Genetic reprogramming of tumor cells by zinc fingertranscription factorsPilar Blancafort*†, Emily I. Chen*‡, Beatriz Gonzalez*, Sharon Bergquist*, Andries Zijlstra§, Daniel Guthy¶,Arndt Brachat¶, Ruud H. Brakenhoff�, James P. Quigley§, Dirk Erdmann¶, and Carlos F. Barbas III*,**

*Department of Molecular Biology and The Skaggs Institute for Chemical Biology, and Departments of ‡Molecular and Experimental Medicine and§Cell Biology, The Scripps Research Institute, La Jolla, CA 92037; ¶Oncology Research, Novartis Institutes for Biomedical Research, Novartis Pharma AG,CH-4002 Basel, Switzerland; and �Section of Tumor Biology, Department of Otolaryngology�Head–Neck Surgery, VU University Medical Center,1081HV, Amsterdam, The Netherlands

Edited by Peter K. Vogt, The Scripps Research Institute, La Jolla, CA, and approved June 3, 2005 (received for review February 10, 2005)

Cancer arises by the accumulation of genetic alterations in DNAleading to aberrant gene transcription. Expression-profiling stud-ies have correlated genomewide expression signatures with ma-lignancy. However, functional analysis elucidating the contribu-tion and synergy of genes in specific cancer cell phenotypesremains a formidable obstacle. Herein, we describe an alternativegenetic approach for identification of genes involved in tumorprogression by using a library of zinc finger artificial transcriptionfactors (ATFs) and functional screening of tumor cells as a sourceof genetic plasticity and clonal selection. We isolated a six-zincfinger transcriptional activator (TF 20-VP, TF 20 containing the VP64activator domain) that acts to reprogram a drug-sensitive, poorlyinvasive, and nonmetastatic cell line into a cell line with a drug-resistant, highly invasive, and metastatic phenotype. Differentialexpression profiles of cells expressing TF 20-VP followed by func-tional studies, both in vitro and in animal models, revealed thatinvasion and metastasis requires coregulation of multiple targetgenes. Significantly, the E48 antigen, associated with poor metas-tasis-free survival in head and neck cancer, was identified as onespecific target of TF 20-VP. We have shown phenotypic modulationof tumor cell behavior by E48 expression, including enhanced cellmigration in vitro and tumor cell dissemination in vivo. This studydemonstrates the use of ATFs to identify the group of genes thatcooperate during tumor progression. By coregulating multipletargets, ATFs can be used as master genetic switches to reprogramand modulate complex neoplastic phenotypes.

drug resistance � metastasis � invasion � transcriptional regulation �RNA inteference

During different stages of a neoplastic disease, phenotypicdiversity and clonal selection of tumor cells generate cell

populations possessing traits that increase the potential for malig-nancy, such as increased mobility and invasiveness. Gene expressionprofiling of tumors has revealed that increased malignancy isassociated with changes in gene expression affecting multiple loci(1–3). Because of their unique ability to orchestrate and coregulatemultiple target genes, transcription factors (TFs) play a crucial rolein generating phenotypic plasticity associated with cancer progres-sion. TFs that play a role in lineage-specific differentiation, such asSTATs [signal of transduction and activator of transcription (4)],and those that affect morphogenesis and embryonic development,such as Twist, Snail, and SIP1, have recently been shown to beinvolved in invasiveness during tumor progression (5, 6).

As a consequence of their inherent potential for altering geneticcascades, artificial TFs (ATFs) may be used to modulate cancer cellphenotypes. TFs possess several distinctive features. First, by in-teracting specifically with endogenous regulatory sequences, TFscan mediate the simultaneous regulation of multiple genes neces-sary for the control of complex phenotypes. Second, TFs have theability either to up-regulate expression of target genes [when theDNA-binding domain (DBD) is linked to an activator of transcrip-tion, i.e., VP16, VP64, or p65] or to down-regulate target gene

expression (when the same DBD is linked to a repressor domainsuch as the KRAB domain) (7). ATFs with zinc finger (ZF) DBDsare particularly useful because a well characterized lexicon of ZFshas been optimized for specific recognition of virtually any tripletof DNA (8, 9). Furthermore, we have recombined the existing ZFDBDs to generate multimodular libraries made of three- and six-ZFbuilding blocks (10–12). When delivered into a tumor cell popu-lation, TF libraries provide phenotypic diversity on the order ofmillions of ATFs capable of ‘‘scanning’’ the tumor cell genome forfunctional, accessible regulatory sequences, each with the potentialto affect transcription of multiple genes involved in tumor progres-sion. A combination of differential DNA profiling and target sitesearch, based on the predicted specificity of ZF units, allowsselected ATFs to be used as genetic probes for the identification ofgenes and genetic interactions involved in malignancy.

In this article we describe the selection and characterization of asix-ZF ATF (TF 20-VP, TF 20 containing the VP64 activatordomain) selected from a six-ZF combinatorial activator librarycapable of inducing drug resistance, cytoskeleton remodeling,matrix-dependent cell migration, and tumor cell invasion in vitro.Furthermore, TF 20-VP enhanced the number of tumor metastasesin animal models. An analysis of the expression profiles of TF20-VP-transduced cells revealed gene expression signatures in-volved in cancer progression. Our data support the use of ATFs asgenetic switches to coregulate genes, discover new gene expressionmarkers of cancer progression, and modulate complex phenotypesassociated with malignancy.

Materials and MethodsZF Selection. TF 20-VP was selected from a six-ZF retroviralactivator library (pMX-6ZFlibrary-IRES-GFP) by treating 108

HeLa cells with 200 �M Taxol (Sigma) for 72 h. Surviving cells weremorphologically examined for the presence of epithelial-mesenchymal transition-like phenotypes. Retroviral DNA was re-covered from genomic DNA and then recloned in the retroviralvector as described (10).

Migration Assays. The cell migration assay was performed by usingtranswell plates (8-�m pore size) (Costar). The undersurface of themembrane was coated at 4°C overnight with 0.25 �g�ml of Laminin(Sigma L-2020) diluted in PBS and then blocked with 2% BSA. Theupper compartment was seeded with 2.5 � 105 transduced HeLacells per well in 100 �l of DMEM. FBS (2%) in DMEM was added

This paper was submitted directly (Track II) to the PNAS office.

Abbreviations: TF, transcription factor; TF 20-SKD, TF 20 containing the SKD repressordomain; TF 20-VP, TF 20 containing the VP64 activator domain; ATF, artificial TF; ZF, zincfinger; DBD, DNA-binding domain; STAT, signal transduction and activator of transcription;NOD, nonobese diabetic; SCID, severe combined immunodeficient; AGT, angiotensinogen;IL-13R�1, IL-13 receptor �1.

†Present address: Department of Pharmacology, University of North Carolina, Chapel Hill,NC 27599-7365.

