Transient Strong Reduction of PTEN Expression by Specific RNAi Induces Loss of Adhesion of the Cells

Transient Strong Reduction of PTEN Expression by Specific RNAi Induces Loss of Adhesion of the Cells
of 9
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Related Documents
  Transient strong reduction of PTEN expression by specificRNAi induces loss of adhesion of the cells Setsuko Mise-Omata a , Yuichi Obata a , Shigeru Iwase b , Nathan Mise c , Takahiro S. Doi a,* a Technology and Development Team for BioSignal Program, Subteam for BioSignal Integration, RIKEN Bioresource Center,RIKEN Tsukuba Institute, 3-1-1 Koyadai Tsukuba, Ibaraki 305-0074, Japan b Bioresource Information Division, RIKEN Bioresource Center, RIKEN Tsukuba Institute, 3-1-1 Koyadai Tsukuba, Ibaraki 305-0074, Japan c Technology and Development Team for Mammalian Cellular Dynamics, RIKEN Bioresource Center, RIKEN Tsukuba Institute,3-1-1 Koyadai Tsukuba, Ibaraki 305-0074, Japan Received 6 January 2005Available online 25 January 2005 Abstract The tumor suppressor gene  pten  encodes a lipid phosphatase that dephosphorylates D3 of phosphatidylinositol(3,4,5)trisphos-phate, producing phosphatidylinositol(4,5)bisphosphate. Although PTEN has been implicated in cell adhesion and migration,the underlying molecular mechanism is unknown. To investigate the role of PTEN in cell adhesion, we designed three differentsiRNAs (siRNA PTEN-a, siRNA PTEN-b, and siRNA PTEN-c) and transfected into 293T cells. Two days later, only the cellstransfected with siRNA PTEN-b became round and detached from the culture dishes, whereas cells transfected with a controlsiRNA against GFP or the two other siRNAs against PTEN did not. Evaluation of the RNAi effect revealed that siRNAPTEN-b inhibited >95% of PTEN expression, the most effective among the three siRNAs. To check for non-specific effects suchas interferon response and inhibition of off-target genes, we then used quantitative PCR analysis and DNA microarray analysis.None was detected, indicating that the RNAi system was highly specific. Immunofluorescence studies using PTEN-knockdownHeLa cells revealed that the loss of adhesion was accompanied by a reduction in the number of focal adhesion plaques and disor-ganization of the actin cytoskeleton. Transient and near-complete loss of PTEN expression induces loss of adhesion of the cells.   2005 Elsevier Inc. All rights reserved. Keywords:  PTEN; Cell adhesion; RNA interference; Specificity of RNAi; Interferon response; Actin cytoskeleton Phosphatase and tensin homolog deleted on chromo-some10(  pten )isatumorsuppressorgeneencodingalipidphosphatase, which converts phosphatidylinosi-tol(3,4,5)trisphosphate (PtdIns(3,4,5)P 3 ) to phosphati-dylinositol(4,5)bisphosphate (PtdIns(4,5)P 2 ), acting inopposition to phosphatidylinositol 3 0 kinase (PI3K) [1].Although PTEN has also been shown to possess tyrosinephosphatase activity and to dephosphorylate focal adhe-sionkinase(FAK)andShcinvitro[2,3],itsinvivoprotein phosphatase activity remains to be investigated [4]. Incontrast, PTEN  s lipid phosphatase activity has beendemonstrated both in vivo and in vitro. The importanceof   pten intumor suppressoractivityhasbeenwelldefinedthrough the highly frequent mutation of   pten  in a widevariety of human cancers, including glioblastoma, mela-noma, and carcinoma [5,6]. Germline mutations in the  pten  locus have been associated with Cowden syndrome,in which patients develop hyperplastic lesions in multipleorgans [6]. In addition, the phenotype of   pten +/  micesupports the conclusion that  pten  functions as a tumorsuppressor gene [7]. Although homozygous disruptionof   pten resultsinearlyembryoniclethality,bothheterozy- 0006-291X/$ - see front matter    2005 