Better Gene Expression by (-)Gene than by (+)Gene in Phage Gene DeliverySystems

Yun Liang,†,‡,§ Bizhi Shi,†,§ Jie Zhang,† Hua Jiang,†,‡ Yuhong Xu,† Zonghai Li,* ,† andJianren Gu* ,†,‡

National Laboratory for Oncogenes and Related Genes, Shanghai Cancer Institute, Shanghai Jiao Tong University,Shanghai 200032, P.R. China, and Medical College of Fudan University

In recent years, capsid-modified filamentous bacteriophage has become a potential vector forgene delivery into mammalian cells. However, little was known about how the carried gene inthe single-stranded genome expressed in mammalian cells. To explore whether the orientationof the carried gene affects its expression in the cells, we prepared EGF-displayed phagemidparticles whose genome carried the GFP gene or luciferase gene. The phagemid carried reportergenes either in the same orientation (called (+)gene) or in the contrary orientation (called (-)gene)to filamentous origin. Using these phagemid particles to infect H1299 cells, we found that thephages with (-) reporter genes had about 2-fold transduction efficiency as those with (+) reportergenes. These results indicated that phagemid carrying (-)gene of interest presented a betterprocedure in phage-mediated gene therapy. Furthermore, camptothecin (CPT) treatment wasalso applied and found to enhance both kinds of phagemid particles, and (-)gene still producedabout 1.5- to 2-fold transduction efficiency compared to those with (+)gene. Thus, it is imperativethat we clone the genes of interest in the reverse orientation to filamentous origin to enhancetheir expressions when preparing phagemid gene delivery vectors. Also, the results suggestedthat CPT could enhance both the replication of single-stranded DNA and its transcription.

IntroductionIn previous studies, researchers successfully modified the

filamentous bacteriophages that are capable of delivering genesto mammalian cells using targeting ligands such as growthfactors (1-5), antibodies (6), and viral capsid proteins (7).However, even modified phages entered most of the target cells,and the carried genes expressed only in a small percentage ofthe cells. Since the phage genome is single-stranded, the carriedgene of interest will be either the sense strand or antisense strandin accordance with its orientation to the bacteriophage originof DNA replication carried in the vector (8, 9). When the gene’sorientation is the same as that of the origin, there will be a sensestrand, (+)gene and when it is the opposite, there will be anantisense strand, (-)gene.

Previous studies implicated that some of the genes weretransformed into double-stranded DNA for transcription and thenfor translation (10). However, it cannot eliminate the possibilitythat single-stranded DNA can be transcribed directly. Therefore,in this study, we explored whether the two forms of the carriedgenes had effects on its expressions. It was found that the(-)gene presented expressions double those of the (+)gene.

Topoisomerase I inhibitor (camptothecin, CPT) treatment waspreviously used to enhance the transduction of phagemid genedelivery particles (1). However, its effects on the transductionof the two forms of phagemid particles in question have notbeen elucidated. Therefore, we also examined the effects of CPTtreatment on the transduction using phagemid particles.

Materials and Methods

Cell Culture. H1299 (large-cell lung carcinoma, ATCC,Manassas, VA) cells were cultured at 37°C in a mediumconsisting of 90% Dulbecco’s modified Eagle’s medium(DMEM) and 10% fetal bovine serum (FBS) in a humidifiedatmosphere of 95% air and 5% CO2.

Construction of Phagemid Gene Transfer Vectors.M13KO7-EGF was obtained by subcloning the EGF fragmentsinto the M13KO7P1 (5) Acc65I and EagI-digested Backbone.EGF was amplified from pAE-8 (11) using oligonucleotideprimers with Acc65I (5′ end) and EagI (3′ end) restrictionendonuclease site extensions (M13EGF-1: 5′-cggggtacctttctat-tctcactctaattccgactctgaatgccc-3′ and M13EGF-2: 5′-gtttCggc-cgaacctccaccacgcagttcccaccatttc-3′). The PCR cycle conditionswere 45 s at 95°C, 45 s at 60°C, and 30 s at 72°C for 20cycles, followed by a 5-min final extension step at 72°C.Consequently, the amplified DNA was purified, digested withAcc65I and EagI, and ligated into Acc65I and EagI-digestedM13KO7P1.