**To whom correspondence should be addressed. E-mail: carlos@scripps.edu.

© 2005 by The National Academy of Sciences of the USA

11716–11721 � PNAS � August 16, 2005 � vol. 102 � no. 33 www.pnas.org�cgi�doi�10.1073�pnas.0501162102

in the lower chamber. Cells were allowed to migrate 22 h, andmigrated cells were stained with 0.25% Crystal violet (Sigma) in20% methanol (wt�vol). Each experiment used triplicate wells and,within each well, counting was done in four randomly selectedmicroscopic high-power fields (�100).

In Vitro Invasion Assays. HeLa cells transduced with differentretroviral vectors expressing TF 20-VP or TF 20-SKD (TF 20containing the SKD repressor domain) or retrovirus containingno ZF (control) were starved overnight, and then 1.25 � 105 cellsof each cell line were loaded in Matrigel invasion chambers (24wells, BD Biocoat Matrigel invasion chamber, BD Biosciences)according to the manufacturer’s instructions. Cells were allowedto invade for 24 h, and then invasive cells were fixed and countedwith an inverted microscope (Leica, McBain Instruments, Chats-worth, CA). The experiment was done two to three times witheach sample in triplicate. For cDNA expression analysis 500 ngof each transient expression vector was transfected in six-wellplates by using Lipofectamin PLUS according to the manufac-turer’s instructions (Invitrogen). cDNA-expressing vectors werepRC-CMV-E48 and pcDNA-IL-13R�1-HA (a kind gift of K.Kohno, University of Occupational and Environmental Health,Kita-Kyushu, Japan). The angiotensinogen (AGT) expressionvector was obtained from Invitrogen (clone ID 4213559). Trans-fected cells overexpressing the corresponding cDNA were as-sessed by real-time PCR quantification.

Quantification of Tumorgenicity and Metastasis. A total of 106 HeLacells transduced with retroviral constructs expressing TF 20-VP(seven mice), TF 20-SKD (seven mice), and no ZF domains(control, seven mice) were implanted s.c. into 3- to 4-week-oldnonobese diabetic (NOD) severe combined immunodeficient(SCID) mice females (The Scripps Research Institute rodentbreeding colony). The weight of each animal and the primary tumorvolume were monitored each week. Animals were killed at day 35postinjection. Lungs harvested from the animals were fixed inBouin’s solution, and the number of macroscopic metastases was

assessed by counting nodules at the surface of the lung under adissecting microscope. For consistency, the upper right lobe of eachlung was used for quantification. In all cases this lobe representedmetastases to the whole lung. Lung metastasis generated from TF20-VP cell injection was significantly higher than the control cellinjection (P � 0.05), but no significant difference in lung metastasiswas found between TF 20-SKD cell injection and the control cellinjection.

Microarray Processing and Data Analysis. RNA samples (see Sup-porting Text, which is published as supporting information on thePNAS web site, for a detailed protocol) were processed for hybrid-ization on Affymetrix HG-U133A microarrays following standardprocedures as recommended by the manufacturer. Microarray datawere retrieved as MAS5 flat files and imported into GENESPRING 6.1(Silicon Genetics, Redwood City, CA) or EXCEL (Microsoft).

Real-Time PCR. Changes in the expression of target genes wereexamined with real-time PCR. The full-length cDNA sequences forgenes of interest were obtained from the National Library ofMedicine (www.ncbi.nlm.nih.gov�UniGene). A detailed descrip-tion of real-time PCR and other procedures are provided inSupporting Text.

Results and DiscussionTF 20–VP Induces Phenotypic Transformations in HeLa Cells. We havedeveloped a functional selection strategy to identify and modulategenes involved in tumor progression (Fig. 1A). The ATF is used asa tool to induce a desired phenotype by perturbing endogenoustranscription and to further understand the genes that are requiredfor neoplastic disease progression. We hypothesized that a deliveryof a TF activator library into a drug-sensitive, poorly invasive, andnonmetastatic cell line would result in the generation of a pheno-typically diverse cancer cell population (or cancer cell library). Acomplex six-ZF activator library (comprising 8.42 � 107 differentATF proteins with unique DNA binding specificities) could regu-late multiple genes involved in tumor progression. Indeed, natural

Fig. 1. TF 20 is able to induce complex phenotypesin HeLa cells. (A) Illustration of TF-mediated repro-gramming of cancer cells. TF 20-VP induces morpho-logical transformations and cytoskeleton remodel-ing in HeLa cells. (B) Aspect of a colony of control cellsexpressing no ZFs. (Magnification: �100.) TF 20-SKD-transduced cells formed the same WT compact colo-nies (data not shown). (C and F) Control cells (C) andTF 20-VP-transduced cells (F) were stained for F-actin(Texas red-phalloidin, red) and nucleus (DAPI, blue).(Magnification: �600.) (D and E) Morphologicaltransformations of HeLa cells transduced with a ret-rovirus expressing TF 20-VP, showing cells migratingout of the colony. (Magnification: �100.) (G) TF 20-VPenhances the migration of HeLa cells on a lamininmigration assay.

Blancafort et al. PNAS � August 16, 2005 � vol. 102 � no. 33 � 11717

CELL

BIO

LOG

Y

TFs have recently been described that are able to induce epithelial-mesenchymal transitions involving dramatic remodeling of thecytoskeleton, loss of epithelial cell adhesion, and acquisition ofmigratory phenotypes (5). One of these natural TFs, Twist, has alsobeen associated with resistance to taxol (13). Likewise, we hypoth-esized that some ATF library members would be able to regulateseveral genes and induce complex malignant phenotypes, includingresistance to anticancer drugs and changes in cell morphology,migration, and invasion. Cells in the population displaying complexmorphological transformations can be selected with phenotypicscreens. The ATF responsible for the phenotypic switch can then beisolated and characterized. Expression profiles of ATF-expressingcells are compared with profiles of the original cell line to determinetarget genes required for the phenotype. One unique feature of thissystem is that the phenotype of cancer cells is transformed withoutchanging their genetic background. Instead, ATFs mediate complexchanges in gene expression profiles that can facilitate reprogram-ming a tumor cell phenotype.

We have chosen HeLa cells as a model system to study genesaffecting neoplastic disease progression because this cell line issensitive to taxol, a commonly used anticancer drug for severalcarcinomas, and because this cell line constitutes a prototypicalmodel of a noninvasive, poorly metastatic cell type. HeLa cellswere transduced with a six-ZF retroviral activator library [pMX-6ZF-library-VP64-IRES-GFP (11)]. Transduced cells werescreened for taxol resistance, and drug-resistant clones weremorphologically examined for the appearance of more complexmorphological transformations, resembling epithelial-mesen-chymal transition (5, 6).