Elsevier Inc. All rights reserved.doi:10.1016/j.bbrc.2005.01.066 * Corresponding author. Fax: +81 29 836 9029. E-mail address:  doi@brc.riken.jp (T.S. Doi). www.elsevier.com/locate/ybbrc Biochemical and Biophysical Research Communications 328 (2005) 1034–1042 BBRC  gous mice and chimaeric mice derived from  pten +/  embryonic stem (ES) cells display hyperplastic changesin the prostate, skin, and colon that are similar to thosein Cowden syndrome. Loss of PTEN results in the accu-mulationofPtdIns(3,4,5)P 3 andactivationofAKT/PBK[8–11]. As a serine/threonine kinase, AKT functions byphosphorylating key intermediate signaling molecules,invoking the properties of tumor cells, such as increasedmetabolism, cell growth, cell survival, and cell invasive-ness. Accumulation of PtdIns(3,4,5)P 3  also activatesother signaling molecules including the phosphatidylino-sitol-dependent kinases (PDKs), S6 kinase, mTOR, andRho family small GTPases [12,13].PTEN controls cell adhesion and migration [4], andnegatively regulates cell migration in culture systems.Overexpression of PTEN inhibits cell migration andspreading, whereas antisense PTEN enhances thesefunctions [2].  pten  /  embryonic fibroblasts show en-hanced motility and exhibit increased cortical actinpolymerization [12]. However, there have been contra-dictory reports regarding the underlying molecularmechanism. One report showed that the overexpressionand antisense system is accompanied by changes in theactivity and phosphorylation of FAK [2], whereas theother reported no difference in the phosphorylation of FAK between  pten  /  fibroblasts and wild-type cells; in-stead, enhanced activity of Rac1 and Cdc42 was de-tected [12]. In contrast to  pten  /  fibroblasts, in an invivo system using the Cre-loxP system, cerebellar gran-ule cells failed to migrate in  pten -disrupted brains[14,15]. The effect of PTEN in cell migration may bemore complicated in tissues with multiple cell types. In Dictyostelium , PTEN is localized opposite side to theleading edge of the cell during chemotaxis, and deletionof PTEN leads to defects in chemotaxis and prolongedactin polymerization [16,17]. Thus, PTEN regulatesorganization of the actin cytoskeleton.RNAi has recently been applied to mammalian cells.It is a powerful tool for investigating the functions of as-yet uncharacterized gene products and those whoseknockout mice show embryonic lethality. Some groupshave confirmed the specificity of RNAi in mammaliancells, in light of microarray analyses [18,19]. However,other groups have pointed out the risk of side effects,in that RNAi in mammalian cells is not target-specific.Bridge et al. [20] observed that a short hairpin RNAdelivered by a lentiviral vector, but not synthetic siRNA,upregulates interferon-inducible genes, including2 0 -5 0 -oligoadenylate synthetase-1 (OAS1). Sledz et al.[21] using microarray analysis and biochemical ap-proaches, detected induction of the interferon responseafter transfection with synthetic siRNA. Jackson et al.[22] found inhibition of    off-target genes   —non-targetedgenes with sequences homologous to those of the de-signed siRNAs. They suggested that an siRNA might in-hibit an off-target gene that has a sequence similar tothose of 14 or 15 nucleotides in the central region of the siRNA or to those of 9 in its 3 0 region. It remainsunknown why there have been contradictory results onthe specificity of RNAi in mammalian cells.To define the cellular function of PTEN, we per-formed RNAi experiments against PTEN. In the courseof our study, we found that the cells transfected with oneof the three designed siRNAs became round and de-tached from culture dishes. This drastic morphologicalchange may be caused by loss of PTEN expression orit may be