pEGFP-Amp was constructed by inserting the ampicillinresistance gene into pEGFP-N1 (BD Biosciences Clontech, PaloAlto, CA, USA) for replacing the kanamycin resistance gene.Ampicillin resistance gene cassette was amplified from pUC119(Takara, Dalian, China) using oligonucleotide primers with pvuII(5′ end) and Eco0109 I (3′ end) restriction endonuclease siteextensions (Amp-pvuII: 5′-CCCCAGCTGCGTCAGGTG-GCACTTTTC-3′ and Amp-Eco0: 5′-CTGAGGGCCTTAAAT-CAATCTAAAGTATATATGAG-3 ′). The conditions for PCRcycles were set as follows: 45 s at 95°C, 45 s at 60°C, and 90s at 72°C for 20 cycles, followed by a 7-min final extensionstep at 72°C. After that, the amplified DNA was purified,

* To whom correspondence should be addressed. E-mail: zonghaili@gmail.com; nlorg@sh163.net.

† Shanghai Jiao Tong University.‡ Medical College of Fudan University.§ Contributed equally to this work.

626 Biotechnol. Prog. 2006, 22, 626−630

10.1021/bp050392j CCC: $33.50 © 2006 American Chemical Society and American Institute of Chemical EngineersPublished on Web 05/03/2006

digested with pvuII and Eco0109 I, and ligated into pvuII andEco0109 I-digested pEGFP-N1.

The orientation of the filamentous origin was reversed in thepEGFP-Amp phagemid to construct pEGFP-AF, and pEGFP-amp was digested with AflII and PvuII. The DNA fragmentsproduced were blunted with Klenow enzyme and then recoveredby agarose gel electrophoresis. The fragments containing thefilamentous origin were ligated with Backbone.

pGl3-AF was constructed as follows. The BglII and BamHIenzymes digested the fragments from pGl3-Control (Promegacat. no. E1741, USA) containing a luciferase gene cassette, andthe remains containing the filamentous origin were blunt endedand ligated to create pGl3-AF, in which the orientation of theluciferase gene cassette was opposite to that of the filamentousorigin.

Helper Phage Preparation.M13KO7-EGF helper phageswere prepared as previously described (5). Phage titers were

determined by infecting ER2738 cells with phage dilutions andcounting the number of kanamycin-resistant clones obtained.

Preparation of Phagemid Particles.Phagemid particles wereprepared following a previous protocol (5), but with somemodification. Briefly, a clone of phagemid transformed ER2738was picked and transferred into 1 mL of Luria Bertani (LB)media containing 50µg/mL ampicillin. The culture was grownat 37 °C for 5 h, after which 100µL of the culture wastransferred into 1 mL of LB media with 70µg/mL kanamycin.Afterward, helper phage M13KO7-EGF [1× 109 plaque-forming units (pfu’s)] was added for immediate mixture with avortex. Shaken at 37°C for 2 h, all cultures were transferredinto 1 L of LB holding 70µg/mL kanamycin and 50µg/mLampicillin, which was then shaken at 37°C for 14-16 h.Phagemid particles were purified with poly(ethylene glycol)(PEG)/NaCl and washed several times.

Figure 1. The reporter gene’s orientation to filamentous origin decides whether the gene is (+) or (-). (A) Plasmids pEGFP-Amp and pEGFP-AFcontained the filamentous origin and (+) or (-) ssDNA packaged in either phagemid particles. (B) Plasmids pGl3-Control and pGl3-AF containedthe filamentous origin and (+) or (-) ssDNA packaged in either phagemid particles; flori, phage f1 origin.