Although several ATF clones were isolated that promoted drugresistance, a unique feature of one of the selected ATFs, TF 20-VP,was its ability to induce particularly aggressive phenotypic trans-formations. When TF 20-VP was expressed in HeLa cells by usingeither retroviral vectors or transient expression vectors, they ac-quired elongated, fibroblast-like morphologies (compare Fig. 1 Bwith D and E) as manifested by the orientation of the actin stressfibers (Fig. 1 C and F). HeLa cells expressing TF 20-VP alsodisplayed enhanced migration in laminin-coated transwell assays,compared with control cells expressing no ZF or cells containingthe same ZF linked to a repressor domain (TF 20-SKD) (Fig. 1G).Interestingly, TF 20-mediated effect on cell migration depended onthe nature of the extracellular matrix protein used in the assay, aswe observed no effect with collagen I, collagen IV, or fibronectin(data not shown). Both TF 20-VP and TF 20-SKD were expressedin HeLa cells at similar levels, as determined by semiquantitativeRNA expression analysis and flow cytometry (Fig. 5, which ispublished as supporting information on the PNAS web site).

TF 20–VP Induces Cell Invasion in Vitro and Enhances the Number ofDistal Metastases in NOD SCID Mice. The enhanced migratory�epithelial-mesenchymal transition type phenotype of cells express-ing TF 20-VP suggested altered invasive behavior. To test TF 20-VPinvasiveness in vitro, we performed Matrigel invasion assays. HeLacells transduced with a control retrovirus without ZFs and cellstransduced with TF20-SKD were poorly invasive in this assay.However, TF 20-VP was able to enhance cell invasion in vitro20-fold relative to cells transduced with a control retrovirus (Fig.2A). Invading cells conserved the morphological signatures of TF20-VP-expressing cells seen in Fig. 1 D and E.

Because invasive phenotypes are thought to be critical to a cell’sability to metastasize we investigated whether or not TF 20-VP wasable to promote metastasis in a mouse model. Approximately 106

HeLa cells transduced with a control virus, TF 20-VP, or TF20-SKD were implanted s.c. into NOD SCID mice (n � 7).Semiquantitative RT-PCR analysis using TF-specific primers andGFP analysis by flow cytometry showed that tumors transducedwith either TF 20-VP or TF 20-SKD expressed TFs with similarexpression levels and that strong TF expression persisted through

the time course of the experiment (Fig. 5A). At day 35 postinjectionmice were killed and the number of lung micrometastases wasdetermined (Fig. 2B). Mice implanted with TF 20-VP-transducedcells developed more lung metastases than mice implanted withcontrol cells (transduced with retrovirus without ZFs) or cellstransduced with TF 20-SKD, suggesting that TF 20-VP modulatedthe expression of genes enhancing the ability of tumor cells toproduce metastasis in the lungs. TF 20–VP-transduced cells had alower proliferation index in vitro than control cells or TF 20–SKD-transduced cells. In addition, TF 20-VP did not promote primarytumor growth compared with control tumors, whereas TF 20–SKDinhibited tumor growth in vivo (Fig. 6, which is published assupporting information on the PNAS web site). This finding isconsistent with recent reports showing that metastatic potential isnot always correlated to the number of cells in the primary tumoror tumor size (1).

TF 20–VP Regulates Specific Gene Expression Signatures. We nextevaluated the altered transcriptional profile mediated by TF 20-VPby determining which genes are differentially regulated by thisATF. We prepared duplicate independent transductions of HeLacells with TF 20-VP, TF 20-SKD, and control retrovirus. Untrans-duced cells were also evaluated. RNA expression profiles wereanalyzed by using a HG-U133A array from Affymetrix with�18,500 genes. Eight genes, listed in Table 1, which is published assupporting information on the PNAS web site, appeared to bedifferentially expressed in TF 20-VP-transduced cells comparedwith control and TF 20-SKD groups. Quantitative real-time ex-pression analysis was used to verify that five of these eight geneswere differentially regulated in TF 20-VP-expressing cells only(Table 2, which is published as supporting information on the PNASweb site). Three of these genes were highly regulated by TF 20-VP:E48 antigen (E48), AGT, and IL-13 receptor �1 (IL-13R�1).

E48 is a glycosylphosphatidylinositol-anchored molecule thatplays a role in cell–cell adhesion. E48 (LY-6D) belongs to the genefamily of Ly-6 antigens (14, 15). E48 is highly expressed in squamouscell carcinomas of the head and neck and constitutes a marker ofdisseminated tumor cells, particularly in lymph nodes and bonemarrow (16, 17). These tumors are characterized by local invasionresulting in poor prognosis. AGT is a precursor of AgtII, a cell-signaling molecule associated with a variety of disorders, such ascardiovascular remodeling and cancer (18–20). IL-13R�1 is a

Fig. 2. TF 20-VP enhances cell invasion and metastasis. (A) Invasion assayswith control and TF-transduced cells were performed in vitro by using Matri-gel chambers. (B) TF 20-VP increased the number of lung micrometastases ina NOD SCID mouse model of spontaneous metastasis. Groups represent HeLacells expressing TF 20 VP64 activator domain (TF 20-VP), the same protein butlinked to a repressor domain (TF 20-SKD), and cells expressing no ZF domains(control). P values between TF 20-VP and control and between TF 20-VP andTF 20-SKD were �0.05; P values between TF 20-SKD and control were �0.4.(Magnifications: �200.)

11718 � www.pnas.org�cgi�doi�10.1073�pnas.0501162102 Blancafort et al.

receptor for IL-13 and IL-4 and has been shown to mediatesignaling processes that result in activation of the Jak1–STATpathway. Interestingly, strong expression of this target has beendetected in 16 cell lines for squamous cell carcinomas of the headand neck (21–23).

A mAb that detects human E48 antigen was used to confirm a500- to 1,000-fold induction of E48 antigen in HeLa cells trans-fected with TF 20-VP. In contrast, untransduced HeLa cells andcells transduced with TF 20-SKD did not express significant E48antigen. Immunofluorescence of TF 20-VP-transduced HeLa cellsconfirmed the E48 up-regulation, particularly in cell–cell junctions(Fig. 3C). We detected 8- to 10-fold up-regulation of IL-13R�1 bothby DNA arrays and real-time PCR. In addition to these threestrongly up-regulated markers, we observed a 3.1-fold up–regulation of the gene ABCC5, a multidrug resistance ABC trans-porter and a 4.7-fold up–regulation of a cDNA of unknown function(cDNA: FLJ22642 fis, clone HS106970). To study the specificity ofgene regulation mediated by TF 20-VP, we analyzed expressionlevels of these five genes in cells expressing four other unrelatedTFs. As shown in Fig. 7, which is published as supporting infor-mation on the PNAS web site, these five genes were specificallyup-regulated in only TF 20-VP-expressing cells.