a side effect of RNAi. To exclude the latterpossibility, we carried out quantification of OAS1mRNA, detection of apoptosis, and gene expressionprofiling. These analyses indicated that our RNAi sys-tem was highly specific. Moreover, we demonstratedthat the siRNA that induced detachment of cells inhib-ited >95% of PTEN expression, the most effectiveamong the three siRNAs. These results suggested thatnear-total loss of PTEN expression results in the lossof adhesion of cells to the surfaces of culture dishes.In light of our data and those of others, we discuss a no-vel role of PTEN in the organization of the cytoskeletalarchitecture. Materials and methods Reagents . siRNAs were synthesized by Eurogentec SA. (Seraing,Belgium), and purified and annealed by Nippon Gene (Toyama,Japan). The sequences of the siRNAs were: siRNA GFP, GCU GACCCU GAA GUU CAU CdTdT; siRNA PTEN-a, CAC CAC AGCUAG AAC UUA UdTdT; siRNA PTEN-b, CCA GUC AGA GGCGCU AUG UdTdT; and siRNA PTEN-c, AGU GGC GGA ACUUGC AAU CdTdT. We purchased the following antibodies fromcommercial sources: rabbit polyclonal anti-PTEN antibody (Cell Sig-naling Technology, Beverly, MA); anti-RelA antibody (Santa CruzBiotechnology, Santa Cruz, CA); monoclonal anti-FAK antibody andanti-paxillin antibody (BD Bioscience, Palo Alto, CA); and rabbitpolyclonal anti-FAK[pY 397 ] antibody (Biosource, Camarillo, CA).Poly(I)–poly(C)(pIC)(Sigma,St.Louis,MO)wasdissolvedin100 mMpotassium acetate, 30 mM Hepes–KOH (pH7.4), and 2 mM magne-sium acetate, incubated at 50   C for 10 min, and then slowly cooled. Cell culture and transfection . HeLa cells were obtained from theRIKEN Bioresource Center Cell Engineering Division. Cells weremaintained in Dulbecco  s modified Eagle  s medium supplemented with10% fetal bovine serum, penicillin, and streptomycin. 293T cells weretransfected with various amounts of siRNAs by using the calciumphosphate procedure. HeLa cells were transfected by using Oligofec-tamine (Invitrogen, Carlsbad, CA), in accordance with the manufac-turer  s instructions. Detection of apoptotic cells . Three days after transfection withsiRNA, 293T cells were stained with annexin V-FITC (MolecularProbe, Eugene, OR) and propidium iodide (PI). Fluorescence inten-sity was measured by a FACS Calibur (BD Bioscience) flowcytometer. RNA analyses . Total RNA was recovered 24 h after transfectionwith siRNA by using Trizol (Invitrogen). Transcription of OAS1 wasdetected by real-time PCR, using the TaqMan Assay-on-DemandSystem (AP Biosystems, Foster City, CA). Microarray analyses werecarried out using the Oligo DNA microarray, Human 1A (G4110A,Agilent Technology, Palo Alto, CA), which carries 17086 clones. S. Mise-Omata et al. / Biochemical and Biophysical Research Communications 328 (2005) 1034–1042  1035  We used 500 ng total RNA from each sample as a template for theamplification and labeled the products with Cy-3 or Cy-5 (Perkin-El-mer, Boston, MA) by using a Low RNA Input Fluorescent LinearAmplification kit (Agilent Technology). Pairs of Cy5/Cy3-labeledcDNA samples were fragmented, combined, and hybridized by usingan In situ Hybridization Kit Plus (Agilent Technology). Afterhybridization, microarrays were washed and scanned using a confocallaser scanner (Agilent Technology). Fluorescence intensities on scan-ned images were quantified, corrected for background noise, andnormalized. The results were analyzed with the Rosetta LuminatorGene Expression Data Analysis System (Agilent Technology). Nucle-otide matches between sequences were identified using the DebianGNU/Linux Med-Bio package. Protein analyses . Cells were lysed with 1% NP-40 lysis buffer(150 mM NaCl, 20 mM Tris–HCl, (pH 7.5), 1 mM MgCl 2 , 1 mMEGTA, 50 mM NaF, 1 mM Na 3 VO 4 , 10 mM Na 4 P 2 O 7 , 10 mM