Figure 2. Analysis of helper phage and phagemid particles. (A) DNA released from phage and phagemid particles was loaded on a 1% agarosegel. Lane 1, pEGFP-Amp phagemid particles; Lanes , pEGFP-AF phagemid particles; Lane 3, pGl3-Control phagemid particles; Lane 4, pGl3-AFphagemid particles; M,λ DNA/HindIII marker (Takara, Dalian, China). (B) Western blot analysis of pIII-EGF fusion proteins. Lane 1, M13KO7helper phage; Lane 2, pEGFP-Amp phagemid particles; Lane 3, pEGFP-AF phagemid particles; Lane 4, pGl3-Control phagemid particles; Lane 5,pGl3-AF phagemid particles.

Biotechnol. Prog., 2006, Vol. 22, No. 3 627

Quantification of Phagemid Particles.Phagemid particleswere quantified by enzyme-linked immunosorbent assay (ELISA)as previously described (12). Briefly, serial dilutions ofphagemid particles in coating buffer (0.1 M NaHCO3, pH 9.1)were coated in microtiter plates (Corning; Corning, NY) at 4°C overnight. Blocked with 1% bovine serum albumin (BSA)(Sigma, St. Louis, MO) in phosphate-buffered saline (PBS), thebound phages were stained with mouse anti-M13 antibodyconjugated with horseradish peroxidase (HRP; AmershamBiosciences) diluted 1:200 in PBST (PBS containing 0.1%Tween 20). The signal was developed using 2,2′-azino-di-(3-ethylbenzthiazoline-6-sulfonic acid) (ABTS) substrate andquantitated in an ELISA reader (Bio-Rad Laboratories, Hercules,CA). M13KO7 helper phage of the known numbers of pfu’swas used for standardization.

Western Blot Analysis. Purified phages or phagemid par-ticles were boiled in Tris-glycine-sodium dodecyl sulfate (SDS)sample buffer with 5% 2-mercaptoethanol for 5 min prior tobeing loaded on a 12% SDS-PAGE gel. Following theirseparation, the proteins were transferred to a nitrocellulose filter(0.2-mm pore; Bio-Rad Laboratories) in 96 mM glycine, 12mM Tris base, 0.01% SDS, and 20% methanol, which wasblocked in 5% skimmed milk in PBST for 2 h and incubatedwith mouse anti-M13 pIII monoclonal antibody (1µg/mL; NewEngland Biolabs) for 1 h atroom temperature, followed by theaddition of goat anti-mouse immunoglobulin (IgG) HRP-streptavidin conjugate (ImmuClub Labs, Sunnyvale, CA) at a1:10,000 dilution in 1% BSA in PBST. After a 1-h incubation,the filters were washed with PBST, developed using a Super-Signal West Pico kit (Pierce Biotechnologies, Rockford, IL) for5 min, and exposed to X-ray film.

CPT Treatment. CPT treatments were performed at 48 hafter the addition of phage in medium containing 10% fetalbovine serum. The treated cells were incubated with CPT at 10uM for 5 h at 37°C, followed by the replacement with freshmedium and an additional incubation of 19 h at 37°C.

In Vitro Phagemid Particle Transfection. The cells wereplaced into 24-well plates at the density of 10,000 cells per welland kept for 24 h before phage particles were added at 1011

pfu/mL for incubation, which lasted for 48 h at 37°C incomplete media. Consequently, the cells were visualized underan epifluorescent inverted microscope (Axioskop 2; Carl Zeiss,Gottingen Germany) equipped with a fluorescein isothiocyanate(FITC) filter set. In another experiment, the cells were lysedand assayed for luciferase activity. The luminescence wasmeasured in a MiniLumat LB 9506 luminometer (EG&GBERTHOLD, Wildbad, Germany); immediately 20µL of celllysate was mixed with 30µL of luciferase substrate (Promega,USA). Relative light units were compared to count the luciferaseactivity. All the transfections were done in triplicate andperformed at least three times.

Statistics.The results were expressed on the basis of tripletexperiments as the means( SE; statistical analysis wasperformed using unpaired two-tailed Student’st test, with apvalue of<0.05 considered to be statistically significant.