Invasiveness and Tumor Cell Dissemination Requires Regulation ofMultiple Targets. Our functional analysis focused on the three genesmost strongly and specifically up–regulated by TF 20–VP: E48,AGT, and IL-13R�1. However, additional experiments are war-ranted to evaluate the specific contribution of ABCC5 and the geneof unknown function. To extend our expression analysis to the invivo model, we first determined whether or not E48, AGT, andIL-13R�1 were differentially regulated in the primary tumors fromNOD SCID mice implanted with HeLa cells expressing TF 20-VP.Cells were recovered from tumors 2 and 6 weeks postinjection. Cellsexpressing GFP were sorted by flow cytometry, and gene expres-sion was analyzed by real-time PCR. E48 and AGT were stronglyexpressed in the tumors at 2 and 6 weeks (Table 2). However, we

did not detect significant up-regulation of IL-13R�1 in tumors ateither time point. One possibility is that IL-13R�1 plays a role in theinitial steps of tumor progression and is silenced in tumors by 2weeks postinjection. In contrast, E48 expression was increased bythree orders of magnitude in cells derived from TF 20-VP-containing tumors at both 2 and 6 weeks compared with control orTF 20-SKD tumors, as assessed by real-time RNA quantificationand flow cytometry (Table 2 and Fig. 3B).

TF 20 VP-mediated induction of E48 expression in vitro and itsincreased expression in late stages of tumor development in vivosuggested a role of E48 in tumor progression. Several glycosylphos-phatidylinositol-anchored proteins have been shown to promotecytoskeleton reorganization, changes in cell shape, cell attachmentand extracellular matrix-specific migration in neutrophil cells (24),pre-B lymphocytes (25), and breast carcinomas (26). These effectsdepend on a cross-talk between the glycosylphosphatidylinositol-containing protein and specific integrins. In light of these obser-vations, we first investigated the possibility that E48 expressioncould influence cell migration. As indicated in Fig. 3D, ectopicexpression of E48 cDNA in HeLa cells induced extracellularmatrix-dependent cell migration. As in TF 20- VP-expressing cells,E48 induced migration in laminin-coated transwells (Fig. 3D), butnot in collagen or fibronectin-coated wells, suggesting an involve-ment of specific integrin signaling. We also observed that, like TF20-VP transduced cells, cells that overexpressed E48 had elongatedfibroblast-like cell morphologies (data not shown), suggesting thatE48 participates in the cytoskeleton-remodeling characteristic ofTF 20-VP-expressing cells.

To better understand the function of E48 and its potentialinvolvement in promoting tumor progression in vivo, we studiedexperimental metastasis formation in chicken embryos. Thechicken embryo system has two important advantages for studyingthe role of potential TF-targeted cDNAs: metastases can begenerated and analyzed much more quickly than in mouse models(7 days in chickens versus 35 days in mouse) and methodology existsto allow precise time course-dependent quantification of dissemi-

Fig. 3. TF 20-VP regulates the endogenous E48 gene, amarker of disseminated tumor cells of squamous cell carcino-mas of the head and neck. (A) HeLa cells were analyzed for E48expression 72 h posttransduction. (B) HeLa cells recoveredfrom mouse primary tumors 35 days postinjection. Flow cy-tometry analyses were performed with a mAb that detectshuman E48. E48 is up-regulated in HeLa cells transduced witha TF 20-VP retrovirus (red); HeLa cells expressing TF 20-SKD(light blue); control cells expressing no ZFs (green); untrans-duced cells (filled blue). (C) Immunofluorescence analysis ofE48 expression in HeLa cells expressing TF 20-VP, control cellsexpressing no ZFs (control), and cells overexpressing an E48cDNA (E48). Expression of E48 was induced in the cell–celljunctions (arrow). (Magnifications: �400.) (D) Ectopic expres-sion of E48 induces cell migration in Laminin-coated tran-swells. HeLa cells were transduced with EGFP, TF 20-VP, andE48 cDNA (E48). Untransduced (HeLa cells) were also evalu-ated. (E) HeLa cells transduced with TF 20-VP, TF 20-SKD, andE48 retroviruses were injected i.v. in chicken embryos; dissem-inated tumor cells were detected in distal organs (lung andlower chorioallantoic membrane, CAM) by real-time alu-PCRas described (27). HeLa cells expressing both TF 20-VP and E48cDNA enhance 10- to 20-fold and 5-fold, respectively, thenumber of experimental metastasis in a chicken embryomodel of organ colonization.

Blancafort et al. PNAS � August 16, 2005 � vol. 102 � no. 33 � 11719

CELL

BIO

LOG

Y

nated tumor cells by using real-time PCR to detect alu-humansequences in chicken organs (27). In the experimental metastasissystem described in Fig. 3E, control HeLa cells and cells transducedwith TF 20-VP, TF 20-SKD, or a retroviral construct expressinghuman E48 cDNA (E48) were injected i.v. into the allantoic vein,and disseminated cells were detected by quantitative PCR. In thismodel, TF 20-VP was able to enhance the process of organcolonization in the lungs and the lower chorioallantoic membraneby 10- and 20-fold, respectively, compared with control and TF20-SKD cells. Furthermore, E48-expressing cells were able toincrease by 5-fold the colonization of HeLa cells in the lowerchorioallantoic membrane relative to the control. Thus, these dataconfirmed the role of TF 20-VP in promoting organ colonizationand also suggested a role of E48 antigen in enhancing tumor celldissemination in specific organs. Nevertheless, E48 by itself was notable to fully recapitulate the TF 20-VP effect in tumor celldissemination, suggesting that the fully metastatic phenotype con-ferred by TF 20-VP requires concerted and synergistic action ofseveral targets.

Subsequently, we analyzed the possibility that coexpression of TF20-VP-targeted genes could cooperate in generating the invasivephenotype of TF 20-VP-expressing cells. We introduced the cDNAscoding for E48, AGT, or IL-13R�1 into HeLa cells by usingtransient expression vectors, and transfected cells were evaluated byusing in vitro invasion assays. Expression of these proteins wasefficiently up-regulated, as confirmed by real-time quantification.As shown in Fig. 4A, transfection of individual targets had littleimpact on cell invasion. However, the simultaneous expression ofAGT, E48, and IL-13R�1 resulted in a cooperative effect thatenhanced the ability of HeLa cells to invade the matrix. Interest-ingly, cotransfection of E48 and AGT recapitulated the elongatedphenotype and the cytoskeleton-remodeling characteristic of TF20-VP expression (Fig. 4B). Cotransfection of IL-13R�1 did notenhance cell shape and affected only the efficiency of invasion.

Nevertheless, cells transiently transfected with TF 20-VP were2.6-fold more efficient in promoting cell invasion than E48, AGT,and IL-13R�1 cDNAs combined. This result further confirms thatTF 20-VP regulates multiple gene targets that are cooperativelynecessary for cell invasion. The failure of cDNA transfection to fullyrecapitulate the phenotype provided by TF 20-VP might also beattributed to contributions provided by splice variants that are notproduced under cDNA delivery.