leu-peptin, 10 mM aprotinin, and 10 mM AEBSF). After quantification of protein concentration by using a Protein Assay (Bio-Rad Laborato-ries, Hercules, CA), 50  l g of cell lysate derived from 293T cells or30  l g of HeLa cell lysate was analyzed using 10% SDS–PAGE.Immunoblots were revealed by the ECL Plus Western blottingDetection System (Amersham Biosciences, Uppsala, Sweden) orSuperSignal West Femto Maximum Sensitivity Substrate (Pierce,Rockford, IL), and the intensities of the signals were quantified byusing the LAS-1000 Luminescent Image Analysis System (Fujifilm,Tokyo, Japan). The percentage PTEN expression was calculated asfollows. Each PTEN signal was normalized against that of RelA, aninternal control for the total amount of protein. Then the normalizedvalue for each transfection was divided by that of the siRNA GFPtransfection. Immunofluorescence . At 24 h after transfection with siRNAs, thecells were plated on coverslips coated with fibronectin. At 67 h aftertransfection, the cells were fixed for 10 min at room temperature with3.7% paraformaldehyde in PBS containing 30 mM sucrose. Afterneutralization of aldehyde groups with 50 mM NH 4 Cl in PBS for10 min, the coverslips were saturated for 30 min with 3% BSA in PBScontaining 0.2% Triton X-100. The coverslips were incubated with theoptimal concentration of the appropriate primary antibody for 1 h atroom temperatures. After serial washing, they were incubated with thesecondary antibody diluted at 1:100 (anti-mouse IgG1-FITC, South-ernBiotech Associates, Birmingham, AL) or 1:400 (anti-rabbit IgG-Alexa Fluor 594, Molecular Probes) for 1 h at room temperatures.After washing, the coverslips were mounted with Vectashield con-taining DAPI (Vector Lab., Burlingame, CA). Results siRNA directed to PTEN induced strong inhibition of PTEN expression and detachment of cells from culturedishes To define the cellular function of PTEN, we designedthree siRNAs (referred as siRNA PTEN-a, siRNAPTEN-b, and siRNA PTEN-c) by the standard method[23]. Two days after transfection of the siRNAs into293Tcells,cellstransfectedwith siRNA PTEN-bbecameround and detached from the culture dishes (Fig. 1A).Most cells had detached by 3 days after transfection.The cells transfected with siRNA PTEN-a or siRNAPTEN-c did not exhibit this morphological change(Fig.1).ToassesstheeffectofRNAi,weperformedWes-tern blot using cell lysates obtained 2 and 3 days aftertransfection.Near-completelossofPTENexpressionoc-curred in the cells transfected with siRNA PTEN-b,whereas the cells transfected with siRNA PTEN-a orsiRNA PTEN-c expressed reduced but still markedamounts of PTEN protein (Fig. 1B). To compare theRNAi effect more precisely, we transfected serial dilu-tions of each siRNA and detected the level of PTENexpression by Western blot. Quantification of PTENexpression revealed the most effective inhibition insiRNA PTEN-b transfected cells: 97.6% ± 0.6% inhibi-tion at 45 pmol siRNA and 94.1% ± 0.8% at 11.3 pmol(Fig. 2). Cells transfected with  = 11.3 pmol siRNAPTEN-b had the morphology shown in Fig. 1A.Although even 5.6 and 2.8 pmol siRNA PTEN-b in-duced >90% inhibition, transfection with 5.6 pmolsiRNA PTEN-b formed mixed populations: some cellsdetached, and some exhibited normal morphology, andmost cells appeared normally at 2.8 pmol (data notshown). In contrast, even with the largest amount of siRNA, the maximum inhibition of PTEN in cells trans- Fig. 1. siRNA PTEN-b induces strong inhibition of PTEN expression and detachment of cells from culture dishes. (A) 293T cells were transfectedwith 45 pmol siRNA PTEN-a, siRNA PTEN-b, or siRNA PTEN-c or 0.6  l g pIC. Morphologies of representative cells 2 days (top row) and 3 days(bottom row) after transfection are shown. siRNA