Results

Producing of (+)Gene or (-)Gene Decided by TheirOrientations to Filamentous Origin. As shown in Figure 1,when the gene’s orientation is the same as the origin’s, therewill be a sense strand, (+)gene; when it is the opposite, therewill be an antisense strand, (-)gene.

EGF Displayed by the Phagemid Particles.To display EGFon the phage, we inserted the EGF encoding sequence directly

into the amino terminus of pIII gene of M13KO7 plasmid toprepare the M13KO7-EGF helper phages. The phagemidscarrying reporter genes were rescued by these M13KO7-EGF

Figure 3. Direct visualization of GFP autofluorescence in H1299 cellstransfected with various EGF-displayed phagemid particles. (A) Panels1 and 3 were H1299 cells transfected with pEGFP and pEGFP-AFphagemid particles, respectively; panels 2 and 4 are phase contrastimages in the same field as panels 1 and 3, respectively. (B) Panels 1and 3 were H1299 cells transfected with pEGFP and pEGFP-AFphagemid particles and then treated with CPT, respectively; panels 2and 4 are phase contrast images in the same field as panels 1 and 3,respectively. (C) Quantitation of GFP positive cells transfected bypEGFP-Amp, pEGFP-AF phagemid particles with or without CPTtreatment. GFP positive cells from duplicate wells were averaged andexperiments were performed in triplicate. In total, 40,000 cells per wellwere counted each time. Error bars: standard deviation of the mean.Standard Error bars: standard error of the mean. Statistical analysiswas performed between pEGFP-AF and pEGFP-Amp or between thetwo CPT treated phagemid transduction. *p value< 0.05.

628 Biotechnol. Prog., 2006, Vol. 22, No. 3

helper phages. To assess the efficiency of phagemid packaging,we extracted ssDNA from the phagemid particles and analyzedit by gel electrophoresis. As shown in Figure 2A, M13KO7-EGF could rescue four kinds of phagemid DNA to formphagemid particles at a similar efficiency. The levels of the EGFdisplayed in the resulted phagemid particles were analyzed byWestern blot using pIII specific antibody (Figure 2B). Theweight of pIII proteins from phagemid particles rescued byM13KO7-EGF was∼6 kd larger than that of those rescued byM13KO7 helper phage, thus demonstrating the high level ofEGF display in the modified phages.

Examination of Reporter Gene Expression Mediated byVarious Phagemid Particles.H1299 cells were incubated withphagemid particles. The endocytosis of the EGF modifiedphages in the H1299 cells was confirmed by fluorescentimmunostaining (data not shown). Following 48 h of incubationwith phagemid particles carrying GFP gene cassette, the numberof GFP-expressing cells were visualized under a fluorescentmicroscopy (Figure 3). The pEGFP-AF phagemid particlesshowed a 1.5- to 2-fold increase in the percentage of GFPpositive cells when compared with pEGFP-amp phagemid ones(Figure 3A and C). We further analyzed the effects of CPTtreatment on the transduction of these two kinds of phagemidparticles. Although CPT increased the transduction of the GFPgene carrying the two kinds of phagemid particles, the (-)EGFPcarrying phagemid ones still remained at 2-fold transductionof (+)EGFP carrying phagemid ones (Figure 3B and C).

To rule out the possibility that the phenomenon was uniquefor GFP gene, we examined EGF-displayed phagemid particlescarrying the luciferase gene. We used the EGF-modifiedM13EGFKO7 helper phages to package phagemid pGl3-Control, which contained (+) luciferase gene, and pGl3-AF,which contains (-) luciferase gene. Without CPT treatment,the luciferase activity induced by pGl3-AF phages transductionwas approximately 2-fold that induced by pGl3-Control phages(Figure 4A). When the CPT was added, the luciferase activityfrom both phagemid particles increased dramatically. Further-more, the (-)gene carrying phagemid particles produced a 1.5-fold luciferase level compared to that of (+)gene carryingphagemid ones (Figure 4B). The results were in concordancewith those from the EGFP carrying phagemid particles. Taken

together, the (-)gene was found to be more efficient forreceptor-mediated phage gene delivery systems.