It is possible that other gene products up-regulated by TF 20-VPbut not detected in this study are necessary to fully recapitulate theTF 20-VP phenotype. By binding to multiple targets, a TF canorchestrate and regulate the expression of each target gene, and itssplice variants, at the proper level. Unlike heterologous cDNAs thatexpress a single splice variant, TFs use the genomic scaffold thatprovides endogenous elements and spatiotemporal cues necessaryfor target gene regulation. A biochemical analysis of the E48, AGT,and IL-13R�1 gene promoters has revealed functional TF 20binding sites for the AGT and IL-13R�1 proximal promoters,suggesting direct regulation. The E48 reporter gene was not regu-lated by a TF 20 site found upstream of the proximal promoter andcould be regulated indirectly by another gene or a more distal site(Fig. 8, which is published as supporting information on the PNASweb site). Although our experiments analyzed the effect of a limitednumber of targets on the process of cell invasion in vitro, in vivothese genes may contribute to any one of many different steps of themetastatic cascade such as mobility, invasion of surrounding tissues,intravasation or extravasation and survival in the vasculature, orcolonization and survival in distal organs.

Overall, the above experiments provide evidence that TF 20-VPregulates expression of E48, AGT, and IL-13R�1. Real-time PCRexperiments on HeLa cells showed that overexpression of eachgiven target individually did not affect transcription of the othermarkers (Table 2). This finding indicates that TF 20 affects tran-scription of these three genes autonomously.

Several reports have suggested an association of E48 antigenexpression with increased tumor progression (28–31). The E48antigen is highly expressed in locally invasive squamous cell carci-nomas of the head and neck and constitutes a marker of dissemi-nated tumor cells both in lymph nodes and bone marrow (32).Recently, the presence of micrometastatic cells in bone marrow ofpatients with squamous cell carcinomas of the head and neck withtwo or more lymph node metastases has been correlated with apoor metastasis-free survival (33). It has been hypothesized thatE48 is a signal transduction protein that plays a role in cell–cellcommunication, and recent reports suggest involvement in medi-ating selectin-dependent binding of premetastatic tumor cells to theendothelium (28). Nevertheless, the signal transduction pathwayand the adhesion system activated by E48 was unknown. Our datademonstrate a role of E48 in promoting matrix-dependent migra-tion of tumor cells, suggesting integrin-dependent signaling. Addi-tionally, we found that E48 is able to facilitate organ colonizationin in vivo models, thus confirming the role of this antigen in tumordissemination.

Expression of AGT, a precursor of the signaling molecule AgtII,is regulated in a tissue-specific manner (34–37). In heart tissue,AgtII activates Jak1, Jak2, and Tyk2, resulting in activation ofSTATs (38). In addition, AgtII can influence important cellularprocesses, such as cytoskeleton remodeling, migration, and sur-vival (19).

Signaling mediated by IL-13R involves binding of cytokines IL-13and IL-4 to the receptor and triggers activation of two pathways: theJak–STAT pathway and the phosphatidylinositol 3-kinase pathway(39). One intriguing possibility is that some AgtII- and IL-13R-signaling cascades are necessary for the TF 20-VP-mediated met-astatic phenotype. In this regard, TF 20-VP-expressing cells offer anexperimental framework to study the effect of activators andinhibitors of signaling pathways in the process of cell invasion andmetastasis. Our preliminary experiments indicate that TF-20-VP-

Fig. 4. In vitro induction of cell invasion requires coexpression of multipletargets. (A) HeLa cells were transiently transfected with individual cDNAsencoding E48, AGT, IL-13R�1, and TF 20-VP. Transfected cells were loaded intoMatrigel chambers, and invading cells were fixed and counted. Values repre-sent averages of two wells, and experiments were done in triplicate. (B) E48and AGT coexpression in HeLa cells suffices to recapitulate the cell morphol-ogy changes mediated by TF 20-VP. (Magnification: �400.)

11720 � www.pnas.org�cgi�doi�10.1073�pnas.0501162102 Blancafort et al.

mediated cell invasion in vitro is stimulated by AgtII and dramat-ically abolished by a Jak2-specific inhibitor (Fig. 9, which is pub-lished as supporting information on the PNAS web site). Thisfinding supports the possibility that Jak2 signaling is important forthe TF 20-VP-mediated invasive phenotype. Recently, activation ofthe Jak2–STAT pathway has been shown to influence survival,invasion, and metastasis in lymphomas (40). We expect that TF-mediated dysregulation of critical signaling molecules will allowinvestigators to transform the phenotype of tumor cells or ‘‘repro-gram’’ tumor cells.

In this study we have described an ATF, TF 20-VP, selected froma combinatorial six-ZF TF library that is able to alter phenotypicproperties of HeLa cells. HeLa cells are taxol-sensitive, noninva-sive, and nonmetastatic. However, upon delivery of TF 20-VP thesecells became drug-resistant, invasive, and metastatic. We demon-strated, using in vivo models, that the ability to enhance metastasisand dissemination of tumor cells depended on linkage to anactivator domain. The fact that the same selected six-ZF DBDdomain linked to a repressor domain or cells not expressing any ZFdid not manifest the phenotype, allowed us to perform an analysisof differentially expressed targets. We validated, by real-time PCR,a group of five genes whose expression was altered only in TF20-VP-expressing cells. Recently, an increasing number of geneticprofiling experiments have been reported with the ultimate goal ofidentifying genetic markers involved in cancer progression andmetastasis (1–3). These studies often compared tumors fromdifferent genetic backgrounds and the resulting heterogeneity ofthe samples complicated the functional analysis of the targets. Theuse of ATFs to artificially modify cancer cell phenotypes is appeal-ing because the TF introduces a transcriptional perturbation of geneexpression but the genetic background of the cell line remains thesame. TFs selected from combinatorial libraries can be used toidentify generic biomarkers of tumor progression. Of the fivedifferentially expressed genes reported here, three were cell surfaceproteins, illustrating the power of this approach in the identificationof cell surface antigens involved in tumor progression.

Functional studies suggest that three of the up-regulated mark-ers, E48, AGT, and IL-13R�1, contributed synergistically in reca-pitulating some aspects of the TF 20-VP-induced phenotype. Thisphenomenon is consistent with the fact that increased malignancyof human tumors seems to require the action of several genesoperating in concert. Mechanistically a TF can achieve simulta-neous regulation of several genes by binding multiple promoters.This ability to regulate multiple defined genes is a distinctive featureof TFs, compared with other strategies to modulate gene expres-sion, such as ribozymes, antisense, and RNA interference. Whereasthe latter strategies typically target a specific RNA sequence froma single gene, ATFs can bind similar or identical DNA sequenceslocated in several regulatory regions, facilitating targeting of mul-tiple genes. Additionally, ATFs can induce either gain-of-functionphenotypes or knockdown phenotypes, whereas RNA-derivedstrategies can achieve only knockdowns.

This work demonstrates that ATFs selected from combinatoriallibraries can be used to transcriptionally reprogram cancer cells tomodify complex phenotypes and identify genes involved in cancerprogression. ATF library screens could also be used to interferewith the regulation of genetic or signaling cascades to modify orrevert certain aspects of a malignant phenotype. In recent experi-ments, we observed that TF 20-SKD is able to efficiently reduce cellinvasion in a highly metastatic melanoma cell line (Fig. 10 which ispublished as supporting information on the PNAS web site).Together with the xenograph model (Fig. 6), these experimentsdemonstrate the use of ATFs to negatively effect cell invasion andtumor growth. In summary, we have shown that ATFs can be usedas tools to dissect the function of genes in tumor progression. Thiswork also suggests potential applications of ATFs in cancer therapy.

We thank Drs. A. Fukamizu and C. M. Perou for discussions, Dr. K.Kohno for the IL-13R�1 cDNA and Luc reporter constructs, and L.Asawapornmongkol for technical support. This work was supported byNational Institutes of Health Grant R01CA086258.

1. Ramaswamy, S., Ross, K. N., Lander, E. S. & Golub, T. R. (2003) Nat. Genet. 33,49–87.

2. Rhodes, D. R., Yu, J., Shanker, K., Deshpande, N., Varambally, R., Ghosh, D.,Barrette. T., Pandey, A. & Chinnaiyan, A. M. (2004) Proc. Natl. Acad. Sci. USA 101,9309–9314.

3. Segal, E., Friedman, N. & Regev, A. (2004) Nat. Genet. 36, 1090–1098.4. Beneklli, M., Baer, M. R., Baumann, H. & Wetzier, M. (2003) Blood 101, 2940–2951.5. Kang, Y. & Massague, J. E. (2004) Cell 118, 277–279.6. Vermon, A. E & LaBonne, C. (2004) Cell 14, R719–R721.7. Blancafort, P., Segal, D. J & Barbas, C. F., III (2004) Mol. Pharmacol. 66, 1361–1371.8. Dreier, B., Segal, D. J. & Barbas, C. F., III (2002) J. Mol. Biol. 303, 489–502.9. Segal, D. J., Beerli, R. R., Blancafort, P., Dreider, B., Effertz, K., Huber, A., Koksch,

B., Magnenat, L., Valente, D. & Barbas, C. F., III (2003) Biochemistry 42, 2137–2148.10. Blancafort, P., Magnenat, L. & Barbas, C. F., III (2003) Nat. Biotechnol. 21, 269–274.11. Lund, C. V., Blancafort, P. & Barbas, C. F., III (2004) J. Mol. Biol. 340, 599–613.12. Magnenat, L., Blancafort, P. & Barbas, C. F., III (2004) J. Mol. Biol. 341, 635–649.13. Wang, X., Ling, M. T., Guan, X.-Y., Tsao, S. W., Cheung, H. W., Lee, D. T. & Wong,

Y. C. (2004) Oncogene 23, 474–482.14. Schrijvers, A. H., Gerretsen, M., Fritz, J. M., van Walsum, M., Quak, J. J., Snow, G. B.

& van Dongen, G. A. (1991) Exp. Cell Res. 196, 264–269.15. Brakenhoff, R. H., Gerretsen, M., Knippels, E. M., van Dijk, M., van Essen, H.,

Weghuis, D. O., Sinke, R. J., Snow, G. B. & van Dongen, G. A. (1995) J. Cell Biol.129, 1677–1689.

16. Brakenhoff, R. H., Stroomer, J. G., ten Brink, C., de Bree, R., Weima, S. M., Snow,G. B. & van Dongen, G. A. (1999) Clin. Cancer Res. 5, 725–732.

17. Pantel, K. & Brakenhoff, R. H. (2004) Nat. Rev. Cancer 4, 448–456.18. Mascareno, E., Dhar, M. & Siddiqui, M. A. Q. (1998) Proc. Natl. Acad. Sci. USA 95,

5590–5594.19. Berry, G. C., Touyz, R., Dominiczak, R. C., Webb, R. C. & Johns, D. G. (2001) Am. J.

Physiol. 281, H2337–H2365.20. Juillerat-Jeanneret, L., Celerier, J., Chapuis Bernasconi, C., Nguyen, G., Wostl, W.,

Maerki, H. P., Janzer, R. C., Corvol, P. & Gasc, J. M. (2004) Br. J. Cancer 90,1059–1068.

21. Murata, T., Obiri, N. I., Debinski, W. & Puri, R. J. (1997) Biochem. Biophys. Res.Commun. 238, 90–94.

22. Ise, T., Izumi, H., Nagatani, G., Takano, H., Wada, M., Kuwano, M. & Kohno, K.(1999) Biochem. Biophys. Res. Commun. 265, 387–394.

23. Joshi, B. H., Kawakami, K., Leland, P. & Puri, R. K. (2002) Clin. Cancer Res. 8,1948–1956.

24. Funaro, A., Ortolan, E., Ferranti, B., Gargiulo, L., Notaro, R., Luzzato, L. &Malavasi, F. (2004) Blood 104, 4269–4278.

25. Kanse, S. M., Chavakis, T., Kuo, A., Bdeir, K., Cines, D. V. & Preissner, K. T. (2004)Cell Biochem. Funct. 22, 257–264.

26. Burbach, B. J. Y. & Rapraeger, A. C. (2004) Exp. Cell Res. 300, 234–247.27. Zijlstra, A., Mellor, R., Panzarella, G., Aimes, R. T., Hooper, J. D., Marchenko, N. D.

& Quigley, J. P. (2002) Cancer Res. 62, 7083–7092.28. Eshel, R., Zanin, A., Sagi-Assif, O., Meshel, T., Smorodinsky, N. I., Dwir, O., Alon,

R., Brakenhoff, R., van Dongen, G. & Witz, I. P. (2000) J. Biol. Chem. 275,12833–12840.

29. Figuerola, R., Nadal Serra, A., Farre Pueyo, X. & Palomar Garcia, V. (2000) ActaOtorrinolaringol. Esp. 51, 288–292.

30. Eshel, R., Zanin, A., Kapon, D, Sagi-Assif, O., Brakenhoff, R., van Dongen, G. &Witz, I. P. (2002) Int. J. Cancer 98, 803–810.

31. Partidge, M., Brakenhoff, R., Phillips, E., Ali K., Francis, R., Hooper, R., Lavery, K.,Brown, A. & Langdon, J. (2003) Clin. Cancer Res. 9, 5287–5294.

32. Nieuwenhuis, E., Leemans, C. R., Kummer, J. A., Denkers, F., Snow, G. B. &Brakenhoff, R. H. (2003) Clin. Cancer Res. 9, 755–761.

33. Colnot, D. R., Nieuwenhuis, E. J., Kuik, D. J., Leemans, C. R., Dijkstra, J., Snow, G.B, van Dongen, G. A. & Brakenhoff, R. H. (2004) Clin. Cancer Res. 10, 7827–7833.

34. Nishii, T., Moriguchi, A., Morishita, R., Yamada, K., Nakamura, S., Tomita, N.,Kaneda, Y., Fukamizu, A., Mikami, H., Higaki, J. & Ogihara, T. (1999) Circ. Res. 85,257–263.

35. Mascareno, E., El-Shafei, M., Maulik, N., Sato, M., Guo, Y., Das, D. K. & Siddiqui,M. A. (2001) Circulation 104, 325–329.

36. Lam, K. Y. & Leung, P. S. (2002) Eur. J. Endocrinol. 146, 567–572.37. Mazak, I., Fiebeler, A., Muller, D. N., Park, J. K., Shagdarsuren, E., Lindschau, C.,

Dechend, R., Viedt, C., Pilz, B., Haller, H. & Luft, F. C. (2004) Circulation 109,2792–2800.

38. Guo, Y., Mascareno, E. & Siddiqui, M. A. (2004) Mol. Endocrinol. 18, 1033–1041.39. Umeshita-Suyama, R., Sugimoto, R., Akaiwa, M., Arima, K., Yu, B., Wada, M.,

Kuwano, M., Nakajima, K., Hamasaki, N. & Izuhara, K. (2000) Int. Immunol. 12,1499–1509.

40. Opdam, F. J. M., Kamp, M., Bruijn, R. & Roos, E. (2004) Oncogene 23, 6647–6653.

Blancafort et al. PNAS � August 16, 2005 � vol. 102 � no. 33 � 11721

CELL

BIO

LOG

Y

Supporting information

Supporting Figure Legends

Fig. 5. TF 20-VP and TF 20-SKD are expressed in both (A) transduced cells andin (B) tumor cells recovered from NOD-SCID mice 35 days post-injection. RT-PCR experiments (left panels) were performed using ZF-specific primers thatamplify the first three ZFs of TF 20. Control samples represent cells transducedwith retrovirus vectors in absence of ZF domains. GAPDH was used as anormalizing control. Flow cytometry measurements (right panels) were performedto detect GFP, a marker for protein expression.

Fig. 6. TF 20-VP do not confer growth advantages to HeLa cells as assessed (A)in vitro by proliferation assays or (B) by monitoring the growth of the primarytumor in NOD SCID mice implanted with TF-20 HeLa transduced cells.

Fig. 7. TF 20-VP up-regulates target genes specifically. Expression levels ofeach of the indicated targets were compared in different TF-transduced cells.Affymetrix DNA arrays were used in two independent experiments usingduplicate biological samples. Normalization was done with Affymetrix MAS5algorithm. VP represents an activator domain and SKD a repressor domain. SS-VP and SS-SKD indicate constructs expressing no ZFs.

Fig. 8. TF 20-VP-mediated cell invasion of TF 20-VP transduced cells isstimulated by Angiotensin II (AgtII) and inhibited by the Jak2 specific inhibitorAG490. Invasion was not inhibited by other kinase inhibitors, such as ERK2/1inhibitor (U0126), a p38 inhibitor (SB 203580), or a Jak3 specific inhibitor (Jak3I).Concentrations of these inhibitors are indicated in mM. TF 20-VP transducedcells were starved overnight in serum free media and treated for 2.5 hr withdifferent concentrations of the indicated inhibitors. Cells were then evaluated inMatrigel invasion assays as described in Materials and Methods. Data wasnormalized to TF 20-VP transduced cells treated in absence of drugs. Datarepresents average of two wells from each of two independent experiments.Drugs were purchased from Calbiochem, San Diego, CA. These compoundswere not toxic at the indicated concentrations (mM) as assessed by survivalassays (XTT assays, Roche). * Indicates p<0.05.

Fig. 9. TF 20-SKD reduces the number of invasive melanoma C8161 cells inmatrigel invasion assays. Melanoma cells were transduced with an emptyretroviral vector (Control) or with a TF 20-SKD a retroviral vector. The percentageof invasive cells was determined as described above.

Supporting Materials and MethodsTotal RNA extraction and RNA quality assessment for Microarray profiling.Cell lysate was homogenized by passing over a QIAshredder spin column(Qiagen) and total RNA was isolated from the cell pellet using the RNeasy MiniKit (Qiagen) according to the manufacturer’s instructions. RNA quantity andquality was analyzed with the RNA 6000 Nano Assay (Agilent Technologies)according to the recommended protocol of the manufacturer. The computationalanalyses were performed with the Agilent 2100 Bioanalyzer software (AgilentTechnologies).

Affymetrix MAS5 data were “normalized” to a constant value of 1 inGeneSpring, effectively keeping the global normalization to a target intensity of150 by the MAS5 algorithm. For clustering analyses, the experimentinterpretation was set to “log of ratio” and Pearson correlation was used as asimilarity measure for both dimensions (Gene tree and experiment tree). Toderive differentially expressed genes, a parametric test, not assuming equalvariances (Welch t-test), was used in the “log of ratio” mode. No multiple testingcorrection was applied. Fold change filtering was either performed in GeneSpringwith the experiment interpretation set to “ratio” or in Excel. Genes were onlyconsidered as differentially expressed when the majority of measurementscorresponded to “present” or “marginal” calls in at least one group ofexperiments. Specific cut-off values for the various filtering steps are given in theresults section.

Real-Time PCRPrimers were designed to amplify the human target genes using a web-basedsoftware (http://www.genome.wi.mit.edu/cgibin/primer/primer3_www.cgi) andwere purchased from MWG. 5 µg of RNA from each tested sample were

reverse-transcribed with Superscript II reverse transcriptase (Invitrogen LifeTechnologies, Inc., CA). The resulting cDNA was diluted 20-fold prior to PCRamplification. Reactions were performed using iQ SYBR Green Supermix (Bio-rad laboratories, Hercules, CA). Each PCR reaction was performed in a finalvolume of 10 L under 10 L of mineral oil with the iCycler iQ real-time PCRdetection system (Bio-rad laboratories, Hercules, CA). A typical protocolinvolved a 2 min of denaturation at 95oC, 40 cycles with annealing at 55oC for 15sec, and extension at 72oC for 15 sec. An automated melting curve analysis wasused to verify that all primers yielded a single PCR product. A quantitativemeasurement of total RNA was obtained by amplification of the human GAPDHprimers (forward: 5’GGGAAGGTGAAGGTCGGAGT3’ and reverse:5’TCCACTTTACCAGAGTTAAAAGCAG3’). Real time PCR was performed usingprimers specific for IL13Ra1 forward: 5’CTCCACCAGTCATTTTTCAG3’ andreverse: 5’ATTATCCTCTGCTCCTCCAG3’, EphB2 forward:

5’TGCAGCTCCAGGTACATATC3’ and reverse:5’AACAAACAAACCCCCTAAAC3’, PLSCR-1 forward:5’TCCATTAAACTGTCCACCTG3’ and reverse: 5’TGCAAAGTAAACCCTCTGTC3’,AGT forward: 5’TGTACATACACCCCTTCCAC3’ and reverse:5’CTCAACTTGTCTTCGGTGTC3’, ABCC5 forward:5’GTCTCACACTGGCGTAGAAG3’ and reverse:5’GTTCAGCAAACATGCTAAGG3’, TNFRSF21 forward:5’GGGATTCCTTCACCAATTAC3’ and reverse: 5’CTTCACCACTACCCACAAAC3’,FLJ22642 fis forward: 5’TTGGCTTGTGGAATTTACTG3’ and reverse:5’CTGCCTCTGTGAAAGAGTTG3’, E48 forward:5’AGATGAGGACAGCATTGCTGC3’ and reverse:5’GCAGACCACAGAATGCTTGC3’. The fluorescence emitted by the reporter dyewas recorded as a quantitative measure of the amount of PCR product in thesample. The Ct is the fractional cycle number at which the fluorescencegenerated by the reporter dye exceeds a fixed level above baseline. Signalsfrom amplification of target genes were normalized against the relative quantity ofGAPDH and expressed as Ct – (CtGAPDH – Ctgene). The changes in target genesignal relative to the total amount of mRNA were expressed as Ct = Ctcontrol -

Ctgene. Relative fold differences in gene expression comparison were calculatedas 2 Ct. Each gene expression analysis was normalized and calculated againstthe indicated control samples in duplicate. Target gene expression waspresented as an average value of fold changes against the control.

Analysis of GFP and E48 expression in primary tumors. For GFPmeasurements in the primary tumors, tumor cells were recovered from primarytumors at 2, 3, and 6 weeks post-injection (2 animals per group). The expressionof GFP in these recovered tumor cells was tracked by flow cytometry. Data wasanalyzed using CELLQuest software (Becton Dickinson). For E48 staining ofprimary tumors, cells were recovered from tumors 6-week post-injection, andGFP positive cells were sorted using a FACSVantage (Becton Dickinson). Thesorted GFP positive cells were then used to detect the expression of E48 usingan anti-human E48 antibody (5 mg/mL) (28) and goat anti-mouse phycoerythrinconjugated secondary antibody (Jackson ImmunoResearch, 1:100 dilution) andanalyzed by flow cytometry as described above.

Semi-quantitative PCR. Approximately 5 x 106 transduced cells were collected72 hr post-transduction and RNA was extracted using the TRI reagent (MolecularResearch Center). Reverse transcription was done using the Superscript kit( Inv i t rogen) . Pr imers used for TF 20 detect ion were5’GCCCAGGCGGCCCTCGAGCCCGGGGAG3’ a n d5’GGCTGGCCAGGTGGCCGGCCTGGCTGAAAG3’ that specifically amplify thefirst three ZFs of TF 20. Conditions for PCR amplification were: 5 min 94°C, 1

min 94°C, 30 sec 56°C, 1 min 30 sec 72°C for 25 cycles, and 10 min 72°C. For

ZF expression analysis in the primary tumors, tissue from primary tumors 35-days post-injection was removed and RNA extracted (Qiagen). RT-PCR was

done as described above. Expression of glyceraldehyde-3-phosphatedehydrogenase (GAPDH) was measured as described (10).

Immunofluorescence. For the immunofluorescent detection of E48, HeLa cellstransduced with retroviral constructs TF 20-VP, TF 20-SKD, control constructs inwithout ZF domains, or a retroviral construct overexpressing E48 cDNA (E48)were cultured on glass coverslips for 24 hr and subsequently fixed with 2%formaldehyde. Samples were blocked with 3% BSA and 5% normal goat serumto prevent non-specific staining. Stained samples were mounted onto glassslides and analyzed by confocal microscopy (MRC1024 laser scanning confocalmicroscope, Biorad). For actin staining 104 cells were cultured in glass coverslipsfor 24 hr, fixed with 2% formaldehyde and stained with Texas red-phalloidin(Molecular Probes) as described by the manufacturer’s instructions. 3-D imageswere obtained using an Olympus IX-70 Delta Vision Deconvolution Microscopeand analyzed using softWoRx 2.5.

5’ GGA GCA GCT GAA GGT CAC 3’ ANG -490 subs

A B

C

D

E

uc

–516

–344

–312

–243

-1222

LucAGT Promoter fragments

–76 Duplex (-516 -> –429)

–516 –344 Duplex (-429-> –344)

*

Table 1. List of differentially regulated genes in TF 20-VP infected samples versus control groups

Gene BankAccession Number

Description Gene Symbol Fold Activation

NM_001560 interleukin 13 receptor, alpha 1 IL13RA1 11.34NM_0021105 phospholipid scramblase 1 PLSCR1 2.59NM_000029 angiotensinogen [serine (or cysteine)

proteinase inhibitor, clade A(alpha–1antiproteinase, antitrypsin)member 8]

AGT 98.51

NM_003695 lymphocyte antigen 6 complex, locus D E48 127.43NM_005688 ATP–binding cassette, subfamily C

(CFTR/MRP), member 5ABCC5 2.484

AK026295 Homosapiens cDNA: FLJ22642 fis,clone HS106970

22.29

NM_017449 Homo sapiens EphB2 EphB2 13.23NM_014452 tumor necrosis factor receptor

superfamily, member 21TNFR21 2.68

Control groups include mock-infected cells and cells infected with retroviral constructs expressing no ZFs.Affymetrix DNA arrays were performed in two independent experiments using duplicate biological samples.RNA samples were processed for hybridization on Affymetrix HG-U133A microarrays (interrogating approximately 18.500 transcripts).

Table 2. Gene target validation by realtime PCR

A: HeLa cells, 72hr post-infection

E48 AGT IL13R 1 EphB2 FLJ22642 ABCC5 TNFSF21 PLSCR

CONTROL 1 0.6 1.3 2.6 0.7 0.4 0.4 0.8SS-VP64 2.4 0.6 1.1 1.5 1.6 0.9 0.8 0.7SS-SKD 8.2 1.4 2.1 2.0 2.1 1.0 0.7 0.6TF 20-VP 791.0 3360.2 8.5 1.7 4.7 3.1 0.5 0.6TF 20-SKD 6.0 0.3 0.7 1.7 0.4 0.3 0.5 0.5

B: Tumors, 2weeks post-infection

E48 AGT IL13R 1

TF 20-VP 11.4 19.4 0.6TF 20-SKD 0.8 0.9 1.9Tumors, 6weeks post-infection

E48 AGT IL13R 1

TF 20-VP 4002.4 39.9 1.2TF 20-SKD 3.9 0.8 0.9

C: TransfectedHeLa cells

E48 AGT IL13R 1

E48 cDNA 4492.6 1.2 0.6AGT cDNA 2.3 74.5 0.9IL13R 1cDNA 0.9 1.4 38.6Mixed cDNA 9863.2 478.5 47.3

A: Gene target validation by real-time PCR using HeLa cells, 72 hr post-infection. Control represents HeLa cells transduced with a retroviralconstruct containing no ZFs (empty pMX-IRES-GFP); SS-VP64, a construct expressing an activator domain only; SS-SKD, a construct with arepressor domain only. The target gene expression was expressed relative to non-infected cells.B: Hela cells recovered from primary tumors, 2 weeks (top Table) and 6 weeks (bottom Table) post-transduction.C: Monitoring the expression of E48, AGT, and IL13R 1 by quantitative PCR using HeLa cells transiently transfected with 500ng of each cDNA.The mixed cDNAs sample was transfected with equal amount (250 ng) of each cDNA. Each gene target was measured in triplicates from two independent experiments.