PTEN-b-transfected cells became round and detached from culture dishes. Bar = 20  l m. (B)PTEN expression was detected by Western blot in non-transfected cells (lane 1), and in cells transfected with siRNA GFP (lane 2), siRNA PTEN-a(lane 3), siRNA PTEN-b (lane 4), siRNA PTEN-c (lane 5), and pIC (lane 6). PTEN expression was reduced in the cells transfected with siRNAPTEN, especially with siRNA PTEN-b, whereas RelA expression was not affected.1036  S. Mise-Omata et al. / Biochemical and Biophysical Research Communications 328 (2005) 1034–1042  fected with siRNA PTEN-a or siRNA PTEN-c was onlyabout 80%, and no morphological changes wereobserved. siRNA PTEN-b does not induce interferon response Although the morphological change induced bysiRNA PTEN-b may be due to loss of PTEN expres-sion, siRNA PTEN-b alternatively could induce theinterferon response, resulting in apoptosis and detach-ment [20,21]. To examine the possibility of the interferonresponse, we checked for cell death. As just described,by 3 days after transfection, cells transfected withsiRNA PTEN-b detached from the culture dishes, butthose transfected with siRNA PTEN-a or siRNAPTEN-c did not (Fig. 3A). Using the same cell popula-tions, we carried out flow cytometry analyses using an-nexin V-FITC and PI. Even though most of thesiRNA PTEN-b-transfected cells had detached fromthe culture dishes, >70% of the cells did not stain withannexin V or PI (Fig. 3B). Furthermore, the proportionsof cells negative for staining did not differ significantlyamong the three siRNAs. This result indicates that thedetachment of siRNA PTEN-b-transfected cells fromthe culture dishes was not caused by cell death.We examined the upregulation of OAS1, a typicalinterferon-inducible gene product, by using pIC as a po-sitive control. Transfection with pIC did not induce thedrastic morphological change that was seen in siRNAPTEN-b-transfected cells, and only a few cells were de-tached from the culture dishes just after transfection(Fig. 1A). We then performed quantitative PCR to mea-sure the amount of OAS1 mRNA at 24 h after transfec-tion. OAS1 mRNA was not detected in the cellstransfected with any of the three siRNAs or in thenon-transfected cells after 50 cycles of amplification,even when 100 ng of total RNA was used as a template(data not shown). In contrast, OSA1 mRNA was upreg-ulated in pIC-transfected cells, and could be detected at35 cycles of   C  t  value, and mRNA of the control gene( b -actin) was detected in all samples at just 20–23 cyclesof   C  t  values. This shows that siRNAs do not upregulateOAS1 expression, but pIC does.To exclude the possibility of the interferon response,we compared the gene expression profiles between non-transfected cells and siRNA PTEN-b-transfected cellsby using the oligo DNA microarray (Fig. 4A). Among17086 genes, 19 interferon-inducible genes are listed inFig. 4B. The fold-change values of all 19 genes wereapproximately 1.0, indicating lack of upregulation of interferon-inducible genes in siRNA PTEN-b-transfec-ted cells. Detachment of cells from culture dishes is not due toinhibition of ‘off-target’ genes To investigate the possibility that the phenotype in-duced by siRNA PTEN-b may be caused by inhibitionof off-target genes as a result of cross-reaction of thesiRNA with non-target genes, we compared the geneexpression profiles of siRNA PTEN-b- and siRNAPTEN-c-transfected cells (Fig. 4C). We compared thesequence of siRNA PTEN-b in both sense and antisensedirections with those of the 100 genes showing the most Fig. 2. Only siRNA PTEN-b induces near-complete loss of PTEN expression. (A) Serial dilutions of siRNA were transfected into 293T cells, and theeffect of RNAi was examined by Western blot 2 days after transfection. (B) Western