Global gene expression profiling of cells overexpressing SMC3
© Ghiselli and Liu. 2005
Received: 20 October 2004
Accepted: 12 September 2005
Published: 12 September 2005
The Structural Maintenance of Chromosome 3 protein (SMC3) plays an essential role during the sister chromatid separation, is involved in DNA repair and recombination and participates in microtubule-mediated intracellular transport. SMC3 is frequently elevated in human colon carcinoma and overexpression of the protein transforms murine NIH3T3 fibroblasts. In order to gain insight into the mechanism of SMC3-mediated tumorigenesis a gene expression profiling was performed on human 293 cells line stably overexpressing SMC3.
Biotinylated complementary RNA (cRNA) was used for hybridization of a cDNAmicroarray chip harboring 18,861 65-mer oligos derived from the published dEST sequences. After filtering, the hybridization data were normalized and statistically analyzed. Sixty-five genes for which a putative function could be assigned displayed at least two-fold change in their expression level. Eighteen of the affected genes is either a transcriptional factor or is involved in DNA and chromatin related mechanisms whereas most of those involved in signal transduction are members or modulators of the ras-rho/GTPase and cAMP signaling pathways. In particular the expression of RhoB and CRE-BPa, two mediators of cellular transformation, was significantly enhanced. This association was confirmed by analyzing the RhoB and CRE-BPa transcript levels in cells transiently transfected with an SMC3 expression vector. Consistent with the idea that the activation of ras-rho/GTPase and cAMP pathways is relevant in the context of the cellular changes following SMC3 overexpression, gene transactivation through the related serum (SRE) and cAMP (CRE) cis-acting response elements was significantly increased.
We have documented a selective effect of the ectopic expression of SMC3 on a set of genes and transcriptional signaling pathways that are relevant for tumorigenesis. The results lead to postulate that RhoB and CRE-BPa two known oncogenic mediators whose expression is significantly increased following SMC3 overexpression play a significant role in mediating SMC3 tumorigenesis.
The Structural Maintenance of Chromosome 3 protein (SMC3) is a key component of the nuclear multimeric protein complex named cohesin. This complex, which also includes SMC1, scc1 and scc3, forms joints between the replicating DNA strands and holds together the sister chromatids throughout G2 phase while opposing the splitting force exerted by the spindle microtubules . In addition to its essential role in mitotic and meiotic chromosome segregation, SMC3 plays an important role in DNA recombination , is a component of the DNA damage repair mechanism  and is involved in the microtubule-based intracellular transport . SMC3 expression is elevated in a large fraction of human colon carcinoma and in the intestinal tumors of mice genetically prone to develop polyps . SMC3 expression level is controlled in intestinal epithelial cells through the APC/β-catenin/TCF4 transactivation pathway a signaling system that is almost invariably altered in colon carcinomas . Furthermore NIH3T3 fibroblasts overexpressing SMC3 lose cell-cell contact inhibition, display anchorage-independent growth and form foci of transformation . These findings support the idea that up-regulation of SMC3 expression is either permissive or sufficient to trigger cell transformation. The mechanism of SMC3-mediated cell transformation has however remained speculative.
In order to identify genes whose expression is affected by SMC3 overexpression, high-density oligonucleotide microarray chip harboring 18,861 human gene-specific oligonucleotides were hybridized with cRNA derived from 293 cells with different expression level of SMC3. The 293 cells are human embryonic kidney cells that have become immortalized following transformation by adenovirus type 5  and display latent tumorigenicity . This represent a well characterized model for human tumorigenesis that has been frequently utilized for in vitro and in vivo assessment of the oncogenic or tumor suppressor potential of a number of genes [9–13]. Statistical analysis of the microarray data has revealed that many of the genes affected by SMC3 overexpression in 293 cells are members or modulators of the ras-rho/GTPase family of proteins and of the cAMP signaling pathway. The analysis of the activity of a panel of reporter vectors monitoring different transactivation pathways further corroborates the idea that ras-rho/GTPase and cAMP response element binding proteins play a predominant role in orchestrating the cell changes subsequent to SMC3 overexpression. In particular RhoB and CRE-BPa, two major modulators of cellular transformation and response to genotoxic stress and whose level is significantly increased following SMC3 overexpression, may act as important mediators of SMC3 activity at cellular level.
Results and Discussion
A microarray analysis of the genome-wide effect of SMC3 overexpression identifies candidate genes mediating SMC3 tumorigenicity
Significantly regulated genes in SMC3 overexpressing 293 cells. Human expressed sequence tags and genes with no known function are not included. Complete results of the array have been submitted to the EBI microarray database.
GPC3, glypican 3
lectin, mannose-binding, 1 like
HAS2, hyaluronan synthase 2
LAMB3, lamin, beta 3
Transporter and ion channels
ATP6V1B1, ATPase, H+ transporting V1 subunit B, kidney isoform
KCTD12, potassium channel tetramerisation domain 12
SLC7A10, solute carrier family 7 member 10, neutral amino acid transporter
MIG12, MID1 interacting G12-like protein
CKB, creatine kinase, B chain
ECHDC3, enoyl Coenzyme A hydratase domain containing 3
ALDH1A3, aldehyde dehydrogenase 1 family, member A3
MOXD1, monooxygenase, DBH-like 1
TXNRD3, thioredoxin reductase 3
Growth factors and receptors
LTBP2, latent transforming growth factor beta binding protein 2
GABRE, gamma-aminobutyric acid A receptor, epsilon
DLL1, delta-like 1, notch ligand
GDF9, growth differentiation factor 9
IGFBP7, insulin-like growth factor binding protein 7
GNRH1, gonadotropin-releasing hormone 1
IRAK1, interleukin-1 receptor associated kinase 1
EGFL3, EGF-like-domain, multiple 3
IL22RA1, interleukin 22 receptor, alpha 1
OR4D1, olfactory receptor, family 4, subfamily D, member 1
GRB7, growth factor receptor-bound protein 7
ARHGEF4, Rho guanine nucleotide exchange factor 4
RGS14, regulator of G-protein signaling 14
RPS6KA5, ribosomal protein S6 kinase, polypeptide 5
PAK6, p21-activated kinase 6
RIN3, Ras and Rab interactor 3
RHOB, ras homolog gene family, member B
SCFD1, sec1 family domain containing 1
LRRK1, leucine-rich repeat kinase 1
RAB40B, GTP-binding protein 40B
PDE6B, phosphodiesterase 6B, cGMP-specific
MCF2L, MCF.2 cell line derived transforming sequence
S100A8, S 100 calcium binding protein A8, calgranulin A
ADCY2, adenylate cyclase 2
MEOX2, mesenchyme homeobox 2, growth arrest specific homeobox
IRF4, interferon regulatory factor 4
KHDRBS3, KH domain, RNA binding, signal transduction associated 3
PPARA, peroxisome proliferative activated receptor, alpha
CRE-BPa, cAMP responsive element binding protein, ATF2-like
NEK9, never-in-mitosis-gene a-related kinase 9
CREM, cAMP responsive element modulator
NCOA6, nuclear receptor coactivator 6
TXB21, T-box 21
POU3F1, POU domain, class 3, octamer-binding TF6
NKX6-1, NK6 transcriptional factor related, locus 1, homeobox 6A
CTNNA1, catenin alpha 1
BCL6B, B-cell CLL/lymphoma 6, member B zinc finger protein
ZNF236, zinc finger protein 236
DNA repair, gene transcription
MHL3, mutL homolog 3
DNMT2, DNA cytosine-5-methyltransferase 2
ADPRTL2, ADP-ribosyltransferase 2
SNRPN, small nuclear ribonuclear polypeptide N
MYOM2, myomesin 2
NOPE, neighbor of Punc E11
C5, complement component 5
COCH, coagulation factor C homolog, cochlin
VMD2L1, vitelliform macular dystrophy 2-like 1
HPN, hepsin serine protease
LPHN1, latrophilin 1
PRND, prion protein 2
HYPM, huntingtin interacting protein M
SMC3 acts as an oncogene in human cells and activates the expression of a set of early-response genes
Further analysis of SMC3-transfected 293 cells and the parent cells provided evidence that in addition to RhoB also CRE-BPa, RGS14 and ARHGEF4 are part of the set of genes activated following transient elevation of SMC3. RGS14 is a target of the p53 tumor suppressor and its overexpression inhibits both Gi- and Gq-coupled growth factor receptor mediated activation of the mitogen-activated protein kinase signaling pathway in mammalian cells . Because p53 transactivation pathway is activated following SMC3 overexpression (see below) we postulate that RGS14 may be involved in counteracting the SMC3 mitogenic activity. ARHGEF4 (also known as Asef) is a Rac-specific guanine nucleotide exchange factor that is activated following binding to APC [37, 38]. APC-ARHGEF4 complex plays an important role in the regulation of the actin cytoskeleton, cell morphology and migration and affects E-cadherin-mediated cell-cell adhesion. Agents such as ARHGEF4 may thus act as mediators of the effect of SMC3 overexpression on cell growth and morphogenesis.
SMC3 overexpression specifically activates the SRE and CRE transactivation pathways
To further examine the cellular response to SMC3 overexpression, the activation level of a number of transactivation pathways was investigated. We reason that selective changes in transcriptional activity could further sort out the key players mediating the SMC3 biological activity. For this purpose 293 cells were transfected with SMC3 expression vector or pcDNA3.1 vector alone (control group) together with a series of luciferase reporters whose expression is driven by target-specific cis-acting elements present in multiple copies in the vector promoter. The serum responsive element (SRE) is a known target of RhoB as well as of other ras-rho/GTPase . Consistent with the idea that elevation of RhoB is functionally significant a 2.8-fold elevation of SRE-transactivation activity was detected in SMC3-overexpressing cells (fig. 5). On the contrary AP1-dependent gene transactivation which is mediated by c-jun, c-fos and ATF2 homo or heterodimers, was significantly suppressed. Given that RhoB, but not other ras-rho/GTPase, suppresses AP1-mediated gene transactivation  this result support the idea that following SMC3 elevation, RhoB plays a central role in mediating gene transactivation. Consistent with a major role of the CREB in the SMC3 mediated cell events, the response of the CRE-dependent reporter was greatly increased (2.6-fold) in cells stably overexpressing SMC3. This cAMP responsive element is a bona-fide target for the CRE-BPa transcriptional factor whose expression is significantly increased following SMC3 elevation (Table I, fig. 4a). CRE-BPa is a member of the ATF2 cAMP-binding proteins family that is activated by a variety of kinases including protein kinase A, JNK/SAPK, p38-MAPK, AKT, and calcium-calmodulin-dependent kinases and is involved in tumorigenesis of endocrine tissues and different forms of leukemia . To the activation of the CREB transactivation pathway may also contribute the downregulation of CREM, a gene encoding several spliced products some of which are CRE-transactivation repressors . The glucocorticoid, TGFβ-activin and of NF-kB transactivation pathways are potential target of RhoB or CRE-BPa, and their activity was also tested [26, 39, 40]. NFkB-mediated transactivation which has been reported to be downregulated by RhoB in NIH3T3 cells  was not significantly affected in our experiments suggesting that this regulatory mechanism is either cell context-specific or that SMC3 elicits other changes that counterbalance the RhoB-dependent NF-kB loss of activity. Likewise gene transactivation from the GRE and TARE cis-acting elements was not affected. Taken together the results are consistent with the idea that following SMC3 elevation, SRE- and CRE- mediated gene transactivation is specifically engaged.
The results presented provide a molecular signature of the changes that occurs in epithelial cells following SMC3 overexpression. In particular we document a selective effect of the ectopic expression of SMC3 on a set of early-response genes such as RhoB and CRE-BPa and on the related transcriptional pathways SRE and CRE that play key roles in tumorigenesis. SMC3 acts as an oncogene in human cells and we show that RhoB and CRE-BPa are part of a set of genes activated following transient SMC3 transfection. These findings provide important initial information on the chain of events occurring following SMC3 overexpression that will allow in future studies to focus on the underlying mechanism of the association between SMC3 deregulation and specific oncogenic pathways.
Establishment of cell lines stably overexpressing SMC3
293 cells were grown to 70% confluence in 10 cm plates and transfected with 3 μg of pcDNA3.1 expression vector harboring the entire human SMC3 coding sequence (SMC3-pcDNA3.1)  and using Lipofectamine as transfecting agent. To generate a control cell line a second batch of cells was transfected instead with the empty pcDNA3.1 vector. After 48 h stably transfected cells were selected in medium containing 500 μg/ml of G418. Clones of the surviving cells were expanded and SMC3 expression examined by semi-quantitative RT-PCR. For this purpose, 1 μg of cell RNA was reverse transcribed with Sensiscript (Qiagen) reverse transcriptase priming with oligo-dT. An aliquot of the RT reaction product was amplified using ExTaq (Takara) DNA polymerase and SMC3-specific primers of sequence: 5'-GAGTAGAAGAACTGGACAGA-3' and 5'-GATTGTACCTCAGTTTGCTG-3'. To ensure that the amplification reaction had not reached saturation, DNA production was monitored after 25 and 30 cycles by analysis on 1% agarose and by staining with ethidium bromide. Gels were photographed, the picture scanned and the band intensity quantified by densitometry with an image scanner. SMC3 protein expression was assessed by Western immunoblotting. Briefly, cell lysates in 150 mM NaCl 1% Nonidet P40 0.5% Na-deoxycholate 50 mM Tris-HCl pH 7.4 were electrophoresed on 8% SDS-PAGE slab gel and the separated proteins transferred onto a nitrocellulose filter. Immunoblotting was performed with goat anti-human SMC3 (1:1,000) antibody (Santa Cruz Biotech) at 25°C for 1 h, followed by incubation in anti-goat IgG horseradish peroxidase conjugated (1:10,000) secondary antibody. Immunocomplexes were identified using an enhanced chemiluminescence (ECL) kit (Pierce) followed by autoradiograph. Three SMC3 overexpressing clones and three control clones were selected for the gene expression profiling.
Microarrays and data analysis
For the target preparation, 5 μg of human untransfected and transfected 293 cell line total RNA were reverse transcribed with SuperscriptT-II/RNaseH- priming with T7-(dT)24 oligonucleotides and the second-strand cDNA synthesized using E. coli DNA polymerase I . Biotinylated cRNA was generated using T7 RNA polymerase and Biotin 11-UTP. Ten μg of purified unfragmented target cRNA was used for hybridization of each KCC/TJU human 18.5 K Expression Bioarray (Compugen Human Oligo Set 1.0) chip containing 18,861 oligos (65-mer) corresponding to 17,260 unique clusters and 18 bacterial control probes. The microarrays were hybridized, washed, and processed using a direct detection method of the biotin-containing transcripts by a Streptavidin-Alexa647 conjugate. Processed slides were scanned using a Perkin Elmer ScanArray XL5K scanner and the spot intensity quantitated using the ScanArray Scanning and QuantArray programs (PerkinElmer) . The chips analyzed presented consistent staining over the entire microarray and had appropriate data distribution. Spots with raw intensity value comprised within one SD from the average background value were excluded from the analysis. The remaining values – about 90% of the whole arrayed genes, were normalized by dividing each spot's intensity (after background subtraction) by the median signal intensity of all test probes. Genes that displayed inter-array variability exceeding one SD unit were furthermore excluded from the final statistical analysis. The data and protocols have been submitted to the EBI ArrayExpress database (samples 171479SUB800 through 171484SUB800).
Statistical analysis of the microarray data
We input the log2 of the gene expression measurements from three sets of microarray experiments each including a control and an SMC3 overexpressing cell sample. Through a series of permutation the program computes a statistic score di for each gene i measuring the strength of the relationship between gene expression and the response variable and creates a profile of observed versus expected values. The values which lie outside a user-defined region that can be adjusted to achieve an optimum of positive vs. false positive values, are considered significantly related to the response and thus regarded as significantly regulated genes (see fig. 2). Since each experiment consisted of the data from two independent chips hybridized with the control and the SMC3-overexpressing cells cDNA, an unpaired two-class analysis was carried out to discover significant changes in gene regulation compared to the control cell line.
Gene transactivation activity assay
Reporter vectors harboring multiple copies of the consensus sequences for the AP1, cAMP (CRE), serum (SRE), p53, TGFβ (TARE), NF-kB and glucocorticoid (GRE) response elements, were obtained from Stratagene or Clontech. Cells cultures at 70% confluence in 12 wells plates were used in all the experiments. The transfection mix contained 10 ng/ml of phRL-SV40 plasmid to monitor the transfection efficiency, 100 ng/ml of the designed plasmid, and 1 μg/ml of SMC3-pcDNA3.1 expression vector. Plasmids were mixed in medium 199 followed by the addition of 10 mg/mg DNA of Tfx-50 (Promega) transfection agent according to the manufacturer directions and finally added to the cultures. After 1 h incubation at 37°C in humidified incubator the cells were supplemented with 2 ml of growth medium and the luciferase activity assayed 24 h later using a Promega dual-luciferase kit. All experiments were carried out with triplicate samples. The statistical difference between groups of data was analyzed by Student's t-test.
SMC3 transient transfection and gene transcript level analysis
Cells (either 293 or NIH3T3) at 70% confluence were transiently transfected with 1 μg/ml of SMC3-pcDNA3.1 expression vector using Lipofectamine as transfection agent. Control cells were transfected instead with 1 μg/ml of pcDNA3.1 empty vector. After 48 h, the cells were washed in PBS and the total RNA extracted with TriReagent. Gene transcripts were amplified by RT-PCR and the products quantified by gel electrophoresis. The primers used had the following sequence: RhoB: 5'-CCTGCTGATCGTGTTCAGTAA-3' and 5'-TCATAGCACCTTGCAGCAGTT-3'; CRE-BPa: 5'-ATGATTTATGAGGAATCCAAGAT G-3' and 5'-TTAAAGAATCGGATTCAGGTCTGT-3'; RGS14: 5'-CTGGTGGGCAATGAACAGAAGGCC-3' and 5'-GGGCTGAGTCGGTGGTGGAGTTCA-3'; ARHGEF4: 5'-AGCCTCAAGCCAAAAGCCAGCAGC-3' and 5'-CTCACTTGCTGGCAGAGGAAGGCCA-3'; G3PDH: 5'-TGAAGGTCGGAGTCAACGGATTTGGT-3' and 5'-CATGTGGGCCATGAGGTCCACCAC-3'. RhoB protein level in cells was examined by Western immunoblotting as described previously using a rabbit polyclonal antibody (1:1,000) from Bethyl.
Cell proliferation assay
Cell proliferation was examined using a CellTiter 96 (Promega) assay kit according to the manufacturer's instructions. Cells growing in log phase were trypsinized and seeded in 96-well plates (2,500 cells/well in a final volume of 150 μl) in replicates of 4 and incubated at 37°C in 5% CO2 and 95% air in DMEM medium containing 1.5% FCS. At 24 h, 72 h or 96 h, 20 μl of the kit dye solution was added to each well and the plates incubated at 37°C for an additional 1 h. The absorbance of the formazan product generated was measured at 490 nm using a 96-well Dynatech MR600 plate reader.
Cell overgrown assay
In order to examine the ability of cells to form foci of transformation, cells were seeded at 30% confluence in 35 mm plates and cultured in DMEM supplemented with 1.5% medium. After 7 days the medium was removed, the cell washed with PBS, and fixed with 70% ethanol on ice. Foci of cell aggregation were evidenced by staining with 0.1% methylene-blue dissolved in water followed by 3 washing in water to decrease background staining.
Anchorage-independent cell growth in soft agar
Anchorage-independent colony formation of cells was assayed as described . Briefly, cells growing in log-phase were trypsinized and resuspended at 37°C in 0.2% agarose in DMEM containing 10% fetal bovine serum and plated on top of solidified agarose (0.4%) dissolved in the same medium in 35 mm dishes. After 3 weeks of culture at 37°C in CO2 humidified incubator, the number of cell aggregates over that of single cells and the number of colonies of diameter >100 μm found in randomly selected areas of 9 mm2, was recorded.
This work was supported by grant RO1-CA82290 to GG.
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.