Discussion

In this study, we employed two reporter gene cassettes,luciferase and EGFP, to quantify the levels of gene tranductionby EGF-targeting phagemid particles, and demonstrated thatphagemid particles carrying (-)gene induced better genetransduction than those carrying (+)gene.

Camptothecin is a type I topoisomerase inhibitor that producesDNA strand breaks (13), which could induce DNA repairfunctions. Therefore, the CPT treatment should enhance thephage transduction by conversion of single-stranded vectorgenomes to double-stranded molecules. The data demonstratedthat the CPT treatment did enhance the transduction of all ofthe tested phagemid particles. However, from the luciferaseactivity obtained with or without CPT treatment, we can seethat although the CPT treatment increased the transduction byover 10-fold, the (-)gene carrying phages still remained at about1.5-fold transduction compared to the (+)gene carrying the samevectors. Therefore, we reasoned that CPT might also enhancethe transcription of the (-)gene.

Although the mechanism of regulation of ssDNA transcriptionin eukaryotes remains largely unknown, it has ever been proventhat yeast RNA polymerase B transcribes a nonrandomly definedsingle-stranded template and the initiation frequency is muchhigher on a linear double-stranded template (14). Also, therehas been ample evidence that ssDNA can be recognized by theeukaryotic cells (15-17). Unfortunately we have not found anystudy showing the process of the (-)/(+)ssDNA transcription.

Therefore, we propose several possible explanations. One isthat when a gene is single-stranded, the host RNA polymerasecan only recognize the (-)gene for transcription. Thus the(-)gene, which is an antisense strand, can be transcribed directlyinto mRNA or when it is replicated into double-stranded DNA.On the other hand, the (+)ssDNA can only be transcribed whenit is replicated into double-stranded DNA. The other is that thepartial elements of (-)ssDNA can selectively bind some factorsto enhance the gene expression. If so, we can insert some cis-element that binds the ssDP (ssDNA binding proteins) to

Figure 4. Luciferase activity of H1299 cells transfected with EGF-displayed phagemid particles. (A) H1299 cells transfected with pGl3-Controland pGl3-AF phagemid particles, respectively. (B) H1299 cells transfected pGl3-Control and pGl3-AF and then treated with CPT, respectively.RLU: relative luciferase units. The data represent the combination of three independent experiments, each done on triplicate culture wells. StandardError bars: standard error of the mean. Statistical analysis was performed between pGl3-control and pGl3-AF or between the two CPT treatedphagemid transduction. *p value< 0.05.

Biotechnol. Prog., 2006, Vol. 22, No. 3 629

enhance DNA binding and gene expression (18). Accordingly,further studies should be performed to elucidate how the(-)gene can enhance the gene expression.

In conclusion, given the advantages of (-)ssDNA phagemid,we should clone the genes of interest in the reverse orientationto filamentous origin to enhance their expression when preparingphagemid gene delivery vectors.

Acknowledgment

This work was supported by National 973 Basic ResearchProgram of China (2004CB518802).

References and Notes(1) Burg, M. A.; Jensen-Pergakes, K.; Gonzalez, A. M.; Ravey, P.;

Baird, A.; Larocca, D. Enhanced Phagemid Particle Gene Transferin Camptothecin-Treated Carcinoma Cells.Cancer. Res.2002, 62(4), 977-981.

(2) Larocca, D.; Witte, A.; Johnson, W.; Pierce, G. F.; Baird, A.Targeting Bacteriophage to Mammalian Cell Surface Receptors forGene Delivery.Hum. Gene Ther.1998, 9 (16), 2393-2399.