blot signals were quantified by using the LAS-1000 LuminescentImage Analysis System, and the percentage inhibition was calculated relative to the signal produced by siRNA GFP transfection. Data are presentedas means ± SD of three independent experiments. siRNA PTEN-b was the most effective among the three siRNAs. S. Mise-Omata et al. / Biochemical and Biophysical Research Communications 328 (2005) 1034–1042  1037  reduced expression in siRNA PTEN-b-transfected cells.We found that only heme oxygenase-1 (HMOX1) hadnoteworthy identity to siRNA PTEN-b (a 9-nucleotidematch to its 3 0 end, Fig. 4D). According to the annota-tion, HMOX1 is not likely to be responsible for detach-ment of cells from the culture dishes [24]. We identifiedno gene with a 14- or 15-nucleotide match in the centralregion of siRNA PTEN-b. Although it is difficult todetermine definitively whether inhibition of off-targetgenes occurs, the morphological change we observed isnot likely to be caused by the cross-reaction of thesiRNA with off-target genes.These results and the correlation between the effi-ciency of RNAi and cell detachment (Fig. 2) led us tothe conclusion that transient loss of PTEN results in celldetachment. PTEN-knockdown cells show reduced focal adhesionand alteration of actin cytoskeletal organization To assess the formation of focal adhesion, we carriedout RNAi experiments using HeLa cells. As with the293T cells, siRNA PTEN-b induced cell detachment(Fig. 5A) and the strongest inhibition of PTEN expres-sion (Fig. 5B, lane 4) 3 days after transfection into HeLacells. On the other hand, cells transfected with siRNAPTEN-a or siRNA PTEN-c did not exhibit morpholog-ical change (data not shown) and had weaker inhibitionof PTEN expression than those with siRNA PTEN-b(Fig. 5B, lanes 3 and 5). Focal adhesion and actin cyto-skeleton were observed by immunofluorescence. PTEN-knockdown cells showed reduced cell spreading onfibronectin-coated coverslips and demonstrated a re-duced number of focal adhesion plaques when stainedwith anti-FAK or anti-paxillin antibody, as comparedwith the negative control, siRNA GFP-transfected cells(Fig. 5C). The FAK localized in the focal adhesionplaques was phosphorylated because the overlay of anti-FAK and anti-FAK[pY 397 ] antibodies became yel-lowish. The amount of phosphorylated FAK at the cellperiphery was reduced in PTEN-knockdown cells, butthey showed strong accumulation of phosphorylatedFAK in the nucleus. PTEN knockdown also dramati-cally altered the actin cytoskeletal organization. In con-trast to siRNA GFP-transfected cells, in which the actinfilaments formed stress fibers, PTEN-knockdown cellshad reduced stress fibers and formed intense corticalactin (Fig. 5D). Double staining of F-actin with phalloi-din and paxillin with anti-paxillin antibody revealedlinkages between focal adhesion plaques and bundlesof actin filaments in control cells, but such linkage wasnot evident in the PTEN knockdown cells (Fig. 5D). Discussion We demonstrated that transient loss of PTEN expres-sion by RNAi induced a drastic morphological change, Fig. 3. Detached cells are not apoptotic. (A) Representative morphologies of 293T cells 3 days after transfection with siRNA PTEN-a, siRNAPTEN-b, or siRNA PTEN-c are shown. Most of the siRNA PTEN-b-transfected cells underwent a drastic morphological change. Bar = 20  l m. (B)Using the cell population indicated in (A), FACS analysis was carried out by PI and annexin V-FITC staining. Percentages of the cells in each squarefrom three different experiments are indicated.1038  S. Mise-Omata et al. / Biochemical and Biophysical Research Communications 328 (2005) 1034–1042
Similar documents
View more...
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks