JL: The ras gene family and human carcinogenesis. Mutat Res. 1988, 195: 255-271.PubMedView ArticleGoogle Scholar
- Shetty S: Regulation of urokinase receptor mRNA stability by hnRNP C in lung epithelial cells. Mol Cell Biochem. 2005, 272: 107-118. 10.1007/s11010-005-7644-2.PubMedView ArticleGoogle Scholar
- Pillay V, Dass CR, Choong PF: The urokinase plasminogen activator receptor as a gene therapy target for cancer. Trends Biotechnol. 2007, 25: 33-39. 10.1016/j.tibtech.2006.10.011.PubMedView ArticleGoogle Scholar
- Edwards AS, Scott JD: A-kinase anchoring proteins: protein kinase A and beyond. Curr Opin Cell Biol. 2000, 12: 217-221. 10.1016/S0955-0674(99)00085-X.PubMedView ArticleGoogle Scholar
- Yoshida Y, Tokunaga F, Chiba T, Iwai K, Tanaka K, Tai T: Fbs2 is a new member of the E3 ubiquitin ligase family that recognizes sugar chains. J Biol Chem. 2003, 278: 43877-43884. 10.1074/jbc.M304157200.PubMedView ArticleGoogle Scholar
- Trolltech website. [http://www.trolltech.com]
- Xas project page on Sourceforge. [http://www.sourceforge.net/projects/xas]
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.