(3) Larocca, D.; Kassner, P. D.; Witte, A.; Ladner, R. C.; Pierce, G.F.; Baird, A. Gene Transfer to Mammalian Cells Using GeneticallyTargeted Filamentous Bacteriophage.FASEB J.1999, 13 (6), 727-734.

(4) Kassner, P. D.; Burg, M. A.; Baird, A.; Larocca, D. GeneticSelection of Phage Engineered for Receptor-Mediated Gene Transferto Mammalian Cells.Biochem. Biophys. Res. Commun.1999, 264(3), 921-928.

(5) Li, Z.; Zhang, J.; Zhao, R.; Xu, Y.; Gu, J. Preparation of Peptide-Targeted Phagemid Particles Using a Protein III-Modified HelperPhage.Biotechniques2005, 39 (4), 493-497.

(6) Poul, M. A.; Marks, J. D. Targeted Gene Delivery to MammalianCells by Filamentous Bacteriophage.J. Mol. Biol. 1999, 288 (2),203-211.

(7) Di Giovine, M.; Salone, B.; Martina, Y.; Amati, V.; Zambruno,G.; Cundari, E.; Failla, C. M.; Saggio, I. Binding Properties, CellDelivery, and Gene Transfer of Adenoviral Penton Base DisplayingBacteriophage.Virology 2001, 282 (1), 102-112.

(8) Dotto, G. P.; Enea, V.; Zinder, N. D. Functional Analysis ofBacteriophage f1 Intergenic Region.Virology 1981, 114 (2), 463-473.

(9) Dotto, G. P.; Horiuchi, K. Replication of A Plasmid ContainingTwo Origins of Bacteriophage.J. Mol. Biol. 1981, 153 (1), 169-176.

(10) Urcelay, E.; Ward, P.; Wiener, S. M.; Safer, B.; Kotin, R. M.Asymmetric Replication In Vitro From A Human Sequence ElementIs Dependent On Adeno-Associated Virus Rep Protein.J. Virol.1995, 69 (4), 2038-2046.

(11) Gan, B.-R.; Huang, P.; Yuan, Y.; Yao, J.; Wen, H.; Wen, Z.Secretion of 0verproduced Human Epidermal Growth Factor inEscherichia coli. Acta Biochim. Biophys. Sin.1992, 24 (6).

(12) Rondot, S.; Koch, J.; Breitling, F.; Dubel, S. A Helper Phage ToImprove Single-chain Antibody Presentation in Phage Display.Nat.Biotechnol.2001, 19 (1), 75-78.

(13) Liu, L. F. DNA Topoisomerase Poisons as Antitumor Drugs.Annu.ReV. Biochem.1989, 58, 351-375.

(14) Nagamine, Y.; Bennetzen, J.; Sentenac, A.; Fromageot, P. Single-Stranded DNA Transcription by Yeast RNA Polymerase B.Biochim.Biophys. Acta1981, 656 (2), 220-227.

(15) Gaillard, C.; Cabannes, E.; Strauss, F. Identity of the RNA-bindingProtein K of hnRNP Particles with Protein H16, a Sequence-SpecificSingle Strand DNA-Binding Protein.Nucleic Acids Res.1994, 22(20), 4183-4186.

(16) Tada, H.; Khalili, K. A Novel Sequence-Specific DNA-BindingProtein, LCP-1, Interacts with Single-Stranded DNA and Differen-tially Regulates Early Gene Expression of the Human NeurotropicJC Virus.J. Virol. 1992, 66 (12), 6885-6892.

(17) Bayarsaihan, D.; Soto, R. J.; Lukens, L. N. Cloning andCharacterization of A Novel Sequence-specific Single-stranded-DNA-binding Protein.Biochem. J.1998, 331 ((Pt 2)), 447-452.

(18) Tomonaga, T.; Levens, D. Activating Transcription From SingleStranded DNA.Proc. Natl. Acad. Sci. U.S.A.1996, 93 (12), 5830-5835.

Accepted for publication April 6, 2006.

BP050392J

630 Biotechnol. Prog., 2006, Vol. 22, No. 3